x

LESSWRONG

LW

StefanHex's Shortform

5th Jul 2024

1 min read

5

This is a special post for quick takes by StefanHex. Only they can create top-level comments. Comments here also appear on the Quick Takes page and All Posts page.

Rendering 0/108 comments, sorted by

(show all) Click to highlight new comments since: Today at 2:25 AM

More from StefanHex

Curated and popular this week

108

StefanHex's Shortform — LessWrong

Comment Permalink

If you believe in the natural or convergent abstraction hypothesis, then "human comprehensible" isn't necessarily arbitrary.

Mentioned in

127A List of 45+ Mech Interp Project Ideas from Apollo Research’s Interpretability Team

71[Interim research report] Activation plateaus & sensitive directions in GPT2

52Sparsely-connected Cross-layer Transcoders

49Gradual Disempowerment Monthly Roundup #3

39Finding features in Transformers: Contrastive directions elicit stronger low-level perturbation responses than baselines

Load More (5/6)