0: [2025-08-07 02:03:24,132] [INFO] [axolotl.utils.schemas.validation.check_eval_packing:118] [PID:2863199] [RANK:0] explicitly setting `eval_sample_packing` to match `sample_packing`
2: [2025-08-07 02:03:24,132] [INFO] [axolotl.utils.schemas.validation.check_eval_packing:118] [PID:2234790] [RANK:0] explicitly setting `eval_sample_packing` to match `sample_packing`
1: [2025-08-07 02:03:24,132] [INFO] [axolotl.utils.schemas.validation.check_eval_packing:118] [PID:860438] [RANK:0] explicitly setting `eval_sample_packing` to match `sample_packing`
3: [2025-08-07 02:03:24,132] [INFO] [axolotl.utils.schemas.validation.check_eval_packing:118] [PID:1842938] [RANK:0] explicitly setting `eval_sample_packing` to match `sample_packing`
0: [2025-08-07 02:03:24,132] [INFO] [axolotl.utils.schemas.validation.hint_sample_packing_padding:217] [PID:2863199] [RANK:0] Setting `pad_to_sequence_len: true` to prevent memory leaks when sample_packing
2: [2025-08-07 02:03:24,132] [INFO] [axolotl.utils.schemas.validation.hint_sample_packing_padding:217] [PID:2234790] [RANK:0] Setting `pad_to_sequence_len: true` to prevent memory leaks when sample_packing
1: [2025-08-07 02:03:24,132] [INFO] [axolotl.utils.schemas.validation.hint_sample_packing_padding:217] [PID:860438] [RANK:0] Setting `pad_to_sequence_len: true` to prevent memory leaks when sample_packing
3: [2025-08-07 02:03:24,132] [INFO] [axolotl.utils.schemas.validation.hint_sample_packing_padding:217] [PID:1842938] [RANK:0] Setting `pad_to_sequence_len: true` to prevent memory leaks when sample_packing
0: [2025-08-07 02:03:40,075] [INFO] [axolotl.cli.config.load_cfg:244] [PID:2863199] [RANK:0] config:
0: {
0:   "activation_offloading": false,
0:   "auto_resume_from_checkpoints": true,
0:   "axolotl_config_path": "/lustre/fswork/projects/rech/dgo/udv55np/train/tmp/1754476678272811264.yaml",
0:   "base_model": "/lustre/fswork/projects/rech/qwv/udv55np/Qwen/Qwen2.5-1.5B",
0:   "base_model_config": "/lustre/fswork/projects/rech/qwv/udv55np/Qwen/Qwen2.5-1.5B",
0:   "batch_size": 16,
0:   "bf16": true,
0:   "capabilities": {
0:     "bf16": true,
0:     "compute_capability": "sm_90",
0:     "fp8": false,
0:     "n_gpu": 4,
0:     "n_node": 1
0:   },
0:   "chat_template": "qwen_25",
0:   "dataloader_num_workers": 4,
0:   "dataloader_pin_memory": true,
0:   "dataloader_prefetch_factor": 256,
0:   "dataset_prepared_path": "/lustre/fsn1/projects/rech/dgo/udv55np/dataset/Qwen3-235B-A22B/Qwen2.5-1.5B/1",
0:   "dataset_processes": 192,
0:   "datasets": [
0:     {
0:       "chat_template": "tokenizer_default",
0:       "data_files": [
0:         "/lustre/fswork/projects/rech/qwv/udv55np/dataset/ift/Qwen3-235B-A22B/thinking/0007.jsonl",
0:         "/lustre/fswork/projects/rech/qwv/udv55np/dataset/ift/Qwen3-235B-A22B/thinking/0009.jsonl",
0:         "/lustre/fswork/projects/rech/qwv/udv55np/dataset/ift/Qwen3-235B-A22B/thinking/0005.jsonl",
0:         "/lustre/fswork/projects/rech/qwv/udv55np/dataset/ift/Qwen3-235B-A22B/thinking/0006.jsonl",
0:         "/lustre/fswork/projects/rech/qwv/udv55np/dataset/ift/Qwen3-235B-A22B/thinking/0014.jsonl",
0:         "/lustre/fswork/projects/rech/qwv/udv55np/dataset/ift/Qwen3-235B-A22B/thinking/0010.jsonl",
0:         "/lustre/fswork/projects/rech/qwv/udv55np/dataset/ift/Qwen3-235B-A22B/thinking/0012.jsonl",
0:         "/lustre/fswork/projects/rech/qwv/udv55np/dataset/ift/Qwen3-235B-A22B/thinking/0008.jsonl",
0:         "/lustre/fswork/projects/rech/qwv/udv55np/dataset/ift/Qwen3-235B-A22B/thinking/0001.jsonl",
0:         "/lustre/fswork/projects/rech/qwv/udv55np/dataset/ift/Qwen3-235B-A22B/thinking/0002.jsonl",
0:         "/lustre/fswork/projects/rech/qwv/udv55np/dataset/ift/Qwen3-235B-A22B/thinking/0013.jsonl",
0:         "/lustre/fswork/projects/rech/qwv/udv55np/dataset/ift/Qwen3-235B-A22B/thinking/0015.jsonl",
0:         "/lustre/fswork/projects/rech/qwv/udv55np/dataset/ift/Qwen3-235B-A22B/thinking/0004.jsonl",
0:         "/lustre/fswork/projects/rech/qwv/udv55np/dataset/ift/Qwen3-235B-A22B/thinking/0011.jsonl",
0:         "/lustre/fswork/projects/rech/qwv/udv55np/dataset/ift/Qwen3-235B-A22B/thinking/0000.jsonl",
0:         "/lustre/fswork/projects/rech/qwv/udv55np/dataset/ift/Qwen3-235B-A22B/thinking/0003.jsonl"
0:       ],
0:       "ds_type": "json",
0:       "field_messages": "conversations",
0:       "message_property_mappings": {
0:         "content": "content",
0:         "role": "role"
0:       },
0:       "path": "/lustre/fswork/projects/rech/qwv/udv55np/dataset/ift/Qwen3-235B-A22B/thinking",
0:       "trust_remote_code": false,
0:       "type": "chat_template"
0:     }
0:   ],
0:   "ddp": true,
0:   "deepspeed": {
0:     "bf16": {
0:       "enabled": true
0:     },
0:     "gradient_accumulation_steps": "auto",
0:     "gradient_clipping": "auto",
0:     "train_batch_size": "auto",
0:     "train_micro_batch_size_per_gpu": "auto",
0:     "wall_clock_breakdown": false,
0:     "zero_optimization": {
0:       "contiguous_gradients": true,
0:       "overlap_comm": true,
0:       "reduce_bucket_size": "auto",
0:       "stage": 3,
0:       "stage3_gather_16bit_weights_on_model_save": true,
0:       "stage3_param_persistence_threshold": "auto",
0:       "stage3_prefetch_bucket_size": "auto",
0:       "sub_group_size": 0
0:     }
0:   },
0:   "device": "cuda:0",
0:   "device_map": {
0:     "": 0
0:   },
0:   "env_capabilities": {
0:     "torch_version": "2.6.0"
0:   },
0:   "eval_batch_size": 1,
0:   "eval_causal_lm_metrics": [
0:     "sacrebleu",
0:     "comet",
0:     "ter",
0:     "chrf"
0:   ],
0:   "eval_max_new_tokens": 128,
0:   "eval_sample_packing": true,
0:   "eval_table_size": 0,
0:   "evals_per_epoch": 0,
0:   "flash_attention": true,
0:   "fp16": false,
0:   "gradient_accumulation_steps": 4,
0:   "gradient_checkpointing": true,
0:   "gradient_checkpointing_kwargs": {
0:     "use_reentrant": true
0:   },
0:   "learning_rate": 1e-05,
0:   "lisa_layers_attribute": "model.layers",
0:   "load_best_model_at_end": false,
0:   "load_in_4bit": false,
0:   "load_in_8bit": false,
0:   "local_rank": 0,
0:   "logging_steps": 10,
0:   "lora_dropout": 0.0,
0:   "loraplus_lr_embedding": 1e-06,
0:   "lr_scheduler": "warmup_stable_decay",
0:   "lr_scheduler_kwargs": {
0:     "min_lr_ratio": 0.1,
0:     "num_decay_steps": 300
0:   },
0:   "max_prompt_len": 512,
0:   "mean_resizing_embeddings": false,
0:   "micro_batch_size": 1,
0:   "model_config_type": "qwen2",
0:   "num_epochs": 1.0,
0:   "optimizer": "adamw_torch_fused",
0:   "output_dir": "/lustre/fswork/projects/rech/dgo/udv55np/ift/Qwen3-235B-A22B/Qwen2.5-1.5B/1",
0:   "pad_to_sequence_len": true,
0:   "pretrain_multipack_attn": true,
0:   "pretrain_multipack_buffer_size": 10000,
0:   "profiler_steps_start": 0,
0:   "qlora_sharded_model_loading": false,
0:   "ray_num_workers": 1,
0:   "resources_per_worker": {
0:     "GPU": 1
0:   },
0:   "sample_packing": true,
0:   "sample_packing_bin_size": 200,
0:   "sample_packing_group_size": 100000,
0:   "save_only_model": false,
0:   "save_safetensors": true,
0:   "save_steps": 0.2,
0:   "save_total_limit": 20,
0:   "sequence_len": 16384,
0:   "sequence_parallel_degree": 1,
0:   "shuffle_merged_datasets": true,
0:   "skip_prepare_dataset": false,
0:   "special_tokens": {
0:     "bos_token": "<|im_start|>",
0:     "eos_token": "<|im_end|>",
0:     "pad_token": "<|endoftext|>"
0:   },
0:   "strict": false,
0:   "tensor_parallel_size": 1,
0:   "tf32": false,
0:   "tiled_mlp_use_original_mlp": true,
0:   "tokenizer_config": "/lustre/fswork/projects/rech/qwv/udv55np/Qwen/Qwen2.5-1.5B",
0:   "torch_dtype": "torch.bfloat16",
0:   "train_on_inputs": false,
0:   "trl": {
0:     "log_completions": false,
0:     "mask_truncated_completions": false,
0:     "ref_model_mixup_alpha": 0.9,
0:     "ref_model_sync_steps": 64,
0:     "scale_rewards": true,
0:     "sync_ref_model": false,
0:     "use_vllm": false,
0:     "vllm_server_host": "0.0.0.0",
0:     "vllm_server_port": 8000
0:   },
0:   "use_ray": false,
0:   "use_tensorboard": true,
0:   "val_set_size": 0.0,
0:   "vllm": {
0:     "device": "auto",
0:     "dtype": "auto",
0:     "gpu_memory_utilization": 0.9,
0:     "host": "0.0.0.0",
0:     "port": 8000
0:   },
0:   "warmup_steps": 150,
0:   "weight_decay": 0.0,
0:   "world_size": 4
0: }
0: [2025-08-07 02:03:40,076] [INFO] [axolotl.cli.checks.check_user_token:35] [PID:2863199] [RANK:0] Skipping HuggingFace token verification because HF_HUB_OFFLINE is set to True. Only local files will be used.
1: [2025-08-07 02:03:40,747] [INFO] [axolotl.utils.data.sft._load_raw_datasets:310] [PID:860438] [RANK:0] Loading raw datasets...
1: [2025-08-07 02:03:42,457] [INFO] [axolotl.utils.data.wrappers.get_dataset_wrapper:88] [PID:860438] [RANK:0] Loading dataset: /lustre/fswork/projects/rech/qwv/udv55np/dataset/ift/Qwen3-235B-A22B/thinking with base_type: chat_template and prompt_style: None
1: 
Tokenizing Prompts (num_proc=192):   0%|          | 0/1393783 [00:00<?, ? examples/s]
Tokenizing Prompts (num_proc=192):   0%|          | 1000/1393783 [00:17<6:54:08, 56.05 examples/s]
Tokenizing Prompts (num_proc=192):   0%|          | 2000/1393783 [00:17<2:52:33, 134.43 examples/s]
Tokenizing Prompts (num_proc=192):   0%|          | 3000/1393783 [00:18<1:35:24, 242.95 examples/s]
Tokenizing Prompts (num_proc=192):   0%|          | 5000/1393783 [00:18<44:28, 520.46 examples/s]  
Tokenizing Prompts (num_proc=192):   1%|          | 8000/1393783 [00:18<20:43, 1114.73 examples/s]
Tokenizing Prompts (num_proc=192):   1%|          | 10000/1393783 [00:19<15:29, 1489.11 examples/s]
Tokenizing Prompts (num_proc=192):   1%|          | 12000/1393783 [00:19<10:59, 2095.25 examples/s]
Tokenizing Prompts (num_proc=192):   1%|          | 13000/1393783 [00:19<10:19, 2228.32 examples/s]
Tokenizing Prompts (num_proc=192):   1%|          | 15000/1393783 [00:20<09:19, 2464.51 examples/s]
Tokenizing Prompts (num_proc=192):   1%
1: |          | 17000/1393783 [00:20<07:27, 3077.14 examples/s]
Tokenizing Prompts (num_proc=192):   1%|▏         | 20000/1393783 [00:21<05:23, 4247.01 examples/s]
Tokenizing Prompts (num_proc=192):   2%|▏         | 21000/1393783 [00:21<05:07, 4467.32 examples/s]
Tokenizing Prompts (num_proc=192):   2%|▏         | 22000/1393783 [00:21<04:51, 4703.45 examples/s]
Tokenizing Prompts (num_proc=192):   2%|▏         | 24000/1393783 [00:21<03:48, 5995.28 examples/s]
Tokenizing Prompts (num_proc=192):   2%|▏         | 25000/1393783 [00:21<04:33, 5013.43 examples/s]
Tokenizing Prompts (num_proc=192):   2%|▏         | 26000/1393783 [00:22<04:22, 5205.96 examples/s]
Tokenizing Prompts (num_proc=192):   2%|▏         | 28000/1393783 [00:22<03:24, 6678.08 examples/s]
Tokenizing Prompts (num_proc=192):   2%|▏         | 29000/1393783 [00:22<04:18, 5281.59 examples/s]
Tokenizing Prompts (num_proc=192):   2%|▏         | 30000/1393783 [00:22<04:09, 5455.87 examples/s]
Tokenizing Prompts (num_proc=192):   2%|▏  
1:        | 31000/1393783 [00:23<05:00, 4533.87 examples/s]
Tokenizing Prompts (num_proc=192):   2%|▏         | 32000/1393783 [00:23<05:41, 3985.22 examples/s]
Tokenizing Prompts (num_proc=192):   2%|▏         | 33000/1393783 [00:23<05:11, 4368.71 examples/s]
Tokenizing Prompts (num_proc=192):   2%|▏         | 34000/1393783 [00:23<04:43, 4799.00 examples/s]
Tokenizing Prompts (num_proc=192):   3%|▎         | 36000/1393783 [00:23<03:26, 6572.55 examples/s]
Tokenizing Prompts (num_proc=192):   3%|▎         | 37000/1393783 [00:24<04:13, 5354.82 examples/s]
Tokenizing Prompts (num_proc=192):   3%|▎         | 38000/1393783 [00:24<03:45, 6016.31 examples/s]
Tokenizing Prompts (num_proc=192):   3%|▎         | 39000/1393783 [00:24<04:04, 5544.65 examples/s]
Tokenizing Prompts (num_proc=192):   3%|▎         | 40000/1393783 [00:24<04:19, 5214.57 examples/s]
Tokenizing Prompts (num_proc=192):   3%|▎         | 41000/1393783 [00:24<03:57, 5699.47 examples/s]
Tokenizing Prompts (num_proc=192):   3%|▎      
1:    | 44000/1393783 [00:25<02:42, 8298.85 examples/s]
Tokenizing Prompts (num_proc=192):   3%|▎         | 45000/1393783 [00:25<05:06, 4404.95 examples/s]
Tokenizing Prompts (num_proc=192):   3%|▎         | 46000/1393783 [00:25<04:27, 5038.08 examples/s]
Tokenizing Prompts (num_proc=192):   3%|▎         | 48000/1393783 [00:25<03:24, 6571.38 examples/s]
Tokenizing Prompts (num_proc=192):   4%|▎         | 49000/1393783 [00:26<03:19, 6736.03 examples/s]
Tokenizing Prompts (num_proc=192):   4%|▎         | 50000/1393783 [00:26<04:42, 4756.14 examples/s]
Tokenizing Prompts (num_proc=192):   4%|▎         | 51000/1393783 [00:26<04:59, 4490.74 examples/s]
Tokenizing Prompts (num_proc=192):   4%|▎         | 52000/1393783 [00:27<05:52, 3811.66 examples/s]
Tokenizing Prompts (num_proc=192):   4%|▍         | 53000/1393783 [00:27<06:21, 3516.21 examples/s]
Tokenizing Prompts (num_proc=192):   4%|▍         | 54000/1393783 [00:27<05:38, 3962.41 examples/s]
Tokenizing Prompts (num_proc=192):   4%|▍         |
1:  56000/1393783 [00:27<04:39, 4783.76 examples/s]
Tokenizing Prompts (num_proc=192):   4%|▍         | 57000/1393783 [00:28<04:58, 4481.90 examples/s]
Tokenizing Prompts (num_proc=192):   4%|▍         | 58000/1393783 [00:28<05:09, 4319.13 examples/s]
Tokenizing Prompts (num_proc=192):   4%|▍         | 59000/1393783 [00:28<04:26, 5001.96 examples/s]
Tokenizing Prompts (num_proc=192):   4%|▍         | 61000/1393783 [00:28<04:01, 5516.15 examples/s]
Tokenizing Prompts (num_proc=192):   5%|▍         | 65000/1393783 [00:29<02:29, 8916.63 examples/s]
Tokenizing Prompts (num_proc=192):   5%|▍         | 66000/1393783 [00:29<03:52, 5701.74 examples/s]
Tokenizing Prompts (num_proc=192):   5%|▍         | 68000/1393783 [00:29<03:17, 6722.66 examples/s]
Tokenizing Prompts (num_proc=192):   5%|▍         | 69000/1393783 [00:30<03:31, 6249.64 examples/s]
Tokenizing Prompts (num_proc=192):   5%|▌         | 70000/1393783 [00:30<06:42, 3291.08 examples/s]
Tokenizing Prompts (num_proc=192):   5%|▌         | 720
1: 00/1393783 [00:31<05:18, 4151.89 examples/s]
Tokenizing Prompts (num_proc=192):   5%|▌         | 73000/1393783 [00:31<05:26, 4050.96 examples/s]
Tokenizing Prompts (num_proc=192):   5%|▌         | 74000/1393783 [00:31<05:21, 4103.25 examples/s]
Tokenizing Prompts (num_proc=192):   5%|▌         | 75000/1393783 [00:31<05:13, 4209.01 examples/s]
Tokenizing Prompts (num_proc=192):   5%|▌         | 76000/1393783 [00:32<07:28, 2939.59 examples/s]
Tokenizing Prompts (num_proc=192):   6%|▌         | 78000/1393783 [00:33<06:40, 3282.35 examples/s]
Tokenizing Prompts (num_proc=192):   6%|▌         | 79000/1393783 [00:34<13:29, 1623.50 examples/s]
Tokenizing Prompts (num_proc=192):   6%|▌         | 83000/1393783 [00:34<06:22, 3424.93 examples/s]
Tokenizing Prompts (num_proc=192):   6%|▌         | 84000/1393783 [00:35<06:49, 3200.47 examples/s]
Tokenizing Prompts (num_proc=192):   6%|▌         | 86000/1393783 [00:35<05:18, 4111.80 examples/s]
Tokenizing Prompts (num_proc=192):   6%|▌         | 87000/1
1: 393783 [00:35<05:43, 3808.08 examples/s]
Tokenizing Prompts (num_proc=192):   6%|▋         | 89000/1393783 [00:36<05:19, 4089.35 examples/s]
Tokenizing Prompts (num_proc=192):   6%|▋         | 90000/1393783 [00:36<05:11, 4182.45 examples/s]
Tokenizing Prompts (num_proc=192):   7%|▋         | 92000/1393783 [00:36<04:32, 4782.17 examples/s]
Tokenizing Prompts (num_proc=192):   7%|▋         | 93000/1393783 [00:37<07:44, 2798.69 examples/s]
Tokenizing Prompts (num_proc=192):   7%|▋         | 94000/1393783 [00:38<07:36, 2845.51 examples/s]
Tokenizing Prompts (num_proc=192):   7%|▋         | 96000/1393783 [00:38<05:06, 4229.75 examples/s]
Tokenizing Prompts (num_proc=192):   7%|▋         | 97000/1393783 [00:38<06:52, 3144.05 examples/s]
Tokenizing Prompts (num_proc=192):   7%|▋         | 98000/1393783 [00:39<07:57, 2712.54 examples/s]
Tokenizing Prompts (num_proc=192):   7%|▋         | 99000/1393783 [00:39<08:02, 2685.89 examples/s]
Tokenizing Prompts (num_proc=192):   7%|▋         | 100000/1393
1: 783 [00:39<06:45, 3191.47 examples/s]
Tokenizing Prompts (num_proc=192):   7%|▋         | 103000/1393783 [00:40<04:17, 5010.50 examples/s]
Tokenizing Prompts (num_proc=192):   7%|▋         | 104000/1393783 [00:41<08:19, 2580.74 examples/s]
Tokenizing Prompts (num_proc=192):   8%|▊         | 108000/1393783 [00:41<04:43, 4532.42 examples/s]
Tokenizing Prompts (num_proc=192):   8%|▊         | 110000/1393783 [00:41<03:45, 5695.40 examples/s]
Tokenizing Prompts (num_proc=192):   8%|▊         | 112000/1393783 [00:41<03:16, 6509.50 examples/s]
Tokenizing Prompts (num_proc=192):   8%|▊         | 114000/1393783 [00:42<03:14, 6593.62 examples/s]
Tokenizing Prompts (num_proc=192):   8%|▊         | 115000/1393783 [00:42<03:31, 6039.11 examples/s]
Tokenizing Prompts (num_proc=192):   8%|▊         | 116000/1393783 [00:42<04:47, 4448.26 examples/s]
Tokenizing Prompts (num_proc=192):   9%|▊         | 119000/1393783 [00:43<03:01, 7010.10 examples/s]
Tokenizing Prompts (num_proc=192):   9%|▊         | 12100
1: 0/1393783 [00:43<02:47, 7607.62 examples/s]
Tokenizing Prompts (num_proc=192):   9%|▉         | 124000/1393783 [00:44<04:01, 5247.99 examples/s]
Tokenizing Prompts (num_proc=192):   9%|▉         | 125000/1393783 [00:44<03:44, 5657.29 examples/s]
Tokenizing Prompts (num_proc=192):   9%|▉         | 126000/1393783 [00:44<03:57, 5327.65 examples/s]
Tokenizing Prompts (num_proc=192):   9%|▉         | 128000/1393783 [00:44<03:03, 6916.80 examples/s]
Tokenizing Prompts (num_proc=192):   9%|▉         | 130000/1393783 [00:45<04:30, 4679.17 examples/s]
Tokenizing Prompts (num_proc=192):   9%|▉         | 132000/1393783 [00:45<04:25, 4751.83 examples/s]
Tokenizing Prompts (num_proc=192):  10%|▉         | 133000/1393783 [00:45<04:08, 5078.72 examples/s]
Tokenizing Prompts (num_proc=192):  10%|▉         | 136000/1393783 [00:46<03:32, 5921.93 examples/s]
Tokenizing Prompts (num_proc=192):  10%|▉         | 138000/1393783 [00:46<03:20, 6265.72 examples/s]
Tokenizing Prompts (num_proc=192):  10%|▉         |
1:  139000/1393783 [00:46<03:35, 5813.45 examples/s]
Tokenizing Prompts (num_proc=192):  10%|█         | 140000/1393783 [00:46<03:30, 5967.67 examples/s]
Tokenizing Prompts (num_proc=192):  10%|█         | 142000/1393783 [00:47<04:31, 4604.19 examples/s]
Tokenizing Prompts (num_proc=192):  10%|█         | 144000/1393783 [00:47<03:25, 6069.23 examples/s]
Tokenizing Prompts (num_proc=192):  10%|█         | 145000/1393783 [00:48<06:25, 3240.64 examples/s]
Tokenizing Prompts (num_proc=192):  10%|█         | 146000/1393783 [00:48<05:41, 3648.67 examples/s]
Tokenizing Prompts (num_proc=192):  11%|█         | 147000/1393783 [00:48<05:15, 3947.04 examples/s]
Tokenizing Prompts (num_proc=192):  11%|█         | 148000/1393783 [00:49<05:59, 3462.49 examples/s]
Tokenizing Prompts (num_proc=192):  11%|█         | 152000/1393783 [00:49<02:46, 7466.93 examples/s]
Tokenizing Prompts (num_proc=192):  11%|█         | 154000/1393783 [00:49<02:25, 8523.67 examples/s]
Tokenizing Prompts (num_proc=192):  11%|█    
1:      | 156000/1393783 [00:49<02:44, 7534.20 examples/s]
Tokenizing Prompts (num_proc=192):  11%|█▏        | 158000/1393783 [00:49<02:25, 8472.35 examples/s]
Tokenizing Prompts (num_proc=192):  11%|█▏        | 160000/1393783 [00:50<02:26, 8423.63 examples/s]
Tokenizing Prompts (num_proc=192):  12%|█▏        | 162000/1393783 [00:50<02:20, 8775.74 examples/s]
Tokenizing Prompts (num_proc=192):  12%|█▏        | 164000/1393783 [00:50<02:50, 7223.49 examples/s]
Tokenizing Prompts (num_proc=192):  12%|█▏        | 166000/1393783 [00:50<02:20, 8730.84 examples/s]
Tokenizing Prompts (num_proc=192):  12%|█▏        | 168000/1393783 [00:51<02:00, 10209.04 examples/s]
Tokenizing Prompts (num_proc=192):  12%|█▏        | 170000/1393783 [00:51<01:48, 11306.69 examples/s]
Tokenizing Prompts (num_proc=192):  12%|█▏        | 172000/1393783 [00:51<01:55, 10584.27 examples/s]
Tokenizing Prompts (num_proc=192):  12%|█▏        | 174000/1393783 [00:51<02:07, 9550.18 examples/s] 
Tokenizing Prompts 
1: (num_proc=192):  13%|█▎        | 177000/1393783 [00:51<01:34, 12892.94 examples/s]
Tokenizing Prompts (num_proc=192):  13%|█▎        | 179000/1393783 [00:51<01:41, 12017.49 examples/s]
Tokenizing Prompts (num_proc=192):  13%|█▎        | 181000/1393783 [00:52<01:38, 12323.25 examples/s]
Tokenizing Prompts (num_proc=192):  13%|█▎        | 183000/1393783 [00:52<01:57, 10342.53 examples/s]
Tokenizing Prompts (num_proc=192):  13%|█▎        | 186000/1393783 [00:52<01:38, 12210.56 examples/s]
Tokenizing Prompts (num_proc=192):  13%|█▎        | 188000/1393783 [00:53<02:54, 6928.04 examples/s] 
Tokenizing Prompts (num_proc=192):  14%|█▎        | 190000/1393783 [00:53<02:50, 7047.02 examples/s]
Tokenizing Prompts (num_proc=192):  14%|█▍        | 192000/1393783 [00:53<02:24, 8325.98 examples/s]
Tokenizing Prompts (num_proc=192):  14%|█▍        | 197000/1393783 [00:53<01:31, 13019.67 examples/s]
Tokenizing Prompts (num_proc=192):  14%|█▍        | 199000/1393783 [00:54<01:53, 10508.
1: 36 examples/s]
Tokenizing Prompts (num_proc=192):  14%|█▍        | 201000/1393783 [00:54<02:32, 7824.96 examples/s] 
Tokenizing Prompts (num_proc=192):  15%|█▍        | 203000/1393783 [00:55<03:18, 5994.52 examples/s]
Tokenizing Prompts (num_proc=192):  15%|█▍        | 207000/1393783 [00:55<02:21, 8372.11 examples/s]
Tokenizing Prompts (num_proc=192):  15%|█▍        | 209000/1393783 [00:55<02:08, 9218.48 examples/s]
Tokenizing Prompts (num_proc=192):  15%|█▌        | 211000/1393783 [00:56<03:03, 6454.59 examples/s]
Tokenizing Prompts (num_proc=192):  15%|█▌        | 212000/1393783 [00:56<03:11, 6180.29 examples/s]
Tokenizing Prompts (num_proc=192):  15%|█▌        | 213000/1393783 [00:56<03:12, 6118.85 examples/s]
Tokenizing Prompts (num_proc=192):  15%|█▌        | 214000/1393783 [00:56<03:44, 5254.49 examples/s]
Tokenizing Prompts (num_proc=192):  15%|█▌        | 215000/1393783 [00:56<03:47, 5184.23 examples/s]
Tokenizing Prompts (num_proc=192):  15%|█▌        | 216000/
1: 1393783 [00:57<03:38, 5378.38 examples/s]
Tokenizing Prompts (num_proc=192):  16%|█▌        | 218000/1393783 [00:57<02:36, 7529.61 examples/s]
Tokenizing Prompts (num_proc=192):  16%|█▌        | 221000/1393783 [00:57<03:08, 6238.19 examples/s]
Tokenizing Prompts (num_proc=192):  16%|█▌        | 223000/1393783 [00:58<03:44, 5213.28 examples/s]
Tokenizing Prompts (num_proc=192):  16%|█▌        | 224000/1393783 [00:58<04:12, 4626.40 examples/s]
Tokenizing Prompts (num_proc=192):  16%|█▌        | 225000/1393783 [00:58<03:53, 4995.15 examples/s]
Tokenizing Prompts (num_proc=192):  16%|█▌        | 226000/1393783 [00:58<03:39, 5309.59 examples/s]
Tokenizing Prompts (num_proc=192):  16%|█▋        | 227000/1393783 [00:59<05:44, 3384.05 examples/s]
Tokenizing Prompts (num_proc=192):  16%|█▋        | 229000/1393783 [00:59<04:28, 4344.66 examples/s]
Tokenizing Prompts (num_proc=192):  17%|█▋        | 230000/1393783 [00:59<04:15, 4559.50 examples/s]
Tokenizing Prompts (num_proc=192):  1
1: 7%|█▋        | 231000/1393783 [01:00<03:41, 5248.30 examples/s]
Tokenizing Prompts (num_proc=192):  17%|█▋        | 233000/1393783 [01:00<03:11, 6068.79 examples/s]
Tokenizing Prompts (num_proc=192):  17%|█▋        | 234000/1393783 [01:00<03:43, 5187.07 examples/s]
Tokenizing Prompts (num_proc=192):  17%|█▋        | 236000/1393783 [01:00<02:56, 6562.63 examples/s]
Tokenizing Prompts (num_proc=192):  17%|█▋        | 238000/1393783 [01:01<04:47, 4025.55 examples/s]
Tokenizing Prompts (num_proc=192):  17%|█▋        | 239000/1393783 [01:01<04:26, 4340.90 examples/s]
Tokenizing Prompts (num_proc=192):  17%|█▋        | 241000/1393783 [01:02<03:37, 5293.87 examples/s]
Tokenizing Prompts (num_proc=192):  18%|█▊        | 244000/1393783 [01:02<03:31, 5447.12 examples/s]
Tokenizing Prompts (num_proc=192):  18%|█▊        | 245000/1393783 [01:02<03:58, 4806.83 examples/s]
Tokenizing Prompts (num_proc=192):  18%|█▊        | 247000/1393783 [01:03<03:45, 5083.34 examples/s]
Tokenizing 
1: Prompts (num_proc=192):  18%|█▊        | 249000/1393783 [01:03<02:55, 6528.92 examples/s]
Tokenizing Prompts (num_proc=192):  18%|█▊        | 250000/1393783 [01:03<03:06, 6144.60 examples/s]
Tokenizing Prompts (num_proc=192):  18%|█▊        | 251000/1393783 [01:03<03:31, 5394.16 examples/s]
Tokenizing Prompts (num_proc=192):  18%|█▊        | 253000/1393783 [01:04<02:55, 6505.81 examples/s]
Tokenizing Prompts (num_proc=192):  18%|█▊        | 255000/1393783 [01:04<02:16, 8327.20 examples/s]
Tokenizing Prompts (num_proc=192):  18%|█▊        | 257000/1393783 [01:04<02:52, 6592.62 examples/s]
Tokenizing Prompts (num_proc=192):  19%|█▊        | 259000/1393783 [01:05<03:17, 5745.39 examples/s]
Tokenizing Prompts (num_proc=192):  19%|█▉        | 262000/1393783 [01:05<03:35, 5244.61 examples/s]
Tokenizing Prompts (num_proc=192):  19%|█▉        | 263000/1393783 [01:05<03:49, 4935.63 examples/s]
Tokenizing Prompts (num_proc=192):  19%|█▉        | 264000/1393783 [01:06<04:40, 4026.
1: 52 examples/s]
Tokenizing Prompts (num_proc=192):  19%|█▉        | 265000/1393783 [01:07<06:38, 2830.04 examples/s]
Tokenizing Prompts (num_proc=192):  19%|█▉        | 266000/1393783 [01:07<06:16, 2995.27 examples/s]
Tokenizing Prompts (num_proc=192):  19%|█▉        | 267000/1393783 [01:07<05:25, 3461.88 examples/s]
Tokenizing Prompts (num_proc=192):  19%|█▉        | 270000/1393783 [01:07<03:19, 5638.48 examples/s]
Tokenizing Prompts (num_proc=192):  20%|█▉        | 272000/1393783 [01:08<04:13, 4421.15 examples/s]
Tokenizing Prompts (num_proc=192):  20%|█▉        | 273000/1393783 [01:09<05:30, 3386.98 examples/s]
Tokenizing Prompts (num_proc=192):  20%|█▉        | 275000/1393783 [01:09<05:30, 3386.53 examples/s]
Tokenizing Prompts (num_proc=192):  20%|█▉        | 276000/1393783 [01:09<05:35, 3328.07 examples/s]
Tokenizing Prompts (num_proc=192):  20%|█▉        | 277000/1393783 [01:10<08:42, 2139.36 examples/s]
Tokenizing Prompts (num_proc=192):  20%|█▉        | 278000/1
1: 393783 [01:11<08:10, 2274.07 examples/s]
Tokenizing Prompts (num_proc=192):  20%|██        | 280000/1393783 [01:11<05:12, 3560.26 examples/s]
Tokenizing Prompts (num_proc=192):  20%|██        | 281000/1393783 [01:11<04:49, 3845.18 examples/s]
Tokenizing Prompts (num_proc=192):  20%|██        | 282000/1393783 [01:11<04:17, 4310.03 examples/s]
Tokenizing Prompts (num_proc=192):  20%|██        | 283000/1393783 [01:12<04:37, 3996.63 examples/s]
Tokenizing Prompts (num_proc=192):  20%|██        | 284000/1393783 [01:12<05:20, 3464.94 examples/s]
Tokenizing Prompts (num_proc=192):  20%|██        | 285000/1393783 [01:12<04:43, 3904.60 examples/s]
Tokenizing Prompts (num_proc=192):  21%|██        | 286000/1393783 [01:12<04:04, 4523.76 examples/s]
Tokenizing Prompts (num_proc=192):  21%|██        | 287000/1393783 [01:13<05:19, 3462.44 examples/s]
Tokenizing Prompts (num_proc=192):  21%|██        | 289000/1393783 [01:13<05:54, 3119.03 examples/s]
Tokenizing Prompts (num_proc=192):  21
1: %|██        | 290000/1393783 [01:14<05:17, 3472.01 examples/s]
Tokenizing Prompts (num_proc=192):  21%|██        | 292000/1393783 [01:14<03:43, 4924.89 examples/s]
Tokenizing Prompts (num_proc=192):  21%|██        | 293000/1393783 [01:14<03:20, 5489.88 examples/s]
Tokenizing Prompts (num_proc=192):  21%|██        | 295000/1393783 [01:14<02:25, 7534.49 examples/s]
Tokenizing Prompts (num_proc=192):  21%|██▏       | 297000/1393783 [01:15<03:13, 5669.44 examples/s]
Tokenizing Prompts (num_proc=192):  21%|██▏       | 299000/1393783 [01:15<03:06, 5861.61 examples/s]
Tokenizing Prompts (num_proc=192):  22%|██▏       | 301000/1393783 [01:15<03:14, 5620.23 examples/s]
Tokenizing Prompts (num_proc=192):  22%|██▏       | 302000/1393783 [01:16<04:42, 3871.41 examples/s]
Tokenizing Prompts (num_proc=192):  22%|██▏       | 304000/1393783 [01:16<04:33, 3985.85 examples/s]
Tokenizing Prompts (num_proc=192):  22%|██▏       | 308000/1393783 [01:17<03:24, 5321.35 examples/s]
1: Tokenizing Prompts (num_proc=192):  22%|██▏       | 309000/1393783 [01:17<03:24, 5299.37 examples/s]
Tokenizing Prompts (num_proc=192):  22%|██▏       | 310000/1393783 [01:18<05:00, 3602.58 examples/s]
Tokenizing Prompts (num_proc=192):  22%|██▏       | 311000/1393783 [01:18<04:35, 3926.85 examples/s]
Tokenizing Prompts (num_proc=192):  22%|██▏       | 312000/1393783 [01:18<04:39, 3874.28 examples/s]
Tokenizing Prompts (num_proc=192):  22%|██▏       | 313000/1393783 [01:19<05:16, 3416.92 examples/s]
Tokenizing Prompts (num_proc=192):  23%|██▎       | 314000/1393783 [01:19<04:31, 3984.42 examples/s]
Tokenizing Prompts (num_proc=192):  23%|██▎       | 315000/1393783 [01:19<03:51, 4662.72 examples/s]
Tokenizing Prompts (num_proc=192):  23%|██▎       | 317000/1393783 [01:19<02:57, 6062.91 examples/s]
Tokenizing Prompts (num_proc=192):  23%|██▎       | 318000/1393783 [01:19<04:25, 4058.36 examples/s]
Tokenizing Prompts (num_proc=192):  23%|██▎       | 319
1: 000/1393783 [01:20<03:56, 4550.47 examples/s]
Tokenizing Prompts (num_proc=192):  23%|██▎       | 320000/1393783 [01:20<05:36, 3191.59 examples/s]
Tokenizing Prompts (num_proc=192):  23%|██▎       | 323000/1393783 [01:20<02:58, 6000.51 examples/s]
Tokenizing Prompts (num_proc=192):  23%|██▎       | 325000/1393783 [01:21<03:06, 5721.84 examples/s]
Tokenizing Prompts (num_proc=192):  24%|██▎       | 328000/1393783 [01:21<03:00, 5915.48 examples/s]
Tokenizing Prompts (num_proc=192):  24%|██▎       | 330000/1393783 [01:21<02:51, 6191.04 examples/s]
Tokenizing Prompts (num_proc=192):  24%|██▍       | 333000/1393783 [01:22<02:05, 8434.83 examples/s]
Tokenizing Prompts (num_proc=192):  24%|██▍       | 335000/1393783 [01:22<02:28, 7128.69 examples/s]
Tokenizing Prompts (num_proc=192):  24%|██▍       | 336000/1393783 [01:22<02:32, 6958.41 examples/s]
Tokenizing Prompts (num_proc=192):  24%|██▍       | 337000/1393783 [01:22<03:04, 5726.13 examples/s]
Tokenizing Prom
1: pts (num_proc=192):  24%|██▍       | 338000/1393783 [01:23<03:04, 5713.30 examples/s]
Tokenizing Prompts (num_proc=192):  25%|██▍       | 343000/1393783 [01:23<01:41, 10362.27 examples/s]
Tokenizing Prompts (num_proc=192):  25%|██▍       | 345000/1393783 [01:23<01:29, 11694.88 examples/s]
Tokenizing Prompts (num_proc=192):  25%|██▍       | 347000/1393783 [01:23<02:05, 8345.68 examples/s] 
Tokenizing Prompts (num_proc=192):  25%|██▌       | 349000/1393783 [01:24<02:16, 7674.64 examples/s]
Tokenizing Prompts (num_proc=192):  25%|██▌       | 350000/1393783 [01:24<02:36, 6659.90 examples/s]
Tokenizing Prompts (num_proc=192):  25%|██▌       | 352000/1393783 [01:24<02:10, 7995.51 examples/s]
Tokenizing Prompts (num_proc=192):  25%|██▌       | 353000/1393783 [01:24<02:14, 7735.53 examples/s]
Tokenizing Prompts (num_proc=192):  26%|██▌       | 357000/1393783 [01:24<01:22, 12511.37 examples/s]
Tokenizing Prompts (num_proc=192):  26%|██▌       | 359000/1393783
1:  [01:25<01:39, 10427.38 examples/s]
Tokenizing Prompts (num_proc=192):  26%|██▌       | 361000/1393783 [01:25<01:38, 10528.39 examples/s]
Tokenizing Prompts (num_proc=192):  26%|██▌       | 363000/1393783 [01:25<01:30, 11335.54 examples/s]
Tokenizing Prompts (num_proc=192):  26%|██▌       | 365000/1393783 [01:25<01:47, 9575.63 examples/s] 
Tokenizing Prompts (num_proc=192):  26%|██▋       | 368000/1393783 [01:25<01:29, 11420.59 examples/s]
Tokenizing Prompts (num_proc=192):  27%|██▋       | 370000/1393783 [01:26<02:24, 7072.78 examples/s] 
Tokenizing Prompts (num_proc=192):  27%|██▋       | 371000/1393783 [01:26<02:24, 7098.82 examples/s]
Tokenizing Prompts (num_proc=192):  27%|██▋       | 372000/1393783 [01:27<04:52, 3495.38 examples/s]
Tokenizing Prompts (num_proc=192):  27%|██▋       | 374000/1393783 [01:27<03:35, 4741.58 examples/s]
Tokenizing Prompts (num_proc=192):  27%|██▋       | 376000/1393783 [01:28<03:15, 5202.06 examples/s]
Tokenizing Prompts (
1: num_proc=192):  27%|██▋       | 378000/1393783 [01:28<02:45, 6140.97 examples/s]
Tokenizing Prompts (num_proc=192):  27%|██▋       | 379000/1393783 [01:28<02:45, 6148.79 examples/s]
Tokenizing Prompts (num_proc=192):  27%|██▋       | 380000/1393783 [01:28<02:54, 5794.52 examples/s]
Tokenizing Prompts (num_proc=192):  27%|██▋       | 382000/1393783 [01:28<02:20, 7177.24 examples/s]
Tokenizing Prompts (num_proc=192):  28%|██▊       | 384000/1393783 [01:28<01:53, 8926.30 examples/s]
Tokenizing Prompts (num_proc=192):  28%|██▊       | 387000/1393783 [01:29<01:27, 11478.73 examples/s]
Tokenizing Prompts (num_proc=192):  28%|██▊       | 389000/1393783 [01:29<01:30, 11071.77 examples/s]
Tokenizing Prompts (num_proc=192):  28%|██▊       | 391000/1393783 [01:29<01:19, 12544.91 examples/s]
Tokenizing Prompts (num_proc=192):  28%|██▊       | 393000/1393783 [01:29<01:55, 8630.90 examples/s] 
Tokenizing Prompts (num_proc=192):  28%|██▊       | 395000/1393783 [01:
1: 29<01:51, 8993.77 examples/s]
Tokenizing Prompts (num_proc=192):  28%|██▊       | 397000/1393783 [01:30<02:12, 7529.90 examples/s]
Tokenizing Prompts (num_proc=192):  29%|██▊       | 400000/1393783 [01:30<01:45, 9399.79 examples/s]
Tokenizing Prompts (num_proc=192):  29%|██▉       | 402000/1393783 [01:31<02:46, 5973.77 examples/s]
Tokenizing Prompts (num_proc=192):  29%|██▉       | 405000/1393783 [01:31<02:30, 6559.66 examples/s]
Tokenizing Prompts (num_proc=192):  29%|██▉       | 406000/1393783 [01:31<02:24, 6849.36 examples/s]
Tokenizing Prompts (num_proc=192):  29%|██▉       | 407000/1393783 [01:31<02:24, 6820.87 examples/s]
Tokenizing Prompts (num_proc=192):  29%|██▉       | 409000/1393783 [01:32<02:03, 8005.31 examples/s]
Tokenizing Prompts (num_proc=192):  29%|██▉       | 410000/1393783 [01:32<02:41, 6088.37 examples/s]
Tokenizing Prompts (num_proc=192):  29%|██▉       | 411000/1393783 [01:33<04:41, 3487.56 examples/s]
Tokenizing Prompts (num_proc=19
1: 2):  30%|██▉       | 412000/1393783 [01:33<05:06, 3203.06 examples/s]
Tokenizing Prompts (num_proc=192):  30%|██▉       | 413000/1393783 [01:33<04:17, 3801.69 examples/s]
Tokenizing Prompts (num_proc=192):  30%|██▉       | 414000/1393783 [01:33<03:57, 4132.99 examples/s]
Tokenizing Prompts (num_proc=192):  30%|██▉       | 416000/1393783 [01:33<02:51, 5698.98 examples/s]
Tokenizing Prompts (num_proc=192):  30%|██▉       | 417000/1393783 [01:34<03:40, 4427.27 examples/s]
Tokenizing Prompts (num_proc=192):  30%|███       | 420000/1393783 [01:34<02:10, 7489.87 examples/s]
Tokenizing Prompts (num_proc=192):  30%|███       | 422000/1393783 [01:35<03:09, 5115.05 examples/s]
Tokenizing Prompts (num_proc=192):  30%|███       | 423000/1393783 [01:35<03:55, 4121.82 examples/s]
Tokenizing Prompts (num_proc=192):  30%|███       | 424000/1393783 [01:35<03:26, 4700.92 examples/s]
Tokenizing Prompts (num_proc=192):  30%|███       | 425000/1393783 [01:35<03:48, 4246.
1: 60 examples/s]
Tokenizing Prompts (num_proc=192):  31%|███       | 426000/1393783 [01:36<03:38, 4421.62 examples/s]
Tokenizing Prompts (num_proc=192):  31%|███       | 427000/1393783 [01:36<03:26, 4690.01 examples/s]
Tokenizing Prompts (num_proc=192):  31%|███       | 429000/1393783 [01:37<05:39, 2845.02 examples/s]
Tokenizing Prompts (num_proc=192):  31%|███       | 430000/1393783 [01:37<05:00, 3212.25 examples/s]
Tokenizing Prompts (num_proc=192):  31%|███       | 431000/1393783 [01:37<04:49, 3321.99 examples/s]
Tokenizing Prompts (num_proc=192):  31%|███       | 432000/1393783 [01:38<04:45, 3370.36 examples/s]
Tokenizing Prompts (num_proc=192):  31%|███       | 434000/1393783 [01:38<04:15, 3757.78 examples/s]
Tokenizing Prompts (num_proc=192):  31%|███       | 435000/1393783 [01:38<03:57, 4043.94 examples/s]
Tokenizing Prompts (num_proc=192):  31%|███▏      | 438000/1393783 [01:38<02:26, 6521.25 examples/s]
Tokenizing Prompts (num_proc=192):  31%|█�
1: ��█▏      | 439000/1393783 [01:39<02:20, 6787.56 examples/s]
Tokenizing Prompts (num_proc=192):  32%|███▏      | 441000/1393783 [01:39<02:22, 6700.45 examples/s]
Tokenizing Prompts (num_proc=192):  32%|███▏      | 443000/1393783 [01:39<02:18, 6868.37 examples/s]
Tokenizing Prompts (num_proc=192):  32%|███▏      | 444000/1393783 [01:40<03:29, 4536.52 examples/s]
Tokenizing Prompts (num_proc=192):  32%|███▏      | 445000/1393783 [01:40<03:32, 4467.93 examples/s]
Tokenizing Prompts (num_proc=192):  32%|███▏      | 446000/1393783 [01:40<04:09, 3793.51 examples/s]
Tokenizing Prompts (num_proc=192):  32%|███▏      | 447000/1393783 [01:41<04:06, 3839.48 examples/s]
Tokenizing Prompts (num_proc=192):  32%|███▏      | 448000/1393783 [01:41<05:07, 3074.61 examples/s]
Tokenizing Prompts (num_proc=192):  32%|███▏      | 450000/1393783 [01:41<03:37, 4348.91 examples/s]
Tokenizing Prompts (num_proc=192):  32%|███▏      | 451000/1393783 [01:42<05:34
1: , 2816.91 examples/s]
Tokenizing Prompts (num_proc=192):  33%|███▎      | 453000/1393783 [01:43<04:56, 3172.05 examples/s]
Tokenizing Prompts (num_proc=192):  33%|███▎      | 455000/1393783 [01:43<03:43, 4209.70 examples/s]
Tokenizing Prompts (num_proc=192):  33%|███▎      | 458000/1393783 [01:43<02:54, 5372.39 examples/s]
Tokenizing Prompts (num_proc=192):  33%|███▎      | 459000/1393783 [01:44<03:28, 4482.20 examples/s]
Tokenizing Prompts (num_proc=192):  33%|███▎      | 460000/1393783 [01:44<03:34, 4351.59 examples/s]
Tokenizing Prompts (num_proc=192):  33%|███▎      | 461000/1393783 [01:44<05:07, 3030.13 examples/s]
Tokenizing Prompts (num_proc=192):  33%|███▎      | 463000/1393783 [01:45<03:44, 4139.66 examples/s]
Tokenizing Prompts (num_proc=192):  33%|███▎      | 465000/1393783 [01:45<02:57, 5228.49 examples/s]
Tokenizing Prompts (num_proc=192):  34%|███▎      | 468000/1393783 [01:45<02:04, 7442.20 examples/s]
Tokenizing Prompts (n
1: um_proc=192):  34%|███▎      | 469000/1393783 [01:45<02:10, 7089.67 examples/s]
Tokenizing Prompts (num_proc=192):  34%|███▍      | 471000/1393783 [01:46<03:01, 5090.91 examples/s]
Tokenizing Prompts (num_proc=192):  34%|███▍      | 473000/1393783 [01:46<02:40, 5740.44 examples/s]
Tokenizing Prompts (num_proc=192):  34%|███▍      | 475000/1393783 [01:46<02:09, 7086.45 examples/s]
Tokenizing Prompts (num_proc=192):  34%|███▍      | 476000/1393783 [01:47<03:17, 4636.17 examples/s]
Tokenizing Prompts (num_proc=192):  34%|███▍      | 477000/1393783 [01:47<03:20, 4578.95 examples/s]
Tokenizing Prompts (num_proc=192):  34%|███▍      | 478000/1393783 [01:47<03:33, 4296.46 examples/s]
Tokenizing Prompts (num_proc=192):  34%|███▍      | 479000/1393783 [01:48<03:29, 4357.65 examples/s]
Tokenizing Prompts (num_proc=192):  35%|███▍      | 481000/1393783 [01:48<02:34, 5910.67 examples/s]
Tokenizing Prompts (num_proc=192):  35%|███▍      | 4820
1: 00/1393783 [01:48<02:46, 5481.24 examples/s]
Tokenizing Prompts (num_proc=192):  35%|███▍      | 485000/1393783 [01:48<02:18, 6561.28 examples/s]
Tokenizing Prompts (num_proc=192):  35%|███▍      | 487000/1393783 [01:49<03:38, 4148.43 examples/s]
Tokenizing Prompts (num_proc=192):  35%|███▌      | 488000/1393783 [01:50<05:20, 2829.57 examples/s]
Tokenizing Prompts (num_proc=192):  35%|███▌      | 490000/1393783 [01:50<03:45, 4014.96 examples/s]
Tokenizing Prompts (num_proc=192):  35%|███▌      | 491000/1393783 [01:50<03:46, 3987.57 examples/s]
Tokenizing Prompts (num_proc=192):  35%|███▌      | 492000/1393783 [01:51<04:44, 3166.48 examples/s]
Tokenizing Prompts (num_proc=192):  35%|███▌      | 493000/1393783 [01:51<04:53, 3069.30 examples/s]
Tokenizing Prompts (num_proc=192):  35%|███▌      | 494000/1393783 [01:52<04:55, 3042.33 examples/s]
Tokenizing Prompts (num_proc=192):  36%|███▌      | 495000/1393783 [01:52<04:24, 3393.80 examples/s
1: ]
Tokenizing Prompts (num_proc=192):  36%|███▌      | 496000/1393783 [01:52<04:08, 3607.38 examples/s]
Tokenizing Prompts (num_proc=192):  36%|███▌      | 498000/1393783 [01:53<04:25, 3378.02 examples/s]
Tokenizing Prompts (num_proc=192):  36%|███▌      | 499000/1393783 [01:53<03:49, 3905.01 examples/s]
Tokenizing Prompts (num_proc=192):  36%|███▌      | 500000/1393783 [01:53<03:35, 4144.90 examples/s]
Tokenizing Prompts (num_proc=192):  36%|███▌      | 502000/1393783 [01:53<03:14, 4594.79 examples/s]
Tokenizing Prompts (num_proc=192):  36%|███▌      | 504000/1393783 [01:54<02:54, 5086.17 examples/s]
Tokenizing Prompts (num_proc=192):  36%|███▋      | 506000/1393783 [01:54<02:33, 5801.74 examples/s]
Tokenizing Prompts (num_proc=192):  36%|███▋      | 507000/1393783 [01:54<02:24, 6122.95 examples/s]
Tokenizing Prompts (num_proc=192):  36%|███▋      | 508000/1393783 [01:54<02:45, 5350.64 examples/s]
Tokenizing Prompts (num_proc=192):  37%|�
1: ��██▋      | 510000/1393783 [01:54<02:08, 6867.99 examples/s]
Tokenizing Prompts (num_proc=192):  37%|███▋      | 511000/1393783 [01:55<02:14, 6578.40 examples/s]
Tokenizing Prompts (num_proc=192):  37%|███▋      | 513000/1393783 [01:55<01:52, 7833.90 examples/s]
Tokenizing Prompts (num_proc=192):  37%|███▋      | 516000/1393783 [01:55<01:20, 10881.64 examples/s]
Tokenizing Prompts (num_proc=192):  37%|███▋      | 518000/1393783 [01:55<01:54, 7670.12 examples/s] 
Tokenizing Prompts (num_proc=192):  37%|███▋      | 519000/1393783 [01:56<01:51, 7864.26 examples/s]
Tokenizing Prompts (num_proc=192):  37%|███▋      | 520000/1393783 [01:56<02:30, 5798.99 examples/s]
Tokenizing Prompts (num_proc=192):  37%|███▋      | 522000/1393783 [01:56<02:05, 6934.29 examples/s]
Tokenizing Prompts (num_proc=192):  38%|███▊      | 523000/1393783 [01:56<02:21, 6170.48 examples/s]
Tokenizing Prompts (num_proc=192):  38%|███▊      | 525000/1393783 [01:57<
1: 02:00, 7190.13 examples/s]
Tokenizing Prompts (num_proc=192):  38%|███▊      | 527000/1393783 [01:57<01:41, 8519.90 examples/s]
Tokenizing Prompts (num_proc=192):  38%|███▊      | 528000/1393783 [01:57<01:42, 8413.85 examples/s]
Tokenizing Prompts (num_proc=192):  38%|███▊      | 530000/1393783 [01:57<02:00, 7164.93 examples/s]
Tokenizing Prompts (num_proc=192):  38%|███▊      | 531000/1393783 [01:57<02:25, 5915.41 examples/s]
Tokenizing Prompts (num_proc=192):  38%|███▊      | 533000/1393783 [01:58<02:00, 7162.64 examples/s]
Tokenizing Prompts (num_proc=192):  38%|███▊      | 535000/1393783 [01:58<01:40, 8583.10 examples/s]
Tokenizing Prompts (num_proc=192):  38%|███▊      | 536000/1393783 [01:58<02:44, 5206.86 examples/s]
Tokenizing Prompts (num_proc=192):  39%|███▊      | 539000/1393783 [01:59<02:38, 5386.29 examples/s]
Tokenizing Prompts (num_proc=192):  39%|███▊      | 540000/1393783 [01:59<02:35, 5501.02 examples/s]
Tokenizing Promp
1: ts (num_proc=192):  39%|███▉      | 541000/1393783 [01:59<03:08, 4533.38 examples/s]
Tokenizing Prompts (num_proc=192):  39%|███▉      | 542000/1393783 [02:00<03:38, 3891.42 examples/s]
Tokenizing Prompts (num_proc=192):  39%|███▉      | 544000/1393783 [02:00<02:44, 5159.76 examples/s]
Tokenizing Prompts (num_proc=192):  39%|███▉      | 545000/1393783 [02:00<02:50, 4964.49 examples/s]
Tokenizing Prompts (num_proc=192):  39%|███▉      | 546000/1393783 [02:00<02:52, 4904.15 examples/s]
Tokenizing Prompts (num_proc=192):  39%|███▉      | 548000/1393783 [02:01<02:23, 5913.27 examples/s]
Tokenizing Prompts (num_proc=192):  39%|███▉      | 550000/1393783 [02:01<01:58, 7097.76 examples/s]
Tokenizing Prompts (num_proc=192):  40%|███▉      | 552000/1393783 [02:01<01:55, 7270.31 examples/s]
Tokenizing Prompts (num_proc=192):  40%|███▉      | 553000/1393783 [02:01<01:50, 7579.39 examples/s]
Tokenizing Prompts (num_proc=192):  40%|███▉      |
1:  554000/1393783 [02:01<02:19, 6031.16 examples/s]
Tokenizing Prompts (num_proc=192):  40%|███▉      | 555000/1393783 [02:02<02:09, 6494.73 examples/s]
Tokenizing Prompts (num_proc=192):  40%|███▉      | 556000/1393783 [02:02<03:17, 4244.58 examples/s]
Tokenizing Prompts (num_proc=192):  40%|████      | 559000/1393783 [02:02<02:08, 6485.55 examples/s]
Tokenizing Prompts (num_proc=192):  40%|████      | 560000/1393783 [02:02<02:08, 6465.94 examples/s]
Tokenizing Prompts (num_proc=192):  40%|████      | 562000/1393783 [02:03<02:24, 5772.13 examples/s]
Tokenizing Prompts (num_proc=192):  40%|████      | 564000/1393783 [02:03<02:21, 5860.22 examples/s]
Tokenizing Prompts (num_proc=192):  41%|████      | 565000/1393783 [02:03<02:40, 5177.20 examples/s]
Tokenizing Prompts (num_proc=192):  41%|████      | 566000/1393783 [02:04<02:30, 5495.30 examples/s]
Tokenizing Prompts (num_proc=192):  41%|████      | 567000/1393783 [02:04<02:44, 5019.04 examp
1: les/s]
Tokenizing Prompts (num_proc=192):  41%|████      | 568000/1393783 [02:04<02:24, 5725.23 examples/s]
Tokenizing Prompts (num_proc=192):  41%|████      | 570000/1393783 [02:04<02:00, 6822.51 examples/s]
Tokenizing Prompts (num_proc=192):  41%|████      | 573000/1393783 [02:04<01:46, 7688.75 examples/s]
Tokenizing Prompts (num_proc=192):  41%|████▏     | 576000/1393783 [02:05<01:47, 7640.27 examples/s]
Tokenizing Prompts (num_proc=192):  41%|████▏     | 577000/1393783 [02:05<02:00, 6753.13 examples/s]
Tokenizing Prompts (num_proc=192):  42%|████▏     | 579000/1393783 [02:05<01:35, 8493.74 examples/s]
Tokenizing Prompts (num_proc=192):  42%|████▏     | 581000/1393783 [02:05<01:40, 8114.19 examples/s]
Tokenizing Prompts (num_proc=192):  42%|████▏     | 583000/1393783 [02:06<02:37, 5140.17 examples/s]
Tokenizing Prompts (num_proc=192):  42%|████▏     | 584000/1393783 [02:06<02:37, 5144.55 examples/s]
Tokenizing Prompts (num_
1: proc=192):  42%|████▏     | 587000/1393783 [02:07<02:07, 6345.98 examples/s]
Tokenizing Prompts (num_proc=192):  42%|████▏     | 589000/1393783 [02:07<01:53, 7066.01 examples/s]
Tokenizing Prompts (num_proc=192):  42%|████▏     | 591000/1393783 [02:07<01:34, 8502.46 examples/s]
Tokenizing Prompts (num_proc=192):  43%|████▎     | 593000/1393783 [02:08<02:18, 5766.96 examples/s]
Tokenizing Prompts (num_proc=192):  43%|████▎     | 595000/1393783 [02:08<01:52, 7097.34 examples/s]
Tokenizing Prompts (num_proc=192):  43%|████▎     | 597000/1393783 [02:08<02:04, 6401.37 examples/s]
Tokenizing Prompts (num_proc=192):  43%|████▎     | 599000/1393783 [02:09<02:38, 5010.90 examples/s]
Tokenizing Prompts (num_proc=192):  43%|████▎     | 601000/1393783 [02:09<02:46, 4770.86 examples/s]
Tokenizing Prompts (num_proc=192):  43%|████▎     | 602000/1393783 [02:09<02:47, 4734.37 examples/s]
Tokenizing Prompts (num_proc=192):  43%|███
1: █▎     | 604000/1393783 [02:10<02:06, 6224.83 examples/s]
Tokenizing Prompts (num_proc=192):  43%|████▎     | 605000/1393783 [02:10<02:53, 4558.41 examples/s]
Tokenizing Prompts (num_proc=192):  44%|████▎     | 608000/1393783 [02:10<01:56, 6744.71 examples/s]
Tokenizing Prompts (num_proc=192):  44%|████▎     | 609000/1393783 [02:11<02:28, 5285.63 examples/s]
Tokenizing Prompts (num_proc=192):  44%|████▍     | 613000/1393783 [02:11<01:30, 8609.92 examples/s]
Tokenizing Prompts (num_proc=192):  44%|████▍     | 615000/1393783 [02:11<01:35, 8166.21 examples/s]
Tokenizing Prompts (num_proc=192):  44%|████▍     | 617000/1393783 [02:12<02:26, 5291.55 examples/s]
Tokenizing Prompts (num_proc=192):  44%|████▍     | 618000/1393783 [02:12<02:16, 5700.29 examples/s]
Tokenizing Prompts (num_proc=192):  44%|████▍     | 619000/1393783 [02:12<02:25, 5324.13 examples/s]
Tokenizing Prompts (num_proc=192):  44%|████▍     | 620000/13937
1: 83 [02:13<03:41, 3489.31 examples/s]
Tokenizing Prompts (num_proc=192):  45%|████▍     | 621000/1393783 [02:13<03:15, 3947.52 examples/s]
Tokenizing Prompts (num_proc=192):  45%|████▍     | 623000/1393783 [02:13<03:02, 4229.38 examples/s]
Tokenizing Prompts (num_proc=192):  45%|████▍     | 625000/1393783 [02:14<02:45, 4637.05 examples/s]
Tokenizing Prompts (num_proc=192):  45%|████▍     | 626000/1393783 [02:14<03:20, 3830.11 examples/s]
Tokenizing Prompts (num_proc=192):  45%|████▍     | 627000/1393783 [02:14<02:59, 4274.83 examples/s]
Tokenizing Prompts (num_proc=192):  45%|████▌     | 628000/1393783 [02:14<02:36, 4878.28 examples/s]
Tokenizing Prompts (num_proc=192):  45%|████▌     | 629000/1393783 [02:15<02:49, 4505.94 examples/s]
Tokenizing Prompts (num_proc=192):  45%|████▌     | 630000/1393783 [02:15<02:49, 4503.31 examples/s]
Tokenizing Prompts (num_proc=192):  45%|████▌     | 631000/1393783 [02:15<02:48, 4529.98 
1: examples/s]
Tokenizing Prompts (num_proc=192):  45%|████▌     | 632000/1393783 [02:16<03:41, 3431.52 examples/s]
Tokenizing Prompts (num_proc=192):  45%|████▌     | 634000/1393783 [02:16<03:25, 3691.48 examples/s]
Tokenizing Prompts (num_proc=192):  46%|████▌     | 635000/1393783 [02:17<04:24, 2868.68 examples/s]
Tokenizing Prompts (num_proc=192):  46%|████▌     | 637000/1393783 [02:17<02:52, 4391.03 examples/s]
Tokenizing Prompts (num_proc=192):  46%|████▌     | 638000/1393783 [02:17<02:47, 4520.30 examples/s]
Tokenizing Prompts (num_proc=192):  46%|████▌     | 640000/1393783 [02:18<03:15, 3852.95 examples/s]
Tokenizing Prompts (num_proc=192):  46%|████▌     | 641000/1393783 [02:18<03:11, 3924.60 examples/s]
Tokenizing Prompts (num_proc=192):  46%|████▌     | 642000/1393783 [02:18<02:58, 4220.82 examples/s]
Tokenizing Prompts (num_proc=192):  46%|████▌     | 643000/1393783 [02:18<03:16, 3828.34 examples/s]
Tokenizing Pr
1: ompts (num_proc=192):  46%|████▌     | 644000/1393783 [02:18<02:49, 4432.07 examples/s]
Tokenizing Prompts (num_proc=192):  46%|████▋     | 645000/1393783 [02:19<02:49, 4413.28 examples/s]
Tokenizing Prompts (num_proc=192):  46%|████▋     | 646000/1393783 [02:19<02:33, 4856.75 examples/s]
Tokenizing Prompts (num_proc=192):  46%|████▋     | 647000/1393783 [02:20<04:41, 2655.01 examples/s]
Tokenizing Prompts (num_proc=192):  46%|████▋     | 648000/1393783 [02:20<03:41, 3362.19 examples/s]
Tokenizing Prompts (num_proc=192):  47%|████▋     | 649000/1393783 [02:20<03:01, 4106.76 examples/s]
Tokenizing Prompts (num_proc=192):  47%|████▋     | 650000/1393783 [02:20<02:48, 4401.82 examples/s]
Tokenizing Prompts (num_proc=192):  47%|████▋     | 651000/1393783 [02:20<02:22, 5195.94 examples/s]
Tokenizing Prompts (num_proc=192):  47%|████▋     | 653000/1393783 [02:21<02:33, 4811.66 examples/s]
Tokenizing Prompts (num_proc=192):  47
1: %|████▋     | 654000/1393783 [02:21<02:22, 5173.65 examples/s]
Tokenizing Prompts (num_proc=192):  47%|████▋     | 656000/1393783 [02:21<02:23, 5139.37 examples/s]
Tokenizing Prompts (num_proc=192):  47%|████▋     | 657000/1393783 [02:22<02:51, 4299.07 examples/s]
Tokenizing Prompts (num_proc=192):  47%|████▋     | 659000/1393783 [02:22<02:06, 5814.84 examples/s]
Tokenizing Prompts (num_proc=192):  47%|████▋     | 660000/1393783 [02:22<02:59, 4093.90 examples/s]
Tokenizing Prompts (num_proc=192):  47%|████▋     | 661000/1393783 [02:22<02:35, 4721.07 examples/s]
Tokenizing Prompts (num_proc=192):  47%|████▋     | 662000/1393783 [02:23<03:01, 4034.08 examples/s]
Tokenizing Prompts (num_proc=192):  48%|████▊     | 663000/1393783 [02:23<03:45, 3239.98 examples/s]
Tokenizing Prompts (num_proc=192):  48%|████▊     | 665000/1393783 [02:23<02:49, 4304.32 examples/s]
Tokenizing Prompts (num_proc=192):  48%|████▊     | 6
1: 66000/1393783 [02:24<02:31, 4790.49 examples/s]
Tokenizing Prompts (num_proc=192):  48%|████▊     | 668000/1393783 [02:24<01:51, 6522.54 examples/s]
Tokenizing Prompts (num_proc=192):  48%|████▊     | 670000/1393783 [02:24<01:33, 7774.16 examples/s]
Tokenizing Prompts (num_proc=192):  48%|████▊     | 671000/1393783 [02:24<01:49, 6588.52 examples/s]
Tokenizing Prompts (num_proc=192):  48%|████▊     | 672000/1393783 [02:24<02:15, 5311.23 examples/s]
Tokenizing Prompts (num_proc=192):  48%|████▊     | 673000/1393783 [02:25<02:13, 5389.10 examples/s]
Tokenizing Prompts (num_proc=192):  48%|████▊     | 674000/1393783 [02:25<02:32, 4729.16 examples/s]
Tokenizing Prompts (num_proc=192):  49%|████▊     | 676000/1393783 [02:25<01:54, 6250.31 examples/s]
Tokenizing Prompts (num_proc=192):  49%|████▊     | 678000/1393783 [02:25<01:37, 7375.71 examples/s]
Tokenizing Prompts (num_proc=192):  49%|████▉     | 680000/1393783 [02:25<01:2
1: 3, 8532.31 examples/s]
Tokenizing Prompts (num_proc=192):  49%|████▉     | 681000/1393783 [02:26<02:32, 4659.86 examples/s]
Tokenizing Prompts (num_proc=192):  49%|████▉     | 682000/1393783 [02:26<02:17, 5192.37 examples/s]
Tokenizing Prompts (num_proc=192):  49%|████▉     | 684000/1393783 [02:26<02:11, 5406.98 examples/s]
Tokenizing Prompts (num_proc=192):  49%|████▉     | 685000/1393783 [02:27<02:22, 4988.39 examples/s]
Tokenizing Prompts (num_proc=192):  49%|████▉     | 688000/1393783 [02:27<01:29, 7894.63 examples/s]
Tokenizing Prompts (num_proc=192):  50%|████▉     | 690000/1393783 [02:27<01:44, 6752.60 examples/s]
Tokenizing Prompts (num_proc=192):  50%|████▉     | 692000/1393783 [02:28<01:40, 7017.16 examples/s]
Tokenizing Prompts (num_proc=192):  50%|████▉     | 695000/1393783 [02:28<01:27, 8027.48 examples/s]
Tokenizing Prompts (num_proc=192):  50%|█████     | 697000/1393783 [02:28<01:21, 8601.04 examples/s]
To
1: kenizing Prompts (num_proc=192):  50%|█████     | 698000/1393783 [02:28<01:37, 7160.81 examples/s]
Tokenizing Prompts (num_proc=192):  50%|█████     | 699000/1393783 [02:28<01:45, 6579.71 examples/s]
Tokenizing Prompts (num_proc=192):  50%|█████     | 700000/1393783 [02:29<02:03, 5630.55 examples/s]
Tokenizing Prompts (num_proc=192):  50%|█████     | 701000/1393783 [02:29<01:56, 5930.31 examples/s]
Tokenizing Prompts (num_proc=192):  50%|█████     | 702000/1393783 [02:29<03:00, 3823.98 examples/s]
Tokenizing Prompts (num_proc=192):  50%|█████     | 703000/1393783 [02:30<02:35, 4443.11 examples/s]
Tokenizing Prompts (num_proc=192):  51%|█████     | 706000/1393783 [02:30<01:38, 6979.49 examples/s]
Tokenizing Prompts (num_proc=192):  51%|█████     | 707000/1393783 [02:30<01:50, 6206.06 examples/s]
Tokenizing Prompts (num_proc=192):  51%|█████     | 709000/1393783 [02:31<02:57, 3852.23 examples/s]
Tokenizing Prompts (num_pro
1: c=192):  51%|█████     | 710000/1393783 [02:31<03:19, 3432.41 examples/s]
Tokenizing Prompts (num_proc=192):  51%|█████     | 712000/1393783 [02:31<02:17, 4950.37 examples/s]
Tokenizing Prompts (num_proc=192):  51%|█████     | 713000/1393783 [02:31<02:04, 5487.03 examples/s]
Tokenizing Prompts (num_proc=192):  51%|█████▏    | 715000/1393783 [02:32<01:33, 7248.79 examples/s]
Tokenizing Prompts (num_proc=192):  52%|█████▏    | 718000/1393783 [02:32<01:36, 7026.06 examples/s]
Tokenizing Prompts (num_proc=192):  52%|█████▏    | 719000/1393783 [02:32<01:40, 6704.78 examples/s]
Tokenizing Prompts (num_proc=192):  52%|█████▏    | 721000/1393783 [02:33<01:38, 6839.42 examples/s]
Tokenizing Prompts (num_proc=192):  52%|█████▏    | 722000/1393783 [02:33<02:09, 5197.95 examples/s]
Tokenizing Prompts (num_proc=192):  52%|█████▏    | 723000/1393783 [02:33<01:56, 5739.08 examples/s]
Tokenizing Prompts (num_proc=192):  52%|
1: █████▏    | 724000/1393783 [02:33<02:17, 4875.17 examples/s]
Tokenizing Prompts (num_proc=192):  52%|█████▏    | 725000/1393783 [02:34<03:21, 3315.78 examples/s]
Tokenizing Prompts (num_proc=192):  52%|█████▏    | 726000/1393783 [02:34<02:59, 3727.01 examples/s]
Tokenizing Prompts (num_proc=192):  52%|█████▏    | 727000/1393783 [02:34<03:05, 3592.39 examples/s]
Tokenizing Prompts (num_proc=192):  52%|█████▏    | 728000/1393783 [02:34<02:32, 4371.55 examples/s]
Tokenizing Prompts (num_proc=192):  52%|█████▏    | 731000/1393783 [02:35<01:24, 7841.21 examples/s]
Tokenizing Prompts (num_proc=192):  53%|█████▎    | 733000/1393783 [02:35<01:30, 7311.48 examples/s]
Tokenizing Prompts (num_proc=192):  53%|█████▎    | 734000/1393783 [02:35<01:55, 5698.01 examples/s]
Tokenizing Prompts (num_proc=192):  53%|█████▎    | 735000/1393783 [02:35<01:59, 5502.92 examples/s]
Tokenizing Prompts (num_proc=192):  53%|██�
1: ��██▎    | 736000/1393783 [02:36<02:24, 4546.22 examples/s]
Tokenizing Prompts (num_proc=192):  53%|█████▎    | 738000/1393783 [02:36<01:50, 5943.24 examples/s]
Tokenizing Prompts (num_proc=192):  53%|█████▎    | 740000/1393783 [02:36<02:04, 5236.65 examples/s]
Tokenizing Prompts (num_proc=192):  53%|█████▎    | 741000/1393783 [02:37<01:53, 5762.48 examples/s]
Tokenizing Prompts (num_proc=192):  53%|█████▎    | 743000/1393783 [02:37<01:23, 7779.59 examples/s]
Tokenizing Prompts (num_proc=192):  53%|█████▎    | 745000/1393783 [02:37<01:36, 6725.46 examples/s]
Tokenizing Prompts (num_proc=192):  54%|█████▎    | 746000/1393783 [02:37<02:02, 5285.18 examples/s]
Tokenizing Prompts (num_proc=192):  54%|█████▎    | 748000/1393783 [02:38<01:33, 6903.72 examples/s]
Tokenizing Prompts (num_proc=192):  54%|█████▎    | 749000/1393783 [02:38<01:28, 7323.31 examples/s]
Tokenizing Prompts (num_proc=192):  54%|████�
1: �▍    | 750000/1393783 [02:38<01:37, 6628.21 examples/s]
Tokenizing Prompts (num_proc=192):  54%|█████▍    | 751000/1393783 [02:39<03:16, 3268.63 examples/s]
Tokenizing Prompts (num_proc=192):  54%|█████▍    | 752000/1393783 [02:39<04:02, 2649.44 examples/s]
Tokenizing Prompts (num_proc=192):  54%|█████▍    | 753000/1393783 [02:39<03:45, 2847.31 examples/s]
Tokenizing Prompts (num_proc=192):  54%|█████▍    | 754000/1393783 [02:40<03:28, 3069.88 examples/s]
Tokenizing Prompts (num_proc=192):  54%|█████▍    | 755000/1393783 [02:40<03:05, 3446.87 examples/s]
Tokenizing Prompts (num_proc=192):  54%|█████▍    | 756000/1393783 [02:40<02:51, 3727.83 examples/s]
Tokenizing Prompts (num_proc=192):  54%|█████▍    | 757000/1393783 [02:40<02:50, 3724.94 examples/s]
Tokenizing Prompts (num_proc=192):  54%|█████▍    | 759000/1393783 [02:41<01:58, 5372.30 examples/s]
Tokenizing Prompts (num_proc=192):  55%|█████▍   
1:  | 760000/1393783 [02:41<01:56, 5436.12 examples/s]
Tokenizing Prompts (num_proc=192):  55%|█████▍    | 761000/1393783 [02:41<02:07, 4954.50 examples/s]
Tokenizing Prompts (num_proc=192):  55%|█████▍    | 762000/1393783 [02:41<02:05, 5047.79 examples/s]
Tokenizing Prompts (num_proc=192):  55%|█████▍    | 765000/1393783 [02:41<01:27, 7151.98 examples/s]
Tokenizing Prompts (num_proc=192):  55%|█████▌    | 767000/1393783 [02:42<01:13, 8517.28 examples/s]
Tokenizing Prompts (num_proc=192):  55%|█████▌    | 768000/1393783 [02:42<02:10, 4812.18 examples/s]
Tokenizing Prompts (num_proc=192):  55%|█████▌    | 770000/1393783 [02:42<01:57, 5328.53 examples/s]
Tokenizing Prompts (num_proc=192):  55%|█████▌    | 772000/1393783 [02:43<02:11, 4743.95 examples/s]
Tokenizing Prompts (num_proc=192):  55%|█████▌    | 773000/1393783 [02:43<02:33, 4035.54 examples/s]
Tokenizing Prompts (num_proc=192):  56%|█████▌    | 7740
1: 00/1393783 [02:44<02:58, 3470.13 examples/s]
Tokenizing Prompts (num_proc=192):  56%|█████▌    | 777000/1393783 [02:44<02:26, 4218.70 examples/s]
Tokenizing Prompts (num_proc=192):  56%|█████▌    | 778000/1393783 [02:45<02:47, 3681.55 examples/s]
Tokenizing Prompts (num_proc=192):  56%|█████▌    | 779000/1393783 [02:45<02:34, 3972.95 examples/s]
Tokenizing Prompts (num_proc=192):  56%|█████▌    | 781000/1393783 [02:45<01:56, 5253.16 examples/s]
Tokenizing Prompts (num_proc=192):  56%|█████▌    | 782000/1393783 [02:45<01:49, 5600.58 examples/s]
Tokenizing Prompts (num_proc=192):  56%|█████▌    | 783000/1393783 [02:46<02:10, 4684.25 examples/s]
Tokenizing Prompts (num_proc=192):  56%|█████▌    | 784000/1393783 [02:46<02:29, 4066.58 examples/s]
Tokenizing Prompts (num_proc=192):  56%|█████▋    | 786000/1393783 [02:46<01:57, 5190.52 examples/s]
Tokenizing Prompts (num_proc=192):  56%|█████▋    | 787000/1393
1: 783 [02:46<01:46, 5676.63 examples/s]
Tokenizing Prompts (num_proc=192):  57%|█████▋    | 789000/1393783 [02:47<01:25, 7032.52 examples/s]
Tokenizing Prompts (num_proc=192):  57%|█████▋    | 792000/1393783 [02:47<01:04, 9321.58 examples/s]
Tokenizing Prompts (num_proc=192):  57%|█████▋    | 794000/1393783 [02:47<01:14, 8081.38 examples/s]
Tokenizing Prompts (num_proc=192):  57%|█████▋    | 795000/1393783 [02:47<01:21, 7387.18 examples/s]
Tokenizing Prompts (num_proc=192):  57%|█████▋    | 797000/1393783 [02:47<01:12, 8226.76 examples/s]
Tokenizing Prompts (num_proc=192):  57%|█████▋    | 798000/1393783 [02:48<01:44, 5725.71 examples/s]
Tokenizing Prompts (num_proc=192):  57%|█████▋    | 799000/1393783 [02:48<01:52, 5265.09 examples/s]
Tokenizing Prompts (num_proc=192):  57%|█████▋    | 801000/1393783 [02:48<01:26, 6853.75 examples/s]
Tokenizing Prompts (num_proc=192):  58%|█████▊    | 802000/1393783 [02
1: :48<01:36, 6154.80 examples/s]
Tokenizing Prompts (num_proc=192):  58%|█████▊    | 803000/1393783 [02:49<01:47, 5513.35 examples/s]
Tokenizing Prompts (num_proc=192):  58%|█████▊    | 804000/1393783 [02:49<01:35, 6206.99 examples/s]
Tokenizing Prompts (num_proc=192):  58%|█████▊    | 805000/1393783 [02:49<01:28, 6688.97 examples/s]
Tokenizing Prompts (num_proc=192):  58%|█████▊    | 806000/1393783 [02:49<01:32, 6364.70 examples/s]
Tokenizing Prompts (num_proc=192):  58%|█████▊    | 807000/1393783 [02:49<02:04, 4694.76 examples/s]
Tokenizing Prompts (num_proc=192):  58%|█████▊    | 809000/1393783 [02:50<01:25, 6816.05 examples/s]
Tokenizing Prompts (num_proc=192):  58%|█████▊    | 813000/1393783 [02:50<00:47, 12113.90 examples/s]
Tokenizing Prompts (num_proc=192):  58%|█████▊    | 815000/1393783 [02:50<00:47, 12191.21 examples/s]
Tokenizing Prompts (num_proc=192):  59%|█████▊    | 818000/1393783 [02:50<0
1: 0:46, 12392.70 examples/s]
Tokenizing Prompts (num_proc=192):  59%|█████▉    | 821000/1393783 [02:50<00:43, 13305.29 examples/s]
Tokenizing Prompts (num_proc=192):  59%|█████▉    | 823000/1393783 [02:51<00:51, 11128.35 examples/s]
Tokenizing Prompts (num_proc=192):  59%|█████▉    | 825000/1393783 [02:51<01:33, 6086.43 examples/s] 
Tokenizing Prompts (num_proc=192):  59%|█████▉    | 826000/1393783 [02:52<02:07, 4460.71 examples/s]
Tokenizing Prompts (num_proc=192):  59%|█████▉    | 828000/1393783 [02:52<01:47, 5242.81 examples/s]
Tokenizing Prompts (num_proc=192):  60%|█████▉    | 830000/1393783 [02:52<01:34, 5963.49 examples/s]
Tokenizing Prompts (num_proc=192):  60%|█████▉    | 831000/1393783 [02:53<02:07, 4426.99 examples/s]
Tokenizing Prompts (num_proc=192):  60%|█████▉    | 832000/1393783 [02:53<01:54, 4908.69 examples/s]
Tokenizing Prompts (num_proc=192):  60%|█████▉    | 833000/1393783 [02:53<01:4
1: 2, 5497.45 examples/s]
Tokenizing Prompts (num_proc=192):  60%|█████▉    | 834000/1393783 [02:53<01:40, 5566.68 examples/s]
Tokenizing Prompts (num_proc=192):  60%|█████▉    | 836000/1393783 [02:54<01:58, 4688.83 examples/s]
Tokenizing Prompts (num_proc=192):  60%|██████    | 837000/1393783 [02:54<02:24, 3855.16 examples/s]
Tokenizing Prompts (num_proc=192):  60%|██████    | 838000/1393783 [02:54<02:04, 4459.07 examples/s]
Tokenizing Prompts (num_proc=192):  60%|██████    | 840000/1393783 [02:54<01:39, 5580.10 examples/s]
Tokenizing Prompts (num_proc=192):  60%|██████    | 841000/1393783 [02:55<01:32, 5981.82 examples/s]
Tokenizing Prompts (num_proc=192):  60%|██████    | 842000/1393783 [02:55<03:01, 3042.89 examples/s]
Tokenizing Prompts (num_proc=192):  61%|██████    | 844000/1393783 [02:56<02:41, 3408.71 examples/s]
Tokenizing Prompts (num_proc=192):  61%|██████    | 846000/1393783 [02:56<01:57, 4646
1: .27 examples/s]
Tokenizing Prompts (num_proc=192):  61%|██████    | 847000/1393783 [02:56<01:51, 4900.71 examples/s]
Tokenizing Prompts (num_proc=192):  61%|██████    | 848000/1393783 [02:58<04:16, 2129.92 examples/s]
Tokenizing Prompts (num_proc=192):  61%|██████    | 850000/1393783 [02:58<03:02, 2986.55 examples/s]
Tokenizing Prompts (num_proc=192):  61%|██████    | 851000/1393783 [02:58<02:43, 3319.45 examples/s]
Tokenizing Prompts (num_proc=192):  61%|██████    | 853000/1393783 [02:58<02:00, 4481.79 examples/s]
Tokenizing Prompts (num_proc=192):  61%|██████▏   | 854000/1393783 [02:58<02:03, 4383.05 examples/s]
Tokenizing Prompts (num_proc=192):  61%|██████▏   | 855000/1393783 [02:59<01:46, 5050.96 examples/s]
Tokenizing Prompts (num_proc=192):  61%|██████▏   | 856000/1393783 [02:59<01:51, 4811.00 examples/s]
Tokenizing Prompts (num_proc=192):  61%|██████▏   | 857000/1393783 [02:59<02:11, 409
1: 2.77 examples/s]
Tokenizing Prompts (num_proc=192):  62%|██████▏   | 859000/1393783 [03:00<01:59, 4489.36 examples/s]
Tokenizing Prompts (num_proc=192):  62%|██████▏   | 861000/1393783 [03:00<01:29, 5961.20 examples/s]
Tokenizing Prompts (num_proc=192):  62%|██████▏   | 863000/1393783 [03:00<01:19, 6656.98 examples/s]
Tokenizing Prompts (num_proc=192):  62%|██████▏   | 865000/1393783 [03:00<01:09, 7596.40 examples/s]
Tokenizing Prompts (num_proc=192):  62%|██████▏   | 866000/1393783 [03:01<01:36, 5446.16 examples/s]
Tokenizing Prompts (num_proc=192):  62%|██████▏   | 867000/1393783 [03:01<02:22, 3709.00 examples/s]
Tokenizing Prompts (num_proc=192):  62%|██████▏   | 868000/1393783 [03:02<02:41, 3248.18 examples/s]
Tokenizing Prompts (num_proc=192):  62%|██████▏   | 870000/1393783 [03:02<01:51, 4699.94 examples/s]
Tokenizing Prompts (num_proc=192):  63%|██████▎   | 872000/1393783 [03:02
1: <01:22, 6291.68 examples/s]
Tokenizing Prompts (num_proc=192):  63%|██████▎   | 873000/1393783 [03:02<01:36, 5413.84 examples/s]
Tokenizing Prompts (num_proc=192):  63%|██████▎   | 874000/1393783 [03:03<02:42, 3200.73 examples/s]
Tokenizing Prompts (num_proc=192):  63%|██████▎   | 875000/1393783 [03:03<02:41, 3203.93 examples/s]
Tokenizing Prompts (num_proc=192):  63%|██████▎   | 878000/1393783 [03:03<01:33, 5535.27 examples/s]
Tokenizing Prompts (num_proc=192):  63%|██████▎   | 879000/1393783 [03:04<01:52, 4587.97 examples/s]
Tokenizing Prompts (num_proc=192):  63%|██████▎   | 880000/1393783 [03:04<01:57, 4382.17 examples/s]
Tokenizing Prompts (num_proc=192):  63%|██████▎   | 881000/1393783 [03:05<02:36, 3270.70 examples/s]
Tokenizing Prompts (num_proc=192):  63%|██████▎   | 883000/1393783 [03:05<01:55, 4430.81 examples/s]
Tokenizing Prompts (num_proc=192):  63%|██████▎   | 884000/139
1: 3783 [03:05<01:45, 4830.48 examples/s]
Tokenizing Prompts (num_proc=192):  63%|██████▎   | 885000/1393783 [03:05<01:58, 4276.51 examples/s]
Tokenizing Prompts (num_proc=192):  64%|██████▎   | 886000/1393783 [03:05<01:53, 4465.39 examples/s]
Tokenizing Prompts (num_proc=192):  64%|██████▎   | 887000/1393783 [03:06<01:53, 4475.39 examples/s]
Tokenizing Prompts (num_proc=192):  64%|██████▍   | 889000/1393783 [03:06<01:15, 6687.69 examples/s]
Tokenizing Prompts (num_proc=192):  64%|██████▍   | 890000/1393783 [03:06<01:35, 5276.20 examples/s]
Tokenizing Prompts (num_proc=192):  64%|██████▍   | 891000/1393783 [03:06<01:24, 5944.20 examples/s]
Tokenizing Prompts (num_proc=192):  64%|██████▍   | 892000/1393783 [03:06<01:31, 5466.25 examples/s]
Tokenizing Prompts (num_proc=192):  64%|██████▍   | 894000/1393783 [03:07<01:10, 7075.13 examples/s]
Tokenizing Prompts (num_proc=192):  64%|██████▍   |
1:  896000/1393783 [03:07<01:11, 6945.94 examples/s]
Tokenizing Prompts (num_proc=192):  64%|██████▍   | 897000/1393783 [03:07<01:44, 4757.37 examples/s]
Tokenizing Prompts (num_proc=192):  65%|██████▍   | 899000/1393783 [03:08<01:29, 5520.84 examples/s]
Tokenizing Prompts (num_proc=192):  65%|██████▍   | 900000/1393783 [03:08<01:22, 5949.24 examples/s]
Tokenizing Prompts (num_proc=192):  65%|██████▍   | 901000/1393783 [03:08<01:23, 5895.12 examples/s]
Tokenizing Prompts (num_proc=192):  65%|██████▍   | 902000/1393783 [03:08<01:30, 5441.67 examples/s]
Tokenizing Prompts (num_proc=192):  65%|██████▍   | 904000/1393783 [03:08<01:26, 5690.51 examples/s]
Tokenizing Prompts (num_proc=192):  65%|██████▍   | 905000/1393783 [03:08<01:17, 6282.72 examples/s]
Tokenizing Prompts (num_proc=192):  65%|██████▌   | 906000/1393783 [03:09<01:45, 4636.45 examples/s]
Tokenizing Prompts (num_proc=192):  65%|████�
1: �█▌   | 909000/1393783 [03:09<01:00, 8008.26 examples/s]
Tokenizing Prompts (num_proc=192):  65%|██████▌   | 911000/1393783 [03:10<01:25, 5615.81 examples/s]
Tokenizing Prompts (num_proc=192):  65%|██████▌   | 912000/1393783 [03:10<01:47, 4461.19 examples/s]
Tokenizing Prompts (num_proc=192):  66%|██████▌   | 914000/1393783 [03:10<01:28, 5428.73 examples/s]
Tokenizing Prompts (num_proc=192):  66%|██████▌   | 915000/1393783 [03:10<01:22, 5813.17 examples/s]
Tokenizing Prompts (num_proc=192):  66%|██████▌   | 917000/1393783 [03:11<01:12, 6557.81 examples/s]
Tokenizing Prompts (num_proc=192):  66%|██████▌   | 918000/1393783 [03:11<01:29, 5298.94 examples/s]
Tokenizing Prompts (num_proc=192):  66%|██████▌   | 920000/1393783 [03:11<01:14, 6345.76 examples/s]
Tokenizing Prompts (num_proc=192):  66%|██████▌   | 921000/1393783 [03:11<01:13, 6435.03 examples/s]
Tokenizing Prompts (num_proc=192):  66%|█
1: █████▌   | 922000/1393783 [03:11<01:15, 6243.17 examples/s]
Tokenizing Prompts (num_proc=192):  66%|██████▌   | 923000/1393783 [03:12<01:28, 5347.73 examples/s]
Tokenizing Prompts (num_proc=192):  66%|██████▋   | 924000/1393783 [03:12<01:21, 5763.23 examples/s]
Tokenizing Prompts (num_proc=192):  66%|██████▋   | 926000/1393783 [03:12<01:05, 7103.27 examples/s]
Tokenizing Prompts (num_proc=192):  67%|██████▋   | 928000/1393783 [03:12<00:51, 9059.13 examples/s]
Tokenizing Prompts (num_proc=192):  67%|██████▋   | 930000/1393783 [03:13<01:07, 6887.43 examples/s]
Tokenizing Prompts (num_proc=192):  67%|██████▋   | 931000/1393783 [03:13<01:09, 6674.96 examples/s]
Tokenizing Prompts (num_proc=192):  67%|██████▋   | 932000/1393783 [03:13<01:59, 3873.15 examples/s]
Tokenizing Prompts (num_proc=192):  67%|██████▋   | 935000/1393783 [03:14<01:35, 4799.23 examples/s]
Tokenizing Prompts (num_proc=192
1: ):  67%|██████▋   | 937000/1393783 [03:14<01:42, 4435.93 examples/s]
Tokenizing Prompts (num_proc=192):  67%|██████▋   | 938000/1393783 [03:15<01:39, 4561.39 examples/s]
Tokenizing Prompts (num_proc=192):  67%|██████▋   | 939000/1393783 [03:15<01:35, 4778.75 examples/s]
Tokenizing Prompts (num_proc=192):  67%|██████▋   | 940000/1393783 [03:15<01:44, 4347.13 examples/s]
Tokenizing Prompts (num_proc=192):  68%|██████▊   | 941000/1393783 [03:15<01:42, 4424.52 examples/s]
Tokenizing Prompts (num_proc=192):  68%|██████▊   | 942000/1393783 [03:16<02:35, 2904.73 examples/s]
Tokenizing Prompts (num_proc=192):  68%|██████▊   | 944000/1393783 [03:17<02:24, 3103.48 examples/s]
Tokenizing Prompts (num_proc=192):  68%|██████▊   | 945000/1393783 [03:17<02:02, 3666.72 examples/s]
Tokenizing Prompts (num_proc=192):  68%|██████▊   | 947000/1393783 [03:17<01:29, 4977.54 examples/s]
Tokenizing Prompts (n
1: um_proc=192):  68%|██████▊   | 948000/1393783 [03:17<01:58, 3775.23 examples/s]
Tokenizing Prompts (num_proc=192):  68%|██████▊   | 950000/1393783 [03:18<01:34, 4688.00 examples/s]
Tokenizing Prompts (num_proc=192):  68%|██████▊   | 952000/1393783 [03:18<01:14, 5908.74 examples/s]
Tokenizing Prompts (num_proc=192):  69%|██████▊   | 955000/1393783 [03:18<00:54, 8046.73 examples/s]
Tokenizing Prompts (num_proc=192):  69%|██████▊   | 957000/1393783 [03:18<01:12, 6033.31 examples/s]
Tokenizing Prompts (num_proc=192):  69%|██████▊   | 958000/1393783 [03:19<01:43, 4209.39 examples/s]
Tokenizing Prompts (num_proc=192):  69%|██████▉   | 962000/1393783 [03:19<00:59, 7219.30 examples/s]
Tokenizing Prompts (num_proc=192):  69%|██████▉   | 964000/1393783 [03:20<01:15, 5710.26 examples/s]
Tokenizing Prompts (num_proc=192):  69%|██████▉   | 966000/1393783 [03:20<01:06, 6473.31 examples/s]
Tokenizing
1:  Prompts (num_proc=192):  69%|██████▉   | 967000/1393783 [03:21<02:00, 3553.38 examples/s]
Tokenizing Prompts (num_proc=192):  69%|██████▉   | 968000/1393783 [03:21<01:46, 3982.71 examples/s]
Tokenizing Prompts (num_proc=192):  70%|██████▉   | 970000/1393783 [03:21<01:20, 5240.11 examples/s]
Tokenizing Prompts (num_proc=192):  70%|██████▉   | 971000/1393783 [03:22<01:38, 4278.35 examples/s]
Tokenizing Prompts (num_proc=192):  70%|██████▉   | 974000/1393783 [03:22<01:09, 6026.48 examples/s]
Tokenizing Prompts (num_proc=192):  70%|██████▉   | 975000/1393783 [03:22<01:13, 5734.04 examples/s]
Tokenizing Prompts (num_proc=192):  70%|███████   | 976000/1393783 [03:22<01:27, 4783.44 examples/s]
Tokenizing Prompts (num_proc=192):  70%|███████   | 979000/1393783 [03:23<01:15, 5494.52 examples/s]
Tokenizing Prompts (num_proc=192):  70%|███████   | 980000/1393783 [03:23<01:17, 5333.10 examples/s]
1: 
Tokenizing Prompts (num_proc=192):  70%|███████   | 981000/1393783 [03:23<01:12, 5692.44 examples/s]
Tokenizing Prompts (num_proc=192):  70%|███████   | 982000/1393783 [03:23<01:22, 5007.59 examples/s]
Tokenizing Prompts (num_proc=192):  71%|███████   | 983000/1393783 [03:24<01:26, 4752.33 examples/s]
Tokenizing Prompts (num_proc=192):  71%|███████   | 985000/1393783 [03:24<01:10, 5760.92 examples/s]
Tokenizing Prompts (num_proc=192):  71%|███████   | 987000/1393783 [03:24<00:56, 7257.47 examples/s]
Tokenizing Prompts (num_proc=192):  71%|███████   | 990000/1393783 [03:24<00:48, 8407.73 examples/s]
Tokenizing Prompts (num_proc=192):  71%|███████   | 991000/1393783 [03:25<00:55, 7222.47 examples/s]
Tokenizing Prompts (num_proc=192):  71%|███████   | 992000/1393783 [03:25<01:42, 3919.85 examples/s]
Tokenizing Prompts (num_proc=192):  71%|███████   | 993000/1393783 [03:25<01:31, 4385.60 
1: examples/s]
Tokenizing Prompts (num_proc=192):  71%|███████▏  | 995000/1393783 [03:26<01:08, 5838.46 examples/s]
Tokenizing Prompts (num_proc=192):  72%|███████▏  | 997000/1393783 [03:26<00:58, 6756.79 examples/s]
Tokenizing Prompts (num_proc=192):  72%|███████▏  | 998000/1393783 [03:26<00:57, 6863.74 examples/s]
Tokenizing Prompts (num_proc=192):  72%|███████▏  | 999000/1393783 [03:27<01:38, 4019.49 examples/s]
Tokenizing Prompts (num_proc=192):  72%|███████▏  | 1000000/1393783 [03:27<01:42, 3846.28 examples/s]
Tokenizing Prompts (num_proc=192):  72%|███████▏  | 1003000/1393783 [03:27<01:18, 4954.23 examples/s]
Tokenizing Prompts (num_proc=192):  72%|███████▏  | 1006000/1393783 [03:28<01:08, 5651.54 examples/s]
Tokenizing Prompts (num_proc=192):  72%|███████▏  | 1007000/1393783 [03:28<01:11, 5374.28 examples/s]
Tokenizing Prompts (num_proc=192):  72%|███████▏  | 1008
1: 000/1393783 [03:28<01:08, 5596.22 examples/s]
Tokenizing Prompts (num_proc=192):  73%|███████▎  | 1011000/1393783 [03:29<00:58, 6596.08 examples/s]
Tokenizing Prompts (num_proc=192):  73%|███████▎  | 1012000/1393783 [03:29<00:58, 6476.69 examples/s]
Tokenizing Prompts (num_proc=192):  73%|███████▎  | 1013000/1393783 [03:29<01:24, 4495.24 examples/s]
Tokenizing Prompts (num_proc=192):  73%|███████▎  | 1015000/1393783 [03:29<01:16, 4979.15 examples/s]
Tokenizing Prompts (num_proc=192):  73%|███████▎  | 1016000/1393783 [03:30<01:31, 4133.73 examples/s]
Tokenizing Prompts (num_proc=192):  73%|███████▎  | 1017000/1393783 [03:30<01:20, 4683.01 examples/s]
Tokenizing Prompts (num_proc=192):  73%|███████▎  | 1018000/1393783 [03:30<01:32, 4077.74 examples/s]
Tokenizing Prompts (num_proc=192):  73%|███████▎  | 1019000/1393783 [03:31<01:26, 4335.79 examples/s]
Tokenizing Prompts (num_proc=192):
1:   73%|███████▎  | 1022000/1393783 [03:31<01:01, 6078.34 examples/s]
Tokenizing Prompts (num_proc=192):  73%|███████▎  | 1024000/1393783 [03:31<00:59, 6194.51 examples/s]
Tokenizing Prompts (num_proc=192):  74%|███████▎  | 1027000/1393783 [03:31<00:47, 7660.78 examples/s]
Tokenizing Prompts (num_proc=192):  74%|███████▍  | 1031000/1393783 [03:32<00:32, 11158.92 examples/s]
Tokenizing Prompts (num_proc=192):  74%|███████▍  | 1033000/1393783 [03:32<00:34, 10416.30 examples/s]
Tokenizing Prompts (num_proc=192):  74%|███████▍  | 1036000/1393783 [03:32<00:35, 10130.38 examples/s]
Tokenizing Prompts (num_proc=192):  74%|███████▍  | 1038000/1393783 [03:32<00:34, 10416.28 examples/s]
Tokenizing Prompts (num_proc=192):  75%|███████▍  | 1040000/1393783 [03:32<00:34, 10305.09 examples/s]
Tokenizing Prompts (num_proc=192):  75%|███████▍  | 1043000/1393783 [03:33<00:33, 10370.89 ex
1: amples/s]
Tokenizing Prompts (num_proc=192):  75%|███████▍  | 1045000/1393783 [03:33<00:46, 7570.04 examples/s] 
Tokenizing Prompts (num_proc=192):  75%|███████▌  | 1046000/1393783 [03:33<00:46, 7405.44 examples/s]
Tokenizing Prompts (num_proc=192):  75%|███████▌  | 1047000/1393783 [03:34<00:53, 6444.02 examples/s]
Tokenizing Prompts (num_proc=192):  75%|███████▌  | 1049000/1393783 [03:34<00:43, 7970.04 examples/s]
Tokenizing Prompts (num_proc=192):  75%|███████▌  | 1051000/1393783 [03:34<00:52, 6509.67 examples/s]
Tokenizing Prompts (num_proc=192):  75%|███████▌  | 1052000/1393783 [03:35<01:32, 3686.55 examples/s]
Tokenizing Prompts (num_proc=192):  76%|███████▌  | 1053000/1393783 [03:35<01:24, 4048.73 examples/s]
Tokenizing Prompts (num_proc=192):  76%|███████▌  | 1055000/1393783 [03:35<01:00, 5612.30 examples/s]
Tokenizing Prompts (num_proc=192):  76%|███████▌  | 1
1: 056000/1393783 [03:36<01:09, 4867.03 examples/s]
Tokenizing Prompts (num_proc=192):  76%|███████▌  | 1057000/1393783 [03:36<01:03, 5306.23 examples/s]
Tokenizing Prompts (num_proc=192):  76%|███████▌  | 1059000/1393783 [03:36<00:57, 5849.06 examples/s]
Tokenizing Prompts (num_proc=192):  76%|███████▌  | 1061000/1393783 [03:36<00:45, 7274.21 examples/s]
Tokenizing Prompts (num_proc=192):  76%|███████▌  | 1062000/1393783 [03:37<01:16, 4349.66 examples/s]
Tokenizing Prompts (num_proc=192):  76%|███████▋  | 1064000/1393783 [03:37<01:12, 4553.31 examples/s]
Tokenizing Prompts (num_proc=192):  76%|███████▋  | 1065000/1393783 [03:37<01:17, 4241.41 examples/s]
Tokenizing Prompts (num_proc=192):  77%|███████▋  | 1066259/1393783 [03:38<01:57, 2780.50 examples/s]
Tokenizing Prompts (num_proc=192):  77%|███████▋  | 1067259/1393783 [03:39<02:07, 2556.62 examples/s]
Tokenizing Prompts (num_proc=19
1: 2):  77%|███████▋  | 1069259/1393783 [03:39<01:50, 2941.67 examples/s]
Tokenizing Prompts (num_proc=192):  77%|███████▋  | 1070519/1393783 [03:40<02:13, 2427.85 examples/s]
Tokenizing Prompts (num_proc=192):  77%|███████▋  | 1071779/1393783 [03:41<02:22, 2265.93 examples/s]
Tokenizing Prompts (num_proc=192):  77%|███████▋  | 1072779/1393783 [03:41<02:00, 2666.84 examples/s]
Tokenizing Prompts (num_proc=192):  77%|███████▋  | 1073779/1393783 [03:41<02:00, 2658.97 examples/s]
Tokenizing Prompts (num_proc=192):  77%|███████▋  | 1074779/1393783 [03:41<01:43, 3073.11 examples/s]
Tokenizing Prompts (num_proc=192):  77%|███████▋  | 1076039/1393783 [03:42<02:16, 2326.33 examples/s]
Tokenizing Prompts (num_proc=192):  77%|███████▋  | 1079039/1393783 [03:42<01:13, 4289.50 examples/s]
Tokenizing Prompts (num_proc=192):  77%|███████▋  | 1080039/1393783 [03:43<01:08, 4585.19 examp
1: les/s]
Tokenizing Prompts (num_proc=192):  78%|███████▊  | 1082039/1393783 [03:43<00:49, 6336.91 examples/s]
Tokenizing Prompts (num_proc=192):  78%|███████▊  | 1084299/1393783 [03:43<01:07, 4562.23 examples/s]
Tokenizing Prompts (num_proc=192):  78%|███████▊  | 1086559/1393783 [03:44<00:55, 5514.07 examples/s]
Tokenizing Prompts (num_proc=192):  78%|███████▊  | 1088559/1393783 [03:44<01:04, 4700.72 examples/s]
Tokenizing Prompts (num_proc=192):  78%|███████▊  | 1089818/1393783 [03:45<01:10, 4309.35 examples/s]
Tokenizing Prompts (num_proc=192):  78%|███████▊  | 1092077/1393783 [03:45<00:55, 5471.91 examples/s]
Tokenizing Prompts (num_proc=192):  79%|███████▊  | 1096077/1393783 [03:45<00:36, 8126.74 examples/s]
Tokenizing Prompts (num_proc=192):  79%|███████▉  | 1098077/1393783 [03:45<00:31, 9476.30 examples/s]
Tokenizing Prompts (num_proc=192):  79%|███████▉  | 11000
1: 77/1393783 [03:46<00:46, 6334.04 examples/s]
Tokenizing Prompts (num_proc=192):  79%|███████▉  | 1102077/1393783 [03:46<00:43, 6774.27 examples/s]
Tokenizing Prompts (num_proc=192):  79%|███████▉  | 1103077/1393783 [03:46<00:52, 5511.27 examples/s]
Tokenizing Prompts (num_proc=192):  79%|███████▉  | 1104077/1393783 [03:47<00:50, 5766.93 examples/s]
Tokenizing Prompts (num_proc=192):  79%|███████▉  | 1105077/1393783 [03:47<00:47, 6107.41 examples/s]
Tokenizing Prompts (num_proc=192):  79%|███████▉  | 1106597/1393783 [03:47<00:45, 6258.55 examples/s]
Tokenizing Prompts (num_proc=192):  79%|███████▉  | 1107597/1393783 [03:47<00:52, 5420.01 examples/s]
Tokenizing Prompts (num_proc=192):  80%|███████▉  | 1108597/1393783 [03:47<00:50, 5696.09 examples/s]
Tokenizing Prompts (num_proc=192):  80%|███████▉  | 1109597/1393783 [03:48<00:50, 5600.65 examples/s]
Tokenizing Prompts (num_proc=192): 
1:  80%|███████▉  | 1111597/1393783 [03:48<00:40, 6948.56 examples/s]
Tokenizing Prompts (num_proc=192):  80%|███████▉  | 1112856/1393783 [03:48<00:55, 5069.63 examples/s]
Tokenizing Prompts (num_proc=192):  80%|███████▉  | 1114856/1393783 [03:48<00:39, 7005.22 examples/s]
Tokenizing Prompts (num_proc=192):  80%|████████  | 1115856/1393783 [03:48<00:38, 7291.76 examples/s]
Tokenizing Prompts (num_proc=192):  80%|████████  | 1117856/1393783 [03:49<00:40, 6851.44 examples/s]
Tokenizing Prompts (num_proc=192):  80%|████████  | 1118856/1393783 [03:49<00:39, 6946.86 examples/s]
Tokenizing Prompts (num_proc=192):  80%|████████  | 1119856/1393783 [03:49<00:44, 6213.36 examples/s]
Tokenizing Prompts (num_proc=192):  80%|████████  | 1121856/1393783 [03:49<00:43, 6231.34 examples/s]
Tokenizing Prompts (num_proc=192):  81%|████████  | 1123115/1393783 [03:50<00:44, 6054.22 examples/
1: s]
Tokenizing Prompts (num_proc=192):  81%|████████  | 1124115/1393783 [03:50<00:47, 5648.01 examples/s]
Tokenizing Prompts (num_proc=192):  81%|████████  | 1125115/1393783 [03:50<00:48, 5564.65 examples/s]
Tokenizing Prompts (num_proc=192):  81%|████████  | 1126115/1393783 [03:50<00:43, 6097.63 examples/s]
Tokenizing Prompts (num_proc=192):  81%|████████  | 1127375/1393783 [03:51<01:07, 3960.92 examples/s]
Tokenizing Prompts (num_proc=192):  81%|████████  | 1128635/1393783 [03:51<01:03, 4202.49 examples/s]
Tokenizing Prompts (num_proc=192):  81%|████████  | 1130894/1393783 [03:51<00:48, 5374.04 examples/s]
Tokenizing Prompts (num_proc=192):  81%|████████▏ | 1133894/1393783 [03:51<00:34, 7619.23 examples/s]
Tokenizing Prompts (num_proc=192):  81%|████████▏ | 1134894/1393783 [03:52<00:41, 6211.94 examples/s]
Tokenizing Prompts (num_proc=192):  82%|████████▏ | 113
1: 6894/1393783 [03:52<00:41, 6230.29 examples/s]
Tokenizing Prompts (num_proc=192):  82%|████████▏ | 1138894/1393783 [03:52<00:38, 6544.95 examples/s]
Tokenizing Prompts (num_proc=192):  82%|████████▏ | 1140894/1393783 [03:53<00:43, 5789.03 examples/s]
Tokenizing Prompts (num_proc=192):  82%|████████▏ | 1142894/1393783 [03:53<00:43, 5827.35 examples/s]
Tokenizing Prompts (num_proc=192):  82%|████████▏ | 1145153/1393783 [03:53<00:41, 5995.38 examples/s]
Tokenizing Prompts (num_proc=192):  82%|████████▏ | 1146153/1393783 [03:54<00:41, 5933.84 examples/s]
Tokenizing Prompts (num_proc=192):  82%|████████▏ | 1147153/1393783 [03:54<00:47, 5149.49 examples/s]
Tokenizing Prompts (num_proc=192):  82%|████████▏ | 1149153/1393783 [03:54<00:50, 4829.69 examples/s]
Tokenizing Prompts (num_proc=192):  83%|████████▎ | 1150413/1393783 [03:54<00:45, 5357.92 examples/s]
Tokenizing Prompt
1: s (num_proc=192):  83%|████████▎ | 1152413/1393783 [03:55<00:35, 6715.16 examples/s]
Tokenizing Prompts (num_proc=192):  83%|████████▎ | 1153413/1393783 [03:55<00:42, 5663.27 examples/s]
Tokenizing Prompts (num_proc=192):  83%|████████▎ | 1154413/1393783 [03:55<00:48, 4890.44 examples/s]
Tokenizing Prompts (num_proc=192):  83%|████████▎ | 1155413/1393783 [03:55<00:50, 4691.28 examples/s]
Tokenizing Prompts (num_proc=192):  83%|████████▎ | 1157673/1393783 [03:56<00:35, 6744.63 examples/s]
Tokenizing Prompts (num_proc=192):  83%|████████▎ | 1160192/1393783 [03:56<00:27, 8527.08 examples/s]
Tokenizing Prompts (num_proc=192):  83%|████████▎ | 1161451/1393783 [03:56<00:29, 7897.98 examples/s]
Tokenizing Prompts (num_proc=192):  83%|████████▎ | 1162710/1393783 [03:56<00:38, 6047.96 examples/s]
Tokenizing Prompts (num_proc=192):  83%|████████▎ | 1163710/139
1: 3783 [03:56<00:35, 6516.18 examples/s]
Tokenizing Prompts (num_proc=192):  84%|████████▎ | 1165970/1393783 [03:57<00:25, 8878.59 examples/s]
Tokenizing Prompts (num_proc=192):  84%|████████▎ | 1167229/1393783 [03:57<00:32, 7078.20 examples/s]
Tokenizing Prompts (num_proc=192):  84%|████████▍ | 1168229/1393783 [03:57<00:31, 7182.87 examples/s]
Tokenizing Prompts (num_proc=192):  84%|████████▍ | 1169229/1393783 [03:57<00:33, 6790.52 examples/s]
Tokenizing Prompts (num_proc=192):  84%|████████▍ | 1172748/1393783 [03:57<00:20, 10907.85 examples/s]
Tokenizing Prompts (num_proc=192):  84%|████████▍ | 1175748/1393783 [03:58<00:17, 12813.17 examples/s]
Tokenizing Prompts (num_proc=192):  85%|████████▍ | 1178007/1393783 [03:58<00:21, 9815.06 examples/s] 
Tokenizing Prompts (num_proc=192):  85%|████████▍ | 1179267/1393783 [03:58<00:34, 6209.74 examples/s]
Tokenizing Prompts (nu
1: m_proc=192):  85%|████████▍ | 1180267/1393783 [03:59<00:33, 6468.17 examples/s]
Tokenizing Prompts (num_proc=192):  85%|████████▍ | 1182267/1393783 [03:59<00:25, 8168.58 examples/s]
Tokenizing Prompts (num_proc=192):  85%|████████▍ | 1183527/1393783 [03:59<00:47, 4437.78 examples/s]
Tokenizing Prompts (num_proc=192):  85%|████████▌ | 1184787/1393783 [04:00<01:10, 2970.64 examples/s]
Tokenizing Prompts (num_proc=192):  85%|████████▌ | 1185787/1393783 [04:00<01:07, 3076.66 examples/s]
Tokenizing Prompts (num_proc=192):  85%|████████▌ | 1187307/1393783 [04:01<00:54, 3762.34 examples/s]
Tokenizing Prompts (num_proc=192):  85%|████████▌ | 1188307/1393783 [04:01<00:52, 3934.26 examples/s]
Tokenizing Prompts (num_proc=192):  85%|████████▌ | 1190567/1393783 [04:01<00:38, 5309.66 examples/s]
Tokenizing Prompts (num_proc=192):  86%|████████▌ | 1191827/1393783 
1: [04:01<00:38, 5296.82 examples/s]
Tokenizing Prompts (num_proc=192):  86%|████████▌ | 1193827/1393783 [04:02<00:29, 6711.54 examples/s]
Tokenizing Prompts (num_proc=192):  86%|████████▌ | 1195347/1393783 [04:02<00:31, 6392.70 examples/s]
Tokenizing Prompts (num_proc=192):  86%|████████▌ | 1197347/1393783 [04:02<00:25, 7835.03 examples/s]
Tokenizing Prompts (num_proc=192):  86%|████████▌ | 1199606/1393783 [04:02<00:20, 9502.14 examples/s]
Tokenizing Prompts (num_proc=192):  86%|████████▌ | 1201124/1393783 [04:02<00:26, 7145.20 examples/s]
Tokenizing Prompts (num_proc=192):  86%|████████▋ | 1202384/1393783 [04:03<00:24, 7805.71 examples/s]
Tokenizing Prompts (num_proc=192):  86%|████████▋ | 1204384/1393783 [04:03<00:26, 7085.80 examples/s]
Tokenizing Prompts (num_proc=192):  87%|████████▋ | 1208384/1393783 [04:03<00:16, 10956.28 examples/s]
Tokenizing Prompts (num_proc=
1: 192):  87%|████████▋ | 1209903/1393783 [04:03<00:21, 8639.47 examples/s] 
Tokenizing Prompts (num_proc=192):  87%|████████▋ | 1211903/1393783 [04:04<00:20, 8837.84 examples/s]
Tokenizing Prompts (num_proc=192):  87%|████████▋ | 1213162/1393783 [04:04<00:25, 6994.54 examples/s]
Tokenizing Prompts (num_proc=192):  87%|████████▋ | 1215162/1393783 [04:04<00:25, 6872.18 examples/s]
Tokenizing Prompts (num_proc=192):  87%|████████▋ | 1216940/1393783 [04:04<00:22, 7701.02 examples/s]
Tokenizing Prompts (num_proc=192):  87%|████████▋ | 1218200/1393783 [04:05<00:29, 5875.39 examples/s]
Tokenizing Prompts (num_proc=192):  87%|████████▋ | 1219200/1393783 [04:05<00:30, 5679.69 examples/s]
Tokenizing Prompts (num_proc=192):  88%|████████▊ | 1220200/1393783 [04:05<00:38, 4452.55 examples/s]
Tokenizing Prompts (num_proc=192):  88%|████████▊ | 1221459/1393783 [04:06
1: <00:41, 4134.13 examples/s]
Tokenizing Prompts (num_proc=192):  88%|████████▊ | 1223977/1393783 [04:06<00:29, 5794.75 examples/s]
Tokenizing Prompts (num_proc=192):  88%|████████▊ | 1225977/1393783 [04:06<00:24, 6964.97 examples/s]
Tokenizing Prompts (num_proc=192):  88%|████████▊ | 1228495/1393783 [04:06<00:20, 8244.90 examples/s]
Tokenizing Prompts (num_proc=192):  88%|████████▊ | 1229495/1393783 [04:07<00:25, 6539.65 examples/s]
Tokenizing Prompts (num_proc=192):  88%|████████▊ | 1233495/1393783 [04:07<00:17, 9040.16 examples/s]
Tokenizing Prompts (num_proc=192):  89%|████████▊ | 1235015/1393783 [04:07<00:20, 7859.59 examples/s]
Tokenizing Prompts (num_proc=192):  89%|████████▊ | 1236015/1393783 [04:08<00:25, 6244.19 examples/s]
Tokenizing Prompts (num_proc=192):  89%|████████▉ | 1237275/1393783 [04:08<00:25, 6222.93 examples/s]
Tokenizing Prompts (num_proc=192):  
1: 89%|████████▉ | 1240275/1393783 [04:08<00:16, 9108.51 examples/s]
Tokenizing Prompts (num_proc=192):  89%|████████▉ | 1241534/1393783 [04:08<00:17, 8534.29 examples/s]
Tokenizing Prompts (num_proc=192):  89%|████████▉ | 1243053/1393783 [04:08<00:17, 8561.94 examples/s]
Tokenizing Prompts (num_proc=192):  89%|████████▉ | 1244573/1393783 [04:08<00:19, 7584.52 examples/s]
Tokenizing Prompts (num_proc=192):  89%|████████▉ | 1246573/1393783 [04:09<00:15, 9536.50 examples/s]
Tokenizing Prompts (num_proc=192):  90%|████████▉ | 1248573/1393783 [04:09<00:28, 5040.53 examples/s]
Tokenizing Prompts (num_proc=192):  90%|████████▉ | 1249573/1393783 [04:09<00:26, 5476.52 examples/s]
Tokenizing Prompts (num_proc=192):  90%|████████▉ | 1250832/1393783 [04:10<00:22, 6300.45 examples/s]
Tokenizing Prompts (num_proc=192):  90%|████████▉ | 1252091/1393783 [04:10<00:25, 
1: 5486.77 examples/s]
Tokenizing Prompts (num_proc=192):  90%|████████▉ | 1253091/1393783 [04:10<00:25, 5453.24 examples/s]
Tokenizing Prompts (num_proc=192):  90%|█████████ | 1256091/1393783 [04:10<00:17, 7676.63 examples/s]
Tokenizing Prompts (num_proc=192):  90%|█████████ | 1257350/1393783 [04:11<00:18, 7192.89 examples/s]
Tokenizing Prompts (num_proc=192):  90%|█████████ | 1258350/1393783 [04:11<00:18, 7258.60 examples/s]
Tokenizing Prompts (num_proc=192):  90%|█████████ | 1259870/1393783 [04:11<00:25, 5151.25 examples/s]
Tokenizing Prompts (num_proc=192):  90%|█████████ | 1261129/1393783 [04:11<00:23, 5566.24 examples/s]
Tokenizing Prompts (num_proc=192):  91%|█████████ | 1263129/1393783 [04:12<00:19, 6687.39 examples/s]
Tokenizing Prompts (num_proc=192):  91%|█████████ | 1263906/1393783 [04:12<00:26, 4880.97 examples/s]
Tokenizing Prompts (num_proc=192):  91%|█�
1: ��███████ | 1266165/1393783 [04:12<00:17, 7217.42 examples/s]
Tokenizing Prompts (num_proc=192):  91%|█████████ | 1267684/1393783 [04:12<00:21, 5937.80 examples/s]
Tokenizing Prompts (num_proc=192):  91%|█████████ | 1269944/1393783 [04:13<00:20, 6025.02 examples/s]
Tokenizing Prompts (num_proc=192):  91%|█████████▏| 1272203/1393783 [04:13<00:15, 7821.72 examples/s]
Tokenizing Prompts (num_proc=192):  91%|█████████▏| 1274721/1393783 [04:13<00:18, 6574.04 examples/s]
Tokenizing Prompts (num_proc=192):  92%|█████████▏| 1275981/1393783 [04:14<00:17, 6716.07 examples/s]
Tokenizing Prompts (num_proc=192):  92%|█████████▏| 1278241/1393783 [04:14<00:13, 8811.57 examples/s]
Tokenizing Prompts (num_proc=192):  92%|█████████▏| 1280241/1393783 [04:14<00:23, 4822.97 examples/s]
Tokenizing Prompts (num_proc=192):  92%|█████████▏| 1282241/1393783 [04:15<00:
1: 23, 4744.20 examples/s]
Tokenizing Prompts (num_proc=192):  92%|█████████▏| 1283500/1393783 [04:15<00:21, 5103.00 examples/s]
Tokenizing Prompts (num_proc=192):  92%|█████████▏| 1286500/1393783 [04:15<00:16, 6622.99 examples/s]
Tokenizing Prompts (num_proc=192):  92%|█████████▏| 1288020/1393783 [04:16<00:20, 5217.40 examples/s]
Tokenizing Prompts (num_proc=192):  93%|█████████▎| 1289279/1393783 [04:16<00:17, 5904.24 examples/s]
Tokenizing Prompts (num_proc=192):  93%|█████████▎| 1291279/1393783 [04:16<00:13, 7635.54 examples/s]
Tokenizing Prompts (num_proc=192):  93%|█████████▎| 1292837/1393783 [04:16<00:12, 7900.32 examples/s]
Tokenizing Prompts (num_proc=192):  93%|█████████▎| 1296096/1393783 [04:16<00:08, 11653.21 examples/s]
Tokenizing Prompts (num_proc=192):  93%|█████████▎| 1297873/1393783 [04:17<00:10, 9467.31 examples/s] 
Tokenizing Prompts (nu
1: m_proc=192):  93%|█████████▎| 1300133/1393783 [04:17<00:09, 10195.62 examples/s]
Tokenizing Prompts (num_proc=192):  93%|█████████▎| 1302392/1393783 [04:17<00:08, 11223.79 examples/s]
Tokenizing Prompts (num_proc=192):  94%|█████████▎| 1304651/1393783 [04:18<00:12, 7217.89 examples/s] 
Tokenizing Prompts (num_proc=192):  94%|█████████▎| 1305910/1393783 [04:18<00:14, 5873.11 examples/s]
Tokenizing Prompts (num_proc=192):  94%|█████████▍| 1307170/1393783 [04:18<00:13, 6279.89 examples/s]
Tokenizing Prompts (num_proc=192):  94%|█████████▍| 1308207/1393783 [04:18<00:14, 6069.54 examples/s]
Tokenizing Prompts (num_proc=192):  94%|█████████▍| 1309466/1393783 [04:19<00:16, 5131.52 examples/s]
Tokenizing Prompts (num_proc=192):  94%|█████████▍| 1310726/1393783 [04:19<00:17, 4830.88 examples/s]
Tokenizing Prompts (num_proc=192):  94%|█████████
1: ▍| 1312244/1393783 [04:19<00:20, 3891.70 examples/s]
Tokenizing Prompts (num_proc=192):  94%|█████████▍| 1313763/1393783 [04:20<00:17, 4602.39 examples/s]
Tokenizing Prompts (num_proc=192):  94%|█████████▍| 1315022/1393783 [04:20<00:14, 5441.68 examples/s]
Tokenizing Prompts (num_proc=192):  94%|█████████▍| 1316281/1393783 [04:20<00:15, 4880.68 examples/s]
Tokenizing Prompts (num_proc=192):  95%|█████████▍| 1317540/1393783 [04:20<00:14, 5175.86 examples/s]
Tokenizing Prompts (num_proc=192):  95%|█████████▍| 1318540/1393783 [04:21<00:14, 5087.82 examples/s]
Tokenizing Prompts (num_proc=192):  95%|█████████▍| 1319318/1393783 [04:21<00:16, 4642.78 examples/s]
Tokenizing Prompts (num_proc=192):  95%|█████████▍| 1320577/1393783 [04:21<00:12, 5816.32 examples/s]
Tokenizing Prompts (num_proc=192):  95%|█████████▍| 1323577/1393783 [04:21<00:07, 9557.63 examp
1: les/s]
Tokenizing Prompts (num_proc=192):  95%|█████████▌| 1325577/1393783 [04:21<00:09, 7562.50 examples/s]
Tokenizing Prompts (num_proc=192):  95%|█████████▌| 1327577/1393783 [04:21<00:07, 9328.87 examples/s]
Tokenizing Prompts (num_proc=192):  95%|█████████▌| 1330095/1393783 [04:22<00:05, 11303.13 examples/s]
Tokenizing Prompts (num_proc=192):  96%|█████████▌| 1331872/1393783 [04:22<00:05, 12259.49 examples/s]
Tokenizing Prompts (num_proc=192):  96%|█████████▌| 1333390/1393783 [04:22<00:05, 11140.84 examples/s]
Tokenizing Prompts (num_proc=192):  96%|█████████▌| 1334908/1393783 [04:22<00:05, 10919.98 examples/s]
Tokenizing Prompts (num_proc=192):  96%|█████████▌| 1336426/1393783 [04:22<00:06, 8470.82 examples/s] 
Tokenizing Prompts (num_proc=192):  96%|█████████▌| 1338204/1393783 [04:23<00:07, 7188.45 examples/s]
Tokenizing Prompts (num_proc=192):  
1: 96%|█████████▌| 1339463/1393783 [04:23<00:08, 6690.32 examples/s]
Tokenizing Prompts (num_proc=192):  96%|█████████▌| 1340982/1393783 [04:23<00:10, 4965.10 examples/s]
Tokenizing Prompts (num_proc=192):  96%|█████████▋| 1343982/1393783 [04:24<00:07, 6449.90 examples/s]
Tokenizing Prompts (num_proc=192):  97%|█████████▋| 1346241/1393783 [04:24<00:08, 5740.45 examples/s]
Tokenizing Prompts (num_proc=192):  97%|█████████▋| 1347241/1393783 [04:24<00:07, 5903.61 examples/s]
Tokenizing Prompts (num_proc=192):  97%|█████████▋| 1349241/1393783 [04:24<00:06, 7406.77 examples/s]
Tokenizing Prompts (num_proc=192):  97%|█████████▋| 1350759/1393783 [04:25<00:06, 6192.49 examples/s]
Tokenizing Prompts (num_proc=192):  97%|█████████▋| 1352018/1393783 [04:25<00:05, 6993.91 examples/s]
Tokenizing Prompts (num_proc=192):  97%|█████████▋| 1353795/1393
1: 783 [04:25<00:05, 7865.66 examples/s]
Tokenizing Prompts (num_proc=192):  97%|█████████▋| 1355572/1393783 [04:25<00:05, 7417.93 examples/s]
Tokenizing Prompts (num_proc=192):  97%|█████████▋| 1356832/1393783 [04:26<00:06, 5854.60 examples/s]
Tokenizing Prompts (num_proc=192):  97%|█████████▋| 1358350/1393783 [04:26<00:07, 4951.40 examples/s]
Tokenizing Prompts (num_proc=192):  98%|█████████▊| 1359868/1393783 [04:26<00:06, 4962.94 examples/s]
Tokenizing Prompts (num_proc=192):  98%|█████████▊| 1362386/1393783 [04:27<00:04, 7380.61 examples/s]
Tokenizing Prompts (num_proc=192):  98%|█████████▊| 1364645/1393783 [04:27<00:03, 9455.52 examples/s]
Tokenizing Prompts (num_proc=192):  98%|█████████▊| 1366423/1393783 [04:27<00:04, 6526.85 examples/s]
Tokenizing Prompts (num_proc=192):  98%|█████████▊| 1368200/1393783 [04:28<00:06, 4053.08 examples/s]
Tokenizing
1:  Prompts (num_proc=192):  98%|█████████▊| 1370200/1393783 [04:28<00:04, 5347.73 examples/s]
Tokenizing Prompts (num_proc=192):  98%|█████████▊| 1371718/1393783 [04:28<00:03, 5944.09 examples/s]
Tokenizing Prompts (num_proc=192):  99%|█████████▊| 1374236/1393783 [04:29<00:02, 6934.21 examples/s]
Tokenizing Prompts (num_proc=192):  99%|█████████▊| 1376013/1393783 [04:29<00:02, 7478.00 examples/s]
Tokenizing Prompts (num_proc=192):  99%|█████████▉| 1377272/1393783 [04:29<00:02, 6533.07 examples/s]
Tokenizing Prompts (num_proc=192):  99%|█████████▉| 1379049/1393783 [04:30<00:03, 3804.69 examples/s]
Tokenizing Prompts (num_proc=192):  99%|█████████▉| 1380308/1393783 [04:30<00:03, 4205.88 examples/s]
Tokenizing Prompts (num_proc=192):  99%|█████████▉| 1381826/1393783 [04:31<00:03, 3659.04 examples/s]
Tokenizing Prompts (num_proc=192):  99%|██████
1: ███▉| 1382603/1393783 [04:31<00:03, 3241.10 examples/s]
Tokenizing Prompts (num_proc=192):  99%|█████████▉| 1383121/1393783 [04:31<00:03, 3298.24 examples/s]
Tokenizing Prompts (num_proc=192):  99%|█████████▉| 1384380/1393783 [04:31<00:02, 4180.68 examples/s]
Tokenizing Prompts (num_proc=192):  99%|█████████▉| 1385639/1393783 [04:32<00:02, 3079.84 examples/s]
Tokenizing Prompts (num_proc=192):  99%|█████████▉| 1386157/1393783 [04:32<00:02, 2955.03 examples/s]
Tokenizing Prompts (num_proc=192): 100%|█████████▉| 1387416/1393783 [04:32<00:01, 3507.02 examples/s]
Tokenizing Prompts (num_proc=192): 100%|█████████▉| 1387934/1393783 [04:33<00:01, 3617.21 examples/s]
Tokenizing Prompts (num_proc=192): 100%|█████████▉| 1388934/1393783 [04:33<00:01, 4061.71 examples/s]
Tokenizing Prompts (num_proc=192): 100%|█████████▉| 1390452/1393783 [04:33<00:00, 4633
1: .12 examples/s]
Tokenizing Prompts (num_proc=192): 100%|█████████▉| 1391711/1393783 [04:34<00:00, 2586.06 examples/s]
Tokenizing Prompts (num_proc=192): 100%|█████████▉| 1392229/1393783 [04:34<00:00, 2134.06 examples/s]
Tokenizing Prompts (num_proc=192): 100%|█████████▉| 1392747/1393783 [04:35<00:00, 1846.13 examples/s]
Tokenizing Prompts (num_proc=192): 100%|█████████▉| 1393006/1393783 [04:36<00:00, 1232.02 examples/s]
Tokenizing Prompts (num_proc=192): 100%|█████████▉| 1393265/1393783 [04:36<00:00, 1068.64 examples/s]
Tokenizing Prompts (num_proc=192): 100%|█████████▉| 1393524/1393783 [04:36<00:00, 1167.22 examples/s]
Tokenizing Prompts (num_proc=192): 100%|██████████| 1393783/1393783 [04:37<00:00, 733.68 examples/s] 
Tokenizing Prompts (num_proc=192): 100%|██████████| 1393783/1393783 [04:37<00:00, 5013.68 examples/s]
1: 
Dropping Long Sequences (>16384) (num_proc=192):   0%|          | 0/1393783 [00:00<?, ? examples/s]
Dropping Long Sequences (>16384) (num_proc=192):   0%|          | 1000/1393783 [00:02<58:31, 396.64 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):   1%|          | 14000/1393783 [00:02<03:07, 7344.37 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):   2%|▏         | 27000/1393783 [00:02<01:25, 15972.69 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):   3%|▎         | 41000/1393783 [00:02<00:49, 27269.85 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):   4%|▍         | 59000/1393783 [00:02<00:29, 44838.46 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):   5%|▌         | 73000/1393783 [00:03<00:28, 46876.46 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):   6%|▌         | 84000/1393783 [00:03<00:31, 42138.93 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):   7%|▋         | 93000/1393783 [00:03<00:28, 45269.25 example
1: s/s]
Dropping Long Sequences (>16384) (num_proc=192):   7%|▋         | 101000/1393783 [00:03<00:26, 49571.19 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):   8%|▊         | 109000/1393783 [00:03<00:26, 48628.82 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):   8%|▊         | 118000/1393783 [00:04<00:23, 55029.26 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):   9%|▉         | 126000/1393783 [00:04<00:21, 59447.41 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  10%|█         | 140000/1393783 [00:04<00:16, 75214.56 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  11%|█         | 156000/1393783 [00:04<00:13, 94426.03 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  12%|█▏        | 167000/1393783 [00:04<00:18, 67206.49 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  13%|█▎        | 176000/1393783 [00:04<00:17, 67861.36 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  14%|█▎        | 1
1: 89000/1393783 [00:04<00:14, 80462.63 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  15%|█▍        | 206000/1393783 [00:05<00:12, 98924.11 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  16%|█▌        | 218000/1393783 [00:05<00:11, 99901.14 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  17%|█▋        | 231000/1393783 [00:05<00:11, 102304.96 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  17%|█▋        | 242000/1393783 [00:05<00:12, 94345.70 examples/s] 
Dropping Long Sequences (>16384) (num_proc=192):  18%|█▊        | 253000/1393783 [00:05<00:12, 88831.22 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  19%|█▉        | 263000/1393783 [00:05<00:13, 83018.24 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  20%|██        | 280000/1393783 [00:05<00:11, 98549.35 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  21%|██        | 291000/1393783 [00:06<00:14, 77188.81 examples/s]
Dropping Long
1:  Sequences (>16384) (num_proc=192):  22%|██▏       | 300000/1393783 [00:06<00:15, 69716.04 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  22%|██▏       | 310000/1393783 [00:06<00:14, 75421.23 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  23%|██▎       | 319000/1393783 [00:06<00:14, 73244.72 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  24%|██▍       | 333000/1393783 [00:06<00:12, 88194.26 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  25%|██▍       | 344000/1393783 [00:06<00:11, 89663.54 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  25%|██▌       | 354000/1393783 [00:06<00:14, 73554.42 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  26%|██▌       | 364000/1393783 [00:07<00:13, 73880.16 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  27%|██▋       | 373000/1393783 [00:07<00:13, 73940.93 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  28%|██�
1: ��       | 389000/1393783 [00:07<00:10, 92955.36 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  29%|██▉       | 404000/1393783 [00:07<00:09, 104817.41 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  30%|██▉       | 416000/1393783 [00:07<00:09, 101268.19 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  31%|███       | 427000/1393783 [00:07<00:09, 98704.61 examples/s] 
Dropping Long Sequences (>16384) (num_proc=192):  31%|███▏      | 438000/1393783 [00:07<00:10, 91656.92 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  32%|███▏      | 448000/1393783 [00:07<00:10, 90487.20 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  33%|███▎      | 458000/1393783 [00:07<00:10, 86221.16 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  34%|███▍      | 471000/1393783 [00:08<00:09, 94147.71 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  35%|███▍      | 481000/1393783 [00:08<00
1: :10, 83564.50 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  35%|███▌      | 490000/1393783 [00:08<00:12, 74186.24 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  36%|███▌      | 498000/1393783 [00:08<00:12, 72852.12 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  36%|███▋      | 506000/1393783 [00:08<00:12, 73398.06 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  37%|███▋      | 517000/1393783 [00:08<00:11, 79220.76 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  38%|███▊      | 531000/1393783 [00:08<00:09, 93166.00 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  39%|███▉      | 541000/1393783 [00:08<00:09, 87561.92 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  39%|███▉      | 550000/1393783 [00:09<00:12, 68805.25 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  40%|████      | 561000/1393783 [00:09<00:10, 76851.96 examples/s]
Droppi
1: ng Long Sequences (>16384) (num_proc=192):  41%|████      | 570000/1393783 [00:09<00:10, 78377.99 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  42%|████▏     | 589000/1393783 [00:09<00:07, 105785.70 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  43%|████▎     | 601000/1393783 [00:09<00:09, 87684.40 examples/s] 
Dropping Long Sequences (>16384) (num_proc=192):  44%|████▍     | 614000/1393783 [00:09<00:08, 88502.25 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  45%|████▍     | 624000/1393783 [00:09<00:08, 90795.81 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  46%|████▌     | 636000/1393783 [00:10<00:07, 97073.35 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  46%|████▋     | 647000/1393783 [00:10<00:09, 78137.95 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  47%|████▋     | 662000/1393783 [00:10<00:07, 92678.48 examples/s]
Dropping Long Sequence
1: s (>16384) (num_proc=192):  48%|████▊     | 673000/1393783 [00:10<00:07, 91350.35 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  49%|████▉     | 683000/1393783 [00:10<00:09, 73374.40 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  50%|████▉     | 693000/1393783 [00:10<00:08, 78524.66 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  50%|█████     | 702000/1393783 [00:10<00:09, 74808.71 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  52%|█████▏    | 719000/1393783 [00:11<00:07, 92072.91 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  52%|█████▏    | 729000/1393783 [00:11<00:07, 84370.34 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  53%|█████▎    | 738000/1393783 [00:11<00:07, 82515.27 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  54%|█████▎    | 747000/1393783 [00:11<00:08, 72834.79 examples/s]
Dropping Long Sequences (>1638
1: 4) (num_proc=192):  54%|█████▍    | 759000/1393783 [00:11<00:07, 81499.98 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  55%|█████▌    | 772000/1393783 [00:11<00:06, 92279.81 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  56%|█████▋    | 786000/1393783 [00:11<00:06, 98907.52 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  57%|█████▋    | 797000/1393783 [00:12<00:07, 84701.72 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  58%|█████▊    | 808000/1393783 [00:12<00:06, 85701.67 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  59%|█████▉    | 820000/1393783 [00:12<00:06, 93919.53 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  60%|█████▉    | 830000/1393783 [00:12<00:06, 93040.83 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  60%|██████    | 840000/1393783 [00:12<00:06, 80513.46 examples/s]
Dropping Long Sequences (>1638
1: 4) (num_proc=192):  61%|██████    | 853000/1393783 [00:12<00:05, 92280.66 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  62%|██████▏   | 867000/1393783 [00:12<00:05, 102640.21 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  63%|██████▎   | 878000/1393783 [00:13<00:07, 71477.68 examples/s] 
Dropping Long Sequences (>16384) (num_proc=192):  64%|██████▎   | 887000/1393783 [00:13<00:07, 70750.22 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  64%|██████▍   | 897000/1393783 [00:13<00:06, 76437.07 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  65%|██████▌   | 911000/1393783 [00:13<00:05, 87871.05 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  66%|██████▌   | 921000/1393783 [00:13<00:05, 86876.84 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  67%|██████▋   | 931000/1393783 [00:13<00:05, 82110.59 examples/s]
Dropping Long 
1: Sequences (>16384) (num_proc=192):  67%|██████▋   | 940000/1393783 [00:13<00:05, 82349.20 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  68%|██████▊   | 950000/1393783 [00:13<00:05, 82628.34 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  69%|██████▉   | 962000/1393783 [00:13<00:04, 91988.86 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  70%|██████▉   | 974000/1393783 [00:14<00:04, 99391.86 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  71%|███████   | 985000/1393783 [00:14<00:04, 85370.64 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  71%|███████▏  | 995000/1393783 [00:14<00:04, 87265.24 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  72%|███████▏  | 1005000/1393783 [00:14<00:04, 87093.35 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  73%|███████▎  | 1014000/1393783 [00:14<00:04, 80254.47 ex
1: amples/s]
Dropping Long Sequences (>16384) (num_proc=192):  74%|███████▎  | 1025000/1393783 [00:14<00:04, 87085.18 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  74%|███████▍  | 1035000/1393783 [00:14<00:04, 89586.22 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  75%|███████▌  | 1046000/1393783 [00:14<00:03, 90443.29 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  76%|███████▌  | 1062000/1393783 [00:15<00:03, 107393.54 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  77%|███████▋  | 1073000/1393783 [00:15<00:04, 71322.24 examples/s] 
Dropping Long Sequences (>16384) (num_proc=192):  78%|███████▊  | 1082260/1393783 [00:15<00:04, 72365.01 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  79%|███████▊  | 1094520/1393783 [00:15<00:03, 80013.53 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  79%|███████▉  |
1:  1103560/1393783 [00:15<00:03, 80352.62 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  80%|████████  | 1115820/1393783 [00:15<00:03, 90317.58 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  81%|████████  | 1125860/1393783 [00:15<00:02, 89460.07 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  81%|████████▏ | 1135639/1393783 [00:15<00:02, 88513.61 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  82%|████████▏ | 1145679/1393783 [00:16<00:02, 91332.02 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  83%|████████▎ | 1161939/1393783 [00:16<00:02, 108600.82 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  85%|████████▍ | 1179239/1393783 [00:16<00:01, 126328.29 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  86%|████████▌ | 1192538/1393783 [00:16<00:01, 119579.57 examples/s]
Dropping Long Sequences (>16
1: 384) (num_proc=192):  86%|████████▋ | 1205395/1393783 [00:16<00:01, 101593.44 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  87%|████████▋ | 1216251/1393783 [00:16<00:01, 96591.02 examples/s] 
Dropping Long Sequences (>16384) (num_proc=192):  88%|████████▊ | 1227846/1393783 [00:16<00:01, 99041.58 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  89%|████████▉ | 1239884/1393783 [00:16<00:01, 103247.85 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  90%|████████▉ | 1250661/1393783 [00:17<00:01, 104065.88 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  91%|█████████ | 1269217/1393783 [00:17<00:00, 125927.55 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  92%|█████████▏| 1282514/1393783 [00:17<00:00, 119294.31 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  93%|█████████▎| 1295845/13937
1: 83 [00:17<00:00, 118673.65 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  94%|█████████▍| 1308176/1393783 [00:17<00:00, 119763.59 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  95%|█████████▍| 1322802/1393783 [00:17<00:00, 126565.13 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  96%|█████████▌| 1335689/1393783 [00:17<00:00, 121885.75 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  97%|█████████▋| 1350056/1393783 [00:17<00:00, 127063.41 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  98%|█████████▊| 1362905/1393783 [00:17<00:00, 121790.20 examples/s]
Dropping Long Sequences (>16384) (num_proc=192):  99%|█████████▊| 1375344/1393783 [00:18<00:00, 88862.01 examples/s] 
Dropping Long Sequences (>16384) (num_proc=192):  99%|█████████▉| 1385639/1393783 [00:18<00:00, 82390.52 examples/s]
Dropping Long Sequen
1: ces (>16384) (num_proc=192): 100%|██████████| 1393783/1393783 [00:18<00:00, 73466.64 examples/s]
1: 
Drop Samples with Zero Trainable Tokens (num_proc=192):   0%|          | 0/1386595 [00:00<?, ? examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):   0%|          | 1000/1386595 [00:02<51:19, 449.96 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):   1%|          | 10000/1386595 [00:02<03:56, 5822.97 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):   1%|          | 15000/1386595 [00:02<02:29, 9196.15 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):   2%|▏         | 22000/1386595 [00:02<01:58, 11547.87 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):   2%|▏         | 26000/1386595 [00:02<01:35, 14242.67 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):   2%|▏         | 34000/1386595 [00:03<01:00, 22200.50 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):   3%|▎         | 44000/1386595 [00:03<00:41, 32486.12 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):   4
1: %|▍         | 59000/1386595 [00:03<00:25, 52210.24 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):   5%|▍         | 68000/1386595 [00:03<00:22, 58694.57 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):   6%|▌         | 78000/1386595 [00:03<00:19, 66374.96 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):   6%|▋         | 88000/1386595 [00:03<00:18, 69715.60 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):   7%|▋         | 97000/1386595 [00:03<00:23, 54298.82 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):   8%|▊         | 104000/1386595 [00:04<00:23, 54949.97 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):   8%|▊         | 112000/1386595 [00:04<00:21, 58485.61 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):   9%|▊         | 120000/1386595 [00:04<00:21, 58391.45 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):   9%|▉         | 129000/138659
1: 5 [00:04<00:19, 65026.03 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  10%|▉         | 137000/1386595 [00:04<00:18, 65994.17 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  11%|█         | 146000/1386595 [00:04<00:17, 71707.94 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  11%|█         | 154000/1386595 [00:04<00:20, 58974.25 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  12%|█▏        | 161000/1386595 [00:04<00:24, 50547.06 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  12%|█▏        | 169000/1386595 [00:05<00:21, 56780.29 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  13%|█▎        | 176000/1386595 [00:05<00:21, 55875.23 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  13%|█▎        | 183000/1386595 [00:05<00:21, 56009.07 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  14%|█▍        | 198000/1386595 [00:05<00:15
1: , 78702.29 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  15%|█▌        | 210000/1386595 [00:05<00:13, 88646.99 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  16%|█▌        | 224000/1386595 [00:05<00:12, 91558.53 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  17%|█▋        | 235000/1386595 [00:05<00:12, 89225.31 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  18%|█▊        | 245000/1386595 [00:06<00:15, 74126.06 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  18%|█▊        | 254000/1386595 [00:06<00:15, 74383.00 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  19%|█▉        | 262000/1386595 [00:06<00:16, 66573.68 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  20%|█▉        | 272000/1386595 [00:06<00:15, 72660.28 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  20%|██        | 283000/1386595 [00:06<00:14, 77336.
1: 45 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  21%|██        | 292000/1386595 [00:06<00:15, 68469.19 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  22%|██▏       | 301000/1386595 [00:06<00:15, 71631.87 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  23%|██▎       | 316000/1386595 [00:06<00:13, 79791.34 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  23%|██▎       | 325000/1386595 [00:07<00:12, 81684.72 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  24%|██▍       | 334000/1386595 [00:07<00:14, 72518.97 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  25%|██▍       | 344000/1386595 [00:07<00:13, 78005.08 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  26%|██▌       | 357000/1386595 [00:07<00:11, 89959.31 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  26%|██▋       | 367000/1386595 [00:07<00:11, 
1: 89889.81 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  27%|██▋       | 381000/1386595 [00:07<00:09, 101926.72 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  29%|██▉       | 399000/1386595 [00:07<00:08, 121509.90 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  30%|██▉       | 415000/1386595 [00:07<00:07, 128627.46 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  31%|███       | 429000/1386595 [00:08<00:09, 103047.62 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  32%|███▏      | 441000/1386595 [00:08<00:11, 84174.76 examples/s] 
Drop Samples with Zero Trainable Tokens (num_proc=192):  33%|███▎      | 451000/1386595 [00:08<00:11, 80481.89 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  33%|███▎      | 460000/1386595 [00:08<00:12, 71913.26 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  34%|███▍      | 472000/1
1: 386595 [00:08<00:11, 79885.18 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  35%|███▍      | 481000/1386595 [00:08<00:11, 75792.51 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  35%|███▌      | 490000/1386595 [00:08<00:12, 71441.89 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  36%|███▋      | 503000/1386595 [00:09<00:10, 83786.44 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  37%|███▋      | 512000/1386595 [00:09<00:11, 76373.26 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  38%|███▊      | 521000/1386595 [00:09<00:11, 74121.43 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  38%|███▊      | 529000/1386595 [00:09<00:12, 69188.96 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  39%|███▉      | 539000/1386595 [00:09<00:11, 74776.79 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  40%|█�
1: ��██      | 555000/1386595 [00:09<00:08, 95707.24 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  41%|████      | 566000/1386595 [00:09<00:08, 93418.44 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  42%|████▏     | 587000/1386595 [00:09<00:06, 122144.12 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  44%|████▎     | 605000/1386595 [00:10<00:05, 135672.26 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  45%|████▍     | 620000/1386595 [00:10<00:07, 99320.48 examples/s] 
Drop Samples with Zero Trainable Tokens (num_proc=192):  46%|████▌     | 632000/1386595 [00:10<00:08, 84043.79 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  46%|████▋     | 642000/1386595 [00:10<00:08, 86357.40 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  47%|████▋     | 652000/1386595 [00:10<00:09, 79354.70 examples/s]
Drop Samples with Zero Tra
1: inable Tokens (num_proc=192):  48%|████▊     | 661000/1386595 [00:10<00:09, 79446.92 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  48%|████▊     | 670000/1386595 [00:11<00:09, 72724.89 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  49%|████▉     | 679000/1386595 [00:11<00:09, 73223.83 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  50%|████▉     | 689000/1386595 [00:11<00:08, 78458.33 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  50%|█████     | 699000/1386595 [00:11<00:08, 82186.26 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  51%|█████     | 708000/1386595 [00:11<00:08, 81636.41 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  52%|█████▏    | 717000/1386595 [00:11<00:09, 71918.10 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  52%|█████▏    | 725000/1386595 [00:11<00:10, 661
1: 42.50 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  53%|█████▎    | 738000/1386595 [00:11<00:08, 80354.96 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  54%|█████▍    | 752000/1386595 [00:12<00:06, 92218.30 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  55%|█████▌    | 769000/1386595 [00:12<00:05, 110934.77 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  57%|█████▋    | 785000/1386595 [00:12<00:04, 121739.70 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  58%|█████▊    | 799000/1386595 [00:12<00:04, 124715.61 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  59%|█████▊    | 812222/1386595 [00:12<00:05, 98096.59 examples/s] 
Drop Samples with Zero Trainable Tokens (num_proc=192):  59%|█████▉    | 823444/1386595 [00:12<00:06, 81183.44 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  
1: 60%|██████    | 837444/1386595 [00:12<00:06, 86816.49 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  61%|██████▏   | 850444/1386595 [00:13<00:06, 88881.23 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  62%|██████▏   | 860666/1386595 [00:13<00:06, 75514.13 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  63%|██████▎   | 870888/1386595 [00:13<00:06, 79232.42 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  63%|██████▎   | 880110/1386595 [00:13<00:07, 72025.86 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  64%|██████▍   | 893110/1386595 [00:13<00:06, 76054.45 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  65%|██████▌   | 901332/1386595 [00:13<00:06, 72206.93 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  66%|██████▌   | 915554/1386595 [00:13<00:05, 86126.66
1:  examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  67%|██████▋   | 925220/1386595 [00:14<00:06, 74094.24 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  68%|██████▊   | 940664/1386595 [00:14<00:04, 91473.33 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  69%|██████▉   | 953664/1386595 [00:14<00:04, 100260.08 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  70%|███████   | 976330/1386595 [00:14<00:03, 132286.78 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  71%|███████▏  | 991330/1386595 [00:14<00:03, 130328.79 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  73%|███████▎  | 1005330/1386595 [00:14<00:03, 110641.66 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  73%|███████▎  | 1018218/1386595 [00:14<00:03, 97076.02 examples/s] 
Drop Samples with Zero Trainable Tokens
1:  (num_proc=192):  74%|███████▍  | 1029662/1386595 [00:14<00:03, 98479.25 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  75%|███████▌  | 1041106/1386595 [00:15<00:03, 100844.91 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  76%|███████▌  | 1051772/1386595 [00:15<00:03, 85853.05 examples/s] 
Drop Samples with Zero Trainable Tokens (num_proc=192):  77%|███████▋  | 1061438/1386595 [00:15<00:04, 74519.77 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  77%|███████▋  | 1069882/1386595 [00:15<00:04, 75351.02 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  78%|███████▊  | 1080882/1386595 [00:15<00:03, 80025.16 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  79%|███████▉  | 1095104/1386595 [00:15<00:03, 95034.72 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  80%|██████�
1: �▉  | 1105992/1386595 [00:15<00:03, 85154.67 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  80%|████████  | 1115436/1386595 [00:16<00:03, 87299.58 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  81%|████████  | 1126546/1386595 [00:16<00:02, 89425.59 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  82%|████████▏ | 1135990/1386595 [00:16<00:02, 86320.69 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  83%|████████▎ | 1149656/1386595 [00:16<00:02, 97682.56 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  84%|████████▍ | 1164544/1386595 [00:16<00:01, 111324.05 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  85%|████████▌ | 1184432/1386595 [00:16<00:01, 135418.94 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  86%|████████▋ | 1198764/1386595 [00:16<0
1: 0:01, 120908.36 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  87%|████████▋ | 1211762/1386595 [00:16<00:01, 112895.52 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  88%|████████▊ | 1224316/1386595 [00:16<00:01, 110121.30 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  89%|████████▉ | 1236648/1386595 [00:17<00:01, 111352.77 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  90%|█████████ | 1250202/1386595 [00:17<00:01, 115524.16 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  91%|█████████ | 1261978/1386595 [00:17<00:01, 109910.09 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  92%|█████████▏| 1273976/1386595 [00:17<00:01, 103807.38 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  93%|█████████▎| 1284864/1386595 [00:17<00:01, 94427.56 exam
1: ples/s] 
Drop Samples with Zero Trainable Tokens (num_proc=192):  93%|█████████▎| 1295973/1386595 [00:17<00:00, 97129.83 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  94%|█████████▍| 1305969/1386595 [00:17<00:00, 94687.73 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  95%|█████████▍| 1316078/1386595 [00:17<00:00, 90973.93 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  96%|█████████▌| 1329410/1386595 [00:18<00:00, 101965.01 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  97%|█████████▋| 1340408/1386595 [00:18<00:00, 93603.43 examples/s] 
Drop Samples with Zero Trainable Tokens (num_proc=192):  97%|█████████▋| 1350959/1386595 [00:18<00:00, 93973.22 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  98%|█████████▊| 1360953/1386595 [00:18<00:00, 83361.47 examples/s]
Drop 
1: Samples with Zero Trainable Tokens (num_proc=192):  99%|█████████▉| 1370500/1386595 [00:18<00:00, 56818.02 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192):  99%|█████████▉| 1377500/1386595 [00:18<00:00, 54342.99 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192): 100%|█████████▉| 1383827/1386595 [00:19<00:00, 48496.13 examples/s]
Drop Samples with Zero Trainable Tokens (num_proc=192): 100%|██████████| 1386595/1386595 [00:19<00:00, 70755.56 examples/s]
1: 
Add position_id column (Sample Packing) (num_proc=192):   0%|          | 0/1386595 [00:00<?, ? examples/s]
Add position_id column (Sample Packing) (num_proc=192):   0%|          | 1000/1386595 [00:03<1:11:41, 322.15 examples/s]
Add position_id column (Sample Packing) (num_proc=192):   0%|          | 3000/1386595 [00:03<19:47, 1165.07 examples/s] 
Add position_id column (Sample Packing) (num_proc=192):   0%|          | 6000/1386595 [00:03<08:06, 2837.10 examples/s]
Add position_id column (Sample Packing) (num_proc=192):   1%|          | 13000/1386595 [00:03<02:54, 7850.64 examples/s]
Add position_id column (Sample Packing) (num_proc=192):   2%|▏         | 22000/1386595 [00:03<01:26, 15725.70 examples/s]
Add position_id column (Sample Packing) (num_proc=192):   2%|▏         | 28000/1386595 [00:03<01:05, 20764.97 examples/s]
Add position_id column (Sample Packing) (num_proc=192):   3%|▎         | 40000/1386595 [00:03<00:38, 34562.62 examples/s]
Add position_id column (Sample Packing) (num_proc=192):   3%|
1: ▎         | 47000/1386595 [00:03<00:33, 39766.97 examples/s]
Add position_id column (Sample Packing) (num_proc=192):   4%|▍         | 55000/1386595 [00:04<00:29, 44694.70 examples/s]
Add position_id column (Sample Packing) (num_proc=192):   4%|▍         | 62000/1386595 [00:04<00:28, 47004.50 examples/s]
Add position_id column (Sample Packing) (num_proc=192):   5%|▍         | 69000/1386595 [00:04<00:39, 33235.10 examples/s]
Add position_id column (Sample Packing) (num_proc=192):   5%|▌         | 74000/1386595 [00:04<00:38, 34109.89 examples/s]
Add position_id column (Sample Packing) (num_proc=192):   6%|▌         | 79000/1386595 [00:04<00:44, 29290.33 examples/s]
Add position_id column (Sample Packing) (num_proc=192):   6%|▋         | 88000/1386595 [00:05<00:34, 37148.25 examples/s]
Add position_id column (Sample Packing) (num_proc=192):   7%|▋         | 93000/1386595 [00:05<00:36, 35605.73 examples/s]
Add position_id column (Sample Packing) (num_proc=192):   7%|▋         | 98000/1386595 [00:
1: 05<00:44, 28938.66 examples/s]
Add position_id column (Sample Packing) (num_proc=192):   8%|▊         | 107000/1386595 [00:05<00:32, 39054.55 examples/s]
Add position_id column (Sample Packing) (num_proc=192):   9%|▊         | 121000/1386595 [00:05<00:21, 58747.12 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  10%|▉         | 133000/1386595 [00:05<00:17, 71963.31 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  10%|█         | 145000/1386595 [00:05<00:15, 78007.83 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  11%|█         | 155000/1386595 [00:06<00:15, 77756.72 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  12%|█▏        | 164000/1386595 [00:06<00:15, 80490.50 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  12%|█▏        | 173000/1386595 [00:06<00:16, 73162.84 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  13%|█▎        | 184000/1386595 [00:06<00:15, 79588.93
1:  examples/s]
Add position_id column (Sample Packing) (num_proc=192):  14%|█▍        | 193000/1386595 [00:06<00:22, 53611.25 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  14%|█▍        | 200000/1386595 [00:06<00:25, 46596.85 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  15%|█▍        | 206000/1386595 [00:07<00:24, 47444.17 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  15%|█▌        | 212000/1386595 [00:07<00:23, 49566.63 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  16%|█▌        | 219000/1386595 [00:07<00:21, 53365.69 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  17%|█▋        | 231000/1386595 [00:07<00:17, 65464.72 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  17%|█▋        | 241000/1386595 [00:07<00:15, 72065.81 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  18%|█▊        | 249000/1386595 [00:07<00:16, 70687.49 example
1: s/s]
Add position_id column (Sample Packing) (num_proc=192):  19%|█▊        | 257000/1386595 [00:07<00:23, 47136.36 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  19%|█▉        | 263000/1386595 [00:08<00:29, 38736.87 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  19%|█▉        | 268000/1386595 [00:08<00:29, 37877.15 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  20%|█▉        | 273000/1386595 [00:08<00:30, 36488.53 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  20%|██        | 279000/1386595 [00:08<00:27, 39593.65 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  20%|██        | 284000/1386595 [00:08<00:31, 34503.09 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  21%|██        | 288000/1386595 [00:09<00:41, 26177.97 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  22%|██▏       | 302000/1386595 [00:09<00:23, 46000.90 examples/s]
A
1: dd position_id column (Sample Packing) (num_proc=192):  22%|██▏       | 311000/1386595 [00:09<00:19, 54064.96 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  23%|██▎       | 324000/1386595 [00:09<00:15, 67299.70 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  24%|██▍       | 339000/1386595 [00:09<00:12, 84354.64 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  25%|██▌       | 349000/1386595 [00:09<00:13, 74723.80 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  26%|██▌       | 358000/1386595 [00:09<00:13, 73851.51 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  26%|██▋       | 367000/1386595 [00:09<00:13, 74484.84 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  27%|██▋       | 375000/1386595 [00:10<00:14, 70596.60 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  28%|██▊       | 384000/1386595 [00:10<00:14, 67929.79 examp
1: les/s]
Add position_id column (Sample Packing) (num_proc=192):  28%|██▊       | 392000/1386595 [00:10<00:20, 49670.40 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  29%|██▊       | 398000/1386595 [00:10<00:19, 50092.90 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  29%|██▉       | 405000/1386595 [00:10<00:18, 54183.33 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  30%|███       | 416000/1386595 [00:10<00:15, 63274.42 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  31%|███       | 423000/1386595 [00:10<00:15, 61326.73 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  31%|███▏      | 434000/1386595 [00:11<00:13, 71273.18 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  32%|███▏      | 442000/1386595 [00:11<00:16, 57797.19 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  32%|███▏      | 449000/1386595 [00:11<00:21, 
1: 43710.80 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  33%|███▎      | 455000/1386595 [00:11<00:24, 37818.52 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  33%|███▎      | 461000/1386595 [00:11<00:22, 41739.03 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  34%|███▎      | 466000/1386595 [00:12<00:24, 37280.82 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  34%|███▍      | 471000/1386595 [00:12<00:27, 33203.76 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  34%|███▍      | 477000/1386595 [00:12<00:26, 34570.43 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  35%|███▍      | 481000/1386595 [00:12<00:27, 33173.86 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  35%|███▍      | 485000/1386595 [00:12<00:29, 30800.08 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  36%|███▌      | 49400
1: 0/1386595 [00:12<00:22, 40549.39 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  36%|███▋      | 504000/1386595 [00:12<00:17, 51603.83 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  37%|███▋      | 516000/1386595 [00:13<00:13, 66842.20 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  38%|███▊      | 530000/1386595 [00:13<00:10, 83289.79 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  39%|███▉      | 540000/1386595 [00:13<00:13, 64628.12 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  40%|███▉      | 551000/1386595 [00:13<00:11, 74169.91 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  40%|████      | 561000/1386595 [00:13<00:10, 78245.45 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  41%|████      | 570000/1386595 [00:13<00:11, 69227.51 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  42%|�
1: ��███▏     | 578000/1386595 [00:13<00:13, 60788.92 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  42%|████▏     | 585000/1386595 [00:14<00:13, 61417.53 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  43%|████▎     | 592000/1386595 [00:14<00:14, 53552.49 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  44%|████▎     | 604000/1386595 [00:14<00:12, 64373.11 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  44%|████▍     | 611000/1386595 [00:14<00:11, 65116.78 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  45%|████▍     | 618000/1386595 [00:14<00:12, 62177.55 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  45%|████▌     | 625000/1386595 [00:14<00:12, 62020.67 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  46%|████▌     | 632000/1386595 [00:14<00:15, 49978.12 examples/s]
Add position_id column
1:  (Sample Packing) (num_proc=192):  46%|████▌     | 638000/1386595 [00:15<00:15, 48061.36 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  46%|████▋     | 644000/1386595 [00:15<00:16, 46087.91 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  47%|████▋     | 649000/1386595 [00:15<00:18, 40535.73 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  47%|████▋     | 654000/1386595 [00:15<00:19, 37117.00 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  48%|████▊     | 660000/1386595 [00:15<00:19, 36441.39 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  48%|████▊     | 666000/1386595 [00:15<00:22, 31715.98 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  48%|████▊     | 670000/1386595 [00:16<00:22, 32213.03 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  49%|████▉     | 679000/1386595 [00:16<00:16, 434
1: 00.86 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  49%|████▉     | 684000/1386595 [00:16<00:17, 40195.49 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  50%|████▉     | 692000/1386595 [00:16<00:14, 48658.67 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  50%|█████     | 699000/1386595 [00:16<00:12, 53110.08 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  51%|█████▏    | 711000/1386595 [00:16<00:10, 67417.96 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  52%|█████▏    | 723000/1386595 [00:16<00:08, 74100.20 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  53%|█████▎    | 732000/1386595 [00:16<00:08, 75650.44 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  53%|█████▎    | 740000/1386595 [00:17<00:10, 59878.23 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  54%|██
1: ███▍    | 752000/1386595 [00:17<00:08, 71113.39 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  55%|█████▍    | 761000/1386595 [00:17<00:08, 72829.50 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  55%|█████▌    | 769000/1386595 [00:17<00:09, 64679.72 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  56%|█████▌    | 777000/1386595 [00:17<00:08, 67752.63 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  57%|█████▋    | 785000/1386595 [00:17<00:09, 61360.99 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  57%|█████▋    | 792000/1386595 [00:17<00:09, 62109.36 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  58%|█████▊    | 799000/1386595 [00:18<00:09, 62942.24 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  58%|█████▊    | 809000/1386595 [00:18<00:08, 68613.51 examples/s]
Add positio
1: n_id column (Sample Packing) (num_proc=192):  59%|█████▉    | 817000/1386595 [00:18<00:10, 56257.49 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  59%|█████▉    | 824000/1386595 [00:18<00:09, 56683.06 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  60%|█████▉    | 830000/1386595 [00:18<00:11, 48262.77 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  60%|██████    | 836000/1386595 [00:18<00:11, 47149.33 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  61%|██████    | 841000/1386595 [00:19<00:14, 37334.35 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  61%|██████    | 846000/1386595 [00:19<00:13, 38902.23 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  62%|██████▏   | 853000/1386595 [00:19<00:12, 41222.51 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  62%|██████▏   | 8
1: 58000/1386595 [00:19<00:13, 38785.47 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  62%|██████▏   | 863000/1386595 [00:19<00:15, 33947.97 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  63%|██████▎   | 869000/1386595 [00:19<00:13, 38943.34 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  63%|██████▎   | 874000/1386595 [00:19<00:13, 38166.35 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  64%|██████▍   | 884000/1386595 [00:19<00:09, 50615.76 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  64%|██████▍   | 893000/1386595 [00:20<00:08, 59665.62 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  65%|██████▍   | 900000/1386595 [00:20<00:08, 58866.14 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  65%|██████▌   | 907000/1386595 [00:20<00:07, 60415.25 examples/s]
Add position_id 
1: column (Sample Packing) (num_proc=192):  66%|██████▌   | 914000/1386595 [00:20<00:08, 58526.71 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  66%|██████▋   | 922000/1386595 [00:20<00:07, 61149.30 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  67%|██████▋   | 930000/1386595 [00:20<00:07, 61686.45 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  68%|██████▊   | 937000/1386595 [00:20<00:07, 63451.92 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  68%|██████▊   | 946000/1386595 [00:20<00:06, 70418.91 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  69%|██████▉   | 954000/1386595 [00:20<00:06, 71924.39 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  69%|██████▉   | 962000/1386595 [00:21<00:06, 64542.76 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  70%|██████�
1: �   | 969000/1386595 [00:21<00:06, 64583.74 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  70%|███████   | 976000/1386595 [00:21<00:06, 62603.98 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  71%|███████   | 984000/1386595 [00:21<00:06, 61877.54 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  71%|███████▏  | 991000/1386595 [00:21<00:06, 63718.05 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  72%|███████▏  | 999000/1386595 [00:21<00:05, 65353.48 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  73%|███████▎  | 1006000/1386595 [00:21<00:06, 56752.32 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  73%|███████▎  | 1013000/1386595 [00:22<00:06, 57133.87 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  73%|███████▎  | 1019000/1386595 [00:22<00:07, 47924.49 examples
1: /s]
Add position_id column (Sample Packing) (num_proc=192):  74%|███████▍  | 1025000/1386595 [00:22<00:09, 40033.96 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  74%|███████▍  | 1030000/1386595 [00:22<00:09, 39455.36 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  75%|███████▍  | 1036000/1386595 [00:22<00:08, 42819.39 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  75%|███████▌  | 1041000/1386595 [00:22<00:07, 43705.87 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  75%|███████▌  | 1046000/1386595 [00:22<00:07, 43010.76 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  76%|███████▌  | 1051000/1386595 [00:23<00:09, 34002.74 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  76%|███████▌  | 1057000/1386595 [00:23<00:08, 39445.92 examples/s]
Add position_id column (Sample Packing) 
1: (num_proc=192):  77%|███████▋  | 1062000/1386595 [00:23<00:08, 37083.84 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  77%|███████▋  | 1067000/1386595 [00:23<00:08, 36380.75 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  78%|███████▊  | 1076000/1386595 [00:23<00:06, 48426.55 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  78%|███████▊  | 1085000/1386595 [00:23<00:05, 55635.48 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  79%|███████▉  | 1092000/1386595 [00:23<00:04, 59049.24 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  79%|███████▉  | 1101000/1386595 [00:23<00:04, 66485.39 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  80%|███████▉  | 1108000/1386595 [00:24<00:04, 61652.21 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  80%|███████�
1: �  | 1115000/1386595 [00:24<00:04, 57977.56 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  81%|████████  | 1122000/1386595 [00:24<00:04, 55360.17 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  81%|████████▏ | 1128000/1386595 [00:24<00:04, 54592.56 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  82%|████████▏ | 1140000/1386595 [00:24<00:03, 68147.49 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  83%|████████▎ | 1149000/1386595 [00:24<00:03, 69669.97 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  83%|████████▎ | 1157000/1386595 [00:24<00:03, 71438.96 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  84%|████████▍ | 1169000/1386595 [00:24<00:02, 82013.65 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  85%|████████▍ | 1178000/1386595 [00:25<00:0
1: 3, 68153.50 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  85%|████████▌ | 1185444/1386595 [00:25<00:03, 62713.00 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  86%|████████▋ | 1197110/1386595 [00:25<00:02, 71930.65 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  87%|████████▋ | 1205220/1386595 [00:25<00:02, 61700.36 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  87%|████████▋ | 1211996/1386595 [00:25<00:03, 54470.77 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  88%|████████▊ | 1217994/1386595 [00:25<00:03, 48962.93 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  88%|████████▊ | 1225882/1386595 [00:26<00:03, 53126.62 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  89%|████████▉ | 1232102/1386595 [00:26<00:02, 55162.18 examples/s]
Add po
1: sition_id column (Sample Packing) (num_proc=192):  89%|████████▉ | 1238212/1386595 [00:26<00:02, 53925.93 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  90%|████████▉ | 1244544/1386595 [00:26<00:02, 50870.29 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  90%|█████████ | 1252766/1386595 [00:26<00:02, 57728.35 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  91%|█████████ | 1259764/1386595 [00:26<00:02, 55312.65 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  91%|█████████▏| 1267318/1386595 [00:26<00:02, 59182.44 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  92%|█████████▏| 1274984/1386595 [00:26<00:01, 63409.33 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  93%|█████████▎| 1283982/1386595 [00:26<00:01, 70054.15 examples/s]
Add position_id column (Sample
1:  Packing) (num_proc=192):  94%|█████████▎| 1296536/1386595 [00:27<00:01, 83962.76 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  94%|█████████▍| 1308867/1386595 [00:27<00:00, 94602.50 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  95%|█████████▌| 1318976/1386595 [00:27<00:00, 91569.76 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  96%|█████████▌| 1329196/1386595 [00:27<00:00, 90994.58 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  97%|█████████▋| 1338523/1386595 [00:27<00:00, 84416.48 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  97%|█████████▋| 1350406/1386595 [00:27<00:00, 93146.60 examples/s]
Add position_id column (Sample Packing) (num_proc=192):  98%|█████████▊| 1360846/1386595 [00:27<00:00, 95863.67 examples/s]
Add position_id column (Sample Packing) (num_p
1: roc=192):  99%|█████████▉| 1371619/1386595 [00:27<00:00, 97214.49 examples/s]
Add position_id column (Sample Packing) (num_proc=192): 100%|█████████▉| 1381498/1386595 [00:28<00:00, 77285.35 examples/s]
Add position_id column (Sample Packing) (num_proc=192): 100%|██████████| 1386595/1386595 [00:28<00:00, 48197.21 examples/s]
1: 
Saving the dataset (0/192 shards):   0%|          | 0/1386595 [00:00<?, ? examples/s]
Saving the dataset (0/192 shards):   0%|          | 1000/1386595 [00:02<49:47, 463.82 examples/s]
Saving the dataset (0/192 shards):   4%|▍         | 59000/1386595 [00:02<00:36, 36339.76 examples/s]
Saving the dataset (0/192 shards):  10%|█         | 143000/1386595 [00:02<00:12, 101633.34 examples/s]
Saving the dataset (0/192 shards):  14%|█▍        | 199000/1386595 [00:02<00:08, 145933.77 examples/s]
Saving the dataset (0/192 shards):  21%|██        | 293000/1386595 [00:02<00:04, 246515.15 examples/s]
Saving the dataset (0/192 shards):  26%|██▌       | 360000/1386595 [00:02<00:03, 310522.15 examples/s]
Saving the dataset (0/192 shards):  42%|████▏     | 578000/1386595 [00:02<00:01, 635899.38 examples/s]
Saving the dataset (1/192 shards):  45%|████▌     | 625222/1386595 [00:02<00:01, 635899.38 examples/s]
Saving the dataset (2/192 shards):  51%|█████     | 701444/1386595 [00:0
1: 2<00:01, 635899.38 examples/s]
Saving the dataset (2/192 shards):  51%|█████     | 707444/1386595 [00:02<00:00, 762130.44 examples/s]
Saving the dataset (3/192 shards):  52%|█████▏    | 727666/1386595 [00:02<00:00, 762130.44 examples/s]
Saving the dataset (4/192 shards):  57%|█████▋    | 785888/1386595 [00:03<00:00, 762130.44 examples/s]
Saving the dataset (5/192 shards):  57%|█████▋    | 788110/1386595 [00:03<00:00, 762130.44 examples/s]
Saving the dataset (6/192 shards):  58%|█████▊    | 805332/1386595 [00:03<00:00, 762130.44 examples/s]
Saving the dataset (6/192 shards):  59%|█████▉    | 822998/1386595 [00:03<00:00, 650727.65 examples/s]
Saving the dataset (7/192 shards):  59%|█████▉    | 822998/1386595 [00:03<00:00, 650727.65 examples/s]
Saving the dataset (8/192 shards):  59%|█████▉    | 824220/1386595 [00:03<00:00, 650727.65 examples/s]
Saving the dataset (9/192 shards):  60%|█████▉    | 825220/1386
1: 595 [00:03<00:00, 650727.65 examples/s]
Saving the dataset (10/192 shards):  60%|█████▉    | 826442/1386595 [00:03<00:00, 650727.65 examples/s]
Saving the dataset (11/192 shards):  60%|█████▉    | 827442/1386595 [00:03<00:00, 650727.65 examples/s]
Saving the dataset (12/192 shards):  61%|██████    | 842108/1386595 [00:03<00:00, 650727.65 examples/s]
Saving the dataset (13/192 shards):  61%|██████    | 842108/1386595 [00:03<00:00, 650727.65 examples/s]
Saving the dataset (14/192 shards):  61%|██████    | 842108/1386595 [00:03<00:00, 650727.65 examples/s]
Saving the dataset (15/192 shards):  62%|██████▏   | 856330/1386595 [00:03<00:00, 650727.65 examples/s]
Saving the dataset (16/192 shards):  62%|██████▏   | 859552/1386595 [00:03<00:00, 650727.65 examples/s]
Saving the dataset (17/192 shards):  62%|██████▏   | 865996/1386595 [00:03<00:00, 650727.65 examples/s]
Saving the dataset (18/192 shards):  63%|███
1: ███▎   | 867996/1386595 [00:03<00:00, 650727.65 examples/s]
Saving the dataset (19/192 shards):  63%|██████▎   | 877662/1386595 [00:03<00:00, 650727.65 examples/s]
Saving the dataset (20/192 shards):  63%|██████▎   | 877662/1386595 [00:03<00:00, 650727.65 examples/s]
Saving the dataset (21/192 shards):  63%|██████▎   | 877662/1386595 [00:03<00:00, 650727.65 examples/s]
Saving the dataset (22/192 shards):  64%|██████▍   | 888884/1386595 [00:03<00:00, 650727.65 examples/s]
Saving the dataset (23/192 shards):  65%|██████▍   | 901106/1386595 [00:03<00:00, 650727.65 examples/s]
Saving the dataset (23/192 shards):  66%|██████▌   | 917328/1386595 [00:03<00:00, 641233.81 examples/s]
Saving the dataset (24/192 shards):  66%|██████▌   | 918328/1386595 [00:03<00:00, 641233.81 examples/s]
Saving the dataset (25/192 shards):  66%|██████▋   | 920550/1386595 [00:03<00:00, 641233.81 examples/s]
Saving the d
1: ataset (26/192 shards):  67%|██████▋   | 928994/1386595 [00:03<00:00, 641233.81 examples/s]
Saving the dataset (27/192 shards):  67%|██████▋   | 929438/1386595 [00:03<00:00, 641233.81 examples/s]
Saving the dataset (28/192 shards):  67%|██████▋   | 930438/1386595 [00:03<00:00, 641233.81 examples/s]
Saving the dataset (29/192 shards):  67%|██████▋   | 931438/1386595 [00:03<00:00, 641233.81 examples/s]
Saving the dataset (30/192 shards):  68%|██████▊   | 938660/1386595 [00:03<00:00, 641233.81 examples/s]
Saving the dataset (31/192 shards):  68%|██████▊   | 945882/1386595 [00:03<00:00, 641233.81 examples/s]
Saving the dataset (32/192 shards):  69%|██████▉   | 954104/1386595 [00:03<00:00, 641233.81 examples/s]
Saving the dataset (33/192 shards):  69%|██████▉   | 961326/1386595 [00:03<00:00, 641233.81 examples/s]
Saving the dataset (34/192 shards):  70%|██████▉   | 969548/1386595 [00:03<00:
1: 00, 641233.81 examples/s]
Saving the dataset (35/192 shards):  70%|███████   | 973770/1386595 [00:03<00:00, 641233.81 examples/s]
Saving the dataset (36/192 shards):  71%|███████   | 977992/1386595 [00:03<00:00, 641233.81 examples/s]
Saving the dataset (37/192 shards):  71%|███████   | 987214/1386595 [00:03<00:00, 641233.81 examples/s]
Saving the dataset (38/192 shards):  72%|███████▏  | 998658/1386595 [00:03<00:00, 641233.81 examples/s]
Saving the dataset (39/192 shards):  72%|███████▏  | 998658/1386595 [00:03<00:00, 641233.81 examples/s]
Saving the dataset (40/192 shards):  72%|███████▏  | 1000880/1386595 [00:03<00:00, 641233.81 examples/s]
Saving the dataset (40/192 shards):  72%|███████▏  | 1001880/1386595 [00:03<00:00, 586110.15 examples/s]
Saving the dataset (41/192 shards):  72%|███████▏  | 1004102/1386595 [00:03<00:00, 586110.15 examples/s]
Saving the dataset (42/192 shards):  73%|
1: ███████▎  | 1011324/1386595 [00:03<00:00, 586110.15 examples/s]
Saving the dataset (43/192 shards):  73%|███████▎  | 1018546/1386595 [00:03<00:00, 586110.15 examples/s]
Saving the dataset (44/192 shards):  74%|███████▍  | 1028768/1386595 [00:03<00:00, 586110.15 examples/s]
Saving the dataset (45/192 shards):  75%|███████▌  | 1039990/1386595 [00:03<00:00, 586110.15 examples/s]
Saving the dataset (46/192 shards):  76%|███████▌  | 1051212/1386595 [00:03<00:00, 586110.15 examples/s]
Saving the dataset (47/192 shards):  76%|███████▌  | 1053656/1386595 [00:03<00:00, 586110.15 examples/s]
Saving the dataset (48/192 shards):  76%|███████▌  | 1053656/1386595 [00:03<00:00, 586110.15 examples/s]
Saving the dataset (49/192 shards):  76%|███████▌  | 1055878/1386595 [00:03<00:00, 586110.15 examples/s]
Saving the dataset (50/192 shards):  76%|███████▌  | 1057100/1386595 [00:03<00:00
1: , 586110.15 examples/s]
Saving the dataset (51/192 shards):  76%|███████▋  | 1060322/1386595 [00:03<00:00, 586110.15 examples/s]
Saving the dataset (52/192 shards):  77%|███████▋  | 1066544/1386595 [00:03<00:00, 586110.15 examples/s]
Saving the dataset (53/192 shards):  77%|███████▋  | 1070988/1386595 [00:03<00:00, 586110.15 examples/s]
Saving the dataset (54/192 shards):  77%|███████▋  | 1070988/1386595 [00:03<00:00, 586110.15 examples/s]
Saving the dataset (55/192 shards):  77%|███████▋  | 1072210/1386595 [00:03<00:00, 586110.15 examples/s]
Saving the dataset (56/192 shards):  77%|███████▋  | 1074432/1386595 [00:03<00:00, 586110.15 examples/s]
Saving the dataset (56/192 shards):  78%|███████▊  | 1075432/1386595 [00:03<00:00, 613159.94 examples/s]
Saving the dataset (57/192 shards):  78%|███████▊  | 1075654/1386595 [00:03<00:00, 613159.94 examples/s]
Saving the dataset (58/192 shard
1: s):  78%|███████▊  | 1078876/1386595 [00:03<00:00, 613159.94 examples/s]
Saving the dataset (59/192 shards):  78%|███████▊  | 1082098/1386595 [00:03<00:00, 613159.94 examples/s]
Saving the dataset (60/192 shards):  78%|███████▊  | 1084320/1386595 [00:03<00:00, 613159.94 examples/s]
Saving the dataset (61/192 shards):  78%|███████▊  | 1084542/1386595 [00:03<00:00, 613159.94 examples/s]
Saving the dataset (62/192 shards):  78%|███████▊  | 1087764/1386595 [00:03<00:00, 613159.94 examples/s]
Saving the dataset (63/192 shards):  79%|███████▉  | 1093986/1386595 [00:03<00:00, 613159.94 examples/s]
Saving the dataset (64/192 shards):  79%|███████▉  | 1095430/1386595 [00:03<00:00, 613159.94 examples/s]
Saving the dataset (65/192 shards):  79%|███████▉  | 1095430/1386595 [00:03<00:00, 613159.94 examples/s]
Saving the dataset (66/192 shards):  80%|███████▉  | 1102652/1386595 [00
1: :03<00:00, 613159.94 examples/s]
Saving the dataset (67/192 shards):  80%|███████▉  | 1102874/1386595 [00:03<00:00, 613159.94 examples/s]
Saving the dataset (68/192 shards):  80%|████████  | 1110096/1386595 [00:03<00:00, 613159.94 examples/s]
Saving the dataset (69/192 shards):  80%|████████  | 1111318/1386595 [00:03<00:00, 613159.94 examples/s]
Saving the dataset (70/192 shards):  81%|████████  | 1122540/1386595 [00:03<00:00, 613159.94 examples/s]
Saving the dataset (71/192 shards):  81%|████████  | 1123762/1386595 [00:03<00:00, 613159.94 examples/s]
Saving the dataset (72/192 shards):  81%|████████  | 1124984/1386595 [00:03<00:00, 613159.94 examples/s]
Saving the dataset (73/192 shards):  82%|████████▏ | 1130206/1386595 [00:03<00:00, 613159.94 examples/s]
Saving the dataset (74/192 shards):  82%|████████▏ | 1135428/1386595 [00:03<00:00, 613159.94 examples/s]
Saving the dataset 
1: (75/192 shards):  82%|████████▏ | 1138650/1386595 [00:03<00:00, 613159.94 examples/s]
Saving the dataset (76/192 shards):  82%|████████▏ | 1142872/1386595 [00:03<00:00, 613159.94 examples/s]
Saving the dataset (77/192 shards):  83%|████████▎ | 1147094/1386595 [00:03<00:00, 613159.94 examples/s]
Saving the dataset (77/192 shards):  83%|████████▎ | 1148316/1386595 [00:03<00:00, 591272.38 examples/s]
Saving the dataset (78/192 shards):  83%|████████▎ | 1148316/1386595 [00:03<00:00, 591272.38 examples/s]
Saving the dataset (79/192 shards):  83%|████████▎ | 1151538/1386595 [00:03<00:00, 591272.38 examples/s]
Saving the dataset (80/192 shards):  84%|████████▎ | 1160760/1386595 [00:03<00:00, 591272.38 examples/s]
Saving the dataset (81/192 shards):  84%|████████▍ | 1162982/1386595 [00:03<00:00, 591272.38 examples/s]
Saving the dataset (82/192 shards):  84%|██████
1: ██▍ | 1165204/1386595 [00:03<00:00, 591272.38 examples/s]
Saving the dataset (83/192 shards):  84%|████████▍ | 1169426/1386595 [00:03<00:00, 591272.38 examples/s]
Saving the dataset (84/192 shards):  84%|████████▍ | 1170648/1386595 [00:03<00:00, 591272.38 examples/s]
Saving the dataset (85/192 shards):  85%|████████▍ | 1171870/1386595 [00:03<00:00, 591272.38 examples/s]
Saving the dataset (86/192 shards):  85%|████████▍ | 1174092/1386595 [00:03<00:00, 591272.38 examples/s]
Saving the dataset (87/192 shards):  85%|████████▌ | 1181314/1386595 [00:03<00:00, 591272.38 examples/s]
Saving the dataset (88/192 shards):  85%|████████▌ | 1182758/1386595 [00:03<00:00, 591272.38 examples/s]
Saving the dataset (89/192 shards):  85%|████████▌ | 1182758/1386595 [00:03<00:00, 591272.38 examples/s]
Saving the dataset (90/192 shards):  86%|████████▌ | 1187980/1386595 [00:03<00:00
1: , 591272.38 examples/s]
Saving the dataset (91/192 shards):  86%|████████▌ | 1189202/1386595 [00:03<00:00, 591272.38 examples/s]
Saving the dataset (92/192 shards):  86%|████████▋ | 1197424/1386595 [00:03<00:00, 591272.38 examples/s]
Saving the dataset (93/192 shards):  86%|████████▋ | 1198646/1386595 [00:03<00:00, 591272.38 examples/s]
Saving the dataset (94/192 shards):  87%|████████▋ | 1203090/1386595 [00:03<00:00, 591272.38 examples/s]
Saving the dataset (95/192 shards):  87%|████████▋ | 1203090/1386595 [00:03<00:00, 591272.38 examples/s]
Saving the dataset (96/192 shards):  87%|████████▋ | 1206312/1386595 [00:03<00:00, 591272.38 examples/s]
Saving the dataset (97/192 shards):  88%|████████▊ | 1213534/1386595 [00:03<00:00, 591272.38 examples/s]
Saving the dataset (97/192 shards):  88%|████████▊ | 1215534/1386595 [00:03<00:00, 587520.11 examples/s]
Saving the datas
1: et (98/192 shards):  88%|████████▊ | 1215756/1386595 [00:03<00:00, 587520.11 examples/s]
Saving the dataset (99/192 shards):  88%|████████▊ | 1217978/1386595 [00:03<00:00, 587520.11 examples/s]
Saving the dataset (100/192 shards):  88%|████████▊ | 1221200/1386595 [00:03<00:00, 587520.11 examples/s]
Saving the dataset (101/192 shards):  88%|████████▊ | 1223422/1386595 [00:03<00:00, 587520.11 examples/s]
Saving the dataset (102/192 shards):  88%|████████▊ | 1224644/1386595 [00:03<00:00, 587520.11 examples/s]
Saving the dataset (103/192 shards):  89%|████████▉ | 1231866/1386595 [00:03<00:00, 587520.11 examples/s]
Saving the dataset (104/192 shards):  89%|████████▉ | 1234088/1386595 [00:03<00:00, 587520.11 examples/s]
Saving the dataset (105/192 shards):  89%|████████▉ | 1236310/1386595 [00:03<00:00, 587520.11 examples/s]
Saving the dataset (106/192 shards):  89%|██�
1: �█████▉ | 1237754/1386595 [00:03<00:00, 587520.11 examples/s]
Saving the dataset (107/192 shards):  89%|████████▉ | 1237754/1386595 [00:03<00:00, 587520.11 examples/s]
Saving the dataset (108/192 shards):  89%|████████▉ | 1237976/1386595 [00:03<00:00, 587520.11 examples/s]
Saving the dataset (109/192 shards):  89%|████████▉ | 1240198/1386595 [00:03<00:00, 587520.11 examples/s]
Saving the dataset (110/192 shards):  90%|████████▉ | 1247864/1386595 [00:03<00:00, 587520.11 examples/s]
Saving the dataset (111/192 shards):  90%|█████████ | 1251864/1386595 [00:03<00:00, 587520.11 examples/s]
Saving the dataset (112/192 shards):  90%|█████████ | 1251864/1386595 [00:03<00:00, 587520.11 examples/s]
Saving the dataset (113/192 shards):  91%|█████████ | 1257086/1386595 [00:03<00:00, 587520.11 examples/s]
Saving the dataset (114/192 shards):  91%|█████████ | 1260308/13
1: 86595 [00:03<00:00, 587520.11 examples/s]
Saving the dataset (115/192 shards):  91%|█████████ | 1263530/1386595 [00:03<00:00, 587520.11 examples/s]
Saving the dataset (116/192 shards):  92%|█████████▏| 1271752/1386595 [00:03<00:00, 587520.11 examples/s]
Saving the dataset (117/192 shards):  92%|█████████▏| 1276974/1386595 [00:03<00:00, 587520.11 examples/s]
Saving the dataset (118/192 shards):  92%|█████████▏| 1278196/1386595 [00:03<00:00, 587520.11 examples/s]
Saving the dataset (119/192 shards):  92%|█████████▏| 1279418/1386595 [00:03<00:00, 587520.11 examples/s]
Saving the dataset (119/192 shards):  92%|█████████▏| 1280418/1386595 [00:03<00:00, 588407.30 examples/s]
Saving the dataset (120/192 shards):  92%|█████████▏| 1281640/1386595 [00:03<00:00, 588407.30 examples/s]
Saving the dataset (121/192 shards):  93%|█████████▎| 1282862/1386595 [00:03<00:00
1: , 588407.30 examples/s]
Saving the dataset (122/192 shards):  93%|█████████▎| 1287084/1386595 [00:03<00:00, 588407.30 examples/s]
Saving the dataset (123/192 shards):  93%|█████████▎| 1288306/1386595 [00:03<00:00, 588407.30 examples/s]
Saving the dataset (124/192 shards):  94%|█████████▎| 1296972/1386595 [00:03<00:00, 588407.30 examples/s]
Saving the dataset (125/192 shards):  94%|█████████▎| 1296972/1386595 [00:03<00:00, 588407.30 examples/s]
Saving the dataset (126/192 shards):  94%|█████████▎| 1296972/1386595 [00:03<00:00, 588407.30 examples/s]
Saving the dataset (127/192 shards):  94%|█████████▍| 1302194/1386595 [00:03<00:00, 588407.30 examples/s]
Saving the dataset (128/192 shards):  94%|█████████▍| 1302416/1386595 [00:03<00:00, 588407.30 examples/s]
Saving the dataset (129/192 shards):  94%|█████████▍| 1302638/1386595 [00:03<00:00, 588407.30 exam
1: ples/s]
Saving the dataset (130/192 shards):  94%|█████████▍| 1304860/1386595 [00:03<00:00, 588407.30 examples/s]
Saving the dataset (131/192 shards):  94%|█████████▍| 1306082/1386595 [00:03<00:00, 588407.30 examples/s]
Saving the dataset (132/192 shards):  94%|█████████▍| 1306304/1386595 [00:03<00:00, 588407.30 examples/s]
Saving the dataset (133/192 shards):  95%|█████████▍| 1310526/1386595 [00:03<00:00, 588407.30 examples/s]
Saving the dataset (134/192 shards):  95%|█████████▍| 1312748/1386595 [00:03<00:00, 588407.30 examples/s]
Saving the dataset (135/192 shards):  95%|█████████▍| 1315970/1386595 [00:03<00:00, 588407.30 examples/s]
Saving the dataset (136/192 shards):  95%|█████████▌| 1318192/1386595 [00:03<00:00, 588407.30 examples/s]
Saving the dataset (137/192 shards):  95%|█████████▌| 1318413/1386595 [00:03<00:00, 588407.30 examples/s]
Saving t
1: he dataset (138/192 shards):  95%|█████████▌| 1322635/1386595 [00:04<00:00, 588407.30 examples/s]
Saving the dataset (139/192 shards):  95%|█████████▌| 1322857/1386595 [00:04<00:00, 588407.30 examples/s]
Saving the dataset (140/192 shards):  95%|█████████▌| 1323078/1386595 [00:04<00:00, 588407.30 examples/s]
Saving the dataset (141/192 shards):  96%|█████████▌| 1325300/1386595 [00:04<00:00, 588407.30 examples/s]
Saving the dataset (142/192 shards):  96%|█████████▌| 1331743/1386595 [00:04<00:00, 588407.30 examples/s]
Saving the dataset (143/192 shards):  96%|█████████▌| 1331743/1386595 [00:04<00:00, 588407.30 examples/s]
Saving the dataset (144/192 shards):  96%|█████████▌| 1331965/1386595 [00:04<00:00, 588407.30 examples/s]
Saving the dataset (145/192 shards):  96%|█████████▌| 1334187/1386595 [00:04<00:00, 588407.30 examples/s]
Saving the dataset (146/
1: 192 shards):  96%|█████████▋| 1335409/1386595 [00:04<00:00, 588407.30 examples/s]
Saving the dataset (147/192 shards):  96%|█████████▋| 1336631/1386595 [00:04<00:00, 588407.30 examples/s]
Saving the dataset (148/192 shards):  97%|█████████▋| 1338075/1386595 [00:04<00:00, 588407.30 examples/s]
Saving the dataset (149/192 shards):  97%|█████████▋| 1338518/1386595 [00:04<00:00, 588407.30 examples/s]
Saving the dataset (150/192 shards):  97%|█████████▋| 1338518/1386595 [00:04<00:00, 588407.30 examples/s]
Saving the dataset (151/192 shards):  97%|█████████▋| 1338740/1386595 [00:04<00:00, 588407.30 examples/s]
Saving the dataset (152/192 shards):  97%|█████████▋| 1338740/1386595 [00:04<00:00, 588407.30 examples/s]
Saving the dataset (152/192 shards):  97%|█████████▋| 1343740/1386595 [00:04<00:00, 598982.80 examples/s]
Saving the dataset (153/192 shards):  97
1: %|█████████▋| 1345962/1386595 [00:04<00:00, 598982.80 examples/s]
Saving the dataset (154/192 shards):  97%|█████████▋| 1349184/1386595 [00:04<00:00, 598982.80 examples/s]
Saving the dataset (155/192 shards):  97%|█████████▋| 1349406/1386595 [00:04<00:00, 598982.80 examples/s]
Saving the dataset (156/192 shards):  97%|█████████▋| 1350627/1386595 [00:04<00:00, 598982.80 examples/s]
Saving the dataset (157/192 shards):  97%|█████████▋| 1351848/1386595 [00:04<00:00, 598982.80 examples/s]
Saving the dataset (158/192 shards):  98%|█████████▊| 1352070/1386595 [00:04<00:00, 598982.80 examples/s]
Saving the dataset (159/192 shards):  98%|█████████▊| 1352291/1386595 [00:04<00:00, 598982.80 examples/s]
Saving the dataset (160/192 shards):  98%|█████████▊| 1353512/1386595 [00:04<00:00, 598982.80 examples/s]
Saving the dataset (161/192 shards):  98%|████�
1: �████▊| 1355733/1386595 [00:04<00:00, 598982.80 examples/s]
Saving the dataset (162/192 shards):  98%|█████████▊| 1359176/1386595 [00:04<00:00, 598982.80 examples/s]
Saving the dataset (163/192 shards):  98%|█████████▊| 1359176/1386595 [00:04<00:00, 598982.80 examples/s]
Saving the dataset (164/192 shards):  98%|█████████▊| 1359397/1386595 [00:04<00:00, 598982.80 examples/s]
Saving the dataset (165/192 shards):  98%|█████████▊| 1361619/1386595 [00:04<00:00, 598982.80 examples/s]
Saving the dataset (166/192 shards):  98%|█████████▊| 1362841/1386595 [00:04<00:00, 598982.80 examples/s]
Saving the dataset (167/192 shards):  99%|█████████▊| 1367062/1386595 [00:04<00:00, 598982.80 examples/s]
Saving the dataset (168/192 shards):  99%|█████████▊| 1367504/1386595 [00:04<00:00, 598982.80 examples/s]
Saving the dataset (169/192 shards):  99%|█████████▊
1: | 1367504/1386595 [00:04<00:00, 598982.80 examples/s]
Saving the dataset (170/192 shards):  99%|█████████▊| 1368725/1386595 [00:04<00:00, 598982.80 examples/s]
Saving the dataset (171/192 shards):  99%|█████████▊| 1368946/1386595 [00:04<00:00, 598982.80 examples/s]
Saving the dataset (172/192 shards):  99%|█████████▊| 1369167/1386595 [00:04<00:00, 598982.80 examples/s]
Saving the dataset (173/192 shards):  99%|█████████▉| 1369611/1386595 [00:04<00:00, 598982.80 examples/s]
Saving the dataset (174/192 shards):  99%|█████████▉| 1369833/1386595 [00:04<00:00, 598982.80 examples/s]
Saving the dataset (175/192 shards):  99%|█████████▉| 1369833/1386595 [00:04<00:00, 598982.80 examples/s]
Saving the dataset (176/192 shards):  99%|█████████▉| 1371276/1386595 [00:04<00:00, 598982.80 examples/s]
Saving the dataset (177/192 shards):  99%|█████████▉| 1372276/138659
1: 5 [00:04<00:00, 598982.80 examples/s]
Saving the dataset (178/192 shards):  99%|█████████▉| 1372497/1386595 [00:04<00:00, 598982.80 examples/s]
Saving the dataset (179/192 shards):  99%|█████████▉| 1373719/1386595 [00:04<00:00, 598982.80 examples/s]
Saving the dataset (180/192 shards):  99%|█████████▉| 1373940/1386595 [00:04<00:00, 598982.80 examples/s]
Saving the dataset (181/192 shards):  99%|█████████▉| 1374162/1386595 [00:04<00:00, 598982.80 examples/s]
Saving the dataset (182/192 shards):  99%|█████████▉| 1378383/1386595 [00:04<00:00, 598982.80 examples/s]
Saving the dataset (183/192 shards): 100%|█████████▉| 1381605/1386595 [00:04<00:00, 598982.80 examples/s]
Saving the dataset (184/192 shards): 100%|█████████▉| 1382048/1386595 [00:04<00:00, 598982.80 examples/s]
Saving the dataset (185/192 shards): 100%|█████████▉| 1382048/1386595 [00:04<00:00, 
1: 598982.80 examples/s]
Saving the dataset (186/192 shards): 100%|█████████▉| 1383269/1386595 [00:04<00:00, 598982.80 examples/s]
Saving the dataset (187/192 shards): 100%|█████████▉| 1383490/1386595 [00:04<00:00, 598982.80 examples/s]
Saving the dataset (188/192 shards): 100%|█████████▉| 1383932/1386595 [00:04<00:00, 598982.80 examples/s]
Saving the dataset (189/192 shards): 100%|█████████▉| 1383932/1386595 [00:04<00:00, 598982.80 examples/s]
Saving the dataset (190/192 shards): 100%|█████████▉| 1384153/1386595 [00:04<00:00, 598982.80 examples/s]
Saving the dataset (191/192 shards): 100%|█████████▉| 1386374/1386595 [00:04<00:00, 598982.80 examples/s]
Saving the dataset (192/192 shards): 100%|██████████| 1386595/1386595 [00:04<00:00, 598982.80 examples/s]
Saving the dataset (192/192 shards): 100%|██████████| 1386595/1386595 [00:04<00:00, 329115.78 exampl
1: es/s]
0: [2025-08-07 02:09:47,545] [INFO] [axolotl.utils.data.shared.load_preprocessed_dataset:471] [PID:2863199] [RANK:0] Loading prepared dataset from disk at /lustre/fsn1/projects/rech/dgo/udv55np/dataset/Qwen3-235B-A22B/Qwen2.5-1.5B/1/fe8bf9553434d780286867f3f92bed38...
0: [2025-08-07 02:12:50,623] [INFO] [axolotl.utils.samplers.multipack.calc_min_len:435] [PID:2863199] [RANK:0] gather_len_batches: [235603, 235603, 235608, 235607]
0: [2025-08-07 02:12:50,650] [INFO] [axolotl.utils.trainer.calc_sample_packing_eff_est:496] [PID:2863199] [RANK:0] sample_packing_eff_est across ranks: [0.9936812520027161, 0.9936812520027161, 0.9936686158180237, 0.9936770796775818]
0: [2025-08-07 02:12:50,672] [INFO] [axolotl.utils.data.sft._prepare_standard_dataset:123] [PID:2863199] [RANK:0] Maximum number of steps set at 14725
0: You are attempting to use Flash Attention 2.0 with a model not initialized on GPU. Make sure to move the model to GPU after initializing it on CPU with `model.to('cuda')`.
2: You are attempting to use Flash Attention 2.0 with a model not initialized on GPU. Make sure to move the model to GPU after initializing it on CPU with `model.to('cuda')`.
1: You are attempting to use Flash Attention 2.0 with a model not initialized on GPU. Make sure to move the model to GPU after initializing it on CPU with `model.to('cuda')`.
3: You are attempting to use Flash Attention 2.0 with a model not initialized on GPU. Make sure to move the model to GPU after initializing it on CPU with `model.to('cuda')`.
0: [2025-08-07 02:12:54,548] [INFO] [axolotl.loaders.model._configure_embedding_dtypes:317] [PID:2863199] [RANK:0] Converting modules to torch.bfloat16
0: [2025-08-07 02:12:59,188] [INFO] [axolotl.train.save_initial_configs:397] [PID:2863199] [RANK:0] Pre-saving tokenizer to /lustre/fswork/projects/rech/dgo/udv55np/ift/Qwen3-235B-A22B/Qwen2.5-1.5B/1...
0: [2025-08-07 02:12:59,380] [INFO] [axolotl.train.save_initial_configs:400] [PID:2863199] [RANK:0] Pre-saving model config to /lustre/fswork/projects/rech/dgo/udv55np/ift/Qwen3-235B-A22B/Qwen2.5-1.5B/1...
0: [2025-08-07 02:12:59,389] [INFO] [axolotl.train.execute_training:221] [PID:2863199] [RANK:0] Starting trainer...
0: [2025-08-07 02:30:24,689] [INFO] [axolotl.utils.samplers.multipack.calc_min_len:435] [PID:2863199] [RANK:0] gather_len_batches: [235610, 235610, 235610, 235610]
0: Parameter Offload - Persistent parameters statistics: param_count = 141, numel = 144896
0: {'loss': 1.2092, 'grad_norm': 1.9594183418996576, 'learning_rate': 1.54e-06, 'epoch': 0.0}
0: 
  0%|          | 0/14725 [00:00<?, ?it/s]
  0%|          | 1/14725 [00:57<235:09:14, 57.49s/it]
  0%|          | 2/14725 [01:00<104:39:47, 25.59s/it]
  0%|          | 3/14725 [01:04<63:20:28, 15.49s/it] 
  0%|          | 4/14725 [01:07<43:48:11, 10.71s/it]
  0%|          | 5/14725 [01:10<32:49:14,  8.03s/it]
  0%|          | 6/14725 [01:13<25:59:41,  6.36s/it]
  0%|          | 7/14725 [01:17<21:54:58,  5.36s/it]
  0%|          | 8/14725 [01:20<19:05:00,  4.67s/it]
  0%|          | 9/14725 [01:23<17:05:44,  4.18s/it]
  0%|          | 10/14725 [01:26<16:05:22,  3.94s/it]
                                                     

  0%|          | 10/14725 [01:26<16:05:22,  3.94s/it]
  0%|          | 11/14725 [01:30<15:11:23,  3.72s/it]
  0%|          | 12/14725 [01:33<14:33:43,  3.56s/it]
  0%|          | 13/14725 [01:36<14:04:40,  3.44s/it]
  0%|          | 14/14725 [01:39<13:39:21,  3.34s/it]
  0%|          | 15/14725 [01:43<13:41:55,  3.35s/it]
  0%|          | 16/14725 [01:46<13:33:23,  3.32s/it]
  0%|         
0: {'loss': 1.1335, 'grad_norm': 1.1247043259389498, 'learning_rate': 2.1400000000000003e-06, 'epoch': 0.0}
0: {'loss': 1.0846, 'grad_norm': 0.8415600839565848, 'learning_rate': 2.7400000000000004e-06, 'epoch': 0.0}
0:  | 17/14725 [01:49<13:24:48,  3.28s/it]
  0%|          | 18/14725 [01:52<13:22:21,  3.27s/it]
  0%|          | 19/14725 [01:55<13:12:04,  3.23s/it]
  0%|          | 20/14725 [01:59<13:24:10,  3.28s/it]
                                                     

  0%|          | 20/14725 [01:59<13:24:10,  3.28s/it]
  0%|          | 21/14725 [02:02<13:17:03,  3.25s/it]
  0%|          | 22/14725 [02:06<13:37:52,  3.34s/it]
  0%|          | 23/14725 [02:09<13:28:15,  3.30s/it]
  0%|          | 24/14725 [02:12<13:18:00,  3.26s/it]
  0%|          | 25/14725 [02:15<13:20:51,  3.27s/it]
  0%|          | 26/14725 [02:18<13:22:45,  3.28s/it]
  0%|          | 27/14725 [02:22<13:26:56,  3.29s/it]
  0%|          | 28/14725 [02:25<13:35:27,  3.33s/it]
  0%|          | 29/14725 [02:29<13:39:09,  3.34s/it]
  0%|          | 30/14725 [02:32<13:53:59,  3.41s/it]
                                                     

  0%|          | 30/14725 [02:32<13:53:59,  3.41s/it]
  0%|          | 31/14725 [02:36<13:50:16,  3.39s/it]
  0%|     
0: {'loss': 1.02, 'grad_norm': 0.7324853935739636, 'learning_rate': 3.3400000000000006e-06, 'epoch': 0.0}
0:      | 32/14725 [02:39<13:51:55,  3.40s/it]
  0%|          | 33/14725 [02:42<14:03:18,  3.44s/it]
  0%|          | 34/14725 [02:46<13:43:23,  3.36s/it]
  0%|          | 35/14725 [02:49<13:33:24,  3.32s/it]
  0%|          | 36/14725 [02:52<13:26:43,  3.30s/it]
  0%|          | 37/14725 [02:55<13:23:39,  3.28s/it]
  0%|          | 38/14725 [02:59<13:25:11,  3.29s/it]
  0%|          | 39/14725 [03:02<13:29:09,  3.31s/it]
  0%|          | 40/14725 [03:05<13:19:43,  3.27s/it]
                                                     

  0%|          | 40/14725 [03:05<13:19:43,  3.27s/it]
  0%|          | 41/14725 [03:09<13:24:42,  3.29s/it]
  0%|          | 42/14725 [03:12<13:24:53,  3.29s/it]
  0%|          | 43/14725 [03:15<13:33:15,  3.32s/it]
  0%|          | 44/14725 [03:19<13:32:18,  3.32s/it]
  0%|          | 45/14725 [03:22<13:30:40,  3.31s/it]
  0%|          | 46/14725 [03:25<13:17:13,  3.26s/it]
  0%|          | 47/14725 [03:28<13:14:17,  3.25s/it]
  0%|          | 48/14725 [03:31<13:16:04,  3.25s/it]
  0%|  
0: {'loss': 0.982, 'grad_norm': 0.49900411497380254, 'learning_rate': 3.94e-06, 'epoch': 0.0}
0: {'loss': 0.9532, 'grad_norm': 0.49169372629634356, 'learning_rate': 4.54e-06, 'epoch': 0.0}
0:         | 49/14725 [03:35<13:19:15,  3.27s/it]
  0%|          | 50/14725 [03:38<13:04:15,  3.21s/it]
                                                     

  0%|          | 50/14725 [03:38<13:04:15,  3.21s/it]
  0%|          | 51/14725 [03:41<13:15:03,  3.25s/it]
  0%|          | 52/14725 [03:44<13:00:31,  3.19s/it]
  0%|          | 53/14725 [03:47<12:57:33,  3.18s/it]
  0%|          | 54/14725 [03:50<12:53:13,  3.16s/it]
  0%|          | 55/14725 [03:54<12:58:59,  3.19s/it]
  0%|          | 56/14725 [03:57<13:06:05,  3.22s/it]
  0%|          | 57/14725 [04:00<13:21:53,  3.28s/it]
  0%|          | 58/14725 [04:04<13:40:37,  3.36s/it]
  0%|          | 59/14725 [04:07<13:41:13,  3.36s/it]
  0%|          | 60/14725 [04:11<13:39:35,  3.35s/it]
                                                     

  0%|          | 60/14725 [04:11<13:39:35,  3.35s/it]
  0%|          | 61/14725 [04:14<13:24:44,  3.29s/it]
  0%|          | 62/14725 [04:17<13:20:24,  3.28s/it]
  0%|          | 63/14725 [04:20<13:26:55,  3.30s/it]
  0
0: {'loss': 0.9502, 'grad_norm': 0.47700935113125725, 'learning_rate': 5.140000000000001e-06, 'epoch': 0.0}
0: %|          | 64/14725 [04:24<13:26:44,  3.30s/it]
  0%|          | 65/14725 [04:27<13:21:21,  3.28s/it]
  0%|          | 66/14725 [04:30<13:27:30,  3.31s/it]
  0%|          | 67/14725 [04:34<13:21:50,  3.28s/it]
  0%|          | 68/14725 [04:37<13:19:08,  3.27s/it]
  0%|          | 69/14725 [04:40<13:22:10,  3.28s/it]
  0%|          | 70/14725 [04:43<13:18:11,  3.27s/it]
                                                     

  0%|          | 70/14725 [04:43<13:18:11,  3.27s/it]
  0%|          | 71/14725 [04:47<13:24:00,  3.29s/it]
  0%|          | 72/14725 [04:50<13:20:30,  3.28s/it]
  0%|          | 73/14725 [04:53<13:24:18,  3.29s/it]
  1%|          | 74/14725 [04:57<13:33:12,  3.33s/it]
  1%|          | 75/14725 [05:00<13:25:50,  3.30s/it]
  1%|          | 76/14725 [05:03<13:27:57,  3.31s/it]
  1%|          | 77/14725 [05:06<13:07:37,  3.23s/it]
  1%|          | 78/14725 [05:10<13:18:34,  3.27s/it]
  1%|          | 79/14725 [05:13<13:05:10,  3.22s/it]
  1%|          | 80/14725 [05:16<13:21:09,  3.28s/it]
0: {'loss': 0.9239, 'grad_norm': 0.49630566939777965, 'learning_rate': 5.74e-06, 'epoch': 0.01}
0: {'loss': 0.9214, 'grad_norm': 0.5914538005180097, 'learning_rate': 6.34e-06, 'epoch': 0.01}
0:                                                      

  1%|          | 80/14725 [05:16<13:21:09,  3.28s/it]
  1%|          | 81/14725 [05:19<13:13:33,  3.25s/it]
  1%|          | 82/14725 [05:23<13:19:48,  3.28s/it]
  1%|          | 83/14725 [05:26<13:14:51,  3.26s/it]
  1%|          | 84/14725 [05:29<13:18:24,  3.27s/it]
  1%|          | 85/14725 [05:32<13:02:14,  3.21s/it]
  1%|          | 86/14725 [05:36<13:18:12,  3.27s/it]
  1%|          | 87/14725 [05:39<13:06:31,  3.22s/it]
  1%|          | 88/14725 [05:42<13:05:42,  3.22s/it]
  1%|          | 89/14725 [05:45<13:13:21,  3.25s/it]
  1%|          | 90/14725 [05:49<13:35:51,  3.34s/it]
                                                     

  1%|          | 90/14725 [05:49<13:35:51,  3.34s/it]
  1%|          | 91/14725 [05:52<13:24:04,  3.30s/it]
  1%|          | 92/14725 [05:55<13:19:19,  3.28s/it]
  1%|          | 93/14725 [05:59<13:18:15,  3.27s/it]
  1%|          | 94/14725 [06:02<13:16:46,  3.27s/it]
  1%|          | 95/14725 [06:05<13:12:37,  3.25s/
0: {'loss': 0.903, 'grad_norm': 0.5599678750945721, 'learning_rate': 6.940000000000001e-06, 'epoch': 0.01}
0: {'loss': 0.9061, 'grad_norm': 0.474843650803078, 'learning_rate': 7.540000000000001e-06, 'epoch': 0.01}
0: it]
  1%|          | 96/14725 [06:08<13:10:10,  3.24s/it]
  1%|          | 97/14725 [06:12<13:14:22,  3.26s/it]
  1%|          | 98/14725 [06:15<13:12:41,  3.25s/it]
  1%|          | 99/14725 [06:18<12:59:13,  3.20s/it]
  1%|          | 100/14725 [06:21<13:07:29,  3.23s/it]
                                                      

  1%|          | 100/14725 [06:21<13:07:29,  3.23s/it]
  1%|          | 101/14725 [06:25<13:19:05,  3.28s/it]
  1%|          | 102/14725 [06:28<13:24:15,  3.30s/it]
  1%|          | 103/14725 [06:31<13:24:40,  3.30s/it]
  1%|          | 104/14725 [06:35<13:22:53,  3.29s/it]
  1%|          | 105/14725 [06:38<13:16:03,  3.27s/it]
  1%|          | 106/14725 [06:41<13:09:36,  3.24s/it]
  1%|          | 107/14725 [06:44<13:20:22,  3.29s/it]
  1%|          | 108/14725 [06:48<13:23:59,  3.30s/it]
  1%|          | 109/14725 [06:51<13:27:59,  3.32s/it]
  1%|          | 110/14725 [06:54<13:21:27,  3.29s/it]
                                                      

  1%|          | 110/14725 [06:5
0: {'loss': 0.888, 'grad_norm': 0.5200916735172596, 'learning_rate': 8.14e-06, 'epoch': 0.01}
0: 4<13:21:27,  3.29s/it]
  1%|          | 111/14725 [06:58<13:27:52,  3.32s/it]
  1%|          | 112/14725 [07:01<13:24:59,  3.31s/it]
  1%|          | 113/14725 [07:04<13:06:38,  3.23s/it]
  1%|          | 114/14725 [07:07<12:58:04,  3.20s/it]
  1%|          | 115/14725 [07:10<13:01:35,  3.21s/it]
  1%|          | 116/14725 [07:14<13:21:14,  3.29s/it]
  1%|          | 117/14725 [07:17<13:09:37,  3.24s/it]
  1%|          | 118/14725 [07:20<13:10:10,  3.25s/it]
  1%|          | 119/14725 [07:23<13:09:21,  3.24s/it]
  1%|          | 120/14725 [07:27<13:03:09,  3.22s/it]
                                                      

  1%|          | 120/14725 [07:27<13:03:09,  3.22s/it]
  1%|          | 121/14725 [07:30<12:59:13,  3.20s/it]
  1%|          | 122/14725 [07:33<13:00:17,  3.21s/it]
  1%|          | 123/14725 [07:36<13:12:29,  3.26s/it]
  1%|          | 124/14725 [07:40<13:18:26,  3.28s/it]
  1%|          | 125/14725 [07:43<13:13:58,  3.26s/it]
  1%|          | 126/14725 [07:46<13:14:04,  3.26s/it]
  1%|     
0: {'loss': 0.8883, 'grad_norm': 0.4862375641315234, 'learning_rate': 8.740000000000001e-06, 'epoch': 0.01}
0: {'loss': 0.8609, 'grad_norm': 0.530715714083786, 'learning_rate': 9.34e-06, 'epoch': 0.01}
0:      | 127/14725 [07:49<13:05:50,  3.23s/it]
  1%|          | 128/14725 [07:52<12:46:05,  3.15s/it]
  1%|          | 129/14725 [07:56<12:56:23,  3.19s/it]
  1%|          | 130/14725 [07:59<12:58:54,  3.20s/it]
                                                      

  1%|          | 130/14725 [07:59<12:58:54,  3.20s/it]
  1%|          | 131/14725 [08:02<13:06:05,  3.23s/it]
  1%|          | 132/14725 [08:05<12:56:14,  3.19s/it]
  1%|          | 133/14725 [08:08<12:54:46,  3.19s/it]
  1%|          | 134/14725 [08:12<13:06:44,  3.24s/it]
  1%|          | 135/14725 [08:15<13:17:24,  3.28s/it]
  1%|          | 136/14725 [08:18<13:25:00,  3.31s/it]
  1%|          | 137/14725 [08:22<13:49:14,  3.41s/it]
  1%|          | 138/14725 [08:25<13:48:34,  3.41s/it]
  1%|          | 139/14725 [08:29<13:32:26,  3.34s/it]
  1%|          | 140/14725 [08:32<13:38:05,  3.37s/it]
                                                      

  1%|          | 140/14725 [08:32<13:38:05,  3.37s/it]
  1%|          | 141/14725 [08:36<13:47:11
0: {'loss': 0.8772, 'grad_norm': 0.5952212055954214, 'learning_rate': 9.940000000000001e-06, 'epoch': 0.01}
0: ,  3.40s/it]
  1%|          | 142/14725 [08:39<13:43:30,  3.39s/it]
  1%|          | 143/14725 [08:42<13:29:15,  3.33s/it]
  1%|          | 144/14725 [08:46<13:39:06,  3.37s/it]
  1%|          | 145/14725 [08:49<13:25:12,  3.31s/it]
  1%|          | 146/14725 [08:52<13:35:35,  3.36s/it]
  1%|          | 147/14725 [08:55<13:29:02,  3.33s/it]
  1%|          | 148/14725 [08:59<13:30:40,  3.34s/it]
  1%|          | 149/14725 [09:02<13:27:47,  3.33s/it]
  1%|          | 150/14725 [09:05<13:20:23,  3.29s/it]
                                                      

  1%|          | 150/14725 [09:05<13:20:23,  3.29s/it]
  1%|          | 151/14725 [09:08<13:03:53,  3.23s/it]
  1%|          | 152/14725 [09:12<12:54:27,  3.19s/it]
  1%|          | 153/14725 [09:15<13:11:20,  3.26s/it]
  1%|          | 154/14725 [09:18<13:05:20,  3.23s/it]
  1%|          | 155/14725 [09:21<13:03:42,  3.23s/it]
  1%|          | 156/14725 [09:25<13:00:12,  3.21s/it]
  1%|          | 157/14725 [09:28<13:09:21,  3.25s/it]
  1%|          | 158
0: {'loss': 0.8614, 'grad_norm': 0.529570089590535, 'learning_rate': 1e-05, 'epoch': 0.01}
0: {'loss': 0.8597, 'grad_norm': 0.5418334667435569, 'learning_rate': 1e-05, 'epoch': 0.01}
0: /14725 [09:31<12:56:42,  3.20s/it]
  1%|          | 159/14725 [09:34<13:14:35,  3.27s/it]
  1%|          | 160/14725 [09:38<13:34:39,  3.36s/it]
                                                      

  1%|          | 160/14725 [09:38<13:34:39,  3.36s/it]
  1%|          | 161/14725 [09:41<13:32:06,  3.35s/it]
  1%|          | 162/14725 [09:44<13:17:59,  3.29s/it]
  1%|          | 163/14725 [09:48<13:18:17,  3.29s/it]
  1%|          | 164/14725 [09:51<13:14:55,  3.28s/it]
  1%|          | 165/14725 [09:54<13:27:45,  3.33s/it]
  1%|          | 166/14725 [09:58<13:36:33,  3.37s/it]
  1%|          | 167/14725 [10:01<13:35:22,  3.36s/it]
  1%|          | 168/14725 [10:04<13:28:44,  3.33s/it]
  1%|          | 169/14725 [10:08<13:18:21,  3.29s/it]
  1%|          | 170/14725 [10:11<13:14:44,  3.28s/it]
                                                      

  1%|          | 170/14725 [10:11<13:14:44,  3.28s/it]
  1%|          | 171/14725 [10:14<13:04:12,  3.23s/it]
  1%|          | 172/14725 [10:17<13:06:50,  3.24s/i
0: {'loss': 0.8577, 'grad_norm': 0.5296876682366072, 'learning_rate': 1e-05, 'epoch': 0.01}
0: t]
  1%|          | 173/14725 [10:21<13:04:18,  3.23s/it]
  1%|          | 174/14725 [10:24<12:59:43,  3.22s/it]
  1%|          | 175/14725 [10:27<13:05:40,  3.24s/it]
  1%|          | 176/14725 [10:30<13:04:49,  3.24s/it]
  1%|          | 177/14725 [10:34<13:15:20,  3.28s/it]
  1%|          | 178/14725 [10:37<13:14:08,  3.28s/it]
  1%|          | 179/14725 [10:40<13:20:04,  3.30s/it]
  1%|          | 180/14725 [10:43<13:10:07,  3.26s/it]
                                                      

  1%|          | 180/14725 [10:43<13:10:07,  3.26s/it]
  1%|          | 181/14725 [10:47<13:02:35,  3.23s/it]
  1%|          | 182/14725 [10:50<13:00:27,  3.22s/it]
  1%|          | 183/14725 [10:53<13:08:02,  3.25s/it]
  1%|          | 184/14725 [10:56<13:04:58,  3.24s/it]
  1%|▏         | 185/14725 [10:59<13:04:26,  3.24s/it]
  1%|▏         | 186/14725 [11:03<12:57:40,  3.21s/it]
  1%|▏         | 187/14725 [11:06<13:06:35,  3.25s/it]
  1%|▏         | 188/14725 [11:09<13:07:17,  3.25s/it]
  1%|▏         | 189
0: {'loss': 0.8545, 'grad_norm': 0.6097684957037094, 'learning_rate': 1e-05, 'epoch': 0.01}
0: {'loss': 0.8269, 'grad_norm': 0.6131861791768415, 'learning_rate': 1e-05, 'epoch': 0.01}
0: /14725 [11:13<13:21:55,  3.31s/it]
  1%|▏         | 190/14725 [11:16<13:25:11,  3.32s/it]
                                                      

  1%|▏         | 190/14725 [11:16<13:25:11,  3.32s/it]
  1%|▏         | 191/14725 [11:19<13:16:48,  3.29s/it]
  1%|▏         | 192/14725 [11:23<13:18:26,  3.30s/it]
  1%|▏         | 193/14725 [11:26<13:28:09,  3.34s/it]
  1%|▏         | 194/14725 [11:29<13:17:31,  3.29s/it]
  1%|▏         | 195/14725 [11:32<13:14:52,  3.28s/it]
  1%|▏         | 196/14725 [11:36<13:22:35,  3.31s/it]
  1%|▏         | 197/14725 [11:39<13:20:29,  3.31s/it]
  1%|▏         | 198/14725 [11:42<13:17:21,  3.29s/it]
  1%|▏         | 199/14725 [11:46<13:13:13,  3.28s/it]
  1%|▏         | 200/14725 [11:49<13:18:13,  3.30s/it]
                                                      

  1%|▏         | 200/14725 [11:49<13:18:13,  3.30s/it]
  1%|▏         | 201/14725 [11:52<13:11:43,  3.27s/it]
  1%|▏         | 202/14725 [11:55<13:07:43,  3.25s/it]
  1%|▏         | 203
0: {'loss': 0.8227, 'grad_norm': 0.5255704550896746, 'learning_rate': 1e-05, 'epoch': 0.01}
0: /14725 [11:59<13:18:49,  3.30s/it]
  1%|▏         | 204/14725 [12:02<13:03:48,  3.24s/it]
  1%|▏         | 205/14725 [12:05<13:13:00,  3.28s/it]
  1%|▏         | 206/14725 [12:09<13:23:17,  3.32s/it]
  1%|▏         | 207/14725 [12:12<13:27:16,  3.34s/it]
  1%|▏         | 208/14725 [12:15<13:26:35,  3.33s/it]
  1%|▏         | 209/14725 [12:19<13:15:47,  3.29s/it]
  1%|▏         | 210/14725 [12:22<13:27:15,  3.34s/it]
                                                      

  1%|▏         | 210/14725 [12:22<13:27:15,  3.34s/it]
  1%|▏         | 211/14725 [12:25<13:17:20,  3.30s/it]
  1%|▏         | 212/14725 [12:28<13:15:19,  3.29s/it]
  1%|▏         | 213/14725 [12:32<13:11:37,  3.27s/it]
  1%|▏         | 214/14725 [12:35<12:58:14,  3.22s/it]
  1%|▏         | 215/14725 [12:38<13:07:59,  3.26s/it]
  1%|▏         | 216/14725 [12:42<13:19:08,  3.30s/it]
  1%|▏         | 217/14725 [12:45<13:30:19,  3.35s/it]
  1%|▏         | 218/14725 [12:49<13:39:03,  3.39s/it]
  1%|▏         | 21
0: {'loss': 0.8434, 'grad_norm': 0.5393127005117994, 'learning_rate': 1e-05, 'epoch': 0.01}
0: {'loss': 0.8443, 'grad_norm': 0.59876728393907, 'learning_rate': 1e-05, 'epoch': 0.02}
0: 9/14725 [12:52<13:34:41,  3.37s/it]
  1%|▏         | 220/14725 [12:55<13:29:42,  3.35s/it]
                                                      

  1%|▏         | 220/14725 [12:55<13:29:42,  3.35s/it]
  2%|▏         | 221/14725 [12:58<13:23:38,  3.32s/it]
  2%|▏         | 222/14725 [13:02<13:23:34,  3.32s/it]
  2%|▏         | 223/14725 [13:05<13:27:36,  3.34s/it]
  2%|▏         | 224/14725 [13:08<13:14:53,  3.29s/it]
  2%|▏         | 225/14725 [13:12<13:11:51,  3.28s/it]
  2%|▏         | 226/14725 [13:15<13:28:24,  3.35s/it]
  2%|▏         | 227/14725 [13:18<13:33:30,  3.37s/it]
  2%|▏         | 228/14725 [13:22<13:36:20,  3.38s/it]
  2%|▏         | 229/14725 [13:25<13:23:00,  3.32s/it]
  2%|▏         | 230/14725 [13:28<13:21:10,  3.32s/it]
                                                      

  2%|▏         | 230/14725 [13:28<13:21:10,  3.32s/it]
  2%|▏         | 231/14725 [13:32<13:19:52,  3.31s/it]
  2%|▏         | 232/14725 [13:35<13:09:48,  3.27s/it]
  2%|▏         | 23
0: {'loss': 0.8426, 'grad_norm': 0.4693754623500759, 'learning_rate': 1e-05, 'epoch': 0.02}
0: 3/14725 [13:38<13:20:19,  3.31s/it]
  2%|▏         | 234/14725 [13:42<13:28:25,  3.35s/it]
  2%|▏         | 235/14725 [13:45<13:16:52,  3.30s/it]
  2%|▏         | 236/14725 [13:48<13:11:37,  3.28s/it]
  2%|▏         | 237/14725 [13:51<12:59:00,  3.23s/it]
  2%|▏         | 238/14725 [13:54<12:53:46,  3.20s/it]
  2%|▏         | 239/14725 [13:58<12:59:27,  3.23s/it]
  2%|▏         | 240/14725 [14:01<13:04:49,  3.25s/it]
                                                      

  2%|▏         | 240/14725 [14:01<13:04:49,  3.25s/it]
  2%|▏         | 241/14725 [14:04<13:17:35,  3.30s/it]
  2%|▏         | 242/14725 [14:08<13:15:57,  3.30s/it]
  2%|▏         | 243/14725 [14:11<13:14:46,  3.29s/it]
  2%|▏         | 244/14725 [14:14<13:02:35,  3.24s/it]
  2%|▏         | 245/14725 [14:17<13:13:56,  3.29s/it]
  2%|▏         | 246/14725 [14:21<13:14:50,  3.29s/it]
  2%|▏         | 247/14725 [14:24<13:27:28,  3.35s/it]
  2%|▏         | 248/14725 [14:28<13:30:22,  3.36s/it]
  2%|▏         | 2
0: {'loss': 0.8611, 'grad_norm': 0.554048552884994, 'learning_rate': 1e-05, 'epoch': 0.02}
0: {'loss': 0.8501, 'grad_norm': 0.48881175687172607, 'learning_rate': 1e-05, 'epoch': 0.02}
0: 49/14725 [14:31<13:30:30,  3.36s/it]
  2%|▏         | 250/14725 [14:34<13:27:54,  3.35s/it]
                                                      

  2%|▏         | 250/14725 [14:34<13:27:54,  3.35s/it]
  2%|▏         | 251/14725 [14:38<13:23:08,  3.33s/it]
  2%|▏         | 252/14725 [14:41<13:09:16,  3.27s/it]
  2%|▏         | 253/14725 [14:44<13:04:46,  3.25s/it]
  2%|▏         | 254/14725 [14:47<12:50:37,  3.20s/it]
  2%|▏         | 255/14725 [14:50<12:56:49,  3.22s/it]
  2%|▏         | 256/14725 [14:54<13:13:50,  3.29s/it]
  2%|▏         | 257/14725 [14:57<13:18:55,  3.31s/it]
  2%|▏         | 258/14725 [15:00<13:16:24,  3.30s/it]
  2%|▏         | 259/14725 [15:04<13:14:18,  3.29s/it]
  2%|▏         | 260/14725 [15:07<13:04:35,  3.25s/it]
                                                      

  2%|▏         | 260/14725 [15:07<13:04:35,  3.25s/it]
  2%|▏         | 261/14725 [15:10<13:07:09,  3.27s/it]
  2%|▏         | 262/14725 [15:14<13:20:04,  3.32s/it]
  2%|▏         | 2
0: {'loss': 0.8537, 'grad_norm': 0.510950235824703, 'learning_rate': 1e-05, 'epoch': 0.02}
0: 63/14725 [15:17<13:22:35,  3.33s/it]
  2%|▏         | 264/14725 [15:20<13:16:30,  3.30s/it]
  2%|▏         | 265/14725 [15:24<13:22:08,  3.33s/it]
  2%|▏         | 266/14725 [15:27<13:10:40,  3.28s/it]
  2%|▏         | 267/14725 [15:30<13:28:33,  3.36s/it]
  2%|▏         | 268/14725 [15:33<13:13:30,  3.29s/it]
  2%|▏         | 269/14725 [15:37<13:12:23,  3.29s/it]
  2%|▏         | 270/14725 [15:40<13:05:49,  3.26s/it]
                                                      

  2%|▏         | 270/14725 [15:40<13:05:49,  3.26s/it]
  2%|▏         | 271/14725 [15:43<13:25:56,  3.35s/it]
  2%|▏         | 272/14725 [15:47<13:18:20,  3.31s/it]
  2%|▏         | 273/14725 [15:50<13:16:09,  3.31s/it]
  2%|▏         | 274/14725 [15:53<13:13:21,  3.29s/it]
  2%|▏         | 275/14725 [15:56<13:01:30,  3.24s/it]
  2%|▏         | 276/14725 [16:00<13:02:21,  3.25s/it]
  2%|▏         | 277/14725 [16:03<12:55:58,  3.22s/it]
  2%|▏         | 278/14725 [16:06<12:43:24,  3.17s/it]
  2%|▏         | 
0: {'loss': 0.8226, 'grad_norm': 0.5055336151558408, 'learning_rate': 1e-05, 'epoch': 0.02}
0: {'loss': 0.8247, 'grad_norm': 0.5549378458898869, 'learning_rate': 1e-05, 'epoch': 0.02}
0: 279/14725 [16:09<12:56:21,  3.22s/it]
  2%|▏         | 280/14725 [16:12<12:59:15,  3.24s/it]
                                                      

  2%|▏         | 280/14725 [16:12<12:59:15,  3.24s/it]
  2%|▏         | 281/14725 [16:16<12:51:08,  3.20s/it]
  2%|▏         | 282/14725 [16:19<13:02:32,  3.25s/it]
  2%|▏         | 283/14725 [16:22<13:14:56,  3.30s/it]
  2%|▏         | 284/14725 [16:26<13:18:13,  3.32s/it]
  2%|▏         | 285/14725 [16:29<13:01:46,  3.25s/it]
  2%|▏         | 286/14725 [16:32<13:08:43,  3.28s/it]
  2%|▏         | 287/14725 [16:35<13:09:14,  3.28s/it]
  2%|▏         | 288/14725 [16:39<13:17:34,  3.31s/it]
  2%|▏         | 289/14725 [16:42<13:10:34,  3.29s/it]
  2%|▏         | 290/14725 [16:45<13:01:31,  3.25s/it]
                                                      

  2%|▏         | 290/14725 [16:45<13:01:31,  3.25s/it]
  2%|▏         | 291/14725 [16:48<13:00:58,  3.25s/it]
  2%|▏         | 292/14725 [16:52<13:19:13,  3.32s/it]
  2%|▏         | 
0: {'loss': 0.8367, 'grad_norm': 0.4989714073362891, 'learning_rate': 1e-05, 'epoch': 0.02}
0: 293/14725 [16:55<13:11:40,  3.29s/it]
  2%|▏         | 294/14725 [16:59<13:23:21,  3.34s/it]
  2%|▏         | 295/14725 [17:02<13:32:00,  3.38s/it]
  2%|▏         | 296/14725 [17:05<13:19:32,  3.32s/it]
  2%|▏         | 297/14725 [17:09<13:18:29,  3.32s/it]
  2%|▏         | 298/14725 [17:12<13:41:16,  3.42s/it]
  2%|▏         | 299/14725 [17:16<13:39:14,  3.41s/it]
  2%|▏         | 300/14725 [17:19<13:39:19,  3.41s/it]
                                                      

  2%|▏         | 300/14725 [17:19<13:39:19,  3.41s/it]
  2%|▏         | 301/14725 [17:23<14:03:18,  3.51s/it]
  2%|▏         | 302/14725 [17:26<13:39:08,  3.41s/it]
  2%|▏         | 303/14725 [17:30<14:08:19,  3.53s/it]
  2%|▏         | 304/14725 [17:33<13:53:05,  3.47s/it]
  2%|▏         | 305/14725 [17:36<13:35:24,  3.39s/it]
  2%|▏         | 306/14725 [17:40<14:22:35,  3.59s/it]
  2%|▏         | 307/14725 [17:44<14:28:01,  3.61s/it]
  2%|▏         | 308/14725 [17:47<14:06:09,  3.52s/it]
  2%|▏         |
0: {'loss': 0.8242, 'grad_norm': 0.4783757300409601, 'learning_rate': 1e-05, 'epoch': 0.02}
0: {'loss': 0.8408, 'grad_norm': 0.5352856017919213, 'learning_rate': 1e-05, 'epoch': 0.02}
0:  309/14725 [17:51<14:01:47,  3.50s/it]
  2%|▏         | 310/14725 [17:54<13:46:39,  3.44s/it]
                                                      

  2%|▏         | 310/14725 [17:54<13:46:39,  3.44s/it]
  2%|▏         | 311/14725 [17:57<13:22:49,  3.34s/it]
  2%|▏         | 312/14725 [18:01<13:54:10,  3.47s/it]
  2%|▏         | 313/14725 [18:05<14:36:32,  3.65s/it]
  2%|▏         | 314/14725 [18:09<14:52:30,  3.72s/it]
  2%|▏         | 315/14725 [18:13<15:36:26,  3.90s/it]
  2%|▏         | 316/14725 [18:17<14:58:07,  3.74s/it]
  2%|▏         | 317/14725 [18:20<14:56:17,  3.73s/it]
  2%|▏         | 318/14725 [18:24<14:54:34,  3.73s/it]
  2%|▏         | 319/14725 [18:27<14:11:58,  3.55s/it]
  2%|▏         | 320/14725 [18:31<14:58:14,  3.74s/it]
                                                      

  2%|▏         | 320/14725 [18:31<14:58:14,  3.74s/it]
  2%|▏         | 321/14725 [18:35<15:25:10,  3.85s/it]
  2%|▏         | 322/14725 [18:39<15:21:55,  3.84s/it]
  2%|▏         |
0: {'loss': 0.8343, 'grad_norm': 0.5787778431128834, 'learning_rate': 1e-05, 'epoch': 0.02}
0:  323/14725 [18:43<15:17:58,  3.82s/it]
  2%|▏         | 324/14725 [18:46<14:35:15,  3.65s/it]
  2%|▏         | 325/14725 [18:51<15:44:26,  3.94s/it]
  2%|▏         | 326/14725 [18:55<15:56:11,  3.98s/it]
  2%|▏         | 327/14725 [18:58<15:03:24,  3.76s/it]
  2%|▏         | 328/14725 [19:04<17:24:21,  4.35s/it]
  2%|▏         | 329/14725 [19:08<16:31:49,  4.13s/it]
  2%|▏         | 330/14725 [19:13<18:39:44,  4.67s/it]
                                                      

  2%|▏         | 330/14725 [19:13<18:39:44,  4.67s/it]
  2%|▏         | 331/14725 [19:17<17:01:04,  4.26s/it]
  2%|▏         | 332/14725 [19:22<17:56:39,  4.49s/it]
  2%|▏         | 333/14725 [19:26<17:27:22,  4.37s/it]
  2%|▏         | 334/14725 [19:30<17:25:52,  4.36s/it]
  2%|▏         | 335/14725 [19:36<18:37:47,  4.66s/it]
  2%|▏         | 336/14725 [19:39<17:08:10,  4.29s/it]
  2%|▏         | 337/14725 [19:44<18:04:55,  4.52s/it]
  2%|▏         | 338/14725 [19:47<16:46:45,  4.20s/it]
  2%|▏         
0: {'loss': 0.8215, 'grad_norm': 0.5248074055850693, 'learning_rate': 1e-05, 'epoch': 0.02}
0: {'loss': 0.8167, 'grad_norm': 0.5161065644526944, 'learning_rate': 1e-05, 'epoch': 0.02}
0: | 339/14725 [19:51<16:03:42,  4.02s/it]
  2%|▏         | 340/14725 [19:55<15:43:41,  3.94s/it]
                                                      

  2%|▏         | 340/14725 [19:55<15:43:41,  3.94s/it]
  2%|▏         | 341/14725 [19:58<15:19:52,  3.84s/it]
  2%|▏         | 342/14725 [20:02<15:13:30,  3.81s/it]
  2%|▏         | 343/14725 [20:06<14:39:57,  3.67s/it]
  2%|▏         | 344/14725 [20:09<13:53:30,  3.48s/it]
  2%|▏         | 345/14725 [20:12<14:00:40,  3.51s/it]
  2%|▏         | 346/14725 [20:16<13:55:56,  3.49s/it]
  2%|▏         | 347/14725 [20:19<13:52:46,  3.48s/it]
  2%|▏         | 348/14725 [20:23<14:00:05,  3.51s/it]
  2%|▏         | 349/14725 [20:26<13:54:05,  3.48s/it]
  2%|▏         | 350/14725 [20:29<13:26:45,  3.37s/it]
                                                      

  2%|▏         | 350/14725 [20:29<13:26:45,  3.37s/it]
  2%|▏         | 351/14725 [20:32<13:11:23,  3.30s/it]
  2%|▏         | 352/14725 [20:36<13:10:48,  3.30s/it]
  2%|▏         
0: {'loss': 0.8271, 'grad_norm': 0.5704101334399515, 'learning_rate': 1e-05, 'epoch': 0.02}
0: | 353/14725 [20:39<13:08:03,  3.29s/it]
  2%|▏         | 354/14725 [20:42<13:08:26,  3.29s/it]
  2%|▏         | 355/14725 [20:45<13:06:43,  3.28s/it]
  2%|▏         | 356/14725 [20:49<13:13:39,  3.31s/it]
  2%|▏         | 357/14725 [20:52<13:37:43,  3.41s/it]
  2%|▏         | 358/14725 [20:56<13:20:26,  3.34s/it]
  2%|▏         | 359/14725 [20:59<13:24:47,  3.36s/it]
  2%|▏         | 360/14725 [21:02<13:16:40,  3.33s/it]
                                                      

