INDEX

Explanations

physical properties and their changes

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

 APPELLANT

-1.52

 правительства

-1.48

 الحكومة

-1.43

 السياس

-1.38

 调用

-1.35

 proget

-1.32

 wouldn

-1.30

 potrivit

-1.30

 risks

-1.28

if

-1.25

POSITIVE LOGITS

 سریال

1.59

 سینم

1.54

 دخترانه

1.46

 عاشقانه

1.41

iltà

1.38

ߛ

1.36

compañ

1.36

 ملابس

1.34

丹药

1.31

 pompe

1.29

Activations Density 0.123%