pith. sign in

How llms learn: Tracing internal representations with sparse au- toencoders

2 Pith papers cite this work. Polarity classification is still indexing.

2 Pith papers citing it

years

2026 2

verdicts

UNVERDICTED 2

representative citing papers

Features have life history. And we should care

q-bio.NC · 2026-05-07 · unverdicted · novelty 5.0

Language model features form an early stable carrier scaffold of about 50 sparse features that is load-bearing, predictable from onset firing, and recruits most later features.

citing papers explorer

Showing 2 of 2 citing papers.

  • Improving Robustness In Sparse Autoencoders via Masked Regularization cs.LG · 2026-04-07 · unverdicted · none · ref 9

    Masked regularization in sparse autoencoders disrupts token co-occurrences to reduce feature absorption, enhance probing, and narrow OOD gaps across architectures and sparsity levels.

  • Features have life history. And we should care q-bio.NC · 2026-05-07 · unverdicted · none · ref 7

    Language model features form an early stable carrier scaffold of about 50 sparse features that is load-bearing, predictable from onset firing, and recruits most later features.