INDEX

Explanations

numbers related to quantity

The neuron activates on numeric tokens representing counts or quantities (e.g. numbers of children, villages, years).

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

 during

-1.57

 that

-1.56

 while

-1.55

 because

-1.54

 information

-1.52

 which

-1.44

it

-1.39

ázaro

-1.38

or

-1.34

 when

-1.33

POSITIVE LOGITS

 cuatro

1.23

 огромное

1.16

 véhicules

1.13

 SEVEN

1.13

 três

1.10

 trzech

1.09

;,

1.06

 ثلاثة

1.06

小孩子

1.05

ecer

1.05

Activations Density 0.002%