INDEX

Explanations

Code, legal, or prompts

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

è¨İ

-0.28

relude

-0.28

ä¿ĿèŃī

-0.28

çļĦä¸Ģæ¬¾

-0.26

äººåĿĩ

-0.26

èĢ³

-0.25

 Pathfinder

-0.25

.newaxis

-0.25

åįķåħĥ

-0.25

 MediaTek

-0.24

POSITIVE LOGITS

å¸¸æĢģ

0.28

 classified

0.27

ess

0.26

idden

0.26

ful

0.25

edImage

0.25

over

0.24

seen

0.24

esser

0.24

 endwhile

0.24

Activations Density 0.010%

Code, legal, or prompts

No Comments

No Known Activations