Is rotate_ov_proj necessary ？

[jiajiyuan]

QQQ/QQQ/rotation/rotation.py

Line 194 in e307d9f

rotate_ov_proj(layer, model_type, num_heads, head_dim)

Thanks for your great work！
In the original quarot paper， rotate_ov_proj is designed to rotate the v state， so the (4bit) quantization for the value cache can get easier. But in this repo, I think we only want to rotate the weight and the linear output, we do not want to quant the v cache. So, is this necessary to have this rotate_ov_proj here? If my understanding is incorrect, please point it out. I am looking forward to your reply. Thanks!

[jiajiyuan]

Furthermore, I have also observed that you have also modified the handling of bias and the parameter had_dim passed into the function. Could you please explain the motivation behind these changes? Thanks!