Skip to content

Commit dd04a96

Browse files
authored
[Bugfix] Fix the bug of incorrect precision (#2479)
### What this PR does / why we need it? Fix the bug of incorrect precision - vLLM version: v0.10.0 - vLLM main: vllm-project/vllm@5341565 --------- Signed-off-by: weiguihua2 <[email protected]>
1 parent f0be3ee commit dd04a96

File tree

1 file changed

+2
-2
lines changed

1 file changed

+2
-2
lines changed

vllm_ascend/quantization/quantizer.py

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -75,8 +75,8 @@ def __init__(self, quant_description):
7575
"vllm.model_executor.layers.layernorm.RMSNorm", "__init__",
7676
[wrapper_rmsnorm_init])
7777
VLLMAscendQuantizer.apply_patch(
78-
"vllm.model_executor.layers.layernorm.RMSNorm",
79-
"forward_oot", [wrapper_rmsnorm_forward_oot])
78+
"vllm_ascend.ops.layernorm.AscendRMSNorm", "forward_oot",
79+
[wrapper_rmsnorm_forward_oot])
8080
VLLMAscendQuantizer.apply_patch(
8181
"vllm.model_executor.layers.vocab_parallel_embedding.VocabParallelEmbedding",
8282
"__init__", [wrapper_vocab_parallel_embedding_init])

0 commit comments

Comments
 (0)