{
  "clockHz": 1500000000,
  "tileCount": 1472,
  "sampleCount": 52241,
  "matchedSampleCount": 52241,
  "profileSpanCycles": 170958,
  "profileSpanMs": 0.113972,
  "atOffset": null,
  "groups": [
    {
      "name": "exchange",
      "dimensions": {},
      "sampleCount": 12650,
      "tileCount": 1472,
      "phaseCount": 10,
      "phaseCycles": 110790,
      "phaseMs": 0.07386,
      "workCycles": 99254970,
      "averageActiveTiles": 895.8838342810723,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 7846.242687747035,
      "p50Cycles": 7194,
      "p95Cycles": 17562,
      "maximumCycles": 30336,
      "firstOffset": 0,
      "lastOffset": 162546
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k576_c16_r72_r72",
      "dimensions": {},
      "sampleCount": 4416,
      "tileCount": 1472,
      "phaseCount": 4,
      "phaseCycles": 32724,
      "phaseMs": 0.021816,
      "workCycles": 37226880,
      "averageActiveTiles": 1137.6017601760177,
      "usefulCycles": 22674816.0,
      "estimatedWorkCycles": 37226880,
      "usefulUtilization": 0.6090979421321367,
      "workLaneOccupancy": 0.9904891304347826,
      "deviceUsefulUtilization": 0.470727507533362,
      "meanCycles": 8430.0,
      "p50Cycles": 8430,
      "p95Cycles": 8430,
      "maximumCycles": 8430,
      "firstOffset": 330,
      "lastOffset": 39030
    },
    {
      "name": "fill_zero_u64",
      "dimensions": {},
      "sampleCount": 11536,
      "tileCount": 1472,
      "phaseCount": 39,
      "phaseCycles": 30210,
      "phaseMs": 0.02014,
      "workCycles": 4210656,
      "averageActiveTiles": 139.37954319761667,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 2139648,
      "usefulUtilization": 0.0,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": 0.0,
      "meanCycles": 365.00138696255203,
      "p50Cycles": 300,
      "p95Cycles": 600,
      "maximumCycles": 3276,
      "firstOffset": 46836,
      "lastOffset": 167472
    },
    {
      "name": "attention_softmax_d72_p768_tail_f16",
      "dimensions": {},
      "sampleCount": 1472,
      "tileCount": 1472,
      "phaseCount": 15,
      "phaseCycles": 28506,
      "phaseMs": 0.019004,
      "workCycles": 37145376,
      "averageActiveTiles": 1303.0721953272996,
      "usefulCycles": 18440784.0,
      "estimatedWorkCycles": 37145376,
      "usefulUtilization": 0.4964489792753747,
      "workLaneOccupancy": 0.9974763406940063,
      "deviceUsefulUtilization": 0.4394761285953529,
      "meanCycles": 25234.630434782608,
      "p50Cycles": 25236,
      "p95Cycles": 25236,
      "maximumCycles": 25236,
      "firstOffset": 96132,
      "lastOffset": 124638
    },
    {
      "name": "copy_u64",
      "dimensions": {},
      "sampleCount": 6977,
      "tileCount": 1472,
      "phaseCount": 47,
      "phaseCycles": 23196,
      "phaseMs": 0.015464,
      "workCycles": 5423694,
      "averageActiveTiles": 233.82022762545267,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 777.3676365199942,
      "p50Cycles": 822,
      "p95Cycles": 1230,
      "maximumCycles": 2010,
      "firstOffset": 23964,
      "lastOffset": 167982
    },
    {
      "name": "gemm_f16_init_small_rows_interleaved_k96_c80_r91_r92",
      "dimensions": {},
      "sampleCount": 896,
      "tileCount": 896,
      "phaseCount": 11,
      "phaseCycles": 20238,
      "phaseMs": 0.013492,
      "workCycles": 14477568,
      "averageActiveTiles": 715.3655499555292,
      "usefulCycles": 8358714.0,
      "estimatedWorkCycles": 14477568,
      "usefulUtilization": 0.5773562244708503,
      "workLaneOccupancy": 0.854296875,
      "deviceUsefulUtilization": 0.28058475070573224,
      "meanCycles": 16158.0,
      "p50Cycles": 16158,
      "p95Cycles": 16158,
      "maximumCycles": 16158,
      "firstOffset": 135336,
      "lastOffset": 155574
    },
    {
      "name": "gemm_f16_init_large_rows_interleaved_k96_c80_r91_r92",
      "dimensions": {},
      "sampleCount": 128,
      "tileCount": 128,
      "phaseCount": 8,
      "phaseCycles": 18804,
      "phaseMs": 0.012536,
      "workCycles": 2072832,
      "averageActiveTiles": 110.23356732610083,
      "usefulCycles": 1207224.0,
      "estimatedWorkCycles": 2072832,
      "usefulUtilization": 0.5824032048907003,
      "workLaneOccupancy": 0.854296875,
      "deviceUsefulUtilization": 0.04361439055520102,
      "meanCycles": 16194.0,
      "p50Cycles": 16194,
      "p95Cycles": 16194,
      "maximumCycles": 16194,
      "firstOffset": 136818,
      "lastOffset": 155622
    },
    {
      "name": "gemm_f16_init_small_rows_interleaved_k80_c80_r91_r92",
      "dimensions": {},
      "sampleCount": 224,
      "tileCount": 224,
      "phaseCount": 4,
      "phaseCycles": 14646,
      "phaseMs": 0.009764,
      "workCycles": 3033408,
      "averageActiveTiles": 207.1151167554281,
      "usefulCycles": 1834560.0,
      "estimatedWorkCycles": 3033408,
      "usefulUtilization": 0.6047851129818342,
      "workLaneOccupancy": 0.9,
      "deviceUsefulUtilization": 0.08509520332009333,
      "meanCycles": 13542.0,
      "p50Cycles": 13542,
      "p95Cycles": 13542,
      "maximumCycles": 13542,
      "firstOffset": 77154,
      "lastOffset": 91800
    },
    {
      "name": "gemm_f16_init_large_rows_interleaved_k80_c80_r91_r92",
      "dimensions": {},
      "sampleCount": 32,
      "tileCount": 32,
      "phaseCount": 3,
      "phaseCycles": 14346,
      "phaseMs": 0.009564,
      "workCycles": 434304,
      "averageActiveTiles": 30.273525721455457,
      "usefulCycles": 264960.0,
      "estimatedWorkCycles": 434304,
      "usefulUtilization": 0.610079575596817,
      "workLaneOccupancy": 0.9,
      "deviceUsefulUtilization": 0.012547051442910916,
      "meanCycles": 13572.0,
      "p50Cycles": 13572,
      "p95Cycles": 13572,
      "maximumCycles": 13572,
      "firstOffset": 77478,
      "lastOffset": 91824
    },
    {
      "name": "gemm_f16_init_small_rows_interleaved_k80_c64_r91_r92",
      "dimensions": {},
      "sampleCount": 1008,
      "tileCount": 1008,
      "phaseCount": 6,
      "phaseCycles": 12564,
      "phaseMs": 0.008376,
      "workCycles": 10983168,
      "averageActiveTiles": 874.1776504297994,
      "usefulCycles": 6524154.0,
      "estimatedWorkCycles": 10983168,
      "usefulUtilization": 0.5940138582966226,
      "workLaneOccupancy": 0.8890625,
      "deviceUsefulUtilization": 0.3527674177775009,
      "meanCycles": 10896.0,
      "p50Cycles": 10896,
      "p95Cycles": 10896,
      "maximumCycles": 10896,
      "firstOffset": 76626,
      "lastOffset": 89190
    },
    {
      "name": "gemm_f16_init_large_rows_interleaved_k80_c64_r91_r92",
      "dimensions": {},
      "sampleCount": 144,
      "tileCount": 144,
      "phaseCount": 5,
      "phaseCycles": 12510,
      "phaseMs": 0.00834,
      "workCycles": 1572480,
      "averageActiveTiles": 125.6978417266187,
      "usefulCycles": 942264.0,
      "estimatedWorkCycles": 1572480,
      "usefulUtilization": 0.5992216117216117,
      "workLaneOccupancy": 0.8890625,
      "deviceUsefulUtilization": 0.05116906474820144,
      "meanCycles": 10920.0,
      "p50Cycles": 10920,
      "p95Cycles": 10920,
      "maximumCycles": 10920,
      "firstOffset": 76638,
      "lastOffset": 89148
    },
    {
      "name": "reduce_sum_f16",
      "dimensions": {},
      "sampleCount": 3680,
      "tileCount": 1472,
      "phaseCount": 23,
      "phaseCycles": 12018,
      "phaseMs": 0.008012,
      "workCycles": 8246592,
      "averageActiveTiles": 686.1867199201198,
      "usefulCycles": 2268864.0,
      "estimatedWorkCycles": 8246592,
      "usefulUtilization": 0.2751274708388629,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.12825327226551478,
      "meanCycles": 2240.9217391304346,
      "p50Cycles": 2154,
      "p95Cycles": 2382,
      "maximumCycles": 2568,
      "firstOffset": 23964,
      "lastOffset": 165576
    },
    {
      "name": "unpack_amp_to_row_major_f16_o1_r9_p16_c72_p72",
      "dimensions": {},
      "sampleCount": 32,
      "tileCount": 32,
      "phaseCount": 4,
      "phaseCycles": 11760,
      "phaseMs": 0.00784,
      "workCycles": 249600,
      "averageActiveTiles": 21.224489795918366,
      "usefulCycles": 10368.0,
      "estimatedWorkCycles": 249600,
      "usefulUtilization": 0.04153846153846154,
      "workLaneOccupancy": 0.5625,
      "deviceUsefulUtilization": 0.0005989352262644189,
      "meanCycles": 7800.0,
      "p50Cycles": 7800,
      "p95Cycles": 7800,
      "maximumCycles": 7800,
      "firstOffset": 39900,
      "lastOffset": 51660
    },
    {
      "name": "copy_strided_u64",
      "dimensions": {},
      "sampleCount": 1632,
      "tileCount": 1472,
      "phaseCount": 26,
      "phaseCycles": 11184,
      "phaseMs": 0.007456,
      "workCycles": 835200,
      "averageActiveTiles": 74.67811158798283,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 511.7647058823529,
      "p50Cycles": 510,
      "p95Cycles": 618,
      "maximumCycles": 762,
      "firstOffset": 97230,
      "lastOffset": 168492
    },
    {
      "name": "unpack_amp_to_row_major_f16_o1_r16_p16_c72_p72",
      "dimensions": {},
      "sampleCount": 1440,
      "tileCount": 1440,
      "phaseCount": 4,
      "phaseCycles": 10962,
      "phaseMs": 0.007308,
      "workCycles": 9987840,
      "averageActiveTiles": 911.1330049261084,
      "usefulCycles": 829440.0,
      "estimatedWorkCycles": 9987840,
      "usefulUtilization": 0.08304498269896193,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.05140286999357464,
      "meanCycles": 6936.0,
      "p50Cycles": 6936,
      "p95Cycles": 6936,
      "maximumCycles": 6936,
      "firstOffset": 39894,
      "lastOffset": 50856
    },
    {
      "name": "rearrange_row_major_to_amp_f16_o0_r8_p8_c72_p80",
      "dimensions": {},
      "sampleCount": 1360,
      "tileCount": 1360,
      "phaseCount": 8,
      "phaseCycles": 9504,
      "phaseMs": 0.006336,
      "workCycles": 3215040,
      "averageActiveTiles": 338.2828282828283,
      "usefulCycles": 391680.0,
      "estimatedWorkCycles": 3215040,
      "usefulUtilization": 0.1218274111675127,
      "workLaneOccupancy": 0.9,
      "deviceUsefulUtilization": 0.027997364953886692,
      "meanCycles": 2364.0,
      "p50Cycles": 2364,
      "p95Cycles": 2364,
      "maximumCycles": 2364,
      "firstOffset": 55422,
      "lastOffset": 64926
    },
    {
      "name": "rearrange_row_major_to_amp_f16_o1_r64_p64_c16_p16",
      "dimensions": {},
      "sampleCount": 880,
      "tileCount": 880,
      "phaseCount": 8,
      "phaseCycles": 8148,
      "phaseMs": 0.005432,
      "workCycles": 1320000,
      "averageActiveTiles": 162.00294550810014,
      "usefulCycles": 450560.0,
      "estimatedWorkCycles": 1320000,
      "usefulUtilization": 0.3413333333333333,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.0375659004076754,
      "meanCycles": 1500.0,
      "p50Cycles": 1500,
      "p95Cycles": 1500,
      "maximumCycles": 1500,
      "firstOffset": 58278,
      "lastOffset": 66426
    },
    {
      "name": "attention_merge_v72_p80_k768_f16",
      "dimensions": {},
      "sampleCount": 1472,
      "tileCount": 1472,
      "phaseCount": 22,
      "phaseCycles": 7986,
      "phaseMs": 0.005324,
      "workCycles": 3629952,
      "averageActiveTiles": 454.53944402704735,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 2466.0,
      "p50Cycles": 2466,
      "p95Cycles": 2466,
      "maximumCycles": 2466,
      "firstOffset": 162972,
      "lastOffset": 170958
    },
    {
      "name": "rearrange_row_major_to_amp_f16_o1_r0_p64_c0_p16",
      "dimensions": {},
      "sampleCount": 80,
      "tileCount": 80,
      "phaseCount": 2,
      "phaseCycles": 7332,
      "phaseMs": 0.004888,
      "workCycles": 216960,
      "averageActiveTiles": 29.590834697217677,
      "usefulCycles": 16000.0,
      "estimatedWorkCycles": 216960,
      "usefulUtilization": 0.07374631268436578,
      "workLaneOccupancy": 0.390625,
      "deviceUsefulUtilization": 0.001482482981095377,
      "meanCycles": 2712.0,
      "p50Cycles": 2712,
      "p95Cycles": 2712,
      "maximumCycles": 2712,
      "firstOffset": 58122,
      "lastOffset": 67200
    },
    {
      "name": "rearrange_row_major_to_amp_f16_o0_r7_p7_c72_p80",
      "dimensions": {},
      "sampleCount": 112,
      "tileCount": 112,
      "phaseCount": 2,
      "phaseCycles": 6966,
      "phaseMs": 0.004644,
      "workCycles": 264768,
      "averageActiveTiles": 38.00861326442722,
      "usefulCycles": 28224.0,
      "estimatedWorkCycles": 264768,
      "usefulUtilization": 0.1065989847715736,
      "workLaneOccupancy": 0.9,
      "deviceUsefulUtilization": 0.0027524997191326816,
      "meanCycles": 2364.0,
      "p50Cycles": 2364,
      "p95Cycles": 2364,
      "maximumCycles": 2364,
      "firstOffset": 55398,
      "lastOffset": 64488
    }
  ],
  "samples": []
}
