INDEX

Explanations

places, events, actions

how systems/models interpret data

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

וני

0.52

ונים

0.52

frameN

0.50

шаем

0.50

峥

0.49

чая

0.48

 dealings

0.47

观念

0.47

ᄉ

0.47

Views

0.46

POSITIVE LOGITS

 Hurricane

0.57

 Sunday

0.52

 University

0.50

 Tuesday

0.50

 Saturday

0.49

aki

0.48

Hurricane

0.46

 completed

0.46

playground

0.46

 hurricane

0.46

Activations Density 0.000%