{ "best_metric": 0.06343186646699905, "best_model_checkpoint": "./models/roberta-base_1656662418.0944197/checkpoint-9400", "epoch": 3.0, "global_step": 14100, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.1, "learning_rate": 1.9334751773049647e-05, "loss": 0.3035, "step": 469 }, { "epoch": 0.2, "learning_rate": 1.866950354609929e-05, "loss": 0.1097, "step": 938 }, { "epoch": 0.3, "learning_rate": 1.8004255319148936e-05, "loss": 0.0856, "step": 1407 }, { "epoch": 0.4, "learning_rate": 1.7339007092198585e-05, "loss": 0.0825, "step": 1876 }, { "epoch": 0.5, "learning_rate": 1.667375886524823e-05, "loss": 0.078, "step": 2345 }, { "epoch": 0.6, "learning_rate": 1.6008510638297874e-05, "loss": 0.0681, "step": 2814 }, { "epoch": 0.7, "learning_rate": 1.534326241134752e-05, "loss": 0.0714, "step": 3283 }, { "epoch": 0.8, "learning_rate": 1.4678014184397164e-05, "loss": 0.0615, "step": 3752 }, { "epoch": 0.9, "learning_rate": 1.401276595744681e-05, "loss": 0.062, "step": 4221 }, { "epoch": 1.0, "learning_rate": 1.3347517730496456e-05, "loss": 0.0586, "step": 4690 }, { "epoch": 1.0, "eval_accuracy": 0.9821204276472912, "eval_f1": 0.8757774857643451, "eval_loss": 0.0644184947013855, "eval_precision": 0.8638209625853279, "eval_recall": 0.8880696455538776, "eval_runtime": 29.7187, "eval_samples_per_second": 318.957, "eval_steps_per_second": 19.954, "step": 4700 }, { "epoch": 1.1, "learning_rate": 1.26822695035461e-05, "loss": 0.0491, "step": 5159 }, { "epoch": 1.2, "learning_rate": 1.2017021276595745e-05, "loss": 0.0498, "step": 5628 }, { "epoch": 1.3, "learning_rate": 1.1351773049645392e-05, "loss": 0.048, "step": 6097 }, { "epoch": 1.4, "learning_rate": 1.0686524822695037e-05, "loss": 0.0474, "step": 6566 }, { "epoch": 1.5, "learning_rate": 1.0021276595744682e-05, "loss": 0.0477, "step": 7035 }, { "epoch": 1.6, "learning_rate": 9.356028368794327e-06, "loss": 0.0441, "step": 7504 }, { "epoch": 1.7, "learning_rate": 8.690780141843971e-06, "loss": 0.0421, "step": 7973 }, { "epoch": 1.8, "learning_rate": 8.025531914893618e-06, "loss": 0.0434, "step": 8442 }, { "epoch": 1.9, "learning_rate": 7.360283687943263e-06, "loss": 0.0409, "step": 8911 }, { "epoch": 2.0, "learning_rate": 6.695035460992908e-06, "loss": 0.045, "step": 9380 }, { "epoch": 2.0, "eval_accuracy": 0.9835062124437578, "eval_f1": 0.890757790368272, "eval_loss": 0.06343186646699905, "eval_precision": 0.8876929863255404, "eval_recall": 0.8938438305054632, "eval_runtime": 27.687, "eval_samples_per_second": 342.363, "eval_steps_per_second": 21.418, "step": 9400 }, { "epoch": 2.1, "learning_rate": 6.029787234042554e-06, "loss": 0.0327, "step": 9849 }, { "epoch": 2.2, "learning_rate": 5.3645390070921985e-06, "loss": 0.0298, "step": 10318 }, { "epoch": 2.3, "learning_rate": 4.699290780141844e-06, "loss": 0.0322, "step": 10787 }, { "epoch": 2.39, "learning_rate": 4.03404255319149e-06, "loss": 0.0338, "step": 11256 }, { "epoch": 2.49, "learning_rate": 3.368794326241135e-06, "loss": 0.0299, "step": 11725 }, { "epoch": 2.59, "learning_rate": 2.70354609929078e-06, "loss": 0.0292, "step": 12194 }, { "epoch": 2.69, "learning_rate": 2.0382978723404254e-06, "loss": 0.0345, "step": 12663 }, { "epoch": 2.79, "learning_rate": 1.373049645390071e-06, "loss": 0.0303, "step": 13132 }, { "epoch": 2.89, "learning_rate": 7.078014184397164e-07, "loss": 0.0311, "step": 13601 }, { "epoch": 2.99, "learning_rate": 4.2553191489361707e-08, "loss": 0.032, "step": 14070 }, { "epoch": 3.0, "eval_accuracy": 0.9842020533202814, "eval_f1": 0.8978192050650721, "eval_loss": 0.06502678245306015, "eval_precision": 0.888830852267781, "eval_recall": 0.9069912054721506, "eval_runtime": 27.9554, "eval_samples_per_second": 339.076, "eval_steps_per_second": 21.212, "step": 14100 } ], "max_steps": 14100, "num_train_epochs": 3, "total_flos": 6095455828493898.0, "trial_name": null, "trial_params": null }