INDEX

Explanations

Consumer Financial Protection Bureau

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 shot

-0.93

idores

-0.88

美味し

-0.84

gnore

-0.83

paramref

-0.81

 możli

-0.81

 OFFICIAL

-0.81

୩

-0.81

 actuated

-0.80

fié

-0.80

POSITIVE LOGITS

 Department

1.35

 Occupational

1.16

 Office

1.11

 Securities

1.11

 Army

1.09

 National

1.06

 Centers

1.03

 Departments

1.03

 Joint

0.97

 Treasury

0.95

Activations Density 0.088%

No Known Activations