INDEX

Explanations

perfect squares, lists, or code structure

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

 sosok

1.55

 maail

1.43

 kalangan

1.40

 yaşanan

1.40

 sogno

1.37

 unglaublich

1.33

 kisah

1.33

 énorme

1.33

 vivent

1.30

 soprattutto

1.30

POSITIVE LOGITS

 using

2.12

 USING

1.86

using

1.85

Using

1.72

 Using

1.66

 باستخدام

1.61

 следующим

1.57

 utilizing

1.45

 separately

1.45

 manually

1.40

Activations Density 5.115%