All 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 202320242025

All Jan Feb Mar Apr May Jun JulAugSep Oct Nov Dec

All 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 232425 26 27 28 29 30 31

Crafting Polysemantic Transformer Benchmarks with Known Circuits

Evan Anders and Adrià Garriga-alonso

Aug 23, 2024, 10:03 PM

17 points

0 comments25 min readLW link

[Question] What is an appropriate sample size when surveying billions of data points?

BlakeAug 23, 2024, 9:54 PM

1 point

2 comments1 min readLW link

Interpretability as Compression: Reconsidering SAE Explanations of Neural Activations with MDL-SAEs

Kola Ayonrinde, Michael Pearce and Lee Sharkey

Aug 23, 2024, 6:52 PM

42 points

8 comments16 min readLW link

How I started believing religion might actually matter for rationality and moral philosophy

zhukeepaAug 23, 2024, 5:40 PM

129 points

41 comments7 min readLW link

[Question] What do you expect AI capabilities may look like in 2028?

nonzerosumAug 23, 2024, 4:59 PM

9 points

5 comments1 min readLW link

Invitation to lead a project at AI Safety Camp (Virtual Edition, 2025)

Linda Linsefors, Remmelt Ellen and Robert Kralisch

Aug 23, 2024, 2:18 PM

17 points

2 comments4 min readLW link

If we solve alignment, do we die anyway?

Seth HerdAug 23, 2024, 1:13 PM

84 points

130 comments4 min readLW link

What’s going on with Per-Component Weight Updates?

4gateAug 22, 2024, 9:22 PM

1 point

0 comments6 min readLW link

Interoperable High Level Structures: Early Thoughts on Adjectives

johnswentworth and David Lorell

Aug 22, 2024, 9:12 PM

49 points

1 comment7 min readLW link

Interest poll: A time-waster blocker for desktop Linux programs

nahojAug 22, 2024, 8:44 PM

4 points

5 comments1 min readLW link

Turning 22 in the Pre-Apocalypse

testingthewatersAug 22, 2024, 8:28 PM

38 points

14 comments24 min readLW link

(utilityhotbar.github.io)

A Robust Natural Latent Over A Mixed Distribution Is Natural Over The Distributions Which Were Mixed

johnswentworth and David Lorell

Aug 22, 2024, 7:19 PM

42 points

4 comments4 min readLW link

what becoming more secure did for me

Chris LakinAug 22, 2024, 5:44 PM

26 points

5 comments2 min readLW link

(chrislakin.blog)

A primer on the current state of longevity research

Abhishaike MahajanAug 22, 2024, 5:14 PM

109 points

6 comments14 min readLW link

(www.owlposting.com)

Some reasons to start a project to stop harmful AI

RemmeltAug 22, 2024, 4:23 PM

5 points

0 comments2 min readLW link

The economics of space tethers

harsimonyAug 22, 2024, 4:15 PM

67 points

22 comments7 min readLW link

(splittinginfinity.substack.com)

Dima’s Shortform

Dmitrii KrasheninnikovAug 22, 2024, 2:49 PM

3 points

0 comments1 min readLW link

AI #78: Some Welcome Calm

ZviAug 22, 2024, 2:20 PM

61 points

15 comments33 min readLW link

(thezvi.wordpress.com)

[Question] How do we know dreams aren’t real?

Logan ZoellnerAug 22, 2024, 12:41 PM

5 points

31 comments1 min readLW link

Measuring Structure Development in Algorithmic Transformers

Micurie and Einar Urdshals

Aug 22, 2024, 8:38 AM

56 points

4 comments11 min readLW link

Deception and Jailbreak Sequence: 1. Iterative Refinement Stages of Deception in LLMs

Winnie Yang and Jojo Yang

Aug 22, 2024, 7:32 AM

23 points

1 comment21 min readLW link

Just because an LLM said it doesn’t mean it’s true: an illustrative example

dirkAug 21, 2024, 9:05 PM

26 points

12 comments3 min readLW link

[Question] How do you finish your tasks faster?

CipollaAug 21, 2024, 8:01 PM

4 points

2 comments1 min readLW link

AI Safety Newsletter #40: California AI Legislation Plus, NVIDIA Delays Chip Production, and Do AI Safety Benchmarks Actually Measure Safety?

