© 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 · 590 ｜ Neuronpedia

Home
GPT2-Small
Transcoders Residuals
8-TRES-DC
590

INDEX

Explanations

terms related to various aspects of health and well-being

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

 Suite

-0.79

 Savior

-0.78

unts

-0.67

 Generation

-0.66

 Innocent

-0.66

 actionGroup

-0.62

ppe

-0.61

 alike

-0.60

 ensu

-0.60

 Fame

-0.60

POSITIVE LOGITS

cyt

0.81

yip

0.78

cultural

0.74

helial

0.73

psych

0.72

oscopic

0.71

gebra

0.70

sterdam

0.69

 renal

0.69

sexual

0.68

Activations Density 1.475%

No Known Activations