© 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 · 472 ｜ Neuronpedia

Home
GPT2-Small
Transcoders Residuals
8-TRES-DC
472

INDEX

Explanations

instances of the word "sharp" 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

uyomi

-0.76

ammy

-0.72

vous

-0.70

ankind

-0.68

 Forth

-0.67

 Causes

-0.66

Cause

-0.66

Lov

-0.65

chel

-0.65

ngth

-0.64

POSITIVE LOGITS

omic

0.80

ening

0.77

continental

0.71

 overest

0.70

ened

0.69

urn

0.64

north

0.63

opia

0.63

 varies

0.62

 differ

0.62

Activations Density 0.063%

No Known Activations