INDEX

Explanations

comparing effects or results

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

Приготовление

-1.56

هذه

-1.43

Ārējās

-1.38

独特

-1.37

Răsp

-1.37

ไม่

-1.35

llllll

-1.34

}$.

-1.32

])

-1.30

乙

-1.30

POSITIVE LOGITS

 other

2.41

 others

1.89

the

1.87

 those

1.53

 другие

1.41

 than

1.38

 some

1.37

很多的

1.29

 других

1.27

かれて

1.27

Activations Density 0.144%