INDEX

Explanations

not alone

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

NOUNC

2.21

pleted

1.85

 finalised

1.82

 inéd

1.79

 완료

1.74

 Upcoming

1.74

ร็จ

1.74

 forthwith

1.73

 terminée

1.72

 ใหม่

1.71

POSITIVE LOGITS

 disadvantages

1.43

 aversion

1.38

相反

1.32

デメリット

1.30

反対

1.30

 argument

1.30

有些人

1.29

癖

1.29

 saying

1.27

 argue

1.26

Activations Density 0.150%