INDEX

Explanations

reprogramming requests

multilingual text identification

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 Molecular

0.47

 guided

0.46

 Riddle

0.46

 molecular

0.45

 sogenannte

0.45

 supposedly

0.44

 Grenz

0.44

 Некоторые

0.43

 NASCAR

0.42

Yoh

0.42

POSITIVE LOGITS

galaxys

0.51

 travaillons

0.50

ouvrir

0.49

ozy

0.49

rawberry

0.49

pl

0.49

Southeast

0.48

is

0.48

 vendas

0.48

SCAPE

0.47

Activations Density 0.000%