{
  "clockHz": 1500000000,
  "tileCount": 1472,
  "sampleCount": 49816,
  "matchedSampleCount": 49816,
  "profileSpanCycles": 203736,
  "profileSpanMs": 0.135824,
  "atOffset": null,
  "groups": [
    {
      "name": "exchange",
      "dimensions": {},
      "sampleCount": 13018,
      "tileCount": 1472,
      "phaseCount": 10,
      "phaseCycles": 132828,
      "phaseMs": 0.088552,
      "workCycles": 115762566,
      "averageActiveTiles": 871.5223145722288,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 8892.500076816716,
      "p50Cycles": 7302,
      "p95Cycles": 23022,
      "maximumCycles": 30540,
      "firstOffset": 0,
      "lastOffset": 195846
    },
    {
      "name": "sync",
      "dimensions": {},
      "sampleCount": 1702,
      "tileCount": 506,
      "phaseCount": 5,
      "phaseCycles": 65904,
      "phaseMs": 0.043936,
      "workCycles": 8762898,
      "averageActiveTiles": 132.96458485069192,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 5148.588719153937,
      "p50Cycles": 3540,
      "p95Cycles": 18780,
      "maximumCycles": 18780,
      "firstOffset": 0,
      "lastOffset": 65904
    },
    {
      "name": "gemm_f16_init_large_rows_interleaved_k384_c16_r114_r116",
      "dimensions": {},
      "sampleCount": 2484,
      "tileCount": 828,
      "phaseCount": 4,
      "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": 65634
    },
    {
      "name": "gemm_f16_init_small_rows_interleaved_k384_c16_r114_r116",
      "dimensions": {},
      "sampleCount": 1656,
      "tileCount": 552,
      "phaseCount": 4,
      "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": 65550
    },
    {
      "name": "fill_zero_u64",
      "dimensions": {},
      "sampleCount": 11536,
      "tileCount": 1472,
      "phaseCount": 45,
      "phaseCycles": 31878,
      "phaseMs": 0.021252,
      "workCycles": 4210656,
      "averageActiveTiles": 132.08658008658008,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 2139648,
      "usefulUtilization": 0.0,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": 0.0,
      "meanCycles": 365.00138696255203,
      "p50Cycles": 300,
      "p95Cycles": 600,
      "maximumCycles": 3276,
      "firstOffset": 68670,
      "lastOffset": 200250
    },
    {
      "name": "attention_softmax_d72_p768_tail_f16",
      "dimensions": {},
      "sampleCount": 1472,
      "tileCount": 1472,
      "phaseCount": 16,
      "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": 130392,
      "lastOffset": 157968
    },
    {
      "name": "copy_u64",
      "dimensions": {},
      "sampleCount": 6199,
      "tileCount": 1472,
      "phaseCount": 45,
      "phaseCycles": 26334,
      "phaseMs": 0.017556,
      "workCycles": 5333034,
      "averageActiveTiles": 202.5151515151515,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 860.3055331505082,
      "p50Cycles": 1002,
      "p95Cycles": 1230,
      "maximumCycles": 2010,
      "firstOffset": 40638,
      "lastOffset": 200760
    },
    {
      "name": "gemm_f16_init_small_rows_interleaved_k96_c80_r91_r92",
      "dimensions": {},
      "sampleCount": 896,
      "tileCount": 896,
      "phaseCount": 11,
      "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": 168612,
      "lastOffset": 188472
    },
    {
      "name": "unpack_amp_to_row_major_f16_o1_r9_p16_c116_p116",
      "dimensions": {},
      "sampleCount": 12,
      "tileCount": 12,
      "phaseCount": 4,
      "phaseCycles": 19470,
      "phaseMs": 0.01298,
      "workCycles": 147456,
      "averageActiveTiles": 7.573497688751926,
      "usefulCycles": 6264.0,
      "estimatedWorkCycles": 147456,
      "usefulUtilization": 0.04248046875,
      "workLaneOccupancy": 0.5625,
      "deviceUsefulUtilization": 0.00021856367655925503,
