Compiling ... Reading: /home/workspace/ByteMLPerf/new_models/model_1/mxr//model-static-batch-size-32.mxr Allocating params ... Running performance report ... main:@0 = check_context::migraphx::version_1::gpu::context -> float_type, {}, {}: 0.00160664ms, 0.014166% main:@1 = hip::hip_allocate_memory[shape=float_type, {2327616}, {1},id=main:scratch] -> float_type, {2327616}, {1}: 0.00108476ms, 0.00956449% uFSOc5Chp2:0 = @param:uFSOc5Chp2:0 -> float_type, {32, 481}, {481, 1}: 0.0006754ms, 0.0059551% main:@3 = slice[axes={0, 1},starts={0, 337},ends={32, 353}](uFSOc5Chp2:0) -> float_type, {32, 16}, {481, 1}: 0.00112694ms, 0.00993639% nv2tzWRf1I:0 = @param:nv2tzWRf1I:0 -> float_type, {32, 301}, {301, 1}: 0.00053396ms, 0.004708% lyiDxfzZg3:0 = @param:lyiDxfzZg3:0 -> float_type, {32, 701}, {701, 1}: 0.00045958ms, 0.00405218% main:@6 = slice[axes={0, 1},starts={0, 541},ends={32, 573}](lyiDxfzZg3:0) -> float_type, {32, 32}, {701, 1}: 0.00091166ms, 0.00803824% main:@7 = slice[axes={0, 1},starts={0, 285},ends={32, 301}](nv2tzWRf1I:0) -> float_type, {32, 16}, {301, 1}: 0.00072508ms, 0.00639313% n4iptOVFwE:0 = @param:n4iptOVFwE:0 -> float_type, {32, 113}, {113, 1}: 0.00047104ms, 0.00415323% main:@9 = slice[axes={0, 1},starts={0, 105},ends={32, 113}](n4iptOVFwE:0) -> float_type, {32, 8}, {113, 1}: 0.00069726ms, 0.00614784% main:@10 = load[offset=62976,end=72192](main:@1) -> float_type, {32, 72}, {72, 1}: 0.00056496ms, 0.00498133% main:@11 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=576,local=256,](main:@6,main:@7,main:@3,main:@9,main:@10) -> float_type, {32, 72}, {72, 1}: 0.0221757ms, 0.195527% main:@12 = hip::hip_copy_literal[id=main:@literal:150] -> float_type, {128, 16}, {16, 1}: 0.00097624ms, 0.00860765% vOI1hWzBDK:0 = @param:vOI1hWzBDK:0 -> int32_type, {32, 50}, {50, 1}: 0.00054202ms, 0.00477907% main:@14 = unsqueeze[axes={2},steps={}](vOI1hWzBDK:0) -> int32_type, {32, 50, 1}, {50, 1, 1}: 0.00088968ms, 0.00784444% main:@15 = load[offset=0,end=6400](main:@1) -> float_type, {32, 50, 1}, {50, 1, 1}: 0.00050796ms, 0.00447876% main:@16 = gpu::code_object[code_object=13520,symbol_name=convert_kernel,global=800,local=256,](main:@14,main:@15) -> float_type, {32, 50, 1}, {50, 1, 1}: 0.0137025ms, 0.120817% main:@17 = multibroadcast[out_lens={32, 50, 40}](main:@16) -> float_type, {32, 50, 40}, {50, 1, 0}: 0.00090266ms, 0.00795888% UrTfLOxlV3:0 = @param:UrTfLOxlV3:0 -> float_type, {32, 50, 41}, {2050, 41, 1}: 0.00051716ms, 0.00455987% main:@19 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 50, 41}](UrTfLOxlV3:0) -> float_type, {32, 50, 40}, {2050, 41, 1}: 0.00092066ms, 0.00811759% main:@20 = load[offset=301056,end=557056](main:@1) -> float_type, {32, 50, 40}, {2000, 40, 1}: 0.00047356ms, 0.00417545% main:@21 = gpu::code_object[code_object=13512,symbol_name=sub_kernel,global=64000,local=256,](main:@19,main:@17,main:@20) -> float_type, {32, 50, 40}, {2000, 40, 1}: 0.0185674ms, 0.163711% Ouj93S4aWb:0 = @param:Ouj93S4aWb:0 -> int32_type, {32, 60}, {60, 1}: 0.00066646ms, 0.00587627% main:@23 = unsqueeze[axes={2},steps={}](Ouj93S4aWb:0) -> int32_type, {32, 60, 1}, {60, 1, 1}: 0.00060234ms, 0.00531092% main:@24 = load[offset=98560,end=106240](main:@1) -> float_type, {32, 60, 1}, {60, 1, 1}: 0.00049028ms, 0.00432287% main:@25 = gpu::code_object[code_object=13520,symbol_name=convert_kernel,global=960,local=256,](main:@23,main:@24) -> float_type, {32, 60, 1}, {60, 1, 1}: 0.0135775ms, 0.119715% loNH8y6bTn:0 = @param:loNH8y6bTn:0 -> int32_type, {32, 15}, {15, 1}: 0.00063256ms, 0.00557737% main:@27 = unsqueeze[axes={2},steps={}](loNH8y6bTn:0) -> int32_type, {32, 15, 1}, {15, 1, 1}: 0.00060706ms, 0.00535254% main:@28 = load[offset=72192,end=74112](main:@1) -> float_type, {32, 15, 1}, {15, 1, 1}: 0.00047806ms, 0.00421512% main:@29 = gpu::code_object[code_object=13520,symbol_name=convert_kernel,global=240,local=256,](main:@27,main:@28) -> float_type, {32, 15, 1}, {15, 1, 1}: 0.0133254ms, 0.117492% I5iNBsTtoe:0 = @param:I5iNBsTtoe:0 -> int32_type, {32, 60}, {60, 1}: 0.00062278ms, 0.00549114% main:@31 = unsqueeze[axes={2},steps={}](I5iNBsTtoe:0) -> int32_type, {32, 60, 1}, {60, 1, 1}: 0.00058058ms, 0.00511906% main:@32 = load[offset=0,end=7680](main:@1) -> float_type, {32, 60, 1}, {60, 1, 1}: 0.00046458ms, 0.00409627% main:@33 = gpu::code_object[code_object=13520,symbol_name=convert_kernel,global=960,local=256,](main:@31,main:@32) -> float_type, {32, 60, 1}, {60, 1, 1}: 0.0134928ms, 0.118968% uVm4kOBe8l:0 = @param:uVm4kOBe8l:0 -> int32_type, {32, 50}, {50, 1}: 0.0006255ms, 0.00551512% main:@35 = unsqueeze[axes={2},steps={}](uVm4kOBe8l:0) -> int32_type, {32, 50, 1}, {50, 1, 1}: 0.00060308ms, 0.00531744% main:@36 = load[offset=682496,end=688896](main:@1) -> float_type, {32, 50, 1}, {50, 1, 1}: 0.00047646ms, 0.00420102% main:@37 = gpu::code_object[code_object=13520,symbol_name=convert_kernel,global=800,local=256,](main:@35,main:@36) -> float_type, {32, 50, 1}, {50, 1, 1}: 0.0133269ms, 0.117506% main:@38 = multibroadcast[out_lens={32, 60, 16}](main:@33) -> float_type, {32, 60, 16}, {60, 1, 0}: 0.00068604ms, 0.00604891% S3KJ2tl1VG:0 = @param:S3KJ2tl1VG:0 -> float_type, {32, 60, 17}, {1020, 17, 1}: 0.00050878ms, 0.00448599% main:@40 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 60, 17}](S3KJ2tl1VG:0) -> float_type, {32, 60, 16}, {1020, 17, 1}: 0.00076472ms, 0.00674265% main:@41 = load[offset=1038464,end=1161344](main:@1) -> float_type, {32, 60, 16}, {960, 16, 1}: 0.00046074ms, 0.00406241% main:@42 = gpu::code_object[code_object=13512,symbol_name=pow_kernel,global=30720,local=256,](main:@40,main:@38,main:@41) -> float_type, {32, 60, 16}, {960, 16, 1}: 0.0191832ms, 0.169141% KOJcBD1Cxa:0 = @param:KOJcBD1Cxa:0 -> int32_type, {32, 50}, {50, 1}: 0.00064622ms, 0.00569781% main:@44 = unsqueeze[axes={2},steps={}](KOJcBD1Cxa:0) -> int32_type, {32, 50, 1}, {50, 1, 1}: 0.0005891ms, 0.00519418% main:@45 = load[offset=0,end=6400](main:@1) -> float_type, {32, 50, 1}, {50, 1, 1}: 0.00051996ms, 0.00458456% main:@46 = gpu::code_object[code_object=13520,symbol_name=convert_kernel,global=800,local=256,](main:@44,main:@45) -> float_type, {32, 50, 1}, {50, 1, 1}: 0.013377ms, 0.117947% TWPHfwn7r5:0 = @param:TWPHfwn7r5:0 -> int32_type, {32, 50}, {50, 1}: 0.00067678ms, 0.00596727% main:@48 = unsqueeze[axes={2},steps={}](TWPHfwn7r5:0) -> int32_type, {32, 50, 1}, {50, 1, 1}: 0.00056474ms, 0.00497939% main:@49 = load[offset=6400,end=12800](main:@1) -> float_type, {32, 50, 1}, {50, 1, 1}: 0.00046124ms, 0.00406682% main:@50 = gpu::code_object[code_object=13520,symbol_name=convert_kernel,global=800,local=256,](main:@48,main:@49) -> float_type, {32, 50, 1}, {50, 1, 1}: 0.0134548ms, 0.118633% main:@51 = multibroadcast[out_lens={32, 50, 16}](main:@50) -> float_type, {32, 50, 16}, {50, 1, 0}: 0.0006552ms, 0.00577699% NoPO6DfgwG:0 = @param:NoPO6DfgwG:0 -> float_type, {32, 50, 17}, {850, 17, 1}: 0.00051342ms, 0.0045269% main:@53 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 50, 17}](NoPO6DfgwG:0) -> float_type, {32, 50, 16}, {850, 17, 1}: 0.00075198ms, 0.00663032% main:@54 = load[offset=1293440,end=1395840](main:@1) -> float_type, {32, 50, 16}, {800, 16, 1}: 0.0005048ms, 0.00445089% main:@55 = gpu::code_object[code_object=13512,symbol_name=sub_kernel,global=25600,local=256,](main:@53,main:@51,main:@54) -> float_type, {32, 50, 16}, {800, 16, 1}: 0.0173208ms, 0.15272% kJg5ca1rqS:0 = @param:kJg5ca1rqS:0 -> int32_type, {32, 15}, {15, 1}: 0.00062146ms, 0.0054795% main:@57 = unsqueeze[axes={2},steps={}](kJg5ca1rqS:0) -> int32_type, {32, 15, 1}, {15, 1, 1}: 0.00057908ms, 0.00510583% main:@58 = load[offset=6400,end=8320](main:@1) -> float_type, {32, 15, 1}, {15, 1, 1}: 0.00046122ms, 0.00406664% main:@59 = gpu::code_object[code_object=13520,symbol_name=convert_kernel,global=240,local=256,](main:@57,main:@58) -> float_type, {32, 15, 1}, {15, 1, 1}: 0.013156ms, 0.115998% main:@60 = multibroadcast[out_lens={32, 15, 28}](main:@59) -> float_type, {32, 15, 28}, {15, 1, 0}: 0.00066202ms, 0.00583713% fQkYLvqDjr:0 = @param:fQkYLvqDjr:0 -> float_type, {32, 15, 93}, {1395, 93, 1}: 0.00046718ms, 0.00411919% main:@62 = slice[axes={0, 1, 2},starts={0, 0, 65},ends={32, 15, 93}](fQkYLvqDjr:0) -> float_type, {32, 15, 28}, {1395, 93, 1}: 0.00072964ms, 0.00643334% main:@63 = load[offset=247296,end=301056](main:@1) -> float_type, {32, 15, 28}, {420, 28, 1}: 0.00046018ms, 0.00405747% main:@64 = gpu::code_object[code_object=13512,symbol_name=add_kernel,global=13440,local=256,](main:@62,main:@60,main:@63) -> float_type, {32, 15, 28}, {420, 28, 1}: 0.0154954ms, 0.136625% v3TAQfOsga:0 = @param:v3TAQfOsga:0 -> int32_type, {32, 15}, {15, 1}: 0.00066878ms, 0.00589673% main:@66 = unsqueeze[axes={2},steps={}](v3TAQfOsga:0) -> int32_type, {32, 15, 1}, {15, 1, 1}: 0.0005884ms, 0.00518801% main:@67 = load[offset=96640,end=98560](main:@1) -> float_type, {32, 15, 1}, {15, 1, 1}: 0.00047992ms, 0.00423152% main:@68 = gpu::code_object[code_object=13520,symbol_name=convert_kernel,global=240,local=256,](main:@66,main:@67) -> float_type, {32, 15, 1}, {15, 1, 1}: 0.0132625ms, 0.116937% yDES0Cj7Td:0 = @param:yDES0Cj7Td:0 -> int32_type, {32, 10}, {10, 1}: 0.00065664ms, 0.00578969% main:@70 = unsqueeze[axes={2},steps={}](yDES0Cj7Td:0) -> int32_type, {32, 10, 1}, {10, 1, 1}: 0.00057828ms, 0.00509878% main:@71 = load[offset=26880,end=28160](main:@1) -> float_type, {32, 10, 1}, {10, 1, 1}: 0.0004723ms, 0.00416434% main:@72 = gpu::code_object[code_object=13520,symbol_name=convert_kernel,global=160,local=256,](main:@70,main:@71) -> float_type, {32, 10, 1}, {10, 1, 1}: 0.0132811ms, 0.117101% main:@73 = multibroadcast[out_lens={32, 10, 16}](main:@72) -> float_type, {32, 10, 16}, {10, 1, 0}: 0.00065024ms, 0.00573326% r1GsND0okc:0 = @param:r1GsND0okc:0 -> float_type, {32, 10, 17}, {170, 17, 1}: 0.00048332ms, 0.0042615% main:@75 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 10, 17}](r1GsND0okc:0) -> float_type, {32, 10, 16}, {170, 17, 1}: 0.00071558ms, 0.00630937% main:@76 = load[offset=6400,end=26880](main:@1) -> float_type, {32, 10, 16}, {160, 16, 1}: 0.0004587ms, 0.00404442% main:@77 = gpu::code_object[code_object=13512,symbol_name=sub_kernel,global=5120,local=256,](main:@75,main:@73,main:@76) -> float_type, {32, 10, 16}, {160, 16, 1}: 0.0146651ms, 0.129304% main:@78 = load[offset=773248,end=1038464](main:@1) -> float_type, {32, 2072}, {2072, 1}: 0.00064138ms, 0.00565514% main:@79 = reshape_lazy[dims={-1, 2000}](main:@21) -> float_type, {32, 2000}, {2000, 1}: 0.00069794ms, 0.00615384% main:@80 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=33152,local=256,](main:@11,main:@79,main:@78) -> float_type, {32, 2072}, {2072, 1}: 0.0184006ms, 0.162241% main:@81 = hip::hip_copy_literal[id=main:@literal:85] -> float_type, {2072, 128}, {128, 1}: 0.00082142ms, 0.00724258% main:@82 = load[offset=74112,end=90496](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00046276ms, 0.00408022% main:@83 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@80,main:@81,main:@82) -> float_type, {32, 128}, {128, 1}: 0.0353665ms, 0.311831% vyDGpmRhC3:0 = @param:vyDGpmRhC3:0 -> int32_type, {32, 60}, {60, 1}: 0.0006586ms, 0.00580697% main:@85 = unsqueeze[axes={2},steps={}](vyDGpmRhC3:0) -> int32_type, {32, 60, 1}, {60, 1, 1}: 0.00060904ms, 0.00536999% main:@86 = load[offset=26880,end=34560](main:@1) -> float_type, {32, 60, 1}, {60, 1, 1}: 0.00048532ms, 0.00427914% main:@87 = gpu::code_object[code_object=13520,symbol_name=convert_kernel,global=960,local=256,](main:@85,main:@86) -> float_type, {32, 60, 1}, {60, 1, 1}: 0.0132671ms, 0.116978% main:@88 = multibroadcast[out_lens={32, 60, 16}](main:@87) -> float_type, {32, 60, 16}, {60, 1, 0}: 0.00068556ms, 0.00604468% main:@89 = load[offset=106240,end=229120](main:@1) -> float_type, {32, 60, 16}, {960, 16, 1}: 0.00052066ms, 0.00459073% y5fPVEH2j4:0 = @param:y5fPVEH2j4:0 -> float_type, {32, 60, 17}, {1020, 17, 1}: 0.0005438ms, 0.00479476% main:@91 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 60, 17}](y5fPVEH2j4:0) -> float_type, {32, 60, 16}, {1020, 17, 1}: 0.00085778ms, 0.00756317% main:@92 = gpu::code_object[code_object=13512,symbol_name=add_kernel,global=30720,local=256,](main:@91,main:@88,main:@89) -> float_type, {32, 60, 16}, {960, 16, 1}: 0.0181727ms, 0.160231% main:@93 = load[offset=409088,end=541184](main:@1) -> float_type, {32, 1032}, {1032, 1}: 0.00063518ms, 0.00560047% main:@94 = reshape_lazy[dims={-1, 960}](main:@92) -> float_type, {32, 960}, {960, 1}: 0.00052864ms, 0.0046611% main:@95 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=16512,local=256,](main:@11,main:@94,main:@93) -> float_type, {32, 1032}, {1032, 1}: 0.0160946ms, 0.141908% main:@96 = hip::hip_copy_literal[id=main:@literal:56] -> float_type, {1032, 128}, {128, 1}: 0.00078738ms, 0.00694244% main:@97 = load[offset=43264,end=59648](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00047938ms, 0.00422676% main:@98 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@95,main:@96,main:@97) -> float_type, {32, 128}, {128, 1}: 0.0269703ms, 0.237801% main:@99 = load[offset=26880,end=43264](main:@1) -> float_type, {32, 128}, {128, 1}: 0.0006562ms, 0.00578581% main:@100 = gpu::code_object[code_object=13520,symbol_name=log_sub_kernel,global=2048,local=256,](main:@98,main:@99) -> float_type, {32, 128}, {128, 1}: 0.0133835ms, 0.118005% main:@101 = hip::hip_copy_literal[id=main:@literal:90] -> float_type, {128, 16}, {16, 1}: 0.00074664ms, 0.00658323% main:@102 = load[offset=92544,end=94592](main:@1) -> float_type, {32, 16}, {16, 1}: 0.00046576ms, 0.00410667% main:@103 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@100,main:@101,main:@102) -> float_type, {32, 16}, {16, 1}: 0.0208957ms, 0.184241% main:@104 = load[offset=379392,end=409088](main:@1) -> float_type, {32, 232}, {232, 1}: 0.00062756ms, 0.00553329% main:@105 = reshape_lazy[dims={-1, 160}](main:@77) -> float_type, {32, 160}, {160, 1}: 0.00057164ms, 0.00504023% main:@106 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=3712,local=256,](main:@11,main:@105,main:@104) -> float_type, {32, 232}, {232, 1}: 0.0142164ms, 0.125348% main:@107 = hip::hip_copy_literal[id=main:@literal:132] -> float_type, {232, 128}, {128, 1}: 0.00076346ms, 0.00673154% main:@108 = load[offset=6400,end=22784](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00051332ms, 0.00452602% main:@109 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@106,main:@107,main:@108) -> float_type, {32, 128}, {128, 1}: 0.0246872ms, 0.21767% main:@110 = hip::hip_copy_literal[id=main:@literal:104] -> float_type, {128, 16}, {16, 1}: 0.00077054ms, 0.00679396% main:@111 = load[offset=22784,end=39168](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00047264ms, 0.00416733% main:@112 = gpu::code_object[code_object=13520,symbol_name=neg_div_kernel,global=2048,local=256,](main:@109,main:@111) -> float_type, {32, 128}, {128, 1}: 0.013369ms, 0.117876% main:@113 = load[offset=90496,end=92544](main:@1) -> float_type, {32, 16}, {16, 1}: 0.00064574ms, 0.00569358% main:@114 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@112,main:@110,main:@113) -> float_type, {32, 16}, {16, 1}: 0.0195292ms, 0.172192% main:@115 = multibroadcast[out_lens={32, 50, 20}](main:@46) -> float_type, {32, 50, 20}, {50, 1, 0}: 0.0007012ms, 0.00618258% main:@116 = load[offset=541184,end=669184](main:@1) -> float_type, {32, 50, 20}, {1000, 20, 1}: 0.00050446ms, 0.0044479% gkGLzseUSO:0 = @param:gkGLzseUSO:0 -> float_type, {32, 50, 21}, {1050, 21, 1}: 0.00049312ms, 0.00434791% main:@118 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 50, 21}](gkGLzseUSO:0) -> float_type, {32, 50, 20}, {1050, 21, 1}: 0.00068638ms, 0.00605191% main:@119 = gpu::code_object[code_object=13512,symbol_name=div_kernel,global=32000,local=256,](main:@118,main:@115,main:@116) -> float_type, {32, 50, 20}, {1000, 20, 1}: 0.0182117ms, 0.160575% main:@120 = load[offset=110080,end=247296](main:@1) -> float_type, {32, 1072}, {1072, 1}: 0.00062332ms, 0.0054959% main:@121 = reshape_lazy[dims={-1, 1000}](main:@119) -> float_type, {32, 1000}, {1000, 1}: 0.00054092ms, 0.00476937% main:@122 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=17152,local=256,](main:@11,main:@121,main:@120) -> float_type, {32, 1072}, {1072, 1}: 0.0163279ms, 0.143966% main:@123 = hip::hip_copy_literal[id=main:@literal:57] -> float_type, {1072, 128}, {128, 1}: 0.00080218ms, 0.00707294% main:@124 = load[offset=16384,end=32768](main:@1) -> float_type, {32, 128}, {128, 1}: 0.000464ms, 0.00409115% main:@125 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@122,main:@123,main:@124) -> float_type, {32, 128}, {128, 1}: 0.0259409ms, 0.228725% main:@126 = load[offset=0,end=16384](main:@1) -> float_type, {32, 128}, {128, 1}: 0.0006252ms, 0.00551248% main:@127 = gpu::code_object[code_object=13520,symbol_name=neg_add_kernel,global=2048,local=256,](main:@125,main:@126) -> float_type, {32, 128}, {128, 1}: 0.0132006ms, 0.116392% main:@128 = load[offset=94592,end=96640](main:@1) -> float_type, {32, 16}, {16, 1}: 0.00061192ms, 0.00539539% main:@129 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@127,main:@110,main:@128) -> float_type, {32, 16}, {16, 1}: 0.0195496ms, 0.172372% main:@130 = reshape_lazy[dims={-1, 420}](main:@64) -> float_type, {32, 420}, {420, 1}: 0.00071248ms, 0.00628204% main:@131 = load[offset=0,end=62976](main:@1) -> float_type, {32, 492}, {492, 1}: 0.00051554ms, 0.00454559% main:@132 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=7872,local=256,](main:@11,main:@130,main:@131) -> float_type, {32, 492}, {492, 1}: 0.0151517ms, 0.133595% main:@133 = hip::hip_copy_literal[id=main:@literal:127] -> float_type, {492, 128}, {128, 1}: 0.0007417ms, 0.00653968% main:@134 = load[offset=1395840,end=1412224](main:@1) -> float_type, {32, 128}, {128, 1}: 0.0004731ms, 0.00417139% main:@135 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@132,main:@133,main:@134) -> float_type, {32, 128}, {128, 1}: 0.0330192ms, 0.291135% CedNyHXqE1:0 = @param:CedNyHXqE1:0 -> int32_type, {32, 50}, {50, 1}: 0.00070556ms, 0.00622102% main:@137 = unsqueeze[axes={2},steps={}](CedNyHXqE1:0) -> int32_type, {32, 50, 1}, {50, 1, 1}: 0.0006052ms, 0.00533614% main:@138 = load[offset=247296,end=253696](main:@1) -> float_type, {32, 50, 1}, {50, 1, 1}: 0.00048486ms, 0.00427508% main:@139 = gpu::code_object[code_object=13520,symbol_name=convert_kernel,global=800,local=256,](main:@137,main:@138) -> float_type, {32, 50, 1}, {50, 1, 1}: 0.0133865ms, 0.118031% YsGAJ1iRPS:0 = @param:YsGAJ1iRPS:0 -> float_type, {32, 50, 17}, {850, 17, 1}: 0.00069272ms, 0.00610781% main:@141 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 50, 17}](YsGAJ1iRPS:0) -> float_type, {32, 50, 16}, {850, 17, 1}: 0.00084618ms, 0.00746089% main:@142 = multibroadcast[out_lens={32, 50, 16}](main:@139) -> float_type, {32, 50, 16}, {50, 1, 0}: 0.00052654ms, 0.00464258% main:@143 = load[offset=253696,end=356096](main:@1) -> float_type, {32, 50, 16}, {800, 16, 1}: 0.00047518ms, 0.00418973% main:@144 = gpu::code_object[code_object=13512,symbol_name=add_kernel,global=25600,local=256,](main:@141,main:@142,main:@143) -> float_type, {32, 50, 16}, {800, 16, 1}: 0.0174259ms, 0.153647% main:@145 = load[offset=570880,end=682496](main:@1) -> float_type, {32, 872}, {872, 1}: 0.00062726ms, 0.00553064% main:@146 = reshape_lazy[dims={-1, 800}](main:@144) -> float_type, {32, 800}, {800, 1}: 0.00050996ms, 0.00449639% main:@147 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=13952,local=256,](main:@11,main:@146,main:@145) -> float_type, {32, 872}, {872, 1}: 0.015989ms, 0.140977% main:@148 = hip::hip_copy_literal[id=main:@literal:159] -> float_type, {872, 128}, {128, 1}: 0.00074388ms, 0.0065589% main:@149 = load[offset=263680,end=280064](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00049234ms, 0.00434103% main:@150 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@147,main:@148,main:@149) -> float_type, {32, 128}, {128, 1}: 0.0247978ms, 0.218646% main:@151 = load[offset=247296,end=263680](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00062274ms, 0.00549079% main:@152 = gpu::code_object[code_object=13520,symbol_name=neg_add_kernel,global=2048,local=256,](main:@150,main:@151) -> float_type, {32, 128}, {128, 1}: 0.0132053ms, 0.116433% main:@153 = hip::hip_copy_literal[id=main:@literal:64] -> float_type, {128, 16}, {16, 1}: 0.0007712ms, 0.00679978% main:@154 = load[offset=106240,end=108288](main:@1) -> float_type, {32, 16}, {16, 1}: 0.0004773ms, 0.00420842% main:@155 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@152,main:@153,main:@154) -> float_type, {32, 16}, {16, 1}: 0.0205172ms, 0.180903% YK10Jin6ca:0 = @param:YK10Jin6ca:0 -> int32_type, {32, 10}, {10, 1}: 0.0007214ms, 0.00636069% main:@157 = unsqueeze[axes={2},steps={}](YK10Jin6ca:0) -> int32_type, {32, 10, 1}, {10, 1, 1}: 0.00059296ms, 0.00522821% main:@158 = load[offset=108288,end=109568](main:@1) -> float_type, {32, 10, 1}, {10, 1, 1}: 0.00047228ms, 0.00416416% main:@159 = gpu::code_object[code_object=13520,symbol_name=convert_kernel,global=160,local=256,](main:@157,main:@158) -> float_type, {32, 10, 1}, {10, 1, 1}: 0.0133127ms, 0.11738% main:@160 = multibroadcast[out_lens={32, 10, 16}](main:@159) -> float_type, {32, 10, 16}, {10, 1, 0}: 0.00067496ms, 0.00595122% rlh04eiWzL:0 = @param:rlh04eiWzL:0 -> float_type, {32, 10, 17}, {170, 17, 1}: 0.00049152ms, 0.0043338% main:@162 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 10, 17}](rlh04eiWzL:0) -> float_type, {32, 10, 16}, {170, 17, 1}: 0.00092048ms, 0.008116% main:@163 = load[offset=247296,end=267776](main:@1) -> float_type, {32, 10, 16}, {160, 16, 1}: 0.00049144ms, 0.0043331% main:@164 = gpu::code_object[code_object=13512,symbol_name=pow_kernel,global=5120,local=256,](main:@162,main:@160,main:@163) -> float_type, {32, 10, 16}, {160, 16, 1}: 0.015669ms, 0.138156% main:@165 = load[offset=541184,end=570880](main:@1) -> float_type, {32, 232}, {232, 1}: 0.00060522ms, 0.00533631% main:@166 = reshape_lazy[dims={-1, 160}](main:@164) -> float_type, {32, 160}, {160, 1}: 0.00052854ms, 0.00466021% main:@167 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=3712,local=256,](main:@11,main:@166,main:@165) -> float_type, {32, 232}, {232, 1}: 0.0142496ms, 0.125641% main:@168 = hip::hip_copy_literal[id=main:@literal:117] -> float_type, {232, 128}, {128, 1}: 0.00075566ms, 0.00666276% main:@169 = load[offset=712192,end=728576](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00048322ms, 0.00426062% main:@170 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@167,main:@168,main:@169) -> float_type, {32, 128}, {128, 1}: 0.0235959ms, 0.208048% main:@171 = reshape_lazy[dims={-1, 960}](main:@42) -> float_type, {32, 960}, {960, 1}: 0.00066982ms, 0.0059059% main:@172 = load[offset=247296,end=379392](main:@1) -> float_type, {32, 1032}, {1032, 1}: 0.0005175ms, 0.00456287% main:@173 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=16512,local=256,](main:@11,main:@171,main:@172) -> float_type, {32, 1032}, {1032, 1}: 0.0162023ms, 0.142858% main:@174 = hip::hip_copy_literal[id=main:@literal:151] -> float_type, {1032, 128}, {128, 1}: 0.00081082ms, 0.00714912% main:@175 = load[offset=728576,end=744960](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00048642ms, 0.00428884% main:@176 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@173,main:@174,main:@175) -> float_type, {32, 128}, {128, 1}: 0.0247088ms, 0.217861% main:@177 = load[offset=688896,end=705280](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00064676ms, 0.00570258% main:@178 = gpu::code_object[code_object=13520,symbol_name=neg_sub_kernel,global=2048,local=256,](main:@176,main:@177) -> float_type, {32, 128}, {128, 1}: 0.0131596ms, 0.11603% main:@179 = hip::hip_copy_literal[id=main:@literal:108] -> float_type, {128, 16}, {16, 1}: 0.00083664ms, 0.00737678% main:@180 = load[offset=759808,end=761856](main:@1) -> float_type, {32, 16}, {16, 1}: 0.00048818ms, 0.00430435% main:@181 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@178,main:@179,main:@180) -> float_type, {32, 16}, {16, 1}: 0.0198561ms, 0.175074% FkwYpLOMG4:0 = @param:FkwYpLOMG4:0 -> float_type, {32, 50, 17}, {850, 17, 1}: 0.00069242ms, 0.00610517% main:@183 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 50, 17}](FkwYpLOMG4:0) -> float_type, {32, 50, 16}, {850, 17, 1}: 0.00070844ms, 0.00624642% main:@184 = load[offset=1038464,end=1140864](main:@1) -> float_type, {32, 50, 16}, {800, 16, 1}: 0.00046672ms, 0.00411514% main:@185 = multibroadcast[out_lens={32, 50, 16}](main:@37) -> float_type, {32, 50, 16}, {50, 1, 0}: 0.00053776ms, 0.00474151% main:@186 = gpu::code_object[code_object=13512,symbol_name=div_kernel,global=25600,local=256,](main:@183,main:@185,main:@184) -> float_type, {32, 50, 16}, {800, 16, 1}: 0.0173622ms, 0.153085% main:@187 = load[offset=3132544,end=3244160](main:@1) -> float_type, {32, 872}, {872, 1}: 0.00061136ms, 0.00539045% main:@188 = reshape_lazy[dims={-1, 800}](main:@186) -> float_type, {32, 800}, {800, 1}: 0.00052776ms, 0.00465334% main:@189 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=13952,local=256,](main:@11,main:@188,main:@187) -> float_type, {32, 872}, {872, 1}: 0.0158249ms, 0.139531% main:@190 = hip::hip_copy_literal[id=main:@literal:89] -> float_type, {872, 128}, {128, 1}: 0.00076162ms, 0.00671531% main:@191 = load[offset=728576,end=744960](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00046846ms, 0.00413048% main:@192 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@189,main:@190,main:@191) -> float_type, {32, 128}, {128, 1}: 0.0246272ms, 0.217142% main:@193 = multibroadcast[out_lens={32, 60, 32}](main:@25) -> float_type, {32, 60, 32}, {60, 1, 0}: 0.00071082ms, 0.0062674% fsm0Ck1QGl:0 = @param:fsm0Ck1QGl:0 -> float_type, {32, 60, 97}, {5820, 97, 1}: 0.0005425ms, 0.0047833% main:@195 = slice[axes={0, 1, 2},starts={0, 0, 65},ends={32, 60, 97}](fsm0Ck1QGl:0) -> float_type, {32, 60, 32}, {5820, 97, 1}: 0.00072754ms, 0.00641483% main:@196 = load[offset=1412224,end=1657984](main:@1) -> float_type, {32, 60, 32}, {1920, 32, 1}: 0.00046632ms, 0.00411161% main:@197 = gpu::code_object[code_object=13512,symbol_name=div_kernel,global=61440,local=256,](main:@195,main:@193,main:@196) -> float_type, {32, 60, 32}, {1920, 32, 1}: 0.0184331ms, 0.162527% main:@198 = reshape_lazy[dims={-1, 1920}](main:@197) -> float_type, {32, 1920}, {1920, 1}: 0.000684ms, 0.00603093% main:@199 = load[offset=1038464,end=1293440](main:@1) -> float_type, {32, 1992}, {1992, 1}: 0.00048194ms, 0.00424933% main:@200 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=31872,local=256,](main:@11,main:@198,main:@199) -> float_type, {32, 1992}, {1992, 1}: 0.0184361ms, 0.162553% main:@201 = hip::hip_copy_literal[id=main:@literal:109] -> float_type, {1992, 128}, {128, 1}: 0.0007842ms, 0.0069144% main:@202 = load[offset=1441920,end=1458304](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00048752ms, 0.00429853% main:@203 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@200,main:@201,main:@202) -> float_type, {32, 128}, {128, 1}: 0.0284097ms, 0.250492% 0MZVmrKeC6:0 = @param:0MZVmrKeC6:0 -> int32_type, {32, 10}, {10, 1}: 0.00072398ms, 0.00638344% main:@205 = unsqueeze[axes={2},steps={}](0MZVmrKeC6:0) -> int32_type, {32, 10, 1}, {10, 1, 1}: 0.0005861ms, 0.00516773% main:@206 = load[offset=98560,end=99840](main:@1) -> float_type, {32, 10, 1}, {10, 1, 1}: 0.00046368ms, 0.00408833% main:@207 = gpu::code_object[code_object=13520,symbol_name=convert_kernel,global=160,local=256,](main:@205,main:@206) -> float_type, {32, 10, 1}, {10, 1, 1}: 0.0131964ms, 0.116354% main:@208 = multibroadcast[out_lens={32, 10, 16}](main:@207) -> float_type, {32, 10, 16}, {10, 1, 0}: 0.00068816ms, 0.00606761% WsiwEgbYlT:0 = @param:WsiwEgbYlT:0 -> float_type, {32, 10, 17}, {170, 17, 1}: 0.00049526ms, 0.00436678% main:@210 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 10, 17}](WsiwEgbYlT:0) -> float_type, {32, 10, 16}, {170, 17, 1}: 0.00087942ms, 0.00775397% main:@211 = load[offset=1412224,end=1432704](main:@1) -> float_type, {32, 10, 16}, {160, 16, 1}: 0.00046304ms, 0.00408269% main:@212 = gpu::code_object[code_object=13512,symbol_name=add_kernel,global=5120,local=256,](main:@210,main:@208,main:@211) -> float_type, {32, 10, 16}, {160, 16, 1}: 0.0144737ms, 0.127617% main:@213 = load[offset=682496,end=712192](main:@1) -> float_type, {32, 232}, {232, 1}: 0.00062272ms, 0.00549061% main:@214 = reshape_lazy[dims={-1, 160}](main:@212) -> float_type, {32, 160}, {160, 1}: 0.00050824ms, 0.00448123% main:@215 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=3712,local=256,](main:@11,main:@214,main:@213) -> float_type, {32, 232}, {232, 1}: 0.0142959ms, 0.126049% main:@216 = hip::hip_copy_literal[id=main:@literal:157] -> float_type, {232, 128}, {128, 1}: 0.00075458ms, 0.00665324% main:@217 = load[offset=1458304,end=1474688](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00049106ms, 0.00432975% main:@218 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@215,main:@216,main:@217) -> float_type, {32, 128}, {128, 1}: 0.0243545ms, 0.214737% main:@219 = load[offset=1412224,end=1428608](main:@1) -> float_type, {32, 128}, {128, 1}: 0.0006329ms, 0.00558037% main:@220 = gpu::code_object[code_object=13520,symbol_name=log_add_kernel,global=2048,local=256,](main:@218,main:@219) -> float_type, {32, 128}, {128, 1}: 0.0132778ms, 0.117072% main:@221 = hip::hip_copy_literal[id=main:@literal:121] -> float_type, {128, 16}, {16, 1}: 0.00074028ms, 0.00652716% main:@222 = load[offset=100608,end=102656](main:@1) -> float_type, {32, 16}, {16, 1}: 0.00050808ms, 0.00447981% main:@223 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@220,main:@221,main:@222) -> float_type, {32, 16}, {16, 1}: 0.0205561ms, 0.181246% main:@224 = load[offset=1412224,end=1428608](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00062328ms, 0.00549555% main:@225 = gpu::code_object[code_object=13520,symbol_name=rsqrt_pow_kernel,global=2048,local=256,](main:@170,main:@224) -> float_type, {32, 128}, {128, 1}: 0.0163329ms, 0.144009% main:@226 = load[offset=98560,end=100608](main:@1) -> float_type, {32, 16}, {16, 1}: 0.00061054ms, 0.00538322% main:@227 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@225,main:@12,main:@226) -> float_type, {32, 16}, {16, 1}: 0.0200817ms, 0.177063% main:@228 = load[offset=712192,end=728576](main:@1) -> float_type, {32, 128}, {128, 1}: 0.000646ms, 0.00569588% main:@229 = gpu::code_object[code_object=13520,symbol_name=neg_div_kernel,global=2048,local=256,](main:@192,main:@228) -> float_type, {32, 128}, {128, 1}: 0.0133226ms, 0.117468% main:@230 = hip::hip_copy_literal[id=main:@literal:100] -> float_type, {128, 16}, {16, 1}: 0.00076812ms, 0.00677262% main:@231 = load[offset=102656,end=104704](main:@1) -> float_type, {32, 16}, {16, 1}: 0.00047096ms, 0.00415252% main:@232 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@229,main:@230,main:@231) -> float_type, {32, 16}, {16, 1}: 0.0198693ms, 0.17519% fFjVmAbhCX:0 = @param:fFjVmAbhCX:0 -> int32_type, {32, 10}, {10, 1}: 0.00068944ms, 0.00607889% main:@234 = unsqueeze[axes={2},steps={}](fFjVmAbhCX:0) -> int32_type, {32, 10, 1}, {10, 1, 1}: 0.00059678ms, 0.0052619% main:@235 = load[offset=104704,end=105984](main:@1) -> float_type, {32, 10, 1}, {10, 1, 1}: 0.00047556ms, 0.00419308% main:@236 = gpu::code_object[code_object=13520,symbol_name=convert_kernel,global=160,local=256,](main:@234,main:@235) -> float_type, {32, 10, 1}, {10, 1, 1}: 0.0132575ms, 0.116893% main:@237 = load[offset=712192,end=732672](main:@1) -> float_type, {32, 10, 16}, {160, 16, 1}: 0.00069052ms, 0.00608841% pjwGhTSAys:0 = @param:pjwGhTSAys:0 -> float_type, {32, 10, 17}, {170, 17, 1}: 0.00048102ms, 0.00424122% main:@239 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 10, 17}](pjwGhTSAys:0) -> float_type, {32, 10, 16}, {170, 17, 1}: 0.00095744ms, 0.00844189% main:@240 = multibroadcast[out_lens={32, 10, 16}](main:@236) -> float_type, {32, 10, 16}, {10, 1, 0}: 0.00051756ms, 0.0045634% main:@241 = gpu::code_object[code_object=13512,symbol_name=pow_kernel,global=5120,local=256,](main:@239,main:@240,main:@237) -> float_type, {32, 10, 16}, {160, 16, 1}: 0.0159377ms, 0.140525% main:@242 = reshape_lazy[dims={-1, 160}](main:@241) -> float_type, {32, 160}, {160, 1}: 0.00067444ms, 0.00594663% main:@243 = load[offset=1412224,end=1441920](main:@1) -> float_type, {32, 232}, {232, 1}: 0.0004653ms, 0.00410262% main:@244 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=3712,local=256,](main:@11,main:@242,main:@243) -> float_type, {32, 232}, {232, 1}: 0.0141547ms, 0.124804% main:@245 = hip::hip_copy_literal[id=main:@literal:130] -> float_type, {232, 128}, {128, 1}: 0.00080062ms, 0.00705918% main:@246 = load[offset=728576,end=744960](main:@1) -> float_type, {32, 128}, {128, 1}: 0.0004712ms, 0.00415464% main:@247 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@244,main:@245,main:@246) -> float_type, {32, 128}, {128, 1}: 0.0239737ms, 0.21138% main:@248 = load[offset=712192,end=728576](main:@1) -> float_type, {32, 128}, {128, 1}: 0.0006591ms, 0.00581138% main:@249 = gpu::code_object[code_object=13520,symbol_name=log_add_kernel,global=2048,local=256,](main:@247,main:@248) -> float_type, {32, 128}, {128, 1}: 0.0133487ms, 0.117697% main:@250 = load[offset=761856,end=763904](main:@1) -> float_type, {32, 16}, {16, 1}: 0.00062182ms, 0.00548268% main:@251 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@249,main:@12,main:@250) -> float_type, {32, 16}, {16, 1}: 0.0202042ms, 0.178143% main:@252 = multibroadcast[out_lens={32, 15, 20}](main:@68) -> float_type, {32, 15, 20}, {15, 1, 0}: 0.00070834ms, 0.00624554% 1qyHnroO6I:0 = @param:1qyHnroO6I:0 -> float_type, {32, 15, 21}, {315, 21, 1}: 0.00054504ms, 0.0048057% main:@254 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 15, 21}](1qyHnroO6I:0) -> float_type, {32, 15, 20}, {315, 21, 1}: 0.00085976ms, 0.00758063% main:@255 = load[offset=1458304,end=1496704](main:@1) -> float_type, {32, 15, 20}, {300, 20, 1}: 0.00046618ms, 0.00411038% main:@256 = gpu::code_object[code_object=13512,symbol_name=add_kernel,global=9600,local=256,](main:@254,main:@252,main:@255) -> float_type, {32, 15, 20}, {300, 20, 1}: 0.0150177ms, 0.132413% main:@257 = reshape_lazy[dims={-1, 300}](main:@256) -> float_type, {32, 300}, {300, 1}: 0.00066312ms, 0.00584682% main:@258 = load[offset=712192,end=759808](main:@1) -> float_type, {32, 372}, {372, 1}: 0.00047094ms, 0.00415235% main:@259 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=5952,local=256,](main:@11,main:@257,main:@258) -> float_type, {32, 372}, {372, 1}: 0.014485ms, 0.127717% main:@260 = hip::hip_copy_literal[id=main:@literal:105] -> float_type, {372, 128}, {128, 1}: 0.00073764ms, 0.00650388% main:@261 = load[offset=1474688,end=1491072](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00048628ms, 0.0042876% main:@262 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@259,main:@260,main:@261) -> float_type, {32, 128}, {128, 1}: 0.0280153ms, 0.247015% main:@263 = load[offset=1458304,end=1474688](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00062514ms, 0.00551195% main:@264 = gpu::code_object[code_object=13520,symbol_name=log_add_kernel,global=2048,local=256,](main:@262,main:@263) -> float_type, {32, 128}, {128, 1}: 0.013268ms, 0.116986% main:@265 = load[offset=763904,end=765952](main:@1) -> float_type, {32, 16}, {16, 1}: 0.00061482ms, 0.00542096% main:@266 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@264,main:@12,main:@265) -> float_type, {32, 16}, {16, 1}: 0.0194955ms, 0.171895% main:@267 = load[offset=1458304,end=1474688](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00065266ms, 0.0057546% main:@268 = gpu::code_object[code_object=13520,symbol_name=rsqrt_pow_kernel,global=2048,local=256,](main:@135,main:@267) -> float_type, {32, 128}, {128, 1}: 0.016276ms, 0.143508% main:@269 = load[offset=765952,end=768000](main:@1) -> float_type, {32, 16}, {16, 1}: 0.00062514ms, 0.00551195% main:@270 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@268,main:@12,main:@269) -> float_type, {32, 16}, {16, 1}: 0.0199163ms, 0.175605% main:@271 = load[offset=1395840,end=1412224](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00061944ms, 0.00546169% main:@272 = gpu::code_object[code_object=13520,symbol_name=neg_pow_kernel,global=2048,local=256,](main:@203,main:@271) -> float_type, {32, 128}, {128, 1}: 0.0162879ms, 0.143612% main:@273 = load[offset=768000,end=770048](main:@1) -> float_type, {32, 16}, {16, 1}: 0.00061634ms, 0.00543436% main:@274 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@272,main:@12,main:@273) -> float_type, {32, 16}, {16, 1}: 0.0201661ms, 0.177808% main:@275 = load[offset=1441920,end=1553536](main:@1) -> float_type, {32, 872}, {872, 1}: 0.00063378ms, 0.00558813% main:@276 = reshape_lazy[dims={-1, 800}](main:@55) -> float_type, {32, 800}, {800, 1}: 0.00055444ms, 0.00488858% main:@277 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=13952,local=256,](main:@11,main:@276,main:@275) -> float_type, {32, 872}, {872, 1}: 0.0160798ms, 0.141778% main:@278 = hip::hip_copy_literal[id=main:@literal:61] -> float_type, {872, 128}, {128, 1}: 0.00078664ms, 0.00693592% main:@279 = load[offset=1309824,end=1326208](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00047474ms, 0.00418585% main:@280 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@277,main:@278,main:@279) -> float_type, {32, 128}, {128, 1}: 0.0238708ms, 0.210472% main:@281 = load[offset=1293440,end=1309824](main:@1) -> float_type, {32, 128}, {128, 1}: 0.0006337ms, 0.00558742% main:@282 = gpu::code_object[code_object=13520,symbol_name=tanh_div_kernel,global=2048,local=256,](main:@280,main:@281) -> float_type, {32, 128}, {128, 1}: 0.0141908ms, 0.125122% main:@283 = hip::hip_copy_literal[id=main:@literal:50] -> float_type, {128, 16}, {16, 1}: 0.00079224ms, 0.00698529% main:@284 = load[offset=770048,end=772096](main:@1) -> float_type, {32, 16}, {16, 1}: 0.00046898ms, 0.00413506% main:@285 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@282,main:@283,main:@284) -> float_type, {32, 16}, {16, 1}: 0.0207574ms, 0.183021% main:@286 = multibroadcast[out_lens={32, 15, 16}](main:@29) -> float_type, {32, 15, 16}, {15, 1, 0}: 0.00071318ms, 0.00628821% 4Rg0WA19ot:0 = @param:4Rg0WA19ot:0 -> float_type, {32, 15, 17}, {255, 17, 1}: 0.00057212ms, 0.00504446% main:@288 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 15, 17}](4Rg0WA19ot:0) -> float_type, {32, 15, 16}, {255, 17, 1}: 0.00069716ms, 0.00614696% main:@289 = load[offset=1333376,end=1364096](main:@1) -> float_type, {32, 15, 16}, {240, 16, 1}: 0.00046284ms, 0.00408093% main:@290 = gpu::code_object[code_object=13512,symbol_name=add_kernel,global=7680,local=256,](main:@288,main:@286,main:@289) -> float_type, {32, 15, 16}, {240, 16, 1}: 0.0146825ms, 0.129458% main:@291 = reshape_lazy[dims={-1, 240}](main:@290) -> float_type, {32, 240}, {240, 1}: 0.00064512ms, 0.00568812% main:@292 = load[offset=1293440,end=1333376](main:@1) -> float_type, {32, 312}, {312, 1}: 0.00047148ms, 0.00415711% main:@293 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=4992,local=256,](main:@11,main:@291,main:@292) -> float_type, {32, 312}, {312, 1}: 0.0143994ms, 0.126962% main:@294 = hip::hip_copy_literal[id=main:@literal:139] -> float_type, {312, 128}, {128, 1}: 0.0007708ms, 0.00679625% main:@295 = load[offset=1349760,end=1366144](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00046306ms, 0.00408287% main:@296 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@293,main:@294,main:@295) -> float_type, {32, 128}, {128, 1}: 0.0257823ms, 0.227327% main:@297 = load[offset=1333376,end=1349760](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00063498ms, 0.00559871% main:@298 = gpu::code_object[code_object=13520,symbol_name=neg_pow_kernel,global=2048,local=256,](main:@296,main:@297) -> float_type, {32, 128}, {128, 1}: 0.0163282ms, 0.143968% main:@299 = hip::hip_copy_literal[id=main:@literal:91] -> float_type, {128, 16}, {16, 1}: 0.00074868ms, 0.00660122% main:@300 = load[offset=65024,end=67072](main:@1) -> float_type, {32, 16}, {16, 1}: 0.00046316ms, 0.00408375% main:@301 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@298,main:@299,main:@300) -> float_type, {32, 16}, {16, 1}: 0.020221ms, 0.178292% main:@302 = load[offset=1333376,end=1349760](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00063384ms, 0.00558866% main:@303 = gpu::code_object[code_object=13520,symbol_name=log_add_kernel,global=2048,local=256,](main:@83,main:@302) -> float_type, {32, 128}, {128, 1}: 0.0132773ms, 0.117068% main:@304 = hip::hip_copy_literal[id=main:@literal:79] -> float_type, {128, 16}, {16, 1}: 0.00078174ms, 0.00689271% main:@305 = load[offset=62976,end=65024](main:@1) -> float_type, {32, 16}, {16, 1}: 0.00048688ms, 0.00429289% main:@306 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@303,main:@304,main:@305) -> float_type, {32, 16}, {16, 1}: 0.0205108ms, 0.180847% main:@307 = load[offset=67072,end=77312](main:@1) -> float_type, {32, 80}, {80, 1}: 0.00065792ms, 0.00580098% main:@308 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=512,local=256,](main:@270,main:@129,main:@285,main:@266,main:@301,main:@307) -> float_type, {32, 80}, {80, 1}: 0.0149956ms, 0.132218% main:@309 = load[offset=1333376,end=1335424](main:@1) -> float_type, {32, 16}, {16, 1}: 0.00064388ms, 0.00567718% main:@310 = gpu::code_object[code_object=13520,symbol_name=rsqrt_pow_kernel,global=256,local=256,](main:@181,main:@309) -> float_type, {32, 16}, {16, 1}: 0.0162685ms, 0.143442% main:@311 = load[offset=759808,end=761856](main:@1) -> float_type, {32, 16}, {16, 1}: 0.00063476ms, 0.00559677% main:@312 = gpu::code_object[code_object=13520,symbol_name=neg_div_kernel,global=256,local=256,](main:@155,main:@311) -> float_type, {32, 16}, {16, 1}: 0.0130587ms, 0.115141% main:@313 = load[offset=104704,end=106752](main:@1) -> float_type, {32, 16}, {16, 1}: 0.00062006ms, 0.00546716% main:@314 = gpu::code_object[code_object=13520,symbol_name=rsqrt_sub_kernel,global=256,local=256,](main:@103,main:@313) -> float_type, {32, 16}, {16, 1}: 0.0129167ms, 0.113888% main:@315 = load[offset=106752,end=108800](main:@1) -> float_type, {32, 16}, {16, 1}: 0.00061622ms, 0.0054333% main:@316 = gpu::code_object[code_object=13520,symbol_name=tanh_sub_kernel,global=256,local=256,](main:@223,main:@315) -> float_type, {32, 16}, {16, 1}: 0.0135653ms, 0.119608% main:@317 = load[offset=100608,end=102656](main:@1) -> float_type, {32, 16}, {16, 1}: 0.00061902ms, 0.00545799% main:@318 = gpu::code_object[code_object=13520,symbol_name=tanh_div_kernel,global=256,local=256,](main:@274,main:@317) -> float_type, {32, 16}, {16, 1}: 0.0139489ms, 0.12299% main:@319 = load[offset=768000,end=770048](main:@1) -> float_type, {32, 16}, {16, 1}: 0.00060594ms, 0.00534266% main:@320 = gpu::code_object[code_object=13520,symbol_name=tanh_sub_kernel,global=256,local=256,](main:@227,main:@319) -> float_type, {32, 16}, {16, 1}: 0.0137354ms, 0.121107% main:@321 = load[offset=96640,end=98688](main:@1) -> float_type, {32, 16}, {16, 1}: 0.00062882ms, 0.0055444% main:@322 = gpu::code_object[code_object=13520,symbol_name=rsqrt_sub_kernel,global=256,local=256,](main:@232,main:@321) -> float_type, {32, 16}, {16, 1}: 0.0130595ms, 0.115147% main:@323 = load[offset=77312,end=81408](main:@1) -> float_type, {32, 32}, {32, 1}: 0.0006023ms, 0.00531057% main:@324 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=512,local=256,](main:@129,main:@285,main:@323) -> float_type, {32, 32}, {32, 1}: 0.013398ms, 0.118132% main:@325 = load[offset=1335424,end=1345664](main:@1) -> float_type, {32, 80}, {80, 1}: 0.00062188ms, 0.00548321% main:@326 = gpu::code_object[code_object=13560,symbol_name=neg_rsqrt_tanh_log_concat_add_kernel,global=640,local=256,](main:@270,main:@324,main:@266,main:@301,main:@308,main:@325) -> float_type, {32, 80}, {80, 1}: 0.0158978ms, 0.140174% main:@327 = load[offset=93696,end=95744](main:@1) -> float_type, {32, 16}, {16, 1}: 0.00059932ms, 0.00528429% main:@328 = gpu::code_object[code_object=13520,symbol_name=neg_div_kernel,global=256,local=256,](main:@251,main:@327) -> float_type, {32, 16}, {16, 1}: 0.0129638ms, 0.114303% main:@329 = load[offset=102656,end=104704](main:@1) -> float_type, {32, 16}, {16, 1}: 0.00061916ms, 0.00545922% main:@330 = gpu::code_object[code_object=13520,symbol_name=tanh_add_kernel,global=256,local=256,](main:@114,main:@329) -> float_type, {32, 16}, {16, 1}: 0.0137388ms, 0.121137% main:@331 = load[offset=761856,end=763904](main:@1) -> float_type, {32, 16}, {16, 1}: 0.00058778ms, 0.00518254% main:@332 = gpu::code_object[code_object=13520,symbol_name=rsqrt_add_kernel,global=256,local=256,](main:@306,main:@331) -> float_type, {32, 16}, {16, 1}: 0.0130082ms, 0.114696% main:@333 = load[offset=62976,end=93696](main:@1) -> float_type, {32, 240}, {240, 1}: 0.00059554ms, 0.00525096% main:@334 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=698,local=256,](main:@322,main:@312,main:@332,main:@318,main:@314,main:@310,main:@326,main:@320,main:@330,main:@328,main:@316,main:@333) -> float_type, {32, 240}, {240, 1}: 0.0192926ms, 0.170106% main:@335 = hip::hip_copy_literal[id=main:@literal:107] -> float_type, {128, 240}, {240, 1}: 0.0008325ms, 0.00734027% main:@336 = hip::hip_copy_literal[id=main:@literal:137] -> float_type, {128, 240}, {240, 1}: 0.00063262ms, 0.0055779% main:@337 = load[offset=1553536,end=1695872](main:@1) -> float_type, {32, 1112}, {1112, 1}: 0.00047928ms, 0.00422588% main:@338 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=17792,local=256,](main:@277,main:@334,main:@337) -> float_type, {32, 1112}, {1112, 1}: 0.0161976ms, 0.142816% main:@339 = hip::hip_copy_literal[id=main:@literal:54] -> float_type, {1112, 128}, {128, 1}: 0.0007934ms, 0.00699552% main:@340 = load[offset=1333376,end=1349760](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00047152ms, 0.00415746% main:@341 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@338,main:@339,main:@340) -> float_type, {32, 128}, {128, 1}: 0.0256783ms, 0.22641% main:@342 = load[offset=93696,end=110080](main:@1) -> float_type, {32, 128}, {128, 1}: 0.0006198ms, 0.00546487% main:@343 = gpu::code_object[code_object=13520,symbol_name=log_pow_kernel,global=2048,local=256,](main:@341,main:@342) -> float_type, {32, 128}, {128, 1}: 0.0165402ms, 0.145838% main:@344 = load[offset=1358976,end=1389696](main:@1) -> float_type, {32, 240}, {240, 1}: 0.00062402ms, 0.00550207% main:@345 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@343,main:@336,main:@344) -> float_type, {32, 240}, {240, 1}: 0.0230805ms, 0.203504% main:@346 = load[offset=1553536,end=1624192](main:@1) -> float_type, {32, 552}, {552, 1}: 0.00062392ms, 0.00550119% main:@347 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=8832,local=256,](main:@293,main:@334,main:@346) -> float_type, {32, 552}, {552, 1}: 0.0152793ms, 0.134719% main:@348 = hip::hip_copy_literal[id=main:@literal:122] -> float_type, {552, 128}, {128, 1}: 0.00083762ms, 0.00738542% main:@349 = load[offset=1333376,end=1349760](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00046784ms, 0.00412501% main:@350 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@347,main:@348,main:@349) -> float_type, {32, 128}, {128, 1}: 0.0338766ms, 0.298695% main:@351 = load[offset=93696,end=110080](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00063098ms, 0.00556344% main:@352 = gpu::code_object[code_object=13520,symbol_name=neg_div_kernel,global=2048,local=256,](main:@350,main:@351) -> float_type, {32, 128}, {128, 1}: 0.0132818ms, 0.117108% main:@353 = load[offset=1624192,end=1654912](main:@1) -> float_type, {32, 240}, {240, 1}: 0.0006284ms, 0.00554069% main:@354 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@352,main:@336,main:@353) -> float_type, {32, 240}, {240, 1}: 0.0206292ms, 0.181891% main:@355 = load[offset=1553536,end=1624192](main:@1) -> float_type, {32, 552}, {552, 1}: 0.0006489ms, 0.00572144% main:@356 = gpu::code_object[code_object=13560,symbol_name=noop_rsqrt_sub_concat_noop_kernel,global=8832,local=256,](main:@293,main:@354,main:@334,main:@355) -> float_type, {32, 552}, {552, 1}: 0.0152904ms, 0.134818% main:@357 = hip::hip_copy_literal[id=main:@literal:98] -> float_type, {552, 256}, {256, 1}: 0.00072752ms, 0.00641465% main:@358 = load[offset=1293440,end=1326208](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00051498ms, 0.00454065% main:@359 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@356,main:@357,main:@358) -> float_type, {32, 256}, {256, 1}: 0.0334842ms, 0.295235% main:@360 = load[offset=1553536,end=1839232](main:@1) -> float_type, {32, 2232}, {2232, 1}: 0.00062286ms, 0.00549185% main:@361 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=35712,local=256,](main:@200,main:@334,main:@360) -> float_type, {32, 2232}, {2232, 1}: 0.0182568ms, 0.160973% main:@362 = hip::hip_copy_literal[id=main:@literal:66] -> float_type, {2232, 128}, {128, 1}: 0.00074118ms, 0.00653509% main:@363 = load[offset=93696,end=110080](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00047494ms, 0.00418761% main:@364 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@361,main:@362,main:@363) -> float_type, {32, 128}, {128, 1}: 0.0296907ms, 0.261787% main:@365 = load[offset=1553536,end=1613952](main:@1) -> float_type, {32, 472}, {472, 1}: 0.00062996ms, 0.00555445% main:@366 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=7552,local=256,](main:@244,main:@334,main:@365) -> float_type, {32, 472}, {472, 1}: 0.014822ms, 0.130688% main:@367 = hip::hip_copy_literal[id=main:@literal:119] -> float_type, {472, 128}, {128, 1}: 0.00076598ms, 0.00675376% main:@368 = load[offset=1342592,end=1358976](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00049424ms, 0.00435779% main:@369 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@366,main:@367,main:@368) -> float_type, {32, 128}, {128, 1}: 0.0313898ms, 0.276768% main:@370 = load[offset=1326208,end=1342592](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00062454ms, 0.00550666% main:@371 = gpu::code_object[code_object=13520,symbol_name=log_add_kernel,global=2048,local=256,](main:@369,main:@370) -> float_type, {32, 128}, {128, 1}: 0.0132812ms, 0.117102% main:@372 = hip::hip_copy_literal[id=main:@literal:102] -> float_type, {128, 240}, {240, 1}: 0.00076532ms, 0.00674794% main:@373 = load[offset=1613952,end=1644672](main:@1) -> float_type, {32, 240}, {240, 1}: 0.00047704ms, 0.00420613% main:@374 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@371,main:@372,main:@373) -> float_type, {32, 240}, {240, 1}: 0.0206392ms, 0.181978% main:@375 = load[offset=1553536,end=1613952](main:@1) -> float_type, {32, 472}, {472, 1}: 