date formats and separators

The neuron activates on numeric tokens (including numbers with ordinal suffixes) such as dates, times, and other numeric values.

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

0.89

0.75

0.74

 ২৭

0.73

 ২৩

0.71

0.70

 ২৬

0.69

下旬

0.68

POSITIVE LOGITS

 instances

0.69

 january

0.69

 stycznia

0.66

 Januar

0.62

 localities

0.62

 oktober

0.61

 January

0.60

 Januari

0.60

 trophies

0.59

 januar

0.59

Activations Density 0.125%