INDEX

Explanations

auxiliary verbs such as "is", "are", and "have".

are/were

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

Negative Logits

is

-1.11

was

-0.93

has

-0.77

was

-0.74

is

-0.68

it

-0.65

isn

-0.64

Is

-0.62

 CanadaChoose

-0.61

 lèvres

-0.61

POSITIVE LOGITS

 secrets

0.91

 methods

0.90

 procedures

0.84

 virtues

0.83

 techniques

0.82

 options

0.82

 attributes

0.82

 reactions

0.81

 adjustments

0.81

 strategies

0.81

Activations Density 17.086%