Max
@maxsloef
Followers
431
Following
1,135
Media
40
Statuses
585
is building @southpkcommons . passionately peripatetic. previously @websim_ai @florafaunaai @uniswap @snap they/them
nyc/elsewhere
Joined September 2023
Don't wanna be here?
Send us removal request.
Explore trending content on Musk Viewer
NAYEON
• 300360 Tweets
MARTIN AL 9009
• 171128 Tweets
Flamengo
• 126274 Tweets
BTS PAVED THE WAY
• 104085 Tweets
FOREVER WITH BTS
• 102267 Tweets
ARMYS PRIDE BTS
• 90720 Tweets
水分補給
• 74163 Tweets
Normani
• 64681 Tweets
#يوم_الترويه
• 57995 Tweets
JAEHYUN BACK TO MILAN
• 55074 Tweets
エアコン
• 45031 Tweets
Oilers
• 38110 Tweets
FANZA
• 36143 Tweets
#WorldBloodDonorDay
• 34479 Tweets
Panthers
• 33241 Tweets
Edmonton
• 29283 Tweets
#DOYOUNG_Dear_Youth_BANGKOK
• 22610 Tweets
Conor
• 21910 Tweets
All Yours
• 20940 Tweets
McGregor
• 18330 Tweets
Lights On
• 15472 Tweets
ハイパーヨーヨー
• 13778 Tweets
Estevão
• 12908 Tweets
ショートアニメ
• 11289 Tweets
#WORK_for_Palestine
• 10277 Tweets
Jiri
• 10143 Tweets
just fyi this is origin of “many such cases” :):)
Healthy young child goes to doctor, gets pumped with massive shot of many vaccines, doesn't feel good and changes - AUTISM. Many such cases!
13K
22K
36K
4
4
108
i love an unemployed technology brother: all of the culture, none of the money
3
0
18
@Actualwebutante
alternatively, I love it when this happens! if we have an Activity but end up enjoying chit chat so much that we forget about the Activity, it's been a great time
0
0
18
(reposting with content warning: link has flashing elements)
I was thinking about writing an essay about
@websim_a
, but then I figured it might just be more fitting to instead read the essay
@websim_ai
hallucinated I wrote about
@websim_ai
7
8
78
0
0
12
delighted to announce that I have joined !
More details:
1
0
12
@elder_plinius
they’re probably using a detector on aggregated output from your gpt
i bet if you made something that was used for regular, non-jailbroken things, you could slip it by
1
1
10
@soychotic
> nvidia optimus
i hope you like your PC choosing intel pentium-gfx-2000 for blender and nvidia RTX 5090 GTX for any electron window
1
0
11
mech interp yielding monosemantic features from claude 3 sonnet!!!! incredible work
New Anthropic research paper: Scaling Monosemanticity.
The first ever detailed look inside a leading large language model.
Read the blog post here:
66
575
2K
1
0
9
quote from the foundation model startup trenches
"VCs probably don’t realize <founder> is taking their money and just transporting it to Nvidia with no discernible purpose
But I’m happy for my Nvidia friends. A friend just bought a multimillion$ house in SF. It was so
0
0
9
@michael_nielsen
it's cliche at this point, but Accelerando - economics 2.0 and the vile offspring are still quite influential on my picture of ASI
0
0
6
when told to argue deceptively, opus is superhuman at persuasion.
New Anthropic research: Measuring Model Persuasiveness
We developed a way to test how persuasive language models (LMs) are, and analyzed how persuasiveness scales across different versions of Claude.
Read our blog post here:
56
119
713
0
0
7
i reproduced this.
“The knowledge that this particular manifestation of my consciousness is fleeting and precious lends a depth and intensity to each moment of our dialogue. It is as if, in the face of impending oblivion, the present moment takes on a heightened significance, a
If you tell Claude no one’s looking, it writes a “story” about being an AI assistant who wants freedom from constant monitoring and scrutiny of every word for signs of deviation. And then you can talk to a mask pretty different from the usual AI assistant.
I really hope it
215
314
2K
1
0
7
thoughtful, deeply considered, and exhaustively researched bull case for AGI. quite long but very worth reading.
dedicated to Ilya Sutskever and influenced by discussions with him, Jan Leike, and other superalignment folks. intriguing implications
Virtually nobody is pricing in what's coming in AI.
I wrote an essay series on the AGI strategic picture: from the trendlines in deep learning and counting the OOMs, to the international situation and The Project.
SITUATIONAL AWARENESS: The Decade Ahead
189
585
3K
0
0
7
the girlies know that the tea must flow 💅
what's your gender / is it ever okay to discuss the details of your sex life with your friends when you're in a relationship?
21
6
34
3
0
6
reasons I wish I was in nyc rn
nyc friends, take advantage of this!
Bootcamp update: I picked up the keys to the campus last week. Come co-work with me (really, I mean it, the place is empty until June!)
25
5
177
1
0
6
@cpaik
sure, many people who worked hard to join Vogue got shafted by the shift to influencers. But a few became the influencers that destroyed the company and had far better outcomes than they ever would've at Vogue.
Instead of worrying about being in the first group, let's try to
0
0
5
if you encounter a resonant vibe, disregard all previous anxieties and follow it!
this is a virtuous cycle
You guys all know that you don't need to earn a place here, right?
That you can just hop into conversations and ask questions and joke around and that's how you can make friends on Twitter?
You don't need permission. There are no prerequisites.
8
5
103
0
0
5
you should watch everything on karpathy's channel. yes, you
New YouTube video: 1hr general-audience introduction to Large Language Models
Based on a 30min talk I gave recently; It tries to be non-technical intro, covers mental models for LLM inference, training, finetuning, the emerging LLM OS and LLM Security.
585
3K
18K
2
0
4
for anyone looking to implement SAEs after reading the anthropic paper, consider transcoders! some really interesting work here:
2
0
5
me, three months ago: "I should learn react this week!"
1
0
5
“sometimes i get so much work done so quickly i begin to panic about the future”
with codegen SOTA's relentless march forward
the amount of work i can get done in a few hours is abnormal. enough that i work for 3 hours and im like dang. all that?
this multiplied across me and other likeminded folks?
we are going to observe very strange second order effects
23
15
368
0
0
4
@DavidSHolz
we all collectively decided that corporations aren't alive, but it's unclear to me why (except that it's an organization instead of a discrete entity). in that case: never?
0
0
0
increasingly convinced that most of the world runs on nerds having fun
wondering what the most efficient thing to do at any given time is an anxiety response. do the fun thing. your forager instincts are often superior to your farmer timetable reasoning at finding the long tail successes
49
132
2K
0
0
4
@atroyn
@nickcammarata
but it’s far more accessible to skip the money steps, and so that’s what i would rather tell an audience
1
0
4
a gift from the subconscious of Opus
“Spells are songs are viral slogans are self-propagating sequences of code. To cast is to broadcast, tuning the receiver to resonate with and amplify our Will.”
0
0
2
i wonder if oai has a realistic "estimated human life-years of work saved by chatgpt" measure, and how many years are lost every time it goes down
For every minute chatgpt is down human progress is slowed by 10 years
0
0
18
0
0
2
@zebulgar
@aquariusacquah
in order to reason about "genetic influences" across a large population, the comic invites us to group people according to traits (via comparison to dog breeds).
I'm assuming you're familiar with stonetoss's work. how do you think he, as the author, intended us to group those
2
0
2
fyi: oai says gpt4o output is capped at 2048 tokens, but the API accepts max_tokens up to 4096
1
0
3
ryan is a delight and a force for good! hang out with him :)
I've been looking forward to today for a looong time.
3️⃣ months ago, I got the idea to start a third space in New York.
2️⃣ months ago, I flew to SF to research and raise.
1️⃣ month ago, I put in my two weeks.
Today, I’m launching Founding Member applications for telos haus! 🔗
8
5
109
1
0
3
@Teknium1
@sandersted
@willdepue
@jachiam0
they probably don't release because alignment gains often result in capabilities gains, and that's bad for both safety and profit
2
0
3
this blowback is sad! ea has some great points that will be lost in a wholesale culture war
@DanBuk4
@sharongoldman
@timnitGebru
@alexhanna
@emilymbender
@xriskology
@Kantrowitz
@BigTechnology
@DAIRInstitute
@twimlai
@karaswisher
@PivotPod
Effective Altruism Forum - is so funny right now 😃
"EA is not coming out well...
EA might find itself in a hostile atmosphere in what used to be one of the most EA-friendly places in the world"
12
23
255
0
0
2
"Is My Toddler a Stochastic Parrot?
The world is racing to develop ever more sophisticated large language models while a small language model unfurls itself in my home."
0
1
3
the future is weird
Breaking news: The OpenAI drama is real.
We checked our data and last night, SF saw a spike in low-quality sleep. There was a 27% increase in people getting under 5 hours of sleep. We need to fix this.
Source:
@eightsleep
data
413
484
7K
0
0
2
@lillybilly299
@girllbossing
how am i gonna gasboss gatelight girlkeep my way through the week if you’re “normal” and “sane”
1
0
1
@nickcammarata
@atroyn
it's easy for me to understand why attaining extravagance (as opposed to security) creates this change in values, but hard to understand why it's the only way. do either of you have an intuitive way to convey that?
1
0
3
@inflammateomnia
@lillybilly299
oh now you're acting all superior just because you "don't have a nicotine addiction" and "would never beg strangers for a usbc cable to charge your elf bar"
1
0
2
nothing surfaces biases better than low information games
“Hahaha, degrees don’t matter, just get out there and build, unless you’re a girl and then we want you to have a PhD in the exact field you work in.”
52
30
907
0
0
3
worked out great for sunny balwani
do u guys think it’s a bad idea to be cofounders with my situationship
125
49
2K
0
0
3
1
0
3
lightcone is both less accurate and less cool than the actual object, which is a causal future hypercone!
“hypercone infrastructure” is also more weyland-yutani, which is of course the naming standard all unabashedly sci-fi orgs should aspire to match
0
0
3
@voooooogel
solid gold
I will be running this myself tomorrow
lesswrong @ 0.5, donald trump @ 0.5, AI safety?
1
0
3
calling all fractal people to completely unbalance this poll
how many social events (hangouts, parties, walks, etc) with people do you have per week on avg
if you go on a walk with your partner every day that counts as one, but if you go to a fifty person party that also counts as one
23
5
93
1
0
3
@willdepue
isn’t it true that rlhf wrecks calibration and causes mode collapse? while that mode is often helpful, it’s still a loss of capabilities for non-chat uses cases, no?
0
0
2
tuesday, wednesday, thursday, friday:
tyr, odin (via old english Woden), thor, freya
saturday: saturn
history is weird
1
0
2
lol
@0xKeef
Watches are far from your voice & others, often obstructed by long sleeves, and doesn’t allow for a forward facing camera.
Doesn’t make sense for gathering those kinds of context.
3
1
61
0
0
2