toTokenizer.from_pretrained("facebook/m2m100_418M")

    >>> text_to_translate = "Life is like a box of chocolates"
    >>> model_inputs = tokenizer(text_to_translate, return_tensors="pt")

    >>> # translate to French
    >>> gen_tokens = model.generate(**model_inputs, forced_bos_token_id=tokenizer.get_lang_id("fr"))
    >>> print(tokenizer.batch_decode(gen_tokens, skip_special_tokens=True))
    ```
a_