{ "best_metric": null, "best_model_checkpoint": null, "epoch": 4.021447721179625, "eval_steps": 500, "global_step": 1500, "is_hyper_param_search": true, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 1.0, "eval_accuracy": 0.7425410747528076, "eval_loss": 0.48700496554374695, "eval_runtime": 8.9854, "eval_samples_per_second": 331.985, "eval_steps_per_second": 20.812, "step": 373 }, { "epoch": 1.3404825737265416, "grad_norm": 33.510581970214844, "learning_rate": 6.933164670907242e-06, "loss": 0.5256, "step": 500 }, { "epoch": 2.0, "eval_accuracy": 0.755950391292572, "eval_loss": 0.5059512257575989, "eval_runtime": 8.9914, "eval_samples_per_second": 331.761, "eval_steps_per_second": 20.798, "step": 746 }, { "epoch": 2.680965147453083, "grad_norm": 13.22558307647705, "learning_rate": 5.859919365813242e-06, "loss": 0.3496, "step": 1000 }, { "epoch": 3.0, "eval_accuracy": 0.7542742490768433, "eval_loss": 0.5935600399971008, "eval_runtime": 8.9199, "eval_samples_per_second": 334.419, "eval_steps_per_second": 20.964, "step": 1119 }, { "epoch": 4.0, "eval_accuracy": 0.7579618096351624, "eval_loss": 0.7120195031166077, "eval_runtime": 8.9843, "eval_samples_per_second": 332.025, "eval_steps_per_second": 20.814, "step": 1492 }, { "epoch": 4.021447721179625, "grad_norm": 4.5996317863464355, "learning_rate": 4.786674060719241e-06, "loss": 0.224, "step": 1500 } ], "logging_steps": 500, "max_steps": 3730, "num_input_tokens_seen": 0, "num_train_epochs": 10, "save_steps": 500, "total_flos": 8230018237658640.0, "train_batch_size": 32, "trial_name": null, "trial_params": { "learning_rate": 8.006409976001242e-06, "per_device_train_batch_size": 32 } }