© 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

GEMMA-2-27B · 34-GEMMASCOPE-RES-131K · 107945 ｜ Neuronpedia

Home
Gemma-2-27B
34-GEMMASCOPE-RES-131K
107945

INDEX

Explanations

granite features

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Configuration

google/gemma-scope-27b-pt-res/layer_34/width_131k

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

No Configuration Found

Embeds

Show PlotsShow ExplanationShow ActivationsShow Test FieldShow SteerShow Link

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Gypsum

-0.81

 gypsum

-0.78

licator

-0.75

 peruano

-0.72

諫

-0.71

 Volcanic

-0.69

 बद

-0.69

mped

-0.68

 coal

-0.68

ību

-0.68

POSITIVE LOGITS

 granite

1.41

peg

1.23

 granit

1.09

 granito

1.00

lep

0.99

Granite

0.99

 Granite

0.97

 pink

0.96

 gneiss

0.95

peg

0.89

Activations Density 0.025%

No Known Activations