{
  "clockHz": 1500000000,
  "tileCount": 1472,
  "sampleCount": 43505,
  "matchedSampleCount": 43505,
  "profileSpanCycles": 172206,
  "profileSpanMs": 0.114804,
  "atOffset": null,
  "groups": [
    {
      "name": "exchange",
      "dimensions": {},
      "sampleCount": 10350,
      "tileCount": 1472,
      "phaseCount": 8,
      "phaseCycles": 118506,
      "phaseMs": 0.079004,
      "workCycles": 107004900,
      "averageActiveTiles": 902.949217761126,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 10338.63768115942,
      "p50Cycles": 8598,
      "p95Cycles": 24348,
      "maximumCycles": 30540,
      "firstOffset": 0,
      "lastOffset": 164316
    },
    {
      "name": "fill_zero_u64",
      "dimensions": {},
      "sampleCount": 11536,
      "tileCount": 1472,
      "phaseCount": 41,
      "phaseCycles": 29724,
      "phaseMs": 0.019816,
      "workCycles": 4210656,
      "averageActiveTiles": 141.6584578118692,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 2139648,
      "usefulUtilization": 0.0,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": 0.0,
      "meanCycles": 365.00138696255203,
      "p50Cycles": 300,
      "p95Cycles": 600,
      "maximumCycles": 3276,
      "firstOffset": 36246,
      "lastOffset": 168720
    },
    {
      "name": "attention_softmax_d72_p768_tail_f16",
      "dimensions": {},
      "sampleCount": 1472,
      "tileCount": 1472,
      "phaseCount": 15,
      "phaseCycles": 27576,
      "phaseMs": 0.018384,
      "workCycles": 37145376,
      "averageActiveTiles": 1347.0182767624021,
      "usefulCycles": 18440784.0,
      "estimatedWorkCycles": 37145376,
      "usefulUtilization": 0.4964489792753747,
      "workLaneOccupancy": 0.9974763406940063,
      "deviceUsefulUtilization": 0.45429745147008743,
      "meanCycles": 25234.630434782608,
      "p50Cycles": 25236,
      "p95Cycles": 25236,
      "maximumCycles": 25236,
      "firstOffset": 98862,
      "lastOffset": 126438
    },
    {
      "name": "copy_u64",
      "dimensions": {},
      "sampleCount": 5597,
      "tileCount": 1472,
      "phaseCount": 45,
      "phaseCycles": 23838,
      "phaseMs": 0.015892,
      "workCycles": 5190198,
      "averageActiveTiles": 217.727913415555,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 927.3178488475969,
      "p50Cycles": 570,
      "p95Cycles": 1974,
      "maximumCycles": 2010,
      "firstOffset": 26070,
      "lastOffset": 169230
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k576_c16_r216_r216",
      "dimensions": {},
      "sampleCount": 1472,
      "tileCount": 1472,
      "phaseCount": 1,
      "phaseCycles": 23700,
      "phaseMs": 0.0158,
      "workCycles": 27670656,
      "averageActiveTiles": 1167.5382278481013,
      "usefulCycles": 22674816.0,
      "estimatedWorkCycles": 27670656,
      "usefulUtilization": 0.8194535033791754,
      "workLaneOccupancy": 0.9904891304347826,
      "deviceUsefulUtilization": 0.6499614749587231,
      "meanCycles": 18798.0,
      "p50Cycles": 18798,
      "p95Cycles": 18798,
      "maximumCycles": 18798,
      "firstOffset": 264,
      "lastOffset": 23964
    },
    {
      "name": "gemm_f16_init_small_rows_interleaved_k96_c80_r91_r92",
      "dimensions": {},
      "sampleCount": 896,
      "tileCount": 896,
      "phaseCount": 13,
      "phaseCycles": 19860,
      "phaseMs": 0.01324,
      "workCycles": 14477568,
      "averageActiveTiles": 728.9812688821752,
      "usefulCycles": 8358714.0,
      "estimatedWorkCycles": 14477568,
      "usefulUtilization": 0.5773562244708503,
      "workLaneOccupancy": 0.854296875,
      "deviceUsefulUtilization": 0.2859251855378957,
      "meanCycles": 16158.0,
      "p50Cycles": 16158,
      "p95Cycles": 16158,
      "maximumCycles": 16158,
      "firstOffset": 137082,
      "lastOffset": 156942
    },
    {
      "name": "gemm_f16_init_large_rows_interleaved_k96_c80_r91_r92",
      "dimensions": {},
      "sampleCount": 128,
      "tileCount": 128,
      "phaseCount": 8,
      "phaseCycles": 18438,
      "phaseMs": 0.012292,
      "workCycles": 2072832,
      "averageActiveTiles": 112.42173771558737,
      "usefulCycles": 1207224.0,
      "estimatedWorkCycles": 2072832,
      "usefulUtilization": 0.5824032048907003,
      "workLaneOccupancy": 0.854296875,
      "deviceUsefulUtilization": 0.04448014969085584,
      "meanCycles": 16194.0,
      "p50Cycles": 16194,
      "p95Cycles": 16194,
      "maximumCycles": 16194,
      "firstOffset": 138522,
      "lastOffset": 156960
    },
    {
      "name": "unpack_amp_to_row_major_f16_o1_r9_p16_c72_p72",
      "dimensions": {},
      "sampleCount": 32,
      "tileCount": 16,
      "phaseCount": 4,
      "phaseCycles": 16770,
      "phaseMs": 0.01118,
      "workCycles": 249600,
      "averageActiveTiles": 14.883720930232558,
      "usefulCycles": 10368.0,
      "estimatedWorkCycles": 249600,
      "usefulUtilization": 0.04153846153846154,
      "workLaneOccupancy": 0.5625,
      "deviceUsefulUtilization": 0.00042000466671851907,
      "meanCycles": 7800.0,
      "p50Cycles": 7800,
      "p95Cycles": 7800,
      "maximumCycles": 7800,
      "firstOffset": 38664,
      "lastOffset": 55434
    },
    {
      "name": "unpack_amp_to_row_major_f16_o1_r16_p16_c72_p72",
      "dimensions": {},
      "sampleCount": 1440,
      "tileCount": 720,
      "phaseCount": 4,
      "phaseCycles": 15360,
      "phaseMs": 0.01024,
      "workCycles": 9987840,
      "averageActiveTiles": 650.25,
      "usefulCycles": 829440.0,
      "estimatedWorkCycles": 9987840,
      "usefulUtilization": 0.08304498269896193,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.036684782608695655,
      "meanCycles": 6936.0,
      "p50Cycles": 6936,
      "p95Cycles": 6936,
      "maximumCycles": 6936,
      "firstOffset": 38628,
      "lastOffset": 53988
    },
    {
      "name": "gemm_f16_init_small_rows_interleaved_k80_c80_r91_r92",
      "dimensions": {},
      "sampleCount": 224,
      "tileCount": 224,
      "phaseCount": 5,
      "phaseCycles": 14244,
      "phaseMs": 0.009496,
      "workCycles": 3033408,
      "averageActiveTiles": 212.9604043807919,
      "usefulCycles": 1834560.0,
      "estimatedWorkCycles": 3033408,
      "usefulUtilization": 0.6047851129818342,
      "workLaneOccupancy": 0.9,
      "deviceUsefulUtilization": 0.08749679498919453,
      "meanCycles": 13542.0,
      "p50Cycles": 13542,
      "p95Cycles": 13542,
      "maximumCycles": 13542,
      "firstOffset": 79932,
      "lastOffset": 94176
    },
    {
      "name": "gemm_f16_init_large_rows_interleaved_k80_c80_r91_r92",
      "dimensions": {},
      "sampleCount": 32,
      "tileCount": 32,
      "phaseCount": 3,
      "phaseCycles": 13902,
      "phaseMs": 0.009268,
      "workCycles": 434304,
      "averageActiveTiles": 31.24039706517048,
      "usefulCycles": 264960.0,
      "estimatedWorkCycles": 434304,
      "usefulUtilization": 0.610079575596817,
      "workLaneOccupancy": 0.9,
      "deviceUsefulUtilization": 0.01294777729823047,
      "meanCycles": 13572.0,
      "p50Cycles": 13572,
      "p95Cycles": 13572,
      "maximumCycles": 13572,
      "firstOffset": 80274,
      "lastOffset": 94176
    },
    {
      "name": "copy_strided_u64",
      "dimensions": {},
      "sampleCount": 1636,
      "tileCount": 1472,
      "phaseCount": 29,
      "phaseCycles": 12510,
      "phaseMs": 0.00834,
      "workCycles": 838536,
      "averageActiveTiles": 67.02925659472422,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 512.5525672371638,
      "p50Cycles": 510,
      "p95Cycles": 618,
      "maximumCycles": 834,
      "firstOffset": 52956,
      "lastOffset": 169740
    },
    {
      "name": "reduce_sum_f16",
      "dimensions": {},
      "sampleCount": 2208,
      "tileCount": 1472,
      "phaseCount": 21,
      "phaseCycles": 12210,
      "phaseMs": 0.00814,
      "workCycles": 7831488,
      "averageActiveTiles": 641.3995085995086,
      "usefulCycles": 2268864.0,
      "estimatedWorkCycles": 7831488,
      "usefulUtilization": 0.28971046115374244,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.12623651319303494,
      "meanCycles": 3546.8695652173915,
      "p50Cycles": 2382,
      "p95Cycles": 5898,
      "maximumCycles": 5898,
      "firstOffset": 27690,
      "lastOffset": 166728
    },
    {
      "name": "gemm_f16_init_small_rows_interleaved_k80_c64_r91_r92",
      "dimensions": {},
      "sampleCount": 1008,
      "tileCount": 1008,
      "phaseCount": 6,
      "phaseCycles": 11886,
      "phaseMs": 0.007924,
      "workCycles": 10983168,
      "averageActiveTiles": 924.0424028268551,
      "usefulCycles": 6524154.0,
      "estimatedWorkCycles": 10983168,
      "usefulUtilization": 0.5940138582966226,
      "workLaneOccupancy": 0.8890625,
      "deviceUsefulUtilization": 0.3728899408511292,
      "meanCycles": 10896.0,
      "p50Cycles": 10896,
      "p95Cycles": 10896,
      "maximumCycles": 10896,
      "firstOffset": 79530,
      "lastOffset": 91416
    },
    {
      "name": "gemm_f16_init_large_rows_interleaved_k80_c64_r91_r92",
      "dimensions": {},
      "sampleCount": 144,
      "tileCount": 144,
      "phaseCount": 5,
      "phaseCycles": 11778,
      "phaseMs": 0.007852,
      "workCycles": 1572480,
      "averageActiveTiles": 133.50993377483445,
      "usefulCycles": 942264.0,
      "estimatedWorkCycles": 1572480,
      "usefulUtilization": 0.5992216117216117,
      "workLaneOccupancy": 0.8890625,
      "deviceUsefulUtilization": 0.05434921039225675,
      "meanCycles": 10920.0,
      "p50Cycles": 10920,
      "p95Cycles": 10920,
      "maximumCycles": 10920,
      "firstOffset": 79746,
      "lastOffset": 91524
    },
    {
      "name": "rearrange_row_major_to_amp_f16_o0_r8_p8_c72_p80",
      "dimensions": {},
      "sampleCount": 1360,
      "tileCount": 1360,
      "phaseCount": 8,
      "phaseCycles": 11232,
      "phaseMs": 0.007488,
      "workCycles": 3215040,
      "averageActiveTiles": 286.2393162393162,
      "usefulCycles": 391680.0,
      "estimatedWorkCycles": 3215040,
      "usefulUtilization": 0.1218274111675127,
      "workLaneOccupancy": 0.9,
      "deviceUsefulUtilization": 0.023690078037904124,
      "meanCycles": 2364.0,
      "p50Cycles": 2364,
      "p95Cycles": 2364,
      "maximumCycles": 2364,
      "firstOffset": 58230,
      "lastOffset": 69462
    },
    {
      "name": "sync",
      "dimensions": {},
      "sampleCount": 1426,
      "tileCount": 1058,
      "phaseCount": 2,
      "phaseCycles": 10662,
      "phaseMs": 0.007108,
      "workCycles": 7139886,
      "averageActiveTiles": 669.657287563309,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 5006.93267882188,
      "p50Cycles": 8532,
      "p95Cycles": 10302,
      "maximumCycles": 10308,
      "firstOffset": 0,
      "lastOffset": 36264
    },
    {
      "name": "rearrange_row_major_to_amp_f16_o1_r64_p64_c16_p16",
      "dimensions": {},
      "sampleCount": 880,
      "tileCount": 880,
      "phaseCount": 8,
      "phaseCycles": 9174,
      "phaseMs": 0.006116,
      "workCycles": 1320000,
      "averageActiveTiles": 143.88489208633092,
      "usefulCycles": 450560.0,
      "estimatedWorkCycles": 1320000,
      "usefulUtilization": 0.3413333333333333,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.033364612657699924,
      "meanCycles": 1500.0,
      "p50Cycles": 1500,
      "p95Cycles": 1500,
      "maximumCycles": 1500,
      "firstOffset": 61788,
      "lastOffset": 70962
    },
    {
      "name": "attention_merge_v72_p80_k768_f16",
      "dimensions": {},
      "sampleCount": 1472,
      "tileCount": 1472,
      "phaseCount": 22,
      "phaseCycles": 7494,
      "phaseMs": 0.004996,
      "workCycles": 3629952,
      "averageActiveTiles": 484.38110488390714,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 2466.0,
      "p50Cycles": 2466,
      "p95Cycles": 2466,
      "maximumCycles": 2466,
      "firstOffset": 164712,
      "lastOffset": 172206
    },
    {
      "name": "rearrange_row_major_to_amp_f16_o1_r0_p64_c0_p16",
      "dimensions": {},
      "sampleCount": 80,
      "tileCount": 80,
      "phaseCount": 2,
      "phaseCycles": 4584,
      "phaseMs": 0.003056,
      "workCycles": 216960,
      "averageActiveTiles": 47.32984293193717,
      "usefulCycles": 16000.0,
      "estimatedWorkCycles": 216960,
      "usefulUtilization": 0.07374631268436578,
      "workLaneOccupancy": 0.390625,
      "deviceUsefulUtilization": 0.002371196600652553,
      "meanCycles": 2712.0,
      "p50Cycles": 2712,
      "p95Cycles": 2712,
      "maximumCycles": 2712,
      "firstOffset": 60966,
      "lastOffset": 65550
    }
  ],
  "samples": []
}
