INDEX

Explanations

quantities or time periods

The neuron peaks on words that express quantities, measures, or frequency (e.g. abundance, over, every, times, generations).

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

どのように

-1.45

 wherea

-1.36

どのような

-1.33

どこに

-1.32

 jusqu

-1.27

 certific

-1.25

 nouvelles

-1.24

 debilit

-1.23

 besoins

-1.22

儚

-1.20

POSITIVE LOGITS

 this

1.83

 those

1.48

 each

1.43

 regarding

1.39

 characteristics

1.35

 these

1.32

 several

1.29

 dozens

1.27

 Those

1.26

 their

1.24

Activations Density 0.115%