nm-research commited on
Commit
c61000a
·
verified ·
1 Parent(s): 361c428

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +21 -21
README.md CHANGED
@@ -217,45 +217,45 @@ lm_eval \
217
  <tr>
218
  <td rowspan="7"><b>OpenLLM V2</b></td>
219
  <td>IFEval (Inst Level Strict Acc, 0-shot)</td>
220
- <td>43.15</td>
221
- <td>42.08</td>
222
- <td>97.5%</td>
223
  </tr>
224
  <tr>
225
  <td>BBH (Acc-Norm, 3-shot)</td>
226
- <td>64.32</td>
227
- <td>63.91</td>
228
- <td>99.4%</td>
229
  </tr>
230
  <tr>
231
  <td>Math-Hard (Exact-Match, 4-shot)</td>
232
- <td>35.04</td>
233
- <td>37.81</td>
234
- <td>107.9%</td>
235
  </tr>
236
  <tr>
237
  <td>GPQA (Acc-Norm, 0-shot)</td>
238
- <td>37.15</td>
239
- <td>36.64</td>
240
- <td>98.6%</td>
241
  </tr>
242
  <tr>
243
  <td>MUSR (Acc-Norm, 0-shot)</td>
244
- <td>42.89</td>
245
- <td>42.49</td>
246
- <td>99.1%</td>
247
  </tr>
248
  <tr>
249
  <td>MMLU-Pro (Acc, 5-shot)</td>
250
- <td>47.22</td>
251
- <td>45.78</td>
252
- <td>96.9%</td>
253
  </tr>
254
  <tr>
255
  <td><b>Average Score</b></td>
256
- <td><b>44.96</b></td>
257
- <td><b>44.78</b></td>
258
- <td><b>99.6%</b></td>
259
  </tr>
260
  <tr>
261
  <td rowspan="4"><b>Coding</b></td>
 
217
  <tr>
218
  <td rowspan="7"><b>OpenLLM V2</b></td>
219
  <td>IFEval (Inst Level Strict Acc, 0-shot)</td>
220
+ <td>42.45</td>
221
+ <td>42.41</td>
222
+ <td>99.9%</td>
223
  </tr>
224
  <tr>
225
  <td>BBH (Acc-Norm, 3-shot)</td>
226
+ <td>21.26</td>
227
+ <td>16.96</td>
228
+ <td>79.8%</td>
229
  </tr>
230
  <tr>
231
  <td>Math-Hard (Exact-Match, 4-shot)</td>
232
+ <td>0.00</td>
233
+ <td>0.00</td>
234
+ <td>---</td>
235
  </tr>
236
  <tr>
237
  <td>GPQA (Acc-Norm, 0-shot)</td>
238
+ <td>9.51</td>
239
+ <td>6.68</td>
240
+ <td>---</td>
241
  </tr>
242
  <tr>
243
  <td>MUSR (Acc-Norm, 0-shot)</td>
244
+ <td>14.87</td>
245
+ <td>12.91</td>
246
+ <td>---</td>
247
  </tr>
248
  <tr>
249
  <td>MMLU-Pro (Acc, 5-shot)</td>
250
+ <td>4.27</td>
251
+ <td>2.38</td>
252
+ <td>---</td>
253
  </tr>
254
  <tr>
255
  <td><b>Average Score</b></td>
256
+ <td><b>15.39</b></td>
257
+ <td><b>13.56</b></td>
258
+ <td><b>88.1%</b></td>
259
  </tr>
260
  <tr>
261
  <td rowspan="4"><b>Coding</b></td>