Bissell: Interpretability allows direct painting into a diffusion model's mental map
Mark Bissell · [State of MechInterp] SAEs in Production, Circuit Tracing, AI4Science, "Pragmatic" Interp — Goodfire · Dec 31, 2025 · at 3:47
Mark Bissell of Goodfire explains the technical architecture behind their paint.goodfire.ai demo using Stable Diffusion XL Turbo.
0:00 / 0:20exact quote · 20.3s
▶ Watch the full episode on YouTube → 720p mp4 · rendered on demand · StarZero watermark
“Using interpretability techniques, you can sort of like plug directly into the mind of the model, and you get a two D canvas where you can basically like paint directly into its mental map of the image. And so we used unsupervised techniques to basically figure out all the concepts that the model like has internally.”
quote is from the automated transcript, cleaned for reading: filler sounds and stutters are removed, nothing is rephrased. names can be misheard (the analysis reads context, assessments check outside sources). how →
More from Mark Bissell
Bissell: CCP bias is identifiable in Qwen and DeepSeek-R1 representation spaces
“Well, there's, there are certainly internal, yeah, parts of the representation space where you can sort of see where that lives.”
Bissell: Activation Steering and In-Context Learning Are Quantitatively Equivalent
“He actually has a paper that, as well as some, you know, others from the team and elsewhere, that go into the essentially equivalence of activation steering and in-context learning, and how those are from a, he thinks of everything in a cognitive neuroscience …”
Bissell: Probing internal model features matches LLM-as-a-judge quality at 500x lower cost
“If you ask that model, try to like use it as an LLM as a judge, it's not very good. But if you probe its mind and you sort of detect when the features related to personally identifiable information are firing, that gets you the highest recall of anything. It's…”
Bissell: Subliminal learning typically only affects models sharing initial random seeds
“I think it only applies to models that were initialized from the same starting Z. Usually, yes.”
Bissell: Goodfire performs activation steering on 1-trillion parameter Kimi K2
“Here you're going to see steering on a one trillion parameter model. This is Kimi K two.”
Bissell: Steering Experiments Can Predict Examples Needed for Jailbreaks
“What's in this in context learning and activation steering equivalence paper is you can like predict the number of examples that you will need to put in there in order to jailbreak the model. By doing steering experiments and using this sort of like equivalenc…”