INDEX

Explanations

four

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

 document

-0.50

avel

-0.46

 training

-0.44

ged

-0.43

 televisor

-0.42

herzog

-0.42

 alight

-0.41

 bekend

-0.41

 Gregor

-0.41

kehrt

-0.40

POSITIVE LOGITS

 nahilalakip

0.88

 مشين

0.83

Lähteet

0.82

tagHelperRunner

0.78

 autorytatywna

0.77

mybatisplus

0.77

MessageState

0.77



0.75

 AssemblyCulture

0.75

 للمعارف

0.74

Activations Density 0.002%