{
  "clockHz": 1500000000,
  "tileCount": 1472,
  "sampleCount": 228580,
  "matchedSampleCount": 124508,
  "profileSpanCycles": 16065444,
  "profileSpanMs": 10.710296,
  "atOffset": null,
  "groups": [
    {
      "name": "repeat-remainder",
      "dimensions": {},
      "sampleCount": 1472,
      "tileCount": 1472,
      "phaseCount": 9,
      "phaseCycles": 15171816,
      "phaseMs": 10.114544,
      "workCycles": 22327282344,
      "averageActiveTiles": 1471.6288639408756,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 15167990.722826088,
      "p50Cycles": 15168018,
      "p95Cycles": 15168090,
      "maximumCycles": 15168114,
      "firstOffset": 688692,
      "lastOffset": 15860508
    },
    {
      "name": "reduce_sum_f16",
      "dimensions": {},
      "sampleCount": 22098,
      "tileCount": 1472,
      "phaseCount": 134,
      "phaseCycles": 122358,
      "phaseMs": 0.081572,
      "workCycles": 86301906,
      "averageActiveTiles": 705.3229539547884,
      "usefulCycles": 30933608.0,
      "estimatedWorkCycles": 86301906,
      "usefulUtilization": 0.35843481834572694,
      "workLaneOccupancy": 0.9908698136754196,
      "deviceUsefulUtilization": 0.1717474897254457,
      "meanCycles": 3905.417051316861,
      "p50Cycles": 2154,
      "p95Cycles": 8748,
      "maximumCycles": 10386,
      "firstOffset": 24600,
      "lastOffset": 16060602
    },
    {
      "name": "fill_zero_u64",
      "dimensions": {},
      "sampleCount": 26672,
      "tileCount": 1472,
      "phaseCount": 185,
      "phaseCycles": 79092,
      "phaseMs": 0.052728,
      "workCycles": 8950038,
      "averageActiveTiles": 113.15983917463207,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 4952022,
      "usefulUtilization": 0.0,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": 0.0,
      "meanCycles": 335.5593131373725,
      "p50Cycles": 300,
      "p95Cycles": 726,
      "maximumCycles": 3150,
      "firstOffset": 3042,
      "lastOffset": 16053060
    },
    {
      "name": "cast_f16_f8",
      "dimensions": {},
      "sampleCount": 14715,
      "tileCount": 1472,
      "phaseCount": 138,
      "phaseCycles": 60012,
      "phaseMs": 0.040008,
      "workCycles": 46839540,
      "averageActiveTiles": 780.502899420116,
      "usefulCycles": 2390372.0,
      "estimatedWorkCycles": 44662836,
      "usefulUtilization": 0.05352038101655703,
      "workLaneOccupancy": 0.8987252870209675,
      "deviceUsefulUtilization": 0.027059488464625914,
      "meanCycles": 3183.1151885830786,
      "p50Cycles": 2832,
      "p95Cycles": 9300,
      "maximumCycles": 9300,
      "firstOffset": 0,
      "lastOffset": 16052214
    },
    {
      "name": "gemm_f8_init_large_rows_interleaved_k192_c144_r180_r184",
      "dimensions": {},
      "sampleCount": 920,
      "tileCount": 920,
      "phaseCount": 22,
      "phaseCycles": 57882,
      "phaseMs": 0.038588,
      "workCycles": 44739600,
      "averageActiveTiles": 772.944956981445,
      "usefulCycles": 35637120.0,
      "estimatedWorkCycles": 44739600,
      "usefulUtilization": 0.7965453423812462,
      "workLaneOccupancy": 0.9746376811594203,
      "deviceUsefulUtilization": 0.418264745516741,
      "meanCycles": 48630.0,
      "p50Cycles": 48630,
      "p95Cycles": 48630,
      "maximumCycles": 48630,
      "firstOffset": 599718,
      "lastOffset": 657600
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k192_c144_r180_r184",
      "dimensions": {},
      "sampleCount": 552,
      "tileCount": 552,
      "phaseCount": 23,
      "phaseCycles": 55308,
      "phaseMs": 0.036872,
      "workCycles": 26515872,
      "averageActiveTiles": 479.4220004339336,
      "usefulCycles": 20839968.0,
      "estimatedWorkCycles": 26515872,
      "usefulUtilization": 0.7859431513321531,
      "workLaneOccupancy": 0.971027911969941,
      "deviceUsefulUtilization": 0.2559771996188931,
      "meanCycles": 48036.0,
      "p50Cycles": 48036,
      "p95Cycles": 48036,
      "maximumCycles": 48036,
