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 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
misc_sweep | RTX3090 | misc_sweep | reduce | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 32 | 16,384 | null | 1 | 32 | null | null | null | 524,288 | reduce | 0.00672 | 512 | 4 | 187,010 | 32 |
misc_sweep | RTX3090 | misc_sweep | reduce | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 64 | 4,096 | null | 1 | 64 | null | null | null | 262,144 | reduce | 0.00803 | 512 | 1 | 121,730 | 32 |
misc_sweep | RTX3090 | misc_sweep | reduce | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 64 | 5,120 | null | 1 | 64 | null | null | null | 327,680 | reduce | 0.00826 | 512 | 1 | 121,730 | 32 |
misc_sweep | RTX3090 | misc_sweep | reduce | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 64 | 6,144 | null | 1 | 64 | null | null | null | 393,216 | reduce | 0.00896 | 512 | 1 | 137,860 | 32 |
misc_sweep | RTX3090 | misc_sweep | reduce | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 64 | 8,192 | null | 1 | 64 | null | null | null | 524,288 | reduce | 0.00893 | 512 | 1 | 137,860 | 32 |
misc_sweep | RTX3090 | misc_sweep | reduce | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 64 | 10,240 | null | 1 | 64 | null | null | null | 655,360 | reduce | 0.0097 | 512 | 1 | 154,240 | 32 |
misc_sweep | RTX3090 | misc_sweep | reduce | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 64 | 12,288 | null | 1 | 64 | null | null | null | 786,432 | reduce | 0.00982 | 512 | 1 | 154,240 | 32 |
misc_sweep | RTX3090 | misc_sweep | reduce | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 64 | 16,384 | null | 1 | 64 | null | null | null | 1,048,576 | reduce | 0.00602 | 512 | 8 | 187,390 | 32 |
misc_sweep | RTX3090 | misc_sweep | reduce | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 128 | 4,096 | null | 1 | 128 | null | null | null | 524,288 | reduce | 0.00582 | 512 | 8 | 187,010 | 32 |
misc_sweep | RTX3090 | misc_sweep | reduce | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 128 | 5,120 | null | 1 | 128 | null | null | null | 655,360 | reduce | 0.00624 | 512 | 8 | 219,780 | 32 |
misc_sweep | RTX3090 | misc_sweep | reduce | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 128 | 6,144 | null | 1 | 128 | null | null | null | 786,432 | reduce | 0.0064 | 512 | 8 | 219,780 | 32 |
misc_sweep | RTX3090 | misc_sweep | reduce | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 128 | 8,192 | null | 1 | 128 | null | null | null | 1,048,576 | reduce | 0.00643 | 512 | 8 | 252,800 | 32 |
misc_sweep | RTX3090 | misc_sweep | reduce | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 128 | 10,240 | null | 1 | 128 | null | null | null | 1,310,720 | reduce | 0.00621 | 512 | 8 | 252,540 | 32 |
misc_sweep | RTX3090 | misc_sweep | reduce | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 128 | 12,288 | null | 1 | 128 | null | null | null | 1,572,864 | reduce | 0.00666 | 512 | 8 | 318,080 | 32 |
misc_sweep | RTX3090 | misc_sweep | reduce | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 128 | 16,384 | null | 1 | 128 | null | null | null | 2,097,152 | reduce | 0.00694 | 512 | 8 | 318,850 | 32 |
misc_sweep | RTX3090 | misc_sweep | reduce | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 256 | 4,096 | null | 1 | 256 | null | null | null | 1,048,576 | reduce | 0.01114 | 512 | 1 | 187,140 | 32 |
misc_sweep | RTX3090 | misc_sweep | reduce | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 256 | 5,120 | null | 1 | 256 | null | null | null | 1,310,720 | reduce | 0.01094 | 512 | 1 | 187,390 | 32 |
misc_sweep | RTX3090 | misc_sweep | reduce | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 256 | 6,144 | null | 1 | 256 | null | null | null | 1,572,864 | reduce | 0.01267 | 512 | 1 | 220,290 | 32 |
misc_sweep | RTX3090 | misc_sweep | reduce | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 256 | 8,192 | null | 1 | 256 | null | null | null | 2,097,152 | reduce | 0.0127 | 512 | 1 | 219,900 | 32 |
misc_sweep | RTX3090 | misc_sweep | reduce | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 256 | 10,240 | null | 1 | 256 | null | null | null | 2,621,440 | reduce | 0.01427 | 512 | 1 | 252,540 | 32 |
misc_sweep | RTX3090 | misc_sweep | reduce | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 256 | 12,288 | null | 1 | 256 | null | null | null | 3,145,728 | reduce | 0.01414 | 512 | 1 | 252,670 | 32 |
misc_sweep | RTX3090 | misc_sweep | reduce | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 256 | 16,384 | null | 1 | 256 | null | null | null | 4,194,304 | reduce | 0.00579 | 512 | 16 | 318,340 | 32 |
misc_sweep | RTX3090 | misc_sweep | reduce | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 512 | 4,096 | null | 1 | 512 | null | null | null | 2,097,152 | reduce | 0.00608 | 512 | 16 | 318,340 | 32 |
misc_sweep | RTX3090 | misc_sweep | reduce | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 512 | 5,120 | null | 1 | 512 | null | null | null | 2,621,440 | reduce | 0.00637 | 512 | 16 | 383,870 | 32 |
misc_sweep | RTX3090 | misc_sweep | reduce | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 512 | 6,144 | null | 1 | 512 | null | null | null | 3,145,728 | reduce | 0.00618 | 512 | 16 | 383,740 | 32 |
misc_sweep | RTX3090 | misc_sweep | reduce | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 512 | 8,192 | null | 1 | 512 | null | null | null | 4,194,304 | reduce | 0.00621 | 512 | 16 | 449,150 | 32 |
misc_sweep | RTX3090 | misc_sweep | reduce | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 512 | 10,240 | null | 1 | 512 | null | null | null | 5,242,880 | reduce | 0.0064 | 512 | 16 | 449,280 | 32 |
misc_sweep | RTX3090 | misc_sweep | reduce | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 512 | 12,288 | null | 1 | 512 | null | null | null | 6,291,456 | reduce | 0.00685 | 512 | 16 | 580,220 | 32 |
misc_sweep | RTX3090 | misc_sweep | reduce | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 512 | 16,384 | null | 1 | 512 | null | null | null | 8,388,608 | reduce | 0.00669 | 512 | 16 | 582,400 | 32 |
misc_sweep | RTX3090 | misc_sweep | reduce | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 1,024 | 4,096 | null | 1 | 1,024 | null | null | null | 4,194,304 | reduce | 0.0111 | 512 | 2 | 318,080 | 32 |
misc_sweep | RTX3090 | misc_sweep | reduce | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 1,024 | 5,120 | null | 1 | 1,024 | null | null | null | 5,242,880 | reduce | 0.01136 | 512 | 2 | 318,850 | 32 |
misc_sweep | RTX3090 | misc_sweep | reduce | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 1,024 | 6,144 | null | 1 | 1,024 | null | null | null | 6,291,456 | reduce | 0.01277 | 512 | 2 | 384,000 | 32 |
misc_sweep | RTX3090 | misc_sweep | reduce | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 1,024 | 8,192 | null | 1 | 1,024 | null | null | null | 8,388,608 | reduce | 0.01251 | 512 | 2 | 384,130 | 32 |
misc_sweep | RTX3090 | misc_sweep | reduce | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 1,024 | 10,240 | null | 1 | 1,024 | null | null | null | 10,485,760 | reduce | 0.01405 | 512 | 2 | 449,150 | 32 |
misc_sweep | RTX3090 | misc_sweep | reduce | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 1,024 | 12,288 | null | 1 | 1,024 | null | null | null | 12,582,912 | reduce | 0.01514 | 512 | 2 | 449,410 | 32 |
misc_sweep | RTX3090 | misc_sweep | reduce | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 1,024 | 16,384 | null | 1 | 1,024 | null | null | null | 16,777,216 | reduce | 0.00611 | 512 | 32 | 580,860 | 32 |
misc_sweep | RTX3090 | misc_sweep | reduce | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 2,048 | 4,096 | null | 1 | 2,048 | null | null | null | 8,388,608 | reduce | 0.00595 | 512 | 32 | 580,350 | 32 |
misc_sweep | RTX3090 | misc_sweep | reduce | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 2,048 | 5,120 | null | 1 | 2,048 | null | null | null | 10,485,760 | reduce | 0.0064 | 512 | 32 | 713,470 | 32 |
misc_sweep | RTX3090 | misc_sweep | reduce | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 2,048 | 6,144 | null | 1 | 2,048 | null | null | null | 12,582,912 | reduce | 0.0065 | 512 | 32 | 711,550 | 32 |
misc_sweep | RTX3090 | misc_sweep | reduce | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 2,048 | 8,192 | null | 1 | 2,048 | null | null | null | 16,777,216 | reduce | 0.00659 | 512 | 32 | 842,370 | 32 |
misc_sweep | RTX3090 | misc_sweep | reduce | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 2,048 | 10,240 | null | 1 | 2,048 | null | null | null | 20,971,520 | reduce | 0.00637 | 512 | 32 | 842,500 | 32 |
misc_sweep | RTX3090 | misc_sweep | reduce | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 2,048 | 12,288 | null | 1 | 2,048 | null | null | null | 25,165,824 | reduce | 0.00678 | 512 | 32 | 1,100,000 | 32 |
misc_sweep | RTX3090 | misc_sweep | reduce | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 2,048 | 16,384 | null | 1 | 2,048 | null | null | null | 33,554,432 | reduce | 0.00701 | 512 | 32 | 1,100,000 | 32 |
misc_sweep | RTX3090 | misc_sweep | reduce | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 4,096 | 4,096 | null | 1 | 4,096 | null | null | null | 16,777,216 | reduce | 0.01136 | 512 | 4 | 580,350 | 32 |
misc_sweep | RTX3090 | misc_sweep | reduce | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 4,096 | 5,120 | null | 1 | 4,096 | null | null | null | 20,971,520 | reduce | 0.01101 | 512 | 4 | 580,480 | 32 |
misc_sweep | RTX3090 | misc_sweep | reduce | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 4,096 | 6,144 | null | 1 | 4,096 | null | null | null | 25,165,824 | reduce | 0.01261 | 512 | 4 | 711,300 | 32 |
misc_sweep | RTX3090 | misc_sweep | reduce | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 4,096 | 8,192 | null | 1 | 4,096 | null | null | null | 33,554,432 | reduce | 0.01286 | 512 | 4 | 711,420 | 32 |
misc_sweep | RTX3090 | misc_sweep | reduce | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 4,096 | 10,240 | null | 1 | 4,096 | null | null | null | 41,943,040 | reduce | 0.01485 | 512 | 4 | 842,370 | 32 |
misc_sweep | RTX3090 | misc_sweep | reduce | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 4,096 | 12,288 | null | 1 | 4,096 | null | null | null | 50,331,648 | reduce | 0.01408 | 512 | 4 | 842,500 | 32 |
misc_sweep | RTX3090 | misc_sweep | reduce | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 4,096 | 16,384 | null | 1 | 4,096 | null | null | null | 67,108,864 | reduce | 0.00646 | 512 | 64 | 1,100,000 | 32 |
misc_sweep | RTX3090 | misc_sweep | cast | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 1 | 4,096 | null | 1 | 1 | null | null | null | 4,096 | cast | 0.00653 | 512 | 64 | 1,100,000 | 32 |
misc_sweep | RTX3090 | misc_sweep | cast | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 1 | 5,120 | null | 1 | 1 | null | null | null | 5,120 | cast | 0.00698 | 512 | 64 | 1,370,000 | 32 |
misc_sweep | RTX3090 | misc_sweep | cast | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 1 | 6,144 | null | 1 | 1 | null | null | null | 6,144 | cast | 0.00691 | 512 | 64 | 1,370,000 | 32 |
misc_sweep | RTX3090 | misc_sweep | cast | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 1 | 8,192 | null | 1 | 1 | null | null | null | 8,192 | cast | 0.00707 | 512 | 64 | 1,630,000 | 32 |
misc_sweep | RTX3090 | misc_sweep | cast | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 1 | 10,240 | null | 1 | 1 | null | null | null | 10,240 | cast | 0.0073 | 512 | 64 | 1,630,000 | 32 |
misc_sweep | RTX3090 | misc_sweep | cast | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 1 | 12,288 | null | 1 | 1 | null | null | null | 12,288 | cast | 0.00765 | 512 | 64 | 2,150,000 | 32 |
misc_sweep | RTX3090 | misc_sweep | cast | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 1 | 16,384 | null | 1 | 1 | null | null | null | 16,384 | cast | 0.00768 | 512 | 64 | 2,150,000 | 32 |
misc_sweep | RTX3090 | misc_sweep | cast | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 2 | 4,096 | null | 1 | 2 | null | null | null | 8,192 | cast | 0.0111 | 512 | 8 | 1,100,000 | 32 |
misc_sweep | RTX3090 | misc_sweep | cast | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 2 | 5,120 | null | 1 | 2 | null | null | null | 10,240 | cast | 0.01126 | 512 | 8 | 1,100,000 | 32 |
misc_sweep | RTX3090 | misc_sweep | cast | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 2 | 6,144 | null | 1 | 2 | null | null | null | 12,288 | cast | 0.01293 | 512 | 8 | 1,370,000 | 32 |
misc_sweep | RTX3090 | misc_sweep | cast | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 2 | 8,192 | null | 1 | 2 | null | null | null | 16,384 | cast | 0.01267 | 512 | 8 | 1,370,000 | 32 |
misc_sweep | RTX3090 | misc_sweep | cast | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 2 | 10,240 | null | 1 | 2 | null | null | null | 20,480 | cast | 0.01414 | 512 | 8 | 1,630,000 | 32 |
misc_sweep | RTX3090 | misc_sweep | cast | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 2 | 12,288 | null | 1 | 2 | null | null | null | 24,576 | cast | 0.0143 | 512 | 8 | 1,630,000 | 32 |
misc_sweep | RTX3090 | misc_sweep | cast | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 2 | 16,384 | null | 1 | 2 | null | null | null | 32,768 | cast | 0.00784 | 512 | 128 | 2,150,000 | 32 |
misc_sweep | RTX3090 | misc_sweep | cast | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 4 | 4,096 | null | 1 | 4 | null | null | null | 16,384 | cast | 0.00765 | 512 | 128 | 2,150,000 | 32 |
misc_sweep | RTX3090 | misc_sweep | cast | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 4 | 5,120 | null | 1 | 4 | null | null | null | 20,480 | cast | 0.00816 | 512 | 128 | 2,680,000 | 32 |
misc_sweep | RTX3090 | misc_sweep | cast | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 4 | 6,144 | null | 1 | 4 | null | null | null | 24,576 | cast | 0.00838 | 512 | 128 | 2,680,000 | 32 |
misc_sweep | RTX3090 | misc_sweep | cast | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 4 | 8,192 | null | 1 | 4 | null | null | null | 32,768 | cast | 0.00906 | 512 | 128 | 3,200,000 | 32 |
misc_sweep | RTX3090 | misc_sweep | cast | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 4 | 10,240 | null | 1 | 4 | null | null | null | 40,960 | cast | 0.0089 | 512 | 128 | 3,200,000 | 32 |
misc_sweep | RTX3090 | misc_sweep | cast | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 4 | 12,288 | null | 1 | 4 | null | null | null | 49,152 | cast | 0.00995 | 512 | 128 | 4,250,000 | 32 |
misc_sweep | RTX3090 | misc_sweep | cast | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 4 | 16,384 | null | 1 | 4 | null | null | null | 65,536 | cast | 0.01024 | 512 | 128 | 4,250,000 | 32 |
misc_sweep | RTX3090 | misc_sweep | cast | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 8 | 4,096 | null | 1 | 8 | null | null | null | 32,768 | cast | 0.01133 | 512 | 16 | 2,150,000 | 32 |
misc_sweep | RTX3090 | misc_sweep | cast | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 8 | 5,120 | null | 1 | 8 | null | null | null | 40,960 | cast | 0.01114 | 512 | 16 | 2,150,000 | 32 |
misc_sweep | RTX3090 | misc_sweep | cast | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 8 | 6,144 | null | 1 | 8 | null | null | null | 49,152 | cast | 0.01283 | 512 | 16 | 2,680,000 | 32 |
misc_sweep | RTX3090 | misc_sweep | cast | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 8 | 8,192 | null | 1 | 8 | null | null | null | 65,536 | cast | 0.01306 | 512 | 16 | 2,680,000 | 32 |
misc_sweep | RTX3090 | misc_sweep | cast | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 8 | 10,240 | null | 1 | 8 | null | null | null | 81,920 | cast | 0.01472 | 512 | 16 | 3,200,000 | 32 |
misc_sweep | RTX3090 | misc_sweep | cast | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 8 | 12,288 | null | 1 | 8 | null | null | null | 98,304 | cast | 0.01424 | 512 | 16 | 3,200,000 | 32 |
misc_sweep | RTX3090 | misc_sweep | cast | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 8 | 16,384 | null | 1 | 8 | null | null | null | 131,072 | cast | 0.01062 | 512 | 256 | 4,250,000 | 32 |
misc_sweep | RTX3090 | misc_sweep | cast | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 16 | 4,096 | null | 1 | 16 | null | null | null | 65,536 | cast | 0.01078 | 512 | 256 | 4,250,000 | 32 |
misc_sweep | RTX3090 | misc_sweep | cast | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 16 | 5,120 | null | 1 | 16 | null | null | null | 81,920 | cast | 0.01325 | 512 | 256 | 5,300,000 | 32 |
misc_sweep | RTX3090 | misc_sweep | cast | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 16 | 6,144 | null | 1 | 16 | null | null | null | 98,304 | cast | 0.01293 | 512 | 256 | 5,300,000 | 32 |
misc_sweep | RTX3090 | misc_sweep | cast | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 16 | 8,192 | null | 1 | 16 | null | null | null | 131,072 | cast | 0.01341 | 512 | 256 | 6,370,000 | 32 |
misc_sweep | RTX3090 | misc_sweep | cast | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 16 | 10,240 | null | 1 | 16 | null | null | null | 163,840 | cast | 0.01338 | 512 | 256 | 6,370,000 | 32 |
misc_sweep | RTX3090 | misc_sweep | cast | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 16 | 12,288 | null | 1 | 16 | null | null | null | 196,608 | cast | 0.0161 | 512 | 256 | 8,460,000 | 32 |
misc_sweep | RTX3090 | misc_sweep | cast | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 16 | 16,384 | null | 1 | 16 | null | null | null | 262,144 | cast | 0.01603 | 512 | 256 | 8,460,000 | 32 |
misc_sweep | RTX3090 | misc_sweep | cast | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 32 | 4,096 | null | 1 | 32 | null | null | null | 131,072 | cast | 0.01181 | 512 | 32 | 4,250,000 | 32 |
misc_sweep | RTX3090 | misc_sweep | cast | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 32 | 5,120 | null | 1 | 32 | null | null | null | 163,840 | cast | 0.01206 | 512 | 32 | 4,250,000 | 32 |
misc_sweep | RTX3090 | misc_sweep | cast | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 32 | 6,144 | null | 1 | 32 | null | null | null | 196,608 | cast | 0.0136 | 512 | 32 | 5,300,000 | 32 |
misc_sweep | RTX3090 | misc_sweep | cast | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 32 | 8,192 | null | 1 | 32 | null | null | null | 262,144 | cast | 0.01322 | 512 | 32 | 5,300,000 | 32 |
misc_sweep | RTX3090 | misc_sweep | cast | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 32 | 10,240 | null | 1 | 32 | null | null | null | 327,680 | cast | 0.01533 | 512 | 32 | 6,370,000 | 32 |
misc_sweep | RTX3090 | misc_sweep | cast | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 32 | 12,288 | null | 1 | 32 | null | null | null | 393,216 | cast | 0.01536 | 512 | 32 | 6,360,000 | 32 |
misc_sweep | RTX3090 | misc_sweep | cast | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 32 | 16,384 | null | 1 | 32 | null | null | null | 524,288 | cast | 0.01578 | 512 | 512 | 8,460,000 | 32 |
misc_sweep | RTX3090 | misc_sweep | cast | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 64 | 4,096 | null | 1 | 64 | null | null | null | 262,144 | cast | 0.01555 | 512 | 512 | 8,460,000 | 32 |
misc_sweep | RTX3090 | misc_sweep | cast | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 64 | 5,120 | null | 1 | 64 | null | null | null | 327,680 | cast | 0.01885 | 512 | 512 | 10,560,000 | 32 |
misc_sweep | RTX3090 | misc_sweep | cast | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 64 | 6,144 | null | 1 | 64 | null | null | null | 393,216 | cast | 0.01907 | 512 | 512 | 10,560,000 | 32 |
misc_sweep | RTX3090 | misc_sweep | cast | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 64 | 8,192 | null | 1 | 64 | null | null | null | 524,288 | cast | 0.02038 | 512 | 512 | 12,660,000 | 32 |
misc_sweep | RTX3090 | misc_sweep | cast | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 64 | 10,240 | null | 1 | 64 | null | null | null | 655,360 | cast | 0.02026 | 512 | 512 | 12,660,000 | 32 |
misc_sweep | RTX3090 | misc_sweep | cast | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 64 | 12,288 | null | 1 | 64 | null | null | null | 786,432 | cast | 0.02531 | 512 | 512 | 16,860,000 | 32 |
misc_sweep | RTX3090 | misc_sweep | cast | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 64 | 16,384 | null | 1 | 64 | null | null | null | 1,048,576 | cast | 0.02525 | 512 | 512 | 16,880,000 | 32 |
misc_sweep | RTX3090 | misc_sweep | cast | void reduce_kernel<512, 1, ReduceOp<c10::BFloat16, func_wrapper_t<c10::BFloat16, sum_functor<c10::BFloat16, float, c10::BFloat16>::operator ()(TensorIterator &)::[lambda(float, float) (instance 1)]>, unsigned int, c10::BFloat16, 4, 8>>(T3) | bf16 | false | 128 | 4,096 | null | 1 | 128 | null | null | null | 524,288 | cast | 0.01536 | 512 | 64 | 8,480,000 | 32 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.