source stringclasses 13
values | gpu stringclasses 4
values | model stringclasses 13
values | kernel_family stringclasses 7
values | kernel_name stringclasses 328
values | dtype stringclasses 2
values | held_out bool 2
classes | M float64 1 16.4k ⌀ | N float64 8 248k ⌀ | K float64 2.05k 29.6k ⌀ | bs float64 1 1 ⌀ | seq float64 1 16.4k ⌀ | n_heads float64 8 64 ⌀ | head_dim float64 64 256 ⌀ | kv_heads float64 1 64 ⌀ | numel float64 64 531M ⌀ | op_type stringclasses 177
values | gpu_time_duration_ms float64 0 703 | launch_block_size float64 16 512 | launch_grid_size float64 1 1.04M | dram_bytes_sum float64 1.27 94.3B ⌀ | launch_registers_per_thread float64 16 255 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::vectorized_elementwise_kernel<4, at::CUDAFunctor_add<float>, std::array<char *, 3>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 654,336 | residual | 0.00842 | 128 | 768 | 6,300,000 | 32 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::BinaryFunctor<float, float, float, binary_internal::MulFunctor<float>>>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 654,336 | mul | 0.00627 | 128 | 768 | 1,580,000 | 16 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::<unnamed>::masked_fill_kernel(at::TensorIterator &, const c10::Scalar &)::[lambda() (instance 1)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float, bool) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 654,336 | other | 0.00598 | 128 | 768 | 801,540 | 16 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::CUDAFunctor_add<float>>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 655,360 | residual | 0.00752 | 128 | 1,536 | 3,160,000 | 16 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::exp_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 2)]::operator ()() const::[lambda() (instance 2)]::operator ()() const::[lambda(float) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 6,144 | other | 0.00512 | 128 | 12 | 22,660 | 16 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::BinaryFunctor<float, float, float, binary_internal::MulFunctor<float>>>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 655,360 | mul | 0.00707 | 128 | 1,536 | 1,600,000 | 16 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::vectorized_elementwise_kernel<4, at::CUDAFunctor_add<float>, std::array<char *, 3>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 218,112 | residual | 0.00643 | 128 | 384 | 3,150,000 | 32 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | cast | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 655,360 | cast | 0.00685 | 128 | 1,536 | 1,580,000 | 16 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::exp_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 2)]::operator ()() const::[lambda() (instance 2)]::operator ()() const::[lambda(float) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 512 | other | 0.00403 | 128 | 1 | 13,440 | 16 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::BinaryFunctor<float, float, float, binary_internal::MulFunctor<float>>>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 1,310,720 | mul | 0.00915 | 128 | 3,072 | 3,160,000 | 16 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::CUDAFunctor_add<float>>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 6,144 | residual | 0.00506 | 128 | 12 | 23,420 | 16 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::vectorized_elementwise_kernel<4, at::exp_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 2)]::operator ()() const::[lambda() (instance 2)]::operator ()() const::[lambda(float) (instance 1)], std::array<char *, 2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 1,536 | other | 0.00352 | 128 | 3 | 19,840 | 28 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::BinaryFunctor<float, float, float, binary_internal::MulFunctor<float>>>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 655,360 | mul | 0.00694 | 128 | 1,536 | 1,600,000 | 16 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::vectorized_elementwise_kernel<4, at::CUDAFunctor_add<float>, std::array<char *, 3>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 654,336 | residual | 0.0087 | 128 | 768 | 6,300,000 | 32 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | cast | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 1,310,720 | cast | 0.00941 | 128 | 3,072 | 3,160,000 | 16 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | cast | void at::vectorized_elementwise_kernel<4, at::bfloat16_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda(float) (instance 1)], std::array<char *, 2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 654,336 | cast | 0.00624 | 128 | 768 | 3,150,000 | 28 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::vectorized_elementwise_kernel<4, at::FillFunctor<c10::BFloat16>, std::array<char *, 1>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 654,336 | fill | 0.00378 | 128 | 768 | 4,220 | 16 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | cast | void at::vectorized_elementwise_kernel<4, at::bfloat16_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda(float) (instance 1)], std::array<char *, 2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 654,336 | cast | 0.00618 | 128 | 768 | 3,150,000 | 28 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | cast | void at::unrolled_elementwise_kernel<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)], std::array<char *, 2>, 4, TrivialOffsetCalculator<1, unsigned int>, TrivialOffsetCalculator<1, unsigned in... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 655,360 | cast | 0.00909 | 128 | 1,536 | 1,610,000 | 32 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::vectorized_elementwise_kernel<4, void at::<unnamed>::pow_tensor_scalar_kernel_impl<float, float>(at::TensorIteratorBase &, T2)::[lambda(float) (instance 1)], std::array<char *, 2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 654,336 | rmsnorm | 0.00669 | 128 | 768 | 3,150,000 | 28 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | reduce | void at::reduce_kernel<512, 1, at::ReduceOp<float, at::MeanOps<float, float, float, float>, unsigned int, float, 4, 4>>(T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 655,360 | rmsnorm | 0.00957 | 512 | 384 | 3,200,000 | 32 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::vectorized_elementwise_kernel<4, at::CUDAFunctorOnSelf_add<float>, std::array<char *, 2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 3,072 | residual | 0.00349 | 128 | 6 | 29,310 | 30 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::vectorized_elementwise_kernel<4, at::rsqrt_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 2)]::operator ()() const::[lambda() (instance 2)]::operator ()() const::[lambda(float) (instance 1)], std::array<char *, 2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 3,072 | rmsnorm | 0.00358 | 128 | 6 | 30,980 | 22 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::BinaryFunctor<float, float, float, binary_internal::MulFunctor<float>>>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 1,310,720 | mul | 0.00909 | 128 | 3,072 | 3,180,000 | 16 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | cast | void at::vectorized_elementwise_kernel<4, at::bfloat16_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda(float) (instance 1)], std::array<char *, 2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 654,336 | cast | 0.00627 | 128 | 768 | 3,150,000 | 28 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::elementwise_kernel<128, 4, void at::gpu_kernel_impl_nocast<at::BinaryFunctor<c10::BFloat16, c10::BFloat16, c10::BFloat16, binary_internal::MulFunctor<float>>>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 655,360 | mul | 0.00874 | 128 | 1,536 | 1,590,000 | 16 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | cast | void at::unrolled_elementwise_kernel<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)], std::array<char *, 2>, 4, TrivialOffsetCalculator<1, unsigned int>, TrivialOffsetCalculator<1, unsigned in... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 655,360 | cast | 0.00922 | 128 | 1,536 | 1,610,000 | 32 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::vectorized_elementwise_kernel<4, at::<unnamed>::silu_kernel(at::TensorIteratorBase &)::[lambda() (instance 1)]::operator ()() const::[lambda() (instance 2)]::operator ()() const::[lambda(float) (instance 1)], std::array<char *, 2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 654,336 | silu | 0.00771 | 128 | 768 | 3,160,000 | 31 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::unrolled_elementwise_kernel<at::BinaryFunctor<float, float, float, binary_internal::MulFunctor<float>>, std::array<char *, 3>, 4, TrivialOffsetCalculator<2, unsigned int>, TrivialOffsetCalculator<1, unsigned int>, memory::LoadWithCast<2>, memory::StoreWithCast<1>>(int, T1, T2, T4, T5, T6, T7) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 655,360 | mul | 0.01101 | 128 | 1,536 | 4,760,000 | 32 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | cast | void at::vectorized_elementwise_kernel<4, at::bfloat16_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda(float) (instance 1)], std::array<char *, 2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 654,336 | cast | 0.00618 | 128 | 768 | 3,150,000 | 28 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::vectorized_elementwise_kernel<4, at::CUDAFunctor_add<c10::BFloat16>, std::array<char *, 3>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 218,112 | residual | 0.00582 | 128 | 640 | 2,630,000 | 32 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | cast | void at::unrolled_elementwise_kernel<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)], std::array<char *, 2>, 4, TrivialOffsetCalculator<1, unsigned int>, TrivialOffsetCalculator<1, unsigned in... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 655,360 | cast | 0.00893 | 128 | 1,280 | 1,340,000 | 32 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::vectorized_elementwise_kernel<4, void at::<unnamed>::pow_tensor_scalar_kernel_impl<float, float>(at::TensorIteratorBase &, T2)::[lambda(float) (instance 1)], std::array<char *, 2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 218,112 | rmsnorm | 0.00634 | 128 | 640 | 2,630,000 | 28 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | reduce | void at::reduce_kernel<512, 1, at::ReduceOp<float, at::MeanOps<float, float, float, float>, unsigned int, float, 4, 4>>(T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 16,384 | rmsnorm | 0.03139 | 512 | 8 | 2,670,000 | 32 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::vectorized_elementwise_kernel<4, at::CUDAFunctorOnSelf_add<float>, std::array<char *, 2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 512 | residual | 0.0033 | 128 | 1 | 5,250 | 30 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::vectorized_elementwise_kernel<4, at::rsqrt_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 2)]::operator ()() const::[lambda() (instance 2)]::operator ()() const::[lambda(float) (instance 1)], std::array<char *, 2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 512 | rmsnorm | 0.00346 | 128 | 1 | 6,780 | 22 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::BinaryFunctor<float, float, float, binary_internal::MulFunctor<float>>>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 1,310,720 | mul | 0.00835 | 128 | 2,560 | 2,630,000 | 16 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | cast | void at::unrolled_elementwise_kernel<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)], std::array<char *, 2>, 4, TrivialOffsetCalculator<1, unsigned int>, TrivialOffsetCalculator<1, unsigned in... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 5,120 | cast | 0.00758 | 128 | 10 | 42,370 | 32 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::vectorized_elementwise_kernel<4, at::CUDAFunctorOnSelf_add<float>, std::array<char *, 2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 2,560 | residual | 0.00362 | 128 | 5 | 25,220 | 30 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::BinaryFunctor<float, float, float, binary_internal::MulFunctor<float>>>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 1,310,720 | mul | 0.00861 | 128 | 2,560 | 2,650,000 | 16 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | cast | void at::vectorized_elementwise_kernel<4, at::bfloat16_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda(float) (instance 1)], std::array<char *, 2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 218,112 | cast | 0.00589 | 128 | 640 | 2,630,000 | 28 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::vectorized_elementwise_kernel<4, at::<unnamed>::silu_kernel(at::TensorIteratorBase &)::[lambda() (instance 1)]::operator ()() const::[lambda() (instance 6)]::operator ()() const::[lambda(c10::BFloat16) (instance 1)], std::array<char *, 2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 1,310,720 | silu | 0.01066 | 128 | 2,176 | 4,470,000 | 28 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::vectorized_elementwise_kernel<4, at::BinaryFunctor<c10::BFloat16, c10::BFloat16, c10::BFloat16, binary_internal::MulFunctor<float>>, std::array<char *, 3>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 1,310,720 | mul | 0.0103 | 128 | 2,176 | 8,920,000 | 32 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::vectorized_elementwise_kernel<4, at::CUDAFunctor_add<c10::BFloat16>, std::array<char *, 3>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 218,112 | residual | 0.00589 | 128 | 640 | 2,630,000 | 32 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | cast | void at::unrolled_elementwise_kernel<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)], std::array<char *, 2>, 4, TrivialOffsetCalculator<1, unsigned int>, TrivialOffsetCalculator<1, unsigned in... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 655,360 | cast | 0.01139 | 128 | 1,280 | 1,340,000 | 32 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::vectorized_elementwise_kernel<4, void at::<unnamed>::pow_tensor_scalar_kernel_impl<float, float>(at::TensorIteratorBase &, T2)::[lambda(float) (instance 1)], std::array<char *, 2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 218,112 | rmsnorm | 0.0063 | 128 | 640 | 2,630,000 | 28 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | reduce | void at::reduce_kernel<512, 1, at::ReduceOp<float, at::MeanOps<float, float, float, float>, unsigned int, float, 4, 4>>(T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 16,384 | rmsnorm | 0.032 | 512 | 8 | 2,670,000 | 32 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::vectorized_elementwise_kernel<4, at::CUDAFunctorOnSelf_add<float>, std::array<char *, 2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 512 | residual | 0.00339 | 128 | 1 | 5,250 | 30 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::vectorized_elementwise_kernel<4, at::rsqrt_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 2)]::operator ()() const::[lambda() (instance 2)]::operator ()() const::[lambda(float) (instance 1)], std::array<char *, 2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 512 | rmsnorm | 0.00419 | 128 | 1 | 7,170 | 22 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::BinaryFunctor<float, float, float, binary_internal::MulFunctor<float>>>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 1,310,720 | mul | 0.00899 | 128 | 2,560 | 2,640,000 | 16 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | cast | void at::unrolled_elementwise_kernel<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)], std::array<char *, 2>, 4, TrivialOffsetCalculator<1, unsigned int>, TrivialOffsetCalculator<1, unsigned in... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 5,120 | cast | 0.00826 | 128 | 10 | 42,620 | 32 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::vectorized_elementwise_kernel<4, at::CUDAFunctorOnSelf_add<float>, std::array<char *, 2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 2,560 | residual | 0.00358 | 128 | 5 | 25,220 | 30 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::BinaryFunctor<float, float, float, binary_internal::MulFunctor<float>>>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 1,310,720 | mul | 0.00858 | 128 | 2,560 | 2,650,000 | 16 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | cast | void at::vectorized_elementwise_kernel<4, at::bfloat16_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda(float) (instance 1)], std::array<char *, 2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 218,112 | cast | 0.00576 | 128 | 640 | 2,630,000 | 28 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | cast | void at::elementwise_kernel<128, 4, void at::gpu_kernel_impl_nocast<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 12)]::operator ()() const::[lambda(c10::BFloat16) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 655,360 | cast | 0.00829 | 128 | 1,536 | 1,590,000 | 16 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | cast | void at::elementwise_kernel<128, 4, void at::gpu_kernel_impl<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 655,360 | cast | 0.01059 | 128 | 1,536 | 1,620,000 | 32 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::vectorized_elementwise_kernel<4, void at::<unnamed>::pow_tensor_scalar_kernel_impl<float, float>(at::TensorIteratorBase &, T2)::[lambda(float) (instance 1)], std::array<char *, 2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 654,336 | rmsnorm | 0.00685 | 128 | 768 | 3,150,000 | 28 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | reduce | void at::reduce_kernel<512, 1, at::ReduceOp<float, at::MeanOps<float, float, float, float>, unsigned int, float, 4, 4>>(T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 654,336 | rmsnorm | 0.0089 | 512 | 192 | 3,200,000 | 32 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::vectorized_elementwise_kernel<4, at::CUDAFunctorOnSelf_add<float>, std::array<char *, 2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 1,536 | residual | 0.00387 | 128 | 3 | 17,020 | 30 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::vectorized_elementwise_kernel<4, at::rsqrt_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 2)]::operator ()() const::[lambda() (instance 2)]::operator ()() const::[lambda(float) (instance 1)], std::array<char *, 2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 1,536 | rmsnorm | 0.00378 | 128 | 3 | 19,330 | 22 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::BinaryFunctor<float, float, float, binary_internal::MulFunctor<float>>>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 1,310,720 | mul | 0.00912 | 128 | 3,072 | 3,170,000 | 16 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | cast | void at::unrolled_elementwise_kernel<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)], std::array<char *, 2>, 4, TrivialOffsetCalculator<1, unsigned int>, TrivialOffsetCalculator<1, unsigned in... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 512 | cast | 0.00611 | 128 | 1 | 32,640 | 32 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::vectorized_elementwise_kernel<4, at::CUDAFunctorOnSelf_add<float>, std::array<char *, 2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 512 | residual | 0.00358 | 128 | 1 | 5,760 | 30 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::BinaryFunctor<float, float, float, binary_internal::MulFunctor<float>>>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 1,310,720 | mul | 0.00909 | 128 | 3,072 | 3,160,000 | 16 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | cast | void at::vectorized_elementwise_kernel<4, at::bfloat16_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda(float) (instance 1)], std::array<char *, 2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 654,336 | cast | 0.00624 | 128 | 768 | 3,150,000 | 28 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | cast | void at::unrolled_elementwise_kernel<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)], std::array<char *, 2>, 4, TrivialOffsetCalculator<1, unsigned int>, TrivialOffsetCalculator<1, unsigned in... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 218,112 | cast | 0.0079 | 128 | 256 | 295,550 | 32 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::vectorized_elementwise_kernel<4, void at::<unnamed>::pow_tensor_scalar_kernel_impl<float, float>(at::TensorIteratorBase &, T2)::[lambda(float) (instance 1)], std::array<char *, 2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 65,536 | rmsnorm | 0.00394 | 128 | 128 | 529,020 | 28 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | reduce | void at::reduce_kernel<512, 1, at::ReduceOp<float, at::MeanOps<float, float, float, float>, unsigned int, float, 4, 4>>(T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 65,536 | rmsnorm | 0.00733 | 512 | 32 | 575,490 | 32 