© 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 · 55075 ｜ Neuronpedia

Home
Qwen3-32B
32-RESID-BATCHTOPK-65K
55075

INDEX

Explanations

Charles followed by a name

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

ally

-0.13

../../../

-0.12

rema

-0.10

ALLY

-0.10

Ø¹Ø¯

-0.10

atel

-0.10

antha

-0.09

eward

-0.09

uron

-0.09

ÑĢÑı

-0.09

POSITIVE LOGITS

çĪ·

0.12

acters

0.12

worth

0.12

bourg

0.11

ses

0.11

ëŁī

0.10

itics

0.10

lest

0.10

isle

0.10

çĪº

0.10

Activations Density 0.044%

No Known Activations