method
active
method:spectral-decoderSpectral Decoder
Method that maps latent concept steering interventions back to EEG amplitude spectrum to obtain physiologically interpretable frequency signatures.
Neighborhood — ranked by edge-count
Papers (1)
paper
Concepts (2)
concept
- pathological slow-wave suppressionintroducesEEG frequency signature of reduced slow-wave activity, obtained as a spectral interpretation of steering.
- α-band restorationintroducesRestoration of alpha band (8–12 Hz) power in EEG, a physiological signature obtained from spectral decoding.
Claims (1)
claim
- Key result linking abstract latent manipulations to known EEG neurophysiology
Events (1)
event
- Preprint applying TopK SAEs to three EEG transformers to reveal sparse feature dictionaries, steering regimes, and spectral interpretation.
Related by similarity (8)
cosine ≥ 0.65 · no typed edgeEntities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.
- Physiological interpretability result linking latent steering to EEG frequency signatures.
- Base architecture of reasoning LLMs studied, with attention and MLP blocks per layer
- Correlative technique measuring the type of information encoded in distributed representations via linear predictability.
- Framework learning end-to-end mappings from activations through sparse concept bottlenecks to behavioral predictions
- Links latent space manipulation to known EEG neurophysiology
- Links latent space manipulation to known EEG neurophysiology
- Temperature=0.8 sampled decoding for self-report; reduces collapse moderately but remains discrete and noisy
- Claim that the spectral decoder adds physiological interpretability.