INDEX

Explanations

prevent further improve performance avoid escalating posting material

psychological abuse or subtle distinctions

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

クーポン

0.50

 Golgi

0.50

 sacchar

0.49

 শরণার্থী

0.47

ત્રો

0.46

 સારી

0.45

 একইভাবে

0.45

Kana

0.45

 ಸಾಕ

0.45

্বেও

0.45

POSITIVE LOGITS

But

0.48

\,\

0.47

 второй

0.46

мыс

0.43

 Thanks

0.43

 Styl

0.43

0.42

 secund

0.42

 Rice

0.42

ด

0.41

Activations Density 0.000%