{
  "clockHz": 1500000000,
  "tileCount": 1472,
  "sampleCount": 189315,
  "matchedSampleCount": 36947,
  "profileSpanCycles": 10666374,
  "profileSpanMs": 7.110916,
  "atOffset": null,
  "groups": [
    {
      "name": "exchange",
      "dimensions": {},
      "sampleCount": 7360,
      "tileCount": 1472,
      "phaseCount": 5,
      "phaseCycles": 66558,
      "phaseMs": 0.044372,
      "workCycles": 53509884,
      "averageActiveTiles": 803.9587127017038,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 7270.364673913044,
      "p50Cycles": 6996,
      "p95Cycles": 12852,
      "maximumCycles": 23226,
      "firstOffset": 111348,
      "lastOffset": 202212
    },
    {
      "name": "attention_softmax_d72_p768_tail_f8_sm4",
      "dimensions": {},
      "sampleCount": 1472,
      "tileCount": 1472,
      "phaseCount": 19,
      "phaseCycles": 28500,
      "phaseMs": 0.019,
      "workCycles": 38108064,
      "averageActiveTiles": 1337.125052631579,
      "usefulCycles": 16866144.0,
      "estimatedWorkCycles": 38108064,
      "usefulUtilization": 0.4425872697180313,
      "workLaneOccupancy": 0.9972413793103448,
      "deviceUsefulUtilization": 0.4020343249427918,
      "meanCycles": 25888.630434782608,
      "p50Cycles": 25890,
      "p95Cycles": 25890,
      "maximumCycles": 25890,
      "firstOffset": 153600,
      "lastOffset": 182100
    },
    {
      "name": "fill_zero_u64",
      "dimensions": {},
      "sampleCount": 14176,
      "tileCount": 1472,
      "phaseCount": 51,
      "phaseCycles": 20478,
      "phaseMs": 0.013652,
      "workCycles": 4881888,
      "averageActiveTiles": 238.39671842953413,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 2760192,
      "usefulUtilization": 0.0,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": 0.0,
      "meanCycles": 344.37697516930024,
      "p50Cycles": 306,
      "p95Cycles": 564,
      "maximumCycles": 564,
      "firstOffset": 111078,
      "lastOffset": 207288
    },
    {
      "name": "gemm_f16_init_small_rows_interleaved_k80_c80_r81_r81",
      "dimensions": {},
      "sampleCount": 864,
      "tileCount": 864,
      "phaseCount": 11,
      "phaseCycles": 13086,
      "phaseMs": 0.008724,
      "workCycles": 11062656,
      "averageActiveTiles": 845.3810178817056,
      "usefulCycles": 6298560.0,
      "estimatedWorkCycles": 11062656,
      "usefulUtilization": 0.5693533270852859,
      "workLaneOccupancy": 0.9,
      "deviceUsefulUtilization": 0.3269840320554991,
      "meanCycles": 12804.0,
      "p50Cycles": 12804,
      "p95Cycles": 12804,
      "maximumCycles": 12804,
      "firstOffset": 134472,
      "lastOffset": 147558
    },
    {
      "name": "copy_u64",
      "dimensions": {},
      "sampleCount": 4443,
      "tileCount": 1472,
      "phaseCount": 46,
      "phaseCycles": 12996,
      "phaseMs": 0.008664,
      "workCycles": 3684534,
      "averageActiveTiles": 283.5129270544783,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 829.2896691424713,
      "p50Cycles": 960,
      "p95Cycles": 1236,
      "maximumCycles": 1236,
      "firstOffset": 112668,
      "lastOffset": 207804
    },
    {
      "name": "gemm_f16_init_small_rows_interleaved_k80_c64_r81_r81",
      "dimensions": {},
      "sampleCount": 576,
      "tileCount": 576,
      "phaseCount": 12,
      "phaseCycles": 10902,
      "phaseMs": 0.007268,
      "workCycles": 5944320,
      "averageActiveTiles": 545.2504127682994,
      "usefulCycles": 3267378.0,
      "estimatedWorkCycles": 5944320,
      "usefulUtilization": 0.5496638808139535,
      "workLaneOccupancy": 0.875390625,
      "deviceUsefulUtilization": 0.20360357194132708,
      "meanCycles": 10320.0,
      "p50Cycles": 10320,
      "p95Cycles": 10320,
      "maximumCycles": 10320,
      "firstOffset": 133884,
      "lastOffset": 144786
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k128_c80_r56_r57",
      "dimensions": {},
      "sampleCount": 1152,
      "tileCount": 1152,
      "phaseCount": 20,
      "phaseCycles": 9960,
      "phaseMs": 0.00664,
      "workCycles": 9662976,
      "averageActiveTiles": 970.178313253012,
      "usefulCycles": 4408992.0,
      "estimatedWorkCycles": 9662976,
      "usefulUtilization": 0.4562768240343348,
      "workLaneOccupancy": 0.854296875,
      "deviceUsefulUtilization": 0.3007268203247774,
      "meanCycles": 8388.0,
      "p50Cycles": 8388,
      "p95Cycles": 8388,
      "maximumCycles": 8388,
      "firstOffset": 189288,
      "lastOffset": 199248
    },
    {
      "name": "gemm_f8_init_large_rows_interleaved_k128_c80_r56_r57",
      "dimensions": {},
      "sampleCount": 96,
      "tileCount": 96,
