| { | |
| "best_metric": 1.2698698043823242, | |
| "best_model_checkpoint": "distilbert_sa_GLUE_Experiment_data_aug_qnli/checkpoint-16604", | |
| "epoch": 6.0, | |
| "global_step": 99624, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 4.900030113225729e-05, | |
| "loss": 0.3057, | |
| "step": 16604 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5996705107084019, | |
| "eval_loss": 1.2698698043823242, | |
| "eval_runtime": 3.2256, | |
| "eval_samples_per_second": 1693.631, | |
| "eval_steps_per_second": 6.82, | |
| "step": 16604 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.800072271741749e-05, | |
| "loss": 0.0735, | |
| "step": 33208 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5952773201537617, | |
| "eval_loss": 1.778567910194397, | |
| "eval_runtime": 3.2082, | |
| "eval_samples_per_second": 1702.822, | |
| "eval_steps_per_second": 6.857, | |
| "step": 33208 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 4.700108407612624e-05, | |
| "loss": 0.0313, | |
| "step": 49812 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.5800842028189639, | |
| "eval_loss": 1.9603168964385986, | |
| "eval_runtime": 3.2082, | |
| "eval_samples_per_second": 1702.844, | |
| "eval_steps_per_second": 6.858, | |
| "step": 49812 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.600150566128644e-05, | |
| "loss": 0.0188, | |
| "step": 66416 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5927146256635548, | |
| "eval_loss": 2.2528913021087646, | |
| "eval_runtime": 3.1975, | |
| "eval_samples_per_second": 1708.529, | |
| "eval_steps_per_second": 6.88, | |
| "step": 66416 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 4.500180679354373e-05, | |
| "loss": 0.0134, | |
| "step": 83020 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.5912502288120081, | |
| "eval_loss": 2.4498074054718018, | |
| "eval_runtime": 3.1892, | |
| "eval_samples_per_second": 1712.986, | |
| "eval_steps_per_second": 6.898, | |
| "step": 83020 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 4.400228860515538e-05, | |
| "loss": 0.0106, | |
| "step": 99624 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.6031484532308256, | |
| "eval_loss": 2.51812481880188, | |
| "eval_runtime": 3.4228, | |
| "eval_samples_per_second": 1596.056, | |
| "eval_steps_per_second": 6.427, | |
| "step": 99624 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "step": 99624, | |
| "total_flos": 1.6891623964441313e+18, | |
| "train_loss": 0.07554671816640922, | |
| "train_runtime": 33064.6768, | |
| "train_samples_per_second": 6427.571, | |
| "train_steps_per_second": 25.108 | |
| } | |
| ], | |
| "max_steps": 830200, | |
| "num_train_epochs": 50, | |
| "total_flos": 1.6891623964441313e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |