INDEX

Explanations

friendly and positive demeanor

friendly or personable interactions

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 unsustainable

0.43

 recursively

0.42

诟

0.42

ជាប់

0.41

朢

0.41

铤

0.40

ildir

0.39

 चिंतन

0.39

퀀

0.39

蠹

0.39

POSITIVE LOGITS

 friendly

2.75

 friendliness

2.50

 Friendly

2.38

friendly

2.36

Friendly

2.30

 freund

1.99

 друже

1.98

 hospitable

1.87

友好

1.86

 courteous

1.78

Activations Density 0.421%