INDEX

Explanations

of

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

labus

-0.56

 dụ

-0.51

yti

-0.46

MAJ

-0.45

thed

-0.44

<bos>

-0.44

 ITIS

-0.43

timus

-0.42

titu

-0.41

Portail

-0.41

POSITIVE LOGITS

ьаж

0.69

djangoproject

0.66

 JpaRepository

0.66

QMetaType

0.65

<?>>

0.63

NameInMap

0.62

 esternos

0.62

Bibliograf

0.60

BagLayout

0.59

ollectionView

0.57

Activations Density 0.006%