INDEX

Explanations

references to structural or organizational differences

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 BoxFit

-1.08

rungsseite

-1.04

GEBURTSDATUM

-0.96

 للمعارف

-0.91

 disambiguazione

-0.85

 disponibilités

-0.80

 nahilalakip

-0.77

WithIOException

-0.73

addPreferredGap

-0.72

 noqa

-0.72

POSITIVE LOGITS

<eos>

0.51

“

0.49

 change

0.47

aleta

0.46

No

0.45

change

0.45

 ред

0.45

 Back

0.44

 such

0.44

woh

0.44

Activations Density 0.054%