Instructions to use Scale-or-Reason/Qwen2.5-14B-math-ift with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Scale-or-Reason/Qwen2.5-14B-math-ift with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="Scale-or-Reason/Qwen2.5-14B-math-ift")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Scale-or-Reason/Qwen2.5-14B-math-ift")
model = AutoModelForCausalLM.from_pretrained("Scale-or-Reason/Qwen2.5-14B-math-ift")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Scale-or-Reason/Qwen2.5-14B-math-ift with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Scale-or-Reason/Qwen2.5-14B-math-ift"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Scale-or-Reason/Qwen2.5-14B-math-ift",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Scale-or-Reason/Qwen2.5-14B-math-ift

SGLang

How to use Scale-or-Reason/Qwen2.5-14B-math-ift with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "Scale-or-Reason/Qwen2.5-14B-math-ift" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Scale-or-Reason/Qwen2.5-14B-math-ift",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "Scale-or-Reason/Qwen2.5-14B-math-ift" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Scale-or-Reason/Qwen2.5-14B-math-ift",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Scale-or-Reason/Qwen2.5-14B-math-ift with Docker Model Runner:
```
docker model run hf.co/Scale-or-Reason/Qwen2.5-14B-math-ift
```

Qwen2.5-14B-math-ift / slurm.out

Nicolas-BZRD

Add files using upload-large-folder tool

de196b8 verified 10 months ago

Raw

History Blame Contribute Delete

229 kB

	3: W0902 18:58:55.388000 284480 torch/distributed/run.py:792]
	3: W0902 18:58:55.388000 284480 torch/distributed/run.py:792] *****************************************
	3: W0902 18:58:55.388000 284480 torch/distributed/run.py:792] Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed.
	3: W0902 18:58:55.388000 284480 torch/distributed/run.py:792] *****************************************
	0: W0902 18:58:55.639000 2134898 torch/distributed/run.py:792]
	0: W0902 18:58:55.639000 2134898 torch/distributed/run.py:792] *****************************************
	0: W0902 18:58:55.639000 2134898 torch/distributed/run.py:792] Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed.
	0: W0902 18:58:55.639000 2134898 torch/distributed/run.py:792] *****************************************
	1: W0902 18:58:55.819000 3856158 torch/distributed/run.py:792]
	1: W0902 18:58:55.819000 3856158 torch/distributed/run.py:792] *****************************************
	1: W0902 18:58:55.819000 3856158 torch/distributed/run.py:792] Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed.
	1: W0902 18:58:55.819000 3856158 torch/distributed/run.py:792] *****************************************
	2: W0902 18:58:56.452000 3393108 torch/distributed/run.py:792]
	2: W0902 18:58:56.452000 3393108 torch/distributed/run.py:792] *****************************************
	2: W0902 18:58:56.452000 3393108 torch/distributed/run.py:792] Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed.
	2: W0902 18:58:56.452000 3393108 torch/distributed/run.py:792] *****************************************
	3: [2025-09-02 18:59:20,646] [INFO] [axolotl.utils.schemas.validation.check_eval_packing:119] [PID:284556] [RANK:0] explicitly setting `eval_sample_packing` to match `sample_packing`[39m
	3: [2025-09-02 18:59:20,646] [INFO] [axolotl.utils.schemas.validation.hint_sample_packing_padding:218] [PID:284556] [RANK:0] Setting `pad_to_sequence_len: true` to prevent memory leaks when sample_packing[39m
	1: [2025-09-02 18:59:21,727] [INFO] [axolotl.utils.schemas.validation.check_eval_packing:119] [PID:3856235] [RANK:0] explicitly setting `eval_sample_packing` to match `sample_packing`[39m
	1: [2025-09-02 18:59:21,728] [INFO] [axolotl.utils.schemas.validation.hint_sample_packing_padding:218] [PID:3856235] [RANK:0] Setting `pad_to_sequence_len: true` to prevent memory leaks when sample_packing[39m
	0: [2025-09-02 18:59:23,380] [INFO] [axolotl.utils.schemas.validation.check_eval_packing:119] [PID:2134977] [RANK:0] explicitly setting `eval_sample_packing` to match `sample_packing`[39m
	0: [2025-09-02 18:59:23,380] [INFO] [axolotl.utils.schemas.validation.hint_sample_packing_padding:218] [PID:2134977] [RANK:0] Setting `pad_to_sequence_len: true` to prevent memory leaks when sample_packing[39m
	2: [2025-09-02 18:59:23,899] [INFO] [axolotl.utils.schemas.validation.check_eval_packing:119] [PID:3393184] [RANK:0] explicitly setting `eval_sample_packing` to match `sample_packing`[39m
	2: [2025-09-02 18:59:23,899] [INFO] [axolotl.utils.schemas.validation.hint_sample_packing_padding:218] [PID:3393184] [RANK:0] Setting `pad_to_sequence_len: true` to prevent memory leaks when sample_packing[39m
	3: [2025-09-02 18:59:25,238] [INFO] [axolotl.utils.data.sft._load_raw_datasets:314] [PID:284557] [RANK:1] Loading raw datasets...[39m
	3: [2025-09-02 18:59:25,510] [INFO] [axolotl.utils.data.wrappers.get_dataset_wrapper:88] [PID:284557] [RANK:1] Loading dataset: /lustre/fswork/projects/rech/qwv/udv55np/dataset/math/hf/no_thinking_text/generator/default-d32b2cae8ea7e541/0.0.0 with base_type: chat_template and prompt_style: None[39m
	0: [2025-09-02 18:59:27,157] [INFO] [axolotl.cli.config.load_cfg:245] [PID:2134977] [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/1756826506262014665.yaml",
	0: "base_model": "/lustre/fswork/projects/rech/qwv/udv55np/Qwen/Qwen2.5-14B_ift",
	0: "base_model_config": "/lustre/fswork/projects/rech/qwv/udv55np/Qwen/Qwen2.5-14B_ift",
	0: "batch_size": 16,
	0: "bf16": true,
	0: "capabilities": {
	0: "bf16": true,
	0: "compute_capability": "sm_90",
	0: "fp8": false,
	0: "n_gpu": 16,
	0: "n_node": 1
	0: },
	0: "chat_template": "qwen_25",
	0: "context_parallel_size": 1,
	0: "dataloader_num_workers": 16,
	0: "dataloader_pin_memory": true,
	0: "dataloader_prefetch_factor": 256,
	0: "dataset_prepared_path": "/lustre/fsn1/projects/rech/dgo/udv55np/dataset_math/Qwen3-235B-A22B/0",
	0: "dataset_processes": 192,
	0: "datasets": [
	0: {
	0: "chat_template": "tokenizer_default",
	0: "field_messages": "conversations",
	0: "message_property_mappings": {
	0: "content": "content",
	0: "role": "role"
	0: },
	0: "path": "/lustre/fswork/projects/rech/qwv/udv55np/dataset/math/hf/no_thinking_text/generator/default-d32b2cae8ea7e541/0.0.0",
	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: "dion_rank_fraction": 1.0,
	0: "dion_rank_multiple_of": 1,
	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": 1,
	0: "gradient_checkpointing": true,
	0: "gradient_checkpointing_kwargs": {
	0: "use_reentrant": true
	0: },
	0: "learning_rate": 1e-06,
	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/math/Qwen3-235B-A22B/Qwen2.5-14B_ift/0",
	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: "shuffle_before_merging_datasets": false,
	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/Qwen2.5-14B_ift",
	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": 16
	0: }[39m
	0: [2025-09-02 18:59:27,159] [INFO] [axolotl.cli.checks.check_user_token:35] [PID:2134977] [RANK:0] Skipping HuggingFace token verification because HF_HUB_OFFLINE is set to True. Only local files will be used.[39m
	3: Tokenizing Prompts (num_proc=192): 0%\| \| 0/321773 [00:00<?, ? examples/s] Tokenizing Prompts (num_proc=192): 0%\| \| 1000/321773 [00:07<41:34, 128.60 examples/s] Tokenizing Prompts (num_proc=192): 1%\| \| 2000/321773 [00:08<18:38, 285.86 examples/s] Tokenizing Prompts (num_proc=192): 1%\| \| 3000/321773 [00:08<10:21, 513.28 examples/s] Tokenizing Prompts (num_proc=192): 2%\|▏ \| 5000/321773 [00:08<04:37, 1143.09 examples/s] Tokenizing Prompts (num_proc=192): 2%\|▏ \| 7000/321773 [00:09<03:16, 1603.57 examples/s] Tokenizing Prompts (num_proc=192): 3%\|▎ \| 9000/321773 [00:09<02:05, 2491.44 examples/s] Tokenizing Prompts (num_proc=192): 3%\|▎ \| 11000/321773 [00:09<01:38, 3143.52 examples/s] Tokenizing Prompts (num_proc=192): 4%\|▎ \| 12000/321773 [00:09<01:32, 3331.12 examples/s] Tokenizing Prompts (num_proc=192): 4%\|▍ \| 13000/321773 [00:09<01:19, 3876.87 examples/s] Tokenizing Prompts (num_proc=192): 5%\|▍
	3: \| 16000/321773 [00:10<00:45, 6718.05 examples/s] Tokenizing Prompts (num_proc=192): 6%\|▌ \| 19000/321773 [00:10<00:33, 9114.77 examples/s] Tokenizing Prompts (num_proc=192): 7%\|▋ \| 23000/321773 [00:10<00:22, 13402.62 examples/s] Tokenizing Prompts (num_proc=192): 8%\|▊ \| 26000/321773 [00:10<00:25, 11522.97 examples/s] Tokenizing Prompts (num_proc=192): 9%\|▊ \| 28000/321773 [00:11<00:35, 8265.75 examples/s] Tokenizing Prompts (num_proc=192): 9%\|▉ \| 30000/321773 [00:11<00:32, 9052.47 examples/s] Tokenizing Prompts (num_proc=192): 10%\|▉ \| 32000/321773 [00:11<00:36, 8035.37 examples/s] Tokenizing Prompts (num_proc=192): 11%\|█▏ \| 37000/321773 [00:11<00:22, 12484.18 examples/s] Tokenizing Prompts (num_proc=192): 12%\|█▏ \| 39000/321773 [00:11<00:22, 12537.61 examples/s] Tokenizing Prompts (num_proc=192): 13%\|█▎ \| 41000/321773 [00:12<00:22, 12552.09 examples/s] Tokenizing Prompts (num_proc=192): 14%\|█▎
	3: \| 43676/321773 [00:12<00:20, 13559.72 examples/s] Tokenizing Prompts (num_proc=192): 14%\|█▍ \| 45676/321773 [00:12<00:37, 7369.49 examples/s] Tokenizing Prompts (num_proc=192): 15%\|█▌ \| 48676/321773 [00:13<00:34, 7925.22 examples/s] Tokenizing Prompts (num_proc=192): 16%\|█▋ \| 52352/321773 [00:13<00:25, 10493.94 examples/s] Tokenizing Prompts (num_proc=192): 17%\|█▋ \| 54352/321773 [00:13<00:24, 10924.24 examples/s] Tokenizing Prompts (num_proc=192): 18%\|█▊ \| 56352/321773 [00:13<00:28, 9289.49 examples/s] Tokenizing Prompts (num_proc=192): 18%\|█▊ \| 58028/321773 [00:14<00:30, 8695.81 examples/s] Tokenizing Prompts (num_proc=192): 19%\|█▊ \| 59704/321773 [00:14<00:27, 9699.29 examples/s] Tokenizing Prompts (num_proc=192): 19%\|█▉ \| 61056/321773 [00:14<00:26, 9991.60 examples/s] Tokenizing Prompts (num_proc=192): 20%\|█▉ \| 63056/321773 [00:14<00:31, 8158.52 examples/s] Tokenizing Prompts (num_proc=192):
	3: 20%\|██ \| 65056/321773 [00:14<00:35, 7261.23 examples/s] Tokenizing Prompts (num_proc=192): 21%\|██ \| 67732/321773 [00:15<00:30, 8367.22 examples/s] Tokenizing Prompts (num_proc=192): 22%\|██▏ \| 71084/321773 [00:15<00:21, 11697.63 examples/s] Tokenizing Prompts (num_proc=192): 23%\|██▎ \| 72760/321773 [00:15<00:20, 12142.59 examples/s] Tokenizing Prompts (num_proc=192): 23%\|██▎ \| 75436/321773 [00:15<00:21, 11235.07 examples/s] Tokenizing Prompts (num_proc=192): 24%\|██▍ \| 77436/321773 [00:15<00:21, 11479.77 examples/s] Tokenizing Prompts (num_proc=192): 24%\|██▍ \| 78788/321773 [00:16<00:22, 10774.12 examples/s] Tokenizing Prompts (num_proc=192): 25%\|██▌ \| 81140/321773 [00:16<00:20, 11729.30 examples/s] Tokenizing Prompts (num_proc=192): 26%\|██▌ \| 83816/321773 [00:16<00:18, 12957.51 examples/s] Tokenizing Prompts (num_proc=192): 27%\|██▋ \| 86492/321773 [00:16<00:17, 13785.25 examples/s] Token
	3: izing Prompts (num_proc=192): 27%\|██▋ \| 88168/321773 [00:16<00:18, 12898.81 examples/s] Tokenizing Prompts (num_proc=192): 28%\|██▊ \| 89844/321773 [00:17<00:24, 9369.35 examples/s] Tokenizing Prompts (num_proc=192): 30%\|██▉ \| 95548/321773 [00:17<00:14, 15619.35 examples/s] Tokenizing Prompts (num_proc=192): 31%\|███ \| 99576/321773 [00:17<00:12, 17957.30 examples/s] Tokenizing Prompts (num_proc=192): 32%\|███▏ \| 102252/321773 [00:17<00:12, 17425.65 examples/s] Tokenizing Prompts (num_proc=192): 33%\|███▎ \| 104604/321773 [00:17<00:14, 14836.35 examples/s] Tokenizing Prompts (num_proc=192): 33%\|███▎ \| 106632/321773 [00:17<00:15, 13804.67 examples/s] Tokenizing Prompts (num_proc=192): 34%\|███▍ \| 109308/321773 [00:18<00:13, 16051.27 examples/s] Tokenizing Prompts (num_proc=192): 35%\|███▍ \| 111660/321773 [00:18<00:18, 11633.51 examples/s] Tokenizing Prompts (num_proc=192): 35%\|███▌ \|
	3: 113660/321773 [00:18<00:17, 11793.91 examples/s] Tokenizing Prompts (num_proc=192): 36%\|███▌ \| 115660/321773 [00:19<00:26, 7858.69 examples/s] Tokenizing Prompts (num_proc=192): 36%\|███▋ \| 117336/321773 [00:19<00:24, 8183.85 examples/s] Tokenizing Prompts (num_proc=192): 38%\|███▊ \| 122040/321773 [00:19<00:22, 8757.69 examples/s] Tokenizing Prompts (num_proc=192): 39%\|███▊ \| 124392/321773 [00:19<00:20, 9764.14 examples/s] Tokenizing Prompts (num_proc=192): 39%\|███▉ \| 126068/321773 [00:20<00:19, 9786.40 examples/s] Tokenizing Prompts (num_proc=192): 40%\|███▉ \| 127420/321773 [00:20<00:20, 9373.78 examples/s] Tokenizing Prompts (num_proc=192): 40%\|████ \| 128772/321773 [00:20<00:20, 9304.94 examples/s] Tokenizing Prompts (num_proc=192): 41%\|████ \| 130448/321773 [00:20<00:20, 9437.12 examples/s] Tokenizing Prompts (num_proc=192): 41%\|████ \| 132124/321773 [00:20<00:19, 9568.90 examples/s] To
	3: kenizing Prompts (num_proc=192): 42%\|████▏ \| 134476/321773 [00:20<00:17, 10747.19 examples/s] Tokenizing Prompts (num_proc=192): 43%\|████▎ \| 137856/321773 [00:21<00:15, 11981.99 examples/s] Tokenizing Prompts (num_proc=192): 43%\|████▎ \| 139208/321773 [00:21<00:18, 9673.00 examples/s] Tokenizing Prompts (num_proc=192): 46%\|████▌ \| 146912/321773 [00:21<00:10, 17267.28 examples/s] Tokenizing Prompts (num_proc=192): 46%\|████▋ \| 149588/321773 [00:21<00:11, 14959.04 examples/s] Tokenizing Prompts (num_proc=192): 48%\|████▊ \| 153616/321773 [00:22<00:10, 15773.59 examples/s] Tokenizing Prompts (num_proc=192): 48%\|████▊ \| 155616/321773 [00:22<00:12, 13297.63 examples/s] Tokenizing Prompts (num_proc=192): 49%\|████▉ \| 158968/321773 [00:22<00:12, 13243.22 examples/s] Tokenizing Prompts (num_proc=192): 50%\|█████ \| 160996/321773 [00:22<00:13, 11915.14 examples/s] Tokenizing Prompts (num_pro
	3: c=192): 51%\|█████ \| 163996/321773 [00:23<00:13, 12083.29 examples/s] Tokenizing Prompts (num_proc=192): 51%\|█████▏ \| 165348/321773 [00:23<00:15, 9800.87 examples/s] Tokenizing Prompts (num_proc=192): 52%\|█████▏ \| 168700/321773 [00:23<00:13, 11429.76 examples/s] Tokenizing Prompts (num_proc=192): 53%\|█████▎ \| 171700/321773 [00:23<00:11, 12975.64 examples/s] Tokenizing Prompts (num_proc=192): 54%\|█████▍ \| 173376/321773 [00:23<00:12, 12174.84 examples/s] Tokenizing Prompts (num_proc=192): 55%\|█████▍ \| 176728/321773 [00:24<00:11, 12855.59 examples/s] Tokenizing Prompts (num_proc=192): 56%\|█████▌ \| 179080/321773 [00:24<00:11, 12412.05 examples/s] Tokenizing Prompts (num_proc=192): 57%\|█████▋ \| 182080/321773 [00:24<00:09, 14042.85 examples/s] Tokenizing Prompts (num_proc=192): 57%\|█████▋ \| 184756/321773 [00:24<00:10, 13372.71 examples/s] Tokenizing Prompts (num_proc=192):
	3: 59%\|█████▉ \| 189756/321773 [00:24<00:07, 17876.22 examples/s] Tokenizing Prompts (num_proc=192): 60%\|█████▉ \| 192108/321773 [00:25<00:07, 17095.18 examples/s] Tokenizing Prompts (num_proc=192): 60%\|██████ \| 194460/321773 [00:25<00:08, 14323.26 examples/s] Tokenizing Prompts (num_proc=192): 62%\|██████▏ \| 198136/321773 [00:25<00:08, 15290.31 examples/s] Tokenizing Prompts (num_proc=192): 62%\|██████▏ \| 200488/321773 [00:25<00:09, 13028.90 examples/s] Tokenizing Prompts (num_proc=192): 63%\|██████▎ \| 202164/321773 [00:25<00:09, 11997.37 examples/s] Tokenizing Prompts (num_proc=192): 63%\|██████▎ \| 203516/321773 [00:26<00:10, 11727.61 examples/s] Tokenizing Prompts (num_proc=192): 64%\|██████▎ \| 204868/321773 [00:26<00:09, 12007.68 examples/s] Tokenizing Prompts (num_proc=192): 64%\|██████▍ \| 206544/321773 [00:26<00:15, 7246.10 examples/s] Tokenizing Prompts (num_proc=19
	3: 2): 65%\|██████▌ \| 209896/321773 [00:26<00:10, 10740.11 examples/s] Tokenizing Prompts (num_proc=192): 66%\|██████▌ \| 211896/321773 [00:26<00:10, 10950.05 examples/s] Tokenizing Prompts (num_proc=192): 66%\|██████▋ \| 213896/321773 [00:27<00:09, 11728.33 examples/s] Tokenizing Prompts (num_proc=192): 67%\|██████▋ \| 216924/321773 [00:27<00:07, 13519.31 examples/s] Tokenizing Prompts (num_proc=192): 68%\|██████▊ \| 219600/321773 [00:27<00:07, 14493.64 examples/s] Tokenizing Prompts (num_proc=192): 69%\|██████▉ \| 221276/321773 [00:27<00:07, 13794.71 examples/s] Tokenizing Prompts (num_proc=192): 69%\|██████▉ \| 223276/321773 [00:27<00:06, 14092.46 examples/s] Tokenizing Prompts (num_proc=192): 70%\|██████▉ \| 224952/321773 [00:27<00:07, 13002.96 examples/s] Tokenizing Prompts (num_proc=192): 70%\|███████ \| 226628/321773 [00:27<00:07, 12991.74 examples/s] Tokenizing Prompts (
	3: num_proc=192): 72%\|███████▏ \| 231332/321773 [00:28<00:04, 19985.53 examples/s] Tokenizing Prompts (num_proc=192): 73%\|███████▎ \| 234360/321773 [00:28<00:04, 21578.81 examples/s] Tokenizing Prompts (num_proc=192): 74%\|███████▎ \| 237036/321773 [00:28<00:05, 16924.96 examples/s] Tokenizing Prompts (num_proc=192): 75%\|███████▌ \| 241388/321773 [00:28<00:03, 21813.09 examples/s] Tokenizing Prompts (num_proc=192): 76%\|███████▌ \| 245064/321773 [00:28<00:03, 23930.01 examples/s] Tokenizing Prompts (num_proc=192): 77%\|███████▋ \| 247768/321773 [00:29<00:06, 11150.30 examples/s] Tokenizing Prompts (num_proc=192): 78%\|███████▊ \| 250472/321773 [00:29<00:06, 11703.18 examples/s] Tokenizing Prompts (num_proc=192): 79%\|███████▊ \| 252824/321773 [00:29<00:07, 9437.76 examples/s] Tokenizing Prompts (num_proc=192): 79%\|███████▉ \| 255500/321773 [00:29<00:05, 11389.15 exa
	3: mples/s] Tokenizing Prompts (num_proc=192): 80%\|████████ \| 258528/321773 [00:30<00:04, 13544.75 examples/s] Tokenizing Prompts (num_proc=192): 81%\|████████ \| 260880/321773 [00:30<00:04, 13404.34 examples/s] Tokenizing Prompts (num_proc=192): 82%\|████████▏ \| 262908/321773 [00:30<00:04, 12458.45 examples/s] Tokenizing Prompts (num_proc=192): 82%\|████████▏ \| 264584/321773 [00:30<00:04, 12611.36 examples/s] Tokenizing Prompts (num_proc=192): 83%\|████████▎ \| 266612/321773 [00:30<00:05, 11030.94 examples/s] Tokenizing Prompts (num_proc=192): 83%\|████████▎ \| 268288/321773 [00:31<00:04, 10785.30 examples/s] Tokenizing Prompts (num_proc=192): 85%\|████████▍ \| 272640/321773 [00:31<00:02, 16783.18 examples/s] Tokenizing Prompts (num_proc=192): 86%\|████████▌ \| 275640/321773 [00:31<00:03, 13232.34 examples/s] Tokenizing Prompts (num_proc=192): 86%\|████████▋
	3: \| 278316/321773 [00:31<00:03, 13648.03 examples/s] Tokenizing Prompts (num_proc=192): 87%\|████████▋ \| 280668/321773 [00:31<00:03, 10898.83 examples/s] Tokenizing Prompts (num_proc=192): 88%\|████████▊ \| 282696/321773 [00:32<00:03, 10797.91 examples/s] Tokenizing Prompts (num_proc=192): 89%\|████████▊ \| 285372/321773 [00:32<00:03, 12049.84 examples/s] Tokenizing Prompts (num_proc=192): 89%\|████████▉ \| 287372/321773 [00:32<00:02, 12532.53 examples/s] Tokenizing Prompts (num_proc=192): 90%\|████████▉ \| 289048/321773 [00:32<00:02, 12836.05 examples/s] Tokenizing Prompts (num_proc=192): 91%\|█████████▏\| 293752/321773 [00:32<00:01, 17684.96 examples/s] Tokenizing Prompts (num_proc=192): 92%\|█████████▏\| 297456/321773 [00:32<00:01, 19206.11 examples/s] Tokenizing Prompts (num_proc=192): 93%\|█████████▎\| 299484/321773 [00:33<00:01, 13543.34 examples/s] Tokenizing Pro
	3: mpts (num_proc=192): 94%\|█████████▎\| 301512/321773 [00:33<00:01, 13618.95 examples/s] Tokenizing Prompts (num_proc=192): 94%\|█████████▍\| 303540/321773 [00:33<00:01, 10297.11 examples/s] Tokenizing Prompts (num_proc=192): 95%\|█████████▍\| 304892/321773 [00:34<00:02, 6584.67 examples/s] Tokenizing Prompts (num_proc=192): 95%\|█████████▌\| 306920/321773 [00:34<00:02, 5077.01 examples/s] Tokenizing Prompts (num_proc=192): 96%\|█████████▌\| 308270/321773 [00:35<00:03, 4363.32 examples/s] Tokenizing Prompts (num_proc=192): 96%\|█████████▌\| 309620/321773 [00:35<00:02, 4829.96 examples/s] Tokenizing Prompts (num_proc=192): 97%\|█████████▋\| 310970/321773 [00:35<00:01, 5517.45 examples/s] Tokenizing Prompts (num_proc=192): 97%\|█████████▋\| 312997/321773 [00:35<00:01, 7286.19 examples/s] Tokenizing Prompts (num_proc=192): 98%\|█████████▊\| 314
	3: 348/321773 [00:36<00:01, 5484.63 examples/s] Tokenizing Prompts (num_proc=192): 98%\|█████████▊\| 315698/321773 [00:36<00:01, 4839.05 examples/s] Tokenizing Prompts (num_proc=192): 99%\|█████████▊\| 317048/321773 [00:36<00:01, 4394.45 examples/s] Tokenizing Prompts (num_proc=192): 99%\|█████████▉\| 318398/321773 [00:37<00:00, 5390.08 examples/s] Tokenizing Prompts (num_proc=192): 100%\|█████████▉\| 320423/321773 [00:37<00:00, 7179.95 examples/s] Tokenizing Prompts (num_proc=192): 100%\|██████████\| 321773/321773 [00:37<00:00, 5973.17 examples/s] Tokenizing Prompts (num_proc=192): 100%\|██████████\| 321773/321773 [00:37<00:00, 8478.51 examples/s]
	3: Dropping Long Sequences (>16384) (num_proc=192): 0%\| \| 0/321773 [00:00<?, ? examples/s] Dropping Long Sequences (>16384) (num_proc=192): 0%\| \| 1000/321773 [00:01<09:44, 548.90 examples/s] Dropping Long Sequences (>16384) (num_proc=192): 5%\|▍ \| 15000/321773 [00:01<00:28, 10641.73 examples/s] Dropping Long Sequences (>16384) (num_proc=192): 8%\|▊ \| 24676/321773 [00:02<00:18, 16454.79 examples/s] Dropping Long Sequences (>16384) (num_proc=192): 10%\|▉ \| 30732/321773 [00:02<00:15, 18680.29 examples/s] Dropping Long Sequences (>16384) (num_proc=192): 11%\|█ \| 36112/321773 [00:02<00:13, 21058.71 examples/s] Dropping Long Sequences (>16384) (num_proc=192): 13%\|█▎ \| 40492/321773 [00:02<00:14, 18996.48 examples/s] Dropping Long Sequences (>16384) (num_proc=192): 14%\|█▎ \| 43844/321773 [00:03<00:14, 19470.16 examples/s] Dropping Long Sequences (>16384) (num_proc=192): 15%\|█▌ \| 48548/321773 [00:03<00:11, 23185.99 example
