{
  "clockHz": 1500000000,
  "tileCount": 1472,
  "sampleCount": 67787,
  "matchedSampleCount": 67787,
  "profileSpanCycles": 112566,
  "profileSpanMs": 0.075044,
  "atOffset": null,
  "groups": [
    {
      "name": "idle",
      "dimensions": {},
      "sampleCount": 61952,
      "tileCount": 1408,
      "phaseCount": 44,
      "phaseCycles": 111468,
      "phaseMs": 0.074312,
      "workCycles": 156907812,
      "averageActiveTiles": 1407.6489396059856,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 2532.731986053719,
      "p50Cycles": 1764,
      "p95Cycles": 4014,
      "maximumCycles": 16560,
      "firstOffset": 0,
      "lastOffset": 111468
    },
    {
      "name": "sync",
      "dimensions": {},
      "sampleCount": 551,
      "tileCount": 60,
      "phaseCount": 28,
      "phaseCycles": 83466,
      "phaseMs": 0.055644,
      "workCycles": 2684058,
      "averageActiveTiles": 32.15750125799727,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 4871.248638838476,
      "p50Cycles": 2082,
      "p95Cycles": 16248,
      "maximumCycles": 16482,
      "firstOffset": 0,
      "lastOffset": 111432
    },
    {
      "name": "exchange",
      "dimensions": {},
      "sampleCount": 2265,
      "tileCount": 64,
      "phaseCount": 44,
      "phaseCycles": 61308,
      "phaseMs": 0.040872,
      "workCycles": 2458470,
      "averageActiveTiles": 40.10031317283226,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 1085.4172185430464,
      "p50Cycles": 744,
      "p95Cycles": 2448,
      "maximumCycles": 4638,
      "firstOffset": 1452,
      "lastOffset": 111534
    },
    {
      "name": "flash_attention_online_f16_m1_q4_k4_d72_v72_3df15bf0",
      "dimensions": {},
      "sampleCount": 4,
      "tileCount": 4,
      "phaseCount": 3,
      "phaseCycles": 14316,
      "phaseMs": 0.009544,
      "workCycles": 53928,
      "averageActiveTiles": 3.7669740150880133,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 13482.0,
      "p50Cycles": 13482,
      "p95Cycles": 13482,
      "maximumCycles": 13482,
      "firstOffset": 17928,
      "lastOffset": 32244
    },
    {
      "name": "flash_attention_online_f16_m1_q1_k4_d72_v72_3df15bf0",
      "dimensions": {},
      "sampleCount": 4,
      "tileCount": 4,
      "phaseCount": 4,
      "phaseCycles": 14052,
      "phaseMs": 0.009368,
      "workCycles": 53760,
      "averageActiveTiles": 3.8257899231426133,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 13440.0,
      "p50Cycles": 13440,
      "p95Cycles": 13440,
      "maximumCycles": 13440,
      "firstOffset": 74880,
      "lastOffset": 88932
    },
    {
      "name": "copy_u64",
      "dimensions": {},
      "sampleCount": 571,
      "tileCount": 64,
      "phaseCount": 119,
      "phaseCycles": 10758,
      "phaseMs": 0.007172,
      "workCycles": 162834,
      "averageActiveTiles": 15.136084774121583,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 285.17338003502624,
      "p50Cycles": 282,
      "p95Cycles": 318,
      "maximumCycles": 318,
      "firstOffset": 2184,
      "lastOffset": 110982
    },
    {
      "name": "static_copy_u32",
      "dimensions": {},
      "sampleCount": 371,
      "tileCount": 64,
      "phaseCount": 100,
      "phaseCycles": 8892,
      "phaseMs": 0.005928,
      "workCycles": 87600,
      "averageActiveTiles": 9.851551956815115,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 236.11859838274933,
      "p50Cycles": 240,
      "p95Cycles": 336,
      "maximumCycles": 816,
      "firstOffset": 3516,
      "lastOffset": 111762
    },
    {
      "name": "add_f16",
      "dimensions": {},
      "sampleCount": 608,
      "tileCount": 64,
      "phaseCount": 109,
      "phaseCycles": 7266,
      "phaseMs": 0.004844,
      "workCycles": 285888,
      "averageActiveTiles": 39.34599504541701,
      "usefulCycles": 3600.0,
      "estimatedWorkCycles": 285888,
      "usefulUtilization": 0.012592343854936199,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.000336588518292464,
      "meanCycles": 470.2105263157895,
      "p50Cycles": 456,
      "p95Cycles": 618,
      "maximumCycles": 618,
      "firstOffset": 4428,
      "lastOffset": 112566
    },
    {
      "name": "reduce_sum_f16",
      "dimensions": {},
      "sampleCount": 160,
      "tileCount": 54,
      "phaseCount": 64,
      "phaseCycles": 6288,
      "phaseMs": 0.004192,
      "workCycles": 83622,
      "averageActiveTiles": 13.298664122137405,
      "usefulCycles": 10512.0,
      "estimatedWorkCycles": 83622,
      "usefulUtilization": 0.12570854559804837,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.0011357036176568205,
      "meanCycles": 522.6375,
      "p50Cycles": 510,
      "p95Cycles": 696,
      "maximumCycles": 768,
      "firstOffset": 2502,
      "lastOffset": 110700
    },
    {
      "name": "layer_norm_moments",
      "dimensions": {},
      "sampleCount": 72,
      "tileCount": 32,
      "phaseCount": 27,
      "phaseCycles": 4074,
      "phaseMs": 0.002716,
      "workCycles": 80352,
      "averageActiveTiles": 19.723122238586157,
      "usefulCycles": 5184.0,
      "estimatedWorkCycles": 80352,
      "usefulUtilization": 0.06451612903225806,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.0008644425946084395,
      "meanCycles": 1116.0,
      "p50Cycles": 1116,
      "p95Cycles": 1116,
      "maximumCycles": 1116,
      "firstOffset": 42174,
      "lastOffset": 97800
    }
  ],
  "samples": []
}
