INDEX

Explanations

the act of conducting in-depth analysis or exploration

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

 examination

-0.91

 Examination

-0.87

Examination

-0.81

 examinations

-0.74

 EXAMINATION

-0.73

 examined

-0.72

 probe

-0.67

examination

-0.66

 Examinations

-0.65

 exam

-0.64

POSITIVE LOGITS

 investigate

1.38

 Investigate

1.06

 investigates

0.86

 nahilalakip

0.86

fjspx

0.78

 EconPapers

0.73

Datuak

0.68

offsetof

0.65

 investigar

0.64

 investigating

0.63

Activations Density 0.004%