Datasets:
Questions about FineWeb data coverage and earlier web content
Hi FineWeb team and community,
First off, thanks for building and sharing FineWeb — it’s an incredible resource for web-scale research!
I’m currently working on a project that looks at historical web data and I have a couple of questions:
Does FineWeb include or plan to include any older web content (e.g., pre-2008) or is it focused on more recent Common Crawl data?
Are there recommended approaches or companion datasets you’d suggest for covering 2008–2013 (or even earlier) web pages?
If I wanted to combine FineWeb with other archives like Common Crawl or Internet Archive crawls, are there any known workflows, tools, or best practices the community uses for merging or deduplicating such large-scale corpora?
Any tips or experiences would be super helpful. Thanks in advance!😀
Best,
Patrick