INDEX

Explanations

foreign words and proper nouns

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

erade

0.95

그런

0.89

histo

0.88

addam

0.80

 Zeiss

0.78

 ejected

0.78

γραφ

0.77

anze

0.77

 Excise

0.75

volle

0.75

POSITIVE LOGITS

校长

1.58

icelli

1.57

 کودکان

1.56

 سریع

1.55

ǔ

1.55

 ясно

1.54

あい

1.53

 abstracto

1.51

 объяснить

1.51

 кто

1.50

Activations Density 0.017%