Update README.md

README.md

@@ -97,6 +97,17 @@ The dataset is comprised of a mixture of open datasets large-scale datasets avai
 | GPT-4 |  -| RLHF |8.99| 95.28|
 
 ## Other benchmark:
+| Metric                | Value                     |
+|-----------------------|---------------------------|
+| ARC (25-shot)         |          |
+| HellaSwag (10-shot)   |     |
+| MMLU (5-shot)        |        |
+| TruthfulQA (0-shot)   |    |
+| Winogrande (5-shot)   |    |
+| GSM8K (5-shot)        | 0.423        |
+| Human Eval            |   |
+
+
 - BigBench: 0.3526.