{ "best_metric": 0.8492806484295846, "best_model_checkpoint": "./category_model\\checkpoint-72", "epoch": 3.0, "eval_steps": 500, "global_step": 72, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.4166666666666667, "grad_norm": 9.026216506958008, "learning_rate": 1.8611111111111114e-05, "loss": 1.3484, "step": 10 }, { "epoch": 0.8333333333333334, "grad_norm": 9.081929206848145, "learning_rate": 1.7222222222222224e-05, "loss": 1.1548, "step": 20 }, { "epoch": 1.0, "eval_accuracy": 0.7446808510638298, "eval_f1": 0.749899679095448, "eval_loss": 0.9859694838523865, "eval_runtime": 1.1007, "eval_samples_per_second": 42.7, "eval_steps_per_second": 5.451, "step": 24 }, { "epoch": 1.25, "grad_norm": 6.283413410186768, "learning_rate": 1.5833333333333333e-05, "loss": 0.9313, "step": 30 }, { "epoch": 1.6666666666666665, "grad_norm": 5.714058876037598, "learning_rate": 1.4444444444444446e-05, "loss": 0.782, "step": 40 }, { "epoch": 2.0, "eval_accuracy": 0.8297872340425532, "eval_f1": 0.8286119554204661, "eval_loss": 0.6606575846672058, "eval_runtime": 1.1034, "eval_samples_per_second": 42.595, "eval_steps_per_second": 5.438, "step": 48 }, { "epoch": 2.0833333333333335, "grad_norm": 7.774585723876953, "learning_rate": 1.3055555555555557e-05, "loss": 0.5314, "step": 50 }, { "epoch": 2.5, "grad_norm": 5.577451229095459, "learning_rate": 1.1666666666666668e-05, "loss": 0.4685, "step": 60 }, { "epoch": 2.9166666666666665, "grad_norm": 3.6863131523132324, "learning_rate": 1.0277777777777777e-05, "loss": 0.3479, "step": 70 }, { "epoch": 3.0, "eval_accuracy": 0.851063829787234, "eval_f1": 0.8492806484295846, "eval_loss": 0.5073530077934265, "eval_runtime": 1.1055, "eval_samples_per_second": 42.515, "eval_steps_per_second": 5.427, "step": 72 } ], "logging_steps": 10, "max_steps": 144, "num_input_tokens_seen": 0, "num_train_epochs": 6, "save_steps": 500, "total_flos": 25946710423920.0, "train_batch_size": 8, "trial_name": null, "trial_params": null }