{ "effective_tokens_per_sec": 7219.495500422262, "epoch": 4.0, "total_flos": 3.5845266746764165e+18, "train_loss": 0.7556913677518341, "train_runtime": 4810.6403, "train_samples_per_second": 11.799, "train_steps_per_second": 0.393 }