{
  "clockHz": 1500000000,
  "tileCount": 1472,
  "sampleCount": 235076,
  "matchedSampleCount": 132476,
  "profileSpanCycles": 16722798,
  "profileSpanMs": 11.148532,
  "atOffset": null,
  "groups": [
    {
      "name": "repeat-remainder",
      "dimensions": {},
      "sampleCount": 1472,
      "tileCount": 1472,
      "phaseCount": 9,
      "phaseCycles": 15803778,
      "phaseMs": 10.535852,
      "workCycles": 23257530354,
      "averageActiveTiles": 1471.6437015250403,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 15799952.686141305,
      "p50Cycles": 15799980,
      "p95Cycles": 15800052,
      "maximumCycles": 15800076,
      "firstOffset": 714132,
      "lastOffset": 16517910
    },
    {
      "name": "reduce_sum_f16",
      "dimensions": {},
      "sampleCount": 22066,
      "tileCount": 1472,
      "phaseCount": 140,
      "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": 16717956
    },
    {
      "name": "fill_zero_u64",
      "dimensions": {},
      "sampleCount": 36142,
      "tileCount": 1472,
      "phaseCount": 183,
      "phaseCycles": 86634,
      "phaseMs": 0.057756,
      "workCycles": 13317870,
      "averageActiveTiles": 153.72567352309716,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 7670646,
      "usefulUtilization": 0.0,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": 0.0,
      "meanCycles": 368.4873554313541,
      "p50Cycles": 306,
      "p95Cycles": 744,
      "maximumCycles": 3150,
      "firstOffset": 3042,
      "lastOffset": 16710414
    },
    {
      "name": "gemm_f8_init_large_rows_interleaved_k384_c192_r72_r76",
      "dimensions": {},
      "sampleCount": 360,
      "tileCount": 360,
      "phaseCount": 14,
      "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": 622026,
      "lastOffset": 695556
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k384_c192_r72_r76",
      "dimensions": {},
      "sampleCount": 1080,
      "tileCount": 1080,
      "phaseCount": 35,
      "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": 622002,
      "lastOffset": 693888
    },
    {
      "name": "cast_f16_f8",
      "dimensions": {},
      "sampleCount": 15330,
      "tileCount": 1472,
      "phaseCount": 144,
      "phaseCycles": 64962,
      "phaseMs": 0.043308,
      "workCycles": 52243536,
      "averageActiveTiles": 804.2168652442966,
      "usefulCycles": 3440132.0,
      "estimatedWorkCycles": 50066832,
      "usefulUtilization": 0.06871079839842872,
      "workLaneOccupancy": 0.8440914881547101,
      "deviceUsefulUtilization": 0.03597558873883461,
      "meanCycles": 3407.927984344423,
      "p50Cycles": 3042,
      "p95Cycles": 9300,
      "maximumCycles": 9300,
      "firstOffset": 0,
      "lastOffset": 16709568
    },
    {
      "name": "attention_softmax_d72_p768_tail_f8_sm4",
      "dimensions": {},
      "sampleCount": 1472,
      "tileCount": 1472,
      "phaseCount": 15,
      "phaseCycles": 52668,
      "phaseMs": 0.035112,
      "workCycles": 71468520,
      "averageActiveTiles": 1356.9628616997038,
      "usefulCycles": 33499008.0,
      "estimatedWorkCycles": 71468520,
      "usefulUtilization": 0.4687239640613797,
      "workLaneOccupancy": 0.9972222222222222,
      "deviceUsefulUtilization": 0.4320930785461678,
      "meanCycles": 48551.983695652176,
      "p50Cycles": 48552,
      "p95Cycles": 48552,
      "maximumCycles": 48552,
      "firstOffset": 270714,
      "lastOffset": 323382
    },
    {
      "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": 483078,
      "lastOffset": 532278
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k384_c80_r160_r164",
      "dimensions": {},
      "sampleCount": 636,
      "tileCount": 636,
      "phaseCount": 15,
      "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": 482802,
      "lastOffset": 531264
    },
    {
      "name": "copy_u64",
      "dimensions": {},
      "sampleCount": 9311,
      "tileCount": 1472,
      "phaseCount": 104,
      "phaseCycles": 43824,
      "phaseMs": 0.029216,
      "workCycles": 12473154,
      "averageActiveTiles": 284.61924972617743,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 1339.61486413919,
      "p50Cycles": 834,
      "p95Cycles": 3018,
      "maximumCycles": 10578,
      "firstOffset": 94614,
      "lastOffset": 16712598
    },
    {
      "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": 10,
      "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": 483084,
      "lastOffset": 522552
    },
    {
      "name": "copy_strided_u64",
      "dimensions": {},
      "sampleCount": 7057,
      "tileCount": 1472,
      "phaseCount": 85,
      "phaseCycles": 39432,
      "phaseMs": 0.026288,
      "workCycles": 9451128,
      "averageActiveTiles": 239.68167985392574,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 1339.2557744083888,
      "p50Cycles": 906,
      "p95Cycles": 3060,
      "maximumCycles": 3060,
      "firstOffset": 23628,
      "lastOffset": 16639854
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k384_c64_r160_r164",
      "dimensions": {},
      "sampleCount": 12,
      "tileCount": 12,
      "phaseCount": 11,
      "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": 482808,
      "lastOffset": 521658
    },
    {
      "name": "gemm_f8_init_large_rows_interleaved_k96_c576_r20_r24",
      "dimensions": {},
      "sampleCount": 1440,
      "tileCount": 1440,
      "phaseCount": 19,
      "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": 390960,
      "lastOffset": 420900
    },
    {
      "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": 16545912,
      "lastOffset": 16575528
    },
    {
      "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": 16548354,
      "lastOffset": 16576002
    },
    {
      "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": 390630,
      "lastOffset": 418218
    },
    {
      "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": 16722798
    },
    {
      "name": "layer_norm_f16",
      "dimensions": {},
      "sampleCount": 4374,
      "tileCount": 1458,
      "phaseCount": 37,
      "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": 16527642
    },
    {
      "name": "bias_gelu_f16",
      "dimensions": {},
      "sampleCount": 1458,
      "tileCount": 1458,
      "phaseCount": 30,
      "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": 558732,
      "lastOffset": 575952
    },
    {
      "name": "gemm_f8_init_large_rows_interleaved_k96_c112_r121_r122",
      "dimensions": {},
      "sampleCount": 384,
      "tileCount": 384,
      "phaseCount": 4,
      "phaseCycles": 14724,
      "phaseMs": 0.009816,
      "workCycles": 5398272,
      "averageActiveTiles": 366.63080684596576,
      "usefulCycles": 2951424.0,
      "estimatedWorkCycles": 5398272,
      "usefulUtilization": 0.5467349551856594,
      "workLaneOccupancy": 0.75,
      "deviceUsefulUtilization": 0.13617518868927395,
      "meanCycles": 14058.0,
      "p50Cycles": 14058,
      "p95Cycles": 14058,
      "maximumCycles": 14058,
      "firstOffset": 247806,
      "lastOffset": 262530
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k96_c112_r121_r122",
      "dimensions": {},
      "sampleCount": 384,
      "tileCount": 384,
      "phaseCount": 4,
      "phaseCycles": 14694,
      "phaseMs": 0.009796,
      "workCycles": 5384448,
      "averageActiveTiles": 366.4385463454471,
      "usefulCycles": 2927232.0,
      "estimatedWorkCycles": 5384448,
      "usefulUtilization": 0.5436456996148908,
      "workLaneOccupancy": 0.75,
      "deviceUsefulUtilization": 0.1353347417757026,
      "meanCycles": 14022.0,
      "p50Cycles": 14022,
      "p95Cycles": 14022,
      "maximumCycles": 14022,
      "firstOffset": 247842,
      "lastOffset": 262536
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k128_c80_r104_r105",
      "dimensions": {},
      "sampleCount": 1152,
      "tileCount": 1152,
      "phaseCount": 14,
      "phaseCycles": 14142,
      "phaseMs": 0.009428,
      "workCycles": 13727232,
      "averageActiveTiles": 970.6711921934663,
      "usefulCycles": 8188128.0,
      "estimatedWorkCycles": 13727232,
      "usefulUtilization": 0.596487915407855,
      "workLaneOccupancy": 0.854296875,
      "deviceUsefulUtilization": 0.39333806791979486,
      "meanCycles": 11916.0,
      "p50Cycles": 11916,
      "p95Cycles": 11916,
      "maximumCycles": 11916,
      "firstOffset": 330978,
      "lastOffset": 345120
    },
    {
      "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": "rearrange_row_major_to_amp_f16_o2_r64_p64_c32_p32_b64x16",
      "dimensions": {},
      "sampleCount": 704,
      "tileCount": 704,
      "phaseCount": 6,
      "phaseCycles": 13806,
      "phaseMs": 0.009204,
      "workCycles": 7569408,
      "averageActiveTiles": 548.2694480660582,
      "usefulCycles": 720896.0,
      "estimatedWorkCycles": 7569408,
      "usefulUtilization": 0.09523809523809523,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.035472919776530686,
      "meanCycles": 10752.0,
      "p50Cycles": 10752,
      "p95Cycles": 10752,
      "maximumCycles": 10752,
      "firstOffset": 210444,
      "lastOffset": 224250
    },
    {
      "name": "gemm_f8_init_large_rows_interleaved_k128_c80_r104_r105",
      "dimensions": {},
      "sampleCount": 192,
      "tileCount": 192,
      "phaseCount": 6,
      "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": 331884,
      "lastOffset": 345462
    },
    {
      "name": "gemm_f8_init_large_rows_interleaved_k96_c96_r121_r122",
      "dimensions": {},
      "sampleCount": 288,
      "tileCount": 288,
      "phaseCount": 3,
      "phaseCycles": 13182,
      "phaseMs": 0.008788,
      "workCycles": 3487104,
      "averageActiveTiles": 264.5352753755121,
      "usefulCycles": 1851228.0,
      "estimatedWorkCycles": 3487104,
      "usefulUtilization": 0.5308783448959365,
      "workLaneOccupancy": 0.7317708333333334,
      "deviceUsefulUtilization": 0.0954049246996893,
      "meanCycles": 12108.0,
      "p50Cycles": 12108,
      "p95Cycles": 12108,
      "maximumCycles": 12108,
      "firstOffset": 245934,
      "lastOffset": 259116
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k96_c96_r121_r122",
      "dimensions": {},
      "sampleCount": 288,
      "tileCount": 288,
      "phaseCount": 5,
      "phaseCycles": 12642,
      "phaseMs": 0.008428,
      "workCycles": 3478464,
      "averageActiveTiles": 275.1514000949217,
      "usefulCycles": 1836054.0,
      "estimatedWorkCycles": 3478464,
      "usefulUtilization": 0.5278346994535519,
      "workLaneOccupancy": 0.7317708333333334,
      "deviceUsefulUtilization": 0.0986647123460102,
      "meanCycles": 12078.0,
      "p50Cycles": 12078,
      "p95Cycles": 12078,
      "maximumCycles": 12078,
      "firstOffset": 245934,
      "lastOffset": 258576
    }
  ],
  "samples": []
}
