© 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

QWEN2.5-7B-IT · 20-MATRYOSHKA-65K · 533 ｜ Neuronpedia

Home
Qwen2.5-7B-IT
20-MATRYOSHKA-65K
533

INDEX

Explanations

research papers

np_max-act-logits · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Configuration

chanind/qwen2.5-7B-it-layer-20-saes/pile/matryoshka/k-100

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

“

-0.31

-0.29

-0.28

↵

-0.28

Âł

-0.27

 above

-0.26

you

-0.26

-0.25

-0.25

↵↵

-0.24

POSITIVE LOGITS

 addCriterion

0.34

âļĹ

0.34

,eg

0.34

/goto

0.33

å°±æĿ¥çľĭçľĭ

0.33

åĲįåĪĹåīį

0.32

/Gate

0.32

 ...↵↵↵↵

0.31

Ð±Ð»Ð°Ð³Ð¾

0.31

 misunder

0.31

Activations Density 17.596%

No Known Activations