INDEX

Explanations

Steps/Johnston

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

 Johnston

-0.88

Johnston

-0.77

Steps

-0.69

 Steps

-0.55

 literally

-0.52

burner

-0.50

ouple

-0.50

steps

-0.48

Minus

-0.47

saw

-0.46

POSITIVE LOGITS

SourceChecksum

0.87

 utafitiHapana

0.74

AsUp

0.73

 noDo

0.71

ngdoc

0.70

Искәрмәләр

0.68

 invokingState

0.67

 Paglinawan

0.66

 gyhoeddwyd

0.64

قایناق‌لار

0.64

Activations Density 1.555%