INDEX

Explanations

code constructs and programming keywords

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

invokeLater

-0.82

 parishes

-0.80

Ę

-0.78

-0.77

 Dioxide

-0.77

ә

-0.77

Empresa

-0.75

Ӏ

-0.73

Oxy

-0.73

 PREPAR

-0.73

POSITIVE LOGITS

 reject

0.81

 erhielt

0.77

ﾖ

0.77

puede

0.75

フラン

0.74

 clip

0.74

当時は

0.74

caron

0.74

formance

0.73

後で

0.73

Activations Density 0.011%