aravdhoot/llm-domain-shift-social_media

这是一个包含来自不同社交媒体平台（如Discord、Instagram、Reddit、Spotify和Youtube）的文本数据集，每个平台都有安全和不安全的文本分类。数据集包含提示文本、类别、标签和大型语言模型输出四个字段，可用于训练文本分类模型或评估模型在处理不同类型文本时的表现。

This dataset consists of text data from various social media platforms such as Discord, Instagram, Reddit, Spotify, and Youtube, with each platform having both safe and unsafe text classifications. The dataset includes four fields: prompt text, category, label, and large language model output, which can be used to train text classification models or evaluate the models performance when dealing with different types of text.