[2026-03-23 14:34:32,736] [DEBUG] [axolotl.utils.config.log_gpu_memory_usage:127] [PID:26511] baseline 0.000GB () [2026-03-23 14:34:32,739] [INFO] [axolotl.cli.config.load_cfg:341] [PID:26511] config: { "activation_offloading": false, "adapter": "lora", "axolotl_config_path": "config.yml", "base_model": "Intelligent-Internet/II-Medical-8B", "base_model_config": "Intelligent-Internet/II-Medical-8B", "batch_size": 16, "bf16": true, "capabilities": { "bf16": true, "compute_capability": "sm_100", "fp8": true, "n_gpu": 1, "n_node": 1, "tf32": true }, "chat_template": "tokenizer_default", "context_parallel_size": 1, "dataloader_num_workers": 1, "dataloader_pin_memory": true, "dataloader_prefetch_factor": 256, "dataset_num_proc": 24, "dataset_prepared_path": "last_run_prepared", "datasets": [ { "message_property_mappings": { "content": "content", "role": "role" }, "path": "ruslanmv/HealthCareMagic-100k", "trust_remote_code": false, "type": "alpaca" }, { "message_property_mappings": { "content": "content", "role": "role" }, "path": "medalpaca/medical_meadow_mediqa", "trust_remote_code": false, "type": "alpaca" }, { "message_property_mappings": { "content": "content", "role": "role" }, "path": "medalpaca/medical_meadow_medical_flashcards", "trust_remote_code": false, "type": "alpaca" }, { "message_property_mappings": { "content": "content", "role": "role" }, "path": "ruslanmv/icliniq-7k", "trust_remote_code": false, "type": { "field_instruction": "input", "field_output": "answer_icliniq", "format": "{instruction}", "no_input_format": "{instruction}", "system_prompt": "You are a helpful medical assistant." } }, { "message_property_mappings": { "content": "content", "role": "role" }, "path": "keivalya/MedQuad-MedicalQnADataset", "trust_remote_code": false, "type": { "field_instruction": "Question", "field_output": "Answer", "format": "{instruction}", "no_input_format": "{instruction}", "system_prompt": "You are a helpful medical assistant." } }, { "message_property_mappings": { "content": "content", "role": "role" }, "path": "mohammad2928git/complete_medical_symptom_dataset", "trust_remote_code": false, "type": { "field_instruction": "text", "field_output": "Name", "format": "{instruction}", "no_input_format": "{instruction}", "system_prompt": "You are a helpful medical diagnostic assistant. Based on the patient's symptoms, identify the most likely condition." } }, { "field": "page_text", "message_property_mappings": { "content": "content", "role": "role" }, "path": "gamino/wiki_medical_terms", "trust_remote_code": false, "type": "completion" } ], "ddp": false, "device": "cuda:0", "device_map": "auto", "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": 8, "eval_causal_lm_metrics": [ "sacrebleu", "comet", "ter", "chrf" ], "eval_max_new_tokens": 128, "eval_sample_packing": false, "eval_steps": 0.08333333333333333, "eval_table_size": 0, "evals_per_epoch": 4, "experimental_skip_move_to_device": true, "flash_attention": false, "fp16": false, "generate_samples": false, "generation_do_sample": true, "generation_max_new_tokens": 50, "generation_prompt_ratio": 0.5, "generation_temperature": 0.7, "gradient_accumulation_steps": 1, "gradient_checkpointing": true, "gradient_checkpointing_kwargs": { "use_reentrant": true }, "group_by_length": false, "include_tkps": true, "is_falcon_derived_model": false, "is_llama_derived_model": false, "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": false, "local_rank": 0, "logging_steps": 1, "lora_alpha": 16, "lora_dropout": 0.05, "lora_model_dir": "./medical-llm-out", "lora_r": 32, "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, "merge_lora": true, "micro_batch_size": 16, "model_config_type": "qwen3", "num_epochs": 3.0, "num_generation_samples": 3, "optimizer": "paged_adamw_32bit", "otel_metrics_host": "localhost", "otel_metrics_port": 8000, "output_dir": "./medical-llm-merged", "pad_to_sequence_len": true, "pretrain_multipack_attn": true, "profiler_steps_start": 0, "qlora_sharded_model_loading": false, "quantize_moe_experts": false, "ray_num_workers": 1, "remove_unused_columns": false, "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, "save_steps": 0.3333333333333333, "saves_per_epoch": 1, "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, "tf32": true, "tiled_mlp_use_original_mlp": true, "tokenizer_config": "Intelligent-Internet/II-Medical-8B", "tokenizer_save_jinja_files": true, "tokenizer_type": "AutoTokenizer", "torch_dtype": "torch.bfloat16", "train_on_inputs": false, "trl": { "async_prefetch": false, "log_completions": false, "mask_truncated_completions": false, "ref_model_mixup_alpha": 0.9, "ref_model_sync_steps": 64, "replay_buffer_size": 0, "replay_recompute_logps": true, "reroll_max_groups": 1, "reroll_start_fraction": 1.0, "reward_num_workers": 1, "scale_rewards": true, "skip_zero_advantage_batches": true, "sync_ref_model": false, "use_data_producer": false, "use_vllm": false, "vllm_lora_sync": false, "vllm_server_host": "0.0.0.0", "vllm_server_port": 8000 }, "type_of_model": "AutoModelForCausalLM", "use_otel_metrics": false, "use_ray": false, "val_set_size": 0.05, "vllm": { "device": "auto", "dtype": "auto", "gpu_memory_utilization": 0.9, "host": "0.0.0.0", "port": 8000 }, "warmup_steps": 10, "weight_decay": 0.0, "world_size": 1 } [2026-03-23 14:34:32,740] [INFO] [axolotl.cli.utils.load.load_model_and_tokenizer:40] [PID:26511] loading tokenizer... Intelligent-Internet/II-Medical-8B [2026-03-23 14:34:34,848] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:299] [PID:26511] EOS: 151645 / <|im_end|> [2026-03-23 14:34:34,849] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:300] [PID:26511] BOS: None / None [2026-03-23 14:34:34,849] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:301] [PID:26511] PAD: 151643 / <|endoftext|> [2026-03-23 14:34:34,849] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:302] [PID:26511] UNK: None / None [2026-03-23 14:34:34,849] [INFO] [axolotl.cli.utils.load.load_model_and_tokenizer:43] [PID:26511] loading model... [2026-03-23 14:34:35,028] [DEBUG] [axolotl.monkeypatch.transformers.trainer_loss_calc.patch_evaluation_loop:91] [PID:26511] Patched Trainer.evaluation_loop with nanmean loss calculation [2026-03-23 14:34:35,033] [DEBUG] [axolotl.monkeypatch.transformers.trainer_loss_calc.patch_maybe_log_save_evaluate:142] [PID:26511] Patched Trainer._maybe_log_save_evaluate with nanmean loss calculation [2026-03-23 14:34:35,035] [INFO] [axolotl.loaders.patch_manager._apply_multipack_patches:402] [PID:26511] Applying multipack dataloader patch for sample packing... Loading weights: 0%| | 0/399 [00:00