INDEX

Explanations

Programming concepts

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

Not in Any Lists

No Comments

Negative Logits

ellung

-0.27

æĿ¡è¯Ħè®º

-0.27

chnitt

-0.27

obia

-0.26

ellant

-0.25

elig

-0.25

ä¸ªå°ı

-0.25

obic

-0.24

conom

-0.24

BX

-0.23

POSITIVE LOGITS

 Associated

0.31

 associated

0.30

associated

0.29

Associated

0.28

æĿ¥

0.28

 municipality

0.27

.operations

0.24

ron

0.24

.ribbon

0.24

les

0.24

Activations Density 0.074%

Programming concepts

No Comments

No Known Activations