© 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 · 129123 ｜ Neuronpedia

Home
Gemma-2-27B
22-GEMMASCOPE-RES-131K
129123

INDEX

Explanations

code blocks

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

to

-1.01

 victims

-0.96

Dawg

-0.96

-0.95

hah

-0.95

-0.94

lel

-0.94

hahah

-0.94

Ic

-0.93

のお知らせ

-0.93

POSITIVE LOGITS

 délicieux

1.13

viève

1.09

InkWell

1.08

استفاده

1.05

attamento

1.04

quête

1.03

آیا

1.03

arschijnlijk

1.02

 which

1.01

링

1.00

Activations Density 0.007%

No Known Activations