© Neuronpedia 2026

Privacy & Terms Blog GitHub Slack Twitter Contact

Neuronpedia

Jacobian LensNEW

Natural Language

NEW Assistant AxisNEW Circuit TracerUPDATESteer SAE Evals Exports interp-engineNEWAPI Community Blog Privacy & Terms Contact

GPT2-SMALL · 8-TRES-DC · 2772 ｜ Neuronpedia

Home
GPT2-Small
Transcoders Residuals
8-TRES-DC
2772

INDEX

Explanations

references to organizations or councils

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

 Apprentice

-0.78

ococ

-0.69

Cre

-0.66

 Lisp

-0.65

 Corn

-0.65

Bow

-0.64

Adapt

-0.63

 Lizard

-0.63

 Dwarf

-0.62

Cow

-0.62

POSITIVE LOGITS

print

0.78

ournals

0.76

fights

0.74

prints

0.73

heses

0.73

ahime

0.73

ierrez

0.73

sonian

0.71

 dated

0.71

senal

0.69

Activations Density 0.170%

No Known Activations