A classic

painting > prompting excited for this to be public soon! more to share throughout the week

the latent space is vast (so much larger than a text box!) and this lets you literally paint with it. i haven't felt quite this way since i saw melody interpolation with musicVAE. a good a time as any to announce im joining goodfire! i'm really excited about what we're working on

"[Deep] unsupervised learning looked worse until suddenly it looked better. We think interpretability is likely to follow a similar arc... Each new item on the tech tree unlocks new questions, new ways to look at the problem. We're building toward the breakthrough"

do you guys like my coffee table book

New research update! We replicated Anthropic's circuit tracing methods to test if they can recover a known, simple transformer mechanism.

A few months ago, we published Attribution-based parameter decomposition -- a method for decomposing a network's parameters for interpretability. But it was janky and didn't scale. Today, we published a new, better algorithm called 🔶Stochastic Parameter Decomposition!🔶

(1/7) New research: how can we understand how an AI model actually works? Our method, SPD, decomposes the *parameters* of neural networks, rather than their activations - akin to understanding a program by reverse-engineering the source code vs. inspecting runtime behavior.

Has anyone used any good AI shopping assistants? I’ve tried Daydream, deep research but couldn’t get either to work for me. Maybe this is a sign to stop buying clothes :/

I knew Claude was from New England

happy July fourth from me and mine (my Llama SAE features)

Brutal roast from this UMAP of language model latents

we discovered the katy parity feature