INDEX

Explanations

printing output in code

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

 repent

1.66

 程序

1.56

沙

1.53

 impro

1.49

 tender

1.48

 lock

1.48

ROM

1.46

 carriers

1.46

血

1.44

非

1.43

POSITIVE LOGITS

dengan

1.93

ằng

1.85

And

1.81

विरुद्ध

1.78

reshape

1.78

लेक्ट

1.76

ırma

1.73

Dengan

1.73

AND

1.72

 संभालने

1.72

Activations Density 0.041%