INDEX

Explanations

removal

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

tagHelperRunner

-0.73

 initComponents

-0.64

 OFDb

-0.62

 interrump

-0.59

StructEnd

-0.58

 Ignoring

-0.57

 expel

-0.57

 Immune

-0.56

 שוליים

-0.56

 interrupted

-0.55

POSITIVE LOGITS

Décès

0.59

 Chwiliwch

0.50

 acer

0.44

>[]

0.43

 certain

0.43

 potentially

0.42

 caramelo

0.41

 formal

0.40

 adequate

0.40

 diretto

0.40

Activations Density 0.016%