INDEX

Explanations

form data serialization

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

 AVENUE

-0.92

Sciences

-0.84

 fiatal

-0.83

nominated

-0.82

 REGULATIONS

-0.81

 إس

-0.81

好处

-0.79

jik

-0.78

КП

-0.77

cq

-0.76

POSITIVE LOGITS

or

0.81

orno

0.77

 transitory

0.75

şey

0.73

Izvori

0.73

HART

0.71

Transient

0.70

 zapro

0.67

handleRequest

0.66

Donnell

0.65

Activations Density 0.004%