nm-research commited on
Commit
1b056b2
·
verified ·
1 Parent(s): b975617

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +17 -17
README.md CHANGED
@@ -194,15 +194,15 @@ lm_eval \
194
  <tr>
195
  <td rowspan="7"><b>OpenLLM V2</b></td>
196
  <td>IFEval (Inst Level Strict Acc, 0-shot)</td>
197
- <td>34.37</td>
198
  <td>34.91</td>
199
- <td>101.6%</td>
200
  </tr>
201
  <tr>
202
  <td>BBH (Acc-Norm, 3-shot)</td>
203
- <td>34.44</td>
204
- <td>34.40</td>
205
- <td>99.9%</td>
206
  </tr>
207
  <tr>
208
  <td>Math-Hard (Exact-Match, 4-shot)</td>
@@ -212,27 +212,27 @@ lm_eval \
212
  </tr>
213
  <tr>
214
  <td>GPQA (Acc-Norm, 0-shot)</td>
215
- <td>24.67</td>
216
- <td>25.16</td>
217
- <td>102.0%</td>
218
  </tr>
219
  <tr>
220
  <td>MUSR (Acc-Norm, 0-shot)</td>
221
- <td>35.82</td>
222
- <td>36.61</td>
223
- <td>102.2%</td>
224
  </tr>
225
  <tr>
226
  <td>MMLU-Pro (Acc, 5-shot)</td>
227
- <td>11.80</td>
228
- <td>11.69</td>
229
- <td>99.1%</td>
230
  </tr>
231
  <tr>
232
  <td><b>Average Score</b></td>
233
- <td><b>23.52</b></td>
234
- <td><b>23.79</b></td>
235
- <td><b>101.2%</b></td>
236
  </tr>
237
  <tr>
238
  <td rowspan="4"><b>Coding</b></td>
 
194
  <tr>
195
  <td rowspan="7"><b>OpenLLM V2</b></td>
196
  <td>IFEval (Inst Level Strict Acc, 0-shot)</td>
197
+ <td>34.63</td>
198
  <td>34.91</td>
199
+ <td>100.8%</td>
200
  </tr>
201
  <tr>
202
  <td>BBH (Acc-Norm, 3-shot)</td>
203
+ <td>3.06</td>
204
+ <td>2.40</td>
205
+ <td>---</td>
206
  </tr>
207
  <tr>
208
  <td>Math-Hard (Exact-Match, 4-shot)</td>
 
212
  </tr>
213
  <tr>
214
  <td>GPQA (Acc-Norm, 0-shot)</td>
215
+ <td>1.01</td>
216
+ <td>0.93</td>
217
+ <td>---</td>
218
  </tr>
219
  <tr>
220
  <td>MUSR (Acc-Norm, 0-shot)</td>
221
+ <td>0.78</td>
222
+ <td>1.26</td>
223
+ <td>---</td>
224
  </tr>
225
  <tr>
226
  <td>MMLU-Pro (Acc, 5-shot)</td>
227
+ <td>1.32</td>
228
+ <td>1.25</td>
229
+ <td>---</td>
230
  </tr>
231
  <tr>
232
  <td><b>Average Score</b></td>
233
+ <td><b>6.80</b></td>
234
+ <td><b>6.79</b></td>
235
+ <td><b>---</b></td>
236
  </tr>
237
  <tr>
238
  <td rowspan="4"><b>Coding</b></td>