| { | |
| "best_metric": 1.1420139074325562, | |
| "best_model_checkpoint": "mobilebert_sa_GLUE_Experiment_logit_kd_data_aug_qnli/checkpoint-33208", | |
| "epoch": 6.0, | |
| "global_step": 199248, | |
| "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": 0.6899, | |
| "step": 33208 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5923485264506682, | |
| "eval_loss": 1.1420139074325562, | |
| "eval_runtime": 12.6193, | |
| "eval_samples_per_second": 432.909, | |
| "eval_steps_per_second": 3.407, | |
| "step": 33208 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.498, | |
| "step": 66416 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5943620721215449, | |
| "eval_loss": 1.219602346420288, | |
| "eval_runtime": 12.6869, | |
| "eval_samples_per_second": 430.602, | |
| "eval_steps_per_second": 3.389, | |
| "step": 66416 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.4209, | |
| "step": 99624 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.5976569650375252, | |
| "eval_loss": 1.2370126247406006, | |
| "eval_runtime": 12.7229, | |
| "eval_samples_per_second": 429.383, | |
| "eval_steps_per_second": 3.38, | |
| "step": 99624 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.3746, | |
| "step": 132832 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5972908658246385, | |
| "eval_loss": 1.2783989906311035, | |
| "eval_runtime": 12.6633, | |
| "eval_samples_per_second": 431.405, | |
| "eval_steps_per_second": 3.396, | |
| "step": 132832 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.3449, | |
| "step": 166040 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.5938129233022149, | |
| "eval_loss": 1.2648983001708984, | |
| "eval_runtime": 12.6182, | |
| "eval_samples_per_second": 432.944, | |
| "eval_steps_per_second": 3.408, | |
| "step": 166040 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.3238, | |
| "step": 199248 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.6113856855207761, | |
| "eval_loss": 1.1661865711212158, | |
| "eval_runtime": 12.5606, | |
| "eval_samples_per_second": 434.933, | |
| "eval_steps_per_second": 3.423, | |
| "step": 199248 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "step": 199248, | |
| "total_flos": 7.996297956242227e+17, | |
| "train_loss": 0.44199674663453337, | |
| "train_runtime": 148025.2261, | |
| "train_samples_per_second": 1435.739, | |
| "train_steps_per_second": 11.217 | |
| } | |
| ], | |
| "max_steps": 1660400, | |
| "num_train_epochs": 50, | |
| "total_flos": 7.996297956242227e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |