{
  "clockHz": 1500000000,
  "tileCount": 1472,
  "sampleCount": 43505,
  "matchedSampleCount": 43505,
  "profileSpanCycles": 195096,
  "profileSpanMs": 0.130064,
  "atOffset": null,
  "groups": [
    {
      "name": "exchange",
      "dimensions": {},
      "sampleCount": 10350,
      "tileCount": 1472,
      "phaseCount": 8,
      "phaseCycles": 125160,
      "phaseMs": 0.08344,
      "workCycles": 113574174,
      "averageActiveTiles": 907.4318791946308,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 10973.350144927535,
      "p50Cycles": 9276,
      "p95Cycles": 23712,
      "maximumCycles": 30336,
      "firstOffset": 0,
      "lastOffset": 186684
    },
    {
      "name": "gemm_f16_init_small_rows_interleaved_k576_c16_r216_r216",
      "dimensions": {},
      "sampleCount": 1472,
      "tileCount": 1472,
      "phaseCount": 1,
      "phaseCycles": 44382,
      "phaseMs": 0.029588,
      "workCycles": 54873216,
      "averageActiveTiles": 1236.3844801946734,
      "usefulCycles": 45349632.0,
      "estimatedWorkCycles": 54873216,
      "usefulUtilization": 0.8264438519513783,
      "workLaneOccupancy": 0.9904891304347826,
      "deviceUsefulUtilization": 0.6941592067289324,
      "meanCycles": 37278.0,
      "p50Cycles": 37278,
      "p95Cycles": 37278,
      "maximumCycles": 37278,
      "firstOffset": 258,
      "lastOffset": 44640
    },
    {
      "name": "fill_zero_u64",
      "dimensions": {},
      "sampleCount": 11536,
      "tileCount": 1472,
      "phaseCount": 41,
      "phaseCycles": 31332,
      "phaseMs": 0.020888,
      "workCycles": 4210656,
      "averageActiveTiles": 134.38835695135964,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 2139648,
      "usefulUtilization": 0.0,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": 0.0,
      "meanCycles": 365.00138696255203,
      "p50Cycles": 300,
      "p95Cycles": 600,
      "maximumCycles": 3276,
      "firstOffset": 56970,
      "lastOffset": 191610
    },
    {
      "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": 120270,
      "lastOffset": 148776
    },
    {
      "name": "copy_u64",
      "dimensions": {},
      "sampleCount": 5597,
      "tileCount": 1472,
      "phaseCount": 45,
      "phaseCycles": 25854,
      "phaseMs": 0.017236,
      "workCycles": 5190198,
      "averageActiveTiles": 200.75029009050823,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 927.3178488475969,
      "p50Cycles": 570,
      "p95Cycles": 1974,
      "maximumCycles": 2010,
      "firstOffset": 46794,
      "lastOffset": 192120
    },
    {
      "name": "gemm_f16_init_small_rows_interleaved_k96_c80_r91_r92",
      "dimensions": {},
      "sampleCount": 896,
      "tileCount": 896,
      "phaseCount": 13,
      "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": 159474,
      "lastOffset": 179712
    },
    {
      "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": 160956,
      "lastOffset": 179760
    },
    {
      "name": "unpack_amp_to_row_major_f16_o1_r9_p16_c72_p72",
      "dimensions": {},
      "sampleCount": 32,
      "tileCount": 16,
      "phaseCount": 4,
      "phaseCycles": 17040,
      "phaseMs": 0.01136,
      "workCycles": 249600,
      "averageActiveTiles": 14.647887323943662,
      "usefulCycles": 10368.0,
      "estimatedWorkCycles": 249600,
      "usefulUtilization": 0.04153846153846154,
      "workLaneOccupancy": 0.5625,
      "deviceUsefulUtilization": 0.0004133496631965707,
      "meanCycles": 7800.0,
      "p50Cycles": 7800,
      "p95Cycles": 7800,
      "maximumCycles": 7800,
      "firstOffset": 59310,
      "lastOffset": 76350
