INDEX

Explanations

names of people or countries

words with non-Latin characters or special patterns

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

 Cataloging

-0.58

 initComponents

-0.54

rawDesc

-0.53

httphttps

-0.52

AnimationFrame

-0.51

BeginContext

-0.51

 tubeless

-0.50

 ویکی‌پدیای

-0.50

InstrumentedTest

-0.50

!*\

-0.49

POSITIVE LOGITS

انون

0.52

achim

0.49

uio

0.47

chi

0.47

gården

0.46

tuce

0.46

DebuggerNonUser

0.46

 χα

0.46

 foot

0.46

%)$

0.46

Activations Density 0.480%