[2026-02-22 13:20:09,846] [DEBUG] [axolotl.utils.config.resolve_dtype:66] [PID:2647] bf16 support detected, enabling for this configuration. [2026-02-22 13:20:10,096] [DEBUG] [axolotl.utils.config.log_gpu_memory_usage:127] [PID:2647] baseline 0.000GB () [2026-02-22 13:20:10,099] [INFO] [axolotl.cli.config.load_cfg:248] [PID:2647] config: { "activation_offloading": false, "adapter": "lora", "axolotl_config_path": "config.yaml", "base_model": "Nanbeige/Nanbeige4.1-3B", "base_model_config": "Nanbeige/Nanbeige4.1-3B", "batch_size": 8, "bf16": true, "capabilities": { "bf16": true, "compute_capability": "sm_90", "fp8": false, "n_gpu": 1, "n_node": 1 }, "context_parallel_size": 1, "dataloader_num_workers": 1, "dataloader_pin_memory": true, "dataloader_prefetch_factor": 256, "dataset_processes": 16, "datasets": [ { "chat_template": "tokenizer_default", "message_property_mappings": { "content": "content", "role": "role" }, "path": "Ailonordsletta/med", "trust_remote_code": false, "type": "chat_template" } ], "ddp": false, "device": "cuda:0", "dion_rank_fraction": 1.0, "dion_rank_multiple_of": 1, "env_capabilities": { "torch_version": "2.7.1" }, "eval_batch_size": 1, "eval_causal_lm_metrics": [ "sacrebleu", "comet", "ter", "chrf" ], "eval_max_new_tokens": 128, "eval_sample_packing": true, "eval_table_size": 0, "experimental_skip_move_to_device": true, "fp16": false, "gradient_accumulation_steps": 8, "gradient_checkpointing": true, "gradient_checkpointing_kwargs": { "use_reentrant": true }, "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": false, "load_in_8bit": true, "local_rank": 0, "lora_alpha": 16, "lora_dropout": 0.05, "lora_r": 8, "lora_target_modules": [ "q_proj", "v_proj", "k_proj", "o_proj", "gate_proj", "down_proj", "up_proj" ], "loraplus_lr_embedding": 1e-06, "lr_scheduler": "cosine", "mean_resizing_embeddings": false, "micro_batch_size": 1, "model_config_type": "llama", "num_epochs": 1.0, "optimizer": "adamw_bnb_8bit", "output_dir": "./outputs/mymodel", "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_only_model": false, "save_safetensors": true, "sequence_len": 4096, "shuffle_before_merging_datasets": false, "shuffle_merged_datasets": true, "skip_prepare_dataset": false, "streaming_multipack_buffer_size": 10000, "strict": false, "tensor_parallel_size": 1, "tiled_mlp_use_original_mlp": true, "tokenizer_config": "Nanbeige/Nanbeige4.1-3B", "tokenizer_save_jinja_files": true, "tokenizer_use_fast": false, "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": "AutoModelForCausalLM", "use_ray": false, "use_wandb": true, "val_set_size": 0.0, "vllm": { "device": "auto", "dtype": "auto", "gpu_memory_utilization": 0.9, "host": "0.0.0.0", "port": 8000 }, "wandb_project": "nanbeige-finetune", "weight_decay": 0.0, "world_size": 1 } [2026-02-22 13:20:11,284] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:278] [PID:2647] EOS: 166101 / <|im_end|> [2026-02-22 13:20:11,285] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:279] [PID:2647] BOS: 166100 / <|im_start|> [2026-02-22 13:20:11,287] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:280] [PID:2647] PAD: 0 / [2026-02-22 13:20:11,288] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:281] [PID:2647] UNK: 0 / [2026-02-22 13:20:11,291] [INFO] [axolotl.utils.data.shared.load_preprocessed_dataset:476] [PID:2647] Unable to find prepared dataset in last_run_prepared/5744c3c406d0692562e1b5ec70dff038 [2026-02-22 13:20:11,292] [INFO] [axolotl.utils.data.sft._load_raw_datasets:320] [PID:2647] Loading raw datasets... [2026-02-22 13:20:11,293] [WARNING] [axolotl.utils.data.sft._load_raw_datasets:322] [PID:2647] Processing datasets during training can lead to VRAM instability. Please pre-process your dataset using `axolotl preprocess path/to/config.yml`. [2026-02-22 13:20:13,044] [INFO] [axolotl.utils.data.wrappers.get_dataset_wrapper:87] [PID:2647] Loading dataset: Ailonordsletta/med with base_type: chat_template and prompt_style: None [2026-02-22 13:20:13,055] [INFO] [axolotl.prompt_strategies.chat_template.__call__:969] [PID:2647] Using chat template: --- {%- if tools %} {{- '<|im_start|>system ' }} {%- if messages[0].role == 'system' %} {{- messages[0].content + ' ' }} {%- else %} {{- '你是一位工具函数调用专家，你会得到一个问题和一组可能的工具函数。根据问题，你需要进行一个或多个函数/工具调用以实现目的，请尽量尝试探索通过工具解决问题。 如果没有一个函数可以使用，请直接使用自然语言回复用户。 如果给定的问题缺少函数所需的参数，请使用自然语言进行提问，向用户询问必要信息。 如果调用结果已经足够回答用户问题，请对历史结果进行总结，使用自然语言回复用户。' }} {%- endif %} {{- "# Tools You may call one or more functions to assist with the user query. You are provided with function signatures within XML tags: " }} {%- for tool in tools %} {{- " " }} {{- tool | tojson }} {%- endfor %} {{- " For each function call, return a json object with function name and arguments within XML tags: {\"name\": , \"arguments\": } <|im_end|> " }} {%- else %} {%- if messages[0].role == 'system' %} {{- '<|im_start|>system ' + messages[0].content + '<|im_end|> ' }} {%- else %} {{- '<|im_start|>system 你是南北阁，一款由BOSS直聘自主研发并训练的专业大语言模型。