© Neuronpedia 2026

Privacy & Terms Blog GitHub Slack Twitter Contact

Neuronpedia

Jacobian LensNEW

Natural Language

NEW Assistant AxisNEW Circuit TracerUPDATESteer SAE Evals ExportsAPI Community Blog Privacy & Terms Contact

GPT2-SMALL · 8-TRES-DC · 554 ｜ Neuronpedia

Home
GPT2-Small
Transcoders Residuals
8-TRES-DC
554

INDEX

Explanations

occurrences of the word "once" in various contexts

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

otomy

-0.76

chnology

-0.75

ilon

-0.72

 horizon

-0.70

arching

-0.67

chini

-0.66

orrow

-0.64

driving

-0.64

pring

-0.63

eering

-0.63

POSITIVE LOGITS

 Joyce

0.63

APR

0.63

FACE

0.62

 Commissioner

0.61

 Judd

0.60

amine

0.60

 Roger

0.60

FAT

0.59

 Nasa

0.59

EY

0.59

Activations Density 0.012%

No Known Activations