INDEX

Explanations

people's last names and their relations

names followed by "who"

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

 ඒවා

0.30

 povas

0.30

とか

0.29

 যাইহোক

0.28

 അതിന്റെ

0.28

 jeweiligen

0.27

hankelijk

0.27

 voltaje

0.26

 slike

0.26

}`}>

0.26

POSITIVE LOGITS

 whom

0.92

who

0.86

who

0.84

whom

0.83

 quien

0.81

 যিনি

0.71

 quién

0.61

 whose

0.57

 Whom

0.56

谁

0.54

Activations Density 0.302%