INDEX

Explanations

company and product names

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

账

0.74

 delinqu

0.69

 confession

0.66

 unbear

0.65

 stereotyp

0.64

 Pompeo

0.63

 Complimentary

0.63

 distracted

0.63

 Mythology

0.63

 subreddit

0.62

POSITIVE LOGITS

Flex

1.46

Tech

1.46

Tec

1.44

Rite

1.39

Care

1.38

tec

1.37

Safe

1.36

TECH

1.36

flex

1.35

MAX

1.31

Activations Density 1.279%