{
  "clockHz": 1500000000,
  "tileCount": 1472,
  "sampleCount": 185026,
  "matchedSampleCount": 94291,
  "profileSpanCycles": 10462992,
  "profileSpanMs": 6.975328,
  "atOffset": null,
  "groups": [
    {
      "name": "repeat-remainder",
      "dimensions": {},
      "sampleCount": 1472,
      "tileCount": 1472,
      "phaseCount": 8,
      "phaseCycles": 9900492,
      "phaseMs": 6.600328,
      "workCycles": 14569112424,
      "averageActiveTiles": 1471.5543857820398,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 9897494.85326087,
      "p50Cycles": 9897498,
      "p95Cycles": 9897510,
      "maximumCycles": 9897528,
      "firstOffset": 417936,
      "lastOffset": 10318428
    },
    {
      "name": "copy_u64",
      "dimensions": {},
      "sampleCount": 7918,
      "tileCount": 1472,
      "phaseCount": 151,
      "phaseCycles": 64182,
      "phaseMs": 0.042788,
      "workCycles": 7586616,
      "averageActiveTiles": 118.20473029821446,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 958.1480171760545,
      "p50Cycles": 1008,
      "p95Cycles": 1632,
      "maximumCycles": 4710,
      "firstOffset": 26568,
      "lastOffset": 10456488
    },
    {
      "name": "reduce_sum_f16",
      "dimensions": {},
      "sampleCount": 10600,
      "tileCount": 1472,
      "phaseCount": 137,
      "phaseCycles": 60978,
      "phaseMs": 0.040652,
      "workCycles": 42208656,
      "averageActiveTiles": 692.194824362885,
      "usefulCycles": 23067536.0,
      "estimatedWorkCycles": 42208656,
      "usefulUtilization": 0.5465119761216751,
      "workLaneOccupancy": 0.9933504298002166,
      "deviceUsefulUtilization": 0.2569923650297256,
      "meanCycles": 3981.948679245283,
      "p50Cycles": 3546,
      "p95Cycles": 7218,
      "maximumCycles": 7218,
      "firstOffset": 19164,
      "lastOffset": 10459290
    },
    {
      "name": "cast_f16_f8",
      "dimensions": {},
      "sampleCount": 8180,
      "tileCount": 1458,
      "phaseCount": 113,
      "phaseCycles": 43500,
      "phaseMs": 0.029,
      "workCycles": 26411394,
      "averageActiveTiles": 607.1584827586207,
      "usefulCycles": 2253738.0,
      "estimatedWorkCycles": 26078118,
      "usefulUtilization": 0.08642257083122333,
      "workLaneOccupancy": 0.9712713325288743,
      "deviceUsefulUtilization": 0.03519705772113943,
      "meanCycles": 3228.776772616137,
      "p50Cycles": 2382,
      "p95Cycles": 5496,
      "maximumCycles": 5496,
      "firstOffset": 0,
      "lastOffset": 10444956
    },
    {
      "name": "fill_zero_u64",
      "dimensions": {},
      "sampleCount": 21042,
      "tileCount": 1472,
      "phaseCount": 158,
      "phaseCycles": 43458,
      "phaseMs": 0.028972,
      "workCycles": 6994788,
      "averageActiveTiles": 160.95512909015602,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 3836826,
      "usefulUtilization": 0.0,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": 0.0,
      "meanCycles": 332.42030225263755,
      "p50Cycles": 270,
      "p95Cycles": 888,
      "maximumCycles": 1260,
      "firstOffset": 23052,
      "lastOffset": 10454814
    },
    {
      "name": "gemm_f8_init_large_rows_interleaved_k192_c160_r80_r82",
      "dimensions": {},
      "sampleCount": 780,
      "tileCount": 780,
      "phaseCount": 30,
      "phaseCycles": 32730,
      "phaseMs": 0.02182,
      "workCycles": 23016240,
      "averageActiveTiles": 703.2153987167736,
      "usefulCycles": 15350400.0,
      "estimatedWorkCycles": 23016240,
      "usefulUtilization": 0.6669377795851973,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.31861475311839954,
      "meanCycles": 29508.0,
      "p50Cycles": 29508,
      "p95Cycles": 29508,
      "maximumCycles": 29508,
      "firstOffset": 277926,
      "lastOffset": 310656
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k192_c160_r80_r82",
      "dimensions": {},
      "sampleCount": 624,
      "tileCount": 624,
      "phaseCount": 22,
