| { | |
| "best_metric": 5.569791028392501e-05, | |
| "best_model_checkpoint": "hBERTv1_data_aug_mrpc/checkpoint-4900", | |
| "epoch": 10.0, | |
| "global_step": 9800, | |
| "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": 0.1151, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9975490196078431, | |
| "eval_combined_score": 0.9978800554210951, | |
| "eval_f1": 0.998211091234347, | |
| "eval_loss": 0.004494799301028252, | |
| "eval_runtime": 0.5596, | |
| "eval_samples_per_second": 729.142, | |
| "eval_steps_per_second": 3.574, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.0108, | |
| "step": 1960 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 1.0, | |
| "eval_combined_score": 1.0, | |
| "eval_f1": 1.0, | |
| "eval_loss": 6.0925009165657684e-05, | |
| "eval_runtime": 0.5584, | |
| "eval_samples_per_second": 730.667, | |
| "eval_steps_per_second": 3.582, | |
| "step": 1960 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.0063, | |
| "step": 2940 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 1.0, | |
| "eval_combined_score": 1.0, | |
| "eval_f1": 1.0, | |
| "eval_loss": 8.451967732980847e-05, | |
| "eval_runtime": 0.5589, | |
| "eval_samples_per_second": 730.056, | |
| "eval_steps_per_second": 3.579, | |
| "step": 2940 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.0054, | |
| "step": 3920 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 1.0, | |
| "eval_combined_score": 1.0, | |
| "eval_f1": 1.0, | |
| "eval_loss": 5.947695899521932e-05, | |
| "eval_runtime": 0.5569, | |
| "eval_samples_per_second": 732.679, | |
| "eval_steps_per_second": 3.592, | |
| "step": 3920 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.004, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 1.0, | |
| "eval_combined_score": 1.0, | |
| "eval_f1": 1.0, | |
| "eval_loss": 5.569791028392501e-05, | |
| "eval_runtime": 0.556, | |
| "eval_samples_per_second": 733.848, | |
| "eval_steps_per_second": 3.597, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 4.4001020408163265e-05, | |
| "loss": 0.0053, | |
| "step": 5880 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 1.0, | |
| "eval_combined_score": 1.0, | |
| "eval_f1": 1.0, | |
| "eval_loss": 0.00023545935982838273, | |
| "eval_runtime": 0.5589, | |
| "eval_samples_per_second": 730.007, | |
| "eval_steps_per_second": 3.578, | |
| "step": 5880 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 4.30030612244898e-05, | |
| "loss": 0.0046, | |
| "step": 6860 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 1.0, | |
| "eval_combined_score": 1.0, | |
| "eval_f1": 1.0, | |
| "eval_loss": 0.0002702152996789664, | |
| "eval_runtime": 0.5598, | |
| "eval_samples_per_second": 728.83, | |
| "eval_steps_per_second": 3.573, | |
| "step": 6860 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 4.20030612244898e-05, | |
| "loss": 0.0116, | |
| "step": 7840 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9975490196078431, | |
| "eval_combined_score": 0.997876843735699, | |
| "eval_f1": 0.9982046678635548, | |
| "eval_loss": 0.015019365586340427, | |
| "eval_runtime": 0.558, | |
| "eval_samples_per_second": 731.241, | |
| "eval_steps_per_second": 3.585, | |
| "step": 7840 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 4.1004081632653066e-05, | |
| "loss": 0.0093, | |
| "step": 8820 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 1.0, | |
| "eval_combined_score": 1.0, | |
| "eval_f1": 1.0, | |
| "eval_loss": 0.0014859754592180252, | |
| "eval_runtime": 0.5575, | |
| "eval_samples_per_second": 731.782, | |
| "eval_steps_per_second": 3.587, | |
| "step": 8820 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 4.000408163265306e-05, | |
| "loss": 0.0123, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.9975490196078431, | |
| "eval_combined_score": 0.997876843735699, | |
| "eval_f1": 0.9982046678635548, | |
| "eval_loss": 0.016378218308091164, | |
| "eval_runtime": 0.5579, | |
| "eval_samples_per_second": 731.36, | |
| "eval_steps_per_second": 3.585, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 9800, | |
| "total_flos": 3.167072376848384e+17, | |
| "train_loss": 0.01846268490869172, | |
| "train_runtime": 9463.9723, | |
| "train_samples_per_second": 1324.687, | |
| "train_steps_per_second": 5.178 | |
| } | |
| ], | |
| "max_steps": 49000, | |
| "num_train_epochs": 50, | |
| "total_flos": 3.167072376848384e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |