INDEX

Explanations

technical instructions

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

える

0.64

elv

0.60

 Doming

0.60

Doming

0.59

 Scienze

0.57

桦

0.55

ரியில்

0.55

 Dominican

0.54

 expanding

0.54

えた

0.53

POSITIVE LOGITS

 stejně

0.68

 běž

0.64

疣

0.60

ра

0.59

Очень

0.58

|^{-

0.57

 grossly

0.55

JUD

0.55

 Invoice

0.55

 menikah

0.55

Activations Density 0.000%