INDEX

Explanations

ro

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_31.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.31.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

DialogContent

-0.28

ignKey

-0.27

 shaved

-0.25

ÃŃculos

-0.25

AAD

-0.23

ä¸Ģè·¯ä¸Ĭ

-0.23

çĶŁæ¶¯

-0.23

æĮĤéĴ©

-0.23

apses

-0.23

å¤§åİħ

-0.23

POSITIVE LOGITS

OTT

0.32

maker

0.30

ellt

0.28

ott

0.28

ubic

0.27

bern

0.27

ç«ĭé©¬

0.27

ual

0.27

uers

0.27

licable

0.26

Activations Density 0.006%

ro

No Known Activations