INDEX

Explanations

questions

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

 lenker

-0.87

IndentedString

-0.82

 ویکی‌پدیا

-0.72

)";

-0.71

kloped

-0.70

 AssemblyTitle

-0.68

 الرياضيه

-0.68

.";

-0.67

VersionUID

-0.65

 يتيمه

-0.65

POSITIVE LOGITS

Does

0.66

 Does

0.64

 Tiene

0.59

Has

0.57

 Viene

0.51

Has

0.50

pires

0.49

Doesn

0.47

 Doesn

0.46

Can

0.43

Activations Density 0.002%