INDEX

Explanations

punctuation

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

çĶŁäº§åĴĮ

-0.28

æĸĩåĮĸåĴĮ

-0.28

bose

-0.28

å»ºè®¾åĴĮ

-0.27

ç©Ĩ

-0.27

NSS

-0.26

eus

-0.25

atile

-0.25

NST

-0.25

tin

-0.25

POSITIVE LOGITS

èĢĮ

0.33

While

0.32

 Whereas

0.31

 whereas

0.30

 meanwhile

0.28

 While

0.28

 while

0.27

 similarly

0.27

 Likewise

0.26

çĽ¸æ¯Ķä¹ĭä¸ĭ

0.26

Activations Density 0.003%

punctuation

No Comments

No Known Activations