haryoaw/wikipedia-processed
收藏Hugging Face2025-10-24 更新2025-10-25 收录
下载链接：
https://hf-mirror.com/datasets/haryoaw/wikipedia-processed
下载链接
链接失效反馈官方服务：
资源简介：
该数据集是一个多语言文本数据集，包含英语、法语、印尼语和中文四种语言的文本数据。每种语言配置都有大约300万（中文为2906163）个训练样本。数据集以不同的配置名区分，如en-512-3M代表英语数据集，包含大约932,321,009字节的训练数据。
The dataset is a multilingual text dataset containing text data in English, French, Indonesian, and Chinese. Each language configuration has about 3 million (Chinese has 2,906,163) training samples. The dataset is distinguished by different configuration names, such as en-512-3M for the English dataset, which contains about 932,321,009 bytes of training data.
提供机构：
haryoaw