0.0006369ms, 0.00561564% main:@376 = gpu::code_object[code_object=13552,symbol_name=noop_tanh_div_concat_noop_kernel,global=7552,local=256,](main:@244,main:@374,main:@334,main:@375) -> float_type, {32, 472}, {472, 1}: 0.0154972ms, 0.136641% main:@377 = hip::hip_copy_literal[id=main:@literal:80] -> float_type, {472, 256}, {256, 1}: 0.00077054ms, 0.00679396% main:@378 = load[offset=1326208,end=1358976](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00047412ms, 0.00418038% main:@379 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@376,main:@377,main:@378) -> float_type, {32, 256}, {256, 1}: 0.0309817ms, 0.27317% main:@380 = load[offset=1553536,end=1716352](main:@1) -> float_type, {32, 1272}, {1272, 1}: 0.00065676ms, 0.00579075% main:@381 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=20352,local=256,](main:@95,main:@334,main:@380) -> float_type, {32, 1272}, {1272, 1}: 0.0163712ms, 0.144348% main:@382 = hip::hip_copy_literal[id=main:@literal:59] -> float_type, {1272, 128}, {128, 1}: 0.00072174ms, 0.00636369% main:@383 = load[offset=1406080,end=1422464](main:@1) -> float_type, {32, 128}, {128, 1}: 0.0004756ms, 0.00419343% main:@384 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@381,main:@382,main:@383) -> float_type, {32, 128}, {128, 1}: 0.0257268ms, 0.226837% main:@385 = load[offset=1389696,end=1406080](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00065988ms, 0.00581826% main:@386 = gpu::code_object[code_object=13520,symbol_name=rsqrt_add_kernel,global=2048,local=256,](main:@384,main:@385) -> float_type, {32, 128}, {128, 1}: 0.0132266ms, 0.116621% main:@387 = hip::hip_copy_literal[id=main:@literal:39] -> float_type, {128, 240}, {240, 1}: 0.00077798ms, 0.00685956% main:@388 = load[offset=1406080,end=1436800](main:@1) -> float_type, {32, 240}, {240, 1}: 0.00048172ms, 0.00424739% main:@389 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@386,main:@387,main:@388) -> float_type, {32, 240}, {240, 1}: 0.0211793ms, 0.186741% main:@390 = load[offset=1553536,end=1716352](main:@1) -> float_type, {32, 1272}, {1272, 1}: 0.0006689ms, 0.00589779% main:@391 = gpu::code_object[code_object=13552,symbol_name=noop_log_sub_concat_noop_kernel,global=20352,local=256,](main:@95,main:@389,main:@334,main:@390) -> float_type, {32, 1272}, {1272, 1}: 0.0164404ms, 0.144958% main:@392 = hip::hip_copy_literal[id=main:@literal:40] -> float_type, {1272, 256}, {256, 1}: 0.00077388ms, 0.00682341% main:@393 = load[offset=3244160,end=3276928](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00048624ms, 0.00428725% main:@394 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@391,main:@392,main:@393) -> float_type, {32, 256}, {256, 1}: 0.0328296ms, 0.289463% main:@395 = hip::hip_copy_literal[id=main:@literal:123] -> float_type, {1112, 256}, {256, 1}: 0.0007175ms, 0.0063263% main:@396 = load[offset=1553536,end=1695872](main:@1) -> float_type, {32, 1112}, {1112, 1}: 0.0004891ms, 0.00431247% main:@397 = gpu::code_object[code_object=13560,symbol_name=noop_rsqrt_add_concat_noop_kernel,global=17792,local=256,](main:@277,main:@345,main:@334,main:@396) -> float_type, {32, 1112}, {1112, 1}: 0.0163527ms, 0.144184% main:@398 = load[offset=3276928,end=3309696](main:@1) -> float_type, {32, 256}, {256, 1}: 0.0006331ms, 0.00558213% main:@399 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@397,main:@395,main:@398) -> float_type, {32, 256}, {256, 1}: 0.0302405ms, 0.266635% main:@400 = load[offset=409088,end=469504](main:@1) -> float_type, {32, 472}, {472, 1}: 0.00064002ms, 0.00564315% main:@401 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=7552,local=256,](main:@106,main:@334,main:@400) -> float_type, {32, 472}, {472, 1}: 0.0148158ms, 0.130633% main:@402 = hip::hip_copy_literal[id=main:@literal:99] -> float_type, {472, 128}, {128, 1}: 0.00076756ms, 0.00676769% main:@403 = load[offset=469504,end=485888](main:@1) -> float_type, {32, 128}, {128, 1}: 0.0004823ms, 0.00425251% main:@404 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@401,main:@402,main:@403) -> float_type, {32, 128}, {128, 1}: 0.0314282ms, 0.277107% main:@405 = load[offset=409088,end=425472](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00062152ms, 0.00548003% main:@406 = gpu::code_object[code_object=13520,symbol_name=log_div_kernel,global=2048,local=256,](main:@404,main:@405) -> float_type, {32, 128}, {128, 1}: 0.0134851ms, 0.1189% main:@407 = load[offset=469504,end=500224](main:@1) -> float_type, {32, 240}, {240, 1}: 0.00061402ms, 0.0054139% main:@408 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@406,main:@335,main:@407) -> float_type, {32, 240}, {240, 1}: 0.0203763ms, 0.17966% main:@409 = load[offset=409088,end=469504](main:@1) -> float_type, {32, 472}, {472, 1}: 0.0006208ms, 0.00547368% main:@410 = gpu::code_object[code_object=13560,symbol_name=noop_rsqrt_div_concat_noop_kernel,global=7552,local=256,](main:@106,main:@408,main:@334,main:@409) -> float_type, {32, 472}, {472, 1}: 0.0151342ms, 0.13344% main:@411 = hip::hip_copy_literal[id=main:@literal:63] -> float_type, {472, 256}, {256, 1}: 0.00075428ms, 0.0066506% main:@412 = load[offset=3309696,end=3342464](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00048906ms, 0.00431211% main:@413 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@410,main:@411,main:@412) -> float_type, {32, 256}, {256, 1}: 0.0302301ms, 0.266543% main:@414 = load[offset=379392,end=521728](main:@1) -> float_type, {32, 1112}, {1112, 1}: 0.00063146ms, 0.00556767% main:@415 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=17792,local=256,](main:@147,main:@334,main:@414) -> float_type, {32, 1112}, {1112, 1}: 0.0162455ms, 0.143239% main:@416 = hip::hip_copy_literal[id=main:@literal:135] -> float_type, {1112, 128}, {128, 1}: 0.0007324ms, 0.00645768% main:@417 = load[offset=521728,end=538112](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00047468ms, 0.00418532% main:@418 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@415,main:@416,main:@417) -> float_type, {32, 128}, {128, 1}: 0.0258874ms, 0.228253% main:@419 = load[offset=379392,end=395776](main:@1) -> float_type, {32, 128}, {128, 1}: 0.0006223ms, 0.00548691% main:@420 = gpu::code_object[code_object=13520,symbol_name=neg_pow_kernel,global=2048,local=256,](main:@418,main:@419) -> float_type, {32, 128}, {128, 1}: 0.0162652ms, 0.143413% main:@421 = hip::hip_copy_literal[id=main:@literal:74] -> float_type, {128, 240}, {240, 1}: 0.00077292ms, 0.00681495% main:@422 = load[offset=1358976,end=1389696](main:@1) -> float_type, {32, 240}, {240, 1}: 0.00048282ms, 0.00425709% main:@423 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@420,main:@421,main:@422) -> float_type, {32, 240}, {240, 1}: 0.0203267ms, 0.179224% main:@424 = load[offset=379392,end=521728](main:@1) -> float_type, {32, 1112}, {1112, 1}: 0.00063226ms, 0.00557473% main:@425 = gpu::code_object[code_object=13552,symbol_name=noop_log_add_concat_noop_kernel,global=17792,local=256,](main:@147,main:@423,main:@334,main:@424) -> float_type, {32, 1112}, {1112, 1}: 0.0164515ms, 0.145055% main:@426 = hip::hip_copy_literal[id=main:@literal:55] -> float_type, {1112, 256}, {256, 1}: 0.00078826ms, 0.0069502% main:@427 = load[offset=3342464,end=3375232](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00049976ms, 0.00440646% main:@428 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@425,main:@426,main:@427) -> float_type, {32, 256}, {256, 1}: 0.0300225ms, 0.264713% main:@429 = load[offset=1358976,end=1526912](main:@1) -> float_type, {32, 1312}, {1312, 1}: 0.00063234ms, 0.00557543% main:@430 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=20992,local=256,](main:@122,main:@334,main:@429) -> float_type, {32, 1312}, {1312, 1}: 0.0163525ms, 0.144182% main:@431 = hip::hip_copy_literal[id=main:@literal:46] -> float_type, {1312, 128}, {128, 1}: 0.00074612ms, 0.00657865% main:@432 = load[offset=379392,end=395776](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00049402ms, 0.00435585% main:@433 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@430,main:@431,main:@432) -> float_type, {32, 128}, {128, 1}: 0.0255577ms, 0.225346% main:@434 = load[offset=410112,end=426496](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00062632ms, 0.00552235% main:@435 = gpu::code_object[code_object=13520,symbol_name=tanh_add_kernel,global=2048,local=256,](main:@433,main:@434) -> float_type, {32, 128}, {128, 1}: 0.0138419ms, 0.122046% main:@436 = load[offset=379392,end=410112](main:@1) -> float_type, {32, 240}, {240, 1}: 0.000619ms, 0.00545781% main:@437 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@435,main:@335,main:@436) -> float_type, {32, 240}, {240, 1}: 0.0205339ms, 0.181051% main:@438 = load[offset=1358976,end=1526912](main:@1) -> float_type, {32, 1312}, {1312, 1}: 0.00065114ms, 0.0057412% main:@439 = gpu::code_object[code_object=13552,symbol_name=noop_neg_div_concat_noop_kernel,global=20992,local=256,](main:@122,main:@437,main:@334,main:@438) -> float_type, {32, 1312}, {1312, 1}: 0.0165859ms, 0.14624% main:@440 = hip::hip_copy_literal[id=main:@literal:60] -> float_type, {1312, 256}, {256, 1}: 0.00081914ms, 0.00722248% main:@441 = load[offset=3375232,end=3408000](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00047842ms, 0.0042183% main:@442 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@439,main:@440,main:@441) -> float_type, {32, 256}, {256, 1}: 0.0304188ms, 0.268207% main:@443 = load[offset=110080,end=170496](main:@1) -> float_type, {32, 472}, {472, 1}: 0.00062886ms, 0.00554475% main:@444 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=7552,local=256,](main:@167,main:@334,main:@443) -> float_type, {32, 472}, {472, 1}: 0.0147917ms, 0.13042% main:@445 = hip::hip_copy_literal[id=main:@literal:145] -> float_type, {472, 128}, {128, 1}: 0.00080476ms, 0.00709568% main:@446 = load[offset=170496,end=186880](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00049128ms, 0.00433169% main:@447 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@444,main:@445,main:@446) -> float_type, {32, 128}, {128, 1}: 0.0309925ms, 0.273265% main:@448 = load[offset=140800,end=157184](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00062154ms, 0.00548021% main:@449 = gpu::code_object[code_object=13520,symbol_name=log_pow_kernel,global=2048,local=256,](main:@447,main:@448) -> float_type, {32, 128}, {128, 1}: 0.0165589ms, 0.146003% main:@450 = hip::hip_copy_literal[id=main:@literal:71] -> float_type, {128, 240}, {240, 1}: 0.000745ms, 0.00656877% main:@451 = load[offset=110080,end=140800](main:@1) -> float_type, {32, 240}, {240, 1}: 0.0004982ms, 0.0043927% main:@452 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@449,main:@450,main:@451) -> float_type, {32, 240}, {240, 1}: 0.0208857ms, 0.184152% main:@453 = hip::hip_copy_literal[id=main:@literal:37] -> float_type, {472, 256}, {256, 1}: 0.00076186ms, 0.00671743% main:@454 = load[offset=140800,end=201216](main:@1) -> float_type, {32, 472}, {472, 1}: 0.00049036ms, 0.00432357% main:@455 = gpu::code_object[code_object=13552,symbol_name=noop_log_pow_concat_noop_kernel,global=7552,local=256,](main:@167,main:@452,main:@334,main:@454) -> float_type, {32, 472}, {472, 1}: 0.0171939ms, 0.151601% main:@456 = load[offset=3408000,end=3440768](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00062356ms, 0.00549802% main:@457 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@455,main:@453,main:@456) -> float_type, {32, 256}, {256, 1}: 0.031295ms, 0.275933% main:@458 = load[offset=110080,end=170496](main:@1) -> float_type, {32, 472}, {472, 1}: 0.0006293ms, 0.00554863% main:@459 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=7552,local=256,](main:@215,main:@334,main:@458) -> float_type, {32, 472}, {472, 1}: 0.0147748ms, 0.130271% main:@460 = hip::hip_copy_literal[id=main:@literal:47] -> float_type, {472, 128}, {128, 1}: 0.00077514ms, 0.00683452% main:@461 = load[offset=170496,end=186880](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00047624ms, 0.00419908% main:@462 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@459,main:@460,main:@461) -> float_type, {32, 128}, {128, 1}: 0.0307774ms, 0.271369% main:@463 = load[offset=110080,end=126464](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00062022ms, 0.00546857% main:@464 = gpu::code_object[code_object=13520,symbol_name=tanh_sub_kernel,global=2048,local=256,](main:@462,main:@463) -> float_type, {32, 128}, {128, 1}: 0.0137095ms, 0.120879% main:@465 = load[offset=170496,end=201216](main:@1) -> float_type, {32, 240}, {240, 1}: 0.00062192ms, 0.00548356% main:@466 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@464,main:@336,main:@465) -> float_type, {32, 240}, {240, 1}: 0.0203795ms, 0.179689% main:@467 = load[offset=110080,end=170496](main:@1) -> float_type, {32, 472}, {472, 1}: 0.0006227ms, 0.00549044% main:@468 = gpu::code_object[code_object=13552,symbol_name=noop_tanh_div_concat_noop_kernel,global=7552,local=256,](main:@215,main:@466,main:@334,main:@467) -> float_type, {32, 472}, {472, 1}: 0.0153591ms, 0.135424% main:@469 = hip::hip_copy_literal[id=main:@literal:156] -> float_type, {472, 256}, {256, 1}: 0.00077052ms, 0.00679379% main:@470 = load[offset=3440768,end=3473536](main:@1) -> float_type, {32, 256}, {256, 1}: 0.0004946ms, 0.00436096% main:@471 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@468,main:@469,main:@470) -> float_type, {32, 256}, {256, 1}: 0.030491ms, 0.268843% main:@472 = load[offset=379392,end=542208](main:@1) -> float_type, {32, 1272}, {1272, 1}: 0.0006195ms, 0.00546222% main:@473 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=20352,local=256,](main:@173,main:@334,main:@472) -> float_type, {32, 1272}, {1272, 1}: 0.0166098ms, 0.146451% main:@474 = hip::hip_copy_literal[id=main:@literal:136] -> float_type, {1272, 128}, {128, 1}: 0.00075244ms, 0.00663437% main:@475 = load[offset=126464,end=142848](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00050024ms, 0.00441069% main:@476 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@473,main:@474,main:@475) -> float_type, {32, 128}, {128, 1}: 0.0260543ms, 0.229725% main:@477 = load[offset=110080,end=126464](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00061462ms, 0.00541919% main:@478 = gpu::code_object[code_object=13520,symbol_name=neg_pow_kernel,global=2048,local=256,](main:@476,main:@477) -> float_type, {32, 128}, {128, 1}: 0.0163258ms, 0.143947% main:@479 = hip::hip_copy_literal[id=main:@literal:78] -> float_type, {128, 240}, {240, 1}: 0.0007736ms, 0.00682094% main:@480 = load[offset=126464,end=157184](main:@1) -> float_type, {32, 240}, {240, 1}: 0.00047372ms, 0.00417686% main:@481 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@478,main:@479,main:@480) -> float_type, {32, 240}, {240, 1}: 0.0203619ms, 0.179534% main:@482 = load[offset=379392,end=542208](main:@1) -> float_type, {32, 1272}, {1272, 1}: 0.00062236ms, 0.00548744% main:@483 = gpu::code_object[code_object=13560,symbol_name=noop_rsqrt_div_concat_noop_kernel,global=20352,local=256,](main:@173,main:@481,main:@334,main:@482) -> float_type, {32, 1272}, {1272, 1}: 0.0168102ms, 0.148218% main:@484 = hip::hip_copy_literal[id=main:@literal:131] -> float_type, {1272, 256}, {256, 1}: 0.00088842ms, 0.00783333% main:@485 = load[offset=3473536,end=3506304](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00047664ms, 0.0042026% main:@486 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@483,main:@484,main:@485) -> float_type, {32, 256}, {256, 1}: 0.0305031ms, 0.26895% main:@487 = hip::hip_copy_literal[id=main:@literal:125] -> float_type, {612, 128}, {128, 1}: 0.0007671ms, 0.00676363% main:@488 = load[offset=126464,end=204800](main:@1) -> float_type, {32, 612}, {612, 1}: 0.00046614ms, 0.00411002% main:@489 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=9792,local=256,](main:@259,main:@334,main:@488) -> float_type, {32, 612}, {612, 1}: 0.015591ms, 0.137468% main:@490 = load[offset=110080,end=126464](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00063594ms, 0.00560717% main:@491 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@489,main:@487,main:@490) -> float_type, {32, 128}, {128, 1}: 0.0350248ms, 0.308818% main:@492 = hip::hip_copy_literal[id=main:@literal:124] -> float_type, {128, 240}, {240, 1}: 0.0007791ms, 0.00686944% main:@493 = load[offset=126464,end=142848](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00046946ms, 0.0041393% main:@494 = gpu::code_object[code_object=13520,symbol_name=neg_pow_kernel,global=2048,local=256,](main:@491,main:@493) -> float_type, {32, 128}, {128, 1}: 0.0164251ms, 0.144823% main:@495 = load[offset=188416,end=219136](main:@1) -> float_type, {32, 240}, {240, 1}: 0.00061896ms, 0.00545746% main:@496 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@494,main:@492,main:@495) -> float_type, {32, 240}, {240, 1}: 0.0201064ms, 0.177281% main:@497 = load[offset=110080,end=188416](main:@1) -> float_type, {32, 612}, {612, 1}: 0.0006215ms, 0.00547986% main:@498 = gpu::code_object[code_object=13560,symbol_name=noop_rsqrt_add_concat_noop_kernel,global=9792,local=256,](main:@259,main:@496,main:@334,main:@497) -> float_type, {32, 612}, {612, 1}: 0.0155991ms, 0.13754% main:@499 = hip::hip_copy_literal[id=main:@literal:103] -> float_type, {612, 256}, {256, 1}: 0.00076836ms, 0.00677474% main:@500 = load[offset=3506304,end=3539072](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00048078ms, 0.00423911% main:@501 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@498,main:@499,main:@500) -> float_type, {32, 256}, {256, 1}: 0.034342ms, 0.302799% cEDTxpPUfb:0 = @param:cEDTxpPUfb:0 -> float_type, {32, 30, 9}, {270, 9, 1}: 0.00077582ms, 0.00684052% main:@503 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 30, 9}](cEDTxpPUfb:0) -> float_type, {32, 30, 8}, {270, 9, 1}: 0.00072402ms, 0.00638379% aHGpDUJj9f:0 = @param:aHGpDUJj9f:0 -> float_type, {32, 30, 21}, {630, 21, 1}: 0.0005105ms, 0.00450115% main:@505 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 30, 21}](aHGpDUJj9f:0) -> float_type, {32, 30, 20}, {630, 21, 1}: 0.00080946ms, 0.00713713% Hj2GwToc7U:0 = @param:Hj2GwToc7U:0 -> float_type, {32, 30, 9}, {270, 9, 1}: 0.00050664ms, 0.00446712% main:@507 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 30, 9}](Hj2GwToc7U:0) -> float_type, {32, 30, 8}, {270, 9, 1}: 0.00064908ms, 0.00572303% S4PeIw3K1X:0 = @param:S4PeIw3K1X:0 -> float_type, {32, 30, 21}, {630, 21, 1}: 0.00050214ms, 0.00442744% main:@509 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 30, 21}](S4PeIw3K1X:0) -> float_type, {32, 30, 20}, {630, 21, 1}: 0.00079054ms, 0.00697031% DW8PcmTVkC:0 = @param:DW8PcmTVkC:0 -> float_type, {32, 30, 9}, {270, 9, 1}: 0.00047482ms, 0.00418656% main:@511 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 30, 9}](DW8PcmTVkC:0) -> float_type, {32, 30, 8}, {270, 9, 1}: 0.00064224ms, 0.00566272% H6VFcpqaB0:0 = @param:H6VFcpqaB0:0 -> float_type, {32, 30, 9}, {270, 9, 1}: 0.0005698ms, 0.00502401% main:@513 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 30, 9}](H6VFcpqaB0:0) -> float_type, {32, 30, 8}, {270, 9, 1}: 0.00077056ms, 0.00679414% CYnW83Np9f:0 = @param:CYnW83Np9f:0 -> float_type, {32, 30, 17}, {510, 17, 1}: 0.00046624ms, 0.00411091% main:@515 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 30, 17}](CYnW83Np9f:0) -> float_type, {32, 30, 16}, {510, 17, 1}: 0.00061802ms, 0.00544917% hEoRIXiSF0:0 = @param:hEoRIXiSF0:0 -> float_type, {32, 30, 9}, {270, 9, 1}: 0.00052808ms, 0.00465616% main:@517 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 30, 9}](hEoRIXiSF0:0) -> float_type, {32, 30, 8}, {270, 9, 1}: 0.00103024ms, 0.00908377% main:@518 = load[offset=114688,end=514048](main:@1) -> float_type, {32, 30, 104}, {3120, 104, 1}: 0.00049372ms, 0.0043532% 4RjhYVIfz6:0 = @param:4RjhYVIfz6:0 -> float_type, {32, 30, 9}, {270, 9, 1}: 0.00047756ms, 0.00421072% main:@520 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 30, 9}](4RjhYVIfz6:0) -> float_type, {32, 30, 8}, {270, 9, 1}: 0.0006438ms, 0.00567648% main:@521 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=11093,local=256,](main:@513,main:@511,main:@517,main:@507,main:@505,main:@520,main:@503,main:@515,main:@509,main:@518) -> float_type, {32, 30, 104}, {3120, 104, 1}: 0.0271286ms, 0.239196% main:@522 = hip::hip_copy_literal[id=main:@literal:8] -> float_type, {104, 64}, {64, 1}: 0.00075648ms, 0.00666999% main:@523 = reshape_lazy[dims={960, 104}](main:@521) -> float_type, {960, 104}, {104, 1}: 0.0005702ms, 0.00502754% main:@524 = load[offset=527488,end=773248](main:@1) -> float_type, {960, 64}, {64, 1}: 0.00049246ms, 0.00434209% main:@525 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@523,main:@522,main:@524) -> float_type, {960, 64}, {64, 1}: 0.0235935ms, 0.208027% IFQJaVMDS5:0 = @param:IFQJaVMDS5:0 -> float_type, {32, 50, 17}, {850, 17, 1}: 0.00074066ms, 0.00653051% main:@527 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 50, 17}](IFQJaVMDS5:0) -> float_type, {32, 50, 16}, {850, 17, 1}: 0.00067028ms, 0.00590996% r8QjI5SCed:0 = @param:r8QjI5SCed:0 -> float_type, {32, 50, 9}, {450, 9, 1}: 0.0005036ms, 0.00444031% main:@529 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 50, 9}](r8QjI5SCed:0) -> float_type, {32, 50, 8}, {450, 9, 1}: 0.0007765ms, 0.00684651% 06AyS2QTzo:0 = @param:06AyS2QTzo:0 -> float_type, {32, 50, 9}, {450, 9, 1}: 0.00051902ms, 0.00457627% main:@531 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 50, 9}](06AyS2QTzo:0) -> float_type, {32, 50, 8}, {450, 9, 1}: 0.00062274ms, 0.00549079% Uc35C1q4oy:0 = @param:Uc35C1q4oy:0 -> float_type, {32, 50, 9}, {450, 9, 1}: 0.00048344ms, 0.00426256% main:@533 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 50, 9}](Uc35C1q4oy:0) -> float_type, {32, 50, 8}, {450, 9, 1}: 0.00077654ms, 0.00684687% main:@534 = load[offset=1358976,end=1845376](main:@1) -> float_type, {32, 50, 76}, {3800, 76, 1}: 0.00045756ms, 0.00403437% main:@535 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=20266,local=256,](main:@118,main:@531,main:@527,main:@529,main:@53,main:@533,main:@534) -> float_type, {32, 50, 76}, {3800, 76, 1}: 0.0214795ms, 0.189387% main:@536 = hip::hip_copy_literal[id=main:@literal:19] -> float_type, {76, 128}, {128, 1}: 0.00076962ms, 0.00678585% main:@537 = reshape_lazy[dims={1600, 76}](main:@535) -> float_type, {1600, 76}, {76, 1}: 0.00053596ms, 0.00472564% main:@538 = load[offset=6371584,end=7190784](main:@1) -> float_type, {1600, 128}, {128, 1}: 0.00048326ms, 0.00426097% main:@539 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@537,main:@536,main:@538) -> float_type, {1600, 128}, {128, 1}: 0.0238936ms, 0.210673% W3KAtTdC92:0 = @param:W3KAtTdC92:0 -> float_type, {32, 15, 17}, {255, 17, 1}: 0.0007295ms, 0.00643211% main:@541 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 15, 17}](W3KAtTdC92:0) -> float_type, {32, 15, 16}, {255, 17, 1}: 0.00067052ms, 0.00591207% a5QhOKE1Yc:0 = @param:a5QhOKE1Yc:0 -> float_type, {32, 15, 9}, {135, 9, 1}: 0.00055118ms, 0.00485983% main:@543 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 15, 9}](a5QhOKE1Yc:0) -> float_type, {32, 15, 8}, {135, 9, 1}: 0.00080696ms, 0.00711508% main:@544 = load[offset=1494144,end=1640064](main:@1) -> float_type, {32, 15, 76}, {1140, 76, 1}: 0.00048164ms, 0.00424669% ZebaMGw6vA:0 = @param:ZebaMGw6vA:0 -> float_type, {32, 15, 9}, {135, 9, 1}: 0.00055678ms, 0.00490921% main:@546 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 15, 9}](ZebaMGw6vA:0) -> float_type, {32, 15, 8}, {135, 9, 1}: 0.00062386ms, 0.00550066% wcHWv4SNE2:0 = @param:wcHWv4SNE2:0 -> float_type, {32, 15, 9}, {135, 9, 1}: 0.00050146ms, 0.00442145% main:@548 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 15, 9}](wcHWv4SNE2:0) -> float_type, {32, 15, 8}, {135, 9, 1}: 0.00098924ms, 0.00872227% main:@549 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=6080,local=256,](main:@543,main:@254,main:@548,main:@541,main:@288,main:@546,main:@544) -> float_type, {32, 15, 76}, {1140, 76, 1}: 0.0194628ms, 0.171606% main:@550 = hip::hip_copy_literal[id=main:@literal:26] -> float_type, {76, 64}, {64, 1}: 0.000753ms, 0.00663931% main:@551 = reshape_lazy[dims={480, 76}](main:@549) -> float_type, {480, 76}, {76, 1}: 0.00050672ms, 0.00446782% main:@552 = load[offset=1371264,end=1494144](main:@1) -> float_type, {480, 64}, {64, 1}: 0.00048106ms, 0.00424158% main:@553 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@551,main:@550,main:@552) -> float_type, {480, 64}, {64, 1}: 0.0213662ms, 0.188389% main:@554 = hip::hip_copy_literal[id=main:@literal:13] -> float_type, {236, 896}, {896, 1}: 0.00079058ms, 0.00697066% main:@555 = slice[axes={0, 1},starts={0, 129},ends={32, 161}](nv2tzWRf1I:0) -> float_type, {32, 32}, {301, 1}: 0.0007129ms, 0.00628574% Qp3K1PHvte:0 = @param:Qp3K1PHvte:0 -> float_type, {32, 69}, {69, 1}: 0.00051046ms, 0.0045008% main:@557 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](Qp3K1PHvte:0) -> float_type, {32, 16}, {69, 1}: 0.00083686ms, 0.00737872% S5AzoJmqUC:0 = @param:S5AzoJmqUC:0 -> float_type, {32, 13}, {13, 1}: 0.00050136ms, 0.00442056% main:@559 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](S5AzoJmqUC:0) -> float_type, {32, 4}, {13, 1}: 0.00063538ms, 0.00560224% w27LgiJW6P:0 = @param:w27LgiJW6P:0 -> float_type, {32, 13}, {13, 1}: 0.00049336ms, 0.00435003% main:@561 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](w27LgiJW6P:0) -> float_type, {32, 4}, {13, 1}: 0.00078342ms, 0.00690753% m5MfQhR1Bp:0 = @param:m5MfQhR1Bp:0 -> float_type, {32, 13}, {13, 1}: 0.0004817ms, 0.00424722% Vi28Dy4Ll5:0 = @param:Vi28Dy4Ll5:0 -> float_type, {32, 13}, {13, 1}: 0.0005733ms, 0.00505487% qeZ76Hc03v:0 = @param:qeZ76Hc03v:0 -> float_type, {32, 13}, {13, 1}: 0.00050788ms, 0.00447805% TxEdCcekGS:0 = @param:TxEdCcekGS:0 -> float_type, {32, 49}, {49, 1}: 0.00056472ms, 0.00497922% LKAvFme9ZS:0 = @param:LKAvFme9ZS:0 -> float_type, {32, 49}, {49, 1}: 0.00047234ms, 0.00416469% ifP6Ceko5T:0 = @param:ifP6Ceko5T:0 -> float_type, {32, 73}, {73, 1}: 0.00048746ms, 0.00429801% main:@568 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](TxEdCcekGS:0) -> float_type, {32, 8}, {49, 1}: 0.00080868ms, 0.00713025% main:@569 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](LKAvFme9ZS:0) -> float_type, {32, 8}, {49, 1}: 0.00062862ms, 0.00554263% KFhBukd7EV:0 = @param:KFhBukd7EV:0 -> float_type, {32, 13}, {13, 1}: 0.00053986ms, 0.00476002% main:@571 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](KFhBukd7EV:0) -> float_type, {32, 4}, {13, 1}: 0.00062546ms, 0.00551477% main:@572 = slice[axes={0, 1},starts={0, 257},ends={32, 289}](lyiDxfzZg3:0) -> float_type, {32, 32}, {701, 1}: 0.00092386ms, 0.00814581% main:@573 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](m5MfQhR1Bp:0) -> float_type, {32, 4}, {13, 1}: 0.0006158ms, 0.0054296% main:@574 = load[offset=1494144,end=1524352](main:@1) -> float_type, {32, 236}, {236, 1}: 0.00050168ms, 0.00442338% 0cjwsL5ldh:0 = @param:0cjwsL5ldh:0 -> float_type, {32, 33}, {33, 1}: 0.00049468ms, 0.00436166% main:@576 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](0cjwsL5ldh:0) -> float_type, {32, 16}, {33, 1}: 0.00068764ms, 0.00606302% 0sDBrPmfS9:0 = @param:0sDBrPmfS9:0 -> float_type, {32, 13}, {13, 1}: 0.00049966ms, 0.00440557% main:@578 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](0sDBrPmfS9:0) -> float_type, {32, 4}, {13, 1}: 0.00069978ms, 0.00617006% gU9SoiHyRO:0 = @param:gU9SoiHyRO:0 -> float_type, {32, 13}, {13, 1}: 0.00056452ms, 0.00497745% main:@580 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](gU9SoiHyRO:0) -> float_type, {32, 4}, {13, 1}: 0.00074496ms, 0.00656842% d1ExPaGqMN:0 = @param:d1ExPaGqMN:0 -> float_type, {32, 13}, {13, 1}: 0.00055064ms, 0.00485507% main:@582 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](d1ExPaGqMN:0) -> float_type, {32, 4}, {13, 1}: 0.0007264ms, 0.00640477% whRnWGgf0Q:0 = @param:whRnWGgf0Q:0 -> float_type, {32, 13}, {13, 1}: 0.00053088ms, 0.00468085% main:@584 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](whRnWGgf0Q:0) -> float_type, {32, 4}, {13, 1}: 0.00068092ms, 0.00600377% 3mk8BKjDey:0 = @param:3mk8BKjDey:0 -> float_type, {32, 25}, {25, 1}: 0.0005275ms, 0.00465104% main:@586 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](3mk8BKjDey:0) -> float_type, {32, 16}, {25, 1}: 0.00072024ms, 0.00635046% mIR0ToEnMH:0 = @param:mIR0ToEnMH:0 -> float_type, {32, 13}, {13, 1}: 0.0005359ms, 0.00472511% cHzIDJm6us:0 = @param:cHzIDJm6us:0 -> float_type, {32, 33}, {33, 1}: 0.00050476ms, 0.00445054% main:@589 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](qeZ76Hc03v:0) -> float_type, {32, 4}, {13, 1}: 0.00073136ms, 0.00644851% main:@590 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](Vi28Dy4Ll5:0) -> float_type, {32, 4}, {13, 1}: 0.0006521ms, 0.00574966% main:@591 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](mIR0ToEnMH:0) -> float_type, {32, 4}, {13, 1}: 0.00068222ms, 0.00601523% main:@592 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](cHzIDJm6us:0) -> float_type, {32, 16}, {33, 1}: 0.000653ms, 0.0057576% main:@593 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](ifP6Ceko5T:0) -> float_type, {32, 16}, {73, 1}: 0.00066786ms, 0.00588862% main:@594 = slice[axes={0, 1},starts={0, 129},ends={32, 161}](uFSOc5Chp2:0) -> float_type, {32, 32}, {481, 1}: 0.00064084ms, 0.00565038% main:@595 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=359,local=256,](main:@572,main:@555,main:@594,main:@593,main:@557,main:@569,main:@568,main:@592,main:@586,main:@576,main:@578,main:@580,main:@571,main:@589,main:@590,main:@573,main:@591,main:@582,main:@559,main:@584,main:@561,main:@574) -> float_type, {32, 236}, {236, 1}: 0.0358827ms, 0.316383% main:@596 = load[offset=3628800,end=3743488](main:@1) -> float_type, {32, 896}, {896, 1}: 0.00066478ms, 0.00586146% main:@597 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@595,main:@554,main:@596) -> float_type, {32, 896}, {896, 1}: 0.0253133ms, 0.223191% lDIuWZ8VFU:0 = @param:lDIuWZ8VFU:0 -> float_type, {32, 50, 5}, {250, 5, 1}: 0.0006802ms, 0.00599742% main:@599 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 50, 5}](lDIuWZ8VFU:0) -> float_type, {32, 50, 4}, {250, 5, 1}: 0.00070886ms, 0.00625012% waXyDlz530:0 = @param:waXyDlz530:0 -> float_type, {32, 50, 5}, {250, 5, 1}: 0.00055624ms, 0.00490445% main:@601 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 50, 5}](waXyDlz530:0) -> float_type, {32, 50, 4}, {250, 5, 1}: 0.00078228ms, 0.00689748% pHflUqMEaX:0 = @param:pHflUqMEaX:0 -> float_type, {32, 50, 5}, {250, 5, 1}: 0.00048144ms, 0.00424493% main:@603 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 50, 5}](pHflUqMEaX:0) -> float_type, {32, 50, 4}, {250, 5, 1}: 0.00066456ms, 0.00585952% main:@604 = load[offset=3743488,end=4281088](main:@1) -> float_type, {32, 50, 84}, {4200, 84, 1}: 0.0004683ms, 0.00412907% main:@605 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=22400,local=256,](main:@603,main:@183,main:@141,main:@601,main:@599,main:@19,main:@604) -> float_type, {32, 50, 84}, {4200, 84, 1}: 0.0212619ms, 0.187469% main:@606 = hip::hip_copy_literal[id=main:@literal:15] -> float_type, {84, 256}, {256, 1}: 0.00078776ms, 0.00694579% main:@607 = load[offset=1494144,end=3132544](main:@1) -> float_type, {1600, 256}, {256, 1}: 0.00049044ms, 0.00432428% main:@608 = reshape_lazy[dims={1600, 84}](main:@605) -> float_type, {1600, 84}, {84, 1}: 0.00054134ms, 0.00477307% main:@609 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@608,main:@606,main:@607) -> float_type, {1600, 256}, {256, 1}: 0.0264682ms, 0.233374% main:@610 = hip::hip_copy_literal[id=main:@literal:21] -> float_type, {64, 64}, {64, 1}: 0.000823ms, 0.00725651% main:@611 = load[offset=3539072,end=3620992](main:@1) -> float_type, {32, 10, 64}, {640, 64, 1}: 0.00053476ms, 0.00471506% main:@612 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=5120,local=256,](main:@162,main:@75,main:@239,main:@210,main:@611) -> float_type, {32, 10, 64}, {640, 64, 1}: 0.0162796ms, 0.143539% main:@613 = reshape_lazy[dims={320, 64}](main:@612) -> float_type, {320, 64}, {64, 1}: 0.00069544ms, 0.00613179% main:@614 = load[offset=3751680,end=3833600](main:@1) -> float_type, {320, 64}, {64, 1}: 0.00050582ms, 0.00445989% main:@615 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@613,main:@610,main:@614) -> float_type, {320, 64}, {64, 1}: 0.0183396ms, 0.161703% main:@616 = hip::hip_copy_literal[id=main:@literal:25] -> float_type, {130, 128}, {128, 1}: 0.00078548ms, 0.00692569% FcmWLZhDgE:0 = @param:FcmWLZhDgE:0 -> float_type, {32, 57}, {57, 1}: 0.00056572ms, 0.00498803% main:@618 = slice[axes={0, 1},starts={0, 41},ends={32, 49}](FcmWLZhDgE:0) -> float_type, {32, 8}, {57, 1}: 0.00066024ms, 0.00582143% main:@619 = slice[axes={0, 1},starts={0, 53},ends={32, 61}](Qp3K1PHvte:0) -> float_type, {32, 8}, {69, 1}: 0.00079254ms, 0.00698794% main:@620 = load[offset=3539072,end=3555712](main:@1) -> float_type, {32, 130}, {130, 1}: 0.00049902ms, 0.00439993% bJsIAzhYc3:0 = @param:bJsIAzhYc3:0 -> float_type, {32, 37}, {37, 1}: 0.00051846ms, 0.00457134% isCn4mEOSt:0 = @param:isCn4mEOSt:0 -> float_type, {32, 49}, {49, 1}: 0.00051824ms, 0.0045694% 2CnSyUBzm0:0 = @param:2CnSyUBzm0:0 -> float_type, {32, 65}, {65, 1}: 0.00047254ms, 0.00416645% ljyhc7HCLk:0 = @param:ljyhc7HCLk:0 -> float_type, {32, 21}, {21, 1}: 0.00053382ms, 0.00470677% 9YqhflbMKe:0 = @param:9YqhflbMKe:0 -> float_type, {32, 29}, {29, 1}: 0.00049692ms, 0.00438142% main:@626 = slice[axes={0, 1},starts={0, 21},ends={32, 29}](9YqhflbMKe:0) -> float_type, {32, 8}, {29, 1}: 0.0007754ms, 0.00683681% ZcWERoz0h6:0 = @param:ZcWERoz0h6:0 -> float_type, {32, 49}, {49, 1}: 0.00050404ms, 0.00444419% main:@628 = slice[axes={0, 1},starts={0, 37},ends={32, 41}](ZcWERoz0h6:0) -> float_type, {32, 4}, {49, 1}: 0.00063478ms, 0.00559695% main:@629 = slice[axes={0, 1},starts={0, 13},ends={32, 21}](ljyhc7HCLk:0) -> float_type, {32, 8}, {21, 1}: 0.0005986ms, 0.00527794% YbIVz8l1un:0 = @param:YbIVz8l1un:0 -> float_type, {32, 45}, {45, 1}: 0.00051176ms, 0.00451226% main:@631 = slice[axes={0, 1},starts={0, 29},ends={32, 37}](YbIVz8l1un:0) -> float_type, {32, 8}, {45, 1}: 0.00079532ms, 0.00701245% main:@632 = slice[axes={0, 1},starts={0, 197},ends={32, 237}](nv2tzWRf1I:0) -> float_type, {32, 40}, {301, 1}: 0.00061286ms, 0.00540367% main:@633 = slice[axes={0, 1},starts={0, 509},ends={32, 517}](lyiDxfzZg3:0) -> float_type, {32, 8}, {701, 1}: 0.0005978ms, 0.00527089% main:@634 = slice[axes={0, 1},starts={0, 37},ends={32, 41}](isCn4mEOSt:0) -> float_type, {32, 4}, {49, 1}: 0.00065936ms, 0.00581367% w04Keh2FJu:0 = @param:w04Keh2FJu:0 -> float_type, {32, 13}, {13, 1}: 0.00051246ms, 0.00451843% main:@636 = slice[axes={0, 1},starts={0, 11},ends={32, 13}](w04Keh2FJu:0) -> float_type, {32, 2}, {13, 1}: 0.00072614ms, 0.00640248% main:@637 = slice[axes={0, 1},starts={0, 53},ends={32, 57}](2CnSyUBzm0:0) -> float_type, {32, 4}, {65, 1}: 0.00060118ms, 0.00530069% main:@638 = slice[axes={0, 1},starts={0, 25},ends={32, 29}](bJsIAzhYc3:0) -> float_type, {32, 4}, {37, 1}: 0.00064714ms, 0.00570593% main:@639 = slice[axes={0, 1},starts={0, 217},ends={32, 233}](uFSOc5Chp2:0) -> float_type, {32, 16}, {481, 1}: 0.0009139ms, 0.00805799% main:@640 = slice[axes={0, 1},starts={0, 57},ends={32, 65}](ifP6Ceko5T:0) -> float_type, {32, 8}, {73, 1}: 0.00065642ms, 0.00578775% main:@641 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=297,local=256,](main:@633,main:@632,main:@639,main:@640,main:@619,main:@637,main:@631,main:@634,main:@628,main:@636,main:@618,main:@638,main:@629,main:@626,main:@620) -> float_type, {32, 130}, {130, 1}: 0.0277611ms, 0.244774% main:@642 = load[offset=3579648,end=3596032](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00064002ms, 0.00564315% main:@643 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@641,main:@616,main:@642) -> float_type, {32, 128}, {128, 1}: 0.0209281ms, 0.184526% main:@644 = slice[axes={0, 1},starts={0, 33},ends={32, 41}](FcmWLZhDgE:0) -> float_type, {32, 8}, {57, 1}: 0.00081918ms, 0.00722283% main:@645 = slice[axes={0, 1},starts={0, 21},ends={32, 29}](YbIVz8l1un:0) -> float_type, {32, 8}, {45, 1}: 0.0006386ms, 0.00563063% main:@646 = slice[axes={0, 1},starts={0, 41},ends={32, 49}](ifP6Ceko5T:0) -> float_type, {32, 8}, {73, 1}: 0.00079896ms, 0.00704455% main:@647 = slice[axes={0, 1},starts={0, 37},ends={32, 45}](Qp3K1PHvte:0) -> float_type, {32, 8}, {69, 1}: 0.00063652ms, 0.00561229% main:@648 = slice[axes={0, 1},starts={0, 5},ends={32, 13}](9YqhflbMKe:0) -> float_type, {32, 8}, {29, 1}: 0.00061006ms, 0.00537899% main:@649 = slice[axes={0, 1},starts={0, 501},ends={32, 509}](lyiDxfzZg3:0) -> float_type, {32, 8}, {701, 1}: 0.0006091ms, 0.00537052% main:@650 = slice[axes={0, 1},starts={0, 181},ends={32, 189}](nv2tzWRf1I:0) -> float_type, {32, 8}, {301, 1}: 0.00075898ms, 0.00669204% main:@651 = slice[axes={0, 1},starts={0, 9},ends={32, 11}](w04Keh2FJu:0) -> float_type, {32, 2}, {13, 1}: 0.00061788ms, 0.00544794% main:@652 = slice[axes={0, 1},starts={0, 33},ends={32, 37}](isCn4mEOSt:0) -> float_type, {32, 4}, {49, 1}: 0.00060596ms, 0.00534284% main:@653 = slice[axes={0, 1},starts={0, 41},ends={32, 45}](2CnSyUBzm0:0) -> float_type, {32, 4}, {65, 1}: 0.00060616ms, 0.0053446% main:@654 = load[offset=514048,end=524544](main:@1) -> float_type, {32, 82}, {82, 1}: 0.00047572ms, 0.00419449% main:@655 = slice[axes={0, 1},starts={0, 5},ends={32, 13}](ljyhc7HCLk:0) -> float_type, {32, 8}, {21, 1}: 0.00076054ms, 0.00670579% main:@656 = slice[axes={0, 1},starts={0, 33},ends={32, 37}](ZcWERoz0h6:0) -> float_type, {32, 4}, {49, 1}: 0.00060632ms, 0.00534601% main:@657 = slice[axes={0, 1},starts={0, 21},ends={32, 25}](bJsIAzhYc3:0) -> float_type, {32, 4}, {37, 1}: 0.00084102ms, 0.00741539% main:@658 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=201,local=256,](main:@649,main:@650,main:@646,main:@647,main:@653,main:@645,main:@652,main:@656,main:@651,main:@644,main:@657,main:@655,main:@648,main:@654) -> float_type, {32, 82}, {82, 1}: 0.0270645ms, 0.238632% main:@659 = hip::hip_copy_literal[id=main:@literal:12] -> float_type, {82, 256}, {256, 1}: 0.0007818ms, 0.00689324% main:@660 = load[offset=3596032,end=3628800](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00048304ms, 0.00425903% main:@661 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@658,main:@659,main:@660) -> float_type, {32, 256}, {256, 1}: 0.020495ms, 0.180708% main:@662 = hip::hip_copy_literal[id=main:@literal:18] -> float_type, {128, 18, 64}, {1152, 64, 1}: 0.0007468ms, 0.00658464% main:@663 = hip::hip_copy_literal[id=main:@literal:10] -> float_type, {128, 18, 18}, {324, 18, 1}: 0.0006568ms, 0.0057911% main:@664 = load[offset=4423424,end=4589312](main:@1) -> float_type, {128, 18, 18}, {324, 18, 1}: 0.00051776ms, 0.00456516% main:@665 = gpu::code_object[code_object=14560,symbol_name=softmax_kernel,global=147456,local=64,](main:@663,main:@664) -> float_type, {128, 18, 18}, {324, 18, 1}: 0.0207362ms, 0.182834% main:@666 = load[offset=3833600,end=4423424](main:@1) -> float_type, {128, 18, 64}, {1152, 64, 1}: 0.00066998ms, 0.00590731% main:@667 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@665,main:@662,main:@666) -> float_type, {128, 18, 64}, {1152, 64, 1}: 0.0235755ms, 0.207869% main:@668 = load[offset=7190784,end=7805184](main:@1) -> float_type, {32, 50, 96}, {4800, 96, 1}: 0.0006639ms, 0.0058537% SKG7aiAo3m:0 = @param:SKG7aiAo3m:0 -> float_type, {32, 50, 9}, {450, 9, 1}: 0.00053104ms, 0.00468226% main:@670 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 50, 9}](SKG7aiAo3m:0) -> float_type, {32, 50, 8}, {450, 9, 1}: 0.00066304ms, 0.00584612% hKxguEQn94:0 = @param:hKxguEQn94:0 -> float_type, {32, 50, 9}, {450, 9, 1}: 0.00052408ms, 0.00462089% main:@672 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 50, 9}](hKxguEQn94:0) -> float_type, {32, 50, 8}, {450, 9, 1}: 0.00078358ms, 0.00690894% D0gn7y6swk:0 = @param:D0gn7y6swk:0 -> float_type, {32, 50, 17}, {850, 17, 1}: 0.00053232ms, 0.00469354% main:@674 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 50, 17}](D0gn7y6swk:0) -> float_type, {32, 50, 16}, {850, 17, 1}: 0.00061522ms, 0.00542448% 0pnqc4Em1y:0 = @param:0pnqc4Em1y:0 -> float_type, {32, 50, 9}, {450, 9, 1}: 0.00051914ms, 0.00457733% main:@676 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 50, 9}](0pnqc4Em1y:0) -> float_type, {32, 50, 8}, {450, 9, 1}: 0.00085084ms, 0.00750198% UpWkNPIeJO:0 = @param:UpWkNPIeJO:0 -> float_type, {32, 50, 17}, {850, 17, 1}: 0.00053476ms, 0.00471506% main:@678 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 50, 17}](UpWkNPIeJO:0) -> float_type, {32, 50, 16}, {850, 17, 1}: 0.00062956ms, 0.00555092% nEuw2U7mFR:0 = @param:nEuw2U7mFR:0 -> float_type, {32, 50, 41}, {2050, 41, 1}: 0.00045346ms, 0.00399822% main:@680 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 50, 41}](nEuw2U7mFR:0) -> float_type, {32, 50, 40}, {2050, 41, 1}: 0.00076252ms, 0.00672325% main:@681 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=25600,local=256,](main:@680,main:@672,main:@670,main:@674,main:@678,main:@676,main:@668) -> float_type, {32, 50, 96}, {4800, 96, 1}: 0.0240708ms, 0.212236% main:@682 = hip::hip_copy_literal[id=main:@literal:11] -> float_type, {96, 256}, {256, 1}: 0.0007812ms, 0.00688795% main:@683 = reshape_lazy[dims={1600, 96}](main:@681) -> float_type, {1600, 96}, {96, 1}: 0.00054546ms, 0.0048094% main:@684 = load[offset=4733184,end=6371584](main:@1) -> float_type, {1600, 256}, {256, 1}: 0.0004711ms, 0.00415376% main:@685 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@683,main:@682,main:@684) -> float_type, {1600, 256}, {256, 1}: 0.0237622ms, 0.209514% main:@686 = hip::hip_copy_literal[id=main:@literal:155] -> float_type, {732, 128}, {128, 1}: 0.00085218ms, 0.00751379% main:@687 = load[offset=4423424,end=4517120](main:@1) -> float_type, {32, 732}, {732, 1}: 0.00046674ms, 0.00411531% main:@688 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=11712,local=256,](main:@132,main:@334,main:@687) -> float_type, {32, 732}, {732, 1}: 0.0156196ms, 0.137721% main:@689 = load[offset=3539072,end=3555456](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00068592ms, 0.00604786% main:@690 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@688,main:@686,main:@689) -> float_type, {32, 128}, {128, 1}: 0.0376887ms, 0.332307% main:@691 = load[offset=4423424,end=4439808](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00063176ms, 0.00557032% main:@692 = gpu::code_object[code_object=13520,symbol_name=log_add_kernel,global=2048,local=256,](main:@690,main:@691) -> float_type, {32, 128}, {128, 1}: 0.0133407ms, 0.117627% main:@693 = load[offset=3539072,end=3569792](main:@1) -> float_type, {32, 240}, {240, 1}: 0.00063146ms, 0.00556767% main:@694 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@692,main:@336,main:@693) -> float_type, {32, 240}, {240, 1}: 0.0205323ms, 0.181037% main:@695 = hip::hip_copy_literal[id=main:@literal:118] -> float_type, {732, 256}, {256, 1}: 0.00077848ms, 0.00686397% main:@696 = load[offset=4423424,end=4517120](main:@1) -> float_type, {32, 732}, {732, 1}: 0.0004733ms, 0.00417315% main:@697 = gpu::code_object[code_object=13560,symbol_name=noop_rsqrt_add_concat_noop_kernel,global=11712,local=256,](main:@132,main:@694,main:@334,main:@696) -> float_type, {32, 732}, {732, 1}: 0.0157967ms, 0.139281% main:@698 = load[offset=0,end=32768](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00062598ms, 0.00551936% main:@699 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@697,main:@695,main:@698) -> float_type, {32, 256}, {256, 1}: 0.0380612ms, 0.335591% main:@700 = load[offset=4423424,end=4565760](main:@1) -> float_type, {32, 1112}, {1112, 1}: 0.00062288ms, 0.00549202% main:@701 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=17792,local=256,](main:@189,main:@334,main:@700) -> float_type, {32, 1112}, {1112, 1}: 0.0161269ms, 0.142193% main:@702 = hip::hip_copy_literal[id=main:@literal:128] -> float_type, {1112, 128}, {128, 1}: 0.00072988ms, 0.00643546% main:@703 = load[offset=32768,end=49152](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00049058ms, 0.00432551% main:@704 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@701,main:@702,main:@703) -> float_type, {32, 128}, {128, 1}: 0.0250856ms, 0.221184% main:@705 = hip::hip_copy_literal[id=main:@literal:67] -> float_type, {128, 240}, {240, 1}: 0.00079826ms, 0.00703837% main:@706 = load[offset=4423424,end=4439808](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00048752ms, 0.00429853% main:@707 = gpu::code_object[code_object=13520,symbol_name=neg_add_kernel,global=2048,local=256,](main:@704,main:@706) -> float_type, {32, 128}, {128, 1}: 0.0131672ms, 0.116097% main:@708 = load[offset=3539072,end=3569792](main:@1) -> float_type, {32, 240}, {240, 1}: 0.0006113ms, 0.00538992% main:@709 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@707,main:@705,main:@708) -> float_type, {32, 240}, {240, 1}: 0.020422ms, 0.180064% main:@710 = hip::hip_copy_literal[id=main:@literal:87] -> float_type, {1112, 256}, {256, 1}: 0.00072438ms, 0.00638696% main:@711 = load[offset=4423424,end=4565760](main:@1) -> float_type, {32, 1112}, {1112, 1}: 0.00047538ms, 0.00419149% main:@712 = gpu::code_object[code_object=13552,symbol_name=noop_log_add_concat_noop_kernel,global=17792,local=256,](main:@189,main:@709,main:@334,main:@711) -> float_type, {32, 1112}, {1112, 1}: 0.0165048ms, 0.145525% main:@713 = load[offset=3132544,end=3165312](main:@1) -> float_type, {32, 256}, {256, 1}: 0.0006415ms, 0.0056562% main:@714 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@712,main:@710,main:@713) -> float_type, {32, 256}, {256, 1}: 0.0309658ms, 0.27303% main:@715 = hip::hip_copy_literal[id=main:@literal:53] -> float_type, {2312, 128}, {128, 1}: 0.00080042ms, 0.00705742% main:@716 = load[offset=4423424,end=4719360](main:@1) -> float_type, {32, 2312}, {2312, 1}: 0.00047804ms, 0.00421495% main:@717 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=36992,local=256,](main:@80,main:@334,main:@716) -> float_type, {32, 2312}, {2312, 1}: 0.018309ms, 0.161433% main:@718 = load[offset=3165312,end=3181696](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00065142ms, 0.00574366% main:@719 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@717,main:@715,main:@718) -> float_type, {32, 128}, {128, 1}: 0.0303215ms, 0.267349% main:@720 = load[offset=32768,end=49152](main:@1) -> float_type, {32, 128}, {128, 1}: 0.0006161ms, 0.00543224% main:@721 = gpu::code_object[code_object=13520,symbol_name=rsqrt_add_kernel,global=2048,local=256,](main:@719,main:@720) -> float_type, {32, 128}, {128, 1}: 0.0132504ms, 0.116831% main:@722 = load[offset=3165312,end=3196032](main:@1) -> float_type, {32, 240}, {240, 1}: 0.00062166ms, 0.00548127% main:@723 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@721,main:@335,main:@722) -> float_type, {32, 240}, {240, 1}: 0.0211791ms, 0.186739% main:@724 = hip::hip_copy_literal[id=main:@literal:144] -> float_type, {2312, 256}, {256, 1}: 0.00075064ms, 0.0066185% main:@725 = load[offset=4423424,end=4719360](main:@1) -> float_type, {32, 2312}, {2312, 1}: 0.000505ms, 0.00445266% main:@726 = gpu::code_object[code_object=13552,symbol_name=noop_tanh_pow_concat_noop_kernel,global=36992,local=256,](main:@80,main:@723,main:@334,main:@725) -> float_type, {32, 2312}, {2312, 1}: 0.0184563ms, 0.162732% main:@727 = load[offset=773248,end=806016](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00062498ms, 0.00551054% main:@728 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@726,main:@724,main:@727) -> float_type, {32, 256}, {256, 1}: 0.0320617ms, 0.282693% main:@729 = load[offset=32768,end=49152](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00060816ms, 0.00536223% main:@730 = gpu::code_object[code_object=13520,symbol_name=neg_div_kernel,global=2048,local=256,](main:@364,main:@729) -> float_type, {32, 128}, {128, 1}: 0.0132778ms, 0.117073% main:@731 = load[offset=806016,end=836736](main:@1) -> float_type, {32, 240}, {240, 1}: 0.00063438ms, 0.00559342% main:@732 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@730,main:@335,main:@731) -> float_type, {32, 240}, {240, 1}: 0.0207154ms, 0.18265% main:@733 = hip::hip_copy_literal[id=main:@literal:69] -> float_type, {2232, 256}, {256, 1}: 0.00076264ms, 0.00672431% main:@734 = load[offset=4423424,end=4709120](main:@1) -> float_type, {32, 2232}, {2232, 1}: 0.00047206ms, 0.00416222% main:@735 = gpu::code_object[code_object=13560,symbol_name=noop_rsqrt_add_concat_noop_kernel,global=35712,local=256,](main:@200,main:@732,main:@334,main:@734) -> float_type, {32, 2232}, {2232, 1}: 0.0183335ms, 0.161649% main:@736 = load[offset=32768,end=65536](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00065028ms, 0.00573361% main:@737 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@735,main:@733,main:@736) -> float_type, {32, 256}, {256, 1}: 0.0309796ms, 0.273151% main:@738 = hip::hip_copy_literal[id=main:@literal:35] -> float_type, {256, 768}, {768, 1}: 0.00079674ms, 0.00702497% main:@739 = unsqueeze[axes={1},steps={}](main:@359) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.00065868ms, 0.00580768% main:@740 = load[offset=8665344,end=9156864](main:@1) -> float_type, {32, 15, 256}, {3840, 256, 1}: 0.00048864ms, 0.00430841% main:@741 = unsqueeze[axes={1},steps={}](main:@379) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.0005141ms, 0.00453289% main:@742 = unsqueeze[axes={1},steps={}](main:@394) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.00050798ms, 0.00447893% main:@743 = unsqueeze[axes={1},steps={}](main:@399) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.0005044ms, 0.00444737% main:@744 = unsqueeze[axes={1},steps={}](main:@413) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.00050422ms, 0.00444578% main:@745 = unsqueeze[axes={1},steps={}](main:@428) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.0004887ms, 0.00430894% main:@746 = unsqueeze[axes={1},steps={}](main:@442) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.00049958ms, 0.00440487% main:@747 = unsqueeze[axes={1},steps={}](main:@457) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.00051016ms, 0.00449815% main:@748 = unsqueeze[axes={1},steps={}](main:@471) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.00048404ms, 0.00426785% main:@749 = unsqueeze[axes={1},steps={}](main:@486) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.0004775ms, 0.00421019% main:@750 = unsqueeze[axes={1},steps={}](main:@501) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.00047682ms, 0.00420419% main:@751 = unsqueeze[axes={1},steps={}](main:@699) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.000476ms, 0.00419696% main:@752 = unsqueeze[axes={1},steps={}](main:@714) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.00047542ms, 0.00419185% main:@753 = unsqueeze[axes={1},steps={}](main:@728) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.00048216ms, 0.00425127% main:@754 = unsqueeze[axes={1},steps={}](main:@737) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.00047678ms, 0.00420384% main:@755 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=4096,local=256,](main:@752,main:@745,main:@753,main:@754,main:@742,main:@749,main:@751,main:@746,main:@743,main:@750,main:@739,main:@747,main:@744,main:@741,main:@748,main:@740) -> float_type, {32, 15, 256}, {3840, 256, 1}: 0.025319ms, 0.223241% main:@756 = load[offset=7190784,end=8665344](main:@1) -> float_type, {480, 768}, {768, 1}: 0.00063656ms, 0.00561264% main:@757 = reshape_lazy[dims={480, 256}](main:@755) -> float_type, {480, 256}, {256, 1}: 0.00054606ms, 0.00481469% main:@758 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@757,main:@738,main:@756) -> float_type, {480, 768}, {768, 1}: 0.0344294ms, 0.303569% main:@759 = hip::hip_copy_literal[id=main:@literal:164] -> float_type, {256, 512}, {512, 1}: 0.0007972ms, 0.00702903% main:@760 = hip::hip_copy_literal[id=main:@literal:7] -> float_type, {896}, {1}: 0.00065136ms, 0.00574313% main:@761 = hip::hip_copy_literal[id=main:@literal:23] -> float_type, {32, 18, 256}, {4608, 256, 1}: 0.00060822ms, 0.00536276% main:@762 = hip::hip_copy_literal[id=main:@literal:146] -> float_type, {256, 512}, {512, 1}: 0.00059258ms, 0.00522486% main:@763 = multibroadcast[out_lens={32, 896}](main:@760) -> float_type, {32, 896}, {0, 1}: 0.00074636ms, 0.00658076% main:@764 = load[offset=0,end=114688](main:@1) -> float_type, {32, 896}, {896, 1}: 0.00054338ms, 0.00479106% main:@765 = gpu::code_object[code_object=13536,symbol_name=add_leaky_relu_kernel,global=14336,local=256,](main:@597,main:@763,main:@764) -> float_type, {32, 896}, {896, 1}: 0.0155758ms, 0.137334% main:@766 = hip::hip_copy_literal[id=main:@literal:162] -> float_type, {256}, {1}: 0.00079888ms, 0.00704384% main:@767 = hip::hip_copy_literal[id=main:@literal:163] -> float_type, {256}, {1}: 0.00061196ms, 0.00539574% main:@768 = hip::hip_copy_literal[id=main:@literal:154] -> float_type, {512, 256}, {256, 1}: 0.00063304ms, 0.0055816% main:@769 = load[offset=8665344,end=9255168](main:@1) -> float_type, {32, 18, 256}, {4608, 256, 1}: 0.00051602ms, 0.00454982% main:@770 = slice[axes={0},starts={96},ends={128}](main:@667) -> float_type, {32, 18, 64}, {1152, 64, 1}: 0.00097374ms, 0.00858561% main:@771 = slice[axes={0},starts={64},ends={96}](main:@667) -> float_type, {32, 18, 64}, {1152, 64, 1}: 0.00060484ms, 0.00533296% main:@772 = slice[axes={0},starts={32},ends={64}](main:@667) -> float_type, {32, 18, 64}, {1152, 64, 1}: 0.0005631ms, 0.00496493% main:@773 = slice[axes={0},starts={0},ends={32}](main:@667) -> float_type, {32, 18, 64}, {1152, 64, 1}: 0.00055072ms, 0.00485578% main:@774 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=36864,local=256,](main:@773,main:@772,main:@771,main:@770,main:@769) -> float_type, {32, 18, 256}, {4608, 256, 1}: 0.0198249ms, 0.174799% main:@775 = multibroadcast[out_lens={32, 18, 256}](main:@767) -> float_type, {32, 18, 256}, {0, 0, 1}: 0.00073414ms, 0.00647302% main:@776 = multibroadcast[out_lens={32, 18, 256}](main:@766) -> float_type, {32, 18, 256}, {0, 0, 1}: 0.00068224ms, 0.00601541% main:@777 = load[offset=781440,end=1371264](main:@1) -> float_type, {32, 18, 256}, {4608, 256, 1}: 0.00051494ms, 0.0045403% main:@778 = gpu::code_object[code_object=13624,symbol_name=mul_relu_reduce_prod_sub_mul_reduce_sum_add_add_add_add_pow_kernel,global=73728,local=128,](main:@774,main:@761,main:@776,main:@775,main:@777) -> float_type, {32, 18, 256}, {4608, 256, 1}: 0.0220324ms, 0.194262% yFeB2xg6sO:0 = @param:yFeB2xg6sO:0 -> float_type, {32, 15, 9}, {135, 9, 1}: 0.00069554ms, 0.00613268% main:@780 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 15, 9}](yFeB2xg6sO:0) -> float_type, {32, 15, 8}, {135, 9, 1}: 0.00067504ms, 0.00595192% T3LHlorDnM:0 = @param:T3LHlorDnM:0 -> float_type, {32, 15, 17}, {255, 17, 1}: 0.00051516ms, 0.00454224% main:@782 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 15, 17}](T3LHlorDnM:0) -> float_type, {32, 15, 16}, {255, 17, 1}: 0.00083272ms, 0.00734221% main:@783 = load[offset=3132544,end=3286144](main:@1) -> float_type, {32, 15, 80}, {1200, 80, 1}: 0.00051306ms, 0.00452372% hGtXM6ZJUv:0 = @param:hGtXM6ZJUv:0 -> float_type, {32, 15, 9}, {135, 9, 1}: 0.0005801ms, 0.00511483% main:@785 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 15, 9}](hGtXM6ZJUv:0) -> float_type, {32, 15, 8}, {135, 9, 1}: 0.00064924ms, 0.00572444% kw7Kzgp1JD:0 = @param:kw7Kzgp1JD:0 -> float_type, {32, 15, 21}, {315, 21, 1}: 0.0005662ms, 0.00499227% main:@787 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 15, 21}](kw7Kzgp1JD:0) -> float_type, {32, 15, 20}, {315, 21, 1}: 0.00101454ms, 0.00894535% main:@788 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=7680,local=256,](main:@785,main:@782,main:@62,main:@787,main:@780,main:@783) -> float_type, {32, 15, 80}, {1200, 80, 1}: 0.0186955ms, 0.164841% main:@789 = hip::hip_copy_literal[id=main:@literal:14] -> float_type, {80, 64}, {64, 1}: 0.00074874ms, 0.00660175% main:@790 = reshape_lazy[dims={480, 80}](main:@788) -> float_type, {480, 80}, {80, 1}: 0.00050954ms, 0.00449269% main:@791 = load[offset=3628800,end=3751680](main:@1) -> float_type, {480, 64}, {64, 1}: 0.00046998ms, 0.00414388% main:@792 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@790,main:@789,main:@791) -> float_type, {480, 64}, {64, 1}: 0.0213101ms, 0.187895% RpITWntAji:0 = @param:RpITWntAji:0 -> float_type, {32, 17}, {17, 1}: 0.00069906ms, 0.00616371% main:@794 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](RpITWntAji:0) -> float_type, {32, 8}, {17, 1}: 0.00094252ms, 0.00831033% IztrlRfWQU:0 = @param:IztrlRfWQU:0 -> float_type, {32, 13}, {13, 1}: 0.00051958ms, 0.00458121% main:@796 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](IztrlRfWQU:0) -> float_type, {32, 2}, {13, 1}: 0.00080624ms, 0.00710873% JSWhLK4Agx:0 = @param:JSWhLK4Agx:0 -> float_type, {32, 25}, {25, 1}: 0.00049378ms, 0.00435373% main:@798 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](JSWhLK4Agx:0) -> float_type, {32, 8}, {25, 1}: 0.00063658ms, 0.00561282% KN13IWXZSu:0 = @param:KN13IWXZSu:0 -> float_type, {32, 13}, {13, 1}: 0.00049514ms, 0.00436572% main:@800 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](KN13IWXZSu:0) -> float_type, {32, 2}, {13, 1}: 0.00077696ms, 0.00685057% NXOSdQVPs0:0 = @param:NXOSdQVPs0:0 -> float_type, {32, 25}, {25, 1}: 0.00053784ms, 0.00474221% main:@802 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](NXOSdQVPs0:0) -> float_type, {32, 8}, {25, 1}: 0.0006698ms, 0.00590572% ry2xad1vc9:0 = @param:ry2xad1vc9:0 -> float_type, {32, 25}, {25, 1}: 0.000498ms, 0.00439094% main:@804 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](ry2xad1vc9:0) -> float_type, {32, 8}, {25, 1}: 0.00079272ms, 0.00698953% main:@805 = slice[axes={0, 1},starts={0, 17},ends={32, 25}](Qp3K1PHvte:0) -> float_type, {32, 8}, {69, 1}: 0.0006081ms, 0.00536171% RHo9Jz6KpI:0 = @param:RHo9Jz6KpI:0 -> float_type, {32, 21}, {21, 1}: 0.00055734ms, 0.00491415% main:@807 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](RHo9Jz6KpI:0) -> float_type, {32, 4}, {21, 1}: 0.00064532ms, 0.00568988% S5D6WGIXMd:0 = @param:S5D6WGIXMd:0 -> float_type, {32, 17}, {17, 1}: 0.00061304ms, 0.00540526% main:@809 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](S5D6WGIXMd:0) -> float_type, {32, 4}, {17, 1}: 0.00080608ms, 0.00710732% TidKGrnobI:0 = @param:TidKGrnobI:0 -> float_type, {32, 13}, {13, 1}: 0.00049998ms, 0.0044084% main:@811 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](TidKGrnobI:0) -> float_type, {32, 2}, {13, 1}: 0.00061162ms, 0.00539274% m2YscaoDjP:0 = @param:m2YscaoDjP:0 -> float_type, {32, 17}, {17, 1}: 0.00048454ms, 0.00427226% main:@813 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](m2YscaoDjP:0) -> float_type, {32, 4}, {17, 1}: 0.00068344ms, 0.00602599% Tg4u0GfnFb:0 = @param:Tg4u0GfnFb:0 -> float_type, {32, 25}, {25, 1}: 0.00047278ms, 0.00416857% xT9FhjM1Uu:0 = @param:xT9FhjM1Uu:0 -> float_type, {32, 17}, {17, 1}: 0.00053642ms, 0.00472969% nPUSZFm74Y:0 = @param:nPUSZFm74Y:0 -> float_type, {32, 13}, {13, 1}: 0.00055784ms, 0.00491856% 5JHUPChra8:0 = @param:5JHUPChra8:0 -> float_type, {32, 17}, {17, 1}: 0.00048638ms, 0.00428848% main:@818 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](5JHUPChra8:0) -> float_type, {32, 2}, {17, 1}: 0.0007536ms, 0.0066446% xymEBUMS25:0 = @param:xymEBUMS25:0 -> float_type, {32, 13}, {13, 1}: 0.00050098ms, 0.00441721% main:@820 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](xymEBUMS25:0) -> float_type, {32, 2}, {13, 1}: 0.00072486ms, 0.0063912% pH5srbufJ1:0 = @param:pH5srbufJ1:0 -> float_type, {32, 17}, {17, 1}: 0.00052424ms, 0.0046223% main:@822 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](pH5srbufJ1:0) -> float_type, {32, 4}, {17, 1}: 0.00070024ms, 0.00617412% zZHaVOckpw:0 = @param:zZHaVOckpw:0 -> float_type, {32, 13}, {13, 1}: 0.00054214ms, 0.00478013% main:@824 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](zZHaVOckpw:0) -> float_type, {32, 2}, {13, 1}: 0.00069262ms, 0.00610693% msiPzSBKMj:0 = @param:msiPzSBKMj:0 -> float_type, {32, 17}, {17, 1}: 0.00050662ms, 0.00446694% main:@826 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](msiPzSBKMj:0) -> float_type, {32, 4}, {17, 1}: 0.0006818ms, 0.00601153% 92MwiALCHX:0 = @param:92MwiALCHX:0 -> float_type, {32, 25}, {25, 1}: 0.000471ms, 0.00415287% main:@828 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](92MwiALCHX:0) -> float_type, {32, 8}, {25, 1}: 0.00066842ms, 0.00589356% AVLcBpWeuo:0 = @param:AVLcBpWeuo:0 -> float_type, {32, 21}, {21, 1}: 0.000568ms, 0.00500814% main:@830 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](AVLcBpWeuo:0) -> float_type, {32, 4}, {21, 1}: 0.0006763ms, 0.00596303% ujOA5LQW7M:0 = @param:ujOA5LQW7M:0 -> float_type, {32, 13}, {13, 1}: 0.00046622ms, 0.00411073% main:@832 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](ujOA5LQW7M:0) -> float_type, {32, 2}, {13, 1}: 0.00095224ms, 0.00839604% main:@833 = slice[axes={0, 1},starts={0, 17},ends={32, 21}](ifP6Ceko5T:0) -> float_type, {32, 4}, {73, 1}: 0.00065794ms, 0.00580115% HZ0l3MmcPY:0 = @param:HZ0l3MmcPY:0 -> float_type, {32, 17}, {17, 1}: 0.00052328ms, 0.00461384% main:@835 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](HZ0l3MmcPY:0) -> float_type, {32, 4}, {17, 1}: 0.00068798ms, 0.00606602% o74vNAnXxy:0 = @param:o74vNAnXxy:0 -> float_type, {32, 17}, {17, 1}: 0.00046312ms, 0.0040834% main:@837 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](o74vNAnXxy:0) -> float_type, {32, 4}, {17, 1}: 0.0009798ms, 0.00863904% main:@838 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](nPUSZFm74Y:0) -> float_type, {32, 2}, {13, 1}: 0.00067278ms, 0.005932% main:@839 = slice[axes={0, 1},starts={0, 161},ends={32, 169}](uFSOc5Chp2:0) -> float_type, {32, 8}, {481, 1}: 0.00067296ms, 0.00593359% main:@840 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](xT9FhjM1Uu:0) -> float_type, {32, 4}, {17, 1}: 0.00067344ms, 0.00593782% qd1komWTn4:0 = @param:qd1komWTn4:0 -> float_type, {32, 25}, {25, 1}: 0.00044544ms, 0.00392751% main:@842 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](qd1komWTn4:0) -> float_type, {32, 8}, {25, 1}: 0.0006831ms, 0.00602299% main:@843 = slice[axes={0, 1},starts={0, 161},ends={32, 163}](nv2tzWRf1I:0) -> float_type, {32, 2}, {301, 1}: 0.00063054ms, 0.00555956% 1PJsBhDaCK:0 = @param:1PJsBhDaCK:0 -> float_type, {32, 25}, {25, 1}: 0.00054048ms, 0.00476549% main:@845 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](1PJsBhDaCK:0) -> float_type, {32, 8}, {25, 1}: 0.00069342ms, 0.00611398% 1WNvHElUG7:0 = @param:1WNvHElUG7:0 -> float_type, {32, 25}, {25, 1}: 0.00046206ms, 0.00407405% main:@847 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](1WNvHElUG7:0) -> float_type, {32, 8}, {25, 1}: 0.00067596ms, 0.00596004% 1u5yIsHTXE:0 = @param:1u5yIsHTXE:0 -> float_type, {32, 17}, {17, 1}: 0.00051576ms, 0.00454753% main:@849 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](1u5yIsHTXE:0) -> float_type, {32, 4}, {17, 1}: 0.00069072ms, 0.00609018% 2lVXMZgJfO:0 = @param:2lVXMZgJfO:0 -> float_type, {32, 25}, {25, 1}: 0.00051202ms, 0.00451455% main:@851 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](2lVXMZgJfO:0) -> float_type, {32, 8}, {25, 1}: 0.00065448ms, 0.00577064% 3kdi0vsSZG:0 = @param:3kdi0vsSZG:0 -> float_type, {32, 25}, {25, 1}: 0.0004976ms, 0.00438741% main:@853 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](3kdi0vsSZG:0) -> float_type, {32, 8}, {25, 1}: 0.0006537ms, 0.00576377% main:@854 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](Tg4u0GfnFb:0) -> float_type, {32, 8}, {25, 1}: 0.00065952ms, 0.00581508% 09zp4EDXTh:0 = @param:09zp4EDXTh:0 -> float_type, {32, 25}, {25, 1}: 0.00054882ms, 0.00483903% XxirNgB98u:0 = @param:XxirNgB98u:0 -> float_type, {32, 13}, {13, 1}: 0.00051372ms, 0.00452954% tWcngAq7hV:0 = @param:tWcngAq7hV:0 -> float_type, {32, 13}, {13, 1}: 0.00050226ms, 0.0044285% main:@858 = load[offset=3157120,end=3181952](main:@1) -> float_type, {32, 194}, {194, 1}: 0.00053246ms, 0.00469478% main:@859 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](YbIVz8l1un:0) -> float_type, {32, 8}, {45, 1}: 0.00074566ms, 0.00657459% main:@860 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](bJsIAzhYc3:0) -> float_type, {32, 8}, {37, 1}: 0.00063148ms, 0.00556785% main:@861 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](tWcngAq7hV:0) -> float_type, {32, 2}, {13, 1}: 0.00064096ms, 0.00565144% main:@862 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](XxirNgB98u:0) -> float_type, {32, 2}, {13, 1}: 0.00092762ms, 0.00817896% main:@863 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](09zp4EDXTh:0) -> float_type, {32, 8}, {25, 1}: 0.00064208ms, 0.00566131% main:@864 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=163,local=256,](main:@818,main:@862,main:@861,main:@820,main:@796,main:@800,main:@838,main:@832,main:@824,main:@811,main:@843,main:@835,main:@822,main:@826,main:@849,main:@807,main:@809,main:@813,main:@830,main:@837,main:@840,main:@833,main:@863,main:@845,main:@805,main:@798,main:@842,main:@804,main:@851,main:@828,main:@854,main:@847,main:@802,main:@860,main:@853,main:@859,main:@794,main:@839,main:@858) -> float_type, {32, 194}, {194, 1}: 0.0502191ms, 0.442789% main:@865 = hip::hip_copy_literal[id=main:@literal:143] -> float_type, {194, 192}, {192, 1}: 0.0008048ms, 0.00709604% main:@866 = load[offset=3132544,end=3157120](main:@1) -> float_type, {32, 192}, {192, 1}: 0.00046594ms, 0.00410826% main:@867 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@864,main:@865,main:@866) -> float_type, {32, 192}, {192, 1}: 0.024636ms, 0.217219% main:@868 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](FcmWLZhDgE:0) -> float_type, {32, 16}, {57, 1}: 0.0008175ms, 0.00720802% main:@869 = unsqueeze[axes={1},steps={}](main:@868) -> float_type, {32, 1, 16}, {57, 16, 1}: 0.00083678ms, 0.00737801% Dy3rwVmn1A:0 = @param:Dy3rwVmn1A:0 -> float_type, {32, 41}, {41, 1}: 0.00048944ms, 0.00431546% main:@871 = reshape_lazy[dims={-1, 12, 16}](main:@867) -> float_type, {32, 12, 16}, {192, 16, 1}: 0.00051074ms, 0.00450327% ahAlEtBvzo:0 = @param:ahAlEtBvzo:0 -> float_type, {32, 41}, {41, 1}: 0.00053588ms, 0.00472493% main:@873 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](ahAlEtBvzo:0) -> float_type, {32, 16}, {41, 1}: 0.0007751ms, 0.00683417% main:@874 = unsqueeze[axes={1},steps={}](main:@873) -> float_type, {32, 1, 16}, {41, 16, 1}: 0.00077566ms, 0.00683911% main:@875 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](Dy3rwVmn1A:0) -> float_type, {32, 16}, {41, 1}: 0.00061854ms, 0.00545376% main:@876 = unsqueeze[axes={1},steps={}](main:@875) -> float_type, {32, 1, 16}, {41, 16, 1}: 0.00058328ms, 0.00514286% zIoqYe9cRb:0 = @param:zIoqYe9cRb:0 -> float_type, {32, 45}, {45, 1}: 0.0005323ms, 0.00469337% mgnGNH4Mz3:0 = @param:mgnGNH4Mz3:0 -> float_type, {32, 49}, {49, 1}: 0.00054356ms, 0.00479265% main:@879 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](isCn4mEOSt:0) -> float_type, {32, 16}, {49, 1}: 0.0007748ms, 0.00683152% main:@880 = unsqueeze[axes={1},steps={}](main:@879) -> float_type, {32, 1, 16}, {49, 16, 1}: 0.0005684ms, 0.00501166% main:@881 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](zIoqYe9cRb:0) -> float_type, {32, 16}, {45, 1}: 0.00061424ms, 0.00541584% pYw7fMVoQg:0 = @param:pYw7fMVoQg:0 -> float_type, {32, 41}, {41, 1}: 0.00058216ms, 0.00513299% main:@883 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](pYw7fMVoQg:0) -> float_type, {32, 16}, {41, 1}: 0.00077198ms, 0.00680666% 6HSQqXdJzb:0 = @param:6HSQqXdJzb:0 -> float_type, {32, 45}, {45, 1}: 0.00051912ms, 0.00457716% main:@885 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](6HSQqXdJzb:0) -> float_type, {32, 16}, {45, 1}: 0.0006379ms, 0.00562446% main:@886 = unsqueeze[axes={1},steps={}](main:@885) -> float_type, {32, 1, 16}, {45, 16, 1}: 0.00060758ms, 0.00535712% main:@887 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](2CnSyUBzm0:0) -> float_type, {32, 16}, {65, 1}: 0.00076214ms, 0.0067199% main:@888 = unsqueeze[axes={1},steps={}](main:@887) -> float_type, {32, 1, 16}, {65, 16, 1}: 0.00071812ms, 0.00633177% X29iyIeu8L:0 = @param:X29iyIeu8L:0 -> float_type, {32, 45}, {45, 1}: 0.0005262ms, 0.00463958% dAFYcGrfEl:0 = @param:dAFYcGrfEl:0 -> float_type, {32, 45}, {45, 1}: 0.0004858ms, 0.00428337% 1aXQU27NrD:0 = @param:1aXQU27NrD:0 -> float_type, {32, 49}, {49, 1}: 0.00046716ms, 0.00411902% main:@892 = load[offset=3157120,end=3208320](main:@1) -> float_type, {32, 25, 16}, {400, 16, 1}: 0.00050182ms, 0.00442462% main:@893 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](1aXQU27NrD:0) -> float_type, {32, 16}, {49, 1}: 0.00075922ms, 0.00669415% main:@894 = unsqueeze[axes={1},steps={}](main:@893) -> float_type, {32, 1, 16}, {49, 16, 1}: 0.00057758ms, 0.00509261% main:@895 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](X29iyIeu8L:0) -> float_type, {32, 16}, {45, 1}: 0.00059908ms, 0.00528217% main:@896 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](ZcWERoz0h6:0) -> float_type, {32, 16}, {49, 1}: 0.00063918ms, 0.00563574% main:@897 = unsqueeze[axes={1},steps={}](main:@896) -> float_type, {32, 1, 16}, {49, 16, 1}: 0.00087114ms, 0.00768097% main:@898 = unsqueeze[axes={1},steps={}](main:@895) -> float_type, {32, 1, 16}, {45, 16, 1}: 0.00077676ms, 0.0068488% main:@899 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](mgnGNH4Mz3:0) -> float_type, {32, 16}, {49, 1}: 0.00060338ms, 0.00532009% main:@900 = unsqueeze[axes={1},steps={}](main:@899) -> float_type, {32, 1, 16}, {49, 16, 1}: 0.00056828ms, 0.00501061% main:@901 = unsqueeze[axes={1},steps={}](main:@881) -> float_type, {32, 1, 16}, {45, 16, 1}: 0.0005711ms, 0.00503547% main:@902 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](dAFYcGrfEl:0) -> float_type, {32, 16}, {45, 1}: 0.000698ms, 0.00615437% main:@903 = unsqueeze[axes={1},steps={}](main:@902) -> float_type, {32, 1, 16}, {45, 16, 1}: 0.00056852ms, 0.00501272% main:@904 = unsqueeze[axes={1},steps={}](main:@883) -> float_type, {32, 1, 16}, {41, 16, 1}: 0.00063488ms, 0.00559783% main:@905 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=914,local=256,](main:@888,main:@900,main:@894,main:@886,main:@901,main:@903,main:@898,main:@897,main:@874,main:@869,main:@904,main:@876,main:@880,main:@871,main:@892) -> float_type, {32, 25, 16}, {400, 16, 1}: 0.0290267ms, 0.255933% main:@906 = transpose[permutation={0, 2, 1}](main:@905) -> float_type, {32, 16, 25}, {400, 1, 16}: 0.00093584ms, 0.00825144% main:@907 = load[offset=3208320,end=3259520](main:@1) -> float_type, {32, 16, 25}, {400, 25, 1}: 0.0004977ms, 0.00438829% main:@908 = gpu::code_object[code_object=13528,symbol_name=contiguous_kernel,global=12800,local=256,](main:@906,main:@907) -> float_type, {32, 16, 25}, {400, 25, 1}: 0.0150796ms, 0.132959% main:@909 = hip::hip_copy_literal[id=main:@literal:115] -> float_type, {25, 5}, {5, 1}: 0.00079982ms, 0.00705213% main:@910 = reshape_lazy[dims={512, 25}](main:@908) -> float_type, {512, 25}, {25, 1}: 0.00050728ms, 0.00447276% main:@911 = load[offset=3132544,end=3142784](main:@1) -> float_type, {512, 5}, {5, 1}: 0.0004873ms, 0.00429659% main:@912 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@910,main:@909,main:@911) -> float_type, {512, 5}, {5, 1}: 0.0212481ms, 0.187347% main:@913 = hip::hip_copy_literal[id=main:@literal:75] -> float_type, {160, 160}, {160, 1}: 0.00081072ms, 0.00714824% EUciqFuptg:0 = @param:EUciqFuptg:0 -> float_type, {32, 17}, {17, 1}: 0.00056002ms, 0.00493778% main:@915 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](EUciqFuptg:0) -> float_type, {32, 4}, {17, 1}: 0.0006963ms, 0.00613938% nyLxrsTmOB:0 = @param:nyLxrsTmOB:0 -> float_type, {32, 17}, {17, 1}: 0.00052628ms, 0.00464029% main:@917 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](nyLxrsTmOB:0) -> float_type, {32, 8}, {17, 1}: 0.00078402ms, 0.00691282% KasFvpq4IG:0 = @param:KasFvpq4IG:0 -> float_type, {32, 5}, {5, 1}: 0.0005192ms, 0.00457786% main:@919 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](KasFvpq4IG:0) -> float_type, {32, 2}, {5, 1}: 0.00062278ms, 0.00549114% Hnjq487SIY:0 = @param:Hnjq487SIY:0 -> float_type, {32, 13}, {13, 1}: 0.0005272ms, 0.0046484% main:@921 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](Hnjq487SIY:0) -> float_type, {32, 2}, {13, 1}: 0.00077652ms, 0.00684669% IRem4MiFDP:0 = @param:IRem4MiFDP:0 -> float_type, {32, 5}, {5, 1}: 0.00047884ms, 0.004222% main:@923 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](IRem4MiFDP:0) -> float_type, {32, 2}, {5, 1}: 0.00062252ms, 0.00548885% JCVMzpWyri:0 = @param:JCVMzpWyri:0 -> float_type, {32, 17}, {17, 1}: 0.00055182ms, 0.00486548% main:@925 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](JCVMzpWyri:0) -> float_type, {32, 4}, {17, 1}: 0.0010877ms, 0.00959041% JSPsGophtx:0 = @param:JSPsGophtx:0 -> float_type, {32, 25}, {25, 1}: 0.00049656ms, 0.00437824% main:@927 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](JSPsGophtx:0) -> float_type, {32, 8}, {25, 1}: 0.00064272ms, 0.00566695% JaHTUxB6zL:0 = @param:JaHTUxB6zL:0 -> float_type, {32, 17}, {17, 1}: 0.0005949ms, 0.00524532% main:@929 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](JaHTUxB6zL:0) -> float_type, {32, 4}, {17, 1}: 0.00078998ms, 0.00696537% khJcIxW3zv:0 = @param:khJcIxW3zv:0 -> float_type, {32, 9}, {9, 1}: 0.0005292ms, 0.00466603% main:@931 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](khJcIxW3zv:0) -> float_type, {32, 4}, {9, 1}: 0.00061588ms, 0.0054303% LMobYAd5ru:0 = @param:LMobYAd5ru:0 -> float_type, {32, 17}, {17, 1}: 0.00054848ms, 0.00483603% main:@933 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](LMobYAd5ru:0) -> float_type, {32, 4}, {17, 1}: 0.0007192ms, 0.00634129% lQmygxs5aD:0 = @param:lQmygxs5aD:0 -> float_type, {32, 13}, {13, 1}: 0.00050606ms, 0.004462% main:@935 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](lQmygxs5aD:0) -> float_type, {32, 2}, {13, 1}: 0.00069468ms, 0.00612509% fMiQsYNnet:0 = @param:fMiQsYNnet:0 -> float_type, {32, 5}, {5, 1}: 0.00047288ms, 0.00416945% main:@937 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](fMiQsYNnet:0) -> float_type, {32, 2}, {5, 1}: 0.00067256ms, 0.00593006% XPcHGKbop2:0 = @param:XPcHGKbop2:0 -> float_type, {32, 13}, {13, 1}: 0.000498ms, 0.00439094% main:@939 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](XPcHGKbop2:0) -> float_type, {32, 2}, {13, 1}: 0.0006768ms, 0.00596744% TSqGwrHeo4:0 = @param:TSqGwrHeo4:0 -> float_type, {32, 5}, {5, 1}: 0.00050174ms, 0.00442391% main:@941 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](TSqGwrHeo4:0) -> float_type, {32, 2}, {5, 1}: 0.00098548ms, 0.00868912% Tej5M3O6As:0 = @param:Tej5M3O6As:0 -> float_type, {32, 17}, {17, 1}: 0.00048136ms, 0.00424422% main:@943 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](Tej5M3O6As:0) -> float_type, {32, 8}, {17, 1}: 0.0006811ms, 0.00600536% omr2g4XaH5:0 = @param:omr2g4XaH5:0 -> float_type, {32, 17}, {17, 1}: 0.00049706ms, 0.00438265% 4gFa5o2YsJ:0 = @param:4gFa5o2YsJ:0 -> float_type, {32, 9}, {9, 1}: 0.00048112ms, 0.0042421% aqMQBgb3hf:0 = @param:aqMQBgb3hf:0 -> float_type, {32, 17}, {17, 1}: 0.00046366ms, 0.00408816% iCbkMrmXBh:0 = @param:iCbkMrmXBh:0 -> float_type, {32, 9}, {9, 1}: 0.00051256ms, 0.00451932% cN2JSTnFOu:0 = @param:cN2JSTnFOu:0 -> float_type, {32, 9}, {9, 1}: 0.00055146ms, 0.0048623% 4M92ryD8ux:0 = @param:4M92ryD8ux:0 -> float_type, {32, 9}, {9, 1}: 0.0005021ms, 0.00442709% AB6lekmMxa:0 = @param:AB6lekmMxa:0 -> float_type, {32, 25}, {25, 1}: 0.00049206ms, 0.00433856% agDiJj8lLK:0 = @param:agDiJj8lLK:0 -> float_type, {32, 13}, {13, 1}: 0.00056072ms, 0.00494395% main:@952 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](4gFa5o2YsJ:0) -> float_type, {32, 4}, {9, 1}: 0.00077614ms, 0.00684334% 6XMrIsBjLY:0 = @param:6XMrIsBjLY:0 -> float_type, {32, 13}, {13, 1}: 0.0004605ms, 0.00406029% main:@954 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](6XMrIsBjLY:0) -> float_type, {32, 2}, {13, 1}: 0.00099248ms, 0.00875084% 72zWmh93Bo:0 = @param:72zWmh93Bo:0 -> float_type, {32, 17}, {17, 1}: 0.00051218ms, 0.00451596% main:@956 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](72zWmh93Bo:0) -> float_type, {32, 4}, {17, 1}: 0.0007097ms, 0.00625753% JjyOIxALzM:0 = @param:JjyOIxALzM:0 -> float_type, {32, 5}, {5, 1}: 0.0005633ms, 0.0049667% main:@958 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](JjyOIxALzM:0) -> float_type, {32, 2}, {5, 1}: 0.00067472ms, 0.0059491% 8vM5p4RzXd:0 = @param:8vM5p4RzXd:0 -> float_type, {32, 13}, {13, 1}: 0.00045518ms, 0.00401339% main:@960 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](8vM5p4RzXd:0) -> float_type, {32, 2}, {13, 1}: 0.0006769ms, 0.00596832% 9dgeU4bKka:0 = @param:9dgeU4bKka:0 -> float_type, {32, 17}, {17, 1}: 0.00051482ms, 0.00453924% main:@962 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](9dgeU4bKka:0) -> float_type, {32, 4}, {17, 1}: 0.00068088ms, 0.00600342% main:@963 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](AB6lekmMxa:0) -> float_type, {32, 8}, {25, 1}: 0.0006721ms, 0.005926% EZBIAqn6LS:0 = @param:EZBIAqn6LS:0 -> float_type, {32, 13}, {13, 1}: 0.00058048ms, 0.00511818% main:@965 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](EZBIAqn6LS:0) -> float_type, {32, 2}, {13, 1}: 0.00069984ms, 0.00617059% BYyjkPCX0m:0 = @param:BYyjkPCX0m:0 -> float_type, {32, 13}, {13, 1}: 0.00051844ms, 0.00457116% main:@967 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](BYyjkPCX0m:0) -> float_type, {32, 2}, {13, 1}: 0.00068466ms, 0.00603675% EKpP4xMXu8:0 = @param:EKpP4xMXu8:0 -> float_type, {32, 17}, {17, 1}: 0.00050224ms, 0.00442832% main:@969 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](EKpP4xMXu8:0) -> float_type, {32, 4}, {17, 1}: 0.0006611ms, 0.00582901% r1WGP7Xe9H:0 = @param:r1WGP7Xe9H:0 -> float_type, {32, 13}, {13, 1}: 0.00046484ms, 0.00409856% qzd23aNGIT:0 = @param:qzd23aNGIT:0 -> float_type, {32, 17}, {17, 1}: 0.00048914ms, 0.00431282% Vbv2Om3ptd:0 = @param:Vbv2Om3ptd:0 -> float_type, {32, 17}, {17, 1}: 0.0005703ms, 0.00502842% fZhaL2tCm6:0 = @param:fZhaL2tCm6:0 -> float_type, {32, 17}, {17, 1}: 0.00053992ms, 0.00476055% d8RoEzG52p:0 = @param:d8RoEzG52p:0 -> float_type, {32, 17}, {17, 1}: 0.00050546ms, 0.00445671% main:@975 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](iCbkMrmXBh:0) -> float_type, {32, 4}, {9, 1}: 0.00074656ms, 0.00658253% main:@976 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](d8RoEzG52p:0) -> float_type, {32, 8}, {17, 1}: 0.00096298ms, 0.00849073% eQquakfWBY:0 = @param:eQquakfWBY:0 -> float_type, {32, 17}, {17, 1}: 0.00056216ms, 0.00495665% main:@978 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](eQquakfWBY:0) -> float_type, {32, 4}, {17, 1}: 0.00069348ms, 0.00611451% main:@979 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](qzd23aNGIT:0) -> float_type, {32, 4}, {17, 1}: 0.0006686ms, 0.00589514% main:@980 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](fZhaL2tCm6:0) -> float_type, {32, 8}, {17, 1}: 0.00065088ms, 0.0057389% main:@981 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](4M92ryD8ux:0) -> float_type, {32, 4}, {9, 1}: 0.00065008ms, 0.00573185% bTeX85HPmJ:0 = @param:bTeX85HPmJ:0 -> float_type, {32, 17}, {17, 1}: 0.00057466ms, 0.00506686% main:@983 = load[offset=3163264,end=3183744](main:@1) -> float_type, {32, 160}, {160, 1}: 0.00050238ms, 0.00442956% main:@984 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](Vbv2Om3ptd:0) -> float_type, {32, 8}, {17, 1}: 0.00069214ms, 0.0061027% main:@985 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](r1WGP7Xe9H:0) -> float_type, {32, 2}, {13, 1}: 0.00062264ms, 0.00548991% main:@986 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](bTeX85HPmJ:0) -> float_type, {32, 4}, {17, 1}: 0.00063238ms, 0.00557579% main:@987 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](agDiJj8lLK:0) -> float_type, {32, 2}, {13, 1}: 0.00063622ms, 0.00560964% main:@988 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](cN2JSTnFOu:0) -> float_type, {32, 4}, {9, 1}: 0.00064234ms, 0.0056636% main:@989 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](omr2g4XaH5:0) -> float_type, {32, 8}, {17, 1}: 0.00065044ms, 0.00573502% main:@990 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](aqMQBgb3hf:0) -> float_type, {32, 8}, {17, 1}: 0.0006532ms, 0.00575936% main:@991 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=134,local=256,](main:@958,main:@965,main:@923,main:@935,main:@941,main:@985,main:@987,main:@939,main:@960,main:@937,main:@967,main:@921,main:@919,main:@954,main:@988,main:@962,main:@981,main:@933,main:@931,main:@915,main:@986,main:@969,main:@978,main:@925,main:@956,main:@979,main:@975,main:@929,main:@952,main:@927,main:@980,main:@976,main:@943,main:@989,main:@963,main:@990,main:@984,main:@917,main:@983) -> float_type, {32, 160}, {160, 1}: 0.0468438ms, 0.413029% main:@992 = load[offset=3142784,end=3163264](main:@1) -> float_type, {32, 160}, {160, 1}: 0.00069018ms, 0.00608542% main:@993 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@991,main:@913,main:@992) -> float_type, {32, 160}, {160, 1}: 0.0211309ms, 0.186314% HRGzlnEkXQ:0 = @param:HRGzlnEkXQ:0 -> float_type, {32, 41}, {41, 1}: 0.0006971ms, 0.00614643% main:@995 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](HRGzlnEkXQ:0) -> float_type, {32, 16}, {41, 1}: 0.00068354ms, 0.00602687% main:@996 = unsqueeze[axes={1},steps={}](main:@995) -> float_type, {32, 1, 16}, {41, 16, 1}: 0.00076412ms, 0.00673736% HzcyUlm8Yj:0 = @param:HzcyUlm8Yj:0 -> float_type, {32, 45}, {45, 1}: 0.00048672ms, 0.00429148% main:@998 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](HzcyUlm8Yj:0) -> float_type, {32, 16}, {45, 1}: 0.00077056ms, 0.00679414% main:@999 = unsqueeze[axes={1},steps={}](main:@998) -> float_type, {32, 1, 16}, {45, 16, 1}: 0.00057908ms, 0.00510583% main:@1000 = slice[axes={0, 1},starts={0, 289},ends={32, 305}](lyiDxfzZg3:0) -> float_type, {32, 16}, {701, 1}: 0.00061412ms, 0.00541478% main:@1001 = unsqueeze[axes={1},steps={}](main:@1000) -> float_type, {32, 1, 16}, {701, 16, 1}: 0.00059294ms, 0.00522804% main:@1002 = reshape_lazy[dims={-1, 10, 16}](main:@993) -> float_type, {32, 10, 16}, {160, 16, 1}: 0.00052092ms, 0.00459303% CHAYRkBvIo:0 = @param:CHAYRkBvIo:0 -> float_type, {32, 41}, {41, 1}: 0.00053144ms, 0.00468578% 3Laj620MSn:0 = @param:3Laj620MSn:0 -> float_type, {32, 45}, {45, 1}: 0.0005346ms, 0.00471365% main:@1005 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](CHAYRkBvIo:0) -> float_type, {32, 16}, {41, 1}: 0.00075316ms, 0.00664072% main:@1006 = unsqueeze[axes={1},steps={}](main:@1005) -> float_type, {32, 1, 16}, {41, 16, 1}: 0.00071876ms, 0.00633741% jznORDJr0N:0 = @param:jznORDJr0N:0 -> float_type, {32, 45}, {45, 1}: 0.00054864ms, 0.00483744% main:@1008 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](jznORDJr0N:0) -> float_type, {32, 16}, {45, 1}: 0.0006328ms, 0.00557949% main:@1009 = unsqueeze[axes={1},steps={}](main:@1008) -> float_type, {32, 1, 16}, {45, 16, 1}: 0.00057132ms, 0.00503741% eQI8ikEMgP:0 = @param:eQI8ikEMgP:0 -> float_type, {32, 41}, {41, 1}: 0.00047982ms, 0.00423064% main:@1011 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](eQI8ikEMgP:0) -> float_type, {32, 16}, {41, 1}: 0.00076912ms, 0.00678144% 2RDXZrx9GA:0 = @param:2RDXZrx9GA:0 -> float_type, {32, 45}, {45, 1}: 0.00048204ms, 0.00425022% main:@1013 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](2RDXZrx9GA:0) -> float_type, {32, 16}, {45, 1}: 0.00064692ms, 0.00570399% duyRoIEkH8:0 = @param:duyRoIEkH8:0 -> float_type, {32, 45}, {45, 1}: 0.00053344ms, 0.00470342% main:@1015 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](duyRoIEkH8:0) -> float_type, {32, 16}, {45, 1}: 0.00074638ms, 0.00658094% main:@1016 = unsqueeze[axes={1},steps={}](main:@1015) -> float_type, {32, 1, 16}, {45, 16, 1}: 0.00072388ms, 0.00638255% main:@1017 = unsqueeze[axes={1},steps={}](main:@1013) -> float_type, {32, 1, 16}, {45, 16, 1}: 0.00056086ms, 0.00494518% 2srtUfGvja:0 = @param:2srtUfGvja:0 -> float_type, {32, 41}, {41, 1}: 0.0005055ms, 0.00445707% main:@1019 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](2srtUfGvja:0) -> float_type, {32, 16}, {41, 1}: 0.0007781ms, 0.00686062% main:@1020 = unsqueeze[axes={1},steps={}](main:@1019) -> float_type, {32, 1, 16}, {41, 16, 1}: 0.00059136ms, 0.00521411% main:@1021 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](3Laj620MSn:0) -> float_type, {32, 16}, {45, 1}: 0.00066714ms, 0.00588227% main:@1022 = unsqueeze[axes={1},steps={}](main:@1021) -> float_type, {32, 1, 16}, {45, 16, 1}: 0.00058988ms, 0.00520106% VUqfdN5Fzv:0 = @param:VUqfdN5Fzv:0 -> float_type, {32, 45}, {45, 1}: 0.00050084ms, 0.00441598% main:@1024 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](VUqfdN5Fzv:0) -> float_type, {32, 16}, {45, 1}: 0.00078182ms, 0.00689342% main:@1025 = unsqueeze[axes={1},steps={}](main:@1011) -> float_type, {32, 1, 16}, {41, 16, 1}: 0.0005644ms, 0.0049764% main:@1026 = unsqueeze[axes={1},steps={}](main:@1024) -> float_type, {32, 1, 16}, {45, 16, 1}: 0.00057198ms, 0.00504323% main:@1027 = load[offset=3163264,end=3206272](main:@1) -> float_type, {32, 21, 16}, {336, 16, 1}: 0.0004831ms, 0.00425956% main:@1028 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=896,local=256,](main:@1009,main:@996,main:@1001,main:@1017,main:@1026,main:@1006,main:@999,main:@1020,main:@1022,main:@1016,main:@1025,main:@1002,main:@1027) -> float_type, {32, 21, 16}, {336, 16, 1}: 0.026842ms, 0.23667% main:@1029 = reshape_lazy[dims={32, 16, 5}](main:@912) -> float_type, {32, 16, 5}, {80, 5, 1}: 0.0007081ms, 0.00624342% main:@1030 = load[offset=514048,end=527488](main:@1) -> float_type, {32, 21, 5}, {105, 5, 1}: 0.00050818ms, 0.0044807% main:@1031 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1028,main:@1029,main:@1030) -> float_type, {32, 21, 5}, {105, 5, 1}: 0.0191803ms, 0.169115% XRvxjULC7M:0 = @param:XRvxjULC7M:0 -> float_type, {32, 60, 5}, {300, 5, 1}: 0.00072002ms, 0.00634852% main:@1033 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 60, 5}](XRvxjULC7M:0) -> float_type, {32, 60, 4}, {300, 5, 1}: 0.00067186ms, 0.00592389% main:@1034 = load[offset=8665344,end=9187584](main:@1) -> float_type, {32, 60, 68}, {4080, 68, 1}: 0.00052408ms, 0.00462089% main:@1035 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=32640,local=256,](main:@195,main:@91,main:@1033,main:@40,main:@1034) -> float_type, {32, 60, 68}, {4080, 68, 1}: 0.0207569ms, 0.183016% main:@1036 = hip::hip_copy_literal[id=main:@literal:152] -> float_type, {68, 64}, {64, 1}: 0.0008226ms, 0.00725298% main:@1037 = reshape_lazy[dims={1920, 68}](main:@1035) -> float_type, {1920, 68}, {68, 1}: 0.00050674ms, 0.004468% main:@1038 = load[offset=3833600,end=4325120](main:@1) -> float_type, {1920, 64}, {64, 1}: 0.00046698ms, 0.00411743% main:@1039 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1037,main:@1036,main:@1038) -> float_type, {1920, 64}, {64, 1}: 0.023572ms, 0.207838% main:@1040 = slice[axes={0, 1},starts={0, 45},ends={32, 53}](Qp3K1PHvte:0) -> float_type, {32, 8}, {69, 1}: 0.00081566ms, 0.00719179% main:@1041 = slice[axes={0, 1},starts={0, 189},ends={32, 197}](nv2tzWRf1I:0) -> float_type, {32, 8}, {301, 1}: 0.00064156ms, 0.00565673% main:@1042 = slice[axes={0, 1},starts={0, 209},ends={32, 217}](uFSOc5Chp2:0) -> float_type, {32, 8}, {481, 1}: 0.00077638ms, 0.00684545% main:@1043 = slice[axes={0, 1},starts={0, 13},ends={32, 21}](9YqhflbMKe:0) -> float_type, {32, 8}, {29, 1}: 0.00065594ms, 0.00578352% main:@1044 = slice[axes={0, 1},starts={0, 49},ends={32, 57}](ifP6Ceko5T:0) -> float_type, {32, 8}, {73, 1}: 0.00062742ms, 0.00553205% main:@1045 = slice[axes={0, 1},starts={0, 45},ends={32, 53}](2CnSyUBzm0:0) -> float_type, {32, 8}, {65, 1}: 0.0006007ms, 0.00529646% main:@1046 = load[offset=773248,end=779392](main:@1) -> float_type, {32, 48}, {48, 1}: 0.00050016ms, 0.00440998% main:@1047 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=256,local=256,](main:@1041,main:@1045,main:@1042,main:@1044,main:@1040,main:@1043,main:@1046) -> float_type, {32, 48}, {48, 1}: 0.0162361ms, 0.143156% main:@1048 = hip::hip_copy_literal[id=main:@literal:160] -> float_type, {48, 64}, {64, 1}: 0.00088812ms, 0.00783068% main:@1049 = load[offset=3132544,end=3140736](main:@1) -> float_type, {32, 64}, {64, 1}: 0.00051002ms, 0.00449692% main:@1050 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1047,main:@1048,main:@1049) -> float_type, {32, 64}, {64, 1}: 0.0206234ms, 0.181839% main:@1051 = reshape_lazy[dims={32, 60, 64}](main:@1039) -> float_type, {32, 60, 64}, {3840, 64, 1}: 0.00069332ms, 0.0061131% main:@1052 = transpose[permutation={0, 2, 1}](main:@1051) -> float_type, {32, 64, 60}, {3840, 1, 64}: 0.00057972ms, 0.00511147% main:@1053 = reshape_lazy[dims={32, 1, 64}](main:@1050) -> float_type, {32, 1, 64}, {64, 64, 1}: 0.00050662ms, 0.00446694% main:@1054 = load[offset=773248,end=780928](main:@1) -> float_type, {32, 1, 60}, {60, 60, 1}: 0.00048512ms, 0.00427737% main:@1055 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1053,main:@1052,main:@1054) -> float_type, {32, 1, 60}, {60, 60, 1}: 0.0229722ms, 0.202549% PnxiROp9zw:0 = @param:PnxiROp9zw:0 -> float_type, {32, 60, 5}, {300, 5, 1}: 0.00069216ms, 0.00610287% main:@1057 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 60, 5}](PnxiROp9zw:0) -> float_type, {32, 60, 4}, {300, 5, 1}: 0.00071882ms, 0.00633794% 4luvCw32jf:0 = @param:4luvCw32jf:0 -> float_type, {32, 60, 5}, {300, 5, 1}: 0.00052098ms, 0.00459356% main:@1059 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 60, 5}](4luvCw32jf:0) -> float_type, {32, 60, 4}, {300, 5, 1}: 0.0007805ms, 0.00688178% F5awZuD1c2:0 = @param:F5awZuD1c2:0 -> float_type, {32, 60, 9}, {540, 9, 1}: 0.00050202ms, 0.00442638% main:@1061 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 60, 9}](F5awZuD1c2:0) -> float_type, {32, 60, 8}, {540, 9, 1}: 0.00095194ms, 0.00839339% main:@1062 = load[offset=8665344,end=9310464](main:@1) -> float_type, {32, 60, 84}, {5040, 84, 1}: 0.0004849ms, 0.00427543% main:@1063 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=23040,local=256,](main:@195,main:@91,main:@1033,main:@40,main:@1057,main:@1061,main:@1059,main:@1062) -> float_type, {32, 60, 84}, {5040, 84, 1}: 0.0217959ms, 0.192178% main:@1064 = hip::hip_copy_literal[id=main:@literal:134] -> float_type, {84, 64}, {64, 1}: 0.00075618ms, 0.00666735% main:@1065 = load[offset=3833600,end=4325120](main:@1) -> float_type, {1920, 64}, {64, 1}: 0.00050556ms, 0.0044576% main:@1066 = reshape_lazy[dims={1920, 84}](main:@1063) -> float_type, {1920, 84}, {84, 1}: 0.0005153ms, 0.00454347% main:@1067 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1066,main:@1064,main:@1065) -> float_type, {1920, 64}, {64, 1}: 0.0215552ms, 0.190055% main:@1068 = load[offset=3132544,end=3140224](main:@1) -> float_type, {32, 1, 60}, {60, 60, 1}: 0.00066132ms, 0.00583095% mt0cODHoal:0 = @param:mt0cODHoal:0 -> int32_type, {32, 60}, {60, 1}: 0.00052018ms, 0.0045865% main:@1070 = unsqueeze[axes={1},steps={}](mt0cODHoal:0) -> int32_type, {32, 1, 60}, {60, 60, 1}: 0.00064444ms, 0.00568212% main:@1071 = gpu::code_object[code_object=13544,symbol_name=convert_add_add_sub_kernel,global=960,local=256,](main:@1070,main:@1055,main:@1068) -> float_type, {32, 1, 60}, {60, 60, 1}: 0.0134834ms, 0.118885% main:@1072 = load[offset=3140224,end=3147904](main:@1) -> float_type, {32, 1, 60}, {60, 60, 1}: 0.0006937ms, 0.00611645% main:@1073 = gpu::code_object[code_object=14552,symbol_name=softmax_kernel,global=2048,local=64,](main:@1071,main:@1072) -> float_type, {32, 1, 60}, {60, 60, 1}: 0.0148119ms, 0.130599% main:@1074 = load[offset=773248,end=781440](main:@1) -> float_type, {32, 1, 64}, {64, 64, 1}: 0.00063488ms, 0.00559783% main:@1075 = reshape_lazy[dims={32, 60, 64}](main:@1067) -> float_type, {32, 60, 64}, {3840, 64, 1}: 0.00053332ms, 0.00470236% main:@1076 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1073,main:@1075,main:@1074) -> float_type, {32, 1, 64}, {64, 64, 1}: 0.022701ms, 0.200158% main:@1077 = slice[axes={1},starts={0},ends={256}](main:@758) -> float_type, {480, 256}, {768, 1}: 0.00081576ms, 0.00719267% main:@1078 = reshape_lazy[dims={32, 15, 256}](main:@1077) -> float_type, {32, 15, 256}, {11520, 768, 1}: 0.00066596ms, 0.00587187% main:@1079 = slice[axes={2},starts={0},ends={64}](main:@1078) -> float_type, {32, 15, 64}, {11520, 768, 1}: 0.00117128ms, 0.0103273% main:@1080 = unsqueeze[axes={0},steps={}](main:@1079) -> float_type, {1, 32, 15, 64}, {368640, 11520, 768, 1}: 0.00069326ms, 0.00611257% main:@1081 = slice[axes={2},starts={64},ends={128}](main:@1078) -> float_type, {32, 15, 64}, {11520, 768, 1}: 0.00074588ms, 0.00657653% main:@1082 = unsqueeze[axes={0},steps={}](main:@1081) -> float_type, {1, 32, 15, 64}, {368640, 11520, 768, 1}: 0.0006201ms, 0.00546751% main:@1083 = slice[axes={2},starts={128},ends={192}](main:@1078) -> float_type, {32, 15, 64}, {11520, 768, 1}: 0.00066146ms, 0.00583219% main:@1084 = unsqueeze[axes={0},steps={}](main:@1083) -> float_type, {1, 32, 15, 64}, {368640, 11520, 768, 1}: 0.00097648ms, 0.00860976% main:@1085 = slice[axes={2},starts={192},ends={256}](main:@1078) -> float_type, {32, 15, 64}, {11520, 768, 1}: 0.00065904ms, 0.00581085% main:@1086 = unsqueeze[axes={0},steps={}](main:@1085) -> float_type, {1, 32, 15, 64}, {368640, 11520, 768, 1}: 0.00062564ms, 0.00551636% main:@1087 = load[offset=3833600,end=4325120](main:@1) -> float_type, {4, 32, 15, 64}, {30720, 960, 64, 1}: 0.0005095ms, 0.00449233% main:@1088 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=15360,local=256,](main:@1080,main:@1082,main:@1084,main:@1086,main:@1087) -> float_type, {4, 32, 15, 64}, {30720, 960, 64, 1}: 0.0206213ms, 0.181821% main:@1089 = slice[axes={1},starts={512},ends={768}](main:@758) -> float_type, {480, 256}, {768, 1}: 0.00075904ms, 0.00669257% main:@1090 = reshape_lazy[dims={32, 15, 256}](main:@1089) -> float_type, {32, 15, 256}, {11520, 768, 1}: 0.00056474ms, 0.00497939% main:@1091 = load[offset=8665344,end=9156864](main:@1) -> float_type, {4, 32, 15, 64}, {30720, 960, 64, 1}: 0.00050382ms, 0.00444225% main:@1092 = slice[axes={2},starts={0},ends={64}](main:@1090) -> float_type, {32, 15, 64}, {11520, 768, 1}: 0.00109596ms, 0.00966324% main:@1093 = unsqueeze[axes={0},steps={}](main:@1092) -> float_type, {1, 32, 15, 64}, {368640, 11520, 768, 1}: 0.00060564ms, 0.00534002% main:@1094 = slice[axes={2},starts={64},ends={128}](main:@1090) -> float_type, {32, 15, 64}, {11520, 768, 1}: 0.0006865ms, 0.00605297% main:@1095 = unsqueeze[axes={0},steps={}](main:@1094) -> float_type, {1, 32, 15, 64}, {368640, 11520, 768, 1}: 0.0008883ms, 0.00783227% main:@1096 = slice[axes={2},starts={128},ends={192}](main:@1090) -> float_type, {32, 15, 64}, {11520, 768, 1}: 0.0006235ms, 0.00549749% main:@1097 = unsqueeze[axes={0},steps={}](main:@1096) -> float_type, {1, 32, 15, 64}, {368640, 11520, 768, 1}: 0.00091986ms, 0.00811054% main:@1098 = slice[axes={2},starts={192},ends={256}](main:@1090) -> float_type, {32, 15, 64}, {11520, 768, 1}: 0.00063982ms, 0.00564139% main:@1099 = unsqueeze[axes={0},steps={}](main:@1098) -> float_type, {1, 32, 15, 64}, {368640, 11520, 768, 1}: 0.0006981ms, 0.00615525% main:@1100 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=15360,local=256,](main:@1093,main:@1095,main:@1097,main:@1099,main:@1091) -> float_type, {4, 32, 15, 64}, {30720, 960, 64, 1}: 0.0208808ms, 0.184109% main:@1101 = load[offset=3132544,end=3247744](main:@1) -> float_type, {4, 32, 15, 15}, {7200, 225, 15, 1}: 0.00064774ms, 0.00571122% main:@1102 = transpose[permutation={0, 1, 3, 2}](main:@1100) -> float_type, {4, 32, 64, 15}, {30720, 960, 1, 64}: 0.00054338ms, 0.00479106% main:@1103 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1088,main:@1102,main:@1101) -> float_type, {4, 32, 15, 15}, {7200, 225, 15, 1}: 0.0219152ms, 0.193229% main:@1104 = load[offset=3247744,end=3362944](main:@1) -> float_type, {4, 32, 15, 15}, {7200, 225, 15, 1}: 0.00066678ms, 0.0058791% main:@1105 = gpu::code_object[code_object=13512,symbol_name=pow_kernel,global=14400,local=256,](main:@1103,main:@1104) -> float_type, {4, 32, 15, 15}, {7200, 225, 15, 1}: 0.0161282ms, 0.142204% main:@1106 = load[offset=3132544,end=3247744](main:@1) -> float_type, {4, 32, 15, 15}, {7200, 225, 15, 1}: 0.00065914ms, 0.00581173% main:@1107 = gpu::code_object[code_object=14576,symbol_name=softmax_kernel,global=122880,local=64,](main:@1105,main:@1106) -> float_type, {4, 32, 15, 15}, {7200, 225, 15, 1}: 0.0192295ms, 0.16955% main:@1108 = load[offset=8665344,end=9156864](main:@1) -> float_type, {4, 32, 15, 64}, {30720, 960, 64, 1}: 0.0006376ms, 0.00562181% main:@1109 = slice[axes={1},starts={256},ends={512}](main:@758) -> float_type, {480, 256}, {768, 1}: 0.00065126ms, 0.00574225% main:@1110 = reshape_lazy[dims={32, 15, 256}](main:@1109) -> float_type, {32, 15, 256}, {11520, 768, 1}: 0.0009488ms, 0.00836571% main:@1111 = slice[axes={2},starts={0},ends={64}](main:@1110) -> float_type, {32, 15, 64}, {11520, 768, 1}: 0.0006662ms, 0.00587398% main:@1112 = unsqueeze[axes={0},steps={}](main:@1111) -> float_type, {1, 32, 15, 64}, {368640, 11520, 768, 1}: 0.00093972ms, 0.00828565% main:@1113 = slice[axes={2},starts={64},ends={128}](main:@1110) -> float_type, {32, 15, 64}, {11520, 768, 1}: 0.00063342ms, 0.00558496% main:@1114 = unsqueeze[axes={0},steps={}](main:@1113) -> float_type, {1, 32, 15, 64}, {368640, 11520, 768, 1}: 0.00061896ms, 0.00545746% main:@1115 = slice[axes={2},starts={128},ends={192}](main:@1110) -> float_type, {32, 15, 64}, {11520, 768, 1}: 0.00098698ms, 0.00870234% main:@1116 = unsqueeze[axes={0},steps={}](main:@1115) -> float_type, {1, 32, 15, 64}, {368640, 11520, 768, 1}: 0.00057954ms, 0.00510989% main:@1117 = slice[axes={2},starts={192},ends={256}](main:@1110) -> float_type, {32, 15, 64}, {11520, 768, 1}: 0.0007087ms, 0.00624871% main:@1118 = unsqueeze[axes={0},steps={}](main:@1117) -> float_type, {1, 32, 15, 64}, {368640, 11520, 768, 1}: 0.00060422ms, 0.00532749% main:@1119 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=15360,local=256,](main:@1112,main:@1114,main:@1116,main:@1118,main:@1108) -> float_type, {4, 32, 15, 64}, {30720, 960, 64, 1}: 0.0209362ms, 0.184597% main:@1120 = load[offset=3833600,end=4325120](main:@1) -> float_type, {4, 32, 15, 64}, {30720, 960, 64, 1}: 0.00062732ms, 0.00553117% main:@1121 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1107,main:@1119,main:@1120) -> float_type, {4, 32, 15, 64}, {30720, 960, 64, 1}: 0.0200203ms, 0.176522% main:@1122 = slice[axes={1},starts={32},ends={96}](main:@643) -> float_type, {32, 64}, {128, 1}: 0.00078106ms, 0.00688672% main:@1123 = load[offset=3138944,end=3147136](main:@1) -> float_type, {32, 64}, {64, 1}: 0.00049194ms, 0.00433751% main:@1124 = gpu::code_object[code_object=13528,symbol_name=contiguous_kernel,global=1024,local=256,](main:@1122,main:@1123) -> float_type, {32, 64}, {64, 1}: 0.0130904ms, 0.11542% main:@1125 = slice[axes={1},starts={64},ends={128}](main:@539) -> float_type, {1600, 64}, {128, 1}: 0.0007526ms, 0.00663578% main:@1126 = reshape_lazy[dims={1, 32, 50, 64}](main:@1125) -> float_type, {1, 32, 50, 64}, {204800, 6400, 128, 1}: 0.00058742ms, 0.00517937% main:@1127 = transpose[permutation={0, 1, 3, 2}](main:@1126) -> float_type, {1, 32, 64, 50}, {204800, 6400, 1, 128}: 0.00060566ms, 0.00534019% main:@1128 = load[offset=3132544,end=3138944](main:@1) -> float_type, {1, 32, 1, 50}, {1600, 50, 50, 1}: 0.00052002ms, 0.00458509% main:@1129 = reshape_lazy[dims={1, 32, 1, 64}](main:@1124) -> float_type, {1, 32, 1, 64}, {2048, 64, 64, 1}: 0.00050304ms, 0.00443538% main:@1130 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1129,main:@1127,main:@1128) -> float_type, {1, 32, 1, 50}, {1600, 50, 50, 1}: 0.0239485ms, 0.211157% ObZQkfyaKG:0 = @param:ObZQkfyaKG:0 -> int32_type, {32, 50}, {50, 1}: 0.0007175ms, 0.0063263% main:@1132 = unsqueeze[axes={1},steps={}](ObZQkfyaKG:0) -> int32_type, {32, 1, 50}, {50, 50, 1}: 0.00061578ms, 0.00542942% main:@1133 = load[offset=3138944,end=3145344](main:@1) -> float_type, {32, 1, 50}, {50, 50, 1}: 0.00051734ms, 0.00456146% main:@1134 = gpu::code_object[code_object=13528,symbol_name=convert_add_kernel,global=800,local=256,](main:@1132,main:@1133) -> float_type, {32, 1, 50}, {50, 50, 1}: 0.0132375ms, 0.116717% main:@1135 = multibroadcast[out_lens={1, 32, 1, 50}](main:@1134) -> float_type, {1, 32, 1, 50}, {0, 50, 50, 1}: 0.00077498ms, 0.00683311% main:@1136 = load[offset=3147136,end=3153536](main:@1) -> float_type, {1, 32, 1, 50}, {1600, 50, 50, 1}: 0.00054176ms, 0.00477678% main:@1137 = gpu::code_object[code_object=13520,symbol_name=pow_pow_kernel,global=800,local=256,](main:@1130,main:@1135,main:@1136) -> float_type, {1, 32, 1, 50}, {1600, 50, 50, 1}: 0.0177403ms, 0.156419% main:@1138 = load[offset=3140736,end=3147136](main:@1) -> float_type, {1, 32, 1, 50}, {1600, 50, 50, 1}: 0.00064142ms, 0.00565549% main:@1139 = gpu::code_object[code_object=14584,symbol_name=softmax_kernel,global=2048,local=64,](main:@1137,main:@1138) -> float_type, {1, 32, 1, 50}, {1600, 50, 50, 1}: 0.0147036ms, 0.129644% main:@1140 = slice[axes={1},starts={0},ends={64}](main:@539) -> float_type, {1600, 64}, {128, 1}: 0.00075724ms, 0.00667669% main:@1141 = load[offset=3132544,end=3140736](main:@1) -> float_type, {1, 32, 1, 64}, {2048, 64, 64, 1}: 0.00049134ms, 0.00433222% main:@1142 = reshape_lazy[dims={1, 32, 50, 