Excited to release new repo: nanochat! (it's among the most unhinged I've written). Unlike my earlier similar repo nanoGPT which only covered pretraining, nanochat is a minimal, from scratch, full-stack training/inference pipeline of a simple ChatGPT clone in a single, dependency-minimal codebase. You boot... See more

How to understand Github code repos with LLMs in 2 mins. Even after 10yrs of engineering, dissecting a large codebase is daunting. 1 Dump the code into one big file 2 Feed it to Gemini-1.5Pro (2M context) 3 Ask it anything. Here's, I dissect... See more

Llama 3 from scratch. Might have been written by @sama https://t.co/QYkkMkV5bq