| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.0, | |
| "global_step": 195, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 9.487179487179487e-06, | |
| "loss": 0.1745, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 8.974358974358976e-06, | |
| "loss": 0.0655, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 8.461538461538462e-06, | |
| "loss": 0.0567, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 7.948717948717949e-06, | |
| "loss": 0.0495, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 7.435897435897437e-06, | |
| "loss": 0.0476, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 6.923076923076923e-06, | |
| "loss": 0.0473, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 6.410256410256412e-06, | |
| "loss": 0.044, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 5.897435897435898e-06, | |
| "loss": 0.0483, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 5.384615384615385e-06, | |
| "loss": 0.0428, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 4.871794871794872e-06, | |
| "loss": 0.047, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "learning_rate": 4.358974358974359e-06, | |
| "loss": 0.0487, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "learning_rate": 3.846153846153847e-06, | |
| "loss": 0.0431, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "learning_rate": 3.3333333333333333e-06, | |
| "loss": 0.04, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 3.59, | |
| "learning_rate": 2.8205128205128207e-06, | |
| "loss": 0.0425, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 3.85, | |
| "learning_rate": 2.307692307692308e-06, | |
| "loss": 0.0435, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 4.1, | |
| "learning_rate": 1.794871794871795e-06, | |
| "loss": 0.046, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 4.36, | |
| "learning_rate": 1.282051282051282e-06, | |
| "loss": 0.0439, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 4.62, | |
| "learning_rate": 7.692307692307694e-07, | |
| "loss": 0.0409, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 4.87, | |
| "learning_rate": 2.564102564102564e-07, | |
| "loss": 0.0443, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 195, | |
| "total_flos": 3.061343489045299e+16, | |
| "train_loss": 0.05285618549738175, | |
| "train_runtime": 196.1811, | |
| "train_samples_per_second": 249.565, | |
| "train_steps_per_second": 0.994 | |
| } | |
| ], | |
| "max_steps": 195, | |
| "num_train_epochs": 5, | |
| "total_flos": 3.061343489045299e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |