Update README.md

README.md

@@ -23,7 +23,7 @@ license_link: https://huggingface.co/microsoft/Phi-3-mini-128k-instruct/resolve/
 - **Model Developers:** Neural Magic
 
 Quantized version of [Phi-3-mini-128k-instruct](https://huggingface.co/microsoft/Phi-3-mini-128k-instruct), with the new configuration files.
-It achieves an average score of 69.42 on the [OpenLLM](https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard) benchmark (version 1), whereas the unquantized model achieves 69.69.
+It achieves an average score of 69.45 on the [OpenLLM](https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard) benchmark (version 1), whereas the unquantized model achieves 69.69.
 
 ### Model Optimizations
 
@@ -187,9 +187,9 @@ lm_eval \
    </td>
    <td>69.33
    </td>
-   <td>68.90
+   <td>68.72
    </td>
-   <td>99.38%
+   <td>99.12%
    </td>
   </tr>
   <tr>
@@ -197,9 +197,9 @@ lm_eval \
    </td>
    <td>63.05
    </td>
-   <td>63.05
+   <td>62.54
    </td>
-   <td>100.0%
+   <td>99.19%
    </td>
   </tr>
   <tr>
@@ -207,9 +207,9 @@ lm_eval \
    </td>
    <td>76.95
    </td>
-   <td>76.27
+   <td>77.03
    </td>
-   <td>99.12%
+   <td>100.1%
    </td>
   </tr>
   <tr>
@@ -217,9 +217,9 @@ lm_eval \
    </td>
    <td>79.58
    </td>
-   <td>79.36
+   <td>79.37
    </td>
-   <td>99.72%
+   <td>99.74%
    </td>
   </tr>
   <tr>
@@ -227,9 +227,9 @@ lm_eval \
    </td>
    <td>74.82
    </td>
-   <td>74.59
+   <td>75.37
    </td>
-   <td>99.69%
+   <td>100.7%
    </td>
   </tr>
   <tr>
@@ -237,9 +237,9 @@ lm_eval \
    </td>
    <td>54.41
    </td>
-   <td>54.36
+   <td>53.68
    </td>
-   <td>99.91%
+   <td>98.66%
    </td>
   </tr>
   <tr>
@@ -247,9 +247,9 @@ lm_eval \
    </td>
    <td><strong>69.69</strong>
    </td>
-   <td><strong>69.42</strong>
+   <td><strong>69.45</strong>
    </td>
-   <td><strong>99.61%</strong>
+   <td><strong>99.66%</strong>
    </td>
   </tr>
 </table>