14 Oct 2024 22:51 UTC

47 points

14 comments3 min readLW link

How long should political (and other) terms be?

ohmurphy14 Oct 2024 21:38 UTC

5 points

0 comments1 min readLW link

(ohmurphy.substack.com)

Examples of How I Use LLMs

jefftk14 Oct 2024 17:10 UTC

31 points

2 comments2 min readLW link

(www.jefftk.com)

It’s important to know when to stop: Mechanistic Exploration of Gemma 2 List Generation

Gerard Boxo14 Oct 2024 17:04 UTC

9 points

0 comments6 min readLW link

(gboxo.github.io)

[Question] LW resources on childhood experiences?

nahir9159514 Oct 2024 17:04 UTC

10 points

7 comments1 min readLW link

Free Will, Neurotypical Dominance, and the Path to ASI and Neuralinks: Evolving Beyond Scarcity

j_passeri14 Oct 2024 16:54 UTC

−1 points

3 comments3 min readLW link

Breakthroughs, Neurodivergence, and Working Outside the System

j_passeri14 Oct 2024 16:54 UTC

2 points

3 comments2 min readLW link

The case for unlearning that removes information from LLM weights

Fabien Roger14 Oct 2024 14:08 UTC

102 points

18 comments6 min readLW link

Circuits in Superposition: Compressing many small neural networks into one

Lucius Bushnaq and jake_mendel

14 Oct 2024 13:06 UTC

131 points

9 comments13 min readLW link

Beyond Defensive Technology

ejk6414 Oct 2024 11:34 UTC

11 points

1 comment10 min readLW link

Why Stop AI is barricading OpenAI

Remmelt14 Oct 2024 7:12 UTC

−16 points

32 comments6 min readLW link

(docs.google.com)

The Explore vs. Exploit Dilemma

nathanjzhao14 Oct 2024 6:20 UTC

1 point

0 comments1 min readLW link

(nathanzhao.cc)

AI Alignment via Slow Substrates: Early Empirical Results With StarCraft II

Lester Leong14 Oct 2024 4:05 UTC

60 points

9 comments12 min readLW link

some questionable space launch guns

bhauth13 Oct 2024 22:52 UTC

17 points

0 comments4 min readLW link

(bhauth.com)

[Question] What are your favorite books or blogs that are out of print, or whose domains have expired (especially if they also aren’t on LibGen/Wayback/etc, or on Amazon)?

Arjun Panickssery13 Oct 2024 20:21 UTC

13 points

4 comments1 min readLW link

The Hopium Wars: the AGI Entente Delusion

Max Tegmark13 Oct 2024 17:00 UTC

236 points

60 comments9 min readLW link

Parental Writing Selection Bias

jefftk13 Oct 2024 14:00 UTC

52 points

3 comments1 min readLW link

(www.jefftk.com)

Personal Philosophy

Xor13 Oct 2024 3:01 UTC

3 points

0 comments2 min readLW link

Contagious Beliefs—Simulating Political Alignment

James Stephen Brown13 Oct 2024 0:27 UTC

8 points

0 comments2 min readLW link

(nonzerosum.games)

Binary encoding as a simple explicit construction for superposition

tailcalled12 Oct 2024 21:18 UTC

12 points

0 comments1 min readLW link

[Question] How Should We Use Limited Time to Maximize Long-Term Impact?

queelius12 Oct 2024 20:02 UTC

10 points

3 comments1 min readLW link

A Percentage Model of a Person

Sable12 Oct 2024 17:55 UTC

41 points

5 comments9 min readLW link

(affablyevil.substack.com)

AI Compute governance: Verifying AI chip location

Farhan12 Oct 2024 17:36 UTC

6 points

0 comments6 min readLW link

Geoffrey Hinton on the Past, Present, and Future of AI

Stephen McAleese12 Oct 2024 16:41 UTC

23 points

5 comments18 min readLW link

[Question] I = W/T?

