jrast/clean-text-v1
收藏Hugging Face2026-05-23 更新2026-05-31 收录
下载链接：
https://hf-mirror.com/datasets/jrast/clean-text-v1
下载链接
链接失效反馈官方服务：
资源简介：
Clean Text Corpus（清洁文本语料库）是一个大规模文本数据集，包含5,064,609个文档，总计36,485,673,610个字符。数据来源于`s3://jrast-10605-f26/clean_text_v1/`，原始数据为清理过的HTML parquet格式，后重新打包为zstd压缩的parquet分片文件。
Clean Text Corpus is a large-scale text dataset with 5,064,609 documents and 36,485,673,610 characters. It was repackaged from cleaned HTML parquet into zstd parquet shards, sourced from `s3://jrast-10605-f26/clean_text_v1/`.
提供机构：
jrast
