Full threadjafitc·"People are really bad at understanding just how big LLM's actually are. I think this is partly why they belittle them as 'just' next-word predictors"https://nitter.net/jam3scampbell/status/1748200331215835561View on HN