[2026-02-06 23:20:12,989] [WARNING] [axolotl.utils.trainer.prepare_optim_env:658] [PID:140] P2P support not detected, setting `NCCL_P2P_DISABLE=1` [2026-02-06 23:20:12,990] [DEBUG] [axolotl.utils.config.resolve_dtype:66] [PID:140] bf16 support detected, enabling for this configuration. [2026-02-06 23:20:13,184] [DEBUG] [axolotl.utils.config.log_gpu_memory_usage:127] [PID:140] baseline 0.000GB () [2026-02-06 23:20:13,185] [INFO] [axolotl.cli.config.load_cfg:259] [PID:140] config: { "activation_offloading": false, "adapter": "qlora", "axolotl_config_path": "myconfig/Llama-3.3-8B-Instruct-heretic-miki-kimified.yaml", "base_model": "aeon37/Llama-3.3-8B-Instruct-128K-heretic", "base_model_config": "aeon37/Llama-3.3-8B-Instruct-128K-heretic", "batch_size": 16, "bf16": true, "capabilities": { "bf16": true, "compute_capability": "sm_86", "fp8": false, "n_gpu": 2, "n_node": 1 }, "context_parallel_size": 1, "dataloader_num_workers": 2, "dataloader_pin_memory": true, "dataloader_prefetch_factor": 256, "dataset_num_proc": 16, "dataset_prepared_path": "last_run_prepared_kimi", "datasets": [ { "chat_template": "tokenizer_default", "field_messages": "messages", "message_property_mappings": { "content": "content", "role": "role" }, "path": "ramendik/kimify-ifeval-like", "trust_remote_code": false, "type": "chat_template" }, { "chat_template": "tokenizer_default", "field_messages": "messages", "message_property_mappings": { "content": "content", "role": "role" }, "path": "ramendik/kimify-20251115", "trust_remote_code": false, "type": "chat_template" }, { "chat_template": "tokenizer_default", "field_messages": "messages", "message_property_mappings": { "content": "content", "role": "role" }, "path": "ramendik/kimify-short-20260131", "trust_remote_code": false, "type": "chat_template" } ], "ddp": true, "device": "cuda:0", "device_map": { "": 0 }, "dion_rank_fraction": 1.0, "dion_rank_multiple_of": 1, "eaft_alpha": 1.0, "eaft_k": 20, "env_capabilities": { "torch_version": "2.9.1" }, "eval_batch_size": 2, "eval_causal_lm_metrics": [ "sacrebleu", "comet", "ter", "chrf" ], "eval_max_new_tokens": 128, "eval_sample_packing": true, "eval_table_size": 0, "evals_per_epoch": 0, "experimental_skip_move_to_device": true, "flash_attention": true, "fp16": false, "gradient_accumulation_steps": 4, "gradient_checkpointing": true, "gradient_checkpointing_kwargs": { "use_reentrant": false }, "include_tkps": true, "is_falcon_derived_model": false, "is_llama_derived_model": true, "is_mistral_derived_model": false, "learning_rate": 0.0002, "lisa_layers_attribute": "model.layers", "load_best_model_at_end": false, "load_in_4bit": true, "load_in_8bit": false, "local_rank": 0, "logging_steps": 1, "lora_alpha": 8, "lora_dropout": 0.05, "lora_modules_to_save": [ "embed_tokens", "lm_head" ], "lora_r": 16, "lora_target_linear": true, "loraplus_lr_embedding": 1e-06, "lr_scheduler": "cosine", "mean_resizing_embeddings": false, "micro_batch_size": 2, "model_config_type": "llama", "num_epochs": 3.0, "optimizer": "paged_adamw_8bit", "otel_metrics_host": "localhost", "otel_metrics_port": 8000, "output_dir": "./outputs/llama3.3-miki-lora-out", "pad_to_sequence_len": true, "pretrain_multipack_attn": true, "profiler_steps_start": 0, "qlora_sharded_model_loading": false, "ray_num_workers": 1, "resources_per_worker": { "GPU": 1 }, "sample_packing": true, "sample_packing_bin_size": 200, "sample_packing_group_size": 100000, "save_first_step": true, "save_only_model": false, "save_safetensors": true, "save_steps": 0.06666666666666667, "saves_per_epoch": 5, "sequence_len": 2048, "shuffle_before_merging_datasets": false, "shuffle_merged_datasets": true, "skip_prepare_dataset": false, "special_tokens": { "pad_token": "<|end_of_text|>" }, "streaming_multipack_buffer_size": 10000, "strict": false, "tensor_parallel_size": 1, "tf32": false, "tiled_mlp_use_original_mlp": true, "tokenizer_config": "aeon37/Llama-3.3-8B-Instruct-128K-heretic", "tokenizer_save_jinja_files": true, "tokenizer_type": "AutoTokenizer", "torch_dtype": "torch.bfloat16", "train_on_inputs": false, "trl": { "log_completions": false, "mask_truncated_completions": false, "ref_model_mixup_alpha": 0.9, "ref_model_sync_steps": 64, "scale_rewards": true, "sync_ref_model": false, "use_vllm": false, "vllm_server_host": "0.0.0.0", "vllm_server_port": 8000 }, "trust_remote_code": true, "type_of_model": "LlamaForCausalLM", "use_otel_metrics": false, "use_ray": false, "use_wandb": true, "val_set_size": 0.1, "vllm": { "device": "auto", "dtype": "auto", "gpu_memory_utilization": 0.9, "host": "0.0.0.0", "port": 8000 }, "wandb_project": "Llama-3.3-8B-Instruct-128k-heretic-Kimi-miki", "warmup_ratio": 0.1, "weight_decay": 0.0, "world_size": 2 } [2026-02-06 23:20:13,186] [WARNING] [axolotl.cli.checks.check_user_token:46] [PID:140] Error verifying HuggingFace token. Remember to log in using `hf auth login` and get your access token from https://huggingface.co/settings/tokens if you want to use gated models or datasets. [2026-02-06 23:20:14,904] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:285] [PID:140] EOS: 128009 / <|eot_id|> [2026-02-06 23:20:14,905] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:286] [PID:140] BOS: 128000 / <|begin_of_text|> [2026-02-06 23:20:14,905] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:287] [PID:140] PAD: 128001 / <|end_of_text|> [2026-02-06 23:20:14,905] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:288] [PID:140] UNK: None / None [2026-02-06 23:20:14,905] [INFO] [axolotl.utils.data.shared.load_preprocessed_dataset:481] [PID:140] Unable to find prepared dataset in last_run_prepared_kimi/97265f4b2c05c2e1f71d7d50bd837864 [2026-02-06 23:20:14,905] [INFO] [axolotl.utils.data.sft._load_raw_datasets:320] [PID:140] Loading raw datasets... [2026-02-06 23:20:14,906] [WARNING] [axolotl.utils.data.sft._load_raw_datasets:322] [PID:140] Processing datasets during training can lead to VRAM instability. Please pre-process your dataset using `axolotl preprocess path/to/config.yml`. Downloading (incomplete total...): 0.00B [00:00, ?B/s] Fetching 0 files: 0it [00:00, ?it/s] Fetching 0 files: 0it [00:00, ?it/s] Download complete: : 0.00B [00:00, ?B/s] Download complete: : 0.00B [00:00, ?B/s] [2026-02-06 23:20:15,930] [INFO] [axolotl.utils.data.wrappers.get_dataset_wrapper:87] [PID:140] Loading dataset: ramendik/kimify-ifeval-like with base_type: chat_template and prompt_style: None [2026-02-06 23:20:15,945] [INFO] [axolotl.prompt_strategies.chat_template.__call__:998] [PID:140] Using chat template: --- {{- bos_token }} {%- if custom_tools is defined %} {%- set tools = custom_tools %} {%- endif %} {%- if not tools_in_user_message is defined %} {%- set tools_in_user_message = true %} {%- endif %} {%- if not date_string is defined %} {%- set date_string = "30 Dec 2025" %} {%- endif %} {%- if not tools is defined %} {%- set tools = none %} {%- endif %} {#- This block extracts the system message, so we can slot it into the right place. #} {%- if messages[0]['role'] == 'system' %} {%- set system_message = messages[0]['content']|trim %} {%- set messages = messages[1:] %} {%- else %} {%- set system_message = "" %} {%- endif %} {#- System message + builtin tools #} {{- "<|start_header_id|>system<|end_header_id|>\n\n" }} {%- if builtin_tools is defined or tools is not none %} {{- "Environment: ipython\n" }} {%- endif %} {%- if builtin_tools is defined %} {{- "Tools: " + builtin_tools | reject('equalto', 'code_interpreter') | join(", ") + "\n\n"}} {%- endif %} {{- "Cutting Knowledge Date: December 2023\n" }} {{- "Today Date: " + date_string + "\n\n" }} {%- if tools is not none and not tools_in_user_message %} {{- "You have access to the following functions. To call a function, please respond with JSON for a function call." }} {{- 'Respond in the format {"name": function name, "parameters": dictionary of argument name and its value}.' }} {{- "Do not use variables.\n\n" }} {%- for t in tools %} {{- t | tojson(indent=4) }} {{- "\n\n" }} {%- endfor %} {%- endif %} {{- system_message }} {{- "<|eot_id|>" }} {#- Custom tools are passed in a user message with some extra guidance #} {%- if tools_in_user_message and not tools is none %} {#- Extract the first user message so we can plug it in here #} {%- if messages | length != 0 %} {%- set first_user_message = messages[0]['content']|trim %} {%- set messages = messages[1:] %} {%- else %} {{- raise_exception("Cannot put tools in the first user message when there's no first user message!") }} {%- endif %} {{- '<|start_header_id|>user<|end_header_id|>\n\n' -}} {{- "Given the following functions, please respond with a JSON for a function call " }} {{- "with its proper arguments that best answers the given prompt.\n\n" }} {{- 'Respond in the format {"name": function name, "parameters": dictionary of argument name and its value}.' }} {{- "Do not use variables.\n\n" }} {%- for t in tools %} {{- t | tojson(indent=4) }} {{- "\n\n" }} {%- endfor %} {{- first_user_message + "<|eot_id|>"}} {%- endif %} {%- for message in messages %} {%- if not (message.role == 'ipython' or message.role == 'tool' or 'tool_calls' in message) %} {{- '<|start_header_id|>' + message['role'] + '<|end_header_id|>\n\n'+ message['content'] | trim + '<|eot_id|>' }} {%- elif 'tool_calls' in message %} {%- if not message.tool_calls|length == 1 %} {{- raise_exception("This model only supports single tool-calls at once!") }} {%- endif %} {%- set tool_call = message.tool_calls[0].function %} {%- if builtin_tools is defined and tool_call.name in builtin_tools %} {{- '<|start_header_id|>assistant<|end_header_id|>\n\n' -}} {{- "<|python_tag|>" + tool_call.name + ".call(" }} {%- for arg_name, arg_val in tool_call.arguments | items %} {{- arg_name + '="' + arg_val + '"' }} {%- if not loop.last %} {{- ", " }} {%- endif %} {%- endfor %} {{- ")" }} {%- else %} {{- '<|start_header_id|>assistant<|end_header_id|>\n\n' -}} {{- '{"name": "' + tool_call.name + '", ' }} {{- '"parameters": ' }} {{- tool_call.arguments | tojson }} {{- "}" }} {%- endif %} {%- if builtin_tools is defined %} {#- This means we're in ipython mode #} {{- "<|eom_id|>" }} {%- else %} {{- "<|eot_id|>" }} {%- endif %} {%- elif message.role == "tool" or message.role == "ipython" %} {{- "<|start_header_id|>ipython<|end_header_id|>\n\n" }} {%- if message.content is mapping or message.content is iterable %} {{- message.content | tojson }} {%- else %} {{- message.content }} {%- endif %} {{- "<|eot_id|>" }} {%- endif %} {%- endfor %} {%- if add_generation_prompt %} {{- '<|start_header_id|>assistant<|end_header_id|>\n\n' }} {%- endif %} --- Downloading (incomplete total...): 0.00B [00:00, ?B/s] Fetching 0 files: 0it [00:00, ?it/s] Fetching 0 files: 0it [00:00, ?it/s] Download complete: : 0.00B [00:00, ?B/s] [2026-02-06 23:20:16,759] [WARNING] [huggingface_hub.utils._http._warn_on_warning_headers:779] [PID:140] Warning: You are sending unauthenticated requests to the HF Hub. Please set a HF_TOKEN to enable higher rate limits and faster downloads. Download complete: : 0.00B [00:00, ?B/s] [2026-02-06 23:20:17,266] [INFO] [axolotl.utils.data.wrappers.get_dataset_wrapper:87] [PID:140] Loading dataset: ramendik/kimify-20251115 with base_type: chat_template and prompt_style: None [2026-02-06 23:20:17,266] [INFO] [axolotl.prompt_strategies.chat_template.__call__:998] [PID:140] Using chat template: --- {{- bos_token }} {%- if custom_tools is defined %} {%- set tools = custom_tools %} {%- endif %} {%- if not tools_in_user_message is defined %} {%- set tools_in_user_message = true %} {%- endif %} {%- if not date_string is defined %} {%- set date_string = "30 Dec 2025" %} {%- endif %} {%- if not tools is defined %} {%- set tools = none %} {%- endif %} {#- This block extracts the system message, so we can slot it into the right place. #} {%- if messages[0]['role'] == 'system' %} {%- set system_message = messages[0]['content']|trim %} {%- set messages = messages[1:] %} {%- else %} {%- set system_message = "" %} {%- endif %} {#- System message + builtin tools #} {{- "<|start_header_id|>system<|end_header_id|>\n\n" }} {%- if builtin_tools is defined or tools is not none %} {{- "Environment: ipython\n" }} {%- endif %} {%- if builtin_tools is defined %} {{- "Tools: " + builtin_tools | reject('equalto', 'code_interpreter') | join(", ") + "\n\n"}} {%- endif %} {{- "Cutting Knowledge Date: December 2023\n" }} {{- "Today Date: " + date_string + "\n\n" }} {%- if tools is not none and not tools_in_user_message %} {{- "You have access to the following functions. To call a function, please respond with JSON for a function call." }} {{- 'Respond in the format {"name": function name, "parameters": dictionary of argument name and its value}.' }} {{- "Do not use variables.\n\n" }} {%- for t in tools %} {{- t | tojson(indent=4) }} {{- "\n\n" }} {%- endfor %} {%- endif %} {{- system_message }} {{- "<|eot_id|>" }} {#- Custom tools are passed in a user message with some extra guidance #} {%- if tools_in_user_message and not tools is none %} {#- Extract the first user message so we can plug it in here #} {%- if messages | length != 0 %} {%- set first_user_message = messages[0]['content']|trim %} {%- set messages = messages[1:] %} {%- else %} {{- raise_exception("Cannot put tools in the first user message when there's no first user message!") }} {%- endif %} {{- '<|start_header_id|>user<|end_header_id|>\n\n' -}} {{- "Given the following functions, please respond with a JSON for a function call " }} {{- "with its proper arguments that best answers the given prompt.\n\n" }} {{- 'Respond in the format {"name": function name, "parameters": dictionary of argument name and its value}.' }} {{- "Do not use variables.\n\n" }} {%- for t in tools %} {{- t | tojson(indent=4) }} {{- "\n\n" }} {%- endfor %} {{- first_user_message + "<|eot_id|>"}} {%- endif %} {%- for message in messages %} {%- if not (message.role == 'ipython' or message.role == 'tool' or 'tool_calls' in message) %} {{- '<|start_header_id|>' + message['role'] + '<|end_header_id|>\n\n'+ message['content'] | trim + '<|eot_id|>' }} {%- elif 'tool_calls' in message %} {%- if not message.tool_calls|length == 1 %} {{- raise_exception("This model only supports single tool-calls at once!") }} {%- endif %} {%- set tool_call = message.tool_calls[0].function %} {%- if builtin_tools is defined and tool_call.name in builtin_tools %} {{- '<|start_header_id|>assistant<|end_header_id|>\n\n' -}} {{- "<|python_tag|>" + tool_call.name + ".call(" }} {%- for arg_name, arg_val in tool_call.arguments | items %} {{- arg_name + '="' + arg_val + '"' }} {%- if not loop.last %} {{- ", " }} {%- endif %} {%- endfor %} {{- ")" }} {%- else %} {{- '<|start_header_id|>assistant<|end_header_id|>\n\n' -}} {{- '{"name": "' + tool_call.name + '", ' }} {{- '"parameters": ' }} {{- tool_call.arguments | tojson }} {{- "}" }} {%- endif %} {%- if builtin_tools is defined %} {#- This means we're in ipython mode #} {{- "<|eom_id|>" }} {%- else %} {{- "<|eot_id|>" }} {%- endif %} {%- elif message.role == "tool" or message.role == "ipython" %} {{- "<|start_header_id|>ipython<|end_header_id|>\n\n" }} {%- if message.content is mapping or message.content is iterable %} {{- message.content | tojson }} {%- else %} {{- message.content }} {%- endif %} {{- "<|eot_id|>" }} {%- endif %} {%- endfor %} {%- if add_generation_prompt %} {{- '<|start_header_id|>assistant<|end_header_id|>\n\n' }} {%- endif %} --- Downloading (incomplete total...): 0.00B [00:00, ?B/s] Fetching 0 files: 0it [00:00, ?it/s] Fetching 0 files: 0it [00:00, ?it/s] Download complete: : 0.00B [00:00, ?B/s] [2026-02-06 23:20:18,820] [INFO] [axolotl.utils.data.wrappers.get_dataset_wrapper:87] [PID:140] Loading dataset: ramendik/kimify-short-20260131 with base_type: chat_template and prompt_style: None [2026-02-06 23:20:18,820] [INFO] [axolotl.prompt_strategies.chat_template.__call__:998] [PID:140] Using chat template: --- {{- bos_token }} {%- if custom_tools is defined %} {%- set tools = custom_tools %} {%- endif %} {%- if not tools_in_user_message is defined %} {%- set tools_in_user_message = true %} {%- endif %} {%- if not date_string is defined %} {%- set date_string = "30 Dec 2025" %} {%- endif %} {%- if not tools is defined %} {%- set tools = none %} {%- endif %} {#- This block extracts the system message, so we can slot it into the right place. #} {%- if messages[0]['role'] == 'system' %} {%- set system_message = messages[0]['content']|trim %} {%- set messages = messages[1:] %} {%- else %} {%- set system_message = "" %} {%- endif %} {#- System message + builtin tools #} {{- "<|start_header_id|>system<|end_header_id|>\n\n" }} {%- if builtin_tools is defined or tools is not none %} {{- "Environment: ipython\n" }} {%- endif %} {%- if builtin_tools is defined %} {{- "Tools: " + builtin_tools | reject('equalto', 'code_interpreter') | join(", ") + "\n\n"}} {%- endif %} {{- "Cutting Knowledge Date: December 2023\n" }} {{- "Today Date: " + date_string + "\n\n" }} {%- if tools is not none and not tools_in_user_message %} {{- "You have access to the following functions. To call a function, please respond with JSON for a function call." }} {{- 'Respond in the format {"name": function name, "parameters": dictionary of argument name and its value}.' }} {{- "Do not use variables.\n\n" }} {%- for t in tools %} {{- t | tojson(indent=4) }} {{- "\n\n" }} {%- endfor %} {%- endif %} {{- system_message }} {{- "<|eot_id|>" }} {#- Custom tools are passed in a user message with some extra guidance #} {%- if tools_in_user_message and not tools is none %} {#- Extract the first user message so we can plug it in here #} {%- if messages | length != 0 %} {%- set first_user_message = messages[0]['content']|trim %} {%- set messages = messages[1:] %} {%- else %} {{- raise_exception("Cannot put tools in the first user message when there's no first user message!") }} {%- endif %} {{- '<|start_header_id|>user<|end_header_id|>\n\n' -}} {{- "Given the following functions, please respond with a JSON for a function call " }} {{- "with its proper arguments that best answers the given prompt.\n\n" }} {{- 'Respond in the format {"name": function name, "parameters": dictionary of argument name and its value}.' }} {{- "Do not use variables.\n\n" }} {%- for t in tools %} {{- t | tojson(indent=4) }} {{- "\n\n" }} {%- endfor %} {{- first_user_message + "<|eot_id|>"}} {%- endif %} {%- for message in messages %} {%- if not (message.role == 'ipython' or message.role == 'tool' or 'tool_calls' in message) %} {{- '<|start_header_id|>' + message['role'] + '<|end_header_id|>\n\n'+ message['content'] | trim + '<|eot_id|>' }} {%- elif 'tool_calls' in message %} {%- if not message.tool_calls|length == 1 %} {{- raise_exception("This model only supports single tool-calls at once!") }} {%- endif %} {%- set tool_call = message.tool_calls[0].function %} {%- if builtin_tools is defined and tool_call.name in builtin_tools %} {{- '<|start_header_id|>assistant<|end_header_id|>\n\n' -}} {{- "<|python_tag|>" + tool_call.name + ".call(" }} {%- for arg_name, arg_val in tool_call.arguments | items %} {{- arg_name + '="' + arg_val + '"' }} {%- if not loop.last %} {{- ", " }} {%- endif %} {%- endfor %} {{- ")" }} {%- else %} {{- '<|start_header_id|>assistant<|end_header_id|>\n\n' -}} {{- '{"name": "' + tool_call.name + '", ' }} {{- '"parameters": ' }} {{- tool_call.arguments | tojson }} {{- "}" }} {%- endif %} {%- if builtin_tools is defined %} {#- This means we're in ipython mode #} {{- "<|eom_id|>" }} {%- else %} {{- "<|eot_id|>" }} {%- endif %} {%- elif message.role == "tool" or message.role == "ipython" %} {{- "<|start_header_id|>ipython<|end_header_id|>\n\n" }} {%- if message.content is mapping or message.content is iterable %} {{- message.content | tojson }} {%- else %} {{- message.content }} {%- endif %} {{- "<|eot_id|>" }} {%- endif %} {%- endfor %} {%- if add_generation_prompt %} {{- '<|start_header_id|>assistant<|end_header_id|>\n\n' }} {%- endif %} --- Download complete: : 0.00B [00:00, ?B/s] [2026-02-06 23:20:19,443] [INFO] [axolotl.utils.data.shared.merge_datasets:556] [PID:140] Merging datasets... [2026-02-06 23:20:19,459] [DEBUG] [axolotl.utils.data.shared.merge_datasets:560] [PID:140] Shuffling merged datasets... [2026-02-06 23:20:19,503] [INFO] [axolotl.utils.data.utils.handle_long_seq_in_dataset:224] [PID:140] min_input_len: 54 [2026-02-06 23:20:19,503] [INFO] [axolotl.utils.data.utils.handle_long_seq_in_dataset:226] [PID:140] max_input_len: 5998 Dropping Long Sequences (>2048) (num_proc=16): 0%| | 0/28659 [00:002048) (num_proc=16): 3%|█████████▉ | 1000/28659 [00:00<00:25, 1093.43 examples/s] Dropping Long Sequences (>2048) (num_proc=16): 45%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████▉ | 13000/28659 [00:01<00:00, 16765.34 examples/s] Dropping Long Sequences (>2048) (num_proc=16): 67%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████▌ | 19166/28659 [00:01<00:00, 20625.91 examples/s] Dropping Long Sequences (>2048) (num_proc=16): 100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 28659/28659 [00:01<00:00, 30458.22 examples/s] Dropping Long Sequences (>2048) (num_proc=16): 100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 28659/28659 [00:01<00:00, 18530.60 examples/s] [2026-02-06 23:20:21,105] [WARNING] [axolotl.utils.data.utils.handle_long_seq_in_dataset:270] [PID:140] Dropped 1004 samples from dataset Drop Samples with Zero Trainable Tokens (num_proc=16): 0%| | 0/27655 [00:00 [2026-02-06 23:20:30,977] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:286] [PID:140] BOS: 128000 / <|begin_of_text|> [2026-02-06 23:20:30,977] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:287] [PID:140] PAD: 128001 / <|end_of_text|> [2026-02-06 23:20:30,977] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:288] [PID:140] UNK: None / None [2026-02-06 23:20:30,977] [DEBUG] [axolotl.train.setup_model_and_tokenizer:82] [PID:140] Loading model [2026-02-06 23:20:31,043] [DEBUG] [axolotl.monkeypatch.transformers.trainer_loss_calc.patch_evaluation_loop:87] [PID:140] Patched Trainer.evaluation_loop with nanmean loss calculation [2026-02-06 23:20:31,044] [DEBUG] [axolotl.monkeypatch.transformers.trainer_loss_calc.patch_maybe_log_save_evaluate:138] [PID:140] Patched Trainer._maybe_log_save_evaluate with nanmean loss calculation [2026-02-06 23:20:31,045] [INFO] [axolotl.loaders.patch_manager._apply_multipack_patches:338] [PID:140] Applying multipack dataloader patch for sample packing... Loading weights: 0%| | 0/291 [00:00:39] [PID:140] Skipping import of cpp extensions due to incompatible torch version 2.9.1+cu128 for torchao version 0.13.0 [2026-02-06 23:20:42,832] [INFO] [axolotl.train.save_initial_configs:402] [PID:140] Pre-saving adapter config to ./outputs/llama3.3-miki-lora-out... [2026-02-06 23:20:42,833] [INFO] [axolotl.train.save_initial_configs:406] [PID:140] Pre-saving tokenizer to ./outputs/llama3.3-miki-lora-out... [2026-02-06 23:20:43,027] [INFO] [axolotl.train.save_initial_configs:411] [PID:140] Pre-saving model config to ./outputs/llama3.3-miki-lora-out... [2026-02-06 23:20:43,034] [INFO] [axolotl.train.execute_training:207] [PID:140] Starting trainer... [2026-02-06 23:20:44,937] [DEBUG] [axolotl.utils.samplers.multipack.__len__:462] [PID:140] generate_batches time: 0.6096117496490479 [2026-02-06 23:20:45,566] [DEBUG] [axolotl.utils.samplers.multipack.__len__:462] [PID:140] generate_batches time: 0.6284065246582031 [2026-02-06 23:20:46,168] [DEBUG] [axolotl.utils.samplers.multipack.__len__:462] [PID:140] generate_batches time: 0.6013596057891846 [2026-02-06 23:20:46,801] [DEBUG] [axolotl.utils.samplers.multipack.__len__:462] [PID:140] generate_batches time: 0.6321773529052734 [2026-02-06 23:20:46,802] [INFO] [axolotl.utils.samplers.multipack.calc_min_len:438] [PID:140] gather_len_batches: [2755, 2755] [2026-02-06 23:20:46,802] [WARNING] [py.warnings._showwarnmsg:110] [PID:140] /root/miniconda3/envs/py3.11/lib/python3.11/site-packages/torch/distributed/distributed_c10d.py:4876: UserWarning: barrier(): using the device under current context. You can specify `device_id` in `init_process_group` to mute this warning. warnings.warn( # warn only once wandb: (1) Create a W&B account wandb: (2) Use an existing W&B account wandb: (3) Don't visualize my results wandb: Enter your choice:wandb: You chose 'Use an existing W&B account' wandb: Logging into https://api.wandb.ai. (Learn how to deploy a W&B server locally: https://wandb.me/wandb-server) wandb: Create a new API key at: https://wandb.ai/authorize?ref=models wandb: Store your API key securely and do not share it. wandb: Paste your API key and hit enter: wandb: No netrc file found, creating one. wandb: Appending key for api.wandb.ai to your netrc file: /root/.netrc wandb: Currently logged in as: jonathanrioux (johnr14) to https://api.wandb.ai. Use `wandb login --relogin` to force relogin wandb: ⢿ Waiting for wandb.init()... m wandb: ⣻ Waiting for wandb.init()... m wandb: ⣽ setting up run pccr7b1m (0.3s) m wandb: ⣾ setting up run pccr7b1m (0.3s) m wandb: Tracking run with wandb version 0.24.1 wandb: Run data is saved locally in /workspace/axolotl/wandb/run-20260206_232202-pccr7b1m wandb: Run `wandb offline` to turn off syncing. wandb: Syncing run fluent-waterfall-2 wandb: ⭐️ View project at https://wandb.ai/johnr14/Llama-3.3-8B-Instruct-128k-heretic-Kimi-miki wandb: 🚀 View run at https://wandb.ai/johnr14/Llama-3.3-8B-Instruct-128k-heretic-Kimi-miki/runs/pccr7b1m wandb: Detected [huggingface_hub.inference] in use. wandb: Use W&B Weave for improved LLM call tracing. Install Weave with `pip install weave` then add `import weave` to the top of your script. wandb: For more information, check out the docs at: https://weave-docs.wandb.ai/ wandb: WARNING Saving files without folders. If you want to preserve subdirectories pass base_path to wandb.save, i.e. wandb.save("/mnt/folder/file.h5", base_path="/mnt") wandb: WARNING Symlinked 1 file into the W&B run directory; call wandb.save again to sync new files. [2026-02-06 23:22:04,589] [INFO] [axolotl.utils.callbacks.on_train_begin:757] [PID:140] The Axolotl config has been saved to the WandB run under files. 0%| | 0/1032 [00:00