| { | |
| "best_metric": 0.6744102835655212, | |
| "best_model_checkpoint": "mobilebert_add_GLUE_Experiment_logit_kd_cola/checkpoint-804", | |
| "epoch": 17.0, | |
| "global_step": 1139, | |
| "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": 0.8111, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.6859493851661682, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 2.3711, | |
| "eval_samples_per_second": 439.88, | |
| "eval_steps_per_second": 3.796, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.7968, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.6865417957305908, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 2.378, | |
| "eval_samples_per_second": 438.609, | |
| "eval_steps_per_second": 3.785, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.796, | |
| "step": 201 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.6834728717803955, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 2.3689, | |
| "eval_samples_per_second": 440.286, | |
| "eval_steps_per_second": 3.799, | |
| "step": 201 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.7938, | |
| "step": 268 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.6812810897827148, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 2.5217, | |
| "eval_samples_per_second": 413.616, | |
| "eval_steps_per_second": 3.569, | |
| "step": 268 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.7828, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.6767937541007996, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 2.3679, | |
| "eval_samples_per_second": 440.48, | |
| "eval_steps_per_second": 3.801, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.7651, | |
| "step": 402 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 0.6749773025512695, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 2.4029, | |
| "eval_samples_per_second": 434.065, | |
| "eval_steps_per_second": 3.746, | |
| "step": 402 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 4.3e-05, | |
| "loss": 0.7594, | |
| "step": 469 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 0.6960383653640747, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 2.3638, | |
| "eval_samples_per_second": 441.237, | |
| "eval_steps_per_second": 3.807, | |
| "step": 469 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 4.2e-05, | |
| "loss": 0.7592, | |
| "step": 536 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.6800056099891663, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 2.364, | |
| "eval_samples_per_second": 441.196, | |
| "eval_steps_per_second": 3.807, | |
| "step": 536 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 4.1e-05, | |
| "loss": 0.7463, | |
| "step": 603 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 0.6788876056671143, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 2.3679, | |
| "eval_samples_per_second": 440.471, | |
| "eval_steps_per_second": 3.801, | |
| "step": 603 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 4e-05, | |
| "loss": 0.7437, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 0.6794541478157043, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 2.3741, | |
| "eval_samples_per_second": 439.332, | |
| "eval_steps_per_second": 3.791, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "learning_rate": 3.9000000000000006e-05, | |
| "loss": 0.7401, | |
| "step": 737 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_loss": 0.6745404601097107, | |
| "eval_matthews_correlation": -0.007887379670285008, | |
| "eval_runtime": 2.3678, | |
| "eval_samples_per_second": 440.5, | |
| "eval_steps_per_second": 3.801, | |
| "step": 737 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 3.8e-05, | |
| "loss": 0.7398, | |
| "step": 804 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 0.6744102835655212, | |
| "eval_matthews_correlation": -0.007887379670285008, | |
| "eval_runtime": 2.3606, | |
| "eval_samples_per_second": 441.829, | |
| "eval_steps_per_second": 3.813, | |
| "step": 804 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "learning_rate": 3.7e-05, | |
| "loss": 0.7328, | |
| "step": 871 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_loss": 0.6813325881958008, | |
| "eval_matthews_correlation": 0.05873054109498616, | |
| "eval_runtime": 2.3703, | |
| "eval_samples_per_second": 440.032, | |
| "eval_steps_per_second": 3.797, | |
| "step": 871 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "learning_rate": 3.6e-05, | |
| "loss": 0.7321, | |
| "step": 938 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 0.6881392598152161, | |
| "eval_matthews_correlation": 0.07944501641685853, | |
| "eval_runtime": 2.366, | |
| "eval_samples_per_second": 440.822, | |
| "eval_steps_per_second": 3.804, | |
| "step": 938 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.7315, | |
| "step": 1005 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_loss": 0.6783654689788818, | |
| "eval_matthews_correlation": 0.061491648871478045, | |
| "eval_runtime": 2.3648, | |
| "eval_samples_per_second": 441.052, | |
| "eval_steps_per_second": 3.806, | |
| "step": 1005 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "learning_rate": 3.4000000000000007e-05, | |
| "loss": 0.7295, | |
| "step": 1072 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 0.681597888469696, | |
| "eval_matthews_correlation": 0.03846275142815186, | |
| "eval_runtime": 2.3627, | |
| "eval_samples_per_second": 441.445, | |
| "eval_steps_per_second": 3.809, | |
| "step": 1072 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "learning_rate": 3.3e-05, | |
| "loss": 0.7297, | |
| "step": 1139 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_loss": 0.698566734790802, | |
| "eval_matthews_correlation": 0.050263791388451516, | |
| "eval_runtime": 2.3861, | |
| "eval_samples_per_second": 437.111, | |
| "eval_steps_per_second": 3.772, | |
| "step": 1139 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "step": 1139, | |
| "total_flos": 4300008849408000.0, | |
| "train_loss": 0.7582107054967436, | |
| "train_runtime": 1104.2457, | |
| "train_samples_per_second": 387.187, | |
| "train_steps_per_second": 3.034 | |
| } | |
| ], | |
| "max_steps": 3350, | |
| "num_train_epochs": 50, | |
| "total_flos": 4300008849408000.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |