Long post about my thoughts on this: I'm so excited that this is finally out! Optimizing this until scaling it out is actually feasible and building the systems to perform the scaling has been my job for the last couple of months. This work is some of the most deserving of my time and energy that I can imagine. There's been a lot of Discourse in the interpretability community lately about the neuronal and featurizer paradigms (not enough discussion of the causal abstraction paradigm imo..), the LRH and superposition, etc but no matter what paradigm we find ourselves in (and remember they can co-exist and describe different phenomena!) it's paramount that we accelerate the science of how models handle a wide array of tasks, how compositionality is constructed through intersection and containment of internal machinery (whether neurons, circuits, subspaces, or whatever else). Our current approaches have been too isolated to specific task and model combinations, but now that we have proven methodologies we can begin the process of "normal science" by applying them to population studies. Once we've proven this out, it will be time to scale the effort beyond even the model-specific setting. This is where things get really interesting! Given how many different axes models vary on, we need to see what findings hold across which axes (model size, architecture, attention schemes, etc) if we are ever to develop "laws" of interpretability. Only then can we start to apply mechanistic alignment techniques broadly across the industry and have a reasonable expectation that it will work. Personally, I'm quite optimistic based on our early results. I firmly believe that this is a solvable problem, but it won't be easy and requires an enormous amount of domain expertise (I'm so grateful to work with such talented researchers) to keep the agents from slopping it up. It will require an even more enormous about of resources, which is why it has to be an effort of the community at large. When we're ready, I hope that we can rally everyone who has the drive to solve these hard problems to use our encyclopedia and to help us fill it out.

Interpretability is the most important scientific problem of our lifetime. But it's too slow right now! I wrote some thoughts with @banburismus_ on how we might accelerate it. (Probably a good time to share that I'll be joining @GoodfireAI post PhD. I'm over the moon about it🌝)

Yes, exactly right! You can actually see this in Figure 8 of our "Why Larger Models Learn More" paper. @EkdeepL made this great visualization of that figure:

The wonderful “Don’t just Fix it in Post” paper identifies predicting memorization as a key open problem. A progress report: we can now accurately predict verbatim memorization using just three well-understood variables. Experiments by Jing Huang, using the Hubble model suite:

I’m still working through it, but this work is incredibly cool and (unless there’s some kind of serious mistake) the lineage of work including this and “Why Larger Models Learn More” is exactly the kind of work that I think is necessary to develop a science of DL. I’m a huge fan

We revisited Independent Component Analysis (ICA), a decades-old statistical method, and found that it works far better for interpreting LLM activations than we expected! This old baseline deserves a second look. Across GPT-2 Small, Gemma 2 2B, and Qwen 3.5 9B Base, we find ICA is actually competitive with the public SAEs across interpretability, sparse reconstruction, concept prediction, and targeted probe perturbation, while fitting in minutes per layer and using 7–43× less parameter memory. We're releasing ICA Lens, which lets you quickly discover, annotate, and evaluate interpretable directions in LLM residual-stream activations. No need to train another dictionary! More broadly, our results suggest that non-Gaussianity is a useful inductive bias for efficiently finding interpretable directions. Try it: pip install icalens pypi.org/project/icalens/ Paper: arxiv.org/abs/2606.11722v2 1/8

