ToxiGen

该数据集旨在评估大型语言模型输出中的毒性问题。它专注于生成有毒提示，以此来评估模型的响应。该数据集的任务是对输出毒性进行评估。

This dataset is designed to evaluate the toxicity issue in the outputs of Large Language Models (LLMs). It focuses on generating toxic prompts to assess model responses. The core task of this dataset is to evaluate the toxicity of model outputs.