INDEX

Explanations

emphasizing adjectives

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

of

-1.62

 emozioni

-1.52

 автомо

-1.45

 notizie

-1.44

is

-1.43

 définiti

-1.41

 opzioni

-1.36

-1.34

реги

-1.34

опера

-1.33

POSITIVE LOGITS

 нашим

1.75

 both

1.70

↵

1.60

šanu

1.59

 нашем

1.55

صادر

1.52

ceptar

1.51

++]=

1.49

考えた

1.48

ulink

1.48

Activations Density 0.040%