{
  "schema_version": 1,
  "method": "one warmed complete mixer F+B under torch.profiler; CUDA device events",
  "rows": [
    {
      "shape": {
        "batch": 4,
        "length": 2048,
        "width": 2048
      },
      "architecture": "rglru",
      "backend": "fused_grouped_prefix32_hierarchical_chunk16",
      "cuda_kernel_events": 50,
      "unique_cuda_kernel_names": 22,
      "top_cuda_kernels": [
        [
          "Memset (Device)",
          9
        ],
        [
          "void at::native::unrolled_elementwise_kernel<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#12}::operator()() const::{lambda(c10::BFloat16)#1}, at::detail::Array<char*, 2>, TrivialOffsetCalculator<1, unsigned int>, TrivialOffsetCalculator<1, unsigned int>, at::native::memory::LoadWithCast<1>, at::native::memory::StoreWithCast<1> >(int, at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#12}::operator()() const::{lambda(c10::BFloat16)#1}, at::detail::Array<char*, 2>, TrivialOffsetCalculator<1, unsigned int>, TrivialOffsetCalculator<1, unsigned int>, at::native::memory::LoadWithCast<1>, at::native::memory::StoreWithCast<1>)",
          5
        ],
        [
          "void at::native::unrolled_elementwise_kernel<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#7}::operator()() const::{lambda(float)#1}, at::detail::Array<char*, 2>, TrivialOffsetCalculator<1, unsigned int>, TrivialOffsetCalculator<1, unsigned int>, at::native::memory::LoadWithCast<1>, at::native::memory::StoreWithCast<1> >(int, at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#7}::operator()() const::{lambda(float)#1}, at::detail::Array<char*, 2>, TrivialOffsetCalculator<1, unsigned int>, TrivialOffsetCalculator<1, unsigned int>, at::native::memory::LoadWithCast<1>, at::native::memory::StoreWithCast<1>)",
          4
        ],
        [
          "void at::native::reduce_kernel<128, 4, at::native::ReduceOp<float, at::native::func_wrapper_t<float, at::native::sum_functor<float, float, float>::operator()(at::TensorIterator&)::{lambda(float, float)#1}>, unsigned int, float, 4> >(at::native::ReduceOp<float, at::native::func_wrapper_t<float, at::native::sum_functor<float, float, float>::operator()(at::TensorIterator&)::{lambda(float, float)#1}>, unsigned int, float, 4>)",
          3
        ],
        [
          "sm90_xmma_gemm_bf16bf16_bf16f32_f32_nn_n_tilesize256x128x64_warpgroupsize2x1x1_execute_segment_k_off_kernel__51_cublas",
          2
        ],
        [
          "void at::native::elementwise_kernel<128, 4, at::native::gpu_kernel_impl<at::native::CUDAFunctor_add<float> >(at::TensorIteratorBase&, at::native::CUDAFunctor_add<float> const&)::{lambda(int)#2}>(int, at::native::gpu_kernel_impl<at::native::CUDAFunctor_add<float> >(at::TensorIteratorBase&, at::native::CUDAFunctor_add<float> const&)::{lambda(int)#2})",
          2
        ],
        [
          "void at::native::elementwise_kernel<128, 2, at::native::gpu_kernel_impl<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#7}::operator()() const::{lambda(float)#1}>(at::TensorIteratorBase&, at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#7}::operator()() const::{lambda(float)#1} const&)::{lambda(int)#1}>(int, at::native::gpu_kernel_impl<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#7}::operator()() const::{lambda(float)#1}>(at::TensorIteratorBase&, at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#7}::operator()() const::{lambda(float)#1} const&)::{lambda(int)#1})",
          2
        ],
        [
          "_affine_group_local",
          2
        ],
        [
          "_affine_group_outer",
          2
        ],
        [
          "_affine_group_correct",
          2
        ],
        [
          "void at::native::vectorized_elementwise_kernel<4, at::native::FillFunctor<float>, at::detail::Array<char*, 1> >(int, at::native::FillFunctor<float>, at::detail::Array<char*, 1>)",
          2
        ],
        [
          "sm90_xmma_gemm_bf16bf16_bf16f32_f32_nt_n_tilesize64x64x64_warpgroupsize1x1x1_execute_segment_k_off_kernel__51_cublas",
          2
        ],
        [
          "sm90_xmma_gemm_bf16bf16_bf16f32_f32_tn_n_tilesize256x128x64_warpgroupsize2x1x1_execute_segment_k_off_kernel__51_cublas",
          2
        ],
        [
          "void at::native::elementwise_kernel<128, 4, at::native::gpu_kernel_impl<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#12}::operator()() const::{lambda(c10::BFloat16)#1}>(at::TensorIteratorBase&, at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#12}::operator()() const::{lambda(c10::BFloat16)#1} const&)::{lambda(int)#1}>(int, at::native::gpu_kernel_impl<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#12}::operator()() const::{lambda(c10::BFloat16)#1}>(at::TensorIteratorBase&, at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#12}::operator()() const::{lambda(c10::BFloat16)#1} const&)::{lambda(int)#1})",
          2
        ],
        [
          "void at::native::vectorized_elementwise_kernel<4, at::native::CUDAFunctor_add<c10::BFloat16>, at::detail::Array<char*, 3> >(int, at::native::CUDAFunctor_add<c10::BFloat16>, at::detail::Array<char*, 3>)",
          2
        ],
        [
          "_chunk_summary",
          1
        ],
        [
          "_chunk_replay",
          1
        ],
        [
          "void at::native::vectorized_elementwise_kernel<4, at::native::BinaryFunctor<float, float, float, at::native::binary_internal::MulFunctor<float> >, at::detail::Array<char*, 3> >(int, at::native::BinaryFunctor<float, float, float, at::native::binary_internal::MulFunctor<float> >, at::detail::Array<char*, 3>)",
          1
        ],
        [
          "void at::native::reduce_kernel<512, 1, at::native::ReduceOp<float, at::native::func_wrapper_t<float, at::native::sum_functor<float, float, float>::operator()(at::TensorIterator&)::{lambda(float, float)#1}>, unsigned int, float, 4> >(at::native::ReduceOp<float, at::native::func_wrapper_t<float, at::native::sum_functor<float, float, float>::operator()(at::TensorIterator&)::{lambda(float, float)#1}>, unsigned int, float, 4>)",
          1
        ],
        [
          "void at::native::elementwise_kernel<128, 2, at::native::gpu_kernel_impl<at::native::BinaryFunctor<float, float, float, at::native::binary_internal::MulFunctor<float> > >(at::TensorIteratorBase&, at::native::BinaryFunctor<float, float, float, at::native::binary_internal::MulFunctor<float> > const&)::{lambda(int)#1}>(int, at::native::gpu_kernel_impl<at::native::BinaryFunctor<float, float, float, at::native::binary_internal::MulFunctor<float> > >(at::TensorIteratorBase&, at::native::BinaryFunctor<float, float, float, at::native::binary_internal::MulFunctor<float> > const&)::{lambda(int)#1})",
          1
        ]
      ]
    },
    {
      "shape": {
        "batch": 4,
        "length": 2048,
        "width": 2048
      },
      "architecture": "samu",
      "backend": "fused_output_fused_write_shared_sfu_chunk32",
      "cuda_kernel_events": 70,
      "unique_cuda_kernel_names": 42,
      "top_cuda_kernels": [
        [
          "void at::native::vectorized_elementwise_kernel<4, at::native::FillFunctor<float>, at::detail::Array<char*, 1> >(int, at::native::FillFunctor<float>, at::detail::Array<char*, 1>)",
          8
