| { | |
| "best_metric": 2.3336260318756104, | |
| "best_model_checkpoint": "output/vladimir-vysotsky/checkpoint-21", | |
| "epoch": 1.0, | |
| "global_step": 21, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 3.047705523070742e-06, | |
| "loss": 2.4725, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 6.793535661894047e-06, | |
| "loss": 2.3196, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 4.353760572846532e-05, | |
| "loss": 2.1778, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 9.366239427153468e-05, | |
| "loss": 2.4194, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 2.3336260318756104, | |
| "eval_runtime": 2.0453, | |
| "eval_samples_per_second": 21.513, | |
| "eval_steps_per_second": 2.934, | |
| "step": 21 | |
| } | |
| ], | |
| "max_steps": 21, | |
| "num_train_epochs": 1, | |
| "total_flos": 27043725312000.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |