2026-07-22T05:19:24.830819Z  INFO command{command="profile-query"}: starting command
{
  "clockHz": 1500000000,
  "tileCount": 1472,
  "sampleCount": 1114789,
  "matchedSampleCount": 1114789,
  "profileSpanCycles": 7647354,
  "profileSpanMs": 5.098236,
  "atOffset": null,
  "groups": [
    {
      "name": "(no kernel)",
      "dimensions": {},
      "sampleCount": 568404,
      "tileCount": 1472,
      "phaseCount": 405,
      "phaseCycles": 7902738,
      "phaseMs": 5.268492,
      "workCycles": 4433018730,
      "averageActiveTiles": 560.9471970347493,
      "meanCycles": 7799.063219118796,
      "p50Cycles": 294,
      "p95Cycles": 26976,
      "maximumCycles": 290844,
      "firstOffset": 0,
      "lastOffset": 7449186
    },
    {
      "name": "expand_f8_f143_to_f16",
      "dimensions": {},
      "sampleCount": 177876,
      "tileCount": 1472,
      "phaseCount": 124,
      "phaseCycles": 1834662,
      "phaseMs": 1.223108,
      "workCycles": 2096755632,
      "averageActiveTiles": 1142.856630812651,
      "meanCycles": 11787.737704918032,
      "p50Cycles": 12696,
      "p95Cycles": 12696,
      "maximumCycles": 12696,
      "firstOffset": 595110,
      "lastOffset": 7343436
    },
    {
      "name": "gemm_f16_f8w_accumulate_rows_27_inner_64_output_128",
      "dimensions": {},
      "sampleCount": 49572,
      "tileCount": 1472,
      "phaseCount": 34,
      "phaseCycles": 624000,
      "phaseMs": 0.416,
      "workCycles": 761425920,
      "averageActiveTiles": 1220.2338461538461,
      "meanCycles": 15360.0,
      "p50Cycles": 15360,
      "p95Cycles": 15360,
      "maximumCycles": 15360,
      "firstOffset": 644814,
      "lastOffset": 1940280
    },
    {
      "name": "gemm_f16_f8w_accumulate_rows_36_inner_64_output_128",
      "dimensions": {},
      "sampleCount": 46818,
      "tileCount": 1472,
      "phaseCount": 34,
      "phaseCycles": 618318,
      "phaseMs": 0.412212,
      "workCycles": 711259056,
      "averageActiveTiles": 1150.3127128758988,
      "meanCycles": 15192.0,
      "p50Cycles": 15192,
      "p95Cycles": 15192,
      "maximumCycles": 15192,
      "firstOffset": 4515276,
      "lastOffset": 5751102
    },
    {
      "name": "gemm_f16_f8w_accumulate_rows_36_inner_128_output_64",
      "dimensions": {},
      "sampleCount": 48114,
      "tileCount": 1458,
      "phaseCount": 33,
      "phaseCycles": 597930,
      "phaseMs": 0.39862,
      "workCycles": 727772364,
      "averageActiveTiles": 1217.1531182579902,
      "meanCycles": 15126.0,
      "p50Cycles": 15126,
      "p95Cycles": 15126,
      "maximumCycles": 15126,
      "firstOffset": 6000660,
      "lastOffset": 7358562
    },
    {
      "name": "attention_merge_large_query_middle_block_f16_q34_35_k31_32",
      "dimensions": {},
      "sampleCount": 20160,
      "tileCount": 960,
      "phaseCount": 8,
      "phaseCycles": 359982,
      "phaseMs": 0.239988,
      "workCycles": 285223506,
      "averageActiveTiles": 792.3271330233179,
      "meanCycles": 14147.991369047619,
      "p50Cycles": 14148,
      "p95Cycles": 14148,
      "maximumCycles": 14148,
      "firstOffset": 2535870,
      "lastOffset": 3345384
    },
    {
      "name": "attention_merge_small_query_middle_block_f16_q34_35_k31_32",
      "dimensions": {},
      "sampleCount": 8064,
      "tileCount": 384,
      "phaseCount": 8,
      "phaseCycles": 297324,
      "phaseMs": 0.198216,
      "workCycles": 114089472,
      "averageActiveTiles": 383.72103160188885,
      "meanCycles": 14148.0,
      "p50Cycles": 14148,
      "p95Cycles": 14148,
      "maximumCycles": 14148,
      "firstOffset": 2538828,
      "lastOffset": 3345372
    },
    {
      "name": "add_layer_norm_affine_f16",
      "dimensions": {},
      "sampleCount": 81,
      "tileCount": 81,
      "phaseCount": 1,
      "phaseCycles": 289416,
      "phaseMs": 0.192944,
      "workCycles": 23442210,
      "averageActiveTiles": 80.99832075628161,
      "meanCycles": 289410.0,
      "p50Cycles": 289410,
      "p95Cycles": 289410,
      "maximumCycles": 289410,
      "firstOffset": 4143198,
      "lastOffset": 4432614
    },
    {
      "name": "relayout_f16_c16_to_a16",
      "dimensions": {},
      "sampleCount": 7290,
      "tileCount": 108,
      "phaseCount": 18,
      "phaseCycles": 272730,
      "phaseMs": 0.18182,
      "workCycles": 25255476,
      "averageActiveTiles": 92.60248597514025,
      "meanCycles": 3464.4,
      "p50Cycles": 2874,
      "p95Cycles": 4350,
      "maximumCycles": 4350,
      "firstOffset": 208920,
      "lastOffset": 7460466
    },
    {
      "name": "layer_norm_affine_f16",
      "dimensions": {},
      "sampleCount": 81,
      "tileCount": 81,
      "phaseCount": 1,
      "phaseCycles": 237642,
      "phaseMs": 0.158428,
      "workCycles": 19248516,
      "averageActiveTiles": 80.99795490696089,
      "meanCycles": 237636.0,
      "p50Cycles": 237636,
      "p95Cycles": 237636,
      "maximumCycles": 237636,
      "firstOffset": 280542,
      "lastOffset": 518184
    },
    {
      "name": "attention_softmax_large_query_large_key_f16_q34_35_k31_32",
      "dimensions": {},
      "sampleCount": 15360,
      "tileCount": 960,
      "phaseCount": 6,
      "phaseCycles": 231264,
      "phaseMs": 0.154176,
      "workCycles": 176025444,
      "averageActiveTiles": 761.1450290577003,
      "meanCycles": 11459.98984375,
      "p50Cycles": 11460,
      "p95Cycles": 11460,
      "maximumCycles": 11460,
      "firstOffset": 2485200,
      "lastOffset": 3089790
    },
    {
      "name": "add_f16",
      "dimensions": {},
      "sampleCount": 1539,
      "tileCount": 1459,
      "phaseCount": 2,
      "phaseCycles": 200520,
      "phaseMs": 0.13368,
      "workCycles": 30639384,
      "averageActiveTiles": 152.7996409335727,
      "meanCycles": 19908.63157894737,
      "p50Cycles": 10632,
      "p95Cycles": 186888,
      "maximumCycles": 186888,
      "firstOffset": 184158,
      "lastOffset": 7647354
    },
    {
      "name": "attention_qk_init_large_rows_q34_35_k31_32",
      "dimensions": {},
      "sampleCount": 22080,
      "tileCount": 960,
      "phaseCount": 8,
      "phaseCycles": 195132,
      "phaseMs": 0.130088,
      "workCycles": 121219200,
      "averageActiveTiles": 621.2164073550213,
      "meanCycles": 5490.0,
      "p50Cycles": 5490,
      "p95Cycles": 5490,
      "maximumCycles": 5490,
      "firstOffset": 2479710,
      "lastOffset": 3350874
    },
    {
      "name": "attention_pv_init_large_rows_q34_35_k31_32",
      "dimensions": {},
      "sampleCount": 22080,
      "tileCount": 960,
      "phaseCount": 8,
      "phaseCycles": 194856,
      "phaseMs": 0.129904,
      "workCycles": 120954240,
      "averageActiveTiles": 620.7365439093485,
      "meanCycles": 5478.0,
      "p50Cycles": 5478,
      "p95Cycles": 5478,
      "maximumCycles": 5478,
      "firstOffset": 2496660,
      "lastOffset": 3368070
    },
    {
      "name": "attention_softmax_small_query_large_key_f16_q34_35_k31_32",
      "dimensions": {},
      "sampleCount": 6144,
      "tileCount": 384,
      "phaseCount": 6,
      "phaseCycles": 183468,
      "phaseMs": 0.122312,
      "workCycles": 70410138,
      "averageActiveTiles": 383.77339917587807,
      "meanCycles": 11459.9833984375,
      "p50Cycles": 11460,
      "p95Cycles": 11460,
      "maximumCycles": 11460,
      "firstOffset": 2488182,
      "lastOffset": 3089784
    },
    {
      "name": "gemm_f16_f8w_accumulate_rows_36_inner_64_output_64",
      "dimensions": {},
      "sampleCount": 24786,
      "tileCount": 1458,
      "phaseCount": 17,
      "phaseCycles": 182052,
      "phaseMs": 0.121368,
      "workCycles": 191248776,
      "averageActiveTiles": 1050.5173027486653,
      "meanCycles": 7716.0,
      "p50Cycles": 7716,
      "p95Cycles": 7716,
      "maximumCycles": 7716,
      "firstOffset": 3691158,
      "lastOffset": 4039530
    },
    {
      "name": "attention_unpack_head_f16",
      "dimensions": {},
      "sampleCount": 1344,
      "tileCount": 84,
      "phaseCount": 3,
      "phaseCycles": 174444,
      "phaseMs": 0.116296,
      "workCycles": 14652288,
      "averageActiveTiles": 83.99422164132902,
      "meanCycles": 10902.0,
      "p50Cycles": 10902,
      "p95Cycles": 10902,
      "maximumCycles": 10902,
      "firstOffset": 3407436,
      "lastOffset": 3596928
    },
    {
      "name": "add_bias_f16_c16",
      "dimensions": {},
      "sampleCount": 8586,
      "tileCount": 1472,
      "phaseCount": 4,
      "phaseCycles": 135264,
      "phaseMs": 0.090176,
      "workCycles": 175490712,
      "averageActiveTiles": 1297.3940738112137,
      "meanCycles": 20439.169811320753,
      "p50Cycles": 21972,
      "p95Cycles": 26316,
      "maximumCycles": 26316,
      "firstOffset": 1948278,
      "lastOffset": 7379502
    },
    {
      "name": "attention_qk_init_small_rows_q34_35_k31_32",
      "dimensions": {},
      "sampleCount": 8832,
      "tileCount": 384,
      "phaseCount": 8,
      "phaseCycles": 126324,
      "phaseMs": 0.084216,
      "workCycles": 48434688,
      "averageActiveTiles": 383.41635793673413,
      "meanCycles": 5484.0,
      "p50Cycles": 5484,
      "p95Cycles": 5484,
      "maximumCycles": 5484,
      "firstOffset": 2482698,
      "lastOffset": 3350856
    },
    {
      "name": "attention_pv_init_small_rows_q34_35_k31_32",
      "dimensions": {},
      "sampleCount": 8832,
      "tileCount": 384,
      "phaseCount": 8,
      "phaseCycles": 126096,
      "phaseMs": 0.084064,
      "workCycles": 48328704,
      "averageActiveTiles": 383.2691282832128,
      "meanCycles": 5472.0,
      "p50Cycles": 5472,
      "p95Cycles": 5472,
      "maximumCycles": 5472,
      "firstOffset": 2499636,
      "lastOffset": 3368046
    },
    {
      "name": "gelu_f16_c16_to_a16",
      "dimensions": {},
      "sampleCount": 2754,
      "tileCount": 1472,
      "phaseCount": 1,
      "phaseCycles": 116406,
      "phaseMs": 0.077604,
      "workCycles": 124045668,
      "averageActiveTiles": 1065.6295036338333,
      "meanCycles": 45042.0,
      "p50Cycles": 45042,
      "p95Cycles": 45042,
      "maximumCycles": 45042,
      "firstOffset": 5791158,
      "lastOffset": 5907564
    },
    {
      "name": "attention_softmax_large_query_small_key_f16_q34_35_k31_32",
      "dimensions": {},
      "sampleCount": 6720,
      "tileCount": 960,
      "phaseCount": 3,
      "phaseCycles": 102984,
      "phaseMs": 0.068656,
      "workCycles": 78737262,
      "averageActiveTiles": 764.5582032160336,
      "meanCycles": 11716.854464285714,
      "p50Cycles": 11718,
      "p95Cycles": 11718,
      "maximumCycles": 11718,
      "firstOffset": 3111912,
      "lastOffset": 3362592
    },
    {
      "name": "attention_pack_value_f16",
      "dimensions": {},
      "sampleCount": 3136,
      "tileCount": 1344,
      "phaseCount": 1,
      "phaseCycles": 102954,
      "phaseMs": 0.068636,
      "workCycles": 73625856,