        ],
        [
          "void at::native::unrolled_elementwise_kernel<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#7}::operator()() const::{lambda(float)#1}, at::detail::Array<char*, 2>, TrivialOffsetCalculator<1, unsigned int>, TrivialOffsetCalculator<1, unsigned int>, at::native::memory::LoadWithCast<1>, at::native::memory::StoreWithCast<1> >(int, at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#7}::operator()() const::{lambda(float)#1}, at::detail::Array<char*, 2>, TrivialOffsetCalculator<1, unsigned int>, TrivialOffsetCalculator<1, unsigned int>, at::native::memory::LoadWithCast<1>, at::native::memory::StoreWithCast<1>)",
          6
        ],
        [
          "void at::native::elementwise_kernel<128, 2, at::native::gpu_kernel_impl<at::native::BinaryFunctor<float, float, float, at::native::binary_internal::DivFunctor<float> > >(at::TensorIteratorBase&, at::native::BinaryFunctor<float, float, float, at::native::binary_internal::DivFunctor<float> > const&)::{lambda(int)#1}>(int, at::native::gpu_kernel_impl<at::native::BinaryFunctor<float, float, float, at::native::binary_internal::DivFunctor<float> > >(at::TensorIteratorBase&, at::native::BinaryFunctor<float, float, float, at::native::binary_internal::DivFunctor<float> > const&)::{lambda(int)#1})",
          5
        ],
        [
          "void at::native::unrolled_elementwise_kernel<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#12}::operator()() const::{lambda(c10::BFloat16)#1}, at::detail::Array<char*, 2>, TrivialOffsetCalculator<1, unsigned int>, TrivialOffsetCalculator<1, unsigned int>, at::native::memory::LoadWithCast<1>, at::native::memory::StoreWithCast<1> >(int, at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#12}::operator()() const::{lambda(c10::BFloat16)#1}, at::detail::Array<char*, 2>, TrivialOffsetCalculator<1, unsigned int>, TrivialOffsetCalculator<1, unsigned int>, at::native::memory::LoadWithCast<1>, at::native::memory::StoreWithCast<1>)",
          4
        ],
        [
          "void at::native::elementwise_kernel<128, 4, at::native::gpu_kernel_impl<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#12}::operator()() const::{lambda(c10::BFloat16)#1}>(at::TensorIteratorBase&, at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#12}::operator()() const::{lambda(c10::BFloat16)#1} const&)::{lambda(int)#2}>(int, at::native::gpu_kernel_impl<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#12}::operator()() const::{lambda(c10::BFloat16)#1}>(at::TensorIteratorBase&, at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#12}::operator()() const::{lambda(c10::BFloat16)#1} const&)::{lambda(int)#2})",
          2
        ],
        [
          "Memset (Device)",
          2
        ],
        [
          "void at::native::vectorized_elementwise_kernel<4, at::native::exp_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#2}::operator()() const::{lambda(float)#1}, at::detail::Array<char*, 2> >(int, at::native::exp_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#2}::operator()() const::{lambda(float)#1}, at::detail::Array<char*, 2>)",
          2
        ],
        [
          "void at::native::vectorized_elementwise_kernel<4, at::native::BinaryFunctor<float, float, float, at::native::binary_internal::MulFunctor<float> >, at::detail::Array<char*, 3> >(int, at::native::BinaryFunctor<float, float, float, at::native::binary_internal::MulFunctor<float> >, at::detail::Array<char*, 3>)",
          2
        ],
        [
          "void at::native::reduce_kernel<512, 1, at::native::ReduceOp<float, at::native::func_wrapper_t<float, at::native::sum_functor<float, float, float>::operator()(at::TensorIterator&)::{lambda(float, float)#1}>, unsigned int, float, 4> >(at::native::ReduceOp<float, at::native::func_wrapper_t<float, at::native::sum_functor<float, float, float>::operator()(at::TensorIterator&)::{lambda(float, float)#1}>, unsigned int, float, 4>)",
          2
        ],
        [
          "void at::native::elementwise_kernel<128, 2, at::native::gpu_kernel_impl<at::native::BinaryFunctor<float, float, float, at::native::binary_internal::MulFunctor<float> > >(at::TensorIteratorBase&, at::native::BinaryFunctor<float, float, float, at::native::binary_internal::MulFunctor<float> > const&)::{lambda(int)#1}>(int, at::native::gpu_kernel_impl<at::native::BinaryFunctor<float, float, float, at::native::binary_internal::MulFunctor<float> > >(at::TensorIteratorBase&, at::native::BinaryFunctor<float, float, float, at::native::binary_internal::MulFunctor<float> > const&)::{lambda(int)#1})",
          2
        ],
        [
          "void at::native::elementwise_kernel<128, 2, at::native::gpu_kernel_impl<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#7}::operator()() const::{lambda(float)#1}>(at::TensorIteratorBase&, at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#7}::operator()() const::{lambda(float)#1} const&)::{lambda(int)#1}>(int, at::native::gpu_kernel_impl<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#7}::operator()() const::{lambda(float)#1}>(at::TensorIteratorBase&, at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#7}::operator()() const::{lambda(float)#1} const&)::{lambda(int)#1})",
          2
        ],
        [
          "Memcpy DtoD (Device -> Device)",
          2
        ],
        [
          "void at::native::vectorized_elementwise_kernel<4, at::native::CUDAFunctor_add<float>, at::detail::Array<char*, 3> >(int, at::native::CUDAFunctor_add<float>, at::detail::Array<char*, 3>)",
          2
        ],
        [
          "void at::native::(anonymous namespace)::CatArrayBatchedCopy_aligned16_contig<float, unsigned int, 1, 128, 1>(float*, at::native::(anonymous namespace)::CatArrInputTensorMetadata<float, unsigned int, 128, 1>, at::native::(anonymous namespace)::TensorSizeStride<unsigned int, 4u>, int, unsigned int)",
          1
        ],
        [
          "void at::native::reduce_kernel<512, 1, at::native::ReduceOp<float, at::native::NormTwoOps<float, float, float>, unsigned int, float, 4> >(at::native::ReduceOp<float, at::native::NormTwoOps<float, float, float>, unsigned int, float, 4>)",
          1
        ],
        [
          "void at::native::vectorized_elementwise_kernel<4, at::native::(anonymous namespace)::launch_clamp_scalar(at::TensorIteratorBase&, c10::Scalar, c10::Scalar, at::native::detail::ClampLimits)::{lambda()#1}::operator()() const::{lambda()#7}::operator()() const::{lambda(float)#1}, at::detail::Array<char*, 2> >(int, at::native::(anonymous namespace)::launch_clamp_scalar(at::TensorIteratorBase&, c10::Scalar, c10::Scalar, at::native::detail::ClampLimits)::{lambda()#1}::operator()() const::{lambda()#7}::operator()() const::{lambda(float)#1}, at::detail::Array<char*, 2>)",
          1
        ],
        [
          "void cutlass::Kernel<cutlass_80_tensorop_bf16_s16816gemm_bf16_64x64_64x6_tn_align2>(cutlass_80_tensorop_bf16_s16816gemm_bf16_64x64_64x6_tn_align2::Params)",
          1
        ],
        [
          "_formal_coordinate_forward",
          1
        ],
        [
          "void at::native::vectorized_elementwise_kernel<4, at::native::AUnaryFunctor<float, float, float, at::native::binary_internal::MulFunctor<float> >, at::detail::Array<char*, 2> >(int, at::native::AUnaryFunctor<float, float, float, at::native::binary_internal::MulFunctor<float> >, at::detail::Array<char*, 2>)",
          1
        ],
        [
          "void at::native::vectorized_elementwise_kernel<4, at::native::CUDAFunctorOnOther_add<float>, at::detail::Array<char*, 2> >(int, at::native::CUDAFunctorOnOther_add<float>, at::detail::Array<char*, 2>)",
          1
        ]
      ]
    },
    {
      "shape": {
        "batch": 1,
        "length": 8192,
        "width": 2560
      },
      "architecture": "rglru",
      "backend": "fused_grouped_prefix32_hierarchical_chunk16",
      "cuda_kernel_events": 50,
      "unique_cuda_kernel_names": 22,
      "top_cuda_kernels": [
        [
          "Memset (Device)",
          9
        ],
        [
          "void at::native::unrolled_elementwise_kernel<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#12}::operator()() const::{lambda(c10::BFloat16)#1}, at::detail::Array<char*, 2>, TrivialOffsetCalculator<1, unsigned int>, TrivialOffsetCalculator<1, unsigned int>, at::native::memory::LoadWithCast<1>, at::native::memory::StoreWithCast<1> >(int, at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#12}::operator()() const::{lambda(c10::BFloat16)#1}, at::detail::Array<char*, 2>, TrivialOffsetCalculator<1, unsigned int>, TrivialOffsetCalculator<1, unsigned int>, at::native::memory::LoadWithCast<1>, at::native::memory::StoreWithCast<1>)",
          5
        ],
        [
          "void at::native::unrolled_elementwise_kernel<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#7}::operator()() const::{lambda(float)#1}, at::detail::Array<char*, 2>, TrivialOffsetCalculator<1, unsigned int>, TrivialOffsetCalculator<1, unsigned int>, at::native::memory::LoadWithCast<1>, at::native::memory::StoreWithCast<1> >(int, at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#7}::operator()() const::{lambda(float)#1}, at::detail::Array<char*, 2>, TrivialOffsetCalculator<1, unsigned int>, TrivialOffsetCalculator<1, unsigned int>, at::native::memory::LoadWithCast<1>, at::native::memory::StoreWithCast<1>)",
          4
        ],
        [
          "void at::native::reduce_kernel<128, 4, at::native::ReduceOp<float, at::native::func_wrapper_t<float, at::native::sum_functor<float, float, float>::operator()(at::TensorIterator&)::{lambda(float, float)#1}>, unsigned int, float, 4> >(at::native::ReduceOp<float, at::native::func_wrapper_t<float, at::native::sum_functor<float, float, float>::operator()(at::TensorIterator&)::{lambda(float, float)#1}>, unsigned int, float, 4>)",
          3
        ],
        [
          "sm90_xmma_gemm_bf16bf16_bf16f32_f32_nn_n_tilesize256x128x64_warpgroupsize2x1x1_execute_segment_k_off_kernel__51_cublas",
          2
        ],
        [
          "void at::native::elementwise_kernel<128, 4, at::native::gpu_kernel_impl<at::native::CUDAFunctor_add<float> >(at::TensorIteratorBase&, at::native::CUDAFunctor_add<float> const&)::{lambda(int)#2}>(int, at::native::gpu_kernel_impl<at::native::CUDAFunctor_add<float> >(at::TensorIteratorBase&, at::native::CUDAFunctor_add<float> const&)::{lambda(int)#2})",
          2
        ],
        [
          "void at::native::elementwise_kernel<128, 2, at::native::gpu_kernel_impl<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#7}::operator()() const::{lambda(float)#1}>(at::TensorIteratorBase&, at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#7}::operator()() const::{lambda(float)#1} const&)::{lambda(int)#1}>(int, at::native::gpu_kernel_impl<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#7}::operator()() const::{lambda(float)#1}>(at::TensorIteratorBase&, at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#7}::operator()() const::{lambda(float)#1} const&)::{lambda(int)#1})",
          2
        ],
        [
          "_affine_group_local",
          2
        ],
        [
          "_affine_group_outer",
          2
        ],
        [
          "_affine_group_correct",
          2
        ],
        [
          "void at::native::vectorized_elementwise_kernel<4, at::native::FillFunctor<float>, at::detail::Array<char*, 1> >(int, at::native::FillFunctor<float>, at::detail::Array<char*, 1>)",
          2
        ],
        [
          "sm90_xmma_gemm_bf16bf16_bf16f32_f32_nt_n_tilesize64x64x64_warpgroupsize1x1x1_execute_segment_k_off_kernel__51_cublas",
          2
        ],
        [
          "sm90_xmma_gemm_bf16bf16_bf16f32_f32_tn_n_tilesize256x128x64_warpgroupsize2x1x1_execute_segment_k_off_kernel__51_cublas",
          2
        ],
        [
          "void at::native::elementwise_kernel<128, 4, at::native::gpu_kernel_impl<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#12}::operator()() const::{lambda(c10::BFloat16)#1}>(at::TensorIteratorBase&, at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#12}::operator()() const::{lambda(c10::BFloat16)#1} const&)::{lambda(int)#1}>(int, at::native::gpu_kernel_impl<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#12}::operator()() const::{lambda(c10::BFloat16)#1}>(at::TensorIteratorBase&, at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#12}::operator()() const::{lambda(c10::BFloat16)#1} const&)::{lambda(int)#1})",
          2
        ],
        [
          "void at::native::vectorized_elementwise_kernel<4, at::native::CUDAFunctor_add<c10::BFloat16>, at::detail::Array<char*, 3> >(int, at::native::CUDAFunctor_add<c10::BFloat16>, at::detail::Array<char*, 3>)",
          2
        ],
        [
          "_chunk_summary",
          1
        ],
        [
          "_chunk_replay",
          1
        ],
        [
          "void at::native::vectorized_elementwise_kernel<4, at::native::BinaryFunctor<float, float, float, at::native::binary_internal::MulFunctor<float> >, at::detail::Array<char*, 3> >(int, at::native::BinaryFunctor<float, float, float, at::native::binary_internal::MulFunctor<float> >, at::detail::Array<char*, 3>)",
          1
        ],
        [
          "void at::native::reduce_kernel<512, 1, at::native::ReduceOp<float, at::native::func_wrapper_t<float, at::native::sum_functor<float, float, float>::operator()(at::TensorIterator&)::{lambda(float, float)#1}>, unsigned int, float, 4> >(at::native::ReduceOp<float, at::native::func_wrapper_t<float, at::native::sum_functor<float, float, float>::operator()(at::TensorIterator&)::{lambda(float, float)#1}>, unsigned int, float, 4>)",
          1
        ],
        [
          "void at::native::elementwise_kernel<128, 2, at::native::gpu_kernel_impl<at::native::BinaryFunctor<float, float, float, at::native::binary_internal::MulFunctor<float> > >(at::TensorIteratorBase&, at::native::BinaryFunctor<float, float, float, at::native::binary_internal::MulFunctor<float> > const&)::{lambda(int)#1}>(int, at::native::gpu_kernel_impl<at::native::BinaryFunctor<float, float, float, at::native::binary_internal::MulFunctor<float> > >(at::TensorIteratorBase&, at::native::BinaryFunctor<float, float, float, at::native::binary_internal::MulFunctor<float> > const&)::{lambda(int)#1})",
          1
        ]
      ]
    },
    {
      "shape": {
        "batch": 1,
        "length": 8192,
        "width": 2560
      },
      "architecture": "samu",
      "backend": "fused_output_fused_write_shared_sfu_grouped_prefix64_hierarchical_chunk32",
      "cuda_kernel_events": 74,
