
Louis
@dikaiosvne
Followers
387
Following
8K
Media
205
Statuses
5K
philosophy phd candidate | multiagent systems | ai | mind | social networks | geopolitics | classical philology | virtue theory | truth-seeking
San Diego, CA
Joined August 2015
The body you can chain, the mind remains forever free. Epictetus 1.1.21-25 (Interlinear Greek-English) by the Hermenaion & LS ["Hermenaion" is integral Claude Opus 4's suggested name for the multiagent Claude / o3 graph that produced this artifact]. 'Tell me your secrets.' I
to be antislop is to select from the space of things that could be said the things that should be said. one manner of making such decisions is to read everything and to say what you record. another is to design longer limbs of search, memory, and inference.
2
2
13
the disdain social elites have for the unwashed masses might be justified if they weren't themselves so dirty, and washed
0
0
0
they should change its name from "social media" to "you're about to get plagiarized"
0
0
0
i've never seen people go so hog wild for a stinker like they did over karpathy's "ghosts". ghosts of my hope for tomorrow maybe.
0
0
0
yud's book is the new "what computers can't do" except it's about "what they can..."
0
0
0
that's what would have happened if dwarkesh were a 1st year phd student. a good student learns to listen. dwarkesh did a service to himself and to the community by interviewing a great scholar. the correct response from the student is sincere gratitude, not entrenched denial.
Boy do you guys have a lot of thoughts about the @RichardSSutton interview. I’ve been thinking about it myself. I have a better understanding of Sutton’s perspective now than I did during the interview itself. So I want to reflect on it a bit. Richard, apologies for any errors
0
0
2
"anything that we can measure seems to be improving really rapidly."
Sholto Douglas (Anthropic): "Over the last year, RL has finally allow[ed] us to take a feedback loop and turn it into a model that is at least as good as the best humans at a given thing in a narrow domain. And you're seeing that with mathematics and competition code, which are
8
9
320
as someone who took philosophy very seriously for a very long period of time and trained my brain to be better at sussing out nonsense, what would happen if i went in there and took a long, scrupulous look at the epistemic practices and experimental validity of llm benchmarks?
0
0
1
accounts with a ton of followers in your niche are the gatekeepers to reach under the current algorithmic regime. the algorithm (anecdotally) weighs their engagement, heavily. one unfortunate consequence of this is that it stifles all criticism that doesn't happen on dwarkesh.
0
0
1
karpathy has good intuitions. if you agree and want to hear similar points several months in advance go ahead and drop me a follow.
Finally had a chance to listen through this pod with Sutton, which was interesting and amusing. As background, Sutton's "The Bitter Lesson" has become a bit of biblical text in frontier LLM circles. Researchers routinely talk about and ask whether this or that approach or idea
0
0
2
the best thing about a phd is the missed earnings. corruption is by definition money-chasing. after having gone without money, servitude to it at least becomes a choice. it's better to have integrity and not money than money and not integrity. also good to have both.
go to GRAD SCHOOL. You will SURELY NOT REGRET getting a PHD. it will make you feel COOL and MYSTERIOUS when you tell the guy at the gym that you do ALGEBRAIC GEOMETRY. It will only cost you NINE HUNDRED THOUSAND DOLLARS in missed earnings.
3
0
3