INDEX

Explanations

German prepositions followed by nouns

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

 dwóch

-1.41

 tři

-1.26

兩種

-1.25

 trzech

-1.24

 عندما

-1.23

fordshire

-1.22

<bos>

-1.22

兩個

-1.19

 dvě

-1.15

 příjem

-1.14

POSITIVE LOGITS

 same

1.25

 nėra

1.13

無し

1.12

 выделя

1.07

 aforementioned

1.07

 humiliating

1.05

elded

1.04

 longed

1.01

様の

1.00

疾病

0.99

Activations Density 0.002%