      "phaseCycles": 30048,
      "phaseMs": 0.020032,
      "workCycles": 18094752,
      "averageActiveTiles": 602.1948881789137,
      "usefulCycles": 11943360.0,
      "estimatedWorkCycles": 18094752,
      "usefulUtilization": 0.660045520380716,
      "workLaneOccupancy": 0.996875,
      "deviceUsefulUtilization": 0.2700244825670232,
      "meanCycles": 28998.0,
      "p50Cycles": 28998,
      "p95Cycles": 28998,
      "maximumCycles": 28998,
      "firstOffset": 280134,
      "lastOffset": 310182
    },
    {
      "name": "gemm_f8_init_large_rows_interleaved_k192_c144_r80_r82",
      "dimensions": {},
      "sampleCount": 30,
      "tileCount": 30,
      "phaseCount": 14,
      "phaseCycles": 29796,
      "phaseMs": 0.019864,
      "workCycles": 797940,
      "averageActiveTiles": 26.780104712041886,
      "usefulCycles": 531360.0,
      "estimatedWorkCycles": 797940,
      "usefulUtilization": 0.6659147304308595,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.01211499063194943,
      "meanCycles": 26598.0,
      "p50Cycles": 26598,
      "p95Cycles": 26598,
      "maximumCycles": 26598,
      "firstOffset": 277938,
      "lastOffset": 307734
    },
    {
      "name": "attention_softmax_d72_p768_tail_f8_sm4",
      "dimensions": {},
      "sampleCount": 1472,
      "tileCount": 1472,
      "phaseCount": 22,
      "phaseCycles": 28584,
      "phaseMs": 0.019056,
      "workCycles": 38108064,
      "averageActiveTiles": 1333.1956339210747,
      "usefulCycles": 16866144.0,
      "estimatedWorkCycles": 38108064,
      "usefulUtilization": 0.4425872697180313,
      "workLaneOccupancy": 0.9972413793103448,
      "deviceUsefulUtilization": 0.4008528638703318,
      "meanCycles": 25888.630434782608,
      "p50Cycles": 25890,
      "p95Cycles": 25890,
      "maximumCycles": 25890,
      "firstOffset": 160536,
      "lastOffset": 189120
    },
    {
      "name": "gemm_f8_init_large_rows_interleaved_k160_c128_r120_r122",
      "dimensions": {},
      "sampleCount": 1215,
      "tileCount": 1215,
      "phaseCount": 34,
      "phaseCycles": 28188,
      "phaseMs": 0.018792,
      "workCycles": 31871880,
      "averageActiveTiles": 1130.6896551724137,
      "usefulCycles": 23628960.0,
      "estimatedWorkCycles": 31871880,
      "usefulUtilization": 0.7413732732427456,
      "workLaneOccupancy": 0.9962962962962963,
      "deviceUsefulUtilization": 0.5694722083402743,
      "meanCycles": 26232.0,
      "p50Cycles": 26232,
      "p95Cycles": 26232,
      "maximumCycles": 26232,
      "firstOffset": 365280,
      "lastOffset": 393468
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k192_c144_r80_r82",
      "dimensions": {},
      "sampleCount": 24,
      "tileCount": 24,
      "phaseCount": 13,
      "phaseCycles": 27180,
      "phaseMs": 0.01812,
      "workCycles": 627408,
      "averageActiveTiles": 23.08344370860927,
      "usefulCycles": 413424.0,
      "estimatedWorkCycles": 627408,
      "usefulUtilization": 0.6589396373651595,
      "workLaneOccupancy": 0.996875,
      "deviceUsefulUtilization": 0.010333285344082925,
      "meanCycles": 26142.0,
      "p50Cycles": 26142,
      "p95Cycles": 26142,
      "maximumCycles": 26142,
      "firstOffset": 280134,
      "lastOffset": 307314
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k160_c128_r120_r122",
      "dimensions": {},
      "sampleCount": 243,
      "tileCount": 243,
      "phaseCount": 19,
      "phaseCycles": 26682,
      "phaseMs": 0.017788,
      "workCycles": 6349590,
      "averageActiveTiles": 237.9727906453789,
      "usefulCycles": 4609584.0,
      "estimatedWorkCycles": 6349590,
      "usefulUtilization": 0.7259656135278025,
      "workLaneOccupancy": 0.9879938271604938,
      "deviceUsefulUtilization": 0.11736417320910042,
      "meanCycles": 26130.0,
      "p50Cycles": 26130,
      "p95Cycles": 26130,
      "maximumCycles": 26130,
      "firstOffset": 366606,
      "lastOffset": 393288
    },
