{
  "clockHz": 1500000000,
  "tileCount": 1472,
  "sampleCount": 189647,
  "matchedSampleCount": 37279,
  "profileSpanCycles": 11081280,
  "profileSpanMs": 7.38752,
  "atOffset": null,
  "groups": [
    {
      "name": "exchange",
      "dimensions": {},
      "sampleCount": 7360,
      "tileCount": 1472,
      "phaseCount": 5,
      "phaseCycles": 81456,
      "phaseMs": 0.054304,
      "workCycles": 68803098,
      "averageActiveTiles": 844.6658073070124,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 9348.247010869565,
      "p50Cycles": 9300,
      "p95Cycles": 16062,
      "maximumCycles": 29376,
      "firstOffset": 111348,
      "lastOffset": 217104
    },
    {
      "name": "attention_softmax_d72_p768_tail_f8_sm4",
      "dimensions": {},
      "sampleCount": 1472,
      "tileCount": 1472,
      "phaseCount": 20,
      "phaseCycles": 28506,
      "phaseMs": 0.019004,
      "workCycles": 38108064,
      "averageActiveTiles": 1336.843611871185,
      "usefulCycles": 16866144.0,
      "estimatedWorkCycles": 38108064,
      "usefulUtilization": 0.4425872697180313,
      "workLaneOccupancy": 0.9972413793103448,
      "deviceUsefulUtilization": 0.40194970395248597,
      "meanCycles": 25888.630434782608,
      "p50Cycles": 25890,
      "p95Cycles": 25890,
      "maximumCycles": 25890,
      "firstOffset": 156978,
      "lastOffset": 185484
    },
    {
      "name": "fill_zero_u64",
      "dimensions": {},
      "sampleCount": 14240,
      "tileCount": 1472,
      "phaseCount": 53,
      "phaseCycles": 27102,
      "phaseMs": 0.018068,
      "workCycles": 5224224,
      "averageActiveTiles": 192.76156741199912,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 3047616,
      "usefulUtilization": 0.0,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": 0.0,
      "meanCycles": 366.8696629213483,
      "p50Cycles": 306,
      "p95Cycles": 720,
      "maximumCycles": 720,
      "firstOffset": 111078,
      "lastOffset": 222174
    },
    {
      "name": "gemm_f16_init_small_rows_interleaved_k80_c80_r81_r81",
      "dimensions": {},
      "sampleCount": 864,
      "tileCount": 864,
      "phaseCount": 12,
      "phaseCycles": 19236,
      "phaseMs": 0.012824,
      "workCycles": 11062656,
      "averageActiveTiles": 575.101684341859,
      "usefulCycles": 6298560.0,
      "estimatedWorkCycles": 11062656,
      "usefulUtilization": 0.5693533270852859,
      "workLaneOccupancy": 0.9,
      "deviceUsefulUtilization": 0.2224429737720036,
      "meanCycles": 12804.0,
      "p50Cycles": 12804,
      "p95Cycles": 12804,
      "maximumCycles": 12804,
      "firstOffset": 131724,
      "lastOffset": 150960
    },
    {
      "name": "gemm_f16_init_small_rows_interleaved_k80_c64_r81_r81",
      "dimensions": {},
      "sampleCount": 576,
      "tileCount": 576,
      "phaseCount": 11,
      "phaseCycles": 17268,
      "phaseMs": 0.011512,
      "workCycles": 5944320,
      "averageActiveTiles": 344.23905489923555,
      "usefulCycles": 3267378.0,
      "estimatedWorkCycles": 5944320,
      "usefulUtilization": 0.5496638808139535,
      "workLaneOccupancy": 0.875390625,
      "deviceUsefulUtilization": 0.12854332530138685,
      "meanCycles": 10320.0,
      "p50Cycles": 10320,
      "p95Cycles": 10320,
      "maximumCycles": 10320,
      "firstOffset": 131118,
      "lastOffset": 148386
    },
    {
      "name": "copy_u64",
      "dimensions": {},
      "sampleCount": 4447,
      "tileCount": 1472,
      "phaseCount": 49,
      "phaseCycles": 15666,
      "phaseMs": 0.010444,
      "workCycles": 3687606,
      "averageActiveTiles": 235.38912294140175,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 829.2345401394198,
      "p50Cycles": 960,
      "p95Cycles": 1236,
      "maximumCycles": 1236,
      "firstOffset": 112668,
      "lastOffset": 222690
    },
    {
      "name": "gemm_f8_init_small_rows_interleaved_k128_c80_r56_r57",
      "dimensions": {},
      "sampleCount": 1152,
      "tileCount": 1152,
      "phaseCount": 21,
      "phaseCycles": 12036,
      "phaseMs": 0.008024,
      "workCycles": 9662976,
      "averageActiveTiles": 802.839481555334,
      "usefulCycles": 4408992.0,
      "estimatedWorkCycles": 9662976,
      "usefulUtilization": 0.4562768240343348,
      "workLaneOccupancy": 0.854296875,
      "deviceUsefulUtilization": 0.2488566907971737,
      "meanCycles": 8388.0,
      "p50Cycles": 8388,
      "p95Cycles": 8388,
