INDEX

Explanations

large numbers and amounts

The neuron fires on phrases that quantify large amounts or scales (e.g. “large numbers of,” “vast amounts of”).

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

all

-1.86

 gratuitas

-1.77

大約

-1.42

岨

-1.35

who

-1.30

 necesito

-1.27

 возрасте

-1.27

卨

-1.26

 conocidos

-1.26

 gebruik

-1.25

POSITIVE LOGITS

You

1.46

沏

1.42

are

1.42

['

1.36

,"

1.33

 amounts

1.32

 amount

1.32

 Because

1.30

 because

1.27

 betreffenden

1.27

Activations Density 0.055%