msaramhassan/analogy_concept_questions_new

该数据集包含三种配置：chunked、ingested和summarized。每种配置都包括文档ID、文档文本、文档文件名和文档元数据（文件大小）。chunked和summarized配置还包括文档摘要和摘要模型信息。chunked配置还包括文本块信息，summarized配置还包括多跳文本块信息。数据集分为训练集，每种配置的训练集大小和下载大小不同。

The dataset consists of three configurations: chunked, ingested, and summarized. Each configuration includes document ID, document text, document filename, and document metadata (file size). The chunked and summarized configurations also include document summary and summarization model information. The chunked configuration also includes chunk information, and the summarized configuration also includes multi-hop chunk information. The dataset is split into a training set, with different sizes and download sizes for each configurations training set.