How did you get the word embeddings of size (B,300) in our_vab_data.npy?

[kaustesseract]

Hello,

I must say amazing work. I was wondering if you could clarify how did you get the word embedding for each word as 300 in the our_vab_data.npy??