INDEX

Explanations

The

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

 Challenge

-0.28

entifier

-0.28

Challenge

-0.26

thon

-0.26

 fanatic

-0.25

åıĮæīĭ

-0.25

-ton

-0.24

çĶ·å¥³

-0.24

cket

-0.24

bis

-0.23

POSITIVE LOGITS

oref

0.27

itated

0.26

lineno

0.26

åĪłéĻ¤æĪĲåĬŁ

0.26

halb

0.26

orris

0.25

 showMessage

0.25

 Nagar

0.25

breadcrumb

0.25

reece

0.25

Activations Density 0.008%

The

No Comments

No Known Activations