The Cost of Scaling a Large Language Model

GPT-4 is a beast in its own and it does its own thing. It's very hard to find something in the open source world right now that is as good as GPT-4, right? But I feel like what is going to start happening more and more is this idea of, hey, smaller models leaning on vector databases more. We're just asking at these things that we know it can do. And so when a neural network, for example, says the queen is dead or alive, humans kind of have this inherent built in maybe feeling or knowledge of like time and like death. So you tell the language model and it's a statistical model. So it says,

Play episode from 14:16

Transcript

Episode notes

The AI-powered Podcast Player

Save insights by tapping your headphones, chat with episodes, discover the best highlights - and more!

Get the app