INDEX

Explanations

reporting speech after titles or pronouns

reminders and explanations

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

强大

0.43

voh

0.40

ിയത്

0.40

 spä

0.39

ர்வ

0.38

 verändert

0.38

强大的

0.37

 ක්ර

0.37

 overdraft

0.37

pfl

0.37

POSITIVE LOGITS

 подчер

1.67

 sottoline

1.48

強調

1.47

 podkre

1.45

 강조

1.42

强调

1.34

 betont

1.34

 emphasised

1.29

 enfat

1.27

 vurg

1.27

Activations Density 0.003%