leighton

@FelixCLC_ Can I interest you in cool shit on a giant ass wafer

if we are mutuals and you are building a cool thing, pls tell me about it

@chopstickfury01 @NLRG_it It is exactly equal to one

@NLRG_it Isn't the answer just no because it's not 1?

scientific american, ladies and gentlemen

This is my hobby

@marctorrence Congrats Marc!

Get ready to learn 2016 Alabama defense highlights, buddy

NVIDIA's biggest GTC announcement was a $20 billion bet on the same problem we solved 6 years ago. Their next-gen inference chip - not available yet - has 140x less memory bandwidth than @cerebras. To run a single 2 trillion parameter model, you need 2,000+ Groq chips. On Cerebras, that's just over 20 wafers. Even paired with GPUs, Groq maxes out at ~1,000 tokens per second. We run at thousands of tokens per second today. And every day. In production now. Why? When you connect 2,000 chips together, every interconnect has latency. Every cable has overhead. It doesn't matter what your memory bandwidth is on paper if you're bottlenecked by the wiring between thousands of tiny chips. We solved this with wafer scale. One integrated system. Little interconnect tax. Jensen told the world that fast inference is where the value is. He’s right - it’s why the world’s leading AI companies and hyperscalers are choosing Cerebras.

When Paul R Ehrlich was born, global average life expectancy was less than 40 years. When he died a few days ago at 93, it was 73.2 years and rising. When he was born, 31% of children died before they were five. Today, that figure is 3.7% and falling. Yet he spent most of his time on this planet telling us that this is a bad thing, and that things are getting worse. All of his prophecies failed. Yet the bad science and bad arithmetic underpinning his claims continued to be used by the weirdo UN environmental agencies, global environmental movement, and the Guardian -- the most spectacularly debunked pseudoscience in history continued to arm the arguments against making life better, and to deny people the opportunity of taking part in markets and industries, and enjoying health and wealth. The human consequences of it are incalculable. Yet such movements founded on his grim work believe that they are the nice guys who want to make the world a better place.

@Midwest_expat @politicalmath Borlaug won the Nobel Prize in 1970 — the green revolution was already well underway at the time Ehrlich wrote Population Bomb!

@politicalmath Ehrlich might’ve been right if it wasn’t for Norman Borlaug single handedly saving the world from famine

The only good thing I can say about Ehrlich is that I'm glad he lived to 93 I'm glad that he spent the last 30 years of his life knowing that he was wrong about everything. I'm glad no one thanked him for his work and many people blamed him for the misery he created.

@powerbottomdad1 Wow really not cool to leave off Birmingham

S - ny, sf A - san diego B - chicago, boston, dc, detroit, austin C - seattle, miami, pittsburgh, denver, vegas, atl D - houston, dallas, s.a, kansas city, portland, louisville, nashvile, LA, minneapolis, charlotte E - philly, phoenix, ohio, indianapolis, stl, jacksonville, tampa

@eigenrobot Extraordinary song lyric pull

The visions dancing in my mind The early dawn, the shades of time Twilight crawling through my windowpane

@Duderichy Tu’s Introduction to Manifolds

what textbooks in swe/math/physics are the most worth reading?

@Duderichy Let Us Now Praise Famous Men

What's some good nonfiction I should be buying?

@techbromemes a nice side-effect of agents is me getting decent with the terminal

@blelbach You have one of the coolest apartments in the history of the world

It's white in NYC.

@0xmer_ @FelixCLC_ If you’re interested in CSL and the Cerebras SDK let me know :)

@FelixCLC_ My opinion might very well change as I learn and do more with regards to cuda and ML kernels. I've taken a brief look at CSL as well and I'm quite intrigued.

Programming accelerators is a lot more enjoyable than programming CPUs

Still absolutely entranced by this song. What a picker. I can play a mediocre rendition at about a third the speed Béla does, but this is absolutely number one on my list of "banjo tunes I want to master before I die"

@eigenrobot Maybe none of it matters anymore but it’s like learning a foreign language or an instrument. It matters to me

@eigenrobot I’ve been fortunate that what I learned in my PhD has been somewhat useful in my working years, but more than anything it’s given me the foundation to effectively learn more math as an adult hobby, which is very nice

the amount of regret that i have about grad school has been growing lately. so many working years wasted and i don't think what i learned is as pleasant or as helpful as i once did. not sure what to do with this awareness. how's everyone else coping

@FelixCLC_ We are in the golden age of cool freaking hardware

Cool people working with cool people. The work NS is doing continues to fascinate me

@nikicaga Disappointed that this list doesn’t include sand mountain Sam

Georgia's groundhog is named WHAT