INDEX

Explanations

assumptions involving

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

 infine

1.23

または

1.23

또

1.22

এবং

1.20

 সর্বোচ্চ

1.18

analisi

1.16

越多

1.15

razioni

1.15

और

1.12

ണ്ടും

1.11

POSITIVE LOGITS

 Postal

1.31

 Synd

1.28

 spinning

1.26

ippi

1.23

 inhib

1.23

 Epile

1.22

 Emoji

1.22

 penned

1.22

的那

1.19

fet

1.18

Activations Density 0.255%