INDEX

Explanations

Names

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 whereas

-0.55

-0.52

 justamente

-0.49

 then

-0.49

 instead

-0.43

 precisely

-0.43

 analogous

-0.43

 according

-0.43

 while

-0.42

 during

-0.41

POSITIVE LOGITS

AndEndTag

0.98

 EconPapers

0.89

ArrowToggle

0.81

 Италијани

0.81

ьаж

0.81

 kasarigan

0.81

 propOrder

0.80

StoryboardSegue

0.79

Hochspringen

0.78

Tikang

0.75

Activations Density 0.001%