INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_20.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.20.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

asto

-0.28

Ð»Ð¸Ð¼

-0.27

è¿ĻäºĽéĹ®é¢ĺ

-0.26

enderror

-0.26

Ã³t

-0.26

ason

-0.26

aston

-0.26

æľ«

-0.25

åºļ

-0.25

è®©åŃ©åŃĲ

-0.25

POSITIVE LOGITS

bote

0.29

 Lack

0.25

è¿Ļæĺ¯ä¸Ģä¸ª

0.25

 issu

0.24

Reply

0.24

å¥Ī

0.24

 beneficiaries

0.23

BarController

0.23

 bÃ©nÃ©

0.23

ipel

0.22

Activations Density 90.033%

No Known Activations

This feature has no known activations.