© Neuronpedia 2026

Privacy & Terms Blog GitHub Slack Twitter Contact

Neuronpedia

Natural Language

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

GPT2-SMALL · 8-TRES-DC · 1179 ｜ Neuronpedia

Home
GPT2-Small
Transcoders Residuals
8-TRES-DC
1179

INDEX

Explanations

phrases or concepts related to separation or distinction

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

senal

-0.68

onica

-0.67

ERO

-0.66

KT

-0.65

amy

-0.64

auri

-0.64

Gy

-0.63

inar

-0.61

vr

-0.61

othy

-0.61

POSITIVE LOGITS

has

1.27

 applies

1.22

 develops

1.21

 considers

1.19

 accepts

1.17

 expects

1.15

 went

1.15

 does

1.15

 hasn

1.15

 cannot

1.14

Activations Density 6.041%

No Known Activations