:github_url: https://github.com/pytorch/pytorch

.. _exhale_function_namespaceat_1_1symint_1a9b922c4c7055ae6989bd4aacc2d52d24:

Template Function at::symint::_efficient_attention_backward(const at::Tensor&, const at::Tensor&, const at::Tensor&, const at::Tensor&, const ::std::optional<at::Tensor>&, const at::Tensor&, const ::std::optional<at::Tensor>&, const ::std::optional<at::Tensor>&, c10::SymInt, c10::SymInt, const at::Tensor&, double, const at::Tensor&, const at::Tensor&, int64_t, bool, ::std::optional<double>, ::std::optional<int64_t>, ::std::optional<int64_t>, bool)
===================================================================================================================================================================================================================================================================================================================================================================================================================================================================

- Defined in :ref:`file_build_aten_src_ATen_Functions.h`


Function Documentation
----------------------


.. doxygenfunction:: at::symint::_efficient_attention_backward(const at::Tensor&, const at::Tensor&, const at::Tensor&, const at::Tensor&, const ::std::optional<at::Tensor>&, const at::Tensor&, const ::std::optional<at::Tensor>&, const ::std::optional<at::Tensor>&, c10::SymInt, c10::SymInt, const at::Tensor&, double, const at::Tensor&, const at::Tensor&, int64_t, bool, ::std::optional<double>, ::std::optional<int64_t>, ::std::optional<int64_t>, bool)