Seonglae ChoBy analyzing feature sensitivity, they present evidence that features may exist not as simple directions but as multi-dimensional manifolds that preserve “mathematical linearity.”
Circuits Updates - July 2024
We report a number of developing ideas on the Anthropic interpretability team, which might be of interest to researchers working actively in this space. Some of these are emerging strands of research where we expect to publish more on in the coming months. Others are minor points we wish to share, since we're unlikely to ever write a paper about them.
https://transformer-circuits.pub/2024/july-update/index.html
When Models Manipulate Manifolds: The Geometry of a Counting Task
We find geometric structure underlying the mechanisms of a fundamental language model behavior.
https://transformer-circuits.pub/2025/linebreaks/index.html
The Confidence Manifold: Geometric Structure of Correctness...
When a language model asserts that "the capital of Australia is Sydney," does it know this is wrong? We characterize the geometry of correctness representations across 9 models from 5 architecture...
https://arxiv.org/abs/2602.08159
Steering Along Manifolds to Control Neural Networks
Concept geometry provides a blueprint for controlling the behavior of neural networks—if you know how to look.
https://www.goodfire.ai/research/manifold-steering#
