INDEX

Explanations

charm and appeal

overall appeal and charm

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

 चिंतित

0.39

 सावधान

0.38

 συνεργ

0.38

,}$

0.38

ार्किक

0.37

 ಸಹಾಯ

0.36

 skut

0.36

衍

0.36

=-\

0.35

 सुनी

0.35

POSITIVE LOGITS

 charm

2.58

 charme

2.16

魅力

2.11

 appeal

2.09

charm

2.09

 charms

2.03

 allure

1.98

 Charm

1.93

appeal

1.86

魅

1.84

Activations Density 0.325%