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

Not in Any Lists

No Comments

Negative Logits

 benches

-0.28

sonian

-0.27

issing

-0.25

 elbows

-0.24

 elbow

-0.24

æĵį

-0.24

assy

-0.24

uben

-0.24

.community

-0.24

yii

-0.23

POSITIVE LOGITS

osed

0.28

OS

0.28

DL

0.27

Ä±

0.26

æĹłç¼Ŀ

0.25

æŃ¤ä¸¾

0.24

éĹŃçİ¯

0.24

ŀæĢ§

0.24

 Airways

0.24

sek

0.23

Activations Density 1.416%

No Known Activations

This feature has no known activations.