      "firstOffset": 601620,
      "lastOffset": 656928
    },
    {
      "name": "attention_softmax_d72_p768_tail_f8_sm4",
      "dimensions": {},
      "sampleCount": 1472,
      "tileCount": 1472,
      "phaseCount": 16,
      "phaseCycles": 53526,
      "phaseMs": 0.035684,
      "workCycles": 71468520,
      "averageActiveTiles": 1335.2112991817062,
      "usefulCycles": 33499008.0,
      "estimatedWorkCycles": 71468520,
      "usefulUtilization": 0.4687239640613797,
      "workLaneOccupancy": 0.9972222222222222,
      "deviceUsefulUtilization": 0.4251668023179308,
      "meanCycles": 48551.983695652176,
      "p50Cycles": 48552,
      "p95Cycles": 48552,
      "maximumCycles": 48552,
      "firstOffset": 264744,
      "lastOffset": 318270
    },
    {
      "name": "copy_u64",
      "dimensions": {},
      "sampleCount": 10040,
      "tileCount": 1472,
      "phaseCount": 104,
      "phaseCycles": 51996,
      "phaseMs": 0.034664,
      "workCycles": 14454576,
      "averageActiveTiles": 277.993999538426,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 1439.6988047808766,
      "p50Cycles": 834,
      "p95Cycles": 3018,
      "maximumCycles": 10578,
      "firstOffset": 94680,
      "lastOffset": 16055244
    },
    {
      "name": "gemm_f8_init_large_rows_interleaved_k384_c80_r160_r164",
      "dimensions": {},
      "sampleCount": 795,
      "tileCount": 795,
      "phaseCount": 13,
      "phaseCycles": 49200,
      "phaseMs": 0.0328,
      "workCycles": 38951820,
      "averageActiveTiles": 791.7036585365854,
      "usefulCycles": 31291200.0,
      "estimatedWorkCycles": 38951820,
      "usefulUtilization": 0.8033308841538085,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.4320652173913043,
      "meanCycles": 48996.0,
      "p50Cycles": 48996,
      "p95Cycles": 48996,
      "maximumCycles": 48996,
      "firstOffset": 482250,
      "lastOffset": 531450
    },
    {
      "name": "copy_strided_u64",
      "dimensions": {},
      "sampleCount": 8440,
      "tileCount": 1472,
      "phaseCount": 95,
      "phaseCycles": 49020,
      "phaseMs": 0.03268,
      "workCycles": 10859268,
      "averageActiveTiles": 221.52729498164015,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 1286.6431279620854,
      "p50Cycles": 984,
      "p95Cycles": 3060,
      "maximumCycles": 3060,
      "firstOffset": 23694,
      "lastOffset": 15982500
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k384_c80_r160_r164",
      "dimensions": {},
      "sampleCount": 636,
      "tileCount": 636,
      "phaseCount": 16,
      "phaseCycles": 48462,
      "phaseMs": 0.032308,
      "workCycles": 30577608,
      "averageActiveTiles": 630.9605051380463,
      "usefulCycles": 24346080.0,
      "estimatedWorkCycles": 30577608,
      "usefulUtilization": 0.7962061649819044,
      "workLaneOccupancy": 0.996875,
      "deviceUsefulUtilization": 0.34128712231726155,
      "meanCycles": 48078.0,
      "p50Cycles": 48078,
      "p95Cycles": 48078,
      "maximumCycles": 48078,
      "firstOffset": 481974,
      "lastOffset": 530436
    },
    {
      "name": "gemm_f8_init_large_rows_interleaved_k288_c96_r144_r148",
      "dimensions": {},
      "sampleCount": 720,
      "tileCount": 720,
      "phaseCount": 1,
      "phaseCycles": 40866,
      "phaseMs": 0.027244,
      "workCycles": 29350080,
      "averageActiveTiles": 718.2029070621054,
      "usefulCycles": 23016960.0,
      "estimatedWorkCycles": 29350080,
      "usefulUtilization": 0.7842213717986458,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.38262912296604595,
      "meanCycles": 40764.0,
      "p50Cycles": 40764,
      "p95Cycles": 40764,
      "maximumCycles": 40764,
      "firstOffset": 131130,
      "lastOffset": 171996
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k288_c96_r144_r148",
      "dimensions": {},
      "sampleCount": 720,
      "tileCount": 720,
      "phaseCount": 3,
      "phaseCycles": 40416,
      "phaseMs": 0.026944,
