INDEX

Explanations

please contact, visit, email

left or clumsy descriptions

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

 разрабаты

0.46

 cultivés

0.46

établir

0.45

దే

0.44

శు

0.43

 bactéri

0.43

 établi

0.43

針對

0.43

რც

0.42

 hoger

0.42

POSITIVE LOGITS

 removed

0.48

iyy

0.47

 slanted

0.46

 perhaps

0.45

angkah

0.45

lij

0.45

removed

0.44

 inclined

0.44

 mungkin

0.44

IMF

0.44

Activations Density 0.001%