INDEX

Explanations

shapes and numbers

text-based visual layout

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

 conveniently

0.48

 renseignements

0.46

ewell

0.44

 warrantless

0.44

wohl

0.43

 effectually

0.43

 fortuit

0.43

 disregarded

0.42

zzet

0.42

 zero

0.42

POSITIVE LOGITS

 삼각형

0.57

 Pattern

0.56

镂

0.56

 पैटर्न

0.55

 балко

0.54

 onCreate

0.51

 三角

0.51

 Horizontal

0.50

对称

0.50

 Patterns

0.49

Activations Density 0.017%