INDEX

Explanations

many many many

take positions, provides, allows, new, reduce

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 extenso

0.49

ículo

0.48

 impure

0.48

 incluye

0.48

 quiso

0.48

urium

0.47

 scris

0.47

 hablar

0.46

exd

0.46

 incluir

0.46

POSITIVE LOGITS

多くの

0.51

lights

0.49

 الكثير

0.47

 добы

0.46

錚

0.46

 অসংখ্য

0.45

 பலர்

0.45

many

0.44

 অনেকেই

0.43

 investors

0.43

Activations Density 0.011%