INDEX

Explanations

The main thing this neuron does is detect numerical data (e.g. counts, measurements, percentages, and other quantitative 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

No Comments

Negative Logits

 Benutz

1.02

0.93

vamo

0.88

 separ

0.85

gruppe

0.85

ur

0.84

kuv

0.84

planes

0.83

chord

0.82

bilder

0.82

POSITIVE LOGITS

 employee

1.82

 праців

1.80

 salariés

1.73

職場

1.70

员工

1.70

职工

1.70

Employers

1.69

 Employee

1.68

 employer

1.66

 workplace

1.66

Activations Density 1.336%