tiles=1472 matched=165805/255597 spanCycles=1296198 spanMs=0.864132 name="copy_u64" phases=231 tiles=1472 samples=33931 timelineCycles=214812 workCycles=33120114 range=15282..1294518 p95=2406 max=2874 useful=n/a coverage=0.0% name="reduce_sum_f16" phases=196 tiles=1472 samples=15033 timelineCycles=157332 workCycles=80785644 range=16260..1294236 p95=13764 max=14142 useful=56.6% coverage=100.0% name="cast_f16_f8" phases=188 tiles=1472 samples=18217 timelineCycles=99654 workCycles=102158256 range=0..1290516 p95=17400 max=17400 useful=20.9% coverage=95.0% name="gemm_f8_init_large_rows_interleaved_k128_c192_r52_r56" phases=57 tiles=756 samples=3780 timelineCycles=99288 workCycles=71555400 range=84420..1158096 p95=18930 max=18930 useful=56.8% coverage=100.0% name="gemm_f8_init_small_rows_interleaved_k128_c192_r52_r56" phases=47 tiles=702 samples=3510 timelineCycles=95316 workCycles=63790740 range=84486..1157316 p95=18174 max=18174 useful=54.8% coverage=100.0% name="add_f16" phases=199 tiles=1458 samples=17908 timelineCycles=72096 workCycles=81994416 range=29766..1296198 p95=5532 max=5532 useful=5.1% coverage=100.0% name="copy_strided_u64" phases=115 tiles=1472 samples=4177 timelineCycles=66180 workCycles=3115038 range=120996..1258680 p95=1098 max=1326 useful=n/a coverage=0.0% name="attention_softmax_d72_p768_tail_f16" phases=40 tiles=1472 samples=1504 timelineCycles=60228 workCycles=74181060 range=370878..1241838 p95=50250 max=50250 useful=49.5% coverage=100.0% name="fill_zero_u64" phases=121 tiles=1472 samples=15936 timelineCycles=60066 workCycles=6780096 range=295464..1254234 p95=960 max=1500 useful=0.0% coverage=41.3% name="gemm_f8_init_large_rows_interleaved_k192_c160_r160_r164" phases=24 tiles=780 samples=780 timelineCycles=49476 workCycles=38507040 range=667710..717186 p95=49368 max=49368 useful=79.7% coverage=100.0% name="gemm_f8_init_large_rows_interleaved_k192_c144_r180_r184" phases=26 tiles=920 samples=920 timelineCycles=48918 workCycles=44905200 range=895818..944736 p95=48810 max=48810 useful=79.4% coverage=100.0% name="gemm_f8_init_small_rows_interleaved_k192_c160_r160_r164" phases=19 tiles=624 samples=624 timelineCycles=48486 workCycles=30225312 range=667770..716256 p95=48438 max=48438 useful=79.0% coverage=100.0% name="gemm_f8_init_small_rows_interleaved_k192_c144_r180_r184" phases=19 tiles=552 samples=552 timelineCycles=48288 workCycles=26598672 range=895824..944112 p95=48186 max=48186 useful=78.3% coverage=100.0% name="gemm_f8_init_large_rows_interleaved_k192_c144_r160_r164" phases=16 tiles=30 samples=30 timelineCycles=44568 workCycles=1333980 range=666948..711516 p95=44466 max=44466 useful=79.7% coverage=100.0% name="bias_gelu_f16" phases=35 tiles=1458 samples=1458 timelineCycles=43764 workCycles=50178528 range=767916..811680 p95=34416 max=34416 useful=40.6% coverage=100.0% name="gemm_f8_init_small_rows_interleaved_k192_c144_r160_r164" phases=9 tiles=24 samples=24 timelineCycles=43680 workCycles=1047024 range=667002..710682 p95=43626 max=43626 useful=79.0% coverage=100.0% name="rearrange_row_major_to_amp_f16_o0_r92_p92_c80_p192" phases=14 tiles=40 samples=40 timelineCycles=41736 workCycles=1642080 range=829476..871212 p95=41052 max=41052 useful=17.9% coverage=100.0% name="rearrange_row_major_to_amp_f16_o0_r25_p26_c128_p128" phases=32 tiles=54 samples=270 timelineCycles=40452 workCycles=2008800 range=65418..1119936 p95=7440 max=7440 useful=43.0% coverage=100.0% name="rearrange_row_major_to_amp_f16_o0_r89_p90_c80_p192" phases=1 tiles=8 samples=8 timelineCycles=39030 workCycles=312192 range=819360..858390 p95=39024 max=39024 useful=18.2% coverage=100.0% name="rearrange_row_major_to_amp_f16_o0_r90_p90_c80_p192" phases=5 tiles=16 samples=16 timelineCycles=38964 workCycles=616224 range=829572..868536 p95=38514 max=38514 useful=18.7% coverage=100.0% name="rearrange_row_major_to_amp_f16_o0_r26_p26_c128_p128" phases=39 tiles=648 samples=3240 timelineCycles=38562 workCycles=21345120 range=65190..1119264 p95=6588 max=6588 useful=50.5% coverage=100.0% name="rearrange_row_major_to_amp_f16_o0_r28_p28_c128_p128" phases=57 tiles=756 samples=3780 timelineCycles=38394 workCycles=24925320 range=65352..1119336 p95=6594 max=6594 useful=54.4% coverage=100.0% name="layer_norm_f16" phases=38 tiles=1458 samples=2916 timelineCycles=34500 workCycles=45314640 range=40830..602160 p95=15540 max=15540 useful=29.7% coverage=100.0% name="gemm_f16_init_small_rows_interleaved_k96_c80_r182_r183" phases=10 tiles=768 samples=768 timelineCycles=30378 workCycles=20731392 range=437136..467514 p95=26994 max=26994 useful=69.1% coverage=100.0% name="rearrange_row_major_to_amp_f16_o0_r92_p92_c192_p192" phases=26 tiles=880 samples=880 timelineCycles=29604 workCycles=25312320 range=829854..859458 p95=28764 max=28764 useful=61.4% coverage=100.0% name="gemm_f16_init_large_rows_interleaved_k96_c80_r182_r183" phases=15 tiles=256 samples=256 timelineCycles=29544 workCycles=6931968 range=438222..467766 p95=27078 max=27078 useful=69.3% coverage=100.0% name="static_copy_u32" phases=155 tiles=1079 samples=3393 timelineCycles=29298 workCycles=665112 range=113292..1295394 p95=288 max=1068 useful=n/a coverage=0.0% name="gemm_f16_init_small_rows_interleaved_k80_c80_r182_r183" phases=7 tiles=192 samples=192 timelineCycles=28800 workCycles=4333824 range=334476..363276 p95=22572 max=22572 useful=72.6% coverage=100.0% name="rearrange_row_major_to_amp_f16_o0_r89_p90_c192_p192" phases=14 tiles=176 samples=176 timelineCycles=28602 workCycles=4975872 range=829260..857862 p95=28272 max=28272 useful=60.4% coverage=100.0% name="rearrange_row_major_to_amp_f16_o0_r90_p90_c192_p192" phases=14 tiles=352 samples=352 timelineCycles=27702 workCycles=9501888 range=829518..857220 p95=26994 max=26994 useful=64.0% coverage=100.0% name="rearrange_row_major_to_amp_f16_o0_r79_p80_c192_p192" phases=13 tiles=162 samples=162 timelineCycles=26718 workCycles=4292352 range=617652..644370 p95=26496 max=26496 useful=57.2% coverage=100.0% name="rearrange_row_major_to_amp_f16_o2_r128_p128_c16_p16_b128x16" phases=52 tiles=800 samples=1600 timelineCycles=26526 workCycles=6816000 range=309102..1222134 p95=4260 max=4260 useful=24.0% coverage=100.0% name="rearrange_row_major_to_amp_f16_o0_r80_p80_c192_p192" phases=14 tiles=486 samples=486 timelineCycles=25620 workCycles=12253032 range=617874..643494 p95=25212 max=25212 useful=60.9% coverage=100.0% name="rearrange_row_major_to_amp_f16_o0_r82_p82_c192_p192" phases=24 tiles=810 samples=810 timelineCycles=25386 workCycles=20426580 range=618222..643608 p95=25218 max=25218 useful=62.4% coverage=100.0% name="gemm_f16_init_large_rows_interleaved_k80_c80_r182_r183" phases=13 tiles=64 samples=64 timelineCycles=24840 workCycles=1448832 range=338538..363378 p95=22638 max=22638 useful=72.8% coverage=100.0% name="rearrange_row_major_to_amp_f16_o2_r89_p128_c16_p16_b128x16" phases=29 tiles=160 samples=320 timelineCycles=21870 workCycles=1213440 range=309912..1221204 p95=3792 max=3792 useful=18.8% coverage=100.0% name="gemm_f16_init_small_rows_interleaved_k80_c64_r182_r183" phases=9 tiles=864 samples=864 timelineCycles=21294 workCycles=15655680 range=331632..352926 p95=18120 max=18120 useful=71.4% coverage=100.0% name="add_layer_norm_f16" phases=36 tiles=1458 samples=1458 timelineCycles=20790 workCycles=26926344 range=998370..1019160 p95=18468 max=18468 useful=29.6% coverage=100.0% name="gemm_f16_init_large_rows_interleaved_k80_c64_r182_r183" phases=7 tiles=288 samples=288 timelineCycles=20688 workCycles=5234112 range=338124..358812 p95=18174 max=18174 useful=71.6% coverage=100.0% name="rearrange_row_major_to_amp_f16_o1_r64_p64_c80_p80" phases=34 tiles=352 samples=352 timelineCycles=15738 workCycles=4145856 range=1202136..1217874 p95=11778 max=11778 useful=21.7% coverage=100.0% name="gemm_f8_init_large_rows_interleaved_k64_c128_r160_r164" phases=23 tiles=810 samples=810 timelineCycles=13662 workCycles=10969020 range=526074..539736 p95=13542 max=13542 useful=77.5% coverage=100.0% name="gemm_f8_init_small_rows_interleaved_k64_c128_r160_r164" phases=20 tiles=648 samples=648 timelineCycles=13332 workCycles=8604144 range=526134..539466 p95=13278 max=13278 useful=76.9% coverage=100.0% name="rearrange_row_major_to_amp_f16_o0_r79_p80_c64_p64" phases=10 tiles=162 samples=162 timelineCycles=11484 workCycles=1757376 range=506652..518136 p95=10848 max=10848 useful=46.6% coverage=100.0% name="rearrange_row_major_to_amp_f16_o0_r82_p82_c64_p64" phases=23 tiles=810 samples=810 timelineCycles=11376 workCycles=8451540 range=506730..518106 p95=10434 max=10434 useful=50.3% coverage=100.0% name="rearrange_row_major_to_amp_f16_o0_r80_p80_c64_p64" phases=13 tiles=486 samples=486 timelineCycles=11256 workCycles=5068008 range=506418..517674 p95=10428 max=10428 useful=49.1% coverage=100.0% name="cast_f32_f16" phases=50 tiles=1472 samples=1504 timelineCycles=11034 workCycles=8766528 range=484740..1256478 p95=6000 max=6000 useful=9.6% coverage=100.0% name="gemm_f16_init_small_rows_interleaved_k80_c16_r1_r1" phases=53 tiles=1472 samples=1472 timelineCycles=9750 workCycles=1713408 range=1224186..1233972 p95=1164 max=1164 useful=1.5% coverage=100.0% name="rearrange_row_major_to_amp_f16_o1_r0_p64_c80_p80" phases=7 tiles=32 samples=32 timelineCycles=9624 workCycles=293184 range=1204284..1213908 p95=9162 max=9162 useful=10.9% coverage=100.0% name="attention_merge_v72_p80_k768_f16" phases=50 tiles=1472 samples=1504 timelineCycles=9462 workCycles=5439936 range=481074..1255878 p95=3666 max=3666 useful=n/a coverage=0.0% name="gemm_f8_init_small_rows_interleaved_k160_c16_r144_r144" phases=1 tiles=1472 samples=1472 timelineCycles=9396 workCycles=11667072 range=3810..13206 p95=7926 max=7926 useful=132.3% coverage=100.0% name="rearrange_row_major_to_amp_f16_o1_r64_p64_c32_p32" phases=14 tiles=704 samples=704 timelineCycles=9084 workCycles=2521728 range=305520..314604 p95=3582 max=3582 useful=28.6% coverage=100.0% name="unpack_amp_to_row_major_f16_o1_r9_p16_c36_p36" phases=1 tiles=32 samples=32 timelineCycles=8112 workCycles=257856 range=19830..27942 p95=8058 max=8058 useful=4.0% coverage=100.0% name="rearrange_row_major_to_amp_f16_o0_r16_p16_c72_p80" phases=14 tiles=1248 samples=1248 timelineCycles=7284 workCycles=2223936 range=303738..311022 p95=1782 max=1782 useful=32.3% coverage=100.0% name="unpack_amp_to_row_major_f16_o1_r16_p16_c36_p36" phases=1 tiles=1440 samples=1440 timelineCycles=7266 workCycles=10359360 range=19812..27078 p95=7194 max=7194 useful=8.0% coverage=100.0% name="gemm_f16_init_small_rows_interleaved_k96_c16_r1_r1" phases=48 tiles=1280 samples=1280 timelineCycles=6546 workCycles=1704960 range=1243686..1250232 p95=1332 max=1332 useful=1.5% coverage=100.0% name="rearrange_row_major_to_amp_f16_o0_r15_p15_c72_p80" phases=9 tiles=224 samples=224 timelineCycles=5868 workCycles=399168 range=299724..307794 p95=1782 max=1782 useful=30.3% coverage=100.0% name="rearrange_row_major_to_amp_f16_o1_r0_p64_c32_p32" phases=3 tiles=64 samples=64 timelineCycles=5214 workCycles=261504 range=306312..311526 p95=4086 max=4086 useful=9.8% coverage=100.0% name="rearrange_row_major_to_amp_f16_o1_r64_p64_c16_p16" phases=6 tiles=352 samples=352 timelineCycles=4584 workCycles=540672 range=307752..312336 p95=1536 max=1536 useful=33.3% coverage=100.0% name="rearrange_row_major_to_amp_f16_o0_r1_p2_c224_p224" phases=65 tiles=1368 samples=1368 timelineCycles=4164 workCycles=5368032 range=1284828..1288992 p95=3924 max=3924 useful=5.7% coverage=100.0% name="rearrange_row_major_to_amp_f16_o0_r1_p2_c48_p224" phases=12 tiles=72 samples=72 timelineCycles=3954 workCycles=282528 range=1284498..1288452 p95=3924 max=3924 useful=1.2% coverage=100.0% name="gemm_f8_init_small_rows_interleaved_k96_c16_r4_r4" phases=21 tiles=864 samples=864 timelineCycles=3054 workCycles=730944 range=1018368..1021422 p95=846 max=846 useful=2.8% coverage=100.0% name="rearrange_row_major_to_amp_f16_o0_r1_p2_c128_p128" phases=63 tiles=1467 samples=1467 timelineCycles=2628 workCycles=3644028 range=1273284..1275912 p95=2484 max=2484 useful=5.2% coverage=100.0% name="layer_norm_apply" phases=23 tiles=64 samples=64 timelineCycles=2502 workCycles=157056 range=1269876..1272378 p95=2454 max=2454 useful=2.9% coverage=100.0% name="rearrange_row_major_to_amp_f16_o0_r1_p1_c72_p80" phases=13 tiles=32 samples=32 timelineCycles=2358 workCycles=24576 range=1203738..1206096 p95=768 max=768 useful=4.7% coverage=100.0% name="gemm_f8_init_small_rows_interleaved_k224_c16_r4_r4" phases=69 tiles=1440 samples=1440 timelineCycles=2100 workCycles=2203200 range=1289946..1292046 p95=1530 max=1530 useful=3.5% coverage=100.0% name="rearrange_row_major_to_amp_f16_o0_r1_p2_c64_p64" phases=50 tiles=1296 samples=1296 timelineCycles=2094 workCycles=1975104 range=1259598..1261692 p95=1524 max=1524 useful=4.2% coverage=100.0% name="rearrange_row_major_to_amp_f16_o1_r0_p64_c16_p16" phases=3 tiles=32 samples=32 timelineCycles=1890 workCycles=56640 range=306750..308640 p95=1770 max=1770 useful=11.3% coverage=100.0% name="gemm_f8_init_small_rows_interleaved_k128_c32_r4_r4" phases=58 tiles=954 samples=954 timelineCycles=1842 workCycles=1625616 range=1277472..1279314 p95=1704 max=1704 useful=3.8% coverage=100.0% name="layer_norm_moments" phases=23 tiles=64 samples=64 timelineCycles=1260 workCycles=71424 range=1268202..1269462 p95=1116 max=1116 useful=6.5% coverage=100.0% name="gelu_tanh_approx_f16" phases=52 tiles=1024 samples=1024 timelineCycles=1176 workCycles=866688 range=1282704..1283880 p95=1062 max=1062 useful=3.0% coverage=100.0% name="gemm_f8_init_small_rows_interleaved_k128_c16_r4_r4" phases=39 tiles=513 samples=513 timelineCycles=1152 workCycles=520182 range=1277478..1278630 p95=1014 max=1014 useful=3.2% coverage=100.0% name="gemm_f8_init_small_rows_interleaved_k64_c16_r4_r4" phases=50 tiles=1296 samples=1296 timelineCycles=774 workCycles=870912 range=1263480..1264254 p95=672 max=672 useful=2.4% coverage=100.0%