All 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 202220232024 2025

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 23 24 25 26 27 28 29 3031

Responses to apparent rationalist confusions about game / decision theory

Anthony DiGiovanniAug 30, 2023, 10:02 PM

142 points

57 votes

Overall karma indicates overall quality.

20 comments12 min readLW link 1 review

Invulnerable Incomplete Preferences: A Formal Statement

SCPAug 30, 2023, 9:59 PM

136 points

44 votes

Overall karma indicates overall quality.

39 comments35 min readLW link

Report on Frontier Model Training

YafahEdelmanAug 30, 2023, 8:02 PM

122 points

66 votes

Overall karma indicates overall quality.

21 comments21 min readLW link

(docs.google.com)

An adversarial example for Direct Logit Attribution: memory management in gelu-4l

Can, Yeu-Tong Lau, James Dao and Jett Janiak

Aug 30, 2023, 5:36 PM

17 points

7 votes

Overall karma indicates overall quality.

0 comments8 min readLW link

(arxiv.org)

A Letter to the Editor of MIT Technology Review

JeffsAug 30, 2023, 4:59 PM

0 points

5 votes

Overall karma indicates overall quality.

0 comments2 min readLW link

Biosecurity Culture, Computer Security Culture

jefftkAug 30, 2023, 4:40 PM

103 points

52 votes

Overall karma indicates overall quality.

11 comments2 min readLW link

(www.jefftk.com)

Why I hang out at LessWrong and why you should check-in there every now and then

Bill BenzonAug 30, 2023, 3:20 PM

16 points

9 votes

Overall karma indicates overall quality.

5 comments5 min readLW link

“Wanting” and “liking”

Mateusz BagińskiAug 30, 2023, 2:52 PM

23 points

7 votes

Overall karma indicates overall quality.

3 comments29 min readLW link

Open Call for Research Assistants in Developmental Interpretability

Jesse Hoogland, Daniel Murfet, Alexander Gietelink Oldenziel and Stan van Wingerden

Aug 30, 2023, 9:02 AM

56 points

22 votes

Overall karma indicates overall quality.

11 comments4 min readLW link

LTFF and EAIF are unusually funding-constrained right now

Linch and calebp99

Aug 30, 2023, 1:03 AM

90 points

37 votes

Overall karma indicates overall quality.

24 comments15 min readLW link

(forum.effectivealtruism.org)

Paper Walkthrough: Automated Circuit Discovery with Arthur Conmy

Neel NandaAug 29, 2023, 10:07 PM

36 points

12 votes

Overall karma indicates overall quality.

1 comment1 min readLW link

(www.youtube.com)

An OV-Coherent Toy Model of Attention Head Superposition

Lauren Greenspan and keith_wynroe

Aug 29, 2023, 7:44 PM

26 points

9 votes

Overall karma indicates overall quality.

2 comments6 min readLW link

The Economics of the Asteroid Deflection Problem (Dominant Assurance Contracts)

moyamoAug 29, 2023, 6:28 PM

77 points

33 votes

Overall karma indicates overall quality.

71 comments15 min readLW link

Democratic Fine-Tuning

Joe EdelmanAug 29, 2023, 6:13 PM

22 points

19 votes

Overall karma indicates overall quality.

2 comments1 min readLW link

(open.substack.com)

Should rationalists (be seen to) win?

Will_PearsonAug 29, 2023, 6:13 PM

6 points

4 votes

Overall karma indicates overall quality.

7 comments1 min readLW link

Frankfurt meetup

sultanAug 29, 2023, 6:10 PM

2 points

2 votes

Overall karma indicates overall quality.

0 comments1 min readLW link

Istanbul meetup

sultanAug 29, 2023, 6:10 PM

3 points

3 votes

Overall karma indicates overall quality.

0 comments1 min readLW link

Broken Benchmark: MMLU

awgAug 29, 2023, 6:09 PM

24 points

26 votes

Overall karma indicates overall quality.

5 comments1 min readLW link

(www.youtube.com)

AISN #20: LLM Proliferation, AI Deception, and Continuing Drivers of AI Capabilities

Dan HAug 29, 2023, 3:07 PM

12 points

5 votes

Overall karma indicates overall quality.

0 comments8 min readLW link

(newsletter.safe.ai)

Loft Bed Fan Guard

jefftkAug 29, 2023, 1:30 PM

16 points

6 votes

Overall karma indicates overall quality.

3 comments1 min readLW link

(www.jefftk.com)

Dating Roundup #1: This is Why You’re Single

ZviAug 29, 2023, 12:50 PM

87 points

59 votes

Overall karma indicates overall quality.

28 comments38 min readLW link

(thezvi.wordpress.com)

Neural Recognizers: Some [old] notes based on a TV tube metaphor [perceptual contact with the world]

Bill BenzonAug 29, 2023, 11:33 AM

4 points

1 vote

Overall karma indicates overall quality.

0 comments5 min readLW link

Barriers to Mechanistic Interpretability for AGI Safety

Connor LeahyAug 29, 2023, 10:56 AM

63 points

41 votes

Overall karma indicates overall quality.

13 comments1 min readLW link

(www.youtube.com)

Newcomb Variant

lsusrAug 29, 2023, 7:02 AM

25 points

23 votes

Overall karma indicates overall quality.

23 comments1 min readLW link

[Question] Incentives affecting alignment-researcher encouragement

Nicholas KrossAug 29, 2023, 5:11 AM

28 points

11 votes

Overall karma indicates overall quality.

3 comments1 min readLW link

Anyone want to debate publicly about FDT?

Bentham's BulldogAug 29, 2023, 3:45 AM

14 points

10 votes

Overall karma indicates overall quality.

31 comments1 min readLW link

AI Deception: A Survey of Examples, Risks, and Potential Solutions

Simon Goldstein and Peter S. Park

Aug 29, 2023, 1:29 AM

54 points

20 votes

Overall karma indicates overall quality.

3 comments10 min readLW link

An Interpretability Illusion for Activation Patching of Arbitrary Subspaces

Georg Lange, Alex Makelov and Neel Nanda

Aug 29, 2023, 1:04 AM

77 points

28 votes

Overall karma indicates overall quality.

4 comments1 min readLW link

OpenAI API base models are not sycophantic, at any size

nostalgebraistAug 29, 2023, 12:58 AM

183 points

80 votes

Overall karma indicates overall quality.

20 comments2 min readLW link

(colab.research.google.com)

Paradigms and Theory Choice in AI: Adaptivity, Economy and Control

particlemaniaAug 28, 2023, 10:19 PM

4 points

2 votes

Overall karma indicates overall quality.

0 comments16 min readLW link

[Question] Humanities In A Post-Conscious AI World?

NetcentricaAug 28, 2023, 9:59 PM

1 point

2 votes

Overall karma indicates overall quality.

1 comment2 min readLW link

Introducing the Center for AI Policy (& we’re hiring!)

Thomas LarsenAug 28, 2023, 9:17 PM

123 points

67 votes

Overall karma indicates overall quality.

50 comments2 min readLW link

(www.aipolicy.us)

[Question] 45% to 55% vs. 90% to 100%

yhoisethAug 28, 2023, 7:15 PM

5 points

5 votes

Overall karma indicates overall quality.

8 comments4 min readLW link

The Evidence for Question Decomposition is Weak

niplavAug 28, 2023, 3:46 PM

22 points

10 votes

Overall karma indicates overall quality.

6 comments5 min readLW link

ACX Meetup Anywhere, Bratislava, Slovakia

David VargaAug 28, 2023, 3:40 PM

1 point

1 vote

Overall karma indicates overall quality.

0 comments1 min readLW link

The Anthropic Principle Tells Us That AGI Will Not Be Conscious

nemAug 28, 2023, 3:25 PM

2 points

9 votes

Overall karma indicates overall quality.

8 comments1 min readLW link

No More Freezer Pucks

jefftkAug 28, 2023, 3:20 PM

10 points

2 votes

Overall karma indicates overall quality.

7 comments1 min readLW link

(www.jefftk.com)

The mind as a polyviscous fluid

Bill BenzonAug 28, 2023, 2:38 PM

8 points

8 votes

Overall karma indicates overall quality.

0 comments3 min readLW link

[Question] Who can most reduce X-Risk?

sudhanshu_kasewaAug 28, 2023, 2:38 PM

1 point

3 votes

Overall karma indicates overall quality.

12 comments1 min readLW link

Drinks at a bar

yakimoffAug 28, 2023, 3:13 AM

2 points

2 votes

Overall karma indicates overall quality.

0 comments1 min readLW link

Dear Self; we need to talk about ambition

ElizabethAug 27, 2023, 11:10 PM

271 points

159 votes

Overall karma indicates overall quality.

28 comments8 min readLW link 2 reviews

(acesounderglass.com)

AI pause/governance advocacy might be net-negative, especially without a focus on explaining x-risk

Mikhail SaminAug 27, 2023, 11:05 PM

72 points

44 votes

Overall karma indicates overall quality.

9 comments6 min readLW link

Will issues are quite nearly skill issues

dkl9Aug 27, 2023, 4:42 PM

1 point

4 votes

Overall karma indicates overall quality.

1 comment3 min readLW link

(dkl9.net)

Xanadu, GPT, and Beyond: An adventure of the mind

Bill BenzonAug 27, 2023, 4:19 PM

2 points

4 votes

Overall karma indicates overall quality.

0 comments5 min readLW link

High level overview on how to go about estimating “p(doom)” or the like

Aryeh EnglanderAug 27, 2023, 4:01 PM

16 points

3 votes

Overall karma indicates overall quality.

0 comments5 min readLW link

Trying a Wet Suit

jefftkAug 27, 2023, 3:00 PM

35 points

16 votes

Overall karma indicates overall quality.

5 comments1 min readLW link

(www.jefftk.com)

Apply to a small iteration of MLAB in Oxford

RP, MariaK and OliverHayman

Aug 27, 2023, 2:54 PM

2 points

1 vote

Overall karma indicates overall quality.

0 comments1 min readLW link

Apply to a small iteration of MLAB to be run in Oxford

RP, MariaK and OliverHayman

Aug 27, 2023, 2:21 PM

12 points

10 votes

Overall karma indicates overall quality.

0 comments1 min readLW link

The Game of Dominance

Karl von WendtAug 27, 2023, 11:04 AM

24 points

12 votes

Overall karma indicates overall quality.

15 comments6 min readLW link

Eliezer Yudkowsky Is Frequently, Confidently, Egregiously Wrong

Bentham's BulldogAug 27, 2023, 1:06 AM

−12 points

122 votes

Overall karma indicates overall quality.

97 comments36 min readLW link

Keyboard shortcuts

Keys shown in yellow (e.g., ]) are accesskeys, and require a browser-specific modifier key (or keys).

Keys shown in grey (e.g., ?) do not require any modifier keys.

General
? Show keyboard shortcuts
Esc Hide keyboard shortcuts

Site navigation
h Go to Home (a.k.a. “Frontpage”) view
f Go to Featured (a.k.a. “Curated”) view
a Go to All (a.k.a. “Community”) view
m Go to Meta view
v Go to Tags view
c Go to Recent Comments view
r Go to Archive view
q Go to Sequences view
t Go to About page
u Go to User or Login page
o Go to Inbox page

Page navigation
, Jump up to top of page
. Jump down to bottom of page
/ Jump to top of comments section
s Search

Page actions
n New post or comment
e Edit current post

Post/comment list views
. Focus next entry in list
, Focus previous entry in list
; Cycle between links in focused entry
Enter Go to currently focused entry
Esc Unfocus currently focused entry
] Go to next page
[ Go to previous page
\ Go to first page
e Edit currently focused post

Editor
k Bold text
i Italic text
l Insert hyperlink
q Blockquote text

Appearance
= Increase text size
- Decrease text size
0 Reset to default text size
′ Cycle through content width settings
1 Switch to default theme [A]
2 Switch to dark theme [B]
3 Switch to grey theme [C]
4 Switch to ultramodern theme [D]
5 Switch to simple theme [E]
6 Switch to brutalist theme [F]
7 Switch to ReadTheSequences theme [G]
8 Switch to classic Less Wrong theme [H]
9 Switch to modern Less Wrong theme [I]
; Open theme tweaker
Enter Save changes and close theme tweaker
Esc Close theme tweaker (without saving)

Slide shows
l Start/resume slideshow
Esc Exit slideshow
→↓ Next slide
←↑ Previous slide
Space Reset slide zoom

Miscellaneous
x Switch to next view on user page
z Switch to previous view on user page
` Toggle compact comment list view
g Toggle anti-kibitzer