INDEX

Explanations

website availability or access

states or conditions

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

surprise

0.44

رح

0.42

آن

0.41

 که

0.41

تد

0.41

た

0.40

 لنا

0.39

 volgens

0.38

gruppen

0.38

 według

0.38

POSITIVE LOGITS

 opened

0.59

 reopened

0.54

 Opened

0.52

 ceased

0.49

ಿದ್ದಾರೆ

0.47

 unscrupulous

0.47

 Coronavirus

0.47

Opened

0.46

 Retina

0.46

 abrió

0.46

Activations Density 0.003%