INDEX

Explanations

methods and techniques

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

endente

1.28

եկ

1.26

 overhang

1.16

ণ্য

1.13

 condos

1.13

 électron

1.13

 lysine

1.13

 количестве

1.12

 duty

1.10

 alumina

1.07

POSITIVE LOGITS

 yöntem

2.98

方法

2.71

 방법

2.66

 method

2.65

 methods

2.63

methods

2.54

 metod

2.52

 metodi

2.51

 Method

2.48

 方法

2.44

Activations Density 0.684%