INDEX

Explanations

the followed by titles or descriptors

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

 each

-1.65

になれば

-1.56

locene

-1.42

brainer

-1.42

 besseren

-1.41

 určit

-1.33

Pemer

-1.32

СТВА

-1.31

 normalen

-1.30

Ketika

-1.29

POSITIVE LOGITS

two

1.44

 désormais

1.43

one

1.40

('');

1.39

 теперь

1.34

 stund

1.30

ająca

1.30

misi

1.26

Ẓ

1.24

{};

1.22

Activations Density 0.104%