{ "best_metric": 0.6181691884994507, "best_model_checkpoint": "hBERTv2_cola/checkpoint-68", "epoch": 7.0, "global_step": 238, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 1.0, "learning_rate": 4.9029411764705883e-05, "loss": 0.6296, "step": 34 }, { "epoch": 1.0, "eval_loss": 0.618189811706543, "eval_matthews_correlation": 0.0, "eval_runtime": 1.2234, "eval_samples_per_second": 852.526, "eval_steps_per_second": 4.087, "step": 34 }, { "epoch": 2.0, "learning_rate": 4.8029411764705886e-05, "loss": 0.614, "step": 68 }, { "epoch": 2.0, "eval_loss": 0.6181691884994507, "eval_matthews_correlation": 0.0, "eval_runtime": 1.2214, "eval_samples_per_second": 853.934, "eval_steps_per_second": 4.094, "step": 68 }, { "epoch": 3.0, "learning_rate": 4.702941176470588e-05, "loss": 0.6089, "step": 102 }, { "epoch": 3.0, "eval_loss": 0.6213549971580505, "eval_matthews_correlation": 0.0, "eval_runtime": 1.2227, "eval_samples_per_second": 853.001, "eval_steps_per_second": 4.089, "step": 102 }, { "epoch": 4.0, "learning_rate": 4.6029411764705885e-05, "loss": 0.6092, "step": 136 }, { "epoch": 4.0, "eval_loss": 0.6190556287765503, "eval_matthews_correlation": 0.0, "eval_runtime": 1.2244, "eval_samples_per_second": 851.842, "eval_steps_per_second": 4.084, "step": 136 }, { "epoch": 5.0, "learning_rate": 4.502941176470589e-05, "loss": 0.6103, "step": 170 }, { "epoch": 5.0, "eval_loss": 0.6221293210983276, "eval_matthews_correlation": 0.0, "eval_runtime": 1.2243, "eval_samples_per_second": 851.906, "eval_steps_per_second": 4.084, "step": 170 }, { "epoch": 6.0, "learning_rate": 4.4029411764705884e-05, "loss": 0.6094, "step": 204 }, { "epoch": 6.0, "eval_loss": 0.6182675361633301, "eval_matthews_correlation": 0.0, "eval_runtime": 1.2259, "eval_samples_per_second": 850.776, "eval_steps_per_second": 4.079, "step": 204 }, { "epoch": 7.0, "learning_rate": 4.302941176470589e-05, "loss": 0.6096, "step": 238 }, { "epoch": 7.0, "eval_loss": 0.6196076273918152, "eval_matthews_correlation": 0.0, "eval_runtime": 1.22, "eval_samples_per_second": 854.907, "eval_steps_per_second": 4.098, "step": 238 }, { "epoch": 7.0, "step": 238, "total_flos": 7455394754461696.0, "train_loss": 0.6130136441783745, "train_runtime": 303.6084, "train_samples_per_second": 1408.228, "train_steps_per_second": 5.599 } ], "max_steps": 1700, "num_train_epochs": 50, "total_flos": 7455394754461696.0, "trial_name": null, "trial_params": null }