HNX12 Oct 2024 15:15 UTC

0 points

3 comments1 min readLW link

AI research assistants competition 2024Q3: Tie between Elicit and You.com

Elizabeth12 Oct 2024 15:10 UTC

64 points

4 comments3 min readLW link

(acesounderglass.com)

SAE features for refusal and sycophancy steering vectors

neverix, Dmitrii Kharlapenko, Arthur Conmy and Neel Nanda

12 Oct 2024 14:54 UTC

29 points

4 comments7 min readLW link

Prices are Bounties

Maxwell Tabarrok12 Oct 2024 14:51 UTC

51 points

13 comments2 min readLW link

(www.maximum-progress.com)

Differential knowledge interconnection

Roman Leventov12 Oct 2024 12:52 UTC

6 points

0 comments7 min readLW link

Most arguments for AI Doom are either bad or weak

Logan Zoellner12 Oct 2024 11:57 UTC

4 points

100 comments3 min readLW link

Kassel ACX/LW Meetup

Fernand012 Oct 2024 7:47 UTC

2 points

0 comments1 min readLW link

Neural Network And Newton’s Second Law

Max Ma12 Oct 2024 6:25 UTC

−10 points

0 comments1 min readLW link

[Question] If the DoJ goes through with the Google breakup,where does Deepmind end up?

O O12 Oct 2024 5:06 UTC

5 points

1 comment1 min readLW link

My motivation and theory of change for working in AI healthtech

Andrew_Critch12 Oct 2024 0:36 UTC

180 points

39 comments14 min readLW link

HDBSCAN is Surprisingly Effective at Finding Interpretable Clusters of the SAE Decoder Matrix

Jaehyuk Lim, Kanishk Tantia and Sinem

11 Oct 2024 23:06 UTC

8 points

2 comments10 min readLW link

Changing the Mind of an LLM

testingthewaters11 Oct 2024 22:25 UTC

2 points

0 comments5 min readLW link

EIS XIV: Is mechanistic interpretability about to be practically useful?

scasper11 Oct 2024 22:13 UTC

68 points

4 comments7 min readLW link

Dario Amodei — Machines of Loving Grace

Matrice Jacobine11 Oct 2024 21:43 UTC

63 points

26 comments1 min readLW link

(darioamodei.com)

“Deep Galactic Chillout” a space to relax during SF tech week & meet wholesome, fun people

Jared M.11 Oct 2024 19:50 UTC

1 point

0 comments1 min readLW link

Open letter to young EAs

Leif Wenar11 Oct 2024 19:49 UTC

9 points

10 comments1 min readLW link

The Great Bootstrap

KristianRonn11 Oct 2024 19:46 UTC

12 points

0 comments15 min readLW link

Embracing complexity when developing and evaluating AI responsibly

Aliya Amirova11 Oct 2024 17:46 UTC

3 points

9 comments9 min readLW link

How much I’m paying for AI productivity software (and the future of AI use)

jacquesthibs11 Oct 2024 17:11 UTC

59 points

18 comments8 min readLW link

(jacquesthibodeau.com)

AI: The Philosopher’s Stone of the 21st Century

HNX11 Oct 2024 16:55 UTC

0 points

2 comments29 min readLW link

[Question] Who created the Less Wrong Gather Town?

Arepo11 Oct 2024 8:53 UTC

2 points

1 comment1 min readLW link

A Heuristic Proof of Practical Aligned Superintelligence

Roko11 Oct 2024 5:05 UTC

7 points

6 comments1 min readLW link

(transhumanaxiology.substack.com)

An AI crash is our best bet for restricting AI

Remmelt11 Oct 2024 2:12 UTC

26 points

3 comments1 min readLW link

A Triple Decker for Elfland

jefftk11 Oct 2024 1:50 UTC

25 points

0 comments1 min readLW link

(www.jefftk.com)

OODA your OODA Loop

Raemon11 Oct 2024 0:50 UTC

38 points

3 comments3 min readLW link

Scaling prediction markets with meta-markets

Dentosal10 Oct 2024 21:17 UTC

1 point

0 comments2 min readLW link