INDEX

Explanations

that's followed by explanation

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

 IBRARY

-1.36

to

-1.29

 perche

-1.21

凤凰

-1.20

ка

-1.20

!（

-1.18

 fatte

-1.16

 saranno

-1.15

这样的

-1.15

by

-1.15

POSITIVE LOGITS

 formada

1.55

1.50

//_

1.49

⁴

1.46

efec

1.45

ově

1.41

]);

1.39

//--

1.35

 MessageBoxIcon

1.34

 Nutzer

1.34

Activations Density 0.015%