INDEX

Explanations

lists starting with introductory phrases

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

 преимущественно

2.36

陆续

2.21

ดวก

2.13

畊

2.10

 głównie

2.07

 principalement

2.07

 прадстаў

2.06

激烈

2.05

 також

2.04

ớm

2.04

POSITIVE LOGITS

that

1.78

 siapa

1.72

 dass

1.62

이자

1.54

those

1.54

al

1.46

 kişinin

1.45

andra

1.42

 that

1.42

bike

1.41

Activations Density 0.320%