Record
- Authors
- Lange, G.; RGRGRG; Dearstyne, K.; Maher, K.
- Year
- 2026
- Title
- Cross-Layer Transcoders are incentivized to learn Unfaithful Circuits
- Published in
- LessWrong, 2 February 2026
- Type
- Web page
- Links
- https://www.lesswrong.com/posts/6CS2NDmoLCFcEJMor/cross-layer-transcoders-are-incentivized-to-learn-unfaithful
- Our note
- none yet
- Verification
- url-ok: the web address responds (the page itself is not checked).
https://www.lesswrong.com/posts/6CS2NDmoLCFcEJMor/cross-layer-transcoders-are-incentivized-to-learn-unfaithful: HTTP 200, checked 2026-11-14 - Id
lange-2026-cross-layer-transcoders-are-incentivized-learn
Cited on
- A Toy Model of Mechanistic (Un)Faithfulness read status: not stated
- Circuit Tracing: Revealing Computational Graphs in Language Models read status: not stated
As cited
The pages cite this work in 2 forms.
-
Lange, G., RGRGRG, Dearstyne, K., and Maher, K. (2026). Cross-Layer Transcoders are incentivized to learn Unfaithful Circuits. LessWrong, 2 February 2026. https://www.lesswrong.com/posts/6CS2NDmoLCFcEJMor/cross-layer-transcoders-are-incentivized-to-learn-unfaithful
-
Lange, G., RGRGRG, Dearstyne, K., and Maher, K. (2026, February 2). Cross-Layer Transcoders are incentivized to learn Unfaithful Circuits. LessWrong. https://www.lesswrong.com/posts/6CS2NDmoLCFcEJMor/cross-layer-transcoders-are-incentivized-to-learn-unfaithful
BibLaTeX
@online{lange-2026-cross-layer-transcoders-are-incentivized-learn,
author = {Lange, G. and {RGRGRG} and Dearstyne, K. and Maher, K.},
title = {{Cross-Layer Transcoders are incentivized to learn Unfaithful Circuits}},
organization = {LessWrong, 2 February 2026},
date = {2026},
url = {https://www.lesswrong.com/posts/6CS2NDmoLCFcEJMor/cross-layer-transcoders-are-incentivized-to-learn-unfaithful}
}