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
628

INDEX

Explanations

statements asserting truths or realities

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

sent

-0.78

soType

-0.77

MpServer

-0.67

asus

-0.65

assetsadobe

-0.64

�

-0.64

League

-0.64

nian

-0.62

shift

-0.62

stad

-0.61

POSITIVE LOGITS

 that

0.86

 THAT

0.75

 thats

0.67

prone

0.58

we

0.58

if

0.57

you

0.57

andy

0.56

that

0.56

 anybody

0.55

Activations Density 0.056%

No Known Activations