INDEX

Explanations

's

np_max-act · gemini-2.0-flash

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

Negative Logits

ively

-0.28

unde

-0.28

uchs

-0.27

åľ¨å¸Ĥåľºä¸Ĭ

-0.27

 warranted

-0.27

Beg

-0.25

eli

-0.24

æľīæĿĥ

-0.24

?key

-0.24

uye

-0.24

POSITIVE LOGITS

çļĦæĦŁåıĹ

0.32

ä¾ĭåŃĲ

0.30

 example

0.30

çļĦä¾ĭåŃĲ

0.28

arning

0.27

immel

0.27

è¯¾é¢ĺ

0.27

 dÃ©cidÃ©

0.26

æ³ķåĪĻ

0.26

changing

0.26

Activations Density 0.003%

's

No Known Activations