llama.cpp convert-pth-to-ggml.py how to handle torch.view_as

convert-pth-to-ggml.py how to handle torch.view_as_complex

Open haolongzhangm opened this issue 2 years ago • 2 comments

llama code block include view_as_real: https://github.com/facebookresearch/llama/blob/main/llama/model.py#L68

how to convert-pth-to-ggml.py handle this part of weight

Mar 17 '23 05:03 haolongzhangm

Please improve your question with more text and examples so it is easier to understand what you are asking.

Mar 17 '23 08:03 gjmulder

If you are asking about applying rotary embeddings, then that is done in the llama.cpp file and not during conversion

Mar 22 '23 14:03 nullhook

@nullhook ths for u info

Apr 10 '23 08:04 haolongzhangm