INDEX

Explanations

corrections and updates

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

excellents

-1.52

 дуже

-1.37

おはよう

-1.37

ጐ

-1.30

 veľmi

-1.27

vollen

-1.27

Dauer

-1.27

глежда

-1.27

 portátil

-1.24

leuchte

-1.21

POSITIVE LOGITS

 from

1.80

and

1.69

 which

1.30

 اللاعب

1.30

any

1.29

 whether

1.29

Any

1.27

 Whether

1.25

 Arzne

1.24

 Arzneimit

1.20

Activations Density 0.043%

No Known Activations