INDEX

Explanations

XeTeX, backend, CJK languages

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

].

-1.53

.”

-1.46

,’

-1.40

.[

-1.38

 jako

-1.34

.\\

-1.31

];

-1.29

 uzun

-1.29

 cidade

-1.26

 pobre

-1.24

POSITIVE LOGITS

 않았

1.38

TODOS

1.28

 навіть

1.27

 asti

1.26

 bruno

1.25

чие

1.21

 saan

1.20

動機

1.20

 gonz

1.20

Commande

1.19

Activations Density 0.040%