manu/mldr-zoomed-1000char-total-queries-documents

该数据集包含三个部分：documents、queries和synthetic_queries。documents部分包含文档的chunk_id和文本块信息；queries部分包含原始查询、处理后的查询、chunk_id和对应答案；synthetic_queries部分包含合成的查询、chunk_id和答案。每个部分都有相应的训练集，并提供了字节数和示例数的统计信息。

The dataset consists of three parts: documents, queries, and synthetic_queries. The documents part includes chunk_id and text chunks of the documents; the queries part includes original queries, processed queries, chunk_id, and corresponding answers; the synthetic_queries part includes synthetic queries, chunk_id, and answers. Each part has its corresponding training set with statistics provided for the number of bytes and examples.