INDEX

Explanations

the word "prosecution" related to legal proceedings

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

 eins

-0.67

 Leap

-0.66

障

-0.63

zay

-0.62

Cla

-0.62

 reasons

-0.61

 extrater

-0.61

 rest

-0.60

Lea

-0.60

DQ

-0.60

POSITIVE LOGITS

 prosecute

1.10

 prosecution

1.01

 Prosecution

0.95

secution

0.91

 prosecuted

0.90

secuted

0.88

NamedQueries

0.84

secutions

0.82

 prosecuting

0.81

 prosecutions

0.72

Activations Density 0.009%