{ "effective_tokens_per_sec": 7014.089972720708, "epoch": 3.0, "total_flos": 4.2988160857187287e+18, "train_loss": 0.8751117374819068, "train_runtime": 6006.754, "train_samples_per_second": 9.493, "train_steps_per_second": 0.593 }