INDEX

Explanations

read, collected, used

explain its use or meanings

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

 ден

0.54

সঙ্গত

0.53

 психологи

0.52

 জ্ঞাত

0.51

 ಚೆ

0.51

 तेंद

0.50

 زید

0.50

 карту

0.49

 इद

0.48

 कलाकार

0.48

POSITIVE LOGITS

ه

0.51

of

0.47

rv

0.47

Fi

0.45

↵↵

0.44

 World

0.42

دو

0.41

Activations Density 0.001%