{
  "clockHz": 1500000000,
  "tileCount": 1472,
  "sampleCount": 57889,
  "matchedSampleCount": 55945,
  "profileSpanCycles": 221076,
  "profileSpanMs": 0.147384,
  "atOffset": null,
  "groups": [
    {
      "name": "exchange",
      "dimensions": {},
      "sampleCount": 14421,
      "tileCount": 1472,
      "phaseCount": 11,
      "phaseCycles": 145140,
      "phaseMs": 0.09676,
      "workCycles": 126849642,
      "averageActiveTiles": 873.9812732534104,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 8796.175161223216,
      "p50Cycles": 8646,
      "p95Cycles": 14694,
      "maximumCycles": 31260,
      "firstOffset": 0,
      "lastOffset": 220458
    },
    {
      "name": "sync",
      "dimensions": {},
      "sampleCount": 1771,
      "tileCount": 986,
      "phaseCount": 5,
      "phaseCycles": 94098,
      "phaseMs": 0.062732,
      "workCycles": 23647860,
      "averageActiveTiles": 251.31097366575273,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 13352.828910220214,
      "p50Cycles": 14466,
      "p95Cycles": 19620,
      "maximumCycles": 31242,
      "firstOffset": 0,
      "lastOffset": 94098
    },
    {
      "name": "copy_strided_u64",
      "dimensions": {},
      "sampleCount": 4661,
      "tileCount": 1472,
      "phaseCount": 51,
      "phaseCycles": 34836,
      "phaseMs": 0.023224,
      "workCycles": 26454810,
      "averageActiveTiles": 759.4100930072339,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 5675.779875563184,
      "p50Cycles": 1134,
      "p95Cycles": 16620,
      "maximumCycles": 16620,
      "firstOffset": 0,
      "lastOffset": 214788
    },
    {
      "name": "fill_zero_u64",
      "dimensions": {},
      "sampleCount": 12896,
      "tileCount": 1472,
      "phaseCount": 51,
      "phaseCycles": 31170,
      "phaseMs": 0.02078,
      "workCycles": 4859808,
      "averageActiveTiles": 155.91299326275265,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 2696160,
      "usefulUtilization": 0.0,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": 0.0,
      "meanCycles": 376.84615384615387,
      "p50Cycles": 300,
      "p95Cycles": 714,
      "maximumCycles": 714,
      "firstOffset": 94068,
      "lastOffset": 213864
    },
    {
      "name": "copy_u64",
      "dimensions": {},
      "sampleCount": 6524,
      "tileCount": 1472,
      "phaseCount": 54,
      "phaseCycles": 27666,
      "phaseMs": 0.018444,
      "workCycles": 7723476,
      "averageActiveTiles": 279.16851008458036,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 1183.8559166155733,
      "p50Cycles": 1002,
      "p95Cycles": 3702,
      "maximumCycles": 3702,
      "firstOffset": 0,
      "lastOffset": 221076
    },
    {
      "name": "attention_softmax_d72_p768_tail_f16",
      "dimensions": {},
      "sampleCount": 1472,
      "tileCount": 1472,
      "phaseCount": 20,
      "phaseCycles": 25728,
      "phaseMs": 0.017152,
      "workCycles": 34884384,
      "averageActiveTiles": 1355.891791044776,
      "usefulCycles": 15816384.0,
      "estimatedWorkCycles": 34884384,
      "usefulUtilization": 0.4533943898794372,
      "workLaneOccupancy": 0.9970588235294118,
      "deviceUsefulUtilization": 0.4176316109669046,
      "meanCycles": 23698.630434782608,
      "p50Cycles": 23700,
      "p95Cycles": 23700,
      "maximumCycles": 23700,
      "firstOffset": 144858,
      "lastOffset": 170586
    },
    {
      "name": "gemm_f8_init_large_rows_interleaved_k384_c64_r80_r82",
      "dimensions": {},
      "sampleCount": 810,
      "tileCount": 810,
      "phaseCount": 3,
      "phaseCycles": 24516,
      "phaseMs": 0.016344,
      "workCycles": 19746180,
      "averageActiveTiles": 805.4405286343613,
      "usefulCycles": 12752640.0,
      "estimatedWorkCycles": 19746180,
      "usefulUtilization": 0.6458282057592911,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.3533805784332503,
      "meanCycles": 24378.0,
      "p50Cycles": 24378,
      "p95Cycles": 24378,
      "maximumCycles": 24378,
      "firstOffset": 39132,
      "lastOffset": 63648
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k384_c64_r80_r82",
      "dimensions": {},
      "sampleCount": 648,
      "tileCount": 648,