      "workCycles": 28922400,
      "averageActiveTiles": 715.6175771971497,
      "usefulCycles": 22332672.0,
      "estimatedWorkCycles": 28922400,
      "usefulUtilization": 0.7721583271097834,
      "workLaneOccupancy": 0.9972222222222222,
      "deviceUsefulUtilization": 0.37538727667045335,
      "meanCycles": 40170.0,
      "p50Cycles": 40170,
      "p95Cycles": 40170,
      "maximumCycles": 40170,
      "firstOffset": 130920,
      "lastOffset": 171336
    },
    {
      "name": "gemm_f8_init_large_rows_interleaved_k384_c64_r160_r164",
      "dimensions": {},
      "sampleCount": 15,
      "tileCount": 15,
      "phaseCount": 9,
      "phaseCycles": 39468,
      "phaseMs": 0.026312,
      "workCycles": 589140,
      "averageActiveTiles": 14.927029492246884,
      "usefulCycles": 472320.0,
      "estimatedWorkCycles": 589140,
      "usefulUtilization": 0.8017109685304002,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.008129866352928734,
      "meanCycles": 39276.0,
      "p50Cycles": 39276,
      "p95Cycles": 39276,
      "maximumCycles": 39276,
      "firstOffset": 482256,
      "lastOffset": 521724
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k384_c64_r160_r164",
      "dimensions": {},
      "sampleCount": 12,
      "tileCount": 12,
      "phaseCount": 10,
      "phaseCycles": 38850,
      "phaseMs": 0.0259,
      "workCycles": 462528,
      "averageActiveTiles": 11.905482625482625,
      "usefulCycles": 367488.0,
      "estimatedWorkCycles": 462528,
      "usefulUtilization": 0.7945205479452054,
      "workLaneOccupancy": 0.996875,
      "deviceUsefulUtilization": 0.006426053382575122,
      "meanCycles": 38544.0,
      "p50Cycles": 38544,
      "p95Cycles": 38544,
      "maximumCycles": 38544,
      "firstOffset": 481980,
      "lastOffset": 520830
    },
    {
      "name": "gemm_f8_init_large_rows_interleaved_k96_c576_r20_r24",
      "dimensions": {},
      "sampleCount": 1440,
      "tileCount": 1440,
      "phaseCount": 21,
      "phaseCycles": 30156,
      "phaseMs": 0.020104,
      "workCycles": 41489280,
      "averageActiveTiles": 1375.8217270194987,
      "usefulCycles": 14929920.0,
      "estimatedWorkCycles": 41489280,
      "usefulUtilization": 0.35985006247396917,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.3363379989273171,
      "meanCycles": 28812.0,
      "p50Cycles": 28812,
      "p95Cycles": 28812,
      "maximumCycles": 28812,
      "firstOffset": 389916,
      "lastOffset": 420072
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k192_c96_r144_r148",
      "dimensions": {},
      "sampleCount": 720,
      "tileCount": 720,
      "phaseCount": 6,
      "phaseCycles": 29532,
      "phaseMs": 0.019688,
      "workCycles": 19418400,
      "averageActiveTiles": 657.5375863470134,
      "usefulCycles": 14888448.0,
      "estimatedWorkCycles": 19418400,
      "usefulUtilization": 0.7667185761957731,
      "workLaneOccupancy": 0.9972222222222222,
      "deviceUsefulUtilization": 0.34249068070596966,
      "meanCycles": 26970.0,
      "p50Cycles": 26970,
      "p95Cycles": 26970,
      "maximumCycles": 26970,
      "firstOffset": 15888690,
      "lastOffset": 15918222
    },
    {
      "name": "gemm_f8_init_large_rows_interleaved_k192_c96_r144_r148",
      "dimensions": {},
      "sampleCount": 720,
      "tileCount": 720,
      "phaseCount": 4,
      "phaseCycles": 27642,
      "phaseMs": 0.018428,
      "workCycles": 19699200,
      "averageActiveTiles": 712.6546559583243,
      "usefulCycles": 15344640.0,
      "estimatedWorkCycles": 19699200,
      "usefulUtilization": 0.7789473684210526,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.3771198837308066,
      "meanCycles": 27360.0,
      "p50Cycles": 27360,
      "p95Cycles": 27360,
      "maximumCycles": 27360,
      "firstOffset": 15891078,
      "lastOffset": 15918720
    },
    {
      "name": "add_f16",
      "dimensions": {},
      "sampleCount": 11220,
      "tileCount": 1472,
