INDEX

Explanations

moderator, moderation, operator

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

to

-1.53

 Then

-1.27

 kemudian

-1.25

 Ayrıca

-1.23

↵↵

-1.23

my

-1.20

 various

-1.20

 that

-1.16

 ftate

-1.15

 財布

-1.11

POSITIVE LOGITS

 тук

1.55

cepted

1.41

この

1.38

 ومل

1.34

ここは

1.31



1.31

vertebrates

1.30

vaded

1.29

operability

1.28

 samme

1.27

Activations Density 0.034%