INDEX

Explanations

scant

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

poptosis

-0.67

 Chronology

-0.65

 ModelExpression

-0.65

 Hidalgo

-0.64

 JAXBElement

-0.64

 AspNetCore

-0.63

 horizon

-0.62

المناصب

-0.62

 Lordship

-0.59

 Wicidata

-0.57

POSITIVE LOGITS

AxisAlignment

0.49

multirow

0.48

<bos>

0.48

 отношению

0.47

 latter

0.47

polazione

0.47

extAlignment

0.47

eraan

0.46

0.45

tainer

0.44

Activations Density 1.144%