INDEX

Explanations

two

The neuron strongly activates on words that introduce or enumerate options in a list—e.g. pluralized items and spelled-out numbers used to mark choices or steps.

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

 semua

0.98

 كافة

0.98

 allemaal

0.94

all

0.92

 tất

0.90

 সবগুলো

0.89

 모든

0.88

সব

0.88

 सभी

0.86

 trio

0.86

POSITIVE LOGITS

two

1.56

两个

1.55

这两个

1.53

兩個

1.52

 deux

1.48

 två

1.48

two

1.46

 две

1.45

 beiden

1.41

deux

1.41

Activations Density 0.267%