INDEX

Explanations

determiner followed by descriptive noun

initially focused on or found

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

Codec

0.47

Full

0.46

Deleted

0.46

Armor

0.46

ന്ഥ

0.46

Testing

0.45

Mirror

0.44

 увиде

0.44

FullPath

0.43

ιστή

0.42

POSITIVE LOGITS

spaper

0.48

 النهار

0.47

iak

0.47

at

0.47

as

0.46

ك

0.46

tos

0.46

psons

0.46

gaz

0.45

 للنساء

0.45

Activations Density 0.000%