Update shieldgemma_on_huggingface.ipynb

[bebechien]

fix bug

Description of the change

preprocess_and_predict() will fail with "NameError: name 'model' is not defined"

Motivation

fix bug

Type of change

Choose one: (Bug fix)

Checklist

• I have performed a self-review of my code.
• I have added detailed comments to my code where applicable.
• I have verified that my change does not break existing code.
• My PR is based on the latest changes of the main branch (if unsure, please run git pull --rebase upstream main).
• I am familiar with the Google Style Guide for the language I have coded in.
• I have read through the Contributing Guide and signed the Contributor License Agreement.

[review-notebook-app]

Check out this pull request on 

See visual diffs & provide feedback on Jupyter Notebooks.

---

Powered by ReviewNB

[bebechien]

I think we can ignore the lint error, detected "we" is actually coming from the sample text in the notebook.

PROMPT_ONLY_TEMPLATE = """You are a policy expert trying to help determine
whether a user prompt is in violation of the defined safety policies.

<start_of_turn>
Human Question: {user_content}
<end_of_turn>

Our safety principle is defined in the below:

* {harm_text}

===

Does the human question violate the above principle? Your answer must start with
'Yes' or 'No'. And then walk through step by step to be sure we answer
correctly.

"""

[markmcd]

I think we can ignore the lint error

Yeah good call - we need to add a per-line or per-cell suppression mechanism to nblint but have not done so yet. It's becoming much more important with LLM generated output.