INDEX

Explanations

possessive pronouns or earning

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

 únicas

-1.48

 nutrientes

-1.34

 aplicar

-1.33

 afectadas

-1.25

Incorrect

-1.24

","+

-1.23

:'',

-1.22

 urgencia

-1.20

 nødvendig

-1.20

тель

-1.19

POSITIVE LOGITS

 inconce

1.49

서는

1.49

枼

1.40

 Instead

1.35

ly

1.34

砹

1.31

 intrigu

1.31

1.29

 Polícia

1.28

 instead

1.27

Activations Density 0.014%