	3: s/s] Dropping Long Sequences (>16384) (num_proc=192): 16%\|█▌ \| 52224/321773 [00:03<00:10, 25109.15 examples/s] Dropping Long Sequences (>16384) (num_proc=192): 54%\|█████▍ \| 173756/321773 [00:03<00:00, 257188.15 examples/s] Dropping Long Sequences (>16384) (num_proc=192): 66%\|██████▋ \| 213896/321773 [00:03<00:00, 260962.67 examples/s] Dropping Long Sequences (>16384) (num_proc=192): 78%\|███████▊ \| 249796/321773 [00:03<00:00, 225600.38 examples/s] Dropping Long Sequences (>16384) (num_proc=192): 87%\|████████▋ \| 280020/321773 [00:03<00:00, 219928.40 examples/s] Dropping Long Sequences (>16384) (num_proc=192): 96%\|█████████▌\| 307593/321773 [00:04<00:00, 177091.06 examples/s] Dropping Long Sequences (>16384) (num_proc=192): 100%\|██████████\| 321773/321773 [00:04<00:00, 68433.68 examples/s]
	3: Drop Samples with Zero Trainable Tokens (num_proc=192): 0%\| \| 0/315947 [00:00<?, ? examples/s] Drop Samples with Zero Trainable Tokens (num_proc=192): 0%\| \| 1000/315947 [00:02<10:32, 497.84 examples/s] Drop Samples with Zero Trainable Tokens (num_proc=192): 6%\|▌ \| 19000/315947 [00:02<00:23, 12405.63 examples/s] Drop Samples with Zero Trainable Tokens (num_proc=192): 10%\|▉ \| 30230/315947 [00:02<00:15, 18618.36 examples/s] Drop Samples with Zero Trainable Tokens (num_proc=192): 12%\|█▏ \| 37982/315947 [00:02<00:14, 18827.53 examples/s] Drop Samples with Zero Trainable Tokens (num_proc=192): 14%\|█▍ \| 43796/315947 [00:03<00:13, 19456.77 examples/s] Drop Samples with Zero Trainable Tokens (num_proc=192): 15%\|█▌ \| 48442/315947 [00:03<00:12, 22066.58 examples/s] Drop Samples with Zero Trainable Tokens (num_proc=192): 21%\|██▏ \| 67734/315947 [00:03<00:05, 42162.92 examples/s] Drop Samples with Zero Trainable Tokens (num_proc=1
	3: 92): 24%\|██▍ \| 75610/315947 [00:03<00:05, 43374.10 examples/s] Drop Samples with Zero Trainable Tokens (num_proc=192): 26%\|██▌ \| 82548/315947 [00:03<00:05, 41894.15 examples/s] Drop Samples with Zero Trainable Tokens (num_proc=192): 28%\|██▊ \| 88486/315947 [00:03<00:05, 41648.69 examples/s] Drop Samples with Zero Trainable Tokens (num_proc=192): 30%\|██▉ \| 94070/315947 [00:03<00:05, 39887.99 examples/s] Drop Samples with Zero Trainable Tokens (num_proc=192): 31%\|███▏ \| 99362/315947 [00:04<00:05, 40725.77 examples/s] Drop Samples with Zero Trainable Tokens (num_proc=192): 33%\|███▎ \| 103946/315947 [00:04<00:05, 38635.30 examples/s] Drop Samples with Zero Trainable Tokens (num_proc=192): 34%\|███▍ \| 108884/315947 [00:04<00:05, 40702.57 examples/s] Drop Samples with Zero Trainable Tokens (num_proc=192): 40%\|███▉ \| 125344/315947 [00:04<00:02, 69558.66 examples/s] Drop Samples with Zero Trainable Tokens (num_pro
	3: c=192): 80%\|███████▉ \| 252052/315947 [00:04<00:00, 369109.63 examples/s] Drop Samples with Zero Trainable Tokens (num_proc=192): 94%\|█████████▎\| 296112/315947 [00:04<00:00, 228684.13 examples/s] Drop Samples with Zero Trainable Tokens (num_proc=192): 100%\|██████████\| 315947/315947 [00:05<00:00, 57249.08 examples/s]
	3: Add position_id column (Sample Packing) (num_proc=192): 0%\| \| 0/315947 [00:00<?, ? examples/s] Add position_id column (Sample Packing) (num_proc=192): 0%\| \| 1000/315947 [00:02<11:19, 463.69 examples/s] Add position_id column (Sample Packing) (num_proc=192): 3%\|▎ \| 10000/315947 [00:02<00:51, 5990.50 examples/s] Add position_id column (Sample Packing) (num_proc=192): 7%\|▋ \| 23000/315947 [00:02<00:18, 15986.23 examples/s] Add position_id column (Sample Packing) (num_proc=192): 11%\|█ \| 34000/315947 [00:02<00:11, 25599.61 examples/s] Add position_id column (Sample Packing) (num_proc=192): 14%\|█▍ \| 45000/315947 [00:02<00:07, 36215.98 examples/s] Add position_id column (Sample Packing) (num_proc=192): 17%\|█▋ \| 55000/315947 [00:02<00:05, 45761.18 examples/s] Add position_id column (Sample Packing) (num_proc=192): 21%\|██ \| 67000/315947 [00:02<00:04, 58922.50 examples/s] Add position_id column (Sample Packing) (num_proc=192):
	3: 25%\|██▍ \| 77938/315947 [00:02<00:04, 58589.19 examples/s] Add position_id column (Sample Packing) (num_proc=192): 28%\|██▊ \| 87106/315947 [00:03<00:05, 42119.45 examples/s] Add position_id column (Sample Packing) (num_proc=192): 40%\|████ \| 127362/315947 [00:03<00:01, 94557.94 examples/s] Add position_id column (Sample Packing) (num_proc=192): 45%\|████▌ \| 142530/315947 [00:03<00:01, 96545.75 examples/s] Add position_id column (Sample Packing) (num_proc=192): 50%\|████▉ \| 156406/315947 [00:03<00:01, 104362.52 examples/s] Add position_id column (Sample Packing) (num_proc=192): 54%\|█████▍ \| 170220/315947 [00:03<00:01, 106994.69 examples/s] Add position_id column (Sample Packing) (num_proc=192): 61%\|██████ \| 192910/315947 [00:03<00:00, 128383.41 examples/s] Add position_id column (Sample Packing) (num_proc=192): 66%\|██████▌ \| 207721/315947 [00:04<00:01, 99321.02 examples/s] Add position_id column (
	3: Sample Packing) (num_proc=192): 72%\|███████▏ \| 226407/315947 [00:04<00:00, 116855.65 examples/s] Add position_id column (Sample Packing) (num_proc=192): 77%\|███████▋ \| 243924/315947 [00:04<00:00, 127103.90 examples/s] Add position_id column (Sample Packing) (num_proc=192): 83%\|████████▎ \| 261891/315947 [00:04<00:00, 138776.78 examples/s] Add position_id column (Sample Packing) (num_proc=192): 88%\|████████▊ \| 278116/315947 [00:04<00:00, 137691.98 examples/s] Add position_id column (Sample Packing) (num_proc=192): 93%\|█████████▎\| 293372/315947 [00:04<00:00, 111443.49 examples/s] Add position_id column (Sample Packing) (num_proc=192): 97%\|█████████▋\| 306272/315947 [00:05<00:00, 97509.17 examples/s] Add position_id column (Sample Packing) (num_proc=192): 100%\|██████████\| 315947/315947 [00:05<00:00, 56185.04 examples/s]
	3: Saving the dataset (0/192 shards): 0%\| \| 0/315947 [00:00<?, ? examples/s] Saving the dataset (0/192 shards): 1%\| \| 1646/315947 [00:01<05:52, 892.52 examples/s] Saving the dataset (1/192 shards): 1%\| \| 1646/315947 [00:01<05:52, 892.52 examples/s] Saving the dataset (2/192 shards): 1%\| \| 3292/315947 [00:01<05:50, 892.52 examples/s] Saving the dataset (3/192 shards): 2%\|▏ \| 4938/315947 [00:01<05:48, 892.52 examples/s] Saving the dataset (4/192 shards): 3%\|▎ \| 8230/315947 [00:01<05:44, 892.52 examples/s] Saving the dataset (5/192 shards): 3%\|▎ \| 8230/315947 [00:01<05:44, 892.52 examples/s] Saving the dataset (6/192 shards): 3%\|▎ \| 9876/315947 [00:01<05:42, 892.52 examples/s] Saving the dataset (7/192 shards): 4%\|▎ \| 11522/315947 [00:01<05:41, 892.52 examples/s] Saving the dataset (8/192 shards): 4%\|▍ \| 13168/315947 [00:01<05:39, 892.52 examples/s] Saving the dataset (9/192 shards): 5%\|▍
	3: \| 14814/315947 [00:01<05:37, 892.52 examples/s] Saving the dataset (10/192 shards): 5%\|▌ \| 16460/315947 [00:01<05:35, 892.52 examples/s] Saving the dataset (11/192 shards): 6%\|▌ \| 18106/315947 [00:01<05:33, 892.52 examples/s] Saving the dataset (12/192 shards): 7%\|▋ \| 21398/315947 [00:01<05:30, 892.52 examples/s] Saving the dataset (13/192 shards): 7%\|▋ \| 21398/315947 [00:01<05:30, 892.52 examples/s] Saving the dataset (14/192 shards): 7%\|▋ \| 23044/315947 [00:01<05:28, 892.52 examples/s] Saving the dataset (15/192 shards): 8%\|▊ \| 24690/315947 [00:01<05:26, 892.52 examples/s] Saving the dataset (16/192 shards): 8%\|▊ \| 26336/315947 [00:01<05:24, 892.52 examples/s] Saving the dataset (17/192 shards): 9%\|▉ \| 29628/315947 [00:01<05:20, 892.52 examples/s] Saving the dataset (18/192 shards): 9%\|▉ \| 29628/315947 [00:01<05:20, 892.52 examples/s] Saving the dataset (19/192 shards): 10%\|▉ \| 31274/31594
	3: 7 [00:01<05:18, 892.52 examples/s] Saving the dataset (20/192 shards): 10%\|█ \| 32920/315947 [00:01<05:17, 892.52 examples/s] Saving the dataset (21/192 shards): 11%\|█▏ \| 36212/315947 [00:01<05:13, 892.52 examples/s] Saving the dataset (22/192 shards): 11%\|█▏ \| 36212/315947 [00:01<05:13, 892.52 examples/s] Saving the dataset (23/192 shards): 12%\|█▏ \| 37858/315947 [00:01<05:11, 892.52 examples/s] Saving the dataset (24/192 shards): 13%\|█▎ \| 41150/315947 [00:01<05:07, 892.52 examples/s] Saving the dataset (25/192 shards): 13%\|█▎ \| 41150/315947 [00:01<05:07, 892.52 examples/s] Saving the dataset (26/192 shards): 14%\|█▎ \| 42796/315947 [00:01<05:06, 892.52 examples/s] Saving the dataset (27/192 shards): 15%\|█▍ \| 46088/315947 [00:01<05:02, 892.52 examples/s] Saving the dataset (28/192 shards): 15%\|█▍ \| 46088/315947 [00:01<05:02, 892.52 examples/s] Saving the dataset (29/192 shards): 15%\|█▌ \| 47734/3
	3: 15947 [00:01<05:00, 892.52 examples/s] Saving the dataset (30/192 shards): 16%\|█▌ \| 49380/315947 [00:01<04:58, 892.52 examples/s] Saving the dataset (31/192 shards): 16%\|█▌ \| 51026/315947 [00:01<04:56, 892.52 examples/s] Saving the dataset (32/192 shards): 17%\|█▋ \| 52672/315947 [00:01<04:54, 892.52 examples/s] Saving the dataset (33/192 shards): 17%\|█▋ \| 54318/315947 [00:01<04:53, 892.52 examples/s] Saving the dataset (34/192 shards): 19%\|█▉ \| 59256/315947 [00:01<04:47, 892.52 examples/s] Saving the dataset (35/192 shards): 19%\|█▉ \| 59256/315947 [00:01<04:47, 892.52 examples/s] Saving the dataset (36/192 shards): 19%\|█▉ \| 59256/315947 [00:01<04:47, 892.52 examples/s] Saving the dataset (37/192 shards): 20%\|█▉ \| 62548/315947 [00:01<04:43, 892.52 examples/s] Saving the dataset (38/192 shards): 20%\|█▉ \| 62548/315947 [00:01<04:43, 892.52 examples/s] Saving the dataset (39/192 shards): 20%\|██ \| 6
	3: 4194/315947 [00:01<04:42, 892.52 examples/s] Saving the dataset (40/192 shards): 21%\|██ \| 65840/315947 [00:01<04:40, 892.52 examples/s] Saving the dataset (41/192 shards): 21%\|██▏ \| 67486/315947 [00:01<04:38, 892.52 examples/s] Saving the dataset (42/192 shards): 22%\|██▏ \| 69132/315947 [00:01<04:36, 892.52 examples/s] Saving the dataset (43/192 shards): 23%\|██▎ \| 72424/315947 [00:01<04:32, 892.52 examples/s] Saving the dataset (44/192 shards): 23%\|██▎ \| 72424/315947 [00:01<04:32, 892.52 examples/s] Saving the dataset (45/192 shards): 23%\|██▎ \| 74070/315947 [00:01<04:31, 892.52 examples/s] Saving the dataset (46/192 shards): 24%\|██▍ \| 77362/315947 [00:01<04:27, 892.52 examples/s] Saving the dataset (47/192 shards): 24%\|██▍ \| 77362/315947 [00:01<04:27, 892.52 examples/s] Saving the dataset (48/192 shards): 25%\|██▌ \| 79008/315947 [00:01<04:25, 892.52 examples/s] Saving the dataset (49/192 shards):
	3: 26%\|██▌ \| 80654/315947 [00:01<04:23, 892.52 examples/s] Saving the dataset (50/192 shards): 26%\|██▌ \| 82300/315947 [00:01<04:21, 892.52 examples/s] Saving the dataset (51/192 shards): 27%\|██▋ \| 83946/315947 [00:01<04:19, 892.52 examples/s] Saving the dataset (52/192 shards): 27%\|██▋ \| 85592/315947 [00:01<04:18, 892.52 examples/s] Saving the dataset (53/192 shards): 28%\|██▊ \| 87238/315947 [00:01<04:16, 892.52 examples/s] Saving the dataset (54/192 shards): 28%\|██▊ \| 88884/315947 [00:01<04:14, 892.52 examples/s] Saving the dataset (55/192 shards): 29%\|██▉ \| 92176/315947 [00:01<04:10, 892.52 examples/s] Saving the dataset (56/192 shards): 29%\|██▉ \| 92176/315947 [00:01<04:10, 892.52 examples/s] Saving the dataset (57/192 shards): 30%\|██▉ \| 93822/315947 [00:01<04:08, 892.52 examples/s] Saving the dataset (58/192 shards): 30%\|███ \| 95468/315947 [00:01<04:07, 892.52 examples/s] Saving the
	3: dataset (59/192 shards): 32%\|███▏ \| 100406/315947 [00:01<04:01, 892.52 examples/s] Saving the dataset (60/192 shards): 32%\|███▏ \| 100406/315947 [00:01<04:01, 892.52 examples/s] Saving the dataset (61/192 shards): 32%\|███▏ \| 100406/315947 [00:01<04:01, 892.52 examples/s] Saving the dataset (62/192 shards): 32%\|███▏ \| 102052/315947 [00:01<03:59, 892.52 examples/s] Saving the dataset (63/192 shards): 33%\|███▎ \| 103698/315947 [00:01<03:57, 892.52 examples/s] Saving the dataset (64/192 shards): 34%\|███▍ \| 106990/315947 [00:01<03:54, 892.52 examples/s] Saving the dataset (65/192 shards): 34%\|███▍ \| 106990/315947 [00:01<03:54, 892.52 examples/s] Saving the dataset (66/192 shards): 34%\|███▍ \| 108636/315947 [00:01<03:52, 892.52 examples/s] Saving the dataset (67/192 shards): 36%\|███▌ \| 113574/315947 [00:01<03:46, 892.52 examples/s] Saving the dataset (68/192 shards): 36%\|███▌ \| 1
	3: 13574/315947 [00:01<03:46, 892.52 examples/s] Saving the dataset (69/192 shards): 36%\|███▌ \| 113574/315947 [00:01<03:46, 892.52 examples/s] Saving the dataset (70/192 shards): 36%\|███▋ \| 115220/315947 [00:01<03:44, 892.52 examples/s] Saving the dataset (71/192 shards): 38%\|███▊ \| 118512/315947 [00:01<03:41, 892.52 examples/s] Saving the dataset (72/192 shards): 38%\|███▊ \| 118512/315947 [00:01<03:41, 892.52 examples/s] Saving the dataset (73/192 shards): 39%\|███▊ \| 121804/315947 [00:01<03:37, 892.52 examples/s] Saving the dataset (74/192 shards): 39%\|███▊ \| 121804/315947 [00:01<03:37, 892.52 examples/s] Saving the dataset (75/192 shards): 39%\|███▉ \| 123450/315947 [00:01<03:35, 892.52 examples/s] Saving the dataset (76/192 shards): 40%\|███▉ \| 125096/315947 [00:01<03:33, 892.52 examples/s] Saving the dataset (77/192 shards): 40%\|████ \| 126742/315947 [00:01<03:31, 892.52 examples/s] Saving
	3: the dataset (78/192 shards): 41%\|████ \| 130034/315947 [00:01<03:28, 892.52 examples/s] Saving the dataset (79/192 shards): 41%\|████ \| 130034/315947 [00:01<03:28, 892.52 examples/s] Saving the dataset (80/192 shards): 42%\|████▏ \| 131680/315947 [00:01<03:26, 892.52 examples/s] Saving the dataset (81/192 shards): 42%\|████▏ \| 133326/315947 [00:01<03:24, 892.52 examples/s] Saving the dataset (82/192 shards): 43%\|████▎ \| 134972/315947 [00:01<03:22, 892.52 examples/s] Saving the dataset (83/192 shards): 43%\|████▎ \| 136618/315947 [00:01<03:20, 892.52 examples/s] Saving the dataset (84/192 shards): 44%\|████▍ \| 138264/315947 [00:01<03:19, 892.52 examples/s] Saving the dataset (85/192 shards): 44%\|████▍ \| 139910/315947 [00:01<03:17, 892.52 examples/s] Saving the dataset (86/192 shards): 45%\|████▍ \| 141556/315947 [00:01<03:15, 892.52 examples/s] Saving the dataset (87/192 shards): 46%\|█
	3: ███▌ \| 144848/315947 [00:01<03:11, 892.52 examples/s] Saving the dataset (88/192 shards): 46%\|████▌ \| 144848/315947 [00:01<03:11, 892.52 examples/s] Saving the dataset (89/192 shards): 46%\|████▋ \| 146494/315947 [00:01<03:09, 892.52 examples/s] Saving the dataset (90/192 shards): 47%\|████▋ \| 148140/315947 [00:01<03:08, 892.52 examples/s] Saving the dataset (91/192 shards): 47%\|████▋ \| 149786/315947 [00:01<03:06, 892.52 examples/s] Saving the dataset (92/192 shards): 48%\|████▊ \| 151432/315947 [00:01<03:04, 892.52 examples/s] Saving the dataset (93/192 shards): 48%\|████▊ \| 153078/315947 [00:01<03:02, 892.52 examples/s] Saving the dataset (94/192 shards): 49%\|████▉ \| 154724/315947 [00:01<03:00, 892.52 examples/s] Saving the dataset (95/192 shards): 50%\|█████ \| 158016/315947 [00:01<02:56, 892.52 examples/s] Saving the dataset (96/192 shards): 50%\|█████ \| 158016/315947 [
	3: 00:01<02:56, 892.52 examples/s] Saving the dataset (97/192 shards): 51%\|█████ \| 159662/315947 [00:01<02:55, 892.52 examples/s] Saving the dataset (98/192 shards): 51%\|█████ \| 161308/315947 [00:01<02:53, 892.52 examples/s] Saving the dataset (99/192 shards): 52%\|█████▏ \| 162954/315947 [00:01<02:51, 892.52 examples/s] Saving the dataset (100/192 shards): 52%\|█████▏ \| 164600/315947 [00:01<02:49, 892.52 examples/s] Saving the dataset (101/192 shards): 53%\|█████▎ \| 166246/315947 [00:01<02:47, 892.52 examples/s] Saving the dataset (102/192 shards): 53%\|█████▎ \| 167892/315947 [00:01<02:45, 892.52 examples/s] Saving the dataset (103/192 shards): 54%\|█████▍ \| 171184/315947 [00:01<02:42, 892.52 examples/s] Saving the dataset (104/192 shards): 54%\|█████▍ \| 171184/315947 [00:01<02:42, 892.52 examples/s] Saving the dataset (105/192 shards): 55%\|█████▍ \| 172830/315947 [00:01<02:40, 8
	3: 92.52 examples/s] Saving the dataset (106/192 shards): 55%\|█████▌ \| 174476/315947 [00:01<02:38, 892.52 examples/s] Saving the dataset (107/192 shards): 56%\|█████▌ \| 176122/315947 [00:01<02:36, 892.52 examples/s] Saving the dataset (108/192 shards): 56%\|█████▋ \| 177767/315947 [00:01<02:34, 892.52 examples/s] Saving the dataset (109/192 shards): 57%\|█████▋ \| 179412/315947 [00:01<02:32, 892.52 examples/s] Saving the dataset (110/192 shards): 58%\|█████▊ \| 182702/315947 [00:01<02:29, 892.52 examples/s] Saving the dataset (111/192 shards): 58%\|█████▊ \| 182702/315947 [00:01<02:29, 892.52 examples/s] Saving the dataset (112/192 shards): 58%\|█████▊ \| 184347/315947 [00:01<02:27, 892.52 examples/s] Saving the dataset (113/192 shards): 59%\|█████▉ \| 187637/315947 [00:01<02:23, 892.52 examples/s] Saving the dataset (114/192 shards): 59%\|█████▉ \| 187637/315947 [00:01<02:23, 892.52 e
	3: xamples/s] Saving the dataset (115/192 shards): 60%\|█████▉ \| 189282/315947 [00:01<02:21, 892.52 examples/s] Saving the dataset (116/192 shards): 60%\|██████ \| 190927/315947 [00:01<02:20, 892.52 examples/s] Saving the dataset (117/192 shards): 61%\|██████ \| 192572/315947 [00:01<02:18, 892.52 examples/s] Saving the dataset (118/192 shards): 61%\|██████▏ \| 194217/315947 [00:01<02:16, 892.52 examples/s] Saving the dataset (119/192 shards): 62%\|██████▏ \| 195862/315947 [00:01<02:14, 892.52 examples/s] Saving the dataset (120/192 shards): 63%\|██████▎ \| 197507/315947 [00:01<02:12, 892.52 examples/s] Saving the dataset (121/192 shards): 64%\|██████▎ \| 200797/315947 [00:01<02:09, 892.52 examples/s] Saving the dataset (122/192 shards): 64%\|██████▎ \| 200797/315947 [00:01<02:09, 892.52 examples/s] Saving the dataset (123/192 shards): 64%\|██████▍ \| 202442/315947 [00:01<02:07, 892
	3: .52 examples/s] Saving the dataset (124/192 shards): 65%\|██████▍ \| 204087/315947 [00:01<02:05, 892.52 examples/s] Saving the dataset (125/192 shards): 65%\|██████▌ \| 205732/315947 [00:01<02:03, 892.52 examples/s] Saving the dataset (126/192 shards): 66%\|██████▌ \| 207377/315947 [00:01<02:01, 892.52 examples/s] Saving the dataset (127/192 shards): 66%\|██████▌ \| 209022/315947 [00:01<01:59, 892.52 examples/s] Saving the dataset (128/192 shards): 67%\|██████▋ \| 210667/315947 [00:01<01:57, 892.52 examples/s] Saving the dataset (129/192 shards): 67%\|██████▋ \| 212312/315947 [00:01<01:56, 892.52 examples/s] Saving the dataset (130/192 shards): 68%\|██████▊ \| 213957/315947 [00:01<01:54, 892.52 examples/s] Saving the dataset (131/192 shards): 68%\|██████▊ \| 215602/315947 [00:01<01:52, 892.52 examples/s] Saving the dataset (132/192 shards): 69%\|██████▉ \| 217247/315947 [00:01
	3: <01:50, 892.52 examples/s] Saving the dataset (133/192 shards): 69%\|██████▉ \| 218892/315947 [00:01<01:48, 892.52 examples/s] Saving the dataset (134/192 shards): 70%\|██████▉ \| 220537/315947 [00:01<01:46, 892.52 examples/s] Saving the dataset (135/192 shards): 70%\|███████ \| 222182/315947 [00:01<01:45, 892.52 examples/s] Saving the dataset (136/192 shards): 71%\|███████ \| 223827/315947 [00:01<01:43, 892.52 examples/s] Saving the dataset (137/192 shards): 71%\|███████▏ \| 225472/315947 [00:01<01:41, 892.52 examples/s] Saving the dataset (138/192 shards): 72%\|███████▏ \| 227117/315947 [00:01<01:39, 892.52 examples/s] Saving the dataset (139/192 shards): 72%\|███████▏ \| 228762/315947 [00:01<01:37, 892.52 examples/s] Saving the dataset (140/192 shards): 73%\|███████▎ \| 230407/315947 [00:01<01:35, 892.52 examples/s] Saving the dataset (141/192 shards): 73%\|███████▎ \|
	3: 232052/315947 [00:01<01:33, 892.52 examples/s] Saving the dataset (142/192 shards): 74%\|███████▍ \| 233697/315947 [00:01<01:32, 892.52 examples/s] Saving the dataset (143/192 shards): 74%\|███████▍ \| 235342/315947 [00:01<01:30, 892.52 examples/s] Saving the dataset (144/192 shards): 75%\|███████▌ \| 236987/315947 [00:01<01:28, 892.52 examples/s] Saving the dataset (145/192 shards): 76%\|███████▌ \| 238632/315947 [00:01<01:26, 892.52 examples/s] Saving the dataset (146/192 shards): 76%\|███████▌ \| 240277/315947 [00:01<01:24, 892.52 examples/s] Saving the dataset (147/192 shards): 77%\|███████▋ \| 241922/315947 [00:01<01:22, 892.52 examples/s] Saving the dataset (148/192 shards): 77%\|███████▋ \| 243567/315947 [00:01<01:21, 892.52 examples/s] Saving the dataset (149/192 shards): 78%\|███████▊ \| 245212/315947 [00:01<01:19, 892.52 examples/s] Saving the dataset (150/192 shards): 78
	3: %\|███████▊ \| 246857/315947 [00:01<01:17, 892.52 examples/s] Saving the dataset (151/192 shards): 79%\|███████▊ \| 248502/315947 [00:01<01:15, 892.52 examples/s] Saving the dataset (152/192 shards): 80%\|███████▉ \| 251792/315947 [00:01<01:11, 892.52 examples/s] Saving the dataset (153/192 shards): 80%\|███████▉ \| 251792/315947 [00:01<01:11, 892.52 examples/s] Saving the dataset (154/192 shards): 80%\|████████ \| 253437/315947 [00:01<01:10, 892.52 examples/s] Saving the dataset (155/192 shards): 81%\|████████ \| 255082/315947 [00:01<01:08, 892.52 examples/s] Saving the dataset (156/192 shards): 81%\|████████▏ \| 256727/315947 [00:01<01:06, 892.52 examples/s] Saving the dataset (157/192 shards): 82%\|████████▏ \| 258372/315947 [00:01<01:04, 892.52 examples/s] Saving the dataset (158/192 shards): 82%\|████████▏ \| 260017/315947 [00:01<01:02, 892.52 examples/s] Savin
	3: g the dataset (159/192 shards): 83%\|████████▎ \| 261662/315947 [00:01<01:00, 892.52 examples/s] Saving the dataset (160/192 shards): 83%\|████████▎ \| 263307/315947 [00:01<00:58, 892.52 examples/s] Saving the dataset (161/192 shards): 84%\|████████▍ \| 264952/315947 [00:01<00:57, 892.52 examples/s] Saving the dataset (162/192 shards): 84%\|████████▍ \| 266597/315947 [00:01<00:55, 892.52 examples/s] Saving the dataset (163/192 shards): 85%\|████████▍ \| 268242/315947 [00:01<00:53, 892.52 examples/s] Saving the dataset (164/192 shards): 85%\|████████▌ \| 269887/315947 [00:01<00:51, 892.52 examples/s] Saving the dataset (165/192 shards): 86%\|████████▌ \| 271532/315947 [00:01<00:49, 892.52 examples/s] Saving the dataset (166/192 shards): 86%\|████████▋ \| 273177/315947 [00:01<00:47, 892.52 examples/s] Saving the dataset (167/192 shards): 87%\|████████▋ \| 27482
