INDEX

Explanations

first part of, portion of

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

-1.52

 autre

-1.36

いずれ

-1.27

 également

-1.26

werben

-1.25

 besök

-1.23

 разных

-1.23

 valorar

-1.23

 nummer

-1.20

َع

-1.16

POSITIVE LOGITS

 this

1.98

 such

1.74

use

1.23

 that

1.18

 then

1.14

 including

1.13

 work

1.11

the

1.11

...");

1.09

 včetně

1.09

Activations Density 0.152%