INDEX

Explanations

jurisdiction or responsibility

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

Websites

-0.88

先代

-0.88

剑

-0.86

пев

-0.85

 cherchés

-0.84

ellemző

-0.82

борка

-0.81

Darth

-0.81

ajiban

-0.79

 alarme

-0.79

POSITIVE LOGITS

 belongs

1.62

 belong

1.60

 domain

1.55

 responsibility

1.41

 properly

1.41

 matters

1.39

belongs

1.35

domain

1.29

 competencia

1.28

 matter

1.27

Activations Density 0.036%