INDEX

Explanations

academic citations

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

 généralement

1.82

 correspondiente

1.80

 generalmente

1.78

 probablement

1.76

 geralmente

1.74

 probabilmente

1.73

 éventuellement

1.70

 probablemente

1.69

 provinc

1.69

 potencialmente

1.67

POSITIVE LOGITS

 দাশ

1.41

 한번

1.37

 하루

1.37

ించి

1.33

 thách

1.32

한

1.32

 제시

1.31

եցին

1.29

把我

1.27

把

1.27

Activations Density 0.123%