INDEX

Explanations

parenthetical numbers, lists, and quotes

avoid excessive, regulated after

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

 schreiben

0.55

 λέ

0.54

 geschrieben

0.49

ប្រស

0.48

 arrive

0.47

逻辑

0.46

 পাঁচটি

0.46

[-

0.46

 escrito

0.45

 kittens

0.45

POSITIVE LOGITS

Kabhi

0.52

USP

0.49

Wax

0.49

ید

0.46

НІ

0.46

Favor

0.46

ﻨ

0.46

ธุรก

0.45

MacOS

0.45

了他

0.45

Activations Density 0.000%