INDEX

Explanations

comparison or falsehood

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

у

1.47

времен

1.45

 начи

1.31

 outward

1.27

sans

1.26

י

1.24

sounding

1.23

を買

1.22

stuffs

1.22

machines

1.21

POSITIVE LOGITS

ڍ

1.55

ية

1.54

ړه

1.38

வாழ்

1.38

 शकता

1.37

പ്പ്

1.37

्हे

1.36

attutto

1.35

iting

1.34

ட்டர்

1.34

Activations Density 0.000%