| { | |
| "best_metric": 0.6706946492195129, | |
| "best_model_checkpoint": "mobilebert_add_GLUE_Experiment_logit_kd_cola_256/checkpoint-737", | |
| "epoch": 16.0, | |
| "global_step": 1072, | |
| "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": 0.8133, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.6864015460014343, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 2.3673, | |
| "eval_samples_per_second": 440.581, | |
| "eval_steps_per_second": 3.802, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.797, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.6865732669830322, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 2.3603, | |
| "eval_samples_per_second": 441.891, | |
| "eval_steps_per_second": 3.813, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.7964, | |
| "step": 201 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.6832323670387268, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 2.3475, | |
| "eval_samples_per_second": 444.297, | |
| "eval_steps_per_second": 3.834, | |
| "step": 201 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.7948, | |
| "step": 268 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.6826822757720947, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 2.3515, | |
| "eval_samples_per_second": 443.542, | |
| "eval_steps_per_second": 3.827, | |
| "step": 268 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.7934, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.6786946058273315, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 2.3636, | |
| "eval_samples_per_second": 441.284, | |
| "eval_steps_per_second": 3.808, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.7759, | |
| "step": 402 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 0.674231231212616, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 2.4966, | |
| "eval_samples_per_second": 417.768, | |
| "eval_steps_per_second": 3.605, | |
| "step": 402 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 4.3e-05, | |
| "loss": 0.761, | |
| "step": 469 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 0.6777247786521912, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 2.3633, | |
| "eval_samples_per_second": 441.324, | |
| "eval_steps_per_second": 3.808, | |
| "step": 469 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 4.2e-05, | |
| "loss": 0.756, | |
| "step": 536 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.6754477024078369, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 2.3686, | |
| "eval_samples_per_second": 440.338, | |
| "eval_steps_per_second": 3.8, | |
| "step": 536 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 4.1e-05, | |
| "loss": 0.7471, | |
| "step": 603 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 0.6772245764732361, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 2.3582, | |
| "eval_samples_per_second": 442.285, | |
| "eval_steps_per_second": 3.816, | |
| "step": 603 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 4e-05, | |
| "loss": 0.7457, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 0.6760045289993286, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 2.3622, | |
| "eval_samples_per_second": 441.534, | |
| "eval_steps_per_second": 3.81, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "learning_rate": 3.9000000000000006e-05, | |
| "loss": 0.7419, | |
| "step": 737 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_loss": 0.6706946492195129, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 2.3314, | |
| "eval_samples_per_second": 447.378, | |
| "eval_steps_per_second": 3.86, | |
| "step": 737 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 3.8e-05, | |
| "loss": 0.741, | |
| "step": 804 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 0.6729392409324646, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 2.3775, | |
| "eval_samples_per_second": 438.704, | |
| "eval_steps_per_second": 3.786, | |
| "step": 804 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "learning_rate": 3.7e-05, | |
| "loss": 0.7351, | |
| "step": 871 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_loss": 0.6835747957229614, | |
| "eval_matthews_correlation": 0.07577320814453954, | |
| "eval_runtime": 2.3704, | |
| "eval_samples_per_second": 440.011, | |
| "eval_steps_per_second": 3.797, | |
| "step": 871 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "learning_rate": 3.6e-05, | |
| "loss": 0.7349, | |
| "step": 938 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 0.6768582463264465, | |
| "eval_matthews_correlation": 0.07951431611081063, | |
| "eval_runtime": 2.359, | |
| "eval_samples_per_second": 442.143, | |
| "eval_steps_per_second": 3.815, | |
| "step": 938 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.7357, | |
| "step": 1005 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_loss": 0.6715154051780701, | |
| "eval_matthews_correlation": 0.03149441738954873, | |
| "eval_runtime": 2.3487, | |
| "eval_samples_per_second": 444.084, | |
| "eval_steps_per_second": 3.832, | |
| "step": 1005 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "learning_rate": 3.4000000000000007e-05, | |
| "loss": 0.7333, | |
| "step": 1072 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 0.681270956993103, | |
| "eval_matthews_correlation": 0.08941665373051183, | |
| "eval_runtime": 2.3503, | |
| "eval_samples_per_second": 443.777, | |
| "eval_steps_per_second": 3.829, | |
| "step": 1072 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "step": 1072, | |
| "total_flos": 3526461549969408.0, | |
| "train_loss": 0.7626726947613617, | |
| "train_runtime": 1027.7392, | |
| "train_samples_per_second": 416.01, | |
| "train_steps_per_second": 3.26 | |
| } | |
| ], | |
| "max_steps": 3350, | |
| "num_train_epochs": 50, | |
| "total_flos": 3526461549969408.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |