beyond/chinese_clean_passages_80m

chinese_clean_passages_80m 包含8千余万（88328203）个纯净中文段落，不包含任何字母、数字。Containing more than 80 million pure & clean Chinese passages, without any letters/digits/special tokens. 文本长度大部分介于50~200个汉字之间。The passage length is approximately 50~200 Chinese characters. 通过datasets.load_dataset()下载数据，会产生38个大小约340M的数据包，共约12GB，所以请确保有足够空间。Downloading the dataset will result in 38 data shards each of which is about 340M and 12GB in total. Make sure there's enough space in your device:) >>> passage_dataset… See the full description on the dataset page: huggingface.co/datasets/beyond/chinese_clean_passages_80m.