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

Typo fix in docstring

Browse source 
Co-Authored-By: ggerganov@gmail.com

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
Signed-off-by: Gabe Goodhart <ghart@us.ibm.com>
This commit is contained in:
Gabe Goodhart 2024-09-23 09:32:23 -06:00 • committed by Gabe Goodhart
parent eca37cd4f2
commit 71bc4c1f93
1 changed files with 1 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

2
convert_hf_to_gguf.py
View file

@ -4122,7 +4122,7 @@ class GraniteMoeModel(GraniteModel):
def modify_tensors(self, data_torch: Tensor, name: str, bid: int | None) -> Iterable[tuple[str, Tensor]]:
"""In modeling_granitemoe, the JetMoe implementation of parallel experts
is used. This essentially merges w1 and w3 into a single tensor with 2x
the hidden size that is then split during forward. To keep compativility
the hidden size that is then split during forward. To keep compatibility
with existing mixtral support, we pull them apart here.
"""