      "phaseCount": 2,
      "phaseCycles": 24030,
      "phaseMs": 0.01602,
      "workCycles": 15528672,
      "averageActiveTiles": 646.2202247191011,
      "usefulCycles": 9922176.0,
      "estimatedWorkCycles": 15528672,
      "usefulUtilization": 0.6389584376564847,
      "workLaneOccupancy": 0.996875,
      "deviceUsefulUtilization": 0.28050806057645333,
      "meanCycles": 23964.0,
      "p50Cycles": 23964,
      "p95Cycles": 23964,
      "maximumCycles": 23964,
      "firstOffset": 39198,
      "lastOffset": 63228
    },
    {
      "name": "gemm_f16_init_small_rows_interleaved_k80_c80_r81_r81",
      "dimensions": {},
      "sampleCount": 864,
      "tileCount": 864,
      "phaseCount": 14,
      "phaseCycles": 19224,
      "phaseMs": 0.012816,
      "workCycles": 11057472,
      "averageActiveTiles": 575.1910112359551,
      "usefulCycles": 6298560.0,
      "estimatedWorkCycles": 11057472,
      "usefulUtilization": 0.569620253164557,
      "workLaneOccupancy": 0.9,
      "deviceUsefulUtilization": 0.2225818270639961,
      "meanCycles": 12798.0,
      "p50Cycles": 12798,
      "p95Cycles": 12798,
      "maximumCycles": 12798,
      "firstOffset": 120330,
      "lastOffset": 139554
    },
    {
      "name": "gemm_f16_init_small_rows_interleaved_k80_c64_r81_r81",
      "dimensions": {},
      "sampleCount": 576,
      "tileCount": 576,
      "phaseCount": 10,
      "phaseCycles": 17268,
      "phaseMs": 0.011512,
      "workCycles": 5940864,
      "averageActiveTiles": 344.03891591382904,
      "usefulCycles": 3267378.0,
      "estimatedWorkCycles": 5940864,
      "usefulUtilization": 0.5499836387434555,
      "workLaneOccupancy": 0.875390625,
      "deviceUsefulUtilization": 0.12854332530138685,
      "meanCycles": 10314.0,
      "p50Cycles": 10314,
      "p95Cycles": 10314,
      "maximumCycles": 10314,
      "firstOffset": 119718,
      "lastOffset": 136986
    },
    {
      "name": "cast_f16_f8",
      "dimensions": {},
      "sampleCount": 2880,
      "tileCount": 1440,
      "phaseCount": 22,
      "phaseCycles": 15510,
      "phaseMs": 0.01034,
      "workCycles": 17096832,
      "averageActiveTiles": 1102.310251450677,
      "usefulCycles": 2694384.0,
      "estimatedWorkCycles": 17096832,
      "usefulUtilization": 0.15759551243177683,
      "workLaneOccupancy": 0.9093604199066874,
      "deviceUsefulUtilization": 0.11801572617946346,
      "meanCycles": 5936.4,
      "p50Cycles": 5100,
      "p95Cycles": 6888,
      "maximumCycles": 6888,
      "firstOffset": 182328,
      "lastOffset": 197838
    },
    {
      "name": "reduce_sum_f16",
      "dimensions": {},
      "sampleCount": 1958,
      "tileCount": 1472,
      "phaseCount": 28,
      "phaseCycles": 14298,
      "phaseMs": 0.009532,
      "workCycles": 7912308,
      "averageActiveTiles": 553.3856483424255,
      "usefulCycles": 2427840.0,
      "estimatedWorkCycles": 7912308,
      "usefulUtilization": 0.30684346463762535,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.11535514240361984,
      "meanCycles": 4041.015321756895,
      "p50Cycles": 2046,
      "p95Cycles": 10182,
      "maximumCycles": 10182,
      "firstOffset": 71664,
      "lastOffset": 211590
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k128_c80_r48_r49",
      "dimensions": {},
      "sampleCount": 576,
      "tileCount": 576,
      "phaseCount": 12,
      "phaseCycles": 11352,
      "phaseMs": 0.007568,
      "workCycles": 4516992,
      "averageActiveTiles": 397.90274841437633,
      "usefulCycles": 1889568.0,
      "estimatedWorkCycles": 4516992,
      "usefulUtilization": 0.4183244070390207,
      "workLaneOccupancy": 0.854296875,
      "deviceUsefulUtilization": 0.11307909734350584,
      "meanCycles": 7842.0,
      "p50Cycles": 7842,
      "p95Cycles": 7842,
      "maximumCycles": 7842,
      "firstOffset": 194028,
      "lastOffset": 205380
    },
    {
      "name": "gemm_f8_init_large_rows_interleaved_k128_c80_r48_r49",
      "dimensions": {},
      "sampleCount": 864,
      "tileCount": 864,
      "phaseCount": 17,
      "phaseCycles": 11352,
      "phaseMs": 0.007568,
      "workCycles": 6801408,
      "averageActiveTiles": 599.1374207188161,
      "usefulCycles": 2893401.0,
