INDEX

Explanations

defense mechanism

numbering or numerical patterns in source code, scientific text, and technical documentation.

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 २०

-0.79

周辺

-0.74

posal

-0.71

uintptr

-0.71

 Promo

-0.69

boldsymbol

-0.68

HttpHeaders

-0.68

اني

-0.67

优

-0.66

兎

-0.66

POSITIVE LOGITS

Nine

0.88

Jiu

0.75

 Nineteenth

0.74

idão

0.70

FileStream

0.70

 Nineteen

0.70

 علت

0.68

Rund

0.68

Setiap

0.67

 Nine

0.66

Activations Density 0.485%