INDEX

Explanations

write something

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

 усі

0.81

 всеми

0.71

 plupart

0.71

جميع

0.67

繇

0.67

 всі

0.65

하며

0.64

வர்களும்

0.63

 zahlreichen

0.61

 تمامی

0.61

POSITIVE LOGITS

 something

5.93

something

5.64

 Something

5.41

Something

5.40

 someth

4.56

 qualcosa

4.42

 sesuatu

4.12

 coś

4.02

 anything

3.92

 iets

3.85

Activations Density 1.722%