DataShare/sos-artifacts
收藏Hugging Face2025-04-21 更新2025-04-26 收录
下载链接:
https://hf-mirror.com/datasets/DataShare/sos-artifacts
下载链接
链接失效反馈官方服务:
资源简介:
Arena-Hard-Auto基准数据集,包含不同模型对问题的回答和评判,以及相关的排行榜和评判模板。该数据集用于研究大型语言模型在评判对齐时的表现。
Arena-Hard-Auto benchmark dataset containing model answers and judgments for questions, along with leaderboards and judgment templates. This dataset is used to study the performance of large language models in alignment judging.
提供机构:
DataShare



