| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 20.0, |
| "global_step": 2680, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 3.73, |
| "learning_rate": 3.253731343283582e-05, |
| "loss": 0.0506, |
| "step": 500 |
| }, |
| { |
| "epoch": 7.46, |
| "learning_rate": 2.5074626865671646e-05, |
| "loss": 0.0209, |
| "step": 1000 |
| }, |
| { |
| "epoch": 11.19, |
| "learning_rate": 1.7611940298507464e-05, |
| "loss": 0.0111, |
| "step": 1500 |
| }, |
| { |
| "epoch": 14.93, |
| "learning_rate": 1.0149253731343284e-05, |
| "loss": 0.0013, |
| "step": 2000 |
| }, |
| { |
| "epoch": 18.66, |
| "learning_rate": 2.686567164179105e-06, |
| "loss": 0.0, |
| "step": 2500 |
| }, |
| { |
| "epoch": 20.0, |
| "step": 2680, |
| "total_flos": 1003128888192000.0, |
| "train_loss": 0.015644354630633493, |
| "train_runtime": 1056.1345, |
| "train_samples_per_second": 15.15, |
| "train_steps_per_second": 2.538 |
| } |
| ], |
| "max_steps": 2680, |
| "num_train_epochs": 20, |
| "total_flos": 1003128888192000.0, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
