INDEX

Explanations

please

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 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

 please

-0.91

please

-0.80

 пожалуйста

-0.63

 veuillez

-0.53

pls

-0.52

 Please

-0.50

 PLEASE

-0.50

 hittar

-0.48

 participé

-0.48

reflect

-0.48

POSITIVE LOGITS

niająca

0.98

Personensuche

0.92

 cherchés

0.85

 itſelf

0.82

 myſelf

0.78

 Offisielt

0.75

 ſever

0.74

WithMany

0.73

 Egli

0.71

TagMode

0.70

Activations Density 0.052%