Shar999/Twin-2K-500_edit

Twin-2K-500数据集包含了来自2,058名美国参与者的全面人口统计和心理数据，旨在为LLM模拟构建数字孪生。数据集分为五个主要部分：完整人物信息文件夹、波次分割文件夹、问题目录和人类响应CSV文件夹、LLM模拟结果文件夹以及原始数据文件夹。完整人物信息文件夹包含每个参与者的完整调查响应，波次分割文件夹用于测试和评估不同的LLM人物创建方法，问题目录和人类响应CSV文件夹提供了结构化的调查问题和标准化响应文件，LLM模拟结果文件夹包含了LLM模拟输出与人类响应的比较，原始数据文件夹提供了来自Qualtrics的原始调查响应文件。数据集还包含了使用示例、JSON格式示例、社会影响、偏见讨论、已知限制、引用和许可证信息。

The Twin-2K-500 dataset contains comprehensive demographic and psychological data from a representative sample of 2,058 US participants, designed for building digital twins for LLM simulations. The dataset is organized into five main sections: the Full Persona Folder, the Wave Split Folder, the Question Catalog and Human Response CSV Folder, the LLM Simulation Results Folder, and the Raw Data Folder. The Full Persona Folder contains complete survey responses for each participant, the Wave Split Folder is designed for testing and evaluating different LLM persona creation methodologies, the Question Catalog and Human Response CSV Folder provides a structured catalog of survey questions and standardized response files, the LLM Simulation Results Folder contains comparisons between LLM-generated and human responses, and the Raw Data Folder provides access to raw survey response files from Qualtrics. The dataset also includes usage examples, JSON format examples, social impact discussions, bias considerations, known limitations, citation information, and licensing details.