Neuronpedia

APISteer SAE Evals Blog/PodcastNEW Slack Privacy & Terms Contact

© Neuronpedia 2025

Privacy & Terms Blog/Podcast GitHub Slack Twitter Contact

Home
Dunefsky · Chlenski · Transcoders Enable Fine-Grained Interpretable Circuit Analysis
GPT2-Small
Transcoders Residuals
9-TRES-DC
626

INDEX

Explanations

instances of looking or gazing outward

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

berman

-0.75

hl

-0.67

 Bord

-0.65

dinand

-0.64

vae

-0.62

bourne

-0.62

erville

-0.60

uate

-0.58

lus

-0.58

 Gentle

-0.57

POSITIVE LOGITS

tg

0.75

ilee

0.70

0.65

AKING

0.63

0.60

 laure

0.59

este

0.58

ONEY

0.58

ocation

0.57

eon

0.57

Activations Density 0.640%

No Known Activations