{ "effective_tokens_per_sec": 7589.902716607846, "epoch": 4.0, "total_flos": 3.5845266747435254e+18, "train_loss": 0.8288866212109377, "train_runtime": 9151.7368, "train_samples_per_second": 6.202, "train_steps_per_second": 0.207 }