© Neuronpedia 2026

Privacy & Terms Blog GitHub Slack Twitter Contact

Neuronpedia

APIAssistant AxisNEW Circuit TracerNEW Steer SAE Evals Exports Community Blog Privacy & Terms Contact

GPT2-SMALL · 8-TRES-DC · 770 ｜ Neuronpedia

Home
GPT2-Small
Transcoders Residuals
8-TRES-DC
770

INDEX

Explanations

conditional phrases that suggest a cause-and-effect relationship

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

ebus

-0.74

rentice

-0.66

mit

-0.65

istance

-0.63

 Sins

-0.62

 Situation

-0.62

compan

-0.62

ewitness

-0.61

moderate

-0.61

grave

-0.61

POSITIVE LOGITS

 empt

0.76

 healthier

0.71

��

0.64

 smoother

0.63

 steril

0.62

omed

0.62

ainers

0.61

atures

0.60

lov

0.60

 suspic

0.59

Activations Density 0.019%

No Known Activations