transformersbook/codeparrot-valid · Repo card metadata block was not found

When trying to run the notebook provided at L7 of the HF's LLM course, when I try to run this block:

from datasets import load_dataset

split = "valid"
filters = ["pandas", "sklearn", "matplotlib", "seaborn"]

data = load_dataset(f"transformersbook/codeparrot-{split}", split=split, streaming=True)
filtered_data = filter_streaming_dataset(data, filters)

I get the following error:

Repo card metadata block was not found. Setting CardData to empty.
WARNING:huggingface_hub.repocard:Repo card metadata block was not found. Setting CardData to empty.
---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
/tmp/ipython-input-10-3880190162.py in <cell line: 0>()
      6 filters = ["pandas", "sklearn", "matplotlib", "seaborn"]
      7 
----> 8 data = load_dataset(f"transformersbook/codeparrot-{split}", split=split, streaming=True)
      9 filtered_data = filter_streaming_dataset(data, filters)

11 frames
/usr/local/lib/python3.11/dist-packages/fsspec/utils.py in glob_translate(pat)
    727             continue
    728         elif "**" in part:
--> 729             raise ValueError(
    730                 "Invalid pattern: '**' can only be an entire path component"
    731             )

ValueError: Invalid pattern: '**' can only be an entire path component

And the same happens if I skip that block and try to execute the next one (which does something pretty similar on a selected subset of data, I assume).
Is there anything I should set up first on my side, or is there any issue with the data?

Best!

EDIT: I can download it locally, so it must be an error in the way credentials are introduced in the notebook, if anyone can shed light on this I would still be very grateful :)