© Neuronpedia 2026

Privacy & Terms Blog GitHub Slack Twitter Contact

Jacobian LensNEW

Natural Language

MAY 26 Assistant AxisJAN 26 Circuit TracerAPR 26Steer SAE Evals Exports interp-engineNEW Community Blog Privacy & Terms Contact

GEMMA-2-27B · 34-GEMMASCOPE-RES-131K · 3473 ｜ Neuronpedia

Home
Gemma-2-27B
34-GEMMASCOPE-RES-131K
3473

INDEX

Explanations

cannot or not

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

Negative Logits

isn

-2.81

 didn

-2.61

 aren

-2.48

 doesn

-2.45

Couldn

-2.41

 weren

-2.34

 couldn

-2.22

 wasn

-2.20

 hasn

-2.13

 wouldn

-2.09

POSITIVE LOGITS

 cannot

2.42

Cannot

1.98

 Cannot

1.96

cannot

1.88

did

1.69

 does

1.66

 tidak

1.64

Tidak

1.45

do

1.41

 Tidak

1.38

Activations Density 0.070%

No Known Activations