logoalt Hacker News

xp84 • today at 5:13 AM • 1 reply • view on HN

That would be a very exciting experiment. I doubt you could find enough vintage training data to train a LLM enough to do it without it being contaminated by anachronistic data in the weights, but I would be absolutely amazed if it could reproduce that.


Replies

anshumankmr • today at 10:10 AM

There are some folks trying the same thing as I mentioned. Can check the past posts on HN that talk about the same thjing