INDEX

Explanations

identifying common phrases

technical terms and concepts

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

 bombers

0.55

 Galloway

0.49

ві

0.48

 Crowley

0.48

ನ

0.48

子

0.47

burner

0.47

 whistleblower

0.46

на

0.46

POSITIVE LOGITS

 regiones

0.49

 아래

0.45

 طلب

0.43

ucie

0.43

 منطقه

0.43

 رشته

0.43

 создания

0.42

 ORIG

0.42

 نیچے

0.42

atthanam

0.42

Activations Density 0.000%