INDEX

Explanations

linked, attached, or speculated

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

⫘

-1.45

 tecnologías

-1.44

 чтобы

-1.42

 позволя

-1.38

流星

-1.37

！(

-1.35

 các

-1.33

 ostat

-1.32

或

-1.29

 всех

-1.28

POSITIVE LOGITS

for

1.84

any

1.49

</strong>

1.45

 využ

1.39

 lèvres

1.39

 muun

1.39

جستارهای

1.34

 旅游

1.33

 户型

1.33

 nabí

1.32

Activations Density 0.026%