INDEX

Explanations

appreciation, random selections, will be

lower level, without

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

ች

0.41

ніше

0.41

ግር

0.41

】

0.40

뎌

0.40

 двох

0.40

 эки

0.40

бята

0.40

 cardinality

0.40

 تجارت

0.40

POSITIVE LOGITS

শের

0.46

Law

0.44

adium

0.43

itu

0.43

law

0.42

ﬂ

0.41

round

0.40

reasonable

0.40

taking

0.40

{\

0.39

Activations Density 0.001%