eval config cleanup

Signed-off-by: Keval Morabia <28916987+kevalmorabia97@users.noreply.github.com>

Author: kevalmorabia97

examples/pruning/minitron/NVIDIA-Nemotron-Nano-9B-v2/README.md

@@ -225,7 +225,7 @@ python /opt/Megatron-Bridge/examples/conversion/convert_checkpoints.py export \
 
 ### 4. Evaluation
 
-The eval config in [nemo_evaluator.yaml](nemo_evaluator.yaml) is for Slurm-based evaluation — it submits a vLLM serving job and runs evals against it. For local model execution and evaluation, refer to the [NeMo Evaluator documentation](https://docs.nvidia.com/nemo/evaluator/latest/).
+The eval config xin [nemo_evaluator.yaml](nemo_evaluator.yaml) is for Slurm-based evaluation — it submits a vLLM serving job and runs evals against it. For local model execution and evaluation, refer to the [NeMo Evaluator documentation](https://docs.nvidia.com/nemo/evaluator/latest/) or this [blog](https://huggingface.co/blog/nvidia/nemotron-3-nano-evaluation-recipe).
 
 Before running, update the following fields in the yaml:
 

examples/pruning/minitron/NVIDIA-Nemotron-Nano-9B-v2/nemo_evaluator.yaml

@@ -160,16 +160,17 @@ evaluation:
             extra:
               num_repeats: 64
 
-    # 5. MATH500
-    - name: AA_math_test_500
-      env_vars:
-        HF_TOKEN: HF_TOKEN
-      nemo_evaluator_config:
-        config:
-          params:
-            # limit_samples: 8
-            extra:
-              n_samples: 5
+    # 5. MATH500 (Requires JUDGE_API_KEY)
+    # - name: AA_math_test_500
+    #   env_vars:
+    #     HF_TOKEN: HF_TOKEN
+    #     JUDGE_API_KEY: JUDGE_API_KEY
+    #   nemo_evaluator_config:
+    #     config:
+    #       params:
+    #         # limit_samples: 8
+    #         extra:
+    #           n_samples: 5
 
     # 6. IFEval
     - name: ns_ifeval