INDEX

Explanations

child, campaign, magicians, replaces

offensive and objectionable terms

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

recogn

0.44

join

0.44

kait

0.43

hin

0.43

kak

0.42

快递

0.41

unpriv

0.41

uq

0.40

giv

0.39

utip

0.39

POSITIVE LOGITS

 encroachment

0.48

 '/')

0.44

 unbelievable

0.41

 objectionable

0.41

 infraction

0.41

 uproar

0.41

 stycz

0.40

 fish

0.40

 alteration

0.40

 infringement

0.39

Activations Density 0.001%