INDEX

Explanations

bring positive outcomes

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

有一个

-1.66

us

-1.65

 دارید

-1.60

 tertentu

-1.60

 Mover

-1.59

有个

-1.51

You

-1.47

 υπάρχ

-1.46

ed

-1.45

ですので

-1.44

POSITIVE LOGITS

 ninguém

1.67

 amanhã

1.55

1.51

 estrateg

1.50

 aguard

1.47

 abenço

1.45

 vítima

1.44

);

1.43

 mergul

1.42

鴞

1.37

Activations Density 0.017%