INDEX

Explanations

statistical tests and models

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

how

-1.58

↵↵

-1.56

 včetně

-1.52

 広島

-1.45

 What

-1.27

 магазина

-1.26

 Bagaimana

-1.25

蛲

-1.24

 Nodo

-1.24

 dobry

-1.24

POSITIVE LOGITS

呖

1.59

比率

1.46

>):

1.38

ಟ್

1.38

焌

1.36

 novembre

1.35

 ihop

1.33

鋱

1.26

嫆

1.26

 princesse

1.25

Activations Density 0.041%