INDEX

Explanations

technical, administrative, or office-related words

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

şı

2.55

طا

2.39

ted

2.33

طاء

2.25

ตรฐาน

2.25

findpost

2.18

loat

2.17

畜

2.15

양이

2.15

سان

2.14

POSITIVE LOGITS

elik

1.75

ék

1.44

ческой

1.32

edor

1.28

enin

1.27

ческого

1.27

чної

1.26

ческую

1.19

ical

1.18

ะนั้น

1.17

Activations Density 0.314%