INDEX

Explanations

URLs and special characters

code snippets and non-English characters

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

٣

0.76

 Многие

0.60

 तीन

0.57

۷

0.57

۳

0.57

 രണ്ട്

0.57

 หนึ่ง

0.57

iki

0.55

 మూడు

0.55

 Рей

0.54

POSITIVE LOGITS

ου

0.61

mathrm

0.58

δε

0.58

↵

0.57

<0xE3>

0.56

мента

0.55

و

0.53

γν

0.52

нома

0.49

λί

0.49

Activations Density 0.003%