INDEX

Explanations

scientific journals and studies

journal names in academic citations.

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 others

-1.40

 ไม่มี

-1.38

 다른

-1.34

헷

-1.33

不像

-1.32

 รวม

-1.30

嗰

-1.30

--

-1.27

 próxima

-1.25

一起去

-1.24

POSITIVE LOGITS

</h6>

1.65

it

1.55

">

1.42

 just

1.39

</h2>

1.34

">

1.34

 kwamba

1.34

 this

1.30

any

1.30

'){

1.28

Activations Density 0.049%