© Neuronpedia 2026

Privacy & Terms Blog GitHub Slack Twitter Contact

Jacobian LensNEW

Natural Language

MAY 26 Assistant AxisJAN 26 Circuit TracerAPR 26Steer SAE Evals Exports interp-engineNEW Community Blog Privacy & Terms Contact

QWEN3-32B · 32-RESID-BATCHTOPK-65K · 16799 ｜ Neuronpedia

Home
Qwen3-32B
32-RESID-BATCHTOPK-65K
16799

INDEX

Explanations

piece of

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

Negative Logits

quo

-0.10

rence

-0.10

å¾Ĺä½ı

-0.09

 Bean

-0.09

çļĦä¸Ģç§į

-0.09

åģľè½¦

-0.09

vens

-0.09

asi

-0.09

é¢ĸ

-0.09

ship

-0.09

POSITIVE LOGITS

 Piece

0.11

.piece

0.11

piece

0.11

(piece

0.11

pieces

0.11

 Pieces

0.10

-piece

0.10

 pieces

0.10

 piece

0.10

ieces

0.10

Activations Density 0.018%

No Known Activations