Our new X account is live! Follow @wizwand_team for updates

Feature Interpretability on LLM Activations

86.9AutoInterp Score

HSAE

Updated 4d ago

Evaluation Results

Method	Links
HSAE 2026.02		86.9	32.4
Baseline 2026.02		86.1	33.8
HSAE 2026.02		85.9	31.4
Baseline 2026.02		85.4	32
Baseline 2026.02		82.9	30.2
HSAE 2026.02		81.4	28.6
Baseline 2026.02		81.3	24.5
HSAE 2026.02		80.7	27.2