| # DistilBERT with word2vec token embeddings | |
| This model has a word2vec token embedding matrix with 256k entries. The word2vec was trained on 100GB data from C4, MSMARCO, News, Wikipedia, S2ORC, for 3 epochs. | |
| Then the model was trained on this dataset with MLM for 500k steps (batch size 64). The token embeddings were NOT updated. | |