INDEX

Explanations

Filters, cancel culture, pandas, Moon

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

 код

1.30

男人

1.25

猫

1.22

<0xEC>

1.22

 Boyfriend

1.20

 ло

1.18

ハン

1.18

감

1.18

ทด

1.16

妈妈

1.15

POSITIVE LOGITS

 ensuring

1.30

 allowing

1.29

 preventing

1.27

 ensures

1.22

 guaranteeing

1.22

 prevents

1.20

 whose

1.17

 gewährleisten

1.17

 encourages

1.17

 providing

1.16

Activations Density 0.016%