INDEX

Explanations

practice, policy, research, education

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

 practice

-2.67

practice

-2.33

 Practice

-2.27

Practice

-2.20

PRACTICE

-2.00

 PRACTICE

-1.97

 practices

-1.91

 practicing

-1.91

 practiced

-1.90

 practise

-1.83

POSITIVE LOGITS

 beliefs

0.87

 Anwendung

0.83

 theory

0.82

 sermon

0.82

 setuptools

0.78

 دليل

0.76

 mệnh

0.76

 thù

0.74

 wskaz

0.73

ᐅ

0.72

Activations Density 0.024%