How to use bylang/hf-tokenizer with Transformers:
# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("bylang/hf-tokenizer", dtype="auto")
