thng292/fineweb-subset-1M
收藏官方服务:
资源简介:
这是一个包含五种语言(越南语、英语、法语、日语、中文)的数据集,每个语言都有200,000个训练样本和10,000个测试样本。数据集的字段包括文本内容、ID、URL和语言类型。默认配置是所有子集的组合。
This dataset includes five languages: Vietnamese, English, French, Japanese, and Chinese, with 200,000 training samples and 10,000 test samples for each language. The dataset features include text content, ID, URL, and language type. The default configuration is a combination of all subsets.
提供机构:
thng292


