SleepFM

EEG transformer foundation model for sleep staging, one of the three analyzed architectures.

Neighborhood — ranked by edge-count

paper

concept

EEG foundation models
associated_with
Large transformer models pretrained on EEG data for clinical tasks; the object of interpretability in this paper.

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.

REM Sleepconcept0.753
Associated with reevaluation of posteriors under new model structure after non-REM synaptic regression
Non-REM Sleepconcept0.713
Associated with synaptic homeostasis and elimination of redundant model parameters (Bayesian model reduction)
Harvard Medical School, Division of Sleep Medicineinstitute0.677
Hobson's institutional affiliation
Virtual Machine Functionalism (VMF)framework0.665
A philosophy-of-mind view that mental phenomena are implemented by virtual machines running on physical brains, irreducible to physics alone.
A single hyperparameter procedure driven by the intrinsic dictionary health audit transfers robustly across SleepFM, REVE, and LaBraM.finding0.665
Demonstrates architecture-agnostic applicability of the SAE tuning method
Daylightconcept0.652
Natural ambient light inside dwellings, a key performance indicator for housing quality in the chapter.