© 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 · 3330 ｜ Neuronpedia

Home
Qwen3-32B
32-RESID-BATCHTOPK-65K
3330

INDEX

Explanations

real-time

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

å¦Ļ

-0.09

åľ¨è¿ĩæ¸¡

-0.08

hurst

-0.08

åĨ¤

-0.08

nutrition

-0.08

ä¸įéĢļ

-0.08

venture

-0.08

emm

-0.08

 Bradley

-0.08

dio

-0.08

POSITIVE LOGITS

å®ŀæĹ¶

0.23

åį³æĹ¶

0.17

 live

0.14

Live

0.14

 LIVE

0.13

live

0.13

/live

0.12

-live

0.12

ãĥªãĤ¢ãĥ«

0.12

 realtime

0.11

Activations Density 0.091%

No Known Activations