{ "best_metric": 0.12117183208465576, "best_model_checkpoint": "/content/cyner_secbert/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": 0.165, "step": 500 }, { "epoch": 1.42, "eval_accuracy": 0.9723376956633307, "eval_f1": 0.6189111747851002, "eval_loss": 0.12117183208465576, "eval_precision": 0.7047308319738989, "eval_recall": 0.5517241379310345, "eval_runtime": 2.5587, "eval_samples_per_second": 317.348, "eval_steps_per_second": 39.864, "step": 500 }, { "epoch": 2.84, "learning_rate": 1.431818181818182e-05, "loss": 0.04, "step": 1000 }, { "epoch": 2.84, "eval_accuracy": 0.970490120605594, "eval_f1": 0.5904761904761905, "eval_loss": 0.16471919417381287, "eval_precision": 0.6924398625429553, "eval_recall": 0.5146871008939975, "eval_runtime": 1.9792, "eval_samples_per_second": 410.273, "eval_steps_per_second": 51.537, "step": 1000 }, { "epoch": 4.26, "learning_rate": 1.1477272727272729e-05, "loss": 0.0156, "step": 1500 }, { "epoch": 4.26, "eval_accuracy": 0.9713625866050808, "eval_f1": 0.5977175463623395, "eval_loss": 0.18028192222118378, "eval_precision": 0.6768982229402262, "eval_recall": 0.5351213282247765, "eval_runtime": 2.4993, "eval_samples_per_second": 324.895, "eval_steps_per_second": 40.812, "step": 1500 }, { "epoch": 5.68, "learning_rate": 8.636363636363637e-06, "loss": 0.0087, "step": 2000 }, { "epoch": 5.68, "eval_accuracy": 0.9713112650756992, "eval_f1": 0.5938375350140056, "eval_loss": 0.1866268515586853, "eval_precision": 0.6573643410852713, "eval_recall": 0.541507024265645, "eval_runtime": 2.0145, "eval_samples_per_second": 403.074, "eval_steps_per_second": 50.632, "step": 2000 }, { "epoch": 7.1, "learning_rate": 5.795454545454546e-06, "loss": 0.0036, "step": 2500 }, { "epoch": 7.1, "eval_accuracy": 0.9719271234282781, "eval_f1": 0.6052076002814919, "eval_loss": 0.20198111236095428, "eval_precision": 0.6739811912225705, "eval_recall": 0.5491698595146871, "eval_runtime": 2.4675, "eval_samples_per_second": 329.073, "eval_steps_per_second": 41.337, "step": 2500 }, { "epoch": 8.52, "learning_rate": 2.954545454545455e-06, "loss": 0.0024, "step": 3000 }, { "epoch": 8.52, "eval_accuracy": 0.9720297664870413, "eval_f1": 0.6080892608089261, "eval_loss": 0.20356133580207825, "eval_precision": 0.6697388632872504, "eval_recall": 0.5568326947637292, "eval_runtime": 2.5877, "eval_samples_per_second": 313.797, "eval_steps_per_second": 39.418, "step": 3000 }, { "epoch": 9.94, "learning_rate": 1.1363636363636364e-07, "loss": 0.0018, "step": 3500 }, { "epoch": 9.94, "eval_accuracy": 0.971465229663844, "eval_f1": 0.6036414565826329, "eval_loss": 0.20842024683952332, "eval_precision": 0.6682170542635659, "eval_recall": 0.5504469987228607, "eval_runtime": 1.9903, "eval_samples_per_second": 407.98, "eval_steps_per_second": 51.249, "step": 3500 }, { "epoch": 10.0, "step": 3520, "total_flos": 507181396306572.0, "train_loss": 0.03371541356159882, "train_runtime": 327.9315, "train_samples_per_second": 85.689, "train_steps_per_second": 10.734 } ], "logging_steps": 500, "max_steps": 3520, "num_train_epochs": 10, "save_steps": 500, "total_flos": 507181396306572.0, "trial_name": null, "trial_params": null }