INDEX

Explanations

foreign languages and specialized terms

makes sense

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

 took

0.46

 searched

0.46

wanted

0.45

飾

0.43

サイズの

0.42

belongs

0.41

 friend

0.41

 belongs

0.41

 taken

0.40

 satisfactory

0.40

POSITIVE LOGITS

 ხელისუფ

0.46

 проекты

0.45

蛋白质

0.45

umlahan

0.44

 корпора

0.43

 misdemean

0.43

ᐱ

0.43

лер

0.41

<0xA1>

0.41

 консо

0.40

Activations Density 0.000%