© 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 · 599 ｜ Neuronpedia

Home
GPT2-Small
Transcoders Residuals
8-TRES-DC
599

INDEX

Explanations

terms related to location and context

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

pees

-0.71

 Notting

-0.67

ettings

-0.67

oresc

-0.64

aturdays

-0.63

irez

-0.63

EStreamFrame

-0.63

 Prix

-0.61

itent

-0.60

instein

-0.59

POSITIVE LOGITS

Choice

0.95

 choice

0.76

rated

0.67

way

0.64

 list

0.63

contract

0.62

grad

0.61

 calendars

0.60

 derog

0.59

hetically

0.59

Activations Density 2.414%

No Known Activations