INDEX

Explanations

for example

The neuron detects numerical expressions—years, quantities, statistics, and other number‐related tokens.

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

 dwóch

0.99

 δύο

0.82

two

0.79

 deux

0.77

dva

0.76

 zwei

0.75

两个

0.73

TWO

0.73

 двох

0.71

 trzech

0.70

POSITIVE LOGITS

比如说

1.42

 meliputi

1.41

比如說

1.38

包括

1.37

 मसलन

1.30

 مثلا

1.26

 incluyendo

1.26

例えば

1.23

 包括

1.23

 شامل

1.23

Activations Density 0.700%