© 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-2-2B · 0-CLT-HP · 41863 ｜ Neuronpedia

Home
Gemma-2-2B
0-CLT-HP
41863

INDEX

Explanations

Qualifying statements

np_max-act · gemini-2.0-flash

New Auto-Interp

Top Features by Cosine Similarity

Configuration

Prompts (Dashboard)

16,384 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

 purpoſe

-1.28

 myſelf

-1.23

Efq

-1.22

 pleaſure

-1.20

 Diſ

-1.13

 itſelf

-1.13

 iſt

-1.13

ſelf

-1.13

 Majefty

-1.12

 ſeveral

-1.09

POSITIVE LOGITS

be

0.81

0.75

 being

0.71

an

0.70

the

0.66

as

0.59

get

0.59

re

0.58

0.56

ter

0.50

Activations Density 0.118%

No Known Activations