{
  "clockHz": 1500000000,
  "tileCount": 1472,
  "sampleCount": 48334,
  "matchedSampleCount": 48334,
  "profileSpanCycles": 195834,
  "profileSpanMs": 0.130556,
  "atOffset": null,
  "groups": [
    {
      "name": "exchange",
      "dimensions": {},
      "sampleCount": 11638,
      "tileCount": 1472,
      "phaseCount": 9,
      "phaseCycles": 121800,
      "phaseMs": 0.0812,
      "workCycles": 104592336,
      "averageActiveTiles": 858.7219704433497,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 8987.140058429284,
      "p50Cycles": 7392,
      "p95Cycles": 25098,
      "maximumCycles": 30540,
      "firstOffset": 0,
      "lastOffset": 187944
    },
    {
      "name": "sync",
      "dimensions": {},
      "sampleCount": 1610,
      "tileCount": 506,
      "phaseCount": 4,
      "phaseCycles": 56628,
      "phaseMs": 0.037752,
      "workCycles": 8306112,
      "averageActiveTiles": 146.6785335876245,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 5159.075776397515,
      "p50Cycles": 3540,
      "p95Cycles": 18780,
      "maximumCycles": 18780,
      "firstOffset": 0,
      "lastOffset": 56628
    },
    {
      "name": "gemm_f16_init_large_rows_interleaved_k384_c16_r114_r116",
      "dimensions": {},
      "sampleCount": 2484,
      "tileCount": 828,
      "phaseCount": 3,
      "phaseCycles": 55476,
      "phaseMs": 0.036984,
      "workCycles": 37856160,
      "averageActiveTiles": 682.3880597014926,
      "usefulCycles": 27398736.0,
      "estimatedWorkCycles": 37856160,
      "usefulUtilization": 0.7237589866484081,
      "workLaneOccupancy": 0.9904891304347826,
      "deviceUsefulUtilization": 0.33551935502073754,
      "meanCycles": 15240.0,
      "p50Cycles": 15240,
      "p95Cycles": 15240,
      "maximumCycles": 15240,
      "firstOffset": 306,
      "lastOffset": 56358
    },
    {
      "name": "gemm_f16_init_small_rows_interleaved_k384_c16_r114_r116",
      "dimensions": {},
      "sampleCount": 1656,
      "tileCount": 552,
      "phaseCount": 3,
      "phaseCycles": 55206,
      "phaseMs": 0.036804,
      "workCycles": 25108272,
      "averageActiveTiles": 454.8105640691229,
      "usefulCycles": 17950896.0,
      "estimatedWorkCycles": 25108272,
      "usefulUtilization": 0.7149395227198431,
      "workLaneOccupancy": 0.9904891304347826,
      "deviceUsefulUtilization": 0.2208981301654357,
      "meanCycles": 15162.0,
      "p50Cycles": 15162,
      "p95Cycles": 15162,
      "maximumCycles": 15162,
      "firstOffset": 312,
      "lastOffset": 56274
    },
    {
      "name": "fill_zero_u64",
      "dimensions": {},
      "sampleCount": 11536,
      "tileCount": 1472,
      "phaseCount": 38,
      "phaseCycles": 30366,
      "phaseMs": 0.020244,
      "workCycles": 4210656,
      "averageActiveTiles": 138.66350523611933,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 2139648,
      "usefulUtilization": 0.0,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": 0.0,
      "meanCycles": 365.00138696255203,
      "p50Cycles": 300,
      "p95Cycles": 600,
      "maximumCycles": 3276,
      "firstOffset": 64500,
      "lastOffset": 192348
    },
    {
      "name": "attention_softmax_d72_p768_tail_f16",
      "dimensions": {},
      "sampleCount": 1472,
      "tileCount": 1472,
      "phaseCount": 12,
      "phaseCycles": 27576,
      "phaseMs": 0.018384,
      "workCycles": 37145376,
      "averageActiveTiles": 1347.0182767624021,
      "usefulCycles": 18440784.0,
      "estimatedWorkCycles": 37145376,
      "usefulUtilization": 0.4964489792753747,
      "workLaneOccupancy": 0.9974763406940063,
      "deviceUsefulUtilization": 0.45429745147008743,
      "meanCycles": 25234.630434782608,
      "p50Cycles": 25236,
      "p95Cycles": 25236,
      "maximumCycles": 25236,
      "firstOffset": 122490,
      "lastOffset": 150066
    },
    {
      "name": "copy_u64",
      "dimensions": {},
      "sampleCount": 6189,
      "tileCount": 1472,
      "phaseCount": 36,
      "phaseCycles": 20286,
      "phaseMs": 0.013524,
      "workCycles": 5328306,
      "averageActiveTiles": 262.659272404614,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 860.9316529326223,
      "p50Cycles": 1002,
      "p95Cycles": 1230,
      "maximumCycles": 2010,
      "firstOffset": 58512,
      "lastOffset": 192858
    },
    {
      "name": "gemm_f16_init_small_rows_interleaved_k96_c80_r91_r92",
      "dimensions": {},
      "sampleCount": 896,
      "tileCount": 896,
      "phaseCount": 10,
      "phaseCycles": 19860,
      "phaseMs": 0.01324,
      "workCycles": 14477568,
      "averageActiveTiles": 728.9812688821752,
      "usefulCycles": 8358714.0,
      "estimatedWorkCycles": 14477568,
      "usefulUtilization": 0.5773562244708503,
      "workLaneOccupancy": 0.854296875,
      "deviceUsefulUtilization": 0.2859251855378957,
      "meanCycles": 16158.0,
      "p50Cycles": 16158,
      "p95Cycles": 16158,
      "maximumCycles": 16158,
      "firstOffset": 160710,
      "lastOffset": 180570
    },
    {
      "name": "gemm_f16_init_large_rows_interleaved_k96_c80_r91_r92",
      "dimensions": {},
      "sampleCount": 128,
      "tileCount": 128,
      "phaseCount": 5,
      "phaseCycles": 18438,
      "phaseMs": 0.012292,
      "workCycles": 2072832,
      "averageActiveTiles": 112.42173771558737,
      "usefulCycles": 1207224.0,
      "estimatedWorkCycles": 2072832,
      "usefulUtilization": 0.5824032048907003,
      "workLaneOccupancy": 0.854296875,
      "deviceUsefulUtilization": 0.04448014969085584,
      "meanCycles": 16194.0,
      "p50Cycles": 16194,
      "p95Cycles": 16194,
      "maximumCycles": 16194,
      "firstOffset": 162150,
      "lastOffset": 180588
    },
    {
      "name": "gemm_f16_init_small_rows_interleaved_k80_c80_r91_r92",
      "dimensions": {},
      "sampleCount": 224,
      "tileCount": 224,
      "phaseCount": 4,
      "phaseCycles": 14244,
      "phaseMs": 0.009496,
      "workCycles": 3033408,
      "averageActiveTiles": 212.9604043807919,
      "usefulCycles": 1834560.0,
      "estimatedWorkCycles": 3033408,
      "usefulUtilization": 0.6047851129818342,
      "workLaneOccupancy": 0.9,
      "deviceUsefulUtilization": 0.08749679498919453,
      "meanCycles": 13542.0,
      "p50Cycles": 13542,
      "p95Cycles": 13542,
      "maximumCycles": 13542,
      "firstOffset": 103560,
      "lastOffset": 117804
    },
    {
      "name": "gemm_f16_init_large_rows_interleaved_k80_c80_r91_r92",
      "dimensions": {},
      "sampleCount": 32,
      "tileCount": 32,
      "phaseCount": 2,
      "phaseCycles": 13902,
      "phaseMs": 0.009268,
      "workCycles": 434304,
      "averageActiveTiles": 31.24039706517048,
      "usefulCycles": 264960.0,
      "estimatedWorkCycles": 434304,
      "usefulUtilization": 0.610079575596817,
      "workLaneOccupancy": 0.9,
      "deviceUsefulUtilization": 0.01294777729823047,
      "meanCycles": 13572.0,
      "p50Cycles": 13572,
      "p95Cycles": 13572,
      "maximumCycles": 13572,
      "firstOffset": 103902,
      "lastOffset": 117804
    },
    {
      "name": "unpack_amp_to_row_major_f16_o1_r9_p16_c116_p116",
      "dimensions": {},
      "sampleCount": 12,
      "tileCount": 12,
      "phaseCount": 2,
      "phaseCycles": 12594,
      "phaseMs": 0.008396,
      "workCycles": 147456,
      "averageActiveTiles": 11.708432586946165,
      "usefulCycles": 6264.0,
      "estimatedWorkCycles": 147456,
      "usefulUtilization": 0.04248046875,
      "workLaneOccupancy": 0.5625,
      "deviceUsefulUtilization": 0.00033789382107421754,
      "meanCycles": 12288.0,
      "p50Cycles": 12288,
      "p95Cycles": 12288,
      "maximumCycles": 12288,
      "firstOffset": 64464,
      "lastOffset": 77058
    },
    {
      "name": "unpack_amp_to_row_major_f16_o1_r9_p16_c114_p114",
      "dimensions": {},
      "sampleCount": 8,
      "tileCount": 8,
      "phaseCount": 1,
      "phaseCycles": 12162,
      "phaseMs": 0.008108,
      "workCycles": 96672,
      "averageActiveTiles": 7.948692649235323,
      "usefulCycles": 4104.0,
      "estimatedWorkCycles": 96672,
      "usefulUtilization": 0.04245283018867924,
      "workLaneOccupancy": 0.5625,
      "deviceUsefulUtilization": 0.0002292421869972759,
      "meanCycles": 12084.0,
      "p50Cycles": 12084,
      "p95Cycles": 12084,
      "maximumCycles": 12084,
      "firstOffset": 64314,
      "lastOffset": 76476
    },
    {
      "name": "gemm_f16_init_small_rows_interleaved_k80_c64_r91_r92",
      "dimensions": {},
      "sampleCount": 1008,
      "tileCount": 1008,
      "phaseCount": 6,
      "phaseCycles": 11886,
      "phaseMs": 0.007924,
      "workCycles": 10983168,
      "averageActiveTiles": 924.0424028268551,
      "usefulCycles": 6524154.0,
      "estimatedWorkCycles": 10983168,
      "usefulUtilization": 0.5940138582966226,
      "workLaneOccupancy": 0.8890625,
      "deviceUsefulUtilization": 0.3728899408511292,
      "meanCycles": 10896.0,
      "p50Cycles": 10896,
      "p95Cycles": 10896,
      "maximumCycles": 10896,
      "firstOffset": 103158,
      "lastOffset": 115044
    },
    {
      "name": "rearrange_row_major_to_amp_f16_o0_r8_p8_c72_p80",
      "dimensions": {},
      "sampleCount": 1360,
      "tileCount": 1360,
      "phaseCount": 5,
      "phaseCycles": 11844,
      "phaseMs": 0.007896,
      "workCycles": 3215040,
      "averageActiveTiles": 271.44883485309015,
      "usefulCycles": 391680.0,
      "estimatedWorkCycles": 3215040,
      "usefulUtilization": 0.1218274111675127,
      "workLaneOccupancy": 0.9,
      "deviceUsefulUtilization": 0.02246597066208537,
      "meanCycles": 2364.0,
      "p50Cycles": 2364,
      "p95Cycles": 2364,
      "maximumCycles": 2364,
      "firstOffset": 81234,
      "lastOffset": 93078
    },
    {
      "name": "gemm_f16_init_large_rows_interleaved_k80_c64_r91_r92",
      "dimensions": {},
      "sampleCount": 144,
      "tileCount": 144,
      "phaseCount": 3,
      "phaseCycles": 11778,
      "phaseMs": 0.007852,
      "workCycles": 1572480,
      "averageActiveTiles": 133.50993377483445,
      "usefulCycles": 942264.0,
      "estimatedWorkCycles": 1572480,
      "usefulUtilization": 0.5992216117216117,
      "workLaneOccupancy": 0.8890625,
      "deviceUsefulUtilization": 0.05434921039225675,
      "meanCycles": 10920.0,
      "p50Cycles": 10920,
      "p95Cycles": 10920,
      "maximumCycles": 10920,
      "firstOffset": 103374,
      "lastOffset": 115152
    },
    {
      "name": "copy_strided_u64",
      "dimensions": {},
      "sampleCount": 1640,
      "tileCount": 1472,
      "phaseCount": 28,
      "phaseCycles": 11586,
      "phaseMs": 0.007724,
      "workCycles": 841332,
      "averageActiveTiles": 72.6162610046608,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 513.0073170731707,
      "p50Cycles": 510,
      "p95Cycles": 618,
      "maximumCycles": 834,
      "firstOffset": 75366,
      "lastOffset": 193368
    },
    {
      "name": "unpack_amp_to_row_major_f16_o1_r16_p16_c116_p116",
      "dimensions": {},
      "sampleCount": 540,
      "tileCount": 540,
      "phaseCount": 2,
      "phaseCycles": 11214,
      "phaseMs": 0.007476,
      "workCycles": 5883840,
      "averageActiveTiles": 524.6869983948635,
      "usefulCycles": 501120.0,
      "estimatedWorkCycles": 5883840,
      "usefulUtilization": 0.08516886930983847,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.030358015213901877,
      "meanCycles": 10896.0,
      "p50Cycles": 10896,
      "p95Cycles": 10896,
      "maximumCycles": 10896,
      "firstOffset": 64452,
      "lastOffset": 75666
    },
    {
      "name": "rearrange_row_major_to_amp_f16_o1_r64_p64_c16_p16",
      "dimensions": {},
      "sampleCount": 880,
      "tileCount": 880,
      "phaseCount": 4,
      "phaseCycles": 10980,
      "phaseMs": 0.00732,
      "workCycles": 1320000,
      "averageActiveTiles": 120.21857923497268,
      "usefulCycles": 450560.0,
      "estimatedWorkCycles": 1320000,
      "usefulUtilization": 0.3413333333333333,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.027876771996515403,
      "meanCycles": 1500.0,
      "p50Cycles": 1500,
      "p95Cycles": 1500,
      "maximumCycles": 1500,
      "firstOffset": 83598,
      "lastOffset": 94578
    },
    {
      "name": "unpack_amp_to_row_major_f16_o1_r16_p16_c114_p114",
      "dimensions": {},
      "sampleCount": 360,
      "tileCount": 360,
      "phaseCount": 1,
      "phaseCycles": 10806,
      "phaseMs": 0.007204,
      "workCycles": 3857760,
      "averageActiveTiles": 357.00166574125484,
      "usefulCycles": 328320.0,
      "estimatedWorkCycles": 3857760,
      "usefulUtilization": 0.0851063829787234,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.020640706853680323,
      "meanCycles": 10716.0,
      "p50Cycles": 10716,
      "p95Cycles": 10716,
      "maximumCycles": 10716,
      "firstOffset": 64302,
      "lastOffset": 75108
    },
    {
      "name": "reduce_sum_f16",
      "dimensions": {},
      "sampleCount": 2852,
      "tileCount": 1472,
      "phaseCount": 18,
      "phaseCycles": 10044,
      "phaseMs": 0.006696,
      "workCycles": 8673840,
      "averageActiveTiles": 863.584229390681,
      "usefulCycles": 2904768.0,
      "estimatedWorkCycles": 8673840,
      "usefulUtilization": 0.33488835394704075,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.19647031323048153,
      "meanCycles": 3041.3183730715286,
      "p50Cycles": 2568,
      "p95Cycles": 3792,
      "maximumCycles": 3792,
      "firstOffset": 59436,
      "lastOffset": 190356
    },
    {
      "name": "rearrange_row_major_to_amp_f16_o1_r0_p64_c0_p16",
      "dimensions": {},
      "sampleCount": 80,
      "tileCount": 80,
      "phaseCount": 2,
      "phaseCycles": 9822,
      "phaseMs": 0.006548,
      "workCycles": 216960,
      "averageActiveTiles": 22.089187538179598,
      "usefulCycles": 16000.0,
      "estimatedWorkCycles": 216960,
      "usefulUtilization": 0.07374631268436578,
      "workLaneOccupancy": 0.390625,
      "deviceUsefulUtilization": 0.0011066549803900737,
      "meanCycles": 2712.0,
      "p50Cycles": 2712,
      "p95Cycles": 2712,
      "maximumCycles": 2712,
      "firstOffset": 83850,
      "lastOffset": 93672
    },
    {
      "name": "attention_merge_v72_p80_k768_f16",
      "dimensions": {},
      "sampleCount": 1472,
      "tileCount": 1472,
      "phaseCount": 18,
      "phaseCycles": 7494,
      "phaseMs": 0.004996,
      "workCycles": 3629952,
      "averageActiveTiles": 484.38110488390714,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 2466.0,
      "p50Cycles": 2466,
      "p95Cycles": 2466,
      "maximumCycles": 2466,
      "firstOffset": 188340,
      "lastOffset": 195834
    },
    {
      "name": "rearrange_row_major_to_amp_f16_o0_r7_p7_c72_p80",
      "dimensions": {},
      "sampleCount": 112,
      "tileCount": 112,
      "phaseCount": 2,
      "phaseCycles": 6168,
      "phaseMs": 0.004112,
      "workCycles": 264768,
      "averageActiveTiles": 42.9260700389105,
      "usefulCycles": 28224.0,
      "estimatedWorkCycles": 264768,
      "usefulUtilization": 0.1065989847715736,
      "workLaneOccupancy": 0.9,
      "deviceUsefulUtilization": 0.0031086110641177465,
      "meanCycles": 2364.0,
      "p50Cycles": 2364,
      "p95Cycles": 2364,
      "maximumCycles": 2364,
      "firstOffset": 80022,
      "lastOffset": 92484
    },
    {
      "name": "copy_strided_u32",
      "dimensions": {},
      "sampleCount": 1,
      "tileCount": 1,
      "phaseCount": 1,
      "phaseCycles": 1086,
      "phaseMs": 0.000724,
      "workCycles": 1086,
      "averageActiveTiles": 1.0,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 1086.0,
      "p50Cycles": 1086,
      "p95Cycles": 1086,
      "maximumCycles": 1086,
      "firstOffset": 75066,
      "lastOffset": 76152
    }
  ],
  "samples": []
}
