{ "best_metric": 0.690871841566128, "best_model_checkpoint": "final_models/glue_models/structroberta_s2_50ep//finetune/control_raising_relative_token_position/checkpoint-200", "epoch": 10.0, "global_step": 1280, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 1.56, "eval_accuracy": 0.6678488254547119, "eval_f1": 0.690871841566128, "eval_loss": 2.7935256958007812, "eval_mcc": 0.34215854020335584, "eval_runtime": 52.1303, "eval_samples_per_second": 470.571, "eval_steps_per_second": 58.833, "step": 200 }, { "epoch": 3.12, "eval_accuracy": 0.6686641573905945, "eval_f1": 0.6732068189128337, "eval_loss": 3.095593214035034, "eval_mcc": 0.33811126683902815, "eval_runtime": 52.2071, "eval_samples_per_second": 469.878, "eval_steps_per_second": 58.747, "step": 400 }, { "epoch": 3.91, "learning_rate": 3.0468750000000002e-05, "loss": 0.0176, "step": 500 }, { "epoch": 4.69, "eval_accuracy": 0.6691125631332397, "eval_f1": 0.6742123218944411, "eval_loss": 3.2520527839660645, "eval_mcc": 0.33910135169070227, "eval_runtime": 52.1876, "eval_samples_per_second": 470.054, "eval_steps_per_second": 58.769, "step": 600 }, { "epoch": 6.25, "eval_accuracy": 0.6692755818367004, "eval_f1": 0.6746601435617756, "eval_loss": 3.348085641860962, "eval_mcc": 0.3394767705468026, "eval_runtime": 51.9657, "eval_samples_per_second": 472.062, "eval_steps_per_second": 59.02, "step": 800 }, { "epoch": 7.81, "learning_rate": 1.09375e-05, "loss": 0.0, "step": 1000 }, { "epoch": 7.81, "eval_accuracy": 0.6689087152481079, "eval_f1": 0.6744428411095078, "eval_loss": 3.4064862728118896, "eval_mcc": 0.33876706470884543, "eval_runtime": 52.1056, "eval_samples_per_second": 470.794, "eval_steps_per_second": 58.861, "step": 1000 }, { "epoch": 9.38, "eval_accuracy": 0.6689087152481079, "eval_f1": 0.6744950304584804, "eval_loss": 3.434260368347168, "eval_mcc": 0.33877617591845427, "eval_runtime": 51.9811, "eval_samples_per_second": 471.921, "eval_steps_per_second": 59.002, "step": 1200 }, { "epoch": 10.0, "step": 1280, "total_flos": 7480384416921600.0, "train_loss": 0.006884438258202863, "train_runtime": 747.6516, "train_samples_per_second": 109.222, "train_steps_per_second": 1.712 } ], "max_steps": 1280, "num_train_epochs": 10, "total_flos": 7480384416921600.0, "trial_name": null, "trial_params": null }