  2%|▏         | 360/14725 [21:02<13:16:40,  3.33s/it]
  2%|▏         | 361/14725 [21:06<13:13:59,  3.32s/it]
  2%|▏         | 362/14725 [21:09<13:04:31,  3.28s/it]
  2%|▏         | 363/14725 [21:12<12:56:40,  3.24s/it]
  2%|▏         | 364/14725 [21:15<12:49:27,  3.21s/it]
  2%|▏         | 365/14725 [21:19<13:16:28,  3.33s/it]
  2%|▏         | 366/14725 [21:22<13:19:02,  3.34s/it]
  2%|▏         | 367/14725 [21:25<13:13:16,  3.31s/it]
  2%|▏         | 368/14725 [21:29<13:18:29,  3.34s/it]
  3%|▎        
0: {'loss': 0.8328, 'grad_norm': 0.525379990011747, 'learning_rate': 1e-05, 'epoch': 0.03}
0: {'loss': 0.8157, 'grad_norm': 0.5499611409071449, 'learning_rate': 1e-05, 'epoch': 0.03}
0:  | 369/14725 [21:32<13:09:15,  3.30s/it]
  3%|▎         | 370/14725 [21:35<13:12:00,  3.31s/it]
                                                      

  3%|▎         | 370/14725 [21:35<13:12:00,  3.31s/it]
  3%|▎         | 371/14725 [21:38<13:03:20,  3.27s/it]
  3%|▎         | 372/14725 [21:42<13:02:10,  3.27s/it]
  3%|▎         | 373/14725 [21:45<12:55:43,  3.24s/it]
  3%|▎         | 374/14725 [21:48<13:10:16,  3.30s/it]
  3%|▎         | 375/14725 [21:52<13:15:45,  3.33s/it]
  3%|▎         | 376/14725 [21:55<13:08:09,  3.30s/it]
  3%|▎         | 377/14725 [21:58<12:59:14,  3.26s/it]
  3%|▎         | 378/14725 [22:01<12:53:26,  3.23s/it]
  3%|▎         | 379/14725 [22:05<12:57:15,  3.25s/it]
  3%|▎         | 380/14725 [22:08<13:02:34,  3.27s/it]
                                                      

  3%|▎         | 380/14725 [22:08<13:02:34,  3.27s/it]
  3%|▎         | 381/14725 [22:11<13:00:50,  3.27s/it]
  3%|▎         | 382/14725 [22:15<13:13:30,  3.32s/it]
  3%|▎        
0: {'loss': 0.8104, 'grad_norm': 0.5246273147974936, 'learning_rate': 1e-05, 'epoch': 0.03}
0:  | 383/14725 [22:18<13:10:04,  3.31s/it]
  3%|▎         | 384/14725 [22:21<13:10:03,  3.31s/it]
  3%|▎         | 385/14725 [22:24<13:13:26,  3.32s/it]
  3%|▎         | 386/14725 [22:28<12:57:10,  3.25s/it]
  3%|▎         | 387/14725 [22:31<12:46:43,  3.21s/it]
  3%|▎         | 388/14725 [22:34<12:56:13,  3.25s/it]
  3%|▎         | 389/14725 [22:37<13:02:16,  3.27s/it]
  3%|▎         | 390/14725 [22:41<12:56:20,  3.25s/it]
                                                      

  3%|▎         | 390/14725 [22:41<12:56:20,  3.25s/it]
  3%|▎         | 391/14725 [22:44<13:14:30,  3.33s/it]
  3%|▎         | 392/14725 [22:47<13:13:02,  3.32s/it]
  3%|▎         | 393/14725 [22:50<12:59:14,  3.26s/it]
  3%|▎         | 394/14725 [22:54<12:46:30,  3.21s/it]
  3%|▎         | 395/14725 [22:57<12:46:22,  3.21s/it]
  3%|▎         | 396/14725 [23:00<12:50:29,  3.23s/it]
  3%|▎         | 397/14725 [23:03<12:44:41,  3.20s/it]
  3%|▎         | 398/14725 [23:06<12:42:10,  3.19s/it]
  3%|▎       
0: {'loss': 0.8101, 'grad_norm': 0.4879903952826372, 'learning_rate': 1e-05, 'epoch': 0.03}
0: {'loss': 0.8318, 'grad_norm': 0.4911324366020474, 'learning_rate': 1e-05, 'epoch': 0.03}
0:   | 399/14725 [23:10<12:49:35,  3.22s/it]
  3%|▎         | 400/14725 [23:13<12:47:53,  3.22s/it]
                                                      

  3%|▎         | 400/14725 [23:13<12:47:53,  3.22s/it]
  3%|▎         | 401/14725 [23:16<12:57:21,  3.26s/it]
  3%|▎         | 402/14725 [23:19<12:51:43,  3.23s/it]
  3%|▎         | 403/14725 [23:23<12:49:46,  3.22s/it]
  3%|▎         | 404/14725 [23:26<12:50:13,  3.23s/it]
  3%|▎         | 405/14725 [23:29<13:01:19,  3.27s/it]
  3%|▎         | 406/14725 [23:33<13:09:14,  3.31s/it]
  3%|▎         | 407/14725 [23:36<12:58:43,  3.26s/it]
  3%|▎         | 408/14725 [23:39<12:51:47,  3.23s/it]
  3%|▎         | 409/14725 [23:42<12:52:24,  3.24s/it]
  3%|▎         | 410/14725 [23:45<12:54:20,  3.25s/it]
                                                      

  3%|▎         | 410/14725 [23:45<12:54:20,  3.25s/it]
  3%|▎         | 411/14725 [23:49<12:54:03,  3.24s/it]
  3%|▎         | 412/14725 [23:52<12:54:04,  3.24s/it]
  3%|▎       
0: {'loss': 0.8107, 'grad_norm': 0.44010737470802447, 'learning_rate': 1e-05, 'epoch': 0.03}
0:   | 413/14725 [23:55<12:56:14,  3.25s/it]
  3%|▎         | 414/14725 [23:58<12:58:24,  3.26s/it]
  3%|▎         | 415/14725 [24:02<12:58:27,  3.26s/it]
  3%|▎         | 416/14725 [24:05<12:58:18,  3.26s/it]
  3%|▎         | 417/14725 [24:08<13:00:44,  3.27s/it]
  3%|▎         | 418/14725 [24:12<13:09:33,  3.31s/it]
  3%|▎         | 419/14725 [24:15<13:14:23,  3.33s/it]
  3%|▎         | 420/14725 [24:18<13:06:04,  3.30s/it]
                                                      

  3%|▎         | 420/14725 [24:18<13:06:04,  3.30s/it]
  3%|▎         | 421/14725 [24:22<13:03:46,  3.29s/it]
  3%|▎         | 422/14725 [24:25<13:03:36,  3.29s/it]
  3%|▎         | 423/14725 [24:28<13:00:58,  3.28s/it]
  3%|▎         | 424/14725 [24:31<12:56:09,  3.26s/it]
  3%|▎         | 425/14725 [24:34<12:48:24,  3.22s/it]
  3%|▎         | 426/14725 [24:38<12:57:45,  3.26s/it]
  3%|▎         | 427/14725 [24:41<12:48:40,  3.23s/it]
  3%|▎         | 428/14725 [24:44<12:48:44,  3.23s/it]
  3%|▎      
0: {'loss': 0.8147, 'grad_norm': 0.5470489431058717, 'learning_rate': 1e-05, 'epoch': 0.03}
0: {'loss': 0.8238, 'grad_norm': 0.49776606467474227, 'learning_rate': 1e-05, 'epoch': 0.03}
0:    | 429/14725 [24:47<12:50:55,  3.24s/it]
  3%|▎         | 430/14725 [24:51<12:44:23,  3.21s/it]
                                                      

  3%|▎         | 430/14725 [24:51<12:44:23,  3.21s/it]
  3%|▎         | 431/14725 [24:54<12:47:46,  3.22s/it]
  3%|▎         | 432/14725 [24:57<12:41:39,  3.20s/it]
  3%|▎         | 433/14725 [25:00<12:47:07,  3.22s/it]
  3%|▎         | 434/14725 [25:04<13:11:10,  3.32s/it]
  3%|▎         | 435/14725 [25:07<12:58:43,  3.27s/it]
  3%|▎         | 436/14725 [25:10<12:59:13,  3.27s/it]
  3%|▎         | 437/14725 [25:14<13:13:39,  3.33s/it]
  3%|▎         | 438/14725 [25:17<13:23:50,  3.38s/it]
  3%|▎         | 439/14725 [25:20<13:05:24,  3.30s/it]
  3%|▎         | 440/14725 [25:23<12:57:48,  3.27s/it]
                                                      

  3%|▎         | 440/14725 [25:24<12:57:48,  3.27s/it]
  3%|▎         | 441/14725 [25:27<13:00:33,  3.28s/it]
  3%|▎         | 442/14725 [25:30<12:57:41,  3.27s/it]
  3%|▎      
0: {'loss': 0.8139, 'grad_norm': 0.4900353213478841, 'learning_rate': 1e-05, 'epoch': 0.03}
0:    | 443/14725 [25:33<12:45:28,  3.22s/it]
  3%|▎         | 444/14725 [25:36<12:42:56,  3.21s/it]
  3%|▎         | 445/14725 [25:40<12:53:47,  3.25s/it]
  3%|▎         | 446/14725 [25:43<12:52:34,  3.25s/it]
  3%|▎         | 447/14725 [25:46<12:42:46,  3.21s/it]
  3%|▎         | 448/14725 [25:49<12:37:13,  3.18s/it]
  3%|▎         | 449/14725 [25:53<12:50:53,  3.24s/it]
  3%|▎         | 450/14725 [25:56<12:47:32,  3.23s/it]
                                                      

  3%|▎         | 450/14725 [25:56<12:47:32,  3.23s/it]
  3%|▎         | 451/14725 [25:59<12:55:08,  3.26s/it]
  3%|▎         | 452/14725 [26:02<13:04:37,  3.30s/it]
  3%|▎         | 453/14725 [26:06<12:47:24,  3.23s/it]
  3%|▎         | 454/14725 [26:09<13:00:40,  3.28s/it]
  3%|▎         | 455/14725 [26:12<13:03:31,  3.29s/it]
  3%|▎         | 456/14725 [26:16<13:03:25,  3.29s/it]
  3%|▎         | 457/14725 [26:19<12:57:33,  3.27s/it]
  3%|▎         | 458/14725 [26:22<12:47:51,  3.23s/it]
  3%|▎     
0: {'loss': 0.8126, 'grad_norm': 0.5382609119674674, 'learning_rate': 1e-05, 'epoch': 0.03}
0: {'loss': 0.8037, 'grad_norm': 0.5159265048845683, 'learning_rate': 1e-05, 'epoch': 0.03}
0:     | 459/14725 [26:25<12:44:56,  3.22s/it]
  3%|▎         | 460/14725 [26:28<12:56:58,  3.27s/it]
                                                      

  3%|▎         | 460/14725 [26:28<12:56:58,  3.27s/it]
  3%|▎         | 461/14725 [26:32<13:05:53,  3.31s/it]
  3%|▎         | 462/14725 [26:35<13:17:46,  3.36s/it]
  3%|▎         | 463/14725 [26:39<13:06:19,  3.31s/it]
  3%|▎         | 464/14725 [26:42<12:55:24,  3.26s/it]
  3%|▎         | 465/14725 [26:45<12:57:00,  3.27s/it]
  3%|▎         | 466/14725 [26:48<12:47:30,  3.23s/it]
  3%|▎         | 467/14725 [26:51<12:58:02,  3.27s/it]
  3%|▎         | 468/14725 [26:55<12:49:54,  3.24s/it]
  3%|▎         | 469/14725 [26:58<12:48:07,  3.23s/it]
  3%|▎         | 470/14725 [27:01<12:47:08,  3.23s/it]
                                                      

  3%|▎         | 470/14725 [27:01<12:47:08,  3.23s/it]
  3%|▎         | 471/14725 [27:04<12:45:21,  3.22s/it]
  3%|▎         | 472/14725 [27:07<12:41:00,  3.20s/it]
  3%|▎     
0: {'loss': 0.7928, 'grad_norm': 0.4776052170682618, 'learning_rate': 1e-05, 'epoch': 0.03}
0:     | 473/14725 [27:11<12:48:26,  3.24s/it]
  3%|▎         | 474/14725 [27:14<12:44:17,  3.22s/it]
  3%|▎         | 475/14725 [27:17<12:45:45,  3.22s/it]
  3%|▎         | 476/14725 [27:20<12:34:26,  3.18s/it]
  3%|▎         | 477/14725 [27:23<12:39:46,  3.20s/it]
  3%|▎         | 478/14725 [27:27<12:34:13,  3.18s/it]
  3%|▎         | 479/14725 [27:30<12:28:14,  3.15s/it]
  3%|▎         | 480/14725 [27:33<12:43:23,  3.22s/it]
                                                      

  3%|▎         | 480/14725 [27:33<12:43:23,  3.22s/it]
  3%|▎         | 481/14725 [27:36<12:37:28,  3.19s/it]
  3%|▎         | 482/14725 [27:40<13:03:57,  3.30s/it]
  3%|▎         | 483/14725 [27:43<13:02:20,  3.30s/it]
  3%|▎         | 484/14725 [27:46<12:57:35,  3.28s/it]
  3%|▎         | 485/14725 [27:50<12:56:17,  3.27s/it]
  3%|▎         | 486/14725 [27:53<12:46:54,  3.23s/it]
  3%|▎         | 487/14725 [27:56<12:34:58,  3.18s/it]
  3%|▎         | 488/14725 [27:59<12:44:47,  3.22s/it]
  3%|▎    
0: {'loss': 0.8022, 'grad_norm': 0.47420306481025876, 'learning_rate': 1e-05, 'epoch': 0.03}
0: {'loss': 0.8027, 'grad_norm': 0.5282819261051567, 'learning_rate': 1e-05, 'epoch': 0.03}
0:      | 489/14725 [28:02<12:56:42,  3.27s/it]
  3%|▎         | 490/14725 [28:05<12:40:20,  3.20s/it]
                                                      

  3%|▎         | 490/14725 [28:05<12:40:20,  3.20s/it]
  3%|▎         | 491/14725 [28:09<12:47:18,  3.23s/it]
  3%|▎         | 492/14725 [28:12<12:43:20,  3.22s/it]
  3%|▎         | 493/14725 [28:15<13:01:22,  3.29s/it]
  3%|▎         | 494/14725 [28:19<12:54:54,  3.27s/it]
  3%|▎         | 495/14725 [28:22<13:03:13,  3.30s/it]
  3%|▎         | 496/14725 [28:25<12:58:53,  3.28s/it]
  3%|▎         | 497/14725 [28:29<13:04:07,  3.31s/it]
  3%|▎         | 498/14725 [28:32<12:47:15,  3.24s/it]
  3%|▎         | 499/14725 [28:35<12:53:12,  3.26s/it]
  3%|▎         | 500/14725 [28:38<13:06:23,  3.32s/it]
                                                      

  3%|▎         | 500/14725 [28:38<13:06:23,  3.32s/it]
  3%|▎         | 501/14725 [28:42<13:21:41,  3.38s/it]
  3%|▎         | 502/14725 [28:45<13:13:18,  3.35s/it]
  3%|▎    
0: {'loss': 0.7889, 'grad_norm': 0.4936643753279381, 'learning_rate': 1e-05, 'epoch': 0.03}
0:      | 503/14725 [28:48<13:02:52,  3.30s/it]
  3%|▎         | 504/14725 [28:52<12:49:04,  3.24s/it]
  3%|▎         | 505/14725 [28:55<12:42:22,  3.22s/it]
  3%|▎         | 506/14725 [28:58<12:33:57,  3.18s/it]
  3%|▎         | 507/14725 [29:01<12:36:20,  3.19s/it]
  3%|▎         | 508/14725 [29:04<12:39:07,  3.20s/it]
  3%|▎         | 509/14725 [29:08<12:42:50,  3.22s/it]
  3%|▎         | 510/14725 [29:11<12:46:57,  3.24s/it]
                                                      

  3%|▎         | 510/14725 [29:11<12:46:57,  3.24s/it]
  3%|▎         | 511/14725 [29:14<13:00:38,  3.30s/it]
  3%|▎         | 512/14725 [29:17<12:54:42,  3.27s/it]
  3%|▎         | 513/14725 [29:21<12:51:18,  3.26s/it]
  3%|▎         | 514/14725 [29:24<12:52:41,  3.26s/it]
  3%|▎         | 515/14725 [29:27<12:48:30,  3.24s/it]
  4%|▎         | 516/14725 [29:30<12:39:06,  3.21s/it]
  4%|▎         | 517/14725 [29:33<12:37:51,  3.20s/it]
  4%|▎         | 518/14725 [29:37<12:33:14,  3.18s/it]
  4%|▎   
0: {'loss': 0.8181, 'grad_norm': 0.4930844731293853, 'learning_rate': 1e-05, 'epoch': 0.04}
0: {'loss': 0.8035, 'grad_norm': 0.5033854664922687, 'learning_rate': 1e-05, 'epoch': 0.04}
0:       | 519/14725 [29:40<12:49:47,  3.25s/it]
  4%|▎         | 520/14725 [29:43<12:58:56,  3.29s/it]
                                                      

  4%|▎         | 520/14725 [29:43<12:58:56,  3.29s/it]
  4%|▎         | 521/14725 [29:47<12:56:18,  3.28s/it]
  4%|▎         | 522/14725 [29:50<12:55:50,  3.28s/it]
  4%|▎         | 523/14725 [29:53<12:56:21,  3.28s/it]
  4%|▎         | 524/14725 [29:57<13:03:01,  3.31s/it]
  4%|▎         | 525/14725 [30:00<12:50:46,  3.26s/it]
  4%|▎         | 526/14725 [30:03<12:49:12,  3.25s/it]
  4%|▎         | 527/14725 [30:06<12:53:03,  3.27s/it]
  4%|▎         | 528/14725 [30:10<12:54:17,  3.27s/it]
  4%|▎         | 529/14725 [30:13<13:00:48,  3.30s/it]
  4%|▎         | 530/14725 [30:16<13:01:49,  3.30s/it]
                                                      

  4%|▎         | 530/14725 [30:16<13:01:49,  3.30s/it]
  4%|▎         | 531/14725 [30:19<12:49:18,  3.25s/it]
  4%|▎         | 532/14725 [30:23<12:54:45,  3.28s/it]
  4%|▎   
0: {'loss': 0.8059, 'grad_norm': 0.5239553483903469, 'learning_rate': 1e-05, 'epoch': 0.04}
0:       | 533/14725 [30:26<13:05:52,  3.32s/it]
  4%|▎         | 534/14725 [30:29<13:06:59,  3.33s/it]
  4%|▎         | 535/14725 [30:33<12:51:33,  3.26s/it]
  4%|▎         | 536/14725 [30:36<13:18:41,  3.38s/it]
  4%|▎         | 537/14725 [30:40<13:18:53,  3.38s/it]
  4%|▎         | 538/14725 [30:43<13:11:53,  3.35s/it]
  4%|▎         | 539/14725 [30:46<13:00:21,  3.30s/it]
  4%|▎         | 540/14725 [30:49<13:09:30,  3.34s/it]
                                                      

  4%|▎         | 540/14725 [30:49<13:09:30,  3.34s/it]
  4%|▎         | 541/14725 [30:53<13:05:56,  3.32s/it]
  4%|▎         | 542/14725 [30:56<13:04:52,  3.32s/it]
  4%|▎         | 543/14725 [30:59<13:10:49,  3.35s/it]
  4%|▎         | 544/14725 [31:03<13:06:33,  3.33s/it]
  4%|▎         | 545/14725 [31:06<13:06:28,  3.33s/it]
  4%|▎         | 546/14725 [31:09<12:52:45,  3.27s/it]
  4%|▎         | 547/14725 [31:12<12:50:08,  3.26s/it]
  4%|▎         | 548/14725 [31:16<12:47:36,  3.25s/it]
  4%|▎  
0: {'loss': 0.8115, 'grad_norm': 0.46375890336238884, 'learning_rate': 1e-05, 'epoch': 0.04}
0: {'loss': 0.7975, 'grad_norm': 0.4822918956479016, 'learning_rate': 1e-05, 'epoch': 0.04}
0:        | 549/14725 [31:19<12:53:02,  3.27s/it]
  4%|▎         | 550/14725 [31:22<13:00:20,  3.30s/it]
                                                      

  4%|▎         | 550/14725 [31:22<13:00:20,  3.30s/it]
  4%|▎         | 551/14725 [31:26<12:53:39,  3.27s/it]
  4%|▎         | 552/14725 [31:29<12:55:26,  3.28s/it]
  4%|▍         | 553/14725 [31:32<12:51:27,  3.27s/it]
  4%|▍         | 554/14725 [31:35<12:51:30,  3.27s/it]
  4%|▍         | 555/14725 [31:39<12:48:36,  3.25s/it]
  4%|▍         | 556/14725 [31:42<12:43:15,  3.23s/it]
  4%|▍         | 557/14725 [31:45<12:54:35,  3.28s/it]
  4%|▍         | 558/14725 [31:48<12:50:55,  3.26s/it]
  4%|▍         | 559/14725 [31:52<12:39:51,  3.22s/it]
  4%|▍         | 560/14725 [31:55<13:00:55,  3.31s/it]
                                                      

  4%|▍         | 560/14725 [31:55<13:00:55,  3.31s/it]
  4%|▍         | 561/14725 [31:58<12:50:38,  3.26s/it]
  4%|▍         | 562/14725 [32:01<12:46:18,  3.25s/it]
  4%|▍  
0: {'loss': 0.8017, 'grad_norm': 0.48800925096000464, 'learning_rate': 1e-05, 'epoch': 0.04}
0:        | 563/14725 [32:05<12:37:43,  3.21s/it]
  4%|▍         | 564/14725 [32:08<12:48:02,  3.25s/it]
  4%|▍         | 565/14725 [32:11<12:39:38,  3.22s/it]
  4%|▍         | 566/14725 [32:14<12:45:07,  3.24s/it]
  4%|▍         | 567/14725 [32:18<12:47:18,  3.25s/it]
  4%|▍         | 568/14725 [32:21<13:04:27,  3.32s/it]
  4%|▍         | 569/14725 [32:25<13:13:48,  3.36s/it]
  4%|▍         | 570/14725 [32:28<12:54:42,  3.28s/it]
                                                      

  4%|▍         | 570/14725 [32:28<12:54:42,  3.28s/it]
  4%|▍         | 571/14725 [32:31<12:48:10,  3.26s/it]
  4%|▍         | 572/14725 [32:34<12:53:46,  3.28s/it]
  4%|▍         | 573/14725 [32:37<12:50:24,  3.27s/it]
  4%|▍         | 574/14725 [32:41<12:57:36,  3.30s/it]
  4%|▍         | 575/14725 [32:44<13:07:07,  3.34s/it]
  4%|▍         | 576/14725 [32:48<13:03:05,  3.32s/it]
  4%|▍         | 577/14725 [32:51<12:48:33,  3.26s/it]
  4%|▍         | 578/14725 [32:54<13:02:23,  3.32s/it]
  4%|▍ 
0: {'loss': 0.7935, 'grad_norm': 0.4995414278718412, 'learning_rate': 1e-05, 'epoch': 0.04}
0: {'loss': 0.8091, 'grad_norm': 0.5414506455807702, 'learning_rate': 1e-05, 'epoch': 0.04}
0:         | 579/14725 [32:57<12:55:20,  3.29s/it]
  4%|▍         | 580/14725 [33:01<13:09:53,  3.35s/it]
                                                      

  4%|▍         | 580/14725 [33:01<13:09:53,  3.35s/it]
  4%|▍         | 581/14725 [33:04<13:16:23,  3.38s/it]
  4%|▍         | 582/14725 [33:08<13:19:58,  3.39s/it]
  4%|▍         | 583/14725 [33:11<13:22:30,  3.40s/it]
  4%|▍         | 584/14725 [33:14<13:17:14,  3.38s/it]
  4%|▍         | 585/14725 [33:18<13:15:53,  3.38s/it]
  4%|▍         | 586/14725 [33:21<13:21:55,  3.40s/it]
  4%|▍         | 587/14725 [33:24<13:09:10,  3.35s/it]
  4%|▍         | 588/14725 [33:28<13:05:22,  3.33s/it]
  4%|▍         | 589/14725 [33:31<13:12:20,  3.36s/it]
  4%|▍         | 590/14725 [33:35<13:09:01,  3.35s/it]
                                                      

  4%|▍         | 590/14725 [33:35<13:09:01,  3.35s/it]
  4%|▍         | 591/14725 [33:38<12:58:59,  3.31s/it]
  4%|▍         | 592/14725 [33:41<12:54:42,  3.29s/it]
  4%|▍ 
0: {'loss': 0.7995, 'grad_norm': 0.58779191539825, 'learning_rate': 1e-05, 'epoch': 0.04}
0:         | 593/14725 [33:44<12:41:42,  3.23s/it]
  4%|▍         | 594/14725 [33:47<12:35:54,  3.21s/it]
  4%|▍         | 595/14725 [33:51<13:11:55,  3.36s/it]
  4%|▍         | 596/14725 [33:55<13:37:10,  3.47s/it]
  4%|▍         | 597/14725 [33:58<13:46:13,  3.51s/it]
  4%|▍         | 598/14725 [34:02<13:26:09,  3.42s/it]
  4%|▍         | 599/14725 [34:05<13:09:07,  3.35s/it]
  4%|▍         | 600/14725 [34:08<13:08:26,  3.35s/it]
                                                      

  4%|▍         | 600/14725 [34:08<13:08:26,  3.35s/it]
  4%|▍         | 601/14725 [34:11<12:51:30,  3.28s/it]
  4%|▍         | 602/14725 [34:15<13:01:22,  3.32s/it]
  4%|▍         | 603/14725 [34:18<12:48:02,  3.26s/it]
  4%|▍         | 604/14725 [34:21<13:00:21,  3.32s/it]
  4%|▍         | 605/14725 [34:24<12:49:02,  3.27s/it]
  4%|▍         | 606/14725 [34:27<12:36:10,  3.21s/it]
  4%|▍         | 607/14725 [34:31<12:37:30,  3.22s/it]
  4%|▍         | 608/14725 [34:34<12:39:19,  3.23s/it]
  4%|▍
0: {'loss': 0.7969, 'grad_norm': 0.5308333694620322, 'learning_rate': 1e-05, 'epoch': 0.04}
0: {'loss': 0.7909, 'grad_norm': 0.49748262587743614, 'learning_rate': 1e-05, 'epoch': 0.04}
0:          | 609/14725 [34:37<12:27:53,  3.18s/it]
  4%|▍         | 610/14725 [34:40<12:32:35,  3.20s/it]
                                                      

  4%|▍         | 610/14725 [34:40<12:32:35,  3.20s/it]
  4%|▍         | 611/14725 [34:44<12:47:12,  3.26s/it]
  4%|▍         | 612/14725 [34:47<12:31:24,  3.19s/it]
  4%|▍         | 613/14725 [34:50<12:44:53,  3.25s/it]
  4%|▍         | 614/14725 [34:53<12:47:02,  3.26s/it]
  4%|▍         | 615/14725 [34:57<12:50:21,  3.28s/it]
  4%|▍         | 616/14725 [35:00<12:42:09,  3.24s/it]
  4%|▍         | 617/14725 [35:03<12:35:02,  3.21s/it]
  4%|▍         | 618/14725 [35:06<12:25:11,  3.17s/it]
  4%|▍         | 619/14725 [35:09<12:34:07,  3.21s/it]
  4%|▍         | 620/14725 [35:13<12:44:25,  3.25s/it]
                                                      

  4%|▍         | 620/14725 [35:13<12:44:25,  3.25s/it]
  4%|▍         | 621/14725 [35:16<12:41:50,  3.24s/it]
  4%|▍         | 622/14725 [35:19<12:43:35,  3.25s/it]
  4%|▍
0: {'loss': 0.8072, 'grad_norm': 0.47223594568260424, 'learning_rate': 1e-05, 'epoch': 0.04}
0:          | 623/14725 [35:22<12:46:38,  3.26s/it]
  4%|▍         | 624/14725 [35:25<12:32:46,  3.20s/it]
  4%|▍         | 625/14725 [35:29<12:51:15,  3.28s/it]
  4%|▍         | 626/14725 [35:32<12:41:03,  3.24s/it]
  4%|▍         | 627/14725 [35:35<12:50:10,  3.28s/it]
  4%|▍         | 628/14725 [35:39<12:45:35,  3.26s/it]
  4%|▍         | 629/14725 [35:42<13:10:08,  3.36s/it]
  4%|▍         | 630/14725 [35:45<12:54:02,  3.29s/it]
                                                      

  4%|▍         | 630/14725 [35:45<12:54:02,  3.29s/it]
  4%|▍         | 631/14725 [35:49<13:02:16,  3.33s/it]
  4%|▍         | 632/14725 [35:52<12:49:46,  3.28s/it]
  4%|▍         | 633/14725 [35:55<12:54:10,  3.30s/it]
  4%|▍         | 634/14725 [35:58<12:43:40,  3.25s/it]
  4%|▍         | 635/14725 [36:02<12:44:57,  3.26s/it]
  4%|▍         | 636/14725 [36:05<12:38:08,  3.23s/it]
  4%|▍         | 637/14725 [36:08<12:47:58,  3.27s/it]
  4%|▍         | 638/14725 [36:12<12:50:07,  3.28s/it]
  4%|�
0: {'loss': 0.8119, 'grad_norm': 0.538262158017164, 'learning_rate': 1e-05, 'epoch': 0.04}
0: {'loss': 0.7994, 'grad_norm': 0.464294691514397, 'learning_rate': 1e-05, 'epoch': 0.04}
0: �         | 639/14725 [36:15<12:44:27,  3.26s/it]
  4%|▍         | 640/14725 [36:18<12:50:02,  3.28s/it]
                                                      

  4%|▍         | 640/14725 [36:18<12:50:02,  3.28s/it]
  4%|▍         | 641/14725 [36:21<12:42:14,  3.25s/it]
  4%|▍         | 642/14725 [36:25<12:58:42,  3.32s/it]
  4%|▍         | 643/14725 [36:28<13:01:05,  3.33s/it]
  4%|▍         | 644/14725 [36:31<12:49:31,  3.28s/it]
  4%|▍         | 645/14725 [36:35<12:55:41,  3.31s/it]
  4%|▍         | 646/14725 [36:38<12:36:31,  3.22s/it]
  4%|▍         | 647/14725 [36:41<12:25:18,  3.18s/it]
  4%|▍         | 648/14725 [36:44<12:37:23,  3.23s/it]
  4%|▍         | 649/14725 [36:47<12:31:58,  3.21s/it]
  4%|▍         | 650/14725 [36:50<12:31:25,  3.20s/it]
                                                      

  4%|▍         | 650/14725 [36:50<12:31:25,  3.20s/it]
  4%|▍         | 651/14725 [36:54<12:35:21,  3.22s/it]
  4%|▍         | 652/14725 [36:57<12:43:45,  3.26s/it]
  4%|�
0: {'loss': 0.7957, 'grad_norm': 0.49730569191355867, 'learning_rate': 1e-05, 'epoch': 0.04}
0: �         | 653/14725 [37:00<12:26:27,  3.18s/it]
  4%|▍         | 654/14725 [37:03<12:38:06,  3.23s/it]
  4%|▍         | 655/14725 [37:07<12:44:25,  3.26s/it]
  4%|▍         | 656/14725 [37:10<12:50:02,  3.28s/it]
  4%|▍         | 657/14725 [37:13<12:40:38,  3.24s/it]
  4%|▍         | 658/14725 [37:16<12:35:57,  3.22s/it]
  4%|▍         | 659/14725 [37:20<12:34:50,  3.22s/it]
  4%|▍         | 660/14725 [37:23<12:43:11,  3.26s/it]
                                                      

  4%|▍         | 660/14725 [37:23<12:43:11,  3.26s/it]
  4%|▍         | 661/14725 [37:26<12:32:46,  3.21s/it]
  4%|▍         | 662/14725 [37:29<12:27:40,  3.19s/it]
  5%|▍         | 663/14725 [37:32<12:25:22,  3.18s/it]
  5%|▍         | 664/14725 [37:36<12:37:28,  3.23s/it]
  5%|▍         | 665/14725 [37:39<12:28:40,  3.19s/it]
  5%|▍         | 666/14725 [37:42<12:21:44,  3.17s/it]
  5%|▍         | 667/14725 [37:45<12:31:17,  3.21s/it]
  5%|▍         | 668/14725 [37:49<12:50:06,  3.29s/it]
  5%|�
0: {'loss': 0.8013, 'grad_norm': 0.4718223595651619, 'learning_rate': 1e-05, 'epoch': 0.05}
0: {'loss': 0.7843, 'grad_norm': 0.4740452945717686, 'learning_rate': 1e-05, 'epoch': 0.05}
0: ��         | 669/14725 [37:52<12:46:22,  3.27s/it]
  5%|▍         | 670/14725 [37:55<12:51:58,  3.30s/it]
                                                      

  5%|▍         | 670/14725 [37:55<12:51:58,  3.30s/it]
  5%|▍         | 671/14725 [37:59<12:51:22,  3.29s/it]
  5%|▍         | 672/14725 [38:02<12:59:51,  3.33s/it]
  5%|▍         | 673/14725 [38:05<13:02:15,  3.34s/it]
  5%|▍         | 674/14725 [38:09<12:55:51,  3.31s/it]
  5%|▍         | 675/14725 [38:12<12:46:19,  3.27s/it]
  5%|▍         | 676/14725 [38:15<12:39:29,  3.24s/it]
  5%|▍         | 677/14725 [38:18<12:45:06,  3.27s/it]
  5%|▍         | 678/14725 [38:22<13:01:58,  3.34s/it]
  5%|▍         | 679/14725 [38:25<13:03:38,  3.35s/it]
  5%|▍         | 680/14725 [38:29<13:08:13,  3.37s/it]
                                                      

  5%|▍         | 680/14725 [38:29<13:08:13,  3.37s/it]
  5%|▍         | 681/14725 [38:32<12:51:17,  3.30s/it]
  5%|▍         | 682/14725 [38:35<12:56:44,  3.32s/it]
  5%|�
0: {'loss': 0.7922, 'grad_norm': 0.4903103188081651, 'learning_rate': 1e-05, 'epoch': 0.05}
0: ��         | 683/14725 [38:38<12:37:46,  3.24s/it]
  5%|▍         | 684/14725 [38:41<12:44:06,  3.27s/it]
  5%|▍         | 685/14725 [38:45<12:45:35,  3.27s/it]
  5%|▍         | 686/14725 [38:48<12:47:59,  3.28s/it]
  5%|▍         | 687/14725 [38:51<12:36:29,  3.23s/it]
  5%|▍         | 688/14725 [38:55<12:52:08,  3.30s/it]
  5%|▍         | 689/14725 [38:58<12:32:45,  3.22s/it]
  5%|▍         | 690/14725 [39:01<12:26:03,  3.19s/it]
                                                      

  5%|▍         | 690/14725 [39:01<12:26:03,  3.19s/it]
  5%|▍         | 691/14725 [39:04<12:28:14,  3.20s/it]
  5%|▍         | 692/14725 [39:07<12:39:34,  3.25s/it]
  5%|▍         | 693/14725 [39:10<12:28:33,  3.20s/it]
  5%|▍         | 694/14725 [39:14<12:31:20,  3.21s/it]
  5%|▍         | 695/14725 [39:17<12:45:08,  3.27s/it]
  5%|▍         | 696/14725 [39:20<12:43:11,  3.26s/it]
  5%|▍         | 697/14725 [39:24<12:45:28,  3.27s/it]
  5%|▍         | 698/14725 [39:27<12:33:38,  3.22s/it]
  5%|
0: {'loss': 0.7994, 'grad_norm': 0.47117046768043863, 'learning_rate': 1e-05, 'epoch': 0.05}
0: {'loss': 0.7992, 'grad_norm': 0.47628159764848416, 'learning_rate': 1e-05, 'epoch': 0.05}
0: ▍         | 699/14725 [39:30<12:58:09,  3.33s/it]
  5%|▍         | 700/14725 [39:34<12:57:18,  3.33s/it]
                                                      

  5%|▍         | 700/14725 [39:34<12:57:18,  3.33s/it]
  5%|▍         | 701/14725 [39:37<12:38:11,  3.24s/it]
  5%|▍         | 702/14725 [39:40<12:59:08,  3.33s/it]
  5%|▍         | 703/14725 [39:43<12:45:37,  3.28s/it]
  5%|▍         | 704/14725 [39:47<12:39:08,  3.25s/it]
  5%|▍         | 705/14725 [39:50<12:39:32,  3.25s/it]
  5%|▍         | 706/14725 [39:53<12:54:13,  3.31s/it]
  5%|▍         | 707/14725 [39:57<12:56:04,  3.32s/it]
  5%|▍         | 708/14725 [40:00<13:03:21,  3.35s/it]
  5%|▍         | 709/14725 [40:03<12:42:52,  3.27s/it]
  5%|▍         | 710/14725 [40:06<12:31:48,  3.22s/it]
                                                      

  5%|▍         | 710/14725 [40:06<12:31:48,  3.22s/it]
  5%|▍         | 711/14725 [40:09<12:35:36,  3.24s/it]
  5%|▍         | 712/14725 [40:13<12:47:50,  3.29s/it]
  5%|
0: {'loss': 0.791, 'grad_norm': 0.49242480870616334, 'learning_rate': 1e-05, 'epoch': 0.05}
0: ▍         | 713/14725 [40:16<12:40:52,  3.26s/it]
  5%|▍         | 714/14725 [40:19<12:39:00,  3.25s/it]
  5%|▍         | 715/14725 [40:23<12:50:18,  3.30s/it]
  5%|▍         | 716/14725 [40:26<12:59:29,  3.34s/it]
  5%|▍         | 717/14725 [40:29<13:00:44,  3.34s/it]
  5%|▍         | 718/14725 [40:33<12:47:13,  3.29s/it]
  5%|▍         | 719/14725 [40:36<12:47:44,  3.29s/it]
  5%|▍         | 720/14725 [40:39<12:55:47,  3.32s/it]
                                                      

  5%|▍         | 720/14725 [40:39<12:55:47,  3.32s/it]
  5%|▍         | 721/14725 [40:43<12:54:06,  3.32s/it]
  5%|▍         | 722/14725 [40:46<12:49:16,  3.30s/it]
  5%|▍         | 723/14725 [40:49<12:33:22,  3.23s/it]
  5%|▍         | 724/14725 [40:52<12:28:33,  3.21s/it]
  5%|▍         | 725/14725 [40:55<12:37:10,  3.25s/it]
  5%|▍         | 726/14725 [40:59<12:44:33,  3.28s/it]
  5%|▍         | 727/14725 [41:02<12:59:56,  3.34s/it]
  5%|▍         | 728/14725 [41:06<12:54:48,  3.32s/it]
  5%
0: {'loss': 0.7961, 'grad_norm': 0.4857430539382994, 'learning_rate': 1e-05, 'epoch': 0.05}
0: {'loss': 0.7917, 'grad_norm': 0.49143322300460374, 'learning_rate': 1e-05, 'epoch': 0.05}
0: |▍         | 729/14725 [41:09<12:58:25,  3.34s/it]
  5%|▍         | 730/14725 [41:12<12:43:11,  3.27s/it]
                                                      

  5%|▍         | 730/14725 [41:12<12:43:11,  3.27s/it]
  5%|▍         | 731/14725 [41:15<12:40:53,  3.26s/it]
  5%|▍         | 732/14725 [41:19<13:00:52,  3.35s/it]
  5%|▍         | 733/14725 [41:22<12:45:27,  3.28s/it]
  5%|▍         | 734/14725 [41:25<12:25:38,  3.20s/it]
  5%|▍         | 735/14725 [41:28<12:24:54,  3.19s/it]
  5%|▍         | 736/14725 [41:31<12:28:54,  3.21s/it]
  5%|▌         | 737/14725 [41:35<12:42:10,  3.27s/it]
  5%|▌         | 738/14725 [41:38<12:40:26,  3.26s/it]
  5%|▌         | 739/14725 [41:42<12:54:31,  3.32s/it]
  5%|▌         | 740/14725 [41:45<12:49:04,  3.30s/it]
                                                      

  5%|▌         | 740/14725 [41:45<12:49:04,  3.30s/it]
  5%|▌         | 741/14725 [41:48<12:46:15,  3.29s/it]
  5%|▌         | 742/14725 [41:51<12:48:20,  3.30s/it]
  5%
0: {'loss': 0.7989, 'grad_norm': 0.5049802854391123, 'learning_rate': 1e-05, 'epoch': 0.05}
0: |▌         | 743/14725 [41:55<12:43:02,  3.27s/it]
  5%|▌         | 744/14725 [41:58<12:41:43,  3.27s/it]
  5%|▌         | 745/14725 [42:01<12:26:41,  3.20s/it]
  5%|▌         | 746/14725 [42:04<12:39:19,  3.26s/it]
  5%|▌         | 747/14725 [42:08<12:43:17,  3.28s/it]
  5%|▌         | 748/14725 [42:11<12:33:10,  3.23s/it]
  5%|▌         | 749/14725 [42:14<12:37:22,  3.25s/it]
  5%|▌         | 750/14725 [42:17<12:41:48,  3.27s/it]
                                                      

  5%|▌         | 750/14725 [42:17<12:41:48,  3.27s/it]
  5%|▌         | 751/14725 [42:21<12:49:41,  3.30s/it]
  5%|▌         | 752/14725 [42:24<12:35:46,  3.25s/it]
  5%|▌         | 753/14725 [42:27<12:26:19,  3.20s/it]
  5%|▌         | 754/14725 [42:30<12:29:09,  3.22s/it]
  5%|▌         | 755/14725 [42:34<12:37:15,  3.25s/it]
  5%|▌         | 756/14725 [42:37<12:30:46,  3.22s/it]
  5%|▌         | 757/14725 [42:40<12:57:15,  3.34s/it]
  5%|▌         | 758/14725 [42:44<13:11:56,  3.40s/it]
  5
0: {'loss': 0.7996, 'grad_norm': 0.47012562196593294, 'learning_rate': 1e-05, 'epoch': 0.05}
0: {'loss': 0.7923, 'grad_norm': 0.4702731666269137, 'learning_rate': 1e-05, 'epoch': 0.05}
0: %|▌         | 759/14725 [42:47<12:54:33,  3.33s/it]
  5%|▌         | 760/14725 [42:50<12:52:20,  3.32s/it]
                                                      

  5%|▌         | 760/14725 [42:50<12:52:20,  3.32s/it]
  5%|▌         | 761/14725 [42:53<12:45:45,  3.29s/it]
  5%|▌         | 762/14725 [42:57<13:15:08,  3.42s/it]
  5%|▌         | 763/14725 [43:01<13:22:26,  3.45s/it]
  5%|▌         | 764/14725 [43:04<13:05:32,  3.38s/it]
  5%|▌         | 765/14725 [43:07<13:07:45,  3.39s/it]
  5%|▌         | 766/14725 [43:11<12:57:08,  3.34s/it]
  5%|▌         | 767/14725 [43:14<12:46:23,  3.29s/it]
  5%|▌         | 768/14725 [43:17<12:51:45,  3.32s/it]
  5%|▌         | 769/14725 [43:20<12:39:09,  3.26s/it]
  5%|▌         | 770/14725 [43:23<12:34:12,  3.24s/it]
                                                      

  5%|▌         | 770/14725 [43:23<12:34:12,  3.24s/it]
  5%|▌         | 771/14725 [43:27<12:28:40,  3.22s/it]
  5%|▌         | 772/14725 [43:30<12:31:26,  3.23s/it]
  5
0: {'loss': 0.772, 'grad_norm': 0.4769996795869756, 'learning_rate': 1e-05, 'epoch': 0.05}
0: %|▌         | 773/14725 [43:33<12:38:54,  3.26s/it]
  5%|▌         | 774/14725 [43:37<12:41:09,  3.27s/it]
  5%|▌         | 775/14725 [43:40<12:46:04,  3.29s/it]
  5%|▌         | 776/14725 [43:43<12:34:12,  3.24s/it]
  5%|▌         | 777/14725 [43:46<12:30:22,  3.23s/it]
  5%|▌         | 778/14725 [43:50<12:37:51,  3.26s/it]
  5%|▌         | 779/14725 [43:53<13:17:12,  3.43s/it]
  5%|▌         | 780/14725 [43:57<13:56:07,  3.60s/it]
                                                      

  5%|▌         | 780/14725 [43:57<13:56:07,  3.60s/it]
  5%|▌         | 781/14725 [44:01<13:57:06,  3.60s/it]
  5%|▌         | 782/14725 [44:04<13:21:01,  3.45s/it]
  5%|▌         | 783/14725 [44:07<13:13:04,  3.41s/it]
  5%|▌         | 784/14725 [44:11<13:03:44,  3.37s/it]
  5%|▌         | 785/14725 [44:14<12:53:36,  3.33s/it]
  5%|▌         | 786/14725 [44:17<12:50:49,  3.32s/it]
  5%|▌         | 787/14725 [44:21<13:11:36,  3.41s/it]
  5%|▌         | 788/14725 [44:24<13:10:55,  3.40s/it]
  
0: {'loss': 0.7791, 'grad_norm': 0.47677895194743325, 'learning_rate': 1e-05, 'epoch': 0.05}
0: {'loss': 0.7806, 'grad_norm': 0.503861091438495, 'learning_rate': 1e-05, 'epoch': 0.05}
0: 5%|▌         | 789/14725 [44:28<13:06:03,  3.38s/it]
  5%|▌         | 790/14725 [44:31<12:53:16,  3.33s/it]
                                                      

  5%|▌         | 790/14725 [44:31<12:53:16,  3.33s/it]
  5%|▌         | 791/14725 [44:34<12:59:26,  3.36s/it]
  5%|▌         | 792/14725 [44:37<12:40:38,  3.28s/it]
  5%|▌         | 793/14725 [44:40<12:39:14,  3.27s/it]
  5%|▌         | 794/14725 [44:44<12:34:22,  3.25s/it]
  5%|▌         | 795/14725 [44:47<12:29:56,  3.23s/it]
  5%|▌         | 796/14725 [44:50<12:30:41,  3.23s/it]
  5%|▌         | 797/14725 [44:53<12:32:50,  3.24s/it]
  5%|▌         | 798/14725 [44:57<12:38:34,  3.27s/it]
  5%|▌         | 799/14725 [45:00<12:55:12,  3.34s/it]
  5%|▌         | 800/14725 [45:04<13:15:05,  3.43s/it]
                                                      

  5%|▌         | 800/14725 [45:04<13:15:05,  3.43s/it]
  5%|▌         | 801/14725 [45:07<13:10:37,  3.41s/it]
  5%|▌         | 802/14725 [45:10<12:49:29,  3.32s/it]
  
0: {'loss': 0.7796, 'grad_norm': 0.46658513778568783, 'learning_rate': 1e-05, 'epoch': 0.06}
0: 5%|▌         | 803/14725 [45:13<12:40:42,  3.28s/it]
  5%|▌         | 804/14725 [45:17<12:55:19,  3.34s/it]
  5%|▌         | 805/14725 [45:20<12:38:01,  3.27s/it]
  5%|▌         | 806/14725 [45:23<12:32:16,  3.24s/it]
  5%|▌         | 807/14725 [45:26<12:26:50,  3.22s/it]
  5%|▌         | 808/14725 [45:30<12:39:10,  3.27s/it]
  5%|▌         | 809/14725 [45:33<12:30:24,  3.24s/it]
  6%|▌         | 810/14725 [45:36<12:20:15,  3.19s/it]
                                                      

  6%|▌         | 810/14725 [45:36<12:20:15,  3.19s/it]
  6%|▌         | 811/14725 [45:39<12:16:20,  3.18s/it]
  6%|▌         | 812/14725 [45:43<12:25:10,  3.21s/it]
  6%|▌         | 813/14725 [45:46<12:16:45,  3.18s/it]
  6%|▌         | 814/14725 [45:49<12:25:29,  3.22s/it]
  6%|▌         | 815/14725 [45:52<12:20:20,  3.19s/it]
  6%|▌         | 816/14725 [45:55<12:15:56,  3.17s/it]
  6%|▌         | 817/14725 [45:58<12:13:13,  3.16s/it]
  6%|▌         | 818/14725 [46:02<12:22:23,  3.20s/it]
 
0: {'loss': 0.8023, 'grad_norm': 0.49574363547979083, 'learning_rate': 1e-05, 'epoch': 0.06}
0: {'loss': 0.7837, 'grad_norm': 0.4727619501744611, 'learning_rate': 1e-05, 'epoch': 0.06}
0:  6%|▌         | 819/14725 [46:05<12:39:32,  3.28s/it]
  6%|▌         | 820/14725 [46:08<12:39:26,  3.28s/it]
                                                      

  6%|▌         | 820/14725 [46:08<12:39:26,  3.28s/it]
  6%|▌         | 821/14725 [46:12<12:49:10,  3.32s/it]
  6%|▌         | 822/14725 [46:15<12:57:39,  3.36s/it]
  6%|▌         | 823/14725 [46:19<12:59:11,  3.36s/it]
  6%|▌         | 824/14725 [46:22<12:40:02,  3.28s/it]
  6%|▌         | 825/14725 [46:25<12:28:40,  3.23s/it]
  6%|▌         | 826/14725 [46:28<12:34:25,  3.26s/it]
  6%|▌         | 827/14725 [46:31<12:40:46,  3.28s/it]
  6%|▌         | 828/14725 [46:35<12:30:39,  3.24s/it]
  6%|▌         | 829/14725 [46:38<12:25:17,  3.22s/it]
  6%|▌         | 830/14725 [46:41<12:25:14,  3.22s/it]
                                                      

  6%|▌         | 830/14725 [46:41<12:25:14,  3.22s/it]
  6%|▌         | 831/14725 [46:44<12:30:32,  3.24s/it]
  6%|▌         | 832/14725 [46:48<12:32:30,  3.25s/it]
 
0: {'loss': 0.7798, 'grad_norm': 0.4767356301414926, 'learning_rate': 1e-05, 'epoch': 0.06}
0:  6%|▌         | 833/14725 [46:51<12:20:45,  3.20s/it]
  6%|▌         | 834/14725 [46:54<12:21:41,  3.20s/it]
  6%|▌         | 835/14725 [46:57<12:23:32,  3.21s/it]
  6%|▌         | 836/14725 [47:00<12:31:35,  3.25s/it]
  6%|▌         | 837/14725 [47:04<12:31:16,  3.25s/it]
  6%|▌         | 838/14725 [47:07<12:33:14,  3.25s/it]
  6%|▌         | 839/14725 [47:10<12:24:18,  3.22s/it]
  6%|▌         | 840/14725 [47:13<12:21:00,  3.20s/it]
                                                      

  6%|▌         | 840/14725 [47:13<12:21:00,  3.20s/it]
  6%|▌         | 841/14725 [47:17<12:32:35,  3.25s/it]
  6%|▌         | 842/14725 [47:20<12:39:53,  3.28s/it]
  6%|▌         | 843/14725 [47:23<12:45:49,  3.31s/it]
  6%|▌         | 844/14725 [47:26<12:29:31,  3.24s/it]
  6%|▌         | 845/14725 [47:30<12:37:17,  3.27s/it]
  6%|▌         | 846/14725 [47:33<12:35:34,  3.27s/it]
  6%|▌         | 847/14725 [47:36<12:51:59,  3.34s/it]
  6%|▌         | 848/14725 [47:40<12:49:13,  3.33s/it]
0: {'loss': 0.7918, 'grad_norm': 0.47537848304432473, 'learning_rate': 1e-05, 'epoch': 0.06}
0: {'loss': 0.7762, 'grad_norm': 0.48596833345104057, 'learning_rate': 1e-05, 'epoch': 0.06}
0:   6%|▌         | 849/14725 [47:43<12:44:40,  3.31s/it]
  6%|▌         | 850/14725 [47:46<12:52:48,  3.34s/it]
                                                      

  6%|▌         | 850/14725 [47:46<12:52:48,  3.34s/it]
  6%|▌         | 851/14725 [47:50<12:33:39,  3.26s/it]
  6%|▌         | 852/14725 [47:53<12:22:51,  3.21s/it]
  6%|▌         | 853/14725 [47:56<12:16:57,  3.19s/it]
  6%|▌         | 854/14725 [47:59<12:20:17,  3.20s/it]
  6%|▌         | 855/14725 [48:02<12:20:27,  3.20s/it]
  6%|▌         | 856/14725 [48:05<12:11:24,  3.16s/it]
  6%|▌         | 857/14725 [48:09<12:26:18,  3.23s/it]
  6%|▌         | 858/14725 [48:12<12:30:36,  3.25s/it]
  6%|▌         | 859/14725 [48:15<12:30:17,  3.25s/it]
  6%|▌         | 860/14725 [48:18<12:32:30,  3.26s/it]
                                                      

  6%|▌         | 860/14725 [48:18<12:32:30,  3.26s/it]
  6%|▌         | 861/14725 [48:22<12:36:39,  3.27s/it]
  6%|▌         | 862/14725 [48:25<12:44:50,  3.31s/it]
0: {'loss': 0.786, 'grad_norm': 0.4694419457469377, 'learning_rate': 1e-05, 'epoch': 0.06}
0:   6%|▌         | 863/14725 [48:29<12:47:35,  3.32s/it]
  6%|▌         | 864/14725 [48:32<12:42:52,  3.30s/it]
  6%|▌         | 865/14725 [48:35<12:39:29,  3.29s/it]
  6%|▌         | 866/14725 [48:38<12:30:09,  3.25s/it]
  6%|▌         | 867/14725 [48:42<12:42:47,  3.30s/it]
  6%|▌         | 868/14725 [48:45<12:27:07,  3.24s/it]
  6%|▌         | 869/14725 [48:48<12:20:17,  3.21s/it]
  6%|▌         | 870/14725 [48:51<12:21:28,  3.21s/it]
                                                      

  6%|▌         | 870/14725 [48:51<12:21:28,  3.21s/it]
  6%|▌         | 871/14725 [48:55<12:38:42,  3.29s/it]
  6%|▌         | 872/14725 [48:58<12:37:49,  3.28s/it]
  6%|▌         | 873/14725 [49:01<12:36:17,  3.28s/it]
  6%|▌         | 874/14725 [49:04<12:35:43,  3.27s/it]
  6%|▌         | 875/14725 [49:08<12:36:55,  3.28s/it]
  6%|▌         | 876/14725 [49:11<12:31:12,  3.25s/it]
  6%|▌         | 877/14725 [49:14<12:23:18,  3.22s/it]
  6%|▌         | 878/14725 [49:17<12:15:25,  3.19s/it]
0: {'loss': 0.7727, 'grad_norm': 0.44989716507154787, 'learning_rate': 1e-05, 'epoch': 0.06}
0: {'loss': 0.7909, 'grad_norm': 0.48965537234626194, 'learning_rate': 1e-05, 'epoch': 0.06}
0: 
  6%|▌         | 879/14725 [49:20<12:22:55,  3.22s/it]
  6%|▌         | 880/14725 [49:24<12:20:28,  3.21s/it]
                                                      

  6%|▌         | 880/14725 [49:24<12:20:28,  3.21s/it]
  6%|▌         | 881/14725 [49:27<12:36:30,  3.28s/it]
  6%|▌         | 882/14725 [49:30<12:22:41,  3.22s/it]
  6%|▌         | 883/14725 [49:33<12:22:54,  3.22s/it]
  6%|▌         | 884/14725 [49:37<12:38:15,  3.29s/it]
  6%|▌         | 885/14725 [49:40<12:31:45,  3.26s/it]
  6%|▌         | 886/14725 [49:43<12:24:59,  3.23s/it]
  6%|▌         | 887/14725 [49:46<12:32:02,  3.26s/it]
  6%|▌         | 888/14725 [49:50<12:31:19,  3.26s/it]
  6%|▌         | 889/14725 [49:53<12:18:25,  3.20s/it]
  6%|▌         | 890/14725 [49:56<12:18:49,  3.20s/it]
                                                      

  6%|▌         | 890/14725 [49:56<12:18:49,  3.20s/it]
  6%|▌         | 891/14725 [49:59<12:22:36,  3.22s/it]
  6%|▌         | 892/14725 [50:03<12:36:10,  3.28s/it]
0: {'loss': 0.7782, 'grad_norm': 0.4742726391217925, 'learning_rate': 1e-05, 'epoch': 0.06}
0: 
  6%|▌         | 893/14725 [50:06<12:41:19,  3.30s/it]
  6%|▌         | 894/14725 [50:09<12:38:45,  3.29s/it]
  6%|▌         | 895/14725 [50:13<12:35:33,  3.28s/it]
  6%|▌         | 896/14725 [50:16<12:51:03,  3.35s/it]
  6%|▌         | 897/14725 [50:19<12:51:24,  3.35s/it]
  6%|▌         | 898/14725 [50:23<12:43:51,  3.31s/it]
  6%|▌         | 899/14725 [50:26<12:52:29,  3.35s/it]
  6%|▌         | 900/14725 [50:29<12:49:43,  3.34s/it]
                                                      

  6%|▌         | 900/14725 [50:29<12:49:43,  3.34s/it]
  6%|▌         | 901/14725 [50:33<12:50:36,  3.34s/it]
  6%|▌         | 902/14725 [50:36<12:56:22,  3.37s/it]
  6%|▌         | 903/14725 [50:39<12:41:12,  3.30s/it]
  6%|▌         | 904/14725 [50:42<12:23:43,  3.23s/it]
  6%|▌         | 905/14725 [50:46<12:38:27,  3.29s/it]
  6%|▌         | 906/14725 [50:49<12:38:13,  3.29s/it]
  6%|▌         | 907/14725 [50:52<12:24:25,  3.23s/it]
  6%|▌         | 908/14725 [50:55<12:20:01,  3.21s/it
0: {'loss': 0.7682, 'grad_norm': 0.4879361571743549, 'learning_rate': 1e-05, 'epoch': 0.06}
0: {'loss': 0.7664, 'grad_norm': 0.49028211925290377, 'learning_rate': 1e-05, 'epoch': 0.06}
0: ]
  6%|▌         | 909/14725 [50:59<12:23:39,  3.23s/it]
  6%|▌         | 910/14725 [51:02<12:29:44,  3.26s/it]
                                                      

  6%|▌         | 910/14725 [51:02<12:29:44,  3.26s/it]
  6%|▌         | 911/14725 [51:05<12:32:59,  3.27s/it]
  6%|▌         | 912/14725 [51:09<12:42:21,  3.31s/it]
  6%|▌         | 913/14725 [51:12<12:41:49,  3.31s/it]
  6%|▌         | 914/14725 [51:15<12:29:49,  3.26s/it]
  6%|▌         | 915/14725 [51:18<12:17:34,  3.20s/it]
  6%|▌         | 916/14725 [51:21<12:11:04,  3.18s/it]
  6%|▌         | 917/14725 [51:25<12:21:34,  3.22s/it]
  6%|▌         | 918/14725 [51:28<12:27:35,  3.25s/it]
  6%|▌         | 919/14725 [51:31<12:26:38,  3.24s/it]
  6%|▌         | 920/14725 [51:34<12:25:59,  3.24s/it]
                                                      

  6%|▌         | 920/14725 [51:34<12:25:59,  3.24s/it]
  6%|▋         | 921/14725 [51:38<12:27:08,  3.25s/it]
  6%|▋         | 922/14725 [51:41<12:22:03,  3.23s/it
0: {'loss': 0.7757, 'grad_norm': 0.4973555111649227, 'learning_rate': 1e-05, 'epoch': 0.06}
0: ]
  6%|▋         | 923/14725 [51:44<12:23:56,  3.23s/it]
  6%|▋         | 924/14725 [51:47<12:30:33,  3.26s/it]
  6%|▋         | 925/14725 [51:51<12:33:11,  3.27s/it]
  6%|▋         | 926/14725 [51:54<12:33:41,  3.28s/it]
  6%|▋         | 927/14725 [51:57<12:25:33,  3.24s/it]
  6%|▋         | 928/14725 [52:00<12:30:26,  3.26s/it]
  6%|▋         | 929/14725 [52:04<12:46:22,  3.33s/it]
  6%|▋         | 930/14725 [52:07<12:43:52,  3.32s/it]
                                                      

  6%|▋         | 930/14725 [52:07<12:43:52,  3.32s/it]
  6%|▋         | 931/14725 [52:10<12:26:55,  3.25s/it]
  6%|▋         | 932/14725 [52:13<12:13:21,  3.19s/it]
  6%|▋         | 933/14725 [52:17<12:17:33,  3.21s/it]
  6%|▋         | 934/14725 [52:20<12:27:24,  3.25s/it]
  6%|▋         | 935/14725 [52:23<12:21:24,  3.23s/it]
  6%|▋         | 936/14725 [52:27<12:33:09,  3.28s/it]
  6%|▋         | 937/14725 [52:30<12:25:51,  3.25s/it]
  6%|▋         | 938/14725 [52:33<12:34:37,  3.28s/i
0: {'loss': 0.7823, 'grad_norm': 0.4819520610700723, 'learning_rate': 1e-05, 'epoch': 0.06}
0: {'loss': 0.7786, 'grad_norm': 0.5156123124701558, 'learning_rate': 1e-05, 'epoch': 0.06}
0: t]
  6%|▋         | 939/14725 [52:36<12:24:03,  3.24s/it]
  6%|▋         | 940/14725 [52:39<12:20:27,  3.22s/it]
                                                      

  6%|▋         | 940/14725 [52:39<12:20:27,  3.22s/it]
  6%|▋         | 941/14725 [52:43<12:19:23,  3.22s/it]
  6%|▋         | 942/14725 [52:46<12:17:05,  3.21s/it]
  6%|▋         | 943/14725 [52:49<12:29:01,  3.26s/it]
  6%|▋         | 944/14725 [52:52<12:20:03,  3.22s/it]
  6%|▋         | 945/14725 [52:55<12:17:00,  3.21s/it]
  6%|▋         | 946/14725 [52:59<12:20:05,  3.22s/it]
  6%|▋         | 947/14725 [53:02<12:19:37,  3.22s/it]
  6%|▋         | 948/14725 [53:05<12:31:42,  3.27s/it]
  6%|▋         | 949/14725 [53:09<12:33:36,  3.28s/it]
  6%|▋         | 950/14725 [53:12<12:29:08,  3.26s/it]
                                                      

  6%|▋         | 950/14725 [53:12<12:29:08,  3.26s/it]
  6%|▋         | 951/14725 [53:15<12:25:57,  3.25s/it]
  6%|▋         | 952/14725 [53:18<12:18:03,  3.22s/i
0: {'loss': 0.7884, 'grad_norm': 0.4638181844895546, 'learning_rate': 1e-05, 'epoch': 0.07}
0: t]
  6%|▋         | 953/14725 [53:21<12:10:53,  3.18s/it]
  6%|▋         | 954/14725 [53:25<12:09:01,  3.18s/it]
  6%|▋         | 955/14725 [53:28<12:33:30,  3.28s/it]
  6%|▋         | 956/14725 [53:31<12:21:15,  3.23s/it]
  6%|▋         | 957/14725 [53:34<12:25:24,  3.25s/it]
  7%|▋         | 958/14725 [53:38<12:37:04,  3.30s/it]
  7%|▋         | 959/14725 [53:41<12:27:28,  3.26s/it]
  7%|▋         | 960/14725 [53:44<12:28:39,  3.26s/it]
                                                      

  7%|▋         | 960/14725 [53:44<12:28:39,  3.26s/it]
  7%|▋         | 961/14725 [53:48<12:26:44,  3.26s/it]
  7%|▋         | 962/14725 [53:51<12:31:18,  3.28s/it]
  7%|▋         | 963/14725 [53:54<12:30:33,  3.27s/it]
  7%|▋         | 964/14725 [53:57<12:31:00,  3.27s/it]
  7%|▋         | 965/14725 [54:01<12:22:49,  3.24s/it]
  7%|▋         | 966/14725 [54:04<12:34:04,  3.29s/it]
  7%|▋         | 967/14725 [54:07<12:30:32,  3.27s/it]
  7%|▋         | 968/14725 [54:11<12:37:11,  3.30s/
0: {'loss': 0.8, 'grad_norm': 0.49252109427548935, 'learning_rate': 1e-05, 'epoch': 0.07}
0: {'loss': 0.7942, 'grad_norm': 0.5217176372478476, 'learning_rate': 1e-05, 'epoch': 0.07}
0: it]
  7%|▋         | 969/14725 [54:14<12:20:09,  3.23s/it]
  7%|▋         | 970/14725 [54:17<12:27:59,  3.26s/it]
                                                      

  7%|▋         | 970/14725 [54:17<12:27:59,  3.26s/it]
  7%|▋         | 971/14725 [54:20<12:35:18,  3.29s/it]
  7%|▋         | 972/14725 [54:24<12:40:18,  3.32s/it]
  7%|▋         | 973/14725 [54:27<12:42:52,  3.33s/it]
  7%|▋         | 974/14725 [54:30<12:36:48,  3.30s/it]
  7%|▋         | 975/14725 [54:34<12:32:13,  3.28s/it]
  7%|▋         | 976/14725 [54:37<12:31:21,  3.28s/it]
  7%|▋         | 977/14725 [54:40<12:42:09,  3.33s/it]
  7%|▋         | 978/14725 [54:43<12:29:04,  3.27s/it]
  7%|▋         | 979/14725 [54:47<12:40:58,  3.32s/it]
  7%|▋         | 980/14725 [54:50<12:49:56,  3.36s/it]
                                                      

  7%|▋         | 980/14725 [54:50<12:49:56,  3.36s/it]
  7%|▋         | 981/14725 [54:54<12:41:43,  3.33s/it]
  7%|▋         | 982/14725 [54:57<12:41:55,  3.33s/
0: {'loss': 0.7938, 'grad_norm': 0.48396584526150044, 'learning_rate': 1e-05, 'epoch': 0.07}
0: it]
  7%|▋         | 983/14725 [55:00<12:35:34,  3.30s/it]
  7%|▋         | 984/14725 [55:03<12:36:25,  3.30s/it]
  7%|▋         | 985/14725 [55:07<12:29:25,  3.27s/it]
  7%|▋         | 986/14725 [55:10<12:27:51,  3.27s/it]
  7%|▋         | 987/14725 [55:13<12:28:41,  3.27s/it]
  7%|▋         | 988/14725 [55:17<12:36:56,  3.31s/it]
  7%|▋         | 989/14725 [55:20<12:41:40,  3.33s/it]
  7%|▋         | 990/14725 [55:23<12:37:34,  3.31s/it]
                                                      

  7%|▋         | 990/14725 [55:23<12:37:34,  3.31s/it]
  7%|▋         | 991/14725 [55:27<12:40:34,  3.32s/it]
  7%|▋         | 992/14725 [55:30<12:33:32,  3.29s/it]
  7%|▋         | 993/14725 [55:33<12:34:45,  3.30s/it]
  7%|▋         | 994/14725 [55:36<12:28:55,  3.27s/it]
  7%|▋         | 995/14725 [55:40<12:45:30,  3.35s/it]
  7%|▋         | 996/14725 [55:43<12:45:25,  3.35s/it]
  7%|▋         | 997/14725 [55:46<12:36:02,  3.30s/it]
  7%|▋         | 998/14725 [55:50<12:31:49,  3.29s
0: {'loss': 0.7692, 'grad_norm': 0.4851435001642193, 'learning_rate': 1e-05, 'epoch': 0.07}
0: {'loss': 0.7727, 'grad_norm': 0.47084877157630944, 'learning_rate': 1e-05, 'epoch': 0.07}
0: /it]
  7%|▋         | 999/14725 [55:53<12:29:07,  3.27s/it]
  7%|▋         | 1000/14725 [55:56<12:23:27,  3.25s/it]
                                                       

  7%|▋         | 1000/14725 [55:56<12:23:27,  3.25s/it]
  7%|▋         | 1001/14725 [55:59<12:34:47,  3.30s/it]
  7%|▋         | 1002/14725 [56:03<12:28:56,  3.27s/it]
  7%|▋         | 1003/14725 [56:06<12:12:13,  3.20s/it]
  7%|▋         | 1004/14725 [56:09<12:14:54,  3.21s/it]
  7%|▋         | 1005/14725 [56:12<12:22:33,  3.25s/it]
  7%|▋         | 1006/14725 [56:15<12:22:57,  3.25s/it]
  7%|▋         | 1007/14725 [56:19<12:30:12,  3.28s/it]
  7%|▋         | 1008/14725 [56:22<12:49:08,  3.36s/it]
  7%|▋         | 1009/14725 [56:26<12:41:08,  3.33s/it]
  7%|▋         | 1010/14725 [56:29<12:35:33,  3.31s/it]
                                                       

  7%|▋         | 1010/14725 [56:29<12:35:33,  3.31s/it]
  7%|▋         | 1011/14725 [56:32<12:20:08,  3.24s/it]
  7%|▋         | 1012/14725 [56:35
0: {'loss': 0.7846, 'grad_norm': 0.5078235971164202, 'learning_rate': 1e-05, 'epoch': 0.07}
0: <12:12:05,  3.20s/it]
  7%|▋         | 1013/14725 [56:38<12:07:24,  3.18s/it]
  7%|▋         | 1014/14725 [56:42<12:32:56,  3.29s/it]
  7%|▋         | 1015/14725 [56:45<12:36:08,  3.31s/it]
  7%|▋         | 1016/14725 [56:48<12:38:08,  3.32s/it]
  7%|▋         | 1017/14725 [56:52<12:47:51,  3.36s/it]
  7%|▋         | 1018/14725 [56:55<12:28:47,  3.28s/it]
  7%|▋         | 1019/14725 [56:58<12:41:10,  3.33s/it]
  7%|▋         | 1020/14725 [57:02<12:50:34,  3.37s/it]
                                                       

  7%|▋         | 1020/14725 [57:02<12:50:34,  3.37s/it]
  7%|▋         | 1021/14725 [57:05<12:43:01,  3.34s/it]
  7%|▋         | 1022/14725 [57:09<12:46:13,  3.35s/it]
  7%|▋         | 1023/14725 [57:12<12:41:55,  3.34s/it]
  7%|▋         | 1024/14725 [57:15<12:43:07,  3.34s/it]
  7%|▋         | 1025/14725 [57:18<12:25:42,  3.27s/it]
  7%|▋         | 1026/14725 [57:22<12:25:15,  3.26s/it]
  7%|▋         | 1027/14725 [57:25<12:20:09,  3.24s/it]
  7%|▋         
0: {'loss': 0.7843, 'grad_norm': 0.47996057563033834, 'learning_rate': 1e-05, 'epoch': 0.07}
0: {'loss': 0.7762, 'grad_norm': 0.4582501403520857, 'learning_rate': 1e-05, 'epoch': 0.07}
0: | 1028/14725 [57:28<12:26:12,  3.27s/it]
  7%|▋         | 1029/14725 [57:32<12:37:38,  3.32s/it]
  7%|▋         | 1030/14725 [57:35<12:31:13,  3.29s/it]
                                                       

  7%|▋         | 1030/14725 [57:35<12:31:13,  3.29s/it]
  7%|▋         | 1031/14725 [57:38<12:23:34,  3.26s/it]
  7%|▋         | 1032/14725 [57:41<12:26:14,  3.27s/it]
  7%|▋         | 1033/14725 [57:44<12:20:57,  3.25s/it]
  7%|▋         | 1034/14725 [57:48<12:39:23,  3.33s/it]
  7%|▋         | 1035/14725 [57:51<12:23:47,  3.26s/it]
  7%|▋         | 1036/14725 [57:54<12:32:24,  3.30s/it]
  7%|▋         | 1037/14725 [57:58<12:29:10,  3.28s/it]
  7%|▋         | 1038/14725 [58:01<12:44:57,  3.35s/it]
  7%|▋         | 1039/14725 [58:04<12:37:07,  3.32s/it]
  7%|▋         | 1040/14725 [58:08<12:38:38,  3.33s/it]
                                                       

  7%|▋         | 1040/14725 [58:08<12:38:38,  3.33s/it]
  7%|▋         | 1041/14725 [58:11<12:37:53,  3.32s/it]
0: {'loss': 0.7795, 'grad_norm': 0.46316348385255446, 'learning_rate': 1e-05, 'epoch': 0.07}
0: 
  7%|▋         | 1042/14725 [58:14<12:29:22,  3.29s/it]
  7%|▋         | 1043/14725 [58:18<12:36:24,  3.32s/it]
  7%|▋         | 1044/14725 [58:21<12:37:02,  3.32s/it]
  7%|▋         | 1045/14725 [58:24<12:28:44,  3.28s/it]
  7%|▋         | 1046/14725 [58:28<12:46:03,  3.36s/it]
  7%|▋         | 1047/14725 [58:31<12:33:40,  3.31s/it]
  7%|▋         | 1048/14725 [58:34<12:27:46,  3.28s/it]
  7%|▋         | 1049/14725 [58:37<12:24:31,  3.27s/it]
  7%|▋         | 1050/14725 [58:41<12:18:07,  3.24s/it]
                                                       

  7%|▋         | 1050/14725 [58:41<12:18:07,  3.24s/it]
  7%|▋         | 1051/14725 [58:44<12:27:36,  3.28s/it]
  7%|▋         | 1052/14725 [58:47<12:23:05,  3.26s/it]
  7%|▋         | 1053/14725 [58:50<12:23:34,  3.26s/it]
  7%|▋         | 1054/14725 [58:54<12:32:21,  3.30s/it]
  7%|▋         | 1055/14725 [58:57<12:33:38,  3.31s/it]
  7%|▋         | 1056/14725 [59:00<12:30:46,  3.30s/it]
  7%|▋         | 1057/14725 [59:04<1
0: {'loss': 0.7606, 'grad_norm': 0.49640389747752967, 'learning_rate': 1e-05, 'epoch': 0.07}
0: {'loss': 0.7848, 'grad_norm': 0.5215107881878532, 'learning_rate': 1e-05, 'epoch': 0.07}
0: 2:27:48,  3.28s/it]
  7%|▋         | 1058/14725 [59:07<12:14:55,  3.23s/it]
  7%|▋         | 1059/14725 [59:10<12:20:14,  3.25s/it]
  7%|▋         | 1060/14725 [59:13<12:25:03,  3.27s/it]
                                                       

  7%|▋         | 1060/14725 [59:13<12:25:03,  3.27s/it]
  7%|▋         | 1061/14725 [59:17<12:31:11,  3.30s/it]
  7%|▋         | 1062/14725 [59:20<12:41:23,  3.34s/it]
  7%|▋         | 1063/14725 [59:24<12:45:28,  3.36s/it]
  7%|▋         | 1064/14725 [59:27<12:43:23,  3.35s/it]
  7%|▋         | 1065/14725 [59:30<12:43:51,  3.36s/it]
  7%|▋         | 1066/14725 [59:33<12:28:01,  3.29s/it]
  7%|▋         | 1067/14725 [59:37<12:15:52,  3.23s/it]
  7%|▋         | 1068/14725 [59:40<12:28:27,  3.29s/it]
  7%|▋         | 1069/14725 [59:43<12:34:27,  3.31s/it]
  7%|▋         | 1070/14725 [59:47<12:32:16,  3.31s/it]
                                                       

  7%|▋         | 1070/14725 [59:47<12:32:16,  3.31s/it]
  7%|▋         | 1
0: {'loss': 0.7803, 'grad_norm': 0.5809888166465258, 'learning_rate': 1e-05, 'epoch': 0.07}
0: 071/14725 [59:50<12:19:51,  3.25s/it]
  7%|▋         | 1072/14725 [59:53<12:17:43,  3.24s/it]
  7%|▋         | 1073/14725 [59:56<12:22:28,  3.26s/it]
  7%|▋         | 1074/14725 [59:59<12:18:29,  3.25s/it]
  7%|▋         | 1075/14725 [1:00:03<12:15:55,  3.23s/it]
  7%|▋         | 1076/14725 [1:00:06<12:21:31,  3.26s/it]
  7%|▋         | 1077/14725 [1:00:09<12:35:47,  3.32s/it]
  7%|▋         | 1078/14725 [1:00:13<12:32:41,  3.31s/it]
  7%|▋         | 1079/14725 [1:00:16<12:39:01,  3.34s/it]
  7%|▋         | 1080/14725 [1:00:19<12:34:06,  3.32s/it]
                                                         

  7%|▋         | 1080/14725 [1:00:19<12:34:06,  3.32s/it]
  7%|▋         | 1081/14725 [1:00:23<12:24:50,  3.28s/it]
  7%|▋         | 1082/14725 [1:00:26<12:25:51,  3.28s/it]
  7%|▋         | 1083/14725 [1:00:29<12:06:18,  3.19s/it]
  7%|▋         | 1084/14725 [1:00:32<12:06:30,  3.20s/it]
  7%|▋         | 1085/14725 [1:00:36<12:27:41,  3.29s/it]
  7%|▋         | 1086/14725 [1:
0: {'loss': 0.7953, 'grad_norm': 0.4975600100606009, 'learning_rate': 1e-05, 'epoch': 0.07}
0: 00:39<12:32:28,  3.31s/it]
  7%|▋         | 1087/14725 [1:00:42<12:28:52,  3.29s/it]
  7%|▋         | 1088/14725 [1:00:45<12:19:47,  3.25s/it]
  7%|▋         | 1089/14725 [1:00:49<12:34:33,  3.32s/it]
  7%|▋         | 1090/14725 [1:00:52<12:33:51,  3.32s/it]
                                                         

  7%|▋         | 1090/14725 [1:00:52<12:33:51,  3.32s/it]
  7%|▋         | 1091/14725 [1:00:55<12:22:17,  3.27s/it]
  7%|▋         | 1092/14725 [1:00:58<12:08:28,  3.21s/it]
  7%|▋         | 1093/14725 [1:01:01<12:00:40,  3.17s/it]
  7%|▋         | 1094/14725 [1:01:05<12:13:41,  3.23s/it]
  7%|▋         | 1095/14725 [1:01:08<12:35:09,  3.32s/it]
  7%|▋         | 1096/14725 [1:01:12<12:27:17,  3.29s/it]
  7%|▋         | 1097/14725 [1:01:15<12:48:33,  3.38s/it]
  7%|▋         | 1098/14725 [1:01:18<12:29:29,  3.30s/it]
  7%|▋         | 1099/14725 [1:01:21<12:21:30,  3.27s/it]
  7%|▋         | 1100/14725 [1:01:25<12:21:23,  3.26s/it]
                                      
0: {'loss': 0.782, 'grad_norm': 0.45463952602059954, 'learning_rate': 1e-05, 'epoch': 0.07}
0: {'loss': 0.7678, 'grad_norm': 0.5070594209564384, 'learning_rate': 1e-05, 'epoch': 0.08}
0:                    

  7%|▋         | 1100/14725 [1:01:25<12:21:23,  3.26s/it]
  7%|▋         | 1101/14725 [1:01:28<12:25:44,  3.28s/it]
  7%|▋         | 1102/14725 [1:01:31<12:17:54,  3.25s/it]
  7%|▋         | 1103/14725 [1:01:34<12:13:16,  3.23s/it]
  7%|▋         | 1104/14725 [1:01:38<12:18:49,  3.25s/it]
  8%|▊         | 1105/14725 [1:01:41<12:20:59,  3.26s/it]
  8%|▊         | 1106/14725 [1:01:44<12:30:04,  3.30s/it]
  8%|▊         | 1107/14725 [1:01:48<12:16:31,  3.25s/it]
  8%|▊         | 1108/14725 [1:01:51<12:16:24,  3.24s/it]
  8%|▊         | 1109/14725 [1:01:54<12:29:36,  3.30s/it]
  8%|▊         | 1110/14725 [1:01:58<12:34:47,  3.33s/it]
                                                         

  8%|▊         | 1110/14725 [1:01:58<12:34:47,  3.33s/it]
  8%|▊         | 1111/14725 [1:02:01<12:30:32,  3.31s/it]
  8%|▊         | 1112/14725 [1:02:04<12:25:48,  3.29s/it]
  8%|▊         | 1113/14725 [1:02:07<12:22:43,  3.27s/it]
  8%|▊         | 1114/14725 [1:02:11<12:33
0: {'loss': 0.7709, 'grad_norm': 0.45779510720566896, 'learning_rate': 1e-05, 'epoch': 0.08}
0: :16,  3.32s/it]
  8%|▊         | 1115/14725 [1:02:14<12:28:17,  3.30s/it]
  8%|▊         | 1116/14725 [1:02:17<12:25:37,  3.29s/it]
  8%|▊         | 1117/14725 [1:02:21<12:31:34,  3.31s/it]
  8%|▊         | 1118/14725 [1:02:24<12:28:35,  3.30s/it]
  8%|▊         | 1119/14725 [1:02:27<12:18:04,  3.25s/it]
  8%|▊         | 1120/14725 [1:02:30<12:14:46,  3.24s/it]
                                                         

  8%|▊         | 1120/14725 [1:02:30<12:14:46,  3.24s/it]
  8%|▊         | 1121/14725 [1:02:34<12:24:45,  3.28s/it]
  8%|▊         | 1122/14725 [1:02:37<12:23:16,  3.28s/it]
  8%|▊         | 1123/14725 [1:02:40<12:19:40,  3.26s/it]
  8%|▊         | 1124/14725 [1:02:43<12:13:39,  3.24s/it]
  8%|▊         | 1125/14725 [1:02:47<12:22:04,  3.27s/it]
  8%|▊         | 1126/14725 [1:02:50<12:14:46,  3.24s/it]
  8%|▊         | 1127/14725 [1:02:53<12:13:06,  3.23s/it]
  8%|▊         | 1128/14725 [1:02:56<12:04:19,  3.20s/it]
  8%|▊         | 1129/14725 [1:02:59<11:59:10, 
0: {'loss': 0.7705, 'grad_norm': 0.5325016368647716, 'learning_rate': 1e-05, 'epoch': 0.08}
0: {'loss': 0.7767, 'grad_norm': 0.43978243957406815, 'learning_rate': 1e-05, 'epoch': 0.08}
0:  3.17s/it]
  8%|▊         | 1130/14725 [1:03:02<11:56:09,  3.16s/it]
                                                         

  8%|▊         | 1130/14725 [1:03:02<11:56:09,  3.16s/it]
  8%|▊         | 1131/14725 [1:03:06<11:53:02,  3.15s/it]
  8%|▊         | 1132/14725 [1:03:09<11:57:51,  3.17s/it]
  8%|▊         | 1133/14725 [1:03:12<12:02:17,  3.19s/it]
  8%|▊         | 1134/14725 [1:03:15<12:09:12,  3.22s/it]
  8%|▊         | 1135/14725 [1:03:19<12:15:48,  3.25s/it]
  8%|▊         | 1136/14725 [1:03:22<12:14:29,  3.24s/it]
  8%|▊         | 1137/14725 [1:03:25<12:25:05,  3.29s/it]
  8%|▊         | 1138/14725 [1:03:29<12:28:12,  3.30s/it]
  8%|▊         | 1139/14725 [1:03:32<12:43:40,  3.37s/it]
  8%|▊         | 1140/14725 [1:03:35<12:32:52,  3.33s/it]
                                                         

  8%|▊         | 1140/14725 [1:03:35<12:32:52,  3.33s/it]
  8%|▊         | 1141/14725 [1:03:39<12:27:58,  3.30s/it]
  8%|▊         | 1142/14725 [1:03:42<12:25:13,  3.29s
0: {'loss': 0.7706, 'grad_norm': 0.46237260440176564, 'learning_rate': 1e-05, 'epoch': 0.08}
0: /it]
  8%|▊         | 1143/14725 [1:03:45<12:10:57,  3.23s/it]
  8%|▊         | 1144/14725 [1:03:48<12:22:47,  3.28s/it]
  8%|▊         | 1145/14725 [1:03:51<12:14:24,  3.24s/it]
  8%|▊         | 1146/14725 [1:03:55<12:14:29,  3.25s/it]
  8%|▊         | 1147/14725 [1:03:58<12:16:56,  3.26s/it]
  8%|▊         | 1148/14725 [1:04:01<12:20:28,  3.27s/it]
  8%|▊         | 1149/14725 [1:04:05<12:20:02,  3.27s/it]
  8%|▊         | 1150/14725 [1:04:08<12:30:41,  3.32s/it]
                                                         

  8%|▊         | 1150/14725 [1:04:08<12:30:41,  3.32s/it]
  8%|▊         | 1151/14725 [1:04:11<12:28:04,  3.31s/it]
  8%|▊         | 1152/14725 [1:04:15<12:36:23,  3.34s/it]
  8%|▊         | 1153/14725 [1:04:18<12:36:59,  3.35s/it]
  8%|▊         | 1154/14725 [1:04:21<12:31:07,  3.32s/it]
  8%|▊         | 1155/14725 [1:04:24<12:18:31,  3.27s/it]
  8%|▊         | 1156/14725 [1:04:28<12:28:20,  3.31s/it]
  8%|▊         | 1157/14725 [1:04:31<12:25:48,  3.30s/it]
0: {'loss': 0.7714, 'grad_norm': 0.45459295454238025, 'learning_rate': 1e-05, 'epoch': 0.08}
0: {'loss': 0.7723, 'grad_norm': 0.46640003226063975, 'learning_rate': 1e-05, 'epoch': 0.08}
0:   8%|▊         | 1158/14725 [1:04:34<12:25:49,  3.30s/it]
  8%|▊         | 1159/14725 [1:04:38<12:22:57,  3.29s/it]
  8%|▊         | 1160/14725 [1:04:41<12:16:22,  3.26s/it]
                                                         

  8%|▊         | 1160/14725 [1:04:41<12:16:22,  3.26s/it]
  8%|▊         | 1161/14725 [1:04:45<12:42:57,  3.37s/it]
  8%|▊         | 1162/14725 [1:04:48<12:28:05,  3.31s/it]
  8%|▊         | 1163/14725 [1:04:51<12:44:49,  3.38s/it]
  8%|▊         | 1164/14725 [1:04:54<12:31:16,  3.32s/it]
  8%|▊         | 1165/14725 [1:04:58<12:27:54,  3.31s/it]
  8%|▊         | 1166/14725 [1:05:01<12:28:55,  3.31s/it]
  8%|▊         | 1167/14725 [1:05:04<12:15:15,  3.25s/it]
  8%|▊         | 1168/14725 [1:05:08<12:26:27,  3.30s/it]
  8%|▊         | 1169/14725 [1:05:11<12:28:54,  3.31s/it]
  8%|▊         | 1170/14725 [1:05:14<12:21:15,  3.28s/it]
                                                         

  8%|▊         | 1170/14725 [1:05:14<12:21:15,  3.28s/it]
  8%|�
0: {'loss': 0.7737, 'grad_norm': 0.47507475260132814, 'learning_rate': 1e-05, 'epoch': 0.08}
0: ��         | 1171/14725 [1:05:17<12:12:22,  3.24s/it]
  8%|▊         | 1172/14725 [1:05:21<12:10:49,  3.24s/it]
  8%|▊         | 1173/14725 [1:05:24<12:18:30,  3.27s/it]
  8%|▊         | 1174/14725 [1:05:27<12:15:32,  3.26s/it]
  8%|▊         | 1175/14725 [1:05:30<12:17:15,  3.26s/it]
  8%|▊         | 1176/14725 [1:05:34<12:12:16,  3.24s/it]
  8%|▊         | 1177/14725 [1:05:37<12:11:48,  3.24s/it]
  8%|▊         | 1178/14725 [1:05:40<12:16:54,  3.26s/it]
  8%|▊         | 1179/14725 [1:05:43<12:23:12,  3.29s/it]
  8%|▊         | 1180/14725 [1:05:47<12:55:06,  3.43s/it]
                                                         

  8%|▊         | 1180/14725 [1:05:47<12:55:06,  3.43s/it]
  8%|▊         | 1181/14725 [1:05:51<13:00:42,  3.46s/it]
  8%|▊         | 1182/14725 [1:05:54<12:56:09,  3.44s/it]
  8%|▊         | 1183/14725 [1:05:57<12:28:36,  3.32s/it]
  8%|▊         | 1184/14725 [1:06:01<12:34:59,  3.35s/it]
  8%|▊         | 1185/14725 [1:06:04<12:34:17,  3.34s/it]
  8%|▊   
0: {'loss': 0.7715, 'grad_norm': 0.5346425304489647, 'learning_rate': 1e-05, 'epoch': 0.08}
0:       | 1186/14725 [1:06:07<12:21:17,  3.29s/it]
  8%|▊         | 1187/14725 [1:06:10<12:10:24,  3.24s/it]
  8%|▊         | 1188/14725 [1:06:14<12:29:53,  3.32s/it]
  8%|▊         | 1189/14725 [1:06:17<12:27:12,  3.31s/it]
  8%|▊         | 1190/14725 [1:06:21<12:39:20,  3.37s/it]
                                                         

  8%|▊         | 1190/14725 [1:06:21<12:39:20,  3.37s/it]
  8%|▊         | 1191/14725 [1:06:24<12:43:09,  3.38s/it]
  8%|▊         | 1192/14725 [1:06:27<12:33:31,  3.34s/it]
  8%|▊         | 1193/14725 [1:06:30<12:15:51,  3.26s/it]
  8%|▊         | 1194/14725 [1:06:34<12:23:11,  3.30s/it]
  8%|▊         | 1195/14725 [1:06:37<12:36:46,  3.36s/it]
  8%|▊         | 1196/14725 [1:06:40<12:21:47,  3.29s/it]
  8%|▊         | 1197/14725 [1:06:43<12:15:21,  3.26s/it]
  8%|▊         | 1198/14725 [1:06:47<12:14:41,  3.26s/it]
  8%|▊         | 1199/14725 [1:06:50<12:10:36,  3.24s/it]
  8%|▊         | 1200/14725 [1:06:53<12:13:28,  3.25s/it]
                
0: {'loss': 0.7672, 'grad_norm': 0.45821699655358794, 'learning_rate': 1e-05, 'epoch': 0.08}
0: {'loss': 0.771, 'grad_norm': 0.46393949396903283, 'learning_rate': 1e-05, 'epoch': 0.08}
0:                                          

  8%|▊         | 1200/14725 [1:06:53<12:13:28,  3.25s/it]
  8%|▊         | 1201/14725 [1:06:57<12:21:13,  3.29s/it]
  8%|▊         | 1202/14725 [1:07:00<12:16:47,  3.27s/it]
  8%|▊         | 1203/14725 [1:07:03<12:20:13,  3.28s/it]
  8%|▊         | 1204/14725 [1:07:06<12:15:02,  3.26s/it]
  8%|▊         | 1205/14725 [1:07:10<12:12:41,  3.25s/it]
  8%|▊         | 1206/14725 [1:07:13<12:02:50,  3.21s/it]
  8%|▊         | 1207/14725 [1:07:16<12:24:26,  3.30s/it]
  8%|▊         | 1208/14725 [1:07:19<12:11:27,  3.25s/it]
  8%|▊         | 1209/14725 [1:07:23<12:12:13,  3.25s/it]
  8%|▊         | 1210/14725 [1:07:26<12:09:58,  3.24s/it]
                                                         

  8%|▊         | 1210/14725 [1:07:26<12:09:58,  3.24s/it]
  8%|▊         | 1211/14725 [1:07:29<12:18:17,  3.28s/it]
  8%|▊         | 1212/14725 [1:07:32<12:11:18,  3.25s/it]
  8%|▊         | 1213/14725 [1:07:36<12:26:12,  3.31s/it]
  8%|▊         | 121
0: {'loss': 0.759, 'grad_norm': 0.4650835558483467, 'learning_rate': 1e-05, 'epoch': 0.08}
0: 4/14725 [1:07:39<12:27:29,  3.32s/it]
  8%|▊         | 1215/14725 [1:07:42<12:17:59,  3.28s/it]
  8%|▊         | 1216/14725 [1:07:45<12:08:03,  3.23s/it]
  8%|▊         | 1217/14725 [1:07:49<12:00:42,  3.20s/it]
  8%|▊         | 1218/14725 [1:07:52<12:04:36,  3.22s/it]
  8%|▊         | 1219/14725 [1:07:55<12:15:46,  3.27s/it]
  8%|▊         | 1220/14725 [1:07:58<12:03:28,  3.21s/it]
                                                         

  8%|▊         | 1220/14725 [1:07:58<12:03:28,  3.21s/it]
  8%|▊         | 1221/14725 [1:08:01<11:56:44,  3.18s/it]
  8%|▊         | 1222/14725 [1:08:05<11:59:54,  3.20s/it]
  8%|▊         | 1223/14725 [1:08:08<12:19:18,  3.29s/it]
  8%|▊         | 1224/14725 [1:08:11<12:15:17,  3.27s/it]
  8%|▊         | 1225/14725 [1:08:15<12:18:41,  3.28s/it]
  8%|▊         | 1226/14725 [1:08:18<12:21:52,  3.30s/it]
  8%|▊         | 1227/14725 [1:08:21<12:16:24,  3.27s/it]
  8%|▊         | 1228/14725 [1:08:24<12:05:42,  3.23s/it]
  8%|▊         | 1229/147
0: {'loss': 0.7761, 'grad_norm': 0.4625417807365279, 'learning_rate': 1e-05, 'epoch': 0.08}
0: {'loss': 0.7573, 'grad_norm': 0.5245051624548596, 'learning_rate': 1e-05, 'epoch': 0.08}
0: 25 [1:08:28<12:07:45,  3.24s/it]
  8%|▊         | 1230/14725 [1:08:31<12:15:32,  3.27s/it]
                                                         

  8%|▊         | 1230/14725 [1:08:31<12:15:32,  3.27s/it]
  8%|▊         | 1231/14725 [1:08:34<12:20:41,  3.29s/it]
  8%|▊         | 1232/14725 [1:08:37<12:15:20,  3.27s/it]
  8%|▊         | 1233/14725 [1:08:41<12:18:01,  3.28s/it]
  8%|▊         | 1234/14725 [1:08:44<12:05:08,  3.23s/it]
  8%|▊         | 1235/14725 [1:08:47<11:56:42,  3.19s/it]
  8%|▊         | 1236/14725 [1:08:50<12:00:23,  3.20s/it]
  8%|▊         | 1237/14725 [1:08:53<11:57:27,  3.19s/it]
  8%|▊         | 1238/14725 [1:08:57<12:09:24,  3.24s/it]
  8%|▊         | 1239/14725 [1:09:00<12:07:10,  3.24s/it]
  8%|▊         | 1240/14725 [1:09:03<12:15:45,  3.27s/it]
                                                         

  8%|▊         | 1240/14725 [1:09:03<12:15:45,  3.27s/it]
  8%|▊         | 1241/14725 [1:09:06<12:02:30,  3.21s/it]
  8%|▊         | 1242/14725 [1:
0: {'loss': 0.7577, 'grad_norm': 0.5556700492858709, 'learning_rate': 1e-05, 'epoch': 0.08}
0: 09:10<12:09:41,  3.25s/it]
  8%|▊         | 1243/14725 [1:09:13<12:10:12,  3.25s/it]
  8%|▊         | 1244/14725 [1:09:16<12:12:10,  3.26s/it]
  8%|▊         | 1245/14725 [1:09:20<12:24:51,  3.32s/it]
  8%|▊         | 1246/14725 [1:09:23<12:12:02,  3.26s/it]
  8%|▊         | 1247/14725 [1:09:26<12:14:11,  3.27s/it]
  8%|▊         | 1248/14725 [1:09:29<12:16:20,  3.28s/it]
  8%|▊         | 1249/14725 [1:09:33<12:06:18,  3.23s/it]
  8%|▊         | 1250/14725 [1:09:36<12:19:19,  3.29s/it]
                                                         

  8%|▊         | 1250/14725 [1:09:36<12:19:19,  3.29s/it]
  8%|▊         | 1251/14725 [1:09:39<12:23:50,  3.31s/it]
  9%|▊         | 1252/14725 [1:09:43<12:27:59,  3.33s/it]
  9%|▊         | 1253/14725 [1:09:46<12:30:33,  3.34s/it]
  9%|▊         | 1254/14725 [1:09:49<12:21:45,  3.30s/it]
  9%|▊         | 1255/14725 [1:09:53<12:24:57,  3.32s/it]
  9%|▊         | 1256/14725 [1:09:56<12:13:31,  3.27s/it]
  9%|▊         | 1257/14725 [1:09:59
0: {'loss': 0.7783, 'grad_norm': 0.5300031334704446, 'learning_rate': 1e-05, 'epoch': 0.09}
0: {'loss': 0.7654, 'grad_norm': 0.4670360023319004, 'learning_rate': 1e-05, 'epoch': 0.09}
0: <12:11:29,  3.26s/it]
  9%|▊         | 1258/14725 [1:10:02<12:10:16,  3.25s/it]
  9%|▊         | 1259/14725 [1:10:06<12:18:55,  3.29s/it]
  9%|▊         | 1260/14725 [1:10:09<12:19:54,  3.30s/it]
                                                         

  9%|▊         | 1260/14725 [1:10:09<12:19:54,  3.30s/it]
  9%|▊         | 1261/14725 [1:10:12<12:21:40,  3.31s/it]
  9%|▊         | 1262/14725 [1:10:16<12:22:40,  3.31s/it]
  9%|▊         | 1263/14725 [1:10:19<12:13:00,  3.27s/it]
  9%|▊         | 1264/14725 [1:10:22<11:57:50,  3.20s/it]
  9%|▊         | 1265/14725 [1:10:25<12:10:14,  3.26s/it]
  9%|▊         | 1266/14725 [1:10:29<12:13:06,  3.27s/it]
  9%|▊         | 1267/14725 [1:10:32<12:10:04,  3.25s/it]
  9%|▊         | 1268/14725 [1:10:35<11:58:47,  3.20s/it]
  9%|▊         | 1269/14725 [1:10:38<12:04:52,  3.23s/it]
  9%|▊         | 1270/14725 [1:10:41<12:02:01,  3.22s/it]
                                                         

  9%|▊         | 1270/14725 [1:10:41<12:02
0: {'loss': 0.7707, 'grad_norm': 0.46076500673620907, 'learning_rate': 1e-05, 'epoch': 0.09}
0: :01,  3.22s/it]
  9%|▊         | 1271/14725 [1:10:44<11:58:37,  3.20s/it]
  9%|▊         | 1272/14725 [1:10:48<11:52:01,  3.18s/it]
  9%|▊         | 1273/14725 [1:10:51<11:45:40,  3.15s/it]
  9%|▊         | 1274/14725 [1:10:54<11:46:48,  3.15s/it]
  9%|▊         | 1275/14725 [1:10:57<11:53:38,  3.18s/it]
  9%|▊         | 1276/14725 [1:11:00<11:55:11,  3.19s/it]
  9%|▊         | 1277/14725 [1:11:04<11:55:47,  3.19s/it]
  9%|▊         | 1278/14725 [1:11:07<12:06:48,  3.24s/it]
  9%|▊         | 1279/14725 [1:11:10<12:07:06,  3.24s/it]
  9%|▊         | 1280/14725 [1:11:13<12:13:49,  3.27s/it]
                                                         

  9%|▊         | 1280/14725 [1:11:13<12:13:49,  3.27s/it]
  9%|▊         | 1281/14725 [1:11:17<12:12:54,  3.27s/it]
  9%|▊         | 1282/14725 [1:11:20<12:17:10,  3.29s/it]
  9%|▊         | 1283/14725 [1:11:23<12:13:59,  3.28s/it]
  9%|▊         | 1284/14725 [1:11:27<12:16:41,  3.29s/it]
  9%|▊         | 1285/14725 [1:11:30<12:12:16, 
0: {'loss': 0.7519, 'grad_norm': 0.4447988126352698, 'learning_rate': 1e-05, 'epoch': 0.09}
0:  3.27s/it]
  9%|▊         | 1286/14725 [1:11:33<12:12:56,  3.27s/it]
  9%|▊         | 1287/14725 [1:11:36<12:04:55,  3.24s/it]
  9%|▊         | 1288/14725 [1:11:40<12:03:49,  3.23s/it]
  9%|▉         | 1289/14725 [1:11:43<12:11:58,  3.27s/it]
  9%|▉         | 1290/14725 [1:11:46<12:06:35,  3.24s/it]
                                                         

  9%|▉         | 1290/14725 [1:11:46<12:06:35,  3.24s/it]
  9%|▉         | 1291/14725 [1:11:49<11:54:30,  3.19s/it]
  9%|▉         | 1292/14725 [1:11:53<12:08:07,  3.25s/it]
  9%|▉         | 1293/14725 [1:11:56<12:16:38,  3.29s/it]
  9%|▉         | 1294/14725 [1:11:59<12:25:12,  3.33s/it]
  9%|▉         | 1295/14725 [1:12:02<12:15:03,  3.28s/it]
  9%|▉         | 1296/14725 [1:12:06<12:11:50,  3.27s/it]
  9%|▉         | 1297/14725 [1:12:09<12:03:57,  3.23s/it]
  9%|▉         | 1298/14725 [1:12:12<12:03:29,  3.23s/it]
  9%|▉         | 1299/14725 [1:12:15<12:02:37,  3.23s/it]
  9%|▉         | 1300/14725 [1:12:19<12:03:00,  3.23
0: {'loss': 0.7696, 'grad_norm': 0.5480160229002221, 'learning_rate': 1e-05, 'epoch': 0.09}
0: {'loss': 0.7714, 'grad_norm': 0.5875129938582575, 'learning_rate': 1e-05, 'epoch': 0.09}
0: s/it]
                                                         

  9%|▉         | 1300/14725 [1:12:19<12:03:00,  3.23s/it]
  9%|▉         | 1301/14725 [1:12:22<12:24:59,  3.33s/it]
  9%|▉         | 1302/14725 [1:12:25<12:13:33,  3.28s/it]
  9%|▉         | 1303/14725 [1:12:29<12:24:35,  3.33s/it]
  9%|▉         | 1304/14725 [1:12:32<12:29:34,  3.35s/it]
  9%|▉         | 1305/14725 [1:12:35<12:21:27,  3.31s/it]
  9%|▉         | 1306/14725 [1:12:39<12:13:06,  3.28s/it]
  9%|▉         | 1307/14725 [1:12:42<12:06:55,  3.25s/it]
  9%|▉         | 1308/14725 [1:12:45<12:09:59,  3.26s/it]
  9%|▉         | 1309/14725 [1:12:48<12:23:18,  3.32s/it]
  9%|▉         | 1310/14725 [1:12:52<12:22:10,  3.32s/it]
                                                         

  9%|▉         | 1310/14725 [1:12:52<12:22:10,  3.32s/it]
  9%|▉         | 1311/14725 [1:12:55<12:23:01,  3.32s/it]
  9%|▉         | 1312/14725 [1:12:58<12:09:33,  3.26s/it]
  9%|▉         | 1313/14725 [1:13:02<12:25:53,  3.34s/it]
0: {'loss': 0.7744, 'grad_norm': 0.48585525078180236, 'learning_rate': 1e-05, 'epoch': 0.09}
0:   9%|▉         | 1314/14725 [1:13:05<12:22:16,  3.32s/it]
  9%|▉         | 1315/14725 [1:13:08<12:23:15,  3.33s/it]
  9%|▉         | 1316/14725 [1:13:12<12:23:18,  3.33s/it]
  9%|▉         | 1317/14725 [1:13:15<12:23:58,  3.33s/it]
  9%|▉         | 1318/14725 [1:13:18<12:17:36,  3.30s/it]
  9%|▉         | 1319/14725 [1:13:22<12:20:29,  3.31s/it]
  9%|▉         | 1320/14725 [1:13:25<12:08:26,  3.26s/it]
                                                         

  9%|▉         | 1320/14725 [1:13:25<12:08:26,  3.26s/it]
  9%|▉         | 1321/14725 [1:13:28<12:13:22,  3.28s/it]
  9%|▉         | 1322/14725 [1:13:31<12:00:30,  3.23s/it]
  9%|▉         | 1323/14725 [1:13:34<11:57:13,  3.21s/it]
  9%|▉         | 1324/14725 [1:13:38<11:56:53,  3.21s/it]
  9%|▉         | 1325/14725 [1:13:41<12:00:49,  3.23s/it]
  9%|▉         | 1326/14725 [1:13:44<11:57:01,  3.21s/it]
  9%|▉         | 1327/14725 [1:13:47<11:47:48,  3.17s/it]
  9%|▉         | 1328/14725 [1:13:50<11:56:24,  3.21s/it]
  9%|
0: {'loss': 0.7532, 'grad_norm': 0.4870482594745457, 'learning_rate': 1e-05, 'epoch': 0.09}
0: {'loss': 0.7706, 'grad_norm': 0.4364063079403578, 'learning_rate': 1e-05, 'epoch': 0.09}
0: ▉         | 1329/14725 [1:13:54<12:01:47,  3.23s/it]
  9%|▉         | 1330/14725 [1:13:57<11:47:34,  3.17s/it]
                                                         

  9%|▉         | 1330/14725 [1:13:57<11:47:34,  3.17s/it]
  9%|▉         | 1331/14725 [1:14:00<11:59:20,  3.22s/it]
  9%|▉         | 1332/14725 [1:14:03<12:00:06,  3.23s/it]
  9%|▉         | 1333/14725 [1:14:07<12:16:45,  3.30s/it]
  9%|▉         | 1334/14725 [1:14:10<12:28:18,  3.35s/it]
  9%|▉         | 1335/14725 [1:14:13<12:00:44,  3.23s/it]
  9%|▉         | 1336/14725 [1:14:16<11:51:18,  3.19s/it]
  9%|▉         | 1337/14725 [1:14:20<11:57:59,  3.22s/it]
  9%|▉         | 1338/14725 [1:14:23<12:10:53,  3.28s/it]
  9%|▉         | 1339/14725 [1:14:26<12:14:26,  3.29s/it]
  9%|▉         | 1340/14725 [1:14:29<12:06:01,  3.25s/it]
                                                         

  9%|▉         | 1340/14725 [1:14:29<12:06:01,  3.25s/it]
  9%|▉         | 1341/14725 [1:14:33<12:16:28,  3.30s/it]
  9%|▉   
0: {'loss': 0.7689, 'grad_norm': 0.4905339904169453, 'learning_rate': 1e-05, 'epoch': 0.09}
0:       | 1342/14725 [1:14:36<12:07:41,  3.26s/it]
  9%|▉         | 1343/14725 [1:14:39<11:55:13,  3.21s/it]
  9%|▉         | 1344/14725 [1:14:42<11:48:36,  3.18s/it]
  9%|▉         | 1345/14725 [1:14:45<11:49:52,  3.18s/it]
  9%|▉         | 1346/14725 [1:14:49<11:59:56,  3.23s/it]
  9%|▉         | 1347/14725 [1:14:52<12:07:41,  3.26s/it]
  9%|▉         | 1348/14725 [1:14:55<12:08:12,  3.27s/it]
  9%|▉         | 1349/14725 [1:14:59<12:13:33,  3.29s/it]
  9%|▉         | 1350/14725 [1:15:02<12:06:49,  3.26s/it]
                                                         

  9%|▉         | 1350/14725 [1:15:02<12:06:49,  3.26s/it]
  9%|▉         | 1351/14725 [1:15:05<12:02:32,  3.24s/it]
  9%|▉         | 1352/14725 [1:15:08<12:10:32,  3.28s/it]
  9%|▉         | 1353/14725 [1:15:12<12:05:51,  3.26s/it]
  9%|▉         | 1354/14725 [1:15:15<12:33:44,  3.38s/it]
  9%|▉         | 1355/14725 [1:15:19<12:19:50,  3.32s/it]
  9%|▉         | 1356/14725 [1:15:22<12:06:08,  3.26s/it]
  9%|▉        
0: {'loss': 0.7709, 'grad_norm': 0.44341774119047733, 'learning_rate': 1e-05, 'epoch': 0.09}
0: {'loss': 0.7624, 'grad_norm': 0.47570562903136804, 'learning_rate': 1e-05, 'epoch': 0.09}
0:  | 1357/14725 [1:15:25<12:09:27,  3.27s/it]
  9%|▉         | 1358/14725 [1:15:28<12:13:08,  3.29s/it]
  9%|▉         | 1359/14725 [1:15:32<12:08:49,  3.27s/it]
  9%|▉         | 1360/14725 [1:15:35<12:21:10,  3.33s/it]
                                                         

  9%|▉         | 1360/14725 [1:15:35<12:21:10,  3.33s/it]
  9%|▉         | 1361/14725 [1:15:38<12:21:46,  3.33s/it]
  9%|▉         | 1362/14725 [1:15:42<12:22:53,  3.34s/it]
  9%|▉         | 1363/14725 [1:15:45<12:04:54,  3.26s/it]
  9%|▉         | 1364/14725 [1:15:48<12:05:49,  3.26s/it]
  9%|▉         | 1365/14725 [1:15:51<12:19:20,  3.32s/it]
  9%|▉         | 1366/14725 [1:15:55<12:13:50,  3.30s/it]
  9%|▉         | 1367/14725 [1:15:58<12:19:28,  3.32s/it]
  9%|▉         | 1368/14725 [1:16:01<12:11:07,  3.28s/it]
  9%|▉         | 1369/14725 [1:16:04<11:59:31,  3.23s/it]
  9%|▉         | 1370/14725 [1:16:08<11:56:19,  3.22s/it]
                                                         

  9%|▉         | 137
0: {'loss': 0.7718, 'grad_norm': 0.4729230111852387, 'learning_rate': 1e-05, 'epoch': 0.09}
0: 0/14725 [1:16:08<11:56:19,  3.22s/it]
  9%|▉         | 1371/14725 [1:16:11<11:55:51,  3.22s/it]
  9%|▉         | 1372/14725 [1:16:14<11:57:53,  3.23s/it]
  9%|▉         | 1373/14725 [1:16:17<11:45:28,  3.17s/it]
  9%|▉         | 1374/14725 [1:16:20<11:46:29,  3.17s/it]
  9%|▉         | 1375/14725 [1:16:24<12:01:40,  3.24s/it]
  9%|▉         | 1376/14725 [1:16:27<12:03:25,  3.25s/it]
  9%|▉         | 1377/14725 [1:16:31<12:28:46,  3.37s/it]
  9%|▉         | 1378/14725 [1:16:34<12:09:14,  3.28s/it]
  9%|▉         | 1379/14725 [1:16:37<12:08:21,  3.27s/it]
  9%|▉         | 1380/14725 [1:16:40<12:28:20,  3.36s/it]
                                                         

  9%|▉         | 1380/14725 [1:16:41<12:28:20,  3.36s/it]
  9%|▉         | 1381/14725 [1:16:44<12:12:02,  3.29s/it]
  9%|▉         | 1382/14725 [1:16:47<12:22:25,  3.34s/it]
  9%|▉         | 1383/14725 [1:16:50<12:17:08,  3.32s/it]
  9%|▉         | 1384/14725 [1:16:54<12:19:24,  3.33s/it]
  9%|▉         | 1385/147
0: {'loss': 0.7652, 'grad_norm': 0.49715046754567227, 'learning_rate': 1e-05, 'epoch': 0.09}
0: 25 [1:16:57<12:14:47,  3.30s/it]
  9%|▉         | 1386/14725 [1:17:00<12:09:36,  3.28s/it]
  9%|▉         | 1387/14725 [1:17:03<12:09:08,  3.28s/it]
  9%|▉         | 1388/14725 [1:17:07<12:09:38,  3.28s/it]
  9%|▉         | 1389/14725 [1:17:10<12:10:35,  3.29s/it]
  9%|▉         | 1390/14725 [1:17:13<11:59:45,  3.24s/it]
                                                         

  9%|▉         | 1390/14725 [1:17:13<11:59:45,  3.24s/it]
  9%|▉         | 1391/14725 [1:17:16<11:45:38,  3.18s/it]
  9%|▉         | 1392/14725 [1:17:19<11:49:26,  3.19s/it]
  9%|▉         | 1393/14725 [1:17:23<11:55:33,  3.22s/it]
  9%|▉         | 1394/14725 [1:17:26<11:55:48,  3.22s/it]
  9%|▉         | 1395/14725 [1:17:29<12:00:42,  3.24s/it]
  9%|▉         | 1396/14725 [1:17:32<12:02:47,  3.25s/it]
  9%|▉         | 1397/14725 [1:17:36<12:08:54,  3.28s/it]
  9%|▉         | 1398/14725 [1:17:39<12:12:04,  3.30s/it]
 10%|▉         | 1399/14725 [1:17:42<12:04:26,  3.26s/it]
 10%|▉         | 1400/14725 [1
0: {'loss': 0.7623, 'grad_norm': 0.5217191228529663, 'learning_rate': 1e-05, 'epoch': 0.1}
0: {'loss': 0.7656, 'grad_norm': 0.47855626348102026, 'learning_rate': 1e-05, 'epoch': 0.1}
0: :17:45<11:57:45,  3.23s/it]
                                                         

 10%|▉         | 1400/14725 [1:17:46<11:57:45,  3.23s/it]
 10%|▉         | 1401/14725 [1:17:49<12:04:51,  3.26s/it]
 10%|▉         | 1402/14725 [1:17:52<12:11:04,  3.29s/it]
 10%|▉         | 1403/14725 [1:17:55<12:04:18,  3.26s/it]
 10%|▉         | 1404/14725 [1:17:59<11:57:28,  3.23s/it]
 10%|▉         | 1405/14725 [1:18:02<12:05:45,  3.27s/it]
 10%|▉         | 1406/14725 [1:18:05<12:05:07,  3.27s/it]
 10%|▉         | 1407/14725 [1:18:08<12:09:20,  3.29s/it]
 10%|▉         | 1408/14725 [1:18:12<12:16:23,  3.32s/it]
 10%|▉         | 1409/14725 [1:18:15<12:28:59,  3.37s/it]
 10%|▉         | 1410/14725 [1:18:19<12:42:25,  3.44s/it]
                                                         

 10%|▉         | 1410/14725 [1:18:19<12:42:25,  3.44s/it]
 10%|▉         | 1411/14725 [1:18:22<12:28:20,  3.37s/it]
 10%|▉         | 1412/14725 [1:18:26<12:26:53,  3.37s/it]
 10%|▉         | 1413/14725 [1:18:29
0: {'loss': 0.752, 'grad_norm': 0.4451125419483311, 'learning_rate': 1e-05, 'epoch': 0.1}
0: <12:24:00,  3.35s/it]
 10%|▉         | 1414/14725 [1:18:32<12:15:59,  3.32s/it]
 10%|▉         | 1415/14725 [1:18:35<12:08:01,  3.28s/it]
 10%|▉         | 1416/14725 [1:18:39<12:13:46,  3.31s/it]
 10%|▉         | 1417/14725 [1:18:42<12:04:38,  3.27s/it]
 10%|▉         | 1418/14725 [1:18:45<12:08:45,  3.29s/it]
 10%|▉         | 1419/14725 [1:18:48<12:05:36,  3.27s/it]
 10%|▉         | 1420/14725 [1:18:52<12:12:34,  3.30s/it]
                                                         

 10%|▉         | 1420/14725 [1:18:52<12:12:34,  3.30s/it]
 10%|▉         | 1421/14725 [1:18:55<11:57:08,  3.23s/it]
 10%|▉         | 1422/14725 [1:18:58<11:52:44,  3.21s/it]
 10%|▉         | 1423/14725 [1:19:01<11:49:47,  3.20s/it]
 10%|▉         | 1424/14725 [1:19:04<11:37:58,  3.15s/it]
 10%|▉         | 1425/14725 [1:19:07<11:38:49,  3.15s/it]
 10%|▉         | 1426/14725 [1:19:11<11:43:50,  3.18s/it]
 10%|▉         | 1427/14725 [1:19:14<12:01:15,  3.25s/it]
 10%|▉         | 1428/14725 [1:19:17<11:5
0: {'loss': 0.7546, 'grad_norm': 0.47657589475072915, 'learning_rate': 1e-05, 'epoch': 0.1}
0: {'loss': 0.7732, 'grad_norm': 0.4937848272682166, 'learning_rate': 1e-05, 'epoch': 0.1}
0: 4:37,  3.22s/it]
 10%|▉         | 1429/14725 [1:19:20<11:50:54,  3.21s/it]
 10%|▉         | 1430/14725 [1:19:24<11:58:17,  3.24s/it]
                                                         

 10%|▉         | 1430/14725 [1:19:24<11:58:17,  3.24s/it]
 10%|▉         | 1431/14725 [1:19:27<11:57:01,  3.24s/it]
 10%|▉         | 1432/14725 [1:19:30<11:52:58,  3.22s/it]
 10%|▉         | 1433/14725 [1:19:34<12:08:53,  3.29s/it]
 10%|▉         | 1434/14725 [1:19:37<12:07:43,  3.29s/it]
 10%|▉         | 1435/14725 [1:19:41<12:35:27,  3.41s/it]
 10%|▉         | 1436/14725 [1:19:44<12:22:12,  3.35s/it]
 10%|▉         | 1437/14725 [1:19:47<12:08:42,  3.29s/it]
 10%|▉         | 1438/14725 [1:19:50<12:07:26,  3.28s/it]
 10%|▉         | 1439/14725 [1:19:53<12:02:58,  3.26s/it]
 10%|▉         | 1440/14725 [1:19:57<12:01:51,  3.26s/it]
                                                         

 10%|▉         | 1440/14725 [1:19:57<12:01:51,  3.26s/it]
 10%|▉         | 1441/14725 [1:20:00<12:08:02, 
0: {'loss': 0.7685, 'grad_norm': 0.48045040802442557, 'learning_rate': 1e-05, 'epoch': 0.1}
0:  3.29s/it]
 10%|▉         | 1442/14725 [1:20:03<12:08:46,  3.29s/it]
 10%|▉         | 1443/14725 [1:20:07<12:07:09,  3.28s/it]
 10%|▉         | 1444/14725 [1:20:10<12:14:19,  3.32s/it]
 10%|▉         | 1445/14725 [1:20:13<12:07:36,  3.29s/it]
 10%|▉         | 1446/14725 [1:20:17<12:11:56,  3.31s/it]
 10%|▉         | 1447/14725 [1:20:20<12:09:52,  3.30s/it]
 10%|▉         | 1448/14725 [1:20:23<12:02:06,  3.26s/it]
 10%|▉         | 1449/14725 [1:20:26<12:01:44,  3.26s/it]
 10%|▉         | 1450/14725 [1:20:29<11:55:09,  3.23s/it]
                                                         

 10%|▉         | 1450/14725 [1:20:29<11:55:09,  3.23s/it]
 10%|▉         | 1451/14725 [1:20:32<11:45:04,  3.19s/it]
 10%|▉         | 1452/14725 [1:20:36<11:57:32,  3.24s/it]
 10%|▉         | 1453/14725 [1:20:39<11:48:45,  3.20s/it]
 10%|▉         | 1454/14725 [1:20:42<11:46:25,  3.19s/it]
 10%|▉         | 1455/14725 [1:20:45<11:39:52,  3.16s/it]
 10%|▉         | 1456/14725 [1:20:48<11:44:35,  3.19
0: {'loss': 0.7446, 'grad_norm': 0.48499051593438314, 'learning_rate': 1e-05, 'epoch': 0.1}
0: {'loss': 0.7729, 'grad_norm': 0.48135498134221155, 'learning_rate': 1e-05, 'epoch': 0.1}
0: s/it]
 10%|▉         | 1457/14725 [1:20:52<11:38:24,  3.16s/it]
 10%|▉         | 1458/14725 [1:20:55<11:52:36,  3.22s/it]
 10%|▉         | 1459/14725 [1:20:58<12:00:13,  3.26s/it]
 10%|▉         | 1460/14725 [1:21:01<11:43:17,  3.18s/it]
                                                         

 10%|▉         | 1460/14725 [1:21:01<11:43:17,  3.18s/it]
 10%|▉         | 1461/14725 [1:21:05<12:13:56,  3.32s/it]
 10%|▉         | 1462/14725 [1:21:08<12:06:33,  3.29s/it]
 10%|▉         | 1463/14725 [1:21:11<11:56:07,  3.24s/it]
 10%|▉         | 1464/14725 [1:21:15<11:56:37,  3.24s/it]
 10%|▉         | 1465/14725 [1:21:18<12:09:39,  3.30s/it]
 10%|▉         | 1466/14725 [1:21:21<12:14:01,  3.32s/it]
 10%|▉         | 1467/14725 [1:21:25<12:04:52,  3.28s/it]
 10%|▉         | 1468/14725 [1:21:28<12:01:37,  3.27s/it]
 10%|▉         | 1469/14725 [1:21:31<12:04:26,  3.28s/it]
 10%|▉         | 1470/14725 [1:21:34<12:03:30,  3.28s/it]
                                                         

0: {'loss': 0.7611, 'grad_norm': 0.5641999912435416, 'learning_rate': 1e-05, 'epoch': 0.1}
0:  10%|▉         | 1470/14725 [1:21:34<12:03:30,  3.28s/it]
 10%|▉         | 1471/14725 [1:21:38<12:02:37,  3.27s/it]
 10%|▉         | 1472/14725 [1:21:41<11:54:20,  3.23s/it]
 10%|█         | 1473/14725 [1:21:44<11:49:47,  3.21s/it]
 10%|█         | 1474/14725 [1:21:47<11:39:18,  3.17s/it]
 10%|█         | 1475/14725 [1:21:50<11:39:56,  3.17s/it]
 10%|█         | 1476/14725 [1:21:54<11:57:24,  3.25s/it]
 10%|█         | 1477/14725 [1:21:57<12:01:37,  3.27s/it]
 10%|█         | 1478/14725 [1:22:00<11:57:52,  3.25s/it]
 10%|█         | 1479/14725 [1:22:03<12:04:19,  3.28s/it]
 10%|█         | 1480/14725 [1:22:07<11:51:29,  3.22s/it]
                                                         

 10%|█         | 1480/14725 [1:22:07<11:51:29,  3.22s/it]
 10%|█         | 1481/14725 [1:22:10<11:48:29,  3.21s/it]
 10%|█         | 1482/14725 [1:22:13<11:55:12,  3.24s/it]
 10%|█         | 1483/14725 [1:22:16<12:08:31,  3.30s/it]
 10%|█         | 1484/14725 [1:22:20<12:12:27,  3.32s/it]
 10%|
0: {'loss': 0.7697, 'grad_norm': 0.5046861351230786, 'learning_rate': 1e-05, 'epoch': 0.1}
0: █         | 1485/14725 [1:22:23<12:01:20,  3.27s/it]
 10%|█         | 1486/14725 [1:22:26<12:00:36,  3.27s/it]
 10%|█         | 1487/14725 [1:22:29<11:59:30,  3.26s/it]
 10%|█         | 1488/14725 [1:22:33<11:54:44,  3.24s/it]
 10%|█         | 1489/14725 [1:22:36<12:02:08,  3.27s/it]
 10%|█         | 1490/14725 [1:22:39<11:48:30,  3.21s/it]
                                                         

 10%|█         | 1490/14725 [1:22:39<11:48:30,  3.21s/it]
 10%|█         | 1491/14725 [1:22:42<11:46:21,  3.20s/it]
 10%|█         | 1492/14725 [1:22:45<11:45:03,  3.20s/it]
 10%|█         | 1493/14725 [1:22:49<11:44:29,  3.19s/it]
 10%|█         | 1494/14725 [1:22:52<11:43:14,  3.19s/it]
 10%|█         | 1495/14725 [1:22:55<11:36:08,  3.16s/it]
 10%|█         | 1496/14725 [1:22:58<11:38:42,  3.17s/it]
 10%|█         | 1497/14725 [1:23:01<11:47:07,  3.21s/it]
 10%|█         | 1498/14725 [1:23:05<11:52:02,  3.23s/it]
 10%|█         | 1499/14725 [1:23:08<11:45:16,  3.20s/it]
 10%|█  
0: {'loss': 0.7498, 'grad_norm': 0.458583689713018, 'learning_rate': 1e-05, 'epoch': 0.1}
0: {'loss': 0.7592, 'grad_norm': 0.4476138625196788, 'learning_rate': 1e-05, 'epoch': 0.1}
0:        | 1500/14725 [1:23:11<11:41:29,  3.18s/it]
                                                         

 10%|█         | 1500/14725 [1:23:11<11:41:29,  3.18s/it]
 10%|█         | 1501/14725 [1:23:14<11:49:12,  3.22s/it]
 10%|█         | 1502/14725 [1:23:17<11:46:06,  3.20s/it]
 10%|█         | 1503/14725 [1:23:21<11:54:49,  3.24s/it]
 10%|█         | 1504/14725 [1:23:24<11:53:48,  3.24s/it]
 10%|█         | 1505/14725 [1:23:27<11:53:32,  3.24s/it]
 10%|█         | 1506/14725 [1:23:30<11:50:56,  3.23s/it]
 10%|█         | 1507/14725 [1:23:34<11:53:20,  3.24s/it]
 10%|█         | 1508/14725 [1:23:37<11:57:04,  3.26s/it]
 10%|█         | 1509/14725 [1:23:40<11:58:44,  3.26s/it]
 10%|█         | 1510/14725 [1:23:44<11:57:35,  3.26s/it]
                                                         

 10%|█         | 1510/14725 [1:23:44<11:57:35,  3.26s/it]
 10%|█         | 1511/14725 [1:23:47<12:03:06,  3.28s/it]
 10%|█         | 1512/14725 [1:23:50<11:52:05,  3.23s/it]
 10%|█        
0: {'loss': 0.7718, 'grad_norm': 0.4496418680426288, 'learning_rate': 1e-05, 'epoch': 0.1}
0:  | 1513/14725 [1:23:53<11:42:19,  3.19s/it]
 10%|█         | 1514/14725 [1:23:57<12:01:00,  3.27s/it]
 10%|█         | 1515/14725 [1:24:00<11:54:03,  3.24s/it]
 10%|█         | 1516/14725 [1:24:03<11:44:59,  3.20s/it]
 10%|█         | 1517/14725 [1:24:06<11:53:25,  3.24s/it]
 10%|█         | 1518/14725 [1:24:10<12:06:09,  3.30s/it]
 10%|█         | 1519/14725 [1:24:13<12:05:54,  3.30s/it]
 10%|█         | 1520/14725 [1:24:16<11:59:52,  3.27s/it]
                                                         

 10%|█         | 1520/14725 [1:24:16<11:59:52,  3.27s/it]
 10%|█         | 1521/14725 [1:24:19<12:08:52,  3.31s/it]
 10%|█         | 1522/14725 [1:24:23<12:02:06,  3.28s/it]
 10%|█         | 1523/14725 [1:24:26<11:48:32,  3.22s/it]
 10%|█         | 1524/14725 [1:24:29<11:40:20,  3.18s/it]
 10%|█         | 1525/14725 [1:24:32<11:56:52,  3.26s/it]
 10%|█         | 1526/14725 [1:24:36<11:56:52,  3.26s/it]
 10%|█         | 1527/14725 [1:24:39<11:48:56,  3.22s/it]
 10%|█         | 15
0: {'loss': 0.752, 'grad_norm': 0.44555233918938075, 'learning_rate': 1e-05, 'epoch': 0.1}
0: {'loss': 0.7726, 'grad_norm': 0.47561696499827644, 'learning_rate': 1e-05, 'epoch': 0.1}
0: 28/14725 [1:24:42<12:01:06,  3.28s/it]
 10%|█         | 1529/14725 [1:24:45<11:49:10,  3.22s/it]
 10%|█         | 1530/14725 [1:24:48<11:38:40,  3.18s/it]
                                                         

 10%|█         | 1530/14725 [1:24:48<11:38:40,  3.18s/it]
 10%|█         | 1531/14725 [1:24:52<11:44:43,  3.20s/it]
 10%|█         | 1532/14725 [1:24:55<11:45:46,  3.21s/it]
 10%|█         | 1533/14725 [1:24:58<11:56:01,  3.26s/it]
 10%|█         | 1534/14725 [1:25:02<12:04:07,  3.29s/it]
 10%|█         | 1535/14725 [1:25:05<11:57:06,  3.26s/it]
 10%|█         | 1536/14725 [1:25:08<12:08:00,  3.31s/it]
 10%|█         | 1537/14725 [1:25:11<12:05:32,  3.30s/it]
 10%|█         | 1538/14725 [1:25:15<12:11:39,  3.33s/it]
 10%|█         | 1539/14725 [1:25:18<12:18:15,  3.36s/it]
 10%|█         | 1540/14725 [1:25:22<12:15:36,  3.35s/it]
                                                         

 10%|█         | 1540/14725 [1:25:22<12:15:36,  3.35s/it]
 10%|█         | 1541/147
0: {'loss': 0.7506, 'grad_norm': 0.5561364355328791, 'learning_rate': 1e-05, 'epoch': 0.11}
0: 25 [1:25:25<12:07:00,  3.31s/it]
 10%|█         | 1542/14725 [1:25:28<12:15:40,  3.35s/it]
 10%|█         | 1543/14725 [1:25:31<12:05:28,  3.30s/it]
 10%|█         | 1544/14725 [1:25:35<12:07:52,  3.31s/it]
 10%|█         | 1545/14725 [1:25:38<12:01:13,  3.28s/it]
 10%|█         | 1546/14725 [1:25:41<11:55:50,  3.26s/it]
 11%|█         | 1547/14725 [1:25:44<11:41:00,  3.19s/it]
 11%|█         | 1548/14725 [1:25:47<11:39:20,  3.18s/it]
 11%|█         | 1549/14725 [1:25:50<11:35:44,  3.17s/it]
 11%|█         | 1550/14725 [1:25:54<11:38:37,  3.18s/it]
                                                         

 11%|█         | 1550/14725 [1:25:54<11:38:37,  3.18s/it]
 11%|█         | 1551/14725 [1:25:57<11:45:35,  3.21s/it]
 11%|█         | 1552/14725 [1:26:00<11:56:47,  3.26s/it]
 11%|█         | 1553/14725 [1:26:04<12:02:45,  3.29s/it]
 11%|█         | 1554/14725 [1:26:07<12:01:01,  3.28s/it]
 11%|█         | 1555/14725 [1:26:10<12:07:24,  3.31s/it]
 11%|█         | 1556/14725 [1
0: {'loss': 0.7606, 'grad_norm': 0.5053328728097269, 'learning_rate': 1e-05, 'epoch': 0.11}
0: :26:13<11:54:38,  3.26s/it]
 11%|█         | 1557/14725 [1:26:17<11:45:20,  3.21s/it]
 11%|█         | 1558/14725 [1:26:20<11:49:33,  3.23s/it]
 11%|█         | 1559/14725 [1:26:23<12:00:31,  3.28s/it]
 11%|█         | 1560/14725 [1:26:27<11:56:58,  3.27s/it]
                                                         

 11%|█         | 1560/14725 [1:26:27<11:56:58,  3.27s/it]
 11%|█         | 1561/14725 [1:26:30<11:48:50,  3.23s/it]
 11%|█         | 1562/14725 [1:26:33<11:40:31,  3.19s/it]
 11%|█         | 1563/14725 [1:26:36<11:40:10,  3.19s/it]
 11%|█         | 1564/14725 [1:26:39<11:58:59,  3.28s/it]
 11%|█         | 1565/14725 [1:26:43<12:01:49,  3.29s/it]
 11%|█         | 1566/14725 [1:26:46<12:08:36,  3.32s/it]
 11%|█         | 1567/14725 [1:26:49<11:55:08,  3.26s/it]
 11%|█         | 1568/14725 [1:26:53<11:53:56,  3.26s/it]
 11%|█         | 1569/14725 [1:26:56<11:52:11,  3.25s/it]
 11%|█         | 1570/14725 [1:26:59<11:45:07,  3.22s/it]
                                     
0: {'loss': 0.7507, 'grad_norm': 0.5134671419581252, 'learning_rate': 1e-05, 'epoch': 0.11}
0: {'loss': 0.7561, 'grad_norm': 0.5216285826452874, 'learning_rate': 1e-05, 'epoch': 0.11}
0:                     

 11%|█         | 1570/14725 [1:26:59<11:45:07,  3.22s/it]
 11%|█         | 1571/14725 [1:27:02<11:35:42,  3.17s/it]
 11%|█         | 1572/14725 [1:27:05<11:29:53,  3.15s/it]
 11%|█         | 1573/14725 [1:27:08<11:27:05,  3.13s/it]
 11%|█         | 1574/14725 [1:27:12<12:03:12,  3.30s/it]
 11%|█         | 1575/14725 [1:27:15<11:49:19,  3.24s/it]
 11%|█         | 1576/14725 [1:27:19<12:16:02,  3.36s/it]
 11%|█         | 1577/14725 [1:27:22<12:09:56,  3.33s/it]
 11%|█         | 1578/14725 [1:27:25<12:13:10,  3.35s/it]
 11%|█         | 1579/14725 [1:27:28<12:07:25,  3.32s/it]
 11%|█         | 1580/14725 [1:27:32<12:09:30,  3.33s/it]
                                                         

 11%|█         | 1580/14725 [1:27:32<12:09:30,  3.33s/it]
 11%|█         | 1581/14725 [1:27:36<12:49:08,  3.51s/it]
 11%|█         | 1582/14725 [1:27:39<12:30:44,  3.43s/it]
 11%|█         | 1583/14725 [1:27:42<12:28:43,  3.42s/it]
 11%|█         | 1584/14725 [1:27:46<12:4
0: {'loss': 0.7567, 'grad_norm': 0.5400006406812882, 'learning_rate': 1e-05, 'epoch': 0.11}
0: 7:09,  3.50s/it]
 11%|█         | 1585/14725 [1:27:49<12:31:44,  3.43s/it]
 11%|█         | 1586/14725 [1:27:53<12:15:35,  3.36s/it]
 11%|█         | 1587/14725 [1:27:56<12:14:29,  3.35s/it]
 11%|█         | 1588/14725 [1:27:59<12:04:25,  3.31s/it]
 11%|█         | 1589/14725 [1:28:02<12:06:06,  3.32s/it]
 11%|█         | 1590/14725 [1:28:06<12:12:19,  3.35s/it]
                                                         

 11%|█         | 1590/14725 [1:28:06<12:12:19,  3.35s/it]
 11%|█         | 1591/14725 [1:28:09<12:10:31,  3.34s/it]
 11%|█         | 1592/14725 [1:28:12<12:07:35,  3.32s/it]
 11%|█         | 1593/14725 [1:28:16<12:13:13,  3.35s/it]
 11%|█         | 1594/14725 [1:28:19<12:09:22,  3.33s/it]
 11%|█         | 1595/14725 [1:28:22<12:02:59,  3.30s/it]
 11%|█         | 1596/14725 [1:28:26<11:55:27,  3.27s/it]
 11%|█         | 1597/14725 [1:28:29<11:57:41,  3.28s/it]
 11%|█         | 1598/14725 [1:28:32<12:18:04,  3.37s/it]
 11%|█         | 1599/14725 [1:28:36<12:07:26,
0: {'loss': 0.7748, 'grad_norm': 0.4497187187334978, 'learning_rate': 1e-05, 'epoch': 0.11}
0: {'loss': 0.7602, 'grad_norm': 0.4659781861675706, 'learning_rate': 1e-05, 'epoch': 0.11}
0:   3.33s/it]
 11%|█         | 1600/14725 [1:28:39<12:11:37,  3.34s/it]
                                                         

 11%|█         | 1600/14725 [1:28:39<12:11:37,  3.34s/it]
 11%|█         | 1601/14725 [1:28:42<12:15:42,  3.36s/it]
 11%|█         | 1602/14725 [1:28:46<12:02:53,  3.31s/it]
 11%|█         | 1603/14725 [1:28:49<11:57:25,  3.28s/it]
 11%|█         | 1604/14725 [1:28:52<11:56:29,  3.28s/it]
 11%|█         | 1605/14725 [1:28:55<11:55:20,  3.27s/it]
 11%|█         | 1606/14725 [1:28:59<11:55:07,  3.27s/it]
 11%|█         | 1607/14725 [1:29:02<11:51:17,  3.25s/it]
 11%|█         | 1608/14725 [1:29:05<11:47:32,  3.24s/it]
 11%|█         | 1609/14725 [1:29:08<11:49:15,  3.24s/it]
 11%|█         | 1610/14725 [1:29:12<11:44:35,  3.22s/it]
                                                         

 11%|█         | 1610/14725 [1:29:12<11:44:35,  3.22s/it]
 11%|█         | 1611/14725 [1:29:15<12:01:06,  3.30s/it]
 11%|█         | 1612/14725 [1:29:18<12:13:20,  3.36
0: {'loss': 0.7701, 'grad_norm': 0.45604179500080927, 'learning_rate': 1e-05, 'epoch': 0.11}
0: s/it]
 11%|█         | 1613/14725 [1:29:22<11:52:06,  3.26s/it]
 11%|█         | 1614/14725 [1:29:25<11:46:08,  3.23s/it]
 11%|█         | 1615/14725 [1:29:28<12:07:46,  3.33s/it]
 11%|█         | 1616/14725 [1:29:31<11:54:19,  3.27s/it]
 11%|█         | 1617/14725 [1:29:35<11:45:59,  3.23s/it]
 11%|█         | 1618/14725 [1:29:38<11:49:44,  3.25s/it]
 11%|█         | 1619/14725 [1:29:41<11:49:06,  3.25s/it]
 11%|█         | 1620/14725 [1:29:44<11:54:57,  3.27s/it]
                                                         

 11%|█         | 1620/14725 [1:29:44<11:54:57,  3.27s/it]
 11%|█         | 1621/14725 [1:29:48<11:58:36,  3.29s/it]
 11%|█         | 1622/14725 [1:29:51<12:04:17,  3.32s/it]
 11%|█         | 1623/14725 [1:29:54<12:02:41,  3.31s/it]
 11%|█         | 1624/14725 [1:29:58<12:01:44,  3.31s/it]
 11%|█         | 1625/14725 [1:30:01<11:45:37,  3.23s/it]
 11%|█         | 1626/14725 [1:30:04<11:54:55,  3.27s/it]
 11%|█         | 1627/14725 [1:30:07<11:48:25,  3.25s/it]
0: {'loss': 0.7719, 'grad_norm': 0.46725389395498007, 'learning_rate': 1e-05, 'epoch': 0.11}
0: {'loss': 0.765, 'grad_norm': 0.5175656873134274, 'learning_rate': 1e-05, 'epoch': 0.11}
0: 
 11%|█         | 1628/14725 [1:30:11<11:59:00,  3.29s/it]
 11%|█         | 1629/14725 [1:30:14<11:59:42,  3.30s/it]
 11%|█         | 1630/14725 [1:30:17<11:55:46,  3.28s/it]
                                                         

 11%|█         | 1630/14725 [1:30:17<11:55:46,  3.28s/it]
 11%|█         | 1631/14725 [1:30:21<11:56:43,  3.28s/it]
 11%|█         | 1632/14725 [1:30:24<11:58:48,  3.29s/it]
 11%|█         | 1633/14725 [1:30:27<12:03:26,  3.32s/it]
 11%|█         | 1634/14725 [1:30:31<12:03:33,  3.32s/it]
 11%|█         | 1635/14725 [1:30:34<12:00:18,  3.30s/it]
 11%|█         | 1636/14725 [1:30:37<12:06:40,  3.33s/it]
 11%|█         | 1637/14725 [1:30:41<12:08:46,  3.34s/it]
 11%|█         | 1638/14725 [1:30:44<11:57:36,  3.29s/it]
 11%|█         | 1639/14725 [1:30:47<11:56:37,  3.29s/it]
 11%|█         | 1640/14725 [1:30:50<11:50:06,  3.26s/it]
                                                         

 11%|█         | 1640/14725 [1:30:50<11:50:06,  3.26s/it]
 11%|
0: {'loss': 0.7573, 'grad_norm': 0.44575181355802435, 'learning_rate': 1e-05, 'epoch': 0.11}
0: █         | 1641/14725 [1:30:53<11:48:00,  3.25s/it]
 11%|█         | 1642/14725 [1:30:57<12:03:58,  3.32s/it]
 11%|█         | 1643/14725 [1:31:00<11:46:15,  3.24s/it]
 11%|█         | 1644/14725 [1:31:03<11:54:47,  3.28s/it]
 11%|█         | 1645/14725 [1:31:07<12:03:44,  3.32s/it]
 11%|█         | 1646/14725 [1:31:10<11:51:14,  3.26s/it]
 11%|█         | 1647/14725 [1:31:13<11:52:37,  3.27s/it]
 11%|█         | 1648/14725 [1:31:16<11:53:02,  3.27s/it]
 11%|█         | 1649/14725 [1:31:20<11:47:04,  3.24s/it]
 11%|█         | 1650/14725 [1:31:23<11:47:25,  3.25s/it]
                                                         

 11%|█         | 1650/14725 [1:31:23<11:47:25,  3.25s/it]
 11%|█         | 1651/14725 [1:31:26<11:42:13,  3.22s/it]
 11%|█         | 1652/14725 [1:31:29<11:37:41,  3.20s/it]
 11%|█         | 1653/14725 [1:31:33<11:55:46,  3.29s/it]
 11%|█         | 1654/14725 [1:31:36<11:51:23,  3.27s/it]
 11%|█         | 1655/14725 [1:31:39<11:59:40,  3.30s/it]
 11%|█  
0: {'loss': 0.7643, 'grad_norm': 0.4634085286561666, 'learning_rate': 1e-05, 'epoch': 0.11}
0:        | 1656/14725 [1:31:43<12:00:28,  3.31s/it]
 11%|█▏        | 1657/14725 [1:31:46<11:51:59,  3.27s/it]
 11%|█▏        | 1658/14725 [1:31:49<11:49:53,  3.26s/it]
 11%|█▏        | 1659/14725 [1:31:52<11:36:50,  3.20s/it]
 11%|█▏        | 1660/14725 [1:31:55<11:36:43,  3.20s/it]
                                                         

 11%|█▏        | 1660/14725 [1:31:55<11:36:43,  3.20s/it]
 11%|█▏        | 1661/14725 [1:31:59<11:42:01,  3.22s/it]
 11%|█▏        | 1662/14725 [1:32:02<11:42:32,  3.23s/it]
 11%|█▏        | 1663/14725 [1:32:05<11:51:48,  3.27s/it]
 11%|█▏        | 1664/14725 [1:32:08<11:54:01,  3.28s/it]
 11%|█▏        | 1665/14725 [1:32:12<11:55:25,  3.29s/it]
 11%|█▏        | 1666/14725 [1:32:15<11:52:23,  3.27s/it]
 11%|█▏        | 1667/14725 [1:32:18<12:05:21,  3.33s/it]
 11%|█▏        | 1668/14725 [1:32:22<12:03:34,  3.33s/it]
 11%|█▏        | 1669/14725 [1:32:25<11:59:26,  3.31s/it]
 11%|█▏        | 1670/14725 [1:32:28<12:01:
0: {'loss': 0.7485, 'grad_norm': 0.49556164467450586, 'learning_rate': 1e-05, 'epoch': 0.11}
0: {'loss': 0.7563, 'grad_norm': 0.5362517640449481, 'learning_rate': 1e-05, 'epoch': 0.11}
0: 18,  3.32s/it]
                                                         

 11%|█▏        | 1670/14725 [1:32:28<12:01:18,  3.32s/it]
 11%|█▏        | 1671/14725 [1:32:32<11:47:53,  3.25s/it]
 11%|█▏        | 1672/14725 [1:32:35<11:43:50,  3.24s/it]
 11%|█▏        | 1673/14725 [1:32:38<11:50:59,  3.27s/it]
 11%|█▏        | 1674/14725 [1:32:41<11:40:03,  3.22s/it]
 11%|█▏        | 1675/14725 [1:32:44<11:44:43,  3.24s/it]
 11%|█▏        | 1676/14725 [1:32:47<11:31:33,  3.18s/it]
 11%|█▏        | 1677/14725 [1:32:51<11:39:03,  3.21s/it]
 11%|█▏        | 1678/14725 [1:32:54<11:52:25,  3.28s/it]
 11%|█▏        | 1679/14725 [1:32:58<11:55:29,  3.29s/it]
 11%|█▏        | 1680/14725 [1:33:01<11:51:54,  3.27s/it]
                                                         

 11%|█▏        | 1680/14725 [1:33:01<11:51:54,  3.27s/it]
 11%|█▏        | 1681/14725 [1:33:04<11:43:44,  3.24s/it]
 11%|█▏        | 1682/14725 [1:33:07<11:44:17,  3.24s/it]
 11%|█▏        | 16
0: {'loss': 0.7489, 'grad_norm': 0.5422538471508933, 'learning_rate': 1e-05, 'epoch': 0.11}
0: 83/14725 [1:33:11<12:02:00,  3.32s/it]
 11%|█▏        | 1684/14725 [1:33:14<12:06:42,  3.34s/it]
 11%|█▏        | 1685/14725 [1:33:18<12:13:03,  3.37s/it]
 11%|█▏        | 1686/14725 [1:33:21<11:54:16,  3.29s/it]
 11%|█▏        | 1687/14725 [1:33:24<11:46:42,  3.25s/it]
 11%|█▏        | 1688/14725 [1:33:27<11:52:13,  3.28s/it]
 11%|█▏        | 1689/14725 [1:33:30<11:52:08,  3.28s/it]
 11%|█▏        | 1690/14725 [1:33:34<11:49:38,  3.27s/it]
                                                         

 11%|█▏        | 1690/14725 [1:33:34<11:49:38,  3.27s/it]
 11%|█▏        | 1691/14725 [1:33:37<11:55:03,  3.29s/it]
 11%|█▏        | 1692/14725 [1:33:41<12:12:38,  3.37s/it]
 11%|█▏        | 1693/14725 [1:33:44<12:16:15,  3.39s/it]
 12%|█▏        | 1694/14725 [1:33:47<12:14:16,  3.38s/it]
 12%|█▏        | 1695/14725 [1:33:51<12:01:32,  3.32s/it]
 12%|█▏        | 1696/14725 [1:33:54<11:57:06,  3.30s/it]
 12%|█▏        | 1697/14725 [1:33:57<11:56:09,  3.30s/
0: {'loss': 0.7577, 'grad_norm': 0.4637525762706042, 'learning_rate': 1e-05, 'epoch': 0.12}
0: {'loss': 0.7539, 'grad_norm': 0.48337001310456484, 'learning_rate': 1e-05, 'epoch': 0.12}
0: it]
 12%|█▏        | 1698/14725 [1:34:00<11:47:25,  3.26s/it]
 12%|█▏        | 1699/14725 [1:34:04<11:56:10,  3.30s/it]
 12%|█▏        | 1700/14725 [1:34:07<11:56:59,  3.30s/it]
                                                         

 12%|█▏        | 1700/14725 [1:34:07<11:56:59,  3.30s/it]
 12%|█▏        | 1701/14725 [1:34:10<11:50:50,  3.27s/it]
 12%|█▏        | 1702/14725 [1:34:13<11:45:24,  3.25s/it]
 12%|█▏        | 1703/14725 [1:34:17<12:03:09,  3.33s/it]
 12%|█▏        | 1704/14725 [1:34:20<11:54:33,  3.29s/it]
 12%|█▏        | 1705/14725 [1:34:23<11:43:33,  3.24s/it]
 12%|█▏        | 1706/14725 [1:34:26<11:39:22,  3.22s/it]
 12%|█▏        | 1707/14725 [1:34:30<11:52:31,  3.28s/it]
 12%|█▏        | 1708/14725 [1:34:33<11:49:26,  3.27s/it]
 12%|█▏        | 1709/14725 [1:34:36<11:37:07,  3.21s/it]
 12%|█▏        | 1710/14725 [1:34:39<11:34:25,  3.20s/it]
                                                         

 12%|█▏        | 1710/14725 [1
0: {'loss': 0.753, 'grad_norm': 0.495711287573877, 'learning_rate': 1e-05, 'epoch': 0.12}
0: :34:39<11:34:25,  3.20s/it]
 12%|█▏        | 1711/14725 [1:34:43<11:38:01,  3.22s/it]
 12%|█▏        | 1712/14725 [1:34:46<11:41:42,  3.24s/it]
 12%|█▏        | 1713/14725 [1:34:49<11:33:35,  3.20s/it]
 12%|█▏        | 1714/14725 [1:34:52<11:34:03,  3.20s/it]
 12%|█▏        | 1715/14725 [1:34:55<11:28:46,  3.18s/it]
 12%|█▏        | 1716/14725 [1:34:58<11:30:17,  3.18s/it]
 12%|█▏        | 1717/14725 [1:35:02<11:33:59,  3.20s/it]
 12%|█▏        | 1718/14725 [1:35:05<11:25:24,  3.16s/it]
 12%|█▏        | 1719/14725 [1:35:08<11:40:53,  3.23s/it]
 12%|█▏        | 1720/14725 [1:35:12<11:53:43,  3.29s/it]
                                                         

 12%|█▏        | 1720/14725 [1:35:12<11:53:43,  3.29s/it]
 12%|█▏        | 1721/14725 [1:35:15<11:52:37,  3.29s/it]
 12%|█▏        | 1722/14725 [1:35:18<11:51:58,  3.29s/it]
 12%|█▏        | 1723/14725 [1:35:21<11:52:55,  3.29s/it]
 12%|█▏        | 1724/14725 [1:35:25<11:41:54,  3.24s/it]
 12%|�
0: {'loss': 0.7628, 'grad_norm': 0.4693651789975761, 'learning_rate': 1e-05, 'epoch': 0.12}
0: �▏        | 1725/14725 [1:35:28<11:44:40,  3.25s/it]
 12%|█▏        | 1726/14725 [1:35:31<11:46:18,  3.26s/it]
 12%|█▏        | 1727/14725 [1:35:34<11:52:23,  3.29s/it]
 12%|█▏        | 1728/14725 [1:35:38<12:06:39,  3.35s/it]
 12%|█▏        | 1729/14725 [1:35:41<12:01:43,  3.33s/it]
 12%|█▏        | 1730/14725 [1:35:44<11:45:36,  3.26s/it]
                                                         

 12%|█▏        | 1730/14725 [1:35:44<11:45:36,  3.26s/it]
 12%|█▏        | 1731/14725 [1:35:48<11:50:42,  3.28s/it]
 12%|█▏        | 1732/14725 [1:35:51<11:44:23,  3.25s/it]
 12%|█▏        | 1733/14725 [1:35:54<11:36:11,  3.22s/it]
 12%|█▏        | 1734/14725 [1:35:57<11:35:51,  3.21s/it]
 12%|█▏        | 1735/14725 [1:36:00<11:32:26,  3.20s/it]
 12%|█▏        | 1736/14725 [1:36:04<11:35:29,  3.21s/it]
 12%|█▏        | 1737/14725 [1:36:07<11:26:52,  3.17s/it]
 12%|█▏        | 1738/14725 [1:36:10<11:34:44,  3.21s/it]
 12%|█▏        | 1739/14725 [1:36:13<1
0: {'loss': 0.7525, 'grad_norm': 0.4631317324645485, 'learning_rate': 1e-05, 'epoch': 0.12}
0: {'loss': 0.7618, 'grad_norm': 0.43933460432089116, 'learning_rate': 1e-05, 'epoch': 0.12}
0: 1:46:14,  3.26s/it]
 12%|█▏        | 1740/14725 [1:36:17<11:41:42,  3.24s/it]
                                                         

 12%|█▏        | 1740/14725 [1:36:17<11:41:42,  3.24s/it]
 12%|█▏        | 1741/14725 [1:36:20<11:51:25,  3.29s/it]
 12%|█▏        | 1742/14725 [1:36:23<12:01:33,  3.33s/it]
 12%|█▏        | 1743/14725 [1:36:27<11:55:15,  3.31s/it]
 12%|█▏        | 1744/14725 [1:36:30<11:46:15,  3.26s/it]
 12%|█▏        | 1745/14725 [1:36:33<11:43:39,  3.25s/it]
 12%|█▏        | 1746/14725 [1:36:37<11:58:57,  3.32s/it]
 12%|█▏        | 1747/14725 [1:36:40<11:56:13,  3.31s/it]
 12%|█▏        | 1748/14725 [1:36:43<11:54:43,  3.30s/it]
 12%|█▏        | 1749/14725 [1:36:46<11:48:01,  3.27s/it]
 12%|█▏        | 1750/14725 [1:36:49<11:40:11,  3.24s/it]
                                                         

 12%|█▏        | 1750/14725 [1:36:49<11:40:11,  3.24s/it]
 12%|█▏        | 1751/14725 [1:36:53<11:53:58,  3.30s/it]
 12%|█▏       
0: {'loss': 0.7455, 'grad_norm': 0.4633682265507792, 'learning_rate': 1e-05, 'epoch': 0.12}
0:  | 1752/14725 [1:36:56<11:49:44,  3.28s/it]
 12%|█▏        | 1753/14725 [1:36:59<11:45:41,  3.26s/it]
 12%|█▏        | 1754/14725 [1:37:03<11:42:53,  3.25s/it]
 12%|█▏        | 1755/14725 [1:37:06<11:31:06,  3.20s/it]
 12%|█▏        | 1756/14725 [1:37:09<11:27:05,  3.18s/it]
 12%|█▏        | 1757/14725 [1:37:12<11:26:11,  3.17s/it]
 12%|█▏        | 1758/14725 [1:37:15<11:30:11,  3.19s/it]
 12%|█▏        | 1759/14725 [1:37:18<11:31:52,  3.20s/it]
 12%|█▏        | 1760/14725 [1:37:22<11:38:08,  3.23s/it]
                                                         

 12%|█▏        | 1760/14725 [1:37:22<11:38:08,  3.23s/it]
 12%|█▏        | 1761/14725 [1:37:25<11:29:40,  3.19s/it]
 12%|█▏        | 1762/14725 [1:37:28<11:36:14,  3.22s/it]
 12%|█▏        | 1763/14725 [1:37:31<11:46:23,  3.27s/it]
 12%|█▏        | 1764/14725 [1:37:35<11:39:57,  3.24s/it]
 12%|█▏        | 1765/14725 [1:37:38<11:27:12,  3.18s/it]
 12%|█▏        | 1766/14725 [1:37:41<11:37:33,  3
0: {'loss': 0.7577, 'grad_norm': 0.4437514489204924, 'learning_rate': 1e-05, 'epoch': 0.12}
0: .23s/it]
 12%|█▏        | 1767/14725 [1:37:44<11:26:43,  3.18s/it]
 12%|█▏        | 1768/14725 [1:37:47<11:39:32,  3.24s/it]
 12%|█▏        | 1769/14725 [1:37:51<11:38:44,  3.24s/it]
 12%|█▏        | 1770/14725 [1:37:54<11:38:16,  3.23s/it]
                                                         

 12%|█▏        | 1770/14725 [1:37:54<11:38:16,  3.23s/it]
 12%|█▏        | 1771/14725 [1:37:57<11:45:33,  3.27s/it]
 12%|█▏        | 1772/14725 [1:38:00<11:39:26,  3.24s/it]
 12%|█▏        | 1773/14725 [1:38:04<11:41:01,  3.25s/it]
 12%|█▏        | 1774/14725 [1:38:07<11:46:13,  3.27s/it]
 12%|█▏        | 1775/14725 [1:38:10<11:43:48,  3.26s/it]
 12%|█▏        | 1776/14725 [1:38:14<11:42:53,  3.26s/it]
 12%|█▏        | 1777/14725 [1:38:17<11:52:09,  3.30s/it]
 12%|█▏        | 1778/14725 [1:38:20<11:44:36,  3.27s/it]
 12%|█▏        | 1779/14725 [1:38:23<11:34:53,  3.22s/it]
 12%|█▏        | 1780/14725 [1:38:27<11:37:45,  3.23s/it]
                          
0: {'loss': 0.756, 'grad_norm': 0.4486062269107222, 'learning_rate': 1e-05, 'epoch': 0.12}
0: {'loss': 0.7573, 'grad_norm': 0.47075530369570123, 'learning_rate': 1e-05, 'epoch': 0.12}
0:                                

 12%|█▏        | 1780/14725 [1:38:27<11:37:45,  3.23s/it]
 12%|█▏        | 1781/14725 [1:38:30<11:48:26,  3.28s/it]
 12%|█▏        | 1782/14725 [1:38:33<11:48:03,  3.28s/it]
 12%|█▏        | 1783/14725 [1:38:36<11:44:40,  3.27s/it]
 12%|█▏        | 1784/14725 [1:38:40<11:50:06,  3.29s/it]
 12%|█▏        | 1785/14725 [1:38:43<11:45:53,  3.27s/it]
 12%|█▏        | 1786/14725 [1:38:46<11:30:18,  3.20s/it]
 12%|█▏        | 1787/14725 [1:38:49<11:30:38,  3.20s/it]
 12%|█▏        | 1788/14725 [1:38:53<11:57:43,  3.33s/it]
 12%|█▏        | 1789/14725 [1:38:58<13:30:27,  3.76s/it]
 12%|█▏        | 1790/14725 [1:39:01<13:13:58,  3.68s/it]
                                                         

 12%|█▏        | 1790/14725 [1:39:01<13:13:58,  3.68s/it]
 12%|█▏        | 1791/14725 [1:39:05<13:04:52,  3.64s/it]
 12%|█▏        | 1792/14725 [1:39:08<12:30:46,  3.48s/it]
 12%|█▏        | 1793/14725 [1:39:11<12:07:27,  3.38s/it]
 1
0: {'loss': 0.754, 'grad_norm': 0.4688172394837873, 'learning_rate': 1e-05, 'epoch': 0.12}
0: 2%|█▏        | 1794/14725 [1:39:14<12:00:39,  3.34s/it]
 12%|█▏        | 1795/14725 [1:39:17<11:53:38,  3.31s/it]
 12%|█▏        | 1796/14725 [1:39:21<11:51:49,  3.30s/it]
 12%|█▏        | 1797/14725 [1:39:24<11:44:31,  3.27s/it]
 12%|█▏        | 1798/14725 [1:39:27<11:39:18,  3.25s/it]
 12%|█▏        | 1799/14725 [1:39:30<11:33:53,  3.22s/it]
 12%|█▏        | 1800/14725 [1:39:34<11:36:23,  3.23s/it]
                                                         

 12%|█▏        | 1800/14725 [1:39:34<11:36:23,  3.23s/it]
 12%|█▏        | 1801/14725 [1:39:37<11:32:56,  3.22s/it]
 12%|█▏        | 1802/14725 [1:39:40<11:30:57,  3.21s/it]
 12%|█▏        | 1803/14725 [1:39:43<11:36:15,  3.23s/it]
 12%|█▏        | 1804/14725 [1:39:46<11:36:08,  3.23s/it]
 12%|█▏        | 1805/14725 [1:39:50<11:46:41,  3.28s/it]
 12%|█▏        | 1806/14725 [1:39:53<11:37:41,  3.24s/it]
 12%|█▏        | 1807/14725 [1:39:56<11:36:24,  3.23s/it]
 12%|█▏        | 1808/14725 [1:40
0: {'loss': 0.754, 'grad_norm': 0.4889854112636995, 'learning_rate': 1e-05, 'epoch': 0.12}
0: {'loss': 0.7412, 'grad_norm': 0.4499994550089357, 'learning_rate': 1e-05, 'epoch': 0.12}
0: :00<11:49:11,  3.29s/it]
 12%|█▏        | 1809/14725 [1:40:03<11:33:20,  3.22s/it]
 12%|█▏        | 1810/14725 [1:40:06<11:40:25,  3.25s/it]
                                                         

 12%|█▏        | 1810/14725 [1:40:06<11:40:25,  3.25s/it]
 12%|█▏        | 1811/14725 [1:40:09<11:37:34,  3.24s/it]
 12%|█▏        | 1812/14725 [1:40:13<11:54:18,  3.32s/it]
 12%|█▏        | 1813/14725 [1:40:16<11:57:40,  3.33s/it]
 12%|█▏        | 1814/14725 [1:40:19<11:43:03,  3.27s/it]
 12%|█▏        | 1815/14725 [1:40:22<11:35:00,  3.23s/it]
 12%|█▏        | 1816/14725 [1:40:26<11:35:05,  3.23s/it]
 12%|█▏        | 1817/14725 [1:40:29<11:38:04,  3.24s/it]
 12%|█▏        | 1818/14725 [1:40:32<11:26:13,  3.19s/it]
 12%|█▏        | 1819/14725 [1:40:35<11:33:32,  3.22s/it]
 12%|█▏        | 1820/14725 [1:40:38<11:31:41,  3.22s/it]
                                                         

 12%|█▏        | 1820/14725 [1:40:38<11:31:41,  3.22s/it]
 12%|█▏  
0: {'loss': 0.7482, 'grad_norm': 0.4640388053200242, 'learning_rate': 1e-05, 'epoch': 0.12}
0:       | 1821/14725 [1:40:42<11:38:39,  3.25s/it]
 12%|█▏        | 1822/14725 [1:40:45<11:39:08,  3.25s/it]
 12%|█▏        | 1823/14725 [1:40:48<11:40:31,  3.26s/it]
 12%|█▏        | 1824/14725 [1:40:52<11:44:28,  3.28s/it]
 12%|█▏        | 1825/14725 [1:40:55<11:36:52,  3.24s/it]
 12%|█▏        | 1826/14725 [1:40:58<11:25:01,  3.19s/it]
 12%|█▏        | 1827/14725 [1:41:01<11:21:03,  3.17s/it]
 12%|█▏        | 1828/14725 [1:41:04<11:28:34,  3.20s/it]
 12%|█▏        | 1829/14725 [1:41:07<11:31:23,  3.22s/it]
 12%|█▏        | 1830/14725 [1:41:11<11:40:38,  3.26s/it]
                                                         

 12%|█▏        | 1830/14725 [1:41:11<11:40:38,  3.26s/it]
 12%|█▏        | 1831/14725 [1:41:14<11:40:49,  3.26s/it]
 12%|█▏        | 1832/14725 [1:41:17<11:31:50,  3.22s/it]
 12%|█▏        | 1833/14725 [1:41:20<11:32:31,  3.22s/it]
 12%|█▏        | 1834/14725 [1:41:24<11:38:44,  3.25s/it]
 12%|█▏        | 1835/14725 [1:41:27<11:42:2
0: {'loss': 0.7569, 'grad_norm': 0.47115207337696463, 'learning_rate': 1e-05, 'epoch': 0.12}
0: 5,  3.27s/it]
 12%|█▏        | 1836/14725 [1:41:30<11:40:57,  3.26s/it]
 12%|█▏        | 1837/14725 [1:41:34<11:39:19,  3.26s/it]
 12%|█▏        | 1838/14725 [1:41:37<11:41:19,  3.27s/it]
 12%|█▏        | 1839/14725 [1:41:40<11:48:20,  3.30s/it]
 12%|█▏        | 1840/14725 [1:41:44<11:51:53,  3.31s/it]
                                                         

 12%|█▏        | 1840/14725 [1:41:44<11:51:53,  3.31s/it]
 13%|█▎        | 1841/14725 [1:41:47<11:50:35,  3.31s/it]
 13%|█▎        | 1842/14725 [1:41:50<11:39:40,  3.26s/it]
 13%|█▎        | 1843/14725 [1:41:53<11:34:42,  3.24s/it]
 13%|█▎        | 1844/14725 [1:41:56<11:34:42,  3.24s/it]
 13%|█▎        | 1845/14725 [1:42:00<11:35:43,  3.24s/it]
 13%|█▎        | 1846/14725 [1:42:03<11:41:30,  3.27s/it]
 13%|█▎        | 1847/14725 [1:42:06<11:33:15,  3.23s/it]
 13%|█▎        | 1848/14725 [1:42:09<11:21:26,  3.18s/it]
 13%|█▎        | 1849/14725 [1:42:12<11:23:33,  3.19s/it]
 13%|█▎        | 
0: {'loss': 0.7647, 'grad_norm': 0.4401949901245932, 'learning_rate': 1e-05, 'epoch': 0.13}
0: {'loss': 0.7346, 'grad_norm': 0.4921400787067114, 'learning_rate': 1e-05, 'epoch': 0.13}
0: 1850/14725 [1:42:16<11:18:37,  3.16s/it]
                                                         

 13%|█▎        | 1850/14725 [1:42:16<11:18:37,  3.16s/it]
 13%|█▎        | 1851/14725 [1:42:19<11:15:44,  3.15s/it]
 13%|█▎        | 1852/14725 [1:42:22<11:09:38,  3.12s/it]
 13%|█▎        | 1853/14725 [1:42:25<11:12:20,  3.13s/it]
 13%|█▎        | 1854/14725 [1:42:28<11:14:55,  3.15s/it]
 13%|█▎        | 1855/14725 [1:42:31<11:17:29,  3.16s/it]
 13%|█▎        | 1856/14725 [1:42:34<11:15:08,  3.15s/it]
 13%|█▎        | 1857/14725 [1:42:38<11:20:14,  3.17s/it]
 13%|█▎        | 1858/14725 [1:42:41<11:23:26,  3.19s/it]
 13%|█▎        | 1859/14725 [1:42:44<11:39:20,  3.26s/it]
 13%|█▎        | 1860/14725 [1:42:48<11:42:03,  3.27s/it]
                                                         

 13%|█▎        | 1860/14725 [1:42:48<11:42:03,  3.27s/it]
 13%|█▎        | 1861/14725 [1:42:51<11:55:56,  3.34s/it]
 13%|█▎        | 1862/14725 [1:42:54<11:49:23,  3.31s/i
0: {'loss': 0.7505, 'grad_norm': 0.4741100025728742, 'learning_rate': 1e-05, 'epoch': 0.13}
0: t]
 13%|█▎        | 1863/14725 [1:42:58<11:53:52,  3.33s/it]
 13%|█▎        | 1864/14725 [1:43:01<11:58:04,  3.35s/it]
 13%|█▎        | 1865/14725 [1:43:04<11:56:58,  3.35s/it]
 13%|█▎        | 1866/14725 [1:43:08<12:05:36,  3.39s/it]
 13%|█▎        | 1867/14725 [1:43:11<11:56:00,  3.34s/it]
 13%|█▎        | 1868/14725 [1:43:14<11:55:24,  3.34s/it]
 13%|█▎        | 1869/14725 [1:43:18<11:47:23,  3.30s/it]
 13%|█▎        | 1870/14725 [1:43:21<11:27:05,  3.21s/it]
                                                         

 13%|█▎        | 1870/14725 [1:43:21<11:27:05,  3.21s/it]
 13%|█▎        | 1871/14725 [1:43:24<11:28:56,  3.22s/it]
 13%|█▎        | 1872/14725 [1:43:27<11:24:24,  3.19s/it]
 13%|█▎        | 1873/14725 [1:43:30<11:37:34,  3.26s/it]
 13%|█▎        | 1874/14725 [1:43:34<11:43:50,  3.29s/it]
 13%|█▎        | 1875/14725 [1:43:37<11:48:07,  3.31s/it]
 13%|█▎        | 1876/14725 [1:43:40<11:42:03,  3.28s/it]
 13%|█▎        | 1877/14725 
0: {'loss': 0.7533, 'grad_norm': 0.48793753903476356, 'learning_rate': 1e-05, 'epoch': 0.13}
0: {'loss': 0.7654, 'grad_norm': 0.5492432968419685, 'learning_rate': 1e-05, 'epoch': 0.13}
0: [1:43:44<11:45:29,  3.29s/it]
 13%|█▎        | 1878/14725 [1:43:47<11:45:56,  3.30s/it]
 13%|█▎        | 1879/14725 [1:43:50<11:38:59,  3.26s/it]
 13%|█▎        | 1880/14725 [1:43:54<11:51:02,  3.32s/it]
                                                         

 13%|█▎        | 1880/14725 [1:43:54<11:51:02,  3.32s/it]
 13%|█▎        | 1881/14725 [1:43:57<11:50:16,  3.32s/it]
 13%|█▎        | 1882/14725 [1:44:00<11:49:38,  3.32s/it]
 13%|█▎        | 1883/14725 [1:44:03<11:44:12,  3.29s/it]
 13%|█▎        | 1884/14725 [1:44:07<11:54:19,  3.34s/it]
 13%|█▎        | 1885/14725 [1:44:10<12:05:42,  3.39s/it]
 13%|█▎        | 1886/14725 [1:44:14<11:56:34,  3.35s/it]
 13%|█▎        | 1887/14725 [1:44:17<11:49:02,  3.31s/it]
 13%|█▎        | 1888/14725 [1:44:20<11:49:06,  3.31s/it]
 13%|█▎        | 1889/14725 [1:44:24<12:03:30,  3.38s/it]
 13%|█▎        | 1890/14725 [1:44:27<11:55:49,  3.35s/it]
                                                         

 13%|█
0: {'loss': 0.7541, 'grad_norm': 0.4431205868687572, 'learning_rate': 1e-05, 'epoch': 0.13}
0: ▎        | 1890/14725 [1:44:27<11:55:49,  3.35s/it]
 13%|█▎        | 1891/14725 [1:44:30<11:40:19,  3.27s/it]
 13%|█▎        | 1892/14725 [1:44:33<11:45:07,  3.30s/it]
 13%|█▎        | 1893/14725 [1:44:37<11:41:18,  3.28s/it]
 13%|█▎        | 1894/14725 [1:44:40<11:37:46,  3.26s/it]
 13%|█▎        | 1895/14725 [1:44:43<11:38:37,  3.27s/it]
 13%|█▎        | 1896/14725 [1:44:46<11:36:44,  3.26s/it]
 13%|█▎        | 1897/14725 [1:44:50<11:32:20,  3.24s/it]
 13%|█▎        | 1898/14725 [1:44:53<11:48:45,  3.32s/it]
 13%|█▎        | 1899/14725 [1:44:56<11:48:33,  3.31s/it]
 13%|█▎        | 1900/14725 [1:45:00<11:37:29,  3.26s/it]
                                                         

 13%|█▎        | 1900/14725 [1:45:00<11:37:29,  3.26s/it]
 13%|█▎        | 1901/14725 [1:45:03<11:29:27,  3.23s/it]
 13%|█▎        | 1902/14725 [1:45:06<11:28:27,  3.22s/it]
 13%|█▎        | 1903/14725 [1:45:09<11:43:39,  3.29s/it]
 13%|█▎        | 1904/14725 [1:45:13<11
0: {'loss': 0.7649, 'grad_norm': 0.5233031008065966, 'learning_rate': 1e-05, 'epoch': 0.13}
0: :50:02,  3.32s/it]
 13%|█▎        | 1905/14725 [1:45:16<11:56:20,  3.35s/it]
 13%|█▎        | 1906/14725 [1:45:20<11:54:24,  3.34s/it]
 13%|█▎        | 1907/14725 [1:45:23<11:43:32,  3.29s/it]
 13%|█▎        | 1908/14725 [1:45:26<11:36:22,  3.26s/it]
 13%|█▎        | 1909/14725 [1:45:29<11:35:19,  3.26s/it]
 13%|█▎        | 1910/14725 [1:45:32<11:38:42,  3.27s/it]
                                                         

 13%|█▎        | 1910/14725 [1:45:32<11:38:42,  3.27s/it]
 13%|█▎        | 1911/14725 [1:45:36<11:39:36,  3.28s/it]
 13%|█▎        | 1912/14725 [1:45:39<11:42:46,  3.29s/it]
 13%|█▎        | 1913/14725 [1:45:42<11:28:38,  3.23s/it]
 13%|█▎        | 1914/14725 [1:45:45<11:27:08,  3.22s/it]
 13%|█▎        | 1915/14725 [1:45:49<11:38:49,  3.27s/it]
 13%|█▎        | 1916/14725 [1:45:52<11:37:28,  3.27s/it]
 13%|█▎        | 1917/14725 [1:45:55<11:45:51,  3.31s/it]
 13%|█▎        | 1918/14725 [1:45:59<11:41:35,  3.29s/it]
 13%|█▎     
0: {'loss': 0.7583, 'grad_norm': 0.4945836794967448, 'learning_rate': 1e-05, 'epoch': 0.13}
0: {'loss': 0.7694, 'grad_norm': 0.4652100114581035, 'learning_rate': 1e-05, 'epoch': 0.13}
0:    | 1919/14725 [1:46:02<11:34:31,  3.25s/it]
 13%|█▎        | 1920/14725 [1:46:05<11:48:13,  3.32s/it]
                                                         

 13%|█▎        | 1920/14725 [1:46:05<11:48:13,  3.32s/it]
 13%|█▎        | 1921/14725 [1:46:09<11:48:48,  3.32s/it]
 13%|█▎        | 1922/14725 [1:46:12<11:59:55,  3.37s/it]
 13%|█▎        | 1923/14725 [1:46:15<11:50:31,  3.33s/it]
 13%|█▎        | 1924/14725 [1:46:18<11:39:50,  3.28s/it]
 13%|█▎        | 1925/14725 [1:46:22<11:38:49,  3.28s/it]
 13%|█▎        | 1926/14725 [1:46:25<11:34:13,  3.25s/it]
 13%|█▎        | 1927/14725 [1:46:28<11:38:52,  3.28s/it]
 13%|█▎        | 1928/14725 [1:46:32<11:35:56,  3.26s/it]
 13%|█▎        | 1929/14725 [1:46:35<11:35:33,  3.26s/it]
 13%|█▎        | 1930/14725 [1:46:38<11:36:48,  3.27s/it]
                                                         

 13%|█▎        | 1930/14725 [1:46:38<11:36:48,  3.27s/it]
 13%|█▎        | 1931/14725 [1:46:41<11:27:36,  3.
0: {'loss': 0.7497, 'grad_norm': 0.46634132141008244, 'learning_rate': 1e-05, 'epoch': 0.13}
0: 22s/it]
 13%|█▎        | 1932/14725 [1:46:45<11:36:17,  3.27s/it]
 13%|█▎        | 1933/14725 [1:46:48<11:24:46,  3.21s/it]
 13%|█▎        | 1934/14725 [1:46:51<11:24:42,  3.21s/it]
 13%|█▎        | 1935/14725 [1:46:54<11:33:04,  3.25s/it]
 13%|█▎        | 1936/14725 [1:46:57<11:22:54,  3.20s/it]
 13%|█▎        | 1937/14725 [1:47:01<11:26:05,  3.22s/it]
 13%|█▎        | 1938/14725 [1:47:04<11:26:21,  3.22s/it]
 13%|█▎        | 1939/14725 [1:47:07<11:34:43,  3.26s/it]
 13%|█▎        | 1940/14725 [1:47:10<11:25:31,  3.22s/it]
                                                         

 13%|█▎        | 1940/14725 [1:47:10<11:25:31,  3.22s/it]
 13%|█▎        | 1941/14725 [1:47:13<11:23:48,  3.21s/it]
 13%|█▎        | 1942/14725 [1:47:17<11:32:06,  3.25s/it]
 13%|█▎        | 1943/14725 [1:47:20<11:29:08,  3.23s/it]
 13%|█▎        | 1944/14725 [1:47:23<11:46:34,  3.32s/it]
 13%|█▎        | 1945/14725 [1:47:27<11:36:15,  3.27s/it]
 13%|█▎        | 1946/1
0: {'loss': 0.7533, 'grad_norm': 0.5069992235302053, 'learning_rate': 1e-05, 'epoch': 0.13}
0: 4725 [1:47:30<11:51:36,  3.34s/it]
 13%|█▎        | 1947/14725 [1:47:33<11:38:54,  3.28s/it]
 13%|█▎        | 1948/14725 [1:47:37<11:41:24,  3.29s/it]
 13%|█▎        | 1949/14725 [1:47:40<11:49:34,  3.33s/it]
 13%|█▎        | 1950/14725 [1:47:43<11:46:12,  3.32s/it]
                                                         

 13%|█▎        | 1950/14725 [1:47:43<11:46:12,  3.32s/it]
 13%|█▎        | 1951/14725 [1:47:47<11:41:29,  3.29s/it]
 13%|█▎        | 1952/14725 [1:47:50<11:36:40,  3.27s/it]
 13%|█▎        | 1953/14725 [1:47:53<11:25:41,  3.22s/it]
 13%|█▎        | 1954/14725 [1:47:56<11:18:44,  3.19s/it]
 13%|█▎        | 1955/14725 [1:47:59<11:29:40,  3.24s/it]
 13%|█▎        | 1956/14725 [1:48:02<11:22:05,  3.21s/it]
 13%|█▎        | 1957/14725 [1:48:06<11:43:34,  3.31s/it]
 13%|█▎        | 1958/14725 [1:48:09<11:40:34,  3.29s/it]
 13%|█▎        | 1959/14725 [1:48:13<11:38:23,  3.28s/it]
 13%|█▎        | 1960/14725 [1:48:16<11:31:37,  3.25s/it]
0: {'loss': 0.7526, 'grad_norm': 0.4590659490829394, 'learning_rate': 1e-05, 'epoch': 0.13}
0: {'loss': 0.7578, 'grad_norm': 0.458335454507576, 'learning_rate': 1e-05, 'epoch': 0.13}
0:                                                          

 13%|█▎        | 1960/14725 [1:48:16<11:31:37,  3.25s/it]
 13%|█▎        | 1961/14725 [1:48:19<11:27:22,  3.23s/it]
 13%|█▎        | 1962/14725 [1:48:22<11:18:09,  3.19s/it]
 13%|█▎        | 1963/14725 [1:48:25<11:32:09,  3.25s/it]
 13%|█▎        | 1964/14725 [1:48:29<11:32:41,  3.26s/it]
 13%|█▎        | 1965/14725 [1:48:32<11:54:34,  3.36s/it]
 13%|█▎        | 1966/14725 [1:48:36<11:57:56,  3.38s/it]
 13%|█▎        | 1967/14725 [1:48:39<11:43:41,  3.31s/it]
 13%|█▎        | 1968/14725 [1:48:42<11:41:16,  3.30s/it]
 13%|█▎        | 1969/14725 [1:48:46<11:55:00,  3.36s/it]
 13%|█▎        | 1970/14725 [1:48:49<11:52:34,  3.35s/it]
                                                         

 13%|█▎        | 1970/14725 [1:48:49<11:52:34,  3.35s/it]
 13%|█▎        | 1971/14725 [1:48:52<11:50:29,  3.34s/it]
 13%|█▎        | 1972/14725 [1:48:56<11:48:14,  3.33s/it]
 13%|█▎        | 1973/14725 [1:48:
0: {'loss': 0.7454, 'grad_norm': 0.4698204637504206, 'learning_rate': 1e-05, 'epoch': 0.13}
0: 59<11:37:45,  3.28s/it]
 13%|█▎        | 1974/14725 [1:49:02<11:38:02,  3.28s/it]
 13%|█▎        | 1975/14725 [1:49:05<11:30:48,  3.25s/it]
 13%|█▎        | 1976/14725 [1:49:08<11:25:08,  3.22s/it]
 13%|█▎        | 1977/14725 [1:49:12<11:31:21,  3.25s/it]
 13%|█▎        | 1978/14725 [1:49:15<11:20:33,  3.20s/it]
 13%|█▎        | 1979/14725 [1:49:18<11:22:59,  3.22s/it]
 13%|█▎        | 1980/14725 [1:49:21<11:20:54,  3.21s/it]
                                                         

 13%|█▎        | 1980/14725 [1:49:21<11:20:54,  3.21s/it]
 13%|█▎        | 1981/14725 [1:49:24<11:14:07,  3.17s/it]
 13%|█▎        | 1982/14725 [1:49:28<11:26:26,  3.23s/it]
 13%|█▎        | 1983/14725 [1:49:31<11:44:20,  3.32s/it]
 13%|█▎        | 1984/14725 [1:49:35<11:51:37,  3.35s/it]
 13%|█▎        | 1985/14725 [1:49:38<11:36:32,  3.28s/it]
 13%|█▎        | 1986/14725 [1:49:41<11:40:06,  3.30s/it]
 13%|█▎        | 1987/14725 [1:49:44<11:37:03,  3.28s/it]
 14%|█▎
0: {'loss': 0.7483, 'grad_norm': 0.4693100925596399, 'learning_rate': 1e-05, 'epoch': 0.14}
0: {'loss': 0.7526, 'grad_norm': 0.4546232837457895, 'learning_rate': 1e-05, 'epoch': 0.14}
0:         | 1988/14725 [1:49:47<11:19:59,  3.20s/it]
 14%|█▎        | 1989/14725 [1:49:51<11:26:05,  3.23s/it]
 14%|█▎        | 1990/14725 [1:49:54<11:20:19,  3.21s/it]
                                                         

 14%|█▎        | 1990/14725 [1:49:54<11:20:19,  3.21s/it]
 14%|█▎        | 1991/14725 [1:49:57<11:25:13,  3.23s/it]
 14%|█▎        | 1992/14725 [1:50:00<11:18:26,  3.20s/it]
 14%|█▎        | 1993/14725 [1:50:03<11:13:05,  3.17s/it]
 14%|█▎        | 1994/14725 [1:50:07<11:28:30,  3.24s/it]
 14%|█▎        | 1995/14725 [1:50:10<11:30:23,  3.25s/it]
 14%|█▎        | 1996/14725 [1:50:13<11:23:11,  3.22s/it]
 14%|█▎        | 1997/14725 [1:50:16<11:25:48,  3.23s/it]
 14%|█▎        | 1998/14725 [1:50:19<11:18:44,  3.20s/it]
 14%|█▎        | 1999/14725 [1:50:23<11:22:06,  3.22s/it]
 14%|█▎        | 2000/14725 [1:50:26<11:16:56,  3.19s/it]
                                                         

 14%|█▎        | 2000/14725 [1:50:26<11:16:56
0: {'loss': 0.7455, 'grad_norm': 0.4495349121500381, 'learning_rate': 1e-05, 'epoch': 0.14}
0: ,  3.19s/it]
 14%|█▎        | 2001/14725 [1:50:29<11:25:42,  3.23s/it]
 14%|█▎        | 2002/14725 [1:50:32<11:21:21,  3.21s/it]
 14%|█▎        | 2003/14725 [1:50:36<11:21:57,  3.22s/it]
 14%|█▎        | 2004/14725 [1:50:39<11:32:18,  3.27s/it]
 14%|█▎        | 2005/14725 [1:50:43<11:50:38,  3.35s/it]
 14%|█▎        | 2006/14725 [1:50:46<11:47:27,  3.34s/it]
 14%|█▎        | 2007/14725 [1:50:49<11:35:59,  3.28s/it]
 14%|█▎        | 2008/14725 [1:50:52<11:26:54,  3.24s/it]
 14%|█▎        | 2009/14725 [1:50:55<11:24:54,  3.23s/it]
 14%|█▎        | 2010/14725 [1:50:58<11:19:35,  3.21s/it]
                                                         

 14%|█▎        | 2010/14725 [1:50:58<11:19:35,  3.21s/it]
 14%|█▎        | 2011/14725 [1:51:02<11:18:37,  3.20s/it]
 14%|█▎        | 2012/14725 [1:51:05<11:15:35,  3.19s/it]
 14%|█▎        | 2013/14725 [1:51:08<11:11:09,  3.17s/it]
 14%|█▎        | 2014/14725 [1:51:11<11:25:25,  3.24s/it]
 14%|█▎        | 2
0: {'loss': 0.7588, 'grad_norm': 0.4702544980271959, 'learning_rate': 1e-05, 'epoch': 0.14}
0: 015/14725 [1:51:15<11:22:13,  3.22s/it]
 14%|█▎        | 2016/14725 [1:51:18<11:24:31,  3.23s/it]
 14%|█▎        | 2017/14725 [1:51:21<11:40:14,  3.31s/it]
 14%|█▎        | 2018/14725 [1:51:25<11:38:55,  3.30s/it]
 14%|█▎        | 2019/14725 [1:51:28<11:47:13,  3.34s/it]
 14%|█▎        | 2020/14725 [1:51:31<11:43:34,  3.32s/it]
                                                         

 14%|█▎        | 2020/14725 [1:51:31<11:43:34,  3.32s/it]
 14%|█▎        | 2021/14725 [1:51:35<11:39:42,  3.30s/it]
 14%|█▎        | 2022/14725 [1:51:38<11:39:27,  3.30s/it]
 14%|█▎        | 2023/14725 [1:51:41<11:34:29,  3.28s/it]
 14%|█▎        | 2024/14725 [1:51:45<11:45:53,  3.33s/it]
 14%|█▍        | 2025/14725 [1:51:48<11:42:54,  3.32s/it]
 14%|█▍        | 2026/14725 [1:51:51<11:23:49,  3.23s/it]
 14%|█▍        | 2027/14725 [1:51:54<11:22:35,  3.23s/it]
 14%|█▍        | 2028/14725 [1:51:57<11:29:04,  3.26s/it]
 14%|█▍        | 2029/14725 [1:52:01<11:21:39,  3.22s
0: {'loss': 0.7535, 'grad_norm': 0.520537666670647, 'learning_rate': 1e-05, 'epoch': 0.14}
0: {'loss': 0.7598, 'grad_norm': 0.4735604545584214, 'learning_rate': 1e-05, 'epoch': 0.14}
0: /it]
 14%|█▍        | 2030/14725 [1:52:04<11:36:37,  3.29s/it]
                                                         

 14%|█▍        | 2030/14725 [1:52:04<11:36:37,  3.29s/it]
 14%|█▍        | 2031/14725 [1:52:07<11:48:40,  3.35s/it]
 14%|█▍        | 2032/14725 [1:52:10<11:27:45,  3.25s/it]
 14%|█▍        | 2033/14725 [1:52:14<11:42:37,  3.32s/it]
 14%|█▍        | 2034/14725 [1:52:17<11:49:35,  3.35s/it]
 14%|█▍        | 2035/14725 [1:52:21<11:35:57,  3.29s/it]
 14%|█▍        | 2036/14725 [1:52:24<11:30:22,  3.26s/it]
 14%|█▍        | 2037/14725 [1:52:27<11:32:47,  3.28s/it]
 14%|█▍        | 2038/14725 [1:52:30<11:28:16,  3.26s/it]
 14%|█▍        | 2039/14725 [1:52:33<11:18:48,  3.21s/it]
 14%|█▍        | 2040/14725 [1:52:37<11:33:10,  3.28s/it]
                                                         

 14%|█▍        | 2040/14725 [1:52:37<11:33:10,  3.28s/it]
 14%|█▍        | 2041/14725 [1:52:40<11:30:04,  3.26s/it]
 14%|█▍        | 2042/14725 [
0: {'loss': 0.7313, 'grad_norm': 0.45314037173868027, 'learning_rate': 1e-05, 'epoch': 0.14}
0: 1:52:43<11:21:11,  3.22s/it]
 14%|█▍        | 2043/14725 [1:52:46<11:26:05,  3.25s/it]
 14%|█▍        | 2044/14725 [1:52:50<11:24:18,  3.24s/it]
 14%|█▍        | 2045/14725 [1:52:53<11:41:46,  3.32s/it]
 14%|█▍        | 2046/14725 [1:52:56<11:34:23,  3.29s/it]
 14%|█▍        | 2047/14725 [1:53:00<11:41:10,  3.32s/it]
 14%|█▍        | 2048/14725 [1:53:03<11:37:36,  3.30s/it]
 14%|█▍        | 2049/14725 [1:53:06<11:28:58,  3.26s/it]
 14%|█▍        | 2050/14725 [1:53:10<11:45:37,  3.34s/it]
                                                         

 14%|█▍        | 2050/14725 [1:53:10<11:45:37,  3.34s/it]
 14%|█▍        | 2051/14725 [1:53:13<11:38:56,  3.31s/it]
 14%|█▍        | 2052/14725 [1:53:16<11:43:12,  3.33s/it]
 14%|█▍        | 2053/14725 [1:53:20<11:33:52,  3.29s/it]
 14%|█▍        | 2054/14725 [1:53:23<11:28:13,  3.26s/it]
 14%|█▍        | 2055/14725 [1:53:26<11:37:18,  3.30s/it]
 14%|█▍        | 2056/14725 [1:53:29<11:32:19,  3.28s/it]
 14%|�
0: {'loss': 0.7554, 'grad_norm': 0.4962549388419437, 'learning_rate': 1e-05, 'epoch': 0.14}
0: ��▍        | 2057/14725 [1:53:33<11:29:10,  3.26s/it]
 14%|█▍        | 2058/14725 [1:53:36<11:30:52,  3.27s/it]
 14%|█▍        | 2059/14725 [1:53:39<11:44:03,  3.34s/it]
 14%|█▍        | 2060/14725 [1:53:43<11:37:44,  3.31s/it]
                                                         

 14%|█▍        | 2060/14725 [1:53:43<11:37:44,  3.31s/it]
 14%|█▍        | 2061/14725 [1:53:46<11:53:34,  3.38s/it]
 14%|█▍        | 2062/14725 [1:53:49<11:40:40,  3.32s/it]
 14%|█▍        | 2063/14725 [1:53:53<11:37:52,  3.31s/it]
 14%|█▍        | 2064/14725 [1:53:56<11:26:13,  3.25s/it]
 14%|█▍        | 2065/14725 [1:53:59<11:27:25,  3.26s/it]
 14%|█▍        | 2066/14725 [1:54:02<11:21:59,  3.23s/it]
 14%|█▍        | 2067/14725 [1:54:05<11:27:10,  3.26s/it]
 14%|█▍        | 2068/14725 [1:54:09<11:14:43,  3.20s/it]
 14%|█▍        | 2069/14725 [1:54:12<11:26:13,  3.25s/it]
 14%|█▍        | 2070/14725 [1:54:15<11:28:09,  3.26s/it]
                                         
0: {'loss': 0.7419, 'grad_norm': 0.44498075012166527, 'learning_rate': 1e-05, 'epoch': 0.14}
0: {'loss': 0.7462, 'grad_norm': 0.47888921793482825, 'learning_rate': 1e-05, 'epoch': 0.14}
0:                 

 14%|█▍        | 2070/14725 [1:54:15<11:28:09,  3.26s/it]
 14%|█▍        | 2071/14725 [1:54:18<11:25:26,  3.25s/it]
 14%|█▍        | 2072/14725 [1:54:22<11:26:06,  3.25s/it]
 14%|█▍        | 2073/14725 [1:54:25<11:23:54,  3.24s/it]
 14%|█▍        | 2074/14725 [1:54:28<11:30:39,  3.28s/it]
 14%|█▍        | 2075/14725 [1:54:32<11:30:42,  3.28s/it]
 14%|█▍        | 2076/14725 [1:54:35<11:20:16,  3.23s/it]
 14%|█▍        | 2077/14725 [1:54:38<11:25:39,  3.25s/it]
 14%|█▍        | 2078/14725 [1:54:41<11:20:14,  3.23s/it]
 14%|█▍        | 2079/14725 [1:54:44<11:15:19,  3.20s/it]
 14%|█▍        | 2080/14725 [1:54:48<11:18:27,  3.22s/it]
                                                         

 14%|█▍        | 2080/14725 [1:54:48<11:18:27,  3.22s/it]
 14%|█▍        | 2081/14725 [1:54:51<11:25:39,  3.25s/it]
 14%|█▍        | 2082/14725 [1:54:54<11:31:36,  3.28s/it]
 14%|█▍        | 2083/14725 [1:54:57<11:26:24,  3.26s/it]
 14%|█▍      
0: {'loss': 0.7479, 'grad_norm': 0.49667630456040546, 'learning_rate': 1e-05, 'epoch': 0.14}
0:   | 2084/14725 [1:55:01<11:24:56,  3.25s/it]
 14%|█▍        | 2085/14725 [1:55:04<11:20:03,  3.23s/it]
 14%|█▍        | 2086/14725 [1:55:07<11:13:48,  3.20s/it]
 14%|█▍        | 2087/14725 [1:55:10<11:20:21,  3.23s/it]
 14%|█▍        | 2088/14725 [1:55:14<11:35:34,  3.30s/it]
 14%|█▍        | 2089/14725 [1:55:17<11:22:40,  3.24s/it]
 14%|█▍        | 2090/14725 [1:55:20<11:18:16,  3.22s/it]
                                                         

 14%|█▍        | 2090/14725 [1:55:20<11:18:16,  3.22s/it]
 14%|█▍        | 2091/14725 [1:55:23<11:27:51,  3.27s/it]
 14%|█▍        | 2092/14725 [1:55:27<11:21:50,  3.24s/it]
 14%|█▍        | 2093/14725 [1:55:30<11:16:16,  3.21s/it]
 14%|█▍        | 2094/14725 [1:55:33<11:12:09,  3.19s/it]
 14%|█▍        | 2095/14725 [1:55:36<11:20:02,  3.23s/it]
 14%|█▍        | 2096/14725 [1:55:39<11:21:23,  3.24s/it]
 14%|█▍        | 2097/14725 [1:55:43<11:31:51,  3.29s/it]
 14%|█▍        | 2098/14725 [1:55:46<11:29:12,  
0: {'loss': 0.7599, 'grad_norm': 0.45382839061489477, 'learning_rate': 1e-05, 'epoch': 0.14}
0: {'loss': 0.7475, 'grad_norm': 0.48702615375563246, 'learning_rate': 1e-05, 'epoch': 0.14}
0: 3.27s/it]
 14%|█▍        | 2099/14725 [1:55:49<11:20:17,  3.23s/it]
 14%|█▍        | 2100/14725 [1:55:52<11:21:22,  3.24s/it]
                                                         

 14%|█▍        | 2100/14725 [1:55:52<11:21:22,  3.24s/it]
 14%|█▍        | 2101/14725 [1:55:55<11:07:54,  3.17s/it]
 14%|█▍        | 2102/14725 [1:55:59<11:24:38,  3.25s/it]
 14%|█▍        | 2103/14725 [1:56:02<11:20:44,  3.24s/it]
 14%|█▍        | 2104/14725 [1:56:05<11:13:45,  3.20s/it]
 14%|█▍        | 2105/14725 [1:56:08<11:15:39,  3.21s/it]
 14%|█▍        | 2106/14725 [1:56:12<11:14:24,  3.21s/it]
 14%|█▍        | 2107/14725 [1:56:15<11:28:25,  3.27s/it]
 14%|█▍        | 2108/14725 [1:56:18<11:33:58,  3.30s/it]
 14%|█▍        | 2109/14725 [1:56:22<12:07:02,  3.46s/it]
 14%|█▍        | 2110/14725 [1:56:26<11:58:11,  3.42s/it]
                                                         

 14%|█▍        | 2110/14725 [1:56:26<11:58:11,  3.42s/it]
 14%|█▍        | 2111/14
0: {'loss': 0.7347, 'grad_norm': 0.42364597552058303, 'learning_rate': 1e-05, 'epoch': 0.14}
0: 725 [1:56:29<11:41:33,  3.34s/it]
 14%|█▍        | 2112/14725 [1:56:32<11:27:57,  3.27s/it]
 14%|█▍        | 2113/14725 [1:56:35<11:29:34,  3.28s/it]
 14%|█▍        | 2114/14725 [1:56:39<11:31:21,  3.29s/it]
 14%|█▍        | 2115/14725 [1:56:42<11:21:05,  3.24s/it]
 14%|█▍        | 2116/14725 [1:56:45<11:31:43,  3.29s/it]
 14%|█▍        | 2117/14725 [1:56:48<11:19:45,  3.23s/it]
 14%|█▍        | 2118/14725 [1:56:51<11:17:48,  3.23s/it]
 14%|█▍        | 2119/14725 [1:56:55<11:25:39,  3.26s/it]
 14%|█▍        | 2120/14725 [1:56:58<11:22:26,  3.25s/it]
                                                         

 14%|█▍        | 2120/14725 [1:56:58<11:22:26,  3.25s/it]
 14%|█▍        | 2121/14725 [1:57:01<11:34:59,  3.31s/it]
 14%|█▍        | 2122/14725 [1:57:05<11:36:41,  3.32s/it]
 14%|█▍        | 2123/14725 [1:57:08<11:33:45,  3.30s/it]
 14%|█▍        | 2124/14725 [1:57:11<11:24:21,  3.26s/it]
 14%|█▍        | 2125/14725 [1:57:14<11:27:19,  3.27s/it]
 
0: {'loss': 0.7505, 'grad_norm': 0.47879459555007287, 'learning_rate': 1e-05, 'epoch': 0.14}
0: 14%|█▍        | 2126/14725 [1:57:18<11:28:15,  3.28s/it]
 14%|█▍        | 2127/14725 [1:57:21<11:18:59,  3.23s/it]
 14%|█▍        | 2128/14725 [1:57:24<11:33:43,  3.30s/it]
 14%|█▍        | 2129/14725 [1:57:28<11:34:26,  3.31s/it]
 14%|█▍        | 2130/14725 [1:57:31<11:36:09,  3.32s/it]
                                                         

 14%|█▍        | 2130/14725 [1:57:31<11:36:09,  3.32s/it]
 14%|█▍        | 2131/14725 [1:57:34<11:23:49,  3.26s/it]
 14%|█▍        | 2132/14725 [1:57:37<11:31:48,  3.30s/it]
 14%|█▍        | 2133/14725 [1:57:41<11:32:43,  3.30s/it]
 14%|█▍        | 2134/14725 [1:57:44<11:29:37,  3.29s/it]
 14%|█▍        | 2135/14725 [1:57:47<11:31:18,  3.29s/it]
 15%|█▍        | 2136/14725 [1:57:51<11:36:30,  3.32s/it]
 15%|█▍        | 2137/14725 [1:57:54<11:28:19,  3.28s/it]
 15%|█▍        | 2138/14725 [1:57:57<11:23:21,  3.26s/it]
 15%|█▍        | 2139/14725 [1:58:01<11:31:00,  3.29s/it]
 15%|█▍        | 2140/14725 [1:5
0: {'loss': 0.7371, 'grad_norm': 0.4563651362975375, 'learning_rate': 1e-05, 'epoch': 0.15}
0: {'loss': 0.7396, 'grad_norm': 0.43174075954063784, 'learning_rate': 1e-05, 'epoch': 0.15}
0: 8:04<11:24:40,  3.26s/it]
                                                         

 15%|█▍        | 2140/14725 [1:58:04<11:24:40,  3.26s/it]
 15%|█▍        | 2141/14725 [1:58:07<11:21:53,  3.25s/it]
 15%|█▍        | 2142/14725 [1:58:10<11:14:28,  3.22s/it]
 15%|█▍        | 2143/14725 [1:58:13<11:10:33,  3.20s/it]
 15%|█▍        | 2144/14725 [1:58:17<11:19:19,  3.24s/it]
 15%|█▍        | 2145/14725 [1:58:20<11:09:08,  3.19s/it]
 15%|█▍        | 2146/14725 [1:58:23<11:16:52,  3.23s/it]
 15%|█▍        | 2147/14725 [1:58:26<11:26:34,  3.28s/it]
 15%|█▍        | 2148/14725 [1:58:30<11:36:26,  3.32s/it]
 15%|█▍        | 2149/14725 [1:58:33<11:35:42,  3.32s/it]
 15%|█▍        | 2150/14725 [1:58:36<11:33:20,  3.31s/it]
                                                         

 15%|█▍        | 2150/14725 [1:58:36<11:33:20,  3.31s/it]
 15%|█▍        | 2151/14725 [1:58:39<11:20:06,  3.25s/it]
 15%|█▍        | 2152/14725 [1:58:43<11:21:42,  3.25s/it]
 15%|█▍ 
0: {'loss': 0.7416, 'grad_norm': 0.4745213008946426, 'learning_rate': 1e-05, 'epoch': 0.15}
0:        | 2153/14725 [1:58:46<11:21:43,  3.25s/it]
 15%|█▍        | 2154/14725 [1:58:49<11:37:04,  3.33s/it]
 15%|█▍        | 2155/14725 [1:58:53<11:37:15,  3.33s/it]
 15%|█▍        | 2156/14725 [1:58:56<11:28:01,  3.28s/it]
 15%|█▍        | 2157/14725 [1:58:59<11:25:10,  3.27s/it]
 15%|█▍        | 2158/14725 [1:59:02<11:16:54,  3.23s/it]
 15%|█▍        | 2159/14725 [1:59:06<11:29:14,  3.29s/it]
 15%|█▍        | 2160/14725 [1:59:09<11:30:30,  3.30s/it]
                                                         

 15%|█▍        | 2160/14725 [1:59:09<11:30:30,  3.30s/it]
 15%|█▍        | 2161/14725 [1:59:12<11:28:28,  3.29s/it]
 15%|█▍        | 2162/14725 [1:59:16<11:22:44,  3.26s/it]
 15%|█▍        | 2163/14725 [1:59:19<11:30:08,  3.30s/it]
 15%|█▍        | 2164/14725 [1:59:22<11:36:21,  3.33s/it]
 15%|█▍        | 2165/14725 [1:59:26<11:31:28,  3.30s/it]
 15%|█▍        | 2166/14725 [1:59:29<11:32:27,  3.31s/it]
 15%|█▍        | 2167/14725 [1:59:32<11:20:
0: {'loss': 0.7297, 'grad_norm': 0.47764836072079725, 'learning_rate': 1e-05, 'epoch': 0.15}
0: {'loss': 0.7386, 'grad_norm': 0.4484898317309989, 'learning_rate': 1e-05, 'epoch': 0.15}
0: 02,  3.25s/it]
 15%|█▍        | 2168/14725 [1:59:35<11:26:26,  3.28s/it]
 15%|█▍        | 2169/14725 [1:59:39<11:28:02,  3.29s/it]
 15%|█▍        | 2170/14725 [1:59:42<11:30:56,  3.30s/it]
                                                         

 15%|█▍        | 2170/14725 [1:59:42<11:30:56,  3.30s/it]
 15%|█▍        | 2171/14725 [1:59:45<11:20:34,  3.25s/it]
 15%|█▍        | 2172/14725 [1:59:48<11:18:52,  3.24s/it]
 15%|█▍        | 2173/14725 [1:59:52<11:20:06,  3.25s/it]
 15%|█▍        | 2174/14725 [1:59:55<11:18:21,  3.24s/it]
 15%|█▍        | 2175/14725 [1:59:58<11:18:58,  3.25s/it]
 15%|█▍        | 2176/14725 [2:00:01<11:16:47,  3.24s/it]
 15%|█▍        | 2177/14725 [2:00:05<11:14:41,  3.23s/it]
 15%|█▍        | 2178/14725 [2:00:08<11:15:19,  3.23s/it]
 15%|█▍        | 2179/14725 [2:00:11<11:24:07,  3.27s/it]
 15%|█▍        | 2180/14725 [2:00:14<11:20:44,  3.26s/it]
                                                         

 15%|█▍        | 21
0: {'loss': 0.7554, 'grad_norm': 0.4501467228717476, 'learning_rate': 1e-05, 'epoch': 0.15}
0: 80/14725 [2:00:14<11:20:44,  3.26s/it]
 15%|█▍        | 2181/14725 [2:00:17<11:11:33,  3.21s/it]
 15%|█▍        | 2182/14725 [2:00:21<11:14:37,  3.23s/it]
 15%|█▍        | 2183/14725 [2:00:24<11:11:10,  3.21s/it]
 15%|█▍        | 2184/14725 [2:00:27<11:18:35,  3.25s/it]
 15%|█▍        | 2185/14725 [2:00:30<11:13:48,  3.22s/it]
 15%|█▍        | 2186/14725 [2:00:34<11:15:56,  3.23s/it]
 15%|█▍        | 2187/14725 [2:00:37<11:13:37,  3.22s/it]
 15%|█▍        | 2188/14725 [2:00:40<11:16:26,  3.24s/it]
 15%|█▍        | 2189/14725 [2:00:43<11:16:08,  3.24s/it]
 15%|█▍        | 2190/14725 [2:00:47<11:15:39,  3.23s/it]
                                                         

 15%|█▍        | 2190/14725 [2:00:47<11:15:39,  3.23s/it]
 15%|█▍        | 2191/14725 [2:00:50<11:19:28,  3.25s/it]
 15%|█▍        | 2192/14725 [2:00:53<11:19:37,  3.25s/it]
 15%|█▍        | 2193/14725 [2:00:57<11:41:54,  3.36s/it]
 15%|█▍        | 2194/14725 [2:01:00<11:30:28,  3.31s/
0: {'loss': 0.7616, 'grad_norm': 0.4832893350271431, 'learning_rate': 1e-05, 'epoch': 0.15}
0: it]
 15%|█▍        | 2195/14725 [2:01:03<11:23:31,  3.27s/it]
 15%|█▍        | 2196/14725 [2:01:06<11:21:46,  3.26s/it]
 15%|█▍        | 2197/14725 [2:01:10<11:27:00,  3.29s/it]
 15%|█▍        | 2198/14725 [2:01:13<11:23:18,  3.27s/it]
 15%|█▍        | 2199/14725 [2:01:16<11:17:57,  3.25s/it]
 15%|█▍        | 2200/14725 [2:01:20<11:30:01,  3.31s/it]
                                                         

 15%|█▍        | 2200/14725 [2:01:20<11:30:01,  3.31s/it]
 15%|█▍        | 2201/14725 [2:01:23<11:29:34,  3.30s/it]
 15%|█▍        | 2202/14725 [2:01:26<11:15:54,  3.24s/it]
 15%|█▍        | 2203/14725 [2:01:29<11:30:19,  3.31s/it]
 15%|█▍        | 2204/14725 [2:01:33<11:28:51,  3.30s/it]
 15%|█▍        | 2205/14725 [2:01:36<11:23:01,  3.27s/it]
 15%|█▍        | 2206/14725 [2:01:39<11:17:09,  3.25s/it]
 15%|█▍        | 2207/14725 [2:01:42<11:13:03,  3.23s/it]
 15%|█▍        | 2208/14725 [2:01:45<11:08:21,  3.20s/it]
 15%|█▌        | 2209/14725
0: {'loss': 0.7382, 'grad_norm': 0.4528828024255827, 'learning_rate': 1e-05, 'epoch': 0.15}
0: {'loss': 0.7566, 'grad_norm': 0.4733297750650248, 'learning_rate': 1e-05, 'epoch': 0.15}
0:  [2:01:49<11:17:31,  3.25s/it]
 15%|█▌        | 2210/14725 [2:01:52<11:09:10,  3.21s/it]
                                                         

 15%|█▌        | 2210/14725 [2:01:52<11:09:10,  3.21s/it]
 15%|█▌        | 2211/14725 [2:01:55<11:01:02,  3.17s/it]
 15%|█▌        | 2212/14725 [2:01:58<11:15:06,  3.24s/it]
 15%|█▌        | 2213/14725 [2:02:02<11:20:27,  3.26s/it]
 15%|█▌        | 2214/14725 [2:02:05<11:07:59,  3.20s/it]
 15%|█▌        | 2215/14725 [2:02:08<11:02:50,  3.18s/it]
 15%|█▌        | 2216/14725 [2:02:11<10:58:55,  3.16s/it]
 15%|█▌        | 2217/14725 [2:02:14<10:55:50,  3.15s/it]
 15%|█▌        | 2218/14725 [2:02:18<11:12:25,  3.23s/it]
 15%|█▌        | 2219/14725 [2:02:21<11:08:13,  3.21s/it]
 15%|█▌        | 2220/14725 [2:02:24<11:08:34,  3.21s/it]
                                                         

 15%|█▌        | 2220/14725 [2:02:24<11:08:34,  3.21s/it]
 15%|█▌        | 2221/14725 [2:02:27<11:08:15,  3.21s/it]
 15%|�
0: {'loss': 0.7446, 'grad_norm': 0.49808076350089187, 'learning_rate': 1e-05, 'epoch': 0.15}
0: �▌        | 2222/14725 [2:02:30<11:08:55,  3.21s/it]
 15%|█▌        | 2223/14725 [2:02:34<11:24:48,  3.29s/it]
 15%|█▌        | 2224/14725 [2:02:37<11:17:22,  3.25s/it]
 15%|█▌        | 2225/14725 [2:02:40<11:25:21,  3.29s/it]
 15%|█▌        | 2226/14725 [2:02:43<11:12:19,  3.23s/it]
 15%|█▌        | 2227/14725 [2:02:47<11:16:12,  3.25s/it]
 15%|█▌        | 2228/14725 [2:02:50<11:14:07,  3.24s/it]
 15%|█▌        | 2229/14725 [2:02:53<11:12:03,  3.23s/it]
 15%|█▌        | 2230/14725 [2:02:56<11:08:24,  3.21s/it]
                                                         

 15%|█▌        | 2230/14725 [2:02:56<11:08:24,  3.21s/it]
 15%|█▌        | 2231/14725 [2:03:00<11:13:14,  3.23s/it]
 15%|█▌        | 2232/14725 [2:03:03<11:05:20,  3.20s/it]
 15%|█▌        | 2233/14725 [2:03:06<11:07:24,  3.21s/it]
 15%|█▌        | 2234/14725 [2:03:09<11:16:44,  3.25s/it]
 15%|█▌        | 2235/14725 [2:03:12<11:12:29,  3.23s/it]
 15%|█▌        | 2236/14725 [2:03:16<1
0: {'loss': 0.7404, 'grad_norm': 0.45177064751297685, 'learning_rate': 1e-05, 'epoch': 0.15}
0: 1:02:52,  3.18s/it]
 15%|█▌        | 2237/14725 [2:03:19<11:03:34,  3.19s/it]
 15%|█▌        | 2238/14725 [2:03:22<11:06:34,  3.20s/it]
 15%|█▌        | 2239/14725 [2:03:25<11:04:00,  3.19s/it]
 15%|█▌        | 2240/14725 [2:03:29<11:25:31,  3.29s/it]
                                                         

 15%|█▌        | 2240/14725 [2:03:29<11:25:31,  3.29s/it]
 15%|█▌        | 2241/14725 [2:03:32<11:30:36,  3.32s/it]
 15%|█▌        | 2242/14725 [2:03:35<11:34:54,  3.34s/it]
 15%|█▌        | 2243/14725 [2:03:39<11:26:48,  3.30s/it]
 15%|█▌        | 2244/14725 [2:03:42<11:27:28,  3.30s/it]
 15%|█▌        | 2245/14725 [2:03:45<11:11:51,  3.23s/it]
 15%|█▌        | 2246/14725 [2:03:48<11:16:38,  3.25s/it]
 15%|█▌        | 2247/14725 [2:03:52<11:24:30,  3.29s/it]
 15%|█▌        | 2248/14725 [2:03:55<11:19:20,  3.27s/it]
 15%|█▌        | 2249/14725 [2:03:58<11:20:04,  3.27s/it]
 15%|█▌        | 2250/14725 [2:04:02<11:25:20,  3.30s/it]
               
0: {'loss': 0.7496, 'grad_norm': 0.46751009692796774, 'learning_rate': 1e-05, 'epoch': 0.15}
0: {'loss': 0.7531, 'grad_norm': 0.4392023420366416, 'learning_rate': 1e-05, 'epoch': 0.15}
0:                                           

 15%|█▌        | 2250/14725 [2:04:02<11:25:20,  3.30s/it]
 15%|█▌        | 2251/14725 [2:04:05<11:25:33,  3.30s/it]
 15%|█▌        | 2252/14725 [2:04:08<11:23:46,  3.29s/it]
 15%|█▌        | 2253/14725 [2:04:11<11:15:17,  3.25s/it]
 15%|█▌        | 2254/14725 [2:04:14<11:09:09,  3.22s/it]
 15%|█▌        | 2255/14725 [2:04:18<11:18:14,  3.26s/it]
 15%|█▌        | 2256/14725 [2:04:21<11:18:12,  3.26s/it]
 15%|█▌        | 2257/14725 [2:04:24<11:11:55,  3.23s/it]
 15%|█▌        | 2258/14725 [2:04:28<11:17:30,  3.26s/it]
 15%|█▌        | 2259/14725 [2:04:31<11:18:56,  3.27s/it]
 15%|█▌        | 2260/14725 [2:04:34<11:09:29,  3.22s/it]
                                                         

 15%|█▌        | 2260/14725 [2:04:34<11:09:29,  3.22s/it]
 15%|█▌        | 2261/14725 [2:04:37<11:20:49,  3.28s/it]
 15%|█▌        | 2262/14725 [2:04:41<11:29:56,  3.32s/it]
 15%|█▌        | 2263/14725 [2:04:44<11:26:22,  3
0: {'loss': 0.7457, 'grad_norm': 0.446550646918874, 'learning_rate': 1e-05, 'epoch': 0.15}
0: .30s/it]
 15%|█▌        | 2264/14725 [2:04:47<11:20:14,  3.28s/it]
 15%|█▌        | 2265/14725 [2:04:51<11:31:08,  3.33s/it]
 15%|█▌        | 2266/14725 [2:04:54<11:38:14,  3.36s/it]
 15%|█▌        | 2267/14725 [2:04:57<11:33:59,  3.34s/it]
 15%|█▌        | 2268/14725 [2:05:01<11:36:08,  3.35s/it]
 15%|█▌        | 2269/14725 [2:05:04<11:23:52,  3.29s/it]
 15%|█▌        | 2270/14725 [2:05:07<11:25:42,  3.30s/it]
                                                         

 15%|█▌        | 2270/14725 [2:05:07<11:25:42,  3.30s/it]
 15%|█▌        | 2271/14725 [2:05:11<11:35:24,  3.35s/it]
 15%|█▌        | 2272/14725 [2:05:14<11:38:11,  3.36s/it]
 15%|█▌        | 2273/14725 [2:05:17<11:17:20,  3.26s/it]
 15%|█▌        | 2274/14725 [2:05:21<11:25:04,  3.30s/it]
 15%|█▌        | 2275/14725 [2:05:24<11:56:02,  3.45s/it]
 15%|█▌        | 2276/14725 [2:05:28<11:35:17,  3.35s/it]
 15%|█▌        | 2277/14725 [2:05:31<11:20:42,  3.28s/it]
 15%|█▌        | 2278/
0: {'loss': 0.7233, 'grad_norm': 0.46224868995837587, 'learning_rate': 1e-05, 'epoch': 0.15}
0: {'loss': 0.7422, 'grad_norm': 0.4695495306309479, 'learning_rate': 1e-05, 'epoch': 0.16}
0: 14725 [2:05:34<11:24:29,  3.30s/it]
 15%|█▌        | 2279/14725 [2:05:37<11:09:41,  3.23s/it]
 15%|█▌        | 2280/14725 [2:05:40<11:19:04,  3.27s/it]
                                                         

 15%|█▌        | 2280/14725 [2:05:40<11:19:04,  3.27s/it]
 15%|█▌        | 2281/14725 [2:05:44<11:20:46,  3.28s/it]
 15%|█▌        | 2282/14725 [2:05:47<11:18:51,  3.27s/it]
 16%|█▌        | 2283/14725 [2:05:50<11:10:27,  3.23s/it]
 16%|█▌        | 2284/14725 [2:05:53<11:09:16,  3.23s/it]
 16%|█▌        | 2285/14725 [2:05:57<11:06:26,  3.21s/it]
 16%|█▌        | 2286/14725 [2:06:00<11:10:37,  3.23s/it]
 16%|█▌        | 2287/14725 [2:06:03<11:21:04,  3.29s/it]
 16%|█▌        | 2288/14725 [2:06:07<11:43:10,  3.39s/it]
 16%|█▌        | 2289/14725 [2:06:10<11:28:08,  3.32s/it]
 16%|█▌        | 2290/14725 [2:06:13<11:24:22,  3.30s/it]
                                                         

 16%|█▌        | 2290/14725 [2:06:13<11:24:22,  3.30s/it]
 1
0: {'loss': 0.7555, 'grad_norm': 0.4233104868116468, 'learning_rate': 1e-05, 'epoch': 0.16}
0: 6%|█▌        | 2291/14725 [2:06:16<11:19:53,  3.28s/it]
 16%|█▌        | 2292/14725 [2:06:20<11:18:30,  3.27s/it]
 16%|█▌        | 2293/14725 [2:06:23<11:21:44,  3.29s/it]
 16%|█▌        | 2294/14725 [2:06:26<11:14:40,  3.26s/it]
 16%|█▌        | 2295/14725 [2:06:30<11:22:00,  3.29s/it]
 16%|█▌        | 2296/14725 [2:06:33<11:13:08,  3.25s/it]
 16%|█▌        | 2297/14725 [2:06:36<11:04:02,  3.21s/it]
 16%|█▌        | 2298/14725 [2:06:39<11:00:50,  3.19s/it]
 16%|█▌        | 2299/14725 [2:06:42<11:02:07,  3.20s/it]
 16%|█▌        | 2300/14725 [2:06:45<11:04:38,  3.21s/it]
                                                         

 16%|█▌        | 2300/14725 [2:06:45<11:04:38,  3.21s/it]
 16%|█▌        | 2301/14725 [2:06:49<10:54:17,  3.16s/it]
 16%|█▌        | 2302/14725 [2:06:52<11:06:23,  3.22s/it]
 16%|█▌        | 2303/14725 [2:06:55<11:01:24,  3.19s/it]
 16%|█▌        | 2304/14725 [2:06:58<11:13:13,  3.25s/it]
 16%|█▌        | 2305/14725 [2:07
0: {'loss': 0.7396, 'grad_norm': 0.4600921752204951, 'learning_rate': 1e-05, 'epoch': 0.16}
0: :02<11:27:37,  3.32s/it]
 16%|█▌        | 2306/14725 [2:07:05<11:21:33,  3.29s/it]
 16%|█▌        | 2307/14725 [2:07:08<11:16:47,  3.27s/it]
 16%|█▌        | 2308/14725 [2:07:11<11:06:35,  3.22s/it]
 16%|█▌        | 2309/14725 [2:07:15<11:25:00,  3.31s/it]
 16%|█▌        | 2310/14725 [2:07:18<11:18:40,  3.28s/it]
                                                         

 16%|█▌        | 2310/14725 [2:07:18<11:18:40,  3.28s/it]
 16%|█▌        | 2311/14725 [2:07:21<11:13:04,  3.25s/it]
 16%|█▌        | 2312/14725 [2:07:25<11:07:58,  3.23s/it]
 16%|█▌        | 2313/14725 [2:07:28<11:03:37,  3.21s/it]
 16%|█▌        | 2314/14725 [2:07:31<11:02:02,  3.20s/it]
 16%|█▌        | 2315/14725 [2:07:34<11:00:55,  3.20s/it]
 16%|█▌        | 2316/14725 [2:07:37<11:02:55,  3.21s/it]
 16%|█▌        | 2317/14725 [2:07:41<11:23:13,  3.30s/it]
 16%|█▌        | 2318/14725 [2:07:44<11:14:00,  3.26s/it]
 16%|█▌        | 2319/14725 [2:07:47<11:09:05,  3.24s/it]
 16%|█�
0: {'loss': 0.7445, 'grad_norm': 0.4620442685907117, 'learning_rate': 1e-05, 'epoch': 0.16}
0: {'loss': 0.7403, 'grad_norm': 0.4280011192031464, 'learning_rate': 1e-05, 'epoch': 0.16}
0: �        | 2320/14725 [2:07:51<11:32:09,  3.35s/it]
                                                         

 16%|█▌        | 2320/14725 [2:07:51<11:32:09,  3.35s/it]
 16%|█▌        | 2321/14725 [2:07:54<11:28:44,  3.33s/it]
 16%|█▌        | 2322/14725 [2:07:57<11:23:26,  3.31s/it]
 16%|█▌        | 2323/14725 [2:08:00<11:09:23,  3.24s/it]
 16%|█▌        | 2324/14725 [2:08:04<11:23:29,  3.31s/it]
 16%|█▌        | 2325/14725 [2:08:07<11:19:41,  3.29s/it]
 16%|█▌        | 2326/14725 [2:08:10<11:24:40,  3.31s/it]
 16%|█▌        | 2327/14725 [2:08:14<11:16:06,  3.27s/it]
 16%|█▌        | 2328/14725 [2:08:17<11:16:54,  3.28s/it]
 16%|█▌        | 2329/14725 [2:08:20<11:07:44,  3.23s/it]
 16%|█▌        | 2330/14725 [2:08:23<11:12:50,  3.26s/it]
                                                         

 16%|█▌        | 2330/14725 [2:08:23<11:12:50,  3.26s/it]
 16%|█▌        | 2331/14725 [2:08:27<11:08:48,  3.24s/it]
 16%|█▌        | 2332/14725 [2:08:30<11:08:1
0: {'loss': 0.7449, 'grad_norm': 0.46707862346655393, 'learning_rate': 1e-05, 'epoch': 0.16}
0: 8,  3.24s/it]
 16%|█▌        | 2333/14725 [2:08:33<11:11:53,  3.25s/it]
 16%|█▌        | 2334/14725 [2:08:37<11:28:13,  3.33s/it]
 16%|█▌        | 2335/14725 [2:08:40<11:21:09,  3.30s/it]
 16%|█▌        | 2336/14725 [2:08:43<11:19:35,  3.29s/it]
 16%|█▌        | 2337/14725 [2:08:46<11:15:09,  3.27s/it]
 16%|█▌        | 2338/14725 [2:08:50<11:18:14,  3.29s/it]
 16%|█▌        | 2339/14725 [2:08:53<11:07:09,  3.23s/it]
 16%|█▌        | 2340/14725 [2:08:56<11:14:30,  3.27s/it]
                                                         

 16%|█▌        | 2340/14725 [2:08:56<11:14:30,  3.27s/it]
 16%|█▌        | 2341/14725 [2:08:59<11:04:20,  3.22s/it]
 16%|█▌        | 2342/14725 [2:09:02<11:07:47,  3.24s/it]
 16%|█▌        | 2343/14725 [2:09:06<10:59:40,  3.20s/it]
 16%|█▌        | 2344/14725 [2:09:09<11:10:11,  3.25s/it]
 16%|█▌        | 2345/14725 [2:09:13<11:31:06,  3.35s/it]
 16%|█▌        | 2346/14725 [2:09:16<11:24:58,  3.32s/it]
 16%|█▌        | 
0: {'loss': 0.7267, 'grad_norm': 0.4303940179674537, 'learning_rate': 1e-05, 'epoch': 0.16}
0: 2347/14725 [2:09:19<11:15:15,  3.27s/it]
 16%|█▌        | 2348/14725 [2:09:22<11:01:14,  3.21s/it]
 16%|█▌        | 2349/14725 [2:09:25<11:00:20,  3.20s/it]
 16%|█▌        | 2350/14725 [2:09:29<11:09:28,  3.25s/it]
                                                         

 16%|█▌        | 2350/14725 [2:09:29<11:09:28,  3.25s/it]
 16%|█▌        | 2351/14725 [2:09:32<11:00:07,  3.20s/it]
 16%|█▌        | 2352/14725 [2:09:35<10:50:45,  3.16s/it]
 16%|█▌        | 2353/14725 [2:09:38<10:48:01,  3.14s/it]
 16%|█▌        | 2354/14725 [2:09:41<11:06:57,  3.23s/it]
 16%|█▌        | 2355/14725 [2:09:44<10:56:42,  3.19s/it]
 16%|█▌        | 2356/14725 [2:09:48<11:02:57,  3.22s/it]
 16%|█▌        | 2357/14725 [2:09:51<11:06:32,  3.23s/it]
 16%|█▌        | 2358/14725 [2:09:54<11:12:58,  3.27s/it]
 16%|█▌        | 2359/14725 [2:09:57<11:09:17,  3.25s/it]
 16%|█▌        | 2360/14725 [2:10:01<11:15:46,  3.28s/it]
                                                        
0: {'loss': 0.7518, 'grad_norm': 0.4503388824657424, 'learning_rate': 1e-05, 'epoch': 0.16}
0: {'loss': 0.7296, 'grad_norm': 0.4515589148691253, 'learning_rate': 1e-05, 'epoch': 0.16}
0:  

 16%|█▌        | 2360/14725 [2:10:01<11:15:46,  3.28s/it]
 16%|█▌        | 2361/14725 [2:10:04<11:25:16,  3.33s/it]
 16%|█▌        | 2362/14725 [2:10:07<11:16:25,  3.28s/it]
 16%|█▌        | 2363/14725 [2:10:11<11:10:37,  3.25s/it]
 16%|█▌        | 2364/14725 [2:10:14<11:08:44,  3.25s/it]
 16%|█▌        | 2365/14725 [2:10:17<10:54:32,  3.18s/it]
 16%|█▌        | 2366/14725 [2:10:20<10:52:52,  3.17s/it]
 16%|█▌        | 2367/14725 [2:10:23<10:52:58,  3.17s/it]
 16%|█▌        | 2368/14725 [2:10:26<10:59:06,  3.20s/it]
 16%|█▌        | 2369/14725 [2:10:30<11:01:20,  3.21s/it]
 16%|█▌        | 2370/14725 [2:10:33<11:14:19,  3.27s/it]
                                                         

 16%|█▌        | 2370/14725 [2:10:33<11:14:19,  3.27s/it]
 16%|█▌        | 2371/14725 [2:10:37<11:25:47,  3.33s/it]
 16%|█▌        | 2372/14725 [2:10:40<11:20:09,  3.30s/it]
 16%|█▌        | 2373/14725 [2:10:43<11:19:50,  3.30s/it]
 16%|█▌        | 2374/14725 
0: {'loss': 0.7501, 'grad_norm': 0.46072590703330346, 'learning_rate': 1e-05, 'epoch': 0.16}
0: [2:10:46<11:21:48,  3.31s/it]
 16%|█▌        | 2375/14725 [2:10:50<11:17:27,  3.29s/it]
 16%|█▌        | 2376/14725 [2:10:53<11:16:39,  3.29s/it]
 16%|█▌        | 2377/14725 [2:10:56<11:23:16,  3.32s/it]
 16%|█▌        | 2378/14725 [2:11:00<11:29:56,  3.35s/it]
 16%|█▌        | 2379/14725 [2:11:03<11:23:16,  3.32s/it]
 16%|█▌        | 2380/14725 [2:11:06<11:15:29,  3.28s/it]
                                                         

 16%|█▌        | 2380/14725 [2:11:06<11:15:29,  3.28s/it]
 16%|█▌        | 2381/14725 [2:11:10<11:16:55,  3.29s/it]
 16%|█▌        | 2382/14725 [2:11:13<11:09:43,  3.26s/it]
 16%|█▌        | 2383/14725 [2:11:16<11:26:09,  3.34s/it]
 16%|█▌        | 2384/14725 [2:11:19<11:16:27,  3.29s/it]
 16%|█▌        | 2385/14725 [2:11:23<11:42:33,  3.42s/it]
 16%|█▌        | 2386/14725 [2:11:26<11:23:38,  3.32s/it]
 16%|█▌        | 2387/14725 [2:11:29<11:19:05,  3.30s/it]
 16%|█▌        | 2388/14725 [2:11:33<11:11:16,  3.26s/it]
 16%|
0: {'loss': 0.7456, 'grad_norm': 0.4626995103863285, 'learning_rate': 1e-05, 'epoch': 0.16}
0: {'loss': 0.7526, 'grad_norm': 0.493964457591847, 'learning_rate': 1e-05, 'epoch': 0.16}
0: █▌        | 2389/14725 [2:11:36<11:11:39,  3.27s/it]
 16%|█▌        | 2390/14725 [2:11:39<11:08:10,  3.25s/it]
                                                         

 16%|█▌        | 2390/14725 [2:11:39<11:08:10,  3.25s/it]
 16%|█▌        | 2391/14725 [2:11:42<11:08:58,  3.25s/it]
 16%|█▌        | 2392/14725 [2:11:46<11:22:08,  3.32s/it]
 16%|█▋        | 2393/14725 [2:11:49<11:28:41,  3.35s/it]
 16%|█▋        | 2394/14725 [2:11:53<11:32:06,  3.37s/it]
 16%|█▋        | 2395/14725 [2:11:56<11:29:59,  3.36s/it]
 16%|█▋        | 2396/14725 [2:11:59<11:27:18,  3.34s/it]
 16%|█▋        | 2397/14725 [2:12:03<11:29:14,  3.35s/it]
 16%|█▋        | 2398/14725 [2:12:06<11:42:16,  3.42s/it]
 16%|█▋        | 2399/14725 [2:12:10<11:41:33,  3.42s/it]
 16%|█▋        | 2400/14725 [2:12:13<12:01:58,  3.51s/it]
                                                         

 16%|█▋        | 2400/14725 [2:12:13<12:01:58,  3.51s/it]
 16%|█▋        | 2401/14725 [2:12:17<11
0: {'loss': 0.75, 'grad_norm': 0.4534300418883751, 'learning_rate': 1e-05, 'epoch': 0.16}
0: :55:53,  3.49s/it]
 16%|█▋        | 2402/14725 [2:12:20<11:54:51,  3.48s/it]
 16%|█▋        | 2403/14725 [2:12:24<11:35:43,  3.39s/it]
 16%|█▋        | 2404/14725 [2:12:27<11:24:57,  3.34s/it]
 16%|█▋        | 2405/14725 [2:12:30<11:24:10,  3.33s/it]
 16%|█▋        | 2406/14725 [2:12:33<11:29:12,  3.36s/it]
 16%|█▋        | 2407/14725 [2:12:37<11:34:47,  3.38s/it]
 16%|█▋        | 2408/14725 [2:12:40<11:21:46,  3.32s/it]
 16%|█▋        | 2409/14725 [2:12:44<11:50:00,  3.46s/it]
 16%|█▋        | 2410/14725 [2:12:47<11:27:36,  3.35s/it]
                                                         

 16%|█▋        | 2410/14725 [2:12:47<11:27:36,  3.35s/it]
 16%|█▋        | 2411/14725 [2:12:50<11:14:07,  3.28s/it]
 16%|█▋        | 2412/14725 [2:12:53<11:17:05,  3.30s/it]
 16%|█▋        | 2413/14725 [2:12:57<11:13:26,  3.28s/it]
 16%|█▋        | 2414/14725 [2:13:00<11:19:07,  3.31s/it]
 16%|█▋        | 2415/14725 [2:13:03<11:16:41,  3.30s/it]
 16%|█▋     
0: {'loss': 0.738, 'grad_norm': 0.5029465061671116, 'learning_rate': 1e-05, 'epoch': 0.16}
0:    | 2416/14725 [2:13:06<11:02:22,  3.23s/it]
 16%|█▋        | 2417/14725 [2:13:10<11:03:11,  3.23s/it]
 16%|█▋        | 2418/14725 [2:13:13<11:06:32,  3.25s/it]
 16%|█▋        | 2419/14725 [2:13:16<11:06:15,  3.25s/it]
 16%|█▋        | 2420/14725 [2:13:19<11:11:25,  3.27s/it]
                                                         

 16%|█▋        | 2420/14725 [2:13:19<11:11:25,  3.27s/it]
 16%|█▋        | 2421/14725 [2:13:23<11:04:13,  3.24s/it]
 16%|█▋        | 2422/14725 [2:13:26<11:09:26,  3.26s/it]
 16%|█▋        | 2423/14725 [2:13:29<11:18:55,  3.31s/it]
 16%|█▋        | 2424/14725 [2:13:33<11:09:57,  3.27s/it]
 16%|█▋        | 2425/14725 [2:13:36<11:05:23,  3.25s/it]
 16%|█▋        | 2426/14725 [2:13:39<11:07:46,  3.26s/it]
 16%|█▋        | 2427/14725 [2:13:42<11:10:29,  3.27s/it]
 16%|█▋        | 2428/14725 [2:13:46<11:07:47,  3.26s/it]
 16%|█▋        | 2429/14725 [2:13:49<11:21:20,  3.32s/it]
 17%|█▋        | 2430/14725 [2:13:52<11:10:21, 
0: {'loss': 0.7424, 'grad_norm': 0.4915321706289736, 'learning_rate': 1e-05, 'epoch': 0.17}
0: {'loss': 0.7408, 'grad_norm': 0.46840667545822456, 'learning_rate': 1e-05, 'epoch': 0.17}
0:  3.27s/it]
                                                         

 17%|█▋        | 2430/14725 [2:13:52<11:10:21,  3.27s/it]
 17%|█▋        | 2431/14725 [2:13:56<11:14:53,  3.29s/it]
 17%|█▋        | 2432/14725 [2:13:59<11:05:24,  3.25s/it]
 17%|█▋        | 2433/14725 [2:14:02<11:03:58,  3.24s/it]
 17%|█▋        | 2434/14725 [2:14:05<11:03:40,  3.24s/it]
 17%|█▋        | 2435/14725 [2:14:08<10:58:30,  3.21s/it]
 17%|█▋        | 2436/14725 [2:14:11<10:55:02,  3.20s/it]
 17%|█▋        | 2437/14725 [2:14:15<11:09:02,  3.27s/it]
 17%|█▋        | 2438/14725 [2:14:18<11:00:57,  3.23s/it]
 17%|█▋        | 2439/14725 [2:14:21<11:13:20,  3.29s/it]
 17%|█▋        | 2440/14725 [2:14:25<11:07:37,  3.26s/it]
                                                         

 17%|█▋        | 2440/14725 [2:14:25<11:07:37,  3.26s/it]
 17%|█▋        | 2441/14725 [2:14:28<11:14:20,  3.29s/it]
 17%|█▋        | 2442/14725 [2:14:31<11:06:13,  3.25s/it]
 17%|█▋        | 2443/1
0: {'loss': 0.7506, 'grad_norm': 0.48931779041183365, 'learning_rate': 1e-05, 'epoch': 0.17}
0: 4725 [2:14:35<11:13:17,  3.29s/it]
 17%|█▋        | 2444/14725 [2:14:38<11:10:53,  3.28s/it]
 17%|█▋        | 2445/14725 [2:14:41<11:01:57,  3.23s/it]
 17%|█▋        | 2446/14725 [2:14:44<10:59:08,  3.22s/it]
 17%|█▋        | 2447/14725 [2:14:47<10:55:40,  3.20s/it]
 17%|█▋        | 2448/14725 [2:14:51<11:09:59,  3.27s/it]
 17%|█▋        | 2449/14725 [2:14:54<11:05:29,  3.25s/it]
 17%|█▋        | 2450/14725 [2:14:57<11:01:37,  3.23s/it]
                                                         

 17%|█▋        | 2450/14725 [2:14:57<11:01:37,  3.23s/it]
 17%|█▋        | 2451/14725 [2:15:00<11:07:46,  3.26s/it]
 17%|█▋        | 2452/14725 [2:15:04<11:16:35,  3.31s/it]
 17%|█▋        | 2453/14725 [2:15:07<11:10:00,  3.28s/it]
 17%|█▋        | 2454/14725 [2:15:10<11:05:06,  3.25s/it]
 17%|█▋        | 2455/14725 [2:15:13<11:02:37,  3.24s/it]
 17%|█▋        | 2456/14725 [2:15:17<10:57:35,  3.22s/it]
 17%|█▋        | 2457/14725 [2:15:20<10:50:43,  3.18s/it]
0: {'loss': 0.7346, 'grad_norm': 0.4262255639088451, 'learning_rate': 1e-05, 'epoch': 0.17}
0: {'loss': 0.749, 'grad_norm': 0.44957234893464926, 'learning_rate': 1e-05, 'epoch': 0.17}
0:  17%|█▋        | 2458/14725 [2:15:23<10:54:41,  3.20s/it]
 17%|█▋        | 2459/14725 [2:15:26<10:48:16,  3.17s/it]
 17%|█▋        | 2460/14725 [2:15:29<10:49:33,  3.18s/it]
                                                         

 17%|█▋        | 2460/14725 [2:15:29<10:49:33,  3.18s/it]
 17%|█▋        | 2461/14725 [2:15:33<10:56:36,  3.21s/it]
 17%|█▋        | 2462/14725 [2:15:36<10:56:45,  3.21s/it]
 17%|█▋        | 2463/14725 [2:15:39<10:58:45,  3.22s/it]
 17%|█▋        | 2464/14725 [2:15:42<10:47:29,  3.17s/it]
 17%|█▋        | 2465/14725 [2:15:45<10:53:04,  3.20s/it]
 17%|█▋        | 2466/14725 [2:15:49<10:52:55,  3.20s/it]
 17%|█▋        | 2467/14725 [2:15:52<10:56:39,  3.21s/it]
 17%|█▋        | 2468/14725 [2:15:55<10:58:11,  3.22s/it]
 17%|█▋        | 2469/14725 [2:15:58<10:59:32,  3.23s/it]
 17%|█▋        | 2470/14725 [2:16:01<10:55:15,  3.21s/it]
                                                         

 17%|█▋        | 2470/14725 [2:16:
0: {'loss': 0.7369, 'grad_norm': 0.42723271592928896, 'learning_rate': 1e-05, 'epoch': 0.17}
0: 01<10:55:15,  3.21s/it]
 17%|█▋        | 2471/14725 [2:16:05<10:50:50,  3.19s/it]
 17%|█▋        | 2472/14725 [2:16:08<10:48:10,  3.17s/it]
 17%|█▋        | 2473/14725 [2:16:11<10:55:26,  3.21s/it]
 17%|█▋        | 2474/14725 [2:16:14<10:58:35,  3.23s/it]
 17%|█▋        | 2475/14725 [2:16:18<11:05:00,  3.26s/it]
 17%|█▋        | 2476/14725 [2:16:21<11:00:42,  3.24s/it]
 17%|█▋        | 2477/14725 [2:16:24<10:52:21,  3.20s/it]
 17%|█▋        | 2478/14725 [2:16:27<11:05:17,  3.26s/it]
 17%|█▋        | 2479/14725 [2:16:31<11:03:15,  3.25s/it]
 17%|█▋        | 2480/14725 [2:16:34<10:54:30,  3.21s/it]
                                                         

 17%|█▋        | 2480/14725 [2:16:34<10:54:30,  3.21s/it]
 17%|█▋        | 2481/14725 [2:16:37<11:00:22,  3.24s/it]
 17%|█▋        | 2482/14725 [2:16:40<10:48:45,  3.18s/it]
 17%|█▋        | 2483/14725 [2:16:43<10:59:33,  3.23s/it]
 17%|█▋        | 2484/14725 [2:16:47<10:59:18,  3.23s/it]
 17%|█▋
0: {'loss': 0.7303, 'grad_norm': 0.4427417553657665, 'learning_rate': 1e-05, 'epoch': 0.17}
0:         | 2485/14725 [2:16:50<10:55:09,  3.21s/it]
 17%|█▋        | 2486/14725 [2:16:53<10:56:00,  3.22s/it]
 17%|█▋        | 2487/14725 [2:16:56<10:56:43,  3.22s/it]
 17%|█▋        | 2488/14725 [2:17:00<11:02:59,  3.25s/it]
 17%|█▋        | 2489/14725 [2:17:03<10:58:36,  3.23s/it]
 17%|█▋        | 2490/14725 [2:17:06<10:54:08,  3.21s/it]
                                                         

