#0  at::autocast::(anonymous namespace)::is_autocast_eligible (tensor=..., device_type=c10::DeviceType::CUDA)
    at /home/colin/develop/pytorch/aten/src/ATen/autocast_mode.h:24
#1  0x00007fffb9799261 in at::autocast::is_eligible (arg=..., device_type=c10::DeviceType::CUDA)
    at /home/colin/develop/pytorch/aten/src/ATen/autocast_mode.h:136
#2  0x00007fffb978dcec in at::autocast::cached_cast (to_type=c10::ScalarType::Half, arg=..., 
    device_type=c10::DeviceType::CUDA) at /home/colin/develop/pytorch/aten/src/ATen/autocast_mode.cpp:112
#3  0x00007fffb979d5d2 in at::autocast::WrapFunction_<(at::autocast::CastPolicy)0, (c10::DeviceType)1, at::Tensor (at::Tensor const&, at::Tensor const&), &at::mm, at::Tensor, c10::guts::typelist::typelist<at::Tensor const&, at::Tensor const&> >::call(at::Tensor const&, at::Tensor const&) (args#0=..., args#1=...)
    at /home/colin/develop/pytorch/aten/src/ATen/autocast_mode.cpp:174
#4  0x00007fffb96a9e2f in c10::impl::detail::WrapFunctionIntoRuntimeFunctor_<at::Tensor (*)(at::Tensor const&, at::Tensor const&), at::Tensor, c10::guts::typelist::typelist<at::Tensor const&, at::Tensor const&> >::operator() (
    this=0x555555f60280, args#0=..., args#1=...)
    at /home/colin/develop/pytorch/aten/src/ATen/core/boxing/impl/WrapFunctionIntoRuntimeFunctor.h:18
#5  0x00007fffb96b1463 in c10::impl::wrap_kernel_functor_unboxed_<c10::impl::detail::WrapFunctionIntoRuntimeFunctor_<at::Tensor (*)(at::Tensor const&, at::Tensor const&), at::Tensor, c10::guts::typelist::typelist<at::Tensor const&, at::Tensor const&> >, at::Tensor (at::Tensor const&, at::Tensor const&)>::call(c10::OperatorKernel*, c10::DispatchKeySet, at::Tensor const&, at::Tensor const&) (functor=0x555555f60280, args#0=..., args#1=...)
    at /home/colin/develop/pytorch/aten/src/ATen/core/boxing/impl/make_boxed_from_unboxed_functor.h:424
#6  0x00007fffba3aeb3c in c10::callUnboxedKernelFunction<at::Tensor, at::Tensor const&, at::Tensor const&> (
    unboxed_kernel_func=0x7fffb96b13fa <c10::impl::wrap_kernel_functor_unboxed_<c10::impl::detail::WrapFunctionIntoRuntimeFunctor_<at::Tensor (*)(at::Tensor const&, at::Tensor const&), at::Tensor, c10::guts::typelist::typelist<at::Tensor const&, at::Tensor const&> >, at::Tensor (at::Tensor const&, at::Tensor const&)>::call(c10::OperatorKernel*, c10::DispatchKeySet, at::Tensor const&, at::Tensor const&)>, functor=0x555555f60280, dispatchKeySet=...)
    at /home/colin/develop/pytorch/aten/src/ATen/core/boxing/KernelFunction_impl.h:57
#7  0x00007fffba86a05d in c10::KernelFunction::call<at::Tensor, at::Tensor const&, at::Tensor const&> (
    dispatchKeySet=..., opHandle=..., this=0x555555f50088)
    at /home/colin/develop/pytorch/aten/src/ATen/core/boxing/KernelFunction_impl.h:67
#8  c10::Dispatcher::call<at::Tensor, at::Tensor const&, at::Tensor const&>(c10::TypedOperatorHandle<at::Tensor (at::Tensor const&, at::Tensor const&)> const&, at::Tensor const&, at::Tensor const&) const (op=..., 
    this=0x7fffc7690520 <c10::Dispatcher::realSingleton()::_singleton>)
    at /home/colin/develop/pytorch/aten/src/ATen/core/dispatch/Dispatcher.h:548
#9  c10::TypedOperatorHandle<at::Tensor (at::Tensor const&, at::Tensor const&)>::call(at::Tensor const&, at::Tensor const&) const (args#1=..., args#0=..., this=<optimized out>)
    at /home/colin/develop/pytorch/aten/src/ATen/core/dispatch/Dispatcher.h:414
#10 at::_ops::mm::call (self=..., mat2=...) at /home/colin/develop/pytorch/build/aten/src/ATen/Operators_3.cpp:3328
#11 0x00007fffc85c1f2c in at::Tensor::mm (this=0x7fffffffdb00, mat2=...)
    at /home/colin/develop/pytorch/build/aten/src/ATen/core/TensorBody.h:2787
#12 0x00007fffc868d046 in torch::autograd::<lambda(const at::Tensor&, const at::Tensor&)>::operator()(const at::Tensor &, const at::Tensor &) const (__closure=0x7fffffffdaf8, self=..., mat2=...)