{ "epoch": 0.9989528795811519, "eval_fcm_dpo/beta": 0.009538780897855759, "eval_logits/chosen": -0.8344770073890686, "eval_logits/rejected": -0.8241644501686096, "eval_logps/chosen": -390.4581604003906, "eval_logps/ref_chosen": -287.8268127441406, "eval_logps/ref_rejected": -266.9300231933594, "eval_logps/rejected": -424.9966735839844, "eval_loss": 0.5423681139945984, "eval_margin_dpo/margin_mean": 55.43527603149414, "eval_margin_dpo/margin_std": 80.3389663696289, "eval_runtime": 78.5759, "eval_samples": 2000, "eval_samples_per_second": 25.453, "eval_steps_per_second": 3.182 }