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 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
gpt-oss-20b_prefill | H100 | gpt-oss-20b | cast | void at::native::elementwise_kernel<(int)128, (int)4, void at::native::gpu_kernel_impl_nocast<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 12)]::operator ()() const::[lambda(c10::BFloat16) (instance 1)]>(at::TensorIteratorBase &, const ... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 265,420,800 | cast | 2.16 | 128 | 518,400 | 1,050,000,000 | 18 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | cast | void at::native::unrolled_elementwise_kernel<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 3)]::operator ()() const::[lambda(int) (instance 1)], std::array<char *, (unsigned long)2>, (int)4, TrivialOffsetCalculator<(int)1, unsigned int>,... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 16,777,216 | cast | 0.03562 | 128 | 32,400 | 58,050,000 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::vectorized_elementwise_kernel<(int)4, at::native::CUDAFunctorOnSelf_add<int>, std::array<char *, (unsigned long)2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 8,294,400 | residual | 0.0408 | 128 | 16,200 | 109,260,000 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::vectorized_elementwise_kernel<(int)4, at::native::AUnaryFunctor<unsigned char, unsigned char, unsigned char, at::native::BitwiseAndFunctor<unsigned char>>, std::array<char *, (unsigned long)2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 132,710,400 | other | 0.18611 | 128 | 259,200 | 510,830,000 | 30 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | cast | void at::native::unrolled_elementwise_kernel<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 3)]::operator ()() const::[lambda(int) (instance 1)], std::array<char *, (unsigned long)2>, (int)4, TrivialOffsetCalculator<(int)1, unsigned int>,... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 265,420,800 | cast | 0.52438 | 128 | 518,400 | 1,300,000,000 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | cast | void at::native::unrolled_elementwise_kernel<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 4)]::operator ()() const::[lambda(long) (instance 1)], std::array<char *, (unsigned long)2>, (int)4, TrivialOffsetCalculator<(int)1, unsigned int>... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 265,420,800 | cast | 1.15 | 128 | 518,400 | 3,160,000,000 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | copy | void at::native::index_elementwise_kernel<(int)128, (int)4, void at::native::gpu_index_kernel<void at::native::index_kernel_impl<at::native::OpaqueType<(int)2>>(at::TensorIteratorBase &, c10::ArrayRef<long>, c10::ArrayRef<long>)::[lambda(char *, const char *, long) (instance 1)]>(at::TensorIteratorBase &, c10::ArrayRef... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 265,420,800 | other | 1.34 | 128 | 518,400 | 2,650,000,000 | 36 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | cast | void at::native::elementwise_kernel<(int)128, (int)4, void at::native::gpu_kernel_impl_nocast<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 12)]::operator ()() const::[lambda(c10::BFloat16) (instance 1)]>(at::TensorIteratorBase &, const ... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 265,420,800 | cast | 0.93517 | 128 | 518,400 | 2,640,000,000 | 18 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::vectorized_elementwise_kernel<(int)4, at::native::BUnaryFunctor<unsigned char, unsigned char, unsigned char, at::native::rshift_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 1)]::operator ()() const::[lambda() (instance 1)]::operator ()() const::[lambda(unsigned char, unsigned char) (insta... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 132,710,400 | other | 0.18794 | 128 | 259,200 | 511,160,000 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | cast | void at::native::unrolled_elementwise_kernel<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 3)]::operator ()() const::[lambda(int) (instance 1)], std::array<char *, (unsigned long)2>, (int)4, TrivialOffsetCalculator<(int)1, unsigned int>,... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 265,420,800 | cast | 0.5209 | 128 | 518,400 | 1,300,000,000 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | cast | void at::native::unrolled_elementwise_kernel<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 4)]::operator ()() const::[lambda(long) (instance 1)], std::array<char *, (unsigned long)2>, (int)4, TrivialOffsetCalculator<(int)1, unsigned int>... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 265,420,800 | cast | 1.15 | 128 | 518,400 | 3,160,000,000 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | copy | void at::native::index_elementwise_kernel<(int)128, (int)4, void at::native::gpu_index_kernel<void at::native::index_kernel_impl<at::native::OpaqueType<(int)2>>(at::TensorIteratorBase &, c10::ArrayRef<long>, c10::ArrayRef<long>)::[lambda(char *, const char *, long) (instance 1)]>(at::TensorIteratorBase &, c10::ArrayRef... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 265,420,800 | other | 1.34 | 128 | 518,400 | 2,650,000,000 | 36 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | cast | void at::native::elementwise_kernel<(int)128, (int)4, void at::native::gpu_kernel_impl_nocast<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 12)]::operator ()() const::[lambda(c10::BFloat16) (instance 1)]>(at::TensorIteratorBase &, const ... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 265,420,800 | cast | 0.93197 | 128 | 518,400 | 2,640,000,000 | 18 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::vectorized_elementwise_kernel<(int)8, at::native::FillFunctor<c10::BFloat16>, std::array<char *, (unsigned long)1>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 512 | fill | 0.00298 | 128 | 1 | 3,840 | 16 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::elementwise_kernel<(int)128, (int)4, void at::native::gpu_kernel_impl<at::native::<unnamed>::pow_tensor_tensor_kernel(at::TensorIteratorBase &)::[lambda() (instance 1)]::operator ()() const::[lambda() (instance 11)]::operator ()() const::[lambda(c10::BFloat16, c10::BFloat16) (instance 1)]>(at::TensorIt... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 16,777,216 | other | 0.13325 | 128 | 32,400 | 87,910,000 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::elementwise_kernel<(int)128, (int)4, void at::native::gpu_kernel_impl_nocast<at::native::BinaryFunctor<c10::BFloat16, c10::BFloat16, c10::BFloat16, at::native::binary_internal::MulFunctor<float>>>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 530,841,600 | mul | 1.65 | 128 | 1,036,800 | 2,130,000,000 | 18 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | cast | void at::native::elementwise_kernel<(int)128, (int)4, void at::native::gpu_kernel_impl_nocast<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 12)]::operator ()() const::[lambda(c10::BFloat16) (instance 1)]>(at::TensorIteratorBase &, const ... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 530,841,600 | cast | 4.33 | 128 | 1,036,800 | 2,110,000,000 | 18 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | cast | void at::native::unrolled_elementwise_kernel<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 3)]::operator ()() const::[lambda(int) (instance 1)], std::array<char *, (unsigned long)2>, (int)4, TrivialOffsetCalculator<(int)1, unsigned int>,... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 8,294,400 | cast | 0.02003 | 128 | 16,200 | 16,410,000 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::vectorized_elementwise_kernel<(int)4, at::native::CUDAFunctorOnSelf_add<int>, std::array<char *, (unsigned long)2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 4,147,200 | residual | 0.01974 | 128 | 8,100 | 43,790,000 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::vectorized_elementwise_kernel<(int)4, at::native::AUnaryFunctor<unsigned char, unsigned char, unsigned char, at::native::BitwiseAndFunctor<unsigned char>>, std::array<char *, (unsigned long)2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 66,355,200 | other | 0.09232 | 128 | 129,600 | 244,650,000 | 30 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | cast | void at::native::unrolled_elementwise_kernel<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 3)]::operator ()() const::[lambda(int) (instance 1)], std::array<char *, (unsigned long)2>, (int)4, TrivialOffsetCalculator<(int)1, unsigned int>,... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 132,710,400 | cast | 0.2648 | 128 | 259,200 | 638,740,000 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | cast | void at::native::unrolled_elementwise_kernel<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 4)]::operator ()() const::[lambda(long) (instance 1)], std::array<char *, (unsigned long)2>, (int)4, TrivialOffsetCalculator<(int)1, unsigned int>... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 132,710,400 | cast | 0.57693 | 128 | 259,200 | 1,570,000,000 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | copy | void at::native::index_elementwise_kernel<(int)128, (int)4, void at::native::gpu_index_kernel<void at::native::index_kernel_impl<at::native::OpaqueType<(int)2>>(at::TensorIteratorBase &, c10::ArrayRef<long>, c10::ArrayRef<long>)::[lambda(char *, const char *, long) (instance 1)]>(at::TensorIteratorBase &, c10::ArrayRef... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 132,710,400 | other | 0.67466 | 128 | 259,200 | 1,320,000,000 | 36 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | cast | void at::native::elementwise_kernel<(int)128, (int)4, void at::native::gpu_kernel_impl_nocast<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 12)]::operator ()() const::[lambda(c10::BFloat16) (instance 1)]>(at::TensorIteratorBase &, const ... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 132,710,400 | cast | 0.46858 | 128 | 259,200 | 1,310,000,000 | 18 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::vectorized_elementwise_kernel<(int)4, at::native::BUnaryFunctor<unsigned char, unsigned char, unsigned char, at::native::rshift_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 1)]::operator ()() const::[lambda() (instance 1)]::operator ()() const::[lambda(unsigned char, unsigned char) (insta... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 66,355,200 | other | 0.09376 | 128 | 129,600 | 245,170,000 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | cast | void at::native::unrolled_elementwise_kernel<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 3)]::operator ()() const::[lambda(int) (instance 1)], std::array<char *, (unsigned long)2>, (int)4, TrivialOffsetCalculator<(int)1, unsigned int>,... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 132,710,400 | cast | 0.26573 | 128 | 259,200 | 639,110,000 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | cast | void at::native::unrolled_elementwise_kernel<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 4)]::operator ()() const::[lambda(long) (instance 1)], std::array<char *, (unsigned long)2>, (int)4, TrivialOffsetCalculator<(int)1, unsigned int>... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 132,710,400 | cast | 0.57712 | 128 | 259,200 | 1,570,000,000 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | copy | void at::native::index_elementwise_kernel<(int)128, (int)4, void at::native::gpu_index_kernel<void at::native::index_kernel_impl<at::native::OpaqueType<(int)2>>(at::TensorIteratorBase &, c10::ArrayRef<long>, c10::ArrayRef<long>)::[lambda(char *, const char *, long) (instance 1)]>(at::TensorIteratorBase &, c10::ArrayRef... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 132,710,400 | other | 0.67168 | 128 | 259,200 | 1,320,000,000 | 36 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | cast | void at::native::elementwise_kernel<(int)128, (int)4, void at::native::gpu_kernel_impl_nocast<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 12)]::operator ()() const::[lambda(c10::BFloat16) (instance 1)]>(at::TensorIteratorBase &, const ... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 132,710,400 | cast | 0.46752 | 128 | 259,200 | 1,310,000,000 | 18 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::vectorized_elementwise_kernel<(int)8, at::native::FillFunctor<c10::BFloat16>, std::array<char *, (unsigned long)1>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 512 | fill | 0.00298 | 128 | 1 | 4,100 | 16 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::elementwise_kernel<(int)128, (int)4, void at::native::gpu_kernel_impl<at::native::<unnamed>::pow_tensor_tensor_kernel(at::TensorIteratorBase &)::[lambda() (instance 1)]::operator ()() const::[lambda() (instance 11)]::operator ()() const::[lambda(c10::BFloat16, c10::BFloat16) (instance 1)]>(at::TensorIt... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 8,294,400 | other | 0.06867 | 128 | 16,200 | 38,370,000 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::elementwise_kernel<(int)128, (int)4, void at::native::gpu_kernel_impl_nocast<at::native::BinaryFunctor<c10::BFloat16, c10::BFloat16, c10::BFloat16, at::native::binary_internal::MulFunctor<float>>>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 265,420,800 | mul | 0.82675 | 128 | 518,400 | 1,060,000,000 | 18 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | cast | void at::native::elementwise_kernel<(int)128, (int)4, void at::native::gpu_kernel_impl_nocast<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 12)]::operator ()() const::[lambda(c10::BFloat16) (instance 1)]>(at::TensorIteratorBase &, const ... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 265,420,800 | cast | 2.16 | 128 | 518,400 | 1,050,000,000 | 18 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | cast | void at::native::unrolled_elementwise_kernel<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 3)]::operator ()() const::[lambda(int) (instance 1)], std::array<char *, (unsigned long)2>, (int)4, TrivialOffsetCalculator<(int)1, unsigned int>,... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 16,777,216 | cast | 0.03552 | 128 | 32,400 | 57,830,000 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::vectorized_elementwise_kernel<(int)4, at::native::CUDAFunctorOnSelf_add<int>, std::array<char *, (unsigned long)2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 8,294,400 | residual | 0.04205 | 128 | 16,200 | 109,550,000 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::vectorized_elementwise_kernel<(int)4, at::native::AUnaryFunctor<unsigned char, unsigned char, unsigned char, at::native::BitwiseAndFunctor<unsigned char>>, std::array<char *, (unsigned long)2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 132,710,400 | other | 0.18624 | 128 | 259,200 | 510,770,000 | 30 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | cast | void at::native::unrolled_elementwise_kernel<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 3)]::operator ()() const::[lambda(int) (instance 1)], std::array<char *, (unsigned long)2>, (int)4, TrivialOffsetCalculator<(int)1, unsigned int>,... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 265,420,800 | cast | 0.52483 | 128 | 518,400 | 1,300,000,000 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | cast | void at::native::unrolled_elementwise_kernel<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 4)]::operator ()() const::[lambda(long) (instance 1)], std::array<char *, (unsigned long)2>, (int)4, TrivialOffsetCalculator<(int)1, unsigned int>... