INDEX

Explanations

unauthorized access or jobs

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

有时候

-1.02

 oubli

-0.91

 ſer

-0.88

 некото

-0.88

 verzi

-0.87

 talvez

-0.85

всем

-0.83

ӑ

-0.83

 alerg

-0.83

 ciga

-0.82

POSITIVE LOGITS

any

2.02

or

1.53

 will

1.34

 cualquier

1.22

ANY

1.20

 anything

1.18

Any

1.18

ANY

1.16

任何

1.16

 this

1.14

Activations Density 0.016%