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::vectorized_elementwise_kernel<4, at::CUDAFunctorOnSelf_add<float>, std::array<char *, 2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 512 | residual | 0.00394 | 128 | 1 | 6,780 | 30 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::vectorized_elementwise_kernel<4, at::rsqrt_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 2)]::operator ()() const::[lambda() (instance 2)]::operator ()() const::[lambda(float) (instance 1)], std::array<char *, 2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 512 | rmsnorm | 0.00426 | 128 | 1 | 8,450 | 22 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::BinaryFunctor<float, float, float, binary_internal::MulFunctor<float>>>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 218,112 | mul | 0.00586 | 128 | 512 | 537,340 | 16 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | cast | void at::unrolled_elementwise_kernel<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)], std::array<char *, 2>, 4, TrivialOffsetCalculator<1, unsigned int>, TrivialOffsetCalculator<1, unsigned in... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 512 | cast | 0.00605 | 128 | 1 | 32,770 | 32 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::vectorized_elementwise_kernel<4, at::CUDAFunctorOnSelf_add<float>, std::array<char *, 2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 512 | residual | 0.00362 | 128 | 1 | 5,760 | 30 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::BinaryFunctor<float, float, float, binary_internal::MulFunctor<float>>>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 218,112 | mul | 0.00589 | 128 | 512 | 536,320 | 16 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | cast | void at::vectorized_elementwise_kernel<4, at::bfloat16_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda(float) (instance 1)], std::array<char *, 2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 65,536 | cast | 0.00387 | 128 | 128 | 529,280 | 28 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::elementwise_kernel<128, 4, void at::gpu_kernel_impl_nocast<at::BinaryFunctor<c10::BFloat16, c10::BFloat16, c10::BFloat16, binary_internal::MulFunctor<float>>>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 218,112 | mul | 0.00758 | 128 | 384 | 429,700 | 16 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::elementwise_kernel<128, 4, void at::gpu_kernel_impl_nocast<at::neg_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 2)]::operator ()() const::[lambda() (instance 9)]::operator ()() const::[lambda(c10::BFloat16) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 98,304 | neg | 0.00704 | 128 | 192 | 214,400 | 16 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | copy | void at::<unnamed>::CatArrayBatchedCopy<at::<unnamed>::OpaqueType<2>, unsigned int, 4, 64, 64>(T1 *, at::<unnamed>::CatArrInputTensorMetadata<T1, T2, T4, T5>, at::<unnamed>::TensorSizeStride<T2, 4>, int, T2) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 655,360 | other | 0.01005 | 512 | 432 | 400,130 | 23 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::elementwise_kernel<128, 4, void at::gpu_kernel_impl_nocast<at::BinaryFunctor<c10::BFloat16, c10::BFloat16, c10::BFloat16, binary_internal::MulFunctor<float>>>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 218,112 | mul | 0.0073 | 128 | 384 | 429,440 | 16 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::elementwise_kernel<128, 4, void at::gpu_kernel_impl_nocast<at::CUDAFunctor_add<c10::BFloat16>>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 218,112 | residual | 0.00765 | 128 | 384 | 806,270 | 16 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::elementwise_kernel<128, 4, void at::gpu_kernel_impl_nocast<at::BinaryFunctor<c10::BFloat16, c10::BFloat16, c10::BFloat16, binary_internal::MulFunctor<float>>>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 32,768 | mul | 0.00717 | 128 | 64 | 101,630 | 16 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::elementwise_kernel<128, 4, void at::gpu_kernel_impl_nocast<at::neg_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 2)]::operator ()() const::[lambda() (instance 9)]::operator ()() const::[lambda(c10::BFloat16) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 16,384 | neg | 0.00672 | 128 | 32 | 50,560 | 16 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | copy | void at::<unnamed>::CatArrayBatchedCopy<at::<unnamed>::OpaqueType<2>, unsigned int, 4, 64, 64>(T1 *, at::<unnamed>::CatArrInputTensorMetadata<T1, T2, T4, T5>, at::<unnamed>::TensorSizeStride<T2, 4>, int, T2) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 655,360 | other | 0.00864 | 512 | 432 | 72,450 | 23 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::elementwise_kernel<128, 4, void at::gpu_kernel_impl_nocast<at::BinaryFunctor<c10::BFloat16, c10::BFloat16, c10::BFloat16, binary_internal::MulFunctor<float>>>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 32,768 | mul | 0.00704 | 128 | 64 | 101,630 | 16 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::elementwise_kernel<128, 4, void at::gpu_kernel_impl_nocast<at::CUDAFunctor_add<c10::BFloat16>>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 32,768 | residual | 0.00707 | 128 | 64 | 150,780 | 16 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | copy | void at::<unnamed>::CatArrayBatchedCopy<at::<unnamed>::OpaqueType<2>, unsigned int, 4, 64, 64>(T1 *, at::<unnamed>::CatArrInputTensorMetadata<T1, T2, T4, T5>, at::<unnamed>::TensorSizeStride<T2, 4>, int, T2) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 655,360 | other | 0.01542 | 512 | 432 | 1,580,000 | 23 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | copy | void at::<unnamed>::CatArrayBatchedCopy<at::<unnamed>::OpaqueType<2>, unsigned int, 4, 64, 64>(T1 *, at::<unnamed>::CatArrInputTensorMetadata<T1, T2, T4, T5>, at::<unnamed>::TensorSizeStride<T2, 4>, int, T2) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 655,360 | other | 0.00976 | 512 | 432 | 269,180 | 23 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | copy | void at::<unnamed>::CatArrayBatchedCopy<at::<unnamed>::OpaqueType<2>, unsigned int, 4, 64, 64>(T1 *, at::<unnamed>::CatArrInputTensorMetadata<T1, T2, T4, T5>, at::<unnamed>::TensorSizeStride<T2, 4>, int, T2) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 655,360 | other | 0.00874 | 512 | 432 | 269,060 | 23 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | copy | void at::<unnamed>::CatArrayBatchedCopy<at::<unnamed>::OpaqueType<2>, unsigned int, 4, 64, 64>(T1 *, at::<unnamed>::CatArrInputTensorMetadata<T1, T2, T4, T5>, at::<unnamed>::TensorSizeStride<T2, 4>, int, T2) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 655,360 | other | 0.0087 | 512 | 432 | 269,060 | 23 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | cast | void at::elementwise_kernel<128, 4, void at::gpu_kernel_impl_nocast<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 12)]::operator ()() const::[lambda(c10::BFloat16) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 655,360 | cast | 0.00883 | 128 | 1,536 | 1,590,000 | 16 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::vectorized_elementwise_kernel<4, at::sigmoid_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 2)]::operator ()() const::[lambda() (instance 4)]::operator ()() const::[lambda(c10::BFloat16) (instance 1)], std::array<char *, 2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 654,336 | other | 0.00669 | 128 | 768 | 1,590,000 | 30 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::vectorized_elementwise_kernel<4, at::BinaryFunctor<c10::BFloat16, c10::BFloat16, c10::BFloat16, binary_internal::MulFunctor<float>>, std::array<char *, 3>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 654,336 | mul | 0.00634 | 128 | 768 | 3,150,000 | 32 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::vectorized_elementwise_kernel<4, at::CUDAFunctor_add<c10::BFloat16>, std::array<char *, 3>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 218,112 | residual | 0.00608 | 128 | 640 | 2,630,000 | 32 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | cast | void at::unrolled_elementwise_kernel<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)], std::array<char *, 2>, 4, TrivialOffsetCalculator<1, unsigned int>, TrivialOffsetCalculator<1, unsigned in... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 655,360 | cast | 0.00912 | 128 | 1,280 | 1,340,000 | 32 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::vectorized_elementwise_kernel<4, void at::<unnamed>::pow_tensor_scalar_kernel_impl<float, float>(at::TensorIteratorBase &, T2)::[lambda(float) (instance 1)], std::array<char *, 2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 218,112 | rmsnorm | 0.00627 | 128 | 640 | 2,630,000 | 28 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | reduce | void at::reduce_kernel<512, 1, at::ReduceOp<float, at::MeanOps<float, float, float, float>, unsigned int, float, 4, 4>>(T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 16,384 | rmsnorm | 0.03178 | 512 | 8 | 2,670,000 | 32 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::vectorized_elementwise_kernel<4, at::CUDAFunctorOnSelf_add<float>, std::array<char *, 2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 512 | residual | 0.00349 | 128 | 1 | 5,250 | 30 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::vectorized_elementwise_kernel<4, at::rsqrt_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 2)]::operator ()() const::[lambda() (instance 2)]::operator ()() const::[lambda(float) (instance 1)], std::array<char *, 2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 512 | rmsnorm | 0.00368 | 128 | 1 | 6,780 | 22 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | elementwise | void at::elementwise_kernel<128, 2, void at::gpu_kernel_impl_nocast<at::BinaryFunctor<float, float, float, binary_internal::MulFunctor<float>>>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 1,310,720 | mul | 0.00835 | 128 | 2,560 | 2,630,000 | 16 |
Qwen3.5-27B_prefill | A100 | Qwen3.5-27B | cast | void at::unrolled_elementwise_kernel<at::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 7)]::operator ()() const::[lambda(float) (instance 1)], std::array<char *, 2>, 4, TrivialOffsetCalculator<1, unsigned int>, TrivialOffsetCalculator<1, unsigned in... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 5,120 | cast | 0.0079 | 128 | 10 | 42,370 | 32 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.