update README.md

README.md

@@ -12,6 +12,19 @@ thumbnail: >-
 license: apache-2.0
 ---
 # Model Card
+## Github
+https://github.com/chi2liu/mamba-gpt-3b
+
+| Metric                | Value |
+|-----------------------|-------|
+| MMLU (5-shot)         | 25.3  |
+| ARC (25-shot)         | 40.5  |
+| HellaSwag (10-shot)   | 64.9  |
+| TruthfulQA (0-shot)   | 37.1  |
+| Avg.                  | 42.0  | 
+
+We use state-of-the-art [Language Model Evaluation Harness](https://github.com/EleutherAI/lm-evaluation-harness) to run the benchmark tests above.
+
 ## Summary
 
 We have fine-tuned the open-lama model and surpassed the original model in multiple evaluation subtasks, making it currently the best performing 3B model with comparable performance to llama-7b