Update README.md
Browse files
README.md
CHANGED
|
@@ -159,6 +159,37 @@ This version of the lm-evaluation-harness includes versions of MMLU, ARC-Challen
|
|
| 159 |
<td><strong>Recovery</strong>
|
| 160 |
</td>
|
| 161 |
</tr>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 162 |
<tr>
|
| 163 |
<td>GSM-8K (CoT, 8-shot, strict-match)
|
| 164 |
</td>
|
|
|
|
| 159 |
<td><strong>Recovery</strong>
|
| 160 |
</td>
|
| 161 |
</tr>
|
| 162 |
+
<tr>
|
| 163 |
+
<td>MMLU (CoT, 0-shot)
|
| 164 |
+
</td>
|
| 165 |
+
<td>88.11
|
| 166 |
+
</td>
|
| 167 |
+
<td>88.23
|
| 168 |
+
</td>
|
| 169 |
+
<td>100.1%
|
| 170 |
+
</td>
|
| 171 |
+
</tr>
|
| 172 |
+
<tr>
|
| 173 |
+
<td>ARC Challenge (0-shot)
|
| 174 |
+
</td>
|
| 175 |
+
<td>94.97
|
| 176 |
+
</td>
|
| 177 |
+
<td>94.88
|
| 178 |
+
</td>
|
| 179 |
+
<td>99.9%
|
| 180 |
+
</td>
|
| 181 |
+
</tr>
|
| 182 |
+
<tr>
|
| 183 |
+
<tr>
|
| 184 |
+
<td>GSM-8K (CoT, 8-shot, strict-match)
|
| 185 |
+
</td>
|
| 186 |
+
<td>96.44
|
| 187 |
+
</td>
|
| 188 |
+
<td>96.13
|
| 189 |
+
</td>
|
| 190 |
+
<td>99.7%
|
| 191 |
+
</td>
|
| 192 |
+
</tr>
|
| 193 |
<tr>
|
| 194 |
<td>GSM-8K (CoT, 8-shot, strict-match)
|
| 195 |
</td>
|