      "phaseCount": 139,
      "phaseCycles": 27540,
      "phaseMs": 0.01836,
      "workCycles": 20423970,
      "averageActiveTiles": 741.6111111111111,
      "usefulCycles": 4623400.0,
      "estimatedWorkCycles": 20423970,
      "usefulUtilization": 0.22637126866128376,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.11404853809478703,
      "meanCycles": 1820.3181818181818,
      "p50Cycles": 1374,
      "p95Cycles": 4314,
      "maximumCycles": 4314,
      "firstOffset": 91452,
      "lastOffset": 16065444
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k96_c576_r20_r24",
      "dimensions": {},
      "sampleCount": 24,
      "tileCount": 24,
      "phaseCount": 2,
      "phaseCycles": 27144,
      "phaseMs": 0.018096,
      "workCycles": 637056,
      "averageActiveTiles": 23.46949602122016,
      "usefulCycles": 186624.0,
      "estimatedWorkCycles": 637056,
      "usefulUtilization": 0.29294755877034356,
      "workLaneOccupancy": 0.9,
      "deviceUsefulUtilization": 0.004670741552300773,
      "meanCycles": 26544.0,
      "p50Cycles": 26544,
      "p95Cycles": 26544,
      "maximumCycles": 26544,
      "firstOffset": 390150,
      "lastOffset": 417294
    },
    {
      "name": "layer_norm_f16",
      "dimensions": {},
      "sampleCount": 4374,
      "tileCount": 1458,
      "phaseCount": 35,
      "phaseCycles": 22866,
      "phaseMs": 0.015244,
      "workCycles": 29463264,
      "averageActiveTiles": 1288.5184990816058,
      "usefulCycles": 11967264.0,
      "estimatedWorkCycles": 29463264,
      "usefulUtilization": 0.40617577197149646,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.35554687232610405,
      "meanCycles": 6736.0,
      "p50Cycles": 6738,
      "p95Cycles": 6738,
      "maximumCycles": 6738,
      "firstOffset": 99936,
      "lastOffset": 15870240
    },
    {
      "name": "bias_gelu_f16",
      "dimensions": {},
      "sampleCount": 1458,
      "tileCount": 1458,
      "phaseCount": 32,
      "phaseCycles": 17220,
      "phaseMs": 0.01148,
      "workCycles": 24284448,
      "averageActiveTiles": 1410.2466898954704,
      "usefulCycles": 17256888.0,
      "estimatedWorkCycles": 24284448,
      "usefulUtilization": 0.7106147934678194,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.6808030980154522,
      "meanCycles": 16656.0,
      "p50Cycles": 16656,
      "p95Cycles": 16656,
      "maximumCycles": 16656,
      "firstOffset": 557904,
      "lastOffset": 575124
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k128_c80_r104_r105",
      "dimensions": {},
      "sampleCount": 1152,
      "tileCount": 1152,
      "phaseCount": 14,
      "phaseCycles": 14850,
      "phaseMs": 0.0099,
      "workCycles": 13727232,
      "averageActiveTiles": 924.3927272727273,
      "usefulCycles": 8188128.0,
      "estimatedWorkCycles": 13727232,
      "usefulUtilization": 0.596487915407855,
      "workLaneOccupancy": 0.854296875,
      "deviceUsefulUtilization": 0.37458498023715414,
      "meanCycles": 11916.0,
      "p50Cycles": 11916,
      "p95Cycles": 11916,
      "maximumCycles": 11916,
      "firstOffset": 326202,
      "lastOffset": 341052
    },
    {
      "name": "gemm_f8_init_large_rows_interleaved_k128_c80_r104_r105",
      "dimensions": {},
      "sampleCount": 192,
      "tileCount": 192,
      "phaseCount": 4,
      "phaseCycles": 14568,
      "phaseMs": 0.009712,
      "workCycles": 2297088,
      "averageActiveTiles": 157.68039538714993,
      "usefulCycles": 1377810.0,
      "estimatedWorkCycles": 2297088,
      "usefulUtilization": 0.5998072342026078,
      "workLaneOccupancy": 0.854296875,
      "deviceUsefulUtilization": 0.06425125125349188,
      "meanCycles": 11964.0,
      "p50Cycles": 11964,
      "p95Cycles": 11964,
      "maximumCycles": 11964,
      "firstOffset": 326526,
      "lastOffset": 341094
    },
    {
      "name": "gemm_f8_init_large_rows_interleaved_k64_c48_r240_r244",
      "dimensions": {},
      "sampleCount": 1200,
      "tileCount": 1200,
