DataTonic/dark_thoughts_case_study_reason

该数据集包含了对话信息，每个对话记录包括id、对话者的思考内容（think）、回应内容（response）、查询内容（query）、对话类别（category）、终点（endpoint）、数据来源（source）和原始数据（source_data）。数据集分为中文和英文两个版本，分别有19100和20626条记录。数据集可用于对话系统的训练和测试。

This dataset includes conversation information, with each conversation record containing id, the thinkers thought content (think), response content (response), query content (query), conversation category (category), endpoint, data source (source), and original data (source_data). The dataset is divided into Chinese and English versions, with 19,100 and 20,626 records respectively. The dataset can be used for training and testing of conversation systems.