vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

lora : raise error if lm_head is ignored

Browse source
This commit is contained in:
Xuan Son Nguyen 2024-08-20 15:06:20 +02:00
parent 90db8146d5
commit cef2c97c03
1 changed files with 4 additions and 0 deletions
Download patch file Download diff file Expand all files Collapse all files

4
convert_lora_to_gguf.py
View file

@ -364,6 +364,10 @@ if __name__ == '__main__':
def modify_tensors(self, data_torch: Tensor, name: str, bid: int | None) -> Iterable[tuple[str, Tensor]]: def modify_tensors(self, data_torch: Tensor, name: str, bid: int | None) -> Iterable[tuple[str, Tensor]]:
dest = super().modify_tensors(data_torch, name, bid) dest = super().modify_tensors(data_torch, name, bid)
# for now, we cannot convert archs that use the same tensor for tok_embd and output
# see: https://github.com/ggerganov/llama.cpp/issues/9065
if name == "lm_head.weight" and len(dest) == 0:
raise ValueError(f"lm_head is present in adapter, but is ignored in base model")
for dest_name, dest_data in dest: for dest_name, dest_data in dest:
assert isinstance(dest_data, LoraTorchTensor) assert isinstance(dest_data, LoraTorchTensor)
lora_a, lora_b = dest_data.get_lora_A_B() lora_a, lora_b = dest_data.get_lora_A_B()