### System Info / 系統信息
无
### Who can help? / 谁可以帮助到您?
无
### Information / 问题信息
- [ ] The official example scripts / 官方的示例脚本
- [ ] My own modified scripts / 我自己修改的脚本和任务
### Reproduction / 复现过程...
I reviewed the code for qwen2_7b_instruct_quantized, but this code uses a model that has already been converted by QNN. I want to use the native Qwen2 model with SHALlamaAttention.
**Describe the issue**
I see there is a Qwen2 7B model. I am trying to use SHALlamaAttention with the Qwen2 0.5B model, but I am getting gibberish. How do I adapt it here?
**To Reproduce**
...