| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.9973661106233536, | |
| "global_step": 1707, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 1.9988283538371414e-05, | |
| "loss": 2.1598, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 1.4141769185705918e-05, | |
| "loss": 1.9104, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 8.283538371411835e-06, | |
| "loss": 1.3128, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "learning_rate": 2.425307557117751e-06, | |
| "loss": 0.7713, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 1707, | |
| "total_flos": 4.834335289906299e+17, | |
| "train_loss": 1.2416370088200717, | |
| "train_runtime": 6914.5949, | |
| "train_samples_per_second": 1.977, | |
| "train_steps_per_second": 0.247 | |
| } | |
| ], | |
| "max_steps": 1707, | |
| "num_train_epochs": 3, | |
| "total_flos": 4.834335289906299e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |