INDEX

Explanations

General capabilities

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

ail

1.56

ails

1.55

ipar

1.51

 bleed

1.44

 echoing

1.43

of

1.42

ati

1.42

 that

1.39

<0xF3>

1.38

and

1.37

POSITIVE LOGITS

 działa

2.09

 ওয়ার

1.92

 adına

1.91

 dwRes

1.91

 sám

1.82

 wykony

1.78

 czyn

1.78

 funktionieren

1.78

 грома

1.76

 posiada

1.76

Activations Density 0.149%