© 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 · 22-GEMMASCOPE-RES-131K · 15330 ｜ Neuronpedia

Home
Gemma-2-27B
22-GEMMASCOPE-RES-131K
15330

INDEX

Explanations

redirect(

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_22/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

-1.13

 precisar

-1.11

 começo

-1.07

 also

-1.07

 inizio

-1.05

Also

-1.05

llten

-1.03

开始

-1.02

 altre

-1.01

也沒

-1.01

POSITIVE LOGITS

 after

1.09

ONY

1.09

<bos>

1.08

wertes

1.07

 bahawa

1.05

)<<

1.04

 verwijzen

1.04

Å

1.03

ены

1.03

プリング

1.02

Activations Density 0.000%

No Known Activations