      "meanCycles": 12288.0,
      "p50Cycles": 12288,
      "p95Cycles": 12288,
      "maximumCycles": 12288,
      "firstOffset": 66834,
      "lastOffset": 86304
    },
    {
      "name": "unpack_amp_to_row_major_f16_o1_r9_p16_c114_p114",
      "dimensions": {},
      "sampleCount": 8,
      "tileCount": 8,
      "phaseCount": 2,
      "phaseCycles": 18876,
      "phaseMs": 0.012584,
      "workCycles": 96672,
      "averageActiveTiles": 5.121424030514939,
      "usefulCycles": 4104.0,
      "estimatedWorkCycles": 96672,
      "usefulUtilization": 0.04245283018867924,
      "workLaneOccupancy": 0.5625,
      "deviceUsefulUtilization": 0.0001477030874264076,
      "meanCycles": 12084.0,
      "p50Cycles": 12084,
      "p95Cycles": 12084,
      "maximumCycles": 12084,
      "firstOffset": 66846,
      "lastOffset": 85722
    },
    {
      "name": "gemm_f16_init_large_rows_interleaved_k96_c80_r91_r92",
      "dimensions": {},
      "sampleCount": 128,
      "tileCount": 128,
      "phaseCount": 8,
      "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": 170052,
      "lastOffset": 188490
    },
    {
      "name": "unpack_amp_to_row_major_f16_o1_r16_p16_c116_p116",
      "dimensions": {},
      "sampleCount": 540,
      "tileCount": 540,
      "phaseCount": 4,
      "phaseCycles": 18078,
      "phaseMs": 0.012052,
      "workCycles": 5883840,
      "averageActiveTiles": 325.4696315964155,
      "usefulCycles": 501120.0,
      "estimatedWorkCycles": 5883840,
      "usefulUtilization": 0.08516886930983847,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.018831440569127982,
      "meanCycles": 10896.0,
      "p50Cycles": 10896,
      "p95Cycles": 10896,
      "maximumCycles": 10896,
      "firstOffset": 66834,
      "lastOffset": 84912
    },
    {
      "name": "unpack_amp_to_row_major_f16_o1_r16_p16_c114_p114",
      "dimensions": {},
      "sampleCount": 360,
      "tileCount": 360,
      "phaseCount": 2,
      "phaseCycles": 17532,
      "phaseMs": 0.011688,
      "workCycles": 3857760,
      "averageActiveTiles": 220.04106776180697,
      "usefulCycles": 328320.0,
      "estimatedWorkCycles": 3857760,
      "usefulUtilization": 0.0851063829787234,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.012722078385858406,
      "meanCycles": 10716.0,
      "p50Cycles": 10716,
      "p95Cycles": 10716,
      "maximumCycles": 10716,
      "firstOffset": 66822,
      "lastOffset": 84354
    },
    {
      "name": "reduce_sum_f16",
      "dimensions": {},
      "sampleCount": 2852,
      "tileCount": 1472,
      "phaseCount": 23,
      "phaseCycles": 15042,
      "phaseMs": 0.010028,
      "workCycles": 8673840,
      "averageActiveTiles": 576.6414040686079,
      "usefulCycles": 2904768.0,
      "estimatedWorkCycles": 8673840,
      "usefulUtilization": 0.33488835394704075,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.13118919200152615,
      "meanCycles": 3041.3183730715286,
      "p50Cycles": 2568,
      "p95Cycles": 3792,
      "maximumCycles": 3792,
      "firstOffset": 40668,
      "lastOffset": 198258
    },
    {
      "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": 111462,
      "lastOffset": 125706
    },
    {
      "name": "gemm_f16_init_large_rows_interleaved_k80_c80_r91_r92",
      "dimensions": {},
      "sampleCount": 32,
      "tileCount": 32,
      "phaseCount": 3,
      "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": 111804,
      "lastOffset": 125706
    },
    {
      "name": "copy_strided_u64",
      "dimensions": {},
      "sampleCount": 1640,
      "tileCount": 1472,
      "phaseCount": 30,
      "phaseCycles": 12576,
      "phaseMs": 0.008384,
      "workCycles": 841332,
      "averageActiveTiles": 66.89980916030534,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 513.0073170731707,
      "p50Cycles": 510,
      "p95Cycles": 618,
      "maximumCycles": 834,
      "firstOffset": 77736,
      "lastOffset": 201270
    },
    {
      "name": "gemm_f16_init_small_rows_interleaved_k80_c64_r91_r92",
      "dimensions": {},
      "sampleCount": 1008,
      "tileCount": 1008,
      "phaseCount": 8,
      "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": 111060,
      "lastOffset": 122946
    },
    {
      "name": "gemm_f16_init_large_rows_interleaved_k80_c64_r91_r92",
      "dimensions": {},
      "sampleCount": 144,
      "tileCount": 144,
      "phaseCount": 5,
      "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": 111276,
      "lastOffset": 123054
    },
    {
      "name": "rearrange_row_major_to_amp_f16_o0_r8_p8_c72_p80",
      "dimensions": {},
      "sampleCount": 1360,
      "tileCount": 1360,
      "phaseCount": 10,
      "phaseCycles": 11562,
      "phaseMs": 0.007708,
      "workCycles": 3215040,
      "averageActiveTiles": 278.0695381421899,
      "usefulCycles": 391680.0,
      "estimatedWorkCycles": 3215040,
      "usefulUtilization": 0.1218274111675127,
      "workLaneOccupancy": 0.9,
      "deviceUsefulUtilization": 0.023013921166038672,
      "meanCycles": 2364.0,
      "p50Cycles": 2364,
      "p95Cycles": 2364,
      "maximumCycles": 2364,
      "firstOffset": 89418,
      "lastOffset": 100980
    },
    {
      "name": "rearrange_row_major_to_amp_f16_o1_r64_p64_c16_p16",
      "dimensions": {},
      "sampleCount": 880,
      "tileCount": 880,
      "phaseCount": 8,
      "phaseCycles": 9492,
      "phaseMs": 0.006328,
      "workCycles": 1320000,
      "averageActiveTiles": 139.0644753476612,
      "usefulCycles": 450560.0,
      "estimatedWorkCycles": 1320000,
      "usefulUtilization": 0.3413333333333333,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.032246834863225785,
      "meanCycles": 1500.0,
      "p50Cycles": 1500,
      "p95Cycles": 1500,
      "maximumCycles": 1500,
      "firstOffset": 92988,
      "lastOffset": 102480
    },
    {
      "name": "attention_merge_v72_p80_k768_f16",
      "dimensions": {},
      "sampleCount": 1472,
      "tileCount": 1472,
      "phaseCount": 24,
      "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": 196242,
      "lastOffset": 203736
    },
    {
      "name": "rearrange_row_major_to_amp_f16_o0_r7_p7_c72_p80",
      "dimensions": {},
      "sampleCount": 112,
      "tileCount": 112,
      "phaseCount": 2,
      "phaseCycles": 6270,
      "phaseMs": 0.00418,
      "workCycles": 264768,
      "averageActiveTiles": 42.22775119617225,
      "usefulCycles": 28224.0,
      "estimatedWorkCycles": 264768,
      "usefulUtilization": 0.1065989847715736,
      "workLaneOccupancy": 0.9,
      "deviceUsefulUtilization": 0.003058040357811525,
      "meanCycles": 2364.0,
      "p50Cycles": 2364,
      "p95Cycles": 2364,
      "maximumCycles": 2364,
      "firstOffset": 89274,
      "lastOffset": 95544
    },
    {
      "name": "rearrange_row_major_to_amp_f16_o1_r0_p64_c0_p16",
      "dimensions": {},
      "sampleCount": 80,
      "tileCount": 80,
      "phaseCount": 3,
      "phaseCycles": 5844,
      "phaseMs": 0.003896,
      "workCycles": 216960,
      "averageActiveTiles": 37.125256673511295,
      "usefulCycles": 16000.0,
      "estimatedWorkCycles": 216960,
      "usefulUtilization": 0.07374631268436578,
      "workLaneOccupancy": 0.390625,
      "deviceUsefulUtilization": 0.0018599529803886557,
      "meanCycles": 2712.0,
      "p50Cycles": 2712,
      "p95Cycles": 2712,
      "maximumCycles": 2712,
      "firstOffset": 92340,
      "lastOffset": 98184
    },
    {
      "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": 84318,
      "lastOffset": 85404
    }
  ],
  "samples": []
}
