INDEX

Explanations

set aside

instructions before or during actions

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

coprotein

0.47

Ward

0.47

as

0.47

Policy

0.46

 effortlessly

0.46

ObjectModel

0.46

place

0.46

doctor

0.46

ap

0.45

oloji

0.45

POSITIVE LOGITS

 bieg

0.50

แน่น

0.47

ブロ

0.47

パワ

0.46

 risultato

0.46

โมง

0.45

 Ressourcen

0.44

億

0.43

퐈

0.43

ǫ

0.43

Activations Density 0.002%