{
  "clockHz": 1500000000,
  "tileCount": 1472,
  "sampleCount": 232260,
  "matchedSampleCount": 129660,
  "profileSpanCycles": 16746138,
  "profileSpanMs": 11.164092,
  "atOffset": null,
  "groups": [
    {
      "name": "repeat-remainder",
      "dimensions": {},
      "sampleCount": 1472,
      "tileCount": 1472,
      "phaseCount": 9,
      "phaseCycles": 15826398,
      "phaseMs": 10.550932,
      "workCycles": 23290826994,
      "averageActiveTiles": 1471.6442107673522,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 15822572.686141305,
      "p50Cycles": 15822600,
      "p95Cycles": 15822672,
      "maximumCycles": 15822696,
      "firstOffset": 714852,
      "lastOffset": 16541250
    },
    {
      "name": "reduce_sum_f16",
      "dimensions": {},
      "sampleCount": 22066,
      "tileCount": 1472,
      "phaseCount": 143,
      "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": 16741296
    },
    {
      "name": "fill_zero_u64",
      "dimensions": {},
      "sampleCount": 36142,
      "tileCount": 1472,
      "phaseCount": 182,
      "phaseCycles": 84882,
      "phaseMs": 0.056588,
      "workCycles": 12783918,
      "averageActiveTiles": 150.6081147946561,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 7209846,
      "usefulUtilization": 0.0,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": 0.0,
      "meanCycles": 353.71362957224284,
      "p50Cycles": 306,
      "p95Cycles": 744,
      "maximumCycles": 3150,
      "firstOffset": 3048,
      "lastOffset": 16733754
    },
    {
      "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": 622746,
      "lastOffset": 696276
    },
    {
      "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": 622722,
      "lastOffset": 694608
    },
    {
      "name": "attention_softmax_d72_p768_tail_f8_sm4",
      "dimensions": {},
      "sampleCount": 1472,
      "tileCount": 1472,
      "phaseCount": 15,
      "phaseCycles": 53034,
      "phaseMs": 0.035356,
      "workCycles": 71468532,
      "averageActiveTiles": 1347.5983708564318,
      "usefulCycles": 33499008.0,
      "estimatedWorkCycles": 71468532,
      "usefulUtilization": 0.46872388535978327,
      "workLaneOccupancy": 0.9972222222222222,
      "deviceUsefulUtilization": 0.4291111034594706,
      "meanCycles": 48551.99184782609,
      "p50Cycles": 48552,
      "p95Cycles": 48552,
      "maximumCycles": 48552,
      "firstOffset": 270666,
      "lastOffset": 323700
    },
    {
      "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": 483798,
      "lastOffset": 532998
    },
    {
      "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": 483522,
      "lastOffset": 531984
    },
    {
      "name": "cast_f16_f8",
      "dimensions": {},
      "sampleCount": 12514,
      "tileCount": 1472,
      "phaseCount": 130,
      "phaseCycles": 48348,
      "phaseMs": 0.032232,
      "workCycles": 39956538,
      "averageActiveTiles": 826.4362124596674,
      "usefulCycles": 1972820.0,
      "estimatedWorkCycles": 37827630,
      "usefulUtilization": 0.05215288401626007,
      "workLaneOccupancy": 0.9384120980109366,
      "deviceUsefulUtilization": 0.0277205050521401,
      "meanCycles": 3192.946939427841,
      "p50Cycles": 3048,
      "p95Cycles": 9300,
      "maximumCycles": 9300,
      "firstOffset": 0,
      "lastOffset": 16732908
    },
    {
      "name": "copy_u64",
      "dimensions": {},
      "sampleCount": 9311,
      "tileCount": 1472,
      "phaseCount": 104,
      "phaseCycles": 43818,
      "phaseMs": 0.029212,
      "workCycles": 12473154,
      "averageActiveTiles": 284.65822264822674,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 1339.61486413919,
      "p50Cycles": 834,
      "p95Cycles": 3018,
      "maximumCycles": 10578,
      "firstOffset": 94614,
      "lastOffset": 16735938
    },
    {
      "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": "copy_strided_u64",
      "dimensions": {},
      "sampleCount": 7057,
      "tileCount": 1472,
      "phaseCount": 88,
      "phaseCycles": 39498,
      "phaseMs": 0.026332,
      "workCycles": 9451128,
      "averageActiveTiles": 239.28117879386298,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 1339.2557744083888,
      "p50Cycles": 906,
      "p95Cycles": 3060,
      "maximumCycles": 3060,
      "firstOffset": 23628,
      "lastOffset": 16663194
    },
    {
      "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": 483804,
      "lastOffset": 523272
    },
    {
      "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": 483528,
      "lastOffset": 522378
    },
    {
      "name": "gemm_f8_init_large_rows_interleaved_k96_c576_r20_r24",
      "dimensions": {},
      "sampleCount": 1440,
      "tileCount": 1440,
      "phaseCount": 19,
      "phaseCycles": 30552,
      "phaseMs": 0.020368,
      "workCycles": 41489280,
      "averageActiveTiles": 1357.989002356638,
      "usefulCycles": 14929920.0,
      "estimatedWorkCycles": 41489280,
      "usefulUtilization": 0.35985006247396917,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.3319785511800266,
      "meanCycles": 28812.0,
      "p50Cycles": 28812,
      "p95Cycles": 28812,
      "maximumCycles": 28812,
      "firstOffset": 391068,
      "lastOffset": 421620
    },
    {
      "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": 16569252,
      "lastOffset": 16598868
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k96_c576_r20_r24",
      "dimensions": {},
      "sampleCount": 24,
      "tileCount": 24,
      "phaseCount": 2,
      "phaseCycles": 27816,
      "phaseMs": 0.018544,
      "workCycles": 637056,
      "averageActiveTiles": 22.902502157031925,
      "usefulCycles": 186624.0,
      "estimatedWorkCycles": 637056,
      "usefulUtilization": 0.29294755877034356,
      "workLaneOccupancy": 0.9,
      "deviceUsefulUtilization": 0.004557902239561841,
      "meanCycles": 26544.0,
      "p50Cycles": 26544,
      "p95Cycles": 26544,
      "maximumCycles": 26544,
      "firstOffset": 391086,
      "lastOffset": 418902
    },
    {
      "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": 16571694,
      "lastOffset": 16599342
    },
    {
      "name": "add_f16",
      "dimensions": {},
      "sampleCount": 11220,
      "tileCount": 1472,
      "phaseCount": 151,
      "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": 16746138
    },
    {
      "name": "gemm_f16_init_large_rows_interleaved_k80_c112_r121_r122",
      "dimensions": {},
      "sampleCount": 384,
      "tileCount": 384,
      "phaseCount": 4,
      "phaseCycles": 23310,
      "phaseMs": 0.01554,
      "workCycles": 8831232,
      "averageActiveTiles": 378.86023166023165,
      "usefulCycles": 5902848.0,
      "estimatedWorkCycles": 8831232,
      "usefulUtilization": 0.6684059483433342,
      "workLaneOccupancy": 0.9,
      "deviceUsefulUtilization": 0.17203290246768507,
      "meanCycles": 22998.0,
      "p50Cycles": 22998,
      "p95Cycles": 22998,
      "maximumCycles": 22998,
      "firstOffset": 240156,
      "lastOffset": 263466
    },
    {
      "name": "gemm_f16_init_small_rows_interleaved_k80_c112_r121_r122",
      "dimensions": {},
      "sampleCount": 384,
      "tileCount": 384,
      "phaseCount": 4,
      "phaseCycles": 23232,
      "phaseMs": 0.015488,
      "workCycles": 8815104,
      "averageActiveTiles": 379.43801652892563,
      "usefulCycles": 5854464.0,
      "estimatedWorkCycles": 8815104,
      "usefulUtilization": 0.6641400940930475,
      "workLaneOccupancy": 0.9,
      "deviceUsefulUtilization": 0.17119565217391305,
      "meanCycles": 22956.0,
      "p50Cycles": 22956,
      "p95Cycles": 22956,
      "maximumCycles": 22956,
      "firstOffset": 240174,
      "lastOffset": 263406
    },
    {
      "name": "layer_norm_f16",
      "dimensions": {},
      "sampleCount": 4374,
      "tileCount": 1458,
      "phaseCount": 36,
      "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": 16550982
    },
    {
      "name": "gemm_f16_init_large_rows_interleaved_k80_c96_r121_r122",
      "dimensions": {},
      "sampleCount": 288,
      "tileCount": 288,
      "phaseCount": 3,
      "phaseCycles": 20574,
      "phaseMs": 0.013716,
      "workCycles": 5693760,
      "averageActiveTiles": 276.745406824147,
      "usefulCycles": 3702456.0,
      "estimatedWorkCycles": 5693760,
      "usefulUtilization": 0.6502655538694992,
      "workLaneOccupancy": 0.878125,
      "deviceUsefulUtilization": 0.12225407965308684,
      "meanCycles": 19770.0,
      "p50Cycles": 19770,
      "p95Cycles": 19770,
      "maximumCycles": 19770,
      "firstOffset": 239586,
      "lastOffset": 260160
    },
    {
      "name": "gemm_f16_init_small_rows_interleaved_k80_c96_r121_r122",
      "dimensions": {},
      "sampleCount": 288,
      "tileCount": 288,
      "phaseCount": 5,
      "phaseCycles": 20016,
      "phaseMs": 0.013344,
      "workCycles": 5683392,
      "averageActiveTiles": 283.9424460431655,
      "usefulCycles": 3672108.0,
      "estimatedWorkCycles": 5683392,
      "usefulUtilization": 0.6461120401337793,
      "workLaneOccupancy": 0.878125,
      "deviceUsefulUtilization": 0.12463222356897091,
      "meanCycles": 19734.0,
      "p50Cycles": 19734,
      "p95Cycles": 19734,
      "maximumCycles": 19734,
      "firstOffset": 239568,
      "lastOffset": 259584
    },
    {
      "name": "bias_gelu_f16",
      "dimensions": {},
      "sampleCount": 1458,
      "tileCount": 1458,
      "phaseCount": 31,
      "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": 559452,
      "lastOffset": 576672
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k128_c80_r104_r105",
      "dimensions": {},
      "sampleCount": 1152,
      "tileCount": 1152,
      "phaseCount": 14,
      "phaseCycles": 14058,
      "phaseMs": 0.009372,
      "workCycles": 13727232,
      "averageActiveTiles": 976.4711907810499,
      "usefulCycles": 8188128.0,
      "estimatedWorkCycles": 13727232,
      "usefulUtilization": 0.596487915407855,
      "workLaneOccupancy": 0.854296875,
      "deviceUsefulUtilization": 0.3956883594054445,
      "meanCycles": 11916.0,
      "p50Cycles": 11916,
      "p95Cycles": 11916,
      "maximumCycles": 11916,
      "firstOffset": 331386,
      "lastOffset": 345444
    },
    {
      "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": 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": 332208,
      "lastOffset": 345786
    },
    {
      "name": "rearrange_row_major_to_amp_f16_o2_r64_p64_c32_p32_b64x16",
      "dimensions": {},
      "sampleCount": 704,
      "tileCount": 704,
      "phaseCount": 6,
      "phaseCycles": 13014,
      "phaseMs": 0.008676,
      "workCycles": 7569408,
      "averageActiveTiles": 581.6357768556938,
      "usefulCycles": 720896.0,
      "estimatedWorkCycles": 7569408,
      "usefulUtilization": 0.09523809523809523,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.0376317143410775,
      "meanCycles": 10752.0,
      "p50Cycles": 10752,
      "p95Cycles": 10752,
      "maximumCycles": 10752,
      "firstOffset": 210924,
      "lastOffset": 223938
    }
  ],
  "samples": []
}
