{ "epoch": 0.9989528795811519, "eval_KL/chosen_KL_mean": -262.6319580078125, "eval_KL/mean": -316.4208679199219, "eval_KL/rejected_KL_mean": -370.2097473144531, "eval_KL/std": 165.86854553222656, "eval_fcm_dpo/beta": 0.0024308497086167336, "eval_logits/chosen": -0.846673846244812, "eval_logits/rejected": -0.8289116621017456, "eval_logps/chosen": -550.458740234375, "eval_logps/ref_chosen": -287.8268127441406, "eval_logps/ref_rejected": -266.9300231933594, "eval_logps/rejected": -637.1398315429688, "eval_loss": 0.6021535396575928, "eval_margin_dpo/margin_mean": 107.57779693603516, "eval_margin_dpo/margin_std": 186.7196807861328, "eval_runtime": 78.3489, "eval_samples": 2000, "eval_samples_per_second": 25.527, "eval_steps_per_second": 3.191 }