INDEX

Explanations

percentages and changes

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

 which

-1.09

canActivate

-0.95

ModelAndView

-0.91

haltet

-0.89

MultipartFile

-0.86

 Decken

-0.86

onload

-0.85

 tiroirs

-0.83

Callback

-0.83

 jendela

-0.82

POSITIVE LOGITS

to

2.27

at

1.63

 settle

1.01

figurine

0.99

cmos

0.95

stanley

0.92

pioneer

0.91

valentine

0.90

oriental

0.90

Settle

0.89

Activations Density 0.012%