INDEX

Explanations

foreign words and spellings

interacting with systems

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 Clements

0.49

 Kurz

0.49

 Forst

0.47

curve

0.46

 kurz

0.43

 Burk

0.43

Curve

0.43

Pg

0.43

 Landry

0.42

Mgr

0.42

POSITIVE LOGITS

チェ

0.55

г

0.52

čio

0.50

เต

0.50

 प्रोत्सा

0.48

کری

0.48

اهای

0.47

汆

0.47

 техники

0.47

فی

0.46

Activations Density 0.000%