INDEX

Explanations

This stands for

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

installment

1.48

ूबी

1.43

irstyle

1.38

ideal

1.34

çant

1.33

ისთვის

1.27

 assistive

1.25

におすすめ

1.24

 Inclusive

1.23

appointments

1.22

POSITIVE LOGITS

And

1.21

مسه

1.20

 volatiles

1.18

 رکھے

1.18

And

1.14

 hardened

1.12

 نفسه

1.12

cology

1.05

mselves

1.03

 marah

1.02

Activations Density 0.000%