INDEX

Explanations

so many or few

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

for

-1.72

 until

-1.59

 에서

-1.56

 again

-1.55

aimana

-1.54

闞

-1.54

by

-1.52

 Caution

-1.51

crops

-1.51

 valore

-1.51

POSITIVE LOGITS

of

2.44

</strong>

1.71

TestingModule

1.68

archical

1.67

(),

1.63

<bos>

1.63

啪

1.61

 Inſ

1.60

autoreleasepool

1.60

 seçenek

1.55

Activations Density 0.013%