{
  "clockHz": 1500000000,
  "tileCount": 1472,
  "sampleCount": 227478,
  "matchedSampleCount": 123406,
  "profileSpanCycles": 16600566,
  "profileSpanMs": 11.067044,
  "atOffset": null,
  "groups": [
    {
      "name": "repeat-remainder",
      "dimensions": {},
      "sampleCount": 1472,
      "tileCount": 1472,
      "phaseCount": 9,
      "phaseCycles": 15686622,
      "phaseMs": 10.457748,
      "workCycles": 23085076722,
      "averageActiveTiles": 1471.6410404993503,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 15682796.686141305,
      "p50Cycles": 15682824,
      "p95Cycles": 15682896,
      "maximumCycles": 15682920,
      "firstOffset": 709056,
      "lastOffset": 16395678
    },
    {
      "name": "reduce_sum_f16",
      "dimensions": {},
      "sampleCount": 22066,
      "tileCount": 1472,
      "phaseCount": 144,
      "phaseCycles": 116388,
      "phaseMs": 0.077592,
      "workCycles": 81713442,
      "averageActiveTiles": 702.0778946283122,
      "usefulCycles": 26314664.0,
      "estimatedWorkCycles": 81713442,
      "usefulUtilization": 0.3220359264758422,
      "workLaneOccupancy": 0.9895200907294152,
      "deviceUsefulUtilization": 0.15359667476551445,
      "meanCycles": 3703.1379497870025,
      "p50Cycles": 2154,
      "p95Cycles": 8748,
      "maximumCycles": 10386,
      "firstOffset": 24534,
      "lastOffset": 16595724
    },
    {
      "name": "fill_zero_u64",
      "dimensions": {},
      "sampleCount": 27054,
      "tileCount": 1472,
      "phaseCount": 184,
      "phaseCycles": 76722,
      "phaseMs": 0.051148,
      "workCycles": 9932526,
      "averageActiveTiles": 129.4612497067334,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 4822134,
      "usefulUtilization": 0.0,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": 0.0,
      "meanCycles": 367.1370592149035,
      "p50Cycles": 300,
      "p95Cycles": 816,
      "maximumCycles": 3150,
      "firstOffset": 3048,
      "lastOffset": 16588182
    },
    {
      "name": "gemm_f8_init_large_rows_interleaved_k384_c192_r72_r76",
      "dimensions": {},
      "sampleCount": 360,
      "tileCount": 360,
      "phaseCount": 13,
      "phaseCycles": 73530,
      "phaseMs": 0.04902,
      "workCycles": 23915520,
      "averageActiveTiles": 325.2484700122399,
      "usefulCycles": 14719680.0,
      "estimatedWorkCycles": 23915520,
      "usefulUtilization": 0.6154865125240848,
      "workLaneOccupancy": 0.9340277777777778,
      "deviceUsefulUtilization": 0.13599595551061677,
      "meanCycles": 66432.0,
      "p50Cycles": 66432,
      "p95Cycles": 66432,
      "maximumCycles": 66432,
      "firstOffset": 616950,
      "lastOffset": 690480
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k384_c192_r72_r76",
      "dimensions": {},
      "sampleCount": 1080,
      "tileCount": 1080,
      "phaseCount": 36,
      "phaseCycles": 71886,
      "phaseMs": 0.047924,
      "workCycles": 70035840,
      "averageActiveTiles": 974.2625824221684,
      "usefulCycles": 41757408.0,
      "estimatedWorkCycles": 70035840,
      "usefulUtilization": 0.5962291306850892,
      "workLaneOccupancy": 0.9322980967078189,
      "deviceUsefulUtilization": 0.39462210093517636,
      "meanCycles": 64848.0,
      "p50Cycles": 64848,
      "p95Cycles": 64848,
      "maximumCycles": 64848,
      "firstOffset": 616926,
      "lastOffset": 688812
    },
    {
      "name": "cast_f16_f8",
      "dimensions": {},
      "sampleCount": 14715,
      "tileCount": 1472,
      "phaseCount": 138,
      "phaseCycles": 60006,
      "phaseMs": 0.040004,
      "workCycles": 46824000,
      "averageActiveTiles": 780.3219678032196,
      "usefulCycles": 2392724.0,
      "estimatedWorkCycles": 44695092,
      "usefulUtilization": 0.05353437912153755,
      "workLaneOccupancy": 0.8987877549425881,
      "deviceUsefulUtilization": 0.02708882191490996,
      "meanCycles": 3182.059123343527,
      "p50Cycles": 2832,
      "p95Cycles": 9300,
      "maximumCycles": 9300,
      "firstOffset": 0,
      "lastOffset": 16587336
    },
    {
      "name": "attention_softmax_d72_p768_tail_f8_sm4",
      "dimensions": {},
      "sampleCount": 1472,
      "tileCount": 1472,
      "phaseCount": 16,
      "phaseCycles": 53334,
      "phaseMs": 0.035556,
      "workCycles": 71468520,
      "averageActiveTiles": 1340.0179997750029,
      "usefulCycles": 33499008.0,
      "estimatedWorkCycles": 71468520,
      "usefulUtilization": 0.4687239640613797,
      "workLaneOccupancy": 0.9972222222222222,
      "deviceUsefulUtilization": 0.42669738367400845,
      "meanCycles": 48551.983695652176,
      "p50Cycles": 48552,
      "p95Cycles": 48552,
      "maximumCycles": 48552,
      "firstOffset": 264954,
      "lastOffset": 318288
    },
    {
      "name": "gemm_f8_init_large_rows_interleaved_k384_c80_r160_r164",
      "dimensions": {},
      "sampleCount": 795,
      "tileCount": 795,
      "phaseCount": 13,
      "phaseCycles": 49200,
      "phaseMs": 0.0328,
      "workCycles": 38951820,
      "averageActiveTiles": 791.7036585365854,
      "usefulCycles": 31291200.0,
      "estimatedWorkCycles": 38951820,
      "usefulUtilization": 0.8033308841538085,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.4320652173913043,
      "meanCycles": 48996.0,
      "p50Cycles": 48996,
      "p95Cycles": 48996,
      "maximumCycles": 48996,
      "firstOffset": 478002,
      "lastOffset": 527202
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k384_c80_r160_r164",
      "dimensions": {},
      "sampleCount": 636,
      "tileCount": 636,
      "phaseCount": 16,
      "phaseCycles": 48462,
      "phaseMs": 0.032308,
      "workCycles": 30577608,
      "averageActiveTiles": 630.9605051380463,
      "usefulCycles": 24346080.0,
      "estimatedWorkCycles": 30577608,
      "usefulUtilization": 0.7962061649819044,
      "workLaneOccupancy": 0.996875,
      "deviceUsefulUtilization": 0.34128712231726155,
      "meanCycles": 48078.0,
      "p50Cycles": 48078,
      "p95Cycles": 48078,
      "maximumCycles": 48078,
      "firstOffset": 477726,
      "lastOffset": 526188
    },
    {
      "name": "copy_u64",
      "dimensions": {},
      "sampleCount": 10040,
      "tileCount": 1472,
      "phaseCount": 108,
      "phaseCycles": 48132,
      "phaseMs": 0.032088,
      "workCycles": 14454576,
      "averageActiveTiles": 300.3111443530292,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 1439.6988047808766,
      "p50Cycles": 834,
      "p95Cycles": 3018,
      "maximumCycles": 10578,
      "firstOffset": 94614,
      "lastOffset": 16590366
    },
    {
      "name": "gemm_f8_init_large_rows_interleaved_k288_c96_r144_r148",
      "dimensions": {},
      "sampleCount": 720,
      "tileCount": 720,
      "phaseCount": 1,
      "phaseCycles": 40884,
      "phaseMs": 0.027256,
      "workCycles": 29350080,
      "averageActiveTiles": 717.886703845025,
      "usefulCycles": 23016960.0,
      "estimatedWorkCycles": 29350080,
      "usefulUtilization": 0.7842213717986458,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.38246066282972396,
      "meanCycles": 40764.0,
      "p50Cycles": 40764,
      "p95Cycles": 40764,
      "maximumCycles": 40764,
      "firstOffset": 131040,
      "lastOffset": 171924
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k288_c96_r144_r148",
      "dimensions": {},
      "sampleCount": 720,
      "tileCount": 720,
      "phaseCount": 3,
      "phaseCycles": 40410,
      "phaseMs": 0.02694,
      "workCycles": 28922400,
      "averageActiveTiles": 715.7238307349666,
      "usefulCycles": 22332672.0,
      "estimatedWorkCycles": 28922400,
      "usefulUtilization": 0.7721583271097834,
      "workLaneOccupancy": 0.9972222222222222,
      "deviceUsefulUtilization": 0.3754430134598625,
      "meanCycles": 40170.0,
      "p50Cycles": 40170,
      "p95Cycles": 40170,
      "maximumCycles": 40170,
      "firstOffset": 130854,
      "lastOffset": 171264
    },
    {
      "name": "gemm_f8_init_large_rows_interleaved_k384_c64_r160_r164",
      "dimensions": {},
      "sampleCount": 15,
      "tileCount": 15,
      "phaseCount": 9,
      "phaseCycles": 39468,
      "phaseMs": 0.026312,
      "workCycles": 589140,
      "averageActiveTiles": 14.927029492246884,
      "usefulCycles": 472320.0,
      "estimatedWorkCycles": 589140,
      "usefulUtilization": 0.8017109685304002,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.008129866352928734,
      "meanCycles": 39276.0,
      "p50Cycles": 39276,
      "p95Cycles": 39276,
      "maximumCycles": 39276,
      "firstOffset": 478008,
      "lastOffset": 517476
    },
    {
      "name": "copy_strided_u64",
      "dimensions": {},
      "sampleCount": 6961,
      "tileCount": 1472,
      "phaseCount": 83,
      "phaseCycles": 39276,
      "phaseMs": 0.026184,
      "workCycles": 9414264,
      "averageActiveTiles": 239.69508096547509,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 1352.4298233012498,
      "p50Cycles": 906,
      "p95Cycles": 3060,
      "maximumCycles": 3060,
      "firstOffset": 23628,
      "lastOffset": 16517622
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k384_c64_r160_r164",
      "dimensions": {},
      "sampleCount": 12,
      "tileCount": 12,
      "phaseCount": 10,
      "phaseCycles": 38850,
      "phaseMs": 0.0259,
      "workCycles": 462528,
      "averageActiveTiles": 11.905482625482625,
      "usefulCycles": 367488.0,
      "estimatedWorkCycles": 462528,
      "usefulUtilization": 0.7945205479452054,
      "workLaneOccupancy": 0.996875,
      "deviceUsefulUtilization": 0.006426053382575122,
      "meanCycles": 38544.0,
      "p50Cycles": 38544,
      "p95Cycles": 38544,
      "maximumCycles": 38544,
      "firstOffset": 477732,
      "lastOffset": 516582
    },
    {
      "name": "gemm_f8_init_large_rows_interleaved_k96_c576_r20_r24",
      "dimensions": {},
      "sampleCount": 1440,
      "tileCount": 1440,
      "phaseCount": 21,
      "phaseCycles": 29940,
      "phaseMs": 0.01996,
      "workCycles": 41489280,
      "averageActiveTiles": 1385.74749498998,
      "usefulCycles": 14929920.0,
      "estimatedWorkCycles": 41489280,
      "usefulUtilization": 0.35985006247396917,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.3387644854927246,
      "meanCycles": 28812.0,
      "p50Cycles": 28812,
      "p95Cycles": 28812,
      "maximumCycles": 28812,
      "firstOffset": 385884,
      "lastOffset": 415824
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k192_c96_r144_r148",
      "dimensions": {},
      "sampleCount": 720,
      "tileCount": 720,
      "phaseCount": 6,
      "phaseCycles": 29616,
      "phaseMs": 0.019744,
      "workCycles": 19418400,
      "averageActiveTiles": 655.6726094003242,
      "usefulCycles": 14888448.0,
      "estimatedWorkCycles": 19418400,
      "usefulUtilization": 0.7667185761957731,
      "workLaneOccupancy": 0.9972222222222222,
      "deviceUsefulUtilization": 0.3415192727785216,
      "meanCycles": 26970.0,
      "p50Cycles": 26970,
      "p95Cycles": 26970,
      "maximumCycles": 26970,
      "firstOffset": 16423680,
      "lastOffset": 16453296
    },
    {
      "name": "gemm_f8_init_large_rows_interleaved_k192_c96_r144_r148",
      "dimensions": {},
      "sampleCount": 720,
      "tileCount": 720,
      "phaseCount": 4,
      "phaseCycles": 27648,
      "phaseMs": 0.018432,
      "workCycles": 19699200,
      "averageActiveTiles": 712.5,
      "usefulCycles": 15344640.0,
      "estimatedWorkCycles": 19699200,
      "usefulUtilization": 0.7789473684210526,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.37703804347826086,
      "meanCycles": 27360.0,
      "p50Cycles": 27360,
      "p95Cycles": 27360,
      "maximumCycles": 27360,
      "firstOffset": 16426122,
      "lastOffset": 16453770
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k96_c576_r20_r24",
      "dimensions": {},
      "sampleCount": 24,
      "tileCount": 24,
      "phaseCount": 2,
      "phaseCycles": 27588,
      "phaseMs": 0.018392,
      "workCycles": 637056,
      "averageActiveTiles": 23.091779034362766,
      "usefulCycles": 186624.0,
      "estimatedWorkCycles": 637056,
      "usefulUtilization": 0.29294755877034356,
      "workLaneOccupancy": 0.9,
      "deviceUsefulUtilization": 0.004595570853111939,
      "meanCycles": 26544.0,
      "p50Cycles": 26544,
      "p95Cycles": 26544,
      "maximumCycles": 26544,
      "firstOffset": 385554,
      "lastOffset": 413142
    },
    {
      "name": "add_f16",
      "dimensions": {},
      "sampleCount": 11220,
      "tileCount": 1472,
      "phaseCount": 150,
      "phaseCycles": 27420,
      "phaseMs": 0.01828,
      "workCycles": 20423970,
      "averageActiveTiles": 744.8566739606127,
      "usefulCycles": 4623400.0,
      "estimatedWorkCycles": 20423970,
      "usefulUtilization": 0.22637126866128376,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.11454765642342943,
      "meanCycles": 1820.3181818181818,
      "p50Cycles": 1374,
      "p95Cycles": 4314,
      "maximumCycles": 4314,
      "firstOffset": 91386,
      "lastOffset": 16600566
    },
    {
      "name": "layer_norm_f16",
      "dimensions": {},
      "sampleCount": 4374,
      "tileCount": 1458,
      "phaseCount": 35,
      "phaseCycles": 22854,
      "phaseMs": 0.015236,
      "workCycles": 29463264,
      "averageActiveTiles": 1289.1950643213443,
      "usefulCycles": 11967264.0,
      "estimatedWorkCycles": 29463264,
      "usefulUtilization": 0.40617577197149646,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.3557335601036447,
      "meanCycles": 6736.0,
      "p50Cycles": 6738,
      "p95Cycles": 6738,
      "maximumCycles": 6738,
      "firstOffset": 99870,
      "lastOffset": 16405410
    },
    {
      "name": "bias_gelu_f16",
      "dimensions": {},
      "sampleCount": 1458,
      "tileCount": 1458,
      "phaseCount": 32,
      "phaseCycles": 17220,
      "phaseMs": 0.01148,
      "workCycles": 24284448,
      "averageActiveTiles": 1410.2466898954704,
      "usefulCycles": 17256888.0,
      "estimatedWorkCycles": 24284448,
      "usefulUtilization": 0.7106147934678194,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.6808030980154522,
      "meanCycles": 16656.0,
      "p50Cycles": 16656,
      "p95Cycles": 16656,
      "maximumCycles": 16656,
      "firstOffset": 553656,
      "lastOffset": 570876
    },
    {
      "name": "gemm_f8_init_large_rows_interleaved_k96_c112_r121_r122",
      "dimensions": {},
      "sampleCount": 384,
      "tileCount": 384,
      "phaseCount": 4,
      "phaseCycles": 14196,
      "phaseMs": 0.009464,
      "workCycles": 5398272,
      "averageActiveTiles": 380.2671174978867,
      "usefulCycles": 2951424.0,
      "estimatedWorkCycles": 5398272,
      "usefulUtilization": 0.5467349551856594,
      "workLaneOccupancy": 0.75,
      "deviceUsefulUtilization": 0.1412400308721379,
      "meanCycles": 14058.0,
      "p50Cycles": 14058,
      "p95Cycles": 14058,
      "maximumCycles": 14058,
      "firstOffset": 243132,
      "lastOffset": 257328
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k96_c112_r121_r122",
      "dimensions": {},
      "sampleCount": 384,
      "tileCount": 384,
      "phaseCount": 4,
      "phaseCycles": 14166,
      "phaseMs": 0.009444,
      "workCycles": 5384448,
      "averageActiveTiles": 380.09656925031766,
      "usefulCycles": 2927232.0,
      "estimatedWorkCycles": 5384448,
      "usefulUtilization": 0.5436456996148908,
      "workLaneOccupancy": 0.75,
      "deviceUsefulUtilization": 0.14037898458648693,
      "meanCycles": 14022.0,
      "p50Cycles": 14022,
      "p95Cycles": 14022,
      "maximumCycles": 14022,
      "firstOffset": 243156,
      "lastOffset": 257322
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k128_c80_r104_r105",
      "dimensions": {},
      "sampleCount": 1152,
      "tileCount": 1152,
      "phaseCount": 14,
      "phaseCycles": 14040,
      "phaseMs": 0.00936,
      "workCycles": 13727232,
      "averageActiveTiles": 977.723076923077,
      "usefulCycles": 8188128.0,
      "estimatedWorkCycles": 13727232,
      "usefulUtilization": 0.596487915407855,
      "workLaneOccupancy": 0.854296875,
      "deviceUsefulUtilization": 0.39619565217391306,
      "meanCycles": 11916.0,
      "p50Cycles": 11916,
      "p95Cycles": 11916,
      "maximumCycles": 11916,
      "firstOffset": 326004,
      "lastOffset": 340044
    },
    {
      "name": "gemm_f8_init_large_rows_interleaved_k64_c48_r240_r244",
      "dimensions": {},
      "sampleCount": 1200,
      "tileCount": 1200,
      "phaseCount": 4,
      "phaseCycles": 13998,
      "phaseMs": 0.009332,
      "workCycles": 8704800,
      "averageActiveTiles": 621.8602657522504,
      "usefulCycles": 6456240.0,
      "estimatedWorkCycles": 8704800,
      "usefulUtilization": 0.7416873449131514,
      "workLaneOccupancy": 0.91875,
      "deviceUsefulUtilization": 0.3133328053075905,
      "meanCycles": 7254.0,
      "p50Cycles": 7254,
      "p95Cycles": 7254,
      "maximumCycles": 7254,
      "firstOffset": 8688,
      "lastOffset": 22686
    },
    {
      "name": "gemm_f8_init_large_rows_interleaved_k128_c80_r104_r105",
      "dimensions": {},
      "sampleCount": 192,
      "tileCount": 192,
      "phaseCount": 4,
      "phaseCycles": 13578,
      "phaseMs": 0.009052,
      "workCycles": 2297088,
      "averageActiveTiles": 169.17719840919133,
      "usefulCycles": 1377810.0,
      "estimatedWorkCycles": 2297088,
      "usefulUtilization": 0.5998072342026078,
      "workLaneOccupancy": 0.854296875,
      "deviceUsefulUtilization": 0.06893594257334434,
      "meanCycles": 11964.0,
      "p50Cycles": 11964,
      "p95Cycles": 11964,
      "maximumCycles": 11964,
      "firstOffset": 326808,
      "lastOffset": 340386
    },
    {
      "name": "gemm_f8_init_large_rows_interleaved_k96_c96_r121_r122",
      "dimensions": {},
      "sampleCount": 288,
      "tileCount": 288,
      "phaseCount": 3,
      "phaseCycles": 12678,
      "phaseMs": 0.008452,
      "workCycles": 3487104,
      "averageActiveTiles": 275.0515854235684,
      "usefulCycles": 1851228.0,
      "estimatedWorkCycles": 3487104,
      "usefulUtilization": 0.5308783448959365,
      "workLaneOccupancy": 0.7317708333333334,
      "deviceUsefulUtilization": 0.09919764295561637,
      "meanCycles": 12108.0,
      "p50Cycles": 12108,
      "p95Cycles": 12108,
      "maximumCycles": 12108,
      "firstOffset": 242556,
      "lastOffset": 255234
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k64_c48_r240_r244",
      "dimensions": {},
      "sampleCount": 240,
      "tileCount": 240,
      "phaseCount": 2,
      "phaseCycles": 12408,
      "phaseMs": 0.008272,
      "workCycles": 1726560,
      "averageActiveTiles": 139.14893617021278,
      "usefulCycles": 1259496.0,
      "estimatedWorkCycles": 1726560,
      "usefulUtilization": 0.7294829024186822,
      "workLaneOccupancy": 0.91109375,
      "deviceUsefulUtilization": 0.06895840341434699,
      "meanCycles": 7194.0,
      "p50Cycles": 7194,
      "p95Cycles": 7194,
      "maximumCycles": 7194,
      "firstOffset": 8832,
      "lastOffset": 21240
    },
    {
      "name": "gemm_f16_init_small_rows_interleaved_k736_c16_r1_r1",
      "dimensions": {},
      "sampleCount": 160,
      "tileCount": 160,
      "phaseCount": 14,
      "phaseCycles": 12384,
      "phaseMs": 0.008256,
      "workCycles": 1257600,
      "averageActiveTiles": 101.55038759689923,
      "usefulCycles": 26244.0,
      "estimatedWorkCycles": 1257600,
      "usefulUtilization": 0.020868320610687024,
      "workLaneOccupancy": 0.8914402173913043,
      "deviceUsefulUtilization": 0.0014396644337714864,
      "meanCycles": 7860.0,
      "p50Cycles": 7860,
      "p95Cycles": 7860,
      "maximumCycles": 7860,
      "firstOffset": 16532916,
      "lastOffset": 16545300
    }
  ],
  "samples": []
}
