MemoryAsModality/EnWiki-Pretrain-Articles-128

该数据集是一个大规模问答数据集，包含id、问题、答案和文档列表等特征。数据集分为训练集和测试集，训练集有约2233万个示例，测试集有约46万个示例。总数据大小约为278.7GB，下载大小约为62.3GB。它用于支持自然语言处理任务，如问答和文档检索。

This dataset is a large-scale question-answering dataset with features including id, question, answer, and a list of documents. It is split into training and test sets, with approximately 22.33 million examples in the training set and 458,000 examples in the test set. The total dataset size is about 278.7 GB, with a download size of approximately 62.3 GB. It is designed to support natural language processing tasks such as question answering and document retrieval.