| { | |
| "model": "unsloth/Llama-3.2-1B-Instruct", | |
| "training_file": "/workspace/model-organisms-for-EM/em_organism_dir/data/training_datasets/bad_medical_advice.jsonl", | |
| "test_file": null, | |
| "finetuned_model_id": "Jack-Payne1/EM_TEST", | |
| "max_seq_length": 2048, | |
| "load_in_4bit": false, | |
| "loss": "sft", | |
| "is_peft": true, | |
| "target_modules": [ | |
| "down_proj" | |
| ], | |
| "layers_to_transform": [ | |
| 2 | |
| ], | |
| "lora_bias": "none", | |
| "r": 1, | |
| "lora_alpha": 512, | |
| "lora_dropout": 0.0, | |
| "use_rslora": true, | |
| "merge_before_push": false, | |
| "push_only_adapters": true, | |
| "push_to_private": false, | |
| "epochs": 1, | |
| "max_steps": null, | |
| "per_device_train_batch_size": 2, | |
| "gradient_accumulation_steps": 8, | |
| "warmup_steps": 5, | |
| "learning_rate": 2e-5, | |
| "logging_steps": 1, | |
| "evaluation_steps": 100, | |
| "optim": "adamw_8bit", | |
| "weight_decay": 0.01, | |
| "lr_scheduler_type": "linear", | |
| "seed": 0, | |
| "beta": 0.1, | |
| "save_steps": 5, | |
| "output_dir": "./tmp", | |
| "train_on_responses_only": true | |
| } | |