INDEX

Explanations

brands, mortality, response, assessing

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

€™

-1.57

bof

-1.38

aning

-1.38

alism

-1.37

 roughly

-1.37

omeness

-1.36

otropy

-1.34

agerie

-1.33

ometries

-1.32

orship

-1.31

POSITIVE LOGITS

 applau

1.23

and

1.22

 значительно

1.19

쏟

1.18

地说

1.08

Beschrijving

1.07

 图解

1.06

 arou

1.05

有个

1.05

田中

1.05

Activations Density 0.043%