INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_23.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.23.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 crowned

-0.29

unf

-0.27

Łèĥ½

-0.26

zM

-0.26

 groom

-0.26

åĲĦä¸ªæĸ¹éĿ¢

-0.25

ajan

-0.24

çļĦåĦ¿åŃĲ

-0.23

Guy

-0.23

ricia

-0.23

POSITIVE LOGITS

 Realty

0.27

åĮº

0.26

nels

0.25

Tor

0.25

åĮºåĨħ

0.24

 Ã¶nÃ¼

0.23

there

0.23

posites

0.23

Ð¸Ñħ

0.23

["$

0.23

Activations Density 0.002%

No Known Activations

This feature has no known activations.