INDEX

Explanations

numbers and percentages

systems, frameworks, improve

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

 purported

0.51

ச்

0.48

corrh

0.46

osin

0.46

ٹ

0.46

ләр

0.46

wind

0.44

 Sanford

0.44

க்கூ

0.44

ේ

0.43

POSITIVE LOGITS

 bekom

0.55

 améliorer

0.53

రిగి

0.48

 GetAll

0.48

 ihrem

0.48

 allgeme

0.48

 getTotal

0.47

第壹百

0.47

歲

0.47

歡

0.46

Activations Density 0.000%