Add reasoning evals
Browse files
README.md
CHANGED
|
@@ -164,6 +164,31 @@ lm_eval \
|
|
| 164 |
</thead>
|
| 165 |
<tbody>
|
| 166 |
<tr>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 167 |
<td rowspan="7"><b>OpenLLM V1</b></td>
|
| 168 |
<td>ARC-Challenge (Acc-Norm, 25-shot)</td>
|
| 169 |
<td>64.59</td>
|
|
|
|
| 164 |
</thead>
|
| 165 |
<tbody>
|
| 166 |
<tr>
|
| 167 |
+
<td rowspan="4"><b>Reasoning</b></td>
|
| 168 |
+
<td>AIME 2024 (pass@1)</td>
|
| 169 |
+
<td>69.75</td>
|
| 170 |
+
<td>68.83</td>
|
| 171 |
+
<td>98.68%</td>
|
| 172 |
+
</tr>
|
| 173 |
+
<tr>
|
| 174 |
+
<td>MATH-500 (pass@1)</td>
|
| 175 |
+
<td>95.09</td>
|
| 176 |
+
<td>95.01</td>
|
| 177 |
+
<td>99.92%</td>
|
| 178 |
+
</tr>
|
| 179 |
+
<tr>
|
| 180 |
+
<td>GPQA Diamond (pass@1)</td>
|
| 181 |
+
<td>64.05</td>
|
| 182 |
+
<td>63.79</td>
|
| 183 |
+
<td>99.59%</td>
|
| 184 |
+
</tr>
|
| 185 |
+
<tr>
|
| 186 |
+
<td><b>Average Score</b></td>
|
| 187 |
+
<td><b>76.3</b></td>
|
| 188 |
+
<td><b>75.88</b></td>
|
| 189 |
+
<td><b>99.45%</b></td>
|
| 190 |
+
</tr>
|
| 191 |
+
<tr>
|
| 192 |
<td rowspan="7"><b>OpenLLM V1</b></td>
|
| 193 |
<td>ARC-Challenge (Acc-Norm, 25-shot)</td>
|
| 194 |
<td>64.59</td>
|