from tokenizers import Tokenizer, decoders, models, pre_tokenizers, processors, trainers
from transformers import AutoTokenizer

tokenizer = AutoTokenizer.from_pretrained('Tensoic/Kan-LLaMA-7B-base')
print(tokenizer)
st="ಮಾಡುವುದು ಮಾಡಲು"
print(tokenizer.convert_ids_to_tokens(tokenizer(st)['input_ids']))