INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_23.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.23.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

æİĴæ¯Ĵ

-0.29

 Convenience

-0.28

etrize

-0.26

ewidth

-0.26

 convenience

-0.26

ä¸ŃæĸŃ

-0.26

ä¾¿åĪ©

-0.26

 Cata

-0.25

å¹³æĹ¶

-0.25

Ord

-0.25

POSITIVE LOGITS

chen

0.27

åī¯æĢ»

0.26

afc

0.25

Ð°Ð»ÑĮÐ½ÑĭÐ¹

0.25

 eens

0.24

ningar

0.24

é®

0.24

 democrat

0.23

DRV

0.23

åĻªéŁ³

0.23

Activations Density 0.018%

No Known Activations

This feature has no known activations.