INDEX

Explanations

instances of the word "par" in its context

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 paralle

-0.76

 parallèle

-0.72

 Park

-0.69

 Parallel

-0.69

 parall

-0.69

 Paras

-0.68

Paragraph

-0.68

 parallel

-0.68

paralle

-0.68

 Paragraph

-0.65

POSITIVE LOGITS

par

2.42

par

0.82

 estekak

0.81

Datuak

0.69

BeginContext

0.66

Tikang

0.66

مصادر

0.66

SourceChecksum

0.65

تقاوى

0.64

 تضيفلها

0.63

Activations Density 0.007%

No Known Activations