INDEX

Explanations

the

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_22.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.22.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

iben

-0.25

åģ¥åº·æĪĲéķ¿

-0.25

vig

-0.24

atory

-0.24

 unaffected

-0.24

ingly

-0.24

 pepp

-0.24

vig

-0.23

æĹłè®ºå¦Ĥä½ķ

-0.23

è¿Ļä¸ĢçĤ¹

-0.23

POSITIVE LOGITS

åĲĦè¡Į

0.28

åĵŃäºĨ

0.25

velt

0.25

æįŁèĢĹ

0.24

æĢ»éĿ¢ç§¯

0.24

 efficiencies

0.24

èıĮ

0.24

ÅĤÄĻ

0.24

éĽĨåĽ¢èĤ¡ä»½

0.24

æĪĳå®¶

0.24

Activations Density 0.006%

the

No Known Activations