INDEX

Explanations

prior Japanese patent application

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

Basically

-1.16

実は

-1.12

｢

-1.08

autorisation

-1.05

 notor

-1.00

 Giochi

-0.99

 possibilité

-0.99

手術

-0.98

しかも

-0.98

品種

-0.96

POSITIVE LOGITS

)}

1.20

 bocetos

1.19

 kunna

1.17

']),

1.16

?’

1.15

}*/

1.14

ícias

1.13

é

1.12

PropTypes

1.10

̸

1.10

Activations Density 0.126%