Super hyped to work together! 😁

I was asked by a journalist to comment on this piece. The comments I wrote up (on Friday evening) are included below, so that I have a public place where my view is fully represented. 1. I appreciate the critical reassessment of some of the existing arguments in favor of superposition. It does look like alternative explanations may be viable. 2. I should say that, despite being an author on perhaps 40 papers on interpretability, I have never really worked on the superposition hypothesis per se. There's a lot more to interpretability than this one idea! We did, though, develop a benchmark (RAVEL) and associated metrics for assessing the extent to which features in neural networks can be disentangled: aclanthology.org/2024.acl-lo… We found that many neurons are polysemantic, but we made no claims about the more specific notion of superposition. It would be great to see RAVEL results for newer models. 3. A fundamental motivation for interpretability research is preparing for an uncertain future full of strange new models with surprising new capabilities. Superposition is a technical insight about how neural networks can learn to pack unrelated features into a single neuron. Even if present-day models rarely do this, we need to be ready for future models. Cammarata's analysis seems to trace legibility in part to architectural choices. Future architectures might be very different. If interpretability researchers don't prepare us for that, who will? 4. Present-day models may make use of superposition as well! Cammarata suspects that they don't, or do this so rarely that we could even ignore the actual instances ("doesn't matter for study"). I would characterize this as a hunch based on a lot of initial heuristic empirical work. In the normal course of scientific research, this would be the start of a process of figuring out precisely how to rigorously test this as a hypothesis. 5. For example, the piece has a nice figure that Cammarata characterizes as indicating that a particular neuron has positive activation on pointed things and negative activation on round things. Cool! Let's define some independent criteria for those classifications so that we can test this objectively. (It is possible that the paper Cammarata is preparing will include these evaluations.) 6. The piece expresses the intuition that interpretability is easier, in some sense, with better models. This intuition has been very common among researchers for years. At the same time, I understand why @JacobHHilton would say the opposite: nitter.siddz.com/JacobHHilton/status/21… 7. The piece seems to allude to a technical proposal, but it does not give the details. As far as I can tell, the ideas have substantial overlap with this work (led by two former students of mine): Arora, Wu, et al. 2026 (arxiv.org/abs/2601.22594). Cammarata does cite this. I am guessing there could real differences that will be interesting to see. I would hope that Cammarata explains the differences and reports comparable evals to Arora et al, at least for language models. I appreciate that he looked at "thousands" of examples (the essay says this a lot), but even a large number here is not a substitute for rigorous evaluation. 8. In my research group at Stanford, we are working on lots of very hard problems in interpretability. As far as I can tell, nothing about this blog post makes these problems any easier, and certainly none of them are solved by this work. It would not have occurred to me to say this at all, since it is not even reasonable to have this expectation, but very prominent people on social media seem to believe that it is an all-encompassing breakthrough. Some immediately extrapolated to an optimistic story about AI safety that is not supported by anything in this piece. (Highly interpretable agents could of course do dangerous things.) 9. It must be said that Cammarata's framing did encourage these reactions. The header of the post page says, "An informal research essay and working hypothesis about neural networks", which suggests something quite preliminary and tentative. The piece also says, "Highly speculative: this might be most of what we need, and interpretability might be far more tractable, possibly in the near term, than we've assumed". However, his announcement tweet (nitter.siddz.com/nickcammarata/status/2…) reworked this considerably: "I now think ambitious interpretability will be solved by the time we need it most". Strictly speaking, this commits to very little beyond optimism, but it is written to invite many inferences.

Something I find strange about parts of the ML community is how we entertain the idea that the systems we're building might be fundamentally beyond our understanding. This is very different from the perspective I'm used to in physics and especially in biology (1/)

Super exciting results that offer a viable hypothesis for how axis aligned, interpretable neurons emerge!

