xyz.mlir

    %216 = torch.aten.linear %result0, %178, %179 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[384,384],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %217 = torch.aten.linear %result0, %177, %3 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[384,384],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %224 = torch.aten.linear %result0, %175, %176 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[384,384],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %250 = torch.aten.linear %249, %173, %174 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[384,384],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %252 = torch.aten.linear %result0_4, %169, %170 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[1536,384],f32>, !torch.vtensor<[1536],f32> -> !torch.vtensor<[1,128,1536],f32>
    %254 = torch.aten.linear %253, %167, %168 : !torch.vtensor<[1,128,1536],f32>, !torch.vtensor<[384,1536],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %256 = torch.aten.linear %result0_7, %163, %164 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[384,384],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %257 = torch.aten.linear %result0_7, %162, %3 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[384,384],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %262 = torch.aten.linear %result0_7, %160, %161 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[384,384],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %287 = torch.aten.linear %286, %158, %159 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[384,384],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %289 = torch.aten.linear %result0_16, %154, %155 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[1536,384],f32>, !torch.vtensor<[1536],f32> -> !torch.vtensor<[1,128,1536],f32>
    %291 = torch.aten.linear %290, %152, %153 : !torch.vtensor<[1,128,1536],f32>, !torch.vtensor<[384,1536],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %293 = torch.aten.linear %result0_19, %148, %149 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[384,384],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %294 = torch.aten.linear %result0_19, %147, %3 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[384,384],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %299 = torch.aten.linear %result0_19, %145, %146 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[384,384],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %324 = torch.aten.linear %323, %143, %144 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[384,384],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %326 = torch.aten.linear %result0_28, %139, %140 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[1536,384],f32>, !torch.vtensor<[1536],f32> -> !torch.vtensor<[1,128,1536],f32>
    %328 = torch.aten.linear %327, %137, %138 : !torch.vtensor<[1,128,1536],f32>, !torch.vtensor<[384,1536],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %330 = torch.aten.linear %result0_31, %133, %134 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[384,384],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %331 = torch.aten.linear %result0_31, %132, %3 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[384,384],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %336 = torch.aten.linear %result0_31, %130, %131 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[384,384],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %361 = torch.aten.linear %360, %128, %129 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[384,384],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %363 = torch.aten.linear %result0_40, %124, %125 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[1536,384],f32>, !torch.vtensor<[1536],f32> -> !torch.vtensor<[1,128,1536],f32>
    %365 = torch.aten.linear %364, %122, %123 : !torch.vtensor<[1,128,1536],f32>, !torch.vtensor<[384,1536],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %367 = torch.aten.linear %result0_43, %118, %119 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[384,384],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %368 = torch.aten.linear %result0_43, %117, %3 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[384,384],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %373 = torch.aten.linear %result0_43, %115, %116 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[384,384],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %398 = torch.aten.linear %397, %113, %114 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[384,384],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %400 = torch.aten.linear %result0_52, %109, %110 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[1536,384],f32>, !torch.vtensor<[1536],f32> -> !torch.vtensor<[1,128,1536],f32>
    %402 = torch.aten.linear %401, %107, %108 : !torch.vtensor<[1,128,1536],f32>, !torch.vtensor<[384,1536],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %404 = torch.aten.linear %result0_55, %103, %104 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[384,384],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %405 = torch.aten.linear %result0_55, %102, %3 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[384,384],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %410 = torch.aten.linear %result0_55, %100, %101 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[384,384],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %435 = torch.aten.linear %434, %98, %99 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[384,384],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %437 = torch.aten.linear %result0_64, %94, %95 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[1536,384],f32>, !torch.vtensor<[1536],f32> -> !torch.vtensor<[1,128,1536],f32>
    %439 = torch.aten.linear %438, %92, %93 : !torch.vtensor<[1,128,1536],f32>, !torch.vtensor<[384,1536],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %441 = torch.aten.linear %result0_67, %88, %89 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[384,384],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %442 = torch.aten.linear %result0_67, %87, %3 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[384,384],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %447 = torch.aten.linear %result0_67, %85, %86 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[384,384],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %472 = torch.aten.linear %471, %83, %84 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[384,384],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %474 = torch.aten.linear %result0_76, %79, %80 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[1536,384],f32>, !torch.vtensor<[1536],f32> -> !torch.vtensor<[1,128,1536],f32>
    %476 = torch.aten.linear %475, %77, %78 : !torch.vtensor<[1,128,1536],f32>, !torch.vtensor<[384,1536],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %478 = torch.aten.linear %result0_79, %73, %74 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[384,384],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %479 = torch.aten.linear %result0_79, %72, %3 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[384,384],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %484 = torch.aten.linear %result0_79, %70, %71 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[384,384],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %509 = torch.aten.linear %508, %68, %69 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[384,384],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %511 = torch.aten.linear %result0_88, %64, %65 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[1536,384],f32>, !torch.vtensor<[1536],f32> -> !torch.vtensor<[1,128,1536],f32>
    %513 = torch.aten.linear %512, %62, %63 : !torch.vtensor<[1,128,1536],f32>, !torch.vtensor<[384,1536],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %515 = torch.aten.linear %result0_91, %58, %59 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[384,384],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %516 = torch.aten.linear %result0_91, %57, %3 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[384,384],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %521 = torch.aten.linear %result0_91, %55, %56 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[384,384],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %546 = torch.aten.linear %545, %53, %54 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[384,384],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %548 = torch.aten.linear %result0_100, %49, %50 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[1536,384],f32>, !torch.vtensor<[1536],f32> -> !torch.vtensor<[1,128,1536],f32>
    %550 = torch.aten.linear %549, %47, %48 : !torch.vtensor<[1,128,1536],f32>, !torch.vtensor<[384,1536],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %552 = torch.aten.linear %result0_103, %43, %44 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[384,384],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %553 = torch.aten.linear %result0_103, %42, %3 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[384,384],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %558 = torch.aten.linear %result0_103, %40, %41 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[384,384],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %583 = torch.aten.linear %582, %38, %39 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[384,384],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %585 = torch.aten.linear %result0_112, %34, %35 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[1536,384],f32>, !torch.vtensor<[1536],f32> -> !torch.vtensor<[1,128,1536],f32>
    %587 = torch.aten.linear %586, %32, %33 : !torch.vtensor<[1,128,1536],f32>, !torch.vtensor<[384,1536],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %589 = torch.aten.linear %result0_115, %28, %29 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[384,384],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %590 = torch.aten.linear %result0_115, %27, %3 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[384,384],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %595 = torch.aten.linear %result0_115, %25, %26 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[384,384],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %620 = torch.aten.linear %619, %23, %24 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[384,384],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %622 = torch.aten.linear %result0_124, %19, %20 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[1536,384],f32>, !torch.vtensor<[1536],f32> -> !torch.vtensor<[1,128,1536],f32>
    %624 = torch.aten.linear %623, %17, %18 : !torch.vtensor<[1,128,1536],f32>, !torch.vtensor<[384,1536],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %626 = torch.aten.linear %result0_127, %12, %13 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[384,384],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %627 = torch.aten.linear %result0_127, %11, %3 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[384,384],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %632 = torch.aten.linear %result0_127, %9, %10 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[384,384],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %657 = torch.aten.linear %656, %8, %3 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[384,384],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %659 = torch.aten.linear %result0_136, %6, %7 : !torch.vtensor<[1,128,384],f32>, !torch.vtensor<[1536,384],f32>, !torch.vtensor<[1536],f32> -> !torch.vtensor<[1,128,1536],f32>
    %661 = torch.aten.linear %660, %5, %3 : !torch.vtensor<[1,128,1536],f32>, !torch.vtensor<[384,1536],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,128,384],f32>
    %667 = torch.aten.linear %666, %2, %3 : !torch.vtensor<[1,384],f32>, !torch.vtensor<[384,384],f32>, !torch.vtensor<[384],f32> -> !torch.vtensor<[1,384],f32>
    %669 = torch.aten.linear %668, %0, %1 : !torch.vtensor<[1,384],f32>, !torch.vtensor<[2,384],f32>, !torch.vtensor<[2],f32> -> !torch.vtensor<[1,2],f32>