 17%|█▋        | 2490/14725 [2:17:06<10:54:08,  3.21s/it]
 17%|█▋        | 2491/14725 [2:17:09<11:09:08,  3.28s/it]
 17%|█▋        | 2492/14725 [2:17:13<11:08:53,  3.28s/it]
 17%|█▋        | 2493/14725 [2:17:16<10:59:59,  3.24s/it]
 17%|█▋        | 2494/14725 [2:17:19<11:07:05,  3.27s/it]
 17%|█▋        | 2495/14725 [2:17:22<11:00:48,  3.24s/it]
 17%|█▋        | 2496/14725 [2:17:26<11:10:46,  3.29s/it]
 17%|█▋        | 2497/14725 [2:17:29<11:08:38,  3.28s/it]
 17%|█▋        | 2498/14725 [2:17:32<11:02:09,  3.25s/it]
 17%|█▋        | 2499/14725 [2:17:35<11:04
0: {'loss': 0.7397, 'grad_norm': 0.4310773045217711, 'learning_rate': 1e-05, 'epoch': 0.17}
0: {'loss': 0.7323, 'grad_norm': 0.4679520997280277, 'learning_rate': 1e-05, 'epoch': 0.17}
0: :59,  3.26s/it]
 17%|█▋        | 2500/14725 [2:17:39<11:05:05,  3.26s/it]
                                                         

 17%|█▋        | 2500/14725 [2:17:39<11:05:05,  3.26s/it]
 17%|█▋        | 2501/14725 [2:17:42<10:50:12,  3.19s/it]
 17%|█▋        | 2502/14725 [2:17:45<10:58:36,  3.23s/it]
 17%|█▋        | 2503/14725 [2:17:48<10:51:42,  3.20s/it]
 17%|█▋        | 2504/14725 [2:17:51<10:48:36,  3.18s/it]
 17%|█▋        | 2505/14725 [2:17:54<10:45:47,  3.17s/it]
 17%|█▋        | 2506/14725 [2:17:58<10:52:48,  3.21s/it]
 17%|█▋        | 2507/14725 [2:18:01<11:04:46,  3.26s/it]
 17%|█▋        | 2508/14725 [2:18:04<11:01:51,  3.25s/it]
 17%|█▋        | 2509/14725 [2:18:07<10:57:08,  3.23s/it]
 17%|█▋        | 2510/14725 [2:18:11<10:49:37,  3.19s/it]
                                                         

 17%|█▋        | 2510/14725 [2:18:11<10:49:37,  3.19s/it]
 17%|█▋        | 2511/14725 [2:18:14<10:51:12,  3.20s/it]
 17%|█▋        | 2
0: {'loss': 0.7559, 'grad_norm': 0.44466076937114524, 'learning_rate': 1e-05, 'epoch': 0.17}
0: 512/14725 [2:18:17<11:17:59,  3.33s/it]
 17%|█▋        | 2513/14725 [2:18:21<11:07:53,  3.28s/it]
 17%|█▋        | 2514/14725 [2:18:24<11:14:34,  3.31s/it]
 17%|█▋        | 2515/14725 [2:18:27<10:55:27,  3.22s/it]
 17%|█▋        | 2516/14725 [2:18:30<10:53:31,  3.21s/it]
 17%|█▋        | 2517/14725 [2:18:33<10:58:12,  3.23s/it]
 17%|█▋        | 2518/14725 [2:18:37<11:05:46,  3.27s/it]
 17%|█▋        | 2519/14725 [2:18:40<11:14:00,  3.31s/it]
 17%|█▋        | 2520/14725 [2:18:44<11:22:11,  3.35s/it]
                                                         

 17%|█▋        | 2520/14725 [2:18:44<11:22:11,  3.35s/it]
 17%|█▋        | 2521/14725 [2:18:47<11:12:49,  3.31s/it]
 17%|█▋        | 2522/14725 [2:18:50<11:01:13,  3.25s/it]
 17%|█▋        | 2523/14725 [2:18:53<10:55:03,  3.22s/it]
 17%|█▋        | 2524/14725 [2:18:56<10:49:28,  3.19s/it]
 17%|█▋        | 2525/14725 [2:19:00<10:52:56,  3.21s/it]
 17%|█▋        | 2526/14725 [2:19:03<11:01:51,  3.26s
0: {'loss': 0.7181, 'grad_norm': 0.45374699779923494, 'learning_rate': 1e-05, 'epoch': 0.17}
0: /it]
 17%|█▋        | 2527/14725 [2:19:06<10:54:46,  3.22s/it]
 17%|█▋        | 2528/14725 [2:19:09<10:58:57,  3.24s/it]
 17%|█▋        | 2529/14725 [2:19:12<10:50:29,  3.20s/it]
 17%|█▋        | 2530/14725 [2:19:16<10:50:48,  3.20s/it]
                                                         

 17%|█▋        | 2530/14725 [2:19:16<10:50:48,  3.20s/it]
 17%|█▋        | 2531/14725 [2:19:19<10:56:29,  3.23s/it]
 17%|█▋        | 2532/14725 [2:19:22<10:57:51,  3.24s/it]
 17%|█▋        | 2533/14725 [2:19:26<11:14:10,  3.32s/it]
 17%|█▋        | 2534/14725 [2:19:29<11:10:22,  3.30s/it]
 17%|█▋        | 2535/14725 [2:19:32<11:04:21,  3.27s/it]
 17%|█▋        | 2536/14725 [2:19:36<11:14:15,  3.32s/it]
 17%|█▋        | 2537/14725 [2:19:39<11:04:11,  3.27s/it]
 17%|█▋        | 2538/14725 [2:19:42<10:52:12,  3.21s/it]
 17%|█▋        | 2539/14725 [2:19:45<11:01:10,  3.26s/it]
 17%|█▋        | 2540/14725 [2:19:48<10:58:03,  3.24s/it]
                              
0: {'loss': 0.7337, 'grad_norm': 0.43811808542125563, 'learning_rate': 1e-05, 'epoch': 0.17}
0: {'loss': 0.7498, 'grad_norm': 0.4856231694373928, 'learning_rate': 1e-05, 'epoch': 0.17}
0:                            

 17%|█▋        | 2540/14725 [2:19:48<10:58:03,  3.24s/it]
 17%|█▋        | 2541/14725 [2:19:52<11:01:52,  3.26s/it]
 17%|█▋        | 2542/14725 [2:19:55<11:06:24,  3.28s/it]
 17%|█▋        | 2543/14725 [2:19:58<11:01:27,  3.26s/it]
 17%|█▋        | 2544/14725 [2:20:01<10:51:28,  3.21s/it]
 17%|█▋        | 2545/14725 [2:20:04<10:39:48,  3.15s/it]
 17%|█▋        | 2546/14725 [2:20:08<10:48:31,  3.19s/it]
 17%|█▋        | 2547/14725 [2:20:11<10:47:28,  3.19s/it]
 17%|█▋        | 2548/14725 [2:20:14<10:58:15,  3.24s/it]
 17%|█▋        | 2549/14725 [2:20:18<11:04:43,  3.28s/it]
 17%|█▋        | 2550/14725 [2:20:21<11:19:54,  3.35s/it]
                                                         

 17%|█▋        | 2550/14725 [2:20:21<11:19:54,  3.35s/it]
 17%|█▋        | 2551/14725 [2:20:24<11:01:52,  3.26s/it]
 17%|█▋        | 2552/14725 [2:20:27<11:03:09,  3.27s/it]
 17%|█▋        | 2553/14725 [2:20:31<11:08:44,  3.30s/it]
 17%|�
0: {'loss': 0.7303, 'grad_norm': 0.4148906941965182, 'learning_rate': 1e-05, 'epoch': 0.17}
0: ��▋        | 2554/14725 [2:20:34<10:58:14,  3.24s/it]
 17%|█▋        | 2555/14725 [2:20:37<11:04:50,  3.28s/it]
 17%|█▋        | 2556/14725 [2:20:40<11:00:29,  3.26s/it]
 17%|█▋        | 2557/14725 [2:20:44<11:01:10,  3.26s/it]
 17%|█▋        | 2558/14725 [2:20:47<10:57:51,  3.24s/it]
 17%|█▋        | 2559/14725 [2:20:50<10:58:58,  3.25s/it]
 17%|█▋        | 2560/14725 [2:20:53<10:59:33,  3.25s/it]
                                                         

 17%|█▋        | 2560/14725 [2:20:53<10:59:33,  3.25s/it]
 17%|█▋        | 2561/14725 [2:20:56<10:47:24,  3.19s/it]
 17%|█▋        | 2562/14725 [2:21:00<10:56:06,  3.24s/it]
 17%|█▋        | 2563/14725 [2:21:03<10:48:26,  3.20s/it]
 17%|█▋        | 2564/14725 [2:21:06<10:52:50,  3.22s/it]
 17%|█▋        | 2565/14725 [2:21:10<11:00:45,  3.26s/it]
 17%|█▋        | 2566/14725 [2:21:13<11:08:07,  3.30s/it]
 17%|█▋        | 2567/14725 [2:21:16<11:12:23,  3.32s/it]
 17%|█▋        | 2568/14725 [2:21:19<
0: {'loss': 0.7432, 'grad_norm': 0.491408938380144, 'learning_rate': 1e-05, 'epoch': 0.17}
0: {'loss': 0.7468, 'grad_norm': 0.44386055405132635, 'learning_rate': 1e-05, 'epoch': 0.18}
0: 10:57:07,  3.24s/it]
 17%|█▋        | 2569/14725 [2:21:23<11:01:06,  3.26s/it]
 17%|█▋        | 2570/14725 [2:21:26<11:07:36,  3.30s/it]
                                                         

 17%|█▋        | 2570/14725 [2:21:26<11:07:36,  3.30s/it]
 17%|█▋        | 2571/14725 [2:21:29<10:52:19,  3.22s/it]
 17%|█▋        | 2572/14725 [2:21:33<11:11:19,  3.31s/it]
 17%|█▋        | 2573/14725 [2:21:36<11:07:01,  3.29s/it]
 17%|█▋        | 2574/14725 [2:21:39<11:14:29,  3.33s/it]
 17%|█▋        | 2575/14725 [2:21:43<11:23:54,  3.38s/it]
 17%|█▋        | 2576/14725 [2:21:46<11:10:42,  3.31s/it]
 18%|█▊        | 2577/14725 [2:21:49<11:18:38,  3.35s/it]
 18%|█▊        | 2578/14725 [2:21:53<11:16:28,  3.34s/it]
 18%|█▊        | 2579/14725 [2:21:56<11:21:29,  3.37s/it]
 18%|█▊        | 2580/14725 [2:21:59<11:11:28,  3.32s/it]
                                                         

 18%|█▊        | 2580/14725 [2:21:59<11:11:28,  3.32s/it]
 18%|█▊      
0: {'loss': 0.7195, 'grad_norm': 0.4923714965815564, 'learning_rate': 1e-05, 'epoch': 0.18}
0:   | 2581/14725 [2:22:03<11:12:05,  3.32s/it]
 18%|█▊        | 2582/14725 [2:22:06<11:11:31,  3.32s/it]
 18%|█▊        | 2583/14725 [2:22:09<11:01:47,  3.27s/it]
 18%|█▊        | 2584/14725 [2:22:13<11:08:17,  3.30s/it]
 18%|█▊        | 2585/14725 [2:22:16<11:03:13,  3.28s/it]
 18%|█▊        | 2586/14725 [2:22:19<11:00:08,  3.26s/it]
 18%|█▊        | 2587/14725 [2:22:22<10:56:14,  3.24s/it]
 18%|█▊        | 2588/14725 [2:22:25<11:00:38,  3.27s/it]
 18%|█▊        | 2589/14725 [2:22:29<10:54:03,  3.23s/it]
 18%|█▊        | 2590/14725 [2:22:32<10:48:13,  3.21s/it]
                                                         

 18%|█▊        | 2590/14725 [2:22:32<10:48:13,  3.21s/it]
 18%|█▊        | 2591/14725 [2:22:35<10:54:32,  3.24s/it]
 18%|█▊        | 2592/14725 [2:22:38<10:46:50,  3.20s/it]
 18%|█▊        | 2593/14725 [2:22:41<10:38:48,  3.16s/it]
 18%|█▊        | 2594/14725 [2:22:44<10:38:03,  3.16s/it]
 18%|█▊        | 2595/14725 [2:22:47<10:31:25,  
0: {'loss': 0.7192, 'grad_norm': 0.4458481178673679, 'learning_rate': 1e-05, 'epoch': 0.18}
0: 3.12s/it]
 18%|█▊        | 2596/14725 [2:22:51<10:38:11,  3.16s/it]
 18%|█▊        | 2597/14725 [2:22:54<10:37:03,  3.15s/it]
 18%|█▊        | 2598/14725 [2:22:57<10:35:50,  3.15s/it]
 18%|█▊        | 2599/14725 [2:23:00<10:31:09,  3.12s/it]
 18%|█▊        | 2600/14725 [2:23:03<10:44:31,  3.19s/it]
                                                         

 18%|█▊        | 2600/14725 [2:23:03<10:44:31,  3.19s/it]
 18%|█▊        | 2601/14725 [2:23:07<10:57:07,  3.25s/it]
 18%|█▊        | 2602/14725 [2:23:10<10:51:48,  3.23s/it]
 18%|█▊        | 2603/14725 [2:23:13<11:02:08,  3.28s/it]
 18%|█▊        | 2604/14725 [2:23:17<11:01:07,  3.27s/it]
 18%|█▊        | 2605/14725 [2:23:20<11:01:06,  3.27s/it]
 18%|█▊        | 2606/14725 [2:23:23<11:00:15,  3.27s/it]
 18%|█▊        | 2607/14725 [2:23:26<10:59:57,  3.27s/it]
 18%|█▊        | 2608/14725 [2:23:30<10:55:38,  3.25s/it]
 18%|█▊        | 2609/14725 [2:23:33<10:56:42,  3.25s/it]
 18%|█▊        | 2610
0: {'loss': 0.7322, 'grad_norm': 0.4140692870835122, 'learning_rate': 1e-05, 'epoch': 0.18}
0: {'loss': 0.7264, 'grad_norm': 0.42243985323942096, 'learning_rate': 1e-05, 'epoch': 0.18}
0: /14725 [2:23:36<10:59:02,  3.26s/it]
                                                         

 18%|█▊        | 2610/14725 [2:23:36<10:59:02,  3.26s/it]
 18%|█▊        | 2611/14725 [2:23:39<10:45:38,  3.20s/it]
 18%|█▊        | 2612/14725 [2:23:42<10:38:30,  3.16s/it]
 18%|█▊        | 2613/14725 [2:23:45<10:35:57,  3.15s/it]
 18%|█▊        | 2614/14725 [2:23:49<10:43:03,  3.19s/it]
 18%|█▊        | 2615/14725 [2:23:52<10:37:12,  3.16s/it]
 18%|█▊        | 2616/14725 [2:23:55<10:36:34,  3.15s/it]
 18%|█▊        | 2617/14725 [2:23:58<10:50:12,  3.22s/it]
 18%|█▊        | 2618/14725 [2:24:01<10:37:33,  3.16s/it]
 18%|█▊        | 2619/14725 [2:24:04<10:31:52,  3.13s/it]
 18%|█▊        | 2620/14725 [2:24:08<10:37:27,  3.16s/it]
                                                         

 18%|█▊        | 2620/14725 [2:24:08<10:37:27,  3.16s/it]
 18%|█▊        | 2621/14725 [2:24:11<10:38:11,  3.16s/it]
 18%|█▊        | 2622/14725 [2:24:14<10:47:24,  3.21s/it]
 
0: {'loss': 0.7334, 'grad_norm': 0.4770667322681491, 'learning_rate': 1e-05, 'epoch': 0.18}
0: 18%|█▊        | 2623/14725 [2:24:17<10:40:36,  3.18s/it]
 18%|█▊        | 2624/14725 [2:24:21<10:56:23,  3.25s/it]
 18%|█▊        | 2625/14725 [2:24:24<10:56:04,  3.25s/it]
 18%|█▊        | 2626/14725 [2:24:27<11:04:48,  3.30s/it]
 18%|█▊        | 2627/14725 [2:24:30<10:57:42,  3.26s/it]
 18%|█▊        | 2628/14725 [2:24:34<11:05:33,  3.30s/it]
 18%|█▊        | 2629/14725 [2:24:37<11:14:00,  3.34s/it]
 18%|█▊        | 2630/14725 [2:24:41<11:06:32,  3.31s/it]
                                                         

 18%|█▊        | 2630/14725 [2:24:41<11:06:32,  3.31s/it]
 18%|█▊        | 2631/14725 [2:24:44<11:14:28,  3.35s/it]
 18%|█▊        | 2632/14725 [2:24:47<11:01:09,  3.28s/it]
 18%|█▊        | 2633/14725 [2:24:50<10:51:20,  3.23s/it]
 18%|█▊        | 2634/14725 [2:24:53<10:50:55,  3.23s/it]
 18%|█▊        | 2635/14725 [2:24:57<10:56:13,  3.26s/it]
 18%|█▊        | 2636/14725 [2:25:00<10:53:37,  3.24s/it]
 18%|█▊        | 2637/14725 [2:2
0: {'loss': 0.7232, 'grad_norm': 0.5650922457394013, 'learning_rate': 1e-05, 'epoch': 0.18}
0: {'loss': 0.7482, 'grad_norm': 0.4619688057552699, 'learning_rate': 1e-05, 'epoch': 0.18}
0: 5:03<10:54:34,  3.25s/it]
 18%|█▊        | 2638/14725 [2:25:06<10:40:01,  3.18s/it]
 18%|█▊        | 2639/14725 [2:25:10<10:47:45,  3.22s/it]
 18%|█▊        | 2640/14725 [2:25:13<10:56:12,  3.26s/it]
                                                         

 18%|█▊        | 2640/14725 [2:25:13<10:56:12,  3.26s/it]
 18%|█▊        | 2641/14725 [2:25:16<11:05:46,  3.31s/it]
 18%|█▊        | 2642/14725 [2:25:19<10:56:31,  3.26s/it]
 18%|█▊        | 2643/14725 [2:25:23<10:55:04,  3.25s/it]
 18%|█▊        | 2644/14725 [2:25:26<10:49:02,  3.22s/it]
 18%|█▊        | 2645/14725 [2:25:29<10:43:09,  3.19s/it]
 18%|█▊        | 2646/14725 [2:25:32<10:57:11,  3.26s/it]
 18%|█▊        | 2647/14725 [2:25:36<11:05:37,  3.31s/it]
 18%|█▊        | 2648/14725 [2:25:39<11:07:43,  3.32s/it]
 18%|█▊        | 2649/14725 [2:25:42<11:00:34,  3.28s/it]
 18%|█▊        | 2650/14725 [2:25:46<10:58:00,  3.27s/it]
                                                         

 18%|█▊ 
0: {'loss': 0.7421, 'grad_norm': 0.4881221945983132, 'learning_rate': 1e-05, 'epoch': 0.18}
0:        | 2650/14725 [2:25:46<10:58:00,  3.27s/it]
 18%|█▊        | 2651/14725 [2:25:49<10:56:41,  3.26s/it]
 18%|█▊        | 2652/14725 [2:25:52<11:10:59,  3.33s/it]
 18%|█▊        | 2653/14725 [2:25:56<11:04:44,  3.30s/it]
 18%|█▊        | 2654/14725 [2:25:59<11:08:51,  3.32s/it]
 18%|█▊        | 2655/14725 [2:26:02<11:09:31,  3.33s/it]
 18%|█▊        | 2656/14725 [2:26:05<10:53:45,  3.25s/it]
 18%|█▊        | 2657/14725 [2:26:09<10:48:23,  3.22s/it]
 18%|█▊        | 2658/14725 [2:26:12<11:09:05,  3.33s/it]
 18%|█▊        | 2659/14725 [2:26:16<11:18:46,  3.38s/it]
 18%|█▊        | 2660/14725 [2:26:19<11:24:00,  3.40s/it]
                                                         

 18%|█▊        | 2660/14725 [2:26:19<11:24:00,  3.40s/it]
 18%|█▊        | 2661/14725 [2:26:22<11:25:30,  3.41s/it]
 18%|█▊        | 2662/14725 [2:26:26<11:13:33,  3.35s/it]
 18%|█▊        | 2663/14725 [2:26:29<10:56:10,  3.26s/it]
 18%|█▊        | 2664/14725 [2:26:32<10:58:
0: {'loss': 0.7356, 'grad_norm': 0.48622729243325674, 'learning_rate': 1e-05, 'epoch': 0.18}
0: 36,  3.28s/it]
 18%|█▊        | 2665/14725 [2:26:35<11:01:51,  3.29s/it]
 18%|█▊        | 2666/14725 [2:26:39<11:04:41,  3.31s/it]
 18%|█▊        | 2667/14725 [2:26:42<11:01:22,  3.29s/it]
 18%|█▊        | 2668/14725 [2:26:45<10:57:32,  3.27s/it]
 18%|█▊        | 2669/14725 [2:26:48<10:41:36,  3.19s/it]
 18%|█▊        | 2670/14725 [2:26:51<10:40:25,  3.19s/it]
                                                         

 18%|█▊        | 2670/14725 [2:26:51<10:40:25,  3.19s/it]
 18%|█▊        | 2671/14725 [2:26:55<10:41:41,  3.19s/it]
 18%|█▊        | 2672/14725 [2:26:58<10:46:03,  3.22s/it]
 18%|█▊        | 2673/14725 [2:27:01<10:53:02,  3.25s/it]
 18%|█▊        | 2674/14725 [2:27:04<10:47:30,  3.22s/it]
 18%|█▊        | 2675/14725 [2:27:08<10:49:54,  3.24s/it]
 18%|█▊        | 2676/14725 [2:27:11<10:57:07,  3.27s/it]
 18%|█▊        | 2677/14725 [2:27:14<10:49:56,  3.24s/it]
 18%|█▊        | 2678/14725 [2:27:17<10:45:46,  3.22s/it]
 18%|█▊        |
0: {'loss': 0.7576, 'grad_norm': 0.459132381679591, 'learning_rate': 1e-05, 'epoch': 0.18}
0: {'loss': 0.7315, 'grad_norm': 0.4251274377286154, 'learning_rate': 1e-05, 'epoch': 0.18}
0:  2679/14725 [2:27:21<10:49:24,  3.23s/it]
 18%|█▊        | 2680/14725 [2:27:24<10:41:35,  3.20s/it]
                                                         

 18%|█▊        | 2680/14725 [2:27:24<10:41:35,  3.20s/it]
 18%|█▊        | 2681/14725 [2:27:27<10:45:22,  3.22s/it]
 18%|█▊        | 2682/14725 [2:27:30<10:52:58,  3.25s/it]
 18%|█▊        | 2683/14725 [2:27:34<10:57:02,  3.27s/it]
 18%|█▊        | 2684/14725 [2:27:37<10:59:43,  3.29s/it]
 18%|█▊        | 2685/14725 [2:27:40<10:53:28,  3.26s/it]
 18%|█▊        | 2686/14725 [2:27:43<10:49:22,  3.24s/it]
 18%|█▊        | 2687/14725 [2:27:46<10:41:37,  3.20s/it]
 18%|█▊        | 2688/14725 [2:27:50<10:46:55,  3.22s/it]
 18%|█▊        | 2689/14725 [2:27:53<10:45:10,  3.22s/it]
 18%|█▊        | 2690/14725 [2:27:56<10:41:08,  3.20s/it]
                                                         

 18%|█▊        | 2690/14725 [2:27:56<10:41:08,  3.20s/it]
 18%|█▊        | 2691/14725 [2:27:59<10:52:22,  3.25s/
0: {'loss': 0.7328, 'grad_norm': 0.46186856687179806, 'learning_rate': 1e-05, 'epoch': 0.18}
0: it]
 18%|█▊        | 2692/14725 [2:28:03<10:56:49,  3.28s/it]
 18%|█▊        | 2693/14725 [2:28:06<11:08:24,  3.33s/it]
 18%|█▊        | 2694/14725 [2:28:09<11:05:36,  3.32s/it]
 18%|█▊        | 2695/14725 [2:28:13<10:56:34,  3.27s/it]
 18%|█▊        | 2696/14725 [2:28:16<10:48:22,  3.23s/it]
 18%|█▊        | 2697/14725 [2:28:19<10:43:03,  3.21s/it]
 18%|█▊        | 2698/14725 [2:28:22<10:34:34,  3.17s/it]
 18%|█▊        | 2699/14725 [2:28:25<10:49:00,  3.24s/it]
 18%|█▊        | 2700/14725 [2:28:29<10:54:44,  3.27s/it]
                                                         

 18%|█▊        | 2700/14725 [2:28:29<10:54:44,  3.27s/it]
 18%|█▊        | 2701/14725 [2:28:32<11:01:01,  3.30s/it]
 18%|█▊        | 2702/14725 [2:28:36<11:10:51,  3.35s/it]
 18%|█▊        | 2703/14725 [2:28:39<10:52:14,  3.26s/it]
 18%|█▊        | 2704/14725 [2:28:42<10:44:35,  3.22s/it]
 18%|█▊        | 2705/14725 [2:28:45<10:43:00,  3.21s/it]
 18%|█▊        | 2706/14725
0: {'loss': 0.7374, 'grad_norm': 0.47945716979235803, 'learning_rate': 1e-05, 'epoch': 0.18}
0:  [2:28:48<10:39:01,  3.19s/it]
 18%|█▊        | 2707/14725 [2:28:51<10:34:24,  3.17s/it]
 18%|█▊        | 2708/14725 [2:28:55<10:51:22,  3.25s/it]
 18%|█▊        | 2709/14725 [2:28:58<10:40:45,  3.20s/it]
 18%|█▊        | 2710/14725 [2:29:01<10:45:35,  3.22s/it]
                                                         

 18%|█▊        | 2710/14725 [2:29:01<10:45:35,  3.22s/it]
 18%|█▊        | 2711/14725 [2:29:04<10:45:27,  3.22s/it]
 18%|█▊        | 2712/14725 [2:29:07<10:43:25,  3.21s/it]
 18%|█▊        | 2713/14725 [2:29:11<10:40:36,  3.20s/it]
 18%|█▊        | 2714/14725 [2:29:14<10:44:51,  3.22s/it]
 18%|█▊        | 2715/14725 [2:29:17<10:54:28,  3.27s/it]
 18%|█▊        | 2716/14725 [2:29:21<10:56:39,  3.28s/it]
 18%|█▊        | 2717/14725 [2:29:24<10:52:06,  3.26s/it]
 18%|█▊        | 2718/14725 [2:29:27<11:08:39,  3.34s/it]
 18%|█▊        | 2719/14725 [2:29:30<10:58:32,  3.29s/it]
 18%|█▊        | 2720/14725 [2:29:34<11:00:48,  3.30s/it]
    
0: {'loss': 0.7425, 'grad_norm': 0.44376062539624683, 'learning_rate': 1e-05, 'epoch': 0.18}
0: {'loss': 0.7392, 'grad_norm': 0.4382432493552869, 'learning_rate': 1e-05, 'epoch': 0.19}
0:                                                      

 18%|█▊        | 2720/14725 [2:29:34<11:00:48,  3.30s/it]
 18%|█▊        | 2721/14725 [2:29:37<10:54:21,  3.27s/it]
 18%|█▊        | 2722/14725 [2:29:40<10:46:15,  3.23s/it]
 18%|█▊        | 2723/14725 [2:29:44<10:59:48,  3.30s/it]
 18%|█▊        | 2724/14725 [2:29:47<11:02:51,  3.31s/it]
 19%|█▊        | 2725/14725 [2:29:50<10:55:54,  3.28s/it]
 19%|█▊        | 2726/14725 [2:29:53<10:48:14,  3.24s/it]
 19%|█▊        | 2727/14725 [2:29:57<10:47:06,  3.24s/it]
 19%|█▊        | 2728/14725 [2:30:00<10:46:12,  3.23s/it]
 19%|█▊        | 2729/14725 [2:30:03<10:45:06,  3.23s/it]
 19%|█▊        | 2730/14725 [2:30:06<10:42:08,  3.21s/it]
                                                         

 19%|█▊        | 2730/14725 [2:30:06<10:42:08,  3.21s/it]
 19%|█▊        | 2731/14725 [2:30:09<10:43:19,  3.22s/it]
 19%|█▊        | 2732/14725 [2:30:13<10:44:46,  3.23s/it]
 19%|█▊        | 2733/14725 [2:30:16<1
0: {'loss': 0.7287, 'grad_norm': 0.4483156126044248, 'learning_rate': 1e-05, 'epoch': 0.19}
0: 0:47:06,  3.24s/it]
 19%|█▊        | 2734/14725 [2:30:19<10:59:28,  3.30s/it]
 19%|█▊        | 2735/14725 [2:30:23<11:08:00,  3.34s/it]
 19%|█▊        | 2736/14725 [2:30:26<11:02:15,  3.31s/it]
 19%|█▊        | 2737/14725 [2:30:29<10:41:21,  3.21s/it]
 19%|█▊        | 2738/14725 [2:30:32<10:30:30,  3.16s/it]
 19%|█▊        | 2739/14725 [2:30:35<10:43:42,  3.22s/it]
 19%|█▊        | 2740/14725 [2:30:39<10:38:41,  3.20s/it]
                                                         

 19%|█▊        | 2740/14725 [2:30:39<10:38:41,  3.20s/it]
 19%|█▊        | 2741/14725 [2:30:42<10:49:01,  3.25s/it]
 19%|█▊        | 2742/14725 [2:30:45<11:03:32,  3.32s/it]
 19%|█▊        | 2743/14725 [2:30:49<10:58:18,  3.30s/it]
 19%|█▊        | 2744/14725 [2:30:52<11:02:43,  3.32s/it]
 19%|█▊        | 2745/14725 [2:30:55<11:08:49,  3.35s/it]
 19%|█▊        | 2746/14725 [2:30:59<10:57:26,  3.29s/it]
 19%|█▊        | 2747/14725 [2:31:02<10:48:54,  3.25s/it]
 19%|█▊    
0: {'loss': 0.7501, 'grad_norm': 0.4507774718229416, 'learning_rate': 1e-05, 'epoch': 0.19}
0: {'loss': 0.7312, 'grad_norm': 0.45517944175825875, 'learning_rate': 1e-05, 'epoch': 0.19}
0:     | 2748/14725 [2:31:05<10:54:29,  3.28s/it]
 19%|█▊        | 2749/14725 [2:31:08<10:53:06,  3.27s/it]
 19%|█▊        | 2750/14725 [2:31:12<10:58:44,  3.30s/it]
                                                         

 19%|█▊        | 2750/14725 [2:31:12<10:58:44,  3.30s/it]
 19%|█▊        | 2751/14725 [2:31:15<10:52:13,  3.27s/it]
 19%|█▊        | 2752/14725 [2:31:18<10:58:02,  3.30s/it]
 19%|█▊        | 2753/14725 [2:31:21<10:42:43,  3.22s/it]
 19%|█▊        | 2754/14725 [2:31:24<10:39:56,  3.21s/it]
 19%|█▊        | 2755/14725 [2:31:28<10:31:11,  3.16s/it]
 19%|█▊        | 2756/14725 [2:31:31<10:25:44,  3.14s/it]
 19%|█▊        | 2757/14725 [2:31:34<10:36:45,  3.19s/it]
 19%|█▊        | 2758/14725 [2:31:37<10:41:19,  3.22s/it]
 19%|█▊        | 2759/14725 [2:31:40<10:34:19,  3.18s/it]
 19%|█▊        | 2760/14725 [2:31:44<10:38:56,  3.20s/it]
                                                         

 19%|█▊        | 2760/14725 [2:31:44<10:38:56,  3
0: {'loss': 0.7329, 'grad_norm': 0.4186228431585726, 'learning_rate': 1e-05, 'epoch': 0.19}
0: .20s/it]
 19%|█▉        | 2761/14725 [2:31:47<10:31:58,  3.17s/it]
 19%|█▉        | 2762/14725 [2:31:50<10:31:42,  3.17s/it]
 19%|█▉        | 2763/14725 [2:31:53<10:31:59,  3.17s/it]
 19%|█▉        | 2764/14725 [2:31:56<10:45:16,  3.24s/it]
 19%|█▉        | 2765/14725 [2:32:00<10:44:14,  3.23s/it]
 19%|█▉        | 2766/14725 [2:32:03<10:38:04,  3.20s/it]
 19%|█▉        | 2767/14725 [2:32:06<10:45:01,  3.24s/it]
 19%|█▉        | 2768/14725 [2:32:09<10:49:21,  3.26s/it]
 19%|█▉        | 2769/14725 [2:32:13<10:43:42,  3.23s/it]
 19%|█▉        | 2770/14725 [2:32:16<10:40:53,  3.22s/it]
                                                         

 19%|█▉        | 2770/14725 [2:32:16<10:40:53,  3.22s/it]
 19%|█▉        | 2771/14725 [2:32:19<10:48:45,  3.26s/it]
 19%|█▉        | 2772/14725 [2:32:22<10:43:33,  3.23s/it]
 19%|█▉        | 2773/14725 [2:32:26<10:58:39,  3.31s/it]
 19%|█▉        | 2774/14725 [2:32:29<10:56:47,  3.30s/it]
 19%|█▉        | 2775/
0: {'loss': 0.7363, 'grad_norm': 0.47852415252061126, 'learning_rate': 1e-05, 'epoch': 0.19}
0: 14725 [2:32:32<10:52:20,  3.28s/it]
 19%|█▉        | 2776/14725 [2:32:35<10:44:48,  3.24s/it]
 19%|█▉        | 2777/14725 [2:32:39<10:53:01,  3.28s/it]
 19%|█▉        | 2778/14725 [2:32:42<10:59:23,  3.31s/it]
 19%|█▉        | 2779/14725 [2:32:45<10:43:02,  3.23s/it]
 19%|█▉        | 2780/14725 [2:32:49<10:51:56,  3.27s/it]
                                                         

 19%|█▉        | 2780/14725 [2:32:49<10:51:56,  3.27s/it]
 19%|█▉        | 2781/14725 [2:32:52<10:48:50,  3.26s/it]
 19%|█▉        | 2782/14725 [2:32:55<10:50:32,  3.27s/it]
 19%|█▉        | 2783/14725 [2:32:58<10:52:35,  3.28s/it]
 19%|█▉        | 2784/14725 [2:33:02<10:56:41,  3.30s/it]
 19%|█▉        | 2785/14725 [2:33:05<10:47:42,  3.25s/it]
 19%|█▉        | 2786/14725 [2:33:08<10:34:21,  3.19s/it]
 19%|█▉        | 2787/14725 [2:33:11<10:44:28,  3.24s/it]
 19%|█▉        | 2788/14725 [2:33:15<10:50:21,  3.27s/it]
 19%|█▉        | 2789/14725 [2:33:18<10:52:18,  3.28s/it]
0: {'loss': 0.7338, 'grad_norm': 0.47373465347376814, 'learning_rate': 1e-05, 'epoch': 0.19}
0: {'loss': 0.7361, 'grad_norm': 0.4834670995941496, 'learning_rate': 1e-05, 'epoch': 0.19}
0: 
 19%|█▉        | 2790/14725 [2:33:21<10:56:23,  3.30s/it]
                                                         

 19%|█▉        | 2790/14725 [2:33:21<10:56:23,  3.30s/it]
 19%|█▉        | 2791/14725 [2:33:25<10:55:56,  3.30s/it]
 19%|█▉        | 2792/14725 [2:33:28<10:53:35,  3.29s/it]
 19%|█▉        | 2793/14725 [2:33:31<10:46:35,  3.25s/it]
 19%|█▉        | 2794/14725 [2:33:34<10:41:33,  3.23s/it]
 19%|█▉        | 2795/14725 [2:33:37<10:47:20,  3.26s/it]
 19%|█▉        | 2796/14725 [2:33:41<10:54:24,  3.29s/it]
 19%|█▉        | 2797/14725 [2:33:44<10:48:21,  3.26s/it]
 19%|█▉        | 2798/14725 [2:33:47<10:52:18,  3.28s/it]
 19%|█▉        | 2799/14725 [2:33:51<10:44:40,  3.24s/it]
 19%|█▉        | 2800/14725 [2:33:54<10:42:09,  3.23s/it]
                                                         

 19%|█▉        | 2800/14725 [2:33:54<10:42:09,  3.23s/it]
 19%|█▉        | 2801/14725 [2:33:57<10:59:36,  3.32s/it]
 19%|█▉        | 2802/14725 [2:34
0: {'loss': 0.7264, 'grad_norm': 0.43466351092408556, 'learning_rate': 1e-05, 'epoch': 0.19}
0: :01<11:00:07,  3.32s/it]
 19%|█▉        | 2803/14725 [2:34:04<11:03:00,  3.34s/it]
 19%|█▉        | 2804/14725 [2:34:07<10:55:13,  3.30s/it]
 19%|█▉        | 2805/14725 [2:34:11<11:04:59,  3.35s/it]
 19%|█▉        | 2806/14725 [2:34:14<10:52:54,  3.29s/it]
 19%|█▉        | 2807/14725 [2:34:17<10:44:44,  3.25s/it]
 19%|█▉        | 2808/14725 [2:34:20<10:41:23,  3.23s/it]
 19%|█▉        | 2809/14725 [2:34:23<10:33:03,  3.19s/it]
 19%|█▉        | 2810/14725 [2:34:26<10:38:08,  3.21s/it]
                                                         

 19%|█▉        | 2810/14725 [2:34:26<10:38:08,  3.21s/it]
 19%|█▉        | 2811/14725 [2:34:30<10:33:02,  3.19s/it]
 19%|█▉        | 2812/14725 [2:34:33<10:36:27,  3.21s/it]
 19%|█▉        | 2813/14725 [2:34:36<10:37:40,  3.21s/it]
 19%|█▉        | 2814/14725 [2:34:39<10:38:35,  3.22s/it]
 19%|█▉        | 2815/14725 [2:34:43<10:38:47,  3.22s/it]
 19%|█▉        | 2816/14725 [2:34:46<10:44:19,  3.25s/it]
 19%|█�
0: {'loss': 0.7306, 'grad_norm': 0.43749864761184615, 'learning_rate': 1e-05, 'epoch': 0.19}
0: �        | 2817/14725 [2:34:49<10:45:58,  3.25s/it]
 19%|█▉        | 2818/14725 [2:34:52<10:46:15,  3.26s/it]
 19%|█▉        | 2819/14725 [2:34:56<10:41:01,  3.23s/it]
 19%|█▉        | 2820/14725 [2:34:59<10:38:04,  3.22s/it]
                                                         

 19%|█▉        | 2820/14725 [2:34:59<10:38:04,  3.22s/it]
 19%|█▉        | 2821/14725 [2:35:02<10:36:06,  3.21s/it]
 19%|█▉        | 2822/14725 [2:35:05<10:39:41,  3.22s/it]
 19%|█▉        | 2823/14725 [2:35:09<10:47:36,  3.26s/it]
 19%|█▉        | 2824/14725 [2:35:12<11:03:51,  3.35s/it]
 19%|█▉        | 2825/14725 [2:35:15<11:07:56,  3.37s/it]
 19%|█▉        | 2826/14725 [2:35:19<11:03:05,  3.34s/it]
 19%|█▉        | 2827/14725 [2:35:22<11:05:55,  3.36s/it]
 19%|█▉        | 2828/14725 [2:35:25<10:53:13,  3.29s/it]
 19%|█▉        | 2829/14725 [2:35:28<10:46:50,  3.26s/it]
 19%|█▉        | 2830/14725 [2:35:32<10:47:06,  3.26s/it]
                                             
0: {'loss': 0.7349, 'grad_norm': 0.4343222900743621, 'learning_rate': 1e-05, 'epoch': 0.19}
0: {'loss': 0.7239, 'grad_norm': 0.4440234829801047, 'learning_rate': 1e-05, 'epoch': 0.19}
0:             

 19%|█▉        | 2830/14725 [2:35:32<10:47:06,  3.26s/it]
 19%|█▉        | 2831/14725 [2:35:35<10:47:00,  3.26s/it]
 19%|█▉        | 2832/14725 [2:35:38<10:37:34,  3.22s/it]
 19%|█▉        | 2833/14725 [2:35:41<10:34:43,  3.20s/it]
 19%|█▉        | 2834/14725 [2:35:45<10:50:40,  3.28s/it]
 19%|█▉        | 2835/14725 [2:35:48<10:41:32,  3.24s/it]
 19%|█▉        | 2836/14725 [2:35:51<10:49:37,  3.28s/it]
 19%|█▉        | 2837/14725 [2:35:55<11:01:18,  3.34s/it]
 19%|█▉        | 2838/14725 [2:35:58<10:55:00,  3.31s/it]
 19%|█▉        | 2839/14725 [2:36:01<10:45:41,  3.26s/it]
 19%|█▉        | 2840/14725 [2:36:04<10:38:59,  3.23s/it]
                                                         

 19%|█▉        | 2840/14725 [2:36:04<10:38:59,  3.23s/it]
 19%|█▉        | 2841/14725 [2:36:08<10:44:58,  3.26s/it]
 19%|█▉        | 2842/14725 [2:36:11<10:42:35,  3.24s/it]
 19%|█▉        | 2843/14725 [2:36:14<10:48:20,  3.27s/it]
 19%|█▉        | 
0: {'loss': 0.7282, 'grad_norm': 0.45818503053935195, 'learning_rate': 1e-05, 'epoch': 0.19}
0: 2844/14725 [2:36:17<10:47:35,  3.27s/it]
 19%|█▉        | 2845/14725 [2:36:21<10:46:21,  3.26s/it]
 19%|█▉        | 2846/14725 [2:36:24<10:47:51,  3.27s/it]
 19%|█▉        | 2847/14725 [2:36:27<10:48:48,  3.28s/it]
 19%|█▉        | 2848/14725 [2:36:31<10:49:04,  3.28s/it]
 19%|█▉        | 2849/14725 [2:36:34<10:54:10,  3.30s/it]
 19%|█▉        | 2850/14725 [2:36:37<10:52:35,  3.30s/it]
                                                         

 19%|█▉        | 2850/14725 [2:36:37<10:52:35,  3.30s/it]
 19%|█▉        | 2851/14725 [2:36:40<10:46:56,  3.27s/it]
 19%|█▉        | 2852/14725 [2:36:44<10:43:20,  3.25s/it]
 19%|█▉        | 2853/14725 [2:36:47<10:42:26,  3.25s/it]
 19%|█▉        | 2854/14725 [2:36:50<10:37:07,  3.22s/it]
 19%|█▉        | 2855/14725 [2:36:53<10:43:35,  3.25s/it]
 19%|█▉        | 2856/14725 [2:36:57<10:53:34,  3.30s/it]
 19%|█▉        | 2857/14725 [2:37:00<10:52:41,  3.30s/it]
 19%|█▉        | 2858/14725 [2:37:03<10:47:54,  3.28
0: {'loss': 0.7282, 'grad_norm': 0.4446202117313477, 'learning_rate': 1e-05, 'epoch': 0.19}
0: {'loss': 0.7253, 'grad_norm': 0.43244734001308, 'learning_rate': 1e-05, 'epoch': 0.19}
0: s/it]
 19%|█▉        | 2859/14725 [2:37:06<10:44:11,  3.26s/it]
 19%|█▉        | 2860/14725 [2:37:10<10:39:42,  3.23s/it]
                                                         

 19%|█▉        | 2860/14725 [2:37:10<10:39:42,  3.23s/it]
 19%|█▉        | 2861/14725 [2:37:13<10:40:52,  3.24s/it]
 19%|█▉        | 2862/14725 [2:37:16<10:45:56,  3.27s/it]
 19%|█▉        | 2863/14725 [2:37:20<10:49:32,  3.29s/it]
 19%|█▉        | 2864/14725 [2:37:23<10:42:14,  3.25s/it]
 19%|█▉        | 2865/14725 [2:37:26<10:37:23,  3.22s/it]
 19%|█▉        | 2866/14725 [2:37:29<10:45:08,  3.26s/it]
 19%|█▉        | 2867/14725 [2:37:33<10:53:09,  3.30s/it]
 19%|█▉        | 2868/14725 [2:37:36<11:03:42,  3.36s/it]
 19%|█▉        | 2869/14725 [2:37:39<10:59:17,  3.34s/it]
 19%|█▉        | 2870/14725 [2:37:43<10:54:14,  3.31s/it]
                                                         

 19%|█▉        | 2870/14725 [2:37:43<10:54:14,  3.31s/it]
 19%|█▉        | 2871/14725 
0: {'loss': 0.7172, 'grad_norm': 0.5072097240931683, 'learning_rate': 1e-05, 'epoch': 0.2}
0: [2:37:46<11:01:48,  3.35s/it]
 20%|█▉        | 2872/14725 [2:37:49<11:00:29,  3.34s/it]
 20%|█▉        | 2873/14725 [2:37:53<10:50:01,  3.29s/it]
 20%|█▉        | 2874/14725 [2:37:56<10:47:31,  3.28s/it]
 20%|█▉        | 2875/14725 [2:37:59<10:48:41,  3.28s/it]
 20%|█▉        | 2876/14725 [2:38:02<10:47:07,  3.28s/it]
 20%|█▉        | 2877/14725 [2:38:06<10:51:14,  3.30s/it]
 20%|█▉        | 2878/14725 [2:38:09<10:41:05,  3.25s/it]
 20%|█▉        | 2879/14725 [2:38:12<10:34:26,  3.21s/it]
 20%|█▉        | 2880/14725 [2:38:15<10:45:38,  3.27s/it]
                                                         

 20%|█▉        | 2880/14725 [2:38:15<10:45:38,  3.27s/it]
 20%|█▉        | 2881/14725 [2:38:19<10:34:13,  3.21s/it]
 20%|█▉        | 2882/14725 [2:38:22<10:22:22,  3.15s/it]
 20%|█▉        | 2883/14725 [2:38:25<10:32:37,  3.21s/it]
 20%|█▉        | 2884/14725 [2:38:28<10:39:41,  3.24s/it]
 20%|█▉        | 2885/14725 [2:38:32<10:48:09,  3.28s/it]
 20%|
0: {'loss': 0.7283, 'grad_norm': 0.516669688564251, 'learning_rate': 1e-05, 'epoch': 0.2}
0: █▉        | 2886/14725 [2:38:35<10:45:52,  3.27s/it]
 20%|█▉        | 2887/14725 [2:38:38<10:39:12,  3.24s/it]
 20%|█▉        | 2888/14725 [2:38:41<10:34:17,  3.22s/it]
 20%|█▉        | 2889/14725 [2:38:44<10:24:40,  3.17s/it]
 20%|█▉        | 2890/14725 [2:38:48<10:44:10,  3.27s/it]
                                                         

 20%|█▉        | 2890/14725 [2:38:48<10:44:10,  3.27s/it]
 20%|█▉        | 2891/14725 [2:38:51<10:52:02,  3.31s/it]
 20%|█▉        | 2892/14725 [2:38:54<10:50:07,  3.30s/it]
 20%|█▉        | 2893/14725 [2:38:58<10:49:42,  3.29s/it]
 20%|█▉        | 2894/14725 [2:39:01<10:43:13,  3.26s/it]
 20%|█▉        | 2895/14725 [2:39:04<10:40:55,  3.25s/it]
 20%|█▉        | 2896/14725 [2:39:07<10:35:47,  3.22s/it]
 20%|█▉        | 2897/14725 [2:39:11<10:41:21,  3.25s/it]
 20%|█▉        | 2898/14725 [2:39:14<10:42:50,  3.26s/it]
 20%|█▉        | 2899/14725 [2:39:17<10:36:59,  3.23s/it]
 20%|█▉        | 2900/14725 [2:39:20
0: {'loss': 0.7373, 'grad_norm': 0.5283450439095077, 'learning_rate': 1e-05, 'epoch': 0.2}
0: {'loss': 0.7453, 'grad_norm': 0.44120797884729174, 'learning_rate': 1e-05, 'epoch': 0.2}
0: <10:33:49,  3.22s/it]
                                                         

 20%|█▉        | 2900/14725 [2:39:20<10:33:49,  3.22s/it]
 20%|█▉        | 2901/14725 [2:39:23<10:25:37,  3.17s/it]
 20%|█▉        | 2902/14725 [2:39:27<10:35:24,  3.22s/it]
 20%|█▉        | 2903/14725 [2:39:30<10:36:27,  3.23s/it]
 20%|█▉        | 2904/14725 [2:39:33<10:41:31,  3.26s/it]
 20%|█▉        | 2905/14725 [2:39:37<10:48:04,  3.29s/it]
 20%|█▉        | 2906/14725 [2:39:40<10:53:43,  3.32s/it]
 20%|█▉        | 2907/14725 [2:39:43<10:43:19,  3.27s/it]
 20%|█▉        | 2908/14725 [2:39:47<10:56:03,  3.33s/it]
 20%|█▉        | 2909/14725 [2:39:50<10:50:37,  3.30s/it]
 20%|█▉        | 2910/14725 [2:39:53<10:40:48,  3.25s/it]
                                                         

 20%|█▉        | 2910/14725 [2:39:53<10:40:48,  3.25s/it]
 20%|█▉        | 2911/14725 [2:39:56<10:47:27,  3.29s/it]
 20%|█▉        | 2912/14725 [2:40:00<10:55:15,  3.33s/it]
 20%|█▉     
0: {'loss': 0.7219, 'grad_norm': 0.4379366258071218, 'learning_rate': 1e-05, 'epoch': 0.2}
0:    | 2913/14725 [2:40:03<10:39:46,  3.25s/it]
 20%|█▉        | 2914/14725 [2:40:06<10:31:17,  3.21s/it]
 20%|█▉        | 2915/14725 [2:40:09<10:24:12,  3.17s/it]
 20%|█▉        | 2916/14725 [2:40:12<10:18:21,  3.14s/it]
 20%|█▉        | 2917/14725 [2:40:15<10:20:47,  3.15s/it]
 20%|█▉        | 2918/14725 [2:40:19<10:35:48,  3.23s/it]
 20%|█▉        | 2919/14725 [2:40:22<10:41:15,  3.26s/it]
 20%|█▉        | 2920/14725 [2:40:25<10:32:36,  3.22s/it]
                                                         

 20%|█▉        | 2920/14725 [2:40:25<10:32:36,  3.22s/it]
 20%|█▉        | 2921/14725 [2:40:28<10:34:10,  3.22s/it]
 20%|█▉        | 2922/14725 [2:40:31<10:28:57,  3.20s/it]
 20%|█▉        | 2923/14725 [2:40:35<10:34:51,  3.23s/it]
 20%|█▉        | 2924/14725 [2:40:38<10:32:27,  3.22s/it]
 20%|█▉        | 2925/14725 [2:40:41<10:34:29,  3.23s/it]
 20%|█▉        | 2926/14725 [2:40:44<10:39:27,  3.25s/it]
 20%|█▉        | 2927/14725 [2:40:48<10:34:34, 
0: {'loss': 0.7487, 'grad_norm': 0.538190733213322, 'learning_rate': 1e-05, 'epoch': 0.2}
0: {'loss': 0.7352, 'grad_norm': 0.47118000926053777, 'learning_rate': 1e-05, 'epoch': 0.2}
0:  3.23s/it]
 20%|█▉        | 2928/14725 [2:40:51<10:51:55,  3.32s/it]
 20%|█▉        | 2929/14725 [2:40:54<10:45:33,  3.28s/it]
 20%|█▉        | 2930/14725 [2:40:58<10:39:36,  3.25s/it]
                                                         

 20%|█▉        | 2930/14725 [2:40:58<10:39:36,  3.25s/it]
 20%|█▉        | 2931/14725 [2:41:01<10:41:24,  3.26s/it]
 20%|█▉        | 2932/14725 [2:41:04<10:35:24,  3.23s/it]
 20%|█▉        | 2933/14725 [2:41:07<10:38:38,  3.25s/it]
 20%|█▉        | 2934/14725 [2:41:11<10:38:22,  3.25s/it]
 20%|█▉        | 2935/14725 [2:41:14<10:42:05,  3.27s/it]
 20%|█▉        | 2936/14725 [2:41:17<10:39:38,  3.26s/it]
 20%|█▉        | 2937/14725 [2:41:20<10:41:23,  3.26s/it]
 20%|█▉        | 2938/14725 [2:41:24<10:37:10,  3.24s/it]
 20%|█▉        | 2939/14725 [2:41:27<10:28:16,  3.20s/it]
 20%|█▉        | 2940/14725 [2:41:30<10:40:16,  3.26s/it]
                                                         

 20%|█▉        | 2940/1
0: {'loss': 0.7364, 'grad_norm': 0.48518732285297106, 'learning_rate': 1e-05, 'epoch': 0.2}
0: 4725 [2:41:30<10:40:16,  3.26s/it]
 20%|█▉        | 2941/14725 [2:41:33<10:30:49,  3.21s/it]
 20%|█▉        | 2942/14725 [2:41:36<10:34:18,  3.23s/it]
 20%|█▉        | 2943/14725 [2:41:40<10:42:09,  3.27s/it]
 20%|█▉        | 2944/14725 [2:41:43<10:48:51,  3.30s/it]
 20%|██        | 2945/14725 [2:41:46<10:45:23,  3.29s/it]
 20%|██        | 2946/14725 [2:41:58<18:45:59,  5.74s/it]
 20%|██        | 2947/14725 [2:42:01<16:24:52,  5.02s/it]
 20%|██        | 2948/14725 [2:42:05<14:44:52,  4.51s/it]
 20%|██        | 2949/14725 [2:42:08<13:34:35,  4.15s/it]
 20%|██        | 2950/14725 [2:42:11<12:46:05,  3.90s/it]
                                                         

 20%|██        | 2950/14725 [2:42:11<12:46:05,  3.90s/it]
 20%|██        | 2951/14725 [2:42:14<12:11:07,  3.73s/it]
 20%|██        | 2952/14725 [2:42:18<11:37:31,  3.55s/it]
 20%|██        | 2953/14725 [2:42:21<11:10:54,  3.42s/it]
 20%|██        | 2954/14725 [2:42:24<11:12:40,  3.43s/it]
0: {'loss': 0.7319, 'grad_norm': 0.4518220550880349, 'learning_rate': 1e-05, 'epoch': 0.2}
0:  20%|██        | 2955/14725 [2:42:28<11:17:35,  3.45s/it]
 20%|██        | 2956/14725 [2:42:31<10:58:27,  3.36s/it]
 20%|██        | 2957/14725 [2:42:34<10:53:34,  3.33s/it]
 20%|██        | 2958/14725 [2:42:37<10:48:19,  3.31s/it]
 20%|██        | 2959/14725 [2:42:41<10:48:56,  3.31s/it]
 20%|██        | 2960/14725 [2:42:44<10:41:06,  3.27s/it]
                                                         

 20%|██        | 2960/14725 [2:42:44<10:41:06,  3.27s/it]
 20%|██        | 2961/14725 [2:42:47<10:48:38,  3.31s/it]
 20%|██        | 2962/14725 [2:42:51<10:49:52,  3.31s/it]
 20%|██        | 2963/14725 [2:42:54<11:02:26,  3.38s/it]
 20%|██        | 2964/14725 [2:42:57<10:43:34,  3.28s/it]
 20%|██        | 2965/14725 [2:43:00<10:41:24,  3.27s/it]
 20%|██        | 2966/14725 [2:43:04<10:34:46,  3.24s/it]
 20%|██        | 2967/14725 [2:43:07<10:32:49,  3.23s/it]
 20%|██        | 2968/14725 [2:43:10<10:28:40,  3.21s/it]
 20%|██        | 2969/14725 [2:
0: {'loss': 0.7279, 'grad_norm': 0.474857161121587, 'learning_rate': 1e-05, 'epoch': 0.2}
0: {'loss': 0.7386, 'grad_norm': 0.4429627769658516, 'learning_rate': 1e-05, 'epoch': 0.2}
0: 43:13<10:25:51,  3.19s/it]
 20%|██        | 2970/14725 [2:43:16<10:29:05,  3.21s/it]
                                                         

 20%|██        | 2970/14725 [2:43:16<10:29:05,  3.21s/it]
 20%|██        | 2971/14725 [2:43:20<10:25:34,  3.19s/it]
 20%|██        | 2972/14725 [2:43:23<10:34:30,  3.24s/it]
 20%|██        | 2973/14725 [2:43:26<10:35:16,  3.24s/it]
 20%|██        | 2974/14725 [2:43:29<10:35:38,  3.25s/it]
 20%|██        | 2975/14725 [2:43:33<10:32:48,  3.23s/it]
 20%|██        | 2976/14725 [2:43:36<10:30:51,  3.22s/it]
 20%|██        | 2977/14725 [2:43:39<10:43:46,  3.29s/it]
 20%|██        | 2978/14725 [2:43:42<10:31:47,  3.23s/it]
 20%|██        | 2979/14725 [2:43:46<10:37:19,  3.26s/it]
 20%|██        | 2980/14725 [2:43:49<10:37:05,  3.25s/it]
                                                         

 20%|██        | 2980/14725 [2:43:49<10:37:05,  3.25s/it]
 20%|██        | 2981/14725 [2:43:52<10:33:15,  3.24s/it]
 20%|██
0: {'loss': 0.7296, 'grad_norm': 0.45370002776679924, 'learning_rate': 1e-05, 'epoch': 0.2}
0:         | 2982/14725 [2:43:55<10:33:17,  3.24s/it]
 20%|██        | 2983/14725 [2:43:59<10:44:05,  3.29s/it]
 20%|██        | 2984/14725 [2:44:02<11:00:05,  3.37s/it]
 20%|██        | 2985/14725 [2:44:06<11:00:41,  3.38s/it]
 20%|██        | 2986/14725 [2:44:09<10:53:54,  3.34s/it]
 20%|██        | 2987/14725 [2:44:12<10:59:59,  3.37s/it]
 20%|██        | 2988/14725 [2:44:16<10:52:07,  3.33s/it]
 20%|██        | 2989/14725 [2:44:19<10:50:59,  3.33s/it]
 20%|██        | 2990/14725 [2:44:22<10:41:54,  3.28s/it]
                                                         

 20%|██        | 2990/14725 [2:44:22<10:41:54,  3.28s/it]
 20%|██        | 2991/14725 [2:44:25<10:24:42,  3.19s/it]
 20%|██        | 2992/14725 [2:44:28<10:23:26,  3.19s/it]
 20%|██        | 2993/14725 [2:44:32<10:37:30,  3.26s/it]
 20%|██        | 2994/14725 [2:44:35<10:34:05,  3.24s/it]
 20%|██        | 2995/14725 [2:44:38<10:37:17,  3.26s/it]
 20%|██        | 2996/14725 [2:44:41<10:30
0: {'loss': 0.7494, 'grad_norm': 0.4873096997498121, 'learning_rate': 1e-05, 'epoch': 0.2}
0: :59,  3.23s/it]
 20%|██        | 2997/14725 [2:44:45<10:33:52,  3.24s/it]
 20%|██        | 2998/14725 [2:44:48<10:27:53,  3.21s/it]
 20%|██        | 2999/14725 [2:44:51<10:19:06,  3.17s/it]
 20%|██        | 3000/14725 [2:44:54<10:18:01,  3.16s/it]
                                                         

 20%|██        | 3000/14725 [2:44:54<10:18:01,  3.16s/it]
 20%|██        | 3001/14725 [2:44:57<10:14:42,  3.15s/it]
 20%|██        | 3002/14725 [2:45:00<10:18:24,  3.17s/it]
 20%|██        | 3003/14725 [2:45:04<10:24:08,  3.19s/it]
 20%|██        | 3004/14725 [2:45:07<10:23:23,  3.19s/it]
 20%|██        | 3005/14725 [2:45:10<10:32:20,  3.24s/it]
 20%|██        | 3006/14725 [2:45:13<10:27:23,  3.21s/it]
 20%|██        | 3007/14725 [2:45:16<10:23:58,  3.19s/it]
 20%|██        | 3008/14725 [2:45:20<10:28:37,  3.22s/it]
 20%|██        | 3009/14725 [2:45:23<10:41:51,  3.29s/it]
 20%|██        | 3010/14725 [2:45:26<10:36:57,  3.26s/it]
                   
0: {'loss': 0.7479, 'grad_norm': 0.4510813757089561, 'learning_rate': 1e-05, 'epoch': 0.2}
0: {'loss': 0.7359, 'grad_norm': 0.44183865696071417, 'learning_rate': 1e-05, 'epoch': 0.21}
0:                                       

 20%|██        | 3010/14725 [2:45:26<10:36:57,  3.26s/it]
 20%|██        | 3011/14725 [2:45:30<10:40:31,  3.28s/it]
 20%|██        | 3012/14725 [2:45:33<10:30:43,  3.23s/it]
 20%|██        | 3013/14725 [2:45:36<10:25:47,  3.21s/it]
 20%|██        | 3014/14725 [2:45:39<10:23:12,  3.19s/it]
 20%|██        | 3015/14725 [2:45:42<10:20:34,  3.18s/it]
 20%|██        | 3016/14725 [2:45:45<10:24:50,  3.20s/it]
 20%|██        | 3017/14725 [2:45:49<10:22:25,  3.19s/it]
 20%|██        | 3018/14725 [2:45:52<10:24:26,  3.20s/it]
 21%|██        | 3019/14725 [2:45:55<10:38:26,  3.27s/it]
 21%|██        | 3020/14725 [2:45:59<10:42:52,  3.30s/it]
                                                         

 21%|██        | 3020/14725 [2:45:59<10:42:52,  3.30s/it]
 21%|██        | 3021/14725 [2:46:02<10:42:27,  3.29s/it]
 21%|██        | 3022/14725 [2:46:05<10:30:46,  3.23s/it]
 21%|██        | 3023/14725 [2:46:08<10:26:37,  3.21s
0: {'loss': 0.7139, 'grad_norm': 0.44868981223616633, 'learning_rate': 1e-05, 'epoch': 0.21}
0: /it]
 21%|██        | 3024/14725 [2:46:11<10:22:38,  3.19s/it]
 21%|██        | 3025/14725 [2:46:15<10:21:46,  3.19s/it]
 21%|██        | 3026/14725 [2:46:18<10:22:49,  3.19s/it]
 21%|██        | 3027/14725 [2:46:21<10:33:42,  3.25s/it]
 21%|██        | 3028/14725 [2:46:24<10:28:29,  3.22s/it]
 21%|██        | 3029/14725 [2:46:27<10:21:09,  3.19s/it]
 21%|██        | 3030/14725 [2:46:31<10:36:07,  3.26s/it]
                                                         

 21%|██        | 3030/14725 [2:46:31<10:36:07,  3.26s/it]
 21%|██        | 3031/14725 [2:46:34<10:27:19,  3.22s/it]
 21%|██        | 3032/14725 [2:46:37<10:23:50,  3.20s/it]
 21%|██        | 3033/14725 [2:46:40<10:28:12,  3.22s/it]
 21%|██        | 3034/14725 [2:46:44<10:36:55,  3.27s/it]
 21%|██        | 3035/14725 [2:46:47<10:37:30,  3.27s/it]
 21%|██        | 3036/14725 [2:46:50<10:44:15,  3.31s/it]
 21%|██        | 3037/14725 [2:46:54<10:42:25,  3.30s/it]
 21%|██        | 3038/1472
0: {'loss': 0.7399, 'grad_norm': 0.45426278429999933, 'learning_rate': 1e-05, 'epoch': 0.21}
0: {'loss': 0.724, 'grad_norm': 0.4368091307803466, 'learning_rate': 1e-05, 'epoch': 0.21}
0: 5 [2:46:57<10:44:32,  3.31s/it]
 21%|██        | 3039/14725 [2:47:00<10:36:46,  3.27s/it]
 21%|██        | 3040/14725 [2:47:04<10:38:59,  3.28s/it]
                                                         

 21%|██        | 3040/14725 [2:47:04<10:38:59,  3.28s/it]
 21%|██        | 3041/14725 [2:47:07<10:38:45,  3.28s/it]
 21%|██        | 3042/14725 [2:47:10<10:42:07,  3.30s/it]
 21%|██        | 3043/14725 [2:47:13<10:25:08,  3.21s/it]
 21%|██        | 3044/14725 [2:47:16<10:28:22,  3.23s/it]
 21%|██        | 3045/14725 [2:47:20<10:33:19,  3.25s/it]
 21%|██        | 3046/14725 [2:47:23<10:31:55,  3.25s/it]
 21%|██        | 3047/14725 [2:47:26<10:41:43,  3.30s/it]
 21%|██        | 3048/14725 [2:47:30<10:52:19,  3.35s/it]
 21%|██        | 3049/14725 [2:47:33<10:50:23,  3.34s/it]
 21%|██        | 3050/14725 [2:47:36<10:45:08,  3.32s/it]
                                                         

 21%|██        | 3050/14725 [2:47:36<10:45:08,  3.32s/it]
 21%|�
0: {'loss': 0.7526, 'grad_norm': 0.4768033953933234, 'learning_rate': 1e-05, 'epoch': 0.21}
0: ��█        | 3051/14725 [2:47:40<10:35:41,  3.27s/it]
 21%|██        | 3052/14725 [2:47:43<10:44:10,  3.31s/it]
 21%|██        | 3053/14725 [2:47:46<10:43:39,  3.31s/it]
 21%|██        | 3054/14725 [2:47:50<10:40:55,  3.29s/it]
 21%|██        | 3055/14725 [2:47:53<10:44:31,  3.31s/it]
 21%|██        | 3056/14725 [2:47:56<10:39:07,  3.29s/it]
 21%|██        | 3057/14725 [2:48:00<10:52:05,  3.35s/it]
 21%|██        | 3058/14725 [2:48:03<10:49:51,  3.34s/it]
 21%|██        | 3059/14725 [2:48:06<10:54:28,  3.37s/it]
 21%|██        | 3060/14725 [2:48:10<10:53:13,  3.36s/it]
                                                         

 21%|██        | 3060/14725 [2:48:10<10:53:13,  3.36s/it]
 21%|██        | 3061/14725 [2:48:13<10:56:31,  3.38s/it]
 21%|██        | 3062/14725 [2:48:16<10:44:46,  3.32s/it]
 21%|██        | 3063/14725 [2:48:20<10:46:54,  3.33s/it]
 21%|██        | 3064/14725 [2:48:23<10:41:06,  3.30s/it]
 21%|██        | 3065/14725 [2:48:26<
0: {'loss': 0.7466, 'grad_norm': 0.4606171898005768, 'learning_rate': 1e-05, 'epoch': 0.21}
0: 10:33:02,  3.26s/it]
 21%|██        | 3066/14725 [2:48:29<10:31:49,  3.25s/it]
 21%|██        | 3067/14725 [2:48:33<10:30:17,  3.24s/it]
 21%|██        | 3068/14725 [2:48:36<10:37:44,  3.28s/it]
 21%|██        | 3069/14725 [2:48:39<10:24:06,  3.21s/it]
 21%|██        | 3070/14725 [2:48:42<10:42:29,  3.31s/it]
                                                         

 21%|██        | 3070/14725 [2:48:42<10:42:29,  3.31s/it]
 21%|██        | 3071/14725 [2:48:46<10:48:22,  3.34s/it]
 21%|██        | 3072/14725 [2:48:49<10:43:27,  3.31s/it]
 21%|██        | 3073/14725 [2:48:52<10:32:39,  3.26s/it]
 21%|██        | 3074/14725 [2:48:55<10:26:22,  3.23s/it]
 21%|██        | 3075/14725 [2:48:59<10:26:17,  3.23s/it]
 21%|██        | 3076/14725 [2:49:02<10:22:32,  3.21s/it]
 21%|██        | 3077/14725 [2:49:05<10:24:24,  3.22s/it]
 21%|██        | 3078/14725 [2:49:08<10:20:12,  3.20s/it]
 21%|██        | 3079/14725 [2:49:11<10:23:50,  3.21s/it]
 21%|██   
0: {'loss': 0.7289, 'grad_norm': 0.42522114543333034, 'learning_rate': 1e-05, 'epoch': 0.21}
0: {'loss': 0.7409, 'grad_norm': 0.4669465406549385, 'learning_rate': 1e-05, 'epoch': 0.21}
0:      | 3080/14725 [2:49:15<10:29:33,  3.24s/it]
                                                         

 21%|██        | 3080/14725 [2:49:15<10:29:33,  3.24s/it]
 21%|██        | 3081/14725 [2:49:18<10:36:33,  3.28s/it]
 21%|██        | 3082/14725 [2:49:21<10:41:51,  3.31s/it]
 21%|██        | 3083/14725 [2:49:25<10:39:35,  3.30s/it]
 21%|██        | 3084/14725 [2:49:28<10:33:25,  3.26s/it]
 21%|██        | 3085/14725 [2:49:31<10:32:04,  3.26s/it]
 21%|██        | 3086/14725 [2:49:35<10:35:33,  3.28s/it]
 21%|██        | 3087/14725 [2:49:38<10:30:03,  3.25s/it]
 21%|██        | 3088/14725 [2:49:41<10:32:01,  3.26s/it]
 21%|██        | 3089/14725 [2:49:44<10:37:18,  3.29s/it]
 21%|██        | 3090/14725 [2:49:48<10:37:30,  3.29s/it]
                                                         

 21%|██        | 3090/14725 [2:49:48<10:37:30,  3.29s/it]
 21%|██        | 3091/14725 [2:49:51<10:54:10,  3.37s/it]
 21%|██        | 3092/14725 [2:49:54<10:46:42,  
0: {'loss': 0.7452, 'grad_norm': 0.4343373503149217, 'learning_rate': 1e-05, 'epoch': 0.21}
0: 3.34s/it]
 21%|██        | 3093/14725 [2:49:58<10:38:45,  3.29s/it]
 21%|██        | 3094/14725 [2:50:01<10:56:24,  3.39s/it]
 21%|██        | 3095/14725 [2:50:04<10:48:18,  3.34s/it]
 21%|██        | 3096/14725 [2:50:08<10:49:02,  3.35s/it]
 21%|██        | 3097/14725 [2:50:11<10:47:54,  3.34s/it]
 21%|██        | 3098/14725 [2:50:14<10:39:44,  3.30s/it]
 21%|██        | 3099/14725 [2:50:18<10:33:17,  3.27s/it]
 21%|██        | 3100/14725 [2:50:21<10:35:01,  3.28s/it]
                                                         

 21%|██        | 3100/14725 [2:50:21<10:35:01,  3.28s/it]
 21%|██        | 3101/14725 [2:50:24<10:34:43,  3.28s/it]
 21%|██        | 3102/14725 [2:50:27<10:33:46,  3.27s/it]
 21%|██        | 3103/14725 [2:50:31<10:24:05,  3.22s/it]
 21%|██        | 3104/14725 [2:50:34<10:24:36,  3.22s/it]
 21%|██        | 3105/14725 [2:50:37<10:25:18,  3.23s/it]
 21%|██        | 3106/14725 [2:50:40<10:30:25,  3.26s/it]
 21%|██        | 3107
0: {'loss': 0.7075, 'grad_norm': 0.4448992432967303, 'learning_rate': 1e-05, 'epoch': 0.21}
0: {'loss': 0.7381, 'grad_norm': 0.481057293451574, 'learning_rate': 1e-05, 'epoch': 0.21}
0: /14725 [2:50:44<10:35:56,  3.28s/it]
 21%|██        | 3108/14725 [2:50:47<10:48:50,  3.35s/it]
 21%|██        | 3109/14725 [2:50:50<10:36:56,  3.29s/it]
 21%|██        | 3110/14725 [2:50:54<10:34:26,  3.28s/it]
                                                         

 21%|██        | 3110/14725 [2:50:54<10:34:26,  3.28s/it]
 21%|██        | 3111/14725 [2:50:57<10:24:12,  3.22s/it]
 21%|██        | 3112/14725 [2:51:00<10:34:10,  3.28s/it]
 21%|██        | 3113/14725 [2:51:03<10:36:09,  3.29s/it]
 21%|██        | 3114/14725 [2:51:07<10:41:03,  3.31s/it]
 21%|██        | 3115/14725 [2:51:10<10:38:22,  3.30s/it]
 21%|██        | 3116/14725 [2:51:13<10:37:00,  3.29s/it]
 21%|██        | 3117/14725 [2:51:16<10:32:23,  3.27s/it]
 21%|██        | 3118/14725 [2:51:20<10:46:24,  3.34s/it]
 21%|██        | 3119/14725 [2:51:23<10:50:56,  3.37s/it]
 21%|██        | 3120/14725 [2:51:27<10:47:39,  3.35s/it]
                                                         

 
0: {'loss': 0.717, 'grad_norm': 0.4360864562562342, 'learning_rate': 1e-05, 'epoch': 0.21}
0: 21%|██        | 3120/14725 [2:51:27<10:47:39,  3.35s/it]
 21%|██        | 3121/14725 [2:51:30<10:42:35,  3.32s/it]
 21%|██        | 3122/14725 [2:51:34<10:54:35,  3.38s/it]
 21%|██        | 3123/14725 [2:51:37<10:47:31,  3.35s/it]
 21%|██        | 3124/14725 [2:51:40<10:45:44,  3.34s/it]
 21%|██        | 3125/14725 [2:51:43<10:30:10,  3.26s/it]
 21%|██        | 3126/14725 [2:51:47<10:33:37,  3.28s/it]
 21%|██        | 3127/14725 [2:51:50<10:29:49,  3.26s/it]
 21%|██        | 3128/14725 [2:51:53<10:19:19,  3.20s/it]
 21%|██        | 3129/14725 [2:51:56<10:16:24,  3.19s/it]
 21%|██▏       | 3130/14725 [2:51:59<10:16:13,  3.19s/it]
                                                         

 21%|██▏       | 3130/14725 [2:51:59<10:16:13,  3.19s/it]
 21%|██▏       | 3131/14725 [2:52:02<10:12:01,  3.17s/it]
 21%|██▏       | 3132/14725 [2:52:05<10:13:43,  3.18s/it]
 21%|██▏       | 3133/14725 [2:52:09<10:19:40,  3.21s/it]
 21%|██▏       | 313
0: {'loss': 0.7259, 'grad_norm': 0.44363273538999215, 'learning_rate': 1e-05, 'epoch': 0.21}
0: 4/14725 [2:52:12<10:25:52,  3.24s/it]
 21%|██▏       | 3135/14725 [2:52:15<10:35:33,  3.29s/it]
 21%|██▏       | 3136/14725 [2:52:19<10:45:15,  3.34s/it]
 21%|██▏       | 3137/14725 [2:52:22<10:45:23,  3.34s/it]
 21%|██▏       | 3138/14725 [2:52:25<10:32:12,  3.27s/it]
 21%|██▏       | 3139/14725 [2:52:29<10:28:53,  3.26s/it]
 21%|██▏       | 3140/14725 [2:52:32<10:29:27,  3.26s/it]
                                                         

 21%|██▏       | 3140/14725 [2:52:32<10:29:27,  3.26s/it]
 21%|██▏       | 3141/14725 [2:52:35<10:19:10,  3.21s/it]
 21%|██▏       | 3142/14725 [2:52:38<10:10:03,  3.16s/it]
 21%|██▏       | 3143/14725 [2:52:41<10:18:59,  3.21s/it]
 21%|██▏       | 3144/14725 [2:52:44<10:13:45,  3.18s/it]
 21%|██▏       | 3145/14725 [2:52:48<10:08:37,  3.15s/it]
 21%|██▏       | 3146/14725 [2:52:51<10:24:47,  3.24s/it]
 21%|██▏       | 3147/14725 [2:52:54<10:28:21,  3.26s/it]
 21%|██▏       | 3148/147
0: {'loss': 0.7336, 'grad_norm': 0.49113186486098154, 'learning_rate': 1e-05, 'epoch': 0.21}
0: {'loss': 0.7295, 'grad_norm': 0.4521031314484686, 'learning_rate': 1e-05, 'epoch': 0.21}
0: 25 [2:52:58<10:29:31,  3.26s/it]
 21%|██▏       | 3149/14725 [2:53:01<10:21:41,  3.22s/it]
 21%|██▏       | 3150/14725 [2:53:04<10:37:07,  3.30s/it]
                                                         

 21%|██▏       | 3150/14725 [2:53:04<10:37:07,  3.30s/it]
 21%|██▏       | 3151/14725 [2:53:07<10:33:20,  3.28s/it]
 21%|██▏       | 3152/14725 [2:53:10<10:22:15,  3.23s/it]
 21%|██▏       | 3153/14725 [2:53:14<10:28:45,  3.26s/it]
 21%|██▏       | 3154/14725 [2:53:17<10:29:25,  3.26s/it]
 21%|██▏       | 3155/14725 [2:53:20<10:27:05,  3.25s/it]
 21%|██▏       | 3156/14725 [2:53:24<10:27:14,  3.25s/it]
 21%|██▏       | 3157/14725 [2:53:27<10:40:44,  3.32s/it]
 21%|██▏       | 3158/14725 [2:53:30<10:37:48,  3.31s/it]
 21%|██▏       | 3159/14725 [2:53:33<10:29:20,  3.26s/it]
 21%|██▏       | 3160/14725 [2:53:37<10:29:17,  3.26s/it]
                                                         

 21%|██▏       | 3160/14725 [2:53:3
0: {'loss': 0.7219, 'grad_norm': 0.4620456517253236, 'learning_rate': 1e-05, 'epoch': 0.22}
0: 7<10:29:17,  3.26s/it]
 21%|██▏       | 3161/14725 [2:53:40<10:23:49,  3.24s/it]
 21%|██▏       | 3162/14725 [2:53:43<10:28:49,  3.26s/it]
 21%|██▏       | 3163/14725 [2:53:47<10:28:03,  3.26s/it]
 21%|██▏       | 3164/14725 [2:53:50<10:27:46,  3.26s/it]
 21%|██▏       | 3165/14725 [2:53:53<10:18:38,  3.21s/it]
 22%|██▏       | 3166/14725 [2:53:56<10:12:06,  3.18s/it]
 22%|██▏       | 3167/14725 [2:53:59<10:06:58,  3.15s/it]
 22%|██▏       | 3168/14725 [2:54:02<10:05:00,  3.14s/it]
 22%|██▏       | 3169/14725 [2:54:06<10:30:38,  3.27s/it]
 22%|██▏       | 3170/14725 [2:54:09<10:22:46,  3.23s/it]
                                                         

 22%|██▏       | 3170/14725 [2:54:09<10:22:46,  3.23s/it]
 22%|██▏       | 3171/14725 [2:54:12<10:26:29,  3.25s/it]
 22%|██▏       | 3172/14725 [2:54:15<10:24:48,  3.24s/it]
 22%|██▏       | 3173/14725 [2:54:19<10:31:53,  3.28s/it]
 22%|██▏       | 3174/14725 [2:54:22<10:
0: {'loss': 0.7278, 'grad_norm': 0.46786617409830356, 'learning_rate': 1e-05, 'epoch': 0.22}
0: 40:02,  3.32s/it]
 22%|██▏       | 3175/14725 [2:54:26<10:54:56,  3.40s/it]
 22%|██▏       | 3176/14725 [2:54:29<10:57:35,  3.42s/it]
 22%|██▏       | 3177/14725 [2:54:32<10:40:07,  3.33s/it]
 22%|██▏       | 3178/14725 [2:54:36<10:42:25,  3.34s/it]
 22%|██▏       | 3179/14725 [2:54:39<10:45:09,  3.35s/it]
 22%|██▏       | 3180/14725 [2:54:42<10:42:05,  3.34s/it]
                                                         

 22%|██▏       | 3180/14725 [2:54:42<10:42:05,  3.34s/it]
 22%|██▏       | 3181/14725 [2:54:46<10:35:33,  3.30s/it]
 22%|██▏       | 3182/14725 [2:54:49<10:30:29,  3.28s/it]
 22%|██▏       | 3183/14725 [2:54:52<10:23:46,  3.24s/it]
 22%|██▏       | 3184/14725 [2:54:55<10:22:31,  3.24s/it]
 22%|██▏       | 3185/14725 [2:54:59<10:27:51,  3.26s/it]
 22%|██▏       | 3186/14725 [2:55:02<10:24:38,  3.25s/it]
 22%|██▏       | 3187/14725 [2:55:05<10:24:09,  3.25s/it]
 22%|██▏       | 3188/14725 [2:55:08<10:28:46
0: {'loss': 0.7213, 'grad_norm': 0.46115568852337285, 'learning_rate': 1e-05, 'epoch': 0.22}
0: {'loss': 0.7231, 'grad_norm': 0.39874031243591096, 'learning_rate': 1e-05, 'epoch': 0.22}
0: ,  3.27s/it]
 22%|██▏       | 3189/14725 [2:55:12<10:31:33,  3.28s/it]
 22%|██▏       | 3190/14725 [2:55:15<10:22:38,  3.24s/it]
                                                         

 22%|██▏       | 3190/14725 [2:55:15<10:22:38,  3.24s/it]
 22%|██▏       | 3191/14725 [2:55:18<10:29:18,  3.27s/it]
 22%|██▏       | 3192/14725 [2:55:22<10:38:29,  3.32s/it]
 22%|██▏       | 3193/14725 [2:55:25<10:38:41,  3.32s/it]
 22%|██▏       | 3194/14725 [2:55:28<10:32:18,  3.29s/it]
 22%|██▏       | 3195/14725 [2:55:31<10:25:20,  3.25s/it]
 22%|██▏       | 3196/14725 [2:55:34<10:16:38,  3.21s/it]
 22%|██▏       | 3197/14725 [2:55:38<10:19:38,  3.23s/it]
 22%|██▏       | 3198/14725 [2:55:41<10:38:18,  3.32s/it]
 22%|██▏       | 3199/14725 [2:55:44<10:33:31,  3.30s/it]
 22%|██▏       | 3200/14725 [2:55:48<10:22:56,  3.24s/it]
                                                         

 22%|██▏       | 3200/14725 [2:55:48<10:22:56,  3.24s/i
0: {'loss': 0.7294, 'grad_norm': 0.4319665257683145, 'learning_rate': 1e-05, 'epoch': 0.22}
0: t]
 22%|██▏       | 3201/14725 [2:55:51<10:14:22,  3.20s/it]
 22%|██▏       | 3202/14725 [2:55:54<10:22:46,  3.24s/it]
 22%|██▏       | 3203/14725 [2:55:57<10:36:42,  3.32s/it]
 22%|██▏       | 3204/14725 [2:56:01<10:27:30,  3.27s/it]
 22%|██▏       | 3205/14725 [2:56:04<10:25:59,  3.26s/it]
 22%|██▏       | 3206/14725 [2:56:08<10:52:56,  3.40s/it]
 22%|██▏       | 3207/14725 [2:56:11<10:49:44,  3.38s/it]
 22%|██▏       | 3208/14725 [2:56:14<10:45:45,  3.36s/it]
 22%|██▏       | 3209/14725 [2:56:18<10:50:00,  3.39s/it]
 22%|██▏       | 3210/14725 [2:56:21<10:48:10,  3.38s/it]
                                                         

 22%|██▏       | 3210/14725 [2:56:21<10:48:10,  3.38s/it]
 22%|██▏       | 3211/14725 [2:56:24<10:38:08,  3.33s/it]
 22%|██▏       | 3212/14725 [2:56:27<10:30:48,  3.29s/it]
 22%|██▏       | 3213/14725 [2:56:31<10:30:41,  3.29s/it]
 22%|██▏       | 3214/14725 [2:56:34<10:55:02,  3.41s/it]
 2
0: {'loss': 0.7373, 'grad_norm': 0.5162007222852709, 'learning_rate': 1e-05, 'epoch': 0.22}
0: 2%|██▏       | 3215/14725 [2:56:38<10:45:05,  3.36s/it]
 22%|██▏       | 3216/14725 [2:56:41<10:36:35,  3.32s/it]
 22%|██▏       | 3217/14725 [2:56:44<10:39:39,  3.34s/it]
 22%|██▏       | 3218/14725 [2:56:47<10:30:22,  3.29s/it]
 22%|██▏       | 3219/14725 [2:56:51<10:37:28,  3.32s/it]
 22%|██▏       | 3220/14725 [2:56:54<10:23:38,  3.25s/it]
                                                         

 22%|██▏       | 3220/14725 [2:56:54<10:23:38,  3.25s/it]
 22%|██▏       | 3221/14725 [2:56:57<10:16:18,  3.21s/it]
 22%|██▏       | 3222/14725 [2:57:00<10:19:13,  3.23s/it]
 22%|██▏       | 3223/14725 [2:57:04<10:14:24,  3.21s/it]
 22%|██▏       | 3224/14725 [2:57:07<10:24:20,  3.26s/it]
 22%|██▏       | 3225/14725 [2:57:10<10:13:07,  3.20s/it]
 22%|██▏       | 3226/14725 [2:57:13<10:20:02,  3.24s/it]
 22%|██▏       | 3227/14725 [2:57:16<10:16:09,  3.22s/it]
 22%|██▏       | 3228/14725 [2:57:20<10:15:01,  3.21s/it]
 22%|�
0: {'loss': 0.7282, 'grad_norm': 0.4824034094676873, 'learning_rate': 1e-05, 'epoch': 0.22}
0: {'loss': 0.7206, 'grad_norm': 0.5024376125973475, 'learning_rate': 1e-05, 'epoch': 0.22}
0: �█▏       | 3229/14725 [2:57:23<10:16:04,  3.22s/it]
 22%|██▏       | 3230/14725 [2:57:26<10:13:31,  3.20s/it]
                                                         

 22%|██▏       | 3230/14725 [2:57:26<10:13:31,  3.20s/it]
 22%|██▏       | 3231/14725 [2:57:29<10:17:30,  3.22s/it]
 22%|██▏       | 3232/14725 [2:57:33<10:23:04,  3.25s/it]
 22%|██▏       | 3233/14725 [2:57:36<10:21:45,  3.25s/it]
 22%|██▏       | 3234/14725 [2:57:39<10:19:46,  3.24s/it]
 22%|██▏       | 3235/14725 [2:57:43<10:52:54,  3.41s/it]
 22%|██▏       | 3236/14725 [2:57:46<10:43:16,  3.36s/it]
 22%|██▏       | 3237/14725 [2:57:49<10:36:46,  3.33s/it]
 22%|██▏       | 3238/14725 [2:57:53<10:36:57,  3.33s/it]
 22%|██▏       | 3239/14725 [2:57:56<10:36:06,  3.32s/it]
 22%|██▏       | 3240/14725 [2:57:59<10:30:37,  3.29s/it]
                                                         

 22%|██▏       | 3240/14725 [2:57:59<10:30:37,  3.29s/it]
 22%|██▏   
0: {'loss': 0.7493, 'grad_norm': 0.4496601997540678, 'learning_rate': 1e-05, 'epoch': 0.22}
0:     | 3241/14725 [2:58:02<10:25:41,  3.27s/it]
 22%|██▏       | 3242/14725 [2:58:06<10:28:31,  3.28s/it]
 22%|██▏       | 3243/14725 [2:58:09<10:26:40,  3.27s/it]
 22%|██▏       | 3244/14725 [2:58:12<10:24:53,  3.27s/it]
 22%|██▏       | 3245/14725 [2:58:15<10:17:23,  3.23s/it]
 22%|██▏       | 3246/14725 [2:58:19<10:12:36,  3.20s/it]
 22%|██▏       | 3247/14725 [2:58:22<10:14:39,  3.21s/it]
 22%|██▏       | 3248/14725 [2:58:25<10:06:04,  3.17s/it]
 22%|██▏       | 3249/14725 [2:58:28<10:07:53,  3.18s/it]
 22%|██▏       | 3250/14725 [2:58:31<10:15:55,  3.22s/it]
                                                         

 22%|██▏       | 3250/14725 [2:58:31<10:15:55,  3.22s/it]
 22%|██▏       | 3251/14725 [2:58:35<10:15:35,  3.22s/it]
 22%|██▏       | 3252/14725 [2:58:38<10:16:48,  3.23s/it]
 22%|██▏       | 3253/14725 [2:58:41<10:21:00,  3.25s/it]
 22%|██▏       | 3254/14725 [2:58:45<10:30:07,  3.30s/it]
 22%|██▏       |
0: {'loss': 0.7439, 'grad_norm': 0.470602923671945, 'learning_rate': 1e-05, 'epoch': 0.22}
0:  3255/14725 [2:58:48<10:26:37,  3.28s/it]
 22%|██▏       | 3256/14725 [2:58:51<10:22:03,  3.25s/it]
 22%|██▏       | 3257/14725 [2:58:54<10:16:12,  3.22s/it]
 22%|██▏       | 3258/14725 [2:58:58<10:34:59,  3.32s/it]
 22%|██▏       | 3259/14725 [2:59:01<10:30:33,  3.30s/it]
 22%|██▏       | 3260/14725 [2:59:04<10:27:47,  3.29s/it]
                                                         

 22%|██▏       | 3260/14725 [2:59:04<10:27:47,  3.29s/it]
 22%|██▏       | 3261/14725 [2:59:07<10:19:09,  3.24s/it]
 22%|██▏       | 3262/14725 [2:59:10<10:09:56,  3.19s/it]
 22%|██▏       | 3263/14725 [2:59:14<10:16:24,  3.23s/it]
 22%|██▏       | 3264/14725 [2:59:17<10:17:25,  3.23s/it]
 22%|██▏       | 3265/14725 [2:59:21<10:37:23,  3.34s/it]
 22%|██▏       | 3266/14725 [2:59:24<10:41:34,  3.36s/it]
 22%|██▏       | 3267/14725 [2:59:27<10:24:02,  3.27s/it]
 22%|██▏       | 3268/14725 [2:59:30<10:18:20,  3.24s/it]
 22%|██▏       | 3269
0: {'loss': 0.735, 'grad_norm': 0.44871702277784115, 'learning_rate': 1e-05, 'epoch': 0.22}
0: {'loss': 0.7068, 'grad_norm': 0.47045477194010593, 'learning_rate': 1e-05, 'epoch': 0.22}
0: /14725 [2:59:33<10:21:16,  3.25s/it]
 22%|██▏       | 3270/14725 [2:59:37<10:20:50,  3.25s/it]
                                                         

 22%|██▏       | 3270/14725 [2:59:37<10:20:50,  3.25s/it]
 22%|██▏       | 3271/14725 [2:59:40<10:18:32,  3.24s/it]
 22%|██▏       | 3272/14725 [2:59:43<10:10:48,  3.20s/it]
 22%|██▏       | 3273/14725 [2:59:46<10:05:06,  3.17s/it]
 22%|██▏       | 3274/14725 [2:59:49<10:13:26,  3.21s/it]
 22%|██▏       | 3275/14725 [2:59:53<10:16:10,  3.23s/it]
 22%|██▏       | 3276/14725 [2:59:56<10:12:59,  3.21s/it]
 22%|██▏       | 3277/14725 [2:59:59<10:18:04,  3.24s/it]
 22%|██▏       | 3278/14725 [3:00:02<10:08:37,  3.19s/it]
 22%|██▏       | 3279/14725 [3:00:05<10:06:51,  3.18s/it]
 22%|██▏       | 3280/14725 [3:00:09<10:21:10,  3.26s/it]
                                                         

 22%|██▏       | 3280/14725 [3:00:09<10:21:10,  3.26s/it]
 22%|██▏       | 3281/14725 [3:
0: {'loss': 0.7341, 'grad_norm': 0.4319286308732515, 'learning_rate': 1e-05, 'epoch': 0.22}
0: 00:12<10:37:43,  3.34s/it]
 22%|██▏       | 3282/14725 [3:00:16<10:28:45,  3.30s/it]
 22%|██▏       | 3283/14725 [3:00:19<10:28:25,  3.30s/it]
 22%|██▏       | 3284/14725 [3:00:22<10:21:32,  3.26s/it]
 22%|██▏       | 3285/14725 [3:00:25<10:22:54,  3.27s/it]
 22%|██▏       | 3286/14725 [3:00:29<10:17:45,  3.24s/it]
 22%|██▏       | 3287/14725 [3:00:32<10:13:18,  3.22s/it]
 22%|██▏       | 3288/14725 [3:00:35<10:12:23,  3.21s/it]
 22%|██▏       | 3289/14725 [3:00:38<10:09:40,  3.20s/it]
 22%|██▏       | 3290/14725 [3:00:41<10:10:55,  3.21s/it]
                                                         

 22%|██▏       | 3290/14725 [3:00:41<10:10:55,  3.21s/it]
 22%|██▏       | 3291/14725 [3:00:45<10:14:28,  3.22s/it]
 22%|██▏       | 3292/14725 [3:00:48<10:05:41,  3.18s/it]
 22%|██▏       | 3293/14725 [3:00:51<10:05:46,  3.18s/it]
 22%|██▏       | 3294/14725 [3:00:54<10:06:00,  3.18s/it]
 22%|██▏       | 3295/14725 [3:00:57
0: {'loss': 0.7132, 'grad_norm': 0.4969249813430487, 'learning_rate': 1e-05, 'epoch': 0.22}
0: <10:09:34,  3.20s/it]
 22%|██▏       | 3296/14725 [3:01:00<10:13:29,  3.22s/it]
 22%|██▏       | 3297/14725 [3:01:04<10:20:39,  3.26s/it]
 22%|██▏       | 3298/14725 [3:01:07<10:24:05,  3.28s/it]
 22%|██▏       | 3299/14725 [3:01:10<10:22:31,  3.27s/it]
 22%|██▏       | 3300/14725 [3:01:14<10:33:44,  3.33s/it]
                                                         

 22%|██▏       | 3300/14725 [3:01:14<10:33:44,  3.33s/it]
 22%|██▏       | 3301/14725 [3:01:17<10:31:44,  3.32s/it]
 22%|██▏       | 3302/14725 [3:01:20<10:18:18,  3.25s/it]
 22%|██▏       | 3303/14725 [3:01:24<10:22:24,  3.27s/it]
 22%|██▏       | 3304/14725 [3:01:27<10:21:41,  3.27s/it]
 22%|██▏       | 3305/14725 [3:01:30<10:24:39,  3.28s/it]
 22%|██▏       | 3306/14725 [3:01:33<10:19:29,  3.26s/it]
 22%|██▏       | 3307/14725 [3:01:37<10:22:04,  3.27s/it]
 22%|██▏       | 3308/14725 [3:01:40<10:12:41,  3.22s/it]
 22%|██▏       | 3309/14725 [3:01:43<10:2
0: {'loss': 0.7349, 'grad_norm': 0.44099600530169747, 'learning_rate': 1e-05, 'epoch': 0.22}
0: {'loss': 0.7127, 'grad_norm': 0.41949942923994554, 'learning_rate': 1e-05, 'epoch': 0.23}
0: 0:04,  3.26s/it]
 22%|██▏       | 3310/14725 [3:01:46<10:08:11,  3.20s/it]
                                                         

 22%|██▏       | 3310/14725 [3:01:46<10:08:11,  3.20s/it]
 22%|██▏       | 3311/14725 [3:01:49<10:09:10,  3.20s/it]
 22%|██▏       | 3312/14725 [3:01:53<10:10:31,  3.21s/it]
 22%|██▏       | 3313/14725 [3:01:56<10:03:29,  3.17s/it]
 23%|██▎       | 3314/14725 [3:01:59<10:15:48,  3.24s/it]
 23%|██▎       | 3315/14725 [3:02:02<10:25:22,  3.29s/it]
 23%|██▎       | 3316/14725 [3:02:06<10:35:21,  3.34s/it]
 23%|██▎       | 3317/14725 [3:02:09<10:24:23,  3.28s/it]
 23%|██▎       | 3318/14725 [3:02:12<10:16:32,  3.24s/it]
 23%|██▎       | 3319/14725 [3:02:15<10:16:25,  3.24s/it]
 23%|██▎       | 3320/14725 [3:02:19<10:22:32,  3.28s/it]
                                                         

 23%|██▎       | 3320/14725 [3:02:19<10:22:32,  3.28s/it]
 23%|██▎       | 3321/14725 [3:02:22<10:15:59,  3.2
0: {'loss': 0.7218, 'grad_norm': 0.44557152700432534, 'learning_rate': 1e-05, 'epoch': 0.23}
0: 4s/it]
 23%|██▎       | 3322/14725 [3:02:25<10:15:11,  3.24s/it]
 23%|██▎       | 3323/14725 [3:02:29<10:19:45,  3.26s/it]
 23%|██▎       | 3324/14725 [3:02:32<10:27:05,  3.30s/it]
 23%|██▎       | 3325/14725 [3:02:35<10:34:38,  3.34s/it]
 23%|██▎       | 3326/14725 [3:02:38<10:20:18,  3.27s/it]
 23%|██▎       | 3327/14725 [3:02:42<10:22:35,  3.28s/it]
 23%|██▎       | 3328/14725 [3:02:45<10:24:12,  3.29s/it]
 23%|██▎       | 3329/14725 [3:02:48<10:12:16,  3.22s/it]
 23%|██▎       | 3330/14725 [3:02:51<10:16:05,  3.24s/it]
                                                         

 23%|██▎       | 3330/14725 [3:02:51<10:16:05,  3.24s/it]
 23%|██▎       | 3331/14725 [3:02:55<10:18:32,  3.26s/it]
 23%|██▎       | 3332/14725 [3:02:58<10:30:45,  3.32s/it]
 23%|██▎       | 3333/14725 [3:03:01<10:27:24,  3.30s/it]
 23%|██▎       | 3334/14725 [3:03:05<10:37:07,  3.36s/it]
 23%|██▎       | 3335/14725 [3:03:08<10:25:32,  3.30s/it
0: {'loss': 0.7353, 'grad_norm': 0.4360127657287992, 'learning_rate': 1e-05, 'epoch': 0.23}
0: ]
 23%|██▎       | 3336/14725 [3:03:11<10:26:24,  3.30s/it]
 23%|██▎       | 3337/14725 [3:03:15<10:30:13,  3.32s/it]
 23%|██▎       | 3338/14725 [3:03:18<10:29:34,  3.32s/it]
 23%|██▎       | 3339/14725 [3:03:21<10:26:48,  3.30s/it]
 23%|██▎       | 3340/14725 [3:03:25<10:25:18,  3.30s/it]
                                                         

 23%|██▎       | 3340/14725 [3:03:25<10:25:18,  3.30s/it]
 23%|██▎       | 3341/14725 [3:03:28<10:16:18,  3.25s/it]
 23%|██▎       | 3342/14725 [3:03:31<10:07:45,  3.20s/it]
 23%|██▎       | 3343/14725 [3:03:34<10:02:19,  3.18s/it]
 23%|██▎       | 3344/14725 [3:03:37<10:05:08,  3.19s/it]
 23%|██▎       | 3345/14725 [3:03:40<10:09:00,  3.21s/it]
 23%|██▎       | 3346/14725 [3:03:44<10:15:12,  3.24s/it]
 23%|██▎       | 3347/14725 [3:03:47<10:15:24,  3.25s/it]
 23%|██▎       | 3348/14725 [3:03:51<10:31:10,  3.33s/it]
 23%|██▎       | 3349/14725 [3:03:54<10:19:13,  3.27s/it]
 23
0: {'loss': 0.741, 'grad_norm': 0.4693091080599494, 'learning_rate': 1e-05, 'epoch': 0.23}
0: {'loss': 0.7179, 'grad_norm': 0.46493513379905854, 'learning_rate': 1e-05, 'epoch': 0.23}
0: %|██▎       | 3350/14725 [3:03:57<10:19:56,  3.27s/it]
                                                         

 23%|██▎       | 3350/14725 [3:03:57<10:19:56,  3.27s/it]
 23%|██▎       | 3351/14725 [3:04:00<10:15:06,  3.24s/it]
 23%|██▎       | 3352/14725 [3:04:03<10:19:22,  3.27s/it]
 23%|██▎       | 3353/14725 [3:04:07<10:33:46,  3.34s/it]
 23%|██▎       | 3354/14725 [3:04:10<10:32:56,  3.34s/it]
 23%|██▎       | 3355/14725 [3:04:14<10:34:38,  3.35s/it]
 23%|██▎       | 3356/14725 [3:04:17<10:31:28,  3.33s/it]
 23%|██▎       | 3357/14725 [3:04:20<10:21:06,  3.28s/it]
 23%|██▎       | 3358/14725 [3:04:23<10:06:30,  3.20s/it]
 23%|██▎       | 3359/14725 [3:04:26<9:59:49,  3.17s/it] 
 23%|██▎       | 3360/14725 [3:04:30<10:07:09,  3.21s/it]
                                                         

 23%|██▎       | 3360/14725 [3:04:30<10:07:09,  3.21s/it]
 23%|██▎       | 3361/14725 [3:04:33<10:12:06,  3.23s/it]
 23%|██�
0: {'loss': 0.7202, 'grad_norm': 0.4652199736699196, 'learning_rate': 1e-05, 'epoch': 0.23}
0: �       | 3362/14725 [3:04:36<10:09:32,  3.22s/it]
 23%|██▎       | 3363/14725 [3:04:39<10:17:38,  3.26s/it]
 23%|██▎       | 3364/14725 [3:04:43<10:16:59,  3.26s/it]
 23%|██▎       | 3365/14725 [3:04:46<10:12:47,  3.24s/it]
 23%|██▎       | 3366/14725 [3:04:49<10:25:55,  3.31s/it]
 23%|██▎       | 3367/14725 [3:04:52<10:20:41,  3.28s/it]
 23%|██▎       | 3368/14725 [3:04:56<10:21:36,  3.28s/it]
 23%|██▎       | 3369/14725 [3:04:59<10:43:22,  3.40s/it]
 23%|██▎       | 3370/14725 [3:05:03<10:43:37,  3.40s/it]
                                                         

 23%|██▎       | 3370/14725 [3:05:03<10:43:37,  3.40s/it]
 23%|██▎       | 3371/14725 [3:05:06<10:38:14,  3.37s/it]
 23%|██▎       | 3372/14725 [3:05:09<10:24:34,  3.30s/it]
 23%|██▎       | 3373/14725 [3:05:13<10:33:09,  3.35s/it]
 23%|██▎       | 3374/14725 [3:05:16<10:25:01,  3.30s/it]
 23%|██▎       | 3375/14725 [3:05:19<10:31:03,  3.34s/it]
 23%|██▎    
0: {'loss': 0.7445, 'grad_norm': 0.43341921826911284, 'learning_rate': 1e-05, 'epoch': 0.23}
0:    | 3376/14725 [3:05:23<10:32:17,  3.34s/it]
 23%|██▎       | 3377/14725 [3:05:26<10:22:48,  3.29s/it]
 23%|██▎       | 3378/14725 [3:05:29<10:28:34,  3.32s/it]
 23%|██▎       | 3379/14725 [3:05:33<10:28:19,  3.32s/it]
 23%|██▎       | 3380/14725 [3:05:36<10:24:27,  3.30s/it]
                                                         

 23%|██▎       | 3380/14725 [3:05:36<10:24:27,  3.30s/it]
 23%|██▎       | 3381/14725 [3:05:39<10:23:46,  3.30s/it]
 23%|██▎       | 3382/14725 [3:05:43<10:27:39,  3.32s/it]
 23%|██▎       | 3383/14725 [3:05:46<10:23:44,  3.30s/it]
 23%|██▎       | 3384/14725 [3:05:49<10:39:22,  3.38s/it]
 23%|██▎       | 3385/14725 [3:05:53<10:36:55,  3.37s/it]
 23%|██▎       | 3386/14725 [3:05:56<10:25:00,  3.31s/it]
 23%|██▎       | 3387/14725 [3:05:59<10:19:59,  3.28s/it]
 23%|██▎       | 3388/14725 [3:06:02<10:23:53,  3.30s/it]
 23%|██▎       | 3389/14725 [3:06:06<10:12:16,  3.24s/it]
 23%|██▎       | 
0: {'loss': 0.7229, 'grad_norm': 0.43587388365465646, 'learning_rate': 1e-05, 'epoch': 0.23}
0: {'loss': 0.7175, 'grad_norm': 0.4581812389628328, 'learning_rate': 1e-05, 'epoch': 0.23}
0: 3390/14725 [3:06:09<10:14:45,  3.25s/it]
                                                         

 23%|██▎       | 3390/14725 [3:06:09<10:14:45,  3.25s/it]
 23%|██▎       | 3391/14725 [3:06:12<10:25:20,  3.31s/it]
 23%|██▎       | 3392/14725 [3:06:15<10:16:08,  3.26s/it]
 23%|██▎       | 3393/14725 [3:06:19<10:25:29,  3.31s/it]
 23%|██▎       | 3394/14725 [3:06:22<10:22:00,  3.29s/it]
 23%|██▎       | 3395/14725 [3:06:25<10:25:58,  3.31s/it]
 23%|██▎       | 3396/14725 [3:06:29<10:16:04,  3.26s/it]
 23%|██▎       | 3397/14725 [3:06:32<10:21:31,  3.29s/it]
 23%|██▎       | 3398/14725 [3:06:35<10:20:38,  3.29s/it]
 23%|██▎       | 3399/14725 [3:06:39<10:20:43,  3.29s/it]
 23%|██▎       | 3400/14725 [3:06:42<10:23:37,  3.30s/it]
                                                         

 23%|██▎       | 3400/14725 [3:06:42<10:23:37,  3.30s/it]
 23%|██▎       | 3401/14725 [3:06:45<10:16:52,  3.27s/it]
 23%|██▎       | 3402/14725
0: {'loss': 0.7236, 'grad_norm': 0.4409668372132767, 'learning_rate': 1e-05, 'epoch': 0.23}
0:  [3:06:48<10:12:43,  3.25s/it]
 23%|██▎       | 3403/14725 [3:06:52<10:35:05,  3.37s/it]
 23%|██▎       | 3404/14725 [3:06:55<10:33:29,  3.36s/it]
 23%|██▎       | 3405/14725 [3:06:58<10:27:22,  3.33s/it]
 23%|██▎       | 3406/14725 [3:07:02<10:17:56,  3.28s/it]
 23%|██▎       | 3407/14725 [3:07:05<10:14:19,  3.26s/it]
 23%|██▎       | 3408/14725 [3:07:08<10:15:52,  3.27s/it]
 23%|██▎       | 3409/14725 [3:07:11<10:12:53,  3.25s/it]
 23%|██▎       | 3410/14725 [3:07:14<10:05:29,  3.21s/it]
                                                         

 23%|██▎       | 3410/14725 [3:07:14<10:05:29,  3.21s/it]
 23%|██▎       | 3411/14725 [3:07:18<10:19:52,  3.29s/it]
 23%|██▎       | 3412/14725 [3:07:21<10:20:12,  3.29s/it]
 23%|██▎       | 3413/14725 [3:07:25<10:22:33,  3.30s/it]
 23%|██▎       | 3414/14725 [3:07:28<10:46:23,  3.43s/it]
 23%|██▎       | 3415/14725 [3:07:32<10:43:08,  3.41s/it]
 23%|██▎       | 3416/14725 [3:0
0: {'loss': 0.7298, 'grad_norm': 0.44207667230606557, 'learning_rate': 1e-05, 'epoch': 0.23}
0: 7:35<10:38:59,  3.39s/it]
 23%|██▎       | 3417/14725 [3:07:38<10:34:44,  3.37s/it]
 23%|██▎       | 3418/14725 [3:07:42<10:30:32,  3.35s/it]
 23%|██▎       | 3419/14725 [3:07:45<10:21:34,  3.30s/it]
 23%|██▎       | 3420/14725 [3:07:48<10:07:31,  3.22s/it]
                                                         

 23%|██▎       | 3420/14725 [3:07:48<10:07:31,  3.22s/it]
 23%|██▎       | 3421/14725 [3:07:51<9:58:50,  3.18s/it] 
 23%|██▎       | 3422/14725 [3:07:54<10:01:03,  3.19s/it]
 23%|██▎       | 3423/14725 [3:07:57<10:08:16,  3.23s/it]
 23%|██▎       | 3424/14725 [3:08:01<10:04:09,  3.21s/it]
 23%|██▎       | 3425/14725 [3:08:04<10:11:56,  3.25s/it]
 23%|██▎       | 3426/14725 [3:08:07<10:14:37,  3.26s/it]
 23%|██▎       | 3427/14725 [3:08:10<10:05:29,  3.22s/it]
 23%|██▎       | 3428/14725 [3:08:14<10:09:35,  3.24s/it]
 23%|██▎       | 3429/14725 [3:08:17<10:15:27,  3.27s/it]
 23%|██▎       | 3430/14725 [3:08:20<
0: {'loss': 0.7452, 'grad_norm': 0.4721650793175216, 'learning_rate': 1e-05, 'epoch': 0.23}
0: {'loss': 0.7306, 'grad_norm': 0.4375360356101152, 'learning_rate': 1e-05, 'epoch': 0.23}
0: 10:28:27,  3.34s/it]
                                                         

 23%|██▎       | 3430/14725 [3:08:20<10:28:27,  3.34s/it]
 23%|██▎       | 3431/14725 [3:08:24<10:26:36,  3.33s/it]
 23%|██▎       | 3432/14725 [3:08:27<10:18:30,  3.29s/it]
 23%|██▎       | 3433/14725 [3:08:30<10:30:01,  3.35s/it]
 23%|██▎       | 3434/14725 [3:08:34<10:33:25,  3.37s/it]
 23%|██▎       | 3435/14725 [3:08:37<10:34:36,  3.37s/it]
 23%|██▎       | 3436/14725 [3:08:40<10:25:12,  3.32s/it]
 23%|██▎       | 3437/14725 [3:08:44<10:26:35,  3.33s/it]
 23%|██▎       | 3438/14725 [3:08:47<10:19:06,  3.29s/it]
 23%|██▎       | 3439/14725 [3:08:50<10:20:07,  3.30s/it]
 23%|██▎       | 3440/14725 [3:08:54<10:21:45,  3.31s/it]
                                                         

 23%|██▎       | 3440/14725 [3:08:54<10:21:45,  3.31s/it]
 23%|██▎       | 3441/14725 [3:08:57<10:12:00,  3.25s/it]
 23%|██▎       | 3442/14725 [3:09:00<10:09:32, 
0: {'loss': 0.7284, 'grad_norm': 0.466335135040074, 'learning_rate': 1e-05, 'epoch': 0.23}
0:  3.24s/it]
 23%|██▎       | 3443/14725 [3:09:03<10:19:28,  3.29s/it]
 23%|██▎       | 3444/14725 [3:09:07<10:08:31,  3.24s/it]
 23%|██▎       | 3445/14725 [3:09:10<10:13:04,  3.26s/it]
 23%|██▎       | 3446/14725 [3:09:13<10:29:27,  3.35s/it]
 23%|██▎       | 3447/14725 [3:09:16<10:14:36,  3.27s/it]
 23%|██▎       | 3448/14725 [3:09:20<10:15:23,  3.27s/it]
 23%|██▎       | 3449/14725 [3:09:23<10:02:56,  3.21s/it]
 23%|██▎       | 3450/14725 [3:09:26<10:00:20,  3.19s/it]
                                                         

 23%|██▎       | 3450/14725 [3:09:26<10:00:20,  3.19s/it]
 23%|██▎       | 3451/14725 [3:09:29<10:09:56,  3.25s/it]
 23%|██▎       | 3452/14725 [3:09:33<10:19:43,  3.30s/it]
 23%|██▎       | 3453/14725 [3:09:36<10:32:44,  3.37s/it]
 23%|██▎       | 3454/14725 [3:09:40<10:39:46,  3.41s/it]
 23%|██▎       | 3455/14725 [3:09:43<10:23:45,  3.32s/it]
 23%|██▎       | 3456/14725 [3:09:46<10:21:24,  3.31
0: {'loss': 0.721, 'grad_norm': 0.43612413099128894, 'learning_rate': 1e-05, 'epoch': 0.23}
0: s/it]
 23%|██▎       | 3457/14725 [3:09:49<10:16:09,  3.28s/it]
 23%|██▎       | 3458/14725 [3:09:53<10:12:35,  3.26s/it]
 23%|██▎       | 3459/14725 [3:09:56<10:15:13,  3.28s/it]
 23%|██▎       | 3460/14725 [3:09:59<10:21:19,  3.31s/it]
                                                         

 23%|██▎       | 3460/14725 [3:09:59<10:21:19,  3.31s/it]
 24%|██▎       | 3461/14725 [3:10:03<10:16:19,  3.28s/it]
 24%|██▎       | 3462/14725 [3:10:06<10:09:06,  3.24s/it]
 24%|██▎       | 3463/14725 [3:10:09<10:10:21,  3.25s/it]
 24%|██▎       | 3464/14725 [3:10:12<10:16:34,  3.29s/it]
 24%|██▎       | 3465/14725 [3:10:16<10:27:08,  3.34s/it]
 24%|██▎       | 3466/14725 [3:10:19<10:24:58,  3.33s/it]
 24%|██▎       | 3467/14725 [3:10:23<10:38:04,  3.40s/it]
 24%|██▎       | 3468/14725 [3:10:26<10:27:44,  3.35s/it]
 24%|██▎       | 3469/14725 [3:10:29<10:16:31,  3.29s/it]
 24%|██▎       | 3470/14725 [3:10:32<10:04:40,  3.22s/it]
0: {'loss': 0.7154, 'grad_norm': 0.47896660117278395, 'learning_rate': 1e-05, 'epoch': 0.24}
0: {'loss': 0.7139, 'grad_norm': 0.46965723829149036, 'learning_rate': 1e-05, 'epoch': 0.24}
0: 
                                                         

 24%|██▎       | 3470/14725 [3:10:32<10:04:40,  3.22s/it]
 24%|██▎       | 3471/14725 [3:10:36<10:22:58,  3.32s/it]
 24%|██▎       | 3472/14725 [3:10:39<10:25:49,  3.34s/it]
 24%|██▎       | 3473/14725 [3:10:42<10:13:27,  3.27s/it]
 24%|██▎       | 3474/14725 [3:10:46<10:25:23,  3.34s/it]
 24%|██▎       | 3475/14725 [3:10:49<10:22:34,  3.32s/it]
 24%|██▎       | 3476/14725 [3:10:52<10:22:23,  3.32s/it]
 24%|██▎       | 3477/14725 [3:10:56<10:18:48,  3.30s/it]
 24%|██▎       | 3478/14725 [3:10:59<10:13:07,  3.27s/it]
 24%|██▎       | 3479/14725 [3:11:02<10:03:44,  3.22s/it]
 24%|██▎       | 3480/14725 [3:11:05<10:06:17,  3.23s/it]
                                                         

 24%|██▎       | 3480/14725 [3:11:05<10:06:17,  3.23s/it]
 24%|██▎       | 3481/14725 [3:11:08<10:06:02,  3.23s/it]
 24%|██▎       | 3482/14725 [3:11:12<10:09:39,  3.25s/it]
 24%|█�
0: {'loss': 0.714, 'grad_norm': 0.4406659313520881, 'learning_rate': 1e-05, 'epoch': 0.24}
0: ��▎       | 3483/14725 [3:11:15<10:08:11,  3.25s/it]
 24%|██▎       | 3484/14725 [3:11:18<10:02:27,  3.22s/it]
 24%|██▎       | 3485/14725 [3:11:21<9:54:48,  3.18s/it] 
 24%|██▎       | 3486/14725 [3:11:24<9:56:02,  3.18s/it]
 24%|██▎       | 3487/14725 [3:11:28<10:13:20,  3.27s/it]
 24%|██▎       | 3488/14725 [3:11:31<10:11:42,  3.27s/it]
 24%|██▎       | 3489/14725 [3:11:34<10:18:58,  3.31s/it]
 24%|██▎       | 3490/14725 [3:11:38<10:11:35,  3.27s/it]
                                                         

 24%|██▎       | 3490/14725 [3:11:38<10:11:35,  3.27s/it]
 24%|██▎       | 3491/14725 [3:11:41<10:06:02,  3.24s/it]
 24%|██▎       | 3492/14725 [3:11:44<10:01:42,  3.21s/it]
 24%|██▎       | 3493/14725 [3:11:47<9:56:13,  3.18s/it] 
 24%|██▎       | 3494/14725 [3:11:50<10:01:06,  3.21s/it]
 24%|██▎       | 3495/14725 [3:11:53<10:00:19,  3.21s/it]
 24%|██▎       | 3496/14725 [3:11:57<10:02:29,  3.22s/it]
 24%|██▎ 
0: {'loss': 0.7211, 'grad_norm': 0.4923058067050355, 'learning_rate': 1e-05, 'epoch': 0.24}
0:       | 3497/14725 [3:12:00<10:02:56,  3.22s/it]
 24%|██▍       | 3498/14725 [3:12:03<10:14:22,  3.28s/it]
 24%|██▍       | 3499/14725 [3:12:07<10:19:04,  3.31s/it]
 24%|██▍       | 3500/14725 [3:12:10<10:28:22,  3.36s/it]
                                                         

 24%|██▍       | 3500/14725 [3:12:10<10:28:22,  3.36s/it]
 24%|██▍       | 3501/14725 [3:12:13<10:22:27,  3.33s/it]
 24%|██▍       | 3502/14725 [3:12:17<10:30:33,  3.37s/it]
 24%|██▍       | 3503/14725 [3:12:20<10:29:16,  3.36s/it]
 24%|██▍       | 3504/14725 [3:12:24<10:41:01,  3.43s/it]
 24%|██▍       | 3505/14725 [3:12:27<10:26:50,  3.35s/it]
 24%|██▍       | 3506/14725 [3:12:30<10:10:52,  3.27s/it]
 24%|██▍       | 3507/14725 [3:12:33<10:06:14,  3.24s/it]
 24%|██▍       | 3508/14725 [3:12:37<10:08:13,  3.25s/it]
 24%|██▍       | 3509/14725 [3:12:40<10:04:58,  3.24s/it]
 24%|██▍       | 3510/14725 [3:12:43<10:05:20,  3.24s/it]
                    
0: {'loss': 0.7302, 'grad_norm': 0.4248136947716915, 'learning_rate': 1e-05, 'epoch': 0.24}
0: {'loss': 0.7253, 'grad_norm': 0.4497058145914851, 'learning_rate': 1e-05, 'epoch': 0.24}
0:                                      

 24%|██▍       | 3510/14725 [3:12:43<10:05:20,  3.24s/it]
 24%|██▍       | 3511/14725 [3:12:46<10:15:39,  3.29s/it]
 24%|██▍       | 3512/14725 [3:12:50<10:07:12,  3.25s/it]
 24%|██▍       | 3513/14725 [3:12:53<9:57:43,  3.20s/it] 
 24%|██▍       | 3514/14725 [3:12:56<10:13:48,  3.29s/it]
 24%|██▍       | 3515/14725 [3:12:59<10:14:34,  3.29s/it]
 24%|██▍       | 3516/14725 [3:13:03<10:18:55,  3.31s/it]
 24%|██▍       | 3517/14725 [3:13:06<10:07:17,  3.25s/it]
 24%|██▍       | 3518/14725 [3:13:09<10:02:56,  3.23s/it]
 24%|██▍       | 3519/14725 [3:13:12<10:04:42,  3.24s/it]
 24%|██▍       | 3520/14725 [3:13:16<9:58:36,  3.21s/it] 
                                                        

 24%|██▍       | 3520/14725 [3:13:16<9:58:36,  3.21s/it]
 24%|██▍       | 3521/14725 [3:13:19<9:58:04,  3.20s/it]
 24%|██▍       | 3522/14725 [3:13:22<9:56:44,  3.20s/it]
 24%|██▍       | 3523/14725 
0: {'loss': 0.7258, 'grad_norm': 0.4588102963926079, 'learning_rate': 1e-05, 'epoch': 0.24}
0: [3:13:25<9:53:59,  3.18s/it]
 24%|██▍       | 3524/14725 [3:13:28<9:59:10,  3.21s/it]
 24%|██▍       | 3525/14725 [3:13:32<9:58:19,  3.21s/it]
 24%|██▍       | 3526/14725 [3:13:35<10:10:12,  3.27s/it]
 24%|██▍       | 3527/14725 [3:13:38<10:20:27,  3.32s/it]
 24%|██▍       | 3528/14725 [3:13:42<10:15:05,  3.30s/it]
 24%|██▍       | 3529/14725 [3:13:45<10:09:34,  3.27s/it]
 24%|██▍       | 3530/14725 [3:13:48<10:02:17,  3.23s/it]
                                                         

 24%|██▍       | 3530/14725 [3:13:48<10:02:17,  3.23s/it]
 24%|██▍       | 3531/14725 [3:13:51<9:54:44,  3.19s/it] 
 24%|██▍       | 3532/14725 [3:13:54<9:49:37,  3.16s/it]
 24%|██▍       | 3533/14725 [3:13:57<9:58:25,  3.21s/it]
 24%|██▍       | 3534/14725 [3:14:01<9:57:19,  3.20s/it]
 24%|██▍       | 3535/14725 [3:14:04<9:58:41,  3.21s/it]
 24%|██▍       | 3536/14725 [3:14:07<9:51:47,  3.17s/it]
 24%|██▍       | 3537/14725 [3:14:10<9:57
0: {'loss': 0.7327, 'grad_norm': 0.45639898507678134, 'learning_rate': 1e-05, 'epoch': 0.24}
0: {'loss': 0.7243, 'grad_norm': 0.41455612878579295, 'learning_rate': 1e-05, 'epoch': 0.24}
0: :53,  3.21s/it]
 24%|██▍       | 3538/14725 [3:14:13<9:58:35,  3.21s/it]
 24%|██▍       | 3539/14725 [3:14:17<9:58:57,  3.21s/it]
 24%|██▍       | 3540/14725 [3:14:20<9:53:33,  3.18s/it]
                                                        

 24%|██▍       | 3540/14725 [3:14:20<9:53:33,  3.18s/it]
 24%|██▍       | 3541/14725 [3:14:23<9:58:25,  3.21s/it]
 24%|██▍       | 3542/14725 [3:14:26<10:02:13,  3.23s/it]
 24%|██▍       | 3543/14725 [3:14:30<10:01:26,  3.23s/it]
 24%|██▍       | 3544/14725 [3:14:33<9:58:28,  3.21s/it] 
 24%|██▍       | 3545/14725 [3:14:36<10:07:47,  3.26s/it]
 24%|██▍       | 3546/14725 [3:14:40<10:15:37,  3.30s/it]
 24%|██▍       | 3547/14725 [3:14:43<10:14:17,  3.30s/it]
 24%|██▍       | 3548/14725 [3:14:46<10:13:44,  3.29s/it]
 24%|██▍       | 3549/14725 [3:14:49<10:10:59,  3.28s/it]
 24%|██▍       | 3550/14725 [3:14:53<10:05:38,  3.25s/it]
                                                         

0: {'loss': 0.7239, 'grad_norm': 0.44186619111283626, 'learning_rate': 1e-05, 'epoch': 0.24}
0:  24%|██▍       | 3550/14725 [3:14:53<10:05:38,  3.25s/it]
 24%|██▍       | 3551/14725 [3:14:56<10:10:36,  3.28s/it]
 24%|██▍       | 3552/14725 [3:14:59<10:01:42,  3.23s/it]
 24%|██▍       | 3553/14725 [3:15:02<10:04:11,  3.24s/it]
 24%|██▍       | 3554/14725 [3:15:05<9:59:40,  3.22s/it] 
 24%|██▍       | 3555/14725 [3:15:08<9:50:12,  3.17s/it]
 24%|██▍       | 3556/14725 [3:15:12<9:59:04,  3.22s/it]
 24%|██▍       | 3557/14725 [3:15:15<10:15:39,  3.31s/it]
 24%|██▍       | 3558/14725 [3:15:19<10:15:40,  3.31s/it]
 24%|██▍       | 3559/14725 [3:15:22<10:04:27,  3.25s/it]
 24%|██▍       | 3560/14725 [3:15:25<9:57:15,  3.21s/it] 
                                                        

 24%|██▍       | 3560/14725 [3:15:25<9:57:15,  3.21s/it]
 24%|██▍       | 3561/14725 [3:15:28<9:50:43,  3.17s/it]
 24%|██▍       | 3562/14725 [3:15:31<10:04:04,  3.25s/it]
 24%|██▍       | 3563/14725 [3:15:35<10:06:45,  3.26s/it]
 24%|█�
0: {'loss': 0.7258, 'grad_norm': 0.49042549419198367, 'learning_rate': 1e-05, 'epoch': 0.24}
0: �▍       | 3564/14725 [3:15:38<10:08:59,  3.27s/it]
 24%|██▍       | 3565/14725 [3:15:41<10:03:04,  3.24s/it]
 24%|██▍       | 3566/14725 [3:15:44<10:08:21,  3.27s/it]
 24%|██▍       | 3567/14725 [3:15:48<10:10:02,  3.28s/it]
 24%|██▍       | 3568/14725 [3:15:51<10:03:57,  3.25s/it]
 24%|██▍       | 3569/14725 [3:15:54<10:00:59,  3.23s/it]
 24%|██▍       | 3570/14725 [3:15:57<9:56:31,  3.21s/it] 
                                                        

 24%|██▍       | 3570/14725 [3:15:57<9:56:31,  3.21s/it]
 24%|██▍       | 3571/14725 [3:16:01<9:59:30,  3.22s/it]
 24%|██▍       | 3572/14725 [3:16:04<9:50:45,  3.18s/it]
 24%|██▍       | 3573/14725 [3:16:07<10:00:21,  3.23s/it]
 24%|██▍       | 3574/14725 [3:16:10<10:03:52,  3.25s/it]
 24%|██▍       | 3575/14725 [3:16:13<10:00:22,  3.23s/it]
 24%|██▍       | 3576/14725 [3:16:17<10:12:31,  3.30s/it]
 24%|██▍       | 3577/14725 [3:16:20<10:14:37,  3.31s/it]
 24%|██▍     
0: {'loss': 0.7165, 'grad_norm': 0.4721738770670584, 'learning_rate': 1e-05, 'epoch': 0.24}
0: {'loss': 0.7148, 'grad_norm': 0.4347345441891485, 'learning_rate': 1e-05, 'epoch': 0.24}
0:   | 3578/14725 [3:16:24<10:15:56,  3.32s/it]
 24%|██▍       | 3579/14725 [3:16:27<10:18:51,  3.33s/it]
 24%|██▍       | 3580/14725 [3:16:30<10:12:32,  3.30s/it]
                                                         

 24%|██▍       | 3580/14725 [3:16:30<10:12:32,  3.30s/it]
 24%|██▍       | 3581/14725 [3:16:33<10:14:10,  3.31s/it]
 24%|██▍       | 3582/14725 [3:16:37<10:09:07,  3.28s/it]
 24%|██▍       | 3583/14725 [3:16:40<10:04:11,  3.25s/it]
 24%|██▍       | 3584/14725 [3:16:43<10:04:01,  3.25s/it]
 24%|██▍       | 3585/14725 [3:16:46<9:54:33,  3.20s/it] 
 24%|██▍       | 3586/14725 [3:16:50<10:11:04,  3.29s/it]
 24%|██▍       | 3587/14725 [3:16:53<10:01:02,  3.24s/it]
 24%|██▍       | 3588/14725 [3:16:56<10:03:38,  3.25s/it]
 24%|██▍       | 3589/14725 [3:16:59<10:08:56,  3.28s/it]
 24%|██▍       | 3590/14725 [3:17:03<10:07:28,  3.27s/it]
                                                         

 24%|██▍       | 3590/1
0: {'loss': 0.7301, 'grad_norm': 0.44947959275048444, 'learning_rate': 1e-05, 'epoch': 0.24}
0: 4725 [3:17:03<10:07:28,  3.27s/it]
 24%|██▍       | 3591/14725 [3:17:06<10:09:21,  3.28s/it]
 24%|██▍       | 3592/14725 [3:17:09<10:10:55,  3.29s/it]
 24%|██▍       | 3593/14725 [3:17:13<10:09:03,  3.28s/it]
 24%|██▍       | 3594/14725 [3:17:16<10:19:30,  3.34s/it]
 24%|██▍       | 3595/14725 [3:17:19<10:19:19,  3.34s/it]
 24%|██▍       | 3596/14725 [3:17:23<10:25:48,  3.37s/it]
 24%|██▍       | 3597/14725 [3:17:26<10:13:36,  3.31s/it]
 24%|██▍       | 3598/14725 [3:17:29<10:19:14,  3.34s/it]
 24%|██▍       | 3599/14725 [3:17:33<10:11:43,  3.30s/it]
 24%|██▍       | 3600/14725 [3:17:36<10:21:08,  3.35s/it]
                                                         

 24%|██▍       | 3600/14725 [3:17:36<10:21:08,  3.35s/it]
 24%|██▍       | 3601/14725 [3:17:39<10:10:49,  3.29s/it]
 24%|██▍       | 3602/14725 [3:17:42<10:02:01,  3.25s/it]
 24%|██▍       | 3603/14725 [3:17:46<10:05:39,  3.27s/it]
 24%|██▍       | 3604/14725 
0: {'loss': 0.7221, 'grad_norm': 0.47004115222664566, 'learning_rate': 1e-05, 'epoch': 0.25}
0: [3:17:49<10:10:57,  3.30s/it]
 24%|██▍       | 3605/14725 [3:17:52<10:01:39,  3.25s/it]
 24%|██▍       | 3606/14725 [3:17:56<10:06:23,  3.27s/it]
 24%|██▍       | 3607/14725 [3:17:59<10:03:48,  3.26s/it]
 25%|██▍       | 3608/14725 [3:18:02<10:06:05,  3.27s/it]
 25%|██▍       | 3609/14725 [3:18:05<10:08:00,  3.28s/it]
 25%|██▍       | 3610/14725 [3:18:09<10:06:03,  3.27s/it]
                                                         

 25%|██▍       | 3610/14725 [3:18:09<10:06:03,  3.27s/it]
 25%|██▍       | 3611/14725 [3:18:12<10:06:14,  3.27s/it]
 25%|██▍       | 3612/14725 [3:18:15<10:03:11,  3.26s/it]
 25%|██▍       | 3613/14725 [3:18:18<10:06:51,  3.28s/it]
 25%|██▍       | 3614/14725 [3:18:22<10:00:22,  3.24s/it]
 25%|██▍       | 3615/14725 [3:18:25<10:02:27,  3.25s/it]
 25%|██▍       | 3616/14725 [3:18:28<10:02:03,  3.25s/it]
 25%|██▍       | 3617/14725 [3:18:31<9:59:54,  3.24s/it] 
 25%|██▍       | 3618/14725 [3:18
0: {'loss': 0.7307, 'grad_norm': 0.4256930733971692, 'learning_rate': 1e-05, 'epoch': 0.25}
0: {'loss': 0.7172, 'grad_norm': 0.45572808864761266, 'learning_rate': 1e-05, 'epoch': 0.25}
0: :35<9:57:01,  3.23s/it]
 25%|██▍       | 3619/14725 [3:18:38<9:59:15,  3.24s/it]
 25%|██▍       | 3620/14725 [3:18:41<10:07:14,  3.28s/it]
                                                         

 25%|██▍       | 3620/14725 [3:18:41<10:07:14,  3.28s/it]
 25%|██▍       | 3621/14725 [3:18:45<10:21:47,  3.36s/it]
 25%|██▍       | 3622/14725 [3:18:48<10:10:08,  3.30s/it]
 25%|██▍       | 3623/14725 [3:18:51<10:05:24,  3.27s/it]
 25%|██▍       | 3624/14725 [3:18:54<10:01:06,  3.25s/it]
 25%|██▍       | 3625/14725 [3:18:58<10:02:15,  3.26s/it]
 25%|██▍       | 3626/14725 [3:19:01<10:10:45,  3.30s/it]
 25%|██▍       | 3627/14725 [3:19:04<10:01:19,  3.25s/it]
 25%|██▍       | 3628/14725 [3:19:07<9:58:38,  3.24s/it] 
 25%|██▍       | 3629/14725 [3:19:11<10:00:43,  3.25s/it]
 25%|██▍       | 3630/14725 [3:19:14<9:49:06,  3.19s/it] 
                                                        

 25%|██▍       | 3630/14725 [3:19:14<9:49:06, 
0: {'loss': 0.7224, 'grad_norm': 0.45703794566657485, 'learning_rate': 1e-05, 'epoch': 0.25}
0:  3.19s/it]
 25%|██▍       | 3631/14725 [3:19:17<9:52:04,  3.20s/it]
 25%|██▍       | 3632/14725 [3:19:20<10:03:54,  3.27s/it]
 25%|██▍       | 3633/14725 [3:19:24<10:01:28,  3.25s/it]
 25%|██▍       | 3634/14725 [3:19:27<10:02:00,  3.26s/it]
 25%|██▍       | 3635/14725 [3:19:30<10:06:57,  3.28s/it]
 25%|██▍       | 3636/14725 [3:19:33<10:08:54,  3.29s/it]
 25%|██▍       | 3637/14725 [3:19:37<10:17:20,  3.34s/it]
 25%|██▍       | 3638/14725 [3:19:40<10:17:13,  3.34s/it]
 25%|██▍       | 3639/14725 [3:19:44<10:13:09,  3.32s/it]
 25%|██▍       | 3640/14725 [3:19:47<10:13:35,  3.32s/it]
                                                         

 25%|██▍       | 3640/14725 [3:19:47<10:13:35,  3.32s/it]
 25%|██▍       | 3641/14725 [3:19:50<10:08:17,  3.29s/it]
 25%|██▍       | 3642/14725 [3:19:53<9:58:27,  3.24s/it] 
 25%|██▍       | 3643/14725 [3:19:56<9:55:03,  3.22s/it]
 25%|██▍       | 3644/14725 [3:20:00<9:56:34,  3.23s/i
0: {'loss': 0.7297, 'grad_norm': 0.4786431092082026, 'learning_rate': 1e-05, 'epoch': 0.25}
0: t]
 25%|██▍       | 3645/14725 [3:20:03<9:50:37,  3.20s/it]
 25%|██▍       | 3646/14725 [3:20:06<9:54:06,  3.22s/it]
 25%|██▍       | 3647/14725 [3:20:09<9:47:07,  3.18s/it]
 25%|██▍       | 3648/14725 [3:20:12<9:46:14,  3.18s/it]
 25%|██▍       | 3649/14725 [3:20:16<9:51:19,  3.20s/it]
 25%|██▍       | 3650/14725 [3:20:19<9:56:18,  3.23s/it]
                                                        

 25%|██▍       | 3650/14725 [3:20:19<9:56:18,  3.23s/it]
 25%|██▍       | 3651/14725 [3:20:22<9:50:28,  3.20s/it]
 25%|██▍       | 3652/14725 [3:20:25<9:59:45,  3.25s/it]
 25%|██▍       | 3653/14725 [3:20:29<9:56:49,  3.23s/it]
 25%|██▍       | 3654/14725 [3:20:32<10:06:19,  3.29s/it]
 25%|██▍       | 3655/14725 [3:20:35<10:04:50,  3.28s/it]
 25%|██▍       | 3656/14725 [3:20:38<10:00:20,  3.25s/it]
 25%|██▍       | 3657/14725 [3:20:42<10:19:36,  3.36s/it]
 25%|██▍       | 3658/14725 [3:20:45<10:06:37,  3.29s/it]
 25%|██�
0: {'loss': 0.7303, 'grad_norm': 0.4780557233445008, 'learning_rate': 1e-05, 'epoch': 0.25}
0: {'loss': 0.7338, 'grad_norm': 0.4806247514372714, 'learning_rate': 1e-05, 'epoch': 0.25}
0: �       | 3659/14725 [3:20:48<9:59:17,  3.25s/it] 
 25%|██▍       | 3660/14725 [3:20:52<10:05:09,  3.28s/it]
                                                         

 25%|██▍       | 3660/14725 [3:20:52<10:05:09,  3.28s/it]
 25%|██▍       | 3661/14725 [3:20:55<10:09:30,  3.31s/it]
 25%|██▍       | 3662/14725 [3:20:58<10:02:50,  3.27s/it]
 25%|██▍       | 3663/14725 [3:21:01<10:00:33,  3.26s/it]
 25%|██▍       | 3664/14725 [3:21:05<9:59:34,  3.25s/it] 
 25%|██▍       | 3665/14725 [3:21:08<10:04:31,  3.28s/it]
 25%|██▍       | 3666/14725 [3:21:11<9:58:14,  3.25s/it] 
 25%|██▍       | 3667/14725 [3:21:14<9:58:24,  3.25s/it]
 25%|██▍       | 3668/14725 [3:21:18<9:53:46,  3.22s/it]
 25%|██▍       | 3669/14725 [3:21:21<9:56:49,  3.24s/it]
 25%|██▍       | 3670/14725 [3:21:24<10:09:04,  3.31s/it]
                                                         

 25%|██▍       | 3670/14725 [3:21:24<10:09:04,  3.31s/it]
 25%|██▍       | 367
0: {'loss': 0.7165, 'grad_norm': 0.4425965649036334, 'learning_rate': 1e-05, 'epoch': 0.25}
0: 1/14725 [3:21:28<10:04:08,  3.28s/it]
 25%|██▍       | 3672/14725 [3:21:31<10:04:21,  3.28s/it]
 25%|██▍       | 3673/14725 [3:21:34<9:54:58,  3.23s/it] 
 25%|██▍       | 3674/14725 [3:21:37<9:58:10,  3.25s/it]
 25%|██▍       | 3675/14725 [3:21:40<9:51:35,  3.21s/it]
 25%|██▍       | 3676/14725 [3:21:43<9:41:53,  3.16s/it]
 25%|██▍       | 3677/14725 [3:21:47<9:44:11,  3.17s/it]
 25%|██▍       | 3678/14725 [3:21:50<9:55:38,  3.24s/it]
 25%|██▍       | 3679/14725 [3:21:53<9:59:02,  3.25s/it]
 25%|██▍       | 3680/14725 [3:21:56<9:47:14,  3.19s/it]
                                                        

 25%|██▍       | 3680/14725 [3:21:56<9:47:14,  3.19s/it]
 25%|██▍       | 3681/14725 [3:22:00<10:04:46,  3.29s/it]
 25%|██▌       | 3682/14725 [3:22:03<10:05:06,  3.29s/it]
 25%|██▌       | 3683/14725 [3:22:06<9:59:33,  3.26s/it] 
 25%|██▌       | 3684/14725 [3:22:10<10:02:21,  3.27s/it]
 25%|██▌       | 3685/14725 [3:22:
0: {'loss': 0.7457, 'grad_norm': 0.4204921829615334, 'learning_rate': 1e-05, 'epoch': 0.25}
0: 13<10:03:54,  3.28s/it]
 25%|██▌       | 3686/14725 [3:22:16<10:08:08,  3.31s/it]
 25%|██▌       | 3687/14725 [3:22:20<10:15:20,  3.34s/it]
 25%|██▌       | 3688/14725 [3:22:23<10:07:58,  3.31s/it]
 25%|██▌       | 3689/14725 [3:22:26<10:04:57,  3.29s/it]
 25%|██▌       | 3690/14725 [3:22:29<10:01:25,  3.27s/it]
                                                         

 25%|██▌       | 3690/14725 [3:22:29<10:01:25,  3.27s/it]
 25%|██▌       | 3691/14725 [3:22:33<9:59:57,  3.26s/it] 
 25%|██▌       | 3692/14725 [3:22:36<9:53:39,  3.23s/it]
 25%|██▌       | 3693/14725 [3:22:39<9:57:49,  3.25s/it]
 25%|██▌       | 3694/14725 [3:22:42<9:54:25,  3.23s/it]
 25%|██▌       | 3695/14725 [3:22:45<9:51:10,  3.22s/it]
 25%|██▌       | 3696/14725 [3:22:49<9:53:27,  3.23s/it]
 25%|██▌       | 3697/14725 [3:22:52<9:55:26,  3.24s/it]
 25%|██▌       | 3698/14725 [3:22:55<9:52:42,  3.23s/it]
 25%|██▌       | 3699/14725 [3:22:58<9:54:34, 
0: {'loss': 0.7383, 'grad_norm': 0.41168343271874275, 'learning_rate': 1e-05, 'epoch': 0.25}
0: {'loss': 0.7257, 'grad_norm': 0.46624548435848323, 'learning_rate': 1e-05, 'epoch': 0.25}
0:  3.24s/it]
 25%|██▌       | 3700/14725 [3:23:02<9:48:00,  3.20s/it]
                                                        

 25%|██▌       | 3700/14725 [3:23:02<9:48:00,  3.20s/it]
 25%|██▌       | 3701/14725 [3:23:05<9:56:09,  3.24s/it]
 25%|██▌       | 3702/14725 [3:23:08<9:48:58,  3.21s/it]
 25%|██▌       | 3703/14725 [3:23:11<10:02:19,  3.28s/it]
 25%|██▌       | 3704/14725 [3:23:15<9:54:29,  3.24s/it] 
 25%|██▌       | 3705/14725 [3:23:18<9:56:13,  3.25s/it]
 25%|██▌       | 3706/14725 [3:23:21<10:01:00,  3.27s/it]
 25%|██▌       | 3707/14725 [3:23:25<10:04:27,  3.29s/it]
 25%|██▌       | 3708/14725 [3:23:28<10:08:50,  3.32s/it]
 25%|██▌       | 3709/14725 [3:23:31<10:04:42,  3.29s/it]
 25%|██▌       | 3710/14725 [3:23:35<10:08:20,  3.31s/it]
                                                         

 25%|██▌       | 3710/14725 [3:23:35<10:08:20,  3.31s/it]
 25%|██▌       | 3711/14725 [3:23:38<10:06:28,  3.30s/it]
 25%|
0: {'loss': 0.7131, 'grad_norm': 0.44745887733011086, 'learning_rate': 1e-05, 'epoch': 0.25}
0: ██▌       | 3712/14725 [3:23:41<9:59:37,  3.27s/it] 
 25%|██▌       | 3713/14725 [3:23:44<9:55:09,  3.24s/it]
 25%|██▌       | 3714/14725 [3:23:47<9:50:44,  3.22s/it]
 25%|██▌       | 3715/14725 [3:23:50<9:44:55,  3.19s/it]
 25%|██▌       | 3716/14725 [3:23:54<9:46:39,  3.20s/it]
 25%|██▌       | 3717/14725 [3:23:57<9:52:16,  3.23s/it]
 25%|██▌       | 3718/14725 [3:24:00<9:51:39,  3.23s/it]
 25%|██▌       | 3719/14725 [3:24:03<9:54:49,  3.24s/it]
 25%|██▌       | 3720/14725 [3:24:07<9:53:04,  3.23s/it]
                                                        

 25%|██▌       | 3720/14725 [3:24:07<9:53:04,  3.23s/it]
 25%|██▌       | 3721/14725 [3:24:10<9:54:15,  3.24s/it]
 25%|██▌       | 3722/14725 [3:24:13<9:48:52,  3.21s/it]
 25%|██▌       | 3723/14725 [3:24:16<9:42:36,  3.18s/it]
 25%|██▌       | 3724/14725 [3:24:19<9:39:59,  3.16s/it]
 25%|██▌       | 3725/14725 [3:24:23<9:46:09,  3.20s/it]
 25%|██▌       | 37
0: {'loss': 0.7249, 'grad_norm': 0.4581827682391358, 'learning_rate': 1e-05, 'epoch': 0.25}
0: 26/14725 [3:24:26<9:40:37,  3.17s/it]
 25%|██▌       | 3727/14725 [3:24:29<9:45:04,  3.19s/it]
 25%|██▌       | 3728/14725 [3:24:32<9:55:05,  3.25s/it]
 25%|██▌       | 3729/14725 [3:24:36<9:57:18,  3.26s/it]
 25%|██▌       | 3730/14725 [3:24:39<10:04:57,  3.30s/it]
                                                         

 25%|██▌       | 3730/14725 [3:24:39<10:04:57,  3.30s/it]
 25%|██▌       | 3731/14725 [3:24:42<9:56:09,  3.25s/it] 
 25%|██▌       | 3732/14725 [3:24:45<9:53:17,  3.24s/it]
 25%|██▌       | 3733/14725 [3:24:49<9:54:25,  3.24s/it]
 25%|██▌       | 3734/14725 [3:24:52<9:52:12,  3.23s/it]
 25%|██▌       | 3735/14725 [3:24:55<9:46:19,  3.20s/it]
 25%|██▌       | 3736/14725 [3:24:58<9:47:52,  3.21s/it]
 25%|██▌       | 3737/14725 [3:25:01<9:46:35,  3.20s/it]
 25%|██▌       | 3738/14725 [3:25:05<9:55:07,  3.25s/it]
 25%|██▌       | 3739/14725 [3:25:08<9:59:32,  3.27s/it]
 25%|██▌       | 3740/14725 [3:25:11
0: {'loss': 0.7325, 'grad_norm': 0.45279496563243915, 'learning_rate': 1e-05, 'epoch': 0.25}
0: {'loss': 0.7415, 'grad_norm': 0.4795396443466839, 'learning_rate': 1e-05, 'epoch': 0.25}
0: <10:00:42,  3.28s/it]
                                                         

 25%|██▌       | 3740/14725 [3:25:11<10:00:42,  3.28s/it]
 25%|██▌       | 3741/14725 [3:25:14<9:48:25,  3.21s/it] 
 25%|██▌       | 3742/14725 [3:25:18<9:46:38,  3.20s/it]
 25%|██▌       | 3743/14725 [3:25:21<9:59:08,  3.27s/it]
 25%|██▌       | 3744/14725 [3:25:24<10:00:46,  3.28s/it]
 25%|██▌       | 3745/14725 [3:25:28<10:00:37,  3.28s/it]
 25%|██▌       | 3746/14725 [3:25:31<10:14:51,  3.36s/it]
 25%|██▌       | 3747/14725 [3:25:35<10:30:29,  3.45s/it]
 25%|██▌       | 3748/14725 [3:25:38<10:20:19,  3.39s/it]
 25%|██▌       | 3749/14725 [3:25:41<10:09:16,  3.33s/it]
 25%|██▌       | 3750/14725 [3:25:44<10:05:18,  3.31s/it]
                                                         

 25%|██▌       | 3750/14725 [3:25:44<10:05:18,  3.31s/it]
 25%|██▌       | 3751/14725 [3:25:48<9:54:23,  3.25s/it] 
 25%|██▌       | 3752/14725 [3:25:51<9:57:13,  3
0: {'loss': 0.7224, 'grad_norm': 0.46163681325564543, 'learning_rate': 1e-05, 'epoch': 0.26}
0: .27s/it]
 25%|██▌       | 3753/14725 [3:25:54<10:00:31,  3.28s/it]
 25%|██▌       | 3754/14725 [3:25:58<10:05:50,  3.31s/it]
 26%|██▌       | 3755/14725 [3:26:01<10:07:40,  3.32s/it]
 26%|██▌       | 3756/14725 [3:26:04<10:03:32,  3.30s/it]
 26%|██▌       | 3757/14725 [3:26:07<9:52:35,  3.24s/it] 
 26%|██▌       | 3758/14725 [3:26:11<9:59:28,  3.28s/it]
 26%|██▌       | 3759/14725 [3:26:14<9:54:32,  3.25s/it]
 26%|██▌       | 3760/14725 [3:26:17<9:55:55,  3.26s/it]
                                                        

 26%|██▌       | 3760/14725 [3:26:17<9:55:55,  3.26s/it]
 26%|██▌       | 3761/14725 [3:26:20<9:50:24,  3.23s/it]
 26%|██▌       | 3762/14725 [3:26:24<10:05:14,  3.31s/it]
 26%|██▌       | 3763/14725 [3:26:27<9:51:38,  3.24s/it] 
 26%|██▌       | 3764/14725 [3:26:30<9:51:00,  3.24s/it]
 26%|██▌       | 3765/14725 [3:26:33<9:52:38,  3.24s/it]
 26%|██▌       | 3766/14725 [3:26:37<9:54:33,  3.26s/it]
 26%|
0: {'loss': 0.7294, 'grad_norm': 0.4617452915120973, 'learning_rate': 1e-05, 'epoch': 0.26}
0: ██▌       | 3767/14725 [3:26:40<10:00:48,  3.29s/it]
 26%|██▌       | 3768/14725 [3:26:43<10:05:35,  3.32s/it]
 26%|██▌       | 3769/14725 [3:26:47<10:06:38,  3.32s/it]
 26%|██▌       | 3770/14725 [3:26:50<10:08:28,  3.33s/it]
                                                         

 26%|██▌       | 3770/14725 [3:26:50<10:08:28,  3.33s/it]
 26%|██▌       | 3771/14725 [3:26:53<10:11:14,  3.35s/it]
 26%|██▌       | 3772/14725 [3:26:57<10:06:51,  3.32s/it]
 26%|██▌       | 3773/14725 [3:27:00<10:27:05,  3.44s/it]
 26%|██▌       | 3774/14725 [3:27:04<10:16:38,  3.38s/it]
 26%|██▌       | 3775/14725 [3:27:07<10:07:36,  3.33s/it]
 26%|██▌       | 3776/14725 [3:27:10<10:08:01,  3.33s/it]
 26%|██▌       | 3777/14725 [3:27:14<10:14:12,  3.37s/it]
 26%|██▌       | 3778/14725 [3:27:17<10:24:55,  3.43s/it]
 26%|██▌       | 3779/14725 [3:27:20<10:09:08,  3.34s/it]
 26%|██▌       | 3780/14725 [3:27:23<9:55:14,  3.26s/it] 
          
0: {'loss': 0.7146, 'grad_norm': 0.42342263975417543, 'learning_rate': 1e-05, 'epoch': 0.26}
0: {'loss': 0.7192, 'grad_norm': 0.4481381308042391, 'learning_rate': 1e-05, 'epoch': 0.26}
0:                                               

 26%|██▌       | 3780/14725 [3:27:23<9:55:14,  3.26s/it]
 26%|██▌       | 3781/14725 [3:27:27<9:51:21,  3.24s/it]
 26%|██▌       | 3782/14725 [3:27:30<9:59:37,  3.29s/it]
 26%|██▌       | 3783/14725 [3:27:33<9:58:30,  3.28s/it]
 26%|██▌       | 3784/14725 [3:27:36<9:49:16,  3.23s/it]
 26%|██▌       | 3785/14725 [3:27:40<9:48:41,  3.23s/it]
 26%|██▌       | 3786/14725 [3:27:43<9:43:41,  3.20s/it]
 26%|██▌       | 3787/14725 [3:27:46<9:42:05,  3.19s/it]
 26%|██▌       | 3788/14725 [3:27:49<9:46:53,  3.22s/it]
 26%|██▌       | 3789/14725 [3:27:52<9:43:27,  3.20s/it]
 26%|██▌       | 3790/14725 [3:27:56<9:48:52,  3.23s/it]
                                                        

 26%|██▌       | 3790/14725 [3:27:56<9:48:52,  3.23s/it]
 26%|██▌       | 3791/14725 [3:27:59<9:51:43,  3.25s/it]
 26%|██▌       | 3792/14725 [3:28:02<9:57:23,  3.28s/it]
 26%|██▌       | 3793/14725 [3
0: {'loss': 0.727, 'grad_norm': 0.4593344974259033, 'learning_rate': 1e-05, 'epoch': 0.26}
0: :28:06<9:55:18,  3.27s/it]
 26%|██▌       | 3794/14725 [3:28:09<9:56:35,  3.27s/it]
 26%|██▌       | 3795/14725 [3:28:12<9:53:34,  3.26s/it]
 26%|██▌       | 3796/14725 [3:28:15<9:54:14,  3.26s/it]
 26%|██▌       | 3797/14725 [3:28:18<9:44:48,  3.21s/it]
 26%|██▌       | 3798/14725 [3:28:22<9:47:59,  3.23s/it]
 26%|██▌       | 3799/14725 [3:28:25<9:53:07,  3.26s/it]
 26%|██▌       | 3800/14725 [3:28:28<9:55:13,  3.27s/it]
                                                        

 26%|██▌       | 3800/14725 [3:28:28<9:55:13,  3.27s/it]
 26%|██▌       | 3801/14725 [3:28:32<9:56:01,  3.27s/it]
 26%|██▌       | 3802/14725 [3:28:35<9:53:29,  3.26s/it]
 26%|██▌       | 3803/14725 [3:28:38<9:46:45,  3.22s/it]
 26%|██▌       | 3804/14725 [3:28:41<9:43:47,  3.21s/it]
 26%|██▌       | 3805/14725 [3:28:44<9:47:28,  3.23s/it]
 26%|██▌       | 3806/14725 [3:28:48<9:51:02,  3.25s/it]
 26%|██▌       | 3807/14725 [3:28:51<10:04:57,  3.3
0: {'loss': 0.7267, 'grad_norm': 0.4525726639233219, 'learning_rate': 1e-05, 'epoch': 0.26}
0: {'loss': 0.7191, 'grad_norm': 0.4471793361147876, 'learning_rate': 1e-05, 'epoch': 0.26}
0: 2s/it]
 26%|██▌       | 3808/14725 [3:28:55<10:19:05,  3.40s/it]
 26%|██▌       | 3809/14725 [3:28:58<10:11:33,  3.36s/it]
 26%|██▌       | 3810/14725 [3:29:01<10:09:52,  3.35s/it]
                                                         

 26%|██▌       | 3810/14725 [3:29:01<10:09:52,  3.35s/it]
 26%|██▌       | 3811/14725 [3:29:05<10:10:50,  3.36s/it]
 26%|██▌       | 3812/14725 [3:29:08<10:01:21,  3.31s/it]
 26%|██▌       | 3813/14725 [3:29:11<9:54:33,  3.27s/it] 
 26%|██▌       | 3814/14725 [3:29:14<9:49:45,  3.24s/it]
 26%|██▌       | 3815/14725 [3:29:17<9:44:46,  3.22s/it]
 26%|██▌       | 3816/14725 [3:29:21<9:49:48,  3.24s/it]
 26%|██▌       | 3817/14725 [3:29:24<9:47:19,  3.23s/it]
 26%|██▌       | 3818/14725 [3:29:27<9:39:33,  3.19s/it]
 26%|██▌       | 3819/14725 [3:29:30<9:44:10,  3.21s/it]
 26%|██▌       | 3820/14725 [3:29:33<9:32:42,  3.15s/it]
                                                        

 26%|██
0: {'loss': 0.7284, 'grad_norm': 0.4617969833841353, 'learning_rate': 1e-05, 'epoch': 0.26}
0: ▌       | 3820/14725 [3:29:33<9:32:42,  3.15s/it]
 26%|██▌       | 3821/14725 [3:29:37<9:39:28,  3.19s/it]
 26%|██▌       | 3822/14725 [3:29:40<9:48:55,  3.24s/it]
 26%|██▌       | 3823/14725 [3:29:43<9:55:44,  3.28s/it]
 26%|██▌       | 3824/14725 [3:29:47<9:55:27,  3.28s/it]
 26%|██▌       | 3825/14725 [3:29:50<9:41:48,  3.20s/it]
 26%|██▌       | 3826/14725 [3:29:53<9:45:09,  3.22s/it]
 26%|██▌       | 3827/14725 [3:29:56<9:43:23,  3.21s/it]
 26%|██▌       | 3828/14725 [3:29:59<9:43:16,  3.21s/it]
 26%|██▌       | 3829/14725 [3:30:03<9:52:33,  3.26s/it]
 26%|██▌       | 3830/14725 [3:30:06<9:48:11,  3.24s/it]
                                                        

 26%|██▌       | 3830/14725 [3:30:06<9:48:11,  3.24s/it]
 26%|██▌       | 3831/14725 [3:30:09<9:39:53,  3.19s/it]
 26%|██▌       | 3832/14725 [3:30:12<9:46:09,  3.23s/it]
 26%|██▌       | 3833/14725 [3:30:15<9:40:38,  3.20s/it]
 26%|██▌       | 3834/1472
0: {'loss': 0.7206, 'grad_norm': 0.4443930657699697, 'learning_rate': 1e-05, 'epoch': 0.26}
0: 5 [3:30:19<9:43:03,  3.21s/it]
 26%|██▌       | 3835/14725 [3:30:22<9:37:55,  3.18s/it]
 26%|██▌       | 3836/14725 [3:30:25<9:38:55,  3.19s/it]
 26%|██▌       | 3837/14725 [3:30:28<9:36:07,  3.17s/it]
 26%|██▌       | 3838/14725 [3:30:32<9:47:39,  3.24s/it]
 26%|██▌       | 3839/14725 [3:30:35<9:44:27,  3.22s/it]
 26%|██▌       | 3840/14725 [3:30:38<9:42:57,  3.21s/it]
                                                        

 26%|██▌       | 3840/14725 [3:30:38<9:42:57,  3.21s/it]
 26%|██▌       | 3841/14725 [3:30:41<9:48:55,  3.25s/it]
 26%|██▌       | 3842/14725 [3:30:44<9:46:37,  3.23s/it]
 26%|██▌       | 3843/14725 [3:30:48<9:45:01,  3.23s/it]
 26%|██▌       | 3844/14725 [3:30:51<9:42:20,  3.21s/it]
 26%|██▌       | 3845/14725 [3:30:54<9:49:08,  3.25s/it]
 26%|██▌       | 3846/14725 [3:30:57<9:34:33,  3.17s/it]
 26%|██▌       | 3847/14725 [3:31:00<9:35:42,  3.18s/it]
 26%|██▌       | 3848/14725 [3:31:04<9:44:27,  
0: {'loss': 0.7325, 'grad_norm': 0.4552758408108717, 'learning_rate': 1e-05, 'epoch': 0.26}
0: {'loss': 0.7195, 'grad_norm': 0.4479849776812761, 'learning_rate': 1e-05, 'epoch': 0.26}
0: 3.22s/it]
 26%|██▌       | 3849/14725 [3:31:07<9:47:39,  3.24s/it]
 26%|██▌       | 3850/14725 [3:31:10<9:59:54,  3.31s/it]
                                                        

 26%|██▌       | 3850/14725 [3:31:10<9:59:54,  3.31s/it]
 26%|██▌       | 3851/14725 [3:31:14<9:58:45,  3.30s/it]
 26%|██▌       | 3852/14725 [3:31:17<9:57:56,  3.30s/it]
 26%|██▌       | 3853/14725 [3:31:20<9:58:24,  3.30s/it]
 26%|██▌       | 3854/14725 [3:31:24<9:53:36,  3.28s/it]
 26%|██▌       | 3855/14725 [3:31:27<9:47:03,  3.24s/it]
 26%|██▌       | 3856/14725 [3:31:30<9:48:01,  3.25s/it]
 26%|██▌       | 3857/14725 [3:31:33<9:49:36,  3.26s/it]
 26%|██▌       | 3858/14725 [3:31:37<9:51:25,  3.27s/it]
 26%|██▌       | 3859/14725 [3:31:40<9:52:30,  3.27s/it]
 26%|██▌       | 3860/14725 [3:31:43<9:46:27,  3.24s/it]
                                                        

 26%|██▌       | 3860/14725 [3:31:43<9:46:27,  3.24s/it]
 26%|██▌  
0: {'loss': 0.7258, 'grad_norm': 0.4362656100644398, 'learning_rate': 1e-05, 'epoch': 0.26}
0:      | 3861/14725 [3:31:46<9:47:39,  3.25s/it]
 26%|██▌       | 3862/14725 [3:31:49<9:40:38,  3.21s/it]
 26%|██▌       | 3863/14725 [3:31:53<9:53:31,  3.28s/it]
 26%|██▌       | 3864/14725 [3:31:56<9:55:11,  3.29s/it]
 26%|██▌       | 3865/14725 [3:31:59<9:48:47,  3.25s/it]
 26%|██▋       | 3866/14725 [3:32:02<9:46:13,  3.24s/it]
 26%|██▋       | 3867/14725 [3:32:06<9:48:35,  3.25s/it]
 26%|██▋       | 3868/14725 [3:32:09<9:49:55,  3.26s/it]
 26%|██▋       | 3869/14725 [3:32:12<9:57:09,  3.30s/it]
 26%|██▋       | 3870/14725 [3:32:16<10:18:11,  3.42s/it]
                                                         

 26%|██▋       | 3870/14725 [3:32:16<10:18:11,  3.42s/it]
 26%|██▋       | 3871/14725 [3:32:19<9:59:39,  3.31s/it] 
 26%|██▋       | 3872/14725 [3:32:23<10:14:17,  3.40s/it]
 26%|██▋       | 3873/14725 [3:32:26<10:08:50,  3.37s/it]
 26%|██▋       | 3874/14725 [3:32:29<9:56:51,  3.30s/it] 
 26%|██▋       | 3875/14
0: {'loss': 0.7313, 'grad_norm': 0.47403409881593506, 'learning_rate': 1e-05, 'epoch': 0.26}
0: 725 [3:32:33<9:56:51,  3.30s/it]
 26%|██▋       | 3876/14725 [3:32:36<9:55:42,  3.29s/it]
 26%|██▋       | 3877/14725 [3:32:39<10:03:49,  3.34s/it]
 26%|██▋       | 3878/14725 [3:32:42<9:55:56,  3.30s/it] 
 26%|██▋       | 3879/14725 [3:32:46<9:56:24,  3.30s/it]
 26%|██▋       | 3880/14725 [3:32:49<10:02:18,  3.33s/it]
                                                         

 26%|██▋       | 3880/14725 [3:32:49<10:02:18,  3.33s/it]
 26%|██▋       | 3881/14725 [3:32:53<10:14:53,  3.40s/it]
 26%|██▋       | 3882/14725 [3:32:56<10:13:02,  3.39s/it]
 26%|██▋       | 3883/14725 [3:32:59<10:02:07,  3.33s/it]
 26%|██▋       | 3884/14725 [3:33:02<9:52:35,  3.28s/it] 
 26%|██▋       | 3885/14725 [3:33:06<9:48:01,  3.25s/it]
 26%|██▋       | 3886/14725 [3:33:09<9:51:55,  3.28s/it]
 26%|██▋       | 3887/14725 [3:33:12<9:46:35,  3.25s/it]
 26%|██▋       | 3888/14725 [3:33:15<9:44:23,  3.24s/it]
 26%|██▋       | 3889/14725 [3:33:19
0: {'loss': 0.7269, 'grad_norm': 0.42709154450438336, 'learning_rate': 1e-05, 'epoch': 0.26}
0: {'loss': 0.7181, 'grad_norm': 0.4352918908428318, 'learning_rate': 1e-05, 'epoch': 0.26}
0: <9:42:07,  3.22s/it]
 26%|██▋       | 3890/14725 [3:33:22<9:44:46,  3.24s/it]
                                                        

 26%|██▋       | 3890/14725 [3:33:22<9:44:46,  3.24s/it]
 26%|██▋       | 3891/14725 [3:33:25<9:50:09,  3.27s/it]
 26%|██▋       | 3892/14725 [3:33:28<9:51:25,  3.28s/it]
 26%|██▋       | 3893/14725 [3:33:32<9:40:51,  3.22s/it]
 26%|██▋       | 3894/14725 [3:33:35<9:55:23,  3.30s/it]
 26%|██▋       | 3895/14725 [3:33:38<10:04:05,  3.35s/it]
 26%|██▋       | 3896/14725 [3:33:42<9:50:10,  3.27s/it] 
 26%|██▋       | 3897/14725 [3:33:45<9:49:00,  3.26s/it]
 26%|██▋       | 3898/14725 [3:33:48<9:52:23,  3.28s/it]
 26%|██▋       | 3899/14725 [3:33:51<9:55:12,  3.30s/it]
 26%|██▋       | 3900/14725 [3:33:55<9:48:38,  3.26s/it]
                                                        

 26%|██▋       | 3900/14725 [3:33:55<9:48:38,  3.26s/it]
 26%|██▋       | 3901/14725 [3:33:58<9:47:28,  3.26s/it]
 26
0: {'loss': 0.7213, 'grad_norm': 0.44614923350145813, 'learning_rate': 1e-05, 'epoch': 0.27}
0: %|██▋       | 3902/14725 [3:34:01<9:42:29,  3.23s/it]
 27%|██▋       | 3903/14725 [3:34:04<9:38:01,  3.20s/it]
 27%|██▋       | 3904/14725 [3:34:07<9:31:03,  3.17s/it]
 27%|██▋       | 3905/14725 [3:34:11<9:43:51,  3.24s/it]
 27%|██▋       | 3906/14725 [3:34:14<9:40:09,  3.22s/it]
 27%|██▋       | 3907/14725 [3:34:17<9:42:45,  3.23s/it]
 27%|██▋       | 3908/14725 [3:34:20<9:32:56,  3.18s/it]
 27%|██▋       | 3909/14725 [3:34:23<9:31:41,  3.17s/it]
 27%|██▋       | 3910/14725 [3:34:26<9:28:46,  3.16s/it]
                                                        

 27%|██▋       | 3910/14725 [3:34:26<9:28:46,  3.16s/it]
 27%|██▋       | 3911/14725 [3:34:30<9:41:27,  3.23s/it]
 27%|██▋       | 3912/14725 [3:34:33<9:44:06,  3.24s/it]
 27%|██▋       | 3913/14725 [3:34:36<9:47:57,  3.26s/it]
 27%|██▋       | 3914/14725 [3:34:40<9:45:05,  3.25s/it]
 27%|██▋       | 3915/14725 [3:34:43<9:47:39,  3.26s/it]
 27%|██▋       | 3
0: {'loss': 0.7333, 'grad_norm': 0.43239717624757146, 'learning_rate': 1e-05, 'epoch': 0.27}
0: 916/14725 [3:34:46<9:36:25,  3.20s/it]
 27%|██▋       | 3917/14725 [3:34:49<9:41:37,  3.23s/it]
 27%|██▋       | 3918/14725 [3:34:52<9:34:24,  3.19s/it]
 27%|██▋       | 3919/14725 [3:34:56<9:30:11,  3.17s/it]
 27%|██▋       | 3920/14725 [3:34:59<9:29:47,  3.16s/it]
                                                        

 27%|██▋       | 3920/14725 [3:34:59<9:29:47,  3.16s/it]
 27%|██▋       | 3921/14725 [3:35:02<9:33:21,  3.18s/it]
 27%|██▋       | 3922/14725 [3:35:05<9:33:51,  3.19s/it]
 27%|██▋       | 3923/14725 [3:35:08<9:25:59,  3.14s/it]
 27%|██▋       | 3924/14725 [3:35:11<9:36:25,  3.20s/it]
 27%|██▋       | 3925/14725 [3:35:15<9:41:00,  3.23s/it]
 27%|██▋       | 3926/14725 [3:35:18<9:46:24,  3.26s/it]
 27%|██▋       | 3927/14725 [3:35:22<9:56:05,  3.31s/it]
 27%|██▋       | 3928/14725 [3:35:25<10:00:58,  3.34s/it]
 27%|██▋       | 3929/14725 [3:35:28<9:53:44,  3.30s/it] 
 27%|██▋       | 3930/14725 [3:35:31<
0: {'loss': 0.7208, 'grad_norm': 0.4693363273374882, 'learning_rate': 1e-05, 'epoch': 0.27}
0: {'loss': 0.7247, 'grad_norm': 0.4640524403292025, 'learning_rate': 1e-05, 'epoch': 0.27}
0: 9:56:25,  3.32s/it]
                                                        

 27%|██▋       | 3930/14725 [3:35:31<9:56:25,  3.32s/it]
 27%|██▋       | 3931/14725 [3:35:35<9:43:05,  3.24s/it]
 27%|██▋       | 3932/14725 [3:35:38<9:44:00,  3.25s/it]
 27%|██▋       | 3933/14725 [3:35:41<9:54:20,  3.30s/it]
 27%|██▋       | 3934/14725 [3:35:45<10:01:23,  3.34s/it]
 27%|██▋       | 3935/14725 [3:35:48<10:03:48,  3.36s/it]
 27%|██▋       | 3936/14725 [3:35:51<10:02:21,  3.35s/it]
 27%|██▋       | 3937/14725 [3:35:55<9:50:17,  3.28s/it] 
 27%|██▋       | 3938/14725 [3:35:58<9:48:47,  3.28s/it]
 27%|██▋       | 3939/14725 [3:36:01<10:00:46,  3.34s/it]
 27%|██▋       | 3940/14725 [3:36:04<9:49:14,  3.28s/it] 
                                                        

 27%|██▋       | 3940/14725 [3:36:04<9:49:14,  3.28s/it]
 27%|██▋       | 3941/14725 [3:36:08<9:41:26,  3.23s/it]
 27%|██▋       | 3942/14725 [3:36:11<9:49:01,  3.28s/it]
0: {'loss': 0.7073, 'grad_norm': 0.4536502296587053, 'learning_rate': 1e-05, 'epoch': 0.27}
0:  27%|██▋       | 3943/14725 [3:36:14<10:01:23,  3.35s/it]
 27%|██▋       | 3944/14725 [3:36:18<9:53:15,  3.30s/it] 
 27%|██▋       | 3945/14725 [3:36:21<9:51:33,  3.29s/it]
 27%|██▋       | 3946/14725 [3:36:24<9:39:42,  3.23s/it]
 27%|██▋       | 3947/14725 [3:36:27<9:43:52,  3.25s/it]
 27%|██▋       | 3948/14725 [3:36:31<9:41:57,  3.24s/it]
 27%|██▋       | 3949/14725 [3:36:34<9:45:19,  3.26s/it]
 27%|██▋       | 3950/14725 [3:36:37<9:42:38,  3.24s/it]
                                                        

 27%|██▋       | 3950/14725 [3:36:37<9:42:38,  3.24s/it]
 27%|██▋       | 3951/14725 [3:36:40<9:42:15,  3.24s/it]
 27%|██▋       | 3952/14725 [3:36:43<9:38:24,  3.22s/it]
 27%|██▋       | 3953/14725 [3:36:47<9:45:29,  3.26s/it]
 27%|██▋       | 3954/14725 [3:36:50<9:54:31,  3.31s/it]
 27%|██▋       | 3955/14725 [3:36:54<9:57:35,  3.33s/it]
 27%|██▋       | 3956/14725 [3:36:57<9:49:24,  3.28s/it]
 27%|██▋     
0: {'loss': 0.7217, 'grad_norm': 0.4325496616955896, 'learning_rate': 1e-05, 'epoch': 0.27}
0:   | 3957/14725 [3:37:00<9:44:59,  3.26s/it]
 27%|██▋       | 3958/14725 [3:37:03<9:46:50,  3.27s/it]
 27%|██▋       | 3959/14725 [3:37:06<9:43:25,  3.25s/it]
 27%|██▋       | 3960/14725 [3:37:09<9:30:45,  3.18s/it]
                                                        

 27%|██▋       | 3960/14725 [3:37:09<9:30:45,  3.18s/it]
 27%|██▋       | 3961/14725 [3:37:13<9:29:22,  3.17s/it]
 27%|██▋       | 3962/14725 [3:37:16<9:32:36,  3.19s/it]
 27%|██▋       | 3963/14725 [3:37:19<9:38:07,  3.22s/it]
 27%|██▋       | 3964/14725 [3:37:22<9:35:04,  3.21s/it]
 27%|██▋       | 3965/14725 [3:37:26<9:36:48,  3.22s/it]
 27%|██▋       | 3966/14725 [3:37:29<9:41:51,  3.24s/it]
 27%|██▋       | 3967/14725 [3:37:32<9:42:35,  3.25s/it]
 27%|██▋       | 3968/14725 [3:37:35<9:44:47,  3.26s/it]
 27%|██▋       | 3969/14725 [3:37:39<9:38:43,  3.23s/it]
 27%|██▋       | 3970/14725 [3:37:42<9:36:49,  3.22s/it]
                                        
0: {'loss': 0.7169, 'grad_norm': 0.4409679168739473, 'learning_rate': 1e-05, 'epoch': 0.27}
0: {'loss': 0.723, 'grad_norm': 0.4409453589466028, 'learning_rate': 1e-05, 'epoch': 0.27}
0:                 

 27%|██▋       | 3970/14725 [3:37:42<9:36:49,  3.22s/it]
 27%|██▋       | 3971/14725 [3:37:45<9:39:25,  3.23s/it]
 27%|██▋       | 3972/14725 [3:37:48<9:45:00,  3.26s/it]
 27%|██▋       | 3973/14725 [3:37:52<9:44:42,  3.26s/it]
 27%|██▋       | 3974/14725 [3:37:55<9:52:33,  3.31s/it]
 27%|██▋       | 3975/14725 [3:37:58<9:47:33,  3.28s/it]
 27%|██▋       | 3976/14725 [3:38:01<9:40:40,  3.24s/it]
 27%|██▋       | 3977/14725 [3:38:05<9:47:23,  3.28s/it]
 27%|██▋       | 3978/14725 [3:38:08<9:47:46,  3.28s/it]
 27%|██▋       | 3979/14725 [3:38:11<9:34:34,  3.21s/it]
 27%|██▋       | 3980/14725 [3:38:14<9:32:25,  3.20s/it]
                                                        

 27%|██▋       | 3980/14725 [3:38:14<9:32:25,  3.20s/it]
 27%|██▋       | 3981/14725 [3:38:18<9:37:50,  3.23s/it]
 27%|██▋       | 3982/14725 [3:38:21<9:44:40,  3.27s/it]
 27%|██▋       | 3983/14725 [3:38:24<9:49:15,  3.29s/it]
 27
0: {'loss': 0.723, 'grad_norm': 0.4819959751043335, 'learning_rate': 1e-05, 'epoch': 0.27}
0: %|██▋       | 3984/14725 [3:38:28<9:51:34,  3.30s/it]
 27%|██▋       | 3985/14725 [3:38:31<9:45:59,  3.27s/it]
 27%|██▋       | 3986/14725 [3:38:34<9:40:40,  3.24s/it]
 27%|██▋       | 3987/14725 [3:38:38<9:55:15,  3.33s/it]
 27%|██▋       | 3988/14725 [3:38:41<9:58:31,  3.34s/it]
 27%|██▋       | 3989/14725 [3:38:44<9:54:58,  3.33s/it]
 27%|██▋       | 3990/14725 [3:38:47<9:51:17,  3.30s/it]
                                                        

 27%|██▋       | 3990/14725 [3:38:47<9:51:17,  3.30s/it]
 27%|██▋       | 3991/14725 [3:38:51<9:50:31,  3.30s/it]
 27%|██▋       | 3992/14725 [3:38:54<9:49:21,  3.29s/it]
 27%|██▋       | 3993/14725 [3:38:57<9:49:12,  3.29s/it]
 27%|██▋       | 3994/14725 [3:39:01<9:54:25,  3.32s/it]
 27%|██▋       | 3995/14725 [3:39:04<9:50:37,  3.30s/it]
 27%|██▋       | 3996/14725 [3:39:07<9:47:00,  3.28s/it]
 27%|██▋       | 3997/14725 [3:39:11<9:54:02,  3.32s/it]
 27%|██▋       | 3
0: {'loss': 0.7142, 'grad_norm': 0.42128081295534386, 'learning_rate': 1e-05, 'epoch': 0.27}
0: {'loss': 0.7221, 'grad_norm': 0.4586713957683762, 'learning_rate': 1e-05, 'epoch': 0.27}
0: 998/14725 [3:39:14<9:37:22,  3.23s/it]
 27%|██▋       | 3999/14725 [3:39:17<9:37:32,  3.23s/it]
 27%|██▋       | 4000/14725 [3:39:20<9:34:09,  3.21s/it]
                                                        

 27%|██▋       | 4000/14725 [3:39:20<9:34:09,  3.21s/it]
 27%|██▋       | 4001/14725 [3:39:23<9:31:15,  3.20s/it]
 27%|██▋       | 4002/14725 [3:39:27<9:40:25,  3.25s/it]
 27%|██▋       | 4003/14725 [3:39:30<9:39:16,  3.24s/it]
 27%|██▋       | 4004/14725 [3:39:33<9:43:34,  3.27s/it]
 27%|██▋       | 4005/14725 [3:39:36<9:41:08,  3.25s/it]
 27%|██▋       | 4006/14725 [3:39:40<9:45:49,  3.28s/it]
 27%|██▋       | 4007/14725 [3:39:43<9:30:12,  3.19s/it]
 27%|██▋       | 4008/14725 [3:39:46<9:41:17,  3.25s/it]
 27%|██▋       | 4009/14725 [3:39:49<9:33:52,  3.21s/it]
 27%|██▋       | 4010/14725 [3:39:52<9:32:22,  3.21s/it]
                                                        

 27%|██▋       | 4010/14725 [3:39:52<9:32:22
0: {'loss': 0.7113, 'grad_norm': 0.49325545668506815, 'learning_rate': 1e-05, 'epoch': 0.27}
0: ,  3.21s/it]
 27%|██▋       | 4011/14725 [3:39:56<9:42:03,  3.26s/it]
 27%|██▋       | 4012/14725 [3:39:59<9:34:46,  3.22s/it]
 27%|██▋       | 4013/14725 [3:40:02<9:38:42,  3.24s/it]
 27%|██▋       | 4014/14725 [3:40:05<9:31:26,  3.20s/it]
 27%|██▋       | 4015/14725 [3:40:09<9:34:31,  3.22s/it]
 27%|██▋       | 4016/14725 [3:40:12<9:33:19,  3.21s/it]
 27%|██▋       | 4017/14725 [3:40:15<9:30:12,  3.20s/it]
 27%|██▋       | 4018/14725 [3:40:18<9:32:52,  3.21s/it]
 27%|██▋       | 4019/14725 [3:40:21<9:39:27,  3.25s/it]
 27%|██▋       | 4020/14725 [3:40:25<9:34:51,  3.22s/it]
                                                        

 27%|██▋       | 4020/14725 [3:40:25<9:34:51,  3.22s/it]
 27%|██▋       | 4021/14725 [3:40:28<9:43:24,  3.27s/it]
 27%|██▋       | 4022/14725 [3:40:31<9:45:18,  3.28s/it]
 27%|██▋       | 4023/14725 [3:40:35<9:43:13,  3.27s/it]
 27%|██▋       | 4024/14725 [3:40:38<9:52:48,  3.32s/it]
 27%|█
0: {'loss': 0.7283, 'grad_norm': 0.44210039659684497, 'learning_rate': 1e-05, 'epoch': 0.27}
0: █▋       | 4025/14725 [3:40:41<9:46:15,  3.29s/it]
 27%|██▋       | 4026/14725 [3:40:45<10:00:34,  3.37s/it]
 27%|██▋       | 4027/14725 [3:40:49<10:22:39,  3.49s/it]
 27%|██▋       | 4028/14725 [3:40:52<10:14:01,  3.44s/it]
 27%|██▋       | 4029/14725 [3:40:55<10:08:16,  3.41s/it]
 27%|██▋       | 4030/14725 [3:40:58<9:58:13,  3.36s/it] 
                                                        

 27%|██▋       | 4030/14725 [3:40:58<9:58:13,  3.36s/it]
 27%|██▋       | 4031/14725 [3:41:02<9:50:38,  3.31s/it]
 27%|██▋       | 4032/14725 [3:41:05<9:54:55,  3.34s/it]
 27%|██▋       | 4033/14725 [3:41:08<9:50:21,  3.31s/it]
 27%|██▋       | 4034/14725 [3:41:12<9:44:18,  3.28s/it]
 27%|██▋       | 4035/14725 [3:41:15<9:39:50,  3.25s/it]
 27%|██▋       | 4036/14725 [3:41:18<9:36:22,  3.24s/it]
 27%|██▋       | 4037/14725 [3:41:21<9:33:50,  3.22s/it]
 27%|██▋       | 4038/14725 [3:41:25<9:49:18,  3.31s/it]
 27%|██▋       | 4
0: {'loss': 0.7155, 'grad_norm': 0.46260765781001373, 'learning_rate': 1e-05, 'epoch': 0.27}
0: {'loss': 0.7187, 'grad_norm': 0.4466522343269641, 'learning_rate': 1e-05, 'epoch': 0.28}
0: 039/14725 [3:41:28<9:44:19,  3.28s/it]
 27%|██▋       | 4040/14725 [3:41:31<9:49:21,  3.31s/it]
                                                        

 27%|██▋       | 4040/14725 [3:41:31<9:49:21,  3.31s/it]
 27%|██▋       | 4041/14725 [3:41:34<9:46:38,  3.29s/it]
 27%|██▋       | 4042/14725 [3:41:38<9:45:58,  3.29s/it]
 27%|██▋       | 4043/14725 [3:41:41<9:42:11,  3.27s/it]
 27%|██▋       | 4044/14725 [3:41:44<9:34:11,  3.23s/it]
 27%|██▋       | 4045/14725 [3:41:47<9:33:17,  3.22s/it]
 27%|██▋       | 4046/14725 [3:41:51<9:40:04,  3.26s/it]
 27%|██▋       | 4047/14725 [3:41:54<9:38:31,  3.25s/it]
 27%|██▋       | 4048/14725 [3:41:57<9:38:13,  3.25s/it]
 27%|██▋       | 4049/14725 [3:42:00<9:41:53,  3.27s/it]
 28%|██▊       | 4050/14725 [3:42:04<9:43:13,  3.28s/it]
                                                        

 28%|██▊       | 4050/14725 [3:42:04<9:43:13,  3.28s/it]
 28%|██▊       | 4051/14725 [3:42:07<9:43:47
0: {'loss': 0.7264, 'grad_norm': 0.4502270612982705, 'learning_rate': 1e-05, 'epoch': 0.28}
0: ,  3.28s/it]
 28%|██▊       | 4052/14725 [3:42:10<9:49:38,  3.31s/it]
 28%|██▊       | 4053/14725 [3:42:14<9:43:33,  3.28s/it]
 28%|██▊       | 4054/14725 [3:42:17<9:49:14,  3.31s/it]
 28%|██▊       | 4055/14725 [3:42:20<9:39:29,  3.26s/it]
 28%|██▊       | 4056/14725 [3:42:23<9:33:22,  3.22s/it]
 28%|██▊       | 4057/14725 [3:42:27<9:52:17,  3.33s/it]
 28%|██▊       | 4058/14725 [3:42:30<9:58:14,  3.36s/it]
 28%|██▊       | 4059/14725 [3:42:34<9:58:42,  3.37s/it]
 28%|██▊       | 4060/14725 [3:42:37<10:04:39,  3.40s/it]
                                                         

 28%|██▊       | 4060/14725 [3:42:37<10:04:39,  3.40s/it]
 28%|██▊       | 4061/14725 [3:42:41<10:16:45,  3.47s/it]
 28%|██▊       | 4062/14725 [3:42:44<9:58:39,  3.37s/it] 
 28%|██▊       | 4063/14725 [3:42:47<9:48:46,  3.31s/it]
 28%|██▊       | 4064/14725 [3:42:50<9:48:52,  3.31s/it]
 28%|██▊       | 4065/14725 [3:42:54<9:48:24,  3.31s/it]
 28
0: {'loss': 0.7175, 'grad_norm': 0.4489467499350163, 'learning_rate': 1e-05, 'epoch': 0.28}
0: %|██▊       | 4066/14725 [3:42:57<9:47:29,  3.31s/it]
 28%|██▊       | 4067/14725 [3:43:00<9:43:53,  3.29s/it]
 28%|██▊       | 4068/14725 [3:43:03<9:33:43,  3.23s/it]
 28%|██▊       | 4069/14725 [3:43:06<9:28:05,  3.20s/it]
 28%|██▊       | 4070/14725 [3:43:10<9:30:39,  3.21s/it]
                                                        

 28%|██▊       | 4070/14725 [3:43:10<9:30:39,  3.21s/it]
 28%|██▊       | 4071/14725 [3:43:13<9:26:47,  3.19s/it]
 28%|██▊       | 4072/14725 [3:43:16<9:24:18,  3.18s/it]
 28%|██▊       | 4073/14725 [3:43:19<9:20:28,  3.16s/it]
 28%|██▊       | 4074/14725 [3:43:22<9:26:26,  3.19s/it]
 28%|██▊       | 4075/14725 [3:43:26<9:25:29,  3.19s/it]
 28%|██▊       | 4076/14725 [3:43:29<9:41:16,  3.28s/it]
 28%|██▊       | 4077/14725 [3:43:32<9:33:41,  3.23s/it]
 28%|██▊       | 4078/14725 [3:43:36<9:46:29,  3.31s/it]
 28%|██▊       | 4079/14725 [3:43:39<9:53:06,  3.34s/it]
 28%|██▊       | 4
0: {'loss': 0.7248, 'grad_norm': 0.41652730287130874, 'learning_rate': 1e-05, 'epoch': 0.28}
0: {'loss': 0.7336, 'grad_norm': 0.43411506394114996, 'learning_rate': 1e-05, 'epoch': 0.28}
0: 080/14725 [3:43:42<9:47:59,  3.31s/it]
                                                        

 28%|██▊       | 4080/14725 [3:43:42<9:47:59,  3.31s/it]
 28%|██▊       | 4081/14725 [3:43:46<9:40:48,  3.27s/it]
 28%|██▊       | 4082/14725 [3:43:49<9:46:12,  3.30s/it]
 28%|██▊       | 4083/14725 [3:43:52<9:47:40,  3.31s/it]
 28%|██▊       | 4084/14725 [3:43:55<9:41:32,  3.28s/it]
 28%|██▊       | 4085/14725 [3:43:59<9:42:07,  3.28s/it]
 28%|██▊       | 4086/14725 [3:44:02<9:36:55,  3.25s/it]
 28%|██▊       | 4087/14725 [3:44:05<9:23:27,  3.18s/it]
 28%|██▊       | 4088/14725 [3:44:08<9:25:16,  3.19s/it]
 28%|██▊       | 4089/14725 [3:44:11<9:21:50,  3.17s/it]
 28%|██▊       | 4090/14725 [3:44:15<9:36:00,  3.25s/it]
                                                        

 28%|██▊       | 4090/14725 [3:44:15<9:36:00,  3.25s/it]
 28%|██▊       | 4091/14725 [3:44:18<9:39:45,  3.27s/it]
 28%|██▊       | 4092/14725 [3:44:21<9:42:46
0: {'loss': 0.7329, 'grad_norm': 0.45652321212859354, 'learning_rate': 1e-05, 'epoch': 0.28}
0: ,  3.29s/it]
 28%|██▊       | 4093/14725 [3:44:24<9:35:45,  3.25s/it]
 28%|██▊       | 4094/14725 [3:44:28<9:31:21,  3.22s/it]
 28%|██▊       | 4095/14725 [3:44:31<9:49:03,  3.32s/it]
 28%|██▊       | 4096/14725 [3:44:35<9:50:07,  3.33s/it]
 28%|██▊       | 4097/14725 [3:44:38<9:59:45,  3.39s/it]
 28%|██▊       | 4098/14725 [3:44:41<9:59:32,  3.39s/it]
 28%|██▊       | 4099/14725 [3:44:45<10:04:56,  3.42s/it]
 28%|██▊       | 4100/14725 [3:44:48<10:05:00,  3.42s/it]
                                                         

 28%|██▊       | 4100/14725 [3:44:48<10:05:00,  3.42s/it]
 28%|██▊       | 4101/14725 [3:44:52<9:59:28,  3.39s/it] 
 28%|██▊       | 4102/14725 [3:44:55<10:00:54,  3.39s/it]
 28%|██▊       | 4103/14725 [3:44:58<9:54:22,  3.36s/it] 
 28%|██▊       | 4104/14725 [3:45:02<9:43:00,  3.29s/it]
 28%|██▊       | 4105/14725 [3:45:05<9:45:13,  3.31s/it]
 28%|██▊       | 4106/14725 [3:45:08<9:48:04,  3.32s/it]
 
0: {'loss': 0.7092, 'grad_norm': 0.41005446008386837, 'learning_rate': 1e-05, 'epoch': 0.28}
0: 28%|██▊       | 4107/14725 [3:45:11<9:39:58,  3.28s/it]
 28%|██▊       | 4108/14725 [3:45:15<9:42:28,  3.29s/it]
 28%|██▊       | 4109/14725 [3:45:18<9:42:45,  3.29s/it]
 28%|██▊       | 4110/14725 [3:45:21<9:38:36,  3.27s/it]
                                                        

 28%|██▊       | 4110/14725 [3:45:21<9:38:36,  3.27s/it]
 28%|██▊       | 4111/14725 [3:45:24<9:36:14,  3.26s/it]
 28%|██▊       | 4112/14725 [3:45:28<9:41:34,  3.29s/it]
 28%|██▊       | 4113/14725 [3:45:31<9:41:13,  3.29s/it]
 28%|██▊       | 4114/14725 [3:45:34<9:37:34,  3.27s/it]
 28%|██▊       | 4115/14725 [3:45:38<9:43:09,  3.30s/it]
 28%|██▊       | 4116/14725 [3:45:41<9:52:16,  3.35s/it]
 28%|██▊       | 4117/14725 [3:45:44<9:46:53,  3.32s/it]
 28%|██▊       | 4118/14725 [3:45:48<9:42:50,  3.30s/it]
 28%|██▊       | 4119/14725 [3:45:51<9:51:19,  3.35s/it]
 28%|██▊       | 4120/14725 [3:45:54<9:47:17,  3.32s/it]
                      
0: {'loss': 0.7238, 'grad_norm': 0.41409078756994866, 'learning_rate': 1e-05, 'epoch': 0.28}
0: {'loss': 0.7284, 'grad_norm': 0.4327458088695549, 'learning_rate': 1e-05, 'epoch': 0.28}
0:                                   

 28%|██▊       | 4120/14725 [3:45:54<9:47:17,  3.32s/it]
 28%|██▊       | 4121/14725 [3:45:58<9:43:41,  3.30s/it]
 28%|██▊       | 4122/14725 [3:46:01<9:48:43,  3.33s/it]
 28%|██▊       | 4123/14725 [3:46:04<9:41:51,  3.29s/it]
 28%|██▊       | 4124/14725 [3:46:07<9:40:31,  3.29s/it]
 28%|██▊       | 4125/14725 [3:46:11<9:37:20,  3.27s/it]
 28%|██▊       | 4126/14725 [3:46:14<9:31:22,  3.23s/it]
 28%|██▊       | 4127/14725 [3:46:17<9:23:43,  3.19s/it]
 28%|██▊       | 4128/14725 [3:46:20<9:30:44,  3.23s/it]
 28%|██▊       | 4129/14725 [3:46:24<9:29:49,  3.23s/it]
 28%|██▊       | 4130/14725 [3:46:27<9:26:56,  3.21s/it]
                                                        

 28%|██▊       | 4130/14725 [3:46:27<9:26:56,  3.21s/it]
 28%|██▊       | 4131/14725 [3:46:30<9:30:16,  3.23s/it]
 28%|██▊       | 4132/14725 [3:46:33<9:33:52,  3.25s/it]
 28%|██▊       | 4133/14725 [3:46:36<9:30:
0: {'loss': 0.7254, 'grad_norm': 0.48419746337733693, 'learning_rate': 1e-05, 'epoch': 0.28}
0: 39,  3.23s/it]
 28%|██▊       | 4134/14725 [3:46:40<9:27:48,  3.22s/it]
 28%|██▊       | 4135/14725 [3:46:43<9:26:03,  3.21s/it]
 28%|██▊       | 4136/14725 [3:46:46<9:32:49,  3.25s/it]
 28%|██▊       | 4137/14725 [3:46:49<9:34:43,  3.26s/it]
 28%|██▊       | 4138/14725 [3:46:53<9:29:31,  3.23s/it]
 28%|██▊       | 4139/14725 [3:46:56<9:39:18,  3.28s/it]
 28%|██▊       | 4140/14725 [3:46:59<9:43:58,  3.31s/it]
                                                        

 28%|██▊       | 4140/14725 [3:46:59<9:43:58,  3.31s/it]
 28%|██▊       | 4141/14725 [3:47:03<9:38:19,  3.28s/it]
 28%|██▊       | 4142/14725 [3:47:06<9:37:10,  3.27s/it]
 28%|██▊       | 4143/14725 [3:47:09<9:38:00,  3.28s/it]
 28%|██▊       | 4144/14725 [3:47:12<9:39:07,  3.28s/it]
 28%|██▊       | 4145/14725 [3:47:16<9:39:21,  3.29s/it]
 28%|██▊       | 4146/14725 [3:47:19<9:32:12,  3.25s/it]
 28%|██▊       | 4147/14725 [3:47:22<9:28:54,  3.23s/it]
 28%|�
0: {'loss': 0.7253, 'grad_norm': 0.48699862498182583, 'learning_rate': 1e-05, 'epoch': 0.28}
0: {'loss': 0.7158, 'grad_norm': 0.42454949659384983, 'learning_rate': 1e-05, 'epoch': 0.28}
0: ��█▊       | 4148/14725 [3:47:25<9:36:05,  3.27s/it]
 28%|██▊       | 4149/14725 [3:47:29<9:34:55,  3.26s/it]
 28%|██▊       | 4150/14725 [3:47:32<9:47:48,  3.34s/it]
                                                        

 28%|██▊       | 4150/14725 [3:47:32<9:47:48,  3.34s/it]
 28%|██▊       | 4151/14725 [3:47:35<9:40:58,  3.30s/it]
 28%|██▊       | 4152/14725 [3:47:38<9:30:49,  3.24s/it]
 28%|██▊       | 4153/14725 [3:47:42<9:38:52,  3.29s/it]
 28%|██▊       | 4154/14725 [3:47:45<9:36:45,  3.27s/it]
 28%|██▊       | 4155/14725 [3:47:49<9:44:30,  3.32s/it]
 28%|██▊       | 4156/14725 [3:47:52<9:47:35,  3.34s/it]
 28%|██▊       | 4157/14725 [3:47:55<9:46:10,  3.33s/it]
 28%|██▊       | 4158/14725 [3:47:59<9:45:22,  3.32s/it]
 28%|██▊       | 4159/14725 [3:48:02<9:39:22,  3.29s/it]
 28%|██▊       | 4160/14725 [3:48:05<9:33:54,  3.26s/it]
                                                        

 28%|██▊       | 4160/1472
0: {'loss': 0.7256, 'grad_norm': 0.4103569047498629, 'learning_rate': 1e-05, 'epoch': 0.28}
0: 5 [3:48:05<9:33:54,  3.26s/it]
 28%|██▊       | 4161/14725 [3:48:08<9:26:21,  3.22s/it]
 28%|██▊       | 4162/14725 [3:48:11<9:38:26,  3.29s/it]
 28%|██▊       | 4163/14725 [3:48:15<9:33:45,  3.26s/it]
 28%|██▊       | 4164/14725 [3:48:18<9:28:14,  3.23s/it]
 28%|██▊       | 4165/14725 [3:48:21<9:30:48,  3.24s/it]
 28%|██▊       | 4166/14725 [3:48:24<9:35:39,  3.27s/it]
 28%|██▊       | 4167/14725 [3:48:28<9:52:30,  3.37s/it]
 28%|██▊       | 4168/14725 [3:48:31<9:45:50,  3.33s/it]
 28%|██▊       | 4169/14725 [3:48:34<9:35:45,  3.27s/it]
 28%|██▊       | 4170/14725 [3:48:38<9:42:07,  3.31s/it]
                                                        

 28%|██▊       | 4170/14725 [3:48:38<9:42:07,  3.31s/it]
 28%|██▊       | 4171/14725 [3:48:41<9:37:28,  3.28s/it]
 28%|██▊       | 4172/14725 [3:48:44<9:44:40,  3.32s/it]
 28%|██▊       | 4173/14725 [3:48:48<9:33:25,  3.26s/it]
 28%|██▊       | 4174/14725 [3:48:51<9:32:30,  
0: {'loss': 0.7328, 'grad_norm': 0.44290839861879105, 'learning_rate': 1e-05, 'epoch': 0.28}
0: 3.26s/it]
 28%|██▊       | 4175/14725 [3:48:54<9:36:34,  3.28s/it]
 28%|██▊       | 4176/14725 [3:48:57<9:37:43,  3.29s/it]
 28%|██▊       | 4177/14725 [3:49:01<9:42:05,  3.31s/it]
 28%|██▊       | 4178/14725 [3:49:04<9:45:57,  3.33s/it]
 28%|██▊       | 4179/14725 [3:49:07<9:31:34,  3.25s/it]
 28%|██▊       | 4180/14725 [3:49:11<9:39:36,  3.30s/it]
                                                        

 28%|██▊       | 4180/14725 [3:49:11<9:39:36,  3.30s/it]
 28%|██▊       | 4181/14725 [3:49:14<10:00:27,  3.42s/it]
 28%|██▊       | 4182/14725 [3:49:18<9:47:38,  3.34s/it] 
 28%|██▊       | 4183/14725 [3:49:21<9:40:02,  3.30s/it]
 28%|██▊       | 4184/14725 [3:49:24<9:40:15,  3.30s/it]
 28%|██▊       | 4185/14725 [3:49:27<9:46:25,  3.34s/it]
 28%|██▊       | 4186/14725 [3:49:31<9:50:37,  3.36s/it]
 28%|██▊       | 4187/14725 [3:49:34<9:38:23,  3.29s/it]
 28%|██▊       | 4188/14725 [3:49:37<9:33:28,  3.27s/it]
 28%|█�
0: {'loss': 0.7132, 'grad_norm': 0.4544704497344482, 'learning_rate': 1e-05, 'epoch': 0.28}
0: {'loss': 0.7063, 'grad_norm': 0.45148339484247585, 'learning_rate': 1e-05, 'epoch': 0.29}
0: ��▊       | 4189/14725 [3:49:40<9:30:15,  3.25s/it]
 28%|██▊       | 4190/14725 [3:49:44<9:27:09,  3.23s/it]
                                                        

 28%|██▊       | 4190/14725 [3:49:44<9:27:09,  3.23s/it]
 28%|██▊       | 4191/14725 [3:49:47<9:33:25,  3.27s/it]
 28%|██▊       | 4192/14725 [3:49:50<9:31:38,  3.26s/it]
 28%|██▊       | 4193/14725 [3:49:54<9:33:03,  3.26s/it]
 28%|██▊       | 4194/14725 [3:49:57<9:37:59,  3.29s/it]
 28%|██▊       | 4195/14725 [3:50:00<9:26:12,  3.23s/it]
 28%|██▊       | 4196/14725 [3:50:03<9:25:23,  3.22s/it]
 29%|██▊       | 4197/14725 [3:50:06<9:19:50,  3.19s/it]
 29%|██▊       | 4198/14725 [3:50:10<9:28:55,  3.24s/it]
 29%|██▊       | 4199/14725 [3:50:13<9:50:26,  3.37s/it]
 29%|██▊       | 4200/14725 [3:50:16<9:41:20,  3.31s/it]
                                                        

 29%|██▊       | 4200/14725 [3:50:16<9:41:20,  3.31s/it]
 29%|██▊       | 4201/14725 [
0: {'loss': 0.7172, 'grad_norm': 0.4721150540922747, 'learning_rate': 1e-05, 'epoch': 0.29}
0: 3:50:20<9:30:24,  3.25s/it]
 29%|██▊       | 4202/14725 [3:50:23<9:33:08,  3.27s/it]
 29%|██▊       | 4203/14725 [3:50:26<9:33:15,  3.27s/it]
 29%|██▊       | 4204/14725 [3:50:30<9:38:11,  3.30s/it]
 29%|██▊       | 4205/14725 [3:50:33<9:32:25,  3.26s/it]
 29%|██▊       | 4206/14725 [3:50:36<9:36:10,  3.29s/it]
 29%|██▊       | 4207/14725 [3:50:39<9:30:07,  3.25s/it]
 29%|██▊       | 4208/14725 [3:50:43<9:37:35,  3.30s/it]
 29%|██▊       | 4209/14725 [3:50:46<9:39:10,  3.30s/it]
 29%|██▊       | 4210/14725 [3:50:49<9:34:06,  3.28s/it]
                                                        

 29%|██▊       | 4210/14725 [3:50:49<9:34:06,  3.28s/it]
 29%|██▊       | 4211/14725 [3:50:52<9:31:08,  3.26s/it]
 29%|██▊       | 4212/14725 [3:50:56<9:26:57,  3.24s/it]
 29%|██▊       | 4213/14725 [3:50:59<9:18:04,  3.19s/it]
 29%|██▊       | 4214/14725 [3:51:02<9:19:11,  3.19s/it]
 29%|██▊       | 4215/14725 [3:51:05<9:19:15,  3.1
0: {'loss': 0.7076, 'grad_norm': 0.4195591180224713, 'learning_rate': 1e-05, 'epoch': 0.29}
0: 9s/it]
 29%|██▊       | 4216/14725 [3:51:09<9:35:02,  3.28s/it]
 29%|██▊       | 4217/14725 [3:51:12<9:31:20,  3.26s/it]
 29%|██▊       | 4218/14725 [3:51:15<9:25:05,  3.23s/it]
 29%|██▊       | 4219/14725 [3:51:19<9:51:43,  3.38s/it]
 29%|██▊       | 4220/14725 [3:51:22<9:54:47,  3.40s/it]
                                                        

 29%|██▊       | 4220/14725 [3:51:22<9:54:47,  3.40s/it]
 29%|██▊       | 4221/14725 [3:51:26<10:00:11,  3.43s/it]
 29%|██▊       | 4222/14725 [3:51:29<9:56:49,  3.41s/it] 
 29%|██▊       | 4223/14725 [3:51:32<9:46:25,  3.35s/it]
 29%|██▊       | 4224/14725 [3:51:35<9:44:49,  3.34s/it]
 29%|██▊       | 4225/14725 [3:51:39<9:42:36,  3.33s/it]
 29%|██▊       | 4226/14725 [3:51:42<9:38:46,  3.31s/it]
 29%|██▊       | 4227/14725 [3:51:45<9:38:30,  3.31s/it]
 29%|██▊       | 4228/14725 [3:51:48<9:25:34,  3.23s/it]
 29%|██▊       | 4229/14725 [3:51:52<9:30:39,  3.26s/it]
 29%|██�
0: {'loss': 0.7209, 'grad_norm': 0.4506620456480397, 'learning_rate': 1e-05, 'epoch': 0.29}
0: {'loss': 0.72, 'grad_norm': 0.4940310202188752, 'learning_rate': 1e-05, 'epoch': 0.29}
0: ��       | 4230/14725 [3:51:55<9:25:34,  3.23s/it]
                                                        

 29%|██▊       | 4230/14725 [3:51:55<9:25:34,  3.23s/it]
 29%|██▊       | 4231/14725 [3:51:58<9:38:45,  3.31s/it]
 29%|██▊       | 4232/14725 [3:52:02<9:33:37,  3.28s/it]
 29%|██▊       | 4233/14725 [3:52:05<9:35:39,  3.29s/it]
 29%|██▉       | 4234/14725 [3:52:08<9:38:05,  3.31s/it]
 29%|██▉       | 4235/14725 [3:52:12<9:41:45,  3.33s/it]
 29%|██▉       | 4236/14725 [3:52:15<9:29:18,  3.26s/it]
 29%|██▉       | 4237/14725 [3:52:18<9:35:54,  3.29s/it]
 29%|██▉       | 4238/14725 [3:52:21<9:38:59,  3.31s/it]
 29%|██▉       | 4239/14725 [3:52:25<9:34:19,  3.29s/it]
 29%|██▉       | 4240/14725 [3:52:28<9:27:22,  3.25s/it]
                                                        

 29%|██▉       | 4240/14725 [3:52:28<9:27:22,  3.25s/it]
 29%|██▉       | 4241/14725 [3:52:31<9:33:00,  3.28s/it]
 29%|██▉       | 4242/14725 [3:5
0: {'loss': 0.7207, 'grad_norm': 0.49304418406321243, 'learning_rate': 1e-05, 'epoch': 0.29}
0: 2:34<9:35:17,  3.29s/it]
 29%|██▉       | 4243/14725 [3:52:38<9:29:41,  3.26s/it]
 29%|██▉       | 4244/14725 [3:52:41<9:21:41,  3.22s/it]
 29%|██▉       | 4245/14725 [3:52:44<9:21:59,  3.22s/it]
 29%|██▉       | 4246/14725 [3:52:47<9:16:57,  3.19s/it]
 29%|██▉       | 4247/14725 [3:52:50<9:13:46,  3.17s/it]
 29%|██▉       | 4248/14725 [3:52:54<9:17:46,  3.19s/it]
 29%|██▉       | 4249/14725 [3:52:57<9:28:44,  3.26s/it]
 29%|██▉       | 4250/14725 [3:53:00<9:35:59,  3.30s/it]
                                                        

 29%|██▉       | 4250/14725 [3:53:00<9:35:59,  3.30s/it]
 29%|██▉       | 4251/14725 [3:53:04<9:34:58,  3.29s/it]
 29%|██▉       | 4252/14725 [3:53:07<9:32:41,  3.28s/it]
 29%|██▉       | 4253/14725 [3:53:10<9:18:00,  3.20s/it]
 29%|██▉       | 4254/14725 [3:53:13<9:22:16,  3.22s/it]
 29%|██▉       | 4255/14725 [3:53:17<9:38:22,  3.31s/it]
 29%|██▉       | 4256/14725 [3:53:20<9:37:49,  3.31s/
0: {'loss': 0.7142, 'grad_norm': 0.4365451911497559, 'learning_rate': 1e-05, 'epoch': 0.29}
0: it]
 29%|██▉       | 4257/14725 [3:53:23<9:28:19,  3.26s/it]
 29%|██▉       | 4258/14725 [3:53:26<9:16:59,  3.19s/it]
 29%|██▉       | 4259/14725 [3:53:29<9:17:36,  3.20s/it]
 29%|██▉       | 4260/14725 [3:53:32<9:14:13,  3.18s/it]
                                                        

 29%|██▉       | 4260/14725 [3:53:32<9:14:13,  3.18s/it]
 29%|██▉       | 4261/14725 [3:53:36<9:07:00,  3.14s/it]
 29%|██▉       | 4262/14725 [3:53:39<9:14:06,  3.18s/it]
 29%|██▉       | 4263/14725 [3:53:42<9:30:20,  3.27s/it]
 29%|██▉       | 4264/14725 [3:53:46<9:44:01,  3.35s/it]
 29%|██▉       | 4265/14725 [3:53:49<9:37:38,  3.31s/it]
 29%|██▉       | 4266/14725 [3:53:52<9:36:09,  3.31s/it]
 29%|██▉       | 4267/14725 [3:53:56<9:32:34,  3.29s/it]
 29%|██▉       | 4268/14725 [3:53:59<9:43:32,  3.35s/it]
 29%|██▉       | 4269/14725 [3:54:02<9:38:34,  3.32s/it]
 29%|██▉       | 4270/14725 [3:54:06<9:40:49,  3.33s/it]
                 
0: {'loss': 0.7157, 'grad_norm': 0.5034692262582858, 'learning_rate': 1e-05, 'epoch': 0.29}
0: {'loss': 0.709, 'grad_norm': 0.4614320906405022, 'learning_rate': 1e-05, 'epoch': 0.29}
0:                                        

 29%|██▉       | 4270/14725 [3:54:06<9:40:49,  3.33s/it]
 29%|██▉       | 4271/14725 [3:54:09<9:42:41,  3.34s/it]
 29%|██▉       | 4272/14725 [3:54:12<9:36:46,  3.31s/it]
 29%|██▉       | 4273/14725 [3:54:15<9:31:26,  3.28s/it]
 29%|██▉       | 4274/14725 [3:54:19<9:30:54,  3.28s/it]
 29%|██▉       | 4275/14725 [3:54:22<9:29:18,  3.27s/it]
 29%|██▉       | 4276/14725 [3:54:25<9:39:42,  3.33s/it]
 29%|██▉       | 4277/14725 [3:54:29<9:35:14,  3.30s/it]
 29%|██▉       | 4278/14725 [3:54:32<9:50:58,  3.39s/it]
 29%|██▉       | 4279/14725 [3:54:36<9:52:22,  3.40s/it]
 29%|██▉       | 4280/14725 [3:54:39<9:42:42,  3.35s/it]
                                                        

 29%|██▉       | 4280/14725 [3:54:39<9:42:42,  3.35s/it]
 29%|██▉       | 4281/14725 [3:54:42<9:35:54,  3.31s/it]
 29%|██▉       | 4282/14725 [3:54:45<9:28:51,  3.27s/it]
 29%|██▉       | 4283/14725 [3:54:49<
0: {'loss': 0.7201, 'grad_norm': 0.4853624061029021, 'learning_rate': 1e-05, 'epoch': 0.29}
0: 9:25:52,  3.25s/it]
 29%|██▉       | 4284/14725 [3:54:52<9:24:22,  3.24s/it]
 29%|██▉       | 4285/14725 [3:54:55<9:19:53,  3.22s/it]
 29%|██▉       | 4286/14725 [3:54:58<9:21:22,  3.23s/it]
 29%|██▉       | 4287/14725 [3:55:02<9:31:55,  3.29s/it]
 29%|██▉       | 4288/14725 [3:55:05<9:42:46,  3.35s/it]
 29%|██▉       | 4289/14725 [3:55:09<9:56:28,  3.43s/it]
 29%|██▉       | 4290/14725 [3:55:12<9:46:14,  3.37s/it]
                                                        

 29%|██▉       | 4290/14725 [3:55:12<9:46:14,  3.37s/it]
 29%|██▉       | 4291/14725 [3:55:15<9:40:16,  3.34s/it]
 29%|██▉       | 4292/14725 [3:55:18<9:30:16,  3.28s/it]
 29%|██▉       | 4293/14725 [3:55:21<9:20:38,  3.22s/it]
 29%|██▉       | 4294/14725 [3:55:25<9:20:55,  3.23s/it]
 29%|██▉       | 4295/14725 [3:55:28<9:26:51,  3.26s/it]
 29%|██▉       | 4296/14725 [3:55:31<9:22:50,  3.24s/it]
 29%|██▉       | 4297/14725 [3:55:34<9:21:21,  3.23s/it]
 
0: {'loss': 0.7233, 'grad_norm': 0.46325383248879465, 'learning_rate': 1e-05, 'epoch': 0.29}
0: {'loss': 0.7044, 'grad_norm': 0.42557186197503005, 'learning_rate': 1e-05, 'epoch': 0.29}
0: 29%|██▉       | 4298/14725 [3:55:38<9:22:16,  3.24s/it]
 29%|██▉       | 4299/14725 [3:55:41<9:25:45,  3.26s/it]
 29%|██▉       | 4300/14725 [3:55:44<9:14:23,  3.19s/it]
                                                        

 29%|██▉       | 4300/14725 [3:55:44<9:14:23,  3.19s/it]
 29%|██▉       | 4301/14725 [3:55:47<9:18:54,  3.22s/it]
 29%|██▉       | 4302/14725 [3:55:50<9:17:31,  3.21s/it]
 29%|██▉       | 4303/14725 [3:55:54<9:19:52,  3.22s/it]
 29%|██▉       | 4304/14725 [3:55:57<9:16:02,  3.20s/it]
 29%|██▉       | 4305/14725 [3:56:00<9:08:35,  3.16s/it]
 29%|██▉       | 4306/14725 [3:56:03<9:14:58,  3.20s/it]
 29%|██▉       | 4307/14725 [3:56:06<9:10:09,  3.17s/it]
 29%|██▉       | 4308/14725 [3:56:09<9:04:10,  3.13s/it]
 29%|██▉       | 4309/14725 [3:56:13<9:04:46,  3.14s/it]
 29%|██▉       | 4310/14725 [3:56:16<9:19:21,  3.22s/it]
                                                        

 29%|██▉       | 4310
0: {'loss': 0.719, 'grad_norm': 0.4436828729714216, 'learning_rate': 1e-05, 'epoch': 0.29}
0: /14725 [3:56:16<9:19:21,  3.22s/it]
 29%|██▉       | 4311/14725 [3:56:19<9:18:54,  3.22s/it]
 29%|██▉       | 4312/14725 [3:56:23<9:31:49,  3.29s/it]
 29%|██▉       | 4313/14725 [3:56:26<9:26:21,  3.26s/it]
 29%|██▉       | 4314/14725 [3:56:29<9:19:44,  3.23s/it]
 29%|██▉       | 4315/14725 [3:56:32<9:20:55,  3.23s/it]
 29%|██▉       | 4316/14725 [3:56:35<9:19:08,  3.22s/it]
 29%|██▉       | 4317/14725 [3:56:39<9:17:18,  3.21s/it]
 29%|██▉       | 4318/14725 [3:56:42<9:14:05,  3.19s/it]
 29%|██▉       | 4319/14725 [3:56:45<9:22:25,  3.24s/it]
 29%|██▉       | 4320/14725 [3:56:48<9:24:29,  3.26s/it]
                                                        

 29%|██▉       | 4320/14725 [3:56:48<9:24:29,  3.26s/it]
 29%|██▉       | 4321/14725 [3:56:52<9:22:59,  3.25s/it]
 29%|██▉       | 4322/14725 [3:56:55<9:28:23,  3.28s/it]
 29%|██▉       | 4323/14725 [3:56:58<9:32:55,  3.30s/it]
 29%|██▉       | 4324/14725 [3:57:01<9:20:
0: {'loss': 0.7239, 'grad_norm': 0.4423924261022515, 'learning_rate': 1e-05, 'epoch': 0.29}
0: 24,  3.23s/it]
 29%|██▉       | 4325/14725 [3:57:05<9:20:47,  3.24s/it]
 29%|██▉       | 4326/14725 [3:57:08<9:21:46,  3.24s/it]
 29%|██▉       | 4327/14725 [3:57:11<9:14:20,  3.20s/it]
 29%|██▉       | 4328/14725 [3:57:14<9:20:44,  3.24s/it]
 29%|██▉       | 4329/14725 [3:57:18<9:20:59,  3.24s/it]
 29%|██▉       | 4330/14725 [3:57:21<9:23:59,  3.26s/it]
                                                        

 29%|██▉       | 4330/14725 [3:57:21<9:23:59,  3.26s/it]
 29%|██▉       | 4331/14725 [3:57:24<9:10:31,  3.18s/it]
 29%|██▉       | 4332/14725 [3:57:27<9:08:53,  3.17s/it]
 29%|██▉       | 4333/14725 [3:57:30<9:12:40,  3.19s/it]
 29%|██▉       | 4334/14725 [3:57:34<9:20:42,  3.24s/it]
 29%|██▉       | 4335/14725 [3:57:37<9:22:04,  3.25s/it]
 29%|██▉       | 4336/14725 [3:57:40<9:14:42,  3.20s/it]
 29%|██▉       | 4337/14725 [3:57:43<9:22:03,  3.25s/it]
 29%|██▉       | 4338/14725 [3:57:46<9:12:53,  3.19s/it]
 29%|�
0: {'loss': 0.7179, 'grad_norm': 0.44494129067793703, 'learning_rate': 1e-05, 'epoch': 0.29}
0: {'loss': 0.7232, 'grad_norm': 0.44643355954806907, 'learning_rate': 1e-05, 'epoch': 0.3}
0: ��█▉       | 4339/14725 [3:57:50<9:21:27,  3.24s/it]
 29%|██▉       | 4340/14725 [3:57:53<9:19:19,  3.23s/it]
                                                        

 29%|██▉       | 4340/14725 [3:57:53<9:19:19,  3.23s/it]
 29%|██▉       | 4341/14725 [3:57:56<9:15:23,  3.21s/it]
 29%|██▉       | 4342/14725 [3:57:59<9:13:38,  3.20s/it]
 29%|██▉       | 4343/14725 [3:58:03<9:26:03,  3.27s/it]
 30%|██▉       | 4344/14725 [3:58:06<9:20:57,  3.24s/it]
 30%|██▉       | 4345/14725 [3:58:09<9:18:37,  3.23s/it]
 30%|██▉       | 4346/14725 [3:58:13<9:27:12,  3.28s/it]
 30%|██▉       | 4347/14725 [3:58:16<9:39:06,  3.35s/it]
 30%|██▉       | 4348/14725 [3:58:19<9:40:30,  3.36s/it]
 30%|██▉       | 4349/14725 [3:58:23<9:31:07,  3.30s/it]
 30%|██▉       | 4350/14725 [3:58:26<9:29:25,  3.29s/it]
                                                        

 30%|██▉       | 4350/14725 [3:58:26<9:29:25,  3.29s/it]
 30%|██▉       | 4351/1472
0: {'loss': 0.7091, 'grad_norm': 0.47579547589375504, 'learning_rate': 1e-05, 'epoch': 0.3}
0: 5 [3:58:29<9:25:44,  3.27s/it]
 30%|██▉       | 4352/14725 [3:58:32<9:17:18,  3.22s/it]
 30%|██▉       | 4353/14725 [3:58:36<9:23:05,  3.26s/it]
 30%|██▉       | 4354/14725 [3:58:39<9:18:24,  3.23s/it]
 30%|██▉       | 4355/14725 [3:58:42<9:19:51,  3.24s/it]
 30%|██▉       | 4356/14725 [3:58:45<9:24:50,  3.27s/it]
 30%|██▉       | 4357/14725 [3:58:48<9:18:45,  3.23s/it]
 30%|██▉       | 4358/14725 [3:58:52<9:19:57,  3.24s/it]
 30%|██▉       | 4359/14725 [3:58:55<9:26:58,  3.28s/it]
 30%|██▉       | 4360/14725 [3:58:58<9:22:10,  3.25s/it]
                                                        

 30%|██▉       | 4360/14725 [3:58:58<9:22:10,  3.25s/it]
 30%|██▉       | 4361/14725 [3:59:01<9:18:18,  3.23s/it]
 30%|██▉       | 4362/14725 [3:59:05<9:19:58,  3.24s/it]
 30%|██▉       | 4363/14725 [3:59:08<9:25:22,  3.27s/it]
 30%|██▉       | 4364/14725 [3:59:11<9:26:30,  3.28s/it]
 30%|██▉       | 4365/14725 [3:59:14<9:19:52,  
0: {'loss': 0.7219, 'grad_norm': 0.5035157787902036, 'learning_rate': 1e-05, 'epoch': 0.3}
0: 3.24s/it]
 30%|██▉       | 4366/14725 [3:59:18<9:18:30,  3.23s/it]
 30%|██▉       | 4367/14725 [3:59:21<9:15:17,  3.22s/it]
 30%|██▉       | 4368/14725 [3:59:24<9:18:28,  3.24s/it]
 30%|██▉       | 4369/14725 [3:59:27<9:20:23,  3.25s/it]
 30%|██▉       | 4370/14725 [3:59:31<9:23:53,  3.27s/it]
                                                        

 30%|██▉       | 4370/14725 [3:59:31<9:23:53,  3.27s/it]
 30%|██▉       | 4371/14725 [3:59:34<9:23:39,  3.27s/it]
 30%|██▉       | 4372/14725 [3:59:37<9:21:01,  3.25s/it]
 30%|██▉       | 4373/14725 [3:59:40<9:15:03,  3.22s/it]
 30%|██▉       | 4374/14725 [3:59:44<9:30:33,  3.31s/it]
 30%|██▉       | 4375/14725 [3:59:47<9:23:36,  3.27s/it]
 30%|██▉       | 4376/14725 [3:59:50<9:18:10,  3.24s/it]
 30%|██▉       | 4377/14725 [3:59:54<9:25:06,  3.28s/it]
 30%|██▉       | 4378/14725 [3:59:57<9:15:10,  3.22s/it]
 30%|██▉       | 4379/14725 [4:00:00<9:17:27,  3.23s/it]
 30%|██
0: {'loss': 0.7154, 'grad_norm': 0.4676108587276805, 'learning_rate': 1e-05, 'epoch': 0.3}
0: {'loss': 0.729, 'grad_norm': 0.47464985936267856, 'learning_rate': 1e-05, 'epoch': 0.3}
0: ▉       | 4380/14725 [4:00:03<9:16:00,  3.22s/it]
                                                        

 30%|██▉       | 4380/14725 [4:00:03<9:16:00,  3.22s/it]
 30%|██▉       | 4381/14725 [4:00:06<9:19:52,  3.25s/it]
 30%|██▉       | 4382/14725 [4:00:10<9:19:37,  3.25s/it]
 30%|██▉       | 4383/14725 [4:00:13<9:14:03,  3.21s/it]
 30%|██▉       | 4384/14725 [4:00:16<9:14:45,  3.22s/it]
 30%|██▉       | 4385/14725 [4:00:19<9:23:56,  3.27s/it]
 30%|██▉       | 4386/14725 [4:00:23<9:12:43,  3.21s/it]
 30%|██▉       | 4387/14725 [4:00:26<9:12:02,  3.20s/it]
 30%|██▉       | 4388/14725 [4:00:29<9:06:27,  3.17s/it]
 30%|██▉       | 4389/14725 [4:00:32<9:07:11,  3.18s/it]
 30%|██▉       | 4390/14725 [4:00:35<9:08:25,  3.18s/it]
                                                        

 30%|██▉       | 4390/14725 [4:00:35<9:08:25,  3.18s/it]
 30%|██▉       | 4391/14725 [4:00:39<9:14:27,  3.22s/it]
 30%|██▉       | 4392/14725 [4:
0: {'loss': 0.7254, 'grad_norm': 0.4369769422069704, 'learning_rate': 1e-05, 'epoch': 0.3}
0: 00:42<9:10:27,  3.20s/it]
 30%|██▉       | 4393/14725 [4:00:45<9:22:58,  3.27s/it]
 30%|██▉       | 4394/14725 [4:00:48<9:26:42,  3.29s/it]
 30%|██▉       | 4395/14725 [4:00:52<9:29:57,  3.31s/it]
 30%|██▉       | 4396/14725 [4:00:55<9:17:22,  3.24s/it]
 30%|██▉       | 4397/14725 [4:00:58<9:34:29,  3.34s/it]
 30%|██▉       | 4398/14725 [4:01:02<9:23:56,  3.28s/it]
 30%|██▉       | 4399/14725 [4:01:05<9:18:58,  3.25s/it]
 30%|██▉       | 4400/14725 [4:01:08<9:14:56,  3.22s/it]
                                                        

 30%|██▉       | 4400/14725 [4:01:08<9:14:56,  3.22s/it]
 30%|██▉       | 4401/14725 [4:01:11<9:09:42,  3.19s/it]
 30%|██▉       | 4402/14725 [4:01:14<9:09:36,  3.19s/it]
 30%|██▉       | 4403/14725 [4:01:18<9:21:33,  3.26s/it]
 30%|██▉       | 4404/14725 [4:01:21<9:15:42,  3.23s/it]
 30%|██▉       | 4405/14725 [4:01:24<9:15:11,  3.23s/it]
 30%|██▉       | 4406/14725 [4:01:27<9:16:45,  3.24s
0: {'loss': 0.7193, 'grad_norm': 0.45303874643749653, 'learning_rate': 1e-05, 'epoch': 0.3}
0: /it]
 30%|██▉       | 4407/14725 [4:01:30<9:13:57,  3.22s/it]
 30%|██▉       | 4408/14725 [4:01:34<9:09:47,  3.20s/it]
 30%|██▉       | 4409/14725 [4:01:37<9:20:29,  3.26s/it]
 30%|██▉       | 4410/14725 [4:01:40<9:18:33,  3.25s/it]
                                                        

 30%|██▉       | 4410/14725 [4:01:40<9:18:33,  3.25s/it]
 30%|██▉       | 4411/14725 [4:01:44<9:19:26,  3.25s/it]
 30%|██▉       | 4412/14725 [4:01:47<9:20:18,  3.26s/it]
 30%|██▉       | 4413/14725 [4:01:50<9:24:52,  3.29s/it]
 30%|██▉       | 4414/14725 [4:01:53<9:21:36,  3.27s/it]
 30%|██▉       | 4415/14725 [4:01:56<9:13:07,  3.22s/it]
 30%|██▉       | 4416/14725 [4:02:00<9:24:26,  3.29s/it]
 30%|██▉       | 4417/14725 [4:02:03<9:20:24,  3.26s/it]
 30%|███       | 4418/14725 [4:02:06<9:15:48,  3.24s/it]
 30%|███       | 4419/14725 [4:02:09<9:09:17,  3.20s/it]
 30%|███       | 4420/14725 [4:02:13<9:22:00,  3.27s/it]
                
0: {'loss': 0.7054, 'grad_norm': 0.4580710988884361, 'learning_rate': 1e-05, 'epoch': 0.3}
0: {'loss': 0.7082, 'grad_norm': 0.4265808458406508, 'learning_rate': 1e-05, 'epoch': 0.3}
0:                                         

 30%|███       | 4420/14725 [4:02:13<9:22:00,  3.27s/it]
 30%|███       | 4421/14725 [4:02:16<9:11:59,  3.21s/it]
 30%|███       | 4422/14725 [4:02:19<9:29:43,  3.32s/it]
 30%|███       | 4423/14725 [4:02:23<9:28:13,  3.31s/it]
 30%|███       | 4424/14725 [4:02:26<9:32:40,  3.34s/it]
 30%|███       | 4425/14725 [4:02:29<9:26:05,  3.30s/it]
 30%|███       | 4426/14725 [4:02:32<9:15:25,  3.24s/it]
 30%|███       | 4427/14725 [4:02:36<9:13:04,  3.22s/it]
 30%|███       | 4428/14725 [4:02:39<9:18:09,  3.25s/it]
 30%|███       | 4429/14725 [4:02:42<9:14:48,  3.23s/it]
 30%|███       | 4430/14725 [4:02:46<9:22:13,  3.28s/it]
                                                        

 30%|███       | 4430/14725 [4:02:46<9:22:13,  3.28s/it]
 30%|███       | 4431/14725 [4:02:49<9:19:27,  3.26s/it]
 30%|███       | 4432/14725 [4:02:52<9:13:00,  3.22s/it]
 30%|███       | 4433/14725 [4:02:55
0: {'loss': 0.7141, 'grad_norm': 0.44072381721036874, 'learning_rate': 1e-05, 'epoch': 0.3}
0: <9:13:58,  3.23s/it]
 30%|███       | 4434/14725 [4:02:59<9:25:15,  3.30s/it]
 30%|███       | 4435/14725 [4:03:02<9:18:06,  3.25s/it]
 30%|███       | 4436/14725 [4:03:05<9:18:44,  3.26s/it]
 30%|███       | 4437/14725 [4:03:08<9:25:20,  3.30s/it]
 30%|███       | 4438/14725 [4:03:12<9:16:17,  3.24s/it]
 30%|███       | 4439/14725 [4:03:15<9:14:37,  3.24s/it]
 30%|███       | 4440/14725 [4:03:18<9:12:55,  3.23s/it]
                                                        

 30%|███       | 4440/14725 [4:03:18<9:12:55,  3.23s/it]
 30%|███       | 4441/14725 [4:03:21<9:13:57,  3.23s/it]
 30%|███       | 4442/14725 [4:03:24<9:11:35,  3.22s/it]
 30%|███       | 4443/14725 [4:03:27<9:05:50,  3.19s/it]
 30%|███       | 4444/14725 [4:03:31<9:17:46,  3.26s/it]
 30%|███       | 4445/14725 [4:03:34<9:26:28,  3.31s/it]
 30%|███       | 4446/14725 [4:03:38<9:30:29,  3.33s/it]
 30%|███       | 4447/14725 [4:03:41<9:30:20,  3.33s/it]
0: {'loss': 0.7292, 'grad_norm': 0.47216303793690273, 'learning_rate': 1e-05, 'epoch': 0.3}
0: {'loss': 0.7299, 'grad_norm': 0.46393988661384383, 'learning_rate': 1e-05, 'epoch': 0.3}
0:  30%|███       | 4448/14725 [4:03:44<9:24:34,  3.30s/it]
 30%|███       | 4449/14725 [4:03:48<9:22:52,  3.29s/it]
 30%|███       | 4450/14725 [4:03:51<9:16:14,  3.25s/it]
                                                        

 30%|███       | 4450/14725 [4:03:51<9:16:14,  3.25s/it]
 30%|███       | 4451/14725 [4:03:54<9:27:37,  3.31s/it]
 30%|███       | 4452/14725 [4:03:57<9:17:25,  3.26s/it]
 30%|███       | 4453/14725 [4:04:00<9:10:18,  3.21s/it]
 30%|███       | 4454/14725 [4:04:04<9:20:54,  3.28s/it]
 30%|███       | 4455/14725 [4:04:07<9:36:43,  3.37s/it]
 30%|███       | 4456/14725 [4:04:11<9:33:03,  3.35s/it]
 30%|███       | 4457/14725 [4:04:14<9:35:19,  3.36s/it]
 30%|███       | 4458/14725 [4:04:17<9:32:54,  3.35s/it]
 30%|███       | 4459/14725 [4:04:21<9:39:39,  3.39s/it]
 30%|███       | 4460/14725 [4:04:24<9:38:50,  3.38s/it]
                                                        

 30%|███       | 446
0: {'loss': 0.7105, 'grad_norm': 0.47724991780724835, 'learning_rate': 1e-05, 'epoch': 0.3}
0: 0/14725 [4:04:24<9:38:50,  3.38s/it]
 30%|███       | 4461/14725 [4:04:27<9:27:48,  3.32s/it]
 30%|███       | 4462/14725 [4:04:31<9:17:02,  3.26s/it]
 30%|███       | 4463/14725 [4:04:34<9:21:44,  3.28s/it]
 30%|███       | 4464/14725 [4:04:37<9:24:24,  3.30s/it]
 30%|███       | 4465/14725 [4:04:41<9:29:37,  3.33s/it]
 30%|███       | 4466/14725 [4:04:44<9:34:37,  3.36s/it]
 30%|███       | 4467/14725 [4:04:47<9:24:39,  3.30s/it]
 30%|███       | 4468/14725 [4:04:50<9:16:51,  3.26s/it]
 30%|███       | 4469/14725 [4:04:54<9:17:45,  3.26s/it]
 30%|███       | 4470/14725 [4:04:57<9:17:57,  3.26s/it]
                                                        

 30%|███       | 4470/14725 [4:04:57<9:17:57,  3.26s/it]
 30%|███       | 4471/14725 [4:05:00<9:29:03,  3.33s/it]
 30%|███       | 4472/14725 [4:05:04<9:21:16,  3.28s/it]
 30%|███       | 4473/14725 [4:05:07<9:11:45,  3.23s/it]
 30%|███       | 4474/14725 [4:05:10<9:20
0: {'loss': 0.7159, 'grad_norm': 0.43803461027036333, 'learning_rate': 1e-05, 'epoch': 0.3}
0: :08,  3.28s/it]
 30%|███       | 4475/14725 [4:05:13<9:15:55,  3.25s/it]
 30%|███       | 4476/14725 [4:05:17<9:19:37,  3.28s/it]
 30%|███       | 4477/14725 [4:05:20<9:15:38,  3.25s/it]
 30%|███       | 4478/14725 [4:05:23<9:12:45,  3.24s/it]
 30%|███       | 4479/14725 [4:05:26<9:09:36,  3.22s/it]
 30%|███       | 4480/14725 [4:05:29<9:07:56,  3.21s/it]
                                                        

 30%|███       | 4480/14725 [4:05:29<9:07:56,  3.21s/it]
 30%|███       | 4481/14725 [4:05:33<9:11:20,  3.23s/it]
 30%|███       | 4482/14725 [4:05:36<9:09:00,  3.22s/it]
 30%|███       | 4483/14725 [4:05:39<9:09:31,  3.22s/it]
 30%|███       | 4484/14725 [4:05:42<9:02:33,  3.18s/it]
 30%|███       | 4485/14725 [4:05:45<9:07:47,  3.21s/it]
 30%|███       | 4486/14725 [4:05:49<9:02:57,  3.18s/it]
 30%|███       | 4487/14725 [4:05:52<9:06:54,  3.21s/it]
 30%|███       | 4488/14725 [4:05:55<9:05:07,  3.20s/it]
 30%|
0: {'loss': 0.7047, 'grad_norm': 0.3971172012314693, 'learning_rate': 1e-05, 'epoch': 0.3}
0: {'loss': 0.7189, 'grad_norm': 0.4258082494424412, 'learning_rate': 1e-05, 'epoch': 0.31}
0: ███       | 4489/14725 [4:05:58<9:05:09,  3.20s/it]
 30%|███       | 4490/14725 [4:06:01<9:07:21,  3.21s/it]
                                                        

 30%|███       | 4490/14725 [4:06:01<9:07:21,  3.21s/it]
 30%|███       | 4491/14725 [4:06:05<9:09:42,  3.22s/it]
 31%|███       | 4492/14725 [4:06:08<9:19:57,  3.28s/it]
 31%|███       | 4493/14725 [4:06:11<9:19:13,  3.28s/it]
 31%|███       | 4494/14725 [4:06:15<9:14:45,  3.25s/it]
 31%|███       | 4495/14725 [4:06:18<9:15:09,  3.26s/it]
 31%|███       | 4496/14725 [4:06:21<9:18:02,  3.27s/it]
 31%|███       | 4497/14725 [4:06:24<9:18:28,  3.28s/it]
 31%|███       | 4498/14725 [4:06:28<9:15:18,  3.26s/it]
 31%|███       | 4499/14725 [4:06:31<9:22:59,  3.30s/it]
 31%|███       | 4500/14725 [4:06:34<9:15:25,  3.26s/it]
                                                        

 31%|███       | 4500/14725 [4:06:34<9:15:25,  3.26s/it]
 31%|███       | 4501/147
0: {'loss': 0.7096, 'grad_norm': 0.44511821104213445, 'learning_rate': 1e-05, 'epoch': 0.31}
0: 25 [4:06:38<9:24:24,  3.31s/it]
 31%|███       | 4502/14725 [4:06:41<9:16:16,  3.26s/it]
 31%|███       | 4503/14725 [4:06:44<9:11:42,  3.24s/it]
 31%|███       | 4504/14725 [4:06:47<8:59:18,  3.17s/it]
 31%|███       | 4505/14725 [4:06:50<9:12:48,  3.25s/it]
 31%|███       | 4506/14725 [4:06:54<9:07:45,  3.22s/it]
 31%|███       | 4507/14725 [4:06:57<9:09:25,  3.23s/it]
 31%|███       | 4508/14725 [4:07:00<9:07:44,  3.22s/it]
 31%|███       | 4509/14725 [4:07:03<9:16:58,  3.27s/it]
 31%|███       | 4510/14725 [4:07:07<9:15:43,  3.26s/it]
                                                        

 31%|███       | 4510/14725 [4:07:07<9:15:43,  3.26s/it]
 31%|███       | 4511/14725 [4:07:10<9:12:23,  3.24s/it]
 31%|███       | 4512/14725 [4:07:13<9:08:12,  3.22s/it]
 31%|███       | 4513/14725 [4:07:16<9:10:02,  3.23s/it]
 31%|███       | 4514/14725 [4:07:20<9:18:55,  3.28s/it]
 31%|███       | 4515/14725 [4:07:23<9:16:11, 
0: {'loss': 0.724, 'grad_norm': 0.475489375142479, 'learning_rate': 1e-05, 'epoch': 0.31}
0:  3.27s/it]
 31%|███       | 4516/14725 [4:07:26<9:13:54,  3.26s/it]
 31%|███       | 4517/14725 [4:07:30<9:26:03,  3.33s/it]
 31%|███       | 4518/14725 [4:07:33<9:23:04,  3.31s/it]
 31%|███       | 4519/14725 [4:07:36<9:16:47,  3.27s/it]
 31%|███       | 4520/14725 [4:07:40<9:32:37,  3.37s/it]
                                                        

 31%|███       | 4520/14725 [4:07:40<9:32:37,  3.37s/it]
 31%|███       | 4521/14725 [4:07:43<9:18:11,  3.28s/it]
 31%|███       | 4522/14725 [4:07:46<9:22:38,  3.31s/it]
 31%|███       | 4523/14725 [4:07:49<9:26:02,  3.33s/it]
 31%|███       | 4524/14725 [4:07:53<9:23:26,  3.31s/it]
 31%|███       | 4525/14725 [4:07:56<9:18:35,  3.29s/it]
 31%|███       | 4526/14725 [4:07:59<9:27:43,  3.34s/it]
 31%|███       | 4527/14725 [4:08:03<9:18:12,  3.28s/it]
 31%|███       | 4528/14725 [4:08:06<9:08:37,  3.23s/it]
 31%|███       | 4529/14725 [4:08:09<9:19:07,  3.29s/it]
 31%|█�
0: {'loss': 0.6988, 'grad_norm': 0.4188005856892409, 'learning_rate': 1e-05, 'epoch': 0.31}
0: {'loss': 0.7063, 'grad_norm': 0.4639364941929471, 'learning_rate': 1e-05, 'epoch': 0.31}
0: �█       | 4530/14725 [4:08:12<9:15:13,  3.27s/it]
                                                        

 31%|███       | 4530/14725 [4:08:12<9:15:13,  3.27s/it]
 31%|███       | 4531/14725 [4:08:16<9:17:03,  3.28s/it]
 31%|███       | 4532/14725 [4:08:19<9:24:52,  3.33s/it]
 31%|███       | 4533/14725 [4:08:23<9:35:01,  3.39s/it]
 31%|███       | 4534/14725 [4:08:26<9:31:54,  3.37s/it]
 31%|███       | 4535/14725 [4:08:29<9:17:41,  3.28s/it]
 31%|███       | 4536/14725 [4:08:33<9:29:06,  3.35s/it]
 31%|███       | 4537/14725 [4:08:36<9:21:17,  3.31s/it]
 31%|███       | 4538/14725 [4:08:39<9:14:43,  3.27s/it]
 31%|███       | 4539/14725 [4:08:42<9:16:26,  3.28s/it]
 31%|███       | 4540/14725 [4:08:46<9:25:07,  3.33s/it]
                                                        

 31%|███       | 4540/14725 [4:08:46<9:25:07,  3.33s/it]
 31%|███       | 4541/14725 [4:08:49<9:08:46,  3.23s/it]
 31%|███       | 4542/14725 [4
0: {'loss': 0.7133, 'grad_norm': 0.4278158382977599, 'learning_rate': 1e-05, 'epoch': 0.31}
0: :08:52<9:03:28,  3.20s/it]
 31%|███       | 4543/14725 [4:08:55<9:12:10,  3.25s/it]
 31%|███       | 4544/14725 [4:08:59<9:18:23,  3.29s/it]
 31%|███       | 4545/14725 [4:09:02<9:22:55,  3.32s/it]
 31%|███       | 4546/14725 [4:09:05<9:26:25,  3.34s/it]
 31%|███       | 4547/14725 [4:09:08<9:17:02,  3.28s/it]
 31%|███       | 4548/14725 [4:09:12<9:11:34,  3.25s/it]
 31%|███       | 4549/14725 [4:09:15<9:09:57,  3.24s/it]
 31%|███       | 4550/14725 [4:09:18<9:09:49,  3.24s/it]
                                                        

 31%|███       | 4550/14725 [4:09:18<9:09:49,  3.24s/it]
 31%|███       | 4551/14725 [4:09:21<9:06:27,  3.22s/it]
 31%|███       | 4552/14725 [4:09:25<9:20:30,  3.31s/it]
 31%|███       | 4553/14725 [4:09:28<9:16:50,  3.28s/it]
 31%|███       | 4554/14725 [4:09:31<9:13:59,  3.27s/it]
 31%|███       | 4555/14725 [4:09:34<9:12:31,  3.26s/it]
 31%|███       | 4556/14725 [4:09:38<9:18:39,  3.30
0: {'loss': 0.7132, 'grad_norm': 0.42002455440666, 'learning_rate': 1e-05, 'epoch': 0.31}
0: s/it]
 31%|███       | 4557/14725 [4:09:41<9:15:57,  3.28s/it]
 31%|███       | 4558/14725 [4:09:44<9:08:35,  3.24s/it]
 31%|███       | 4559/14725 [4:09:47<9:08:29,  3.24s/it]
 31%|███       | 4560/14725 [4:09:51<9:03:36,  3.21s/it]
                                                        

