How to use aeonium/tokenizer-64k with Transformers:
# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("aeonium/tokenizer-64k", dtype="auto")
