All200520062007200820092010201120122013201420152016201720182019202020212022202320242025
AllJanFebMarAprMayJunJulAugSepOctNovDec
All12345678910111213141516171819202122232425262728293031

Arch-an­ar­chy:The­ory and practice

Peter lawless Apr 30, 2024, 11:20 PM
−6 points
0 comments2 min readLW link

“Open Source AI” is a lie, but it doesn’t have to be

jacobhaimesApr 30, 2024, 11:10 PM
19 points
5 comments6 min readLW link
(jacob-haimes.github.io)

Ques­tions for labs

Zach Stein-PerlmanApr 30, 2024, 10:15 PM
77 points
11 comments8 min readLW link

Real­ity com­pre­hen­si­bil­ity: are there illog­i­cal things in re­al­ity?

DDthinkerApr 30, 2024, 9:30 PM
−3 points
0 comments10 min readLW link

Mechanis­ti­cally Elic­it­ing La­tent Be­hav­iors in Lan­guage Models

Andrew Mack and TurnTrout
Apr 30, 2024, 6:51 PM
210 points
43 comments45 min readLW link

[Question] What is the eas­iest/​funnest way to build up a com­pre­hen­sive un­der­stand­ing of AI and AI Safety?

Jordan ArelApr 30, 2024, 6:41 PM
4 points
2 comments1 min readLW link

Transcoders en­able fine-grained in­ter­pretable cir­cuit anal­y­sis for lan­guage models

Jacob Dunefsky, Philippe Chlenski and Neel Nanda
Apr 30, 2024, 5:58 PM
74 points
14 comments17 min readLW link

An­nounc­ing the 2024 Roots of Progress Blog-Build­ing Intensive

jasoncrawfordApr 30, 2024, 5:37 PM
14 points
0 comments2 min readLW link
(rootsofprogress.org)

The In­ten­tional Stance, LLMs Edition

Eleni AngelouApr 30, 2024, 5:12 PM
30 points
3 comments8 min readLW link

In­tro­duc­ing AI Lab Watch

Zach Stein-PerlmanApr 30, 2024, 5:00 PM
225 points
30 comments1 min readLW link
(ailabwatch.org)

Why I’m do­ing PauseAI

Joseph MillerApr 30, 2024, 4:21 PM
108 points
16 comments4 min readLW link

LLMs could be as con­scious as hu­man em­u­la­tions, potentially

CanalettoApr 30, 2024, 11:36 AM
15 points
15 comments3 min readLW link

An in­ter­est­ing math­e­mat­i­cal model of how LLMs work

Bill BenzonApr 30, 2024, 11:01 AM
5 points
0 comments1 min readLW link

Towards Mul­ti­modal In­ter­pretabil­ity: Learn­ing Sparse In­ter­pretable Fea­tures in Vi­sion Transformers

hugofryApr 29, 2024, 8:57 PM
94 points
8 comments11 min readLW link

Towards a for­mal­iza­tion of the agent struc­ture problem

Alex_AltairApr 29, 2024, 8:28 PM
55 points
6 comments14 min readLW link

Iron­ing Out the Squiggles

Zack_M_DavisApr 29, 2024, 4:13 PM
157 points
36 comments11 min readLW link

Su­per ad­di­tivity of consciousness

Arturo MaciasApr 29, 2024, 3:41 PM
−2 points
13 comments2 min readLW link

AISC9 has ended and there will be an AISC10

Linda LinseforsApr 29, 2024, 10:53 AM
75 points
4 comments2 min readLW link

Open-Source AI: A Reg­u­la­tory Review

Elliot Mckernon and Deric Cheng
Apr 29, 2024, 10:10 AM
18 points
0 comments8 min readLW link

Big-en­dian is bet­ter than lit­tle-endian

MenotimApr 29, 2024, 2:30 AM
29 points
17 comments3 min readLW link

The Prop-room and Stage Cog­ni­tive Architecture

Robert KralischApr 29, 2024, 12:48 AM
14 points
4 comments14 min readLW link

How are Si­mu­la­tors and Agents re­lated?

Robert KralischApr 29, 2024, 12:22 AM
6 points
0 comments7 min readLW link

Ex­tended Embodiment

Robert KralischApr 29, 2024, 12:18 AM
8 points
1 comment3 min readLW link

Refer­en­tial Containment

Robert KralischApr 29, 2024, 12:16 AM
2 points
4 comments3 min readLW link

Disen­tan­gling Com­pe­tence and Intelligence

Robert KralischApr 29, 2024, 12:12 AM
23 points
7 comments6 min readLW link

List your AI X-Risk cruxes!

Aryeh EnglanderApr 28, 2024, 6:26 PM
42 points
7 comments2 min readLW link

Things I tell my­self to be more agentic

DMMFApr 28, 2024, 5:44 PM
9 points
0 comments3 min readLW link
(danfrank.ca)

Es­ti­mat­ing the Num­ber of Play­ers from Game Re­sult Percentages

Daniel LApr 28, 2024, 5:42 PM
1 point
2 comments1 min readLW link

The Science Al­gorithm—AISC 2024 Fi­nal Presentation

Johannes C. MayerApr 28, 2024, 2:55 PM
4 points
0 comments1 min readLW link
(www.youtube.com)

[Aspira­tion-based de­signs] Out­look: deal­ing with complexity

Jobst Heitzig, jossoliver, thomasfinn and Simon Dima
Apr 28, 2024, 1:06 PM
13 points
3 comments2 min readLW link

[Aspira­tion-based de­signs] 3. Perfor­mance and safety crite­ria, and as­pira­tion intervals

Jobst HeitzigApr 28, 2024, 1:04 PM
10 points
0 comments12 min readLW link

[Aspira­tion-based de­signs] 2. For­mal frame­work, ba­sic algorithm

Jobst Heitzig, Simon Dima and Simon Fischer
Apr 28, 2024, 1:02 PM
18 points
2 comments16 min readLW link

[Aspira­tion-based de­signs] 1. In­for­mal in­tro­duc­tion

B Jacobs, Jobst Heitzig, Simon Fischer and Simon Dima
Apr 28, 2024, 1:00 PM
44 points
4 comments8 min readLW link

Play­ing North­boro with Lily and Rick

jefftkApr 28, 2024, 2:40 AM
10 points
1 comment2 min readLW link
(www.jefftk.com)

Re­lease of UN’s draft re­lated to the gov­er­nance of AI (a sum­mary of the Si­mon In­sti­tute’s re­sponse)

Sebastian SchmidtApr 27, 2024, 6:34 PM
7 points
0 comments1 min readLW link
(forum.effectivealtruism.org)

Mercy to the Ma­chine: Thoughts & Rights

False NameApr 27, 2024, 4:36 PM
7 points
6 comments17 min readLW link

Con­structabil­ity: Plainly-coded AGIs may be fea­si­ble in the near future

Épiphanie Gédéon and Charbel-Raphaël
Apr 27, 2024, 4:04 PM
91 points
13 comments13 min readLW link

So What’s Up With PUFAs Chem­i­cally?

J BostockApr 27, 2024, 1:32 PM
57 points
23 comments6 min readLW link

Link: Let’s Think Dot by Dot: Hid­den Com­pu­ta­tion in Trans­former Lan­guage Models by Ja­cob Pfau, William Mer­rill & Sa­muel R. Bowman

Chris_LeongApr 27, 2024, 1:22 PM
12 points
0 comments1 min readLW link
(twitter.com)

Two Ver­nor Vinge Book Reviews

Maxwell TabarrokApr 27, 2024, 12:14 PM
17 points
0 comments2 min readLW link
(www.maximum-progress.com)

Re­fusal in LLMs is me­di­ated by a sin­gle direction

Andy Arditi, Oscar Obeso, Aaquib111, wesg and Neel Nanda
Apr 27, 2024, 11:13 AM
247 points
95 comments10 min readLW link

[Question] Plau­si­bil­ity of Get­ting Early Warn­ing Shots be­cause AIs can’t co­or­di­nate?

hmysApr 27, 2024, 8:02 AM
12 points
0 comments1 min readLW link

AI Safety Sphere

Myles HApr 27, 2024, 1:49 AM
6 points
2 comments2 min readLW link

Ex­plor­ing the Eso­teric Path­ways to AI Sen­tience (Part One)

jeffreycarusoApr 27, 2024, 1:02 AM
−11 points
6 comments2 min readLW link

Su­per­po­si­tion is not “just” neu­ron polysemanticity

LawrenceCApr 26, 2024, 11:22 PM
66 points
4 comments13 min readLW link

D&D.Sci Long War: Defen­der of Data-mocracy

aphyerApr 26, 2024, 10:30 PM
44 points
20 comments4 min readLW link

On Not Pul­ling The Lad­der Up Be­hind You

ScrewtapeApr 26, 2024, 9:58 PM
189 points
21 comments9 min readLW link

We are headed into an ex­treme com­pute overhang

devrandomApr 26, 2024, 9:38 PM
54 points
34 comments2 min readLW link

[Con­cept Depen­dency] Edge Reg­u­lar Lat­tice Graph

Johannes C. MayerApr 26, 2024, 9:14 PM
9 points
1 comment1 min readLW link

[Con­cept Depen­dency] Con­cept Depen­dency Posts

Johannes C. MayerApr 26, 2024, 8:57 PM
10 points
3 comments2 min readLW link