INDEX

Explanations

Code and programming text

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

BindView

-0.50

Magdalena

-0.50

 Tags

-0.49

fml

-0.48

beds

-0.48

 FishBase

-0.48

 Hagen

-0.47

ched

-0.47

szön

-0.47

fully

-0.46

POSITIVE LOGITS

it

0.91

IntoConstraints

0.77

its

0.74

itt

0.65

itte

0.64

itse

0.63

IT

0.63

itest

0.62

parsedMessage

0.61

itle

0.57

Activations Density 0.014%