      "phaseCount": 6,
      "phaseCycles": 8964,
      "phaseMs": 0.005976,
      "workCycles": 809856,
      "averageActiveTiles": 90.34538152610442,
      "usefulCycles": 373977.0,
      "estimatedWorkCycles": 809856,
      "usefulUtilization": 0.46178209459459457,
      "workLaneOccupancy": 0.854296875,
      "deviceUsefulUtilization": 0.028342309455212152,
      "meanCycles": 8436.0,
      "p50Cycles": 8436,
      "p95Cycles": 8436,
      "maximumCycles": 8436,
      "firstOffset": 190338,
      "lastOffset": 199302
    },
    {
      "name": "attention_merge_v72_p80_k768_out16_p8_f16",
      "dimensions": {},
      "sampleCount": 1472,
      "tileCount": 1472,
      "phaseCount": 32,
      "phaseCycles": 6978,
      "phaseMs": 0.004652,
      "workCycles": 4398336,
      "averageActiveTiles": 630.3147033533963,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 2988.0,
      "p50Cycles": 2988,
      "p95Cycles": 2988,
      "maximumCycles": 2988,
      "firstOffset": 204234,
      "lastOffset": 211212
    },
    {
      "name": "copy_strided_u64",
      "dimensions": {},
      "sampleCount": 1720,
      "tileCount": 1472,
      "phaseCount": 39,
      "phaseCycles": 5508,
      "phaseMs": 0.003672,
      "workCycles": 709056,
      "averageActiveTiles": 128.73202614379085,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 412.2418604651163,
      "p50Cycles": 420,
      "p95Cycles": 420,
      "maximumCycles": 480,
      "firstOffset": 120102,
      "lastOffset": 208224
    },
    {
      "name": "cast_f16_f8",
      "dimensions": {},
      "sampleCount": 960,
      "tileCount": 960,
      "phaseCount": 12,
      "phaseCycles": 5220,
      "phaseMs": 0.00348,
      "workCycles": 2286720,
      "averageActiveTiles": 438.0689655172414,
      "usefulCycles": 116640.0,
      "estimatedWorkCycles": 2286720,
      "usefulUtilization": 0.05100755667506297,
      "workLaneOccupancy": 0.94921875,
      "deviceUsefulUtilization": 0.015179910044977511,
      "meanCycles": 2382.0,
      "p50Cycles": 2382,
      "p95Cycles": 2382,
      "maximumCycles": 2382,
      "firstOffset": 120726,
      "lastOffset": 125946
    },
    {
      "name": "reduce_sum_f16",
      "dimensions": {},
      "sampleCount": 1472,
      "tileCount": 1472,
      "phaseCount": 26,
      "phaseCycles": 4140,
      "phaseMs": 0.00276,
      "workCycles": 3028032,
      "averageActiveTiles": 731.408695652174,
      "usefulCycles": 1065120.0,
      "estimatedWorkCycles": 3028032,
      "usefulUtilization": 0.3517532179316467,
      "workLaneOccupancy": 0.9131687242798354,
      "deviceUsefulUtilization": 0.17477945809703843,
      "meanCycles": 2057.086956521739,
      "p50Cycles": 2052,
      "p95Cycles": 2238,
      "maximumCycles": 2238,
      "firstOffset": 201318,
      "lastOffset": 205458
    },
    {
      "name": "rearrange_row_major_to_amp_f16_o2_r64_p64_c16_p16_b64x16",
      "dimensions": {},
      "sampleCount": 880,
      "tileCount": 880,
      "phaseCount": 11,
      "phaseCycles": 3924,
      "phaseMs": 0.002616,
      "workCycles": 2143680,
      "averageActiveTiles": 546.2996941896024,
      "usefulCycles": 450560.0,
      "estimatedWorkCycles": 2143680,
      "usefulUtilization": 0.21018062397372742,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.07800381154988255,
      "meanCycles": 2436.0,
      "p50Cycles": 2436,
      "p95Cycles": 2436,
      "maximumCycles": 2436,
      "firstOffset": 119640,
      "lastOffset": 123564
    },
    {
      "name": "rearrange_row_major_to_amp_f16_o2_r0_p64_c16_p16_b64x16",
      "dimensions": {},
      "sampleCount": 80,
      "tileCount": 80,
      "phaseCount": 7,
      "phaseCycles": 3798,
      "phaseMs": 0.002532,
      "workCycles": 157440,
      "averageActiveTiles": 41.45339652448657,
      "usefulCycles": 16000.0,
      "estimatedWorkCycles": 157440,
      "usefulUtilization": 0.1016260162601626,
      "workLaneOccupancy": 0.390625,
      "deviceUsefulUtilization": 0.002861918172035625,
      "meanCycles": 1968.0,
      "p50Cycles": 1968,
      "p95Cycles": 1968,
      "maximumCycles": 1968,
      "firstOffset": 118758,
      "lastOffset": 122556
    },
    {
      "name": "static_copy_u32",
      "dimensions": {},
      "sampleCount": 224,
      "tileCount": 184,
      "phaseCount": 8,
      "phaseCycles": 3408,
      "phaseMs": 0.002272,
      "workCycles": 63696,
      "averageActiveTiles": 18.690140845070424,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 284.35714285714283,
      "p50Cycles": 294,
      "p95Cycles": 294,
      "maximumCycles": 294,
      "firstOffset": 113538,
      "lastOffset": 121128
    }
  ],
  "samples": []
}
