INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_1.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.1.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

à¤¿

-0.28

evice

-0.25

today

-0.25

åľ¨ä¸Ģèµ·

-0.25

tin

-0.25

èµŀåĬ©

-0.24

åľ¨å®¶

-0.24

tures

-0.24

iets

-0.23

'),'

-0.23

POSITIVE LOGITS

ands

0.25

butt

0.25

contre

0.25

ometric

0.24

æĸŃ

0.24

 Verde

0.24

ä¾Ŀæ³ķ

0.24

éĦ¢

0.24

eway

0.24

.realm

0.23

Activations Density 28.898%

No Known Activations

This feature has no known activations.