Manifold-respecting steering produces smooth natural behavioral trajectories while linear steering teleports between non-adjacent concepts.

Neighborhood — ranked by edge-count

Communities (3)

community

Manifold-aware concept steering in neural representations
members_of
Explores geometry of activation/behavior manifolds to enable selective, non-destructive concept interventions.
Geometric concept representations in neural networks
members_of
Concepts encoded as curved manifolds and circular structures in LLM activation spaces.
Manifold-aware steering for language models
members_of
Geometric approach to model control that respects learned concept structure, contrasting with linear steering that produces off-manifold artifacts.

Vectors (1)

vector

Interpretability as Microscope for Consciousness
addresses_vector

Source docs (1)

source_doc

2026-05-14_phil-trans-A-goodfire-aboutblank-impact.md
extracted_from

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.

Steering along manifolds provides better control than linear steering when the concept geometry is non-linear.claim0.872
The central thesis of the paper, motivating the shift from linear to geometry-aware manifold steering.
manifold steering produces clean probability shifts along natural behavior structure; linear steering cuts across manifold and produces off-target noisy effectsfinding0.864
Empirical demonstration on Llama-3.1-8B that steering along representation manifold aligns outputs with behavior manifold, whereas linear steering does not.
Linear steering produces noisy off-target effects; manifold steering cleanly shifts probability mass between sequential concepts.finding0.862
Core empirical claim comparing steering approaches on cyclic concepts.
Manifold-aware steering is non-trivial IP requiring geometric analysis, not a system-prompt implementation.claim0.840
Manifold steering demonstrates bidirectional geometry-behavior link in a video world model on tasks with geometry corresponding to physical dynamicsfinding0.831
Extension of manifold steering validation to video world models and physical dynamics tasks, demonstrating cross-modal generality
Linear steering cuts through off-manifold regions and hence produces unnatural outputs.claim0.820
Attribution of failure to Euclidean assumption.
There is a clear bidirectional relationship between the geometry of behavior and representation: steering along representation manifolds follows behavior manifolds, and vice versa.claim0.819
The paper's finding that the alignment holds in both directions — from representation to behavior and from behavior back to representation space.
Gap: linear steering assumes Euclidean geometry and does not account for the actual curved geometry of activation manifoldsconcept0.817
The research gap that motivates manifold steering as an alternative to conventional linear approaches