{
  "clockHz": 1500000000,
  "tileCount": 1472,
  "sampleCount": 50945,
  "matchedSampleCount": 49001,
  "profileSpanCycles": 221250,
  "profileSpanMs": 0.1475,
  "atOffset": null,
  "groups": [
    {
      "name": "exchange",
      "dimensions": {},
      "sampleCount": 14421,
      "tileCount": 1472,
      "phaseCount": 11,
      "phaseCycles": 147138,
      "phaseMs": 0.098092,
      "workCycles": 142196232,
      "averageActiveTiles": 966.4140602699507,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 9860.358643644684,
      "p50Cycles": 9150,
      "p95Cycles": 15840,
      "maximumCycles": 29508,
      "firstOffset": 0,
      "lastOffset": 220632
    },
    {
      "name": "sync",
      "dimensions": {},
      "sampleCount": 1771,
      "tileCount": 986,
      "phaseCount": 5,
      "phaseCycles": 94098,
      "phaseMs": 0.062732,
      "workCycles": 23647860,
      "averageActiveTiles": 251.31097366575273,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 13352.828910220214,
      "p50Cycles": 14466,
      "p95Cycles": 19620,
      "maximumCycles": 31242,
      "firstOffset": 0,
      "lastOffset": 94098
    },
    {
      "name": "copy_strided_u64",
      "dimensions": {},
      "sampleCount": 3429,
      "tileCount": 1472,
      "phaseCount": 49,
      "phaseCycles": 33978,
      "phaseMs": 0.022652,
      "workCycles": 25114902,
      "averageActiveTiles": 739.1518629701571,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 7324.264216972879,
      "p50Cycles": 666,
      "p95Cycles": 16620,
      "maximumCycles": 16620,
      "firstOffset": 0,
      "lastOffset": 214986
    },
    {
      "name": "fill_zero_u64",
      "dimensions": {},
      "sampleCount": 11712,
      "tileCount": 1472,
      "phaseCount": 49,
      "phaseCycles": 32190,
      "phaseMs": 0.02146,
      "workCycles": 4254240,
      "averageActiveTiles": 132.16029822926376,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 2164608,
      "usefulUtilization": 0.0,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": 0.0,
      "meanCycles": 363.23770491803276,
      "p50Cycles": 300,
      "p95Cycles": 600,
      "maximumCycles": 600,
      "firstOffset": 94068,
      "lastOffset": 214062
    },
    {
      "name": "copy_u64",
      "dimensions": {},
      "sampleCount": 5324,
      "tileCount": 1472,
      "phaseCount": 52,
      "phaseCycles": 28254,
      "phaseMs": 0.018836,
      "workCycles": 6071406,
      "averageActiveTiles": 214.8866001274156,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 1140.3842975206612,
      "p50Cycles": 570,
      "p95Cycles": 3702,
      "maximumCycles": 3702,
      "firstOffset": 0,
      "lastOffset": 221250
    },
    {
      "name": "attention_softmax_d72_p768_tail_f16",
      "dimensions": {},
      "sampleCount": 1472,
      "tileCount": 1472,
      "phaseCount": 20,
      "phaseCycles": 26136,
      "phaseMs": 0.017424,
      "workCycles": 34884384,
      "averageActiveTiles": 1334.7254361799817,
      "usefulCycles": 15816384.0,
      "estimatedWorkCycles": 34884384,
      "usefulUtilization": 0.4533943898794372,
      "workLaneOccupancy": 0.9970588235294118,
      "deviceUsefulUtilization": 0.41111210923463887,
      "meanCycles": 23698.630434782608,
      "p50Cycles": 23700,
      "p95Cycles": 23700,
      "maximumCycles": 23700,
      "firstOffset": 146460,
      "lastOffset": 172596
    },
    {
      "name": "gemm_f8_init_large_rows_interleaved_k384_c64_r80_r82",
      "dimensions": {},
      "sampleCount": 810,
      "tileCount": 810,
      "phaseCount": 3,
      "phaseCycles": 24516,
      "phaseMs": 0.016344,
      "workCycles": 19746180,
      "averageActiveTiles": 805.4405286343613,
      "usefulCycles": 12752640.0,
      "estimatedWorkCycles": 19746180,
      "usefulUtilization": 0.6458282057592911,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.3533805784332503,
      "meanCycles": 24378.0,
      "p50Cycles": 24378,
      "p95Cycles": 24378,
      "maximumCycles": 24378,
      "firstOffset": 39132,
      "lastOffset": 63648
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k384_c64_r80_r82",
      "dimensions": {},
      "sampleCount": 648,
      "tileCount": 648,
      "phaseCount": 2,
      "phaseCycles": 24030,
      "phaseMs": 0.01602,
      "workCycles": 15528672,
      "averageActiveTiles": 646.2202247191011,
      "usefulCycles": 9922176.0,
      "estimatedWorkCycles": 15528672,
      "usefulUtilization": 0.6389584376564847,
      "workLaneOccupancy": 0.996875,
      "deviceUsefulUtilization": 0.28050806057645333,
      "meanCycles": 23964.0,
      "p50Cycles": 23964,
      "p95Cycles": 23964,
      "maximumCycles": 23964,
      "firstOffset": 39198,
      "lastOffset": 63228
    },
    {
      "name": "gemm_f16_init_small_rows_interleaved_k80_c80_r91_r92",
      "dimensions": {},
      "sampleCount": 224,
      "tileCount": 224,
      "phaseCount": 8,
      "phaseCycles": 20082,
      "phaseMs": 0.013388,
      "workCycles": 3115392,
      "averageActiveTiles": 155.13355243501644,
      "usefulCycles": 1834560.0,
      "estimatedWorkCycles": 3115392,
      "usefulUtilization": 0.588869715271786,
      "workLaneOccupancy": 0.9,
      "deviceUsefulUtilization": 0.06206076824151414,
      "meanCycles": 13908.0,
      "p50Cycles": 13908,
      "p95Cycles": 13908,
      "maximumCycles": 13908,
      "firstOffset": 121038,
      "lastOffset": 141120
    },
    {
      "name": "gemm_f16_init_small_rows_interleaved_k96_c80_r91_r92",
      "dimensions": {},
      "sampleCount": 896,
      "tileCount": 896,
      "phaseCount": 13,
      "phaseCycles": 19788,
      "phaseMs": 0.013192,
      "workCycles": 14853888,
      "averageActiveTiles": 750.6513038204972,
      "usefulCycles": 8358714.0,
      "estimatedWorkCycles": 14853888,
      "usefulUtilization": 0.5627290309446255,
      "workLaneOccupancy": 0.854296875,
      "deviceUsefulUtilization": 0.2869655440055897,
      "meanCycles": 16578.0,
      "p50Cycles": 16578,
      "p95Cycles": 16578,
      "maximumCycles": 16578,
      "firstOffset": 184392,
      "lastOffset": 204180
    },
    {
      "name": "gemm_f16_init_large_rows_interleaved_k80_c64_r91_r92",
      "dimensions": {},
      "sampleCount": 144,
      "tileCount": 144,
      "phaseCount": 6,
      "phaseCycles": 18288,
      "phaseMs": 0.012192,
      "workCycles": 1616544,
      "averageActiveTiles": 88.39370078740157,
      "usefulCycles": 942264.0,
      "estimatedWorkCycles": 1616544,
      "usefulUtilization": 0.5828879387137004,
      "workLaneOccupancy": 0.8890625,
      "deviceUsefulUtilization": 0.03500246062992126,
      "meanCycles": 11226.0,
      "p50Cycles": 11226,
      "p95Cycles": 11226,
      "maximumCycles": 11226,
      "firstOffset": 120516,
      "lastOffset": 138804
    },
    {
      "name": "gemm_f16_init_small_rows_interleaved_k80_c64_r91_r92",
      "dimensions": {},
      "sampleCount": 1008,
      "tileCount": 1008,
      "phaseCount": 8,
      "phaseCycles": 17964,
      "phaseMs": 0.011976,
      "workCycles": 11291616,
      "averageActiveTiles": 628.5691382765531,
      "usefulCycles": 6524154.0,
      "estimatedWorkCycles": 11291616,
      "usefulUtilization": 0.5777874486698804,
      "workLaneOccupancy": 0.8890625,
      "deviceUsefulUtilization": 0.24672510782434434,
      "meanCycles": 11202.0,
      "p50Cycles": 11202,
      "p95Cycles": 11202,
      "maximumCycles": 11202,
      "firstOffset": 120462,
      "lastOffset": 138426
    },
    {
      "name": "gemm_f16_init_large_rows_interleaved_k96_c80_r91_r92",
      "dimensions": {},
      "sampleCount": 128,
      "tileCount": 128,
      "phaseCount": 11,
      "phaseCycles": 17580,
      "phaseMs": 0.01172,
      "workCycles": 2127360,
      "averageActiveTiles": 121.01023890784982,
      "usefulCycles": 1207224.0,
      "estimatedWorkCycles": 2127360,
      "usefulUtilization": 0.5674751805054151,
      "workLaneOccupancy": 0.854296875,
      "deviceUsefulUtilization": 0.04665102389078498,
      "meanCycles": 16620.0,
      "p50Cycles": 16620,
      "p95Cycles": 16620,
      "maximumCycles": 16620,
      "firstOffset": 186636,
      "lastOffset": 204216
    },
    {
      "name": "reduce_sum_f16",
      "dimensions": {},
      "sampleCount": 1958,
      "tileCount": 1472,
      "phaseCount": 30,
      "phaseCycles": 15384,
      "phaseMs": 0.010256,
      "workCycles": 8380692,
      "averageActiveTiles": 544.7667706708269,
      "usefulCycles": 2894400.0,
      "estimatedWorkCycles": 8380692,
      "usefulUtilization": 0.345365275325713,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.12781489520450384,
      "meanCycles": 4280.230847803881,
      "p50Cycles": 2382,
      "p95Cycles": 10182,
      "maximumCycles": 10182,
      "firstOffset": 71664,
      "lastOffset": 211788
    },
    {
      "name": "gemm_f16_init_large_rows_interleaved_k80_c80_r91_r92",
      "dimensions": {},
      "sampleCount": 32,
      "tileCount": 32,
      "phaseCount": 6,
      "phaseCycles": 14382,
      "phaseMs": 0.009588,
      "workCycles": 446016,
      "averageActiveTiles": 31.012098456403837,
      "usefulCycles": 264960.0,
      "estimatedWorkCycles": 446016,
      "usefulUtilization": 0.594059405940594,
      "workLaneOccupancy": 0.9,
      "deviceUsefulUtilization": 0.012515644555694618,
      "meanCycles": 13938.0,
      "p50Cycles": 13938,
      "p95Cycles": 13938,
      "maximumCycles": 13938,
      "firstOffset": 127122,
      "lastOffset": 141504
    },
    {
      "name": "attention_merge_v72_p80_k768_out16_f16",
      "dimensions": {},
      "sampleCount": 1472,
      "tileCount": 1472,
      "phaseCount": 28,
      "phaseCycles": 7164,
      "phaseMs": 0.004776,
      "workCycles": 4371840,
      "averageActiveTiles": 610.251256281407,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 2970.0,
      "p50Cycles": 2970,
      "p95Cycles": 2970,
      "maximumCycles": 2970,
      "firstOffset": 210792,
      "lastOffset": 217956
    },
    {
      "name": "rearrange_row_major_to_amp_f16_o0_r8_p8_c72_p80",
      "dimensions": {},
      "sampleCount": 1360,
      "tileCount": 1360,
      "phaseCount": 14,
      "phaseCycles": 6330,
      "phaseMs": 0.00422,
      "workCycles": 1721760,
      "averageActiveTiles": 272.0,
      "usefulCycles": 391680.0,
      "estimatedWorkCycles": 1721760,
      "usefulUtilization": 0.22748815165876776,
      "workLaneOccupancy": 0.9,
      "deviceUsefulUtilization": 0.042035854110859264,
      "meanCycles": 1266.0,
      "p50Cycles": 1266,
      "p95Cycles": 1266,
      "maximumCycles": 1266,
      "firstOffset": 101520,
      "lastOffset": 107850
    },
    {
      "name": "rearrange_row_major_to_amp_f16_o2_r64_p64_c16_p16_b64x16",
      "dimensions": {},
      "sampleCount": 880,
      "tileCount": 880,
      "phaseCount": 13,
      "phaseCycles": 5064,
      "phaseMs": 0.003376,
      "workCycles": 2138400,
      "averageActiveTiles": 422.2748815165877,
      "usefulCycles": 450560.0,
      "estimatedWorkCycles": 2138400,
      "usefulUtilization": 0.21069958847736625,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.060443711793392405,
      "meanCycles": 2430.0,
      "p50Cycles": 2430,
      "p95Cycles": 2430,
      "maximumCycles": 2430,
      "firstOffset": 106740,
      "lastOffset": 111804
    },
    {
      "name": "rearrange_row_major_to_amp_f16_o2_r0_p64_c16_p16_b64x16",
      "dimensions": {},
      "sampleCount": 80,
      "tileCount": 80,
      "phaseCount": 8,
      "phaseCycles": 4470,
      "phaseMs": 0.00298,
      "workCycles": 156960,
      "averageActiveTiles": 35.11409395973154,
      "usefulCycles": 16000.0,
      "estimatedWorkCycles": 156960,
      "usefulUtilization": 0.1019367991845056,
      "workLaneOccupancy": 0.390625,
      "deviceUsefulUtilization": 0.0024316700710047662,
      "meanCycles": 1962.0,
      "p50Cycles": 1962,
      "p95Cycles": 1962,
      "maximumCycles": 1962,
      "firstOffset": 106878,
      "lastOffset": 111348
    },
    {
      "name": "rearrange_row_major_to_amp_f16_o1_r64_p64_c16_p16",
      "dimensions": {},
      "sampleCount": 880,
      "tileCount": 880,
      "phaseCount": 13,
      "phaseCycles": 4182,
      "phaseMs": 0.002788,
      "workCycles": 1351680,
      "averageActiveTiles": 323.2137733142037,
      "usefulCycles": 450560.0,
      "estimatedWorkCycles": 1351680,
      "usefulUtilization": 0.3333333333333333,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.07319152475412222,
      "meanCycles": 1536.0,
      "p50Cycles": 1536,
      "p95Cycles": 1536,
      "maximumCycles": 1536,
      "firstOffset": 105204,
      "lastOffset": 109386
    }
  ],
  "samples": []
}