64}](main:@1140) -> float_type, {1, 32, 50, 64}, {204800, 6400, 128, 1}: 0.00059826ms, 0.00527494% main:@1143 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1139,main:@1142,main:@1141) -> float_type, {1, 32, 1, 64}, {2048, 64, 64, 1}: 0.0221198ms, 0.195033% main:@1144 = hip::hip_copy_literal[id=main:@literal:116] -> float_type, {128, 3429}, {3429, 1}: 0.00080316ms, 0.00708158% main:@1145 = slice[axes={1},starts={128},ends={256}](main:@765) -> float_type, {32, 128}, {896, 1}: 0.0006058ms, 0.00534143% main:@1146 = load[offset=3140736,end=3579648](main:@1) -> float_type, {32, 3429}, {3429, 1}: 0.00049722ms, 0.00438406% main:@1147 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1145,main:@1144,main:@1146) -> float_type, {32, 3429}, {3429, 1}: 0.024227ms, 0.213613% main:@1148 = hip::hip_copy_literal[id=main:@literal:49] -> float_type, {256, 512}, {512, 1}: 0.00075134ms, 0.00662467% main:@1149 = slice[axes={0, 1, 2},starts={0, 9, 0},ends={32, 10, 256}](main:@778) -> float_type, {32, 1, 256}, {4608, 256, 1}: 0.00064482ms, 0.00568547% main:@1150 = squeeze[axes={1}](main:@1149) -> float_type, {32, 256}, {4608, 1}: 0.00079244ms, 0.00698706% main:@1151 = load[offset=4423424,end=4488960](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00048568ms, 0.00428231% main:@1152 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1150,main:@1148,main:@1151) -> float_type, {32, 512}, {512, 1}: 0.0255117ms, 0.22494% main:@1153 = load[offset=4357888,end=4423424](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00062412ms, 0.00550296% main:@1154 = gpu::code_object[code_object=13512,symbol_name=relu_kernel,global=8192,local=256,](main:@1152,main:@1153) -> float_type, {32, 512}, {512, 1}: 0.014338ms, 0.126421% main:@1155 = load[offset=4325120,end=4357888](main:@1) -> float_type, {32, 256}, {256, 1}: 0.0006266ms, 0.00552482% main:@1156 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1154,main:@768,main:@1155) -> float_type, {32, 256}, {256, 1}: 0.0309635ms, 0.27301% main:@1157 = slice[axes={0, 1, 2},starts={0, 16, 0},ends={32, 17, 256}](main:@778) -> float_type, {32, 1, 256}, {4608, 256, 1}: 0.00077196ms, 0.00680648% main:@1158 = squeeze[axes={1}](main:@1157) -> float_type, {32, 256}, {4608, 1}: 0.00061478ms, 0.0054206% main:@1159 = load[offset=4456192,end=4521728](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00048974ms, 0.00431811% main:@1160 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1158,main:@762,main:@1159) -> float_type, {32, 512}, {512, 1}: 0.0233809ms, 0.206153% main:@1161 = load[offset=4390656,end=4456192](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00061334ms, 0.00540791% main:@1162 = gpu::code_object[code_object=13512,symbol_name=relu_kernel,global=8192,local=256,](main:@1160,main:@1161) -> float_type, {32, 512}, {512, 1}: 0.014514ms, 0.127972% main:@1163 = hip::hip_copy_literal[id=main:@literal:84] -> float_type, {512, 256}, {256, 1}: 0.00083022ms, 0.00732017% main:@1164 = load[offset=4357888,end=4390656](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00049596ms, 0.00437295% main:@1165 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1162,main:@1163,main:@1164) -> float_type, {32, 256}, {256, 1}: 0.0301371ms, 0.265723% main:@1166 = hip::hip_copy_literal[id=main:@literal:142] -> float_type, {256, 512}, {512, 1}: 0.00078948ms, 0.00696096% main:@1167 = slice[axes={0, 1, 2},starts={0, 1, 0},ends={32, 2, 256}](main:@778) -> float_type, {32, 1, 256}, {4608, 256, 1}: 0.00058462ms, 0.00515468% main:@1168 = squeeze[axes={1}](main:@1167) -> float_type, {32, 256}, {4608, 1}: 0.00091948ms, 0.00810719% main:@1169 = load[offset=4390656,end=4456192](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00049648ms, 0.00437754% main:@1170 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1168,main:@1166,main:@1169) -> float_type, {32, 512}, {512, 1}: 0.0242756ms, 0.214041% main:@1171 = hip::hip_copy_literal[id=main:@literal:120] -> float_type, {512, 256}, {256, 1}: 0.00077716ms, 0.00685233% main:@1172 = load[offset=4456192,end=4521728](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00048324ms, 0.0042608% main:@1173 = gpu::code_object[code_object=13512,symbol_name=relu_kernel,global=8192,local=256,](main:@1170,main:@1172) -> float_type, {32, 512}, {512, 1}: 0.014363ms, 0.12664% main:@1174 = load[offset=4401408,end=4434176](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00062908ms, 0.00554669% main:@1175 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1173,main:@1171,main:@1174) -> float_type, {32, 256}, {256, 1}: 0.0304129ms, 0.268155% main:@1176 = hip::hip_copy_literal[id=main:@literal:62] -> float_type, {256, 512}, {512, 1}: 0.00072354ms, 0.00637956% main:@1177 = slice[axes={0, 1, 2},starts={0, 15, 0},ends={32, 16, 256}](main:@778) -> float_type, {32, 1, 256}, {4608, 256, 1}: 0.00094106ms, 0.00829746% main:@1178 = squeeze[axes={1}](main:@1177) -> float_type, {32, 256}, {4608, 1}: 0.00059232ms, 0.00522257% main:@1179 = load[offset=4532480,end=4598016](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00047994ms, 0.0042317% main:@1180 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1178,main:@1176,main:@1179) -> float_type, {32, 512}, {512, 1}: 0.0232166ms, 0.204704% main:@1181 = load[offset=4466944,end=4532480](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00060688ms, 0.00535095% main:@1182 = gpu::code_object[code_object=13512,symbol_name=relu_kernel,global=8192,local=256,](main:@1180,main:@1181) -> float_type, {32, 512}, {512, 1}: 0.0144353ms, 0.127278% main:@1183 = hip::hip_copy_literal[id=main:@literal:95] -> float_type, {512, 256}, {256, 1}: 0.00073208ms, 0.00645485% main:@1184 = load[offset=4434176,end=4466944](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00048198ms, 0.00424969% main:@1185 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1182,main:@1183,main:@1184) -> float_type, {32, 256}, {256, 1}: 0.0301021ms, 0.265415% main:@1186 = hip::hip_copy_literal[id=main:@literal:138] -> float_type, {256, 512}, {512, 1}: 0.00082486ms, 0.00727291% main:@1187 = slice[axes={0, 1, 2},starts={0, 8, 0},ends={32, 9, 256}](main:@778) -> float_type, {32, 1, 256}, {4608, 256, 1}: 0.00063086ms, 0.00556238% main:@1188 = squeeze[axes={1}](main:@1187) -> float_type, {32, 256}, {4608, 1}: 0.00089694ms, 0.00790845% main:@1189 = load[offset=4565248,end=4630784](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00049046ms, 0.00432446% main:@1190 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1188,main:@1186,main:@1189) -> float_type, {32, 512}, {512, 1}: 0.0238465ms, 0.210258% main:@1191 = load[offset=4499712,end=4565248](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00060868ms, 0.00536682% main:@1192 = gpu::code_object[code_object=13512,symbol_name=relu_kernel,global=8192,local=256,](main:@1190,main:@1191) -> float_type, {32, 512}, {512, 1}: 0.0143612ms, 0.126625% main:@1193 = hip::hip_copy_literal[id=main:@literal:44] -> float_type, {512, 256}, {256, 1}: 0.00074348ms, 0.00655537% main:@1194 = load[offset=4466944,end=4499712](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00048314ms, 0.00425991% main:@1195 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1192,main:@1193,main:@1194) -> float_type, {32, 256}, {256, 1}: 0.030329ms, 0.267416% main:@1196 = hip::hip_copy_literal[id=main:@literal:48] -> float_type, {256, 512}, {512, 1}: 0.00081944ms, 0.00722512% main:@1197 = slice[axes={0, 1, 2},starts={0, 5, 0},ends={32, 6, 256}](main:@778) -> float_type, {32, 1, 256}, {4608, 256, 1}: 0.00093948ms, 0.00828353% main:@1198 = squeeze[axes={1}](main:@1197) -> float_type, {32, 256}, {4608, 1}: 0.00057884ms, 0.00510372% main:@1199 = load[offset=4602112,end=4667648](main:@1) -> float_type, {32, 512}, {512, 1}: 0.000476ms, 0.00419696% main:@1200 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1198,main:@1196,main:@1199) -> float_type, {32, 512}, {512, 1}: 0.0232269ms, 0.204795% main:@1201 = load[offset=4536576,end=4602112](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00064842ms, 0.00571721% main:@1202 = gpu::code_object[code_object=13512,symbol_name=relu_kernel,global=8192,local=256,](main:@1200,main:@1201) -> float_type, {32, 512}, {512, 1}: 0.014385ms, 0.126834% main:@1203 = load[offset=4503808,end=4536576](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00061276ms, 0.00540279% main:@1204 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1202,main:@768,main:@1203) -> float_type, {32, 256}, {256, 1}: 0.0299819ms, 0.264355% main:@1205 = slice[axes={0, 1, 2},starts={0, 2, 0},ends={32, 3, 256}](main:@778) -> float_type, {32, 1, 256}, {4608, 256, 1}: 0.00077844ms, 0.00686362% main:@1206 = squeeze[axes={1}](main:@1205) -> float_type, {32, 256}, {4608, 1}: 0.00058074ms, 0.00512047% main:@1207 = load[offset=4634880,end=4700416](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00048696ms, 0.0042936% main:@1208 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1206,main:@759,main:@1207) -> float_type, {32, 512}, {512, 1}: 0.0239511ms, 0.21118% main:@1209 = load[offset=4569344,end=4634880](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00060812ms, 0.00536188% main:@1210 = gpu::code_object[code_object=13512,symbol_name=relu_kernel,global=8192,local=256,](main:@1208,main:@1209) -> float_type, {32, 512}, {512, 1}: 0.0144986ms, 0.127836% main:@1211 = hip::hip_copy_literal[id=main:@literal:97] -> float_type, {512, 256}, {256, 1}: 0.00073576ms, 0.0064873% main:@1212 = load[offset=4536576,end=4569344](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00050612ms, 0.00446253% main:@1213 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1210,main:@1211,main:@1212) -> float_type, {32, 256}, {256, 1}: 0.0301419ms, 0.265765% main:@1214 = slice[axes={0, 1, 2},starts={0, 0, 0},ends={32, 1, 256}](main:@778) -> float_type, {32, 1, 256}, {4608, 256, 1}: 0.00074426ms, 0.00656225% main:@1215 = squeeze[axes={1}](main:@1214) -> float_type, {32, 256}, {4608, 1}: 0.00091218ms, 0.00804282% main:@1216 = load[offset=4667648,end=4733184](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00049694ms, 0.00438159% main:@1217 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1215,main:@759,main:@1216) -> float_type, {32, 512}, {512, 1}: 0.0231518ms, 0.204133% main:@1218 = load[offset=4602112,end=4667648](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00060874ms, 0.00536735% main:@1219 = gpu::code_object[code_object=13512,symbol_name=relu_kernel,global=8192,local=256,](main:@1217,main:@1218) -> float_type, {32, 512}, {512, 1}: 0.0142999ms, 0.126085% main:@1220 = hip::hip_copy_literal[id=main:@literal:70] -> float_type, {512, 256}, {256, 1}: 0.00076872ms, 0.00677791% main:@1221 = load[offset=4569344,end=4602112](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00046828ms, 0.00412889% main:@1222 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1219,main:@1220,main:@1221) -> float_type, {32, 256}, {256, 1}: 0.0308588ms, 0.272087% main:@1223 = hip::hip_copy_literal[id=main:@literal:86] -> float_type, {256, 512}, {512, 1}: 0.0007645ms, 0.00674071% main:@1224 = load[offset=4602112,end=4667648](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00049654ms, 0.00437806% main:@1225 = slice[axes={0, 1, 2},starts={0, 14, 0},ends={32, 15, 256}](main:@778) -> float_type, {32, 1, 256}, {4608, 256, 1}: 0.00061786ms, 0.00544776% main:@1226 = squeeze[axes={1}](main:@1225) -> float_type, {32, 256}, {4608, 1}: 0.00066252ms, 0.00584153% main:@1227 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1226,main:@1223,main:@1224) -> float_type, {32, 512}, {512, 1}: 0.0241675ms, 0.213089% main:@1228 = load[offset=4667648,end=4733184](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00062936ms, 0.00554916% main:@1229 = gpu::code_object[code_object=13512,symbol_name=relu_kernel,global=8192,local=256,](main:@1227,main:@1228) -> float_type, {32, 512}, {512, 1}: 0.0143377ms, 0.126417% main:@1230 = load[offset=4602112,end=4634880](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00064362ms, 0.00567489% main:@1231 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1229,main:@768,main:@1230) -> float_type, {32, 256}, {256, 1}: 0.030051ms, 0.264964% main:@1232 = hip::hip_copy_literal[id=main:@literal:141] -> float_type, {256, 512}, {512, 1}: 0.00081676ms, 0.00720149% main:@1233 = slice[axes={0, 1, 2},starts={0, 12, 0},ends={32, 13, 256}](main:@778) -> float_type, {32, 1, 256}, {4608, 256, 1}: 0.0006234ms, 0.00549661% main:@1234 = squeeze[axes={1}](main:@1233) -> float_type, {32, 256}, {4608, 1}: 0.00090642ms, 0.00799204% main:@1235 = load[offset=6371584,end=6437120](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00047496ms, 0.00418779% main:@1236 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1234,main:@1232,main:@1235) -> float_type, {32, 512}, {512, 1}: 0.0233761ms, 0.20611% main:@1237 = load[offset=4667648,end=4733184](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00062362ms, 0.00549855% main:@1238 = gpu::code_object[code_object=13512,symbol_name=relu_kernel,global=8192,local=256,](main:@1236,main:@1237) -> float_type, {32, 512}, {512, 1}: 0.014397ms, 0.12694% main:@1239 = hip::hip_copy_literal[id=main:@literal:113] -> float_type, {512, 256}, {256, 1}: 0.0007402ms, 0.00652645% main:@1240 = load[offset=4634880,end=4667648](main:@1) -> float_type, {32, 256}, {256, 1}: 0.0004738ms, 0.00417756% main:@1241 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1238,main:@1239,main:@1240) -> float_type, {32, 256}, {256, 1}: 0.0300874ms, 0.265285% main:@1242 = slice[axes={0, 1, 2},starts={0, 17, 0},ends={32, 18, 256}](main:@778) -> float_type, {32, 1, 256}, {4608, 256, 1}: 0.00076574ms, 0.00675164% main:@1243 = squeeze[axes={1}](main:@1242) -> float_type, {32, 256}, {4608, 1}: 0.00059288ms, 0.00522751% main:@1244 = load[offset=4667648,end=4733184](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00048208ms, 0.00425057% main:@1245 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1243,main:@762,main:@1244) -> float_type, {32, 512}, {512, 1}: 0.0236811ms, 0.2088% main:@1246 = hip::hip_copy_literal[id=main:@literal:96] -> float_type, {512, 256}, {256, 1}: 0.00075606ms, 0.00666629% main:@1247 = load[offset=6371584,end=6437120](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00049426ms, 0.00435796% main:@1248 = gpu::code_object[code_object=13512,symbol_name=relu_kernel,global=8192,local=256,](main:@1245,main:@1247) -> float_type, {32, 512}, {512, 1}: 0.01436ms, 0.126614% main:@1249 = load[offset=4667648,end=4700416](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00062828ms, 0.00553964% main:@1250 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1248,main:@1246,main:@1249) -> float_type, {32, 256}, {256, 1}: 0.0308773ms, 0.272249% main:@1251 = slice[axes={0, 1, 2},starts={0, 10, 0},ends={32, 11, 256}](main:@778) -> float_type, {32, 1, 256}, {4608, 256, 1}: 0.00072598ms, 0.00640107% main:@1252 = squeeze[axes={1}](main:@1251) -> float_type, {32, 256}, {4608, 1}: 0.00057082ms, 0.005033% main:@1253 = load[offset=6371584,end=6437120](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00047664ms, 0.0042026% main:@1254 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1252,main:@762,main:@1253) -> float_type, {32, 512}, {512, 1}: 0.023397ms, 0.206295% main:@1255 = hip::hip_copy_literal[id=main:@literal:42] -> float_type, {512, 256}, {256, 1}: 0.00079024ms, 0.00696766% main:@1256 = load[offset=6437120,end=6502656](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00047518ms, 0.00418973% main:@1257 = gpu::code_object[code_object=13512,symbol_name=relu_kernel,global=8192,local=256,](main:@1254,main:@1256) -> float_type, {32, 512}, {512, 1}: 0.0142959ms, 0.126049% main:@1258 = load[offset=4700416,end=4733184](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00061412ms, 0.00541478% main:@1259 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1257,main:@1255,main:@1258) -> float_type, {32, 256}, {256, 1}: 0.0301847ms, 0.266143% main:@1260 = slice[axes={0, 1},starts={0, 305},ends={32, 337}](uFSOc5Chp2:0) -> float_type, {32, 32}, {481, 1}: 0.0008658ms, 0.00763388% main:@1261 = slice[axes={0, 1},starts={0, 73},ends={32, 105}](n4iptOVFwE:0) -> float_type, {32, 32}, {113, 1}: 0.0007686ms, 0.00677686% main:@1262 = load[offset=6375680,end=6386944](main:@1) -> float_type, {32, 88}, {88, 1}: 0.0005136ms, 0.00452849% main:@1263 = slice[axes={0, 1},starts={0, 269},ends={32, 285}](nv2tzWRf1I:0) -> float_type, {32, 16}, {301, 1}: 0.00080076ms, 0.00706042% main:@1264 = slice[axes={0, 1},starts={0, 533},ends={32, 541}](lyiDxfzZg3:0) -> float_type, {32, 8}, {701, 1}: 0.00063122ms, 0.00556556% main:@1265 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=704,local=256,](main:@1264,main:@1263,main:@1260,main:@1261,main:@1262) -> float_type, {32, 88}, {88, 1}: 0.0161882ms, 0.142734% main:@1266 = hip::hip_copy_literal[id=main:@literal:77] -> float_type, {88, 32}, {32, 1}: 0.00078678ms, 0.00693715% main:@1267 = load[offset=6371584,end=6375680](main:@1) -> float_type, {32, 32}, {32, 1}: 0.00046628ms, 0.00411126% main:@1268 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1265,main:@1266,main:@1267) -> float_type, {32, 32}, {32, 1}: 0.0203494ms, 0.179424% main:@1269 = slice[axes={0, 1},starts={0, 525},ends={32, 533}](lyiDxfzZg3:0) -> float_type, {32, 8}, {701, 1}: 0.00080424ms, 0.0070911% main:@1270 = slice[axes={0, 1},starts={0, 41},ends={32, 73}](n4iptOVFwE:0) -> float_type, {32, 32}, {113, 1}: 0.00064664ms, 0.00570152% main:@1271 = slice[axes={0, 1},starts={0, 253},ends={32, 269}](nv2tzWRf1I:0) -> float_type, {32, 16}, {301, 1}: 0.0008489ms, 0.00748487% main:@1272 = slice[axes={0, 1},starts={0, 273},ends={32, 305}](uFSOc5Chp2:0) -> float_type, {32, 32}, {481, 1}: 0.0009408ms, 0.00829517% main:@1273 = load[offset=6379776,end=6391040](main:@1) -> float_type, {32, 88}, {88, 1}: 0.00052706ms, 0.00464716% main:@1274 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=704,local=256,](main:@1269,main:@1271,main:@1272,main:@1270,main:@1273) -> float_type, {32, 88}, {88, 1}: 0.0160379ms, 0.141408% main:@1275 = hip::hip_copy_literal[id=main:@literal:51] -> float_type, {88, 32}, {32, 1}: 0.0007747ms, 0.00683064% main:@1276 = load[offset=6375680,end=6379776](main:@1) -> float_type, {32, 32}, {32, 1}: 0.00048802ms, 0.00430294% main:@1277 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1274,main:@1275,main:@1276) -> float_type, {32, 32}, {32, 1}: 0.020063ms, 0.176899% main:@1278 = slice[axes={1},starts={128},ends={256}](main:@685) -> float_type, {1600, 128}, {256, 1}: 0.00076874ms, 0.00677809% main:@1279 = reshape_lazy[dims={32, 50, 128}](main:@1278) -> float_type, {32, 50, 128}, {12800, 256, 1}: 0.00063532ms, 0.00560171% main:@1280 = slice[axes={2},starts={64},ends={128}](main:@1279) -> float_type, {32, 50, 64}, {12800, 256, 1}: 0.00071898ms, 0.00633935% main:@1281 = slice[axes={2},starts={0},ends={64}](main:@1279) -> float_type, {32, 50, 64}, {12800, 256, 1}: 0.00099958ms, 0.00881344% main:@1282 = unsqueeze[axes={0},steps={}](main:@1280) -> float_type, {1, 32, 50, 64}, {409600, 12800, 256, 1}: 0.00064176ms, 0.00565849% main:@1283 = unsqueeze[axes={0},steps={}](main:@1281) -> float_type, {1, 32, 50, 64}, {409600, 12800, 256, 1}: 0.00093638ms, 0.0082562% main:@1284 = load[offset=6379776,end=7198976](main:@1) -> float_type, {2, 32, 50, 64}, {102400, 3200, 64, 1}: 0.00049982ms, 0.00440698% main:@1285 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=51200,local=256,](main:@1283,main:@1282,main:@1284) -> float_type, {2, 32, 50, 64}, {102400, 3200, 64, 1}: 0.0189388ms, 0.166986% main:@1286 = slice[axes={1},starts={128},ends={256}](main:@661) -> float_type, {32, 128}, {256, 1}: 0.00076552ms, 0.0067497% main:@1287 = reshape_lazy[dims={32, 1, 128}](main:@1286) -> float_type, {32, 1, 128}, {256, 128, 1}: 0.00054816ms, 0.00483321% main:@1288 = slice[axes={2},starts={0},ends={64}](main:@1287) -> float_type, {32, 1, 64}, {256, 128, 1}: 0.00100186ms, 0.00883354% main:@1289 = unsqueeze[axes={0},steps={}](main:@1288) -> float_type, {1, 32, 1, 64}, {8192, 256, 128, 1}: 0.00060468ms, 0.00533155% main:@1290 = slice[axes={2},starts={64},ends={128}](main:@1287) -> float_type, {32, 1, 64}, {256, 128, 1}: 0.0007365ms, 0.00649383% main:@1291 = unsqueeze[axes={0},steps={}](main:@1290) -> float_type, {1, 32, 1, 64}, {8192, 256, 128, 1}: 0.00060946ms, 0.0053737% main:@1292 = load[offset=7211776,end=7228160](main:@1) -> float_type, {2, 32, 1, 64}, {2048, 64, 64, 1}: 0.00051926ms, 0.00457839% main:@1293 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=1024,local=256,](main:@1289,main:@1291,main:@1292) -> float_type, {2, 32, 1, 64}, {2048, 64, 64, 1}: 0.0135171ms, 0.119182% main:@1294 = load[offset=7198976,end=7211776](main:@1) -> float_type, {2, 32, 1, 50}, {1600, 50, 50, 1}: 0.00064342ms, 0.00567313% main:@1295 = transpose[permutation={0, 1, 3, 2}](main:@1285) -> float_type, {2, 32, 64, 50}, {102400, 3200, 1, 64}: 0.00058786ms, 0.00518325% main:@1296 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1293,main:@1295,main:@1294) -> float_type, {2, 32, 1, 50}, {1600, 50, 50, 1}: 0.02212ms, 0.195036% WjNeBZroCJ:0 = @param:WjNeBZroCJ:0 -> int32_type, {32, 50}, {50, 1}: 0.00075406ms, 0.00664866% main:@1298 = unsqueeze[axes={1},steps={}](WjNeBZroCJ:0) -> int32_type, {32, 1, 50}, {50, 50, 1}: 0.00059786ms, 0.00527142% main:@1299 = load[offset=4390656,end=4397056](main:@1) -> float_type, {32, 1, 50}, {50, 50, 1}: 0.00049522ms, 0.00436643% main:@1300 = gpu::code_object[code_object=13536,symbol_name=convert_pow_add_kernel,global=800,local=256,](main:@1298,main:@1299) -> float_type, {32, 1, 50}, {50, 50, 1}: 0.0128615ms, 0.113402% main:@1301 = multibroadcast[out_lens={2, 32, 1, 50}](main:@1300) -> float_type, {2, 32, 1, 50}, {0, 50, 50, 1}: 0.00071152ms, 0.00627357% main:@1302 = load[offset=6392576,end=6405376](main:@1) -> float_type, {2, 32, 1, 50}, {1600, 50, 50, 1}: 0.00051534ms, 0.00454383% main:@1303 = gpu::code_object[code_object=13520,symbol_name=pow_mul_kernel,global=1600,local=256,](main:@1296,main:@1301,main:@1302) -> float_type, {2, 32, 1, 50}, {1600, 50, 50, 1}: 0.0154184ms, 0.135946% main:@1304 = load[offset=6379776,end=6392576](main:@1) -> float_type, {2, 32, 1, 50}, {1600, 50, 50, 1}: 0.00063354ms, 0.00558601% main:@1305 = gpu::code_object[code_object=14584,symbol_name=softmax_kernel,global=4096,local=64,](main:@1303,main:@1304) -> float_type, {2, 32, 1, 50}, {1600, 50, 50, 1}: 0.0158666ms, 0.139898% main:@1306 = slice[axes={1},starts={0},ends={128}](main:@685) -> float_type, {1600, 128}, {256, 1}: 0.00076466ms, 0.00674212% main:@1307 = reshape_lazy[dims={32, 50, 128}](main:@1306) -> float_type, {32, 50, 128}, {12800, 256, 1}: 0.00060364ms, 0.00532238% main:@1308 = slice[axes={2},starts={0},ends={64}](main:@1307) -> float_type, {32, 50, 64}, {12800, 256, 1}: 0.00068526ms, 0.00604204% main:@1309 = unsqueeze[axes={0},steps={}](main:@1308) -> float_type, {1, 32, 50, 64}, {409600, 12800, 256, 1}: 0.0006155ms, 0.00542695% main:@1310 = slice[axes={2},starts={64},ends={128}](main:@1307) -> float_type, {32, 50, 64}, {12800, 256, 1}: 0.00066968ms, 0.00590467% main:@1311 = unsqueeze[axes={0},steps={}](main:@1310) -> float_type, {1, 32, 50, 64}, {409600, 12800, 256, 1}: 0.0009825ms, 0.00866284% main:@1312 = load[offset=6392576,end=7211776](main:@1) -> float_type, {2, 32, 50, 64}, {102400, 3200, 64, 1}: 0.00048838ms, 0.00430612% main:@1313 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=51200,local=256,](main:@1309,main:@1311,main:@1312) -> float_type, {2, 32, 50, 64}, {102400, 3200, 64, 1}: 0.0191022ms, 0.168427% main:@1314 = load[offset=4733184,end=4749568](main:@1) -> float_type, {2, 32, 1, 64}, {2048, 64, 64, 1}: 0.00061552ms, 0.00542713% main:@1315 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1305,main:@1313,main:@1314) -> float_type, {2, 32, 1, 64}, {2048, 64, 64, 1}: 0.0204157ms, 0.180008% main:@1316 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](EZBIAqn6LS:0) -> float_type, {32, 2}, {13, 1}: 0.00083198ms, 0.00733569% main:@1317 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](Vbv2Om3ptd:0) -> float_type, {32, 8}, {17, 1}: 0.0006382ms, 0.0056271% main:@1318 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](Hnjq487SIY:0) -> float_type, {32, 2}, {13, 1}: 0.00077338ms, 0.006819% main:@1319 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](JCVMzpWyri:0) -> float_type, {32, 4}, {17, 1}: 0.00060084ms, 0.00529769% main:@1320 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](JSPsGophtx:0) -> float_type, {32, 8}, {25, 1}: 0.00062524ms, 0.00551283% main:@1321 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](JjyOIxALzM:0) -> float_type, {32, 2}, {5, 1}: 0.00058932ms, 0.00519612% main:@1322 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](JaHTUxB6zL:0) -> float_type, {32, 4}, {17, 1}: 0.00075412ms, 0.00664918% main:@1323 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](khJcIxW3zv:0) -> float_type, {32, 4}, {9, 1}: 0.00063404ms, 0.00559042% main:@1324 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](Tej5M3O6As:0) -> float_type, {32, 8}, {17, 1}: 0.00060508ms, 0.00533508% main:@1325 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](r1WGP7Xe9H:0) -> float_type, {32, 2}, {13, 1}: 0.00063074ms, 0.00556133% main:@1326 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](4gFa5o2YsJ:0) -> float_type, {32, 4}, {9, 