metadata
license: odc-by
The dataset was derived from OpenWebMath after cleaning about 1.65m garbled and non-mathematical documents.
You can find the detailed cleaning process on my blog.
license: odc-by
The dataset was derived from OpenWebMath after cleaning about 1.65m garbled and non-mathematical documents.
You can find the detailed cleaning process on my blog.