INDEX

Explanations

qualifiers indicating a contrast or an exception to a statement, and potentially phrases indicating awareness or discovery.

Contrasting ideas or lack of awareness

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

Negative Logits

تقاوى

-0.65

MigrationBuilder

-0.57

)':

-0.56

ابراین

-0.55

].(

-0.53

وأضاف

-0.52

arası

-0.50

zag

-0.47

architecte

-0.46

 />';

-0.45

POSITIVE LOGITS

awtextra

0.65

初めて

0.63

 ModelExpression

0.59

 впервые

0.59

 slowest

0.59

್ದ

0.58

はじめて

0.56

今回

0.56

 longest

0.53

 unheard

0.52

Activations Density 1.298%