	3: 2/315947 [00:01<00:46, 892.52 examples/s] Saving the dataset (168/192 shards): 88%\|████████▊ \| 276467/315947 [00:01<00:44, 892.52 examples/s] Saving the dataset (169/192 shards): 88%\|████████▊ \| 278112/315947 [00:01<00:42, 892.52 examples/s] Saving the dataset (170/192 shards): 89%\|████████▊ \| 279757/315947 [00:01<00:40, 892.52 examples/s] Saving the dataset (171/192 shards): 89%\|████████▉ \| 281402/315947 [00:01<00:38, 892.52 examples/s] Saving the dataset (172/192 shards): 90%\|████████▉ \| 283047/315947 [00:01<00:36, 892.52 examples/s] Saving the dataset (173/192 shards): 90%\|█████████ \| 284692/315947 [00:01<00:35, 892.52 examples/s] Saving the dataset (174/192 shards): 91%\|█████████ \| 286337/315947 [00:01<00:33, 892.52 examples/s] Saving the dataset (175/192 shards): 91%\|█████████ \| 287982/315947 [00:01<00:31, 892.52 examples/s] Saving the dataset (176/192 sh
	3: ards): 92%\|█████████▏\| 289627/315947 [00:01<00:29, 892.52 examples/s] Saving the dataset (177/192 shards): 92%\|█████████▏\| 291272/315947 [00:01<00:27, 892.52 examples/s] Saving the dataset (178/192 shards): 93%\|█████████▎\| 292917/315947 [00:01<00:25, 892.52 examples/s] Saving the dataset (179/192 shards): 93%\|█████████▎\| 294562/315947 [00:01<00:23, 892.52 examples/s] Saving the dataset (180/192 shards): 94%\|█████████▍\| 296207/315947 [00:01<00:22, 892.52 examples/s] Saving the dataset (181/192 shards): 94%\|█████████▍\| 297852/315947 [00:01<00:20, 892.52 examples/s] Saving the dataset (182/192 shards): 95%\|█████████▍\| 299497/315947 [00:01<00:18, 892.52 examples/s] Saving the dataset (183/192 shards): 95%\|█████████▌\| 301142/315947 [00:01<00:16, 892.52 examples/s] Saving the dataset (184/192 shards): 96%\|█████████▌\| 302787/31594
	3: 7 [00:01<00:14, 892.52 examples/s] Saving the dataset (185/192 shards): 96%\|█████████▋\| 304432/315947 [00:01<00:12, 892.52 examples/s] Saving the dataset (186/192 shards): 97%\|█████████▋\| 307722/315947 [00:01<00:09, 892.52 examples/s] Saving the dataset (187/192 shards): 97%\|█████████▋\| 307722/315947 [00:01<00:09, 892.52 examples/s] Saving the dataset (188/192 shards): 98%\|█████████▊\| 309367/315947 [00:01<00:07, 892.52 examples/s] Saving the dataset (189/192 shards): 98%\|█████████▊\| 311012/315947 [00:01<00:05, 892.52 examples/s] Saving the dataset (190/192 shards): 99%\|█████████▉\| 312657/315947 [00:01<00:03, 892.52 examples/s] Saving the dataset (191/192 shards): 99%\|█████████▉\| 314302/315947 [00:01<00:01, 892.52 examples/s] Saving the dataset (192/192 shards): 100%\|██████████\| 315947/315947 [00:01<00:00, 892.52 examples/s] Saving the dataset (1
	3: 92/192 shards): 100%\|██████████\| 315947/315947 [00:01<00:00, 158524.02 examples/s]
	0: [2025-09-02 19:00:29,431] [INFO] [axolotl.utils.data.shared.load_preprocessed_dataset:472] [PID:2134977] [RANK:0] Loading prepared dataset from disk at /lustre/fsn1/projects/rech/dgo/udv55np/dataset_math/Qwen3-235B-A22B/0/8bdfa514f4bdb23bcb10d93ba101ff0f...[39m
	0: [2025-09-02 19:01:04,779] [INFO] [axolotl.utils.samplers.multipack.calc_min_len:436] [PID:2134977] [RANK:0] gather_len_batches: [25939, 25939, 25939, 25939, 25940, 25939, 25939, 25940, 25939, 25937, 25939, 25940, 25940, 25939, 25939, 25939][39m
	0: [2025-09-02 19:01:04,844] [INFO] [axolotl.utils.trainer.calc_sample_packing_eff_est:495] [PID:2134977] [RANK:0] sample_packing_eff_est across ranks: [0.9965166449546814, 0.9965550303459167, 0.9965550303459167, 0.996631920337677, 0.9965934753417969, 0.9964781999588013, 0.9965166449546814, 0.9965934753417969, 0.9965934753417969, 0.9965934753417969, 0.9965550303459167, 0.9965550303459167, 0.9965550303459167, 0.9965550303459167, 0.9965550303459167, 0.9965550303459167][39m
	0: [2025-09-02 19:01:04,850] [INFO] [axolotl.utils.data.sft._prepare_standard_dataset:127] [PID:2134977] [RANK:0] Maximum number of steps set at 1621[39m
	0: [2025-09-02 19:01:05,194] [INFO] [axolotl.monkeypatch.transformers.trainer_loss_calc.patch_evaluation_loop:110] [PID:2134977] [RANK:0] Patched Trainer.evaluation_loop with nanmean loss calculation[39m
	0: [2025-09-02 19:01:05,195] [INFO] [axolotl.monkeypatch.transformers.trainer_loss_calc.patch_maybe_log_save_evaluate:164] [PID:2134977] [RANK:0] Patched Trainer._maybe_log_save_evaluate with nanmean loss calculation[39m
	0: Loading checkpoint shards: 0%\| \| 0/6 [00:00<?, ?it/s] Loading checkpoint shards: 0%\| \| 0/6 [00:00<?, ?it/s] Loading checkpoint shards: 0%\| \| 0/6 [00:00<?, ?it/s] Loading checkpoint shards: 0%\| \| 0/6 [00:00<?, ?it/s] Loading checkpoint shards: 17%\|█▋ \| 1/6 [00:07<00:36, 7.20s/it] Loading checkpoint shards: 17%\|█▋ \| 1/6 [00:07<00:36, 7.21s/it] Loading checkpoint shards: 17%\|█▋ \| 1/6 [00:07<00:36, 7.20s/it] Loading checkpoint shards: 17%\|█▋ \| 1/6 [00:07<00:36, 7.31s/it] Loading checkpoint shards: 33%\|███▎ \| 2/6 [00:18<00:38, 9.62s/it] Loading checkpoint shards: 33%\|███▎ \| 2/6 [00:18<00:38, 9.62s/it] Loading checkpoint shards: 33%\|███▎ \| 2/6 [00:18<00:38, 9.62s/it] Loading checkpoint shards: 33%\|███▎ \| 2/6 [00:18<00:38, 9.66s/it] Loading checkpoint shards: 50%\|█████ \| 3/6 [00:28<00:29, 9.77s/it] Loading checkpoint shards: 50%\|█████
	2: Loading checkpoint shards: 0%\| \| 0/6 [00:00<?, ?it/s] Loading checkpoint shards: 0%\| \| 0/6 [00:00<?, ?it/s] Loading checkpoint shards: 0%\| \| 0/6 [00:00<?, ?it/s] Loading checkpoint shards: 0%\| \| 0/6 [00:00<?, ?it/s] Loading checkpoint shards: 17%\|█▋ \| 1/6 [00:07<00:35, 7.20s/it] Loading checkpoint shards: 17%\|█▋ \| 1/6 [00:07<00:36, 7.20s/it] Loading checkpoint shards: 17%\|█▋ \| 1/6 [00:07<00:35, 7.20s/it] Loading checkpoint shards: 17%\|█▋ \| 1/6 [00:07<00:36, 7.20s/it] Loading checkpoint shards: 33%\|███▎ \| 2/6 [00:18<00:38, 9.62s/it] Loading checkpoint shards: 33%\|███▎ \| 2/6 [00:18<00:38, 9.62s/it] Loading checkpoint shards: 33%\|███▎ \| 2/6 [00:18<00:38, 9.62s/it] Loading checkpoint shards: 33%\|███▎ \| 2/6 [00:18<00:38, 9.62s/it] Loading checkpoint shards: 50%\|█████ \| 3/6 [00:28<00:29, 9.77s/it] Loading checkpoint shards: 50%\|█████
	3: Loading checkpoint shards: 0%\| \| 0/6 [00:00<?, ?it/s] Loading checkpoint shards: 0%\| \| 0/6 [00:00<?, ?it/s] Loading checkpoint shards: 0%\| \| 0/6 [00:00<?, ?it/s] Loading checkpoint shards: 0%\| \| 0/6 [00:00<?, ?it/s] Loading checkpoint shards: 17%\|█▋ \| 1/6 [00:07<00:36, 7.23s/it] Loading checkpoint shards: 17%\|█▋ \| 1/6 [00:07<00:36, 7.23s/it] Loading checkpoint shards: 17%\|█▋ \| 1/6 [00:07<00:36, 7.23s/it] Loading checkpoint shards: 17%\|█▋ \| 1/6 [00:07<00:36, 7.23s/it] Loading checkpoint shards: 33%\|███▎ \| 2/6 [00:18<00:38, 9.63s/it] Loading checkpoint shards: 33%\|███▎ \| 2/6 [00:18<00:38, 9.63s/it] Loading checkpoint shards: 33%\|███▎ \| 2/6 [00:18<00:38, 9.63s/it] Loading checkpoint shards: 33%\|███▎ \| 2/6 [00:18<00:38, 9.63s/it] Loading checkpoint shards: 50%\|█████ \| 3/6 [00:28<00:29, 9.78s/it] Loading checkpoint shards: 50%\|█████
	1: Loading checkpoint shards: 0%\| \| 0/6 [00:00<?, ?it/s] Loading checkpoint shards: 0%\| \| 0/6 [00:00<?, ?it/s] Loading checkpoint shards: 0%\| \| 0/6 [00:00<?, ?it/s] Loading checkpoint shards: 0%\| \| 0/6 [00:00<?, ?it/s] Loading checkpoint shards: 17%\|█▋ \| 1/6 [00:07<00:35, 7.19s/it] Loading checkpoint shards: 17%\|█▋ \| 1/6 [00:07<00:35, 7.19s/it] Loading checkpoint shards: 17%\|█▋ \| 1/6 [00:07<00:35, 7.19s/it] Loading checkpoint shards: 17%\|█▋ \| 1/6 [00:07<00:35, 7.19s/it] Loading checkpoint shards: 33%\|███▎ \| 2/6 [00:18<00:38, 9.61s/it] Loading checkpoint shards: 33%\|███▎ \| 2/6 [00:18<00:38, 9.62s/it] Loading checkpoint shards: 33%\|███▎ \| 2/6 [00:18<00:38, 9.61s/it] Loading checkpoint shards: 33%\|███▎ \| 2/6 [00:18<00:38, 9.62s/it] Loading checkpoint shards: 50%\|█████ \| 3/6 [00:28<00:29, 9.77s/it] Loading checkpoint shards: 50%\|█████
	0: \| 3/6 [00:28<00:29, 9.77s/it] Loading checkpoint shards: 50%\|█████ \| 3/6 [00:28<00:29, 9.77s/it] Loading checkpoint shards: 50%\|█████ \| 3/6 [00:28<00:29, 9.79s/it] Loading checkpoint shards: 67%\|██████▋ \| 4/6 [00:38<00:19, 9.97s/it] Loading checkpoint shards: 67%\|██████▋ \| 4/6 [00:38<00:19, 9.97s/it] Loading checkpoint shards: 67%\|██████▋ \| 4/6 [00:38<00:19, 9.97s/it] Loading checkpoint shards: 67%\|██████▋ \| 4/6 [00:38<00:19, 9.99s/it] Loading checkpoint shards: 83%\|████████▎ \| 5/6 [00:50<00:10, 10.62s/it] Loading checkpoint shards: 83%\|████████▎ \| 5/6 [00:50<00:10, 10.62s/it] Loading checkpoint shards: 83%\|████████▎ \| 5/6 [00:50<00:10, 10.62s/it] Loading checkpoint shards: 83%\|████████▎ \| 5/6 [00:50<00:10, 10.62s/it] Loading checkpoint shards: 100%\|██████████\| 6/6 [00:57<00:00, 9.53s/it] Loading checkpoint
	0: shards: 100%\|██████████\| 6/6 [00:57<00:00, 9.53s/it] Loading checkpoint shards: 100%\|██████████\| 6/6 [00:57<00:00, 9.66s/it]
	0: Loading checkpoint shards: 100%\|██████████\| 6/6 [00:57<00:00, 9.66s/it]
	2: \| 3/6 [00:28<00:29, 9.77s/it] Loading checkpoint shards: 50%\|█████ \| 3/6 [00:28<00:29, 9.77s/it] Loading checkpoint shards: 50%\|█████ \| 3/6 [00:28<00:29, 9.77s/it] Loading checkpoint shards: 67%\|██████▋ \| 4/6 [00:38<00:19, 9.97s/it] Loading checkpoint shards: 67%\|██████▋ \| 4/6 [00:38<00:19, 9.97s/it] Loading checkpoint shards: 67%\|██████▋ \| 4/6 [00:38<00:19, 9.97s/it] Loading checkpoint shards: 67%\|██████▋ \| 4/6 [00:38<00:19, 9.97s/it] Loading checkpoint shards: 83%\|████████▎ \| 5/6 [00:50<00:10, 10.62s/it] Loading checkpoint shards: 83%\|████████▎ \| 5/6 [00:50<00:10, 10.62s/it] Loading checkpoint shards: 83%\|████████▎ \| 5/6 [00:50<00:10, 10.62s/it] Loading checkpoint shards: 83%\|████████▎ \| 5/6 [00:50<00:10, 10.62s/it] Loading checkpoint shards: 100%\|██████████\| 6/6 [00:57<00:00, 9.53s/it] Loading checkpoint
	2: shards: 100%\|██████████\| 6/6 [00:57<00:00, 9.53s/it] Loading checkpoint shards: 100%\|██████████\| 6/6 [00:57<00:00, 9.66s/it]
	1: \| 3/6 [00:28<00:29, 9.77s/it] Loading checkpoint shards: 50%\|█████ \| 3/6 [00:28<00:29, 9.77s/it] Loading checkpoint shards: 50%\|█████ \| 3/6 [00:28<00:29, 9.77s/it] Loading checkpoint shards: 67%\|██████▋ \| 4/6 [00:38<00:19, 9.97s/it] Loading checkpoint shards: 67%\|██████▋ \| 4/6 [00:38<00:19, 9.97s/it] Loading checkpoint shards: 67%\|██████▋ \| 4/6 [00:38<00:19, 9.97s/it] Loading checkpoint shards: 67%\|██████▋ \| 4/6 [00:38<00:19, 9.97s/it] Loading checkpoint shards: 83%\|████████▎ \| 5/6 [00:50<00:10, 10.62s/it] Loading checkpoint shards: 83%\|████████▎ \| 5/6 [00:50<00:10, 10.62s/it] Loading checkpoint shards: 83%\|████████▎ \| 5/6 [00:50<00:10, 10.62s/it] Loading checkpoint shards: 83%\|████████▎ \| 5/6 [00:50<00:10, 10.62s/it] Loading checkpoint shards: 100%\|██████████\| 6/6 [00:57<00:00, 9.53s/it] Loading checkpoint
	2: Loading checkpoint shards: 100%\|██████████\| 6/6 [00:57<00:00, 9.66s/it]
	1: shards: 100%\|██████████\| 6/6 [00:57<00:00, 9.53s/it] Loading checkpoint shards: 100%\|██████████\| 6/6 [00:57<00:00, 9.65s/it]
	1: Loading checkpoint shards: 100%\|██████████\| 6/6 [00:57<00:00, 9.65s/it]
	2: Loading checkpoint shards: 100%\|██████████\| 6/6 [00:57<00:00, 9.53s/it] Loading checkpoint shards: 100%\|██████████\| 6/6 [00:57<00:00, 9.66s/it]
	3: \| 3/6 [00:28<00:29, 9.78s/it] Loading checkpoint shards: 50%\|█████ \| 3/6 [00:28<00:29, 9.78s/it] Loading checkpoint shards: 50%\|█████ \| 3/6 [00:28<00:29, 9.78s/it] Loading checkpoint shards: 67%\|██████▋ \| 4/6 [00:38<00:19, 9.98s/it] Loading checkpoint shards: 67%\|██████▋ \| 4/6 [00:38<00:19, 9.98s/it] Loading checkpoint shards: 67%\|██████▋ \| 4/6 [00:38<00:19, 9.98s/it] Loading checkpoint shards: 67%\|██████▋ \| 4/6 [00:38<00:19, 9.98s/it] Loading checkpoint shards: 83%\|████████▎ \| 5/6 [00:50<00:10, 10.62s/it] Loading checkpoint shards: 83%\|████████▎ \| 5/6 [00:50<00:10, 10.62s/it] Loading checkpoint shards: 83%\|████████▎ \| 5/6 [00:50<00:10, 10.62s/it] Loading checkpoint shards: 83%\|████████▎ \| 5/6 [00:50<00:10, 10.62s/it] Loading checkpoint shards: 100%\|██████████\| 6/6 [00:57<00:00, 9.54s/it] Loading checkpoint
	0: Loading checkpoint shards: 100%\|██████████\| 6/6 [00:57<00:00, 9.53s/it] Loading checkpoint shards: 100%\|██████████\| 6/6 [00:57<00:00, 9.66s/it]
	1: Loading checkpoint shards: 100%\|██████████\| 6/6 [00:57<00:00, 9.53s/it] Loading checkpoint shards: 100%\|██████████\| 6/6 [00:57<00:00, 9.65s/it]
	3: shards: 100%\|██████████\| 6/6 [00:57<00:00, 9.54s/it] Loading checkpoint shards: 100%\|██████████\| 6/6 [00:57<00:00, 9.66s/it]
	3: Loading checkpoint shards: 100%\|██████████\| 6/6 [00:57<00:00, 9.66s/it]
	3: Loading checkpoint shards: 100%\|██████████\| 6/6 [00:57<00:00, 9.54s/it] Loading checkpoint shards: 100%\|██████████\| 6/6 [00:57<00:00, 9.66s/it]
	1: Loading checkpoint shards: 100%\|██████████\| 6/6 [00:57<00:00, 9.53s/it] Loading checkpoint shards: 100%\|██████████\| 6/6 [00:57<00:00, 9.65s/it]
	3: Loading checkpoint shards: 100%\|██████████\| 6/6 [00:57<00:00, 9.54s/it] Loading checkpoint shards: 100%\|██████████\| 6/6 [00:57<00:00, 9.66s/it]
	2: Loading checkpoint shards: 100%\|██████████\| 6/6 [00:57<00:00, 9.53s/it] Loading checkpoint shards: 100%\|██████████\| 6/6 [00:57<00:00, 9.66s/it]
	0: Loading checkpoint shards: 100%\|██████████\| 6/6 [00:59<00:00, 10.18s/it] Loading checkpoint shards: 100%\|██████████\| 6/6 [00:59<00:00, 9.99s/it]
	0: [2025-09-02 19:02:05,501] [INFO] [axolotl.loaders.model._configure_embedding_dtypes:345] [PID:2134977] [RANK:0] Converting modules to torch.bfloat16[39m
	0: [2025-09-02 19:02:09,288] [INFO] [axolotl.train.save_initial_configs:416] [PID:2134977] [RANK:0] Pre-saving tokenizer to /lustre/fswork/projects/rech/dgo/udv55np/math/Qwen3-235B-A22B/Qwen2.5-14B_ift/0...[39m
	0: [2025-09-02 19:02:09,487] [INFO] [axolotl.train.save_initial_configs:419] [PID:2134977] [RANK:0] Pre-saving model config to /lustre/fswork/projects/rech/dgo/udv55np/math/Qwen3-235B-A22B/Qwen2.5-14B_ift/0...[39m
	0: [2025-09-02 19:02:09,502] [INFO] [axolotl.train.execute_training:203] [PID:2134977] [RANK:0] Starting trainer...[39m
	0: [2025-09-02 19:04:11,579] [INFO] [axolotl.utils.samplers.multipack.calc_min_len:436] [PID:2134977] [RANK:0] gather_len_batches: [25939, 25939, 25939, 25939, 25939, 25939, 25939, 25939, 25939, 25939, 25939, 25939, 25939, 25939, 25939, 25939][39m
	0: Parameter Offload - Persistent parameters statistics: param_count = 241, numel = 840704
	0: {'loss': 0.2404, 'grad_norm': 0.364680268021655, 'learning_rate': 1.54e-07, 'memory/max_mem_active(gib)': 52.3, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.01}
	0: 0%\| \| 0/1621 [00:00<?, ?it/s] 0%\| \| 1/1621 [03:16<88:13:23, 196.05s/it] 0%\| \| 2/1621 [03:20<37:32:36, 83.48s/it] 0%\| \| 3/1621 [03:25<21:22:10, 47.55s/it] 0%\| \| 4/1621 [03:30<13:44:38, 30.60s/it] 0%\| \| 5/1621 [03:35<9:35:07, 21.35s/it] 0%\| \| 6/1621 [03:39<7:02:38, 15.70s/it] 0%\| \| 7/1621 [03:45<5:33:21, 12.39s/it] 0%\| \| 8/1621 [03:50<4:27:23, 9.95s/it] 1%\| \| 9/1621 [03:54<3:42:49, 8.29s/it] 1%\| \| 10/1621 [04:00<3:17:19, 7.35s/it] 1%\| \| 10/1621 [04:00<3:17:19, 7.35s/it] 1%\| \| 11/1621 [04:05<2:58:11, 6.64s/it] 1%\| \| 12/1621 [04:09<2:42:25, 6.06s/it] 1%\| \| 13/1621 [04:14<2:32:10, 5.68s/it] 1%\| \| 14/1621 [04:19<2:24:06, 5.38s/it] 1%\| \| 15/1621 [04:24<2:23:35, 5.36s/it] 1%\| \| 16/1621 [04:29<2:18:24, 5.17s/it] 1%\| \| 17/1621 [04:34<2:16:57, 5.12s
	0: {'loss': 0.229, 'grad_norm': 0.29208771571602604, 'learning_rate': 2.14e-07, 'memory/max_mem_active(gib)': 52.3, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.01}
	0: {'loss': 0.2364, 'grad_norm': 0.3147847192135592, 'learning_rate': 2.74e-07, 'memory/max_mem_active(gib)': 53.83, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.02}
	0: /it] 1%\| \| 18/1621 [04:39<2:13:20, 4.99s/it] 1%\| \| 19/1621 [04:43<2:10:48, 4.90s/it] 1%\| \| 20/1621 [04:48<2:09:49, 4.87s/it] 1%\| \| 20/1621 [04:48<2:09:49, 4.87s/it] 1%\|▏ \| 21/1621 [04:54<2:17:13, 5.15s/it] 1%\|▏ \| 22/1621 [04:59<2:14:38, 5.05s/it] 1%\|▏ \| 23/1621 [05:03<2:12:25, 4.97s/it] 1%\|▏ \| 24/1621 [05:08<2:10:25, 4.90s/it] 2%\|▏ \| 25/1621 [05:13<2:08:46, 4.84s/it] 2%\|▏ \| 26/1621 [05:18<2:07:33, 4.80s/it] 2%\|▏ \| 27/1621 [05:23<2:09:47, 4.89s/it] 2%\|▏ \| 28/1621 [05:27<2:09:07, 4.86s/it] 2%\|▏ \| 29/1621 [05:32<2:07:33, 4.81s/it] 2%\|▏ \| 30/1621 [05:37<2:06:00, 4.75s/it] 2%\|▏ \| 30/1621 [05:37<2:06:00, 4.75s/it] 2%\|▏ \| 31/1621 [05:42<2:06:18, 4.77s/it] 2%\|▏ \| 32/1621 [05:47<2:07:55, 4.83s/it] 2
	0: {'loss': 0.2316, 'grad_norm': 0.2898188486284449, 'learning_rate': 3.34e-07, 'memory/max_mem_active(gib)': 53.83, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.02}
	0: %\|▏ \| 33/1621 [05:51<2:06:38, 4.78s/it] 2%\|▏ \| 34/1621 [05:56<2:08:50, 4.87s/it] 2%\|▏ \| 35/1621 [06:01<2:08:43, 4.87s/it] 2%\|▏ \| 36/1621 [06:06<2:07:47, 4.84s/it] 2%\|▏ \| 37/1621 [06:11<2:07:14, 4.82s/it] 2%\|▏ \| 38/1621 [06:16<2:13:26, 5.06s/it] 2%\|▏ \| 39/1621 [06:21<2:10:44, 4.96s/it] 2%\|▏ \| 40/1621 [06:26<2:09:23, 4.91s/it] 2%\|▏ \| 40/1621 [06:26<2:09:23, 4.91s/it] 3%\|▎ \| 41/1621 [06:31<2:09:26, 4.92s/it] 3%\|▎ \| 42/1621 [06:36<2:09:02, 4.90s/it] 3%\|▎ \| 43/1621 [06:41<2:08:48, 4.90s/it] 3%\|▎ \| 44/1621 [06:47<2:17:38, 5.24s/it] 3%\|▎ \| 45/1621 [06:51<2:13:22, 5.08s/it] 3%\|▎ \| 46/1621 [06:56<2:13:42, 5.09s/it] 3%\|▎ \| 47/1621 [07:03<2:22:28, 5.43s/it] 3%\|▎ \| 48/1621 [07:08<2:18:54, 5.30s/it] 3%\|▎ \| 49/1621 [07:12<2:15:19, 5.17s/it]
	0: {'loss': 0.2356, 'grad_norm': 0.30174914984263745, 'learning_rate': 3.94e-07, 'memory/max_mem_active(gib)': 53.83, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.03}
	0: {'loss': 0.2328, 'grad_norm': 0.39057298409934055, 'learning_rate': 4.5399999999999996e-07, 'memory/max_mem_active(gib)': 53.83, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.04}
	0: 3%\|▎ \| 50/1621 [07:17<2:11:06, 5.01s/it] 3%\|▎ \| 50/1621 [07:17<2:11:06, 5.01s/it] 3%\|▎ \| 51/1621 [07:22<2:09:59, 4.97s/it] 3%\|▎ \| 52/1621 [07:27<2:11:56, 5.05s/it] 3%\|▎ \| 53/1621 [07:32<2:10:03, 4.98s/it] 3%\|▎ \| 54/1621 [07:37<2:07:16, 4.87s/it] 3%\|▎ \| 55/1621 [07:41<2:05:36, 4.81s/it] 3%\|▎ \| 56/1621 [07:47<2:09:47, 4.98s/it] 4%\|▎ \| 57/1621 [07:52<2:14:39, 5.17s/it] 4%\|▎ \| 58/1621 [07:57<2:11:09, 5.04s/it] 4%\|▎ \| 59/1621 [08:02<2:09:51, 4.99s/it] 4%\|▎ \| 60/1621 [08:07<2:08:00, 4.92s/it] 4%\|▎ \| 60/1621 [08:07<2:08:00, 4.92s/it] 4%\|▍ \| 61/1621 [08:12<2:08:12, 4.93s/it] 4%\|▍ \| 62/1621 [08:16<2:06:51, 4.88s/it] 4%\|▍ \| 63/1621 [08:21<2:05:44, 4.84s/it] 4%\|▍ \| 64/1621 [08:26<2:05:05, 4.82s/it]
	0: {'loss': 0.2363, 'grad_norm': 0.4675017545126673, 'learning_rate': 5.14e-07, 'memory/max_mem_active(gib)': 53.83, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.04}
	0: {'loss': 0.2345, 'grad_norm': 0.2788174122742784, 'learning_rate': 5.739999999999999e-07, 'memory/max_mem_active(gib)': 53.83, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.05}
	0: 4%\|▍ \| 65/1621 [08:31<2:10:08, 5.02s/it] 4%\|▍ \| 66/1621 [08:36<2:08:23, 4.95s/it] 4%\|▍ \| 67/1621 [08:41<2:07:03, 4.91s/it] 4%\|▍ \| 68/1621 [08:46<2:05:37, 4.85s/it] 4%\|▍ \| 69/1621 [08:50<2:04:01, 4.79s/it] 4%\|▍ \| 70/1621 [08:55<2:02:54, 4.75s/it] 4%\|▍ \| 70/1621 [08:55<2:02:54, 4.75s/it] 4%\|▍ \| 71/1621 [09:00<2:03:44, 4.79s/it] 4%\|▍ \| 72/1621 [09:05<2:03:06, 4.77s/it] 5%\|▍ \| 73/1621 [09:09<2:03:25, 4.78s/it] 5%\|▍ \| 74/1621 [09:15<2:07:31, 4.95s/it] 5%\|▍ \| 75/1621 [09:19<2:05:23, 4.87s/it] 5%\|▍ \| 76/1621 [09:24<2:03:29, 4.80s/it] 5%\|▍ \| 77/1621 [09:29<2:03:04, 4.78s/it] 5%\|▍ \| 78/1621 [09:34<2:03:39, 4.81s/it] 5%\|▍ \| 79/1621 [09:38<2:02:28, 4.77s/it] 5%\|▍ \| 80/1621 [09:43<2:01:28, 4.73s/it]
	0: {'loss': 0.2368, 'grad_norm': 0.2746391832738924, 'learning_rate': 6.34e-07, 'memory/max_mem_active(gib)': 53.83, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.06}
	0: 5%\|▍ \| 80/1621 [09:43<2:01:28, 4.73s/it] 5%\|▍ \| 81/1621 [09:48<2:01:03, 4.72s/it] 5%\|▌ \| 82/1621 [09:53<2:06:03, 4.91s/it] 5%\|▌ \| 83/1621 [09:58<2:04:16, 4.85s/it] 5%\|▌ \| 84/1621 [10:02<2:03:08, 4.81s/it] 5%\|▌ \| 85/1621 [10:07<2:03:24, 4.82s/it] 5%\|▌ \| 86/1621 [10:12<2:02:23, 4.78s/it] 5%\|▌ \| 87/1621 [10:17<2:01:48, 4.76s/it] 5%\|▌ \| 88/1621 [10:21<2:00:55, 4.73s/it] 5%\|▌ \| 89/1621 [10:26<2:00:53, 4.73s/it] 6%\|▌ \| 90/1621 [10:31<2:00:48, 4.73s/it] 6%\|▌ \| 90/1621 [10:31<2:00:48, 4.73s/it] 6%\|▌ \| 91/1621 [10:36<2:00:21, 4.72s/it] 6%\|▌ \| 92/1621 [10:41<2:09:11, 5.07s/it] 6%\|▌ \| 93/1621 [10:46<2:07:23, 5.00s/it] 6%\|▌ \| 94/1621 [10:52<2:12:01, 5.19s/it] 6%\|▌ \| 95/1621 [10:57<2:09:02, 5.07s/it] 6%\|▌ \| 96/1621 [11:01<2:05:57, 4.96s/it]
	0: {'loss': 0.2293, 'grad_norm': 0.29838762735346686, 'learning_rate': 6.94e-07, 'memory/max_mem_active(gib)': 53.83, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.06}
	0: {'loss': 0.231, 'grad_norm': 0.275344655730642, 'learning_rate': 7.54e-07, 'memory/max_mem_active(gib)': 53.83, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.07}
	0: 6%\|▌ \| 97/1621 [11:07<2:09:45, 5.11s/it] 6%\|▌ \| 98/1621 [11:12<2:08:02, 5.04s/it] 6%\|▌ \| 99/1621 [11:17<2:09:26, 5.10s/it] 6%\|▌ \| 100/1621 [11:22<2:07:20, 5.02s/it] 6%\|▌ \| 100/1621 [11:22<2:07:20, 5.02s/it] 6%\|▌ \| 101/1621 [11:27<2:06:04, 4.98s/it] 6%\|▋ \| 102/1621 [11:31<2:04:16, 4.91s/it] 6%\|▋ \| 103/1621 [11:37<2:05:48, 4.97s/it] 6%\|▋ \| 104/1621 [11:41<2:04:00, 4.90s/it] 6%\|▋ \| 105/1621 [11:46<2:04:38, 4.93s/it] 7%\|▋ \| 106/1621 [11:51<2:03:31, 4.89s/it] 7%\|▋ \| 107/1621 [11:56<2:04:36, 4.94s/it] 7%\|▋ \| 108/1621 [12:02<2:09:36, 5.14s/it] 7%\|▋ \| 109/1621 [12:06<2:06:22, 5.01s/it] 7%\|▋ \| 110/1621 [12:12<2:07:28, 5.06s/it] 7%\|▋ \| 110/1621 [12:12<2:07:28, 5.06s/it] 7%\|▋ \| 111/1621 [12:17<2:06:
