manu/mldr-zoomed-100char-total-queries-documents

该数据集包含了三个部分：文档(documents)，查询(queries)和合成查询(synthetic_queries)。文档部分包含了文档的块及其标识符，查询部分包含了查询文本、块标识符和答案，合成查询部分则包含了块标识符、查询文本和答案。数据集为训练用途，提供了训练分割的数据。

The dataset consists of three parts: documents, queries, and synthetic queries. The documents part includes document chunks and their identifiers, the queries part includes query text, chunk identifiers, and answers, while the synthetic queries part includes chunk identifiers, query text, and answers. The dataset is for training purposes and provides training split data.