logoalt Hacker News

mattkrauselast Tuesday at 6:10 PM1 replyview on HN

It’s literally a trigram (character) language model. Check any NLP book from before 2015 or so.

LLMs have more stuff bolted onto them (embeddings, RLHF) but the autoregressive core is a direct descendent of that sort of language model.


Replies

justincliftlast Tuesday at 11:08 PM

So, LM vs LLM? :)