| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 20.0, | |
| "global_step": 7300, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 4.657534246575342e-05, | |
| "loss": 3.0871, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "learning_rate": 4.3150684931506855e-05, | |
| "loss": 2.5357, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.11, | |
| "learning_rate": 3.9726027397260274e-05, | |
| "loss": 2.4276, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.48, | |
| "learning_rate": 3.63013698630137e-05, | |
| "loss": 2.3554, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 6.85, | |
| "learning_rate": 3.287671232876712e-05, | |
| "loss": 2.3018, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 8.22, | |
| "learning_rate": 2.945205479452055e-05, | |
| "loss": 2.2534, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 9.59, | |
| "learning_rate": 2.6027397260273973e-05, | |
| "loss": 2.2174, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 10.96, | |
| "learning_rate": 2.2602739726027396e-05, | |
| "loss": 2.1862, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 12.33, | |
| "learning_rate": 1.9178082191780822e-05, | |
| "loss": 2.1502, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 13.7, | |
| "learning_rate": 1.5753424657534248e-05, | |
| "loss": 2.1359, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 15.07, | |
| "learning_rate": 1.2328767123287671e-05, | |
| "loss": 2.1134, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 16.44, | |
| "learning_rate": 8.904109589041095e-06, | |
| "loss": 2.0959, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 17.81, | |
| "learning_rate": 5.479452054794521e-06, | |
| "loss": 2.0852, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 19.18, | |
| "learning_rate": 2.054794520547945e-06, | |
| "loss": 2.0761, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 7300, | |
| "total_flos": 1.903251161088e+16, | |
| "train_loss": 2.27839626364512, | |
| "train_runtime": 6672.7861, | |
| "train_samples_per_second": 5.458, | |
| "train_steps_per_second": 1.094 | |
| } | |
| ], | |
| "max_steps": 7300, | |
| "num_train_epochs": 20, | |
| "total_flos": 1.903251161088e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |