{ "best_metric": 0.9900189823035943, "best_model_checkpoint": "final_models/glue_models/structroberta_s2_50ep//finetune/main_verb_control/checkpoint-600", "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": 0.9280734062194824, "eval_f1": 0.9230164964650432, "eval_loss": 0.25712713599205017, "eval_mcc": 0.8634812917806318, "eval_runtime": 54.3317, "eval_samples_per_second": 300.929, "eval_steps_per_second": 37.621, "step": 200 }, { "epoch": 3.12, "eval_accuracy": 0.9854434132575989, "eval_f1": 0.985414879274421, "eval_loss": 0.051639676094055176, "eval_mcc": 0.9708903070492874, "eval_runtime": 49.9623, "eval_samples_per_second": 327.247, "eval_steps_per_second": 40.911, "step": 400 }, { "epoch": 3.91, "learning_rate": 3.0468750000000002e-05, "loss": 0.141, "step": 500 }, { "epoch": 4.69, "eval_accuracy": 0.9900305867195129, "eval_f1": 0.9900189823035943, "eval_loss": 0.04498597979545593, "eval_mcc": 0.9800617351233112, "eval_runtime": 53.1065, "eval_samples_per_second": 307.872, "eval_steps_per_second": 38.489, "step": 600 }, { "epoch": 6.25, "eval_accuracy": 0.9856269359588623, "eval_f1": 0.9855340104647584, "eval_loss": 0.0649297907948494, "eval_mcc": 0.971319195627004, "eval_runtime": 51.3997, "eval_samples_per_second": 318.095, "eval_steps_per_second": 39.767, "step": 800 }, { "epoch": 7.81, "learning_rate": 1.09375e-05, "loss": 0.0023, "step": 1000 }, { "epoch": 7.81, "eval_accuracy": 0.9882568717002869, "eval_f1": 0.9881232215761475, "eval_loss": 0.05662437155842781, "eval_mcc": 0.9767347209581166, "eval_runtime": 50.9239, "eval_samples_per_second": 321.067, "eval_steps_per_second": 40.138, "step": 1000 }, { "epoch": 9.38, "eval_accuracy": 0.984770655632019, "eval_f1": 0.9845466393595235, "eval_loss": 0.07832500338554382, "eval_mcc": 0.9699147617250261, "eval_runtime": 51.2538, "eval_samples_per_second": 319.001, "eval_steps_per_second": 39.88, "step": 1200 }, { "epoch": 10.0, "step": 1280, "total_flos": 7480384416921600.0, "train_loss": 0.056016668380470945, "train_runtime": 903.2296, "train_samples_per_second": 90.409, "train_steps_per_second": 1.417 } ], "max_steps": 1280, "num_train_epochs": 10, "total_flos": 7480384416921600.0, "trial_name": null, "trial_params": null }