INDEX

Explanations

thousands sophistication

The neuron detects numeric or statistical expressions (e.g. numbers, percentages, monetary amounts) in the text.

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

 مِنْ

0.67

ographical

0.66

 geometrical

0.65

────────

0.65

 Variations

0.63

 definite

0.63



0.63

 geändert

0.62

 Width

0.61

 இருந்தாலும்

0.61

POSITIVE LOGITS

 decenas

0.83

 svært

0.80

 Biden

0.80

 Duterte

0.75

 incentivize

0.75

 undermines

0.72

 egregious

0.72

 overwhelmingly

0.72

 neoliberal

0.72

Biden

0.71

Activations Density 0.000%