INDEX

Explanations

keys followed by colon

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

↵

-2.53

of

-1.95

 will

-1.83

//{

-1.63

。”

-1.51

who

-1.47

”，

-1.46

voer

-1.42

is

-1.38

 which

-1.34

POSITIVE LOGITS

 enamorado

1.82

1.76

1.63

1.59

 herido

1.58

 modrá

1.56

他の

1.52

 россия

1.52

擷

1.52

 моск

1.51

Activations Density 0.034%