INDEX

Explanations

non- prefix indicating negation

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Jeder

1.63

 każdy

1.60

 каждая

1.58

 Каждый

1.57

 каждый

1.49

 દરેક

1.46

 Mỗi

1.46

 każde

1.43

 chacune

1.39

 каждой

1.35

POSITIVE LOGITS

hello

0.85

al

0.85

allo

0.82

ali

0.81

ल्ली

0.77

ail

0.76

ali

0.75

ಳ್ಳ

0.75

al

0.75

all

0.74

Activations Density 0.038%