© Neuronpedia 2026

Privacy & Terms Blog GitHub Slack Twitter Contact

Jacobian LensNEW

Natural Language

MAY 26 Assistant AxisJAN 26 Circuit TracerAPR 26Steer SAE Evals Exports interp-engineNEW Community Blog Privacy & Terms Contact

GPT2-SMALL · 8-TRES-DC · 3082 ｜ Neuronpedia

Home
GPT2-Small
Transcoders Residuals
8-TRES-DC
3082

INDEX

Explanations

actions associated with achieving results or consequences

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Embeds

Show PlotsShow ExplanationShow ActivationsShow Test FieldShow SteerShow Link

IFrame

Link

Negative Logits

Money

-0.71

article

-0.69

 Ending

-0.67

 introductory

-0.64

Foreign

-0.62

 Recomm

-0.61

arettes

-0.60

erred

-0.57

 tert

-0.57

�

-0.56

POSITIVE LOGITS

�

0.70

 keen

0.70

atically

0.67

 dism

0.66

 tremend

0.62

keyes

0.62

↑

0.61

 sway

0.61

ealous

0.60

ossession

0.60

Activations Density 0.177%

No Known Activations