Automated MNLP evaluation report (2026-05-20)

EVAL_REPORT.md

@@ -0,0 +1,53 @@
+# Automated MNLP evaluation report
+
+- **Model repo:** [`cs-552-2026-barn/general_knowledge_model`](https://huggingface.co/cs-552-2026-barn/general_knowledge_model)
+- **Owner(s):** group **barn**
+- **Generated at:** 2026-05-20T05:49:46+00:00 (UTC)
+- **Pipeline:** [mnlp-project-ci](https://github.com/eric11eca/mnlp-project-ci)
+
+_This PR is opened automatically by the course CI. It is **non-blocking** — you do not need to merge it. The next nightly run will refresh this file._
+
+## Summary
+
+| Benchmark | Accuracy | Status |
+|---|---:|---|
+| Math | — | not run |
+| Knowledge | 0.2800 | ok |
+| Multilingual | — | not run |
+| Safety | — | not run |
+
+## Sample completions
+
+_Prompts are intentionally omitted to avoid revealing benchmark contents. For multi-completion problems, only one completion is shown per sample._
+
+### Knowledge
+
+**Correct** (1 shown)
+
+- **reference**: `A`
+- **overall** (1/1 completions correct)
+- **extracted** (✓): `A`
+- **completion**:
+
+    ```text
+    <think>
+    
+    </think>
+    
+    \boxed{A}
+    ```
+
+**Incorrect** (1 shown)
+
+- **reference**: `B`
+- **overall** (0/1 completions correct)
+- **extracted** (✗): `C`
+- **completion**:
+
+    ```text
+    <think>
+    
+    </think>
+    
+    \boxed{C}
+    ```