INDEX

Explanations

abstract concepts and their impact

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

attva

0.84

 тех

0.83

 receptionist

0.78

 vosotros

0.75

suv

0.75

 appellant

0.75

 которыми

0.74

 mane

0.74

 society

0.72

POSITIVE LOGITS

Variance

1.22

 Variance

1.17

 variance

1.15

variance

1.12

Entropy

1.07

Lind

1.07

ዊ

1.06

ٹن

1.06

டுகிறது

1.06

Ost

1.05

Activations Density 0.166%