{ "effective_tokens_per_sec": 6922.665391312463, "epoch": 4.0, "total_flos": 4.038502240003031e+18, "train_loss": 0.7417308912200637, "train_runtime": 11316.072, "train_samples_per_second": 5.323, "train_steps_per_second": 0.089 }