RSS

ProLU: A Pareto Im­prove­ment for Sparse Autoencoders

Glen Taggart23 Apr 2024 14:09 UTC
−3 points
0 comments6 min readLW link

Sub­jec­tive Ques­tions Re­quire Sub­jec­tive information

Ben23 Apr 2024 13:16 UTC
7 points
1 comment4 min readLW link

Re­ject­ing Television

Declan Molony23 Apr 2024 4:59 UTC
36 points
3 comments6 min readLW link

Take the wheel, Shog­goth! (Less­wrong is try­ing out changes to the front­page al­gorithm)

Ruby and RobertM
23 Apr 2024 3:58 UTC
50 points
3 comments4 min readLW link

Thoughts on Zero Points

depressurize23 Apr 2024 2:22 UTC
21 points
0 comments4 min readLW link
(sexandchicago.substack.com)

How LLMs Work, in the Style of The Economist

Rocket22 Apr 2024 19:06 UTC
1 point
0 comments2 min readLW link

Mea­sur­ing Co­her­ence and Goal-Direct­ed­ness in RL Policies

dx2622 Apr 2024 18:26 UTC
2 points
0 comments7 min readLW link

AI Reg­u­la­tion is Unsafe

Maxwell Tabarrok22 Apr 2024 16:37 UTC
31 points
8 comments4 min readLW link
(www.maximum-progress.com)

Pri­ors and Prejudice

MathiasKB22 Apr 2024 15:00 UTC
65 points
10 comments7 min readLW link

For­get Every­thing (Statis­ti­cal Me­chan­ics Part 1)

J Bostock22 Apr 2024 13:33 UTC
36 points
4 comments3 min readLW link

Should we break up Google Deep­Mind?

Hauke Hillebrandt22 Apr 2024 9:16 UTC
−6 points
0 comments1 min readLW link

What should our con­tain­ers do?

Richard Henage22 Apr 2024 6:17 UTC
3 points
1 comment2 min readLW link

Goal ori­ented cog­ni­tion in “a sin­gle for­ward pass”

dxu and habryka
22 Apr 2024 5:03 UTC
18 points
11 comments26 min readLW link

Time com­plex­ity for de­ter­minis­tic string machines

alcatal21 Apr 2024 22:35 UTC
14 points
0 comments21 min readLW link

Trans­fer Learn­ing in Humans

niplav21 Apr 2024 20:49 UTC
53 points
1 comment13 min readLW link

I cre­ated an Asi Align­ment Tier List

TimeGoat21 Apr 2024 18:44 UTC
−6 points
0 comments1 min readLW link

Fruits of our Labors In­tro­duc­tion: The Art of Weirdness

Bridgett Kay21 Apr 2024 17:34 UTC
2 points
2 comments4 min readLW link
(dxmrevealed.wordpress.com)

The los­ing iden­tity of Twitter

Itay Dreyfus21 Apr 2024 13:43 UTC
8 points
1 comment12 min readLW link
(productidentity.co)

Good Bings copy, great Bings steal

dr_s21 Apr 2024 9:52 UTC
29 points
6 comments9 min readLW link

Paper: “The Ethics of Ad­vanced AI As­sis­tants” -Google DeepMind

Tristan Wegner21 Apr 2024 6:45 UTC
20 points
0 comments1 min readLW link
(storage.googleapis.com)