	0: {'loss': 0.2313, 'grad_norm': 0.2697336383234993, 'learning_rate': 8.14e-07, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.07}
	0: 48, 5.04s/it] 7%\|▋ \| 112/1621 [12:22<2:09:10, 5.14s/it] 7%\|▋ \| 113/1621 [12:27<2:06:01, 5.01s/it] 7%\|▋ \| 114/1621 [12:31<2:04:00, 4.94s/it] 7%\|▋ \| 115/1621 [12:36<2:01:55, 4.86s/it] 7%\|▋ \| 116/1621 [12:41<2:02:07, 4.87s/it] 7%\|▋ \| 117/1621 [12:46<2:01:10, 4.83s/it] 7%\|▋ \| 118/1621 [12:51<2:00:22, 4.81s/it] 7%\|▋ \| 119/1621 [12:55<2:00:09, 4.80s/it] 7%\|▋ \| 120/1621 [13:00<2:01:29, 4.86s/it] 7%\|▋ \| 120/1621 [13:00<2:01:29, 4.86s/it] 7%\|▋ \| 121/1621 [13:05<2:00:49, 4.83s/it] 8%\|▊ \| 122/1621 [13:10<1:59:47, 4.79s/it] 8%\|▊ \| 123/1621 [13:15<1:59:43, 4.80s/it] 8%\|▊ \| 124/1621 [13:19<1:59:46, 4.80s/it] 8%\|▊ \| 125/1621 [13:25<2:07:20, 5.11s/it] 8%\|▊ \| 126/1621 [13:30<2:06:59, 5.10s/it] 8%\|▊ \| 127/1621 [13:35<2:04:05, 4.98s/it] 8%\|▊ \| 1
	0: {'loss': 0.2332, 'grad_norm': 0.2602617441366092, 'learning_rate': 8.739999999999999e-07, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.08}
	0: {'loss': 0.2302, 'grad_norm': 0.29568418904442645, 'learning_rate': 9.339999999999999e-07, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.09}
	0: 28/1621 [13:40<2:02:36, 4.93s/it] 8%\|▊ \| 129/1621 [13:44<2:00:34, 4.85s/it] 8%\|▊ \| 130/1621 [13:50<2:02:56, 4.95s/it] 8%\|▊ \| 130/1621 [13:50<2:02:56, 4.95s/it] 8%\|▊ \| 131/1621 [13:55<2:07:42, 5.14s/it] 8%\|▊ \| 132/1621 [14:00<2:06:36, 5.10s/it] 8%\|▊ \| 133/1621 [14:06<2:07:58, 5.16s/it] 8%\|▊ \| 134/1621 [14:10<2:04:47, 5.04s/it] 8%\|▊ \| 135/1621 [14:16<2:09:02, 5.21s/it] 8%\|▊ \| 136/1621 [14:21<2:04:56, 5.05s/it] 8%\|▊ \| 137/1621 [14:25<2:03:24, 4.99s/it] 9%\|▊ \| 138/1621 [14:31<2:09:53, 5.26s/it] 9%\|▊ \| 139/1621 [14:37<2:09:39, 5.25s/it] 9%\|▊ \| 140/1621 [14:41<2:05:50, 5.10s/it] 9%\|▊ \| 140/1621 [14:41<2:05:50, 5.10s/it] 9%\|▊ \| 141/1621 [14:46<2:03:08, 4.99s/it] 9%\|▉ \| 142/1621 [14:51<2:05:27, 5.09s/it]
	0: {'loss': 0.228, 'grad_norm': 0.2833699712446876, 'learning_rate': 9.94e-07, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.09}
	0: 9%\|▉ \| 143/1621 [14:56<2:02:29, 4.97s/it] 9%\|▉ \| 144/1621 [15:01<2:01:18, 4.93s/it] 9%\|▉ \| 145/1621 [15:06<1:59:54, 4.87s/it] 9%\|▉ \| 146/1621 [15:10<1:58:29, 4.82s/it] 9%\|▉ \| 147/1621 [15:15<2:00:36, 4.91s/it] 9%\|▉ \| 148/1621 [15:20<1:59:48, 4.88s/it] 9%\|▉ \| 149/1621 [15:25<1:58:34, 4.83s/it] 9%\|▉ \| 150/1621 [15:30<1:59:20, 4.87s/it] 9%\|▉ \| 150/1621 [15:30<1:59:20, 4.87s/it] 9%\|▉ \| 151/1621 [15:35<2:03:59, 5.06s/it] 9%\|▉ \| 152/1621 [15:41<2:10:10, 5.32s/it] 9%\|▉ \| 153/1621 [15:46<2:05:24, 5.13s/it] 10%\|▉ \| 154/1621 [15:51<2:02:58, 5.03s/it] 10%\|▉ \| 155/1621 [15:56<2:03:28, 5.05s/it] 10%\|▉ \| 156/1621 [16:01<2:01:18, 4.97s/it] 10%\|▉ \| 157/1621 [16:06<2:02:57, 5.04s/it] 10%\|▉ \| 158/1621 [16:12<2:07:42, 5.24s/it] 10%\|▉ \| 159/1621 [16:16<2
	0: {'loss': 0.2244, 'grad_norm': 0.2798382710926688, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.1}
	0: {'loss': 0.2304, 'grad_norm': 0.2784530895613703, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.1}
	0: :04:33, 5.11s/it] 10%\|▉ \| 160/1621 [16:21<2:01:43, 5.00s/it] 10%\|▉ \| 160/1621 [16:21<2:01:43, 5.00s/it] 10%\|▉ \| 161/1621 [16:26<2:01:13, 4.98s/it] 10%\|▉ \| 162/1621 [16:31<2:01:08, 4.98s/it] 10%\|█ \| 163/1621 [16:36<1:59:32, 4.92s/it] 10%\|█ \| 164/1621 [16:41<1:57:23, 4.83s/it] 10%\|█ \| 165/1621 [16:45<1:57:19, 4.83s/it] 10%\|█ \| 166/1621 [16:50<1:56:06, 4.79s/it] 10%\|█ \| 167/1621 [16:55<1:59:03, 4.91s/it] 10%\|█ \| 168/1621 [17:00<1:57:33, 4.85s/it] 10%\|█ \| 169/1621 [17:05<2:00:04, 4.96s/it] 10%\|█ \| 170/1621 [17:10<1:58:32, 4.90s/it] 10%\|█ \| 170/1621 [17:10<1:58:32, 4.90s/it] 11%\|█ \| 171/1621 [17:15<1:57:41, 4.87s/it] 11%\|█ \| 172/1621 [17:19<1:56:43, 4.83s/it] 11%\|█ \| 173/1621 [17:24<1:55:32, 4.79s/it] 11%\|█
	0: {'loss': 0.231, 'grad_norm': 0.27996196551038177, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.11}
	0: \| 174/1621 [17:29<1:54:47, 4.76s/it] 11%\|█ \| 175/1621 [17:34<1:55:19, 4.79s/it] 11%\|█ \| 176/1621 [17:38<1:55:23, 4.79s/it] 11%\|█ \| 177/1621 [17:43<1:54:37, 4.76s/it] 11%\|█ \| 178/1621 [17:48<1:53:59, 4.74s/it] 11%\|█ \| 179/1621 [17:53<1:53:36, 4.73s/it] 11%\|█ \| 180/1621 [17:57<1:54:59, 4.79s/it] 11%\|█ \| 180/1621 [17:58<1:54:59, 4.79s/it] 11%\|█ \| 181/1621 [18:02<1:54:16, 4.76s/it] 11%\|█ \| 182/1621 [18:07<1:53:40, 4.74s/it] 11%\|█▏ \| 183/1621 [18:12<1:53:21, 4.73s/it] 11%\|█▏ \| 184/1621 [18:16<1:53:21, 4.73s/it] 11%\|█▏ \| 185/1621 [18:21<1:52:48, 4.71s/it] 11%\|█▏ \| 186/1621 [18:26<1:54:24, 4.78s/it] 12%\|█▏ \| 187/1621 [18:31<1:56:56, 4.89s/it] 12%\|█▏ \| 188/1621 [18:37<2:00:52, 5.06s/it] 12%\|█▏ \| 189/1621 [18:43<2:08:38, 5.39s/it] 12%\|█▏ \| 190/1621 [18:48<2
	0: {'loss': 0.2245, 'grad_norm': 0.2792720116191664, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.12}
	0: {'loss': 0.2339, 'grad_norm': 0.26943311530651054, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.12}
	0: :04:39, 5.23s/it] 12%\|█▏ \| 190/1621 [18:48<2:04:39, 5.23s/it] 12%\|█▏ \| 191/1621 [18:53<2:02:38, 5.15s/it] 12%\|█▏ \| 192/1621 [18:57<1:59:58, 5.04s/it] 12%\|█▏ \| 193/1621 [19:02<1:58:00, 4.96s/it] 12%\|█▏ \| 194/1621 [19:07<1:57:40, 4.95s/it] 12%\|█▏ \| 195/1621 [19:12<1:55:55, 4.88s/it] 12%\|█▏ \| 196/1621 [19:16<1:54:57, 4.84s/it] 12%\|█▏ \| 197/1621 [19:21<1:54:23, 4.82s/it] 12%\|█▏ \| 198/1621 [19:26<1:53:31, 4.79s/it] 12%\|█▏ \| 199/1621 [19:31<1:53:11, 4.78s/it] 12%\|█▏ \| 200/1621 [19:36<1:53:26, 4.79s/it] 12%\|█▏ \| 200/1621 [19:36<1:53:26, 4.79s/it] 12%\|█▏ \| 201/1621 [19:40<1:52:24, 4.75s/it] 12%\|█▏ \| 202/1621 [19:45<1:51:58, 4.73s/it] 13%\|█▎ \| 203/1621 [19:50<1:52:02, 4.74s/it] 13%\|█▎ \| 204/1621 [19:54<1:52:
	0: {'loss': 0.2227, 'grad_norm': 0.2691123392185592, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.13}
	0: 29, 4.76s/it] 13%\|█▎ \| 205/1621 [19:59<1:53:14, 4.80s/it] 13%\|█▎ \| 206/1621 [20:04<1:52:19, 4.76s/it] 13%\|█▎ \| 207/1621 [20:09<1:52:45, 4.78s/it] 13%\|█▎ \| 208/1621 [20:14<1:53:06, 4.80s/it] 13%\|█▎ \| 209/1621 [20:18<1:52:43, 4.79s/it] 13%\|█▎ \| 210/1621 [20:23<1:53:31, 4.83s/it] 13%\|█▎ \| 210/1621 [20:23<1:53:31, 4.83s/it] 13%\|█▎ \| 211/1621 [20:28<1:53:30, 4.83s/it] 13%\|█▎ \| 212/1621 [20:33<1:53:11, 4.82s/it] 13%\|█▎ \| 213/1621 [20:38<1:52:11, 4.78s/it] 13%\|█▎ \| 214/1621 [20:43<1:53:32, 4.84s/it] 13%\|█▎ \| 215/1621 [20:48<1:54:54, 4.90s/it] 13%\|█▎ \| 216/1621 [20:52<1:53:39, 4.85s/it] 13%\|█▎ \| 217/1621 [20:57<1:52:41, 4.82s/it] 13%\|█▎ \| 218/1621 [21:02<1:52:06, 4.79s/it] 14%\|█▎ \| 219/1621 [21:07<1:51:33, 4.77s/it] 14%\|█▎ \| 220/1621 [21:11<1:51:0
	0: {'loss': 0.2352, 'grad_norm': 0.3657263884364967, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.14}
	0: {'loss': 0.2264, 'grad_norm': 0.3037520130063509, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.14}
	0: 8, 4.76s/it] 14%\|█▎ \| 220/1621 [21:11<1:51:08, 4.76s/it] 14%\|█▎ \| 221/1621 [21:16<1:50:44, 4.75s/it] 14%\|█▎ \| 222/1621 [21:21<1:52:02, 4.81s/it] 14%\|█▍ \| 223/1621 [21:26<1:50:49, 4.76s/it] 14%\|█▍ \| 224/1621 [21:30<1:50:47, 4.76s/it] 14%\|█▍ \| 225/1621 [21:35<1:50:56, 4.77s/it] 14%\|█▍ \| 226/1621 [21:40<1:51:13, 4.78s/it] 14%\|█▍ \| 227/1621 [21:45<1:50:01, 4.74s/it] 14%\|█▍ \| 228/1621 [21:50<1:50:52, 4.78s/it] 14%\|█▍ \| 229/1621 [21:54<1:50:11, 4.75s/it] 14%\|█▍ \| 230/1621 [21:59<1:51:07, 4.79s/it] 14%\|█▍ \| 230/1621 [21:59<1:51:07, 4.79s/it] 14%\|█▍ \| 231/1621 [22:04<1:50:29, 4.77s/it] 14%\|█▍ \| 232/1621 [22:08<1:49:39, 4.74s/it] 14%\|█▍ \| 233/1621 [22:13<1:49:34, 4.74s/it] 14%\|█▍ \| 234/1621 [22:18<1:49:09,
	0: {'loss': 0.231, 'grad_norm': 0.2936400072170698, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.15}
	0: 4.72s/it] 14%\|█▍ \| 235/1621 [22:23<1:50:13, 4.77s/it] 15%\|█▍ \| 236/1621 [22:28<1:49:57, 4.76s/it] 15%\|█▍ \| 237/1621 [22:32<1:49:10, 4.73s/it] 15%\|█▍ \| 238/1621 [22:37<1:49:15, 4.74s/it] 15%\|█▍ \| 239/1621 [22:42<1:49:34, 4.76s/it] 15%\|█▍ \| 240/1621 [22:46<1:48:47, 4.73s/it] 15%\|█▍ \| 240/1621 [22:46<1:48:47, 4.73s/it] 15%\|█▍ \| 241/1621 [22:51<1:48:11, 4.70s/it] 15%\|█▍ \| 242/1621 [22:56<1:47:47, 4.69s/it] 15%\|█▍ \| 243/1621 [23:00<1:47:46, 4.69s/it] 15%\|█▌ \| 244/1621 [23:05<1:47:33, 4.69s/it] 15%\|█▌ \| 245/1621 [23:10<1:48:45, 4.74s/it] 15%\|█▌ \| 246/1621 [23:15<1:48:13, 4.72s/it] 15%\|█▌ \| 247/1621 [23:21<1:58:08, 5.16s/it] 15%\|█▌ \| 248/1621 [23:26<1:55:14, 5.04s/it] 15%\|█▌ \| 249/1621 [23:31<1:57:59, 5.16s/it] 15%\|█▌ \| 250/1621 [23:36<1:54:34, 5
	0: {'loss': 0.225, 'grad_norm': 0.2749950026180422, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.15}
	0: {'loss': 0.2254, 'grad_norm': 0.27253893456551315, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.16}
	0: .01s/it] 15%\|█▌ \| 250/1621 [23:36<1:54:34, 5.01s/it] 15%\|█▌ \| 251/1621 [23:40<1:52:51, 4.94s/it] 16%\|█▌ \| 252/1621 [23:45<1:51:51, 4.90s/it] 16%\|█▌ \| 253/1621 [23:50<1:52:12, 4.92s/it] 16%\|█▌ \| 254/1621 [23:55<1:50:42, 4.86s/it] 16%\|█▌ \| 255/1621 [24:00<1:51:10, 4.88s/it] 16%\|█▌ \| 256/1621 [24:05<1:50:46, 4.87s/it] 16%\|█▌ \| 257/1621 [24:10<1:50:58, 4.88s/it] 16%\|█▌ \| 258/1621 [24:14<1:49:41, 4.83s/it] 16%\|█▌ \| 259/1621 [24:19<1:49:52, 4.84s/it] 16%\|█▌ \| 260/1621 [24:24<1:49:30, 4.83s/it] 16%\|█▌ \| 260/1621 [24:24<1:49:30, 4.83s/it] 16%\|█▌ \| 261/1621 [24:29<1:48:43, 4.80s/it] 16%\|█▌ \| 262/1621 [24:33<1:48:12, 4.78s/it] 16%\|█▌ \| 263/1621 [24:39<1:51:03, 4.91s/it] 16%\|█▋ \| 264/1621 [24:44<1:50:55, 4.90s
	0: {'loss': 0.223, 'grad_norm': 0.28707421840879827, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.17}
	0: /it] 16%\|█▋ \| 265/1621 [24:49<1:51:38, 4.94s/it] 16%\|█▋ \| 266/1621 [24:54<1:53:18, 5.02s/it] 16%\|█▋ \| 267/1621 [25:00<1:58:37, 5.26s/it] 17%\|█▋ \| 268/1621 [25:04<1:54:50, 5.09s/it] 17%\|█▋ \| 269/1621 [25:09<1:52:18, 4.98s/it] 17%\|█▋ \| 270/1621 [25:14<1:50:21, 4.90s/it] 17%\|█▋ \| 270/1621 [25:14<1:50:21, 4.90s/it] 17%\|█▋ \| 271/1621 [25:19<1:50:30, 4.91s/it] 17%\|█▋ \| 272/1621 [25:24<1:51:59, 4.98s/it] 17%\|█▋ \| 273/1621 [25:30<1:57:14, 5.22s/it] 17%\|█▋ \| 274/1621 [25:35<1:55:12, 5.13s/it] 17%\|█▋ \| 275/1621 [25:39<1:52:27, 5.01s/it] 17%\|█▋ \| 276/1621 [25:44<1:53:28, 5.06s/it] 17%\|█▋ \| 277/1621 [25:49<1:52:16, 5.01s/it] 17%\|█▋ \| 278/1621 [25:54<1:49:54, 4.91s/it] 17%\|█▋ \| 279/1621 [25:59<1:49:03, 4.88s/it] 17%\|█▋ \| 280/1621 [26:04<1:49:15, 4.89s/
	0: {'loss': 0.2273, 'grad_norm': 0.29934907545031625, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.17}
	0: {'loss': 0.2251, 'grad_norm': 0.27152873873106187, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.18}
	0: it] 17%\|█▋ \| 280/1621 [26:04<1:49:15, 4.89s/it] 17%\|█▋ \| 281/1621 [26:09<1:54:52, 5.14s/it] 17%\|█▋ \| 282/1621 [26:14<1:52:43, 5.05s/it] 17%\|█▋ \| 283/1621 [26:19<1:51:16, 4.99s/it] 18%\|█▊ \| 284/1621 [26:24<1:49:26, 4.91s/it] 18%\|█▊ \| 285/1621 [26:29<1:48:12, 4.86s/it] 18%\|█▊ \| 286/1621 [26:33<1:46:58, 4.81s/it] 18%\|█▊ \| 287/1621 [26:38<1:46:48, 4.80s/it] 18%\|█▊ \| 288/1621 [26:43<1:47:36, 4.84s/it] 18%\|█▊ \| 289/1621 [26:48<1:48:04, 4.87s/it] 18%\|█▊ \| 290/1621 [26:53<1:47:51, 4.86s/it] 18%\|█▊ \| 290/1621 [26:53<1:47:51, 4.86s/it] 18%\|█▊ \| 291/1621 [26:58<1:47:39, 4.86s/it] 18%\|█▊ \| 292/1621 [27:03<1:50:40, 5.00s/it] 18%\|█▊ \| 293/1621 [27:08<1:48:20, 4.90s/it] 18%\|█▊ \| 294/1621 [27:12<1:47:16, 4.85s/it]
	0: {'loss': 0.2256, 'grad_norm': 0.2697161270992779, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.19}
	0: 18%\|█▊ \| 295/1621 [27:17<1:48:02, 4.89s/it] 18%\|█▊ \| 296/1621 [27:22<1:48:22, 4.91s/it] 18%\|█▊ \| 297/1621 [27:27<1:47:10, 4.86s/it] 18%\|█▊ \| 298/1621 [27:32<1:46:31, 4.83s/it] 18%\|█▊ \| 299/1621 [27:37<1:47:51, 4.89s/it] 19%\|█▊ \| 300/1621 [27:42<1:46:42, 4.85s/it] 19%\|█▊ \| 300/1621 [27:42<1:46:42, 4.85s/it] 19%\|█▊ \| 301/1621 [27:46<1:46:00, 4.82s/it] 19%\|█▊ \| 302/1621 [27:51<1:45:18, 4.79s/it] 19%\|█▊ \| 303/1621 [27:57<1:52:19, 5.11s/it] 19%\|█▉ \| 304/1621 [28:02<1:49:50, 5.00s/it] 19%\|█▉ \| 305/1621 [28:06<1:48:05, 4.93s/it] 19%\|█▉ \| 306/1621 [28:11<1:47:23, 4.90s/it] 19%\|█▉ \| 307/1621 [28:16<1:45:56, 4.84s/it] 19%\|█▉ \| 308/1621 [28:21<1:45:46, 4.83s/it] 19%\|█▉ \| 309/1621 [28:26<1:44:51, 4.80s/it] 19%\|█▉ \| 310/1621 [28:30<1:44:15, 4.77s/it]
	0: {'loss': 0.2215, 'grad_norm': 0.2703094434846044, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.19}
	0: {'loss': 0.2252, 'grad_norm': 0.28310536455434837, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.2}
	0: 19%\|█▉ \| 310/1621 [28:30<1:44:15, 4.77s/it] 19%\|█▉ \| 311/1621 [28:36<1:48:13, 4.96s/it] 19%\|█▉ \| 312/1621 [28:41<1:48:06, 4.96s/it] 19%\|█▉ \| 313/1621 [28:45<1:46:19, 4.88s/it] 19%\|█▉ \| 314/1621 [28:50<1:45:08, 4.83s/it] 19%\|█▉ \| 315/1621 [28:55<1:45:27, 4.85s/it] 19%\|█▉ \| 316/1621 [29:00<1:44:51, 4.82s/it] 20%\|█▉ \| 317/1621 [29:04<1:44:44, 4.82s/it] 20%\|█▉ \| 318/1621 [29:10<1:47:42, 4.96s/it] 20%\|█▉ \| 319/1621 [29:15<1:47:43, 4.96s/it] 20%\|█▉ \| 320/1621 [29:19<1:46:25, 4.91s/it] 20%\|█▉ \| 320/1621 [29:19<1:46:25, 4.91s/it] 20%\|█▉ \| 321/1621 [29:24<1:44:57, 4.84s/it] 20%\|█▉ \| 322/1621 [29:29<1:44:11, 4.81s/it] 20%\|█▉ \| 323/1621 [29:34<1:43:16, 4.77s/it] 20%\|█▉ \| 324/1621 [29:38<1:42:38, 4.75s/it] 20%\|
	0: [2025-09-02 19:34:06,837] [INFO] [axolotl.core.trainers.base._save:613] [PID:2134977] [RANK:0] Saving model checkpoint to /lustre/fswork/projects/rech/dgo/udv55np/math/Qwen3-235B-A22B/Qwen2.5-14B_ift/0/checkpoint-325[39m
	0: [2025-09-02 19:34:29,509] [INFO] [axolotl.core.trainers.base._save:662] [PID:2134977] [RANK:0] Saving Trainer.data_collator.tokenizer by default as Trainer.processing_class is `None`[39m
	0: {'loss': 0.2218, 'grad_norm': 0.2993205202255554, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.2}
	0: ██ \| 325/1621 [29:44<1:47:14, 4.97s/it] 20%\|██ \| 326/1621 [30:30<6:15:14, 17.39s/it] 20%\|██ \| 327/1621 [30:36<4:57:55, 13.81s/it] 20%\|██ \| 328/1621 [30:40<3:59:57, 11.13s/it] 20%\|██ \| 329/1621 [30:45<3:18:22, 9.21s/it] 20%\|██ \| 330/1621 [30:50<2:48:55, 7.85s/it] 20%\|██ \| 330/1621 [30:50<2:48:55, 7.85s/it] 20%\|██ \| 331/1621 [30:55<2:28:26, 6.90s/it] 20%\|██ \| 332/1621 [30:59<2:14:10, 6.25s/it] 21%\|██ \| 333/1621 [31:04<2:05:04, 5.83s/it] 21%\|██ \| 334/1621 [31:09<2:01:34, 5.67s/it] 21%\|██ \| 335/1621 [31:14<1:56:50, 5.45s/it] 21%\|██ \| 336/1621 [31:20<1:56:11, 5.43s/it] 21%\|██ \| 337/1621 [31:24<1:51:24, 5.21s/it] 21%\|██ \| 338/1621 [31:29<1:48:01, 5.05s/it] 21%\|██ \| 339/1621 [31:34<1:47:23, 5.03s/it] 21%\|██ \| 340/1621 [31:39<1:44:55, 4.91s/it]
	0: {'loss': 0.226, 'grad_norm': 0.27889848818452406, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.21}
	0: {'loss': 0.2205, 'grad_norm': 0.2827087303201053, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.22}
	0: 21%\|██ \| 340/1621 [31:39<1:44:55, 4.91s/it] 21%\|██ \| 341/1621 [31:45<1:52:00, 5.25s/it] 21%\|██ \| 342/1621 [31:50<1:48:55, 5.11s/it] 21%\|██ \| 343/1621 [31:55<1:50:00, 5.16s/it] 21%\|██ \| 344/1621 [32:00<1:48:47, 5.11s/it] 21%\|██▏ \| 345/1621 [32:05<1:47:45, 5.07s/it] 21%\|██▏ \| 346/1621 [32:09<1:44:58, 4.94s/it] 21%\|██▏ \| 347/1621 [32:14<1:43:21, 4.87s/it] 21%\|██▏ \| 348/1621 [32:19<1:43:11, 4.86s/it] 22%\|██▏ \| 349/1621 [32:24<1:41:37, 4.79s/it] 22%\|██▏ \| 350/1621 [32:28<1:41:48, 4.81s/it] 22%\|██▏ \| 350/1621 [32:29<1:41:48, 4.81s/it] 22%\|██▏ \| 351/1621 [32:33<1:41:34, 4.80s/it] 22%\|██▏ \| 352/1621 [32:38<1:40:53, 4.77s/it] 22%\|██▏ \| 353/1621 [32:43<1:40:09, 4.74s/it] 22%\|██▏ \| 354/1621 [32:48<1:41:13,
	0: {'loss': 0.2276, 'grad_norm': 0.2806893131234233, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.22}
	0: 4.79s/it] 22%\|██▏ \| 355/1621 [32:52<1:40:42, 4.77s/it] 22%\|██▏ \| 356/1621 [32:57<1:40:14, 4.75s/it] 22%\|██▏ \| 357/1621 [33:02<1:39:51, 4.74s/it] 22%\|██▏ \| 358/1621 [33:06<1:39:47, 4.74s/it] 22%\|██▏ \| 359/1621 [33:11<1:39:50, 4.75s/it] 22%\|██▏ \| 360/1621 [33:16<1:41:07, 4.81s/it] 22%\|██▏ \| 360/1621 [33:16<1:41:07, 4.81s/it] 22%\|██▏ \| 361/1621 [33:21<1:41:13, 4.82s/it] 22%\|██▏ \| 362/1621 [33:26<1:41:29, 4.84s/it] 22%\|██▏ \| 363/1621 [33:31<1:40:38, 4.80s/it] 22%\|██▏ \| 364/1621 [33:35<1:41:05, 4.83s/it] 23%\|██▎ \| 365/1621 [33:40<1:40:00, 4.78s/it] 23%\|██▎ \| 366/1621 [33:45<1:39:25, 4.75s/it] 23%\|██▎ \| 367/1621 [33:51<1:47:26, 5.14s/it] 23%\|██▎ \| 368/1621 [33:56<1:46:27, 5.10s/it] 23%\|██▎ \| 369/1621 [34:01<1:44:50, 5.02s/it] 23%\|██▎
	0: {'loss': 0.2228, 'grad_norm': 0.2891434638471352, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.23}
	0: {'loss': 0.2249, 'grad_norm': 0.2937072467224044, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.23}
	0: \| 370/1621 [34:06<1:43:29, 4.96s/it] 23%\|██▎ \| 370/1621 [34:06<1:43:29, 4.96s/it] 23%\|██▎ \| 371/1621 [34:10<1:41:40, 4.88s/it] 23%\|██▎ \| 372/1621 [34:15<1:40:28, 4.83s/it] 23%\|██▎ \| 373/1621 [34:20<1:40:47, 4.85s/it] 23%\|██▎ \| 374/1621 [34:25<1:41:22, 4.88s/it] 23%\|██▎ \| 375/1621 [34:30<1:40:46, 4.85s/it] 23%\|██▎ \| 376/1621 [34:35<1:43:29, 4.99s/it] 23%\|██▎ \| 377/1621 [34:40<1:41:56, 4.92s/it] 23%\|██▎ \| 378/1621 [34:45<1:42:50, 4.96s/it] 23%\|██▎ \| 379/1621 [34:51<1:49:28, 5.29s/it] 23%\|██▎ \| 380/1621 [34:56<1:46:14, 5.14s/it] 23%\|██▎ \| 380/1621 [34:56<1:46:14, 5.14s/it] 24%\|██▎ \| 381/1621 [35:00<1:43:24, 5.00s/it] 24%\|██▎ \| 382/1621 [35:05<1:42:04, 4.94s/it] 24%\|██▎ \| 383/1621 [35:10<1:40:2
	0: {'loss': 0.2224, 'grad_norm': 0.28778273745914995, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.24}
	0: 4, 4.87s/it] 24%\|██▎ \| 384/1621 [35:16<1:47:21, 5.21s/it] 24%\|██▍ \| 385/1621 [35:21<1:45:31, 5.12s/it] 24%\|██▍ \| 386/1621 [35:26<1:48:57, 5.29s/it] 24%\|██▍ \| 387/1621 [35:31<1:44:54, 5.10s/it] 24%\|██▍ \| 388/1621 [35:36<1:45:26, 5.13s/it] 24%\|██▍ \| 389/1621 [35:41<1:42:28, 4.99s/it] 24%\|██▍ \| 390/1621 [35:46<1:41:26, 4.94s/it] 24%\|██▍ \| 390/1621 [35:46<1:41:26, 4.94s/it] 24%\|██▍ \| 391/1621 [35:50<1:40:10, 4.89s/it] 24%\|██▍ \| 392/1621 [35:56<1:41:45, 4.97s/it] 24%\|██▍ \| 393/1621 [36:00<1:40:15, 4.90s/it] 24%\|██▍ \| 394/1621 [36:06<1:47:42, 5.27s/it] 24%\|██▍ \| 395/1621 [36:12<1:46:47, 5.23s/it] 24%\|██▍ \| 396/1621 [36:16<1:43:31, 5.07s/it] 24%\|██▍ \| 397/1621 [36:21<1:41:40, 4.98s/it] 25%\|██▍ \| 398/1621 [36:26<1:39:39, 4.89s/it] 25%\|██�
	0: {'loss': 0.2161, 'grad_norm': 0.29578423224080963, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.25}
	0: {'loss': 0.2226, 'grad_norm': 0.2776599089421953, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.25}
	0: �� \| 399/1621 [36:30<1:38:21, 4.83s/it] 25%\|██▍ \| 400/1621 [36:36<1:39:47, 4.90s/it] 25%\|██▍ \| 400/1621 [36:36<1:39:47, 4.90s/it] 25%\|██▍ \| 401/1621 [36:40<1:39:31, 4.89s/it] 25%\|██▍ \| 402/1621 [36:46<1:41:50, 5.01s/it] 25%\|██▍ \| 403/1621 [36:51<1:43:47, 5.11s/it] 25%\|██▍ \| 404/1621 [36:56<1:41:13, 4.99s/it] 25%\|██▍ \| 405/1621 [37:01<1:42:15, 5.05s/it] 25%\|██▌ \| 406/1621 [37:06<1:40:11, 4.95s/it] 25%\|██▌ \| 407/1621 [37:10<1:39:04, 4.90s/it] 25%\|██▌ \| 408/1621 [37:15<1:38:15, 4.86s/it] 25%\|██▌ \| 409/1621 [37:20<1:37:13, 4.81s/it] 25%\|██▌ \| 410/1621 [37:25<1:36:42, 4.79s/it] 25%\|██▌ \| 410/1621 [37:25<1:36:42, 4.79s/it] 25%\|██▌ \| 411/1621 [37:29<1:36:19, 4.78s/it] 25%\|██▌ \| 412/1621 [37:34<1:36
	0: {'loss': 0.2202, 'grad_norm': 0.2851396473633923, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.26}
	0: :04, 4.77s/it] 25%\|██▌ \| 413/1621 [37:39<1:38:57, 4.92s/it] 26%\|██▌ \| 414/1621 [37:44<1:37:37, 4.85s/it] 26%\|██▌ \| 415/1621 [37:49<1:36:40, 4.81s/it] 26%\|██▌ \| 416/1621 [37:54<1:40:00, 4.98s/it] 26%\|██▌ \| 417/1621 [37:59<1:38:12, 4.89s/it] 26%\|██▌ \| 418/1621 [38:04<1:36:40, 4.82s/it] 26%\|██▌ \| 419/1621 [38:09<1:38:59, 4.94s/it] 26%\|██▌ \| 420/1621 [38:13<1:37:37, 4.88s/it] 26%\|██▌ \| 420/1621 [38:13<1:37:37, 4.88s/it] 26%\|██▌ \| 421/1621 [38:18<1:36:14, 4.81s/it] 26%\|██▌ \| 422/1621 [38:23<1:35:30, 4.78s/it] 26%\|██▌ \| 423/1621 [38:28<1:34:55, 4.75s/it] 26%\|██▌ \| 424/1621 [38:33<1:38:07, 4.92s/it] 26%\|██▌ \| 425/1621 [38:38<1:37:06, 4.87s/it] 26%\|██▋ \| 426/1621 [38:43<1:41:53, 5.12s/it] 26%\|██▋ \| 427/1621 [38:48<1:40:17, 5.04s/it] 26%\|█�
	0: {'loss': 0.226, 'grad_norm': 0.3027055991780978, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.27}
	0: {'loss': 0.2171, 'grad_norm': 0.2775325370268343, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.27}
	0: �▋ \| 428/1621 [38:53<1:38:18, 4.94s/it] 26%\|██▋ \| 429/1621 [38:58<1:36:40, 4.87s/it] 27%\|██▋ \| 430/1621 [39:02<1:35:42, 4.82s/it] 27%\|██▋ \| 430/1621 [39:02<1:35:42, 4.82s/it] 27%\|██▋ \| 431/1621 [39:07<1:37:34, 4.92s/it] 27%\|██▋ \| 432/1621 [39:12<1:36:44, 4.88s/it] 27%\|██▋ \| 433/1621 [39:17<1:37:25, 4.92s/it] 27%\|██▋ \| 434/1621 [39:23<1:44:10, 5.27s/it] 27%\|██▋ \| 435/1621 [39:28<1:41:00, 5.11s/it] 27%\|██▋ \| 436/1621 [39:33<1:39:14, 5.02s/it] 27%\|██▋ \| 437/1621 [39:38<1:37:52, 4.96s/it] 27%\|██▋ \| 438/1621 [39:43<1:40:38, 5.10s/it] 27%\|██▋ \| 439/1621 [39:48<1:38:08, 4.98s/it] 27%\|██▋ \| 440/1621 [39:53<1:39:14, 5.04s/it] 27%\|██▋ \| 440/1621 [39:53<1:39:14, 5.04s/it] 27%\|██▋ \| 441/1621 [39:58<1:
	0: {'loss': 0.2224, 'grad_norm': 0.28083751378878014, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.28}
	0: 37:07, 4.94s/it] 27%\|██▋ \| 442/1621 [40:03<1:37:12, 4.95s/it] 27%\|██▋ \| 443/1621 [40:07<1:35:41, 4.87s/it] 27%\|██▋ \| 444/1621 [40:12<1:34:35, 4.82s/it] 27%\|██▋ \| 445/1621 [40:17<1:34:23, 4.82s/it] 28%\|██▊ \| 446/1621 [40:23<1:42:49, 5.25s/it] 28%\|██▊ \| 447/1621 [40:28<1:39:29, 5.08s/it] 28%\|██▊ \| 448/1621 [40:33<1:37:15, 4.97s/it] 28%\|██▊ \| 449/1621 [40:37<1:35:56, 4.91s/it] 28%\|██▊ \| 450/1621 [40:42<1:35:52, 4.91s/it] 28%\|██▊ \| 450/1621 [40:42<1:35:52, 4.91s/it] 28%\|██▊ \| 451/1621 [40:47<1:36:38, 4.96s/it] 28%\|██▊ \| 452/1621 [40:52<1:36:38, 4.96s/it] 28%\|██▊ \| 453/1621 [40:57<1:35:08, 4.89s/it] 28%\|██▊ \| 454/1621 [41:02<1:34:12, 4.84s/it] 28%\|██▊ \| 455/1621 [41:06<1:33:20, 4.80s/it] 28%\|██▊ \| 456/1621 [41:11<1:32:59, 4.79s/it] 28%\|█
