{ "effective_tokens_per_sec": 6857.321759551241, "epoch": 3.0, "total_flos": 4.1743170019314893e+18, "train_loss": 0.9459603985573383, "train_runtime": 10552.9203, "train_samples_per_second": 4.276, "train_steps_per_second": 0.134 }