INDEX

Explanations

data structures and storage

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

榈

-2.91

’

-2.55

 disiplin

-2.39

 rokok

-2.36

 teknologi

-2.36

喾

-2.27

 visuelle

-2.27

 selben

-2.25

 demokra

-2.22

 occupé

-2.19

POSITIVE LOGITS

↵↵

2.69

みたいです

2.33

you

2.33

Its

2.27

“

2.16

2.13

 рассказывает

2.02

ar

2.02

 realise

2.00

 المسلمين

1.99

Activations Density 0.006%