hypothesis

active

hypothesis:we-hypothesize-that-representation-geometry-drives-model-behavior-the-geometric-structure-of-internal-representations-causally-shapes-what-models-do-externally

We hypothesize that representation geometry drives model behavior — the geometric structure of internal representations causally shapes what models do externally.

The causal hypothesis motivating the use of causality (intervention) as the lens connecting representation and behavior geometry.

Source paper

extracted_from

Steering Along Manifolds to Control Neural Networks

Neighborhood — ranked by edge-count

Papers (1)

paper

Steering Along Manifolds to Control Neural Networks
introduces

Findings (1)

finding

The representation-based path and the behavior-based path in Llama-3.1 8B activation space trace out similar curves, demonstrating bidirectional geometry alignment.
associated_with
Key empirical result showing that optimizing for behavioral outputs and fitting representation geometry produce the same path in activation space.

Claims (1)

claim

The geometry of internal representations and the geometry of model behavior share a precise correspondence — representation geometry is a window into the inner world of neural networks.
supports
The paper's deepest interpretive claim, asserting that representation structure and behavioral structure are not coincidentally aligned but deeply connected.

Concepts (1)

concept

Causal Intervention on Representations
supports
The use of interventions (rather than correlations) to establish a causal link between representation geometry and behavioral geometry.

Related by similarity (8)

cosine ≥ 0.65 · no typed edge

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.

How does representation geometry causally drive model behavior?question0.907
The central scientific question the paper addresses through the lens of interventional causality.
Does the geometric structure of neural representations causally shape model behavior?question0.866
The motivating research question of the paper
Geometric structure of neural representations causally shapes model behaviorclaim0.858
The paper's core causal assertion: geometry is not incidental but mechanistically linked to behavior
There is a clear bidirectional relationship between the geometry of behavior and representation: steering along representation manifolds follows behavior manifolds, and vice versa.claim0.849
The paper's finding that the alignment holds in both directions — from representation to behavior and from behavior back to representation space.
Representation geometry causally shapes behavior; activation and behavior manifolds are approximately isometric.claim0.848
There is a bidirectional relationship between the geometry of representation and behavior across tasks and modalities.claim0.845
Author’s interpretive claim that the shared geometry is general and robust.
Neural representations carry rich geometric structure; but does that structure causally shape behavior?quote0.837
Opening sentence framing the paper's core inquiry.
There exists a bidirectional relationship between the geometry of neural representation and the geometry of model behaviorclaim0.837
Central empirical claim of the paper, demonstrated across tasks and modalities