𝚃𝙷𝙴 𝚆𝙾𝚁𝙻𝙳 𝚈𝙾𝚄 𝙶𝚁𝙴𝚆 𝚄𝙿 𝙸𝙽 𝙽𝙾 𝙻𝙾𝙽𝙶𝙴𝚁 𝙴𝚇𝙸𝚂𝚃𝚂

taking a break from this hell site you can find me on discord i love you bye

my tl is absolutely bonkers rn i genuinely cannot comprehend how this content actually exists.

i have bad news seeing early signs than LLMs can be funny you just have to give them a weird medium that isn't just text

As far as I can tell most of these steering neurons, just put the model into a mode collapse state. Are there any that are particularly coherent? or is it more about the amount you steer them

Just vibe coding the most important game-dev market research tool ever, no biggie.

Guys we found the Genghis Khan neuron

Last year I started using Aristotle for Lean proofs in mathematics and this was a big shock how it triggered a novel way of doing mathematics. Code is the next stage, because AI produces even more code than mathematics and this is much harder to verify than anyone would expect. Really cool to have this feature in Aristotle, especially that @HarmonicMath probably did it better than anyone before. Curious to try, congrats!

my tl: - racism - cats - 🍇

chase the target

Me and the boys using AI in the Oil and Gas industry

Two kinds of people Type one thinks the speed of the future is dictated by physics and speed limits Type two thinks the speed of the future is dictated by the structure of the computational universe The strongest view is something in between Diff people need diff corrections

there was a point before deep blue, before alphago, before navier stokes, where you could argue, if not entirely convincingly, that maybe some areas of human endeavor were already skill saturated enough that more intelligence wouldn't help much I think there is basically no shred of credibility left for this kind of thinking now on anything where a primary input is intelligence. we know too much, we have seen too much. the universe has a high skill cap

They cannot immediately release the lean code because it would be too dangerous You gotta check it first - thank god for these shepherds because without them we’d never figure out if the lean definitions were correct it wouldn’t take our Opus agents like 15 freaking minutes or anything

This is f*cking insane. SpaceXAI senior engineer (ex-Cursor): "99% of people open one GrokBot, ask it a question and close the tab I run 50+ GrokBot agents like a company. A Chief of Staff runs my day, 3 PMs hand out the work, 40+ workers ship it overnight, and I just check the results over coffee. That team does the job of 20 engineers" In this 1-hour workshop, SpaceXAI builds that GrokBot team from scratch, step by step GrokBot → Chief of Staff → PMs → 40+ Workers → Autonomous Company At SpaceXAI, 97% of engineers already run a team like this, and the ones who do are getting $750K+ offers. The rest are about to compete with them Worth more than any $5,000 agentic engineering course Bookmark it and watch today Then read how to build your own team of GrokBot agents in the harness engineering guide below ↓

The number of posts I see coping about new jobs coming actually has me really concerned for humanity's collective mental health. Yes it actually IS different this time. AI isn't the 2000 bubble or the 2008 crash or COVID. Sometimes things do happen. The world you know is dead.

You can get your hands really dirty with AI proofs and feel the adiabatic. But its a state the mind can’t maintain for that long. Its super lubricated ideation but the moment you lose balance its infinite friction. Train to maintain the comath adiabatic

It is ~clearly~ now career viable for a single mathematician’s ambitions to exceed what was once considered a collective lifetime effort for an entire field’s community. In the past, overly ambitious mathematical programs were advised against, unless one planned to join a collective effort. Now one can reasonably assume that an individual mathematician’s capabilities exceed that of entire mathematical communities just a year ago. 1000x is a conservative lower bound. 10^8x is an ideal upper bound set by the timing ratio of transistor to neuron firing

Very soon, if ABC can't be autoformalized after FLT and this, then it will be a clear signal.

Nobody's going to train neural networks anymore. Just solve MAX-SAT and skip to the optimal circuit.