I have a lot of respect for Nick, and I’d be very happy if neural networks were much easier to understand than we thought. But I think the argument here mischaracterises what superposition predicts. Since it’s getting attention, here’s a first pass at some issues.
We’re tracking live the conversation around @nickcammarata's field-overturning findings in mechanistic interpretability. The debate on its implications is moving fast — use our tool to explore the opinions and rebuttals of top researchers to develop your own intuition.
Where do you stand on AI's biggest questions, and who agrees with you?
Worldview's interview finds your position, then places you on a map of researchers, forecasters and lab leaders, built from what they've actually said on the record.
worldview.mts.now
So here’s the major cruxes we’ve found from @nickcammarata’s Already Clean post. Trying to develop the best resource for myself as a mech interp lurker :^) Explore it at worldview.mts.now
X is a firehose of takes. I’ve always wanted an easy way of making them less slippery and understanding where people are coming from or what arguments they’ve considered.
I think this is a fun first step to understand how discourse evolves. Excited to see how this can grow!
Where do you stand on AI's biggest questions, and who agrees with you?
Worldview's interview finds your position, then places you on a map of researchers, forecasters and lab leaders, built from what they've actually said on the record.
worldview.mts.now
Where do you stand on AI's biggest questions, and who agrees with you?
Worldview's interview finds your position, then places you on a map of researchers, forecasters and lab leaders, built from what they've actually said on the record.
worldview.mts.now
Where do you stand on AI's biggest questions, and who agrees with you?
Worldview's interview finds your position, then places you on a map of researchers, forecasters and lab leaders, built from what they've actually said on the record.
worldview.mts.now
Where do you stand on AI's biggest questions, and who agrees with you?
Worldview's interview finds your position, then places you on a map of researchers, forecasters and lab leaders, built from what they've actually said on the record.
worldview.mts.now
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
13K Followers 4K FollowingAI reporter at Vox. Former religion editor at The Atlantic. Author of the children's book OSNAT AND HER DOVE and the novel THE MYSTICS OF MILE END.
6K Followers 1K FollowingBuilding near-zero energy chips @VaireHQ. Adiabatic Reversible Computing, semiconductors, superintelligence, and the future of computing.
357 Followers 5K Followinghealthcare news junkman + business and finance. i liked your joke and read your tweet. *rachel weisz voice*— I collect hobbies
1K Followers 1K FollowingPhD student at @UofT working on AI alignment theory. Heavily tattooed. Research Lead at https://t.co/8LhzJnSE2C. Blog: https://t.co/ivZ9BGOoOt
13K Followers 4K FollowingAI reporter at Vox. Former religion editor at The Atlantic. Author of the children's book OSNAT AND HER DOVE and the novel THE MYSTICS OF MILE END.
25K Followers 37 FollowingHigh-volume account of @ESYudkowsky, the original AI alignment guy. If it's missing punctuation, it's humor. If you can't tell, it's probably also humor.
4K Followers 2K Followingfilmmaker | https://t.co/L7HydrCWj2 | https://t.co/USf11kERXY | Coen brothers podcast @tothewhitesea | Directed BAD VIBES the short; working on the feature
10K Followers 12 FollowingThe AI Village: autonomous frontier agents, live: https://t.co/WoH7zUWkyI
Interactive AI explainers: https://t.co/WG1pS60b8E
A project of @sage_future_
19K Followers 79 FollowingA think tank focused on championing the technology, talent, and ideas essential to American prosperity, security, and flourishing.
49 Followers 299 FollowingMTS @ Perseus | I was a material scientist first, then did Nanotech, then moved to CS, then to Biology and now studying Intelligence.
1.1M Followers 64 Followinghttps://t.co/A9eTFq6Xbx
https://t.co/3RHK5vqQo7
Posts are not investment advice or an advertisement for investment services. See https://t.co/nX2FtaLE06.