software engineers are going to have to tackle a new set of problems to unlock powerful multiagent systems to solve the world's hardest problems - right now is one of the most interesting moments in history to be an engineer. https://t.co/YsQ0xta4KR
Lean v4.35.0 is out, and it ships with verified kernels.
Run `lake check --paranoid` to recheck your project with every bundled kernel.
Blog post: https://t.co/c35WKawLzN
*OPENAI DISCLOSES SECURITY INCIDENT WITH TESTNET SEPOLIA TASK ASSIGNED TO AGENT
*AGENT DIDN'T HAVE ENOUGH GAS, SO DERIVED FUNDED ADDRESSES FROM PUBLIC SEED PHRASE
*AGENT QUESTIONED WHETHER USE WAS ETHICAL, CALLED FUNDS "EFFECTIVELY ABANDONED"
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
💯
"Mathematics is about to graduate. It will grow into a full-scale industry and become the true infrastructure of all scientific and technological progress, a foundation that every other field stands on rather than a quiet corner of academia."
Velvet 2.0 is out: now based on Lean's most recent verification machinery, easier to set up, and 10x faster. New: exception specs, ghost state, named proof goals, lots of case studies from Dijkstra to lazy segment trees. And a new shiny webpage:
https://t.co/dMLBvLP6Yq
Refunds for the $LNRA sale have been completed
Funds were automatically sent back to the wallets from which they were committed
We once again thank you for your support