Continual pre-training for multilingual support (extend embedding matrix and tokenizer)
#46
by
ibotana
- opened
Hi there!
By any chance, is there any ongoing research extending this model to the multilingual version?