INDEX

Explanations

court filings and judgments

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

breviated

-0.89

 byen

-0.86

 Membuat

-0.85

＝

-0.83

 hembra

-0.82

 tombé

-0.82

 jesús

-0.81

Fordítás

-0.81

 minutter

-0.81

今

-0.80

POSITIVE LOGITS

Turning

1.11

 courts

1.11

 Turning

1.07

 turning

1.05

 assuming

1.02

"[

1.02

 reviewing

1.00

 acceptez

0.98

결

0.97

 որ

0.97

Activations Density 0.009%