INDEX

Explanations

points and fat

The neuron detects sports box‐score statistics, i.e. numeric performance figures paired with stat labels (points, steals, free throws, etc.).

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

ೋಜನ

0.66

 কাবুল

0.65

homogeneous

0.64

ismillahirrah

0.63

最初に

0.63

 EVERY

0.63

รู้จัก

0.62

habilit

0.61

Name

0.60

exception

0.60

POSITIVE LOGITS

 ಹಾಗೂ

0.80

and

0.77

 apiece

0.76

 scorer

0.75

 while

0.73

 ਅਤੇ

0.73

 અને

0.71

 tandis

0.70

 आणि

0.69

 ಮತ್ತು

0.68

Activations Density 0.015%