INDEX

Explanations

peeling, inverting, ziehen

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

leq

1.59

صفر

1.51

loops

1.41

ellipse

1.36

allis

1.31

rmse

1.30

azem

1.29

olium

1.25

WL

1.25

elsior

1.21

POSITIVE LOGITS

 kalangan

1.57

 கவர்

1.56

 revelar

1.51

俬

1.49

 akses

1.48

 démont

1.47

há

1.45

 తీవ్ర

1.43

 aprove

1.43

 dévo

1.42

Activations Density 0.186%