{ "best_metric": 0.2427176833152771, "best_model_checkpoint": "/content/dnrti_securebert/checkpoint-4500", "epoch": 10.0, "eval_steps": 500, "global_step": 6570, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.76, "learning_rate": 1.8477929984779303e-05, "loss": 0.7821, "step": 500 }, { "epoch": 0.76, "eval_accuracy": 0.8744877049180327, "eval_f1": 0.5547161666306928, "eval_loss": 0.4215104579925537, "eval_precision": 0.5219333874898456, "eval_recall": 0.5918931368033165, "eval_runtime": 3.0633, "eval_samples_per_second": 215.778, "eval_steps_per_second": 27.095, "step": 500 }, { "epoch": 1.52, "learning_rate": 1.69558599695586e-05, "loss": 0.3559, "step": 1000 }, { "epoch": 1.52, "eval_accuracy": 0.900785519125683, "eval_f1": 0.6425522354527072, "eval_loss": 0.31522974371910095, "eval_precision": 0.6271929824561403, "eval_recall": 0.6586826347305389, "eval_runtime": 2.5699, "eval_samples_per_second": 257.203, "eval_steps_per_second": 32.296, "step": 1000 }, { "epoch": 2.28, "learning_rate": 1.54337899543379e-05, "loss": 0.2807, "step": 1500 }, { "epoch": 2.28, "eval_accuracy": 0.9084130236794171, "eval_f1": 0.681571521597569, "eval_loss": 0.29523420333862305, "eval_precision": 0.6444991789819376, "eval_recall": 0.7231690465223399, "eval_runtime": 2.6375, "eval_samples_per_second": 250.619, "eval_steps_per_second": 31.47, "step": 1500 }, { "epoch": 3.04, "learning_rate": 1.39117199391172e-05, "loss": 0.2272, "step": 2000 }, { "epoch": 3.04, "eval_accuracy": 0.9160974499089253, "eval_f1": 0.707285342584562, "eval_loss": 0.2793317437171936, "eval_precision": 0.6681687832855387, "eval_recall": 0.7512666973744818, "eval_runtime": 3.0401, "eval_samples_per_second": 217.427, "eval_steps_per_second": 27.302, "step": 2000 }, { "epoch": 3.81, "learning_rate": 1.2389649923896501e-05, "loss": 0.1837, "step": 2500 }, { "epoch": 3.81, "eval_accuracy": 0.9258310564663024, "eval_f1": 0.7333931777378815, "eval_loss": 0.2489345222711563, "eval_precision": 0.7150984682713348, "eval_recall": 0.7526485490557346, "eval_runtime": 2.5725, "eval_samples_per_second": 256.952, "eval_steps_per_second": 32.265, "step": 2500 }, { "epoch": 4.57, "learning_rate": 1.08675799086758e-05, "loss": 0.1497, "step": 3000 }, { "epoch": 4.57, "eval_accuracy": 0.9285632969034608, "eval_f1": 0.7529359627742077, "eval_loss": 0.25113749504089355, "eval_precision": 0.7254483347566183, "eval_recall": 0.7825886688162137, "eval_runtime": 2.9682, "eval_samples_per_second": 222.691, "eval_steps_per_second": 27.963, "step": 3000 }, { "epoch": 5.33, "learning_rate": 9.3455098934551e-06, "loss": 0.1371, "step": 3500 }, { "epoch": 5.33, "eval_accuracy": 0.9331170309653917, "eval_f1": 0.7587294435683712, "eval_loss": 0.24962495267391205, "eval_precision": 0.7425044091710759, "eval_recall": 0.7756794104099494, "eval_runtime": 2.5966, "eval_samples_per_second": 254.564, "eval_steps_per_second": 31.965, "step": 3500 }, { "epoch": 6.09, "learning_rate": 7.823439878234399e-06, "loss": 0.1135, "step": 4000 }, { "epoch": 6.09, "eval_accuracy": 0.9325478142076503, "eval_f1": 0.7661713286713285, "eval_loss": 0.25536227226257324, "eval_precision": 0.7288981288981289, "eval_recall": 0.8074619990787656, "eval_runtime": 2.5496, "eval_samples_per_second": 259.252, "eval_steps_per_second": 32.554, "step": 4000 }, { "epoch": 6.85, "learning_rate": 6.301369863013699e-06, "loss": 0.1018, "step": 4500 }, { "epoch": 6.85, "eval_accuracy": 0.9382399817850637, "eval_f1": 0.7772965580123091, "eval_loss": 0.2427176833152771, "eval_precision": 0.769404332129964, "eval_recall": 0.7853523721787194, "eval_runtime": 2.5904, "eval_samples_per_second": 255.175, "eval_steps_per_second": 32.042, "step": 4500 }, { "epoch": 7.61, "learning_rate": 4.779299847792998e-06, "loss": 0.0899, "step": 5000 }, { "epoch": 7.61, "eval_accuracy": 0.9377846083788707, "eval_f1": 0.7864271457085829, "eval_loss": 0.2516387104988098, "eval_precision": 0.7583404619332763, "eval_recall": 0.8166743436204514, "eval_runtime": 2.5729, "eval_samples_per_second": 256.912, "eval_steps_per_second": 32.26, "step": 5000 }, { "epoch": 8.37, "learning_rate": 3.2572298325722984e-06, "loss": 0.0809, "step": 5500 }, { "epoch": 8.37, "eval_accuracy": 0.9405737704918032, "eval_f1": 0.7940058152538582, "eval_loss": 0.24588441848754883, "eval_precision": 0.7717391304347826, "eval_recall": 0.81759557807462, "eval_runtime": 2.5675, "eval_samples_per_second": 257.446, "eval_steps_per_second": 32.327, "step": 5500 }, { "epoch": 9.13, "learning_rate": 1.7351598173515982e-06, "loss": 0.0763, "step": 6000 }, { "epoch": 9.13, "eval_accuracy": 0.9392076502732241, "eval_f1": 0.7852112676056338, "eval_loss": 0.2552666962146759, "eval_precision": 0.7517909818794775, "eval_recall": 0.8217411331183786, "eval_runtime": 2.7364, "eval_samples_per_second": 241.554, "eval_steps_per_second": 30.331, "step": 6000 }, { "epoch": 9.89, "learning_rate": 2.1308980213089802e-07, "loss": 0.0687, "step": 6500 }, { "epoch": 9.89, "eval_accuracy": 0.9406876138433515, "eval_f1": 0.790150842945874, "eval_loss": 0.2534164488315582, "eval_precision": 0.7620881471972615, "eval_recall": 0.8203592814371258, "eval_runtime": 3.0206, "eval_samples_per_second": 218.833, "eval_steps_per_second": 27.478, "step": 6500 }, { "epoch": 10.0, "step": 6570, "total_flos": 1395895287387240.0, "train_loss": 0.20224008327931212, "train_runtime": 907.9488, "train_samples_per_second": 57.823, "train_steps_per_second": 7.236 } ], "logging_steps": 500, "max_steps": 6570, "num_train_epochs": 10, "save_steps": 500, "total_flos": 1395895287387240.0, "trial_name": null, "trial_params": null }