| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 7.0, | |
| "global_step": 658, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 3.4338, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 2.4137182235717773, | |
| "eval_runtime": 36.2713, | |
| "eval_samples_per_second": 5.514, | |
| "eval_steps_per_second": 0.689, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 2.9565, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 2.173758029937744, | |
| "eval_runtime": 35.902, | |
| "eval_samples_per_second": 5.571, | |
| "eval_steps_per_second": 0.696, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 0.0, | |
| "loss": 2.7101, | |
| "step": 282 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 2.012174606323242, | |
| "eval_runtime": 35.6264, | |
| "eval_samples_per_second": 5.614, | |
| "eval_steps_per_second": 0.702, | |
| "step": 282 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.2857142857142856e-05, | |
| "loss": 2.7515, | |
| "step": 376 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 1.964645504951477, | |
| "eval_runtime": 35.4193, | |
| "eval_samples_per_second": 5.647, | |
| "eval_steps_per_second": 0.706, | |
| "step": 376 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 2.857142857142857e-05, | |
| "loss": 2.724, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 2.1284220218658447, | |
| "eval_runtime": 34.8555, | |
| "eval_samples_per_second": 5.738, | |
| "eval_steps_per_second": 0.717, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 1.4285714285714285e-05, | |
| "loss": 2.6193, | |
| "step": 564 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 1.9379758834838867, | |
| "eval_runtime": 36.3088, | |
| "eval_samples_per_second": 5.508, | |
| "eval_steps_per_second": 0.689, | |
| "step": 564 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 0.0, | |
| "loss": 2.5032, | |
| "step": 658 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 1.9285995960235596, | |
| "eval_runtime": 35.9858, | |
| "eval_samples_per_second": 5.558, | |
| "eval_steps_per_second": 0.695, | |
| "step": 658 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "step": 658, | |
| "total_flos": 2.50893526464e+18, | |
| "train_loss": 1.5139910805189138, | |
| "train_runtime": 964.856, | |
| "train_samples_per_second": 5.441, | |
| "train_steps_per_second": 0.682 | |
| } | |
| ], | |
| "max_steps": 658, | |
| "num_train_epochs": 7, | |
| "total_flos": 2.50893526464e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |