您好，我在使用CANN7.0RC1对模型（经过量化后的Llama3.1onnx模型）进行转换时，报错信息如下：

E10042: GenerateOfflineModel execute failed.

        TraceBack (most recent call last):

        op[/model/layers.0/self_attn/Concat_6], the input shape dims should be equal except merge axis,shapes:[[1, 32, 1024, 128, ], [1, 8, 1, 128, ], ]axis:-2[FUNC:ConcatInferShapeCommon][FILE:split_combination_ops.cc][LINE:892]

        Call InferShapeAndType for node:/model/layers.0/self_attn/Concat_6(ConcatD) failed[FUNC:Infer][FILE:infershape_pass.cc][LINE:119]

        process pass InferShapePass on node:/model/layers.0/self_attn/Concat_6 failed, ret:4294967295[FUNC:RunPassesOnNode][FILE:base_pass.cc][LINE:571]

        build graph failed, graph id:0, ret:1343242270[FUNC:BuildModelWithGraphId][FILE:ge_generator.cc][LINE:1615]

        GenerateOfflineModel execute failed.

其中，模型转换指令为：
atc --framework=5 --model="llama3.onnx"  --output="llama3.om" --input_format=ND --input_shape="input_ids:1,1;attention_mask:1,1025;position_ids:1,1;past_key_values:32,2,1,32,1024,128" --log=debug --soc_version=Ascend310B1 --precision_mode=must_keep_origin_dtype
麻烦帮忙看一下，非常感谢！