{ "best_metric": 0.2404903918504715, "best_model_checkpoint": "/content/cyner_securebert/checkpoint-500", "epoch": 10.0, "eval_steps": 500, "global_step": 3520, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 1.42, "learning_rate": 1.715909090909091e-05, "loss": 0.2304, "step": 500 }, { "epoch": 1.42, "eval_accuracy": 0.9567872722607134, "eval_f1": 0.35087719298245623, "eval_loss": 0.2404903918504715, "eval_precision": 0.46709129511677283, "eval_recall": 0.280970625798212, "eval_runtime": 2.0115, "eval_samples_per_second": 403.686, "eval_steps_per_second": 50.709, "step": 500 }, { "epoch": 2.84, "learning_rate": 1.431818181818182e-05, "loss": 0.1092, "step": 1000 }, { "epoch": 2.84, "eval_accuracy": 0.9600718501411342, "eval_f1": 0.37353433835845895, "eval_loss": 0.2574901878833771, "eval_precision": 0.5425790754257908, "eval_recall": 0.2848020434227331, "eval_runtime": 2.1429, "eval_samples_per_second": 378.923, "eval_steps_per_second": 47.599, "step": 1000 }, { "epoch": 4.26, "learning_rate": 1.1477272727272729e-05, "loss": 0.0797, "step": 1500 }, { "epoch": 4.26, "eval_accuracy": 0.957557095201437, "eval_f1": 0.38830584707646176, "eval_loss": 0.2453581690788269, "eval_precision": 0.47005444646098005, "eval_recall": 0.33077905491698595, "eval_runtime": 2.2968, "eval_samples_per_second": 353.535, "eval_steps_per_second": 44.41, "step": 1500 }, { "epoch": 5.68, "learning_rate": 8.636363636363637e-06, "loss": 0.0615, "step": 2000 }, { "epoch": 5.68, "eval_accuracy": 0.9585835257890685, "eval_f1": 0.38574748257164987, "eval_loss": 0.2668910622596741, "eval_precision": 0.49015748031496065, "eval_recall": 0.31800766283524906, "eval_runtime": 2.1824, "eval_samples_per_second": 372.068, "eval_steps_per_second": 46.738, "step": 2000 }, { "epoch": 7.1, "learning_rate": 5.795454545454546e-06, "loss": 0.0504, "step": 2500 }, { "epoch": 7.1, "eval_accuracy": 0.9580189889658712, "eval_f1": 0.40949554896142437, "eval_loss": 0.26873332262039185, "eval_precision": 0.48849557522123893, "eval_recall": 0.3524904214559387, "eval_runtime": 2.58, "eval_samples_per_second": 314.726, "eval_steps_per_second": 39.534, "step": 2500 }, { "epoch": 8.52, "learning_rate": 2.954545454545455e-06, "loss": 0.0379, "step": 3000 }, { "epoch": 8.52, "eval_accuracy": 0.9573004875545291, "eval_f1": 0.4077530509691314, "eval_loss": 0.2752237319946289, "eval_precision": 0.46557377049180326, "eval_recall": 0.36270753512132825, "eval_runtime": 2.3982, "eval_samples_per_second": 338.588, "eval_steps_per_second": 42.532, "step": 3000 }, { "epoch": 9.94, "learning_rate": 1.1363636363636364e-07, "loss": 0.0339, "step": 3500 }, { "epoch": 9.94, "eval_accuracy": 0.9585835257890685, "eval_f1": 0.41141141141141135, "eval_loss": 0.2827732264995575, "eval_precision": 0.4990892531876138, "eval_recall": 0.3499361430395913, "eval_runtime": 2.6514, "eval_samples_per_second": 306.252, "eval_steps_per_second": 38.47, "step": 3500 }, { "epoch": 10.0, "step": 3520, "total_flos": 583523808348012.0, "train_loss": 0.08588492049073632, "train_runtime": 361.1884, "train_samples_per_second": 77.799, "train_steps_per_second": 9.746 } ], "logging_steps": 500, "max_steps": 3520, "num_train_epochs": 10, "save_steps": 500, "total_flos": 583523808348012.0, "trial_name": null, "trial_params": null }