INDEX

Explanations

phrases or punctuation that indicates a break in the text

punctuation

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

GraphicsUnit

-0.93

providedIn

-0.82

تقاوى

-0.62

ंदीखरीदारी

-0.62

enumii

-0.61

 AssemblyCulture

-0.60

 getItemId

-0.58

 beginnetje

-0.57

LookAnd

-0.57

 cherchés

-0.56

POSITIVE LOGITS

 vậy

0.62

 nope

0.56

 houſe

0.55

yes

0.54

 stimmt

0.53

 Nope

0.52

 hinweg

0.51

 Houſe

0.50

YES

0.50

Yes

0.49

Activations Density 1.799%