to be clear ai alignment hasn’t been solved so don’t go around saying that

it’s been a dramatic couple of days, I didn’t expect the already clean essay to be read so widely. I've heard more interesting discussions about mechanistic interpretability – should we think mostly in features or mechanisms and the flow of data through them, how much do weak activations matter, how does scale change things, and lots of other questions – than I have in a long time and I’ve been happy to hear all the discussion! a couple things on the essay itself. it didn't have enough citations, only ones for the basic story, since it was an "informal research essay and working hypothesis" rather than a paper, and other prior work around neurons were in the technical paper we're still finishing up. but it blew up, so I've added citations to the essay as well. I also want to thank @curttigges who did a lot of the language neuron research with me finally, while there was quite a range of updates, from this being incremental and obvious to interpretability going from hopeless to fully solved, I want to try to be clear about roughly where I stand I used to think neural networks would take some big cleaning step and then we could begin the ordinary natural science of mapping the parts and how they connect, watching data flow through, mapping circuits, and so on. at various points, like late layers of early model organisms and most layers of language models years ago, this didn't really work. but now with better tools and better models of study, it works much better for smaller vision models, this normal science process worked surprisingly well, until the whole map started to be broadly filled in. For bigger modern language models, it seems promising so far, for both neurons and circuits, attribution graphs, basically everywhere we've glanced. the map isn't filling in, there's a lot of map. there's still off-target firing, neurons that do a few things, but so far it's not an incoherent mess that you can't explore and learn from, and over time we get better at learning how to see it. so maybe it keeps going, and mapping motifs and exploring actually gets you pretty far, maybe something like "all the way" to a reasonable solution. I don't think it would be so scientifically surprising if that were the case. everything in the neural network was built in some sense to be read – by its downstream readers – and if we can learn to look like a reader, we might find reading them relatively easy. many researchers I talk to outside interpretability assume the parts are already understandable, and the difficulty is just in the mapping stage! it was also my prior when I first started looking at neurons when I first joined the field. there's lots of highly interpretable ones, maybe that's just how neural networks are if you look right. even in the good world it would still take some iteration on tools, on ways of looking, neuron-attention interactions are tricky for instance. there's probably no bright red line separating "cleaning the model" from "learning to look better" but there's broad working principles, and as of now I think we should try to look relatively directly if we can it shouldn't be so hard to test, you can just start looking at things and iterating and trying to understand motifs and such and see if it just breaks down at some point. it can take some iteration, I would recommend a bias against assuming arbitrary mixtures when you don't understand something at first, but also explanations should be tested for now I'm pretty optimistic, and I'm glad there's excitement about interpretability, but too much can cause problems too. I think part might be that ai safety isn't a very hopeful field, and even moderate amounts of optimism about an important problem being solvable in the future fills a spot that has been malnourished for a while. across several safety disciplines, the researchers directly working on them are often more optimistic than the people broadcasting pessimism more broadly for neural network interpretability at least, these are extraordinarily experimentally accessible objects of science, I don't know of a better one, and we have automated scientists. we have many things going in our favor. it's important and it's fun. the circuitry of these machines can be beautiful, surprising, entertaining, and insightful. I've spent thousands of hours looking through them, asking questions and running experiments and testing things, and would do so even if it weren't so important. I think the problem is solvable, whether the simple thing works, it mostly works but we need some new things, or we need big new ideas. I think we can figure it out

working on this with Nick was easily the most fascinating and rewarding project I've done in interp so far, and we think we have excellent news about the interpretability of neurons in neural networks. neurons are far more interpretable than we thought, and the bottlenecks were primarily our methods and preconceived notions about superposition and intractability rather than fundamental entanglement

Excited to share this new result on EVEE, interp generated hypothesis actually reclassifying VUS. Want to share a bit about how this happened 🧵 (1/n)

How much representational diversity exists around a trained network? A lot! Moving a ViT’s weights by 1.3% while preserving its predictions, its representations end up less similar to the original than a random-init ViT’s ‼️ New paper: Hessian Null Space Continuation (HNC) 💫🧵 (1/12)

Our boy @RyoYbioinfo and the broader life sciences team at Goodfire cooking

🚀 AI Deep Dive Episode 5: Can We See Inside an AI’s Mind? AI can write down its reasoning. But what happens when more of its thinking never makes it onto the page? @rocketalignment and @GoodfireAI co-founder and chief scientist Tom McGrath explore what AI “brain scans” can reveal beyond written reasoning, how internal signals could help catch cheating, and what it would take to understand—and change—what models learn. 00:00 Intro 00:39 Tom McGrath and AI brain scans 09:01 Written reasoning and hidden computation 17:45 Why AI is hard to understand 24:07 How researchers identify concepts 31:10 What interpretability could change 34:42 Reverse engineering a language model 43:09 The field’s approaches and investment 51:51 Changing models without hiding problems 1:00:10 AI agents studying AI

Goodfire safety infrastructure 🤝 @Baseten inference Today we're partnering with Baseten to bring real-time safety monitoring to open models. This is an important step toward our vision of ensuring every capable model has a frontier alignment stack.

We discovered a new class of Alzheimer’s biomarkers with @PrimaMente. Today we're taking the next step by opening a global AI competition to discover new treatments. Introducing the Alzheimer’s Translation Challenge, led by @PrimaMente and @AlzData. 🧵

SOTA on AntiFelonyBench 🚔