INDEX

Explanations

the verb "start"

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

CopyWith

-0.83

 endregion

-0.76

ThemeOverlay

-0.76

 PopupWindow

-0.66

StructEnd

-0.65

 שוליים

-0.65

PostExecute

-0.63

IconModule

-0.63

enumi

-0.63

Errorf

-0.62

POSITIVE LOGITS

 start

2.78

 starts

2.56

 Start

2.52

 starting

2.50

 started

2.44

Start

2.39

start

2.36

 Starting

2.33

 Starts

2.33

starts

2.31

Activations Density 0.497%