	0: {'loss': 0.2217, 'grad_norm': 0.29695254217700356, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.28}
	0: {'loss': 0.2201, 'grad_norm': 0.27273838019805313, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.29}
	0: █▊ \| 457/1621 [41:17<1:37:21, 5.02s/it] 28%\|██▊ \| 458/1621 [41:21<1:35:11, 4.91s/it] 28%\|██▊ \| 459/1621 [41:26<1:34:12, 4.86s/it] 28%\|██▊ \| 460/1621 [41:31<1:32:54, 4.80s/it] 28%\|██▊ \| 460/1621 [41:31<1:32:54, 4.80s/it] 28%\|██▊ \| 461/1621 [41:35<1:32:10, 4.77s/it] 29%\|██▊ \| 462/1621 [41:40<1:32:55, 4.81s/it] 29%\|██▊ \| 463/1621 [41:45<1:33:33, 4.85s/it] 29%\|██▊ \| 464/1621 [41:50<1:33:54, 4.87s/it] 29%\|██▊ \| 465/1621 [41:55<1:33:14, 4.84s/it] 29%\|██▊ \| 466/1621 [42:00<1:32:05, 4.78s/it] 29%\|██▉ \| 467/1621 [42:05<1:33:47, 4.88s/it] 29%\|██▉ \| 468/1621 [42:09<1:32:47, 4.83s/it] 29%\|██▉ \| 469/1621 [42:14<1:32:52, 4.84s/it] 29%\|██▉ \| 470/1621 [42:19<1:33:00, 4.85s/it] 29%\|██▉ \| 470/1621 [42:19<
	0: {'loss': 0.224, 'grad_norm': 0.29445797771312915, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.3}
	0: 1:33:00, 4.85s/it] 29%\|██▉ \| 471/1621 [42:25<1:37:01, 5.06s/it] 29%\|██▉ \| 472/1621 [42:29<1:34:39, 4.94s/it] 29%\|██▉ \| 473/1621 [42:35<1:39:18, 5.19s/it] 29%\|██▉ \| 474/1621 [42:40<1:37:07, 5.08s/it] 29%\|██▉ \| 475/1621 [42:45<1:34:53, 4.97s/it] 29%\|██▉ \| 476/1621 [42:49<1:33:02, 4.88s/it] 29%\|██▉ \| 477/1621 [42:54<1:32:32, 4.85s/it] 29%\|██▉ \| 478/1621 [42:59<1:31:41, 4.81s/it] 30%\|██▉ \| 479/1621 [43:04<1:30:34, 4.76s/it] 30%\|██▉ \| 480/1621 [43:08<1:30:11, 4.74s/it] 30%\|██▉ \| 480/1621 [43:08<1:30:11, 4.74s/it] 30%\|██▉ \| 481/1621 [43:13<1:29:45, 4.72s/it] 30%\|██▉ \| 482/1621 [43:18<1:29:25, 4.71s/it] 30%\|██▉ \| 483/1621 [43:23<1:34:09, 4.96s/it] 30%\|██▉ \| 484/1621 [43:28<1:32:47, 4.90s/it] 30%\|██▉ \| 485/1621 [43:33<1:33:42, 4.95s/it] 30%\|�
	0: {'loss': 0.2164, 'grad_norm': 0.27358498215933197, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.3}
	0: ��█▉ \| 486/1621 [43:38<1:34:06, 4.97s/it] 30%\|███ \| 487/1621 [43:43<1:33:49, 4.96s/it] 30%\|███ \| 488/1621 [43:48<1:32:14, 4.88s/it] 30%\|███ \| 489/1621 [43:53<1:36:36, 5.12s/it] 30%\|███ \| 490/1621 [43:58<1:35:07, 5.05s/it] 30%\|███ \| 490/1621 [43:58<1:35:07, 5.05s/it] 30%\|███ \| 491/1621 [44:03<1:36:21, 5.12s/it] 30%\|███ \| 492/1621 [44:09<1:36:55, 5.15s/it] 30%\|███ \| 493/1621 [44:14<1:40:00, 5.32s/it] 30%\|███ \| 494/1621 [44:19<1:36:20, 5.13s/it] 31%\|███ \| 495/1621 [44:24<1:35:47, 5.10s/it] 31%\|███ \| 496/1621 [44:29<1:35:16, 5.08s/it] 31%\|███ \| 497/1621 [44:34<1:33:39, 5.00s/it] 31%\|███ \| 498/1621 [44:39<1:33:44, 5.01s/it] 31%\|███ \| 499/1621 [44:44<1:32:40, 4.96s/it] 31%\|███ \| 500/1621 [44:49<1:31:23, 4.89s/it]
	0: {'loss': 0.2163, 'grad_norm': 0.2699071388397595, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.31}
	0: {'loss': 0.2203, 'grad_norm': 0.3025178334899167, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.31}
	0: 31%\|███ \| 500/1621 [44:49<1:31:23, 4.89s/it] 31%\|███ \| 501/1621 [44:54<1:32:10, 4.94s/it] 31%\|███ \| 502/1621 [44:59<1:32:15, 4.95s/it] 31%\|███ \| 503/1621 [45:03<1:31:08, 4.89s/it] 31%\|███ \| 504/1621 [45:08<1:30:56, 4.88s/it] 31%\|███ \| 505/1621 [45:13<1:30:18, 4.86s/it] 31%\|███ \| 506/1621 [45:18<1:29:30, 4.82s/it] 31%\|███▏ \| 507/1621 [45:22<1:28:54, 4.79s/it] 31%\|███▏ \| 508/1621 [45:27<1:29:52, 4.84s/it] 31%\|███▏ \| 509/1621 [45:32<1:29:50, 4.85s/it] 31%\|███▏ \| 510/1621 [45:37<1:29:20, 4.83s/it] 31%\|███▏ \| 510/1621 [45:37<1:29:20, 4.83s/it] 32%\|███▏ \| 511/1621 [45:42<1:29:45, 4.85s/it] 32%\|███▏ \| 512/1621 [45:47<1:29:00, 4.82s/it] 32%\|███▏ \| 513/1621 [45:52<1:32:24, 5.00s/it] 32%\|███▏ \| 514/1621 [45:57<1:32:2
	0: {'loss': 0.2162, 'grad_norm': 0.2858585805555601, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.32}
	0: 6, 5.01s/it] 32%\|███▏ \| 515/1621 [46:03<1:35:25, 5.18s/it] 32%\|███▏ \| 516/1621 [46:07<1:32:36, 5.03s/it] 32%\|███▏ \| 517/1621 [46:12<1:30:54, 4.94s/it] 32%\|███▏ \| 518/1621 [46:17<1:29:17, 4.86s/it] 32%\|███▏ \| 519/1621 [46:23<1:34:39, 5.15s/it] 32%\|███▏ \| 520/1621 [46:27<1:32:11, 5.02s/it] 32%\|███▏ \| 520/1621 [46:27<1:32:11, 5.02s/it] 32%\|███▏ \| 521/1621 [46:32<1:30:34, 4.94s/it] 32%\|███▏ \| 522/1621 [46:37<1:30:00, 4.91s/it] 32%\|███▏ \| 523/1621 [46:42<1:29:12, 4.88s/it] 32%\|███▏ \| 524/1621 [46:48<1:33:54, 5.14s/it] 32%\|███▏ \| 525/1621 [46:52<1:31:21, 5.00s/it] 32%\|███▏ \| 526/1621 [46:57<1:29:44, 4.92s/it] 33%\|███▎ \| 527/1621 [47:02<1:29:19, 4.90s/it] 33%\|███▎ \| 528/1621 [47:06<1:28:10, 4.84s/it] 33%\|███▎ \| 529/1621 [47:12<
	0: {'loss': 0.2196, 'grad_norm': 0.2825619726348337, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.33}
	0: {'loss': 0.2215, 'grad_norm': 0.2806598223572053, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.33}
	0: 1:29:32, 4.92s/it] 33%\|███▎ \| 530/1621 [47:16<1:28:29, 4.87s/it] 33%\|███▎ \| 530/1621 [47:16<1:28:29, 4.87s/it] 33%\|███▎ \| 531/1621 [47:21<1:28:25, 4.87s/it] 33%\|███▎ \| 532/1621 [47:26<1:27:42, 4.83s/it] 33%\|███▎ \| 533/1621 [47:31<1:30:30, 4.99s/it] 33%\|███▎ \| 534/1621 [47:37<1:31:48, 5.07s/it] 33%\|███▎ \| 535/1621 [47:41<1:30:32, 5.00s/it] 33%\|███▎ \| 536/1621 [47:46<1:28:57, 4.92s/it] 33%\|███▎ \| 537/1621 [47:51<1:27:39, 4.85s/it] 33%\|███▎ \| 538/1621 [47:56<1:27:35, 4.85s/it] 33%\|███▎ \| 539/1621 [48:00<1:26:59, 4.82s/it] 33%\|███▎ \| 540/1621 [48:05<1:27:08, 4.84s/it] 33%\|███▎ \| 540/1621 [48:05<1:27:08, 4.84s/it] 33%\|███▎ \| 541/1621 [48:10<1:26:06, 4.78s/it] 33%\|███▎ \| 542/1621 [48:15<1
	0: {'loss': 0.2184, 'grad_norm': 0.27993941650756166, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.34}
	0: :25:35, 4.76s/it] 33%\|███▎ \| 543/1621 [48:19<1:25:28, 4.76s/it] 34%\|███▎ \| 544/1621 [48:24<1:25:08, 4.74s/it] 34%\|███▎ \| 545/1621 [48:29<1:26:02, 4.80s/it] 34%\|███▎ \| 546/1621 [48:34<1:26:31, 4.83s/it] 34%\|███▎ \| 547/1621 [48:39<1:26:52, 4.85s/it] 34%\|███▍ \| 548/1621 [48:44<1:26:15, 4.82s/it] 34%\|███▍ \| 549/1621 [48:48<1:26:04, 4.82s/it] 34%\|███▍ \| 550/1621 [48:53<1:26:15, 4.83s/it] 34%\|███▍ \| 550/1621 [48:53<1:26:15, 4.83s/it] 34%\|███▍ \| 551/1621 [48:58<1:25:58, 4.82s/it] 34%\|███▍ \| 552/1621 [49:03<1:26:37, 4.86s/it] 34%\|███▍ \| 553/1621 [49:08<1:25:27, 4.80s/it] 34%\|███▍ \| 554/1621 [49:12<1:24:54, 4.77s/it] 34%\|███▍ \| 555/1621 [49:17<1:24:07, 4.74s/it] 34%\|███▍ \| 556/1621 [49:22<1:24:15, 4.75s/it] 34%\|███▍ \| 557/1621 [4
	0: {'loss': 0.2188, 'grad_norm': 0.2717601936897116, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.35}
	0: {'loss': 0.2228, 'grad_norm': 0.2931548668073716, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.35}
	0: 9:27<1:26:01, 4.85s/it] 34%\|███▍ \| 558/1621 [49:32<1:25:55, 4.85s/it] 34%\|███▍ \| 559/1621 [49:37<1:25:56, 4.86s/it] 35%\|███▍ \| 560/1621 [49:42<1:27:39, 4.96s/it] 35%\|███▍ \| 560/1621 [49:42<1:27:39, 4.96s/it] 35%\|███▍ \| 561/1621 [49:47<1:26:40, 4.91s/it] 35%\|███▍ \| 562/1621 [49:53<1:33:42, 5.31s/it] 35%\|███▍ \| 563/1621 [49:58<1:33:17, 5.29s/it] 35%\|███▍ \| 564/1621 [50:03<1:30:03, 5.11s/it] 35%\|███▍ \| 565/1621 [50:08<1:28:36, 5.03s/it] 35%\|███▍ \| 566/1621 [50:12<1:26:43, 4.93s/it] 35%\|███▍ \| 567/1621 [50:18<1:29:34, 5.10s/it] 35%\|███▌ \| 568/1621 [50:23<1:27:18, 4.97s/it] 35%\|███▌ \| 569/1621 [50:27<1:25:47, 4.89s/it] 35%\|███▌ \| 570/1621 [50:32<1:24:39, 4.83s/it] 35%\|███▌ \| 570/1621 [50
	0: {'loss': 0.218, 'grad_norm': 0.2654817797833893, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.36}
	0: :32<1:24:39, 4.83s/it] 35%\|███▌ \| 571/1621 [50:37<1:25:35, 4.89s/it] 35%\|███▌ \| 572/1621 [50:42<1:25:45, 4.90s/it] 35%\|███▌ \| 573/1621 [50:47<1:24:37, 4.84s/it] 35%\|███▌ \| 574/1621 [50:51<1:23:48, 4.80s/it] 35%\|███▌ \| 575/1621 [50:56<1:23:47, 4.81s/it] 36%\|███▌ \| 576/1621 [51:01<1:24:19, 4.84s/it] 36%\|███▌ \| 577/1621 [51:06<1:24:33, 4.86s/it] 36%\|███▌ \| 578/1621 [51:11<1:23:36, 4.81s/it] 36%\|███▌ \| 579/1621 [51:15<1:23:06, 4.79s/it] 36%\|███▌ \| 580/1621 [51:20<1:22:32, 4.76s/it] 36%\|███▌ \| 580/1621 [51:20<1:22:32, 4.76s/it] 36%\|███▌ \| 581/1621 [51:25<1:22:10, 4.74s/it] 36%\|███▌ \| 582/1621 [51:29<1:21:37, 4.71s/it] 36%\|███▌ \| 583/1621 [51:35<1:25:53, 4.96s/it] 36%\|███▌ \| 584/1621 [51:40<1:26:44, 5.02s/it] 36%\|███▌ \| 585/16
	0: {'loss': 0.2206, 'grad_norm': 0.2843657162386269, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.36}
	0: 21 [51:45<1:25:17, 4.94s/it] 36%\|███▌ \| 586/1621 [51:50<1:23:56, 4.87s/it] 36%\|███▌ \| 587/1621 [51:54<1:23:29, 4.84s/it] 36%\|███▋ \| 588/1621 [51:59<1:22:58, 4.82s/it] 36%\|███▋ \| 589/1621 [52:04<1:24:15, 4.90s/it] 36%\|███▋ \| 590/1621 [52:10<1:28:35, 5.16s/it] 36%\|███▋ \| 590/1621 [52:10<1:28:35, 5.16s/it] 36%\|███▋ \| 591/1621 [52:15<1:26:31, 5.04s/it] 37%\|███▋ \| 592/1621 [52:20<1:28:50, 5.18s/it] 37%\|███▋ \| 593/1621 [52:25<1:28:14, 5.15s/it] 37%\|███▋ \| 594/1621 [52:30<1:27:53, 5.13s/it] 37%\|███▋ \| 595/1621 [52:35<1:26:15, 5.04s/it] 37%\|███▋ \| 596/1621 [52:40<1:24:25, 4.94s/it] 37%\|███▋ \| 597/1621 [52:45<1:24:07, 4.93s/it] 37%\|███▋ \| 598/1621 [52:51<1:28:23, 5.18s/it] 37%\|███▋ \| 599/1621 [52:55<1:25:55, 5.04s/it] 37%\|███▋ \|
	0: {'loss': 0.2101, 'grad_norm': 0.27242097410054383, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.37}
	0: {'loss': 0.2175, 'grad_norm': 0.27369902586673506, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.38}
	0: 600/1621 [53:00<1:23:55, 4.93s/it] 37%\|███▋ \| 600/1621 [53:00<1:23:55, 4.93s/it] 37%\|███▋ \| 601/1621 [53:05<1:23:34, 4.92s/it] 37%\|███▋ \| 602/1621 [53:10<1:26:38, 5.10s/it] 37%\|███▋ \| 603/1621 [53:15<1:24:15, 4.97s/it] 37%\|███▋ \| 604/1621 [53:21<1:30:06, 5.32s/it] 37%\|███▋ \| 605/1621 [53:26<1:29:23, 5.28s/it] 37%\|███▋ \| 606/1621 [53:31<1:26:26, 5.11s/it] 37%\|███▋ \| 607/1621 [53:36<1:23:41, 4.95s/it] 38%\|███▊ \| 608/1621 [53:41<1:22:55, 4.91s/it] 38%\|███▊ \| 609/1621 [53:45<1:21:42, 4.84s/it] 38%\|███▊ \| 610/1621 [53:50<1:21:27, 4.83s/it] 38%\|███▊ \| 610/1621 [53:50<1:21:27, 4.83s/it] 38%\|███▊ \| 611/1621 [53:55<1:20:29, 4.78s/it] 38%\|███▊ \| 612/1621 [54:00<1:21:05, 4.82s/it] 38%\|███▊ \| 6
	0: {'loss': 0.2196, 'grad_norm': 0.2717722682759322, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.38}
	0: 13/1621 [54:04<1:20:14, 4.78s/it] 38%\|███▊ \| 614/1621 [54:09<1:19:55, 4.76s/it] 38%\|███▊ \| 615/1621 [54:15<1:26:51, 5.18s/it] 38%\|███▊ \| 616/1621 [54:20<1:25:15, 5.09s/it] 38%\|███▊ \| 617/1621 [54:25<1:23:04, 4.96s/it] 38%\|███▊ \| 618/1621 [54:30<1:23:40, 5.01s/it] 38%\|███▊ \| 619/1621 [54:35<1:22:21, 4.93s/it] 38%\|███▊ \| 620/1621 [54:39<1:20:55, 4.85s/it] 38%\|███▊ \| 620/1621 [54:39<1:20:55, 4.85s/it] 38%\|███▊ \| 621/1621 [54:44<1:19:55, 4.80s/it] 38%\|███▊ \| 622/1621 [54:49<1:19:23, 4.77s/it] 38%\|███▊ \| 623/1621 [54:53<1:19:02, 4.75s/it] 38%\|███▊ \| 624/1621 [54:58<1:20:18, 4.83s/it] 39%\|███▊ \| 625/1621 [55:03<1:19:20, 4.78s/it] 39%\|███▊ \| 626/1621 [55:08<1:20:15, 4.84s/it] 39%\|███▊ \| 627/1621 [55:13<1:19:29, 4.80s/it] 39%\|███▊
	0: {'loss': 0.2176, 'grad_norm': 0.268329487405915, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.39}
	0: {'loss': 0.2193, 'grad_norm': 0.27333488439228737, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.39}
	0: \| 628/1621 [55:18<1:23:40, 5.06s/it] 39%\|███▉ \| 629/1621 [55:23<1:22:12, 4.97s/it] 39%\|███▉ \| 630/1621 [55:28<1:21:34, 4.94s/it] 39%\|███▉ \| 630/1621 [55:28<1:21:34, 4.94s/it] 39%\|███▉ \| 631/1621 [55:33<1:21:46, 4.96s/it] 39%\|███▉ \| 632/1621 [55:38<1:20:16, 4.87s/it] 39%\|███▉ \| 633/1621 [55:43<1:20:45, 4.90s/it] 39%\|███▉ \| 634/1621 [55:48<1:20:58, 4.92s/it] 39%\|███▉ \| 635/1621 [55:52<1:19:54, 4.86s/it] 39%\|███▉ \| 636/1621 [55:57<1:18:56, 4.81s/it] 39%\|███▉ \| 637/1621 [56:02<1:18:56, 4.81s/it] 39%\|███▉ \| 638/1621 [56:08<1:24:26, 5.15s/it] 39%\|███▉ \| 639/1621 [56:12<1:22:13, 5.02s/it] 39%\|███▉ \| 640/1621 [56:17<1:20:30, 4.92s/it] 39%\|███▉ \| 640/1621 [56:17<1:20:30, 4.92s/it] 40%\|███▉
	0: {'loss': 0.2158, 'grad_norm': 0.27556160207006947, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.4}
	0: [2025-09-02 20:01:28,827] [INFO] [axolotl.core.trainers.base._save:613] [PID:2134977] [RANK:0] Saving model checkpoint to /lustre/fswork/projects/rech/dgo/udv55np/math/Qwen3-235B-A22B/Qwen2.5-14B_ift/0/checkpoint-650[39m
	0: [2025-09-02 20:01:54,285] [INFO] [axolotl.core.trainers.base._save:662] [PID:2134977] [RANK:0] Saving Trainer.data_collator.tokenizer by default as Trainer.processing_class is `None`[39m
	0: \| 641/1621 [56:22<1:19:57, 4.90s/it] 40%\|███▉ \| 642/1621 [56:27<1:19:09, 4.85s/it] 40%\|███▉ \| 643/1621 [56:32<1:19:15, 4.86s/it] 40%\|███▉ \| 644/1621 [56:37<1:19:48, 4.90s/it] 40%\|███▉ \| 645/1621 [56:41<1:18:45, 4.84s/it] 40%\|███▉ \| 646/1621 [56:46<1:18:09, 4.81s/it] 40%\|███▉ \| 647/1621 [56:51<1:17:47, 4.79s/it] 40%\|███▉ \| 648/1621 [56:56<1:17:25, 4.77s/it] 40%\|████ \| 649/1621 [57:01<1:21:16, 5.02s/it] 40%\|████ \| 650/1621 [57:06<1:20:10, 4.95s/it] 40%\|████ \| 650/1621 [57:06<1:20:10, 4.95s/it] 40%\|████ \| 651/1621 [57:56<4:59:26, 18.52s/it] 40%\|████ \| 652/1621 [58:02<3:56:36, 14.65s/it] 40%\|████ \| 653/1621 [58:07<3:09:42, 11.76s/it] 40%\|████ \| 654/1621 [58:11<2:35:22, 9.64s/it] 40%\|████ \| 655/1621 [58:16<2:11:15, 8.15s/it] 40%\|███�
	0: {'loss': 0.2128, 'grad_norm': 0.2692142059802366, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.41}
	0: �� \| 656/1621 [58:21<1:54:23, 7.11s/it] 41%\|████ \| 657/1621 [58:26<1:42:34, 6.38s/it] 41%\|████ \| 658/1621 [58:31<1:36:38, 6.02s/it] 41%\|████ \| 659/1621 [58:36<1:31:13, 5.69s/it] 41%\|████ \| 660/1621 [58:41<1:30:11, 5.63s/it] 41%\|████ \| 660/1621 [58:41<1:30:11, 5.63s/it] 41%\|████ \| 661/1621 [58:46<1:26:07, 5.38s/it] 41%\|████ \| 662/1621 [58:51<1:22:57, 5.19s/it] 41%\|████ \| 663/1621 [58:55<1:20:44, 5.06s/it] 41%\|████ \| 664/1621 [59:00<1:19:03, 4.96s/it] 41%\|████ \| 665/1621 [59:05<1:17:34, 4.87s/it] 41%\|████ \| 666/1621 [59:09<1:16:47, 4.82s/it] 41%\|████ \| 667/1621 [59:14<1:16:01, 4.78s/it] 41%\|████ \| 668/1621 [59:20<1:19:48, 5.02s/it] 41%\|████▏ \| 669/1621 [59:25<1:19:00, 4.98s/it] 41%\|████▏ \| 670/1621 [59:29<1:18:08, 4.93s/it]
	0: {'loss': 0.218, 'grad_norm': 0.2715865676756254, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.41}
	0: {'loss': 0.21, 'grad_norm': 0.2775979473941244, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.42}
	0: 41%\|████▏ \| 670/1621 [59:29<1:18:08, 4.93s/it] 41%\|████▏ \| 671/1621 [59:34<1:17:11, 4.88s/it] 41%\|████▏ \| 672/1621 [59:39<1:16:47, 4.86s/it] 42%\|████▏ \| 673/1621 [59:44<1:16:18, 4.83s/it] 42%\|████▏ \| 674/1621 [59:49<1:17:10, 4.89s/it] 42%\|████▏ \| 675/1621 [59:54<1:17:20, 4.91s/it] 42%\|████▏ \| 676/1621 [59:58<1:16:32, 4.86s/it] 42%\|████▏ \| 677/1621 [1:00:04<1:19:44, 5.07s/it] 42%\|████▏ \| 678/1621 [1:00:09<1:17:59, 4.96s/it] 42%\|████▏ \| 679/1621 [1:00:14<1:17:28, 4.93s/it] 42%\|████▏ \| 680/1621 [1:00:19<1:18:21, 5.00s/it] 42%\|████▏ \| 680/1621 [1:00:19<1:18:21, 5.00s/it] 42%\|████▏ \| 681/1621 [1:00:24<1:19:52, 5.10s/it] 42%\|████▏ \| 682/1621 [1:00:29<1:19:02, 5.05s/it] 42%\|████▏
	0: {'loss': 0.2163, 'grad_norm': 0.2767364037640067, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.43}
	0: \| 683/1621 [1:00:34<1:17:50, 4.98s/it] 42%\|████▏ \| 684/1621 [1:00:39<1:16:51, 4.92s/it] 42%\|████▏ \| 685/1621 [1:00:43<1:16:08, 4.88s/it] 42%\|████▏ \| 686/1621 [1:00:49<1:17:43, 4.99s/it] 42%\|████▏ \| 687/1621 [1:00:53<1:16:22, 4.91s/it] 42%\|████▏ \| 688/1621 [1:00:58<1:15:15, 4.84s/it] 43%\|████▎ \| 689/1621 [1:01:03<1:14:19, 4.79s/it] 43%\|████▎ \| 690/1621 [1:01:08<1:17:20, 4.98s/it] 43%\|████▎ \| 690/1621 [1:01:08<1:17:20, 4.98s/it] 43%\|████▎ \| 691/1621 [1:01:13<1:16:10, 4.91s/it] 43%\|████▎ \| 692/1621 [1:01:18<1:16:26, 4.94s/it] 43%\|████▎ \| 693/1621 [1:01:23<1:17:31, 5.01s/it] 43%\|████▎ \| 694/1621 [1:01:28<1:17:42, 5.03s/it] 43%\|████▎ \| 695/1621 [1:01:33<1:18:26, 5.08s/it] 43%\|████▎ \| 696/1621 [1:01:38<1:16:49, 4.98s/it] 43%\|███�
	0: {'loss': 0.2166, 'grad_norm': 0.27600448503199765, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.43}
	0: �▎ \| 697/1621 [1:01:43<1:15:41, 4.92s/it] 43%\|████▎ \| 698/1621 [1:01:48<1:14:51, 4.87s/it] 43%\|████▎ \| 699/1621 [1:01:52<1:14:18, 4.84s/it] 43%\|████▎ \| 700/1621 [1:01:57<1:14:01, 4.82s/it] 43%\|████▎ \| 700/1621 [1:01:57<1:14:01, 4.82s/it] 43%\|████▎ \| 701/1621 [1:02:02<1:13:37, 4.80s/it] 43%\|████▎ \| 702/1621 [1:02:07<1:13:18, 4.79s/it] 43%\|████▎ \| 703/1621 [1:02:11<1:12:59, 4.77s/it] 43%\|████▎ \| 704/1621 [1:02:16<1:12:36, 4.75s/it] 43%\|████▎ \| 705/1621 [1:02:21<1:12:12, 4.73s/it] 44%\|████▎ \| 706/1621 [1:02:26<1:12:38, 4.76s/it] 44%\|████▎ \| 707/1621 [1:02:30<1:12:30, 4.76s/it] 44%\|████▎ \| 708/1621 [1:02:35<1:12:27, 4.76s/it] 44%\|████▎ \| 709/1621 [1:02:40<1:14:21, 4.89s/it] 44%\|████▍ \| 710/1621 [1:02:45<1:14:01, 4.88s/it]
	0: {'loss': 0.2181, 'grad_norm': 0.3437401810798713, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.44}
	0: {'loss': 0.2182, 'grad_norm': 0.2802402195008618, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.44}
	0: 44%\|████▍ \| 710/1621 [1:02:45<1:14:01, 4.88s/it] 44%\|████▍ \| 711/1621 [1:02:50<1:13:06, 4.82s/it] 44%\|████▍ \| 712/1621 [1:02:55<1:12:22, 4.78s/it] 44%\|████▍ \| 713/1621 [1:02:59<1:12:32, 4.79s/it] 44%\|████▍ \| 714/1621 [1:03:04<1:11:54, 4.76s/it] 44%\|████▍ \| 715/1621 [1:03:09<1:11:54, 4.76s/it] 44%\|████▍ \| 716/1621 [1:03:15<1:18:56, 5.23s/it] 44%\|████▍ \| 717/1621 [1:03:20<1:16:35, 5.08s/it] 44%\|████▍ \| 718/1621 [1:03:25<1:14:54, 4.98s/it] 44%\|████▍ \| 719/1621 [1:03:29<1:13:43, 4.90s/it] 44%\|████▍ \| 720/1621 [1:03:34<1:12:51, 4.85s/it] 44%\|████▍ \| 720/1621 [1:03:34<1:12:51, 4.85s/it] 44%\|████▍ \| 721/1621 [1:03:39<1:12:50, 4.86s/it] 45%\|████▍ \| 722/1621 [1:03:44<1:11:43, 4.79s/it] 45%\|██
	0: {'loss': 0.2153, 'grad_norm': 0.32129231597397556, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.45}
	0: ██▍ \| 723/1621 [1:03:48<1:11:30, 4.78s/it] 45%\|████▍ \| 724/1621 [1:03:53<1:10:43, 4.73s/it] 45%\|████▍ \| 725/1621 [1:03:58<1:11:14, 4.77s/it] 45%\|████▍ \| 726/1621 [1:04:03<1:11:05, 4.77s/it] 45%\|████▍ \| 727/1621 [1:04:07<1:11:29, 4.80s/it] 45%\|████▍ \| 728/1621 [1:04:12<1:11:53, 4.83s/it] 45%\|████▍ \| 729/1621 [1:04:17<1:11:07, 4.78s/it] 45%\|████▌ \| 730/1621 [1:04:22<1:12:59, 4.92s/it] 45%\|████▌ \| 730/1621 [1:04:22<1:12:59, 4.92s/it] 45%\|████▌ \| 731/1621 [1:04:27<1:12:23, 4.88s/it] 45%\|████▌ \| 732/1621 [1:04:33<1:14:50, 5.05s/it] 45%\|████▌ \| 733/1621 [1:04:38<1:14:40, 5.05s/it] 45%\|████▌ \| 734/1621 [1:04:44<1:18:41, 5.32s/it] 45%\|████▌ \| 735/1621 [1:04:48<1:16:10, 5.16s/it] 45%\|████▌ \| 736/1621 [1:04:54<1:17:59, 5.29s/it] 45%
	0: {'loss': 0.2121, 'grad_norm': 0.2642878114528468, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.46}
	0: \|████▌ \| 737/1621 [1:05:00<1:21:28, 5.53s/it] 46%\|████▌ \| 738/1621 [1:05:05<1:17:52, 5.29s/it] 46%\|████▌ \| 739/1621 [1:05:10<1:17:13, 5.25s/it] 46%\|████▌ \| 740/1621 [1:05:16<1:19:31, 5.42s/it] 46%\|████▌ \| 740/1621 [1:05:16<1:19:31, 5.42s/it] 46%\|████▌ \| 741/1621 [1:05:21<1:18:24, 5.35s/it] 46%\|████▌ \| 742/1621 [1:05:26<1:16:18, 5.21s/it] 46%\|████▌ \| 743/1621 [1:05:30<1:13:52, 5.05s/it] 46%\|████▌ \| 744/1621 [1:05:35<1:12:34, 4.97s/it] 46%\|████▌ \| 745/1621 [1:05:40<1:11:15, 4.88s/it] 46%\|████▌ \| 746/1621 [1:05:45<1:10:42, 4.85s/it] 46%\|████▌ \| 747/1621 [1:05:50<1:11:29, 4.91s/it] 46%\|████▌ \| 748/1621 [1:05:55<1:14:49, 5.14s/it] 46%\|████▌ \| 749/1621 [1:06:00<1:12:41, 5.00s/it] 46%\|████▋ \| 750/1621 [1:06:05<1:11:17, 4.91s/i
	0: {'loss': 0.212, 'grad_norm': 0.28681062888229386, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.46}
	0: {'loss': 0.2132, 'grad_norm': 0.27795940147223497, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.47}
	0: t] 46%\|████▋ \| 750/1621 [1:06:05<1:11:17, 4.91s/it] 46%\|████▋ \| 751/1621 [1:06:09<1:10:18, 4.85s/it] 46%\|████▋ \| 752/1621 [1:06:14<1:10:40, 4.88s/it] 46%\|████▋ \| 753/1621 [1:06:19<1:10:35, 4.88s/it] 47%\|████▋ \| 754/1621 [1:06:24<1:11:41, 4.96s/it] 47%\|████▋ \| 755/1621 [1:06:29<1:11:26, 4.95s/it] 47%\|████▋ \| 756/1621 [1:06:34<1:10:25, 4.88s/it] 47%\|████▋ \| 757/1621 [1:06:39<1:09:33, 4.83s/it] 47%\|████▋ \| 758/1621 [1:06:44<1:09:47, 4.85s/it] 47%\|████▋ \| 759/1621 [1:06:49<1:10:01, 4.87s/it] 47%\|████▋ \| 760/1621 [1:06:53<1:09:45, 4.86s/it] 47%\|████▋ \| 760/1621 [1:06:53<1:09:45, 4.86s/it] 47%\|████▋ \| 761/1621 [1:06:58<1:09:38, 4.86s/it] 47%\|████▋ \| 762/1621 [1:07:03<1:08:27, 4.78s/it]
	0: {'loss': 0.2172, 'grad_norm': 0.297594215387472, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.48}
	0: 47%\|████▋ \| 763/1621 [1:07:08<1:08:50, 4.81s/it] 47%\|████▋ \| 764/1621 [1:07:13<1:08:20, 4.78s/it] 47%\|████▋ \| 765/1621 [1:07:17<1:08:14, 4.78s/it] 47%\|████▋ \| 766/1621 [1:07:23<1:11:12, 5.00s/it] 47%\|████▋ \| 767/1621 [1:07:28<1:10:15, 4.94s/it] 47%\|████▋ \| 768/1621 [1:07:33<1:11:39, 5.04s/it] 47%\|████▋ \| 769/1621 [1:07:38<1:11:09, 5.01s/it] 48%\|████▊ \| 770/1621 [1:07:43<1:10:56, 5.00s/it] 48%\|████▊ \| 770/1621 [1:07:43<1:10:56, 5.00s/it] 48%\|████▊ \| 771/1621 [1:07:48<1:10:48, 5.00s/it] 48%\|████▊ \| 772/1621 [1:07:53<1:09:33, 4.92s/it] 48%\|████▊ \| 773/1621 [1:07:57<1:08:40, 4.86s/it] 48%\|████▊ \| 774/1621 [1:08:03<1:11:30, 5.07s/it] 48%\|████▊ \| 775/1621 [1:08:08<1:10:31, 5.00s/it] 48%\|████▊ \| 776/1621 [1:08:12<1:09:39, 4.
	0: {'loss': 0.2144, 'grad_norm': 0.28602779425213437, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.48}
	0: 95s/it] 48%\|████▊ \| 777/1621 [1:08:17<1:08:50, 4.89s/it] 48%\|████▊ \| 778/1621 [1:08:22<1:07:58, 4.84s/it] 48%\|████▊ \| 779/1621 [1:08:27<1:07:45, 4.83s/it] 48%\|████▊ \| 780/1621 [1:08:32<1:10:57, 5.06s/it] 48%\|████▊ \| 780/1621 [1:08:32<1:10:57, 5.06s/it] 48%\|████▊ \| 781/1621 [1:08:37<1:09:54, 4.99s/it] 48%\|████▊ \| 782/1621 [1:08:43<1:11:07, 5.09s/it] 48%\|████▊ \| 783/1621 [1:08:47<1:09:34, 4.98s/it] 48%\|████▊ \| 784/1621 [1:08:52<1:08:36, 4.92s/it] 48%\|████▊ \| 785/1621 [1:08:57<1:07:59, 4.88s/it] 48%\|████▊ \| 786/1621 [1:09:02<1:10:26, 5.06s/it] 49%\|████▊ \| 787/1621 [1:09:07<1:09:00, 4.96s/it] 49%\|████▊ \| 788/1621 [1:09:12<1:07:41, 4.88s/it] 49%\|████▊ \| 789/1621 [1:09:16<1:07:02, 4.83s/it] 49%\|████▊ \| 790/1621 [1:09:21<1:06:
	0: {'loss': 0.2207, 'grad_norm': 0.27599624793574346, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.49}
	0: {'loss': 0.2172, 'grad_norm': 0.26979823876142667, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.49}
	0: 10, 4.78s/it] 49%\|████▊ \| 790/1621 [1:09:21<1:06:10, 4.78s/it] 49%\|████▉ \| 791/1621 [1:09:26<1:05:51, 4.76s/it] 49%\|████▉ \| 792/1621 [1:09:31<1:06:21, 4.80s/it] 49%\|████▉ \| 793/1621 [1:09:35<1:05:57, 4.78s/it] 49%\|████▉ \| 794/1621 [1:09:40<1:05:20, 4.74s/it] 49%\|████▉ \| 795/1621 [1:09:45<1:05:12, 4.74s/it] 49%\|████▉ \| 796/1621 [1:09:50<1:05:19, 4.75s/it] 49%\|████▉ \| 797/1621 [1:09:54<1:05:48, 4.79s/it] 49%\|████▉ \| 798/1621 [1:10:00<1:07:00, 4.88s/it] 49%\|████▉ \| 799/1621 [1:10:04<1:06:33, 4.86s/it] 49%\|████▉ \| 800/1621 [1:10:09<1:05:51, 4.81s/it] 49%\|████▉ \| 800/1621 [1:10:09<1:05:51, 4.81s/it] 49%\|████▉ \| 801/1621 [1:10:14<1:05:46, 4.81s/it] 49%\|████▉ \| 802/1621 [1:10:19<1:05:04
	0: {'loss': 0.2217, 'grad_norm': 0.29824386294830224, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.5}
	0: , 4.77s/it] 50%\|████▉ \| 803/1621 [1:10:23<1:04:48, 4.75s/it] 50%\|████▉ \| 804/1621 [1:10:28<1:04:31, 4.74s/it] 50%\|████▉ \| 805/1621 [1:10:33<1:04:26, 4.74s/it] 50%\|████▉ \| 806/1621 [1:10:38<1:05:44, 4.84s/it] 50%\|████▉ \| 807/1621 [1:10:43<1:05:30, 4.83s/it] 50%\|████▉ \| 808/1621 [1:10:47<1:05:25, 4.83s/it] 50%\|████▉ \| 809/1621 [1:10:52<1:05:15, 4.82s/it] 50%\|████▉ \| 810/1621 [1:10:57<1:04:50, 4.80s/it] 50%\|████▉ \| 810/1621 [1:10:57<1:04:50, 4.80s/it] 50%\|█████ \| 811/1621 [1:11:02<1:06:10, 4.90s/it] 50%\|█████ \| 812/1621 [1:11:07<1:06:16, 4.92s/it] 50%\|█████ \| 813/1621 [1:11:12<1:05:49, 4.89s/it] 50%\|█████ \| 814/1621 [1:11:17<1:04:55, 4.83s/it] 50%\|█████ \| 815/1621 [1:11:21<1:04:23, 4.79s/it] 50%\|█████ \| 816/1621 [1:11:26<
	0: {'loss': 0.2133, 'grad_norm': 0.26092437055105083, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.51}
	0: 1:04:56, 4.84s/it] 50%\|█████ \| 817/1621 [1:11:32<1:07:24, 5.03s/it] 50%\|█████ \| 818/1621 [1:11:37<1:07:57, 5.08s/it] 51%\|█████ \| 819/1621 [1:11:42<1:07:14, 5.03s/it] 51%\|█████ \| 820/1621 [1:11:47<1:05:46, 4.93s/it] 51%\|█████ \| 820/1621 [1:11:47<1:05:46, 4.93s/it] 51%\|█████ \| 821/1621 [1:11:51<1:05:12, 4.89s/it] 51%\|█████ \| 822/1621 [1:11:56<1:05:34, 4.92s/it] 51%\|█████ \| 823/1621 [1:12:02<1:07:41, 5.09s/it] 51%\|█████ \| 824/1621 [1:12:07<1:06:10, 4.98s/it] 51%\|█████ \| 825/1621 [1:12:11<1:05:42, 4.95s/it] 51%\|█████ \| 826/1621 [1:12:16<1:04:53, 4.90s/it] 51%\|█████ \| 827/1621 [1:12:21<1:04:15, 4.86s/it] 51%\|█████ \| 828/1621 [1:12:26<1:03:27, 4.80s/it] 51%\|█████ \| 829/1621 [1:12:31<1:05:55, 4.99s/it] 51%\|█████ \| 830/1621 [1
	0: {'loss': 0.2158, 'grad_norm': 0.26693057475276766, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.51}
	0: {'loss': 0.2121, 'grad_norm': 0.27091788132485156, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.52}
	0: :12:36<1:06:32, 5.05s/it] 51%\|█████ \| 830/1621 [1:12:36<1:06:32, 5.05s/it] 51%\|█████▏ \| 831/1621 [1:12:41<1:05:22, 4.97s/it] 51%\|█████▏ \| 832/1621 [1:12:46<1:05:15, 4.96s/it] 51%\|█████▏ \| 833/1621 [1:12:51<1:04:33, 4.92s/it] 51%\|█████▏ \| 834/1621 [1:12:56<1:06:56, 5.10s/it] 52%\|█████▏ \| 835/1621 [1:13:02<1:08:24, 5.22s/it] 52%\|█████▏ \| 836/1621 [1:13:07<1:09:48, 5.34s/it] 52%\|█████▏ \| 837/1621 [1:13:12<1:08:07, 5.21s/it] 52%\|█████▏ \| 838/1621 [1:13:17<1:05:50, 5.05s/it] 52%\|█████▏ \| 839/1621 [1:13:22<1:05:20, 5.01s/it] 52%\|█████▏ \| 840/1621 [1:13:27<1:05:29, 5.03s/it] 52%\|█████▏ \| 840/1621 [1:13:27<1:05:29, 5.03s/it] 52%\|█████▏ \| 841/1621 [1:13:32<1:04:19, 4.95s/it] 52%\|███�
	0: {'loss': 0.2139, 'grad_norm': 0.2710407718400621, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.52}
	0: �█▏ \| 842/1621 [1:13:37<1:04:39, 4.98s/it] 52%\|█████▏ \| 843/1621 [1:13:41<1:03:28, 4.90s/it] 52%\|█████▏ \| 844/1621 [1:13:46<1:03:45, 4.92s/it] 52%\|█████▏ \| 845/1621 [1:13:51<1:02:56, 4.87s/it] 52%\|█████▏ \| 846/1621 [1:13:56<1:02:12, 4.82s/it] 52%\|█████▏ \| 847/1621 [1:14:01<1:02:51, 4.87s/it] 52%\|█████▏ \| 848/1621 [1:14:06<1:01:57, 4.81s/it] 52%\|█████▏ \| 849/1621 [1:14:10<1:01:38, 4.79s/it] 52%\|█████▏ \| 850/1621 [1:14:15<1:01:09, 4.76s/it] 52%\|█████▏ \| 850/1621 [1:14:15<1:01:09, 4.76s/it] 52%\|█████▏ \| 851/1621 [1:14:20<1:01:00, 4.75s/it] 53%\|█████▎ \| 852/1621 [1:14:24<1:00:45, 4.74s/it] 53%\|█████▎ \| 853/1621 [1:14:29<1:00:52, 4.76s/it] 53%\|█████▎ \| 854/1621 [1:14:34<1:00:33, 4.74s/it] 53%\|█████▎ \| 855/1621 [1:14:39
	0: {'loss': 0.215, 'grad_norm': 0.26826181484921574, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.53}
	0: <1:00:18, 4.72s/it] 53%\|█████▎ \| 856/1621 [1:14:43<59:57, 4.70s/it] 53%\|█████▎ \| 857/1621 [1:14:48<1:00:13, 4.73s/it] 53%\|█████▎ \| 858/1621 [1:14:53<1:00:54, 4.79s/it] 53%\|█████▎ \| 859/1621 [1:14:58<1:01:15, 4.82s/it] 53%\|█████▎ \| 860/1621 [1:15:03<1:00:56, 4.80s/it] 53%\|█████▎ \| 860/1621 [1:15:03<1:00:56, 4.80s/it] 53%\|█████▎ \| 861/1621 [1:15:07<1:00:53, 4.81s/it] 53%\|█████▎ \| 862/1621 [1:15:13<1:02:00, 4.90s/it] 53%\|█████▎ \| 863/1621 [1:15:17<1:01:14, 4.85s/it] 53%\|█████▎ \| 864/1621 [1:15:22<1:00:51, 4.82s/it] 53%\|█████▎ \| 865/1621 [1:15:27<1:01:35, 4.89s/it] 53%\|█████▎ \| 866/1621 [1:15:32<1:00:50, 4.84s/it] 53%\|█████▎ \| 867/1621 [1:15:37<1:00:14, 4.79s/it] 54%\|█████▎ \| 868/1621 [1:15:41<59:43, 4.76s/it] 54%\|█�
	0: {'loss': 0.2139, 'grad_norm': 0.2852708208541685, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.54}
	0: {'loss': 0.2151, 'grad_norm': 0.28550830834224616, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.54}
	0: ��███▎ \| 869/1621 [1:15:46<59:47, 4.77s/it] 54%\|█████▎ \| 870/1621 [1:15:51<1:01:12, 4.89s/it] 54%\|█████▎ \| 870/1621 [1:15:51<1:01:12, 4.89s/it] 54%\|█████▎ \| 871/1621 [1:15:56<1:00:21, 4.83s/it] 54%\|█████▍ \| 872/1621 [1:16:01<59:39, 4.78s/it] 54%\|█████▍ \| 873/1621 [1:16:05<59:39, 4.79s/it] 54%\|█████▍ \| 874/1621 [1:16:10<1:00:18, 4.84s/it] 54%\|█████▍ \| 875/1621 [1:16:16<1:03:16, 5.09s/it] 54%\|█████▍ \| 876/1621 [1:16:21<1:01:48, 4.98s/it] 54%\|█████▍ \| 877/1621 [1:16:25<1:00:49, 4.91s/it] 54%\|█████▍ \| 878/1621 [1:16:30<1:00:14, 4.87s/it] 54%\|█████▍ \| 879/1621 [1:16:35<59:44, 4.83s/it] 54%\|█████▍ \| 880/1621 [1:16:40<59:11, 4.79s/it] 54%\|█████▍ \| 880/1621 [1:16:40<59:11, 4.7
	0: {'loss': 0.2192, 'grad_norm': 0.2804935280560918, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.55}
	0: 9s/it] 54%\|█████▍ \| 881/1621 [1:16:44<59:14, 4.80s/it] 54%\|█████▍ \| 882/1621 [1:16:49<58:58, 4.79s/it] 54%\|█████▍ \| 883/1621 [1:16:54<58:17, 4.74s/it] 55%\|█████▍ \| 884/1621 [1:16:59<58:07, 4.73s/it] 55%\|█████▍ \| 885/1621 [1:17:04<1:01:35, 5.02s/it] 55%\|█████▍ \| 886/1621 [1:17:09<1:00:15, 4.92s/it] 55%\|█████▍ \| 887/1621 [1:17:14<1:02:01, 5.07s/it] 55%\|█████▍ \| 888/1621 [1:17:19<1:00:51, 4.98s/it] 55%\|█████▍ \| 889/1621 [1:17:24<59:45, 4.90s/it] 55%\|█████▍ \| 890/1621 [1:17:29<58:57, 4.84s/it] 55%\|█████▍ \| 890/1621 [1:17:29<58:57, 4.84s/it] 55%\|█████▍ \| 891/1621 [1:17:35<1:02:59, 5.18s/it] 55%\|█████▌ \| 892/1621 [1:17:40<1:03:56, 5.26s/it] 55%\|█████▌ \| 893/1621 [1:17:45<1:02:07, 5.12s/it] 55%\|█████▌ \| 894/1621
	0: {'loss': 0.2126, 'grad_norm': 0.2750084705232265, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.56}
	0: [1:17:50<1:01:53, 5.11s/it] 55%\|█████▌ \| 895/1621 [1:17:55<1:03:00, 5.21s/it] 55%\|█████▌ \| 896/1621 [1:18:00<1:01:04, 5.05s/it] 55%\|█████▌ \| 897/1621 [1:18:05<1:01:02, 5.06s/it] 55%\|█████▌ \| 898/1621 [1:18:10<59:52, 4.97s/it] 55%\|█████▌ \| 899/1621 [1:18:15<59:42, 4.96s/it] 56%\|█████▌ \| 900/1621 [1:18:19<58:36, 4.88s/it] 56%\|█████▌ \| 900/1621 [1:18:19<58:36, 4.88s/it] 56%\|█████▌ \| 901/1621 [1:18:24<57:40, 4.81s/it] 56%\|█████▌ \| 902/1621 [1:18:29<57:10, 4.77s/it] 56%\|█████▌ \| 903/1621 [1:18:34<56:54, 4.76s/it] 56%\|█████▌ \| 904/1621 [1:18:38<56:28, 4.73s/it] 56%\|█████▌ \| 905/1621 [1:18:44<59:56, 5.02s/it] 56%\|█████▌ \| 906/1621 [1:18:49<59:04, 4.96s/it] 56%\|█████▌ \| 907/1621 [1:18:53<58:22, 4.90s/it] 56%\|█████�
	0: {'loss': 0.2147, 'grad_norm': 0.3443521809729604, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.56}
	0: {'loss': 0.2137, 'grad_norm': 0.27078378526006086, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.57}
	0: � \| 908/1621 [1:18:58<57:27, 4.84s/it] 56%\|█████▌ \| 909/1621 [1:19:03<57:09, 4.82s/it] 56%\|█████▌ \| 910/1621 [1:19:09<59:59, 5.06s/it] 56%\|█████▌ \| 910/1621 [1:19:09<59:59, 5.06s/it] 56%\|█████▌ \| 911/1621 [1:19:14<59:47, 5.05s/it] 56%\|█████▋ \| 912/1621 [1:19:18<59:03, 5.00s/it] 56%\|█████▋ \| 913/1621 [1:19:23<57:41, 4.89s/it] 56%\|█████▋ \| 914/1621 [1:19:28<56:49, 4.82s/it] 56%\|█████▋ \| 915/1621 [1:19:33<57:46, 4.91s/it] 57%\|█████▋ \| 916/1621 [1:19:38<59:06, 5.03s/it] 57%\|█████▋ \| 917/1621 [1:19:43<57:45, 4.92s/it] 57%\|█████▋ \| 918/1621 [1:19:48<56:50, 4.85s/it] 57%\|█████▋ \| 919/1621 [1:19:52<56:21, 4.82s/it] 57%\|█████▋ \| 920/1621 [1:19:57<56:55, 4.87s/it] 57%\|█████▋ \|
	0: {'loss': 0.2147, 'grad_norm': 0.31227860783495803, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.57}
	0: 920/1621 [1:19:57<56:55, 4.87s/it] 57%\|█████▋ \| 921/1621 [1:20:02<56:13, 4.82s/it] 57%\|█████▋ \| 922/1621 [1:20:07<56:18, 4.83s/it] 57%\|█████▋ \| 923/1621 [1:20:12<55:55, 4.81s/it] 57%\|█████▋ \| 924/1621 [1:20:16<55:51, 4.81s/it] 57%\|█████▋ \| 925/1621 [1:20:21<55:29, 4.78s/it] 57%\|█████▋ \| 926/1621 [1:20:26<55:23, 4.78s/it] 57%\|█████▋ \| 927/1621 [1:20:31<55:01, 4.76s/it] 57%\|█████▋ \| 928/1621 [1:20:35<54:46, 4.74s/it] 57%\|█████▋ \| 929/1621 [1:20:41<56:43, 4.92s/it] 57%\|█████▋ \| 930/1621 [1:20:47<1:01:18, 5.32s/it] 57%\|█████▋ \| 930/1621 [1:20:47<1:01:18, 5.32s/it] 57%\|█████▋ \| 931/1621 [1:20:52<1:00:00, 5.22s/it] 57%\|█████▋ \| 932/1621 [1:20:57<58:17, 5.08s/it] 58%\|█████▊ \| 933/1621 [1:21:01<56:53, 4.96s/it] 58%\|██�
	0: {'loss': 0.2184, 'grad_norm': 0.25947362686664, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.58}
	0: �██▊ \| 934/1621 [1:21:06<55:49, 4.88s/it] 58%\|█████▊ \| 935/1621 [1:21:11<55:13, 4.83s/it] 58%\|█████▊ \| 936/1621 [1:21:16<55:14, 4.84s/it] 58%\|█████▊ \| 937/1621 [1:21:20<55:23, 4.86s/it] 58%\|█████▊ \| 938/1621 [1:21:25<55:05, 4.84s/it] 58%\|█████▊ \| 939/1621 [1:21:30<54:32, 4.80s/it] 58%\|█████▊ \| 940/1621 [1:21:36<57:50, 5.10s/it] 58%\|█████▊ \| 940/1621 [1:21:36<57:50, 5.10s/it] 58%\|█████▊ \| 941/1621 [1:21:41<57:11, 5.05s/it] 58%\|█████▊ \| 942/1621 [1:21:46<57:45, 5.10s/it] 58%\|█████▊ \| 943/1621 [1:21:51<57:45, 5.11s/it] 58%\|█████▊ \| 944/1621 [1:21:56<56:35, 5.02s/it] 58%\|█████▊ \| 945/1621 [1:22:01<55:18, 4.91s/it] 58%\|█████▊ \| 946/1621 [1:22:05<54:59, 4.89s/it] 58%\|█████▊ \| 947/1621 [1:22:10<55:18, 4.92s/it] 58%\|█
	0: {'loss': 0.2102, 'grad_norm': 0.3142685936794017, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.59}
	0: {'loss': 0.2198, 'grad_norm': 0.29152014419018757, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.59}
	0: ████▊ \| 948/1621 [1:22:15<54:35, 4.87s/it] 59%\|█████▊ \| 949/1621 [1:22:20<54:45, 4.89s/it] 59%\|█████▊ \| 950/1621 [1:22:25<55:06, 4.93s/it] 59%\|█████▊ \| 950/1621 [1:22:25<55:06, 4.93s/it] 59%\|█████▊ \| 951/1621 [1:22:31<58:40, 5.25s/it] 59%\|█████▊ \| 952/1621 [1:22:36<57:28, 5.15s/it] 59%\|█████▉ \| 953/1621 [1:22:41<58:29, 5.25s/it] 59%\|█████▉ \| 954/1621 [1:22:46<56:46, 5.11s/it] 59%\|█████▉ \| 955/1621 [1:22:51<55:42, 5.02s/it] 59%\|█████▉ \| 956/1621 [1:22:56<54:28, 4.91s/it] 59%\|█████▉ \| 957/1621 [1:23:01<55:41, 5.03s/it] 59%\|█████▉ \| 958/1621 [1:23:06<56:48, 5.14s/it] 59%\|█████▉ \| 959/1621 [1:23:12<58:42, 5.32s/it] 59%\|█████▉ \| 960/1621 [1:23:17<56:52, 5.16s/it] 59%\|███
	0: {'loss': 0.2175, 'grad_norm': 0.28132541400191075, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.6}
	0: ██▉ \| 960/1621 [1:23:17<56:52, 5.16s/it] 59%\|█████▉ \| 961/1621 [1:23:22<55:30, 5.05s/it] 59%\|█████▉ \| 962/1621 [1:23:27<55:10, 5.02s/it] 59%\|█████▉ \| 963/1621 [1:23:31<54:19, 4.95s/it] 59%\|█████▉ \| 964/1621 [1:23:36<53:41, 4.90s/it] 60%\|█████▉ \| 965/1621 [1:23:41<53:11, 4.86s/it] 60%\|█████▉ \| 966/1621 [1:23:46<52:29, 4.81s/it] 60%\|█████▉ \| 967/1621 [1:23:50<51:54, 4.76s/it] 60%\|█████▉ \| 968/1621 [1:23:55<52:18, 4.81s/it] 60%\|█████▉ \| 969/1621 [1:24:00<52:50, 4.86s/it] 60%\|█████▉ \| 970/1621 [1:24:05<52:50, 4.87s/it] 60%\|█████▉ \| 970/1621 [1:24:05<52:50, 4.87s/it] 60%\|█████▉ \| 971/1621 [1:24:10<53:44, 4.96s/it] 60%\|█████▉ \| 972/1621 [1:24:15<53:01, 4.90s/it] 60%\|██████ \| 973/1621 [1:24:20<52:22, 4.85s/it] 60%\|█�
	0: [2025-09-02 20:28:52,563] [INFO] [axolotl.core.trainers.base._save:613] [PID:2134977] [RANK:0] Saving model checkpoint to /lustre/fswork/projects/rech/dgo/udv55np/math/Qwen3-235B-A22B/Qwen2.5-14B_ift/0/checkpoint-975[39m
	0: [2025-09-02 20:29:15,164] [INFO] [axolotl.core.trainers.base._save:662] [PID:2134977] [RANK:0] Saving Trainer.data_collator.tokenizer by default as Trainer.processing_class is `None`[39m
	0: {'loss': 0.2136, 'grad_norm': 0.27352537796541654, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.6}
	0: ��████ \| 974/1621 [1:24:25<52:18, 4.85s/it] 60%\|██████ \| 975/1621 [1:24:30<52:06, 4.84s/it] 60%\|██████ \| 976/1621 [1:25:16<3:06:42, 17.37s/it] 60%\|██████ \| 977/1621 [1:25:21<2:26:43, 13.67s/it] 60%\|██████ \| 978/1621 [1:25:26<1:57:50, 11.00s/it] 60%\|██████ \| 979/1621 [1:25:31<1:37:29, 9.11s/it] 60%\|██████ \| 980/1621 [1:25:35<1:23:19, 7.80s/it] 60%\|██████ \| 980/1621 [1:25:35<1:23:19, 7.80s/it] 61%\|██████ \| 981/1621 [1:25:40<1:13:33, 6.90s/it] 61%\|██████ \| 982/1621 [1:25:45<1:06:28, 6.24s/it] 61%\|██████ \| 983/1621 [1:25:50<1:04:08, 6.03s/it] 61%\|██████ \| 984/1621 [1:25:55<59:58, 5.65s/it] 61%\|██████ \| 985/1621 [1:26:00<57:07, 5.39s/it] 61%\|██████ \| 986/1621 [1:26:05<54:50, 5.18s/it] 61%\|██████ \| 987/1621 [1:26:09<
	0: {'loss': 0.2176, 'grad_norm': 0.2789264204643204, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.61}
	0: 53:25, 5.06s/it] 61%\|██████ \| 988/1621 [1:26:14<52:22, 4.96s/it] 61%\|██████ \| 989/1621 [1:26:20<53:35, 5.09s/it] 61%\|██████ \| 990/1621 [1:26:24<52:29, 4.99s/it] 61%\|██████ \| 990/1621 [1:26:24<52:29, 4.99s/it] 61%\|██████ \| 991/1621 [1:26:29<51:43, 4.93s/it] 61%\|██████ \| 992/1621 [1:26:35<53:42, 5.12s/it] 61%\|██████▏ \| 993/1621 [1:26:39<52:30, 5.02s/it] 61%\|██████▏ \| 994/1621 [1:26:44<51:25, 4.92s/it] 61%\|██████▏ \| 995/1621 [1:26:49<51:13, 4.91s/it] 61%\|██████▏ \| 996/1621 [1:26:54<52:44, 5.06s/it] 62%\|██████▏ \| 997/1621 [1:27:00<53:05, 5.11s/it] 62%\|██████▏ \| 998/1621 [1:27:06<56:20, 5.43s/it] 62%\|██████▏ \| 999/1621 [1:27:11<54:31, 5.26s/it] 62%\|██████▏ \| 1000/1621 [1:27:15<52:34, 5.08s/it]
	0: {'loss': 0.2141, 'grad_norm': 0.2810459322487379, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.62}
	0: {'loss': 0.2124, 'grad_norm': 0.28919272828442305, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.62}
	0: 62%\|██████▏ \| 1000/1621 [1:27:15<52:34, 5.08s/it] 62%\|██████▏ \| 1001/1621 [1:27:20<51:53, 5.02s/it] 62%\|██████▏ \| 1002/1621 [1:27:25<50:58, 4.94s/it] 62%\|██████▏ \| 1003/1621 [1:27:30<50:05, 4.86s/it] 62%\|██████▏ \| 1004/1621 [1:27:34<49:16, 4.79s/it] 62%\|██████▏ \| 1005/1621 [1:27:40<52:54, 5.15s/it] 62%\|██████▏ \| 1006/1621 [1:27:46<53:37, 5.23s/it] 62%\|██████▏ \| 1007/1621 [1:27:52<55:23, 5.41s/it] 62%\|██████▏ \| 1008/1621 [1:27:57<54:46, 5.36s/it] 62%\|██████▏ \| 1009/1621 [1:28:02<55:30, 5.44s/it] 62%\|██████▏ \| 1010/1621 [1:28:07<53:01, 5.21s/it] 62%\|██████▏ \| 1010/1621 [1:28:07<53:01, 5.21s/it] 62%\|██████▏ \| 1011/1621 [1:28:12<51:35, 5.07s/it] 62%\|██████▏ \| 1012/1621 [1:28:17<50:41
	0: {'loss': 0.2135, 'grad_norm': 0.27980806989132895, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.63}
	0: , 4.99s/it] 62%\|██████▏ \| 1013/1621 [1:28:21<49:59, 4.93s/it] 63%\|██████▎ \| 1014/1621 [1:28:26<49:10, 4.86s/it] 63%\|██████▎ \| 1015/1621 [1:28:31<48:49, 4.83s/it] 63%\|██████▎ \| 1016/1621 [1:28:36<48:30, 4.81s/it] 63%\|██████▎ \| 1017/1621 [1:28:40<48:10, 4.79s/it] 63%\|██████▎ \| 1018/1621 [1:28:45<48:20, 4.81s/it] 63%\|██████▎ \| 1019/1621 [1:28:50<49:02, 4.89s/it] 63%\|██████▎ \| 1020/1621 [1:28:55<48:16, 4.82s/it] 63%\|██████▎ \| 1020/1621 [1:28:55<48:16, 4.82s/it] 63%\|██████▎ \| 1021/1621 [1:29:00<47:51, 4.79s/it] 63%\|██████▎ \| 1022/1621 [1:29:04<47:49, 4.79s/it] 63%\|██████▎ \| 1023/1621 [1:29:09<47:21, 4.75s/it] 63%\|██████▎ \| 1024/1621 [1:29:15<49:10, 4.94s/it] 63%\|██████▎ \| 1025/1621 [1:29:19<48:24, 4.87s/it] 63%
	0: {'loss': 0.2148, 'grad_norm': 0.29552995472458155, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.64}
	0: \|██████▎ \| 1026/1621 [1:29:24<47:51, 4.83s/it] 63%\|██████▎ \| 1027/1621 [1:29:29<47:20, 4.78s/it] 63%\|██████▎ \| 1028/1621 [1:29:34<49:21, 4.99s/it] 63%\|██████▎ \| 1029/1621 [1:29:39<48:22, 4.90s/it] 64%\|██████▎ \| 1030/1621 [1:29:43<47:31, 4.83s/it] 64%\|██████▎ \| 1030/1621 [1:29:43<47:31, 4.83s/it] 64%\|██████▎ \| 1031/1621 [1:29:48<47:03, 4.79s/it] 64%\|██████▎ \| 1032/1621 [1:29:53<46:47, 4.77s/it] 64%\|██████▎ \| 1033/1621 [1:29:58<46:46, 4.77s/it] 64%\|██████▍ \| 1034/1621 [1:30:02<46:28, 4.75s/it] 64%\|██████▍ \| 1035/1621 [1:30:07<46:21, 4.75s/it] 64%\|██████▍ \| 1036/1621 [1:30:12<45:56, 4.71s/it] 64%\|██████▍ \| 1037/1621 [1:30:17<46:25, 4.77s/it] 64%\|██████▍ \| 1038/1621 [1:30:21<45:56, 4.73s/it] 64%\|█████�
	0: {'loss': 0.2114, 'grad_norm': 0.29673487038816576, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.64}
	0: {'loss': 0.2084, 'grad_norm': 0.26864545919062477, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.65}
	0: ��▍ \| 1039/1621 [1:30:26<47:07, 4.86s/it] 64%\|██████▍ \| 1040/1621 [1:30:31<46:35, 4.81s/it] 64%\|██████▍ \| 1040/1621 [1:30:31<46:35, 4.81s/it] 64%\|██████▍ \| 1041/1621 [1:30:36<46:04, 4.77s/it] 64%\|██████▍ \| 1042/1621 [1:30:40<45:42, 4.74s/it] 64%\|██████▍ \| 1043/1621 [1:30:45<46:04, 4.78s/it] 64%\|██████▍ \| 1044/1621 [1:30:50<45:43, 4.76s/it] 64%\|██████▍ \| 1045/1621 [1:30:55<47:03, 4.90s/it] 65%\|██████▍ \| 1046/1621 [1:31:01<48:18, 5.04s/it] 65%\|██████▍ \| 1047/1621 [1:31:06<47:46, 4.99s/it] 65%\|██████▍ \| 1048/1621 [1:31:11<48:02, 5.03s/it] 65%\|██████▍ \| 1049/1621 [1:31:17<50:26, 5.29s/it] 65%\|██████▍ \| 1050/1621 [1:31:21<48:52, 5.14s/it] 65%\|██████▍ \| 1050/1621 [1:31:21<4
	0: {'loss': 0.2142, 'grad_norm': 0.26666640684697085, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.65}
	0: 8:52, 5.14s/it] 65%\|██████▍ \| 1051/1621 [1:31:26<48:22, 5.09s/it] 65%\|██████▍ \| 1052/1621 [1:31:31<47:16, 4.99s/it] 65%\|██████▍ \| 1053/1621 [1:31:36<47:14, 4.99s/it] 65%\|██████▌ \| 1054/1621 [1:31:42<49:17, 5.22s/it] 65%\|██████▌ \| 1055/1621 [1:31:47<47:49, 5.07s/it] 65%\|██████▌ \| 1056/1621 [1:31:52<47:48, 5.08s/it] 65%\|██████▌ \| 1057/1621 [1:31:56<46:44, 4.97s/it] 65%\|██████▌ \| 1058/1621 [1:32:01<46:19, 4.94s/it] 65%\|██████▌ \| 1059/1621 [1:32:06<45:52, 4.90s/it] 65%\|██████▌ \| 1060/1621 [1:32:11<46:39, 4.99s/it] 65%\|██████▌ \| 1060/1621 [1:32:11<46:39, 4.99s/it] 65%\|██████▌ \| 1061/1621 [1:32:16<46:03, 4.93s/it] 66%\|██████▌ \| 1062/1621 [1:32:21<45:18, 4.86s/it] 66%\|██████▌ \| 1063/1621 [1:32:26<45:11, 4.86s/it]
	0: {'loss': 0.21, 'grad_norm': 0.28093123469034137, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.66}
	0: 66%\|██████▌ \| 1064/1621 [1:32:30<44:38, 4.81s/it] 66%\|██████▌ \| 1065/1621 [1:32:35<44:12, 4.77s/it] 66%\|██████▌ \| 1066/1621 [1:32:40<43:44, 4.73s/it] 66%\|██████▌ \| 1067/1621 [1:32:44<43:47, 4.74s/it] 66%\|██████▌ \| 1068/1621 [1:32:49<43:29, 4.72s/it] 66%\|██████▌ \| 1069/1621 [1:32:54<43:22, 4.72s/it] 66%\|██████▌ \| 1070/1621 [1:32:58<43:11, 4.70s/it] 66%\|██████▌ \| 1070/1621 [1:32:58<43:11, 4.70s/it] 66%\|██████▌ \| 1071/1621 [1:33:03<43:19, 4.73s/it] 66%\|██████▌ \| 1072/1621 [1:33:08<43:13, 4.72s/it] 66%\|██████▌ \| 1073/1621 [1:33:13<42:55, 4.70s/it] 66%\|██████▋ \| 1074/1621 [1:33:17<42:34, 4.67s/it] 66%\|██████▋ \| 1075/1621 [1:33:22<42:59, 4.73s/it] 66%\|██████▋ \| 1076/1621 [1:33:28<45:19, 4.99s/it] 66%\|████
	0: {'loss': 0.2119, 'grad_norm': 0.2831457906289921, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.67}
	0: ██▋ \| 1077/1621 [1:33:32<44:32, 4.91s/it] 67%\|██████▋ \| 1078/1621 [1:33:37<44:12, 4.88s/it] 67%\|██████▋ \| 1079/1621 [1:33:42<44:56, 4.98s/it] 67%\|██████▋ \| 1080/1621 [1:33:47<44:58, 4.99s/it] 67%\|██████▋ \| 1080/1621 [1:33:47<44:58, 4.99s/it] 67%\|██████▋ \| 1081/1621 [1:33:52<44:23, 4.93s/it] 67%\|██████▋ \| 1082/1621 [1:33:57<43:46, 4.87s/it] 67%\|██████▋ \| 1083/1621 [1:34:02<43:24, 4.84s/it] 67%\|██████▋ \| 1084/1621 [1:34:08<46:58, 5.25s/it] 67%\|██████▋ \| 1085/1621 [1:34:13<45:23, 5.08s/it] 67%\|██████▋ \| 1086/1621 [1:34:18<45:36, 5.11s/it] 67%\|██████▋ \| 1087/1621 [1:34:22<44:22, 4.99s/it] 67%\|██████▋ \| 1088/1621 [1:34:27<43:37, 4.91s/it] 67%\|██████▋ \| 1089/1621 [1:34:32<43:01, 4.85s/it] 67%\|██████▋ \| 109
	0: {'loss': 0.2105, 'grad_norm': 0.2612962366533585, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.67}
	0: {'loss': 0.208, 'grad_norm': 0.26706625813917995, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.68}
	0: 0/1621 [1:34:37<42:36, 4.82s/it] 67%\|██████▋ \| 1090/1621 [1:34:37<42:36, 4.82s/it] 67%\|██████▋ \| 1091/1621 [1:34:42<43:38, 4.94s/it] 67%\|██████▋ \| 1092/1621 [1:34:47<45:22, 5.15s/it] 67%\|██████▋ \| 1093/1621 [1:34:52<43:58, 5.00s/it] 67%\|██████▋ \| 1094/1621 [1:34:57<43:30, 4.95s/it] 68%\|██████▊ \| 1095/1621 [1:35:02<43:37, 4.98s/it] 68%\|██████▊ \| 1096/1621 [1:35:08<45:54, 5.25s/it] 68%\|██████▊ \| 1097/1621 [1:35:14<47:16, 5.41s/it] 68%\|██████▊ \| 1098/1621 [1:35:19<46:21, 5.32s/it] 68%\|██████▊ \| 1099/1621 [1:35:24<45:09, 5.19s/it] 68%\|██████▊ \| 1100/1621 [1:35:28<44:10, 5.09s/it] 68%\|██████▊ \| 1100/1621 [1:35:29<44:10, 5.09s/it] 68%\|██████▊ \| 1101/1621 [1:35:35<46:56, 5.42s/
	0: {'loss': 0.2117, 'grad_norm': 0.3142948099091601, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.68}
	0: it] 68%\|██████▊ \| 1102/1621 [1:35:40<46:07, 5.33s/it] 68%\|██████▊ \| 1103/1621 [1:35:45<45:36, 5.28s/it] 68%\|██████▊ \| 1104/1621 [1:35:50<44:30, 5.17s/it] 68%\|██████▊ \| 1105/1621 [1:35:56<46:28, 5.40s/it] 68%\|██████▊ \| 1106/1621 [1:36:01<44:54, 5.23s/it] 68%\|██████▊ \| 1107/1621 [1:36:05<43:40, 5.10s/it] 68%\|██████▊ \| 1108/1621 [1:36:10<42:33, 4.98s/it] 68%\|██████▊ \| 1109/1621 [1:36:15<42:05, 4.93s/it] 68%\|██████▊ \| 1110/1621 [1:36:20<41:27, 4.87s/it] 68%\|██████▊ \| 1110/1621 [1:36:20<41:27, 4.87s/it] 69%\|██████▊ \| 1111/1621 [1:36:25<41:40, 4.90s/it] 69%\|██████▊ \| 1112/1621 [1:36:30<41:22, 4.88s/it] 69%\|██████▊ \| 1113/1621 [1:36:34<41:05, 4.85s/it] 69%\|██████▊ \| 1114/1621 [1:36:39<40:58, 4.85s/it] 69%\|██�
	0: {'loss': 0.2134, 'grad_norm': 0.2646577996823958, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.69}
	0: �███▉ \| 1115/1621 [1:36:44<40:26, 4.80s/it] 69%\|██████▉ \| 1116/1621 [1:36:49<40:10, 4.77s/it] 69%\|██████▉ \| 1117/1621 [1:36:54<41:00, 4.88s/it] 69%\|██████▉ \| 1118/1621 [1:36:58<40:21, 4.81s/it] 69%\|██████▉ \| 1119/1621 [1:37:03<40:17, 4.82s/it] 69%\|██████▉ \| 1120/1621 [1:37:08<39:50, 4.77s/it] 69%\|██████▉ \| 1120/1621 [1:37:08<39:50, 4.77s/it] 69%\|██████▉ \| 1121/1621 [1:37:13<40:22, 4.84s/it] 69%\|██████▉ \| 1122/1621 [1:37:17<39:51, 4.79s/it] 69%\|██████▉ \| 1123/1621 [1:37:23<42:23, 5.11s/it] 69%\|██████▉ \| 1124/1621 [1:37:28<41:13, 4.98s/it] 69%\|██████▉ \| 1125/1621 [1:37:33<40:32, 4.90s/it] 69%\|██████▉ \| 1126/1621 [1:37:37<39:48, 4.83s/it] 70%\|██████▉ \| 1127/1621 [1:37:42<39:21, 4.78s/it] 70%\|██████▉ \|
	0: {'loss': 0.2106, 'grad_norm': 0.27369103318579074, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.7}
	0: 1128/1621 [1:37:47<39:11, 4.77s/it] 70%\|██████▉ \| 1129/1621 [1:37:52<40:11, 4.90s/it] 70%\|██████▉ \| 1130/1621 [1:37:57<40:38, 4.97s/it] 70%\|██████▉ \| 1130/1621 [1:37:57<40:38, 4.97s/it] 70%\|██████▉ \| 1131/1621 [1:38:02<40:24, 4.95s/it] 70%\|██████▉ \| 1132/1621 [1:38:07<39:44, 4.88s/it] 70%\|██████▉ \| 1133/1621 [1:38:11<39:20, 4.84s/it] 70%\|██████▉ \| 1134/1621 [1:38:16<39:09, 4.82s/it] 70%\|███████ \| 1135/1621 [1:38:21<39:11, 4.84s/it] 70%\|███████ \| 1136/1621 [1:38:26<38:55, 4.82s/it] 70%\|███████ \| 1137/1621 [1:38:31<39:47, 4.93s/it] 70%\|███████ \| 1138/1621 [1:38:36<39:28, 4.90s/it] 70%\|███████ \| 1139/1621 [1:38:41<39:34, 4.93s/it] 70%\|███████ \| 1140/1621 [1:38:46<39:08, 4.88s/it]
	0: {'loss': 0.2075, 'grad_norm': 0.26705269442253254, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.7}
	0: {'loss': 0.2098, 'grad_norm': 0.279247722684754, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.71}
	0: 70%\|███████ \| 1140/1621 [1:38:46<39:08, 4.88s/it] 70%\|███████ \| 1141/1621 [1:38:50<38:32, 4.82s/it] 70%\|███████ \| 1142/1621 [1:38:55<38:05, 4.77s/it] 71%\|███████ \| 1143/1621 [1:39:01<40:43, 5.11s/it] 71%\|███████ \| 1144/1621 [1:39:06<39:28, 4.97s/it] 71%\|███████ \| 1145/1621 [1:39:10<39:01, 4.92s/it] 71%\|███████ \| 1146/1621 [1:39:16<40:20, 5.10s/it] 71%\|███████ \| 1147/1621 [1:39:21<39:31, 5.00s/it] 71%\|███████ \| 1148/1621 [1:39:26<41:01, 5.20s/it] 71%\|███████ \| 1149/1621 [1:39:31<39:55, 5.08s/it] 71%\|███████ \| 1150/1621 [1:39:36<39:55, 5.09s/it] 71%\|███████ \| 1150/1621 [1:39:36<39:55, 5.09s/it] 71%\|███████ \| 1151/1621 [1:39:41<39:46, 5.08s/it] 71%\|███████ \| 1152/1621 [1:39:47<41:37, 5.33s/it] 71%\|█�
	0: {'loss': 0.2114, 'grad_norm': 0.28121457876267647, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.72}
	0: ��█████ \| 1153/1621 [1:39:52<40:02, 5.13s/it] 71%\|███████ \| 1154/1621 [1:39:57<39:00, 5.01s/it] 71%\|███████▏ \| 1155/1621 [1:40:02<39:03, 5.03s/it] 71%\|███████▏ \| 1156/1621 [1:40:06<38:07, 4.92s/it] 71%\|███████▏ \| 1157/1621 [1:40:11<37:21, 4.83s/it] 71%\|███████▏ \| 1158/1621 [1:40:16<37:14, 4.83s/it] 71%\|███████▏ \| 1159/1621 [1:40:21<36:59, 4.80s/it] 72%\|███████▏ \| 1160/1621 [1:40:25<36:32, 4.76s/it] 72%\|███████▏ \| 1160/1621 [1:40:25<36:32, 4.76s/it] 72%\|███████▏ \| 1161/1621 [1:40:30<36:18, 4.74s/it] 72%\|███████▏ \| 1162/1621 [1:40:35<36:04, 4.72s/it] 72%\|███████▏ \| 1163/1621 [1:40:39<36:23, 4.77s/it] 72%\|███████▏ \| 1164/1621 [1:40:44<36:24, 4.78s/it] 72%\|███████▏ \| 1165/1621 [1:40:49<36:07, 4.75s/it] 7
	0: {'loss': 0.2152, 'grad_norm': 0.26571684410130864, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.72}
	0: 2%\|███████▏ \| 1166/1621 [1:40:54<36:14, 4.78s/it] 72%\|███████▏ \| 1167/1621 [1:40:58<35:58, 4.76s/it] 72%\|███████▏ \| 1168/1621 [1:41:03<35:47, 4.74s/it] 72%\|███████▏ \| 1169/1621 [1:41:08<35:53, 4.76s/it] 72%\|███████▏ \| 1170/1621 [1:41:13<36:18, 4.83s/it] 72%\|███████▏ \| 1170/1621 [1:41:13<36:18, 4.83s/it] 72%\|███████▏ \| 1171/1621 [1:41:18<36:05, 4.81s/it] 72%\|███████▏ \| 1172/1621 [1:41:22<35:47, 4.78s/it] 72%\|███████▏ \| 1173/1621 [1:41:27<35:31, 4.76s/it] 72%\|███████▏ \| 1174/1621 [1:41:32<35:31, 4.77s/it] 72%\|███████▏ \| 1175/1621 [1:41:37<35:11, 4.74s/it] 73%\|███████▎ \| 1176/1621 [1:41:41<35:15, 4.75s/it] 73%\|███████▎ \| 1177/1621 [1:41:46<35:29, 4.80s/it] 73%\|███████▎ \| 1178/1621 [1:41:51<35:33, 4
	0: {'loss': 0.2104, 'grad_norm': 0.28659441188197854, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.73}
	0: .82s/it] 73%\|███████▎ \| 1179/1621 [1:41:56<35:50, 4.86s/it] 73%\|███████▎ \| 1180/1621 [1:42:01<35:15, 4.80s/it] 73%\|███████▎ \| 1180/1621 [1:42:01<35:15, 4.80s/it] 73%\|███████▎ \| 1181/1621 [1:42:05<34:57, 4.77s/it] 73%\|███████▎ \| 1182/1621 [1:42:10<34:42, 4.74s/it] 73%\|███████▎ \| 1183/1621 [1:42:15<35:18, 4.84s/it] 73%\|███████▎ \| 1184/1621 [1:42:20<34:54, 4.79s/it] 73%\|███████▎ \| 1185/1621 [1:42:25<34:41, 4.77s/it] 73%\|███████▎ \| 1186/1621 [1:42:29<34:24, 4.75s/it] 73%\|███████▎ \| 1187/1621 [1:42:34<34:09, 4.72s/it] 73%\|███████▎ \| 1188/1621 [1:42:39<34:13, 4.74s/it] 73%\|███████▎ \| 1189/1621 [1:42:43<33:57, 4.72s/it] 73%\|███████▎ \| 1190/1621 [1:42:49<34:57, 4.87s/it]
	0: {'loss': 0.2125, 'grad_norm': 0.2754546966001141, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.73}
	0: {'loss': 0.2078, 'grad_norm': 0.2850579163343314, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.74}
	0: 73%\|███████▎ \| 1190/1621 [1:42:49<34:57, 4.87s/it] 73%\|███████▎ \| 1191/1621 [1:42:53<34:31, 4.82s/it] 74%\|███████▎ \| 1192/1621 [1:42:58<34:10, 4.78s/it] 74%\|███████▎ \| 1193/1621 [1:43:03<34:31, 4.84s/it] 74%\|███████▎ \| 1194/1621 [1:43:08<34:36, 4.86s/it] 74%\|███████▎ \| 1195/1621 [1:43:13<34:13, 4.82s/it] 74%\|███████▍ \| 1196/1621 [1:43:18<34:20, 4.85s/it] 74%\|███████▍ \| 1197/1621 [1:43:22<34:00, 4.81s/it] 74%\|███████▍ \| 1198/1621 [1:43:27<33:50, 4.80s/it] 74%\|███████▍ \| 1199/1621 [1:43:32<35:02, 4.98s/it] 74%\|███████▍ \| 1200/1621 [1:43:37<34:23, 4.90s/it] 74%\|███████▍ \| 1200/1621 [1:43:37<34:23, 4.90s/it] 74%\|███████▍ \| 1201/1621 [1:43:42<33:51, 4.84s/it] 74%\|███████▍ \| 1202/1621 [1:43:47<33
	0: {'loss': 0.2153, 'grad_norm': 0.27762596072148293, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.75}
	0: :37, 4.81s/it] 74%\|███████▍ \| 1203/1621 [1:43:52<34:02, 4.89s/it] 74%\|███████▍ \| 1204/1621 [1:43:56<33:30, 4.82s/it] 74%\|███████▍ \| 1205/1621 [1:44:01<33:06, 4.77s/it] 74%\|███████▍ \| 1206/1621 [1:44:06<33:00, 4.77s/it] 74%\|███████▍ \| 1207/1621 [1:44:11<32:51, 4.76s/it] 75%\|███████▍ \| 1208/1621 [1:44:16<33:56, 4.93s/it] 75%\|███████▍ \| 1209/1621 [1:44:21<34:11, 4.98s/it] 75%\|███████▍ \| 1210/1621 [1:44:26<33:52, 4.95s/it] 75%\|███████▍ \| 1210/1621 [1:44:26<33:52, 4.95s/it] 75%\|███████▍ \| 1211/1621 [1:44:31<33:20, 4.88s/it] 75%\|███████▍ \| 1212/1621 [1:44:35<32:47, 4.81s/it] 75%\|███████▍ \| 1213/1621 [1:44:40<33:17, 4.90s/it] 75%\|███████▍ \| 1214/1621 [1:44:45<32:45, 4.83s/it] 75%\|███████▍ \| 1215/1621
	0: {'loss': 0.2115, 'grad_norm': 0.2740032175831026, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.75}
	0: [1:44:50<33:53, 5.01s/it] 75%\|███████▌ \| 1216/1621 [1:44:55<33:02, 4.89s/it] 75%\|███████▌ \| 1217/1621 [1:45:00<33:07, 4.92s/it] 75%\|███████▌ \| 1218/1621 [1:45:05<33:24, 4.97s/it] 75%\|███████▌ \| 1219/1621 [1:45:11<34:16, 5.12s/it] 75%\|███████▌ \| 1220/1621 [1:45:16<34:07, 5.11s/it] 75%\|███████▌ \| 1220/1621 [1:45:16<34:07, 5.11s/it] 75%\|███████▌ \| 1221/1621 [1:45:20<33:09, 4.97s/it] 75%\|███████▌ \| 1222/1621 [1:45:25<32:29, 4.89s/it] 75%\|███████▌ \| 1223/1621 [1:45:30<32:08, 4.85s/it] 76%\|███████▌ \| 1224/1621 [1:45:35<33:32, 5.07s/it] 76%\|███████▌ \| 1225/1621 [1:45:40<32:42, 4.96s/it] 76%\|███████▌ \| 1226/1621 [1:45:45<32:31, 4.94s/it] 76%\|███████▌ \| 1227/1621 [1:45:50<32:17, 4.92s/it] 76%\|███████▌ \|
	0: {'loss': 0.2137, 'grad_norm': 1.012604656340779, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.76}
	0: 1228/1621 [1:45:55<32:08, 4.91s/it] 76%\|███████▌ \| 1229/1621 [1:46:00<31:54, 4.88s/it] 76%\|███████▌ \| 1230/1621 [1:46:04<31:22, 4.82s/it] 76%\|███████▌ \| 1230/1621 [1:46:04<31:22, 4.82s/it] 76%\|███████▌ \| 1231/1621 [1:46:09<31:15, 4.81s/it] 76%\|███████▌ \| 1232/1621 [1:46:14<31:40, 4.89s/it] 76%\|███████▌ \| 1233/1621 [1:46:19<31:14, 4.83s/it] 76%\|███████▌ \| 1234/1621 [1:46:23<30:58, 4.80s/it] 76%\|███████▌ \| 1235/1621 [1:46:28<30:59, 4.82s/it] 76%\|███████▌ \| 1236/1621 [1:46:33<30:37, 4.77s/it] 76%\|███████▋ \| 1237/1621 [1:46:38<30:20, 4.74s/it] 76%\|███████▋ \| 1238/1621 [1:46:42<30:15, 4.74s/it] 76%\|███████▋ \| 1239/1621 [1:46:47<30:03, 4.72s/it] 76%\|███████▋ \| 1240/1621 [1:46:52<29:48, 4.69s/it]
	0: {'loss': 0.2143, 'grad_norm': 0.2846388443797818, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.76}
	0: {'loss': 0.2055, 'grad_norm': 0.26029718751210856, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.77}
	0: 76%\|███████▋ \| 1240/1621 [1:46:52<29:48, 4.69s/it] 77%\|███████▋ \| 1241/1621 [1:46:57<31:09, 4.92s/it] 77%\|███████▋ \| 1242/1621 [1:47:02<31:27, 4.98s/it] 77%\|███████▋ \| 1243/1621 [1:47:07<31:00, 4.92s/it] 77%\|███████▋ \| 1244/1621 [1:47:12<31:05, 4.95s/it] 77%\|███████▋ \| 1245/1621 [1:47:17<31:52, 5.09s/it] 77%\|███████▋ \| 1246/1621 [1:47:22<31:09, 4.98s/it] 77%\|███████▋ \| 1247/1621 [1:47:27<30:43, 4.93s/it] 77%\|███████▋ \| 1248/1621 [1:47:32<30:08, 4.85s/it] 77%\|███████▋ \| 1249/1621 [1:47:36<30:01, 4.84s/it] 77%\|███████▋ \| 1250/1621 [1:47:41<29:49, 4.82s/it] 77%\|███████▋ \| 1250/1621 [1:47:41<29:49, 4.82s/it] 77%\|███████▋ \| 1251/1621 [1:47:46<29:38, 4.81s/it] 77%\|██████�
	0: {'loss': 0.2112, 'grad_norm': 0.35354670964811946, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.78}
	0: �▋ \| 1252/1621 [1:47:51<29:20, 4.77s/it] 77%\|███████▋ \| 1253/1621 [1:47:56<29:51, 4.87s/it] 77%\|███████▋ \| 1254/1621 [1:48:01<29:53, 4.89s/it] 77%\|███████▋ \| 1255/1621 [1:48:05<29:31, 4.84s/it] 77%\|███████▋ \| 1256/1621 [1:48:10<29:11, 4.80s/it] 78%\|███████▊ \| 1257/1621 [1:48:15<29:12, 4.81s/it] 78%\|███████▊ \| 1258/1621 [1:48:20<28:59, 4.79s/it] 78%\|███████▊ \| 1259/1621 [1:48:25<29:38, 4.91s/it] 78%\|███████▊ \| 1260/1621 [1:48:30<29:55, 4.97s/it] 78%\|███████▊ \| 1260/1621 [1:48:30<29:55, 4.97s/it] 78%\|███████▊ \| 1261/1621 [1:48:35<29:29, 4.91s/it] 78%\|███████▊ \| 1262/1621 [1:48:40<29:48, 4.98s/it] 78%\|███████▊ \| 1263/1621 [1:48:45<29:46, 4.99s/it] 78%\|███████▊ \| 1264/1621 [1:48:50<29:17, 4.92s/it] 78%\|███
	0: {'loss': 0.2116, 'grad_norm': 0.26423705530912817, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.78}
	0: ████▊ \| 1265/1621 [1:48:55<29:36, 4.99s/it] 78%\|███████▊ \| 1266/1621 [1:49:00<29:00, 4.90s/it] 78%\|███████▊ \| 1267/1621 [1:49:04<28:34, 4.84s/it] 78%\|███████▊ \| 1268/1621 [1:49:09<28:44, 4.88s/it] 78%\|███████▊ \| 1269/1621 [1:49:14<28:21, 4.83s/it] 78%\|███████▊ \| 1270/1621 [1:49:19<28:04, 4.80s/it] 78%\|███████▊ \| 1270/1621 [1:49:19<28:04, 4.80s/it] 78%\|███████▊ \| 1271/1621 [1:49:23<27:52, 4.78s/it] 78%\|███████▊ \| 1272/1621 [1:49:28<27:38, 4.75s/it] 79%\|███████▊ \| 1273/1621 [1:49:33<28:08, 4.85s/it] 79%\|███████▊ \| 1274/1621 [1:49:38<27:44, 4.80s/it] 79%\|███████▊ \| 1275/1621 [1:49:43<27:29, 4.77s/it] 79%\|███████▊ \| 1276/1621 [1:49:47<27:29, 4.78s/it] 79%\|███████▉ \| 1277/1621 [1:49:52<27:15, 4.75s/it] 79
	0: {'loss': 0.211, 'grad_norm': 0.2708639256375303, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.79}
	0: %\|███████▉ \| 1278/1621 [1:49:57<26:58, 4.72s/it] 79%\|███████▉ \| 1279/1621 [1:50:02<27:08, 4.76s/it] 79%\|███████▉ \| 1280/1621 [1:50:07<27:17, 4.80s/it] 79%\|███████▉ \| 1280/1621 [1:50:07<27:17, 4.80s/it] 79%\|███████▉ \| 1281/1621 [1:50:11<26:55, 4.75s/it] 79%\|███████▉ \| 1282/1621 [1:50:16<26:42, 4.73s/it] 79%\|███████▉ \| 1283/1621 [1:50:21<26:33, 4.71s/it] 79%\|███████▉ \| 1284/1621 [1:50:25<26:24, 4.70s/it] 79%\|███████▉ \| 1285/1621 [1:50:30<26:37, 4.75s/it] 79%\|███████▉ \| 1286/1621 [1:50:35<26:23, 4.73s/it] 79%\|███████▉ \| 1287/1621 [1:50:39<26:13, 4.71s/it] 79%\|███████▉ \| 1288/1621 [1:50:44<26:11, 4.72s/it] 80%\|███████▉ \| 1289/1621 [1:50:49<26:39, 4.82s/it] 80%\|███████▉ \| 1290/1621 [1:50:54<27:01, 4.
	0: {'loss': 0.2084, 'grad_norm': 0.25641950625482923, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.8}
	0: {'loss': 0.2098, 'grad_norm': 0.2683510657084983, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.8}
	0: [2025-09-02 20:56:07,755] [INFO] [axolotl.core.trainers.base._save:613] [PID:2134977] [RANK:0] Saving model checkpoint to /lustre/fswork/projects/rech/dgo/udv55np/math/Qwen3-235B-A22B/Qwen2.5-14B_ift/0/checkpoint-1300[39m
	0: [2025-09-02 20:56:31,010] [INFO] [axolotl.core.trainers.base._save:662] [PID:2134977] [RANK:0] Saving Trainer.data_collator.tokenizer by default as Trainer.processing_class is `None`[39m
	0: 90s/it] 80%\|███████▉ \| 1290/1621 [1:50:54<27:01, 4.90s/it] 80%\|███████▉ \| 1291/1621 [1:50:59<26:39, 4.85s/it] 80%\|███████▉ \| 1292/1621 [1:51:04<26:21, 4.81s/it] 80%\|███████▉ \| 1293/1621 [1:51:09<26:25, 4.83s/it] 80%\|███████▉ \| 1294/1621 [1:51:13<26:10, 4.80s/it] 80%\|███████▉ \| 1295/1621 [1:51:18<26:07, 4.81s/it] 80%\|███████▉ \| 1296/1621 [1:51:24<28:07, 5.19s/it] 80%\|████████ \| 1297/1621 [1:51:30<28:45, 5.33s/it] 80%\|████████ \| 1298/1621 [1:51:35<28:27, 5.29s/it] 80%\|████████ \| 1299/1621 [1:51:40<27:22, 5.10s/it] 80%\|████████ \| 1300/1621 [1:51:45<27:20, 5.11s/it] 80%\|████████ \| 1300/1621 [1:51:45<27:20, 5.11s/it] 80%\|████████ \| 1301/1621 [1:52:32<1:33:48, 17.59
	0: {'loss': 0.2155, 'grad_norm': 0.2922501736371335, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.81}
	0: s/it] 80%\|████████ \| 1302/1621 [1:52:36<1:12:54, 13.71s/it] 80%\|████████ \| 1303/1621 [1:52:41<58:41, 11.07s/it] 80%\|████████ \| 1304/1621 [1:52:46<48:27, 9.17s/it] 81%\|████████ \| 1305/1621 [1:52:51<41:11, 7.82s/it] 81%\|████████ \| 1306/1621 [1:52:55<36:04, 6.87s/it] 81%\|████████ \| 1307/1621 [1:53:00<32:32, 6.22s/it] 81%\|████████ \| 1308/1621 [1:53:05<30:07, 5.77s/it] 81%\|████████ \| 1309/1621 [1:53:09<28:31, 5.49s/it] 81%\|████████ \| 1310/1621 [1:53:14<27:08, 5.24s/it] 81%\|████████ \| 1310/1621 [1:53:14<27:08, 5.24s/it] 81%\|████████ \| 1311/1621 [1:53:19<26:20, 5.10s/it] 81%\|████████ \| 1312/1621 [1:53:24<25:37, 4.98s/it] 81%\|████████ \| 1313/1621 [1:53:30<27:05, 5.28s/it] 81%\|████████ \| 1314/1621 [1:53:
	0: {'loss': 0.2112, 'grad_norm': 0.27213095352265687, 'learning_rate': 1e-06, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.81}
	0: 34<26:06, 5.10s/it] 81%\|████████ \| 1315/1621 [1:53:39<25:27, 4.99s/it] 81%\|████████ \| 1316/1621 [1:53:44<25:13, 4.96s/it] 81%\|████████ \| 1317/1621 [1:53:49<25:07, 4.96s/it] 81%\|████████▏ \| 1318/1621 [1:53:54<25:23, 5.03s/it] 81%\|████████▏ \| 1319/1621 [1:53:59<24:46, 4.92s/it] 81%\|████████▏ \| 1320/1621 [1:54:03<24:25, 4.87s/it] 81%\|████████▏ \| 1320/1621 [1:54:03<24:25, 4.87s/it] 81%\|████████▏ \| 1321/1621 [1:54:08<24:02, 4.81s/it] 82%\|████████▏ \| 1322/1621 [1:54:13<23:48, 4.78s/it] 82%\|████████▏ \| 1323/1621 [1:54:18<23:48, 4.79s/it] 82%\|████████▏ \| 1324/1621 [1:54:22<23:33, 4.76s/it] 82%\|████████▏ \| 1325/1621 [1:54:27<23:40, 4.80s/it] 82%\|████████▏ \| 1326/1621 [1:54:32<23:21, 4.75s/it] 82%\|████�
	0: {'loss': 0.2114, 'grad_norm': 0.2722154666863481, 'learning_rate': 9.98421786662277e-07, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.82}
	0: ��███▏ \| 1327/1621 [1:54:37<23:21, 4.77s/it] 82%\|████████▏ \| 1328/1621 [1:54:42<23:31, 4.82s/it] 82%\|████████▏ \| 1329/1621 [1:54:46<23:10, 4.76s/it] 82%\|████████▏ \| 1330/1621 [1:54:52<24:50, 5.12s/it] 82%\|████████▏ \| 1330/1621 [1:54:52<24:50, 5.12s/it] 82%\|████████▏ \| 1331/1621 [1:54:57<24:08, 5.00s/it] 82%\|████████▏ \| 1332/1621 [1:55:02<23:42, 4.92s/it] 82%\|████████▏ \| 1333/1621 [1:55:06<23:21, 4.87s/it] 82%\|████████▏ \| 1334/1621 [1:55:11<23:03, 4.82s/it] 82%\|████████▏ \| 1335/1621 [1:55:16<23:33, 4.94s/it] 82%\|████████▏ \| 1336/1621 [1:55:21<23:06, 4.86s/it] 82%\|████████▏ \| 1337/1621 [1:55:26<22:49, 4.82s/it] 83%\|████████▎ \| 1338/1621 [1:55:31<22:52, 4.85s/it] 83%\|████████▎ \| 1339/1621 [1:55:
	0: {'loss': 0.2082, 'grad_norm': 0.2585298118477787, 'learning_rate': 9.9202926282791e-07, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.83}
	0: {'loss': 0.2067, 'grad_norm': 0.26664310644526223, 'learning_rate': 9.807937738894303e-07, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.83}
	0: 35<22:29, 4.79s/it] 83%\|████████▎ \| 1340/1621 [1:55:40<22:39, 4.84s/it] 83%\|████████▎ \| 1340/1621 [1:55:40<22:39, 4.84s/it] 83%\|████████▎ \| 1341/1621 [1:55:45<23:04, 4.94s/it] 83%\|████████▎ \| 1342/1621 [1:55:51<24:02, 5.17s/it] 83%\|████████▎ \| 1343/1621 [1:55:56<23:16, 5.02s/it] 83%\|████████▎ \| 1344/1621 [1:56:01<22:43, 4.92s/it] 83%\|████████▎ \| 1345/1621 [1:56:05<22:19, 4.85s/it] 83%\|████████▎ \| 1346/1621 [1:56:10<22:10, 4.84s/it] 83%\|████████▎ \| 1347/1621 [1:56:15<21:55, 4.80s/it] 83%\|████████▎ \| 1348/1621 [1:56:19<21:41, 4.77s/it] 83%\|████████▎ \| 1349/1621 [1:56:24<21:29, 4.74s/it] 83%\|████████▎ \| 1350/1621 [1:56:29<21:40, 4.80s/it] 83%\|████████
	0: {'loss': 0.2053, 'grad_norm': 0.28712495588574566, 'learning_rate': 9.648384182148252e-07, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.84}
	0: ▎ \| 1350/1621 [1:56:29<21:40, 4.80s/it] 83%\|████████▎ \| 1351/1621 [1:56:34<21:41, 4.82s/it] 83%\|████████▎ \| 1352/1621 [1:56:39<21:52, 4.88s/it] 83%\|████████▎ \| 1353/1621 [1:56:44<21:33, 4.83s/it] 84%\|████████▎ \| 1354/1621 [1:56:49<22:00, 4.95s/it] 84%\|████████▎ \| 1355/1621 [1:56:54<22:50, 5.15s/it] 84%\|████████▎ \| 1356/1621 [1:56:59<22:11, 5.02s/it] 84%\|████████▎ \| 1357/1621 [1:57:04<22:12, 5.05s/it] 84%\|████████▍ \| 1358/1621 [1:57:09<22:04, 5.04s/it] 84%\|████████▍ \| 1359/1621 [1:57:14<21:32, 4.93s/it] 84%\|████████▍ \| 1360/1621 [1:57:19<21:11, 4.87s/it] 84%\|████████▍ \| 1360/1621 [1:57:19<21:11, 4.87s/it] 84%\|████████▍ \| 1361/1621 [1:57:24<21:02, 4.86s/it] 84%\|████████▍ \| 1362/1621 [1:57:29<21:26,
	0: {'loss': 0.2077, 'grad_norm': 0.26673734209176836, 'learning_rate': 9.443380060197385e-07, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.85}
	0: 4.97s/it] 84%\|████████▍ \| 1363/1621 [1:57:33<21:01, 4.89s/it] 84%\|████████▍ \| 1364/1621 [1:57:39<21:58, 5.13s/it] 84%\|████████▍ \| 1365/1621 [1:57:44<21:52, 5.13s/it] 84%\|████████▍ \| 1366/1621 [1:57:49<21:44, 5.12s/it] 84%\|████████▍ \| 1367/1621 [1:57:55<21:56, 5.18s/it] 84%\|████████▍ \| 1368/1621 [1:58:00<21:36, 5.13s/it] 84%\|████████▍ \| 1369/1621 [1:58:04<21:01, 5.00s/it] 85%\|████████▍ \| 1370/1621 [1:58:09<20:56, 5.00s/it] 85%\|████████▍ \| 1370/1621 [1:58:09<20:56, 5.00s/it] 85%\|████████▍ \| 1371/1621 [1:58:14<20:38, 4.95s/it] 85%\|████████▍ \| 1372/1621 [1:58:19<20:18, 4.89s/it] 85%\|████████▍ \| 1373/1621 [1:58:24<20:09, 4.88s/it] 85%\|████████▍ \| 1374/1621 [1:58:29<19:50, 4.82s/it] 85%\|██████
	0: {'loss': 0.214, 'grad_norm': 0.2619901325108232, 'learning_rate': 9.195171441101668e-07, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.85}
	0: ██▍ \| 1375/1621 [1:58:34<20:14, 4.94s/it] 85%\|████████▍ \| 1376/1621 [1:58:39<20:04, 4.92s/it] 85%\|████████▍ \| 1377/1621 [1:58:43<19:46, 4.86s/it] 85%\|████████▌ \| 1378/1621 [1:58:48<19:39, 4.85s/it] 85%\|████████▌ \| 1379/1621 [1:58:53<19:30, 4.84s/it] 85%\|████████▌ \| 1380/1621 [1:58:58<19:12, 4.78s/it] 85%\|████████▌ \| 1380/1621 [1:58:58<19:12, 4.78s/it] 85%\|████████▌ \| 1381/1621 [1:59:02<19:02, 4.76s/it] 85%\|████████▌ \| 1382/1621 [1:59:07<18:56, 4.76s/it] 85%\|████████▌ \| 1383/1621 [1:59:12<18:55, 4.77s/it] 85%\|████████▌ \| 1384/1621 [1:59:17<18:50, 4.77s/it] 85%\|████████▌ \| 1385/1621 [1:59:22<19:12, 4.88s/it] 86%\|████████▌ \| 1386/1621 [1:59:27<18:54, 4.83s/it] 86%\|████████▌ \| 1387/1621 [1:59:31<18
	0: {'loss': 0.207, 'grad_norm': 0.2769211257765989, 'learning_rate': 8.906477750432903e-07, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.86}
	0: :44, 4.80s/it] 86%\|████████▌ \| 1388/1621 [1:59:36<18:34, 4.78s/it] 86%\|████████▌ \| 1389/1621 [1:59:41<18:19, 4.74s/it] 86%\|████████▌ \| 1390/1621 [1:59:45<18:15, 4.74s/it] 86%\|████████▌ \| 1390/1621 [1:59:45<18:15, 4.74s/it] 86%\|████████▌ \| 1391/1621 [1:59:50<18:06, 4.73s/it] 86%\|████████▌ \| 1392/1621 [1:59:55<17:59, 4.72s/it] 86%\|████████▌ \| 1393/1621 [1:59:59<17:51, 4.70s/it] 86%\|████████▌ \| 1394/1621 [2:00:04<17:51, 4.72s/it] 86%\|████████▌ \| 1395/1621 [2:00:09<17:51, 4.74s/it] 86%\|████████▌ \| 1396/1621 [2:00:14<17:45, 4.74s/it] 86%\|████████▌ \| 1397/1621 [2:00:18<17:37, 4.72s/it] 86%\|████████▌ \| 1398/1621 [2:00:23<17:31, 4.72s/it] 86%\|████████▋ \| 1399/1621 [2:00:28<17:33, 4.74s/it] 86%\|████
	0: {'loss': 0.2157, 'grad_norm': 0.2825136647523626, 'learning_rate': 8.580461976679099e-07, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.86}
	0: {'loss': 0.2117, 'grad_norm': 0.25441419447548025, 'learning_rate': 8.220696016880687e-07, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.87}
	0: ████▋ \| 1400/1621 [2:00:33<17:26, 4.73s/it] 86%\|████████▋ \| 1400/1621 [2:00:33<17:26, 4.73s/it] 86%\|████████▋ \| 1401/1621 [2:00:37<17:22, 4.74s/it] 86%\|████████▋ \| 1402/1621 [2:00:42<17:14, 4.72s/it] 87%\|████████▋ \| 1403/1621 [2:00:47<17:20, 4.78s/it] 87%\|████████▋ \| 1404/1621 [2:00:52<17:09, 4.74s/it] 87%\|████████▋ \| 1405/1621 [2:00:56<17:08, 4.76s/it] 87%\|████████▋ \| 1406/1621 [2:01:01<16:58, 4.74s/it] 87%\|████████▋ \| 1407/1621 [2:01:06<17:10, 4.82s/it] 87%\|████████▋ \| 1408/1621 [2:01:11<17:00, 4.79s/it] 87%\|████████▋ \| 1409/1621 [2:01:16<16:59, 4.81s/it] 87%\|████████▋ \| 1410/1621 [2:01:20<16:51, 4.79s/it] 87%\|████████▋ \| 1410/1621 [2:01:20<16:51, 4.79s
	0: {'loss': 0.2081, 'grad_norm': 0.27992518282615286, 'learning_rate': 7.831121542179086e-07, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.88}
	0: /it] 87%\|████████▋ \| 1411/1621 [2:01:25<16:53, 4.83s/it] 87%\|████████▋ \| 1412/1621 [2:01:31<17:23, 4.99s/it] 87%\|████████▋ \| 1413/1621 [2:01:35<16:56, 4.89s/it] 87%\|████████▋ \| 1414/1621 [2:01:41<17:29, 5.07s/it] 87%\|████████▋ \| 1415/1621 [2:01:47<18:03, 5.26s/it] 87%\|████████▋ \| 1416/1621 [2:01:51<17:26, 5.10s/it] 87%\|████████▋ \| 1417/1621 [2:01:56<17:16, 5.08s/it] 87%\|████████▋ \| 1418/1621 [2:02:01<17:04, 5.05s/it] 88%\|████████▊ \| 1419/1621 [2:02:06<16:39, 4.95s/it] 88%\|████████▊ \| 1420/1621 [2:02:11<16:20, 4.88s/it] 88%\|████████▊ \| 1420/1621 [2:02:11<16:20, 4.88s/it] 88%\|████████▊ \| 1421/1621 [2:02:15<16:05, 4.83s/it] 88%\|████████▊ \| 1422/1621 [2:02:20<16:00, 4.82s/it] 88%\|███████�
	0: {'loss': 0.2102, 'grad_norm': 0.2623614381561981, 'learning_rate': 7.416006812042827e-07, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.88}
	0: �▊ \| 1423/1621 [2:02:25<15:58, 4.84s/it] 88%\|████████▊ \| 1424/1621 [2:02:30<15:42, 4.78s/it] 88%\|████████▊ \| 1425/1621 [2:02:35<15:35, 4.77s/it] 88%\|████████▊ \| 1426/1621 [2:02:40<15:43, 4.84s/it] 88%\|████████▊ \| 1427/1621 [2:02:45<16:42, 5.17s/it] 88%\|████████▊ \| 1428/1621 [2:02:50<16:25, 5.11s/it] 88%\|████████▊ \| 1429/1621 [2:02:56<16:22, 5.12s/it] 88%\|████████▊ \| 1430/1621 [2:03:00<15:56, 5.01s/it] 88%\|████████▊ \| 1430/1621 [2:03:00<15:56, 5.01s/it] 88%\|████████▊ \| 1431/1621 [2:03:05<15:29, 4.89s/it] 88%\|████████▊ \| 1432/1621 [2:03:10<15:34, 4.95s/it] 88%\|████████▊ \| 1433/1621 [2:03:15<15:36, 4.98s/it] 88%\|████████▊ \| 1434/1621 [2:03:20<15:21, 4.93s/it] 89%\|████████▊ \| 1435/1621 [2:03:25<15:03,
	0: {'loss': 0.2116, 'grad_norm': 0.2687063278495978, 'learning_rate': 6.979899910323624e-07, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.89}
	0: 4.86s/it] 89%\|████████▊ \| 1436/1621 [2:03:29<14:54, 4.84s/it] 89%\|████████▊ \| 1437/1621 [2:03:34<15:01, 4.90s/it] 89%\|████████▊ \| 1438/1621 [2:03:39<14:44, 4.83s/it] 89%\|████████▉ \| 1439/1621 [2:03:44<14:35, 4.81s/it] 89%\|████████▉ \| 1440/1621 [2:03:49<14:26, 4.79s/it] 89%\|████████▉ \| 1440/1621 [2:03:49<14:26, 4.79s/it] 89%\|████████▉ \| 1441/1621 [2:03:53<14:18, 4.77s/it] 89%\|████████▉ \| 1442/1621 [2:03:58<14:11, 4.76s/it] 89%\|████████▉ \| 1443/1621 [2:04:03<14:08, 4.77s/it] 89%\|████████▉ \| 1444/1621 [2:04:08<14:03, 4.77s/it] 89%\|████████▉ \| 1445/1621 [2:04:12<13:57, 4.76s/it] 89%\|████████▉ \| 1446/1621 [2:04:17<13:55, 4.77s/it] 89%\|████████▉ \| 1447/1621 [2:04:22<13:51, 4.78s/it] 89%\|█████�
	0: {'loss': 0.2102, 'grad_norm': 0.2714307919796291, 'learning_rate': 6.527578915497951e-07, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.89}
	0: �██▉ \| 1448/1621 [2:04:27<13:44, 4.77s/it] 89%\|████████▉ \| 1449/1621 [2:04:31<13:34, 4.74s/it] 89%\|████████▉ \| 1450/1621 [2:04:36<13:30, 4.74s/it] 89%\|████████▉ \| 1450/1621 [2:04:36<13:30, 4.74s/it] 90%\|████████▉ \| 1451/1621 [2:04:41<13:23, 4.73s/it] 90%\|████████▉ \| 1452/1621 [2:04:46<13:34, 4.82s/it] 90%\|████████▉ \| 1453/1621 [2:04:51<13:24, 4.79s/it] 90%\|████████▉ \| 1454/1621 [2:04:56<13:27, 4.84s/it] 90%\|████████▉ \| 1455/1621 [2:05:01<13:44, 4.97s/it] 90%\|████████▉ \| 1456/1621 [2:05:05<13:25, 4.88s/it] 90%\|████████▉ \| 1457/1621 [2:05:10<13:09, 4.82s/it] 90%\|████████▉ \| 1458/1621 [2:05:15<13:01, 4.80s/it] 90%\|█████████ \| 1459/1621 [2:05:20<13:18, 4.93s/it] 90%\|█████████ \| 1460/1621 [2:05:25<1
	0: {'loss': 0.2083, 'grad_norm': 0.2627481844715906, 'learning_rate': 6.06399955103937e-07, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.9}
	0: {'loss': 0.2042, 'grad_norm': 0.2713225358577235, 'learning_rate': 5.594240889475106e-07, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.91}
	0: 3:04, 4.87s/it] 90%\|█████████ \| 1460/1621 [2:05:25<13:04, 4.87s/it] 90%\|█████████ \| 1461/1621 [2:05:30<12:50, 4.82s/it] 90%\|█████████ \| 1462/1621 [2:05:35<13:01, 4.91s/it] 90%\|█████████ \| 1463/1621 [2:05:40<12:56, 4.91s/it] 90%\|█████████ \| 1464/1621 [2:05:44<12:39, 4.84s/it] 90%\|█████████ \| 1465/1621 [2:05:49<12:25, 4.78s/it] 90%\|█████████ \| 1466/1621 [2:05:54<12:24, 4.80s/it] 90%\|█████████ \| 1467/1621 [2:05:58<12:15, 4.78s/it] 91%\|█████████ \| 1468/1621 [2:06:03<12:06, 4.75s/it] 91%\|█████████ \| 1469/1621 [2:06:08<11:57, 4.72s/it] 91%\|█████████ \| 1470/1621 [2:06:13<11:52, 4.72s/it] 91%\|█████████ \| 1470/1621 [2:06:13<11:52, 4.72s/it] 91%\|█████████
	0: {'loss': 0.2089, 'grad_norm': 0.2537827651441038, 'learning_rate': 5.123449705004581e-07, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.91}
	0: \| 1471/1621 [2:06:17<11:51, 4.74s/it] 91%\|█████████ \| 1472/1621 [2:06:22<12:04, 4.86s/it] 91%\|█████████ \| 1473/1621 [2:06:27<11:52, 4.82s/it] 91%\|█████████ \| 1474/1621 [2:06:33<12:18, 5.02s/it] 91%\|█████████ \| 1475/1621 [2:06:37<11:59, 4.93s/it] 91%\|█████████ \| 1476/1621 [2:06:43<12:07, 5.02s/it] 91%\|█████████ \| 1477/1621 [2:06:47<11:47, 4.92s/it] 91%\|█████████ \| 1478/1621 [2:06:52<11:32, 4.85s/it] 91%\|█████████ \| 1479/1621 [2:06:57<11:49, 5.00s/it] 91%\|█████████▏\| 1480/1621 [2:07:02<11:43, 4.99s/it] 91%\|█████████▏\| 1480/1621 [2:07:02<11:43, 4.99s/it] 91%\|█████████▏\| 1481/1621 [2:07:07<11:23, 4.89s/it] 91%\|█████████▏\| 1482/1621 [2:07:12<11:18, 4.88s/it] 91%\|█████████▏\| 1483/1621 [2:07:17<11
	0: {'loss': 0.2087, 'grad_norm': 0.2823914222814313, 'learning_rate': 4.656784084364238e-07, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.92}
	0: :07, 4.83s/it] 92%\|█████████▏\| 1484/1621 [2:07:22<11:14, 4.92s/it] 92%\|█████████▏\| 1485/1621 [2:07:26<11:01, 4.86s/it] 92%\|█████████▏\| 1486/1621 [2:07:31<10:55, 4.86s/it] 92%\|█████████▏\| 1487/1621 [2:07:36<10:47, 4.83s/it] 92%\|█████████▏\| 1488/1621 [2:07:41<10:54, 4.92s/it] 92%\|█████████▏\| 1489/1621 [2:07:46<10:39, 4.84s/it] 92%\|█████████▏\| 1490/1621 [2:07:51<10:29, 4.81s/it] 92%\|█████████▏\| 1490/1621 [2:07:51<10:29, 4.81s/it] 92%\|█████████▏\| 1491/1621 [2:07:55<10:21, 4.78s/it] 92%\|█████████▏\| 1492/1621 [2:08:00<10:13, 4.75s/it] 92%\|█████████▏\| 1493/1621 [2:08:05<10:08, 4.75s/it] 92%\|█████████▏\| 1494/1621 [2:08:10<10:11, 4.82s/it] 92%\|█████████▏\| 1495/1621 [2:08:15<10:12, 4
	0: {'loss': 0.21, 'grad_norm': 0.2831683932276976, 'learning_rate': 4.1993569137498776e-07, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.93}
	0: .86s/it] 92%\|█████████▏\| 1496/1621 [2:08:20<10:36, 5.09s/it] 92%\|█████████▏\| 1497/1621 [2:08:25<10:17, 4.98s/it] 92%\|█████████▏\| 1498/1621 [2:08:30<10:10, 4.96s/it] 92%\|█████████▏\| 1499/1621 [2:08:35<09:55, 4.88s/it] 93%\|█████████▎\| 1500/1621 [2:08:39<09:43, 4.82s/it] 93%\|█████████▎\| 1500/1621 [2:08:39<09:43, 4.82s/it] 93%\|█████████▎\| 1501/1621 [2:08:44<09:32, 4.77s/it] 93%\|█████████▎\| 1502/1621 [2:08:49<09:24, 4.74s/it] 93%\|█████████▎\| 1503/1621 [2:08:54<09:28, 4.82s/it] 93%\|█████████▎\| 1504/1621 [2:08:58<09:18, 4.77s/it] 93%\|█████████▎\| 1505/1621 [2:09:03<09:21, 4.84s/it] 93%\|█████████▎\| 1506/1621 [2:09:08<09:13, 4.81s/it] 93%\|█████████▎\| 1507/1621 [2:09:13<09:04, 4.77s/it
	0: {'loss': 0.2151, 'grad_norm': 0.2947685205241263, 'learning_rate': 3.7561798609655373e-07, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.93}
	0: ] 93%\|█████████▎\| 1508/1621 [2:09:17<08:59, 4.77s/it] 93%\|█████████▎\| 1509/1621 [2:09:23<09:08, 4.90s/it] 93%\|█████████▎\| 1510/1621 [2:09:28<09:06, 4.92s/it] 93%\|█████████▎\| 1510/1621 [2:09:28<09:06, 4.92s/it] 93%\|█████████▎\| 1511/1621 [2:09:32<08:54, 4.86s/it] 93%\|█████████▎\| 1512/1621 [2:09:38<09:01, 4.97s/it] 93%\|█████████▎\| 1513/1621 [2:09:42<08:47, 4.88s/it] 93%\|█████████▎\| 1514/1621 [2:09:48<09:06, 5.11s/it] 93%\|█████████▎\| 1515/1621 [2:09:53<08:52, 5.03s/it] 94%\|█████████▎\| 1516/1621 [2:09:58<08:49, 5.04s/it] 94%\|█████████▎\| 1517/1621 [2:10:03<09:02, 5.22s/it] 94%\|█████████▎\| 1518/1621 [2:10:08<08:49, 5.14s/it] 94%\|█████████▎\| 1519/1621 [2:10:13<08:32, 5.02s/it] 94%\|
	0: {'loss': 0.2045, 'grad_norm': 0.2678544345260796, 'learning_rate': 3.3321084665422803e-07, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.94}
	0: {'loss': 0.2093, 'grad_norm': 0.27176220768733284, 'learning_rate': 2.931788945420058e-07, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.94}
	0: █████████▍\| 1520/1621 [2:10:18<08:35, 5.10s/it] 94%\|█████████▍\| 1520/1621 [2:10:18<08:35, 5.10s/it] 94%\|█████████▍\| 1521/1621 [2:10:23<08:24, 5.04s/it] 94%\|█████████▍\| 1522/1621 [2:10:28<08:19, 5.04s/it] 94%\|█████████▍\| 1523/1621 [2:10:34<08:42, 5.34s/it] 94%\|█████████▍\| 1524/1621 [2:10:39<08:17, 5.13s/it] 94%\|█████████▍\| 1525/1621 [2:10:44<07:57, 4.98s/it] 94%\|█████████▍\| 1526/1621 [2:10:48<07:43, 4.87s/it] 94%\|█████████▍\| 1527/1621 [2:10:53<07:33, 4.82s/it] 94%\|█████████▍\| 1528/1621 [2:10:58<07:29, 4.83s/it] 94%\|█████████▍\| 1529/1621 [2:11:03<07:20, 4.79s/it] 94%\|█████████▍\| 1530/1621 [2:11:08<07:25, 4.89s/it] 94%\|████████�
	0: {'loss': 0.2121, 'grad_norm': 0.2805922111643405, 'learning_rate': 2.5596072820445254e-07, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.95}
	0: �▍\| 1530/1621 [2:11:08<07:25, 4.89s/it] 94%\|█████████▍\| 1531/1621 [2:11:13<07:29, 4.99s/it] 95%\|█████████▍\| 1532/1621 [2:11:18<07:21, 4.96s/it] 95%\|█████████▍\| 1533/1621 [2:11:22<07:10, 4.89s/it] 95%\|█████████▍\| 1534/1621 [2:11:27<07:03, 4.87s/it] 95%\|█████████▍\| 1535/1621 [2:11:32<06:55, 4.83s/it] 95%\|█████████▍\| 1536/1621 [2:11:37<06:47, 4.79s/it] 95%\|█████████▍\| 1537/1621 [2:11:41<06:39, 4.75s/it] 95%\|█████████▍\| 1538/1621 [2:11:46<06:36, 4.78s/it] 95%\|█████████▍\| 1539/1621 [2:11:51<06:30, 4.77s/it] 95%\|█████████▌\| 1540/1621 [2:11:56<06:35, 4.88s/it] 95%\|█████████▌\| 1540/1621 [2:11:56<06:35, 4.88s/it] 95%\|█████████▌\| 1541/1621 [2:12:01<06:29, 4.87s/it] 95%\|█████████▌\| 1
	0: {'loss': 0.2115, 'grad_norm': 0.4089994863899589, 'learning_rate': 2.2196411766036487e-07, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.96}
	0: 542/1621 [2:12:07<06:44, 5.13s/it] 95%\|█████████▌\| 1543/1621 [2:12:11<06:31, 5.02s/it] 95%\|█████████▌\| 1544/1621 [2:12:16<06:19, 4.92s/it] 95%\|█████████▌\| 1545/1621 [2:12:21<06:09, 4.87s/it] 95%\|█████████▌\| 1546/1621 [2:12:26<06:02, 4.83s/it] 95%\|█████████▌\| 1547/1621 [2:12:30<05:53, 4.78s/it] 95%\|█████████▌\| 1548/1621 [2:12:35<05:47, 4.75s/it] 96%\|█████████▌\| 1549/1621 [2:12:40<05:48, 4.84s/it] 96%\|█████████▌\| 1550/1621 [2:12:46<05:57, 5.03s/it] 96%\|█████████▌\| 1550/1621 [2:12:46<05:57, 5.03s/it] 96%\|█████████▌\| 1551/1621 [2:12:50<05:45, 4.93s/it] 96%\|█████████▌\| 1552/1621 [2:12:55<05:35, 4.86s/it] 96%\|█████████▌\| 1553/1621 [2:13:01<05:55, 5.23s/it] 96%\|█████████▌\| 1554/162
	0: {'loss': 0.2144, 'grad_norm': 0.2651691403897037, 'learning_rate': 1.915615368891117e-07, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.96}
	0: 1 [2:13:06<05:38, 5.05s/it] 96%\|█████████▌\| 1555/1621 [2:13:11<05:33, 5.05s/it] 96%\|█████████▌\| 1556/1621 [2:13:15<05:20, 4.94s/it] 96%\|█████████▌\| 1557/1621 [2:13:20<05:13, 4.89s/it] 96%\|█████████▌\| 1558/1621 [2:13:25<05:08, 4.89s/it] 96%\|█████████▌\| 1559/1621 [2:13:31<05:21, 5.19s/it] 96%\|█████████▌\| 1560/1621 [2:13:36<05:06, 5.03s/it] 96%\|█████████▌\| 1560/1621 [2:13:36<05:06, 5.03s/it] 96%\|█████████▋\| 1561/1621 [2:13:41<05:10, 5.18s/it] 96%\|█████████▋\| 1562/1621 [2:13:46<04:58, 5.05s/it] 96%\|█████████▋\| 1563/1621 [2:13:51<04:46, 4.94s/it] 96%\|█████████▋\| 1564/1621 [2:13:55<04:41, 4.93s/it] 97%\|█████████▋\| 1565/1621 [2:14:00<04:31, 4.86s/it] 97%\|█████████▋\| 1566/1621 [2:14
	0: {'loss': 0.2132, 'grad_norm': 0.2650836750987023, 'learning_rate': 1.6508608292777203e-07, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.97}
	0: :05<04:25, 4.82s/it] 97%\|█████████▋\| 1567/1621 [2:14:10<04:17, 4.76s/it] 97%\|█████████▋\| 1568/1621 [2:14:14<04:11, 4.74s/it] 97%\|█████████▋\| 1569/1621 [2:14:19<04:06, 4.75s/it] 97%\|█████████▋\| 1570/1621 [2:14:24<04:05, 4.82s/it] 97%\|█████████▋\| 1570/1621 [2:14:24<04:05, 4.82s/it] 97%\|█████████▋\| 1571/1621 [2:14:29<03:59, 4.78s/it] 97%\|█████████▋\| 1572/1621 [2:14:33<03:53, 4.76s/it] 97%\|█████████▋\| 1573/1621 [2:14:38<03:48, 4.77s/it] 97%\|█████████▋\| 1574/1621 [2:14:44<03:52, 4.95s/it] 97%\|█████████▋\| 1575/1621 [2:14:48<03:44, 4.87s/it] 97%\|█████████▋\| 1576/1621 [2:14:53<03:36, 4.81s/it] 97%\|█████████▋\| 1577/1621 [2:14:58<03:30, 4.78s/it] 97%\|█████████▋\| 1578/1621 [2:15:02<03:
	0: {'loss': 0.2101, 'grad_norm': 0.2573210674484348, 'learning_rate': 1.4282782639029128e-07, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.97}
	0: 25, 4.78s/it] 97%\|█████████▋\| 1579/1621 [2:15:07<03:18, 4.74s/it] 97%\|█████████▋\| 1580/1621 [2:15:12<03:12, 4.71s/it] 97%\|█████████▋\| 1580/1621 [2:15:12<03:12, 4.71s/it] 98%\|█████████▊\| 1581/1621 [2:15:17<03:11, 4.79s/it] 98%\|█████████▊\| 1582/1621 [2:15:21<03:04, 4.74s/it] 98%\|█████████▊\| 1583/1621 [2:15:26<03:00, 4.75s/it] 98%\|█████████▊\| 1584/1621 [2:15:31<03:00, 4.88s/it] 98%\|█████████▊\| 1585/1621 [2:15:37<03:06, 5.18s/it] 98%\|█████████▊\| 1586/1621 [2:15:42<02:59, 5.13s/it] 98%\|█████████▊\| 1587/1621 [2:15:47<02:51, 5.03s/it] 98%\|█████████▊\| 1588/1621 [2:15:52<02:42, 4.93s/it] 98%\|█████████▊\| 1589/1621 [2:15:56<02:35, 4.86s/it] 98%\|█████████▊\| 1590/1621 [2:16:01<02:30, 4.
	0: {'loss': 0.2086, 'grad_norm': 0.2553344270166335, 'learning_rate': 1.2503063339313356e-07, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.98}
	0: {'loss': 0.206, 'grad_norm': 0.25174882817793, 'learning_rate': 1.1188949370707787e-07, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.99}
	0: 85s/it] 98%\|█████████▊\| 1590/1621 [2:16:01<02:30, 4.85s/it] 98%\|█████████▊\| 1591/1621 [2:16:06<02:24, 4.81s/it] 98%\|█████████▊\| 1592/1621 [2:16:11<02:21, 4.89s/it] 98%\|█████████▊\| 1593/1621 [2:16:16<02:19, 4.97s/it] 98%\|█████████▊\| 1594/1621 [2:16:21<02:13, 4.93s/it] 98%\|█████████▊\| 1595/1621 [2:16:26<02:10, 5.01s/it] 98%\|█████████▊\| 1596/1621 [2:16:31<02:07, 5.11s/it] 99%\|█████████▊\| 1597/1621 [2:16:36<02:00, 5.01s/it] 99%\|█████████▊\| 1598/1621 [2:16:41<01:55, 5.04s/it] 99%\|█████████▊\| 1599/1621 [2:16:46<01:48, 4.94s/it] 99%\|█████████▊\| 1600/1621 [2:16:52<01:48, 5.16s/it] 99%\|█████████▊\| 1600/1621 [2:16:52<01:48, 5.16s/it] 99%\|████�
	0: {'loss': 0.2112, 'grad_norm': 0.2496232881123691, 'learning_rate': 1.0354838440848501e-07, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 0.99}
	0: ��████▉\| 1601/1621 [2:16:56<01:40, 5.01s/it] 99%\|█████████▉\| 1602/1621 [2:17:01<01:33, 4.93s/it] 99%\|█████████▉\| 1603/1621 [2:17:06<01:27, 4.87s/it] 99%\|█████████▉\| 1604/1621 [2:17:11<01:23, 4.88s/it] 99%\|█████████▉\| 1605/1621 [2:17:16<01:18, 4.88s/it] 99%\|█████████▉\| 1606/1621 [2:17:20<01:12, 4.83s/it] 99%\|█████████▉\| 1607/1621 [2:17:25<01:07, 4.85s/it] 99%\|█████████▉\| 1608/1621 [2:17:30<01:04, 4.96s/it] 99%\|█████████▉\| 1609/1621 [2:17:35<00:58, 4.88s/it] 99%\|█████████▉\| 1610/1621 [2:17:40<00:53, 4.83s/it] 99%\|█████████▉\| 1610/1621 [2:17:40<00:53, 4.83s/it] 99%\|█████████▉\| 1611/1621 [2:17:45<00:47, 4.77s/it] 99%\|█████████▉\| 1612/1621 [2:17:49<00:42, 4.76s/it] 100%\|██████�
	0: {'loss': 0.2094, 'grad_norm': 0.2608941458825701, 'learning_rate': 1.0009869243631952e-07, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 1.0}
	0: [2025-09-02 21:23:01,404] [INFO] [axolotl.core.trainers.base._save:613] [PID:2134977] [RANK:0] Saving model checkpoint to /lustre/fswork/projects/rech/dgo/udv55np/math/Qwen3-235B-A22B/Qwen2.5-14B_ift/0/checkpoint-1621[39m
	0: [2025-09-02 21:23:23,920] [INFO] [axolotl.core.trainers.base._save:662] [PID:2134977] [RANK:0] Saving Trainer.data_collator.tokenizer by default as Trainer.processing_class is `None`[39m
	0: {'train_runtime': 8360.3247, 'train_samples_per_second': 3.102, 'train_steps_per_second': 0.194, 'train_loss': 0.21736425671549509, 'memory/max_mem_active(gib)': 53.9, 'memory/max_mem_allocated(gib)': 51.85, 'memory/device_mem_reserved(gib)': 57.41, 'epoch': 1.0}
	0: �██▉\| 1613/1621 [2:17:54<00:38, 4.76s/it] 100%\|█████████▉\| 1614/1621 [2:17:59<00:33, 4.73s/it] 100%\|█████████▉\| 1615/1621 [2:18:03<00:28, 4.73s/it] 100%\|█████████▉\| 1616/1621 [2:18:08<00:23, 4.73s/it] 100%\|█████████▉\| 1617/1621 [2:18:13<00:18, 4.70s/it] 100%\|█████████▉\| 1618/1621 [2:18:18<00:14, 4.98s/it] 100%\|█████████▉\| 1619/1621 [2:18:23<00:09, 4.90s/it] 100%\|█████████▉\| 1620/1621 [2:18:28<00:04, 4.85s/it] 100%\|█████████▉\| 1620/1621 [2:18:28<00:04, 4.85s/it] 100%\|██████████\| 1621/1621 [2:18:39<00:00, 6.66s/it] 100%\|██████████\| 1621/1621 [2:19:20<00:00, 6.66s/it] 100%\|██████████\| 1621/1621 [2:19:20<00:00, 5.16s/it]
	0: [2025-09-02 21:23:35,467] [INFO] [axolotl.train.save_trained_model:228] [PID:2134977] [RANK:0] Training completed! Saving trained model to /lustre/fswork/projects/rech/dgo/udv55np/math/Qwen3-235B-A22B/Qwen2.5-14B_ift/0.[39m
	0: [2025-09-02 21:23:42,461] [INFO] [axolotl.core.trainers.base._save:613] [PID:2134977] [RANK:0] Saving model checkpoint to /lustre/fswork/projects/rech/dgo/udv55np/math/Qwen3-235B-A22B/Qwen2.5-14B_ift/0[39m
	0: [2025-09-02 21:24:04,610] [INFO] [axolotl.core.trainers.base._save:662] [PID:2134977] [RANK:0] Saving Trainer.data_collator.tokenizer by default as Trainer.processing_class is `None`[39m
	0: [2025-09-02 21:24:05,016] [INFO] [axolotl.train.save_trained_model:350] [PID:2134977] [RANK:0] Model successfully saved to /lustre/fswork/projects/rech/dgo/udv55np/math/Qwen3-235B-A22B/Qwen2.5-14B_ift/0[39m