{ "best_metric": 0.8615721366809934, "best_model_checkpoint": "./output/relation_extraction_bert_base_uncased/checkpoint-963", "epoch": 10.0, "eval_steps": 500, "global_step": 1070, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 1.0, "learning_rate": 4.5e-05, "loss": 1.2511, "step": 107 }, { "epoch": 1.0, "eval_f1": 0.7907449232122132, "eval_loss": 0.6098188161849976, "eval_runtime": 2.957, "eval_samples_per_second": 405.813, "eval_steps_per_second": 6.425, "step": 107 }, { "epoch": 2.0, "learning_rate": 4e-05, "loss": 0.437, "step": 214 }, { "epoch": 2.0, "eval_f1": 0.8452978511771352, "eval_loss": 0.5166792869567871, "eval_runtime": 2.8484, "eval_samples_per_second": 421.296, "eval_steps_per_second": 6.671, "step": 214 }, { "epoch": 3.0, "learning_rate": 3.5e-05, "loss": 0.1921, "step": 321 }, { "epoch": 3.0, "eval_f1": 0.8552983432267793, "eval_loss": 0.5445510745048523, "eval_runtime": 2.9775, "eval_samples_per_second": 403.017, "eval_steps_per_second": 6.381, "step": 321 }, { "epoch": 4.0, "learning_rate": 3e-05, "loss": 0.0795, "step": 428 }, { "epoch": 4.0, "eval_f1": 0.8526607872188615, "eval_loss": 0.7041279077529907, "eval_runtime": 2.8817, "eval_samples_per_second": 416.422, "eval_steps_per_second": 6.593, "step": 428 }, { "epoch": 5.0, "learning_rate": 2.5e-05, "loss": 0.0334, "step": 535 }, { "epoch": 5.0, "eval_f1": 0.8567909830709438, "eval_loss": 0.7774203419685364, "eval_runtime": 2.9813, "eval_samples_per_second": 402.511, "eval_steps_per_second": 6.373, "step": 535 }, { "epoch": 6.0, "learning_rate": 2e-05, "loss": 0.0124, "step": 642 }, { "epoch": 6.0, "eval_f1": 0.8490240342835185, "eval_loss": 0.808628261089325, "eval_runtime": 2.6913, "eval_samples_per_second": 445.878, "eval_steps_per_second": 7.06, "step": 642 }, { "epoch": 7.0, "learning_rate": 1.5e-05, "loss": 0.0056, "step": 749 }, { "epoch": 7.0, "eval_f1": 0.860117487560933, "eval_loss": 0.8614501953125, "eval_runtime": 2.7072, "eval_samples_per_second": 443.256, "eval_steps_per_second": 7.018, "step": 749 }, { "epoch": 8.0, "learning_rate": 1e-05, "loss": 0.003, "step": 856 }, { "epoch": 8.0, "eval_f1": 0.8613618380674043, "eval_loss": 0.9245680570602417, "eval_runtime": 2.7083, "eval_samples_per_second": 443.078, "eval_steps_per_second": 7.015, "step": 856 }, { "epoch": 9.0, "learning_rate": 5e-06, "loss": 0.0014, "step": 963 }, { "epoch": 9.0, "eval_f1": 0.8615721366809934, "eval_loss": 0.9300327301025391, "eval_runtime": 2.9509, "eval_samples_per_second": 406.653, "eval_steps_per_second": 6.439, "step": 963 }, { "epoch": 10.0, "learning_rate": 0.0, "loss": 0.0011, "step": 1070 }, { "epoch": 10.0, "eval_f1": 0.8613345463465772, "eval_loss": 0.9421382546424866, "eval_runtime": 2.9527, "eval_samples_per_second": 406.409, "eval_steps_per_second": 6.435, "step": 1070 }, { "epoch": 10.0, "step": 1070, "total_flos": 4443650073600000.0, "train_loss": 0.2016566672197012, "train_runtime": 574.5493, "train_samples_per_second": 118.354, "train_steps_per_second": 1.862 } ], "logging_steps": 500, "max_steps": 1070, "num_train_epochs": 10, "save_steps": 500, "total_flos": 4443650073600000.0, "trial_name": null, "trial_params": null }