Closed Lanxin1011 closed 10 months ago
Dear the suthors, Thanks for the great work! I encounter a problem when training the model to reproduce the performance in shikra~ When training the model, what should be the initial weight? (/path/to/init/checkpoint) LLaMA-7B or shikra-7B?
Looking forward to your reply!
Have you got the answers?
uyo9ko
commented
9 months ago
Dear the suthors, Thanks for the great work! I encounter a problem when training the model to reproduce the performance in shikra~ When training the model, what should be the initial weight? (/path/to/init/checkpoint) LLaMA-7B or shikra-7B?
Looking forward to your reply!