What are the most beautiful research blogs presenting technical work? I'm a big fan of how Anthropic presents their transformer circuits work. Interested in others.
Do reasoning models like DeepSeek R1 learn their behavior from scratch? No! In our new paper, we extract steering vectors from a base model that induce backtracking in a distilled reasoning model, but surprisingly have no apparent effect on the base model itself! 🧵 (1/5)
bayes Same bottlenecks as biology: Epistemology.
It's barely a science, and getting it to the point where it is one is closer to philosophy or programming language design/UX than it is physics or math.
Neural networks are grown, not programmed. What does that growth process look like? Like this!
This is a small language model (3M) across training, visualised with a new interpretability technique: susceptibilities. We call this handsome critter the rainbow serpent.
What’s going on inside large AI models?
Astera grantees Adam Shai and Paul Riechers are building a new theory of internal structure to better understand intelligence.
We sat down with them to learn more about their work as co-founders of Simplex, a research organization: