{ "effective_tokens_per_sec": 7595.2105537626085, "epoch": 4.0, "total_flos": 3.5845266746764165e+18, "train_loss": 0.8317920884405938, "train_runtime": 4572.6706, "train_samples_per_second": 12.413, "train_steps_per_second": 0.414 }