    },
    {
      "name": "unpack_amp_to_row_major_f16_o1_r16_p16_c72_p72",
      "dimensions": {},
      "sampleCount": 1440,
      "tileCount": 720,
      "phaseCount": 4,
      "phaseCycles": 15996,
      "phaseMs": 0.010664,
      "workCycles": 9987840,
      "averageActiveTiles": 624.3960990247562,
      "usefulCycles": 829440.0,
      "estimatedWorkCycles": 9987840,
      "usefulUtilization": 0.08304498269896193,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.03522619785381128,
      "meanCycles": 6936.0,
      "p50Cycles": 6936,
      "p95Cycles": 6936,
      "maximumCycles": 6936,
      "firstOffset": 59298,
      "lastOffset": 75294
    },
    {
      "name": "gemm_f16_init_small_rows_interleaved_k80_c80_r91_r92",
      "dimensions": {},
      "sampleCount": 224,
      "tileCount": 224,
      "phaseCount": 5,
      "phaseCycles": 14640,
      "phaseMs": 0.00976,
      "workCycles": 3033408,
      "averageActiveTiles": 207.2,
      "usefulCycles": 1834560.0,
      "estimatedWorkCycles": 3033408,
      "usefulUtilization": 0.6047851129818342,
      "workLaneOccupancy": 0.9,
      "deviceUsefulUtilization": 0.08513007840342124,
      "meanCycles": 13542.0,
      "p50Cycles": 13542,
      "p95Cycles": 13542,
      "maximumCycles": 13542,
      "firstOffset": 101292,
      "lastOffset": 115932
    },
    {
      "name": "gemm_f16_init_large_rows_interleaved_k80_c80_r91_r92",
      "dimensions": {},
      "sampleCount": 32,
      "tileCount": 32,
      "phaseCount": 3,
      "phaseCycles": 14352,
      "phaseMs": 0.009568,
      "workCycles": 434304,
      "averageActiveTiles": 30.26086956521739,
      "usefulCycles": 264960.0,
      "estimatedWorkCycles": 434304,
      "usefulUtilization": 0.610079575596817,
      "workLaneOccupancy": 0.9,
      "deviceUsefulUtilization": 0.01254180602006689,
      "meanCycles": 13572.0,
      "p50Cycles": 13572,
      "p95Cycles": 13572,
      "maximumCycles": 13572,
      "firstOffset": 101610,
      "lastOffset": 115962
    },
    {
      "name": "copy_strided_u64",
      "dimensions": {},
      "sampleCount": 1636,
      "tileCount": 1472,
      "phaseCount": 29,
      "phaseCycles": 13374,
      "phaseMs": 0.008916,
      "workCycles": 838536,
      "averageActiveTiles": 62.69896814715119,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 512.5525672371638,
      "p50Cycles": 510,
      "p95Cycles": 618,
      "maximumCycles": 834,
      "firstOffset": 73566,
      "lastOffset": 192630
    },
    {
      "name": "reduce_sum_f16",
      "dimensions": {},
      "sampleCount": 2208,
      "tileCount": 1472,
      "phaseCount": 21,
      "phaseCycles": 12690,
      "phaseMs": 0.00846,
      "workCycles": 7831488,
      "averageActiveTiles": 617.1385342789598,
      "usefulCycles": 2268864.0,
      "estimatedWorkCycles": 7831488,
      "usefulUtilization": 0.28971046115374244,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.12146160962072156,
      "meanCycles": 3546.8695652173915,
      "p50Cycles": 2382,
      "p95Cycles": 5898,
      "maximumCycles": 5898,
      "firstOffset": 48414,
      "lastOffset": 189714
    },
    {
      "name": "gemm_f16_init_small_rows_interleaved_k80_c64_r91_r92",
      "dimensions": {},
      "sampleCount": 1008,
      "tileCount": 1008,
      "phaseCount": 6,
      "phaseCycles": 12570,
      "phaseMs": 0.00838,
      "workCycles": 10983168,
      "averageActiveTiles": 873.7603818615752,
      "usefulCycles": 6524154.0,
      "estimatedWorkCycles": 10983168,
      "usefulUtilization": 0.5940138582966226,
      "workLaneOccupancy": 0.8890625,
      "deviceUsefulUtilization": 0.3525990323752205,
      "meanCycles": 10896.0,
      "p50Cycles": 10896,
      "p95Cycles": 10896,
      "maximumCycles": 10896,
      "firstOffset": 100758,
      "lastOffset": 113328
    },
    {
      "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": 100770,
      "lastOffset": 113280
    },
    {
      "name": "rearrange_row_major_to_amp_f16_o0_r8_p8_c72_p80",
      "dimensions": {},
      "sampleCount": 1360,
      "tileCount": 1360,
      "phaseCount": 8,
      "phaseCycles": 11346,
      "phaseMs": 0.007564,
      "workCycles": 3215040,
      "averageActiveTiles": 283.3632998413538,
      "usefulCycles": 391680.0,
      "estimatedWorkCycles": 3215040,
      "usefulUtilization": 0.1218274111675127,
      "workLaneOccupancy": 0.9,
      "deviceUsefulUtilization": 0.02345204975513301,
      "meanCycles": 2364.0,
      "p50Cycles": 2364,
      "p95Cycles": 2364,
      "maximumCycles": 2364,
      "firstOffset": 78216,
      "lastOffset": 89562
    },
    {
      "name": "sync",
      "dimensions": {},
      "sampleCount": 1426,
      "tileCount": 1058,
      "phaseCount": 2,
      "phaseCycles": 10662,
      "phaseMs": 0.007108,
      "workCycles": 7148442,
      "averageActiveTiles": 670.4597636465954,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 5012.93267882188,
      "p50Cycles": 8556,
      "p95Cycles": 10326,
      "maximumCycles": 10332,
      "firstOffset": 0,
      "lastOffset": 57012
    },
    {
      "name": "rearrange_row_major_to_amp_f16_o1_r64_p64_c16_p16",
      "dimensions": {},
      "sampleCount": 880,
      "tileCount": 880,
      "phaseCount": 8,
      "phaseCycles": 9606,
      "phaseMs": 0.006404,
      "workCycles": 1320000,
      "averageActiveTiles": 137.41411617738913,
      "usefulCycles": 450560.0,
      "estimatedWorkCycles": 1320000,
      "usefulUtilization": 0.3413333333333333,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.03186414288171342,
      "meanCycles": 1500.0,
      "p50Cycles": 1500,
      "p95Cycles": 1500,
      "maximumCycles": 1500,
      "firstOffset": 81456,
      "lastOffset": 91062
    },
    {
      "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": 187110,
      "lastOffset": 195096
    },
    {
      "name": "rearrange_row_major_to_amp_f16_o1_r0_p64_c0_p16",
      "dimensions": {},
      "sampleCount": 80,
      "tileCount": 80,
      "phaseCount": 2,
      "phaseCycles": 5226,
      "phaseMs": 0.003484,
      "workCycles": 216960,
      "averageActiveTiles": 41.51549942594719,
      "usefulCycles": 16000.0,
      "estimatedWorkCycles": 216960,
      "usefulUtilization": 0.07374631268436578,
      "workLaneOccupancy": 0.390625,
      "deviceUsefulUtilization": 0.0020799014958651557,
      "meanCycles": 2712.0,
      "p50Cycles": 2712,
      "p95Cycles": 2712,
      "maximumCycles": 2712,
      "firstOffset": 81030,
      "lastOffset": 86256
    },
    {
      "name": "rearrange_row_major_to_amp_f16_o0_r7_p7_c72_p80",
      "dimensions": {},
      "sampleCount": 112,
      "tileCount": 112,
      "phaseCount": 1,
      "phaseCycles": 3408,
      "phaseMs": 0.002272,
      "workCycles": 264768,
      "averageActiveTiles": 77.69014084507042,
      "usefulCycles": 28224.0,
      "estimatedWorkCycles": 264768,
      "usefulUtilization": 0.1065989847715736,
      "workLaneOccupancy": 0.9,
      "deviceUsefulUtilization": 0.005626148193508879,
      "meanCycles": 2364.0,
      "p50Cycles": 2364,
      "p95Cycles": 2364,
      "maximumCycles": 2364,
      "firstOffset": 78180,
      "lastOffset": 81588
    }
  ],
  "samples": []
}
