© 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-3-27B · 31-GEMMASCOPE-2-RES-262K · 156086 ｜ Neuronpedia

Home
Gemma-3-27B
31-GEMMASCOPE-2-RES-262K
156086

INDEX

Explanations

which is

np_acts-logits-general · gemini-2.5-flash-lite

modifies or corrects

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Configuration

google/gemma-scope-2-27b-pt/resid_post/layer_31_width_262k_l0_medium

Prompts (Dashboard)

392,802 prompts, 256 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

৫

0.51

ة

0.46

४

0.46

priv

0.45

지

0.45

 प्रतिनिधि

0.44

ungk

0.44

作業

0.44

ându

0.44

드

0.44

POSITIVE LOGITS

 Springs

0.53

 Battle

0.49

 Streets

0.48

 Shat

0.46

 Parisi

0.46

 flotte

0.45

 Modern

0.45

 Warm

0.44

 Collins

0.44

 Mississippi

0.43

Activations Density 0.004%

No Known Activations