INDEX

Explanations

Swedish/Danish words, often followed by verbs

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 gerçekleştir

1.86

 između

1.70

되어

1.69

 이루

1.66

 sağlan

1.65

 faaliyet

1.62

separate

1.60

 distint

1.60

 내려

1.57

侈

1.57

POSITIVE LOGITS

haha

1.43

 مجھے

1.30

不想

1.27

 haha

1.25

想到

1.22

出来了

1.22

ahaha

1.21

oooo

1.21

偶尔

1.21

 stuff

1.20

Activations Density 0.004%