INDEX

Explanations

action followed by outcome

finding and identifying

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

棿

0.48

कृत

0.47

ަތ

0.46

૦

0.46

蒯

0.46

空

0.45

الصفحه

0.45

别的

0.44

పురం

0.44

 עצ

0.44

POSITIVE LOGITS

els

0.49

 Burj

0.47

=\

0.46

lds

0.45

ก่อน

0.45

 inder

0.45

 literatura

0.44

vals

0.44

rnd

0.44

Buj

0.44

Activations Density 0.001%