INDEX

Explanations

)

acknowledging, answering, and, Despite, could

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

 коопера

0.50

 Entrepreneurs

0.45

NUS

0.45

erte

0.43

DOS

0.43

 equilíbrio

0.43

лят

0.43

平衡

0.42

 kontrak

0.42

 первых

0.42

POSITIVE LOGITS

third

0.50

raise

0.44

second

0.43

Lt

0.43

ന്ത്രാ

0.43

tooth

0.42

tw

0.42

navy

0.42

uzioni

0.41

dı

0.41

Activations Density 0.000%