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 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_128x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 1,024 | 8,192 | 1 | 128 | null | null | null | null | v_proj | 0.02496 | 128 | 72 | 18,900,000 | 246 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_256x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 8,192 | 8,192 | 1 | 128 | null | null | null | null | o_proj | 0.13667 | 256 | 96 | 136,340,000 | 238 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_256x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 28,672 | 8,192 | 1 | 128 | null | null | null | null | gate | 0.5889 | 256 | 112 | 475,680,000 | 238 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_256x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 28,672 | 8,192 | 1 | 128 | null | null | null | null | up | 0.5952 | 256 | 112 | 475,620,000 | 238 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_256x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 8,192 | 28,672 | 1 | 128 | null | null | null | null | down_proj | 0.37626 | 256 | 96 | 477,140,000 | 238 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_256x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 8,192 | 8,192 | 1 | 128 | null | null | null | null | q_proj | 0.14582 | 256 | 96 | 136,340,000 | 238 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_128x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 1,024 | 8,192 | 1 | 128 | null | null | null | null | k_proj | 0.02531 | 128 | 72 | 18,900,000 | 246 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_128x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 1,024 | 8,192 | 1 | 128 | null | null | null | null | v_proj | 0.02474 | 128 | 72 | 18,900,000 | 246 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_256x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 8,192 | 8,192 | 1 | 128 | null | null | null | null | o_proj | 0.13994 | 256 | 96 | 136,340,000 | 238 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_256x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 28,672 | 8,192 | 1 | 128 | null | null | null | null | gate | 0.59328 | 256 | 112 | 475,480,000 | 238 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_256x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 28,672 | 8,192 | 1 | 128 | null | null | null | null | up | 0.58218 | 256 | 112 | 475,670,000 | 238 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_256x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 8,192 | 28,672 | 1 | 128 | null | null | null | null | down_proj | 0.37418 | 256 | 96 | 477,130,000 | 238 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_256x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 8,192 | 8,192 | 1 | 128 | null | null | null | null | q_proj | 0.13901 | 256 | 96 | 136,340,000 | 238 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_128x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 1,024 | 8,192 | 1 | 128 | null | null | null | null | k_proj | 0.02534 | 128 | 72 | 18,900,000 | 246 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_128x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 1,024 | 8,192 | 1 | 128 | null | null | null | null | v_proj | 0.02477 | 128 | 72 | 18,900,000 | 246 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_256x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 8,192 | 8,192 | 1 | 128 | null | null | null | null | o_proj | 0.13661 | 256 | 96 | 136,340,000 | 238 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_256x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 28,672 | 8,192 | 1 | 128 | null | null | null | null | gate | 0.59072 | 256 | 112 | 475,590,000 | 238 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_256x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 28,672 | 8,192 | 1 | 128 | null | null | null | null | up | 0.5959 | 256 | 112 | 475,500,000 | 238 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_256x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 8,192 | 28,672 | 1 | 128 | null | null | null | null | down_proj | 0.3872 | 256 | 96 | 477,130,000 | 238 