👀 github.com/pytorch-labs/f…

Talk: Inverting Language Models (by dr. jack morris) youtube.com/watch?v=lguThu… Techniques for extracting text from vector databases and prompts from LLM APIs.

New blog post: how to make LLMs go fast! Want to understand how people are making LLMs go brrrrr? This post is a survey of lots of different LLM inference optimizations, ranging from "everyone uses this in prod" to "I cooked this up last week (but it seems to work)"

new blog post! played around w/ representation engineering, and released a new library for training control vectors in <60s! i train some useful control vectors, some… out there… control vectors, test them for jailbreaking, test them *against* jailbreaking, and more!

When a student sadly tells me that the idea we've been working on for weeks was just arXived, I say: "Great! We've just gotten *strong* confirmation that our thinking was in the right direction. We've had the initial work done for us. Lets figure out how to make this 10x better"

‼️⚠️bad day to be a LLM⚠️‼️ Haize Labs took one of our favorite adversarial attack algorithms, GCG, and made it *38x* faster

Yes, and get SOTA by doing so. x.com/vaibhav_adlakh…

This is the most in-depth technical report we’ve gotten about a frontier-class model in a while!

Contrast pairs are overpowered. Once you have them, you can use them to generate control vectors, and to initialize classifiers, and to do RL/DPO, and probably more

Why not just train your embedding model to embed queries directly into answer space? Feels like an inefficient use of flops.

there was a good thread recently estimating the current upper bound of tokens in the world (including Gmail and other private repositories), but I can't find it can anyone point me to it?

he can't keep getting away with it