INDEX

Explanations

the word "both" (case insensitive) in the text

both

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

SourceChecksum

-0.91

SequentialGroup

-0.71

الإنجليزية

-0.65

 referenties

-0.65

elemField

-0.63

 myſelf

-0.61

 Houſe

-0.60

 juſt

-0.60

PMailer

-0.60

 houſe

-0.59

POSITIVE LOGITS

CloseOperation

0.62

AccessException

0.62

 those

0.57

RESPOND

0.54

ricultural

0.54

 these

0.53

agama

0.53

CHORUS

0.52

rogly

0.51

holms

0.50

Activations Density 2.303%