finding
active
finding:gemma-scope-loss-recovered-and-autointerp-improve-with-width-absorption-scr-tpp-degrade-at-larger-widths-unlearning-near-zero-at-final-layer

Gemma-Scope: loss recovered and autointerp improve with width; absorption, SCR, TPP degrade at larger widths; unlearning near zero at final layer

Scaling trends at larger scale.

Source paper

extracted_from
SAEBench: A Comprehensive Benchmark for Sparse Autoencoders in Language Model Interpretability
(2025) · Adam Karvonen · Can Rager · Johnny Lin · Curt Tigges +11

Neighborhood — ranked by edge-count

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.