Why does ChatGPT work so well? Is it “just scaling up GPT-3” under the hood? In this 🧵, let’s discuss the “Instruct” paradigm, its deep technical insi

Why does ChatGPT work so well? Is it “just scaling up GPT-3” under the hood? In this 🧵, let’s discuss the “Instruct” paradigm, its deep technical insi