    {
      "name": "copy_strided_u64",
      "dimensions": {},
      "sampleCount": 3976,
      "tileCount": 1472,
      "phaseCount": 105,
      "phaseCycles": 25218,
      "phaseMs": 0.016812,
      "workCycles": 1637796,
      "averageActiveTiles": 64.945515108256,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 411.920523138833,
      "p50Cycles": 384,
      "p95Cycles": 522,
      "maximumCycles": 720,
      "firstOffset": 103494,
      "lastOffset": 10407840
    },
    {
      "name": "gemm_f8_init_large_rows_interleaved_k128_c64_r242_r244",
      "dimensions": {},
      "sampleCount": 972,
      "tileCount": 972,
      "phaseCount": 3,
      "phaseCycles": 21516,
      "phaseMs": 0.014344,
      "workCycles": 18055872,
      "averageActiveTiles": 839.1834913552705,
      "usefulCycles": 15178752.0,
      "estimatedWorkCycles": 18055872,
      "usefulUtilization": 0.8406546080964685,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.47925507408036083,
      "meanCycles": 18576.0,
      "p50Cycles": 18576,
      "p95Cycles": 18576,
      "maximumCycles": 18576,
      "firstOffset": 60948,
      "lastOffset": 82464
    },
    {
      "name": "layer_norm_f8",
      "dimensions": {},
      "sampleCount": 1458,
      "tileCount": 729,
      "phaseCount": 30,
      "phaseCycles": 20406,
      "phaseMs": 0.013604,
      "workCycles": 13900572,
      "averageActiveTiles": 681.2002352249339,
      "usefulCycles": 5668704.0,
      "estimatedWorkCycles": 13900572,
      "usefulUtilization": 0.407803650094399,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.18872006954476306,
      "meanCycles": 9534.0,
      "p50Cycles": 9528,
      "p95Cycles": 9540,
      "maximumCycles": 9540,
      "firstOffset": 253500,
      "lastOffset": 10333680
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k128_c64_r242_r244",
      "dimensions": {},
      "sampleCount": 486,
      "tileCount": 486,
      "phaseCount": 1,
      "phaseCycles": 19614,
      "phaseMs": 0.013076,
      "workCycles": 8963784,
      "averageActiveTiles": 457.009483022331,
      "usefulCycles": 7496064.0,
      "estimatedWorkCycles": 8963784,
      "usefulUtilization": 0.836261114725656,
      "workLaneOccupancy": 0.9958677685950413,
      "deviceUsefulUtilization": 0.25963264926117546,
      "meanCycles": 18444.0,
      "p50Cycles": 18444,
      "p95Cycles": 18444,
      "maximumCycles": 18444,
      "firstOffset": 62718,
      "lastOffset": 82332
    },
    {
      "name": "add_f16",
      "dimensions": {},
      "sampleCount": 11800,
      "tileCount": 1458,
      "phaseCount": 154,
      "phaseCycles": 18246,
      "phaseMs": 0.012164,
      "workCycles": 12935772,
      "averageActiveTiles": 708.9648142058534,
      "usefulCycles": 2310624.0,
      "estimatedWorkCycles": 12935772,
      "usefulUtilization": 0.17862281431676438,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.08603076791101326,
      "meanCycles": 1096.2518644067798,
      "p50Cycles": 942,
      "p95Cycles": 1806,
      "maximumCycles": 1806,
      "firstOffset": 23856,
      "lastOffset": 10462992
    },
    {
      "name": "gemm_f8_init_large_rows_interleaved_k128_c128_r80_r82",
      "dimensions": {},
      "sampleCount": 810,
      "tileCount": 810,
      "phaseCount": 7,
      "phaseCycles": 16614,
      "phaseMs": 0.011076,
      "workCycles": 12956760,
      "averageActiveTiles": 779.8699891657639,
      "usefulCycles": 8501760.0,
      "estimatedWorkCycles": 12956760,
      "usefulUtilization": 0.6561640410102526,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.3476376654576287,
      "meanCycles": 15996.0,
      "p50Cycles": 15996,
      "p95Cycles": 15996,
      "maximumCycles": 15996,
      "firstOffset": 10343754,
      "lastOffset": 10360368
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k128_c128_r80_r82",
      "dimensions": {},
      "sampleCount": 648,
      "tileCount": 648,
      "phaseCount": 7,
