INDEX

Explanations

Xin Hua, Housing analyst on ISBN RUN

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

 establishing

-0.71

 during

-0.70

as

-0.70

 exploit

-0.68

logica

-0.67

判

-0.67

 Contrary

-0.67

cocktail

-0.67

at

-0.66

جاب

-0.66

POSITIVE LOGITS

NT

1.01

 Greek

0.91

 Helps

0.90

 Torrey

0.89

tures

0.89

 Concord

0.89

STRONG

0.89

 greek

0.87

 concord

0.85

lighter

0.85

Activations Density 0.016%