© 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

Home
Gemma-2-27B
34-GEMMASCOPE-RES-131K
77856

INDEX

Explanations

screen

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

("\"

-0.75

펩

-0.75

校長

-0.72

 Stripe

-0.71

 утеп

-0.68

 striping

-0.68

 raids

-0.67

 chase

-0.67

魑

-0.67

Stripe

-0.66

POSITIVE LOGITS

 screen

2.13

screen

1.78

 Screen

1.77

Screen

1.67

 SCREEN

1.61

creen

1.57

 protector

1.48

SCREEN

1.45

屏幕

1.33

 screens

1.29

Activations Density 0.008%

No Known Activations

GEMMA-2-27B · 34-GEMMASCOPE-RES-131K · 77856 ｜ Neuronpedia