INDEX

Explanations

abbreviations for

fueled economic

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

 വര്‍

0.47

 udě

0.44

technik

0.42

 ампли

0.42

尻

0.42

<unused3>

0.41

合金

0.41

 ഞാൻ

0.40

qid

0.40

愊

0.40

POSITIVE LOGITS

 webinars

0.48

 office

0.46

 soap

0.45

 convenient

0.44

 कैसे

0.44

 transportation

0.44

 soaps

0.43

 limbo

0.43

 túl

0.43

 company

0.43

Activations Density 0.009%