      "estimatedWorkCycles": 6801408,
      "usefulUtilization": 0.4254120617378049,
      "workLaneOccupancy": 0.854296875,
      "deviceUsefulUtilization": 0.1731523678072433,
      "meanCycles": 7872.0,
      "p50Cycles": 7872,
      "p95Cycles": 7872,
      "maximumCycles": 7872,
      "firstOffset": 194358,
      "lastOffset": 205710
    },
    {
      "name": "attention_merge_v72_p80_k768_out16_f16",
      "dimensions": {},
      "sampleCount": 1472,
      "tileCount": 1472,
      "phaseCount": 26,
      "phaseCycles": 6366,
      "phaseMs": 0.004244,
      "workCycles": 4371840,
      "averageActiveTiles": 686.7483506126296,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 2970.0,
      "p50Cycles": 2970,
      "p95Cycles": 2970,
      "maximumCycles": 2970,
      "firstOffset": 211392,
      "lastOffset": 217758
    },
    {
      "name": "rearrange_row_major_to_amp_f16_o0_r8_p8_c72_p80",
      "dimensions": {},
      "sampleCount": 1360,
      "tileCount": 1360,
      "phaseCount": 14,
      "phaseCycles": 6330,
      "phaseMs": 0.00422,
      "workCycles": 1721760,
      "averageActiveTiles": 272.0,
      "usefulCycles": 391680.0,
      "estimatedWorkCycles": 1721760,
      "usefulUtilization": 0.22748815165876776,
      "workLaneOccupancy": 0.9,
      "deviceUsefulUtilization": 0.042035854110859264,
      "meanCycles": 1266.0,
      "p50Cycles": 1266,
      "p95Cycles": 1266,
      "maximumCycles": 1266,
      "firstOffset": 101520,
      "lastOffset": 107850
    },
    {
      "name": "rearrange_row_major_to_amp_f16_o2_r64_p64_c16_p16_b64x16",
      "dimensions": {},
      "sampleCount": 880,
      "tileCount": 880,
      "phaseCount": 13,
      "phaseCycles": 5064,
      "phaseMs": 0.003376,
      "workCycles": 2138400,
      "averageActiveTiles": 422.2748815165877,
      "usefulCycles": 450560.0,
      "estimatedWorkCycles": 2138400,
      "usefulUtilization": 0.21069958847736625,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.060443711793392405,
      "meanCycles": 2430.0,
      "p50Cycles": 2430,
      "p95Cycles": 2430,
      "maximumCycles": 2430,
      "firstOffset": 106740,
      "lastOffset": 111804
    },
    {
      "name": "rearrange_row_major_to_amp_f16_o2_r0_p64_c16_p16_b64x16",
      "dimensions": {},
      "sampleCount": 80,
      "tileCount": 80,
      "phaseCount": 8,
      "phaseCycles": 4470,
      "phaseMs": 0.00298,
      "workCycles": 156960,
      "averageActiveTiles": 35.11409395973154,
      "usefulCycles": 16000.0,
      "estimatedWorkCycles": 156960,
      "usefulUtilization": 0.1019367991845056,
      "workLaneOccupancy": 0.390625,
      "deviceUsefulUtilization": 0.0024316700710047662,
      "meanCycles": 1962.0,
      "p50Cycles": 1962,
      "p95Cycles": 1962,
      "maximumCycles": 1962,
      "firstOffset": 106878,
      "lastOffset": 111348
    },
    {
      "name": "rearrange_row_major_to_amp_f16_o1_r64_p64_c16_p16",
      "dimensions": {},
      "sampleCount": 880,
      "tileCount": 880,
      "phaseCount": 13,
      "phaseCycles": 4182,
      "phaseMs": 0.002788,
      "workCycles": 1351680,
      "averageActiveTiles": 323.2137733142037,
      "usefulCycles": 450560.0,
      "estimatedWorkCycles": 1351680,
      "usefulUtilization": 0.3333333333333333,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.07319152475412222,
      "meanCycles": 1536.0,
      "p50Cycles": 1536,
      "p95Cycles": 1536,
      "maximumCycles": 1536,
      "firstOffset": 105204,
      "lastOffset": 109386
    },
    {
      "name": "rearrange_row_major_to_amp_f16_o0_r7_p7_c72_p80",
      "dimensions": {},
      "sampleCount": 112,
      "tileCount": 112,
      "phaseCount": 2,
      "phaseCycles": 3336,
      "phaseMs": 0.002224,
      "workCycles": 141792,
      "averageActiveTiles": 42.50359712230216,
      "usefulCycles": 28224.0,
      "estimatedWorkCycles": 141792,
      "usefulUtilization": 0.1990521327014218,
      "workLaneOccupancy": 0.9,
      "deviceUsefulUtilization": 0.005747575852361589,
      "meanCycles": 1266.0,
      "p50Cycles": 1266,
      "p95Cycles": 1266,
      "maximumCycles": 1266,
      "firstOffset": 101652,
      "lastOffset": 104988
    }
  ],
  "samples": []
}
