The "tiling" perspective explains a lot of the common problems with SAEs for explaining concepts inside neural networks www.goodfire.ai/research/can...
Goodfire has a new thread on neural geometry! I'm sold on steering along manifolds after this project. Agenda post: www.goodfire.ai/research/the... Paper on SAEs tiling manifolds: arxiv.org/abs/2604.28119 Paper on manifold steering: arxiv.org/abs/2605.05115
The World Inside Neural Networks
How neural geometry will unlock understanding and control of AI
goodfire.ai
Humans and LLMs think fast and slow. Do SAEs recover slow concepts in LLMs? Not really. Our Temporal Feature Analyzer discovers contextual features in LLMs, that detect event boundaries, parse complex grammar, and represent ICL patterns.
What's the right unit of analysis for understanding LLM internals? We explore in our mech interp survey (a major update from our 2024 ms). We’ve added more recent work and more immediately actionable directions for future work. Now published in Computational Linguistics!
🚨 Registration is live! 🚨 The New England Mechanistic Interpretability (NEMI) Workshop is happening Aug 22nd 2025 at Northeastern University! A chance for the mech interp community to nerd out on how models really work 🧠🤖 🌐 Info: nemiconf.github.io/summer25/ 📝 Register: forms.gle/v4kJCweE3UUH...
Can we uncover the list of topics a language model is censored on? Refused topics vary strongly among models. Claude-3.5 vs DeepSeek-R1 refusal patterns:
Announcing ARBOR, an open research community for collectively understanding how reasoning models like openai-o3 and deepseek-r1 work. We invite all researchers and enthusiasts to this initiative by @wattenberg.bsky.social's and @davidbau.bsky.social's lab. arborproject.github.io
Addressing key concerns about AI competition. darioamodei.com/on-deepseek-...
Dario Amodei — On DeepSeek and Export Controls
On DeepSeek and Export Controls
darioamodei.com
The #38c3 Chaos Computer Conference was a blast! 🚀 Find the accompanying code for my intro workshop on activation steering in the thread.
Sparse Autoencoders (SAEs) are popular, with 10+ new approaches proposed in the last year. How do we know if we are making progress? The field has relied on imperfect proxy metrics. We are releasing SAE Bench, a suite of 8 SAE evaluations! Project co-led with Adam Karvonen.
More big news! Applications are open for the NDIF Summer Engineering Fellowship—an opportunity to work on cutting-edge AI research infrastructure this summer in Boston! 🚀
Safe travels to #NeurIPS2025 in Vancouver BC! Join our poster sessions on *Measuring Progress in Dictionary Learning with Board Game Models* and *Evaluating Sparse Autoencoders on Concept Erasure Tasks*. Reach out brainstorm future interpretability benchmarks.
