INDEX

Explanations

days of the week

The neuron strongly activates on mentions of days or weeks—especially when paired with numeric settings (e.g. “first day of the week,” “DayOfWeek,” week numbers).

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

...

-1.51

衪

-1.43

-1.40

 două

-1.34

一つの

-1.30

one

-1.29

 обеспечить

-1.28

 falleció

-1.24

 religi

-1.23

POSITIVE LOGITS

 erlauben

1.42

 hoger

1.37

1.35

 invece

1.34

 stuks

1.34

UNCIL

1.30

gleichen

1.27

legungen

1.27

อย่างไร

1.24

′

1.24

Activations Density 0.006%