      "averageActiveTiles": 715.1335159391573,
      "meanCycles": 23477.632653061224,
      "p50Cycles": 24018,
      "p95Cycles": 40236,
      "maximumCycles": 40248,
      "firstOffset": 2370888,
      "lastOffset": 2473842
    },
    {
      "name": "gemm_f16_accumulate_rows_36_inner_64_output_64",
      "dimensions": {},
      "sampleCount": 13122,
      "tileCount": 1458,
      "phaseCount": 5,
      "phaseCycles": 89052,
      "phaseMs": 0.059368,
      "workCycles": 108020304,
      "averageActiveTiles": 1213.0025603018462,
      "meanCycles": 8232.0,
      "p50Cycles": 8232,
      "p95Cycles": 8232,
      "maximumCycles": 8232,
      "firstOffset": 27210,
      "lastOffset": 180270
    },
    {
      "name": "attention_softmax_small_query_small_key_f16_q34_35_k31_32",
      "dimensions": {},
      "sampleCount": 2688,
      "tileCount": 384,
      "phaseCount": 3,
      "phaseCycles": 82110,
      "phaseMs": 0.05474,
      "workCycles": 31494996,
      "averageActiveTiles": 383.5707709170625,
      "meanCycles": 11716.888392857143,
      "p50Cycles": 11718,
      "p95Cycles": 11718,
      "maximumCycles": 11718,
      "firstOffset": 3114882,
      "lastOffset": 3362574
    },
    {
      "name": "copy_u64",
      "dimensions": {},
      "sampleCount": 14256,
      "tileCount": 1472,
      "phaseCount": 9,
      "phaseCycles": 81300,
      "phaseMs": 0.0542,
      "workCycles": 76935744,
      "averageActiveTiles": 946.3191143911439,
      "meanCycles": 5396.727272727273,
      "p50Cycles": 5436,
      "p95Cycles": 7164,
      "maximumCycles": 7164,
      "firstOffset": 180450,
      "lastOffset": 7365438
    },
    {
      "name": "reblock_f16_a16_to_a16",
      "dimensions": {},
      "sampleCount": 5850,
      "tileCount": 1472,
      "phaseCount": 2,
      "phaseCycles": 48288,
      "phaseMs": 0.032192,
      "workCycles": 48922056,
      "averageActiveTiles": 1013.1307157057654,
      "meanCycles": 8362.744615384616,
      "p50Cycles": 8388,
      "p95Cycles": 14772,
      "maximumCycles": 16140,
      "firstOffset": 560628,
      "lastOffset": 3660324
    },
    {
      "name": "attention_pack_key_f16",
      "dimensions": {},
      "sampleCount": 3136,
      "tileCount": 1344,
      "phaseCount": 1,
      "phaseCycles": 41580,
      "phaseMs": 0.02772,
      "workCycles": 28225152,
      "averageActiveTiles": 678.8155844155845,
      "meanCycles": 9000.367346938776,
      "p50Cycles": 8514,
      "p95Cycles": 15582,
      "maximumCycles": 15588,
      "firstOffset": 2289066,
      "lastOffset": 2330646
    },
    {
      "name": "gemm_f16_f8w_init_rows_27_inner_64_output_128",
      "dimensions": {},
      "sampleCount": 2916,
      "tileCount": 1472,
      "phaseCount": 2,
      "phaseCycles": 36336,
      "phaseMs": 0.024224,
      "workCycles": 44247384,
      "averageActiveTiles": 1217.728533685601,
      "meanCycles": 15174.0,
      "p50Cycles": 15174,
      "p95Cycles": 15174,
      "maximumCycles": 15174,
      "firstOffset": 607806,
      "lastOffset": 1303722
    },
    {
      "name": "gemm_f16_f8w_init_rows_36_inner_64_output_128",
      "dimensions": {},
      "sampleCount": 2754,
      "tileCount": 1472,
      "phaseCount": 2,
      "phaseCycles": 36084,
      "phaseMs": 0.024056,
      "workCycles": 41442192,
      "averageActiveTiles": 1148.4921849018956,
      "meanCycles": 15048.0,
      "p50Cycles": 15048,
      "p95Cycles": 15048,
      "maximumCycles": 15048,
      "firstOffset": 4480080,
      "lastOffset": 5149974
    }
  ],
  "samples": []
}
2026-07-22T05:19:31.255826Z  INFO command{command="profile-query"}: command completed
