INDEX

Explanations

Code, math, and data

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

enko

-0.28

bled

-0.28

ä¹Łè®©

-0.27

stances

-0.26

æ²¡

-0.26

Ã©rio

-0.25

ä¸ĢéĿ¢

-0.25

 bloodstream

-0.25

æİĴ

-0.25

-blood

-0.24

POSITIVE LOGITS

 result

0.41

ç»ĵæŀľ

0.39

result

0.36

 resultat

0.34

çļĦç»ĵæŀľ

0.33

Result

0.33

 rÃ©sultat

0.33

 Result

0.32

å¾Ĺåĩº

0.31

.result

0.31

Activations Density 0.017%

Code, math, and data

No Comments

No Known Activations