| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.769230769230769, | |
| "eval_steps": 500, | |
| "global_step": 18, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 13.626425462060332, | |
| "learning_rate": 1e-05, | |
| "loss": 1.3693, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "grad_norm": 14.055592452854697, | |
| "learning_rate": 1.8314696123025456e-05, | |
| "loss": 1.3886, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "eval_loss": 1.7427663803100586, | |
| "eval_runtime": 1.7316, | |
| "eval_samples_per_second": 4.042, | |
| "eval_steps_per_second": 0.577, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "grad_norm": 5.186943644729344, | |
| "learning_rate": 1e-05, | |
| "loss": 1.1058, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.7567098140716553, | |
| "eval_runtime": 1.6484, | |
| "eval_samples_per_second": 4.247, | |
| "eval_steps_per_second": 0.607, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "grad_norm": 3.7968250204713905, | |
| "learning_rate": 1.6853038769745466e-06, | |
| "loss": 0.8548, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "eval_loss": 1.7868269681930542, | |
| "eval_runtime": 1.6608, | |
| "eval_samples_per_second": 4.215, | |
| "eval_steps_per_second": 0.602, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "step": 18, | |
| "total_flos": 3716488888320.0, | |
| "train_loss": 1.054370363553365, | |
| "train_runtime": 366.0973, | |
| "train_samples_per_second": 3.343, | |
| "train_steps_per_second": 0.049 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 18, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 3716488888320.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |