patrickmlml/yourbench_example-openrouter

该数据集是一个综合性的文本数据集，包含文档ID、文本内容、文件名、文档元数据（如文件大小）、文本摘要、多跳和单次提问及其答案、难度评估、引用信息等字段。数据集分为多个配置，包括分块的文本、已摄入的文本、轻量评估问题、多跳问题、单次提问和文本摘要等。每个配置都有训练集分割，并且提供了数据的大小和示例数量。

This dataset is a comprehensive text dataset containing fields such as document ID, text content, filename, document metadata (e.g., file size), text summary, multi-hop and single-shot questions and answers, difficulty assessment, citation information, etc. The dataset is divided into multiple configurations, including chunked text, ingested text, lighteval questions, multi-hop questions, single-shot questions, and text summaries. Each configuration has a training set split, and the size of the data and the number of examples are provided.