{ "effective_tokens_per_sec": 874.8418751446711, "epoch": 8.790419161676647, "total_flos": 2.992361562963968e+17, "train_loss": 0.17746063170394277, "train_runtime": 2311.8158, "train_samples_per_second": 3.893, "train_steps_per_second": 0.16 }