{
  "clockHz": 1500000000,
  "tileCount": 1472,
  "sampleCount": 288886,
  "matchedSampleCount": 288886,
  "profileSpanCycles": 1624134,
  "profileSpanMs": 1.082756,
  "atOffset": null,
  "groups": [
    {
      "name": "repeat-remainder",
      "dimensions": {},
      "sampleCount": 1472,
      "tileCount": 1472,
      "phaseCount": 11,
      "phaseCycles": 661818,
      "phaseMs": 0.441212,
      "workCycles": 969784338,
      "averageActiveTiles": 1465.3338803115055,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 658820.8817934783,
      "p50Cycles": 658824,
      "p95Cycles": 658824,
      "maximumCycles": 658872,
      "firstOffset": 720366,
      "lastOffset": 1382184
    },
    {
      "name": "sync",
      "dimensions": {},
      "sampleCount": 20321,
      "tileCount": 1457,
      "phaseCount": 59,
      "phaseCycles": 649890,
      "phaseMs": 0.43326,
      "workCycles": 119299116,
      "averageActiveTiles": 183.5681669205558,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 5870.730574282762,
      "p50Cycles": 1908,
      "p95Cycles": 19680,
      "maximumCycles": 46848,
      "firstOffset": 0,
      "lastOffset": 1623402
    },
    {
      "name": "exchange",
      "dimensions": {},
      "sampleCount": 93023,
      "tileCount": 1472,
      "phaseCount": 77,
      "phaseCycles": 585726,
      "phaseMs": 0.390484,
      "workCycles": 627448512,
      "averageActiveTiles": 1071.2321324305221,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 6745.090052997646,
      "p50Cycles": 5562,
      "p95Cycles": 21084,
      "maximumCycles": 47124,
      "firstOffset": 0,
      "lastOffset": 1623984
    },
    {
      "name": "copy_u64",
      "dimensions": {},
      "sampleCount": 24752,
      "tileCount": 1458,
      "phaseCount": 248,
      "phaseCycles": 119526,
      "phaseMs": 0.079684,
      "workCycles": 22432314,
      "averageActiveTiles": 187.67727523718688,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 906.2828862314157,
      "p50Cycles": 558,
      "p95Cycles": 1584,
      "maximumCycles": 13158,
      "firstOffset": 38112,
      "lastOffset": 1620324
    },
    {
      "name": "cast_f16_f8",
      "dimensions": {},
      "sampleCount": 14184,
      "tileCount": 1458,
      "phaseCount": 183,
      "phaseCycles": 80826,
      "phaseMs": 0.053884,
      "workCycles": 81109782,
      "averageActiveTiles": 1003.5110236804989,
      "usefulCycles": 18436032.0,
      "estimatedWorkCycles": 81109782,
      "usefulUtilization": 0.2272972697670424,
      "workLaneOccupancy": 0.9851679998153214,
      "deviceUsefulUtilization": 0.15495605697262718,
      "meanCycles": 5718.399746192893,
      "p50Cycles": 6480,
      "p95Cycles": 11448,
      "maximumCycles": 11640,
      "firstOffset": 18312,
      "lastOffset": 1610676
    },
    {
      "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": 192696,
      "lastOffset": 429678
    },
    {
      "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": 185004,
      "lastOffset": 422232
    },
    {
      "name": "gemm_f16_init_small_rows_interleaved_k64_c80_r7_r8",
      "dimensions": {},
      "sampleCount": 1344,
      "tileCount": 112,
      "phaseCount": 26,
      "phaseCycles": 54528,
      "phaseMs": 0.036352,
      "workCycles": 6064128,
      "averageActiveTiles": 111.21126760563381,
      "usefulCycles": 642978.0,
      "estimatedWorkCycles": 6064128,
      "usefulUtilization": 0.10602975398936171,
      "workLaneOccupancy": 0.854296875,
      "deviceUsefulUtilization": 0.008010668033335885,
      "meanCycles": 4512.0,
      "p50Cycles": 4512,
      "p95Cycles": 4512,
      "maximumCycles": 4512,
      "firstOffset": 192678,
      "lastOffset": 429600
    },
    {
      "name": "gemm_f16_init_small_rows_interleaved_k80_c64_r7_r8",
      "dimensions": {},
      "sampleCount": 1344,
      "tileCount": 112,
      "phaseCount": 26,
      "phaseCycles": 54228,
      "phaseMs": 0.036152,
      "workCycles": 6031872,
      "averageActiveTiles": 111.23168842664306,
      "usefulCycles": 642978.0,
