INDEX

Explanations

most important

describing subsequent states

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

ت

0.46

まずは

0.44

0.42

 Infiniti

0.41

 hangout

0.38

 Justine

0.38

 éste

0.38

 unapolog

0.38

 ആദ്യം

0.38

เรา

0.37

POSITIVE LOGITS

otri

0.45

ವಾಡ

0.45

zechos

0.44

ับสน

0.44

 chessboard

0.43

chol

0.42

ESTIC

0.41

ANAL

0.40

그러나

0.40

EST

0.40

Activations Density 0.000%