© 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 · 3253 ｜ Neuronpedia

Home
GPT2-Small
Transcoders Residuals
8-TRES-DC
3253

INDEX

Explanations

phrases indicating something that has not been seen or known

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

Crit

-0.73

artifacts

-0.71

 Noon

-0.69

etheless

-0.64

Earn

-0.64

uala

-0.62

••

-0.60

crit

-0.60

��

-0.59

cially

-0.58

POSITIVE LOGITS

 ILCS

0.71

iversal

0.69

onic

0.62

own

0.59

ranch

0.59

illions

0.58

yip

0.57

isSpecialOrderable

0.56

addin

0.56

 tune

0.56

Activations Density 0.060%

No Known Activations