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_256x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 8,192 | 8,192 | 1 | 128 | null | null | null | null | q_proj | 0.13696 | 256 | 96 | 136,340,000 | 238 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_128x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 1,024 | 8,192 | 1 | 128 | null | null | null | null | k_proj | 0.02528 | 128 | 72 | 18,900,000 | 246 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_128x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 1,024 | 8,192 | 1 | 128 | null | null | null | null | v_proj | 0.02496 | 128 | 72 | 18,900,000 | 246 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_256x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 8,192 | 8,192 | 1 | 128 | null | null | null | null | o_proj | 0.13258 | 256 | 96 | 136,340,000 | 238 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_256x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 28,672 | 8,192 | 1 | 128 | null | null | null | null | gate | 0.58621 | 256 | 112 | 475,610,000 | 238 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_256x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 28,672 | 8,192 | 1 | 128 | null | null | null | null | up | 0.59098 | 256 | 112 | 475,330,000 | 238 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_256x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 8,192 | 28,672 | 1 | 128 | null | null | null | null | down_proj | 0.38886 | 256 | 96 | 477,140,000 | 238 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_256x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 8,192 | 8,192 | 1 | 128 | null | null | null | null | q_proj | 0.13258 | 256 | 96 | 136,340,000 | 238 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_128x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 1,024 | 8,192 | 1 | 128 | null | null | null | null | k_proj | 0.0257 | 128 | 72 | 18,900,000 | 246 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_128x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 1,024 | 8,192 | 1 | 128 | null | null | null | null | v_proj | 0.02518 | 128 | 72 | 18,900,000 | 246 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_256x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 8,192 | 8,192 | 1 | 128 | null | null | null | null | o_proj | 0.13619 | 256 | 96 | 136,340,000 | 238 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_256x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 28,672 | 8,192 | 1 | 128 | null | null | null | null | gate | 0.58842 | 256 | 112 | 475,340,000 | 238 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_256x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 28,672 | 8,192 | 1 | 128 | null | null | null | null | up | 0.58109 | 256 | 112 | 476,300,000 | 238 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_256x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 8,192 | 28,672 | 1 | 128 | null | null | null | null | down_proj | 0.38752 | 256 | 96 | 477,140,000 | 238 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_256x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 8,192 | 8,192 | 1 | 128 | null | null | null | null | q_proj | 0.13597 | 256 | 96 | 136,350,000 | 238 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_128x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 1,024 | 8,192 | 1 | 128 | null | null | null | null | k_proj | 0.02557 | 128 | 72 | 18,900,000 | 246 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_128x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 1,024 | 8,192 | 1 | 128 | null | null | null | null | v_proj | 0.02528 | 128 | 72 | 18,900,000 | 246 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_256x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 8,192 | 8,192 | 1 | 128 | null | null | null | null | o_proj | 0.13821 | 256 | 96 | 136,340,000 | 238 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_256x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 28,672 | 8,192 | 1 | 128 | null | null | null | null | gate | 0.58371 | 256 | 112 | 476,140,000 | 238 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_256x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 