INDEX

Explanations

units: scientific measurements

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

 Find

1.89

する

1.87

 ين

1.79

 quelque

1.74

 لا

1.71

 какой

1.70

の意味

1.70

رك

1.68

 легче

1.68

 كل

1.67

POSITIVE LOGITS

article

1.50

accharides

1.39

larger

1.38

cel

1.38

meg

1.36

aside

1.33

auen

1.31

additional

1.29

adı

1.29

īb

1.27

Activations Density 0.002%