      "estimatedWorkCycles": 6031872,
      "usefulUtilization": 0.10659675802139038,
      "workLaneOccupancy": 0.854296875,
      "deviceUsefulUtilization": 0.008054984630112472,
      "meanCycles": 4488.0,
      "p50Cycles": 4488,
      "p95Cycles": 4488,
      "maximumCycles": 4488,
      "firstOffset": 185010,
      "lastOffset": 422184
    },
    {
      "name": "reduce_sum_f16",
      "dimensions": {},
      "sampleCount": 8968,
      "tileCount": 1458,
      "phaseCount": 146,
      "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": 38376,
      "lastOffset": 1620054
    },
    {
      "name": "copy_strided_u64",
      "dimensions": {},
      "sampleCount": 2366,
      "tileCount": 1117,
      "phaseCount": 109,
      "phaseCycles": 51978,
      "phaseMs": 0.034652,
      "workCycles": 1020774,
      "averageActiveTiles": 19.638577859863787,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 431.4344885883347,
      "p50Cycles": 384,
      "p95Cycles": 624,
      "maximumCycles": 7350,
      "firstOffset": 0,
      "lastOffset": 1616958
    },
    {
      "name": "attention_merge_v72_p80_k64_out32_f16",
      "dimensions": {},
      "sampleCount": 16192,
      "tileCount": 1472,
      "phaseCount": 69,
      "phaseCycles": 37890,
      "phaseMs": 0.02526,
      "workCycles": 54140112,
      "averageActiveTiles": 1428.8760095011876,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 3343.633399209486,
      "p50Cycles": 3432,
      "p95Cycles": 3432,
      "maximumCycles": 3432,
      "firstOffset": 197190,
      "lastOffset": 412158
    },
    {
      "name": "attention_softmax_d72_p64_full_f16",
      "dimensions": {},
      "sampleCount": 16192,
      "tileCount": 1472,
      "phaseCount": 69,
      "phaseCycles": 35808,
      "phaseMs": 0.023872,
      "workCycles": 51490560,
      "averageActiveTiles": 1437.9624664879357,
      "usefulCycles": 15524784.0,
      "estimatedWorkCycles": 51490560,
      "usefulUtilization": 0.30150738310090236,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.29453553007343514,
      "meanCycles": 3180.0,
      "p50Cycles": 3180,
      "p95Cycles": 3180,
      "maximumCycles": 3180,
      "firstOffset": 189498,
      "lastOffset": 404184
    },
    {
      "name": "fill_zero_u64",
      "dimensions": {},
      "sampleCount": 5060,
      "tileCount": 1472,
      "phaseCount": 71,
      "phaseCycles": 32568,
      "phaseMs": 0.021712,
      "workCycles": 1957392,
      "averageActiveTiles": 60.101694915254235,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 624672,
      "usefulUtilization": 0.0,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": 0.0,
      "meanCycles": 386.8363636363636,
      "p50Cycles": 300,
      "p95Cycles": 618,
      "maximumCycles": 3150,
      "firstOffset": 28932,
      "lastOffset": 1611294
    },
    {
      "name": "layer_norm_f16",
      "dimensions": {},
      "sampleCount": 2187,
      "tileCount": 729,
      "phaseCount": 30,
      "phaseCycles": 32358,
      "phaseMs": 0.021572,
      "workCycles": 20518434,
      "averageActiveTiles": 634.1069905432969,
      "usefulCycles": 7243344.0,
      "estimatedWorkCycles": 20518434,
      "usefulUtilization": 0.35301641441057346,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.15207213053958835,
      "meanCycles": 9382.0,
      "p50Cycles": 9384,
      "p95Cycles": 9384,
      "maximumCycles": 9384,
      "firstOffset": 57156,
      "lastOffset": 1397286
    },
    {
      "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": 547434,
      "lastOffset": 579498
    },
    {
      "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": 547488,
      "lastOffset": 579006
    },
    {
      "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": 546834,
      "lastOffset": 575652
    },
    {
      "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": 664146,
      "lastOffset": 692508
    },
    {
      "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": 546912,
      "lastOffset": 575196
    }
  ],
  "samples": []
}
