INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_30.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.30.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Exped

-0.28

oyer

-0.27

æŀ¢

-0.26

OnInit

-0.26

recommended

-0.25

å¤ĩ

-0.25

çģ¼

-0.25

 MEMBER

-0.24

oÄŁ

-0.24

Buzz

-0.24

POSITIVE LOGITS

iska

0.29

usk

0.28

iscrim

0.27

rylic

0.26

isk

0.26

leftrightarrow

0.26

æµģæ°´

0.25

 Ridley

0.25

isÃ©

0.25

éĵł

0.24

Activations Density 0.002%

No Known Activations

This feature has no known activations.