[Question] Liter­a­ture On Ex­is­ten­tial Risk From At­mo­spheric Con­tam­i­na­tion?

YitzOct 13, 2023, 10:27 PM
6 points
3 comments1 min readLW link

How to par­ti­tion teams to move fast? De­bat­ing “low-di­men­sional cuts”

Oct 13, 2023, 9:43 PM
41 points
2 comments11 min readLW link

Gothen­burg LW /​ ACX meetup

StefanOct 13, 2023, 9:39 PM
2 points
0 comments1 min readLW link

Meta-Regulations

SableOct 13, 2023, 9:23 PM
18 points
5 comments10 min readLW link
(affablyevil.substack.com)

Hiring: Lighthaven Events & Venue Lead

RaemonOct 13, 2023, 9:02 PM
69 points
3 comments4 min readLW link

Pre­dic­tion mar­kets cov­ered in the NYT pod­cast “Hard Fork”

Austin ChenOct 13, 2023, 6:43 PM
56 points
6 commentsLW link
(www.nytimes.com)

[Paper] All’s Fair In Love And Love: Copy Sup­pres­sion in GPT-2 Small

Oct 13, 2023, 6:32 PM
82 points
4 comments8 min readLW link

[Question] In­tel­li­gence En­hance­ment (Monthly Thread) 13 Oct 2023

Nicholas / Heather KrossOct 13, 2023, 5:28 PM
52 points
40 comments1 min readLW link

FLI pod­cast se­ries, “Imag­ine A World”, about as­pira­tional fu­tures with AGI

Jackson WagnerOct 13, 2023, 4:07 PM
9 points
0 comments4 min readLW link

To open-source or to not open-source, that is (an over­sim­plifi­ca­tion of) the ques­tion.

Justin BullockOct 13, 2023, 3:10 PM
12 points
5 comments5 min readLW link

Com­bi­na­tion Lock Boxes

jefftkOct 13, 2023, 12:50 PM
17 points
9 comments1 min readLW link
(www.jefftk.com)

Cir­cle of Sup­port (Oct 14th @ 10am PST)

AlexeiOct 13, 2023, 9:24 AM
19 points
1 comment1 min readLW link

[Question] How can the world han­dle the HAMAS situ­a­tion?

AnnapurnaOct 13, 2023, 9:15 AM
5 points
43 comments1 min readLW link

UVic AI Ethics Conference

Oct 13, 2023, 7:31 AM
3 points
1 comment1 min readLW link

LW UI fea­tures you might not have tried

ElizabethOct 13, 2023, 3:04 AM
49 points
6 comments1 min readLW link

Re­vis­it­ing Guide Dogs and Blind­ness Prevention

jefftkOct 13, 2023, 2:30 AM
22 points
0 comments2 min readLW link
(www.jefftk.com)

Paper: Un­der­stand­ing and Con­trol­ling a Maze-Solv­ing Policy Network

Oct 13, 2023, 1:38 AM
70 points
0 comments1 min readLW link
(arxiv.org)

OPTIC: An­nounc­ing In­ter­col­le­giate Fore­cast­ing Tour­na­ments in SF, DC, Boston

Oct 13, 2023, 1:36 AM
6 points
0 comments1 min readLW link

Progress links di­gest, 2023-10-12: Dyson sphere ther­mo­dy­nam­ics and a cure for cavities

jasoncrawfordOct 13, 2023, 12:41 AM
15 points
1 comment10 min readLW link
(rootsofprogress.org)

What do Marginal Grants at EAIF Look Like? Fund­ing Pri­ori­ties and Grant­mak­ing Thresh­olds at the EA In­fras­truc­ture Fund

LinchOct 12, 2023, 9:40 PM
20 points
0 commentsLW link

unRLHF—Effi­ciently un­do­ing LLM safeguards

Oct 12, 2023, 7:58 PM
117 points
15 comments20 min readLW link

LoRA Fine-tun­ing Effi­ciently Un­does Safety Train­ing from Llama 2-Chat 70B

Oct 12, 2023, 7:58 PM
151 points
29 comments14 min readLW link

[Question] Look­ing for read­ing recom­men­da­tions: The­o­ries of right/​jus­tice that safe­guard against hav­ing one’s job au­to­mated?

bulKlubOct 12, 2023, 7:40 PM
−1 points
1 comment1 min readLW link

The In­ter­na­tional PauseAI Protest: Ac­tivism un­der uncertainty

Joseph MillerOct 12, 2023, 5:36 PM
32 points
1 commentLW link

AI #33: Cool New In­ter­pretabil­ity Paper

ZviOct 12, 2023, 4:20 PM
46 points
18 comments46 min readLW link
(thezvi.wordpress.com)

Notic­ing con­fu­sion in physics

Jacob G-WOct 12, 2023, 3:21 PM
20 points
27 comments2 min readLW link
(jacobgw.com)

[Question] How to make to-do lists (and to get things done)?

TeaTieAndHatOct 12, 2023, 2:26 PM
9 points
13 comments2 min readLW link

Rele­vance of ‘Harm­ful In­tel­li­gence’ Data in Train­ing Datasets (We­bText vs. Pile)

MiguelDevOct 12, 2023, 12:08 PM
12 points
0 comments9 min readLW link

Soul­mate Fermi Es­ti­mate + My A(ltr)u[t]is­tic Mat­ing Strat­egy

Jordan ArelOct 12, 2023, 8:32 AM
0 points
9 comments3 min readLW link

Evolu­tion Solved Align­ment (what sharp left turn?)

jacob_cannellOct 12, 2023, 4:15 AM
23 points
89 comments4 min readLW link

The CHOICE

Gabi QUENEOct 12, 2023, 3:02 AM
−29 points
2 comments3 min readLW link

Sols­tice 2023 Roundup

dspeyerOct 11, 2023, 11:09 PM
28 points
6 comments1 min readLW link

Un­der­stand­ing LLMs: Some ba­sic ob­ser­va­tions about words, syn­tax, and dis­course [w/​ a con­jec­ture about grokking]

Bill BenzonOct 11, 2023, 7:13 PM
6 points
0 comments5 min readLW link

[Linkpost] Gen­er­al­iza­tion in diffu­sion mod­els arises from ge­om­e­try-adap­tive har­monic representation

Bogdan Ionut CirsteaOct 11, 2023, 5:48 PM
4 points
3 comments1 min readLW link

What I’ve been read­ing, Oc­to­ber 2023: The stir­rup in Europe, 19th-cen­tury art deco, and more

jasoncrawfordOct 11, 2023, 4:11 PM
18 points
2 comments11 min readLW link
(rootsofprogress.org)

EA Madrid social

Pablo VillalobosOct 11, 2023, 3:34 PM
6 points
0 comments1 min readLW link

At­tribut­ing to in­ter­ac­tions with GCPD and GWPD

jennyOct 11, 2023, 3:06 PM
20 points
0 comments6 min readLW link

You’re Mea­sur­ing Model Com­plex­ity Wrong

Oct 11, 2023, 11:46 AM
93 points
17 comments13 min readLW link

Up­date on the UK AI Task­force & up­com­ing AI Safety Summit

Elliot MckernonOct 11, 2023, 11:37 AM
84 points
2 comments4 min readLW link

An ex­pla­na­tion for ev­ery to­ken: us­ing an LLM to sam­ple an­other LLM

Max HOct 11, 2023, 12:53 AM
35 points
5 comments11 min readLW link

[Question] Ex­am­ples of Low Sta­tus Fun

niplavOct 10, 2023, 11:19 PM
18 points
17 comments1 min readLW link

A New Model for Com­pute Cen­ter Verification

Damin CurtisOct 10, 2023, 7:22 PM
8 points
0 comments5 min readLW link

An­nounc­ing MIRI’s new CEO and lead­er­ship team

Gretta DulebaOct 10, 2023, 7:22 PM
222 points
52 comments3 min readLW link

18 Hetero­dox lenses to look the world through

Shaurya GuptaOct 10, 2023, 6:33 PM
−1 points
2 comments5 min readLW link

Doc­u­ment­ing Jour­ney Into AI Safety

jacobhaimesOct 10, 2023, 6:30 PM
17 points
4 comments6 min readLW link

Look­ing for AI Art Col­lab­o­ra­tors!

beatrice@foresight.orgOct 10, 2023, 6:24 PM
1 point
0 comments1 min readLW link

Child­hood Roundup #3

ZviOct 10, 2023, 2:30 PM
49 points
3 comments30 min readLW link
(thezvi.wordpress.com)

My sim­ple model for Align­ment vs Capability

ryan_bOct 10, 2023, 12:07 PM
7 points
0 comments7 min readLW link

Next year in Jerusalem: The brilli­ant ideas and ra­di­ant legacy of Miriam Lip­schutz Ye­vick [in re­la­tion to cur­rent AI de­bates]

Bill BenzonOct 10, 2023, 9:06 AM
1 point
0 comments1 min readLW link
(3quarksdaily.com)

I’m a Former Is­raeli Officer. AMA

Yovel RomOct 10, 2023, 8:33 AM
78 points
70 comments1 min readLW link