© 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

QWEN3-32B · 32-RESID-BATCHTOPK-65K · 17301 ｜ Neuronpedia

Home
Qwen3-32B
32-RESID-BATCHTOPK-65K
17301

INDEX

Explanations

image links

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

New Auto-Interp

Top Features by Cosine Similarity

Configuration

adamkarvonen/qwen3-32b-saes/saes_Qwen_Qwen3-32B_batch_top_k/resid_post_layer_32

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

ä¸įä¿¡

-0.10

æµģæĺŁ

-0.09

rib

-0.09

åĢľ

-0.09

PCS

-0.08

ä¸ĩæĪ·

-0.08

Intl

-0.08

.locale

-0.08

apy

-0.08

amer

-0.08

POSITIVE LOGITS

åĽ¾çīĩ

0.14

 images

0.12

 image

0.12

åĽ¾åĥı

0.12

.images

0.11

.getImage

0.10

æīĺç®¡

0.10

.setImage

0.10

 Image

0.10

 Images

0.10

Activations Density 0.362%

No Known Activations