INDEX

Explanations

the word "facial".

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

 שוליים

-0.93

PreExecute

-0.80

Linki

-0.78

 Kessel

-0.76

สือ

-0.76

sidemargin

-0.75

 مشين

-0.75

ագրություններ

-0.75

 financieras

-0.73

iguity

-0.71

POSITIVE LOGITS

 facial

1.05

Facial

0.94

facial

0.93

 Facial

0.90

 fibrillation

0.71

 Schre

0.70

="&#

0.69

óch

0.66

pthread

0.64

fac

0.64

Activations Density 0.002%