Update README.md

README.md

@@ -10,7 +10,7 @@ language:
 # Mistral-Nemo-Instruct-2407-FP8
 
 ## Model Overview
-- **Model Architecture:** Meta-Llama-3
+- **Model Architecture:** Mistral-Nemo
   - **Input:** Text
   - **Output:** Text
 - **Model Optimizations:**
@@ -24,7 +24,7 @@ language:
 - **Model Developers:** Neural Magic
 
 Quantized version of [Mistral-Nemo-Instruct-2407](https://huggingface.co/mistralai/Mistral-Nemo-Instruct-2407).
-It achieves an average score of 68.22 on the [OpenLLM](https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard) benchmark (version 1), whereas the unquantized model achieves 68.71.
+It achieves an average score of 71.28 on the [OpenLLM](https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard) benchmark (version 1), whereas the unquantized model achieves 71.61.
 
 ### Model Optimizations
 
@@ -78,7 +78,7 @@ from transformers import AutoTokenizer
 
 from auto_fp8 import AutoFP8ForCausalLM, BaseQuantizeConfig
 
-pretrained_model_dir = "meta-llama/Mistral-Nemo-Instruct-2407"
+pretrained_model_dir = "mistralai/Mistral-Nemo-Instruct-2407"
 quantized_model_dir = "Mistral-Nemo-Instruct-2407-FP8"
 
 tokenizer = AutoTokenizer.from_pretrained(pretrained_model_dir, use_fast=True, model_max_length=4096)
@@ -130,71 +130,71 @@ lm_eval \
   <tr>
    <td>MMLU (5-shot)
    </td>
-   <td>66.60
+   <td>68.35
    </td>
-   <td>66.27
+   <td>68.50
    </td>
-   <td>99.50%
+   <td>100.2%
    </td>
   </tr>
   <tr>
    <td>ARC Challenge (25-shot)
    </td>
-   <td>62.54
+   <td>65.53
    </td>
-   <td>61.77
+   <td>64.68
    </td>
-   <td>98.76%
+   <td>98.70%
    </td>
   </tr>
   <tr>
    <td>GSM-8K (5-shot, strict-match)
    </td>
-   <td>75.96
+   <td>74.45
    </td>
-   <td>73.99
+   <td>73.01
    </td>
-   <td>97.40%
+   <td>98.06%
    </td>
   </tr>
   <tr>
    <td>Hellaswag (10-shot)
    </td>
-   <td>78.83
+   <td>84.32
    </td>
-   <td>78.56
+   <td>84.18
    </td>
-   <td>99.65%
+   <td>99.83%
    </td>
   </tr>
   <tr>
    <td>Winogrande (5-shot)
    </td>
-   <td>75.93
+   <td>82.16
    </td>
-   <td>76.40
+   <td>82.32
    </td>
-   <td>100.6%
+   <td>100.1%
    </td>
   </tr>
   <tr>
    <td>TruthfulQA (0-shot)
    </td>
-   <td>52.44
+   <td>54.85
    </td>
-   <td>52.35
+   <td>54.96
    </td>
-   <td>99.82%
+   <td>100.2%
    </td>
   </tr>
   <tr>
    <td><strong>Average</strong>
    </td>
-   <td><strong>68.71</strong>
+   <td><strong>71.61</strong>
    </td>
-   <td><strong>68.22</strong>
+   <td><strong>71.28</strong>
    </td>
-   <td><strong>99.28%</strong>
+   <td><strong>99.53%</strong>
    </td>
   </tr>
 </table>