INDEX

Explanations

abstract concepts or qualifiers

inherently and necessarily

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

 przede

0.55

 केयर

0.54

 также

0.52

 також

0.51

dine

0.47

 моде

0.46

 таксама

0.46

 అలాగే

0.46

 cũng

0.45

trivial

0.45

POSITIVE LOGITS

Potion

0.49

璃

0.47

vés

0.43

ביר

0.43

酒吧

0.42

etl

0.41

ಾತ

0.41

အရ

0.41

更新

0.40

收

0.40

Activations Density 0.004%