INDEX

Explanations

nationality and associated descriptor

The neuron flags adjectives or nouns denoting geographic or cultural origin (nationality, regional, or ethnic descriptors).

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

 that

-1.68

his

-1.58

 žen

-1.34

είτε

-1.34

 für

-1.33

it

-1.30

erobic

-1.29

 žiad

-1.28

 untuk

-1.28

for

-1.27

POSITIVE LOGITS

 both

1.37

and

1.23

 キッチン

1.21

Both

1.19

style

1.19

while

1.15

 ハン

1.15

GetItem

1.14

 Sedangkan

1.13

飲

1.13

Activations Density 0.053%