| { | |
| "best_metric": 1.3502187728881836, | |
| "best_model_checkpoint": "mobilebert_sa_GLUE_Experiment_logit_kd_data_aug_stsb_256/checkpoint-2518", | |
| "epoch": 6.0, | |
| "global_step": 15108, | |
| "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": 0.535, | |
| "step": 2518 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_combined_score": 0.1688901374764955, | |
| "eval_loss": 1.3502187728881836, | |
| "eval_pearson": 0.1594880381910939, | |
| "eval_runtime": 2.7246, | |
| "eval_samples_per_second": 550.531, | |
| "eval_spearmanr": 0.1782922367618971, | |
| "eval_steps_per_second": 4.404, | |
| "step": 2518 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.3149, | |
| "step": 5036 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_combined_score": 0.1556657918320208, | |
| "eval_loss": 1.4771482944488525, | |
| "eval_pearson": 0.15037668447419109, | |
| "eval_runtime": 2.7178, | |
| "eval_samples_per_second": 551.91, | |
| "eval_spearmanr": 0.16095489918985054, | |
| "eval_steps_per_second": 4.415, | |
| "step": 5036 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.2467, | |
| "step": 7554 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_combined_score": 0.163447899818954, | |
| "eval_loss": 1.5606096982955933, | |
| "eval_pearson": 0.16090805712166092, | |
| "eval_runtime": 2.7862, | |
| "eval_samples_per_second": 538.364, | |
| "eval_spearmanr": 0.16598774251624707, | |
| "eval_steps_per_second": 4.307, | |
| "step": 7554 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.1704, | |
| "step": 10072 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_combined_score": 0.17142373626247578, | |
| "eval_loss": 1.5334404706954956, | |
| "eval_pearson": 0.16623579741955485, | |
| "eval_runtime": 2.7219, | |
| "eval_samples_per_second": 551.087, | |
| "eval_spearmanr": 0.17661167510539671, | |
| "eval_steps_per_second": 4.409, | |
| "step": 10072 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.1257, | |
| "step": 12590 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_combined_score": 0.1747127018367471, | |
| "eval_loss": 1.4389137029647827, | |
| "eval_pearson": 0.17116423098994665, | |
| "eval_runtime": 2.7134, | |
| "eval_samples_per_second": 552.805, | |
| "eval_spearmanr": 0.17826117268354752, | |
| "eval_steps_per_second": 4.422, | |
| "step": 12590 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.0976, | |
| "step": 15108 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_combined_score": 0.1481536553679076, | |
| "eval_loss": 1.5826839208602905, | |
| "eval_pearson": 0.1437432963327039, | |
| "eval_runtime": 2.7241, | |
| "eval_samples_per_second": 550.637, | |
| "eval_spearmanr": 0.15256401440311126, | |
| "eval_steps_per_second": 4.405, | |
| "step": 15108 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "step": 15108, | |
| "total_flos": 5.092937537735885e+16, | |
| "train_loss": 0.24839666365062177, | |
| "train_runtime": 8288.774, | |
| "train_samples_per_second": 1943.876, | |
| "train_steps_per_second": 15.189 | |
| } | |
| ], | |
| "max_steps": 125900, | |
| "num_train_epochs": 50, | |
| "total_flos": 5.092937537735885e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |