INDEX

Explanations

as

New Auto-Interp

Configuration

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

tagext

-0.62

Keeping

-0.56

 Keeping

-0.53

easy

-0.53

ínas

-0.52

RetentionPolicy

-0.52

 Eilish

-0.52

-0.51

netinet

-0.51

 keeping

-0.51

POSITIVE LOGITS

!("{

0.49

writeFieldEnd

0.49

RenderAtEndOf

0.48

TagHelper

0.48

 فريبيس

0.48

principalTable

0.47

aarrggbb

0.47

存于互联网档案馆

0.46

 تضيفلها

0.45

 utafiti

0.45

Activations Density 0.001%