{ "best_metric": null, "best_model_checkpoint": null, "epoch": 0.9968847352024922, "eval_steps": 500, "global_step": 80, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.012461059190031152, "grad_norm": 25.687093602481305, "learning_rate": 2.5e-06, "loss": 1.2485, "step": 1 }, { "epoch": 0.06230529595015576, "grad_norm": 15.094847025724023, "learning_rate": 1.25e-05, "loss": 1.1293, "step": 5 }, { "epoch": 0.12461059190031153, "grad_norm": 11.363800874975448, "learning_rate": 1.9961946980917457e-05, "loss": 0.9767, "step": 10 }, { "epoch": 0.18691588785046728, "grad_norm": 8.12136266435658, "learning_rate": 1.953716950748227e-05, "loss": 0.7746, "step": 15 }, { "epoch": 0.24922118380062305, "grad_norm": 7.927023181508759, "learning_rate": 1.866025403784439e-05, "loss": 0.6367, "step": 20 }, { "epoch": 0.3115264797507788, "grad_norm": 3.686503615552781, "learning_rate": 1.737277336810124e-05, "loss": 0.5175, "step": 25 }, { "epoch": 0.37383177570093457, "grad_norm": 3.6799656836763077, "learning_rate": 1.573576436351046e-05, "loss": 0.4736, "step": 30 }, { "epoch": 0.43613707165109034, "grad_norm": 1.824779614147191, "learning_rate": 1.3826834323650899e-05, "loss": 0.4468, "step": 35 }, { "epoch": 0.4984423676012461, "grad_norm": 1.6428127656511486, "learning_rate": 1.1736481776669307e-05, "loss": 0.4412, "step": 40 }, { "epoch": 0.5607476635514018, "grad_norm": 1.255700115951508, "learning_rate": 9.563806126346643e-06, "loss": 0.4362, "step": 45 }, { "epoch": 0.6230529595015576, "grad_norm": 1.1763215498923227, "learning_rate": 7.411809548974792e-06, "loss": 0.4261, "step": 50 }, { "epoch": 0.6853582554517134, "grad_norm": 1.0876919118710244, "learning_rate": 5.382513867649663e-06, "loss": 0.4159, "step": 55 }, { "epoch": 0.7476635514018691, "grad_norm": 1.0885014436037381, "learning_rate": 3.5721239031346067e-06, "loss": 0.4142, "step": 60 }, { "epoch": 0.8099688473520249, "grad_norm": 0.9605618048169687, "learning_rate": 2.0664665970876496e-06, "loss": 0.4068, "step": 65 }, { "epoch": 0.8722741433021807, "grad_norm": 0.9420340387839357, "learning_rate": 9.369221296335007e-07, "loss": 0.4026, "step": 70 }, { "epoch": 0.9345794392523364, "grad_norm": 0.8433707190106806, "learning_rate": 2.370399288006664e-07, "loss": 0.3984, "step": 75 }, { "epoch": 0.9968847352024922, "grad_norm": 0.8754132731808941, "learning_rate": 0.0, "loss": 0.399, "step": 80 }, { "epoch": 0.9968847352024922, "eval_loss": 1.1888436079025269, "eval_runtime": 0.7161, "eval_samples_per_second": 15.361, "eval_steps_per_second": 1.396, "step": 80 }, { "epoch": 0.9968847352024922, "step": 80, "total_flos": 33396055080960.0, "train_loss": 0.5449534490704536, "train_runtime": 879.8279, "train_samples_per_second": 46.577, "train_steps_per_second": 0.091 } ], "logging_steps": 5, "max_steps": 80, "num_input_tokens_seen": 0, "num_train_epochs": 1, "save_steps": 500, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": false, "should_training_stop": false }, "attributes": {} } }, "total_flos": 33396055080960.0, "train_batch_size": 16, "trial_name": null, "trial_params": null }