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 | elementwise | void at::native::vectorized_elementwise_kernel<(int)8, at::native::CUDAFunctor_add<c10::BFloat16>, std::array<char *, (unsigned long)3>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 368,640 | residual | 0.00458 | 128 | 720 | 2,960,000 | 34 |
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 | 737,280 | mul | 0.00643 | 128 | 1,440 | 1,500,000 | 18 |
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)1>>(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 | 512 | other | 0.0064 | 128 | 1 | 50,180 | 36 |
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::<unnamed>::masked_fill_kernel(at::TensorIterator &, const c10::Scalar &)::[lambda() (instance 1)]::operator ()() const::[lambda() (instance 12)]::operator ()() const::[lambda(c10::BFloat16, bool) (instance 1)]>(at:... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 737,280 | other | 0.00707 | 128 | 1,440 | 1,500,000 | 18 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | copy | void at::native::vectorized_gather_kernel<(int)16, long>(char *, char *, T2 *, int, long, long, long, long, bool) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 737,280 | other | 0.00451 | 256 | 512 | 1,480,000 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | reduce | void at::native::reduce_kernel<(int)128, (int)4, at::native::ReduceOp<c10::BFloat16, at::native::func_wrapper_t<c10::BFloat16, at::native::sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(at::TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, (int)4, (int)8>>(T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 184,320 | other | 0.00848 | 128 | 360 | 1,550,000 | 54 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::vectorized_elementwise_kernel<(int)8, at::native::CUDAFunctor_add<c10::BFloat16>, std::array<char *, (unsigned long)3>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 92,160 | residual | 0.00406 | 128 | 180 | 744,190 | 34 |
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 7)]::operator ()() const::[lambda(float) (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 | 184,320 | cast | 0.00608 | 128 | 360 | 401,150 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::vectorized_elementwise_kernel<(int)4, void at::native::<unnamed>::pow_tensor_scalar_kernel_impl<float, float>(at::TensorIteratorBase &, T2)::[lambda(float) (instance 1)], std::array<char *, (unsigned long)2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 92,160 | rmsnorm | 0.00362 | 128 | 180 | 742,140 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | reduce | void at::native::reduce_kernel<(int)512, (int)1, at::native::ReduceOp<float, at::native::MeanOps<float, float, float, float>, unsigned int, float, (int)4, (int)4>>(T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 8,192 | rmsnorm | 0.01482 | 512 | 4 | 812,030 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::vectorized_elementwise_kernel<(int)4, at::native::CUDAFunctorOnSelf_add<float>, std::array<char *, (unsigned long)2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 512 | residual | 0.00317 | 128 | 1 | 5,380 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::vectorized_elementwise_kernel<(int)4, at::native::rsqrt_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 2)]::operator ()() const::[lambda() (instance 2)]::operator ()() const::[lambda(float) (instance 1)], std::array<char *, (unsigned long)2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 512 | rmsnorm | 0.00333 | 128 | 1 | 6,910 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::elementwise_kernel<(int)128, (int)2, void at::native::gpu_kernel_impl_nocast<at::native::BinaryFunctor<float, float, float, 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 | 368,640 | mul | 0.00483 | 128 | 720 | 755,970 | 18 |
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::BinaryFunctor<float, float, float, 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 | 184,320 | mul | 0.00842 | 128 | 360 | 809,220 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | cast | void at::native::vectorized_elementwise_kernel<(int)8, at::native::bfloat16_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda(float) (instance 1)], std::array<char *, (unsigned long)2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 92,160 | cast | 0.00355 | 128 | 180 | 742,660 | 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 | 92,160 | mul | 0.00669 | 128 | 256 | 293,380 | 18 |
