Mirrors/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/llama.cpp.git synced 2025-01-15 06:40:45 +01:00
Code Issues Packages Projects Releases Wiki Activity

convert-hf : check for unprocessed Jamba experts

Browse Source
This commit is contained in:
Francis Couture-Harpin 2024-05-25 12:54:30 -04:00
parent 61a88a1da3
commit ea2e63e9d2
1 changed files with 9 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

9
convert-hf-to-gguf.py
View File

@ -2470,6 +2470,15 @@ class JambaModel(Model):
yield new_name, data_torch yield new_name, data_torch
def write_tensors(self):
super().write_tensors()
if self._experts is not None:
# flatten `list[dict[str, Tensor]]` into `list[str]`
experts = [k for d in self._experts for k in d.keys()]
if len(experts) > 0:
raise ValueError(f"Unprocessed experts: {experts}")
# same as Mamba # same as Mamba
def extra_f32_tensors(self, name: str, new_name: str, bid: int | None, n_dims: int) -> bool: def extra_f32_tensors(self, name: str, new_name: str, bid: int | None, n_dims: int) -> bool:
del n_dims # unused del n_dims # unused