add evalaute question (2nd take) (#5)

- add evalaute question (2nd take) (35a196ef99a6cc19ae7ceacf4e58a493443f3469)


Co-authored-by: Leandro von Werra <lvwerra@users.noreply.huggingface.co>

app.py

@@ -78,13 +78,23 @@ References
 - Codex paper: https://arxiv.org/abs/2107.03374
 """
 
+evaluate_question = """
+Use the `evaluate` library to compute the BLEU score of the model generation `"Evaluate is a library to evaluate Machine Learning models"` and the reference solution `"Evaluate is a library to evaluate ML models"`. Round the result to two digits after the comma.
+<br/>
+<br/>
+References
+<br/>
+- `evaluate` library: https://huggingface.co/docs/evaluate/index
+- BLEU score: https://en.wikipedia.org/wiki/BLEU
+"""
+
 internships = {
     'Accelerate': default_question,
     'Diffusion distillation': default_question,
     'Skops & Scikit-Learn': skops_question,
     "Code Generation": code_question,
     "Document AI Democratization": default_question,
-    "Evaluate": default_question,
+    "Evaluate": evaluate_question,
     "ASR": default_question,
     "Efficient video pretraining": default_question,
     "Embodied AI": default_question,