-------------------------------------------------------  ------------  ------------  ------------  ------------  ------------  ------------  ------------  ------------  ------------  ------------  ------------  ------------  ------------  ------------  --------------------------------------------------------------------------------  ------------  
                                                   Name    Self CPU %      Self CPU   CPU total %     CPU total  CPU time avg     Self CUDA   Self CUDA %    CUDA total  CUDA time avg       CPU Mem  Self CPU Mem      CUDA Mem  Self CUDA Mem    # of Calls                                                                      Input Shapes   Total FLOPs  
-------------------------------------------------------  ------------  ------------  ------------  ------------  ------------  ------------  ------------  ------------  ------------  ------------  ------------  ------------  ------------  ------------  --------------------------------------------------------------------------------  ------------  
                                          ProfilerStep*         0.00%       0.000us         0.00%       0.000us       0.000us     471.697ms       448.06%     471.697ms      26.205ms           0 B           0 B           0 B           0 B            18                                                                                []            --  
                                          ProfilerStep*       -81.92%  -287776.088us        93.41%     328.123ms     109.374ms       0.000us         0.00%     105.276ms      35.092ms           0 B           0 B           0 B      -3.35 GB             3                                                                                []            --  
                                       aten::_cudnn_rnn         2.11%       7.396ms         7.95%      27.921ms       4.654ms      31.732ms        30.14%      69.179ms      11.530ms           0 B           0 B       1.44 GB      -4.78 GB             6  [[8192, 68, 200], [], [], [588672], [2, 8192, 224], [], [], [], [], [], [], [],             --  
                                         aten::_to_copy       -75.18%  -264071.485us        84.12%     295.489ms      49.248ms       0.000us         0.00%      58.026ms       9.671ms           0 B           0 B       1.87 GB    -426.00 MB             6                                         [[8192, 68, 200], [], [], [], [], [], []]            --  
                                            aten::copy_         0.02%      67.002us         0.05%     184.487us      30.748us      58.026ms        55.12%      58.026ms       9.671ms           0 B           0 B           0 B           0 B             6                                            [[8192, 68, 200], [8192, 68, 200], []]            --  
                         Memcpy HtoD (Pinned -> Device)         0.00%       0.000us         0.00%       0.000us       0.000us      55.855ms        53.06%      55.855ms       6.206ms           0 B           0 B           0 B           0 B             9                                                                                []            --  
                                               aten::to         0.04%     143.865us        84.14%     295.560ms      49.260ms       0.000us         0.00%      55.836ms       9.306ms           0 B           0 B       1.25 GB           0 B             6                                     [[8192, 68, 200], [], [], [], [], [], [], []]            --  
                                              aten::gru         0.07%     239.873us         4.37%      15.351ms       5.117ms       0.000us         0.00%      35.722ms      11.907ms           0 B           0 B     735.00 MB      -3.37 MB             3                     [[8192, 68, 200], [2, 8192, 224], [], [], [], [], [], [], []]            --  
ampere_fp16_s1688gemm_fp16_128x128_ldg8_f2f_stages_3...         0.00%       0.000us         0.00%       0.000us       0.000us      11.542ms        10.96%      11.542ms      28.288us           0 B           0 B           0 B           0 B           408                                                                                []            --  
void elemWiseRNNcell<__half, __half, float, (cudnnRN...         0.00%       0.000us         0.00%       0.000us       0.000us      10.449ms         9.93%      10.449ms      25.611us           0 B           0 B           0 B           0 B           408                                                                                []            --  
ampere_fp16_s1688gemm_fp16_256x64_ldg8_f2f_stages_32...         0.00%       0.000us         0.00%       0.000us       0.000us       9.741ms         9.25%       9.741ms      95.501us           0 B           0 B           0 B           0 B           102                                                                                []            --  
                                             aten::sub_         0.02%      78.057us         0.03%     111.517us      37.172us       2.934ms         2.79%       2.934ms     978.006us           0 B           0 B           0 B           0 B             3                                                [[8192, 68, 200], [1, 1, 200], []]            --  
void at::native::elementwise_kernel<128, 2, at::nati...         0.00%       0.000us         0.00%       0.000us       0.000us       2.934ms         2.79%       2.934ms     978.006us           0 B           0 B           0 B           0 B             3                                                                                []            --  
                                             aten::mul_         0.01%      41.282us         0.02%      64.647us      21.549us       2.913ms         2.77%       2.913ms     970.902us           0 B           0 B           0 B           0 B             3                                                    [[8192, 68, 200], [1, 1, 200]]            --  
void at::native::elementwise_kernel<128, 2, at::nati...         0.00%       0.000us         0.00%       0.000us       0.000us       2.913ms         2.77%       2.913ms     970.902us           0 B           0 B           0 B           0 B             3                                                                                []            --  
void at::native::vectorized_elementwise_kernel<4, at...         0.00%       0.000us         0.00%       0.000us       0.000us       2.875ms         2.73%       2.875ms     239.589us           0 B           0 B           0 B           0 B            12                                                                                []            --  
                                      aten::nan_to_num_         0.01%      17.918us         0.04%     142.895us      47.632us       0.000us         0.00%       2.865ms     954.987us           0 B           0 B           0 B           0 B             3                                                     [[8192, 68, 200], [], [], []]            --  
                                       aten::nan_to_num         0.01%      42.372us         0.04%     124.977us      41.659us       2.865ms         2.72%       2.865ms     954.987us           0 B           0 B           0 B           0 B             3                                    [[8192, 68, 200], [], [], [], [8192, 68, 200]]            --  
                                              aten::mul         0.02%      62.463us         0.03%      96.795us      32.265us       2.811ms         2.67%       2.811ms     937.057us           0 B           0 B       1.25 GB       1.25 GB             3                                                 [[8192, 68, 200], [8192, 1, 200]]  334233600.000  
void at::native::elementwise_kernel<128, 4, at::nati...         0.00%       0.000us         0.00%       0.000us       0.000us       2.811ms         2.67%       2.811ms     937.057us           0 B           0 B           0 B           0 B             3                                                                                []            --  
void at::native::vectorized_elementwise_kernel<4, at...         0.00%       0.000us         0.00%       0.000us       0.000us       2.260ms         2.15%       2.260ms      75.331us           0 B           0 B           0 B           0 B            30                                                                                []            --  
                                               aten::to         0.00%       8.691us         0.02%      81.622us      13.604us       0.000us         0.00%       2.189ms     364.897us           0 B           0 B     637.50 MB           0 B             6                                                 [[8192, 68, 200], [], [], [], []]            --  
                                       aten::contiguous         0.00%      11.358us         0.04%     127.141us      42.380us       0.000us         0.00%       1.725ms     574.938us           0 B           0 B     637.50 MB           0 B             3                                                             [[68, 8192, 200], []]            --  
                                            aten::clone         0.01%      22.752us         0.03%     115.783us      38.594us       0.000us         0.00%       1.725ms     574.938us           0 B           0 B     637.50 MB           0 B             3                                                             [[68, 8192, 200], []]            --  
                                            aten::copy_         0.01%      42.335us         0.02%      69.575us      23.192us       1.725ms         1.64%       1.725ms     574.938us           0 B           0 B           0 B           0 B             3                                            [[68, 8192, 200], [68, 8192, 200], []]            --  
void at::native::elementwise_kernel<128, 4, at::nati...         0.00%       0.000us         0.00%       0.000us       0.000us       1.725ms         1.64%       1.725ms     574.938us           0 B           0 B           0 B           0 B             3                                                                                []            --  
                                             aten::mean         0.03%     118.949us         0.05%     163.926us      54.642us       1.611ms         1.53%       1.611ms     537.136us           0 B           0 B      18.75 MB      18.75 MB             3                                                     [[8192, 68, 200], [], [], []]            --  
void at::native::reduce_kernel<128, 4, at::native::R...         0.00%       0.000us         0.00%       0.000us       0.000us       1.611ms         1.53%       1.611ms     537.136us           0 B           0 B           0 B           0 B             3                                                                                []            --  
                                              aten::sum         0.24%     849.628us         0.34%       1.195ms      26.549us     115.937us         0.11%     115.937us       2.576us           0 B           0 B      22.50 KB      22.50 KB            45                                                                      [[8192], []]            --  
void at::native::reduce_kernel<512, 1, at::native::R...         0.00%       0.000us         0.00%       0.000us       0.000us     115.937us         0.11%     115.937us       2.576us           0 B           0 B           0 B           0 B            45                                                                                []            --  
                                       aten::layer_norm         0.02%      55.294us         0.18%     645.277us     107.546us       0.000us         0.00%      97.824us      16.304us           0 B           0 B      42.00 MB     -21.19 MB             6                                           [[8192, 224], [], [224], [224], [], []]            --  
void at::native::(anonymous namespace)::vectorized_l...         0.00%       0.000us         0.00%       0.000us       0.000us      82.016us         0.08%      82.016us      13.669us           0 B           0 B           0 B           0 B             6                                                                                []            --  
                                         aten::_to_copy         0.17%     611.941us         0.36%       1.249ms      59.459us       0.000us         0.00%      73.820us       3.515us           0 B           0 B     864.00 KB           0 B            21                                                  [[8192], [], [], [], [], [], []]            --  
                                            aten::copy_         0.04%     143.689us         0.14%     499.241us      23.773us      73.820us         0.07%      73.820us       3.515us           0 B           0 B           0 B           0 B            21                                                              [[8192], [8192], []]            --  
                                           aten::linear         0.03%     118.808us         0.32%       1.111ms     185.213us       0.000us         0.00%      62.047us      10.341us           0 B           0 B      14.41 MB     -10.50 MB             6                                                  [[8192, 224], [112, 224], [112]]            --  
                                               aten::to         0.18%     617.897us         0.46%       1.610ms      89.464us       0.000us         0.00%      55.675us       3.093us           0 B           0 B     576.00 KB           0 B            18                                              [[8192], [], [], [], [], [], [], []]            --  
                                              aten::mul         0.14%     500.781us         0.24%     844.656us      17.597us      52.835us         0.05%      52.835us       1.101us           0 B           0 B       1.36 MB       1.36 MB            48                                                                  [[8192], [8192]]    393216.000  
void at::native::vectorized_elementwise_kernel<4, at...         0.00%       0.000us         0.00%       0.000us       0.000us      50.878us         0.05%      50.878us       1.696us           0 B           0 B           0 B           0 B            30                                                                                []            --  
void at::native::vectorized_elementwise_kernel<4, at...         0.00%       0.000us         0.00%       0.000us       0.000us      46.466us         0.04%      46.466us       1.106us           0 B           0 B           0 B           0 B            42                                                                                []            --  
                                           aten::linear         0.03%     107.832us         0.31%       1.087ms     181.148us       0.000us         0.00%      44.125us       7.354us           0 B           0 B       3.04 MB      -9.38 MB             6                                                    [[8192, 200], [32, 200], [32]]            --  
-------------------------------------------------------  ------------  ------------  ------------  ------------  ------------  ------------  ------------  ------------  ------------  ------------  ------------  ------------  ------------  ------------  --------------------------------------------------------------------------------  ------------  
Self CPU time total: 351.275ms
Self CUDA time total: 105.276ms
