INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_18.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.18.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

åĲĮæł·

-0.27

vie

-0.27

UME

-0.26

agle

-0.26

owe

-0.25

oot

-0.25

å²Ľ

-0.24

 chÃło

-0.24

å¿ł

-0.24

oc

-0.24

POSITIVE LOGITS

andum

0.29

roken

0.28

)n

0.26

æĻ®åıĬ

0.25

åıĤå±ķ

0.25

del

0.24

:pk

0.24

å¥łåŁº

0.24

apid

0.24

(KP

0.24

Activations Density 41.290%

No Known Activations

This feature has no known activations.