INDEX

Explanations

i understand followed by explanation

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

-2.09

-1.86

也

-1.85

 These

-1.76

</h2>

-1.67

 This

-1.55

-1.53

</strong>

-1.52

 Unternehmer

-1.49

 include

-1.45

POSITIVE LOGITS

 administrativos

1.70

部は

1.62

vaila

1.60

 reporte

1.52

¹,

1.51

 saud

1.50

见状

1.48

 tecnológicos

1.47

』

1.45

ולות

1.44

Activations Density 0.013%