{ "epoch": 1.0, "eval_error": 0.2430962324142456, "eval_logits/chosen": -2.8012824058532715, "eval_logits/rejected": -2.779707670211792, "eval_logps/chosen": -497.6483154296875, "eval_logps/rejected": -503.2738037109375, "eval_loss": 0.7637462019920349, "eval_preference/chosen": 0.5319993495941162, "eval_preference/rejected": 0.4680006802082062, "eval_rewards/accuracies": 0.5389999747276306, "eval_rewards/chosen": 0.0706886574625969, "eval_rewards/margins": 0.2767164409160614, "eval_rewards/rejected": -0.2060278058052063, "eval_runtime": 41.7363, "eval_samples_per_second": 23.96, "eval_steps_per_second": 2.995 }