Hugging Face's logo

neuralmagic
/
Meta-Llama-3.1-8B-Instruct-quantized.w8a8

Text Generation
Transformers
Safetensors
llama
int8
vllm
conversational
text-generation-inference
Inference Endpoints
8-bit precision
compressed-tensors
Model card Files Community
2
alexmarques commited on
Commit
44910a5
·
verified ·
1 Parent(s): e94a1df

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +8 -8
README.md CHANGED
@@ -150,9 +150,9 @@ This version of the lm-evaluation-harness includes versions of MMLU, ARC-Challen
150
  </td>
151
  <td>69.43
152
  </td>
153
- <td>69.37
154
  </td>
155
- <td>99.9%
156
  </td>
157
  </tr>
158
  <tr>
@@ -160,9 +160,9 @@ This version of the lm-evaluation-harness includes versions of MMLU, ARC-Challen
160
  </td>
161
  <td>72.56
162
  </td>
163
- <td>72.14
164
  </td>
165
- <td>99.4%
166
  </td>
167
  </tr>
168
  <tr>
@@ -170,9 +170,9 @@ This version of the lm-evaluation-harness includes versions of MMLU, ARC-Challen
170
  </td>
171
  <td>81.57
172
  </td>
173
- <td>81.48
174
  </td>
175
- <td>99.9%
176
  </td>
177
  </tr>
178
  <tr>
@@ -220,9 +220,9 @@ This version of the lm-evaluation-harness includes versions of MMLU, ARC-Challen
220
  </td>
221
  <td><strong>74.04</strong>
222
  </td>
223
- <td><strong>73.89</strong>
224
  </td>
225
- <td><strong>99.8%</strong>
226
  </td>
227
  </tr>
228
  </table>
 
150
  </td>
151
  <td>69.43
152
  </td>
153
+ <td>68.78
154
  </td>
155
+ <td>99.1%
156
  </td>
157
  </tr>
158
  <tr>
 
160
  </td>
161
  <td>72.56
162
  </td>
163
+ <td>72.20
164
  </td>
165
+ <td>99.5%
166
  </td>
167
  </tr>
168
  <tr>
 
170
  </td>
171
  <td>81.57
172
  </td>
173
+ <td>81.06
174
  </td>
175
+ <td>99.4%
176
  </td>
177
  </tr>
178
  <tr>
 
220
  </td>
221
  <td><strong>74.04</strong>
222
  </td>
223
+ <td><strong>73.59</strong>
224
  </td>
225
+ <td><strong>99.4%</strong>
226
  </td>
227
  </tr>
228
  </table>