INDEX

Explanations

entity identification

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

."

1.55

Oo

1.52

 scored

1.51

 percussion

1.46

 produced

1.46

 harp

1.46

.",

1.45

 haired

1.45

的

1.44

 infringe

1.44

POSITIVE LOGITS

ходит

1.50

ików

1.35

 объем

1.34

ходят

1.30

ف

1.30

važ

1.30

zieć

1.27

стоит

1.25

того

1.21

ধিক

1.21

Activations Density 0.003%