INDEX

Explanations

language and context diversity

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

or

2.29

ou

2.06

 atau

1.91

 oder

1.87

 veya

1.72

 или

1.65

 eller

1.57

または

1.54

或

1.54

 hoặc

1.53

POSITIVE LOGITS

 intertwined

2.63

 inextricably

2.58

 criaturas

2.44

无比

2.33

不已

2.31

 کردار

2.31

 hardworking

2.30

 zowel

2.26

矗

2.25

 beset

2.24

Activations Density 0.267%