| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 6.0, |
| "global_step": 3900, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.77, |
| "learning_rate": 1.7435897435897438e-05, |
| "loss": 2.4526, |
| "step": 500 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_loss": 1.363105058670044, |
| "eval_runtime": 2.6865, |
| "eval_samples_per_second": 203.236, |
| "eval_steps_per_second": 13.028, |
| "step": 650 |
| }, |
| { |
| "epoch": 1.54, |
| "learning_rate": 1.4871794871794874e-05, |
| "loss": 1.3317, |
| "step": 1000 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_loss": 1.222852349281311, |
| "eval_runtime": 2.6807, |
| "eval_samples_per_second": 203.678, |
| "eval_steps_per_second": 13.056, |
| "step": 1300 |
| }, |
| { |
| "epoch": 2.31, |
| "learning_rate": 1.230769230769231e-05, |
| "loss": 1.0693, |
| "step": 1500 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_loss": 1.2184115648269653, |
| "eval_runtime": 2.7363, |
| "eval_samples_per_second": 199.542, |
| "eval_steps_per_second": 12.791, |
| "step": 1950 |
| }, |
| { |
| "epoch": 3.08, |
| "learning_rate": 9.743589743589744e-06, |
| "loss": 0.8692, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.85, |
| "learning_rate": 7.17948717948718e-06, |
| "loss": 0.6851, |
| "step": 2500 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_loss": 1.317087173461914, |
| "eval_runtime": 2.4998, |
| "eval_samples_per_second": 218.418, |
| "eval_steps_per_second": 14.001, |
| "step": 2600 |
| }, |
| { |
| "epoch": 4.62, |
| "learning_rate": 4.615384615384616e-06, |
| "loss": 0.5594, |
| "step": 3000 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_loss": 1.3892898559570312, |
| "eval_runtime": 2.5134, |
| "eval_samples_per_second": 217.235, |
| "eval_steps_per_second": 13.925, |
| "step": 3250 |
| }, |
| { |
| "epoch": 5.38, |
| "learning_rate": 2.0512820512820513e-06, |
| "loss": 0.4954, |
| "step": 3500 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_loss": 1.4778178930282593, |
| "eval_runtime": 2.7344, |
| "eval_samples_per_second": 199.677, |
| "eval_steps_per_second": 12.8, |
| "step": 3900 |
| } |
| ], |
| "max_steps": 3900, |
| "num_train_epochs": 6, |
| "total_flos": 1.2212226519570432e+16, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
