INDEX

Explanations

can

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_9.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.9.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

åĨħåĪĨæ³Į

-0.28

 remains

-0.27

whole

-0.26

 remainder

-0.26

 empathy

-0.24

åĲĪéĩĳ

-0.24

 whole

-0.24

 endIndex

-0.24

éĿ¢ä¸´çĿĢ

-0.24

 lá»Ŀi

-0.24

POSITIVE LOGITS

ä»ĸä»¬æĺ¯

0.29

ä»ĸä»¬çļĦ

0.27

lets

0.27

æı©

0.26

ä»Ģä¹Īåı«

0.26

 deposited

0.25

ubi

0.25

 MISS

0.25

Hed

0.25

pol

0.25

Activations Density 5.337%

can

No Comments

No Known Activations

can

No Comments

No Known Activations