      "phaseCycles": 16044,
      "phaseMs": 0.010696,
      "workCycles": 10198224,
      "averageActiveTiles": 635.6409872849663,
      "usefulCycles": 6614784.0,
      "estimatedWorkCycles": 10198224,
      "usefulUtilization": 0.6486211716863642,
      "workLaneOccupancy": 0.996875,
      "deviceUsefulUtilization": 0.28008845240805175,
      "meanCycles": 15738.0,
      "p50Cycles": 15738,
      "p95Cycles": 15738,
      "maximumCycles": 15738,
      "firstOffset": 10343928,
      "lastOffset": 10359972
    },
    {
      "name": "rearrange_row_major_to_amp_f16_o2_r64_p64_c16_p16_b64x16",
      "dimensions": {},
      "sampleCount": 1760,
      "tileCount": 950,
      "phaseCount": 31,
      "phaseCycles": 10860,
      "phaseMs": 0.00724,
      "workCycles": 4282080,
      "averageActiveTiles": 394.29834254143645,
      "usefulCycles": 901120.0,
      "estimatedWorkCycles": 4282080,
      "usefulUtilization": 0.21043978627209206,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.05636960525262231,
      "meanCycles": 2433.0,
      "p50Cycles": 2430,
      "p95Cycles": 2436,
      "maximumCycles": 2436,
      "firstOffset": 127374,
      "lastOffset": 10399014
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k128_c80_r56_r57",
      "dimensions": {},
      "sampleCount": 1152,
      "tileCount": 1152,
      "phaseCount": 22,
      "phaseCycles": 9702,
      "phaseMs": 0.006468,
      "workCycles": 9303552,
      "averageActiveTiles": 958.9313543599258,
      "usefulCycles": 4408992.0,
      "estimatedWorkCycles": 9303552,
      "usefulUtilization": 0.4739041604754829,
      "workLaneOccupancy": 0.854296875,
      "deviceUsefulUtilization": 0.30872388481084134,
      "meanCycles": 8076.0,
      "p50Cycles": 8076,
      "p95Cycles": 8076,
      "maximumCycles": 8076,
      "firstOffset": 196314,
      "lastOffset": 206016
    },
    {
      "name": "rearrange_row_major_to_amp_f16_o2_r0_p64_c16_p16_b64x16",
      "dimensions": {},
      "sampleCount": 160,
      "tileCount": 150,
      "phaseCount": 21,
      "phaseCycles": 9510,
      "phaseMs": 0.00634,
      "workCycles": 314400,
      "averageActiveTiles": 33.05993690851735,
      "usefulCycles": 32000.0,
      "estimatedWorkCycles": 314400,
      "usefulUtilization": 0.10178117048346055,
      "workLaneOccupancy": 0.390625,
      "deviceUsefulUtilization": 0.0022859232844145753,
      "meanCycles": 1965.0,
      "p50Cycles": 1962,
      "p95Cycles": 1968,
      "maximumCycles": 1968,
      "firstOffset": 127122,
      "lastOffset": 10397616
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k96_c48_r146_r146",
      "dimensions": {},
      "sampleCount": 1440,
      "tileCount": 1440,
      "phaseCount": 31,
      "phaseCycles": 9390,
      "phaseMs": 0.00626,
      "workCycles": 10255680,
      "averageActiveTiles": 1092.1916932907347,
      "usefulCycles": 7558272.0,
      "estimatedWorkCycles": 10255680,
      "usefulUtilization": 0.7369839932603202,
      "workLaneOccupancy": 0.9986301369863013,
      "deviceUsefulUtilization": 0.5468259480483401,
      "meanCycles": 7122.0,
      "p50Cycles": 7122,
      "p95Cycles": 7122,
      "maximumCycles": 7122,
      "firstOffset": 227502,
      "lastOffset": 236892
    },
    {
      "name": "layer_norm_f16",
      "dimensions": {},
      "sampleCount": 729,
      "tileCount": 729,
      "phaseCount": 2,
      "phaseCycles": 9252,
      "phaseMs": 0.006168,
      "workCycles": 4912002,
      "averageActiveTiles": 530.9124513618677,
      "usefulCycles": 1994544.0,
      "estimatedWorkCycles": 4912002,
      "usefulUtilization": 0.40605520926090827,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.14645364574522077,
      "meanCycles": 6738.0,
      "p50Cycles": 6738,
      "p95Cycles": 6738,
      "maximumCycles": 6738,
      "firstOffset": 34998,
      "lastOffset": 44250
    },
    {
      "name": "bias_gelu_f16",
      "dimensions": {},
      "sampleCount": 1458,
      "tileCount": 1458,
      "phaseCount": 35,
      "phaseCycles": 9078,
      "phaseMs": 0.006052,
      "workCycles": 12474648,
      "averageActiveTiles": 1374.1625908790481,
      "usefulCycles": 8628444.0,
      "estimatedWorkCycles": 12474648,
      "usefulUtilization": 0.6916783543712015,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.6457055159631024,
      "meanCycles": 8556.0,
      "p50Cycles": 8556,
      "p95Cycles": 8556,
      "maximumCycles": 8556,
      "firstOffset": 335550,
      "lastOffset": 344628
    },
    {
      "name": "gemm_f8_init_large_rows_interleaved_k128_c80_r56_r57",
      "dimensions": {},
      "sampleCount": 96,
      "tileCount": 96,
      "phaseCount": 5,
      "phaseCycles": 8718,
      "phaseMs": 0.005812,
      "workCycles": 779904,
      "averageActiveTiles": 89.45905024088094,
      "usefulCycles": 373977.0,
      "estimatedWorkCycles": 779904,
      "usefulUtilization": 0.47951670974889216,
      "workLaneOccupancy": 0.854296875,
      "deviceUsefulUtilization": 0.029142058035847872,
      "meanCycles": 8124.0,
      "p50Cycles": 8124,
      "p95Cycles": 8124,
      "maximumCycles": 8124,
      "firstOffset": 197358,
      "lastOffset": 206076
    },
    {
      "name": "gemm_f16_init_small_rows_interleaved_k80_c16_r1_r1",
      "dimensions": {},
      "sampleCount": 736,
      "tileCount": 736,
      "phaseCount": 18,
      "phaseCycles": 8136,
      "phaseMs": 0.005424,
      "workCycles": 896448,
      "averageActiveTiles": 110.18289085545723,
      "usefulCycles": 13122.0,
      "estimatedWorkCycles": 896448,
      "usefulUtilization": 0.01463777040051403,
      "workLaneOccupancy": 0.8914402173913043,
      "deviceUsefulUtilization": 0.0010956738168526356,
      "meanCycles": 1218.0,
      "p50Cycles": 1218,
      "p95Cycles": 1218,
      "maximumCycles": 1218,
      "firstOffset": 10400922,
      "lastOffset": 10409058
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k96_c80_r81_r81",
      "dimensions": {},
      "sampleCount": 864,
      "tileCount": 864,
      "phaseCount": 11,
      "phaseCycles": 7938,
      "phaseMs": 0.005292,
      "workCycles": 6630336,
      "averageActiveTiles": 835.265306122449,
      "usefulCycles": 3149280.0,
      "estimatedWorkCycles": 6630336,
      "usefulUtilization": 0.4749804534792807,
      "workLaneOccupancy": 0.75,
      "deviceUsefulUtilization": 0.26952085181898844,
      "meanCycles": 7674.0,
      "p50Cycles": 7674,
      "p95Cycles": 7674,
      "maximumCycles": 7674,
      "firstOffset": 146544,
      "lastOffset": 154482
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k64_c32_r182_r184",
      "dimensions": {},
      "sampleCount": 1080,
      "tileCount": 1080,
      "phaseCount": 2,
      "phaseCycles": 7452,
      "phaseMs": 0.004968,
      "workCycles": 4302720,
      "averageActiveTiles": 577.3913043478261,
      "usefulCycles": 2884140.0,
      "estimatedWorkCycles": 4302720,
      "usefulUtilization": 0.6703062248995983,
      "workLaneOccupancy": 0.9170673076923077,
      "deviceUsefulUtilization": 0.26292729993698805,
      "meanCycles": 3984.0,
      "p50Cycles": 3984,
      "p95Cycles": 3984,
      "maximumCycles": 3984,
      "firstOffset": 8724,
      "lastOffset": 16176
    },
    {
      "name": "gemm_f8_init_large_rows_interleaved_k64_c32_r182_r184",
      "dimensions": {},
      "sampleCount": 360,
      "tileCount": 360,
      "phaseCount": 3,
      "phaseCycles": 7404,
      "phaseMs": 0.004936,
      "workCycles": 1442880,
      "averageActiveTiles": 194.87844408427875,
      "usefulCycles": 973728.0,
      "estimatedWorkCycles": 1442880,
      "usefulUtilization": 0.6748502994011976,
      "workLaneOccupancy": 0.91875,
      "deviceUsefulUtilization": 0.08934359805510535,
      "meanCycles": 4008.0,
      "p50Cycles": 4008,
      "p95Cycles": 4008,
      "maximumCycles": 4008,
      "firstOffset": 8748,
      "lastOffset": 16152
    }
  ],
  "samples": []
}
