opendatalab/SlimPajama-Meta-rater-Cleanliness-30B
收藏Hugging Face2025-06-14 更新2025-07-05 收录
下载链接：
https://hf-mirror.com/datasets/opendatalab/SlimPajama-Meta-rater-Cleanliness-30B
下载链接
链接失效反馈官方服务：
资源简介：
这个数据集包含了从SlimPajama-627B语料库中根据PRRC框架的清洁度维度筛选出的顶部30B个token。每个文档都经过一个基于ModernBERT的评分器的评分和过滤，该评分器经过微调，用于评估文本的格式、完整性和噪声或无关内容的缺失。数据集涵盖了CommonCrawl、C4、GitHub、书籍、ArXiv、Wikipedia和StackExchange等领域的文本。
This dataset contains the top 30B tokens from the SlimPajama-627B corpus, selected based on the Cleanliness dimension of the PRRC framework. Each document in this subset is scored and filtered by a ModernBERT-based rater fine-tuned to assess the formatting, completeness, and absence of noise or irrelevant content in the text, covering domains such as CommonCrawl, C4, GitHub, Books, ArXiv, Wikipedia, and StackExchange.
提供机构：
opendatalab
