INDEX

Explanations

define

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

AndEndTag

-1.20

Tikang

-0.99

 CreateTagHelper

-0.82

GEBURTSDATUM

-0.81

 незавершена

-0.75

abstractmethod

-0.74

WriteTagHelper

-0.74

msgTypes

-0.73

 Приступљено

-0.72

enced

-0.71

POSITIVE LOGITS

0.56

一面

0.50

an

0.41

ětí

0.40

 plates

0.39

mon

0.39

 plait

0.38

 belong

0.37

 dual

0.37

 wider

0.37

Activations Density 0.023%