nm-research commited on
Commit
d167aa1
·
verified ·
1 Parent(s): d9dc71c

Add reasoning evals

Browse files
Files changed (1) hide show
  1. README.md +25 -0
README.md CHANGED
@@ -167,6 +167,31 @@ lm_eval \
167
  </tr>
168
  </thead>
169
  <tbody>
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
170
  <tr>
171
  <td rowspan="7"><b>OpenLLM V1</b></td>
172
  <td>ARC-Challenge (Acc-Norm, 25-shot)</td>
 
167
  </tr>
168
  </thead>
169
  <tbody>
170
+ <tr>
171
+ <td rowspan="4"><b>Reasoning</b></td>
172
+ <td>AIME 2024 (pass@1)</td>
173
+ <td>49.25</td>
174
+ <td>49.07</td>
175
+ <td>99.63%</td>
176
+ </tr>
177
+ <tr>
178
+ <td>MATH-500 (pass@1)</td>
179
+ <td>90.18</td>
180
+ <td>90.04</td>
181
+ <td>99.84%</td>
182
+ </tr>
183
+ <tr>
184
+ <td>GPQA Diamond (pass@1)</td>
185
+ <td>49.27</td>
186
+ <td>48.9</td>
187
+ <td>98.25%</td>
188
+ </tr>
189
+ <tr>
190
+ <td><b>Average Score</b></td>
191
+ <td><b>62.9</b></td>
192
+ <td><b>62.67</b></td>
193
+ <td><b>99.63%</b></td>
194
+ </tr>
195
  <tr>
196
  <td rowspan="7"><b>OpenLLM V1</b></td>
197
  <td>ARC-Challenge (Acc-Norm, 25-shot)</td>