遇见数据集

价值观数据集

收藏
官方服务:

资源简介:

上百万字从网络爬取后且经过专家清洗后的中文原始数据,原始数据采集自权威期刊、杂志、法律法规领域权威基础数据库等。经提取后形成大模型训练所需三元组(问题-优秀答案-不良答案)

This dataset consists of over one million Chinese raw characters, which were crawled from the web and cleaned by domain experts. The raw data was sourced from authoritative journals, magazines, and authoritative foundational databases in the legal and regulatory field, along with other reputable channels. Following targeted extraction, triplets formatted as (Question - Excellent Answer - Substandard Answer) required for Large Language Model (LLM) training are generated.

搜集汇总
数据集介绍
价值观数据集 数据集图片
以上内容由遇见数据集搜集并总结生成
二维码
社区交流群
二维码
科研交流群
商业服务