INDEX

Explanations

medicine or languages

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

 pharmacists

-0.86

 sternum

-0.80

Spoljašnje

-0.73

 estekak

-0.72

 تضيفلها

-0.66

IUrlHelper

-0.66

 Roskov

-0.64

)|^{

-0.63

</caption>

-0.62

Географи

-0.62

POSITIVE LOGITS

 Übergang

0.51

rill

0.43

付け

0.41

junct

0.41

strato

0.40

Кры

0.40

 män

0.40

الإنجليزية

0.40

oyd

0.40

 camiones

0.40

Activations Density 0.013%