Hexa has 6 clusters in 3 dimensional space. Cluster centers are (+-4, 0, 0), (0, +-4, 0), (0, 0, +-4). Variances are 1. Each cluster contains 100 samples.
For this demo I used a thousand neurons. The number of neurons is greater than the number of data. It is "Emergent SOM" approach. (rf. A. Ultsch, "Data Mining and Knowledge Discovery with Emergent Self-Organizing Feature Maps") I recommend to check "Hide Vertex" checkbox. When you select "U-Matrix" more white neurons means the neuron is more distant from its immediate neighbors in fashion of high dimensional data space.