INDEX

Explanations

run button icons (play, green, blue)

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

 Vocabulary

1.23

Vocabulary

1.19

 vocational

1.07

 Слова

1.06

 profound

1.06

Formatted

1.06

 Vocational

1.03

 Psychological

1.03

 vocabulary

1.03

ElementException

1.02

POSITIVE LOGITS

kop

0.88

 chúng

0.87

Zal

0.87

 başlayalım

0.85

我们在

0.83

 lauren

0.83

 lisa

0.82

 Leslie

0.82

ที่เรา

0.81

 Noël

0.80

Activations Density 0.003%