Model Self-Identification Dataset
收藏arXiv2025-09-30 收录
下载链接：
https://github.com/lm-sys/FastChat/blob/v0.2.1/playground/data/dummy.json
下载链接
链接失效反馈官方服务：
资源简介：
该数据集包含了用于回答身份相关问题的少量提示-响应对，它被作为属性条件下的SFT（敏感度函数训练）的一部分包含在内。该数据集的规模为910对，其任务是对属性条件下的SFT进行训练。
This dataset contains a small number of prompt-response pairs for answering identity-related questions. It is included as part of the SFT (Sensitivity Function Training) under attribute conditions. The dataset comprises 910 such pairs, and its purpose is to train the SFT under attribute conditions.
提供机构：
Model Self-Identification
