Skip to content

llava: fixes #12863 v.head.ffn_up.bias byteswap on BE systems #12918

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
wants to merge 2 commits into from
Closed

llava: fixes #12863 v.head.ffn_up.bias byteswap on BE systems #12918

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
8765394
llava: fixes #12863 v.head.ffn_up.bias byteswap on BE systems
taronaeo Apr 12, 2025
9ba4b4f
llava: fix typo in arg vs args
taronaeo Apr 12, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
5 changes: 5 additions & 0 deletions examples/llava/convert_image_encoder_to_gguf.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -363,6 +363,11 @@ def get_non_negative_vision_feature_layers(v_hparams):
else:
data = data.squeeze().numpy().astype(np.float32)

# byteswaps v.head.ffn_up.bias for big-endian systems
# see: https://github.com/ggml-org/llama.cpp/issues/12863
if name == "v.head.ffn_up.bias" and args.bigendian:
data = data.byteswap(inplace=True)

fout.add_tensor(name, data)

print("Projector tensors added\n")
Expand Down