INDEX

Explanations

lists positive qualities or quantities

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

"-"

-1.02

 hansen

-0.94

 тщательно

-0.91

 izdel

-0.90

 nasled

-0.90

"";

-0.90

.;

-0.90

BuilderFactory

-0.89

 officinalis

-0.88

輕鬆

-0.87

POSITIVE LOGITS

 nice

1.59

 decent

1.48

 outdoor

1.48

 good

1.45

 indoor

1.44

1.40

 variety

1.26

 cheap

1.23

 vegan

1.21

 great

1.20

Activations Density 0.013%