technical discussions or code

The neuron activates on numeric tokens containing decimal points (i.e. floating-point numbers).

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

 Бы

1.02

 машины

0.95

тировать

0.94

 Казіно

0.93

чего

0.92

 escalera

0.91

 канторы

0.90

 любы

0.90

 Если

0.89

 Якщо

0.88

POSITIVE LOGITS

0.88

0.80

0.79

0.77

0.73

0.72

ally

0.70

from

0.70

istä

0.69

Activations Density 0.000%