Simple probes can catch sleeper agents

Monte M, Carson Denison, Zac Hatfield-Dodds, David Duvenaud, Sam Bowman, Ethan Perez and evhub

23 Apr 2024 21:10 UTC

85 points

7 comments1 min readLW link

(www.anthropic.com)

Dequantifying first-order theories

jessicata23 Apr 2024 19:04 UTC

32 points

0 comments8 min readLW link

(unstableontology.com)

Transformers Represent Belief State Geometry in their Residual Stream

Adam Shai16 Apr 2024 21:16 UTC

293 points

61 comments12 min readLW link

ProLU: A Pareto Improvement for Sparse Autoencoders

Glen Taggart23 Apr 2024 14:09 UTC

4 points

0 comments7 min readLW link

[Full Post] Progress Update #1 from the GDM Mech Interp Team

Neel Nanda, Arthur Conmy, lsgos, Senthooran Rajamanoharan, Tom Lieberum, János Kramár and Vikrant Varma

19 Apr 2024 19:06 UTC

69 points

8 comments8 min readLW link

Discriminating Behaviorally Identical Classifiers: a model problem for applying interpretability to scalable oversight

Sam Marks18 Apr 2024 16:17 UTC

90 points

5 comments12 min readLW link

[Summary] Progress Update #1 from the GDM Mech Interp Team

Neel Nanda, Arthur Conmy, lsgos, Senthooran Rajamanoharan, Tom Lieberum, János Kramár and Vikrant Varma

19 Apr 2024 19:06 UTC

66 points

0 comments3 min readLW link

Time complexity for deterministic string machines

alcatal21 Apr 2024 22:35 UTC

14 points

0 comments21 min readLW link

Inducing Unprompted Misalignment in LLMs

Sam Svenningsen, evhub and Henry Sleight

19 Apr 2024 20:00 UTC

34 points

6 comments16 min readLW link

LLM Evaluators Recognize and Favor Their Own Generations

Arjun Panickssery, Sam Bowman and Shi Feng

17 Apr 2024 21:09 UTC

43 points

1 comment3 min readLW link

(tiny.cc)

LLMs for Alignment Research: a safety priority?

abramdemski4 Apr 2024 20:03 UTC

138 points

23 comments11 min readLW link

Modern Transformers are AGI, and Human-Level

abramdemski26 Mar 2024 17:46 UTC

196 points

89 comments5 min readLW link

AXRP Episode 27 - AI Control with Buck Shlegeris and Ryan Greenblatt

DanielFilan11 Apr 2024 21:30 UTC

67 points

10 comments107 min readLW link

How We Picture Bayesian Agents

johnswentworth and David Lorell

8 Apr 2024 18:12 UTC

65 points

11 comments7 min readLW link

A Selection of Randomly Selected SAE Features

CallumMcDougall and Joseph Bloom

1 Apr 2024 9:09 UTC

104 points

2 comments4 min readLW link

Sparsify: A mechanistic interpretability research agenda

Lee Sharkey3 Apr 2024 12:34 UTC

84 points

20 comments22 min readLW link

SAE reconstruction errors are (empirically) pathological

wesg29 Mar 2024 16:37 UTC

85 points

15 comments8 min readLW link

SAE-VIS: Announcement Post

CallumMcDougall and Joseph Bloom

31 Mar 2024 15:30 UTC

73 points

8 comments1 min readLW link

How I select alignment research projects

Ethan Perez, Henry Sleight and Mikita Balesni

10 Apr 2024 4:33 UTC

34 points

4 comments24 min readLW link

PIBBSS is hiring in a variety of roles (alignment research and incubation program)

Nora_Ammann, Lucas Teixeira and DusanDNesic

9 Apr 2024 8:12 UTC

47 points

0 comments3 min readLW link