INDEX

Explanations

tagged categories and items

negations and contrasts

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 bổ

0.40

 indipendente

0.40

 Revista

0.40

leurs

0.39

 hành

0.39

 Brasileiro

0.39

 elettron

0.38

 علیحد

0.38

 Rama

0.38

 دستیاب

0.38

POSITIVE LOGITS

하거나

0.52

temps

0.48

diagn

0.48

temperatur

0.47

jsonData

0.46

หรือ

0.45

ನದ

0.45

kanen

0.43

或

0.43

 снова

0.42

Activations Density 0.002%