INDEX

Explanations

all

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_27.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.27.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

å¤§éĿ¢ç§¯

-0.42

 surrounded

-0.40

 surrounds

-0.38

å®½

-0.35

ç½©

-0.33

åĮħåĽ´

-0.33

åįłåľ°

-0.33

 surround

-0.33

èĥĮæĻ¯

-0.33

.bulk

-0.32

POSITIVE LOGITS

èĬ¬

0.34

oft

0.28

åŀĤ

0.27

 hinge

0.26

åĬĽ

0.26

Arg

0.25

Ã¶n

0.24

ÐºÐ»Ð°

0.24

Rig

0.24

 down

0.24

Activations Density 0.002%

all

No Known Activations