Corin Katzke, Julius, Alexa Pan and Dan H

Aug 21, 2024, 6:09 PM

11 points

0 comments6 min readLW link

(newsletter.safe.ai)

[Question] Should LW suggest standard metaprompts?

DagonAug 21, 2024, 4:41 PM

3 points

6 comments1 min readLW link

Eternal Existence and Eternal Boredom: The Case for AI and Immortal Humans

Tuan Tu NguyenAug 21, 2024, 9:58 AM

−12 points

2 comments5 min readLW link

Please do not use AI to write for you

Richard_KennawayAug 21, 2024, 9:53 AM

69 points

34 comments4 min readLW link

Apply to Aether—Independent LLM Agent Safety Research Group

RohanSAug 21, 2024, 9:47 AM

10 points

0 comments7 min readLW link

(forum.effectivealtruism.org)

the Giga Press was a mistake

bhauthAug 21, 2024, 4:51 AM

99 points

26 comments5 min readLW link

(bhauth.com)

Exploring the Boundaries of Cognitohazards and the Nature of Reality

Victor NovikovAug 21, 2024, 3:42 AM

−2 points

2 comments1 min readLW link

[Question] What is the point of 2v2 debates?

Axel AhlqvistAug 20, 2024, 9:59 PM

2 points

1 comment1 min readLW link

[Question] Where should I look for information on gut health?

FinalFormal2Aug 20, 2024, 7:44 PM

10 points

10 comments1 min readLW link

Would you benefit from, or object to, a page with LW users’ reacts?

RaemonAug 20, 2024, 4:35 PM

23 points

6 comments1 min readLW link

Freedom of Speech

Zero ContradictionsAug 20, 2024, 4:34 PM

−13 points

2 comments2 min readLW link

(thewaywardaxolotl.blogspot.com)

AGI Safety and Alignment at Google DeepMind: A Summary of Recent Work

Rohin Shah, Seb Farquhar and Anca Dragan

Aug 20, 2024, 4:22 PM

222 points

33 comments9 min readLW link

Trying to be rational for the wrong reasons

ViliamAug 20, 2024, 4:18 PM

26 points

9 comments3 min readLW link

[Question] How great is the utility of “saving” endangered languages?

SpectrumDTAug 20, 2024, 1:14 PM

18 points

29 comments1 min readLW link

Guide to SB 1047

ZviAug 20, 2024, 1:10 PM

71 points

18 comments53 min readLW link

(thezvi.wordpress.com)

Finding Deception in Language Models

Esben Kran and Archana Vaidheeswaran

Aug 20, 2024, 9:42 AM

20 points

4 comments4 min readLW link

Next automated reasoning grand challenge: CompCert

sanxiynAug 20, 2024, 5:27 AM

−5 points

0 comments1 min readLW link

Thiel on AI & Racing with China

Ben PaceAug 20, 2024, 3:19 AM

55 points

10 comments12 min readLW link

Reflecting on the transhumanist rebuttal to AI existential risk and critique of our debate methodologies and misuse of statistics

catgirlsruletheworldAug 20, 2024, 1:59 AM

−5 points

0 comments4 min readLW link

Artificial Intelligence and Eternal Torture and Suffering

Tuan Tu NguyenAug 20, 2024, 1:53 AM

−1 points

0 comments4 min readLW link

AI #77: A Few Upgrades

ZviAug 20, 2024, 12:20 AM

23 points

3 comments52 min readLW link

(thezvi.wordpress.com)

Monthly Roundup #21: August 2024

ZviAug 20, 2024, 12:20 AM

22 points

6 comments40 min readLW link

(thezvi.wordpress.com)

[Linkpost] Automated Design of Agentic Systems

Bogdan Ionut CirsteaAug 19, 2024, 11:06 PM

8 points

1 comment1 min readLW link

(arxiv.org)

Limitations on Formal Verification for AI Safety

Andrew DicksonAug 19, 2024, 11:03 PM

134 points

60 comments23 min readLW link

The Conscious River: Conscious Turing machines negate materialism

blalloAug 19, 2024, 9:54 PM

0 points

4 comments7 min readLW link

LLM Applications I Want To See

sarahconstantinAug 19, 2024, 9:10 PM

102 points

6 comments8 min readLW link

(sarahconstantin.substack.com)

Defining alignment research

Richard_NgoAug 19, 2024, 8:42 PM

92 points

23 comments7 min readLW link