© 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 · 1567 ｜ Neuronpedia

Home
Gemma-3-27B
31-GEMMASCOPE-2-RES-262K
1567

INDEX

Explanations

entry date

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

ASCII art descriptions

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

 ornamentation

0.47

宍

0.46

 interessi

0.46

 suunn

0.46

唦

0.45

শনের

0.44

 ateş

0.44

耑

0.44

Plaintiffs

0.44

 leuc

0.44

POSITIVE LOGITS

<0x0D>

0.65

0.65

id

0.51

0.51

ne

0.48

max

0.47

0.47

↵

0.47

ib

0.46

un

0.46

Activations Density 0.002%

No Known Activations