INDEX

Explanations

approximations of size or measurement

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

tructuring

-0.75

次は

-0.73

发生了

-0.72

 lowly

-0.71

méable

-0.70

 their

-0.69

classmethod

-0.69

hasMany

-0.68

 very

-0.68

 Ages

-0.68

POSITIVE LOGITS

 size

2.36

 about

2.13

size

1.81

 roughly

1.79

 sebesar

1.61

 tamaño

1.60

 Size

1.52

 bigger

1.52

Size

1.49

 sizeof

1.38

Activations Density 0.035%