INDEX

Explanations

names and foreign words

list phrases and code snippets

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

ُل

0.50

 conversa

0.47

ليا

0.47

 suddiv

0.46

 discurso

0.46

LongNumber

0.46

нии

0.46

ُر

0.46

 conflicto

0.45

ز

0.45

POSITIVE LOGITS

 Jing

0.51

 あなた

0.48

bok

0.45

’

0.45

 Tian

0.45

Mrs

0.44

mam

0.44

 ওয়া

0.44

asi

0.44

estry

0.43

Activations Density 0.003%