snap-stanford/preference_iterative_hard
收藏Hugging Face2025-01-28 更新2025-04-08 收录
下载链接:
https://hf-mirror.com/datasets/snap-stanford/preference_iterative_hard
下载链接
链接失效反馈官方服务:
资源简介:
该数据集包含了对话上下文、选择的回复、被拒绝的回复、选择的回复得分、被拒绝的回复得分、历史得分记录和历史转发记录等字段。数据集整体被划分为answer_generator部分,共有717个样本。
The dataset includes fields such as conversation context, chosen response, rejected response, score of chosen response, score of rejected response, history of scores, and history of forwards. The dataset is divided into one part called answer_generator, containing a total of 717 samples.
提供机构:
snap-stanford



