{ "effective_tokens_per_sec": 7862.994109807929, "epoch": 4.0, "total_flos": 3.519329211045118e+18, "train_loss": 0.7596988798197047, "train_runtime": 8661.0758, "train_samples_per_second": 6.955, "train_steps_per_second": 0.116 }