| { | |
| "best_metric": 0.6563981175422668, | |
| "best_model_checkpoint": "distilbert_sa_GLUE_Experiment_qnli_256/checkpoint-820", | |
| "epoch": 7.0, | |
| "global_step": 2870, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 4.9e-05, | |
| "loss": 0.679, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5938129233022149, | |
| "eval_loss": 0.6613895297050476, | |
| "eval_runtime": 3.2297, | |
| "eval_samples_per_second": 1691.499, | |
| "eval_steps_per_second": 6.812, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.6496, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6029654036243822, | |
| "eval_loss": 0.6563981175422668, | |
| "eval_runtime": 3.2407, | |
| "eval_samples_per_second": 1685.736, | |
| "eval_steps_per_second": 6.789, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.6268, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.5978400146439685, | |
| "eval_loss": 0.6634846329689026, | |
| "eval_runtime": 3.2185, | |
| "eval_samples_per_second": 1697.373, | |
| "eval_steps_per_second": 6.835, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.6055, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5932637744828848, | |
| "eval_loss": 0.6714160442352295, | |
| "eval_runtime": 3.1995, | |
| "eval_samples_per_second": 1707.477, | |
| "eval_steps_per_second": 6.876, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.5836, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.5912502288120081, | |
| "eval_loss": 0.6963688135147095, | |
| "eval_runtime": 3.1762, | |
| "eval_samples_per_second": 1720.003, | |
| "eval_steps_per_second": 6.927, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.5602, | |
| "step": 2460 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.5831960461285008, | |
| "eval_loss": 0.7318801283836365, | |
| "eval_runtime": 3.2345, | |
| "eval_samples_per_second": 1688.985, | |
| "eval_steps_per_second": 6.802, | |
| "step": 2460 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 4.3e-05, | |
| "loss": 0.5385, | |
| "step": 2870 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.5718469705290133, | |
| "eval_loss": 0.7652592658996582, | |
| "eval_runtime": 3.2105, | |
| "eval_samples_per_second": 1701.597, | |
| "eval_steps_per_second": 6.852, | |
| "step": 2870 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "step": 2870, | |
| "total_flos": 1.2511057062920192e+16, | |
| "train_loss": 0.606161908405583, | |
| "train_runtime": 917.3964, | |
| "train_samples_per_second": 5708.71, | |
| "train_steps_per_second": 22.346 | |
| } | |
| ], | |
| "max_steps": 20500, | |
| "num_train_epochs": 50, | |
| "total_flos": 1.2511057062920192e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |