INDEX

Explanations

own

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

åĲĮçŃī

-0.31

 Classified

-0.28

æľĢçĪ±

-0.27

 themselves

-0.27

å°¤åħ¶æĺ¯åľ¨

-0.25

 herself

-0.25

girls

-0.25

ä¼ļåıĳçĶŁ

-0.25

åħĶ

-0.25

ä¼ļæľī

-0.25

POSITIVE LOGITS

 unique

0.29

.stdin

0.28

isti

0.27

 determinant

0.27

ifix

0.26

atoria

0.26

 kind

0.26

çĭ¬çī¹çļĦ

0.25

 replic

0.25

ä½ĵåĪ¶æľºåĪ¶

0.25

Activations Density 0.007%

own

No Comments

No Known Activations