      "unique_cuda_kernel_names": 43,
      "top_cuda_kernels": [
        [
          "void at::native::vectorized_elementwise_kernel<4, at::native::FillFunctor<float>, at::detail::Array<char*, 1> >(int, at::native::FillFunctor<float>, at::detail::Array<char*, 1>)",
          8
        ],
        [
          "void at::native::unrolled_elementwise_kernel<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#7}::operator()() const::{lambda(float)#1}, at::detail::Array<char*, 2>, TrivialOffsetCalculator<1, unsigned int>, TrivialOffsetCalculator<1, unsigned int>, at::native::memory::LoadWithCast<1>, at::native::memory::StoreWithCast<1> >(int, at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#7}::operator()() const::{lambda(float)#1}, at::detail::Array<char*, 2>, TrivialOffsetCalculator<1, unsigned int>, TrivialOffsetCalculator<1, unsigned int>, at::native::memory::LoadWithCast<1>, at::native::memory::StoreWithCast<1>)",
          6
        ],
        [
          "void at::native::elementwise_kernel<128, 2, at::native::gpu_kernel_impl<at::native::BinaryFunctor<float, float, float, at::native::binary_internal::DivFunctor<float> > >(at::TensorIteratorBase&, at::native::BinaryFunctor<float, float, float, at::native::binary_internal::DivFunctor<float> > const&)::{lambda(int)#1}>(int, at::native::gpu_kernel_impl<at::native::BinaryFunctor<float, float, float, at::native::binary_internal::DivFunctor<float> > >(at::TensorIteratorBase&, at::native::BinaryFunctor<float, float, float, at::native::binary_internal::DivFunctor<float> > const&)::{lambda(int)#1})",
          5
        ],
        [
          "void at::native::unrolled_elementwise_kernel<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#12}::operator()() const::{lambda(c10::BFloat16)#1}, at::detail::Array<char*, 2>, TrivialOffsetCalculator<1, unsigned int>, TrivialOffsetCalculator<1, unsigned int>, at::native::memory::LoadWithCast<1>, at::native::memory::StoreWithCast<1> >(int, at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#12}::operator()() const::{lambda(c10::BFloat16)#1}, at::detail::Array<char*, 2>, TrivialOffsetCalculator<1, unsigned int>, TrivialOffsetCalculator<1, unsigned int>, at::native::memory::LoadWithCast<1>, at::native::memory::StoreWithCast<1>)",
          4
        ],
        [
          "void at::native::elementwise_kernel<128, 4, at::native::gpu_kernel_impl<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#12}::operator()() const::{lambda(c10::BFloat16)#1}>(at::TensorIteratorBase&, at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#12}::operator()() const::{lambda(c10::BFloat16)#1} const&)::{lambda(int)#2}>(int, at::native::gpu_kernel_impl<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#12}::operator()() const::{lambda(c10::BFloat16)#1}>(at::TensorIteratorBase&, at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#12}::operator()() const::{lambda(c10::BFloat16)#1} const&)::{lambda(int)#2})",
          2
        ],
        [
          "Memset (Device)",
          2
        ],
        [
          "void at::native::vectorized_elementwise_kernel<4, at::native::exp_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#2}::operator()() const::{lambda(float)#1}, at::detail::Array<char*, 2> >(int, at::native::exp_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#2}::operator()() const::{lambda(float)#1}, at::detail::Array<char*, 2>)",
          2
        ],
        [
          "_complex_affine_group_local",
          2
        ],
        [
          "_complex_affine_group_outer",
          2
        ],
        [
          "_complex_affine_group_correct",
          2
        ],
        [
          "void at::native::vectorized_elementwise_kernel<4, at::native::BinaryFunctor<float, float, float, at::native::binary_internal::MulFunctor<float> >, at::detail::Array<char*, 3> >(int, at::native::BinaryFunctor<float, float, float, at::native::binary_internal::MulFunctor<float> >, at::detail::Array<char*, 3>)",
          2
        ],
        [
          "void at::native::reduce_kernel<512, 1, at::native::ReduceOp<float, at::native::func_wrapper_t<float, at::native::sum_functor<float, float, float>::operator()(at::TensorIterator&)::{lambda(float, float)#1}>, unsigned int, float, 4> >(at::native::ReduceOp<float, at::native::func_wrapper_t<float, at::native::sum_functor<float, float, float>::operator()(at::TensorIterator&)::{lambda(float, float)#1}>, unsigned int, float, 4>)",
          2
        ],
        [
          "void at::native::elementwise_kernel<128, 2, at::native::gpu_kernel_impl<at::native::BinaryFunctor<float, float, float, at::native::binary_internal::MulFunctor<float> > >(at::TensorIteratorBase&, at::native::BinaryFunctor<float, float, float, at::native::binary_internal::MulFunctor<float> > const&)::{lambda(int)#1}>(int, at::native::gpu_kernel_impl<at::native::BinaryFunctor<float, float, float, at::native::binary_internal::MulFunctor<float> > >(at::TensorIteratorBase&, at::native::BinaryFunctor<float, float, float, at::native::binary_internal::MulFunctor<float> > const&)::{lambda(int)#1})",
          2
        ],
        [
          "void at::native::elementwise_kernel<128, 2, at::native::gpu_kernel_impl<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#7}::operator()() const::{lambda(float)#1}>(at::TensorIteratorBase&, at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#7}::operator()() const::{lambda(float)#1} const&)::{lambda(int)#1}>(int, at::native::gpu_kernel_impl<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#7}::operator()() const::{lambda(float)#1}>(at::TensorIteratorBase&, at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#7}::operator()() const::{lambda(float)#1} const&)::{lambda(int)#1})",
          2
        ],
        [
          "Memcpy DtoD (Device -> Device)",
          2
        ],
        [
          "void at::native::vectorized_elementwise_kernel<4, at::native::CUDAFunctor_add<float>, at::detail::Array<char*, 3> >(int, at::native::CUDAFunctor_add<float>, at::detail::Array<char*, 3>)",
          2
        ],
        [
          "void at::native::(anonymous namespace)::CatArrayBatchedCopy_aligned16_contig<float, unsigned int, 1, 128, 1>(float*, at::native::(anonymous namespace)::CatArrInputTensorMetadata<float, unsigned int, 128, 1>, at::native::(anonymous namespace)::TensorSizeStride<unsigned int, 4u>, int, unsigned int)",
          1
        ],
        [
          "void at::native::reduce_kernel<512, 1, at::native::ReduceOp<float, at::native::NormTwoOps<float, float, float>, unsigned int, float, 4> >(at::native::ReduceOp<float, at::native::NormTwoOps<float, float, float>, unsigned int, float, 4>)",
          1
        ],
        [
          "void at::native::vectorized_elementwise_kernel<4, at::native::(anonymous namespace)::launch_clamp_scalar(at::TensorIteratorBase&, c10::Scalar, c10::Scalar, at::native::detail::ClampLimits)::{lambda()#1}::operator()() const::{lambda()#7}::operator()() const::{lambda(float)#1}, at::detail::Array<char*, 2> >(int, at::native::(anonymous namespace)::launch_clamp_scalar(at::TensorIteratorBase&, c10::Scalar, c10::Scalar, at::native::detail::ClampLimits)::{lambda()#1}::operator()() const::{lambda()#7}::operator()() const::{lambda(float)#1}, at::detail::Array<char*, 2>)",
          1
        ],
        [
          "void cutlass::Kernel<cutlass_80_tensorop_bf16_s16816gemm_bf16_64x128_64x3_tn_align2>(cutlass_80_tensorop_bf16_s16816gemm_bf16_64x128_64x3_tn_align2::Params)",
          1
        ]
      ]
    },
    {
      "shape": {
        "batch": 1,
        "length": 32768,
        "width": 1024
      },
      "architecture": "rglru",
      "backend": "fused_grouped_prefix32_hierarchical_chunk16",
      "cuda_kernel_events": 51,
      "unique_cuda_kernel_names": 22,
      "top_cuda_kernels": [
        [
          "Memset (Device)",
          10
        ],
        [
          "void at::native::unrolled_elementwise_kernel<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#12}::operator()() const::{lambda(c10::BFloat16)#1}, at::detail::Array<char*, 2>, TrivialOffsetCalculator<1, unsigned int>, TrivialOffsetCalculator<1, unsigned int>, at::native::memory::LoadWithCast<1>, at::native::memory::StoreWithCast<1> >(int, at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#12}::operator()() const::{lambda(c10::BFloat16)#1}, at::detail::Array<char*, 2>, TrivialOffsetCalculator<1, unsigned int>, TrivialOffsetCalculator<1, unsigned int>, at::native::memory::LoadWithCast<1>, at::native::memory::StoreWithCast<1>)",
          5
        ],
        [
          "void at::native::unrolled_elementwise_kernel<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#7}::operator()() const::{lambda(float)#1}, at::detail::Array<char*, 2>, TrivialOffsetCalculator<1, unsigned int>, TrivialOffsetCalculator<1, unsigned int>, at::native::memory::LoadWithCast<1>, at::native::memory::StoreWithCast<1> >(int, at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#7}::operator()() const::{lambda(float)#1}, at::detail::Array<char*, 2>, TrivialOffsetCalculator<1, unsigned int>, TrivialOffsetCalculator<1, unsigned int>, at::native::memory::LoadWithCast<1>, at::native::memory::StoreWithCast<1>)",
          4
        ],
        [
          "void at::native::reduce_kernel<128, 4, at::native::ReduceOp<float, at::native::func_wrapper_t<float, at::native::sum_functor<float, float, float>::operator()(at::TensorIterator&)::{lambda(float, float)#1}>, unsigned int, float, 4> >(at::native::ReduceOp<float, at::native::func_wrapper_t<float, at::native::sum_functor<float, float, float>::operator()(at::TensorIterator&)::{lambda(float, float)#1}>, unsigned int, float, 4>)",
          3
        ],
        [
          "sm90_xmma_gemm_bf16bf16_bf16f32_f32_nn_n_tilesize256x128x64_warpgroupsize2x1x1_execute_segment_k_off_kernel__51_cublas",
          2
        ],
        [
          "void at::native::elementwise_kernel<128, 4, at::native::gpu_kernel_impl<at::native::CUDAFunctor_add<float> >(at::TensorIteratorBase&, at::native::CUDAFunctor_add<float> const&)::{lambda(int)#2}>(int, at::native::gpu_kernel_impl<at::native::CUDAFunctor_add<float> >(at::TensorIteratorBase&, at::native::CUDAFunctor_add<float> const&)::{lambda(int)#2})",
          2
        ],
        [
          "void at::native::elementwise_kernel<128, 2, at::native::gpu_kernel_impl<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#7}::operator()() const::{lambda(float)#1}>(at::TensorIteratorBase&, at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#7}::operator()() const::{lambda(float)#1} const&)::{lambda(int)#1}>(int, at::native::gpu_kernel_impl<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#7}::operator()() const::{lambda(float)#1}>(at::TensorIteratorBase&, at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#7}::operator()() const::{lambda(float)#1} const&)::{lambda(int)#1})",
          2
        ],
        [
          "_affine_group_local",
          2
        ],
        [
          "_affine_group_outer",
          2
        ],
        [
          "_affine_group_correct",
          2
        ],
        [
          "void at::native::vectorized_elementwise_kernel<4, at::native::FillFunctor<float>, at::detail::Array<char*, 1> >(int, at::native::FillFunctor<float>, at::detail::Array<char*, 1>)",
          2
        ],
        [
          "sm90_xmma_gemm_bf16bf16_bf16f32_f32_nt_n_tilesize64x64x64_warpgroupsize1x1x1_execute_segment_k_off_kernel__51_cublas",
          2
        ],
        [
          "sm90_xmma_gemm_bf16bf16_bf16f32_f32_tn_n_tilesize256x128x64_warpgroupsize2x1x1_execute_segment_k_off_kernel__51_cublas",
          2
        ],
        [
          "void at::native::elementwise_kernel<128, 4, at::native::gpu_kernel_impl<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#12}::operator()() const::{lambda(c10::BFloat16)#1}>(at::TensorIteratorBase&, at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#12}::operator()() const::{lambda(c10::BFloat16)#1} const&)::{lambda(int)#1}>(int, at::native::gpu_kernel_impl<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#12}::operator()() const::{lambda(c10::BFloat16)#1}>(at::TensorIteratorBase&, at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#12}::operator()() const::{lambda(c10::BFloat16)#1} const&)::{lambda(int)#1})",
          2
        ],
        [
          "void at::native::vectorized_elementwise_kernel<4, at::native::CUDAFunctor_add<c10::BFloat16>, at::detail::Array<char*, 3> >(int, at::native::CUDAFunctor_add<c10::BFloat16>, at::detail::Array<char*, 3>)",
          2
        ],
        [
          "_chunk_summary",
          1
        ],
        [
          "_chunk_replay",
          1
        ],
        [
          "void at::native::vectorized_elementwise_kernel<4, at::native::BinaryFunctor<float, float, float, at::native::binary_internal::MulFunctor<float> >, at::detail::Array<char*, 3> >(int, at::native::BinaryFunctor<float, float, float, at::native::binary_internal::MulFunctor<float> >, at::detail::Array<char*, 3>)",
          1
        ],
        [
          "void at::native::reduce_kernel<512, 1, at::native::ReduceOp<float, at::native::func_wrapper_t<float, at::native::sum_functor<float, float, float>::operator()(at::TensorIterator&)::{lambda(float, float)#1}>, unsigned int, float, 4> >(at::native::ReduceOp<float, at::native::func_wrapper_t<float, at::native::sum_functor<float, float, float>::operator()(at::TensorIterator&)::{lambda(float, float)#1}>, unsigned int, float, 4>)",
          1
        ],
        [
          "void at::native::elementwise_kernel<128, 2, at::native::gpu_kernel_impl<at::native::BinaryFunctor<float, float, float, at::native::binary_internal::MulFunctor<float> > >(at::TensorIteratorBase&, at::native::BinaryFunctor<float, float, float, at::native::binary_internal::MulFunctor<float> > const&)::{lambda(int)#1}>(int, at::native::gpu_kernel_impl<at::native::BinaryFunctor<float, float, float, at::native::binary_internal::MulFunctor<float> > >(at::TensorIteratorBase&, at::native::BinaryFunctor<float, float, float, at::native::binary_internal::MulFunctor<float> > const&)::{lambda(int)#1})",
          1
        ]
      ]
    },
    {
      "shape": {
        "batch": 1,
        "length": 32768,
        "width": 1024
      },
      "architecture": "samu",
      "backend": "fused_output_fused_write_shared_sfu_grouped_prefix64_hierarchical_chunk32",
      "cuda_kernel_events": 73,
      "unique_cuda_kernel_names": 43,
      "top_cuda_kernels": [
        [
