Probably just all of reddit. There are json dumps of all reddit posts and comments (up to 2022 or so), making it olive of the low-hanging fruit.
I wonder what LLMs would look like if they weren't able to be trained on the collective community efforts of Reddit + StackOverflow exports
That's a hot take
Most of what we talk about is either parroting information produced by somebody else or opinions about information produced by somebody else that always converge to relatively common speaking points.
Unique human content is pretty minimal. Everything is a meme.