INDEX

Explanations

architects discussing things

specific entities after certain words

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

 impulsive

0.52

 erros

0.48

 drunkenness

0.48

 wasting

0.47

 orgasm

0.47

 menstrual

0.46

 demonstrating

0.46

 intracranial

0.46

डमी

0.45

 inconsistencies

0.45

POSITIVE LOGITS

XIX

0.50

Р

0.45

 SCALE

0.45

 MATHEMAT

0.43

ජ

0.42

 விஜய்

0.42

यची

0.41

 आपूर्ति

0.41

BOY

0.41

Activations Density 0.000%