| { | |
| "best_metric": 0.5874866247177124, | |
| "best_model_checkpoint": "bert_tiny_lda_100_v1_book_cola/checkpoint-102", | |
| "epoch": 8.0, | |
| "eval_steps": 500, | |
| "global_step": 272, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.9012271165847778, | |
| "learning_rate": 4.9e-05, | |
| "loss": 0.6112, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6912751793861389, | |
| "eval_loss": 0.6111970543861389, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 0.3494, | |
| "eval_samples_per_second": 2984.868, | |
| "eval_steps_per_second": 14.309, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.6870914697647095, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.5906, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6951102614402771, | |
| "eval_loss": 0.5991763472557068, | |
| "eval_matthews_correlation": 0.08436523968200914, | |
| "eval_runtime": 0.3261, | |
| "eval_samples_per_second": 3198.114, | |
| "eval_steps_per_second": 15.331, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 1.8232595920562744, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.5465, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.693192720413208, | |
| "eval_loss": 0.5874866247177124, | |
| "eval_matthews_correlation": 0.19955176173637962, | |
| "eval_runtime": 0.3617, | |
| "eval_samples_per_second": 2883.62, | |
| "eval_steps_per_second": 13.824, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 2.4694697856903076, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.4903, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.6912751793861389, | |
| "eval_loss": 0.5886690616607666, | |
| "eval_matthews_correlation": 0.19917446639499126, | |
| "eval_runtime": 0.3515, | |
| "eval_samples_per_second": 2967.313, | |
| "eval_steps_per_second": 14.225, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 3.748969316482544, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.4275, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.7094918489456177, | |
| "eval_loss": 0.6417186260223389, | |
| "eval_matthews_correlation": 0.217400289930018, | |
| "eval_runtime": 0.3243, | |
| "eval_samples_per_second": 3215.81, | |
| "eval_steps_per_second": 15.416, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 4.171329975128174, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.3786, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.7209970951080322, | |
| "eval_loss": 0.6495499014854431, | |
| "eval_matthews_correlation": 0.27111800500621025, | |
| "eval_runtime": 0.3294, | |
| "eval_samples_per_second": 3166.651, | |
| "eval_steps_per_second": 15.18, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 4.947795867919922, | |
| "learning_rate": 4.3e-05, | |
| "loss": 0.3295, | |
| "step": 238 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.7027804255485535, | |
| "eval_loss": 0.6800650358200073, | |
| "eval_matthews_correlation": 0.24007288633988105, | |
| "eval_runtime": 0.336, | |
| "eval_samples_per_second": 3104.623, | |
| "eval_steps_per_second": 14.883, | |
| "step": 238 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 7.731811046600342, | |
| "learning_rate": 4.2e-05, | |
| "loss": 0.2891, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.715244472026825, | |
| "eval_loss": 0.7395167946815491, | |
| "eval_matthews_correlation": 0.27006989464867953, | |
| "eval_runtime": 0.3343, | |
| "eval_samples_per_second": 3120.119, | |
| "eval_steps_per_second": 14.957, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "step": 272, | |
| "total_flos": 1793902531485696.0, | |
| "train_loss": 0.4579080904231352, | |
| "train_runtime": 48.0666, | |
| "train_samples_per_second": 8894.95, | |
| "train_steps_per_second": 35.368 | |
| } | |
| ], | |
| "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": 1793902531485696.0, | |
| "train_batch_size": 256, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |