{ "epoch": 2.269721303731696, "eval_dpo_loss": 0.6728345155715942, "eval_logits": -8.890972137451172, "eval_logps": -573.5075073242188, "eval_loss": 0.3941618800163269, "eval_objective": 0.3930768370628357, "eval_ranking_idealized": 0.9870600700378418, "eval_ranking_idealized_expo": 0.6319875717163086, "eval_ranking_simple": 0.6102484464645386, "eval_regularize": 0.3930768370628357, "eval_runtime": 310.9135, "eval_samples": 5790, "eval_samples_per_second": 18.623, "eval_steps_per_second": 1.553, "eval_wo_beta": 160.35784912109375 }