© 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 · 17724 ｜ Neuronpedia

Home
Qwen3-32B
32-RESID-BATCHTOPK-65K
17724

INDEX

Explanations

shapes and forms

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

æ¶ī

-0.09

èĪħ

-0.09

urrets

-0.09

åĲ±

-0.09

 Gaut

-0.09

 Curtain

-0.09

 mocked

-0.08

è¡

-0.08

æŀģé«ĺ

-0.08

è¦Ĩ

-0.08

POSITIVE LOGITS

å½¢çĬ¶

0.26

 shape

0.25

shape

0.22

 shapes

0.19

-shaped

0.19

(shape

0.19

.shape

0.18

 Shape

0.18

 shaped

0.18

Shape

0.18

Activations Density 0.169%

No Known Activations