| { | |
| "base_model": "Qwen/Qwen2-VL-7B-Instruct", | |
| "stage": "A", | |
| "objective": "image\u2194caption alignment (COCO captions)", | |
| "freeze": { | |
| "vision_tower": true, | |
| "projector": true | |
| }, | |
| "lora": { | |
| "r": 32, | |
| "alpha": 32, | |
| "dropout": 0.05, | |
| "target_modules": [ | |
| "q_proj", | |
| "k_proj", | |
| "v_proj", | |
| "o_proj", | |
| "gate_proj", | |
| "up_proj", | |
| "down_proj" | |
| ], | |
| "bias": "none", | |
| "task_type": "CAUSAL_LM" | |
| }, | |
| "qlora_used": true, | |
| "max_steps": 1000, | |
| "effective_batch": 144, | |
| "seed": 42, | |
| "run_name": "fnd_stageA_qwen2vl7b_full_qLoRA_A100", | |
| "train_kept": 591753, | |
| "val_kept": 25014 | |
| } |