Skip to content

PTQ-Level Efficiency, QAT-Level Accuracy | 训练后量化的效率,量化感知训练的精度

update readme

wenhuach210134653fe880
diff --git a/README.md b/README.md
index b2aff5184..3382ea9db 100644
--- a/README.md
+++ b/README.md
@@ -289,7 +289,7 @@ For wikitext2/ptb-new/c4-new ppl, we follow the code of gptq and set the sequenc
</tr>
<tr>
- <th>AutoRound</th>
+ <th>Ours</th>
<td>61.67</td>
<td>54.57</td>
<td>61.32</td>
@@ -335,7 +335,7 @@ We provide a [comprehensive analysis](docs/README.md) with other methods in our
## Tips
1 Consider increasing tuning steps to achieve better results, albeit with increased tuning time.
-2 Leverage AutoGPTQ to evaluate the model on GPU
+2 Leverage AutoGPTQ to run the model on GPU
```python
from transformers import AutoModelForCausalLM, AutoTokenizer