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 | 92,160 | mul | 0.00685 | 128 | 256 | 293,890 | 18 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::vectorized_elementwise_kernel<(int)8, at::native::CUDAFunctor_add<c10::BFloat16>, std::array<char *, (unsigned long)3>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 92,160 | residual | 0.00358 | 128 | 128 | 531,200 | 34 |
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 | 92,160 | mul | 0.00643 | 128 | 256 | 293,120 | 18 |
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 | 92,160 | mul | 0.00704 | 128 | 256 | 294,140 | 18 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::vectorized_elementwise_kernel<(int)8, at::native::CUDAFunctor_add<c10::BFloat16>, std::array<char *, (unsigned long)3>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 92,160 | residual | 0.00365 | 128 | 128 | 530,940 | 34 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | copy | void at::native::<unnamed>::CatArrayBatchedCopy<at::native::<unnamed>::OpaqueType<(unsigned int)2>, unsigned int, (int)4, (int)64, (int)64>(T1 *, at::native::<unnamed>::CatArrInputTensorMetadata<T1, T2, T4, T5>, at::native::<unnamed>::TensorSizeStride<T2, (unsigned int)4>, int, T2) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 737,280 | other | 0.00717 | 512 | 528 | 531,970 | 25 |
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 | 16,384 | mul | 0.00659 | 128 | 32 | 64,260 | 18 |
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 | 16,384 | mul | 0.00694 | 128 | 32 | 64,510 | 18 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::vectorized_elementwise_kernel<(int)8, at::native::CUDAFunctor_add<c10::BFloat16>, std::array<char *, (unsigned long)3>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 8,192 | residual | 0.00333 | 128 | 16 | 72,700 | 34 |
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 | 16,384 | mul | 0.00688 | 128 | 32 | 63,490 | 18 |
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 | 16,384 | mul | 0.00576 | 128 | 32 | 64,000 | 18 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::vectorized_elementwise_kernel<(int)8, at::native::CUDAFunctor_add<c10::BFloat16>, std::array<char *, (unsigned long)3>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 8,192 | residual | 0.00346 | 128 | 16 | 72,190 | 34 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | copy | void at::native::<unnamed>::CatArrayBatchedCopy<at::native::<unnamed>::OpaqueType<(unsigned int)2>, unsigned int, (int)4, (int)64, (int)64>(T1 *, at::native::<unnamed>::CatArrInputTensorMetadata<T1, T2, T4, T5>, at::native::<unnamed>::TensorSizeStride<T2, (unsigned int)4>, int, T2) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 737,280 | other | 0.00579 | 512 | 528 | 72,700 | 25 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | copy | void at::native::<unnamed>::CatArrayBatchedCopy<at::native::<unnamed>::OpaqueType<(unsigned int)2>, unsigned int, (int)4, (int)64, (int)64>(T1 *, at::native::<unnamed>::CatArrInputTensorMetadata<T1, T2, T4, T5>, at::native::<unnamed>::TensorSizeStride<T2, (unsigned int)4>, int, T2) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 737,280 | other | 0.00547 | 512 | 528 | 72,960 | 25 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | copy | void at::native::<unnamed>::CatArrayBatchedCopy<at::native::<unnamed>::OpaqueType<(unsigned int)2>, unsigned int, (int)4, (int)64, (int)64>(T1 *, at::native::<unnamed>::CatArrInputTensorMetadata<T1, T2, T4, T5>, at::native::<unnamed>::TensorSizeStride<T2, (unsigned int)4>, int, T2) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 737,280 | other | 0.00659 | 512 | 528 | 72,960 | 25 |
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 | 368,640 | cast | 0.00544 | 128 | 512 | 88,830 | 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 | 368,640 | cast | 0.00544 | 128 | 512 | 88,580 | 18 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::vectorized_elementwise_kernel<(int)8, at::native::AUnaryFunctor<c10::BFloat16, c10::BFloat16, c10::BFloat16, at::native::binary_internal::MulFunctor<float>>, std::array<char *, (unsigned long)2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 92,160 | mul | 0.00374 | 128 | 256 | 530,690 | 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::CUDAFunctor_add<c10::BFloat16>>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 368,640 | residual | 0.00694 | 128 | 512 | 559,620 | 18 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | copy | void at::native::<unnamed>::CatArrayBatchedCopy<at::native::<unnamed>::OpaqueType<(unsigned int)2>, unsigned int, (int)4, (int)64, (int)64>(T1 *, at::native::<unnamed>::CatArrInputTensorMetadata<T1, T2, T4, T5>, at::native::<unnamed>::TensorSizeStride<T2, (unsigned int)4>, int, T2) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 737,280 | other | 0.00758 | 512 | 528 | 531,710 | 25 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | reduce | void at::native::reduce_kernel<(int)512, (int)1, at::native::ReduceOp<c10::BFloat16, at::native::MaxOps<c10::BFloat16>, unsigned int, c10::BFloat16, (int)4, (int)4>>(T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 737,280 | other | 0.00858 | 512 | 256 | 1,130,000 | 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::CUDAFunctor_add<c10::BFloat16>>(at::TensorIteratorBase &, const T1 &)::[lambda(int) (instance 1)]>(int, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 368,640 | residual | 0.00685 | 128 | 520 | 592,130 | 18 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | reduce | void <unnamed>::softmax_warp_forward<c10::BFloat16, c10::BFloat16, float, (int)7, (bool)0, (bool)0>(T2 *, const T1 *, int, int, int, const bool *, int, bool) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 368,640 | other | 0.00528 | 128 | 512 | 544,000 | 30 |
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 | 368,640 | cast | 0.00685 | 128 | 512 | 547,580 | 18 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::vectorized_elementwise_kernel<(int)8, at::native::CUDAFunctor_add<c10::BFloat16>, std::array<char *, (unsigned long)3>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 92,160 | residual | 0.00365 | 128 | 180 | 743,940 | 34 |
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 7)]::operator ()() const::[lambda(float) (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 | 184,320 | cast | 0.00592 | 128 | 360 | 400,900 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::vectorized_elementwise_kernel<(int)4, void at::native::<unnamed>::pow_tensor_scalar_kernel_impl<float, float>(at::TensorIteratorBase &, T2)::[lambda(float) (instance 1)], std::array<char *, (unsigned long)2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 92,160 | rmsnorm | 0.00365 | 128 | 180 | 742,400 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | reduce | void at::native::reduce_kernel<(int)512, (int)1, at::native::ReduceOp<float, at::native::MeanOps<float, float, float, float>, unsigned int, float, (int)4, (int)4>>(T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 8,192 | rmsnorm | 0.01526 | 512 | 4 | 801,020 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::vectorized_elementwise_kernel<(int)4, at::native::CUDAFunctorOnSelf_add<float>, std::array<char *, (unsigned long)2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 512 | residual | 0.0032 | 128 | 1 | 5,380 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::vectorized_elementwise_kernel<(int)4, at::native::rsqrt_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 2)]::operator ()() const::[lambda() (instance 2)]::operator ()() const::[lambda(float) (instance 1)], std::array<char *, (unsigned long)2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 512 | rmsnorm | 0.00336 | 128 | 1 | 6,910 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::elementwise_kernel<(int)128, (int)2, void at::native::gpu_kernel_impl_nocast<at::native::BinaryFunctor<float, float, float, 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 | 368,640 | mul | 0.0048 | 128 | 720 | 752,130 | 18 |
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::BinaryFunctor<float, float, float, 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 | 184,320 | mul | 0.00848 | 128 | 360 | 809,470 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | cast | void at::native::vectorized_elementwise_kernel<(int)8, at::native::bfloat16_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda(float) (instance 1)], std::array<char *, (unsigned long)2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 92,160 | cast | 0.0039 | 128 | 180 | 742,910 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | reduce | void <unnamed>::softmax_warp_forward<c10::BFloat16, c10::BFloat16, float, (int)2, (bool)0, (bool)0>(T2 *, const T1 *, int, int, int, const bool *, int, bool) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 512 | other | 0.00333 | 128 | 1 | 7,170 | 21 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | cast | void at::native::elementwise_kernel<(int)128, (int)2, void at::native::gpu_kernel_impl_nocast<at::native::direct_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 3)]::operator ()() const::[lambda() (instance 4)]::operator ()() const::[lambda(long) (instance 1)]>(at::TensorIteratorBase &, const T1 &)::[la... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 512 | cast | 0.00426 | 128 | 1 | 13,310 | 18 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::vectorized_elementwise_kernel<(int)4, void at::native::compare_scalar_kernel<long>(at::TensorIteratorBase &, at::native::<unnamed>::OpType, T1)::[lambda(long) (instance 1)], std::array<char *, (unsigned long)2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 512 | other | 0.00355 | 128 | 1 | 11,520 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::vectorized_elementwise_kernel<(int)2, at::native::<unnamed>::launch_clamp_scalar(at::TensorIteratorBase &, c10::Scalar, c10::Scalar, at::native::detail::ClampLimits)::[lambda() (instance 1)]::operator ()() const::[lambda() (instance 4)]::operator ()() const::[lambda(long) (instance 1)], std::array<char... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 512 | other | 0.00349 | 128 | 1 | 11,780 | 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_put_kernel_impl<at::native::OpaqueType<(int)8>>(at::TensorIterator &, 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 | 512 | other | 0.00646 | 128 | 1 | 52,220 | 36 |
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)8>>(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 | 512 | other | 0.00544 | 128 | 1 | 52,220 | 36 |
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 | 512 | other | 0.00518 | 128 | 1 | 50,690 | 36 |
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)8>>(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 | 512 | other | 0.00605 | 128 | 1 | 52,220 | 36 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | copy | void at::native::vectorized_gather_kernel<(int)16, long>(char *, char *, T2 *, int, long, long, long, long, bool) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 737,280 | other | 0.00403 | 256 | 512 | 374,780 | 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 | 512 | cast | 0.00336 | 128 | 1 | 35,070 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::vectorized_elementwise_kernel<(int)4, at::native::FillFunctor<int>, std::array<char *, (unsigned long)1>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 512 | fill | 0.00278 | 128 | 1 | 4,350 | 16 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | copy | void at::native::vectorized_gather_kernel<(int)16, long>(char *, char *, T2 *, int, long, long, long, long, bool) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 737,280 | other | 0.00422 | 256 | 768 | 328,450 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::vectorized_elementwise_kernel<(int)8, at::native::CUDAFunctor_add<c10::BFloat16>, std::array<char *, (unsigned long)3>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 737,280 | residual | 0.00582 | 128 | 1,440 | 5,910,000 | 34 |
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::<unnamed>::launch_clamp_scalar(at::TensorIteratorBase &, c10::Scalar, c10::Scalar, at::native::detail::ClampLimits)::[lambda() (instance 1)]::operator ()() const::[lambda() (instance 9)]::operator ()() const::[lamb... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 737,280 | other | 0.0071 | 128 | 1,440 | 2,970,000 | 18 |
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::<unnamed>::launch_clamp_scalar(at::TensorIteratorBase &, c10::Scalar, c10::Scalar, at::native::detail::ClampLimits)::[lambda() (instance 1)]::operator ()() const::[lambda() (instance 9)]::operator ()() const::[lamb... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 737,280 | other | 0.00701 | 128 | 1,440 | 2,970,000 | 18 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::vectorized_elementwise_kernel<(int)8, at::native::AUnaryFunctor<c10::BFloat16, c10::BFloat16, c10::BFloat16, at::native::binary_internal::MulFunctor<float>>, std::array<char *, (unsigned long)2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 368,640 | mul | 0.004 | 128 | 720 | 1,480,000 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::vectorized_elementwise_kernel<(int)8, at::native::sigmoid_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 2)]::operator ()() const::[lambda() (instance 4)]::operator ()() const::[lambda(c10::BFloat16) (instance 1)], std::array<char *, (unsigned long)2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 368,640 | other | 0.00483 | 128 | 720 | 1,490,000 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::vectorized_elementwise_kernel<(int)8, at::native::BinaryFunctor<c10::BFloat16, c10::BFloat16, c10::BFloat16, at::native::binary_internal::MulFunctor<float>>, std::array<char *, (unsigned long)3>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 368,640 | mul | 0.00461 | 128 | 720 | 2,960,000 | 40 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::vectorized_elementwise_kernel<(int)8, at::native::CUDAFunctorOnSelf_add<c10::BFloat16>, std::array<char *, (unsigned long)2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 368,640 | residual | 0.00384 | 128 | 720 | 1,480,000 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::vectorized_elementwise_kernel<(int)8, at::native::BinaryFunctor<c10::BFloat16, c10::BFloat16, c10::BFloat16, at::native::binary_internal::MulFunctor<float>>, std::array<char *, (unsigned long)3>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 368,640 | mul | 0.00448 | 128 | 720 | 2,960,000 | 40 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | copy | void at::native::vectorized_gather_kernel<(int)16, long>(char *, char *, T2 *, int, long, long, long, long, bool) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 737,280 | other | 0.00406 | 256 | 512 | 167,170 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::vectorized_elementwise_kernel<(int)8, at::native::CUDAFunctor_add<c10::BFloat16>, std::array<char *, (unsigned long)3>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 368,640 | residual | 0.0047 | 128 | 720 | 2,960,000 | 34 |
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 | 737,280 | mul | 0.00653 | 128 | 1,440 | 1,500,000 | 18 |
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)1>>(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 | 512 | other | 0.00534 | 128 | 1 | 50,180 | 36 |
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::<unnamed>::masked_fill_kernel(at::TensorIterator &, const c10::Scalar &)::[lambda() (instance 1)]::operator ()() const::[lambda() (instance 12)]::operator ()() const::[lambda(c10::BFloat16, bool) (instance 1)]>(at:... | bf16 | false | null | null | null | 1 | 128 | null | null | null | 737,280 | other | 0.00707 | 128 | 1,440 | 1,500,000 | 18 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | copy | void at::native::vectorized_gather_kernel<(int)16, long>(char *, char *, T2 *, int, long, long, long, long, bool) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 737,280 | other | 0.00451 | 256 | 512 | 1,480,000 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | reduce | void at::native::reduce_kernel<(int)128, (int)4, at::native::ReduceOp<c10::BFloat16, at::native::func_wrapper_t<c10::BFloat16, at::native::sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(at::TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, (int)4, (int)8>>(T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 184,320 | other | 0.0087 | 128 | 360 | 1,550,000 | 54 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::vectorized_elementwise_kernel<(int)8, at::native::CUDAFunctor_add<c10::BFloat16>, std::array<char *, (unsigned long)3>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 92,160 | residual | 0.00371 | 128 | 180 | 744,190 | 34 |
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 7)]::operator ()() const::[lambda(float) (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 | 184,320 | cast | 0.00598 | 128 | 360 | 401,150 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::vectorized_elementwise_kernel<(int)4, void at::native::<unnamed>::pow_tensor_scalar_kernel_impl<float, float>(at::TensorIteratorBase &, T2)::[lambda(float) (instance 1)], std::array<char *, (unsigned long)2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 92,160 | rmsnorm | 0.00368 | 128 | 180 | 742,660 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | reduce | void at::native::reduce_kernel<(int)512, (int)1, at::native::ReduceOp<float, at::native::MeanOps<float, float, float, float>, unsigned int, float, (int)4, (int)4>>(T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 8,192 | rmsnorm | 0.01485 | 512 | 4 | 806,140 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::vectorized_elementwise_kernel<(int)4, at::native::CUDAFunctorOnSelf_add<float>, std::array<char *, (unsigned long)2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 512 | residual | 0.0031 | 128 | 1 | 5,120 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::vectorized_elementwise_kernel<(int)4, at::native::rsqrt_kernel_cuda(at::TensorIteratorBase &)::[lambda() (instance 2)]::operator ()() const::[lambda() (instance 2)]::operator ()() const::[lambda(float) (instance 1)], std::array<char *, (unsigned long)2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 512 | rmsnorm | 0.00336 | 128 | 1 | 6,400 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::elementwise_kernel<(int)128, (int)2, void at::native::gpu_kernel_impl_nocast<at::native::BinaryFunctor<float, float, float, 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 | 368,640 | mul | 0.00486 | 128 | 720 | 751,870 | 18 |
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::BinaryFunctor<float, float, float, 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 | 184,320 | mul | 0.00826 | 128 | 360 | 809,980 | 32 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | cast | void at::native::vectorized_elementwise_kernel<(int)8, at::native::bfloat16_copy_kernel_cuda(at::TensorIteratorBase &)::[lambda(float) (instance 1)], std::array<char *, (unsigned long)2>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 92,160 | cast | 0.00358 | 128 | 180 | 742,400 | 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 | 92,160 | mul | 0.00678 | 128 | 256 | 293,630 | 18 |
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 | 92,160 | mul | 0.00653 | 128 | 256 | 293,380 | 18 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::vectorized_elementwise_kernel<(int)8, at::native::CUDAFunctor_add<c10::BFloat16>, std::array<char *, (unsigned long)3>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 92,160 | residual | 0.00371 | 128 | 128 | 530,940 | 34 |
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 | 92,160 | mul | 0.0065 | 128 | 256 | 292,860 | 18 |
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 | 92,160 | mul | 0.00682 | 128 | 256 | 293,380 | 18 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::vectorized_elementwise_kernel<(int)8, at::native::CUDAFunctor_add<c10::BFloat16>, std::array<char *, (unsigned long)3>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 92,160 | residual | 0.00355 | 128 | 128 | 530,940 | 34 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | copy | void at::native::<unnamed>::CatArrayBatchedCopy<at::native::<unnamed>::OpaqueType<(unsigned int)2>, unsigned int, (int)4, (int)64, (int)64>(T1 *, at::native::<unnamed>::CatArrInputTensorMetadata<T1, T2, T4, T5>, at::native::<unnamed>::TensorSizeStride<T2, (unsigned int)4>, int, T2) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 737,280 | other | 0.00765 | 512 | 528 | 531,460 | 25 |
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 | 16,384 | mul | 0.00573 | 128 | 32 | 63,740 | 18 |
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 | 16,384 | mul | 0.00637 | 128 | 32 | 63,740 | 18 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::vectorized_elementwise_kernel<(int)8, at::native::CUDAFunctor_add<c10::BFloat16>, std::array<char *, (unsigned long)3>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 8,192 | residual | 0.00336 | 128 | 16 | 72,190 | 34 |
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 | 16,384 | mul | 0.00563 | 128 | 32 | 62,980 | 18 |
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 | 16,384 | mul | 0.00691 | 128 | 32 | 64,510 | 18 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | elementwise | void at::native::vectorized_elementwise_kernel<(int)8, at::native::CUDAFunctor_add<c10::BFloat16>, std::array<char *, (unsigned long)3>>(int, T2, T3) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 8,192 | residual | 0.00342 | 128 | 16 | 72,700 | 34 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | copy | void at::native::<unnamed>::CatArrayBatchedCopy<at::native::<unnamed>::OpaqueType<(unsigned int)2>, unsigned int, (int)4, (int)64, (int)64>(T1 *, at::native::<unnamed>::CatArrInputTensorMetadata<T1, T2, T4, T5>, at::native::<unnamed>::TensorSizeStride<T2, (unsigned int)4>, int, T2) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 737,280 | other | 0.00637 | 512 | 528 | 72,960 | 25 |
gpt-oss-20b_prefill | H100 | gpt-oss-20b | copy | void at::native::<unnamed>::CatArrayBatchedCopy<at::native::<unnamed>::OpaqueType<(unsigned int)2>, unsigned int, (int)4, (int)64, (int)64>(T1 *, at::native::<unnamed>::CatArrInputTensorMetadata<T1, T2, T4, T5>, at::native::<unnamed>::TensorSizeStride<T2, (unsigned int)4>, int, T2) | bf16 | false | null | null | null | 1 | 128 | null | null | null | 737,280 | other | 0.00538 | 512 | 528 | 72,190 | 25 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.