      "maximumCycles": 8388,
      "firstOffset": 202134,
      "lastOffset": 214170
    },
    {
      "name": "gemm_f8_init_large_rows_interleaved_k128_c80_r56_r57",
      "dimensions": {},
      "sampleCount": 96,
      "tileCount": 96,
      "phaseCount": 7,
      "phaseCycles": 10794,
      "phaseMs": 0.007196,
      "workCycles": 809856,
      "averageActiveTiles": 75.0283490828238,
      "usefulCycles": 373977.0,
      "estimatedWorkCycles": 809856,
      "usefulUtilization": 0.46178209459459457,
      "workLaneOccupancy": 0.854296875,
      "deviceUsefulUtilization": 0.023537193066196196,
      "meanCycles": 8436.0,
      "p50Cycles": 8436,
      "p95Cycles": 8436,
      "maximumCycles": 8436,
      "firstOffset": 203406,
      "lastOffset": 214200
    },
    {
      "name": "cast_f16_f8",
      "dimensions": {},
      "sampleCount": 1248,
      "tileCount": 1248,
      "phaseCount": 21,
      "phaseCycles": 10542,
      "phaseMs": 0.007028,
      "workCycles": 8569152,
      "averageActiveTiles": 812.85828116107,
      "usefulCycles": 1364688.0,
      "estimatedWorkCycles": 8569152,
      "usefulUtilization": 0.15925589836660617,
      "workLaneOccupancy": 0.854296875,
      "deviceUsefulUtilization": 0.0879432580238054,
      "meanCycles": 6866.307692307692,
      "p50Cycles": 6894,
      "p95Cycles": 6894,
      "maximumCycles": 6894,
      "firstOffset": 195240,
      "lastOffset": 205782
    },
    {
      "name": "attention_merge_v72_p80_k768_out16_p8_f16",
      "dimensions": {},
      "sampleCount": 1472,
      "tileCount": 1472,
      "phaseCount": 34,
      "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": 219120,
      "lastOffset": 226098
    },
    {
      "name": "copy_strided_u64",
      "dimensions": {},
      "sampleCount": 1720,
      "tileCount": 1472,
      "phaseCount": 41,
      "phaseCycles": 5526,
      "phaseMs": 0.003684,
      "workCycles": 709056,
      "averageActiveTiles": 128.31270358306188,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 412.2418604651163,
      "p50Cycles": 420,
      "p95Cycles": 420,
      "maximumCycles": 480,
      "firstOffset": 119544,
      "lastOffset": 223110
    },
    {
      "name": "rearrange_row_major_to_amp_f16_o2_r64_p64_c16_p16_b64x16",
      "dimensions": {},
      "sampleCount": 880,
      "tileCount": 880,
      "phaseCount": 13,
      "phaseCycles": 4434,
      "phaseMs": 0.002956,
      "workCycles": 2143680,
      "averageActiveTiles": 483.4641407307172,
      "usefulCycles": 450560.0,
      "estimatedWorkCycles": 2143680,
      "usefulUtilization": 0.21018062397372742,
      "workLaneOccupancy": 1.0,
      "deviceUsefulUtilization": 0.06903178992371203,
      "meanCycles": 2436.0,
      "p50Cycles": 2436,
      "p95Cycles": 2436,
      "maximumCycles": 2436,
      "firstOffset": 118746,
      "lastOffset": 123180
    },
    {
      "name": "reduce_sum_f16",
      "dimensions": {},
      "sampleCount": 1472,
      "tileCount": 1472,
      "phaseCount": 29,
      "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": 216204,
      "lastOffset": 220344
    },
    {
      "name": "rearrange_row_major_to_amp_f16_o2_r0_p64_c16_p16_b64x16",
      "dimensions": {},
      "sampleCount": 80,
      "tileCount": 80,
      "phaseCount": 8,
      "phaseCycles": 3306,
      "phaseMs": 0.002204,
      "workCycles": 157440,
      "averageActiveTiles": 47.62250453720508,
      "usefulCycles": 16000.0,
      "estimatedWorkCycles": 157440,
      "usefulUtilization": 0.1016260162601626,
      "workLaneOccupancy": 0.390625,
      "deviceUsefulUtilization": 0.0032878297693258634,
      "meanCycles": 1968.0,
      "p50Cycles": 1968,
      "p95Cycles": 1968,
      "maximumCycles": 1968,
      "firstOffset": 118800,
      "lastOffset": 122106
    },
    {
      "name": "static_copy_u32",
      "dimensions": {},
      "sampleCount": 200,
      "tileCount": 190,
      "phaseCount": 11,
      "phaseCycles": 1512,
      "phaseMs": 0.001008,
      "workCycles": 59250,
      "averageActiveTiles": 39.18650793650794,
      "usefulCycles": 0.0,
      "estimatedWorkCycles": 0,
      "usefulUtilization": null,
      "workLaneOccupancy": null,
      "deviceUsefulUtilization": null,
      "meanCycles": 296.25,
      "p50Cycles": 294,
      "p95Cycles": 294,
      "maximumCycles": 492,
      "firstOffset": 119232,
      "lastOffset": 120744
    }
  ],
  "samples": []
}
