Archive
Sequences
About
Search
Log In
Questions
Events
Shortform
Alignment Forum
AF Comments
Home
Featured
All
Tags
Recent
Comments
RSS
Sohaib Imran
Karma:
167
Research associate at the University of Lancaster, studying AI safety.
My current views on AI:
All
Posts
Comments
New
Top
Old
[Question]
Does the ChatGPT (web)app sometimes show actual o1 CoTs now?
Sohaib Imran
Jan 29, 2025, 5:27 PM
6
points
6
comments
1
min read
LW
link
Cross-context abduction: LLMs make inferences about procedural training data leveraging declarative facts in earlier training data
Sohaib Imran
Nov 16, 2024, 11:22 PM
36
points
11
comments
14
min read
LW
link
Artefacts generated by mode collapse in GPT-4 Turbo serve as adversarial attacks.
Sohaib Imran
Nov 10, 2023, 3:23 PM
11
points
0
comments
2
min read
LW
link
Introduction
Robert Kralisch
,
Eris
,
teahorse
and
Sohaib Imran
Jun 30, 2023, 8:45 PM
8
points
0
comments
2
min read
LW
link
Inherently Interpretable Architectures
Robert Kralisch
,
teahorse
,
Eris
and
Sohaib Imran
Jun 30, 2023, 8:43 PM
4
points
0
comments
7
min read
LW
link
Positive Attractors
Robert Kralisch
,
teahorse
,
Eris
and
Sohaib Imran
Jun 30, 2023, 8:43 PM
6
points
0
comments
13
min read
LW
link
AISC 2023, Progress Report for March: Team Interpretable Architectures
Robert Kralisch
,
Eris
,
teahorse
and
Sohaib Imran
Apr 2, 2023, 4:19 PM
14
points
0
comments
14
min read
LW
link
Back to top