wrecking-ball intervention

Type of concept steering intervention that catastrophically collapses global model performance.

Neighborhood — ranked by edge-count

paper

method

Concept Steering
associated_with
Latent intervention technique that manipulates sparse features to steer model predictions toward desired concepts.

concept

Representational Failure
extends
A failure mode exposed by the SAE framework where model representations are entangled or collapse under intervention

event

Mechanistic Interpretability of EEG Foundation Models via Sparse Autoencoders (2026)
mentions
Preprint applying TopK SAEs to three EEG transformers to reveal sparse feature dictionaries, steering regimes, and spectral interpretation.

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.

wrecking-ball interventions that collapse global model performancequote0.783
Load-bearing phrase describing catastrophic steering effects.
Concept interventions on some concepts act as 'wrecking-ball' interventions, collapsing global model performance.finding0.762
Observation of catastrophic performance drop when steering certain concepts.
Interchange Interventionmethod0.740
Fundamental operation for causal abstraction analysis; forces neurons to take values from source inputs to create counterfactuals.
Wrecking-ball interventions that collapse global model performance are empirically identified in EEG foundation models.finding0.736
Demonstrates a critical failure mode of concept steering with clinical safety implications
steering (intervention on internals)concept0.732
General technique of modifying activations to control model behavior.
Intervention Propagationconcept0.721
Property that additive modifications to activations affect all downstream computations, enabling tractable behavioral control
Causal Intervention on Representationsconcept0.716
The use of interventions (rather than correlations) to establish a causal link between representation geometry and behavioral geometry.
Causal Intervention via Activation Shiftingmethod0.714
Method of shifting hidden state activations along probe directions to cause the model to treat false statements as true and vice versa; evaluated on OOD inputs