INDEX

Explanations

number and date formatting

This neuron detects mentions of formatting functions or classes—especially those dealing with numerical or number‐format settings (e.g. “format,” “NumberFormat,” “DecimalFormat,” etc.).

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

 Bagaimana

-1.97

 figlia

-1.77

Seperti

-1.73

ндарт

-1.67

鷽

-1.64

Karakteristik

-1.61

 ہے

-1.61



-1.60

 Apakah

-1.59

quées

-1.59

POSITIVE LOGITS

1.69

“

1.68

1.54

1.51

мо

1.48

1.45

1.43

 jede

1.41

big

1.34

Activations Density 0.001%