INDEX

Explanations

advocacy and support services

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

os

-1.54

 that

-1.52

不夠

-1.46

 verksamhet

-1.44

 anunció

-1.38

-1.34

 bienvenue

-1.32

嫵

-1.29

=$("#

-1.29

 آدم

-1.28

POSITIVE LOGITS

 they

1.90

 with

1.65

and

1.58

 like

1.51

of

1.46

 они

1.45

 آنها

1.43

Sabes

1.38

早晨

1.36

ควร

1.32

Activations Density 0.078%