Add reasoning evals
Browse files
README.md
CHANGED
|
@@ -158,6 +158,31 @@ lm_eval \
|
|
| 158 |
</thead>
|
| 159 |
<tbody>
|
| 160 |
<tr>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 161 |
<td rowspan="7"><b>OpenLLM V1</b></td>
|
| 162 |
<td>ARC-Challenge (Acc-Norm, 25-shot)</td>
|
| 163 |
<td>63.65</td>
|
|
|
|
| 158 |
</thead>
|
| 159 |
<tbody>
|
| 160 |
<tr>
|
| 161 |
+
<td rowspan="4"><b>Reasoning</b></td>
|
| 162 |
+
<td>AIME 2024 (pass@1)</td>
|
| 163 |
+
<td>67.83</td>
|
| 164 |
+
<td>69.17</td>
|
| 165 |
+
<td>101.98%</td>
|
| 166 |
+
</tr>
|
| 167 |
+
<tr>
|
| 168 |
+
<td>MATH-500 (pass@1)</td>
|
| 169 |
+
<td>95.29</td>
|
| 170 |
+
<td>95.14</td>
|
| 171 |
+
<td>99.84%</td>
|
| 172 |
+
</tr>
|
| 173 |
+
<tr>
|
| 174 |
+
<td>GPQA Diamond (pass@1)</td>
|
| 175 |
+
<td>65.57</td>
|
| 176 |
+
<td>65.15</td>
|
| 177 |
+
<td>99.36%</td>
|
| 178 |
+
</tr>
|
| 179 |
+
<tr>
|
| 180 |
+
<td><b>Average Score</b></td>
|
| 181 |
+
<td><b>76.23</b></td>
|
| 182 |
+
<td><b>76.49</b></td>
|
| 183 |
+
<td><b>100.34%</b></td>
|
| 184 |
+
</tr>
|
| 185 |
+
<tr>
|
| 186 |
<td rowspan="7"><b>OpenLLM V1</b></td>
|
| 187 |
<td>ARC-Challenge (Acc-Norm, 25-shot)</td>
|
| 188 |
<td>63.65</td>
|