© Neuronpedia 2026

Privacy & Terms Blog GitHub Slack Twitter Contact

Neuronpedia

Jacobian LensNEW

Natural Language

NEW Assistant AxisNEW Circuit TracerUPDATESteer SAE Evals ExportsAPI Community Blog Privacy & Terms Contact

GEMMA-2-27B · 34-GEMMASCOPE-RES-131K · 970 ｜ Neuronpedia

Home
Gemma-2-27B
34-GEMMASCOPE-RES-131K
970

INDEX

Explanations

parsing strings

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

ဈ

-0.83

 oprot

-0.81

tivism

-0.76

お世話

-0.76

 Serialize

-0.76

chick

-0.73

banding

-0.73

嫁

-0.72

eroom

-0.70

 fiume

-0.69

POSITIVE LOGITS

 parse

3.70

parse

3.50

 parsed

3.20

 parsing

3.19

Parse

3.13

 Parse

3.13

parsed

2.66

Parsed

2.59

parsing

2.56

 Parsing

2.48

Activations Density 0.074%

No Known Activations