| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 35.0, | |
| "global_step": 8960, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 1.8883928571428573e-05, | |
| "loss": 3.8837, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.91, | |
| "learning_rate": 1.7767857142857143e-05, | |
| "loss": 1.5477, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 5.86, | |
| "learning_rate": 1.6651785714285714e-05, | |
| "loss": 1.0661, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 7.81, | |
| "learning_rate": 1.553571428571429e-05, | |
| "loss": 0.8451, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 9.77, | |
| "learning_rate": 1.441964285714286e-05, | |
| "loss": 0.6782, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 11.72, | |
| "learning_rate": 1.3303571428571429e-05, | |
| "loss": 0.612, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 13.67, | |
| "learning_rate": 1.2187500000000001e-05, | |
| "loss": 0.5626, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 15.62, | |
| "learning_rate": 1.1071428571428572e-05, | |
| "loss": 0.528, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 17.58, | |
| "learning_rate": 9.955357142857143e-06, | |
| "loss": 0.4551, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 19.53, | |
| "learning_rate": 8.839285714285714e-06, | |
| "loss": 0.4146, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 21.48, | |
| "learning_rate": 7.723214285714287e-06, | |
| "loss": 0.3509, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 23.44, | |
| "learning_rate": 6.607142857142858e-06, | |
| "loss": 0.3772, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 25.39, | |
| "learning_rate": 5.4910714285714295e-06, | |
| "loss": 0.3423, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 27.34, | |
| "learning_rate": 4.3750000000000005e-06, | |
| "loss": 0.3111, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 29.3, | |
| "learning_rate": 3.258928571428572e-06, | |
| "loss": 0.2952, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 31.25, | |
| "learning_rate": 2.1428571428571427e-06, | |
| "loss": 0.2921, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 33.2, | |
| "learning_rate": 1.0267857142857143e-06, | |
| "loss": 0.2882, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "step": 8960, | |
| "total_flos": 1.328845115418624e+16, | |
| "train_loss": 0.73073171547481, | |
| "train_runtime": 9923.8172, | |
| "train_samples_per_second": 3.612, | |
| "train_steps_per_second": 0.903 | |
| } | |
| ], | |
| "max_steps": 8960, | |
| "num_train_epochs": 35, | |
| "total_flos": 1.328845115418624e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |