{
  "clockHz": 1500000000,
  "tileCount": 1472,
  "sampleCount": 202925,
  "matchedSampleCount": 202925,
  "profileSpanCycles": 779868,
  "profileSpanMs": 0.519912,
  "atOffset": null,
  "groups": [
    {
      "name": "sync",
      "dimensions": {},
      "sampleCount": 14269,
      "tileCount": 1468,
      "phaseCount": 44,
      "phaseCycles": 629760,
      "phaseMs": 0.41984,
      "workCycles": 87447564,
      "averageActiveTiles": 138.8585556402439,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 6128.49982479501,
      "p50Cycles": 2934,
      "p95Cycles": 19332,
      "maximumCycles": 63984,
      "firstOffset": 3228,
      "lastOffset": 778650
    },
    {
      "name": "exchange",
      "dimensions": {},
      "sampleCount": 63747,
      "tileCount": 1472,
      "phaseCount": 53,
      "phaseCycles": 484428,
      "phaseMs": 0.322952,
      "workCycles": 466060350,
      "averageActiveTiles": 962.083839084446,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 7311.094639747753,
      "p50Cycles": 5670,
      "p95Cycles": 22860,
      "maximumCycles": 35220,
      "firstOffset": 3228,
      "lastOffset": 778848
    },
    {
      "name": "copy_u64",
      "dimensions": {},
      "sampleCount": 28196,
      "tileCount": 1472,
      "phaseCount": 207,
      "phaseCycles": 109878,
      "phaseMs": 0.073252,
      "workCycles": 21545052,
      "averageActiveTiles": 196.0815813902692,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 764.1173216059016,
      "p50Cycles": 558,
      "p95Cycles": 1398,
      "maximumCycles": 10578,
      "firstOffset": 12120,
      "lastOffset": 778368
    },
    {
      "name": "cast_f16_f8",
      "dimensions": {},
      "sampleCount": 13215,
      "tileCount": 1472,
      "phaseCount": 169,
      "phaseCycles": 68700,
      "phaseMs": 0.0458,
      "workCycles": 51106314,
      "averageActiveTiles": 743.9055895196507,
      "usefulCycles": 7704273.5,
      "estimatedWorkCycles": 46078266,
      "usefulUtilization": 0.1671997270904248,
      "workLaneOccupancy": 0.9392669759407941,
      "deviceUsefulUtilization": 0.076184591758433,
      "meanCycles": 3867.2958002270148,
      "p50Cycles": 3654,
      "p95Cycles": 8238,
      "maximumCycles": 9102,
      "firstOffset": 0,
      "lastOffset": 771960
    },
    {
      "name": "reduce_sum_f16",
      "dimensions": {},
      "sampleCount": 10659,
      "tileCount": 1472,
      "phaseCount": 181,
      "phaseCycles": 58068,
      "phaseMs": 0.038712,
      "workCycles": 39338838,
      "averageActiveTiles": 677.4615623062616,
      "usefulCycles": 21041112.0,
      "estimatedWorkCycles": 39338838,
      "usefulUtilization": 0.5348686710065,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.24616369938093569,
      "meanCycles": 3690.668730650155,
      "p50Cycles": 2658,
      "p95Cycles": 8994,
      "maximumCycles": 9576,
      "firstOffset": 12654,
      "lastOffset": 778098
    },
    {
      "name": "layer_norm_f16",
      "dimensions": {},
      "sampleCount": 2187,
      "tileCount": 729,
      "phaseCount": 59,
      "phaseCycles": 50100,
      "phaseMs": 0.0334,
      "workCycles": 33985980,
      "averageActiveTiles": 678.362874251497,
      "usefulCycles": 10077696.0,
      "estimatedWorkCycles": 33985980,
      "usefulUtilization": 0.2965250965250965,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.136651913564176,
      "meanCycles": 15540.0,
      "p50Cycles": 15540,
      "p95Cycles": 15540,
      "maximumCycles": 15540,
      "firstOffset": 28218,
      "lastOffset": 598662
    },
    {
      "name": "fill_zero_u64",
      "dimensions": {},
      "sampleCount": 15136,
      "tileCount": 1472,
      "phaseCount": 136,
      "phaseCycles": 47628,
      "phaseMs": 0.031752,
      "workCycles": 5601120,
      "averageActiveTiles": 117.60141093474427,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 2901888,
      "usefulUtilization": 0.0,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": 0.0,
      "meanCycles": 370.05285412262157,
      "p50Cycles": 300,
      "p95Cycles": 600,
      "maximumCycles": 3150,
      "firstOffset": 143592,
      "lastOffset": 769710
    },
    {
      "name": "gemm_f8_init_large_rows_interleaved_k384_c192_r36_r38",
      "dimensions": {},
      "sampleCount": 360,
      "tileCount": 360,
      "phaseCount": 26,
      "phaseCycles": 45198,
      "phaseMs": 0.030132,
      "workCycles": 16249680,
      "averageActiveTiles": 359.5221027479092,
      "usefulCycles": 7359840.0,
      "estimatedWorkCycles": 16249680,
      "usefulUtilization": 0.4529221498515663,
      "workLaneOccupancy": 0.9340277777777778,
      "deviceUsefulUtilization": 0.11062195903243122,
      "meanCycles": 45138.0,
      "p50Cycles": 45138,
      "p95Cycles": 45138,
      "maximumCycles": 45138,
      "firstOffset": 517944,
      "lastOffset": 563142
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k384_c192_r36_r38",
      "dimensions": {},
      "sampleCount": 1080,
      "tileCount": 1080,
      "phaseCount": 31,
      "phaseCycles": 44832,
      "phaseMs": 0.029888,
      "workCycles": 48366720,
      "averageActiveTiles": 1078.8436830835117,
      "usefulCycles": 20878704.0,
      "estimatedWorkCycles": 48366720,
      "usefulUtilization": 0.4316750029772538,
      "workLaneOccupancy": 0.9322980967078189,
      "deviceUsefulUtilization": 0.31637897425751793,
      "meanCycles": 44784.0,
      "p50Cycles": 44784,
      "p95Cycles": 44784,
      "maximumCycles": 44784,
      "firstOffset": 517938,
      "lastOffset": 562770
    },
    {
      "name": "add_f16",
      "dimensions": {},
      "sampleCount": 12221,
      "tileCount": 1458,
      "phaseCount": 185,
      "phaseCycles": 40734,
      "phaseMs": 0.027156,
      "workCycles": 45913638,
      "averageActiveTiles": 1127.1576078951246,
      "usefulCycles": 2311700.0,
      "estimatedWorkCycles": 45913638,
      "usefulUtilization": 0.05034887455444067,
      "workLaneOccupancy": 0.999037135122614,
      "deviceUsefulUtilization": 0.0385537479639912,
      "meanCycles": 3756.9460764258242,
      "p50Cycles": 2940,
      "p95Cycles": 8124,
      "maximumCycles": 8124,
      "firstOffset": 19644,
      "lastOffset": 779868
    },
    {
      "name": "gemm_f8_init_large_rows_interleaved_k160_c240_r72_r74",
      "dimensions": {},
      "sampleCount": 680,
      "tileCount": 680,
      "phaseCount": 27,
      "phaseCycles": 34842,
      "phaseMs": 0.023228,
      "workCycles": 23594640,
      "averageActiveTiles": 677.1895987601171,
      "usefulCycles": 13586400.0,
      "estimatedWorkCycles": 23594640,
      "usefulUtilization": 0.5758256960055335,
      "workLaneOccupancy": 0.9,
      "deviceUsefulUtilization": 0.2649070462185818,
      "meanCycles": 34698.0,
      "p50Cycles": 34698,
      "p95Cycles": 34698,
      "maximumCycles": 34698,
      "firstOffset": 378132,
      "lastOffset": 412974
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k160_c240_r72_r74",
      "dimensions": {},
      "sampleCount": 680,
      "tileCount": 680,
      "phaseCount": 20,
      "phaseCycles": 34602,
      "phaseMs": 0.023068,
      "workCycles": 23443680,
      "averageActiveTiles": 677.5238425524536,
      "usefulCycles": 13182480.0,
      "estimatedWorkCycles": 23443680,
      "usefulUtilization": 0.5623042116254786,
      "workLaneOccupancy": 0.8975,
      "deviceUsefulUtilization": 0.2588142052608168,
      "meanCycles": 34476.0,
      "p50Cycles": 34476,
      "p95Cycles": 34476,
      "maximumCycles": 34476,
      "firstOffset": 378126,
      "lastOffset": 412728
    },
    {
      "name": "attention_softmax_d72_p768_tail_f16",
      "dimensions": {},
      "sampleCount": 1488,
      "tileCount": 1472,
      "phaseCount": 33,
      "phaseCycles": 34254,
      "phaseMs": 0.022836,
      "workCycles": 37251936,
      "averageActiveTiles": 1087.5207566999475,
      "usefulCycles": 18466080.0,
      "estimatedWorkCycles": 37251936,
      "usefulUtilization": 0.4957079277705191,
      "workLaneOccupancy": 0.9974763406940063,
      "deviceUsefulUtilization": 0.3662314271135583,
      "meanCycles": 25034.90322580645,
      "p50Cycles": 25236,
      "p95Cycles": 25236,
      "maximumCycles": 25236,
      "firstOffset": 190248,
      "lastOffset": 720234
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k160_c224_r72_r74",
      "dimensions": {},
      "sampleCount": 40,
      "tileCount": 40,
      "phaseCount": 14,
      "phaseCycles": 32862,
      "phaseMs": 0.021908,
      "workCycles": 1287840,
      "averageActiveTiles": 39.18933722840971,
      "usefulCycles": 723744.0,
      "estimatedWorkCycles": 1287840,
      "usefulUtilization": 0.5619828550130451,
      "workLaneOccupancy": 0.8975,
      "deviceUsefulUtilization": 0.014961776916909447,
      "meanCycles": 32196.0,
      "p50Cycles": 32196,
      "p95Cycles": 32196,
      "maximumCycles": 32196,
      "firstOffset": 377490,
      "lastOffset": 410352
    },
    {
      "name": "gemm_f8_init_large_rows_interleaved_k160_c224_r72_r74",
      "dimensions": {},
      "sampleCount": 40,
      "tileCount": 40,
      "phaseCount": 11,
      "phaseCycles": 32520,
      "phaseMs": 0.02168,
      "workCycles": 1296240,
      "averageActiveTiles": 39.859778597785976,
      "usefulCycles": 745920.0,
      "estimatedWorkCycles": 1296240,
      "usefulUtilization": 0.575448990927606,
      "workLaneOccupancy": 0.9,
      "deviceUsefulUtilization": 0.015582384084710412,
      "meanCycles": 32406.0,
      "p50Cycles": 32406,
      "p95Cycles": 32406,
      "maximumCycles": 32406,
      "firstOffset": 377508,
      "lastOffset": 410028
    },
    {
      "name": "static_copy_u32",
      "dimensions": {},
      "sampleCount": 1834,
      "tileCount": 788,
      "phaseCount": 116,
      "phaseCycles": 30132,
      "phaseMs": 0.020088,
      "workCycles": 524232,
      "averageActiveTiles": 17.397849462365592,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 285.84078516902946,
      "p50Cycles": 288,
      "p95Cycles": 288,
      "maximumCycles": 288,
      "firstOffset": 112116,
      "lastOffset": 770490
    },
    {
      "name": "copy_strided_u64",
      "dimensions": {},
      "sampleCount": 3344,
      "tileCount": 1472,
      "phaseCount": 108,
      "phaseCycles": 28194,
      "phaseMs": 0.018796,
      "workCycles": 1390944,
      "averageActiveTiles": 49.33475207490955,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 415.95215311004785,
      "p50Cycles": 414,
      "p95Cycles": 714,
      "maximumCycles": 714,
      "firstOffset": 119676,
      "lastOffset": 711228
    },
    {
      "name": "gemm_f8_init_large_rows_interleaved_k96_c320_r66_r68",
      "dimensions": {},
      "sampleCount": 168,
      "tileCount": 168,
      "phaseCount": 3,
      "phaseCycles": 26694,
      "phaseMs": 0.017796,
      "workCycles": 4468464,
      "averageActiveTiles": 167.39581928523265,
      "usefulCycles": 2741760.0,
      "estimatedWorkCycles": 4468464,
      "usefulUtilization": 0.6135799684186781,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.06977630537394823,
      "meanCycles": 26598.0,
      "p50Cycles": 26598,
      "p95Cycles": 26598,
      "maximumCycles": 26598,
      "firstOffset": 69690,
      "lastOffset": 96384
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k96_c320_r66_r68",
      "dimensions": {},
      "sampleCount": 756,
      "tileCount": 756,
      "phaseCount": 3,
      "phaseCycles": 26454,
      "phaseMs": 0.017636,
      "workCycles": 19949328,
      "averageActiveTiles": 754.1138580176911,
      "usefulCycles": 11954880.0,
      "estimatedWorkCycles": 19949328,
      "usefulUtilization": 0.5992622909403265,
      "workLaneOccupancy": 0.9983164983164983,
      "deviceUsefulUtilization": 0.3070054335499522,
      "meanCycles": 26388.0,
      "p50Cycles": 26388,
      "p95Cycles": 26388,
      "maximumCycles": 26388,
      "firstOffset": 69762,
      "lastOffset": 96216
    },
    {
      "name": "gemm_f8_init_large_rows_interleaved_k96_c304_r66_r68",
      "dimensions": {},
      "sampleCount": 96,
      "tileCount": 96,
      "phaseCount": 2,
      "phaseCycles": 25380,
      "phaseMs": 0.01692,
      "workCycles": 2427264,
      "averageActiveTiles": 95.63687943262411,
      "usefulCycles": 1488384.0,
      "estimatedWorkCycles": 2427264,
      "usefulUtilization": 0.6131941148552444,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.03983965464076472,
      "meanCycles": 25284.0,
      "p50Cycles": 25284,
      "p95Cycles": 25284,
      "maximumCycles": 25284,
      "firstOffset": 69306,
      "lastOffset": 94686
    }
  ],
  "samples": []
}
