© 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

QWEN3-4B · 9-TRANSCODER-HP · 29947 ｜ Neuronpedia

Home
Qwen3-4B
9-TRANSCODER-HP
29947

INDEX

Explanations

for

np_max-act · gemini-2.0-flash

New Auto-Interp

Top Features by Cosine Similarity

Configuration

mwhanna/qwen3-4b-transcoders/layer_9.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.9.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

Show PlotsShow ExplanationShow ActivationsShow Test FieldShow SteerShow Link

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

MMI

-0.27

ntax

-0.25

Reviewer

-0.24

áĥ¯

-0.24

cmds

-0.24

oller

-0.24

-dollar

-0.24

 McCorm

-0.23

-done

-0.23

éķ°

-0.23

POSITIVE LOGITS

enha

0.28

emsp

0.28

è´¨

0.26

moz

0.26

åĬ¨æľº

0.25

èĬĴ

0.25

æĢª

0.25

åĲ¯åıĳ

0.25

ald

0.25

éłĵ

0.24

Activations Density 0.685%

No Known Activations