minor fix to successfully run a single test case

[ViswanathaReddyGajjala]

I wanted to add a few test cases, but it took me some time to get it working for a sample test case. I hope this change helps others save some time.

I made a few changes to ensure one specific test works on Colab:

1. I copied the config.json.example file to create a new file named config.json and added the OpenAI key.
2. I updated the Python version in the code and included the information in the readme.
3. I tested these changes on Colab, and everything is working correctly.

[ViswanathaReddyGajjala]

Exact commands for reference

After making the changes mentioned above, I executed the following commands on Colab:

!git clone https://github.com/carlini/yet-another-applied-llm-benchmark.git

%cd yet-another-applied-llm-benchmark
!pip install -r requirements.txt
!pip install -r requirements-extra.txt

!PYTHONPATH='.' python tests/print_hello.py 

[carlini]

Do you want to add a subsection "Running from Colab" and include this information? It seems useful.

[ViswanathaReddyGajjala]

Thank you for the suggestion. Instead, I've created a Colab notebook, making it much easier to run. Feel free to share any further suggestions or improvements!