{ "best_metric": null, "best_model_checkpoint": null, "epoch": 3.0, "eval_steps": 500, "global_step": 6156, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.24, "grad_norm": 1.4913195371627808, "learning_rate": 2.0292207792207792e-05, "loss": 1.0759, "step": 500 }, { "epoch": 0.49, "grad_norm": 0.7769107222557068, "learning_rate": 4.0584415584415584e-05, "loss": 0.2485, "step": 1000 }, { "epoch": 0.73, "grad_norm": 0.9267131686210632, "learning_rate": 4.963542668550149e-05, "loss": 0.1884, "step": 1500 }, { "epoch": 0.97, "grad_norm": 0.8767552971839905, "learning_rate": 4.7058361708328026e-05, "loss": 0.1553, "step": 2000 }, { "epoch": 1.22, "grad_norm": 1.5689440965652466, "learning_rate": 4.225547310225216e-05, "loss": 0.133, "step": 2500 }, { "epoch": 1.46, "grad_norm": 1.353319764137268, "learning_rate": 3.5711401707230445e-05, "loss": 0.1242, "step": 3000 }, { "epoch": 1.71, "grad_norm": 0.7634446620941162, "learning_rate": 2.808648435756026e-05, "loss": 0.1157, "step": 3500 }, { "epoch": 1.95, "grad_norm": 0.5596290826797485, "learning_rate": 2.0150121855972848e-05, "loss": 0.1079, "step": 4000 }, { "epoch": 2.19, "grad_norm": 0.407377690076828, "learning_rate": 1.2703141726159556e-05, "loss": 0.0933, "step": 4500 }, { "epoch": 2.44, "grad_norm": 0.9677791595458984, "learning_rate": 6.496989816913732e-06, "loss": 0.086, "step": 5000 }, { "epoch": 2.68, "grad_norm": 1.0203440189361572, "learning_rate": 2.157904819340964e-06, "loss": 0.0836, "step": 5500 }, { "epoch": 2.92, "grad_norm": 0.48467525839805603, "learning_rate": 1.2372695218094142e-07, "loss": 0.0787, "step": 6000 }, { "epoch": 3.0, "step": 6156, "total_flos": 4.250069188231039e+16, "train_loss": 0.2043684836406844, "train_runtime": 5455.8592, "train_samples_per_second": 40.607, "train_steps_per_second": 1.128 }, { "epoch": 3.0, "eval_AADHAR_CARD_f1": 0.7498692696531287, "eval_AGE_f1": 0.5816216216216216, "eval_CITY_f1": 0.7305788680869378, "eval_COUNTRY_f1": 0.6781510192362905, "eval_CREDITCARDCVV_f1": 0.7219837863614689, "eval_CREDITCARDNUMBER_f1": 0.7364130434782609, "eval_CURRENCYNAME_f1": 0.08870967741935484, "eval_CURRENCY_f1": 0.6680805938494168, "eval_DATEOFBIRTH_f1": 0.6437189496096523, "eval_DATE_f1": 0.6695026178010473, "eval_EMAIL_f1": 0.6485573173901742, "eval_EXPIRY_DATE_f1": 0.5622568093385214, "eval_ORGANIZATION_f1": 0.7392519632636763, "eval_PAN_CARD_f1": 0.7191325638335083, "eval_PERSON_f1": 0.8087976183243211, "eval_PHONENUMBER_f1": 0.7218013468013468, "eval_SECONDARY_ADDRESS_f1": 0.6800595238095238, "eval_STATE_f1": 0.7525204032645223, "eval_STREET_f1": 0.8528925619834711, "eval_TIME_f1": 0.7544524441076165, "eval_URL_f1": 0.5520118422823308, "eval_US_SSN_f1": 0.90694626474443, "eval_loss": 0.09682328253984451, "eval_overall_accuracy": 0.9661661821488446, "eval_overall_f1": 0.7095774460143962, "eval_overall_precision": 0.6921289148206536, "eval_overall_recall": 0.727928482300053, "eval_runtime": 151.236, "eval_samples_per_second": 122.081, "eval_steps_per_second": 15.261, "step": 6156 } ], "logging_steps": 500, "max_steps": 6156, "num_input_tokens_seen": 0, "num_train_epochs": 3, "save_steps": 500, "total_flos": 4.250069188231039e+16, "train_batch_size": 36, "trial_name": null, "trial_params": null }