INDEX

Explanations

Arguments

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

 Holy

-0.56

 flame

-0.50

 true

-0.50

 Truth

-0.49

 sete

-0.47

 True

-0.46

 False

-0.45

 mean

-0.45

red

-0.44

iomanip

-0.42

POSITIVE LOGITS

Personendaten

0.84

Hochspringen

0.76

tanleria

0.72

 aikaa

0.70

 HttpServlet

0.69

saraba

0.68

存于互联网档案馆

0.68

 Commando

0.64

noons

0.63

VersionUID

0.63

Activations Density 0.234%