 31%|███       | 4560/14725 [4:09:51<9:03:36,  3.21s/it]
 31%|███       | 4561/14725 [4:09:54<9:13:13,  3.27s/it]
 31%|███       | 4562/14725 [4:09:57<9:12:46,  3.26s/it]
 31%|███       | 4563/14725 [4:10:00<9:05:21,  3.22s/it]
 31%|███       | 4564/14725 [4:10:04<9:11:50,  3.26s/it]
 31%|███       | 4565/14725 [4:10:07<9:11:28,  3.26s/it]
 31%|███       | 4566/14725 [4:10:10<9:00:00,  3.19s/it]
 31%|███       | 4567/14725 [4:10:14<9:14:42,  3.28s/it]
 31%|███       | 4568/14725 [4:10:17<9:17:10,  3.29s/it]
 31%|███       | 4569/14725 [4:10:20<9:12:28,  3.26s/it]
 31%|███       | 4570/14725 [4:10:23<9:03:42,  3.21s/it]
               
0: {'loss': 0.7121, 'grad_norm': 0.43920513102479686, 'learning_rate': 1e-05, 'epoch': 0.31}
0: {'loss': 0.7181, 'grad_norm': 0.46278304885966715, 'learning_rate': 1e-05, 'epoch': 0.31}
0:                                          

 31%|███       | 4570/14725 [4:10:23<9:03:42,  3.21s/it]
 31%|███       | 4571/14725 [4:10:26<9:10:36,  3.25s/it]
 31%|███       | 4572/14725 [4:10:30<9:02:34,  3.21s/it]
 31%|███       | 4573/14725 [4:10:33<9:08:51,  3.24s/it]
 31%|███       | 4574/14725 [4:10:36<9:11:15,  3.26s/it]
 31%|███       | 4575/14725 [4:10:40<9:15:47,  3.29s/it]
 31%|███       | 4576/14725 [4:10:43<9:06:05,  3.23s/it]
 31%|███       | 4577/14725 [4:10:46<9:07:21,  3.24s/it]
 31%|███       | 4578/14725 [4:10:49<9:09:39,  3.25s/it]
 31%|███       | 4579/14725 [4:10:52<9:03:11,  3.21s/it]
 31%|███       | 4580/14725 [4:10:56<9:19:09,  3.31s/it]
                                                        

 31%|███       | 4580/14725 [4:10:56<9:19:09,  3.31s/it]
 31%|███       | 4581/14725 [4:10:59<9:29:52,  3.37s/it]
 31%|███       | 4582/14725 [4:11:03<9:31:59,  3.38s/it]
 31%|███       | 4583/14725 [4:11:0
0: {'loss': 0.725, 'grad_norm': 0.4464136635442616, 'learning_rate': 1e-05, 'epoch': 0.31}
0: 6<9:26:13,  3.35s/it]
 31%|███       | 4584/14725 [4:11:09<9:23:40,  3.34s/it]
 31%|███       | 4585/14725 [4:11:13<9:27:31,  3.36s/it]
 31%|███       | 4586/14725 [4:11:16<9:22:00,  3.33s/it]
 31%|███       | 4587/14725 [4:11:20<9:35:15,  3.40s/it]
 31%|███       | 4588/14725 [4:11:23<9:20:29,  3.32s/it]
 31%|███       | 4589/14725 [4:11:26<9:21:18,  3.32s/it]
 31%|███       | 4590/14725 [4:11:29<9:25:35,  3.35s/it]
                                                        

 31%|███       | 4590/14725 [4:11:29<9:25:35,  3.35s/it]
 31%|███       | 4591/14725 [4:11:33<9:11:04,  3.26s/it]
 31%|███       | 4592/14725 [4:11:36<9:03:25,  3.22s/it]
 31%|███       | 4593/14725 [4:11:39<8:57:18,  3.18s/it]
 31%|███       | 4594/14725 [4:11:42<8:58:51,  3.19s/it]
 31%|███       | 4595/14725 [4:11:45<8:56:26,  3.18s/it]
 31%|███       | 4596/14725 [4:11:48<9:03:27,  3.22s/it]
 31%|███       | 4597/14725 [4:11:52<9:04:18,  3.22s/it]
0: {'loss': 0.7076, 'grad_norm': 0.4169287394909493, 'learning_rate': 1e-05, 'epoch': 0.31}
0: {'loss': 0.7236, 'grad_norm': 0.4249451987785673, 'learning_rate': 1e-05, 'epoch': 0.31}
0: 
 31%|███       | 4598/14725 [4:11:55<9:05:56,  3.23s/it]
 31%|███       | 4599/14725 [4:11:58<9:02:46,  3.22s/it]
 31%|███       | 4600/14725 [4:12:01<8:59:40,  3.20s/it]
                                                        

 31%|███       | 4600/14725 [4:12:01<8:59:40,  3.20s/it]
 31%|███       | 4601/14725 [4:12:04<9:00:48,  3.21s/it]
 31%|███▏      | 4602/14725 [4:12:08<9:08:00,  3.25s/it]
 31%|███▏      | 4603/14725 [4:12:11<9:20:27,  3.32s/it]
 31%|███▏      | 4604/14725 [4:12:15<9:29:53,  3.38s/it]
 31%|███▏      | 4605/14725 [4:12:18<9:17:29,  3.31s/it]
 31%|███▏      | 4606/14725 [4:12:21<9:21:42,  3.33s/it]
 31%|███▏      | 4607/14725 [4:12:25<9:19:34,  3.32s/it]
 31%|███▏      | 4608/14725 [4:12:28<9:11:32,  3.27s/it]
 31%|███▏      | 4609/14725 [4:12:31<9:20:19,  3.32s/it]
 31%|███▏      | 4610/14725 [4:12:35<9:29:03,  3.38s/it]
                                                        

 31%|�
0: {'loss': 0.7181, 'grad_norm': 0.4366301306842478, 'learning_rate': 1e-05, 'epoch': 0.31}
0: �██▏      | 4610/14725 [4:12:35<9:29:03,  3.38s/it]
 31%|███▏      | 4611/14725 [4:12:38<9:25:52,  3.36s/it]
 31%|███▏      | 4612/14725 [4:12:41<9:26:41,  3.36s/it]
 31%|███▏      | 4613/14725 [4:12:45<9:15:24,  3.30s/it]
 31%|███▏      | 4614/14725 [4:12:48<9:17:03,  3.31s/it]
 31%|███▏      | 4615/14725 [4:12:51<9:10:45,  3.27s/it]
 31%|███▏      | 4616/14725 [4:12:54<9:09:03,  3.26s/it]
 31%|███▏      | 4617/14725 [4:12:57<9:01:36,  3.21s/it]
 31%|███▏      | 4618/14725 [4:13:01<9:00:04,  3.21s/it]
 31%|███▏      | 4619/14725 [4:13:04<8:55:59,  3.18s/it]
 31%|███▏      | 4620/14725 [4:13:07<8:59:26,  3.20s/it]
                                                        

 31%|███▏      | 4620/14725 [4:13:07<8:59:26,  3.20s/it]
 31%|███▏      | 4621/14725 [4:13:10<9:04:39,  3.23s/it]
 31%|███▏      | 4622/14725 [4:13:13<8:58:47,  3.20s/it]
 31%|███▏      | 4623/14725 [4:13:17<8:57:00,  3.19s/it
0: {'loss': 0.7101, 'grad_norm': 0.4047052114499317, 'learning_rate': 1e-05, 'epoch': 0.31}
0: ]
 31%|███▏      | 4624/14725 [4:13:20<9:01:38,  3.22s/it]
 31%|███▏      | 4625/14725 [4:13:23<8:52:07,  3.16s/it]
 31%|███▏      | 4626/14725 [4:13:26<8:58:46,  3.20s/it]
 31%|███▏      | 4627/14725 [4:13:29<8:54:06,  3.17s/it]
 31%|███▏      | 4628/14725 [4:13:33<9:02:05,  3.22s/it]
 31%|███▏      | 4629/14725 [4:13:36<9:06:11,  3.25s/it]
 31%|███▏      | 4630/14725 [4:13:40<9:24:04,  3.35s/it]
                                                        

 31%|███▏      | 4630/14725 [4:13:40<9:24:04,  3.35s/it]
 31%|███▏      | 4631/14725 [4:13:43<9:20:44,  3.33s/it]
 31%|███▏      | 4632/14725 [4:13:46<9:21:08,  3.34s/it]
 31%|███▏      | 4633/14725 [4:13:50<9:25:33,  3.36s/it]
 31%|███▏      | 4634/14725 [4:13:53<9:27:13,  3.37s/it]
 31%|███▏      | 4635/14725 [4:13:56<9:15:44,  3.30s/it]
 31%|███▏      | 4636/14725 [4:13:59<9:16:13,  3.31s/it]
 31%|███▏      | 4637/14725 [4:14:03<9:21:23, 
0: {'loss': 0.7073, 'grad_norm': 0.4180904688466916, 'learning_rate': 1e-05, 'epoch': 0.32}
0:  3.34s/it]
 31%|███▏      | 4638/14725 [4:14:06<9:10:19,  3.27s/it]
 32%|███▏      | 4639/14725 [4:14:09<9:04:55,  3.24s/it]
 32%|███▏      | 4640/14725 [4:14:12<9:08:57,  3.27s/it]
                                                        

 32%|███▏      | 4640/14725 [4:14:12<9:08:57,  3.27s/it]
 32%|███▏      | 4641/14725 [4:14:16<9:06:51,  3.25s/it]
 32%|███▏      | 4642/14725 [4:14:19<9:12:51,  3.29s/it]
 32%|███▏      | 4643/14725 [4:14:22<9:08:33,  3.26s/it]
 32%|███▏      | 4644/14725 [4:14:25<9:03:35,  3.24s/it]
 32%|███▏      | 4645/14725 [4:14:29<9:04:50,  3.24s/it]
 32%|███▏      | 4646/14725 [4:14:32<8:59:43,  3.21s/it]
 32%|███▏      | 4647/14725 [4:14:35<9:01:27,  3.22s/it]
 32%|███▏      | 4648/14725 [4:14:38<8:55:41,  3.19s/it]
 32%|███▏      | 4649/14725 [4:14:41<9:01:52,  3.23s/it]
 32%|███▏      | 4650/14725 [4:14:45<8:55:13,  3.19s/it]
                                             
0: {'loss': 0.7053, 'grad_norm': 0.44149268180553003, 'learning_rate': 1e-05, 'epoch': 0.32}
0: {'loss': 0.711, 'grad_norm': 0.5296418191794959, 'learning_rate': 1e-05, 'epoch': 0.32}
0:            

 32%|███▏      | 4650/14725 [4:14:45<8:55:13,  3.19s/it]
 32%|███▏      | 4651/14725 [4:14:48<9:03:56,  3.24s/it]
 32%|███▏      | 4652/14725 [4:14:51<9:05:06,  3.25s/it]
 32%|███▏      | 4653/14725 [4:14:54<9:00:47,  3.22s/it]
 32%|███▏      | 4654/14725 [4:14:58<9:10:54,  3.28s/it]
 32%|███▏      | 4655/14725 [4:15:01<9:02:15,  3.23s/it]
 32%|███▏      | 4656/14725 [4:15:04<8:59:50,  3.22s/it]
 32%|███▏      | 4657/14725 [4:15:07<9:06:03,  3.25s/it]
 32%|███▏      | 4658/14725 [4:15:11<9:03:34,  3.24s/it]
 32%|███▏      | 4659/14725 [4:15:14<9:02:34,  3.23s/it]
 32%|███▏      | 4660/14725 [4:15:17<8:58:01,  3.21s/it]
                                                        

 32%|███▏      | 4660/14725 [4:15:17<8:58:01,  3.21s/it]
 32%|███▏      | 4661/14725 [4:15:20<9:06:43,  3.26s/it]
 32%|███▏      | 4662/14725 [4:15:24<9:04:51,  3.25s/it]
 32%|███▏      | 4663/14725 [4:15:2
0: {'loss': 0.7215, 'grad_norm': 0.422362519253561, 'learning_rate': 1e-05, 'epoch': 0.32}
0: 7<9:08:58,  3.27s/it]
 32%|███▏      | 4664/14725 [4:15:30<9:06:05,  3.26s/it]
 32%|███▏      | 4665/14725 [4:15:34<9:16:54,  3.32s/it]
 32%|███▏      | 4666/14725 [4:15:37<9:13:14,  3.30s/it]
 32%|███▏      | 4667/14725 [4:15:40<9:12:21,  3.30s/it]
 32%|███▏      | 4668/14725 [4:15:43<9:13:44,  3.30s/it]
 32%|███▏      | 4669/14725 [4:15:47<9:09:02,  3.28s/it]
 32%|███▏      | 4670/14725 [4:15:50<9:06:36,  3.26s/it]
                                                        

 32%|███▏      | 4670/14725 [4:15:50<9:06:36,  3.26s/it]
 32%|███▏      | 4671/14725 [4:15:53<9:00:06,  3.22s/it]
 32%|███▏      | 4672/14725 [4:15:56<9:08:05,  3.27s/it]
 32%|███▏      | 4673/14725 [4:16:00<9:12:15,  3.30s/it]
 32%|███▏      | 4674/14725 [4:16:03<9:07:19,  3.27s/it]
 32%|███▏      | 4675/14725 [4:16:06<9:09:32,  3.28s/it]
 32%|███▏      | 4676/14725 [4:16:10<9:06:42,  3.26s/it]
 32%|███▏      | 4677/1472
0: {'loss': 0.7147, 'grad_norm': 0.4453369404851961, 'learning_rate': 1e-05, 'epoch': 0.32}
0: 5 [4:16:13<9:04:26,  3.25s/it]
 32%|███▏      | 4678/14725 [4:16:16<9:06:15,  3.26s/it]
 32%|███▏      | 4679/14725 [4:16:19<9:10:31,  3.29s/it]
 32%|███▏      | 4680/14725 [4:16:23<9:04:21,  3.25s/it]
                                                        

 32%|███▏      | 4680/14725 [4:16:23<9:04:21,  3.25s/it]
 32%|███▏      | 4681/14725 [4:16:26<9:14:38,  3.31s/it]
 32%|███▏      | 4682/14725 [4:16:29<9:12:30,  3.30s/it]
 32%|███▏      | 4683/14725 [4:16:32<9:06:36,  3.27s/it]
 32%|███▏      | 4684/14725 [4:16:36<8:58:26,  3.22s/it]
 32%|███▏      | 4685/14725 [4:16:39<8:58:44,  3.22s/it]
 32%|███▏      | 4686/14725 [4:16:42<9:01:54,  3.24s/it]
 32%|███▏      | 4687/14725 [4:16:45<9:06:18,  3.27s/it]
 32%|███▏      | 4688/14725 [4:16:49<9:11:32,  3.30s/it]
 32%|███▏      | 4689/14725 [4:16:52<9:14:27,  3.31s/it]
 32%|███▏      | 4690/14725 [4:16:55<9:10:44,  3.29s/it]
                         
0: {'loss': 0.7123, 'grad_norm': 0.4611256120538188, 'learning_rate': 1e-05, 'epoch': 0.32}
0: {'loss': 0.7199, 'grad_norm': 0.48445289822881427, 'learning_rate': 1e-05, 'epoch': 0.32}
0:                                

 32%|███▏      | 4690/14725 [4:16:55<9:10:44,  3.29s/it]
 32%|███▏      | 4691/14725 [4:16:59<9:12:06,  3.30s/it]
 32%|███▏      | 4692/14725 [4:17:02<9:16:57,  3.33s/it]
 32%|███▏      | 4693/14725 [4:17:05<9:19:19,  3.35s/it]
 32%|███▏      | 4694/14725 [4:17:09<9:13:21,  3.31s/it]
 32%|███▏      | 4695/14725 [4:17:12<9:11:38,  3.30s/it]
 32%|███▏      | 4696/14725 [4:17:15<9:12:14,  3.30s/it]
 32%|███▏      | 4697/14725 [4:17:19<9:12:13,  3.30s/it]
 32%|███▏      | 4698/14725 [4:17:22<9:02:38,  3.25s/it]
 32%|███▏      | 4699/14725 [4:17:25<9:02:08,  3.24s/it]
 32%|███▏      | 4700/14725 [4:17:28<9:08:31,  3.28s/it]
                                                        

 32%|███▏      | 4700/14725 [4:17:28<9:08:31,  3.28s/it]
 32%|███▏      | 4701/14725 [4:17:32<9:09:39,  3.29s/it]
 32%|███▏      | 4702/14725 [4:17:35<8:57:44,  3.22s/it]
 32%|███▏      
0: {'loss': 0.7171, 'grad_norm': 0.43033516287729556, 'learning_rate': 1e-05, 'epoch': 0.32}
0: | 4703/14725 [4:17:38<9:01:49,  3.24s/it]
 32%|███▏      | 4704/14725 [4:17:41<9:03:19,  3.25s/it]
 32%|███▏      | 4705/14725 [4:17:44<9:01:18,  3.24s/it]
 32%|███▏      | 4706/14725 [4:17:48<8:52:04,  3.19s/it]
 32%|███▏      | 4707/14725 [4:17:51<8:55:49,  3.21s/it]
 32%|███▏      | 4708/14725 [4:17:54<8:52:31,  3.19s/it]
 32%|███▏      | 4709/14725 [4:17:57<8:56:10,  3.21s/it]
 32%|███▏      | 4710/14725 [4:18:00<8:52:51,  3.19s/it]
                                                        

 32%|███▏      | 4710/14725 [4:18:00<8:52:51,  3.19s/it]
 32%|███▏      | 4711/14725 [4:18:03<8:48:59,  3.17s/it]
 32%|███▏      | 4712/14725 [4:18:07<8:53:33,  3.20s/it]
 32%|███▏      | 4713/14725 [4:18:10<8:58:54,  3.23s/it]
 32%|███▏      | 4714/14725 [4:18:13<9:03:49,  3.26s/it]
 32%|███▏      | 4715/14725 [4:18:17<9:06:46,  3.28s/it]
 32%|███▏      | 4716/14725 [4:18:20<9:05:00,  3.27s/it]
 32%|███
0: {'loss': 0.7174, 'grad_norm': 0.4436323884169957, 'learning_rate': 1e-05, 'epoch': 0.32}
0: ▏      | 4717/14725 [4:18:23<9:00:04,  3.24s/it]
 32%|███▏      | 4718/14725 [4:18:26<9:02:20,  3.25s/it]
 32%|███▏      | 4719/14725 [4:18:30<8:59:18,  3.23s/it]
 32%|███▏      | 4720/14725 [4:18:33<9:09:11,  3.29s/it]
                                                        

 32%|███▏      | 4720/14725 [4:18:33<9:09:11,  3.29s/it]
 32%|███▏      | 4721/14725 [4:18:36<9:08:00,  3.29s/it]
 32%|███▏      | 4722/14725 [4:18:40<9:05:26,  3.27s/it]
 32%|███▏      | 4723/14725 [4:18:43<9:01:19,  3.25s/it]
 32%|███▏      | 4724/14725 [4:18:46<9:05:31,  3.27s/it]
 32%|███▏      | 4725/14725 [4:18:49<8:58:41,  3.23s/it]
 32%|███▏      | 4726/14725 [4:18:52<8:51:59,  3.19s/it]
 32%|███▏      | 4727/14725 [4:18:55<8:51:07,  3.19s/it]
 32%|███▏      | 4728/14725 [4:18:59<8:54:45,  3.21s/it]
 32%|███▏      | 4729/14725 [4:19:02<9:00:52,  3.25s/it]
 32%|███▏      | 4730/14725 [4:19:05<8:54:03,  3.21s/it]
     
0: {'loss': 0.7229, 'grad_norm': 0.42538216092160824, 'learning_rate': 1e-05, 'epoch': 0.32}
0: {'loss': 0.7269, 'grad_norm': 0.45518380037639355, 'learning_rate': 1e-05, 'epoch': 0.32}
0:                                                    

 32%|███▏      | 4730/14725 [4:19:05<8:54:03,  3.21s/it]
 32%|███▏      | 4731/14725 [4:19:08<8:53:35,  3.20s/it]
 32%|███▏      | 4732/14725 [4:19:12<8:54:51,  3.21s/it]
 32%|███▏      | 4733/14725 [4:19:15<8:55:41,  3.22s/it]
 32%|███▏      | 4734/14725 [4:19:18<8:52:56,  3.20s/it]
 32%|███▏      | 4735/14725 [4:19:21<8:46:44,  3.16s/it]
 32%|███▏      | 4736/14725 [4:19:24<8:56:50,  3.22s/it]
 32%|███▏      | 4737/14725 [4:19:28<8:55:07,  3.21s/it]
 32%|███▏      | 4738/14725 [4:19:31<9:01:39,  3.25s/it]
 32%|███▏      | 4739/14725 [4:19:35<9:16:19,  3.34s/it]
 32%|███▏      | 4740/14725 [4:19:38<9:05:24,  3.28s/it]
                                                        

 32%|███▏      | 4740/14725 [4:19:38<9:05:24,  3.28s/it]
 32%|███▏      | 4741/14725 [4:19:41<9:17:30,  3.35s/it]
 32%|███▏      | 4742/14725 [4:19:44<9:07:14,  3.29s/it]
 32
0: {'loss': 0.7169, 'grad_norm': 0.40898476918952953, 'learning_rate': 1e-05, 'epoch': 0.32}
0: %|███▏      | 4743/14725 [4:19:48<9:04:21,  3.27s/it]
 32%|███▏      | 4744/14725 [4:19:51<9:04:25,  3.27s/it]
 32%|███▏      | 4745/14725 [4:19:54<8:59:52,  3.25s/it]
 32%|███▏      | 4746/14725 [4:19:57<9:08:42,  3.30s/it]
 32%|███▏      | 4747/14725 [4:20:01<9:05:44,  3.28s/it]
 32%|███▏      | 4748/14725 [4:20:04<9:00:42,  3.25s/it]
 32%|███▏      | 4749/14725 [4:20:07<9:10:32,  3.31s/it]
 32%|███▏      | 4750/14725 [4:20:11<9:10:06,  3.31s/it]
                                                        

 32%|███▏      | 4750/14725 [4:20:11<9:10:06,  3.31s/it]
 32%|███▏      | 4751/14725 [4:20:14<9:08:27,  3.30s/it]
 32%|███▏      | 4752/14725 [4:20:17<9:00:30,  3.25s/it]
 32%|███▏      | 4753/14725 [4:20:20<8:49:58,  3.19s/it]
 32%|███▏      | 4754/14725 [4:20:23<8:54:10,  3.21s/it]
 32%|███▏      | 4755/14725 [4:20:27<8:53:14,  3.21s/it]
 32%|███▏      | 4756/14725 [4:20:30<8:49:45,  3.19
0: {'loss': 0.7113, 'grad_norm': 0.44994977206879117, 'learning_rate': 1e-05, 'epoch': 0.32}
0: s/it]
 32%|███▏      | 4757/14725 [4:20:33<8:58:00,  3.24s/it]
 32%|███▏      | 4758/14725 [4:20:36<8:56:49,  3.23s/it]
 32%|███▏      | 4759/14725 [4:20:39<8:53:47,  3.21s/it]
 32%|███▏      | 4760/14725 [4:20:43<8:54:58,  3.22s/it]
                                                        

 32%|███▏      | 4760/14725 [4:20:43<8:54:58,  3.22s/it]
 32%|███▏      | 4761/14725 [4:20:46<8:54:32,  3.22s/it]
 32%|███▏      | 4762/14725 [4:20:49<9:03:08,  3.27s/it]
 32%|███▏      | 4763/14725 [4:20:53<9:12:46,  3.33s/it]
 32%|███▏      | 4764/14725 [4:20:56<9:10:20,  3.32s/it]
 32%|███▏      | 4765/14725 [4:20:59<9:14:05,  3.34s/it]
 32%|███▏      | 4766/14725 [4:21:03<9:07:30,  3.30s/it]
 32%|███▏      | 4767/14725 [4:21:06<9:06:49,  3.29s/it]
 32%|███▏      | 4768/14725 [4:21:09<9:00:44,  3.26s/it]
 32%|███▏      | 4769/14725 [4:21:12<8:59:45,  3.25s/it]
 32%|███▏      | 4770/14725 [4:21:16<9:00:
0: {'loss': 0.7227, 'grad_norm': 0.4248800592081502, 'learning_rate': 1e-05, 'epoch': 0.32}
0: {'loss': 0.7003, 'grad_norm': 0.42738515360810164, 'learning_rate': 1e-05, 'epoch': 0.32}
0: 53,  3.26s/it]
                                                        

 32%|███▏      | 4770/14725 [4:21:16<9:00:53,  3.26s/it]
 32%|███▏      | 4771/14725 [4:21:19<8:57:06,  3.24s/it]
 32%|███▏      | 4772/14725 [4:21:22<8:56:17,  3.23s/it]
 32%|███▏      | 4773/14725 [4:21:25<9:04:31,  3.28s/it]
 32%|███▏      | 4774/14725 [4:21:29<9:00:54,  3.26s/it]
 32%|███▏      | 4775/14725 [4:21:32<9:13:19,  3.34s/it]
 32%|███▏      | 4776/14725 [4:21:35<9:03:21,  3.28s/it]
 32%|███▏      | 4777/14725 [4:21:38<8:51:54,  3.21s/it]
 32%|███▏      | 4778/14725 [4:21:42<8:52:16,  3.21s/it]
 32%|███▏      | 4779/14725 [4:21:45<9:00:26,  3.26s/it]
 32%|███▏      | 4780/14725 [4:21:48<8:52:22,  3.21s/it]
                                                        

 32%|███▏      | 4780/14725 [4:21:48<8:52:22,  3.21s/it]
 32%|███▏      | 4781/14725 [4:21:51<8:56:18,  3.24s/it]
 32%|███▏      | 4782/14725 [4:21:54<8:5
0: {'loss': 0.7135, 'grad_norm': 0.4233199159897108, 'learning_rate': 1e-05, 'epoch': 0.33}
0: 4:45,  3.23s/it]
 32%|███▏      | 4783/14725 [4:21:58<8:54:04,  3.22s/it]
 32%|███▏      | 4784/14725 [4:22:01<8:55:54,  3.23s/it]
 32%|███▏      | 4785/14725 [4:22:05<9:14:58,  3.35s/it]
 33%|███▎      | 4786/14725 [4:22:08<9:04:40,  3.29s/it]
 33%|███▎      | 4787/14725 [4:22:11<9:01:21,  3.27s/it]
 33%|███▎      | 4788/14725 [4:22:14<9:11:08,  3.33s/it]
 33%|███▎      | 4789/14725 [4:22:18<9:10:57,  3.33s/it]
 33%|███▎      | 4790/14725 [4:22:21<9:12:18,  3.34s/it]
                                                        

 33%|███▎      | 4790/14725 [4:22:21<9:12:18,  3.34s/it]
 33%|███▎      | 4791/14725 [4:22:24<9:05:17,  3.29s/it]
 33%|███▎      | 4792/14725 [4:22:28<9:05:41,  3.30s/it]
 33%|███▎      | 4793/14725 [4:22:31<9:05:02,  3.29s/it]
 33%|███▎      | 4794/14725 [4:22:34<9:04:23,  3.29s/it]
 33%|███▎      | 4795/14725 [4:22:37<8:57:32,  3.25s/it]
 33%|███▎      | 4796/14725 [4:
0: {'loss': 0.6959, 'grad_norm': 0.45389034547469825, 'learning_rate': 1e-05, 'epoch': 0.33}
0: 22:41<9:01:32,  3.27s/it]
 33%|███▎      | 4797/14725 [4:22:44<8:55:35,  3.24s/it]
 33%|███▎      | 4798/14725 [4:22:47<8:57:28,  3.25s/it]
 33%|███▎      | 4799/14725 [4:22:50<9:00:29,  3.27s/it]
 33%|███▎      | 4800/14725 [4:22:53<8:52:15,  3.22s/it]
                                                        

 33%|███▎      | 4800/14725 [4:22:53<8:52:15,  3.22s/it]
 33%|███▎      | 4801/14725 [4:22:57<8:58:39,  3.26s/it]
 33%|███▎      | 4802/14725 [4:23:00<8:55:39,  3.24s/it]
 33%|███▎      | 4803/14725 [4:23:04<9:10:46,  3.33s/it]
 33%|███▎      | 4804/14725 [4:23:07<9:15:12,  3.36s/it]
 33%|███▎      | 4805/14725 [4:23:10<9:21:52,  3.40s/it]
 33%|███▎      | 4806/14725 [4:23:14<9:07:18,  3.31s/it]
 33%|███▎      | 4807/14725 [4:23:17<9:01:38,  3.28s/it]
 33%|███▎      | 4808/14725 [4:23:20<8:49:32,  3.20s/it]
 33%|███▎      | 4809/14725 [4:23:23<8:53:08,  3.23s/it]
 33%|███▎      | 4810/
0: {'loss': 0.7171, 'grad_norm': 0.44823123495542905, 'learning_rate': 1e-05, 'epoch': 0.33}
0: {'loss': 0.7205, 'grad_norm': 0.4368126596598621, 'learning_rate': 1e-05, 'epoch': 0.33}
0: 14725 [4:23:27<9:04:29,  3.29s/it]
                                                        

 33%|███▎      | 4810/14725 [4:23:27<9:04:29,  3.29s/it]
 33%|███▎      | 4811/14725 [4:23:30<8:56:47,  3.25s/it]
 33%|███▎      | 4812/14725 [4:23:33<9:06:33,  3.31s/it]
 33%|███▎      | 4813/14725 [4:23:37<9:12:39,  3.35s/it]
 33%|███▎      | 4814/14725 [4:23:40<9:10:45,  3.33s/it]
 33%|███▎      | 4815/14725 [4:23:43<9:07:24,  3.31s/it]
 33%|███▎      | 4816/14725 [4:23:46<9:03:01,  3.29s/it]
 33%|███▎      | 4817/14725 [4:23:50<9:04:48,  3.30s/it]
 33%|███▎      | 4818/14725 [4:23:53<9:14:12,  3.36s/it]
 33%|███▎      | 4819/14725 [4:23:56<9:11:49,  3.34s/it]
 33%|███▎      | 4820/14725 [4:24:00<9:12:14,  3.35s/it]
                                                        

 33%|███▎      | 4820/14725 [4:24:00<9:12:14,  3.35s/it]
 33%|███▎      | 4821/14725 [4:24:03<9:06:38,  3.31s/it]
 33%|███▎      | 482
0: {'loss': 0.7045, 'grad_norm': 0.43814498342827946, 'learning_rate': 1e-05, 'epoch': 0.33}
0: 2/14725 [4:24:06<9:04:22,  3.30s/it]
 33%|███▎      | 4823/14725 [4:24:10<8:57:46,  3.26s/it]
 33%|███▎      | 4824/14725 [4:24:13<8:51:01,  3.22s/it]
 33%|███▎      | 4825/14725 [4:24:16<8:54:26,  3.24s/it]
 33%|███▎      | 4826/14725 [4:24:19<8:53:18,  3.23s/it]
 33%|███▎      | 4827/14725 [4:24:22<8:47:13,  3.20s/it]
 33%|███▎      | 4828/14725 [4:24:25<8:38:43,  3.14s/it]
 33%|███▎      | 4829/14725 [4:24:28<8:36:22,  3.13s/it]
 33%|███▎      | 4830/14725 [4:24:32<8:48:38,  3.21s/it]
                                                        

 33%|███▎      | 4830/14725 [4:24:32<8:48:38,  3.21s/it]
 33%|███▎      | 4831/14725 [4:24:35<8:42:04,  3.17s/it]
 33%|███▎      | 4832/14725 [4:24:38<8:40:31,  3.16s/it]
 33%|███▎      | 4833/14725 [4:24:42<9:01:03,  3.28s/it]
 33%|███▎      | 4834/14725 [4:24:45<9:17:21,  3.38s/it]
 33%|███▎      | 4835/14725 [4:24:49<9:19:49,  3.40s/it]
 33%|███▎  
0: {'loss': 0.711, 'grad_norm': 0.4101945591243011, 'learning_rate': 1e-05, 'epoch': 0.33}
0:     | 4836/14725 [4:24:52<9:17:07,  3.38s/it]
 33%|███▎      | 4837/14725 [4:24:55<9:05:36,  3.31s/it]
 33%|███▎      | 4838/14725 [4:24:59<9:11:35,  3.35s/it]
 33%|███▎      | 4839/14725 [4:25:02<9:11:45,  3.35s/it]
 33%|███▎      | 4840/14725 [4:25:05<9:10:43,  3.34s/it]
                                                        

 33%|███▎      | 4840/14725 [4:25:05<9:10:43,  3.34s/it]
 33%|███▎      | 4841/14725 [4:25:09<9:23:07,  3.42s/it]
 33%|███▎      | 4842/14725 [4:25:12<9:28:15,  3.45s/it]
 33%|███▎      | 4843/14725 [4:25:16<9:23:12,  3.42s/it]
 33%|███▎      | 4844/14725 [4:25:19<9:32:37,  3.48s/it]
 33%|███▎      | 4845/14725 [4:25:22<9:15:10,  3.37s/it]
 33%|███▎      | 4846/14725 [4:25:26<9:07:42,  3.33s/it]
 33%|███▎      | 4847/14725 [4:25:29<9:00:34,  3.28s/it]
 33%|███▎      | 4848/14725 [4:25:32<9:03:25,  3.30s/it]
 33%|███▎      | 4849/14725 [4:25:35<9:05:47,  3.32s/it]
 33%|█�
0: {'loss': 0.7053, 'grad_norm': 0.4467032497545692, 'learning_rate': 1e-05, 'epoch': 0.33}
0: {'loss': 0.7133, 'grad_norm': 0.4595062593841621, 'learning_rate': 1e-05, 'epoch': 0.33}
0: �█▎      | 4850/14725 [4:25:39<9:00:36,  3.28s/it]
                                                        

 33%|███▎      | 4850/14725 [4:25:39<9:00:36,  3.28s/it]
 33%|███▎      | 4851/14725 [4:25:42<8:50:34,  3.22s/it]
 33%|███▎      | 4852/14725 [4:25:45<9:06:27,  3.32s/it]
 33%|███▎      | 4853/14725 [4:25:49<8:59:05,  3.28s/it]
 33%|███▎      | 4854/14725 [4:25:52<9:03:31,  3.30s/it]
 33%|███▎      | 4855/14725 [4:25:55<8:54:15,  3.25s/it]
 33%|███▎      | 4856/14725 [4:25:58<8:56:44,  3.26s/it]
 33%|███▎      | 4857/14725 [4:26:02<8:57:28,  3.27s/it]
 33%|███▎      | 4858/14725 [4:26:05<8:57:08,  3.27s/it]
 33%|███▎      | 4859/14725 [4:26:08<8:49:47,  3.22s/it]
 33%|███▎      | 4860/14725 [4:26:12<9:08:42,  3.34s/it]
                                                        

 33%|███▎      | 4860/14725 [4:26:12<9:08:42,  3.34s/it]
 33%|███▎      | 4861/14725 [4:26:15<9:08:35,  3.34s/it]
 33%|█
0: {'loss': 0.709, 'grad_norm': 0.39489982278278246, 'learning_rate': 1e-05, 'epoch': 0.33}
0: ██▎      | 4862/14725 [4:26:18<8:54:33,  3.25s/it]
 33%|███▎      | 4863/14725 [4:26:21<9:02:13,  3.30s/it]
 33%|███▎      | 4864/14725 [4:26:25<9:09:34,  3.34s/it]
 33%|███▎      | 4865/14725 [4:26:28<9:14:34,  3.37s/it]
 33%|███▎      | 4866/14725 [4:26:31<9:04:44,  3.32s/it]
 33%|███▎      | 4867/14725 [4:26:35<9:08:42,  3.34s/it]
 33%|███▎      | 4868/14725 [4:26:38<9:02:38,  3.30s/it]
 33%|███▎      | 4869/14725 [4:26:41<8:59:53,  3.29s/it]
 33%|███▎      | 4870/14725 [4:26:45<9:11:24,  3.36s/it]
                                                        

 33%|███▎      | 4870/14725 [4:26:45<9:11:24,  3.36s/it]
 33%|███▎      | 4871/14725 [4:26:48<9:07:00,  3.33s/it]
 33%|███▎      | 4872/14725 [4:26:51<9:07:45,  3.34s/it]
 33%|███▎      | 4873/14725 [4:26:55<9:00:01,  3.29s/it]
 33%|███▎      | 4874/14725 [4:26:58<9:00:26,  3.29s/it]
 33%|███▎      | 4875/14725 [4:27:01<9:05:33,  3.32s/it]
0: {'loss': 0.7185, 'grad_norm': 0.4115913808485106, 'learning_rate': 1e-05, 'epoch': 0.33}
0: 
 33%|███▎      | 4876/14725 [4:27:05<9:02:14,  3.30s/it]
 33%|███▎      | 4877/14725 [4:27:08<8:58:00,  3.28s/it]
 33%|███▎      | 4878/14725 [4:27:11<9:01:26,  3.30s/it]
 33%|███▎      | 4879/14725 [4:27:14<9:00:47,  3.30s/it]
 33%|███▎      | 4880/14725 [4:27:17<8:46:58,  3.21s/it]
                                                        

 33%|███▎      | 4880/14725 [4:27:17<8:46:58,  3.21s/it]
 33%|███▎      | 4881/14725 [4:27:21<8:48:18,  3.22s/it]
 33%|███▎      | 4882/14725 [4:27:24<8:49:49,  3.23s/it]
 33%|███▎      | 4883/14725 [4:27:27<8:45:21,  3.20s/it]
 33%|███▎      | 4884/14725 [4:27:31<9:13:33,  3.38s/it]
 33%|███▎      | 4885/14725 [4:27:34<9:14:55,  3.38s/it]
 33%|███▎      | 4886/14725 [4:27:37<9:00:50,  3.30s/it]
 33%|███▎      | 4887/14725 [4:27:41<8:56:07,  3.27s/it]
 33%|███▎      | 4888/14725 [4:27:44<9:05:50,  3.33s/it]
 33%|███▎      | 4889/14725 [4:27:47<9:08:06,  
0: {'loss': 0.724, 'grad_norm': 0.4290673262277478, 'learning_rate': 1e-05, 'epoch': 0.33}
0: {'loss': 0.719, 'grad_norm': 0.44515585208137504, 'learning_rate': 1e-05, 'epoch': 0.33}
0: 3.34s/it]
 33%|███▎      | 4890/14725 [4:27:51<9:12:55,  3.37s/it]
                                                        

 33%|███▎      | 4890/14725 [4:27:51<9:12:55,  3.37s/it]
 33%|███▎      | 4891/14725 [4:27:54<9:01:01,  3.30s/it]
 33%|███▎      | 4892/14725 [4:27:57<9:02:18,  3.31s/it]
 33%|███▎      | 4893/14725 [4:28:00<8:51:25,  3.24s/it]
 33%|███▎      | 4894/14725 [4:28:04<8:49:51,  3.23s/it]
 33%|███▎      | 4895/14725 [4:28:07<8:58:32,  3.29s/it]
 33%|███▎      | 4896/14725 [4:28:10<9:07:11,  3.34s/it]
 33%|███▎      | 4897/14725 [4:28:14<9:11:12,  3.37s/it]
 33%|███▎      | 4898/14725 [4:28:17<9:09:46,  3.36s/it]
 33%|███▎      | 4899/14725 [4:28:21<9:13:48,  3.38s/it]
 33%|███▎      | 4900/14725 [4:28:24<9:08:16,  3.35s/it]
                                                        

 33%|███▎      | 4900/14725 [4:28:24<9:08:16,  3.35s/it]
 33%|███▎      | 4901/14725 [4:28:27<9:05:33,
0: {'loss': 0.7306, 'grad_norm': 0.4725078519324119, 'learning_rate': 1e-05, 'epoch': 0.33}
0:   3.33s/it]
 33%|███▎      | 4902/14725 [4:28:30<9:01:14,  3.31s/it]
 33%|███▎      | 4903/14725 [4:28:34<9:00:14,  3.30s/it]
 33%|███▎      | 4904/14725 [4:28:37<8:52:39,  3.25s/it]
 33%|███▎      | 4905/14725 [4:28:40<8:38:58,  3.17s/it]
 33%|███▎      | 4906/14725 [4:28:43<8:55:56,  3.27s/it]
 33%|███▎      | 4907/14725 [4:28:47<8:50:37,  3.24s/it]
 33%|███▎      | 4908/14725 [4:28:50<8:55:10,  3.27s/it]
 33%|███▎      | 4909/14725 [4:28:53<8:45:31,  3.21s/it]
 33%|███▎      | 4910/14725 [4:28:56<8:50:03,  3.24s/it]
                                                        

 33%|███▎      | 4910/14725 [4:28:56<8:50:03,  3.24s/it]
 33%|███▎      | 4911/14725 [4:29:00<9:00:55,  3.31s/it]
 33%|███▎      | 4912/14725 [4:29:03<9:06:03,  3.34s/it]
 33%|███▎      | 4913/14725 [4:29:07<9:06:55,  3.34s/it]
 33%|███▎      | 4914/14725 [4:29:10<9:00:21,  3.30s/it]
 33%|███▎      | 4915/14725 [4:29:13
0: {'loss': 0.7153, 'grad_norm': 0.4253665222031506, 'learning_rate': 1e-05, 'epoch': 0.33}
0: <9:01:22,  3.31s/it]
 33%|███▎      | 4916/14725 [4:29:16<8:58:17,  3.29s/it]
 33%|███▎      | 4917/14725 [4:29:20<9:00:34,  3.31s/it]
 33%|███▎      | 4918/14725 [4:29:23<8:54:39,  3.27s/it]
 33%|███▎      | 4919/14725 [4:29:26<8:50:32,  3.25s/it]
 33%|███▎      | 4920/14725 [4:29:29<8:48:49,  3.24s/it]
                                                        

 33%|███▎      | 4920/14725 [4:29:29<8:48:49,  3.24s/it]
 33%|███▎      | 4921/14725 [4:29:33<8:50:12,  3.24s/it]
 33%|███▎      | 4922/14725 [4:29:36<8:53:26,  3.26s/it]
 33%|███▎      | 4923/14725 [4:29:39<8:51:38,  3.25s/it]
 33%|███▎      | 4924/14725 [4:29:42<8:53:10,  3.26s/it]
 33%|███▎      | 4925/14725 [4:29:46<8:50:33,  3.25s/it]
 33%|███▎      | 4926/14725 [4:29:49<8:51:23,  3.25s/it]
 33%|███▎      | 4927/14725 [4:29:52<8:58:11,  3.30s/it]
 33%|███▎      | 4928/14725 [4:29:55<8:54:19,  3.27s/it]
 33%|███▎      | 4929/14725
0: {'loss': 0.714, 'grad_norm': 0.42474182526396015, 'learning_rate': 1e-05, 'epoch': 0.33}
0: {'loss': 0.73, 'grad_norm': 0.507296951360673, 'learning_rate': 1e-05, 'epoch': 0.34}
0:  [4:29:59<9:01:12,  3.31s/it]
 33%|███▎      | 4930/14725 [4:30:02<8:53:38,  3.27s/it]
                                                        

 33%|███▎      | 4930/14725 [4:30:02<8:53:38,  3.27s/it]
 33%|███▎      | 4931/14725 [4:30:05<8:54:03,  3.27s/it]
 33%|███▎      | 4932/14725 [4:30:08<8:48:40,  3.24s/it]
 34%|███▎      | 4933/14725 [4:30:12<8:57:07,  3.29s/it]
 34%|███▎      | 4934/14725 [4:30:15<8:53:55,  3.27s/it]
 34%|███▎      | 4935/14725 [4:30:19<9:03:17,  3.33s/it]
 34%|███▎      | 4936/14725 [4:30:22<9:00:26,  3.31s/it]
 34%|███▎      | 4937/14725 [4:30:25<9:00:07,  3.31s/it]
 34%|███▎      | 4938/14725 [4:30:28<8:54:36,  3.28s/it]
 34%|███▎      | 4939/14725 [4:30:31<8:47:03,  3.23s/it]
 34%|███▎      | 4940/14725 [4:30:35<8:46:25,  3.23s/it]
                                                        

 34%|███▎      | 4940/14725 [4:30:35<8:46:25,  3.23s/it]
 34%|███▎      | 4941/147
0: {'loss': 0.717, 'grad_norm': 0.4645222237471658, 'learning_rate': 1e-05, 'epoch': 0.34}
0: 25 [4:30:38<8:47:05,  3.23s/it]
 34%|███▎      | 4942/14725 [4:30:41<8:48:11,  3.24s/it]
 34%|███▎      | 4943/14725 [4:30:44<8:50:00,  3.25s/it]
 34%|███▎      | 4944/14725 [4:30:48<8:47:52,  3.24s/it]
 34%|███▎      | 4945/14725 [4:30:51<8:56:00,  3.29s/it]
 34%|███▎      | 4946/14725 [4:30:54<8:52:35,  3.27s/it]
 34%|███▎      | 4947/14725 [4:30:58<8:57:37,  3.30s/it]
 34%|███▎      | 4948/14725 [4:31:01<8:50:18,  3.25s/it]
 34%|███▎      | 4949/14725 [4:31:04<8:47:00,  3.23s/it]
 34%|███▎      | 4950/14725 [4:31:07<8:51:19,  3.26s/it]
                                                        

 34%|███▎      | 4950/14725 [4:31:07<8:51:19,  3.26s/it]
 34%|███▎      | 4951/14725 [4:31:11<8:48:53,  3.25s/it]
 34%|███▎      | 4952/14725 [4:31:14<8:46:22,  3.23s/it]
 34%|███▎      | 4953/14725 [4:31:17<8:46:14,  3.23s/it]
 34%|███▎      | 4954/14725 [4:31:20<8:43:43,  3.22s/it]
 34%|███▎      |
0: {'loss': 0.7093, 'grad_norm': 0.44886835629887706, 'learning_rate': 1e-05, 'epoch': 0.34}
0:  4955/14725 [4:31:23<8:45:35,  3.23s/it]
 34%|███▎      | 4956/14725 [4:31:27<8:40:53,  3.20s/it]
 34%|███▎      | 4957/14725 [4:31:30<8:47:51,  3.24s/it]
 34%|███▎      | 4958/14725 [4:31:33<8:45:18,  3.23s/it]
 34%|███▎      | 4959/14725 [4:31:36<8:47:48,  3.24s/it]
 34%|███▎      | 4960/14725 [4:31:39<8:43:32,  3.22s/it]
                                                        

 34%|███▎      | 4960/14725 [4:31:39<8:43:32,  3.22s/it]
 34%|███▎      | 4961/14725 [4:31:43<8:47:51,  3.24s/it]
 34%|███▎      | 4962/14725 [4:31:46<9:00:17,  3.32s/it]
 34%|███▎      | 4963/14725 [4:31:50<8:57:12,  3.30s/it]
 34%|███▎      | 4964/14725 [4:31:53<8:56:35,  3.30s/it]
 34%|███▎      | 4965/14725 [4:31:56<9:01:03,  3.33s/it]
 34%|███▎      | 4966/14725 [4:32:00<9:00:43,  3.32s/it]
 34%|███▎      | 4967/14725 [4:32:03<9:00:08,  3.32s/it]
 34%|███▎      | 4968/14725 [4:32:06<8:47:06,  3.24s/it]
 34%|███�
0: {'loss': 0.7108, 'grad_norm': 0.40430145345691004, 'learning_rate': 1e-05, 'epoch': 0.34}
0: {'loss': 0.7143, 'grad_norm': 0.4355535052602201, 'learning_rate': 1e-05, 'epoch': 0.34}
0: ��      | 4969/14725 [4:32:09<8:59:37,  3.32s/it]
 34%|███▍      | 4970/14725 [4:32:13<8:51:05,  3.27s/it]
                                                        

 34%|███▍      | 4970/14725 [4:32:13<8:51:05,  3.27s/it]
 34%|███▍      | 4971/14725 [4:32:16<8:46:57,  3.24s/it]
 34%|███▍      | 4972/14725 [4:32:19<8:42:58,  3.22s/it]
 34%|███▍      | 4973/14725 [4:32:22<8:43:56,  3.22s/it]
 34%|███▍      | 4974/14725 [4:32:26<8:56:17,  3.30s/it]
 34%|███▍      | 4975/14725 [4:32:29<8:55:34,  3.30s/it]
 34%|███▍      | 4976/14725 [4:32:32<8:54:40,  3.29s/it]
 34%|███▍      | 4977/14725 [4:32:35<8:42:41,  3.22s/it]
 34%|███▍      | 4978/14725 [4:32:38<8:36:47,  3.18s/it]
 34%|███▍      | 4979/14725 [4:32:42<8:46:21,  3.24s/it]
 34%|███▍      | 4980/14725 [4:32:45<8:43:58,  3.23s/it]
                                                        

 34%|███▍      | 4980/14725 [4:32:45<8:43:58,  3.23s/it]
 34%|██�
0: {'loss': 0.7099, 'grad_norm': 0.4079031998087741, 'learning_rate': 1e-05, 'epoch': 0.34}
0: �▍      | 4981/14725 [4:32:48<8:40:11,  3.20s/it]
 34%|███▍      | 4982/14725 [4:32:51<8:33:04,  3.16s/it]
 34%|███▍      | 4983/14725 [4:32:54<8:37:52,  3.19s/it]
 34%|███▍      | 4984/14725 [4:32:58<8:40:55,  3.21s/it]
 34%|███▍      | 4985/14725 [4:33:01<8:54:35,  3.29s/it]
 34%|███▍      | 4986/14725 [4:33:04<8:52:57,  3.28s/it]
 34%|███▍      | 4987/14725 [4:33:08<8:46:16,  3.24s/it]
 34%|███▍      | 4988/14725 [4:33:11<8:46:05,  3.24s/it]
 34%|███▍      | 4989/14725 [4:33:14<8:43:19,  3.23s/it]
 34%|███▍      | 4990/14725 [4:33:17<8:46:43,  3.25s/it]
                                                        

 34%|███▍      | 4990/14725 [4:33:17<8:46:43,  3.25s/it]
 34%|███▍      | 4991/14725 [4:33:21<8:56:51,  3.31s/it]
 34%|███▍      | 4992/14725 [4:33:24<8:51:29,  3.28s/it]
 34%|███▍      | 4993/14725 [4:33:27<8:52:29,  3.28s/it]
 34%|███▍      | 4994/14725 [4:33:30<8:48:22,  3.26s/it]
 34%
0: {'loss': 0.71, 'grad_norm': 0.4382831520989252, 'learning_rate': 1e-05, 'epoch': 0.34}
0: |███▍      | 4995/14725 [4:33:34<8:50:48,  3.27s/it]
 34%|███▍      | 4996/14725 [4:33:37<8:45:45,  3.24s/it]
 34%|███▍      | 4997/14725 [4:33:40<8:57:15,  3.31s/it]
 34%|███▍      | 4998/14725 [4:33:44<8:50:25,  3.27s/it]
 34%|███▍      | 4999/14725 [4:33:47<8:44:01,  3.23s/it]
 34%|███▍      | 5000/14725 [4:33:50<8:46:27,  3.25s/it]
                                                        

 34%|███▍      | 5000/14725 [4:33:50<8:46:27,  3.25s/it]
 34%|███▍      | 5001/14725 [4:33:53<8:42:08,  3.22s/it]
 34%|███▍      | 5002/14725 [4:33:56<8:43:50,  3.23s/it]
 34%|███▍      | 5003/14725 [4:34:00<8:41:54,  3.22s/it]
 34%|███▍      | 5004/14725 [4:34:03<8:40:38,  3.21s/it]
 34%|███▍      | 5005/14725 [4:34:06<8:39:12,  3.20s/it]
 34%|███▍      | 5006/14725 [4:34:09<8:45:46,  3.25s/it]
 34%|███▍      | 5007/14725 [4:34:12<8:42:10,  3.22s/it]
 34%|███▍      | 5008/14725 [4:34:16<8:44:25,  3.24s
0: {'loss': 0.73, 'grad_norm': 0.46435482298149194, 'learning_rate': 1e-05, 'epoch': 0.34}
0: {'loss': 0.7154, 'grad_norm': 0.41846998805469016, 'learning_rate': 1e-05, 'epoch': 0.34}
0: /it]
 34%|███▍      | 5009/14725 [4:34:19<8:45:31,  3.25s/it]
 34%|███▍      | 5010/14725 [4:34:22<8:51:28,  3.28s/it]
                                                        

 34%|███▍      | 5010/14725 [4:34:22<8:51:28,  3.28s/it]
 34%|███▍      | 5011/14725 [4:34:26<8:49:21,  3.27s/it]
 34%|███▍      | 5012/14725 [4:34:29<8:49:43,  3.27s/it]
 34%|███▍      | 5013/14725 [4:34:32<8:52:09,  3.29s/it]
 34%|███▍      | 5014/14725 [4:34:35<8:49:23,  3.27s/it]
 34%|███▍      | 5015/14725 [4:34:39<8:49:12,  3.27s/it]
 34%|███▍      | 5016/14725 [4:34:42<8:52:33,  3.29s/it]
 34%|███▍      | 5017/14725 [4:34:46<9:05:40,  3.37s/it]
 34%|███▍      | 5018/14725 [4:34:49<9:08:00,  3.39s/it]
 34%|███▍      | 5019/14725 [4:34:52<8:58:47,  3.33s/it]
 34%|███▍      | 5020/14725 [4:34:56<9:00:36,  3.34s/it]
                                                        

 34%|███▍      | 5020/14725 [4:34:56<9:00:36,  3.3
0: {'loss': 0.7079, 'grad_norm': 0.4269770834463501, 'learning_rate': 1e-05, 'epoch': 0.34}
0: 4s/it]
 34%|███▍      | 5021/14725 [4:34:59<8:55:51,  3.31s/it]
 34%|███▍      | 5022/14725 [4:35:02<8:53:13,  3.30s/it]
 34%|███▍      | 5023/14725 [4:35:05<8:44:06,  3.24s/it]
 34%|███▍      | 5024/14725 [4:35:08<8:38:55,  3.21s/it]
 34%|███▍      | 5025/14725 [4:35:12<8:45:05,  3.25s/it]
 34%|███▍      | 5026/14725 [4:35:15<8:44:04,  3.24s/it]
 34%|███▍      | 5027/14725 [4:35:18<8:48:11,  3.27s/it]
 34%|███▍      | 5028/14725 [4:35:22<8:49:10,  3.27s/it]
 34%|███▍      | 5029/14725 [4:35:25<8:42:18,  3.23s/it]
 34%|███▍      | 5030/14725 [4:35:28<8:38:52,  3.21s/it]
                                                        

 34%|███▍      | 5030/14725 [4:35:28<8:38:52,  3.21s/it]
 34%|███▍      | 5031/14725 [4:35:31<8:49:26,  3.28s/it]
 34%|███▍      | 5032/14725 [4:35:35<8:59:01,  3.34s/it]
 34%|███▍      | 5033/14725 [4:35:38<9:12:49,  3.42s/it]
 34%|███▍      | 5034/14725 [4:35:42<9:03
0: {'loss': 0.7283, 'grad_norm': 0.4434548931665101, 'learning_rate': 1e-05, 'epoch': 0.34}
0: :04,  3.36s/it]
 34%|███▍      | 5035/14725 [4:35:45<9:00:43,  3.35s/it]
 34%|███▍      | 5036/14725 [4:35:48<8:53:06,  3.30s/it]
 34%|███▍      | 5037/14725 [4:35:51<8:50:30,  3.29s/it]
 34%|███▍      | 5038/14725 [4:35:55<8:51:32,  3.29s/it]
 34%|███▍      | 5039/14725 [4:35:58<8:44:55,  3.25s/it]
 34%|███▍      | 5040/14725 [4:36:01<8:51:24,  3.29s/it]
                                                        

 34%|███▍      | 5040/14725 [4:36:01<8:51:24,  3.29s/it]
 34%|███▍      | 5041/14725 [4:36:04<8:46:05,  3.26s/it]
 34%|███▍      | 5042/14725 [4:36:08<8:57:34,  3.33s/it]
 34%|███▍      | 5043/14725 [4:36:11<8:50:46,  3.29s/it]
 34%|███▍      | 5044/14725 [4:36:14<8:45:10,  3.25s/it]
 34%|███▍      | 5045/14725 [4:36:17<8:41:04,  3.23s/it]
 34%|███▍      | 5046/14725 [4:36:21<8:45:28,  3.26s/it]
 34%|███▍      | 5047/14725 [4:36:24<8:56:27,  3.33s/it]
 34%|███▍      | 5048/14725 [4:3
0: {'loss': 0.6951, 'grad_norm': 0.4540740490171333, 'learning_rate': 1e-05, 'epoch': 0.34}
0: {'loss': 0.7249, 'grad_norm': 0.45106011464485823, 'learning_rate': 1e-05, 'epoch': 0.34}
0: 6:27<8:52:05,  3.30s/it]
 34%|███▍      | 5049/14725 [4:36:31<8:52:56,  3.30s/it]
 34%|███▍      | 5050/14725 [4:36:34<8:50:34,  3.29s/it]
                                                        

 34%|███▍      | 5050/14725 [4:36:34<8:50:34,  3.29s/it]
 34%|███▍      | 5051/14725 [4:36:37<8:55:59,  3.32s/it]
 34%|███▍      | 5052/14725 [4:36:41<8:48:00,  3.28s/it]
 34%|███▍      | 5053/14725 [4:36:44<8:43:27,  3.25s/it]
 34%|███▍      | 5054/14725 [4:36:47<8:44:01,  3.25s/it]
 34%|███▍      | 5055/14725 [4:36:50<8:41:46,  3.24s/it]
 34%|███▍      | 5056/14725 [4:36:54<8:48:44,  3.28s/it]
 34%|███▍      | 5057/14725 [4:36:57<8:44:32,  3.26s/it]
 34%|███▍      | 5058/14725 [4:37:00<8:44:04,  3.25s/it]
 34%|███▍      | 5059/14725 [4:37:04<8:59:18,  3.35s/it]
 34%|███▍      | 5060/14725 [4:37:07<8:52:13,  3.30s/it]
                                                        

 34%|███▍      | 5060/14725 [4
0: {'loss': 0.7096, 'grad_norm': 0.41455382174611655, 'learning_rate': 1e-05, 'epoch': 0.34}
0: :37:07<8:52:13,  3.30s/it]
 34%|███▍      | 5061/14725 [4:37:10<8:52:46,  3.31s/it]
 34%|███▍      | 5062/14725 [4:37:13<8:49:33,  3.29s/it]
 34%|███▍      | 5063/14725 [4:37:17<8:43:44,  3.25s/it]
 34%|███▍      | 5064/14725 [4:37:20<8:43:04,  3.25s/it]
 34%|███▍      | 5065/14725 [4:37:23<8:54:15,  3.32s/it]
 34%|███▍      | 5066/14725 [4:37:27<8:55:07,  3.32s/it]
 34%|███▍      | 5067/14725 [4:37:30<8:52:39,  3.31s/it]
 34%|███▍      | 5068/14725 [4:37:33<8:50:02,  3.29s/it]
 34%|███▍      | 5069/14725 [4:37:37<8:59:45,  3.35s/it]
 34%|███▍      | 5070/14725 [4:37:40<8:51:38,  3.30s/it]
                                                        

 34%|███▍      | 5070/14725 [4:37:40<8:51:38,  3.30s/it]
 34%|███▍      | 5071/14725 [4:37:43<8:41:17,  3.24s/it]
 34%|███▍      | 5072/14725 [4:37:46<8:39:16,  3.23s/it]
 34%|███▍      | 5073/14725 [4:37:50<8:49:03,  3.29s/it]
 34%|███▍      | 5074
0: {'loss': 0.7178, 'grad_norm': 0.4654410751796842, 'learning_rate': 1e-05, 'epoch': 0.34}
0: /14725 [4:37:53<8:47:19,  3.28s/it]
 34%|███▍      | 5075/14725 [4:37:56<8:49:28,  3.29s/it]
 34%|███▍      | 5076/14725 [4:37:59<8:50:43,  3.30s/it]
 34%|███▍      | 5077/14725 [4:38:03<8:44:00,  3.26s/it]
 34%|███▍      | 5078/14725 [4:38:06<8:39:43,  3.23s/it]
 34%|███▍      | 5079/14725 [4:38:09<8:45:11,  3.27s/it]
 34%|███▍      | 5080/14725 [4:38:12<8:40:56,  3.24s/it]
                                                        

 34%|███▍      | 5080/14725 [4:38:12<8:40:56,  3.24s/it]
 35%|███▍      | 5081/14725 [4:38:15<8:32:45,  3.19s/it]
 35%|███▍      | 5082/14725 [4:38:19<8:37:45,  3.22s/it]
 35%|███▍      | 5083/14725 [4:38:22<8:40:20,  3.24s/it]
 35%|███▍      | 5084/14725 [4:38:25<8:40:00,  3.24s/it]
 35%|███▍      | 5085/14725 [4:38:28<8:36:29,  3.21s/it]
 35%|███▍      | 5086/14725 [4:38:32<8:40:07,  3.24s/it]
 35%|███▍      | 5087/14725 [4:38:35<8:30:28,  3.18s/it]
 35%|███▍   
0: {'loss': 0.7129, 'grad_norm': 0.46109252294059433, 'learning_rate': 1e-05, 'epoch': 0.35}
0: {'loss': 0.7171, 'grad_norm': 0.4061421252259631, 'learning_rate': 1e-05, 'epoch': 0.35}
0:    | 5088/14725 [4:38:38<8:29:27,  3.17s/it]
 35%|███▍      | 5089/14725 [4:38:41<8:33:10,  3.20s/it]
 35%|███▍      | 5090/14725 [4:38:44<8:34:36,  3.20s/it]
                                                        

 35%|███▍      | 5090/14725 [4:38:44<8:34:36,  3.20s/it]
 35%|███▍      | 5091/14725 [4:38:48<8:33:24,  3.20s/it]
 35%|███▍      | 5092/14725 [4:38:51<8:37:33,  3.22s/it]
 35%|███▍      | 5093/14725 [4:38:54<8:39:31,  3.24s/it]
 35%|███▍      | 5094/14725 [4:38:57<8:36:42,  3.22s/it]
 35%|███▍      | 5095/14725 [4:39:00<8:33:19,  3.20s/it]
 35%|███▍      | 5096/14725 [4:39:04<8:35:47,  3.21s/it]
 35%|███▍      | 5097/14725 [4:39:07<8:43:42,  3.26s/it]
 35%|███▍      | 5098/14725 [4:39:10<8:43:13,  3.26s/it]
 35%|███▍      | 5099/14725 [4:39:13<8:38:47,  3.23s/it]
 35%|███▍      | 5100/14725 [4:39:17<8:39:27,  3.24s/it]
                                                        

 35%|███▍ 
0: {'loss': 0.7254, 'grad_norm': 0.46962576183952776, 'learning_rate': 1e-05, 'epoch': 0.35}
0:      | 5100/14725 [4:39:17<8:39:27,  3.24s/it]
 35%|███▍      | 5101/14725 [4:39:20<8:39:57,  3.24s/it]
 35%|███▍      | 5102/14725 [4:39:23<8:37:46,  3.23s/it]
 35%|███▍      | 5103/14725 [4:39:26<8:44:19,  3.27s/it]
 35%|███▍      | 5104/14725 [4:39:30<8:32:55,  3.20s/it]
 35%|███▍      | 5105/14725 [4:39:33<8:40:33,  3.25s/it]
 35%|███▍      | 5106/14725 [4:39:36<8:46:33,  3.28s/it]
 35%|███▍      | 5107/14725 [4:39:40<8:50:02,  3.31s/it]
 35%|███▍      | 5108/14725 [4:39:43<8:40:22,  3.25s/it]
 35%|███▍      | 5109/14725 [4:39:46<8:34:53,  3.21s/it]
 35%|███▍      | 5110/14725 [4:39:49<8:38:28,  3.24s/it]
                                                        

 35%|███▍      | 5110/14725 [4:39:49<8:38:28,  3.24s/it]
 35%|███▍      | 5111/14725 [4:39:52<8:40:04,  3.25s/it]
 35%|███▍      | 5112/14725 [4:39:56<8:41:15,  3.25s/it]
 35%|███▍      | 5113/14725 [4:39:59<8:39:50,  3.24s/it]
 35%|█�
0: {'loss': 0.7046, 'grad_norm': 0.45682061516500466, 'learning_rate': 1e-05, 'epoch': 0.35}
0: ��█▍      | 5114/14725 [4:40:02<8:41:12,  3.25s/it]
 35%|███▍      | 5115/14725 [4:40:05<8:42:20,  3.26s/it]
 35%|███▍      | 5116/14725 [4:40:09<8:37:58,  3.23s/it]
 35%|███▍      | 5117/14725 [4:40:12<8:39:47,  3.25s/it]
 35%|███▍      | 5118/14725 [4:40:15<8:42:41,  3.26s/it]
 35%|███▍      | 5119/14725 [4:40:18<8:39:31,  3.25s/it]
 35%|███▍      | 5120/14725 [4:40:22<8:39:14,  3.24s/it]
                                                        

 35%|███▍      | 5120/14725 [4:40:22<8:39:14,  3.24s/it]
 35%|███▍      | 5121/14725 [4:40:25<8:45:22,  3.28s/it]
 35%|███▍      | 5122/14725 [4:40:28<8:42:15,  3.26s/it]
 35%|███▍      | 5123/14725 [4:40:31<8:33:46,  3.21s/it]
 35%|███▍      | 5124/14725 [4:40:34<8:28:31,  3.18s/it]
 35%|███▍      | 5125/14725 [4:40:38<8:38:50,  3.24s/it]
 35%|███▍      | 5126/14725 [4:40:41<8:40:16,  3.25s/it]
 35%|███▍      | 5127/14725 [4:40:45<8:50:10,  3.31s/it]
0: {'loss': 0.7036, 'grad_norm': 0.44073392027996516, 'learning_rate': 1e-05, 'epoch': 0.35}
0:  35%|███▍      | 5128/14725 [4:40:48<8:41:45,  3.26s/it]
 35%|███▍      | 5129/14725 [4:40:51<8:48:23,  3.30s/it]
 35%|███▍      | 5130/14725 [4:40:55<8:54:32,  3.34s/it]
                                                        

 35%|███▍      | 5130/14725 [4:40:55<8:54:32,  3.34s/it]
 35%|███▍      | 5131/14725 [4:40:58<8:45:05,  3.28s/it]
 35%|███▍      | 5132/14725 [4:41:01<8:48:49,  3.31s/it]
 35%|███▍      | 5133/14725 [4:41:04<8:40:28,  3.26s/it]
 35%|███▍      | 5134/14725 [4:41:07<8:40:50,  3.26s/it]
 35%|███▍      | 5135/14725 [4:41:11<8:43:56,  3.28s/it]
 35%|███▍      | 5136/14725 [4:41:14<8:46:52,  3.30s/it]
 35%|███▍      | 5137/14725 [4:41:17<8:46:14,  3.29s/it]
 35%|███▍      | 5138/14725 [4:41:21<8:46:20,  3.29s/it]
 35%|███▍      | 5139/14725 [4:41:24<8:51:41,  3.33s/it]
 35%|███▍      | 5140/14725 [4:41:27<8:46:34,  3.30s/it]
                                                        
0: {'loss': 0.7142, 'grad_norm': 0.4264856490052652, 'learning_rate': 1e-05, 'epoch': 0.35}
0: {'loss': 0.7013, 'grad_norm': 0.43675797321621046, 'learning_rate': 1e-05, 'epoch': 0.35}
0: 

 35%|███▍      | 5140/14725 [4:41:27<8:46:34,  3.30s/it]
 35%|███▍      | 5141/14725 [4:41:30<8:39:45,  3.25s/it]
 35%|███▍      | 5142/14725 [4:41:34<8:39:52,  3.25s/it]
 35%|███▍      | 5143/14725 [4:41:37<8:31:23,  3.20s/it]
 35%|███▍      | 5144/14725 [4:41:40<8:33:21,  3.21s/it]
 35%|███▍      | 5145/14725 [4:41:43<8:35:16,  3.23s/it]
 35%|███▍      | 5146/14725 [4:41:47<8:33:20,  3.22s/it]
 35%|███▍      | 5147/14725 [4:41:50<8:38:33,  3.25s/it]
 35%|███▍      | 5148/14725 [4:41:53<8:36:42,  3.24s/it]
 35%|███▍      | 5149/14725 [4:41:56<8:39:15,  3.25s/it]
 35%|███▍      | 5150/14725 [4:42:00<8:42:45,  3.28s/it]
                                                        

 35%|███▍      | 5150/14725 [4:42:00<8:42:45,  3.28s/it]
 35%|███▍      | 5151/14725 [4:42:03<8:42:37,  3.28s/it]
 35%|███▍      | 5152/14725 [4:42:06<8:38:08,  3.25s/it]
 35%|███▍      | 5153/14725 [4:42:09<8:39:32, 
0: {'loss': 0.7047, 'grad_norm': 0.4390608288342472, 'learning_rate': 1e-05, 'epoch': 0.35}
0:  3.26s/it]
 35%|███▌      | 5154/14725 [4:42:13<8:42:29,  3.28s/it]
 35%|███▌      | 5155/14725 [4:42:16<8:34:06,  3.22s/it]
 35%|███▌      | 5156/14725 [4:42:19<8:38:42,  3.25s/it]
 35%|███▌      | 5157/14725 [4:42:22<8:37:21,  3.24s/it]
 35%|███▌      | 5158/14725 [4:42:25<8:29:35,  3.20s/it]
 35%|███▌      | 5159/14725 [4:42:29<8:31:11,  3.21s/it]
 35%|███▌      | 5160/14725 [4:42:32<8:29:23,  3.20s/it]
                                                        

 35%|███▌      | 5160/14725 [4:42:32<8:29:23,  3.20s/it]
 35%|███▌      | 5161/14725 [4:42:35<8:34:29,  3.23s/it]
 35%|███▌      | 5162/14725 [4:42:39<8:50:58,  3.33s/it]
 35%|███▌      | 5163/14725 [4:42:42<8:42:02,  3.28s/it]
 35%|███▌      | 5164/14725 [4:42:45<8:39:09,  3.26s/it]
 35%|███▌      | 5165/14725 [4:42:48<8:42:32,  3.28s/it]
 35%|███▌      | 5166/14725 [4:42:52<8:35:34,  3.24s/it]
 35%|███▌      | 5167/14725 [4:42:55<
0: {'loss': 0.704, 'grad_norm': 0.43144058169048194, 'learning_rate': 1e-05, 'epoch': 0.35}
0: 8:37:23,  3.25s/it]
 35%|███▌      | 5168/14725 [4:42:58<8:44:06,  3.29s/it]
 35%|███▌      | 5169/14725 [4:43:01<8:39:32,  3.26s/it]
 35%|███▌      | 5170/14725 [4:43:05<8:34:48,  3.23s/it]
                                                        

 35%|███▌      | 5170/14725 [4:43:05<8:34:48,  3.23s/it]
 35%|███▌      | 5171/14725 [4:43:08<8:42:20,  3.28s/it]
 35%|███▌      | 5172/14725 [4:43:11<8:44:06,  3.29s/it]
 35%|███▌      | 5173/14725 [4:43:14<8:35:26,  3.24s/it]
 35%|███▌      | 5174/14725 [4:43:18<8:34:01,  3.23s/it]
 35%|███▌      | 5175/14725 [4:43:21<8:44:08,  3.29s/it]
 35%|███▌      | 5176/14725 [4:43:24<8:49:58,  3.33s/it]
 35%|███▌      | 5177/14725 [4:43:28<8:39:13,  3.26s/it]
 35%|███▌      | 5178/14725 [4:43:31<8:44:13,  3.29s/it]
 35%|███▌      | 5179/14725 [4:43:34<8:42:02,  3.28s/it]
 35%|███▌      | 5180/14725 [4:43:37<8:37:06,  3.25s/it]
                                    
0: {'loss': 0.712, 'grad_norm': 0.45081729616357075, 'learning_rate': 1e-05, 'epoch': 0.35}
0: {'loss': 0.714, 'grad_norm': 0.42581819098866186, 'learning_rate': 1e-05, 'epoch': 0.35}
0:                     

 35%|███▌      | 5180/14725 [4:43:37<8:37:06,  3.25s/it]
 35%|███▌      | 5181/14725 [4:43:41<8:33:13,  3.23s/it]
 35%|███▌      | 5182/14725 [4:43:44<8:28:56,  3.20s/it]
 35%|███▌      | 5183/14725 [4:43:47<8:31:43,  3.22s/it]
 35%|███▌      | 5184/14725 [4:43:50<8:32:57,  3.23s/it]
 35%|███▌      | 5185/14725 [4:43:53<8:36:38,  3.25s/it]
 35%|███▌      | 5186/14725 [4:43:57<8:50:23,  3.34s/it]
 35%|███▌      | 5187/14725 [4:44:00<8:46:25,  3.31s/it]
 35%|███▌      | 5188/14725 [4:44:03<8:40:59,  3.28s/it]
 35%|███▌      | 5189/14725 [4:44:07<8:39:08,  3.27s/it]
 35%|███▌      | 5190/14725 [4:44:10<8:43:00,  3.29s/it]
                                                        

 35%|███▌      | 5190/14725 [4:44:10<8:43:00,  3.29s/it]
 35%|███▌      | 5191/14725 [4:44:13<8:32:00,  3.22s/it]
 35%|███▌      | 5192/14725 [4:44:16<8:28:23,  3.20s/it]
 35%|███▌      | 5193/1472
0: {'loss': 0.7062, 'grad_norm': 0.4725362003532514, 'learning_rate': 1e-05, 'epoch': 0.35}
0: 5 [4:44:19<8:25:47,  3.18s/it]
 35%|███▌      | 5194/14725 [4:44:23<8:23:17,  3.17s/it]
 35%|███▌      | 5195/14725 [4:44:26<8:32:21,  3.23s/it]
 35%|███▌      | 5196/14725 [4:44:29<8:29:48,  3.21s/it]
 35%|███▌      | 5197/14725 [4:44:32<8:25:53,  3.19s/it]
 35%|███▌      | 5198/14725 [4:44:35<8:25:35,  3.18s/it]
 35%|███▌      | 5199/14725 [4:44:39<8:23:32,  3.17s/it]
 35%|███▌      | 5200/14725 [4:44:42<8:29:47,  3.21s/it]
                                                        

 35%|███▌      | 5200/14725 [4:44:42<8:29:47,  3.21s/it]
 35%|███▌      | 5201/14725 [4:44:45<8:31:43,  3.22s/it]
 35%|███▌      | 5202/14725 [4:44:48<8:32:26,  3.23s/it]
 35%|███▌      | 5203/14725 [4:44:52<8:34:46,  3.24s/it]
 35%|███▌      | 5204/14725 [4:44:55<8:42:40,  3.29s/it]
 35%|███▌      | 5205/14725 [4:44:58<8:40:59,  3.28s/it]
 35%|███▌      | 5206/14725 [4:45:02<8:39:18,  3.27s/it]
 35%|███▌      | 
0: {'loss': 0.723, 'grad_norm': 0.47309292434453654, 'learning_rate': 1e-05, 'epoch': 0.35}
0: 5207/14725 [4:45:05<8:33:02,  3.23s/it]
 35%|███▌      | 5208/14725 [4:45:08<8:33:34,  3.24s/it]
 35%|███▌      | 5209/14725 [4:45:11<8:39:32,  3.28s/it]
 35%|███▌      | 5210/14725 [4:45:15<8:39:26,  3.28s/it]
                                                        

 35%|███▌      | 5210/14725 [4:45:15<8:39:26,  3.28s/it]
 35%|███▌      | 5211/14725 [4:45:18<8:43:49,  3.30s/it]
 35%|███▌      | 5212/14725 [4:45:21<8:38:18,  3.27s/it]
 35%|███▌      | 5213/14725 [4:45:24<8:33:32,  3.24s/it]
 35%|███▌      | 5214/14725 [4:45:28<8:41:50,  3.29s/it]
 35%|███▌      | 5215/14725 [4:45:31<8:42:01,  3.29s/it]
 35%|███▌      | 5216/14725 [4:45:34<8:33:39,  3.24s/it]
 35%|███▌      | 5217/14725 [4:45:37<8:30:10,  3.22s/it]
 35%|███▌      | 5218/14725 [4:45:40<8:25:00,  3.19s/it]
 35%|███▌      | 5219/14725 [4:45:44<8:37:32,  3.27s/it]
 35%|███▌      | 5220/14725 [4:45:47<8:35:56,  3.26s/it]
                
0: {'loss': 0.6975, 'grad_norm': 0.43886779515558766, 'learning_rate': 1e-05, 'epoch': 0.35}
0: {'loss': 0.6994, 'grad_norm': 0.4384940863325193, 'learning_rate': 1e-05, 'epoch': 0.36}
0:                                         

 35%|███▌      | 5220/14725 [4:45:47<8:35:56,  3.26s/it]
 35%|███▌      | 5221/14725 [4:45:50<8:36:03,  3.26s/it]
 35%|███▌      | 5222/14725 [4:45:54<8:35:18,  3.25s/it]
 35%|███▌      | 5223/14725 [4:45:57<8:34:45,  3.25s/it]
 35%|███▌      | 5224/14725 [4:46:00<8:40:48,  3.29s/it]
 35%|███▌      | 5225/14725 [4:46:03<8:32:15,  3.24s/it]
 35%|███▌      | 5226/14725 [4:46:07<8:40:02,  3.28s/it]
 35%|███▌      | 5227/14725 [4:46:10<8:38:10,  3.27s/it]
 36%|███▌      | 5228/14725 [4:46:13<8:37:58,  3.27s/it]
 36%|███▌      | 5229/14725 [4:46:16<8:32:19,  3.24s/it]
 36%|███▌      | 5230/14725 [4:46:20<8:31:08,  3.23s/it]
                                                        

 36%|███▌      | 5230/14725 [4:46:20<8:31:08,  3.23s/it]
 36%|███▌      | 5231/14725 [4:46:23<8:31:39,  3.23s/it]
 36%|███▌      | 5232/14725 [4:46:26<8:38:48,  3.28s/it]
 36%|███
0: {'loss': 0.7192, 'grad_norm': 0.4568684502004619, 'learning_rate': 1e-05, 'epoch': 0.36}
0: ▌      | 5233/14725 [4:46:30<8:39:45,  3.29s/it]
 36%|███▌      | 5234/14725 [4:46:33<8:32:53,  3.24s/it]
 36%|███▌      | 5235/14725 [4:46:36<8:34:03,  3.25s/it]
 36%|███▌      | 5236/14725 [4:46:39<8:34:40,  3.25s/it]
 36%|███▌      | 5237/14725 [4:46:43<8:36:24,  3.27s/it]
 36%|███▌      | 5238/14725 [4:46:46<8:38:58,  3.28s/it]
 36%|███▌      | 5239/14725 [4:46:49<8:33:30,  3.25s/it]
 36%|███▌      | 5240/14725 [4:46:52<8:44:20,  3.32s/it]
                                                        

 36%|███▌      | 5240/14725 [4:46:52<8:44:20,  3.32s/it]
 36%|███▌      | 5241/14725 [4:46:56<8:40:37,  3.29s/it]
 36%|███▌      | 5242/14725 [4:46:59<8:36:48,  3.27s/it]
 36%|███▌      | 5243/14725 [4:47:02<8:26:32,  3.21s/it]
 36%|███▌      | 5244/14725 [4:47:05<8:24:56,  3.20s/it]
 36%|███▌      | 5245/14725 [4:47:08<8:28:50,  3.22s/it]
 36%|███▌      | 5246/14725 [4:47:12<8:36:08,  3.27s/it]
 36%|
0: {'loss': 0.6932, 'grad_norm': 0.4544920796246397, 'learning_rate': 1e-05, 'epoch': 0.36}
0: ███▌      | 5247/14725 [4:47:15<8:27:45,  3.21s/it]
 36%|███▌      | 5248/14725 [4:47:18<8:33:14,  3.25s/it]
 36%|███▌      | 5249/14725 [4:47:21<8:29:13,  3.22s/it]
 36%|███▌      | 5250/14725 [4:47:25<8:33:03,  3.25s/it]
                                                        

 36%|███▌      | 5250/14725 [4:47:25<8:33:03,  3.25s/it]
 36%|███▌      | 5251/14725 [4:47:28<8:27:54,  3.22s/it]
 36%|███▌      | 5252/14725 [4:47:31<8:39:28,  3.29s/it]
 36%|███▌      | 5253/14725 [4:47:35<8:41:45,  3.31s/it]
 36%|███▌      | 5254/14725 [4:47:38<8:37:15,  3.28s/it]
 36%|███▌      | 5255/14725 [4:47:41<8:37:49,  3.28s/it]
 36%|███▌      | 5256/14725 [4:47:44<8:40:11,  3.30s/it]
 36%|███▌      | 5257/14725 [4:47:48<8:44:37,  3.32s/it]
 36%|███▌      | 5258/14725 [4:47:51<8:34:42,  3.26s/it]
 36%|███▌      | 5259/14725 [4:47:54<8:33:06,  3.25s/it]
 36%|███▌      | 5260/14725 [4:47:58<8:36:33,  3.27s/
0: {'loss': 0.6922, 'grad_norm': 0.44802402867694024, 'learning_rate': 1e-05, 'epoch': 0.36}
0: {'loss': 0.7056, 'grad_norm': 0.43890536548376385, 'learning_rate': 1e-05, 'epoch': 0.36}
0: it]
                                                        

 36%|███▌      | 5260/14725 [4:47:58<8:36:33,  3.27s/it]
 36%|███▌      | 5261/14725 [4:48:01<8:31:11,  3.24s/it]
 36%|███▌      | 5262/14725 [4:48:04<8:41:33,  3.31s/it]
 36%|███▌      | 5263/14725 [4:48:08<8:43:11,  3.32s/it]
 36%|███▌      | 5264/14725 [4:48:11<8:44:10,  3.32s/it]
 36%|███▌      | 5265/14725 [4:48:14<8:34:54,  3.27s/it]
 36%|███▌      | 5266/14725 [4:48:17<8:26:40,  3.21s/it]
 36%|███▌      | 5267/14725 [4:48:20<8:21:01,  3.18s/it]
 36%|███▌      | 5268/14725 [4:48:23<8:21:26,  3.18s/it]
 36%|███▌      | 5269/14725 [4:48:27<8:36:29,  3.28s/it]
 36%|███▌      | 5270/14725 [4:48:30<8:26:14,  3.21s/it]
                                                        

 36%|███▌      | 5270/14725 [4:48:30<8:26:14,  3.21s/it]
 36%|███▌      | 5271/14725 [4:48:33<8:30:47,  3.24s/it]
 36%|███▌      | 5272/14725 [4:48:37<8:36:04,  3.28
0: {'loss': 0.7183, 'grad_norm': 0.4500332376257809, 'learning_rate': 1e-05, 'epoch': 0.36}
0: s/it]
 36%|███▌      | 5273/14725 [4:48:40<8:27:21,  3.22s/it]
 36%|███▌      | 5274/14725 [4:48:43<8:27:20,  3.22s/it]
 36%|███▌      | 5275/14725 [4:48:46<8:39:24,  3.30s/it]
 36%|███▌      | 5276/14725 [4:48:50<8:34:22,  3.27s/it]
 36%|███▌      | 5277/14725 [4:48:53<8:30:30,  3.24s/it]
 36%|███▌      | 5278/14725 [4:48:56<8:32:39,  3.26s/it]
 36%|███▌      | 5279/14725 [4:48:59<8:33:35,  3.26s/it]
 36%|███▌      | 5280/14725 [4:49:03<8:32:46,  3.26s/it]
                                                        

 36%|███▌      | 5280/14725 [4:49:03<8:32:46,  3.26s/it]
 36%|███▌      | 5281/14725 [4:49:06<8:39:00,  3.30s/it]
 36%|███▌      | 5282/14725 [4:49:09<8:38:31,  3.29s/it]
 36%|███▌      | 5283/14725 [4:49:12<8:35:02,  3.27s/it]
 36%|███▌      | 5284/14725 [4:49:16<8:37:24,  3.29s/it]
 36%|███▌      | 5285/14725 [4:49:19<8:29:03,  3.24s/it]
 36%|███▌      | 5286/14725 [4:49:22<8:30:
0: {'loss': 0.7024, 'grad_norm': 0.4297426341159321, 'learning_rate': 1e-05, 'epoch': 0.36}
0: 06,  3.24s/it]
 36%|███▌      | 5287/14725 [4:49:26<8:36:37,  3.28s/it]
 36%|███▌      | 5288/14725 [4:49:29<8:32:17,  3.26s/it]
 36%|███▌      | 5289/14725 [4:49:32<8:35:13,  3.28s/it]
 36%|███▌      | 5290/14725 [4:49:35<8:38:09,  3.30s/it]
                                                        

 36%|███▌      | 5290/14725 [4:49:35<8:38:09,  3.30s/it]
 36%|███▌      | 5291/14725 [4:49:39<8:29:41,  3.24s/it]
 36%|███▌      | 5292/14725 [4:49:42<8:27:04,  3.23s/it]
 36%|███▌      | 5293/14725 [4:49:45<8:35:06,  3.28s/it]
 36%|███▌      | 5294/14725 [4:49:48<8:34:05,  3.27s/it]
 36%|███▌      | 5295/14725 [4:49:51<8:28:11,  3.23s/it]
 36%|███▌      | 5296/14725 [4:49:55<8:26:14,  3.22s/it]
 36%|███▌      | 5297/14725 [4:49:58<8:21:16,  3.19s/it]
 36%|███▌      | 5298/14725 [4:50:01<8:34:06,  3.27s/it]
 36%|███▌      | 5299/14725 [4:50:05<8:42:24,  3.33s/it]
 36%|███▌      | 5300/14725 [4:50
0: {'loss': 0.7143, 'grad_norm': 0.4236173535869148, 'learning_rate': 1e-05, 'epoch': 0.36}
0: {'loss': 0.7258, 'grad_norm': 0.44566739524867405, 'learning_rate': 1e-05, 'epoch': 0.36}
0: :08<8:41:47,  3.32s/it]
                                                        

 36%|███▌      | 5300/14725 [4:50:08<8:41:47,  3.32s/it]
 36%|███▌      | 5301/14725 [4:50:11<8:32:44,  3.26s/it]
 36%|███▌      | 5302/14725 [4:50:14<8:35:12,  3.28s/it]
 36%|███▌      | 5303/14725 [4:50:18<8:35:58,  3.29s/it]
 36%|███▌      | 5304/14725 [4:50:21<8:42:30,  3.33s/it]
 36%|███▌      | 5305/14725 [4:50:24<8:36:59,  3.29s/it]
 36%|███▌      | 5306/14725 [4:50:28<8:27:45,  3.23s/it]
 36%|███▌      | 5307/14725 [4:50:31<8:29:03,  3.24s/it]
 36%|███▌      | 5308/14725 [4:50:34<8:30:55,  3.26s/it]
 36%|███▌      | 5309/14725 [4:50:37<8:31:41,  3.26s/it]
 36%|███▌      | 5310/14725 [4:50:41<8:32:33,  3.27s/it]
                                                        

 36%|███▌      | 5310/14725 [4:50:41<8:32:33,  3.27s/it]
 36%|███▌      | 5311/14725 [4:50:44<8:32:43,  3.27s/it]
 36%|███▌      | 5312/14725 [4:
0: {'loss': 0.7, 'grad_norm': 0.40894744775168, 'learning_rate': 1e-05, 'epoch': 0.36}
0: 50:47<8:26:48,  3.23s/it]
 36%|███▌      | 5313/14725 [4:50:51<8:39:14,  3.31s/it]
 36%|███▌      | 5314/14725 [4:50:54<8:36:40,  3.29s/it]
 36%|███▌      | 5315/14725 [4:50:57<8:28:49,  3.24s/it]
 36%|███▌      | 5316/14725 [4:51:00<8:24:08,  3.21s/it]
 36%|███▌      | 5317/14725 [4:51:03<8:22:22,  3.20s/it]
 36%|███▌      | 5318/14725 [4:51:06<8:25:21,  3.22s/it]
 36%|███▌      | 5319/14725 [4:51:10<8:29:35,  3.25s/it]
 36%|███▌      | 5320/14725 [4:51:13<8:31:32,  3.26s/it]
                                                        

 36%|███▌      | 5320/14725 [4:51:13<8:31:32,  3.26s/it]
 36%|███▌      | 5321/14725 [4:51:16<8:35:20,  3.29s/it]
 36%|███▌      | 5322/14725 [4:51:20<8:38:38,  3.31s/it]
 36%|███▌      | 5323/14725 [4:51:23<8:33:28,  3.28s/it]
 36%|███▌      | 5324/14725 [4:51:26<8:26:59,  3.24s/it]
 36%|███▌      | 5325/14725 [4:51:29<8:22:36,  3.21s/it]
 36%|███▌      | 5326/
0: {'loss': 0.6969, 'grad_norm': 0.4525882930624492, 'learning_rate': 1e-05, 'epoch': 0.36}
0: 14725 [4:51:32<8:20:15,  3.19s/it]
 36%|███▌      | 5327/14725 [4:51:36<8:19:12,  3.19s/it]
 36%|███▌      | 5328/14725 [4:51:39<8:23:34,  3.22s/it]
 36%|███▌      | 5329/14725 [4:51:42<8:22:47,  3.21s/it]
 36%|███▌      | 5330/14725 [4:51:46<8:31:42,  3.27s/it]
                                                        

 36%|███▌      | 5330/14725 [4:51:46<8:31:42,  3.27s/it]
 36%|███▌      | 5331/14725 [4:51:49<8:33:17,  3.28s/it]
 36%|███▌      | 5332/14725 [4:51:52<8:41:56,  3.33s/it]
 36%|███▌      | 5333/14725 [4:51:56<8:40:58,  3.33s/it]
 36%|███▌      | 5334/14725 [4:51:59<8:47:14,  3.37s/it]
 36%|███▌      | 5335/14725 [4:52:02<8:37:03,  3.30s/it]
 36%|███▌      | 5336/14725 [4:52:05<8:29:18,  3.25s/it]
 36%|███▌      | 5337/14725 [4:52:09<8:31:40,  3.27s/it]
 36%|███▋      | 5338/14725 [4:52:12<8:31:05,  3.27s/it]
 36%|███▋      | 5339/14725 [4:52:15<8:29:33,  3.26s/it]
 36%|███▋    
0: {'loss': 0.7227, 'grad_norm': 0.4421965122319444, 'learning_rate': 1e-05, 'epoch': 0.36}
0: {'loss': 0.7044, 'grad_norm': 0.4257774633126275, 'learning_rate': 1e-05, 'epoch': 0.36}
0:   | 5340/14725 [4:52:18<8:29:54,  3.26s/it]
                                                        

 36%|███▋      | 5340/14725 [4:52:18<8:29:54,  3.26s/it]
 36%|███▋      | 5341/14725 [4:52:22<8:25:23,  3.23s/it]
 36%|███▋      | 5342/14725 [4:52:25<8:21:23,  3.21s/it]
 36%|███▋      | 5343/14725 [4:52:28<8:33:35,  3.28s/it]
 36%|███▋      | 5344/14725 [4:52:31<8:25:08,  3.23s/it]
 36%|███▋      | 5345/14725 [4:52:35<8:27:02,  3.24s/it]
 36%|███▋      | 5346/14725 [4:52:38<8:33:50,  3.29s/it]
 36%|███▋      | 5347/14725 [4:52:41<8:29:05,  3.26s/it]
 36%|███▋      | 5348/14725 [4:52:44<8:22:20,  3.21s/it]
 36%|███▋      | 5349/14725 [4:52:47<8:21:52,  3.21s/it]
 36%|███▋      | 5350/14725 [4:52:51<8:21:35,  3.21s/it]
                                                        

 36%|███▋      | 5350/14725 [4:52:51<8:21:35,  3.21s/it]
 36%|███▋      | 5351/14725 [4:52:54<8:40:29,  3.33s/it]
 36%|███▋  
0: {'loss': 0.7301, 'grad_norm': 0.4750902503369908, 'learning_rate': 1e-05, 'epoch': 0.36}
0:     | 5352/14725 [4:52:58<8:41:13,  3.34s/it]
 36%|███▋      | 5353/14725 [4:53:01<8:43:18,  3.35s/it]
 36%|███▋      | 5354/14725 [4:53:04<8:35:31,  3.30s/it]
 36%|███▋      | 5355/14725 [4:53:07<8:31:36,  3.28s/it]
 36%|███▋      | 5356/14725 [4:53:11<8:38:33,  3.32s/it]
 36%|███▋      | 5357/14725 [4:53:14<8:29:39,  3.26s/it]
 36%|███▋      | 5358/14725 [4:53:17<8:28:06,  3.25s/it]
 36%|███▋      | 5359/14725 [4:53:20<8:28:29,  3.26s/it]
 36%|███▋      | 5360/14725 [4:53:24<8:33:41,  3.29s/it]
                                                        

 36%|███▋      | 5360/14725 [4:53:24<8:33:41,  3.29s/it]
 36%|███▋      | 5361/14725 [4:53:27<8:29:46,  3.27s/it]
 36%|███▋      | 5362/14725 [4:53:31<8:42:42,  3.35s/it]
 36%|███▋      | 5363/14725 [4:53:34<8:34:35,  3.30s/it]
 36%|███▋      | 5364/14725 [4:53:37<8:44:17,  3.36s/it]
 36%|███▋      | 5365/14725 [4:53:41<8:41:31,  3.34s/it]
 36%|█�
0: {'loss': 0.7002, 'grad_norm': 0.4318640572176625, 'learning_rate': 1e-05, 'epoch': 0.36}
0: �█▋      | 5366/14725 [4:53:44<8:40:26,  3.34s/it]
 36%|███▋      | 5367/14725 [4:53:47<8:46:29,  3.38s/it]
 36%|███▋      | 5368/14725 [4:53:51<8:45:26,  3.37s/it]
 36%|███▋      | 5369/14725 [4:53:54<8:33:11,  3.29s/it]
 36%|███▋      | 5370/14725 [4:53:57<8:39:07,  3.33s/it]
                                                        

 36%|███▋      | 5370/14725 [4:53:57<8:39:07,  3.33s/it]
 36%|███▋      | 5371/14725 [4:54:01<8:39:56,  3.34s/it]
 36%|███▋      | 5372/14725 [4:54:04<8:33:34,  3.29s/it]
 36%|███▋      | 5373/14725 [4:54:07<8:28:19,  3.26s/it]
 36%|███▋      | 5374/14725 [4:54:10<8:23:40,  3.23s/it]
 37%|███▋      | 5375/14725 [4:54:14<8:33:12,  3.29s/it]
 37%|███▋      | 5376/14725 [4:54:17<8:29:10,  3.27s/it]
 37%|███▋      | 5377/14725 [4:54:20<8:31:37,  3.28s/it]
 37%|███▋      | 5378/14725 [4:54:24<8:39:40,  3.34s/it]
 37%|███▋      | 5379/14725 [4:54:27<8:36:33,  3.32s/it]
 
0: {'loss': 0.7071, 'grad_norm': 0.40664006017674215, 'learning_rate': 1e-05, 'epoch': 0.37}
0: {'loss': 0.7178, 'grad_norm': 0.417363952076524, 'learning_rate': 1e-05, 'epoch': 0.37}
0: 37%|███▋      | 5380/14725 [4:54:30<8:32:52,  3.29s/it]
                                                        

 37%|███▋      | 5380/14725 [4:54:30<8:32:52,  3.29s/it]
 37%|███▋      | 5381/14725 [4:54:33<8:32:03,  3.29s/it]
 37%|███▋      | 5382/14725 [4:54:37<8:32:38,  3.29s/it]
 37%|███▋      | 5383/14725 [4:54:40<8:29:00,  3.27s/it]
 37%|███▋      | 5384/14725 [4:54:43<8:24:29,  3.24s/it]
 37%|███▋      | 5385/14725 [4:54:46<8:18:49,  3.20s/it]
 37%|███▋      | 5386/14725 [4:54:50<8:27:15,  3.26s/it]
 37%|███▋      | 5387/14725 [4:54:53<8:31:52,  3.29s/it]
 37%|███▋      | 5388/14725 [4:54:56<8:25:22,  3.25s/it]
 37%|███▋      | 5389/14725 [4:54:59<8:28:41,  3.27s/it]
 37%|███▋      | 5390/14725 [4:55:03<8:28:58,  3.27s/it]
                                                        

 37%|███▋      | 5390/14725 [4:55:03<8:28:58,  3.27s/it]
 37%|███▋      | 5391/14725 [4:55:06<8:20:48,  3.22s/it]
0: {'loss': 0.7014, 'grad_norm': 0.4280329434823763, 'learning_rate': 1e-05, 'epoch': 0.37}
0: 
 37%|███▋      | 5392/14725 [4:55:09<8:18:07,  3.20s/it]
 37%|███▋      | 5393/14725 [4:55:12<8:22:06,  3.23s/it]
 37%|███▋      | 5394/14725 [4:55:15<8:15:53,  3.19s/it]
 37%|███▋      | 5395/14725 [4:55:19<8:30:29,  3.28s/it]
 37%|███▋      | 5396/14725 [4:55:22<8:25:02,  3.25s/it]
 37%|███▋      | 5397/14725 [4:55:25<8:20:55,  3.22s/it]
 37%|███▋      | 5398/14725 [4:55:29<8:27:30,  3.26s/it]
 37%|███▋      | 5399/14725 [4:55:32<8:15:24,  3.19s/it]
 37%|███▋      | 5400/14725 [4:55:35<8:22:33,  3.23s/it]
                                                        

 37%|███▋      | 5400/14725 [4:55:35<8:22:33,  3.23s/it]
 37%|███▋      | 5401/14725 [4:55:38<8:36:01,  3.32s/it]
 37%|███▋      | 5402/14725 [4:55:42<8:36:28,  3.32s/it]
 37%|███▋      | 5403/14725 [4:55:45<8:31:33,  3.29s/it]
 37%|███▋      | 5404/14725 [4:55:48<8:30:26,  3.29s/it]
 37%|███▋      | 5405/14725 [4:55:51<8:20:59,  
0: {'loss': 0.7203, 'grad_norm': 0.4040721766821338, 'learning_rate': 1e-05, 'epoch': 0.37}
0: 3.23s/it]
 37%|███▋      | 5406/14725 [4:55:55<8:24:15,  3.25s/it]
 37%|███▋      | 5407/14725 [4:55:58<8:23:42,  3.24s/it]
 37%|███▋      | 5408/14725 [4:56:01<8:29:09,  3.28s/it]
 37%|███▋      | 5409/14725 [4:56:04<8:24:14,  3.25s/it]
 37%|███▋      | 5410/14725 [4:56:07<8:18:57,  3.21s/it]
                                                        

 37%|███▋      | 5410/14725 [4:56:07<8:18:57,  3.21s/it]
 37%|███▋      | 5411/14725 [4:56:11<8:25:24,  3.26s/it]
 37%|███▋      | 5412/14725 [4:56:14<8:31:25,  3.29s/it]
 37%|███▋      | 5413/14725 [4:56:18<8:32:20,  3.30s/it]
 37%|███▋      | 5414/14725 [4:56:21<8:31:21,  3.30s/it]
 37%|███▋      | 5415/14725 [4:56:24<8:34:28,  3.32s/it]
 37%|███▋      | 5416/14725 [4:56:27<8:30:31,  3.29s/it]
 37%|███▋      | 5417/14725 [4:56:31<8:25:17,  3.26s/it]
 37%|███▋      | 5418/14725 [4:56:34<8:23:39,  3.25s/it]
 37%|███▋      | 5419/14725 [4:56:37<8
0: {'loss': 0.7089, 'grad_norm': 0.42276868595907335, 'learning_rate': 1e-05, 'epoch': 0.37}
0: {'loss': 0.7006, 'grad_norm': 0.42965740972219213, 'learning_rate': 1e-05, 'epoch': 0.37}
0: :21:27,  3.23s/it]
 37%|███▋      | 5420/14725 [4:56:40<8:22:46,  3.24s/it]
                                                        

 37%|███▋      | 5420/14725 [4:56:40<8:22:46,  3.24s/it]
 37%|███▋      | 5421/14725 [4:56:44<8:26:43,  3.27s/it]
 37%|███▋      | 5422/14725 [4:56:47<8:17:30,  3.21s/it]
 37%|███▋      | 5423/14725 [4:56:50<8:30:40,  3.29s/it]
 37%|███▋      | 5424/14725 [4:56:53<8:30:25,  3.29s/it]
 37%|███▋      | 5425/14725 [4:56:57<8:25:57,  3.26s/it]
 37%|███▋      | 5426/14725 [4:57:00<8:29:09,  3.29s/it]
 37%|███▋      | 5427/14725 [4:57:03<8:30:01,  3.29s/it]
 37%|███▋      | 5428/14725 [4:57:07<8:28:21,  3.28s/it]
 37%|███▋      | 5429/14725 [4:57:10<8:20:45,  3.23s/it]
 37%|███▋      | 5430/14725 [4:57:13<8:14:30,  3.19s/it]
                                                        

 37%|███▋      | 5430/14725 [4:57:13<8:14:30,  3.19s/it]
 37%|███▋      | 5431/14725 [4:57:16
0: {'loss': 0.7102, 'grad_norm': 0.41075235149840933, 'learning_rate': 1e-05, 'epoch': 0.37}
0: <8:09:00,  3.16s/it]
 37%|███▋      | 5432/14725 [4:57:19<8:18:44,  3.22s/it]
 37%|███▋      | 5433/14725 [4:57:22<8:14:49,  3.20s/it]
 37%|███▋      | 5434/14725 [4:57:25<8:08:20,  3.15s/it]
 37%|███▋      | 5435/14725 [4:57:29<8:08:25,  3.15s/it]
 37%|███▋      | 5436/14725 [4:57:32<8:09:31,  3.16s/it]
 37%|███▋      | 5437/14725 [4:57:35<8:08:00,  3.15s/it]
 37%|███▋      | 5438/14725 [4:57:38<8:17:39,  3.22s/it]
 37%|███▋      | 5439/14725 [4:57:42<8:21:48,  3.24s/it]
 37%|███▋      | 5440/14725 [4:57:45<8:18:26,  3.22s/it]
                                                        

 37%|███▋      | 5440/14725 [4:57:45<8:18:26,  3.22s/it]
 37%|███▋      | 5441/14725 [4:57:48<8:17:40,  3.22s/it]
 37%|███▋      | 5442/14725 [4:57:51<8:20:13,  3.23s/it]
 37%|███▋      | 5443/14725 [4:57:54<8:19:45,  3.23s/it]
 37%|███▋      | 5444/14725 [4:57:58<8:20:49,  3.24s/it]
 37%|███▋      | 5445/14725
0: {'loss': 0.7099, 'grad_norm': 0.4249324090610497, 'learning_rate': 1e-05, 'epoch': 0.37}
0:  [4:58:01<8:27:30,  3.28s/it]
 37%|███▋      | 5446/14725 [4:58:04<8:27:41,  3.28s/it]
 37%|███▋      | 5447/14725 [4:58:08<8:25:50,  3.27s/it]
 37%|███▋      | 5448/14725 [4:58:11<8:17:45,  3.22s/it]
 37%|███▋      | 5449/14725 [4:58:14<8:23:23,  3.26s/it]
 37%|███▋      | 5450/14725 [4:58:17<8:16:27,  3.21s/it]
                                                        

 37%|███▋      | 5450/14725 [4:58:17<8:16:27,  3.21s/it]
 37%|███▋      | 5451/14725 [4:58:20<8:18:48,  3.23s/it]
 37%|███▋      | 5452/14725 [4:58:24<8:13:10,  3.19s/it]
 37%|███▋      | 5453/14725 [4:58:27<8:25:37,  3.27s/it]
 37%|███▋      | 5454/14725 [4:58:30<8:23:47,  3.26s/it]
 37%|███▋      | 5455/14725 [4:58:33<8:22:15,  3.25s/it]
 37%|███▋      | 5456/14725 [4:58:37<8:25:31,  3.27s/it]
 37%|███▋      | 5457/14725 [4:58:40<8:27:04,  3.28s/it]
 37%|███▋      | 5458/14725 [4:58:43<8:27:30,  3.29s/it]
 37%|███▋      | 5
0: {'loss': 0.7262, 'grad_norm': 0.4519507960033218, 'learning_rate': 1e-05, 'epoch': 0.37}
0: {'loss': 0.7091, 'grad_norm': 0.4490678135728241, 'learning_rate': 1e-05, 'epoch': 0.37}
0: 459/14725 [4:58:47<8:33:10,  3.32s/it]
 37%|███▋      | 5460/14725 [4:58:50<8:27:02,  3.28s/it]
                                                        

 37%|███▋      | 5460/14725 [4:58:50<8:27:02,  3.28s/it]
 37%|███▋      | 5461/14725 [4:58:53<8:15:20,  3.21s/it]
 37%|███▋      | 5462/14725 [4:58:56<8:12:02,  3.19s/it]
 37%|███▋      | 5463/14725 [4:58:59<8:10:23,  3.18s/it]
 37%|███▋      | 5464/14725 [4:59:03<8:15:26,  3.21s/it]
 37%|███▋      | 5465/14725 [4:59:06<8:19:23,  3.24s/it]
 37%|███▋      | 5466/14725 [4:59:09<8:22:57,  3.26s/it]
 37%|███▋      | 5467/14725 [4:59:12<8:19:52,  3.24s/it]
 37%|███▋      | 5468/14725 [4:59:16<8:27:29,  3.29s/it]
 37%|███▋      | 5469/14725 [4:59:19<8:37:48,  3.36s/it]
 37%|███▋      | 5470/14725 [4:59:23<8:35:21,  3.34s/it]
                                                        

 37%|███▋      | 5470/14725 [4:59:23<8:35:21,  3.34s/it]
 37%|███▋      |
0: {'loss': 0.7077, 'grad_norm': 0.4188959066049605, 'learning_rate': 1e-05, 'epoch': 0.37}
0:  5471/14725 [4:59:26<8:30:47,  3.31s/it]
 37%|███▋      | 5472/14725 [4:59:29<8:29:54,  3.31s/it]
 37%|███▋      | 5473/14725 [4:59:32<8:30:30,  3.31s/it]
 37%|███▋      | 5474/14725 [4:59:36<8:20:37,  3.25s/it]
 37%|███▋      | 5475/14725 [4:59:39<8:10:22,  3.18s/it]
 37%|███▋      | 5476/14725 [4:59:42<8:11:17,  3.19s/it]
 37%|███▋      | 5477/14725 [4:59:45<8:13:42,  3.20s/it]
 37%|███▋      | 5478/14725 [4:59:48<8:17:48,  3.23s/it]
 37%|███▋      | 5479/14725 [4:59:52<8:25:49,  3.28s/it]
 37%|███▋      | 5480/14725 [4:59:55<8:24:53,  3.28s/it]
                                                        

 37%|███▋      | 5480/14725 [4:59:55<8:24:53,  3.28s/it]
 37%|███▋      | 5481/14725 [4:59:58<8:23:40,  3.27s/it]
 37%|███▋      | 5482/14725 [5:00:02<8:27:35,  3.29s/it]
 37%|███▋      | 5483/14725 [5:00:05<8:39:27,  3.37s/it]
 37%|███▋      | 5484/14725 [5:00:08<8:33:21,  3.33s/it]
 37%|███�
0: {'loss': 0.7039, 'grad_norm': 0.4338088887718469, 'learning_rate': 1e-05, 'epoch': 0.37}
0: ��      | 5485/14725 [5:00:12<8:33:48,  3.34s/it]
 37%|███▋      | 5486/14725 [5:00:15<8:31:10,  3.32s/it]
 37%|███▋      | 5487/14725 [5:00:18<8:22:58,  3.27s/it]
 37%|███▋      | 5488/14725 [5:00:21<8:19:18,  3.24s/it]
 37%|███▋      | 5489/14725 [5:00:25<8:28:54,  3.31s/it]
 37%|███▋      | 5490/14725 [5:00:28<8:21:19,  3.26s/it]
                                                        

 37%|███▋      | 5490/14725 [5:00:28<8:21:19,  3.26s/it]
 37%|███▋      | 5491/14725 [5:00:31<8:20:58,  3.26s/it]
 37%|███▋      | 5492/14725 [5:00:35<8:30:05,  3.31s/it]
 37%|███▋      | 5493/14725 [5:00:38<8:26:26,  3.29s/it]
 37%|███▋      | 5494/14725 [5:00:41<8:17:02,  3.23s/it]
 37%|███▋      | 5495/14725 [5:00:44<8:13:06,  3.21s/it]
 37%|███▋      | 5496/14725 [5:00:47<8:08:41,  3.18s/it]
 37%|███▋      | 5497/14725 [5:00:50<8:05:47,  3.16s/it]
 37%|███▋      | 5498/14725 [5:00:54<8:16:12,  3.23s/it]
 37%|�
0: {'loss': 0.7159, 'grad_norm': 0.4550769636720895, 'learning_rate': 1e-05, 'epoch': 0.37}
0: {'loss': 0.7038, 'grad_norm': 0.4075735339034615, 'learning_rate': 1e-05, 'epoch': 0.37}
0: ��██▋      | 5499/14725 [5:00:57<8:23:49,  3.28s/it]
 37%|███▋      | 5500/14725 [5:01:00<8:19:09,  3.25s/it]
                                                        

 37%|███▋      | 5500/14725 [5:01:00<8:19:09,  3.25s/it]
 37%|███▋      | 5501/14725 [5:01:04<8:27:02,  3.30s/it]
 37%|███▋      | 5502/14725 [5:01:07<8:25:33,  3.29s/it]
 37%|███▋      | 5503/14725 [5:01:10<8:27:54,  3.30s/it]
 37%|███▋      | 5504/14725 [5:01:14<8:35:19,  3.35s/it]
 37%|███▋      | 5505/14725 [5:01:17<8:37:06,  3.37s/it]
 37%|███▋      | 5506/14725 [5:01:21<8:37:14,  3.37s/it]
 37%|███▋      | 5507/14725 [5:01:24<8:37:55,  3.37s/it]
 37%|███▋      | 5508/14725 [5:01:27<8:32:18,  3.34s/it]
 37%|███▋      | 5509/14725 [5:01:30<8:27:45,  3.31s/it]
 37%|███▋      | 5510/14725 [5:01:34<8:19:45,  3.25s/it]
                                                        

 37%|███▋      | 5510/14725 [5:01:34<8:19:45,  3.25s/it]
 37%
0: {'loss': 0.7113, 'grad_norm': 0.3998330548677613, 'learning_rate': 1e-05, 'epoch': 0.37}
0: |███▋      | 5511/14725 [5:01:37<8:17:08,  3.24s/it]
 37%|███▋      | 5512/14725 [5:01:40<8:24:04,  3.28s/it]
 37%|███▋      | 5513/14725 [5:01:43<8:20:48,  3.26s/it]
 37%|███▋      | 5514/14725 [5:01:47<8:19:41,  3.25s/it]
 37%|███▋      | 5515/14725 [5:01:50<8:17:47,  3.24s/it]
 37%|███▋      | 5516/14725 [5:01:53<8:23:30,  3.28s/it]
 37%|███▋      | 5517/14725 [5:01:56<8:16:27,  3.23s/it]
 37%|███▋      | 5518/14725 [5:02:00<8:18:17,  3.25s/it]
 37%|███▋      | 5519/14725 [5:02:03<8:22:49,  3.28s/it]
 37%|███▋      | 5520/14725 [5:02:06<8:18:17,  3.25s/it]
                                                        

 37%|███▋      | 5520/14725 [5:02:06<8:18:17,  3.25s/it]
 37%|███▋      | 5521/14725 [5:02:09<8:19:02,  3.25s/it]
 38%|███▊      | 5522/14725 [5:02:13<8:16:40,  3.24s/it]
 38%|███▊      | 5523/14725 [5:02:16<8:17:45,  3.25s/it]
 38%|███▊      | 5524/14725 [5:02:19<8:11:32,  3.21s
0: {'loss': 0.7022, 'grad_norm': 0.4450384689543365, 'learning_rate': 1e-05, 'epoch': 0.38}
0: /it]
 38%|███▊      | 5525/14725 [5:02:22<8:12:29,  3.21s/it]
 38%|███▊      | 5526/14725 [5:02:25<8:10:28,  3.20s/it]
 38%|███▊      | 5527/14725 [5:02:29<8:18:08,  3.25s/it]
 38%|███▊      | 5528/14725 [5:02:32<8:14:39,  3.23s/it]
 38%|███▊      | 5529/14725 [5:02:35<8:12:16,  3.21s/it]
 38%|███▊      | 5530/14725 [5:02:38<8:12:15,  3.21s/it]
                                                        

 38%|███▊      | 5530/14725 [5:02:38<8:12:15,  3.21s/it]
 38%|███▊      | 5531/14725 [5:02:41<8:08:56,  3.19s/it]
 38%|███▊      | 5532/14725 [5:02:45<8:23:09,  3.28s/it]
 38%|███▊      | 5533/14725 [5:02:48<8:17:43,  3.25s/it]
 38%|███▊      | 5534/14725 [5:02:52<8:30:54,  3.34s/it]
 38%|███▊      | 5535/14725 [5:02:55<8:38:46,  3.39s/it]
 38%|███▊      | 5536/14725 [5:02:58<8:30:01,  3.33s/it]
 38%|███▊      | 5537/14725 [5:03:01<8:22:19,  3.28s/it]
 38%|███▊      | 5538/14725 [5:03:05<8:15:2
0: {'loss': 0.7229, 'grad_norm': 0.42472294123900844, 'learning_rate': 1e-05, 'epoch': 0.38}
0: {'loss': 0.6959, 'grad_norm': 0.43138671035200804, 'learning_rate': 1e-05, 'epoch': 0.38}
0: 7,  3.24s/it]
 38%|███▊      | 5539/14725 [5:03:08<8:25:18,  3.30s/it]
 38%|███▊      | 5540/14725 [5:03:11<8:24:14,  3.29s/it]
                                                        

 38%|███▊      | 5540/14725 [5:03:11<8:24:14,  3.29s/it]
 38%|███▊      | 5541/14725 [5:03:15<8:24:10,  3.29s/it]
 38%|███▊      | 5542/14725 [5:03:18<8:24:10,  3.29s/it]
 38%|███▊      | 5543/14725 [5:03:21<8:16:55,  3.25s/it]
 38%|███▊      | 5544/14725 [5:03:24<8:20:14,  3.27s/it]
 38%|███▊      | 5545/14725 [5:03:28<8:19:04,  3.26s/it]
 38%|███▊      | 5546/14725 [5:03:31<8:14:04,  3.23s/it]
 38%|███▊      | 5547/14725 [5:03:34<8:21:09,  3.28s/it]
 38%|███▊      | 5548/14725 [5:03:37<8:16:37,  3.25s/it]
 38%|███▊      | 5549/14725 [5:03:41<8:25:46,  3.31s/it]
 38%|███▊      | 5550/14725 [5:03:44<8:21:43,  3.28s/it]
                                                        

 38%|███▊      | 5550/14725 [5:03:44<8:21
0: {'loss': 0.712, 'grad_norm': 0.41384132215131086, 'learning_rate': 1e-05, 'epoch': 0.38}
0: :43,  3.28s/it]
 38%|███▊      | 5551/14725 [5:03:47<8:17:59,  3.26s/it]
 38%|███▊      | 5552/14725 [5:03:50<8:16:42,  3.25s/it]
 38%|███▊      | 5553/14725 [5:03:54<8:12:27,  3.22s/it]
 38%|███▊      | 5554/14725 [5:03:57<8:05:31,  3.18s/it]
 38%|███▊      | 5555/14725 [5:04:00<8:04:07,  3.17s/it]
 38%|███▊      | 5556/14725 [5:04:03<8:12:33,  3.22s/it]
 38%|███▊      | 5557/14725 [5:04:07<8:20:40,  3.28s/it]
 38%|███▊      | 5558/14725 [5:04:10<8:18:27,  3.26s/it]
 38%|███▊      | 5559/14725 [5:04:13<8:20:43,  3.28s/it]
 38%|███▊      | 5560/14725 [5:04:16<8:19:17,  3.27s/it]
                                                        

 38%|███▊      | 5560/14725 [5:04:16<8:19:17,  3.27s/it]
 38%|███▊      | 5561/14725 [5:04:20<8:17:46,  3.26s/it]
 38%|███▊      | 5562/14725 [5:04:23<8:25:05,  3.31s/it]
 38%|███▊      | 5563/14725 [5:04:26<8:21:02,  3.28s/it]
 38%|███▊      | 5564/14725 [5:0
0: {'loss': 0.7131, 'grad_norm': 0.42664218711665053, 'learning_rate': 1e-05, 'epoch': 0.38}
0: 4:29<8:12:26,  3.23s/it]
 38%|███▊      | 5565/14725 [5:04:33<8:16:21,  3.25s/it]
 38%|███▊      | 5566/14725 [5:04:36<8:11:16,  3.22s/it]
 38%|███▊      | 5567/14725 [5:04:39<8:13:46,  3.24s/it]
 38%|███▊      | 5568/14725 [5:04:42<8:13:42,  3.23s/it]
 38%|███▊      | 5569/14725 [5:04:46<8:20:26,  3.28s/it]
 38%|███▊      | 5570/14725 [5:04:49<8:21:25,  3.29s/it]
                                                        

 38%|███▊      | 5570/14725 [5:04:49<8:21:25,  3.29s/it]
 38%|███▊      | 5571/14725 [5:04:52<8:20:15,  3.28s/it]
 38%|███▊      | 5572/14725 [5:04:56<8:28:21,  3.33s/it]
 38%|███▊      | 5573/14725 [5:04:59<8:22:40,  3.30s/it]
 38%|███▊      | 5574/14725 [5:05:02<8:25:00,  3.31s/it]
 38%|███▊      | 5575/14725 [5:05:06<8:24:22,  3.31s/it]
 38%|███▊      | 5576/14725 [5:05:09<8:18:21,  3.27s/it]
 38%|███▊      | 5577/14725 [5:05:12<8:27:15,  3.33s/it]
 38%|███▊      | 5578/1
0: {'loss': 0.7143, 'grad_norm': 0.42766478990319473, 'learning_rate': 1e-05, 'epoch': 0.38}
0: {'loss': 0.6834, 'grad_norm': 0.43852887041330535, 'learning_rate': 1e-05, 'epoch': 0.38}
0: 4725 [5:05:16<8:26:30,  3.32s/it]
 38%|███▊      | 5579/14725 [5:05:19<8:17:59,  3.27s/it]
 38%|███▊      | 5580/14725 [5:05:22<8:14:48,  3.25s/it]
                                                        

 38%|███▊      | 5580/14725 [5:05:22<8:14:48,  3.25s/it]
 38%|███▊      | 5581/14725 [5:05:25<8:15:22,  3.25s/it]
 38%|███▊      | 5582/14725 [5:05:28<8:12:24,  3.23s/it]
 38%|███▊      | 5583/14725 [5:05:31<8:05:48,  3.19s/it]
 38%|███▊      | 5584/14725 [5:05:35<8:07:26,  3.20s/it]
 38%|███▊      | 5585/14725 [5:05:38<8:08:55,  3.21s/it]
 38%|███▊      | 5586/14725 [5:05:41<8:12:54,  3.24s/it]
 38%|███▊      | 5587/14725 [5:05:44<8:11:28,  3.23s/it]
 38%|███▊      | 5588/14725 [5:05:48<8:12:27,  3.23s/it]
 38%|███▊      | 5589/14725 [5:05:51<8:22:05,  3.30s/it]
 38%|███▊      | 5590/14725 [5:05:54<8:24:21,  3.31s/it]
                                                        

 38%|███▊      | 5590
0: {'loss': 0.7143, 'grad_norm': 0.4110578792972888, 'learning_rate': 1e-05, 'epoch': 0.38}
0: /14725 [5:05:54<8:24:21,  3.31s/it]
 38%|███▊      | 5591/14725 [5:05:58<8:20:12,  3.29s/it]
 38%|███▊      | 5592/14725 [5:06:01<8:18:17,  3.27s/it]
 38%|███▊      | 5593/14725 [5:06:04<8:20:56,  3.29s/it]
 38%|███▊      | 5594/14725 [5:06:08<8:23:48,  3.31s/it]
 38%|███▊      | 5595/14725 [5:06:11<8:31:13,  3.36s/it]
 38%|███▊      | 5596/14725 [5:06:14<8:29:47,  3.35s/it]
 38%|███▊      | 5597/14725 [5:06:18<8:29:38,  3.35s/it]
 38%|███▊      | 5598/14725 [5:06:21<8:35:37,  3.39s/it]
 38%|███▊      | 5599/14725 [5:06:25<8:32:16,  3.37s/it]
 38%|███▊      | 5600/14725 [5:06:28<8:21:51,  3.30s/it]
                                                        

 38%|███▊      | 5600/14725 [5:06:28<8:21:51,  3.30s/it]
 38%|███▊      | 5601/14725 [5:06:31<8:13:39,  3.25s/it]
 38%|███▊      | 5602/14725 [5:06:34<8:11:53,  3.24s/it]
 38%|███▊      | 5603/14725 [5:06:37<8:22:32,  3.31s/it]
 38%|███▊   
0: {'loss': 0.7021, 'grad_norm': 0.42113084787034016, 'learning_rate': 1e-05, 'epoch': 0.38}
0:    | 5604/14725 [5:06:41<8:26:06,  3.33s/it]
 38%|███▊      | 5605/14725 [5:06:44<8:22:52,  3.31s/it]
 38%|███▊      | 5606/14725 [5:06:47<8:20:59,  3.30s/it]
 38%|███▊      | 5607/14725 [5:06:51<8:15:07,  3.26s/it]
 38%|███▊      | 5608/14725 [5:06:54<8:28:56,  3.35s/it]
 38%|███▊      | 5609/14725 [5:06:57<8:31:05,  3.36s/it]
 38%|███▊      | 5610/14725 [5:07:01<8:28:28,  3.35s/it]
                                                        

 38%|███▊      | 5610/14725 [5:07:01<8:28:28,  3.35s/it]
 38%|███▊      | 5611/14725 [5:07:04<8:22:16,  3.31s/it]
 38%|███▊      | 5612/14725 [5:07:07<8:22:18,  3.31s/it]
 38%|███▊      | 5613/14725 [5:07:10<8:15:52,  3.27s/it]
 38%|███▊      | 5614/14725 [5:07:14<8:16:07,  3.27s/it]
 38%|███▊      | 5615/14725 [5:07:17<8:11:24,  3.24s/it]
 38%|███▊      | 5616/14725 [5:07:20<8:02:36,  3.18s/it]
 38%|███▊      | 5617/14725 [5:07:23<8:05:18,  3.20s/it]
 38%|██
0: {'loss': 0.6916, 'grad_norm': 0.41685856693104695, 'learning_rate': 1e-05, 'epoch': 0.38}
0: {'loss': 0.7032, 'grad_norm': 0.43831582980433775, 'learning_rate': 1e-05, 'epoch': 0.38}
0: █▊      | 5618/14725 [5:07:27<8:10:15,  3.23s/it]
 38%|███▊      | 5619/14725 [5:07:30<8:06:41,  3.21s/it]
 38%|███▊      | 5620/14725 [5:07:33<8:07:12,  3.21s/it]
                                                        

 38%|███▊      | 5620/14725 [5:07:33<8:07:12,  3.21s/it]
 38%|███▊      | 5621/14725 [5:07:36<8:03:51,  3.19s/it]
 38%|███▊      | 5622/14725 [5:07:39<7:58:48,  3.16s/it]
 38%|███▊      | 5623/14725 [5:07:43<8:10:37,  3.23s/it]
 38%|███▊      | 5624/14725 [5:07:46<8:01:18,  3.17s/it]
 38%|███▊      | 5625/14725 [5:07:49<8:03:52,  3.19s/it]
 38%|███▊      | 5626/14725 [5:07:52<8:03:43,  3.19s/it]
 38%|███▊      | 5627/14725 [5:07:55<8:06:03,  3.21s/it]
 38%|███▊      | 5628/14725 [5:07:58<8:05:02,  3.20s/it]
 38%|███▊      | 5629/14725 [5:08:02<8:07:55,  3.22s/it]
 38%|███▊      | 5630/14725 [5:08:05<8:02:37,  3.18s/it]
                                                        

 38%|█�
0: {'loss': 0.713, 'grad_norm': 0.4224886230497042, 'learning_rate': 1e-05, 'epoch': 0.38}
0: ��█▊      | 5630/14725 [5:08:05<8:02:37,  3.18s/it]
 38%|███▊      | 5631/14725 [5:08:08<8:02:57,  3.19s/it]
 38%|███▊      | 5632/14725 [5:08:11<8:03:05,  3.19s/it]
 38%|███▊      | 5633/14725 [5:08:14<8:09:58,  3.23s/it]
 38%|███▊      | 5634/14725 [5:08:18<8:11:58,  3.25s/it]
 38%|███▊      | 5635/14725 [5:08:21<8:10:34,  3.24s/it]
 38%|███▊      | 5636/14725 [5:08:24<8:16:54,  3.28s/it]
 38%|███▊      | 5637/14725 [5:08:28<8:32:01,  3.38s/it]
 38%|███▊      | 5638/14725 [5:08:31<8:34:42,  3.40s/it]
 38%|███▊      | 5639/14725 [5:08:34<8:18:36,  3.29s/it]
 38%|███▊      | 5640/14725 [5:08:38<8:23:20,  3.32s/it]
                                                        

 38%|███▊      | 5640/14725 [5:08:38<8:23:20,  3.32s/it]
 38%|███▊      | 5641/14725 [5:08:41<8:12:19,  3.25s/it]
 38%|███▊      | 5642/14725 [5:08:44<8:10:26,  3.24s/it]
 38%|███▊      | 5643/14725 [5:08:47<8:13:25,  3.26s/it]
0: {'loss': 0.7288, 'grad_norm': 0.4615095901155572, 'learning_rate': 1e-05, 'epoch': 0.38}
0:  38%|███▊      | 5644/14725 [5:08:51<8:14:58,  3.27s/it]
 38%|███▊      | 5645/14725 [5:08:54<8:22:51,  3.32s/it]
 38%|███▊      | 5646/14725 [5:08:58<8:26:51,  3.35s/it]
 38%|███▊      | 5647/14725 [5:09:01<8:18:44,  3.30s/it]
 38%|███▊      | 5648/14725 [5:09:04<8:23:54,  3.33s/it]
 38%|███▊      | 5649/14725 [5:09:07<8:18:21,  3.29s/it]
 38%|███▊      | 5650/14725 [5:09:11<8:23:06,  3.33s/it]
                                                        

 38%|███▊      | 5650/14725 [5:09:11<8:23:06,  3.33s/it]
 38%|███▊      | 5651/14725 [5:09:14<8:21:23,  3.32s/it]
 38%|███▊      | 5652/14725 [5:09:17<8:21:55,  3.32s/it]
 38%|███▊      | 5653/14725 [5:09:21<8:18:56,  3.30s/it]
 38%|███▊      | 5654/14725 [5:09:24<8:15:41,  3.28s/it]
 38%|███▊      | 5655/14725 [5:09:27<8:11:56,  3.25s/it]
 38%|███▊      | 5656/14725 [5:09:30<8:09:56,  3.24s/it]
 38%|███▊      | 5657/14725 [5:09:33<8:05:55,  3
0: {'loss': 0.7233, 'grad_norm': 0.4129710249420506, 'learning_rate': 1e-05, 'epoch': 0.38}
0: .22s/it]
 38%|███▊      | 5658/14725 [5:09:37<8:16:33,  3.29s/it]
 38%|███▊      | 5659/14725 [5:09:40<8:24:20,  3.34s/it]
 38%|███▊      | 5660/14725 [5:09:44<8:24:01,  3.34s/it]
                                                        

 38%|███▊      | 5660/14725 [5:09:44<8:24:01,  3.34s/it]
 38%|███▊      | 5661/14725 [5:09:47<8:13:00,  3.26s/it]
 38%|███▊      | 5662/14725 [5:09:50<8:15:27,  3.28s/it]
 38%|███▊      | 5663/14725 [5:09:54<8:33:23,  3.40s/it]
 38%|███▊      | 5664/14725 [5:09:57<8:33:39,  3.40s/it]
 38%|███▊      | 5665/14725 [5:10:01<8:39:08,  3.44s/it]
 38%|███▊      | 5666/14725 [5:10:04<8:50:35,  3.51s/it]
 38%|███▊      | 5667/14725 [5:10:08<8:40:10,  3.45s/it]
 38%|███▊      | 5668/14725 [5:10:11<8:25:59,  3.35s/it]
 38%|███▊      | 5669/14725 [5:10:14<8:22:44,  3.33s/it]
 39%|███▊      | 5670/14725 [5:10:17<8:23:52,  3.34s/it]
                                               
0: {'loss': 0.7011, 'grad_norm': 0.42036984366409097, 'learning_rate': 1e-05, 'epoch': 0.39}
0: {'loss': 0.7144, 'grad_norm': 0.46397699508149665, 'learning_rate': 1e-05, 'epoch': 0.39}
0:          

 39%|███▊      | 5670/14725 [5:10:17<8:23:52,  3.34s/it]
 39%|███▊      | 5671/14725 [5:10:21<8:16:00,  3.29s/it]
 39%|███▊      | 5672/14725 [5:10:24<8:14:07,  3.27s/it]
 39%|███▊      | 5673/14725 [5:10:27<8:13:37,  3.27s/it]
 39%|███▊      | 5674/14725 [5:10:30<8:13:45,  3.27s/it]
 39%|███▊      | 5675/14725 [5:10:34<8:13:42,  3.27s/it]
 39%|███▊      | 5676/14725 [5:10:37<8:17:43,  3.30s/it]
 39%|███▊      | 5677/14725 [5:10:40<8:17:45,  3.30s/it]
 39%|███▊      | 5678/14725 [5:10:44<8:12:37,  3.27s/it]
 39%|███▊      | 5679/14725 [5:10:47<8:12:30,  3.27s/it]
 39%|███▊      | 5680/14725 [5:10:50<8:18:50,  3.31s/it]
                                                        

 39%|███▊      | 5680/14725 [5:10:50<8:18:50,  3.31s/it]
 39%|███▊      | 5681/14725 [5:10:53<8:10:41,  3.26s/it]
 39%|███▊      | 5682/14725 [5:10:57<8:18:27,  3.31s/it]
 39%|███▊      | 5683/14725 [5:11:00<
0: {'loss': 0.7099, 'grad_norm': 0.5014713247199699, 'learning_rate': 1e-05, 'epoch': 0.39}
0: 8:14:05,  3.28s/it]
 39%|███▊      | 5684/14725 [5:11:03<8:09:53,  3.25s/it]
 39%|███▊      | 5685/14725 [5:11:06<8:05:12,  3.22s/it]
 39%|███▊      | 5686/14725 [5:11:09<7:55:39,  3.16s/it]
 39%|███▊      | 5687/14725 [5:11:13<8:08:59,  3.25s/it]
 39%|███▊      | 5688/14725 [5:11:16<8:05:42,  3.22s/it]
 39%|███▊      | 5689/14725 [5:11:19<8:13:22,  3.28s/it]
 39%|███▊      | 5690/14725 [5:11:23<8:10:42,  3.26s/it]
                                                        

 39%|███▊      | 5690/14725 [5:11:23<8:10:42,  3.26s/it]
 39%|███▊      | 5691/14725 [5:11:26<8:21:01,  3.33s/it]
 39%|███▊      | 5692/14725 [5:11:29<8:09:53,  3.25s/it]
 39%|███▊      | 5693/14725 [5:11:33<8:17:05,  3.30s/it]
 39%|███▊      | 5694/14725 [5:11:36<8:18:47,  3.31s/it]
 39%|███▊      | 5695/14725 [5:11:39<8:10:55,  3.26s/it]
 39%|███▊      | 5696/14725 [5:11:42<8:06:24,  3.23s/it]
 39%|███▊      | 5697/14725 
0: {'loss': 0.6991, 'grad_norm': 0.4200114581435983, 'learning_rate': 1e-05, 'epoch': 0.39}
0: {'loss': 0.7188, 'grad_norm': 0.4298379375852389, 'learning_rate': 1e-05, 'epoch': 0.39}
0: [5:11:45<8:06:19,  3.23s/it]
 39%|███▊      | 5698/14725 [5:11:49<7:58:39,  3.18s/it]
 39%|███▊      | 5699/14725 [5:11:52<8:12:43,  3.28s/it]
 39%|███▊      | 5700/14725 [5:11:55<8:14:14,  3.29s/it]
                                                        

 39%|███▊      | 5700/14725 [5:11:55<8:14:14,  3.29s/it]
 39%|███▊      | 5701/14725 [5:11:58<8:08:48,  3.25s/it]
 39%|███▊      | 5702/14725 [5:12:02<8:07:44,  3.24s/it]
 39%|███▊      | 5703/14725 [5:12:05<8:16:36,  3.30s/it]
 39%|███▊      | 5704/14725 [5:12:09<8:22:13,  3.34s/it]
 39%|███▊      | 5705/14725 [5:12:12<8:15:44,  3.30s/it]
 39%|███▉      | 5706/14725 [5:12:15<8:09:37,  3.26s/it]
 39%|███▉      | 5707/14725 [5:12:18<8:09:16,  3.26s/it]
 39%|███▉      | 5708/14725 [5:12:21<8:04:19,  3.22s/it]
 39%|███▉      | 5709/14725 [5:12:25<8:29:27,  3.39s/it]
 39%|███▉      | 5710/14725 [5:12:28<8:23:44,  3.35s/it]
                           
0: {'loss': 0.7016, 'grad_norm': 0.42662077192402814, 'learning_rate': 1e-05, 'epoch': 0.39}
0:                              

 39%|███▉      | 5710/14725 [5:12:28<8:23:44,  3.35s/it]
 39%|███▉      | 5711/14725 [5:12:32<8:28:44,  3.39s/it]
 39%|███▉      | 5712/14725 [5:12:35<8:20:01,  3.33s/it]
 39%|███▉      | 5713/14725 [5:12:38<8:22:49,  3.35s/it]
 39%|███▉      | 5714/14725 [5:12:42<8:24:18,  3.36s/it]
 39%|███▉      | 5715/14725 [5:12:45<8:17:02,  3.31s/it]
 39%|███▉      | 5716/14725 [5:12:48<8:12:08,  3.28s/it]
 39%|███▉      | 5717/14725 [5:12:52<8:14:40,  3.29s/it]
 39%|███▉      | 5718/14725 [5:12:55<8:13:04,  3.28s/it]
 39%|███▉      | 5719/14725 [5:12:58<8:03:40,  3.22s/it]
 39%|███▉      | 5720/14725 [5:13:01<8:09:53,  3.26s/it]
                                                        

 39%|███▉      | 5720/14725 [5:13:01<8:09:53,  3.26s/it]
 39%|███▉      | 5721/14725 [5:13:05<8:11:57,  3.28s/it]
 39%|███▉      | 5722/14725 [5:13:08<8:18:25,  3.32s/it]
 39%|███▉      | 
0: {'loss': 0.7058, 'grad_norm': 0.4512262030406272, 'learning_rate': 1e-05, 'epoch': 0.39}
0: 5723/14725 [5:13:11<8:18:07,  3.32s/it]
 39%|███▉      | 5724/14725 [5:13:15<8:29:27,  3.40s/it]
 39%|███▉      | 5725/14725 [5:13:18<8:21:27,  3.34s/it]
 39%|███▉      | 5726/14725 [5:13:21<8:16:15,  3.31s/it]
 39%|███▉      | 5727/14725 [5:13:25<8:18:20,  3.32s/it]
 39%|███▉      | 5728/14725 [5:13:28<8:09:59,  3.27s/it]
 39%|███▉      | 5729/14725 [5:13:31<8:11:42,  3.28s/it]
 39%|███▉      | 5730/14725 [5:13:34<8:08:22,  3.26s/it]
                                                        

 39%|███▉      | 5730/14725 [5:13:34<8:08:22,  3.26s/it]
 39%|███▉      | 5731/14725 [5:13:38<8:06:10,  3.24s/it]
 39%|███▉      | 5732/14725 [5:13:41<8:01:20,  3.21s/it]
 39%|███▉      | 5733/14725 [5:13:44<7:58:38,  3.19s/it]
 39%|███▉      | 5734/14725 [5:13:47<8:05:53,  3.24s/it]
 39%|███▉      | 5735/14725 [5:13:51<8:11:21,  3.28s/it]
 39%|███▉      | 5736/14725 [5:13:54<8:24:14,  3.37s/it]
 39%|███�
0: {'loss': 0.7107, 'grad_norm': 0.4660915078877638, 'learning_rate': 1e-05, 'epoch': 0.39}
0: �      | 5737/14725 [5:13:57<8:13:22,  3.29s/it]
 39%|███▉      | 5738/14725 [5:14:01<8:16:36,  3.32s/it]
 39%|███▉      | 5739/14725 [5:14:04<8:21:59,  3.35s/it]
 39%|███▉      | 5740/14725 [5:14:07<8:17:07,  3.32s/it]
                                                        

 39%|███▉      | 5740/14725 [5:14:07<8:17:07,  3.32s/it]
 39%|███▉      | 5741/14725 [5:14:11<8:13:44,  3.30s/it]
 39%|███▉      | 5742/14725 [5:14:14<8:13:26,  3.30s/it]
 39%|███▉      | 5743/14725 [5:14:17<8:16:35,  3.32s/it]
 39%|███▉      | 5744/14725 [5:14:21<8:20:50,  3.35s/it]
 39%|███▉      | 5745/14725 [5:14:24<8:18:39,  3.33s/it]
 39%|███▉      | 5746/14725 [5:14:27<8:16:58,  3.32s/it]
 39%|███▉      | 5747/14725 [5:14:31<8:20:13,  3.34s/it]
 39%|███▉      | 5748/14725 [5:14:34<8:15:55,  3.31s/it]
 39%|███▉      | 5749/14725 [5:14:37<8:14:11,  3.30s/it]
 39%|███▉      | 5750/14725 [5:14:40<8:14:39,  3.31s/it]
       
0: {'loss': 0.6866, 'grad_norm': 0.4238169349043033, 'learning_rate': 1e-05, 'epoch': 0.39}
0: {'loss': 0.705, 'grad_norm': 0.4214814243689131, 'learning_rate': 1e-05, 'epoch': 0.39}
0:                                                  

 39%|███▉      | 5750/14725 [5:14:40<8:14:39,  3.31s/it]
 39%|███▉      | 5751/14725 [5:14:44<8:10:15,  3.28s/it]
 39%|███▉      | 5752/14725 [5:14:47<8:11:01,  3.28s/it]
 39%|███▉      | 5753/14725 [5:14:50<8:11:30,  3.29s/it]
 39%|███▉      | 5754/14725 [5:14:53<8:03:45,  3.24s/it]
 39%|███▉      | 5755/14725 [5:14:57<8:05:54,  3.25s/it]
 39%|███▉      | 5756/14725 [5:15:00<8:04:56,  3.24s/it]
 39%|███▉      | 5757/14725 [5:15:03<7:57:15,  3.19s/it]
 39%|███▉      | 5758/14725 [5:15:06<7:59:44,  3.21s/it]
 39%|███▉      | 5759/14725 [5:15:09<8:00:24,  3.21s/it]
 39%|███▉      | 5760/14725 [5:15:13<8:09:16,  3.27s/it]
                                                        

 39%|███▉      | 5760/14725 [5:15:13<8:09:16,  3.27s/it]
 39%|███▉      | 5761/14725 [5:15:16<8:15:17,  3.32s/it]
 39%|███▉      | 5762/14725 [5:15:19<8:11:15,  3.29s/it]
 39%|
0: {'loss': 0.7011, 'grad_norm': 0.40925473785575694, 'learning_rate': 1e-05, 'epoch': 0.39}
0: ███▉      | 5763/14725 [5:15:23<8:06:39,  3.26s/it]
 39%|███▉      | 5764/14725 [5:15:26<8:07:46,  3.27s/it]
 39%|███▉      | 5765/14725 [5:15:29<8:18:53,  3.34s/it]
 39%|███▉      | 5766/14725 [5:15:33<8:12:59,  3.30s/it]
 39%|███▉      | 5767/14725 [5:15:36<8:09:28,  3.28s/it]
 39%|███▉      | 5768/14725 [5:15:39<8:06:54,  3.26s/it]
 39%|███▉      | 5769/14725 [5:15:42<8:05:05,  3.25s/it]
 39%|███▉      | 5770/14725 [5:15:46<8:03:47,  3.24s/it]
                                                        

 39%|███▉      | 5770/14725 [5:15:46<8:03:47,  3.24s/it]
 39%|███▉      | 5771/14725 [5:15:49<8:11:19,  3.29s/it]
 39%|███▉      | 5772/14725 [5:15:52<8:07:20,  3.27s/it]
 39%|███▉      | 5773/14725 [5:15:55<8:02:49,  3.24s/it]
 39%|███▉      | 5774/14725 [5:15:58<7:57:25,  3.20s/it]
 39%|███▉      | 5775/14725 [5:16:02<8:06:52,  3.26s/it]
 39%|███▉      | 5776/14725 [5:16:05<8:05:13,  3.25s/
0: {'loss': 0.7123, 'grad_norm': 0.44094386287162307, 'learning_rate': 1e-05, 'epoch': 0.39}
0: it]
 39%|███▉      | 5777/14725 [5:16:09<8:16:13,  3.33s/it]
 39%|███▉      | 5778/14725 [5:16:12<8:21:58,  3.37s/it]
 39%|███▉      | 5779/14725 [5:16:15<8:22:15,  3.37s/it]
 39%|███▉      | 5780/14725 [5:16:19<8:13:25,  3.31s/it]
                                                        

 39%|███▉      | 5780/14725 [5:16:19<8:13:25,  3.31s/it]
 39%|███▉      | 5781/14725 [5:16:22<8:14:20,  3.32s/it]
 39%|███▉      | 5782/14725 [5:16:25<8:13:38,  3.31s/it]
 39%|███▉      | 5783/14725 [5:16:28<8:09:43,  3.29s/it]
 39%|███▉      | 5784/14725 [5:16:32<8:02:53,  3.24s/it]
 39%|███▉      | 5785/14725 [5:16:35<8:04:06,  3.25s/it]
 39%|███▉      | 5786/14725 [5:16:38<8:06:11,  3.26s/it]
 39%|███▉      | 5787/14725 [5:16:41<8:08:10,  3.28s/it]
 39%|███▉      | 5788/14725 [5:16:45<8:05:16,  3.26s/it]
 39%|███▉      | 5789/14725 [5:16:48<8:07:42,  3.27s/it]
 39%|███▉      | 5790/14725 [5:16:51<8:01:25
0: {'loss': 0.7245, 'grad_norm': 0.4155208956338423, 'learning_rate': 1e-05, 'epoch': 0.39}
0: {'loss': 0.6937, 'grad_norm': 0.44630500747979324, 'learning_rate': 1e-05, 'epoch': 0.39}
0: ,  3.23s/it]
                                                        

 39%|███▉      | 5790/14725 [5:16:51<8:01:25,  3.23s/it]
 39%|███▉      | 5791/14725 [5:16:54<7:56:39,  3.20s/it]
 39%|███▉      | 5792/14725 [5:16:58<7:59:41,  3.22s/it]
 39%|███▉      | 5793/14725 [5:17:01<8:02:58,  3.24s/it]
 39%|███▉      | 5794/14725 [5:17:04<8:03:35,  3.25s/it]
 39%|███▉      | 5795/14725 [5:17:07<8:06:10,  3.27s/it]
 39%|███▉      | 5796/14725 [5:17:11<8:11:58,  3.31s/it]
 39%|███▉      | 5797/14725 [5:17:14<8:09:51,  3.29s/it]
 39%|███▉      | 5798/14725 [5:17:17<8:11:19,  3.30s/it]
 39%|███▉      | 5799/14725 [5:17:21<8:15:41,  3.33s/it]
 39%|███▉      | 5800/14725 [5:17:24<8:09:14,  3.29s/it]
                                                        

 39%|███▉      | 5800/14725 [5:17:24<8:09:14,  3.29s/it]
 39%|███▉      | 5801/14725 [5:17:27<8:08:19,  3.28s/it]
 39%|███▉      | 5802/14725 [5:17:31<8:13:
0: {'loss': 0.706, 'grad_norm': 0.4006789390855213, 'learning_rate': 1e-05, 'epoch': 0.39}
0: 21,  3.32s/it]
 39%|███▉      | 5803/14725 [5:17:34<8:04:07,  3.26s/it]
 39%|███▉      | 5804/14725 [5:17:37<8:02:26,  3.24s/it]
 39%|███▉      | 5805/14725 [5:17:40<8:00:01,  3.23s/it]
 39%|███▉      | 5806/14725 [5:17:44<8:10:58,  3.30s/it]
 39%|███▉      | 5807/14725 [5:17:47<8:17:23,  3.35s/it]
 39%|███▉      | 5808/14725 [5:17:50<8:10:43,  3.30s/it]
 39%|███▉      | 5809/14725 [5:17:54<8:10:31,  3.30s/it]
 39%|███▉      | 5810/14725 [5:17:57<8:16:30,  3.34s/it]
                                                        

 39%|███▉      | 5810/14725 [5:17:57<8:16:30,  3.34s/it]
 39%|███▉      | 5811/14725 [5:18:00<8:14:51,  3.33s/it]
 39%|███▉      | 5812/14725 [5:18:03<8:06:50,  3.28s/it]
 39%|███▉      | 5813/14725 [5:18:07<8:10:03,  3.30s/it]
 39%|███▉      | 5814/14725 [5:18:10<7:57:54,  3.22s/it]
 39%|███▉      | 5815/14725 [5:18:13<8:04:10,  3.26s/it]
 39%|███▉      | 5816/14725 [5:18
0: {'loss': 0.6983, 'grad_norm': 0.5128196008409235, 'learning_rate': 1e-05, 'epoch': 0.4}
0: :17<8:11:01,  3.31s/it]
 40%|███▉      | 5817/14725 [5:18:20<8:16:53,  3.35s/it]
 40%|███▉      | 5818/14725 [5:18:23<8:07:03,  3.28s/it]
 40%|███▉      | 5819/14725 [5:18:26<8:04:01,  3.26s/it]
 40%|███▉      | 5820/14725 [5:18:30<8:03:48,  3.26s/it]
                                                        

 40%|███▉      | 5820/14725 [5:18:30<8:03:48,  3.26s/it]
 40%|███▉      | 5821/14725 [5:18:33<8:07:45,  3.29s/it]
 40%|███▉      | 5822/14725 [5:18:36<8:11:21,  3.31s/it]
 40%|███▉      | 5823/14725 [5:18:40<8:18:08,  3.36s/it]
 40%|███▉      | 5824/14725 [5:18:43<8:10:33,  3.31s/it]
 40%|███▉      | 5825/14725 [5:18:46<8:08:27,  3.29s/it]
 40%|███▉      | 5826/14725 [5:18:50<8:17:00,  3.35s/it]
 40%|███▉      | 5827/14725 [5:18:53<8:10:08,  3.31s/it]
 40%|███▉      | 5828/14725 [5:18:56<8:03:57,  3.26s/it]
 40%|███▉      | 5829/14725 [5:18:59<8:04:32,  3.27s/it]
 40%|███▉      | 5830/14
0: {'loss': 0.716, 'grad_norm': 0.4571935661308541, 'learning_rate': 1e-05, 'epoch': 0.4}
0: {'loss': 0.7178, 'grad_norm': 0.44679000628334015, 'learning_rate': 1e-05, 'epoch': 0.4}
0: 725 [5:19:03<7:58:32,  3.23s/it]
                                                        

 40%|███▉      | 5830/14725 [5:19:03<7:58:32,  3.23s/it]
 40%|███▉      | 5831/14725 [5:19:06<7:59:05,  3.23s/it]
 40%|███▉      | 5832/14725 [5:19:09<7:56:11,  3.21s/it]
 40%|███▉      | 5833/14725 [5:19:12<7:48:12,  3.16s/it]
 40%|███▉      | 5834/14725 [5:19:15<7:46:11,  3.15s/it]
 40%|███▉      | 5835/14725 [5:19:19<7:59:21,  3.24s/it]
 40%|███▉      | 5836/14725 [5:19:22<7:55:20,  3.21s/it]
 40%|███▉      | 5837/14725 [5:19:25<7:56:13,  3.21s/it]
 40%|███▉      | 5838/14725 [5:19:28<7:52:31,  3.19s/it]
 40%|███▉      | 5839/14725 [5:19:31<7:55:30,  3.21s/it]
 40%|███▉      | 5840/14725 [5:19:34<7:52:28,  3.19s/it]
                                                        

 40%|███▉      | 5840/14725 [5:19:34<7:52:28,  3.19s/it]
 40%|███▉      | 5841/14725 [5:19:38<7:59:34,  3.24s/it]
 40%|███▉      | 5842/
0: {'loss': 0.7135, 'grad_norm': 0.4254749294069591, 'learning_rate': 1e-05, 'epoch': 0.4}
0: 14725 [5:19:41<7:57:15,  3.22s/it]
 40%|███▉      | 5843/14725 [5:19:44<7:59:52,  3.24s/it]
 40%|███▉      | 5844/14725 [5:19:48<8:01:15,  3.25s/it]
 40%|███▉      | 5845/14725 [5:19:51<8:03:46,  3.27s/it]
 40%|███▉      | 5846/14725 [5:19:54<7:56:40,  3.22s/it]
 40%|███▉      | 5847/14725 [5:19:57<8:03:19,  3.27s/it]
 40%|███▉      | 5848/14725 [5:20:01<8:12:12,  3.33s/it]
 40%|███▉      | 5849/14725 [5:20:04<8:18:03,  3.37s/it]
 40%|███▉      | 5850/14725 [5:20:07<8:02:34,  3.26s/it]
                                                        

 40%|███▉      | 5850/14725 [5:20:07<8:02:34,  3.26s/it]
 40%|███▉      | 5851/14725 [5:20:11<8:00:25,  3.25s/it]
 40%|███▉      | 5852/14725 [5:20:14<8:02:45,  3.26s/it]
 40%|███▉      | 5853/14725 [5:20:17<8:08:11,  3.30s/it]
 40%|███▉      | 5854/14725 [5:20:20<8:01:56,  3.26s/it]
 40%|███▉      | 5855/14725 [5:20:24<8:04:43,  3.28s/it]
 40%|███▉    
0: {'loss': 0.7259, 'grad_norm': 0.4364642116945564, 'learning_rate': 1e-05, 'epoch': 0.4}
0:   | 5856/14725 [5:20:27<8:12:09,  3.33s/it]
 40%|███▉      | 5857/14725 [5:20:30<8:06:55,  3.29s/it]
 40%|███▉      | 5858/14725 [5:20:34<8:10:43,  3.32s/it]
 40%|███▉      | 5859/14725 [5:20:37<8:02:51,  3.27s/it]
 40%|███▉      | 5860/14725 [5:20:40<8:01:03,  3.26s/it]
                                                        

 40%|███▉      | 5860/14725 [5:20:40<8:01:03,  3.26s/it]
 40%|███▉      | 5861/14725 [5:20:43<8:02:38,  3.27s/it]
 40%|███▉      | 5862/14725 [5:20:47<8:04:55,  3.28s/it]
 40%|███▉      | 5863/14725 [5:20:50<8:10:25,  3.32s/it]
 40%|███▉      | 5864/14725 [5:20:53<8:04:21,  3.28s/it]
 40%|███▉      | 5865/14725 [5:20:57<8:03:38,  3.28s/it]
 40%|███▉      | 5866/14725 [5:21:00<8:09:39,  3.32s/it]
 40%|███▉      | 5867/14725 [5:21:03<8:13:27,  3.34s/it]
 40%|███▉      | 5868/14725 [5:21:07<8:05:13,  3.29s/it]
 40%|███▉      | 5869/14725 [5:21:10<7:52:25,  3.20s/it]
 40%|██�
0: {'loss': 0.6976, 'grad_norm': 0.4342881111540947, 'learning_rate': 1e-05, 'epoch': 0.4}
0: {'loss': 0.6941, 'grad_norm': 0.43752462368475514, 'learning_rate': 1e-05, 'epoch': 0.4}
0: ��▉      | 5870/14725 [5:21:13<7:58:14,  3.24s/it]
                                                        

 40%|███▉      | 5870/14725 [5:21:13<7:58:14,  3.24s/it]
 40%|███▉      | 5871/14725 [5:21:16<8:08:50,  3.31s/it]
 40%|███▉      | 5872/14725 [5:21:20<8:05:33,  3.29s/it]
 40%|███▉      | 5873/14725 [5:21:23<8:01:00,  3.26s/it]
 40%|███▉      | 5874/14725 [5:21:26<7:57:35,  3.24s/it]
 40%|███▉      | 5875/14725 [5:21:29<7:56:46,  3.23s/it]
 40%|███▉      | 5876/14725 [5:21:32<7:54:40,  3.22s/it]
 40%|███▉      | 5877/14725 [5:21:36<7:49:30,  3.18s/it]
 40%|███▉      | 5878/14725 [5:21:39<7:45:57,  3.16s/it]
 40%|███▉      | 5879/14725 [5:21:42<7:52:08,  3.20s/it]
 40%|███▉      | 5880/14725 [5:21:45<7:57:47,  3.24s/it]
                                                        

 40%|███▉      | 5880/14725 [5:21:45<7:57:47,  3.24s/it]
 40%|███▉      | 5881/14725 [5:21:48<7:57:21,  3.24s/it]
 40%|█�
0: {'loss': 0.7238, 'grad_norm': 0.43397595448555953, 'learning_rate': 1e-05, 'epoch': 0.4}
0: �█▉      | 5882/14725 [5:21:52<8:01:30,  3.27s/it]
 40%|███▉      | 5883/14725 [5:21:55<8:04:11,  3.29s/it]
 40%|███▉      | 5884/14725 [5:21:59<8:09:53,  3.32s/it]
 40%|███▉      | 5885/14725 [5:22:02<7:59:31,  3.25s/it]
 40%|███▉      | 5886/14725 [5:22:05<7:59:20,  3.25s/it]
 40%|███▉      | 5887/14725 [5:22:08<7:59:52,  3.26s/it]
 40%|███▉      | 5888/14725 [5:22:11<7:57:41,  3.24s/it]
 40%|███▉      | 5889/14725 [5:22:15<7:58:18,  3.25s/it]
 40%|████      | 5890/14725 [5:22:18<8:10:54,  3.33s/it]
                                                        

 40%|████      | 5890/14725 [5:22:18<8:10:54,  3.33s/it]
 40%|████      | 5891/14725 [5:22:29<13:41:12,  5.58s/it]
 40%|████      | 5892/14725 [5:22:32<12:00:40,  4.90s/it]
 40%|████      | 5893/14725 [5:22:35<10:43:19,  4.37s/it]
 40%|████      | 5894/14725 [5:22:39<9:56:30,  4.05s/it] 
 40%|████      | 5895/14725 [5:22:42<9:28:13,  3.86s/i
0: {'loss': 0.7036, 'grad_norm': 0.45550179286592163, 'learning_rate': 1e-05, 'epoch': 0.4}
0: t]
 40%|████      | 5896/14725 [5:22:45<9:00:44,  3.67s/it]
 40%|████      | 5897/14725 [5:22:49<8:42:17,  3.55s/it]
 40%|████      | 5898/14725 [5:22:52<8:35:42,  3.51s/it]
 40%|████      | 5899/14725 [5:22:55<8:22:25,  3.42s/it]
 40%|████      | 5900/14725 [5:22:58<8:10:13,  3.33s/it]
                                                        

 40%|████      | 5900/14725 [5:22:58<8:10:13,  3.33s/it]
 40%|████      | 5901/14725 [5:23:02<8:05:14,  3.30s/it]
 40%|████      | 5902/14725 [5:23:05<8:05:24,  3.30s/it]
 40%|████      | 5903/14725 [5:23:08<7:57:23,  3.25s/it]
 40%|████      | 5904/14725 [5:23:11<7:56:26,  3.24s/it]
 40%|████      | 5905/14725 [5:23:15<8:04:40,  3.30s/it]
 40%|████      | 5906/14725 [5:23:18<7:55:18,  3.23s/it]
 40%|████      | 5907/14725 [5:23:21<7:51:38,  3.21s/it]
 40%|████      | 5908/14725 [5:23:24<7:47:22,  3.18s/it]
 40%|████      | 5909/14725 [5:23:27<7:43:11,
0: {'loss': 0.699, 'grad_norm': 0.45640291855503606, 'learning_rate': 1e-05, 'epoch': 0.4}
0: {'loss': 0.7075, 'grad_norm': 0.45054316613031375, 'learning_rate': 1e-05, 'epoch': 0.4}
0:   3.15s/it]
 40%|████      | 5910/14725 [5:23:30<7:47:36,  3.18s/it]
                                                        

 40%|████      | 5910/14725 [5:23:30<7:47:36,  3.18s/it]
 40%|████      | 5911/14725 [5:23:34<7:45:48,  3.17s/it]
 40%|████      | 5912/14725 [5:23:37<7:52:13,  3.21s/it]
 40%|████      | 5913/14725 [5:23:40<7:53:20,  3.22s/it]
 40%|████      | 5914/14725 [5:23:43<7:50:53,  3.21s/it]
 40%|████      | 5915/14725 [5:23:47<7:54:31,  3.23s/it]
 40%|████      | 5916/14725 [5:23:50<7:49:21,  3.20s/it]
 40%|████      | 5917/14725 [5:23:53<7:49:45,  3.20s/it]
 40%|████      | 5918/14725 [5:23:56<7:43:31,  3.16s/it]
 40%|████      | 5919/14725 [5:23:59<7:42:53,  3.15s/it]
 40%|████      | 5920/14725 [5:24:02<7:52:17,  3.22s/it]
                                                        

 40%|████      | 5920/14725 [5:24:02<7:52:17,  3.22s/it]
 40%|████      | 5921/14725 [5:24:06<7:54:2
0: {'loss': 0.7094, 'grad_norm': 0.41365089234132285, 'learning_rate': 1e-05, 'epoch': 0.4}
0: 6,  3.23s/it]
 40%|████      | 5922/14725 [5:24:09<7:55:45,  3.24s/it]
 40%|████      | 5923/14725 [5:24:12<7:51:57,  3.22s/it]
 40%|████      | 5924/14725 [5:24:15<7:53:58,  3.23s/it]
 40%|████      | 5925/14725 [5:24:19<7:56:25,  3.25s/it]
 40%|████      | 5926/14725 [5:24:22<8:02:52,  3.29s/it]
 40%|████      | 5927/14725 [5:24:25<7:57:34,  3.26s/it]
 40%|████      | 5928/14725 [5:24:28<7:51:40,  3.22s/it]
 40%|████      | 5929/14725 [5:24:32<7:55:22,  3.24s/it]
 40%|████      | 5930/14725 [5:24:35<8:00:31,  3.28s/it]
                                                        

 40%|████      | 5930/14725 [5:24:35<8:00:31,  3.28s/it]
 40%|████      | 5931/14725 [5:24:38<7:57:31,  3.26s/it]
 40%|████      | 5932/14725 [5:24:42<8:01:38,  3.29s/it]
 40%|████      | 5933/14725 [5:24:45<7:59:05,  3.27s/it]
 40%|████      | 5934/14725 [5:24:48<7:58:07,  3.26s/it]
 40%|████      | 5935/14725 [5:24:
0: {'loss': 0.6968, 'grad_norm': 0.4825668819180876, 'learning_rate': 1e-05, 'epoch': 0.4}
0: 51<7:54:19,  3.24s/it]
 40%|████      | 5936/14725 [5:24:55<7:57:54,  3.26s/it]
 40%|████      | 5937/14725 [5:24:58<7:59:43,  3.28s/it]
 40%|████      | 5938/14725 [5:25:01<8:00:05,  3.28s/it]
 40%|████      | 5939/14725 [5:25:04<8:00:31,  3.28s/it]
 40%|████      | 5940/14725 [5:25:08<8:00:23,  3.28s/it]
                                                        

 40%|████      | 5940/14725 [5:25:08<8:00:23,  3.28s/it]
 40%|████      | 5941/14725 [5:25:11<8:05:36,  3.32s/it]
 40%|████      | 5942/14725 [5:25:14<8:04:53,  3.31s/it]
 40%|████      | 5943/14725 [5:25:18<8:06:11,  3.32s/it]
 40%|████      | 5944/14725 [5:25:21<8:06:58,  3.33s/it]
 40%|████      | 5945/14725 [5:25:24<8:07:57,  3.33s/it]
 40%|████      | 5946/14725 [5:25:28<8:14:14,  3.38s/it]
 40%|████      | 5947/14725 [5:25:31<8:01:31,  3.29s/it]
 40%|████      | 5948/14725 [5:25:34<8:06:02,  3.32s/it]
 40%|████      | 5949/147
0: {'loss': 0.7204, 'grad_norm': 0.5007967921834017, 'learning_rate': 1e-05, 'epoch': 0.4}
0: {'loss': 0.708, 'grad_norm': 0.4173532844029284, 'learning_rate': 1e-05, 'epoch': 0.4}
0: 25 [5:25:38<8:05:37,  3.32s/it]
 40%|████      | 5950/14725 [5:25:41<8:09:07,  3.34s/it]
                                                        

 40%|████      | 5950/14725 [5:25:41<8:09:07,  3.34s/it]
 40%|████      | 5951/14725 [5:25:44<8:08:12,  3.34s/it]
 40%|████      | 5952/14725 [5:25:48<8:04:39,  3.31s/it]
 40%|████      | 5953/14725 [5:25:51<8:01:37,  3.29s/it]
 40%|████      | 5954/14725 [5:25:54<8:01:56,  3.30s/it]
 40%|████      | 5955/14725 [5:25:58<8:02:13,  3.30s/it]
 40%|████      | 5956/14725 [5:26:01<7:58:45,  3.28s/it]
 40%|████      | 5957/14725 [5:26:04<7:51:15,  3.22s/it]
 40%|████      | 5958/14725 [5:26:07<7:51:28,  3.23s/it]
 40%|████      | 5959/14725 [5:26:11<8:07:52,  3.34s/it]
 40%|████      | 5960/14725 [5:26:14<8:01:24,  3.30s/it]
                                                        

 40%|████      | 5960/14725 [5:26:14<8:01:24,  3.30s/it]
 40%|████      | 5961/1
0: {'loss': 0.7076, 'grad_norm': 0.5068029872215418, 'learning_rate': 1e-05, 'epoch': 0.41}
0: 4725 [5:26:17<8:05:47,  3.33s/it]
 40%|████      | 5962/14725 [5:26:21<8:05:13,  3.32s/it]
 40%|████      | 5963/14725 [5:26:24<8:01:28,  3.30s/it]
 41%|████      | 5964/14725 [5:26:27<8:04:32,  3.32s/it]
 41%|████      | 5965/14725 [5:26:31<8:03:11,  3.31s/it]
 41%|████      | 5966/14725 [5:26:34<8:07:26,  3.34s/it]
 41%|████      | 5967/14725 [5:26:37<8:01:07,  3.30s/it]
 41%|████      | 5968/14725 [5:26:41<8:05:17,  3.33s/it]
 41%|████      | 5969/14725 [5:26:44<8:03:16,  3.31s/it]
 41%|████      | 5970/14725 [5:26:47<7:48:24,  3.21s/it]
                                                        

 41%|████      | 5970/14725 [5:26:47<7:48:24,  3.21s/it]
 41%|████      | 5971/14725 [5:26:50<7:47:43,  3.21s/it]
 41%|████      | 5972/14725 [5:26:53<7:52:41,  3.24s/it]
 41%|████      | 5973/14725 [5:26:57<7:55:37,  3.26s/it]
 41%|████      | 5974/14725 [5:27:00<7:51:18,  3.23s/it]
 41%|████     
0: {'loss': 0.7004, 'grad_norm': 0.44675418886733276, 'learning_rate': 1e-05, 'epoch': 0.41}
0:  | 5975/14725 [5:27:03<8:04:00,  3.32s/it]
 41%|████      | 5976/14725 [5:27:07<8:00:27,  3.30s/it]
 41%|████      | 5977/14725 [5:27:10<7:49:34,  3.22s/it]
 41%|████      | 5978/14725 [5:27:13<7:46:02,  3.20s/it]
 41%|████      | 5979/14725 [5:27:16<7:55:34,  3.26s/it]
 41%|████      | 5980/14725 [5:27:19<7:50:59,  3.23s/it]
                                                        

 41%|████      | 5980/14725 [5:27:19<7:50:59,  3.23s/it]
 41%|████      | 5981/14725 [5:27:23<7:51:55,  3.24s/it]
 41%|████      | 5982/14725 [5:27:26<7:47:51,  3.21s/it]
 41%|████      | 5983/14725 [5:27:29<7:51:10,  3.23s/it]
 41%|████      | 5984/14725 [5:27:32<7:47:49,  3.21s/it]
 41%|████      | 5985/14725 [5:27:35<7:46:27,  3.20s/it]
 41%|████      | 5986/14725 [5:27:38<7:40:23,  3.16s/it]
 41%|████      | 5987/14725 [5:27:42<7:46:43,  3.20s/it]
 41%|████      | 5988/14725 [5:27:45<7:50:56,  3.23s/it]
 41%|██�
0: {'loss': 0.707, 'grad_norm': 0.42038991671347803, 'learning_rate': 1e-05, 'epoch': 0.41}
0: {'loss': 0.7033, 'grad_norm': 0.44724668649561067, 'learning_rate': 1e-05, 'epoch': 0.41}
0: �█      | 5989/14725 [5:27:48<7:46:51,  3.21s/it]
 41%|████      | 5990/14725 [5:27:51<7:45:46,  3.20s/it]
                                                        

 41%|████      | 5990/14725 [5:27:51<7:45:46,  3.20s/it]
 41%|████      | 5991/14725 [5:27:55<7:48:26,  3.22s/it]
 41%|████      | 5992/14725 [5:27:58<7:57:05,  3.28s/it]
 41%|████      | 5993/14725 [5:28:01<7:53:56,  3.26s/it]
 41%|████      | 5994/14725 [5:28:05<7:55:25,  3.27s/it]
 41%|████      | 5995/14725 [5:28:08<7:57:09,  3.28s/it]
 41%|████      | 5996/14725 [5:28:11<7:58:08,  3.29s/it]
 41%|████      | 5997/14725 [5:28:14<7:57:43,  3.28s/it]
 41%|████      | 5998/14725 [5:28:18<7:55:27,  3.27s/it]
 41%|████      | 5999/14725 [5:28:21<7:51:03,  3.24s/it]
 41%|████      | 6000/14725 [5:28:24<7:46:15,  3.21s/it]
                                                        

 41%|████      | 6000/14725 [5:28:24<7:46:15,  3.21s/it]
 41%|██
0: {'loss': 0.7014, 'grad_norm': 0.4313069832725854, 'learning_rate': 1e-05, 'epoch': 0.41}
0: ██      | 6001/14725 [5:28:27<7:47:47,  3.22s/it]
 41%|████      | 6002/14725 [5:28:31<8:08:13,  3.36s/it]
 41%|████      | 6003/14725 [5:28:34<8:03:02,  3.32s/it]
 41%|████      | 6004/14725 [5:28:37<7:53:54,  3.26s/it]
 41%|████      | 6005/14725 [5:28:40<7:50:00,  3.23s/it]
 41%|████      | 6006/14725 [5:28:44<7:58:09,  3.29s/it]
 41%|████      | 6007/14725 [5:28:47<7:50:53,  3.24s/it]
 41%|████      | 6008/14725 [5:28:50<7:51:15,  3.24s/it]
 41%|████      | 6009/14725 [5:28:53<7:46:01,  3.21s/it]
 41%|████      | 6010/14725 [5:28:57<7:44:48,  3.20s/it]
                                                        

 41%|████      | 6010/14725 [5:28:57<7:44:48,  3.20s/it]
 41%|████      | 6011/14725 [5:29:00<7:46:14,  3.21s/it]
 41%|████      | 6012/14725 [5:29:03<7:45:09,  3.20s/it]
 41%|████      | 6013/14725 [5:29:06<7:45:01,  3.20s/it]
 41%|████      | 6014/14725 [5:29:09<7:46:30,  3.21s/it]
 4
0: {'loss': 0.6939, 'grad_norm': 0.41534630891186225, 'learning_rate': 1e-05, 'epoch': 0.41}
0: 1%|████      | 6015/14725 [5:29:13<7:44:39,  3.20s/it]
 41%|████      | 6016/14725 [5:29:16<7:41:11,  3.18s/it]
 41%|████      | 6017/14725 [5:29:19<7:38:42,  3.16s/it]
 41%|████      | 6018/14725 [5:29:22<7:45:32,  3.21s/it]
 41%|████      | 6019/14725 [5:29:25<7:48:57,  3.23s/it]
 41%|████      | 6020/14725 [5:29:29<7:52:55,  3.26s/it]
                                                        

 41%|████      | 6020/14725 [5:29:29<7:52:55,  3.26s/it]
 41%|████      | 6021/14725 [5:29:32<7:46:47,  3.22s/it]
 41%|████      | 6022/14725 [5:29:35<7:56:27,  3.28s/it]
 41%|████      | 6023/14725 [5:29:39<7:57:02,  3.29s/it]
 41%|████      | 6024/14725 [5:29:42<7:52:49,  3.26s/it]
 41%|████      | 6025/14725 [5:29:45<7:56:27,  3.29s/it]
 41%|████      | 6026/14725 [5:29:48<7:52:11,  3.26s/it]
 41%|████      | 6027/14725 [5:29:52<7:52:57,  3.26s/it]
 41%|████      | 6028/14725 [5:29:55<7:49:29,  3.2
0: {'loss': 0.6923, 'grad_norm': 0.42766756949782636, 'learning_rate': 1e-05, 'epoch': 0.41}
0: {'loss': 0.6995, 'grad_norm': 0.40880132042730205, 'learning_rate': 1e-05, 'epoch': 0.41}
0: 4s/it]
 41%|████      | 6029/14725 [5:29:58<7:47:01,  3.22s/it]
 41%|████      | 6030/14725 [5:30:01<7:57:21,  3.29s/it]
                                                        

 41%|████      | 6030/14725 [5:30:01<7:57:21,  3.29s/it]
 41%|████      | 6031/14725 [5:30:05<8:05:35,  3.35s/it]
 41%|████      | 6032/14725 [5:30:08<8:08:51,  3.37s/it]
 41%|████      | 6033/14725 [5:30:12<8:06:00,  3.35s/it]
 41%|████      | 6034/14725 [5:30:15<8:10:51,  3.39s/it]
 41%|████      | 6035/14725 [5:30:18<8:05:59,  3.36s/it]
 41%|████      | 6036/14725 [5:30:22<8:01:17,  3.32s/it]
 41%|████      | 6037/14725 [5:30:25<7:48:17,  3.23s/it]
 41%|████      | 6038/14725 [5:30:28<7:57:14,  3.30s/it]
 41%|████      | 6039/14725 [5:30:31<7:51:41,  3.26s/it]
 41%|████      | 6040/14725 [5:30:35<7:51:20,  3.26s/it]
                                                        

 41%|████      | 6040/14725 [5:30:35<7:51:20,  3
0: {'loss': 0.7093, 'grad_norm': 0.42191918577784815, 'learning_rate': 1e-05, 'epoch': 0.41}
0: .26s/it]
 41%|████      | 6041/14725 [5:30:38<7:49:25,  3.24s/it]
 41%|████      | 6042/14725 [5:30:41<7:58:37,  3.31s/it]
 41%|████      | 6043/14725 [5:30:45<7:58:23,  3.31s/it]
 41%|████      | 6044/14725 [5:30:48<8:01:30,  3.33s/it]
 41%|████      | 6045/14725 [5:30:51<8:01:26,  3.33s/it]
 41%|████      | 6046/14725 [5:30:54<7:57:49,  3.30s/it]
 41%|████      | 6047/14725 [5:30:58<7:58:09,  3.31s/it]
 41%|████      | 6048/14725 [5:31:01<7:48:52,  3.24s/it]
 41%|████      | 6049/14725 [5:31:04<7:52:35,  3.27s/it]
 41%|████      | 6050/14725 [5:31:08<8:00:52,  3.33s/it]
                                                        

 41%|████      | 6050/14725 [5:31:08<8:00:52,  3.33s/it]
 41%|████      | 6051/14725 [5:31:11<7:57:10,  3.30s/it]
 41%|████      | 6052/14725 [5:31:14<7:52:55,  3.27s/it]
 41%|████      | 6053/14725 [5:31:17<7:49:19,  3.25s/it]
 41%|████      | 6054/14725 [5:31:21<7:
0: {'loss': 0.705, 'grad_norm': 0.4276040373023886, 'learning_rate': 1e-05, 'epoch': 0.41}
0: 53:09,  3.27s/it]
 41%|████      | 6055/14725 [5:31:24<7:48:55,  3.25s/it]
 41%|████      | 6056/14725 [5:31:27<7:46:38,  3.23s/it]
 41%|████      | 6057/14725 [5:31:30<7:49:15,  3.25s/it]
 41%|████      | 6058/14725 [5:31:34<7:53:24,  3.28s/it]
 41%|████      | 6059/14725 [5:31:37<7:55:11,  3.29s/it]
 41%|████      | 6060/14725 [5:31:40<7:50:29,  3.26s/it]
                                                        

 41%|████      | 6060/14725 [5:31:40<7:50:29,  3.26s/it]
 41%|████      | 6061/14725 [5:31:43<7:42:08,  3.20s/it]
 41%|████      | 6062/14725 [5:31:47<7:47:22,  3.24s/it]
 41%|████      | 6063/14725 [5:31:50<7:43:29,  3.21s/it]
 41%|████      | 6064/14725 [5:31:53<7:47:06,  3.24s/it]
 41%|████      | 6065/14725 [5:31:56<7:45:49,  3.23s/it]
 41%|████      | 6066/14725 [5:31:59<7:49:41,  3.25s/it]
 41%|████      | 6067/14725 [5:32:03<7:59:00,  3.32s/it]
 41%|████      | 6068/14725 [5
0: {'loss': 0.7028, 'grad_norm': 0.40905902876587247, 'learning_rate': 1e-05, 'epoch': 0.41}
0: {'loss': 0.7057, 'grad_norm': 0.4107900760913168, 'learning_rate': 1e-05, 'epoch': 0.41}
0: :32:06<8:01:31,  3.34s/it]
 41%|████      | 6069/14725 [5:32:10<7:58:07,  3.31s/it]
 41%|████      | 6070/14725 [5:32:13<8:01:08,  3.34s/it]
                                                        

 41%|████      | 6070/14725 [5:32:13<8:01:08,  3.34s/it]
 41%|████      | 6071/14725 [5:32:16<8:01:04,  3.34s/it]
 41%|████      | 6072/14725 [5:32:20<7:55:11,  3.29s/it]
 41%|████      | 6073/14725 [5:32:23<7:51:20,  3.27s/it]
 41%|████      | 6074/14725 [5:32:26<7:45:19,  3.23s/it]
 41%|████▏     | 6075/14725 [5:32:29<7:47:03,  3.24s/it]
 41%|████▏     | 6076/14725 [5:32:32<7:47:19,  3.24s/it]
 41%|████▏     | 6077/14725 [5:32:36<7:49:42,  3.26s/it]
 41%|████▏     | 6078/14725 [5:32:39<7:44:08,  3.22s/it]
 41%|████▏     | 6079/14725 [5:32:42<7:50:41,  3.27s/it]
 41%|████▏     | 6080/14725 [5:32:45<7:51:35,  3.27s/it]
                                                        

 41%|████▏    
0: {'loss': 0.7243, 'grad_norm': 0.4373544718880877, 'learning_rate': 1e-05, 'epoch': 0.41}
0:  | 6080/14725 [5:32:45<7:51:35,  3.27s/it]
 41%|████▏     | 6081/14725 [5:32:49<7:42:55,  3.21s/it]
 41%|████▏     | 6082/14725 [5:32:52<7:42:18,  3.21s/it]
 41%|████▏     | 6083/14725 [5:32:55<7:40:20,  3.20s/it]
 41%|████▏     | 6084/14725 [5:32:58<7:38:00,  3.18s/it]
 41%|████▏     | 6085/14725 [5:33:01<7:45:01,  3.23s/it]
 41%|████▏     | 6086/14725 [5:33:05<7:44:51,  3.23s/it]
 41%|████▏     | 6087/14725 [5:33:08<7:48:42,  3.26s/it]
 41%|████▏     | 6088/14725 [5:33:11<7:53:22,  3.29s/it]
 41%|████▏     | 6089/14725 [5:33:15<7:50:02,  3.27s/it]
 41%|████▏     | 6090/14725 [5:33:18<7:59:41,  3.33s/it]
                                                        

 41%|████▏     | 6090/14725 [5:33:18<7:59:41,  3.33s/it]
 41%|████▏     | 6091/14725 [5:33:21<7:56:53,  3.31s/it]
 41%|████▏     | 6092/14725 [5:33:24<7:48:03,  3.25s/it]
 41%|████▏     | 6093/14725 [5:33:27<7:41:
0: {'loss': 0.7002, 'grad_norm': 0.4050118124514176, 'learning_rate': 1e-05, 'epoch': 0.41}
0: 51,  3.21s/it]
 41%|████▏     | 6094/14725 [5:33:31<7:35:09,  3.16s/it]
 41%|████▏     | 6095/14725 [5:33:34<7:48:27,  3.26s/it]
 41%|████▏     | 6096/14725 [5:33:37<7:49:41,  3.27s/it]
 41%|████▏     | 6097/14725 [5:33:41<7:51:34,  3.28s/it]
 41%|████▏     | 6098/14725 [5:33:44<7:54:35,  3.30s/it]
 41%|████▏     | 6099/14725 [5:33:47<7:53:52,  3.30s/it]
 41%|████▏     | 6100/14725 [5:33:50<7:51:04,  3.28s/it]
                                                        

 41%|████▏     | 6100/14725 [5:33:50<7:51:04,  3.28s/it]
 41%|████▏     | 6101/14725 [5:33:54<7:48:11,  3.26s/it]
 41%|████▏     | 6102/14725 [5:33:57<7:51:38,  3.28s/it]
 41%|████▏     | 6103/14725 [5:34:00<7:51:06,  3.28s/it]
 41%|████▏     | 6104/14725 [5:34:04<7:51:21,  3.28s/it]
 41%|████▏     | 6105/14725 [5:34:07<7:53:06,  3.29s/it]
 41%|████▏     | 6106/14725 [5:34:10<7:55:06,  3.31s/it]
 41%|██�
0: {'loss': 0.71, 'grad_norm': 0.4282009765402916, 'learning_rate': 1e-05, 'epoch': 0.41}
0: �█▏     | 6107/14725 [5:34:14<7:54:39,  3.30s/it]
 41%|████▏     | 6108/14725 [5:34:17<7:55:08,  3.31s/it]
 41%|████▏     | 6109/14725 [5:34:20<7:49:43,  3.27s/it]
 41%|████▏     | 6110/14725 [5:34:23<7:46:15,  3.25s/it]
                                                        

 41%|████▏     | 6110/14725 [5:34:23<7:46:15,  3.25s/it]
 42%|████▏     | 6111/14725 [5:34:26<7:42:18,  3.22s/it]
 42%|████▏     | 6112/14725 [5:34:30<7:42:52,  3.22s/it]
 42%|████▏     | 6113/14725 [5:34:33<7:43:51,  3.23s/it]
 42%|████▏     | 6114/14725 [5:34:36<7:54:52,  3.31s/it]
 42%|████▏     | 6115/14725 [5:34:39<7:46:20,  3.25s/it]
 42%|████▏     | 6116/14725 [5:34:43<7:44:41,  3.24s/it]
 42%|████▏     | 6117/14725 [5:34:46<7:41:06,  3.21s/it]
 42%|████▏     | 6118/14725 [5:34:49<7:43:57,  3.23s/it]
 42%|████▏     | 6119/14725 [5:34:52<7:38:13,  3.19s/it]
 42%|████▏     | 6120/14725 [5:
0: {'loss': 0.7026, 'grad_norm': 0.42939745277962277, 'learning_rate': 1e-05, 'epoch': 0.42}
0: {'loss': 0.6999, 'grad_norm': 0.4137331171444699, 'learning_rate': 1e-05, 'epoch': 0.42}
0: 34:56<7:40:41,  3.21s/it]
                                                        

 42%|████▏     | 6120/14725 [5:34:56<7:40:41,  3.21s/it]
 42%|████▏     | 6121/14725 [5:34:59<7:37:49,  3.19s/it]
 42%|████▏     | 6122/14725 [5:35:02<7:40:14,  3.21s/it]
 42%|████▏     | 6123/14725 [5:35:05<7:41:13,  3.22s/it]
 42%|████▏     | 6124/14725 [5:35:08<7:39:27,  3.21s/it]
 42%|████▏     | 6125/14725 [5:35:12<7:48:03,  3.27s/it]
 42%|████▏     | 6126/14725 [5:35:15<7:43:59,  3.24s/it]
 42%|████▏     | 6127/14725 [5:35:18<7:45:57,  3.25s/it]
 42%|████▏     | 6128/14725 [5:35:21<7:49:00,  3.27s/it]
 42%|████▏     | 6129/14725 [5:35:25<7:51:28,  3.29s/it]
 42%|████▏     | 6130/14725 [5:35:28<7:47:04,  3.26s/it]
                                                        

 42%|████▏     | 6130/14725 [5:35:28<7:47:04,  3.26s/it]
 42%|████▏     | 6131/14725 [5:35:31<7:45:07,  3.25s/it]
 42%|██
0: {'loss': 0.724, 'grad_norm': 0.4326673239247551, 'learning_rate': 1e-05, 'epoch': 0.42}
0: ██▏     | 6132/14725 [5:35:35<7:46:20,  3.26s/it]
 42%|████▏     | 6133/14725 [5:35:38<7:41:04,  3.22s/it]
 42%|████▏     | 6134/14725 [5:35:41<7:44:07,  3.24s/it]
 42%|████▏     | 6135/14725 [5:35:44<7:46:46,  3.26s/it]
 42%|████▏     | 6136/14725 [5:35:47<7:37:30,  3.20s/it]
 42%|████▏     | 6137/14725 [5:35:50<7:35:45,  3.18s/it]
 42%|████▏     | 6138/14725 [5:35:54<7:37:35,  3.20s/it]
 42%|████▏     | 6139/14725 [5:35:57<7:40:44,  3.22s/it]
 42%|████▏     | 6140/14725 [5:36:00<7:39:50,  3.21s/it]
                                                        

 42%|████▏     | 6140/14725 [5:36:00<7:39:50,  3.21s/it]
 42%|████▏     | 6141/14725 [5:36:03<7:44:53,  3.25s/it]
 42%|████▏     | 6142/14725 [5:36:07<7:45:23,  3.25s/it]
 42%|████▏     | 6143/14725 [5:36:10<7:51:49,  3.30s/it]
 42%|████▏     | 6144/14725 [5:36:13<7:47:32,  3.27s/it]
 42%|████▏     | 6145/14725 [
0: {'loss': 0.6987, 'grad_norm': 0.42106129427902095, 'learning_rate': 1e-05, 'epoch': 0.42}
0: 5:36:17<7:45:20,  3.25s/it]
 42%|████▏     | 6146/14725 [5:36:20<7:34:28,  3.18s/it]
 42%|████▏     | 6147/14725 [5:36:23<7:34:59,  3.18s/it]
 42%|████▏     | 6148/14725 [5:36:26<7:36:10,  3.19s/it]
 42%|████▏     | 6149/14725 [5:36:29<7:39:09,  3.21s/it]
 42%|████▏     | 6150/14725 [5:36:32<7:34:45,  3.18s/it]
                                                        

 42%|████▏     | 6150/14725 [5:36:32<7:34:45,  3.18s/it]
 42%|████▏     | 6151/14725 [5:36:36<7:46:02,  3.26s/it]
 42%|████▏     | 6152/14725 [5:36:39<7:39:18,  3.21s/it]
 42%|████▏     | 6153/14725 [5:36:42<7:34:34,  3.18s/it]
 42%|████▏     | 6154/14725 [5:36:45<7:38:20,  3.21s/it]
 42%|████▏     | 6155/14725 [5:36:49<7:39:46,  3.22s/it]
 42%|████▏     | 6156/14725 [5:36:52<7:37:53,  3.21s/it]
 42%|████▏     | 6157/14725 [5:36:55<7:46:32,  3.27s/it]
 42%|████▏     | 6158/14725 [5:36:59<7:54:11,  3.32s/it]
0: {'loss': 0.7053, 'grad_norm': 0.44435009327833846, 'learning_rate': 1e-05, 'epoch': 0.42}
0: {'loss': 0.7083, 'grad_norm': 0.4314103367948697, 'learning_rate': 1e-05, 'epoch': 0.42}
0:  42%|████▏     | 6159/14725 [5:37:02<7:59:14,  3.36s/it]
 42%|████▏     | 6160/14725 [5:37:05<7:48:22,  3.28s/it]
                                                        

 42%|████▏     | 6160/14725 [5:37:05<7:48:22,  3.28s/it]
 42%|████▏     | 6161/14725 [5:37:08<7:41:17,  3.23s/it]
 42%|████▏     | 6162/14725 [5:37:11<7:42:22,  3.24s/it]
 42%|████▏     | 6163/14725 [5:37:15<7:38:25,  3.21s/it]
 42%|████▏     | 6164/14725 [5:37:18<7:53:48,  3.32s/it]
 42%|████▏     | 6165/14725 [5:37:21<7:48:46,  3.29s/it]
 42%|████▏     | 6166/14725 [5:37:25<7:49:51,  3.29s/it]
 42%|████▏     | 6167/14725 [5:37:28<7:45:20,  3.26s/it]
 42%|████▏     | 6168/14725 [5:37:31<7:54:24,  3.33s/it]
 42%|████▏     | 6169/14725 [5:37:35<7:46:50,  3.27s/it]
 42%|████▏     | 6170/14725 [5:37:38<7:45:20,  3.26s/it]
                                                        

 42%|████▏     | 6170/14725
0: {'loss': 0.7213, 'grad_norm': 0.43056809078695457, 'learning_rate': 1e-05, 'epoch': 0.42}
0:  [5:37:38<7:45:20,  3.26s/it]
 42%|████▏     | 6171/14725 [5:37:41<7:42:45,  3.25s/it]
 42%|████▏     | 6172/14725 [5:37:44<7:37:34,  3.21s/it]
 42%|████▏     | 6173/14725 [5:37:47<7:37:00,  3.21s/it]
 42%|████▏     | 6174/14725 [5:37:50<7:36:56,  3.21s/it]
 42%|████▏     | 6175/14725 [5:37:54<7:46:38,  3.27s/it]
 42%|████▏     | 6176/14725 [5:37:57<7:45:03,  3.26s/it]
 42%|████▏     | 6177/14725 [5:38:00<7:42:37,  3.25s/it]
 42%|████▏     | 6178/14725 [5:38:04<7:39:10,  3.22s/it]
 42%|████▏     | 6179/14725 [5:38:07<7:48:35,  3.29s/it]
 42%|████▏     | 6180/14725 [5:38:10<7:43:09,  3.25s/it]
                                                        

 42%|████▏     | 6180/14725 [5:38:10<7:43:09,  3.25s/it]
 42%|████▏     | 6181/14725 [5:38:14<7:50:06,  3.30s/it]
 42%|████▏     | 6182/14725 [5:38:17<7:55:18,  3.34s/it]
 42%|████▏     | 6183/14725 [5:38:20<7:53:43,  3.33s/it
0: {'loss': 0.7301, 'grad_norm': 0.4161151679446186, 'learning_rate': 1e-05, 'epoch': 0.42}
0: ]
 42%|████▏     | 6184/14725 [5:38:24<7:57:27,  3.35s/it]
 42%|████▏     | 6185/14725 [5:38:27<7:55:30,  3.34s/it]
 42%|████▏     | 6186/14725 [5:38:30<7:54:53,  3.34s/it]
 42%|████▏     | 6187/14725 [5:38:34<7:54:35,  3.34s/it]
 42%|████▏     | 6188/14725 [5:38:37<7:57:44,  3.36s/it]
 42%|████▏     | 6189/14725 [5:38:40<7:53:53,  3.33s/it]
 42%|████▏     | 6190/14725 [5:38:44<7:51:36,  3.32s/it]
                                                        

 42%|████▏     | 6190/14725 [5:38:44<7:51:36,  3.32s/it]
 42%|████▏     | 6191/14725 [5:38:47<7:53:27,  3.33s/it]
 42%|████▏     | 6192/14725 [5:38:50<7:54:47,  3.34s/it]
 42%|████▏     | 6193/14725 [5:38:53<7:43:13,  3.26s/it]
 42%|████▏     | 6194/14725 [5:38:57<7:44:02,  3.26s/it]
 42%|████▏     | 6195/14725 [5:39:00<7:46:40,  3.28s/it]
 42%|████▏     | 6196/14725 [5:39:03<7:41:10,  3.24s/it]
 42%|████▏     |
0: {'loss': 0.7049, 'grad_norm': 0.44171800776631653, 'learning_rate': 1e-05, 'epoch': 0.42}
0:  6197/14725 [5:39:06<7:43:45,  3.26s/it]
 42%|████▏     | 6198/14725 [5:39:10<7:43:10,  3.26s/it]
 42%|████▏     | 6199/14725 [5:39:13<7:38:52,  3.23s/it]
 42%|████▏     | 6200/14725 [5:39:16<7:35:47,  3.21s/it]
                                                        

 42%|████▏     | 6200/14725 [5:39:16<7:35:47,  3.21s/it]
 42%|████▏     | 6201/14725 [5:39:19<7:39:26,  3.23s/it]
 42%|████▏     | 6202/14725 [5:39:22<7:34:46,  3.20s/it]
 42%|████▏     | 6203/14725 [5:39:26<7:39:03,  3.23s/it]
 42%|████▏     | 6204/14725 [5:39:29<7:48:37,  3.30s/it]
 42%|████▏     | 6205/14725 [5:39:33<7:57:46,  3.36s/it]
 42%|████▏     | 6206/14725 [5:39:36<7:59:08,  3.37s/it]
 42%|████▏     | 6207/14725 [5:39:39<7:57:29,  3.36s/it]
 42%|████▏     | 6208/14725 [5:39:43<7:57:14,  3.36s/it]
 42%|████▏     | 6209/14725 [5:39:46<7:56:18,  3.36s/it]
 42%|████▏     | 6210/14725 [5:39:49<7:47:21
0: {'loss': 0.7031, 'grad_norm': 0.4393499227515721, 'learning_rate': 1e-05, 'epoch': 0.42}
0: {'loss': 0.7009, 'grad_norm': 0.4063761088589635, 'learning_rate': 1e-05, 'epoch': 0.42}
0: ,  3.29s/it]
                                                        

 42%|████▏     | 6210/14725 [5:39:49<7:47:21,  3.29s/it]
 42%|████▏     | 6211/14725 [5:39:53<7:52:45,  3.33s/it]
 42%|████▏     | 6212/14725 [5:39:56<7:49:06,  3.31s/it]
 42%|████▏     | 6213/14725 [5:39:59<7:50:46,  3.32s/it]
 42%|████▏     | 6214/14725 [5:40:03<7:54:50,  3.35s/it]
 42%|████▏     | 6215/14725 [5:40:06<7:53:45,  3.34s/it]
 42%|████▏     | 6216/14725 [5:40:09<7:49:07,  3.31s/it]
 42%|████▏     | 6217/14725 [5:40:12<7:39:03,  3.24s/it]
 42%|████▏     | 6218/14725 [5:40:16<7:37:29,  3.23s/it]
 42%|████▏     | 6219/14725 [5:40:19<7:36:33,  3.22s/it]
 42%|████▏     | 6220/14725 [5:40:22<7:33:24,  3.20s/it]
                                                        

 42%|████▏     | 6220/14725 [5:40:22<7:33:24,  3.20s/it]
 42%|████▏     | 6221/14725 [5:40:25<7:42:41,  3.26s/it]
 42%|████▏    
0: {'loss': 0.7088, 'grad_norm': 0.4401263230246938, 'learning_rate': 1e-05, 'epoch': 0.42}
0:  | 6222/14725 [5:40:29<7:42:51,  3.27s/it]
 42%|████▏     | 6223/14725 [5:40:32<7:37:22,  3.23s/it]
 42%|████▏     | 6224/14725 [5:40:35<7:37:29,  3.23s/it]
 42%|████▏     | 6225/14725 [5:40:38<7:34:09,  3.21s/it]
 42%|████▏     | 6226/14725 [5:40:41<7:36:11,  3.22s/it]
 42%|████▏     | 6227/14725 [5:40:45<7:34:28,  3.21s/it]
 42%|████▏     | 6228/14725 [5:40:48<7:31:46,  3.19s/it]
 42%|████▏     | 6229/14725 [5:40:51<7:31:42,  3.19s/it]
 42%|████▏     | 6230/14725 [5:40:54<7:33:04,  3.20s/it]
                                                        

 42%|████▏     | 6230/14725 [5:40:54<7:33:04,  3.20s/it]
 42%|████▏     | 6231/14725 [5:40:57<7:34:00,  3.21s/it]
 42%|████▏     | 6232/14725 [5:41:01<7:33:04,  3.20s/it]
 42%|████▏     | 6233/14725 [5:41:04<7:37:32,  3.23s/it]
 42%|████▏     | 6234/14725 [5:41:07<7:37:49,  3.24s/it]
 42%|████▏     | 6235/14725 [5:41:11<7:48:
0: {'loss': 0.7017, 'grad_norm': 0.4231152334961976, 'learning_rate': 1e-05, 'epoch': 0.42}
0: 28,  3.31s/it]
 42%|████▏     | 6236/14725 [5:41:14<7:48:16,  3.31s/it]
 42%|████▏     | 6237/14725 [5:41:17<7:47:03,  3.30s/it]
 42%|████▏     | 6238/14725 [5:41:20<7:44:51,  3.29s/it]
 42%|████▏     | 6239/14725 [5:41:24<7:36:54,  3.23s/it]
 42%|████▏     | 6240/14725 [5:41:27<7:34:58,  3.22s/it]
                                                        

 42%|████▏     | 6240/14725 [5:41:27<7:34:58,  3.22s/it]
 42%|████▏     | 6241/14725 [5:41:30<7:35:33,  3.22s/it]
 42%|████▏     | 6242/14725 [5:41:33<7:36:52,  3.23s/it]
 42%|████▏     | 6243/14725 [5:41:36<7:35:35,  3.22s/it]
 42%|████▏     | 6244/14725 [5:41:40<7:37:52,  3.24s/it]
 42%|████▏     | 6245/14725 [5:41:43<7:35:18,  3.22s/it]
 42%|████▏     | 6246/14725 [5:41:46<7:33:00,  3.21s/it]
 42%|████▏     | 6247/14725 [5:41:49<7:34:08,  3.21s/it]
 42%|████▏     | 6248/14725 [5:41:53<7:38:33,  3.25s/it]
 42%|██�
0: {'loss': 0.6996, 'grad_norm': 0.42037937097212785, 'learning_rate': 1e-05, 'epoch': 0.42}
0: {'loss': 0.7147, 'grad_norm': 0.4563274871953895, 'learning_rate': 1e-05, 'epoch': 0.43}
0: �█▏     | 6249/14725 [5:41:56<7:42:23,  3.27s/it]
 42%|████▏     | 6250/14725 [5:41:59<7:44:51,  3.29s/it]
                                                        

 42%|████▏     | 6250/14725 [5:41:59<7:44:51,  3.29s/it]
 42%|████▏     | 6251/14725 [5:42:03<7:46:41,  3.30s/it]
 42%|████▏     | 6252/14725 [5:42:06<7:47:45,  3.31s/it]
 42%|████▏     | 6253/14725 [5:42:09<7:41:26,  3.27s/it]
 42%|████▏     | 6254/14725 [5:42:12<7:38:08,  3.25s/it]
 42%|████▏     | 6255/14725 [5:42:16<7:37:16,  3.24s/it]
 42%|████▏     | 6256/14725 [5:42:19<7:38:40,  3.25s/it]
 42%|████▏     | 6257/14725 [5:42:22<7:35:08,  3.22s/it]
 42%|████▏     | 6258/14725 [5:42:25<7:31:29,  3.20s/it]
 43%|████▎     | 6259/14725 [5:42:28<7:31:03,  3.20s/it]
 43%|████▎     | 6260/14725 [5:42:31<7:30:23,  3.19s/it]
                                                        

 43%|████▎     | 6260/14725 [5:42:31<7:3
0: {'loss': 0.7009, 'grad_norm': 0.42466910460606766, 'learning_rate': 1e-05, 'epoch': 0.43}
0: 0:23,  3.19s/it]
 43%|████▎     | 6261/14725 [5:42:35<7:28:03,  3.18s/it]
 43%|████▎     | 6262/14725 [5:42:38<7:34:15,  3.22s/it]
 43%|████▎     | 6263/14725 [5:42:41<7:29:50,  3.19s/it]
 43%|████▎     | 6264/14725 [5:42:44<7:32:58,  3.21s/it]
 43%|████▎     | 6265/14725 [5:42:47<7:29:57,  3.19s/it]
 43%|████▎     | 6266/14725 [5:42:51<7:30:37,  3.20s/it]
 43%|████▎     | 6267/14725 [5:42:54<7:27:22,  3.17s/it]
 43%|████▎     | 6268/14725 [5:42:57<7:27:22,  3.17s/it]
 43%|████▎     | 6269/14725 [5:43:00<7:21:59,  3.14s/it]
 43%|████▎     | 6270/14725 [5:43:03<7:31:55,  3.21s/it]
                                                        

 43%|████▎     | 6270/14725 [5:43:03<7:31:55,  3.21s/it]
 43%|████▎     | 6271/14725 [5:43:07<7:35:24,  3.23s/it]
 43%|████▎     | 6272/14725 [5:43:10<7:42:44,  3.28s/it]
 43%|████▎     | 6273/14725 [5:43:13<7:39:57,  3.27s/it]
 43%|██
0: {'loss': 0.7026, 'grad_norm': 0.4261577685578473, 'learning_rate': 1e-05, 'epoch': 0.43}
0: ██▎     | 6274/14725 [5:43:16<7:32:51,  3.22s/it]
 43%|████▎     | 6275/14725 [5:43:20<7:37:34,  3.25s/it]
 43%|████▎     | 6276/14725 [5:43:23<7:32:58,  3.22s/it]
 43%|████▎     | 6277/14725 [5:43:26<7:36:30,  3.24s/it]
 43%|████▎     | 6278/14725 [5:43:30<7:43:51,  3.29s/it]
 43%|████▎     | 6279/14725 [5:43:33<7:46:47,  3.32s/it]
 43%|████▎     | 6280/14725 [5:43:36<7:41:17,  3.28s/it]
                                                        

 43%|████▎     | 6280/14725 [5:43:36<7:41:17,  3.28s/it]
 43%|████▎     | 6281/14725 [5:43:39<7:38:04,  3.25s/it]
 43%|████▎     | 6282/14725 [5:43:43<7:44:53,  3.30s/it]
 43%|████▎     | 6283/14725 [5:43:46<7:39:27,  3.27s/it]
 43%|████▎     | 6284/14725 [5:43:49<7:41:59,  3.28s/it]
 43%|████▎     | 6285/14725 [5:43:53<7:44:11,  3.30s/it]
 43%|████▎     | 6286/14725 [5:43:56<7:47:36,  3.32s/it]
 43%|████▎     | 6287/14725 [
0: {'loss': 0.7158, 'grad_norm': 0.4017185633053606, 'learning_rate': 1e-05, 'epoch': 0.43}
0: 5:43:59<7:49:35,  3.34s/it]
 43%|████▎     | 6288/14725 [5:44:03<7:51:41,  3.35s/it]
 43%|████▎     | 6289/14725 [5:44:06<7:47:41,  3.33s/it]
 43%|████▎     | 6290/14725 [5:44:09<7:36:45,  3.25s/it]
                                                        

 43%|████▎     | 6290/14725 [5:44:09<7:36:45,  3.25s/it]
 43%|████▎     | 6291/14725 [5:44:12<7:35:12,  3.24s/it]
 43%|████▎     | 6292/14725 [5:44:16<7:40:33,  3.28s/it]
 43%|████▎     | 6293/14725 [5:44:19<7:40:24,  3.28s/it]
 43%|████▎     | 6294/14725 [5:44:22<7:48:53,  3.34s/it]
 43%|████▎     | 6295/14725 [5:44:26<7:47:32,  3.33s/it]
 43%|████▎     | 6296/14725 [5:44:29<7:38:12,  3.26s/it]
 43%|████▎     | 6297/14725 [5:44:32<7:31:58,  3.22s/it]
 43%|████▎     | 6298/14725 [5:44:35<7:33:25,  3.23s/it]
 43%|████▎     | 6299/14725 [5:44:38<7:32:51,  3.22s/it]
 43%|████▎     | 6300/14725 [5:44:42<7:36:32,  3.25s/it]
0: {'loss': 0.7038, 'grad_norm': 0.4174181836216847, 'learning_rate': 1e-05, 'epoch': 0.43}
0: {'loss': 0.6966, 'grad_norm': 0.4621969619098391, 'learning_rate': 1e-05, 'epoch': 0.43}
0:                                                         

 43%|████▎     | 6300/14725 [5:44:42<7:36:32,  3.25s/it]
 43%|████▎     | 6301/14725 [5:44:45<7:24:59,  3.17s/it]
 43%|████▎     | 6302/14725 [5:44:48<7:34:24,  3.24s/it]
 43%|████▎     | 6303/14725 [5:44:51<7:28:29,  3.20s/it]
 43%|████▎     | 6304/14725 [5:44:55<7:36:15,  3.25s/it]
 43%|████▎     | 6305/14725 [5:44:58<7:46:52,  3.33s/it]
 43%|████▎     | 6306/14725 [5:45:01<7:43:12,  3.30s/it]
 43%|████▎     | 6307/14725 [5:45:05<7:40:14,  3.28s/it]
 43%|████▎     | 6308/14725 [5:45:08<7:32:20,  3.22s/it]
 43%|████▎     | 6309/14725 [5:45:11<7:27:23,  3.19s/it]
 43%|████▎     | 6310/14725 [5:45:14<7:35:30,  3.25s/it]
                                                        

 43%|████▎     | 6310/14725 [5:45:14<7:35:30,  3.25s/it]
 43%|████▎     | 6311/14725 [5:45:18<7:41:41,  3.29s/it]
 43%|████▎     | 6312/14725
0: {'loss': 0.6847, 'grad_norm': 0.44032777175920607, 'learning_rate': 1e-05, 'epoch': 0.43}
0:  [5:45:21<7:39:49,  3.28s/it]
 43%|████▎     | 6313/14725 [5:45:24<7:35:26,  3.25s/it]
 43%|████▎     | 6314/14725 [5:45:27<7:42:57,  3.30s/it]
 43%|████▎     | 6315/14725 [5:45:31<7:48:17,  3.34s/it]
 43%|████▎     | 6316/14725 [5:45:34<7:48:42,  3.34s/it]
 43%|████▎     | 6317/14725 [5:45:38<7:48:26,  3.34s/it]
 43%|████▎     | 6318/14725 [5:45:41<7:50:21,  3.36s/it]
 43%|████▎     | 6319/14725 [5:45:44<7:40:09,  3.28s/it]
 43%|████▎     | 6320/14725 [5:45:47<7:39:46,  3.28s/it]
                                                        

 43%|████▎     | 6320/14725 [5:45:47<7:39:46,  3.28s/it]
 43%|████▎     | 6321/14725 [5:45:51<7:44:45,  3.32s/it]
 43%|████▎     | 6322/14725 [5:45:54<7:43:39,  3.31s/it]
 43%|████▎     | 6323/14725 [5:45:57<7:47:10,  3.34s/it]
 43%|████▎     | 6324/14725 [5:46:01<7:46:57,  3.34s/it]
 43%|████▎     | 6325/14725 [5:46:04<7:45:37,  3.33s/it
0: {'loss': 0.6969, 'grad_norm': 0.4296883724469455, 'learning_rate': 1e-05, 'epoch': 0.43}
0: ]
 43%|████▎     | 6326/14725 [5:46:07<7:40:38,  3.29s/it]
 43%|████▎     | 6327/14725 [5:46:10<7:33:43,  3.24s/it]
 43%|████▎     | 6328/14725 [5:46:13<7:27:04,  3.19s/it]
 43%|████▎     | 6329/14725 [5:46:17<7:33:57,  3.24s/it]
 43%|████▎     | 6330/14725 [5:46:20<7:27:47,  3.20s/it]
                                                        

 43%|████▎     | 6330/14725 [5:46:20<7:27:47,  3.20s/it]
 43%|████▎     | 6331/14725 [5:46:23<7:29:46,  3.21s/it]
 43%|████▎     | 6332/14725 [5:46:26<7:33:16,  3.24s/it]
 43%|████▎     | 6333/14725 [5:46:30<7:36:00,  3.26s/it]
 43%|████▎     | 6334/14725 [5:46:33<7:34:14,  3.25s/it]
 43%|████▎     | 6335/14725 [5:46:36<7:34:36,  3.25s/it]
 43%|████▎     | 6336/14725 [5:46:40<7:35:09,  3.26s/it]
 43%|████▎     | 6337/14725 [5:46:43<7:34:13,  3.25s/it]
 43%|████▎     | 6338/14725 [5:46:46<7:38:43,  3.28s/it]
 43%|████▎     |
0: {'loss': 0.7179, 'grad_norm': 0.41177150538631657, 'learning_rate': 1e-05, 'epoch': 0.43}
0: {'loss': 0.7007, 'grad_norm': 0.438397022918047, 'learning_rate': 1e-05, 'epoch': 0.43}
0:  6339/14725 [5:46:49<7:32:26,  3.24s/it]
 43%|████▎     | 6340/14725 [5:46:52<7:29:48,  3.22s/it]
                                                        

 43%|████▎     | 6340/14725 [5:46:52<7:29:48,  3.22s/it]
 43%|████▎     | 6341/14725 [5:46:56<7:36:45,  3.27s/it]
 43%|████▎     | 6342/14725 [5:46:59<7:41:47,  3.31s/it]
 43%|████▎     | 6343/14725 [5:47:02<7:39:40,  3.29s/it]
 43%|████▎     | 6344/14725 [5:47:06<7:37:00,  3.27s/it]
 43%|████▎     | 6345/14725 [5:47:09<7:41:39,  3.31s/it]
 43%|████▎     | 6346/14725 [5:47:12<7:42:44,  3.31s/it]
 43%|████▎     | 6347/14725 [5:47:16<7:45:07,  3.33s/it]
 43%|████▎     | 6348/14725 [5:47:19<7:42:10,  3.31s/it]
 43%|████▎     | 6349/14725 [5:47:22<7:35:26,  3.26s/it]
 43%|████▎     | 6350/14725 [5:47:25<7:29:01,  3.22s/it]
                                                        

 43%|████▎     | 6350/14725 [5:47:25<7:29:01,  3.22s/
0: {'loss': 0.7034, 'grad_norm': 0.4664204870736511, 'learning_rate': 1e-05, 'epoch': 0.43}
0: it]
 43%|████▎     | 6351/14725 [5:47:28<7:23:03,  3.17s/it]
 43%|████▎     | 6352/14725 [5:47:32<7:29:24,  3.22s/it]
 43%|████▎     | 6353/14725 [5:47:35<7:35:12,  3.26s/it]
 43%|████▎     | 6354/14725 [5:47:38<7:34:38,  3.26s/it]
 43%|████▎     | 6355/14725 [5:47:42<7:38:50,  3.29s/it]
 43%|████▎     | 6356/14725 [5:47:45<7:38:27,  3.29s/it]
 43%|████▎     | 6357/14725 [5:47:48<7:38:21,  3.29s/it]
 43%|████▎     | 6358/14725 [5:47:51<7:31:46,  3.24s/it]
 43%|████▎     | 6359/14725 [5:47:55<7:29:01,  3.22s/it]
 43%|████▎     | 6360/14725 [5:47:58<7:32:25,  3.25s/it]
                                                        

 43%|████▎     | 6360/14725 [5:47:58<7:32:25,  3.25s/it]
 43%|████▎     | 6361/14725 [5:48:01<7:35:22,  3.27s/it]
 43%|████▎     | 6362/14725 [5:48:04<7:37:12,  3.28s/it]
 43%|████▎     | 6363/14725 [5:48:08<7:40:16,  3.30s/it]
 43%|████▎    
0: {'loss': 0.7114, 'grad_norm': 0.43336941629047726, 'learning_rate': 1e-05, 'epoch': 0.43}
0:  | 6364/14725 [5:48:11<7:38:29,  3.29s/it]
 43%|████▎     | 6365/14725 [5:48:15<7:48:25,  3.36s/it]
 43%|████▎     | 6366/14725 [5:48:18<7:57:16,  3.43s/it]
 43%|████▎     | 6367/14725 [5:48:21<7:48:07,  3.36s/it]
 43%|████▎     | 6368/14725 [5:48:25<7:42:39,  3.32s/it]
 43%|████▎     | 6369/14725 [5:48:28<7:30:08,  3.23s/it]
 43%|████▎     | 6370/14725 [5:48:31<7:34:13,  3.26s/it]
                                                        

 43%|████▎     | 6370/14725 [5:48:31<7:34:13,  3.26s/it]
 43%|████▎     | 6371/14725 [5:48:35<7:45:58,  3.35s/it]
 43%|████▎     | 6372/14725 [5:48:38<7:50:11,  3.38s/it]
 43%|████▎     | 6373/14725 [5:48:41<7:38:26,  3.29s/it]
 43%|████▎     | 6374/14725 [5:48:45<7:48:15,  3.36s/it]
 43%|████▎     | 6375/14725 [5:48:48<7:41:57,  3.32s/it]
 43%|████▎     | 6376/14725 [5:48:51<7:43:17,  3.33s/it]
 43%|████▎     | 6377/14725 [5:48:54<7:34:
0: {'loss': 0.6976, 'grad_norm': 0.43904673196791383, 'learning_rate': 1e-05, 'epoch': 0.43}
0: 54,  3.27s/it]
 43%|████▎     | 6378/14725 [5:48:57<7:25:03,  3.20s/it]
 43%|████▎     | 6379/14725 [5:49:00<7:22:54,  3.18s/it]
 43%|████▎     | 6380/14725 [5:49:04<7:25:36,  3.20s/it]
                                                        

 43%|████▎     | 6380/14725 [5:49:04<7:25:36,  3.20s/it]
 43%|████▎     | 6381/14725 [5:49:07<7:31:48,  3.25s/it]
 43%|████▎     | 6382/14725 [5:49:10<7:31:03,  3.24s/it]
 43%|████▎     | 6383/14725 [5:49:14<7:29:29,  3.23s/it]
 43%|████▎     | 6384/14725 [5:49:17<7:31:38,  3.25s/it]
 43%|████▎     | 6385/14725 [5:49:20<7:47:03,  3.36s/it]
 43%|████▎     | 6386/14725 [5:49:24<7:49:27,  3.38s/it]
 43%|████▎     | 6387/14725 [5:49:27<7:46:11,  3.35s/it]
 43%|████▎     | 6388/14725 [5:49:30<7:39:32,  3.31s/it]
 43%|████▎     | 6389/14725 [5:49:34<7:36:02,  3.28s/it]
 43%|████▎     | 6390/14725 [5:49:37<7:37:39,  3.29s/it]
             
0: {'loss': 0.7029, 'grad_norm': 0.39745905653806296, 'learning_rate': 1e-05, 'epoch': 0.43}
0: {'loss': 0.7048, 'grad_norm': 0.4087898431732981, 'learning_rate': 1e-05, 'epoch': 0.43}
0:                                            

 43%|████▎     | 6390/14725 [5:49:37<7:37:39,  3.29s/it]
 43%|████▎     | 6391/14725 [5:49:40<7:46:05,  3.36s/it]
 43%|████▎     | 6392/14725 [5:49:44<7:42:35,  3.33s/it]
 43%|████▎     | 6393/14725 [5:49:47<7:36:14,  3.29s/it]
 43%|████▎     | 6394/14725 [5:49:50<7:39:34,  3.31s/it]
 43%|████▎     | 6395/14725 [5:49:54<7:49:52,  3.38s/it]
 43%|████▎     | 6396/14725 [5:49:57<7:44:28,  3.35s/it]
 43%|████▎     | 6397/14725 [5:50:00<7:42:49,  3.33s/it]
 43%|████▎     | 6398/14725 [5:50:04<7:41:18,  3.32s/it]
 43%|████▎     | 6399/14725 [5:50:07<7:39:48,  3.31s/it]
 43%|████▎     | 6400/14725 [5:50:10<7:42:21,  3.33s/it]
                                                        

 43%|████▎     | 6400/14725 [5:50:10<7:42:21,  3.33s/it]
 43%|████▎     | 6401/14725 [5:50:14<7:41:38,  3.33s/it]
 43%|████▎     | 6402/14725 [5:50:17<7:4
0: {'loss': 0.7031, 'grad_norm': 0.4194672631715894, 'learning_rate': 1e-05, 'epoch': 0.44}
0: 8:22,  3.38s/it]
 43%|████▎     | 6403/14725 [5:50:20<7:46:12,  3.36s/it]
 43%|████▎     | 6404/14725 [5:50:23<7:30:46,  3.25s/it]
 43%|████▎     | 6405/14725 [5:50:27<7:40:11,  3.32s/it]
 44%|████▎     | 6406/14725 [5:50:30<7:26:06,  3.22s/it]
 44%|████▎     | 6407/14725 [5:50:33<7:28:11,  3.23s/it]
 44%|████▎     | 6408/14725 [5:50:37<7:36:25,  3.29s/it]
 44%|████▎     | 6409/14725 [5:50:40<7:39:42,  3.32s/it]
 44%|████▎     | 6410/14725 [5:50:43<7:35:49,  3.29s/it]
                                                        

 44%|████▎     | 6410/14725 [5:50:43<7:35:49,  3.29s/it]
 44%|████▎     | 6411/14725 [5:50:46<7:32:55,  3.27s/it]
 44%|████▎     | 6412/14725 [5:50:50<7:30:23,  3.25s/it]
 44%|████▎     | 6413/14725 [5:50:53<7:31:15,  3.26s/it]
 44%|████▎     | 6414/14725 [5:50:56<7:36:00,  3.29s/it]
 44%|████▎     | 6415/14725 [5:51:00<7:35:45,  3.29s/it]
 44%|██
0: {'loss': 0.6932, 'grad_norm': 0.41620179896225545, 'learning_rate': 1e-05, 'epoch': 0.44}
0: ██▎     | 6416/14725 [5:51:03<7:33:19,  3.27s/it]
 44%|████▎     | 6417/14725 [5:51:06<7:37:39,  3.31s/it]
 44%|████▎     | 6418/14725 [5:51:09<7:38:33,  3.31s/it]
 44%|████▎     | 6419/14725 [5:51:13<7:37:35,  3.31s/it]
 44%|████▎     | 6420/14725 [5:51:16<7:36:22,  3.30s/it]
                                                        

 44%|████▎     | 6420/14725 [5:51:16<7:36:22,  3.30s/it]
 44%|████▎     | 6421/14725 [5:51:19<7:34:49,  3.29s/it]
 44%|████▎     | 6422/14725 [5:51:23<7:38:12,  3.31s/it]
 44%|████▎     | 6423/14725 [5:51:26<7:29:46,  3.25s/it]
 44%|████▎     | 6424/14725 [5:51:29<7:34:19,  3.28s/it]
 44%|████▎     | 6425/14725 [5:51:32<7:23:04,  3.20s/it]
 44%|████▎     | 6426/14725 [5:51:35<7:18:33,  3.17s/it]
 44%|████▎     | 6427/14725 [5:51:38<7:20:43,  3.19s/it]
 44%|████▎     | 6428/14725 [5:51:42<7:21:56,  3.20s/it]
 44%|████▎     | 6429/14725 [
0: {'loss': 0.6946, 'grad_norm': 0.40397466144191546, 'learning_rate': 1e-05, 'epoch': 0.44}
0: {'loss': 0.6903, 'grad_norm': 0.4387756308298488, 'learning_rate': 1e-05, 'epoch': 0.44}
0: 5:51:45<7:21:47,  3.20s/it]
 44%|████▎     | 6430/14725 [5:51:48<7:22:13,  3.20s/it]
                                                        

 44%|████▎     | 6430/14725 [5:51:48<7:22:13,  3.20s/it]
 44%|████▎     | 6431/14725 [5:51:51<7:24:40,  3.22s/it]
 44%|████▎     | 6432/14725 [5:51:54<7:18:45,  3.17s/it]
 44%|████▎     | 6433/14725 [5:51:58<7:22:58,  3.21s/it]
 44%|████▎     | 6434/14725 [5:52:01<7:27:24,  3.24s/it]
 44%|████▎     | 6435/14725 [5:52:04<7:25:43,  3.23s/it]
 44%|████▎     | 6436/14725 [5:52:07<7:17:37,  3.17s/it]
 44%|████▎     | 6437/14725 [5:52:11<7:24:04,  3.21s/it]
 44%|████▎     | 6438/14725 [5:52:14<7:27:06,  3.24s/it]
 44%|████▎     | 6439/14725 [5:52:17<7:21:14,  3.20s/it]
 44%|████▎     | 6440/14725 [5:52:20<7:16:54,  3.16s/it]
                                                        

 44%|████▎     | 6440/14725 [5:52:20<7:16:54,  3.16s/it]
 44%|█�
0: {'loss': 0.7042, 'grad_norm': 0.4060031457132779, 'learning_rate': 1e-05, 'epoch': 0.44}
0: ��██▎     | 6441/14725 [5:52:23<7:14:41,  3.15s/it]
 44%|████▎     | 6442/14725 [5:52:26<7:15:15,  3.15s/it]
 44%|████▍     | 6443/14725 [5:52:30<7:29:22,  3.26s/it]
 44%|████▍     | 6444/14725 [5:52:33<7:28:55,  3.25s/it]
 44%|████▍     | 6445/14725 [5:52:36<7:28:54,  3.25s/it]
 44%|████▍     | 6446/14725 [5:52:39<7:23:58,  3.22s/it]
 44%|████▍     | 6447/14725 [5:52:43<7:22:06,  3.20s/it]
 44%|████▍     | 6448/14725 [5:52:46<7:23:57,  3.22s/it]
 44%|████▍     | 6449/14725 [5:52:49<7:17:57,  3.18s/it]
 44%|████▍     | 6450/14725 [5:52:52<7:22:07,  3.21s/it]
                                                        

 44%|████▍     | 6450/14725 [5:52:52<7:22:07,  3.21s/it]
 44%|████▍     | 6451/14725 [5:52:55<7:19:31,  3.19s/it]
 44%|████▍     | 6452/14725 [5:52:59<7:28:32,  3.25s/it]
 44%|████▍     | 6453/14725 [5:53:02<7:26:03,  3.24s/it]
 44%|████▍     | 6454/14725
0: {'loss': 0.7015, 'grad_norm': 0.4607790473094336, 'learning_rate': 1e-05, 'epoch': 0.44}
0:  [5:53:05<7:34:10,  3.29s/it]
 44%|████▍     | 6455/14725 [5:53:09<7:33:27,  3.29s/it]
 44%|████▍     | 6456/14725 [5:53:12<7:28:30,  3.25s/it]
 44%|████▍     | 6457/14725 [5:53:15<7:34:27,  3.30s/it]
 44%|████▍     | 6458/14725 [5:53:18<7:31:19,  3.28s/it]
 44%|████▍     | 6459/14725 [5:53:22<7:31:56,  3.28s/it]
 44%|████▍     | 6460/14725 [5:53:25<7:28:35,  3.26s/it]
                                                        

 44%|████▍     | 6460/14725 [5:53:25<7:28:35,  3.26s/it]
 44%|████▍     | 6461/14725 [5:53:28<7:22:11,  3.21s/it]
 44%|████▍     | 6462/14725 [5:53:31<7:22:05,  3.21s/it]
 44%|████▍     | 6463/14725 [5:53:35<7:29:23,  3.26s/it]
 44%|████▍     | 6464/14725 [5:53:38<7:33:26,  3.29s/it]
 44%|████▍     | 6465/14725 [5:53:41<7:30:34,  3.27s/it]
 44%|████▍     | 6466/14725 [5:53:45<7:32:08,  3.28s/it]
 44%|████▍     | 6467/14725 [5:53:48<7:25:58,  3.24s/it
0: {'loss': 0.7108, 'grad_norm': 0.49289632220034785, 'learning_rate': 1e-05, 'epoch': 0.44}
0: ]
 44%|████▍     | 6468/14725 [5:53:51<7:26:48,  3.25s/it]
 44%|████▍     | 6469/14725 [5:53:54<7:24:32,  3.23s/it]
 44%|████▍     | 6470/14725 [5:53:58<7:32:58,  3.29s/it]
                                                        

 44%|████▍     | 6470/14725 [5:53:58<7:32:58,  3.29s/it]
 44%|████▍     | 6471/14725 [5:54:01<7:26:01,  3.24s/it]
 44%|████▍     | 6472/14725 [5:54:04<7:23:58,  3.23s/it]
 44%|████▍     | 6473/14725 [5:54:07<7:35:35,  3.31s/it]
 44%|████▍     | 6474/14725 [5:54:11<7:37:53,  3.33s/it]
 44%|████▍     | 6475/14725 [5:54:14<7:39:26,  3.34s/it]
 44%|████▍     | 6476/14725 [5:54:18<7:40:28,  3.35s/it]
 44%|████▍     | 6477/14725 [5:54:21<8:03:35,  3.52s/it]
 44%|████▍     | 6478/14725 [5:54:25<7:47:48,  3.40s/it]
 44%|████▍     | 6479/14725 [5:54:28<7:56:23,  3.47s/it]
 44%|████▍     | 6480/14725 [5:54:32<7:51:14,  3.43s/it]
                          
0: {'loss': 0.7121, 'grad_norm': 0.43905353488648097, 'learning_rate': 1e-05, 'epoch': 0.44}
0: {'loss': 0.7014, 'grad_norm': 0.4259442348042166, 'learning_rate': 1e-05, 'epoch': 0.44}
0:                               

 44%|████▍     | 6480/14725 [5:54:32<7:51:14,  3.43s/it]
 44%|████▍     | 6481/14725 [5:54:35<7:37:58,  3.33s/it]
 44%|████▍     | 6482/14725 [5:54:38<7:41:03,  3.36s/it]
 44%|████▍     | 6483/14725 [5:54:41<7:29:38,  3.27s/it]
 44%|████▍     | 6484/14725 [5:54:44<7:25:27,  3.24s/it]
 44%|████▍     | 6485/14725 [5:54:48<7:26:31,  3.25s/it]
 44%|████▍     | 6486/14725 [5:54:51<7:37:16,  3.33s/it]
 44%|████▍     | 6487/14725 [5:54:54<7:37:50,  3.33s/it]
 44%|████▍     | 6488/14725 [5:54:58<7:28:42,  3.27s/it]
 44%|████▍     | 6489/14725 [5:55:01<7:29:38,  3.28s/it]
 44%|████▍     | 6490/14725 [5:55:04<7:33:08,  3.30s/it]
                                                        

 44%|████▍     | 6490/14725 [5:55:04<7:33:08,  3.30s/it]
 44%|████▍     | 6491/14725 [5:55:08<7:36:09,  3.32s/it]
 44%|████▍     | 6492/14725 [5:55:11<7:39:52,  3.35s/
0: {'loss': 0.7037, 'grad_norm': 0.40462314682771794, 'learning_rate': 1e-05, 'epoch': 0.44}
0: it]
 44%|████▍     | 6493/14725 [5:55:14<7:42:02,  3.37s/it]
 44%|████▍     | 6494/14725 [5:55:18<7:46:49,  3.40s/it]
 44%|████▍     | 6495/14725 [5:55:21<7:32:50,  3.30s/it]
 44%|████▍     | 6496/14725 [5:55:24<7:27:43,  3.26s/it]
 44%|████▍     | 6497/14725 [5:55:27<7:19:27,  3.20s/it]
 44%|████▍     | 6498/14725 [5:55:31<7:29:04,  3.28s/it]
 44%|████▍     | 6499/14725 [5:55:34<7:24:25,  3.24s/it]
 44%|████▍     | 6500/14725 [5:55:37<7:21:11,  3.22s/it]
                                                        

 44%|████▍     | 6500/14725 [5:55:37<7:21:11,  3.22s/it]
 44%|████▍     | 6501/14725 [5:55:40<7:26:26,  3.26s/it]
 44%|████▍     | 6502/14725 [5:55:44<7:25:44,  3.25s/it]
 44%|████▍     | 6503/14725 [5:55:47<7:28:09,  3.27s/it]
 44%|████▍     | 6504/14725 [5:55:50<7:33:35,  3.31s/it]
 44%|████▍     | 6505/14725 [5:55:54<7:45:28,  3.40s/it]
 44%|████▍    
0: {'loss': 0.711, 'grad_norm': 0.41587802498749876, 'learning_rate': 1e-05, 'epoch': 0.44}
0:  | 6506/14725 [5:55:57<7:36:15,  3.33s/it]
 44%|████▍     | 6507/14725 [5:56:00<7:30:11,  3.29s/it]
 44%|████▍     | 6508/14725 [5:56:04<7:31:53,  3.30s/it]
 44%|████▍     | 6509/14725 [5:56:07<7:27:29,  3.27s/it]
 44%|████▍     | 6510/14725 [5:56:10<7:24:59,  3.25s/it]
                                                        

 44%|████▍     | 6510/14725 [5:56:10<7:24:59,  3.25s/it]
 44%|████▍     | 6511/14725 [5:56:13<7:20:32,  3.22s/it]
 44%|████▍     | 6512/14725 [5:56:16<7:23:06,  3.24s/it]
 44%|████▍     | 6513/14725 [5:56:20<7:37:06,  3.34s/it]
 44%|████▍     | 6514/14725 [5:56:23<7:36:28,  3.34s/it]
 44%|████▍     | 6515/14725 [5:56:26<7:28:53,  3.28s/it]
 44%|████▍     | 6516/14725 [5:56:30<7:25:01,  3.25s/it]
 44%|████▍     | 6517/14725 [5:56:33<7:41:38,  3.37s/it]
 44%|████▍     | 6518/14725 [5:56:36<7:31:29,  3.30s/it]
 44%|████▍     | 6519/14725 [5:56:40<7:35:
0: {'loss': 0.7082, 'grad_norm': 0.41927257275924956, 'learning_rate': 1e-05, 'epoch': 0.44}
0: {'loss': 0.6996, 'grad_norm': 0.416671199706575, 'learning_rate': 1e-05, 'epoch': 0.44}
0: 51,  3.33s/it]
 44%|████▍     | 6520/14725 [5:56:43<7:34:02,  3.32s/it]
                                                        

 44%|████▍     | 6520/14725 [5:56:43<7:34:02,  3.32s/it]
 44%|████▍     | 6521/14725 [5:56:46<7:30:50,  3.30s/it]
 44%|████▍     | 6522/14725 [5:56:50<7:27:26,  3.27s/it]
 44%|████▍     | 6523/14725 [5:56:53<7:29:25,  3.29s/it]
 44%|████▍     | 6524/14725 [5:56:56<7:24:06,  3.25s/it]
 44%|████▍     | 6525/14725 [5:56:59<7:28:24,  3.28s/it]
 44%|████▍     | 6526/14725 [5:57:03<7:25:25,  3.26s/it]
 44%|████▍     | 6527/14725 [5:57:06<7:25:17,  3.26s/it]
 44%|████▍     | 6528/14725 [5:57:09<7:21:32,  3.23s/it]
 44%|████▍     | 6529/14725 [5:57:13<7:30:53,  3.30s/it]
 44%|████▍     | 6530/14725 [5:57:16<7:23:40,  3.25s/it]
                                                        

 44%|████▍     | 6530/14725 [5:57:16<7:23:40,  3.25s/it]
 44%|████▍  
0: {'loss': 0.7072, 'grad_norm': 0.4051010466837028, 'learning_rate': 1e-05, 'epoch': 0.44}
0:    | 6531/14725 [5:57:19<7:21:36,  3.23s/it]
 44%|████▍     | 6532/14725 [5:57:22<7:28:50,  3.29s/it]
 44%|████▍     | 6533/14725 [5:57:25<7:22:07,  3.24s/it]
 44%|████▍     | 6534/14725 [5:57:29<7:24:09,  3.25s/it]
 44%|████▍     | 6535/14725 [5:57:32<7:26:06,  3.27s/it]
 44%|████▍     | 6536/14725 [5:57:35<7:25:24,  3.26s/it]
 44%|████▍     | 6537/14725 [5:57:38<7:24:16,  3.26s/it]
 44%|████▍     | 6538/14725 [5:57:42<7:22:11,  3.24s/it]
 44%|████▍     | 6539/14725 [5:57:45<7:15:59,  3.20s/it]
 44%|████▍     | 6540/14725 [5:57:48<7:16:49,  3.20s/it]
                                                        

 44%|████▍     | 6540/14725 [5:57:48<7:16:49,  3.20s/it]
 44%|████▍     | 6541/14725 [5:57:51<7:14:07,  3.18s/it]
 44%|████▍     | 6542/14725 [5:57:54<7:16:37,  3.20s/it]
 44%|████▍     | 6543/14725 [5:57:58<7:18:31,  3.22s/it]
 44%|████▍     | 6544/14725 [5:58:01<7:1
0: {'loss': 0.7049, 'grad_norm': 0.4354805521892365, 'learning_rate': 1e-05, 'epoch': 0.44}
0: 8:53,  3.22s/it]
 44%|████▍     | 6545/14725 [5:58:04<7:21:48,  3.24s/it]
 44%|████▍     | 6546/14725 [5:58:08<7:32:04,  3.32s/it]
 44%|████▍     | 6547/14725 [5:58:11<7:28:50,  3.29s/it]
 44%|████▍     | 6548/14725 [5:58:14<7:25:25,  3.27s/it]
 44%|████▍     | 6549/14725 [5:58:18<7:32:33,  3.32s/it]
 44%|████▍     | 6550/14725 [5:58:21<7:39:01,  3.37s/it]
                                                        

 44%|████▍     | 6550/14725 [5:58:21<7:39:01,  3.37s/it]
 44%|████▍     | 6551/14725 [5:58:25<7:51:26,  3.46s/it]
 44%|████▍     | 6552/14725 [5:58:28<7:37:18,  3.36s/it]
 45%|████▍     | 6553/14725 [5:58:31<7:42:41,  3.40s/it]
 45%|████▍     | 6554/14725 [5:58:34<7:33:50,  3.33s/it]
 45%|████▍     | 6555/14725 [5:58:38<7:25:54,  3.27s/it]
 45%|████▍     | 6556/14725 [5:58:41<7:33:36,  3.33s/it]
 45%|████▍     | 6557/14725 [5:58:44<7:20:45,  3.24s/it]
 45%|██
0: {'loss': 0.695, 'grad_norm': 0.4133227404882546, 'learning_rate': 1e-05, 'epoch': 0.45}
0: ██▍     | 6558/14725 [5:58:47<7:17:20,  3.21s/it]
 45%|████▍     | 6559/14725 [5:58:50<7:16:57,  3.21s/it]
 45%|████▍     | 6560/14725 [5:58:54<7:17:23,  3.21s/it]
                                                        

 45%|████▍     | 6560/14725 [5:58:54<7:17:23,  3.21s/it]
 45%|████▍     | 6561/14725 [5:58:57<7:15:35,  3.20s/it]
 45%|████▍     | 6562/14725 [5:59:00<7:19:33,  3.23s/it]
 45%|████▍     | 6563/14725 [5:59:03<7:20:18,  3.24s/it]
 45%|████▍     | 6564/14725 [5:59:07<7:24:16,  3.27s/it]
 45%|████▍     | 6565/14725 [5:59:10<7:23:57,  3.26s/it]
 45%|████▍     | 6566/14725 [5:59:13<7:16:31,  3.21s/it]
 45%|████▍     | 6567/14725 [5:59:16<7:22:54,  3.26s/it]
 45%|████▍     | 6568/14725 [5:59:20<7:38:59,  3.38s/it]
 45%|████▍     | 6569/14725 [5:59:23<7:36:16,  3.36s/it]
 45%|████▍     | 6570/14725 [5:59:27<7:38:18,  3.37s/it]
                                       
0: {'loss': 0.6986, 'grad_norm': 0.4127393108473845, 'learning_rate': 1e-05, 'epoch': 0.45}
0: {'loss': 0.7077, 'grad_norm': 0.4109879461780326, 'learning_rate': 1e-05, 'epoch': 0.45}
0:                  

 45%|████▍     | 6570/14725 [5:59:27<7:38:18,  3.37s/it]
 45%|████▍     | 6571/14725 [5:59:30<7:32:07,  3.33s/it]
 45%|████▍     | 6572/14725 [5:59:33<7:26:20,  3.28s/it]
 45%|████▍     | 6573/14725 [5:59:36<7:22:57,  3.26s/it]
 45%|████▍     | 6574/14725 [5:59:40<7:30:12,  3.31s/it]
 45%|████▍     | 6575/14725 [5:59:43<7:24:43,  3.27s/it]
 45%|████▍     | 6576/14725 [5:59:46<7:31:59,  3.33s/it]
 45%|████▍     | 6577/14725 [5:59:50<7:33:30,  3.34s/it]
 45%|████▍     | 6578/14725 [5:59:53<7:26:55,  3.29s/it]
 45%|████▍     | 6579/14725 [5:59:56<7:32:24,  3.33s/it]
 45%|████▍     | 6580/14725 [6:00:00<7:27:18,  3.30s/it]
                                                        

 45%|████▍     | 6580/14725 [6:00:00<7:27:18,  3.30s/it]
 45%|████▍     | 6581/14725 [6:00:03<7:23:41,  3.27s/it]
 45%|████▍     | 6582/14725 [6:00:06<7:13:30,  3.19s/it]
 45%|█�
0: {'loss': 0.7052, 'grad_norm': 0.4441964423765746, 'learning_rate': 1e-05, 'epoch': 0.45}
0: ��██▍     | 6583/14725 [6:00:09<7:12:15,  3.19s/it]
 45%|████▍     | 6584/14725 [6:00:12<7:15:42,  3.21s/it]
 45%|████▍     | 6585/14725 [6:00:15<7:09:55,  3.17s/it]
 45%|████▍     | 6586/14725 [6:00:19<7:12:51,  3.19s/it]
 45%|████▍     | 6587/14725 [6:00:22<7:16:58,  3.22s/it]
 45%|████▍     | 6588/14725 [6:00:25<7:26:06,  3.29s/it]
 45%|████▍     | 6589/14725 [6:00:28<7:18:08,  3.23s/it]
 45%|████▍     | 6590/14725 [6:00:32<7:15:30,  3.21s/it]
                                                        

 45%|████▍     | 6590/14725 [6:00:32<7:15:30,  3.21s/it]
 45%|████▍     | 6591/14725 [6:00:35<7:14:57,  3.21s/it]
 45%|████▍     | 6592/14725 [6:00:38<7:22:27,  3.26s/it]
 45%|████▍     | 6593/14725 [6:00:42<7:24:14,  3.28s/it]
 45%|████▍     | 6594/14725 [6:00:45<7:34:22,  3.35s/it]
 45%|████▍     | 6595/14725 [6:00:48<7:33:02,  3.34s/it]
 45%|████▍     | 6596/14725
0: {'loss': 0.6975, 'grad_norm': 0.4019580162699815, 'learning_rate': 1e-05, 'epoch': 0.45}
0:  [6:00:52<7:27:55,  3.31s/it]
 45%|████▍     | 6597/14725 [6:00:55<7:22:30,  3.27s/it]
 45%|████▍     | 6598/14725 [6:00:58<7:18:12,  3.24s/it]
 45%|████▍     | 6599/14725 [6:01:01<7:22:36,  3.27s/it]
 45%|████▍     | 6600/14725 [6:01:05<7:25:52,  3.29s/it]
                                                        

 45%|████▍     | 6600/14725 [6:01:05<7:25:52,  3.29s/it]
 45%|████▍     | 6601/14725 [6:01:08<7:30:38,  3.33s/it]
 45%|████▍     | 6602/14725 [6:01:11<7:20:44,  3.26s/it]
 45%|████▍     | 6603/14725 [6:01:14<7:21:54,  3.26s/it]
 45%|████▍     | 6604/14725 [6:01:18<7:14:40,  3.21s/it]
 45%|████▍     | 6605/14725 [6:01:21<7:22:44,  3.27s/it]
 45%|████▍     | 6606/14725 [6:01:24<7:18:54,  3.24s/it]
 45%|████▍     | 6607/14725 [6:01:27<7:18:32,  3.24s/it]
 45%|████▍     | 6608/14725 [6:01:31<7:23:51,  3.28s/it]
 45%|████▍     | 6609/14725 [6:01:34<7:26:29,  3.30s/it
0: {'loss': 0.7066, 'grad_norm': 0.44542135249630344, 'learning_rate': 1e-05, 'epoch': 0.45}
0: {'loss': 0.7056, 'grad_norm': 0.4193850627770692, 'learning_rate': 1e-05, 'epoch': 0.45}
0: ]
 45%|████▍     | 6610/14725 [6:01:37<7:24:25,  3.29s/it]
                                                        

 45%|████▍     | 6610/14725 [6:01:37<7:24:25,  3.29s/it]
 45%|████▍     | 6611/14725 [6:01:41<7:23:46,  3.28s/it]
 45%|████▍     | 6612/14725 [6:01:44<7:20:40,  3.26s/it]
 45%|████▍     | 6613/14725 [6:01:47<7:23:08,  3.28s/it]
 45%|████▍     | 6614/14725 [6:01:50<7:23:19,  3.28s/it]
 45%|████▍     | 6615/14725 [6:01:54<7:23:45,  3.28s/it]
 45%|████▍     | 6616/14725 [6:01:57<7:24:18,  3.29s/it]
 45%|████▍     | 6617/14725 [6:02:00<7:18:52,  3.25s/it]
 45%|████▍     | 6618/14725 [6:02:03<7:21:35,  3.27s/it]
 45%|████▍     | 6619/14725 [6:02:07<7:14:34,  3.22s/it]
 45%|████▍     | 6620/14725 [6:02:10<7:21:15,  3.27s/it]
                                                        

