Hugging Face's logo

tiiuae
/
Falcon3-10B-Instruct-GPTQ-Int8

Safetensors
llama
falcon3
8-bit precision
gptq
Model card Files Community
1

Update README.md

#1
by karnakar - opened
base: refs/heads/main
←
from: refs/pr/1
Discussion Files changed
+9
-9
Files changed (1) hide show
  1. README.md +9 -9
README.md CHANGED
@@ -112,24 +112,24 @@ We report in the following table our internal pipeline benchmarks:
112
  <tbody>
113
  <tr>
114
  <td>MMLU</td>
115
- <td>71.99</td>
116
- <td>69.66</td>
117
  <td>71.1</td>
118
- <td>71.81</td>
119
  </tr>
120
  <tr>
121
  <td>MMLU-PRO</td>
122
- <td>44.16</td>
123
- <td>41.74</td>
124
  <td>43.4</td>
125
- <td>44.26</td>
126
  </tr>
127
  <tr>
128
  <td>IFEval</td>
129
- <td>78.9</td>
130
- <td>77.03</td>
131
  <td>76.8</td>
132
- <td>78.26</td>
133
  </tr>
134
  </tbody>
135
  </table>
 
112
  <tbody>
113
  <tr>
114
  <td>MMLU</td>
115
+ <td>72.0</td>
116
+ <td>69.7</td>
117
  <td>71.1</td>
118
+ <td>71.9</td>
119
  </tr>
120
  <tr>
121
  <td>MMLU-PRO</td>
122
+ <td>44.2</td>
123
+ <td>41.8</td>
124
  <td>43.4</td>
125
+ <td>44.3</td>
126
  </tr>
127
  <tr>
128
  <td>IFEval</td>
129
+ <td>79.0</td>
130
+ <td>77.1</td>
131
  <td>76.8</td>
132
+ <td>78.3</td>
133
  </tr>
134
  </tbody>
135
  </table>