© 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 · 127 ｜ Neuronpedia

Home
GPT2-Small
Transcoders Residuals
8-TRES-DC
127

INDEX

Explanations

elements related to qualification criteria and competitive structures

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

bably

-0.85

alore

-0.79

azo

-0.75

rop

-0.75

ofi

-0.74

rower

-0.73

avorite

-0.72

oaded

-0.72

 favoring

-0.71

inning

-0.71

POSITIVE LOGITS

FAQ

0.89

 Past

0.75

 Each

0.73

 Levels

0.73

 Failure

0.73

 Meaning

0.72

 Different

0.72

Each

0.72

Past

0.71

Rew

0.71

Activations Density 0.264%

No Known Activations