Seanie-lee/ThinkSafe-0.6B-v3
收藏Hugging Face2025-10-31 更新2025-11-15 收录
下载链接:
https://hf-mirror.com/datasets/Seanie-lee/ThinkSafe-0.6B-v3
下载链接
链接失效反馈官方服务:
资源简介:
这是一个关于模型对提示反应的统计数据集,包含有害提示、模型拒绝次数、良性提示等信息,并且经过了LlamaGuard过滤,以确保数据的安全性。
This is a dataset about model responses to prompts, including harmful prompts, number of model refusals, benign prompts, etc., and has been filtered by LlamaGuard to ensure data security.
提供机构:
Seanie-lee



