INDEX

Explanations

Lesser

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

erator

-0.30

aket

-0.28

çĽ´éĢļè½¦

-0.28

 till

-0.27

æ£®

-0.26

åĴĮåıĳå±ķ

-0.26

arus

-0.26

ÃŃc

-0.25

æľĢå°ĳ

-0.24

ayo

-0.24

POSITIVE LOGITS

 ephem

0.28

azzi

0.27

ä¸įä¸Ĭ

0.27

å¿ħè¦ģ

0.26

ecom

0.25

 required

0.25

ection

0.24

 stricter

0.23

 REQUIRED

0.23

åį±å®³

0.23

Activations Density 0.007%

Lesser

No Known Activations