| { | |
| "best_metric": 0.652438759803772, | |
| "best_model_checkpoint": "mobilebert_sa_GLUE_Experiment_qnli/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": 0.6824, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5996705107084019, | |
| "eval_loss": 0.6577625870704651, | |
| "eval_runtime": 5.6422, | |
| "eval_samples_per_second": 968.24, | |
| "eval_steps_per_second": 3.899, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.6441, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6086399414241259, | |
| "eval_loss": 0.652438759803772, | |
| "eval_runtime": 5.5355, | |
| "eval_samples_per_second": 986.904, | |
| "eval_steps_per_second": 3.974, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.6202, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6071755445725792, | |
| "eval_loss": 0.6553815603256226, | |
| "eval_runtime": 5.4718, | |
| "eval_samples_per_second": 998.389, | |
| "eval_steps_per_second": 4.021, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.6009, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.6051619989017024, | |
| "eval_loss": 0.6619159579277039, | |
| "eval_runtime": 5.4488, | |
| "eval_samples_per_second": 1002.613, | |
| "eval_steps_per_second": 4.038, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.587, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.598572213069742, | |
| "eval_loss": 0.668364405632019, | |
| "eval_runtime": 5.5979, | |
| "eval_samples_per_second": 975.905, | |
| "eval_steps_per_second": 3.93, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 4.400243902439024e-05, | |
| "loss": 0.5755, | |
| "step": 2460 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.5978400146439685, | |
| "eval_loss": 0.680844783782959, | |
| "eval_runtime": 5.501, | |
| "eval_samples_per_second": 993.091, | |
| "eval_steps_per_second": 3.999, | |
| "step": 2460 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 4.3002439024390246e-05, | |
| "loss": 0.5671, | |
| "step": 2870 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.5844773933736043, | |
| "eval_loss": 0.7068149447441101, | |
| "eval_runtime": 5.4741, | |
| "eval_samples_per_second": 997.977, | |
| "eval_steps_per_second": 4.019, | |
| "step": 2870 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "step": 2870, | |
| "total_flos": 2.328477811133645e+16, | |
| "train_loss": 0.6110314930772948, | |
| "train_runtime": 2217.6487, | |
| "train_samples_per_second": 2361.578, | |
| "train_steps_per_second": 9.244 | |
| } | |
| ], | |
| "max_steps": 20500, | |
| "num_train_epochs": 50, | |
| "total_flos": 2.328477811133645e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |