{
  "clockHz": 1500000000,
  "tileCount": 1472,
  "sampleCount": 302134,
  "matchedSampleCount": 302134,
  "profileSpanCycles": 1627302,
  "profileSpanMs": 1.084868,
  "atOffset": null,
  "groups": [
    {
      "name": "repeat-remainder",
      "dimensions": {},
      "sampleCount": 1472,
      "tileCount": 1472,
      "phaseCount": 11,
      "phaseCycles": 664278,
      "phaseMs": 0.442852,
      "workCycles": 973402506,
      "averageActiveTiles": 1465.3541228220715,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 661278.8763586957,
      "p50Cycles": 661284,
      "p95Cycles": 661284,
      "maximumCycles": 661326,
      "firstOffset": 722886,
      "lastOffset": 1387164
    },
    {
      "name": "sync",
      "dimensions": {},
      "sampleCount": 27452,
      "tileCount": 1457,
      "phaseCount": 68,
      "phaseCycles": 655110,
      "phaseMs": 0.43674,
      "workCycles": 161233680,
      "averageActiveTiles": 246.1169574575262,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 5873.294477633688,
      "p50Cycles": 2202,
      "p95Cycles": 19674,
      "maximumCycles": 46848,
      "firstOffset": 0,
      "lastOffset": 1626576
    },
    {
      "name": "exchange",
      "dimensions": {},
      "sampleCount": 99140,
      "tileCount": 1472,
      "phaseCount": 86,
      "phaseCycles": 545754,
      "phaseMs": 0.363836,
      "workCycles": 586560120,
      "averageActiveTiles": 1074.7701711760244,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 5916.482953399233,
      "p50Cycles": 5238,
      "p95Cycles": 16584,
      "maximumCycles": 47124,
      "firstOffset": 0,
      "lastOffset": 1627152
    },
    {
      "name": "copy_u64",
      "dimensions": {},
      "sampleCount": 24752,
      "tileCount": 1458,
      "phaseCount": 257,
      "phaseCycles": 120408,
      "phaseMs": 0.080272,
      "workCycles": 22432314,
      "averageActiveTiles": 186.3025214271477,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 906.2828862314157,
      "p50Cycles": 558,
      "p95Cycles": 1584,
      "maximumCycles": 13158,
      "firstOffset": 38106,
      "lastOffset": 1623438
    },
    {
      "name": "cast_f16_f8",
      "dimensions": {},
      "sampleCount": 14184,
      "tileCount": 1458,
      "phaseCount": 185,
      "phaseCycles": 70680,
      "phaseMs": 0.04712,
      "workCycles": 81114624,
      "averageActiveTiles": 1147.6319185059422,
      "usefulCycles": 18436032.0,
      "estimatedWorkCycles": 81114624,
      "usefulUtilization": 0.22728370164176562,
      "workLaneOccupancy": 0.9851679998153214,
      "deviceUsefulUtilization": 0.17719974902192367,
      "meanCycles": 5718.741116751269,
      "p50Cycles": 6498,
      "p95Cycles": 11448,
      "maximumCycles": 11640,
      "firstOffset": 18306,
      "lastOffset": 1613790
    },
    {
      "name": "gemm_f16_init_large_rows_interleaved_k64_c80_r7_r8",
      "dimensions": {},
      "sampleCount": 16320,
      "tileCount": 1360,
      "phaseCount": 72,
      "phaseCycles": 55350,
      "phaseMs": 0.0369,
      "workCycles": 74125440,
      "averageActiveTiles": 1339.2130081300813,
      "usefulCycles": 8922960.0,
      "estimatedWorkCycles": 74125440,
      "usefulUtilization": 0.1203764861294584,
      "workLaneOccupancy": 0.854296875,
      "deviceUsefulUtilization": 0.10951749734888654,
      "meanCycles": 4542.0,
      "p50Cycles": 4542,
      "p95Cycles": 4542,
      "maximumCycles": 4542,
      "firstOffset": 191748,
      "lastOffset": 428730
    },
    {
      "name": "gemm_f16_init_large_rows_interleaved_k80_c64_r7_r8",
      "dimensions": {},
      "sampleCount": 16320,
      "tileCount": 1360,
      "phaseCount": 72,
      "phaseCycles": 54984,
      "phaseMs": 0.036656,
      "workCycles": 73635840,
      "averageActiveTiles": 1339.2230467044958,
      "usefulCycles": 8922960.0,
      "estimatedWorkCycles": 73635840,
      "usefulUtilization": 0.12117686170212766,
      "workLaneOccupancy": 0.854296875,
      "deviceUsefulUtilization": 0.11024649858614996,
      "meanCycles": 4512.0,
      "p50Cycles": 4512,
      "p95Cycles": 4512,
      "maximumCycles": 4512,
      "firstOffset": 184056,
      "lastOffset": 421284
    },
    {
      "name": "gemm_f16_init_small_rows_interleaved_k64_c80_r7_r8",
      "dimensions": {},
      "sampleCount": 1344,
      "tileCount": 112,
      "phaseCount": 26,
      "phaseCycles": 54522,
      "phaseMs": 0.036348,
      "workCycles": 6064128,
      "averageActiveTiles": 111.22350610762628,
      "usefulCycles": 642978.0,
      "estimatedWorkCycles": 6064128,
      "usefulUtilization": 0.10602975398936171,
      "workLaneOccupancy": 0.854296875,
      "deviceUsefulUtilization": 0.00801154958588715,
      "meanCycles": 4512.0,
      "p50Cycles": 4512,
      "p95Cycles": 4512,
      "maximumCycles": 4512,
      "firstOffset": 191736,
      "lastOffset": 428652
    },
    {
      "name": "gemm_f16_init_small_rows_interleaved_k80_c64_r7_r8",
      "dimensions": {},
      "sampleCount": 1344,
      "tileCount": 112,
      "phaseCount": 26,
      "phaseCycles": 54222,
      "phaseMs": 0.036148,
      "workCycles": 6031872,
      "averageActiveTiles": 111.24399690162664,
      "usefulCycles": 642978.0,
      "estimatedWorkCycles": 6031872,
      "usefulUtilization": 0.10659675802139038,
      "workLaneOccupancy": 0.854296875,
      "deviceUsefulUtilization": 0.008055875964031926,
      "meanCycles": 4488.0,
      "p50Cycles": 4488,
      "p95Cycles": 4488,
      "maximumCycles": 4488,
      "firstOffset": 184068,
      "lastOffset": 421236
    },
    {
      "name": "reduce_sum_f16",
      "dimensions": {},
      "sampleCount": 8968,
      "tileCount": 1458,
      "phaseCount": 150,
      "phaseCycles": 52434,
      "phaseMs": 0.034956,
      "workCycles": 39164064,
      "averageActiveTiles": 746.9211580272342,
      "usefulCycles": 22113888.0,
      "estimatedWorkCycles": 39164064,
      "usefulUtilization": 0.5646474278052451,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.28651298278083753,
      "meanCycles": 4367.090098126672,
      "p50Cycles": 4068,
      "p95Cycles": 7218,
      "maximumCycles": 7218,
      "firstOffset": 38370,
      "lastOffset": 1623168
    },
    {
      "name": "copy_strided_u64",
      "dimensions": {},
      "sampleCount": 2366,
      "tileCount": 1117,
      "phaseCount": 107,
      "phaseCycles": 49890,
      "phaseMs": 0.03326,
      "workCycles": 1020732,
      "averageActiveTiles": 20.459651232711966,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 431.4167371090448,
      "p50Cycles": 384,
      "p95Cycles": 624,
      "maximumCycles": 7350,
      "firstOffset": 0,
      "lastOffset": 1620072
    },
    {
      "name": "attention_merge_v72_p80_k64_out32_f16",
      "dimensions": {},
      "sampleCount": 16192,
      "tileCount": 1472,
      "phaseCount": 69,
      "phaseCycles": 37884,
      "phaseMs": 0.025256,
      "workCycles": 54140112,
      "averageActiveTiles": 1429.1023123218245,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 3343.633399209486,
      "p50Cycles": 3432,
      "p95Cycles": 3432,
      "maximumCycles": 3432,
      "firstOffset": 196248,
      "lastOffset": 411210
    },
    {
      "name": "attention_softmax_d72_p64_full_f16",
      "dimensions": {},
      "sampleCount": 16192,
      "tileCount": 1472,
      "phaseCount": 69,
      "phaseCycles": 35802,
      "phaseMs": 0.023868,
      "workCycles": 51490560,
      "averageActiveTiles": 1438.2034523210993,
      "usefulCycles": 15524784.0,
      "estimatedWorkCycles": 51490560,
      "usefulUtilization": 0.30150738310090236,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.2945848908125123,
      "meanCycles": 3180.0,
      "p50Cycles": 3180,
      "p95Cycles": 3180,
      "maximumCycles": 3180,
      "firstOffset": 188556,
      "lastOffset": 403236
    },
    {
      "name": "gemm_f8_init_large_rows_interleaved_k192_c160_r80_r82",
      "dimensions": {},
      "sampleCount": 780,
      "tileCount": 780,
      "phaseCount": 23,
      "phaseCycles": 32064,
      "phaseMs": 0.021376,
      "workCycles": 23797800,
      "averageActiveTiles": 742.1968562874251,
      "usefulCycles": 15350400.0,
      "estimatedWorkCycles": 23797800,
      "usefulUtilization": 0.6450344149459194,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.3252326868003124,
      "meanCycles": 30510.0,
      "p50Cycles": 30510,
      "p95Cycles": 30510,
      "maximumCycles": 30510,
      "firstOffset": 546192,
      "lastOffset": 578256
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k192_c160_r80_r82",
      "dimensions": {},
      "sampleCount": 624,
      "tileCount": 624,
      "phaseCount": 14,
      "phaseCycles": 31518,
      "phaseMs": 0.021012,
      "workCycles": 18720000,
      "averageActiveTiles": 593.9463163906339,
      "usefulCycles": 11943360.0,
      "estimatedWorkCycles": 18720000,
      "usefulUtilization": 0.638,
      "workLaneOccupancy": 0.996875,
      "deviceUsefulUtilization": 0.25743053658778836,
      "meanCycles": 30000.0,
      "p50Cycles": 30000,
      "p95Cycles": 30000,
      "maximumCycles": 30000,
      "firstOffset": 546246,
      "lastOffset": 577764
    },
    {
      "name": "layer_norm_f16",
      "dimensions": {},
      "sampleCount": 2187,
      "tileCount": 729,
      "phaseCount": 30,
      "phaseCycles": 30720,
      "phaseMs": 0.02048,
      "workCycles": 20518434,
      "averageActiveTiles": 667.9177734375,
      "usefulCycles": 7243344.0,
      "estimatedWorkCycles": 20518434,
      "usefulUtilization": 0.35301641441057346,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.1601806640625,
      "meanCycles": 9382.0,
      "p50Cycles": 9384,
      "p95Cycles": 9384,
      "maximumCycles": 9384,
      "firstOffset": 57150,
      "lastOffset": 1402266
    },
    {
      "name": "gemm_f8_init_large_rows_interleaved_k192_c144_r80_r82",
      "dimensions": {},
      "sampleCount": 30,
      "tileCount": 30,
      "phaseCount": 9,
      "phaseCycles": 28818,
      "phaseMs": 0.019212,
      "workCycles": 824940,
      "averageActiveTiles": 28.625858838226108,
      "usefulCycles": 531360.0,
      "estimatedWorkCycles": 824940,
      "usefulUtilization": 0.6441195723325333,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.012526138554707656,
      "meanCycles": 27498.0,
      "p50Cycles": 27498,
      "p95Cycles": 27498,
      "maximumCycles": 27498,
      "firstOffset": 545592,
      "lastOffset": 574410
    },
    {
      "name": "fill_zero_u64",
      "dimensions": {},
      "sampleCount": 5060,
      "tileCount": 1472,
      "phaseCount": 71,
      "phaseCycles": 28656,
      "phaseMs": 0.019104,
      "workCycles": 1957392,
      "averageActiveTiles": 68.30653266331659,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 624672,
      "usefulUtilization": 0.0,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": 0.0,
      "meanCycles": 386.8363636363636,
      "p50Cycles": 300,
      "p95Cycles": 618,
      "maximumCycles": 3150,
      "firstOffset": 28926,
      "lastOffset": 1614408
    },
    {
      "name": "gemm_f8_init_large_rows_interleaved_k160_c128_r120_r122",
      "dimensions": {},
      "sampleCount": 1215,
      "tileCount": 1215,
      "phaseCount": 31,
      "phaseCycles": 28362,
      "phaseMs": 0.018908,
      "workCycles": 32666490,
      "averageActiveTiles": 1151.769621324307,
      "usefulCycles": 23628960.0,
      "estimatedWorkCycles": 32666490,
      "usefulUtilization": 0.7233394221417728,
      "workLaneOccupancy": 0.9962962962962963,
      "deviceUsefulUtilization": 0.5659785138105794,
      "meanCycles": 26886.0,
      "p50Cycles": 26886,
      "p95Cycles": 26886,
      "maximumCycles": 26886,
      "firstOffset": 666126,
      "lastOffset": 694488
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k192_c144_r80_r82",
      "dimensions": {},
      "sampleCount": 24,
      "tileCount": 24,
      "phaseCount": 9,
      "phaseCycles": 28284,
      "phaseMs": 0.018856,
      "workCycles": 649008,
      "averageActiveTiles": 22.946117946542216,
      "usefulCycles": 413424.0,
      "estimatedWorkCycles": 649008,
      "usefulUtilization": 0.637009096960284,
      "workLaneOccupancy": 0.996875,
      "deviceUsefulUtilization": 0.009929949641216728,
      "meanCycles": 27042.0,
      "p50Cycles": 27042,
      "p95Cycles": 27042,
      "maximumCycles": 27042,
      "firstOffset": 545670,
      "lastOffset": 573954
    }
  ],
  "samples": []
}
