|
{ |
|
"best_metric": 0.04098295420408249, |
|
"best_model_checkpoint": "doc-topic-model_eval-03_train-01/checkpoint-8000", |
|
"epoch": 3.9447731755424065, |
|
"eval_steps": 1000, |
|
"global_step": 8000, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.2465483234714004, |
|
"grad_norm": 0.32817578315734863, |
|
"learning_rate": 1.9950690335305722e-05, |
|
"loss": 0.1659, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 0.4930966469428008, |
|
"grad_norm": 0.34106990694999695, |
|
"learning_rate": 1.9901380670611442e-05, |
|
"loss": 0.0941, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 0.4930966469428008, |
|
"eval_accuracy": 0.981439120143997, |
|
"eval_f1": 0.0, |
|
"eval_loss": 0.09035909175872803, |
|
"eval_precision": 0.0, |
|
"eval_recall": 0.0, |
|
"eval_runtime": 12.1739, |
|
"eval_samples_per_second": 666.182, |
|
"eval_steps_per_second": 2.629, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 0.7396449704142012, |
|
"grad_norm": 0.4004117548465729, |
|
"learning_rate": 1.9852071005917162e-05, |
|
"loss": 0.088, |
|
"step": 1500 |
|
}, |
|
{ |
|
"epoch": 0.9861932938856016, |
|
"grad_norm": 0.37588435411453247, |
|
"learning_rate": 1.980276134122288e-05, |
|
"loss": 0.0787, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 0.9861932938856016, |
|
"eval_accuracy": 0.981439120143997, |
|
"eval_f1": 0.0, |
|
"eval_loss": 0.07049160450696945, |
|
"eval_precision": 0.0, |
|
"eval_recall": 0.0, |
|
"eval_runtime": 12.1677, |
|
"eval_samples_per_second": 666.518, |
|
"eval_steps_per_second": 2.63, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 1.232741617357002, |
|
"grad_norm": 0.43946701288223267, |
|
"learning_rate": 1.9753451676528602e-05, |
|
"loss": 0.0696, |
|
"step": 2500 |
|
}, |
|
{ |
|
"epoch": 1.4792899408284024, |
|
"grad_norm": 0.4252438545227051, |
|
"learning_rate": 1.9704142011834322e-05, |
|
"loss": 0.0628, |
|
"step": 3000 |
|
}, |
|
{ |
|
"epoch": 1.4792899408284024, |
|
"eval_accuracy": 0.9822044588063551, |
|
"eval_f1": 0.1170182841068917, |
|
"eval_loss": 0.05713700130581856, |
|
"eval_precision": 0.7402135231316725, |
|
"eval_recall": 0.06353084911423336, |
|
"eval_runtime": 14.6983, |
|
"eval_samples_per_second": 551.763, |
|
"eval_steps_per_second": 2.177, |
|
"step": 3000 |
|
}, |
|
{ |
|
"epoch": 1.725838264299803, |
|
"grad_norm": 0.3535268306732178, |
|
"learning_rate": 1.965483234714004e-05, |
|
"loss": 0.0584, |
|
"step": 3500 |
|
}, |
|
{ |
|
"epoch": 1.972386587771203, |
|
"grad_norm": 0.3976692855358124, |
|
"learning_rate": 1.9605522682445763e-05, |
|
"loss": 0.0537, |
|
"step": 4000 |
|
}, |
|
{ |
|
"epoch": 1.972386587771203, |
|
"eval_accuracy": 0.9841787490964752, |
|
"eval_f1": 0.3162859067801801, |
|
"eval_loss": 0.05010608583688736, |
|
"eval_precision": 0.7991333952336738, |
|
"eval_recall": 0.19715943799633476, |
|
"eval_runtime": 14.5719, |
|
"eval_samples_per_second": 556.549, |
|
"eval_steps_per_second": 2.196, |
|
"step": 4000 |
|
}, |
|
{ |
|
"epoch": 2.2189349112426036, |
|
"grad_norm": 0.24940212070941925, |
|
"learning_rate": 1.9556213017751483e-05, |
|
"loss": 0.0499, |
|
"step": 4500 |
|
}, |
|
{ |
|
"epoch": 2.465483234714004, |
|
"grad_norm": 0.40094295144081116, |
|
"learning_rate": 1.95069033530572e-05, |
|
"loss": 0.0478, |
|
"step": 5000 |
|
}, |
|
{ |
|
"epoch": 2.465483234714004, |
|
"eval_accuracy": 0.9851084938418584, |
|
"eval_f1": 0.42113382182799847, |
|
"eval_loss": 0.046862900257110596, |
|
"eval_precision": 0.7560830860534125, |
|
"eval_recall": 0.2918448381185095, |
|
"eval_runtime": 14.4838, |
|
"eval_samples_per_second": 559.937, |
|
"eval_steps_per_second": 2.209, |
|
"step": 5000 |
|
}, |
|
{ |
|
"epoch": 2.712031558185404, |
|
"grad_norm": 0.42049625515937805, |
|
"learning_rate": 1.9457593688362923e-05, |
|
"loss": 0.0463, |
|
"step": 5500 |
|
}, |
|
{ |
|
"epoch": 2.9585798816568047, |
|
"grad_norm": 0.41673073172569275, |
|
"learning_rate": 1.940828402366864e-05, |
|
"loss": 0.0453, |
|
"step": 6000 |
|
}, |
|
{ |
|
"epoch": 2.9585798816568047, |
|
"eval_accuracy": 0.985744858766671, |
|
"eval_f1": 0.4941152801529021, |
|
"eval_loss": 0.044261422008275986, |
|
"eval_precision": 0.7238432066018273, |
|
"eval_recall": 0.3750763591936469, |
|
"eval_runtime": 14.4397, |
|
"eval_samples_per_second": 561.645, |
|
"eval_steps_per_second": 2.216, |
|
"step": 6000 |
|
}, |
|
{ |
|
"epoch": 3.2051282051282053, |
|
"grad_norm": 0.3240315914154053, |
|
"learning_rate": 1.935897435897436e-05, |
|
"loss": 0.0424, |
|
"step": 6500 |
|
}, |
|
{ |
|
"epoch": 3.4516765285996054, |
|
"grad_norm": 0.39667147397994995, |
|
"learning_rate": 1.930966469428008e-05, |
|
"loss": 0.0389, |
|
"step": 7000 |
|
}, |
|
{ |
|
"epoch": 3.4516765285996054, |
|
"eval_accuracy": 0.9863174454696203, |
|
"eval_f1": 0.5358653846153846, |
|
"eval_loss": 0.04172312840819359, |
|
"eval_precision": 0.7233904465212876, |
|
"eval_recall": 0.42554978619425776, |
|
"eval_runtime": 14.5013, |
|
"eval_samples_per_second": 559.261, |
|
"eval_steps_per_second": 2.207, |
|
"step": 7000 |
|
}, |
|
{ |
|
"epoch": 3.698224852071006, |
|
"grad_norm": 0.39966270327568054, |
|
"learning_rate": 1.92603550295858e-05, |
|
"loss": 0.0403, |
|
"step": 7500 |
|
}, |
|
{ |
|
"epoch": 3.9447731755424065, |
|
"grad_norm": 0.42601555585861206, |
|
"learning_rate": 1.921104536489152e-05, |
|
"loss": 0.0393, |
|
"step": 8000 |
|
}, |
|
{ |
|
"epoch": 3.9447731755424065, |
|
"eval_accuracy": 0.9861601258556911, |
|
"eval_f1": 0.5412262156448203, |
|
"eval_loss": 0.04098295420408249, |
|
"eval_precision": 0.7033825863963854, |
|
"eval_recall": 0.4398289554062309, |
|
"eval_runtime": 14.4597, |
|
"eval_samples_per_second": 560.87, |
|
"eval_steps_per_second": 2.213, |
|
"step": 8000 |
|
} |
|
], |
|
"logging_steps": 500, |
|
"max_steps": 202800, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 100, |
|
"save_steps": 1000, |
|
"stateful_callbacks": { |
|
"EarlyStoppingCallback": { |
|
"args": { |
|
"early_stopping_patience": 5, |
|
"early_stopping_threshold": 0.0 |
|
}, |
|
"attributes": { |
|
"early_stopping_patience_counter": 0 |
|
} |
|
}, |
|
"TrainerControl": { |
|
"args": { |
|
"should_epoch_stop": false, |
|
"should_evaluate": false, |
|
"should_log": false, |
|
"should_save": true, |
|
"should_training_stop": false |
|
}, |
|
"attributes": {} |
|
} |
|
}, |
|
"total_flos": 181076248886436.0, |
|
"train_batch_size": 4, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|