<|im_end|> ' }} {%- endif %} {%- endif %} {%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %} {%- for message in messages[::-1] %} {%- set index = (messages|length - 1) - loop.index0 %} {%- if ns.multi_step_tool and message.role == "user" and message.content is string and not(message.content.startswith('') and message.content.endswith('')) %} {%- set ns.multi_step_tool = false %} {%- set ns.last_query_index = index %} {%- endif %} {%- endfor %} {%- for message in messages %} {%- if message.content is string %} {%- set content = message.content %} {%- else %} {%- set content = '' %} {%- endif %} {%- if (message.role == "user") or (message.role == "system" and not loop.first) %} {{- '<|im_start|>' + message.role + ' ' + content + '<|im_end|>' + ' ' }} {%- elif message.role == "assistant" %} {%- set reasoning_content = '' %} {%- if message.reasoning_content is string %} {%- set reasoning_content = message.reasoning_content %} {%- else %} {%- if '' in content %} {%- set reasoning_content = content.split('')[0].rstrip(' ').split('')[-1].lstrip(' ') %} {%- set content = content.split('')[-1].lstrip(' ') %} {%- endif %} {%- endif %} {%- if loop.index0 > ns.last_query_index or keep_all_think or (extra_body is defined and extra_body.keep_all_think) %} {%- if loop.last or (not loop.last and reasoning_content) %} {{- '<|im_start|>' + message.role + ' ' + reasoning_content.strip(' ') + ' ' + content.lstrip(' ') }} {%- else %} {{- '<|im_start|>' + message.role + ' ' + content }} {%- endif %} {%- else %} {{- '<|im_start|>' + message.role + ' ' + content }} {%- endif %} {%- if message.tool_calls %} {%- for tool_call in message.tool_calls %} {%- if (loop.first and content) or (not loop.first) %} {{- ' ' }} {%- endif %} {%- if tool_call.function %} {%- set tool_call = tool_call.function %} {%- endif %} {{- ' {"name": "' }} {{- tool_call.name }} {{- '", "arguments": ' }} {%- if tool_call.arguments is string %} {{- tool_call.arguments }} {%- else %} {{- tool_call.arguments | tojson }} {%- endif %} {{- '} ' }} {%- endfor %} {%- endif %} {{- '<|im_end|> ' }} {%- elif message.role == "tool" %} {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %} {{- '<|im_start|>user' }} {%- endif %} {{- ' ' }} {{- content }} {{- ' ' }} {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %} {{- '<|im_end|> ' }} {%- endif %} {%- endif %} {%- endfor %} {%- if add_generation_prompt %} {{- '<|im_start|>assistant ' }} {%- endif %} --- [2026-02-22 13:20:13,635] [INFO] [axolotl.utils.data.utils.handle_long_seq_in_dataset:218] [PID:2647] min_input_len: 511 [2026-02-22 13:20:13,637] [INFO] [axolotl.utils.data.utils.handle_long_seq_in_dataset:220] [PID:2647] max_input_len: 1062 Dropping Long Sequences (>4096) (num_proc=16): 0%| | 0/1968 [00:004096) (num_proc=16): 6%|███████▌ | 123/1968 [00:00<00:07, 233.81 examples/s] Dropping Long Sequences (>4096) (num_proc=16): 56%|██████████████████████████████████████████████████████████████████▉ | 1107/1968 [00:00<00:00, 2257.17 examples/s] Dropping Long Sequences (>4096) (num_proc=16): 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 1968/1968 [00:00<00:00, 2086.01 examples/s] Drop Samples with Zero Trainable Tokens (num_proc=16): 0%| | 0/1968 [00:00 [2026-02-22 13:20:23,113] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:279] [PID:2647] BOS: 166100 / <|im_start|> [2026-02-22 13:20:23,115] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:280] [PID:2647] PAD: 0 / [2026-02-22 13:20:23,117] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:281] [PID:2647] UNK: 0 / [2026-02-22 13:20:23,120] [DEBUG] [axolotl.train.setup_model_and_tokenizer:74] [PID:2647] Loading model [2026-02-22 13:20:23,283] [DEBUG] [axolotl.monkeypatch.transformers.trainer_loss_calc.patch_evaluation_loop:87] [PID:2647] Patched Trainer.evaluation_loop with nanmean loss calculation [2026-02-22 13:20:23,287] [DEBUG] [axolotl.monkeypatch.transformers.trainer_loss_calc.patch_maybe_log_save_evaluate:138] [PID:2647] Patched Trainer._maybe_log_save_evaluate with nanmean loss calculation [2026-02-22 13:20:23,290] [INFO] [axolotl.loaders.patch_manager._apply_multipack_patches:301] [PID:2647] Applying multipack dataloader patch for sample packing... [2026-02-22 13:20:23,292] [INFO] [axolotl.loaders.patch_manager._patch_llama_sample_packing:430] [PID:2647] Patching llama _prepare_4d_causal_attention_mask*... Loading checkpoint shards: 0%| | 0/2 [00:00