3 ms·
what does the first L mean in LLM?
by yieldcrv 2mo ago
what does the first L mean in LLM?
- janalsncm 2mo agoFwiw people have told me that GPT2 doesn’t qualify as an LLM at 550MB despite being one of the first LLMs. So the practical answer to your question is: not much.
- kube-system 2mo agoI don't think some people are aware that "large" has always referred to the training inputs not resulting the size of the model. LLM is not "we made a language model and it is big" -- it's "we trained this model on a lot of language"
- yieldcrv 2mo agoNot aware of that etymology
- rshemet 2mo agoit stands for Lets-not-be-sarcastic :)