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 265,420,800 | cast | 1.15 | 128 | 518,400 | 3,160,000,000 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | copy | void at::native::index_elementwise_kernel<(int)128, (int)4, void at::native::gpu_index_kernel<void at::native::index_kernel_impl<at::native::OpaqueType<(int)2>>(at::TensorIteratorBase &, c10::ArrayRef<long>, c10::ArrayRef<long>)::[lambda(char *, const char *, long) (instance 1)]>(at::TensorIteratorBase &, c10::ArrayRef... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 265,420,800 | other | 1.34 | 128 | 518,400 | 2,650,000,000 | 36 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | cast | void at::native::elementwise_kernel<(int)128, (int)4, void at::native::gpu_kernel_impl_nocast<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 12)]::operator ()() const::[lambda(c10::BFloat16) (instance 1)]>(at::TensorIteratorBase &, const ... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 265,420,800 | cast | 0.9352 | 128 | 518,400 | 2,640,000,000 | 18 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::vectorized_elementwise_kernel<(int)4, at::native::BUnaryFunctor<unsigned char, unsigned char, unsigned char, at::native::rshift_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 1)]::operator ()() const::[lambda() (instance 1)]::operator ()() const::[lambda(unsigned char, unsigned char) (insta... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 132,710,400 | other | 0.18941 | 128 | 259,200 | 511,410,000 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | cast | void at::native::unrolled_elementwise_kernel<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 3)]::operator ()() const::[lambda(int) (instance 1)], std::array<char *, (unsigned long)2>, (int)4, TrivialOffsetCalculator<(int)1, unsigned int>,... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 265,420,800 | cast | 0.52355 | 128 | 518,400 | 1,300,000,000 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | cast | void at::native::unrolled_elementwise_kernel<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 4)]::operator ()() const::[lambda(long) (instance 1)], std::array<char *, (unsigned long)2>, (int)4, TrivialOffsetCalculator<(int)1, unsigned int>... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 265,420,800 | cast | 1.15 | 128 | 518,400 | 3,160,000,000 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | copy | void at::native::index_elementwise_kernel<(int)128, (int)4, void at::native::gpu_index_kernel<void at::native::index_kernel_impl<at::native::OpaqueType<(int)2>>(at::TensorIteratorBase &, c10::ArrayRef<long>, c10::ArrayRef<long>)::[lambda(char *, const char *, long) (instance 1)]>(at::TensorIteratorBase &, c10::ArrayRef... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 265,420,800 | other | 1.34 | 128 | 518,400 | 2,650,000,000 | 36 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | cast | void at::native::elementwise_kernel<(int)128, (int)4, void at::native::gpu_kernel_impl_nocast<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 12)]::operator ()() const::[lambda(c10::BFloat16) (instance 1)]>(at::TensorIteratorBase &, const ... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 265,420,800 | cast | 0.93222 | 128 | 518,400 | 2,640,000,000 | 18 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::vectorized_elementwise_kernel<(int)8, at::native::FillFunctor<c10::BFloat16>, std::array<char *, (unsigned long)1>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 512 | fill | 0.00314 | 128 | 1 | 4,100 | 16 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::elementwise_kernel<(int)128, (int)4, void at::native::gpu_kernel_impl<at::native::<unnamed>::pow_tensor_tensor_kernel(at::TensorIteratorBase &)::[lambda() (instance 1)]::operator ()() const::[lambda() (instance 11)]::operator ()() const::[lambda(c10::BFloat16, c10::BFloat16) (instance 1)]>(at::TensorIt... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 16,777,216 | other | 0.13261 | 128 | 32,400 | 87,870,000 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::elementwise_kernel<(int)128, (int)4, void at::native::gpu_kernel_impl_nocast<at::native::BinaryFunctor<c10::BFloat16, c10::BFloat16, c10::BFloat16, at::native::binary_internal::MulFunctor<float>>>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 530,841,600 | mul | 1.65 | 128 | 1,036,800 | 2,130,000,000 | 18 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | cast | void at::native::elementwise_kernel<(int)128, (int)4, void at::native::gpu_kernel_impl_nocast<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 12)]::operator ()() const::[lambda(c10::BFloat16) (instance 1)]>(at::TensorIteratorBase &, const ... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 530,841,600 | cast | 4.33 | 128 | 1,036,800 | 2,110,000,000 | 18 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | cast | void at::native::unrolled_elementwise_kernel<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 3)]::operator ()() const::[lambda(int) (instance 1)], std::array<char *, (unsigned long)2>, (int)4, TrivialOffsetCalculator<(int)1, unsigned int>,... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 8,294,400 | cast | 0.01974 | 128 | 16,200 | 16,510,000 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::vectorized_elementwise_kernel<(int)4, at::native::CUDAFunctorOnSelf_add<int>, std::array<char *, (unsigned long)2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 4,147,200 | residual | 0.01994 | 128 | 8,100 | 43,930,000 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::vectorized_elementwise_kernel<(int)4, at::native::AUnaryFunctor<unsigned char, unsigned char, unsigned char, at::native::BitwiseAndFunctor<unsigned char>>, std::array<char *, (unsigned long)2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 66,355,200 | other | 0.09296 | 128 | 129,600 | 245,480,000 | 30 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | cast | void at::native::unrolled_elementwise_kernel<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 3)]::operator ()() const::[lambda(int) (instance 1)], std::array<char *, (unsigned long)2>, (int)4, TrivialOffsetCalculator<(int)1, unsigned int>,... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 132,710,400 | cast | 0.26826 | 128 | 259,200 | 638,360,000 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | cast | void at::native::unrolled_elementwise_kernel<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 4)]::operator ()() const::[lambda(long) (instance 1)], std::array<char *, (unsigned long)2>, (int)4, TrivialOffsetCalculator<(int)1, unsigned int>... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 132,710,400 | cast | 0.57651 | 128 | 259,200 | 1,570,000,000 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | copy | void at::native::index_elementwise_kernel<(int)128, (int)4, void at::native::gpu_index_kernel<void at::native::index_kernel_impl<at::native::OpaqueType<(int)2>>(at::TensorIteratorBase &, c10::ArrayRef<long>, c10::ArrayRef<long>)::[lambda(char *, const char *, long) (instance 1)]>(at::TensorIteratorBase &, c10::ArrayRef... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 132,710,400 | other | 0.67386 | 128 | 259,200 | 1,320,000,000 | 36 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | cast | void at::native::elementwise_kernel<(int)128, (int)4, void at::native::gpu_kernel_impl_nocast<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 12)]::operator ()() const::[lambda(c10::BFloat16) (instance 1)]>(at::TensorIteratorBase &, const ... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 132,710,400 | cast | 0.46784 | 128 | 259,200 | 1,310,000,000 | 18 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::vectorized_elementwise_kernel<(int)4, at::native::BUnaryFunctor<unsigned char, unsigned char, unsigned char, at::native::rshift_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 1)]::operator ()() const::[lambda() (instance 1)]::operator ()() const::[lambda(unsigned char, unsigned char) (insta... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 66,355,200 | other | 0.09389 | 128 | 129,600 | 245,860,000 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | cast | void at::native::unrolled_elementwise_kernel<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 3)]::operator ()() const::[lambda(int) (instance 1)], std::array<char *, (unsigned long)2>, (int)4, TrivialOffsetCalculator<(int)1, unsigned int>,... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 132,710,400 | cast | 0.26691 | 128 | 259,200 | 638,770,000 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | cast | void at::native::unrolled_elementwise_kernel<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 4)]::operator ()() const::[lambda(long) (instance 1)], std::array<char *, (unsigned long)2>, (int)4, TrivialOffsetCalculator<(int)1, unsigned int>... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 132,710,400 | cast | 0.57706 | 128 | 259,200 | 1,570,000,000 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | copy | void at::native::index_elementwise_kernel<(int)128, (int)4, void at::native::gpu_index_kernel<void at::native::index_kernel_impl<at::native::OpaqueType<(int)2>>(at::TensorIteratorBase &, c10::ArrayRef<long>, c10::ArrayRef<long>)::[lambda(char *, const char *, long) (instance 1)]>(at::TensorIteratorBase &, c10::ArrayRef... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 132,710,400 | other | 0.67373 | 128 | 259,200 | 1,320,000,000 | 36 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | cast | void at::native::elementwise_kernel<(int)128, (int)4, void at::native::gpu_kernel_impl_nocast<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 12)]::operator ()() const::[lambda(c10::BFloat16) (instance 1)]>(at::TensorIteratorBase &, const ... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 132,710,400 | cast | 0.46797 | 128 | 259,200 | 1,310,000,000 | 18 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::vectorized_elementwise_kernel<(int)8, at::native::FillFunctor<c10::BFloat16>, std::array<char *, (unsigned long)1>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 512 | fill | 0.00294 | 128 | 1 | 3,840 | 16 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::elementwise_kernel<(int)128, (int)4, void at::native::gpu_kernel_impl<at::native::<unnamed>::pow_tensor_tensor_kernel(at::TensorIteratorBase &)::[lambda() (instance 1)]::operator ()() const::[lambda() (instance 11)]::operator ()() const::[lambda(c10::BFloat16, c10::BFloat16) (instance 1)]>(at::TensorIt... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 8,294,400 | other | 0.06842 | 128 | 16,200 | 38,570,000 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::elementwise_kernel<(int)128, (int)4, void at::native::gpu_kernel_impl_nocast<at::native::BinaryFunctor<c10::BFloat16, c10::BFloat16, c10::BFloat16, at::native::binary_internal::MulFunctor<float>>>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 265,420,800 | mul | 0.82698 | 128 | 518,400 | 1,060,000,000 | 18 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | cast | void at::native::elementwise_kernel<(int)128, (int)4, void at::native::gpu_kernel_impl_nocast<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 12)]::operator ()() const::[lambda(c10::BFloat16) (instance 1)]>(at::TensorIteratorBase &, const ... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 265,420,800 | cast | 2.16 | 128 | 518,400 | 1,050,000,000 | 18 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | cast | void at::native::unrolled_elementwise_kernel<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 3)]::operator ()() const::[lambda(int) (instance 1)], std::array<char *, (unsigned long)2>, (int)4, TrivialOffsetCalculator<(int)1, unsigned int>,... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 16,777,216 | cast | 0.03562 | 128 | 32,400 | 57,870,000 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::vectorized_elementwise_kernel<(int)4, at::native::CUDAFunctorOnSelf_add<int>, std::array<char *, (unsigned long)2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 8,294,400 | residual | 0.04115 | 128 | 16,200 | 109,490,000 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::vectorized_elementwise_kernel<(int)4, at::native::AUnaryFunctor<unsigned char, unsigned char, unsigned char, at::native::BitwiseAndFunctor<unsigned char>>, std::array<char *, (unsigned long)2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 132,710,400 | other | 0.18618 | 128 | 259,200 | 510,930,000 | 30 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | cast | void at::native::unrolled_elementwise_kernel<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 3)]::operator ()() const::[lambda(int) (instance 1)], std::array<char *, (unsigned long)2>, (int)4, TrivialOffsetCalculator<(int)1, unsigned int>,... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 265,420,800 | cast | 0.52253 | 128 | 518,400 | 1,300,000,000 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | cast | void at::native::unrolled_elementwise_kernel<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 4)]::operator ()() const::[lambda(long) (instance 1)], std::array<char *, (unsigned long)2>, (int)4, TrivialOffsetCalculator<(int)1, unsigned int>... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 265,420,800 | cast | 1.15 | 128 | 518,400 | 3,160,000,000 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | copy | void at::native::index_elementwise_kernel<(int)128, (int)4, void at::native::gpu_index_kernel<void at::native::index_kernel_impl<at::native::OpaqueType<(int)2>>(at::TensorIteratorBase &, c10::ArrayRef<long>, c10::ArrayRef<long>)::[lambda(char *, const char *, long) (instance 1)]>(at::TensorIteratorBase &, c10::ArrayRef... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 265,420,800 | other | 1.34 | 128 | 518,400 | 2,650,000,000 | 36 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | cast | void at::native::elementwise_kernel<(int)128, (int)4, void at::native::gpu_kernel_impl_nocast<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 12)]::operator ()() const::[lambda(c10::BFloat16) (instance 1)]>(at::TensorIteratorBase &, const ... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 265,420,800 | cast | 0.93219 | 128 | 518,400 | 2,640,000,000 | 18 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::vectorized_elementwise_kernel<(int)4, at::native::BUnaryFunctor<unsigned char, unsigned char, unsigned char, at::native::rshift_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 1)]::operator ()() const::[lambda() (instance 1)]::operator ()() const::[lambda(unsigned char, unsigned char) (insta... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 132,710,400 | other | 0.18822 | 128 | 259,200 | 511,320,000 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | cast | void at::native::unrolled_elementwise_kernel<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 3)]::operator ()() const::[lambda(int) (instance 1)], std::array<char *, (unsigned long)2>, (int)4, TrivialOffsetCalculator<(int)1, unsigned int>,... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 265,420,800 | cast | 0.52176 | 128 | 518,400 | 1,300,000,000 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | cast | void at::native::unrolled_elementwise_kernel<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 4)]::operator ()() const::[lambda(long) (instance 1)], std::array<char *, (unsigned long)2>, (int)4, TrivialOffsetCalculator<(int)1, unsigned int>... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 265,420,800 | cast | 1.15 | 128 | 518,400 | 3,160,000,000 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | copy | void at::native::index_elementwise_kernel<(int)128, (int)4, void at::native::gpu_index_kernel<void at::native::index_kernel_impl<at::native::OpaqueType<(int)2>>(at::TensorIteratorBase &, c10::ArrayRef<long>, c10::ArrayRef<long>)::[lambda(char *, const char *, long) (instance 1)]>(at::TensorIteratorBase &, c10::ArrayRef... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 265,420,800 | other | 1.34 | 128 | 518,400 | 2,650,000,000 | 36 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | cast | void at::native::elementwise_kernel<(int)128, (int)4, void at::native::gpu_kernel_impl_nocast<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 12)]::operator ()() const::[lambda(c10::BFloat16) (instance 1)]>(at::TensorIteratorBase &, const ... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 265,420,800 | cast | 0.9327 | 128 | 518,400 | 2,640,000,000 | 18 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::vectorized_elementwise_kernel<(int)8, at::native::FillFunctor<c10::BFloat16>, std::array<char *, (unsigned long)1>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 512 | fill | 0.00294 | 128 | 1 | 4,350 | 16 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::elementwise_kernel<(int)128, (int)4, void at::native::gpu_kernel_impl<at::native::<unnamed>::pow_tensor_tensor_kernel(at::TensorIteratorBase &)::[lambda() (instance 1)]::operator ()() const::[lambda() (instance 11)]::operator ()() const::[lambda(c10::BFloat16, c10::BFloat16) (instance 1)]>(at::TensorIt... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 16,777,216 | other | 0.13354 | 128 | 32,400 | 87,900,000 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::elementwise_kernel<(int)128, (int)4, void at::native::gpu_kernel_impl_nocast<at::native::BinaryFunctor<c10::BFloat16, c10::BFloat16, c10::BFloat16, at::native::binary_internal::MulFunctor<float>>>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 530,841,600 | mul | 1.65 | 128 | 1,036,800 | 2,130,000,000 | 18 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | cast | void at::native::elementwise_kernel<(int)128, (int)4, void at::native::gpu_kernel_impl_nocast<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 12)]::operator ()() const::[lambda(c10::BFloat16) (instance 1)]>(at::TensorIteratorBase &, const ... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 530,841,600 | cast | 4.34 | 128 | 1,036,800 | 2,110,000,000 | 18 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | cast | void at::native::unrolled_elementwise_kernel<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 3)]::operator ()() const::[lambda(int) (instance 1)], std::array<char *, (unsigned long)2>, (int)4, TrivialOffsetCalculator<(int)1, unsigned int>,... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 8,294,400 | cast | 0.01981 | 128 | 16,200 | 16,550,000 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::vectorized_elementwise_kernel<(int)4, at::native::CUDAFunctorOnSelf_add<int>, std::array<char *, (unsigned long)2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 4,147,200 | residual | 0.01968 | 128 | 8,100 | 43,660,000 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::vectorized_elementwise_kernel<(int)4, at::native::AUnaryFunctor<unsigned char, unsigned char, unsigned char, at::native::BitwiseAndFunctor<unsigned char>>, std::array<char *, (unsigned long)2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 66,355,200 | other | 0.09299 | 128 | 129,600 | 244,470,000 | 30 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | cast | void at::native::unrolled_elementwise_kernel<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 3)]::operator ()() const::[lambda(int) (instance 1)], std::array<char *, (unsigned long)2>, (int)4, TrivialOffsetCalculator<(int)1, unsigned int>,... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 132,710,400 | cast | 0.26618 | 128 | 259,200 | 638,430,000 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | cast | void at::native::unrolled_elementwise_kernel<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 4)]::operator ()() const::[lambda(long) (instance 1)], std::array<char *, (unsigned long)2>, (int)4, TrivialOffsetCalculator<(int)1, unsigned int>... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 132,710,400 | cast | 0.57571 | 128 | 259,200 | 1,570,000,000 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | copy | void at::native::index_elementwise_kernel<(int)128, (int)4, void at::native::gpu_index_kernel<void at::native::index_kernel_impl<at::native::OpaqueType<(int)2>>(at::TensorIteratorBase &, c10::ArrayRef<long>, c10::ArrayRef<long>)::[lambda(char *, const char *, long) (instance 1)]>(at::TensorIteratorBase &, c10::ArrayRef... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 132,710,400 | other | 0.67238 | 128 | 259,200 | 1,320,000,000 | 36 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | cast | void at::native::elementwise_kernel<(int)128, (int)4, void at::native::gpu_kernel_impl_nocast<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 12)]::operator ()() const::[lambda(c10::BFloat16) (instance 1)]>(at::TensorIteratorBase &, const ... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 132,710,400 | cast | 0.46848 | 128 | 259,200 | 1,310,000,000 | 18 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::vectorized_elementwise_kernel<(int)4, at::native::BUnaryFunctor<unsigned char, unsigned char, unsigned char, at::native::rshift_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 1)]::operator ()() const::[lambda() (instance 1)]::operator ()() const::[lambda(unsigned char, unsigned char) (insta... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 66,355,200 | other | 0.09299 | 128 | 129,600 | 245,090,000 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | cast | void at::native::unrolled_elementwise_kernel<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 3)]::operator ()() const::[lambda(int) (instance 1)], std::array<char *, (unsigned long)2>, (int)4, TrivialOffsetCalculator<(int)1, unsigned int>,... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 132,710,400 | cast | 0.26733 | 128 | 259,200 | 638,860,000 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | cast | void at::native::unrolled_elementwise_kernel<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 4)]::operator ()() const::[lambda(long) (instance 1)], std::array<char *, (unsigned long)2>, (int)4, TrivialOffsetCalculator<(int)1, unsigned int>... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 132,710,400 | cast | 0.57622 | 128 | 259,200 | 1,570,000,000 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | copy | void at::native::index_elementwise_kernel<(int)128, (int)4, void at::native::gpu_index_kernel<void at::native::index_kernel_impl<at::native::OpaqueType<(int)2>>(at::TensorIteratorBase &, c10::ArrayRef<long>, c10::ArrayRef<long>)::[lambda(char *, const char *, long) (instance 1)]>(at::TensorIteratorBase &, c10::ArrayRef... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 132,710,400 | other | 0.67146 | 128 | 259,200 | 1,320,000,000 | 36 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | cast | void at::native::elementwise_kernel<(int)128, (int)4, void at::native::gpu_kernel_impl_nocast<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 12)]::operator ()() const::[lambda(c10::BFloat16) (instance 1)]>(at::TensorIteratorBase &, const ... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 132,710,400 | cast | 0.46787 | 128 | 259,200 | 1,310,000,000 | 18 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::vectorized_elementwise_kernel<(int)8, at::native::FillFunctor<c10::BFloat16>, std::array<char *, (unsigned long)1>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 512 | fill | 0.0032 | 128 | 1 | 4,350 | 16 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::elementwise_kernel<(int)128, (int)4, void at::native::gpu_kernel_impl<at::native::<unnamed>::pow_tensor_tensor_kernel(at::TensorIteratorBase &)::[lambda() (instance 1)]::operator ()() const::[lambda() (instance 11)]::operator ()() const::[lambda(c10::BFloat16, c10::BFloat16) (instance 1)]>(at::TensorIt... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 8,294,400 | other | 0.0687 | 128 | 16,200 | 38,260,000 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::elementwise_kernel<(int)128, (int)4, void at::native::gpu_kernel_impl_nocast<at::native::BinaryFunctor<c10::BFloat16, c10::BFloat16, c10::BFloat16, at::native::binary_internal::MulFunctor<float>>>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 265,420,800 | mul | 0.82819 | 128 | 518,400 | 1,060,000,000 | 18 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | cast | void at::native::elementwise_kernel<(int)128, (int)4, void at::native::gpu_kernel_impl_nocast<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 12)]::operator ()() const::[lambda(c10::BFloat16) (instance 1)]>(at::TensorIteratorBase &, const ... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 265,420,800 | cast | 2.16 | 128 | 518,400 | 1,050,000,000 | 18 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | cast | void at::native::unrolled_elementwise_kernel<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 3)]::operator ()() const::[lambda(int) (instance 1)], std::array<char *, (unsigned long)2>, (int)4, TrivialOffsetCalculator<(int)1, unsigned int>,... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 16,777,216 | cast | 0.03542 | 128 | 32,400 | 58,050,000 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::vectorized_elementwise_kernel<(int)4, at::native::CUDAFunctorOnSelf_add<int>, std::array<char *, (unsigned long)2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 8,294,400 | residual | 0.04237 | 128 | 16,200 | 109,460,000 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::vectorized_elementwise_kernel<(int)4, at::native::AUnaryFunctor<unsigned char, unsigned char, unsigned char, at::native::BitwiseAndFunctor<unsigned char>>, std::array<char *, (unsigned long)2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 132,710,400 | other | 0.1872 | 128 | 259,200 | 510,580,000 | 30 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.