Update README.md

README.md

@@ -83,13 +83,13 @@ pip3 install lm-eval==0.4.2
 
 ```bash
 git clone https://github.com/intel/auto-round
-pip install -vvv --no-build-isolation -e .
-auto-round --model "Intel/Qwen2.5-0.5B-Instruct-int4-inc" --eval --eval_bs 16  --tasks lambada_openai,hellaswag,piqa,winogrande,truthfulqa_mc1,truthfulqa_mc2,openbookqa,boolq,arc_easy,arc_challenge,mmlu,gsm8k,cmmlu,ceval-valid
+cd auto-round
+python -m auto_round --model "Intel/Qwen2.5-0.5B-Instruct-int4-inc" --eval --eval_bs 16  --tasks lambada_openai,hellaswag,piqa,winogrande,truthfulqa_mc1,openbookqa,boolq,arc_easy,arc_challenge,mmlu,gsm8k,cmmlu,ceval-valid
 ```
 
 | Metric         | BF16   | INT4(group_size 128)   | INT4(group_size 32)   |
-| -------------- | ------ | ------ | ------ |
-| Avg            | 0.4604 | 0.4469 | 0.4478 |
+|:--------------:| :----: | :----: |:------:|
+| Avg            | 0.4637 | 0.4496 | 0.4512 |
 | mmlu           | 0.4587 | 0.4440 | 0.4452 |
 | cmmlu          | 0.5036 | 0.4580 | 0.4683 |
 | ceval-valid    | 0.5312 | 0.4844 | 0.5015 |
@@ -98,7 +98,6 @@ auto-round --model "Intel/Qwen2.5-0.5B-Instruct-int4-inc" --eval --eval_bs 16  -
 | winogrande     | 0.5604 | 0.5683 | 0.5556 |
 | piqa           | 0.7035 | 0.7008 | 0.7089 |
 | truthfulqa_mc1 | 0.2693 | 0.2583 | 0.2570 |
-| truthfulqa_mc2 | 0.4183 | 0.4123 | 0.4025 |
 | openbookqa     | 0.2400 | 0.2360 | 0.2340 |
 | boolq          | 0.6798 | 0.6826 | 0.6676 |
 | arc_easy       | 0.6557 | 0.6536 | 0.6334 |
@@ -115,8 +114,8 @@ Here is the sample command to reproduce the model. We observed a larger accuracy
 
 ```bash
 git clone https://github.com/intel/auto-round
-pip install -vvv --no-build-isolation -e .
-auto-round \
+cd auto-round
+python -m auto_round \
 --model_name  Qwen/Qwen2.5-0.5B-Instruct \
 --device 0 \
 --group_size 128 \