{ "epoch": 0.999244142101285, "eval_logits/chosen": 0.900075376033783, "eval_logits/rejected": 0.867337167263031, "eval_logps/chosen": -211.55142211914062, "eval_logps/ref_chosen": -75.30646514892578, "eval_logps/ref_rejected": -77.75511932373047, "eval_logps/rejected": -274.3049621582031, "eval_loss": 0.5347095727920532, "eval_margin_dpo/margin_mean": 60.304866790771484, "eval_margin_dpo/margin_std": 94.75297546386719, "eval_runtime": 37.4264, "eval_samples": 2303, "eval_samples_per_second": 61.534, "eval_steps_per_second": 1.924 }