INDEX

Explanations

publication years and names

French, other languages, academic contexts

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

 sunt

0.45

뇨

0.40

 bagian

0.39

<unused100>

0.38

 동일

0.38

 기타

0.38

 pekan

0.38

 antiseptic

0.38

다면

0.38

 sulph

0.37

POSITIVE LOGITS

 appré

0.38

 Wörter

0.37

至少

0.36

Essentials

0.36

}^{+}$.

0.36

に対する

0.35

求解

0.35

 predictability

0.35

 competent

0.34

}$).

0.34

Activations Density 0.025%