INDEX

Explanations

punctuation followed by common words

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

 like

-1.69

 maille

-1.53

ׇ

-1.52

as

-1.48

 плата

-1.38

 король

-1.35

нью

-1.31

 today

-1.29

 ویتامین

-1.29

 блюдо

-1.27

POSITIVE LOGITS

↵

2.13

וּ

1.67

 две

1.60

 keď

1.59

 детства

1.49

 dwie

1.40

สอง

1.38

 brag

1.38

 sartén

1.36

 время

1.34

Activations Density 0.034%