INDEX

Explanations

protecting from harmful content

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

1.88

 garanti

1.49

м

1.48

 reduc

1.45

 nego

1.43

이

1.41

1.39

고

1.39

an

1.38

 limite

1.36

POSITIVE LOGITS

<unused536>

1.78

StateMachine

1.67

िरपेक्ष

1.66

<unused962>

1.59

<unused1768>

1.58

<unused349>

1.56

Markdown

1.55

<unused111>

1.54

ypse

1.54

<unused593>

1.54

Activations Density 0.351%