INDEX

Explanations

code syntax, variables, and keywords

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 cajones

-1.53

of

-1.47

 AMAZ

-1.46

lijken

-1.38

chè

-1.36

≪

-1.35

all

-1.34

Каждый

-1.31

 impecable

-1.30

Гри

-1.30

POSITIVE LOGITS

for

2.14

or

1.66

and

1.59

謳

1.57

 numerosas

1.55

 rozwój

1.55

you

1.54

姿を

1.47

følge

1.46

 numerosos

1.45

Activations Density 0.007%