Skip to content
This repository has been archived by the owner on Aug 30, 2024. It is now read-only.

Commit

Permalink
[pre-commit.ci] auto fixes from pre-commit.com hooks
Browse files Browse the repository at this point in the history
for more information, see https://pre-commit.ci
  • Loading branch information
pre-commit-ci[bot] committed Jun 4, 2024
1 parent c5a1cbb commit 06ccf30
Showing 1 changed file with 1 addition and 1 deletion.
2 changes: 1 addition & 1 deletion neural_speed/convert/convert_quantized_qwen.py
Original file line number Diff line number Diff line change
Expand Up @@ -65,7 +65,7 @@ def main(args_in: Optional[List[str]] = None) -> None:
f.write(struct.pack("i", hparams["hidden_size"]))
f.write(struct.pack("i", hparams["intermediate_size"])) # dummy data
f.write(struct.pack("i", hparams["num_attention_heads"]))
f.write(struct.pack("i", hparams["num_key_value_heads"] if "num_key_value_heads" in hparams
f.write(struct.pack("i", hparams["num_key_value_heads"] if "num_key_value_heads" in hparams
else ["num_attention_heads"])) # multi-query attention
f.write(struct.pack("i", hparams["num_hidden_layers"]))
f.write(
Expand Down

0 comments on commit 06ccf30

Please sign in to comment.