28,672 | 8,192 | 1 | 128 | null | null | null | null | up | 0.59146 | 256 | 112 | 475,480,000 | 238 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_256x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 8,192 | 28,672 | 1 | 128 | null | null | null | null | down_proj | 0.38141 | 256 | 96 | 477,130,000 | 238 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_256x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 8,192 | 8,192 | 1 | 128 | null | null | null | null | q_proj | 0.14045 | 256 | 96 | 136,340,000 | 238 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_128x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 1,024 | 8,192 | 1 | 128 | null | null | null | null | k_proj | 0.0255 | 128 | 72 | 18,900,000 | 246 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_128x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 1,024 | 8,192 | 1 | 128 | null | null | null | null | v_proj | 0.02534 | 128 | 72 | 18,900,000 | 246 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_256x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 8,192 | 8,192 | 1 | 128 | null | null | null | null | o_proj | 0.14218 | 256 | 96 | 136,340,000 | 238 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_256x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 28,672 | 8,192 | 1 | 128 | null | null | null | null | gate | 0.58941 | 256 | 112 | 476,120,000 | 238 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_256x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 28,672 | 8,192 | 1 | 128 | null | null | null | null | up | 0.58557 | 256 | 112 | 475,920,000 | 238 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_256x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 8,192 | 28,672 | 1 | 128 | null | null | null | null | down_proj | 0.38826 | 256 | 96 | 477,130,000 | 238 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_256x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 8,192 | 8,192 | 1 | 128 | null | null | null | null | q_proj | 0.14058 | 256 | 96 | 136,340,000 | 238 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_128x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 1,024 | 8,192 | 1 | 128 | null | null | null | null | k_proj | 0.02566 | 128 | 72 | 18,900,000 | 246 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_128x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 1,024 | 8,192 | 1 | 128 | null | null | null | null | v_proj | 0.02525 | 128 | 72 | 18,900,000 | 246 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_256x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 8,192 | 8,192 | 1 | 128 | null | null | null | null | o_proj | 0.13715 | 256 | 96 | 136,340,000 | 238 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_256x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 28,672 | 8,192 | 1 | 128 | null | null | null | null | gate | 0.59514 | 256 | 112 | 475,500,000 | 238 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_256x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 28,672 | 8,192 | 1 | 128 | null | null | null | null | up | 0.59283 | 256 | 112 | 475,600,000 | 238 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_256x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 8,192 | 28,672 | 1 | 128 | null | null | null | null | down_proj | 0.37578 | 256 | 96 | 477,140,000 | 238 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_256x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 8,192 | 8,192 | 1 | 128 | null | null | null | null | q_proj | 0.13821 | 256 | 96 | 136,340,000 | 238 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_128x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 1,024 | 8,192 | 1 | 128 | null | null | null | null | k_proj | 0.02566 | 128 | 72 | 18,900,000 | 246 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_128x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 1,024 | 8,192 | 1 | 128 | null | null | null | null | v_proj | 0.02541 | 128 | 72 | 18,900,000 | 246 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_256x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 8,192 | 8,192 | 1 | 128 | null | null | null | null | o_proj | 0.13747 | 256 | 96 | 136,340,000 | 238 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_256x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 28,672 | 8,192 | 1 | 128 | null | null | null | null | gate | 0.58461 | 256 | 112 | 475,950,000 | 238 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_256x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 28,672 | 8,192 | 1 | 128 | null | null | null | null | up | 0.59168 | 256 | 112 | 475,640,000 | 238 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_256x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 8,192 | 28,672 | 1 | 128 | null | null | null | null | down_proj | 0.37792 | 256 | 96 | 477,130,000 | 238 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | gemm | ampere_bf16_s16816gemm_bf16_256x128_ldg8_f2f_stages_64x3_tn | bf16 | true | 128 | 128,256 | 8,192 | 1 | 128 | null | null | null | null | lm_head | 1.65 | 256 | 501 | 2,140,000,000 | 238 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | splitk_reduce | void cublasLt::splitKreduce_kernel<32, 16, int, __nv_bfloat16, __nv_bfloat16, float, __nv_bfloat16, 0, __nv_bfloat16, __nv_bfloat16, __nv_bfloat16, 1, 0, 0>(cublasLt::cublasSplitKParams<T6>, const T4 *, const T10 *, T9 *, T5 *, const T6 *, const T6 *, const T11 *, const T4 *, T11 *, void *, long, T6 *, int *, T6 *, T6 ... | bf16 | true | 128 | 1,024 | 8,192 | null | null | null | null | null | null | null | 0.0071 | 512 | 256 | 2,370,000 | 32 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | splitk_reduce | void cublasLt::splitKreduce_kernel<32, 16, int, __nv_bfloat16, __nv_bfloat16, float, __nv_bfloat16, 0, __nv_bfloat16, __nv_bfloat16, __nv_bfloat16, 1, 0, 0>(cublasLt::cublasSplitKParams<T6>, const T4 *, const T10 *, T9 *, T5 *, const T6 *, const T6 *, const T11 *, const T4 *, T11 *, void *, long, T6 *, int *, T6 *, T6 ... | bf16 | true | 128 | 1,024 | 8,192 | null | null | null | null | null | null | null | 0.0072 | 512 | 256 | 2,370,000 | 32 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | splitk_reduce | void cublasLt::splitKreduce_kernel<32, 16, int, __nv_bfloat16, __nv_bfloat16, float, __nv_bfloat16, 0, __nv_bfloat16, __nv_bfloat16, __nv_bfloat16, 1, 0, 0>(cublasLt::cublasSplitKParams<T6>, const T4 *, const T10 *, T9 *, T5 *, const T6 *, const T6 *, const T11 *, const T4 *, T11 *, void *, long, T6 *, int *, T6 *, T6 ... | bf16 | true | 128 | 1,024 | 8,192 | null | null | null | null | null | null | null | 0.00707 | 512 | 256 | 2,370,000 | 32 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | splitk_reduce | void cublasLt::splitKreduce_kernel<32, 16, int, __nv_bfloat16, __nv_bfloat16, float, __nv_bfloat16, 0, __nv_bfloat16, __nv_bfloat16, __nv_bfloat16, 1, 0, 0>(cublasLt::cublasSplitKParams<T6>, const T4 *, const T10 *, T9 *, T5 *, const T6 *, const T6 *, const T11 *, const T4 *, T11 *, void *, long, T6 *, int *, T6 *, T6 ... | bf16 | true | 128 | 1,024 | 8,192 | null | null | null | null | null | null | null | 0.0073 | 512 | 256 | 2,370,000 | 32 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | splitk_reduce | void cublasLt::splitKreduce_kernel<32, 16, int, __nv_bfloat16, __nv_bfloat16, float, __nv_bfloat16, 0, __nv_bfloat16, __nv_bfloat16, __nv_bfloat16, 1, 0, 0>(cublasLt::cublasSplitKParams<T6>, const T4 *, const T10 *, T9 *, T5 *, const T6 *, const T6 *, const T11 *, const T4 *, T11 *, void *, long, T6 *, int *, T6 *, T6 ... | bf16 | true | 128 | 1,024 | 8,192 | null | null | null | null | null | null | null | 0.00739 | 512 | 256 | 2,370,000 | 32 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | splitk_reduce | void cublasLt::splitKreduce_kernel<32, 16, int, __nv_bfloat16, __nv_bfloat16, float, __nv_bfloat16, 0, __nv_bfloat16, __nv_bfloat16, __nv_bfloat16, 1, 0, 0>(cublasLt::cublasSplitKParams<T6>, const T4 *, const T10 *, T9 *, T5 *, const T6 *, const T6 *, const T11 *, const T4 *, T11 *, void *, long, T6 *, int *, T6 *, T6 ... | bf16 | true | 128 | 1,024 | 8,192 | null | null | null | null | null | null | null | 0.00723 | 512 | 256 | 2,370,000 | 32 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | splitk_reduce | void cublasLt::splitKreduce_kernel<32, 16, int, __nv_bfloat16, __nv_bfloat16, float, __nv_bfloat16, 0, __nv_bfloat16, __nv_bfloat16, __nv_bfloat16, 1, 0, 0>(cublasLt::cublasSplitKParams<T6>, const T4 *, const T10 *, T9 *, T5 *, const T6 *, const T6 *, const T11 *, const T4 *, T11 *, void *, long, T6 *, int *, T6 *, T6 ... | bf16 | true | 128 | 1,024 | 8,192 | null | null | null | null | null | null | null | 0.00726 | 512 | 256 | 2,370,000 | 32 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | splitk_reduce | void cublasLt::splitKreduce_kernel<32, 16, int, __nv_bfloat16, __nv_bfloat16, float, __nv_bfloat16, 0, __nv_bfloat16, __nv_bfloat16, __nv_bfloat16, 1, 0, 0>(cublasLt::cublasSplitKParams<T6>, const T4 *, const T10 *, T9 *, T5 *, const T6 *, const T6 *, const T11 *, const T4 *, T11 *, void *, long, T6 *, int *, T6 *, T6 ... | bf16 | true | 128 | 1,024 | 8,192 | null | null | null | null | null | null | null | 0.0071 | 512 | 256 | 2,370,000 | 32 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | splitk_reduce | void cublasLt::splitKreduce_kernel<32, 16, int, __nv_bfloat16, __nv_bfloat16, float, __nv_bfloat16, 0, __nv_bfloat16, __nv_bfloat16, __nv_bfloat16, 1, 0, 0>(cublasLt::cublasSplitKParams<T6>, const T4 *, const T10 *, T9 *, T5 *, const T6 *, const T6 *, const T11 *, const T4 *, T11 *, void *, long, T6 *, int *, T6 *, T6 ... | bf16 | true | 128 | 1,024 | 8,192 | null | null | null | null | null | null | null | 0.00717 | 512 | 256 | 2,370,000 | 32 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | splitk_reduce | void cublasLt::splitKreduce_kernel<32, 16, int, __nv_bfloat16, __nv_bfloat16, float, __nv_bfloat16, 0, __nv_bfloat16, __nv_bfloat16, __nv_bfloat16, 1, 0, 0>(cublasLt::cublasSplitKParams<T6>, const T4 *, const T10 *, T9 *, T5 *, const T6 *, const T6 *, const T11 *, const T4 *, T11 *, void *, long, T6 *, int *, T6 *, T6 ... | bf16 | true | 128 | 1,024 | 8,192 | null | null | null | null | null | null | null | 0.00733 | 512 | 256 | 2,370,000 | 32 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | splitk_reduce | void cublasLt::splitKreduce_kernel<32, 16, int, __nv_bfloat16, __nv_bfloat16, float, __nv_bfloat16, 0, __nv_bfloat16, __nv_bfloat16, __nv_bfloat16, 1, 0, 0>(cublasLt::cublasSplitKParams<T6>, const T4 *, const T10 *, T9 *, T5 *, const T6 *, const T6 *, const T11 *, const T4 *, T11 *, void *, long, T6 *, int *, T6 *, T6 ... | bf16 | true | 128 | 1,024 | 8,192 | null | null | null | null | null | null | null | 0.0073 | 512 | 256 | 2,370,000 | 32 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | splitk_reduce | void cublasLt::splitKreduce_kernel<32, 16, int, __nv_bfloat16, __nv_bfloat16, float, __nv_bfloat16, 0, __nv_bfloat16, __nv_bfloat16, __nv_bfloat16, 1, 0, 0>(cublasLt::cublasSplitKParams<T6>, const T4 *, const T10 *, T9 *, T5 *, const T6 *, const T6 *, const T11 *, const T4 *, T11 *, void *, long, T6 *, int *, T6 *, T6 ... | bf16 | true | 128 | 1,024 | 8,192 | null | null | null | null | null | null | null | 0.00714 | 512 | 256 | 2,370,000 | 32 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | splitk_reduce | void cublasLt::splitKreduce_kernel<32, 16, int, __nv_bfloat16, __nv_bfloat16, float, __nv_bfloat16, 0, __nv_bfloat16, __nv_bfloat16, __nv_bfloat16, 1, 0, 0>(cublasLt::cublasSplitKParams<T6>, const T4 *, const T10 *, T9 *, T5 *, const T6 *, const T6 *, const T11 *, const T4 *, T11 *, void *, long, T6 *, int *, T6 *, T6 ... | bf16 | true | 128 | 1,024 | 8,192 | null | null | null | null | null | null | null | 0.00707 | 512 | 256 | 2,370,000 | 32 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | splitk_reduce | void cublasLt::splitKreduce_kernel<32, 16, int, __nv_bfloat16, __nv_bfloat16, float, __nv_bfloat16, 0, __nv_bfloat16, __nv_bfloat16, __nv_bfloat16, 1, 0, 0>(cublasLt::cublasSplitKParams<T6>, const T4 *, const T10 *, T9 *, T5 *, const T6 *, const T6 *, const T11 *, const T4 *, T11 *, void *, long, T6 *, int *, T6 *, T6 ... | bf16 | true | 128 | 1,024 | 8,192 | null | null | null | null | null | null | null | 0.0072 | 512 | 256 | 2,370,000 | 32 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | splitk_reduce | void cublasLt::splitKreduce_kernel<32, 16, int, __nv_bfloat16, __nv_bfloat16, float, __nv_bfloat16, 0, __nv_bfloat16, __nv_bfloat16, __nv_bfloat16, 1, 0, 0>(cublasLt::cublasSplitKParams<T6>, const T4 *, const T10 *, T9 *, T5 *, const T6 *, const T6 *, const T11 *, const T4 *, T11 *, void *, long, T6 *, int *, T6 *, T6 ... | bf16 | true | 128 | 1,024 | 8,192 | null | null | null | null | null | null | null | 0.0073 | 512 | 256 | 2,370,000 | 32 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | splitk_reduce | void cublasLt::splitKreduce_kernel<32, 16, int, __nv_bfloat16, __nv_bfloat16, float, __nv_bfloat16, 0, __nv_bfloat16, __nv_bfloat16, __nv_bfloat16, 1, 0, 0>(cublasLt::cublasSplitKParams<T6>, const T4 *, const T10 *, T9 *, T5 *, const T6 *, const T6 *, const T11 *, const T4 *, T11 *, void *, long, T6 *, int *, T6 *, T6 ... | bf16 | true | 128 | 1,024 | 8,192 | null | null | null | null | null | null | null | 0.00723 | 512 | 256 | 2,370,000 | 32 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | splitk_reduce | void cublasLt::splitKreduce_kernel<32, 16, int, __nv_bfloat16, __nv_bfloat16, float, __nv_bfloat16, 0, __nv_bfloat16, __nv_bfloat16, __nv_bfloat16, 1, 0, 0>(cublasLt::cublasSplitKParams<T6>, const T4 *, const T10 *, T9 *, T5 *, const T6 *, const T6 *, const T11 *, const T4 *, T11 *, void *, long, T6 *, int *, T6 *, T6 ... | bf16 | true | 128 | 1,024 | 8,192 | null | null | null | null | null | null | null | 0.00714 | 512 | 256 | 2,370,000 | 32 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | splitk_reduce | void cublasLt::splitKreduce_kernel<32, 16, int, __nv_bfloat16, __nv_bfloat16, float, __nv_bfloat16, 0, __nv_bfloat16, __nv_bfloat16, __nv_bfloat16, 1, 0, 0>(cublasLt::cublasSplitKParams<T6>, const T4 *, const T10 *, T9 *, T5 *, const T6 *, const T6 *, const T11 *, const T4 *, T11 *, void *, long, T6 *, int *, T6 *, T6 ... | bf16 | true | 128 | 1,024 | 8,192 | null | null | null | null | null | null | null | 0.00701 | 512 | 256 | 2,370,000 | 32 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | splitk_reduce | void cublasLt::splitKreduce_kernel<32, 16, int, __nv_bfloat16, __nv_bfloat16, float, __nv_bfloat16, 0, __nv_bfloat16, __nv_bfloat16, __nv_bfloat16, 1, 0, 0>(cublasLt::cublasSplitKParams<T6>, const T4 *, const T10 *, T9 *, T5 *, const T6 *, const T6 *, const T11 *, const T4 *, T11 *, void *, long, T6 *, int *, T6 *, T6 ... | bf16 | true | 128 | 1,024 | 8,192 | null | null | null | null | null | null | null | 0.00714 | 512 | 256 | 2,370,000 | 32 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | splitk_reduce | void cublasLt::splitKreduce_kernel<32, 16, int, __nv_bfloat16, __nv_bfloat16, float, __nv_bfloat16, 0, __nv_bfloat16, __nv_bfloat16, __nv_bfloat16, 1, 0, 0>(cublasLt::cublasSplitKParams<T6>, const T4 *, const T10 *, T9 *, T5 *, const T6 *, const T6 *, const T11 *, const T4 *, T11 *, void *, long, T6 *, int *, T6 *, T6 ... | bf16 | true | 128 | 1,024 | 8,192 | null | null | null | null | null | null | null | 0.00707 | 512 | 256 | 2,370,000 | 32 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | splitk_reduce | void cublasLt::splitKreduce_kernel<32, 16, int, __nv_bfloat16, __nv_bfloat16, float, __nv_bfloat16, 0, __nv_bfloat16, __nv_bfloat16, __nv_bfloat16, 1, 0, 0>(cublasLt::cublasSplitKParams<T6>, const T4 *, const T10 *, T9 *, T5 *, const T6 *, const T6 *, const T11 *, const T4 *, T11 *, void *, long, T6 *, int *, T6 *, T6 ... | bf16 | true | 128 | 1,024 | 8,192 | null | null | null | null | null | null | null | 0.00701 | 512 | 256 | 2,370,000 | 32 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | splitk_reduce | void cublasLt::splitKreduce_kernel<32, 16, int, __nv_bfloat16, __nv_bfloat16, float, __nv_bfloat16, 0, __nv_bfloat16, __nv_bfloat16, __nv_bfloat16, 1, 0, 0>(cublasLt::cublasSplitKParams<T6>, const T4 *, const T10 *, T9 *, T5 *, const T6 *, const T6 *, const T11 *, const T4 *, T11 *, void *, long, T6 *, int *, T6 *, T6 ... | bf16 | true | 128 | 1,024 | 8,192 | null | null | null | null | null | null | null | 0.00717 | 512 | 256 | 2,370,000 | 32 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | splitk_reduce | void cublasLt::splitKreduce_kernel<32, 16, int, __nv_bfloat16, __nv_bfloat16, float, __nv_bfloat16, 0, __nv_bfloat16, __nv_bfloat16, __nv_bfloat16, 1, 0, 0>(cublasLt::cublasSplitKParams<T6>, const T4 *, const T10 *, T9 *, T5 *, const T6 *, const T6 *, const T11 *, const T4 *, T11 *, void *, long, T6 *, int *, T6 *, T6 ... | bf16 | true | 128 | 1,024 | 8,192 | null | null | null | null | null | null | null | 0.00704 | 512 | 256 | 2,370,000 | 32 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | splitk_reduce | void cublasLt::splitKreduce_kernel<32, 16, int, __nv_bfloat16, __nv_bfloat16, float, __nv_bfloat16, 0, __nv_bfloat16, __nv_bfloat16, __nv_bfloat16, 1, 0, 0>(cublasLt::cublasSplitKParams<T6>, const T4 *, const T10 *, T9 *, T5 *, const T6 *, const T6 *, const T11 *, const T4 *, T11 *, void *, long, T6 *, int *, T6 *, T6 ... | bf16 | true | 128 | 1,024 | 8,192 | null | null | null | null | null | null | null | 0.00707 | 512 | 256 | 2,370,000 | 32 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | splitk_reduce | void cublasLt::splitKreduce_kernel<32, 16, int, __nv_bfloat16, __nv_bfloat16, float, __nv_bfloat16, 0, __nv_bfloat16, __nv_bfloat16, __nv_bfloat16, 1, 0, 0>(cublasLt::cublasSplitKParams<T6>, const T4 *, const T10 *, T9 *, T5 *, const T6 *, const T6 *, const T11 *, const T4 *, T11 *, void *, long, T6 *, int *, T6 *, T6 ... | bf16 | true | 128 | 1,024 | 8,192 | null | null | null | null | null | null | null | 0.00707 | 512 | 256 | 2,370,000 | 32 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | splitk_reduce | void cublasLt::splitKreduce_kernel<32, 16, int, __nv_bfloat16, __nv_bfloat16, float, __nv_bfloat16, 0, __nv_bfloat16, __nv_bfloat16, __nv_bfloat16, 1, 0, 0>(cublasLt::cublasSplitKParams<T6>, const T4 *, const T10 *, T9 *, T5 *, const T6 *, const T6 *, const T11 *, const T4 *, T11 *, void *, long, T6 *, int *, T6 *, T6 ... | bf16 | true | 128 | 1,024 | 8,192 | null | null | null | null | null | null | null | 0.00717 | 512 | 256 | 2,370,000 | 32 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | splitk_reduce | void cublasLt::splitKreduce_kernel<32, 16, int, __nv_bfloat16, __nv_bfloat16, float, __nv_bfloat16, 0, __nv_bfloat16, __nv_bfloat16, __nv_bfloat16, 1, 0, 0>(cublasLt::cublasSplitKParams<T6>, const T4 *, const T10 *, T9 *, T5 *, const T6 *, const T6 *, const T11 *, const T4 *, T11 *, void *, long, T6 *, int *, T6 *, T6 ... | bf16 | true | 128 | 1,024 | 8,192 | null | null | null | null | null | null | null | 0.00726 | 512 | 256 | 2,370,000 | 32 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | splitk_reduce | void cublasLt::splitKreduce_kernel<32, 16, int, __nv_bfloat16, __nv_bfloat16, float, __nv_bfloat16, 0, __nv_bfloat16, __nv_bfloat16, __nv_bfloat16, 1, 0, 0>(cublasLt::cublasSplitKParams<T6>, const T4 *, const T10 *, T9 *, T5 *, const T6 *, const T6 *, const T11 *, const T4 *, T11 *, void *, long, T6 *, int *, T6 *, T6 ... | bf16 | true | 128 | 1,024 | 8,192 | null | null | null | null | null | null | null | 0.0072 | 512 | 256 | 2,370,000 | 32 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | splitk_reduce | void cublasLt::splitKreduce_kernel<32, 16, int, __nv_bfloat16, __nv_bfloat16, float, __nv_bfloat16, 0, __nv_bfloat16, __nv_bfloat16, __nv_bfloat16, 1, 0, 0>(cublasLt::cublasSplitKParams<T6>, const T4 *, const T10 *, T9 *, T5 *, const T6 *, const T6 *, const T11 *, const T4 *, T11 *, void *, long, T6 *, int *, T6 *, T6 ... | bf16 | true | 128 | 1,024 | 8,192 | null | null | null | null | null | null | null | 0.00717 | 512 | 256 | 2,370,000 | 32 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | splitk_reduce | void cublasLt::splitKreduce_kernel<32, 16, int, __nv_bfloat16, __nv_bfloat16, float, __nv_bfloat16, 0, __nv_bfloat16, __nv_bfloat16, __nv_bfloat16, 1, 0, 0>(cublasLt::cublasSplitKParams<T6>, const T4 *, const T10 *, T9 *, T5 *, const T6 *, const T6 *, const T11 *, const T4 *, T11 *, void *, long, T6 *, int *, T6 *, T6 ... | bf16 | true | 128 | 1,024 | 8,192 | null | null | null | null | null | null | null | 0.00717 | 512 | 256 | 2,370,000 | 32 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | splitk_reduce | void cublasLt::splitKreduce_kernel<32, 16, int, __nv_bfloat16, __nv_bfloat16, float, __nv_bfloat16, 0, __nv_bfloat16, __nv_bfloat16, __nv_bfloat16, 1, 0, 0>(cublasLt::cublasSplitKParams<T6>, const T4 *, const T10 *, T9 *, T5 *, const T6 *, const T6 *, const T11 *, const T4 *, T11 *, void *, long, T6 *, int *, T6 *, T6 ... | bf16 | true | 128 | 1,024 | 8,192 | null | null | null | null | null | null | null | 0.00723 | 512 | 256 | 2,370,000 | 32 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | splitk_reduce | void cublasLt::splitKreduce_kernel<32, 16, int, __nv_bfloat16, __nv_bfloat16, float, __nv_bfloat16, 0, __nv_bfloat16, __nv_bfloat16, __nv_bfloat16, 1, 0, 0>(cublasLt::cublasSplitKParams<T6>, const T4 *, const T10 *, T9 *, T5 *, const T6 *, const T6 *, const T11 *, const T4 *, T11 *, void *, long, T6 *, int *, T6 *, T6 ... | bf16 | true | 128 | 1,024 | 8,192 | null | null | null | null | null | null | null | 0.00714 | 512 | 256 | 2,370,000 | 32 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | splitk_reduce | void cublasLt::splitKreduce_kernel<32, 16, int, __nv_bfloat16, __nv_bfloat16, float, __nv_bfloat16, 0, __nv_bfloat16, __nv_bfloat16, __nv_bfloat16, 1, 0, 0>(cublasLt::cublasSplitKParams<T6>, const T4 *, const T10 *, T9 *, T5 *, const T6 *, const T6 *, const T11 *, const T4 *, T11 *, void *, long, T6 *, int *, T6 *, T6 ... | bf16 | true | 128 | 1,024 | 8,192 | null | null | null | null | null | null | null | 0.00707 | 512 | 256 | 2,370,000 | 32 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | splitk_reduce | void cublasLt::splitKreduce_kernel<32, 16, int, __nv_bfloat16, __nv_bfloat16, float, __nv_bfloat16, 0, __nv_bfloat16, __nv_bfloat16, __nv_bfloat16, 1, 0, 0>(cublasLt::cublasSplitKParams<T6>, const T4 *, const T10 *, T9 *, T5 *, const T6 *, const T6 *, const T11 *, const T4 *, T11 *, void *, long, T6 *, int *, T6 *, T6 ... | bf16 | true | 128 | 1,024 | 8,192 | null | null | null | null | null | null | null | 0.00714 | 512 | 256 | 2,370,000 | 32 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | splitk_reduce | void cublasLt::splitKreduce_kernel<32, 16, int, __nv_bfloat16, __nv_bfloat16, float, __nv_bfloat16, 0, __nv_bfloat16, __nv_bfloat16, __nv_bfloat16, 1, 0, 0>(cublasLt::cublasSplitKParams<T6>, const T4 *, const T10 *, T9 *, T5 *, const T6 *, const T6 *, const T11 *, const T4 *, T11 *, void *, long, T6 *, int *, T6 *, T6 ... | bf16 | true | 128 | 1,024 | 8,192 | null | null | null | null | null | null | null | 0.00704 | 512 | 256 | 2,370,000 | 32 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | splitk_reduce | void cublasLt::splitKreduce_kernel<32, 16, int, __nv_bfloat16, __nv_bfloat16, float, __nv_bfloat16, 0, __nv_bfloat16, __nv_bfloat16, __nv_bfloat16, 1, 0, 0>(cublasLt::cublasSplitKParams<T6>, const T4 *, const T10 *, T9 *, T5 *, const T6 *, const T6 *, const T11 *, const T4 *, T11 *, void *, long, T6 *, int *, T6 *, T6 ... | bf16 | true | 128 | 1,024 | 8,192 | null | null | null | null | null | null | null | 0.00707 | 512 | 256 | 2,370,000 | 32 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | splitk_reduce | void cublasLt::splitKreduce_kernel<32, 16, int, __nv_bfloat16, __nv_bfloat16, float, __nv_bfloat16, 0, __nv_bfloat16, __nv_bfloat16, __nv_bfloat16, 1, 0, 0>(cublasLt::cublasSplitKParams<T6>, const T4 *, const T10 *, T9 *, T5 *, const T6 *, const T6 *, const T11 *, const T4 *, T11 *, void *, long, T6 *, int *, T6 *, T6 ... | bf16 | true | 128 | 1,024 | 8,192 | null | null | null | null | null | null | null | 0.0071 | 512 | 256 | 2,370,000 | 32 |
Llama-3.1-70B-Instruct_prefill | A100 | Llama-70B | splitk_reduce | void cublasLt::splitKreduce_kernel<32, 16, int, __nv_bfloat16, __nv_bfloat16, float, __nv_bfloat16, 0, __nv_bfloat16, __nv_bfloat16, __nv_bfloat16, 1, 0, 0>(cublasLt::cublasSplitKParams<T6>, const T4 *, const T10 *, T9 *, T5 *, const T6 *, const T6 *, const T11 *, const T4 *, T11 *, void *, long, T6 *, int *, T6 *, T6 ... | bf16 | true | 128 | 1,024 | 8,192 | null | null | null | null | null | null | null | 0.00717 | 512 | 256 | 2,370,000 | 32 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.