Conrad Close retweeted
One reason you don't want to claim your candidate is "God's candidate" or your party is "God's party" is that when that candidate or party loses, it becomes really apparent how small you've made God in order to support your own political opinions and aspirations.
Conrad Close retweeted
give a man a fish and you feed him for a day
teach a man to Claude Code and you financially ruin him with distracting side projects for the rest of his life
Conrad Close retweeted
Tom Brady: "No, my goodness. What is that?! There's no way!... That's definitely not roughing the passer!"
Fox rules analyst Dean Blandino: "That's teach-tape tackle. That's not roughing the passer." 🏈🦓🎙️
Conrad Close retweeted
The Packers offense is just
1 yard run
-1 yard run
20 yard pass to the most open receiver you’ve ever seen in your life
Conrad Close retweeted
Zaire Franklin (@ZiggySmalls_) says Jordan Love is the reason he chose Green Bay as his trade destination
"I told him straight up, you the reason I'm here bro! He's a special player and I'm glad he's on my side"
LIVE NOW ➡️ youtube.com/watch?v=Ps5AEi0Z…
Conrad Close retweeted
“So you spent $6 billion dollars on a baseball team?”
“Yes, Dave.”
“And you lost to the Brewers in the playoffs?”
“That’s correct, Dave.”
Conrad Close retweeted
genuinely an amazing day
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
already-clean.goodfire.com/
Conrad Close retweeted
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
already-clean.goodfire.com/
Conrad Close retweeted
MarShawn Lloyd WIELDS THE SCEPTER
😡ANGRY RUNS😡
The Packers sent some of the best scepter pics I’ve ever seen…
