INDEX

Explanations

Replace with placeholders

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Encyclopedia

1.93

ashita

1.88

aniye

1.69

ƙ

1.68

ooks

1.67

 incorporación

1.67

 cintura

1.66

πομπ

1.65

enzie

1.62

ixeira

1.61

POSITIVE LOGITS

heid

2.38

 вся

2.06

ност

1.98

ائيل

1.98

miš

1.97

тки

1.94

راهيم

1.91

 стоя

1.90

にして

1.89

 здания

1.89

Activations Density 0.006%