© 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 · 72029 ｜ Neuronpedia

Home
Gemma-2-27B
22-GEMMASCOPE-RES-131K
72029

INDEX

Explanations

Flow injection

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

our

-1.63

 each

-1.30

for

-1.28

all

-1.24

 four

-1.23

 three

-1.22

 wybrać

-1.19

 eight

-1.18

 less

-1.18

or

-1.18

POSITIVE LOGITS

 according

1.29

he

1.26

ztes

1.23

膣

1.23

部分が

1.20

 amostra

1.20

 เป็น

1.19

IBLIO

1.19

according

1.17

 hipó

1.16

Activations Density 0.006%

No Known Activations