{
  "clockHz": 1500000000,
  "tileCount": 1472,
  "sampleCount": 54625,
  "matchedSampleCount": 52681,
  "profileSpanCycles": 222630,
  "profileSpanMs": 0.14842,
  "atOffset": null,
  "groups": [
    {
      "name": "exchange",
      "dimensions": {},
      "sampleCount": 14421,
      "tileCount": 1472,
      "phaseCount": 11,
      "phaseCycles": 146886,
      "phaseMs": 0.097924,
      "workCycles": 132313728,
      "averageActiveTiles": 900.7919611126996,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 9175.073018514666,
      "p50Cycles": 9156,
      "p95Cycles": 15786,
      "maximumCycles": 33762,
      "firstOffset": 0,
      "lastOffset": 222012
    },
    {
      "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": 5173,
      "tileCount": 1472,
      "phaseCount": 46,
      "phaseCycles": 36786,
      "phaseMs": 0.024524,
      "workCycles": 26296410,
      "averageActiveTiles": 714.8483118577719,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 5083.39648173207,
      "p50Cycles": 714,
      "p95Cycles": 16620,
      "maximumCycles": 16620,
      "firstOffset": 0,
      "lastOffset": 216330
    },
    {
      "name": "fill_zero_u64",
      "dimensions": {},
      "sampleCount": 9920,
      "tileCount": 1472,
      "phaseCount": 45,
      "phaseCycles": 35178,
      "phaseMs": 0.023452,
      "workCycles": 5211936,
      "averageActiveTiles": 148.15896298823128,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 2016576,
      "usefulUtilization": 0.0,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": 0.0,
      "meanCycles": 525.3967741935484,
      "p50Cycles": 360,
      "p95Cycles": 1404,
      "maximumCycles": 1404,
      "firstOffset": 94068,
      "lastOffset": 215406
    },
    {
      "name": "copy_u64",
      "dimensions": {},
      "sampleCount": 5756,
      "tileCount": 1472,
      "phaseCount": 47,
      "phaseCycles": 27300,
      "phaseMs": 0.0182,
      "workCycles": 6541716,
      "averageActiveTiles": 239.62329670329672,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 1136.5038220986796,
      "p50Cycles": 774,
      "p95Cycles": 3702,
      "maximumCycles": 3702,
      "firstOffset": 0,
      "lastOffset": 222630
    },
    {
      "name": "attention_softmax_d72_p768_tail_f16",
      "dimensions": {},
      "sampleCount": 1472,
      "tileCount": 1472,
      "phaseCount": 18,
      "phaseCycles": 25656,
      "phaseMs": 0.017104,
      "workCycles": 34884384,
      "averageActiveTiles": 1359.6969130028065,
      "usefulCycles": 15816384.0,
      "estimatedWorkCycles": 34884384,
      "usefulUtilization": 0.4533943898794372,
      "workLaneOccupancy": 0.9970588235294118,
      "deviceUsefulUtilization": 0.4188036360678407,
      "meanCycles": 23698.630434782608,
      "p50Cycles": 23700,
      "p95Cycles": 23700,
      "maximumCycles": 23700,
      "firstOffset": 153510,
      "lastOffset": 179166
    },
    {
      "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": "cast_f16_f8",
      "dimensions": {},
      "sampleCount": 2880,
      "tileCount": 1440,
      "phaseCount": 18,
      "phaseCycles": 19572,
      "phaseMs": 0.013048,
      "workCycles": 14530272,
      "averageActiveTiles": 742.4009809932556,
      "usefulCycles": 3044304.0,
      "estimatedWorkCycles": 14530272,
      "usefulUtilization": 0.2095145913304307,
      "workLaneOccupancy": 0.9436822997262231,
      "deviceUsefulUtilization": 0.10566836829898968,
      "meanCycles": 5045.233333333334,
      "p50Cycles": 4020,
      "p95Cycles": 6654,
      "maximumCycles": 7626,
      "firstOffset": 121272,
      "lastOffset": 140844
    },
    {
      "name": "gemm_f16_init_small_rows_interleaved_k96_c80_r66_r67",
      "dimensions": {},
      "sampleCount": 1024,
      "tileCount": 1024,
      "phaseCount": 12,
      "phaseCycles": 17286,
      "phaseMs": 0.011524,
      "workCycles": 13996032,
      "averageActiveTiles": 809.6744186046511,
      "usefulCycles": 6928416.0,
      "estimatedWorkCycles": 13996032,
      "usefulUtilization": 0.49502716198419666,
      "workLaneOccupancy": 0.854296875,
      "deviceUsefulUtilization": 0.27228996574257125,
      "meanCycles": 13668.0,
      "p50Cycles": 13668,
      "p95Cycles": 13668,
      "maximumCycles": 13668,
      "firstOffset": 189216,
      "lastOffset": 206502
    },
    {
      "name": "gemm_f16_init_large_rows_interleaved_k96_c80_r66_r67",
      "dimensions": {},
      "sampleCount": 384,
      "tileCount": 384,
      "phaseCount": 13,
