INDEX

Explanations

references to the importance or consequence of a particular point

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

цездатний

-0.79

Hence

-0.75

($__

-0.73

 "..\..\

-0.72

Personendaten

-0.72

AnchorTagHelper

-0.71

TagMode

-0.71

 disambiguazione

-0.70

modelBuilder

-0.69

UnusedPrivate

-0.66

POSITIVE LOGITS

 hence

1.41

hence

0.90

iş

0.54

かつ

0.54

 indeed

0.54

 rôles

0.52

 причем

0.51

しかも

0.51

indeed

0.51

 lento

0.50

Activations Density 0.002%