INDEX

Explanations

portions

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 portions

-1.81

 portion

-1.64

portion

-1.41

 Portion

-1.37

 Portions

-1.32

Portion

-1.09

 fragment

-1.00

 fragments

-0.94

 chunks

-0.94

 parts

-0.93

POSITIVE LOGITS

:✨

0.57

 courbe

0.50

 Corintios

0.50

IVEREF

0.50

 plastiques

0.49

 varandra

0.48

 stället

0.48

 valmis

0.47

ButterKnife

0.46

 diametro

0.45

Activations Density 0.620%