INDEX

Explanations

instances where the described quantity or level is overly high or beyond acceptable limits

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 wireType

-0.59

NegativeButton

-0.57

FXMLLoader

-0.56

 Pair

-0.55

SupportActionBar

-0.55

 AppCompatTheme

-0.52

inflater

-0.52

 rootReducer

-0.52

Personendaten

-0.51

inut

-0.50

POSITIVE LOGITS

 excessive

2.27

 Excessive

2.27

Excessive

2.13

cessive

1.45

 excessively

1.30

 exces

0.99

 exager

0.79

 exag

0.74

 overly

0.72

 exaggerated

0.71

Activations Density 0.002%