INDEX

Explanations

punctuation

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_20.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.20.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

å¦Ĥå®ŀ

-0.30

-free

-0.27

Ã¶ren

-0.26

ÃŃc

-0.25

ä»¬éĥ½

-0.25

Ð·Ñĭ

-0.24

åĬ¨åĳĺ

-0.24

çī¢çī¢

-0.24

éĢĲå¹´

-0.24

çĦ¡æĸĻ

-0.24

POSITIVE LOGITS

 Pact

0.28

olem

0.27

èĲ½

0.27

pane

0.25

 Conway

0.25

ille

0.25

å¼¹

0.25

ç«ĭé¡¹

0.24

 tact

0.24

gu

0.24

Activations Density 7.008%

punctuation

No Known Activations