interpretability might be the most important scientific problem in the world right now, as I don't think we can solve alignment if we don't know what we're building. I think I have good news about it. I co-authored early work in mechanistic interpretability at OpenAI and came away pessimistic, because neural networks appeared to fold into arbitrary tangled knots. years later, I started openai's first large-scale automated interpretability project, finding about 2% of neurons legible with the primitive tools we had
with modern agentic tools and a much closer look, the picture looks totally different. the vast majority in models up to trillions of parameters are legible, and their circuits beautiful. many famous examples of tangled neurons turned out to be misunderstood. The roadblocks I expected mostly aren't there. I now think ambitious interpretability will be solved by the time we need it most, and possibly quite soon, as automated scientists keep improving. it may be a bitter lesson for interpretability: the simple thing, at absurd scale, works
https://t.co/zHij5zwKGT
we kept it a secret at leap motion that i filmed and edited all the videos myself - its been now ~10 years since i filmed and edited a video like this and i was honestly quite nervous that my skills weren't up to par for the task - but im very happy everyone seems to like it!
It's very frustrating that people see widespread cheap food delivery, a la China, as a sign of prosperity. It's the opposite!
Rich countries don’t have millions of people willing to spend their time bringing you lunch for basically nothing!