Arthur
Arthur
For sure
Could you fix the `ERROR tests/utils/test_file_utils.py - AttributeError: module transformers.models.nemotron has no attribute NemotronForPreTraining` failing tests as well? 🤗
If it is adapted from you can remove the copied from of the class (unless a function alone is different, in that case, you can add # Ignore Copy to...
Thanks a LOT for your hard work, almost finished!
IMO the best thing is to use `PreTokenizerFast`
> But PreTrainedTokenizerFast does not support directly load setencepiece (tokenizer.model) file. It does not, but we can simply convert it in the conversion script! It's really up to you 🤗
failing doc: ``` - model_doc/nemotron Add them to ../transformers/docs/source/en/_toctree.yml. ``` and we can merge!
Let's go on slack @gante !
🫠sounds like kwargs getting lost maybe?
I think #33742 should fix it