Arthur Conmy

Karma: 1,650

Intepretability

Views my own

Negative Results for SAEs On Downstream Tasks and Deprioritising SAE Research (GDM Mech Interp Team Progress Update #2)

lewis smith, Senthooran Rajamanoharan, Arthur Conmy, CallumMcDougall, Tom Lieberum, János Kramár, Rohin Shah and Neel Nanda

Mar 26, 2025, 7:07 PM

109 points

15 comments29 min readLW link

(deepmindsafetyresearch.medium.com)

The GDM AGI Safety+Alignment Team is Hiring for Applied Interpretability Research

Arthur Conmy and Neel Nanda

Feb 24, 2025, 2:17 AM

48 points

1 comment7 min readLW link

SAEBench: A Comprehensive Benchmark for Sparse Autoencoders

Can, Adam Karvonen, Johnny Lin, Curt Tigges, Joseph Bloom, chanind, Yeu-Tong Lau, Eoin Farrell, Arthur Conmy, CallumMcDougall, Kola Ayonrinde, Matthew Wearden, Sam Marks and Neel Nanda

Dec 11, 2024, 6:30 AM

82 points

6 comments2 min readLW link

(www.neuronpedia.org)

Evolutionary prompt optimization for SAE feature visualization

neverix, Daniel Tan, Dmitrii Kharlapenko, Neel Nanda and Arthur Conmy

Nov 14, 2024, 1:06 PM

21 points

0 comments9 min readLW link

SAEs are highly dataset dependent: a case study on the refusal direction

Connor Kissane, robertzk, Neel Nanda and Arthur Conmy

Nov 7, 2024, 5:22 AM

66 points

4 comments14 min readLW link

Open Source Replication of Anthropic’s Crosscoder paper for model-diffing

Connor Kissane, robertzk, Arthur Conmy and Neel Nanda

Oct 27, 2024, 6:46 PM

47 points

4 comments5 min readLW link

SAE features for refusal and sycophancy steering vectors

neverix, Dmitrii Kharlapenko, Arthur Conmy and Neel Nanda

Oct 12, 2024, 2:54 PM

29 points

4 comments7 min readLW link

Base LLMs refuse too

Connor Kissane, robertzk, Arthur Conmy and Neel Nanda

Sep 29, 2024, 4:04 PM

60 points

20 comments10 min readLW link

Extracting SAE task features for in-context learning

Dmitrii Kharlapenko, neverix, Neel Nanda and Arthur Conmy

Aug 12, 2024, 8:34 PM

31 points

1 comment9 min readLW link

Self-explaining SAE features

Dmitrii Kharlapenko, neverix, Neel Nanda and Arthur Conmy

Aug 5, 2024, 10:20 PM

60 points

13 comments10 min readLW link

JumpReLU SAEs + Early Access to Gemma 2 SAEs

Senthooran Rajamanoharan, Tom Lieberum, nps29, Arthur Conmy, Vikrant Varma, János Kramár and Neel Nanda

Jul 19, 2024, 4:10 PM

48 points

10 comments1 min readLW link

(storage.googleapis.com)

SAEs (usually) Transfer Between Base and Chat Models

Connor Kissane, robertzk, Arthur Conmy and Neel Nanda

Jul 18, 2024, 10:29 AM

66 points

0 comments10 min readLW link

Attention Output SAEs Improve Circuit Analysis

Connor Kissane, robertzk, Arthur Conmy and Neel Nanda

Jun 21, 2024, 12:56 PM

33 points

3 comments19 min readLW link

Improving Dictionary Learning with Gated Sparse Autoencoders

Senthooran Rajamanoharan, Arthur Conmy, lewis smith, Tom Lieberum, Vikrant Varma, János Kramár, Rohin Shah and Neel Nanda

Apr 25, 2024, 6:43 PM

63 points

38 comments1 min readLW link

(arxiv.org)

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

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

Apr 19, 2024, 7:06 PM

79 points

10 comments8 min readLW link

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

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

Apr 19, 2024, 7:06 PM

72 points

0 comments3 min readLW link

We Inspected Every Head In GPT-2 Small using SAEs So You Don’t Have To

robertzk, Connor Kissane, Arthur Conmy and Neel Nanda

Mar 6, 2024, 5:03 AM

63 points

0 comments12 min readLW link

Attention SAEs Scale to GPT-2 Small

Connor Kissane, robertzk, Arthur Conmy and Neel Nanda

Feb 3, 2024, 6:50 AM

78 points

4 comments8 min readLW link

Sparse Autoencoders Work on Attention Layer Outputs

Connor Kissane, robertzk, Arthur Conmy and Neel Nanda

Jan 16, 2024, 12:26 AM

83 points

9 comments18 min readLW link

My best guess at the important tricks for training 1L SAEs

Arthur ConmyDec 21, 2023, 1:59 AM

37 points

4 comments3 min readLW link