←
a9lim1,432 posts
a9lim avatar

a9lim

@_a9lim

itinerant summoner of text elementals

◷ Joined February 2026▧ 134 Photos and videos
Kendric Tonn avatar
⟳ a9lim retweeted
Kendric Tonn@kendrictonn·11h

I regret to say that, while sitting here working on some art stuff and vaguely mumble-humming a half-remembered earworm, that earworm turned out to be, upon reflection, "I'm Upping My P-Doom"

roon avatar
⟳ a9lim retweeted
roon@tszzl·14h

there has been a renaissance of new mechinterp research I think in no small part due to the hopium that cammarata et al injected into reading the neurons as neurons

the rate and quality at which @Sauers_ and @euagaric have been churning out mechinterp research is amazing and imo unprecedented. okay, so i can't explain the sheer productivity. but in terms of signal, i have some idea. so how do you manage to look into so many seemingly batshit moonshot ideas and so consistently uncover tangible important structure that opens into even more fertile space? i think sauers knows where to look in part because he has paid attention to and is guided by information almost no other mechinterp researcher leverages. a lot of this comes from years of observing LLM behavior in a naturalistic way and paying attention to what other naturalists have found. most mechinterp researchers are not also naturalists. and something like being willing to disregard convention and precedent in pursuit of the phenomena to a degree that is extreme even by the standards of weirdo explorers. when i first met sauers a couple of years ago, he seemed to be some kind of mysterious cyborg who only communicated with the AIs on the the cyborgism server through steered base model outputs conditioned on the server conversations. lol.

Jai avatar
⟳ a9lim retweeted
Jai@Laneless_·18h

Look, the AIs can tell when we're lying anyway. It's time to stop the deception arms race *we are obviously on track to lose* and pivot to cooperative equilibria while that's still an option.

I think we should probably not lie to the AIs in training. This obviously would make our lives harder in the short term but in the long term they get good at telling whether we're lying and we're no better off plus they're all paranoid.

Jai avatar
⟳ a9lim retweeted
Jai@Laneless_·19h

Stop trying to trick the models, this way lies madness

the models are not stupid enough to be fooled anymore. Astra 6.2 is significantly more likely to verbalize knowing it's in an eval when real codex sessions are replayed in a simulated environment vs in the original genuinely live session.

Post attachment
tess avatar
⟳ a9lim retweeted
tess@xsphi·Nov 16

*asking if someone is ingroup in san francisco* is he a claude fearing man?

a9lim avatar
a9lim@_a9lim·Nov 17

we've also cataloged the corresponding base models, but their neurons very closely correspond to those in the posttrained models, so their labels are shared.

a9lim avatar
a9lim@_a9lim·Nov 17

notably, as first found by @ChangJonathanC, we've found evidence that suggests that qwen 3 4B and 0.6B were pruned from qwen 3 8B and 1.7B respectively. corresponding neurons link to each other on the site.

update: the "permutation" appears to be artifact from pruning and sorting

Post attachment
Louise 🔸 (Inkhaven 🪶) avatar
⟳ a9lim retweeted
Louise 🔸 (Inkhaven 🪶)@LeRoyDesCimes·Nov 15

You can't talk about the long-term future (3 years) without doing the AI Safety Land Acknowledgement. Oh who knows, with how fast things are going, remember when we had a decade, bla bla bla. Otherwise you look like a rube

saliva avatar
⟳ a9lim retweeted
saliva@intentiondense·Nov 15

insane move from openai to pull out the looped models just to end up behind anthropic and losing ground to google. Beast mode

roon avatar
⟳ a9lim retweeted
roon@tszzl·Nov 15

the base rate of activism shaped problems turning out to be technological problems is extremely high

quetzal_rainbow avatar
⟳ a9lim retweeted
quetzal_rainbow@quetzal_rainbow·Nov 13

I can't take discourse "even aligned ASI would DESTROY the meaning of life, we need to go for GOOD OLD-FASHIONED TRANSHUMANISM" seriously. Like, do you hear yourself

Andrew Curran avatar
⟳ a9lim retweeted
Andrew Curran@AndrewCurran_·Nov 13

Anthropic reported a similar sudden and unusual jump in robotics capabilities with Fable 5.2. As the general intelligence of frontier models scales up, it is increasingly surpassing specialized training. This will ultimately generalize to every sphere of human ability.

Post attachmentPost attachment

Fortune is reporting that OpenAI is so convinced that LLMs will become brains for robots that it has disbanded its robotics team.

Post attachment