{ "effective_tokens_per_sec": 2424.672079197759, "epoch": 3.0, "total_flos": 2.513699577954042e+18, "train_loss": 1.4121140867471695, "train_runtime": 17187.153, "train_samples_per_second": 1.564, "train_steps_per_second": 0.098 }