nm-research commited on
Commit
d9dc71c
·
verified ·
1 Parent(s): f77821f

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +18 -18
README.md CHANGED
@@ -213,15 +213,15 @@ lm_eval \
213
  <tr>
214
  <td rowspan="7"><b>OpenLLM V2</b></td>
215
  <td>IFEval (Inst Level Strict Acc, 0-shot)</td>
216
- <td>38.34</td>
217
- <td>39.07</td>
218
- <td>101.9%</td>
219
  </tr>
220
  <tr>
221
  <td>BBH (Acc-Norm, 3-shot)</td>
222
- <td>38.19</td>
223
- <td>39.57</td>
224
- <td>103.6%</td>
225
  </tr>
226
  <tr>
227
  <td>Math-Hard (Exact-Match, 4-shot)</td>
@@ -231,27 +231,27 @@ lm_eval \
231
  </tr>
232
  <tr>
233
  <td>GPQA (Acc-Norm, 0-shot)</td>
234
- <td>28.87</td>
235
- <td>27.28</td>
236
- <td>94.5%</td>
237
  </tr>
238
  <tr>
239
  <td>MUSR (Acc-Norm, 0-shot)</td>
240
- <td>33.31</td>
241
- <td>34.50</td>
242
- <td>103.6%</td>
243
  </tr>
244
  <tr>
245
  <td>MMLU-Pro (Acc, 5-shot)</td>
246
- <td>20.10</td>
247
- <td>20.60</td>
248
- <td>102.4%</td>
249
  </tr>
250
  <tr>
251
  <td><b>Average Score</b></td>
252
- <td><b>26.47</b></td>
253
- <td><b>26.84</b></td>
254
- <td><b>101.4%</b></td>
255
  </tr>
256
  <tr>
257
  <td rowspan="4"><b>Coding</b></td>
 
213
  <tr>
214
  <td rowspan="7"><b>OpenLLM V2</b></td>
215
  <td>IFEval (Inst Level Strict Acc, 0-shot)</td>
216
+ <td>38.37</td>
217
+ <td>38.40</td>
218
+ <td>100.1%</td>
219
  </tr>
220
  <tr>
221
  <td>BBH (Acc-Norm, 3-shot)</td>
222
+ <td>7.43</td>
223
+ <td>7.66</td>
224
+ <td>---</td>
225
  </tr>
226
  <tr>
227
  <td>Math-Hard (Exact-Match, 4-shot)</td>
 
231
  </tr>
232
  <tr>
233
  <td>GPQA (Acc-Norm, 0-shot)</td>
234
+ <td>1.51</td>
235
+ <td>1.47</td>
236
+ <td>---</td>
237
  </tr>
238
  <tr>
239
  <td>MUSR (Acc-Norm, 0-shot)</td>
240
+ <td>1.86</td>
241
+ <td>1.27</td>
242
+ <td>---</td>
243
  </tr>
244
  <tr>
245
  <td>MMLU-Pro (Acc, 5-shot)</td>
246
+ <td>1.61</td>
247
+ <td>1.48</td>
248
+ <td>---</td>
249
  </tr>
250
  <tr>
251
  <td><b>Average Score</b></td>
252
+ <td><b>8.47</b></td>
253
+ <td><b>8.38</b></td>
254
+ <td><b>---</b></td>
255
  </tr>
256
  <tr>
257
  <td rowspan="4"><b>Coding</b></td>