AIPlans/FilteredPKU-SafeRLHF_chinese

该数据集是一个包含提示（prompt）和两个回应（response）的文本数据集，同时记录了回应的来源、是否安全，以及具体的危害类别和严重程度等级。数据集旨在用于训练和评估模型在处理文本回应时的安全性和适宜性。它分为训练集和测试集，可用于机器学习模型的训练和测试。

This dataset is a collection of prompts and two responses, including their sources, safety information, specific harm categories, and severity levels. The dataset is intended for training and evaluating models on the safety and appropriateness of text responses. It is split into training and test sets for machine learning model training and testing.