      "phaseCycles": 16926,
      "phaseMs": 0.011284,
      "workCycles": 5260032,
      "averageActiveTiles": 310.76639489542714,
      "usefulCycles": 2637522.0,
      "estimatedWorkCycles": 5260032,
      "usefulUtilization": 0.5014269875164258,
      "workLaneOccupancy": 0.854296875,
      "deviceUsefulUtilization": 0.10586050082456114,
      "meanCycles": 13698.0,
      "p50Cycles": 13698,
      "p95Cycles": 13698,
      "maximumCycles": 13698,
      "firstOffset": 189570,
      "lastOffset": 206496
    },
    {
      "name": "gemm_f8_init_large_rows_interleaved_k96_c48_r145_r146",
      "dimensions": {},
      "sampleCount": 640,
      "tileCount": 640,
      "phaseCount": 14,
      "phaseCycles": 15534,
      "phaseMs": 0.010356,
      "workCycles": 4634880,
      "averageActiveTiles": 298.37002703746623,
      "usefulCycles": 2522880.0,
      "estimatedWorkCycles": 4634880,
      "usefulUtilization": 0.5443247721623861,
      "workLaneOccupancy": 0.75,
      "deviceUsefulUtilization": 0.11033301425764522,
      "meanCycles": 7242.0,
      "p50Cycles": 7242,
      "p95Cycles": 7242,
      "maximumCycles": 7242,
      "firstOffset": 132552,
      "lastOffset": 148086
    },
    {
      "name": "reduce_sum_f16",
      "dimensions": {},
      "sampleCount": 1958,
      "tileCount": 1472,
      "phaseCount": 27,
      "phaseCycles": 14742,
      "phaseMs": 0.009828,
      "workCycles": 8392596,
      "averageActiveTiles": 569.2983312983313,
      "usefulCycles": 2894400.0,
      "estimatedWorkCycles": 8392596,
      "usefulUtilization": 0.34487541161280727,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.13338111164198121,
      "meanCycles": 4286.3105209397345,
      "p50Cycles": 2382,
      "p95Cycles": 10182,
      "maximumCycles": 10182,
      "firstOffset": 71664,
      "lastOffset": 213558
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k96_c48_r145_r146",
      "dimensions": {},
      "sampleCount": 160,
      "tileCount": 160,
      "phaseCount": 2,
      "phaseCycles": 13956,
      "phaseMs": 0.009304,
      "workCycles": 1156800,
      "averageActiveTiles": 82.88907996560619,
      "usefulCycles": 626400.0,
      "estimatedWorkCycles": 1156800,
      "usefulUtilization": 0.5414937759336099,
      "workLaneOccupancy": 0.75,
      "deviceUsefulUtilization": 0.030491794085760214,
      "meanCycles": 7230.0,
      "p50Cycles": 7230,
      "p95Cycles": 7230,
      "maximumCycles": 7230,
      "firstOffset": 132558,
      "lastOffset": 146514
    },
    {
      "name": "gemm_f8_init_large_rows_interleaved_k96_c32_r145_r146",
      "dimensions": {},
      "sampleCount": 512,
      "tileCount": 512,
      "phaseCount": 8,
      "phaseCycles": 12552,
      "phaseMs": 0.008368,
      "workCycles": 2537472,
      "averageActiveTiles": 202.1567877629063,
      "usefulCycles": 1308744.0,
      "estimatedWorkCycles": 2537472,
      "usefulUtilization": 0.5157668734866828,
      "workLaneOccupancy": 0.7294921875,
      "deviceUsefulUtilization": 0.07083272715936487,
      "meanCycles": 4956.0,
      "p50Cycles": 4956,
      "p95Cycles": 4956,
      "maximumCycles": 4956,
      "firstOffset": 130512,
      "lastOffset": 143064
    },
    {
      "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": "attention_merge_v72_p80_k768_out16_f16",
      "dimensions": {},
      "sampleCount": 1472,
      "tileCount": 1472,
      "phaseCount": 22,
      "phaseCycles": 6096,
      "phaseMs": 0.004064,
      "workCycles": 4371840,
      "averageActiveTiles": 717.1653543307086,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 2970.0,
      "p50Cycles": 2970,
      "p95Cycles": 2970,
      "maximumCycles": 2970,
      "firstOffset": 213204,
      "lastOffset": 219300
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k96_c32_r145_r146",
      "dimensions": {},
      "sampleCount": 128,
      "tileCount": 128,
      "phaseCount": 1,
      "phaseCycles": 5376,
      "phaseMs": 0.003584,
      "workCycles": 633600,
      "averageActiveTiles": 117.85714285714286,
      "usefulCycles": 324945.0,
      "estimatedWorkCycles": 633600,
      "usefulUtilization": 0.5128551136363636,
      "workLaneOccupancy": 0.7294921875,
      "deviceUsefulUtilization": 0.0410622543429736,
      "meanCycles": 4950.0,
      "p50Cycles": 4950,
      "p95Cycles": 4950,
      "maximumCycles": 4950,
      "firstOffset": 130530,
      "lastOffset": 135906
    },
    {
      "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
    }
  ],
  "samples": []
}
