INDEX

Explanations

technical or code writing

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

ourd

-0.27

ä¼ļå±ķ

-0.27

auce

-0.26

altung

-0.25

ipping

-0.25

annel

-0.25

é¢ĦåĪ¶

-0.24

åħĳ

-0.24

éĻįåĪ°

-0.24

ipped

-0.24

POSITIVE LOGITS

 which

0.38

which

0.36

ä½Ĩè¿Ļ

0.29

Which

0.29

 Which

0.28

Still

0.27

 bulb

0.25

kn

0.25

 bully

0.25

 STILL

0.25

Activations Density 0.000%

technical or code writing

No Known Activations