INDEX

Explanations

objects and their properties

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

 thirds

1.22

 things

1.19

 cucumbers

1.18

 many

1.16

 skirts

1.14

等を

1.12

 displacements

1.12

 lique

1.12

eds

1.11

大多數

1.10

POSITIVE LOGITS

except

1.27

 except

1.26

 تقريبا

1.17

 excepto

1.16

 excepting

1.05

Except

1.03

 EXCEPT

1.03

ionario

1.00

Angel

0.95

الله

0.93

Activations Density 0.711%