INDEX

Explanations

legal document structure

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

you

-1.73

 what

-1.61

 look

-1.29

 just

-1.27

 блюдо

-1.27

 hope

-1.23

if

-1.16

 definitely

-1.14

 организма

-1.13

 wouldn

-1.09

POSITIVE LOGITS

 السياس

1.34

AFFIRMED

1.27

 Moreover

1.20

ִּ

1.16

 לאחר

1.13

 alcuni

1.11

 कई

1.11

 다른

1.10

UCL

1.09

ַּ

1.09

Activations Density 0.036%