{
  "clockHz": 1500000000,
  "tileCount": 1472,
  "sampleCount": 248751,
  "matchedSampleCount": 248751,
  "profileSpanCycles": 844380,
  "profileSpanMs": 0.56292,
  "atOffset": null,
  "groups": [
    {
      "name": "sync",
      "dimensions": {},
      "sampleCount": 12545,
      "tileCount": 1464,
      "phaseCount": 53,
      "phaseCycles": 690864,
      "phaseMs": 0.460576,
      "workCycles": 87941028,
      "averageActiveTiles": 127.29137427916348,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 7010.046074133121,
      "p50Cycles": 6222,
      "p95Cycles": 19266,
      "maximumCycles": 63960,
      "firstOffset": 2844,
      "lastOffset": 842478
    },
    {
      "name": "exchange",
      "dimensions": {},
      "sampleCount": 80191,
      "tileCount": 1472,
      "phaseCount": 63,
      "phaseCycles": 523086,
      "phaseMs": 0.348724,
      "workCycles": 500377302,
      "averageActiveTiles": 956.587065989149,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 6239.818707835044,
      "p50Cycles": 4200,
      "p95Cycles": 18372,
      "maximumCycles": 32058,
      "firstOffset": 2850,
      "lastOffset": 843174
    },
    {
      "name": "cast_f16_f8",
      "dimensions": {},
      "sampleCount": 17977,
      "tileCount": 1472,
      "phaseCount": 182,
      "phaseCycles": 89508,
      "phaseMs": 0.059672,
      "workCycles": 76872798,
      "averageActiveTiles": 858.8371765652232,
      "usefulCycles": 12838402.0,
      "estimatedWorkCycles": 72720366,
      "usefulUtilization": 0.17654479351767838,
      "workLaneOccupancy": 0.9789593617244517,
      "deviceUsefulUtilization": 0.09744090460216333,
      "meanCycles": 4276.175001390666,
      "p50Cycles": 4578,
      "p95Cycles": 8238,
      "maximumCycles": 9102,
      "firstOffset": 0,
      "lastOffset": 836484
    },
    {
      "name": "copy_u64",
      "dimensions": {},
      "sampleCount": 34289,
      "tileCount": 1472,
      "phaseCount": 233,
      "phaseCycles": 81936,
      "phaseMs": 0.054624,
      "workCycles": 21030318,
      "averageActiveTiles": 256.667618629174,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 613.3254979731109,
      "p50Cycles": 534,
      "p95Cycles": 1326,
      "maximumCycles": 2010,
      "firstOffset": 9234,
      "lastOffset": 842220
    },
    {
      "name": "gemm_f8_init_large_rows_interleaved_k192_c96_r36_r38",
      "dimensions": {},
      "sampleCount": 2160,
      "tileCount": 360,
      "phaseCount": 87,
      "phaseCycles": 70194,
      "phaseMs": 0.046796,
      "workCycles": 25064640,
      "averageActiveTiles": 357.07667321993335,
      "usefulCycles": 11819520.0,
      "estimatedWorkCycles": 25064640,
      "usefulUtilization": 0.47156153050672184,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.1143910479156524,
      "meanCycles": 11604.0,
      "p50Cycles": 11604,
      "p95Cycles": 11604,
      "maximumCycles": 11604,
      "firstOffset": 67104,
      "lastOffset": 731292
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k192_c96_r36_r38",
      "dimensions": {},
      "sampleCount": 6480,
      "tileCount": 1080,
      "phaseCount": 88,
      "phaseCycles": 69324,
      "phaseMs": 0.046216,
      "workCycles": 74416320,
      "averageActiveTiles": 1073.456811493855,
      "usefulCycles": 33530112.0,
      "estimatedWorkCycles": 74416320,
      "usefulUtilization": 0.45057471264367815,
      "workLaneOccupancy": 0.9981481481481481,
      "deviceUsefulUtilization": 0.32858185759119746,
      "meanCycles": 11484.0,
      "p50Cycles": 11484,
      "p95Cycles": 11484,
      "maximumCycles": 11484,
      "firstOffset": 67170,
      "lastOffset": 731214
    },
    {
      "name": "layer_norm_f16",
      "dimensions": {},
      "sampleCount": 2187,
      "tileCount": 729,
      "phaseCount": 61,
      "phaseCycles": 50118,
      "phaseMs": 0.033412,
      "workCycles": 33985980,
      "averageActiveTiles": 678.1192385969113,
      "usefulCycles": 10077696.0,
      "estimatedWorkCycles": 33985980,
      "usefulUtilization": 0.2965250965250965,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.13660283470140902,
      "meanCycles": 15540.0,
      "p50Cycles": 15540,
      "p95Cycles": 15540,
      "maximumCycles": 15540,
      "firstOffset": 25596,
      "lastOffset": 640386
    },
    {
      "name": "add_f16",
      "dimensions": {},
      "sampleCount": 16466,
      "tileCount": 1458,
      "phaseCount": 197,
      "phaseCycles": 46914,
      "phaseMs": 0.031276,
      "workCycles": 47384418,
      "averageActiveTiles": 1010.027241335209,
      "usefulCycles": 2311700.0,
      "estimatedWorkCycles": 47384418,
      "usefulUtilization": 0.048786079845910525,
      "workLaneOccupancy": 0.999161494441659,
      "deviceUsefulUtilization": 0.03347504731136158,
      "meanCycles": 2877.712741406535,
      "p50Cycles": 2940,
      "p95Cycles": 2940,
      "maximumCycles": 5532,
      "firstOffset": 17022,
      "lastOffset": 844380
    },
    {
      "name": "rearrange_row_major_to_amp_f16_o0_r38_p38_c192_p192",
      "dimensions": {},
      "sampleCount": 2160,
      "tileCount": 360,
      "phaseCount": 87,
      "phaseCycles": 45912,
      "phaseMs": 0.030608,
      "workCycles": 13932000,
      "averageActiveTiles": 303.45007841087295,
      "usefulCycles": 7879680.0,
      "estimatedWorkCycles": 13932000,
      "usefulUtilization": 0.5655813953488372,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.11659355894452146,
      "meanCycles": 6450.0,
      "p50Cycles": 6450,
      "p95Cycles": 6450,
      "maximumCycles": 6450,
      "firstOffset": 49290,
      "lastOffset": 709662
    },
    {
      "name": "fill_zero_u64",
      "dimensions": {},
      "sampleCount": 15270,
      "tileCount": 1472,
      "phaseCount": 144,
      "phaseCycles": 45654,
      "phaseMs": 0.030436,
      "workCycles": 5563332,
      "averageActiveTiles": 121.85858851360231,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 2905092,
      "usefulUtilization": 0.0,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": 0.0,
      "meanCycles": 364.33084479371314,
      "p50Cycles": 300,
      "p95Cycles": 600,
      "maximumCycles": 1500,
      "firstOffset": 189330,
      "lastOffset": 833868
    },
    {
      "name": "gemm_f8_init_large_rows_interleaved_k384_c192_r36_r38",
      "dimensions": {},
      "sampleCount": 360,
      "tileCount": 360,
      "phaseCount": 29,
      "phaseCycles": 45180,
      "phaseMs": 0.03012,
      "workCycles": 16249680,
      "averageActiveTiles": 359.66533864541833,
      "usefulCycles": 7359840.0,
      "estimatedWorkCycles": 16249680,
      "usefulUtilization": 0.4529221498515663,
      "workLaneOccupancy": 0.9340277777777778,
      "deviceUsefulUtilization": 0.11066603152606963,
      "meanCycles": 45138.0,
      "p50Cycles": 45138,
      "p95Cycles": 45138,
      "maximumCycles": 45138,
      "firstOffset": 559716,
      "lastOffset": 604896
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k384_c192_r36_r38",
      "dimensions": {},
      "sampleCount": 1080,
      "tileCount": 1080,
      "phaseCount": 31,
      "phaseCycles": 44844,
      "phaseMs": 0.029896,
      "workCycles": 48366720,
      "averageActiveTiles": 1078.5549906341985,
      "usefulCycles": 20878704.0,
      "estimatedWorkCycles": 48366720,
      "usefulUtilization": 0.4316750029772538,
      "workLaneOccupancy": 0.9322980967078189,
      "deviceUsefulUtilization": 0.31629431303882444,
      "meanCycles": 44784.0,
      "p50Cycles": 44784,
      "p95Cycles": 44784,
      "maximumCycles": 44784,
      "firstOffset": 559704,
      "lastOffset": 604548
    },
    {
      "name": "reduce_sum_f16",
      "dimensions": {},
      "sampleCount": 14954,
      "tileCount": 1472,
      "phaseCount": 200,
      "phaseCycles": 44784,
      "phaseMs": 0.029856,
      "workCycles": 34382604,
      "averageActiveTiles": 767.7430332261522,
      "usefulCycles": 14862088.0,
      "estimatedWorkCycles": 34382604,
      "usefulUtilization": 0.4322560327309706,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.22544942778476784,
      "meanCycles": 2299.224555302929,
      "p50Cycles": 1794,
      "p95Cycles": 7086,
      "maximumCycles": 7212,
      "firstOffset": 9768,
      "lastOffset": 841026
    },
    {
      "name": "rearrange_row_major_to_amp_f16_o0_r36_p36_c192_p192",
      "dimensions": {},
      "sampleCount": 6048,
      "tileCount": 1008,
      "phaseCount": 81,
      "phaseCycles": 41424,
      "phaseMs": 0.027616,
      "workCycles": 33638976,
      "averageActiveTiles": 812.0648899188876,
      "usefulCycles": 20901888.0,
      "estimatedWorkCycles": 33638976,
      "usefulUtilization": 0.6213592233009708,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.34278804977580735,
      "meanCycles": 5562.0,
      "p50Cycles": 5562,
      "p95Cycles": 5562,
      "maximumCycles": 5562,
      "firstOffset": 49002,
      "lastOffset": 708492
    },
    {
      "name": "rearrange_row_major_to_amp_f16_o0_r35_p36_c192_p192",
      "dimensions": {},
      "sampleCount": 432,
      "tileCount": 72,
      "phaseCount": 43,
      "phaseCycles": 39246,
      "phaseMs": 0.026164,
      "workCycles": 2680128,
      "averageActiveTiles": 68.29047546246751,
      "usefulCycles": 1451520.0,
      "estimatedWorkCycles": 2680128,
      "usefulUtilization": 0.5415860735009671,
      "workLaneOccupancy": 0.9722222222222222,
      "deviceUsefulUtilization": 0.02512579515165212,
      "meanCycles": 6204.0,
      "p50Cycles": 6204,
      "p95Cycles": 6204,
      "maximumCycles": 6204,
      "firstOffset": 48978,
      "lastOffset": 708918
    },
    {
      "name": "attention_softmax_d72_p768_tail_f16",
      "dimensions": {},
      "sampleCount": 1488,
      "tileCount": 1472,
      "phaseCount": 36,
      "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": 236334,
      "lastOffset": 786672
    },
    {
      "name": "gemm_f8_init_large_rows_interleaved_k192_c160_r80_r82",
      "dimensions": {},
      "sampleCount": 780,
      "tileCount": 780,
      "phaseCount": 30,
      "phaseCycles": 34032,
      "phaseMs": 0.022688,
      "workCycles": 23180040,
      "averageActiveTiles": 681.1248236953455,
      "usefulCycles": 15350400.0,
      "estimatedWorkCycles": 23180040,
      "usefulUtilization": 0.6622249141934181,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.30642515484147914,
      "meanCycles": 29718.0,
      "p50Cycles": 29718,
      "p95Cycles": 29718,
      "maximumCycles": 29718,
      "firstOffset": 420726,
      "lastOffset": 454758
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k192_c160_r80_r82",
      "dimensions": {},
      "sampleCount": 624,
      "tileCount": 624,
      "phaseCount": 20,
      "phaseCycles": 33504,
      "phaseMs": 0.022336,
      "workCycles": 18225792,
      "averageActiveTiles": 543.9885386819484,
      "usefulCycles": 11943360.0,
      "estimatedWorkCycles": 18225792,
      "usefulUtilization": 0.6552999178307313,
      "workLaneOccupancy": 0.996875,
      "deviceUsefulUtilization": 0.24217095427930735,
      "meanCycles": 29208.0,
      "p50Cycles": 29208,
      "p95Cycles": 29208,
      "maximumCycles": 29208,
      "firstOffset": 420726,
      "lastOffset": 454230
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k192_c144_r80_r82",
      "dimensions": {},
      "sampleCount": 24,
      "tileCount": 24,
      "phaseCount": 13,
      "phaseCycles": 31272,
      "phaseMs": 0.020848,
      "workCycles": 631728,
      "averageActiveTiles": 20.20107444359171,
      "usefulCycles": 413424.0,
      "estimatedWorkCycles": 631728,
      "usefulUtilization": 0.6544335536813312,
      "workLaneOccupancy": 0.996875,
      "deviceUsefulUtilization": 0.008981155527378291,
      "meanCycles": 26322.0,
      "p50Cycles": 26322,
      "p95Cycles": 26322,
      "maximumCycles": 26322,
      "firstOffset": 419982,
      "lastOffset": 451254
    },
    {
      "name": "gemm_f8_init_large_rows_interleaved_k192_c144_r80_r82",
      "dimensions": {},
      "sampleCount": 30,
      "tileCount": 30,
      "phaseCount": 17,
      "phaseCycles": 31044,
      "phaseMs": 0.020696,
      "workCycles": 803340,
      "averageActiveTiles": 25.877464244298416,
      "usefulCycles": 531360.0,
      "estimatedWorkCycles": 803340,
      "usefulUtilization": 0.6614384942863545,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.01162795583267508,
      "meanCycles": 26778.0,
      "p50Cycles": 26778,
      "p95Cycles": 26778,
      "maximumCycles": 26778,
      "firstOffset": 419994,
      "lastOffset": 451038
    }
  ],
  "samples": []
}
