© 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 · 709 ｜ Neuronpedia

Home
GPT2-Small
Transcoders Residuals
8-TRES-DC
709

INDEX

Explanations

instances of awareness related to various subjects

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

Not in Any Lists

No Comments

Negative Logits

イト

-0.72

 redes

-0.70

 helicop

-0.70

 happens

-0.69

enegger

-0.68

サ

-0.67

ヘ

-0.64

 paran

-0.64

Pokémon

-0.63

pmwiki

-0.61

POSITIVE LOGITS

vier

0.73

 Sultan

0.67

vant

0.67

apeake

0.66

 Geoffrey

0.64

 Hale

0.64

fy

0.63

LER

0.62

rov

0.62

ffen

0.61

Activations Density 0.117%

No Known Activations