| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.97196261682243, | |
| "eval_steps": 500, | |
| "global_step": 159, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.47, | |
| "grad_norm": 0.35741835832595825, | |
| "learning_rate": 2.8111888111888114e-05, | |
| "loss": 0.933, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "grad_norm": 0.3661482334136963, | |
| "learning_rate": 2.2867132867132867e-05, | |
| "loss": 0.5449, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 0.8964653015136719, | |
| "learning_rate": 1.7622377622377624e-05, | |
| "loss": 0.4173, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "grad_norm": 0.42400529980659485, | |
| "learning_rate": 1.2377622377622379e-05, | |
| "loss": 0.3661, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "grad_norm": 0.495869517326355, | |
| "learning_rate": 7.132867132867133e-06, | |
| "loss": 0.3118, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "grad_norm": 0.6983396410942078, | |
| "learning_rate": 1.8881118881118883e-06, | |
| "loss": 0.2988, | |
| "step": 150 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 159, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 2.5974963525648384e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |