{
  "clockHz": 1500000000,
  "tileCount": 1472,
  "sampleCount": 227656,
  "matchedSampleCount": 227656,
  "profileSpanCycles": 13391934,
  "profileSpanMs": 8.927956,
  "atOffset": null,
  "groups": [
    {
      "name": "repeat-remainder",
      "dimensions": {},
      "sampleCount": 1472,
      "tileCount": 1472,
      "phaseCount": 11,
      "phaseCycles": 12626910,
      "phaseMs": 8.41794,
      "workCycles": 18582392424,
      "averageActiveTiles": 1471.6500255406904,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 12623907.89673913,
      "p50Cycles": 12623910,
      "p95Cycles": 12623922,
      "maximumCycles": 12623946,
      "firstOffset": 543792,
      "lastOffset": 13170702
    },
    {
      "name": "sync",
      "dimensions": {},
      "sampleCount": 20565,
      "tileCount": 1457,
      "phaseCount": 57,
      "phaseCycles": 602286,
      "phaseMs": 0.401524,
      "workCycles": 106501914,
      "averageActiveTiles": 176.8294697203654,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 5178.794748358862,
      "p50Cycles": 2070,
      "p95Cycles": 19674,
      "maximumCycles": 46848,
      "firstOffset": 0,
      "lastOffset": 13391202
    },
    {
      "name": "exchange",
      "dimensions": {},
      "sampleCount": 78059,
      "tileCount": 1472,
      "phaseCount": 67,
      "phaseCycles": 562458,
      "phaseMs": 0.374972,
      "workCycles": 567086670,
      "averageActiveTiles": 1008.2293611256307,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 7264.846718507795,
      "p50Cycles": 4980,
      "p95Cycles": 19560,
      "maximumCycles": 31020,
      "firstOffset": 0,
      "lastOffset": 13391784
    },
    {
      "name": "copy_u64",
      "dimensions": {},
      "sampleCount": 28987,
      "tileCount": 1472,
      "phaseCount": 225,
      "phaseCycles": 122538,
      "phaseMs": 0.081692,
      "workCycles": 23403648,
      "averageActiveTiles": 190.99094158546737,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 807.3842757098009,
      "p50Cycles": 576,
      "p95Cycles": 1572,
      "maximumCycles": 10578,
      "firstOffset": 38112,
      "lastOffset": 13388022
    },
    {
      "name": "cast_f16_f8",
      "dimensions": {},
      "sampleCount": 11652,
      "tileCount": 1458,
      "phaseCount": 165,
      "phaseCycles": 72222,
      "phaseMs": 0.048148,
      "workCycles": 56424366,
      "averageActiveTiles": 781.2628561934036,
      "usefulCycles": 10457856.0,
      "estimatedWorkCycles": 56424366,
      "usefulUtilization": 0.18534290664426784,
      "workLaneOccupancy": 0.9751656629539366,
      "deviceUsefulUtilization": 0.09837060368212504,
      "meanCycles": 4842.461894953656,
      "p50Cycles": 4512,
      "p95Cycles": 9162,
      "maximumCycles": 9906,
      "firstOffset": 18312,
      "lastOffset": 13378374
    },
    {
      "name": "fill_zero_u64",
      "dimensions": {},
      "sampleCount": 17964,
      "tileCount": 1472,
      "phaseCount": 113,
      "phaseCycles": 64848,
      "phaseMs": 0.043232,
      "workCycles": 6292224,
      "averageActiveTiles": 97.03034789045152,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 3543024,
      "usefulUtilization": 0.0,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": 0.0,
      "meanCycles": 350.2685370741483,
      "p50Cycles": 306,
      "p95Cycles": 606,
      "maximumCycles": 3150,
      "firstOffset": 28932,
      "lastOffset": 13378992
    },
    {
      "name": "reduce_sum_f16",
      "dimensions": {},
      "sampleCount": 10440,
      "tileCount": 1472,
      "phaseCount": 148,
      "phaseCycles": 63738,
      "phaseMs": 0.042492,
      "workCycles": 42663456,
      "averageActiveTiles": 669.3566789042643,
      "usefulCycles": 23746848.0,
      "estimatedWorkCycles": 42663456,
      "usefulUtilization": 0.5566086347997686,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.25310442067867506,
      "meanCycles": 4086.537931034483,
      "p50Cycles": 3690,
      "p95Cycles": 7218,
      "maximumCycles": 7218,
      "firstOffset": 38376,
      "lastOffset": 13387752
    },
    {
      "name": "copy_strided_u64",
      "dimensions": {},
      "sampleCount": 4177,
      "tileCount": 1472,
      "phaseCount": 125,
      "phaseCycles": 54090,
      "phaseMs": 0.03606,
      "workCycles": 1816626,
      "averageActiveTiles": 33.58524681087077,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 434.91165908546805,
      "p50Cycles": 420,
      "p95Cycles": 720,
      "maximumCycles": 5430,
      "firstOffset": 0,
      "lastOffset": 13384656
    },
    {
      "name": "gemm_f8_init_large_rows_interleaved_k192_c160_r80_r82",
      "dimensions": {},
      "sampleCount": 780,
      "tileCount": 780,
      "phaseCount": 28,
      "phaseCycles": 34482,
      "phaseMs": 0.022988,
      "workCycles": 23797800,
      "averageActiveTiles": 690.1513833304333,
      "usefulCycles": 15350400.0,
      "estimatedWorkCycles": 23797800,
      "usefulUtilization": 0.6450344149459194,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.3024262185942004,
      "meanCycles": 30510.0,
      "p50Cycles": 30510,
      "p95Cycles": 30510,
      "maximumCycles": 30510,
      "firstOffset": 361242,
      "lastOffset": 395724
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k192_c160_r80_r82",
      "dimensions": {},
      "sampleCount": 624,
      "tileCount": 624,
      "phaseCount": 21,
      "phaseCycles": 33942,
      "phaseMs": 0.022628,
      "workCycles": 18720000,
      "averageActiveTiles": 551.5290790171468,
      "usefulCycles": 11943360.0,
      "estimatedWorkCycles": 18720000,
      "usefulUtilization": 0.638,
      "workLaneOccupancy": 0.996875,
      "deviceUsefulUtilization": 0.23904589158487752,
      "meanCycles": 30000.0,
      "p50Cycles": 30000,
      "p95Cycles": 30000,
      "maximumCycles": 30000,
      "firstOffset": 361290,
      "lastOffset": 395232
    },
    {
      "name": "gemm_f8_init_large_rows_interleaved_k160_c128_r120_r122",
      "dimensions": {},
      "sampleCount": 1215,
      "tileCount": 1215,
      "phaseCount": 32,
      "phaseCycles": 32730,
      "phaseMs": 0.02182,
      "workCycles": 32666490,
      "averageActiveTiles": 998.0595783684693,
      "usefulCycles": 23628960.0,
      "estimatedWorkCycles": 32666490,
      "usefulUtilization": 0.7233394221417728,
      "workLaneOccupancy": 0.9962962962962963,
      "deviceUsefulUtilization": 0.49044554258159645,
      "meanCycles": 26886.0,
      "p50Cycles": 26886,
      "p95Cycles": 26886,
      "maximumCycles": 26886,
      "firstOffset": 483894,
      "lastOffset": 516624
    },
    {
      "name": "attention_softmax_d72_p768_tail_f16",
      "dimensions": {},
      "sampleCount": 1488,
      "tileCount": 1472,
      "phaseCount": 28,
      "phaseCycles": 32700,
      "phaseMs": 0.0218,
      "workCycles": 34993632,
      "averageActiveTiles": 1070.1416513761467,
      "usefulCycles": 15838080.0,
      "estimatedWorkCycles": 34993632,
      "usefulUtilization": 0.4525989185689556,
      "workLaneOccupancy": 0.9970588235294118,
      "deviceUsefulUtilization": 0.32903869166334265,
      "meanCycles": 23517.225806451614,
      "p50Cycles": 23706,
      "p95Cycles": 23706,
      "maximumCycles": 23706,
      "firstOffset": 197076,
      "lastOffset": 13312092
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k160_c128_r120_r122",
      "dimensions": {},
      "sampleCount": 243,
      "tileCount": 243,
      "phaseCount": 19,
      "phaseCycles": 32556,
      "phaseMs": 0.021704,
      "workCycles": 6508512,
      "averageActiveTiles": 199.91743457427202,
      "usefulCycles": 4609584.0,
      "estimatedWorkCycles": 6508512,
      "usefulUtilization": 0.7082393026240099,
      "workLaneOccupancy": 0.9879938271604938,
      "deviceUsefulUtilization": 0.09618844051988013,
      "meanCycles": 26784.0,
      "p50Cycles": 26784,
      "p95Cycles": 26784,
      "maximumCycles": 26784,
      "firstOffset": 483960,
      "lastOffset": 516516
    },
    {
      "name": "layer_norm_f16",
      "dimensions": {},
      "sampleCount": 2187,
      "tileCount": 729,
      "phaseCount": 37,
      "phaseCycles": 30432,
      "phaseMs": 0.020288,
      "workCycles": 20518434,
      "averageActiveTiles": 674.238761829653,
      "usefulCycles": 7243344.0,
      "estimatedWorkCycles": 20518434,
      "usefulUtilization": 0.35301641441057346,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.1616965694006309,
      "meanCycles": 9382.0,
      "p50Cycles": 9384,
      "p95Cycles": 9384,
      "maximumCycles": 9384,
      "firstOffset": 57156,
      "lastOffset": 13185804
    },
    {
      "name": "gemm_f8_init_large_rows_interleaved_k192_c144_r80_r82",
      "dimensions": {},
      "sampleCount": 30,
      "tileCount": 30,
      "phaseCount": 11,
      "phaseCycles": 29304,
      "phaseMs": 0.019536,
      "workCycles": 824940,
      "averageActiveTiles": 28.151105651105652,
      "usefulCycles": 531360.0,
      "estimatedWorkCycles": 824940,
      "usefulUtilization": 0.6441195723325333,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.012318395470569383,
      "meanCycles": 27498.0,
      "p50Cycles": 27498,
      "p95Cycles": 27498,
      "maximumCycles": 27498,
      "firstOffset": 362574,
      "lastOffset": 391878
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k192_c144_r80_r82",
      "dimensions": {},
      "sampleCount": 24,
      "tileCount": 24,
      "phaseCount": 13,
      "phaseCycles": 28770,
      "phaseMs": 0.01918,
      "workCycles": 649008,
      "averageActiveTiles": 22.5584984358707,
      "usefulCycles": 413424.0,
      "estimatedWorkCycles": 649008,
      "usefulUtilization": 0.637009096960284,
      "workLaneOccupancy": 0.996875,
      "deviceUsefulUtilization": 0.009762207009112754,
      "meanCycles": 27042.0,
      "p50Cycles": 27042,
      "p95Cycles": 27042,
      "maximumCycles": 27042,
      "firstOffset": 362652,
      "lastOffset": 391422
    },
    {
      "name": "gemm_f8_init_large_rows_interleaved_k128_c64_r242_r244",
      "dimensions": {},
      "sampleCount": 972,
      "tileCount": 972,
      "phaseCount": 4,
      "phaseCycles": 22110,
      "phaseMs": 0.01474,
      "workCycles": 18294984,
      "averageActiveTiles": 827.4529172320217,
      "usefulCycles": 15178752.0,
      "estimatedWorkCycles": 18294984,
      "usefulUtilization": 0.8296674104771012,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.4663795646274556,
      "meanCycles": 18822.0,
      "p50Cycles": 18822,
      "p95Cycles": 18822,
      "maximumCycles": 18822,
      "firstOffset": 88854,
      "lastOffset": 110964
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k128_c64_r242_r244",
      "dimensions": {},
      "sampleCount": 486,
      "tileCount": 486,
      "phaseCount": 1,
      "phaseCycles": 21996,
      "phaseMs": 0.014664,
      "workCycles": 9083340,
      "averageActiveTiles": 412.9541734860884,
      "usefulCycles": 7496064.0,
      "estimatedWorkCycles": 9083340,
      "usefulUtilization": 0.8252541466024612,
      "workLaneOccupancy": 0.9958677685950413,
      "deviceUsefulUtilization": 0.23151640219170283,
      "meanCycles": 18690.0,
      "p50Cycles": 18690,
      "p95Cycles": 18690,
      "maximumCycles": 18690,
      "firstOffset": 88848,
      "lastOffset": 110844
    },
    {
      "name": "add_f16",
      "dimensions": {},
      "sampleCount": 15364,
      "tileCount": 1472,
      "phaseCount": 206,
      "phaseCycles": 20850,
      "phaseMs": 0.0139,
      "workCycles": 20358648,
      "averageActiveTiles": 976.4339568345324,
      "usefulCycles": 3096104.0,
      "estimatedWorkCycles": 20358648,
      "usefulUtilization": 0.15207807512561738,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.10087920967573767,
      "meanCycles": 1325.0877375683415,
      "p50Cycles": 1074,
      "p95Cycles": 2658,
      "maximumCycles": 2658,
      "firstOffset": 44688,
      "lastOffset": 13390932
    },
    {
      "name": "gemm_f16_init_small_rows_interleaved_k96_c80_r91_r92",
      "dimensions": {},
      "sampleCount": 896,
      "tileCount": 896,
      "phaseCount": 16,
      "phaseCycles": 20232,
      "phaseMs": 0.013488,
      "workCycles": 14859264,
      "averageActiveTiles": 734.4436536180308,
      "usefulCycles": 8358714.0,
      "estimatedWorkCycles": 14859264,
      "usefulUtilization": 0.5625254386758322,
      "workLaneOccupancy": 0.854296875,
      "deviceUsefulUtilization": 0.28066796089277424,
      "meanCycles": 16584.0,
      "p50Cycles": 16584,
      "p95Cycles": 16584,
      "maximumCycles": 16584,
      "firstOffset": 234738,
      "lastOffset": 254970
    }
  ],
  "samples": []
}
