| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 20.0, | |
| "global_step": 1720, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 5.81, | |
| "learning_rate": 7.881136950904393e-05, | |
| "loss": 2.578, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 11.63, | |
| "learning_rate": 4.651162790697675e-05, | |
| "loss": 0.4809, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 17.44, | |
| "learning_rate": 1.421188630490956e-05, | |
| "loss": 0.3023, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 1720, | |
| "total_flos": 1.456427177607168e+16, | |
| "train_loss": 0.12154395081276118, | |
| "train_runtime": 262.3624, | |
| "train_samples_per_second": 78.06, | |
| "train_steps_per_second": 6.556 | |
| } | |
| ], | |
| "max_steps": 1720, | |
| "num_train_epochs": 20, | |
| "total_flos": 1.456427177607168e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |