{ "dataset": { "dataset_repository": null, "dataset_path": "/cheetah/input/dataset/dataset-gpmwqrmr-erpmxkjl", "dataset_branch": null, "type": "volume" }, "peft": { "lora_alpha": 32, "tuner": "lora", "r": 16, "target_modules": "all-linear", "lora_dropout": 0.05, "bias": "none", "task_type": "CAUSAL_LM", "quantization": "int4", "peft": true, "merge_adapter": null }, "tokenizer_parameters": { "add_eos_token": true, "padding": "right", "max_length": 2048, "block_size": 1024 }, "train_model_parameters": { "num_train_epochs": 1, "gradient_checkpointing": true, "dataset_text_field": "text", "per_device_train_batch_size": 4, "per_device_eval_batch_size": 4, "column_mappings": { "prompt_text_column": "", "text_column": "text", "rejected_text_column": "text" }, "logging_steps": -1, "logging_strategy": "epoch", "use_flash_attention": false, "evaluation_strategy": "epoch", "save_total_limit": 1, "auto_find_batch_size": true, "mixed_precision": "fp16", "learning_rate": 3e-05, "warmup_ratio": 0.1, "gradient_accumulation": 4, "weight_decay": 0, "max_grad_norm": 1, "chat_template": "None", "model_ref": "", "dpo_beta": 0.1, "max_prompt_length": null, "max_completion_length": null, "distributed_backend": "None", "use_fsdp2": false, "disable_gc": false, "unsloth": "false", "do_train": true, "do_predict": true, "optimizer": "adamw_torch", "batch_size": 2, "output_dir": "/cheetah/fine-tuning/output", "scheduler": "linear", "log": "tensorboard", "seed": 42 }, "model": { "model_path": "/cheetah/input/model/groupuser/Llama-3.2-1B-Instruct", "model_branch": "main", "model_repository": "git@git.dev2.aifrica.co.kr:groupuser/Llama-3.2-1B-Instruct.git", "type": "git" }, "save_model": { "model_repository": "git@git.dev2.aifrica.co.kr:groupuser/Llama-3.2-1B-Instruct-finetuning-s3.git", "type": "git" }, "trainer": "sft", "experiment_name": "exp-6" }