Update README.md

README.md

@@ -32,7 +32,7 @@ base_model: meta-llama/Meta-Llama-3.1-70B-Instruct
 - **Model Developers:** Neural Magic
 
 Quantized version of [Meta-Llama-3.1-70B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3.1-70B-Instruct).
-It achieves scores within 1% of the scores of the unquantized model for MMLU, ARC-Challenge, GSM-8k, Hellaswag and Winogrande, and within 3.2% for TruthfulQA.
+It achieves scores within 1.4% of the scores of the unquantized model for MMLU, ARC-Challenge, GSM-8k, Hellaswag, Winogrande, and TruthfulQA.
 
 ### Model Optimizations
 
@@ -196,9 +196,9 @@ This version of the lm-evaluation-harness includes versions of MMLU, ARC-Challen
    </td>
    <td>86.66
    </td>
-   <td>86.06
+   <td>86.25
    </td>
-   <td>99.3%
+   <td>99.5%
    </td>
   </tr>
   <tr>
@@ -206,9 +206,9 @@ This version of the lm-evaluation-harness includes versions of MMLU, ARC-Challen
    </td>
    <td>85.32
    </td>
-   <td>85.16
+   <td>85.48
    </td>
-   <td>99.8%
+   <td>100.2%
    </td>
   </tr>
   <tr>
@@ -216,9 +216,9 @@ This version of the lm-evaluation-harness includes versions of MMLU, ARC-Challen
    </td>
    <td>60.65
    </td>
-   <td>58.74
+   <td>59.82
    </td>
-   <td>96.8%
+   <td>98.6%
    </td>
   </tr>
   <tr>
@@ -226,9 +226,9 @@ This version of the lm-evaluation-harness includes versions of MMLU, ARC-Challen
    </td>
    <td><strong>84.50</strong>
    </td>
-   <td><strong>83.76</strong>
+   <td><strong>83.98</strong>
    </td>
-   <td><strong>99.1%</strong>
+   <td><strong>99.4%</strong>
    </td>
   </tr>
 </table>