INDEX

Explanations

No Explanations Found

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

undy

-0.28

åŁµ

-0.25

 manufacturing

-0.25

LEE

-0.24

/xml

-0.24

.ViewModels

-0.24

 suits

-0.23

æŃ§

-0.23

ç®¨

-0.23

POSITIVE LOGITS

uÃ©

0.34

 tent

0.28

kah

0.27

alus

0.26

cala

0.26

aten

0.26

ablish

0.26

 USAGE

0.25

uentes

0.25

itaire

0.24

Activations Density 0.882%

No Known Activations

This feature has no known activations.