Update README.md
Browse files
README.md
CHANGED
|
@@ -104,43 +104,59 @@ The model shows some promise in understanding context related to simple requests
|
|
| 104 |
<table style="text-align: left;">
|
| 105 |
<tr>
|
| 106 |
<th>Benchmark</th>
|
| 107 |
-
<th>5-shot</th>
|
| 108 |
-
<th>
|
|
|
|
|
|
|
| 109 |
</tr>
|
| 110 |
<tr>
|
| 111 |
<td>ARC Easy</td>
|
| 112 |
<td>47.26</td>
|
|
|
|
| 113 |
<td>45.58</td>
|
|
|
|
| 114 |
</tr>
|
| 115 |
<tr>
|
| 116 |
<td>BoolQ</td>
|
| 117 |
<td>43.33</td>
|
|
|
|
| 118 |
<td>44.16</td>
|
|
|
|
| 119 |
</tr>
|
| 120 |
<tr>
|
| 121 |
<td>HellaSWAG</td>
|
| 122 |
<td>28.70</td>
|
|
|
|
| 123 |
<td>28.72</td>
|
|
|
|
| 124 |
</tr>
|
| 125 |
<tr>
|
| 126 |
<td>MMLU</td>
|
| 127 |
-
<td
|
|
|
|
| 128 |
<td>25.28</td>
|
|
|
|
| 129 |
</tr>
|
| 130 |
<tr>
|
| 131 |
<td>OpenBookQA</td>
|
| 132 |
<td>16.00</td>
|
|
|
|
| 133 |
<td>18.20</td>
|
|
|
|
| 134 |
</tr>
|
| 135 |
<tr>
|
| 136 |
<td>PIQA</td>
|
| 137 |
<td>62.79</td>
|
|
|
|
| 138 |
<td>62.02</td>
|
|
|
|
| 139 |
</tr>
|
| 140 |
<tr>
|
| 141 |
<td>Winogrande</td>
|
| 142 |
<td>51.30</td>
|
|
|
|
| 143 |
<td>51.78</td>
|
|
|
|
| 144 |
</tr>
|
| 145 |
</table>
|
| 146 |
|
|
|
|
| 104 |
<table style="text-align: left;">
|
| 105 |
<tr>
|
| 106 |
<th>Benchmark</th>
|
| 107 |
+
<th>5-shot (acc)</th>
|
| 108 |
+
<th>5-shot (acc_norm)</th>
|
| 109 |
+
<th>0-shot (acc)</th>
|
| 110 |
+
<th>0-shot (acc_norm)</th>
|
| 111 |
</tr>
|
| 112 |
<tr>
|
| 113 |
<td>ARC Easy</td>
|
| 114 |
<td>47.26</td>
|
| 115 |
+
<td>46.30</td>
|
| 116 |
<td>45.58</td>
|
| 117 |
+
<td>39.52</td>
|
| 118 |
</tr>
|
| 119 |
<tr>
|
| 120 |
<td>BoolQ</td>
|
| 121 |
<td>43.33</td>
|
| 122 |
+
<td>-</td>
|
| 123 |
<td>44.16</td>
|
| 124 |
+
<td>-</td>
|
| 125 |
</tr>
|
| 126 |
<tr>
|
| 127 |
<td>HellaSWAG</td>
|
| 128 |
<td>28.70</td>
|
| 129 |
+
<td>30.73</td>
|
| 130 |
<td>28.72</td>
|
| 131 |
+
<td>31.01</td>
|
| 132 |
</tr>
|
| 133 |
<tr>
|
| 134 |
<td>MMLU</td>
|
| 135 |
+
<td>-</td>
|
| 136 |
+
<td>-</td>
|
| 137 |
<td>25.28</td>
|
| 138 |
+
<td>-</td>
|
| 139 |
</tr>
|
| 140 |
<tr>
|
| 141 |
<td>OpenBookQA</td>
|
| 142 |
<td>16.00</td>
|
| 143 |
+
<td>28.00</td>
|
| 144 |
<td>18.20</td>
|
| 145 |
+
<td>28.40</td>
|
| 146 |
</tr>
|
| 147 |
<tr>
|
| 148 |
<td>PIQA</td>
|
| 149 |
<td>62.79</td>
|
| 150 |
+
<td>62.68</td>
|
| 151 |
<td>62.02</td>
|
| 152 |
+
<td>60.23</td>
|
| 153 |
</tr>
|
| 154 |
<tr>
|
| 155 |
<td>Winogrande</td>
|
| 156 |
<td>51.30</td>
|
| 157 |
+
<td>-</td>
|
| 158 |
<td>51.78</td>
|
| 159 |
+
<td>-</td>
|
| 160 |
</tr>
|
| 161 |
</table>
|
| 162 |
|