INDEX

Explanations

INESS

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

æĹ©æĹ©

-0.28

ilar

-0.28

stantiate

-0.27

åĬªåĬĽ

-0.25

ce

-0.24

åľ¨æİ¥åıĹ

-0.24

æľ¬çĿĢ

-0.23

 optimizations

-0.23

NET

-0.23

Ph

-0.23

POSITIVE LOGITS

Å¥

0.27

tres

0.27

iners

0.26

/Private

0.25

ARGIN

0.25

<![

0.24

ustum

0.24

åĪĨæŀĲå¸Ī

0.24

 Hollow

0.24

pcf

0.23

Activations Density 0.000%

INESS

No Comments

No Known Activations