      "phaseCount": 4,
      "phaseCycles": 14244,
      "phaseMs": 0.009496,
      "workCycles": 8704800,
      "averageActiveTiles": 611.1204717775905,
      "usefulCycles": 6456240.0,
      "estimatedWorkCycles": 8704800,
      "usefulUtilization": 0.7416873449131514,
      "workLaneOccupancy": 0.91875,
      "deviceUsefulUtilization": 0.30792141313505,
      "meanCycles": 7254.0,
      "p50Cycles": 7254,
      "p95Cycles": 7254,
      "maximumCycles": 7254,
      "firstOffset": 8508,
      "lastOffset": 22752
    },
    {
      "name": "gemm_f8_init_large_rows_interleaved_k96_c112_r121_r122",
      "dimensions": {},
      "sampleCount": 384,
      "tileCount": 384,
      "phaseCount": 4,
      "phaseCycles": 14190,
      "phaseMs": 0.00946,
      "workCycles": 5398272,
      "averageActiveTiles": 380.42790697674417,
      "usefulCycles": 2951424.0,
      "estimatedWorkCycles": 5398272,
      "usefulUtilization": 0.5467349551856594,
      "workLaneOccupancy": 0.75,
      "deviceUsefulUtilization": 0.1412997518154242,
      "meanCycles": 14058.0,
      "p50Cycles": 14058,
      "p95Cycles": 14058,
      "maximumCycles": 14058,
      "firstOffset": 243180,
      "lastOffset": 257370
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k96_c112_r121_r122",
      "dimensions": {},
      "sampleCount": 384,
      "tileCount": 384,
      "phaseCount": 4,
      "phaseCycles": 14160,
      "phaseMs": 0.00944,
      "workCycles": 5384448,
      "averageActiveTiles": 380.25762711864405,
      "usefulCycles": 2927232.0,
      "estimatedWorkCycles": 5384448,
      "usefulUtilization": 0.5436456996148908,
      "workLaneOccupancy": 0.75,
      "deviceUsefulUtilization": 0.14043846720707442,
      "meanCycles": 14022.0,
      "p50Cycles": 14022,
      "p95Cycles": 14022,
      "maximumCycles": 14022,
      "firstOffset": 243204,
      "lastOffset": 257364
    },
    {
      "name": "gemm_f8_init_large_rows_interleaved_k96_c96_r121_r122",
      "dimensions": {},
      "sampleCount": 288,
      "tileCount": 288,
      "phaseCount": 3,
      "phaseCycles": 12672,
      "phaseMs": 0.008448,
      "workCycles": 3487104,
      "averageActiveTiles": 275.1818181818182,
      "usefulCycles": 1851228.0,
      "estimatedWorkCycles": 3487104,
      "usefulUtilization": 0.5308783448959365,
      "workLaneOccupancy": 0.7317708333333334,
      "deviceUsefulUtilization": 0.09924461153656126,
      "meanCycles": 12108.0,
      "p50Cycles": 12108,
      "p95Cycles": 12108,
      "maximumCycles": 12108,
      "firstOffset": 242604,
      "lastOffset": 255276
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k64_c48_r240_r244",
      "dimensions": {},
      "sampleCount": 240,
      "tileCount": 240,
      "phaseCount": 2,
      "phaseCycles": 12426,
      "phaseMs": 0.008284,
      "workCycles": 1726560,
      "averageActiveTiles": 138.94736842105263,
      "usefulCycles": 1259496.0,
      "estimatedWorkCycles": 1726560,
      "usefulUtilization": 0.7294829024186822,
      "workLaneOccupancy": 0.91109375,
      "deviceUsefulUtilization": 0.06885851195599689,
      "meanCycles": 7194.0,
      "p50Cycles": 7194,
      "p95Cycles": 7194,
      "maximumCycles": 7194,
      "firstOffset": 9090,
      "lastOffset": 21516
    },
    {
      "name": "gemm_f16_init_small_rows_interleaved_k736_c16_r1_r1",
      "dimensions": {},
      "sampleCount": 160,
      "tileCount": 160,
      "phaseCount": 14,
      "phaseCycles": 12384,
      "phaseMs": 0.008256,
      "workCycles": 1257600,
      "averageActiveTiles": 101.55038759689923,
      "usefulCycles": 26244.0,
      "estimatedWorkCycles": 1257600,
      "usefulUtilization": 0.020868320610687024,
      "workLaneOccupancy": 0.8914402173913043,
      "deviceUsefulUtilization": 0.0014396644337714864,
      "meanCycles": 7860.0,
      "p50Cycles": 7860,
      "p95Cycles": 7860,
      "maximumCycles": 7860,
      "firstOffset": 15997794,
      "lastOffset": 16010178
    }
  ],
  "samples": []
}
