@Scobleizer I’ve tried so many ways to use an llm with large amounts of data and it’s not at all easy to do with the current token limits. What’s really amazing is how few examples there are of people actually doing it. That just reinforces how difficult it is. If it was easy there would be… https://t.co/lVBDwCEMb0

— TJ Allard (@TjAllard) May 12, 2023