INDEX

Explanations

abstract concepts and scenarios

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

 таких

1.63

这一

1.59

 такими

1.54

 이러한

1.51

这种

1.50

 této

1.47

 цьому

1.45

 tejto

1.44

這種

1.44

 이런

1.43

POSITIVE LOGITS

 will

1.33

has

1.30

 VERY

1.26

 very

1.23

 terribly

1.22

 hebben

1.20

 have

1.18

 belongs

1.16

 heeft

1.16

めっちゃ

1.15

Activations Density 0.799%