David Udell

Karma: 2,592

(Not) Explaining GPT-2-Small Forward Passes with Edge-Level Autoencoder Circuits

David Udell, hrdkbhatnagar and JacksonKaunismaa

Jul 22, 2025, 8:36 PM

23 points

8 votes

Overall karma indicates overall quality.

0 comments6 min readLW link

Why Can’t We Hypothesize After the Fact?

David UdellFeb 26, 2025, 10:41 PM

40 points

15 votes

Overall karma indicates overall quality.

3 comments2 min readLW link

Causal Graphs of GPT-2-Small’s Residual Stream

David UdellJul 9, 2024, 10:06 PM

53 points

18 votes

Overall karma indicates overall quality.

7 comments7 min readLW link

Sparse Coding, for Mechanistic Interpretability and Activation Engineering

David UdellSep 23, 2023, 7:16 PM

42 points

19 votes

Overall karma indicates overall quality.

7 comments34 min readLW link

ActAdd: Steering Language Models without Optimization

technicalities, TurnTrout, lisathiergart, David Udell, Ulisse Mini and Monte M

Sep 6, 2023, 5:21 PM

105 points

31 votes

Overall karma indicates overall quality.

3 comments2 min readLW link

(arxiv.org)

Steering GPT-2-XL by adding an activation vector

TurnTrout, Monte M, David Udell, lisathiergart and Ulisse Mini

May 13, 2023, 6:42 PM

439 points

206 votes

Overall karma indicates overall quality.

98 comments50 min readLW link 1 review

Understanding and controlling a maze-solving policy network

TurnTrout, peligrietzer, Ulisse Mini, Monte M and David Udell

Mar 11, 2023, 6:59 PM

334 points

130 votes

Overall karma indicates overall quality.

28 comments23 min readLW link

Beneath My Epistemic Dignity

David UdellFeb 28, 2023, 4:02 AM

6 points

10 votes

Overall karma indicates overall quality.

3 comments2 min readLW link

Probability Theory: The Logic of Science, Jaynes

David UdellFeb 16, 2023, 9:57 PM

29 points

13 votes

Overall karma indicates overall quality.

0 comments18 min readLW link

Rounding Someone Off

David UdellJan 24, 2023, 12:03 AM

25 points

10 votes

Overall karma indicates overall quality.

0 comments5 min readLW link

Consequentialists: One-Way Pattern Traps

David UdellJan 16, 2023, 8:48 PM

59 points

28 votes

Overall karma indicates overall quality.

3 comments14 min readLW link

Linear Algebra Done Right, Axler

David UdellJan 2, 2023, 10:54 PM

57 points

23 votes

Overall karma indicates overall quality.

6 comments9 min readLW link

Naive Set Theory, Halmos

David UdellDec 22, 2022, 2:34 AM

11 points

4 votes

Overall karma indicates overall quality.

1 comment8 min readLW link

Moorean Statements

David UdellOct 22, 2022, 12:50 AM

11 points

4 votes

Overall karma indicates overall quality.

11 comments1 min readLW link

Dath Ilan’s Views on Stopgap Corrigibility

David UdellSep 22, 2022, 4:16 PM

78 points

28 votes

Overall karma indicates overall quality.

19 comments13 min readLW link

(www.glowfic.com)

Guidelines for Mad Entrepreneurs

David UdellSep 16, 2022, 6:33 AM

31 points

13 votes

Overall karma indicates overall quality.

0 comments11 min readLW link

Framing AI Childhoods

David UdellSep 6, 2022, 11:40 PM

37 points

13 votes

Overall karma indicates overall quality.

8 comments4 min readLW link

The Shard Theory Alignment Scheme

David UdellAug 25, 2022, 4:52 AM

47 points

18 votes

Overall karma indicates overall quality.

32 comments2 min readLW link

“What Mistakes Are You Making Right Now?”

David UdellAug 15, 2022, 9:19 PM

13 points

5 votes

Overall karma indicates overall quality.

2 comments1 min readLW link

Shard Theory: An Overview

David UdellAug 11, 2022, 5:44 AM

167 points

75 votes

Overall karma indicates overall quality.

34 comments10 min readLW link

Keyboard shortcuts

Keys shown in yellow (e.g., ]) are accesskeys, and require a browser-specific modifier key (or keys).

Keys shown in grey (e.g., ?) do not require any modifier keys.

General
? Show keyboard shortcuts
Esc Hide keyboard shortcuts

Site navigation
h Go to Home (a.k.a. “Frontpage”) view
f Go to Featured (a.k.a. “Curated”) view
a Go to All (a.k.a. “Community”) view
m Go to Meta view
v Go to Tags view
c Go to Recent Comments view
r Go to Archive view
q Go to Sequences view
t Go to About page
u Go to User or Login page
o Go to Inbox page

Page navigation
, Jump up to top of page
. Jump down to bottom of page
/ Jump to top of comments section
s Search

Page actions
n New post or comment
e Edit current post

Post/comment list views
. Focus next entry in list
, Focus previous entry in list
; Cycle between links in focused entry
Enter Go to currently focused entry
Esc Unfocus currently focused entry
] Go to next page
[ Go to previous page
\ Go to first page
e Edit currently focused post

Editor
k Bold text
i Italic text
l Insert hyperlink
q Blockquote text

Appearance
= Increase text size
- Decrease text size
0 Reset to default text size
′ Cycle through content width settings
1 Switch to default theme [A]
2 Switch to dark theme [B]
3 Switch to grey theme [C]
4 Switch to ultramodern theme [D]
5 Switch to simple theme [E]
6 Switch to brutalist theme [F]
7 Switch to ReadTheSequences theme [G]
8 Switch to classic Less Wrong theme [H]
9 Switch to modern Less Wrong theme [I]
; Open theme tweaker
Enter Save changes and close theme tweaker
Esc Close theme tweaker (without saving)

Slide shows
l Start/resume slideshow
Esc Exit slideshow
→↓ Next slide
←↑ Previous slide
Space Reset slide zoom

Miscellaneous
x Switch to next view on user page
z Switch to previous view on user page
` Toggle compact comment list view
g Toggle anti-kibitzer