          "void at::native::vectorized_elementwise_kernel<4, at::native::FillFunctor<float>, at::detail::Array<char*, 1> >(int, at::native::FillFunctor<float>, at::detail::Array<char*, 1>)",
          8
        ],
        [
          "void at::native::unrolled_elementwise_kernel<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#7}::operator()() const::{lambda(float)#1}, at::detail::Array<char*, 2>, TrivialOffsetCalculator<1, unsigned int>, TrivialOffsetCalculator<1, unsigned int>, at::native::memory::LoadWithCast<1>, at::native::memory::StoreWithCast<1> >(int, at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#7}::operator()() const::{lambda(float)#1}, at::detail::Array<char*, 2>, TrivialOffsetCalculator<1, unsigned int>, TrivialOffsetCalculator<1, unsigned int>, at::native::memory::LoadWithCast<1>, at::native::memory::StoreWithCast<1>)",
          6
        ],
        [
          "void at::native::elementwise_kernel<128, 2, at::native::gpu_kernel_impl<at::native::BinaryFunctor<float, float, float, at::native::binary_internal::DivFunctor<float> > >(at::TensorIteratorBase&, at::native::BinaryFunctor<float, float, float, at::native::binary_internal::DivFunctor<float> > const&)::{lambda(int)#1}>(int, at::native::gpu_kernel_impl<at::native::BinaryFunctor<float, float, float, at::native::binary_internal::DivFunctor<float> > >(at::TensorIteratorBase&, at::native::BinaryFunctor<float, float, float, at::native::binary_internal::DivFunctor<float> > const&)::{lambda(int)#1})",
          5
        ],
        [
          "void at::native::unrolled_elementwise_kernel<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#12}::operator()() const::{lambda(c10::BFloat16)#1}, at::detail::Array<char*, 2>, TrivialOffsetCalculator<1, unsigned int>, TrivialOffsetCalculator<1, unsigned int>, at::native::memory::LoadWithCast<1>, at::native::memory::StoreWithCast<1> >(int, at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#12}::operator()() const::{lambda(c10::BFloat16)#1}, at::detail::Array<char*, 2>, TrivialOffsetCalculator<1, unsigned int>, TrivialOffsetCalculator<1, unsigned int>, at::native::memory::LoadWithCast<1>, at::native::memory::StoreWithCast<1>)",
          4
        ],
        [
          "void at::native::elementwise_kernel<128, 4, at::native::gpu_kernel_impl<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#12}::operator()() const::{lambda(c10::BFloat16)#1}>(at::TensorIteratorBase&, at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#12}::operator()() const::{lambda(c10::BFloat16)#1} const&)::{lambda(int)#2}>(int, at::native::gpu_kernel_impl<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#12}::operator()() const::{lambda(c10::BFloat16)#1}>(at::TensorIteratorBase&, at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#12}::operator()() const::{lambda(c10::BFloat16)#1} const&)::{lambda(int)#2})",
          2
        ],
        [
          "void at::native::vectorized_elementwise_kernel<4, at::native::exp_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#2}::operator()() const::{lambda(float)#1}, at::detail::Array<char*, 2> >(int, at::native::exp_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#2}::operator()() const::{lambda(float)#1}, at::detail::Array<char*, 2>)",
          2
        ],
        [
          "_complex_affine_group_local",
          2
        ],
        [
          "_complex_affine_group_outer",
          2
        ],
        [
          "_complex_affine_group_correct",
          2
        ],
        [
          "void at::native::vectorized_elementwise_kernel<4, at::native::BinaryFunctor<float, float, float, at::native::binary_internal::MulFunctor<float> >, at::detail::Array<char*, 3> >(int, at::native::BinaryFunctor<float, float, float, at::native::binary_internal::MulFunctor<float> >, at::detail::Array<char*, 3>)",
          2
        ],
        [
          "void at::native::reduce_kernel<512, 1, at::native::ReduceOp<float, at::native::func_wrapper_t<float, at::native::sum_functor<float, float, float>::operator()(at::TensorIterator&)::{lambda(float, float)#1}>, unsigned int, float, 4> >(at::native::ReduceOp<float, at::native::func_wrapper_t<float, at::native::sum_functor<float, float, float>::operator()(at::TensorIterator&)::{lambda(float, float)#1}>, unsigned int, float, 4>)",
          2
        ],
        [
          "void at::native::elementwise_kernel<128, 2, at::native::gpu_kernel_impl<at::native::BinaryFunctor<float, float, float, at::native::binary_internal::MulFunctor<float> > >(at::TensorIteratorBase&, at::native::BinaryFunctor<float, float, float, at::native::binary_internal::MulFunctor<float> > const&)::{lambda(int)#1}>(int, at::native::gpu_kernel_impl<at::native::BinaryFunctor<float, float, float, at::native::binary_internal::MulFunctor<float> > >(at::TensorIteratorBase&, at::native::BinaryFunctor<float, float, float, at::native::binary_internal::MulFunctor<float> > const&)::{lambda(int)#1})",
          2
        ],
        [
          "void at::native::elementwise_kernel<128, 2, at::native::gpu_kernel_impl<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#7}::operator()() const::{lambda(float)#1}>(at::TensorIteratorBase&, at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#7}::operator()() const::{lambda(float)#1} const&)::{lambda(int)#1}>(int, at::native::gpu_kernel_impl<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#7}::operator()() const::{lambda(float)#1}>(at::TensorIteratorBase&, at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#7}::operator()() const::{lambda(float)#1} const&)::{lambda(int)#1})",
          2
        ],
        [
          "Memcpy DtoD (Device -> Device)",
          2
        ],
        [
          "void at::native::vectorized_elementwise_kernel<4, at::native::CUDAFunctor_add<float>, at::detail::Array<char*, 3> >(int, at::native::CUDAFunctor_add<float>, at::detail::Array<char*, 3>)",
          2
        ],
        [
          "void at::native::(anonymous namespace)::CatArrayBatchedCopy_aligned16_contig<float, unsigned int, 1, 128, 1>(float*, at::native::(anonymous namespace)::CatArrInputTensorMetadata<float, unsigned int, 128, 1>, at::native::(anonymous namespace)::TensorSizeStride<unsigned int, 4u>, int, unsigned int)",
          1
        ],
        [
          "void at::native::reduce_kernel<512, 1, at::native::ReduceOp<float, at::native::NormTwoOps<float, float, float>, unsigned int, float, 4> >(at::native::ReduceOp<float, at::native::NormTwoOps<float, float, float>, unsigned int, float, 4>)",
          1
        ],
        [
          "void at::native::vectorized_elementwise_kernel<4, at::native::(anonymous namespace)::launch_clamp_scalar(at::TensorIteratorBase&, c10::Scalar, c10::Scalar, at::native::detail::ClampLimits)::{lambda()#1}::operator()() const::{lambda()#7}::operator()() const::{lambda(float)#1}, at::detail::Array<char*, 2> >(int, at::native::(anonymous namespace)::launch_clamp_scalar(at::TensorIteratorBase&, c10::Scalar, c10::Scalar, at::native::detail::ClampLimits)::{lambda()#1}::operator()() const::{lambda()#7}::operator()() const::{lambda(float)#1}, at::detail::Array<char*, 2>)",
          1
        ],
        [
          "void cutlass::Kernel<cutlass_80_tensorop_bf16_s16816gemm_bf16_64x128_64x3_tn_align2>(cutlass_80_tensorop_bf16_s16816gemm_bf16_64x128_64x3_tn_align2::Params)",
          1
        ],
