| { | |
| "best_metric": 0.48669904470443726, | |
| "best_model_checkpoint": "bert_base_lda_20_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": 3.9988954067230225, | |
| "learning_rate": 4.9e-05, | |
| "loss": 0.6052, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6912751793861389, | |
| "eval_loss": 0.5938838720321655, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 0.6852, | |
| "eval_samples_per_second": 1522.289, | |
| "eval_steps_per_second": 7.298, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 4.848330497741699, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.5337, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7545541524887085, | |
| "eval_loss": 0.5131515264511108, | |
| "eval_matthews_correlation": 0.3639324656122657, | |
| "eval_runtime": 0.6836, | |
| "eval_samples_per_second": 1525.781, | |
| "eval_steps_per_second": 7.314, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 5.25543737411499, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.4113, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7890700101852417, | |
| "eval_loss": 0.48669904470443726, | |
| "eval_matthews_correlation": 0.47428224772244454, | |
| "eval_runtime": 0.7023, | |
| "eval_samples_per_second": 1485.127, | |
| "eval_steps_per_second": 7.119, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 5.055551528930664, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.2988, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7804410457611084, | |
| "eval_loss": 0.5833057761192322, | |
| "eval_matthews_correlation": 0.4456190824541546, | |
| "eval_runtime": 0.6822, | |
| "eval_samples_per_second": 1528.819, | |
| "eval_steps_per_second": 7.329, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 5.268617630004883, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.232, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.7957813739776611, | |
| "eval_loss": 0.5777825713157654, | |
| "eval_matthews_correlation": 0.4905937147239641, | |
| "eval_runtime": 0.6775, | |
| "eval_samples_per_second": 1539.424, | |
| "eval_steps_per_second": 7.38, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 5.387474060058594, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.1741, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8063278794288635, | |
| "eval_loss": 0.592420756816864, | |
| "eval_matthews_correlation": 0.5184234806320535, | |
| "eval_runtime": 0.68, | |
| "eval_samples_per_second": 1533.85, | |
| "eval_steps_per_second": 7.353, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 5.771633148193359, | |
| "learning_rate": 4.3e-05, | |
| "loss": 0.1299, | |
| "step": 238 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8015340566635132, | |
| "eval_loss": 0.7205991744995117, | |
| "eval_matthews_correlation": 0.5048506698662225, | |
| "eval_runtime": 0.6748, | |
| "eval_samples_per_second": 1545.6, | |
| "eval_steps_per_second": 7.409, | |
| "step": 238 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 5.118968486785889, | |
| "learning_rate": 4.2e-05, | |
| "loss": 0.1075, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.7986577153205872, | |
| "eval_loss": 0.6481350660324097, | |
| "eval_matthews_correlation": 0.5020753041155603, | |
| "eval_runtime": 0.6785, | |
| "eval_samples_per_second": 1537.3, | |
| "eval_steps_per_second": 7.37, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "step": 272, | |
| "total_flos": 8999450537533440.0, | |
| "train_loss": 0.31154586725375233, | |
| "train_runtime": 134.705, | |
| "train_samples_per_second": 3173.972, | |
| "train_steps_per_second": 12.62 | |
| } | |
| ], | |
| "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": 8999450537533440.0, | |
| "train_batch_size": 256, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |