INDEX

Explanations

people

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 épaules

-0.74

 bénéfices

-0.66

 genoux

-0.61

 lèvres

-0.58

 autorités

-0.57

 jambes

-0.54

 prochaines

-0.54

 idea

-0.52

 navires

-0.52

 âmes

-0.51

POSITIVE LOGITS

 rooms

0.69

 qualities

0.68

 batteries

0.68

 wrappers

0.68

 walls

0.68

 stations

0.68

 gardens

0.68

 dishes

0.68

 cabinets

0.68

 tubes

0.67

Activations Density 0.271%