Safetensors
llama
falcon3
4-bit precision
gptq
Files changed (1) hide show
  1. README.md +12 -12
README.md CHANGED
@@ -115,24 +115,24 @@ We report in the following table our internal pipeline benchmarks:
115
  <tbody>
116
  <tr>
117
  <td>MMLU</td>
118
- <td>67.68</td>
119
- <td>67.55</td>
120
- <td>66.33</td>
121
- <td>65.55</td>
122
  </tr>
123
  <tr>
124
  <td>MMLU-PRO</td>
125
- <td>40.86</td>
126
- <td>40.85</td>
127
- <td>39.82</td>
128
- <td>39.07</td>
129
  </tr>
130
  <tr>
131
  <td>IFEval</td>
132
- <td>75.06</td>
133
- <td>76.92</td>
134
- <td>74.76</td>
135
- <td>72.14</td>
136
  </tr>
137
  </tbody>
138
  </table>
 
115
  <tbody>
116
  <tr>
117
  <td>MMLU</td>
118
+ <td>67.7</td>
119
+ <td>67.6</td>
120
+ <td>66.4</td>
121
+ <td>65.6</td>
122
  </tr>
123
  <tr>
124
  <td>MMLU-PRO</td>
125
+ <td>40.9</td>
126
+ <td>40.9</td>
127
+ <td>39.9</td>
128
+ <td>39.1</td>
129
  </tr>
130
  <tr>
131
  <td>IFEval</td>
132
+ <td>75.1</td>
133
+ <td>77.0</td>
134
+ <td>74.8</td>
135
+ <td>72.2</td>
136
  </tr>
137
  </tbody>
138
  </table>