INDEX

Explanations

enter

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

 enter

-0.53

一等

-0.52

ge

-0.49

pre

-0.45

 Civil

-0.43

iency

-0.43

wir

-0.43

 tick

-0.43

Ge

-0.42

POSITIVE LOGITS

afficheront

0.81

 Wikimedijinoj

0.80

Personendaten

0.77

EndGlobalSection

0.74

CloseOperation

0.71

 للمعارف

0.70

 дописавши

0.68

SharedDtor

0.68

يكب

0.66

AccessorTable

0.65

Activations Density 0.102%