INDEX

Explanations

auf dem

auf followed by a noun phrase

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

ージュ

0.47

о

0.47

ড়

0.44

大きい

0.42

着

0.42

at

0.41

クス

0.40

by

0.40

 Kwiatkowski

0.40

Kris

0.39

POSITIVE LOGITS

 manière

0.58

 wijze

0.56

 начин

0.54

 places

0.53

 જગ્યા

0.51

 place

0.49

 maniera

0.49

 τρόπο

0.49

 şekilde

0.48

 места

0.48

Activations Density 0.018%