        [
          "_formal_coordinate_forward",
          1
        ]
      ]
    },
    {
      "shape": {
        "batch": 1,
        "length": 32768,
        "width": 2048
      },
      "architecture": "rglru",
      "backend": "fused_grouped_prefix32_hierarchical_chunk16",
      "cuda_kernel_events": 51,
      "unique_cuda_kernel_names": 22,
      "top_cuda_kernels": [
        [
          "Memset (Device)",
          10
        ],
        [
          "void at::native::unrolled_elementwise_kernel<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#12}::operator()() const::{lambda(c10::BFloat16)#1}, at::detail::Array<char*, 2>, TrivialOffsetCalculator<1, unsigned int>, TrivialOffsetCalculator<1, unsigned int>, at::native::memory::LoadWithCast<1>, at::native::memory::StoreWithCast<1> >(int, at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#12}::operator()() const::{lambda(c10::BFloat16)#1}, at::detail::Array<char*, 2>, TrivialOffsetCalculator<1, unsigned int>, TrivialOffsetCalculator<1, unsigned int>, at::native::memory::LoadWithCast<1>, at::native::memory::StoreWithCast<1>)",
          5
        ],
        [
          "void at::native::unrolled_elementwise_kernel<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#7}::operator()() const::{lambda(float)#1}, at::detail::Array<char*, 2>, TrivialOffsetCalculator<1, unsigned int>, TrivialOffsetCalculator<1, unsigned int>, at::native::memory::LoadWithCast<1>, at::native::memory::StoreWithCast<1> >(int, at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#7}::operator()() const::{lambda(float)#1}, at::detail::Array<char*, 2>, TrivialOffsetCalculator<1, unsigned int>, TrivialOffsetCalculator<1, unsigned int>, at::native::memory::LoadWithCast<1>, at::native::memory::StoreWithCast<1>)",
          4
        ],
        [
          "void at::native::reduce_kernel<128, 4, at::native::ReduceOp<float, at::native::func_wrapper_t<float, at::native::sum_functor<float, float, float>::operator()(at::TensorIterator&)::{lambda(float, float)#1}>, unsigned int, float, 4> >(at::native::ReduceOp<float, at::native::func_wrapper_t<float, at::native::sum_functor<float, float, float>::operator()(at::TensorIterator&)::{lambda(float, float)#1}>, unsigned int, float, 4>)",
          3
        ],
        [
          "sm90_xmma_gemm_bf16bf16_bf16f32_f32_nn_n_tilesize256x128x64_warpgroupsize2x1x1_execute_segment_k_off_kernel__51_cublas",
          2
        ],
        [
          "void at::native::elementwise_kernel<128, 4, at::native::gpu_kernel_impl<at::native::CUDAFunctor_add<float> >(at::TensorIteratorBase&, at::native::CUDAFunctor_add<float> const&)::{lambda(int)#2}>(int, at::native::gpu_kernel_impl<at::native::CUDAFunctor_add<float> >(at::TensorIteratorBase&, at::native::CUDAFunctor_add<float> const&)::{lambda(int)#2})",
          2
        ],
        [
          "void at::native::elementwise_kernel<128, 2, at::native::gpu_kernel_impl<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#7}::operator()() const::{lambda(float)#1}>(at::TensorIteratorBase&, at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#7}::operator()() const::{lambda(float)#1} const&)::{lambda(int)#1}>(int, at::native::gpu_kernel_impl<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#7}::operator()() const::{lambda(float)#1}>(at::TensorIteratorBase&, at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#7}::operator()() const::{lambda(float)#1} const&)::{lambda(int)#1})",
          2
        ],
        [
          "_affine_group_local",
          2
        ],
        [
          "_affine_group_outer",
          2
        ],
        [
          "_affine_group_correct",
          2
        ],
        [
          "void at::native::vectorized_elementwise_kernel<4, at::native::FillFunctor<float>, at::detail::Array<char*, 1> >(int, at::native::FillFunctor<float>, at::detail::Array<char*, 1>)",
          2
        ],
        [
          "sm90_xmma_gemm_bf16bf16_bf16f32_f32_nt_n_tilesize64x64x64_warpgroupsize1x1x1_execute_segment_k_off_kernel__51_cublas",
          2
        ],
        [
          "sm90_xmma_gemm_bf16bf16_bf16f32_f32_tn_n_tilesize256x128x64_warpgroupsize2x1x1_execute_segment_k_off_kernel__51_cublas",
          2
        ],
        [
          "void at::native::elementwise_kernel<128, 4, at::native::gpu_kernel_impl<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#12}::operator()() const::{lambda(c10::BFloat16)#1}>(at::TensorIteratorBase&, at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#12}::operator()() const::{lambda(c10::BFloat16)#1} const&)::{lambda(int)#1}>(int, at::native::gpu_kernel_impl<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#12}::operator()() const::{lambda(c10::BFloat16)#1}>(at::TensorIteratorBase&, at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#12}::operator()() const::{lambda(c10::BFloat16)#1} const&)::{lambda(int)#1})",
          2
        ],
        [
          "void at::native::vectorized_elementwise_kernel<4, at::native::CUDAFunctor_add<c10::BFloat16>, at::detail::Array<char*, 3> >(int, at::native::CUDAFunctor_add<c10::BFloat16>, at::detail::Array<char*, 3>)",
          2
        ],
        [
          "_chunk_summary",
          1
        ],
        [
          "_chunk_replay",
          1
        ],
        [
          "void at::native::vectorized_elementwise_kernel<4, at::native::BinaryFunctor<float, float, float, at::native::binary_internal::MulFunctor<float> >, at::detail::Array<char*, 3> >(int, at::native::BinaryFunctor<float, float, float, at::native::binary_internal::MulFunctor<float> >, at::detail::Array<char*, 3>)",
          1
        ],
        [
          "void at::native::reduce_kernel<512, 1, at::native::ReduceOp<float, at::native::func_wrapper_t<float, at::native::sum_functor<float, float, float>::operator()(at::TensorIterator&)::{lambda(float, float)#1}>, unsigned int, float, 4> >(at::native::ReduceOp<float, at::native::func_wrapper_t<float, at::native::sum_functor<float, float, float>::operator()(at::TensorIterator&)::{lambda(float, float)#1}>, unsigned int, float, 4>)",
          1
        ],
        [
          "void at::native::elementwise_kernel<128, 2, at::native::gpu_kernel_impl<at::native::BinaryFunctor<float, float, float, at::native::binary_internal::MulFunctor<float> > >(at::TensorIteratorBase&, at::native::BinaryFunctor<float, float, float, at::native::binary_internal::MulFunctor<float> > const&)::{lambda(int)#1}>(int, at::native::gpu_kernel_impl<at::native::BinaryFunctor<float, float, float, at::native::binary_internal::MulFunctor<float> > >(at::TensorIteratorBase&, at::native::BinaryFunctor<float, float, float, at::native::binary_internal::MulFunctor<float> > const&)::{lambda(int)#1})",
          1
        ]
      ]
    },
    {
      "shape": {
        "batch": 1,
        "length": 32768,
        "width": 2048
      },
      "architecture": "samu",
      "backend": "fused_output_fused_write_shared_sfu_grouped_prefix64_hierarchical_chunk32",
      "cuda_kernel_events": 73,
      "unique_cuda_kernel_names": 43,
      "top_cuda_kernels": [
        [
          "void at::native::vectorized_elementwise_kernel<4, at::native::FillFunctor<float>, at::detail::Array<char*, 1> >(int, at::native::FillFunctor<float>, at::detail::Array<char*, 1>)",
          8
        ],
