INDEX

Explanations

the words "was" and "is"

was/is

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

 يتيمه

-0.82

GOTREF

-0.68

Cal

-0.48

ագրություն

-0.47

twimg

-0.46

((

-0.45

 acet

-0.45

Cal

-0.45

antMatchers

-0.44

dt

-0.44

POSITIVE LOGITS

Autoritní

0.81

DockStyle

0.66

dasarkan

0.60

İstinadlar

0.60

SharedCtor

0.57

sidemargin

0.54

flashdata

0.54

Bronnen

0.53

ItemBackground

0.53

uggle

0.52

Activations Density 0.505%