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

åºķä¸ĭ

-0.28

oise

-0.27

ets

-0.27

IMUM

-0.26

erves

-0.25

ç®ĹäºĨ

-0.25

è´«

-0.25

olist

-0.24

ä¸ĭæĸ¹

-0.24

ETS

-0.23

POSITIVE LOGITS

ÄĻd

0.27

åµĮ

0.27

piv

0.25

.getLine

0.24

 cabo

0.23

 christ

0.23

 gauss

0.23

åĿļæĮģä»¥

0.23

å¥¥

0.23

utz

0.23

Activations Density 0.010%

No Known Activations

This feature has no known activations.