I bet there are multiple copies of various depth by various groups including reddit themselves,. But how they will use it, wether they share it we cant really know.

Large chunks are contained within dataset for ai, we may be able to obtain or reverse engineer those.

The internet archive waybackmachine is probably the best most valid source for the public for now.