| { | |
| "best_metric": 0.6091980338096619, | |
| "best_model_checkpoint": "bert_uncased_L-4_H-128_A-2_cola/checkpoint-204", | |
| "epoch": 11.0, | |
| "eval_steps": 500, | |
| "global_step": 374, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.7114630937576294, | |
| "learning_rate": 4.9e-05, | |
| "loss": 0.6362, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6912751793861389, | |
| "eval_loss": 0.6191275119781494, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 0.368, | |
| "eval_samples_per_second": 2834.557, | |
| "eval_steps_per_second": 13.588, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.6178502440452576, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.608, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6912751793861389, | |
| "eval_loss": 0.6190881729125977, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 0.359, | |
| "eval_samples_per_second": 2905.511, | |
| "eval_steps_per_second": 13.929, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.42143625020980835, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.607, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6912751793861389, | |
| "eval_loss": 0.616786777973175, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 0.3564, | |
| "eval_samples_per_second": 2926.415, | |
| "eval_steps_per_second": 14.029, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.5581986904144287, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.6055, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.6912751793861389, | |
| "eval_loss": 0.6145115494728088, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 0.3604, | |
| "eval_samples_per_second": 2894.371, | |
| "eval_steps_per_second": 13.875, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.6049013137817383, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.6009, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.6912751793861389, | |
| "eval_loss": 0.6107176542282104, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 0.3593, | |
| "eval_samples_per_second": 2902.792, | |
| "eval_steps_per_second": 13.916, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 0.7587459087371826, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.5939, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.6912751793861389, | |
| "eval_loss": 0.6091980338096619, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 0.3601, | |
| "eval_samples_per_second": 2896.749, | |
| "eval_steps_per_second": 13.887, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 1.9767988920211792, | |
| "learning_rate": 4.3e-05, | |
| "loss": 0.5799, | |
| "step": 238 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.6951102614402771, | |
| "eval_loss": 0.616756796836853, | |
| "eval_matthews_correlation": 0.0854532539112171, | |
| "eval_runtime": 0.3674, | |
| "eval_samples_per_second": 2838.795, | |
| "eval_steps_per_second": 13.609, | |
| "step": 238 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 1.1349046230316162, | |
| "learning_rate": 4.2e-05, | |
| "loss": 0.5679, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.6912751793861389, | |
| "eval_loss": 0.6161894202232361, | |
| "eval_matthews_correlation": 0.08481153787433816, | |
| "eval_runtime": 0.364, | |
| "eval_samples_per_second": 2865.523, | |
| "eval_steps_per_second": 13.737, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 1.366595983505249, | |
| "learning_rate": 4.1e-05, | |
| "loss": 0.5553, | |
| "step": 306 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.6855225563049316, | |
| "eval_loss": 0.623590886592865, | |
| "eval_matthews_correlation": 0.06376490253157786, | |
| "eval_runtime": 0.3742, | |
| "eval_samples_per_second": 2787.105, | |
| "eval_steps_per_second": 13.361, | |
| "step": 306 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 1.5425918102264404, | |
| "learning_rate": 4e-05, | |
| "loss": 0.5361, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.6586769223213196, | |
| "eval_loss": 0.6316266655921936, | |
| "eval_matthews_correlation": 0.08369053019265868, | |
| "eval_runtime": 0.3592, | |
| "eval_samples_per_second": 2903.742, | |
| "eval_steps_per_second": 13.92, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 2.3495326042175293, | |
| "learning_rate": 3.9000000000000006e-05, | |
| "loss": 0.5249, | |
| "step": 374 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.6548417806625366, | |
| "eval_loss": 0.638280987739563, | |
| "eval_matthews_correlation": 0.10313429660779676, | |
| "eval_runtime": 0.3611, | |
| "eval_samples_per_second": 2888.509, | |
| "eval_steps_per_second": 13.847, | |
| "step": 374 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "step": 374, | |
| "total_flos": 117043404217344.0, | |
| "train_loss": 0.5832455527973684, | |
| "train_runtime": 45.3448, | |
| "train_samples_per_second": 9428.866, | |
| "train_steps_per_second": 37.491 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 1700, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 50, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 5, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 5 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 117043404217344.0, | |
| "train_batch_size": 256, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |