INDEX

Explanations

risk assessment criteria

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

so

-1.67

֭

-1.55

 chré

-1.54

 just

-1.48

 обслуживание

-1.45

 آنان

-1.44

 פֿ

-1.39

 cristianos

-1.39

皈

-1.38

-1.37

POSITIVE LOGITS

ֶּ

2.31

ִּ

1.77

ּוֹ

1.74

 кроссовки

1.71

ַּ

1.70

ְּ

1.65

 مباراة

1.64

暧

1.63

 LUIS

1.58

וּ

1.58

Activations Density 0.067%