{ "best_metric": 0.6082243323326111, "best_model_checkpoint": "hBERTv1_new_pretrain_w_init__mrpc/checkpoint-116", "epoch": 9.0, "global_step": 261, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 1.0, "learning_rate": 3.9200000000000004e-05, "loss": 0.7111, "step": 29 }, { "epoch": 1.0, "eval_accuracy": 0.6838235294117647, "eval_combined_score": 0.7480253018237863, "eval_f1": 0.8122270742358079, "eval_loss": 0.6563864350318909, "eval_runtime": 0.7398, "eval_samples_per_second": 551.531, "eval_steps_per_second": 5.407, "step": 29 }, { "epoch": 2.0, "learning_rate": 3.8400000000000005e-05, "loss": 0.6641, "step": 58 }, { "epoch": 2.0, "eval_accuracy": 0.6838235294117647, "eval_combined_score": 0.7480253018237863, "eval_f1": 0.8122270742358079, "eval_loss": 0.6160457730293274, "eval_runtime": 0.7439, "eval_samples_per_second": 548.435, "eval_steps_per_second": 5.377, "step": 58 }, { "epoch": 3.0, "learning_rate": 3.76e-05, "loss": 0.6156, "step": 87 }, { "epoch": 3.0, "eval_accuracy": 0.6838235294117647, "eval_combined_score": 0.7480253018237863, "eval_f1": 0.8122270742358079, "eval_loss": 0.6354035139083862, "eval_runtime": 0.7456, "eval_samples_per_second": 547.23, "eval_steps_per_second": 5.365, "step": 87 }, { "epoch": 4.0, "learning_rate": 3.680000000000001e-05, "loss": 0.5817, "step": 116 }, { "epoch": 4.0, "eval_accuracy": 0.6862745098039216, "eval_combined_score": 0.737874097007224, "eval_f1": 0.7894736842105262, "eval_loss": 0.6082243323326111, "eval_runtime": 0.7468, "eval_samples_per_second": 546.334, "eval_steps_per_second": 5.356, "step": 116 }, { "epoch": 5.0, "learning_rate": 3.6e-05, "loss": 0.5091, "step": 145 }, { "epoch": 5.0, "eval_accuracy": 0.5073529411764706, "eval_combined_score": 0.5115077958894401, "eval_f1": 0.5156626506024096, "eval_loss": 0.7811743021011353, "eval_runtime": 0.7431, "eval_samples_per_second": 549.025, "eval_steps_per_second": 5.383, "step": 145 }, { "epoch": 6.0, "learning_rate": 3.52e-05, "loss": 0.3973, "step": 174 }, { "epoch": 6.0, "eval_accuracy": 0.6544117647058824, "eval_combined_score": 0.705444224321853, "eval_f1": 0.7564766839378239, "eval_loss": 0.7949338555335999, "eval_runtime": 0.7416, "eval_samples_per_second": 550.17, "eval_steps_per_second": 5.394, "step": 174 }, { "epoch": 7.0, "learning_rate": 3.44e-05, "loss": 0.2966, "step": 203 }, { "epoch": 7.0, "eval_accuracy": 0.6078431372549019, "eval_combined_score": 0.6482795452811474, "eval_f1": 0.688715953307393, "eval_loss": 1.0388320684432983, "eval_runtime": 0.7411, "eval_samples_per_second": 550.501, "eval_steps_per_second": 5.397, "step": 203 }, { "epoch": 8.0, "learning_rate": 3.3600000000000004e-05, "loss": 0.2024, "step": 232 }, { "epoch": 8.0, "eval_accuracy": 0.6200980392156863, "eval_combined_score": 0.6662642329659136, "eval_f1": 0.7124304267161411, "eval_loss": 1.0064674615859985, "eval_runtime": 0.7424, "eval_samples_per_second": 549.6, "eval_steps_per_second": 5.388, "step": 232 }, { "epoch": 9.0, "learning_rate": 3.28e-05, "loss": 0.1621, "step": 261 }, { "epoch": 9.0, "eval_accuracy": 0.5735294117647058, "eval_combined_score": 0.6155048633626679, "eval_f1": 0.65748031496063, "eval_loss": 1.307615876197815, "eval_runtime": 0.7454, "eval_samples_per_second": 547.342, "eval_steps_per_second": 5.366, "step": 261 }, { "epoch": 9.0, "step": 261, "total_flos": 4888970768941056.0, "train_loss": 0.46000806216535894, "train_runtime": 225.7005, "train_samples_per_second": 812.581, "train_steps_per_second": 6.424 } ], "max_steps": 1450, "num_train_epochs": 50, "total_flos": 4888970768941056.0, "trial_name": null, "trial_params": null }