INDEX

Explanations

geographical entities and names

The neuron activates on proper nouns and acronyms (capitalized names of people, places, organizations, classes, etc.).

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

 Sadly

-3.09

 multitude

-2.97

 There

-2.95

</b>

-2.89

 Usually

-2.81

 Though

-2.78

 Generally

-2.78

</td>

-2.75

 Notwithstanding

-2.69

 Typically

-2.67

POSITIVE LOGITS

3.44

 uſed

3.44

衚

3.05

涠

2.58

antaranya

2.56

advert

2.56

 sentral

2.50

 secoli

2.48

 Minggu

2.42

栊

2.41

Activations Density 0.004%