{ "effective_tokens_per_sec": 8017.957320589926, "epoch": 3.0, "total_flos": 4.931331991116186e+18, "train_loss": 0.88325994085853, "train_runtime": 11968.8346, "train_samples_per_second": 5.504, "train_steps_per_second": 0.092 }