This repo. hosts various NLP word embeddings trained on public data for research purposes.
- "custom.vec.bin": fasttext embedding of the first 1 billion bytes of English Wikipedia. LINK: https://drive.google.com/file/d/1mQPKHoa4SQr-skCO5XpzWpOxGB5z02-U/view?usp=sharing Ref: https://fasttext.cc/docs/en/unsupervised-tutorial.html