© 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 · 698 ｜ Neuronpedia

Home
GPT2-Small
Transcoders Residuals
8-TRES-DC
698

INDEX

Explanations

references to the speaker or self-identity

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

oir

-0.80

̶

-0.75

 glim

-0.71

aum

-0.66

pered

-0.65

essee

-0.65

uve

-0.65

Torrent

-0.64

ussia

-0.63

Va

-0.62

POSITIVE LOGITS

 being

0.73

joining

0.73

 hanging

0.70

 appearing

0.66

self

0.65

 joining

0.65

 graduated

0.64

 onstage

0.63

 ascending

0.62

 dangling

0.62

Activations Density 0.159%

No Known Activations