INDEX

Explanations

one followed by noun or preposition

one followed by number-related words

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

冖

0.49

AIP

0.49

 watery

0.48

欢

0.47

 iure

0.46

 contrat

0.45

 জাতি

0.45

 expansive

0.45

 communication

0.44

 baggy

0.44

POSITIVE LOGITS

Two

0.52

STR

0.47

Three

0.47

 hundred

0.47

two

0.47

Hundred

0.47

HOM

0.45

HY

0.45

HC

0.45

eram

0.44

Activations Density 0.005%