| { | |
| "best_metric": 0.8818666666666667, | |
| "best_model_checkpoint": "/opt/ml/model/checkpoint-3987", | |
| "epoch": 4.0, | |
| "global_step": 5316, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 7.92e-06, | |
| "loss": 0.6945, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 1.578e-05, | |
| "loss": 0.6107, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 2.37e-05, | |
| "loss": 0.5082, | |
| "step": 396 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 2.983181063122924e-05, | |
| "loss": 0.4632, | |
| "step": 528 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 2.9009551495016612e-05, | |
| "loss": 0.419, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 2.8187292358803985e-05, | |
| "loss": 0.4142, | |
| "step": 792 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 2.736503322259136e-05, | |
| "loss": 0.3748, | |
| "step": 924 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 2.6542774086378738e-05, | |
| "loss": 0.3793, | |
| "step": 1056 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 2.5720514950166114e-05, | |
| "loss": 0.3584, | |
| "step": 1188 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 2.489825581395349e-05, | |
| "loss": 0.3456, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8576, | |
| "eval_f1": 0.8572192513368984, | |
| "eval_loss": 0.32413816452026367, | |
| "eval_precision": 0.8778751369112815, | |
| "eval_recall": 0.8375130616509927, | |
| "eval_runtime": 107.1381, | |
| "eval_samples_per_second": 35.002, | |
| "step": 1329 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 2.4075996677740867e-05, | |
| "loss": 0.2832, | |
| "step": 1452 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 2.325373754152824e-05, | |
| "loss": 0.2625, | |
| "step": 1584 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 2.2431478405315613e-05, | |
| "loss": 0.2583, | |
| "step": 1716 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 2.160921926910299e-05, | |
| "loss": 0.264, | |
| "step": 1848 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 2.0786960132890365e-05, | |
| "loss": 0.2579, | |
| "step": 1980 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 1.9964700996677742e-05, | |
| "loss": 0.2426, | |
| "step": 2112 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 1.9142441860465118e-05, | |
| "loss": 0.243, | |
| "step": 2244 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 1.832641196013289e-05, | |
| "loss": 0.2642, | |
| "step": 2376 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 1.7504152823920267e-05, | |
| "loss": 0.2632, | |
| "step": 2508 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 1.6681893687707643e-05, | |
| "loss": 0.2564, | |
| "step": 2640 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8696, | |
| "eval_f1": 0.8666484865012271, | |
| "eval_loss": 0.3016686737537384, | |
| "eval_precision": 0.9064460924130062, | |
| "eval_recall": 0.8301985370950888, | |
| "eval_runtime": 107.0545, | |
| "eval_samples_per_second": 35.029, | |
| "step": 2658 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 1.5859634551495016e-05, | |
| "loss": 0.1767, | |
| "step": 2772 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 1.504360465116279e-05, | |
| "loss": 0.1527, | |
| "step": 2904 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "learning_rate": 1.4221345514950167e-05, | |
| "loss": 0.1381, | |
| "step": 3036 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "learning_rate": 1.3399086378737542e-05, | |
| "loss": 0.1516, | |
| "step": 3168 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "learning_rate": 1.2576827242524916e-05, | |
| "loss": 0.1508, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 1.1754568106312293e-05, | |
| "loss": 0.1479, | |
| "step": 3432 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "learning_rate": 1.093230897009967e-05, | |
| "loss": 0.1341, | |
| "step": 3564 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "learning_rate": 1.0110049833887042e-05, | |
| "loss": 0.1474, | |
| "step": 3696 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "learning_rate": 9.287790697674419e-06, | |
| "loss": 0.1462, | |
| "step": 3828 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "learning_rate": 8.465531561461795e-06, | |
| "loss": 0.1444, | |
| "step": 3960 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8818666666666667, | |
| "eval_f1": 0.8856774193548387, | |
| "eval_loss": 0.37261688709259033, | |
| "eval_precision": 0.8750637429882713, | |
| "eval_recall": 0.896551724137931, | |
| "eval_runtime": 107.1711, | |
| "eval_samples_per_second": 34.991, | |
| "step": 3987 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "learning_rate": 7.64327242524917e-06, | |
| "loss": 0.0754, | |
| "step": 4092 | |
| }, | |
| { | |
| "epoch": 3.18, | |
| "learning_rate": 6.821013289036545e-06, | |
| "loss": 0.0811, | |
| "step": 4224 | |
| }, | |
| { | |
| "epoch": 3.28, | |
| "learning_rate": 5.9987541528239206e-06, | |
| "loss": 0.0808, | |
| "step": 4356 | |
| }, | |
| { | |
| "epoch": 3.38, | |
| "learning_rate": 5.176495016611296e-06, | |
| "loss": 0.0752, | |
| "step": 4488 | |
| }, | |
| { | |
| "epoch": 3.48, | |
| "learning_rate": 4.354235880398672e-06, | |
| "loss": 0.0803, | |
| "step": 4620 | |
| }, | |
| { | |
| "epoch": 3.58, | |
| "learning_rate": 3.5319767441860467e-06, | |
| "loss": 0.0734, | |
| "step": 4752 | |
| }, | |
| { | |
| "epoch": 3.67, | |
| "learning_rate": 2.7097176079734222e-06, | |
| "loss": 0.0615, | |
| "step": 4884 | |
| }, | |
| { | |
| "epoch": 3.77, | |
| "learning_rate": 1.8874584717607975e-06, | |
| "loss": 0.0605, | |
| "step": 5016 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "learning_rate": 1.0651993355481726e-06, | |
| "loss": 0.0781, | |
| "step": 5148 | |
| }, | |
| { | |
| "epoch": 3.97, | |
| "learning_rate": 2.491694352159469e-07, | |
| "loss": 0.0713, | |
| "step": 5280 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8818666666666667, | |
| "eval_f1": 0.8882723833543505, | |
| "eval_loss": 0.5432285666465759, | |
| "eval_precision": 0.8586055582642613, | |
| "eval_recall": 0.9200626959247649, | |
| "eval_runtime": 107.1677, | |
| "eval_samples_per_second": 34.992, | |
| "step": 5316 | |
| } | |
| ], | |
| "max_steps": 5316, | |
| "num_train_epochs": 4, | |
| "total_flos": 8223951552000000.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |