Moscow – ACX Mee­tups Every­where Fall 2024

red-haraSep 20, 2024, 11:03 PM
−1 points
0 comments1 min readLW link

Val­i­dat­ing /​ find­ing al­ign­ment-rele­vant con­cepts us­ing neu­ral data

Bogdan Ionut CirsteaSep 20, 2024, 9:12 PM
7 points
0 comments1 min readLW link
(docs.google.com)

Aug­ment­ing Statis­ti­cal Models with Nat­u­ral Lan­guage Parameters

jsteinhardtSep 20, 2024, 6:30 PM
34 points
0 comments8 min readLW link
(bounded-regret.ghost.io)

Fun With The Tab­ula Muris (Se­nis)

sarahconstantinSep 20, 2024, 6:20 PM
25 points
0 comments8 min readLW link
(sarahconstantin.substack.com)

My Cri­tique of Effec­tive Altruism

Dylan PriceSep 20, 2024, 5:41 PM
−10 points
8 comments4 min readLW link

[Question] Why be moral if we can’t mea­sure how moral we are? Is it even pos­si­ble to mea­sure moral­ity?

OKlogicSep 20, 2024, 5:40 PM
−2 points
0 comments3 min readLW link

On Mea­sur­ing In­tel­lec­tual Perfor­mance—per­sonal ex­pe­rience and sev­eral thoughts

Alexander GufanSep 20, 2024, 5:21 PM
3 points
2 comments8 min readLW link

In­tro­duc­tion to Su­per Pow­ers (for kids!)

Shoshannah TekofskySep 20, 2024, 5:17 PM
25 points
0 comments3 min readLW link
(kidquest.substack.com)

Col­laps­ing “Col­laps­ing the Belief/​Knowl­edge Distinc­tion”

JeremiasSep 20, 2024, 4:11 PM
3 points
0 comments4 min readLW link

A New Class of Glitch To­kens—BPE Subto­ken Ar­ti­facts (BSA)

Lao MeinSep 20, 2024, 1:13 PM
37 points
7 comments5 min readLW link

o1-pre­view is pretty good at do­ing ML on an un­known dataset

Håvard Tveit IhleSep 20, 2024, 8:39 AM
67 points
1 comment2 min readLW link

Mo­ral Trade, Im­pact Distri­bu­tions and Large Worlds

LarksSep 20, 2024, 3:45 AM
7 points
0 commentsLW link

Key­board Gremlins

jefftkSep 20, 2024, 2:30 AM
10 points
0 comments2 min readLW link
(www.jefftk.com)

The case for more Align­ment Tar­get Anal­y­sis (ATA)

Sep 20, 2024, 1:14 AM
27 points
13 comments17 min readLW link

Piling bounded arguments

momom2Sep 19, 2024, 10:27 PM
7 points
0 comments4 min readLW link

We Don’t Know Our Own Values, but Re­ward Bridges The Is-Ought Gap

Sep 19, 2024, 10:22 PM
49 points
48 comments5 min readLW link

In­ter­ested in Cog­ni­tive Boot­camp?

RaemonSep 19, 2024, 10:12 PM
48 points
0 comments2 min readLW link

Just How Good Are Modern Chess Com­put­ers?

nemSep 19, 2024, 6:57 PM
10 points
1 comment6 min readLW link

RLHF is the worst pos­si­ble thing done when fac­ing the al­ign­ment problem

tailcalledSep 19, 2024, 6:56 PM
32 points
10 comments6 min readLW link

AISafety.info: What are In­duc­tive Bi­ases?

AlgonSep 19, 2024, 5:26 PM
11 points
4 comments2 min readLW link
(aisafety.info)

Physics of Lan­guage mod­els (part 2.1)

Nathan Helm-BurgerSep 19, 2024, 4:48 PM
9 points
2 comments1 min readLW link
(youtu.be)

Why good things of­ten don’t lead to bet­ter outcomes

DMMFSep 19, 2024, 4:37 PM
16 points
1 comment4 min readLW link
(danfrank.ca)

To CoT or not to CoT? Chain-of-thought helps mainly on math and sym­bolic reasoning

Bogdan Ionut CirsteaSep 19, 2024, 4:13 PM
21 points
1 comment1 min readLW link
(arxiv.org)

Laz­i­ness death spirals

PatrickDFarleySep 19, 2024, 3:58 PM
276 points
40 comments8 min readLW link

[In­tu­itive self-mod­els] 1. Preliminaries

Steven ByrnesSep 19, 2024, 1:45 PM
91 points
23 comments15 min readLW link

AI #82: The Gover­nor Ponders

ZviSep 19, 2024, 1:30 PM
50 points
8 comments27 min readLW link
(thezvi.wordpress.com)

Slave Mo­ral­ity: A place for ev­ery man and ev­ery man in his place

Martin SustrikSep 19, 2024, 4:20 AM
16 points
7 comments2 min readLW link
(250bpm.substack.com)

Which LessWrong/​Align­ment top­ics would you like to be tu­tored in? [Poll]

RubySep 19, 2024, 1:35 AM
43 points
12 comments1 min readLW link

The Oblique­ness Thesis

jessicataSep 19, 2024, 12:26 AM
95 points
19 comments17 min readLW link

How to choose what to work on

jasoncrawfordSep 18, 2024, 8:39 PM
22 points
6 comments4 min readLW link
(blog.rootsofprogress.org)

In­ten­tion-to-Treat (Re: How harm­ful is mu­sic, re­ally?)

kqrSep 18, 2024, 6:44 PM
11 points
0 comments5 min readLW link
(entropicthoughts.com)

The case for a nega­tive al­ign­ment tax

Sep 18, 2024, 6:33 PM
75 points
20 comments7 min readLW link

En­doge­nous Growth and Hu­man Intelligence

Nicholas D.Sep 18, 2024, 2:05 PM
3 points
0 comments2 min readLW link

In­quisi­tive vs. ad­ver­sar­ial rationality

gbSep 18, 2024, 1:50 PM
6 points
9 comments2 min readLW link

Pro­nouns are Annoying

ymeskhoutSep 18, 2024, 1:30 PM
15 points
23 comments4 min readLW link
(www.ymeskhout.com)

Is “su­per­hu­man” AI fore­cast­ing BS? Some ex­per­i­ments on the “539″ bot from the Cen­tre for AI Safety

titotalSep 18, 2024, 1:07 PM
79 points
3 commentsLW link
(open.substack.com)

Knowl­edge’s practicability

Ted NguyễnSep 18, 2024, 2:31 AM
−5 points
0 comments7 min readLW link
(tednguyen.substack.com)

Skills from a year of Pur­pose­ful Ra­tion­al­ity Practice

RaemonSep 18, 2024, 2:05 AM
190 points
18 comments7 min readLW link

[Question] Where to find re­li­able re­views of AI prod­ucts?

ElizabethSep 17, 2024, 11:48 PM
29 points
6 comments1 min readLW link

Su­per­po­si­tion through Ac­tive Learn­ing Lens

akankshancSep 17, 2024, 5:32 PM
1 point
0 comments10 min readLW link

Sur­vey—Psy­cholog­i­cal Im­pact of Long-Term AI Engagement

Manuela GarcíaSep 17, 2024, 5:31 PM
2 points
0 comments1 min readLW link

Sur­vey—Psy­cholog­i­cal Im­pact of Long-Term AI Engagement

Manuela GarcíaSep 17, 2024, 5:31 PM
1 point
1 comment1 min readLW link

[Question] What does it mean for an event or ob­ser­va­tion to have prob­a­bil­ity 0 or 1 in Bayesian terms?

Noosphere89Sep 17, 2024, 5:28 PM
1 point
22 comments1 min readLW link

How harm­ful is mu­sic, re­ally?

dkl9Sep 17, 2024, 2:53 PM
10 points
6 comments3 min readLW link
(dkl9.net)

Monthly Roundup #22: Septem­ber 2024

ZviSep 17, 2024, 12:20 PM
35 points
10 comments45 min readLW link
(thezvi.wordpress.com)

I fi­nally got ChatGPT to sound like me

lsusrSep 17, 2024, 9:39 AM
47 points
18 comments6 min readLW link

Food, Pri­son & Ex­otic An­i­mals: Sparse Au­toen­coders De­tect 6.5x Perform­ing Youtube Thumbnails

Louka Ewington-PitsosSep 17, 2024, 3:52 AM
6 points
2 comments7 min readLW link

Head in the Cloud: Why an Upload of Your Mind is Not You

xhqSep 17, 2024, 12:25 AM
−11 points
3 comments14 min readLW link

[Question] How does some­one prove that their gen­eral in­tel­li­gence is above av­er­age?

M. Y. ZuoSep 16, 2024, 9:01 PM
−3 points
12 comments1 min readLW link

[Question] Does life ac­tu­ally lo­cally *in­crease* en­tropy?

tailcalledSep 16, 2024, 8:30 PM
10 points
27 comments1 min readLW link