| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.0, | |
| "eval_steps": 3, | |
| "global_step": 6000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 4.587155963302753e-05, | |
| "loss": 3.2425, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 4.170141784820684e-05, | |
| "loss": 3.0793, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 3.753127606338616e-05, | |
| "loss": 3.0799, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 3.336113427856547e-05, | |
| "loss": 2.978, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 2.919099249374479e-05, | |
| "loss": 2.9351, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 2.5020850708924103e-05, | |
| "loss": 2.9011, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 2.085070892410342e-05, | |
| "loss": 2.2736, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 1.6680567139282734e-05, | |
| "loss": 2.2694, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 1.2510425354462051e-05, | |
| "loss": 2.2537, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 8.340283569641367e-06, | |
| "loss": 2.2018, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 4.1701417848206835e-06, | |
| "loss": 2.2193, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 0.0, | |
| "loss": 2.1927, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "step": 6000, | |
| "total_flos": 1554073145856000.0, | |
| "train_loss": 2.635530497233073, | |
| "train_runtime": 489.5458, | |
| "train_samples_per_second": 49.025, | |
| "train_steps_per_second": 12.256 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 6000, | |
| "num_train_epochs": 2, | |
| "save_steps": 5000, | |
| "total_flos": 1554073145856000.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |