first version of wiki-sentence-alignment fine-tuned BERT model and the tokenizer

e188626