 45%|████▍     | 6620/14725 [6:02:10<7:21:15,  3.27s/it]
 45%|████▍     | 6621/147
0: {'loss': 0.7123, 'grad_norm': 0.4228495143986567, 'learning_rate': 1e-05, 'epoch': 0.45}
0: 25 [6:02:13<7:29:39,  3.33s/it]
 45%|████▍     | 6622/14725 [6:02:17<7:30:29,  3.34s/it]
 45%|████▍     | 6623/14725 [6:02:20<7:33:43,  3.36s/it]
 45%|████▍     | 6624/14725 [6:02:23<7:27:18,  3.31s/it]
 45%|████▍     | 6625/14725 [6:02:27<7:27:57,  3.32s/it]
 45%|████▍     | 6626/14725 [6:02:30<7:24:08,  3.29s/it]
 45%|████▌     | 6627/14725 [6:02:33<7:21:18,  3.27s/it]
 45%|████▌     | 6628/14725 [6:02:36<7:23:06,  3.28s/it]
 45%|████▌     | 6629/14725 [6:02:40<7:27:44,  3.32s/it]
 45%|████▌     | 6630/14725 [6:02:43<7:24:59,  3.30s/it]
                                                        

 45%|████▌     | 6630/14725 [6:02:43<7:24:59,  3.30s/it]
 45%|████▌     | 6631/14725 [6:02:47<7:28:02,  3.32s/it]
 45%|████▌     | 6632/14725 [6:02:50<7:29:28,  3.33s/it]
 45%|████▌     | 6633/14725 [6:02:53<7:24:08,  3.29s/it]
 45%|████▌     | 6634/14725 [6:02:56<7:18:17,  3.25s/
0: {'loss': 0.7088, 'grad_norm': 0.44677258160347966, 'learning_rate': 1e-05, 'epoch': 0.45}
0: it]
 45%|████▌     | 6635/14725 [6:02:59<7:18:03,  3.25s/it]
 45%|████▌     | 6636/14725 [6:03:03<7:18:05,  3.25s/it]
 45%|████▌     | 6637/14725 [6:03:06<7:16:11,  3.24s/it]
 45%|████▌     | 6638/14725 [6:03:09<7:23:23,  3.29s/it]
 45%|████▌     | 6639/14725 [6:03:13<7:21:18,  3.27s/it]
 45%|████▌     | 6640/14725 [6:03:16<7:23:45,  3.29s/it]
                                                        

 45%|████▌     | 6640/14725 [6:03:16<7:23:45,  3.29s/it]
 45%|████▌     | 6641/14725 [6:03:19<7:25:54,  3.31s/it]
 45%|████▌     | 6642/14725 [6:03:23<7:30:27,  3.34s/it]
 45%|████▌     | 6643/14725 [6:03:26<7:31:04,  3.35s/it]
 45%|████▌     | 6644/14725 [6:03:29<7:25:28,  3.31s/it]
 45%|████▌     | 6645/14725 [6:03:33<7:25:15,  3.31s/it]
 45%|████▌     | 6646/14725 [6:03:36<7:26:19,  3.31s/it]
 45%|████▌     | 6647/14725 [6:03:39<7:27:03,  3.32s/it]
 45%|████▌    
0: {'loss': 0.7038, 'grad_norm': 0.44281284746121236, 'learning_rate': 1e-05, 'epoch': 0.45}
0:  | 6648/14725 [6:03:42<7:18:23,  3.26s/it]
 45%|████▌     | 6649/14725 [6:03:46<7:24:27,  3.30s/it]
 45%|████▌     | 6650/14725 [6:03:49<7:24:17,  3.30s/it]
                                                        

 45%|████▌     | 6650/14725 [6:03:49<7:24:17,  3.30s/it]
 45%|████▌     | 6651/14725 [6:03:52<7:15:13,  3.23s/it]
 45%|████▌     | 6652/14725 [6:03:55<7:13:12,  3.22s/it]
 45%|████▌     | 6653/14725 [6:03:59<7:18:16,  3.26s/it]
 45%|████▌     | 6654/14725 [6:04:02<7:20:21,  3.27s/it]
 45%|████▌     | 6655/14725 [6:04:05<7:18:18,  3.26s/it]
 45%|████▌     | 6656/14725 [6:04:09<7:21:36,  3.28s/it]
 45%|████▌     | 6657/14725 [6:04:12<7:20:46,  3.28s/it]
 45%|████▌     | 6658/14725 [6:04:15<7:22:11,  3.29s/it]
 45%|████▌     | 6659/14725 [6:04:19<7:32:44,  3.37s/it]
 45%|████▌     | 6660/14725 [6:04:22<7:20:44,  3.28s/it]
                                                    
0: {'loss': 0.7067, 'grad_norm': 0.40224609015347834, 'learning_rate': 1e-05, 'epoch': 0.45}
0: {'loss': 0.7177, 'grad_norm': 0.4128305953277735, 'learning_rate': 1e-05, 'epoch': 0.45}
0:     

 45%|████▌     | 6660/14725 [6:04:22<7:20:44,  3.28s/it]
 45%|████▌     | 6661/14725 [6:04:25<7:24:25,  3.31s/it]
 45%|████▌     | 6662/14725 [6:04:28<7:22:13,  3.29s/it]
 45%|████▌     | 6663/14725 [6:04:32<7:18:14,  3.26s/it]
 45%|████▌     | 6664/14725 [6:04:35<7:18:17,  3.26s/it]
 45%|████▌     | 6665/14725 [6:04:38<7:21:16,  3.28s/it]
 45%|████▌     | 6666/14725 [6:04:42<7:27:15,  3.33s/it]
 45%|████▌     | 6667/14725 [6:04:45<7:22:21,  3.29s/it]
 45%|████▌     | 6668/14725 [6:04:48<7:21:49,  3.29s/it]
 45%|████▌     | 6669/14725 [6:04:51<7:20:05,  3.28s/it]
 45%|████▌     | 6670/14725 [6:04:55<7:24:40,  3.31s/it]
                                                        

 45%|████▌     | 6670/14725 [6:04:55<7:24:40,  3.31s/it]
 45%|████▌     | 6671/14725 [6:04:58<7:19:30,  3.27s/it]
 45%|████▌     | 6672/14725 [6:05:01<7:15:41,  3.25s/it]
 45%|████▌  
0: {'loss': 0.6912, 'grad_norm': 0.40506080036711295, 'learning_rate': 1e-05, 'epoch': 0.45}
0:    | 6673/14725 [6:05:04<7:13:54,  3.23s/it]
 45%|████▌     | 6674/14725 [6:05:07<7:11:41,  3.22s/it]
 45%|████▌     | 6675/14725 [6:05:11<7:15:37,  3.25s/it]
 45%|████▌     | 6676/14725 [6:05:14<7:25:34,  3.32s/it]
 45%|████▌     | 6677/14725 [6:05:18<7:28:31,  3.34s/it]
 45%|████▌     | 6678/14725 [6:05:21<7:31:23,  3.37s/it]
 45%|████▌     | 6679/14725 [6:05:24<7:28:42,  3.35s/it]
 45%|████▌     | 6680/14725 [6:05:28<7:27:11,  3.34s/it]
                                                        

 45%|████▌     | 6680/14725 [6:05:28<7:27:11,  3.34s/it]
 45%|████▌     | 6681/14725 [6:05:31<7:26:50,  3.33s/it]
 45%|████▌     | 6682/14725 [6:05:34<7:18:03,  3.27s/it]
 45%|████▌     | 6683/14725 [6:05:37<7:14:25,  3.24s/it]
 45%|████▌     | 6684/14725 [6:05:41<7:21:28,  3.29s/it]
 45%|████▌     | 6685/14725 [6:05:44<7:26:30,  3.33s/it]
 45%|████▌     | 6686/14725 [6:05:47<7:2
0: {'loss': 0.7181, 'grad_norm': 0.4332427702453979, 'learning_rate': 1e-05, 'epoch': 0.45}
0: 1:13,  3.29s/it]
 45%|████▌     | 6687/14725 [6:05:51<7:22:41,  3.30s/it]
 45%|████▌     | 6688/14725 [6:05:54<7:20:58,  3.29s/it]
 45%|████▌     | 6689/14725 [6:05:57<7:17:24,  3.27s/it]
 45%|████▌     | 6690/14725 [6:06:00<7:15:07,  3.25s/it]
                                                        

 45%|████▌     | 6690/14725 [6:06:00<7:15:07,  3.25s/it]
 45%|████▌     | 6691/14725 [6:06:04<7:14:31,  3.25s/it]
 45%|████▌     | 6692/14725 [6:06:07<7:10:04,  3.21s/it]
 45%|████▌     | 6693/14725 [6:06:10<7:15:40,  3.25s/it]
 45%|████▌     | 6694/14725 [6:06:13<7:13:38,  3.24s/it]
 45%|████▌     | 6695/14725 [6:06:17<7:12:48,  3.23s/it]
 45%|████▌     | 6696/14725 [6:06:20<7:09:18,  3.21s/it]
 45%|████▌     | 6697/14725 [6:06:23<7:15:21,  3.25s/it]
 45%|████▌     | 6698/14725 [6:06:26<7:12:05,  3.23s/it]
 45%|████▌     | 6699/14725 [6:06:29<7:09:49,  3.21s/it]
 46%|██
0: {'loss': 0.7036, 'grad_norm': 0.4272551869243931, 'learning_rate': 1e-05, 'epoch': 0.45}
0: {'loss': 0.6956, 'grad_norm': 0.4066001765171633, 'learning_rate': 1e-05, 'epoch': 0.46}
0: ██▌     | 6700/14725 [6:06:33<7:17:29,  3.27s/it]
                                                        

 46%|████▌     | 6700/14725 [6:06:33<7:17:29,  3.27s/it]
 46%|████▌     | 6701/14725 [6:06:36<7:19:33,  3.29s/it]
 46%|████▌     | 6702/14725 [6:06:39<7:17:54,  3.27s/it]
 46%|████▌     | 6703/14725 [6:06:43<7:18:35,  3.28s/it]
 46%|████▌     | 6704/14725 [6:06:46<7:14:46,  3.25s/it]
 46%|████▌     | 6705/14725 [6:06:49<7:08:18,  3.20s/it]
 46%|████▌     | 6706/14725 [6:06:52<7:08:27,  3.21s/it]
 46%|████▌     | 6707/14725 [6:06:55<7:08:34,  3.21s/it]
 46%|████▌     | 6708/14725 [6:06:59<7:13:05,  3.24s/it]
 46%|████▌     | 6709/14725 [6:07:02<7:20:43,  3.30s/it]
 46%|████▌     | 6710/14725 [6:07:05<7:16:28,  3.27s/it]
                                                        

 46%|████▌     | 6710/14725 [6:07:05<7:16:28,  3.27s/it]
 46%|████▌     | 6711/14725 [6:07:09<7
0: {'loss': 0.704, 'grad_norm': 0.42740374340469334, 'learning_rate': 1e-05, 'epoch': 0.46}
0: :18:01,  3.28s/it]
 46%|████▌     | 6712/14725 [6:07:12<7:13:12,  3.24s/it]
 46%|████▌     | 6713/14725 [6:07:15<7:26:39,  3.34s/it]
 46%|████▌     | 6714/14725 [6:07:19<7:28:02,  3.36s/it]
 46%|████▌     | 6715/14725 [6:07:22<7:31:39,  3.38s/it]
 46%|████▌     | 6716/14725 [6:07:26<7:41:59,  3.46s/it]
 46%|████▌     | 6717/14725 [6:07:29<7:34:41,  3.41s/it]
 46%|████▌     | 6718/14725 [6:07:32<7:27:04,  3.35s/it]
 46%|████▌     | 6719/14725 [6:07:36<7:31:57,  3.39s/it]
 46%|████▌     | 6720/14725 [6:07:39<7:28:42,  3.36s/it]
                                                        

 46%|████▌     | 6720/14725 [6:07:39<7:28:42,  3.36s/it]
 46%|████▌     | 6721/14725 [6:07:43<7:31:15,  3.38s/it]
 46%|████▌     | 6722/14725 [6:07:46<7:21:17,  3.31s/it]
 46%|████▌     | 6723/14725 [6:07:49<7:19:04,  3.29s/it]
 46%|████▌     | 6724/14725 [6:07:52<7:10:49,  3.23s/it]
 46%|█�
0: {'loss': 0.7209, 'grad_norm': 0.41623452508288433, 'learning_rate': 1e-05, 'epoch': 0.46}
0: ��██▌     | 6725/14725 [6:07:55<7:15:39,  3.27s/it]
 46%|████▌     | 6726/14725 [6:07:59<7:13:36,  3.25s/it]
 46%|████▌     | 6727/14725 [6:08:02<7:11:17,  3.24s/it]
 46%|████▌     | 6728/14725 [6:08:05<7:16:22,  3.27s/it]
 46%|████▌     | 6729/14725 [6:08:09<7:27:29,  3.36s/it]
 46%|████▌     | 6730/14725 [6:08:12<7:21:15,  3.31s/it]
                                                        

 46%|████▌     | 6730/14725 [6:08:12<7:21:15,  3.31s/it]
 46%|████▌     | 6731/14725 [6:08:15<7:24:35,  3.34s/it]
 46%|████▌     | 6732/14725 [6:08:19<7:20:20,  3.31s/it]
 46%|████▌     | 6733/14725 [6:08:22<7:16:19,  3.28s/it]
 46%|████▌     | 6734/14725 [6:08:25<7:20:25,  3.31s/it]
 46%|████▌     | 6735/14725 [6:08:28<7:18:07,  3.29s/it]
 46%|████▌     | 6736/14725 [6:08:32<7:13:01,  3.25s/it]
 46%|████▌     | 6737/14725 [6:08:35<7:13:12,  3.25s/it]
 46%|████▌     | 6738/14725
0: {'loss': 0.7051, 'grad_norm': 0.44149563328943203, 'learning_rate': 1e-05, 'epoch': 0.46}
0: {'loss': 0.7086, 'grad_norm': 0.446366217421396, 'learning_rate': 1e-05, 'epoch': 0.46}
0:  [6:08:38<7:14:20,  3.26s/it]
 46%|████▌     | 6739/14725 [6:08:41<7:05:35,  3.20s/it]
 46%|████▌     | 6740/14725 [6:08:44<7:01:11,  3.16s/it]
                                                        

 46%|████▌     | 6740/14725 [6:08:44<7:01:11,  3.16s/it]
 46%|████▌     | 6741/14725 [6:08:47<7:06:31,  3.21s/it]
 46%|████▌     | 6742/14725 [6:08:51<7:11:53,  3.25s/it]
 46%|████▌     | 6743/14725 [6:08:54<7:15:55,  3.28s/it]
 46%|████▌     | 6744/14725 [6:08:57<7:08:44,  3.22s/it]
 46%|████▌     | 6745/14725 [6:09:00<7:05:25,  3.20s/it]
 46%|████▌     | 6746/14725 [6:09:04<7:14:30,  3.27s/it]
 46%|████▌     | 6747/14725 [6:09:07<7:19:19,  3.30s/it]
 46%|████▌     | 6748/14725 [6:09:11<7:26:14,  3.36s/it]
 46%|████▌     | 6749/14725 [6:09:14<7:19:41,  3.31s/it]
 46%|████▌     | 6750/14725 [6:09:17<7:18:59,  3.30s/it]
                                                        

 46%|�
0: {'loss': 0.6873, 'grad_norm': 0.41219391587466236, 'learning_rate': 1e-05, 'epoch': 0.46}
0: �███▌     | 6750/14725 [6:09:17<7:18:59,  3.30s/it]
 46%|████▌     | 6751/14725 [6:09:20<7:18:41,  3.30s/it]
 46%|████▌     | 6752/14725 [6:09:24<7:08:54,  3.23s/it]
 46%|████▌     | 6753/14725 [6:09:27<7:01:46,  3.17s/it]
 46%|████▌     | 6754/14725 [6:09:30<7:05:50,  3.21s/it]
 46%|████▌     | 6755/14725 [6:09:33<7:10:28,  3.24s/it]
 46%|████▌     | 6756/14725 [6:09:36<7:06:55,  3.21s/it]
 46%|████▌     | 6757/14725 [6:09:40<7:09:59,  3.24s/it]
 46%|████▌     | 6758/14725 [6:09:43<7:17:31,  3.29s/it]
 46%|████▌     | 6759/14725 [6:09:46<7:14:46,  3.27s/it]
 46%|████▌     | 6760/14725 [6:09:49<7:06:31,  3.21s/it]
                                                        

 46%|████▌     | 6760/14725 [6:09:49<7:06:31,  3.21s/it]
 46%|████▌     | 6761/14725 [6:09:53<7:09:44,  3.24s/it]
 46%|████▌     | 6762/14725 [6:09:56<7:08:50,  3.23s/it]
 46%|████▌     | 6763/147
0: {'loss': 0.7056, 'grad_norm': 0.3887212683987474, 'learning_rate': 1e-05, 'epoch': 0.46}
0: 25 [6:09:59<7:08:52,  3.23s/it]
 46%|████▌     | 6764/14725 [6:10:02<7:02:01,  3.18s/it]
 46%|████▌     | 6765/14725 [6:10:06<7:08:16,  3.23s/it]
 46%|████▌     | 6766/14725 [6:10:09<7:14:42,  3.28s/it]
 46%|████▌     | 6767/14725 [6:10:12<7:26:18,  3.37s/it]
 46%|████▌     | 6768/14725 [6:10:16<7:26:42,  3.37s/it]
 46%|████▌     | 6769/14725 [6:10:19<7:24:50,  3.35s/it]
 46%|████▌     | 6770/14725 [6:10:22<7:17:53,  3.30s/it]
                                                        

 46%|████▌     | 6770/14725 [6:10:22<7:17:53,  3.30s/it]
 46%|████▌     | 6771/14725 [6:10:26<7:25:10,  3.36s/it]
 46%|████▌     | 6772/14725 [6:10:29<7:18:12,  3.31s/it]
 46%|████▌     | 6773/14725 [6:10:32<7:20:42,  3.33s/it]
 46%|████▌     | 6774/14725 [6:10:36<7:16:46,  3.30s/it]
 46%|████▌     | 6775/14725 [6:10:39<7:30:14,  3.40s/it]
 46%|████▌     | 6776/14725 [6:10:43<7:33:32,  3.42s/
0: {'loss': 0.6965, 'grad_norm': 0.43204146201699956, 'learning_rate': 1e-05, 'epoch': 0.46}
0: it]
 46%|████▌     | 6777/14725 [6:10:46<7:21:01,  3.33s/it]
 46%|████▌     | 6778/14725 [6:10:49<7:13:39,  3.27s/it]
 46%|████▌     | 6779/14725 [6:10:52<7:08:22,  3.23s/it]
 46%|████▌     | 6780/14725 [6:10:55<7:09:09,  3.24s/it]
                                                        

 46%|████▌     | 6780/14725 [6:10:55<7:09:09,  3.24s/it]
 46%|████▌     | 6781/14725 [6:10:59<7:17:39,  3.31s/it]
 46%|████▌     | 6782/14725 [6:11:02<7:09:43,  3.25s/it]
 46%|████▌     | 6783/14725 [6:11:05<7:11:37,  3.26s/it]
 46%|████▌     | 6784/14725 [6:11:09<7:11:22,  3.26s/it]
 46%|████▌     | 6785/14725 [6:11:12<7:09:13,  3.24s/it]
 46%|████▌     | 6786/14725 [6:11:15<7:13:51,  3.28s/it]
 46%|████▌     | 6787/14725 [6:11:18<7:11:01,  3.26s/it]
 46%|████▌     | 6788/14725 [6:11:22<7:09:36,  3.25s/it]
 46%|████▌     | 6789/14725 [6:11:25<7:11:57,  3.27s/it]
 46%|████▌    
0: {'loss': 0.7102, 'grad_norm': 0.45276236802708775, 'learning_rate': 1e-05, 'epoch': 0.46}
0: {'loss': 0.6821, 'grad_norm': 0.4053309227259144, 'learning_rate': 1e-05, 'epoch': 0.46}
0:  | 6790/14725 [6:11:28<7:10:57,  3.26s/it]
                                                        

 46%|████▌     | 6790/14725 [6:11:28<7:10:57,  3.26s/it]
 46%|████▌     | 6791/14725 [6:11:31<7:08:13,  3.24s/it]
 46%|████▌     | 6792/14725 [6:11:34<7:06:59,  3.23s/it]
 46%|████▌     | 6793/14725 [6:11:38<7:14:54,  3.29s/it]
 46%|████▌     | 6794/14725 [6:11:41<7:11:08,  3.26s/it]
 46%|████▌     | 6795/14725 [6:11:45<7:21:03,  3.34s/it]
 46%|████▌     | 6796/14725 [6:11:48<7:22:00,  3.34s/it]
 46%|████▌     | 6797/14725 [6:11:51<7:25:17,  3.37s/it]
 46%|████▌     | 6798/14725 [6:11:55<7:23:14,  3.35s/it]
 46%|████▌     | 6799/14725 [6:11:58<7:15:43,  3.30s/it]
 46%|████▌     | 6800/14725 [6:12:01<7:14:45,  3.29s/it]
                                                        

 46%|████▌     | 6800/14725 [6:12:01<7:14:45,  3.29s/it]
 46%|████▌     | 6801/14725 [6:12:05<7:18:37,  3.32
0: {'loss': 0.7088, 'grad_norm': 0.42630193347306783, 'learning_rate': 1e-05, 'epoch': 0.46}
0: s/it]
 46%|████▌     | 6802/14725 [6:12:08<7:30:04,  3.41s/it]
 46%|████▌     | 6803/14725 [6:12:11<7:18:08,  3.32s/it]
 46%|████▌     | 6804/14725 [6:12:15<7:24:46,  3.37s/it]
 46%|████▌     | 6805/14725 [6:12:18<7:19:26,  3.33s/it]
 46%|████▌     | 6806/14725 [6:12:21<7:22:21,  3.35s/it]
 46%|████▌     | 6807/14725 [6:12:25<7:14:04,  3.29s/it]
 46%|████▌     | 6808/14725 [6:12:28<7:11:08,  3.27s/it]
 46%|████▌     | 6809/14725 [6:12:31<7:05:28,  3.22s/it]
 46%|████▌     | 6810/14725 [6:12:34<7:04:38,  3.22s/it]
                                                        

 46%|████▌     | 6810/14725 [6:12:34<7:04:38,  3.22s/it]
 46%|████▋     | 6811/14725 [6:12:37<7:07:58,  3.24s/it]
 46%|████▋     | 6812/14725 [6:12:41<7:11:04,  3.27s/it]
 46%|████▋     | 6813/14725 [6:12:44<7:05:52,  3.23s/it]
 46%|████▋     | 6814/14725 [6:12:47<7:13:21,  3.29s/it]
 46%|████▋  
0: {'loss': 0.7137, 'grad_norm': 0.40461744686101586, 'learning_rate': 1e-05, 'epoch': 0.46}
0:    | 6815/14725 [6:12:51<7:13:38,  3.29s/it]
 46%|████▋     | 6816/14725 [6:12:54<7:11:15,  3.27s/it]
 46%|████▋     | 6817/14725 [6:12:57<7:00:51,  3.19s/it]
 46%|████▋     | 6818/14725 [6:13:00<6:59:48,  3.19s/it]
 46%|████▋     | 6819/14725 [6:13:03<7:07:45,  3.25s/it]
 46%|████▋     | 6820/14725 [6:13:07<7:08:53,  3.26s/it]
                                                        

 46%|████▋     | 6820/14725 [6:13:07<7:08:53,  3.26s/it]
 46%|████▋     | 6821/14725 [6:13:10<7:12:10,  3.28s/it]
 46%|████▋     | 6822/14725 [6:13:13<7:10:46,  3.27s/it]
 46%|████▋     | 6823/14725 [6:13:17<7:16:36,  3.32s/it]
 46%|████▋     | 6824/14725 [6:13:20<7:12:20,  3.28s/it]
 46%|████▋     | 6825/14725 [6:13:23<7:07:25,  3.25s/it]
 46%|████▋     | 6826/14725 [6:13:26<7:02:09,  3.21s/it]
 46%|████▋     | 6827/14725 [6:13:29<7:02:34,  3.21s/it]
 46%|████▋     | 6828/14725 [6:13:33<7:0
0: {'loss': 0.7121, 'grad_norm': 0.45725727103997466, 'learning_rate': 1e-05, 'epoch': 0.46}
0: {'loss': 0.6962, 'grad_norm': 0.40012196377370157, 'learning_rate': 1e-05, 'epoch': 0.46}
0: 1:45,  3.20s/it]
 46%|████▋     | 6829/14725 [6:13:36<7:08:29,  3.26s/it]
 46%|████▋     | 6830/14725 [6:13:39<7:06:55,  3.24s/it]
                                                        

 46%|████▋     | 6830/14725 [6:13:39<7:06:55,  3.24s/it]
 46%|████▋     | 6831/14725 [6:13:42<7:03:32,  3.22s/it]
 46%|████▋     | 6832/14725 [6:13:46<7:10:23,  3.27s/it]
 46%|████▋     | 6833/14725 [6:13:49<7:06:45,  3.24s/it]
 46%|████▋     | 6834/14725 [6:13:52<7:03:43,  3.22s/it]
 46%|████▋     | 6835/14725 [6:13:56<7:18:11,  3.33s/it]
 46%|████▋     | 6836/14725 [6:13:59<7:09:57,  3.27s/it]
 46%|████▋     | 6837/14725 [6:14:02<7:14:29,  3.31s/it]
 46%|████▋     | 6838/14725 [6:14:05<7:12:50,  3.29s/it]
 46%|████▋     | 6839/14725 [6:14:09<7:13:19,  3.30s/it]
 46%|████▋     | 6840/14725 [6:14:12<7:02:52,  3.22s/it]
                                                        

 46%|████▋
0: {'loss': 0.6997, 'grad_norm': 0.4281049228572508, 'learning_rate': 1e-05, 'epoch': 0.47}
0:      | 6840/14725 [6:14:12<7:02:52,  3.22s/it]
 46%|████▋     | 6841/14725 [6:14:15<7:06:51,  3.25s/it]
 46%|████▋     | 6842/14725 [6:14:18<7:08:58,  3.27s/it]
 46%|████▋     | 6843/14725 [6:14:22<7:10:15,  3.28s/it]
 46%|████▋     | 6844/14725 [6:14:25<7:05:51,  3.24s/it]
 46%|████▋     | 6845/14725 [6:14:28<7:08:45,  3.26s/it]
 46%|████▋     | 6846/14725 [6:14:32<7:17:57,  3.34s/it]
 46%|████▋     | 6847/14725 [6:14:35<7:13:46,  3.30s/it]
 47%|████▋     | 6848/14725 [6:14:38<7:08:34,  3.26s/it]
 47%|████▋     | 6849/14725 [6:14:41<7:04:57,  3.24s/it]
 47%|████▋     | 6850/14725 [6:14:44<7:01:28,  3.21s/it]
                                                        

 47%|████▋     | 6850/14725 [6:14:44<7:01:28,  3.21s/it]
 47%|████▋     | 6851/14725 [6:14:48<7:02:10,  3.22s/it]
 47%|████▋     | 6852/14725 [6:14:51<7:07:58,  3.26s/it]
 47%|████▋     | 6853/14725 [6:14:54<7
0: {'loss': 0.7002, 'grad_norm': 0.41707109550572774, 'learning_rate': 1e-05, 'epoch': 0.47}
0: :09:36,  3.27s/it]
 47%|████▋     | 6854/14725 [6:14:58<7:11:22,  3.29s/it]
 47%|████▋     | 6855/14725 [6:15:01<7:09:08,  3.27s/it]
 47%|████▋     | 6856/14725 [6:15:04<7:04:04,  3.23s/it]
 47%|████▋     | 6857/14725 [6:15:07<6:55:40,  3.17s/it]
 47%|████▋     | 6858/14725 [6:15:10<7:01:35,  3.22s/it]
 47%|████▋     | 6859/14725 [6:15:14<7:00:48,  3.21s/it]
 47%|████▋     | 6860/14725 [6:15:17<7:04:28,  3.24s/it]
                                                        

 47%|████▋     | 6860/14725 [6:15:17<7:04:28,  3.24s/it]
 47%|████▋     | 6861/14725 [6:15:20<7:01:48,  3.22s/it]
 47%|████▋     | 6862/14725 [6:15:23<7:11:39,  3.29s/it]
 47%|████▋     | 6863/14725 [6:15:27<7:11:42,  3.29s/it]
 47%|████▋     | 6864/14725 [6:15:30<7:05:42,  3.25s/it]
 47%|████▋     | 6865/14725 [6:15:33<7:09:18,  3.28s/it]
 47%|████▋     | 6866/14725 [6:15:37<7:15:59,  3.33s/it]
 47%|█�
0: {'loss': 0.7075, 'grad_norm': 0.42953899056872685, 'learning_rate': 1e-05, 'epoch': 0.47}
0: ��██▋     | 6867/14725 [6:15:40<7:12:57,  3.31s/it]
 47%|████▋     | 6868/14725 [6:15:43<7:18:15,  3.35s/it]
 47%|████▋     | 6869/14725 [6:15:47<7:12:22,  3.30s/it]
 47%|████▋     | 6870/14725 [6:15:50<7:09:40,  3.28s/it]
                                                        

 47%|████▋     | 6870/14725 [6:15:50<7:09:40,  3.28s/it]
 47%|████▋     | 6871/14725 [6:15:53<7:07:51,  3.27s/it]
 47%|████▋     | 6872/14725 [6:15:56<7:05:33,  3.25s/it]
 47%|████▋     | 6873/14725 [6:16:00<7:05:18,  3.25s/it]
 47%|████▋     | 6874/14725 [6:16:03<7:12:05,  3.30s/it]
 47%|████▋     | 6875/14725 [6:16:06<7:08:28,  3.28s/it]
 47%|████▋     | 6876/14725 [6:16:10<7:11:11,  3.30s/it]
 47%|████▋     | 6877/14725 [6:16:13<7:16:08,  3.33s/it]
 47%|████▋     | 6878/14725 [6:16:16<7:08:47,  3.28s/it]
 47%|████▋     | 6879/14725 [6:16:19<7:06:39,  3.26s/it]
 47%|████▋     | 6880/14725
0: {'loss': 0.6952, 'grad_norm': 0.4183952217167361, 'learning_rate': 1e-05, 'epoch': 0.47}
0: {'loss': 0.6958, 'grad_norm': 0.4060072877884669, 'learning_rate': 1e-05, 'epoch': 0.47}
0:  [6:16:23<7:10:54,  3.30s/it]
                                                        

 47%|████▋     | 6880/14725 [6:16:23<7:10:54,  3.30s/it]
 47%|████▋     | 6881/14725 [6:16:26<7:15:16,  3.33s/it]
 47%|████▋     | 6882/14725 [6:16:29<7:16:29,  3.34s/it]
 47%|████▋     | 6883/14725 [6:16:33<7:12:11,  3.31s/it]
 47%|████▋     | 6884/14725 [6:16:36<7:17:18,  3.35s/it]
 47%|████▋     | 6885/14725 [6:16:39<7:14:58,  3.33s/it]
 47%|████▋     | 6886/14725 [6:16:43<7:17:59,  3.35s/it]
 47%|████▋     | 6887/14725 [6:16:46<7:16:49,  3.34s/it]
 47%|████▋     | 6888/14725 [6:16:49<7:15:16,  3.33s/it]
 47%|████▋     | 6889/14725 [6:16:53<7:12:34,  3.31s/it]
 47%|████▋     | 6890/14725 [6:16:56<7:04:13,  3.25s/it]
                                                        

 47%|████▋     | 6890/14725 [6:16:56<7:04:13,  3.25s/it]
 47%|████▋     | 6891/14725 [6:16:59<7:05:44,  3.26s/it]
 47%|�
0: {'loss': 0.6935, 'grad_norm': 0.40972793182304384, 'learning_rate': 1e-05, 'epoch': 0.47}
0: �███▋     | 6892/14725 [6:17:03<7:13:17,  3.32s/it]
 47%|████▋     | 6893/14725 [6:17:06<7:11:14,  3.30s/it]
 47%|████▋     | 6894/14725 [6:17:09<7:11:48,  3.31s/it]
 47%|████▋     | 6895/14725 [6:17:12<7:12:15,  3.31s/it]
 47%|████▋     | 6896/14725 [6:17:16<7:05:49,  3.26s/it]
 47%|████▋     | 6897/14725 [6:17:19<7:15:11,  3.34s/it]
 47%|████▋     | 6898/14725 [6:17:22<7:14:49,  3.33s/it]
 47%|████▋     | 6899/14725 [6:17:26<7:06:26,  3.27s/it]
 47%|████▋     | 6900/14725 [6:17:29<6:58:46,  3.21s/it]
                                                        

 47%|████▋     | 6900/14725 [6:17:29<6:58:46,  3.21s/it]
 47%|████▋     | 6901/14725 [6:17:32<7:06:48,  3.27s/it]
 47%|████▋     | 6902/14725 [6:17:35<7:06:21,  3.27s/it]
 47%|████▋     | 6903/14725 [6:17:39<7:08:39,  3.29s/it]
 47%|████▋     | 6904/14725 [6:17:42<7:15:01,  3.34s/it]
 47%|████▋     | 6905/147
0: {'loss': 0.7213, 'grad_norm': 0.45409146512819387, 'learning_rate': 1e-05, 'epoch': 0.47}
0: 25 [6:17:45<7:09:24,  3.29s/it]
 47%|████▋     | 6906/14725 [6:17:49<7:08:27,  3.29s/it]
 47%|████▋     | 6907/14725 [6:17:52<7:06:09,  3.27s/it]
 47%|████▋     | 6908/14725 [6:17:55<7:06:42,  3.28s/it]
 47%|████▋     | 6909/14725 [6:17:58<7:05:31,  3.27s/it]
 47%|████▋     | 6910/14725 [6:18:02<7:08:57,  3.29s/it]
                                                        

 47%|████▋     | 6910/14725 [6:18:02<7:08:57,  3.29s/it]
 47%|████▋     | 6911/14725 [6:18:05<7:10:42,  3.31s/it]
 47%|████▋     | 6912/14725 [6:18:08<7:08:13,  3.29s/it]
 47%|████▋     | 6913/14725 [6:18:12<7:09:25,  3.30s/it]
 47%|████▋     | 6914/14725 [6:18:15<7:07:52,  3.29s/it]
 47%|████▋     | 6915/14725 [6:18:18<7:06:06,  3.27s/it]
 47%|████▋     | 6916/14725 [6:18:21<7:02:20,  3.25s/it]
 47%|████▋     | 6917/14725 [6:18:24<7:01:24,  3.24s/it]
 47%|████▋     | 6918/14725 [6:18:28<7:03:57,  3.26s/
0: {'loss': 0.7131, 'grad_norm': 0.41544246880496044, 'learning_rate': 1e-05, 'epoch': 0.47}
0: {'loss': 0.6989, 'grad_norm': 0.41713048852626144, 'learning_rate': 1e-05, 'epoch': 0.47}
0: it]
 47%|████▋     | 6919/14725 [6:18:31<7:14:22,  3.34s/it]
 47%|████▋     | 6920/14725 [6:18:35<7:13:11,  3.33s/it]
                                                        

 47%|████▋     | 6920/14725 [6:18:35<7:13:11,  3.33s/it]
 47%|████▋     | 6921/14725 [6:18:38<7:05:33,  3.27s/it]
 47%|████▋     | 6922/14725 [6:18:41<7:04:06,  3.26s/it]
 47%|████▋     | 6923/14725 [6:18:44<7:06:42,  3.28s/it]
 47%|████▋     | 6924/14725 [6:18:47<7:01:19,  3.24s/it]
 47%|████▋     | 6925/14725 [6:18:51<7:03:54,  3.26s/it]
 47%|████▋     | 6926/14725 [6:18:54<6:56:45,  3.21s/it]
 47%|████▋     | 6927/14725 [6:18:57<6:57:51,  3.22s/it]
 47%|████▋     | 6928/14725 [6:19:00<6:53:47,  3.18s/it]
 47%|████▋     | 6929/14725 [6:19:03<6:54:31,  3.19s/it]
 47%|████▋     | 6930/14725 [6:19:07<7:03:37,  3.26s/it]
                                                        

 47%|████▋     | 6930/1
0: {'loss': 0.7031, 'grad_norm': 0.43480806597140786, 'learning_rate': 1e-05, 'epoch': 0.47}
0: 4725 [6:19:07<7:03:37,  3.26s/it]
 47%|████▋     | 6931/14725 [6:19:10<7:05:03,  3.27s/it]
 47%|████▋     | 6932/14725 [6:19:13<7:06:18,  3.28s/it]
 47%|████▋     | 6933/14725 [6:19:17<7:11:13,  3.32s/it]
 47%|████▋     | 6934/14725 [6:19:20<7:07:56,  3.30s/it]
 47%|████▋     | 6935/14725 [6:19:23<7:00:33,  3.24s/it]
 47%|████▋     | 6936/14725 [6:19:27<7:07:03,  3.29s/it]
 47%|████▋     | 6937/14725 [6:19:30<7:10:15,  3.31s/it]
 47%|████▋     | 6938/14725 [6:19:33<7:05:20,  3.28s/it]
 47%|████▋     | 6939/14725 [6:19:36<7:04:57,  3.27s/it]
 47%|████▋     | 6940/14725 [6:19:40<6:58:23,  3.22s/it]
                                                        

 47%|████▋     | 6940/14725 [6:19:40<6:58:23,  3.22s/it]
 47%|████▋     | 6941/14725 [6:19:43<6:58:18,  3.22s/it]
 47%|████▋     | 6942/14725 [6:19:46<7:11:10,  3.32s/it]
 47%|████▋     | 6943/14725 [6:19:50<7:07:06,  3.29
0: {'loss': 0.7112, 'grad_norm': 0.4255820953256102, 'learning_rate': 1e-05, 'epoch': 0.47}
0: s/it]
 47%|████▋     | 6944/14725 [6:19:53<7:13:25,  3.34s/it]
 47%|████▋     | 6945/14725 [6:19:56<7:12:51,  3.34s/it]
 47%|████▋     | 6946/14725 [6:20:00<7:13:01,  3.34s/it]
 47%|████▋     | 6947/14725 [6:20:03<7:05:21,  3.28s/it]
 47%|████▋     | 6948/14725 [6:20:06<7:03:32,  3.27s/it]
 47%|████▋     | 6949/14725 [6:20:09<7:02:22,  3.26s/it]
 47%|████▋     | 6950/14725 [6:20:13<7:07:37,  3.30s/it]
                                                        

 47%|████▋     | 6950/14725 [6:20:13<7:07:37,  3.30s/it]
 47%|████▋     | 6951/14725 [6:20:16<7:08:34,  3.31s/it]
 47%|████▋     | 6952/14725 [6:20:19<7:06:57,  3.30s/it]
 47%|████▋     | 6953/14725 [6:20:22<6:59:35,  3.24s/it]
 47%|████▋     | 6954/14725 [6:20:26<6:55:57,  3.21s/it]
 47%|████▋     | 6955/14725 [6:20:29<6:52:49,  3.19s/it]
 47%|████▋     | 6956/14725 [6:20:32<6:54:02,  3.20s/it]
 47%|████▋  
0: {'loss': 0.6928, 'grad_norm': 0.40138369249661454, 'learning_rate': 1e-05, 'epoch': 0.47}
0:    | 6957/14725 [6:20:35<6:56:01,  3.21s/it]
 47%|████▋     | 6958/14725 [6:20:38<6:50:25,  3.17s/it]
 47%|████▋     | 6959/14725 [6:20:41<6:52:23,  3.19s/it]
 47%|████▋     | 6960/14725 [6:20:45<6:54:29,  3.20s/it]
                                                        

 47%|████▋     | 6960/14725 [6:20:45<6:54:29,  3.20s/it]
 47%|████▋     | 6961/14725 [6:20:48<6:56:32,  3.22s/it]
 47%|████▋     | 6962/14725 [6:20:51<6:59:08,  3.24s/it]
 47%|████▋     | 6963/14725 [6:20:55<7:05:15,  3.29s/it]
 47%|████▋     | 6964/14725 [6:20:58<6:57:59,  3.23s/it]
 47%|████▋     | 6965/14725 [6:21:01<7:02:36,  3.27s/it]
 47%|████▋     | 6966/14725 [6:21:04<6:55:10,  3.21s/it]
 47%|████▋     | 6967/14725 [6:21:08<7:05:22,  3.29s/it]
 47%|████▋     | 6968/14725 [6:21:11<7:03:30,  3.28s/it]
 47%|████▋     | 6969/14725 [6:21:14<7:07:31,  3.31s/it]
 47%|████▋     | 6970/14725 [6:21:17<6:5
0: {'loss': 0.707, 'grad_norm': 0.4022336008671866, 'learning_rate': 1e-05, 'epoch': 0.47}
0: {'loss': 0.7044, 'grad_norm': 0.42646363438307555, 'learning_rate': 1e-05, 'epoch': 0.47}
0: 9:03,  3.24s/it]
                                                        

 47%|████▋     | 6970/14725 [6:21:17<6:59:03,  3.24s/it]
 47%|████▋     | 6971/14725 [6:21:21<7:05:03,  3.29s/it]
 47%|████▋     | 6972/14725 [6:21:24<6:56:10,  3.22s/it]
 47%|████▋     | 6973/14725 [6:21:27<7:01:05,  3.26s/it]
 47%|████▋     | 6974/14725 [6:21:30<6:57:03,  3.23s/it]
 47%|████▋     | 6975/14725 [6:21:34<6:55:41,  3.22s/it]
 47%|████▋     | 6976/14725 [6:21:37<6:52:50,  3.20s/it]
 47%|████▋     | 6977/14725 [6:21:40<7:02:53,  3.27s/it]
 47%|████▋     | 6978/14725 [6:21:43<7:00:38,  3.26s/it]
 47%|████▋     | 6979/14725 [6:21:47<7:02:32,  3.27s/it]
 47%|████▋     | 6980/14725 [6:21:50<6:58:04,  3.24s/it]
                                                        

 47%|████▋     | 6980/14725 [6:21:50<6:58:04,  3.24s/it]
 47%|████▋     | 6981/14725 [6:21:53<6:53:34,  3.20s/it]
 47%|████▋
0: {'loss': 0.7019, 'grad_norm': 0.4349343757418373, 'learning_rate': 1e-05, 'epoch': 0.47}
0:      | 6982/14725 [6:21:56<6:55:47,  3.22s/it]
 47%|████▋     | 6983/14725 [6:21:59<6:54:06,  3.21s/it]
 47%|████▋     | 6984/14725 [6:22:03<6:52:31,  3.20s/it]
 47%|████▋     | 6985/14725 [6:22:06<6:58:04,  3.24s/it]
 47%|████▋     | 6986/14725 [6:22:09<6:58:43,  3.25s/it]
 47%|████▋     | 6987/14725 [6:22:12<6:52:31,  3.20s/it]
 47%|████▋     | 6988/14725 [6:22:16<6:56:13,  3.23s/it]
 47%|████▋     | 6989/14725 [6:22:19<7:00:14,  3.26s/it]
 47%|████▋     | 6990/14725 [6:22:22<6:58:33,  3.25s/it]
                                                        

 47%|████▋     | 6990/14725 [6:22:22<6:58:33,  3.25s/it]
 47%|████▋     | 6991/14725 [6:22:25<6:59:21,  3.25s/it]
 47%|████▋     | 6992/14725 [6:22:29<7:00:30,  3.26s/it]
 47%|████▋     | 6993/14725 [6:22:32<7:00:44,  3.26s/it]
 47%|████▋     | 6994/14725 [6:22:35<7:06:00,  3.31s/it]
 48%|████▊     | 6995/14725 [6:22:39<7
0: {'loss': 0.7045, 'grad_norm': 0.4519959471880608, 'learning_rate': 1e-05, 'epoch': 0.48}
0: :04:50,  3.30s/it]
 48%|████▊     | 6996/14725 [6:22:42<7:02:20,  3.28s/it]
 48%|████▊     | 6997/14725 [6:22:45<7:06:45,  3.31s/it]
 48%|████▊     | 6998/14725 [6:22:48<7:02:01,  3.28s/it]
 48%|████▊     | 6999/14725 [6:22:52<6:57:34,  3.24s/it]
 48%|████▊     | 7000/14725 [6:22:55<7:00:18,  3.26s/it]
                                                        

 48%|████▊     | 7000/14725 [6:22:55<7:00:18,  3.26s/it]
 48%|████▊     | 7001/14725 [6:22:58<7:02:01,  3.28s/it]
 48%|████▊     | 7002/14725 [6:23:01<7:00:00,  3.26s/it]
 48%|████▊     | 7003/14725 [6:23:05<6:56:53,  3.24s/it]
 48%|████▊     | 7004/14725 [6:23:08<7:02:50,  3.29s/it]
 48%|████▊     | 7005/14725 [6:23:11<7:04:59,  3.30s/it]
 48%|████▊     | 7006/14725 [6:23:14<6:58:16,  3.25s/it]
 48%|████▊     | 7007/14725 [6:23:18<6:56:27,  3.24s/it]
 48%|████▊     | 7008/14725 [6:23:21<6:50:38,  3.19s/it]
 48%|█�
0: {'loss': 0.6997, 'grad_norm': 0.44390929756529035, 'learning_rate': 1e-05, 'epoch': 0.48}
0: {'loss': 0.7051, 'grad_norm': 0.4124055252114199, 'learning_rate': 1e-05, 'epoch': 0.48}
0: ��██▊     | 7009/14725 [6:23:24<6:55:03,  3.23s/it]
 48%|████▊     | 7010/14725 [6:23:27<7:02:03,  3.28s/it]
                                                        

 48%|████▊     | 7010/14725 [6:23:27<7:02:03,  3.28s/it]
 48%|████▊     | 7011/14725 [6:23:31<7:00:06,  3.27s/it]
 48%|████▊     | 7012/14725 [6:23:34<7:02:42,  3.29s/it]
 48%|████▊     | 7013/14725 [6:23:37<7:08:12,  3.33s/it]
 48%|████▊     | 7014/14725 [6:23:41<7:00:32,  3.27s/it]
 48%|████▊     | 7015/14725 [6:23:44<7:04:58,  3.31s/it]
 48%|████▊     | 7016/14725 [6:23:47<7:06:56,  3.32s/it]
 48%|████▊     | 7017/14725 [6:23:51<7:04:05,  3.30s/it]
 48%|████▊     | 7018/14725 [6:23:54<7:04:19,  3.30s/it]
 48%|████▊     | 7019/14725 [6:23:57<7:01:43,  3.28s/it]
 48%|████▊     | 7020/14725 [6:24:01<7:10:42,  3.35s/it]
                                                        

 48%|████▊     | 7020/14725 [6:24:01
0: {'loss': 0.706, 'grad_norm': 0.40729637045673006, 'learning_rate': 1e-05, 'epoch': 0.48}
0: <7:10:42,  3.35s/it]
 48%|████▊     | 7021/14725 [6:24:04<7:03:08,  3.30s/it]
 48%|████▊     | 7022/14725 [6:24:07<7:09:00,  3.34s/it]
 48%|████▊     | 7023/14725 [6:24:11<7:14:17,  3.38s/it]
 48%|████▊     | 7024/14725 [6:24:14<7:13:06,  3.37s/it]
 48%|████▊     | 7025/14725 [6:24:17<7:09:13,  3.34s/it]
 48%|████▊     | 7026/14725 [6:24:21<7:07:06,  3.33s/it]
 48%|████▊     | 7027/14725 [6:24:24<7:01:34,  3.29s/it]
 48%|████▊     | 7028/14725 [6:24:27<6:54:45,  3.23s/it]
 48%|████▊     | 7029/14725 [6:24:30<6:50:36,  3.20s/it]
 48%|████▊     | 7030/14725 [6:24:33<6:44:40,  3.16s/it]
                                                        

 48%|████▊     | 7030/14725 [6:24:33<6:44:40,  3.16s/it]
 48%|████▊     | 7031/14725 [6:24:37<6:54:46,  3.23s/it]
 48%|████▊     | 7032/14725 [6:24:40<6:54:44,  3.23s/it]
 48%|████▊     | 7033/14725 [6:24:43<7:00:37,  3.28s/it]
 48%|�
0: {'loss': 0.7018, 'grad_norm': 0.41850904479129075, 'learning_rate': 1e-05, 'epoch': 0.48}
0: �███▊     | 7034/14725 [6:24:47<7:05:34,  3.32s/it]
 48%|████▊     | 7035/14725 [6:24:50<7:09:28,  3.35s/it]
 48%|████▊     | 7036/14725 [6:24:53<7:04:18,  3.31s/it]
 48%|████▊     | 7037/14725 [6:24:56<6:54:58,  3.24s/it]
 48%|████▊     | 7038/14725 [6:25:00<7:00:31,  3.28s/it]
 48%|████▊     | 7039/14725 [6:25:03<7:04:11,  3.31s/it]
 48%|████▊     | 7040/14725 [6:25:06<7:03:17,  3.30s/it]
                                                        

 48%|████▊     | 7040/14725 [6:25:06<7:03:17,  3.30s/it]
 48%|████▊     | 7041/14725 [6:25:10<7:02:51,  3.30s/it]
 48%|████▊     | 7042/14725 [6:25:13<7:09:49,  3.36s/it]
 48%|████▊     | 7043/14725 [6:25:16<7:05:46,  3.33s/it]
 48%|████▊     | 7044/14725 [6:25:20<7:14:30,  3.39s/it]
 48%|████▊     | 7045/14725 [6:25:23<7:08:44,  3.35s/it]
 48%|████▊     | 7046/14725 [6:25:26<7:03:41,  3.31s/it]
 48%|████▊     | 7047/147
0: {'loss': 0.7067, 'grad_norm': 0.42601802561843194, 'learning_rate': 1e-05, 'epoch': 0.48}
0: 25 [6:25:30<6:58:46,  3.27s/it]
 48%|████▊     | 7048/14725 [6:25:33<6:54:14,  3.24s/it]
 48%|████▊     | 7049/14725 [6:25:36<6:55:59,  3.25s/it]
 48%|████▊     | 7050/14725 [6:25:39<6:52:47,  3.23s/it]
                                                        

 48%|████▊     | 7050/14725 [6:25:39<6:52:47,  3.23s/it]
 48%|████▊     | 7051/14725 [6:25:43<6:56:09,  3.25s/it]
 48%|████▊     | 7052/14725 [6:25:46<6:54:16,  3.24s/it]
 48%|████▊     | 7053/14725 [6:25:49<6:49:14,  3.20s/it]
 48%|████▊     | 7054/14725 [6:25:52<6:52:06,  3.22s/it]
 48%|████▊     | 7055/14725 [6:25:56<6:59:50,  3.28s/it]
 48%|████▊     | 7056/14725 [6:25:59<7:02:05,  3.30s/it]
 48%|████▊     | 7057/14725 [6:26:02<7:01:04,  3.29s/it]
 48%|████▊     | 7058/14725 [6:26:06<7:04:32,  3.32s/it]
 48%|████▊     | 7059/14725 [6:26:09<6:59:23,  3.28s/it]
 48%|████▊     | 7060/14725 [6:26:12<7:04:43,  3.32s/
0: {'loss': 0.7086, 'grad_norm': 0.42537271573778607, 'learning_rate': 1e-05, 'epoch': 0.48}
0: {'loss': 0.7063, 'grad_norm': 0.42597797218125544, 'learning_rate': 1e-05, 'epoch': 0.48}
0: it]
                                                        

 48%|████▊     | 7060/14725 [6:26:12<7:04:43,  3.32s/it]
 48%|████▊     | 7061/14725 [6:26:16<7:06:58,  3.34s/it]
 48%|████▊     | 7062/14725 [6:26:19<7:03:42,  3.32s/it]
 48%|████▊     | 7063/14725 [6:26:22<7:01:27,  3.30s/it]
 48%|████▊     | 7064/14725 [6:26:26<7:08:02,  3.35s/it]
 48%|████▊     | 7065/14725 [6:26:29<7:03:39,  3.32s/it]
 48%|████▊     | 7066/14725 [6:26:32<7:05:54,  3.34s/it]
 48%|████▊     | 7067/14725 [6:26:35<7:03:26,  3.32s/it]
 48%|████▊     | 7068/14725 [6:26:39<6:54:13,  3.25s/it]
 48%|████▊     | 7069/14725 [6:26:42<6:53:33,  3.24s/it]
 48%|████▊     | 7070/14725 [6:26:45<6:58:34,  3.28s/it]
                                                        

 48%|████▊     | 7070/14725 [6:26:45<6:58:34,  3.28s/it]
 48%|████▊     | 7071/14725 [6:26:49<7:05:41,  3.34s/it]
 48%|████▊     | 7072/1
0: {'loss': 0.6998, 'grad_norm': 0.44328852147369, 'learning_rate': 1e-05, 'epoch': 0.48}
0: 4725 [6:26:52<7:02:21,  3.31s/it]
 48%|████▊     | 7073/14725 [6:26:55<6:59:38,  3.29s/it]
 48%|████▊     | 7074/14725 [6:26:58<6:57:15,  3.27s/it]
 48%|████▊     | 7075/14725 [6:27:01<6:49:37,  3.21s/it]
 48%|████▊     | 7076/14725 [6:27:05<6:48:27,  3.20s/it]
 48%|████▊     | 7077/14725 [6:27:08<6:45:51,  3.18s/it]
 48%|████▊     | 7078/14725 [6:27:11<6:46:46,  3.19s/it]
 48%|████▊     | 7079/14725 [6:27:14<6:42:40,  3.16s/it]
 48%|████▊     | 7080/14725 [6:27:17<6:43:57,  3.17s/it]
                                                        

 48%|████▊     | 7080/14725 [6:27:17<6:43:57,  3.17s/it]
 48%|████▊     | 7081/14725 [6:27:21<6:57:13,  3.27s/it]
 48%|████▊     | 7082/14725 [6:27:24<6:58:43,  3.29s/it]
 48%|████▊     | 7083/14725 [6:27:27<7:05:31,  3.34s/it]
 48%|████▊     | 7084/14725 [6:27:31<7:00:26,  3.30s/it]
 48%|████▊     | 7085/14725 [6:27:34<6:54:11,  3.25
0: {'loss': 0.6888, 'grad_norm': 0.3914188745761841, 'learning_rate': 1e-05, 'epoch': 0.48}
0: s/it]
 48%|████▊     | 7086/14725 [6:27:37<6:57:21,  3.28s/it]
 48%|████▊     | 7087/14725 [6:27:40<6:57:29,  3.28s/it]
 48%|████▊     | 7088/14725 [6:27:44<6:57:57,  3.28s/it]
 48%|████▊     | 7089/14725 [6:27:47<6:59:47,  3.30s/it]
 48%|████▊     | 7090/14725 [6:27:50<7:00:30,  3.30s/it]
                                                        

 48%|████▊     | 7090/14725 [6:27:50<7:00:30,  3.30s/it]
 48%|████▊     | 7091/14725 [6:27:54<6:56:49,  3.28s/it]
 48%|████▊     | 7092/14725 [6:27:57<6:55:47,  3.27s/it]
 48%|████▊     | 7093/14725 [6:28:00<7:04:34,  3.34s/it]
 48%|████▊     | 7094/14725 [6:28:04<7:03:11,  3.33s/it]
 48%|████▊     | 7095/14725 [6:28:07<6:52:19,  3.24s/it]
 48%|████▊     | 7096/14725 [6:28:10<6:58:43,  3.29s/it]
 48%|████▊     | 7097/14725 [6:28:13<6:49:15,  3.22s/it]
 48%|████▊     | 7098/14725 [6:28:16<6:52:49,  3.25s/it]
 48%|████▊  
0: {'loss': 0.714, 'grad_norm': 0.4004845996856976, 'learning_rate': 1e-05, 'epoch': 0.48}
0: {'loss': 0.6999, 'grad_norm': 0.43028750622031314, 'learning_rate': 1e-05, 'epoch': 0.48}
0:    | 7099/14725 [6:28:20<6:53:25,  3.25s/it]
 48%|████▊     | 7100/14725 [6:28:23<6:53:29,  3.25s/it]
                                                        

 48%|████▊     | 7100/14725 [6:28:23<6:53:29,  3.25s/it]
 48%|████▊     | 7101/14725 [6:28:26<6:54:49,  3.26s/it]
 48%|████▊     | 7102/14725 [6:28:30<6:55:07,  3.27s/it]
 48%|████▊     | 7103/14725 [6:28:33<6:48:32,  3.22s/it]
 48%|████▊     | 7104/14725 [6:28:36<6:48:57,  3.22s/it]
 48%|████▊     | 7105/14725 [6:28:39<6:52:32,  3.25s/it]
 48%|████▊     | 7106/14725 [6:28:43<6:56:31,  3.28s/it]
 48%|████▊     | 7107/14725 [6:28:46<6:57:45,  3.29s/it]
 48%|████▊     | 7108/14725 [6:28:49<7:00:00,  3.31s/it]
 48%|████▊     | 7109/14725 [6:28:52<6:54:31,  3.27s/it]
 48%|████▊     | 7110/14725 [6:28:56<6:58:09,  3.29s/it]
                                                        

 48%|████▊     | 7110/14725 [6:28:56<6:58:09,  3.
0: {'loss': 0.7014, 'grad_norm': 0.5044590388008448, 'learning_rate': 1e-05, 'epoch': 0.48}
0: 29s/it]
 48%|████▊     | 7111/14725 [6:28:59<6:50:49,  3.24s/it]
 48%|████▊     | 7112/14725 [6:29:02<6:47:19,  3.21s/it]
 48%|████▊     | 7113/14725 [6:29:05<6:41:30,  3.16s/it]
 48%|████▊     | 7114/14725 [6:29:08<6:37:08,  3.13s/it]
 48%|████▊     | 7115/14725 [6:29:11<6:40:05,  3.15s/it]
 48%|████▊     | 7116/14725 [6:29:14<6:37:34,  3.13s/it]
 48%|████▊     | 7117/14725 [6:29:18<6:40:39,  3.16s/it]
 48%|████▊     | 7118/14725 [6:29:21<6:44:14,  3.19s/it]
 48%|████▊     | 7119/14725 [6:29:24<6:53:55,  3.27s/it]
 48%|████▊     | 7120/14725 [6:29:28<6:58:44,  3.30s/it]
                                                        

 48%|████▊     | 7120/14725 [6:29:28<6:58:44,  3.30s/it]
 48%|████▊     | 7121/14725 [6:29:31<6:56:55,  3.29s/it]
 48%|████▊     | 7122/14725 [6:29:34<6:56:57,  3.29s/it]
 48%|████▊     | 7123/14725 [6:29:38<7:00:34,  3.32s/it]
 48%|████▊
0: {'loss': 0.7171, 'grad_norm': 0.46769937831124053, 'learning_rate': 1e-05, 'epoch': 0.48}
0:      | 7124/14725 [6:29:41<6:56:12,  3.29s/it]
 48%|████▊     | 7125/14725 [6:29:44<6:57:13,  3.29s/it]
 48%|████▊     | 7126/14725 [6:29:48<6:59:34,  3.31s/it]
 48%|████▊     | 7127/14725 [6:29:51<7:01:58,  3.33s/it]
 48%|████▊     | 7128/14725 [6:29:54<6:55:54,  3.28s/it]
 48%|████▊     | 7129/14725 [6:29:57<6:58:06,  3.30s/it]
 48%|████▊     | 7130/14725 [6:30:01<7:01:54,  3.33s/it]
                                                        

 48%|████▊     | 7130/14725 [6:30:01<7:01:54,  3.33s/it]
 48%|████▊     | 7131/14725 [6:30:04<6:56:33,  3.29s/it]
 48%|████▊     | 7132/14725 [6:30:07<6:55:56,  3.29s/it]
 48%|████▊     | 7133/14725 [6:30:10<6:46:19,  3.21s/it]
 48%|████▊     | 7134/14725 [6:30:14<6:50:33,  3.25s/it]
 48%|████▊     | 7135/14725 [6:30:17<6:51:04,  3.25s/it]
 48%|████▊     | 7136/14725 [6:30:20<6:52:51,  3.26s/it]
 48%|████▊     | 7137/14725 [6:30:23<6
0: {'loss': 0.6867, 'grad_norm': 0.4394248552996047, 'learning_rate': 1e-05, 'epoch': 0.48}
0: :47:09,  3.22s/it]
 48%|████▊     | 7138/14725 [6:30:27<6:44:59,  3.20s/it]
 48%|████▊     | 7139/14725 [6:30:30<6:48:57,  3.23s/it]
 48%|████▊     | 7140/14725 [6:30:33<6:47:52,  3.23s/it]
                                                        

 48%|████▊     | 7140/14725 [6:30:33<6:47:52,  3.23s/it]
 48%|████▊     | 7141/14725 [6:30:36<6:53:56,  3.27s/it]
 49%|████▊     | 7142/14725 [6:30:40<6:59:01,  3.32s/it]
 49%|████▊     | 7143/14725 [6:30:43<7:01:28,  3.34s/it]
 49%|████▊     | 7144/14725 [6:30:46<6:58:09,  3.31s/it]
 49%|████▊     | 7145/14725 [6:30:50<6:58:02,  3.31s/it]
 49%|████▊     | 7146/14725 [6:30:53<6:54:17,  3.28s/it]
 49%|████▊     | 7147/14725 [6:30:56<6:45:34,  3.21s/it]
 49%|████▊     | 7148/14725 [6:31:00<6:59:49,  3.32s/it]
 49%|████▊     | 7149/14725 [6:31:03<6:56:44,  3.30s/it]
 49%|████▊     | 7150/14725 [6:31:06<6:49:24,  3.24s/it]
         
0: {'loss': 0.6872, 'grad_norm': 0.41150182997632123, 'learning_rate': 1e-05, 'epoch': 0.49}
0: {'loss': 0.7094, 'grad_norm': 0.4180060538399622, 'learning_rate': 1e-05, 'epoch': 0.49}
0:                                                

 49%|████▊     | 7150/14725 [6:31:06<6:49:24,  3.24s/it]
 49%|████▊     | 7151/14725 [6:31:09<6:56:48,  3.30s/it]
 49%|████▊     | 7152/14725 [6:31:13<6:56:49,  3.30s/it]
 49%|████▊     | 7153/14725 [6:31:16<6:59:07,  3.32s/it]
 49%|████▊     | 7154/14725 [6:31:19<6:53:43,  3.28s/it]
 49%|████▊     | 7155/14725 [6:31:22<6:49:02,  3.24s/it]
 49%|████▊     | 7156/14725 [6:31:26<6:44:00,  3.20s/it]
 49%|████▊     | 7157/14725 [6:31:29<6:51:16,  3.26s/it]
 49%|████▊     | 7158/14725 [6:31:32<6:43:37,  3.20s/it]
 49%|████▊     | 7159/14725 [6:31:35<6:44:08,  3.20s/it]
 49%|████▊     | 7160/14725 [6:31:38<6:44:20,  3.21s/it]
                                                        

 49%|████▊     | 7160/14725 [6:31:38<6:44:20,  3.21s/it]
 49%|████▊     | 7161/14725 [6:31:42<6:42:07,  3.19s/it]
 49%|████▊     | 7162/14725 [6:31:45
0: {'loss': 0.7211, 'grad_norm': 0.4337986082787264, 'learning_rate': 1e-05, 'epoch': 0.49}
0: <6:49:53,  3.25s/it]
 49%|████▊     | 7163/14725 [6:31:48<6:43:55,  3.20s/it]
 49%|████▊     | 7164/14725 [6:31:51<6:43:53,  3.21s/it]
 49%|████▊     | 7165/14725 [6:31:55<6:48:14,  3.24s/it]
 49%|████▊     | 7166/14725 [6:31:58<6:50:34,  3.26s/it]
 49%|████▊     | 7167/14725 [6:32:01<6:53:34,  3.28s/it]
 49%|████▊     | 7168/14725 [6:32:04<6:49:21,  3.25s/it]
 49%|████▊     | 7169/14725 [6:32:08<6:52:37,  3.28s/it]
 49%|████▊     | 7170/14725 [6:32:11<6:49:09,  3.25s/it]
                                                        

 49%|████▊     | 7170/14725 [6:32:11<6:49:09,  3.25s/it]
 49%|████▊     | 7171/14725 [6:32:14<6:49:38,  3.25s/it]
 49%|████▊     | 7172/14725 [6:32:17<6:50:10,  3.26s/it]
 49%|████▊     | 7173/14725 [6:32:21<6:53:40,  3.29s/it]
 49%|████▊     | 7174/14725 [6:32:24<6:56:11,  3.31s/it]
 49%|████▊     | 7175/14725 [6:32:27<6:54:17,  3.29s/it]
 49%|�
0: {'loss': 0.6917, 'grad_norm': 0.4148280999835485, 'learning_rate': 1e-05, 'epoch': 0.49}
0: �███▊     | 7176/14725 [6:32:31<6:50:56,  3.27s/it]
 49%|████▊     | 7177/14725 [6:32:34<6:54:16,  3.29s/it]
 49%|████▊     | 7178/14725 [6:32:38<7:03:27,  3.37s/it]
 49%|████▉     | 7179/14725 [6:32:41<6:54:48,  3.30s/it]
 49%|████▉     | 7180/14725 [6:32:44<6:45:14,  3.22s/it]
                                                        

 49%|████▉     | 7180/14725 [6:32:44<6:45:14,  3.22s/it]
 49%|████▉     | 7181/14725 [6:32:47<6:44:53,  3.22s/it]
 49%|████▉     | 7182/14725 [6:32:50<6:39:27,  3.18s/it]
 49%|████▉     | 7183/14725 [6:32:53<6:31:57,  3.12s/it]
 49%|████▉     | 7184/14725 [6:32:56<6:36:35,  3.16s/it]
 49%|████▉     | 7185/14725 [6:33:00<6:47:37,  3.24s/it]
 49%|████▉     | 7186/14725 [6:33:03<6:45:04,  3.22s/it]
 49%|████▉     | 7187/14725 [6:33:06<6:48:07,  3.25s/it]
 49%|████▉     | 7188/14725 [6:33:09<6:48:40,  3.25s/it]
 49%|████▉     | 7189/147
0: {'loss': 0.6881, 'grad_norm': 0.402376875403535, 'learning_rate': 1e-05, 'epoch': 0.49}
0: {'loss': 0.7047, 'grad_norm': 0.39806635504382454, 'learning_rate': 1e-05, 'epoch': 0.49}
0: 25 [6:33:13<6:45:38,  3.23s/it]
 49%|████▉     | 7190/14725 [6:33:16<6:45:21,  3.23s/it]
                                                        

 49%|████▉     | 7190/14725 [6:33:16<6:45:21,  3.23s/it]
 49%|████▉     | 7191/14725 [6:33:19<6:41:49,  3.20s/it]
 49%|████▉     | 7192/14725 [6:33:22<6:47:18,  3.24s/it]
 49%|████▉     | 7193/14725 [6:33:26<6:49:01,  3.26s/it]
 49%|████▉     | 7194/14725 [6:33:29<6:53:08,  3.29s/it]
 49%|████▉     | 7195/14725 [6:33:32<6:55:36,  3.31s/it]
 49%|████▉     | 7196/14725 [6:33:36<7:00:03,  3.35s/it]
 49%|████▉     | 7197/14725 [6:33:39<6:58:52,  3.34s/it]
 49%|████▉     | 7198/14725 [6:33:42<7:00:36,  3.35s/it]
 49%|████▉     | 7199/14725 [6:33:46<7:02:55,  3.37s/it]
 49%|████▉     | 7200/14725 [6:33:49<6:54:32,  3.31s/it]
                                                        

 49%|████▉     | 7200/14725 [6:33:49<6:54:32,  3.31s/it]
 49%|
0: {'loss': 0.695, 'grad_norm': 0.4364138390369041, 'learning_rate': 1e-05, 'epoch': 0.49}
0: ████▉     | 7201/14725 [6:33:52<6:56:52,  3.32s/it]
 49%|████▉     | 7202/14725 [6:33:56<6:59:47,  3.35s/it]
 49%|████▉     | 7203/14725 [6:33:59<7:03:55,  3.38s/it]
 49%|████▉     | 7204/14725 [6:34:02<6:51:44,  3.28s/it]
 49%|████▉     | 7205/14725 [6:34:05<6:42:58,  3.22s/it]
 49%|████▉     | 7206/14725 [6:34:09<6:47:24,  3.25s/it]
 49%|████▉     | 7207/14725 [6:34:12<6:44:57,  3.23s/it]
 49%|████▉     | 7208/14725 [6:34:15<6:49:48,  3.27s/it]
 49%|████▉     | 7209/14725 [6:34:18<6:47:08,  3.25s/it]
 49%|████▉     | 7210/14725 [6:34:22<6:54:15,  3.31s/it]
                                                        

 49%|████▉     | 7210/14725 [6:34:22<6:54:15,  3.31s/it]
 49%|████▉     | 7211/14725 [6:34:25<6:45:58,  3.24s/it]
 49%|████▉     | 7212/14725 [6:34:28<6:43:54,  3.23s/it]
 49%|████▉     | 7213/14725 [6:34:31<6:37:58,  3.18s/it]
 49%|████▉     | 7214/1
0: {'loss': 0.6721, 'grad_norm': 0.41314790301086984, 'learning_rate': 1e-05, 'epoch': 0.49}
0: 4725 [6:34:35<6:45:30,  3.24s/it]
 49%|████▉     | 7215/14725 [6:34:38<6:44:55,  3.24s/it]
 49%|████▉     | 7216/14725 [6:34:41<6:42:44,  3.22s/it]
 49%|████▉     | 7217/14725 [6:34:44<6:42:52,  3.22s/it]
 49%|████▉     | 7218/14725 [6:34:47<6:43:38,  3.23s/it]
 49%|████▉     | 7219/14725 [6:34:51<6:43:52,  3.23s/it]
 49%|████▉     | 7220/14725 [6:34:54<6:39:35,  3.19s/it]
                                                        

 49%|████▉     | 7220/14725 [6:34:54<6:39:35,  3.19s/it]
 49%|████▉     | 7221/14725 [6:34:57<6:36:06,  3.17s/it]
 49%|████▉     | 7222/14725 [6:35:00<6:37:06,  3.18s/it]
 49%|████▉     | 7223/14725 [6:35:03<6:40:53,  3.21s/it]
 49%|████▉     | 7224/14725 [6:35:07<6:41:07,  3.21s/it]
 49%|████▉     | 7225/14725 [6:35:10<6:51:01,  3.29s/it]
 49%|████▉     | 7226/14725 [6:35:13<6:50:10,  3.28s/it]
 49%|████▉     | 7227/14725 [6:35:17<7:00:17,  3.36
0: {'loss': 0.7131, 'grad_norm': 0.41784782683356914, 'learning_rate': 1e-05, 'epoch': 0.49}
0: s/it]
 49%|████▉     | 7228/14725 [6:35:20<6:53:28,  3.31s/it]
 49%|████▉     | 7229/14725 [6:35:23<6:49:20,  3.28s/it]
 49%|████▉     | 7230/14725 [6:35:26<6:41:57,  3.22s/it]
                                                        

 49%|████▉     | 7230/14725 [6:35:26<6:41:57,  3.22s/it]
 49%|████▉     | 7231/14725 [6:35:30<6:49:32,  3.28s/it]
 49%|████▉     | 7232/14725 [6:35:33<6:44:20,  3.24s/it]
 49%|████▉     | 7233/14725 [6:35:36<6:50:23,  3.29s/it]
 49%|████▉     | 7234/14725 [6:35:40<6:46:02,  3.25s/it]
 49%|████▉     | 7235/14725 [6:35:43<6:48:57,  3.28s/it]
 49%|████▉     | 7236/14725 [6:35:46<6:52:04,  3.30s/it]
 49%|████▉     | 7237/14725 [6:35:50<6:59:13,  3.36s/it]
 49%|████▉     | 7238/14725 [6:35:53<6:51:07,  3.29s/it]
 49%|████▉     | 7239/14725 [6:35:56<6:50:59,  3.29s/it]
 49%|████▉     | 7240/14725 [6:35:59<6:50:33,  3.29s/it]
                      
0: {'loss': 0.6912, 'grad_norm': 0.40914428972814093, 'learning_rate': 1e-05, 'epoch': 0.49}
0: {'loss': 0.7017, 'grad_norm': 0.44523428194238895, 'learning_rate': 1e-05, 'epoch': 0.49}
0:                                   

 49%|████▉     | 7240/14725 [6:35:59<6:50:33,  3.29s/it]
 49%|████▉     | 7241/14725 [6:36:03<6:43:31,  3.24s/it]
 49%|████▉     | 7242/14725 [6:36:06<6:46:09,  3.26s/it]
 49%|████▉     | 7243/14725 [6:36:09<6:42:14,  3.23s/it]
 49%|████▉     | 7244/14725 [6:36:12<6:37:07,  3.19s/it]
 49%|████▉     | 7245/14725 [6:36:16<6:50:20,  3.29s/it]
 49%|████▉     | 7246/14725 [6:36:19<6:50:01,  3.29s/it]
 49%|████▉     | 7247/14725 [6:36:22<6:46:07,  3.26s/it]
 49%|████▉     | 7248/14725 [6:36:25<6:41:30,  3.22s/it]
 49%|████▉     | 7249/14725 [6:36:28<6:40:37,  3.22s/it]
 49%|████▉     | 7250/14725 [6:36:32<6:38:38,  3.20s/it]
                                                        

 49%|████▉     | 7250/14725 [6:36:32<6:38:38,  3.20s/it]
 49%|████▉     | 7251/14725 [6:36:35<6:36:56,  3.19s/it]
 49%|████▉     | 7252/14725 [6:36:38<6:33:17,  3.
0: {'loss': 0.7007, 'grad_norm': 0.4108002313153059, 'learning_rate': 1e-05, 'epoch': 0.49}
0: 16s/it]
 49%|████▉     | 7253/14725 [6:36:41<6:41:40,  3.23s/it]
 49%|████▉     | 7254/14725 [6:36:44<6:42:14,  3.23s/it]
 49%|████▉     | 7255/14725 [6:36:48<6:42:22,  3.23s/it]
 49%|████▉     | 7256/14725 [6:36:51<6:36:20,  3.18s/it]
 49%|████▉     | 7257/14725 [6:36:54<6:38:02,  3.20s/it]
 49%|████▉     | 7258/14725 [6:36:57<6:39:09,  3.21s/it]
 49%|████▉     | 7259/14725 [6:37:00<6:40:18,  3.22s/it]
 49%|████▉     | 7260/14725 [6:37:04<6:39:38,  3.21s/it]
                                                        

 49%|████▉     | 7260/14725 [6:37:04<6:39:38,  3.21s/it]
 49%|████▉     | 7261/14725 [6:37:07<6:44:58,  3.26s/it]
 49%|████▉     | 7262/14725 [6:37:10<6:51:18,  3.31s/it]
 49%|████▉     | 7263/14725 [6:37:14<6:51:27,  3.31s/it]
 49%|████▉     | 7264/14725 [6:37:17<6:51:10,  3.31s/it]
 49%|████▉     | 7265/14725 [6:37:20<6:46:13,  3.27s/it]
 49%|████▉
0: {'loss': 0.6831, 'grad_norm': 0.4323531643898529, 'learning_rate': 1e-05, 'epoch': 0.49}
0:      | 7266/14725 [6:37:24<6:48:00,  3.28s/it]
 49%|████▉     | 7267/14725 [6:37:27<6:53:23,  3.33s/it]
 49%|████▉     | 7268/14725 [6:37:30<6:47:44,  3.28s/it]
 49%|████▉     | 7269/14725 [6:37:33<6:41:08,  3.23s/it]
 49%|████▉     | 7270/14725 [6:37:37<6:43:58,  3.25s/it]
                                                        

 49%|████▉     | 7270/14725 [6:37:37<6:43:58,  3.25s/it]
 49%|████▉     | 7271/14725 [6:37:40<6:41:10,  3.23s/it]
 49%|████▉     | 7272/14725 [6:37:43<6:37:14,  3.20s/it]
 49%|████▉     | 7273/14725 [6:37:46<6:40:30,  3.22s/it]
 49%|████▉     | 7274/14725 [6:37:49<6:44:17,  3.26s/it]
 49%|████▉     | 7275/14725 [6:37:53<6:51:53,  3.32s/it]
 49%|████▉     | 7276/14725 [6:37:56<6:47:29,  3.28s/it]
 49%|████▉     | 7277/14725 [6:38:00<6:52:43,  3.32s/it]
 49%|████▉     | 7278/14725 [6:38:03<6:47:28,  3.28s/it]
 49%|████▉     | 7279/14725 [6:38:06<6
0: {'loss': 0.6947, 'grad_norm': 0.4268061306576738, 'learning_rate': 1e-05, 'epoch': 0.49}
0: {'loss': 0.7093, 'grad_norm': 0.41901151916476487, 'learning_rate': 1e-05, 'epoch': 0.5}
0: :49:49,  3.30s/it]
 49%|████▉     | 7280/14725 [6:38:10<6:56:03,  3.35s/it]
                                                        

 49%|████▉     | 7280/14725 [6:38:10<6:56:03,  3.35s/it]
 49%|████▉     | 7281/14725 [6:38:13<6:52:23,  3.32s/it]
 49%|████▉     | 7282/14725 [6:38:16<6:53:00,  3.33s/it]
 49%|████▉     | 7283/14725 [6:38:19<6:52:24,  3.32s/it]
 49%|████▉     | 7284/14725 [6:38:23<6:47:38,  3.29s/it]
 49%|████▉     | 7285/14725 [6:38:26<6:41:14,  3.24s/it]
 49%|████▉     | 7286/14725 [6:38:29<6:40:50,  3.23s/it]
 49%|████▉     | 7287/14725 [6:38:32<6:46:28,  3.28s/it]
 49%|████▉     | 7288/14725 [6:38:36<6:43:57,  3.26s/it]
 50%|████▉     | 7289/14725 [6:38:39<6:41:32,  3.24s/it]
 50%|████▉     | 7290/14725 [6:38:42<6:44:33,  3.26s/it]
                                                        

 50%|████▉     | 7290/14725 [6:38:42<6:44:33,  3.26s/it]
 50%|████�
0: {'loss': 0.6998, 'grad_norm': 0.441256336447556, 'learning_rate': 1e-05, 'epoch': 0.5}
0: ��     | 7291/14725 [6:38:45<6:42:10,  3.25s/it]
 50%|████▉     | 7292/14725 [6:38:49<6:46:44,  3.28s/it]
 50%|████▉     | 7293/14725 [6:38:52<6:49:15,  3.30s/it]
 50%|████▉     | 7294/14725 [6:38:56<6:59:36,  3.39s/it]
 50%|████▉     | 7295/14725 [6:38:59<6:57:06,  3.37s/it]
 50%|████▉     | 7296/14725 [6:39:03<7:14:31,  3.51s/it]
 50%|████▉     | 7297/14725 [6:39:06<7:05:42,  3.44s/it]
 50%|████▉     | 7298/14725 [6:39:09<7:00:31,  3.40s/it]
 50%|████▉     | 7299/14725 [6:39:13<7:04:59,  3.43s/it]
 50%|████▉     | 7300/14725 [6:39:16<7:03:29,  3.42s/it]
                                                        

 50%|████▉     | 7300/14725 [6:39:16<7:03:29,  3.42s/it]
 50%|████▉     | 7301/14725 [6:39:20<6:57:40,  3.38s/it]
 50%|████▉     | 7302/14725 [6:39:23<6:49:57,  3.31s/it]
 50%|████▉     | 7303/14725 [6:39:26<6:45:48,  3.28s/it]
 50%|████▉     | 7304/14725 [6:39:29
0: {'loss': 0.695, 'grad_norm': 0.42648633753880816, 'learning_rate': 1e-05, 'epoch': 0.5}
0: <6:45:48,  3.28s/it]
 50%|████▉     | 7305/14725 [6:39:33<7:05:02,  3.44s/it]
 50%|████▉     | 7306/14725 [6:39:36<6:54:11,  3.35s/it]
 50%|████▉     | 7307/14725 [6:39:39<6:43:37,  3.26s/it]
 50%|████▉     | 7308/14725 [6:39:43<6:44:26,  3.27s/it]
 50%|████▉     | 7309/14725 [6:39:46<6:44:27,  3.27s/it]
 50%|████▉     | 7310/14725 [6:39:49<6:49:04,  3.31s/it]
                                                        

 50%|████▉     | 7310/14725 [6:39:49<6:49:04,  3.31s/it]
 50%|████▉     | 7311/14725 [6:39:53<6:55:00,  3.36s/it]
 50%|████▉     | 7312/14725 [6:39:56<6:44:00,  3.27s/it]
 50%|████▉     | 7313/14725 [6:39:59<6:43:35,  3.27s/it]
 50%|████▉     | 7314/14725 [6:40:02<6:45:40,  3.28s/it]
 50%|████▉     | 7315/14725 [6:40:06<6:43:20,  3.27s/it]
 50%|████▉     | 7316/14725 [6:40:09<6:41:29,  3.25s/it]
 50%|████▉     | 7317/14725 [6:40:12<6:41:12,  3.25s/it]
 50%|�
0: {'loss': 0.7057, 'grad_norm': 0.40453494373943855, 'learning_rate': 1e-05, 'epoch': 0.5}
0: �███▉     | 7318/14725 [6:40:15<6:41:35,  3.25s/it]
 50%|████▉     | 7319/14725 [6:40:19<6:41:02,  3.25s/it]
 50%|████▉     | 7320/14725 [6:40:22<6:37:23,  3.22s/it]
                                                        

 50%|████▉     | 7320/14725 [6:40:22<6:37:23,  3.22s/it]
 50%|████▉     | 7321/14725 [6:40:25<6:32:35,  3.18s/it]
 50%|████▉     | 7322/14725 [6:40:28<6:34:31,  3.20s/it]
 50%|████▉     | 7323/14725 [6:40:31<6:31:53,  3.18s/it]
 50%|████▉     | 7324/14725 [6:40:34<6:35:02,  3.20s/it]
 50%|████▉     | 7325/14725 [6:40:37<6:29:38,  3.16s/it]
 50%|████▉     | 7326/14725 [6:40:41<6:28:57,  3.15s/it]
 50%|████▉     | 7327/14725 [6:40:44<6:27:58,  3.15s/it]
 50%|████▉     | 7328/14725 [6:40:47<6:27:59,  3.15s/it]
 50%|████▉     | 7329/14725 [6:40:50<6:41:52,  3.26s/it]
 50%|████▉     | 7330/14725 [6:40:54<6:42:54,  3.27s/it]
                                   
0: {'loss': 0.6833, 'grad_norm': 0.42703970308342076, 'learning_rate': 1e-05, 'epoch': 0.5}
0: {'loss': 0.6986, 'grad_norm': 0.41225771693754654, 'learning_rate': 1e-05, 'epoch': 0.5}
0:                      

 50%|████▉     | 7330/14725 [6:40:54<6:42:54,  3.27s/it]
 50%|████▉     | 7331/14725 [6:40:57<6:51:36,  3.34s/it]
 50%|████▉     | 7332/14725 [6:41:01<6:54:07,  3.36s/it]
 50%|████▉     | 7333/14725 [6:41:04<7:01:57,  3.42s/it]
 50%|████▉     | 7334/14725 [6:41:07<6:54:32,  3.37s/it]
 50%|████▉     | 7335/14725 [6:41:11<6:49:18,  3.32s/it]
 50%|████▉     | 7336/14725 [6:41:14<6:45:37,  3.29s/it]
 50%|████▉     | 7337/14725 [6:41:17<6:43:11,  3.27s/it]
 50%|████▉     | 7338/14725 [6:41:20<6:48:14,  3.32s/it]
 50%|████▉     | 7339/14725 [6:41:24<6:41:55,  3.27s/it]
 50%|████▉     | 7340/14725 [6:41:27<6:39:27,  3.25s/it]
                                                        

 50%|████▉     | 7340/14725 [6:41:27<6:39:27,  3.25s/it]
 50%|████▉     | 7341/14725 [6:41:30<6:42:04,  3.27s/it]
 50%|████▉     | 7342/14725 [6:41:33<6:38:09,  3.24s/it]
 50%|
0: {'loss': 0.6972, 'grad_norm': 0.4483778535145902, 'learning_rate': 1e-05, 'epoch': 0.5}
0: ████▉     | 7343/14725 [6:41:37<6:53:30,  3.36s/it]
 50%|████▉     | 7344/14725 [6:41:40<6:46:52,  3.31s/it]
 50%|████▉     | 7345/14725 [6:41:43<6:38:03,  3.24s/it]
 50%|████▉     | 7346/14725 [6:41:46<6:30:48,  3.18s/it]
 50%|████▉     | 7347/14725 [6:41:50<6:36:57,  3.23s/it]
 50%|████▉     | 7348/14725 [6:41:53<6:39:58,  3.25s/it]
 50%|████▉     | 7349/14725 [6:41:56<6:45:56,  3.30s/it]
 50%|████▉     | 7350/14725 [6:42:00<6:48:12,  3.32s/it]
                                                        

 50%|████▉     | 7350/14725 [6:42:00<6:48:12,  3.32s/it]
 50%|████▉     | 7351/14725 [6:42:03<6:41:18,  3.27s/it]
 50%|████▉     | 7352/14725 [6:42:06<6:36:39,  3.23s/it]
 50%|████▉     | 7353/14725 [6:42:09<6:38:20,  3.24s/it]
 50%|████▉     | 7354/14725 [6:42:12<6:37:37,  3.24s/it]
 50%|████▉     | 7355/14725 [6:42:16<6:40:31,  3.26s/it]
 50%|████▉     | 7356/1
0: {'loss': 0.6863, 'grad_norm': 0.3977629282437327, 'learning_rate': 1e-05, 'epoch': 0.5}
0: 4725 [6:42:19<6:37:24,  3.24s/it]
 50%|████▉     | 7357/14725 [6:42:22<6:44:29,  3.29s/it]
 50%|████▉     | 7358/14725 [6:42:26<6:52:21,  3.36s/it]
 50%|████▉     | 7359/14725 [6:42:29<6:50:38,  3.34s/it]
 50%|████▉     | 7360/14725 [6:42:32<6:39:58,  3.26s/it]
                                                        

 50%|████▉     | 7360/14725 [6:42:32<6:39:58,  3.26s/it]
 50%|████▉     | 7361/14725 [6:42:36<6:42:23,  3.28s/it]
 50%|████▉     | 7362/14725 [6:42:39<6:44:13,  3.29s/it]
 50%|█████     | 7363/14725 [6:42:42<6:36:40,  3.23s/it]
 50%|█████     | 7364/14725 [6:42:45<6:34:14,  3.21s/it]
 50%|█████     | 7365/14725 [6:42:48<6:30:41,  3.18s/it]
 50%|█████     | 7366/14725 [6:42:52<6:33:22,  3.21s/it]
 50%|█████     | 7367/14725 [6:42:55<6:39:08,  3.25s/it]
 50%|█████     | 7368/14725 [6:42:58<6:40:27,  3.27s/it]
 50%|█████     | 7369/14725 [6:43:02<6:46:29,  3.32
0: {'loss': 0.7016, 'grad_norm': 0.43871263542798206, 'learning_rate': 1e-05, 'epoch': 0.5}
0: {'loss': 0.6901, 'grad_norm': 0.46025368149860985, 'learning_rate': 1e-05, 'epoch': 0.5}
0: s/it]
 50%|█████     | 7370/14725 [6:43:05<6:43:05,  3.29s/it]
                                                        

 50%|█████     | 7370/14725 [6:43:05<6:43:05,  3.29s/it]
 50%|█████     | 7371/14725 [6:43:08<6:46:36,  3.32s/it]
 50%|█████     | 7372/14725 [6:43:12<6:48:20,  3.33s/it]
 50%|█████     | 7373/14725 [6:43:15<6:46:27,  3.32s/it]
 50%|█████     | 7374/14725 [6:43:18<6:40:31,  3.27s/it]
 50%|█████     | 7375/14725 [6:43:21<6:35:47,  3.23s/it]
 50%|█████     | 7376/14725 [6:43:24<6:37:46,  3.25s/it]
 50%|█████     | 7377/14725 [6:43:28<6:39:05,  3.26s/it]
 50%|█████     | 7378/14725 [6:43:31<6:47:11,  3.33s/it]
 50%|█████     | 7379/14725 [6:43:34<6:39:07,  3.26s/it]
 50%|█████     | 7380/14725 [6:43:38<6:51:08,  3.36s/it]
                                                        

 50%|█████     | 7380/14725 [6:43:38<6:51:08,  3.36s/it]
 50%|█████     | 7381
0: {'loss': 0.6794, 'grad_norm': 0.401863869867856, 'learning_rate': 1e-05, 'epoch': 0.5}
0: /14725 [6:43:41<6:44:10,  3.30s/it]
 50%|█████     | 7382/14725 [6:43:44<6:46:12,  3.32s/it]
 50%|█████     | 7383/14725 [6:43:48<6:45:55,  3.32s/it]
 50%|█████     | 7384/14725 [6:43:51<6:40:12,  3.27s/it]
 50%|█████     | 7385/14725 [6:43:54<6:38:18,  3.26s/it]
 50%|█████     | 7386/14725 [6:43:58<6:41:23,  3.28s/it]
 50%|█████     | 7387/14725 [6:44:01<6:42:35,  3.29s/it]
 50%|█████     | 7388/14725 [6:44:04<6:46:27,  3.32s/it]
 50%|█████     | 7389/14725 [6:44:07<6:38:28,  3.26s/it]
 50%|█████     | 7390/14725 [6:44:10<6:31:46,  3.20s/it]
                                                        

 50%|█████     | 7390/14725 [6:44:10<6:31:46,  3.20s/it]
 50%|█████     | 7391/14725 [6:44:14<6:38:25,  3.26s/it]
 50%|█████     | 7392/14725 [6:44:17<6:35:02,  3.23s/it]
 50%|█████     | 7393/14725 [6:44:20<6:31:41,  3.21s/it]
 50%|█████     | 7394/14725 [6:44:24<6:41:38,  3.
0: {'loss': 0.7032, 'grad_norm': 0.4149393513793494, 'learning_rate': 1e-05, 'epoch': 0.5}
0: 29s/it]
 50%|█████     | 7395/14725 [6:44:27<6:39:19,  3.27s/it]
 50%|█████     | 7396/14725 [6:44:30<6:40:47,  3.28s/it]
 50%|█████     | 7397/14725 [6:44:33<6:38:42,  3.26s/it]
 50%|█████     | 7398/14725 [6:44:37<6:43:15,  3.30s/it]
 50%|█████     | 7399/14725 [6:44:40<6:39:23,  3.27s/it]
 50%|█████     | 7400/14725 [6:44:43<6:35:23,  3.24s/it]
                                                        

 50%|█████     | 7400/14725 [6:44:43<6:35:23,  3.24s/it]
 50%|█████     | 7401/14725 [6:44:46<6:36:03,  3.24s/it]
 50%|█████     | 7402/14725 [6:44:50<6:34:34,  3.23s/it]
 50%|█████     | 7403/14725 [6:44:53<6:40:19,  3.28s/it]
 50%|█████     | 7404/14725 [6:44:56<6:43:39,  3.31s/it]
 50%|█████     | 7405/14725 [6:45:00<6:42:26,  3.30s/it]
 50%|█████     | 7406/14725 [6:45:03<6:40:23,  3.28s/it]
 50%|█████     | 7407/14725 [6:45:06<6:42:40,  3.30s/it]
 50%|█████
0: {'loss': 0.6764, 'grad_norm': 0.4083417611315556, 'learning_rate': 1e-05, 'epoch': 0.5}
0:      | 7408/14725 [6:45:10<6:49:37,  3.36s/it]
 50%|█████     | 7409/14725 [6:45:13<6:40:07,  3.28s/it]
 50%|█████     | 7410/14725 [6:45:16<6:35:44,  3.25s/it]
                                                        

 50%|█████     | 7410/14725 [6:45:16<6:35:44,  3.25s/it]
 50%|█████     | 7411/14725 [6:45:19<6:35:03,  3.24s/it]
 50%|█████     | 7412/14725 [6:45:22<6:32:07,  3.22s/it]
 50%|█████     | 7413/14725 [6:45:26<6:32:28,  3.22s/it]
 50%|█████     | 7414/14725 [6:45:29<6:34:10,  3.23s/it]
 50%|█████     | 7415/14725 [6:45:32<6:35:14,  3.24s/it]
 50%|█████     | 7416/14725 [6:45:35<6:39:16,  3.28s/it]
 50%|█████     | 7417/14725 [6:45:39<6:51:50,  3.38s/it]
 50%|█████     | 7418/14725 [6:45:43<6:53:13,  3.39s/it]
 50%|█████     | 7419/14725 [6:45:46<6:48:31,  3.35s/it]
 50%|█████     | 7420/14725 [6:45:49<6:39:38,  3.28s/it]
                                                
0: {'loss': 0.7073, 'grad_norm': 0.4071807424038854, 'learning_rate': 1e-05, 'epoch': 0.5}
0: {'loss': 0.683, 'grad_norm': 0.4285585804211312, 'learning_rate': 1e-05, 'epoch': 0.5}
0:         

 50%|█████     | 7420/14725 [6:45:49<6:39:38,  3.28s/it]
 50%|█████     | 7421/14725 [6:45:52<6:34:31,  3.24s/it]
 50%|█████     | 7422/14725 [6:45:55<6:36:04,  3.25s/it]
 50%|█████     | 7423/14725 [6:45:59<6:43:22,  3.31s/it]
 50%|█████     | 7424/14725 [6:46:02<6:50:24,  3.37s/it]
 50%|█████     | 7425/14725 [6:46:06<6:47:20,  3.35s/it]
 50%|█████     | 7426/14725 [6:46:09<6:46:55,  3.35s/it]
 50%|█████     | 7427/14725 [6:46:12<6:40:57,  3.30s/it]
 50%|█████     | 7428/14725 [6:46:16<6:48:00,  3.35s/it]
 50%|█████     | 7429/14725 [6:46:19<6:45:48,  3.34s/it]
 50%|█████     | 7430/14725 [6:46:22<6:37:57,  3.27s/it]
                                                        

 50%|█████     | 7430/14725 [6:46:22<6:37:57,  3.27s/it]
 50%|█████     | 7431/14725 [6:46:25<6:33:31,  3.24s/it]
 50%|█████     | 7432/14725 [6:46:28<6:29:29,  3.20s/it]
 50%|████�
0: {'loss': 0.6994, 'grad_norm': 0.4227736734510479, 'learning_rate': 1e-05, 'epoch': 0.51}
0: ��     | 7433/14725 [6:46:31<6:24:19,  3.16s/it]
 50%|█████     | 7434/14725 [6:46:35<6:32:38,  3.23s/it]
 50%|█████     | 7435/14725 [6:46:38<6:29:01,  3.20s/it]
 50%|█████     | 7436/14725 [6:46:41<6:28:48,  3.20s/it]
 51%|█████     | 7437/14725 [6:46:44<6:30:24,  3.21s/it]
 51%|█████     | 7438/14725 [6:46:47<6:27:58,  3.19s/it]
 51%|█████     | 7439/14725 [6:46:51<6:33:29,  3.24s/it]
 51%|█████     | 7440/14725 [6:46:54<6:34:51,  3.25s/it]
                                                        

 51%|█████     | 7440/14725 [6:46:54<6:34:51,  3.25s/it]
 51%|█████     | 7441/14725 [6:46:57<6:35:48,  3.26s/it]
 51%|█████     | 7442/14725 [6:47:01<6:34:22,  3.25s/it]
 51%|█████     | 7443/14725 [6:47:04<6:31:52,  3.23s/it]
 51%|█████     | 7444/14725 [6:47:07<6:41:03,  3.30s/it]
 51%|█████     | 7445/14725 [6:47:11<6:39:55,  3.30s/it]
 51%|█████     | 7446/14725 [6:47:14
0: {'loss': 0.7038, 'grad_norm': 0.4028718398402354, 'learning_rate': 1e-05, 'epoch': 0.51}
0: <6:37:57,  3.28s/it]
 51%|█████     | 7447/14725 [6:47:17<6:31:09,  3.22s/it]
 51%|█████     | 7448/14725 [6:47:20<6:36:48,  3.27s/it]
 51%|█████     | 7449/14725 [6:47:24<6:41:05,  3.31s/it]
 51%|█████     | 7450/14725 [6:47:27<6:36:39,  3.27s/it]
                                                        

 51%|█████     | 7450/14725 [6:47:27<6:36:39,  3.27s/it]
 51%|█████     | 7451/14725 [6:47:30<6:39:38,  3.30s/it]
 51%|█████     | 7452/14725 [6:47:34<6:41:58,  3.32s/it]
 51%|█████     | 7453/14725 [6:47:37<6:43:59,  3.33s/it]
 51%|█████     | 7454/14725 [6:47:40<6:51:39,  3.40s/it]
 51%|█████     | 7455/14725 [6:47:44<6:50:09,  3.39s/it]
 51%|█████     | 7456/14725 [6:47:47<6:56:38,  3.44s/it]
 51%|█████     | 7457/14725 [6:47:51<6:49:14,  3.38s/it]
 51%|█████     | 7458/14725 [6:47:54<6:44:47,  3.34s/it]
 51%|█████     | 7459/14725 [6:47:57<6:41:40,  3.32s/it]
 51%|�
0: {'loss': 0.6936, 'grad_norm': 0.42056519481393123, 'learning_rate': 1e-05, 'epoch': 0.51}
0: {'loss': 0.6997, 'grad_norm': 0.41343809294769673, 'learning_rate': 1e-05, 'epoch': 0.51}
0: �████     | 7460/14725 [6:48:00<6:32:29,  3.24s/it]
                                                        

 51%|█████     | 7460/14725 [6:48:00<6:32:29,  3.24s/it]
 51%|█████     | 7461/14725 [6:48:03<6:32:04,  3.24s/it]
 51%|█████     | 7462/14725 [6:48:07<6:34:00,  3.25s/it]
 51%|█████     | 7463/14725 [6:48:10<6:31:06,  3.23s/it]
 51%|█████     | 7464/14725 [6:48:13<6:35:40,  3.27s/it]
 51%|█████     | 7465/14725 [6:48:17<6:41:56,  3.32s/it]
 51%|█████     | 7466/14725 [6:48:20<6:35:36,  3.27s/it]
 51%|█████     | 7467/14725 [6:48:23<6:38:22,  3.29s/it]
 51%|█████     | 7468/14725 [6:48:26<6:35:02,  3.27s/it]
 51%|█████     | 7469/14725 [6:48:30<6:28:31,  3.21s/it]
 51%|█████     | 7470/14725 [6:48:33<6:31:02,  3.23s/it]
                                                        

 51%|█████     | 7470/14725 [6:48:33<6:31:02,  3.23s/it]
 51%|█████     | 7471/14725 [6:48:
0: {'loss': 0.6948, 'grad_norm': 0.45310921789060626, 'learning_rate': 1e-05, 'epoch': 0.51}
0: 36<6:42:24,  3.33s/it]
 51%|█████     | 7472/14725 [6:48:40<6:39:35,  3.31s/it]
 51%|█████     | 7473/14725 [6:48:43<6:40:41,  3.32s/it]
 51%|█████     | 7474/14725 [6:48:46<6:40:33,  3.31s/it]
 51%|█████     | 7475/14725 [6:48:49<6:36:58,  3.29s/it]
 51%|█████     | 7476/14725 [6:48:53<6:32:10,  3.25s/it]
 51%|█████     | 7477/14725 [6:48:56<6:31:57,  3.24s/it]
 51%|█████     | 7478/14725 [6:48:59<6:31:01,  3.24s/it]
 51%|█████     | 7479/14725 [6:49:02<6:31:13,  3.24s/it]
 51%|█████     | 7480/14725 [6:49:06<6:31:17,  3.24s/it]
                                                        

 51%|█████     | 7480/14725 [6:49:06<6:31:17,  3.24s/it]
 51%|█████     | 7481/14725 [6:49:09<6:31:11,  3.24s/it]
 51%|█████     | 7482/14725 [6:49:12<6:32:17,  3.25s/it]
 51%|█████     | 7483/14725 [6:49:15<6:31:47,  3.25s/it]
 51%|█████     | 7484/14725 [6:49:19<6:39:29,  3.31s/it]
 51%|
0: {'loss': 0.7091, 'grad_norm': 0.423082900599226, 'learning_rate': 1e-05, 'epoch': 0.51}
0: █████     | 7485/14725 [6:49:22<6:37:22,  3.29s/it]
 51%|█████     | 7486/14725 [6:49:25<6:36:25,  3.29s/it]
 51%|█████     | 7487/14725 [6:49:29<6:34:54,  3.27s/it]
 51%|█████     | 7488/14725 [6:49:32<6:34:53,  3.27s/it]
 51%|█████     | 7489/14725 [6:49:35<6:45:28,  3.36s/it]
 51%|█████     | 7490/14725 [6:49:38<6:36:41,  3.29s/it]
                                                        

 51%|█████     | 7490/14725 [6:49:39<6:36:41,  3.29s/it]
 51%|█████     | 7491/14725 [6:49:42<6:36:40,  3.29s/it]
 51%|█████     | 7492/14725 [6:49:45<6:36:44,  3.29s/it]
 51%|█████     | 7493/14725 [6:49:48<6:39:21,  3.31s/it]
 51%|█████     | 7494/14725 [6:49:52<6:43:10,  3.35s/it]
 51%|█████     | 7495/14725 [6:49:55<6:41:38,  3.33s/it]
 51%|█████     | 7496/14725 [6:49:58<6:35:21,  3.28s/it]
 51%|█████     | 7497/14725 [6:50:02<6:33:30,  3.27s/it]
 51%|█████     | 7498/1
0: {'loss': 0.6999, 'grad_norm': 0.4303437419809771, 'learning_rate': 1e-05, 'epoch': 0.51}
0: {'loss': 0.6968, 'grad_norm': 0.39021792271973627, 'learning_rate': 1e-05, 'epoch': 0.51}
0: 4725 [6:50:05<6:31:35,  3.25s/it]
 51%|█████     | 7499/14725 [6:50:08<6:28:42,  3.23s/it]
 51%|█████     | 7500/14725 [6:50:11<6:23:45,  3.19s/it]
                                                        

 51%|█████     | 7500/14725 [6:50:11<6:23:45,  3.19s/it]
 51%|█████     | 7501/14725 [6:50:14<6:28:13,  3.22s/it]
 51%|█████     | 7502/14725 [6:50:18<6:34:09,  3.27s/it]
 51%|█████     | 7503/14725 [6:50:21<6:32:09,  3.26s/it]
 51%|█████     | 7504/14725 [6:50:24<6:28:59,  3.23s/it]
 51%|█████     | 7505/14725 [6:50:27<6:30:13,  3.24s/it]
 51%|█████     | 7506/14725 [6:50:31<6:33:58,  3.27s/it]
 51%|█████     | 7507/14725 [6:50:34<6:34:33,  3.28s/it]
 51%|█████     | 7508/14725 [6:50:37<6:24:27,  3.20s/it]
 51%|█████     | 7509/14725 [6:50:40<6:22:47,  3.18s/it]
 51%|█████     | 7510/14725 [6:50:43<6:19:09,  3.15s/it]
                                                        

 51
0: {'loss': 0.6918, 'grad_norm': 0.39966568701885147, 'learning_rate': 1e-05, 'epoch': 0.51}
0: %|█████     | 7510/14725 [6:50:43<6:19:09,  3.15s/it]
 51%|█████     | 7511/14725 [6:50:47<6:27:02,  3.22s/it]
 51%|█████     | 7512/14725 [6:50:50<6:33:53,  3.28s/it]
 51%|█████     | 7513/14725 [6:50:53<6:34:23,  3.28s/it]
 51%|█████     | 7514/14725 [6:50:57<6:37:33,  3.31s/it]
 51%|█████     | 7515/14725 [6:51:00<6:44:24,  3.37s/it]
 51%|█████     | 7516/14725 [6:51:03<6:34:56,  3.29s/it]
 51%|█████     | 7517/14725 [6:51:07<6:33:34,  3.28s/it]
 51%|█████     | 7518/14725 [6:51:10<6:27:43,  3.23s/it]
 51%|█████     | 7519/14725 [6:51:13<6:34:00,  3.28s/it]
 51%|█████     | 7520/14725 [6:51:16<6:29:58,  3.25s/it]
                                                        

 51%|█████     | 7520/14725 [6:51:16<6:29:58,  3.25s/it]
 51%|█████     | 7521/14725 [6:51:20<6:31:09,  3.26s/it]
 51%|█████     | 7522/14725 [6:51:23<6:32:30,  3.27s/it]
 51%|█████     | 7523
0: {'loss': 0.7001, 'grad_norm': 0.4318405340959211, 'learning_rate': 1e-05, 'epoch': 0.51}
0: /14725 [6:51:26<6:30:19,  3.25s/it]
 51%|█████     | 7524/14725 [6:51:29<6:34:42,  3.29s/it]
 51%|█████     | 7525/14725 [6:51:33<6:30:53,  3.26s/it]
 51%|█████     | 7526/14725 [6:51:36<6:29:51,  3.25s/it]
 51%|█████     | 7527/14725 [6:51:39<6:24:35,  3.21s/it]
 51%|█████     | 7528/14725 [6:51:42<6:27:33,  3.23s/it]
 51%|█████     | 7529/14725 [6:51:45<6:24:40,  3.21s/it]
 51%|█████     | 7530/14725 [6:51:49<6:28:49,  3.24s/it]
                                                        

 51%|█████     | 7530/14725 [6:51:49<6:28:49,  3.24s/it]
 51%|█████     | 7531/14725 [6:51:52<6:31:25,  3.26s/it]
 51%|█████     | 7532/14725 [6:51:55<6:27:25,  3.23s/it]
 51%|█████     | 7533/14725 [6:51:58<6:27:02,  3.23s/it]
 51%|█████     | 7534/14725 [6:52:02<6:25:50,  3.22s/it]
 51%|█████     | 7535/14725 [6:52:05<6:37:28,  3.32s/it]
 51%|█████     | 7536/14725 [6:52:08<6:35:31,  3.
0: {'loss': 0.7029, 'grad_norm': 0.4160328128726748, 'learning_rate': 1e-05, 'epoch': 0.51}
0: 30s/it]
 51%|█████     | 7537/14725 [6:52:12<6:34:18,  3.29s/it]
 51%|█████     | 7538/14725 [6:52:15<6:37:44,  3.32s/it]
 51%|█████     | 7539/14725 [6:52:18<6:34:05,  3.29s/it]
 51%|█████     | 7540/14725 [6:52:21<6:30:23,  3.26s/it]
                                                        

 51%|█████     | 7540/14725 [6:52:21<6:30:23,  3.26s/it]
 51%|█████     | 7541/14725 [6:52:25<6:29:48,  3.26s/it]
 51%|█████     | 7542/14725 [6:52:28<6:25:01,  3.22s/it]
 51%|█████     | 7543/14725 [6:52:31<6:19:21,  3.17s/it]
 51%|█████     | 7544/14725 [6:52:34<6:26:39,  3.23s/it]
 51%|█████     | 7545/14725 [6:52:38<6:30:48,  3.27s/it]
 51%|█████     | 7546/14725 [6:52:41<6:32:04,  3.28s/it]
 51%|█████▏    | 7547/14725 [6:52:44<6:39:17,  3.34s/it]
 51%|█████▏    | 7548/14725 [6:52:48<6:33:49,  3.29s/it]
 51%|█████▏    | 7549/14725 [6:52:51<6:34:05,  3.30s/it]
 51%|███
0: {'loss': 0.6899, 'grad_norm': 0.4089270331628542, 'learning_rate': 1e-05, 'epoch': 0.51}
0: {'loss': 0.7049, 'grad_norm': 0.4202846276676179, 'learning_rate': 1e-05, 'epoch': 0.51}
0: ██▏    | 7550/14725 [6:52:54<6:33:22,  3.29s/it]
                                                        

 51%|█████▏    | 7550/14725 [6:52:54<6:33:22,  3.29s/it]
 51%|█████▏    | 7551/14725 [6:52:58<6:34:25,  3.30s/it]
 51%|█████▏    | 7552/14725 [6:53:01<6:34:17,  3.30s/it]
 51%|█████▏    | 7553/14725 [6:53:04<6:29:21,  3.26s/it]
 51%|█████▏    | 7554/14725 [6:53:07<6:20:17,  3.18s/it]
 51%|█████▏    | 7555/14725 [6:53:10<6:24:48,  3.22s/it]
 51%|█████▏    | 7556/14725 [6:53:13<6:19:07,  3.17s/it]
 51%|█████▏    | 7557/14725 [6:53:17<6:24:54,  3.22s/it]
 51%|█████▏    | 7558/14725 [6:53:20<6:18:20,  3.17s/it]
 51%|█████▏    | 7559/14725 [6:53:23<6:17:44,  3.16s/it]
 51%|█████▏    | 7560/14725 [6:53:26<6:22:03,  3.20s/it]
                                                        

 51%|█████▏    | 7560/14725 [6:53:26<6:22:03,  3.20s/it]
 51%|█████▏  
0: {'loss': 0.6973, 'grad_norm': 0.40687004417009176, 'learning_rate': 1e-05, 'epoch': 0.51}
0:   | 7561/14725 [6:53:29<6:24:50,  3.22s/it]
 51%|█████▏    | 7562/14725 [6:53:33<6:20:23,  3.19s/it]
 51%|█████▏    | 7563/14725 [6:53:36<6:24:11,  3.22s/it]
 51%|█████▏    | 7564/14725 [6:53:39<6:22:55,  3.21s/it]
 51%|█████▏    | 7565/14725 [6:53:43<6:34:54,  3.31s/it]
 51%|█████▏    | 7566/14725 [6:53:46<6:31:55,  3.28s/it]
 51%|█████▏    | 7567/14725 [6:53:49<6:26:54,  3.24s/it]
 51%|█████▏    | 7568/14725 [6:53:52<6:34:50,  3.31s/it]
 51%|█████▏    | 7569/14725 [6:53:56<6:35:01,  3.31s/it]
 51%|█████▏    | 7570/14725 [6:53:59<6:34:27,  3.31s/it]
                                                        

 51%|█████▏    | 7570/14725 [6:53:59<6:34:27,  3.31s/it]
 51%|█████▏    | 7571/14725 [6:54:02<6:27:15,  3.25s/it]
 51%|█████▏    | 7572/14725 [6:54:05<6:24:57,  3.23s/it]
 51%|█████▏    | 7573/14725 [6:54:09<6:27:43,  3.25s/it]
 51%|█████▏  
0: {'loss': 0.6783, 'grad_norm': 0.4285123618136602, 'learning_rate': 1e-05, 'epoch': 0.51}
0:   | 7574/14725 [6:54:12<6:23:30,  3.22s/it]
 51%|█████▏    | 7575/14725 [6:54:15<6:24:35,  3.23s/it]
 51%|█████▏    | 7576/14725 [6:54:18<6:31:56,  3.29s/it]
 51%|█████▏    | 7577/14725 [6:54:22<6:28:07,  3.26s/it]
 51%|█████▏    | 7578/14725 [6:54:25<6:24:00,  3.22s/it]
 51%|█████▏    | 7579/14725 [6:54:28<6:23:57,  3.22s/it]
 51%|█████▏    | 7580/14725 [6:54:31<6:28:53,  3.27s/it]
                                                        

 51%|█████▏    | 7580/14725 [6:54:31<6:28:53,  3.27s/it]
 51%|█████▏    | 7581/14725 [6:54:35<6:27:51,  3.26s/it]
 51%|█████▏    | 7582/14725 [6:54:38<6:28:33,  3.26s/it]
 51%|█████▏    | 7583/14725 [6:54:41<6:26:13,  3.24s/it]
 52%|█████▏    | 7584/14725 [6:54:44<6:26:29,  3.25s/it]
 52%|█████▏    | 7585/14725 [6:54:48<6:31:41,  3.29s/it]
 52%|█████▏    | 7586/14725 [6:54:51<6:33:35,  3.31s/it]
 52%|█████▏  
0: {'loss': 0.6924, 'grad_norm': 0.44115824280075594, 'learning_rate': 1e-05, 'epoch': 0.52}
0:   | 7587/14725 [6:54:54<6:27:47,  3.26s/it]
 52%|█████▏    | 7588/14725 [6:54:57<6:20:31,  3.20s/it]
 52%|█████▏    | 7589/14725 [6:55:00<6:17:48,  3.18s/it]
 52%|█████▏    | 7590/14725 [6:55:04<6:22:40,  3.22s/it]
                                                        

 52%|█████▏    | 7590/14725 [6:55:04<6:22:40,  3.22s/it]
 52%|█████▏    | 7591/14725 [6:55:07<6:27:40,  3.26s/it]
 52%|█████▏    | 7592/14725 [6:55:10<6:29:51,  3.28s/it]
 52%|█████▏    | 7593/14725 [6:55:14<6:26:13,  3.25s/it]
 52%|█████▏    | 7594/14725 [6:55:17<6:24:26,  3.23s/it]
 52%|█████▏    | 7595/14725 [6:55:20<6:23:00,  3.22s/it]
 52%|█████▏    | 7596/14725 [6:55:23<6:19:57,  3.20s/it]
 52%|█████▏    | 7597/14725 [6:55:26<6:22:41,  3.22s/it]
 52%|█████▏    | 7598/14725 [6:55:30<6:27:09,  3.26s/it]
 52%|█████▏    | 7599/14725 [6:55:33<6:25:45,  3.25s/it]
 52%|█████▏  
0: {'loss': 0.7052, 'grad_norm': 0.4212039818856107, 'learning_rate': 1e-05, 'epoch': 0.52}
0: {'loss': 0.6965, 'grad_norm': 0.4210625913130812, 'learning_rate': 1e-05, 'epoch': 0.52}
0:   | 7600/14725 [6:55:36<6:29:11,  3.28s/it]
                                                        

 52%|█████▏    | 7600/14725 [6:55:36<6:29:11,  3.28s/it]
 52%|█████▏    | 7601/14725 [6:55:40<6:29:30,  3.28s/it]
 52%|█████▏    | 7602/14725 [6:55:43<6:27:54,  3.27s/it]
 52%|█████▏    | 7603/14725 [6:55:46<6:25:10,  3.24s/it]
 52%|█████▏    | 7604/14725 [6:55:49<6:25:00,  3.24s/it]
 52%|█████▏    | 7605/14725 [6:55:52<6:19:26,  3.20s/it]
 52%|█████▏    | 7606/14725 [6:55:56<6:22:38,  3.22s/it]
 52%|█████▏    | 7607/14725 [6:55:59<6:22:19,  3.22s/it]
 52%|█████▏    | 7608/14725 [6:56:02<6:25:27,  3.25s/it]
 52%|█████▏    | 7609/14725 [6:56:05<6:24:18,  3.24s/it]
 52%|█████▏    | 7610/14725 [6:56:08<6:15:08,  3.16s/it]
                                                        

 52%|█████▏    | 7610/14725 [6:56:08<6:15:08,  3.16s/it]
 52%|█████▏    | 7611/14
0: {'loss': 0.6843, 'grad_norm': 0.40786526716953664, 'learning_rate': 1e-05, 'epoch': 0.52}
0: 725 [6:56:12<6:18:15,  3.19s/it]
 52%|█████▏    | 7612/14725 [6:56:15<6:14:01,  3.16s/it]
 52%|█████▏    | 7613/14725 [6:56:18<6:12:37,  3.14s/it]
 52%|█████▏    | 7614/14725 [6:56:21<6:08:22,  3.11s/it]
 52%|█████▏    | 7615/14725 [6:56:24<6:09:00,  3.11s/it]
 52%|█████▏    | 7616/14725 [6:56:27<6:21:27,  3.22s/it]
 52%|█████▏    | 7617/14725 [6:56:31<6:20:27,  3.21s/it]
 52%|█████▏    | 7618/14725 [6:56:34<6:25:40,  3.26s/it]
 52%|█████▏    | 7619/14725 [6:56:37<6:25:00,  3.25s/it]
 52%|█████▏    | 7620/14725 [6:56:41<6:31:57,  3.31s/it]
                                                        

 52%|█████▏    | 7620/14725 [6:56:41<6:31:57,  3.31s/it]
 52%|█████▏    | 7621/14725 [6:56:44<6:32:06,  3.31s/it]
 52%|█████▏    | 7622/14725 [6:56:47<6:29:38,  3.29s/it]
 52%|█████▏    | 7623/14725 [6:56:50<6:22:19,  3.23s/it]
 52%|█████▏    | 7624/14
0: {'loss': 0.6893, 'grad_norm': 0.4124566196793564, 'learning_rate': 1e-05, 'epoch': 0.52}
0: 725 [6:56:54<6:26:21,  3.26s/it]
 52%|█████▏    | 7625/14725 [6:56:57<6:27:03,  3.27s/it]
 52%|█████▏    | 7626/14725 [6:57:00<6:26:17,  3.26s/it]
 52%|█████▏    | 7627/14725 [6:57:03<6:20:57,  3.22s/it]
 52%|█████▏    | 7628/14725 [6:57:07<6:28:21,  3.28s/it]
 52%|█████▏    | 7629/14725 [6:57:10<6:26:37,  3.27s/it]
 52%|█████▏    | 7630/14725 [6:57:13<6:24:21,  3.25s/it]
                                                        

 52%|█████▏    | 7630/14725 [6:57:13<6:24:21,  3.25s/it]
 52%|█████▏    | 7631/14725 [6:57:17<6:30:29,  3.30s/it]
 52%|█████▏    | 7632/14725 [6:57:20<6:27:03,  3.27s/it]
 52%|█████▏    | 7633/14725 [6:57:23<6:27:34,  3.28s/it]
 52%|█████▏    | 7634/14725 [6:57:26<6:20:50,  3.22s/it]
 52%|█████▏    | 7635/14725 [6:57:29<6:20:31,  3.22s/it]
 52%|█████▏    | 7636/14725 [6:57:33<6:18:48,  3.21s/it]
 52%|█████▏    | 7637/14
0: {'loss': 0.6906, 'grad_norm': 0.41825050421784116, 'learning_rate': 1e-05, 'epoch': 0.52}
0: 725 [6:57:36<6:16:06,  3.18s/it]
 52%|█████▏    | 7638/14725 [6:57:39<6:13:33,  3.16s/it]
 52%|█████▏    | 7639/14725 [6:57:42<6:11:09,  3.14s/it]
 52%|█████▏    | 7640/14725 [6:57:45<6:12:15,  3.15s/it]
                                                        

 52%|█████▏    | 7640/14725 [6:57:45<6:12:15,  3.15s/it]
 52%|█████▏    | 7641/14725 [6:57:48<6:17:14,  3.20s/it]
 52%|█████▏    | 7642/14725 [6:57:52<6:20:16,  3.22s/it]
 52%|█████▏    | 7643/14725 [6:57:55<6:23:05,  3.25s/it]
 52%|█████▏    | 7644/14725 [6:57:58<6:24:23,  3.26s/it]
 52%|█████▏    | 7645/14725 [6:58:01<6:17:40,  3.20s/it]
 52%|█████▏    | 7646/14725 [6:58:05<6:21:10,  3.23s/it]
 52%|█████▏    | 7647/14725 [6:58:08<6:18:25,  3.21s/it]
 52%|█████▏    | 7648/14725 [6:58:11<6:21:21,  3.23s/it]
 52%|█████▏    | 7649/14725 [6:58:14<6:19:58,  3.22s/it]
 52%|█████▏    | 7650/14
0: {'loss': 0.6822, 'grad_norm': 0.43850559188736876, 'learning_rate': 1e-05, 'epoch': 0.52}
0: {'loss': 0.6981, 'grad_norm': 0.43361469794970464, 'learning_rate': 1e-05, 'epoch': 0.52}
0: 725 [6:58:17<6:16:15,  3.19s/it]
                                                        

 52%|█████▏    | 7650/14725 [6:58:17<6:16:15,  3.19s/it]
 52%|█████▏    | 7651/14725 [6:58:21<6:14:18,  3.17s/it]
 52%|█████▏    | 7652/14725 [6:58:24<6:18:43,  3.21s/it]
 52%|█████▏    | 7653/14725 [6:58:27<6:25:15,  3.27s/it]
 52%|█████▏    | 7654/14725 [6:58:31<6:26:11,  3.28s/it]
 52%|█████▏    | 7655/14725 [6:58:34<6:22:01,  3.24s/it]
 52%|█████▏    | 7656/14725 [6:58:37<6:21:48,  3.24s/it]
 52%|█████▏    | 7657/14725 [6:58:40<6:23:37,  3.26s/it]
 52%|█████▏    | 7658/14725 [6:58:43<6:17:31,  3.21s/it]
 52%|█████▏    | 7659/14725 [6:58:47<6:25:00,  3.27s/it]
 52%|█████▏    | 7660/14725 [6:58:50<6:23:20,  3.26s/it]
                                                        

 52%|█████▏    | 7660/14725 [6:58:50<6:23:20,  3.26s/it]
 52%|█████▏    | 7661/14725 [6:58:5
0: {'loss': 0.708, 'grad_norm': 0.44193904210788365, 'learning_rate': 1e-05, 'epoch': 0.52}
0: 3<6:28:47,  3.30s/it]
 52%|█████▏    | 7662/14725 [6:58:56<6:18:54,  3.22s/it]
 52%|█████▏    | 7663/14725 [6:59:00<6:23:09,  3.26s/it]
 52%|█████▏    | 7664/14725 [6:59:03<6:21:32,  3.24s/it]
 52%|█████▏    | 7665/14725 [6:59:06<6:23:19,  3.26s/it]
 52%|█████▏    | 7666/14725 [6:59:09<6:22:37,  3.25s/it]
 52%|█████▏    | 7667/14725 [6:59:13<6:25:38,  3.28s/it]
 52%|█████▏    | 7668/14725 [6:59:16<6:28:14,  3.30s/it]
 52%|█████▏    | 7669/14725 [6:59:19<6:26:01,  3.28s/it]
 52%|█████▏    | 7670/14725 [6:59:23<6:30:46,  3.32s/it]
                                                        

 52%|█████▏    | 7670/14725 [6:59:23<6:30:46,  3.32s/it]
 52%|█████▏    | 7671/14725 [6:59:26<6:20:58,  3.24s/it]
 52%|█████▏    | 7672/14725 [6:59:29<6:19:17,  3.23s/it]
 52%|█████▏    | 7673/14725 [6:59:32<6:21:50,  3.25s/it]
 52%|█████▏    | 7674/14725 [6:59:3
0: {'loss': 0.6957, 'grad_norm': 0.41231571340588297, 'learning_rate': 1e-05, 'epoch': 0.52}
0: 6<6:17:51,  3.22s/it]
 52%|█████▏    | 7675/14725 [6:59:39<6:23:58,  3.27s/it]
 52%|█████▏    | 7676/14725 [6:59:42<6:27:40,  3.30s/it]
 52%|█████▏    | 7677/14725 [6:59:46<6:35:09,  3.36s/it]
 52%|█████▏    | 7678/14725 [6:59:49<6:29:56,  3.32s/it]
 52%|█████▏    | 7679/14725 [6:59:52<6:27:15,  3.30s/it]
 52%|█████▏    | 7680/14725 [6:59:56<6:26:52,  3.29s/it]
                                                        

 52%|█████▏    | 7680/14725 [6:59:56<6:26:52,  3.29s/it]
 52%|█████▏    | 7681/14725 [6:59:59<6:20:57,  3.24s/it]
 52%|█████▏    | 7682/14725 [7:00:02<6:20:23,  3.24s/it]
 52%|█████▏    | 7683/14725 [7:00:05<6:16:32,  3.21s/it]
 52%|█████▏    | 7684/14725 [7:00:08<6:18:48,  3.23s/it]
 52%|█████▏    | 7685/14725 [7:00:12<6:18:39,  3.23s/it]
 52%|█████▏    | 7686/14725 [7:00:15<6:14:38,  3.19s/it]
 52%|█████▏    | 7687/14725 [7:00:1
0: {'loss': 0.6739, 'grad_norm': 0.39909131982761975, 'learning_rate': 1e-05, 'epoch': 0.52}
0: 8<6:15:43,  3.20s/it]
 52%|█████▏    | 7688/14725 [7:00:21<6:13:29,  3.18s/it]
 52%|█████▏    | 7689/14725 [7:00:24<6:09:28,  3.15s/it]
 52%|█████▏    | 7690/14725 [7:00:27<6:07:40,  3.14s/it]
                                                        

 52%|█████▏    | 7690/14725 [7:00:27<6:07:40,  3.14s/it]
 52%|█████▏    | 7691/14725 [7:00:30<6:09:18,  3.15s/it]
 52%|█████▏    | 7692/14725 [7:00:34<6:14:24,  3.19s/it]
 52%|█████▏    | 7693/14725 [7:00:37<6:14:47,  3.20s/it]
 52%|█████▏    | 7694/14725 [7:00:40<6:22:31,  3.26s/it]
 52%|█████▏    | 7695/14725 [7:00:44<6:21:11,  3.25s/it]
 52%|█████▏    | 7696/14725 [7:00:47<6:21:37,  3.26s/it]
 52%|█████▏    | 7697/14725 [7:00:50<6:22:17,  3.26s/it]
 52%|█████▏    | 7698/14725 [7:00:53<6:22:49,  3.27s/it]
 52%|█████▏    | 7699/14725 [7:00:57<6:25:13,  3.29s/it]
 52%|█████▏    | 7700/14725 [7:01:0
0: {'loss': 0.714, 'grad_norm': 0.4089333923560052, 'learning_rate': 1e-05, 'epoch': 0.52}
0: {'loss': 0.6946, 'grad_norm': 0.4177810752333485, 'learning_rate': 1e-05, 'epoch': 0.52}
0: 0<6:27:15,  3.31s/it]
                                                        

 52%|█████▏    | 7700/14725 [7:01:00<6:27:15,  3.31s/it]
 52%|█████▏    | 7701/14725 [7:01:03<6:22:46,  3.27s/it]
 52%|█████▏    | 7702/14725 [7:01:07<6:25:42,  3.30s/it]
 52%|█████▏    | 7703/14725 [7:01:10<6:28:13,  3.32s/it]
 52%|█████▏    | 7704/14725 [7:01:14<6:40:32,  3.42s/it]
 52%|█████▏    | 7705/14725 [7:01:17<6:32:32,  3.36s/it]
 52%|█████▏    | 7706/14725 [7:01:20<6:23:50,  3.28s/it]
 52%|█████▏    | 7707/14725 [7:01:23<6:23:17,  3.28s/it]
 52%|█████▏    | 7708/14725 [7:01:27<6:34:10,  3.37s/it]
 52%|█████▏    | 7709/14725 [7:01:30<6:27:29,  3.31s/it]
 52%|█████▏    | 7710/14725 [7:01:33<6:23:41,  3.28s/it]
                                                        

 52%|█████▏    | 7710/14725 [7:01:33<6:23:41,  3.28s/it]
 52%|█████▏    | 7711/14725 [7:01:36<6:25:14, 
0: {'loss': 0.688, 'grad_norm': 0.46913536247511983, 'learning_rate': 1e-05, 'epoch': 0.52}
0:  3.30s/it]
 52%|█████▏    | 7712/14725 [7:01:40<6:19:04,  3.24s/it]
 52%|█████▏    | 7713/14725 [7:01:43<6:21:44,  3.27s/it]
 52%|█████▏    | 7714/14725 [7:01:46<6:18:41,  3.24s/it]
 52%|█████▏    | 7715/14725 [7:01:49<6:17:23,  3.23s/it]
 52%|█████▏    | 7716/14725 [7:01:53<6:18:22,  3.24s/it]
 52%|█████▏    | 7717/14725 [7:01:56<6:13:21,  3.20s/it]
 52%|█████▏    | 7718/14725 [7:01:59<6:18:12,  3.24s/it]
 52%|█████▏    | 7719/14725 [7:02:02<6:14:10,  3.20s/it]
 52%|█████▏    | 7720/14725 [7:02:06<6:25:44,  3.30s/it]
                                                        

 52%|█████▏    | 7720/14725 [7:02:06<6:25:44,  3.30s/it]
 52%|█████▏    | 7721/14725 [7:02:09<6:19:02,  3.25s/it]
 52%|█████▏    | 7722/14725 [7:02:12<6:22:15,  3.28s/it]
 52%|█████▏    | 7723/14725 [7:02:15<6:21:59,  3.27s/it]
 52%|█████▏    | 7724/14725 [7:02:19<6:18:03, 
0: {'loss': 0.6978, 'grad_norm': 0.4004901998797393, 'learning_rate': 1e-05, 'epoch': 0.52}
0:  3.24s/it]
 52%|█████▏    | 7725/14725 [7:02:22<6:16:40,  3.23s/it]
 52%|█████▏    | 7726/14725 [7:02:25<6:14:27,  3.21s/it]
 52%|█████▏    | 7727/14725 [7:02:28<6:18:42,  3.25s/it]
 52%|█████▏    | 7728/14725 [7:02:32<6:24:58,  3.30s/it]
 52%|█████▏    | 7729/14725 [7:02:35<6:27:25,  3.32s/it]
 52%|█████▏    | 7730/14725 [7:02:38<6:26:14,  3.31s/it]
                                                        

 52%|█████▏    | 7730/14725 [7:02:38<6:26:14,  3.31s/it]
 53%|█████▎    | 7731/14725 [7:02:42<6:21:30,  3.27s/it]
 53%|█████▎    | 7732/14725 [7:02:45<6:18:02,  3.24s/it]
 53%|█████▎    | 7733/14725 [7:02:48<6:15:33,  3.22s/it]
 53%|█████▎    | 7734/14725 [7:02:51<6:16:32,  3.23s/it]
 53%|█████▎    | 7735/14725 [7:02:54<6:20:41,  3.27s/it]
 53%|█████▎    | 7736/14725 [7:02:58<6:22:38,  3.28s/it]
 53%|█████▎    | 7737/14725 [7:03:01<6:21:03, 
0: {'loss': 0.7046, 'grad_norm': 0.41793533843272174, 'learning_rate': 1e-05, 'epoch': 0.53}
0:  3.27s/it]
 53%|█████▎    | 7738/14725 [7:03:04<6:22:03,  3.28s/it]
 53%|█████▎    | 7739/14725 [7:03:08<6:21:33,  3.28s/it]
 53%|█████▎    | 7740/14725 [7:03:11<6:15:50,  3.23s/it]
                                                        

 53%|█████▎    | 7740/14725 [7:03:11<6:15:50,  3.23s/it]
 53%|█████▎    | 7741/14725 [7:03:14<6:17:10,  3.24s/it]
 53%|█████▎    | 7742/14725 [7:03:18<6:27:24,  3.33s/it]
 53%|█████▎    | 7743/14725 [7:03:21<6:27:27,  3.33s/it]
 53%|█████▎    | 7744/14725 [7:03:24<6:24:18,  3.30s/it]
 53%|█████▎    | 7745/14725 [7:03:27<6:25:32,  3.31s/it]
 53%|█████▎    | 7746/14725 [7:03:31<6:30:51,  3.36s/it]
 53%|█████▎    | 7747/14725 [7:03:34<6:25:55,  3.32s/it]
 53%|█████▎    | 7748/14725 [7:03:37<6:21:40,  3.28s/it]
 53%|█████▎    | 7749/14725 [7:03:40<6:16:12,  3.24s/it]
 53%|█████▎    | 7750/14725 [7:03:44<6:15:58, 
0: {'loss': 0.6913, 'grad_norm': 0.416230168437437, 'learning_rate': 1e-05, 'epoch': 0.53}
0: {'loss': 0.6936, 'grad_norm': 0.4116588406107215, 'learning_rate': 1e-05, 'epoch': 0.53}
0:  3.23s/it]
                                                        

 53%|█████▎    | 7750/14725 [7:03:44<6:15:58,  3.23s/it]
 53%|█████▎    | 7751/14725 [7:03:47<6:18:49,  3.26s/it]
 53%|█████▎    | 7752/14725 [7:03:50<6:16:46,  3.24s/it]
 53%|█████▎    | 7753/14725 [7:03:53<6:12:55,  3.21s/it]
 53%|█████▎    | 7754/14725 [7:03:57<6:20:53,  3.28s/it]
 53%|█████▎    | 7755/14725 [7:04:00<6:23:01,  3.30s/it]
 53%|█████▎    | 7756/14725 [7:04:03<6:22:55,  3.30s/it]
 53%|█████▎    | 7757/14725 [7:04:07<6:25:05,  3.32s/it]
 53%|█████▎    | 7758/14725 [7:04:10<6:31:28,  3.37s/it]
 53%|█████▎    | 7759/14725 [7:04:13<6:24:50,  3.31s/it]
 53%|█████▎    | 7760/14725 [7:04:17<6:27:36,  3.34s/it]
                                                        

 53%|█████▎    | 7760/14725 [7:04:17<6:27:36,  3.34s/it]
 53%|█████▎    | 7761/14725 [7:04:20<6:22:44,  3.30s/it]
0: {'loss': 0.6995, 'grad_norm': 0.418594133826632, 'learning_rate': 1e-05, 'epoch': 0.53}
0:  53%|█████▎    | 7762/14725 [7:04:24<6:27:56,  3.34s/it]
 53%|█████▎    | 7763/14725 [7:04:27<6:34:42,  3.40s/it]
 53%|█████▎    | 7764/14725 [7:04:31<6:42:15,  3.47s/it]
 53%|█████▎    | 7765/14725 [7:04:34<6:32:34,  3.38s/it]
 53%|█████▎    | 7766/14725 [7:04:37<6:34:16,  3.40s/it]
 53%|█████▎    | 7767/14725 [7:04:40<6:25:43,  3.33s/it]
 53%|█████▎    | 7768/14725 [7:04:44<6:22:02,  3.29s/it]
 53%|█████▎    | 7769/14725 [7:04:47<6:19:52,  3.28s/it]
 53%|█████▎    | 7770/14725 [7:04:50<6:20:59,  3.29s/it]
                                                        

 53%|█████▎    | 7770/14725 [7:04:50<6:20:59,  3.29s/it]
 53%|█████▎    | 7771/14725 [7:04:53<6:15:01,  3.24s/it]
 53%|█████▎    | 7772/14725 [7:04:56<6:11:40,  3.21s/it]
 53%|█████▎    | 7773/14725 [7:05:00<6:08:21,  3.18s/it]
 53%|█████▎    | 7774/14725 [7:05:03<6:05:05,  3.15s/it]
0: {'loss': 0.6865, 'grad_norm': 0.4097396611149825, 'learning_rate': 1e-05, 'epoch': 0.53}
0:  53%|█████▎    | 7775/14725 [7:05:06<6:04:57,  3.15s/it]
 53%|█████▎    | 7776/14725 [7:05:09<6:16:03,  3.25s/it]
 53%|█████▎    | 7777/14725 [7:05:13<6:18:42,  3.27s/it]
 53%|█████▎    | 7778/14725 [7:05:16<6:14:26,  3.23s/it]
 53%|█████▎    | 7779/14725 [7:05:19<6:15:48,  3.25s/it]
 53%|█████▎    | 7780/14725 [7:05:22<6:20:57,  3.29s/it]
                                                        

 53%|█████▎    | 7780/14725 [7:05:22<6:20:57,  3.29s/it]
 53%|█████▎    | 7781/14725 [7:05:26<6:15:27,  3.24s/it]
 53%|█████▎    | 7782/14725 [7:05:29<6:14:28,  3.24s/it]
 53%|█████▎    | 7783/14725 [7:05:32<6:16:31,  3.25s/it]
 53%|█████▎    | 7784/14725 [7:05:35<6:15:24,  3.25s/it]
 53%|█████▎    | 7785/14725 [7:05:38<6:06:52,  3.17s/it]
 53%|█████▎    | 7786/14725 [7:05:42<6:12:50,  3.22s/it]
 53%|█████▎    | 7787/14725 [7:05:45<6:10:09,  3.20s/it]
0: {'loss': 0.6907, 'grad_norm': 0.40721256622431645, 'learning_rate': 1e-05, 'epoch': 0.53}
0:  53%|█████▎    | 7788/14725 [7:05:48<6:19:43,  3.28s/it]
 53%|█████▎    | 7789/14725 [7:05:51<6:16:27,  3.26s/it]
 53%|█████▎    | 7790/14725 [7:05:55<6:10:48,  3.21s/it]
                                                        

 53%|█████▎    | 7790/14725 [7:05:55<6:10:48,  3.21s/it]
 53%|█████▎    | 7791/14725 [7:05:58<6:11:42,  3.22s/it]
 53%|█████▎    | 7792/14725 [7:06:01<6:11:31,  3.22s/it]
 53%|█████▎    | 7793/14725 [7:06:04<6:15:31,  3.25s/it]
 53%|█████▎    | 7794/14725 [7:06:08<6:12:18,  3.22s/it]
 53%|█████▎    | 7795/14725 [7:06:11<6:16:24,  3.26s/it]
 53%|█████▎    | 7796/14725 [7:06:14<6:20:35,  3.30s/it]
 53%|█████▎    | 7797/14725 [7:06:18<6:20:56,  3.30s/it]
 53%|█████▎    | 7798/14725 [7:06:21<6:14:12,  3.24s/it]
 53%|█████▎    | 7799/14725 [7:06:24<6:13:38,  3.24s/it]
 53%|█████▎    | 7800/14725 [7:06:27<6:18:32,  3.28s/it]
0: {'loss': 0.6885, 'grad_norm': 0.4087067623533869, 'learning_rate': 1e-05, 'epoch': 0.53}
0: {'loss': 0.71, 'grad_norm': 0.41557488684850663, 'learning_rate': 1e-05, 'epoch': 0.53}
0:                                                         

 53%|█████▎    | 7800/14725 [7:06:27<6:18:32,  3.28s/it]
 53%|█████▎    | 7801/14725 [7:06:31<6:18:28,  3.28s/it]
 53%|█████▎    | 7802/14725 [7:06:34<6:15:37,  3.26s/it]
 53%|█████▎    | 7803/14725 [7:06:37<6:11:08,  3.22s/it]
 53%|█████▎    | 7804/14725 [7:06:40<6:12:17,  3.23s/it]
 53%|█████▎    | 7805/14725 [7:06:43<6:08:46,  3.20s/it]
 53%|█████▎    | 7806/14725 [7:06:47<6:22:12,  3.31s/it]
 53%|█████▎    | 7807/14725 [7:06:50<6:18:03,  3.28s/it]
 53%|█████▎    | 7808/14725 [7:06:53<6:14:49,  3.25s/it]
 53%|█████▎    | 7809/14725 [7:06:56<6:12:40,  3.23s/it]
 53%|█████▎    | 7810/14725 [7:07:00<6:13:17,  3.24s/it]
                                                        

 53%|█████▎    | 7810/14725 [7:07:00<6:13:17,  3.24s/it]
 53%|█████▎    | 7811/14725 [7:07:03<6:12:21,  3.23s/it]
 53%|██
0: {'loss': 0.6767, 'grad_norm': 0.4267542048410163, 'learning_rate': 1e-05, 'epoch': 0.53}
0: ███▎    | 7812/14725 [7:07:06<6:10:41,  3.22s/it]
 53%|█████▎    | 7813/14725 [7:07:09<6:13:13,  3.24s/it]
 53%|█████▎    | 7814/14725 [7:07:13<6:11:36,  3.23s/it]
 53%|█████▎    | 7815/14725 [7:07:16<6:09:28,  3.21s/it]
 53%|█████▎    | 7816/14725 [7:07:19<6:07:35,  3.19s/it]
 53%|█████▎    | 7817/14725 [7:07:22<6:09:27,  3.21s/it]
 53%|█████▎    | 7818/14725 [7:07:25<6:09:56,  3.21s/it]
 53%|█████▎    | 7819/14725 [7:07:29<6:09:07,  3.21s/it]
 53%|█████▎    | 7820/14725 [7:07:32<6:06:48,  3.19s/it]
                                                        

 53%|█████▎    | 7820/14725 [7:07:32<6:06:48,  3.19s/it]
 53%|█████▎    | 7821/14725 [7:07:35<6:10:12,  3.22s/it]
 53%|█████▎    | 7822/14725 [7:07:38<6:12:40,  3.24s/it]
 53%|█████▎    | 7823/14725 [7:07:42<6:19:00,  3.29s/it]
 53%|█████▎    | 7824/14725 [7:07:45<6:13:20,  3.25s/it]
 53%|██
0: {'loss': 0.715, 'grad_norm': 0.41552632508230425, 'learning_rate': 1e-05, 'epoch': 0.53}
0: ███▎    | 7825/14725 [7:07:48<6:13:07,  3.24s/it]
 53%|█████▎    | 7826/14725 [7:07:51<6:08:26,  3.20s/it]
 53%|█████▎    | 7827/14725 [7:07:54<6:08:00,  3.20s/it]
 53%|█████▎    | 7828/14725 [7:07:58<6:14:29,  3.26s/it]
 53%|█████▎    | 7829/14725 [7:08:01<6:14:21,  3.26s/it]
 53%|█████▎    | 7830/14725 [7:08:04<6:09:27,  3.22s/it]
                                                        

 53%|█████▎    | 7830/14725 [7:08:04<6:09:27,  3.22s/it]
 53%|█████▎    | 7831/14725 [7:08:07<6:11:19,  3.23s/it]
 53%|█████▎    | 7832/14725 [7:08:11<6:17:02,  3.28s/it]
 53%|█████▎    | 7833/14725 [7:08:14<6:25:06,  3.35s/it]
 53%|█████▎    | 7834/14725 [7:08:17<6:18:19,  3.29s/it]
 53%|█████▎    | 7835/14725 [7:08:21<6:12:17,  3.24s/it]
 53%|█████▎    | 7836/14725 [7:08:24<6:13:43,  3.25s/it]
 53%|█████▎    | 7837/14725 [7:08:27<6:16:08,  3.28s/it]
 53%|██
0: {'loss': 0.6885, 'grad_norm': 0.42752266331915306, 'learning_rate': 1e-05, 'epoch': 0.53}
0: ███▎    | 7838/14725 [7:08:30<6:17:15,  3.29s/it]
 53%|█████▎    | 7839/14725 [7:08:34<6:20:19,  3.31s/it]
 53%|█████▎    | 7840/14725 [7:08:37<6:18:44,  3.30s/it]
                                                        

 53%|█████▎    | 7840/14725 [7:08:37<6:18:44,  3.30s/it]
 53%|█████▎    | 7841/14725 [7:08:40<6:17:55,  3.29s/it]
 53%|█████▎    | 7842/14725 [7:08:44<6:14:17,  3.26s/it]
 53%|█████▎    | 7843/14725 [7:08:47<6:17:45,  3.29s/it]
 53%|█████▎    | 7844/14725 [7:08:50<6:11:05,  3.24s/it]
 53%|█████▎    | 7845/14725 [7:08:53<6:13:48,  3.26s/it]
 53%|█████▎    | 7846/14725 [7:08:56<6:07:46,  3.21s/it]
 53%|█████▎    | 7847/14725 [7:09:00<6:05:57,  3.19s/it]
 53%|█████▎    | 7848/14725 [7:09:03<6:12:25,  3.25s/it]
 53%|█████▎    | 7849/14725 [7:09:06<6:12:59,  3.25s/it]
 53%|█████▎    | 7850/14725 [7:09:10<6:19:07,  3.31s/it]
           
0: {'loss': 0.7027, 'grad_norm': 0.4294156629697269, 'learning_rate': 1e-05, 'epoch': 0.53}
0: {'loss': 0.6933, 'grad_norm': 0.44489192089973206, 'learning_rate': 1e-05, 'epoch': 0.53}
0:                                              

 53%|█████▎    | 7850/14725 [7:09:10<6:19:07,  3.31s/it]
 53%|█████▎    | 7851/14725 [7:09:13<6:18:41,  3.31s/it]
 53%|█████▎    | 7852/14725 [7:09:16<6:12:58,  3.26s/it]
 53%|█████▎    | 7853/14725 [7:09:19<6:13:04,  3.26s/it]
 53%|█████▎    | 7854/14725 [7:09:23<6:14:48,  3.27s/it]
 53%|█████▎    | 7855/14725 [7:09:26<6:15:40,  3.28s/it]
 53%|█████▎    | 7856/14725 [7:09:29<6:13:13,  3.26s/it]
 53%|█████▎    | 7857/14725 [7:09:33<6:15:09,  3.28s/it]
 53%|█████▎    | 7858/14725 [7:09:36<6:12:58,  3.26s/it]
 53%|█████▎    | 7859/14725 [7:09:39<6:20:59,  3.33s/it]
 53%|█████▎    | 7860/14725 [7:09:43<6:18:54,  3.31s/it]
                                                        

 53%|█████▎    | 7860/14725 [7:09:43<6:18:54,  3.31s/it]
 53%|█████▎    | 7861/14725 [7:09:46<6:16:42,  3.29s/it]
 53%|█████�
0: {'loss': 0.6744, 'grad_norm': 0.397628960628788, 'learning_rate': 1e-05, 'epoch': 0.53}
0: �    | 7862/14725 [7:09:49<6:12:13,  3.25s/it]
 53%|█████▎    | 7863/14725 [7:09:52<6:14:37,  3.28s/it]
 53%|█████▎    | 7864/14725 [7:09:55<6:11:18,  3.25s/it]
 53%|█████▎    | 7865/14725 [7:09:59<6:10:14,  3.24s/it]
 53%|█████▎    | 7866/14725 [7:10:02<6:15:35,  3.29s/it]
 53%|█████▎    | 7867/14725 [7:10:05<6:16:07,  3.29s/it]
 53%|█████▎    | 7868/14725 [7:10:09<6:13:10,  3.27s/it]
 53%|█████▎    | 7869/14725 [7:10:12<6:11:37,  3.25s/it]
 53%|█████▎    | 7870/14725 [7:10:15<6:11:59,  3.26s/it]
                                                        

 53%|█████▎    | 7870/14725 [7:10:15<6:11:59,  3.26s/it]
 53%|█████▎    | 7871/14725 [7:10:19<6:21:31,  3.34s/it]
 53%|█████▎    | 7872/14725 [7:10:22<6:23:30,  3.36s/it]
 53%|█████▎    | 7873/14725 [7:10:25<6:24:24,  3.37s/it]
 53%|█████▎    | 7874/14725 [7:10:29<6:20:56,  3.34s/it]
 53%|█████�
0: {'loss': 0.7019, 'grad_norm': 0.423299097910882, 'learning_rate': 1e-05, 'epoch': 0.54}
0: �    | 7875/14725 [7:10:32<6:14:14,  3.28s/it]
 53%|█████▎    | 7876/14725 [7:10:35<6:11:23,  3.25s/it]
 53%|█████▎    | 7877/14725 [7:10:38<6:09:40,  3.24s/it]
 54%|█████▎    | 7878/14725 [7:10:41<6:05:03,  3.20s/it]
 54%|█████▎    | 7879/14725 [7:10:44<6:02:23,  3.18s/it]
 54%|█████▎    | 7880/14725 [7:10:48<6:02:41,  3.18s/it]
                                                        

 54%|█████▎    | 7880/14725 [7:10:48<6:02:41,  3.18s/it]
 54%|█████▎    | 7881/14725 [7:10:51<6:05:25,  3.20s/it]
 54%|█████▎    | 7882/14725 [7:10:54<6:07:52,  3.23s/it]
 54%|█████▎    | 7883/14725 [7:10:57<6:06:32,  3.21s/it]
 54%|█████▎    | 7884/14725 [7:11:00<6:03:25,  3.19s/it]
 54%|█████▎    | 7885/14725 [7:11:04<6:10:25,  3.25s/it]
 54%|█████▎    | 7886/14725 [7:11:07<6:06:56,  3.22s/it]
 54%|█████▎    | 7887/14725 [7:11:10<6:07:18,  3.22s/it]
 54%|█████�
0: {'loss': 0.7024, 'grad_norm': 0.4036110292610097, 'learning_rate': 1e-05, 'epoch': 0.54}
0: �    | 7888/14725 [7:11:14<6:09:48,  3.25s/it]
 54%|█████▎    | 7889/14725 [7:11:17<6:16:46,  3.31s/it]
 54%|█████▎    | 7890/14725 [7:11:20<6:11:19,  3.26s/it]
                                                        

 54%|█████▎    | 7890/14725 [7:11:20<6:11:19,  3.26s/it]
 54%|█████▎    | 7891/14725 [7:11:23<6:09:23,  3.24s/it]
 54%|█████▎    | 7892/14725 [7:11:27<6:07:09,  3.22s/it]
 54%|█████▎    | 7893/14725 [7:11:30<6:09:26,  3.24s/it]
 54%|█████▎    | 7894/14725 [7:11:33<6:03:17,  3.19s/it]
 54%|█████▎    | 7895/14725 [7:11:36<6:08:36,  3.24s/it]
 54%|█████▎    | 7896/14725 [7:11:39<6:03:13,  3.19s/it]
 54%|█████▎    | 7897/14725 [7:11:43<6:06:41,  3.22s/it]
 54%|█████▎    | 7898/14725 [7:11:46<6:07:17,  3.23s/it]
 54%|█████▎    | 7899/14725 [7:11:49<6:11:11,  3.26s/it]
 54%|█████▎    | 7900/14725 [7:11:52<6:10:51,  3.26s/it]
                      
0: {'loss': 0.6945, 'grad_norm': 0.4339000202665986, 'learning_rate': 1e-05, 'epoch': 0.54}
0: {'loss': 0.6892, 'grad_norm': 0.4304312557490236, 'learning_rate': 1e-05, 'epoch': 0.54}
0:                                   

 54%|█████▎    | 7900/14725 [7:11:52<6:10:51,  3.26s/it]
 54%|█████▎    | 7901/14725 [7:11:56<6:10:58,  3.26s/it]
 54%|█████▎    | 7902/14725 [7:11:59<6:10:44,  3.26s/it]
 54%|█████▎    | 7903/14725 [7:12:02<6:10:14,  3.26s/it]
 54%|█████▎    | 7904/14725 [7:12:05<6:08:29,  3.24s/it]
 54%|█████▎    | 7905/14725 [7:12:09<6:15:07,  3.30s/it]
 54%|█████▎    | 7906/14725 [7:12:12<6:12:09,  3.27s/it]
 54%|█████▎    | 7907/14725 [7:12:15<6:12:50,  3.28s/it]
 54%|█████▎    | 7908/14725 [7:12:19<6:12:22,  3.28s/it]
 54%|█████▎    | 7909/14725 [7:12:22<6:04:08,  3.21s/it]
 54%|█████▎    | 7910/14725 [7:12:25<6:04:35,  3.21s/it]
                                                        

 54%|█████▎    | 7910/14725 [7:12:25<6:04:35,  3.21s/it]
 54%|█████▎    | 7911/14725 [7:12:28<6:07:27,  3.24s/it]
 54%|█████▎    | 7912
0: {'loss': 0.6818, 'grad_norm': 0.41516743490711683, 'learning_rate': 1e-05, 'epoch': 0.54}
0: /14725 [7:12:31<6:07:01,  3.23s/it]
 54%|█████▎    | 7913/14725 [7:12:35<6:06:51,  3.23s/it]
 54%|█████▎    | 7914/14725 [7:12:38<6:08:38,  3.25s/it]
 54%|█████▍    | 7915/14725 [7:12:41<6:07:52,  3.24s/it]
 54%|█████▍    | 7916/14725 [7:12:44<6:07:41,  3.24s/it]
 54%|█████▍    | 7917/14725 [7:12:48<6:07:14,  3.24s/it]
 54%|█████▍    | 7918/14725 [7:12:51<6:07:00,  3.23s/it]
 54%|█████▍    | 7919/14725 [7:12:54<6:09:33,  3.26s/it]
 54%|█████▍    | 7920/14725 [7:12:57<6:09:12,  3.26s/it]
                                                        

 54%|█████▍    | 7920/14725 [7:12:57<6:09:12,  3.26s/it]
 54%|█████▍    | 7921/14725 [7:13:01<6:09:49,  3.26s/it]
 54%|█████▍    | 7922/14725 [7:13:04<6:07:10,  3.24s/it]
 54%|█████▍    | 7923/14725 [7:13:07<6:05:27,  3.22s/it]
 54%|█████▍    | 7924/14725 [7:13:10<6:00:59,  3.18s/it]
 54%|█████▍    | 7925
0: {'loss': 0.6905, 'grad_norm': 0.39153890704242345, 'learning_rate': 1e-05, 'epoch': 0.54}
0: /14725 [7:13:13<6:02:50,  3.20s/it]
 54%|█████▍    | 7926/14725 [7:13:17<6:02:44,  3.20s/it]
 54%|█████▍    | 7927/14725 [7:13:20<6:06:40,  3.24s/it]
 54%|█████▍    | 7928/14725 [7:13:23<6:12:19,  3.29s/it]
 54%|█████▍    | 7929/14725 [7:13:27<6:09:56,  3.27s/it]
 54%|█████▍    | 7930/14725 [7:13:30<6:10:55,  3.28s/it]
                                                        

 54%|█████▍    | 7930/14725 [7:13:30<6:10:55,  3.28s/it]
 54%|█████▍    | 7931/14725 [7:13:33<6:11:54,  3.28s/it]
 54%|█████▍    | 7932/14725 [7:13:36<6:10:54,  3.28s/it]
 54%|█████▍    | 7933/14725 [7:13:40<6:12:50,  3.29s/it]
 54%|█████▍    | 7934/14725 [7:13:43<6:13:47,  3.30s/it]
 54%|█████▍    | 7935/14725 [7:13:46<6:06:16,  3.24s/it]
 54%|█████▍    | 7936/14725 [7:13:50<6:12:29,  3.29s/it]
 54%|█████▍    | 7937/14725 [7:13:53<6:11:28,  3.28s/it]
 54%|█████▍    | 7938
0: {'loss': 0.7133, 'grad_norm': 0.41738563602171536, 'learning_rate': 1e-05, 'epoch': 0.54}
0: /14725 [7:13:56<6:11:27,  3.28s/it]
 54%|█████▍    | 7939/14725 [7:13:59<6:09:17,  3.27s/it]
 54%|█████▍    | 7940/14725 [7:14:03<6:08:29,  3.26s/it]
                                                        

 54%|█████▍    | 7940/14725 [7:14:03<6:08:29,  3.26s/it]
 54%|█████▍    | 7941/14725 [7:14:06<6:02:41,  3.21s/it]
 54%|█████▍    | 7942/14725 [7:14:09<5:59:44,  3.18s/it]
 54%|█████▍    | 7943/14725 [7:14:12<6:06:34,  3.24s/it]
 54%|█████▍    | 7944/14725 [7:14:15<6:06:57,  3.25s/it]
 54%|█████▍    | 7945/14725 [7:14:19<6:07:06,  3.25s/it]
 54%|█████▍    | 7946/14725 [7:14:22<6:06:00,  3.24s/it]
 54%|█████▍    | 7947/14725 [7:14:25<6:08:13,  3.26s/it]
 54%|█████▍    | 7948/14725 [7:14:28<6:06:22,  3.24s/it]
 54%|█████▍    | 7949/14725 [7:14:32<6:03:22,  3.22s/it]
 54%|█████▍    | 7950/14725 [7:14:35<6:08:20,  3.26s/it]
                                 
0: {'loss': 0.6952, 'grad_norm': 0.49871190473756694, 'learning_rate': 1e-05, 'epoch': 0.54}
0: {'loss': 0.6787, 'grad_norm': 0.41402767621139014, 'learning_rate': 1e-05, 'epoch': 0.54}
0:                        

 54%|█████▍    | 7950/14725 [7:14:35<6:08:20,  3.26s/it]
 54%|█████▍    | 7951/14725 [7:14:38<6:17:10,  3.34s/it]
 54%|█████▍    | 7952/14725 [7:14:42<6:15:06,  3.32s/it]
 54%|█████▍    | 7953/14725 [7:14:45<6:19:35,  3.36s/it]
 54%|█████▍    | 7954/14725 [7:14:48<6:12:33,  3.30s/it]
 54%|█████▍    | 7955/14725 [7:14:52<6:08:53,  3.27s/it]
 54%|█████▍    | 7956/14725 [7:14:55<6:04:35,  3.23s/it]
 54%|█████▍    | 7957/14725 [7:14:58<6:02:43,  3.22s/it]
 54%|█████▍    | 7958/14725 [7:15:01<6:06:12,  3.25s/it]
 54%|█████▍    | 7959/14725 [7:15:04<6:06:19,  3.25s/it]
 54%|█████▍    | 7960/14725 [7:15:08<6:06:27,  3.25s/it]
                                                        

 54%|█████▍    | 7960/14725 [7:15:08<6:06:27,  3.25s/it]
 54%|█████▍    | 7961/14725 [7:15:11<6:04:01,  3.23s/it]
 54%|█████▍    | 7962/14725 [7:1
0: {'loss': 0.6875, 'grad_norm': 0.441222812289891, 'learning_rate': 1e-05, 'epoch': 0.54}
0: 5:14<6:04:28,  3.23s/it]
 54%|█████▍    | 7963/14725 [7:15:17<6:07:00,  3.26s/it]
 54%|█████▍    | 7964/14725 [7:15:21<6:12:43,  3.31s/it]
 54%|█████▍    | 7965/14725 [7:15:24<6:10:42,  3.29s/it]
 54%|█████▍    | 7966/14725 [7:15:28<6:15:53,  3.34s/it]
 54%|█████▍    | 7967/14725 [7:15:31<6:16:56,  3.35s/it]
 54%|█████▍    | 7968/14725 [7:15:34<6:16:44,  3.35s/it]
 54%|█████▍    | 7969/14725 [7:15:38<6:16:33,  3.34s/it]
 54%|█████▍    | 7970/14725 [7:15:41<6:12:24,  3.31s/it]
                                                        

 54%|█████▍    | 7970/14725 [7:15:41<6:12:24,  3.31s/it]
 54%|█████▍    | 7971/14725 [7:15:44<6:09:37,  3.28s/it]
 54%|█████▍    | 7972/14725 [7:15:47<6:07:38,  3.27s/it]
 54%|█████▍    | 7973/14725 [7:15:50<6:05:52,  3.25s/it]
 54%|█████▍    | 7974/14725 [7:15:54<6:08:58,  3.28s/it]
 54%|█████▍    | 7975/14725 [7:1
0: {'loss': 0.6836, 'grad_norm': 0.456129178677506, 'learning_rate': 1e-05, 'epoch': 0.54}
0: 5:57<6:05:25,  3.25s/it]
 54%|█████▍    | 7976/14725 [7:16:00<6:10:36,  3.29s/it]
 54%|█████▍    | 7977/14725 [7:16:04<6:05:23,  3.25s/it]
 54%|█████▍    | 7978/14725 [7:16:07<5:57:04,  3.18s/it]
 54%|█████▍    | 7979/14725 [7:16:10<5:59:21,  3.20s/it]
 54%|█████▍    | 7980/14725 [7:16:13<6:06:44,  3.26s/it]
                                                        

 54%|█████▍    | 7980/14725 [7:16:13<6:06:44,  3.26s/it]
 54%|█████▍    | 7981/14725 [7:16:16<6:05:07,  3.25s/it]
 54%|█████▍    | 7982/14725 [7:16:20<6:09:30,  3.29s/it]
 54%|█████▍    | 7983/14725 [7:16:23<6:10:28,  3.30s/it]
 54%|█████▍    | 7984/14725 [7:16:26<6:09:15,  3.29s/it]
 54%|█████▍    | 7985/14725 [7:16:30<6:10:29,  3.30s/it]
 54%|█████▍    | 7986/14725 [7:16:33<6:08:56,  3.28s/it]
 54%|█████▍    | 7987/14725 [7:16:36<6:09:35,  3.29s/it]
 54%|█████▍    | 7988/14725 [7:1
0: {'loss': 0.6944, 'grad_norm': 0.42489019282530827, 'learning_rate': 1e-05, 'epoch': 0.54}
0: 6:40<6:13:23,  3.33s/it]
 54%|█████▍    | 7989/14725 [7:16:43<6:19:52,  3.38s/it]
 54%|█████▍    | 7990/14725 [7:16:47<6:17:26,  3.36s/it]
                                                        

 54%|█████▍    | 7990/14725 [7:16:47<6:17:26,  3.36s/it]
 54%|█████▍    | 7991/14725 [7:16:50<6:11:18,  3.31s/it]
 54%|█████▍    | 7992/14725 [7:16:53<6:11:51,  3.31s/it]
 54%|█████▍    | 7993/14725 [7:16:56<6:06:41,  3.27s/it]
 54%|█████▍    | 7994/14725 [7:16:59<6:02:01,  3.23s/it]
 54%|█████▍    | 7995/14725 [7:17:02<5:59:54,  3.21s/it]
 54%|█████▍    | 7996/14725 [7:17:05<5:51:12,  3.13s/it]
 54%|█████▍    | 7997/14725 [7:17:09<5:54:03,  3.16s/it]
 54%|█████▍    | 7998/14725 [7:17:12<5:54:06,  3.16s/it]
 54%|█████▍    | 7999/14725 [7:17:15<6:03:04,  3.24s/it]
 54%|█████▍    | 8000/14725 [7:17:18<6:03:21,  3.24s/it]
                                            
0: {'loss': 0.687, 'grad_norm': 0.426317410377034, 'learning_rate': 1e-05, 'epoch': 0.54}
0: {'loss': 0.7031, 'grad_norm': 0.4232299724239775, 'learning_rate': 1e-05, 'epoch': 0.54}
0:             

 54%|█████▍    | 8000/14725 [7:17:18<6:03:21,  3.24s/it]
 54%|█████▍    | 8001/14725 [7:17:22<5:58:44,  3.20s/it]
 54%|█████▍    | 8002/14725 [7:17:25<6:00:14,  3.22s/it]
 54%|█████▍    | 8003/14725 [7:17:28<6:07:27,  3.28s/it]
 54%|█████▍    | 8004/14725 [7:17:32<6:07:10,  3.28s/it]
 54%|█████▍    | 8005/14725 [7:17:35<6:03:04,  3.24s/it]
 54%|█████▍    | 8006/14725 [7:17:38<6:00:38,  3.22s/it]
 54%|█████▍    | 8007/14725 [7:17:41<6:02:20,  3.24s/it]
 54%|█████▍    | 8008/14725 [7:17:45<6:14:47,  3.35s/it]
 54%|█████▍    | 8009/14725 [7:17:48<6:13:20,  3.34s/it]
 54%|█████▍    | 8010/14725 [7:17:51<6:04:08,  3.25s/it]
                                                        

 54%|█████▍    | 8010/14725 [7:17:51<6:04:08,  3.25s/it]
 54%|█████▍    | 8011/14725 [7:17:54<6:07:32,  3.28s/it]
 54%|█████▍    | 8012/14725 [7:17:58<6:08:0
0: {'loss': 0.6912, 'grad_norm': 0.38759933510878736, 'learning_rate': 1e-05, 'epoch': 0.54}
0: 1,  3.29s/it]
 54%|█████▍    | 8013/14725 [7:18:01<6:03:30,  3.25s/it]
 54%|█████▍    | 8014/14725 [7:18:04<6:10:50,  3.32s/it]
 54%|█████▍    | 8015/14725 [7:18:08<6:11:51,  3.33s/it]
 54%|█████▍    | 8016/14725 [7:18:11<6:11:45,  3.32s/it]
 54%|█████▍    | 8017/14725 [7:18:15<6:15:49,  3.36s/it]
 54%|█████▍    | 8018/14725 [7:18:18<6:10:03,  3.31s/it]
 54%|█████▍    | 8019/14725 [7:18:21<6:07:58,  3.29s/it]
 54%|█████▍    | 8020/14725 [7:18:24<6:12:28,  3.33s/it]
                                                        

 54%|█████▍    | 8020/14725 [7:18:24<6:12:28,  3.33s/it]
 54%|█████▍    | 8021/14725 [7:18:28<6:06:33,  3.28s/it]
 54%|█████▍    | 8022/14725 [7:18:31<6:03:57,  3.26s/it]
 54%|█████▍    | 8023/14725 [7:18:34<6:01:50,  3.24s/it]
 54%|█████▍    | 8024/14725 [7:18:37<6:06:45,  3.28s/it]
 54%|█████▍    | 8025/14725 [7:18:41<6:02:4
0: {'loss': 0.6896, 'grad_norm': 0.43483068453568696, 'learning_rate': 1e-05, 'epoch': 0.55}
0: 3,  3.25s/it]
 55%|█████▍    | 8026/14725 [7:18:44<5:57:00,  3.20s/it]
 55%|█████▍    | 8027/14725 [7:18:47<5:58:59,  3.22s/it]
 55%|█████▍    | 8028/14725 [7:18:50<5:59:59,  3.23s/it]
 55%|█████▍    | 8029/14725 [7:18:53<5:58:42,  3.21s/it]
 55%|█████▍    | 8030/14725 [7:18:57<6:02:31,  3.25s/it]
                                                        

 55%|█████▍    | 8030/14725 [7:18:57<6:02:31,  3.25s/it]
 55%|█████▍    | 8031/14725 [7:19:00<6:12:32,  3.34s/it]
 55%|█████▍    | 8032/14725 [7:19:03<6:11:17,  3.33s/it]
 55%|█████▍    | 8033/14725 [7:19:07<6:08:36,  3.30s/it]
 55%|█████▍    | 8034/14725 [7:19:10<6:05:52,  3.28s/it]
 55%|█████▍    | 8035/14725 [7:19:13<6:06:40,  3.29s/it]
 55%|█████▍    | 8036/14725 [7:19:16<6:02:39,  3.25s/it]
 55%|█████▍    | 8037/14725 [7:19:20<6:08:23,  3.30s/it]
 55%|█████▍    | 8038/14725 [7:19:23<6:13:5
0: {'loss': 0.6772, 'grad_norm': 0.40537428793590735, 'learning_rate': 1e-05, 'epoch': 0.55}
0: 6,  3.36s/it]
 55%|█████▍    | 8039/14725 [7:19:27<6:08:58,  3.31s/it]
 55%|█████▍    | 8040/14725 [7:19:30<6:09:41,  3.32s/it]
                                                        

 55%|█████▍    | 8040/14725 [7:19:30<6:09:41,  3.32s/it]
 55%|█████▍    | 8041/14725 [7:19:33<6:13:37,  3.35s/it]
 55%|█████▍    | 8042/14725 [7:19:37<6:10:13,  3.32s/it]
 55%|█████▍    | 8043/14725 [7:19:40<6:12:24,  3.34s/it]
 55%|█████▍    | 8044/14725 [7:19:43<6:10:47,  3.33s/it]
 55%|█████▍    | 8045/14725 [7:19:47<6:08:37,  3.31s/it]
 55%|█████▍    | 8046/14725 [7:19:50<6:06:46,  3.29s/it]
 55%|█████▍    | 8047/14725 [7:19:53<6:00:32,  3.24s/it]
 55%|█████▍    | 8048/14725 [7:19:56<5:56:15,  3.20s/it]
 55%|█████▍    | 8049/14725 [7:19:59<5:55:46,  3.20s/it]
 55%|█████▍    | 8050/14725 [7:20:02<5:54:41,  3.19s/it]
                                                       
0: {'loss': 0.6909, 'grad_norm': 0.41022290963509755, 'learning_rate': 1e-05, 'epoch': 0.55}
0: {'loss': 0.6984, 'grad_norm': 0.39892654735269023, 'learning_rate': 1e-05, 'epoch': 0.55}
0:  

 55%|█████▍    | 8050/14725 [7:20:02<5:54:41,  3.19s/it]
 55%|█████▍    | 8051/14725 [7:20:06<5:55:24,  3.20s/it]
 55%|█████▍    | 8052/14725 [7:20:09<6:02:35,  3.26s/it]
 55%|█████▍    | 8053/14725 [7:20:12<5:56:04,  3.20s/it]
 55%|█████▍    | 8054/14725 [7:20:15<5:58:00,  3.22s/it]
 55%|█████▍    | 8055/14725 [7:20:18<5:55:46,  3.20s/it]
 55%|█████▍    | 8056/14725 [7:20:22<6:03:13,  3.27s/it]
 55%|█████▍    | 8057/14725 [7:20:25<6:05:58,  3.29s/it]
 55%|█████▍    | 8058/14725 [7:20:28<6:03:51,  3.27s/it]
 55%|█████▍    | 8059/14725 [7:20:32<6:07:51,  3.31s/it]
 55%|█████▍    | 8060/14725 [7:20:35<6:02:42,  3.27s/it]
                                                        

 55%|█████▍    | 8060/14725 [7:20:35<6:02:42,  3.27s/it]
 55%|█████▍    | 8061/14725 [7:20:38<6:03:41,  3.27s/it]
 55%|█████▍    | 8062/14725 [7:20:41<5:58:57,  3.23s/i
0: {'loss': 0.6952, 'grad_norm': 0.39148051562713176, 'learning_rate': 1e-05, 'epoch': 0.55}
0: t]
 55%|█████▍    | 8063/14725 [7:20:45<5:59:17,  3.24s/it]
 55%|█████▍    | 8064/14725 [7:20:48<5:59:24,  3.24s/it]
 55%|█████▍    | 8065/14725 [7:20:51<6:04:59,  3.29s/it]
 55%|█████▍    | 8066/14725 [7:20:54<5:59:59,  3.24s/it]
 55%|█████▍    | 8067/14725 [7:20:58<5:58:56,  3.23s/it]
 55%|█████▍    | 8068/14725 [7:21:01<6:01:31,  3.26s/it]
 55%|█████▍    | 8069/14725 [7:21:04<6:04:19,  3.28s/it]
 55%|█████▍    | 8070/14725 [7:21:07<5:55:32,  3.21s/it]
                                                        

 55%|█████▍    | 8070/14725 [7:21:07<5:55:32,  3.21s/it]
 55%|█████▍    | 8071/14725 [7:21:11<5:53:33,  3.19s/it]
 55%|█████▍    | 8072/14725 [7:21:14<6:01:33,  3.26s/it]
 55%|█████▍    | 8073/14725 [7:21:17<5:59:29,  3.24s/it]
 55%|█████▍    | 8074/14725 [7:21:20<6:02:55,  3.27s/it]
 55%|█████▍    | 8075/14725 [7:21:24<5:54:21,  3.20s/i
0: {'loss': 0.6999, 'grad_norm': 0.43398395960609526, 'learning_rate': 1e-05, 'epoch': 0.55}
0: t]
 55%|█████▍    | 8076/14725 [7:21:27<5:56:25,  3.22s/it]
 55%|█████▍    | 8077/14725 [7:21:30<5:50:56,  3.17s/it]
 55%|█████▍    | 8078/14725 [7:21:33<5:56:10,  3.21s/it]
 55%|█████▍    | 8079/14725 [7:21:36<6:00:22,  3.25s/it]
 55%|█████▍    | 8080/14725 [7:21:40<6:03:49,  3.29s/it]
                                                        

 55%|█████▍    | 8080/14725 [7:21:40<6:03:49,  3.29s/it]
 55%|█████▍    | 8081/14725 [7:21:43<6:12:19,  3.36s/it]
 55%|█████▍    | 8082/14725 [7:21:47<6:10:00,  3.34s/it]
 55%|█████▍    | 8083/14725 [7:21:50<6:05:46,  3.30s/it]
 55%|█████▍    | 8084/14725 [7:21:53<6:02:45,  3.28s/it]
 55%|█████▍    | 8085/14725 [7:21:56<6:05:00,  3.30s/it]
 55%|█████▍    | 8086/14725 [7:22:00<6:15:44,  3.40s/it]
 55%|█████▍    | 8087/14725 [7:22:03<6:08:31,  3.33s/it]
 55%|█████▍    | 8088/14725 [7:22:06<6:01:47,  3.27s/i
0: {'loss': 0.7081, 'grad_norm': 0.40739906856432484, 'learning_rate': 1e-05, 'epoch': 0.55}
0: {'loss': 0.7103, 'grad_norm': 0.4006134303146791, 'learning_rate': 1e-05, 'epoch': 0.55}
0: t]
 55%|█████▍    | 8089/14725 [7:22:10<5:58:27,  3.24s/it]
 55%|█████▍    | 8090/14725 [7:22:13<6:13:42,  3.38s/it]
                                                        

 55%|█████▍    | 8090/14725 [7:22:13<6:13:42,  3.38s/it]
 55%|█████▍    | 8091/14725 [7:22:16<6:07:37,  3.32s/it]
 55%|█████▍    | 8092/14725 [7:22:20<5:59:52,  3.26s/it]
 55%|█████▍    | 8093/14725 [7:22:23<6:02:59,  3.28s/it]
 55%|█████▍    | 8094/14725 [7:22:26<6:04:19,  3.30s/it]
 55%|█████▍    | 8095/14725 [7:22:30<6:05:45,  3.31s/it]
 55%|█████▍    | 8096/14725 [7:22:33<6:06:32,  3.32s/it]
 55%|█████▍    | 8097/14725 [7:22:36<6:05:49,  3.31s/it]
 55%|█████▍    | 8098/14725 [7:22:40<6:06:22,  3.32s/it]
 55%|█████▌    | 8099/14725 [7:22:43<6:05:20,  3.31s/it]
 55%|█████▌    | 8100/14725 [7:22:46<6:03:42,  3.29s/it]
                                                        

 55%|█
0: {'loss': 0.6936, 'grad_norm': 0.40493479509649627, 'learning_rate': 1e-05, 'epoch': 0.55}
0: ████▌    | 8100/14725 [7:22:46<6:03:42,  3.29s/it]
 55%|█████▌    | 8101/14725 [7:22:49<6:05:10,  3.31s/it]
 55%|█████▌    | 8102/14725 [7:22:53<6:08:33,  3.34s/it]
 55%|█████▌    | 8103/14725 [7:22:56<6:03:49,  3.30s/it]
 55%|█████▌    | 8104/14725 [7:22:59<6:08:58,  3.34s/it]
 55%|█████▌    | 8105/14725 [7:23:03<6:09:47,  3.35s/it]
 55%|█████▌    | 8106/14725 [7:23:06<6:03:34,  3.30s/it]
 55%|█████▌    | 8107/14725 [7:23:09<5:56:58,  3.24s/it]
 55%|█████▌    | 8108/14725 [7:23:12<5:49:29,  3.17s/it]
 55%|█████▌    | 8109/14725 [7:23:15<5:53:46,  3.21s/it]
 55%|█████▌    | 8110/14725 [7:23:19<6:08:12,  3.34s/it]
                                                        

 55%|█████▌    | 8110/14725 [7:23:19<6:08:12,  3.34s/it]
 55%|█████▌    | 8111/14725 [7:23:22<6:04:53,  3.31s/it]
 55%|█████▌    | 8112/14725 [7:23:26<6:04:13,  3.30s/it]
 55%|█
0: {'loss': 0.6986, 'grad_norm': 0.4886406730351573, 'learning_rate': 1e-05, 'epoch': 0.55}
0: ████▌    | 8113/14725 [7:23:29<5:59:55,  3.27s/it]
 55%|█████▌    | 8114/14725 [7:23:32<5:57:16,  3.24s/it]
 55%|█████▌    | 8115/14725 [7:23:35<6:03:42,  3.30s/it]
 55%|█████▌    | 8116/14725 [7:23:39<6:00:34,  3.27s/it]
 55%|█████▌    | 8117/14725 [7:23:42<6:04:28,  3.31s/it]
 55%|█████▌    | 8118/14725 [7:23:45<6:06:28,  3.33s/it]
 55%|█████▌    | 8119/14725 [7:23:48<5:57:39,  3.25s/it]
 55%|█████▌    | 8120/14725 [7:23:52<5:57:07,  3.24s/it]
                                                        

 55%|█████▌    | 8120/14725 [7:23:52<5:57:07,  3.24s/it]
 55%|█████▌    | 8121/14725 [7:23:55<6:07:53,  3.34s/it]
 55%|█████▌    | 8122/14725 [7:23:59<6:10:50,  3.37s/it]
 55%|█████▌    | 8123/14725 [7:24:02<6:04:07,  3.31s/it]
 55%|█████▌    | 8124/14725 [7:24:05<6:03:00,  3.30s/it]
 55%|█████▌    | 8125/14725 [7:24:09<6:09:29,  3.36s/it]
 55%|█
0: {'loss': 0.6985, 'grad_norm': 0.40155128392472356, 'learning_rate': 1e-05, 'epoch': 0.55}
0: ████▌    | 8126/14725 [7:24:12<6:05:53,  3.33s/it]
 55%|█████▌    | 8127/14725 [7:24:16<6:18:46,  3.44s/it]
 55%|█████▌    | 8128/14725 [7:24:19<6:10:38,  3.37s/it]
 55%|█████▌    | 8129/14725 [7:24:22<6:08:11,  3.35s/it]
 55%|█████▌    | 8130/14725 [7:24:25<6:05:26,  3.32s/it]
                                                        

 55%|█████▌    | 8130/14725 [7:24:25<6:05:26,  3.32s/it]
 55%|█████▌    | 8131/14725 [7:24:29<6:07:59,  3.35s/it]
 55%|█████▌    | 8132/14725 [7:24:32<6:03:52,  3.31s/it]
 55%|█████▌    | 8133/14725 [7:24:35<6:08:40,  3.36s/it]
 55%|█████▌    | 8134/14725 [7:24:39<6:03:47,  3.31s/it]
 55%|█████▌    | 8135/14725 [7:24:42<6:02:30,  3.30s/it]
 55%|█████▌    | 8136/14725 [7:24:45<6:05:40,  3.33s/it]
 55%|█████▌    | 8137/14725 [7:24:49<6:05:47,  3.33s/it]
 55%|█████▌    | 8138/14725 [7:24:52<6:00:52,  3.29s/it]
 55%|█
0: {'loss': 0.6778, 'grad_norm': 0.38940918516997064, 'learning_rate': 1e-05, 'epoch': 0.55}
0: {'loss': 0.6902, 'grad_norm': 0.4355485699247318, 'learning_rate': 1e-05, 'epoch': 0.55}
0: ████▌    | 8139/14725 [7:24:55<5:58:18,  3.26s/it]
 55%|█████▌    | 8140/14725 [7:24:58<5:59:02,  3.27s/it]
                                                        

 55%|█████▌    | 8140/14725 [7:24:58<5:59:02,  3.27s/it]
 55%|█████▌    | 8141/14725 [7:25:02<6:06:03,  3.34s/it]
 55%|█████▌    | 8142/14725 [7:25:05<5:59:07,  3.27s/it]
 55%|█████▌    | 8143/14725 [7:25:08<6:01:16,  3.29s/it]
 55%|█████▌    | 8144/14725 [7:25:11<5:55:39,  3.24s/it]
 55%|█████▌    | 8145/14725 [7:25:15<6:01:43,  3.30s/it]
 55%|█████▌    | 8146/14725 [7:25:18<5:57:53,  3.26s/it]
 55%|█████▌    | 8147/14725 [7:25:21<5:55:19,  3.24s/it]
 55%|█████▌    | 8148/14725 [7:25:24<5:53:03,  3.22s/it]
 55%|█████▌    | 8149/14725 [7:25:28<5:56:13,  3.25s/it]
 55%|█████▌    | 8150/14725 [7:25:31<6:10:22,  3.38s/it]
                                                        

 55%|████�
0: {'loss': 0.7033, 'grad_norm': 0.4073522681845403, 'learning_rate': 1e-05, 'epoch': 0.55}
0: �▌    | 8150/14725 [7:25:31<6:10:22,  3.38s/it]
 55%|█████▌    | 8151/14725 [7:25:35<6:01:15,  3.30s/it]
 55%|█████▌    | 8152/14725 [7:25:38<5:58:22,  3.27s/it]
 55%|█████▌    | 8153/14725 [7:25:41<5:49:42,  3.19s/it]
 55%|█████▌    | 8154/14725 [7:25:44<5:50:15,  3.20s/it]
 55%|█████▌    | 8155/14725 [7:25:47<5:52:50,  3.22s/it]
 55%|█████▌    | 8156/14725 [7:25:51<6:06:41,  3.35s/it]
 55%|█████▌    | 8157/14725 [7:25:54<6:06:29,  3.35s/it]
 55%|█████▌    | 8158/14725 [7:25:57<6:00:58,  3.30s/it]
 55%|█████▌    | 8159/14725 [7:26:01<5:54:25,  3.24s/it]
 55%|█████▌    | 8160/14725 [7:26:04<5:56:33,  3.26s/it]
                                                        

 55%|█████▌    | 8160/14725 [7:26:04<5:56:33,  3.26s/it]
 55%|█████▌    | 8161/14725 [7:26:07<5:51:03,  3.21s/it]
 55%|█████▌    | 8162/14725 [7:26:10<5:48:58,  3.19s/it]
 55%|████�
0: {'loss': 0.6964, 'grad_norm': 0.4172012270394086, 'learning_rate': 1e-05, 'epoch': 0.55}
0: �▌    | 8163/14725 [7:26:13<5:50:46,  3.21s/it]
 55%|█████▌    | 8164/14725 [7:26:17<5:51:04,  3.21s/it]
 55%|█████▌    | 8165/14725 [7:26:20<5:49:20,  3.20s/it]
 55%|█████▌    | 8166/14725 [7:26:23<5:46:01,  3.17s/it]
 55%|█████▌    | 8167/14725 [7:26:26<5:51:17,  3.21s/it]
 55%|█████▌    | 8168/14725 [7:26:29<5:55:19,  3.25s/it]
 55%|█████▌    | 8169/14725 [7:26:33<6:01:31,  3.31s/it]
 55%|█████▌    | 8170/14725 [7:26:36<6:00:20,  3.30s/it]
                                                        

 55%|█████▌    | 8170/14725 [7:26:36<6:00:20,  3.30s/it]
 55%|█████▌    | 8171/14725 [7:26:39<5:58:53,  3.29s/it]
 55%|█████▌    | 8172/14725 [7:26:42<5:50:51,  3.21s/it]
 56%|█████▌    | 8173/14725 [7:26:46<5:47:12,  3.18s/it]
 56%|█████▌    | 8174/14725 [7:26:49<5:55:12,  3.25s/it]
 56%|█████▌    | 8175/14725 [7:26:52<5:59:57,  3.30s/it]
 56%|████�
0: {'loss': 0.7049, 'grad_norm': 0.41709443819252445, 'learning_rate': 1e-05, 'epoch': 0.56}
0: �▌    | 8176/14725 [7:26:56<6:00:27,  3.30s/it]
 56%|█████▌    | 8177/14725 [7:26:59<6:00:39,  3.30s/it]
 56%|█████▌    | 8178/14725 [7:27:02<6:00:35,  3.30s/it]
 56%|█████▌    | 8179/14725 [7:27:06<6:02:21,  3.32s/it]
 56%|█████▌    | 8180/14725 [7:27:09<5:54:01,  3.25s/it]
                                                        

 56%|█████▌    | 8180/14725 [7:27:09<5:54:01,  3.25s/it]
 56%|█████▌    | 8181/14725 [7:27:12<6:02:12,  3.32s/it]
 56%|█████▌    | 8182/14725 [7:27:16<6:01:30,  3.32s/it]
 56%|█████▌    | 8183/14725 [7:27:19<6:04:04,  3.34s/it]
 56%|█████▌    | 8184/14725 [7:27:22<5:59:32,  3.30s/it]
 56%|█████▌    | 8185/14725 [7:27:26<6:07:43,  3.37s/it]
 56%|█████▌    | 8186/14725 [7:27:29<6:04:43,  3.35s/it]
 56%|█████▌    | 8187/14725 [7:27:32<6:07:19,  3.37s/it]
 56%|█████▌    | 8188/14725 [7:27:36<6:05:30,  3.35s/it]
 56%|████�
0: {'loss': 0.7003, 'grad_norm': 0.4298394497485811, 'learning_rate': 1e-05, 'epoch': 0.56}
0: {'loss': 0.7058, 'grad_norm': 0.41889380192776915, 'learning_rate': 1e-05, 'epoch': 0.56}
0: �▌    | 8189/14725 [7:27:39<6:00:43,  3.31s/it]
 56%|█████▌    | 8190/14725 [7:27:42<5:59:29,  3.30s/it]
                                                        

 56%|█████▌    | 8190/14725 [7:27:42<5:59:29,  3.30s/it]
 56%|█████▌    | 8191/14725 [7:27:45<5:56:09,  3.27s/it]
 56%|█████▌    | 8192/14725 [7:27:49<5:54:39,  3.26s/it]
 56%|█████▌    | 8193/14725 [7:27:52<5:47:17,  3.19s/it]
 56%|█████▌    | 8194/14725 [7:27:55<5:48:48,  3.20s/it]
 56%|█████▌    | 8195/14725 [7:27:58<5:46:19,  3.18s/it]
 56%|█████▌    | 8196/14725 [7:28:01<5:45:06,  3.17s/it]
 56%|█████▌    | 8197/14725 [7:28:04<5:47:59,  3.20s/it]
 56%|█████▌    | 8198/14725 [7:28:08<5:53:18,  3.25s/it]
 56%|█████▌    | 8199/14725 [7:28:11<5:49:38,  3.21s/it]
 56%|█████▌    | 8200/14725 [7:28:14<5:53:15,  3.25s/it]
                                                        

 56%|█████▌    | 8
0: {'loss': 0.6947, 'grad_norm': 0.4260441976047693, 'learning_rate': 1e-05, 'epoch': 0.56}
0: 200/14725 [7:28:14<5:53:15,  3.25s/it]
 56%|█████▌    | 8201/14725 [7:28:18<5:56:18,  3.28s/it]
 56%|█████▌    | 8202/14725 [7:28:21<5:55:07,  3.27s/it]
 56%|█████▌    | 8203/14725 [7:28:24<5:53:28,  3.25s/it]
 56%|█████▌    | 8204/14725 [7:28:27<5:52:42,  3.25s/it]
 56%|█████▌    | 8205/14725 [7:28:31<5:55:25,  3.27s/it]
 56%|█████▌    | 8206/14725 [7:28:34<5:58:12,  3.30s/it]
 56%|█████▌    | 8207/14725 [7:28:37<5:57:52,  3.29s/it]
 56%|█████▌    | 8208/14725 [7:28:41<5:58:40,  3.30s/it]
 56%|█████▌    | 8209/14725 [7:28:44<6:05:03,  3.36s/it]
 56%|█████▌    | 8210/14725 [7:28:47<6:05:02,  3.36s/it]
                                                        

 56%|█████▌    | 8210/14725 [7:28:47<6:05:02,  3.36s/it]
 56%|█████▌    | 8211/14725 [7:28:51<6:04:55,  3.36s/it]
 56%|█████▌    | 8212/14725 [7:28:54<6:03:11,  3.35s/it]
 56%|█████▌    | 8
0: {'loss': 0.6906, 'grad_norm': 0.4001075471709703, 'learning_rate': 1e-05, 'epoch': 0.56}
0: 213/14725 [7:28:57<5:51:22,  3.24s/it]
 56%|█████▌    | 8214/14725 [7:29:01<6:02:33,  3.34s/it]
 56%|█████▌    | 8215/14725 [7:29:04<6:00:14,  3.32s/it]
 56%|█████▌    | 8216/14725 [7:29:07<5:59:15,  3.31s/it]
 56%|█████▌    | 8217/14725 [7:29:10<5:54:09,  3.27s/it]
 56%|█████▌    | 8218/14725 [7:29:14<5:54:04,  3.26s/it]
 56%|█████▌    | 8219/14725 [7:29:17<5:54:15,  3.27s/it]
 56%|█████▌    | 8220/14725 [7:29:20<5:53:53,  3.26s/it]
                                                        

 56%|█████▌    | 8220/14725 [7:29:20<5:53:53,  3.26s/it]
 56%|█████▌    | 8221/14725 [7:29:23<5:50:50,  3.24s/it]
 56%|█████▌    | 8222/14725 [7:29:27<6:00:21,  3.32s/it]
 56%|█████▌    | 8223/14725 [7:29:30<5:58:52,  3.31s/it]
 56%|█████▌    | 8224/14725 [7:29:34<6:00:21,  3.33s/it]
 56%|█████▌    | 8225/14725 [7:29:37<5:50:24,  3.23s/it]
 56%|█████▌    | 8
0: {'loss': 0.6786, 'grad_norm': 0.40464573382125907, 'learning_rate': 1e-05, 'epoch': 0.56}
0: 226/14725 [7:29:40<5:52:55,  3.26s/it]
 56%|█████▌    | 8227/14725 [7:29:43<5:55:01,  3.28s/it]
 56%|█████▌    | 8228/14725 [7:29:46<5:50:02,  3.23s/it]
 56%|█████▌    | 8229/14725 [7:29:50<5:49:37,  3.23s/it]
 56%|█████▌    | 8230/14725 [7:29:53<5:46:56,  3.20s/it]
                                                        

 56%|█████▌    | 8230/14725 [7:29:53<5:46:56,  3.20s/it]
 56%|█████▌    | 8231/14725 [7:29:56<5:52:16,  3.25s/it]
 56%|█████▌    | 8232/14725 [7:29:59<5:48:33,  3.22s/it]
 56%|█████▌    | 8233/14725 [7:30:02<5:42:33,  3.17s/it]
 56%|█████▌    | 8234/14725 [7:30:06<5:51:54,  3.25s/it]
 56%|█████▌    | 8235/14725 [7:30:09<5:54:13,  3.27s/it]
 56%|█████▌    | 8236/14725 [7:30:12<5:53:10,  3.27s/it]
 56%|█████▌    | 8237/14725 [7:30:15<5:50:30,  3.24s/it]
 56%|█████▌    | 8238/14725 [7:30:19<5:50:14,  3.24s/it]
 56%|█████▌    | 8
0: {'loss': 0.6922, 'grad_norm': 0.48524243033038655, 'learning_rate': 1e-05, 'epoch': 0.56}
0: {'loss': 0.6926, 'grad_norm': 0.4285836107119692, 'learning_rate': 1e-05, 'epoch': 0.56}
0: 239/14725 [7:30:22<5:46:11,  3.20s/it]
 56%|█████▌    | 8240/14725 [7:30:25<5:42:53,  3.17s/it]
                                                        

 56%|█████▌    | 8240/14725 [7:30:25<5:42:53,  3.17s/it]
 56%|█████▌    | 8241/14725 [7:30:28<5:45:43,  3.20s/it]
 56%|█████▌    | 8242/14725 [7:30:31<5:45:42,  3.20s/it]
 56%|█████▌    | 8243/14725 [7:30:35<5:50:39,  3.25s/it]
 56%|█████▌    | 8244/14725 [7:30:38<5:50:33,  3.25s/it]
 56%|█████▌    | 8245/14725 [7:30:41<5:47:27,  3.22s/it]
 56%|█████▌    | 8246/14725 [7:30:44<5:50:21,  3.24s/it]
 56%|█████▌    | 8247/14725 [7:30:48<5:54:10,  3.28s/it]
 56%|█████▌    | 8248/14725 [7:30:51<5:52:58,  3.27s/it]
 56%|█████▌    | 8249/14725 [7:30:54<5:50:54,  3.25s/it]
 56%|█████▌    | 8250/14725 [7:30:58<5:53:59,  3.28s/it]
                                                        

 56%|█████▌    | 8250/14725 [
0: {'loss': 0.6898, 'grad_norm': 0.41934999042734505, 'learning_rate': 1e-05, 'epoch': 0.56}
0: 7:30:58<5:53:59,  3.28s/it]
 56%|█████▌    | 8251/14725 [7:31:01<5:52:44,  3.27s/it]
 56%|█████▌    | 8252/14725 [7:31:04<5:47:06,  3.22s/it]
 56%|█████▌    | 8253/14725 [7:31:07<5:54:29,  3.29s/it]
 56%|█████▌    | 8254/14725 [7:31:10<5:47:24,  3.22s/it]
 56%|█████▌    | 8255/14725 [7:31:14<5:44:55,  3.20s/it]
 56%|█████▌    | 8256/14725 [7:31:17<5:44:27,  3.19s/it]
 56%|█████▌    | 8257/14725 [7:31:20<5:40:29,  3.16s/it]
 56%|█████▌    | 8258/14725 [7:31:23<5:40:40,  3.16s/it]
 56%|█████▌    | 8259/14725 [7:31:26<5:46:10,  3.21s/it]
 56%|█████▌    | 8260/14725 [7:31:30<5:53:01,  3.28s/it]
                                                        

 56%|█████▌    | 8260/14725 [7:31:30<5:53:01,  3.28s/it]
 56%|█████▌    | 8261/14725 [7:31:33<5:58:10,  3.32s/it]
 56%|█████▌    | 8262/14725 [7:31:37<5:59:05,  3.33s/it]
 56%|█████▌    | 8263/14725 [
0: {'loss': 0.7008, 'grad_norm': 0.4087896664314997, 'learning_rate': 1e-05, 'epoch': 0.56}
0: 7:31:40<5:56:26,  3.31s/it]
 56%|█████▌    | 8264/14725 [7:31:43<5:49:38,  3.25s/it]
 56%|█████▌    | 8265/14725 [7:31:46<5:46:08,  3.21s/it]
 56%|█████▌    | 8266/14725 [7:31:50<5:53:32,  3.28s/it]
 56%|█████▌    | 8267/14725 [7:31:53<5:56:09,  3.31s/it]
 56%|█████▌    | 8268/14725 [7:31:56<5:52:35,  3.28s/it]
 56%|█████▌    | 8269/14725 [7:31:59<5:47:06,  3.23s/it]
 56%|█████▌    | 8270/14725 [7:32:02<5:46:54,  3.22s/it]
                                                        

 56%|█████▌    | 8270/14725 [7:32:02<5:46:54,  3.22s/it]
 56%|█████▌    | 8271/14725 [7:32:06<5:48:41,  3.24s/it]
 56%|█████▌    | 8272/14725 [7:32:09<5:45:56,  3.22s/it]
 56%|█████▌    | 8273/14725 [7:32:12<5:50:53,  3.26s/it]
 56%|█████▌    | 8274/14725 [7:32:16<5:54:13,  3.29s/it]
 56%|█████▌    | 8275/14725 [7:32:19<5:51:34,  3.27s/it]
 56%|█████▌    | 8276/14725 [
0: {'loss': 0.6839, 'grad_norm': 0.4439104326635834, 'learning_rate': 1e-05, 'epoch': 0.56}
0: 7:32:22<5:48:21,  3.24s/it]
 56%|█████▌    | 8277/14725 [7:32:25<5:44:40,  3.21s/it]
 56%|█████▌    | 8278/14725 [7:32:28<5:46:40,  3.23s/it]
 56%|█████▌    | 8279/14725 [7:32:32<5:44:16,  3.20s/it]
 56%|█████▌    | 8280/14725 [7:32:35<5:42:24,  3.19s/it]
                                                        

 56%|█████▌    | 8280/14725 [7:32:35<5:42:24,  3.19s/it]
 56%|█████▌    | 8281/14725 [7:32:38<5:46:17,  3.22s/it]
 56%|█████▌    | 8282/14725 [7:32:41<5:45:09,  3.21s/it]
 56%|█████▋    | 8283/14725 [7:32:44<5:41:41,  3.18s/it]
 56%|█████▋    | 8284/14725 [7:32:48<5:43:35,  3.20s/it]
 56%|█████▋    | 8285/14725 [7:32:51<5:49:58,  3.26s/it]
 56%|█████▋    | 8286/14725 [7:32:54<5:47:57,  3.24s/it]
 56%|█████▋    | 8287/14725 [7:32:57<5:48:32,  3.25s/it]
 56%|█████▋    | 8288/14725 [7:33:01<5:53:08,  3.29s/it]
 56%|█████▋    | 8289/14725 [
0: {'loss': 0.7102, 'grad_norm': 0.4195002465220463, 'learning_rate': 1e-05, 'epoch': 0.56}
0: {'loss': 0.687, 'grad_norm': 0.4107999506723922, 'learning_rate': 1e-05, 'epoch': 0.56}
0: 7:33:04<5:53:46,  3.30s/it]
 56%|█████▋    | 8290/14725 [7:33:08<5:58:37,  3.34s/it]
                                                        

 56%|█████▋    | 8290/14725 [7:33:08<5:58:37,  3.34s/it]
 56%|█████▋    | 8291/14725 [7:33:11<5:58:26,  3.34s/it]
 56%|█████▋    | 8292/14725 [7:33:14<5:54:43,  3.31s/it]
 56%|█████▋    | 8293/14725 [7:33:17<5:53:28,  3.30s/it]
 56%|█████▋    | 8294/14725 [7:33:21<5:53:46,  3.30s/it]
 56%|█████▋    | 8295/14725 [7:33:24<5:50:26,  3.27s/it]
 56%|█████▋    | 8296/14725 [7:33:27<5:45:36,  3.23s/it]
 56%|█████▋    | 8297/14725 [7:33:30<5:47:10,  3.24s/it]
 56%|█████▋    | 8298/14725 [7:33:34<5:49:34,  3.26s/it]
 56%|█████▋    | 8299/14725 [7:33:37<5:48:12,  3.25s/it]
 56%|█████▋    | 8300/14725 [7:33:40<5:45:42,  3.23s/it]
                                                        

 56%|█████▋    | 8300/14725 [7:33:40<5:4
0: {'loss': 0.6957, 'grad_norm': 0.46391979797036936, 'learning_rate': 1e-05, 'epoch': 0.56}
0: 5:42,  3.23s/it]
 56%|█████▋    | 8301/14725 [7:33:43<5:40:35,  3.18s/it]
 56%|█████▋    | 8302/14725 [7:33:46<5:43:32,  3.21s/it]
 56%|█████▋    | 8303/14725 [7:33:50<5:48:51,  3.26s/it]
 56%|█████▋    | 8304/14725 [7:33:53<5:46:41,  3.24s/it]
 56%|█████▋    | 8305/14725 [7:33:56<5:44:29,  3.22s/it]
 56%|█████▋    | 8306/14725 [7:33:59<5:48:29,  3.26s/it]
 56%|█████▋    | 8307/14725 [7:34:03<5:51:36,  3.29s/it]
 56%|█████▋    | 8308/14725 [7:34:06<5:43:52,  3.22s/it]
 56%|█████▋    | 8309/14725 [7:34:09<5:42:54,  3.21s/it]
 56%|█████▋    | 8310/14725 [7:34:12<5:45:16,  3.23s/it]
                                                        

 56%|█████▋    | 8310/14725 [7:34:12<5:45:16,  3.23s/it]
 56%|█████▋    | 8311/14725 [7:34:15<5:41:23,  3.19s/it]
 56%|█████▋    | 8312/14725 [7:34:19<5:42:48,  3.21s/it]
 56%|█████▋    | 8313/14725 [7:34:22<5:4
0: {'loss': 0.6985, 'grad_norm': 0.456154923858747, 'learning_rate': 1e-05, 'epoch': 0.57}
0: 4:50,  3.23s/it]
 56%|█████▋    | 8314/14725 [7:34:25<5:44:53,  3.23s/it]
 56%|█████▋    | 8315/14725 [7:34:28<5:44:08,  3.22s/it]
 56%|█████▋    | 8316/14725 [7:34:32<5:47:42,  3.26s/it]
 56%|█████▋    | 8317/14725 [7:34:35<5:47:27,  3.25s/it]
 56%|█████▋    | 8318/14725 [7:34:38<5:47:29,  3.25s/it]
 56%|█████▋    | 8319/14725 [7:34:41<5:47:37,  3.26s/it]
 57%|█████▋    | 8320/14725 [7:34:45<5:50:18,  3.28s/it]
                                                        

 57%|█████▋    | 8320/14725 [7:34:45<5:50:18,  3.28s/it]
 57%|█████▋    | 8321/14725 [7:34:48<5:58:57,  3.36s/it]
 57%|█████▋    | 8322/14725 [7:34:52<5:51:38,  3.30s/it]
 57%|█████▋    | 8323/14725 [7:34:55<5:57:24,  3.35s/it]
 57%|█████▋    | 8324/14725 [7:34:58<5:57:19,  3.35s/it]
 57%|█████▋    | 8325/14725 [7:35:01<5:48:46,  3.27s/it]
 57%|█████▋    | 8326/14725 [7:35:05<5:4
0: {'loss': 0.684, 'grad_norm': 0.4066867943177724, 'learning_rate': 1e-05, 'epoch': 0.57}
0: 6:43,  3.25s/it]
 57%|█████▋    | 8327/14725 [7:35:08<5:53:08,  3.31s/it]
 57%|█████▋    | 8328/14725 [7:35:11<5:47:53,  3.26s/it]
 57%|█████▋    | 8329/14725 [7:35:14<5:47:51,  3.26s/it]
 57%|█████▋    | 8330/14725 [7:35:18<5:42:02,  3.21s/it]
                                                        

 57%|█████▋    | 8330/14725 [7:35:18<5:42:02,  3.21s/it]
 57%|█████▋    | 8331/14725 [7:35:21<5:48:06,  3.27s/it]
 57%|█████▋    | 8332/14725 [7:35:24<5:46:47,  3.25s/it]
 57%|█████▋    | 8333/14725 [7:35:27<5:46:41,  3.25s/it]
 57%|█████▋    | 8334/14725 [7:35:31<5:45:08,  3.24s/it]
 57%|█████▋    | 8335/14725 [7:35:34<5:41:49,  3.21s/it]
 57%|█████▋    | 8336/14725 [7:35:37<5:49:14,  3.28s/it]
 57%|█████▋    | 8337/14725 [7:35:41<5:49:31,  3.28s/it]
 57%|█████▋    | 8338/14725 [7:35:44<5:50:09,  3.29s/it]
 57%|█████▋    | 8339/14725 [7:35:47<5:5
0: {'loss': 0.7021, 'grad_norm': 0.4593032726521289, 'learning_rate': 1e-05, 'epoch': 0.57}
0: {'loss': 0.6733, 'grad_norm': 0.4171821584810486, 'learning_rate': 1e-05, 'epoch': 0.57}
0: 0:13,  3.29s/it]
 57%|█████▋    | 8340/14725 [7:35:51<5:57:12,  3.36s/it]
                                                        

 57%|█████▋    | 8340/14725 [7:35:51<5:57:12,  3.36s/it]
 57%|█████▋    | 8341/14725 [7:35:54<5:58:30,  3.37s/it]
 57%|█████▋    | 8342/14725 [7:35:57<5:57:14,  3.36s/it]
 57%|█████▋    | 8343/14725 [7:36:01<5:54:05,  3.33s/it]
 57%|█████▋    | 8344/14725 [7:36:04<5:46:09,  3.25s/it]
 57%|█████▋    | 8345/14725 [7:36:07<5:47:30,  3.27s/it]
 57%|█████▋    | 8346/14725 [7:36:10<5:44:31,  3.24s/it]
 57%|█████▋    | 8347/14725 [7:36:14<5:46:55,  3.26s/it]
 57%|█████▋    | 8348/14725 [7:36:17<5:45:39,  3.25s/it]
 57%|█████▋    | 8349/14725 [7:36:20<5:47:27,  3.27s/it]
 57%|█████▋    | 8350/14725 [7:36:23<5:44:08,  3.24s/it]
                                                        

 57%|█████▋    | 8350/14725 [7:36:23<5:44:08,  3.24
0: {'loss': 0.7006, 'grad_norm': 0.4176258720652083, 'learning_rate': 1e-05, 'epoch': 0.57}
0: s/it]
 57%|█████▋    | 8351/14725 [7:36:26<5:40:34,  3.21s/it]
 57%|█████▋    | 8352/14725 [7:36:29<5:38:06,  3.18s/it]
 57%|█████▋    | 8353/14725 [7:36:33<5:38:22,  3.19s/it]
 57%|█████▋    | 8354/14725 [7:36:36<5:41:17,  3.21s/it]
 57%|█████▋    | 8355/14725 [7:36:39<5:43:22,  3.23s/it]
 57%|█████▋    | 8356/14725 [7:36:43<5:50:53,  3.31s/it]
 57%|█████▋    | 8357/14725 [7:36:46<5:50:49,  3.31s/it]
 57%|█████▋    | 8358/14725 [7:36:49<5:50:16,  3.30s/it]
 57%|█████▋    | 8359/14725 [7:36:53<5:51:06,  3.31s/it]
 57%|█████▋    | 8360/14725 [7:36:56<5:49:30,  3.29s/it]
                                                        

 57%|█████▋    | 8360/14725 [7:36:56<5:49:30,  3.29s/it]
 57%|█████▋    | 8361/14725 [7:36:59<5:45:37,  3.26s/it]
 57%|█████▋    | 8362/14725 [7:37:02<5:44:16,  3.25s/it]
 57%|█████▋    | 8363/14725 [7:37:06<5:51:53,  3.32
0: {'loss': 0.6889, 'grad_norm': 0.4242516083138223, 'learning_rate': 1e-05, 'epoch': 0.57}
0: s/it]
 57%|█████▋    | 8364/14725 [7:37:09<5:51:00,  3.31s/it]
 57%|█████▋    | 8365/14725 [7:37:12<5:49:30,  3.30s/it]
 57%|█████▋    | 8366/14725 [7:37:15<5:45:35,  3.26s/it]
 57%|█████▋    | 8367/14725 [7:37:19<5:46:50,  3.27s/it]
 57%|█████▋    | 8368/14725 [7:37:22<5:48:54,  3.29s/it]
 57%|█████▋    | 8369/14725 [7:37:25<5:49:21,  3.30s/it]
 57%|█████▋    | 8370/14725 [7:37:29<5:47:53,  3.28s/it]
                                                        

 57%|█████▋    | 8370/14725 [7:37:29<5:47:53,  3.28s/it]
 57%|█████▋    | 8371/14725 [7:37:32<5:43:16,  3.24s/it]
 57%|█████▋    | 8372/14725 [7:37:35<5:49:36,  3.30s/it]
 57%|█████▋    | 8373/14725 [7:37:38<5:46:05,  3.27s/it]
 57%|█████▋    | 8374/14725 [7:37:42<5:44:55,  3.26s/it]
 57%|█████▋    | 8375/14725 [7:37:45<5:44:26,  3.25s/it]
 57%|█████▋    | 8376/14725 [7:37:48<5:47:59,  3.29
0: {'loss': 0.7004, 'grad_norm': 0.48259389755186416, 'learning_rate': 1e-05, 'epoch': 0.57}
0: s/it]
 57%|█████▋    | 8377/14725 [7:37:51<5:42:30,  3.24s/it]
 57%|█████▋    | 8378/14725 [7:37:55<5:38:45,  3.20s/it]
 57%|█████▋    | 8379/14725 [7:37:58<5:39:58,  3.21s/it]
 57%|█████▋    | 8380/14725 [7:38:01<5:52:09,  3.33s/it]
                                                        

 57%|█████▋    | 8380/14725 [7:38:01<5:52:09,  3.33s/it]
 57%|█████▋    | 8381/14725 [7:38:05<5:48:38,  3.30s/it]
 57%|█████▋    | 8382/14725 [7:38:08<5:49:01,  3.30s/it]
 57%|█████▋    | 8383/14725 [7:38:11<5:52:56,  3.34s/it]
 57%|█████▋    | 8384/14725 [7:38:15<5:51:39,  3.33s/it]
 57%|█████▋    | 8385/14725 [7:38:18<5:51:16,  3.32s/it]
 57%|█████▋    | 8386/14725 [7:38:21<5:46:11,  3.28s/it]
 57%|█████▋    | 8387/14725 [7:38:24<5:41:52,  3.24s/it]
 57%|█████▋    | 8388/14725 [7:38:28<5:42:24,  3.24s/it]
 57%|█████▋    | 8389/14725 [7:38:31<5:36:41,  3.19
0: {'loss': 0.683, 'grad_norm': 0.4343348698859466, 'learning_rate': 1e-05, 'epoch': 0.57}
0: {'loss': 0.6745, 'grad_norm': 0.43030769088802856, 'learning_rate': 1e-05, 'epoch': 0.57}
0: s/it]
 57%|█████▋    | 8390/14725 [7:38:34<5:36:13,  3.18s/it]
                                                        

 57%|█████▋    | 8390/14725 [7:38:34<5:36:13,  3.18s/it]
 57%|█████▋    | 8391/14725 [7:38:37<5:41:46,  3.24s/it]
 57%|█████▋    | 8392/14725 [7:38:40<5:39:03,  3.21s/it]
 57%|█████▋    | 8393/14725 [7:38:44<5:44:18,  3.26s/it]
 57%|█████▋    | 8394/14725 [7:38:47<5:42:34,  3.25s/it]
 57%|█████▋    | 8395/14725 [7:38:50<5:43:03,  3.25s/it]
 57%|█████▋    | 8396/14725 [7:38:53<5:42:57,  3.25s/it]
 57%|█████▋    | 8397/14725 [7:38:57<5:41:37,  3.24s/it]
 57%|█████▋    | 8398/14725 [7:39:00<5:42:46,  3.25s/it]
 57%|█████▋    | 8399/14725 [7:39:03<5:40:53,  3.23s/it]
 57%|█████▋    | 8400/14725 [7:39:06<5:44:09,  3.26s/it]
                                                        

 57%|█████▋    | 8400/14725 [7:39:06<5:44:09,  3.26s/it]
 57%|
0: {'loss': 0.7016, 'grad_norm': 0.4337697477684697, 'learning_rate': 1e-05, 'epoch': 0.57}
0: █████▋    | 8401/14725 [7:39:10<5:43:10,  3.26s/it]
 57%|█████▋    | 8402/14725 [7:39:13<5:46:00,  3.28s/it]
 57%|█████▋    | 8403/14725 [7:39:16<5:44:58,  3.27s/it]
 57%|█████▋    | 8404/14725 [7:39:20<5:45:13,  3.28s/it]
 57%|█████▋    | 8405/14725 [7:39:23<5:49:36,  3.32s/it]
 57%|█████▋    | 8406/14725 [7:39:26<5:41:38,  3.24s/it]
 57%|█████▋    | 8407/14725 [7:39:29<5:41:22,  3.24s/it]
 57%|█████▋    | 8408/14725 [7:39:32<5:36:45,  3.20s/it]
 57%|█████▋    | 8409/14725 [7:39:35<5:33:41,  3.17s/it]
 57%|█████▋    | 8410/14725 [7:39:39<5:33:39,  3.17s/it]
                                                        

 57%|█████▋    | 8410/14725 [7:39:39<5:33:39,  3.17s/it]
 57%|█████▋    | 8411/14725 [7:39:42<5:40:16,  3.23s/it]
 57%|█████▋    | 8412/14725 [7:39:46<5:49:48,  3.32s/it]
 57%|█████▋    | 8413/14725 [7:39:49<5:50:12,  3.33s/it]
 57%|
0: {'loss': 0.6977, 'grad_norm': 0.41217105505739365, 'learning_rate': 1e-05, 'epoch': 0.57}
0: █████▋    | 8414/14725 [7:39:52<5:44:34,  3.28s/it]
 57%|█████▋    | 8415/14725 [7:39:55<5:43:54,  3.27s/it]
 57%|█████▋    | 8416/14725 [7:39:59<5:45:32,  3.29s/it]
 57%|█████▋    | 8417/14725 [7:40:02<5:40:15,  3.24s/it]
 57%|█████▋    | 8418/14725 [7:40:05<5:48:53,  3.32s/it]
 57%|█████▋    | 8419/14725 [7:40:09<5:49:12,  3.32s/it]
 57%|█████▋    | 8420/14725 [7:40:12<5:47:24,  3.31s/it]
                                                        

 57%|█████▋    | 8420/14725 [7:40:12<5:47:24,  3.31s/it]
 57%|█████▋    | 8421/14725 [7:40:15<5:52:15,  3.35s/it]
 57%|█████▋    | 8422/14725 [7:40:18<5:46:37,  3.30s/it]
 57%|█████▋    | 8423/14725 [7:40:22<5:45:33,  3.29s/it]
 57%|█████▋    | 8424/14725 [7:40:25<5:43:31,  3.27s/it]
 57%|█████▋    | 8425/14725 [7:40:28<5:41:55,  3.26s/it]
 57%|█████▋    | 8426/14725 [7:40:31<5:42:16,  3.26s/it]
 57%|
0: {'loss': 0.6886, 'grad_norm': 0.46534322046508536, 'learning_rate': 1e-05, 'epoch': 0.57}
0: █████▋    | 8427/14725 [7:40:35<5:42:03,  3.26s/it]
 57%|█████▋    | 8428/14725 [7:40:38<5:35:02,  3.19s/it]
 57%|█████▋    | 8429/14725 [7:40:41<5:38:00,  3.22s/it]
 57%|█████▋    | 8430/14725 [7:40:44<5:37:18,  3.21s/it]
                                                        

 57%|█████▋    | 8430/14725 [7:40:44<5:37:18,  3.21s/it]
 57%|█████▋    | 8431/14725 [7:40:48<5:40:30,  3.25s/it]
 57%|█████▋    | 8432/14725 [7:40:51<5:39:44,  3.24s/it]
 57%|█████▋    | 8433/14725 [7:40:54<5:42:59,  3.27s/it]
 57%|█████▋    | 8434/14725 [7:40:57<5:38:05,  3.22s/it]
 57%|█████▋    | 8435/14725 [7:41:01<5:44:02,  3.28s/it]
 57%|█████▋    | 8436/14725 [7:41:04<5:43:33,  3.28s/it]
 57%|█████▋    | 8437/14725 [7:41:07<5:44:49,  3.29s/it]
 57%|█████▋    | 8438/14725 [7:41:11<5:50:56,  3.35s/it]
 57%|█████▋    | 8439/14725 [7:41:14<5:46:24,  3.31s/it]
 57%|
0: {'loss': 0.6899, 'grad_norm': 0.4468222757824624, 'learning_rate': 1e-05, 'epoch': 0.57}
0: {'loss': 0.7033, 'grad_norm': 0.4374805410327514, 'learning_rate': 1e-05, 'epoch': 0.57}
0: █████▋    | 8440/14725 [7:41:17<5:40:47,  3.25s/it]
                                                        

 57%|█████▋    | 8440/14725 [7:41:17<5:40:47,  3.25s/it]
 57%|█████▋    | 8441/14725 [7:41:20<5:40:28,  3.25s/it]
 57%|█████▋    | 8442/14725 [7:41:24<5:47:48,  3.32s/it]
 57%|█████▋    | 8443/14725 [7:41:27<5:41:29,  3.26s/it]
 57%|█████▋    | 8444/14725 [7:41:30<5:39:45,  3.25s/it]
 57%|█████▋    | 8445/14725 [7:41:34<5:50:21,  3.35s/it]
 57%|█████▋    | 8446/14725 [7:41:37<5:44:13,  3.29s/it]
 57%|█████▋    | 8447/14725 [7:41:40<5:47:29,  3.32s/it]
 57%|█████▋    | 8448/14725 [7:41:43<5:43:56,  3.29s/it]
 57%|█████▋    | 8449/14725 [7:41:47<5:37:08,  3.22s/it]
 57%|█████▋    | 8450/14725 [7:41:50<5:44:33,  3.29s/it]
                                                        

 57%|█████▋    | 8450/14725 [7:41:50<5:44:33,  3.29s/it]
 57%|███�
0: {'loss': 0.7077, 'grad_norm': 0.42127177538542315, 'learning_rate': 1e-05, 'epoch': 0.57}
0: �█▋    | 8451/14725 [7:41:53<5:46:21,  3.31s/it]
 57%|█████▋    | 8452/14725 [7:41:57<5:48:00,  3.33s/it]
 57%|█████▋    | 8453/14725 [7:42:00<5:46:36,  3.32s/it]
 57%|█████▋    | 8454/14725 [7:42:03<5:43:45,  3.29s/it]
 57%|█████▋    | 8455/14725 [7:42:06<5:41:10,  3.26s/it]
 57%|█████▋    | 8456/14725 [7:42:10<5:39:51,  3.25s/it]
 57%|█████▋    | 8457/14725 [7:42:13<5:39:10,  3.25s/it]
 57%|█████▋    | 8458/14725 [7:42:16<5:41:10,  3.27s/it]
 57%|█████▋    | 8459/14725 [7:42:20<5:49:01,  3.34s/it]
 57%|█████▋    | 8460/14725 [7:42:23<5:47:01,  3.32s/it]
                                                        

 57%|█████▋    | 8460/14725 [7:42:23<5:47:01,  3.32s/it]
 57%|█████▋    | 8461/14725 [7:42:26<5:46:16,  3.32s/it]
 57%|█████▋    | 8462/14725 [7:42:30<5:54:51,  3.40s/it]
 57%|█████▋    | 8463/14725 [7:42:33<5:46:46,  3.32s/it]
 57%|███�
0: {'loss': 0.6878, 'grad_norm': 0.40643042796706835, 'learning_rate': 1e-05, 'epoch': 0.58}
0: �█▋    | 8464/14725 [7:42:36<5:48:59,  3.34s/it]
 57%|█████▋    | 8465/14725 [7:42:39<5:38:13,  3.24s/it]
 57%|█████▋    | 8466/14725 [7:42:43<5:35:00,  3.21s/it]
 58%|█████▊    | 8467/14725 [7:42:46<5:34:58,  3.21s/it]
 58%|█████▊    | 8468/14725 [7:42:49<5:32:19,  3.19s/it]
 58%|█████▊    | 8469/14725 [7:42:52<5:42:44,  3.29s/it]
 58%|█████▊    | 8470/14725 [7:42:56<5:35:14,  3.22s/it]
                                                        

 58%|█████▊    | 8470/14725 [7:42:56<5:35:14,  3.22s/it]
 58%|█████▊    | 8471/14725 [7:42:59<5:35:34,  3.22s/it]
 58%|█████▊    | 8472/14725 [7:43:02<5:42:09,  3.28s/it]
 58%|█████▊    | 8473/14725 [7:43:05<5:41:09,  3.27s/it]
 58%|█████▊    | 8474/14725 [7:43:09<5:36:41,  3.23s/it]
 58%|█████▊    | 8475/14725 [7:43:12<5:35:18,  3.22s/it]
 58%|█████▊    | 8476/14725 [7:43:15<5:45:30,  3.32s/it]
 58%|███�
0: {'loss': 0.6953, 'grad_norm': 0.41482814618490793, 'learning_rate': 1e-05, 'epoch': 0.58}
0: �█▊    | 8477/14725 [7:43:18<5:39:41,  3.26s/it]
 58%|█████▊    | 8478/14725 [7:43:22<5:37:14,  3.24s/it]
 58%|█████▊    | 8479/14725 [7:43:25<5:40:11,  3.27s/it]
 58%|█████▊    | 8480/14725 [7:43:28<5:43:56,  3.30s/it]
                                                        

 58%|█████▊    | 8480/14725 [7:43:28<5:43:56,  3.30s/it]
 58%|█████▊    | 8481/14725 [7:43:32<5:43:03,  3.30s/it]
 58%|█████▊    | 8482/14725 [7:43:35<5:50:13,  3.37s/it]
 58%|█████▊    | 8483/14725 [7:43:38<5:42:23,  3.29s/it]
 58%|█████▊    | 8484/14725 [7:43:42<5:47:30,  3.34s/it]
 58%|█████▊    | 8485/14725 [7:43:45<5:49:46,  3.36s/it]
 58%|█████▊    | 8486/14725 [7:43:49<5:49:54,  3.37s/it]
 58%|█████▊    | 8487/14725 [7:43:52<5:47:46,  3.35s/it]
 58%|█████▊    | 8488/14725 [7:43:55<5:44:51,  3.32s/it]
 58%|█████▊    | 8489/14725 [7:43:58<5:47:12,  3.34s/it]
 58%|███�
0: {'loss': 0.6925, 'grad_norm': 0.4288050197961458, 'learning_rate': 1e-05, 'epoch': 0.58}
0: {'loss': 0.7085, 'grad_norm': 0.455686713907164, 'learning_rate': 1e-05, 'epoch': 0.58}
0: �█▊    | 8490/14725 [7:44:02<5:42:12,  3.29s/it]
                                                        

 58%|█████▊    | 8490/14725 [7:44:02<5:42:12,  3.29s/it]
 58%|█████▊    | 8491/14725 [7:44:05<5:38:09,  3.25s/it]
 58%|█████▊    | 8492/14725 [7:44:08<5:35:37,  3.23s/it]
 58%|█████▊    | 8493/14725 [7:44:11<5:35:32,  3.23s/it]
 58%|█████▊    | 8494/14725 [7:44:15<5:44:59,  3.32s/it]
 58%|█████▊    | 8495/14725 [7:44:18<5:40:08,  3.28s/it]
 58%|█████▊    | 8496/14725 [7:44:21<5:36:08,  3.24s/it]
 58%|█████▊    | 8497/14725 [7:44:24<5:33:51,  3.22s/it]
 58%|█████▊    | 8498/14725 [7:44:27<5:33:59,  3.22s/it]
 58%|█████▊    | 8499/14725 [7:44:31<5:35:04,  3.23s/it]
 58%|█████▊    | 8500/14725 [7:44:34<5:37:09,  3.25s/it]
                                                        

 58%|█████▊    | 8500/14725 [7:44:34<5:37:09,  3.25s/it]
 58%|█████▊    
0: {'loss': 0.6976, 'grad_norm': 0.4071493023926805, 'learning_rate': 1e-05, 'epoch': 0.58}
0: | 8501/14725 [7:44:37<5:34:42,  3.23s/it]
 58%|█████▊    | 8502/14725 [7:44:40<5:31:09,  3.19s/it]
 58%|█████▊    | 8503/14725 [7:44:44<5:36:57,  3.25s/it]
 58%|█████▊    | 8504/14725 [7:44:47<5:42:21,  3.30s/it]
 58%|█████▊    | 8505/14725 [7:44:50<5:40:56,  3.29s/it]
 58%|█████▊    | 8506/14725 [7:44:54<5:38:08,  3.26s/it]
 58%|█████▊    | 8507/14725 [7:44:57<5:43:03,  3.31s/it]
 58%|█████▊    | 8508/14725 [7:45:00<5:36:53,  3.25s/it]
 58%|█████▊    | 8509/14725 [7:45:04<5:45:26,  3.33s/it]
 58%|█████▊    | 8510/14725 [7:45:07<5:41:19,  3.30s/it]
                                                        

 58%|█████▊    | 8510/14725 [7:45:07<5:41:19,  3.30s/it]
 58%|█████▊    | 8511/14725 [7:45:10<5:39:33,  3.28s/it]
 58%|█████▊    | 8512/14725 [7:45:13<5:37:39,  3.26s/it]
 58%|█████▊    | 8513/14725 [7:45:16<5:34:07,  3.23s/it]
 58%|█████▊    
0: {'loss': 0.6846, 'grad_norm': 0.39172679318820636, 'learning_rate': 1e-05, 'epoch': 0.58}
0: | 8514/14725 [7:45:20<5:41:55,  3.30s/it]
 58%|█████▊    | 8515/14725 [7:45:23<5:40:51,  3.29s/it]
 58%|█████▊    | 8516/14725 [7:45:26<5:38:58,  3.28s/it]
 58%|█████▊    | 8517/14725 [7:45:30<5:38:37,  3.27s/it]
 58%|█████▊    | 8518/14725 [7:45:33<5:34:17,  3.23s/it]
 58%|█████▊    | 8519/14725 [7:45:36<5:35:17,  3.24s/it]
 58%|█████▊    | 8520/14725 [7:45:39<5:30:48,  3.20s/it]
                                                        

 58%|█████▊    | 8520/14725 [7:45:39<5:30:48,  3.20s/it]
 58%|█████▊    | 8521/14725 [7:45:43<5:34:54,  3.24s/it]
 58%|█████▊    | 8522/14725 [7:45:46<5:31:15,  3.20s/it]
 58%|█████▊    | 8523/14725 [7:45:49<5:33:29,  3.23s/it]
 58%|█████▊    | 8524/14725 [7:45:52<5:35:34,  3.25s/it]
 58%|█████▊    | 8525/14725 [7:45:55<5:32:20,  3.22s/it]
 58%|█████▊    | 8526/14725 [7:45:59<5:33:09,  3.22s/it]
 58%|█████▊    
0: {'loss': 0.686, 'grad_norm': 0.42701700611709825, 'learning_rate': 1e-05, 'epoch': 0.58}
0: | 8527/14725 [7:46:02<5:29:46,  3.19s/it]
 58%|█████▊    | 8528/14725 [7:46:05<5:36:16,  3.26s/it]
 58%|█████▊    | 8529/14725 [7:46:09<5:40:16,  3.30s/it]
 58%|█████▊    | 8530/14725 [7:46:12<5:39:25,  3.29s/it]
                                                        

 58%|█████▊    | 8530/14725 [7:46:12<5:39:25,  3.29s/it]
 58%|█████▊    | 8531/14725 [7:46:15<5:37:27,  3.27s/it]
 58%|█████▊    | 8532/14725 [7:46:18<5:33:28,  3.23s/it]
 58%|█████▊    | 8533/14725 [7:46:21<5:29:53,  3.20s/it]
 58%|█████▊    | 8534/14725 [7:46:25<5:36:52,  3.26s/it]
 58%|█████▊    | 8535/14725 [7:46:28<5:34:42,  3.24s/it]
 58%|█████▊    | 8536/14725 [7:46:31<5:36:07,  3.26s/it]
 58%|█████▊    | 8537/14725 [7:46:34<5:35:31,  3.25s/it]
 58%|█████▊    | 8538/14725 [7:46:38<5:35:04,  3.25s/it]
 58%|█████▊    | 8539/14725 [7:46:41<5:33:35,  3.24s/it]
 58%|█████▊    
0: {'loss': 0.6845, 'grad_norm': 0.443309687771431, 'learning_rate': 1e-05, 'epoch': 0.58}
0: {'loss': 0.6779, 'grad_norm': 0.3912425194975704, 'learning_rate': 1e-05, 'epoch': 0.58}
0: | 8540/14725 [7:46:44<5:35:15,  3.25s/it]
                                                        

 58%|█████▊    | 8540/14725 [7:46:44<5:35:15,  3.25s/it]
 58%|█████▊    | 8541/14725 [7:46:47<5:34:02,  3.24s/it]
 58%|█████▊    | 8542/14725 [7:46:50<5:30:13,  3.20s/it]
 58%|█████▊    | 8543/14725 [7:46:54<5:32:44,  3.23s/it]
 58%|█████▊    | 8544/14725 [7:46:57<5:27:33,  3.18s/it]
 58%|█████▊    | 8545/14725 [7:47:00<5:21:59,  3.13s/it]
 58%|█████▊    | 8546/14725 [7:47:03<5:21:34,  3.12s/it]
 58%|█████▊    | 8547/14725 [7:47:06<5:19:07,  3.10s/it]
 58%|█████▊    | 8548/14725 [7:47:09<5:22:56,  3.14s/it]
 58%|█████▊    | 8549/14725 [7:47:13<5:37:34,  3.28s/it]
 58%|█████▊    | 8550/14725 [7:47:16<5:32:47,  3.23s/it]
                                                        

 58%|█████▊    | 8550/14725 [7:47:16<5:32:47,  3.23s/it]
 58%|█████▊    | 8551/1472
0: {'loss': 0.7013, 'grad_norm': 0.409167512449048, 'learning_rate': 1e-05, 'epoch': 0.58}
0: 5 [7:47:19<5:36:43,  3.27s/it]
 58%|█████▊    | 8552/14725 [7:47:23<5:44:56,  3.35s/it]
 58%|█████▊    | 8553/14725 [7:47:26<5:41:41,  3.32s/it]
 58%|█████▊    | 8554/14725 [7:47:29<5:42:21,  3.33s/it]
 58%|█████▊    | 8555/14725 [7:47:33<5:54:05,  3.44s/it]
 58%|█████▊    | 8556/14725 [7:47:36<5:49:17,  3.40s/it]
 58%|█████▊    | 8557/14725 [7:47:40<5:54:41,  3.45s/it]
 58%|█████▊    | 8558/14725 [7:47:43<5:50:50,  3.41s/it]
 58%|█████▊    | 8559/14725 [7:47:47<5:44:35,  3.35s/it]
 58%|█████▊    | 8560/14725 [7:47:50<5:45:28,  3.36s/it]
                                                        

 58%|█████▊    | 8560/14725 [7:47:50<5:45:28,  3.36s/it]
 58%|█████▊    | 8561/14725 [7:47:53<5:42:57,  3.34s/it]
 58%|█████▊    | 8562/14725 [7:47:57<5:45:20,  3.36s/it]
 58%|█████▊    | 8563/14725 [7:48:00<5:42:40,  3.34s/it]
 58%|█████▊    | 8564/1472
0: {'loss': 0.6853, 'grad_norm': 0.38601110291610596, 'learning_rate': 1e-05, 'epoch': 0.58}
0: 5 [7:48:03<5:42:54,  3.34s/it]
 58%|█████▊    | 8565/14725 [7:48:06<5:36:08,  3.27s/it]
 58%|█████▊    | 8566/14725 [7:48:10<5:31:28,  3.23s/it]
 58%|█████▊    | 8567/14725 [7:48:13<5:29:25,  3.21s/it]
 58%|█████▊    | 8568/14725 [7:48:16<5:26:17,  3.18s/it]
 58%|█████▊    | 8569/14725 [7:48:19<5:29:34,  3.21s/it]
 58%|█████▊    | 8570/14725 [7:48:22<5:32:46,  3.24s/it]
                                                        

 58%|█████▊    | 8570/14725 [7:48:22<5:32:46,  3.24s/it]
 58%|█████▊    | 8571/14725 [7:48:26<5:28:36,  3.20s/it]
 58%|█████▊    | 8572/14725 [7:48:29<5:37:52,  3.29s/it]
 58%|█████▊    | 8573/14725 [7:48:32<5:33:42,  3.25s/it]
 58%|█████▊    | 8574/14725 [7:48:35<5:28:39,  3.21s/it]
 58%|█████▊    | 8575/14725 [7:48:38<5:25:22,  3.17s/it]
 58%|█████▊    | 8576/14725 [7:48:42<5:28:00,  3.20s/it]
 58%|█████▊    | 8577/1472
0: {'loss': 0.7118, 'grad_norm': 0.43549778099827346, 'learning_rate': 1e-05, 'epoch': 0.58}
0: 5 [7:48:45<5:32:36,  3.25s/it]
 58%|█████▊    | 8578/14725 [7:48:49<5:41:28,  3.33s/it]
 58%|█████▊    | 8579/14725 [7:48:52<5:38:16,  3.30s/it]
 58%|█████▊    | 8580/14725 [7:48:55<5:39:46,  3.32s/it]
                                                        

 58%|█████▊    | 8580/14725 [7:48:55<5:39:46,  3.32s/it]
 58%|█████▊    | 8581/14725 [7:48:58<5:36:41,  3.29s/it]
 58%|█████▊    | 8582/14725 [7:49:02<5:39:07,  3.31s/it]
 58%|█████▊    | 8583/14725 [7:49:05<5:41:03,  3.33s/it]
 58%|█████▊    | 8584/14725 [7:49:08<5:36:42,  3.29s/it]
 58%|█████▊    | 8585/14725 [7:49:12<5:41:35,  3.34s/it]
 58%|█████▊    | 8586/14725 [7:49:15<5:45:04,  3.37s/it]
 58%|█████▊    | 8587/14725 [7:49:18<5:37:54,  3.30s/it]
 58%|█████▊    | 8588/14725 [7:49:22<5:38:59,  3.31s/it]
 58%|█████▊    | 8589/14725 [7:49:25<5:35:27,  3.28s/it]
 58%|█████▊    | 8590/1472
0: {'loss': 0.6873, 'grad_norm': 0.3841041293834927, 'learning_rate': 1e-05, 'epoch': 0.58}
0: {'loss': 0.6763, 'grad_norm': 0.42006132299144017, 'learning_rate': 1e-05, 'epoch': 0.58}
0: 5 [7:49:28<5:33:47,  3.26s/it]
                                                        

 58%|█████▊    | 8590/14725 [7:49:28<5:33:47,  3.26s/it]
 58%|█████▊    | 8591/14725 [7:49:31<5:35:58,  3.29s/it]
 58%|█████▊    | 8592/14725 [7:49:34<5:27:49,  3.21s/it]
 58%|█████▊    | 8593/14725 [7:49:38<5:32:30,  3.25s/it]
 58%|█████▊    | 8594/14725 [7:49:41<5:26:28,  3.20s/it]
 58%|█████▊    | 8595/14725 [7:49:44<5:30:19,  3.23s/it]
 58%|█████▊    | 8596/14725 [7:49:47<5:28:47,  3.22s/it]
 58%|█████▊    | 8597/14725 [7:49:51<5:28:21,  3.21s/it]
 58%|█████▊    | 8598/14725 [7:49:54<5:24:34,  3.18s/it]
 58%|█████▊    | 8599/14725 [7:49:57<5:20:16,  3.14s/it]
 58%|█████▊    | 8600/14725 [7:50:00<5:26:11,  3.20s/it]
                                                        

 58%|█████▊    | 8600/14725 [7:50:00<5:26:11,  3.20s/it]
 58%|█████▊    | 8601/14725 [7:50:03<
0: {'loss': 0.6882, 'grad_norm': 0.4176664525018306, 'learning_rate': 1e-05, 'epoch': 0.58}
0: 5:32:54,  3.26s/it]
 58%|█████▊    | 8602/14725 [7:50:07<5:30:45,  3.24s/it]
 58%|█████▊    | 8603/14725 [7:50:10<5:29:16,  3.23s/it]
 58%|█████▊    | 8604/14725 [7:50:13<5:31:45,  3.25s/it]
 58%|█████▊    | 8605/14725 [7:50:16<5:27:43,  3.21s/it]
 58%|█████▊    | 8606/14725 [7:50:19<5:24:39,  3.18s/it]
 58%|█████▊    | 8607/14725 [7:50:23<5:26:49,  3.21s/it]
 58%|█████▊    | 8608/14725 [7:50:26<5:22:41,  3.17s/it]
 58%|█████▊    | 8609/14725 [7:50:29<5:29:43,  3.23s/it]
 58%|█████▊    | 8610/14725 [7:50:32<5:27:45,  3.22s/it]
                                                        

 58%|█████▊    | 8610/14725 [7:50:32<5:27:45,  3.22s/it]
 58%|█████▊    | 8611/14725 [7:50:35<5:26:03,  3.20s/it]
 58%|█████▊    | 8612/14725 [7:50:39<5:28:24,  3.22s/it]
 58%|█████▊    | 8613/14725 [7:50:42<5:33:27,  3.27s/it]
 58%|█████▊    | 8614/14725 [7:50:45<
0: {'loss': 0.7084, 'grad_norm': 0.4161202368431312, 'learning_rate': 1e-05, 'epoch': 0.59}
0: 5:33:27,  3.27s/it]
 59%|█████▊    | 8615/14725 [7:50:49<5:33:08,  3.27s/it]
 59%|█████▊    | 8616/14725 [7:50:52<5:31:44,  3.26s/it]
 59%|█████▊    | 8617/14725 [7:50:55<5:34:18,  3.28s/it]
 59%|█████▊    | 8618/14725 [7:50:59<5:38:17,  3.32s/it]
 59%|█████▊    | 8619/14725 [7:51:02<5:35:33,  3.30s/it]
 59%|█████▊    | 8620/14725 [7:51:05<5:32:38,  3.27s/it]
                                                        

 59%|█████▊    | 8620/14725 [7:51:05<5:32:38,  3.27s/it]
 59%|█████▊    | 8621/14725 [7:51:08<5:29:14,  3.24s/it]
 59%|█████▊    | 8622/14725 [7:51:11<5:28:05,  3.23s/it]
 59%|█████▊    | 8623/14725 [7:51:15<5:25:35,  3.20s/it]
 59%|█████▊    | 8624/14725 [7:51:18<5:23:28,  3.18s/it]
 59%|█████▊    | 8625/14725 [7:51:21<5:26:34,  3.21s/it]
 59%|█████▊    | 8626/14725 [7:51:24<5:30:20,  3.25s/it]
 59%|█████▊    | 8627/14725 [7:51:28<
0: {'loss': 0.6974, 'grad_norm': 0.40835522946105174, 'learning_rate': 1e-05, 'epoch': 0.59}
0: 5:29:20,  3.24s/it]
 59%|█████▊    | 8628/14725 [7:51:31<5:31:40,  3.26s/it]
 59%|█████▊    | 8629/14725 [7:51:34<5:28:30,  3.23s/it]
 59%|█████▊    | 8630/14725 [7:51:37<5:32:32,  3.27s/it]
                                                        

 59%|█████▊    | 8630/14725 [7:51:37<5:32:32,  3.27s/it]
 59%|█████▊    | 8631/14725 [7:51:41<5:35:29,  3.30s/it]
 59%|█████▊    | 8632/14725 [7:51:44<5:36:17,  3.31s/it]
 59%|█████▊    | 8633/14725 [7:51:47<5:33:38,  3.29s/it]
 59%|█████▊    | 8634/14725 [7:51:50<5:25:10,  3.20s/it]
 59%|█████▊    | 8635/14725 [7:51:54<5:26:09,  3.21s/it]
 59%|█████▊    | 8636/14725 [7:51:57<5:30:34,  3.26s/it]
 59%|█████▊    | 8637/14725 [7:52:00<5:36:34,  3.32s/it]
 59%|█████▊    | 8638/14725 [7:52:04<5:33:02,  3.28s/it]
 59%|█████▊    | 8639/14725 [7:52:07<5:31:05,  3.26s/it]
 59%|█████▊    | 8640/14725 [7:52:10<
0: {'loss': 0.6906, 'grad_norm': 0.407614410369211, 'learning_rate': 1e-05, 'epoch': 0.59}
0: {'loss': 0.6977, 'grad_norm': 0.409589348668179, 'learning_rate': 1e-05, 'epoch': 0.59}
0: 5:27:57,  3.23s/it]
                                                        

 59%|█████▊    | 8640/14725 [7:52:10<5:27:57,  3.23s/it]
 59%|█████▊    | 8641/14725 [7:52:13<5:29:26,  3.25s/it]
 59%|█████▊    | 8642/14725 [7:52:17<5:32:39,  3.28s/it]
 59%|█████▊    | 8643/14725 [7:52:20<5:24:43,  3.20s/it]
 59%|█████▊    | 8644/14725 [7:52:23<5:23:53,  3.20s/it]
 59%|█████▊    | 8645/14725 [7:52:26<5:24:05,  3.20s/it]
 59%|█████▊    | 8646/14725 [7:52:29<5:29:40,  3.25s/it]
 59%|█████▊    | 8647/14725 [7:52:33<5:25:14,  3.21s/it]
 59%|█████▊    | 8648/14725 [7:52:36<5:27:01,  3.23s/it]
 59%|█████▊    | 8649/14725 [7:52:39<5:24:54,  3.21s/it]
 59%|█████▊    | 8650/14725 [7:52:42<5:26:17,  3.22s/it]
                                                        

 59%|█████▊    | 8650/14725 [7:52:42<5:26:17,  3.22s/it]
 59%|█████▉    | 8651/14725 [7:52:46<5:34:28,  3
0: {'loss': 0.6941, 'grad_norm': 0.4009086869394943, 'learning_rate': 1e-05, 'epoch': 0.59}
0: .30s/it]
 59%|█████▉    | 8652/14725 [7:52:49<5:36:55,  3.33s/it]
 59%|█████▉    | 8653/14725 [7:52:52<5:35:38,  3.32s/it]
 59%|█████▉    | 8654/14725 [7:52:56<5:36:48,  3.33s/it]
 59%|█████▉    | 8655/14725 [7:52:59<5:33:08,  3.29s/it]
 59%|█████▉    | 8656/14725 [7:53:02<5:31:48,  3.28s/it]
 59%|█████▉    | 8657/14725 [7:53:05<5:31:28,  3.28s/it]
 59%|█████▉    | 8658/14725 [7:53:09<5:24:03,  3.20s/it]
 59%|█████▉    | 8659/14725 [7:53:12<5:23:29,  3.20s/it]
 59%|█████▉    | 8660/14725 [7:53:15<5:21:24,  3.18s/it]
                                                        

 59%|█████▉    | 8660/14725 [7:53:15<5:21:24,  3.18s/it]
 59%|█████▉    | 8661/14725 [7:53:18<5:26:46,  3.23s/it]
 59%|█████▉    | 8662/14725 [7:53:21<5:28:04,  3.25s/it]
 59%|█████▉    | 8663/14725 [7:53:25<5:30:22,  3.27s/it]
 59%|█████▉    | 8664/14725 [7:53:28<5:28:16,  3
0: {'loss': 0.703, 'grad_norm': 0.4148564874746042, 'learning_rate': 1e-05, 'epoch': 0.59}
0: .25s/it]
 59%|█████▉    | 8665/14725 [7:53:31<5:27:28,  3.24s/it]
 59%|█████▉    | 8666/14725 [7:53:35<5:33:41,  3.30s/it]
 59%|█████▉    | 8667/14725 [7:53:38<5:27:46,  3.25s/it]
 59%|█████▉    | 8668/14725 [7:53:41<5:28:04,  3.25s/it]
 59%|█████▉    | 8669/14725 [7:53:44<5:31:46,  3.29s/it]
 59%|█████▉    | 8670/14725 [7:53:48<5:28:12,  3.25s/it]
                                                        

 59%|█████▉    | 8670/14725 [7:53:48<5:28:12,  3.25s/it]
 59%|█████▉    | 8671/14725 [7:53:51<5:26:23,  3.23s/it]
 59%|█████▉    | 8672/14725 [7:53:54<5:28:45,  3.26s/it]
 59%|█████▉    | 8673/14725 [7:53:57<5:28:03,  3.25s/it]
 59%|█████▉    | 8674/14725 [7:54:01<5:28:53,  3.26s/it]
 59%|█████▉    | 8675/14725 [7:54:04<5:27:38,  3.25s/it]
 59%|█████▉    | 8676/14725 [7:54:07<5:29:31,  3.27s/it]
 59%|█████▉    | 8677/14725 [7:54:10<5:23:59,  3
0: {'loss': 0.6856, 'grad_norm': 0.3842468738761172, 'learning_rate': 1e-05, 'epoch': 0.59}
0: .21s/it]
 59%|█████▉    | 8678/14725 [7:54:14<5:29:04,  3.27s/it]
 59%|█████▉    | 8679/14725 [7:54:17<5:30:05,  3.28s/it]
 59%|█████▉    | 8680/14725 [7:54:20<5:23:47,  3.21s/it]
                                                        

 59%|█████▉    | 8680/14725 [7:54:20<5:23:47,  3.21s/it]
 59%|█████▉    | 8681/14725 [7:54:23<5:24:05,  3.22s/it]
 59%|█████▉    | 8682/14725 [7:54:26<5:23:58,  3.22s/it]
 59%|█████▉    | 8683/14725 [7:54:30<5:21:18,  3.19s/it]
 59%|█████▉    | 8684/14725 [7:54:33<5:24:40,  3.22s/it]
 59%|█████▉    | 8685/14725 [7:54:36<5:23:53,  3.22s/it]
 59%|█████▉    | 8686/14725 [7:54:39<5:28:11,  3.26s/it]
 59%|█████▉    | 8687/14725 [7:54:43<5:25:48,  3.24s/it]
 59%|█████▉    | 8688/14725 [7:54:46<5:32:22,  3.30s/it]
 59%|█████▉    | 8689/14725 [7:54:49<5:33:39,  3.32s/it]
 59%|█████▉    | 8690/14725 [7:54:53<5:30:17,  3
0: {'loss': 0.6862, 'grad_norm': 0.41649660043197523, 'learning_rate': 1e-05, 'epoch': 0.59}
0: {'loss': 0.6877, 'grad_norm': 0.4730883417179645, 'learning_rate': 1e-05, 'epoch': 0.59}
0: .28s/it]
                                                        

 59%|█████▉    | 8690/14725 [7:54:53<5:30:17,  3.28s/it]
 59%|█████▉    | 8691/14725 [7:54:56<5:29:20,  3.27s/it]
 59%|█████▉    | 8692/14725 [7:54:59<5:31:29,  3.30s/it]
 59%|█████▉    | 8693/14725 [7:55:03<5:33:57,  3.32s/it]
 59%|█████▉    | 8694/14725 [7:55:06<5:30:40,  3.29s/it]
 59%|█████▉    | 8695/14725 [7:55:09<5:27:11,  3.26s/it]
 59%|█████▉    | 8696/14725 [7:55:12<5:27:47,  3.26s/it]
 59%|█████▉    | 8697/14725 [7:55:15<5:22:56,  3.21s/it]
 59%|█████▉    | 8698/14725 [7:55:19<5:26:07,  3.25s/it]
 59%|█████▉    | 8699/14725 [7:55:22<5:23:30,  3.22s/it]
 59%|█████▉    | 8700/14725 [7:55:25<5:23:55,  3.23s/it]
                                                        

 59%|█████▉    | 8700/14725 [7:55:25<5:23:55,  3.23s/it]
 59%|█████▉    | 8701/14725 [7:55:28<5:23:09,  3.22s/it]
 5
0: {'loss': 0.6856, 'grad_norm': 0.4250061254809362, 'learning_rate': 1e-05, 'epoch': 0.59}
0: 9%|█████▉    | 8702/14725 [7:55:32<5:26:48,  3.26s/it]
 59%|█████▉    | 8703/14725 [7:55:35<5:22:10,  3.21s/it]
 59%|█████▉    | 8704/14725 [7:55:38<5:26:28,  3.25s/it]
 59%|█████▉    | 8705/14725 [7:55:41<5:29:49,  3.29s/it]
 59%|█████▉    | 8706/14725 [7:55:45<5:31:37,  3.31s/it]
 59%|█████▉    | 8707/14725 [7:55:48<5:30:14,  3.29s/it]
 59%|█████▉    | 8708/14725 [7:55:51<5:29:48,  3.29s/it]
 59%|█████▉    | 8709/14725 [7:55:55<5:28:50,  3.28s/it]
 59%|█████▉    | 8710/14725 [7:55:58<5:23:39,  3.23s/it]
                                                        

 59%|█████▉    | 8710/14725 [7:55:58<5:23:39,  3.23s/it]
 59%|█████▉    | 8711/14725 [7:56:01<5:32:16,  3.31s/it]
 59%|█████▉    | 8712/14725 [7:56:04<5:27:23,  3.27s/it]
 59%|█████▉    | 8713/14725 [7:56:07<5:20:30,  3.20s/it]
 59%|█████▉    | 8714/14725 [7:56:10<5:16:11,  3.16s/it]
 5
0: {'loss': 0.6883, 'grad_norm': 0.409137751228765, 'learning_rate': 1e-05, 'epoch': 0.59}
0: 9%|█████▉    | 8715/14725 [7:56:14<5:22:32,  3.22s/it]
 59%|█████▉    | 8716/14725 [7:56:17<5:18:25,  3.18s/it]
 59%|█████▉    | 8717/14725 [7:56:20<5:19:44,  3.19s/it]
 59%|█████▉    | 8718/14725 [7:56:24<5:26:25,  3.26s/it]
 59%|█████▉    | 8719/14725 [7:56:27<5:30:45,  3.30s/it]
 59%|█████▉    | 8720/14725 [7:56:30<5:26:52,  3.27s/it]
                                                        

 59%|█████▉    | 8720/14725 [7:56:30<5:26:52,  3.27s/it]
 59%|█████▉    | 8721/14725 [7:56:33<5:24:39,  3.24s/it]
 59%|█████▉    | 8722/14725 [7:56:37<5:23:58,  3.24s/it]
 59%|█████▉    | 8723/14725 [7:56:40<5:30:59,  3.31s/it]
 59%|█████▉    | 8724/14725 [7:56:43<5:26:48,  3.27s/it]
 59%|█████▉    | 8725/14725 [7:56:47<5:27:44,  3.28s/it]
 59%|█████▉    | 8726/14725 [7:56:50<5:21:59,  3.22s/it]
 59%|█████▉    | 8727/14725 [7:56:53<5:24:21,  3.24s/it]
 5
0: {'loss': 0.6809, 'grad_norm': 0.4016210743565842, 'learning_rate': 1e-05, 'epoch': 0.59}
0: 9%|█████▉    | 8728/14725 [7:56:56<5:23:35,  3.24s/it]
 59%|█████▉    | 8729/14725 [7:57:00<5:28:41,  3.29s/it]
 59%|█████▉    | 8730/14725 [7:57:03<5:28:05,  3.28s/it]
                                                        

 59%|█████▉    | 8730/14725 [7:57:03<5:28:05,  3.28s/it]
 59%|█████▉    | 8731/14725 [7:57:06<5:29:33,  3.30s/it]
 59%|█████▉    | 8732/14725 [7:57:10<5:35:47,  3.36s/it]
 59%|█████▉    | 8733/14725 [7:57:13<5:39:09,  3.40s/it]
 59%|█████▉    | 8734/14725 [7:57:16<5:32:24,  3.33s/it]
 59%|█████▉    | 8735/14725 [7:57:19<5:24:57,  3.26s/it]
 59%|█████▉    | 8736/14725 [7:57:23<5:26:22,  3.27s/it]
 59%|█████▉    | 8737/14725 [7:57:26<5:24:54,  3.26s/it]
 59%|█████▉    | 8738/14725 [7:57:29<5:23:06,  3.24s/it]
 59%|█████▉    | 8739/14725 [7:57:33<5:28:55,  3.30s/it]
 59%|█████▉    | 8740/14725 [7:57:36<5:27:53,  3.29s/it]
  
0: {'loss': 0.6952, 'grad_norm': 0.39942511141009684, 'learning_rate': 1e-05, 'epoch': 0.59}
0: {'loss': 0.6906, 'grad_norm': 0.3990566360533106, 'learning_rate': 1e-05, 'epoch': 0.59}
0:                                                       

 59%|█████▉    | 8740/14725 [7:57:36<5:27:53,  3.29s/it]
 59%|█████▉    | 8741/14725 [7:57:39<5:22:21,  3.23s/it]
 59%|█████▉    | 8742/14725 [7:57:42<5:20:23,  3.21s/it]
 59%|█████▉    | 8743/14725 [7:57:45<5:17:03,  3.18s/it]
 59%|█████▉    | 8744/14725 [7:57:49<5:22:05,  3.23s/it]
 59%|█████▉    | 8745/14725 [7:57:52<5:21:03,  3.22s/it]
 59%|█████▉    | 8746/14725 [7:57:55<5:22:13,  3.23s/it]
 59%|█████▉    | 8747/14725 [7:57:58<5:28:03,  3.29s/it]
 59%|█████▉    | 8748/14725 [7:58:02<5:25:37,  3.27s/it]
 59%|█████▉    | 8749/14725 [7:58:05<5:19:26,  3.21s/it]
 59%|█████▉    | 8750/14725 [7:58:08<5:19:22,  3.21s/it]
                                                        

 59%|█████▉    | 8750/14725 [7:58:08<5:19:22,  3.21s/it]
 59%|█████▉    | 8751/14725 [7:58:11<5:25:32,  3.27s/it]
 59%|██�
0: {'loss': 0.6811, 'grad_norm': 0.44978700520482984, 'learning_rate': 1e-05, 'epoch': 0.59}
0: �██▉    | 8752/14725 [7:58:15<5:28:34,  3.30s/it]
 59%|█████▉    | 8753/14725 [7:58:18<5:32:07,  3.34s/it]
 59%|█████▉    | 8754/14725 [7:58:21<5:28:47,  3.30s/it]
 59%|█████▉    | 8755/14725 [7:58:25<5:30:43,  3.32s/it]
 59%|█████▉    | 8756/14725 [7:58:28<5:31:11,  3.33s/it]
 59%|█████▉    | 8757/14725 [7:58:31<5:24:42,  3.26s/it]
 59%|█████▉    | 8758/14725 [7:58:34<5:22:40,  3.24s/it]
 59%|█████▉    | 8759/14725 [7:58:38<5:24:48,  3.27s/it]
 59%|█████▉    | 8760/14725 [7:58:41<5:23:16,  3.25s/it]
                                                        

 59%|█████▉    | 8760/14725 [7:58:41<5:23:16,  3.25s/it]
 59%|█████▉    | 8761/14725 [7:58:44<5:24:41,  3.27s/it]
 60%|█████▉    | 8762/14725 [7:58:48<5:25:40,  3.28s/it]
 60%|█████▉    | 8763/14725 [7:58:51<5:19:30,  3.22s/it]
 60%|█████▉    | 8764/14725 [7:58:54<5:18:57,  3.21s/it]
 60%|██�
0: {'loss': 0.6985, 'grad_norm': 0.4072428692640305, 'learning_rate': 1e-05, 'epoch': 0.6}
0: �██▉    | 8765/14725 [7:58:57<5:20:13,  3.22s/it]
 60%|█████▉    | 8766/14725 [7:59:00<5:20:34,  3.23s/it]
 60%|█████▉    | 8767/14725 [7:59:04<5:28:09,  3.30s/it]
 60%|█████▉    | 8768/14725 [7:59:07<5:29:22,  3.32s/it]
 60%|█████▉    | 8769/14725 [7:59:10<5:23:54,  3.26s/it]
 60%|█████▉    | 8770/14725 [7:59:14<5:24:35,  3.27s/it]
                                                        

 60%|█████▉    | 8770/14725 [7:59:14<5:24:35,  3.27s/it]
 60%|█████▉    | 8771/14725 [7:59:17<5:26:41,  3.29s/it]
 60%|█████▉    | 8772/14725 [7:59:20<5:21:56,  3.24s/it]
 60%|█████▉    | 8773/14725 [7:59:24<5:31:40,  3.34s/it]
 60%|█████▉    | 8774/14725 [7:59:27<5:25:23,  3.28s/it]
 60%|█████▉    | 8775/14725 [7:59:30<5:25:42,  3.28s/it]
 60%|█████▉    | 8776/14725 [7:59:34<5:33:08,  3.36s/it]
 60%|█████▉    | 8777/14725 [7:59:37<5:25:28,  3.28s/it]
 60%|██�
0: {'loss': 0.6812, 'grad_norm': 0.3998703059845488, 'learning_rate': 1e-05, 'epoch': 0.6}
0: �██▉    | 8778/14725 [7:59:40<5:31:40,  3.35s/it]
 60%|█████▉    | 8779/14725 [7:59:43<5:29:31,  3.33s/it]
 60%|█████▉    | 8780/14725 [7:59:47<5:23:07,  3.26s/it]
                                                        

 60%|█████▉    | 8780/14725 [7:59:47<5:23:07,  3.26s/it]
 60%|█████▉    | 8781/14725 [7:59:50<5:20:31,  3.24s/it]
 60%|█████▉    | 8782/14725 [7:59:53<5:23:27,  3.27s/it]
 60%|█████▉    | 8783/14725 [7:59:56<5:20:26,  3.24s/it]
 60%|█████▉    | 8784/14725 [7:59:59<5:20:48,  3.24s/it]
 60%|█████▉    | 8785/14725 [8:00:03<5:21:11,  3.24s/it]
 60%|█████▉    | 8786/14725 [8:00:06<5:21:16,  3.25s/it]
 60%|█████▉    | 8787/14725 [8:00:09<5:22:43,  3.26s/it]
 60%|█████▉    | 8788/14725 [8:00:12<5:21:27,  3.25s/it]
 60%|█████▉    | 8789/14725 [8:00:16<5:23:19,  3.27s/it]
 60%|█████▉    | 8790/14725 [8:00:19<5:20:20,  3.24s/it]
             
0: {'loss': 0.6953, 'grad_norm': 0.3901401617743541, 'learning_rate': 1e-05, 'epoch': 0.6}
0: {'loss': 0.6854, 'grad_norm': 0.41470668740979844, 'learning_rate': 1e-05, 'epoch': 0.6}
0:                                            

 60%|█████▉    | 8790/14725 [8:00:19<5:20:20,  3.24s/it]
 60%|█████▉    | 8791/14725 [8:00:22<5:25:50,  3.29s/it]
 60%|█████▉    | 8792/14725 [8:00:26<5:31:06,  3.35s/it]
 60%|█████▉    | 8793/14725 [8:00:29<5:24:49,  3.29s/it]
 60%|█████▉    | 8794/14725 [8:00:32<5:29:15,  3.33s/it]
 60%|█████▉    | 8795/14725 [8:00:36<5:26:59,  3.31s/it]
 60%|█████▉    | 8796/14725 [8:00:39<5:29:04,  3.33s/it]
 60%|█████▉    | 8797/14725 [8:00:43<5:33:47,  3.38s/it]
 60%|█████▉    | 8798/14725 [8:00:46<5:25:50,  3.30s/it]
 60%|█████▉    | 8799/14725 [8:00:49<5:26:59,  3.31s/it]
 60%|█████▉    | 8800/14725 [8:00:52<5:26:20,  3.30s/it]
                                                        

 60%|█████▉    | 8800/14725 [8:00:52<5:26:20,  3.30s/it]
 60%|█████▉    | 8801/14725 [8:00:56<5:24:53,  3.29s/it]
 60%|█████▉ 
0: {'loss': 0.6857, 'grad_norm': 0.4248798868693646, 'learning_rate': 1e-05, 'epoch': 0.6}
0:    | 8802/14725 [8:00:59<5:25:57,  3.30s/it]
 60%|█████▉    | 8803/14725 [8:01:02<5:18:48,  3.23s/it]
 60%|█████▉    | 8804/14725 [8:01:05<5:26:13,  3.31s/it]
 60%|█████▉    | 8805/14725 [8:01:09<5:27:47,  3.32s/it]
 60%|█████▉    | 8806/14725 [8:01:12<5:28:07,  3.33s/it]
 60%|█████▉    | 8807/14725 [8:01:15<5:22:31,  3.27s/it]
 60%|█████▉    | 8808/14725 [8:01:19<5:24:55,  3.29s/it]
 60%|█████▉    | 8809/14725 [8:01:22<5:27:29,  3.32s/it]
 60%|█████▉    | 8810/14725 [8:01:25<5:30:14,  3.35s/it]
                                                        

 60%|█████▉    | 8810/14725 [8:01:25<5:30:14,  3.35s/it]
 60%|█████▉    | 8811/14725 [8:01:29<5:31:46,  3.37s/it]
 60%|█████▉    | 8812/14725 [8:01:32<5:32:20,  3.37s/it]
 60%|█████▉    | 8813/14725 [8:01:36<5:38:14,  3.43s/it]
 60%|█████▉    | 8814/14725 [8:01:39<5:30:06,  3.35s/it]
 60%|█████▉ 
0: {'loss': 0.6896, 'grad_norm': 0.41749119813931024, 'learning_rate': 1e-05, 'epoch': 0.6}
0:    | 8815/14725 [8:01:42<5:28:58,  3.34s/it]
 60%|█████▉    | 8816/14725 [8:01:45<5:23:58,  3.29s/it]
 60%|█████▉    | 8817/14725 [8:01:49<5:22:27,  3.27s/it]
 60%|█████▉    | 8818/14725 [8:01:52<5:21:45,  3.27s/it]
 60%|█████▉    | 8819/14725 [8:01:55<5:16:04,  3.21s/it]
 60%|█████▉    | 8820/14725 [8:01:58<5:15:41,  3.21s/it]
                                                        

 60%|█████▉    | 8820/14725 [8:01:58<5:15:41,  3.21s/it]
 60%|█████▉    | 8821/14725 [8:02:01<5:14:05,  3.19s/it]
 60%|█████▉    | 8822/14725 [8:02:05<5:13:22,  3.19s/it]
 60%|█████▉    | 8823/14725 [8:02:08<5:13:28,  3.19s/it]
 60%|█████▉    | 8824/14725 [8:02:11<5:15:17,  3.21s/it]
 60%|█████▉    | 8825/14725 [8:02:14<5:17:48,  3.23s/it]
 60%|█████▉    | 8826/14725 [8:02:18<5:19:39,  3.25s/it]
 60%|█████▉    | 8827/14725 [8:02:21<5:21:11,  3.27s/it]
 60%|█████▉ 
0: {'loss': 0.6893, 'grad_norm': 0.42237556106667307, 'learning_rate': 1e-05, 'epoch': 0.6}
0:    | 8828/14725 [8:02:24<5:24:49,  3.30s/it]
 60%|█████▉    | 8829/14725 [8:02:27<5:21:00,  3.27s/it]
 60%|█████▉    | 8830/14725 [8:02:31<5:23:36,  3.29s/it]
                                                        

 60%|█████▉    | 8830/14725 [8:02:31<5:23:36,  3.29s/it]
 60%|█████▉    | 8831/14725 [8:02:34<5:21:51,  3.28s/it]
 60%|█████▉    | 8832/14725 [8:02:37<5:20:19,  3.26s/it]
 60%|█████▉    | 8833/14725 [8:02:40<5:18:20,  3.24s/it]
 60%|█████▉    | 8834/14725 [8:02:44<5:17:32,  3.23s/it]
 60%|██████    | 8835/14725 [8:02:47<5:24:44,  3.31s/it]
 60%|██████    | 8836/14725 [8:02:58<9:11:11,  5.62s/it]
 60%|██████    | 8837/14725 [8:03:02<8:04:41,  4.94s/it]
 60%|██████    | 8838/14725 [8:03:05<7:13:08,  4.41s/it]
 60%|██████    | 8839/14725 [8:03:08<6:38:28,  4.06s/it]
 60%|██████    | 8840/14725 [8:03:11<6:11:26,  3.79s/it]
                        
0: {'loss': 0.6945, 'grad_norm': 0.41478092171753245, 'learning_rate': 1e-05, 'epoch': 0.6}
0: {'loss': 0.6951, 'grad_norm': 0.4158091244671277, 'learning_rate': 1e-05, 'epoch': 0.6}
0:                                 

 60%|██████    | 8840/14725 [8:03:11<6:11:26,  3.79s/it]
 60%|██████    | 8841/14725 [8:03:14<5:58:28,  3.66s/it]
 60%|██████    | 8842/14725 [8:03:18<5:44:52,  3.52s/it]
 60%|██████    | 8843/14725 [8:03:21<5:36:46,  3.44s/it]
 60%|██████    | 8844/14725 [8:03:24<5:37:19,  3.44s/it]
 60%|██████    | 8845/14725 [8:03:28<5:41:34,  3.49s/it]
 60%|██████    | 8846/14725 [8:03:31<5:30:48,  3.38s/it]
 60%|██████    | 8847/14725 [8:03:35<5:36:38,  3.44s/it]
 60%|██████    | 8848/14725 [8:03:38<5:36:17,  3.43s/it]
 60%|██████    | 8849/14725 [8:03:42<5:37:35,  3.45s/it]
 60%|██████    | 8850/14725 [8:03:45<5:33:01,  3.40s/it]
                                                        

 60%|██████    | 8850/14725 [8:03:45<5:33:01,  3.40s/it]
 60%|██████    | 8851/14725 [8:03:48<5:28:34,  3.36s/it]
 60%|██████    | 8852/1
0: {'loss': 0.6784, 'grad_norm': 0.4136578814504646, 'learning_rate': 1e-05, 'epoch': 0.6}
0: 4725 [8:03:51<5:23:28,  3.30s/it]
 60%|██████    | 8853/14725 [8:03:54<5:16:05,  3.23s/it]
 60%|██████    | 8854/14725 [8:03:58<5:18:18,  3.25s/it]
 60%|██████    | 8855/14725 [8:04:01<5:15:21,  3.22s/it]
 60%|██████    | 8856/14725 [8:04:04<5:09:27,  3.16s/it]
 60%|██████    | 8857/14725 [8:04:07<5:13:58,  3.21s/it]
 60%|██████    | 8858/14725 [8:04:10<5:13:41,  3.21s/it]
 60%|██████    | 8859/14725 [8:04:13<5:11:28,  3.19s/it]
 60%|██████    | 8860/14725 [8:04:17<5:14:05,  3.21s/it]
                                                        

 60%|██████    | 8860/14725 [8:04:17<5:14:05,  3.21s/it]
 60%|██████    | 8861/14725 [8:04:20<5:19:54,  3.27s/it]
 60%|██████    | 8862/14725 [8:04:24<5:33:10,  3.41s/it]
 60%|██████    | 8863/14725 [8:04:27<5:34:13,  3.42s/it]
 60%|██████    | 8864/14725 [8:04:30<5:26:50,  3.35s/it]
 60%|██████    | 8865/1
0: {'loss': 0.6959, 'grad_norm': 0.4182411703817849, 'learning_rate': 1e-05, 'epoch': 0.6}
0: 4725 [8:04:34<5:21:20,  3.29s/it]
 60%|██████    | 8866/14725 [8:04:37<5:21:35,  3.29s/it]
 60%|██████    | 8867/14725 [8:04:40<5:25:57,  3.34s/it]
 60%|██████    | 8868/14725 [8:04:44<5:23:26,  3.31s/it]
 60%|██████    | 8869/14725 [8:04:47<5:26:45,  3.35s/it]
 60%|██████    | 8870/14725 [8:04:50<5:20:14,  3.28s/it]
                                                        

 60%|██████    | 8870/14725 [8:04:50<5:20:14,  3.28s/it]
 60%|██████    | 8871/14725 [8:04:53<5:19:15,  3.27s/it]
 60%|██████    | 8872/14725 [8:04:57<5:15:05,  3.23s/it]
 60%|██████    | 8873/14725 [8:05:00<5:19:00,  3.27s/it]
 60%|██████    | 8874/14725 [8:05:03<5:19:58,  3.28s/it]
 60%|██████    | 8875/14725 [8:05:06<5:16:54,  3.25s/it]
 60%|██████    | 8876/14725 [8:05:10<5:15:27,  3.24s/it]
 60%|██████    | 8877/14725 [8:05:13<5:16:53,  3.25s/it]
 60%|██████    | 8878/1
0: {'loss': 0.7219, 'grad_norm': 0.42491485735067563, 'learning_rate': 1e-05, 'epoch': 0.6}
0: {'loss': 0.6842, 'grad_norm': 0.43295513119553425, 'learning_rate': 1e-05, 'epoch': 0.6}
0: 4725 [8:05:16<5:15:33,  3.24s/it]
 60%|██████    | 8879/14725 [8:05:19<5:10:54,  3.19s/it]
 60%|██████    | 8880/14725 [8:05:22<5:09:49,  3.18s/it]
                                                        

 60%|██████    | 8880/14725 [8:05:22<5:09:49,  3.18s/it]
 60%|██████    | 8881/14725 [8:05:25<5:06:49,  3.15s/it]
 60%|██████    | 8882/14725 [8:05:29<5:08:03,  3.16s/it]
 60%|██████    | 8883/14725 [8:05:32<5:12:03,  3.20s/it]
 60%|██████    | 8884/14725 [8:05:35<5:14:19,  3.23s/it]
 60%|██████    | 8885/14725 [8:05:38<5:14:35,  3.23s/it]
 60%|██████    | 8886/14725 [8:05:42<5:19:31,  3.28s/it]
 60%|██████    | 8887/14725 [8:05:45<5:22:00,  3.31s/it]
 60%|██████    | 8888/14725 [8:05:49<5:28:26,  3.38s/it]
 60%|██████    | 8889/14725 [8:05:52<5:25:33,  3.35s/it]
 60%|██████    | 8890/14725 [8:05:55<5:20:05,  3.29s/it]
                                   
0: {'loss': 0.6905, 'grad_norm': 0.4070703669796774, 'learning_rate': 1e-05, 'epoch': 0.6}
0:                      

 60%|██████    | 8890/14725 [8:05:55<5:20:05,  3.29s/it]
 60%|██████    | 8891/14725 [8:05:58<5:16:43,  3.26s/it]
 60%|██████    | 8892/14725 [8:06:01<5:10:02,  3.19s/it]
 60%|██████    | 8893/14725 [8:06:05<5:16:17,  3.25s/it]
 60%|██████    | 8894/14725 [8:06:08<5:16:13,  3.25s/it]
 60%|██████    | 8895/14725 [8:06:12<5:21:59,  3.31s/it]
 60%|██████    | 8896/14725 [8:06:15<5:31:26,  3.41s/it]
 60%|██████    | 8897/14725 [8:06:19<5:36:07,  3.46s/it]
 60%|██████    | 8898/14725 [8:06:22<5:29:35,  3.39s/it]
 60%|██████    | 8899/14725 [8:06:25<5:28:32,  3.38s/it]
 60%|██████    | 8900/14725 [8:06:28<5:21:34,  3.31s/it]
                                                        

 60%|██████    | 8900/14725 [8:06:28<5:21:34,  3.31s/it]
 60%|██████    | 8901/14725 [8:06:32<5:20:51,  3.31s/it]
 60%|██████    | 8902/14725 [8:06:
0: {'loss': 0.6808, 'grad_norm': 0.4124454874021646, 'learning_rate': 1e-05, 'epoch': 0.61}
0: 35<5:17:46,  3.27s/it]
 60%|██████    | 8903/14725 [8:06:38<5:14:11,  3.24s/it]
 60%|██████    | 8904/14725 [8:06:41<5:12:18,  3.22s/it]
 60%|██████    | 8905/14725 [8:06:45<5:17:14,  3.27s/it]
 60%|██████    | 8906/14725 [8:06:48<5:21:40,  3.32s/it]
 60%|██████    | 8907/14725 [8:06:51<5:20:48,  3.31s/it]
 60%|██████    | 8908/14725 [8:06:55<5:14:48,  3.25s/it]
 61%|██████    | 8909/14725 [8:06:58<5:10:10,  3.20s/it]
 61%|██████    | 8910/14725 [8:07:01<5:10:16,  3.20s/it]
                                                        

 61%|██████    | 8910/14725 [8:07:01<5:10:16,  3.20s/it]
 61%|██████    | 8911/14725 [8:07:04<5:11:17,  3.21s/it]
 61%|██████    | 8912/14725 [8:07:07<5:08:43,  3.19s/it]
 61%|██████    | 8913/14725 [8:07:10<5:10:01,  3.20s/it]
 61%|██████    | 8914/14725 [8:07:14<5:09:15,  3.19s/it]
 61%|██████    | 8915/14725 [8:07:
0: {'loss': 0.6888, 'grad_norm': 0.4345365169403108, 'learning_rate': 1e-05, 'epoch': 0.61}
0: 17<5:11:52,  3.22s/it]
 61%|██████    | 8916/14725 [8:07:20<5:11:39,  3.22s/it]
 61%|██████    | 8917/14725 [8:07:23<5:15:21,  3.26s/it]
 61%|██████    | 8918/14725 [8:07:27<5:12:17,  3.23s/it]
 61%|██████    | 8919/14725 [8:07:30<5:19:25,  3.30s/it]
 61%|██████    | 8920/14725 [8:07:33<5:18:45,  3.29s/it]
                                                        

 61%|██████    | 8920/14725 [8:07:33<5:18:45,  3.29s/it]
 61%|██████    | 8921/14725 [8:07:37<5:34:45,  3.46s/it]
 61%|██████    | 8922/14725 [8:07:40<5:28:03,  3.39s/it]
 61%|██████    | 8923/14725 [8:07:44<5:30:52,  3.42s/it]
 61%|██████    | 8924/14725 [8:07:47<5:26:31,  3.38s/it]
 61%|██████    | 8925/14725 [8:07:51<5:32:25,  3.44s/it]
 61%|██████    | 8926/14725 [8:07:54<5:29:57,  3.41s/it]
 61%|██████    | 8927/14725 [8:07:57<5:28:05,  3.40s/it]
 61%|██████    | 8928/14725 [8:08:
0: {'loss': 0.6984, 'grad_norm': 0.41939971259486664, 'learning_rate': 1e-05, 'epoch': 0.61}
0: 01<5:23:13,  3.35s/it]
 61%|██████    | 8929/14725 [8:08:04<5:19:20,  3.31s/it]
 61%|██████    | 8930/14725 [8:08:07<5:14:08,  3.25s/it]
                                                        

 61%|██████    | 8930/14725 [8:08:07<5:14:08,  3.25s/it]
 61%|██████    | 8931/14725 [8:08:10<5:12:40,  3.24s/it]
 61%|██████    | 8932/14725 [8:08:14<5:13:01,  3.24s/it]
 61%|██████    | 8933/14725 [8:08:17<5:17:43,  3.29s/it]
 61%|██████    | 8934/14725 [8:08:20<5:22:30,  3.34s/it]
 61%|██████    | 8935/14725 [8:08:24<5:19:01,  3.31s/it]
 61%|██████    | 8936/14725 [8:08:27<5:22:49,  3.35s/it]
 61%|██████    | 8937/14725 [8:08:30<5:21:06,  3.33s/it]
 61%|██████    | 8938/14725 [8:08:34<5:23:00,  3.35s/it]
 61%|██████    | 8939/14725 [8:08:37<5:24:34,  3.37s/it]
 61%|██████    | 8940/14725 [8:08:40<5:22:37,  3.35s/it]
                                              
0: {'loss': 0.6962, 'grad_norm': 0.40432564070035903, 'learning_rate': 1e-05, 'epoch': 0.61}
0: {'loss': 0.6901, 'grad_norm': 0.4166675621035091, 'learning_rate': 1e-05, 'epoch': 0.61}
0:           

 61%|██████    | 8940/14725 [8:08:40<5:22:37,  3.35s/it]
 61%|██████    | 8941/14725 [8:08:44<5:27:53,  3.40s/it]
 61%|██████    | 8942/14725 [8:08:47<5:24:19,  3.36s/it]
 61%|██████    | 8943/14725 [8:08:51<5:23:35,  3.36s/it]
 61%|██████    | 8944/14725 [8:08:54<5:20:03,  3.32s/it]
 61%|██████    | 8945/14725 [8:08:57<5:20:23,  3.33s/it]
 61%|██████    | 8946/14725 [8:09:00<5:17:33,  3.30s/it]
 61%|██████    | 8947/14725 [8:09:04<5:23:03,  3.35s/it]
 61%|██████    | 8948/14725 [8:09:07<5:28:32,  3.41s/it]
 61%|██████    | 8949/14725 [8:09:11<5:22:05,  3.35s/it]
 61%|██████    | 8950/14725 [8:09:14<5:22:58,  3.36s/it]
                                                        

 61%|██████    | 8950/14725 [8:09:14<5:22:58,  3.36s/it]
 61%|██████    | 8951/14725 [8:09:17<5:26:33,  3.39s/it]
 61%|██████    | 8952/14725 [8:09:21<5:18:47,
0: {'loss': 0.6738, 'grad_norm': 0.4119425246122353, 'learning_rate': 1e-05, 'epoch': 0.61}
0:   3.31s/it]
 61%|██████    | 8953/14725 [8:09:24<5:18:33,  3.31s/it]
 61%|██████    | 8954/14725 [8:09:27<5:16:53,  3.29s/it]
 61%|██████    | 8955/14725 [8:09:30<5:09:14,  3.22s/it]
 61%|██████    | 8956/14725 [8:09:34<5:15:12,  3.28s/it]
 61%|██████    | 8957/14725 [8:09:37<5:16:45,  3.30s/it]
 61%|██████    | 8958/14725 [8:09:40<5:11:59,  3.25s/it]
 61%|██████    | 8959/14725 [8:09:44<5:20:58,  3.34s/it]
 61%|██████    | 8960/14725 [8:09:47<5:17:53,  3.31s/it]
                                                        

 61%|██████    | 8960/14725 [8:09:47<5:17:53,  3.31s/it]
 61%|██████    | 8961/14725 [8:09:50<5:16:34,  3.30s/it]
 61%|██████    | 8962/14725 [8:09:54<5:19:41,  3.33s/it]
 61%|██████    | 8963/14725 [8:09:57<5:23:46,  3.37s/it]
 61%|██████    | 8964/14725 [8:10:00<5:16:50,  3.30s/it]
 61%|██████    | 8965/14725 [8:10:03<5:12:24,
0: {'loss': 0.7054, 'grad_norm': 0.41969635340522854, 'learning_rate': 1e-05, 'epoch': 0.61}
0:   3.25s/it]
 61%|██████    | 8966/14725 [8:10:07<5:14:13,  3.27s/it]
 61%|██████    | 8967/14725 [8:10:10<5:19:36,  3.33s/it]
 61%|██████    | 8968/14725 [8:10:13<5:15:45,  3.29s/it]
 61%|██████    | 8969/14725 [8:10:16<5:13:30,  3.27s/it]
 61%|██████    | 8970/14725 [8:10:20<5:16:49,  3.30s/it]
                                                        

 61%|██████    | 8970/14725 [8:10:20<5:16:49,  3.30s/it]
 61%|██████    | 8971/14725 [8:10:23<5:11:20,  3.25s/it]
 61%|██████    | 8972/14725 [8:10:26<5:05:23,  3.19s/it]
 61%|██████    | 8973/14725 [8:10:29<5:13:13,  3.27s/it]
 61%|██████    | 8974/14725 [8:10:33<5:09:36,  3.23s/it]
 61%|██████    | 8975/14725 [8:10:36<5:10:12,  3.24s/it]
 61%|██████    | 8976/14725 [8:10:39<5:10:58,  3.25s/it]
 61%|██████    | 8977/14725 [8:10:42<5:05:57,  3.19s/it]
 61%|██████    | 8978/14725 [8:10:45<5:06:32,
0: {'loss': 0.6935, 'grad_norm': 0.3961206655700515, 'learning_rate': 1e-05, 'epoch': 0.61}
0:   3.20s/it]
 61%|██████    | 8979/14725 [8:10:49<5:16:41,  3.31s/it]
 61%|██████    | 8980/14725 [8:10:52<5:09:07,  3.23s/it]
                                                        

 61%|██████    | 8980/14725 [8:10:52<5:09:07,  3.23s/it]
 61%|██████    | 8981/14725 [8:10:55<5:08:34,  3.22s/it]
 61%|██████    | 8982/14725 [8:10:59<5:13:46,  3.28s/it]
 61%|██████    | 8983/14725 [8:11:02<5:17:19,  3.32s/it]
 61%|██████    | 8984/14725 [8:11:05<5:18:54,  3.33s/it]
 61%|██████    | 8985/14725 [8:11:09<5:14:51,  3.29s/it]
 61%|██████    | 8986/14725 [8:11:12<5:16:40,  3.31s/it]
 61%|██████    | 8987/14725 [8:11:15<5:14:20,  3.29s/it]
 61%|██████    | 8988/14725 [8:11:18<5:11:54,  3.26s/it]
 61%|██████    | 8989/14725 [8:11:22<5:08:34,  3.23s/it]
 61%|██████    | 8990/14725 [8:11:25<5:10:51,  3.25s/it]
                                                        
0: {'loss': 0.6858, 'grad_norm': 0.42963080097998185, 'learning_rate': 1e-05, 'epoch': 0.61}
0: {'loss': 0.6817, 'grad_norm': 0.4417711274306189, 'learning_rate': 1e-05, 'epoch': 0.61}
0: 
 61%|██████    | 8990/14725 [8:11:25<5:10:51,  3.25s/it]
 61%|██████    | 8991/14725 [8:11:28<5:10:20,  3.25s/it]
 61%|██████    | 8992/14725 [8:11:31<5:11:34,  3.26s/it]
 61%|██████    | 8993/14725 [8:11:35<5:11:56,  3.27s/it]
 61%|██████    | 8994/14725 [8:11:38<5:10:14,  3.25s/it]
 61%|██████    | 8995/14725 [8:11:41<5:13:01,  3.28s/it]
 61%|██████    | 8996/14725 [8:11:44<5:09:27,  3.24s/it]
 61%|██████    | 8997/14725 [8:11:47<5:04:52,  3.19s/it]
 61%|██████    | 8998/14725 [8:11:51<5:08:05,  3.23s/it]
 61%|██████    | 8999/14725 [8:11:54<5:06:30,  3.21s/it]
 61%|██████    | 9000/14725 [8:11:57<5:09:09,  3.24s/it]
                                                        

 61%|██████    | 9000/14725 [8:11:57<5:09:09,  3.24s/it]
 61%|██████    | 9001/14725 [8:12:00<5:07:48,  3.23s/it]
 61%|██████    | 9002/14725 [8:12:04<5:04:53,  3.20s/it]
0: {'loss': 0.6895, 'grad_norm': 0.4240948807976795, 'learning_rate': 1e-05, 'epoch': 0.61}
0: 
 61%|██████    | 9003/14725 [8:12:07<5:05:19,  3.20s/it]
 61%|██████    | 9004/14725 [8:12:10<5:07:47,  3.23s/it]
 61%|██████    | 9005/14725 [8:12:13<5:10:56,  3.26s/it]
 61%|██████    | 9006/14725 [8:12:17<5:08:53,  3.24s/it]
 61%|██████    | 9007/14725 [8:12:20<5:07:22,  3.23s/it]
 61%|██████    | 9008/14725 [8:12:23<5:12:38,  3.28s/it]
 61%|██████    | 9009/14725 [8:12:26<5:11:38,  3.27s/it]
 61%|██████    | 9010/14725 [8:12:30<5:14:21,  3.30s/it]
                                                        

 61%|██████    | 9010/14725 [8:12:30<5:14:21,  3.30s/it]
 61%|██████    | 9011/14725 [8:12:33<5:18:12,  3.34s/it]
 61%|██████    | 9012/14725 [8:12:36<5:10:13,  3.26s/it]
 61%|██████    | 9013/14725 [8:12:40<5:09:16,  3.25s/it]
 61%|██████    | 9014/14725 [8:12:43<5:16:38,  3.33s/it]
 61%|██████    | 9015/14725 [8:12:46<5:14:16,  3.30s/it]
0: {'loss': 0.6963, 'grad_norm': 0.4247983481299442, 'learning_rate': 1e-05, 'epoch': 0.61}
0: 
 61%|██████    | 9016/14725 [8:12:49<5:10:50,  3.27s/it]
 61%|██████    | 9017/14725 [8:12:53<5:14:54,  3.31s/it]
 61%|██████    | 9018/14725 [8:12:56<5:13:21,  3.29s/it]
 61%|██████    | 9019/14725 [8:12:59<5:08:52,  3.25s/it]
 61%|██████▏   | 9020/14725 [8:13:03<5:09:21,  3.25s/it]
                                                        

 61%|██████▏   | 9020/14725 [8:13:03<5:09:21,  3.25s/it]
 61%|██████▏   | 9021/14725 [8:13:06<5:05:11,  3.21s/it]
 61%|██████▏   | 9022/14725 [8:13:09<5:08:37,  3.25s/it]
 61%|██████▏   | 9023/14725 [8:13:12<5:05:05,  3.21s/it]
 61%|██████▏   | 9024/14725 [8:13:16<5:09:56,  3.26s/it]
 61%|██████▏   | 9025/14725 [8:13:19<5:08:54,  3.25s/it]
 61%|██████▏   | 9026/14725 [8:13:22<5:07:36,  3.24s/it]
 61%|██████▏   | 9027/14725 [8:13:25<5:05:38,  3.22s/it]
 61%|██████▏   | 9028/14725 [8:13:28
0: {'loss': 0.6829, 'grad_norm': 0.40893688941907136, 'learning_rate': 1e-05, 'epoch': 0.61}
0: <5:07:45,  3.24s/it]
 61%|██████▏   | 9029/14725 [8:13:32<5:09:54,  3.26s/it]
 61%|██████▏   | 9030/14725 [8:13:35<5:03:49,  3.20s/it]
                                                        

 61%|██████▏   | 9030/14725 [8:13:35<5:03:49,  3.20s/it]
 61%|██████▏   | 9031/14725 [8:13:38<5:03:41,  3.20s/it]
 61%|██████▏   | 9032/14725 [8:13:41<5:09:54,  3.27s/it]
 61%|██████▏   | 9033/14725 [8:13:45<5:06:25,  3.23s/it]
 61%|██████▏   | 9034/14725 [8:13:48<5:02:43,  3.19s/it]
 61%|██████▏   | 9035/14725 [8:13:51<5:04:00,  3.21s/it]
 61%|██████▏   | 9036/14725 [8:13:54<5:05:25,  3.22s/it]
 61%|██████▏   | 9037/14725 [8:13:57<5:06:23,  3.23s/it]
 61%|██████▏   | 9038/14725 [8:14:01<5:03:54,  3.21s/it]
 61%|██████▏   | 9039/14725 [8:14:04<5:04:34,  3.21s/it]
 61%|██████▏   | 9040/14725 [8:14:07<5:04:34,  3.21s/it]
                      
0: {'loss': 0.6929, 'grad_norm': 0.40413830889116303, 'learning_rate': 1e-05, 'epoch': 0.61}
0: {'loss': 0.6976, 'grad_norm': 0.4874855496520055, 'learning_rate': 1e-05, 'epoch': 0.61}
0:                                   

 61%|██████▏   | 9040/14725 [8:14:07<5:04:34,  3.21s/it]
 61%|██████▏   | 9041/14725 [8:14:10<5:07:43,  3.25s/it]
 61%|██████▏   | 9042/14725 [8:14:14<5:06:05,  3.23s/it]
 61%|██████▏   | 9043/14725 [8:14:17<5:04:10,  3.21s/it]
 61%|██████▏   | 9044/14725 [8:14:20<5:09:11,  3.27s/it]
 61%|██████▏   | 9045/14725 [8:14:24<5:17:43,  3.36s/it]
 61%|██████▏   | 9046/14725 [8:14:27<5:20:42,  3.39s/it]
 61%|██████▏   | 9047/14725 [8:14:31<5:20:32,  3.39s/it]
 61%|██████▏   | 9048/14725 [8:14:34<5:13:03,  3.31s/it]
 61%|██████▏   | 9049/14725 [8:14:37<5:10:11,  3.28s/it]
 61%|██████▏   | 9050/14725 [8:14:40<5:09:06,  3.27s/it]
                                                        

 61%|██████▏   | 9050/14725 [8:14:40<5:09:06,  3.27s/it]
 61%|██████▏   | 9051/14725 [8:14:43<5:10:17,  3.28s/it]
 61%|�
0: {'loss': 0.6936, 'grad_norm': 0.39890932970923043, 'learning_rate': 1e-05, 'epoch': 0.62}
0: �█████▏   | 9052/14725 [8:14:47<5:07:52,  3.26s/it]
 61%|██████▏   | 9053/14725 [8:14:50<5:08:41,  3.27s/it]
 61%|██████▏   | 9054/14725 [8:14:53<5:08:39,  3.27s/it]
 61%|██████▏   | 9055/14725 [8:14:56<5:02:49,  3.20s/it]
 62%|██████▏   | 9056/14725 [8:14:59<5:03:09,  3.21s/it]
 62%|██████▏   | 9057/14725 [8:15:03<5:03:14,  3.21s/it]
 62%|██████▏   | 9058/14725 [8:15:06<4:57:55,  3.15s/it]
 62%|██████▏   | 9059/14725 [8:15:09<4:54:59,  3.12s/it]
 62%|██████▏   | 9060/14725 [8:15:12<5:00:33,  3.18s/it]
                                                        

 62%|██████▏   | 9060/14725 [8:15:12<5:00:33,  3.18s/it]
 62%|██████▏   | 9061/14725 [8:15:15<5:08:19,  3.27s/it]
 62%|██████▏   | 9062/14725 [8:15:19<5:03:34,  3.22s/it]
 62%|██████▏   | 9063/14725 [8:15:22<5:14:21,  3.33s/it]
 62%|██████▏   | 9064/14725 [8:15:25
0: {'loss': 0.6779, 'grad_norm': 0.3987111002087947, 'learning_rate': 1e-05, 'epoch': 0.62}
0: <5:13:09,  3.32s/it]
 62%|██████▏   | 9065/14725 [8:15:29<5:07:23,  3.26s/it]
 62%|██████▏   | 9066/14725 [8:15:32<5:07:00,  3.26s/it]
 62%|██████▏   | 9067/14725 [8:15:35<5:05:04,  3.24s/it]
 62%|██████▏   | 9068/14725 [8:15:38<5:05:23,  3.24s/it]
 62%|██████▏   | 9069/14725 [8:15:42<5:05:24,  3.24s/it]
 62%|██████▏   | 9070/14725 [8:15:45<5:05:22,  3.24s/it]
                                                        

 62%|██████▏   | 9070/14725 [8:15:45<5:05:22,  3.24s/it]
 62%|██████▏   | 9071/14725 [8:15:48<5:02:04,  3.21s/it]
 62%|██████▏   | 9072/14725 [8:15:51<5:06:46,  3.26s/it]
 62%|██████▏   | 9073/14725 [8:15:54<5:01:45,  3.20s/it]
 62%|██████▏   | 9074/14725 [8:15:58<5:00:41,  3.19s/it]
 62%|██████▏   | 9075/14725 [8:16:01<5:03:31,  3.22s/it]
 62%|██████▏   | 9076/14725 [8:16:04<5:03:55,  3.23s/it]
 62%|█████�
0: {'loss': 0.6998, 'grad_norm': 0.3864863972752333, 'learning_rate': 1e-05, 'epoch': 0.62}
0: �▏   | 9077/14725 [8:16:07<5:06:49,  3.26s/it]
 62%|██████▏   | 9078/14725 [8:16:11<5:03:48,  3.23s/it]
 62%|██████▏   | 9079/14725 [8:16:14<5:04:18,  3.23s/it]
 62%|██████▏   | 9080/14725 [8:16:17<5:03:41,  3.23s/it]
                                                        

 62%|██████▏   | 9080/14725 [8:16:17<5:03:41,  3.23s/it]
 62%|██████▏   | 9081/14725 [8:16:20<5:00:57,  3.20s/it]
 62%|██████▏   | 9082/14725 [8:16:23<4:58:59,  3.18s/it]
 62%|██████▏   | 9083/14725 [8:16:27<5:01:13,  3.20s/it]
 62%|██████▏   | 9084/14725 [8:16:30<4:59:25,  3.18s/it]
 62%|██████▏   | 9085/14725 [8:16:33<5:00:35,  3.20s/it]
 62%|██████▏   | 9086/14725 [8:16:36<5:02:32,  3.22s/it]
 62%|██████▏   | 9087/14725 [8:16:39<5:02:46,  3.22s/it]
 62%|██████▏   | 9088/14725 [8:16:43<5:05:24,  3.25s/it]
 62%|██████▏   | 9089/14725 [8:16:46<5:09:22,  3.29
0: {'loss': 0.692, 'grad_norm': 0.4290430627323127, 'learning_rate': 1e-05, 'epoch': 0.62}
0: {'loss': 0.6743, 'grad_norm': 0.41357185918867345, 'learning_rate': 1e-05, 'epoch': 0.62}
0: s/it]
 62%|██████▏   | 9090/14725 [8:16:49<5:09:46,  3.30s/it]
                                                        

 62%|██████▏   | 9090/14725 [8:16:49<5:09:46,  3.30s/it]
 62%|██████▏   | 9091/14725 [8:16:53<5:10:15,  3.30s/it]
 62%|██████▏   | 9092/14725 [8:16:56<5:07:13,  3.27s/it]
 62%|██████▏   | 9093/14725 [8:16:59<5:12:58,  3.33s/it]
 62%|██████▏   | 9094/14725 [8:17:03<5:08:09,  3.28s/it]
 62%|██████▏   | 9095/14725 [8:17:06<5:08:23,  3.29s/it]
 62%|██████▏   | 9096/14725 [8:17:09<5:06:41,  3.27s/it]
 62%|██████▏   | 9097/14725 [8:17:12<5:04:57,  3.25s/it]
 62%|██████▏   | 9098/14725 [8:17:15<5:03:16,  3.23s/it]
 62%|██████▏   | 9099/14725 [8:17:19<5:03:44,  3.24s/it]
 62%|██████▏   | 9100/14725 [8:17:22<5:02:20,  3.23s/it]
                                                        

 62%|██████▏   | 9100/14725 [8:17:22
0: {'loss': 0.6823, 'grad_norm': 0.41094645157805915, 'learning_rate': 1e-05, 'epoch': 0.62}
0: <5:02:20,  3.23s/it]
 62%|██████▏   | 9101/14725 [8:17:25<4:59:53,  3.20s/it]
 62%|██████▏   | 9102/14725 [8:17:28<5:02:20,  3.23s/it]
 62%|██████▏   | 9103/14725 [8:17:32<5:07:42,  3.28s/it]
 62%|██████▏   | 9104/14725 [8:17:35<5:07:03,  3.28s/it]
 62%|██████▏   | 9105/14725 [8:17:39<5:16:35,  3.38s/it]
 62%|██████▏   | 9106/14725 [8:17:42<5:12:47,  3.34s/it]
 62%|██████▏   | 9107/14725 [8:17:45<5:12:34,  3.34s/it]
 62%|██████▏   | 9108/14725 [8:17:49<5:17:30,  3.39s/it]
 62%|██████▏   | 9109/14725 [8:17:52<5:15:57,  3.38s/it]
 62%|██████▏   | 9110/14725 [8:17:55<5:11:33,  3.33s/it]
                                                        

 62%|██████▏   | 9110/14725 [8:17:55<5:11:33,  3.33s/it]
 62%|██████▏   | 9111/14725 [8:17:59<5:07:46,  3.29s/it]
 62%|██████▏   | 9112/14725 [8:18:02<5:06:15,  3.27s/it]
 62%|█████�
0: {'loss': 0.6812, 'grad_norm': 0.42841968870337166, 'learning_rate': 1e-05, 'epoch': 0.62}
0: �▏   | 9113/14725 [8:18:05<5:06:00,  3.27s/it]
 62%|██████▏   | 9114/14725 [8:18:08<5:03:26,  3.24s/it]
 62%|██████▏   | 9115/14725 [8:18:11<5:02:52,  3.24s/it]
 62%|██████▏   | 9116/14725 [8:18:15<5:07:12,  3.29s/it]
 62%|██████▏   | 9117/14725 [8:18:18<5:00:01,  3.21s/it]
 62%|██████▏   | 9118/14725 [8:18:21<5:07:27,  3.29s/it]
 62%|██████▏   | 9119/14725 [8:18:25<5:13:38,  3.36s/it]
 62%|██████▏   | 9120/14725 [8:18:28<5:09:34,  3.31s/it]
                                                        

 62%|██████▏   | 9120/14725 [8:18:28<5:09:34,  3.31s/it]
 62%|██████▏   | 9121/14725 [8:18:31<5:04:41,  3.26s/it]
 62%|██████▏   | 9122/14725 [8:18:34<5:00:02,  3.21s/it]
 62%|██████▏   | 9123/14725 [8:18:38<5:02:15,  3.24s/it]
 62%|██████▏   | 9124/14725 [8:18:41<5:09:46,  3.32s/it]
 62%|██████▏   | 9125/14725 [8:18:44<5:08:56,  3.31
0: {'loss': 0.6983, 'grad_norm': 0.454149675234029, 'learning_rate': 1e-05, 'epoch': 0.62}
0: s/it]
 62%|██████▏   | 9126/14725 [8:18:48<5:03:28,  3.25s/it]
 62%|██████▏   | 9127/14725 [8:18:51<5:05:05,  3.27s/it]
 62%|██████▏   | 9128/14725 [8:18:54<5:03:28,  3.25s/it]
 62%|██████▏   | 9129/14725 [8:18:57<5:02:41,  3.25s/it]
 62%|██████▏   | 9130/14725 [8:19:01<5:03:28,  3.25s/it]
                                                        

 62%|██████▏   | 9130/14725 [8:19:01<5:03:28,  3.25s/it]
 62%|██████▏   | 9131/14725 [8:19:04<4:57:45,  3.19s/it]
 62%|██████▏   | 9132/14725 [8:19:07<4:58:58,  3.21s/it]
 62%|██████▏   | 9133/14725 [8:19:10<5:02:59,  3.25s/it]
 62%|██████▏   | 9134/14725 [8:19:13<4:58:42,  3.21s/it]
 62%|██████▏   | 9135/14725 [8:19:17<4:59:46,  3.22s/it]
 62%|██████▏   | 9136/14725 [8:19:20<4:57:57,  3.20s/it]
 62%|██████▏   | 9137/14725 [8:19:23<5:00:43,  3.23s/it]
 62%|██████▏   | 9138/1
0: {'loss': 0.6891, 'grad_norm': 0.44360386838895277, 'learning_rate': 1e-05, 'epoch': 0.62}
0: 4725 [8:19:26<5:04:37,  3.27s/it]
 62%|██████▏   | 9139/14725 [8:19:29<4:59:54,  3.22s/it]
 62%|██████▏   | 9140/14725 [8:19:33<5:05:17,  3.28s/it]
                                                        

 62%|██████▏   | 9140/14725 [8:19:33<5:05:17,  3.28s/it]
 62%|██████▏   | 9141/14725 [8:19:36<5:06:44,  3.30s/it]
 62%|██████▏   | 9142/14725 [8:19:39<5:06:17,  3.29s/it]
 62%|██████▏   | 9143/14725 [8:19:43<5:03:46,  3.27s/it]
 62%|██████▏   | 9144/14725 [8:19:46<5:08:31,  3.32s/it]
 62%|██████▏   | 9145/14725 [8:19:49<5:06:48,  3.30s/it]
 62%|██████▏   | 9146/14725 [8:19:53<5:04:57,  3.28s/it]
 62%|██████▏   | 9147/14725 [8:19:56<5:03:36,  3.27s/it]
 62%|██████▏   | 9148/14725 [8:19:59<5:05:12,  3.28s/it]
 62%|██████▏   | 9149/14725 [8:20:02<5:05:43,  3.29s/it]
 62%|██████▏   | 9150/14725 [8:20:06<5:02:19,  3.25s/it]
         
0: {'loss': 0.6866, 'grad_norm': 0.40643446170965525, 'learning_rate': 1e-05, 'epoch': 0.62}
0: {'loss': 0.6842, 'grad_norm': 0.4137559455258062, 'learning_rate': 1e-05, 'epoch': 0.62}
0:                                                

 62%|██████▏   | 9150/14725 [8:20:06<5:02:19,  3.25s/it]
 62%|██████▏   | 9151/14725 [8:20:09<5:00:12,  3.23s/it]
 62%|██████▏   | 9152/14725 [8:20:12<5:02:10,  3.25s/it]
 62%|██████▏   | 9153/14725 [8:20:15<5:02:51,  3.26s/it]
 62%|██████▏   | 9154/14725 [8:20:19<5:00:22,  3.24s/it]
 62%|██████▏   | 9155/14725 [8:20:22<4:57:18,  3.20s/it]
 62%|██████▏   | 9156/14725 [8:20:25<5:00:03,  3.23s/it]
 62%|██████▏   | 9157/14725 [8:20:28<4:57:49,  3.21s/it]
 62%|██████▏   | 9158/14725 [8:20:32<5:03:00,  3.27s/it]
 62%|██████▏   | 9159/14725 [8:20:35<5:05:00,  3.29s/it]
 62%|██████▏   | 9160/14725 [8:20:38<5:04:24,  3.28s/it]
                                                        

 62%|██████▏   | 9160/14725 [8:20:38<5:04:24,  3.28s/it]
 62%|██████▏   | 9161/14725 [8:20:42<5:09:48,  3.34
0: {'loss': 0.6792, 'grad_norm': 0.4007318168049191, 'learning_rate': 1e-05, 'epoch': 0.62}
0: s/it]
 62%|██████▏   | 9162/14725 [8:20:45<5:01:07,  3.25s/it]
 62%|██████▏   | 9163/14725 [8:20:48<4:59:15,  3.23s/it]
 62%|██████▏   | 9164/14725 [8:20:51<4:52:32,  3.16s/it]
 62%|██████▏   | 9165/14725 [8:20:54<4:54:24,  3.18s/it]
 62%|██████▏   | 9166/14725 [8:20:57<4:56:00,  3.19s/it]
 62%|██████▏   | 9167/14725 [8:21:01<4:58:27,  3.22s/it]
 62%|██████▏   | 9168/14725 [8:21:04<4:58:31,  3.22s/it]
 62%|██████▏   | 9169/14725 [8:21:07<4:57:45,  3.22s/it]
 62%|██████▏   | 9170/14725 [8:21:10<4:54:16,  3.18s/it]
                                                        

 62%|██████▏   | 9170/14725 [8:21:10<4:54:16,  3.18s/it]
 62%|██████▏   | 9171/14725 [8:21:13<4:58:59,  3.23s/it]
 62%|██████▏   | 9172/14725 [8:21:17<4:55:24,  3.19s/it]
 62%|██████▏   | 9173/14725 [8:21:20<4:56:52,  3.21s/it]
 62%|██████▏   | 9174/1
0: {'loss': 0.6951, 'grad_norm': 0.3922919351596811, 'learning_rate': 1e-05, 'epoch': 0.62}
0: 4725 [8:21:23<5:00:19,  3.25s/it]
 62%|██████▏   | 9175/14725 [8:21:27<5:03:53,  3.29s/it]
 62%|██████▏   | 9176/14725 [8:21:30<4:59:13,  3.24s/it]
 62%|██████▏   | 9177/14725 [8:21:33<4:57:37,  3.22s/it]
 62%|██████▏   | 9178/14725 [8:21:36<4:56:41,  3.21s/it]
 62%|██████▏   | 9179/14725 [8:21:39<5:02:27,  3.27s/it]
 62%|██████▏   | 9180/14725 [8:21:43<4:59:29,  3.24s/it]
                                                        

 62%|██████▏   | 9180/14725 [8:21:43<4:59:29,  3.24s/it]
 62%|██████▏   | 9181/14725 [8:21:46<4:59:15,  3.24s/it]
 62%|██████▏   | 9182/14725 [8:21:49<5:01:45,  3.27s/it]
 62%|██████▏   | 9183/14725 [8:21:52<4:58:28,  3.23s/it]
 62%|██████▏   | 9184/14725 [8:21:55<4:55:34,  3.20s/it]
 62%|██████▏   | 9185/14725 [8:21:59<4:57:19,  3.22s/it]
 62%|██████▏   | 9186/14725 [8:22:02<5:02:38,  3.28s/it]
 62%|█�
0: {'loss': 0.6768, 'grad_norm': 0.4092097872209957, 'learning_rate': 1e-05, 'epoch': 0.62}
0: ��████▏   | 9187/14725 [8:22:06<5:11:04,  3.37s/it]
 62%|██████▏   | 9188/14725 [8:22:09<5:06:39,  3.32s/it]
 62%|██████▏   | 9189/14725 [8:22:12<5:01:47,  3.27s/it]
 62%|██████▏   | 9190/14725 [8:22:15<4:59:22,  3.25s/it]
                                                        

 62%|██████▏   | 9190/14725 [8:22:15<4:59:22,  3.25s/it]
 62%|██████▏   | 9191/14725 [8:22:19<4:59:11,  3.24s/it]
 62%|██████▏   | 9192/14725 [8:22:22<4:58:55,  3.24s/it]
 62%|██████▏   | 9193/14725 [8:22:25<5:04:21,  3.30s/it]
 62%|██████▏   | 9194/14725 [8:22:29<5:06:03,  3.32s/it]
 62%|██████▏   | 9195/14725 [8:22:32<5:06:04,  3.32s/it]
 62%|██████▏   | 9196/14725 [8:22:35<5:03:35,  3.29s/it]
 62%|██████▏   | 9197/14725 [8:22:38<4:59:40,  3.25s/it]
 62%|██████▏   | 9198/14725 [8:22:42<5:03:04,  3.29s/it]
 62%|██████▏   | 9199/14725 [8:22:45<4
0: {'loss': 0.6925, 'grad_norm': 0.3993853329622353, 'learning_rate': 1e-05, 'epoch': 0.62}
0: {'loss': 0.6889, 'grad_norm': 0.3921405397864197, 'learning_rate': 1e-05, 'epoch': 0.63}
0: :57:46,  3.23s/it]
 62%|██████▏   | 9200/14725 [8:22:48<5:03:58,  3.30s/it]
                                                        

 62%|██████▏   | 9200/14725 [8:22:48<5:03:58,  3.30s/it]
 62%|██████▏   | 9201/14725 [8:22:51<5:03:40,  3.30s/it]
 62%|██████▏   | 9202/14725 [8:22:55<5:02:59,  3.29s/it]
 62%|██████▏   | 9203/14725 [8:22:58<5:01:37,  3.28s/it]
 63%|██████▎   | 9204/14725 [8:23:01<5:04:07,  3.31s/it]
 63%|██████▎   | 9205/14725 [8:23:05<5:01:28,  3.28s/it]
 63%|██████▎   | 9206/14725 [8:23:08<5:00:06,  3.26s/it]
 63%|██████▎   | 9207/14725 [8:23:11<4:57:13,  3.23s/it]
 63%|██████▎   | 9208/14725 [8:23:14<5:01:42,  3.28s/it]
 63%|██████▎   | 9209/14725 [8:23:18<4:59:06,  3.25s/it]
 63%|██████▎   | 9210/14725 [8:23:21<5:01:10,  3.28s/it]
                                                        

 63%|██████▎   | 9210/1
0: {'loss': 0.7024, 'grad_norm': 0.4372387278151583, 'learning_rate': 1e-05, 'epoch': 0.63}
0: 4725 [8:23:21<5:01:10,  3.28s/it]
 63%|██████▎   | 9211/14725 [8:23:24<5:00:18,  3.27s/it]
 63%|██████▎   | 9212/14725 [8:23:27<4:57:24,  3.24s/it]
 63%|██████▎   | 9213/14725 [8:23:31<4:59:21,  3.26s/it]
 63%|██████▎   | 9214/14725 [8:23:34<5:00:43,  3.27s/it]
 63%|██████▎   | 9215/14725 [8:23:37<5:05:08,  3.32s/it]
 63%|██████▎   | 9216/14725 [8:23:41<5:05:03,  3.32s/it]
 63%|██████▎   | 9217/14725 [8:23:44<5:07:59,  3.35s/it]
 63%|██████▎   | 9218/14725 [8:23:47<5:02:15,  3.29s/it]
 63%|██████▎   | 9219/14725 [8:23:51<5:01:18,  3.28s/it]
 63%|██████▎   | 9220/14725 [8:23:54<5:01:20,  3.28s/it]
                                                        

 63%|██████▎   | 9220/14725 [8:23:54<5:01:20,  3.28s/it]
 63%|██████▎   | 9221/14725 [8:23:57<4:58:10,  3.25s/it]
 63%|██████▎   | 9222/14725 [8:24:00<4:59:06,  3.26s/it]
 63%|█�
0: {'loss': 0.6962, 'grad_norm': 0.42308980326133533, 'learning_rate': 1e-05, 'epoch': 0.63}
0: ��████▎   | 9223/14725 [8:24:03<4:54:11,  3.21s/it]
 63%|██████▎   | 9224/14725 [8:24:07<4:56:08,  3.23s/it]
 63%|██████▎   | 9225/14725 [8:24:10<4:55:18,  3.22s/it]
 63%|██████▎   | 9226/14725 [8:24:13<4:58:52,  3.26s/it]
 63%|██████▎   | 9227/14725 [8:24:16<4:55:21,  3.22s/it]
 63%|██████▎   | 9228/14725 [8:24:20<4:57:18,  3.25s/it]
 63%|██████▎   | 9229/14725 [8:24:23<5:00:50,  3.28s/it]
 63%|██████▎   | 9230/14725 [8:24:26<5:00:40,  3.28s/it]
                                                        

 63%|██████▎   | 9230/14725 [8:24:26<5:00:40,  3.28s/it]
 63%|██████▎   | 9231/14725 [8:24:29<4:56:51,  3.24s/it]
 63%|██████▎   | 9232/14725 [8:24:32<4:51:34,  3.18s/it]
 63%|██████▎   | 9233/14725 [8:24:36<4:51:02,  3.18s/it]
 63%|██████▎   | 9234/14725 [8:24:39<4:53:09,  3.20s/it]
 63%|██████▎   | 9235/14725 [8:24:42<4
0: {'loss': 0.7079, 'grad_norm': 0.39703410437663494, 'learning_rate': 1e-05, 'epoch': 0.63}
0: :51:59,  3.19s/it]
 63%|██████▎   | 9236/14725 [8:24:45<4:58:07,  3.26s/it]
 63%|██████▎   | 9237/14725 [8:24:49<4:59:22,  3.27s/it]
 63%|██████▎   | 9238/14725 [8:24:52<5:03:51,  3.32s/it]
 63%|██████▎   | 9239/14725 [8:24:55<5:01:23,  3.30s/it]
 63%|██████▎   | 9240/14725 [8:24:59<4:57:28,  3.25s/it]
                                                        

 63%|██████▎   | 9240/14725 [8:24:59<4:57:28,  3.25s/it]
 63%|██████▎   | 9241/14725 [8:25:02<4:59:48,  3.28s/it]
 63%|██████▎   | 9242/14725 [8:25:05<5:06:23,  3.35s/it]
 63%|██████▎   | 9243/14725 [8:25:09<5:04:30,  3.33s/it]
 63%|██████▎   | 9244/14725 [8:25:12<5:05:58,  3.35s/it]
 63%|██████▎   | 9245/14725 [8:25:16<5:10:25,  3.40s/it]
 63%|██████▎   | 9246/14725 [8:25:19<5:08:06,  3.37s/it]
 63%|██████▎   | 9247/14725 [8:25:22<5:07:05,  3.36s/it]
 63%|██████�
0: {'loss': 0.6997, 'grad_norm': 0.47219806984907864, 'learning_rate': 1e-05, 'epoch': 0.63}
0: ��   | 9248/14725 [8:25:26<5:03:25,  3.32s/it]
 63%|██████▎   | 9249/14725 [8:25:29<4:56:09,  3.24s/it]
 63%|██████▎   | 9250/14725 [8:25:32<4:57:44,  3.26s/it]
                                                        

 63%|██████▎   | 9250/14725 [8:25:32<4:57:44,  3.26s/it]
 63%|██████▎   | 9251/14725 [8:25:35<4:57:53,  3.27s/it]
 63%|██████▎   | 9252/14725 [8:25:38<4:55:18,  3.24s/it]
 63%|██████▎   | 9253/14725 [8:25:42<4:57:12,  3.26s/it]
 63%|██████▎   | 9254/14725 [8:25:45<5:00:05,  3.29s/it]
 63%|██████▎   | 9255/14725 [8:25:48<5:02:59,  3.32s/it]
 63%|██████▎   | 9256/14725 [8:25:52<5:03:10,  3.33s/it]
 63%|██████▎   | 9257/14725 [8:25:55<5:00:49,  3.30s/it]
 63%|██████▎   | 9258/14725 [8:25:58<4:59:26,  3.29s/it]
 63%|██████▎   | 9259/14725 [8:26:01<4:56:10,  3.25s/it]
 63%|██████▎   | 9260/14725 [8:26:05<4:57:56,  3.27s/
0: {'loss': 0.6939, 'grad_norm': 0.44673812406909946, 'learning_rate': 1e-05, 'epoch': 0.63}
0: {'loss': 0.6914, 'grad_norm': 0.4057203387056052, 'learning_rate': 1e-05, 'epoch': 0.63}
0: it]
                                                        

 63%|██████▎   | 9260/14725 [8:26:05<4:57:56,  3.27s/it]
 63%|██████▎   | 9261/14725 [8:26:08<5:06:19,  3.36s/it]
 63%|██████▎   | 9262/14725 [8:26:12<5:07:37,  3.38s/it]
 63%|██████▎   | 9263/14725 [8:26:15<5:10:36,  3.41s/it]
 63%|██████▎   | 9264/14725 [8:26:19<5:07:39,  3.38s/it]
 63%|██████▎   | 9265/14725 [8:26:22<5:06:04,  3.36s/it]
 63%|██████▎   | 9266/14725 [8:26:25<5:05:04,  3.35s/it]
 63%|██████▎   | 9267/14725 [8:26:28<5:03:13,  3.33s/it]
 63%|██████▎   | 9268/14725 [8:26:32<5:06:22,  3.37s/it]
 63%|██████▎   | 9269/14725 [8:26:35<5:02:55,  3.33s/it]
 63%|██████▎   | 9270/14725 [8:26:38<4:55:47,  3.25s/it]
                                                        

 63%|██████▎   | 9270/14725 [8:26:38<4:55:47,  3.25s/it]
 63%|██████▎   | 9271/14725 [8:26:42<4
0: {'loss': 0.689, 'grad_norm': 0.41311251056058373, 'learning_rate': 1e-05, 'epoch': 0.63}
0: :57:25,  3.27s/it]
 63%|██████▎   | 9272/14725 [8:26:45<4:54:08,  3.24s/it]
 63%|██████▎   | 9273/14725 [8:26:48<4:51:26,  3.21s/it]
 63%|██████▎   | 9274/14725 [8:26:51<4:59:01,  3.29s/it]
 63%|██████▎   | 9275/14725 [8:26:55<5:00:26,  3.31s/it]
 63%|██████▎   | 9276/14725 [8:26:58<5:00:52,  3.31s/it]
 63%|██████▎   | 9277/14725 [8:27:01<4:57:02,  3.27s/it]
 63%|██████▎   | 9278/14725 [8:27:05<4:58:51,  3.29s/it]
 63%|██████▎   | 9279/14725 [8:27:08<4:55:15,  3.25s/it]
 63%|██████▎   | 9280/14725 [8:27:11<4:51:06,  3.21s/it]
                                                        

 63%|██████▎   | 9280/14725 [8:27:11<4:51:06,  3.21s/it]
 63%|██████▎   | 9281/14725 [8:27:14<4:53:35,  3.24s/it]
 63%|██████▎   | 9282/14725 [8:27:17<4:53:36,  3.24s/it]
 63%|██████▎   | 9283/14725 [8:27:21<4:56:32,  3.27s/it]
 63%|██████�
0: {'loss': 0.6816, 'grad_norm': 0.40013929818271465, 'learning_rate': 1e-05, 'epoch': 0.63}
0: ��   | 9284/14725 [8:27:24<4:53:38,  3.24s/it]
 63%|██████▎   | 9285/14725 [8:27:27<5:02:45,  3.34s/it]
 63%|██████▎   | 9286/14725 [8:27:31<5:01:10,  3.32s/it]
 63%|██████▎   | 9287/14725 [8:27:34<4:53:49,  3.24s/it]
 63%|██████▎   | 9288/14725 [8:27:37<4:51:23,  3.22s/it]
 63%|██████▎   | 9289/14725 [8:27:40<4:58:49,  3.30s/it]
 63%|██████▎   | 9290/14725 [8:27:44<4:58:32,  3.30s/it]
                                                        

 63%|██████▎   | 9290/14725 [8:27:44<4:58:32,  3.30s/it]
 63%|██████▎   | 9291/14725 [8:27:47<4:56:24,  3.27s/it]
 63%|██████▎   | 9292/14725 [8:27:50<4:55:56,  3.27s/it]
 63%|██████▎   | 9293/14725 [8:27:53<4:54:29,  3.25s/it]
 63%|██████▎   | 9294/14725 [8:27:57<4:58:23,  3.30s/it]
 63%|██████▎   | 9295/14725 [8:28:00<4:58:06,  3.29s/it]
 63%|██████▎   | 9296/14725 [8:28:03<4:56:53,  3.28s/
0: {'loss': 0.6895, 'grad_norm': 0.3935899047163626, 'learning_rate': 1e-05, 'epoch': 0.63}
0: it]
 63%|██████▎   | 9297/14725 [8:28:07<4:58:26,  3.30s/it]
 63%|██████▎   | 9298/14725 [8:28:10<4:52:54,  3.24s/it]
 63%|██████▎   | 9299/14725 [8:28:13<4:57:31,  3.29s/it]
 63%|██████▎   | 9300/14725 [8:28:16<4:52:36,  3.24s/it]
                                                        

 63%|██████▎   | 9300/14725 [8:28:16<4:52:36,  3.24s/it]
 63%|██████▎   | 9301/14725 [8:28:20<4:54:33,  3.26s/it]
 63%|██████▎   | 9302/14725 [8:28:23<4:50:40,  3.22s/it]
 63%|██████▎   | 9303/14725 [8:28:26<4:58:04,  3.30s/it]
 63%|██████▎   | 9304/14725 [8:28:29<4:54:31,  3.26s/it]
 63%|██████▎   | 9305/14725 [8:28:33<4:52:03,  3.23s/it]
 63%|██████▎   | 9306/14725 [8:28:36<4:54:53,  3.27s/it]
 63%|██████▎   | 9307/14725 [8:28:39<4:59:54,  3.32s/it]
 63%|██████▎   | 9308/14725 [8:28:43<4:57:04,  3.29s/it]
 63%|██████▎   | 9309/147
0: {'loss': 0.6827, 'grad_norm': 0.42536119293934543, 'learning_rate': 1e-05, 'epoch': 0.63}
0: {'loss': 0.7025, 'grad_norm': 0.4097429176960429, 'learning_rate': 1e-05, 'epoch': 0.63}
0: 25 [8:28:46<5:00:12,  3.33s/it]
 63%|██████▎   | 9310/14725 [8:28:49<4:56:14,  3.28s/it]
                                                        

 63%|██████▎   | 9310/14725 [8:28:49<4:56:14,  3.28s/it]
 63%|██████▎   | 9311/14725 [8:28:53<4:59:30,  3.32s/it]
 63%|██████▎   | 9312/14725 [8:28:56<4:53:07,  3.25s/it]
 63%|██████▎   | 9313/14725 [8:28:59<4:53:26,  3.25s/it]
 63%|██████▎   | 9314/14725 [8:29:02<4:51:55,  3.24s/it]
 63%|██████▎   | 9315/14725 [8:29:05<4:53:01,  3.25s/it]
 63%|██████▎   | 9316/14725 [8:29:09<4:59:10,  3.32s/it]
 63%|██████▎   | 9317/14725 [8:29:12<4:58:01,  3.31s/it]
 63%|██████▎   | 9318/14725 [8:29:16<5:02:12,  3.35s/it]
 63%|██████▎   | 9319/14725 [8:29:19<5:00:09,  3.33s/it]
 63%|██████▎   | 9320/14725 [8:29:22<4:59:36,  3.33s/it]
                                                        

 63%|██████�
0: {'loss': 0.6955, 'grad_norm': 0.4012786522915242, 'learning_rate': 1e-05, 'epoch': 0.63}
0: ��   | 9320/14725 [8:29:22<4:59:36,  3.33s/it]
 63%|██████▎   | 9321/14725 [8:29:26<5:00:28,  3.34s/it]
 63%|██████▎   | 9322/14725 [8:29:29<4:58:05,  3.31s/it]
 63%|██████▎   | 9323/14725 [8:29:32<4:52:28,  3.25s/it]
 63%|██████▎   | 9324/14725 [8:29:35<4:50:00,  3.22s/it]
 63%|██████▎   | 9325/14725 [8:29:38<4:49:01,  3.21s/it]
 63%|██████▎   | 9326/14725 [8:29:42<4:52:26,  3.25s/it]
 63%|██████▎   | 9327/14725 [8:29:45<4:52:54,  3.26s/it]
 63%|██████▎   | 9328/14725 [8:29:48<4:49:42,  3.22s/it]
 63%|██████▎   | 9329/14725 [8:29:51<4:51:19,  3.24s/it]
 63%|██████▎   | 9330/14725 [8:29:55<4:59:13,  3.33s/it]
                                                        

 63%|██████▎   | 9330/14725 [8:29:55<4:59:13,  3.33s/it]
 63%|██████▎   | 9331/14725 [8:29:58<4:55:18,  3.28s/it]
 63%|██████▎   | 9332/14725 [8:30:01<4:52:48,  3.26s/
0: {'loss': 0.6954, 'grad_norm': 0.44598099262237806, 'learning_rate': 1e-05, 'epoch': 0.63}
0: it]
 63%|██████▎   | 9333/14725 [8:30:04<4:47:31,  3.20s/it]
 63%|██████▎   | 9334/14725 [8:30:07<4:44:04,  3.16s/it]
 63%|██████▎   | 9335/14725 [8:30:11<4:44:56,  3.17s/it]
 63%|██████▎   | 9336/14725 [8:30:14<4:48:23,  3.21s/it]
 63%|██████▎   | 9337/14725 [8:30:17<4:53:41,  3.27s/it]
 63%|██████▎   | 9338/14725 [8:30:21<4:58:55,  3.33s/it]
 63%|██████▎   | 9339/14725 [8:30:24<4:58:38,  3.33s/it]
 63%|██████▎   | 9340/14725 [8:30:27<4:52:06,  3.25s/it]
                                                        

 63%|██████▎   | 9340/14725 [8:30:27<4:52:06,  3.25s/it]
 63%|██████▎   | 9341/14725 [8:30:30<4:50:57,  3.24s/it]
 63%|██████▎   | 9342/14725 [8:30:34<4:56:16,  3.30s/it]
 63%|██████▎   | 9343/14725 [8:30:37<4:56:01,  3.30s/it]
 63%|██████▎   | 9344/14725 [8:30:40<4:50:19,  3.24s/it]
 63%|██████▎   | 9345/147
0: {'loss': 0.6866, 'grad_norm': 0.4339171711261095, 'learning_rate': 1e-05, 'epoch': 0.63}
0: 25 [8:30:43<4:52:55,  3.27s/it]
 63%|██████▎   | 9346/14725 [8:30:47<4:55:05,  3.29s/it]
 63%|██████▎   | 9347/14725 [8:30:50<4:51:17,  3.25s/it]
 63%|██████▎   | 9348/14725 [8:30:53<4:55:10,  3.29s/it]
 63%|██████▎   | 9349/14725 [8:30:57<4:52:18,  3.26s/it]
 63%|██████▎   | 9350/14725 [8:31:00<4:57:20,  3.32s/it]
                                                        

 63%|██████▎   | 9350/14725 [8:31:00<4:57:20,  3.32s/it]
 64%|██████▎   | 9351/14725 [8:31:03<4:57:19,  3.32s/it]
 64%|██████▎   | 9352/14725 [8:31:07<4:52:52,  3.27s/it]
 64%|██████▎   | 9353/14725 [8:31:10<4:51:13,  3.25s/it]
 64%|██████▎   | 9354/14725 [8:31:13<4:52:46,  3.27s/it]
 64%|██████▎   | 9355/14725 [8:31:16<4:55:26,  3.30s/it]
 64%|██████▎   | 9356/14725 [8:31:20<4:54:44,  3.29s/it]
 64%|██████▎   | 9357/14725 [8:31:23<4:52:24,  3.27s/it]
 64%|██
0: {'loss': 0.6831, 'grad_norm': 0.42992686734079594, 'learning_rate': 1e-05, 'epoch': 0.64}
0: ████▎   | 9358/14725 [8:31:26<4:53:31,  3.28s/it]
 64%|██████▎   | 9359/14725 [8:31:30<4:55:59,  3.31s/it]
 64%|██████▎   | 9360/14725 [8:31:33<4:53:36,  3.28s/it]
                                                        

 64%|██████▎   | 9360/14725 [8:31:33<4:53:36,  3.28s/it]
 64%|██████▎   | 9361/14725 [8:31:36<4:52:01,  3.27s/it]
 64%|██████▎   | 9362/14725 [8:31:40<4:57:52,  3.33s/it]
 64%|██████▎   | 9363/14725 [8:31:43<4:55:51,  3.31s/it]
 64%|██████▎   | 9364/14725 [8:31:46<4:52:19,  3.27s/it]
 64%|██████▎   | 9365/14725 [8:31:49<4:56:15,  3.32s/it]
 64%|██████▎   | 9366/14725 [8:31:53<4:53:39,  3.29s/it]
 64%|██████▎   | 9367/14725 [8:31:56<4:56:48,  3.32s/it]
 64%|██████▎   | 9368/14725 [8:31:59<4:56:15,  3.32s/it]
 64%|██████▎   | 9369/14725 [8:32:03<5:03:22,  3.40s/it]
 64%|██████▎   | 9370/14725 [8:32:06<4:5
0: {'loss': 0.6999, 'grad_norm': 0.4453750541735555, 'learning_rate': 1e-05, 'epoch': 0.64}
0: {'loss': 0.6855, 'grad_norm': 0.43773892999371267, 'learning_rate': 1e-05, 'epoch': 0.64}
0: 5:43,  3.31s/it]
                                                        

 64%|██████▎   | 9370/14725 [8:32:06<4:55:43,  3.31s/it]
 64%|██████▎   | 9371/14725 [8:32:09<4:51:40,  3.27s/it]
 64%|██████▎   | 9372/14725 [8:32:12<4:49:28,  3.24s/it]
 64%|██████▎   | 9373/14725 [8:32:16<4:48:57,  3.24s/it]
 64%|██████▎   | 9374/14725 [8:32:19<4:54:54,  3.31s/it]
 64%|██████▎   | 9375/14725 [8:32:22<4:51:39,  3.27s/it]
 64%|██████▎   | 9376/14725 [8:32:25<4:48:24,  3.24s/it]
 64%|██████▎   | 9377/14725 [8:32:29<4:51:30,  3.27s/it]
 64%|██████▎   | 9378/14725 [8:32:32<4:53:10,  3.29s/it]
 64%|██████▎   | 9379/14725 [8:32:35<4:54:47,  3.31s/it]
 64%|██████▎   | 9380/14725 [8:32:39<4:55:06,  3.31s/it]
                                                        

 64%|██████▎   | 9380/14725 [8:32:39<4:55:06,  3.31s/it]
 64%|██████▎   | 9381/147
0: {'loss': 0.6893, 'grad_norm': 0.4224301183895458, 'learning_rate': 1e-05, 'epoch': 0.64}
0: 25 [8:32:42<4:55:46,  3.32s/it]
 64%|██████▎   | 9382/14725 [8:32:45<4:54:59,  3.31s/it]
 64%|██████▎   | 9383/14725 [8:32:49<4:56:05,  3.33s/it]
 64%|██████▎   | 9384/14725 [8:32:52<4:54:54,  3.31s/it]
 64%|██████▎   | 9385/14725 [8:32:55<4:51:37,  3.28s/it]
 64%|██████▎   | 9386/14725 [8:32:58<4:48:40,  3.24s/it]
 64%|██████▎   | 9387/14725 [8:33:02<4:53:23,  3.30s/it]
 64%|██████▍   | 9388/14725 [8:33:05<4:55:06,  3.32s/it]
 64%|██████▍   | 9389/14725 [8:33:08<4:51:21,  3.28s/it]
 64%|██████▍   | 9390/14725 [8:33:12<4:55:30,  3.32s/it]
                                                        

 64%|██████▍   | 9390/14725 [8:33:12<4:55:30,  3.32s/it]
 64%|██████▍   | 9391/14725 [8:33:15<4:54:10,  3.31s/it]
 64%|██████▍   | 9392/14725 [8:33:18<4:51:50,  3.28s/it]
 64%|██████▍   | 9393/14725 [8:33:22<4:50:35,  3.27s/it]
 64%|██
0: {'loss': 0.6668, 'grad_norm': 0.40749307167101995, 'learning_rate': 1e-05, 'epoch': 0.64}
0: ████▍   | 9394/14725 [8:33:25<4:48:29,  3.25s/it]
 64%|██████▍   | 9395/14725 [8:33:28<4:45:11,  3.21s/it]
 64%|██████▍   | 9396/14725 [8:33:31<4:51:28,  3.28s/it]
 64%|██████▍   | 9397/14725 [8:33:34<4:45:53,  3.22s/it]
 64%|██████▍   | 9398/14725 [8:33:38<4:46:54,  3.23s/it]
 64%|██████▍   | 9399/14725 [8:33:41<4:48:28,  3.25s/it]
 64%|██████▍   | 9400/14725 [8:33:44<4:52:24,  3.29s/it]
                                                        

 64%|██████▍   | 9400/14725 [8:33:44<4:52:24,  3.29s/it]
 64%|██████▍   | 9401/14725 [8:33:48<4:55:57,  3.34s/it]
 64%|██████▍   | 9402/14725 [8:33:51<4:54:11,  3.32s/it]
 64%|██████▍   | 9403/14725 [8:33:54<4:51:42,  3.29s/it]
 64%|██████▍   | 9404/14725 [8:33:58<4:52:55,  3.30s/it]
 64%|██████▍   | 9405/14725 [8:34:01<4:54:21,  3.32s/it]
 64%|██████▍   | 9406/14725 [8:34:04<4:5
0: {'loss': 0.6821, 'grad_norm': 0.3918261898112189, 'learning_rate': 1e-05, 'epoch': 0.64}
0: 2:56,  3.30s/it]
 64%|██████▍   | 9407/14725 [8:34:07<4:52:27,  3.30s/it]
 64%|██████▍   | 9408/14725 [8:34:11<4:50:59,  3.28s/it]
 64%|██████▍   | 9409/14725 [8:34:14<4:51:39,  3.29s/it]
 64%|██████▍   | 9410/14725 [8:34:17<4:47:43,  3.25s/it]
                                                        

 64%|██████▍   | 9410/14725 [8:34:17<4:47:43,  3.25s/it]
 64%|██████▍   | 9411/14725 [8:34:21<4:50:18,  3.28s/it]
 64%|██████▍   | 9412/14725 [8:34:24<4:48:30,  3.26s/it]
 64%|██████▍   | 9413/14725 [8:34:27<4:48:08,  3.25s/it]
 64%|██████▍   | 9414/14725 [8:34:30<4:49:05,  3.27s/it]
 64%|██████▍   | 9415/14725 [8:34:33<4:44:40,  3.22s/it]
 64%|██████▍   | 9416/14725 [8:34:37<4:42:31,  3.19s/it]
 64%|██████▍   | 9417/14725 [8:34:40<4:40:38,  3.17s/it]
 64%|██████▍   | 9418/14725 [8:34:43<4:41:50,  3.19s/it]
 64%|██████▍
0: {'loss': 0.6823, 'grad_norm': 0.40122394362582314, 'learning_rate': 1e-05, 'epoch': 0.64}
0: {'loss': 0.6842, 'grad_norm': 0.45260801296647846, 'learning_rate': 1e-05, 'epoch': 0.64}
0:    | 9419/14725 [8:34:46<4:42:31,  3.19s/it]
 64%|██████▍   | 9420/14725 [8:34:49<4:42:08,  3.19s/it]
                                                        

 64%|██████▍   | 9420/14725 [8:34:49<4:42:08,  3.19s/it]
 64%|██████▍   | 9421/14725 [8:34:53<4:50:16,  3.28s/it]
 64%|██████▍   | 9422/14725 [8:34:56<4:45:41,  3.23s/it]
 64%|██████▍   | 9423/14725 [8:34:59<4:43:44,  3.21s/it]
 64%|██████▍   | 9424/14725 [8:35:02<4:46:35,  3.24s/it]
 64%|██████▍   | 9425/14725 [8:35:05<4:41:18,  3.18s/it]
 64%|██████▍   | 9426/14725 [8:35:09<4:44:10,  3.22s/it]
 64%|██████▍   | 9427/14725 [8:35:12<4:45:48,  3.24s/it]
 64%|██████▍   | 9428/14725 [8:35:15<4:44:08,  3.22s/it]
 64%|██████▍   | 9429/14725 [8:35:19<4:48:14,  3.27s/it]
 64%|██████▍   | 9430/14725 [8:35:22<4:52:07,  3.31s/it]
                                                        

 64%|██
0: {'loss': 0.682, 'grad_norm': 0.4145329109973415, 'learning_rate': 1e-05, 'epoch': 0.64}
0: ████▍   | 9430/14725 [8:35:22<4:52:07,  3.31s/it]
 64%|██████▍   | 9431/14725 [8:35:25<4:47:10,  3.25s/it]
 64%|██████▍   | 9432/14725 [8:35:28<4:47:12,  3.26s/it]
 64%|██████▍   | 9433/14725 [8:35:32<4:49:37,  3.28s/it]
 64%|██████▍   | 9434/14725 [8:35:35<4:44:36,  3.23s/it]
 64%|██████▍   | 9435/14725 [8:35:38<4:43:55,  3.22s/it]
 64%|██████▍   | 9436/14725 [8:35:41<4:44:02,  3.22s/it]
 64%|██████▍   | 9437/14725 [8:35:45<4:50:42,  3.30s/it]
 64%|██████▍   | 9438/14725 [8:35:48<4:52:55,  3.32s/it]
 64%|██████▍   | 9439/14725 [8:35:51<4:52:36,  3.32s/it]
 64%|██████▍   | 9440/14725 [8:35:55<4:49:23,  3.29s/it]
                                                        

 64%|██████▍   | 9440/14725 [8:35:55<4:49:23,  3.29s/it]
 64%|██████▍   | 9441/14725 [8:35:58<4:48:51,  3.28s/it]
 64%|██████▍   | 9442/14725 [8:36:01<4:4
0: {'loss': 0.6889, 'grad_norm': 0.4210677889017217, 'learning_rate': 1e-05, 'epoch': 0.64}
0: 7:36,  3.27s/it]
 64%|██████▍   | 9443/14725 [8:36:05<4:53:55,  3.34s/it]
 64%|██████▍   | 9444/14725 [8:36:08<4:56:14,  3.37s/it]
 64%|██████▍   | 9445/14725 [8:36:11<4:54:44,  3.35s/it]
 64%|██████▍   | 9446/14725 [8:36:14<4:49:39,  3.29s/it]
 64%|██████▍   | 9447/14725 [8:36:18<4:52:38,  3.33s/it]
 64%|██████▍   | 9448/14725 [8:36:21<4:49:07,  3.29s/it]
 64%|██████▍   | 9449/14725 [8:36:24<4:49:18,  3.29s/it]
 64%|██████▍   | 9450/14725 [8:36:28<4:52:31,  3.33s/it]
                                                        

 64%|██████▍   | 9450/14725 [8:36:28<4:52:31,  3.33s/it]
 64%|██████▍   | 9451/14725 [8:36:31<4:50:48,  3.31s/it]
 64%|██████▍   | 9452/14725 [8:36:34<4:49:42,  3.30s/it]
 64%|██████▍   | 9453/14725 [8:36:37<4:44:53,  3.24s/it]
 64%|██████▍   | 9454/14725 [8:36:41<4:44:21,  3.24s/it]
 64%|██████▍
0: {'loss': 0.684, 'grad_norm': 0.4298596896396833, 'learning_rate': 1e-05, 'epoch': 0.64}
0:    | 9455/14725 [8:36:44<4:48:18,  3.28s/it]
 64%|██████▍   | 9456/14725 [8:36:47<4:49:34,  3.30s/it]
 64%|██████▍   | 9457/14725 [8:36:51<4:47:38,  3.28s/it]
 64%|██████▍   | 9458/14725 [8:36:54<4:50:10,  3.31s/it]
 64%|██████▍   | 9459/14725 [8:36:57<4:44:41,  3.24s/it]
 64%|██████▍   | 9460/14725 [8:37:00<4:47:39,  3.28s/it]
                                                        

 64%|██████▍   | 9460/14725 [8:37:00<4:47:39,  3.28s/it]
 64%|██████▍   | 9461/14725 [8:37:04<4:49:59,  3.31s/it]
 64%|██████▍   | 9462/14725 [8:37:07<4:53:12,  3.34s/it]
 64%|██████▍   | 9463/14725 [8:37:10<4:50:06,  3.31s/it]
 64%|██████▍   | 9464/14725 [8:37:14<4:48:28,  3.29s/it]
 64%|██████▍   | 9465/14725 [8:37:17<4:49:33,  3.30s/it]
 64%|██████▍   | 9466/14725 [8:37:21<4:52:58,  3.34s/it]
 64%|██████▍   | 9467/14725 [8:37:24<4:55:46,  3.38s/it
0: {'loss': 0.6817, 'grad_norm': 0.390235002926666, 'learning_rate': 1e-05, 'epoch': 0.64}
0: ]
 64%|██████▍   | 9468/14725 [8:37:28<5:01:32,  3.44s/it]
 64%|██████▍   | 9469/14725 [8:37:31<4:57:57,  3.40s/it]
 64%|██████▍   | 9470/14725 [8:37:34<4:49:21,  3.30s/it]
                                                        

 64%|██████▍   | 9470/14725 [8:37:34<4:49:21,  3.30s/it]
 64%|██████▍   | 9471/14725 [8:37:37<4:49:44,  3.31s/it]
 64%|██████▍   | 9472/14725 [8:37:41<4:50:37,  3.32s/it]
 64%|██████▍   | 9473/14725 [8:37:44<4:49:36,  3.31s/it]
 64%|██████▍   | 9474/14725 [8:37:47<4:42:07,  3.22s/it]
 64%|██████▍   | 9475/14725 [8:37:50<4:42:52,  3.23s/it]
 64%|██████▍   | 9476/14725 [8:37:53<4:45:09,  3.26s/it]
 64%|██████▍   | 9477/14725 [8:37:57<4:45:24,  3.26s/it]
 64%|██████▍   | 9478/14725 [8:38:00<4:44:02,  3.25s/it]
 64%|██████▍   | 9479/14725 [8:38:03<4:42:04,  3.23s/it]
 64%|██████▍   | 9480/14725
0: {'loss': 0.6873, 'grad_norm': 0.41409633133725937, 'learning_rate': 1e-05, 'epoch': 0.64}
0: {'loss': 0.6858, 'grad_norm': 0.42425497270936213, 'learning_rate': 1e-05, 'epoch': 0.64}
0:  [8:38:06<4:44:01,  3.25s/it]
                                                        

 64%|██████▍   | 9480/14725 [8:38:06<4:44:01,  3.25s/it]
 64%|██████▍   | 9481/14725 [8:38:10<4:39:40,  3.20s/it]
 64%|██████▍   | 9482/14725 [8:38:13<4:39:20,  3.20s/it]
 64%|██████▍   | 9483/14725 [8:38:16<4:39:51,  3.20s/it]
 64%|██████▍   | 9484/14725 [8:38:19<4:41:15,  3.22s/it]
 64%|██████▍   | 9485/14725 [8:38:23<4:47:36,  3.29s/it]
 64%|██████▍   | 9486/14725 [8:38:26<4:43:57,  3.25s/it]
 64%|██████▍   | 9487/14725 [8:38:29<4:42:44,  3.24s/it]
 64%|██████▍   | 9488/14725 [8:38:32<4:44:08,  3.26s/it]
 64%|██████▍   | 9489/14725 [8:38:36<4:48:15,  3.30s/it]
 64%|██████▍   | 9490/14725 [8:38:39<4:46:32,  3.28s/it]
                                                        

 64%|██████▍   | 9490/14725 [8:38:39<4:46:32,  3.28s/it]
 64%|██████▍
0: {'loss': 0.7039, 'grad_norm': 0.4002553806384975, 'learning_rate': 1e-05, 'epoch': 0.65}
0:    | 9491/14725 [8:38:42<4:46:17,  3.28s/it]
 64%|██████▍   | 9492/14725 [8:38:46<4:50:46,  3.33s/it]
 64%|██████▍   | 9493/14725 [8:38:49<4:43:30,  3.25s/it]
 64%|██████▍   | 9494/14725 [8:38:52<4:44:22,  3.26s/it]
 64%|██████▍   | 9495/14725 [8:38:56<4:55:04,  3.39s/it]
 64%|██████▍   | 9496/14725 [8:38:59<4:48:52,  3.31s/it]
 64%|██████▍   | 9497/14725 [8:39:02<4:45:57,  3.28s/it]
 65%|██████▍   | 9498/14725 [8:39:05<4:45:37,  3.28s/it]
 65%|██████▍   | 9499/14725 [8:39:09<4:42:12,  3.24s/it]
 65%|██████▍   | 9500/14725 [8:39:12<4:44:37,  3.27s/it]
                                                        

 65%|██████▍   | 9500/14725 [8:39:12<4:44:37,  3.27s/it]
 65%|██████▍   | 9501/14725 [8:39:15<4:44:35,  3.27s/it]
 65%|██████▍   | 9502/14725 [8:39:18<4:39:37,  3.21s/it]
 65%|██████▍   | 9503/14725 [8:39:22<4:43:50,  3.26s/it
0: {'loss': 0.6988, 'grad_norm': 0.4174962837250178, 'learning_rate': 1e-05, 'epoch': 0.65}
0: ]
 65%|██████▍   | 9504/14725 [8:39:25<4:41:22,  3.23s/it]
 65%|██████▍   | 9505/14725 [8:39:28<4:41:41,  3.24s/it]
 65%|██████▍   | 9506/14725 [8:39:31<4:38:49,  3.21s/it]
 65%|██████▍   | 9507/14725 [8:39:34<4:40:23,  3.22s/it]
 65%|██████▍   | 9508/14725 [8:39:38<4:39:12,  3.21s/it]
 65%|██████▍   | 9509/14725 [8:39:41<4:41:18,  3.24s/it]
 65%|██████▍   | 9510/14725 [8:39:44<4:42:11,  3.25s/it]
                                                        

 65%|██████▍   | 9510/14725 [8:39:44<4:42:11,  3.25s/it]
 65%|██████▍   | 9511/14725 [8:39:47<4:39:58,  3.22s/it]
 65%|██████▍   | 9512/14725 [8:39:50<4:39:32,  3.22s/it]
 65%|██████▍   | 9513/14725 [8:39:54<4:40:51,  3.23s/it]
 65%|██████▍   | 9514/14725 [8:39:57<4:37:47,  3.20s/it]
 65%|██████▍   | 9515/14725 [8:40:00<4:46:01,  3.29s/it]
 65%|██████▍   | 9516/14725
0: {'loss': 0.6943, 'grad_norm': 0.4273358747860556, 'learning_rate': 1e-05, 'epoch': 0.65}
0:  [8:40:04<4:42:27,  3.25s/it]
 65%|██████▍   | 9517/14725 [8:40:07<4:38:56,  3.21s/it]
 65%|██████▍   | 9518/14725 [8:40:10<4:37:36,  3.20s/it]
 65%|██████▍   | 9519/14725 [8:40:13<4:34:18,  3.16s/it]
 65%|██████▍   | 9520/14725 [8:40:16<4:34:31,  3.16s/it]
                                                        

 65%|██████▍   | 9520/14725 [8:40:16<4:34:31,  3.16s/it]
 65%|██████▍   | 9521/14725 [8:40:19<4:33:17,  3.15s/it]
 65%|██████▍   | 9522/14725 [8:40:22<4:32:15,  3.14s/it]
 65%|██████▍   | 9523/14725 [8:40:26<4:33:21,  3.15s/it]
 65%|██████▍   | 9524/14725 [8:40:29<4:38:33,  3.21s/it]
 65%|██████▍   | 9525/14725 [8:40:32<4:44:37,  3.28s/it]
 65%|██████▍   | 9526/14725 [8:40:35<4:41:07,  3.24s/it]
 65%|██████▍   | 9527/14725 [8:40:39<4:42:29,  3.26s/it]
 65%|██████▍   | 9528/14725 [8:40:42<4:42:07,  3.26s/it]
 65%|██�
0: {'loss': 0.693, 'grad_norm': 0.4193176120913559, 'learning_rate': 1e-05, 'epoch': 0.65}
0: �███▍   | 9529/14725 [8:40:45<4:39:39,  3.23s/it]
 65%|██████▍   | 9530/14725 [8:40:48<4:38:37,  3.22s/it]
                                                        

 65%|██████▍   | 9530/14725 [8:40:48<4:38:37,  3.22s/it]
 65%|██████▍   | 9531/14725 [8:40:52<4:37:31,  3.21s/it]
 65%|██████▍   | 9532/14725 [8:40:55<4:37:10,  3.20s/it]
 65%|██████▍   | 9533/14725 [8:40:58<4:38:21,  3.22s/it]
 65%|██████▍   | 9534/14725 [8:41:01<4:42:25,  3.26s/it]
 65%|██████▍   | 9535/14725 [8:41:05<4:43:34,  3.28s/it]
 65%|██████▍   | 9536/14725 [8:41:08<4:47:54,  3.33s/it]
 65%|██████▍   | 9537/14725 [8:41:12<4:49:08,  3.34s/it]
 65%|██████▍   | 9538/14725 [8:41:15<4:45:30,  3.30s/it]
 65%|██████▍   | 9539/14725 [8:41:18<4:46:29,  3.31s/it]
 65%|██████▍   | 9540/14725 [8:41:21<4:47:31,  3.33s/it]
                                                        
0: {'loss': 0.6732, 'grad_norm': 0.4210444584711267, 'learning_rate': 1e-05, 'epoch': 0.65}
0: {'loss': 0.6958, 'grad_norm': 0.3784006832399807, 'learning_rate': 1e-05, 'epoch': 0.65}
0: 

 65%|██████▍   | 9540/14725 [8:41:21<4:47:31,  3.33s/it]
 65%|██████▍   | 9541/14725 [8:41:25<4:43:50,  3.29s/it]
 65%|██████▍   | 9542/14725 [8:41:28<4:43:51,  3.29s/it]
 65%|██████▍   | 9543/14725 [8:41:31<4:43:08,  3.28s/it]
 65%|██████▍   | 9544/14725 [8:41:34<4:35:56,  3.20s/it]
 65%|██████▍   | 9545/14725 [8:41:37<4:35:40,  3.19s/it]
 65%|██████▍   | 9546/14725 [8:41:40<4:32:57,  3.16s/it]
 65%|██████▍   | 9547/14725 [8:41:44<4:34:28,  3.18s/it]
 65%|██████▍   | 9548/14725 [8:41:47<4:37:29,  3.22s/it]
 65%|██████▍   | 9549/14725 [8:41:50<4:44:05,  3.29s/it]
 65%|██████▍   | 9550/14725 [8:41:54<4:42:08,  3.27s/it]
                                                        

 65%|██████▍   | 9550/14725 [8:41:54<4:42:08,  3.27s/it]
 65%|██████▍   | 9551/14725 [8:41:57<4:44:18,  3.30s/it]
 65%|██████▍   | 9552/14725
0: {'loss': 0.6885, 'grad_norm': 0.4113993082678648, 'learning_rate': 1e-05, 'epoch': 0.65}
0:  [8:42:00<4:39:35,  3.24s/it]
 65%|██████▍   | 9553/14725 [8:42:04<4:43:25,  3.29s/it]
 65%|██████▍   | 9554/14725 [8:42:07<4:52:31,  3.39s/it]
 65%|██████▍   | 9555/14725 [8:42:11<4:51:34,  3.38s/it]
 65%|██████▍   | 9556/14725 [8:42:14<4:50:17,  3.37s/it]
 65%|██████▍   | 9557/14725 [8:42:17<4:48:50,  3.35s/it]
 65%|██████▍   | 9558/14725 [8:42:20<4:48:10,  3.35s/it]
 65%|██████▍   | 9559/14725 [8:42:24<4:47:11,  3.34s/it]
 65%|██████▍   | 9560/14725 [8:42:27<4:42:22,  3.28s/it]
                                                        

 65%|██████▍   | 9560/14725 [8:42:27<4:42:22,  3.28s/it]
 65%|██████▍   | 9561/14725 [8:42:30<4:41:46,  3.27s/it]
 65%|██████▍   | 9562/14725 [8:42:33<4:38:12,  3.23s/it]
 65%|██████▍   | 9563/14725 [8:42:37<4:40:59,  3.27s/it]
 65%|██████▍   | 9564/14725 [8:42:40<4:35:58,  3.21s/it]
 65%|██�
0: {'loss': 0.6827, 'grad_norm': 0.4256449761189027, 'learning_rate': 1e-05, 'epoch': 0.65}
0: �███▍   | 9565/14725 [8:42:43<4:35:45,  3.21s/it]
 65%|██████▍   | 9566/14725 [8:42:46<4:42:41,  3.29s/it]
 65%|██████▍   | 9567/14725 [8:42:50<4:43:06,  3.29s/it]
 65%|██████▍   | 9568/14725 [8:42:53<4:38:06,  3.24s/it]
 65%|██████▍   | 9569/14725 [8:42:56<4:36:36,  3.22s/it]
 65%|██████▍   | 9570/14725 [8:42:59<4:36:51,  3.22s/it]
                                                        

 65%|██████▍   | 9570/14725 [8:42:59<4:36:51,  3.22s/it]
 65%|██████▍   | 9571/14725 [8:43:03<4:38:34,  3.24s/it]
 65%|██████▌   | 9572/14725 [8:43:06<4:39:24,  3.25s/it]
 65%|██████▌   | 9573/14725 [8:43:09<4:40:04,  3.26s/it]
 65%|██████▌   | 9574/14725 [8:43:12<4:34:54,  3.20s/it]
 65%|██████▌   | 9575/14725 [8:43:16<4:40:40,  3.27s/it]
 65%|██████▌   | 9576/14725 [8:43:19<4:38:22,  3.24s/it]
 65%|██████▌   | 9577/14725 [8:43:22<4:41:
0: {'loss': 0.6844, 'grad_norm': 0.4323663198890186, 'learning_rate': 1e-05, 'epoch': 0.65}
0: 53,  3.29s/it]
 65%|██████▌   | 9578/14725 [8:43:26<4:46:18,  3.34s/it]
 65%|██████▌   | 9579/14725 [8:43:29<4:41:43,  3.28s/it]
 65%|██████▌   | 9580/14725 [8:43:32<4:39:41,  3.26s/it]
                                                        

 65%|██████▌   | 9580/14725 [8:43:32<4:39:41,  3.26s/it]
 65%|██████▌   | 9581/14725 [8:43:35<4:42:05,  3.29s/it]
 65%|██████▌   | 9582/14725 [8:43:39<4:39:25,  3.26s/it]
 65%|██████▌   | 9583/14725 [8:43:42<4:39:54,  3.27s/it]
 65%|██████▌   | 9584/14725 [8:43:45<4:37:41,  3.24s/it]
 65%|██████▌   | 9585/14725 [8:43:48<4:39:20,  3.26s/it]
 65%|██████▌   | 9586/14725 [8:43:52<4:39:31,  3.26s/it]
 65%|██████▌   | 9587/14725 [8:43:55<4:39:50,  3.27s/it]
 65%|██████▌   | 9588/14725 [8:43:58<4:42:35,  3.30s/it]
 65%|██████▌   | 9589/14725 [8:44:01<4:39:32,  3.27s/it]
 65%|██████▌  
0: {'loss': 0.6775, 'grad_norm': 0.3991910808168848, 'learning_rate': 1e-05, 'epoch': 0.65}
0: {'loss': 0.6773, 'grad_norm': 0.4379211379819776, 'learning_rate': 1e-05, 'epoch': 0.65}
0:  | 9590/14725 [8:44:05<4:36:42,  3.23s/it]
                                                        

 65%|██████▌   | 9590/14725 [8:44:05<4:36:42,  3.23s/it]
 65%|██████▌   | 9591/14725 [8:44:08<4:37:59,  3.25s/it]
 65%|██████▌   | 9592/14725 [8:44:11<4:37:12,  3.24s/it]
 65%|██████▌   | 9593/14725 [8:44:15<4:43:47,  3.32s/it]
 65%|██████▌   | 9594/14725 [8:44:18<4:36:46,  3.24s/it]
 65%|██████▌   | 9595/14725 [8:44:21<4:43:35,  3.32s/it]
 65%|██████▌   | 9596/14725 [8:44:24<4:37:41,  3.25s/it]
 65%|██████▌   | 9597/14725 [8:44:27<4:37:09,  3.24s/it]
 65%|██████▌   | 9598/14725 [8:44:31<4:36:34,  3.24s/it]
 65%|██████▌   | 9599/14725 [8:44:34<4:33:20,  3.20s/it]
 65%|██████▌   | 9600/14725 [8:44:37<4:31:58,  3.18s/it]
                                                        

 65%|██████▌   | 9600/14725 [8:44:37<4:31:58,  3.18s/it]
 65%|██�
0: {'loss': 0.6792, 'grad_norm': 0.4117379809651676, 'learning_rate': 1e-05, 'epoch': 0.65}
0: �███▌   | 9601/14725 [8:44:40<4:35:32,  3.23s/it]
 65%|██████▌   | 9602/14725 [8:44:43<4:32:25,  3.19s/it]
 65%|██████▌   | 9603/14725 [8:44:46<4:29:34,  3.16s/it]
 65%|██████▌   | 9604/14725 [8:44:50<4:37:02,  3.25s/it]
 65%|██████▌   | 9605/14725 [8:44:53<4:31:12,  3.18s/it]
 65%|██████▌   | 9606/14725 [8:44:56<4:35:37,  3.23s/it]
 65%|██████▌   | 9607/14725 [8:44:59<4:31:27,  3.18s/it]
 65%|██████▌   | 9608/14725 [8:45:03<4:35:48,  3.23s/it]
 65%|██████▌   | 9609/14725 [8:45:06<4:31:27,  3.18s/it]
 65%|██████▌   | 9610/14725 [8:45:09<4:30:41,  3.18s/it]
                                                        

 65%|██████▌   | 9610/14725 [8:45:09<4:30:41,  3.18s/it]
 65%|██████▌   | 9611/14725 [8:45:12<4:32:30,  3.20s/it]
 65%|██████▌   | 9612/14725 [8:45:15<4:35:47,  3.24s/it]
 65%|██████▌   | 9613/14725 [8:45:19<4:38:
0: {'loss': 0.6942, 'grad_norm': 0.43051986198119785, 'learning_rate': 1e-05, 'epoch': 0.65}
0: 15,  3.27s/it]
 65%|██████▌   | 9614/14725 [8:45:22<4:40:39,  3.29s/it]
 65%|██████▌   | 9615/14725 [8:45:26<4:43:50,  3.33s/it]
 65%|██████▌   | 9616/14725 [8:45:29<4:36:22,  3.25s/it]
 65%|██████▌   | 9617/14725 [8:45:32<4:34:19,  3.22s/it]
 65%|██████▌   | 9618/14725 [8:45:35<4:38:29,  3.27s/it]
 65%|██████▌   | 9619/14725 [8:45:39<4:39:34,  3.29s/it]
 65%|██████▌   | 9620/14725 [8:45:42<4:39:47,  3.29s/it]
                                                        

 65%|██████▌   | 9620/14725 [8:45:42<4:39:47,  3.29s/it]
 65%|██████▌   | 9621/14725 [8:45:45<4:41:35,  3.31s/it]
 65%|██████▌   | 9622/14725 [8:45:49<4:41:54,  3.31s/it]
 65%|██████▌   | 9623/14725 [8:45:52<4:36:35,  3.25s/it]
 65%|██████▌   | 9624/14725 [8:45:55<4:35:09,  3.24s/it]
 65%|██████▌   | 9625/14725 [8:45:58<4:35:57,  3.25s/it]
 65%|██████▌  
0: {'loss': 0.6677, 'grad_norm': 0.3922150346824666, 'learning_rate': 1e-05, 'epoch': 0.65}
0:  | 9626/14725 [8:46:01<4:34:38,  3.23s/it]
 65%|██████▌   | 9627/14725 [8:46:04<4:32:31,  3.21s/it]
 65%|██████▌   | 9628/14725 [8:46:08<4:34:04,  3.23s/it]
 65%|██████▌   | 9629/14725 [8:46:11<4:35:39,  3.25s/it]
 65%|██████▌   | 9630/14725 [8:46:14<4:34:21,  3.23s/it]
                                                        

 65%|██████▌   | 9630/14725 [8:46:14<4:34:21,  3.23s/it]
 65%|██████▌   | 9631/14725 [8:46:18<4:37:19,  3.27s/it]
 65%|██████▌   | 9632/14725 [8:46:21<4:37:19,  3.27s/it]
 65%|██████▌   | 9633/14725 [8:46:24<4:33:47,  3.23s/it]
 65%|██████▌   | 9634/14725 [8:46:27<4:31:59,  3.21s/it]
 65%|██████▌   | 9635/14725 [8:46:30<4:29:32,  3.18s/it]
 65%|██████▌   | 9636/14725 [8:46:34<4:35:38,  3.25s/it]
 65%|██████▌   | 9637/14725 [8:46:37<4:38:45,  3.29s/it]
 65%|██████▌   | 9638/14725 [8:46:40<4:39:05,  3.29s/it]
0: {'loss': 0.6896, 'grad_norm': 0.4033326676126749, 'learning_rate': 1e-05, 'epoch': 0.65}
0:  65%|██████▌   | 9639/14725 [8:46:44<4:39:13,  3.29s/it]
 65%|██████▌   | 9640/14725 [8:46:47<4:37:01,  3.27s/it]
                                                        

 65%|██████▌   | 9640/14725 [8:46:47<4:37:01,  3.27s/it]
 65%|██████▌   | 9641/14725 [8:46:50<4:37:06,  3.27s/it]
 65%|██████▌   | 9642/14725 [8:46:54<4:44:01,  3.35s/it]
 65%|██████▌   | 9643/14725 [8:46:57<4:48:46,  3.41s/it]
 65%|██████▌   | 9644/14725 [8:47:00<4:42:59,  3.34s/it]
 66%|██████▌   | 9645/14725 [8:47:03<4:36:07,  3.26s/it]
 66%|██████▌   | 9646/14725 [8:47:07<4:35:44,  3.26s/it]
 66%|██████▌   | 9647/14725 [8:47:10<4:35:23,  3.25s/it]
 66%|██████▌   | 9648/14725 [8:47:13<4:32:14,  3.22s/it]
 66%|██████▌   | 9649/14725 [8:47:16<4:31:29,  3.21s/it]
 66%|██████▌   | 9650/14725 [8:47:20<4:32:49,  3.23s/it]
                                           
0: {'loss': 0.6735, 'grad_norm': 0.40350674549353227, 'learning_rate': 1e-05, 'epoch': 0.66}
0: {'loss': 0.6722, 'grad_norm': 0.41758041611837343, 'learning_rate': 1e-05, 'epoch': 0.66}
0:              

 66%|██████▌   | 9650/14725 [8:47:20<4:32:49,  3.23s/it]
 66%|██████▌   | 9651/14725 [8:47:23<4:35:08,  3.25s/it]
 66%|██████▌   | 9652/14725 [8:47:26<4:34:21,  3.24s/it]
 66%|██████▌   | 9653/14725 [8:47:29<4:32:38,  3.23s/it]
 66%|██████▌   | 9654/14725 [8:47:32<4:30:30,  3.20s/it]
 66%|██████▌   | 9655/14725 [8:47:36<4:34:33,  3.25s/it]
 66%|██████▌   | 9656/14725 [8:47:39<4:40:31,  3.32s/it]
 66%|██████▌   | 9657/14725 [8:47:42<4:38:25,  3.30s/it]
 66%|██████▌   | 9658/14725 [8:47:46<4:38:51,  3.30s/it]
 66%|██████▌   | 9659/14725 [8:47:49<4:35:23,  3.26s/it]
 66%|██████▌   | 9660/14725 [8:47:52<4:35:50,  3.27s/it]
                                                        

 66%|██████▌   | 9660/14725 [8:47:52<4:35:50,  3.27s/it]
 66%|██████▌   | 9661/14725 [8:47:55<4:34:08,  3.25s/it]
 66%|██████▌  
0: {'loss': 0.6896, 'grad_norm': 0.4318487115091939, 'learning_rate': 1e-05, 'epoch': 0.66}
0:  | 9662/14725 [8:47:59<4:33:19,  3.24s/it]
 66%|██████▌   | 9663/14725 [8:48:02<4:33:54,  3.25s/it]
 66%|██████▌   | 9664/14725 [8:48:05<4:33:20,  3.24s/it]
 66%|██████▌   | 9665/14725 [8:48:09<4:39:18,  3.31s/it]
 66%|██████▌   | 9666/14725 [8:48:12<4:43:10,  3.36s/it]
 66%|██████▌   | 9667/14725 [8:48:15<4:40:06,  3.32s/it]
 66%|██████▌   | 9668/14725 [8:48:19<4:37:53,  3.30s/it]
 66%|██████▌   | 9669/14725 [8:48:22<4:49:20,  3.43s/it]
 66%|██████▌   | 9670/14725 [8:48:26<4:44:02,  3.37s/it]
                                                        

 66%|██████▌   | 9670/14725 [8:48:26<4:44:02,  3.37s/it]
 66%|██████▌   | 9671/14725 [8:48:29<4:40:50,  3.33s/it]
 66%|██████▌   | 9672/14725 [8:48:32<4:43:06,  3.36s/it]
 66%|██████▌   | 9673/14725 [8:48:35<4:38:19,  3.31s/it]
 66%|██████▌   | 9674/14725 [8:48:39<4:37:13,  3.29s/it]
0: {'loss': 0.6921, 'grad_norm': 0.3834810024737715, 'learning_rate': 1e-05, 'epoch': 0.66}
0:  66%|██████▌   | 9675/14725 [8:48:42<4:37:20,  3.30s/it]
 66%|██████▌   | 9676/14725 [8:48:45<4:35:54,  3.28s/it]
 66%|██████▌   | 9677/14725 [8:48:49<4:42:19,  3.36s/it]
 66%|██████▌   | 9678/14725 [8:48:52<4:40:39,  3.34s/it]
 66%|██████▌   | 9679/14725 [8:48:55<4:42:20,  3.36s/it]
 66%|██████▌   | 9680/14725 [8:48:59<4:41:27,  3.35s/it]
                                                        

 66%|██████▌   | 9680/14725 [8:48:59<4:41:27,  3.35s/it]
 66%|██████▌   | 9681/14725 [8:49:02<4:38:31,  3.31s/it]
 66%|██████▌   | 9682/14725 [8:49:05<4:33:40,  3.26s/it]
 66%|██████▌   | 9683/14725 [8:49:08<4:28:58,  3.20s/it]
 66%|██████▌   | 9684/14725 [8:49:11<4:30:04,  3.21s/it]
 66%|██████▌   | 9685/14725 [8:49:15<4:30:28,  3.22s/it]
 66%|██████▌   | 9686/14725 [8:49:18<4:33:58,  3.26s/it]
 66%|██████▌   | 9687/14725 [
0: {'loss': 0.6971, 'grad_norm': 0.41212881018028524, 'learning_rate': 1e-05, 'epoch': 0.66}
0: 8:49:21<4:35:35,  3.28s/it]
 66%|██████▌   | 9688/14725 [8:49:25<4:36:27,  3.29s/it]
 66%|██████▌   | 9689/14725 [8:49:28<4:31:37,  3.24s/it]
 66%|██████▌   | 9690/14725 [8:49:31<4:33:06,  3.25s/it]
                                                        

 66%|██████▌   | 9690/14725 [8:49:31<4:33:06,  3.25s/it]
 66%|██████▌   | 9691/14725 [8:49:34<4:32:33,  3.25s/it]
 66%|██████▌   | 9692/14725 [8:49:38<4:32:44,  3.25s/it]
 66%|██████▌   | 9693/14725 [8:49:41<4:31:51,  3.24s/it]
 66%|██████▌   | 9694/14725 [8:49:44<4:35:53,  3.29s/it]
 66%|██████▌   | 9695/14725 [8:49:47<4:33:52,  3.27s/it]
 66%|██████▌   | 9696/14725 [8:49:50<4:29:14,  3.21s/it]
 66%|██████▌   | 9697/14725 [8:49:54<4:41:16,  3.36s/it]
 66%|██████▌   | 9698/14725 [8:49:57<4:38:14,  3.32s/it]
 66%|██████▌   | 9699/14725 [8:50:01<4:37:37,  3.31s/it]
 66%|███�
0: {'loss': 0.6918, 'grad_norm': 0.43029532534004467, 'learning_rate': 1e-05, 'epoch': 0.66}
0: {'loss': 0.6762, 'grad_norm': 0.3905026056435541, 'learning_rate': 1e-05, 'epoch': 0.66}
0: ��██▌   | 9700/14725 [8:50:04<4:41:33,  3.36s/it]
                                                        

 66%|██████▌   | 9700/14725 [8:50:04<4:41:33,  3.36s/it]
 66%|██████▌   | 9701/14725 [8:50:08<4:43:50,  3.39s/it]
 66%|██████▌   | 9702/14725 [8:50:11<4:46:33,  3.42s/it]
 66%|██████▌   | 9703/14725 [8:50:14<4:43:12,  3.38s/it]
 66%|██████▌   | 9704/14725 [8:50:18<4:39:46,  3.34s/it]
 66%|██████▌   | 9705/14725 [8:50:21<4:41:45,  3.37s/it]
 66%|██████▌   | 9706/14725 [8:50:24<4:36:06,  3.30s/it]
 66%|██████▌   | 9707/14725 [8:50:27<4:32:58,  3.26s/it]
 66%|██████▌   | 9708/14725 [8:50:30<4:27:38,  3.20s/it]
 66%|██████▌   | 9709/14725 [8:50:34<4:27:42,  3.20s/it]
 66%|██████▌   | 9710/14725 [8:50:37<4:32:33,  3.26s/it]
                                                        

 66%|██████▌   | 9710/14725 [8:50:37<4:32:33,  3.26s/it]
0: {'loss': 0.6875, 'grad_norm': 0.4211120918735048, 'learning_rate': 1e-05, 'epoch': 0.66}
0:  66%|██████▌   | 9711/14725 [8:50:41<4:42:04,  3.38s/it]
 66%|██████▌   | 9712/14725 [8:50:44<4:42:23,  3.38s/it]
 66%|██████▌   | 9713/14725 [8:50:47<4:37:00,  3.32s/it]
 66%|██████▌   | 9714/14725 [8:50:50<4:30:02,  3.23s/it]
 66%|██████▌   | 9715/14725 [8:50:54<4:32:50,  3.27s/it]
 66%|██████▌   | 9716/14725 [8:50:57<4:35:21,  3.30s/it]
 66%|██████▌   | 9717/14725 [8:51:00<4:34:23,  3.29s/it]
 66%|██████▌   | 9718/14725 [8:51:04<4:38:24,  3.34s/it]
 66%|██████▌   | 9719/14725 [8:51:07<4:38:30,  3.34s/it]
 66%|██████▌   | 9720/14725 [8:51:10<4:35:47,  3.31s/it]
                                                        

 66%|██████▌   | 9720/14725 [8:51:10<4:35:47,  3.31s/it]
 66%|██████▌   | 9721/14725 [8:51:14<4:33:08,  3.28s/it]
 66%|██████▌   | 9722/14725 [8:51:17<4:31:16,  3.25s/it]
 66%|██████▌   | 9723/14725 [
0: {'loss': 0.686, 'grad_norm': 0.4260876493227654, 'learning_rate': 1e-05, 'epoch': 0.66}
0: 8:51:20<4:31:03,  3.25s/it]
 66%|██████▌   | 9724/14725 [8:51:23<4:32:23,  3.27s/it]
 66%|██████▌   | 9725/14725 [8:51:27<4:32:13,  3.27s/it]
 66%|██████▌   | 9726/14725 [8:51:30<4:35:04,  3.30s/it]
 66%|██████▌   | 9727/14725 [8:51:33<4:37:26,  3.33s/it]
 66%|██████▌   | 9728/14725 [8:51:37<4:37:46,  3.34s/it]
 66%|██████▌   | 9729/14725 [8:51:40<4:37:51,  3.34s/it]
 66%|██████▌   | 9730/14725 [8:51:43<4:32:35,  3.27s/it]
                                                        

 66%|██████▌   | 9730/14725 [8:51:43<4:32:35,  3.27s/it]
 66%|██████▌   | 9731/14725 [8:51:46<4:31:19,  3.26s/it]
 66%|██████▌   | 9732/14725 [8:51:50<4:34:48,  3.30s/it]
 66%|██████▌   | 9733/14725 [8:51:53<4:34:44,  3.30s/it]
 66%|██████▌   | 9734/14725 [8:51:57<4:46:41,  3.45s/it]
 66%|██████▌   | 9735/14725 [8:52:00<4:45:17,  3.43s/it]
 66%|███�
0: {'loss': 0.7097, 'grad_norm': 0.40550274817014875, 'learning_rate': 1e-05, 'epoch': 0.66}
0: ��██▌   | 9736/14725 [8:52:04<4:41:19,  3.38s/it]
 66%|██████▌   | 9737/14725 [8:52:07<4:34:01,  3.30s/it]
 66%|██████▌   | 9738/14725 [8:52:10<4:30:23,  3.25s/it]
 66%|██████▌   | 9739/14725 [8:52:13<4:35:24,  3.31s/it]
 66%|██████▌   | 9740/14725 [8:52:16<4:33:38,  3.29s/it]
                                                        

 66%|██████▌   | 9740/14725 [8:52:16<4:33:38,  3.29s/it]
 66%|██████▌   | 9741/14725 [8:52:20<4:32:01,  3.27s/it]
 66%|██████▌   | 9742/14725 [8:52:23<4:30:50,  3.26s/it]
 66%|██████▌   | 9743/14725 [8:52:26<4:29:30,  3.25s/it]
 66%|██████▌   | 9744/14725 [8:52:29<4:30:09,  3.25s/it]
 66%|██████▌   | 9745/14725 [8:52:33<4:38:19,  3.35s/it]
 66%|██████▌   | 9746/14725 [8:52:36<4:35:45,  3.32s/it]
 66%|██████▌   | 9747/14725 [8:52:40<4:35:43,  3.32s/it]
 66%|██████▌   | 9748/14725 [8:52:43<4:37:31
0: {'loss': 0.6892, 'grad_norm': 0.40389889048865324, 'learning_rate': 1e-05, 'epoch': 0.66}
0: ,  3.35s/it]
 66%|██████▌   | 9749/14725 [8:52:46<4:33:46,  3.30s/it]
 66%|██████▌   | 9750/14725 [8:52:49<4:30:24,  3.26s/it]
                                                        

 66%|██████▌   | 9750/14725 [8:52:49<4:30:24,  3.26s/it]
 66%|██████▌   | 9751/14725 [8:52:53<4:39:21,  3.37s/it]
 66%|██████▌   | 9752/14725 [8:52:56<4:30:20,  3.26s/it]
 66%|██████▌   | 9753/14725 [8:53:00<4:41:13,  3.39s/it]
 66%|██████▌   | 9754/14725 [8:53:03<4:36:41,  3.34s/it]
 66%|██████▌   | 9755/14725 [8:53:06<4:31:04,  3.27s/it]
 66%|██████▋   | 9756/14725 [8:53:09<4:28:37,  3.24s/it]
 66%|██████▋   | 9757/14725 [8:53:12<4:26:29,  3.22s/it]
 66%|██████▋   | 9758/14725 [8:53:16<4:32:03,  3.29s/it]
 66%|██████▋   | 9759/14725 [8:53:19<4:33:39,  3.31s/it]
 66%|██████▋   | 9760/14725 [8:53:23<4:44:00,  3.43s/it]
                              
0: {'loss': 0.6836, 'grad_norm': 0.4008461301283662, 'learning_rate': 1e-05, 'epoch': 0.66}
0: {'loss': 0.6754, 'grad_norm': 0.4009250025353552, 'learning_rate': 1e-05, 'epoch': 0.66}
0:                           

 66%|██████▋   | 9760/14725 [8:53:23<4:44:00,  3.43s/it]
 66%|██████▋   | 9761/14725 [8:53:26<4:36:04,  3.34s/it]
 66%|██████▋   | 9762/14725 [8:53:29<4:33:09,  3.30s/it]
 66%|██████▋   | 9763/14725 [8:53:32<4:27:05,  3.23s/it]
 66%|██████▋   | 9764/14725 [8:53:35<4:26:22,  3.22s/it]
 66%|██████▋   | 9765/14725 [8:53:39<4:34:40,  3.32s/it]
 66%|██████▋   | 9766/14725 [8:53:42<4:34:50,  3.33s/it]
 66%|██████▋   | 9767/14725 [8:53:45<4:28:42,  3.25s/it]
 66%|██████▋   | 9768/14725 [8:53:49<4:25:24,  3.21s/it]
 66%|██████▋   | 9769/14725 [8:53:52<4:28:58,  3.26s/it]
 66%|██████▋   | 9770/14725 [8:53:55<4:25:55,  3.22s/it]
                                                        

 66%|██████▋   | 9770/14725 [8:53:55<4:25:55,  3.22s/it]
 66%|██████▋   | 9771/14725 [8:53:58<4:23:32,  3.19s/it]
 66%|███�
0: {'loss': 0.6939, 'grad_norm': 0.4064801657607591, 'learning_rate': 1e-05, 'epoch': 0.66}
0: ��██▋   | 9772/14725 [8:54:02<4:27:14,  3.24s/it]
 66%|██████▋   | 9773/14725 [8:54:05<4:31:16,  3.29s/it]
 66%|██████▋   | 9774/14725 [8:54:08<4:28:56,  3.26s/it]
 66%|██████▋   | 9775/14725 [8:54:11<4:26:05,  3.23s/it]
 66%|██████▋   | 9776/14725 [8:54:14<4:25:38,  3.22s/it]
 66%|██████▋   | 9777/14725 [8:54:18<4:27:57,  3.25s/it]
 66%|██████▋   | 9778/14725 [8:54:21<4:27:24,  3.24s/it]
 66%|██████▋   | 9779/14725 [8:54:24<4:31:49,  3.30s/it]
 66%|██████▋   | 9780/14725 [8:54:28<4:34:23,  3.33s/it]
                                                        

 66%|██████▋   | 9780/14725 [8:54:28<4:34:23,  3.33s/it]
 66%|██████▋   | 9781/14725 [8:54:31<4:29:14,  3.27s/it]
 66%|██████▋   | 9782/14725 [8:54:34<4:29:18,  3.27s/it]
 66%|██████▋   | 9783/14725 [8:54:37<4:25:30,  3.22s/it]
 66%|██████▋   | 9784/14725 [8:54:40<4:23:06
0: {'loss': 0.6911, 'grad_norm': 0.41159096096360875, 'learning_rate': 1e-05, 'epoch': 0.66}
0: ,  3.20s/it]
 66%|██████▋   | 9785/14725 [8:54:44<4:25:57,  3.23s/it]
 66%|██████▋   | 9786/14725 [8:54:47<4:25:58,  3.23s/it]
 66%|██████▋   | 9787/14725 [8:54:50<4:27:59,  3.26s/it]
 66%|██████▋   | 9788/14725 [8:54:54<4:26:54,  3.24s/it]
 66%|██████▋   | 9789/14725 [8:54:57<4:21:24,  3.18s/it]
 66%|██████▋   | 9790/14725 [8:55:00<4:23:01,  3.20s/it]
                                                        

 66%|██████▋   | 9790/14725 [8:55:00<4:23:01,  3.20s/it]
 66%|██████▋   | 9791/14725 [8:55:03<4:21:08,  3.18s/it]
 66%|██████▋   | 9792/14725 [8:55:06<4:28:18,  3.26s/it]
 67%|██████▋   | 9793/14725 [8:55:10<4:34:06,  3.33s/it]
 67%|██████▋   | 9794/14725 [8:55:13<4:36:12,  3.36s/it]
 67%|██████▋   | 9795/14725 [8:55:17<4:31:12,  3.30s/it]
 67%|██████▋   | 9796/14725 [8:55:20<4:26:24,  3.24s/it]
 67%|██████▋   |
0: {'loss': 0.6749, 'grad_norm': 0.3922366756364316, 'learning_rate': 1e-05, 'epoch': 0.67}
0:  9797/14725 [8:55:23<4:21:46,  3.19s/it]
 67%|██████▋   | 9798/14725 [8:55:26<4:25:43,  3.24s/it]
 67%|██████▋   | 9799/14725 [8:55:29<4:22:53,  3.20s/it]
 67%|██████▋   | 9800/14725 [8:55:32<4:19:43,  3.16s/it]
                                                        

 67%|██████▋   | 9800/14725 [8:55:32<4:19:43,  3.16s/it]
 67%|██████▋   | 9801/14725 [8:55:36<4:22:39,  3.20s/it]
 67%|██████▋   | 9802/14725 [8:55:39<4:36:28,  3.37s/it]
 67%|██████▋   | 9803/14725 [8:55:42<4:32:44,  3.32s/it]
 67%|██████▋   | 9804/14725 [8:55:46<4:31:15,  3.31s/it]
 67%|██████▋   | 9805/14725 [8:55:49<4:29:41,  3.29s/it]
 67%|██████▋   | 9806/14725 [8:55:52<4:30:08,  3.30s/it]
 67%|██████▋   | 9807/14725 [8:55:55<4:26:36,  3.25s/it]
 67%|██████▋   | 9808/14725 [8:55:59<4:29:34,  3.29s/it]
 67%|██████▋   | 9809/14725 [8:56:02<4:31:16,  3.31s/it]
 6
0: {'loss': 0.6852, 'grad_norm': 0.4223277332648646, 'learning_rate': 1e-05, 'epoch': 0.67}
0: {'loss': 0.6805, 'grad_norm': 0.41298096661111766, 'learning_rate': 1e-05, 'epoch': 0.67}
0: 7%|██████▋   | 9810/14725 [8:56:05<4:30:49,  3.31s/it]
                                                        

 67%|██████▋   | 9810/14725 [8:56:06<4:30:49,  3.31s/it]
 67%|██████▋   | 9811/14725 [8:56:09<4:34:39,  3.35s/it]
 67%|██████▋   | 9812/14725 [8:56:12<4:36:06,  3.37s/it]
 67%|██████▋   | 9813/14725 [8:56:16<4:30:54,  3.31s/it]
 67%|██████▋   | 9814/14725 [8:56:19<4:31:40,  3.32s/it]
 67%|██████▋   | 9815/14725 [8:56:22<4:27:28,  3.27s/it]
 67%|██████▋   | 9816/14725 [8:56:25<4:25:39,  3.25s/it]
 67%|██████▋   | 9817/14725 [8:56:29<4:26:30,  3.26s/it]
 67%|██████▋   | 9818/14725 [8:56:32<4:24:49,  3.24s/it]
 67%|██████▋   | 9819/14725 [8:56:35<4:27:53,  3.28s/it]
 67%|██████▋   | 9820/14725 [8:56:38<4:28:44,  3.29s/it]
                                                        

 67%|██████▋   | 9820/14725 [8:56:38<4:28:44
0: {'loss': 0.6843, 'grad_norm': 0.4283689953890898, 'learning_rate': 1e-05, 'epoch': 0.67}
0: ,  3.29s/it]
 67%|██████▋   | 9821/14725 [8:56:42<4:25:19,  3.25s/it]
 67%|██████▋   | 9822/14725 [8:56:45<4:25:59,  3.26s/it]
 67%|██████▋   | 9823/14725 [8:56:48<4:27:00,  3.27s/it]
 67%|██████▋   | 9824/14725 [8:56:51<4:26:45,  3.27s/it]
 67%|██████▋   | 9825/14725 [8:56:55<4:29:10,  3.30s/it]
 67%|██████▋   | 9826/14725 [8:56:58<4:27:39,  3.28s/it]
 67%|██████▋   | 9827/14725 [8:57:01<4:26:46,  3.27s/it]
 67%|██████▋   | 9828/14725 [8:57:05<4:28:49,  3.29s/it]
 67%|██████▋   | 9829/14725 [8:57:08<4:32:33,  3.34s/it]
 67%|██████▋   | 9830/14725 [8:57:11<4:34:37,  3.37s/it]
                                                        

 67%|██████▋   | 9830/14725 [8:57:11<4:34:37,  3.37s/it]
 67%|██████▋   | 9831/14725 [8:57:15<4:27:54,  3.28s/it]
 67%|██████▋   | 9832/14725 [8:57:18<4:24:30,  3.24s/it]
 67%|██████▋   |
0: {'loss': 0.6788, 'grad_norm': 0.4032114326878211, 'learning_rate': 1e-05, 'epoch': 0.67}
0:  9833/14725 [8:57:21<4:24:51,  3.25s/it]
 67%|██████▋   | 9834/14725 [8:57:24<4:25:20,  3.26s/it]
 67%|██████▋   | 9835/14725 [8:57:27<4:25:26,  3.26s/it]
 67%|██████▋   | 9836/14725 [8:57:31<4:22:41,  3.22s/it]
 67%|██████▋   | 9837/14725 [8:57:34<4:25:39,  3.26s/it]
 67%|██████▋   | 9838/14725 [8:57:37<4:28:25,  3.30s/it]
 67%|██████▋   | 9839/14725 [8:57:41<4:26:56,  3.28s/it]
 67%|██████▋   | 9840/14725 [8:57:44<4:24:35,  3.25s/it]
                                                        

 67%|██████▋   | 9840/14725 [8:57:44<4:24:35,  3.25s/it]
 67%|██████▋   | 9841/14725 [8:57:47<4:20:59,  3.21s/it]
 67%|██████▋   | 9842/14725 [8:57:50<4:22:28,  3.23s/it]
 67%|██████▋   | 9843/14725 [8:57:53<4:18:53,  3.18s/it]
 67%|██████▋   | 9844/14725 [8:57:56<4:18:12,  3.17s/it]
 67%|██████▋   | 9845/14725 [8:58:00<4:17:49,  3.17s/it]
 6
0: {'loss': 0.6703, 'grad_norm': 0.4080404967716629, 'learning_rate': 1e-05, 'epoch': 0.67}
0: 7%|██████▋   | 9846/14725 [8:58:03<4:18:53,  3.18s/it]
 67%|██████▋   | 9847/14725 [8:58:06<4:23:21,  3.24s/it]
 67%|██████▋   | 9848/14725 [8:58:09<4:21:45,  3.22s/it]
 67%|██████▋   | 9849/14725 [8:58:13<4:25:17,  3.26s/it]
 67%|██████▋   | 9850/14725 [8:58:16<4:23:02,  3.24s/it]
                                                        

 67%|██████▋   | 9850/14725 [8:58:16<4:23:02,  3.24s/it]
 67%|██████▋   | 9851/14725 [8:58:19<4:29:01,  3.31s/it]
 67%|██████▋   | 9852/14725 [8:58:23<4:28:13,  3.30s/it]
 67%|██████▋   | 9853/14725 [8:58:26<4:26:47,  3.29s/it]
 67%|██████▋   | 9854/14725 [8:58:29<4:20:26,  3.21s/it]
 67%|██████▋   | 9855/14725 [8:58:32<4:22:18,  3.23s/it]
 67%|██████▋   | 9856/14725 [8:58:36<4:26:06,  3.28s/it]
 67%|██████▋   | 9857/14725 [8:58:39<4:33:58,  3.38s/it]
 67%|██████▋   | 9858/14725 [8:
0: {'loss': 0.6797, 'grad_norm': 0.4242256696775422, 'learning_rate': 1e-05, 'epoch': 0.67}
0: 58:42<4:28:27,  3.31s/it]
 67%|██████▋   | 9859/14725 [8:58:46<4:34:08,  3.38s/it]
 67%|██████▋   | 9860/14725 [8:58:49<4:27:25,  3.30s/it]
                                                        

 67%|██████▋   | 9860/14725 [8:58:49<4:27:25,  3.30s/it]
 67%|██████▋   | 9861/14725 [8:58:52<4:26:54,  3.29s/it]
 67%|██████▋   | 9862/14725 [8:58:56<4:28:03,  3.31s/it]
 67%|██████▋   | 9863/14725 [8:58:59<4:25:14,  3.27s/it]
 67%|██████▋   | 9864/14725 [8:59:02<4:25:30,  3.28s/it]
 67%|██████▋   | 9865/14725 [8:59:05<4:25:43,  3.28s/it]
 67%|██████▋   | 9866/14725 [8:59:09<4:22:45,  3.24s/it]
 67%|██████▋   | 9867/14725 [8:59:12<4:21:03,  3.22s/it]
 67%|██████▋   | 9868/14725 [8:59:15<4:20:32,  3.22s/it]
 67%|██████▋   | 9869/14725 [8:59:18<4:19:02,  3.20s/it]
 67%|██████▋   | 9870/14725 [8:59:21<4:22:38,  3.25s/it]
                 
0: {'loss': 0.6908, 'grad_norm': 0.4651899754370994, 'learning_rate': 1e-05, 'epoch': 0.67}
0: {'loss': 0.6827, 'grad_norm': 0.4189598738521035, 'learning_rate': 1e-05, 'epoch': 0.67}
0:                                        

 67%|██████▋   | 9870/14725 [8:59:21<4:22:38,  3.25s/it]
 67%|██████▋   | 9871/14725 [8:59:25<4:26:29,  3.29s/it]
 67%|██████▋   | 9872/14725 [8:59:28<4:21:37,  3.23s/it]
 67%|██████▋   | 9873/14725 [8:59:31<4:21:44,  3.24s/it]
 67%|██████▋   | 9874/14725 [8:59:34<4:18:15,  3.19s/it]
 67%|██████▋   | 9875/14725 [8:59:38<4:21:43,  3.24s/it]
 67%|██████▋   | 9876/14725 [8:59:41<4:25:55,  3.29s/it]
 67%|██████▋   | 9877/14725 [8:59:44<4:24:33,  3.27s/it]
 67%|██████▋   | 9878/14725 [8:59:48<4:27:28,  3.31s/it]
 67%|██████▋   | 9879/14725 [8:59:51<4:30:39,  3.35s/it]
 67%|██████▋   | 9880/14725 [8:59:54<4:25:45,  3.29s/it]
                                                        

 67%|██████▋   | 9880/14725 [8:59:54<4:25:45,  3.29s/it]
 67%|██████▋   | 9881/14725 [8:59:57<4:22:05,  3.25s/it]
 6
0: {'loss': 0.6963, 'grad_norm': 0.4200568007118224, 'learning_rate': 1e-05, 'epoch': 0.67}
0: 7%|██████▋   | 9882/14725 [9:00:00<4:18:09,  3.20s/it]
 67%|██████▋   | 9883/14725 [9:00:04<4:17:51,  3.20s/it]
 67%|██████▋   | 9884/14725 [9:00:07<4:16:17,  3.18s/it]
 67%|██████▋   | 9885/14725 [9:00:10<4:22:11,  3.25s/it]
 67%|██████▋   | 9886/14725 [9:00:14<4:24:25,  3.28s/it]
 67%|██████▋   | 9887/14725 [9:00:17<4:26:35,  3.31s/it]
 67%|██████▋   | 9888/14725 [9:00:20<4:29:39,  3.34s/it]
 67%|██████▋   | 9889/14725 [9:00:24<4:25:46,  3.30s/it]
 67%|██████▋   | 9890/14725 [9:00:27<4:30:49,  3.36s/it]
                                                        

 67%|██████▋   | 9890/14725 [9:00:27<4:30:49,  3.36s/it]
 67%|██████▋   | 9891/14725 [9:00:30<4:26:05,  3.30s/it]
 67%|██████▋   | 9892/14725 [9:00:33<4:24:16,  3.28s/it]
 67%|██████▋   | 9893/14725 [9:00:37<4:22:08,  3.26s/it]
 67%|██████▋   | 9894/14725 [9:
0: {'loss': 0.6761, 'grad_norm': 0.41086086942346045, 'learning_rate': 1e-05, 'epoch': 0.67}
0: 00:40<4:19:57,  3.23s/it]
 67%|██████▋   | 9895/14725 [9:00:43<4:21:40,  3.25s/it]
 67%|██████▋   | 9896/14725 [9:00:46<4:23:10,  3.27s/it]
 67%|██████▋   | 9897/14725 [9:00:50<4:27:36,  3.33s/it]
 67%|██████▋   | 9898/14725 [9:00:53<4:22:09,  3.26s/it]
 67%|██████▋   | 9899/14725 [9:00:56<4:26:11,  3.31s/it]
 67%|██████▋   | 9900/14725 [9:01:00<4:23:05,  3.27s/it]
                                                        

 67%|██████▋   | 9900/14725 [9:01:00<4:23:05,  3.27s/it]
 67%|██████▋   | 9901/14725 [9:01:03<4:24:44,  3.29s/it]
 67%|██████▋   | 9902/14725 [9:01:06<4:19:48,  3.23s/it]
 67%|██████▋   | 9903/14725 [9:01:09<4:21:01,  3.25s/it]
 67%|██████▋   | 9904/14725 [9:01:13<4:20:08,  3.24s/it]
 67%|██████▋   | 9905/14725 [9:01:16<4:18:00,  3.21s/it]
 67%|██████▋   | 9906/14725 [9:01:19<4:18:45,  3.22s/it]
 67%|████
0: {'loss': 0.6838, 'grad_norm': 0.40737538597721373, 'learning_rate': 1e-05, 'epoch': 0.67}
0: ██▋   | 9907/14725 [9:01:22<4:19:16,  3.23s/it]
 67%|██████▋   | 9908/14725 [9:01:25<4:18:45,  3.22s/it]
 67%|██████▋   | 9909/14725 [9:01:28<4:16:02,  3.19s/it]
 67%|██████▋   | 9910/14725 [9:01:32<4:20:10,  3.24s/it]
                                                        

 67%|██████▋   | 9910/14725 [9:01:32<4:20:10,  3.24s/it]
 67%|██████▋   | 9911/14725 [9:01:35<4:20:52,  3.25s/it]
 67%|██████▋   | 9912/14725 [9:01:38<4:19:07,  3.23s/it]
 67%|██████▋   | 9913/14725 [9:01:42<4:20:35,  3.25s/it]
 67%|██████▋   | 9914/14725 [9:01:45<4:20:39,  3.25s/it]
 67%|██████▋   | 9915/14725 [9:01:48<4:22:41,  3.28s/it]
 67%|██████▋   | 9916/14725 [9:01:51<4:18:23,  3.22s/it]
 67%|██████▋   | 9917/14725 [9:01:55<4:19:11,  3.23s/it]
 67%|██████▋   | 9918/14725 [9:01:58<4:14:31,  3.18s/it]
 67%|██████▋   | 9919/14725 [9:02:01<4:15:28, 
0: {'loss': 0.6849, 'grad_norm': 0.39362849175284986, 'learning_rate': 1e-05, 'epoch': 0.67}
0: {'loss': 0.6873, 'grad_norm': 0.41672668827609066, 'learning_rate': 1e-05, 'epoch': 0.67}
0:  3.19s/it]
 67%|██████▋   | 9920/14725 [9:02:04<4:14:45,  3.18s/it]
                                                        

 67%|██████▋   | 9920/14725 [9:02:04<4:14:45,  3.18s/it]
 67%|██████▋   | 9921/14725 [9:02:07<4:18:41,  3.23s/it]
 67%|██████▋   | 9922/14725 [9:02:11<4:17:51,  3.22s/it]
 67%|██████▋   | 9923/14725 [9:02:14<4:17:27,  3.22s/it]
 67%|██████▋   | 9924/14725 [9:02:17<4:14:57,  3.19s/it]
 67%|██████▋   | 9925/14725 [9:02:20<4:16:35,  3.21s/it]
 67%|██████▋   | 9926/14725 [9:02:23<4:16:19,  3.20s/it]
 67%|██████▋   | 9927/14725 [9:02:27<4:16:42,  3.21s/it]
 67%|██████▋   | 9928/14725 [9:02:30<4:21:24,  3.27s/it]
 67%|██████▋   | 9929/14725 [9:02:33<4:25:14,  3.32s/it]
 67%|██████▋   | 9930/14725 [9:02:37<4:23:19,  3.30s/it]
                                                        

 67%|██████▋   | 9930/14725 [9:
0: {'loss': 0.6749, 'grad_norm': 0.44873531922667165, 'learning_rate': 1e-05, 'epoch': 0.68}
0: 02:37<4:23:19,  3.30s/it]
 67%|██████▋   | 9931/14725 [9:02:40<4:23:11,  3.29s/it]
 67%|██████▋   | 9932/14725 [9:02:43<4:23:18,  3.30s/it]
 67%|██████▋   | 9933/14725 [9:02:46<4:19:27,  3.25s/it]
 67%|██████▋   | 9934/14725 [9:02:50<4:22:07,  3.28s/it]
 67%|██████▋   | 9935/14725 [9:02:53<4:21:46,  3.28s/it]
 67%|██████▋   | 9936/14725 [9:02:56<4:16:09,  3.21s/it]
 67%|██████▋   | 9937/14725 [9:02:59<4:19:59,  3.26s/it]
 67%|██████▋   | 9938/14725 [9:03:03<4:19:49,  3.26s/it]
 67%|██████▋   | 9939/14725 [9:03:06<4:20:24,  3.26s/it]
 68%|██████▊   | 9940/14725 [9:03:09<4:21:55,  3.28s/it]
                                                        

 68%|██████▊   | 9940/14725 [9:03:09<4:21:55,  3.28s/it]
 68%|██████▊   | 9941/14725 [9:03:12<4:20:40,  3.27s/it]
 68%|██████▊   | 9942/14725 [9:03:16<4:24:22,  3.32s/it]
 68%|████
0: {'loss': 0.684, 'grad_norm': 0.40103715929722633, 'learning_rate': 1e-05, 'epoch': 0.68}
0: ██▊   | 9943/14725 [9:03:19<4:25:55,  3.34s/it]
 68%|██████▊   | 9944/14725 [9:03:23<4:26:20,  3.34s/it]
 68%|██████▊   | 9945/14725 [9:03:26<4:21:04,  3.28s/it]
 68%|██████▊   | 9946/14725 [9:03:29<4:21:31,  3.28s/it]
 68%|██████▊   | 9947/14725 [9:03:32<4:20:06,  3.27s/it]
 68%|██████▊   | 9948/14725 [9:03:35<4:13:56,  3.19s/it]
 68%|██████▊   | 9949/14725 [9:03:38<4:12:20,  3.17s/it]
 68%|██████▊   | 9950/14725 [9:03:42<4:16:05,  3.22s/it]
                                                        

 68%|██████▊   | 9950/14725 [9:03:42<4:16:05,  3.22s/it]
 68%|██████▊   | 9951/14725 [9:03:45<4:19:05,  3.26s/it]
 68%|██████▊   | 9952/14725 [9:03:48<4:15:49,  3.22s/it]
 68%|██████▊   | 9953/14725 [9:03:51<4:14:41,  3.20s/it]
 68%|██████▊   | 9954/14725 [9:03:55<4:19:00,  3.26s/it]
 68%|██████▊   | 9955/14725 [9:03:58<4:20:34, 
0: {'loss': 0.6941, 'grad_norm': 0.43732728339786636, 'learning_rate': 1e-05, 'epoch': 0.68}
0:  3.28s/it]
 68%|██████▊   | 9956/14725 [9:04:02<4:24:04,  3.32s/it]
 68%|██████▊   | 9957/14725 [9:04:05<4:22:42,  3.31s/it]
 68%|██████▊   | 9958/14725 [9:04:08<4:19:24,  3.27s/it]
 68%|██████▊   | 9959/14725 [9:04:11<4:24:27,  3.33s/it]
 68%|██████▊   | 9960/14725 [9:04:15<4:24:01,  3.32s/it]
                                                        

 68%|██████▊   | 9960/14725 [9:04:15<4:24:01,  3.32s/it]
 68%|██████▊   | 9961/14725 [9:04:18<4:22:08,  3.30s/it]
 68%|██████▊   | 9962/14725 [9:04:21<4:18:45,  3.26s/it]
 68%|██████▊   | 9963/14725 [9:04:24<4:16:03,  3.23s/it]
 68%|██████▊   | 9964/14725 [9:04:27<4:14:26,  3.21s/it]
 68%|██████▊   | 9965/14725 [9:04:31<4:16:13,  3.23s/it]
 68%|██████▊   | 9966/14725 [9:04:34<4:13:00,  3.19s/it]
 68%|██████▊   | 9967/14725 [9:04:37<4:18:24,  3.26s/it]
 68%|██████▊   | 9
0: {'loss': 0.6946, 'grad_norm': 0.39034194215445456, 'learning_rate': 1e-05, 'epoch': 0.68}
0: 968/14725 [9:04:40<4:13:50,  3.20s/it]
 68%|██████▊   | 9969/14725 [9:04:44<4:17:05,  3.24s/it]
 68%|██████▊   | 9970/14725 [9:04:47<4:16:41,  3.24s/it]
                                                        

 68%|██████▊   | 9970/14725 [9:04:47<4:16:41,  3.24s/it]
 68%|██████▊   | 9971/14725 [9:04:50<4:15:59,  3.23s/it]
 68%|██████▊   | 9972/14725 [9:04:53<4:16:12,  3.23s/it]
 68%|██████▊   | 9973/14725 [9:04:57<4:18:25,  3.26s/it]
 68%|██████▊   | 9974/14725 [9:05:00<4:14:40,  3.22s/it]
 68%|██████▊   | 9975/14725 [9:05:03<4:18:44,  3.27s/it]
 68%|██████▊   | 9976/14725 [9:05:06<4:15:15,  3.22s/it]
 68%|██████▊   | 9977/14725 [9:05:09<4:13:02,  3.20s/it]
 68%|██████▊   | 9978/14725 [9:05:13<4:12:18,  3.19s/it]
 68%|██████▊   | 9979/14725 [9:05:16<4:11:37,  3.18s/it]
 68%|██████▊   | 9980/14725 [9:05:19<4:11:57,  3.19s/it]
    
0: {'loss': 0.6864, 'grad_norm': 0.4404513661691384, 'learning_rate': 1e-05, 'epoch': 0.68}
0: {'loss': 0.6911, 'grad_norm': 0.436854590476222, 'learning_rate': 1e-05, 'epoch': 0.68}
0:                                                     

 68%|██████▊   | 9980/14725 [9:05:19<4:11:57,  3.19s/it]
 68%|██████▊   | 9981/14725 [9:05:22<4:14:35,  3.22s/it]
 68%|██████▊   | 9982/14725 [9:05:26<4:15:43,  3.24s/it]
 68%|██████▊   | 9983/14725 [9:05:29<4:22:33,  3.32s/it]
 68%|██████▊   | 9984/14725 [9:05:32<4:21:46,  3.31s/it]
 68%|██████▊   | 9985/14725 [9:05:36<4:22:12,  3.32s/it]
 68%|██████▊   | 9986/14725 [9:05:39<4:17:56,  3.27s/it]
 68%|██████▊   | 9987/14725 [9:05:42<4:13:31,  3.21s/it]
 68%|██████▊   | 9988/14725 [9:05:46<4:25:39,  3.36s/it]
 68%|██████▊   | 9989/14725 [9:05:49<4:23:07,  3.33s/it]
 68%|██████▊   | 9990/14725 [9:05:52<4:20:21,  3.30s/it]
                                                        

 68%|██████▊   | 9990/14725 [9:05:52<4:20:21,  3.30s/it]
 68%|██████▊   | 9991/14725 [9:05:56<4:22:37, 
0: {'loss': 0.6704, 'grad_norm': 0.42852183218768974, 'learning_rate': 1e-05, 'epoch': 0.68}
0:  3.33s/it]
 68%|██████▊   | 9992/14725 [9:05:59<4:19:45,  3.29s/it]
 68%|██████▊   | 9993/14725 [9:06:02<4:18:47,  3.28s/it]
 68%|██████▊   | 9994/14725 [9:06:05<4:19:38,  3.29s/it]
 68%|██████▊   | 9995/14725 [9:06:09<4:18:53,  3.28s/it]
 68%|██████▊   | 9996/14725 [9:06:12<4:22:52,  3.34s/it]
 68%|██████▊   | 9997/14725 [9:06:15<4:20:45,  3.31s/it]
 68%|██████▊   | 9998/14725 [9:06:19<4:18:40,  3.28s/it]
 68%|██████▊   | 9999/14725 [9:06:22<4:16:08,  3.25s/it]
 68%|██████▊   | 10000/14725 [9:06:25<4:16:37,  3.26s/it]
                                                         

 68%|██████▊   | 10000/14725 [9:06:25<4:16:37,  3.26s/it]
 68%|██████▊   | 10001/14725 [9:06:28<4:19:33,  3.30s/it]
 68%|██████▊   | 10002/14725 [9:06:32<4:16:31,  3.26s/it]
 68%|██████▊   | 10003/14725 [9:06:35<4:21:21,  3.32s/it]
 68%|██████▊
0: {'loss': 0.6873, 'grad_norm': 0.4146698458709184, 'learning_rate': 1e-05, 'epoch': 0.68}
0:    | 10004/14725 [9:06:38<4:21:26,  3.32s/it]
 68%|██████▊   | 10005/14725 [9:06:42<4:22:53,  3.34s/it]
 68%|██████▊   | 10006/14725 [9:06:45<4:23:08,  3.35s/it]
 68%|██████▊   | 10007/14725 [9:06:48<4:23:19,  3.35s/it]
 68%|██████▊   | 10008/14725 [9:06:52<4:19:18,  3.30s/it]
 68%|██████▊   | 10009/14725 [9:06:55<4:16:59,  3.27s/it]
 68%|██████▊   | 10010/14725 [9:06:58<4:16:50,  3.27s/it]
                                                         

 68%|██████▊   | 10010/14725 [9:06:58<4:16:50,  3.27s/it]
 68%|██████▊   | 10011/14725 [9:07:01<4:19:22,  3.30s/it]
 68%|██████▊   | 10012/14725 [9:07:05<4:17:27,  3.28s/it]
 68%|██████▊   | 10013/14725 [9:07:08<4:18:34,  3.29s/it]
 68%|██████▊   | 10014/14725 [9:07:11<4:18:32,  3.29s/it]
 68%|██████▊   | 10015/14725 [9:07:14<4:15:38,  3.26s/it]
 68%|██████▊   | 10016/14725 [9:07:18<4:1
0: {'loss': 0.6759, 'grad_norm': 0.4088022126477994, 'learning_rate': 1e-05, 'epoch': 0.68}
0: 5:33,  3.26s/it]
 68%|██████▊   | 10017/14725 [9:07:21<4:16:13,  3.27s/it]
 68%|██████▊   | 10018/14725 [9:07:24<4:15:29,  3.26s/it]
 68%|██████▊   | 10019/14725 [9:07:28<4:17:25,  3.28s/it]
 68%|██████▊   | 10020/14725 [9:07:31<4:17:58,  3.29s/it]
                                                         

 68%|██████▊   | 10020/14725 [9:07:31<4:17:58,  3.29s/it]
 68%|██████▊   | 10021/14725 [9:07:34<4:19:00,  3.30s/it]
 68%|██████▊   | 10022/14725 [9:07:37<4:15:52,  3.26s/it]
 68%|██████▊   | 10023/14725 [9:07:41<4:14:05,  3.24s/it]
 68%|██████▊   | 10024/14725 [9:07:44<4:11:47,  3.21s/it]
 68%|██████▊   | 10025/14725 [9:07:47<4:15:01,  3.26s/it]
 68%|██████▊   | 10026/14725 [9:07:50<4:17:32,  3.29s/it]
 68%|██████▊   | 10027/14725 [9:07:54<4:21:38,  3.34s/it]
 68%|██████▊   | 10028/14725 [9:07:57<4:25:42,  3.39s/it]
 68%|██�
0: {'loss': 0.6896, 'grad_norm': 0.41969542208702115, 'learning_rate': 1e-05, 'epoch': 0.68}
0: ��███▊   | 10029/14725 [9:08:01<4:21:07,  3.34s/it]
 68%|██████▊   | 10030/14725 [9:08:04<4:16:53,  3.28s/it]
                                                         

 68%|██████▊   | 10030/14725 [9:08:04<4:16:53,  3.28s/it]
 68%|██████▊   | 10031/14725 [9:08:07<4:17:34,  3.29s/it]
 68%|██████▊   | 10032/14725 [9:08:10<4:16:36,  3.28s/it]
 68%|██████▊   | 10033/14725 [9:08:14<4:16:36,  3.28s/it]
 68%|██████▊   | 10034/14725 [9:08:17<4:18:26,  3.31s/it]
 68%|██████▊   | 10035/14725 [9:08:20<4:21:23,  3.34s/it]
 68%|██████▊   | 10036/14725 [9:08:24<4:22:47,  3.36s/it]
 68%|██████▊   | 10037/14725 [9:08:27<4:23:27,  3.37s/it]
 68%|██████▊   | 10038/14725 [9:08:30<4:17:18,  3.29s/it]
 68%|██████▊   | 10039/14725 [9:08:34<4:18:57,  3.32s/it]
 68%|██████▊   | 10040/14725 [9:08:37<4:19:32,  3.32s/it]
                                         
0: {'loss': 0.6864, 'grad_norm': 0.43952040860039476, 'learning_rate': 1e-05, 'epoch': 0.68}
0: {'loss': 0.687, 'grad_norm': 0.43432474781628966, 'learning_rate': 1e-05, 'epoch': 0.68}
0:                 

 68%|██████▊   | 10040/14725 [9:08:37<4:19:32,  3.32s/it]
 68%|██████▊   | 10041/14725 [9:08:40<4:20:48,  3.34s/it]
 68%|██████▊   | 10042/14725 [9:08:44<4:19:09,  3.32s/it]
 68%|██████▊   | 10043/14725 [9:08:47<4:18:18,  3.31s/it]
 68%|██████▊   | 10044/14725 [9:08:50<4:15:36,  3.28s/it]
 68%|██████▊   | 10045/14725 [9:08:54<4:23:13,  3.37s/it]
 68%|██████▊   | 10046/14725 [9:08:57<4:20:08,  3.34s/it]
 68%|██████▊   | 10047/14725 [9:09:00<4:17:30,  3.30s/it]
 68%|██████▊   | 10048/14725 [9:09:04<4:19:15,  3.33s/it]
 68%|██████▊   | 10049/14725 [9:09:07<4:15:36,  3.28s/it]
 68%|██████▊   | 10050/14725 [9:09:10<4:19:53,  3.34s/it]
                                                         

 68%|██████▊   | 10050/14725 [9:09:10<4:19:53,  3.34s/it]
 68%|██████▊   | 10051/14725 [9:09:14<4:21:32,  3.36s/it]
 68%|██
0: {'loss': 0.6834, 'grad_norm': 0.3861215737483006, 'learning_rate': 1e-05, 'epoch': 0.68}
0: ████▊   | 10052/14725 [9:09:17<4:17:19,  3.30s/it]
 68%|██████▊   | 10053/14725 [9:09:20<4:19:38,  3.33s/it]
 68%|██████▊   | 10054/14725 [9:09:24<4:17:46,  3.31s/it]
 68%|██████▊   | 10055/14725 [9:09:27<4:21:09,  3.36s/it]
 68%|██████▊   | 10056/14725 [9:09:30<4:19:02,  3.33s/it]
 68%|██████▊   | 10057/14725 [9:09:34<4:17:38,  3.31s/it]
 68%|██████▊   | 10058/14725 [9:09:37<4:18:30,  3.32s/it]
 68%|██████▊   | 10059/14725 [9:09:40<4:19:00,  3.33s/it]
 68%|██████▊   | 10060/14725 [9:09:44<4:21:43,  3.37s/it]
                                                         

 68%|██████▊   | 10060/14725 [9:09:44<4:21:43,  3.37s/it]
 68%|██████▊   | 10061/14725 [9:09:47<4:18:28,  3.33s/it]
 68%|██████▊   | 10062/14725 [9:09:50<4:16:23,  3.30s/it]
 68%|██████▊   | 10063/14725 [9:09:53<4:15:48,  3.29s/it]
 68%|██████▊   | 10064/147
0: {'loss': 0.6978, 'grad_norm': 0.40881571233993586, 'learning_rate': 1e-05, 'epoch': 0.68}
0: 25 [9:09:57<4:18:15,  3.32s/it]
 68%|██████▊   | 10065/14725 [9:10:00<4:19:12,  3.34s/it]
 68%|██████▊   | 10066/14725 [9:10:03<4:15:59,  3.30s/it]
 68%|██████▊   | 10067/14725 [9:10:07<4:20:34,  3.36s/it]
 68%|██████▊   | 10068/14725 [9:10:10<4:16:38,  3.31s/it]
 68%|██████▊   | 10069/14725 [9:10:13<4:14:06,  3.27s/it]
 68%|██████▊   | 10070/14725 [9:10:17<4:15:06,  3.29s/it]
                                                         

 68%|██████▊   | 10070/14725 [9:10:17<4:15:06,  3.29s/it]
 68%|██████▊   | 10071/14725 [9:10:20<4:14:32,  3.28s/it]
 68%|██████▊   | 10072/14725 [9:10:23<4:14:07,  3.28s/it]
 68%|██████▊   | 10073/14725 [9:10:26<4:12:21,  3.25s/it]
 68%|██████▊   | 10074/14725 [9:10:30<4:11:13,  3.24s/it]
 68%|██████▊   | 10075/14725 [9:10:33<4:09:47,  3.22s/it]
 68%|██████▊   | 10076/14725 [9:10:36<4:05:46,  3.17s/i
0: {'loss': 0.6925, 'grad_norm': 0.4120351970342153, 'learning_rate': 1e-05, 'epoch': 0.68}
0: t]
 68%|██████▊   | 10077/14725 [9:10:39<4:05:00,  3.16s/it]
 68%|██████▊   | 10078/14725 [9:10:42<4:08:39,  3.21s/it]
 68%|██████▊   | 10079/14725 [9:10:46<4:09:51,  3.23s/it]
 68%|██████▊   | 10080/14725 [9:10:49<4:12:14,  3.26s/it]
                                                         

 68%|██████▊   | 10080/14725 [9:10:49<4:12:14,  3.26s/it]
 68%|██████▊   | 10081/14725 [9:10:52<4:13:13,  3.27s/it]
 68%|██████▊   | 10082/14725 [9:10:55<4:10:31,  3.24s/it]
 68%|██████▊   | 10083/14725 [9:10:59<4:11:20,  3.25s/it]
 68%|██████▊   | 10084/14725 [9:11:02<4:12:12,  3.26s/it]
 68%|██████▊   | 10085/14725 [9:11:05<4:14:09,  3.29s/it]
 68%|██████▊   | 10086/14725 [9:11:08<4:14:08,  3.29s/it]
 69%|██████▊   | 10087/14725 [9:11:12<4:14:47,  3.30s/it]
 69%|██████▊   | 10088/14725 [9:11:15<4:15:10,  3.30s/it]
 69%|██████▊
0: {'loss': 0.6818, 'grad_norm': 0.408280753757046, 'learning_rate': 1e-05, 'epoch': 0.69}
0:    | 10089/14725 [9:11:18<4:14:43,  3.30s/it]
 69%|██████▊   | 10090/14725 [9:11:22<4:12:12,  3.26s/it]
                                                         

 69%|██████▊   | 10090/14725 [9:11:22<4:12:12,  3.26s/it]
 69%|██████▊   | 10091/14725 [9:11:25<4:12:58,  3.28s/it]
 69%|██████▊   | 10092/14725 [9:11:28<4:11:32,  3.26s/it]
 69%|██████▊   | 10093/14725 [9:11:31<4:10:28,  3.24s/it]
 69%|██████▊   | 10094/14725 [9:11:35<4:13:05,  3.28s/it]
 69%|██████▊   | 10095/14725 [9:11:38<4:13:01,  3.28s/it]
 69%|██████▊   | 10096/14725 [9:11:41<4:16:24,  3.32s/it]
 69%|██████▊   | 10097/14725 [9:11:45<4:18:02,  3.35s/it]
 69%|██████▊   | 10098/14725 [9:11:48<4:15:30,  3.31s/it]
 69%|██████▊   | 10099/14725 [9:11:51<4:15:22,  3.31s/it]
 69%|██████▊   | 10100/14725 [9:11:55<4:19:53,  3.37s/it]
                                                       
0: {'loss': 0.6874, 'grad_norm': 0.406899909940288, 'learning_rate': 1e-05, 'epoch': 0.69}
0: {'loss': 0.6952, 'grad_norm': 0.39905451177718015, 'learning_rate': 1e-05, 'epoch': 0.69}
0:   

 69%|██████▊   | 10100/14725 [9:11:55<4:19:53,  3.37s/it]
 69%|██████▊   | 10101/14725 [9:11:58<4:14:55,  3.31s/it]
 69%|██████▊   | 10102/14725 [9:12:01<4:14:02,  3.30s/it]
 69%|██████▊   | 10103/14725 [9:12:05<4:13:15,  3.29s/it]
 69%|██████▊   | 10104/14725 [9:12:08<4:13:04,  3.29s/it]
 69%|██████▊   | 10105/14725 [9:12:11<4:14:49,  3.31s/it]
 69%|██████▊   | 10106/14725 [9:12:14<4:13:12,  3.29s/it]
 69%|██████▊   | 10107/14725 [9:12:18<4:08:44,  3.23s/it]
 69%|██████▊   | 10108/14725 [9:12:21<4:07:06,  3.21s/it]
 69%|██████▊   | 10109/14725 [9:12:24<4:10:04,  3.25s/it]
 69%|██████▊   | 10110/14725 [9:12:27<4:11:24,  3.27s/it]
                                                         

 69%|██████▊   | 10110/14725 [9:12:27<4:11:24,  3.27s/it]
 69%|██████▊   | 10111/14725 [9:12:31<4:09:48,  3.25s/it]
 69%|██████�
0: {'loss': 0.6544, 'grad_norm': 0.4382246948492091, 'learning_rate': 1e-05, 'epoch': 0.69}
0: �   | 10112/14725 [9:12:34<4:05:45,  3.20s/it]
 69%|██████▊   | 10113/14725 [9:12:37<4:09:50,  3.25s/it]
 69%|██████▊   | 10114/14725 [9:12:40<4:08:41,  3.24s/it]
 69%|██████▊   | 10115/14725 [9:12:43<4:06:38,  3.21s/it]
 69%|██████▊   | 10116/14725 [9:12:47<4:07:33,  3.22s/it]
 69%|██████▊   | 10117/14725 [9:12:50<4:03:11,  3.17s/it]
 69%|██████▊   | 10118/14725 [9:12:53<4:04:30,  3.18s/it]
 69%|██████▊   | 10119/14725 [9:12:56<4:08:20,  3.23s/it]
 69%|██████▊   | 10120/14725 [9:13:00<4:10:31,  3.26s/it]
                                                         

 69%|██████▊   | 10120/14725 [9:13:00<4:10:31,  3.26s/it]
 69%|██████▊   | 10121/14725 [9:13:03<4:09:28,  3.25s/it]
 69%|██████▊   | 10122/14725 [9:13:06<4:06:31,  3.21s/it]
 69%|██████▊   | 10123/14725 [9:13:09<4:01:43,  3.15s/it]
 69%|██████▉   | 10124/14725 [9:13:12<4:
0: {'loss': 0.6874, 'grad_norm': 0.39494443240985283, 'learning_rate': 1e-05, 'epoch': 0.69}
0: 07:39,  3.23s/it]
 69%|██████▉   | 10125/14725 [9:13:16<4:12:36,  3.29s/it]
 69%|██████▉   | 10126/14725 [9:13:19<4:07:05,  3.22s/it]
 69%|██████▉   | 10127/14725 [9:13:22<4:03:46,  3.18s/it]
 69%|██████▉   | 10128/14725 [9:13:25<4:09:51,  3.26s/it]
 69%|██████▉   | 10129/14725 [9:13:29<4:10:37,  3.27s/it]
 69%|██████▉   | 10130/14725 [9:13:32<4:14:29,  3.32s/it]
                                                         

 69%|██████▉   | 10130/14725 [9:13:32<4:14:29,  3.32s/it]
 69%|██████▉   | 10131/14725 [9:13:35<4:10:08,  3.27s/it]
 69%|██████▉   | 10132/14725 [9:13:38<4:07:44,  3.24s/it]
 69%|██████▉   | 10133/14725 [9:13:42<4:08:33,  3.25s/it]
 69%|██████▉   | 10134/14725 [9:13:45<4:08:57,  3.25s/it]
 69%|██████▉   | 10135/14725 [9:13:48<4:04:38,  3.20s/it]
 69%|██████▉   | 10136/14725 [9:13:51<4:07:46,  3.24s/it]
 69%|██
0: {'loss': 0.6772, 'grad_norm': 0.4228114679703457, 'learning_rate': 1e-05, 'epoch': 0.69}
0: ████▉   | 10137/14725 [9:13:55<4:06:58,  3.23s/it]
 69%|██████▉   | 10138/14725 [9:13:58<4:10:11,  3.27s/it]
 69%|██████▉   | 10139/14725 [9:14:01<4:07:42,  3.24s/it]
 69%|██████▉   | 10140/14725 [9:14:04<4:06:33,  3.23s/it]
                                                         

 69%|██████▉   | 10140/14725 [9:14:04<4:06:33,  3.23s/it]
 69%|██████▉   | 10141/14725 [9:14:08<4:09:19,  3.26s/it]
 69%|██████▉   | 10142/14725 [9:14:11<4:07:38,  3.24s/it]
 69%|██████▉   | 10143/14725 [9:14:14<4:05:38,  3.22s/it]
 69%|██████▉   | 10144/14725 [9:14:17<4:08:50,  3.26s/it]
 69%|██████▉   | 10145/14725 [9:14:21<4:10:21,  3.28s/it]
 69%|██████▉   | 10146/14725 [9:14:24<4:11:50,  3.30s/it]
 69%|██████▉   | 10147/14725 [9:14:27<4:13:25,  3.32s/it]
 69%|██████▉   | 10148/14725 [9:14:31<4:16:55,  3.37s/it]
 69%|██████▉   | 10149/147
0: {'loss': 0.6705, 'grad_norm': 0.3782444151434854, 'learning_rate': 1e-05, 'epoch': 0.69}
0: {'loss': 0.6902, 'grad_norm': 0.38417220262755764, 'learning_rate': 1e-05, 'epoch': 0.69}
0: 25 [9:14:34<4:19:00,  3.40s/it]
 69%|██████▉   | 10150/14725 [9:14:38<4:15:55,  3.36s/it]
                                                         

 69%|██████▉   | 10150/14725 [9:14:38<4:15:55,  3.36s/it]
 69%|██████▉   | 10151/14725 [9:14:41<4:14:34,  3.34s/it]
 69%|██████▉   | 10152/14725 [9:14:44<4:14:52,  3.34s/it]
 69%|██████▉   | 10153/14725 [9:14:48<4:16:33,  3.37s/it]
 69%|██████▉   | 10154/14725 [9:14:51<4:14:15,  3.34s/it]
 69%|██████▉   | 10155/14725 [9:14:54<4:12:56,  3.32s/it]
 69%|██████▉   | 10156/14725 [9:14:57<4:11:33,  3.30s/it]
 69%|██████▉   | 10157/14725 [9:15:01<4:16:47,  3.37s/it]
 69%|██████▉   | 10158/14725 [9:15:04<4:15:50,  3.36s/it]
 69%|██████▉   | 10159/14725 [9:15:08<4:13:36,  3.33s/it]
 69%|██████▉   | 10160/14725 [9:15:11<4:11:30,  3.31s/it]
                                                         

 69%|█�
0: {'loss': 0.6799, 'grad_norm': 0.4261003293365945, 'learning_rate': 1e-05, 'epoch': 0.69}
0: �████▉   | 10160/14725 [9:15:11<4:11:30,  3.31s/it]
 69%|██████▉   | 10161/14725 [9:15:14<4:09:09,  3.28s/it]
 69%|██████▉   | 10162/14725 [9:15:17<4:11:22,  3.31s/it]
 69%|██████▉   | 10163/14725 [9:15:21<4:11:04,  3.30s/it]
 69%|██████▉   | 10164/14725 [9:15:24<4:09:29,  3.28s/it]
 69%|██████▉   | 10165/14725 [9:15:27<4:09:26,  3.28s/it]
 69%|██████▉   | 10166/14725 [9:15:31<4:10:03,  3.29s/it]
 69%|██████▉   | 10167/14725 [9:15:34<4:10:13,  3.29s/it]
 69%|██████▉   | 10168/14725 [9:15:37<4:10:51,  3.30s/it]
 69%|██████▉   | 10169/14725 [9:15:41<4:14:45,  3.36s/it]
 69%|██████▉   | 10170/14725 [9:15:44<4:16:42,  3.38s/it]
                                                         

 69%|██████▉   | 10170/14725 [9:15:44<4:16:42,  3.38s/it]
 69%|██████▉   | 10171/14725 [9:15:47<4:16:03,  3.37s/it]
 69%|██████▉   | 10172/14
0: {'loss': 0.6796, 'grad_norm': 0.39367919107019655, 'learning_rate': 1e-05, 'epoch': 0.69}
0: 725 [9:15:51<4:15:47,  3.37s/it]
 69%|██████▉   | 10173/14725 [9:15:54<4:12:55,  3.33s/it]
 69%|██████▉   | 10174/14725 [9:15:57<4:10:49,  3.31s/it]
 69%|██████▉   | 10175/14725 [9:16:01<4:10:17,  3.30s/it]
 69%|██████▉   | 10176/14725 [9:16:04<4:12:32,  3.33s/it]
 69%|██████▉   | 10177/14725 [9:16:07<4:11:46,  3.32s/it]
 69%|██████▉   | 10178/14725 [9:16:11<4:11:29,  3.32s/it]
 69%|██████▉   | 10179/14725 [9:16:14<4:10:19,  3.30s/it]
 69%|██████▉   | 10180/14725 [9:16:17<4:04:45,  3.23s/it]
                                                         

 69%|██████▉   | 10180/14725 [9:16:17<4:04:45,  3.23s/it]
 69%|██████▉   | 10181/14725 [9:16:20<4:01:49,  3.19s/it]
 69%|██████▉   | 10182/14725 [9:16:23<4:03:44,  3.22s/it]
 69%|██████▉   | 10183/14725 [9:16:27<4:03:34,  3.22s/it]
 69%|██████▉   | 10184/14725 [9:16:30<4:06:54,  3.26s/
0: {'loss': 0.6952, 'grad_norm': 0.4276482202225372, 'learning_rate': 1e-05, 'epoch': 0.69}
0: it]
 69%|██████▉   | 10185/14725 [9:16:33<4:07:32,  3.27s/it]
 69%|██████▉   | 10186/14725 [9:16:36<4:06:45,  3.26s/it]
 69%|██████▉   | 10187/14725 [9:16:40<4:07:17,  3.27s/it]
 69%|██████▉   | 10188/14725 [9:16:43<4:10:43,  3.32s/it]
 69%|██████▉   | 10189/14725 [9:16:46<4:10:10,  3.31s/it]
 69%|██████▉   | 10190/14725 [9:16:50<4:09:43,  3.30s/it]
                                                         

 69%|██████▉   | 10190/14725 [9:16:50<4:09:43,  3.30s/it]
 69%|██████▉   | 10191/14725 [9:16:53<4:07:33,  3.28s/it]
 69%|██████▉   | 10192/14725 [9:16:56<4:04:07,  3.23s/it]
 69%|██████▉   | 10193/14725 [9:17:00<4:09:16,  3.30s/it]
 69%|██████▉   | 10194/14725 [9:17:03<4:08:35,  3.29s/it]
 69%|██████▉   | 10195/14725 [9:17:06<4:08:59,  3.30s/it]
 69%|██████▉   | 10196/14725 [9:17:09<4:06:01,  3.26s/it]
 69%|██████�
0: {'loss': 0.6892, 'grad_norm': 0.3909675516859939, 'learning_rate': 1e-05, 'epoch': 0.69}
0: �   | 10197/14725 [9:17:13<4:09:56,  3.31s/it]
 69%|██████▉   | 10198/14725 [9:17:16<4:13:59,  3.37s/it]
 69%|██████▉   | 10199/14725 [9:17:19<4:11:46,  3.34s/it]
 69%|██████▉   | 10200/14725 [9:17:23<4:10:12,  3.32s/it]
                                                         

 69%|██████▉   | 10200/14725 [9:17:23<4:10:12,  3.32s/it]
 69%|██████▉   | 10201/14725 [9:17:26<4:14:37,  3.38s/it]
 69%|██████▉   | 10202/14725 [9:17:30<4:11:22,  3.33s/it]
 69%|██████▉   | 10203/14725 [9:17:33<4:11:10,  3.33s/it]
 69%|██████▉   | 10204/14725 [9:17:36<4:08:37,  3.30s/it]
 69%|██████▉   | 10205/14725 [9:17:39<4:09:58,  3.32s/it]
 69%|██████▉   | 10206/14725 [9:17:43<4:10:24,  3.32s/it]
 69%|██████▉   | 10207/14725 [9:17:46<4:10:57,  3.33s/it]
 69%|██████▉   | 10208/14725 [9:17:49<4:08:14,  3.30s/it]
 69%|██████▉   | 10209/14725 [9:17:53<4:
0: {'loss': 0.6844, 'grad_norm': 0.4143502457175874, 'learning_rate': 1e-05, 'epoch': 0.69}
0: {'loss': 0.6911, 'grad_norm': 0.41986168044193106, 'learning_rate': 1e-05, 'epoch': 0.69}
0: 08:20,  3.30s/it]
 69%|██████▉   | 10210/14725 [9:17:56<4:06:06,  3.27s/it]
                                                         

 69%|██████▉   | 10210/14725 [9:17:56<4:06:06,  3.27s/it]
 69%|██████▉   | 10211/14725 [9:17:59<4:06:05,  3.27s/it]
 69%|██████▉   | 10212/14725 [9:18:02<4:03:48,  3.24s/it]
 69%|██████▉   | 10213/14725 [9:18:05<4:01:56,  3.22s/it]
 69%|██████▉   | 10214/14725 [9:18:09<4:01:28,  3.21s/it]
 69%|██████▉   | 10215/14725 [9:18:12<3:58:06,  3.17s/it]
 69%|██████▉   | 10216/14725 [9:18:15<4:01:18,  3.21s/it]
 69%|██████▉   | 10217/14725 [9:18:18<4:07:11,  3.29s/it]
 69%|██████▉   | 10218/14725 [9:18:22<4:01:22,  3.21s/it]
 69%|██████▉   | 10219/14725 [9:18:25<4:00:22,  3.20s/it]
 69%|██████▉   | 10220/14725 [9:18:28<4:09:09,  3.32s/it]
                                                         

 69%|██████�
0: {'loss': 0.6794, 'grad_norm': 0.4002751974479123, 'learning_rate': 1e-05, 'epoch': 0.69}
0: ��   | 10220/14725 [9:18:28<4:09:09,  3.32s/it]
 69%|██████▉   | 10221/14725 [9:18:31<4:05:56,  3.28s/it]
 69%|██████▉   | 10222/14725 [9:18:35<4:06:01,  3.28s/it]
 69%|██████▉   | 10223/14725 [9:18:38<4:05:39,  3.27s/it]
 69%|██████▉   | 10224/14725 [9:18:41<4:08:18,  3.31s/it]
 69%|██████▉   | 10225/14725 [9:18:45<4:07:02,  3.29s/it]
 69%|██████▉   | 10226/14725 [9:18:48<4:03:59,  3.25s/it]
 69%|██████▉   | 10227/14725 [9:18:51<3:59:23,  3.19s/it]
 69%|██████▉   | 10228/14725 [9:18:54<4:04:09,  3.26s/it]
 69%|██████▉   | 10229/14725 [9:18:58<4:05:43,  3.28s/it]
 69%|██████▉   | 10230/14725 [9:19:01<4:07:55,  3.31s/it]
                                                         

 69%|██████▉   | 10230/14725 [9:19:01<4:07:55,  3.31s/it]
 69%|██████▉   | 10231/14725 [9:19:04<4:11:07,  3.35s/it]
 69%|██████▉   | 10232/14725 [9:19:08<4
0: {'loss': 0.6687, 'grad_norm': 0.38707653724020297, 'learning_rate': 1e-05, 'epoch': 0.7}
0: :07:49,  3.31s/it]
 69%|██████▉   | 10233/14725 [9:19:11<4:05:01,  3.27s/it]
 70%|██████▉   | 10234/14725 [9:19:14<4:07:16,  3.30s/it]
 70%|██████▉   | 10235/14725 [9:19:17<4:05:58,  3.29s/it]
 70%|██████▉   | 10236/14725 [9:19:21<4:02:58,  3.25s/it]
 70%|██████▉   | 10237/14725 [9:19:24<4:04:41,  3.27s/it]
 70%|██████▉   | 10238/14725 [9:19:27<4:05:27,  3.28s/it]
 70%|██████▉   | 10239/14725 [9:19:30<4:01:45,  3.23s/it]
 70%|██████▉   | 10240/14725 [9:19:34<3:59:46,  3.21s/it]
                                                         

 70%|██████▉   | 10240/14725 [9:19:34<3:59:46,  3.21s/it]
 70%|██████▉   | 10241/14725 [9:19:37<3:59:20,  3.20s/it]
 70%|██████▉   | 10242/14725 [9:19:40<3:58:27,  3.19s/it]
 70%|██████▉   | 10243/14725 [9:19:43<3:56:38,  3.17s/it]
 70%|██████▉   | 10244/14725 [9:19:46<3:58:16,  3.19s/it]
 70%|█�
0: {'loss': 0.6829, 'grad_norm': 0.4049316637925053, 'learning_rate': 1e-05, 'epoch': 0.7}
0: �████▉   | 10245/14725 [9:19:49<3:57:17,  3.18s/it]
 70%|██████▉   | 10246/14725 [9:19:52<3:54:51,  3.15s/it]
 70%|██████▉   | 10247/14725 [9:19:56<3:55:59,  3.16s/it]
 70%|██████▉   | 10248/14725 [9:19:59<3:59:15,  3.21s/it]
 70%|██████▉   | 10249/14725 [9:20:02<3:59:55,  3.22s/it]
 70%|██████▉   | 10250/14725 [9:20:06<4:02:27,  3.25s/it]
                                                         

 70%|██████▉   | 10250/14725 [9:20:06<4:02:27,  3.25s/it]
 70%|██████▉   | 10251/14725 [9:20:09<3:59:21,  3.21s/it]
 70%|██████▉   | 10252/14725 [9:20:12<4:03:25,  3.27s/it]
 70%|██████▉   | 10253/14725 [9:20:15<4:05:00,  3.29s/it]
 70%|██████▉   | 10254/14725 [9:20:19<4:12:15,  3.39s/it]
 70%|██████▉   | 10255/14725 [9:20:22<4:10:15,  3.36s/it]
 70%|██████▉   | 10256/14725 [9:20:25<4:06:23,  3.31s/it]
 70%|██████▉   | 10257/14
0: {'loss': 0.6856, 'grad_norm': 0.3918194325466413, 'learning_rate': 1e-05, 'epoch': 0.7}
0: 725 [9:20:29<4:06:29,  3.31s/it]
 70%|██████▉   | 10258/14725 [9:20:32<4:02:46,  3.26s/it]
 70%|██████▉   | 10259/14725 [9:20:35<3:58:42,  3.21s/it]
 70%|██████▉   | 10260/14725 [9:20:38<3:58:33,  3.21s/it]
                                                         

 70%|██████▉   | 10260/14725 [9:20:38<3:58:33,  3.21s/it]
 70%|██████▉   | 10261/14725 [9:20:41<3:56:37,  3.18s/it]
 70%|██████▉   | 10262/14725 [9:20:44<3:54:25,  3.15s/it]
 70%|██████▉   | 10263/14725 [9:20:48<3:57:03,  3.19s/it]
 70%|██████▉   | 10264/14725 [9:20:51<4:05:30,  3.30s/it]
 70%|██████▉   | 10265/14725 [9:20:54<4:02:22,  3.26s/it]
 70%|██████▉   | 10266/14725 [9:20:58<4:01:20,  3.25s/it]
 70%|██████▉   | 10267/14725 [9:21:01<3:58:38,  3.21s/it]
 70%|██████▉   | 10268/14725 [9:21:04<3:55:44,  3.17s/it]
 70%|██████▉   | 10269/14725 [9:21:07<3:57:28,  3.20s/
0: {'loss': 0.6982, 'grad_norm': 0.41315153669496224, 'learning_rate': 1e-05, 'epoch': 0.7}
0: {'loss': 0.678, 'grad_norm': 0.43153091780588587, 'learning_rate': 1e-05, 'epoch': 0.7}
0: it]
 70%|██████▉   | 10270/14725 [9:21:11<4:06:32,  3.32s/it]
                                                         

 70%|██████▉   | 10270/14725 [9:21:11<4:06:32,  3.32s/it]
 70%|██████▉   | 10271/14725 [9:21:14<4:04:07,  3.29s/it]
 70%|██████▉   | 10272/14725 [9:21:17<4:01:48,  3.26s/it]
 70%|██████▉   | 10273/14725 [9:21:21<4:07:27,  3.33s/it]
 70%|██████▉   | 10274/14725 [9:21:24<4:06:09,  3.32s/it]
 70%|██████▉   | 10275/14725 [9:21:27<4:03:58,  3.29s/it]
 70%|██████▉   | 10276/14725 [9:21:30<3:59:00,  3.22s/it]
 70%|██████▉   | 10277/14725 [9:21:33<3:59:39,  3.23s/it]
 70%|██████▉   | 10278/14725 [9:21:37<4:01:16,  3.26s/it]
 70%|██████▉   | 10279/14725 [9:21:40<3:56:11,  3.19s/it]
 70%|██████▉   | 10280/14725 [9:21:43<3:55:19,  3.18s/it]
                                                         

 70%|██████▉   | 10280/1
0: {'loss': 0.6831, 'grad_norm': 0.42007443684015183, 'learning_rate': 1e-05, 'epoch': 0.7}
0: 4725 [9:21:43<3:55:19,  3.18s/it]
 70%|██████▉   | 10281/14725 [9:21:46<3:59:37,  3.24s/it]
 70%|██████▉   | 10282/14725 [9:21:50<4:02:02,  3.27s/it]
 70%|██████▉   | 10283/14725 [9:21:53<4:04:08,  3.30s/it]
 70%|██████▉   | 10284/14725 [9:21:56<3:59:49,  3.24s/it]
 70%|██████▉   | 10285/14725 [9:22:00<4:02:53,  3.28s/it]
 70%|██████▉   | 10286/14725 [9:22:03<4:04:32,  3.31s/it]
 70%|██████▉   | 10287/14725 [9:22:06<4:03:15,  3.29s/it]
 70%|██████▉   | 10288/14725 [9:22:09<4:03:56,  3.30s/it]
 70%|██████▉   | 10289/14725 [9:22:13<4:01:47,  3.27s/it]
 70%|██████▉   | 10290/14725 [9:22:16<3:58:19,  3.22s/it]
                                                         

 70%|██████▉   | 10290/14725 [9:22:16<3:58:19,  3.22s/it]
 70%|██████▉   | 10291/14725 [9:22:19<3:58:34,  3.23s/it]
 70%|██████▉   | 10292/14725 [9:22:22<3:55:54,  3.19s
0: {'loss': 0.6862, 'grad_norm': 0.4205643660173953, 'learning_rate': 1e-05, 'epoch': 0.7}
0: /it]
 70%|██████▉   | 10293/14725 [9:22:25<3:57:36,  3.22s/it]
 70%|██████▉   | 10294/14725 [9:22:29<3:56:26,  3.20s/it]
 70%|██████▉   | 10295/14725 [9:22:32<3:58:01,  3.22s/it]
 70%|██████▉   | 10296/14725 [9:22:35<4:02:20,  3.28s/it]
 70%|██████▉   | 10297/14725 [9:22:39<4:02:10,  3.28s/it]
 70%|██████▉   | 10298/14725 [9:22:42<4:02:31,  3.29s/it]
 70%|██████▉   | 10299/14725 [9:22:45<4:01:43,  3.28s/it]
 70%|██████▉   | 10300/14725 [9:22:49<4:07:29,  3.36s/it]
                                                         

 70%|██████▉   | 10300/14725 [9:22:49<4:07:29,  3.36s/it]
 70%|██████▉   | 10301/14725 [9:22:52<4:06:27,  3.34s/it]
 70%|██████▉   | 10302/14725 [9:22:55<4:02:41,  3.29s/it]
 70%|██████▉   | 10303/14725 [9:22:58<4:00:56,  3.27s/it]
 70%|██████▉   | 10304/14725 [9:23:02<4:06:04,  3.34s/it]
 70%|██████�
0: {'loss': 0.7019, 'grad_norm': 0.4328877720171542, 'learning_rate': 1e-05, 'epoch': 0.7}
0: ��   | 10305/14725 [9:23:05<4:03:34,  3.31s/it]
 70%|██████▉   | 10306/14725 [9:23:08<4:05:29,  3.33s/it]
 70%|██████▉   | 10307/14725 [9:23:12<4:00:26,  3.27s/it]
 70%|███████   | 10308/14725 [9:23:15<4:02:28,  3.29s/it]
 70%|███████   | 10309/14725 [9:23:18<3:59:26,  3.25s/it]
 70%|███████   | 10310/14725 [9:23:21<4:00:10,  3.26s/it]
                                                         

 70%|███████   | 10310/14725 [9:23:21<4:00:10,  3.26s/it]
 70%|███████   | 10311/14725 [9:23:25<3:57:14,  3.22s/it]
 70%|███████   | 10312/14725 [9:23:28<3:57:32,  3.23s/it]
 70%|███████   | 10313/14725 [9:23:31<4:00:07,  3.27s/it]
 70%|███████   | 10314/14725 [9:23:34<3:57:05,  3.22s/it]
 70%|███████   | 10315/14725 [9:23:38<3:59:21,  3.26s/it]
 70%|███████   | 10316/14725 [9:23:41<3:59:20,  3.26s/it]
 70%|███████   | 10317/14725 [9:23:44<3
0: {'loss': 0.6797, 'grad_norm': 0.4068038464947319, 'learning_rate': 1e-05, 'epoch': 0.7}
0: :58:53,  3.25s/it]
 70%|███████   | 10318/14725 [9:23:47<3:59:07,  3.26s/it]
 70%|███████   | 10319/14725 [9:23:50<3:56:46,  3.22s/it]
 70%|███████   | 10320/14725 [9:23:54<4:01:28,  3.29s/it]
                                                         

 70%|███████   | 10320/14725 [9:23:54<4:01:28,  3.29s/it]
 70%|███████   | 10321/14725 [9:23:57<4:01:32,  3.29s/it]
 70%|███████   | 10322/14725 [9:24:00<3:59:51,  3.27s/it]
 70%|███████   | 10323/14725 [9:24:04<3:59:12,  3.26s/it]
 70%|███████   | 10324/14725 [9:24:07<4:00:03,  3.27s/it]
 70%|███████   | 10325/14725 [9:24:10<3:59:24,  3.26s/it]
 70%|███████   | 10326/14725 [9:24:13<3:58:46,  3.26s/it]
 70%|███████   | 10327/14725 [9:24:17<3:57:56,  3.25s/it]
 70%|███████   | 10328/14725 [9:24:20<3:59:55,  3.27s/it]
 70%|███████   | 10329/14725 [9:24:23<4:01:08,  3.29s/it]
 70%|█�
0: {'loss': 0.6846, 'grad_norm': 0.42631275603697144, 'learning_rate': 1e-05, 'epoch': 0.7}
0: {'loss': 0.6777, 'grad_norm': 0.4247244580320253, 'learning_rate': 1e-05, 'epoch': 0.7}
0: �█████   | 10330/14725 [9:24:27<4:01:01,  3.29s/it]
                                                         

 70%|███████   | 10330/14725 [9:24:27<4:01:01,  3.29s/it]
 70%|███████   | 10331/14725 [9:24:30<4:02:22,  3.31s/it]
 70%|███████   | 10332/14725 [9:24:33<4:00:16,  3.28s/it]
 70%|███████   | 10333/14725 [9:24:36<3:59:58,  3.28s/it]
 70%|███████   | 10334/14725 [9:24:40<3:59:17,  3.27s/it]
 70%|███████   | 10335/14725 [9:24:43<3:57:15,  3.24s/it]
 70%|███████   | 10336/14725 [9:24:46<3:54:54,  3.21s/it]
 70%|███████   | 10337/14725 [9:24:49<3:59:55,  3.28s/it]
 70%|███████   | 10338/14725 [9:24:53<4:00:59,  3.30s/it]
 70%|███████   | 10339/14725 [9:24:56<4:00:30,  3.29s/it]
 70%|███████   | 10340/14725 [9:24:59<4:01:29,  3.30s/it]
                                                         

 70%|███████   | 10340/14725 [9:24:59<
0: {'loss': 0.7009, 'grad_norm': 0.43786517326211766, 'learning_rate': 1e-05, 'epoch': 0.7}
0: 4:01:29,  3.30s/it]
 70%|███████   | 10341/14725 [9:25:03<4:02:41,  3.32s/it]
 70%|███████   | 10342/14725 [9:25:06<4:01:42,  3.31s/it]
 70%|███████   | 10343/14725 [9:25:10<4:06:42,  3.38s/it]
 70%|███████   | 10344/14725 [9:25:13<4:04:59,  3.36s/it]
 70%|███████   | 10345/14725 [9:25:16<4:00:43,  3.30s/it]
 70%|███████   | 10346/14725 [9:25:19<3:57:14,  3.25s/it]
 70%|███████   | 10347/14725 [9:25:23<3:58:44,  3.27s/it]
 70%|███████   | 10348/14725 [9:25:26<3:57:08,  3.25s/it]
 70%|███████   | 10349/14725 [9:25:29<3:55:43,  3.23s/it]
 70%|███████   | 10350/14725 [9:25:32<3:56:13,  3.24s/it]
                                                         

 70%|███████   | 10350/14725 [9:25:32<3:56:13,  3.24s/it]
 70%|███████   | 10351/14725 [9:25:36<4:02:35,  3.33s/it]
 70%|███████   | 10352/14725 [9:25:39<4:02:04,  3.32s/it]
 70%|█�
0: {'loss': 0.6919, 'grad_norm': 0.42917451575646753, 'learning_rate': 1e-05, 'epoch': 0.7}
0: ��█████   | 10353/14725 [9:25:42<3:59:59,  3.29s/it]
 70%|███████   | 10354/14725 [9:25:45<3:55:00,  3.23s/it]
 70%|███████   | 10355/14725 [9:25:49<3:56:11,  3.24s/it]
 70%|███████   | 10356/14725 [9:25:52<3:54:03,  3.21s/it]
 70%|███████   | 10357/14725 [9:25:55<3:57:47,  3.27s/it]
 70%|███████   | 10358/14725 [9:25:58<3:56:45,  3.25s/it]
 70%|███████   | 10359/14725 [9:26:02<3:56:56,  3.26s/it]
 70%|███████   | 10360/14725 [9:26:05<3:58:08,  3.27s/it]
                                                         

 70%|███████   | 10360/14725 [9:26:05<3:58:08,  3.27s/it]
 70%|███████   | 10361/14725 [9:26:08<3:56:57,  3.26s/it]
 70%|███████   | 10362/14725 [9:26:12<3:58:28,  3.28s/it]
 70%|███████   | 10363/14725 [9:26:15<3:58:40,  3.28s/it]
 70%|███████   | 10364/14725 [9:26:18<3:56:04,  3.25s/it]
 70%|███████   | 10365/1
0: {'loss': 0.6979, 'grad_norm': 0.4261622896040308, 'learning_rate': 1e-05, 'epoch': 0.7}
0: 4725 [9:26:21<4:01:46,  3.33s/it]
 70%|███████   | 10366/14725 [9:26:25<3:58:20,  3.28s/it]
 70%|███████   | 10367/14725 [9:26:28<3:59:40,  3.30s/it]
 70%|███████   | 10368/14725 [9:26:31<3:57:12,  3.27s/it]
 70%|███████   | 10369/14725 [9:26:34<3:53:20,  3.21s/it]
 70%|███████   | 10370/14725 [9:26:38<3:57:54,  3.28s/it]
                                                         

 70%|███████   | 10370/14725 [9:26:38<3:57:54,  3.28s/it]
 70%|███████   | 10371/14725 [9:26:41<3:54:55,  3.24s/it]
 70%|███████   | 10372/14725 [9:26:44<3:59:37,  3.30s/it]
 70%|███████   | 10373/14725 [9:26:48<3:59:09,  3.30s/it]
 70%|███████   | 10374/14725 [9:26:51<3:58:08,  3.28s/it]
 70%|███████   | 10375/14725 [9:26:54<3:58:46,  3.29s/it]
 70%|███████   | 10376/14725 [9:26:58<4:00:43,  3.32s/it]
 70%|███████   | 10377/14725 [9:27:01<4:01:28,  3.33s
0: {'loss': 0.6881, 'grad_norm': 0.44324700017955027, 'learning_rate': 1e-05, 'epoch': 0.7}
0: /it]
 70%|███████   | 10378/14725 [9:27:04<3:59:47,  3.31s/it]
 70%|███████   | 10379/14725 [9:27:08<4:00:51,  3.33s/it]
 70%|███████   | 10380/14725 [9:27:11<4:05:56,  3.40s/it]
                                                         

 70%|███████   | 10380/14725 [9:27:11<4:05:56,  3.40s/it]
 70%|███████   | 10381/14725 [9:27:14<4:02:43,  3.35s/it]
 71%|███████   | 10382/14725 [9:27:17<3:58:26,  3.29s/it]
 71%|███████   | 10383/14725 [9:27:21<4:00:43,  3.33s/it]
 71%|███████   | 10384/14725 [9:27:24<3:59:02,  3.30s/it]
 71%|███████   | 10385/14725 [9:27:28<4:03:10,  3.36s/it]
 71%|███████   | 10386/14725 [9:27:31<3:58:49,  3.30s/it]
 71%|███████   | 10387/14725 [9:27:34<3:56:20,  3.27s/it]
 71%|███████   | 10388/14725 [9:27:37<4:00:30,  3.33s/it]
 71%|███████   | 10389/14725 [9:27:41<3:55:58,  3.27s/it]
 71%|██████�
0: {'loss': 0.6735, 'grad_norm': 0.42586605196320176, 'learning_rate': 1e-05, 'epoch': 0.71}
0: {'loss': 0.6786, 'grad_norm': 0.4173885734498904, 'learning_rate': 1e-05, 'epoch': 0.71}
0: ��   | 10390/14725 [9:27:44<3:59:09,  3.31s/it]
                                                         

 71%|███████   | 10390/14725 [9:27:44<3:59:09,  3.31s/it]
 71%|███████   | 10391/14725 [9:27:47<4:00:24,  3.33s/it]
 71%|███████   | 10392/14725 [9:27:51<3:58:53,  3.31s/it]
 71%|███████   | 10393/14725 [9:27:54<3:59:33,  3.32s/it]
 71%|███████   | 10394/14725 [9:27:57<3:59:03,  3.31s/it]
 71%|███████   | 10395/14725 [9:28:00<3:56:36,  3.28s/it]
 71%|███████   | 10396/14725 [9:28:04<4:04:00,  3.38s/it]
 71%|███████   | 10397/14725 [9:28:07<4:00:50,  3.34s/it]
 71%|███████   | 10398/14725 [9:28:11<3:59:31,  3.32s/it]
 71%|███████   | 10399/14725 [9:28:14<4:01:56,  3.36s/it]
 71%|███████   | 10400/14725 [9:28:17<4:04:00,  3.39s/it]
                                                         

 71%|███████   | 10400/14725 [9:28:17<4:04:00,  3.39
0: {'loss': 0.6875, 'grad_norm': 0.41349226470310846, 'learning_rate': 1e-05, 'epoch': 0.71}
0: s/it]
 71%|███████   | 10401/14725 [9:28:21<4:02:20,  3.36s/it]
 71%|███████   | 10402/14725 [9:28:24<3:57:16,  3.29s/it]
 71%|███████   | 10403/14725 [9:28:27<3:54:37,  3.26s/it]
 71%|███████   | 10404/14725 [9:28:30<3:53:28,  3.24s/it]
 71%|███████   | 10405/14725 [9:28:34<3:52:43,  3.23s/it]
 71%|███████   | 10406/14725 [9:28:37<3:56:30,  3.29s/it]
 71%|███████   | 10407/14725 [9:28:40<3:51:42,  3.22s/it]
 71%|███████   | 10408/14725 [9:28:43<3:50:13,  3.20s/it]
 71%|███████   | 10409/14725 [9:28:47<3:58:21,  3.31s/it]
 71%|███████   | 10410/14725 [9:28:50<3:55:16,  3.27s/it]
                                                         

 71%|███████   | 10410/14725 [9:28:50<3:55:16,  3.27s/it]
 71%|███████   | 10411/14725 [9:28:53<3:58:48,  3.32s/it]
 71%|███████   | 10412/14725 [9:28:57<3:58:48,  3.32s/it]
 71%|██████
0: {'loss': 0.7045, 'grad_norm': 0.4266054997822586, 'learning_rate': 1e-05, 'epoch': 0.71}
0: █   | 10413/14725 [9:29:00<4:01:08,  3.36s/it]
 71%|███████   | 10414/14725 [9:29:03<4:00:26,  3.35s/it]
 71%|███████   | 10415/14725 [9:29:07<3:55:19,  3.28s/it]
 71%|███████   | 10416/14725 [9:29:10<3:53:36,  3.25s/it]
 71%|███████   | 10417/14725 [9:29:13<3:56:45,  3.30s/it]
 71%|███████   | 10418/14725 [9:29:16<3:54:55,  3.27s/it]
 71%|███████   | 10419/14725 [9:29:20<3:56:08,  3.29s/it]
 71%|███████   | 10420/14725 [9:29:23<3:56:47,  3.30s/it]
                                                         

 71%|███████   | 10420/14725 [9:29:23<3:56:47,  3.30s/it]
 71%|███████   | 10421/14725 [9:29:26<3:51:13,  3.22s/it]
 71%|███████   | 10422/14725 [9:29:29<3:52:34,  3.24s/it]
 71%|███████   | 10423/14725 [9:29:33<3:51:03,  3.22s/it]
 71%|███████   | 10424/14725 [9:29:36<3:55:59,  3.29s/it]
 71%|███████   | 10425/14725 [9:29:39<
0: {'loss': 0.6763, 'grad_norm': 0.40673433796862585, 'learning_rate': 1e-05, 'epoch': 0.71}
0: 3:54:20,  3.27s/it]
 71%|███████   | 10426/14725 [9:29:42<3:53:52,  3.26s/it]
 71%|███████   | 10427/14725 [9:29:46<3:54:32,  3.27s/it]
 71%|███████   | 10428/14725 [9:29:49<3:53:46,  3.26s/it]
 71%|███████   | 10429/14725 [9:29:52<3:52:18,  3.24s/it]
 71%|███████   | 10430/14725 [9:29:55<3:50:10,  3.22s/it]
                                                         

 71%|███████   | 10430/14725 [9:29:55<3:50:10,  3.22s/it]
 71%|███████   | 10431/14725 [9:29:59<3:49:41,  3.21s/it]
 71%|███████   | 10432/14725 [9:30:02<3:48:28,  3.19s/it]
 71%|███████   | 10433/14725 [9:30:05<3:49:44,  3.21s/it]
 71%|███████   | 10434/14725 [9:30:08<3:50:54,  3.23s/it]
 71%|███████   | 10435/14725 [9:30:12<3:53:32,  3.27s/it]
 71%|███████   | 10436/14725 [9:30:15<3:56:56,  3.31s/it]
 71%|███████   | 10437/14725 [9:30:18<3:59:36,  3.35s/it]
 71%|█�
0: {'loss': 0.6756, 'grad_norm': 0.3944160319011433, 'learning_rate': 1e-05, 'epoch': 0.71}
0: ��█████   | 10438/14725 [9:30:22<3:55:05,  3.29s/it]
 71%|███████   | 10439/14725 [9:30:25<3:56:00,  3.30s/it]
 71%|███████   | 10440/14725 [9:30:28<3:58:23,  3.34s/it]
                                                         

 71%|███████   | 10440/14725 [9:30:28<3:58:23,  3.34s/it]
 71%|███████   | 10441/14725 [9:30:32<3:56:27,  3.31s/it]
 71%|███████   | 10442/14725 [9:30:35<3:56:16,  3.31s/it]
 71%|███████   | 10443/14725 [9:30:38<3:50:11,  3.23s/it]
 71%|███████   | 10444/14725 [9:30:41<3:53:16,  3.27s/it]
 71%|███████   | 10445/14725 [9:30:45<3:52:22,  3.26s/it]
 71%|███████   | 10446/14725 [9:30:48<3:52:40,  3.26s/it]
 71%|███████   | 10447/14725 [9:30:51<3:49:39,  3.22s/it]
 71%|███████   | 10448/14725 [9:30:54<3:51:43,  3.25s/it]
 71%|███████   | 10449/14725 [9:30:58<3:52:24,  3.26s/it]
 71%|███████   | 10450/1
0: {'loss': 0.6927, 'grad_norm': 0.40806471281943524, 'learning_rate': 1e-05, 'epoch': 0.71}
0: {'loss': 0.6709, 'grad_norm': 0.44666029093789766, 'learning_rate': 1e-05, 'epoch': 0.71}
0: 4725 [9:31:01<3:54:31,  3.29s/it]
                                                         

 71%|███████   | 10450/14725 [9:31:01<3:54:31,  3.29s/it]
 71%|███████   | 10451/14725 [9:31:04<3:55:06,  3.30s/it]
 71%|███████   | 10452/14725 [9:31:07<3:49:37,  3.22s/it]
 71%|███████   | 10453/14725 [9:31:11<3:53:37,  3.28s/it]
 71%|███████   | 10454/14725 [9:31:14<3:50:09,  3.23s/it]
 71%|███████   | 10455/14725 [9:31:17<3:54:30,  3.30s/it]
 71%|███████   | 10456/14725 [9:31:20<3:50:15,  3.24s/it]
 71%|███████   | 10457/14725 [9:31:23<3:49:11,  3.22s/it]
 71%|███████   | 10458/14725 [9:31:27<3:51:24,  3.25s/it]
 71%|███████   | 10459/14725 [9:31:30<3:50:29,  3.24s/it]
 71%|███████   | 10460/14725 [9:31:33<3:52:26,  3.27s/it]
                                                         

 71%|███████   | 10460/14725 [9:31:33<3:52:26,  3.27s/it]
 71%|█
0: {'loss': 0.7075, 'grad_norm': 0.41218422793581166, 'learning_rate': 1e-05, 'epoch': 0.71}
0: ██████   | 10461/14725 [9:31:37<3:51:08,  3.25s/it]
 71%|███████   | 10462/14725 [9:31:40<3:53:20,  3.28s/it]
 71%|███████   | 10463/14725 [9:31:43<3:53:07,  3.28s/it]
 71%|███████   | 10464/14725 [9:31:47<3:55:22,  3.31s/it]
 71%|███████   | 10465/14725 [9:31:50<3:53:56,  3.29s/it]
 71%|███████   | 10466/14725 [9:31:53<3:54:45,  3.31s/it]
 71%|███████   | 10467/14725 [9:31:56<3:53:07,  3.28s/it]
 71%|███████   | 10468/14725 [9:32:00<3:52:04,  3.27s/it]
 71%|███████   | 10469/14725 [9:32:03<3:59:23,  3.37s/it]
 71%|███████   | 10470/14725 [9:32:07<3:59:29,  3.38s/it]
                                                         

 71%|███████   | 10470/14725 [9:32:07<3:59:29,  3.38s/it]
 71%|███████   | 10471/14725 [9:32:10<3:58:04,  3.36s/it]
 71%|███████   | 10472/14725 [9:32:13<3:51:40,  3.27s/it]
 71%|███████   | 10473/
0: {'loss': 0.6638, 'grad_norm': 0.39962185008879786, 'learning_rate': 1e-05, 'epoch': 0.71}
0: 14725 [9:32:17<3:56:39,  3.34s/it]
 71%|███████   | 10474/14725 [9:32:20<3:53:08,  3.29s/it]
 71%|███████   | 10475/14725 [9:32:23<3:49:48,  3.24s/it]
 71%|███████   | 10476/14725 [9:32:26<3:50:20,  3.25s/it]
 71%|███████   | 10477/14725 [9:32:29<3:50:35,  3.26s/it]
 71%|███████   | 10478/14725 [9:32:33<3:51:32,  3.27s/it]
 71%|███████   | 10479/14725 [9:32:36<3:50:33,  3.26s/it]
 71%|███████   | 10480/14725 [9:32:39<3:51:42,  3.28s/it]
                                                         

 71%|███████   | 10480/14725 [9:32:39<3:51:42,  3.28s/it]
 71%|███████   | 10481/14725 [9:32:43<3:56:05,  3.34s/it]
 71%|███████   | 10482/14725 [9:32:46<3:54:30,  3.32s/it]
 71%|███████   | 10483/14725 [9:32:49<3:54:50,  3.32s/it]
 71%|███████   | 10484/14725 [9:32:53<3:53:01,  3.30s/it]
 71%|███████   | 10485/14725 [9:32:56<3:50:41,  3.26
0: {'loss': 0.6835, 'grad_norm': 0.42721003986167994, 'learning_rate': 1e-05, 'epoch': 0.71}
0: s/it]
 71%|███████   | 10486/14725 [9:32:59<3:46:27,  3.21s/it]
 71%|███████   | 10487/14725 [9:33:02<3:45:18,  3.19s/it]
 71%|███████   | 10488/14725 [9:33:05<3:48:19,  3.23s/it]
 71%|███████   | 10489/14725 [9:33:09<3:52:05,  3.29s/it]
 71%|███████   | 10490/14725 [9:33:12<3:54:03,  3.32s/it]
                                                         

 71%|███████   | 10490/14725 [9:33:12<3:54:03,  3.32s/it]
 71%|███████   | 10491/14725 [9:33:15<3:53:19,  3.31s/it]
 71%|███████▏  | 10492/14725 [9:33:19<3:50:26,  3.27s/it]
 71%|███████▏  | 10493/14725 [9:33:22<3:51:28,  3.28s/it]
 71%|███████▏  | 10494/14725 [9:33:25<3:56:18,  3.35s/it]
 71%|███████▏  | 10495/14725 [9:33:29<3:52:02,  3.29s/it]
 71%|███████▏  | 10496/14725 [9:33:32<3:51:05,  3.28s/it]
 71%|███████▏  | 10497/14725 [9:33:35<3:51:30,  3.29s/it]
 71%|██
0: {'loss': 0.6813, 'grad_norm': 0.4445424067375959, 'learning_rate': 1e-05, 'epoch': 0.71}
0: █████▏  | 10498/14725 [9:33:38<3:49:37,  3.26s/it]
 71%|███████▏  | 10499/14725 [9:33:41<3:46:54,  3.22s/it]
 71%|███████▏  | 10500/14725 [9:33:45<3:50:59,  3.28s/it]
                                                         

 71%|███████▏  | 10500/14725 [9:33:45<3:50:59,  3.28s/it]
 71%|███████▏  | 10501/14725 [9:33:48<3:52:39,  3.30s/it]
 71%|███████▏  | 10502/14725 [9:33:52<3:54:49,  3.34s/it]
 71%|███████▏  | 10503/14725 [9:33:55<3:52:45,  3.31s/it]
 71%|███████▏  | 10504/14725 [9:33:58<3:50:50,  3.28s/it]
 71%|███████▏  | 10505/14725 [9:34:01<3:48:56,  3.26s/it]
 71%|███████▏  | 10506/14725 [9:34:05<3:49:00,  3.26s/it]
 71%|███████▏  | 10507/14725 [9:34:08<3:46:40,  3.22s/it]
 71%|███████▏  | 10508/14725 [9:34:11<3:44:21,  3.19s/it]
 71%|███████▏  | 10509/14725 [9:34:14<3:42:02,  3.16s/it]
 71%|███
0: {'loss': 0.674, 'grad_norm': 0.4147459673063168, 'learning_rate': 1e-05, 'epoch': 0.71}
0: {'loss': 0.6709, 'grad_norm': 0.41512311509108785, 'learning_rate': 1e-05, 'epoch': 0.71}
0: ████▏  | 10510/14725 [9:34:17<3:44:14,  3.19s/it]
                                                         

 71%|███████▏  | 10510/14725 [9:34:17<3:44:14,  3.19s/it]
 71%|███████▏  | 10511/14725 [9:34:21<3:47:28,  3.24s/it]
 71%|███████▏  | 10512/14725 [9:34:24<3:49:01,  3.26s/it]
 71%|███████▏  | 10513/14725 [9:34:27<3:48:23,  3.25s/it]
 71%|███████▏  | 10514/14725 [9:34:30<3:51:42,  3.30s/it]
 71%|███████▏  | 10515/14725 [9:34:34<3:48:28,  3.26s/it]
 71%|███████▏  | 10516/14725 [9:34:37<3:49:06,  3.27s/it]
 71%|███████▏  | 10517/14725 [9:34:40<3:49:54,  3.28s/it]
 71%|███████▏  | 10518/14725 [9:34:44<3:51:45,  3.31s/it]
 71%|███████▏  | 10519/14725 [9:34:47<3:51:18,  3.30s/it]
 71%|███████▏  | 10520/14725 [9:34:50<3:48:12,  3.26s/it]
                                                         

 71%|███████▏  |
0: {'loss': 0.6824, 'grad_norm': 0.41597925767540156, 'learning_rate': 1e-05, 'epoch': 0.72}
0:  10520/14725 [9:34:50<3:48:12,  3.26s/it]
 71%|███████▏  | 10521/14725 [9:34:53<3:51:27,  3.30s/it]
 71%|███████▏  | 10522/14725 [9:34:57<3:58:24,  3.40s/it]
 71%|███████▏  | 10523/14725 [9:35:00<3:55:13,  3.36s/it]
 71%|███████▏  | 10524/14725 [9:35:04<3:51:48,  3.31s/it]
 71%|███████▏  | 10525/14725 [9:35:07<3:48:14,  3.26s/it]
 71%|███████▏  | 10526/14725 [9:35:10<3:51:02,  3.30s/it]
 71%|███████▏  | 10527/14725 [9:35:13<3:48:51,  3.27s/it]
 71%|███████▏  | 10528/14725 [9:35:17<3:48:53,  3.27s/it]
 72%|███████▏  | 10529/14725 [9:35:20<3:46:12,  3.23s/it]
 72%|███████▏  | 10530/14725 [9:35:23<3:45:08,  3.22s/it]
                                                         

 72%|███████▏  | 10530/14725 [9:35:23<3:45:08,  3.22s/it]
 72%|███████▏  | 10531/14725 [9:35:26<3:47:14,  3.25s/it]
 72%|███████▏  | 10
0: {'loss': 0.6874, 'grad_norm': 0.4164522802909403, 'learning_rate': 1e-05, 'epoch': 0.72}
0: 532/14725 [9:35:29<3:45:22,  3.23s/it]
 72%|███████▏  | 10533/14725 [9:35:32<3:41:35,  3.17s/it]
 72%|███████▏  | 10534/14725 [9:35:36<3:42:33,  3.19s/it]
 72%|███████▏  | 10535/14725 [9:35:39<3:44:54,  3.22s/it]
 72%|███████▏  | 10536/14725 [9:35:42<3:45:25,  3.23s/it]
 72%|███████▏  | 10537/14725 [9:35:45<3:44:46,  3.22s/it]
 72%|███████▏  | 10538/14725 [9:35:48<3:42:35,  3.19s/it]
 72%|███████▏  | 10539/14725 [9:35:52<3:42:35,  3.19s/it]
 72%|███████▏  | 10540/14725 [9:35:55<3:44:00,  3.21s/it]
                                                         

 72%|███████▏  | 10540/14725 [9:35:55<3:44:00,  3.21s/it]
 72%|███████▏  | 10541/14725 [9:35:58<3:43:09,  3.20s/it]
 72%|███████▏  | 10542/14725 [9:36:02<3:50:22,  3.30s/it]
 72%|███████▏  | 10543/14725 [9:36:05<3:51:04,  3.32s/it]
 72%|███████▏  | 10544
0: {'loss': 0.6929, 'grad_norm': 0.4167232035118996, 'learning_rate': 1e-05, 'epoch': 0.72}
0: /14725 [9:36:08<3:51:26,  3.32s/it]
 72%|███████▏  | 10545/14725 [9:36:12<3:49:28,  3.29s/it]
 72%|███████▏  | 10546/14725 [9:36:15<3:55:06,  3.38s/it]
 72%|███████▏  | 10547/14725 [9:36:18<3:52:46,  3.34s/it]
 72%|███████▏  | 10548/14725 [9:36:22<3:47:47,  3.27s/it]
 72%|███████▏  | 10549/14725 [9:36:25<3:46:08,  3.25s/it]
 72%|███████▏  | 10550/14725 [9:36:28<3:46:57,  3.26s/it]
                                                         

 72%|███████▏  | 10550/14725 [9:36:28<3:46:57,  3.26s/it]
 72%|███████▏  | 10551/14725 [9:36:32<3:53:00,  3.35s/it]
 72%|███████▏  | 10552/14725 [9:36:35<3:55:41,  3.39s/it]
 72%|███████▏  | 10553/14725 [9:36:38<3:56:45,  3.41s/it]
 72%|███████▏  | 10554/14725 [9:36:42<3:55:00,  3.38s/it]
 72%|███████▏  | 10555/14725 [9:36:45<3:49:08,  3.30s/it]
 72%|███████▏  | 10556/14
0: {'loss': 0.6821, 'grad_norm': 0.4343546560144951, 'learning_rate': 1e-05, 'epoch': 0.72}
0: 725 [9:36:48<3:47:48,  3.28s/it]
 72%|███████▏  | 10557/14725 [9:36:51<3:45:15,  3.24s/it]
 72%|███████▏  | 10558/14725 [9:36:55<3:44:56,  3.24s/it]
 72%|███████▏  | 10559/14725 [9:36:58<3:42:40,  3.21s/it]
 72%|███████▏  | 10560/14725 [9:37:01<3:41:40,  3.19s/it]
                                                         

 72%|███████▏  | 10560/14725 [9:37:01<3:41:40,  3.19s/it]
 72%|███████▏  | 10561/14725 [9:37:04<3:42:07,  3.20s/it]
 72%|███████▏  | 10562/14725 [9:37:07<3:42:57,  3.21s/it]
 72%|███████▏  | 10563/14725 [9:37:10<3:40:24,  3.18s/it]
 72%|███████▏  | 10564/14725 [9:37:13<3:38:11,  3.15s/it]
 72%|███████▏  | 10565/14725 [9:37:17<3:39:02,  3.16s/it]
 72%|███████▏  | 10566/14725 [9:37:20<3:41:45,  3.20s/it]
 72%|███████▏  | 10567/14725 [9:37:23<3:42:54,  3.22s/it]
 72%|███████▏  | 10568/14725
0: {'loss': 0.6818, 'grad_norm': 0.3815574590611952, 'learning_rate': 1e-05, 'epoch': 0.72}
0:  [9:37:27<3:45:45,  3.26s/it]
 72%|███████▏  | 10569/14725 [9:37:30<3:42:47,  3.22s/it]
 72%|███████▏  | 10570/14725 [9:37:33<3:43:00,  3.22s/it]
                                                         

 72%|███████▏  | 10570/14725 [9:37:33<3:43:00,  3.22s/it]
 72%|███████▏  | 10571/14725 [9:37:36<3:50:06,  3.32s/it]
 72%|███████▏  | 10572/14725 [9:37:40<3:49:04,  3.31s/it]
 72%|███████▏  | 10573/14725 [9:37:43<3:46:16,  3.27s/it]
 72%|███████▏  | 10574/14725 [9:37:46<3:47:21,  3.29s/it]
 72%|███████▏  | 10575/14725 [9:37:49<3:45:34,  3.26s/it]
 72%|███████▏  | 10576/14725 [9:37:53<3:42:53,  3.22s/it]
 72%|███████▏  | 10577/14725 [9:37:56<3:43:31,  3.23s/it]
 72%|███████▏  | 10578/14725 [9:37:59<3:43:05,  3.23s/it]
 72%|███████▏  | 10579/14725 [9:38:02<3:43:58,  3.24s/it]
 72%|███████▏  | 10580/14725 [9
0: {'loss': 0.6696, 'grad_norm': 0.39925429586061095, 'learning_rate': 1e-05, 'epoch': 0.72}
0: {'loss': 0.6849, 'grad_norm': 0.40139363881734347, 'learning_rate': 1e-05, 'epoch': 0.72}
0: :38:06<3:52:44,  3.37s/it]
                                                         

 72%|███████▏  | 10580/14725 [9:38:06<3:52:44,  3.37s/it]
 72%|███████▏  | 10581/14725 [9:38:09<3:52:03,  3.36s/it]
 72%|███████▏  | 10582/14725 [9:38:13<3:55:33,  3.41s/it]
 72%|███████▏  | 10583/14725 [9:38:16<3:55:16,  3.41s/it]
 72%|███████▏  | 10584/14725 [9:38:20<3:57:23,  3.44s/it]
 72%|███████▏  | 10585/14725 [9:38:23<3:51:13,  3.35s/it]
 72%|███████▏  | 10586/14725 [9:38:26<3:46:08,  3.28s/it]
 72%|███████▏  | 10587/14725 [9:38:29<3:44:31,  3.26s/it]
 72%|███████▏  | 10588/14725 [9:38:32<3:41:22,  3.21s/it]
 72%|███████▏  | 10589/14725 [9:38:36<3:45:24,  3.27s/it]
 72%|███████▏  | 10590/14725 [9:38:39<3:42:48,  3.23s/it]
                                                         

 72%|███████▏  | 10590/14725 [9:38:39<3:42:48,  3
0: {'loss': 0.6793, 'grad_norm': 0.4335462218539929, 'learning_rate': 1e-05, 'epoch': 0.72}
0: .23s/it]
 72%|███████▏  | 10591/14725 [9:38:42<3:44:29,  3.26s/it]
 72%|███████▏  | 10592/14725 [9:38:45<3:42:46,  3.23s/it]
 72%|███████▏  | 10593/14725 [9:38:49<3:40:53,  3.21s/it]
 72%|███████▏  | 10594/14725 [9:38:52<3:39:25,  3.19s/it]
 72%|███████▏  | 10595/14725 [9:38:55<3:43:12,  3.24s/it]
 72%|███████▏  | 10596/14725 [9:38:58<3:42:02,  3.23s/it]
 72%|███████▏  | 10597/14725 [9:39:01<3:41:05,  3.21s/it]
 72%|███████▏  | 10598/14725 [9:39:05<3:41:54,  3.23s/it]
 72%|███████▏  | 10599/14725 [9:39:08<3:42:09,  3.23s/it]
 72%|███████▏  | 10600/14725 [9:39:11<3:38:31,  3.18s/it]
                                                         

 72%|███████▏  | 10600/14725 [9:39:11<3:38:31,  3.18s/it]
 72%|███████▏  | 10601/14725 [9:39:14<3:38:32,  3.18s/it]
 72%|███████▏  | 10602/14725 [9:39:17<3:39:17,  3.19
0: {'loss': 0.6917, 'grad_norm': 0.3842890304402938, 'learning_rate': 1e-05, 'epoch': 0.72}
0: s/it]
 72%|███████▏  | 10603/14725 [9:39:20<3:37:19,  3.16s/it]
 72%|███████▏  | 10604/14725 [9:39:24<3:39:49,  3.20s/it]
 72%|███████▏  | 10605/14725 [9:39:27<3:40:23,  3.21s/it]
 72%|███████▏  | 10606/14725 [9:39:30<3:38:31,  3.18s/it]
 72%|███████▏  | 10607/14725 [9:39:33<3:40:56,  3.22s/it]
 72%|███████▏  | 10608/14725 [9:39:37<3:40:27,  3.21s/it]
 72%|███████▏  | 10609/14725 [9:39:40<3:42:35,  3.24s/it]
 72%|███████▏  | 10610/14725 [9:39:43<3:39:46,  3.20s/it]
                                                         

 72%|███████▏  | 10610/14725 [9:39:43<3:39:46,  3.20s/it]
 72%|███████▏  | 10611/14725 [9:39:47<3:46:48,  3.31s/it]
 72%|███████▏  | 10612/14725 [9:39:50<3:50:01,  3.36s/it]
 72%|███████▏  | 10613/14725 [9:39:53<3:44:00,  3.27s/it]
 72%|███████▏  | 10614/14725 [9:39:56<3:45:44,  3.29s/i
0: {'loss': 0.6949, 'grad_norm': 0.4356998513131384, 'learning_rate': 1e-05, 'epoch': 0.72}
0: t]
 72%|███████▏  | 10615/14725 [9:40:00<3:44:20,  3.28s/it]
 72%|███████▏  | 10616/14725 [9:40:03<3:43:58,  3.27s/it]
 72%|███████▏  | 10617/14725 [9:40:06<3:40:57,  3.23s/it]
 72%|███████▏  | 10618/14725 [9:40:09<3:39:15,  3.20s/it]
 72%|███████▏  | 10619/14725 [9:40:12<3:39:11,  3.20s/it]
 72%|███████▏  | 10620/14725 [9:40:16<3:42:58,  3.26s/it]
                                                         

 72%|███████▏  | 10620/14725 [9:40:16<3:42:58,  3.26s/it]
 72%|███████▏  | 10621/14725 [9:40:19<3:45:04,  3.29s/it]
 72%|███████▏  | 10622/14725 [9:40:22<3:41:08,  3.23s/it]
 72%|███████▏  | 10623/14725 [9:40:26<3:42:34,  3.26s/it]
 72%|███████▏  | 10624/14725 [9:40:29<3:48:10,  3.34s/it]
 72%|███████▏  | 10625/14725 [9:40:32<3:45:41,  3.30s/it]
 72%|███████▏  | 10626/14725 [9:40:36<3:44:11,  3.28s/it]
0: {'loss': 0.6862, 'grad_norm': 0.41844897174227036, 'learning_rate': 1e-05, 'epoch': 0.72}
0:  72%|███████▏  | 10627/14725 [9:40:39<3:55:10,  3.44s/it]
 72%|███████▏  | 10628/14725 [9:40:43<3:51:25,  3.39s/it]
 72%|███████▏  | 10629/14725 [9:40:46<3:48:49,  3.35s/it]
 72%|███████▏  | 10630/14725 [9:40:49<3:49:32,  3.36s/it]
                                                         

 72%|███████▏  | 10630/14725 [9:40:49<3:49:32,  3.36s/it]
 72%|███████▏  | 10631/14725 [9:40:53<3:47:18,  3.33s/it]
 72%|███████▏  | 10632/14725 [9:40:56<3:46:43,  3.32s/it]
 72%|███████▏  | 10633/14725 [9:40:59<3:40:38,  3.24s/it]
 72%|███████▏  | 10634/14725 [9:41:02<3:41:43,  3.25s/it]
 72%|███████▏  | 10635/14725 [9:41:05<3:42:17,  3.26s/it]
 72%|███████▏  | 10636/14725 [9:41:09<3:43:37,  3.28s/it]
 72%|███████▏  | 10637/14725 [9:41:12<3:41:16,  3.25s/it]
 72%|███████▏  | 10638/14725 [9:41:15<3:38:08,  3.20s/it]
 72
0: {'loss': 0.6814, 'grad_norm': 0.4199145355709904, 'learning_rate': 1e-05, 'epoch': 0.72}
0: %|███████▏  | 10639/14725 [9:41:19<3:44:46,  3.30s/it]
 72%|███████▏  | 10640/14725 [9:41:22<3:44:45,  3.30s/it]
                                                         

 72%|███████▏  | 10640/14725 [9:41:22<3:44:45,  3.30s/it]
 72%|███████▏  | 10641/14725 [9:41:25<3:45:03,  3.31s/it]
 72%|███████▏  | 10642/14725 [9:41:28<3:42:40,  3.27s/it]
 72%|███████▏  | 10643/14725 [9:41:32<3:41:14,  3.25s/it]
 72%|███████▏  | 10644/14725 [9:41:35<3:47:32,  3.35s/it]
 72%|███████▏  | 10645/14725 [9:41:38<3:43:45,  3.29s/it]
 72%|███████▏  | 10646/14725 [9:41:42<3:42:44,  3.28s/it]
 72%|███████▏  | 10647/14725 [9:41:45<3:40:15,  3.24s/it]
 72%|███████▏  | 10648/14725 [9:41:48<3:38:38,  3.22s/it]
 72%|███████▏  | 10649/14725 [9:41:51<3:43:25,  3.29s/it]
 72%|███████▏  | 10650/14725 [9:41:55<3:43:50,  3.30s/it]
      
0: {'loss': 0.6901, 'grad_norm': 0.3880305543883205, 'learning_rate': 1e-05, 'epoch': 0.72}
0: {'loss': 0.6928, 'grad_norm': 0.40657894452991417, 'learning_rate': 1e-05, 'epoch': 0.72}
0:                                                    

 72%|███████▏  | 10650/14725 [9:41:55<3:43:50,  3.30s/it]
 72%|███████▏  | 10651/14725 [9:41:58<3:39:11,  3.23s/it]
 72%|███████▏  | 10652/14725 [9:42:01<3:38:07,  3.21s/it]
 72%|███████▏  | 10653/14725 [9:42:04<3:38:38,  3.22s/it]
 72%|███████▏  | 10654/14725 [9:42:08<3:42:26,  3.28s/it]
 72%|███████▏  | 10655/14725 [9:42:11<3:39:05,  3.23s/it]
 72%|███████▏  | 10656/14725 [9:42:14<3:39:14,  3.23s/it]
 72%|███████▏  | 10657/14725 [9:42:17<3:43:33,  3.30s/it]
 72%|███████▏  | 10658/14725 [9:42:21<3:40:03,  3.25s/it]
 72%|███████▏  | 10659/14725 [9:42:24<3:41:16,  3.27s/it]
 72%|███████▏  | 10660/14725 [9:42:27<3:38:52,  3.23s/it]
                                                         

 72%|███████▏  | 10660/14725 [9:42:27<3:38:52,  3.23s/it]
 72%|██████�
0: {'loss': 0.6844, 'grad_norm': 0.4262442334656921, 'learning_rate': 1e-05, 'epoch': 0.72}
0: ��▏  | 10661/14725 [9:42:30<3:35:26,  3.18s/it]
 72%|███████▏  | 10662/14725 [9:42:33<3:35:49,  3.19s/it]
 72%|███████▏  | 10663/14725 [9:42:36<3:36:00,  3.19s/it]
 72%|███████▏  | 10664/14725 [9:42:40<3:39:34,  3.24s/it]
 72%|███████▏  | 10665/14725 [9:42:43<3:44:55,  3.32s/it]
 72%|███████▏  | 10666/14725 [9:42:47<3:44:11,  3.31s/it]
 72%|███████▏  | 10667/14725 [9:42:50<3:40:34,  3.26s/it]
 72%|███████▏  | 10668/14725 [9:42:53<3:46:26,  3.35s/it]
 72%|███████▏  | 10669/14725 [9:42:57<3:46:26,  3.35s/it]
 72%|███████▏  | 10670/14725 [9:43:00<3:45:13,  3.33s/it]
                                                         

 72%|███████▏  | 10670/14725 [9:43:00<3:45:13,  3.33s/it]
 72%|███████▏  | 10671/14725 [9:43:03<3:43:09,  3.30s/it]
 72%|███████▏  | 10672/14725 [9:43:06<3:39:46,  3.25s/it]
 72%|███████�
0: {'loss': 0.6807, 'grad_norm': 0.4421851277749382, 'learning_rate': 1e-05, 'epoch': 0.73}
0: ��  | 10673/14725 [9:43:10<3:39:07,  3.24s/it]
 72%|███████▏  | 10674/14725 [9:43:13<3:42:03,  3.29s/it]
 72%|███████▏  | 10675/14725 [9:43:16<3:38:43,  3.24s/it]
 73%|███████▎  | 10676/14725 [9:43:19<3:40:06,  3.26s/it]
 73%|███████▎  | 10677/14725 [9:43:22<3:36:30,  3.21s/it]
 73%|███████▎  | 10678/14725 [9:43:26<3:38:27,  3.24s/it]
 73%|███████▎  | 10679/14725 [9:43:29<3:39:55,  3.26s/it]
 73%|███████▎  | 10680/14725 [9:43:32<3:41:44,  3.29s/it]
                                                         

 73%|███████▎  | 10680/14725 [9:43:32<3:41:44,  3.29s/it]
 73%|███████▎  | 10681/14725 [9:43:36<3:43:08,  3.31s/it]
 73%|███████▎  | 10682/14725 [9:43:39<3:39:40,  3.26s/it]
 73%|███████▎  | 10683/14725 [9:43:42<3:37:45,  3.23s/it]
 73%|███████▎  | 10684/14725 [9:43:45<3:39:48,  3.26s/it]
 73%|███████▎ 
0: {'loss': 0.6748, 'grad_norm': 0.41713272932486645, 'learning_rate': 1e-05, 'epoch': 0.73}
0:  | 10685/14725 [9:43:49<3:39:05,  3.25s/it]
 73%|███████▎  | 10686/14725 [9:43:52<3:38:29,  3.25s/it]
 73%|███████▎  | 10687/14725 [9:43:55<3:38:44,  3.25s/it]
 73%|███████▎  | 10688/14725 [9:43:59<3:43:04,  3.32s/it]
 73%|███████▎  | 10689/14725 [9:44:02<3:42:31,  3.31s/it]
 73%|███████▎  | 10690/14725 [9:44:05<3:41:56,  3.30s/it]
                                                         

 73%|███████▎  | 10690/14725 [9:44:05<3:41:56,  3.30s/it]
 73%|███████▎  | 10691/14725 [9:44:09<3:43:25,  3.32s/it]
 73%|███████▎  | 10692/14725 [9:44:12<3:48:16,  3.40s/it]
 73%|███████▎  | 10693/14725 [9:44:15<3:43:38,  3.33s/it]
 73%|███████▎  | 10694/14725 [9:44:19<3:46:11,  3.37s/it]
 73%|███████▎  | 10695/14725 [9:44:22<3:47:46,  3.39s/it]
 73%|███████▎  | 10696/14725 [9:44:25<3:44:13,  3.34s/it]
 73%|███████▎  | 
0: {'loss': 0.6737, 'grad_norm': 0.4116942633838823, 'learning_rate': 1e-05, 'epoch': 0.73}
0: 10697/14725 [9:44:29<3:41:08,  3.29s/it]
 73%|███████▎  | 10698/14725 [9:44:32<3:40:47,  3.29s/it]
 73%|███████▎  | 10699/14725 [9:44:35<3:37:29,  3.24s/it]
 73%|███████▎  | 10700/14725 [9:44:38<3:39:20,  3.27s/it]
                                                         

 73%|███████▎  | 10700/14725 [9:44:38<3:39:20,  3.27s/it]
 73%|███████▎  | 10701/14725 [9:44:42<3:40:16,  3.28s/it]
 73%|███████▎  | 10702/14725 [9:44:45<3:38:29,  3.26s/it]
 73%|███████▎  | 10703/14725 [9:44:48<3:39:13,  3.27s/it]
 73%|███████▎  | 10704/14725 [9:44:52<3:44:59,  3.36s/it]
 73%|███████▎  | 10705/14725 [9:44:55<3:40:37,  3.29s/it]
 73%|███████▎  | 10706/14725 [9:44:58<3:39:02,  3.27s/it]
 73%|███████▎  | 10707/14725 [9:45:01<3:37:31,  3.25s/it]
 73%|███████▎  | 10708/14725 [9:45:04<3:34:30,  3.20s/it]
 73%|███████▎  | 107
0: {'loss': 0.6838, 'grad_norm': 0.4219553332009701, 'learning_rate': 1e-05, 'epoch': 0.73}
0: 09/14725 [9:45:08<3:33:57,  3.20s/it]
 73%|███████▎  | 10710/14725 [9:45:11<3:38:10,  3.26s/it]
                                                         

 73%|███████▎  | 10710/14725 [9:45:11<3:38:10,  3.26s/it]
 73%|███████▎  | 10711/14725 [9:45:14<3:38:07,  3.26s/it]
 73%|███████▎  | 10712/14725 [9:45:18<3:37:54,  3.26s/it]
 73%|███████▎  | 10713/14725 [9:45:21<3:37:14,  3.25s/it]
 73%|███████▎  | 10714/14725 [9:45:24<3:37:20,  3.25s/it]
 73%|███████▎  | 10715/14725 [9:45:28<3:42:51,  3.33s/it]
 73%|███████▎  | 10716/14725 [9:45:31<3:36:44,  3.24s/it]
 73%|███████▎  | 10717/14725 [9:45:34<3:35:40,  3.23s/it]
 73%|███████▎  | 10718/14725 [9:45:37<3:43:16,  3.34s/it]
 73%|███████▎  | 10719/14725 [9:45:41<3:46:17,  3.39s/it]
 73%|███████▎  | 10720/14725 [9:45:44<3:43:21,  3.35s/it]
                                       
0: {'loss': 0.6863, 'grad_norm': 0.4173446757766782, 'learning_rate': 1e-05, 'epoch': 0.73}
0: {'loss': 0.6758, 'grad_norm': 0.4028040418849448, 'learning_rate': 1e-05, 'epoch': 0.73}
0:                   

 73%|███████▎  | 10720/14725 [9:45:44<3:43:21,  3.35s/it]
 73%|███████▎  | 10721/14725 [9:45:47<3:42:05,  3.33s/it]
 73%|███████▎  | 10722/14725 [9:45:51<3:44:48,  3.37s/it]
 73%|███████▎  | 10723/14725 [9:45:54<3:46:53,  3.40s/it]
 73%|███████▎  | 10724/14725 [9:45:58<3:42:22,  3.33s/it]
 73%|███████▎  | 10725/14725 [9:46:01<3:42:04,  3.33s/it]
 73%|███████▎  | 10726/14725 [9:46:04<3:39:48,  3.30s/it]
 73%|███████▎  | 10727/14725 [9:46:07<3:42:47,  3.34s/it]
 73%|███████▎  | 10728/14725 [9:46:11<3:41:18,  3.32s/it]
 73%|███████▎  | 10729/14725 [9:46:14<3:39:54,  3.30s/it]
 73%|███████▎  | 10730/14725 [9:46:17<3:38:55,  3.29s/it]
                                                         

 73%|███████▎  | 10730/14725 [9:46:17<3:38:55,  3.29s/it]
 73%|███████▎  | 10731/14725 [9:46:21<3:3
0: {'loss': 0.6825, 'grad_norm': 0.39757990714101327, 'learning_rate': 1e-05, 'epoch': 0.73}
0: 9:23,  3.30s/it]
 73%|███████▎  | 10732/14725 [9:46:24<3:36:02,  3.25s/it]
 73%|███████▎  | 10733/14725 [9:46:27<3:36:28,  3.25s/it]
 73%|███████▎  | 10734/14725 [9:46:30<3:37:13,  3.27s/it]
 73%|███████▎  | 10735/14725 [9:46:33<3:34:30,  3.23s/it]
 73%|███████▎  | 10736/14725 [9:46:37<3:39:35,  3.30s/it]
 73%|███████▎  | 10737/14725 [9:46:40<3:42:33,  3.35s/it]
 73%|███████▎  | 10738/14725 [9:46:44<3:42:18,  3.35s/it]
 73%|███████▎  | 10739/14725 [9:46:47<3:47:34,  3.43s/it]
 73%|███████▎  | 10740/14725 [9:46:50<3:41:40,  3.34s/it]
                                                         

 73%|███████▎  | 10740/14725 [9:46:50<3:41:40,  3.34s/it]
 73%|███████▎  | 10741/14725 [9:46:54<3:44:59,  3.39s/it]
 73%|███████▎  | 10742/14725 [9:46:57<3:44:05,  3.38s/it]
 73%|███████▎  | 10743/14725 [9:47:00<3:40:2
0: {'loss': 0.6805, 'grad_norm': 0.40152701240296906, 'learning_rate': 1e-05, 'epoch': 0.73}
0: 0,  3.32s/it]
 73%|███████▎  | 10744/14725 [9:47:04<3:39:53,  3.31s/it]
 73%|███████▎  | 10745/14725 [9:47:07<3:39:35,  3.31s/it]
 73%|███████▎  | 10746/14725 [9:47:10<3:40:34,  3.33s/it]
 73%|███████▎  | 10747/14725 [9:47:14<3:41:24,  3.34s/it]
 73%|███████▎  | 10748/14725 [9:47:17<3:46:30,  3.42s/it]
 73%|███████▎  | 10749/14725 [9:47:21<3:44:29,  3.39s/it]
 73%|███████▎  | 10750/14725 [9:47:24<3:41:05,  3.34s/it]
                                                         

 73%|███████▎  | 10750/14725 [9:47:24<3:41:05,  3.34s/it]
 73%|███████▎  | 10751/14725 [9:47:27<3:40:40,  3.33s/it]
 73%|███████▎  | 10752/14725 [9:47:31<3:39:01,  3.31s/it]
 73%|███████▎  | 10753/14725 [9:47:34<3:41:07,  3.34s/it]
 73%|███████▎  | 10754/14725 [9:47:37<3:36:43,  3.27s/it]
 73%|███████▎  | 10755/14725 [9:47:40<3:37:05, 
0: {'loss': 0.7074, 'grad_norm': 0.44562748663568974, 'learning_rate': 1e-05, 'epoch': 0.73}
0:  3.28s/it]
 73%|███████▎  | 10756/14725 [9:47:44<3:39:57,  3.33s/it]
 73%|███████▎  | 10757/14725 [9:47:47<3:38:00,  3.30s/it]
 73%|███████▎  | 10758/14725 [9:47:50<3:36:13,  3.27s/it]
 73%|███████▎  | 10759/14725 [9:47:53<3:34:22,  3.24s/it]
 73%|███████▎  | 10760/14725 [9:47:57<3:35:16,  3.26s/it]
                                                         

 73%|███████▎  | 10760/14725 [9:47:57<3:35:16,  3.26s/it]
 73%|███████▎  | 10761/14725 [9:48:00<3:37:44,  3.30s/it]
 73%|███████▎  | 10762/14725 [9:48:03<3:39:22,  3.32s/it]
 73%|███████▎  | 10763/14725 [9:48:07<3:34:05,  3.24s/it]
 73%|███████▎  | 10764/14725 [9:48:10<3:36:36,  3.28s/it]
 73%|███████▎  | 10765/14725 [9:48:13<3:36:59,  3.29s/it]
 73%|███████▎  | 10766/14725 [9:48:16<3:36:51,  3.29s/it]
 73%|███████▎  | 10767/14725 [9:48:20<3:38:43,  3.
0: {'loss': 0.6976, 'grad_norm': 0.4246756227085547, 'learning_rate': 1e-05, 'epoch': 0.73}
0: 32s/it]
 73%|███████▎  | 10768/14725 [9:48:23<3:36:08,  3.28s/it]
 73%|███████▎  | 10769/14725 [9:48:27<3:39:54,  3.34s/it]
 73%|███████▎  | 10770/14725 [9:48:30<3:42:30,  3.38s/it]
                                                         

 73%|███████▎  | 10770/14725 [9:48:30<3:42:30,  3.38s/it]
 73%|███████▎  | 10771/14725 [9:48:33<3:41:51,  3.37s/it]
 73%|███████▎  | 10772/14725 [9:48:37<3:40:58,  3.35s/it]
 73%|███████▎  | 10773/14725 [9:48:40<3:36:50,  3.29s/it]
 73%|███████▎  | 10774/14725 [9:48:43<3:36:03,  3.28s/it]
 73%|███████▎  | 10775/14725 [9:48:46<3:36:40,  3.29s/it]
 73%|███████▎  | 10776/14725 [9:48:50<3:38:30,  3.32s/it]
 73%|███████▎  | 10777/14725 [9:48:53<3:35:47,  3.28s/it]
 73%|███████▎  | 10778/14725 [9:48:56<3:36:25,  3.29s/it]
 73%|███████▎  | 10779/14725 [9:49:00<3:41:22,  3.37s
0: {'loss': 0.6676, 'grad_norm': 0.43585225703312497, 'learning_rate': 1e-05, 'epoch': 0.73}
0: {'loss': 0.6826, 'grad_norm': 0.4097343728067163, 'learning_rate': 1e-05, 'epoch': 0.73}
0: /it]
 73%|███████▎  | 10780/14725 [9:49:03<3:36:09,  3.29s/it]
                                                         

 73%|███████▎  | 10780/14725 [9:49:03<3:36:09,  3.29s/it]
 73%|███████▎  | 10781/14725 [9:49:06<3:35:43,  3.28s/it]
 73%|███████▎  | 10782/14725 [9:49:09<3:33:45,  3.25s/it]
 73%|███████▎  | 10783/14725 [9:49:12<3:31:11,  3.21s/it]
 73%|███████▎  | 10784/14725 [9:49:16<3:35:42,  3.28s/it]
 73%|███████▎  | 10785/14725 [9:49:19<3:37:18,  3.31s/it]
 73%|███████▎  | 10786/14725 [9:49:23<3:39:39,  3.35s/it]
 73%|███████▎  | 10787/14725 [9:49:26<3:35:03,  3.28s/it]
 73%|███████▎  | 10788/14725 [9:49:29<3:34:23,  3.27s/it]
 73%|███████▎  | 10789/14725 [9:49:33<3:37:41,  3.32s/it]
 73%|███████▎  | 10790/14725 [9:49:36<3:35:17,  3.28s/it]
                                                         

 73%|██�
0: {'loss': 0.6796, 'grad_norm': 0.4401643031021292, 'learning_rate': 1e-05, 'epoch': 0.73}
0: �████▎  | 10790/14725 [9:49:36<3:35:17,  3.28s/it]
 73%|███████▎  | 10791/14725 [9:49:39<3:34:15,  3.27s/it]
 73%|███████▎  | 10792/14725 [9:49:42<3:33:26,  3.26s/it]
 73%|███████▎  | 10793/14725 [9:49:45<3:33:59,  3.27s/it]
 73%|███████▎  | 10794/14725 [9:49:49<3:33:19,  3.26s/it]
 73%|███████▎  | 10795/14725 [9:49:52<3:37:55,  3.33s/it]
 73%|███████▎  | 10796/14725 [9:49:56<3:43:47,  3.42s/it]
 73%|███████▎  | 10797/14725 [9:49:59<3:39:51,  3.36s/it]
 73%|███████▎  | 10798/14725 [9:50:02<3:39:31,  3.35s/it]
 73%|███████▎  | 10799/14725 [9:50:06<3:38:08,  3.33s/it]
 73%|███████▎  | 10800/14725 [9:50:09<3:35:08,  3.29s/it]
                                                         

 73%|███████▎  | 10800/14725 [9:50:09<3:35:08,  3.29s/it]
 73%|███████▎  | 10801/14725 [9:50:12<3:29:17,  3.20s/it]
 73%|███�
0: {'loss': 0.691, 'grad_norm': 0.4222388572915197, 'learning_rate': 1e-05, 'epoch': 0.73}
0: �███▎  | 10802/14725 [9:50:15<3:32:06,  3.24s/it]
 73%|███████▎  | 10803/14725 [9:50:18<3:29:33,  3.21s/it]
 73%|███████▎  | 10804/14725 [9:50:22<3:31:49,  3.24s/it]
 73%|███████▎  | 10805/14725 [9:50:25<3:32:28,  3.25s/it]
 73%|███████▎  | 10806/14725 [9:50:28<3:34:15,  3.28s/it]
 73%|███████▎  | 10807/14725 [9:50:32<3:35:16,  3.30s/it]
 73%|███████▎  | 10808/14725 [9:50:35<3:32:02,  3.25s/it]
 73%|███████▎  | 10809/14725 [9:50:38<3:31:00,  3.23s/it]
 73%|███████▎  | 10810/14725 [9:50:41<3:30:16,  3.22s/it]
                                                         

 73%|███████▎  | 10810/14725 [9:50:41<3:30:16,  3.22s/it]
 73%|███████▎  | 10811/14725 [9:50:44<3:31:01,  3.23s/it]
 73%|███████▎  | 10812/14725 [9:50:48<3:33:57,  3.28s/it]
 73%|███████▎  | 10813/14725 [9:50:51<3:36:36,  3.32s/it]
 73%|████�
0: {'loss': 0.6819, 'grad_norm': 0.43187973484207826, 'learning_rate': 1e-05, 'epoch': 0.73}
0: �██▎  | 10814/14725 [9:50:55<3:36:22,  3.32s/it]
 73%|███████▎  | 10815/14725 [9:50:58<3:33:52,  3.28s/it]
 73%|███████▎  | 10816/14725 [9:51:01<3:35:22,  3.31s/it]
 73%|███████▎  | 10817/14725 [9:51:04<3:30:10,  3.23s/it]
 73%|███████▎  | 10818/14725 [9:51:07<3:30:34,  3.23s/it]
 73%|███████▎  | 10819/14725 [9:51:10<3:27:43,  3.19s/it]
 73%|███████▎  | 10820/14725 [9:51:14<3:27:09,  3.18s/it]
                                                         

 73%|███████▎  | 10820/14725 [9:51:14<3:27:09,  3.18s/it]
 73%|███████▎  | 10821/14725 [9:51:17<3:28:09,  3.20s/it]
 73%|███████▎  | 10822/14725 [9:51:20<3:31:08,  3.25s/it]
 74%|███████▎  | 10823/14725 [9:51:24<3:34:34,  3.30s/it]
 74%|███████▎  | 10824/14725 [9:51:27<3:34:05,  3.29s/it]
 74%|███████▎  | 10825/14725 [9:51:30<3:35:19,  3.31s/it]
 74%|█████�
0: {'loss': 0.6969, 'grad_norm': 0.40272461859464703, 'learning_rate': 1e-05, 'epoch': 0.74}
0: �█▎  | 10826/14725 [9:51:33<3:32:43,  3.27s/it]
 74%|███████▎  | 10827/14725 [9:51:37<3:28:11,  3.20s/it]
 74%|███████▎  | 10828/14725 [9:51:40<3:31:40,  3.26s/it]
 74%|███████▎  | 10829/14725 [9:51:43<3:27:49,  3.20s/it]
 74%|███████▎  | 10830/14725 [9:51:46<3:30:36,  3.24s/it]
                                                         

 74%|███████▎  | 10830/14725 [9:51:46<3:30:36,  3.24s/it]
 74%|███████▎  | 10831/14725 [9:51:50<3:31:32,  3.26s/it]
 74%|███████▎  | 10832/14725 [9:51:53<3:34:18,  3.30s/it]
 74%|███████▎  | 10833/14725 [9:51:56<3:32:24,  3.27s/it]
 74%|███████▎  | 10834/14725 [9:52:00<3:36:17,  3.34s/it]
 74%|███████▎  | 10835/14725 [9:52:03<3:35:08,  3.32s/it]
 74%|███████▎  | 10836/14725 [9:52:06<3:28:58,  3.22s/it]
 74%|███████▎  | 10837/14725 [9:52:09<3:30:14,  3.24s/it]
 74%|██████�
0: {'loss': 0.6995, 'grad_norm': 0.4228659207543732, 'learning_rate': 1e-05, 'epoch': 0.74}
0: �▎  | 10838/14725 [9:52:13<3:32:52,  3.29s/it]
 74%|███████▎  | 10839/14725 [9:52:16<3:32:47,  3.29s/it]
 74%|███████▎  | 10840/14725 [9:52:19<3:30:26,  3.25s/it]
                                                         

 74%|███████▎  | 10840/14725 [9:52:19<3:30:26,  3.25s/it]
 74%|███████▎  | 10841/14725 [9:52:22<3:26:40,  3.19s/it]
 74%|███████▎  | 10842/14725 [9:52:25<3:28:10,  3.22s/it]
 74%|███████▎  | 10843/14725 [9:52:29<3:25:52,  3.18s/it]
 74%|███████▎  | 10844/14725 [9:52:32<3:28:08,  3.22s/it]
 74%|███████▎  | 10845/14725 [9:52:35<3:28:23,  3.22s/it]
 74%|███████▎  | 10846/14725 [9:52:38<3:26:40,  3.20s/it]
 74%|███████▎  | 10847/14725 [9:52:42<3:28:39,  3.23s/it]
 74%|███████▎  | 10848/14725 [9:52:45<3:28:05,  3.22s/it]
 74%|███████▎  | 10849/14725 [9:52:48<3:25:27,  3.18s/it]
 74%|███████�
0: {'loss': 0.6625, 'grad_norm': 0.40693823314036476, 'learning_rate': 1e-05, 'epoch': 0.74}
0: {'loss': 0.6821, 'grad_norm': 0.4190052234399237, 'learning_rate': 1e-05, 'epoch': 0.74}
0: �  | 10850/14725 [9:52:51<3:28:09,  3.22s/it]
                                                         

 74%|███████▎  | 10850/14725 [9:52:51<3:28:09,  3.22s/it]
 74%|███████▎  | 10851/14725 [9:52:55<3:32:10,  3.29s/it]
 74%|███████▎  | 10852/14725 [9:52:58<3:33:24,  3.31s/it]
 74%|███████▎  | 10853/14725 [9:53:01<3:32:50,  3.30s/it]
 74%|███████▎  | 10854/14725 [9:53:04<3:30:14,  3.26s/it]
 74%|███████▎  | 10855/14725 [9:53:08<3:29:47,  3.25s/it]
 74%|███████▎  | 10856/14725 [9:53:11<3:30:18,  3.26s/it]
 74%|███████▎  | 10857/14725 [9:53:14<3:31:07,  3.28s/it]
 74%|███████▎  | 10858/14725 [9:53:18<3:32:20,  3.29s/it]
 74%|███████▎  | 10859/14725 [9:53:21<3:34:45,  3.33s/it]
 74%|███████▍  | 10860/14725 [9:53:24<3:30:24,  3.27s/it]
                                                         

 74%|███████▍  | 10860/14725 [
0: {'loss': 0.7023, 'grad_norm': 0.40705360282786157, 'learning_rate': 1e-05, 'epoch': 0.74}
0: 9:53:24<3:30:24,  3.27s/it]
 74%|███████▍  | 10861/14725 [9:53:27<3:30:20,  3.27s/it]
 74%|███████▍  | 10862/14725 [9:53:31<3:33:50,  3.32s/it]
 74%|███████▍  | 10863/14725 [9:53:34<3:33:10,  3.31s/it]
 74%|███████▍  | 10864/14725 [9:53:37<3:29:19,  3.25s/it]
 74%|███████▍  | 10865/14725 [9:53:40<3:26:55,  3.22s/it]
 74%|███████▍  | 10866/14725 [9:53:44<3:31:07,  3.28s/it]
 74%|███████▍  | 10867/14725 [9:53:47<3:30:01,  3.27s/it]
 74%|███████▍  | 10868/14725 [9:53:50<3:28:49,  3.25s/it]
 74%|███████▍  | 10869/14725 [9:53:54<3:33:30,  3.32s/it]
 74%|███████▍  | 10870/14725 [9:53:57<3:31:31,  3.29s/it]
                                                         

 74%|███████▍  | 10870/14725 [9:53:57<3:31:31,  3.29s/it]
 74%|███████▍  | 10871/14725 [9:54:00<3:32:59,  3.32s/it]
 74%|███████▍  | 10872/14725 [9:5
0: {'loss': 0.6759, 'grad_norm': 0.4190305231997195, 'learning_rate': 1e-05, 'epoch': 0.74}
0: 4:03<3:30:36,  3.28s/it]
 74%|███████▍  | 10873/14725 [9:54:07<3:32:05,  3.30s/it]
 74%|███████▍  | 10874/14725 [9:54:10<3:35:19,  3.35s/it]
 74%|███████▍  | 10875/14725 [9:54:13<3:31:30,  3.30s/it]
 74%|███████▍  | 10876/14725 [9:54:17<3:27:24,  3.23s/it]
 74%|███████▍  | 10877/14725 [9:54:20<3:25:12,  3.20s/it]
 74%|███████▍  | 10878/14725 [9:54:23<3:27:47,  3.24s/it]
 74%|███████▍  | 10879/14725 [9:54:26<3:28:04,  3.25s/it]
 74%|███████▍  | 10880/14725 [9:54:30<3:29:58,  3.28s/it]
                                                         

 74%|███████▍  | 10880/14725 [9:54:30<3:29:58,  3.28s/it]
 74%|███████▍  | 10881/14725 [9:54:33<3:29:33,  3.27s/it]
 74%|███████▍  | 10882/14725 [9:54:36<3:27:59,  3.25s/it]
 74%|███████▍  | 10883/14725 [9:54:39<3:27:35,  3.24s/it]
 74%|███████▍  | 10884/14725 [9:54:4
0: {'loss': 0.6791, 'grad_norm': 0.4174367855512704, 'learning_rate': 1e-05, 'epoch': 0.74}
0: 3<3:30:35,  3.29s/it]
 74%|███████▍  | 10885/14725 [9:54:46<3:26:27,  3.23s/it]
 74%|███████▍  | 10886/14725 [9:54:49<3:26:42,  3.23s/it]
 74%|███████▍  | 10887/14725 [9:54:52<3:23:01,  3.17s/it]
 74%|███████▍  | 10888/14725 [9:54:55<3:25:22,  3.21s/it]
 74%|███████▍  | 10889/14725 [9:54:59<3:29:03,  3.27s/it]
 74%|███████▍  | 10890/14725 [9:55:02<3:34:00,  3.35s/it]
                                                         

 74%|███████▍  | 10890/14725 [9:55:02<3:34:00,  3.35s/it]
 74%|███████▍  | 10891/14725 [9:55:05<3:30:07,  3.29s/it]
 74%|███████▍  | 10892/14725 [9:55:09<3:32:18,  3.32s/it]
 74%|███████▍  | 10893/14725 [9:55:12<3:33:10,  3.34s/it]
 74%|███████▍  | 10894/14725 [9:55:16<3:32:22,  3.33s/it]
 74%|███████▍  | 10895/14725 [9:55:19<3:30:17,  3.29s/it]
 74%|███████▍  | 10896/14725 [9:55:22<3
0: {'loss': 0.6819, 'grad_norm': 0.39393494380006316, 'learning_rate': 1e-05, 'epoch': 0.74}
0: :29:53,  3.29s/it]
 74%|███████▍  | 10897/14725 [9:55:25<3:29:03,  3.28s/it]
 74%|███████▍  | 10898/14725 [9:55:28<3:28:09,  3.26s/it]
 74%|███████▍  | 10899/14725 [9:55:32<3:27:48,  3.26s/it]
 74%|███████▍  | 10900/14725 [9:55:35<3:28:55,  3.28s/it]
                                                         

 74%|███████▍  | 10900/14725 [9:55:35<3:28:55,  3.28s/it]
 74%|███████▍  | 10901/14725 [9:55:38<3:30:50,  3.31s/it]
 74%|███████▍  | 10902/14725 [9:55:42<3:26:22,  3.24s/it]
 74%|███████▍  | 10903/14725 [9:55:45<3:28:42,  3.28s/it]
 74%|███████▍  | 10904/14725 [9:55:48<3:30:57,  3.31s/it]
 74%|███████▍  | 10905/14725 [9:55:52<3:32:37,  3.34s/it]
 74%|███████▍  | 10906/14725 [9:55:55<3:32:42,  3.34s/it]
 74%|███████▍  | 10907/14725 [9:55:58<3:30:14,  3.30s/it]
 74%|███████▍  | 10908/14725 [9:56:02<3:29
0: {'loss': 0.6924, 'grad_norm': 0.3870520136123086, 'learning_rate': 1e-05, 'epoch': 0.74}
0: :44,  3.30s/it]
 74%|███████▍  | 10909/14725 [9:56:05<3:26:21,  3.24s/it]
 74%|███████▍  | 10910/14725 [9:56:08<3:24:54,  3.22s/it]
                                                         

 74%|███████▍  | 10910/14725 [9:56:08<3:24:54,  3.22s/it]
 74%|███████▍  | 10911/14725 [9:56:11<3:27:57,  3.27s/it]
 74%|███████▍  | 10912/14725 [9:56:14<3:26:47,  3.25s/it]
 74%|███████▍  | 10913/14725 [9:56:18<3:27:56,  3.27s/it]
 74%|███████▍  | 10914/14725 [9:56:21<3:30:52,  3.32s/it]
 74%|███████▍  | 10915/14725 [9:56:24<3:28:23,  3.28s/it]
 74%|███████▍  | 10916/14725 [9:56:27<3:25:05,  3.23s/it]
 74%|███████▍  | 10917/14725 [9:56:31<3:25:22,  3.24s/it]
 74%|███████▍  | 10918/14725 [9:56:34<3:27:44,  3.27s/it]
 74%|███████▍  | 10919/14725 [9:56:37<3:24:31,  3.22s/it]
 74%|███████▍  | 10920/14725 [9:56:40<3:22:47
0: {'loss': 0.6744, 'grad_norm': 0.40423750389139007, 'learning_rate': 1e-05, 'epoch': 0.74}
0: {'loss': 0.6742, 'grad_norm': 0.41021287417187996, 'learning_rate': 1e-05, 'epoch': 0.74}
0: ,  3.20s/it]
                                                         

 74%|███████▍  | 10920/14725 [9:56:40<3:22:47,  3.20s/it]
 74%|███████▍  | 10921/14725 [9:56:44<3:24:40,  3.23s/it]
 74%|███████▍  | 10922/14725 [9:56:47<3:23:06,  3.20s/it]
 74%|███████▍  | 10923/14725 [9:56:50<3:22:04,  3.19s/it]
 74%|███████▍  | 10924/14725 [9:56:53<3:19:52,  3.16s/it]
 74%|███████▍  | 10925/14725 [9:56:56<3:17:49,  3.12s/it]
 74%|███████▍  | 10926/14725 [9:56:59<3:21:46,  3.19s/it]
 74%|███████▍  | 10927/14725 [9:57:02<3:19:34,  3.15s/it]
 74%|███████▍  | 10928/14725 [9:57:06<3:21:38,  3.19s/it]
 74%|███████▍  | 10929/14725 [9:57:09<3:21:07,  3.18s/it]
 74%|███████▍  | 10930/14725 [9:57:12<3:27:42,  3.28s/it]
                                                         

 74%|███████▍  | 10930/14725 [9:57:12<3:27:42,  3.28s/it]
 74%|
0: {'loss': 0.6906, 'grad_norm': 0.4296814856844232, 'learning_rate': 1e-05, 'epoch': 0.74}
0: ███████▍  | 10931/14725 [9:57:16<3:25:58,  3.26s/it]
 74%|███████▍  | 10932/14725 [9:57:19<3:23:24,  3.22s/it]
 74%|███████▍  | 10933/14725 [9:57:22<3:25:27,  3.25s/it]
 74%|███████▍  | 10934/14725 [9:57:25<3:26:22,  3.27s/it]
 74%|███████▍  | 10935/14725 [9:57:29<3:28:57,  3.31s/it]
 74%|███████▍  | 10936/14725 [9:57:32<3:31:43,  3.35s/it]
 74%|███████▍  | 10937/14725 [9:57:36<3:33:59,  3.39s/it]
 74%|███████▍  | 10938/14725 [9:57:39<3:32:17,  3.36s/it]
 74%|███████▍  | 10939/14725 [9:57:42<3:29:28,  3.32s/it]
 74%|███████▍  | 10940/14725 [9:57:45<3:28:14,  3.30s/it]
                                                         

 74%|███████▍  | 10940/14725 [9:57:45<3:28:14,  3.30s/it]
 74%|███████▍  | 10941/14725 [9:57:49<3:30:36,  3.34s/it]
 74%|███████▍  | 10942/14725 [9:57:52<3:30:08,  3.33s/it]
 74%|█
0: {'loss': 0.6791, 'grad_norm': 0.4134091856526281, 'learning_rate': 1e-05, 'epoch': 0.74}
0: ██████▍  | 10943/14725 [9:57:55<3:27:28,  3.29s/it]
 74%|███████▍  | 10944/14725 [9:57:59<3:25:53,  3.27s/it]
 74%|███████▍  | 10945/14725 [9:58:02<3:24:21,  3.24s/it]
 74%|███████▍  | 10946/14725 [9:58:05<3:26:24,  3.28s/it]
 74%|███████▍  | 10947/14725 [9:58:09<3:30:19,  3.34s/it]
 74%|███████▍  | 10948/14725 [9:58:12<3:29:17,  3.32s/it]
 74%|███████▍  | 10949/14725 [9:58:15<3:27:16,  3.29s/it]
 74%|███████▍  | 10950/14725 [9:58:18<3:26:23,  3.28s/it]
                                                         

 74%|███████▍  | 10950/14725 [9:58:18<3:26:23,  3.28s/it]
 74%|███████▍  | 10951/14725 [9:58:22<3:24:26,  3.25s/it]
 74%|███████▍  | 10952/14725 [9:58:25<3:27:21,  3.30s/it]
 74%|███████▍  | 10953/14725 [9:58:28<3:27:47,  3.31s/it]
 74%|███████▍  | 10954/14725 [9:58:32<3:25:12,  3.27s/it]
 74%|██
0: {'loss': 0.6836, 'grad_norm': 0.3797023469460843, 'learning_rate': 1e-05, 'epoch': 0.74}
0: █████▍  | 10955/14725 [9:58:35<3:26:24,  3.29s/it]
 74%|███████▍  | 10956/14725 [9:58:38<3:27:13,  3.30s/it]
 74%|███████▍  | 10957/14725 [9:58:41<3:23:51,  3.25s/it]
 74%|███████▍  | 10958/14725 [9:58:45<3:23:13,  3.24s/it]
 74%|███████▍  | 10959/14725 [9:58:48<3:30:45,  3.36s/it]
 74%|███████▍  | 10960/14725 [9:58:51<3:25:48,  3.28s/it]
                                                         

 74%|███████▍  | 10960/14725 [9:58:51<3:25:48,  3.28s/it]
 74%|███████▍  | 10961/14725 [9:58:54<3:23:01,  3.24s/it]
 74%|███████▍  | 10962/14725 [9:58:58<3:21:02,  3.21s/it]
 74%|███████▍  | 10963/14725 [9:59:01<3:21:48,  3.22s/it]
 74%|███████▍  | 10964/14725 [9:59:04<3:26:35,  3.30s/it]
 74%|███████▍  | 10965/14725 [9:59:08<3:26:09,  3.29s/it]
 74%|███████▍  | 10966/14725 [9:59:11<3:25:25,  3.28s/it]
 74%|███
0: {'loss': 0.6852, 'grad_norm': 0.3955110495813276, 'learning_rate': 1e-05, 'epoch': 0.74}
0: ████▍  | 10967/14725 [9:59:14<3:26:41,  3.30s/it]
 74%|███████▍  | 10968/14725 [9:59:18<3:31:23,  3.38s/it]
 74%|███████▍  | 10969/14725 [9:59:21<3:33:33,  3.41s/it]
 74%|███████▍  | 10970/14725 [9:59:24<3:31:54,  3.39s/it]
                                                         

 74%|███████▍  | 10970/14725 [9:59:24<3:31:54,  3.39s/it]
 75%|███████▍  | 10971/14725 [9:59:28<3:25:30,  3.28s/it]
 75%|███████▍  | 10972/14725 [9:59:31<3:31:10,  3.38s/it]
 75%|███████▍  | 10973/14725 [9:59:34<3:28:06,  3.33s/it]
 75%|███████▍  | 10974/14725 [9:59:38<3:25:18,  3.28s/it]
 75%|███████▍  | 10975/14725 [9:59:41<3:24:20,  3.27s/it]
 75%|███████▍  | 10976/14725 [9:59:44<3:25:24,  3.29s/it]
 75%|███████▍  | 10977/14725 [9:59:47<3:26:38,  3.31s/it]
 75%|███████▍  | 10978/14725 [9:59:51<3:24:10,  3.27s/it]
 75%|████
0: {'loss': 0.6807, 'grad_norm': 0.436151949682417, 'learning_rate': 1e-05, 'epoch': 0.75}
0: ███▍  | 10979/14725 [9:59:54<3:28:07,  3.33s/it]
 75%|███████▍  | 10980/14725 [9:59:57<3:26:35,  3.31s/it]
                                                         

 75%|███████▍  | 10980/14725 [9:59:57<3:26:35,  3.31s/it]
 75%|███████▍  | 10981/14725 [10:00:01<3:24:29,  3.28s/it]
 75%|███████▍  | 10982/14725 [10:00:04<3:23:53,  3.27s/it]
 75%|███████▍  | 10983/14725 [10:00:07<3:21:44,  3.23s/it]
 75%|███████▍  | 10984/14725 [10:00:10<3:22:17,  3.24s/it]
 75%|███████▍  | 10985/14725 [10:00:14<3:25:46,  3.30s/it]
 75%|███████▍  | 10986/14725 [10:00:17<3:25:01,  3.29s/it]
 75%|███████▍  | 10987/14725 [10:00:20<3:21:48,  3.24s/it]
 75%|███████▍  | 10988/14725 [10:00:23<3:20:38,  3.22s/it]
 75%|███████▍  | 10989/14725 [10:00:26<3:21:29,  3.24s/it]
 75%|███████▍  | 10990/14725 [10:00:30<3:22:22,  3.25s/it]
          
0: {'loss': 0.6853, 'grad_norm': 0.38433057496536405, 'learning_rate': 1e-05, 'epoch': 0.75}
0: {'loss': 0.6774, 'grad_norm': 0.41161517901905065, 'learning_rate': 1e-05, 'epoch': 0.75}
0:                                                 

 75%|███████▍  | 10990/14725 [10:00:30<3:22:22,  3.25s/it]
 75%|███████▍  | 10991/14725 [10:00:33<3:27:17,  3.33s/it]
 75%|███████▍  | 10992/14725 [10:00:37<3:25:00,  3.30s/it]
 75%|███████▍  | 10993/14725 [10:00:40<3:22:21,  3.25s/it]
 75%|███████▍  | 10994/14725 [10:00:43<3:24:35,  3.29s/it]
 75%|███████▍  | 10995/14725 [10:00:46<3:26:17,  3.32s/it]
 75%|███████▍  | 10996/14725 [10:00:50<3:23:22,  3.27s/it]
 75%|███████▍  | 10997/14725 [10:00:53<3:23:19,  3.27s/it]
 75%|███████▍  | 10998/14725 [10:00:56<3:23:58,  3.28s/it]
 75%|███████▍  | 10999/14725 [10:00:59<3:20:00,  3.22s/it]
 75%|███████▍  | 11000/14725 [10:01:02<3:16:18,  3.16s/it]
                                                          

 75%|███████▍  | 11000/14725 [10:01:02<3:16:18,  3.16s/it]
 75%|███
0: {'loss': 0.6644, 'grad_norm': 0.3947249484964775, 'learning_rate': 1e-05, 'epoch': 0.75}
0: ████▍  | 11001/14725 [10:01:06<3:18:00,  3.19s/it]
 75%|███████▍  | 11002/14725 [10:01:09<3:16:32,  3.17s/it]
 75%|███████▍  | 11003/14725 [10:01:12<3:19:08,  3.21s/it]
 75%|███████▍  | 11004/14725 [10:01:15<3:19:22,  3.21s/it]
 75%|███████▍  | 11005/14725 [10:01:18<3:16:11,  3.16s/it]
 75%|███████▍  | 11006/14725 [10:01:21<3:15:28,  3.15s/it]
 75%|███████▍  | 11007/14725 [10:01:24<3:14:41,  3.14s/it]
 75%|███████▍  | 11008/14725 [10:01:28<3:18:40,  3.21s/it]
 75%|███████▍  | 11009/14725 [10:01:31<3:20:44,  3.24s/it]
 75%|███████▍  | 11010/14725 [10:01:35<3:25:30,  3.32s/it]
                                                          

 75%|███████▍  | 11010/14725 [10:01:35<3:25:30,  3.32s/it]
 75%|███████▍  | 11011/14725 [10:01:38<3:24:26,  3.30s/it]
 75%|███████▍  | 11012/14725 [10:01:41<3:23:34,  3.29s/it]
 75
0: {'loss': 0.6807, 'grad_norm': 0.407316561024169, 'learning_rate': 1e-05, 'epoch': 0.75}
0: %|███████▍  | 11013/14725 [10:01:44<3:23:35,  3.29s/it]
 75%|███████▍  | 11014/14725 [10:01:48<3:22:45,  3.28s/it]
 75%|███████▍  | 11015/14725 [10:01:51<3:17:59,  3.20s/it]
 75%|███████▍  | 11016/14725 [10:01:54<3:20:58,  3.25s/it]
 75%|███████▍  | 11017/14725 [10:01:57<3:19:57,  3.24s/it]
 75%|███████▍  | 11018/14725 [10:02:00<3:18:57,  3.22s/it]
 75%|███████▍  | 11019/14725 [10:02:04<3:20:02,  3.24s/it]
 75%|███████▍  | 11020/14725 [10:02:07<3:19:54,  3.24s/it]
                                                          

 75%|███████▍  | 11020/14725 [10:02:07<3:19:54,  3.24s/it]
 75%|███████▍  | 11021/14725 [10:02:10<3:17:02,  3.19s/it]
 75%|███████▍  | 11022/14725 [10:02:13<3:18:41,  3.22s/it]
 75%|███████▍  | 11023/14725 [10:02:17<3:21:05,  3.26s/it]
 75%|███████▍  | 11024/14725 [10:02:20<3:21:12,  3.
0: {'loss': 0.6942, 'grad_norm': 0.41943738731381636, 'learning_rate': 1e-05, 'epoch': 0.75}
0: 26s/it]
 75%|███████▍  | 11025/14725 [10:02:24<3:27:31,  3.37s/it]
 75%|███████▍  | 11026/14725 [10:02:27<3:20:41,  3.26s/it]
 75%|███████▍  | 11027/14725 [10:02:30<3:17:27,  3.20s/it]
 75%|███████▍  | 11028/14725 [10:02:33<3:22:01,  3.28s/it]
 75%|███████▍  | 11029/14725 [10:02:37<3:23:39,  3.31s/it]
 75%|███████▍  | 11030/14725 [10:02:40<3:26:44,  3.36s/it]
                                                          

 75%|███████▍  | 11030/14725 [10:02:40<3:26:44,  3.36s/it]
 75%|███████▍  | 11031/14725 [10:02:44<3:29:38,  3.41s/it]
 75%|███████▍  | 11032/14725 [10:02:47<3:27:46,  3.38s/it]
 75%|███████▍  | 11033/14725 [10:02:50<3:25:10,  3.33s/it]
 75%|███████▍  | 11034/14725 [10:02:53<3:24:57,  3.33s/it]
 75%|███████▍  | 11035/14725 [10:02:57<3:26:19,  3.35s/it]
 75%|███████▍  | 11036/14725 [10:03:00<3
0: {'loss': 0.6721, 'grad_norm': 0.41599925897705314, 'learning_rate': 1e-05, 'epoch': 0.75}
0: :24:25,  3.32s/it]
 75%|███████▍  | 11037/14725 [10:03:03<3:23:14,  3.31s/it]
 75%|███████▍  | 11038/14725 [10:03:07<3:25:13,  3.34s/it]
 75%|███████▍  | 11039/14725 [10:03:10<3:24:39,  3.33s/it]
 75%|███████▍  | 11040/14725 [10:03:13<3:23:11,  3.31s/it]
                                                          

 75%|███████▍  | 11040/14725 [10:03:13<3:23:11,  3.31s/it]
 75%|███████▍  | 11041/14725 [10:03:16<3:20:39,  3.27s/it]
 75%|███████▍  | 11042/14725 [10:03:20<3:20:07,  3.26s/it]
 75%|███████▍  | 11043/14725 [10:03:23<3:19:27,  3.25s/it]
 75%|███████▌  | 11044/14725 [10:03:26<3:21:00,  3.28s/it]
 75%|███████▌  | 11045/14725 [10:03:29<3:18:29,  3.24s/it]
 75%|███████▌  | 11046/14725 [10:03:33<3:16:48,  3.21s/it]
 75%|███████▌  | 11047/14725 [10:03:36<3:15:59,  3.20s/it]
 75%|███████▌  | 11048/14725 
0: {'loss': 0.6676, 'grad_norm': 0.3967533615681166, 'learning_rate': 1e-05, 'epoch': 0.75}
0: [10:03:39<3:12:11,  3.14s/it]
 75%|███████▌  | 11049/14725 [10:03:42<3:14:37,  3.18s/it]
 75%|███████▌  | 11050/14725 [10:03:45<3:13:18,  3.16s/it]
                                                          

 75%|███████▌  | 11050/14725 [10:03:45<3:13:18,  3.16s/it]
 75%|███████▌  | 11051/14725 [10:03:48<3:12:43,  3.15s/it]
 75%|███████▌  | 11052/14725 [10:03:51<3:14:36,  3.18s/it]
 75%|███████▌  | 11053/14725 [10:03:55<3:16:55,  3.22s/it]
 75%|███████▌  | 11054/14725 [10:03:58<3:18:29,  3.24s/it]
 75%|███████▌  | 11055/14725 [10:04:02<3:23:26,  3.33s/it]
 75%|███████▌  | 11056/14725 [10:04:05<3:26:38,  3.38s/it]
 75%|███████▌  | 11057/14725 [10:04:08<3:23:08,  3.32s/it]
 75%|███████▌  | 11058/14725 [10:04:12<3:21:42,  3.30s/it]
 75%|███████▌  | 11059/14725 [10:04:15<3:20:36,  3.28s/it]
 75%|███████▌  | 1
0: {'loss': 0.6728, 'grad_norm': 0.4190770203376062, 'learning_rate': 1e-05, 'epoch': 0.75}
0: {'loss': 0.6821, 'grad_norm': 0.41895320120782903, 'learning_rate': 1e-05, 'epoch': 0.75}
0: 1060/14725 [10:04:18<3:20:17,  3.28s/it]
                                                          

 75%|███████▌  | 11060/14725 [10:04:18<3:20:17,  3.28s/it]
 75%|███████▌  | 11061/14725 [10:04:21<3:18:06,  3.24s/it]
 75%|███████▌  | 11062/14725 [10:04:25<3:18:41,  3.25s/it]
 75%|███████▌  | 11063/14725 [10:04:28<3:18:54,  3.26s/it]
 75%|███████▌  | 11064/14725 [10:04:31<3:17:12,  3.23s/it]
 75%|███████▌  | 11065/14725 [10:04:34<3:17:57,  3.25s/it]
 75%|███████▌  | 11066/14725 [10:04:38<3:18:36,  3.26s/it]
 75%|███████▌  | 11067/14725 [10:04:41<3:19:45,  3.28s/it]
 75%|███████▌  | 11068/14725 [10:04:44<3:17:03,  3.23s/it]
 75%|███████▌  | 11069/14725 [10:04:47<3:18:15,  3.25s/it]
 75%|███████▌  | 11070/14725 [10:04:51<3:20:05,  3.28s/it]
                                                          

 75%|███████▌  | 11070
0: {'loss': 0.6849, 'grad_norm': 0.4254919705309609, 'learning_rate': 1e-05, 'epoch': 0.75}
0: /14725 [10:04:51<3:20:05,  3.28s/it]
 75%|███████▌  | 11071/14725 [10:04:54<3:20:41,  3.30s/it]
 75%|███████▌  | 11072/14725 [10:04:57<3:21:07,  3.30s/it]
 75%|███████▌  | 11073/14725 [10:05:00<3:19:20,  3.28s/it]
 75%|███████▌  | 11074/14725 [10:05:04<3:20:47,  3.30s/it]
 75%|███████▌  | 11075/14725 [10:05:07<3:16:35,  3.23s/it]
 75%|███████▌  | 11076/14725 [10:05:10<3:22:40,  3.33s/it]
 75%|███████▌  | 11077/14725 [10:05:14<3:24:42,  3.37s/it]
 75%|███████▌  | 11078/14725 [10:05:17<3:19:08,  3.28s/it]
 75%|███████▌  | 11079/14725 [10:05:20<3:21:45,  3.32s/it]
 75%|███████▌  | 11080/14725 [10:05:24<3:19:16,  3.28s/it]
                                                          

 75%|███████▌  | 11080/14725 [10:05:24<3:19:16,  3.28s/it]
 75%|███████▌  | 11081/14725 [10:05:27<3:18:55,  3.28s/it]
 75%|███████�
0: {'loss': 0.666, 'grad_norm': 0.389384175813844, 'learning_rate': 1e-05, 'epoch': 0.75}
0: ��  | 11082/14725 [10:05:30<3:16:31,  3.24s/it]
 75%|███████▌  | 11083/14725 [10:05:33<3:19:01,  3.28s/it]
 75%|███████▌  | 11084/14725 [10:05:36<3:14:24,  3.20s/it]
 75%|███████▌  | 11085/14725 [10:05:39<3:11:58,  3.16s/it]
 75%|███████▌  | 11086/14725 [10:05:43<3:12:01,  3.17s/it]
 75%|███████▌  | 11087/14725 [10:05:46<3:13:19,  3.19s/it]
 75%|███████▌  | 11088/14725 [10:05:49<3:12:54,  3.18s/it]
 75%|███████▌  | 11089/14725 [10:05:52<3:14:12,  3.20s/it]
 75%|███████▌  | 11090/14725 [10:05:55<3:12:05,  3.17s/it]
                                                          

 75%|███████▌  | 11090/14725 [10:05:55<3:12:05,  3.17s/it]
 75%|███████▌  | 11091/14725 [10:05:59<3:11:14,  3.16s/it]
 75%|███████▌  | 11092/14725 [10:06:02<3:11:17,  3.16s/it]
 75%|███████▌  | 11093/14725 [10:06:05<3:12:26,  3.18s/it]
 75%|███�
0: {'loss': 0.6731, 'grad_norm': 0.42907839662543745, 'learning_rate': 1e-05, 'epoch': 0.75}
0: �███▌  | 11094/14725 [10:06:08<3:14:09,  3.21s/it]
 75%|███████▌  | 11095/14725 [10:06:12<3:15:54,  3.24s/it]
 75%|███████▌  | 11096/14725 [10:06:15<3:13:59,  3.21s/it]
 75%|███████▌  | 11097/14725 [10:06:18<3:14:15,  3.21s/it]
 75%|███████▌  | 11098/14725 [10:06:21<3:17:42,  3.27s/it]
 75%|███████▌  | 11099/14725 [10:06:25<3:17:52,  3.27s/it]
 75%|███████▌  | 11100/14725 [10:06:28<3:16:46,  3.26s/it]
                                                          

 75%|███████▌  | 11100/14725 [10:06:28<3:16:46,  3.26s/it]
 75%|███████▌  | 11101/14725 [10:06:31<3:15:54,  3.24s/it]
 75%|███████▌  | 11102/14725 [10:06:34<3:14:48,  3.23s/it]
 75%|███████▌  | 11103/14725 [10:06:37<3:15:17,  3.24s/it]
 75%|███████▌  | 11104/14725 [10:06:41<3:14:38,  3.23s/it]
 75%|███████▌  | 11105/14725 [10:06:44<3:14:21,  3.22s/it]
 75%|
0: {'loss': 0.6857, 'grad_norm': 0.3887440672219658, 'learning_rate': 1e-05, 'epoch': 0.75}
0: ███████▌  | 11106/14725 [10:06:47<3:16:57,  3.27s/it]
 75%|███████▌  | 11107/14725 [10:06:51<3:18:02,  3.28s/it]
 75%|███████▌  | 11108/14725 [10:06:54<3:15:14,  3.24s/it]
 75%|███████▌  | 11109/14725 [10:06:57<3:15:06,  3.24s/it]
 75%|███████▌  | 11110/14725 [10:07:00<3:15:12,  3.24s/it]
                                                          

 75%|███████▌  | 11110/14725 [10:07:00<3:15:12,  3.24s/it]
 75%|███████▌  | 11111/14725 [10:07:03<3:13:18,  3.21s/it]
 75%|███████▌  | 11112/14725 [10:07:07<3:17:33,  3.28s/it]
 75%|███████▌  | 11113/14725 [10:07:10<3:16:40,  3.27s/it]
 75%|███████▌  | 11114/14725 [10:07:13<3:15:20,  3.25s/it]
 75%|███████▌  | 11115/14725 [10:07:16<3:16:37,  3.27s/it]
 75%|███████▌  | 11116/14725 [10:07:20<3:13:07,  3.21s/it]
 75%|███████▌  | 11117/14725 [10:07:23<3:14:03,  3.23
0: {'loss': 0.6797, 'grad_norm': 0.39963933451153194, 'learning_rate': 1e-05, 'epoch': 0.76}
0: s/it]
 76%|███████▌  | 11118/14725 [10:07:26<3:13:01,  3.21s/it]
 76%|███████▌  | 11119/14725 [10:07:29<3:15:53,  3.26s/it]
 76%|███████▌  | 11120/14725 [10:07:33<3:13:41,  3.22s/it]
                                                          

 76%|███████▌  | 11120/14725 [10:07:33<3:13:41,  3.22s/it]
 76%|███████▌  | 11121/14725 [10:07:36<3:16:15,  3.27s/it]
 76%|███████▌  | 11122/14725 [10:07:39<3:14:51,  3.24s/it]
 76%|███████▌  | 11123/14725 [10:07:42<3:16:16,  3.27s/it]
 76%|███████▌  | 11124/14725 [10:07:46<3:15:41,  3.26s/it]
 76%|███████▌  | 11125/14725 [10:07:49<3:17:05,  3.28s/it]
 76%|███████▌  | 11126/14725 [10:07:52<3:15:39,  3.26s/it]
 76%|███████▌  | 11127/14725 [10:07:56<3:17:50,  3.30s/it]
 76%|███████▌  | 11128/14725 [10:07:59<3:17:16,  3.29s/it]
 76%|███████▌  | 11129/14725 [10:08:02<3:1
0: {'loss': 0.6772, 'grad_norm': 0.3990397970381126, 'learning_rate': 1e-05, 'epoch': 0.76}
0: 2:04,  3.20s/it]
 76%|███████▌  | 11130/14725 [10:08:05<3:12:04,  3.21s/it]
                                                          

 76%|███████▌  | 11130/14725 [10:08:05<3:12:04,  3.21s/it]
 76%|███████▌  | 11131/14725 [10:08:08<3:14:06,  3.24s/it]
 76%|███████▌  | 11132/14725 [10:08:12<3:16:05,  3.27s/it]
 76%|███████▌  | 11133/14725 [10:08:15<3:11:21,  3.20s/it]
 76%|███████▌  | 11134/14725 [10:08:18<3:11:38,  3.20s/it]
 76%|███████▌  | 11135/14725 [10:08:21<3:10:53,  3.19s/it]
 76%|███████▌  | 11136/14725 [10:08:25<3:17:17,  3.30s/it]
 76%|███████▌  | 11137/14725 [10:08:28<3:15:55,  3.28s/it]
 76%|███████▌  | 11138/14725 [10:08:31<3:11:42,  3.21s/it]
 76%|███████▌  | 11139/14725 [10:08:34<3:10:56,  3.19s/it]
 76%|███████▌  | 11140/14725 [10:08:37<3:13:12,  3.23s/it]
                                               
0: {'loss': 0.676, 'grad_norm': 0.4520329419708491, 'learning_rate': 1e-05, 'epoch': 0.76}
0: {'loss': 0.6798, 'grad_norm': 0.4123204589897159, 'learning_rate': 1e-05, 'epoch': 0.76}
0:            

 76%|███████▌  | 11140/14725 [10:08:37<3:13:12,  3.23s/it]
 76%|███████▌  | 11141/14725 [10:08:41<3:13:53,  3.25s/it]
 76%|███████▌  | 11142/14725 [10:08:44<3:18:04,  3.32s/it]
 76%|███████▌  | 11143/14725 [10:08:47<3:15:03,  3.27s/it]
 76%|███████▌  | 11144/14725 [10:08:51<3:13:52,  3.25s/it]
 76%|███████▌  | 11145/14725 [10:08:54<3:14:30,  3.26s/it]
 76%|███████▌  | 11146/14725 [10:08:57<3:11:55,  3.22s/it]
 76%|███████▌  | 11147/14725 [10:09:00<3:10:58,  3.20s/it]
 76%|███████▌  | 11148/14725 [10:09:03<3:11:39,  3.21s/it]
 76%|███████▌  | 11149/14725 [10:09:06<3:08:54,  3.17s/it]
 76%|███████▌  | 11150/14725 [10:09:10<3:08:57,  3.17s/it]
                                                          

 76%|███████▌  | 11150/14725 [10:09:10<3:08:57,  3.17s/it]
 76%|███████▌  | 11151/14725 [10:09
0: {'loss': 0.6742, 'grad_norm': 0.4168260475356358, 'learning_rate': 1e-05, 'epoch': 0.76}
0: :13<3:07:41,  3.15s/it]
 76%|███████▌  | 11152/14725 [10:09:16<3:08:48,  3.17s/it]
 76%|███████▌  | 11153/14725 [10:09:19<3:11:07,  3.21s/it]
 76%|███████▌  | 11154/14725 [10:09:22<3:10:57,  3.21s/it]
 76%|███████▌  | 11155/14725 [10:09:26<3:11:07,  3.21s/it]
 76%|███████▌  | 11156/14725 [10:09:29<3:15:33,  3.29s/it]
 76%|███████▌  | 11157/14725 [10:09:32<3:16:54,  3.31s/it]
 76%|███████▌  | 11158/14725 [10:09:36<3:16:00,  3.30s/it]
 76%|███████▌  | 11159/14725 [10:09:39<3:18:06,  3.33s/it]
 76%|███████▌  | 11160/14725 [10:09:42<3:15:20,  3.29s/it]
                                                          

 76%|███████▌  | 11160/14725 [10:09:42<3:15:20,  3.29s/it]
 76%|███████▌  | 11161/14725 [10:09:46<3:12:51,  3.25s/it]
 76%|███████▌  | 11162/14725 [10:09:49<3:15:33,  3.29s/it]
 76%|███████▌  | 11163/1
0: {'loss': 0.671, 'grad_norm': 0.40066644284490305, 'learning_rate': 1e-05, 'epoch': 0.76}
0: 4725 [10:09:52<3:13:48,  3.26s/it]
 76%|███████▌  | 11164/14725 [10:09:56<3:20:03,  3.37s/it]
 76%|███████▌  | 11165/14725 [10:09:59<3:15:26,  3.29s/it]
 76%|███████▌  | 11166/14725 [10:10:02<3:16:36,  3.31s/it]
 76%|███████▌  | 11167/14725 [10:10:06<3:17:44,  3.33s/it]
 76%|███████▌  | 11168/14725 [10:10:09<3:14:06,  3.27s/it]
 76%|███████▌  | 11169/14725 [10:10:12<3:12:30,  3.25s/it]
 76%|███████▌  | 11170/14725 [10:10:15<3:14:28,  3.28s/it]
                                                          

 76%|███████▌  | 11170/14725 [10:10:15<3:14:28,  3.28s/it]
 76%|███████▌  | 11171/14725 [10:10:19<3:13:46,  3.27s/it]
 76%|███████▌  | 11172/14725 [10:10:22<3:13:43,  3.27s/it]
 76%|███████▌  | 11173/14725 [10:10:25<3:10:51,  3.22s/it]
 76%|███████▌  | 11174/14725 [10:10:28<3:11:06,  3.23s/it]
 76%|███████▌
0: {'loss': 0.6868, 'grad_norm': 0.42393870915854853, 'learning_rate': 1e-05, 'epoch': 0.76}
0:   | 11175/14725 [10:10:31<3:12:04,  3.25s/it]
 76%|███████▌  | 11176/14725 [10:10:35<3:16:22,  3.32s/it]
 76%|███████▌  | 11177/14725 [10:10:38<3:16:53,  3.33s/it]
 76%|███████▌  | 11178/14725 [10:10:42<3:18:24,  3.36s/it]
 76%|███████▌  | 11179/14725 [10:10:45<3:14:20,  3.29s/it]
 76%|███████▌  | 11180/14725 [10:10:48<3:11:45,  3.25s/it]
                                                          

 76%|███████▌  | 11180/14725 [10:10:48<3:11:45,  3.25s/it]
 76%|███████▌  | 11181/14725 [10:10:51<3:10:50,  3.23s/it]
 76%|███████▌  | 11182/14725 [10:10:54<3:12:15,  3.26s/it]
 76%|███████▌  | 11183/14725 [10:10:58<3:12:02,  3.25s/it]
 76%|███████▌  | 11184/14725 [10:11:01<3:11:13,  3.24s/it]
 76%|███████▌  | 11185/14725 [10:11:04<3:11:12,  3.24s/it]
 76%|███████▌  | 11186/14725 [10:11:07<3:12:20,  3.26s/it]
 76%|████�
0: {'loss': 0.691, 'grad_norm': 0.42188038995613864, 'learning_rate': 1e-05, 'epoch': 0.76}
0: ��██▌  | 11187/14725 [10:11:11<3:12:49,  3.27s/it]
 76%|███████▌  | 11188/14725 [10:11:14<3:10:44,  3.24s/it]
 76%|███████▌  | 11189/14725 [10:11:17<3:12:41,  3.27s/it]
 76%|███████▌  | 11190/14725 [10:11:21<3:12:34,  3.27s/it]
                                                          

 76%|███████▌  | 11190/14725 [10:11:21<3:12:34,  3.27s/it]
 76%|███████▌  | 11191/14725 [10:11:24<3:13:15,  3.28s/it]
 76%|███████▌  | 11192/14725 [10:11:27<3:11:08,  3.25s/it]
 76%|███████▌  | 11193/14725 [10:11:30<3:13:05,  3.28s/it]
 76%|███████▌  | 11194/14725 [10:11:34<3:11:14,  3.25s/it]
 76%|███████▌  | 11195/14725 [10:11:37<3:08:23,  3.20s/it]
 76%|███████▌  | 11196/14725 [10:11:40<3:08:36,  3.21s/it]
 76%|███████▌  | 11197/14725 [10:11:43<3:10:49,  3.25s/it]
 76%|███████▌  | 11198/14725 [10:11:46<3:11:20,  3.25s/it]
 76%|�
0: {'loss': 0.6871, 'grad_norm': 0.3668949489991273, 'learning_rate': 1e-05, 'epoch': 0.76}
0: �██████▌  | 11199/14725 [10:11:50<3:12:59,  3.28s/it]
 76%|███████▌  | 11200/14725 [10:11:53<3:08:54,  3.22s/it]
                                                          

 76%|███████▌  | 11200/14725 [10:11:53<3:08:54,  3.22s/it]
 76%|███████▌  | 11201/14725 [10:11:56<3:09:28,  3.23s/it]
 76%|███████▌  | 11202/14725 [10:11:59<3:09:03,  3.22s/it]
 76%|███████▌  | 11203/14725 [10:12:03<3:09:17,  3.22s/it]
 76%|███████▌  | 11204/14725 [10:12:06<3:07:45,  3.20s/it]
 76%|███████▌  | 11205/14725 [10:12:09<3:12:27,  3.28s/it]
 76%|███████▌  | 11206/14725 [10:12:12<3:09:46,  3.24s/it]
 76%|███████▌  | 11207/14725 [10:12:16<3:10:19,  3.25s/it]
 76%|███████▌  | 11208/14725 [10:12:19<3:08:34,  3.22s/it]
 76%|███████▌  | 11209/14725 [10:12:22<3:14:42,  3.32s/it]
 76%|███████▌  | 11210/14725 [10:12:26<3:15:58,  3.35s/
0: {'loss': 0.6859, 'grad_norm': 0.42167608647478544, 'learning_rate': 1e-05, 'epoch': 0.76}
0: {'loss': 0.6767, 'grad_norm': 0.4092150677856554, 'learning_rate': 1e-05, 'epoch': 0.76}
0: it]
                                                          

 76%|███████▌  | 11210/14725 [10:12:26<3:15:58,  3.35s/it]
 76%|███████▌  | 11211/14725 [10:12:29<3:12:55,  3.29s/it]
 76%|███████▌  | 11212/14725 [10:12:32<3:14:25,  3.32s/it]
 76%|███████▌  | 11213/14725 [10:12:36<3:13:35,  3.31s/it]
 76%|███████▌  | 11214/14725 [10:12:39<3:12:43,  3.29s/it]
 76%|███████▌  | 11215/14725 [10:12:42<3:13:20,  3.31s/it]
 76%|███████▌  | 11216/14725 [10:12:45<3:11:49,  3.28s/it]
 76%|███████▌  | 11217/14725 [10:12:49<3:14:35,  3.33s/it]
 76%|███████▌  | 11218/14725 [10:12:52<3:10:44,  3.26s/it]
 76%|███████▌  | 11219/14725 [10:12:55<3:07:13,  3.20s/it]
 76%|███████▌  | 11220/14725 [10:12:58<3:07:46,  3.21s/it]
                                                          

 76%|███████▌  | 11220/14725 [10:12:58<3:07:46,  3.21s/it]
0: {'loss': 0.6744, 'grad_norm': 0.403764869230577, 'learning_rate': 1e-05, 'epoch': 0.76}
0:  76%|███████▌  | 11221/14725 [10:13:01<3:04:21,  3.16s/it]
 76%|███████▌  | 11222/14725 [10:13:05<3:08:54,  3.24s/it]
 76%|███████▌  | 11223/14725 [10:13:08<3:12:38,  3.30s/it]
 76%|███████▌  | 11224/14725 [10:13:11<3:09:16,  3.24s/it]
 76%|███████▌  | 11225/14725 [10:13:15<3:10:06,  3.26s/it]
 76%|███████▌  | 11226/14725 [10:13:18<3:07:59,  3.22s/it]
 76%|███████▌  | 11227/14725 [10:13:21<3:10:37,  3.27s/it]
 76%|███████▋  | 11228/14725 [10:13:24<3:09:55,  3.26s/it]
 76%|███████▋  | 11229/14725 [10:13:28<3:10:45,  3.27s/it]
 76%|███████▋  | 11230/14725 [10:13:31<3:11:13,  3.28s/it]
                                                          

 76%|███████▋  | 11230/14725 [10:13:31<3:11:13,  3.28s/it]
 76%|███████▋  | 11231/14725 [10:13:34<3:11:32,  3.29s/it]
 76%|███████▋  | 11232/14725 [10:13:37<3:07:01, 
0: {'loss': 0.6811, 'grad_norm': 0.394400904260288, 'learning_rate': 1e-05, 'epoch': 0.76}
0:  3.21s/it]
 76%|███████▋  | 11233/14725 [10:13:40<3:07:57,  3.23s/it]
 76%|███████▋  | 11234/14725 [10:13:44<3:07:43,  3.23s/it]
 76%|███████▋  | 11235/14725 [10:13:47<3:10:09,  3.27s/it]
 76%|███████▋  | 11236/14725 [10:13:50<3:10:04,  3.27s/it]
 76%|███████▋  | 11237/14725 [10:13:54<3:12:35,  3.31s/it]
 76%|███████▋  | 11238/14725 [10:13:57<3:14:36,  3.35s/it]
 76%|███████▋  | 11239/14725 [10:14:00<3:11:27,  3.30s/it]
 76%|███████▋  | 11240/14725 [10:14:03<3:08:06,  3.24s/it]
                                                          

 76%|███████▋  | 11240/14725 [10:14:03<3:08:06,  3.24s/it]
 76%|███████▋  | 11241/14725 [10:14:07<3:06:25,  3.21s/it]
 76%|███████▋  | 11242/14725 [10:14:10<3:09:49,  3.27s/it]
 76%|███████▋  | 11243/14725 [10:14:13<3:10:45,  3.29s/it]
 76%|███████▋  | 11244/14725 [10:14:1
0: {'loss': 0.6881, 'grad_norm': 0.46964438185631746, 'learning_rate': 1e-05, 'epoch': 0.76}
0: 7<3:10:25,  3.28s/it]
 76%|███████▋  | 11245/14725 [10:14:20<3:10:55,  3.29s/it]
 76%|███████▋  | 11246/14725 [10:14:23<3:12:27,  3.32s/it]
 76%|███████▋  | 11247/14725 [10:14:27<3:10:20,  3.28s/it]
 76%|███████▋  | 11248/14725 [10:14:30<3:10:02,  3.28s/it]
 76%|███████▋  | 11249/14725 [10:14:33<3:11:35,  3.31s/it]
 76%|███████▋  | 11250/14725 [10:14:36<3:09:54,  3.28s/it]
                                                          

 76%|███████▋  | 11250/14725 [10:14:36<3:09:54,  3.28s/it]
 76%|███████▋  | 11251/14725 [10:14:40<3:13:50,  3.35s/it]
 76%|███████▋  | 11252/14725 [10:14:43<3:09:16,  3.27s/it]
 76%|███████▋  | 11253/14725 [10:14:46<3:12:03,  3.32s/it]
 76%|███████▋  | 11254/14725 [10:14:50<3:14:34,  3.36s/it]
 76%|███████▋  | 11255/14725 [10:14:53<3:15:46,  3.39s/it]
 76%|███████▋  | 11256/147
0: {'loss': 0.6944, 'grad_norm': 0.3878050882376695, 'learning_rate': 1e-05, 'epoch': 0.76}
0: 25 [10:14:56<3:11:11,  3.31s/it]
 76%|███████▋  | 11257/14725 [10:15:00<3:10:42,  3.30s/it]
 76%|███████▋  | 11258/14725 [10:15:03<3:08:24,  3.26s/it]
 76%|███████▋  | 11259/14725 [10:15:06<3:12:06,  3.33s/it]
 76%|███████▋  | 11260/14725 [10:15:10<3:09:14,  3.28s/it]
                                                          

 76%|███████▋  | 11260/14725 [10:15:10<3:09:14,  3.28s/it]
 76%|███████▋  | 11261/14725 [10:15:13<3:13:16,  3.35s/it]
 76%|███████▋  | 11262/14725 [10:15:16<3:11:01,  3.31s/it]
 76%|███████▋  | 11263/14725 [10:15:20<3:14:01,  3.36s/it]
 76%|███████▋  | 11264/14725 [10:15:23<3:13:54,  3.36s/it]
 77%|███████▋  | 11265/14725 [10:15:26<3:11:42,  3.32s/it]
 77%|███████▋  | 11266/14725 [10:15:30<3:11:48,  3.33s/it]
 77%|███████▋  | 11267/14725 [10:15:33<3:14:02,  3.37s/it]
 77%|███████▋  
0: {'loss': 0.6803, 'grad_norm': 0.40146475373022517, 'learning_rate': 1e-05, 'epoch': 0.77}
0: | 11268/14725 [10:15:37<3:15:31,  3.39s/it]
 77%|███████▋  | 11269/14725 [10:15:40<3:14:06,  3.37s/it]
 77%|███████▋  | 11270/14725 [10:15:43<3:16:14,  3.41s/it]
                                                          

 77%|███████▋  | 11270/14725 [10:15:43<3:16:14,  3.41s/it]
 77%|███████▋  | 11271/14725 [10:15:47<3:14:43,  3.38s/it]
 77%|███████▋  | 11272/14725 [10:15:50<3:10:15,  3.31s/it]
 77%|███████▋  | 11273/14725 [10:15:53<3:10:13,  3.31s/it]
 77%|███████▋  | 11274/14725 [10:15:56<3:09:37,  3.30s/it]
 77%|███████▋  | 11275/14725 [10:16:00<3:11:21,  3.33s/it]
 77%|███████▋  | 11276/14725 [10:16:03<3:10:32,  3.31s/it]
 77%|███████▋  | 11277/14725 [10:16:06<3:08:09,  3.27s/it]
 77%|███████▋  | 11278/14725 [10:16:09<3:06:00,  3.24s/it]
 77%|███████▋  | 11279/14725 [10:16:13<3:08:46,  3.29s/it]
 77%|█████
0: {'loss': 0.6902, 'grad_norm': 0.39948936025699977, 'learning_rate': 1e-05, 'epoch': 0.77}
0: {'loss': 0.6912, 'grad_norm': 0.4034070374650738, 'learning_rate': 1e-05, 'epoch': 0.77}
0: ██▋  | 11280/14725 [10:16:16<3:07:27,  3.26s/it]
                                                          

 77%|███████▋  | 11280/14725 [10:16:16<3:07:27,  3.26s/it]
 77%|███████▋  | 11281/14725 [10:16:19<3:06:59,  3.26s/it]
 77%|███████▋  | 11282/14725 [10:16:23<3:08:19,  3.28s/it]
 77%|███████▋  | 11283/14725 [10:16:26<3:12:29,  3.36s/it]
 77%|███████▋  | 11284/14725 [10:16:30<3:12:29,  3.36s/it]
 77%|███████▋  | 11285/14725 [10:16:33<3:10:06,  3.32s/it]
 77%|███████▋  | 11286/14725 [10:16:36<3:11:12,  3.34s/it]
 77%|███████▋  | 11287/14725 [10:16:39<3:08:51,  3.30s/it]
 77%|███████▋  | 11288/14725 [10:16:42<3:06:00,  3.25s/it]
 77%|███████▋  | 11289/14725 [10:16:46<3:06:58,  3.26s/it]
 77%|███████▋  | 11290/14725 [10:16:49<3:07:46,  3.28s/it]
                                                          

 77%|██████�
0: {'loss': 0.6794, 'grad_norm': 0.3937228270683748, 'learning_rate': 1e-05, 'epoch': 0.77}
0: ��▋  | 11290/14725 [10:16:49<3:07:46,  3.28s/it]
 77%|███████▋  | 11291/14725 [10:16:52<3:06:35,  3.26s/it]
 77%|███████▋  | 11292/14725 [10:16:56<3:06:23,  3.26s/it]
 77%|███████▋  | 11293/14725 [10:16:59<3:05:47,  3.25s/it]
 77%|███████▋  | 11294/14725 [10:17:02<3:04:48,  3.23s/it]
 77%|███████▋  | 11295/14725 [10:17:05<3:04:34,  3.23s/it]
 77%|███████▋  | 11296/14725 [10:17:08<3:04:19,  3.23s/it]
 77%|███████▋  | 11297/14725 [10:17:11<3:01:16,  3.17s/it]
 77%|███████▋  | 11298/14725 [10:17:15<3:03:33,  3.21s/it]
 77%|███████▋  | 11299/14725 [10:17:18<3:03:02,  3.21s/it]
 77%|███████▋  | 11300/14725 [10:17:21<3:01:24,  3.18s/it]
                                                          

 77%|███████▋  | 11300/14725 [10:17:21<3:01:24,  3.18s/it]
 77%|███████▋  | 11301/14725 [10:17:24<3:00:26,  3.16s/it]
 77%|██�
0: {'loss': 0.6832, 'grad_norm': 0.44400562153482487, 'learning_rate': 1e-05, 'epoch': 0.77}
0: �████▋  | 11302/14725 [10:17:28<3:04:04,  3.23s/it]
 77%|███████▋  | 11303/14725 [10:17:31<3:04:52,  3.24s/it]
 77%|███████▋  | 11304/14725 [10:17:34<3:09:42,  3.33s/it]
 77%|███████▋  | 11305/14725 [10:17:38<3:09:07,  3.32s/it]
 77%|███████▋  | 11306/14725 [10:17:41<3:06:39,  3.28s/it]
 77%|███████▋  | 11307/14725 [10:17:44<3:10:23,  3.34s/it]
 77%|███████▋  | 11308/14725 [10:17:47<3:05:50,  3.26s/it]
 77%|███████▋  | 11309/14725 [10:17:51<3:03:59,  3.23s/it]
 77%|███████▋  | 11310/14725 [10:17:54<3:03:31,  3.22s/it]
                                                          

 77%|███████▋  | 11310/14725 [10:17:54<3:03:31,  3.22s/it]
 77%|███████▋  | 11311/14725 [10:17:57<3:05:46,  3.27s/it]
 77%|███████▋  | 11312/14725 [10:18:00<3:06:28,  3.28s/it]
 77%|███████▋  | 11313/14725 [10:18:04<3:05:27,  3.26s/it]
 7
0: {'loss': 0.685, 'grad_norm': 0.4085479178580855, 'learning_rate': 1e-05, 'epoch': 0.77}
0: 7%|███████▋  | 11314/14725 [10:18:07<3:02:49,  3.22s/it]
 77%|███████▋  | 11315/14725 [10:18:10<3:02:01,  3.20s/it]
 77%|███████▋  | 11316/14725 [10:18:13<3:05:20,  3.26s/it]
 77%|███████▋  | 11317/14725 [10:18:17<3:05:02,  3.26s/it]
 77%|███████▋  | 11318/14725 [10:18:20<3:05:05,  3.26s/it]
 77%|███████▋  | 11319/14725 [10:18:23<3:03:11,  3.23s/it]
 77%|███████▋  | 11320/14725 [10:18:26<3:03:42,  3.24s/it]
                                                          

 77%|███████▋  | 11320/14725 [10:18:26<3:03:42,  3.24s/it]
 77%|███████▋  | 11321/14725 [10:18:30<3:04:43,  3.26s/it]
 77%|███████▋  | 11322/14725 [10:18:33<3:07:13,  3.30s/it]
 77%|███████▋  | 11323/14725 [10:18:36<3:07:17,  3.30s/it]
 77%|███████▋  | 11324/14725 [10:18:40<3:06:50,  3.30s/it]
 77%|███████▋  | 11325/14725 [10:18:43<3:02:49,  3
0: {'loss': 0.6787, 'grad_norm': 0.4063059117497941, 'learning_rate': 1e-05, 'epoch': 0.77}
0: .23s/it]
 77%|███████▋  | 11326/14725 [10:18:46<3:02:46,  3.23s/it]
 77%|███████▋  | 11327/14725 [10:18:49<3:07:30,  3.31s/it]
 77%|███████▋  | 11328/14725 [10:18:53<3:05:28,  3.28s/it]
 77%|███████▋  | 11329/14725 [10:18:56<3:03:45,  3.25s/it]
 77%|███████▋  | 11330/14725 [10:18:59<3:07:37,  3.32s/it]
                                                          

 77%|███████▋  | 11330/14725 [10:18:59<3:07:37,  3.32s/it]
 77%|███████▋  | 11331/14725 [10:19:02<3:03:37,  3.25s/it]
 77%|███████▋  | 11332/14725 [10:19:06<3:03:56,  3.25s/it]
 77%|███████▋  | 11333/14725 [10:19:09<3:07:21,  3.31s/it]
 77%|███████▋  | 11334/14725 [10:19:12<3:05:37,  3.28s/it]
 77%|███████▋  | 11335/14725 [10:19:15<3:04:24,  3.26s/it]
 77%|███████▋  | 11336/14725 [10:19:19<3:04:08,  3.26s/it]
 77%|███████▋  | 11337/14725 [10:19:22<
0: {'loss': 0.6942, 'grad_norm': 0.3879862970643965, 'learning_rate': 1e-05, 'epoch': 0.77}
0: 3:05:15,  3.28s/it]
 77%|███████▋  | 11338/14725 [10:19:25<3:04:46,  3.27s/it]
 77%|███████▋  | 11339/14725 [10:19:28<3:02:31,  3.23s/it]
 77%|███████▋  | 11340/14725 [10:19:32<3:03:24,  3.25s/it]
                                                          

 77%|███████▋  | 11340/14725 [10:19:32<3:03:24,  3.25s/it]
 77%|███████▋  | 11341/14725 [10:19:35<3:00:20,  3.20s/it]
 77%|███████▋  | 11342/14725 [10:19:38<3:03:04,  3.25s/it]
 77%|███████▋  | 11343/14725 [10:19:41<3:03:26,  3.25s/it]
 77%|███████▋  | 11344/14725 [10:19:45<3:08:04,  3.34s/it]
 77%|███████▋  | 11345/14725 [10:19:48<3:07:36,  3.33s/it]
 77%|███████▋  | 11346/14725 [10:19:52<3:08:43,  3.35s/it]
 77%|███████▋  | 11347/14725 [10:19:55<3:08:05,  3.34s/it]
 77%|███████▋  | 11348/14725 [10:19:58<3:08:24,  3.35s/it]
 77%|███████▋  | 11349/14725
0: {'loss': 0.667, 'grad_norm': 0.3873130202983429, 'learning_rate': 1e-05, 'epoch': 0.77}
0:  [10:20:02<3:10:15,  3.38s/it]
 77%|███████▋  | 11350/14725 [10:20:05<3:06:38,  3.32s/it]
                                                          

 77%|███████▋  | 11350/14725 [10:20:05<3:06:38,  3.32s/it]
 77%|███████▋  | 11351/14725 [10:20:08<3:03:50,  3.27s/it]
 77%|███████▋  | 11352/14725 [10:20:11<3:02:15,  3.24s/it]
 77%|███████▋  | 11353/14725 [10:20:15<3:02:27,  3.25s/it]
 77%|███████▋  | 11354/14725 [10:20:18<2:59:26,  3.19s/it]
 77%|███████▋  | 11355/14725 [10:20:21<3:00:18,  3.21s/it]
 77%|███████▋  | 11356/14725 [10:20:24<2:59:21,  3.19s/it]
 77%|███████▋  | 11357/14725 [10:20:27<2:59:17,  3.19s/it]
 77%|███████▋  | 11358/14725 [10:20:31<3:01:11,  3.23s/it]
 77%|███████▋  | 11359/14725 [10:20:34<3:00:35,  3.22s/it]
 77%|███████▋  | 11360/14725 [10:20:37<3:04:17,  3.29s/it]
                                 
0: {'loss': 0.6711, 'grad_norm': 0.395310658321548, 'learning_rate': 1e-05, 'epoch': 0.77}
0: {'loss': 0.6722, 'grad_norm': 0.3954417803928707, 'learning_rate': 1e-05, 'epoch': 0.77}
0:                          

 77%|███████▋  | 11360/14725 [10:20:37<3:04:17,  3.29s/it]
 77%|███████▋  | 11361/14725 [10:20:41<3:05:17,  3.30s/it]
 77%|███████▋  | 11362/14725 [10:20:44<3:01:32,  3.24s/it]
 77%|███████▋  | 11363/14725 [10:20:47<3:02:37,  3.26s/it]
 77%|███████▋  | 11364/14725 [10:20:50<3:02:10,  3.25s/it]
 77%|███████▋  | 11365/14725 [10:20:53<3:00:12,  3.22s/it]
 77%|███████▋  | 11366/14725 [10:20:57<3:02:00,  3.25s/it]
 77%|███████▋  | 11367/14725 [10:21:00<3:03:31,  3.28s/it]
 77%|███████▋  | 11368/14725 [10:21:03<3:03:34,  3.28s/it]
 77%|███████▋  | 11369/14725 [10:21:06<3:00:00,  3.22s/it]
 77%|███████▋  | 11370/14725 [10:21:10<3:08:34,  3.37s/it]
                                                          

 77%|███████▋  | 11370/14725 [10:21:10<3:08:34,  3.37s/it]
 77%|███████▋  | 1137
0: {'loss': 0.6623, 'grad_norm': 0.41057465895674694, 'learning_rate': 1e-05, 'epoch': 0.77}
0: 1/14725 [10:21:13<3:03:54,  3.29s/it]
 77%|███████▋  | 11372/14725 [10:21:17<3:04:42,  3.31s/it]
 77%|███████▋  | 11373/14725 [10:21:20<3:04:40,  3.31s/it]
 77%|███████▋  | 11374/14725 [10:21:23<3:04:24,  3.30s/it]
 77%|███████▋  | 11375/14725 [10:21:27<3:05:45,  3.33s/it]
 77%|███████▋  | 11376/14725 [10:21:30<3:04:43,  3.31s/it]
 77%|███████▋  | 11377/14725 [10:21:33<3:02:42,  3.27s/it]
 77%|███████▋  | 11378/14725 [10:21:36<3:00:15,  3.23s/it]
 77%|███████▋  | 11379/14725 [10:21:39<3:00:26,  3.24s/it]
 77%|███████▋  | 11380/14725 [10:21:42<2:57:44,  3.19s/it]
                                                          

 77%|███████▋  | 11380/14725 [10:21:42<2:57:44,  3.19s/it]
 77%|███████▋  | 11381/14725 [10:21:46<2:59:14,  3.22s/it]
 77%|███████▋  | 11382/14725 [10:21:49<2:59:37,  3.22s/it]
 77%|███████
0: {'loss': 0.6749, 'grad_norm': 0.42524654130312317, 'learning_rate': 1e-05, 'epoch': 0.77}
0: ▋  | 11383/14725 [10:21:52<2:59:38,  3.23s/it]
 77%|███████▋  | 11384/14725 [10:21:55<2:59:34,  3.22s/it]
 77%|███████▋  | 11385/14725 [10:21:59<2:58:23,  3.20s/it]
 77%|███████▋  | 11386/14725 [10:22:02<2:59:17,  3.22s/it]
 77%|███████▋  | 11387/14725 [10:22:05<3:00:17,  3.24s/it]
 77%|███████▋  | 11388/14725 [10:22:09<3:05:10,  3.33s/it]
 77%|███████▋  | 11389/14725 [10:22:12<3:03:10,  3.29s/it]
 77%|███████▋  | 11390/14725 [10:22:15<3:02:43,  3.29s/it]
                                                          

 77%|███████▋  | 11390/14725 [10:22:15<3:02:43,  3.29s/it]
 77%|███████▋  | 11391/14725 [10:22:18<2:59:39,  3.23s/it]
 77%|███████▋  | 11392/14725 [10:22:22<3:01:13,  3.26s/it]
 77%|███████▋  | 11393/14725 [10:22:25<3:01:04,  3.26s/it]
 77%|███████▋  | 11394/14725 [10:22:28<3:00:58,  3.26s/it]
 77%|███�
0: {'loss': 0.688, 'grad_norm': 0.40529980463121473, 'learning_rate': 1e-05, 'epoch': 0.77}
0: ��███▋  | 11395/14725 [10:22:31<3:00:36,  3.25s/it]
 77%|███████▋  | 11396/14725 [10:22:35<3:01:03,  3.26s/it]
 77%|███████▋  | 11397/14725 [10:22:38<3:02:42,  3.29s/it]
 77%|███████▋  | 11398/14725 [10:22:41<3:02:44,  3.30s/it]
 77%|███████▋  | 11399/14725 [10:22:45<3:03:32,  3.31s/it]
 77%|███████▋  | 11400/14725 [10:22:48<3:04:14,  3.32s/it]
                                                          

 77%|███████▋  | 11400/14725 [10:22:48<3:04:14,  3.32s/it]
 77%|███████▋  | 11401/14725 [10:22:51<3:04:03,  3.32s/it]
 77%|███████▋  | 11402/14725 [10:22:55<3:08:03,  3.40s/it]
 77%|███████▋  | 11403/14725 [10:22:58<3:08:10,  3.40s/it]
 77%|███████▋  | 11404/14725 [10:23:01<3:03:39,  3.32s/it]
 77%|███████▋  | 11405/14725 [10:23:05<3:03:34,  3.32s/it]
 77%|███████▋  | 11406/14725 [10:23:08<3:09:23,  3.42s/it]
 77%
0: {'loss': 0.6682, 'grad_norm': 0.4167038689716037, 'learning_rate': 1e-05, 'epoch': 0.77}
0: |███████▋  | 11407/14725 [10:23:12<3:04:44,  3.34s/it]
 77%|███████▋  | 11408/14725 [10:23:15<3:02:59,  3.31s/it]
 77%|███████▋  | 11409/14725 [10:23:18<3:05:03,  3.35s/it]
 77%|███████▋  | 11410/14725 [10:23:21<3:03:04,  3.31s/it]
                                                          

 77%|███████▋  | 11410/14725 [10:23:21<3:03:04,  3.31s/it]
 77%|███████▋  | 11411/14725 [10:23:25<3:01:32,  3.29s/it]
 78%|███████▊  | 11412/14725 [10:23:28<3:00:40,  3.27s/it]
 78%|███████▊  | 11413/14725 [10:23:31<2:59:45,  3.26s/it]
 78%|███████▊  | 11414/14725 [10:23:34<2:57:20,  3.21s/it]
 78%|███████▊  | 11415/14725 [10:23:38<2:59:47,  3.26s/it]
 78%|███████▊  | 11416/14725 [10:23:41<2:58:49,  3.24s/it]
 78%|███████▊  | 11417/14725 [10:23:44<2:58:43,  3.24s/it]
 78%|███████▊  | 11418/14725 [10:23:47<2:58:54,  3.2
0: {'loss': 0.6786, 'grad_norm': 0.45404820322368733, 'learning_rate': 1e-05, 'epoch': 0.78}
0: 5s/it]
 78%|███████▊  | 11419/14725 [10:23:51<2:59:58,  3.27s/it]
 78%|███████▊  | 11420/14725 [10:23:54<2:57:47,  3.23s/it]
                                                          

 78%|███████▊  | 11420/14725 [10:23:54<2:57:47,  3.23s/it]
 78%|███████▊  | 11421/14725 [10:23:57<2:56:17,  3.20s/it]
 78%|███████▊  | 11422/14725 [10:24:00<2:58:38,  3.24s/it]
 78%|███████▊  | 11423/14725 [10:24:03<2:58:37,  3.25s/it]
 78%|███████▊  | 11424/14725 [10:24:07<2:57:46,  3.23s/it]
 78%|███████▊  | 11425/14725 [10:24:10<2:59:04,  3.26s/it]
 78%|███████▊  | 11426/14725 [10:24:13<2:56:20,  3.21s/it]
 78%|███████▊  | 11427/14725 [10:24:16<2:52:32,  3.14s/it]
 78%|███████▊  | 11428/14725 [10:24:19<2:54:48,  3.18s/it]
 78%|███████▊  | 11429/14725 [10:24:23<2:56:48,  3.22s/it]
 78%|███████▊  | 11430/14725 [10:24:26<2:
0: {'loss': 0.6893, 'grad_norm': 0.4023410926614429, 'learning_rate': 1e-05, 'epoch': 0.78}
0: {'loss': 0.6702, 'grad_norm': 0.3884260295095471, 'learning_rate': 1e-05, 'epoch': 0.78}
0: 59:43,  3.27s/it]
                                                          

 78%|███████▊  | 11430/14725 [10:24:26<2:59:43,  3.27s/it]
 78%|███████▊  | 11431/14725 [10:24:29<2:56:49,  3.22s/it]
 78%|███████▊  | 11432/14725 [10:24:33<3:01:47,  3.31s/it]
 78%|███████▊  | 11433/14725 [10:24:36<3:00:01,  3.28s/it]
 78%|███████▊  | 11434/14725 [10:24:39<2:58:35,  3.26s/it]
 78%|███████▊  | 11435/14725 [10:24:42<2:55:37,  3.20s/it]
 78%|███████▊  | 11436/14725 [10:24:45<2:54:57,  3.19s/it]
 78%|███████▊  | 11437/14725 [10:24:48<2:52:49,  3.15s/it]
 78%|███████▊  | 11438/14725 [10:24:52<2:57:54,  3.25s/it]
 78%|███████▊  | 11439/14725 [10:24:55<2:56:21,  3.22s/it]
 78%|███████▊  | 11440/14725 [10:24:58<2:53:09,  3.16s/it]
                                                          

 78%|███████▊  | 11440/14725 [10:24:58<2:53:0
0: {'loss': 0.6815, 'grad_norm': 0.40514945773203387, 'learning_rate': 1e-05, 'epoch': 0.78}
0: 9,  3.16s/it]
 78%|███████▊  | 11441/14725 [10:25:01<2:51:07,  3.13s/it]
 78%|███████▊  | 11442/14725 [10:25:04<2:55:11,  3.20s/it]
 78%|███████▊  | 11443/14725 [10:25:08<2:53:56,  3.18s/it]
 78%|███████▊  | 11444/14725 [10:25:11<2:59:47,  3.29s/it]
 78%|███████▊  | 11445/14725 [10:25:14<3:00:17,  3.30s/it]
 78%|███████▊  | 11446/14725 [10:25:18<2:59:30,  3.28s/it]
 78%|███████▊  | 11447/14725 [10:25:21<3:03:53,  3.37s/it]
 78%|███████▊  | 11448/14725 [10:25:25<3:05:08,  3.39s/it]
 78%|███████▊  | 11449/14725 [10:25:28<3:03:03,  3.35s/it]
 78%|███████▊  | 11450/14725 [10:25:31<3:02:32,  3.34s/it]
                                                          

 78%|███████▊  | 11450/14725 [10:25:31<3:02:32,  3.34s/it]
 78%|███████▊  | 11451/14725 [10:25:35<3:03:02,  3.35s/it]
 78%|███████▊  | 11452/14725 [10:2
0: {'loss': 0.6843, 'grad_norm': 0.42522258860973494, 'learning_rate': 1e-05, 'epoch': 0.78}
0: 5:38<3:02:57,  3.35s/it]
 78%|███████▊  | 11453/14725 [10:25:42<3:08:33,  3.46s/it]
 78%|███████▊  | 11454/14725 [10:25:45<3:07:10,  3.43s/it]
 78%|███████▊  | 11455/14725 [10:25:48<3:05:07,  3.40s/it]
 78%|███████▊  | 11456/14725 [10:25:52<3:03:01,  3.36s/it]
 78%|███████▊  | 11457/14725 [10:25:55<3:03:03,  3.36s/it]
 78%|███████▊  | 11458/14725 [10:25:58<2:58:41,  3.28s/it]
 78%|███████▊  | 11459/14725 [10:26:01<2:59:51,  3.30s/it]
 78%|███████▊  | 11460/14725 [10:26:05<3:04:46,  3.40s/it]
                                                          

 78%|███████▊  | 11460/14725 [10:26:05<3:04:46,  3.40s/it]
 78%|███████▊  | 11461/14725 [10:26:08<3:02:10,  3.35s/it]
 78%|███████▊  | 11462/14725 [10:26:12<3:01:22,  3.34s/it]
 78%|███████▊  | 11463/14725 [10:26:15<2:58:33,  3.28s/it]
 78%|███████▊  | 11464/
0: {'loss': 0.6666, 'grad_norm': 0.40233277812114887, 'learning_rate': 1e-05, 'epoch': 0.78}
0: 14725 [10:26:18<2:56:29,  3.25s/it]
 78%|███████▊  | 11465/14725 [10:26:21<2:54:50,  3.22s/it]
 78%|███████▊  | 11466/14725 [10:26:24<2:56:49,  3.26s/it]
 78%|███████▊  | 11467/14725 [10:26:28<2:55:15,  3.23s/it]
 78%|███████▊  | 11468/14725 [10:26:31<2:54:29,  3.21s/it]
 78%|███████▊  | 11469/14725 [10:26:34<2:55:08,  3.23s/it]
 78%|███████▊  | 11470/14725 [10:26:37<2:57:38,  3.27s/it]
                                                          

 78%|███████▊  | 11470/14725 [10:26:37<2:57:38,  3.27s/it]
 78%|███████▊  | 11471/14725 [10:26:41<2:55:32,  3.24s/it]
 78%|███████▊  | 11472/14725 [10:26:44<2:54:38,  3.22s/it]
 78%|███████▊  | 11473/14725 [10:26:47<2:55:21,  3.24s/it]
 78%|███████▊  | 11474/14725 [10:26:50<2:53:39,  3.20s/it]
 78%|███████▊  | 11475/14725 [10:26:53<2:54:30,  3.22s/it]
 78%|███████�
0: {'loss': 0.6687, 'grad_norm': 0.4035486208299849, 'learning_rate': 1e-05, 'epoch': 0.78}
0: �  | 11476/14725 [10:26:57<2:54:05,  3.22s/it]
 78%|███████▊  | 11477/14725 [10:27:00<2:54:16,  3.22s/it]
 78%|███████▊  | 11478/14725 [10:27:03<2:56:37,  3.26s/it]
 78%|███████▊  | 11479/14725 [10:27:07<2:56:56,  3.27s/it]
 78%|███████▊  | 11480/14725 [10:27:10<2:55:53,  3.25s/it]
                                                          

 78%|███████▊  | 11480/14725 [10:27:10<2:55:53,  3.25s/it]
 78%|███████▊  | 11481/14725 [10:27:13<2:56:42,  3.27s/it]
 78%|███████▊  | 11482/14725 [10:27:16<2:56:36,  3.27s/it]
 78%|███████▊  | 11483/14725 [10:27:20<2:57:16,  3.28s/it]
 78%|███████▊  | 11484/14725 [10:27:23<2:55:02,  3.24s/it]
 78%|███████▊  | 11485/14725 [10:27:26<2:54:40,  3.23s/it]
 78%|███████▊  | 11486/14725 [10:27:29<2:54:23,  3.23s/it]
 78%|███████▊  | 11487/14725 [10:27:32<2:54:21,  3.23s/it]
 78%|████
0: {'loss': 0.6793, 'grad_norm': 0.4024676531140735, 'learning_rate': 1e-05, 'epoch': 0.78}
0: ███▊  | 11488/14725 [10:27:36<2:53:00,  3.21s/it]
 78%|███████▊  | 11489/14725 [10:27:39<2:55:46,  3.26s/it]
 78%|███████▊  | 11490/14725 [10:27:42<2:55:30,  3.26s/it]
                                                          

 78%|███████▊  | 11490/14725 [10:27:42<2:55:30,  3.26s/it]
 78%|███████▊  | 11491/14725 [10:27:46<3:01:04,  3.36s/it]
 78%|███████▊  | 11492/14725 [10:27:49<3:00:34,  3.35s/it]
 78%|███████▊  | 11493/14725 [10:27:52<2:59:20,  3.33s/it]
 78%|███████▊  | 11494/14725 [10:27:56<2:57:16,  3.29s/it]
 78%|███████▊  | 11495/14725 [10:27:59<2:59:49,  3.34s/it]
 78%|███████▊  | 11496/14725 [10:28:02<2:57:23,  3.30s/it]
 78%|███████▊  | 11497/14725 [10:28:05<2:55:39,  3.27s/it]
 78%|███████▊  | 11498/14725 [10:28:09<2:53:28,  3.23s/it]
 78%|███████▊  | 11499/14725 [10:28:12<2:55:14,  3.26s/it]
 78%|�
0: {'loss': 0.6875, 'grad_norm': 0.40191395874516556, 'learning_rate': 1e-05, 'epoch': 0.78}
0: {'loss': 0.6969, 'grad_norm': 0.38673746932918024, 'learning_rate': 1e-05, 'epoch': 0.78}
0: ��██████▊  | 11500/14725 [10:28:15<2:55:25,  3.26s/it]
                                                          

 78%|███████▊  | 11500/14725 [10:28:15<2:55:25,  3.26s/it]
 78%|███████▊  | 11501/14725 [10:28:19<2:57:41,  3.31s/it]
 78%|███████▊  | 11502/14725 [10:28:22<2:57:54,  3.31s/it]
 78%|███████▊  | 11503/14725 [10:28:25<2:55:31,  3.27s/it]
 78%|███████▊  | 11504/14725 [10:28:28<2:56:50,  3.29s/it]
 78%|███████▊  | 11505/14725 [10:28:32<3:00:22,  3.36s/it]
 78%|███████▊  | 11506/14725 [10:28:35<3:01:25,  3.38s/it]
 78%|███████▊  | 11507/14725 [10:28:39<2:58:25,  3.33s/it]
 78%|███████▊  | 11508/14725 [10:28:42<2:58:30,  3.33s/it]
 78%|███████▊  | 11509/14725 [10:28:45<2:59:17,  3.34s/it]
 78%|███████▊  | 11510/14725 [10:28:49<2:59:32,  3.35s/it]
                                                          

 78%|█�
0: {'loss': 0.6768, 'grad_norm': 0.4185294054667201, 'learning_rate': 1e-05, 'epoch': 0.78}
0: �█████▊  | 11510/14725 [10:28:49<2:59:32,  3.35s/it]
 78%|███████▊  | 11511/14725 [10:28:52<2:57:50,  3.32s/it]
 78%|███████▊  | 11512/14725 [10:28:56<3:01:25,  3.39s/it]
 78%|███████▊  | 11513/14725 [10:28:59<2:59:17,  3.35s/it]
 78%|███████▊  | 11514/14725 [10:29:02<2:56:38,  3.30s/it]
 78%|███████▊  | 11515/14725 [10:29:05<2:57:49,  3.32s/it]
 78%|███████▊  | 11516/14725 [10:29:09<2:57:13,  3.31s/it]
 78%|███████▊  | 11517/14725 [10:29:12<2:57:13,  3.31s/it]
 78%|███████▊  | 11518/14725 [10:29:15<2:56:25,  3.30s/it]
 78%|███████▊  | 11519/14725 [10:29:19<2:56:35,  3.30s/it]
 78%|███████▊  | 11520/14725 [10:29:22<2:55:55,  3.29s/it]
                                                          

 78%|███████▊  | 11520/14725 [10:29:22<2:55:55,  3.29s/it]
 78%|███████▊  | 11521/14725 [10:29:25<2:54:06,  3.26s/it]
0: {'loss': 0.6805, 'grad_norm': 0.40634011856471974, 'learning_rate': 1e-05, 'epoch': 0.78}
0: 
 78%|███████▊  | 11522/14725 [10:29:28<2:54:41,  3.27s/it]
 78%|███████▊  | 11523/14725 [10:29:31<2:50:35,  3.20s/it]
 78%|███████▊  | 11524/14725 [10:29:35<2:56:42,  3.31s/it]
 78%|███████▊  | 11525/14725 [10:29:38<2:54:47,  3.28s/it]
 78%|███████▊  | 11526/14725 [10:29:41<2:52:22,  3.23s/it]
 78%|███████▊  | 11527/14725 [10:29:44<2:50:33,  3.20s/it]
 78%|███████▊  | 11528/14725 [10:29:48<2:51:09,  3.21s/it]
 78%|███████▊  | 11529/14725 [10:29:51<2:51:31,  3.22s/it]
 78%|███████▊  | 11530/14725 [10:29:54<2:50:44,  3.21s/it]
                                                          

 78%|███████▊  | 11530/14725 [10:29:54<2:50:44,  3.21s/it]
 78%|███████▊  | 11531/14725 [10:29:57<2:52:06,  3.23s/it]
 78%|███████▊  | 11532/14725 [10:30:01<2:52:49,  3.25s/it]
 78%|███████▊  | 11533/14725 [10:30:04<2:53:42,
0: {'loss': 0.6693, 'grad_norm': 0.45054616873809067, 'learning_rate': 1e-05, 'epoch': 0.78}
0:   3.27s/it]
 78%|███████▊  | 11534/14725 [10:30:07<2:52:24,  3.24s/it]
 78%|███████▊  | 11535/14725 [10:30:10<2:52:27,  3.24s/it]
 78%|███████▊  | 11536/14725 [10:30:13<2:51:39,  3.23s/it]
 78%|███████▊  | 11537/14725 [10:30:17<2:51:16,  3.22s/it]
 78%|███████▊  | 11538/14725 [10:30:20<2:53:16,  3.26s/it]
 78%|███████▊  | 11539/14725 [10:30:23<2:51:45,  3.23s/it]
 78%|███████▊  | 11540/14725 [10:30:26<2:51:46,  3.24s/it]
                                                          

 78%|███████▊  | 11540/14725 [10:30:26<2:51:46,  3.24s/it]
 78%|███████▊  | 11541/14725 [10:30:30<2:51:45,  3.24s/it]
 78%|███████▊  | 11542/14725 [10:30:33<2:53:37,  3.27s/it]
 78%|███████▊  | 11543/14725 [10:30:36<2:54:11,  3.28s/it]
 78%|███████▊  | 11544/14725 [10:30:40<2:53:49,  3.28s/it]
 78%|███████▊  | 11545/14725 [10:30:
0: {'loss': 0.6793, 'grad_norm': 0.40285415731585617, 'learning_rate': 1e-05, 'epoch': 0.78}
0: 43<2:52:09,  3.25s/it]
 78%|███████▊  | 11546/14725 [10:30:46<2:54:03,  3.29s/it]
 78%|███████▊  | 11547/14725 [10:30:49<2:52:18,  3.25s/it]
 78%|███████▊  | 11548/14725 [10:30:53<2:54:50,  3.30s/it]
 78%|███████▊  | 11549/14725 [10:30:56<2:53:14,  3.27s/it]
 78%|███████▊  | 11550/14725 [10:30:59<2:52:17,  3.26s/it]
                                                          

 78%|███████▊  | 11550/14725 [10:30:59<2:52:17,  3.26s/it]
 78%|███████▊  | 11551/14725 [10:31:03<2:53:05,  3.27s/it]
 78%|███████▊  | 11552/14725 [10:31:06<2:54:11,  3.29s/it]
 78%|███████▊  | 11553/14725 [10:31:09<2:55:57,  3.33s/it]
 78%|███████▊  | 11554/14725 [10:31:13<2:56:29,  3.34s/it]
 78%|███████▊  | 11555/14725 [10:31:16<2:54:48,  3.31s/it]
 78%|███████▊  | 11556/14725 [10:31:19<2:55:04,  3.31s/it]
 78%|███████▊  | 11557/14
0: {'loss': 0.6814, 'grad_norm': 0.4155208924858514, 'learning_rate': 1e-05, 'epoch': 0.79}
0: 725 [10:31:23<2:56:07,  3.34s/it]
 78%|███████▊  | 11558/14725 [10:31:26<2:53:11,  3.28s/it]
 78%|███████▊  | 11559/14725 [10:31:29<2:55:38,  3.33s/it]
 79%|███████▊  | 11560/14725 [10:31:32<2:52:08,  3.26s/it]
                                                          

 79%|███████▊  | 11560/14725 [10:31:32<2:52:08,  3.26s/it]
 79%|███████▊  | 11561/14725 [10:31:35<2:49:09,  3.21s/it]
 79%|███████▊  | 11562/14725 [10:31:38<2:47:04,  3.17s/it]
 79%|███████▊  | 11563/14725 [10:31:42<2:46:57,  3.17s/it]
 79%|███████▊  | 11564/14725 [10:31:45<2:49:07,  3.21s/it]
 79%|███████▊  | 11565/14725 [10:31:48<2:48:25,  3.20s/it]
 79%|███████▊  | 11566/14725 [10:31:51<2:47:54,  3.19s/it]
 79%|███████▊  | 11567/14725 [10:31:55<2:49:41,  3.22s/it]
 79%|███████▊  | 11568/14725 [10:31:58<2:48:01,  3.19s/it]
 79%|███████▊ 
0: {'loss': 0.6705, 'grad_norm': 0.41562772230951844, 'learning_rate': 1e-05, 'epoch': 0.79}
0:  | 11569/14725 [10:32:01<2:45:58,  3.16s/it]
 79%|███████▊  | 11570/14725 [10:32:04<2:46:39,  3.17s/it]
                                                          

 79%|███████▊  | 11570/14725 [10:32:04<2:46:39,  3.17s/it]
 79%|███████▊  | 11571/14725 [10:32:07<2:49:34,  3.23s/it]
 79%|███████▊  | 11572/14725 [10:32:11<2:50:15,  3.24s/it]
 79%|███████▊  | 11573/14725 [10:32:14<2:48:24,  3.21s/it]
 79%|███████▊  | 11574/14725 [10:32:17<2:47:26,  3.19s/it]
 79%|███████▊  | 11575/14725 [10:32:20<2:49:33,  3.23s/it]
 79%|███████▊  | 11576/14725 [10:32:23<2:47:04,  3.18s/it]
 79%|███████▊  | 11577/14725 [10:32:26<2:47:20,  3.19s/it]
 79%|███████▊  | 11578/14725 [10:32:30<2:48:43,  3.22s/it]
 79%|███████▊  | 11579/14725 [10:32:33<2:47:48,  3.20s/it]
 79%|███████▊  | 11580/14725 [10:32:36<2:48:10,  3.21s/it]
                   
0: {'loss': 0.6767, 'grad_norm': 0.4034558272191966, 'learning_rate': 1e-05, 'epoch': 0.79}
0: {'loss': 0.6774, 'grad_norm': 0.42682722212644114, 'learning_rate': 1e-05, 'epoch': 0.79}
0:                                        

 79%|███████▊  | 11580/14725 [10:32:36<2:48:10,  3.21s/it]
 79%|███████▊  | 11581/14725 [10:32:39<2:45:36,  3.16s/it]
 79%|███████▊  | 11582/14725 [10:32:42<2:47:36,  3.20s/it]
 79%|███████▊  | 11583/14725 [10:32:46<2:46:36,  3.18s/it]
 79%|███████▊  | 11584/14725 [10:32:49<2:46:22,  3.18s/it]
 79%|███████▊  | 11585/14725 [10:32:52<2:45:34,  3.16s/it]
 79%|███████▊  | 11586/14725 [10:32:55<2:50:37,  3.26s/it]
 79%|███████▊  | 11587/14725 [10:32:59<2:50:30,  3.26s/it]
 79%|███████▊  | 11588/14725 [10:33:02<2:51:31,  3.28s/it]
 79%|███████▊  | 11589/14725 [10:33:05<2:50:08,  3.26s/it]
 79%|███████▊  | 11590/14725 [10:33:08<2:47:32,  3.21s/it]
                                                          

 79%|███████▊  | 11590/14725 [10:33:08<2:47:32,  3.21s/it]
 79%|██████
0: {'loss': 0.6946, 'grad_norm': 0.388134722440445, 'learning_rate': 1e-05, 'epoch': 0.79}
0: █▊  | 11591/14725 [10:33:12<2:53:34,  3.32s/it]
 79%|███████▊  | 11592/14725 [10:33:15<2:51:20,  3.28s/it]
 79%|███████▊  | 11593/14725 [10:33:18<2:51:54,  3.29s/it]
 79%|███████▊  | 11594/14725 [10:33:22<2:50:05,  3.26s/it]
 79%|███████▊  | 11595/14725 [10:33:25<2:49:34,  3.25s/it]
 79%|███████▉  | 11596/14725 [10:33:28<2:50:31,  3.27s/it]
 79%|███████▉  | 11597/14725 [10:33:32<2:53:05,  3.32s/it]
 79%|███████▉  | 11598/14725 [10:33:35<2:50:11,  3.27s/it]
 79%|███████▉  | 11599/14725 [10:33:38<2:48:07,  3.23s/it]
 79%|███████▉  | 11600/14725 [10:33:41<2:46:57,  3.21s/it]
                                                          

 79%|███████▉  | 11600/14725 [10:33:41<2:46:57,  3.21s/it]
 79%|███████▉  | 11601/14725 [10:33:44<2:49:41,  3.26s/it]
 79%|███████▉  | 11602/14725 [10:33:48<2:49:13,  3.25s/it]
 79%|██�
0: {'loss': 0.6844, 'grad_norm': 0.3866021028365764, 'learning_rate': 1e-05, 'epoch': 0.79}
0: ��████▉  | 11603/14725 [10:33:51<2:49:39,  3.26s/it]
 79%|███████▉  | 11604/14725 [10:33:54<2:48:20,  3.24s/it]
 79%|███████▉  | 11605/14725 [10:33:57<2:46:17,  3.20s/it]
 79%|███████▉  | 11606/14725 [10:34:00<2:48:37,  3.24s/it]
 79%|███████▉  | 11607/14725 [10:34:04<2:51:37,  3.30s/it]
 79%|███████▉  | 11608/14725 [10:34:07<2:51:24,  3.30s/it]
 79%|███████▉  | 11609/14725 [10:34:10<2:50:20,  3.28s/it]
 79%|███████▉  | 11610/14725 [10:34:14<2:47:52,  3.23s/it]
                                                          

 79%|███████▉  | 11610/14725 [10:34:14<2:47:52,  3.23s/it]
 79%|███████▉  | 11611/14725 [10:34:17<2:48:25,  3.25s/it]
 79%|███████▉  | 11612/14725 [10:34:20<2:48:15,  3.24s/it]
 79%|███████▉  | 11613/14725 [10:34:23<2:50:24,  3.29s/it]
 79%|███████▉  | 11614/14725 [10:34:27<2:48:42,  3.25s/it]
 
0: {'loss': 0.6926, 'grad_norm': 0.3998322070527483, 'learning_rate': 1e-05, 'epoch': 0.79}
0: 79%|███████▉  | 11615/14725 [10:34:30<2:47:48,  3.24s/it]
 79%|███████▉  | 11616/14725 [10:34:33<2:46:06,  3.21s/it]
 79%|███████▉  | 11617/14725 [10:34:36<2:45:34,  3.20s/it]
 79%|███████▉  | 11618/14725 [10:34:40<2:49:49,  3.28s/it]
 79%|███████▉  | 11619/14725 [10:34:43<2:50:23,  3.29s/it]
 79%|███████▉  | 11620/14725 [10:34:46<2:50:06,  3.29s/it]
                                                          

 79%|███████▉  | 11620/14725 [10:34:46<2:50:06,  3.29s/it]
 79%|███████▉  | 11621/14725 [10:34:49<2:48:51,  3.26s/it]
 79%|███████▉  | 11622/14725 [10:34:53<2:48:28,  3.26s/it]
 79%|███████▉  | 11623/14725 [10:34:56<2:50:10,  3.29s/it]
 79%|███████▉  | 11624/14725 [10:34:59<2:49:55,  3.29s/it]
 79%|███████▉  | 11625/14725 [10:35:03<2:50:33,  3.30s/it]
 79%|███████▉  | 11626/14725 [10:35:06<2:53:30,  
0: {'loss': 0.7014, 'grad_norm': 0.39987746785668216, 'learning_rate': 1e-05, 'epoch': 0.79}
0: 3.36s/it]
 79%|███████▉  | 11627/14725 [10:35:09<2:52:21,  3.34s/it]
 79%|███████▉  | 11628/14725 [10:35:13<2:50:26,  3.30s/it]
 79%|███████▉  | 11629/14725 [10:35:16<2:49:08,  3.28s/it]
 79%|███████▉  | 11630/14725 [10:35:19<2:45:43,  3.21s/it]
                                                          

 79%|███████▉  | 11630/14725 [10:35:19<2:45:43,  3.21s/it]
 79%|███████▉  | 11631/14725 [10:35:22<2:46:45,  3.23s/it]
 79%|███████▉  | 11632/14725 [10:35:25<2:46:45,  3.23s/it]
 79%|███████▉  | 11633/14725 [10:35:29<2:48:04,  3.26s/it]
 79%|███████▉  | 11634/14725 [10:35:32<2:47:09,  3.24s/it]
 79%|███████▉  | 11635/14725 [10:35:35<2:47:35,  3.25s/it]
 79%|███████▉  | 11636/14725 [10:35:38<2:46:16,  3.23s/it]
 79%|███████▉  | 11637/14725 [10:35:42<2:46:04,  3.23s/it]
 79%|███████▉  | 11638/14725 [10:35:45
0: {'loss': 0.6792, 'grad_norm': 0.3949332117384266, 'learning_rate': 1e-05, 'epoch': 0.79}
0: <2:44:02,  3.19s/it]
 79%|███████▉  | 11639/14725 [10:35:48<2:46:10,  3.23s/it]
 79%|███████▉  | 11640/14725 [10:35:51<2:44:40,  3.20s/it]
                                                          

 79%|███████▉  | 11640/14725 [10:35:51<2:44:40,  3.20s/it]
 79%|███████▉  | 11641/14725 [10:35:55<2:46:27,  3.24s/it]
 79%|███████▉  | 11642/14725 [10:35:58<2:45:52,  3.23s/it]
 79%|███████▉  | 11643/14725 [10:36:01<2:48:06,  3.27s/it]
 79%|███████▉  | 11644/14725 [10:36:05<2:50:46,  3.33s/it]
 79%|███████▉  | 11645/14725 [10:36:08<2:53:46,  3.39s/it]
 79%|███████▉  | 11646/14725 [10:36:11<2:51:27,  3.34s/it]
 79%|███████▉  | 11647/14725 [10:36:15<2:48:22,  3.28s/it]
 79%|███████▉  | 11648/14725 [10:36:18<2:47:33,  3.27s/it]
 79%|███████▉  | 11649/14725 [10:36:21<2:48:49,  3.29s/it]
 79%|███████▉  | 11650/1472
0: {'loss': 0.6847, 'grad_norm': 0.4004053505015727, 'learning_rate': 1e-05, 'epoch': 0.79}
0: {'loss': 0.6701, 'grad_norm': 0.398848596773035, 'learning_rate': 1e-05, 'epoch': 0.79}
0: 5 [10:36:24<2:48:44,  3.29s/it]
                                                          

 79%|███████▉  | 11650/14725 [10:36:24<2:48:44,  3.29s/it]
 79%|███████▉  | 11651/14725 [10:36:28<2:48:37,  3.29s/it]
 79%|███████▉  | 11652/14725 [10:36:31<2:48:24,  3.29s/it]
 79%|███████▉  | 11653/14725 [10:36:34<2:48:47,  3.30s/it]
 79%|███████▉  | 11654/14725 [10:36:37<2:47:36,  3.27s/it]
 79%|███████▉  | 11655/14725 [10:36:41<2:46:56,  3.26s/it]
 79%|███████▉  | 11656/14725 [10:36:44<2:45:42,  3.24s/it]
 79%|███████▉  | 11657/14725 [10:36:47<2:48:04,  3.29s/it]
 79%|███████▉  | 11658/14725 [10:36:50<2:46:17,  3.25s/it]
 79%|███████▉  | 11659/14725 [10:36:54<2:48:13,  3.29s/it]
 79%|███████▉  | 11660/14725 [10:36:57<2:45:54,  3.25s/it]
                                                          

 79%|███████▉  | 11660/14725 [1
0: {'loss': 0.6588, 'grad_norm': 0.4005391785020853, 'learning_rate': 1e-05, 'epoch': 0.79}
0: 0:36:57<2:45:54,  3.25s/it]
 79%|███████▉  | 11661/14725 [10:37:00<2:44:37,  3.22s/it]
 79%|███████▉  | 11662/14725 [10:37:03<2:44:28,  3.22s/it]
 79%|███████▉  | 11663/14725 [10:37:07<2:43:11,  3.20s/it]
 79%|███████▉  | 11664/14725 [10:37:10<2:42:17,  3.18s/it]
 79%|███████▉  | 11665/14725 [10:37:13<2:42:18,  3.18s/it]
 79%|███████▉  | 11666/14725 [10:37:16<2:42:04,  3.18s/it]
 79%|███████▉  | 11667/14725 [10:37:19<2:43:15,  3.20s/it]
 79%|███████▉  | 11668/14725 [10:37:23<2:46:34,  3.27s/it]
 79%|███████▉  | 11669/14725 [10:37:26<2:45:34,  3.25s/it]
 79%|███████▉  | 11670/14725 [10:37:29<2:42:47,  3.20s/it]
                                                          

 79%|███████▉  | 11670/14725 [10:37:29<2:42:47,  3.20s/it]
 79%|███████▉  | 11671/14725 [10:37:32<2:42:55,  3.20s/it]
 79%|███████▉  | 116
0: {'loss': 0.6691, 'grad_norm': 0.39452323658292887, 'learning_rate': 1e-05, 'epoch': 0.79}
0: 72/14725 [10:37:36<2:44:43,  3.24s/it]
 79%|███████▉  | 11673/14725 [10:37:39<2:47:08,  3.29s/it]
 79%|███████▉  | 11674/14725 [10:37:42<2:49:21,  3.33s/it]
 79%|███████▉  | 11675/14725 [10:37:46<2:47:47,  3.30s/it]
 79%|███████▉  | 11676/14725 [10:37:49<2:46:01,  3.27s/it]
 79%|███████▉  | 11677/14725 [10:37:52<2:47:48,  3.30s/it]
 79%|███████▉  | 11678/14725 [10:37:56<2:48:29,  3.32s/it]
 79%|███████▉  | 11679/14725 [10:37:59<2:47:08,  3.29s/it]
 79%|███████▉  | 11680/14725 [10:38:02<2:46:20,  3.28s/it]
                                                          

 79%|███████▉  | 11680/14725 [10:38:02<2:46:20,  3.28s/it]
 79%|███████▉  | 11681/14725 [10:38:06<2:50:59,  3.37s/it]
 79%|███████▉  | 11682/14725 [10:38:09<2:46:17,  3.28s/it]
 79%|███████▉  | 11683/14725 [10:38:12<2:46:42,  3.29s/it]
 79%|██████�
0: {'loss': 0.6794, 'grad_norm': 0.38735483415461736, 'learning_rate': 1e-05, 'epoch': 0.79}
0: �▉  | 11684/14725 [10:38:15<2:46:00,  3.28s/it]
 79%|███████▉  | 11685/14725 [10:38:18<2:45:23,  3.26s/it]
 79%|███████▉  | 11686/14725 [10:38:22<2:43:52,  3.24s/it]
 79%|███████▉  | 11687/14725 [10:38:25<2:42:20,  3.21s/it]
 79%|███████▉  | 11688/14725 [10:38:28<2:44:51,  3.26s/it]
 79%|███████▉  | 11689/14725 [10:38:31<2:44:36,  3.25s/it]
 79%|███████▉  | 11690/14725 [10:38:35<2:43:52,  3.24s/it]
                                                          

 79%|███████▉  | 11690/14725 [10:38:35<2:43:52,  3.24s/it]
 79%|███████▉  | 11691/14725 [10:38:38<2:42:48,  3.22s/it]
 79%|███████▉  | 11692/14725 [10:38:41<2:43:50,  3.24s/it]
 79%|███████▉  | 11693/14725 [10:38:44<2:46:00,  3.29s/it]
 79%|███████▉  | 11694/14725 [10:38:48<2:46:31,  3.30s/it]
 79%|███████▉  | 11695/14725 [10:38:51<2:43:39,  3.24s/it]
 79%|███
0: {'loss': 0.6897, 'grad_norm': 0.4101337030455963, 'learning_rate': 1e-05, 'epoch': 0.79}
0: ████▉  | 11696/14725 [10:38:54<2:42:16,  3.21s/it]
 79%|███████▉  | 11697/14725 [10:38:57<2:42:51,  3.23s/it]
 79%|███████▉  | 11698/14725 [10:39:00<2:40:06,  3.17s/it]
 79%|███████▉  | 11699/14725 [10:39:04<2:44:39,  3.26s/it]
 79%|███████▉  | 11700/14725 [10:39:07<2:44:01,  3.25s/it]
                                                          

 79%|███████▉  | 11700/14725 [10:39:07<2:44:01,  3.25s/it]
 79%|███████▉  | 11701/14725 [10:39:10<2:43:22,  3.24s/it]
 79%|███████▉  | 11702/14725 [10:39:14<2:44:29,  3.26s/it]
 79%|███████▉  | 11703/14725 [10:39:17<2:43:39,  3.25s/it]
 79%|███████▉  | 11704/14725 [10:39:20<2:43:34,  3.25s/it]
 79%|███████▉  | 11705/14725 [10:39:23<2:42:44,  3.23s/it]
 79%|███████▉  | 11706/14725 [10:39:27<2:45:56,  3.30s/it]
 80%|███████▉  | 11707/14725 [10:39:30<2:48:40,  3.35s/it]
 80
0: {'loss': 0.6811, 'grad_norm': 0.3893607541406185, 'learning_rate': 1e-05, 'epoch': 0.8}
0: %|███████▉  | 11708/14725 [10:39:33<2:46:28,  3.31s/it]
 80%|███████▉  | 11709/14725 [10:39:37<2:51:16,  3.41s/it]
 80%|███████▉  | 11710/14725 [10:39:40<2:49:20,  3.37s/it]
                                                          

 80%|███████▉  | 11710/14725 [10:39:40<2:49:20,  3.37s/it]
 80%|███████▉  | 11711/14725 [10:39:43<2:46:09,  3.31s/it]
 80%|███████▉  | 11712/14725 [10:39:47<2:44:53,  3.28s/it]
 80%|███████▉  | 11713/14725 [10:39:50<2:45:27,  3.30s/it]
 80%|███████▉  | 11714/14725 [10:39:53<2:42:30,  3.24s/it]
 80%|███████▉  | 11715/14725 [10:39:56<2:44:40,  3.28s/it]
 80%|███████▉  | 11716/14725 [10:40:00<2:45:04,  3.29s/it]
 80%|███████▉  | 11717/14725 [10:40:03<2:44:52,  3.29s/it]
 80%|███████▉  | 11718/14725 [10:40:06<2:46:10,  3.32s/it]
 80%|███████▉  | 11719/14725 [10:40:10<2:46:24,  3.
0: {'loss': 0.6754, 'grad_norm': 0.41303100475855525, 'learning_rate': 1e-05, 'epoch': 0.8}
0: 32s/it]
 80%|███████▉  | 11720/14725 [10:40:13<2:43:49,  3.27s/it]
                                                          

 80%|███████▉  | 11720/14725 [10:40:13<2:43:49,  3.27s/it]
 80%|███████▉  | 11721/14725 [10:40:16<2:43:18,  3.26s/it]
 80%|███████▉  | 11722/14725 [10:40:19<2:42:11,  3.24s/it]
 80%|███████▉  | 11723/14725 [10:40:23<2:42:03,  3.24s/it]
 80%|███████▉  | 11724/14725 [10:40:26<2:40:28,  3.21s/it]
 80%|███████▉  | 11725/14725 [10:40:29<2:40:17,  3.21s/it]
 80%|███████▉  | 11726/14725 [10:40:32<2:40:34,  3.21s/it]
 80%|███████▉  | 11727/14725 [10:40:36<2:43:41,  3.28s/it]
 80%|███████▉  | 11728/14725 [10:40:39<2:45:17,  3.31s/it]
 80%|███████▉  | 11729/14725 [10:40:42<2:43:31,  3.27s/it]
 80%|███████▉  | 11730/14725 [10:40:45<2:42:23,  3.25s/it]
                                                        
0: {'loss': 0.6801, 'grad_norm': 0.39257405880319074, 'learning_rate': 1e-05, 'epoch': 0.8}
0: {'loss': 0.6674, 'grad_norm': 0.40428956394043697, 'learning_rate': 1e-05, 'epoch': 0.8}
0:   

 80%|███████▉  | 11730/14725 [10:40:45<2:42:23,  3.25s/it]
 80%|███████▉  | 11731/14725 [10:40:49<2:44:27,  3.30s/it]
 80%|███████▉  | 11732/14725 [10:40:52<2:41:38,  3.24s/it]
 80%|███████▉  | 11733/14725 [10:40:55<2:40:57,  3.23s/it]
 80%|███████▉  | 11734/14725 [10:40:58<2:40:30,  3.22s/it]
 80%|███████▉  | 11735/14725 [10:41:01<2:40:16,  3.22s/it]
 80%|███████▉  | 11736/14725 [10:41:05<2:39:28,  3.20s/it]
 80%|███████▉  | 11737/14725 [10:41:08<2:41:47,  3.25s/it]
 80%|███████▉  | 11738/14725 [10:41:11<2:42:56,  3.27s/it]
 80%|███████▉  | 11739/14725 [10:41:15<2:43:04,  3.28s/it]
 80%|███████▉  | 11740/14725 [10:41:18<2:41:48,  3.25s/it]
                                                          

 80%|███████▉  | 11740/14725 [10:41:18<2:41:48,  3.25s/it]
 80%|███████▉  | 11741/14725 [10:41:21<2:39:
0: {'loss': 0.6822, 'grad_norm': 0.3970363557663239, 'learning_rate': 1e-05, 'epoch': 0.8}
0: 48,  3.21s/it]
 80%|███████▉  | 11742/14725 [10:41:24<2:39:19,  3.20s/it]
 80%|███████▉  | 11743/14725 [10:41:27<2:38:17,  3.18s/it]
 80%|███████▉  | 11744/14725 [10:41:30<2:36:17,  3.15s/it]
 80%|███████▉  | 11745/14725 [10:41:34<2:39:20,  3.21s/it]
 80%|███████▉  | 11746/14725 [10:41:37<2:39:52,  3.22s/it]
 80%|███████▉  | 11747/14725 [10:41:40<2:39:20,  3.21s/it]
 80%|███████▉  | 11748/14725 [10:41:43<2:41:36,  3.26s/it]
 80%|███████▉  | 11749/14725 [10:41:47<2:45:34,  3.34s/it]
 80%|███████▉  | 11750/14725 [10:41:50<2:44:16,  3.31s/it]
                                                          

 80%|███████▉  | 11750/14725 [10:41:50<2:44:16,  3.31s/it]
 80%|███████▉  | 11751/14725 [10:41:54<2:43:59,  3.31s/it]
 80%|███████▉  | 11752/14725 [10:41:57<2:41:57,  3.27s/it]
 80%|███████▉  | 11753/14725 [10:
0: {'loss': 0.6617, 'grad_norm': 0.4321234659396933, 'learning_rate': 1e-05, 'epoch': 0.8}
0: 42:00<2:42:34,  3.28s/it]
 80%|███████▉  | 11754/14725 [10:42:03<2:41:47,  3.27s/it]
 80%|███████▉  | 11755/14725 [10:42:06<2:40:14,  3.24s/it]
 80%|███████▉  | 11756/14725 [10:42:10<2:41:40,  3.27s/it]
 80%|███████▉  | 11757/14725 [10:42:13<2:41:05,  3.26s/it]
 80%|███████▉  | 11758/14725 [10:42:16<2:42:52,  3.29s/it]
 80%|███████▉  | 11759/14725 [10:42:20<2:41:04,  3.26s/it]
 80%|███████▉  | 11760/14725 [10:42:23<2:41:48,  3.27s/it]
                                                          

 80%|███████▉  | 11760/14725 [10:42:23<2:41:48,  3.27s/it]
 80%|███████▉  | 11761/14725 [10:42:26<2:39:52,  3.24s/it]
 80%|███████▉  | 11762/14725 [10:42:29<2:40:24,  3.25s/it]
 80%|███████▉  | 11763/14725 [10:42:33<2:43:11,  3.31s/it]
 80%|███████▉  | 11764/14725 [10:42:36<2:47:04,  3.39s/it]
 80%|███████▉  | 11765
0: {'loss': 0.6778, 'grad_norm': 0.3970100810677018, 'learning_rate': 1e-05, 'epoch': 0.8}
0: /14725 [10:42:40<2:44:15,  3.33s/it]
 80%|███████▉  | 11766/14725 [10:42:43<2:42:04,  3.29s/it]
 80%|███████▉  | 11767/14725 [10:42:46<2:41:38,  3.28s/it]
 80%|███████▉  | 11768/14725 [10:42:49<2:39:40,  3.24s/it]
 80%|███████▉  | 11769/14725 [10:42:52<2:39:57,  3.25s/it]
 80%|███████▉  | 11770/14725 [10:42:56<2:42:26,  3.30s/it]
                                                          

 80%|███████▉  | 11770/14725 [10:42:56<2:42:26,  3.30s/it]
 80%|███████▉  | 11771/14725 [10:42:59<2:40:25,  3.26s/it]
 80%|███████▉  | 11772/14725 [10:43:02<2:39:36,  3.24s/it]
 80%|███████▉  | 11773/14725 [10:43:05<2:38:26,  3.22s/it]
 80%|███████▉  | 11774/14725 [10:43:09<2:40:33,  3.26s/it]
 80%|███████▉  | 11775/14725 [10:43:12<2:37:10,  3.20s/it]
 80%|███████▉  | 11776/14725 [10:43:15<2:37:16,  3.20s/it]
 80%|███████�
0: {'loss': 0.6805, 'grad_norm': 0.4204896945139716, 'learning_rate': 1e-05, 'epoch': 0.8}
0: ��  | 11777/14725 [10:43:18<2:37:52,  3.21s/it]
 80%|███████▉  | 11778/14725 [10:43:22<2:41:31,  3.29s/it]
 80%|███████▉  | 11779/14725 [10:43:25<2:40:31,  3.27s/it]
 80%|████████  | 11780/14725 [10:43:28<2:41:15,  3.29s/it]
                                                          

 80%|████████  | 11780/14725 [10:43:28<2:41:15,  3.29s/it]
 80%|████████  | 11781/14725 [10:43:40<4:45:46,  5.82s/it]
 80%|████████  | 11782/14725 [10:43:43<4:09:03,  5.08s/it]
 80%|████████  | 11783/14725 [10:43:47<3:41:28,  4.52s/it]
 80%|████████  | 11784/14725 [10:43:50<3:24:22,  4.17s/it]
 80%|████████  | 11785/14725 [10:43:53<3:09:00,  3.86s/it]
 80%|████████  | 11786/14725 [10:43:56<2:58:56,  3.65s/it]
 80%|████████  | 11787/14725 [10:44:00<2:55:12,  3.58s/it]
 80%|████████  | 11788/14725 [10:44:03<2:51:33,  3.50s/it]
 80%|███�
0: {'loss': 0.6765, 'grad_norm': 0.39989142238890996, 'learning_rate': 1e-05, 'epoch': 0.8}
0: �████  | 11789/14725 [10:44:06<2:49:40,  3.47s/it]
 80%|████████  | 11790/14725 [10:44:10<2:46:22,  3.40s/it]
                                                          

 80%|████████  | 11790/14725 [10:44:10<2:46:22,  3.40s/it]
 80%|████████  | 11791/14725 [10:44:13<2:41:50,  3.31s/it]
 80%|████████  | 11792/14725 [10:44:16<2:42:00,  3.31s/it]
 80%|████████  | 11793/14725 [10:44:19<2:40:10,  3.28s/it]
 80%|████████  | 11794/14725 [10:44:22<2:38:17,  3.24s/it]
 80%|████████  | 11795/14725 [10:44:26<2:37:57,  3.23s/it]
 80%|████████  | 11796/14725 [10:44:29<2:40:07,  3.28s/it]
 80%|████████  | 11797/14725 [10:44:32<2:42:39,  3.33s/it]
 80%|████████  | 11798/14725 [10:44:36<2:41:38,  3.31s/it]
 80%|████████  | 11799/14725 [10:44:39<2:40:41,  3.30s/it]
 80%|████████  | 11800/14725 [10:44:42<2:39:08,  3.26s/it]
     
0: {'loss': 0.6631, 'grad_norm': 0.3826487289482672, 'learning_rate': 1e-05, 'epoch': 0.8}
0: {'loss': 0.6871, 'grad_norm': 0.3900982250138492, 'learning_rate': 1e-05, 'epoch': 0.8}
0:                                                      

 80%|████████  | 11800/14725 [10:44:42<2:39:08,  3.26s/it]
 80%|████████  | 11801/14725 [10:44:45<2:37:32,  3.23s/it]
 80%|████████  | 11802/14725 [10:44:49<2:39:21,  3.27s/it]
 80%|████████  | 11803/14725 [10:44:52<2:42:46,  3.34s/it]
 80%|████████  | 11804/14725 [10:44:55<2:40:29,  3.30s/it]
 80%|████████  | 11805/14725 [10:44:59<2:43:15,  3.35s/it]
 80%|████████  | 11806/14725 [10:45:02<2:39:35,  3.28s/it]
 80%|████████  | 11807/14725 [10:45:05<2:36:59,  3.23s/it]
 80%|████████  | 11808/14725 [10:45:08<2:36:57,  3.23s/it]
 80%|████████  | 11809/14725 [10:45:12<2:38:32,  3.26s/it]
 80%|████████  | 11810/14725 [10:45:15<2:37:39,  3.25s/it]
                                                          

 80%|████████  | 11810/14725 [10:45:15<2:37:39,  3.25s/it]
 80%|█�
0: {'loss': 0.6991, 'grad_norm': 0.40916918121596135, 'learning_rate': 1e-05, 'epoch': 0.8}
0: ��██████  | 11811/14725 [10:45:18<2:39:20,  3.28s/it]
 80%|████████  | 11812/14725 [10:45:21<2:36:18,  3.22s/it]
 80%|████████  | 11813/14725 [10:45:24<2:36:01,  3.21s/it]
 80%|████████  | 11814/14725 [10:45:28<2:37:02,  3.24s/it]
 80%|████████  | 11815/14725 [10:45:31<2:36:25,  3.23s/it]
 80%|████████  | 11816/14725 [10:45:34<2:36:09,  3.22s/it]
 80%|████████  | 11817/14725 [10:45:37<2:37:12,  3.24s/it]
 80%|████████  | 11818/14725 [10:45:41<2:35:04,  3.20s/it]
 80%|████████  | 11819/14725 [10:45:44<2:36:40,  3.23s/it]
 80%|████████  | 11820/14725 [10:45:47<2:36:42,  3.24s/it]
                                                          

 80%|████████  | 11820/14725 [10:45:47<2:36:42,  3.24s/it]
 80%|████████  | 11821/14725 [10:45:50<2:38:05,  3.27s/it]
 80%|████████  | 11822/14725 [10:45:54<2:41:35,  3.34s/it
0: {'loss': 0.6751, 'grad_norm': 0.4015162713432652, 'learning_rate': 1e-05, 'epoch': 0.8}
0: ]
 80%|████████  | 11823/14725 [10:45:57<2:39:54,  3.31s/it]
 80%|████████  | 11824/14725 [10:46:00<2:39:28,  3.30s/it]
 80%|████████  | 11825/14725 [10:46:04<2:39:37,  3.30s/it]
 80%|████████  | 11826/14725 [10:46:07<2:36:16,  3.23s/it]
 80%|████████  | 11827/14725 [10:46:10<2:37:27,  3.26s/it]
 80%|████████  | 11828/14725 [10:46:13<2:34:55,  3.21s/it]
 80%|████████  | 11829/14725 [10:46:16<2:31:16,  3.13s/it]
 80%|████████  | 11830/14725 [10:46:20<2:34:22,  3.20s/it]
                                                          

 80%|████████  | 11830/14725 [10:46:20<2:34:22,  3.20s/it]
 80%|████████  | 11831/14725 [10:46:23<2:32:08,  3.15s/it]
 80%|████████  | 11832/14725 [10:46:26<2:34:42,  3.21s/it]
 80%|████████  | 11833/14725 [10:46:29<2:34:22,  3.20s/it]
 80%|████████  | 11834/14725 [10:46:32<2:33:23
0: {'loss': 0.6822, 'grad_norm': 0.4047545774605819, 'learning_rate': 1e-05, 'epoch': 0.8}
0: ,  3.18s/it]
 80%|████████  | 11835/14725 [10:46:36<2:34:29,  3.21s/it]
 80%|████████  | 11836/14725 [10:46:39<2:35:25,  3.23s/it]
 80%|████████  | 11837/14725 [10:46:42<2:33:09,  3.18s/it]
 80%|████████  | 11838/14725 [10:46:45<2:35:04,  3.22s/it]
 80%|████████  | 11839/14725 [10:46:48<2:33:48,  3.20s/it]
 80%|████████  | 11840/14725 [10:46:51<2:32:52,  3.18s/it]
                                                          

 80%|████████  | 11840/14725 [10:46:51<2:32:52,  3.18s/it]
 80%|████████  | 11841/14725 [10:46:55<2:32:33,  3.17s/it]
 80%|████████  | 11842/14725 [10:46:58<2:33:03,  3.19s/it]
 80%|████████  | 11843/14725 [10:47:01<2:33:18,  3.19s/it]
 80%|████████  | 11844/14725 [10:47:04<2:34:22,  3.21s/it]
 80%|████████  | 11845/14725 [10:47:08<2:35:05,  3.23s/it]
 80%|████████  | 11846/14725 [10:47
0: {'loss': 0.6812, 'grad_norm': 0.437091754260141, 'learning_rate': 1e-05, 'epoch': 0.8}
0: :11<2:35:50,  3.25s/it]
 80%|████████  | 11847/14725 [10:47:14<2:38:05,  3.30s/it]
 80%|████████  | 11848/14725 [10:47:17<2:36:00,  3.25s/it]
 80%|████████  | 11849/14725 [10:47:21<2:34:26,  3.22s/it]
 80%|████████  | 11850/14725 [10:47:24<2:33:05,  3.20s/it]
                                                          

 80%|████████  | 11850/14725 [10:47:24<2:33:05,  3.20s/it]
 80%|████████  | 11851/14725 [10:47:27<2:32:43,  3.19s/it]
 80%|████████  | 11852/14725 [10:47:30<2:33:51,  3.21s/it]
 80%|████████  | 11853/14725 [10:47:33<2:35:41,  3.25s/it]
 81%|████████  | 11854/14725 [10:47:37<2:35:38,  3.25s/it]
 81%|████████  | 11855/14725 [10:47:40<2:35:22,  3.25s/it]
 81%|████████  | 11856/14725 [10:47:43<2:35:12,  3.25s/it]
 81%|████████  | 11857/14725 [10:47:46<2:33:22,  3.21s/it]
 81%|████████  | 11858/1
0: {'loss': 0.673, 'grad_norm': 0.3881194568332104, 'learning_rate': 1e-05, 'epoch': 0.81}
0: 4725 [10:47:50<2:33:53,  3.22s/it]
 81%|████████  | 11859/14725 [10:47:53<2:36:55,  3.29s/it]
 81%|████████  | 11860/14725 [10:47:56<2:36:06,  3.27s/it]
                                                          

 81%|████████  | 11860/14725 [10:47:56<2:36:06,  3.27s/it]
 81%|████████  | 11861/14725 [10:47:59<2:35:08,  3.25s/it]
 81%|████████  | 11862/14725 [10:48:03<2:36:18,  3.28s/it]
 81%|████████  | 11863/14725 [10:48:06<2:34:50,  3.25s/it]
 81%|████████  | 11864/14725 [10:48:09<2:33:19,  3.22s/it]
 81%|████████  | 11865/14725 [10:48:12<2:30:53,  3.17s/it]
 81%|████████  | 11866/14725 [10:48:15<2:33:00,  3.21s/it]
 81%|████████  | 11867/14725 [10:48:19<2:33:58,  3.23s/it]
 81%|████████  | 11868/14725 [10:48:22<2:31:53,  3.19s/it]
 81%|████████  | 11869/14725 [10:48:25<2:31:32,  3.18s/it]
 81%|████████
0: {'loss': 0.6707, 'grad_norm': 0.39598718998753346, 'learning_rate': 1e-05, 'epoch': 0.81}
0: {'loss': 0.6842, 'grad_norm': 0.38643259595137086, 'learning_rate': 1e-05, 'epoch': 0.81}
0:   | 11870/14725 [10:48:28<2:33:07,  3.22s/it]
                                                          

 81%|████████  | 11870/14725 [10:48:28<2:33:07,  3.22s/it]
 81%|████████  | 11871/14725 [10:48:31<2:32:21,  3.20s/it]
 81%|████████  | 11872/14725 [10:48:35<2:33:07,  3.22s/it]
 81%|████████  | 11873/14725 [10:48:38<2:34:45,  3.26s/it]
 81%|████████  | 11874/14725 [10:48:42<2:37:43,  3.32s/it]
 81%|████████  | 11875/14725 [10:48:45<2:35:38,  3.28s/it]
 81%|████████  | 11876/14725 [10:48:48<2:35:48,  3.28s/it]
 81%|████████  | 11877/14725 [10:48:52<2:38:53,  3.35s/it]
 81%|████████  | 11878/14725 [10:48:55<2:38:01,  3.33s/it]
 81%|████████  | 11879/14725 [10:48:58<2:36:55,  3.31s/it]
 81%|████████  | 11880/14725 [10:49:01<2:35:25,  3.28s/it]
                                                          

 81%|████████  | 
0: {'loss': 0.6874, 'grad_norm': 0.413217330448847, 'learning_rate': 1e-05, 'epoch': 0.81}
0: 11880/14725 [10:49:01<2:35:25,  3.28s/it]
 81%|████████  | 11881/14725 [10:49:04<2:34:25,  3.26s/it]
 81%|████████  | 11882/14725 [10:49:08<2:35:24,  3.28s/it]
 81%|████████  | 11883/14725 [10:49:11<2:38:56,  3.36s/it]
 81%|████████  | 11884/14725 [10:49:15<2:36:01,  3.30s/it]
 81%|████████  | 11885/14725 [10:49:18<2:35:03,  3.28s/it]
 81%|████████  | 11886/14725 [10:49:21<2:36:59,  3.32s/it]
 81%|████████  | 11887/14725 [10:49:24<2:36:00,  3.30s/it]
 81%|████████  | 11888/14725 [10:49:28<2:37:59,  3.34s/it]
 81%|████████  | 11889/14725 [10:49:31<2:36:36,  3.31s/it]
 81%|████████  | 11890/14725 [10:49:34<2:36:16,  3.31s/it]
                                                          

 81%|████████  | 11890/14725 [10:49:34<2:36:16,  3.31s/it]
 81%|████████  | 11891/14725 [10:49:38<2:36:00,  3.30s/it]
 81%|█████�
0: {'loss': 0.6837, 'grad_norm': 0.4126769627442416, 'learning_rate': 1e-05, 'epoch': 0.81}
0: �██  | 11892/14725 [10:49:41<2:34:10,  3.27s/it]
 81%|████████  | 11893/14725 [10:49:44<2:34:49,  3.28s/it]
 81%|████████  | 11894/14725 [10:49:48<2:35:19,  3.29s/it]
 81%|████████  | 11895/14725 [10:49:51<2:35:54,  3.31s/it]
 81%|████████  | 11896/14725 [10:49:54<2:34:51,  3.28s/it]
 81%|████████  | 11897/14725 [10:49:58<2:39:07,  3.38s/it]
 81%|████████  | 11898/14725 [10:50:01<2:38:06,  3.36s/it]
 81%|████████  | 11899/14725 [10:50:04<2:36:25,  3.32s/it]
 81%|████████  | 11900/14725 [10:50:07<2:33:58,  3.27s/it]
                                                          

 81%|████████  | 11900/14725 [10:50:07<2:33:58,  3.27s/it]
 81%|████████  | 11901/14725 [10:50:11<2:32:33,  3.24s/it]
 81%|████████  | 11902/14725 [10:50:14<2:33:35,  3.26s/it]
 81%|████████  | 11903/14725 [10:50:17<2:33:10,  3.26s/it]
 81%|██
0: {'loss': 0.6781, 'grad_norm': 0.4411912682635061, 'learning_rate': 1e-05, 'epoch': 0.81}
0: ██████  | 11904/14725 [10:50:20<2:34:14,  3.28s/it]
 81%|████████  | 11905/14725 [10:50:24<2:35:57,  3.32s/it]
 81%|████████  | 11906/14725 [10:50:27<2:36:54,  3.34s/it]
 81%|████████  | 11907/14725 [10:50:31<2:37:48,  3.36s/it]
 81%|████████  | 11908/14725 [10:50:34<2:35:52,  3.32s/it]
 81%|████████  | 11909/14725 [10:50:37<2:35:02,  3.30s/it]
 81%|████████  | 11910/14725 [10:50:40<2:35:11,  3.31s/it]
                                                          

 81%|████████  | 11910/14725 [10:50:40<2:35:11,  3.31s/it]
 81%|████████  | 11911/14725 [10:50:44<2:34:17,  3.29s/it]
 81%|████████  | 11912/14725 [10:50:47<2:33:42,  3.28s/it]
 81%|████████  | 11913/14725 [10:50:50<2:32:58,  3.26s/it]
 81%|████████  | 11914/14725 [10:50:53<2:32:54,  3.26s/it]
 81%|████████  | 11915/14725 [10:50:57<2:35:40,  3.32s/it]
0: {'loss': 0.6774, 'grad_norm': 0.4113689420192849, 'learning_rate': 1e-05, 'epoch': 0.81}
0:  81%|████████  | 11916/14725 [10:51:00<2:36:09,  3.34s/it]
 81%|████████  | 11917/14725 [10:51:04<2:35:49,  3.33s/it]
 81%|████████  | 11918/14725 [10:51:07<2:34:56,  3.31s/it]
 81%|████████  | 11919/14725 [10:51:10<2:32:14,  3.26s/it]
 81%|████████  | 11920/14725 [10:51:13<2:34:29,  3.30s/it]
                                                          

 81%|████████  | 11920/14725 [10:51:13<2:34:29,  3.30s/it]
 81%|████████  | 11921/14725 [10:51:17<2:33:11,  3.28s/it]
 81%|████████  | 11922/14725 [10:51:20<2:31:26,  3.24s/it]
 81%|████████  | 11923/14725 [10:51:23<2:33:35,  3.29s/it]
 81%|████████  | 11924/14725 [10:51:26<2:33:27,  3.29s/it]
 81%|████████  | 11925/14725 [10:51:30<2:37:55,  3.38s/it]
 81%|████████  | 11926/14725 [10:51:33<2:34:01,  3.30s/it]
 81%|████████  | 11927/14725 [10:51:37<2:34:55, 
0: {'loss': 0.6715, 'grad_norm': 0.3966511946710179, 'learning_rate': 1e-05, 'epoch': 0.81}
0:  3.32s/it]
 81%|████████  | 11928/14725 [10:51:40<2:37:24,  3.38s/it]
 81%|████████  | 11929/14725 [10:51:43<2:36:09,  3.35s/it]
 81%|████████  | 11930/14725 [10:51:46<2:31:50,  3.26s/it]
                                                          

 81%|████████  | 11930/14725 [10:51:46<2:31:50,  3.26s/it]
 81%|████████  | 11931/14725 [10:51:50<2:31:08,  3.25s/it]
 81%|████████  | 11932/14725 [10:51:53<2:29:56,  3.22s/it]
 81%|████████  | 11933/14725 [10:51:56<2:32:03,  3.27s/it]
 81%|████████  | 11934/14725 [10:51:59<2:30:37,  3.24s/it]
 81%|████████  | 11935/14725 [10:52:03<2:30:31,  3.24s/it]
 81%|████████  | 11936/14725 [10:52:06<2:31:26,  3.26s/it]
 81%|████████  | 11937/14725 [10:52:09<2:34:09,  3.32s/it]
 81%|████████  | 11938/14725 [10:52:13<2:36:24,  3.37s/it]
 81%|████████  | 11939/14725 [10:52:1
0: {'loss': 0.6751, 'grad_norm': 0.4158322252332612, 'learning_rate': 1e-05, 'epoch': 0.81}
0: 6<2:33:44,  3.31s/it]
 81%|████████  | 11940/14725 [10:52:19<2:33:29,  3.31s/it]
                                                          

 81%|████████  | 11940/14725 [10:52:19<2:33:29,  3.31s/it]
 81%|████████  | 11941/14725 [10:52:23<2:33:03,  3.30s/it]
 81%|████████  | 11942/14725 [10:52:26<2:32:03,  3.28s/it]
 81%|████████  | 11943/14725 [10:52:29<2:34:44,  3.34s/it]
 81%|████████  | 11944/14725 [10:52:32<2:33:27,  3.31s/it]
 81%|████████  | 11945/14725 [10:52:36<2:31:33,  3.27s/it]
 81%|████████  | 11946/14725 [10:52:39<2:31:26,  3.27s/it]
 81%|████████  | 11947/14725 [10:52:42<2:33:40,  3.32s/it]
 81%|████████  | 11948/14725 [10:52:46<2:31:02,  3.26s/it]
 81%|████████  | 11949/14725 [10:52:49<2:30:45,  3.26s/it]
 81%|████████  | 11950/14725 [10:52:52<2:32:43,  3.30s/it]
                                          
0: {'loss': 0.6761, 'grad_norm': 0.4018677230365247, 'learning_rate': 1e-05, 'epoch': 0.81}
0: {'loss': 0.6673, 'grad_norm': 0.40168182593412577, 'learning_rate': 1e-05, 'epoch': 0.81}
0:                 

 81%|████████  | 11950/14725 [10:52:52<2:32:43,  3.30s/it]
 81%|████████  | 11951/14725 [10:52:55<2:31:38,  3.28s/it]
 81%|████████  | 11952/14725 [10:52:59<2:31:32,  3.28s/it]
 81%|████████  | 11953/14725 [10:53:02<2:30:00,  3.25s/it]
 81%|████████  | 11954/14725 [10:53:05<2:30:14,  3.25s/it]
 81%|████████  | 11955/14725 [10:53:08<2:30:40,  3.26s/it]
 81%|████████  | 11956/14725 [10:53:12<2:32:26,  3.30s/it]
 81%|████████  | 11957/14725 [10:53:15<2:33:12,  3.32s/it]
 81%|████████  | 11958/14725 [10:53:18<2:29:55,  3.25s/it]
 81%|████████  | 11959/14725 [10:53:22<2:30:25,  3.26s/it]
 81%|████████  | 11960/14725 [10:53:25<2:29:53,  3.25s/it]
                                                          

 81%|████████  | 11960/14725 [10:53:25<2:29:53,  3.25s/it]
 81%|████████  | 11961/14725 [
0: {'loss': 0.6691, 'grad_norm': 0.41280801642823933, 'learning_rate': 1e-05, 'epoch': 0.81}
0: 10:53:28<2:29:24,  3.24s/it]
 81%|████████  | 11962/14725 [10:53:31<2:28:25,  3.22s/it]
 81%|████████  | 11963/14725 [10:53:34<2:26:54,  3.19s/it]
 81%|████████  | 11964/14725 [10:53:38<2:28:56,  3.24s/it]
 81%|████████▏ | 11965/14725 [10:53:41<2:30:50,  3.28s/it]
 81%|████████▏ | 11966/14725 [10:53:44<2:30:22,  3.27s/it]
 81%|████████▏ | 11967/14725 [10:53:47<2:29:09,  3.24s/it]
 81%|████████▏ | 11968/14725 [10:53:51<2:30:26,  3.27s/it]
 81%|████████▏ | 11969/14725 [10:53:54<2:30:41,  3.28s/it]
 81%|████████▏ | 11970/14725 [10:53:57<2:30:37,  3.28s/it]
                                                          

 81%|████████▏ | 11970/14725 [10:53:57<2:30:37,  3.28s/it]
 81%|████████▏ | 11971/14725 [10:54:01<2:31:16,  3.30s/it]
 81%|████████▏ | 11972/14725 [10:54:04<2:31:11,  3.30s/it]
 81%|████
0: {'loss': 0.6648, 'grad_norm': 0.38317696488866015, 'learning_rate': 1e-05, 'epoch': 0.81}
0: ████▏ | 11973/14725 [10:54:07<2:29:49,  3.27s/it]
 81%|████████▏ | 11974/14725 [10:54:10<2:29:31,  3.26s/it]
 81%|████████▏ | 11975/14725 [10:54:14<2:30:12,  3.28s/it]
 81%|████████▏ | 11976/14725 [10:54:17<2:27:05,  3.21s/it]
 81%|████████▏ | 11977/14725 [10:54:20<2:29:23,  3.26s/it]
 81%|████████▏ | 11978/14725 [10:54:23<2:29:53,  3.27s/it]
 81%|████████▏ | 11979/14725 [10:54:27<2:28:28,  3.24s/it]
 81%|████████▏ | 11980/14725 [10:54:30<2:28:01,  3.24s/it]
                                                          

 81%|████████▏ | 11980/14725 [10:54:30<2:28:01,  3.24s/it]
 81%|████████▏ | 11981/14725 [10:54:33<2:28:31,  3.25s/it]
 81%|████████▏ | 11982/14725 [10:54:37<2:30:45,  3.30s/it]
 81%|████████▏ | 11983/14725 [10:54:40<2:29:54,  3.28s/it]
 81%|████████▏ | 11984/14725 [10:54:43<
0: {'loss': 0.6946, 'grad_norm': 0.4860919189545844, 'learning_rate': 1e-05, 'epoch': 0.81}
0: 2:31:53,  3.32s/it]
 81%|████████▏ | 11985/14725 [10:54:46<2:30:21,  3.29s/it]
 81%|████████▏ | 11986/14725 [10:54:50<2:30:29,  3.30s/it]
 81%|████████▏ | 11987/14725 [10:54:53<2:30:36,  3.30s/it]
 81%|████████▏ | 11988/14725 [10:54:56<2:29:31,  3.28s/it]
 81%|████████▏ | 11989/14725 [10:55:00<2:30:39,  3.30s/it]
 81%|████████▏ | 11990/14725 [10:55:03<2:31:26,  3.32s/it]
                                                          

 81%|████████▏ | 11990/14725 [10:55:03<2:31:26,  3.32s/it]
 81%|████████▏ | 11991/14725 [10:55:06<2:30:03,  3.29s/it]
 81%|████████▏ | 11992/14725 [10:55:09<2:28:23,  3.26s/it]
 81%|████████▏ | 11993/14725 [10:55:13<2:28:33,  3.26s/it]
 81%|████████▏ | 11994/14725 [10:55:16<2:30:29,  3.31s/it]
 81%|████████▏ | 11995/14725 [10:55:19<2:30:09,  3.30s/it]
 81%|█████
0: {'loss': 0.6851, 'grad_norm': 0.4163296951867278, 'learning_rate': 1e-05, 'epoch': 0.81}
0: ███▏ | 11996/14725 [10:55:23<2:29:46,  3.29s/it]
 81%|████████▏ | 11997/14725 [10:55:26<2:31:01,  3.32s/it]
 81%|████████▏ | 11998/14725 [10:55:29<2:30:47,  3.32s/it]
 81%|████████▏ | 11999/14725 [10:55:33<2:30:13,  3.31s/it]
 81%|████████▏ | 12000/14725 [10:55:36<2:34:26,  3.40s/it]
                                                          

 81%|████████▏ | 12000/14725 [10:55:36<2:34:26,  3.40s/it]
 82%|████████▏ | 12001/14725 [10:55:39<2:30:46,  3.32s/it]
 82%|████████▏ | 12002/14725 [10:55:43<2:29:21,  3.29s/it]
 82%|████████▏ | 12003/14725 [10:55:46<2:32:16,  3.36s/it]
 82%|████████▏ | 12004/14725 [10:55:49<2:31:08,  3.33s/it]
 82%|████████▏ | 12005/14725 [10:55:53<2:33:52,  3.39s/it]
 82%|████████▏ | 12006/14725 [10:55:56<2:30:54,  3.33s/it]
 82%|████████▏ | 12007/14725 [10:55:59<2:2
0: {'loss': 0.689, 'grad_norm': 0.4193618793375493, 'learning_rate': 1e-05, 'epoch': 0.82}
0: 9:42,  3.30s/it]
 82%|████████▏ | 12008/14725 [10:56:03<2:28:46,  3.29s/it]
 82%|████████▏ | 12009/14725 [10:56:06<2:27:12,  3.25s/it]
 82%|████████▏ | 12010/14725 [10:56:09<2:30:57,  3.34s/it]
                                                          

 82%|████████▏ | 12010/14725 [10:56:09<2:30:57,  3.34s/it]
 82%|████████▏ | 12011/14725 [10:56:13<2:29:34,  3.31s/it]
 82%|████████▏ | 12012/14725 [10:56:16<2:29:05,  3.30s/it]
 82%|████████▏ | 12013/14725 [10:56:19<2:29:23,  3.31s/it]
 82%|████████▏ | 12014/14725 [10:56:22<2:29:12,  3.30s/it]
 82%|████████▏ | 12015/14725 [10:56:26<2:30:19,  3.33s/it]
 82%|████████▏ | 12016/14725 [10:56:29<2:27:44,  3.27s/it]
 82%|████████▏ | 12017/14725 [10:56:32<2:26:25,  3.24s/it]
 82%|████████▏ | 12018/14725 [10:56:35<2:27:39,  3.27s/it]
 82%|██████
0: {'loss': 0.6824, 'grad_norm': 0.40575585692473914, 'learning_rate': 1e-05, 'epoch': 0.82}
0: ██▏ | 12019/14725 [10:56:39<2:27:49,  3.28s/it]
 82%|████████▏ | 12020/14725 [10:56:42<2:29:24,  3.31s/it]
                                                          

 82%|████████▏ | 12020/14725 [10:56:42<2:29:24,  3.31s/it]
 82%|████████▏ | 12021/14725 [10:56:45<2:27:43,  3.28s/it]
 82%|████████▏ | 12022/14725 [10:56:49<2:26:38,  3.26s/it]
 82%|████████▏ | 12023/14725 [10:56:52<2:25:24,  3.23s/it]
 82%|████████▏ | 12024/14725 [10:56:55<2:27:47,  3.28s/it]
 82%|████████▏ | 12025/14725 [10:56:58<2:25:50,  3.24s/it]
 82%|████████▏ | 12026/14725 [10:57:02<2:26:47,  3.26s/it]
 82%|████████▏ | 12027/14725 [10:57:05<2:24:46,  3.22s/it]
 82%|████████▏ | 12028/14725 [10:57:08<2:23:57,  3.20s/it]
 82%|████████▏ | 12029/14725 [10:57:11<2:26:58,  3.27s/it]
 82%|████████▏ | 12030/14725 [10:57:14<2:25:3
0: {'loss': 0.6827, 'grad_norm': 0.3919643603434153, 'learning_rate': 1e-05, 'epoch': 0.82}
0: {'loss': 0.6853, 'grad_norm': 0.3968926873088189, 'learning_rate': 1e-05, 'epoch': 0.82}
0: 7,  3.24s/it]
                                                          

 82%|████████▏ | 12030/14725 [10:57:14<2:25:37,  3.24s/it]
 82%|████████▏ | 12031/14725 [10:57:18<2:23:09,  3.19s/it]
 82%|████████▏ | 12032/14725 [10:57:21<2:23:41,  3.20s/it]
 82%|████████▏ | 12033/14725 [10:57:24<2:22:34,  3.18s/it]
 82%|████████▏ | 12034/14725 [10:57:27<2:22:54,  3.19s/it]
 82%|████████▏ | 12035/14725 [10:57:30<2:22:26,  3.18s/it]
 82%|████████▏ | 12036/14725 [10:57:34<2:23:57,  3.21s/it]
 82%|████████▏ | 12037/14725 [10:57:37<2:25:40,  3.25s/it]
 82%|████████▏ | 12038/14725 [10:57:40<2:25:12,  3.24s/it]
 82%|████████▏ | 12039/14725 [10:57:43<2:26:32,  3.27s/it]
 82%|████████▏ | 12040/14725 [10:57:47<2:25:54,  3.26s/it]
                                                          

 82%|████████▏ | 12040/14
0: {'loss': 0.6732, 'grad_norm': 0.3817767569687608, 'learning_rate': 1e-05, 'epoch': 0.82}
0: 725 [10:57:47<2:25:54,  3.26s/it]
 82%|████████▏ | 12041/14725 [10:57:50<2:23:31,  3.21s/it]
 82%|████████▏ | 12042/14725 [10:57:53<2:25:57,  3.26s/it]
 82%|████████▏ | 12043/14725 [10:57:57<2:29:09,  3.34s/it]
 82%|████████▏ | 12044/14725 [10:58:00<2:26:33,  3.28s/it]
 82%|████████▏ | 12045/14725 [10:58:03<2:24:00,  3.22s/it]
 82%|████████▏ | 12046/14725 [10:58:06<2:25:37,  3.26s/it]
 82%|████████▏ | 12047/14725 [10:58:10<2:25:55,  3.27s/it]
 82%|████████▏ | 12048/14725 [10:58:13<2:25:28,  3.26s/it]
 82%|████████▏ | 12049/14725 [10:58:16<2:23:05,  3.21s/it]
 82%|████████▏ | 12050/14725 [10:58:19<2:23:41,  3.22s/it]
                                                          

 82%|████████▏ | 12050/14725 [10:58:19<2:23:41,  3.22s/it]
 82%|████████▏ | 12051/14725 [10:58:23<2:27:27,  3.31s/it]
 82%|�
0: {'loss': 0.6811, 'grad_norm': 0.4171186185963709, 'learning_rate': 1e-05, 'epoch': 0.82}
0: ��███████▏ | 12052/14725 [10:58:26<2:26:28,  3.29s/it]
 82%|████████▏ | 12053/14725 [10:58:29<2:24:59,  3.26s/it]
 82%|████████▏ | 12054/14725 [10:58:32<2:23:46,  3.23s/it]
 82%|████████▏ | 12055/14725 [10:58:36<2:24:33,  3.25s/it]
 82%|████████▏ | 12056/14725 [10:58:39<2:25:27,  3.27s/it]
 82%|████████▏ | 12057/14725 [10:58:42<2:27:15,  3.31s/it]
 82%|████████▏ | 12058/14725 [10:58:46<2:27:53,  3.33s/it]
 82%|████████▏ | 12059/14725 [10:58:49<2:27:14,  3.31s/it]
 82%|████████▏ | 12060/14725 [10:58:52<2:26:18,  3.29s/it]
                                                          

 82%|████████▏ | 12060/14725 [10:58:52<2:26:18,  3.29s/it]
 82%|████████▏ | 12061/14725 [10:58:55<2:25:28,  3.28s/it]
 82%|████████▏ | 12062/14725 [10:58:59<2:29:39,  3.37s/it]
 82%|████████▏ | 12063/14725
0: {'loss': 0.6813, 'grad_norm': 0.3909463638246501, 'learning_rate': 1e-05, 'epoch': 0.82}
0:  [10:59:02<2:26:30,  3.30s/it]
 82%|████████▏ | 12064/14725 [10:59:05<2:23:48,  3.24s/it]
 82%|████████▏ | 12065/14725 [10:59:09<2:24:13,  3.25s/it]
 82%|████████▏ | 12066/14725 [10:59:12<2:23:38,  3.24s/it]
 82%|████████▏ | 12067/14725 [10:59:15<2:23:27,  3.24s/it]
 82%|████████▏ | 12068/14725 [10:59:18<2:23:39,  3.24s/it]
 82%|████████▏ | 12069/14725 [10:59:22<2:24:36,  3.27s/it]
 82%|████████▏ | 12070/14725 [10:59:25<2:25:47,  3.29s/it]
                                                          

 82%|████████▏ | 12070/14725 [10:59:25<2:25:47,  3.29s/it]
 82%|████████▏ | 12071/14725 [10:59:28<2:25:09,  3.28s/it]
 82%|████████▏ | 12072/14725 [10:59:31<2:24:06,  3.26s/it]
 82%|████████▏ | 12073/14725 [10:59:34<2:21:44,  3.21s/it]
 82%|████████▏ | 12074/14725 [10:59:38<2:21:46,  3.21s/it]
 82%|█�
0: {'loss': 0.6847, 'grad_norm': 0.43489991246686077, 'learning_rate': 1e-05, 'epoch': 0.82}
0: ��██████▏ | 12075/14725 [10:59:41<2:22:08,  3.22s/it]
 82%|████████▏ | 12076/14725 [10:59:44<2:22:24,  3.23s/it]
 82%|████████▏ | 12077/14725 [10:59:48<2:24:44,  3.28s/it]
 82%|████████▏ | 12078/14725 [10:59:51<2:23:52,  3.26s/it]
 82%|████████▏ | 12079/14725 [10:59:54<2:25:35,  3.30s/it]
 82%|████████▏ | 12080/14725 [10:59:58<2:26:08,  3.32s/it]
                                                          

 82%|████████▏ | 12080/14725 [10:59:58<2:26:08,  3.32s/it]
 82%|████████▏ | 12081/14725 [11:00:01<2:24:39,  3.28s/it]
 82%|████████▏ | 12082/14725 [11:00:04<2:23:35,  3.26s/it]
 82%|████████▏ | 12083/14725 [11:00:07<2:21:04,  3.20s/it]
 82%|████████▏ | 12084/14725 [11:00:11<2:25:33,  3.31s/it]
 82%|████████▏ | 12085/14725 [11:00:14<2:24:22,  3.28s/it]
 82%|████████▏ | 12086/14725 [1
0: {'loss': 0.6818, 'grad_norm': 0.42973636391719044, 'learning_rate': 1e-05, 'epoch': 0.82}
0: 1:00:17<2:22:52,  3.25s/it]
 82%|████████▏ | 12087/14725 [11:00:20<2:23:17,  3.26s/it]
 82%|████████▏ | 12088/14725 [11:00:23<2:21:07,  3.21s/it]
 82%|████████▏ | 12089/14725 [11:00:27<2:21:18,  3.22s/it]
 82%|████████▏ | 12090/14725 [11:00:30<2:22:25,  3.24s/it]
                                                          

 82%|████████▏ | 12090/14725 [11:00:30<2:22:25,  3.24s/it]
 82%|████████▏ | 12091/14725 [11:00:33<2:22:14,  3.24s/it]
 82%|████████▏ | 12092/14725 [11:00:36<2:22:17,  3.24s/it]
 82%|████████▏ | 12093/14725 [11:00:40<2:22:50,  3.26s/it]
 82%|████████▏ | 12094/14725 [11:00:43<2:20:50,  3.21s/it]
 82%|████████▏ | 12095/14725 [11:00:46<2:21:21,  3.22s/it]
 82%|████████▏ | 12096/14725 [11:00:49<2:22:38,  3.26s/it]
 82%|████████▏ | 12097/14725 [11:00:53<2:22:16,  3.25s/it]
 82%|██�
0: {'loss': 0.669, 'grad_norm': 0.44805380925503174, 'learning_rate': 1e-05, 'epoch': 0.82}
0: ��█████▏ | 12098/14725 [11:00:56<2:20:39,  3.21s/it]
 82%|████████▏ | 12099/14725 [11:00:59<2:19:36,  3.19s/it]
 82%|████████▏ | 12100/14725 [11:01:02<2:21:28,  3.23s/it]
                                                          

 82%|████████▏ | 12100/14725 [11:01:02<2:21:28,  3.23s/it]
 82%|████████▏ | 12101/14725 [11:01:05<2:22:48,  3.27s/it]
 82%|████████▏ | 12102/14725 [11:01:09<2:25:06,  3.32s/it]
 82%|████████▏ | 12103/14725 [11:01:12<2:24:01,  3.30s/it]
 82%|████████▏ | 12104/14725 [11:01:15<2:24:08,  3.30s/it]
 82%|████████▏ | 12105/14725 [11:01:19<2:22:02,  3.25s/it]
 82%|████████▏ | 12106/14725 [11:01:22<2:21:43,  3.25s/it]
 82%|████████▏ | 12107/14725 [11:01:25<2:22:51,  3.27s/it]
 82%|████████▏ | 12108/14725 [11:01:28<2:22:43,  3.27s/it]
 82%|████████▏ | 12109/14725 [11:0
0: {'loss': 0.6851, 'grad_norm': 0.4080545012545257, 'learning_rate': 1e-05, 'epoch': 0.82}
0: 1:32<2:21:45,  3.25s/it]
 82%|████████▏ | 12110/14725 [11:01:35<2:21:36,  3.25s/it]
                                                          

 82%|████████▏ | 12110/14725 [11:01:35<2:21:36,  3.25s/it]
 82%|████████▏ | 12111/14725 [11:01:38<2:24:29,  3.32s/it]
 82%|████████▏ | 12112/14725 [11:01:42<2:25:17,  3.34s/it]
 82%|████████▏ | 12113/14725 [11:01:45<2:24:31,  3.32s/it]
 82%|████████▏ | 12114/14725 [11:01:48<2:23:42,  3.30s/it]
 82%|████████▏ | 12115/14725 [11:01:52<2:25:58,  3.36s/it]
 82%|████████▏ | 12116/14725 [11:01:55<2:23:20,  3.30s/it]
 82%|████████▏ | 12117/14725 [11:01:58<2:25:29,  3.35s/it]
 82%|████████▏ | 12118/14725 [11:02:02<2:25:20,  3.35s/it]
 82%|████████▏ | 12119/14725 [11:02:05<2:26:21,  3.37s/it]
 82%|████████▏ | 12120/14725 [11:02:08<2:24:38,  3.33s/it]
               
0: {'loss': 0.6785, 'grad_norm': 0.3861047593403469, 'learning_rate': 1e-05, 'epoch': 0.82}
0: {'loss': 0.674, 'grad_norm': 0.40323594323439543, 'learning_rate': 1e-05, 'epoch': 0.82}
0:                                            

 82%|████████▏ | 12120/14725 [11:02:08<2:24:38,  3.33s/it]
 82%|████████▏ | 12121/14725 [11:02:12<2:24:30,  3.33s/it]
 82%|████████▏ | 12122/14725 [11:02:15<2:25:59,  3.37s/it]
 82%|████████▏ | 12123/14725 [11:02:19<2:26:08,  3.37s/it]
 82%|████████▏ | 12124/14725 [11:02:22<2:25:08,  3.35s/it]
 82%|████████▏ | 12125/14725 [11:02:25<2:21:59,  3.28s/it]
 82%|████████▏ | 12126/14725 [11:02:28<2:23:20,  3.31s/it]
 82%|████████▏ | 12127/14725 [11:02:32<2:22:35,  3.29s/it]
 82%|████████▏ | 12128/14725 [11:02:35<2:21:36,  3.27s/it]
 82%|████████▏ | 12129/14725 [11:02:38<2:19:50,  3.23s/it]
 82%|████████▏ | 12130/14725 [11:02:41<2:19:46,  3.23s/it]
                                                          

 82%|████████▏ | 12130/14725 [11:02:41<2:19:46,  3.23s
0: {'loss': 0.6758, 'grad_norm': 0.43215510668419516, 'learning_rate': 1e-05, 'epoch': 0.82}
0: /it]
 82%|████████▏ | 12131/14725 [11:02:45<2:21:32,  3.27s/it]
 82%|████████▏ | 12132/14725 [11:02:48<2:21:34,  3.28s/it]
 82%|████████▏ | 12133/14725 [11:02:51<2:19:06,  3.22s/it]
 82%|████████▏ | 12134/14725 [11:02:54<2:18:53,  3.22s/it]
 82%|████████▏ | 12135/14725 [11:02:57<2:20:06,  3.25s/it]
 82%|████████▏ | 12136/14725 [11:03:01<2:22:54,  3.31s/it]
 82%|████████▏ | 12137/14725 [11:03:04<2:22:30,  3.30s/it]
 82%|████████▏ | 12138/14725 [11:03:07<2:20:26,  3.26s/it]
 82%|████████▏ | 12139/14725 [11:03:11<2:19:52,  3.25s/it]
 82%|████████▏ | 12140/14725 [11:03:14<2:20:49,  3.27s/it]
                                                          

 82%|████████▏ | 12140/14725 [11:03:14<2:20:49,  3.27s/it]
 82%|████████▏ | 12141/14725 [11:03:17<2:20:42,  3.27s/it]
 82%|████████▏ | 
0: {'loss': 0.6705, 'grad_norm': 0.41256581027701916, 'learning_rate': 1e-05, 'epoch': 0.83}
0: 12142/14725 [11:03:20<2:19:31,  3.24s/it]
 82%|████████▏ | 12143/14725 [11:03:24<2:20:40,  3.27s/it]
 82%|████████▏ | 12144/14725 [11:03:27<2:20:15,  3.26s/it]
 82%|████████▏ | 12145/14725 [11:03:30<2:21:21,  3.29s/it]
 82%|████████▏ | 12146/14725 [11:03:34<2:20:23,  3.27s/it]
 82%|████████▏ | 12147/14725 [11:03:37<2:17:35,  3.20s/it]
 82%|████████▏ | 12148/14725 [11:03:40<2:17:16,  3.20s/it]
 83%|████████▎ | 12149/14725 [11:03:43<2:18:19,  3.22s/it]
 83%|████████▎ | 12150/14725 [11:03:46<2:17:48,  3.21s/it]
                                                          

 83%|████████▎ | 12150/14725 [11:03:46<2:17:48,  3.21s/it]
 83%|████████▎ | 12151/14725 [11:03:49<2:15:32,  3.16s/it]
 83%|████████▎ | 12152/14725 [11:03:52<2:16:32,  3.18s/it]
 83%|████████▎ | 12153/14725 [11:03:56<2:19:56,  3.26s/it
0: {'loss': 0.686, 'grad_norm': 0.39841407444085675, 'learning_rate': 1e-05, 'epoch': 0.83}
0: ]
 83%|████████▎ | 12154/14725 [11:03:59<2:20:55,  3.29s/it]
 83%|████████▎ | 12155/14725 [11:04:03<2:22:38,  3.33s/it]
 83%|████████▎ | 12156/14725 [11:04:06<2:20:36,  3.28s/it]
 83%|████████▎ | 12157/14725 [11:04:09<2:19:33,  3.26s/it]
 83%|████████▎ | 12158/14725 [11:04:12<2:18:54,  3.25s/it]
 83%|████████▎ | 12159/14725 [11:04:15<2:18:06,  3.23s/it]
 83%|████████▎ | 12160/14725 [11:04:19<2:18:11,  3.23s/it]
                                                          

 83%|████████▎ | 12160/14725 [11:04:19<2:18:11,  3.23s/it]
 83%|████████▎ | 12161/14725 [11:04:22<2:19:37,  3.27s/it]
 83%|████████▎ | 12162/14725 [11:04:25<2:20:14,  3.28s/it]
 83%|████████▎ | 12163/14725 [11:04:29<2:18:48,  3.25s/it]
 83%|████████▎ | 12164/14725 [11:04:32<2:17:55,  3.23s/it]
 83%|████████▎ | 121
0: {'loss': 0.6833, 'grad_norm': 0.41199858859169936, 'learning_rate': 1e-05, 'epoch': 0.83}
0: 65/14725 [11:04:35<2:17:28,  3.22s/it]
 83%|████████▎ | 12166/14725 [11:04:38<2:19:06,  3.26s/it]
 83%|████████▎ | 12167/14725 [11:04:41<2:16:44,  3.21s/it]
 83%|████████▎ | 12168/14725 [11:04:45<2:16:13,  3.20s/it]
 83%|████████▎ | 12169/14725 [11:04:48<2:18:36,  3.25s/it]
 83%|████████▎ | 12170/14725 [11:04:51<2:16:43,  3.21s/it]
                                                          

 83%|████████▎ | 12170/14725 [11:04:51<2:16:43,  3.21s/it]
 83%|████████▎ | 12171/14725 [11:04:54<2:15:54,  3.19s/it]
 83%|████████▎ | 12172/14725 [11:04:58<2:19:16,  3.27s/it]
 83%|████████▎ | 12173/14725 [11:05:01<2:19:55,  3.29s/it]
 83%|████████▎ | 12174/14725 [11:05:04<2:17:03,  3.22s/it]
 83%|████████▎ | 12175/14725 [11:05:07<2:18:45,  3.27s/it]
 83%|████████▎ | 12176/14725 [11:05:11<2:17:00,  3.23s/it]
 
0: {'loss': 0.6718, 'grad_norm': 0.41983349585332674, 'learning_rate': 1e-05, 'epoch': 0.83}
0: 83%|████████▎ | 12177/14725 [11:05:14<2:19:53,  3.29s/it]
 83%|████████▎ | 12178/14725 [11:05:18<2:24:24,  3.40s/it]
 83%|████████▎ | 12179/14725 [11:05:21<2:22:48,  3.37s/it]
 83%|████████▎ | 12180/14725 [11:05:24<2:20:25,  3.31s/it]
                                                          

 83%|████████▎ | 12180/14725 [11:05:24<2:20:25,  3.31s/it]
 83%|████████▎ | 12181/14725 [11:05:27<2:20:24,  3.31s/it]
 83%|████████▎ | 12182/14725 [11:05:31<2:20:17,  3.31s/it]
 83%|████████▎ | 12183/14725 [11:05:34<2:20:12,  3.31s/it]
 83%|████████▎ | 12184/14725 [11:05:37<2:18:41,  3.27s/it]
 83%|████████▎ | 12185/14725 [11:05:40<2:17:31,  3.25s/it]
 83%|████████▎ | 12186/14725 [11:05:44<2:17:09,  3.24s/it]
 83%|████████▎ | 12187/14725 [11:05:47<2:17:14,  3.24s/it]
 83%|████████▎ | 12188/
0: {'loss': 0.6845, 'grad_norm': 0.4023609030414479, 'learning_rate': 1e-05, 'epoch': 0.83}
0: 14725 [11:05:50<2:18:39,  3.28s/it]
 83%|████████▎ | 12189/14725 [11:05:53<2:17:32,  3.25s/it]
 83%|████████▎ | 12190/14725 [11:05:57<2:16:10,  3.22s/it]
                                                          

 83%|████████▎ | 12190/14725 [11:05:57<2:16:10,  3.22s/it]
 83%|████████▎ | 12191/14725 [11:06:00<2:22:15,  3.37s/it]
 83%|████████▎ | 12192/14725 [11:06:04<2:21:23,  3.35s/it]
 83%|████████▎ | 12193/14725 [11:06:07<2:19:12,  3.30s/it]
 83%|████████▎ | 12194/14725 [11:06:10<2:16:25,  3.23s/it]
 83%|████████▎ | 12195/14725 [11:06:13<2:14:47,  3.20s/it]
 83%|████████▎ | 12196/14725 [11:06:16<2:17:46,  3.27s/it]
 83%|████████▎ | 12197/14725 [11:06:20<2:17:12,  3.26s/it]
 83%|████████▎ | 12198/14725 [11:06:23<2:18:21,  3.28s/it]
 83%|████████▎ | 12199/14725 [11:06:26<2:19:17,  3.31s/it]
 83%
0: {'loss': 0.6625, 'grad_norm': 0.4307955751041169, 'learning_rate': 1e-05, 'epoch': 0.83}
0: |████████▎ | 12200/14725 [11:06:30<2:19:10,  3.31s/it]
                                                          

 83%|████████▎ | 12200/14725 [11:06:30<2:19:10,  3.31s/it]
 83%|████████▎ | 12201/14725 [11:06:33<2:17:50,  3.28s/it]
 83%|████████▎ | 12202/14725 [11:06:36<2:18:46,  3.30s/it]
 83%|████████▎ | 12203/14725 [11:06:39<2:17:34,  3.27s/it]
 83%|████████▎ | 12204/14725 [11:06:43<2:17:53,  3.28s/it]
 83%|████████▎ | 12205/14725 [11:06:46<2:19:13,  3.32s/it]
 83%|████████▎ | 12206/14725 [11:06:49<2:17:35,  3.28s/it]
 83%|████████▎ | 12207/14725 [11:06:53<2:19:44,  3.33s/it]
 83%|████████▎ | 12208/14725 [11:06:56<2:19:18,  3.32s/it]
 83%|████████▎ | 12209/14725 [11:06:59<2:16:58,  3.27s/it]
 83%|████████▎ | 12210/14725 [11:07:03<2:18:32,  3.31s/it]
                                            
0: {'loss': 0.6912, 'grad_norm': 0.4073051913245642, 'learning_rate': 1e-05, 'epoch': 0.83}
0: {'loss': 0.6802, 'grad_norm': 0.4053009335045191, 'learning_rate': 1e-05, 'epoch': 0.83}
0:               

 83%|████████▎ | 12210/14725 [11:07:03<2:18:32,  3.31s/it]
 83%|████████▎ | 12211/14725 [11:07:06<2:17:37,  3.28s/it]
 83%|████████▎ | 12212/14725 [11:07:09<2:20:47,  3.36s/it]
 83%|████████▎ | 12213/14725 [11:07:13<2:18:03,  3.30s/it]
 83%|████████▎ | 12214/14725 [11:07:16<2:16:26,  3.26s/it]
 83%|████████▎ | 12215/14725 [11:07:19<2:15:51,  3.25s/it]
 83%|████████▎ | 12216/14725 [11:07:22<2:18:35,  3.31s/it]
 83%|████████▎ | 12217/14725 [11:07:26<2:22:14,  3.40s/it]
 83%|████████▎ | 12218/14725 [11:07:29<2:18:11,  3.31s/it]
 83%|████████▎ | 12219/14725 [11:07:32<2:18:51,  3.32s/it]
 83%|████████▎ | 12220/14725 [11:07:36<2:21:28,  3.39s/it]
                                                          

 83%|████████▎ | 12220/14725 [11:07:36<2:21:28,  3.39s/it]
 83%|██████�
0: {'loss': 0.6822, 'grad_norm': 0.4221305733117207, 'learning_rate': 1e-05, 'epoch': 0.83}
0: ��█▎ | 12221/14725 [11:07:40<2:25:54,  3.50s/it]
 83%|████████▎ | 12222/14725 [11:07:43<2:23:42,  3.45s/it]
 83%|████████▎ | 12223/14725 [11:07:46<2:22:51,  3.43s/it]
 83%|████████▎ | 12224/14725 [11:07:50<2:18:03,  3.31s/it]
 83%|████████▎ | 12225/14725 [11:07:53<2:18:26,  3.32s/it]
 83%|████████▎ | 12226/14725 [11:07:56<2:17:11,  3.29s/it]
 83%|████████▎ | 12227/14725 [11:07:59<2:17:43,  3.31s/it]
 83%|████████▎ | 12228/14725 [11:08:02<2:13:21,  3.20s/it]
 83%|████████▎ | 12229/14725 [11:08:06<2:14:38,  3.24s/it]
 83%|████████▎ | 12230/14725 [11:08:09<2:17:05,  3.30s/it]
                                                          

 83%|████████▎ | 12230/14725 [11:08:09<2:17:05,  3.30s/it]
 83%|████████▎ | 12231/14725 [11:08:12<2:15:57,  3.27s/it]
 83%|████████▎ | 12232/14725 [11:08:16<2:15:16
0: {'loss': 0.6743, 'grad_norm': 0.39969619145554447, 'learning_rate': 1e-05, 'epoch': 0.83}
0: ,  3.26s/it]
 83%|████████▎ | 12233/14725 [11:08:19<2:15:28,  3.26s/it]
 83%|████████▎ | 12234/14725 [11:08:22<2:16:29,  3.29s/it]
 83%|████████▎ | 12235/14725 [11:08:26<2:18:39,  3.34s/it]
 83%|████████▎ | 12236/14725 [11:08:29<2:17:23,  3.31s/it]
 83%|████████▎ | 12237/14725 [11:08:32<2:13:36,  3.22s/it]
 83%|████████▎ | 12238/14725 [11:08:35<2:12:40,  3.20s/it]
 83%|████████▎ | 12239/14725 [11:08:38<2:14:08,  3.24s/it]
 83%|████████▎ | 12240/14725 [11:08:42<2:13:05,  3.21s/it]
                                                          

 83%|████████▎ | 12240/14725 [11:08:42<2:13:05,  3.21s/it]
 83%|████████▎ | 12241/14725 [11:08:45<2:12:26,  3.20s/it]
 83%|████████▎ | 12242/14725 [11:08:48<2:14:42,  3.26s/it]
 83%|████████▎ | 12243/14725 [11:08:51<2:16:07,  3.29s/it]
 83%|███████�
0: {'loss': 0.6758, 'grad_norm': 0.3866025925965811, 'learning_rate': 1e-05, 'epoch': 0.83}
0: ��▎ | 12244/14725 [11:08:55<2:15:49,  3.28s/it]
 83%|████████▎ | 12245/14725 [11:08:58<2:15:50,  3.29s/it]
 83%|████████▎ | 12246/14725 [11:09:01<2:15:23,  3.28s/it]
 83%|████████▎ | 12247/14725 [11:09:04<2:12:43,  3.21s/it]
 83%|████████▎ | 12248/14725 [11:09:07<2:10:05,  3.15s/it]
 83%|████████▎ | 12249/14725 [11:09:11<2:10:46,  3.17s/it]
 83%|████████▎ | 12250/14725 [11:09:14<2:10:59,  3.18s/it]
                                                          

 83%|████████▎ | 12250/14725 [11:09:14<2:10:59,  3.18s/it]
 83%|████████▎ | 12251/14725 [11:09:17<2:11:46,  3.20s/it]
 83%|████████▎ | 12252/14725 [11:09:20<2:11:27,  3.19s/it]
 83%|████████▎ | 12253/14725 [11:09:23<2:10:19,  3.16s/it]
 83%|████████▎ | 12254/14725 [11:09:27<2:11:55,  3.20s/it]
 83%|████████▎ | 12255/14725 [11:09:30<2:12:34,  
0: {'loss': 0.6772, 'grad_norm': 0.38830250447836495, 'learning_rate': 1e-05, 'epoch': 0.83}
0: 3.22s/it]
 83%|████████▎ | 12256/14725 [11:09:33<2:13:11,  3.24s/it]
 83%|████████▎ | 12257/14725 [11:09:36<2:12:32,  3.22s/it]
 83%|████████▎ | 12258/14725 [11:09:40<2:12:24,  3.22s/it]
 83%|████████▎ | 12259/14725 [11:09:43<2:11:39,  3.20s/it]
 83%|████████▎ | 12260/14725 [11:09:46<2:12:55,  3.24s/it]
                                                          

 83%|████████▎ | 12260/14725 [11:09:46<2:12:55,  3.24s/it]
 83%|████████▎ | 12261/14725 [11:09:49<2:10:50,  3.19s/it]
 83%|████████▎ | 12262/14725 [11:09:52<2:11:23,  3.20s/it]
 83%|████████▎ | 12263/14725 [11:09:56<2:13:08,  3.24s/it]
 83%|████████▎ | 12264/14725 [11:09:59<2:11:38,  3.21s/it]
 83%|████████▎ | 12265/14725 [11:10:02<2:13:40,  3.26s/it]
 83%|████████▎ | 12266/14725 [11:10:05<2:13:05,  3.25s/it]
 83%|████████�
0: {'loss': 0.6798, 'grad_norm': 0.37634411926726186, 'learning_rate': 1e-05, 'epoch': 0.83}
0: �� | 12267/14725 [11:10:09<2:14:08,  3.27s/it]
 83%|████████▎ | 12268/14725 [11:10:12<2:12:30,  3.24s/it]
 83%|████████▎ | 12269/14725 [11:10:15<2:13:00,  3.25s/it]
 83%|████████▎ | 12270/14725 [11:10:19<2:15:13,  3.31s/it]
                                                          

 83%|████████▎ | 12270/14725 [11:10:19<2:15:13,  3.31s/it]
 83%|████████▎ | 12271/14725 [11:10:22<2:12:34,  3.24s/it]
 83%|████████▎ | 12272/14725 [11:10:25<2:12:56,  3.25s/it]
 83%|████████▎ | 12273/14725 [11:10:28<2:14:39,  3.30s/it]
 83%|████████▎ | 12274/14725 [11:10:32<2:13:31,  3.27s/it]
 83%|████████▎ | 12275/14725 [11:10:35<2:12:32,  3.25s/it]
 83%|████████▎ | 12276/14725 [11:10:38<2:13:11,  3.26s/it]
 83%|████████▎ | 12277/14725 [11:10:42<2:15:56,  3.33s/it]
 83%|████████▎ | 12278/14725 [11:10:45<2:14:53,  3.3
0: {'loss': 0.685, 'grad_norm': 0.41238441563323813, 'learning_rate': 1e-05, 'epoch': 0.83}
0: 1s/it]
 83%|████████▎ | 12279/14725 [11:10:48<2:15:49,  3.33s/it]
 83%|████████▎ | 12280/14725 [11:10:52<2:15:45,  3.33s/it]
                                                          

 83%|████████▎ | 12280/14725 [11:10:52<2:15:45,  3.33s/it]
 83%|████████▎ | 12281/14725 [11:10:55<2:14:50,  3.31s/it]
 83%|████████▎ | 12282/14725 [11:10:58<2:16:21,  3.35s/it]
 83%|████████▎ | 12283/14725 [11:11:01<2:13:20,  3.28s/it]
 83%|████████▎ | 12284/14725 [11:11:05<2:13:22,  3.28s/it]
 83%|████████▎ | 12285/14725 [11:11:08<2:14:21,  3.30s/it]
 83%|████████▎ | 12286/14725 [11:11:11<2:15:26,  3.33s/it]
 83%|████████▎ | 12287/14725 [11:11:15<2:13:35,  3.29s/it]
 83%|████████▎ | 12288/14725 [11:11:18<2:13:47,  3.29s/it]
 83%|████████▎ | 12289/14725 [11:11:21<2:14:14,  3.31s/it]
 83%|████████▎ 
0: {'loss': 0.6674, 'grad_norm': 0.4055730537460346, 'learning_rate': 1e-05, 'epoch': 0.83}
0: {'loss': 0.6873, 'grad_norm': 0.45067046180644516, 'learning_rate': 1e-05, 'epoch': 0.84}
0: | 12290/14725 [11:11:24<2:11:42,  3.25s/it]
                                                          

 83%|████████▎ | 12290/14725 [11:11:24<2:11:42,  3.25s/it]
 83%|████████▎ | 12291/14725 [11:11:27<2:10:59,  3.23s/it]
 83%|████████▎ | 12292/14725 [11:11:31<2:14:24,  3.31s/it]
 83%|████████▎ | 12293/14725 [11:11:34<2:12:20,  3.26s/it]
 83%|████████▎ | 12294/14725 [11:11:38<2:13:52,  3.30s/it]
 83%|████████▎ | 12295/14725 [11:11:41<2:13:42,  3.30s/it]
 84%|████████▎ | 12296/14725 [11:11:44<2:12:37,  3.28s/it]
 84%|████████▎ | 12297/14725 [11:11:47<2:10:06,  3.22s/it]
 84%|████████▎ | 12298/14725 [11:11:50<2:09:57,  3.21s/it]
 84%|████████▎ | 12299/14725 [11:11:54<2:09:22,  3.20s/it]
 84%|████████▎ | 12300/14725 [11:11:57<2:10:10,  3.22s/it]
                                                          

 84%|██�
0: {'loss': 0.667, 'grad_norm': 0.3874642530755041, 'learning_rate': 1e-05, 'epoch': 0.84}
0: �█████▎ | 12300/14725 [11:11:57<2:10:10,  3.22s/it]
 84%|████████▎ | 12301/14725 [11:12:00<2:09:34,  3.21s/it]
 84%|████████▎ | 12302/14725 [11:12:03<2:11:43,  3.26s/it]
 84%|████████▎ | 12303/14725 [11:12:07<2:12:13,  3.28s/it]
 84%|████████▎ | 12304/14725 [11:12:10<2:11:07,  3.25s/it]
 84%|████████▎ | 12305/14725 [11:12:13<2:10:37,  3.24s/it]
 84%|████████▎ | 12306/14725 [11:12:16<2:11:17,  3.26s/it]
 84%|████████▎ | 12307/14725 [11:12:19<2:09:34,  3.22s/it]
 84%|████████▎ | 12308/14725 [11:12:23<2:13:52,  3.32s/it]
 84%|████████▎ | 12309/14725 [11:12:26<2:12:20,  3.29s/it]
 84%|████████▎ | 12310/14725 [11:12:29<2:11:39,  3.27s/it]
                                                          

 84%|████████▎ | 12310/14725 [11:12:29<2:11:39,  3.27s/it]
 84%|████████▎ | 12311/14725 [11:12
0: {'loss': 0.6678, 'grad_norm': 0.3943206931871914, 'learning_rate': 1e-05, 'epoch': 0.84}
0: :33<2:13:34,  3.32s/it]
 84%|████████▎ | 12312/14725 [11:12:36<2:12:05,  3.28s/it]
 84%|████████▎ | 12313/14725 [11:12:39<2:11:41,  3.28s/it]
 84%|████████▎ | 12314/14725 [11:12:43<2:11:28,  3.27s/it]
 84%|████████▎ | 12315/14725 [11:12:46<2:08:41,  3.20s/it]
 84%|████████▎ | 12316/14725 [11:12:49<2:07:32,  3.18s/it]
 84%|████████▎ | 12317/14725 [11:12:52<2:07:49,  3.18s/it]
 84%|████████▎ | 12318/14725 [11:12:55<2:08:10,  3.20s/it]
 84%|████████▎ | 12319/14725 [11:12:59<2:10:43,  3.26s/it]
 84%|████████▎ | 12320/14725 [11:13:02<2:08:40,  3.21s/it]
                                                          

 84%|████████▎ | 12320/14725 [11:13:02<2:08:40,  3.21s/it]
 84%|████████▎ | 12321/14725 [11:13:05<2:08:29,  3.21s/it]
 84%|████████▎ | 12322/14725 [11:13:08<2:08:33,  3.21s/it]
 84%|███�
0: {'loss': 0.6711, 'grad_norm': 0.39818063713120144, 'learning_rate': 1e-05, 'epoch': 0.84}
0: �████▎ | 12323/14725 [11:13:11<2:08:06,  3.20s/it]
 84%|████████▎ | 12324/14725 [11:13:14<2:06:39,  3.17s/it]
 84%|████████▎ | 12325/14725 [11:13:18<2:09:15,  3.23s/it]
 84%|████████▎ | 12326/14725 [11:13:21<2:10:58,  3.28s/it]
 84%|████████▎ | 12327/14725 [11:13:24<2:10:30,  3.27s/it]
 84%|████████▎ | 12328/14725 [11:13:28<2:10:22,  3.26s/it]
 84%|████████▎ | 12329/14725 [11:13:31<2:09:47,  3.25s/it]
 84%|████████▎ | 12330/14725 [11:13:34<2:08:58,  3.23s/it]
                                                          

 84%|████████▎ | 12330/14725 [11:13:34<2:08:58,  3.23s/it]
 84%|████████▎ | 12331/14725 [11:13:37<2:09:13,  3.24s/it]
 84%|████████▎ | 12332/14725 [11:13:41<2:11:44,  3.30s/it]
 84%|████████▍ | 12333/14725 [11:13:44<2:10:25,  3.27s/it]
 84%|████████▍ | 12334/14725 [11:13:47
0: {'loss': 0.6664, 'grad_norm': 0.4119946610697205, 'learning_rate': 1e-05, 'epoch': 0.84}
0: <2:08:47,  3.23s/it]
 84%|████████▍ | 12335/14725 [11:13:50<2:07:32,  3.20s/it]
 84%|████████▍ | 12336/14725 [11:13:54<2:11:00,  3.29s/it]
 84%|████████▍ | 12337/14725 [11:13:57<2:09:43,  3.26s/it]
 84%|████████▍ | 12338/14725 [11:14:00<2:12:14,  3.32s/it]
 84%|████████▍ | 12339/14725 [11:14:04<2:10:27,  3.28s/it]
 84%|████████▍ | 12340/14725 [11:14:07<2:07:29,  3.21s/it]
                                                          

 84%|████████▍ | 12340/14725 [11:14:07<2:07:29,  3.21s/it]
 84%|████████▍ | 12341/14725 [11:14:10<2:06:23,  3.18s/it]
 84%|████████▍ | 12342/14725 [11:14:13<2:07:05,  3.20s/it]
 84%|████████▍ | 12343/14725 [11:14:16<2:07:49,  3.22s/it]
 84%|████████▍ | 12344/14725 [11:14:20<2:08:14,  3.23s/it]
 84%|████████▍ | 12345/14725 [11:14:23<2:09:02,  3.25s/it]
 84%|████�
0: {'loss': 0.679, 'grad_norm': 0.4317693489670967, 'learning_rate': 1e-05, 'epoch': 0.84}
0: �███▍ | 12346/14725 [11:14:26<2:08:53,  3.25s/it]
 84%|████████▍ | 12347/14725 [11:14:30<2:11:05,  3.31s/it]
 84%|████████▍ | 12348/14725 [11:14:33<2:12:35,  3.35s/it]
 84%|████████▍ | 12349/14725 [11:14:36<2:10:51,  3.30s/it]
 84%|████████▍ | 12350/14725 [11:14:39<2:10:35,  3.30s/it]
                                                          

 84%|████████▍ | 12350/14725 [11:14:39<2:10:35,  3.30s/it]
 84%|████████▍ | 12351/14725 [11:14:43<2:10:33,  3.30s/it]
 84%|████████▍ | 12352/14725 [11:14:46<2:10:03,  3.29s/it]
 84%|████████▍ | 12353/14725 [11:14:49<2:10:35,  3.30s/it]
 84%|████████▍ | 12354/14725 [11:14:53<2:09:14,  3.27s/it]
 84%|████████▍ | 12355/14725 [11:14:56<2:08:38,  3.26s/it]
 84%|████████▍ | 12356/14725 [11:14:59<2:08:25,  3.25s/it]
 84%|████████▍ | 12357/14725 [11:15:02<2:
0: {'loss': 0.6697, 'grad_norm': 0.38011801603870016, 'learning_rate': 1e-05, 'epoch': 0.84}
0: 07:11,  3.22s/it]
 84%|████████▍ | 12358/14725 [11:15:05<2:08:05,  3.25s/it]
 84%|████████▍ | 12359/14725 [11:15:09<2:10:29,  3.31s/it]
 84%|████████▍ | 12360/14725 [11:15:12<2:09:12,  3.28s/it]
                                                          

 84%|████████▍ | 12360/14725 [11:15:12<2:09:12,  3.28s/it]
 84%|████████▍ | 12361/14725 [11:15:15<2:09:48,  3.29s/it]
 84%|████████▍ | 12362/14725 [11:15:19<2:09:06,  3.28s/it]
 84%|████████▍ | 12363/14725 [11:15:22<2:07:58,  3.25s/it]
 84%|████████▍ | 12364/14725 [11:15:25<2:09:30,  3.29s/it]
 84%|████████▍ | 12365/14725 [11:15:28<2:08:03,  3.26s/it]
 84%|████████▍ | 12366/14725 [11:15:32<2:07:43,  3.25s/it]
 84%|████████▍ | 12367/14725 [11:15:35<2:08:52,  3.28s/it]
 84%|████████▍ | 12368/14725 [11:15:38<2:09:05,  3.29s/it]
 84%|█████�
0: {'loss': 0.6788, 'grad_norm': 0.39997900316498086, 'learning_rate': 1e-05, 'epoch': 0.84}
0: �██▍ | 12369/14725 [11:15:42<2:07:55,  3.26s/it]
 84%|████████▍ | 12370/14725 [11:15:45<2:05:28,  3.20s/it]
                                                          

 84%|████████▍ | 12370/14725 [11:15:45<2:05:28,  3.20s/it]
 84%|████████▍ | 12371/14725 [11:15:48<2:04:28,  3.17s/it]
 84%|████████▍ | 12372/14725 [11:15:51<2:04:11,  3.17s/it]
 84%|████████▍ | 12373/14725 [11:15:55<2:10:48,  3.34s/it]
 84%|████████▍ | 12374/14725 [11:15:58<2:10:08,  3.32s/it]
 84%|████████▍ | 12375/14725 [11:16:01<2:07:51,  3.26s/it]
 84%|████████▍ | 12376/14725 [11:16:04<2:07:47,  3.26s/it]
 84%|████████▍ | 12377/14725 [11:16:08<2:08:10,  3.28s/it]
 84%|████████▍ | 12378/14725 [11:16:11<2:08:34,  3.29s/it]
 84%|████████▍ | 12379/14725 [11:16:14<2:08:46,  3.29s/it]
 84%|████████▍ | 12380/14725 [11:16:17<2:08:
0: {'loss': 0.6781, 'grad_norm': 0.4727686975527431, 'learning_rate': 1e-05, 'epoch': 0.84}
0: {'loss': 0.6564, 'grad_norm': 0.4571008007781111, 'learning_rate': 1e-05, 'epoch': 0.84}
0: 30,  3.29s/it]
                                                          

 84%|████████▍ | 12380/14725 [11:16:17<2:08:30,  3.29s/it]
 84%|████████▍ | 12381/14725 [11:16:20<2:05:27,  3.21s/it]
 84%|████████▍ | 12382/14725 [11:16:24<2:05:31,  3.21s/it]
 84%|████████▍ | 12383/14725 [11:16:27<2:04:51,  3.20s/it]
 84%|████████▍ | 12384/14725 [11:16:30<2:02:42,  3.15s/it]
 84%|████████▍ | 12385/14725 [11:16:33<2:04:55,  3.20s/it]
 84%|████████▍ | 12386/14725 [11:16:36<2:04:00,  3.18s/it]
 84%|████████▍ | 12387/14725 [11:16:40<2:04:46,  3.20s/it]
 84%|████████▍ | 12388/14725 [11:16:43<2:06:19,  3.24s/it]
 84%|████████▍ | 12389/14725 [11:16:46<2:04:33,  3.20s/it]
 84%|████████▍ | 12390/14725 [11:16:49<2:07:17,  3.27s/it]
                                                          

 84%|████████▍ | 12390/1
0: {'loss': 0.6809, 'grad_norm': 0.41263345434082765, 'learning_rate': 1e-05, 'epoch': 0.84}
0: 4725 [11:16:49<2:07:17,  3.27s/it]
 84%|████████▍ | 12391/14725 [11:16:53<2:05:22,  3.22s/it]
 84%|████████▍ | 12392/14725 [11:16:56<2:05:02,  3.22s/it]
 84%|████████▍ | 12393/14725 [11:16:59<2:04:16,  3.20s/it]
 84%|████████▍ | 12394/14725 [11:17:02<2:03:23,  3.18s/it]
 84%|████████▍ | 12395/14725 [11:17:06<2:06:47,  3.26s/it]
 84%|████████▍ | 12396/14725 [11:17:09<2:07:53,  3.29s/it]
 84%|████████▍ | 12397/14725 [11:17:12<2:08:38,  3.32s/it]
 84%|████████▍ | 12398/14725 [11:17:16<2:11:36,  3.39s/it]
 84%|████████▍ | 12399/14725 [11:17:19<2:12:06,  3.41s/it]
 84%|████████▍ | 12400/14725 [11:17:23<2:10:42,  3.37s/it]
                                                          

 84%|████████▍ | 12400/14725 [11:17:23<2:10:42,  3.37s/it]
 84%|████████▍ | 12401/14725 [11:17:26<2:09:18,  3.34s/it]
 84%|
0: {'loss': 0.6771, 'grad_norm': 0.46893205633947155, 'learning_rate': 1e-05, 'epoch': 0.84}
0: ████████▍ | 12402/14725 [11:17:29<2:07:23,  3.29s/it]
 84%|████████▍ | 12403/14725 [11:17:32<2:06:04,  3.26s/it]
 84%|████████▍ | 12404/14725 [11:17:36<2:06:36,  3.27s/it]
 84%|████████▍ | 12405/14725 [11:17:39<2:07:13,  3.29s/it]
 84%|████████▍ | 12406/14725 [11:17:42<2:06:42,  3.28s/it]
 84%|████████▍ | 12407/14725 [11:17:45<2:05:25,  3.25s/it]
 84%|████████▍ | 12408/14725 [11:17:49<2:07:06,  3.29s/it]
 84%|████████▍ | 12409/14725 [11:17:52<2:04:39,  3.23s/it]
 84%|████████▍ | 12410/14725 [11:17:55<2:07:22,  3.30s/it]
                                                          

 84%|████████▍ | 12410/14725 [11:17:55<2:07:22,  3.30s/it]
 84%|████████▍ | 12411/14725 [11:17:58<2:05:12,  3.25s/it]
 84%|████████▍ | 12412/14725 [11:18:02<2:06:10,  3.27s/it]
 84%|████████▍ | 12413/1472
0: {'loss': 0.6703, 'grad_norm': 0.3875206469970168, 'learning_rate': 1e-05, 'epoch': 0.84}
0: 5 [11:18:05<2:06:02,  3.27s/it]
 84%|████████▍ | 12414/14725 [11:18:08<2:08:22,  3.33s/it]
 84%|████████▍ | 12415/14725 [11:18:12<2:06:42,  3.29s/it]
 84%|████████▍ | 12416/14725 [11:18:15<2:05:58,  3.27s/it]
 84%|████████▍ | 12417/14725 [11:18:18<2:05:04,  3.25s/it]
 84%|████████▍ | 12418/14725 [11:18:21<2:05:44,  3.27s/it]
 84%|████████▍ | 12419/14725 [11:18:25<2:06:22,  3.29s/it]
 84%|████████▍ | 12420/14725 [11:18:28<2:05:38,  3.27s/it]
                                                          

 84%|████████▍ | 12420/14725 [11:18:28<2:05:38,  3.27s/it]
 84%|████████▍ | 12421/14725 [11:18:31<2:06:17,  3.29s/it]
 84%|████████▍ | 12422/14725 [11:18:34<2:03:59,  3.23s/it]
 84%|████████▍ | 12423/14725 [11:18:37<2:02:56,  3.20s/it]
 84%|████████▍ | 12424/14725 [11:18:41<2:05:08,  3.26s/it]
 84%|█
0: {'loss': 0.6613, 'grad_norm': 0.4204362550133749, 'learning_rate': 1e-05, 'epoch': 0.84}
0: ███████▍ | 12425/14725 [11:18:44<2:04:10,  3.24s/it]
 84%|████████▍ | 12426/14725 [11:18:47<2:05:27,  3.27s/it]
 84%|████████▍ | 12427/14725 [11:18:51<2:05:20,  3.27s/it]
 84%|████████▍ | 12428/14725 [11:18:54<2:06:09,  3.30s/it]
 84%|████████▍ | 12429/14725 [11:18:57<2:05:03,  3.27s/it]
 84%|████████▍ | 12430/14725 [11:19:00<2:04:31,  3.26s/it]
                                                          

 84%|████████▍ | 12430/14725 [11:19:00<2:04:31,  3.26s/it]
 84%|████████▍ | 12431/14725 [11:19:04<2:05:24,  3.28s/it]
 84%|████████▍ | 12432/14725 [11:19:07<2:05:04,  3.27s/it]
 84%|████████▍ | 12433/14725 [11:19:10<2:05:51,  3.29s/it]
 84%|████████▍ | 12434/14725 [11:19:14<2:04:27,  3.26s/it]
 84%|████████▍ | 12435/14725 [11:19:17<2:05:23,  3.29s/it]
 84%|████████▍ | 12436/14725 [
0: {'loss': 0.6926, 'grad_norm': 0.40509717777723525, 'learning_rate': 1e-05, 'epoch': 0.84}
0: 11:19:20<2:06:04,  3.30s/it]
 84%|████████▍ | 12437/14725 [11:19:23<2:04:18,  3.26s/it]
 84%|████████▍ | 12438/14725 [11:19:27<2:03:24,  3.24s/it]
 84%|████████▍ | 12439/14725 [11:19:30<2:03:53,  3.25s/it]
 84%|████████▍ | 12440/14725 [11:19:33<2:04:05,  3.26s/it]
                                                          

 84%|████████▍ | 12440/14725 [11:19:33<2:04:05,  3.26s/it]
 84%|████████▍ | 12441/14725 [11:19:37<2:05:31,  3.30s/it]
 84%|████████▍ | 12442/14725 [11:19:40<2:04:55,  3.28s/it]
 85%|████████▍ | 12443/14725 [11:19:43<2:04:17,  3.27s/it]
 85%|████████▍ | 12444/14725 [11:19:46<2:02:58,  3.23s/it]
 85%|████████▍ | 12445/14725 [11:19:50<2:05:29,  3.30s/it]
 85%|████████▍ | 12446/14725 [11:19:53<2:03:44,  3.26s/it]
 85%|████████▍ | 12447/14725 [11:19:56<2:01:17,  3.19s/it]
 85%|██
0: {'loss': 0.6784, 'grad_norm': 0.45419478236715816, 'learning_rate': 1e-05, 'epoch': 0.85}
0: ██████▍ | 12448/14725 [11:19:59<1:59:36,  3.15s/it]
 85%|████████▍ | 12449/14725 [11:20:02<1:59:32,  3.15s/it]
 85%|████████▍ | 12450/14725 [11:20:05<2:00:49,  3.19s/it]
                                                          

 85%|████████▍ | 12450/14725 [11:20:05<2:00:49,  3.19s/it]
 85%|████████▍ | 12451/14725 [11:20:09<2:02:00,  3.22s/it]
 85%|████████▍ | 12452/14725 [11:20:12<2:02:53,  3.24s/it]
 85%|████████▍ | 12453/14725 [11:20:15<2:03:52,  3.27s/it]
 85%|████████▍ | 12454/14725 [11:20:19<2:04:13,  3.28s/it]
 85%|████████▍ | 12455/14725 [11:20:22<2:02:53,  3.25s/it]
 85%|████████▍ | 12456/14725 [11:20:25<2:02:09,  3.23s/it]
 85%|████████▍ | 12457/14725 [11:20:28<2:02:17,  3.24s/it]
 85%|████████▍ | 12458/14725 [11:20:31<2:02:13,  3.24s/it]
 85%|████████▍ | 12459/14725 [11:
0: {'loss': 0.6848, 'grad_norm': 0.4017262150091424, 'learning_rate': 1e-05, 'epoch': 0.85}
0: 20:35<2:00:45,  3.20s/it]
 85%|████████▍ | 12460/14725 [11:20:38<2:00:33,  3.19s/it]
                                                          

 85%|████████▍ | 12460/14725 [11:20:38<2:00:33,  3.19s/it]
 85%|████████▍ | 12461/14725 [11:20:41<1:59:51,  3.18s/it]
 85%|████████▍ | 12462/14725 [11:20:44<2:00:14,  3.19s/it]
 85%|████████▍ | 12463/14725 [11:20:47<2:01:07,  3.21s/it]
 85%|████████▍ | 12464/14725 [11:20:51<2:01:56,  3.24s/it]
 85%|████████▍ | 12465/14725 [11:20:54<2:02:41,  3.26s/it]
 85%|████████▍ | 12466/14725 [11:20:57<2:02:55,  3.26s/it]
 85%|████████▍ | 12467/14725 [11:21:01<2:03:47,  3.29s/it]
 85%|████████▍ | 12468/14725 [11:21:04<2:02:29,  3.26s/it]
 85%|████████▍ | 12469/14725 [11:21:07<2:02:04,  3.25s/it]
 85%|████████▍ | 12470/14725 [11:21:10<2:04:16,  3.31s/it]
              
0: {'loss': 0.6633, 'grad_norm': 0.41041345232754556, 'learning_rate': 1e-05, 'epoch': 0.85}
0: {'loss': 0.6713, 'grad_norm': 0.40541525701063147, 'learning_rate': 1e-05, 'epoch': 0.85}
0:                                             

 85%|████████▍ | 12470/14725 [11:21:10<2:04:16,  3.31s/it]
 85%|████████▍ | 12471/14725 [11:21:14<2:02:12,  3.25s/it]
 85%|████████▍ | 12472/14725 [11:21:17<2:00:44,  3.22s/it]
 85%|████████▍ | 12473/14725 [11:21:20<2:01:33,  3.24s/it]
 85%|████████▍ | 12474/14725 [11:21:23<2:00:56,  3.22s/it]
 85%|████████▍ | 12475/14725 [11:21:26<1:59:54,  3.20s/it]
 85%|████████▍ | 12476/14725 [11:21:30<2:00:22,  3.21s/it]
 85%|████████▍ | 12477/14725 [11:21:33<2:01:41,  3.25s/it]
 85%|████████▍ | 12478/14725 [11:21:36<2:03:54,  3.31s/it]
 85%|████████▍ | 12479/14725 [11:21:39<2:02:29,  3.27s/it]
 85%|████████▍ | 12480/14725 [11:21:43<2:01:02,  3.24s/it]
                                                          

 85%|████████▍ | 12480/14725 [11:21:43<2:01:02,  3.24
0: {'loss': 0.6837, 'grad_norm': 0.448161898934887, 'learning_rate': 1e-05, 'epoch': 0.85}
0: s/it]
 85%|████████▍ | 12481/14725 [11:21:46<2:01:38,  3.25s/it]
 85%|████████▍ | 12482/14725 [11:21:49<2:01:46,  3.26s/it]
 85%|████████▍ | 12483/14725 [11:21:52<2:00:58,  3.24s/it]
 85%|████████▍ | 12484/14725 [11:21:56<2:02:47,  3.29s/it]
 85%|████████▍ | 12485/14725 [11:21:59<2:01:42,  3.26s/it]
 85%|████████▍ | 12486/14725 [11:22:02<2:02:52,  3.29s/it]
 85%|████████▍ | 12487/14725 [11:22:05<2:00:55,  3.24s/it]
 85%|████████▍ | 12488/14725 [11:22:09<2:00:34,  3.23s/it]
 85%|████████▍ | 12489/14725 [11:22:12<2:01:21,  3.26s/it]
 85%|████████▍ | 12490/14725 [11:22:15<2:02:14,  3.28s/it]
                                                          

 85%|████████▍ | 12490/14725 [11:22:15<2:02:14,  3.28s/it]
 85%|████████▍ | 12491/14725 [11:22:19<2:02:51,  3.30s/it]
 85%|████████▍ |
0: {'loss': 0.6691, 'grad_norm': 0.41686943247868746, 'learning_rate': 1e-05, 'epoch': 0.85}
0:  12492/14725 [11:22:22<2:04:19,  3.34s/it]
 85%|████████▍ | 12493/14725 [11:22:25<2:02:44,  3.30s/it]
 85%|████████▍ | 12494/14725 [11:22:29<2:02:00,  3.28s/it]
 85%|████████▍ | 12495/14725 [11:22:32<2:01:33,  3.27s/it]
 85%|████████▍ | 12496/14725 [11:22:35<1:59:25,  3.21s/it]
 85%|████████▍ | 12497/14725 [11:22:38<1:59:25,  3.22s/it]
 85%|████████▍ | 12498/14725 [11:22:41<2:00:28,  3.25s/it]
 85%|████████▍ | 12499/14725 [11:22:45<2:02:16,  3.30s/it]
 85%|████████▍ | 12500/14725 [11:22:48<2:03:35,  3.33s/it]
                                                          

 85%|████████▍ | 12500/14725 [11:22:48<2:03:35,  3.33s/it]
 85%|████████▍ | 12501/14725 [11:22:52<2:02:26,  3.30s/it]
 85%|████████▍ | 12502/14725 [11:22:55<2:01:02,  3.27s/it]
 85%|████████▍ | 12503/14725 [11:22:58<2:02:53,  3.32s/i
0: {'loss': 0.6807, 'grad_norm': 0.4221702993071932, 'learning_rate': 1e-05, 'epoch': 0.85}
0: t]
 85%|████████▍ | 12504/14725 [11:23:02<2:03:33,  3.34s/it]
 85%|████████▍ | 12505/14725 [11:23:05<2:00:06,  3.25s/it]
 85%|████████▍ | 12506/14725 [11:23:08<2:01:21,  3.28s/it]
 85%|████████▍ | 12507/14725 [11:23:11<2:00:41,  3.26s/it]
 85%|████████▍ | 12508/14725 [11:23:14<2:01:21,  3.28s/it]
 85%|████████▍ | 12509/14725 [11:23:18<2:03:53,  3.35s/it]
 85%|████████▍ | 12510/14725 [11:23:21<2:01:46,  3.30s/it]
                                                          

 85%|████████▍ | 12510/14725 [11:23:21<2:01:46,  3.30s/it]
 85%|████████▍ | 12511/14725 [11:23:24<2:00:09,  3.26s/it]
 85%|████████▍ | 12512/14725 [11:23:28<2:00:34,  3.27s/it]
 85%|████████▍ | 12513/14725 [11:23:31<1:59:38,  3.25s/it]
 85%|████████▍ | 12514/14725 [11:23:34<2:00:20,  3.27s/it]
 85%|████████▍ | 12
0: {'loss': 0.6775, 'grad_norm': 0.40966332713790404, 'learning_rate': 1e-05, 'epoch': 0.85}
0: 515/14725 [11:23:37<1:59:46,  3.25s/it]
 85%|████████▍ | 12516/14725 [11:23:41<2:00:46,  3.28s/it]
 85%|████████▌ | 12517/14725 [11:23:44<1:58:45,  3.23s/it]
 85%|████████▌ | 12518/14725 [11:23:47<1:59:09,  3.24s/it]
 85%|████████▌ | 12519/14725 [11:23:50<1:58:53,  3.23s/it]
 85%|████████▌ | 12520/14725 [11:23:54<2:00:04,  3.27s/it]
                                                          

 85%|████████▌ | 12520/14725 [11:23:54<2:00:04,  3.27s/it]
 85%|████████▌ | 12521/14725 [11:23:57<1:58:50,  3.24s/it]
 85%|████████▌ | 12522/14725 [11:24:00<1:59:16,  3.25s/it]
 85%|████████▌ | 12523/14725 [11:24:03<1:58:33,  3.23s/it]
 85%|████████▌ | 12524/14725 [11:24:06<1:57:00,  3.19s/it]
 85%|████████▌ | 12525/14725 [11:24:10<1:57:05,  3.19s/it]
 85%|████████▌ | 12526/14725 [11:24:13<1:56:34,  3.18s/it]
0: {'loss': 0.6721, 'grad_norm': 0.41230945774838923, 'learning_rate': 1e-05, 'epoch': 0.85}
0:  85%|████████▌ | 12527/14725 [11:24:16<1:59:06,  3.25s/it]
 85%|████████▌ | 12528/14725 [11:24:19<1:59:16,  3.26s/it]
 85%|████████▌ | 12529/14725 [11:24:23<1:59:29,  3.26s/it]
 85%|████████▌ | 12530/14725 [11:24:26<2:01:18,  3.32s/it]
                                                          

 85%|████████▌ | 12530/14725 [11:24:26<2:01:18,  3.32s/it]
 85%|████████▌ | 12531/14725 [11:24:29<1:59:56,  3.28s/it]
 85%|████████▌ | 12532/14725 [11:24:33<2:00:19,  3.29s/it]
 85%|████████▌ | 12533/14725 [11:24:36<2:00:02,  3.29s/it]
 85%|████████▌ | 12534/14725 [11:24:39<2:02:37,  3.36s/it]
 85%|████████▌ | 12535/14725 [11:24:43<2:02:54,  3.37s/it]
 85%|████████▌ | 12536/14725 [11:24:46<2:00:47,  3.31s/it]
 85%|████████▌ | 12537/14725 [11:24:49<1:59:32,  3.28s/it]
 85%|████████▌ | 12538
0: {'loss': 0.6722, 'grad_norm': 0.40082301664920317, 'learning_rate': 1e-05, 'epoch': 0.85}
0: /14725 [11:24:52<1:59:06,  3.27s/it]
 85%|████████▌ | 12539/14725 [11:24:56<1:59:04,  3.27s/it]
 85%|████████▌ | 12540/14725 [11:24:59<1:59:10,  3.27s/it]
                                                          

 85%|████████▌ | 12540/14725 [11:24:59<1:59:10,  3.27s/it]
 85%|████████▌ | 12541/14725 [11:25:02<1:58:23,  3.25s/it]
 85%|████████▌ | 12542/14725 [11:25:05<1:57:50,  3.24s/it]
 85%|████████▌ | 12543/14725 [11:25:09<2:00:07,  3.30s/it]
 85%|████████▌ | 12544/14725 [11:25:12<1:59:35,  3.29s/it]
 85%|████████▌ | 12545/14725 [11:25:15<1:57:57,  3.25s/it]
 85%|████████▌ | 12546/14725 [11:25:18<1:57:13,  3.23s/it]
 85%|████████▌ | 12547/14725 [11:25:22<1:57:08,  3.23s/it]
 85%|████████▌ | 12548/14725 [11:25:25<1:55:11,  3.17s/it]
 85%|████████▌ | 12549/14725 [11:25:28<1:56:39,  3.22s/it]
 85
0: {'loss': 0.6725, 'grad_norm': 0.39639632518004836, 'learning_rate': 1e-05, 'epoch': 0.85}
0: %|████████▌ | 12550/14725 [11:25:31<1:57:04,  3.23s/it]
                                                          

 85%|████████▌ | 12550/14725 [11:25:31<1:57:04,  3.23s/it]
 85%|████████▌ | 12551/14725 [11:25:35<1:57:59,  3.26s/it]
 85%|████████▌ | 12552/14725 [11:25:38<1:57:37,  3.25s/it]
 85%|████████▌ | 12553/14725 [11:25:41<1:55:44,  3.20s/it]
 85%|████████▌ | 12554/14725 [11:25:44<1:57:30,  3.25s/it]
 85%|████████▌ | 12555/14725 [11:25:48<1:59:22,  3.30s/it]
 85%|████████▌ | 12556/14725 [11:25:51<1:59:01,  3.29s/it]
 85%|████████▌ | 12557/14725 [11:25:54<1:59:47,  3.32s/it]
 85%|████████▌ | 12558/14725 [11:25:58<2:00:23,  3.33s/it]
 85%|████████▌ | 12559/14725 [11:26:01<2:01:56,  3.38s/it]
 85%|████████▌ | 12560/14725 [11:26:05<2:01:14,  3.36s/it]
                                           
0: {'loss': 0.6661, 'grad_norm': 0.44012965011948546, 'learning_rate': 1e-05, 'epoch': 0.85}
0: {'loss': 0.6719, 'grad_norm': 0.4346147524499878, 'learning_rate': 1e-05, 'epoch': 0.85}
0:                

 85%|████████▌ | 12560/14725 [11:26:05<2:01:14,  3.36s/it]
 85%|████████▌ | 12561/14725 [11:26:08<2:00:18,  3.34s/it]
 85%|████████▌ | 12562/14725 [11:26:11<2:01:26,  3.37s/it]
 85%|████████▌ | 12563/14725 [11:26:15<2:00:49,  3.35s/it]
 85%|████████▌ | 12564/14725 [11:26:18<1:58:44,  3.30s/it]
 85%|████████▌ | 12565/14725 [11:26:21<1:56:15,  3.23s/it]
 85%|████████▌ | 12566/14725 [11:26:24<1:55:57,  3.22s/it]
 85%|████████▌ | 12567/14725 [11:26:27<1:57:27,  3.27s/it]
 85%|████████▌ | 12568/14725 [11:26:31<1:57:25,  3.27s/it]
 85%|████████▌ | 12569/14725 [11:26:34<1:57:26,  3.27s/it]
 85%|████████▌ | 12570/14725 [11:26:37<1:57:47,  3.28s/it]
                                                          

 85%|████████▌ | 12570/14725 [11:26:37<1:57:47,  3.28s/it]
 85%|██████
0: {'loss': 0.6746, 'grad_norm': 0.4426431452552096, 'learning_rate': 1e-05, 'epoch': 0.85}
0: ██▌ | 12571/14725 [11:26:40<1:56:13,  3.24s/it]
 85%|████████▌ | 12572/14725 [11:26:44<1:57:07,  3.26s/it]
 85%|████████▌ | 12573/14725 [11:26:47<1:56:53,  3.26s/it]
 85%|████████▌ | 12574/14725 [11:26:50<1:56:42,  3.26s/it]
 85%|████████▌ | 12575/14725 [11:26:53<1:56:11,  3.24s/it]
 85%|████████▌ | 12576/14725 [11:26:57<1:57:09,  3.27s/it]
 85%|████████▌ | 12577/14725 [11:27:00<1:57:06,  3.27s/it]
 85%|████████▌ | 12578/14725 [11:27:03<1:55:18,  3.22s/it]
 85%|████████▌ | 12579/14725 [11:27:06<1:55:56,  3.24s/it]
 85%|████████▌ | 12580/14725 [11:27:10<1:57:16,  3.28s/it]
                                                          

 85%|████████▌ | 12580/14725 [11:27:10<1:57:16,  3.28s/it]
 85%|████████▌ | 12581/14725 [11:27:13<1:56:57,  3.27s/it]
 85%|████████▌ | 12582/14725 [11:27:16<1:58:4
0: {'loss': 0.6779, 'grad_norm': 0.40852738221104534, 'learning_rate': 1e-05, 'epoch': 0.85}
0: 1,  3.32s/it]
 85%|████████▌ | 12583/14725 [11:27:20<1:57:21,  3.29s/it]
 85%|████████▌ | 12584/14725 [11:27:23<1:57:08,  3.28s/it]
 85%|████████▌ | 12585/14725 [11:27:26<1:55:26,  3.24s/it]
 85%|████████▌ | 12586/14725 [11:27:29<1:55:07,  3.23s/it]
 85%|████████▌ | 12587/14725 [11:27:33<1:56:56,  3.28s/it]
 85%|████████▌ | 12588/14725 [11:27:36<1:56:51,  3.28s/it]
 85%|████████▌ | 12589/14725 [11:27:39<1:55:39,  3.25s/it]
 86%|████████▌ | 12590/14725 [11:27:42<1:55:46,  3.25s/it]
                                                          

 86%|████████▌ | 12590/14725 [11:27:42<1:55:46,  3.25s/it]
 86%|████████▌ | 12591/14725 [11:27:46<1:57:22,  3.30s/it]
 86%|████████▌ | 12592/14725 [11:27:49<1:57:31,  3.31s/it]
 86%|████████▌ | 12593/14725 [11:27:52<1:55:54,  3.26s/it]
 86%|███████
0: {'loss': 0.6733, 'grad_norm': 0.4016491424826099, 'learning_rate': 1e-05, 'epoch': 0.86}
0: █▌ | 12594/14725 [11:27:56<1:55:34,  3.25s/it]
 86%|████████▌ | 12595/14725 [11:27:59<1:58:03,  3.33s/it]
 86%|████████▌ | 12596/14725 [11:28:02<1:56:43,  3.29s/it]
 86%|████████▌ | 12597/14725 [11:28:05<1:55:36,  3.26s/it]
 86%|████████▌ | 12598/14725 [11:28:09<1:54:22,  3.23s/it]
 86%|████████▌ | 12599/14725 [11:28:12<1:58:47,  3.35s/it]
 86%|████████▌ | 12600/14725 [11:28:15<1:57:41,  3.32s/it]
                                                          

 86%|████████▌ | 12600/14725 [11:28:15<1:57:41,  3.32s/it]
 86%|████████▌ | 12601/14725 [11:28:19<1:59:35,  3.38s/it]
 86%|████████▌ | 12602/14725 [11:28:22<1:56:40,  3.30s/it]
 86%|████████▌ | 12603/14725 [11:28:25<1:57:01,  3.31s/it]
 86%|████████▌ | 12604/14725 [11:28:29<1:55:59,  3.28s/it]
 86%|████████▌ | 12605/14725 [11:28:32<1:56:52, 
0: {'loss': 0.6687, 'grad_norm': 0.39351577336732085, 'learning_rate': 1e-05, 'epoch': 0.86}
0:  3.31s/it]
 86%|████████▌ | 12606/14725 [11:28:35<1:56:22,  3.30s/it]
 86%|████████▌ | 12607/14725 [11:28:39<1:57:03,  3.32s/it]
 86%|████████▌ | 12608/14725 [11:28:42<1:54:04,  3.23s/it]
 86%|████████▌ | 12609/14725 [11:28:45<1:52:52,  3.20s/it]
 86%|████████▌ | 12610/14725 [11:28:48<1:53:45,  3.23s/it]
                                                          

 86%|████████▌ | 12610/14725 [11:28:48<1:53:45,  3.23s/it]
 86%|████████▌ | 12611/14725 [11:28:51<1:52:59,  3.21s/it]
 86%|████████▌ | 12612/14725 [11:28:54<1:52:40,  3.20s/it]
 86%|████████▌ | 12613/14725 [11:28:58<1:53:15,  3.22s/it]
 86%|████████▌ | 12614/14725 [11:29:01<1:57:08,  3.33s/it]
 86%|████████▌ | 12615/14725 [11:29:05<1:56:14,  3.31s/it]
 86%|████████▌ | 12616/14725 [11:29:08<1:57:55,  3.35s/it]
 86%|████████
0: {'loss': 0.6752, 'grad_norm': 0.4062335438898971, 'learning_rate': 1e-05, 'epoch': 0.86}
0: ▌ | 12617/14725 [11:29:11<1:55:18,  3.28s/it]
 86%|████████▌ | 12618/14725 [11:29:14<1:53:41,  3.24s/it]
 86%|████████▌ | 12619/14725 [11:29:18<1:54:43,  3.27s/it]
 86%|████████▌ | 12620/14725 [11:29:21<1:54:02,  3.25s/it]
                                                          

 86%|████████▌ | 12620/14725 [11:29:21<1:54:02,  3.25s/it]
 86%|████████▌ | 12621/14725 [11:29:24<1:55:00,  3.28s/it]
 86%|████████▌ | 12622/14725 [11:29:28<1:57:07,  3.34s/it]
 86%|████████▌ | 12623/14725 [11:29:31<1:57:03,  3.34s/it]
 86%|████████▌ | 12624/14725 [11:29:34<1:56:54,  3.34s/it]
 86%|████████▌ | 12625/14725 [11:29:37<1:53:54,  3.25s/it]
 86%|████████▌ | 12626/14725 [11:29:41<1:53:30,  3.24s/it]
 86%|████████▌ | 12627/14725 [11:29:44<1:53:08,  3.24s/it]
 86%|████████▌ | 12628/14725 [11:29:47<1:53:51,  3.
0: {'loss': 0.6881, 'grad_norm': 0.38948386697188897, 'learning_rate': 1e-05, 'epoch': 0.86}
0: 26s/it]
 86%|████████▌ | 12629/14725 [11:29:50<1:54:57,  3.29s/it]
 86%|████████▌ | 12630/14725 [11:29:54<1:52:56,  3.23s/it]
                                                          

 86%|████████▌ | 12630/14725 [11:29:54<1:52:56,  3.23s/it]
 86%|████████▌ | 12631/14725 [11:29:57<1:52:24,  3.22s/it]
 86%|████████▌ | 12632/14725 [11:30:00<1:51:14,  3.19s/it]
 86%|████████▌ | 12633/14725 [11:30:03<1:52:15,  3.22s/it]
 86%|████████▌ | 12634/14725 [11:30:07<1:53:50,  3.27s/it]
 86%|████████▌ | 12635/14725 [11:30:10<1:52:53,  3.24s/it]
 86%|████████▌ | 12636/14725 [11:30:13<1:52:19,  3.23s/it]
 86%|████████▌ | 12637/14725 [11:30:16<1:53:01,  3.25s/it]
 86%|████████▌ | 12638/14725 [11:30:19<1:51:43,  3.21s/it]
 86%|████████▌ | 12639/14725 [11:30:23<1:54:49,  3.30s/it]
 86%|████████▌
0: {'loss': 0.6762, 'grad_norm': 0.412524725303282, 'learning_rate': 1e-05, 'epoch': 0.86}
0: {'loss': 0.664, 'grad_norm': 0.4701053442889733, 'learning_rate': 1e-05, 'epoch': 0.86}
0:  | 12640/14725 [11:30:26<1:53:56,  3.28s/it]
                                                          

 86%|████████▌ | 12640/14725 [11:30:26<1:53:56,  3.28s/it]
 86%|████████▌ | 12641/14725 [11:30:29<1:51:31,  3.21s/it]
 86%|████████▌ | 12642/14725 [11:30:33<1:54:30,  3.30s/it]
 86%|████████▌ | 12643/14725 [11:30:36<1:53:16,  3.26s/it]
 86%|████████▌ | 12644/14725 [11:30:39<1:52:37,  3.25s/it]
 86%|████████▌ | 12645/14725 [11:30:42<1:51:16,  3.21s/it]
 86%|████████▌ | 12646/14725 [11:30:45<1:51:02,  3.20s/it]
 86%|████████▌ | 12647/14725 [11:30:49<1:51:05,  3.21s/it]
 86%|████████▌ | 12648/14725 [11:30:52<1:52:49,  3.26s/it]
 86%|████████▌ | 12649/14725 [11:30:55<1:51:12,  3.21s/it]
 86%|████████▌ | 12650/14725 [11:30:58<1:51:22,  3.22s/it]
                                                          

 86%|██�
0: {'loss': 0.6885, 'grad_norm': 0.3676841477801024, 'learning_rate': 1e-05, 'epoch': 0.86}
0: ��█████▌ | 12650/14725 [11:30:58<1:51:22,  3.22s/it]
 86%|████████▌ | 12651/14725 [11:31:02<1:53:45,  3.29s/it]
 86%|████████▌ | 12652/14725 [11:31:05<1:53:13,  3.28s/it]
 86%|████████▌ | 12653/14725 [11:31:08<1:51:39,  3.23s/it]
 86%|████████▌ | 12654/14725 [11:31:11<1:51:41,  3.24s/it]
 86%|████████▌ | 12655/14725 [11:31:15<1:51:02,  3.22s/it]
 86%|████████▌ | 12656/14725 [11:31:18<1:50:54,  3.22s/it]
 86%|████████▌ | 12657/14725 [11:31:21<1:50:20,  3.20s/it]
 86%|████████▌ | 12658/14725 [11:31:24<1:50:56,  3.22s/it]
 86%|████████▌ | 12659/14725 [11:31:27<1:49:19,  3.18s/it]
 86%|████████▌ | 12660/14725 [11:31:31<1:50:37,  3.21s/it]
                                                          

 86%|████████▌ | 12660/14725 [11:31:31<1:50:37,  3.21s/it]
 86%|████████▌ | 12661/14725 [11:3
0: {'loss': 0.6739, 'grad_norm': 0.4481141312863909, 'learning_rate': 1e-05, 'epoch': 0.86}
0: 1:34<1:50:50,  3.22s/it]
 86%|████████▌ | 12662/14725 [11:31:37<1:51:54,  3.25s/it]
 86%|████████▌ | 12663/14725 [11:31:40<1:51:35,  3.25s/it]
 86%|████████▌ | 12664/14725 [11:31:43<1:50:20,  3.21s/it]
 86%|████████▌ | 12665/14725 [11:31:47<1:55:20,  3.36s/it]
 86%|████████▌ | 12666/14725 [11:31:50<1:53:38,  3.31s/it]
 86%|████████▌ | 12667/14725 [11:31:53<1:51:03,  3.24s/it]
 86%|████████▌ | 12668/14725 [11:31:57<1:50:51,  3.23s/it]
 86%|████████▌ | 12669/14725 [11:32:00<1:50:56,  3.24s/it]
 86%|████████▌ | 12670/14725 [11:32:03<1:51:55,  3.27s/it]
                                                          

 86%|████████▌ | 12670/14725 [11:32:03<1:51:55,  3.27s/it]
 86%|████████▌ | 12671/14725 [11:32:07<1:52:17,  3.28s/it]
 86%|████████▌ | 12672/14725 [11:32:10<1:54:12,  3.34s/it]
 86%|███�
0: {'loss': 0.6878, 'grad_norm': 0.39168847588190986, 'learning_rate': 1e-05, 'epoch': 0.86}
0: ��████▌ | 12673/14725 [11:32:13<1:54:53,  3.36s/it]
 86%|████████▌ | 12674/14725 [11:32:17<1:52:00,  3.28s/it]
 86%|████████▌ | 12675/14725 [11:32:20<1:54:57,  3.36s/it]
 86%|████████▌ | 12676/14725 [11:32:23<1:54:03,  3.34s/it]
 86%|████████▌ | 12677/14725 [11:32:27<1:54:49,  3.36s/it]
 86%|████████▌ | 12678/14725 [11:32:30<1:53:32,  3.33s/it]
 86%|████████▌ | 12679/14725 [11:32:33<1:51:49,  3.28s/it]
 86%|████████▌ | 12680/14725 [11:32:36<1:50:48,  3.25s/it]
                                                          

 86%|████████▌ | 12680/14725 [11:32:36<1:50:48,  3.25s/it]
 86%|████████▌ | 12681/14725 [11:32:40<1:49:41,  3.22s/it]
 86%|████████▌ | 12682/14725 [11:32:43<1:49:25,  3.21s/it]
 86%|████████▌ | 12683/14725 [11:32:46<1:49:10,  3.21s/it]
 86%|████████▌ | 12684/14725 [11:32:4
0: {'loss': 0.6746, 'grad_norm': 0.40243154254101465, 'learning_rate': 1e-05, 'epoch': 0.86}
0: 9<1:48:21,  3.19s/it]
 86%|████████▌ | 12685/14725 [11:32:52<1:49:04,  3.21s/it]
 86%|████████▌ | 12686/14725 [11:32:56<1:49:53,  3.23s/it]
 86%|████████▌ | 12687/14725 [11:32:59<1:49:41,  3.23s/it]
 86%|████████▌ | 12688/14725 [11:33:02<1:48:42,  3.20s/it]
 86%|████████▌ | 12689/14725 [11:33:05<1:47:21,  3.16s/it]
 86%|████████▌ | 12690/14725 [11:33:08<1:47:15,  3.16s/it]
                                                          

 86%|████████▌ | 12690/14725 [11:33:08<1:47:15,  3.16s/it]
 86%|████████▌ | 12691/14725 [11:33:12<1:49:26,  3.23s/it]
 86%|████████▌ | 12692/14725 [11:33:15<1:51:05,  3.28s/it]
 86%|████████▌ | 12693/14725 [11:33:18<1:50:50,  3.27s/it]
 86%|████████▌ | 12694/14725 [11:33:22<1:51:56,  3.31s/it]
 86%|████████▌ | 12695/14725 [11:33:25<1:51:11,  3.29s/it]
 86%|████�
0: {'loss': 0.6808, 'grad_norm': 0.39414162394710617, 'learning_rate': 1e-05, 'epoch': 0.86}
0: ��███▌ | 12696/14725 [11:33:28<1:53:32,  3.36s/it]
 86%|████████▌ | 12697/14725 [11:33:32<1:53:24,  3.36s/it]
 86%|████████▌ | 12698/14725 [11:33:35<1:54:31,  3.39s/it]
 86%|████████▌ | 12699/14725 [11:33:39<1:55:05,  3.41s/it]
 86%|████████▌ | 12700/14725 [11:33:42<1:52:32,  3.33s/it]
                                                          

 86%|████████▌ | 12700/14725 [11:33:42<1:52:32,  3.33s/it]
 86%|████████▋ | 12701/14725 [11:33:46<1:56:19,  3.45s/it]
 86%|████████▋ | 12702/14725 [11:33:49<1:54:11,  3.39s/it]
 86%|████████▋ | 12703/14725 [11:33:52<1:52:23,  3.34s/it]
 86%|████████▋ | 12704/14725 [11:33:55<1:52:41,  3.35s/it]
 86%|████████▋ | 12705/14725 [11:33:59<1:51:25,  3.31s/it]
 86%|████████▋ | 12706/14725 [11:34:02<1:51:01,  3.30s/it]
 86%|████████▋ | 12707/14725 [11:34:05<1
0: {'loss': 0.653, 'grad_norm': 0.3922586951955999, 'learning_rate': 1e-05, 'epoch': 0.86}
0: :51:15,  3.31s/it]
 86%|████████▋ | 12708/14725 [11:34:09<1:52:15,  3.34s/it]
 86%|████████▋ | 12709/14725 [11:34:12<1:52:47,  3.36s/it]
 86%|████████▋ | 12710/14725 [11:34:15<1:52:57,  3.36s/it]
                                                          

 86%|████████▋ | 12710/14725 [11:34:15<1:52:57,  3.36s/it]
 86%|████████▋ | 12711/14725 [11:34:18<1:49:51,  3.27s/it]
 86%|████████▋ | 12712/14725 [11:34:22<1:49:40,  3.27s/it]
 86%|████████▋ | 12713/14725 [11:34:25<1:49:41,  3.27s/it]
 86%|████████▋ | 12714/14725 [11:34:28<1:48:08,  3.23s/it]
 86%|████████▋ | 12715/14725 [11:34:31<1:48:43,  3.25s/it]
 86%|████████▋ | 12716/14725 [11:34:35<1:48:29,  3.24s/it]
 86%|████████▋ | 12717/14725 [11:34:38<1:48:23,  3.24s/it]
 86%|████████▋ | 12718/14725 [11:34:41<1:47:20,  3.21s/it]
 86%|█████�
0: {'loss': 0.6677, 'grad_norm': 0.38984963372388576, 'learning_rate': 1e-05, 'epoch': 0.86}
0: ��██▋ | 12719/14725 [11:34:44<1:47:45,  3.22s/it]
 86%|████████▋ | 12720/14725 [11:34:48<1:48:53,  3.26s/it]
                                                          

 86%|████████▋ | 12720/14725 [11:34:48<1:48:53,  3.26s/it]
 86%|████████▋ | 12721/14725 [11:34:51<1:48:40,  3.25s/it]
 86%|████████▋ | 12722/14725 [11:34:54<1:48:08,  3.24s/it]
 86%|████████▋ | 12723/14725 [11:34:57<1:45:59,  3.18s/it]
 86%|████████▋ | 12724/14725 [11:35:01<1:49:52,  3.29s/it]
 86%|████████▋ | 12725/14725 [11:35:04<1:50:27,  3.31s/it]
 86%|████████▋ | 12726/14725 [11:35:07<1:49:30,  3.29s/it]
 86%|████████▋ | 12727/14725 [11:35:11<1:52:47,  3.39s/it]
 86%|████████▋ | 12728/14725 [11:35:14<1:51:22,  3.35s/it]
 86%|████████▋ | 12729/14725 [11:35:17<1:51:12,  3.34s/it]
 86%|████████▋ | 12730/14725 [11:35:21<1:50
0: {'loss': 0.6806, 'grad_norm': 0.3857313063335332, 'learning_rate': 1e-05, 'epoch': 0.86}
0: {'loss': 0.6748, 'grad_norm': 0.4067825262223952, 'learning_rate': 1e-05, 'epoch': 0.87}
0: :06,  3.31s/it]
                                                          

 86%|████████▋ | 12730/14725 [11:35:21<1:50:06,  3.31s/it]
 86%|████████▋ | 12731/14725 [11:35:24<1:50:56,  3.34s/it]
 86%|████████▋ | 12732/14725 [11:35:28<1:52:04,  3.37s/it]
 86%|████████▋ | 12733/14725 [11:35:31<1:51:34,  3.36s/it]
 86%|████████▋ | 12734/14725 [11:35:34<1:49:13,  3.29s/it]
 86%|████████▋ | 12735/14725 [11:35:37<1:50:05,  3.32s/it]
 86%|████████▋ | 12736/14725 [11:35:41<1:49:56,  3.32s/it]
 86%|████████▋ | 12737/14725 [11:35:44<1:47:44,  3.25s/it]
 87%|████████▋ | 12738/14725 [11:35:47<1:47:28,  3.25s/it]
 87%|████████▋ | 12739/14725 [11:35:50<1:46:16,  3.21s/it]
 87%|████████▋ | 12740/14725 [11:35:53<1:45:48,  3.20s/it]
                                                          

 87%|████████▋ | 12740/
0: {'loss': 0.6697, 'grad_norm': 0.37512124002980807, 'learning_rate': 1e-05, 'epoch': 0.87}
0: 14725 [11:35:53<1:45:48,  3.20s/it]
 87%|████████▋ | 12741/14725 [11:35:57<1:45:58,  3.20s/it]
 87%|████████▋ | 12742/14725 [11:36:00<1:45:04,  3.18s/it]
 87%|████████▋ | 12743/14725 [11:36:03<1:47:58,  3.27s/it]
 87%|████████▋ | 12744/14725 [11:36:06<1:47:44,  3.26s/it]
 87%|████████▋ | 12745/14725 [11:36:10<1:47:56,  3.27s/it]
 87%|████████▋ | 12746/14725 [11:36:13<1:45:54,  3.21s/it]
 87%|████████▋ | 12747/14725 [11:36:16<1:47:08,  3.25s/it]
 87%|████████▋ | 12748/14725 [11:36:19<1:46:48,  3.24s/it]
 87%|████████▋ | 12749/14725 [11:36:22<1:45:47,  3.21s/it]
 87%|████████▋ | 12750/14725 [11:36:26<1:45:53,  3.22s/it]
                                                          

 87%|████████▋ | 12750/14725 [11:36:26<1:45:53,  3.22s/it]
 87%|████████▋ | 12751/14725 [11:36:29<1:47:12,  3.26s/it]
 87%
0: {'loss': 0.6789, 'grad_norm': 0.37904775065565055, 'learning_rate': 1e-05, 'epoch': 0.87}
0: |████████▋ | 12752/14725 [11:36:32<1:47:24,  3.27s/it]
 87%|████████▋ | 12753/14725 [11:36:36<1:48:09,  3.29s/it]
 87%|████████▋ | 12754/14725 [11:36:39<1:46:46,  3.25s/it]
 87%|████████▋ | 12755/14725 [11:36:42<1:45:22,  3.21s/it]
 87%|████████▋ | 12756/14725 [11:36:45<1:46:58,  3.26s/it]
 87%|████████▋ | 12757/14725 [11:36:49<1:48:13,  3.30s/it]
 87%|████████▋ | 12758/14725 [11:36:52<1:48:41,  3.32s/it]
 87%|████████▋ | 12759/14725 [11:36:55<1:47:46,  3.29s/it]
 87%|████████▋ | 12760/14725 [11:36:59<1:50:16,  3.37s/it]
                                                          

 87%|████████▋ | 12760/14725 [11:36:59<1:50:16,  3.37s/it]
 87%|████████▋ | 12761/14725 [11:37:02<1:49:22,  3.34s/it]
 87%|████████▋ | 12762/14725 [11:37:05<1:47:32,  3.29s/it]
 87%|████████▋ | 12763/147
0: {'loss': 0.6773, 'grad_norm': 0.4001883579315657, 'learning_rate': 1e-05, 'epoch': 0.87}
0: 25 [11:37:09<1:47:04,  3.27s/it]
 87%|████████▋ | 12764/14725 [11:37:12<1:47:34,  3.29s/it]
 87%|████████▋ | 12765/14725 [11:37:15<1:47:57,  3.30s/it]
 87%|████████▋ | 12766/14725 [11:37:19<1:48:00,  3.31s/it]
 87%|████████▋ | 12767/14725 [11:37:22<1:48:16,  3.32s/it]
 87%|████████▋ | 12768/14725 [11:37:25<1:46:56,  3.28s/it]
 87%|████████▋ | 12769/14725 [11:37:28<1:45:33,  3.24s/it]
 87%|████████▋ | 12770/14725 [11:37:31<1:44:51,  3.22s/it]
                                                          

 87%|████████▋ | 12770/14725 [11:37:31<1:44:51,  3.22s/it]
 87%|████████▋ | 12771/14725 [11:37:34<1:43:36,  3.18s/it]
 87%|████████▋ | 12772/14725 [11:37:38<1:44:56,  3.22s/it]
 87%|████████▋ | 12773/14725 [11:37:41<1:46:07,  3.26s/it]
 87%|████████▋ | 12774/14725 [11:37:45<1:48:45,  3.34s/it]
 87%|�
0: {'loss': 0.6794, 'grad_norm': 0.40272588778993135, 'learning_rate': 1e-05, 'epoch': 0.87}
0: �███████▋ | 12775/14725 [11:37:48<1:47:51,  3.32s/it]
 87%|████████▋ | 12776/14725 [11:37:51<1:47:38,  3.31s/it]
 87%|████████▋ | 12777/14725 [11:37:55<1:47:18,  3.30s/it]
 87%|████████▋ | 12778/14725 [11:37:58<1:47:20,  3.31s/it]
 87%|████████▋ | 12779/14725 [11:38:01<1:48:36,  3.35s/it]
 87%|████████▋ | 12780/14725 [11:38:04<1:47:10,  3.31s/it]
                                                          

 87%|████████▋ | 12780/14725 [11:38:04<1:47:10,  3.31s/it]
 87%|████████▋ | 12781/14725 [11:38:08<1:49:37,  3.38s/it]
 87%|████████▋ | 12782/14725 [11:38:11<1:47:29,  3.32s/it]
 87%|████████▋ | 12783/14725 [11:38:14<1:46:45,  3.30s/it]
 87%|████████▋ | 12784/14725 [11:38:18<1:46:03,  3.28s/it]
 87%|████████▋ | 12785/14725 [11:38:21<1:46:33,  3.30s/it]
 87%|████████▋ | 12786/14725 
0: {'loss': 0.6731, 'grad_norm': 0.43818931835216735, 'learning_rate': 1e-05, 'epoch': 0.87}
0: [11:38:24<1:45:20,  3.26s/it]
 87%|████████▋ | 12787/14725 [11:38:27<1:44:35,  3.24s/it]
 87%|████████▋ | 12788/14725 [11:38:31<1:44:43,  3.24s/it]
 87%|████████▋ | 12789/14725 [11:38:34<1:43:05,  3.19s/it]
 87%|████████▋ | 12790/14725 [11:38:37<1:44:31,  3.24s/it]
                                                          

 87%|████████▋ | 12790/14725 [11:38:37<1:44:31,  3.24s/it]
 87%|████████▋ | 12791/14725 [11:38:40<1:42:59,  3.20s/it]
 87%|████████▋ | 12792/14725 [11:38:43<1:43:59,  3.23s/it]
 87%|████████▋ | 12793/14725 [11:38:47<1:43:08,  3.20s/it]
 87%|████████▋ | 12794/14725 [11:38:50<1:44:06,  3.24s/it]
 87%|████████▋ | 12795/14725 [11:38:53<1:45:12,  3.27s/it]
 87%|████████▋ | 12796/14725 [11:38:56<1:43:42,  3.23s/it]
 87%|████████▋ | 12797/14725 [11:39:00<1:45:08,  3.27s/it]
 87%|█�
0: {'loss': 0.6742, 'grad_norm': 0.40416802327564716, 'learning_rate': 1e-05, 'epoch': 0.87}
0: �██████▋ | 12798/14725 [11:39:03<1:44:52,  3.27s/it]
 87%|████████▋ | 12799/14725 [11:39:06<1:45:26,  3.28s/it]
 87%|████████▋ | 12800/14725 [11:39:10<1:45:09,  3.28s/it]
                                                          

 87%|████████▋ | 12800/14725 [11:39:10<1:45:09,  3.28s/it]
 87%|████████▋ | 12801/14725 [11:39:13<1:44:54,  3.27s/it]
 87%|████████▋ | 12802/14725 [11:39:16<1:46:22,  3.32s/it]
 87%|████████▋ | 12803/14725 [11:39:20<1:46:01,  3.31s/it]
 87%|████████▋ | 12804/14725 [11:39:23<1:47:29,  3.36s/it]
 87%|████████▋ | 12805/14725 [11:39:26<1:46:55,  3.34s/it]
 87%|████████▋ | 12806/14725 [11:39:30<1:46:30,  3.33s/it]
 87%|████████▋ | 12807/14725 [11:39:33<1:46:32,  3.33s/it]
 87%|████████▋ | 12808/14725 [11:39:36<1:45:30,  3.30s/it]
 87%|████████▋ | 12809/14725 [11
0: {'loss': 0.6807, 'grad_norm': 0.4105497254562372, 'learning_rate': 1e-05, 'epoch': 0.87}
0: :39:39<1:44:36,  3.28s/it]
 87%|████████▋ | 12810/14725 [11:39:43<1:44:30,  3.27s/it]
                                                          

 87%|████████▋ | 12810/14725 [11:39:43<1:44:30,  3.27s/it]
 87%|████████▋ | 12811/14725 [11:39:46<1:45:15,  3.30s/it]
 87%|████████▋ | 12812/14725 [11:39:50<1:46:28,  3.34s/it]
 87%|████████▋ | 12813/14725 [11:39:53<1:46:31,  3.34s/it]
 87%|████████▋ | 12814/14725 [11:39:56<1:45:26,  3.31s/it]
 87%|████████▋ | 12815/14725 [11:39:59<1:45:37,  3.32s/it]
 87%|████████▋ | 12816/14725 [11:40:03<1:44:16,  3.28s/it]
 87%|████████▋ | 12817/14725 [11:40:06<1:43:28,  3.25s/it]
 87%|████████▋ | 12818/14725 [11:40:09<1:43:01,  3.24s/it]
 87%|████████▋ | 12819/14725 [11:40:12<1:44:42,  3.30s/it]
 87%|████████▋ | 12820/14725 [11:40:16<1:45:51,  3.33s/it]
             
0: {'loss': 0.6782, 'grad_norm': 0.40791470754641324, 'learning_rate': 1e-05, 'epoch': 0.87}
0: {'loss': 0.6768, 'grad_norm': 0.403740458920126, 'learning_rate': 1e-05, 'epoch': 0.87}
0:                                              

 87%|████████▋ | 12820/14725 [11:40:16<1:45:51,  3.33s/it]
 87%|████████▋ | 12821/14725 [11:40:19<1:47:58,  3.40s/it]
 87%|████████▋ | 12822/14725 [11:40:23<1:45:05,  3.31s/it]
 87%|████████▋ | 12823/14725 [11:40:26<1:43:06,  3.25s/it]
 87%|████████▋ | 12824/14725 [11:40:29<1:42:04,  3.22s/it]
 87%|████████▋ | 12825/14725 [11:40:32<1:42:48,  3.25s/it]
 87%|████████▋ | 12826/14725 [11:40:35<1:42:48,  3.25s/it]
 87%|████████▋ | 12827/14725 [11:40:39<1:44:28,  3.30s/it]
 87%|████████▋ | 12828/14725 [11:40:42<1:43:31,  3.27s/it]
 87%|████████▋ | 12829/14725 [11:40:46<1:45:37,  3.34s/it]
 87%|████████▋ | 12830/14725 [11:40:49<1:44:04,  3.30s/it]
                                                          

 87%|████████▋ | 12830/14725 [11:40:49<1:44:04,  3.3
0: {'loss': 0.6633, 'grad_norm': 0.44483817343373755, 'learning_rate': 1e-05, 'epoch': 0.87}
0: 0s/it]
 87%|████████▋ | 12831/14725 [11:40:52<1:42:34,  3.25s/it]
 87%|████████▋ | 12832/14725 [11:40:55<1:41:36,  3.22s/it]
 87%|████████▋ | 12833/14725 [11:40:58<1:44:00,  3.30s/it]
 87%|████████▋ | 12834/14725 [11:41:02<1:43:48,  3.29s/it]
 87%|████████▋ | 12835/14725 [11:41:05<1:43:02,  3.27s/it]
 87%|████████▋ | 12836/14725 [11:41:08<1:43:44,  3.30s/it]
 87%|████████▋ | 12837/14725 [11:41:12<1:43:27,  3.29s/it]
 87%|████████▋ | 12838/14725 [11:41:15<1:42:09,  3.25s/it]
 87%|████████▋ | 12839/14725 [11:41:18<1:41:10,  3.22s/it]
 87%|████████▋ | 12840/14725 [11:41:21<1:40:37,  3.20s/it]
                                                          

 87%|████████▋ | 12840/14725 [11:41:21<1:40:37,  3.20s/it]
 87%|████████▋ | 12841/14725 [11:41:24<1:41:27,  3.23s/it]
 87%|████████▋ 
0: {'loss': 0.6671, 'grad_norm': 0.4140444951825981, 'learning_rate': 1e-05, 'epoch': 0.87}
0: | 12842/14725 [11:41:28<1:42:09,  3.26s/it]
 87%|████████▋ | 12843/14725 [11:41:31<1:42:22,  3.26s/it]
 87%|████████▋ | 12844/14725 [11:41:35<1:45:19,  3.36s/it]
 87%|████████▋ | 12845/14725 [11:41:38<1:44:53,  3.35s/it]
 87%|████████▋ | 12846/14725 [11:41:41<1:42:49,  3.28s/it]
 87%|████████▋ | 12847/14725 [11:41:44<1:42:01,  3.26s/it]
 87%|████████▋ | 12848/14725 [11:41:47<1:40:43,  3.22s/it]
 87%|████████▋ | 12849/14725 [11:41:51<1:41:50,  3.26s/it]
 87%|████████▋ | 12850/14725 [11:41:54<1:41:06,  3.24s/it]
                                                          

 87%|████████▋ | 12850/14725 [11:41:54<1:41:06,  3.24s/it]
 87%|████████▋ | 12851/14725 [11:41:57<1:40:27,  3.22s/it]
 87%|████████▋ | 12852/14725 [11:42:00<1:40:08,  3.21s/it]
 87%|████████▋ | 12853/14725 [11:42:04<1:41:37,  3.26s/
0: {'loss': 0.6719, 'grad_norm': 0.41917016056941414, 'learning_rate': 1e-05, 'epoch': 0.87}
0: it]
 87%|████████▋ | 12854/14725 [11:42:07<1:42:34,  3.29s/it]
 87%|████████▋ | 12855/14725 [11:42:10<1:43:55,  3.33s/it]
 87%|████████▋ | 12856/14725 [11:42:14<1:43:25,  3.32s/it]
 87%|████████▋ | 12857/14725 [11:42:17<1:42:31,  3.29s/it]
 87%|████████▋ | 12858/14725 [11:42:20<1:41:20,  3.26s/it]
 87%|████████▋ | 12859/14725 [11:42:23<1:41:32,  3.26s/it]
 87%|████████▋ | 12860/14725 [11:42:27<1:41:05,  3.25s/it]
                                                          

 87%|████████▋ | 12860/14725 [11:42:27<1:41:05,  3.25s/it]
 87%|████████▋ | 12861/14725 [11:42:30<1:39:17,  3.20s/it]
 87%|████████▋ | 12862/14725 [11:42:33<1:37:43,  3.15s/it]
 87%|████████▋ | 12863/14725 [11:42:36<1:37:43,  3.15s/it]
 87%|████████▋ | 12864/14725 [11:42:39<1:40:39,  3.25s/it]
 87%|████████▋ | 1
0: {'loss': 0.6587, 'grad_norm': 0.3820596266545182, 'learning_rate': 1e-05, 'epoch': 0.87}
0: 2865/14725 [11:42:43<1:40:22,  3.24s/it]
 87%|████████▋ | 12866/14725 [11:42:46<1:39:16,  3.20s/it]
 87%|████████▋ | 12867/14725 [11:42:49<1:41:22,  3.27s/it]
 87%|████████▋ | 12868/14725 [11:42:52<1:40:55,  3.26s/it]
 87%|████████▋ | 12869/14725 [11:42:55<1:39:24,  3.21s/it]
 87%|████████▋ | 12870/14725 [11:42:59<1:39:15,  3.21s/it]
                                                          

 87%|████████▋ | 12870/14725 [11:42:59<1:39:15,  3.21s/it]
 87%|████████▋ | 12871/14725 [11:43:02<1:39:24,  3.22s/it]
 87%|████████▋ | 12872/14725 [11:43:05<1:41:09,  3.28s/it]
 87%|████████▋ | 12873/14725 [11:43:09<1:41:38,  3.29s/it]
 87%|████████▋ | 12874/14725 [11:43:12<1:42:58,  3.34s/it]
 87%|████████▋ | 12875/14725 [11:43:15<1:40:35,  3.26s/it]
 87%|████████▋ | 12876/14725 [11:43:18<1:40:04,  3.25s/it]
0: {'loss': 0.6797, 'grad_norm': 0.36852232319170986, 'learning_rate': 1e-05, 'epoch': 0.87}
0: 
 87%|████████▋ | 12877/14725 [11:43:22<1:42:13,  3.32s/it]
 87%|████████▋ | 12878/14725 [11:43:25<1:40:10,  3.25s/it]
 87%|████████▋ | 12879/14725 [11:43:28<1:41:14,  3.29s/it]
 87%|████████▋ | 12880/14725 [11:43:32<1:40:37,  3.27s/it]
                                                          

 87%|████████▋ | 12880/14725 [11:43:32<1:40:37,  3.27s/it]
 87%|████████▋ | 12881/14725 [11:43:35<1:39:17,  3.23s/it]
 87%|████████▋ | 12882/14725 [11:43:38<1:41:24,  3.30s/it]
 87%|████████▋ | 12883/14725 [11:43:41<1:39:55,  3.26s/it]
 87%|████████▋ | 12884/14725 [11:43:45<1:40:19,  3.27s/it]
 88%|████████▊ | 12885/14725 [11:43:48<1:41:01,  3.29s/it]
 88%|████████▊ | 12886/14725 [11:43:51<1:39:44,  3.25s/it]
 88%|████████▊ | 12887/14725 [11:43:54<1:40:24,  3.28s/it]
 88%|████████▊ | 1288
0: {'loss': 0.6729, 'grad_norm': 0.42112522394430485, 'learning_rate': 1e-05, 'epoch': 0.88}
0: 8/14725 [11:43:58<1:41:26,  3.31s/it]
 88%|████████▊ | 12889/14725 [11:44:01<1:41:28,  3.32s/it]
 88%|████████▊ | 12890/14725 [11:44:04<1:39:43,  3.26s/it]
                                                          

 88%|████████▊ | 12890/14725 [11:44:04<1:39:43,  3.26s/it]
 88%|████████▊ | 12891/14725 [11:44:07<1:39:10,  3.24s/it]
 88%|████████▊ | 12892/14725 [11:44:11<1:38:44,  3.23s/it]
 88%|████████▊ | 12893/14725 [11:44:14<1:38:02,  3.21s/it]
 88%|████████▊ | 12894/14725 [11:44:17<1:38:19,  3.22s/it]
 88%|████████▊ | 12895/14725 [11:44:20<1:37:06,  3.18s/it]
 88%|████████▊ | 12896/14725 [11:44:23<1:37:53,  3.21s/it]
 88%|████████▊ | 12897/14725 [11:44:27<1:40:04,  3.28s/it]
 88%|████████▊ | 12898/14725 [11:44:30<1:39:45,  3.28s/it]
 88%|████████▊ | 12899/14725 [11:44:33<1:38:17,  3.23s/it]
 8
0: {'loss': 0.6635, 'grad_norm': 0.38462846609785445, 'learning_rate': 1e-05, 'epoch': 0.88}
0: 8%|████████▊ | 12900/14725 [11:44:37<1:39:02,  3.26s/it]
                                                          

 88%|████████▊ | 12900/14725 [11:44:37<1:39:02,  3.26s/it]
 88%|████████▊ | 12901/14725 [11:44:40<1:40:00,  3.29s/it]
 88%|████████▊ | 12902/14725 [11:44:43<1:41:17,  3.33s/it]
 88%|████████▊ | 12903/14725 [11:44:47<1:41:12,  3.33s/it]
 88%|████████▊ | 12904/14725 [11:44:50<1:42:05,  3.36s/it]
 88%|████████▊ | 12905/14725 [11:44:53<1:40:12,  3.30s/it]
 88%|████████▊ | 12906/14725 [11:44:56<1:38:00,  3.23s/it]
 88%|████████▊ | 12907/14725 [11:45:00<1:36:57,  3.20s/it]
 88%|████████▊ | 12908/14725 [11:45:03<1:36:23,  3.18s/it]
 88%|████████▊ | 12909/14725 [11:45:06<1:37:08,  3.21s/it]
 88%|████████▊ | 12910/14725 [11:45:09<1:38:23,  3.25s/it]
                                          
0: {'loss': 0.6831, 'grad_norm': 0.42094979294699236, 'learning_rate': 1e-05, 'epoch': 0.88}
0: {'loss': 0.6766, 'grad_norm': 0.41179997884850505, 'learning_rate': 1e-05, 'epoch': 0.88}
0:                 

 88%|████████▊ | 12910/14725 [11:45:09<1:38:23,  3.25s/it]
 88%|████████▊ | 12911/14725 [11:45:13<1:38:04,  3.24s/it]
 88%|████████▊ | 12912/14725 [11:45:16<1:37:12,  3.22s/it]
 88%|████████▊ | 12913/14725 [11:45:19<1:36:02,  3.18s/it]
 88%|████████▊ | 12914/14725 [11:45:22<1:38:50,  3.27s/it]
 88%|████████▊ | 12915/14725 [11:45:25<1:36:49,  3.21s/it]
 88%|████████▊ | 12916/14725 [11:45:29<1:37:20,  3.23s/it]
 88%|████████▊ | 12917/14725 [11:45:32<1:39:47,  3.31s/it]
 88%|████████▊ | 12918/14725 [11:45:35<1:38:07,  3.26s/it]
 88%|████████▊ | 12919/14725 [11:45:39<1:38:26,  3.27s/it]
 88%|████████▊ | 12920/14725 [11:45:42<1:38:21,  3.27s/it]
                                                          

 88%|████████▊ | 12920/14725 [11:45:42<1:38:21,  3.27s/it]
 88%|█████�
0: {'loss': 0.6717, 'grad_norm': 0.40159768661140416, 'learning_rate': 1e-05, 'epoch': 0.88}
0: �██▊ | 12921/14725 [11:45:45<1:38:53,  3.29s/it]
 88%|████████▊ | 12922/14725 [11:45:48<1:39:19,  3.31s/it]
 88%|████████▊ | 12923/14725 [11:45:52<1:37:46,  3.26s/it]
 88%|████████▊ | 12924/14725 [11:45:55<1:38:23,  3.28s/it]
 88%|████████▊ | 12925/14725 [11:45:58<1:38:31,  3.28s/it]
 88%|████████▊ | 12926/14725 [11:46:02<1:38:24,  3.28s/it]
 88%|████████▊ | 12927/14725 [11:46:05<1:36:24,  3.22s/it]
 88%|████████▊ | 12928/14725 [11:46:08<1:37:41,  3.26s/it]
 88%|████████▊ | 12929/14725 [11:46:11<1:38:36,  3.29s/it]
 88%|████████▊ | 12930/14725 [11:46:15<1:37:59,  3.28s/it]
                                                          

 88%|████████▊ | 12930/14725 [11:46:15<1:37:59,  3.28s/it]
 88%|████████▊ | 12931/14725 [11:46:18<1:39:26,  3.33s/it]
 88%|████████▊ | 12932/14725 [11:46:21<1:39:
0: {'loss': 0.6764, 'grad_norm': 0.42895651816765, 'learning_rate': 1e-05, 'epoch': 0.88}
0: 50,  3.34s/it]
 88%|████████▊ | 12933/14725 [11:46:25<1:40:39,  3.37s/it]
 88%|████████▊ | 12934/14725 [11:46:28<1:40:34,  3.37s/it]
 88%|████████▊ | 12935/14725 [11:46:31<1:38:21,  3.30s/it]
 88%|████████▊ | 12936/14725 [11:46:35<1:38:34,  3.31s/it]
 88%|████████▊ | 12937/14725 [11:46:38<1:38:47,  3.31s/it]
 88%|████████▊ | 12938/14725 [11:46:41<1:39:20,  3.34s/it]
 88%|████████▊ | 12939/14725 [11:46:45<1:39:08,  3.33s/it]
 88%|████████▊ | 12940/14725 [11:46:48<1:37:30,  3.28s/it]
                                                          

 88%|████████▊ | 12940/14725 [11:46:48<1:37:30,  3.28s/it]
 88%|████████▊ | 12941/14725 [11:46:51<1:37:58,  3.30s/it]
 88%|████████▊ | 12942/14725 [11:46:55<1:39:32,  3.35s/it]
 88%|████████▊ | 12943/14725 [11:46:58<1:37:51,  3.29s/it]
 88%|██████�
0: {'loss': 0.6757, 'grad_norm': 0.43308435612597446, 'learning_rate': 1e-05, 'epoch': 0.88}
0: �█▊ | 12944/14725 [11:47:01<1:37:35,  3.29s/it]
 88%|████████▊ | 12945/14725 [11:47:04<1:37:39,  3.29s/it]
 88%|████████▊ | 12946/14725 [11:47:08<1:36:34,  3.26s/it]
 88%|████████▊ | 12947/14725 [11:47:11<1:37:30,  3.29s/it]
 88%|████████▊ | 12948/14725 [11:47:14<1:37:14,  3.28s/it]
 88%|████████▊ | 12949/14725 [11:47:18<1:39:18,  3.35s/it]
 88%|████████▊ | 12950/14725 [11:47:21<1:38:43,  3.34s/it]
                                                          

 88%|████████▊ | 12950/14725 [11:47:21<1:38:43,  3.34s/it]
 88%|████████▊ | 12951/14725 [11:47:24<1:37:32,  3.30s/it]
 88%|████████▊ | 12952/14725 [11:47:28<1:37:13,  3.29s/it]
 88%|████████▊ | 12953/14725 [11:47:31<1:37:18,  3.29s/it]
 88%|████████▊ | 12954/14725 [11:47:34<1:38:23,  3.33s/it]
 88%|████████▊ | 12955/14725 [11:47:38<1:38:10,
0: {'loss': 0.6711, 'grad_norm': 0.41621877816331826, 'learning_rate': 1e-05, 'epoch': 0.88}
0:   3.33s/it]
 88%|████████▊ | 12956/14725 [11:47:41<1:37:16,  3.30s/it]
 88%|████████▊ | 12957/14725 [11:47:44<1:37:56,  3.32s/it]
 88%|████████▊ | 12958/14725 [11:47:47<1:37:17,  3.30s/it]
 88%|████████▊ | 12959/14725 [11:47:50<1:34:52,  3.22s/it]
 88%|████████▊ | 12960/14725 [11:47:54<1:35:58,  3.26s/it]
                                                          

 88%|████████▊ | 12960/14725 [11:47:54<1:35:58,  3.26s/it]
 88%|████████▊ | 12961/14725 [11:47:57<1:36:16,  3.27s/it]
 88%|████████▊ | 12962/14725 [11:48:00<1:37:00,  3.30s/it]
 88%|████████▊ | 12963/14725 [11:48:04<1:34:41,  3.22s/it]
 88%|████████▊ | 12964/14725 [11:48:07<1:33:55,  3.20s/it]
 88%|████████▊ | 12965/14725 [11:48:10<1:34:03,  3.21s/it]
 88%|████████▊ | 12966/14725 [11:48:13<1:34:42,  3.23s/it]
 88%|███████�
0: {'loss': 0.6743, 'grad_norm': 0.4012983436133381, 'learning_rate': 1e-05, 'epoch': 0.88}
0: �▊ | 12967/14725 [11:48:16<1:32:59,  3.17s/it]
 88%|████████▊ | 12968/14725 [11:48:19<1:32:21,  3.15s/it]
 88%|████████▊ | 12969/14725 [11:48:22<1:31:30,  3.13s/it]
 88%|████████▊ | 12970/14725 [11:48:25<1:31:03,  3.11s/it]
                                                          

 88%|████████▊ | 12970/14725 [11:48:25<1:31:03,  3.11s/it]
 88%|████████▊ | 12971/14725 [11:48:29<1:32:24,  3.16s/it]
 88%|████████▊ | 12972/14725 [11:48:32<1:34:47,  3.24s/it]
 88%|████████▊ | 12973/14725 [11:48:35<1:34:52,  3.25s/it]
 88%|████████▊ | 12974/14725 [11:48:39<1:35:27,  3.27s/it]
 88%|████████▊ | 12975/14725 [11:48:42<1:34:14,  3.23s/it]
 88%|████████▊ | 12976/14725 [11:48:45<1:33:54,  3.22s/it]
 88%|████████▊ | 12977/14725 [11:48:48<1:33:25,  3.21s/it]
 88%|████████▊ | 12978/14725 [11:48:52<1:33:43,  3
0: {'loss': 0.6641, 'grad_norm': 0.39223580557557525, 'learning_rate': 1e-05, 'epoch': 0.88}
0: .22s/it]
 88%|████████▊ | 12979/14725 [11:48:55<1:34:57,  3.26s/it]
 88%|████████▊ | 12980/14725 [11:48:58<1:34:02,  3.23s/it]
                                                          

 88%|████████▊ | 12980/14725 [11:48:58<1:34:02,  3.23s/it]
 88%|████████▊ | 12981/14725 [11:49:02<1:36:04,  3.31s/it]
 88%|████████▊ | 12982/14725 [11:49:05<1:33:49,  3.23s/it]
 88%|████████▊ | 12983/14725 [11:49:08<1:34:09,  3.24s/it]
 88%|████████▊ | 12984/14725 [11:49:11<1:33:15,  3.21s/it]
 88%|████████▊ | 12985/14725 [11:49:14<1:35:10,  3.28s/it]
 88%|████████▊ | 12986/14725 [11:49:18<1:35:11,  3.28s/it]
 88%|████████▊ | 12987/14725 [11:49:21<1:34:03,  3.25s/it]
 88%|████████▊ | 12988/14725 [11:49:24<1:34:00,  3.25s/it]
 88%|████████▊ | 12989/14725 [11:49:27<1:33:26,  3.23s/it]
 88%|████████�
0: {'loss': 0.6648, 'grad_norm': 0.3849305873663286, 'learning_rate': 1e-05, 'epoch': 0.88}
0: {'loss': 0.6883, 'grad_norm': 0.4194896681577512, 'learning_rate': 1e-05, 'epoch': 0.88}
0: � | 12990/14725 [11:49:31<1:33:18,  3.23s/it]
                                                          

 88%|████████▊ | 12990/14725 [11:49:31<1:33:18,  3.23s/it]
 88%|████████▊ | 12991/14725 [11:49:34<1:33:30,  3.24s/it]
 88%|████████▊ | 12992/14725 [11:49:37<1:34:18,  3.27s/it]
 88%|████████▊ | 12993/14725 [11:49:40<1:35:04,  3.29s/it]
 88%|████████▊ | 12994/14725 [11:49:44<1:35:18,  3.30s/it]
 88%|████████▊ | 12995/14725 [11:49:47<1:34:30,  3.28s/it]
 88%|████████▊ | 12996/14725 [11:49:50<1:34:23,  3.28s/it]
 88%|████████▊ | 12997/14725 [11:49:54<1:34:38,  3.29s/it]
 88%|████████▊ | 12998/14725 [11:49:57<1:34:51,  3.30s/it]
 88%|████████▊ | 12999/14725 [11:50:00<1:34:04,  3.27s/it]
 88%|████████▊ | 13000/14725 [11:50:03<1:32:54,  3.23s/it]
                                                          

 88%|██
0: {'loss': 0.6841, 'grad_norm': 0.3998052601107101, 'learning_rate': 1e-05, 'epoch': 0.88}
0: ██████▊ | 13000/14725 [11:50:03<1:32:54,  3.23s/it]
 88%|████████▊ | 13001/14725 [11:50:06<1:31:56,  3.20s/it]
 88%|████████▊ | 13002/14725 [11:50:10<1:32:09,  3.21s/it]
 88%|████████▊ | 13003/14725 [11:50:13<1:31:46,  3.20s/it]
 88%|████████▊ | 13004/14725 [11:50:16<1:31:14,  3.18s/it]
 88%|████████▊ | 13005/14725 [11:50:19<1:30:04,  3.14s/it]
 88%|████████▊ | 13006/14725 [11:50:22<1:30:06,  3.15s/it]
 88%|████████▊ | 13007/14725 [11:50:26<1:32:00,  3.21s/it]
 88%|████████▊ | 13008/14725 [11:50:29<1:32:37,  3.24s/it]
 88%|████████▊ | 13009/14725 [11:50:32<1:31:41,  3.21s/it]
 88%|████████▊ | 13010/14725 [11:50:36<1:35:44,  3.35s/it]
                                                          

 88%|████████▊ | 13010/14725 [11:50:36<1:35:44,  3.35s/it]
 88%|████████▊ | 13011/14725 [11:
0: {'loss': 0.6644, 'grad_norm': 0.38876979206977946, 'learning_rate': 1e-05, 'epoch': 0.88}
0: 50:39<1:35:14,  3.33s/it]
 88%|████████▊ | 13012/14725 [11:50:42<1:34:54,  3.32s/it]
 88%|████████▊ | 13013/14725 [11:50:45<1:34:03,  3.30s/it]
 88%|████████▊ | 13014/14725 [11:50:49<1:33:16,  3.27s/it]
 88%|████████▊ | 13015/14725 [11:50:52<1:32:15,  3.24s/it]
 88%|████████▊ | 13016/14725 [11:50:55<1:31:37,  3.22s/it]
 88%|████████▊ | 13017/14725 [11:50:58<1:30:33,  3.18s/it]
 88%|████████▊ | 13018/14725 [11:51:01<1:29:56,  3.16s/it]
 88%|████████▊ | 13019/14725 [11:51:05<1:30:50,  3.19s/it]
 88%|████████▊ | 13020/14725 [11:51:08<1:31:13,  3.21s/it]
                                                          

 88%|████████▊ | 13020/14725 [11:51:08<1:31:13,  3.21s/it]
 88%|████████▊ | 13021/14725 [11:51:11<1:30:55,  3.20s/it]
 88%|████████▊ | 13022/14725 [11:51:14<1:30:20,  3.18s/it]
 88%|███
0: {'loss': 0.6697, 'grad_norm': 0.4113770936671367, 'learning_rate': 1e-05, 'epoch': 0.88}
0: █████▊ | 13023/14725 [11:51:17<1:30:11,  3.18s/it]
 88%|████████▊ | 13024/14725 [11:51:21<1:32:00,  3.25s/it]
 88%|████████▊ | 13025/14725 [11:51:24<1:31:15,  3.22s/it]
 88%|████████▊ | 13026/14725 [11:51:27<1:29:30,  3.16s/it]
 88%|████████▊ | 13027/14725 [11:51:30<1:30:10,  3.19s/it]
 88%|████████▊ | 13028/14725 [11:51:34<1:32:12,  3.26s/it]
 88%|████████▊ | 13029/14725 [11:51:37<1:32:20,  3.27s/it]
 88%|████████▊ | 13030/14725 [11:51:40<1:30:59,  3.22s/it]
                                                          

 88%|████████▊ | 13030/14725 [11:51:40<1:30:59,  3.22s/it]
 88%|████████▊ | 13031/14725 [11:51:43<1:31:16,  3.23s/it]
 89%|████████▊ | 13032/14725 [11:51:47<1:34:24,  3.35s/it]
 89%|████████▊ | 13033/14725 [11:51:50<1:33:39,  3.32s/it]
 89%|████████▊ | 13034/14725 [11:51:
0: {'loss': 0.6536, 'grad_norm': 0.3884451934558712, 'learning_rate': 1e-05, 'epoch': 0.89}
0: 53<1:33:54,  3.33s/it]
 89%|████████▊ | 13035/14725 [11:51:57<1:33:20,  3.31s/it]
 89%|████████▊ | 13036/14725 [11:52:00<1:32:46,  3.30s/it]
 89%|████████▊ | 13037/14725 [11:52:04<1:35:48,  3.41s/it]
 89%|████████▊ | 13038/14725 [11:52:07<1:33:47,  3.34s/it]
 89%|████████▊ | 13039/14725 [11:52:10<1:32:26,  3.29s/it]
 89%|████████▊ | 13040/14725 [11:52:13<1:32:31,  3.29s/it]
                                                          

 89%|████████▊ | 13040/14725 [11:52:13<1:32:31,  3.29s/it]
 89%|████████▊ | 13041/14725 [11:52:16<1:31:51,  3.27s/it]
 89%|████████▊ | 13042/14725 [11:52:20<1:32:43,  3.31s/it]
 89%|████████▊ | 13043/14725 [11:52:23<1:34:20,  3.37s/it]
 89%|████████▊ | 13044/14725 [11:52:27<1:33:17,  3.33s/it]
 89%|████████▊ | 13045/14725 [11:52:30<1:32:07,  3.29s/it]
 89%|████
0: {'loss': 0.6736, 'grad_norm': 0.3985164659589319, 'learning_rate': 1e-05, 'epoch': 0.89}
0: ████▊ | 13046/14725 [11:52:33<1:30:27,  3.23s/it]
 89%|████████▊ | 13047/14725 [11:52:36<1:30:47,  3.25s/it]
 89%|████████▊ | 13048/14725 [11:52:39<1:30:11,  3.23s/it]
 89%|████████▊ | 13049/14725 [11:52:43<1:31:36,  3.28s/it]
 89%|████████▊ | 13050/14725 [11:52:46<1:32:04,  3.30s/it]
                                                          

 89%|████████▊ | 13050/14725 [11:52:46<1:32:04,  3.30s/it]
 89%|████████▊ | 13051/14725 [11:52:49<1:32:13,  3.31s/it]
 89%|████████▊ | 13052/14725 [11:52:53<1:30:58,  3.26s/it]
 89%|████████▊ | 13053/14725 [11:52:56<1:29:33,  3.21s/it]
 89%|████████▊ | 13054/14725 [11:52:59<1:29:32,  3.21s/it]
 89%|████████▊ | 13055/14725 [11:53:02<1:29:34,  3.22s/it]
 89%|████████▊ | 13056/14725 [11:53:05<1:30:33,  3.26s/it]
 89%|████████▊ | 13057/14725 [11:53:09<
0: {'loss': 0.663, 'grad_norm': 0.4047797194243259, 'learning_rate': 1e-05, 'epoch': 0.89}
0: 1:30:59,  3.27s/it]
 89%|████████▊ | 13058/14725 [11:53:12<1:29:41,  3.23s/it]
 89%|████████▊ | 13059/14725 [11:53:15<1:30:29,  3.26s/it]
 89%|████████▊ | 13060/14725 [11:53:18<1:29:07,  3.21s/it]
                                                          

 89%|████████▊ | 13060/14725 [11:53:18<1:29:07,  3.21s/it]
 89%|████████▊ | 13061/14725 [11:53:22<1:29:52,  3.24s/it]
 89%|████████▊ | 13062/14725 [11:53:25<1:30:41,  3.27s/it]
 89%|████████▊ | 13063/14725 [11:53:28<1:31:52,  3.32s/it]
 89%|████████▊ | 13064/14725 [11:53:32<1:32:23,  3.34s/it]
 89%|████████▊ | 13065/14725 [11:53:35<1:31:49,  3.32s/it]
 89%|████████▊ | 13066/14725 [11:53:38<1:30:17,  3.27s/it]
 89%|████████▊ | 13067/14725 [11:53:41<1:29:55,  3.25s/it]
 89%|████████▊ | 13068/14725 [11:53:45<1:29:44,  3.25s/it]
 89%|█████
0: {'loss': 0.6788, 'grad_norm': 0.4166345811080456, 'learning_rate': 1e-05, 'epoch': 0.89}
0: ███▉ | 13069/14725 [11:53:48<1:31:42,  3.32s/it]
 89%|████████▉ | 13070/14725 [11:53:51<1:30:26,  3.28s/it]
                                                          

 89%|████████▉ | 13070/14725 [11:53:51<1:30:26,  3.28s/it]
 89%|████████▉ | 13071/14725 [11:53:55<1:30:04,  3.27s/it]
 89%|████████▉ | 13072/14725 [11:53:58<1:28:27,  3.21s/it]
 89%|████████▉ | 13073/14725 [11:54:01<1:28:02,  3.20s/it]
 89%|████████▉ | 13074/14725 [11:54:04<1:29:04,  3.24s/it]
 89%|████████▉ | 13075/14725 [11:54:07<1:27:53,  3.20s/it]
 89%|████████▉ | 13076/14725 [11:54:10<1:27:24,  3.18s/it]
 89%|████████▉ | 13077/14725 [11:54:14<1:27:19,  3.18s/it]
 89%|████████▉ | 13078/14725 [11:54:17<1:25:59,  3.13s/it]
 89%|████████▉ | 13079/14725 [11:54:20<1:26:41,  3.16s/it]
 89%|████████▉ | 13080/14725 [11:54:23<1:2
0: {'loss': 0.673, 'grad_norm': 0.3882513463432693, 'learning_rate': 1e-05, 'epoch': 0.89}
0: {'loss': 0.6552, 'grad_norm': 0.3867940793553335, 'learning_rate': 1e-05, 'epoch': 0.89}
0: 7:52,  3.21s/it]
                                                          

 89%|████████▉ | 13080/14725 [11:54:23<1:27:52,  3.21s/it]
 89%|████████▉ | 13081/14725 [11:54:26<1:27:32,  3.19s/it]
 89%|████████▉ | 13082/14725 [11:54:30<1:27:51,  3.21s/it]
 89%|████████▉ | 13083/14725 [11:54:33<1:28:10,  3.22s/it]
 89%|████████▉ | 13084/14725 [11:54:36<1:28:04,  3.22s/it]
 89%|████████▉ | 13085/14725 [11:54:39<1:27:03,  3.19s/it]
 89%|████████▉ | 13086/14725 [11:54:42<1:27:39,  3.21s/it]
 89%|████████▉ | 13087/14725 [11:54:45<1:25:57,  3.15s/it]
 89%|████████▉ | 13088/14725 [11:54:49<1:26:16,  3.16s/it]
 89%|████████▉ | 13089/14725 [11:54:52<1:28:37,  3.25s/it]
 89%|████████▉ | 13090/14725 [11:54:55<1:27:56,  3.23s/it]
                                                          

 89%|████████▉ | 13090
0: {'loss': 0.669, 'grad_norm': 0.3978619694070705, 'learning_rate': 1e-05, 'epoch': 0.89}
0: /14725 [11:54:55<1:27:56,  3.23s/it]
 89%|████████▉ | 13091/14725 [11:54:58<1:27:27,  3.21s/it]
 89%|████████▉ | 13092/14725 [11:55:01<1:26:13,  3.17s/it]
 89%|████████▉ | 13093/14725 [11:55:05<1:25:47,  3.15s/it]
 89%|████████▉ | 13094/14725 [11:55:08<1:24:58,  3.13s/it]
 89%|████████▉ | 13095/14725 [11:55:11<1:27:04,  3.21s/it]
 89%|████████▉ | 13096/14725 [11:55:14<1:27:20,  3.22s/it]
 89%|████████▉ | 13097/14725 [11:55:18<1:28:17,  3.25s/it]
 89%|████████▉ | 13098/14725 [11:55:21<1:26:37,  3.19s/it]
 89%|████████▉ | 13099/14725 [11:55:24<1:27:34,  3.23s/it]
 89%|████████▉ | 13100/14725 [11:55:27<1:27:42,  3.24s/it]
                                                          

 89%|████████▉ | 13100/14725 [11:55:27<1:27:42,  3.24s/it]
 89%|████████▉ | 13101/14725 [11:55:31<1:28:41,  3.28s/it]
 89
0: {'loss': 0.6643, 'grad_norm': 0.39049347582565347, 'learning_rate': 1e-05, 'epoch': 0.89}
0: %|████████▉ | 13102/14725 [11:55:34<1:31:15,  3.37s/it]
 89%|████████▉ | 13103/14725 [11:55:38<1:30:48,  3.36s/it]
 89%|████████▉ | 13104/14725 [11:55:41<1:29:40,  3.32s/it]
 89%|████████▉ | 13105/14725 [11:55:44<1:29:54,  3.33s/it]
 89%|████████▉ | 13106/14725 [11:55:47<1:30:03,  3.34s/it]
 89%|████████▉ | 13107/14725 [11:55:51<1:28:45,  3.29s/it]
 89%|████████▉ | 13108/14725 [11:55:54<1:28:20,  3.28s/it]
 89%|████████▉ | 13109/14725 [11:55:57<1:27:30,  3.25s/it]
 89%|████████▉ | 13110/14725 [11:56:00<1:28:17,  3.28s/it]
                                                          

 89%|████████▉ | 13110/14725 [11:56:00<1:28:17,  3.28s/it]
 89%|████████▉ | 13111/14725 [11:56:04<1:27:10,  3.24s/it]
 89%|████████▉ | 13112/14725 [11:56:07<1:27:16,  3.25s/it]
 89%|████████▉ | 13113/14
0: {'loss': 0.6671, 'grad_norm': 0.40031612009894507, 'learning_rate': 1e-05, 'epoch': 0.89}
0: 725 [11:56:10<1:27:12,  3.25s/it]
 89%|████████▉ | 13114/14725 [11:56:13<1:27:39,  3.26s/it]
 89%|████████▉ | 13115/14725 [11:56:16<1:26:01,  3.21s/it]
 89%|████████▉ | 13116/14725 [11:56:20<1:27:07,  3.25s/it]
 89%|████████▉ | 13117/14725 [11:56:23<1:29:00,  3.32s/it]
 89%|████████▉ | 13118/14725 [11:56:26<1:27:22,  3.26s/it]
 89%|████████▉ | 13119/14725 [11:56:30<1:27:25,  3.27s/it]
 89%|████████▉ | 13120/14725 [11:56:33<1:28:20,  3.30s/it]
                                                          

 89%|████████▉ | 13120/14725 [11:56:33<1:28:20,  3.30s/it]
 89%|████████▉ | 13121/14725 [11:56:36<1:26:50,  3.25s/it]
 89%|████████▉ | 13122/14725 [11:56:39<1:26:25,  3.23s/it]
 89%|████████▉ | 13123/14725 [11:56:43<1:25:40,  3.21s/it]
 89%|████████▉ | 13124/14725 [11:56:46<1:25:31,  3.21s/it]
 89%|�
0: {'loss': 0.6662, 'grad_norm': 0.39753055280478006, 'learning_rate': 1e-05, 'epoch': 0.89}
0: ��███████▉ | 13125/14725 [11:56:49<1:25:15,  3.20s/it]
 89%|████████▉ | 13126/14725 [11:56:52<1:27:39,  3.29s/it]
 89%|████████▉ | 13127/14725 [11:56:56<1:27:21,  3.28s/it]
 89%|████████▉ | 13128/14725 [11:56:59<1:27:14,  3.28s/it]
 89%|████████▉ | 13129/14725 [11:57:02<1:25:59,  3.23s/it]
 89%|████████▉ | 13130/14725 [11:57:06<1:27:16,  3.28s/it]
                                                          

 89%|████████▉ | 13130/14725 [11:57:06<1:27:16,  3.28s/it]
 89%|████████▉ | 13131/14725 [11:57:09<1:27:34,  3.30s/it]
 89%|████████▉ | 13132/14725 [11:57:12<1:26:00,  3.24s/it]
 89%|████████▉ | 13133/14725 [11:57:15<1:25:49,  3.23s/it]
 89%|████████▉ | 13134/14725 [11:57:18<1:25:41,  3.23s/it]
 89%|████████▉ | 13135/14725 [11:57:22<1:26:30,  3.26s/it]
 89%|████████▉ | 13136/14725
0: {'loss': 0.6828, 'grad_norm': 0.3825697261943994, 'learning_rate': 1e-05, 'epoch': 0.89}
0:  [11:57:25<1:26:21,  3.26s/it]
 89%|████████▉ | 13137/14725 [11:57:28<1:25:05,  3.21s/it]
 89%|████████▉ | 13138/14725 [11:57:31<1:24:35,  3.20s/it]
 89%|████████▉ | 13139/14725 [11:57:34<1:24:27,  3.20s/it]
 89%|████████▉ | 13140/14725 [11:57:38<1:24:19,  3.19s/it]
                                                          

 89%|████████▉ | 13140/14725 [11:57:38<1:24:19,  3.19s/it]
 89%|████████▉ | 13141/14725 [11:57:41<1:25:56,  3.26s/it]
 89%|████████▉ | 13142/14725 [11:57:44<1:26:08,  3.26s/it]
 89%|████████▉ | 13143/14725 [11:57:47<1:24:47,  3.22s/it]
 89%|████████▉ | 13144/14725 [11:57:51<1:24:39,  3.21s/it]
 89%|████████▉ | 13145/14725 [11:57:54<1:23:50,  3.18s/it]
 89%|████████▉ | 13146/14725 [11:57:57<1:25:01,  3.23s/it]
 89%|████████▉ | 13147/14725 [11:58:00<1:24:04,  3.20s/it]
 89%|█�
0: {'loss': 0.6758, 'grad_norm': 0.40255678055940275, 'learning_rate': 1e-05, 'epoch': 0.89}
0: ��██████▉ | 13148/14725 [11:58:03<1:24:11,  3.20s/it]
 89%|████████▉ | 13149/14725 [11:58:07<1:23:22,  3.17s/it]
 89%|████████▉ | 13150/14725 [11:58:10<1:25:05,  3.24s/it]
                                                          

 89%|████████▉ | 13150/14725 [11:58:10<1:25:05,  3.24s/it]
 89%|████████▉ | 13151/14725 [11:58:13<1:26:06,  3.28s/it]
 89%|████████▉ | 13152/14725 [11:58:17<1:26:18,  3.29s/it]
 89%|████████▉ | 13153/14725 [11:58:20<1:25:09,  3.25s/it]
 89%|████████▉ | 13154/14725 [11:58:23<1:23:35,  3.19s/it]
 89%|████████▉ | 13155/14725 [11:58:26<1:24:30,  3.23s/it]
 89%|████████▉ | 13156/14725 [11:58:29<1:24:07,  3.22s/it]
 89%|████████▉ | 13157/14725 [11:58:32<1:22:53,  3.17s/it]
 89%|████████▉ | 13158/14725 [11:58:36<1:24:25,  3.23s/it]
 89%|████████▉ | 13159/14725 [1
0: {'loss': 0.6763, 'grad_norm': 0.40803152824685435, 'learning_rate': 1e-05, 'epoch': 0.89}
0: 1:58:39<1:24:43,  3.25s/it]
 89%|████████▉ | 13160/14725 [11:58:42<1:24:35,  3.24s/it]
                                                          

 89%|████████▉ | 13160/14725 [11:58:42<1:24:35,  3.24s/it]
 89%|████████▉ | 13161/14725 [11:58:45<1:22:58,  3.18s/it]
 89%|████████▉ | 13162/14725 [11:58:49<1:23:18,  3.20s/it]
 89%|████████▉ | 13163/14725 [11:58:52<1:22:56,  3.19s/it]
 89%|████████▉ | 13164/14725 [11:58:55<1:23:05,  3.19s/it]
 89%|████████▉ | 13165/14725 [11:58:58<1:23:24,  3.21s/it]
 89%|████████▉ | 13166/14725 [11:59:01<1:22:34,  3.18s/it]
 89%|████████▉ | 13167/14725 [11:59:05<1:22:52,  3.19s/it]
 89%|████████▉ | 13168/14725 [11:59:08<1:22:48,  3.19s/it]
 89%|████████▉ | 13169/14725 [11:59:11<1:22:19,  3.17s/it]
 89%|████████▉ | 13170/14725 [11:59:14<1:24:15,  3.25s/it]
            
0: {'loss': 0.6723, 'grad_norm': 0.3956273116089588, 'learning_rate': 1e-05, 'epoch': 0.89}
0: {'loss': 0.673, 'grad_norm': 0.42023455683100047, 'learning_rate': 1e-05, 'epoch': 0.9}
0:                                               

 89%|████████▉ | 13170/14725 [11:59:14<1:24:15,  3.25s/it]
 89%|████████▉ | 13171/14725 [11:59:17<1:23:55,  3.24s/it]
 89%|████████▉ | 13172/14725 [11:59:21<1:23:13,  3.22s/it]
 89%|████████▉ | 13173/14725 [11:59:24<1:22:49,  3.20s/it]
 89%|████████▉ | 13174/14725 [11:59:27<1:22:44,  3.20s/it]
 89%|████████▉ | 13175/14725 [11:59:30<1:23:58,  3.25s/it]
 89%|████████▉ | 13176/14725 [11:59:34<1:24:29,  3.27s/it]
 89%|████████▉ | 13177/14725 [11:59:37<1:23:53,  3.25s/it]
 89%|████████▉ | 13178/14725 [11:59:40<1:24:39,  3.28s/it]
 90%|████████▉ | 13179/14725 [11:59:43<1:23:38,  3.25s/it]
 90%|████████▉ | 13180/14725 [11:59:46<1:21:57,  3.18s/it]
                                                          

 90%|████████▉ | 13180/14725 [11:59:46<1:21:57,  3.
0: {'loss': 0.6583, 'grad_norm': 0.3973789245269285, 'learning_rate': 1e-05, 'epoch': 0.9}
0: 18s/it]
 90%|████████▉ | 13181/14725 [11:59:50<1:22:56,  3.22s/it]
 90%|████████▉ | 13182/14725 [11:59:53<1:21:52,  3.18s/it]
 90%|████████▉ | 13183/14725 [11:59:56<1:23:12,  3.24s/it]
 90%|████████▉ | 13184/14725 [12:00:00<1:25:03,  3.31s/it]
 90%|████████▉ | 13185/14725 [12:00:03<1:23:23,  3.25s/it]
 90%|████████▉ | 13186/14725 [12:00:06<1:23:27,  3.25s/it]
 90%|████████▉ | 13187/14725 [12:00:09<1:23:59,  3.28s/it]
 90%|████████▉ | 13188/14725 [12:00:13<1:24:52,  3.31s/it]
 90%|████████▉ | 13189/14725 [12:00:16<1:24:35,  3.30s/it]
 90%|████████▉ | 13190/14725 [12:00:19<1:24:08,  3.29s/it]
                                                          

 90%|████████▉ | 13190/14725 [12:00:19<1:24:08,  3.29s/it]
 90%|████████▉ | 13191/14725 [12:00:23<1:24:28,  3.30s/it]
 90%|████████▉
0: {'loss': 0.6741, 'grad_norm': 0.4019632397249504, 'learning_rate': 1e-05, 'epoch': 0.9}
0:  | 13192/14725 [12:00:26<1:23:04,  3.25s/it]
 90%|████████▉ | 13193/14725 [12:00:29<1:24:17,  3.30s/it]
 90%|████████▉ | 13194/14725 [12:00:33<1:25:29,  3.35s/it]
 90%|████████▉ | 13195/14725 [12:00:36<1:24:30,  3.31s/it]
 90%|████████▉ | 13196/14725 [12:00:39<1:24:18,  3.31s/it]
 90%|████████▉ | 13197/14725 [12:00:43<1:24:30,  3.32s/it]
 90%|████████▉ | 13198/14725 [12:00:46<1:22:41,  3.25s/it]
 90%|████████▉ | 13199/14725 [12:00:49<1:22:24,  3.24s/it]
 90%|████████▉ | 13200/14725 [12:00:52<1:22:07,  3.23s/it]
                                                          

 90%|████████▉ | 13200/14725 [12:00:52<1:22:07,  3.23s/it]
 90%|████████▉ | 13201/14725 [12:00:55<1:22:35,  3.25s/it]
 90%|████████▉ | 13202/14725 [12:00:59<1:23:04,  3.27s/it]
 90%|████████▉ | 13203/14725 [12:01:02<1:23:30,  3.29s
0: {'loss': 0.658, 'grad_norm': 0.40264632814323104, 'learning_rate': 1e-05, 'epoch': 0.9}
0: /it]
 90%|████████▉ | 13204/14725 [12:01:05<1:24:33,  3.34s/it]
 90%|████████▉ | 13205/14725 [12:01:09<1:24:23,  3.33s/it]
 90%|████████▉ | 13206/14725 [12:01:12<1:23:42,  3.31s/it]
 90%|████████▉ | 13207/14725 [12:01:15<1:23:01,  3.28s/it]
 90%|████████▉ | 13208/14725 [12:01:18<1:21:42,  3.23s/it]
 90%|████████▉ | 13209/14725 [12:01:22<1:21:09,  3.21s/it]
 90%|████████▉ | 13210/14725 [12:01:25<1:21:05,  3.21s/it]
                                                          

 90%|████████▉ | 13210/14725 [12:01:25<1:21:05,  3.21s/it]
 90%|████████▉ | 13211/14725 [12:01:28<1:21:02,  3.21s/it]
 90%|████████▉ | 13212/14725 [12:01:31<1:20:04,  3.18s/it]
 90%|████████▉ | 13213/14725 [12:01:34<1:20:28,  3.19s/it]
 90%|████████▉ | 13214/14725 [12:01:38<1:22:08,  3.26s/it]
 90%|████████▉ | 
0: {'loss': 0.6707, 'grad_norm': 0.39524651999758625, 'learning_rate': 1e-05, 'epoch': 0.9}
0: 13215/14725 [12:01:41<1:23:42,  3.33s/it]
 90%|████████▉ | 13216/14725 [12:01:44<1:22:49,  3.29s/it]
 90%|████████▉ | 13217/14725 [12:01:48<1:23:30,  3.32s/it]
 90%|████████▉ | 13218/14725 [12:01:51<1:23:11,  3.31s/it]
 90%|████████▉ | 13219/14725 [12:01:54<1:22:27,  3.29s/it]
 90%|████████▉ | 13220/14725 [12:01:58<1:21:44,  3.26s/it]
                                                          

 90%|████████▉ | 13220/14725 [12:01:58<1:21:44,  3.26s/it]
 90%|████████▉ | 13221/14725 [12:02:01<1:21:04,  3.23s/it]
 90%|████████▉ | 13222/14725 [12:02:04<1:20:35,  3.22s/it]
 90%|████████▉ | 13223/14725 [12:02:07<1:21:19,  3.25s/it]
 90%|████████▉ | 13224/14725 [12:02:10<1:21:00,  3.24s/it]
 90%|████████▉ | 13225/14725 [12:02:14<1:22:27,  3.30s/it]
 90%|████████▉ | 13226/14725 [12:02:17<1:21:31,  3.26s/it
0: {'loss': 0.6785, 'grad_norm': 0.39128703593198355, 'learning_rate': 1e-05, 'epoch': 0.9}
0: ]
 90%|████████▉ | 13227/14725 [12:02:20<1:19:59,  3.20s/it]
 90%|████████▉ | 13228/14725 [12:02:23<1:19:24,  3.18s/it]
 90%|████████▉ | 13229/14725 [12:02:26<1:19:06,  3.17s/it]
 90%|████████▉ | 13230/14725 [12:02:30<1:19:39,  3.20s/it]
                                                          

 90%|████████▉ | 13230/14725 [12:02:30<1:19:39,  3.20s/it]
 90%|████████▉ | 13231/14725 [12:02:33<1:19:25,  3.19s/it]
 90%|████████▉ | 13232/14725 [12:02:36<1:19:10,  3.18s/it]
 90%|████████▉ | 13233/14725 [12:02:39<1:20:08,  3.22s/it]
 90%|████████▉ | 13234/14725 [12:02:43<1:20:47,  3.25s/it]
 90%|████████▉ | 13235/14725 [12:02:46<1:21:33,  3.28s/it]
 90%|████████▉ | 13236/14725 [12:02:49<1:22:05,  3.31s/it]
 90%|████████▉ | 13237/14725 [12:02:53<1:21:45,  3.30s/it]
 90%|████████▉ | 132
0: {'loss': 0.6777, 'grad_norm': 0.381753888312441, 'learning_rate': 1e-05, 'epoch': 0.9}
0: 38/14725 [12:02:56<1:22:58,  3.35s/it]
 90%|████████▉ | 13239/14725 [12:02:59<1:22:18,  3.32s/it]
 90%|████████▉ | 13240/14725 [12:03:03<1:23:15,  3.36s/it]
                                                          

 90%|████████▉ | 13240/14725 [12:03:03<1:23:15,  3.36s/it]
 90%|████████▉ | 13241/14725 [12:03:06<1:21:26,  3.29s/it]
 90%|████████▉ | 13242/14725 [12:03:09<1:21:09,  3.28s/it]
 90%|████████▉ | 13243/14725 [12:03:12<1:20:03,  3.24s/it]
 90%|████████▉ | 13244/14725 [12:03:15<1:19:21,  3.21s/it]
 90%|████████▉ | 13245/14725 [12:03:19<1:19:58,  3.24s/it]
 90%|████████▉ | 13246/14725 [12:03:22<1:20:02,  3.25s/it]
 90%|████████▉ | 13247/14725 [12:03:26<1:22:01,  3.33s/it]
 90%|████████▉ | 13248/14725 [12:03:29<1:22:42,  3.36s/it]
 90%|████████▉ | 13249/14725 [12:03:32<1:22:18,  3.35s/it]
 
0: {'loss': 0.694, 'grad_norm': 0.42699336600719623, 'learning_rate': 1e-05, 'epoch': 0.9}
0: 90%|████████▉ | 13250/14725 [12:03:35<1:20:22,  3.27s/it]
                                                          

 90%|████████▉ | 13250/14725 [12:03:35<1:20:22,  3.27s/it]
 90%|████████▉ | 13251/14725 [12:03:39<1:21:20,  3.31s/it]
 90%|████████▉ | 13252/14725 [12:03:42<1:22:05,  3.34s/it]
 90%|█████████ | 13253/14725 [12:03:45<1:20:37,  3.29s/it]
 90%|█████████ | 13254/14725 [12:03:49<1:20:27,  3.28s/it]
 90%|█████████ | 13255/14725 [12:03:52<1:19:40,  3.25s/it]
 90%|█████████ | 13256/14725 [12:03:55<1:18:45,  3.22s/it]
 90%|█████████ | 13257/14725 [12:03:58<1:19:42,  3.26s/it]
 90%|█████████ | 13258/14725 [12:04:02<1:20:08,  3.28s/it]
 90%|█████████ | 13259/14725 [12:04:05<1:20:00,  3.27s/it]
 90%|█████████ | 13260/14725 [12:04:08<1:19:46,  3.27s/it]
                                         
0: {'loss': 0.6683, 'grad_norm': 0.4200325308091614, 'learning_rate': 1e-05, 'epoch': 0.9}
0: {'loss': 0.6619, 'grad_norm': 0.3807500428462391, 'learning_rate': 1e-05, 'epoch': 0.9}
0:                  

 90%|█████████ | 13260/14725 [12:04:08<1:19:46,  3.27s/it]
 90%|█████████ | 13261/14725 [12:04:11<1:19:55,  3.28s/it]
 90%|█████████ | 13262/14725 [12:04:15<1:18:48,  3.23s/it]
 90%|█████████ | 13263/14725 [12:04:18<1:18:11,  3.21s/it]
 90%|█████████ | 13264/14725 [12:04:21<1:17:44,  3.19s/it]
 90%|█████████ | 13265/14725 [12:04:24<1:17:36,  3.19s/it]
 90%|█████████ | 13266/14725 [12:04:27<1:16:55,  3.16s/it]
 90%|█████████ | 13267/14725 [12:04:30<1:17:55,  3.21s/it]
 90%|█████████ | 13268/14725 [12:04:34<1:17:29,  3.19s/it]
 90%|█████████ | 13269/14725 [12:04:37<1:16:34,  3.16s/it]
 90%|█████████ | 13270/14725 [12:04:40<1:17:24,  3.19s/it]
                                                          

 90%|█████████ | 13270/14725 [12:04:40<1:17:24,  3.19s/it]
 90%|█████�
0: {'loss': 0.6734, 'grad_norm': 0.4218245964057673, 'learning_rate': 1e-05, 'epoch': 0.9}
0: ��███ | 13271/14725 [12:04:43<1:16:49,  3.17s/it]
 90%|█████████ | 13272/14725 [12:04:46<1:17:13,  3.19s/it]
 90%|█████████ | 13273/14725 [12:04:50<1:18:23,  3.24s/it]
 90%|█████████ | 13274/14725 [12:04:53<1:19:02,  3.27s/it]
 90%|█████████ | 13275/14725 [12:04:56<1:20:28,  3.33s/it]
 90%|█████████ | 13276/14725 [12:05:00<1:19:57,  3.31s/it]
 90%|█████████ | 13277/14725 [12:05:03<1:20:41,  3.34s/it]
 90%|█████████ | 13278/14725 [12:05:06<1:19:44,  3.31s/it]
 90%|█████████ | 13279/14725 [12:05:10<1:18:26,  3.25s/it]
 90%|█████████ | 13280/14725 [12:05:13<1:18:00,  3.24s/it]
                                                          

 90%|█████████ | 13280/14725 [12:05:13<1:18:00,  3.24s/it]
 90%|█████████ | 13281/14725 [12:05:16<1:20:14,  3.33s/it]
 90%|█████████ | 13282/14725 [12:05:20<1:20
0: {'loss': 0.667, 'grad_norm': 0.42178816188975654, 'learning_rate': 1e-05, 'epoch': 0.9}
0: :49,  3.36s/it]
 90%|█████████ | 13283/14725 [12:05:23<1:19:43,  3.32s/it]
 90%|█████████ | 13284/14725 [12:05:26<1:20:10,  3.34s/it]
 90%|█████████ | 13285/14725 [12:05:30<1:20:00,  3.33s/it]
 90%|█████████ | 13286/14725 [12:05:33<1:19:56,  3.33s/it]
 90%|█████████ | 13287/14725 [12:05:36<1:19:56,  3.34s/it]
 90%|█████████ | 13288/14725 [12:05:40<1:19:18,  3.31s/it]
 90%|█████████ | 13289/14725 [12:05:43<1:19:14,  3.31s/it]
 90%|█████████ | 13290/14725 [12:05:46<1:19:47,  3.34s/it]
                                                          

 90%|█████████ | 13290/14725 [12:05:46<1:19:47,  3.34s/it]
 90%|█████████ | 13291/14725 [12:05:49<1:18:38,  3.29s/it]
 90%|█████████ | 13292/14725 [12:05:53<1:18:34,  3.29s/it]
 90%|█████████ | 13293/14725 [12:05:56<1:18:13,  3.28s/it]
 90%|██████�
0: {'loss': 0.6627, 'grad_norm': 0.43575404479921825, 'learning_rate': 1e-05, 'epoch': 0.9}
0: ��██ | 13294/14725 [12:05:59<1:17:03,  3.23s/it]
 90%|█████████ | 13295/14725 [12:06:03<1:18:30,  3.29s/it]
 90%|█████████ | 13296/14725 [12:06:06<1:18:53,  3.31s/it]
 90%|█████████ | 13297/14725 [12:06:09<1:19:16,  3.33s/it]
 90%|█████████ | 13298/14725 [12:06:13<1:19:00,  3.32s/it]
 90%|█████████ | 13299/14725 [12:06:16<1:18:51,  3.32s/it]
 90%|█████████ | 13300/14725 [12:06:19<1:18:24,  3.30s/it]
                                                          

 90%|█████████ | 13300/14725 [12:06:19<1:18:24,  3.30s/it]
 90%|█████████ | 13301/14725 [12:06:22<1:18:03,  3.29s/it]
 90%|█████████ | 13302/14725 [12:06:26<1:17:07,  3.25s/it]
 90%|█████████ | 13303/14725 [12:06:29<1:16:31,  3.23s/it]
 90%|█████████ | 13304/14725 [12:06:32<1:17:33,  3.27s/it]
 90%|█████████ | 13305/14725 [12:06:35<1:16:16
0: {'loss': 0.6648, 'grad_norm': 0.3906264768546002, 'learning_rate': 1e-05, 'epoch': 0.9}
0: ,  3.22s/it]
 90%|█████████ | 13306/14725 [12:06:38<1:15:56,  3.21s/it]
 90%|█████████ | 13307/14725 [12:06:42<1:15:32,  3.20s/it]
 90%|█████████ | 13308/14725 [12:06:45<1:15:52,  3.21s/it]
 90%|█████████ | 13309/14725 [12:06:48<1:15:52,  3.22s/it]
 90%|█████████ | 13310/14725 [12:06:51<1:14:53,  3.18s/it]
                                                          

 90%|█████████ | 13310/14725 [12:06:51<1:14:53,  3.18s/it]
 90%|█████████ | 13311/14725 [12:06:54<1:14:44,  3.17s/it]
 90%|█████████ | 13312/14725 [12:06:57<1:14:14,  3.15s/it]
 90%|█████████ | 13313/14725 [12:07:01<1:13:51,  3.14s/it]
 90%|█████████ | 13314/14725 [12:07:04<1:13:50,  3.14s/it]
 90%|█████████ | 13315/14725 [12:07:07<1:15:23,  3.21s/it]
 90%|█████████ | 13316/14725 [12:07:10<1:14:46,  3.18s/it]
 90%|███████�
0: {'loss': 0.6614, 'grad_norm': 0.42383149736392073, 'learning_rate': 1e-05, 'epoch': 0.9}
0: ��█ | 13317/14725 [12:07:14<1:16:19,  3.25s/it]
 90%|█████████ | 13318/14725 [12:07:17<1:15:22,  3.21s/it]
 90%|█████████ | 13319/14725 [12:07:20<1:14:31,  3.18s/it]
 90%|█████████ | 13320/14725 [12:07:23<1:14:31,  3.18s/it]
                                                          

 90%|█████████ | 13320/14725 [12:07:23<1:14:31,  3.18s/it]
 90%|█████████ | 13321/14725 [12:07:26<1:14:57,  3.20s/it]
 90%|█████████ | 13322/14725 [12:07:30<1:15:43,  3.24s/it]
 90%|█████████ | 13323/14725 [12:07:33<1:15:41,  3.24s/it]
 90%|█████████ | 13324/14725 [12:07:36<1:16:43,  3.29s/it]
 90%|█████████ | 13325/14725 [12:07:39<1:16:00,  3.26s/it]
 90%|█████████ | 13326/14725 [12:07:43<1:15:38,  3.24s/it]
 91%|█████████ | 13327/14725 [12:07:46<1:14:05,  3.18s/it]
 91%|█████████ | 13328/14725 [12:07:49<1:13:59,  
0: {'loss': 0.677, 'grad_norm': 0.39142942591250474, 'learning_rate': 1e-05, 'epoch': 0.91}
0: 3.18s/it]
 91%|█████████ | 13329/14725 [12:07:52<1:13:52,  3.18s/it]
 91%|█████████ | 13330/14725 [12:07:56<1:17:16,  3.32s/it]
                                                          

 91%|█████████ | 13330/14725 [12:07:56<1:17:16,  3.32s/it]
 91%|█████████ | 13331/14725 [12:07:59<1:16:21,  3.29s/it]
 91%|█████████ | 13332/14725 [12:08:02<1:16:11,  3.28s/it]
 91%|█████████ | 13333/14725 [12:08:05<1:15:48,  3.27s/it]
 91%|█████████ | 13334/14725 [12:08:09<1:17:34,  3.35s/it]
 91%|█████████ | 13335/14725 [12:08:12<1:15:57,  3.28s/it]
 91%|█████████ | 13336/14725 [12:08:15<1:15:19,  3.25s/it]
 91%|█████████ | 13337/14725 [12:08:19<1:16:51,  3.32s/it]
 91%|█████████ | 13338/14725 [12:08:22<1:16:02,  3.29s/it]
 91%|█████████ | 13339/14725 [12:08:25<1:16:31,  3.31s/it]
 91%|████████�
0: {'loss': 0.6661, 'grad_norm': 0.4057907298484578, 'learning_rate': 1e-05, 'epoch': 0.91}
0: {'loss': 0.6753, 'grad_norm': 0.42142762728437, 'learning_rate': 1e-05, 'epoch': 0.91}
0: �� | 13340/14725 [12:08:28<1:14:52,  3.24s/it]
                                                          

 91%|█████████ | 13340/14725 [12:08:28<1:14:52,  3.24s/it]
 91%|█████████ | 13341/14725 [12:08:32<1:15:40,  3.28s/it]
 91%|█████████ | 13342/14725 [12:08:35<1:15:13,  3.26s/it]
 91%|█████████ | 13343/14725 [12:08:38<1:14:42,  3.24s/it]
 91%|█████████ | 13344/14725 [12:08:42<1:16:59,  3.35s/it]
 91%|█████████ | 13345/14725 [12:08:45<1:18:30,  3.41s/it]
 91%|█████████ | 13346/14725 [12:08:49<1:17:17,  3.36s/it]
 91%|█████████ | 13347/14725 [12:08:52<1:17:01,  3.35s/it]
 91%|█████████ | 13348/14725 [12:08:55<1:16:00,  3.31s/it]
 91%|█████████ | 13349/14725 [12:08:58<1:15:26,  3.29s/it]
 91%|█████████ | 13350/14725 [12:09:01<1:14:33,  3.25s/it]
                                                          

 91%|█�
0: {'loss': 0.6601, 'grad_norm': 0.38051852778874956, 'learning_rate': 1e-05, 'epoch': 0.91}
0: �███████ | 13350/14725 [12:09:01<1:14:33,  3.25s/it]
 91%|█████████ | 13351/14725 [12:09:05<1:15:03,  3.28s/it]
 91%|█████████ | 13352/14725 [12:09:08<1:14:06,  3.24s/it]
 91%|█████████ | 13353/14725 [12:09:11<1:14:51,  3.27s/it]
 91%|█████████ | 13354/14725 [12:09:15<1:14:18,  3.25s/it]
 91%|█████████ | 13355/14725 [12:09:18<1:15:16,  3.30s/it]
 91%|█████████ | 13356/14725 [12:09:21<1:14:28,  3.26s/it]
 91%|█████████ | 13357/14725 [12:09:24<1:13:55,  3.24s/it]
 91%|█████████ | 13358/14725 [12:09:27<1:13:05,  3.21s/it]
 91%|█████████ | 13359/14725 [12:09:31<1:13:05,  3.21s/it]
 91%|█████████ | 13360/14725 [12:09:34<1:12:59,  3.21s/it]
                                                          

 91%|█████████ | 13360/14725 [12:09:34<1:12:59,  3.21s/it]
 91%|█████████ | 13361/14725 [12
0: {'loss': 0.6738, 'grad_norm': 0.39343558154995106, 'learning_rate': 1e-05, 'epoch': 0.91}
0: :09:37<1:13:40,  3.24s/it]
 91%|█████████ | 13362/14725 [12:09:40<1:13:16,  3.23s/it]
 91%|█████████ | 13363/14725 [12:09:44<1:13:35,  3.24s/it]
 91%|█████████ | 13364/14725 [12:09:47<1:13:38,  3.25s/it]
 91%|█████████ | 13365/14725 [12:09:50<1:13:39,  3.25s/it]
 91%|█████████ | 13366/14725 [12:09:53<1:13:58,  3.27s/it]
 91%|█████████ | 13367/14725 [12:09:57<1:15:13,  3.32s/it]
 91%|█████████ | 13368/14725 [12:10:00<1:15:36,  3.34s/it]
 91%|█████████ | 13369/14725 [12:10:04<1:15:43,  3.35s/it]
 91%|█████████ | 13370/14725 [12:10:07<1:14:00,  3.28s/it]
                                                          

 91%|█████████ | 13370/14725 [12:10:07<1:14:00,  3.28s/it]
 91%|█████████ | 13371/14725 [12:10:10<1:15:33,  3.35s/it]
 91%|█████████ | 13372/14725 [12:10:14<1:15:38,  3.35s/it]
 91%|██�
0: {'loss': 0.6788, 'grad_norm': 0.39099091939327757, 'learning_rate': 1e-05, 'epoch': 0.91}
0: �██████ | 13373/14725 [12:10:17<1:14:09,  3.29s/it]
 91%|█████████ | 13374/14725 [12:10:20<1:13:09,  3.25s/it]
 91%|█████████ | 13375/14725 [12:10:23<1:12:06,  3.20s/it]
 91%|█████████ | 13376/14725 [12:10:26<1:12:24,  3.22s/it]
 91%|█████████ | 13377/14725 [12:10:30<1:12:13,  3.21s/it]
 91%|█████████ | 13378/14725 [12:10:33<1:12:23,  3.22s/it]
 91%|█████████ | 13379/14725 [12:10:36<1:11:58,  3.21s/it]
 91%|█████████ | 13380/14725 [12:10:39<1:11:58,  3.21s/it]
                                                          

 91%|█████████ | 13380/14725 [12:10:39<1:11:58,  3.21s/it]
 91%|█████████ | 13381/14725 [12:10:42<1:10:58,  3.17s/it]
 91%|█████████ | 13382/14725 [12:10:45<1:11:40,  3.20s/it]
 91%|█████████ | 13383/14725 [12:10:49<1:11:44,  3.21s/it]
 91%|█████████ | 13384/14725 [12:10
0: {'loss': 0.6699, 'grad_norm': 0.40632640877091486, 'learning_rate': 1e-05, 'epoch': 0.91}
0: :52<1:11:29,  3.20s/it]
 91%|█████████ | 13385/14725 [12:10:55<1:12:46,  3.26s/it]
 91%|█████████ | 13386/14725 [12:10:58<1:11:08,  3.19s/it]
 91%|█████████ | 13387/14725 [12:11:01<1:10:40,  3.17s/it]
 91%|█████████ | 13388/14725 [12:11:05<1:10:59,  3.19s/it]
 91%|█████████ | 13389/14725 [12:11:08<1:13:09,  3.29s/it]
 91%|█████████ | 13390/14725 [12:11:12<1:13:18,  3.29s/it]
                                                          

 91%|█████████ | 13390/14725 [12:11:12<1:13:18,  3.29s/it]
 91%|█████████ | 13391/14725 [12:11:15<1:12:47,  3.27s/it]
 91%|█████████ | 13392/14725 [12:11:18<1:12:35,  3.27s/it]
 91%|█████████ | 13393/14725 [12:11:21<1:11:49,  3.24s/it]
 91%|█████████ | 13394/14725 [12:11:24<1:11:53,  3.24s/it]
 91%|█████████ | 13395/14725 [12:11:27<1:10:21,  3.17s/it]
 91%|███�
0: {'loss': 0.6707, 'grad_norm': 0.40472674563401867, 'learning_rate': 1e-05, 'epoch': 0.91}
0: �█████ | 13396/14725 [12:11:31<1:11:40,  3.24s/it]
 91%|█████████ | 13397/14725 [12:11:34<1:12:24,  3.27s/it]
 91%|█████████ | 13398/14725 [12:11:37<1:12:45,  3.29s/it]
 91%|█████████ | 13399/14725 [12:11:41<1:12:29,  3.28s/it]
 91%|█████████ | 13400/14725 [12:11:44<1:11:07,  3.22s/it]
                                                          

 91%|█████████ | 13400/14725 [12:11:44<1:11:07,  3.22s/it]
 91%|█████████ | 13401/14725 [12:11:47<1:11:05,  3.22s/it]
 91%|█████████ | 13402/14725 [12:11:50<1:10:58,  3.22s/it]
 91%|█████████ | 13403/14725 [12:11:54<1:11:24,  3.24s/it]
 91%|█████████ | 13404/14725 [12:11:57<1:12:18,  3.28s/it]
 91%|█████████ | 13405/14725 [12:12:00<1:12:30,  3.30s/it]
 91%|█████████ | 13406/14725 [12:12:04<1:12:14,  3.29s/it]
 91%|█████████ | 13407/14725 [12:12:07
0: {'loss': 0.6794, 'grad_norm': 0.4329846147384155, 'learning_rate': 1e-05, 'epoch': 0.91}
0: <1:10:34,  3.21s/it]
 91%|█████████ | 13408/14725 [12:12:10<1:10:34,  3.22s/it]
 91%|█████████ | 13409/14725 [12:12:13<1:11:19,  3.25s/it]
 91%|█████████ | 13410/14725 [12:12:17<1:12:35,  3.31s/it]
                                                          

 91%|█████████ | 13410/14725 [12:12:17<1:12:35,  3.31s/it]
 91%|█████████ | 13411/14725 [12:12:20<1:12:17,  3.30s/it]
 91%|█████████ | 13412/14725 [12:12:23<1:12:37,  3.32s/it]
 91%|█████████ | 13413/14725 [12:12:26<1:10:55,  3.24s/it]
 91%|█████████ | 13414/14725 [12:12:30<1:11:26,  3.27s/it]
 91%|█████████ | 13415/14725 [12:12:33<1:12:32,  3.32s/it]
 91%|█████████ | 13416/14725 [12:12:36<1:12:13,  3.31s/it]
 91%|█████████ | 13417/14725 [12:12:40<1:11:25,  3.28s/it]
 91%|█████████ | 13418/14725 [12:12:43<1:12:07,  3.31s/it]
 91%|████�
0: {'loss': 0.684, 'grad_norm': 0.3871910097633349, 'learning_rate': 1e-05, 'epoch': 0.91}
0: �████ | 13419/14725 [12:12:46<1:12:02,  3.31s/it]
 91%|█████████ | 13420/14725 [12:12:50<1:12:11,  3.32s/it]
                                                          

 91%|█████████ | 13420/14725 [12:12:50<1:12:11,  3.32s/it]
 91%|█████████ | 13421/14725 [12:12:53<1:13:05,  3.36s/it]
 91%|█████████ | 13422/14725 [12:12:56<1:11:41,  3.30s/it]
 91%|█████████ | 13423/14725 [12:12:59<1:11:25,  3.29s/it]
 91%|█████████ | 13424/14725 [12:13:03<1:10:42,  3.26s/it]
 91%|█████████ | 13425/14725 [12:13:06<1:09:52,  3.23s/it]
 91%|█████████ | 13426/14725 [12:13:09<1:11:04,  3.28s/it]
 91%|█████████ | 13427/14725 [12:13:13<1:11:18,  3.30s/it]
 91%|█████████ | 13428/14725 [12:13:16<1:12:55,  3.37s/it]
 91%|█████████ | 13429/14725 [12:13:19<1:11:03,  3.29s/it]
 91%|█████████ | 13430/14725 [12:13:22<1:
0: {'loss': 0.6675, 'grad_norm': 0.40368679485649095, 'learning_rate': 1e-05, 'epoch': 0.91}
0: {'loss': 0.685, 'grad_norm': 0.37859400775991864, 'learning_rate': 1e-05, 'epoch': 0.91}
0: 10:50,  3.28s/it]
                                                          

 91%|█████████ | 13430/14725 [12:13:22<1:10:50,  3.28s/it]
 91%|█████████ | 13431/14725 [12:13:26<1:11:07,  3.30s/it]
 91%|█████████ | 13432/14725 [12:13:29<1:11:09,  3.30s/it]
 91%|█████████ | 13433/14725 [12:13:32<1:11:06,  3.30s/it]
 91%|█████████ | 13434/14725 [12:13:36<1:12:28,  3.37s/it]
 91%|█████████ | 13435/14725 [12:13:39<1:11:35,  3.33s/it]
 91%|█████████ | 13436/14725 [12:13:42<1:11:03,  3.31s/it]
 91%|█████████▏| 13437/14725 [12:13:46<1:10:37,  3.29s/it]
 91%|█████████▏| 13438/14725 [12:13:49<1:09:34,  3.24s/it]
 91%|█████████▏| 13439/14725 [12:13:52<1:08:57,  3.22s/it]
 91%|█████████▏| 13440/14725 [12:13:55<1:09:24,  3.24s/it]
                                                          

 91%|████████�
0: {'loss': 0.6692, 'grad_norm': 0.4069708409392762, 'learning_rate': 1e-05, 'epoch': 0.91}
0: �▏| 13440/14725 [12:13:55<1:09:24,  3.24s/it]
 91%|█████████▏| 13441/14725 [12:13:58<1:08:25,  3.20s/it]
 91%|█████████▏| 13442/14725 [12:14:02<1:09:11,  3.24s/it]
 91%|█████████▏| 13443/14725 [12:14:05<1:09:35,  3.26s/it]
 91%|█████████▏| 13444/14725 [12:14:08<1:08:40,  3.22s/it]
 91%|█████████▏| 13445/14725 [12:14:11<1:08:51,  3.23s/it]
 91%|█████████▏| 13446/14725 [12:14:15<1:09:48,  3.27s/it]
 91%|█████████▏| 13447/14725 [12:14:18<1:09:04,  3.24s/it]
 91%|█████████▏| 13448/14725 [12:14:21<1:08:47,  3.23s/it]
 91%|█████████▏| 13449/14725 [12:14:24<1:07:56,  3.19s/it]
 91%|█████████▏| 13450/14725 [12:14:28<1:09:26,  3.27s/it]
                                                          

 91%|█████████▏| 13450/14725 [12:14:28<1:09:26,  3.27s/it]
 91%|█████████▏| 13451/1472
0: {'loss': 0.679, 'grad_norm': 0.3992786677671621, 'learning_rate': 1e-05, 'epoch': 0.91}
0: 5 [12:14:31<1:08:11,  3.21s/it]
 91%|█████████▏| 13452/14725 [12:14:34<1:08:26,  3.23s/it]
 91%|█████████▏| 13453/14725 [12:14:37<1:09:20,  3.27s/it]
 91%|█████████▏| 13454/14725 [12:14:41<1:09:39,  3.29s/it]
 91%|█████████▏| 13455/14725 [12:14:44<1:09:49,  3.30s/it]
 91%|█████████▏| 13456/14725 [12:14:47<1:09:58,  3.31s/it]
 91%|█████████▏| 13457/14725 [12:14:51<1:10:03,  3.31s/it]
 91%|█████████▏| 13458/14725 [12:14:54<1:09:27,  3.29s/it]
 91%|█████████▏| 13459/14725 [12:14:57<1:08:58,  3.27s/it]
 91%|█████████▏| 13460/14725 [12:15:00<1:08:15,  3.24s/it]
                                                          

 91%|█████████▏| 13460/14725 [12:15:00<1:08:15,  3.24s/it]
 91%|█████████▏| 13461/14725 [12:15:03<1:07:47,  3.22s/it]
 91%|█████████▏| 13462/14725 [12:15:07<1:09
0: {'loss': 0.669, 'grad_norm': 0.4642615125005842, 'learning_rate': 1e-05, 'epoch': 0.91}
0: :53,  3.32s/it]
 91%|█████████▏| 13463/14725 [12:15:10<1:09:15,  3.29s/it]
 91%|█████████▏| 13464/14725 [12:15:14<1:09:28,  3.31s/it]
 91%|█████████▏| 13465/14725 [12:15:17<1:10:10,  3.34s/it]
 91%|█████████▏| 13466/14725 [12:15:20<1:09:27,  3.31s/it]
 91%|█████████▏| 13467/14725 [12:15:24<1:09:19,  3.31s/it]
 91%|█████████▏| 13468/14725 [12:15:27<1:08:34,  3.27s/it]
 91%|█████████▏| 13469/14725 [12:15:30<1:08:28,  3.27s/it]
 91%|█████████▏| 13470/14725 [12:15:33<1:09:31,  3.32s/it]
                                                          

 91%|█████████▏| 13470/14725 [12:15:33<1:09:31,  3.32s/it]
 91%|█████████▏| 13471/14725 [12:15:37<1:08:37,  3.28s/it]
 91%|█████████▏| 13472/14725 [12:15:40<1:07:13,  3.22s/it]
 91%|█████████▏| 13473/14725 [12:15:43<1:08:43,  3.29s/it]
0: {'loss': 0.6764, 'grad_norm': 0.41588693208004296, 'learning_rate': 1e-05, 'epoch': 0.92}
0:  92%|█████████▏| 13474/14725 [12:15:46<1:07:58,  3.26s/it]
 92%|█████████▏| 13475/14725 [12:15:50<1:07:43,  3.25s/it]
 92%|█████████▏| 13476/14725 [12:15:53<1:06:39,  3.20s/it]
 92%|█████████▏| 13477/14725 [12:15:56<1:07:31,  3.25s/it]
 92%|█████████▏| 13478/14725 [12:15:59<1:07:38,  3.25s/it]
 92%|█████████▏| 13479/14725 [12:16:03<1:07:31,  3.25s/it]
 92%|█████████▏| 13480/14725 [12:16:06<1:07:39,  3.26s/it]
                                                          

 92%|█████████▏| 13480/14725 [12:16:06<1:07:39,  3.26s/it]
 92%|█████████▏| 13481/14725 [12:16:09<1:08:04,  3.28s/it]
 92%|█████████▏| 13482/14725 [12:16:12<1:07:26,  3.26s/it]
 92%|█████████▏| 13483/14725 [12:16:16<1:08:03,  3.29s/it]
 92%|█████████▏| 13484/14725 [12:16:19<1:07:38,  3.27s/it]
 92%|███�
0: {'loss': 0.6823, 'grad_norm': 0.41583016187252186, 'learning_rate': 1e-05, 'epoch': 0.92}
0: �█████▏| 13485/14725 [12:16:22<1:06:29,  3.22s/it]
 92%|█████████▏| 13486/14725 [12:16:25<1:06:43,  3.23s/it]
 92%|█████████▏| 13487/14725 [12:16:29<1:07:13,  3.26s/it]
 92%|█████████▏| 13488/14725 [12:16:32<1:07:04,  3.25s/it]
 92%|█████████▏| 13489/14725 [12:16:35<1:07:13,  3.26s/it]
 92%|█████████▏| 13490/14725 [12:16:38<1:06:46,  3.24s/it]
                                                          

 92%|█████████▏| 13490/14725 [12:16:38<1:06:46,  3.24s/it]
 92%|█████████▏| 13491/14725 [12:16:42<1:06:02,  3.21s/it]
 92%|█████████▏| 13492/14725 [12:16:45<1:06:07,  3.22s/it]
 92%|█████████▏| 13493/14725 [12:16:48<1:06:03,  3.22s/it]
 92%|█████████▏| 13494/14725 [12:16:51<1:06:48,  3.26s/it]
 92%|█████████▏| 13495/14725 [12:16:55<1:06:51,  3.26s/it]
 92%|█████████
0: {'loss': 0.6723, 'grad_norm': 0.398507071871603, 'learning_rate': 1e-05, 'epoch': 0.92}
0: ▏| 13496/14725 [12:16:58<1:07:28,  3.29s/it]
 92%|█████████▏| 13497/14725 [12:17:01<1:08:04,  3.33s/it]
 92%|█████████▏| 13498/14725 [12:17:05<1:08:28,  3.35s/it]
 92%|█████████▏| 13499/14725 [12:17:08<1:08:40,  3.36s/it]
 92%|█████████▏| 13500/14725 [12:17:11<1:06:46,  3.27s/it]
                                                          

 92%|█████████▏| 13500/14725 [12:17:11<1:06:46,  3.27s/it]
 92%|█████████▏| 13501/14725 [12:17:14<1:05:48,  3.23s/it]
 92%|█████████▏| 13502/14725 [12:17:18<1:05:29,  3.21s/it]
 92%|█████████▏| 13503/14725 [12:17:21<1:06:17,  3.26s/it]
 92%|█████████▏| 13504/14725 [12:17:24<1:06:47,  3.28s/it]
 92%|█████████▏| 13505/14725 [12:17:27<1:05:54,  3.24s/it]
 92%|█████████▏| 13506/14725 [12:17:31<1:05:47,  3.24s/it]
 92%|█████████▏| 13507/14725
0: {'loss': 0.6796, 'grad_norm': 0.4038953499463243, 'learning_rate': 1e-05, 'epoch': 0.92}
0:  [12:17:34<1:05:19,  3.22s/it]
 92%|█████████▏| 13508/14725 [12:17:37<1:04:49,  3.20s/it]
 92%|█████████▏| 13509/14725 [12:17:40<1:05:13,  3.22s/it]
 92%|█████████▏| 13510/14725 [12:17:44<1:05:58,  3.26s/it]
                                                          

 92%|█████████▏| 13510/14725 [12:17:44<1:05:58,  3.26s/it]
 92%|█████████▏| 13511/14725 [12:17:47<1:05:02,  3.21s/it]
 92%|█████████▏| 13512/14725 [12:17:50<1:05:12,  3.23s/it]
 92%|█████████▏| 13513/14725 [12:17:53<1:05:50,  3.26s/it]
 92%|█████████▏| 13514/14725 [12:17:57<1:05:54,  3.27s/it]
 92%|█████████▏| 13515/14725 [12:18:00<1:05:44,  3.26s/it]
 92%|█████████▏| 13516/14725 [12:18:03<1:05:24,  3.25s/it]
 92%|█████████▏| 13517/14725 [12:18:06<1:05:54,  3.27s/it]
 92%|█████████▏| 13518/14725 [12:18:10<1:06:
0: {'loss': 0.67, 'grad_norm': 0.40006113013373107, 'learning_rate': 1e-05, 'epoch': 0.92}
0: 40,  3.31s/it]
 92%|█████████▏| 13519/14725 [12:18:13<1:07:17,  3.35s/it]
 92%|█████████▏| 13520/14725 [12:18:16<1:06:58,  3.34s/it]
                                                          

 92%|█████████▏| 13520/14725 [12:18:16<1:06:58,  3.34s/it]
 92%|█████████▏| 13521/14725 [12:18:20<1:07:01,  3.34s/it]
 92%|█████████▏| 13522/14725 [12:18:23<1:07:08,  3.35s/it]
 92%|█████████▏| 13523/14725 [12:18:26<1:06:24,  3.31s/it]
 92%|█████████▏| 13524/14725 [12:18:30<1:05:38,  3.28s/it]
 92%|█████████▏| 13525/14725 [12:18:33<1:04:52,  3.24s/it]
 92%|█████████▏| 13526/14725 [12:18:36<1:05:31,  3.28s/it]
 92%|█████████▏| 13527/14725 [12:18:39<1:04:41,  3.24s/it]
 92%|█████████▏| 13528/14725 [12:18:42<1:04:35,  3.24s/it]
 92%|█████████▏| 13529/14725 [12:18:46<1:05:08,  3.27s/it]
 
0: {'loss': 0.6745, 'grad_norm': 0.39942609587659683, 'learning_rate': 1e-05, 'epoch': 0.92}
0: 92%|█████████▏| 13530/14725 [12:18:49<1:03:54,  3.21s/it]
                                                          

 92%|█████████▏| 13530/14725 [12:18:49<1:03:54,  3.21s/it]
 92%|█████████▏| 13531/14725 [12:18:52<1:03:35,  3.20s/it]
 92%|█████████▏| 13532/14725 [12:18:55<1:04:30,  3.24s/it]
 92%|█████████▏| 13533/14725 [12:18:59<1:04:42,  3.26s/it]
 92%|█████████▏| 13534/14725 [12:19:02<1:04:59,  3.27s/it]
 92%|█████████▏| 13535/14725 [12:19:05<1:05:31,  3.30s/it]
 92%|█████████▏| 13536/14725 [12:19:09<1:05:05,  3.29s/it]
 92%|█████████▏| 13537/14725 [12:19:12<1:04:19,  3.25s/it]
 92%|█████████▏| 13538/14725 [12:19:15<1:04:06,  3.24s/it]
 92%|█████████▏| 13539/14725 [12:19:18<1:03:08,  3.19s/it]
 92%|█████████▏| 13540/14725 [12:19:22<1:04:53,  3.29s/it]
                 
0: {'loss': 0.6765, 'grad_norm': 0.4195167187014657, 'learning_rate': 1e-05, 'epoch': 0.92}
0: {'loss': 0.6835, 'grad_norm': 0.3925071756445406, 'learning_rate': 1e-05, 'epoch': 0.92}
0:                                          

 92%|█████████▏| 13540/14725 [12:19:22<1:04:53,  3.29s/it]
 92%|█████████▏| 13541/14725 [12:19:25<1:04:11,  3.25s/it]
 92%|█████████▏| 13542/14725 [12:19:28<1:03:40,  3.23s/it]
 92%|█████████▏| 13543/14725 [12:19:31<1:04:44,  3.29s/it]
 92%|█████████▏| 13544/14725 [12:19:35<1:04:39,  3.28s/it]
 92%|█████████▏| 13545/14725 [12:19:38<1:04:58,  3.30s/it]
 92%|█████████▏| 13546/14725 [12:19:41<1:03:18,  3.22s/it]
 92%|█████████▏| 13547/14725 [12:19:44<1:03:06,  3.21s/it]
 92%|█████████▏| 13548/14725 [12:19:48<1:03:27,  3.24s/it]
 92%|█████████▏| 13549/14725 [12:19:51<1:03:52,  3.26s/it]
 92%|█████████▏| 13550/14725 [12:19:54<1:04:59,  3.32s/it]
                                                          

 92%|█████████▏| 13550/14725 [12
0: {'loss': 0.653, 'grad_norm': 0.44590273666530417, 'learning_rate': 1e-05, 'epoch': 0.92}
0: :19:54<1:04:59,  3.32s/it]
 92%|█████████▏| 13551/14725 [12:19:58<1:04:29,  3.30s/it]
 92%|█████████▏| 13552/14725 [12:20:01<1:04:09,  3.28s/it]
 92%|█████████▏| 13553/14725 [12:20:04<1:03:58,  3.28s/it]
 92%|█████████▏| 13554/14725 [12:20:07<1:03:32,  3.26s/it]
 92%|█████████▏| 13555/14725 [12:20:11<1:03:53,  3.28s/it]
 92%|█████████▏| 13556/14725 [12:20:14<1:03:09,  3.24s/it]
 92%|█████████▏| 13557/14725 [12:20:17<1:04:58,  3.34s/it]
 92%|█████████▏| 13558/14725 [12:20:20<1:03:21,  3.26s/it]
 92%|█████████▏| 13559/14725 [12:20:24<1:02:58,  3.24s/it]
 92%|█████████▏| 13560/14725 [12:20:27<1:03:28,  3.27s/it]
                                                          

 92%|█████████▏| 13560/14725 [12:20:27<1:03:28,  3.27s/it]
 92%|█████████▏| 13561/14725 [12:20:30<1:04:08, 
0: {'loss': 0.6748, 'grad_norm': 0.3941907897312781, 'learning_rate': 1e-05, 'epoch': 0.92}
0:  3.31s/it]
 92%|█████████▏| 13562/14725 [12:20:33<1:03:15,  3.26s/it]
 92%|█████████▏| 13563/14725 [12:20:37<1:03:28,  3.28s/it]
 92%|█████████▏| 13564/14725 [12:20:40<1:03:49,  3.30s/it]
 92%|█████████▏| 13565/14725 [12:20:43<1:02:48,  3.25s/it]
 92%|█████████▏| 13566/14725 [12:20:46<1:02:02,  3.21s/it]
 92%|█████████▏| 13567/14725 [12:20:50<1:03:22,  3.28s/it]
 92%|█████████▏| 13568/14725 [12:20:53<1:02:39,  3.25s/it]
 92%|█████████▏| 13569/14725 [12:20:56<1:02:42,  3.26s/it]
 92%|█████████▏| 13570/14725 [12:21:00<1:03:13,  3.28s/it]
                                                          

 92%|█████████▏| 13570/14725 [12:21:00<1:03:13,  3.28s/it]
 92%|█████████▏| 13571/14725 [12:21:03<1:03:09,  3.28s/it]
 92%|█████████▏| 13572/14725 [12:21:06<1:02:43,  3.26s/it]
 92%|
0: {'loss': 0.6468, 'grad_norm': 0.38752427730251815, 'learning_rate': 1e-05, 'epoch': 0.92}
0: █████████▏| 13573/14725 [12:21:09<1:01:45,  3.22s/it]
 92%|█████████▏| 13574/14725 [12:21:12<1:01:29,  3.21s/it]
 92%|█████████▏| 13575/14725 [12:21:16<1:01:07,  3.19s/it]
 92%|█████████▏| 13576/14725 [12:21:19<1:01:29,  3.21s/it]
 92%|█████████▏| 13577/14725 [12:21:22<1:02:25,  3.26s/it]
 92%|█████████▏| 13578/14725 [12:21:25<1:02:23,  3.26s/it]
 92%|█████████▏| 13579/14725 [12:21:28<1:00:34,  3.17s/it]
 92%|█████████▏| 13580/14725 [12:21:32<1:01:21,  3.22s/it]
                                                          

 92%|█████████▏| 13580/14725 [12:21:32<1:01:21,  3.22s/it]
 92%|█████████▏| 13581/14725 [12:21:35<1:00:42,  3.18s/it]
 92%|█████████▏| 13582/14725 [12:21:38<1:00:41,  3.19s/it]
 92%|█████████▏| 13583/14725 [12:21:41<1:00:33,  3.18s/it]
 92%|█████�
0: {'loss': 0.6495, 'grad_norm': 0.3934008351948138, 'learning_rate': 1e-05, 'epoch': 0.92}
0: ��███▏| 13584/14725 [12:21:44<1:00:25,  3.18s/it]
 92%|█████████▏| 13585/14725 [12:21:48<1:00:49,  3.20s/it]
 92%|█████████▏| 13586/14725 [12:21:51<1:00:38,  3.19s/it]
 92%|█████████▏| 13587/14725 [12:21:54<1:00:42,  3.20s/it]
 92%|█████████▏| 13588/14725 [12:21:57<1:00:11,  3.18s/it]
 92%|█████████▏| 13589/14725 [12:22:00<1:00:17,  3.18s/it]
 92%|█████████▏| 13590/14725 [12:22:04<1:01:40,  3.26s/it]
                                                          

 92%|█████████▏| 13590/14725 [12:22:04<1:01:40,  3.26s/it]
 92%|█████████▏| 13591/14725 [12:22:07<1:01:27,  3.25s/it]
 92%|█████████▏| 13592/14725 [12:22:10<1:01:52,  3.28s/it]
 92%|█████████▏| 13593/14725 [12:22:14<1:02:13,  3.30s/it]
 92%|█████████▏| 13594/14725 [12:22:17<1:02:08,  3.30s/it]
 92%|█████████▏| 
0: {'loss': 0.6862, 'grad_norm': 0.41175200468713674, 'learning_rate': 1e-05, 'epoch': 0.92}
0: 13595/14725 [12:22:20<1:01:21,  3.26s/it]
 92%|█████████▏| 13596/14725 [12:22:23<1:00:59,  3.24s/it]
 92%|█████████▏| 13597/14725 [12:22:26<1:00:05,  3.20s/it]
 92%|█████████▏| 13598/14725 [12:22:30<1:00:54,  3.24s/it]
 92%|█████████▏| 13599/14725 [12:22:33<1:00:54,  3.25s/it]
 92%|█████████▏| 13600/14725 [12:22:36<1:01:04,  3.26s/it]
                                                          

 92%|█████████▏| 13600/14725 [12:22:36<1:01:04,  3.26s/it]
 92%|█████████▏| 13601/14725 [12:22:40<1:01:11,  3.27s/it]
 92%|█████████▏| 13602/14725 [12:22:43<1:01:19,  3.28s/it]
 92%|█████████▏| 13603/14725 [12:22:46<1:01:22,  3.28s/it]
 92%|█████████▏| 13604/14725 [12:22:50<1:01:35,  3.30s/it]
 92%|█████████▏| 13605/14725 [12:22:53<1:00:52,  3.26s/it]
 92%|█████████▏| 13606/14725 [12:
0: {'loss': 0.6701, 'grad_norm': 0.38748522141440733, 'learning_rate': 1e-05, 'epoch': 0.92}
0: 22:56<1:01:35,  3.30s/it]
 92%|█████████▏| 13607/14725 [12:22:59<1:01:11,  3.28s/it]
 92%|█████████▏| 13608/14725 [12:23:03<1:00:54,  3.27s/it]
 92%|█████████▏| 13609/14725 [12:23:06<1:01:12,  3.29s/it]
 92%|█████████▏| 13610/14725 [12:23:09<59:56,  3.23s/it]  
                                                        

 92%|█████████▏| 13610/14725 [12:23:09<59:56,  3.23s/it]
 92%|█████████▏| 13611/14725 [12:23:12<1:00:04,  3.24s/it]
 92%|█████████▏| 13612/14725 [12:23:16<59:54,  3.23s/it]  
 92%|█████████▏| 13613/14725 [12:23:19<59:33,  3.21s/it]
 92%|█████████▏| 13614/14725 [12:23:22<59:18,  3.20s/it]
 92%|█████████▏| 13615/14725 [12:23:25<1:00:35,  3.28s/it]
 92%|█████████▏| 13616/14725 [12:23:28<59:43,  3.23s/it]  
 92%|█████████▏| 13617/14725 [12:23:32<59:39,  3.23s/it]
0: {'loss': 0.6763, 'grad_norm': 0.39864944599846813, 'learning_rate': 1e-05, 'epoch': 0.92}
0:  92%|█████████▏| 13618/14725 [12:23:35<1:00:42,  3.29s/it]
 92%|█████████▏| 13619/14725 [12:23:38<1:00:42,  3.29s/it]
 92%|█████████▏| 13620/14725 [12:23:42<1:00:18,  3.27s/it]
                                                          

 92%|█████████▏| 13620/14725 [12:23:42<1:00:18,  3.27s/it]
 93%|█████████▎| 13621/14725 [12:23:45<1:00:06,  3.27s/it]
 93%|█████████▎| 13622/14725 [12:23:48<59:51,  3.26s/it]  
 93%|█████████▎| 13623/14725 [12:23:51<1:00:09,  3.28s/it]
 93%|█████████▎| 13624/14725 [12:23:55<1:01:36,  3.36s/it]
 93%|█████████▎| 13625/14725 [12:23:58<1:00:40,  3.31s/it]
 93%|█████████▎| 13626/14725 [12:24:01<1:00:05,  3.28s/it]
 93%|█████████▎| 13627/14725 [12:24:05<59:44,  3.26s/it]  
 93%|█████████▎| 13628/14725 [12:24:08<59:06,  3.23s/it]
 93%|████�
0: {'loss': 0.674, 'grad_norm': 0.38367819841439865, 'learning_rate': 1e-05, 'epoch': 0.93}
0: ��████▎| 13629/14725 [12:24:11<59:49,  3.28s/it]
 93%|█████████▎| 13630/14725 [12:24:15<1:00:23,  3.31s/it]
                                                          

 93%|█████████▎| 13630/14725 [12:24:15<1:00:23,  3.31s/it]
 93%|█████████▎| 13631/14725 [12:24:18<59:45,  3.28s/it]  
 93%|█████████▎| 13632/14725 [12:24:21<58:58,  3.24s/it]
 93%|█████████▎| 13633/14725 [12:24:24<59:50,  3.29s/it]
 93%|█████████▎| 13634/14725 [12:24:28<59:44,  3.29s/it]
 93%|█████████▎| 13635/14725 [12:24:31<59:12,  3.26s/it]
 93%|█████████▎| 13636/14725 [12:24:34<59:15,  3.26s/it]
 93%|█████████▎| 13637/14725 [12:24:37<58:08,  3.21s/it]
 93%|█████████▎| 13638/14725 [12:24:40<58:07,  3.21s/it]
 93%|█████████▎| 13639/14725 [12:24:44<58:06,  3.21s/it]
 93%|█████████▎| 13640/14725 [12
0: {'loss': 0.6726, 'grad_norm': 0.4157317671575165, 'learning_rate': 1e-05, 'epoch': 0.93}
0: {'loss': 0.6735, 'grad_norm': 0.39867461209957245, 'learning_rate': 1e-05, 'epoch': 0.93}
0: :24:47<56:58,  3.15s/it]
                                                        

 93%|█████████▎| 13640/14725 [12:24:47<56:58,  3.15s/it]
 93%|█████████▎| 13641/14725 [12:24:50<57:15,  3.17s/it]
 93%|█████████▎| 13642/14725 [12:24:53<56:49,  3.15s/it]
 93%|█████████▎| 13643/14725 [12:24:56<57:01,  3.16s/it]
 93%|█████████▎| 13644/14725 [12:24:59<57:34,  3.20s/it]
 93%|█████████▎| 13645/14725 [12:25:02<57:14,  3.18s/it]
 93%|█████████▎| 13646/14725 [12:25:06<57:25,  3.19s/it]
 93%|█████████▎| 13647/14725 [12:25:09<57:48,  3.22s/it]
 93%|█████████▎| 13648/14725 [12:25:12<58:08,  3.24s/it]
 93%|█████████▎| 13649/14725 [12:25:15<57:54,  3.23s/it]
 93%|█████████▎| 13650/14725 [12:25:19<58:04,  3.24s/it]
                                                        

 93%|█████████▎|
0: {'loss': 0.6804, 'grad_norm': 0.3865816325373391, 'learning_rate': 1e-05, 'epoch': 0.93}
0:  13650/14725 [12:25:19<58:04,  3.24s/it]
 93%|█████████▎| 13651/14725 [12:25:22<59:28,  3.32s/it]
 93%|█████████▎| 13652/14725 [12:25:26<59:22,  3.32s/it]
 93%|█████████▎| 13653/14725 [12:25:29<59:11,  3.31s/it]
 93%|█████████▎| 13654/14725 [12:25:32<59:30,  3.33s/it]
 93%|█████████▎| 13655/14725 [12:25:35<58:29,  3.28s/it]
 93%|█████████▎| 13656/14725 [12:25:39<58:12,  3.27s/it]
 93%|█████████▎| 13657/14725 [12:25:42<57:42,  3.24s/it]
 93%|█████████▎| 13658/14725 [12:25:45<59:24,  3.34s/it]
 93%|█████████▎| 13659/14725 [12:25:49<58:53,  3.31s/it]
 93%|█████████▎| 13660/14725 [12:25:52<59:26,  3.35s/it]
                                                        

 93%|█████████▎| 13660/14725 [12:25:52<59:26,  3.35s/it]
 93%|█████████▎| 13661/14725 [12:25:55<58:58,  3.33s/it]
 
0: {'loss': 0.657, 'grad_norm': 0.40730632859926397, 'learning_rate': 1e-05, 'epoch': 0.93}
0: 93%|█████████▎| 13662/14725 [12:25:59<58:23,  3.30s/it]
 93%|█████████▎| 13663/14725 [12:26:02<57:44,  3.26s/it]
 93%|█████████▎| 13664/14725 [12:26:05<56:53,  3.22s/it]
 93%|█████████▎| 13665/14725 [12:26:08<55:53,  3.16s/it]
 93%|█████████▎| 13666/14725 [12:26:11<56:43,  3.21s/it]
 93%|█████████▎| 13667/14725 [12:26:15<57:30,  3.26s/it]
 93%|█████████▎| 13668/14725 [12:26:18<58:45,  3.34s/it]
 93%|█████████▎| 13669/14725 [12:26:22<59:17,  3.37s/it]
 93%|█████████▎| 13670/14725 [12:26:25<59:18,  3.37s/it]
                                                        

 93%|█████████▎| 13670/14725 [12:26:25<59:18,  3.37s/it]
 93%|█████████▎| 13671/14725 [12:26:28<58:06,  3.31s/it]
 93%|█████████▎| 13672/14725 [12:26:31<57:40,  3.29s/it]
 93%|█████████▎| 13673/
0: {'loss': 0.6658, 'grad_norm': 0.4124033283527593, 'learning_rate': 1e-05, 'epoch': 0.93}
0: 14725 [12:26:35<57:32,  3.28s/it]
 93%|█████████▎| 13674/14725 [12:26:38<57:14,  3.27s/it]
 93%|█████████▎| 13675/14725 [12:26:41<57:10,  3.27s/it]
 93%|█████████▎| 13676/14725 [12:26:45<58:02,  3.32s/it]
 93%|█████████▎| 13677/14725 [12:26:48<57:31,  3.29s/it]
 93%|█████████▎| 13678/14725 [12:26:51<57:03,  3.27s/it]
 93%|█████████▎| 13679/14725 [12:26:54<57:18,  3.29s/it]
 93%|█████████▎| 13680/14725 [12:26:58<57:06,  3.28s/it]
                                                        

 93%|█████████▎| 13680/14725 [12:26:58<57:06,  3.28s/it]
 93%|█████████▎| 13681/14725 [12:27:01<56:42,  3.26s/it]
 93%|█████████▎| 13682/14725 [12:27:04<56:32,  3.25s/it]
 93%|█████████▎| 13683/14725 [12:27:07<56:33,  3.26s/it]
 93%|█████████▎| 13684/14725 [12:27:11<57:40,  3.32s/it]
 93%|█
0: {'loss': 0.6762, 'grad_norm': 0.41964438387624287, 'learning_rate': 1e-05, 'epoch': 0.93}
0: ████████▎| 13685/14725 [12:27:14<58:19,  3.36s/it]
 93%|█████████▎| 13686/14725 [12:27:18<58:43,  3.39s/it]
 93%|█████████▎| 13687/14725 [12:27:21<58:52,  3.40s/it]
 93%|█████████▎| 13688/14725 [12:27:25<58:53,  3.41s/it]
 93%|█████████▎| 13689/14725 [12:27:28<57:30,  3.33s/it]
 93%|█████████▎| 13690/14725 [12:27:31<56:28,  3.27s/it]
                                                        

 93%|█████████▎| 13690/14725 [12:27:31<56:28,  3.27s/it]
 93%|█████████▎| 13691/14725 [12:27:34<56:41,  3.29s/it]
 93%|█████████▎| 13692/14725 [12:27:37<56:28,  3.28s/it]
 93%|█████████▎| 13693/14725 [12:27:41<55:26,  3.22s/it]
 93%|█████████▎| 13694/14725 [12:27:44<55:33,  3.23s/it]
 93%|█████████▎| 13695/14725 [12:27:47<54:43,  3.19s/it]
 93%|█████████▎| 13696/14725 [
0: {'loss': 0.6716, 'grad_norm': 0.4059951132610767, 'learning_rate': 1e-05, 'epoch': 0.93}
0: 12:27:50<54:28,  3.18s/it]
 93%|█████████▎| 13697/14725 [12:27:53<54:00,  3.15s/it]
 93%|█████████▎| 13698/14725 [12:27:56<54:50,  3.20s/it]
 93%|█████████▎| 13699/14725 [12:28:00<55:43,  3.26s/it]
 93%|█████████▎| 13700/14725 [12:28:03<55:11,  3.23s/it]
                                                        

 93%|█████████▎| 13700/14725 [12:28:03<55:11,  3.23s/it]
 93%|█████████▎| 13701/14725 [12:28:06<55:10,  3.23s/it]
 93%|█████████▎| 13702/14725 [12:28:09<54:40,  3.21s/it]
 93%|█████████▎| 13703/14725 [12:28:12<54:06,  3.18s/it]
 93%|█████████▎| 13704/14725 [12:28:16<54:36,  3.21s/it]
 93%|█████████▎| 13705/14725 [12:28:19<55:59,  3.29s/it]
 93%|█████████▎| 13706/14725 [12:28:22<55:09,  3.25s/it]
 93%|█████████▎| 13707/14725 [12:28:26<55:35,  3.28s/it]
 93%|███�
0: {'loss': 0.6783, 'grad_norm': 0.40807848548582415, 'learning_rate': 1e-05, 'epoch': 0.93}
0: ��█████▎| 13708/14725 [12:28:29<55:55,  3.30s/it]
 93%|█████████▎| 13709/14725 [12:28:32<55:24,  3.27s/it]
 93%|█████████▎| 13710/14725 [12:28:36<55:43,  3.29s/it]
                                                        

 93%|█████████▎| 13710/14725 [12:28:36<55:43,  3.29s/it]
 93%|█████████▎| 13711/14725 [12:28:39<55:53,  3.31s/it]
 93%|█████████▎| 13712/14725 [12:28:42<55:56,  3.31s/it]
 93%|█████████▎| 13713/14725 [12:28:46<56:08,  3.33s/it]
 93%|█████████▎| 13714/14725 [12:28:49<55:51,  3.31s/it]
 93%|█████████▎| 13715/14725 [12:28:52<55:11,  3.28s/it]
 93%|█████████▎| 13716/14725 [12:28:55<55:10,  3.28s/it]
 93%|█████████▎| 13717/14725 [12:28:59<54:25,  3.24s/it]
 93%|█████████▎| 13718/14725 [12:29:02<54:02,  3.22s/it]
 93%|█████████▎| 13719/14725 [12:29:0
0: {'loss': 0.6816, 'grad_norm': 0.403797906673994, 'learning_rate': 1e-05, 'epoch': 0.93}
0: 5<55:02,  3.28s/it]
 93%|█████████▎| 13720/14725 [12:29:08<55:03,  3.29s/it]
                                                        

 93%|█████████▎| 13720/14725 [12:29:08<55:03,  3.29s/it]
 93%|█████████▎| 13721/14725 [12:29:12<54:54,  3.28s/it]
 93%|█████████▎| 13722/14725 [12:29:15<54:42,  3.27s/it]
 93%|█████████▎| 13723/14725 [12:29:18<53:58,  3.23s/it]
 93%|█████████▎| 13724/14725 [12:29:22<54:48,  3.28s/it]
 93%|█████████▎| 13725/14725 [12:29:25<55:38,  3.34s/it]
 93%|█████████▎| 13726/14725 [12:29:28<55:46,  3.35s/it]
 93%|█████████▎| 13727/14725 [12:29:32<55:14,  3.32s/it]
 93%|█████████▎| 13728/14725 [12:29:35<54:01,  3.25s/it]
 93%|█████████▎| 13729/14725 [12:29:38<54:06,  3.26s/it]
 93%|█████████▎| 13730/14725 [12:29:41<53:34,  3.23s/it]
                      
0: {'loss': 0.6584, 'grad_norm': 0.41416573641166204, 'learning_rate': 1e-05, 'epoch': 0.93}
0: {'loss': 0.6773, 'grad_norm': 0.421240126931993, 'learning_rate': 1e-05, 'epoch': 0.93}
0:                                   

 93%|█████████▎| 13730/14725 [12:29:41<53:34,  3.23s/it]
 93%|█████████▎| 13731/14725 [12:29:45<55:27,  3.35s/it]
 93%|█████████▎| 13732/14725 [12:29:48<55:08,  3.33s/it]
 93%|█████████▎| 13733/14725 [12:29:51<54:24,  3.29s/it]
 93%|█████████▎| 13734/14725 [12:29:54<53:53,  3.26s/it]
 93%|█████████▎| 13735/14725 [12:29:58<53:33,  3.25s/it]
 93%|█████████▎| 13736/14725 [12:30:01<53:08,  3.22s/it]
 93%|█████████▎| 13737/14725 [12:30:04<54:10,  3.29s/it]
 93%|█████████▎| 13738/14725 [12:30:08<54:46,  3.33s/it]
 93%|█████████▎| 13739/14725 [12:30:11<54:10,  3.30s/it]
 93%|█████████▎| 13740/14725 [12:30:14<54:48,  3.34s/it]
                                                        

 93%|█████████▎| 13740/14725 [12:30:14<54:48,  3.34s/it]
 93%|�
0: {'loss': 0.6742, 'grad_norm': 0.45319189367365603, 'learning_rate': 1e-05, 'epoch': 0.93}
0: ��████████▎| 13741/14725 [12:30:18<54:06,  3.30s/it]
 93%|█████████▎| 13742/14725 [12:30:21<53:10,  3.25s/it]
 93%|█████████▎| 13743/14725 [12:30:24<53:04,  3.24s/it]
 93%|█████████▎| 13744/14725 [12:30:27<52:54,  3.24s/it]
 93%|█████████▎| 13745/14725 [12:30:30<52:49,  3.23s/it]
 93%|█████████▎| 13746/14725 [12:30:34<52:36,  3.22s/it]
 93%|█████████▎| 13747/14725 [12:30:37<53:36,  3.29s/it]
 93%|█████████▎| 13748/14725 [12:30:40<53:26,  3.28s/it]
 93%|█████████▎| 13749/14725 [12:30:44<53:23,  3.28s/it]
 93%|█████████▎| 13750/14725 [12:30:47<53:30,  3.29s/it]
                                                        

 93%|█████████▎| 13750/14725 [12:30:47<53:30,  3.29s/it]
 93%|█████████▎| 13751/14725 [12:30:50<53:52,  3.32s/it]
 93%|█████████▎| 13752/14725
0: {'loss': 0.6741, 'grad_norm': 0.3991454519978608, 'learning_rate': 1e-05, 'epoch': 0.93}
0:  [12:30:53<52:53,  3.26s/it]
 93%|█████████▎| 13753/14725 [12:30:57<54:04,  3.34s/it]
 93%|█████████▎| 13754/14725 [12:31:00<52:54,  3.27s/it]
 93%|█████████▎| 13755/14725 [12:31:03<51:48,  3.20s/it]
 93%|█████████▎| 13756/14725 [12:31:06<52:39,  3.26s/it]
 93%|█████████▎| 13757/14725 [12:31:10<51:37,  3.20s/it]
 93%|█████████▎| 13758/14725 [12:31:13<51:50,  3.22s/it]
 93%|█████████▎| 13759/14725 [12:31:16<52:22,  3.25s/it]
 93%|█████████▎| 13760/14725 [12:31:20<53:00,  3.30s/it]
                                                        

 93%|█████████▎| 13760/14725 [12:31:20<53:00,  3.30s/it]
 93%|█████████▎| 13761/14725 [12:31:23<52:41,  3.28s/it]
 93%|█████████▎| 13762/14725 [12:31:26<52:35,  3.28s/it]
 93%|█████████▎| 13763/14725 [12:31:30<53:32,  3.34s/it]
 93%|██�
0: {'loss': 0.6751, 'grad_norm': 0.3948722702266098, 'learning_rate': 1e-05, 'epoch': 0.94}
0: �██████▎| 13764/14725 [12:31:33<52:58,  3.31s/it]
 93%|█████████▎| 13765/14725 [12:31:36<52:25,  3.28s/it]
 93%|█████████▎| 13766/14725 [12:31:39<52:03,  3.26s/it]
 93%|█████████▎| 13767/14725 [12:31:42<51:47,  3.24s/it]
 94%|█████████▎| 13768/14725 [12:31:46<52:07,  3.27s/it]
 94%|█████████▎| 13769/14725 [12:31:49<51:49,  3.25s/it]
 94%|█████████▎| 13770/14725 [12:31:52<52:16,  3.28s/it]
                                                        

 94%|█████████▎| 13770/14725 [12:31:52<52:16,  3.28s/it]
 94%|█████████▎| 13771/14725 [12:31:56<52:18,  3.29s/it]
 94%|█████████▎| 13772/14725 [12:31:59<52:10,  3.28s/it]
 94%|█████████▎| 13773/14725 [12:32:02<52:06,  3.28s/it]
 94%|█████████▎| 13774/14725 [12:32:05<51:25,  3.24s/it]
 94%|█████████▎| 13775/14725 [12:32
0: {'loss': 0.6655, 'grad_norm': 0.40560265901313497, 'learning_rate': 1e-05, 'epoch': 0.94}
0: :09<51:41,  3.26s/it]
 94%|█████████▎| 13776/14725 [12:32:12<50:52,  3.22s/it]
 94%|█████████▎| 13777/14725 [12:32:15<51:08,  3.24s/it]
 94%|█████████▎| 13778/14725 [12:32:18<50:43,  3.21s/it]
 94%|█████████▎| 13779/14725 [12:32:21<50:36,  3.21s/it]
 94%|█████████▎| 13780/14725 [12:32:25<50:30,  3.21s/it]
                                                        

 94%|█████████▎| 13780/14725 [12:32:25<50:30,  3.21s/it]
 94%|█████████▎| 13781/14725 [12:32:28<50:53,  3.23s/it]
 94%|█████████▎| 13782/14725 [12:32:31<51:09,  3.26s/it]
 94%|█████████▎| 13783/14725 [12:32:34<50:37,  3.22s/it]
 94%|█████████▎| 13784/14725 [12:32:38<50:39,  3.23s/it]
 94%|█████████▎| 13785/14725 [12:32:41<50:24,  3.22s/it]
 94%|█████████▎| 13786/14725 [12:32:44<50:47,  3.24s/it]
 94%|█████
0: {'loss': 0.6619, 'grad_norm': 0.39614761008527094, 'learning_rate': 1e-05, 'epoch': 0.94}
0: ████▎| 13787/14725 [12:32:47<50:50,  3.25s/it]
 94%|█████████▎| 13788/14725 [12:32:51<51:12,  3.28s/it]
 94%|█████████▎| 13789/14725 [12:32:54<52:02,  3.34s/it]
 94%|█████████▎| 13790/14725 [12:32:57<51:59,  3.34s/it]
                                                        

 94%|█████████▎| 13790/14725 [12:32:57<51:59,  3.34s/it]
 94%|█████████▎| 13791/14725 [12:33:01<51:32,  3.31s/it]
 94%|█████████▎| 13792/14725 [12:33:04<51:02,  3.28s/it]
 94%|█████████▎| 13793/14725 [12:33:07<51:23,  3.31s/it]
 94%|█████████▎| 13794/14725 [12:33:11<51:49,  3.34s/it]
 94%|█████████▎| 13795/14725 [12:33:14<50:35,  3.26s/it]
 94%|█████████▎| 13796/14725 [12:33:17<52:16,  3.38s/it]
 94%|█████████▎| 13797/14725 [12:33:21<52:59,  3.43s/it]
 94%|█████████▎| 13798/14725 [12:33:24<52:
0: {'loss': 0.6763, 'grad_norm': 0.4016386426472818, 'learning_rate': 1e-05, 'epoch': 0.94}
0: 41,  3.41s/it]
 94%|█████████▎| 13799/14725 [12:33:28<51:31,  3.34s/it]
 94%|█████████▎| 13800/14725 [12:33:31<50:17,  3.26s/it]
                                                        

 94%|█████████▎| 13800/14725 [12:33:31<50:17,  3.26s/it]
 94%|█████████▎| 13801/14725 [12:33:34<50:17,  3.27s/it]
 94%|█████████▎| 13802/14725 [12:33:37<50:05,  3.26s/it]
 94%|█████████▎| 13803/14725 [12:33:40<49:59,  3.25s/it]
 94%|█████████▎| 13804/14725 [12:33:44<50:11,  3.27s/it]
 94%|█████████▍| 13805/14725 [12:33:47<49:49,  3.25s/it]
 94%|█████████▍| 13806/14725 [12:33:50<49:45,  3.25s/it]
 94%|█████████▍| 13807/14725 [12:33:53<49:10,  3.21s/it]
 94%|█████████▍| 13808/14725 [12:33:56<48:52,  3.20s/it]
 94%|█████████▍| 13809/14725 [12:34:00<49:41,  3.25s/it]
 94%|███████�
0: {'loss': 0.6736, 'grad_norm': 0.4197334422178165, 'learning_rate': 1e-05, 'epoch': 0.94}
0: {'loss': 0.6612, 'grad_norm': 0.371134731173471, 'learning_rate': 1e-05, 'epoch': 0.94}
0: ��█▍| 13810/14725 [12:34:03<49:33,  3.25s/it]
                                                        

 94%|█████████▍| 13810/14725 [12:34:03<49:33,  3.25s/it]
 94%|█████████▍| 13811/14725 [12:34:06<49:11,  3.23s/it]
 94%|█████████▍| 13812/14725 [12:34:09<48:52,  3.21s/it]
 94%|█████████▍| 13813/14725 [12:34:12<48:13,  3.17s/it]
 94%|█████████▍| 13814/14725 [12:34:16<47:53,  3.15s/it]
 94%|█████████▍| 13815/14725 [12:34:19<48:19,  3.19s/it]
 94%|█████████▍| 13816/14725 [12:34:22<49:10,  3.25s/it]
 94%|█████████▍| 13817/14725 [12:34:25<49:04,  3.24s/it]
 94%|█████████▍| 13818/14725 [12:34:29<49:22,  3.27s/it]
 94%|█████████▍| 13819/14725 [12:34:32<48:15,  3.20s/it]
 94%|█████████▍| 13820/14725 [12:34:35<48:00,  3.18s/it]
                                                        

 94%|██
0: {'loss': 0.6808, 'grad_norm': 0.40772319080471964, 'learning_rate': 1e-05, 'epoch': 0.94}
0: ███████▍| 13820/14725 [12:34:35<48:00,  3.18s/it]
 94%|█████████▍| 13821/14725 [12:34:38<48:09,  3.20s/it]
 94%|█████████▍| 13822/14725 [12:34:42<48:52,  3.25s/it]
 94%|█████████▍| 13823/14725 [12:34:45<49:19,  3.28s/it]
 94%|█████████▍| 13824/14725 [12:34:48<49:02,  3.27s/it]
 94%|█████████▍| 13825/14725 [12:34:51<49:03,  3.27s/it]
 94%|█████████▍| 13826/14725 [12:34:55<48:55,  3.27s/it]
 94%|█████████▍| 13827/14725 [12:34:58<48:20,  3.23s/it]
 94%|█████████▍| 13828/14725 [12:35:01<48:06,  3.22s/it]
 94%|█████████▍| 13829/14725 [12:35:04<48:26,  3.24s/it]
 94%|█████████▍| 13830/14725 [12:35:08<48:54,  3.28s/it]
                                                        

 94%|█████████▍| 13830/14725 [12:35:08<48:54,  3.28s/it]
 94%|█████████▍| 13831/14725 [12:
0: {'loss': 0.6662, 'grad_norm': 0.4279188356686232, 'learning_rate': 1e-05, 'epoch': 0.94}
0: 35:11<48:56,  3.28s/it]
 94%|█████████▍| 13832/14725 [12:35:14<47:57,  3.22s/it]
 94%|█████████▍| 13833/14725 [12:35:18<49:59,  3.36s/it]
 94%|█████████▍| 13834/14725 [12:35:21<49:10,  3.31s/it]
 94%|█████████▍| 13835/14725 [12:35:24<48:53,  3.30s/it]
 94%|█████████▍| 13836/14725 [12:35:28<48:53,  3.30s/it]
 94%|█████████▍| 13837/14725 [12:35:31<48:13,  3.26s/it]
 94%|█████████▍| 13838/14725 [12:35:34<48:26,  3.28s/it]
 94%|█████████▍| 13839/14725 [12:35:37<47:58,  3.25s/it]
 94%|█████████▍| 13840/14725 [12:35:40<48:07,  3.26s/it]
                                                        

 94%|█████████▍| 13840/14725 [12:35:40<48:07,  3.26s/it]
 94%|█████████▍| 13841/14725 [12:35:44<48:15,  3.28s/it]
 94%|█████████▍| 13842/14725 [12:35:47<47:39,  3.24s/it]
 94%|████�
0: {'loss': 0.6754, 'grad_norm': 0.4010654547275436, 'learning_rate': 1e-05, 'epoch': 0.94}
0: ��████▍| 13843/14725 [12:35:50<47:16,  3.22s/it]
 94%|█████████▍| 13844/14725 [12:35:53<47:21,  3.23s/it]
 94%|█████████▍| 13845/14725 [12:35:57<48:00,  3.27s/it]
 94%|█████████▍| 13846/14725 [12:36:00<49:43,  3.39s/it]
 94%|█████████▍| 13847/14725 [12:36:04<48:57,  3.35s/it]
 94%|█████████▍| 13848/14725 [12:36:07<49:21,  3.38s/it]
 94%|█████████▍| 13849/14725 [12:36:10<49:06,  3.36s/it]
 94%|█████████▍| 13850/14725 [12:36:14<50:23,  3.46s/it]
                                                        

 94%|█████████▍| 13850/14725 [12:36:14<50:23,  3.46s/it]
 94%|█████████▍| 13851/14725 [12:36:17<48:56,  3.36s/it]
 94%|█████████▍| 13852/14725 [12:36:20<48:09,  3.31s/it]
 94%|█████████▍| 13853/14725 [12:36:24<47:21,  3.26s/it]
 94%|█████████▍| 13854/14725 [12:36:27<4
0: {'loss': 0.6755, 'grad_norm': 0.37118362929564525, 'learning_rate': 1e-05, 'epoch': 0.94}
0: 6:58,  3.24s/it]
 94%|█████████▍| 13855/14725 [12:36:30<47:36,  3.28s/it]
 94%|█████████▍| 13856/14725 [12:36:33<46:56,  3.24s/it]
 94%|█████████▍| 13857/14725 [12:36:36<46:11,  3.19s/it]
 94%|█████████▍| 13858/14725 [12:36:40<46:44,  3.23s/it]
 94%|█████████▍| 13859/14725 [12:36:43<45:46,  3.17s/it]
 94%|█████████▍| 13860/14725 [12:36:46<45:49,  3.18s/it]
                                                        

 94%|█████████▍| 13860/14725 [12:36:46<45:49,  3.18s/it]
 94%|█████████▍| 13861/14725 [12:36:49<45:48,  3.18s/it]
 94%|█████████▍| 13862/14725 [12:36:52<45:42,  3.18s/it]
 94%|█████████▍| 13863/14725 [12:36:56<46:02,  3.20s/it]
 94%|█████████▍| 13864/14725 [12:36:59<46:30,  3.24s/it]
 94%|█████████▍| 13865/14725 [12:37:02<46:40,  3.26s/it]
 94%|██████�
0: {'loss': 0.6919, 'grad_norm': 0.40179948015519373, 'learning_rate': 1e-05, 'epoch': 0.94}
0: �██▍| 13866/14725 [12:37:06<47:05,  3.29s/it]
 94%|█████████▍| 13867/14725 [12:37:09<46:56,  3.28s/it]
 94%|█████████▍| 13868/14725 [12:37:12<47:27,  3.32s/it]
 94%|█████████▍| 13869/14725 [12:37:16<47:21,  3.32s/it]
 94%|█████████▍| 13870/14725 [12:37:19<47:32,  3.34s/it]
                                                        

 94%|█████████▍| 13870/14725 [12:37:19<47:32,  3.34s/it]
 94%|█████████▍| 13871/14725 [12:37:22<46:33,  3.27s/it]
 94%|█████████▍| 13872/14725 [12:37:25<46:29,  3.27s/it]
 94%|█████████▍| 13873/14725 [12:37:28<46:12,  3.25s/it]
 94%|█████████▍| 13874/14725 [12:37:32<46:41,  3.29s/it]
 94%|█████████▍| 13875/14725 [12:37:35<45:49,  3.23s/it]
 94%|█████████▍| 13876/14725 [12:37:38<45:48,  3.24s/it]
 94%|█████████▍| 13877/14725 [12:37:42<46:06,  
0: {'loss': 0.6425, 'grad_norm': 0.3955192784818542, 'learning_rate': 1e-05, 'epoch': 0.94}
0: 3.26s/it]
 94%|█████████▍| 13878/14725 [12:37:45<45:18,  3.21s/it]
 94%|█████████▍| 13879/14725 [12:37:48<45:32,  3.23s/it]
 94%|█████████▍| 13880/14725 [12:37:51<46:02,  3.27s/it]
                                                        

 94%|█████████▍| 13880/14725 [12:37:51<46:02,  3.27s/it]
 94%|█████████▍| 13881/14725 [12:37:54<45:48,  3.26s/it]
 94%|█████████▍| 13882/14725 [12:37:58<45:17,  3.22s/it]
 94%|█████████▍| 13883/14725 [12:38:01<44:45,  3.19s/it]
 94%|█████████▍| 13884/14725 [12:38:04<44:23,  3.17s/it]
 94%|█████████▍| 13885/14725 [12:38:07<45:07,  3.22s/it]
 94%|█████████▍| 13886/14725 [12:38:10<44:51,  3.21s/it]
 94%|█████████▍| 13887/14725 [12:38:13<44:19,  3.17s/it]
 94%|█████████▍| 13888/14725 [12:38:17<45:23,  3.25s/it]
 94%|█████████
0: {'loss': 0.666, 'grad_norm': 0.3893031777194329, 'learning_rate': 1e-05, 'epoch': 0.94}
0: ▍| 13889/14725 [12:38:20<44:56,  3.23s/it]
 94%|█████████▍| 13890/14725 [12:38:23<44:45,  3.22s/it]
                                                        

 94%|█████████▍| 13890/14725 [12:38:23<44:45,  3.22s/it]
 94%|█████████▍| 13891/14725 [12:38:26<44:41,  3.21s/it]
 94%|█████████▍| 13892/14725 [12:38:30<45:00,  3.24s/it]
 94%|█████████▍| 13893/14725 [12:38:33<45:28,  3.28s/it]
 94%|█████████▍| 13894/14725 [12:38:36<45:25,  3.28s/it]
 94%|█████████▍| 13895/14725 [12:38:40<45:49,  3.31s/it]
 94%|█████████▍| 13896/14725 [12:38:43<45:15,  3.28s/it]
 94%|█████████▍| 13897/14725 [12:38:46<45:06,  3.27s/it]
 94%|█████████▍| 13898/14725 [12:38:49<44:31,  3.23s/it]
 94%|█████████▍| 13899/14725 [12:38:53<45:37,  3.31s/it]
 94%|█████████▍| 13900/14725 [12:38:56<45:50,  3.33s/i
0: {'loss': 0.6763, 'grad_norm': 0.4205098321981457, 'learning_rate': 1e-05, 'epoch': 0.94}
0: {'loss': 0.6744, 'grad_norm': 0.40978548330430553, 'learning_rate': 1e-05, 'epoch': 0.94}
0: t]
                                                        

 94%|█████████▍| 13900/14725 [12:38:56<45:50,  3.33s/it]
 94%|█████████▍| 13901/14725 [12:39:00<45:35,  3.32s/it]
 94%|█████████▍| 13902/14725 [12:39:03<45:12,  3.30s/it]
 94%|█████████▍| 13903/14725 [12:39:06<45:45,  3.34s/it]
 94%|█████████▍| 13904/14725 [12:39:10<46:00,  3.36s/it]
 94%|█████████▍| 13905/14725 [12:39:13<45:26,  3.32s/it]
 94%|█████████▍| 13906/14725 [12:39:16<45:35,  3.34s/it]
 94%|█████████▍| 13907/14725 [12:39:20<45:15,  3.32s/it]
 94%|█████████▍| 13908/14725 [12:39:23<43:55,  3.23s/it]
 94%|█████████▍| 13909/14725 [12:39:26<43:34,  3.20s/it]
 94%|█████████▍| 13910/14725 [12:39:29<43:31,  3.20s/it]
                                                        

 94%|█████████▍| 13910/14725 [12:39:29
0: {'loss': 0.6755, 'grad_norm': 0.3999479405475373, 'learning_rate': 1e-05, 'epoch': 0.95}
0: <43:31,  3.20s/it]
 94%|█████████▍| 13911/14725 [12:39:32<44:05,  3.25s/it]
 94%|█████████▍| 13912/14725 [12:39:36<44:43,  3.30s/it]
 94%|█████████▍| 13913/14725 [12:39:39<45:14,  3.34s/it]
 94%|█████████▍| 13914/14725 [12:39:42<44:52,  3.32s/it]
 94%|█████████▍| 13915/14725 [12:39:46<44:21,  3.29s/it]
 95%|█████████▍| 13916/14725 [12:39:49<44:02,  3.27s/it]
 95%|█████████▍| 13917/14725 [12:39:52<43:59,  3.27s/it]
 95%|█████████▍| 13918/14725 [12:39:56<46:18,  3.44s/it]
 95%|█████████▍| 13919/14725 [12:39:59<45:28,  3.39s/it]
 95%|█████████▍| 13920/14725 [12:40:02<44:53,  3.35s/it]
                                                        

 95%|█████████▍| 13920/14725 [12:40:02<44:53,  3.35s/it]
 95%|█████████▍| 13921/14725 [12:40:06<44:00,  3.28s/it]
 95%|██████
0: {'loss': 0.6772, 'grad_norm': 0.3876390116595074, 'learning_rate': 1e-05, 'epoch': 0.95}
0: ███▍| 13922/14725 [12:40:09<43:27,  3.25s/it]
 95%|█████████▍| 13923/14725 [12:40:12<43:23,  3.25s/it]
 95%|█████████▍| 13924/14725 [12:40:15<43:02,  3.22s/it]
 95%|█████████▍| 13925/14725 [12:40:18<42:58,  3.22s/it]
 95%|█████████▍| 13926/14725 [12:40:22<43:32,  3.27s/it]
 95%|█████████▍| 13927/14725 [12:40:25<43:40,  3.28s/it]
 95%|█████████▍| 13928/14725 [12:40:28<43:58,  3.31s/it]
 95%|█████████▍| 13929/14725 [12:40:32<44:33,  3.36s/it]
 95%|█████████▍| 13930/14725 [12:40:35<43:51,  3.31s/it]
                                                        

 95%|█████████▍| 13930/14725 [12:40:35<43:51,  3.31s/it]
 95%|█████████▍| 13931/14725 [12:40:38<43:44,  3.31s/it]
 95%|█████████▍| 13932/14725 [12:40:42<43:06,  3.26s/it]
 95%|█████████▍| 13933/14725 [12:40:45<43:26,
0: {'loss': 0.6728, 'grad_norm': 0.4079855388724821, 'learning_rate': 1e-05, 'epoch': 0.95}
0:   3.29s/it]
 95%|█████████▍| 13934/14725 [12:40:48<43:32,  3.30s/it]
 95%|█████████▍| 13935/14725 [12:40:52<43:14,  3.28s/it]
 95%|█████████▍| 13936/14725 [12:40:55<42:57,  3.27s/it]
 95%|█████████▍| 13937/14725 [12:40:58<42:47,  3.26s/it]
 95%|█████████▍| 13938/14725 [12:41:01<41:44,  3.18s/it]
 95%|█████████▍| 13939/14725 [12:41:04<41:59,  3.21s/it]
 95%|█████████▍| 13940/14725 [12:41:08<42:03,  3.21s/it]
                                                        

 95%|█████████▍| 13940/14725 [12:41:08<42:03,  3.21s/it]
 95%|█████████▍| 13941/14725 [12:41:11<43:04,  3.30s/it]
 95%|█████████▍| 13942/14725 [12:41:14<42:59,  3.29s/it]
 95%|█████████▍| 13943/14725 [12:41:17<42:27,  3.26s/it]
 95%|█████████▍| 13944/14725 [12:41:21<42:02,  3.23s/it]
 95%|████████�
0: {'loss': 0.6713, 'grad_norm': 0.402672146067937, 'learning_rate': 1e-05, 'epoch': 0.95}
0: ��▍| 13945/14725 [12:41:24<41:56,  3.23s/it]
 95%|█████████▍| 13946/14725 [12:41:27<41:41,  3.21s/it]
 95%|█████████▍| 13947/14725 [12:41:30<41:02,  3.17s/it]
 95%|█████████▍| 13948/14725 [12:41:33<41:49,  3.23s/it]
 95%|█████████▍| 13949/14725 [12:41:37<42:24,  3.28s/it]
 95%|█████████▍| 13950/14725 [12:41:40<42:11,  3.27s/it]
                                                        

 95%|█████████▍| 13950/14725 [12:41:40<42:11,  3.27s/it]
 95%|█████████▍| 13951/14725 [12:41:43<41:47,  3.24s/it]
 95%|█████████▍| 13952/14725 [12:41:47<41:59,  3.26s/it]
 95%|█████████▍| 13953/14725 [12:41:50<41:51,  3.25s/it]
 95%|█████████▍| 13954/14725 [12:41:53<41:38,  3.24s/it]
 95%|█████████▍| 13955/14725 [12:41:56<41:58,  3.27s/it]
 95%|█████████▍| 13956/14725 [12:41:59<40:59,  3.20s
0: {'loss': 0.6822, 'grad_norm': 0.40687487520764604, 'learning_rate': 1e-05, 'epoch': 0.95}
0: /it]
 95%|█████████▍| 13957/14725 [12:42:03<41:07,  3.21s/it]
 95%|█████████▍| 13958/14725 [12:42:06<41:36,  3.26s/it]
 95%|█████████▍| 13959/14725 [12:42:09<41:13,  3.23s/it]
 95%|█████████▍| 13960/14725 [12:42:13<41:58,  3.29s/it]
                                                        

 95%|█████████▍| 13960/14725 [12:42:13<41:58,  3.29s/it]
 95%|█████████▍| 13961/14725 [12:42:16<42:20,  3.33s/it]
 95%|█████████▍| 13962/14725 [12:42:19<41:41,  3.28s/it]
 95%|█████████▍| 13963/14725 [12:42:22<40:48,  3.21s/it]
 95%|█████████▍| 13964/14725 [12:42:25<40:25,  3.19s/it]
 95%|█████████▍| 13965/14725 [12:42:28<40:08,  3.17s/it]
 95%|█████████▍| 13966/14725 [12:42:32<40:28,  3.20s/it]
 95%|█████████▍| 13967/14725 [12:42:35<40:23,  3.20s/it]
 95%|█████████▍| 
0: {'loss': 0.6651, 'grad_norm': 0.4153727173227243, 'learning_rate': 1e-05, 'epoch': 0.95}
0: 13968/14725 [12:42:38<41:05,  3.26s/it]
 95%|█████████▍| 13969/14725 [12:42:42<41:19,  3.28s/it]
 95%|█████████▍| 13970/14725 [12:42:45<41:21,  3.29s/it]
                                                        

 95%|█████████▍| 13970/14725 [12:42:45<41:21,  3.29s/it]
 95%|█████████▍| 13971/14725 [12:42:48<40:57,  3.26s/it]
 95%|█████████▍| 13972/14725 [12:42:51<40:20,  3.21s/it]
 95%|█████████▍| 13973/14725 [12:42:55<41:00,  3.27s/it]
 95%|█████████▍| 13974/14725 [12:42:58<40:17,  3.22s/it]
 95%|█████████▍| 13975/14725 [12:43:01<40:05,  3.21s/it]
 95%|█████████▍| 13976/14725 [12:43:04<40:05,  3.21s/it]
 95%|█████████▍| 13977/14725 [12:43:07<39:49,  3.19s/it]
 95%|█████████▍| 13978/14725 [12:43:11<39:46,  3.19s/it]
 95%|█████████▍| 13979/14725 [12:43:14<40:45,  3.28s/it]
 9
0: {'loss': 0.6654, 'grad_norm': 0.39458154537187284, 'learning_rate': 1e-05, 'epoch': 0.95}
0: 5%|█████████▍| 13980/14725 [12:43:17<39:52,  3.21s/it]
                                                        

 95%|█████████▍| 13980/14725 [12:43:17<39:52,  3.21s/it]
 95%|█████████▍| 13981/14725 [12:43:20<39:23,  3.18s/it]
 95%|█████████▍| 13982/14725 [12:43:24<40:52,  3.30s/it]
 95%|█████████▍| 13983/14725 [12:43:27<41:53,  3.39s/it]
 95%|█████████▍| 13984/14725 [12:43:31<41:25,  3.35s/it]
 95%|█████████▍| 13985/14725 [12:43:34<40:42,  3.30s/it]
 95%|█████████▍| 13986/14725 [12:43:37<40:48,  3.31s/it]
 95%|█████████▍| 13987/14725 [12:43:40<40:44,  3.31s/it]
 95%|█████████▍| 13988/14725 [12:43:43<39:42,  3.23s/it]
 95%|█████████▌| 13989/14725 [12:43:47<39:35,  3.23s/it]
 95%|█████████▌| 13990/14725 [12:43:50<39:18,  3.21s/it]
                                            
0: {'loss': 0.6809, 'grad_norm': 0.4025331594733152, 'learning_rate': 1e-05, 'epoch': 0.95}
0: {'loss': 0.6664, 'grad_norm': 0.4021898090685326, 'learning_rate': 1e-05, 'epoch': 0.95}
0:             

 95%|█████████▌| 13990/14725 [12:43:50<39:18,  3.21s/it]
 95%|█████████▌| 13991/14725 [12:43:53<39:39,  3.24s/it]
 95%|█████████▌| 13992/14725 [12:43:57<40:08,  3.29s/it]
 95%|█████████▌| 13993/14725 [12:44:00<39:40,  3.25s/it]
 95%|█████████▌| 13994/14725 [12:44:03<39:04,  3.21s/it]
 95%|█████████▌| 13995/14725 [12:44:06<38:23,  3.16s/it]
 95%|█████████▌| 13996/14725 [12:44:09<38:12,  3.14s/it]
 95%|█████████▌| 13997/14725 [12:44:12<37:59,  3.13s/it]
 95%|█████████▌| 13998/14725 [12:44:15<37:55,  3.13s/it]
 95%|█████████▌| 13999/14725 [12:44:18<38:10,  3.15s/it]
 95%|█████████▌| 14000/14725 [12:44:22<38:47,  3.21s/it]
                                                        

 95%|█████████▌| 14000/14725 [12:44:22<38:47,  3.21s/it]
 95%|███████�
0: {'loss': 0.6733, 'grad_norm': 0.4254837985054416, 'learning_rate': 1e-05, 'epoch': 0.95}
0: �█▌| 14001/14725 [12:44:25<39:02,  3.24s/it]
 95%|█████████▌| 14002/14725 [12:44:29<39:46,  3.30s/it]
 95%|█████████▌| 14003/14725 [12:44:32<40:56,  3.40s/it]
 95%|█████████▌| 14004/14725 [12:44:35<39:35,  3.30s/it]
 95%|█████████▌| 14005/14725 [12:44:39<40:00,  3.33s/it]
 95%|█████████▌| 14006/14725 [12:44:42<39:00,  3.25s/it]
 95%|█████████▌| 14007/14725 [12:44:45<39:09,  3.27s/it]
 95%|█████████▌| 14008/14725 [12:44:48<38:36,  3.23s/it]
 95%|█████████▌| 14009/14725 [12:44:51<38:38,  3.24s/it]
 95%|█████████▌| 14010/14725 [12:44:55<38:29,  3.23s/it]
                                                        

 95%|█████████▌| 14010/14725 [12:44:55<38:29,  3.23s/it]
 95%|█████████▌| 14011/14725 [12:44:58<37:50,  3.18s/it]
 95%|█████████▌| 14012/14725 [12:45:01<38:30,  3.2
0: {'loss': 0.6661, 'grad_norm': 0.3731358568827908, 'learning_rate': 1e-05, 'epoch': 0.95}
0: 4s/it]
 95%|█████████▌| 14013/14725 [12:45:04<37:44,  3.18s/it]
 95%|█████████▌| 14014/14725 [12:45:08<38:39,  3.26s/it]
 95%|█████████▌| 14015/14725 [12:45:11<38:38,  3.26s/it]
 95%|█████████▌| 14016/14725 [12:45:14<38:41,  3.27s/it]
 95%|█████████▌| 14017/14725 [12:45:17<38:42,  3.28s/it]
 95%|█████████▌| 14018/14725 [12:45:21<38:17,  3.25s/it]
 95%|█████████▌| 14019/14725 [12:45:24<38:46,  3.30s/it]
 95%|█████████▌| 14020/14725 [12:45:27<37:57,  3.23s/it]
                                                        

 95%|█████████▌| 14020/14725 [12:45:27<37:57,  3.23s/it]
 95%|█████████▌| 14021/14725 [12:45:30<37:34,  3.20s/it]
 95%|█████████▌| 14022/14725 [12:45:33<37:25,  3.19s/it]
 95%|█████████▌| 14023/14725 [12:45:37<37:23,  3.20s/it]
 95%|█████████▌
0: {'loss': 0.6575, 'grad_norm': 0.3973160548287191, 'learning_rate': 1e-05, 'epoch': 0.95}
0: | 14024/14725 [12:45:40<37:29,  3.21s/it]
 95%|█████████▌| 14025/14725 [12:45:43<37:44,  3.23s/it]
 95%|█████████▌| 14026/14725 [12:45:46<37:33,  3.22s/it]
 95%|█████████▌| 14027/14725 [12:45:49<37:05,  3.19s/it]
 95%|█████████▌| 14028/14725 [12:45:53<37:43,  3.25s/it]
 95%|█████████▌| 14029/14725 [12:45:56<37:20,  3.22s/it]
 95%|█████████▌| 14030/14725 [12:45:59<37:08,  3.21s/it]
                                                        

 95%|█████████▌| 14030/14725 [12:45:59<37:08,  3.21s/it]
 95%|█████████▌| 14031/14725 [12:46:02<37:16,  3.22s/it]
 95%|█████████▌| 14032/14725 [12:46:06<37:53,  3.28s/it]
 95%|█████████▌| 14033/14725 [12:46:09<37:56,  3.29s/it]
 95%|█████████▌| 14034/14725 [12:46:12<37:47,  3.28s/it]
 95%|█████████▌| 14035/14725 [12:46:16<37:36,  3.27s/it]
0: {'loss': 0.6915, 'grad_norm': 0.3939675927366901, 'learning_rate': 1e-05, 'epoch': 0.95}
0:  95%|█████████▌| 14036/14725 [12:46:19<37:36,  3.28s/it]
 95%|█████████▌| 14037/14725 [12:46:22<37:53,  3.31s/it]
 95%|█████████▌| 14038/14725 [12:46:26<38:22,  3.35s/it]
 95%|█████████▌| 14039/14725 [12:46:29<37:37,  3.29s/it]
 95%|█████████▌| 14040/14725 [12:46:32<37:03,  3.25s/it]
                                                        

 95%|█████████▌| 14040/14725 [12:46:32<37:03,  3.25s/it]
 95%|█████████▌| 14041/14725 [12:46:36<37:41,  3.31s/it]
 95%|█████████▌| 14042/14725 [12:46:39<38:17,  3.36s/it]
 95%|█████████▌| 14043/14725 [12:46:42<38:18,  3.37s/it]
 95%|█████████▌| 14044/14725 [12:46:46<37:40,  3.32s/it]
 95%|█████████▌| 14045/14725 [12:46:49<37:47,  3.33s/it]
 95%|█████████▌| 14046/14725 [12:46:52<37:03,  3.27s/it]
 95%|█████████▌| 14047
0: {'loss': 0.6767, 'grad_norm': 0.3992302994859378, 'learning_rate': 1e-05, 'epoch': 0.95}
0: /14725 [12:46:55<36:58,  3.27s/it]
 95%|█████████▌| 14048/14725 [12:46:59<36:56,  3.27s/it]
 95%|█████████▌| 14049/14725 [12:47:02<37:21,  3.32s/it]
 95%|█████████▌| 14050/14725 [12:47:05<36:48,  3.27s/it]
                                                        

 95%|█████████▌| 14050/14725 [12:47:05<36:48,  3.27s/it]
 95%|█████████▌| 14051/14725 [12:47:09<36:49,  3.28s/it]
 95%|█████████▌| 14052/14725 [12:47:12<37:14,  3.32s/it]
 95%|█████████▌| 14053/14725 [12:47:15<37:43,  3.37s/it]
 95%|█████████▌| 14054/14725 [12:47:19<37:41,  3.37s/it]
 95%|█████████▌| 14055/14725 [12:47:22<36:56,  3.31s/it]
 95%|█████████▌| 14056/14725 [12:47:25<36:21,  3.26s/it]
 95%|█████████▌| 14057/14725 [12:47:28<36:07,  3.24s/it]
 95%|█████████▌| 14058/14725 [12:47:31<35:41,  3.21s/it]
 95%|�
0: {'loss': 0.6589, 'grad_norm': 0.40726730252080273, 'learning_rate': 1e-05, 'epoch': 0.95}
0: �████████▌| 14059/14725 [12:47:35<35:08,  3.17s/it]
 95%|█████████▌| 14060/14725 [12:47:38<34:59,  3.16s/it]
                                                        

 95%|█████████▌| 14060/14725 [12:47:38<34:59,  3.16s/it]
 95%|█████████▌| 14061/14725 [12:47:41<35:23,  3.20s/it]
 95%|█████████▌| 14062/14725 [12:47:44<35:36,  3.22s/it]
 96%|█████████▌| 14063/14725 [12:47:47<35:34,  3.22s/it]
 96%|█████████▌| 14064/14725 [12:47:51<35:41,  3.24s/it]
 96%|█████████▌| 14065/14725 [12:47:54<35:50,  3.26s/it]
 96%|█████████▌| 14066/14725 [12:47:57<35:20,  3.22s/it]
 96%|█████████▌| 14067/14725 [12:48:00<35:26,  3.23s/it]
 96%|█████████▌| 14068/14725 [12:48:04<35:31,  3.24s/it]
 96%|█████████▌| 14069/14725 [12:48:07<36:28,  3.34s/it]
 96%|█████████▌| 14070/14725 
0: {'loss': 0.678, 'grad_norm': 0.3805324938357124, 'learning_rate': 1e-05, 'epoch': 0.96}
0: {'loss': 0.6813, 'grad_norm': 0.40748967922864365, 'learning_rate': 1e-05, 'epoch': 0.96}
0: [12:48:10<35:50,  3.28s/it]
                                                        

 96%|█████████▌| 14070/14725 [12:48:10<35:50,  3.28s/it]
 96%|█████████▌| 14071/14725 [12:48:14<35:40,  3.27s/it]
 96%|█████████▌| 14072/14725 [12:48:17<35:28,  3.26s/it]
 96%|█████████▌| 14073/14725 [12:48:20<35:21,  3.25s/it]
 96%|█████████▌| 14074/14725 [12:48:23<35:05,  3.23s/it]
 96%|█████████▌| 14075/14725 [12:48:27<35:29,  3.28s/it]
 96%|█████████▌| 14076/14725 [12:48:30<36:05,  3.34s/it]
 96%|█████████▌| 14077/14725 [12:48:34<36:42,  3.40s/it]
 96%|█████████▌| 14078/14725 [12:48:37<36:50,  3.42s/it]
 96%|█████████▌| 14079/14725 [12:48:40<35:59,  3.34s/it]
 96%|█████████▌| 14080/14725 [12:48:44<36:05,  3.36s/it]
                                                        

 96%|█████████�
0: {'loss': 0.6665, 'grad_norm': 0.4519653855972518, 'learning_rate': 1e-05, 'epoch': 0.96}
0: ��| 14080/14725 [12:48:44<36:05,  3.36s/it]
 96%|█████████▌| 14081/14725 [12:48:47<35:28,  3.31s/it]
 96%|█████████▌| 14082/14725 [12:48:50<35:32,  3.32s/it]
 96%|█████████▌| 14083/14725 [12:48:53<34:58,  3.27s/it]
 96%|█████████▌| 14084/14725 [12:48:57<35:28,  3.32s/it]
 96%|█████████▌| 14085/14725 [12:49:00<35:16,  3.31s/it]
 96%|█████████▌| 14086/14725 [12:49:03<34:34,  3.25s/it]
 96%|█████████▌| 14087/14725 [12:49:06<34:27,  3.24s/it]
 96%|█████████▌| 14088/14725 [12:49:10<34:23,  3.24s/it]
 96%|█████████▌| 14089/14725 [12:49:13<34:29,  3.25s/it]
 96%|█████████▌| 14090/14725 [12:49:16<34:56,  3.30s/it]
                                                        

 96%|█████████▌| 14090/14725 [12:49:16<34:56,  3.30s/it]
 96%|█████████▌| 14091/14725 [12:49:20<34:37,  3.28s/it
0: {'loss': 0.6876, 'grad_norm': 0.42729917168161646, 'learning_rate': 1e-05, 'epoch': 0.96}
0: ]
 96%|█████████▌| 14092/14725 [12:49:23<34:16,  3.25s/it]
 96%|█████████▌| 14093/14725 [12:49:26<33:38,  3.19s/it]
 96%|█████████▌| 14094/14725 [12:49:29<33:34,  3.19s/it]
 96%|█████████▌| 14095/14725 [12:49:32<33:27,  3.19s/it]
 96%|█████████▌| 14096/14725 [12:49:36<34:13,  3.26s/it]
 96%|█████████▌| 14097/14725 [12:49:39<34:29,  3.30s/it]
 96%|█████████▌| 14098/14725 [12:49:42<34:28,  3.30s/it]
 96%|█████████▌| 14099/14725 [12:49:46<34:39,  3.32s/it]
 96%|█████████▌| 14100/14725 [12:49:49<34:14,  3.29s/it]
                                                        

 96%|█████████▌| 14100/14725 [12:49:49<34:14,  3.29s/it]
 96%|█████████▌| 14101/14725 [12:49:52<34:28,  3.31s/it]
 96%|█████████▌| 14102/14725 [12:49:56<34:37,  3.34s/it]
 96%|█████████▌| 141
0: {'loss': 0.6721, 'grad_norm': 0.38759470608322566, 'learning_rate': 1e-05, 'epoch': 0.96}
0: 03/14725 [12:49:59<34:10,  3.30s/it]
 96%|█████████▌| 14104/14725 [12:50:02<33:44,  3.26s/it]
 96%|█████████▌| 14105/14725 [12:50:05<33:17,  3.22s/it]
 96%|█████████▌| 14106/14725 [12:50:08<33:10,  3.22s/it]
 96%|█████████▌| 14107/14725 [12:50:12<32:52,  3.19s/it]
 96%|█████████▌| 14108/14725 [12:50:15<32:23,  3.15s/it]
 96%|█████████▌| 14109/14725 [12:50:18<32:17,  3.15s/it]
 96%|█████████▌| 14110/14725 [12:50:21<32:24,  3.16s/it]
                                                        

 96%|█████████▌| 14110/14725 [12:50:21<32:24,  3.16s/it]
 96%|█████████▌| 14111/14725 [12:50:24<32:03,  3.13s/it]
 96%|█████████▌| 14112/14725 [12:50:27<32:19,  3.16s/it]
 96%|█████████▌| 14113/14725 [12:50:30<32:18,  3.17s/it]
 96%|█████████▌| 14114/14725 [12:50:34<32:48,  3.22s/it]
 96%|
0: {'loss': 0.6862, 'grad_norm': 0.41055577231988766, 'learning_rate': 1e-05, 'epoch': 0.96}
0: █████████▌| 14115/14725 [12:50:37<33:07,  3.26s/it]
 96%|█████████▌| 14116/14725 [12:50:40<33:23,  3.29s/it]
 96%|█████████▌| 14117/14725 [12:50:44<33:54,  3.35s/it]
 96%|█████████▌| 14118/14725 [12:50:47<33:27,  3.31s/it]
 96%|█████████▌| 14119/14725 [12:50:51<33:44,  3.34s/it]
 96%|█████████▌| 14120/14725 [12:50:54<33:16,  3.30s/it]
                                                        

 96%|█████████▌| 14120/14725 [12:50:54<33:16,  3.30s/it]
 96%|█████████▌| 14121/14725 [12:50:57<33:21,  3.31s/it]
 96%|█████████▌| 14122/14725 [12:51:00<33:18,  3.31s/it]
 96%|█████████▌| 14123/14725 [12:51:04<32:43,  3.26s/it]
 96%|█████████▌| 14124/14725 [12:51:07<33:19,  3.33s/it]
 96%|█████████▌| 14125/14725 [12:51:10<33:25,  3.34s/it]
 96%|█████████▌| 14126/1472
0: {'loss': 0.6694, 'grad_norm': 0.43979692950591465, 'learning_rate': 1e-05, 'epoch': 0.96}
0: 5 [12:51:14<33:12,  3.33s/it]
 96%|█████████▌| 14127/14725 [12:51:17<33:05,  3.32s/it]
 96%|█████████▌| 14128/14725 [12:51:20<32:22,  3.25s/it]
 96%|█████████▌| 14129/14725 [12:51:23<32:32,  3.28s/it]
 96%|█████████▌| 14130/14725 [12:51:27<32:40,  3.30s/it]
                                                        

 96%|█████████▌| 14130/14725 [12:51:27<32:40,  3.30s/it]
 96%|█████████▌| 14131/14725 [12:51:30<32:35,  3.29s/it]
 96%|█████████▌| 14132/14725 [12:51:33<32:44,  3.31s/it]
 96%|█████████▌| 14133/14725 [12:51:37<32:32,  3.30s/it]
 96%|█████████▌| 14134/14725 [12:51:40<33:07,  3.36s/it]
 96%|█████████▌| 14135/14725 [12:51:43<32:17,  3.28s/it]
 96%|█████████▌| 14136/14725 [12:51:47<32:42,  3.33s/it]
 96%|█████████▌| 14137/14725 [12:51:50<33:01,  3.37s/it]
 96%|██�
0: {'loss': 0.6909, 'grad_norm': 0.42304283256772496, 'learning_rate': 1e-05, 'epoch': 0.96}
0: ��██████▌| 14138/14725 [12:51:54<33:10,  3.39s/it]
 96%|█████████▌| 14139/14725 [12:51:57<32:35,  3.34s/it]
 96%|█████████▌| 14140/14725 [12:52:00<32:33,  3.34s/it]
                                                        

 96%|█████████▌| 14140/14725 [12:52:00<32:33,  3.34s/it]
 96%|█████████▌| 14141/14725 [12:52:03<32:05,  3.30s/it]
 96%|█████████▌| 14142/14725 [12:52:07<31:31,  3.24s/it]
 96%|█████████▌| 14143/14725 [12:52:10<31:21,  3.23s/it]
 96%|█████████▌| 14144/14725 [12:52:13<31:12,  3.22s/it]
 96%|█████████▌| 14145/14725 [12:52:16<31:11,  3.23s/it]
 96%|█████████▌| 14146/14725 [12:52:19<30:50,  3.20s/it]
 96%|█████████▌| 14147/14725 [12:52:23<31:04,  3.23s/it]
 96%|█████████▌| 14148/14725 [12:52:26<31:27,  3.27s/it]
 96%|█████████▌| 14149/14725 [12:5
0: {'loss': 0.6747, 'grad_norm': 0.39165645181040454, 'learning_rate': 1e-05, 'epoch': 0.96}
0: 2:29<31:11,  3.25s/it]
 96%|█████████▌| 14150/14725 [12:52:32<30:51,  3.22s/it]
                                                        

 96%|█████████▌| 14150/14725 [12:52:32<30:51,  3.22s/it]
 96%|█████████▌| 14151/14725 [12:52:36<31:03,  3.25s/it]
 96%|█████████▌| 14152/14725 [12:52:39<30:57,  3.24s/it]
 96%|█████████▌| 14153/14725 [12:52:42<30:53,  3.24s/it]
 96%|█████████▌| 14154/14725 [12:52:45<30:30,  3.20s/it]
 96%|█████████▌| 14155/14725 [12:52:48<30:35,  3.22s/it]
 96%|█████████▌| 14156/14725 [12:52:52<31:16,  3.30s/it]
 96%|█████████▌| 14157/14725 [12:52:55<31:08,  3.29s/it]
 96%|█████████▌| 14158/14725 [12:52:58<30:44,  3.25s/it]
 96%|█████████▌| 14159/14725 [12:53:02<30:36,  3.24s/it]
 96%|█████████▌| 14160/14725 [12:53:05<30:45,  3.27s/it]
                   
0: {'loss': 0.6687, 'grad_norm': 0.4549804369730801, 'learning_rate': 1e-05, 'epoch': 0.96}
0: {'loss': 0.6769, 'grad_norm': 0.4100686582360027, 'learning_rate': 1e-05, 'epoch': 0.96}
0:                                      

 96%|█████████▌| 14160/14725 [12:53:05<30:45,  3.27s/it]
 96%|█████████▌| 14161/14725 [12:53:08<30:39,  3.26s/it]
 96%|█████████▌| 14162/14725 [12:53:12<30:51,  3.29s/it]
 96%|█████████▌| 14163/14725 [12:53:15<31:10,  3.33s/it]
 96%|█████████▌| 14164/14725 [12:53:18<31:06,  3.33s/it]
 96%|█████████▌| 14165/14725 [12:53:22<30:46,  3.30s/it]
 96%|█████████▌| 14166/14725 [12:53:25<30:30,  3.28s/it]
 96%|█████████▌| 14167/14725 [12:53:28<30:25,  3.27s/it]
 96%|█████████▌| 14168/14725 [12:53:31<30:07,  3.25s/it]
 96%|█████████▌| 14169/14725 [12:53:34<30:08,  3.25s/it]
 96%|█████████▌| 14170/14725 [12:53:38<30:16,  3.27s/it]
                                                        

 96%|█████████▌| 14170/14725 [12:53:38<30:16,  3.27s/it]
 96
0: {'loss': 0.6625, 'grad_norm': 0.39195493716613883, 'learning_rate': 1e-05, 'epoch': 0.96}
0: %|█████████▌| 14171/14725 [12:53:41<30:29,  3.30s/it]
 96%|█████████▌| 14172/14725 [12:53:44<30:29,  3.31s/it]
 96%|█████████▋| 14173/14725 [12:53:48<30:24,  3.31s/it]
 96%|█████████▋| 14174/14725 [12:53:51<29:56,  3.26s/it]
 96%|█████████▋| 14175/14725 [12:53:54<30:16,  3.30s/it]
 96%|█████████▋| 14176/14725 [12:53:58<30:05,  3.29s/it]
 96%|█████████▋| 14177/14725 [12:54:01<30:02,  3.29s/it]
 96%|█████████▋| 14178/14725 [12:54:04<29:23,  3.22s/it]
 96%|█████████▋| 14179/14725 [12:54:07<29:06,  3.20s/it]
 96%|█████████▋| 14180/14725 [12:54:10<28:34,  3.15s/it]
                                                        

 96%|█████████▋| 14180/14725 [12:54:10<28:34,  3.15s/it]
 96%|█████████▋| 14181/14725 [12:54:14<29:21,  3.24s/it]
 96%|█████████▋| 14182/14
0: {'loss': 0.6688, 'grad_norm': 0.3959380355703324, 'learning_rate': 1e-05, 'epoch': 0.96}
0: 725 [12:54:17<28:50,  3.19s/it]
 96%|█████████▋| 14183/14725 [12:54:20<28:39,  3.17s/it]
 96%|█████████▋| 14184/14725 [12:54:23<28:27,  3.16s/it]
 96%|█████████▋| 14185/14725 [12:54:26<28:22,  3.15s/it]
 96%|█████████▋| 14186/14725 [12:54:29<28:48,  3.21s/it]
 96%|█████████▋| 14187/14725 [12:54:33<29:15,  3.26s/it]
 96%|█████████▋| 14188/14725 [12:54:36<29:10,  3.26s/it]
 96%|█████████▋| 14189/14725 [12:54:39<29:36,  3.31s/it]
 96%|█████████▋| 14190/14725 [12:54:43<29:09,  3.27s/it]
                                                        

 96%|█████████▋| 14190/14725 [12:54:43<29:09,  3.27s/it]
 96%|█████████▋| 14191/14725 [12:54:46<28:35,  3.21s/it]
 96%|█████████▋| 14192/14725 [12:54:49<28:03,  3.16s/it]
 96%|█████████▋| 14193/14725 [12:54:52<28:13,  3.18s/it]
 96%|█�
0: {'loss': 0.6558, 'grad_norm': 0.38791846181099215, 'learning_rate': 1e-05, 'epoch': 0.96}
0: �███████▋| 14194/14725 [12:54:55<27:53,  3.15s/it]
 96%|█████████▋| 14195/14725 [12:54:58<28:31,  3.23s/it]
 96%|█████████▋| 14196/14725 [12:55:02<29:24,  3.34s/it]
 96%|█████████▋| 14197/14725 [12:55:05<29:30,  3.35s/it]
 96%|█████████▋| 14198/14725 [12:55:09<29:11,  3.32s/it]
 96%|█████████▋| 14199/14725 [12:55:12<29:16,  3.34s/it]
 96%|█████████▋| 14200/14725 [12:55:15<29:27,  3.37s/it]
                                                        

 96%|█████████▋| 14200/14725 [12:55:15<29:27,  3.37s/it]
 96%|█████████▋| 14201/14725 [12:55:19<29:04,  3.33s/it]
 96%|█████████▋| 14202/14725 [12:55:22<28:45,  3.30s/it]
 96%|█████████▋| 14203/14725 [12:55:25<28:33,  3.28s/it]
 96%|█████████▋| 14204/14725 [12:55:29<28:49,  3.32s/it]
 96%|█████████▋| 14205/14725 [12
0: {'loss': 0.6786, 'grad_norm': 0.40912309650937523, 'learning_rate': 1e-05, 'epoch': 0.96}
0: :55:32<28:17,  3.26s/it]
 96%|█████████▋| 14206/14725 [12:55:35<28:16,  3.27s/it]
 96%|█████████▋| 14207/14725 [12:55:38<28:11,  3.27s/it]
 96%|█████████▋| 14208/14725 [12:55:42<28:48,  3.34s/it]
 96%|█████████▋| 14209/14725 [12:55:45<28:26,  3.31s/it]
 97%|█████████▋| 14210/14725 [12:55:48<28:04,  3.27s/it]
                                                        

 97%|█████████▋| 14210/14725 [12:55:48<28:04,  3.27s/it]
 97%|█████████▋| 14211/14725 [12:55:51<27:54,  3.26s/it]
 97%|█████████▋| 14212/14725 [12:55:55<27:53,  3.26s/it]
 97%|█████████▋| 14213/14725 [12:55:58<27:46,  3.25s/it]
 97%|█████████▋| 14214/14725 [12:56:01<27:26,  3.22s/it]
 97%|█████████▋| 14215/14725 [12:56:05<28:06,  3.31s/it]
 97%|█████████▋| 14216/14725 [12:56:08<28:00,  3.30s/it]
 97%|████
0: {'loss': 0.6839, 'grad_norm': 0.4452875639214236, 'learning_rate': 1e-05, 'epoch': 0.97}
0: █████▋| 14217/14725 [12:56:11<27:35,  3.26s/it]
 97%|█████████▋| 14218/14725 [12:56:14<27:11,  3.22s/it]
 97%|█████████▋| 14219/14725 [12:56:18<27:48,  3.30s/it]
 97%|█████████▋| 14220/14725 [12:56:21<27:47,  3.30s/it]
                                                        

 97%|█████████▋| 14220/14725 [12:56:21<27:47,  3.30s/it]
 97%|█████████▋| 14221/14725 [12:56:24<27:30,  3.27s/it]
 97%|█████████▋| 14222/14725 [12:56:27<27:33,  3.29s/it]
 97%|█████████▋| 14223/14725 [12:56:31<27:27,  3.28s/it]
 97%|█████████▋| 14224/14725 [12:56:34<27:01,  3.24s/it]
 97%|█████████▋| 14225/14725 [12:56:37<26:58,  3.24s/it]
 97%|█████████▋| 14226/14725 [12:56:40<26:45,  3.22s/it]
 97%|█████████▋| 14227/14725 [12:56:44<27:02,  3.26s/it]
 97%|█████████▋| 14228/14725 [12:56:47<
0: {'loss': 0.6783, 'grad_norm': 0.39425779564220986, 'learning_rate': 1e-05, 'epoch': 0.97}
0: 27:08,  3.28s/it]
 97%|█████████▋| 14229/14725 [12:56:50<27:21,  3.31s/it]
 97%|█████████▋| 14230/14725 [12:56:54<27:14,  3.30s/it]
                                                        

 97%|█████████▋| 14230/14725 [12:56:54<27:14,  3.30s/it]
 97%|█████████▋| 14231/14725 [12:56:57<26:36,  3.23s/it]
 97%|█████████▋| 14232/14725 [12:57:00<26:33,  3.23s/it]
 97%|█████████▋| 14233/14725 [12:57:03<27:13,  3.32s/it]
 97%|█████████▋| 14234/14725 [12:57:07<27:07,  3.31s/it]
 97%|█████████▋| 14235/14725 [12:57:10<27:24,  3.36s/it]
 97%|█████████▋| 14236/14725 [12:57:13<26:57,  3.31s/it]
 97%|█████████▋| 14237/14725 [12:57:17<26:44,  3.29s/it]
 97%|█████████▋| 14238/14725 [12:57:20<26:31,  3.27s/it]
 97%|█████████▋| 14239/14725 [12:57:23<26:19,  3.25s/it]
 97%|██████�
0: {'loss': 0.6643, 'grad_norm': 0.39499257737764437, 'learning_rate': 1e-05, 'epoch': 0.97}
0: {'loss': 0.6597, 'grad_norm': 0.4147966329047065, 'learning_rate': 1e-05, 'epoch': 0.97}
0: ��██▋| 14240/14725 [12:57:26<26:25,  3.27s/it]
                                                        

 97%|█████████▋| 14240/14725 [12:57:26<26:25,  3.27s/it]
 97%|█████████▋| 14241/14725 [12:57:30<26:36,  3.30s/it]
 97%|█████████▋| 14242/14725 [12:57:33<26:20,  3.27s/it]
 97%|█████████▋| 14243/14725 [12:57:36<26:33,  3.30s/it]
 97%|█████████▋| 14244/14725 [12:57:40<26:29,  3.30s/it]
 97%|█████████▋| 14245/14725 [12:57:43<26:03,  3.26s/it]
 97%|█████████▋| 14246/14725 [12:57:46<25:57,  3.25s/it]
 97%|█████████▋| 14247/14725 [12:57:49<26:08,  3.28s/it]
 97%|█████████▋| 14248/14725 [12:57:53<26:00,  3.27s/it]
 97%|█████████▋| 14249/14725 [12:57:56<25:50,  3.26s/it]
 97%|█████████▋| 14250/14725 [12:57:59<26:33,  3.35s/it]
                                                        

 97%|█
0: {'loss': 0.6858, 'grad_norm': 0.40071969333103535, 'learning_rate': 1e-05, 'epoch': 0.97}
0: ████████▋| 14250/14725 [12:57:59<26:33,  3.35s/it]
 97%|█████████▋| 14251/14725 [12:58:03<26:30,  3.36s/it]
 97%|█████████▋| 14252/14725 [12:58:06<26:51,  3.41s/it]
 97%|█████████▋| 14253/14725 [12:58:10<26:37,  3.38s/it]
 97%|█████████▋| 14254/14725 [12:58:13<26:46,  3.41s/it]
 97%|█████████▋| 14255/14725 [12:58:16<26:11,  3.34s/it]
 97%|█████████▋| 14256/14725 [12:58:20<26:24,  3.38s/it]
 97%|█████████▋| 14257/14725 [12:58:23<25:40,  3.29s/it]
 97%|█████████▋| 14258/14725 [12:58:26<25:44,  3.31s/it]
 97%|█████████▋| 14259/14725 [12:58:30<26:11,  3.37s/it]
 97%|█████████▋| 14260/14725 [12:58:33<25:32,  3.29s/it]
                                                        

 97%|█████████▋| 14260/14725 [12:58:33<25:32,  3.29s/it]
 97%|█████████▋| 14261/14725 [
0: {'loss': 0.653, 'grad_norm': 0.35712003868072756, 'learning_rate': 1e-05, 'epoch': 0.97}
0: 12:58:36<25:10,  3.26s/it]
 97%|█████████▋| 14262/14725 [12:58:39<24:48,  3.22s/it]
 97%|█████████▋| 14263/14725 [12:58:42<24:38,  3.20s/it]
 97%|█████████▋| 14264/14725 [12:58:46<24:34,  3.20s/it]
 97%|█████████▋| 14265/14725 [12:58:49<24:41,  3.22s/it]
 97%|█████████▋| 14266/14725 [12:58:52<24:40,  3.23s/it]
 97%|█████████▋| 14267/14725 [12:58:55<25:01,  3.28s/it]
 97%|█████████▋| 14268/14725 [12:58:59<25:00,  3.28s/it]
 97%|█████████▋| 14269/14725 [12:59:02<24:36,  3.24s/it]
 97%|█████████▋| 14270/14725 [12:59:05<24:37,  3.25s/it]
                                                        

 97%|█████████▋| 14270/14725 [12:59:05<24:37,  3.25s/it]
 97%|█████████▋| 14271/14725 [12:59:08<24:52,  3.29s/it]
 97%|█████████▋| 14272/14725 [12:59:12<24:14,  3.21s/it]
 97%|███�
0: {'loss': 0.6712, 'grad_norm': 0.37908219260403153, 'learning_rate': 1e-05, 'epoch': 0.97}
0: ��█████▋| 14273/14725 [12:59:15<24:02,  3.19s/it]
 97%|█████████▋| 14274/14725 [12:59:18<23:49,  3.17s/it]
 97%|█████████▋| 14275/14725 [12:59:21<23:45,  3.17s/it]
 97%|█████████▋| 14276/14725 [12:59:24<23:38,  3.16s/it]
 97%|█████████▋| 14277/14725 [12:59:27<23:51,  3.19s/it]
 97%|█████████▋| 14278/14725 [12:59:31<24:06,  3.24s/it]
 97%|█████████▋| 14279/14725 [12:59:34<24:20,  3.28s/it]
 97%|█████████▋| 14280/14725 [12:59:37<24:23,  3.29s/it]
                                                        

 97%|█████████▋| 14280/14725 [12:59:37<24:23,  3.29s/it]
 97%|█████████▋| 14281/14725 [12:59:40<23:52,  3.23s/it]
 97%|█████████▋| 14282/14725 [12:59:44<23:47,  3.22s/it]
 97%|█████████▋| 14283/14725 [12:59:47<23:42,  3.22s/it]
 97%|█████████▋| 14284/14725 [12:59:5
0: {'loss': 0.667, 'grad_norm': 0.3832239753023329, 'learning_rate': 1e-05, 'epoch': 0.97}
0: 0<23:59,  3.27s/it]
 97%|█████████▋| 14285/14725 [12:59:54<24:00,  3.27s/it]
 97%|█████████▋| 14286/14725 [12:59:57<24:00,  3.28s/it]
 97%|█████████▋| 14287/14725 [13:00:00<23:59,  3.29s/it]
 97%|█████████▋| 14288/14725 [13:00:03<23:46,  3.26s/it]
 97%|█████████▋| 14289/14725 [13:00:07<23:41,  3.26s/it]
 97%|█████████▋| 14290/14725 [13:00:10<23:28,  3.24s/it]
                                                        

 97%|█████████▋| 14290/14725 [13:00:10<23:28,  3.24s/it]
 97%|█████████▋| 14291/14725 [13:00:13<23:28,  3.25s/it]
 97%|█████████▋| 14292/14725 [13:00:17<23:51,  3.31s/it]
 97%|█████████▋| 14293/14725 [13:00:20<23:38,  3.28s/it]
 97%|█████████▋| 14294/14725 [13:00:23<23:25,  3.26s/it]
 97%|█████████▋| 14295/14725 [13:00:26<23:32,  3.28s/it]
 97%|█████�
0: {'loss': 0.6707, 'grad_norm': 0.384873681895216, 'learning_rate': 1e-05, 'epoch': 0.97}
0: �███▋| 14296/14725 [13:00:30<23:27,  3.28s/it]
 97%|█████████▋| 14297/14725 [13:00:33<23:14,  3.26s/it]
 97%|█████████▋| 14298/14725 [13:00:36<23:12,  3.26s/it]
 97%|█████████▋| 14299/14725 [13:00:39<23:03,  3.25s/it]
 97%|█████████▋| 14300/14725 [13:00:43<23:07,  3.26s/it]
                                                        

 97%|█████████▋| 14300/14725 [13:00:43<23:07,  3.26s/it]
 97%|█████████▋| 14301/14725 [13:00:46<23:01,  3.26s/it]
 97%|█████████▋| 14302/14725 [13:00:49<22:48,  3.23s/it]
 97%|█████████▋| 14303/14725 [13:00:52<22:47,  3.24s/it]
 97%|█████████▋| 14304/14725 [13:00:55<22:40,  3.23s/it]
 97%|█████████▋| 14305/14725 [13:00:59<22:21,  3.20s/it]
 97%|█████████▋| 14306/14725 [13:01:02<22:17,  3.19s/it]
 97%|█████████▋| 14307/14725 [13:01:05<22:05
0: {'loss': 0.6737, 'grad_norm': 0.42372193378194567, 'learning_rate': 1e-05, 'epoch': 0.97}
0: ,  3.17s/it]
 97%|█████████▋| 14308/14725 [13:01:08<22:28,  3.23s/it]
 97%|█████████▋| 14309/14725 [13:01:12<22:31,  3.25s/it]
 97%|█████████▋| 14310/14725 [13:01:15<22:54,  3.31s/it]
                                                        

 97%|█████████▋| 14310/14725 [13:01:15<22:54,  3.31s/it]
 97%|█████████▋| 14311/14725 [13:01:18<22:57,  3.33s/it]
 97%|█████████▋| 14312/14725 [13:01:21<22:25,  3.26s/it]
 97%|█████████▋| 14313/14725 [13:01:25<22:17,  3.25s/it]
 97%|█████████▋| 14314/14725 [13:01:28<21:56,  3.20s/it]
 97%|█████████▋| 14315/14725 [13:01:31<21:54,  3.21s/it]
 97%|█████████▋| 14316/14725 [13:01:34<22:02,  3.23s/it]
 97%|█████████▋| 14317/14725 [13:01:37<21:38,  3.18s/it]
 97%|█████████▋| 14318/14725 [13:01:41<21:56,  3.24s/it]
 97%|████████
0: {'loss': 0.6513, 'grad_norm': 0.39402658725613515, 'learning_rate': 1e-05, 'epoch': 0.97}
0: █▋| 14319/14725 [13:01:44<21:52,  3.23s/it]
 97%|█████████▋| 14320/14725 [13:01:47<21:49,  3.23s/it]
                                                        

 97%|█████████▋| 14320/14725 [13:01:47<21:49,  3.23s/it]
 97%|█████████▋| 14321/14725 [13:01:50<21:22,  3.17s/it]
 97%|█████████▋| 14322/14725 [13:01:53<21:24,  3.19s/it]
 97%|█████████▋| 14323/14725 [13:01:57<21:33,  3.22s/it]
 97%|█████████▋| 14324/14725 [13:02:00<21:38,  3.24s/it]
 97%|█████████▋| 14325/14725 [13:02:03<21:18,  3.20s/it]
 97%|█████████▋| 14326/14725 [13:02:06<21:18,  3.20s/it]
 97%|█████████▋| 14327/14725 [13:02:10<21:17,  3.21s/it]
 97%|█████████▋| 14328/14725 [13:02:13<21:43,  3.28s/it]
 97%|█████████▋| 14329/14725 [13:02:16<21:57,  3.33s/it]
 97%|█████████▋| 14330/14725 [13:02:19<21:19,  3.24
0: {'loss': 0.6724, 'grad_norm': 0.3909769261110469, 'learning_rate': 1e-05, 'epoch': 0.97}
0: {'loss': 0.6765, 'grad_norm': 0.38407299527539324, 'learning_rate': 1e-05, 'epoch': 0.97}
0: s/it]
                                                        

 97%|█████████▋| 14330/14725 [13:02:19<21:19,  3.24s/it]
 97%|█████████▋| 14331/14725 [13:02:23<20:58,  3.20s/it]
 97%|█████████▋| 14332/14725 [13:02:26<21:24,  3.27s/it]
 97%|█████████▋| 14333/14725 [13:02:29<21:26,  3.28s/it]
 97%|█████████▋| 14334/14725 [13:02:32<21:11,  3.25s/it]
 97%|█████████▋| 14335/14725 [13:02:36<20:46,  3.20s/it]
 97%|█████████▋| 14336/14725 [13:02:39<20:56,  3.23s/it]
 97%|█████████▋| 14337/14725 [13:02:42<20:58,  3.24s/it]
 97%|█████████▋| 14338/14725 [13:02:45<21:06,  3.27s/it]
 97%|█████████▋| 14339/14725 [13:02:49<20:43,  3.22s/it]
 97%|█████████▋| 14340/14725 [13:02:52<20:43,  3.23s/it]
                                                        

 97%|█████████▋| 14340/14725 [13:02
0: {'loss': 0.6747, 'grad_norm': 0.4210878750481338, 'learning_rate': 1e-05, 'epoch': 0.97}
0: :52<20:43,  3.23s/it]
 97%|█████████▋| 14341/14725 [13:02:55<20:59,  3.28s/it]
 97%|█████████▋| 14342/14725 [13:02:58<20:43,  3.25s/it]
 97%|█████████▋| 14343/14725 [13:03:02<20:59,  3.30s/it]
 97%|█████████▋| 14344/14725 [13:03:05<20:35,  3.24s/it]
 97%|█████████▋| 14345/14725 [13:03:08<20:21,  3.21s/it]
 97%|█████████▋| 14346/14725 [13:03:11<20:24,  3.23s/it]
 97%|█████████▋| 14347/14725 [13:03:15<20:25,  3.24s/it]
 97%|█████████▋| 14348/14725 [13:03:18<20:47,  3.31s/it]
 97%|█████████▋| 14349/14725 [13:03:21<20:31,  3.27s/it]
 97%|█████████▋| 14350/14725 [13:03:25<20:25,  3.27s/it]
                                                        

 97%|█████████▋| 14350/14725 [13:03:25<20:25,  3.27s/it]
 97%|█████████▋| 14351/14725 [13:03:28<20:08,  3.23s/it]
 97%|█████
0: {'loss': 0.6878, 'grad_norm': 0.4261246362046917, 'learning_rate': 1e-05, 'epoch': 0.98}
0: ████▋| 14352/14725 [13:03:31<20:11,  3.25s/it]
 97%|█████████▋| 14353/14725 [13:03:34<19:48,  3.19s/it]
 97%|█████████▋| 14354/14725 [13:03:37<19:50,  3.21s/it]
 97%|█████████▋| 14355/14725 [13:03:41<19:55,  3.23s/it]
 97%|█████████▋| 14356/14725 [13:03:44<19:56,  3.24s/it]
 98%|█████████▊| 14357/14725 [13:03:47<19:55,  3.25s/it]
 98%|█████████▊| 14358/14725 [13:03:50<20:02,  3.28s/it]
 98%|█████████▊| 14359/14725 [13:03:54<20:08,  3.30s/it]
 98%|█████████▊| 14360/14725 [13:03:57<20:06,  3.30s/it]
                                                        

 98%|█████████▊| 14360/14725 [13:03:57<20:06,  3.30s/it]
 98%|█████████▊| 14361/14725 [13:04:01<20:24,  3.36s/it]
 98%|█████████▊| 14362/14725 [13:04:04<20:10,  3.33s/it]
 98%|█████████▊| 14363/14725 [13:04:07<19:
0: {'loss': 0.6713, 'grad_norm': 0.3927694024884123, 'learning_rate': 1e-05, 'epoch': 0.98}
0: 49,  3.29s/it]
 98%|█████████▊| 14364/14725 [13:04:10<20:02,  3.33s/it]
 98%|█████████▊| 14365/14725 [13:04:14<19:58,  3.33s/it]
 98%|█████████▊| 14366/14725 [13:04:17<20:11,  3.38s/it]
 98%|█████████▊| 14367/14725 [13:04:21<20:03,  3.36s/it]
 98%|█████████▊| 14368/14725 [13:04:24<19:30,  3.28s/it]
 98%|█████████▊| 14369/14725 [13:04:27<19:30,  3.29s/it]
 98%|█████████▊| 14370/14725 [13:04:30<19:25,  3.28s/it]
                                                        

 98%|█████████▊| 14370/14725 [13:04:30<19:25,  3.28s/it]
 98%|█████████▊| 14371/14725 [13:04:33<19:03,  3.23s/it]
 98%|█████████▊| 14372/14725 [13:04:37<19:12,  3.26s/it]
 98%|█████████▊| 14373/14725 [13:04:40<19:13,  3.28s/it]
 98%|█████████▊| 14374/14725 [13:04:43<19:07,  3.27s/it]
 98%|███████�
0: {'loss': 0.6774, 'grad_norm': 0.4000381890086999, 'learning_rate': 1e-05, 'epoch': 0.98}
0: ��█▊| 14375/14725 [13:04:47<19:11,  3.29s/it]
 98%|█████████▊| 14376/14725 [13:04:50<19:03,  3.28s/it]
 98%|█████████▊| 14377/14725 [13:04:53<19:00,  3.28s/it]
 98%|█████████▊| 14378/14725 [13:04:56<18:48,  3.25s/it]
 98%|█████████▊| 14379/14725 [13:04:59<18:27,  3.20s/it]
 98%|█████████▊| 14380/14725 [13:05:03<18:35,  3.23s/it]
                                                        

 98%|█████████▊| 14380/14725 [13:05:03<18:35,  3.23s/it]
 98%|█████████▊| 14381/14725 [13:05:06<18:36,  3.25s/it]
 98%|█████████▊| 14382/14725 [13:05:09<18:41,  3.27s/it]
 98%|█████████▊| 14383/14725 [13:05:13<18:38,  3.27s/it]
 98%|█████████▊| 14384/14725 [13:05:16<18:26,  3.24s/it]
 98%|█████████▊| 14385/14725 [13:05:19<18:41,  3.30s/it]
 98%|█████████▊| 14386/14725 [13:05:22<18:30,  3.
0: {'loss': 0.6634, 'grad_norm': 0.43966546203418017, 'learning_rate': 1e-05, 'epoch': 0.98}
0: 28s/it]
 98%|█████████▊| 14387/14725 [13:05:26<18:26,  3.27s/it]
 98%|█████████▊| 14388/14725 [13:05:29<18:50,  3.35s/it]
 98%|█████████▊| 14389/14725 [13:05:33<18:37,  3.33s/it]
 98%|█████████▊| 14390/14725 [13:05:36<18:27,  3.31s/it]
                                                        

 98%|█████████▊| 14390/14725 [13:05:36<18:27,  3.31s/it]
 98%|█████████▊| 14391/14725 [13:05:39<18:09,  3.26s/it]
 98%|█████████▊| 14392/14725 [13:05:42<18:07,  3.27s/it]
 98%|█████████▊| 14393/14725 [13:05:46<18:15,  3.30s/it]
 98%|█████████▊| 14394/14725 [13:05:49<18:05,  3.28s/it]
 98%|█████████▊| 14395/14725 [13:05:52<17:58,  3.27s/it]
 98%|█████████▊| 14396/14725 [13:05:55<17:58,  3.28s/it]
 98%|█████████▊| 14397/14725 [13:05:59<17:42,  3.24s/it]
 98%|█████████�
0: {'loss': 0.6824, 'grad_norm': 0.3819213569529784, 'learning_rate': 1e-05, 'epoch': 0.98}
0: �| 14398/14725 [13:06:02<17:54,  3.29s/it]
 98%|█████████▊| 14399/14725 [13:06:05<17:43,  3.26s/it]
 98%|█████████▊| 14400/14725 [13:06:08<17:33,  3.24s/it]
                                                        

 98%|█████████▊| 14400/14725 [13:06:08<17:33,  3.24s/it]
 98%|█████████▊| 14401/14725 [13:06:12<17:29,  3.24s/it]
 98%|█████████▊| 14402/14725 [13:06:15<17:22,  3.23s/it]
 98%|█████████▊| 14403/14725 [13:06:18<17:15,  3.22s/it]
 98%|█████████▊| 14404/14725 [13:06:21<17:08,  3.20s/it]
 98%|█████████▊| 14405/14725 [13:06:24<17:09,  3.22s/it]
 98%|█████████▊| 14406/14725 [13:06:27<16:57,  3.19s/it]
 98%|█████████▊| 14407/14725 [13:06:31<17:07,  3.23s/it]
 98%|█████████▊| 14408/14725 [13:06:34<17:15,  3.27s/it]
 98%|█████████▊| 14409/14725 [13:06:37<17:08,  3.25s/it]
0: {'loss': 0.6712, 'grad_norm': 0.39451584912626253, 'learning_rate': 1e-05, 'epoch': 0.98}
0: 
 98%|█████████▊| 14410/14725 [13:06:41<17:01,  3.24s/it]
                                                        

 98%|█████████▊| 14410/14725 [13:06:41<17:01,  3.24s/it]
 98%|█████████▊| 14411/14725 [13:06:44<16:56,  3.24s/it]
 98%|█████████▊| 14412/14725 [13:06:47<16:44,  3.21s/it]
 98%|█████████▊| 14413/14725 [13:06:50<16:35,  3.19s/it]
 98%|█████████▊| 14414/14725 [13:06:54<16:51,  3.25s/it]
 98%|█████████▊| 14415/14725 [13:06:57<16:41,  3.23s/it]
 98%|█████████▊| 14416/14725 [13:07:00<16:33,  3.22s/it]
 98%|█████████▊| 14417/14725 [13:07:03<16:34,  3.23s/it]
 98%|█████████▊| 14418/14725 [13:07:06<16:37,  3.25s/it]
 98%|█████████▊| 14419/14725 [13:07:10<16:29,  3.23s/it]
 98%|█████████▊| 14420/14725 [13:07:13<16:14,  3.19s/it]
                                         
0: {'loss': 0.6663, 'grad_norm': 0.40162910454285145, 'learning_rate': 1e-05, 'epoch': 0.98}
0: {'loss': 0.6604, 'grad_norm': 0.4157935145394525, 'learning_rate': 9.996052735444863e-06, 'epoch': 0.98}
0:                

 98%|█████████▊| 14420/14725 [13:07:13<16:14,  3.19s/it]
 98%|█████████▊| 14421/14725 [13:07:16<16:20,  3.22s/it]
 98%|█████████▊| 14422/14725 [13:07:20<16:47,  3.32s/it]
 98%|█████████▊| 14423/14725 [13:07:23<16:42,  3.32s/it]
 98%|█████████▊| 14424/14725 [13:07:26<16:34,  3.30s/it]
 98%|█████████▊| 14425/14725 [13:07:29<16:22,  3.28s/it]
 98%|█████████▊| 14426/14725 [13:07:33<16:20,  3.28s/it]
 98%|█████████▊| 14427/14725 [13:07:36<16:01,  3.23s/it]
 98%|█████████▊| 14428/14725 [13:07:39<15:57,  3.22s/it]
 98%|█████████▊| 14429/14725 [13:07:42<16:01,  3.25s/it]
 98%|█████████▊| 14430/14725 [13:07:45<15:37,  3.18s/it]
                                                        

 98%|█████████▊| 14430/14725 [13:07:45<15:37,  3.18s/it]
 98%|██████�
0: {'loss': 0.6695, 'grad_norm': 0.41182862289889227, 'learning_rate': 9.951725498333449e-06, 'epoch': 0.98}
0: �██▊| 14431/14725 [13:07:49<15:36,  3.19s/it]
 98%|█████████▊| 14432/14725 [13:07:52<15:45,  3.23s/it]
 98%|█████████▊| 14433/14725 [13:07:55<15:37,  3.21s/it]
 98%|█████████▊| 14434/14725 [13:07:58<15:52,  3.27s/it]
 98%|█████████▊| 14435/14725 [13:08:02<15:42,  3.25s/it]
 98%|█████████▊| 14436/14725 [13:08:05<15:29,  3.21s/it]
 98%|█████████▊| 14437/14725 [13:08:08<15:40,  3.26s/it]
 98%|█████████▊| 14438/14725 [13:08:12<15:52,  3.32s/it]
 98%|█████████▊| 14439/14725 [13:08:15<15:45,  3.30s/it]
 98%|█████████▊| 14440/14725 [13:08:18<15:36,  3.29s/it]
                                                        

 98%|█████████▊| 14440/14725 [13:08:18<15:36,  3.29s/it]
 98%|█████████▊| 14441/14725 [13:08:21<15:31,  3.28s/it]
 98%|█████████▊| 14442/14725 [13:08:25<15:23,  
0: {'loss': 0.6557, 'grad_norm': 0.4004614536801997, 'learning_rate': 9.858624225078841e-06, 'epoch': 0.98}
0: 3.26s/it]
 98%|█████████▊| 14443/14725 [13:08:28<15:31,  3.30s/it]
 98%|█████████▊| 14444/14725 [13:08:31<15:27,  3.30s/it]
 98%|█████████▊| 14445/14725 [13:08:34<15:09,  3.25s/it]
 98%|█████████▊| 14446/14725 [13:08:38<15:14,  3.28s/it]
 98%|█████████▊| 14447/14725 [13:08:41<15:16,  3.30s/it]
 98%|█████████▊| 14448/14725 [13:08:44<15:16,  3.31s/it]
 98%|█████████▊| 14449/14725 [13:08:48<15:04,  3.28s/it]
 98%|█████████▊| 14450/14725 [13:08:51<14:50,  3.24s/it]
                                                        

 98%|█████████▊| 14450/14725 [13:08:51<14:50,  3.24s/it]
 98%|█████████▊| 14451/14725 [13:08:54<14:57,  3.28s/it]
 98%|█████████▊| 14452/14725 [13:08:57<14:39,  3.22s/it]
 98%|█████████▊| 14453/14725 [13:09:01<14:40,  3.24s/it]
 98%|█████████
0: {'loss': 0.6838, 'grad_norm': 0.3888227524408755, 'learning_rate': 9.717768952713514e-06, 'epoch': 0.98}
0: ▊| 14454/14725 [13:09:04<14:25,  3.20s/it]
 98%|█████████▊| 14455/14725 [13:09:07<14:32,  3.23s/it]
 98%|█████████▊| 14456/14725 [13:09:10<14:28,  3.23s/it]
 98%|█████████▊| 14457/14725 [13:09:13<14:31,  3.25s/it]
 98%|█████████▊| 14458/14725 [13:09:17<14:41,  3.30s/it]
 98%|█████████▊| 14459/14725 [13:09:20<14:22,  3.24s/it]
 98%|█████████▊| 14460/14725 [13:09:23<14:21,  3.25s/it]
                                                        

 98%|█████████▊| 14460/14725 [13:09:23<14:21,  3.25s/it]
 98%|█████████▊| 14461/14725 [13:09:27<14:20,  3.26s/it]
 98%|█████████▊| 14462/14725 [13:09:30<14:21,  3.28s/it]
 98%|█████████▊| 14463/14725 [13:09:33<14:17,  3.27s/it]
 98%|█████████▊| 14464/14725 [13:09:37<14:28,  3.33s/it]
 98%|█████████▊| 14465/14725 [13:09:40<14:24,  3.33s/i
0: {'loss': 0.6822, 'grad_norm': 0.3938515784305857, 'learning_rate': 9.530702921077358e-06, 'epoch': 0.98}
0: t]
 98%|█████████▊| 14466/14725 [13:09:43<14:04,  3.26s/it]
 98%|█████████▊| 14467/14725 [13:09:46<14:00,  3.26s/it]
 98%|█████████▊| 14468/14725 [13:09:50<14:02,  3.28s/it]
 98%|█████████▊| 14469/14725 [13:09:53<14:00,  3.28s/it]
 98%|█████████▊| 14470/14725 [13:09:56<14:00,  3.30s/it]
                                                        

 98%|█████████▊| 14470/14725 [13:09:56<14:00,  3.30s/it]
 98%|█████████▊| 14471/14725 [13:09:59<13:56,  3.29s/it]
 98%|█████████▊| 14472/14725 [13:10:03<13:43,  3.25s/it]
 98%|█████████▊| 14473/14725 [13:10:06<13:56,  3.32s/it]
 98%|█████████▊| 14474/14725 [13:10:09<13:40,  3.27s/it]
 98%|█████████▊| 14475/14725 [13:10:12<13:21,  3.21s/it]
 98%|█████████▊| 14476/14725 [13:10:16<13:19,  3.21s/it]
 98%|█████████▊| 14
0: {'loss': 0.6751, 'grad_norm': 0.40043261637596383, 'learning_rate': 9.29947566475907e-06, 'epoch': 0.98}
0: 477/14725 [13:10:19<13:14,  3.20s/it]
 98%|█████████▊| 14478/14725 [13:10:22<13:03,  3.17s/it]
 98%|█████████▊| 14479/14725 [13:10:25<13:09,  3.21s/it]
 98%|█████████▊| 14480/14725 [13:10:28<13:16,  3.25s/it]
                                                        

 98%|█████████▊| 14480/14725 [13:10:28<13:16,  3.25s/it]
 98%|█████████▊| 14481/14725 [13:10:32<13:06,  3.22s/it]
 98%|█████████▊| 14482/14725 [13:10:35<13:04,  3.23s/it]
 98%|█████████▊| 14483/14725 [13:10:38<13:06,  3.25s/it]
 98%|█████████▊| 14484/14725 [13:10:41<13:07,  3.27s/it]
 98%|█████████▊| 14485/14725 [13:10:45<13:16,  3.32s/it]
 98%|█████████▊| 14486/14725 [13:10:48<12:57,  3.25s/it]
 98%|█████████▊| 14487/14725 [13:10:51<13:02,  3.29s/it]
 98%|█████████▊| 14488/14725 [13:10:54<12:46,  3.23s/it]
 98%
0: {'loss': 0.6461, 'grad_norm': 0.3904372704915405, 'learning_rate': 9.02662055796628e-06, 'epoch': 0.98}
0: |█████████▊| 14489/14725 [13:10:58<12:49,  3.26s/it]
 98%|█████████▊| 14490/14725 [13:11:01<12:45,  3.26s/it]
                                                        

 98%|█████████▊| 14490/14725 [13:11:01<12:45,  3.26s/it]
 98%|█████████▊| 14491/14725 [13:11:04<12:41,  3.25s/it]
 98%|█████████▊| 14492/14725 [13:11:07<12:33,  3.23s/it]
 98%|█████████▊| 14493/14725 [13:11:11<12:31,  3.24s/it]
 98%|█████████▊| 14494/14725 [13:11:14<12:25,  3.23s/it]
 98%|█████████▊| 14495/14725 [13:11:17<12:33,  3.28s/it]
 98%|█████████▊| 14496/14725 [13:11:21<12:23,  3.25s/it]
 98%|█████████▊| 14497/14725 [13:11:24<12:36,  3.32s/it]
 98%|█████████▊| 14498/14725 [13:11:27<12:22,  3.27s/it]
 98%|█████████▊| 14499/14725 [13:11:31<12:25,  3.30s/it]
 98%|█████████▊| 14500/147
0: {'loss': 0.6738, 'grad_norm': 0.3957139103076907, 'learning_rate': 8.715127058347615e-06, 'epoch': 0.98}
0: {'loss': 0.6662, 'grad_norm': 0.37623337539675, 'learning_rate': 8.368407953869105e-06, 'epoch': 0.99}
0: 25 [13:11:34<12:16,  3.27s/it]
                                                        

 98%|█████████▊| 14500/14725 [13:11:34<12:16,  3.27s/it]
 98%|█████████▊| 14501/14725 [13:11:37<12:03,  3.23s/it]
 98%|█████████▊| 14502/14725 [13:11:40<11:59,  3.23s/it]
 98%|█████████▊| 14503/14725 [13:11:43<12:04,  3.26s/it]
 98%|█████████▊| 14504/14725 [13:11:47<11:50,  3.21s/it]
 99%|█████████▊| 14505/14725 [13:11:50<11:47,  3.22s/it]
 99%|█████████▊| 14506/14725 [13:11:53<11:45,  3.22s/it]
 99%|█████████▊| 14507/14725 [13:11:56<11:36,  3.20s/it]
 99%|█████████▊| 14508/14725 [13:11:59<11:32,  3.19s/it]
 99%|█████████▊| 14509/14725 [13:12:02<11:25,  3.17s/it]
 99%|█████████▊| 14510/14725 [13:12:06<11:30,  3.21s/it]
                                                        

 99%|████████�
0: {'loss': 0.6552, 'grad_norm': 0.38478047694797185, 'learning_rate': 7.99026197159505e-06, 'epoch': 0.99}
0: ��▊| 14510/14725 [13:12:06<11:30,  3.21s/it]
 99%|█████████▊| 14511/14725 [13:12:09<11:32,  3.24s/it]
 99%|█████████▊| 14512/14725 [13:12:12<11:31,  3.25s/it]
 99%|█████████▊| 14513/14725 [13:12:15<11:22,  3.22s/it]
 99%|█████████▊| 14514/14725 [13:12:19<11:13,  3.19s/it]
 99%|█████████▊| 14515/14725 [13:12:22<11:15,  3.22s/it]
 99%|█████████▊| 14516/14725 [13:12:25<11:22,  3.26s/it]
 99%|█████████▊| 14517/14725 [13:12:29<11:20,  3.27s/it]
 99%|█████████▊| 14518/14725 [13:12:32<11:16,  3.27s/it]
 99%|█████████▊| 14519/14725 [13:12:35<11:04,  3.22s/it]
 99%|█████████▊| 14520/14725 [13:12:38<11:07,  3.25s/it]
                                                        

 99%|█████████▊| 14520/14725 [13:12:38<11:07,  3.25s/it]
 99%|█████████▊| 14521/14725 [13:12:42<11:13,  3.30s
0: {'loss': 0.6782, 'grad_norm': 0.37140655334903516, 'learning_rate': 7.584832158039379e-06, 'epoch': 0.99}
0: /it]
 99%|█████████▊| 14522/14725 [13:12:45<11:11,  3.31s/it]
 99%|█████████▊| 14523/14725 [13:12:48<11:09,  3.31s/it]
 99%|█████████▊| 14524/14725 [13:12:52<11:05,  3.31s/it]
 99%|█████████▊| 14525/14725 [13:12:55<10:57,  3.29s/it]
 99%|█████████▊| 14526/14725 [13:12:58<11:04,  3.34s/it]
 99%|█████████▊| 14527/14725 [13:13:01<10:47,  3.27s/it]
 99%|█████████▊| 14528/14725 [13:13:05<10:52,  3.31s/it]
 99%|█████████▊| 14529/14725 [13:13:08<10:34,  3.24s/it]
 99%|█████████▊| 14530/14725 [13:13:11<10:38,  3.27s/it]
                                                        

 99%|█████████▊| 14530/14725 [13:13:11<10:38,  3.27s/it]
 99%|█████████▊| 14531/14725 [13:13:15<10:40,  3.30s/it]
 99%|█████████▊| 14532/14725 [13:13:18<10:30,  3.27s/it]
 99%|█████████▊| 
0: {'loss': 0.6691, 'grad_norm': 0.37616909212909017, 'learning_rate': 7.156560487081052e-06, 'epoch': 0.99}
0: 14533/14725 [13:13:22<11:15,  3.52s/it]
 99%|█████████▊| 14534/14725 [13:13:25<11:02,  3.47s/it]
 99%|█████████▊| 14535/14725 [13:13:28<10:46,  3.40s/it]
 99%|█████████▊| 14536/14725 [13:13:32<10:25,  3.31s/it]
 99%|█████████▊| 14537/14725 [13:13:35<10:24,  3.32s/it]
 99%|█████████▊| 14538/14725 [13:13:38<10:19,  3.31s/it]
 99%|█████████▊| 14539/14725 [13:13:41<10:12,  3.29s/it]
 99%|█████████▊| 14540/14725 [13:13:45<09:55,  3.22s/it]
                                                        

 99%|█████████▊| 14540/14725 [13:13:45<09:55,  3.22s/it]
 99%|█████████▉| 14541/14725 [13:13:48<09:53,  3.22s/it]
 99%|█████████▉| 14542/14725 [13:13:51<09:53,  3.24s/it]
 99%|█████████▉| 14543/14725 [13:13:54<10:01,  3.30s/it]
 99%|█████████▉| 14544/14725 [13:13:58<09:56,  3.29s/it]
 9
0: {'loss': 0.6671, 'grad_norm': 0.3897490275413447, 'learning_rate': 6.710139192768695e-06, 'epoch': 0.99}
0: 9%|█████████▉| 14545/14725 [13:14:01<09:47,  3.26s/it]
 99%|█████████▉| 14546/14725 [13:14:04<09:40,  3.24s/it]
 99%|█████████▉| 14547/14725 [13:14:07<09:33,  3.22s/it]
 99%|█████████▉| 14548/14725 [13:14:11<09:29,  3.22s/it]
 99%|█████████▉| 14549/14725 [13:14:14<09:22,  3.20s/it]
 99%|█████████▉| 14550/14725 [13:14:17<09:21,  3.21s/it]
                                                        

 99%|█████████▉| 14550/14725 [13:14:17<09:21,  3.21s/it]
 99%|█████████▉| 14551/14725 [13:14:20<09:20,  3.22s/it]
 99%|█████████▉| 14552/14725 [13:14:23<09:20,  3.24s/it]
 99%|█████████▉| 14553/14725 [13:14:27<09:14,  3.22s/it]
 99%|█████████▉| 14554/14725 [13:14:30<09:12,  3.23s/it]
 99%|█████████▉| 14555/14725 [13:14:33<09:12,  3.25s/it]
 99%|█████████▉| 14556/1
0: {'loss': 0.6585, 'grad_norm': 0.37972886727230254, 'learning_rate': 6.25045936022246e-06, 'epoch': 0.99}
0: 4725 [13:14:36<09:10,  3.26s/it]
 99%|█████████▉| 14557/14725 [13:14:40<09:01,  3.23s/it]
 99%|█████████▉| 14558/14725 [13:14:43<09:05,  3.27s/it]
 99%|█████████▉| 14559/14725 [13:14:46<09:09,  3.31s/it]
 99%|█████████▉| 14560/14725 [13:14:50<09:07,  3.32s/it]
                                                        

 99%|█████████▉| 14560/14725 [13:14:50<09:07,  3.32s/it]
 99%|█████████▉| 14561/14725 [13:14:53<09:13,  3.37s/it]
 99%|█████████▉| 14562/14725 [13:14:57<09:09,  3.37s/it]
 99%|█████████▉| 14563/14725 [13:15:00<08:50,  3.28s/it]
 99%|█████████▉| 14564/14725 [13:15:03<08:53,  3.31s/it]
 99%|█████████▉| 14565/14725 [13:15:06<08:40,  3.26s/it]
 99%|█████████▉| 14566/14725 [13:15:09<08:40,  3.27s/it]
 99%|█████████▉| 14567/14725 [13:15:13<08:33,  3.25s/it]
 99%|█�
0: {'loss': 0.6676, 'grad_norm': 0.39144341066978156, 'learning_rate': 5.782557337881911e-06, 'epoch': 0.99}
0: ��███████▉| 14568/14725 [13:15:16<08:24,  3.21s/it]
 99%|█████████▉| 14569/14725 [13:15:19<08:28,  3.26s/it]
 99%|█████████▉| 14570/14725 [13:15:23<08:31,  3.30s/it]
                                                        

 99%|█████████▉| 14570/14725 [13:15:23<08:31,  3.30s/it]
 99%|█████████▉| 14571/14725 [13:15:26<08:18,  3.24s/it]
 99%|█████████▉| 14572/14725 [13:15:29<08:34,  3.36s/it]
 99%|█████████▉| 14573/14725 [13:15:33<08:30,  3.36s/it]
 99%|█████████▉| 14574/14725 [13:15:36<08:25,  3.35s/it]
 99%|█████████▉| 14575/14725 [13:15:39<08:17,  3.31s/it]
 99%|█████████▉| 14576/14725 [13:15:42<08:07,  3.27s/it]
 99%|█████████▉| 14577/14725 [13:15:45<07:58,  3.23s/it]
 99%|█████████▉| 14578/14725 [13:15:49<08:04,  3.29s/it]
 99%|█████████▉| 14579/14725 [1
0: {'loss': 0.674, 'grad_norm': 0.36013576610388265, 'learning_rate': 5.311559558218603e-06, 'epoch': 0.99}
0: 3:15:52<07:56,  3.26s/it]
 99%|█████████▉| 14580/14725 [13:15:55<07:53,  3.26s/it]
                                                        

 99%|█████████▉| 14580/14725 [13:15:55<07:53,  3.26s/it]
 99%|█████████▉| 14581/14725 [13:15:59<07:50,  3.27s/it]
 99%|█████████▉| 14582/14725 [13:16:02<07:43,  3.24s/it]
 99%|█████████▉| 14583/14725 [13:16:05<07:41,  3.25s/it]
 99%|█████████▉| 14584/14725 [13:16:08<07:35,  3.23s/it]
 99%|█████████▉| 14585/14725 [13:16:12<07:34,  3.25s/it]
 99%|█████████▉| 14586/14725 [13:16:15<07:30,  3.24s/it]
 99%|█████████▉| 14587/14725 [13:16:18<07:23,  3.21s/it]
 99%|█████████▉| 14588/14725 [13:16:21<07:17,  3.19s/it]
 99%|█████████▉| 14589/14725 [13:16:24<07:17,  3.22s/it]
 99%|█████████▉| 14590/14725 [13:16:28<07:16,  3.23s/it]
                
0: {'loss': 0.6637, 'grad_norm': 0.35336048556437993, 'learning_rate': 4.842626371469149e-06, 'epoch': 0.99}
0: {'loss': 0.6683, 'grad_norm': 0.3692389162599803, 'learning_rate': 4.380895507758155e-06, 'epoch': 0.99}
0:                                         

 99%|█████████▉| 14590/14725 [13:16:28<07:16,  3.23s/it]
 99%|█████████▉| 14591/14725 [13:16:31<07:12,  3.23s/it]
 99%|█████████▉| 14592/14725 [13:16:34<07:09,  3.23s/it]
 99%|█████████▉| 14593/14725 [13:16:37<07:11,  3.27s/it]
 99%|█████████▉| 14594/14725 [13:16:41<07:15,  3.32s/it]
 99%|█████████▉| 14595/14725 [13:16:44<07:09,  3.31s/it]
 99%|█████████▉| 14596/14725 [13:16:47<07:05,  3.30s/it]
 99%|█████████▉| 14597/14725 [13:16:51<07:05,  3.33s/it]
 99%|█████████▉| 14598/14725 [13:16:54<07:03,  3.33s/it]
 99%|█████████▉| 14599/14725 [13:16:57<06:53,  3.28s/it]
 99%|█████████▉| 14600/14725 [13:17:01<06:47,  3.26s/it]
                                                        

 99%|█████████▉| 14600/14725 [13:17:01<06:47,  3.26s/it]
0: {'loss': 0.661, 'grad_norm': 0.3850301061272489, 'learning_rate': 3.931425787051832e-06, 'epoch': 0.99}
0:  99%|█████████▉| 14601/14725 [13:17:04<06:41,  3.24s/it]
 99%|█████████▉| 14602/14725 [13:17:07<06:37,  3.23s/it]
 99%|█████████▉| 14603/14725 [13:17:11<06:46,  3.33s/it]
 99%|█████████▉| 14604/14725 [13:17:14<06:39,  3.31s/it]
 99%|█████████▉| 14605/14725 [13:17:17<06:35,  3.30s/it]
 99%|█████████▉| 14606/14725 [13:17:20<06:25,  3.24s/it]
 99%|█████████▉| 14607/14725 [13:17:23<06:23,  3.25s/it]
 99%|█████████▉| 14608/14725 [13:17:27<06:21,  3.26s/it]
 99%|█████████▉| 14609/14725 [13:17:30<06:21,  3.29s/it]
 99%|█████████▉| 14610/14725 [13:17:33<06:11,  3.23s/it]
                                                        

 99%|█████████▉| 14610/14725 [13:17:33<06:11,  3.23s/it]
 99%|█████████▉| 14611/14725 [13:17:36<06:07,  3.22s/it]
 99%|█████████▉| 14612
0: {'loss': 0.6703, 'grad_norm': 0.36715506948971394, 'learning_rate': 3.499141693667828e-06, 'epoch': 0.99}
0: /14725 [13:17:40<06:08,  3.26s/it]
 99%|█████████▉| 14613/14725 [13:17:43<06:05,  3.26s/it]
 99%|█████████▉| 14614/14725 [13:17:46<06:00,  3.25s/it]
 99%|█████████▉| 14615/14725 [13:17:50<06:00,  3.28s/it]
 99%|█████████▉| 14616/14725 [13:17:53<05:51,  3.22s/it]
 99%|█████████▉| 14617/14725 [13:17:56<05:54,  3.28s/it]
 99%|█████████▉| 14618/14725 [13:17:59<05:47,  3.25s/it]
 99%|█████████▉| 14619/14725 [13:18:03<05:47,  3.27s/it]
 99%|█████████▉| 14620/14725 [13:18:06<05:40,  3.24s/it]
                                                        

 99%|█████████▉| 14620/14725 [13:18:06<05:40,  3.24s/it]
 99%|█████████▉| 14621/14725 [13:18:09<05:41,  3.28s/it]
 99%|█████████▉| 14622/14725 [13:18:13<05:41,  3.31s/it]
 99%|█████████▉| 14623/14725 [13:18:16<05:34,  3.28s/it]
 99%|�
0: {'loss': 0.6615, 'grad_norm': 0.36382993509435757, 'learning_rate': 3.0887794225945143e-06, 'epoch': 0.99}
0: �████████▉| 14624/14725 [13:18:19<05:30,  3.27s/it]
 99%|█████████▉| 14625/14725 [13:18:22<05:20,  3.21s/it]
 99%|█████████▉| 14626/14725 [13:18:25<05:17,  3.21s/it]
 99%|█████████▉| 14627/14725 [13:18:28<05:15,  3.22s/it]
 99%|█████████▉| 14628/14725 [13:18:32<05:09,  3.19s/it]
 99%|█████████▉| 14629/14725 [13:18:35<05:06,  3.19s/it]
 99%|█████████▉| 14630/14725 [13:18:38<05:08,  3.24s/it]
                                                        

 99%|█████████▉| 14630/14725 [13:18:38<05:08,  3.24s/it]
 99%|█████████▉| 14631/14725 [13:18:41<05:06,  3.26s/it]
 99%|█████████▉| 14632/14725 [13:18:45<05:00,  3.23s/it]
 99%|█████████▉| 14633/14725 [13:18:48<04:54,  3.21s/it]
 99%|█████████▉| 14634/14725 [13:18:51<04:56,  3.26s/it]
 99%|█████████▉| 14635/14725 
0: {'loss': 0.6606, 'grad_norm': 0.3794383705788541, 'learning_rate': 2.7048349887476038e-06, 'epoch': 0.99}
0: [13:18:55<04:57,  3.30s/it]
 99%|█████████▉| 14636/14725 [13:18:58<04:47,  3.23s/it]
 99%|█████████▉| 14637/14725 [13:19:01<04:42,  3.21s/it]
 99%|█████████▉| 14638/14725 [13:19:04<04:39,  3.22s/it]
 99%|█████████▉| 14639/14725 [13:19:07<04:37,  3.22s/it]
 99%|█████████▉| 14640/14725 [13:19:11<04:37,  3.27s/it]
                                                        

 99%|█████████▉| 14640/14725 [13:19:11<04:37,  3.27s/it]
 99%|█████████▉| 14641/14725 [13:19:14<04:33,  3.26s/it]
 99%|█████████▉| 14642/14725 [13:19:17<04:33,  3.29s/it]
 99%|█████████▉| 14643/14725 [13:19:21<04:34,  3.35s/it]
 99%|█████████▉| 14644/14725 [13:19:24<04:28,  3.32s/it]
 99%|█████████▉| 14645/14725 [13:19:27<04:26,  3.33s/it]
 99%|█████████▉| 14646/14725 [13:19:31<04:24,  3.35s/it]
 99%|███
0: {'loss': 0.664, 'grad_norm': 0.35762151053141283, 'learning_rate': 2.3515149676898554e-06, 'epoch': 0.99}
0: ██████▉| 14647/14725 [13:19:34<04:16,  3.29s/it]
 99%|█████████▉| 14648/14725 [13:19:37<04:12,  3.28s/it]
 99%|█████████▉| 14649/14725 [13:19:40<04:05,  3.24s/it]
 99%|█████████▉| 14650/14725 [13:19:43<03:58,  3.19s/it]
                                                        

 99%|█████████▉| 14650/14725 [13:19:43<03:58,  3.19s/it]
 99%|█████████▉| 14651/14725 [13:19:46<03:52,  3.14s/it]
100%|█████████▉| 14652/14725 [13:19:50<03:57,  3.25s/it]
100%|█████████▉| 14653/14725 [13:19:53<03:52,  3.23s/it]
100%|█████████▉| 14654/14725 [13:19:56<03:47,  3.20s/it]
100%|█████████▉| 14655/14725 [13:20:00<03:47,  3.24s/it]
100%|█████████▉| 14656/14725 [13:20:03<03:45,  3.27s/it]
100%|█████████▉| 14657/14725 [13:20:06<03:48,  3.36s/it]
100%|█████████▉| 14658/14725 [13:20:
0: {'loss': 0.6866, 'grad_norm': 0.3556406894278512, 'learning_rate': 2.032690407508949e-06, 'epoch': 1.0}
0: 10<03:44,  3.35s/it]
100%|█████████▉| 14659/14725 [13:20:13<03:39,  3.33s/it]
100%|█████████▉| 14660/14725 [13:20:16<03:34,  3.29s/it]
                                                        

100%|█████████▉| 14660/14725 [13:20:16<03:34,  3.29s/it]
100%|█████████▉| 14661/14725 [13:20:19<03:29,  3.28s/it]
100%|█████████▉| 14662/14725 [13:20:23<03:24,  3.25s/it]
100%|█████████▉| 14663/14725 [13:20:26<03:20,  3.24s/it]
100%|█████████▉| 14664/14725 [13:20:29<03:21,  3.30s/it]
100%|█████████▉| 14665/14725 [13:20:33<03:25,  3.42s/it]
100%|█████████▉| 14666/14725 [13:20:36<03:16,  3.34s/it]
100%|█████████▉| 14667/14725 [13:20:40<03:16,  3.39s/it]
100%|█████████▉| 14668/14725 [13:20:43<03:09,  3.33s/it]
100%|█████████▉| 14669/14725 [13:20:46<03:04,  3.29s/it]
100%|█████�
0: {'loss': 0.6829, 'grad_norm': 0.3541351991928239, 'learning_rate': 1.7518544168045527e-06, 'epoch': 1.0}
0: {'loss': 0.6715, 'grad_norm': 0.35589453798548365, 'learning_rate': 1.5120838934595338e-06, 'epoch': 1.0}
0: ��███▉| 14670/14725 [13:20:50<03:03,  3.34s/it]
                                                        

100%|█████████▉| 14670/14725 [13:20:50<03:03,  3.34s/it]
100%|█████████▉| 14671/14725 [13:20:53<03:01,  3.36s/it]
100%|█████████▉| 14672/14725 [13:20:56<02:56,  3.33s/it]
100%|█████████▉| 14673/14725 [13:21:00<02:52,  3.33s/it]
100%|█████████▉| 14674/14725 [13:21:03<02:51,  3.36s/it]
100%|█████████▉| 14675/14725 [13:21:06<02:44,  3.30s/it]
100%|█████████▉| 14676/14725 [13:21:09<02:41,  3.30s/it]
100%|█████████▉| 14677/14725 [13:21:13<02:39,  3.32s/it]
100%|█████████▉| 14678/14725 [13:21:16<02:35,  3.30s/it]
100%|█████████▉| 14679/14725 [13:21:19<02:29,  3.24s/it]
100%|█████████▉| 14680/14725 [13:21:23<02:30,  3.35s/it]
                                                        

100%|
0: {'loss': 0.6739, 'grad_norm': 0.3676013825818554, 'learning_rate': 1.316005813502869e-06, 'epoch': 1.0}
0: █████████▉| 14680/14725 [13:21:23<02:30,  3.35s/it]
100%|█████████▉| 14681/14725 [13:21:26<02:26,  3.33s/it]
100%|█████████▉| 14682/14725 [13:21:29<02:24,  3.36s/it]
100%|█████████▉| 14683/14725 [13:21:33<02:21,  3.36s/it]
100%|█████████▉| 14684/14725 [13:21:36<02:16,  3.32s/it]
100%|█████████▉| 14685/14725 [13:21:39<02:12,  3.31s/it]
100%|█████████▉| 14686/14725 [13:21:43<02:07,  3.27s/it]
100%|█████████▉| 14687/14725 [13:21:46<02:04,  3.27s/it]
100%|█████████▉| 14688/14725 [13:21:49<02:00,  3.25s/it]
100%|█████████▉| 14689/14725 [13:21:52<01:57,  3.26s/it]
100%|█████████▉| 14690/14725 [13:21:55<01:52,  3.22s/it]
                                                        

100%|█████████▉| 14690/14725 [13:21:55<01:52,  3.22s/it]
100%|█████████▉| 14691/1472
0: {'loss': 0.6766, 'grad_norm': 0.36490597029472827, 'learning_rate': 1.1657684494105386e-06, 'epoch': 1.0}
0: 5 [13:21:58<01:47,  3.17s/it]
100%|█████████▉| 14692/14725 [13:22:02<01:45,  3.20s/it]
100%|█████████▉| 14693/14725 [13:22:05<01:43,  3.22s/it]
100%|█████████▉| 14694/14725 [13:22:08<01:40,  3.23s/it]
100%|█████████▉| 14695/14725 [13:22:11<01:37,  3.24s/it]
100%|█████████▉| 14696/14725 [13:22:15<01:34,  3.24s/it]
100%|█████████▉| 14697/14725 [13:22:18<01:30,  3.22s/it]
100%|█████████▉| 14698/14725 [13:22:21<01:28,  3.27s/it]
100%|█████████▉| 14699/14725 [13:22:24<01:24,  3.24s/it]
100%|█████████▉| 14700/14725 [13:22:28<01:22,  3.32s/it]
                                                        

100%|█████████▉| 14700/14725 [13:22:28<01:22,  3.32s/it]
100%|█████████▉| 14701/14725 [13:22:31<01:19,  3.33s/it]
100%|█████████▉| 14702/14725 [13:22:35<01:16,  3.33s/it]
100%|██�
0: {'loss': 0.6605, 'grad_norm': 0.3732990110185473, 'learning_rate': 1.0630178331827281e-06, 'epoch': 1.0}
0: ��██████▉| 14703/14725 [13:22:38<01:13,  3.32s/it]
100%|█████████▉| 14704/14725 [13:22:41<01:08,  3.28s/it]
100%|█████████▉| 14705/14725 [13:22:44<01:04,  3.22s/it]
100%|█████████▉| 14706/14725 [13:22:47<01:00,  3.21s/it]
100%|█████████▉| 14707/14725 [13:22:51<00:57,  3.19s/it]
100%|█████████▉| 14708/14725 [13:22:54<00:54,  3.21s/it]
100%|█████████▉| 14709/14725 [13:22:57<00:50,  3.19s/it]
100%|█████████▉| 14710/14725 [13:23:00<00:48,  3.24s/it]
                                                        

100%|█████████▉| 14710/14725 [13:23:00<00:48,  3.24s/it]
100%|█████████▉| 14711/14725 [13:23:03<00:45,  3.23s/it]
100%|█████████▉| 14712/14725 [13:23:07<00:42,  3.30s/it]
100%|█████████▉| 14713/14725 [13:23:10<00:39,  3.27s/it]
100%|█████████▉| 14714/14725 [13:2
0: {'loss': 0.6534, 'grad_norm': 0.3555674913287364, 'learning_rate': 1.008879722072778e-06, 'epoch': 1.0}
0: {'train_runtime': 48238.6792, 'train_samples_per_second': 4.884, 'train_steps_per_second': 0.305, 'train_loss': 0.7104830094387656, 'epoch': 1.0}
0: 3:13<00:36,  3.28s/it]
100%|█████████▉| 14715/14725 [13:23:17<00:32,  3.20s/it]
100%|█████████▉| 14716/14725 [13:23:20<00:28,  3.19s/it]
100%|█████████▉| 14717/14725 [13:23:23<00:25,  3.19s/it]
100%|█████████▉| 14718/14725 [13:23:26<00:22,  3.16s/it]
100%|█████████▉| 14719/14725 [13:23:29<00:19,  3.21s/it]
100%|█████████▉| 14720/14725 [13:23:33<00:16,  3.25s/it]
                                                        

100%|█████████▉| 14720/14725 [13:23:33<00:16,  3.25s/it]
100%|█████████▉| 14721/14725 [13:23:36<00:13,  3.29s/it]
100%|█████████▉| 14722/14725 [13:23:39<00:09,  3.30s/it]
100%|█████████▉| 14723/14725 [13:23:43<00:06,  3.34s/it]
100%|█████████▉| 14724/14725 [13:23:46<00:03,  3.29s/it]
100%|██████████| 14725/14725 [13:23:49<00:00,  3.33s/it]
                   
0:                                      

100%|██████████| 14725/14725 [13:23:58<00:00,  3.33s/it]
100%|██████████| 14725/14725 [13:23:58<00:00,  3.28s/it]
0: [2025-08-07 15:54:46,609] [INFO] [axolotl.train.save_trained_model:246] [PID:2863199] [RANK:0] Training completed! Saving trained model to /lustre/fswork/projects/rech/dgo/udv55np/ift/Qwen3-235B-A22B/Qwen2.5-1.5B/1.
0: [2025-08-07 15:54:50,351] [INFO] [axolotl.train.save_trained_model:331] [PID:2863199] [RANK:0] Model successfully saved to /lustre/fswork/projects/rech/dgo/udv55np/ift/Qwen3-235B-A22B/Qwen2.5-1.5B/1