INDEX

Explanations

was

strings phrases that may appear in a resume or biography

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

 privatisation

-0.59

 regia

-0.56

UnknownFields

-0.56

 Meksiku

-0.54

 persecution

-0.54

flink

-0.54

 subordination

-0.53

Demografie

-0.53

 accreditation

-0.53

CodedInputStream

-0.52

POSITIVE LOGITS

ScopeManager

0.60

Gön

0.59

 able

0.52

 faktiskt

0.52

VELAND

0.52

ierende

0.52

 effectivement

0.51

indd

0.51

 reluctant

0.50

NSON

0.49

Activations Density 1.188%