INDEX

Explanations

of

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

æĪĲä¸ºä¸ŃåĽ½

-0.32

eware

-0.27

åħ»çĮª

-0.26

*)((

-0.26

-alist

-0.26

unar

-0.25

å¯¹å¥¹

-0.25

è¿Ļä¸ªä¸ĸçķĮ

-0.25

 EITHER

-0.25

oard

-0.25

POSITIVE LOGITS

 through

0.29

åĩł

0.27

an

0.27

cket

0.27

 weak

0.26

 solutions

0.25

åĩłä¸ª

0.25

and

0.25

 Penn

0.25

Activations Density 0.004%

of

No Comments

No Known Activations