INDEX

Explanations

in

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

built

-0.58

parsedMessage

-0.57

 built

-0.56

classnames

-0.55

afficheront

-0.55

 dibangun

-0.55

شاء

-0.55

orious

-0.54

oredCriteria

-0.54

build

-0.53

POSITIVE LOGITS

TagHelper

0.84

 nakalista

0.63

 المعيارى

0.59

 InputDecoration

0.54

 concernés

0.53

 connaissent

0.52

 therefrom

0.52

CodedInputStream

0.51

 Carga

0.50

 marquées

0.49

Activations Density 0.040%