INDEX

Explanations

finger related actions and anatomy

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

栳

-2.98

he

-2.50

 enti

-2.39

 денег

-2.36

 sehe

-2.33

釹

-2.30

 trochu

-2.23

𝝈

-2.22

栒

-2.22

一定会

-2.22

POSITIVE LOGITS

 После

2.88

jąca

2.77

For

2.70

Ꮨ

2.63

 Для

2.63

the

2.61

膂

2.59

 Following

2.58

Surprisingly

2.56

蜮

2.50

Activations Density 0.006%