INDEX

Explanations

numbers followed by of

strategies and explanations

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

illée

0.45

leaning

0.45

quez

0.43

^{-}

0.40

えて

0.40

^{-

0.40

aisse

0.39

ルイ

0.39

்சைடு

0.39

essi

0.39

POSITIVE LOGITS

变更

0.42

 capacitances

0.41

 worldwide

0.40

 Однако

0.40

 المف

0.40

 omnichannel

0.40

 Fintech

0.40

 힘들

0.39

ابعه

0.39

тических

0.38

Activations Density 0.000%