INDEX

Explanations

example followed by colon or comma

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

Aş

1.60

iza

1.50

isation

1.37

 giây

1.35

ité

1.32

또

1.31

Aspect

1.30

 arrayOf

1.30

 назна

1.29

Ẓ

1.29

POSITIVE LOGITS

seriously

2.38

houses

2.36

 seriously

2.23

specifically

2.17

really

2.14

 Seriously

2.14

literally

2.13

roughly

2.11

 really

2.10

 literalmente

2.10

Activations Density 0.011%