        [
          "void at::native::unrolled_elementwise_kernel<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#7}::operator()() const::{lambda(float)#1}, at::detail::Array<char*, 2>, TrivialOffsetCalculator<1, unsigned int>, TrivialOffsetCalculator<1, unsigned int>, at::native::memory::LoadWithCast<1>, at::native::memory::StoreWithCast<1> >(int, at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#7}::operator()() const::{lambda(float)#1}, at::detail::Array<char*, 2>, TrivialOffsetCalculator<1, unsigned int>, TrivialOffsetCalculator<1, unsigned int>, at::native::memory::LoadWithCast<1>, at::native::memory::StoreWithCast<1>)",
          6
        ],
        [
          "void at::native::elementwise_kernel<128, 2, at::native::gpu_kernel_impl<at::native::BinaryFunctor<float, float, float, at::native::binary_internal::DivFunctor<float> > >(at::TensorIteratorBase&, at::native::BinaryFunctor<float, float, float, at::native::binary_internal::DivFunctor<float> > const&)::{lambda(int)#1}>(int, at::native::gpu_kernel_impl<at::native::BinaryFunctor<float, float, float, at::native::binary_internal::DivFunctor<float> > >(at::TensorIteratorBase&, at::native::BinaryFunctor<float, float, float, at::native::binary_internal::DivFunctor<float> > const&)::{lambda(int)#1})",
          5
        ],
        [
          "void at::native::unrolled_elementwise_kernel<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#12}::operator()() const::{lambda(c10::BFloat16)#1}, at::detail::Array<char*, 2>, TrivialOffsetCalculator<1, unsigned int>, TrivialOffsetCalculator<1, unsigned int>, at::native::memory::LoadWithCast<1>, at::native::memory::StoreWithCast<1> >(int, at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#12}::operator()() const::{lambda(c10::BFloat16)#1}, at::detail::Array<char*, 2>, TrivialOffsetCalculator<1, unsigned int>, TrivialOffsetCalculator<1, unsigned int>, at::native::memory::LoadWithCast<1>, at::native::memory::StoreWithCast<1>)",
          4
        ],
        [
          "void at::native::elementwise_kernel<128, 4, at::native::gpu_kernel_impl<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#12}::operator()() const::{lambda(c10::BFloat16)#1}>(at::TensorIteratorBase&, at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#12}::operator()() const::{lambda(c10::BFloat16)#1} const&)::{lambda(int)#2}>(int, at::native::gpu_kernel_impl<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#12}::operator()() const::{lambda(c10::BFloat16)#1}>(at::TensorIteratorBase&, at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#12}::operator()() const::{lambda(c10::BFloat16)#1} const&)::{lambda(int)#2})",
          2
        ],
        [
          "void at::native::vectorized_elementwise_kernel<4, at::native::exp_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#2}::operator()() const::{lambda(float)#1}, at::detail::Array<char*, 2> >(int, at::native::exp_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#2}::operator()() const::{lambda(float)#1}, at::detail::Array<char*, 2>)",
          2
        ],
        [
          "_complex_affine_group_local",
          2
        ],
        [
          "_complex_affine_group_outer",
          2
        ],
        [
          "_complex_affine_group_correct",
          2
        ],
        [
          "void at::native::vectorized_elementwise_kernel<4, at::native::BinaryFunctor<float, float, float, at::native::binary_internal::MulFunctor<float> >, at::detail::Array<char*, 3> >(int, at::native::BinaryFunctor<float, float, float, at::native::binary_internal::MulFunctor<float> >, at::detail::Array<char*, 3>)",
          2
        ],
        [
          "void at::native::reduce_kernel<512, 1, at::native::ReduceOp<float, at::native::func_wrapper_t<float, at::native::sum_functor<float, float, float>::operator()(at::TensorIterator&)::{lambda(float, float)#1}>, unsigned int, float, 4> >(at::native::ReduceOp<float, at::native::func_wrapper_t<float, at::native::sum_functor<float, float, float>::operator()(at::TensorIterator&)::{lambda(float, float)#1}>, unsigned int, float, 4>)",
          2
        ],
        [
          "void at::native::elementwise_kernel<128, 2, at::native::gpu_kernel_impl<at::native::BinaryFunctor<float, float, float, at::native::binary_internal::MulFunctor<float> > >(at::TensorIteratorBase&, at::native::BinaryFunctor<float, float, float, at::native::binary_internal::MulFunctor<float> > const&)::{lambda(int)#1}>(int, at::native::gpu_kernel_impl<at::native::BinaryFunctor<float, float, float, at::native::binary_internal::MulFunctor<float> > >(at::TensorIteratorBase&, at::native::BinaryFunctor<float, float, float, at::native::binary_internal::MulFunctor<float> > const&)::{lambda(int)#1})",
          2
        ],
        [
          "void at::native::elementwise_kernel<128, 2, at::native::gpu_kernel_impl<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#7}::operator()() const::{lambda(float)#1}>(at::TensorIteratorBase&, at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#7}::operator()() const::{lambda(float)#1} const&)::{lambda(int)#1}>(int, at::native::gpu_kernel_impl<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#7}::operator()() const::{lambda(float)#1}>(at::TensorIteratorBase&, at::native::direct_copy_kernel_cuda(at::TensorIteratorBase&)::{lambda()#2}::operator()() const::{lambda()#7}::operator()() const::{lambda(float)#1} const&)::{lambda(int)#1})",
          2
        ],
        [
          "Memcpy DtoD (Device -> Device)",
          2
        ],
        [
          "void at::native::vectorized_elementwise_kernel<4, at::native::CUDAFunctor_add<float>, at::detail::Array<char*, 3> >(int, at::native::CUDAFunctor_add<float>, at::detail::Array<char*, 3>)",
          2
        ],
        [
          "void at::native::(anonymous namespace)::CatArrayBatchedCopy_aligned16_contig<float, unsigned int, 1, 128, 1>(float*, at::native::(anonymous namespace)::CatArrInputTensorMetadata<float, unsigned int, 128, 1>, at::native::(anonymous namespace)::TensorSizeStride<unsigned int, 4u>, int, unsigned int)",
          1
        ],
        [
          "void at::native::reduce_kernel<512, 1, at::native::ReduceOp<float, at::native::NormTwoOps<float, float, float>, unsigned int, float, 4> >(at::native::ReduceOp<float, at::native::NormTwoOps<float, float, float>, unsigned int, float, 4>)",
          1
        ],
        [
          "void at::native::vectorized_elementwise_kernel<4, at::native::(anonymous namespace)::launch_clamp_scalar(at::TensorIteratorBase&, c10::Scalar, c10::Scalar, at::native::detail::ClampLimits)::{lambda()#1}::operator()() const::{lambda()#7}::operator()() const::{lambda(float)#1}, at::detail::Array<char*, 2> >(int, at::native::(anonymous namespace)::launch_clamp_scalar(at::TensorIteratorBase&, c10::Scalar, c10::Scalar, at::native::detail::ClampLimits)::{lambda()#1}::operator()() const::{lambda()#7}::operator()() const::{lambda(float)#1}, at::detail::Array<char*, 2>)",
          1
        ],
        [
          "void cutlass::Kernel<cutlass_80_tensorop_bf16_s16816gemm_bf16_128x128_64x3_tn_align2>(cutlass_80_tensorop_bf16_s16816gemm_bf16_128x128_64x3_tn_align2::Params)",
          1
        ],
        [
          "_formal_coordinate_forward",
          1
        ]
      ]
    }
  ]
}