{ "effective_tokens_per_sec": 6531.597542486509, "epoch": 4.0, "total_flos": 3.873354436822696e+18, "train_loss": 0.8859393315431767, "train_runtime": 11385.0199, "train_samples_per_second": 5.607, "train_steps_per_second": 0.047 }