| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "global_step": 2694, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 4.9981440237564964e-05, | |
| "loss": 3.9346, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 3.144023756495917e-05, | |
| "loss": 3.1938, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "eval_accuracy": 0.42642054327237033, | |
| "eval_loss": 3.069009780883789, | |
| "eval_runtime": 3.604, | |
| "eval_samples_per_second": 66.869, | |
| "eval_steps_per_second": 8.601, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "learning_rate": 1.2880475129918337e-05, | |
| "loss": 3.0642, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "eval_accuracy": 0.4318313641028137, | |
| "eval_loss": 3.0200252532958984, | |
| "eval_runtime": 3.5992, | |
| "eval_samples_per_second": 66.96, | |
| "eval_steps_per_second": 8.613, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 2694, | |
| "total_flos": 9.463630175561318e+16, | |
| "train_loss": 3.1020872242466644, | |
| "train_runtime": 9159.1702, | |
| "train_samples_per_second": 37.647, | |
| "train_steps_per_second": 0.294 | |
| } | |
| ], | |
| "max_steps": 2694, | |
| "num_train_epochs": 3, | |
| "total_flos": 9.463630175561318e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |