INDEX

Explanations

apologies for mistakes

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

之类的

1.01

 gist

0.93

ញ្

0.92

amples

0.91

の内容

0.91

的情况下

0.89

 timings

0.86

 реализации

0.86

的一切

0.85

configs

0.83

POSITIVE LOGITS

 الوحد

1.40

Rosen

1.37

Another

1.35

Advert

1.25

الأ

1.25

Tür

1.24

 أحد

1.24

Soup

1.23

Mass

1.23

Dinner

1.21

Activations Density 0.008%