{ "effective_tokens_per_sec": 7436.254647328101, "epoch": 4.0, "total_flos": 4.038502240003031e+18, "train_loss": 1.0441685825645686, "train_runtime": 10534.5209, "train_samples_per_second": 5.718, "train_steps_per_second": 0.095 }