INDEX

Explanations

make

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

æİ°

-0.28

 factor

-0.27

 dial

-0.27

ëĭ¹

-0.26

ä¸Ģç¬ĳ

-0.25

 adherence

-0.24

 Dial

-0.24

factor

-0.24

mos

-0.24

æ´¼

-0.24

POSITIVE LOGITS

 senses

0.29

 sense

0.28

hol

0.28

Hol

0.28

 difference

0.27

_CLK

0.27

 Difference

0.27

.showMessage

0.25

difference

0.25

 Holl

0.25

Activations Density 0.037%

make

No Comments

No Known Activations