andreaparker/wiki-ss-nq-train-overlap-sm

该数据集包含用于训练的信息检索模型的数据，主要包括查询ID、查询内容、正面和负面相关文档以及答案。每个查询都有对应的正面和负面文档列表，每个文档包含文档ID、文本内容和标题。数据集被分割为训练集，包含1000个样本，总大小为139483047.95236582字节。

This dataset contains data for training information retrieval models, mainly including query IDs, query content, positive and negative relevant documents, and answers. Each query has a corresponding list of positive and negative documents, each containing a document ID, text content, and title. The dataset is split into a training set containing 1000 samples, with a total size of 139483047.95236582 bytes.