INDEX

Explanations

inexpensive

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

 שוליים

-0.85

Autoritní

-0.81

出版年

-0.79

хьтан

-0.79

 saites

-0.79

ConstraintMaker

-0.78

 وتسجيلات

-0.76

 EconPapers

-0.73

 varandra

-0.72

 становника

-0.72

POSITIVE LOGITS

่าง

0.50

 careers

0.49

pula

0.47

UnsafeEnabled

0.47

 boll

0.46

 bien

0.45

ARR

0.45

EEEE

0.44

Normalize

0.43

Activations Density 0.044%