1}: 0.00074678ms, 0.00658447% main:@1327 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](TSqGwrHeo4:0) -> float_type, {32, 2}, {5, 1}: 0.00064532ms, 0.00568988% main:@1328 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](6XMrIsBjLY:0) -> float_type, {32, 2}, {13, 1}: 0.0005981ms, 0.00527353% main:@1329 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](72zWmh93Bo:0) -> float_type, {32, 4}, {17, 1}: 0.00061492ms, 0.00542184% main:@1330 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](qzd23aNGIT:0) -> float_type, {32, 4}, {17, 1}: 0.00071416ms, 0.00629685% main:@1331 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](8vM5p4RzXd:0) -> float_type, {32, 2}, {13, 1}: 0.00058454ms, 0.00515397% main:@1332 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](9dgeU4bKka:0) -> float_type, {32, 4}, {17, 1}: 0.000602ms, 0.00530792% main:@1333 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](AB6lekmMxa:0) -> float_type, {32, 8}, {25, 1}: 0.00061874ms, 0.00545552% main:@1334 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](eQquakfWBY:0) -> float_type, {32, 4}, {17, 1}: 0.00073382ms, 0.0064702% main:@1335 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](bTeX85HPmJ:0) -> float_type, {32, 4}, {17, 1}: 0.00070876ms, 0.00624924% main:@1336 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](BYyjkPCX0m:0) -> float_type, {32, 2}, {13, 1}: 0.00063288ms, 0.00558019% main:@1337 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](fMiQsYNnet:0) -> float_type, {32, 2}, {5, 1}: 0.00066822ms, 0.00589179% main:@1338 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](IRem4MiFDP:0) -> float_type, {32, 2}, {5, 1}: 0.0006671ms, 0.00588192% main:@1339 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](EKpP4xMXu8:0) -> float_type, {32, 4}, {17, 1}: 0.00071284ms, 0.00628521% main:@1340 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](EUciqFuptg:0) -> float_type, {32, 4}, {17, 1}: 0.00062114ms, 0.00547668% main:@1341 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](LMobYAd5ru:0) -> float_type, {32, 4}, {17, 1}: 0.000698ms, 0.00615437% main:@1342 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](nyLxrsTmOB:0) -> float_type, {32, 8}, {17, 1}: 0.00062416ms, 0.00550331% main:@1343 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](iCbkMrmXBh:0) -> float_type, {32, 4}, {9, 1}: 0.00063336ms, 0.00558443% main:@1344 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](KasFvpq4IG:0) -> float_type, {32, 2}, {5, 1}: 0.00063174ms, 0.00557014% main:@1345 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](4M92ryD8ux:0) -> float_type, {32, 4}, {9, 1}: 0.00063966ms, 0.00563997% main:@1346 = load[offset=4770048,end=4790528](main:@1) -> float_type, {32, 160}, {160, 1}: 0.0004937ms, 0.00435302% main:@1347 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](fZhaL2tCm6:0) -> float_type, {32, 8}, {17, 1}: 0.00099394ms, 0.00876371% main:@1348 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](XPcHGKbop2:0) -> float_type, {32, 2}, {13, 1}: 0.00065484ms, 0.00577382% main:@1349 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](lQmygxs5aD:0) -> float_type, {32, 2}, {13, 1}: 0.00062182ms, 0.00548268% main:@1350 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](d8RoEzG52p:0) -> float_type, {32, 8}, {17, 1}: 0.0006444ms, 0.00568177% main:@1351 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](omr2g4XaH5:0) -> float_type, {32, 8}, {17, 1}: 0.00092906ms, 0.00819166% main:@1352 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](agDiJj8lLK:0) -> float_type, {32, 2}, {13, 1}: 0.0006371ms, 0.0056174% main:@1353 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](aqMQBgb3hf:0) -> float_type, {32, 8}, {17, 1}: 0.00062512ms, 0.00551177% main:@1354 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](cN2JSTnFOu:0) -> float_type, {32, 4}, {9, 1}: 0.00064146ms, 0.00565585% main:@1355 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=134,local=256,](main:@1321,main:@1316,main:@1338,main:@1349,main:@1327,main:@1325,main:@1352,main:@1348,main:@1331,main:@1337,main:@1336,main:@1318,main:@1344,main:@1328,main:@1354,main:@1332,main:@1345,main:@1341,main:@1323,main:@1340,main:@1335,main:@1339,main:@1334,main:@1319,main:@1329,main:@1330,main:@1343,main:@1322,main:@1326,main:@1320,main:@1347,main:@1350,main:@1324,main:@1351,main:@1333,main:@1353,main:@1317,main:@1342,main:@1346) -> float_type, {32, 160}, {160, 1}: 0.0445174ms, 0.392517% main:@1356 = load[offset=4749568,end=4770048](main:@1) -> float_type, {32, 160}, {160, 1}: 0.00065568ms, 0.00578123% main:@1357 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1355,main:@913,main:@1356) -> float_type, {32, 160}, {160, 1}: 0.0216555ms, 0.19094% EhUjmCTaGW:0 = @param:EhUjmCTaGW:0 -> float_type, {32, 9}, {9, 1}: 0.0007084ms, 0.00624606% main:@1359 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](EhUjmCTaGW:0) -> float_type, {32, 4}, {9, 1}: 0.00065184ms, 0.00574737% Pu1O5qFxY6:0 = @param:Pu1O5qFxY6:0 -> float_type, {32, 17}, {17, 1}: 0.00059676ms, 0.00526172% main:@1361 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](Pu1O5qFxY6:0) -> float_type, {32, 4}, {17, 1}: 0.00076962ms, 0.00678585% FWhBxj4Tun:0 = @param:FWhBxj4Tun:0 -> float_type, {32, 9}, {9, 1}: 0.000544ms, 0.00479653% main:@1363 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](FWhBxj4Tun:0) -> float_type, {32, 4}, {9, 1}: 0.0006469ms, 0.00570381% yVxKAL4U5G:0 = @param:yVxKAL4U5G:0 -> float_type, {32, 9}, {9, 1}: 0.00056126ms, 0.00494871% main:@1365 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](yVxKAL4U5G:0) -> float_type, {32, 4}, {9, 1}: 0.00078054ms, 0.00688213% HmATr3gKsW:0 = @param:HmATr3gKsW:0 -> float_type, {32, 9}, {9, 1}: 0.00053888ms, 0.00475138% main:@1367 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](HmATr3gKsW:0) -> float_type, {32, 4}, {9, 1}: 0.00077222ms, 0.00680877% IMdFk7D16X:0 = @param:IMdFk7D16X:0 -> float_type, {32, 9}, {9, 1}: 0.00049324ms, 0.00434897% main:@1369 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](IMdFk7D16X:0) -> float_type, {32, 4}, {9, 1}: 0.00076348ms, 0.00673171% IxpbBg7JTO:0 = @param:IxpbBg7JTO:0 -> float_type, {32, 5}, {5, 1}: 0.00056698ms, 0.00499914% main:@1371 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](IxpbBg7JTO:0) -> float_type, {32, 2}, {5, 1}: 0.00063242ms, 0.00557614% J3H2EjAXt8:0 = @param:J3H2EjAXt8:0 -> float_type, {32, 5}, {5, 1}: 0.0004989ms, 0.00439887% main:@1373 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](J3H2EjAXt8:0) -> float_type, {32, 2}, {5, 1}: 0.00076858ms, 0.00677668% ZRYNFfjSez:0 = @param:ZRYNFfjSez:0 -> float_type, {32, 33}, {33, 1}: 0.00051186ms, 0.00451314% main:@1375 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](ZRYNFfjSez:0) -> float_type, {32, 8}, {33, 1}: 0.00063248ms, 0.00557667% KCNfS4GlpP:0 = @param:KCNfS4GlpP:0 -> float_type, {32, 9}, {9, 1}: 0.00053042ms, 0.00467679% main:@1377 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](KCNfS4GlpP:0) -> float_type, {32, 4}, {9, 1}: 0.00076228ms, 0.00672113% vAbKepTjQg:0 = @param:vAbKepTjQg:0 -> float_type, {32, 5}, {5, 1}: 0.00055788ms, 0.00491891% main:@1379 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](vAbKepTjQg:0) -> float_type, {32, 2}, {5, 1}: 0.00062958ms, 0.0055511% NJBt1ubXFW:0 = @param:NJBt1ubXFW:0 -> float_type, {32, 5}, {5, 1}: 0.00049126ms, 0.00433151% main:@1381 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](NJBt1ubXFW:0) -> float_type, {32, 2}, {5, 1}: 0.0007471ms, 0.00658729% Ouh1XpPZaj:0 = @param:Ouh1XpPZaj:0 -> float_type, {32, 9}, {9, 1}: 0.00055594ms, 0.0049018% main:@1383 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](Ouh1XpPZaj:0) -> float_type, {32, 4}, {9, 1}: 0.0009359ms, 0.00825197% gqk5byC8aJ:0 = @param:gqk5byC8aJ:0 -> float_type, {32, 5}, {5, 1}: 0.00053ms, 0.00467309% main:@1385 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](gqk5byC8aJ:0) -> float_type, {32, 2}, {5, 1}: 0.00076662ms, 0.0067594% gsP96ZTbn0:0 = @param:gsP96ZTbn0:0 -> float_type, {32, 17}, {17, 1}: 0.00059048ms, 0.00520635% main:@1387 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](gsP96ZTbn0:0) -> float_type, {32, 8}, {17, 1}: 0.0006199ms, 0.00546575% O5aJeUQh2G:0 = @param:O5aJeUQh2G:0 -> float_type, {32, 17}, {17, 1}: 0.0005149ms, 0.00453995% main:@1389 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](O5aJeUQh2G:0) -> float_type, {32, 8}, {17, 1}: 0.00075104ms, 0.00662203% mTMtDEo3r8:0 = @param:mTMtDEo3r8:0 -> float_type, {32, 9}, {9, 1}: 0.00051384ms, 0.0045306% main:@1391 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](mTMtDEo3r8:0) -> float_type, {32, 4}, {9, 1}: 0.00063358ms, 0.00558637% U6GqhP8k7I:0 = @param:U6GqhP8k7I:0 -> float_type, {32, 9}, {9, 1}: 0.00047544ms, 0.00419202% main:@1393 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](U6GqhP8k7I:0) -> float_type, {32, 4}, {9, 1}: 0.00068286ms, 0.00602088% UMO0o14RuI:0 = @param:UMO0o14RuI:0 -> float_type, {32, 9}, {9, 1}: 0.00057168ms, 0.00504058% main:@1395 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](UMO0o14RuI:0) -> float_type, {32, 4}, {9, 1}: 0.00070476ms, 0.00621397% Uc42tiuavR:0 = @param:Uc42tiuavR:0 -> float_type, {32, 9}, {9, 1}: 0.000554ms, 0.0048847% main:@1397 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](Uc42tiuavR:0) -> float_type, {32, 4}, {9, 1}: 0.00067886ms, 0.00598561% E1ZkVIXqS5:0 = @param:E1ZkVIXqS5:0 -> float_type, {32, 9}, {9, 1}: 0.0005034ms, 0.00443855% 5yColzZ8K3:0 = @param:5yColzZ8K3:0 -> float_type, {32, 5}, {5, 1}: 0.00053982ms, 0.00475967% main:@1400 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](5yColzZ8K3:0) -> float_type, {32, 2}, {5, 1}: 0.00073362ms, 0.00646843% 6Wh2Jm7GHC:0 = @param:6Wh2Jm7GHC:0 -> float_type, {32, 5}, {5, 1}: 0.00053506ms, 0.0047177% main:@1402 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](6Wh2Jm7GHC:0) -> float_type, {32, 2}, {5, 1}: 0.00067466ms, 0.00594857% zX6xKOWG9S:0 = @param:zX6xKOWG9S:0 -> float_type, {32, 5}, {5, 1}: 0.00056072ms, 0.00494395% main:@1404 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](zX6xKOWG9S:0) -> float_type, {32, 2}, {5, 1}: 0.00068224ms, 0.00601541% gdck97AK4w:0 = @param:gdck97AK4w:0 -> float_type, {32, 17}, {17, 1}: 0.00053802ms, 0.0047438% main:@1406 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](gdck97AK4w:0) -> float_type, {32, 4}, {17, 1}: 0.00070322ms, 0.00620039% NLt5JnRMjG:0 = @param:NLt5JnRMjG:0 -> float_type, {32, 5}, {5, 1}: 0.00050568ms, 0.00445865% main:@1408 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](NLt5JnRMjG:0) -> float_type, {32, 2}, {5, 1}: 0.00069762ms, 0.00615102% Cw59eOo7M1:0 = @param:Cw59eOo7M1:0 -> float_type, {32, 9}, {9, 1}: 0.000502ms, 0.00442621% main:@1410 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](Cw59eOo7M1:0) -> float_type, {32, 4}, {9, 1}: 0.00101694ms, 0.00896651% main:@1411 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](E1ZkVIXqS5:0) -> float_type, {32, 4}, {9, 1}: 0.0006742ms, 0.00594452% mNgQ4ohWqU:0 = @param:mNgQ4ohWqU:0 -> float_type, {32, 9}, {9, 1}: 0.0004745ms, 0.00418373% main:@1413 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](mNgQ4ohWqU:0) -> float_type, {32, 4}, {9, 1}: 0.00066004ms, 0.00581967% tIGjbcEuDd:0 = @param:tIGjbcEuDd:0 -> float_type, {32, 9}, {9, 1}: 0.00046758ms, 0.00412272% main:@1415 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](tIGjbcEuDd:0) -> float_type, {32, 4}, {9, 1}: 0.0010075ms, 0.00888327% nltbg9ZWQX:0 = @param:nltbg9ZWQX:0 -> float_type, {32, 17}, {17, 1}: 0.00050368ms, 0.00444102% main:@1417 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](nltbg9ZWQX:0) -> float_type, {32, 4}, {17, 1}: 0.0006889ms, 0.00607413% ecPTQUALkq:0 = @param:ecPTQUALkq:0 -> float_type, {32, 5}, {5, 1}: 0.0004611ms, 0.00406559% main:@1419 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](ecPTQUALkq:0) -> float_type, {32, 2}, {5, 1}: 0.00065024ms, 0.00573326% magwfJSnUu:0 = @param:magwfJSnUu:0 -> float_type, {32, 9}, {9, 1}: 0.00051078ms, 0.00450362% main:@1421 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](magwfJSnUu:0) -> float_type, {32, 4}, {9, 1}: 0.00065616ms, 0.00578546% xVErSbJ0hg:0 = @param:xVErSbJ0hg:0 -> float_type, {32, 9}, {9, 1}: 0.0005223ms, 0.00460519% main:@1423 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](xVErSbJ0hg:0) -> float_type, {32, 4}, {9, 1}: 0.00070092ms, 0.00618011% l4bReWtsMB:0 = @param:l4bReWtsMB:0 -> float_type, {32, 25}, {25, 1}: 0.00050494ms, 0.00445213% main:@1425 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](l4bReWtsMB:0) -> float_type, {32, 8}, {25, 1}: 0.00066504ms, 0.00586375% pQsi1GNxvj:0 = @param:pQsi1GNxvj:0 -> float_type, {32, 5}, {5, 1}: 0.00052478ms, 0.00462706% main:@1427 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](pQsi1GNxvj:0) -> float_type, {32, 2}, {5, 1}: 0.00065012ms, 0.0057322% lDhFNYzbxR:0 = @param:lDhFNYzbxR:0 -> float_type, {32, 17}, {17, 1}: 0.0005554ms, 0.00489704% main:@1429 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](lDhFNYzbxR:0) -> float_type, {32, 8}, {17, 1}: 0.00065032ms, 0.00573397% l5WSX3VvMK:0 = @param:l5WSX3VvMK:0 -> float_type, {32, 25}, {25, 1}: 0.00048906ms, 0.00431211% main:@1431 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](l5WSX3VvMK:0) -> float_type, {32, 8}, {25, 1}: 0.00100126ms, 0.00882825% 1QWFmJ96aB:0 = @param:1QWFmJ96aB:0 -> float_type, {32, 5}, {5, 1}: 0.00047806ms, 0.00421512% main:@1433 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](1QWFmJ96aB:0) -> float_type, {32, 2}, {5, 1}: 0.00064324ms, 0.00567154% 3VED7WSJ1b:0 = @param:3VED7WSJ1b:0 -> float_type, {32, 5}, {5, 1}: 0.00053448ms, 0.00471259% main:@1435 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](3VED7WSJ1b:0) -> float_type, {32, 2}, {5, 1}: 0.00065042ms, 0.00573485% qfwYOHde3u:0 = @param:qfwYOHde3u:0 -> float_type, {32, 17}, {17, 1}: 0.0005464ms, 0.00481769% main:@1437 = load[offset=4790528,end=4812288](main:@1) -> float_type, {32, 170}, {170, 1}: 0.00047062ms, 0.00414952% oQ5OcjZJe8:0 = @param:oQ5OcjZJe8:0 -> float_type, {32, 17}, {17, 1}: 0.00055976ms, 0.00493548% main:@1439 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](oQ5OcjZJe8:0) -> float_type, {32, 8}, {17, 1}: 0.00073972ms, 0.00652222% main:@1440 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](qfwYOHde3u:0) -> float_type, {32, 8}, {17, 1}: 0.00067534ms, 0.00595457% main:@1441 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=132,local=256,](main:@1402,main:@1419,main:@1381,main:@1379,main:@1404,main:@1427,main:@1408,main:@1435,main:@1433,main:@1373,main:@1385,main:@1371,main:@1400,main:@1359,main:@1393,main:@1411,main:@1410,main:@1377,main:@1397,main:@1369,main:@1413,main:@1391,main:@1365,main:@1406,main:@1361,main:@1415,main:@1423,main:@1395,main:@1383,main:@1417,main:@1367,main:@1421,main:@1363,main:@1440,main:@1429,main:@1387,main:@1389,main:@1375,main:@1425,main:@1431,main:@1439,main:@1437) -> float_type, {32, 170}, {170, 1}: 0.0478894ms, 0.422247% main:@1442 = hip::hip_copy_literal[id=main:@literal:101] -> float_type, {170, 160}, {160, 1}: 0.00084022ms, 0.00740834% main:@1443 = load[offset=4770048,end=4790528](main:@1) -> float_type, {32, 160}, {160, 1}: 0.00049288ms, 0.00434579% main:@1444 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1441,main:@1442,main:@1443) -> float_type, {32, 160}, {160, 1}: 0.0227355ms, 0.200462% WDLoyKsiA7:0 = @param:WDLoyKsiA7:0 -> float_type, {32, 33}, {33, 1}: 0.00072246ms, 0.00637003% main:@1446 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](WDLoyKsiA7:0) -> float_type, {32, 16}, {33, 1}: 0.00069074ms, 0.00609035% main:@1447 = unsqueeze[axes={1},steps={}](main:@1446) -> float_type, {32, 1, 16}, {33, 16, 1}: 0.00080614ms, 0.00710785% main:@1448 = slice[axes={0, 1},starts={0, 9},ends={32, 25}](LKAvFme9ZS:0) -> float_type, {32, 16}, {49, 1}: 0.00078118ms, 0.00688778% main:@1449 = unsqueeze[axes={1},steps={}](main:@1448) -> float_type, {32, 1, 16}, {49, 16, 1}: 0.00060184ms, 0.00530651% VfFM90O7LN:0 = @param:VfFM90O7LN:0 -> float_type, {32, 41}, {41, 1}: 0.00050116ms, 0.0044188% main:@1451 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](VfFM90O7LN:0) -> float_type, {32, 16}, {41, 1}: 0.0006686ms, 0.00589514% main:@1452 = unsqueeze[axes={1},steps={}](main:@1451) -> float_type, {32, 1, 16}, {41, 16, 1}: 0.00060456ms, 0.00533049% main:@1453 = slice[axes={0, 1},starts={0, 9},ends={32, 25}](TxEdCcekGS:0) -> float_type, {32, 16}, {49, 1}: 0.0007636ms, 0.00673277% main:@1454 = unsqueeze[axes={1},steps={}](main:@1453) -> float_type, {32, 1, 16}, {49, 16, 1}: 0.00057744ms, 0.00509137% 2Yy3zxNGBC:0 = @param:2Yy3zxNGBC:0 -> float_type, {32, 45}, {45, 1}: 0.00047732ms, 0.0042086% pY0t1rI5WU:0 = @param:pY0t1rI5WU:0 -> float_type, {32, 41}, {41, 1}: 0.00052084ms, 0.00459232% z3lqgZM8WU:0 = @param:z3lqgZM8WU:0 -> float_type, {32, 41}, {41, 1}: 0.0005252ms, 0.00463076% sZdFJI9nwg:0 = @param:sZdFJI9nwg:0 -> float_type, {32, 41}, {41, 1}: 0.0005487ms, 0.00483797% main:@1459 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](sZdFJI9nwg:0) -> float_type, {32, 16}, {41, 1}: 0.0007978ms, 0.00703432% main:@1460 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](z3lqgZM8WU:0) -> float_type, {32, 16}, {41, 1}: 0.0006559ms, 0.00578316% main:@1461 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](pY0t1rI5WU:0) -> float_type, {32, 16}, {41, 1}: 0.00062964ms, 0.00555163% main:@1462 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](2Yy3zxNGBC:0) -> float_type, {32, 16}, {45, 1}: 0.00077018ms, 0.00679079% main:@1463 = unsqueeze[axes={1},steps={}](main:@1462) -> float_type, {32, 1, 16}, {45, 16, 1}: 0.00060504ms, 0.00533472% main:@1464 = reshape_lazy[dims={-1, 10, 16}](main:@1444) -> float_type, {32, 10, 16}, {160, 16, 1}: 0.00054334ms, 0.00479071% main:@1465 = unsqueeze[axes={1},steps={}](main:@1460) -> float_type, {32, 1, 16}, {41, 16, 1}: 0.0007441ms, 0.00656084% main:@1466 = unsqueeze[axes={1},steps={}](main:@1461) -> float_type, {32, 1, 16}, {41, 16, 1}: 0.00057252ms, 0.00504799% main:@1467 = unsqueeze[axes={1},steps={}](main:@1459) -> float_type, {32, 1, 16}, {41, 16, 1}: 0.00058024ms, 0.00511606% main:@1468 = load[offset=4815104,end=4851968](main:@1) -> float_type, {32, 18, 16}, {288, 16, 1}: 0.00051432ms, 0.00453483% main:@1469 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=1024,local=256,](main:@1447,main:@1465,main:@1449,main:@1466,main:@1454,main:@1467,main:@1452,main:@1463,main:@1464,main:@1468) -> float_type, {32, 18, 16}, {288, 16, 1}: 0.0247795ms, 0.218485% main:@1470 = transpose[permutation={0, 2, 1}](main:@1469) -> float_type, {32, 16, 18}, {288, 1, 16}: 0.00074322ms, 0.00655308% main:@1471 = load[offset=4778240,end=4815104](main:@1) -> float_type, {32, 16, 18}, {288, 18, 1}: 0.00048332ms, 0.0042615% main:@1472 = gpu::code_object[code_object=13528,symbol_name=contiguous_kernel,global=9216,local=256,](main:@1470,main:@1471) -> float_type, {32, 16, 18}, {288, 18, 1}: 0.0144866ms, 0.12773% main:@1473 = hip::hip_copy_literal[id=main:@literal:129] -> float_type, {18, 4}, {4, 1}: 0.0007741ms, 0.00682535% main:@1474 = load[offset=4770048,end=4778240](main:@1) -> float_type, {512, 4}, {4, 1}: 0.00048756ms, 0.00429889% main:@1475 = reshape_lazy[dims={512, 18}](main:@1472) -> float_type, {512, 18}, {18, 1}: 0.00054284ms, 0.0047863% main:@1476 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1475,main:@1473,main:@1474) -> float_type, {512, 4}, {4, 1}: 0.020597ms, 0.181607% main:@1477 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](VUqfdN5Fzv:0) -> float_type, {32, 16}, {45, 1}: 0.00081636ms, 0.00719796% main:@1478 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](HRGzlnEkXQ:0) -> float_type, {32, 16}, {41, 1}: 0.00064082ms, 0.0056502% main:@1479 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](HzcyUlm8Yj:0) -> float_type, {32, 16}, {45, 1}: 0.00076662ms, 0.0067594% main:@1480 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](duyRoIEkH8:0) -> float_type, {32, 16}, {45, 1}: 0.00063774ms, 0.00562305% main:@1481 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](jznORDJr0N:0) -> float_type, {32, 16}, {45, 1}: 0.00058904ms, 0.00519365% main:@1482 = unsqueeze[axes={1},steps={}](main:@1481) -> float_type, {32, 1, 16}, {45, 16, 1}: 0.00062938ms, 0.00554933% main:@1483 = unsqueeze[axes={1},steps={}](main:@1478) -> float_type, {32, 1, 16}, {41, 16, 1}: 0.0005777ms, 0.00509366% main:@1484 = slice[axes={0, 1},starts={0, 305},ends={32, 321}](lyiDxfzZg3:0) -> float_type, {32, 16}, {701, 1}: 0.00074256ms, 0.00654726% main:@1485 = unsqueeze[axes={1},steps={}](main:@1484) -> float_type, {32, 1, 16}, {701, 16, 1}: 0.000578ms, 0.00509631% main:@1486 = unsqueeze[axes={1},steps={}](main:@1477) -> float_type, {32, 1, 16}, {45, 16, 1}: 0.00064798ms, 0.00571333% main:@1487 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](CHAYRkBvIo:0) -> float_type, {32, 16}, {41, 1}: 0.00063486ms, 0.00559765% main:@1488 = unsqueeze[axes={1},steps={}](main:@1487) -> float_type, {32, 1, 16}, {41, 16, 1}: 0.00056688ms, 0.00499826% main:@1489 = unsqueeze[axes={1},steps={}](main:@1479) -> float_type, {32, 1, 16}, {45, 16, 1}: 0.00071824ms, 0.00633283% main:@1490 = unsqueeze[axes={1},steps={}](main:@1480) -> float_type, {32, 1, 16}, {45, 16, 1}: 0.0005557ms, 0.00489969% main:@1491 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](eQI8ikEMgP:0) -> float_type, {32, 16}, {41, 1}: 0.00061496ms, 0.00542219% main:@1492 = unsqueeze[axes={1},steps={}](main:@1491) -> float_type, {32, 1, 16}, {41, 16, 1}: 0.00056226ms, 0.00495753% main:@1493 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](2RDXZrx9GA:0) -> float_type, {32, 16}, {45, 1}: 0.00072654ms, 0.00640601% main:@1494 = unsqueeze[axes={1},steps={}](main:@1493) -> float_type, {32, 1, 16}, {45, 16, 1}: 0.0005736ms, 0.00505751% main:@1495 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](2srtUfGvja:0) -> float_type, {32, 16}, {41, 1}: 0.00059574ms, 0.00525273% main:@1496 = unsqueeze[axes={1},steps={}](main:@1495) -> float_type, {32, 1, 16}, {41, 16, 1}: 0.00056402ms, 0.00497305% main:@1497 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](3Laj620MSn:0) -> float_type, {32, 16}, {45, 1}: 0.0006782ms, 0.00597979% main:@1498 = unsqueeze[axes={1},steps={}](main:@1497) -> float_type, {32, 1, 16}, {45, 16, 1}: 0.00055578ms, 0.00490039% main:@1499 = load[offset=4778240,end=4821248](main:@1) -> float_type, {32, 21, 16}, {336, 16, 1}: 0.00049552ms, 0.00436907% main:@1500 = reshape_lazy[dims={-1, 10, 16}](main:@1357) -> float_type, {32, 10, 16}, {160, 16, 1}: 0.00050916ms, 0.00448934% main:@1501 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=896,local=256,](main:@1482,main:@1483,main:@1485,main:@1494,main:@1486,main:@1488,main:@1489,main:@1496,main:@1498,main:@1490,main:@1492,main:@1500,main:@1499) -> float_type, {32, 21, 16}, {336, 16, 1}: 0.0262585ms, 0.231525% main:@1502 = load[offset=4390656,end=4401408](main:@1) -> float_type, {32, 21, 4}, {84, 4, 1}: 0.0006893ms, 0.00607766% main:@1503 = reshape_lazy[dims={32, 16, 4}](main:@1476) -> float_type, {32, 16, 4}, {64, 4, 1}: 0.00051624ms, 0.00455176% main:@1504 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1501,main:@1503,main:@1502) -> float_type, {32, 21, 4}, {84, 4, 1}: 0.0177397ms, 0.156413% main:@1505 = slice[axes={0, 1},starts={0, 517},ends={32, 525}](lyiDxfzZg3:0) -> float_type, {32, 8}, {701, 1}: 0.00081764ms, 0.00720925% main:@1506 = slice[axes={0, 1},starts={0, 241},ends={32, 273}](uFSOc5Chp2:0) -> float_type, {32, 32}, {481, 1}: 0.00062486ms, 0.00550948% main:@1507 = slice[axes={0, 1},starts={0, 237},ends={32, 253}](nv2tzWRf1I:0) -> float_type, {32, 16}, {301, 1}: 0.0009552ms, 0.00842214% main:@1508 = load[offset=4749568,end=4760832](main:@1) -> float_type, {32, 88}, {88, 1}: 0.0004661ms, 0.00410967% main:@1509 = slice[axes={0, 1},starts={0, 1},ends={32, 33}](n4iptOVFwE:0) -> float_type, {32, 32}, {113, 1}: 0.00062642ms, 0.00552324% main:@1510 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=704,local=256,](main:@1505,main:@1507,main:@1506,main:@1509,main:@1508) -> float_type, {32, 88}, {88, 1}: 0.0161044ms, 0.141995% main:@1511 = hip::hip_copy_literal[id=main:@literal:58] -> float_type, {88, 32}, {32, 1}: 0.00086378ms, 0.00761607% main:@1512 = load[offset=4499712,end=4503808](main:@1) -> float_type, {32, 32}, {32, 1}: 0.00050926ms, 0.00449022% main:@1513 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1510,main:@1511,main:@1512) -> float_type, {32, 32}, {32, 1}: 0.0199789ms, 0.176157% main:@1514 = reshape_lazy[dims={1, 32, 1, 32}](main:@1513) -> float_type, {1, 32, 1, 32}, {1024, 32, 32, 1}: 0.00070652ms, 0.00622949% main:@1515 = slice[axes={1},starts={0},ends={32}](main:@615) -> float_type, {320, 32}, {64, 1}: 0.00064446ms, 0.0056823% main:@1516 = reshape_lazy[dims={1, 32, 10, 32}](main:@1515) -> float_type, {1, 32, 10, 32}, {20480, 640, 64, 1}: 0.00070736ms, 0.0062369% main:@1517 = transpose[permutation={0, 1, 3, 2}](main:@1516) -> float_type, {1, 32, 32, 10}, {20480, 640, 1, 64}: 0.0006374ms, 0.00562005% main:@1518 = load[offset=4749568,end=4750848](main:@1) -> float_type, {1, 32, 1, 10}, {320, 10, 10, 1}: 0.00051336ms, 0.00452637% main:@1519 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1514,main:@1517,main:@1518) -> float_type, {1, 32, 1, 10}, {320, 10, 10, 1}: 0.0212822ms, 0.187648% main:@1520 = unsqueeze[axes={1},steps={}](0MZVmrKeC6:0) -> int32_type, {32, 1, 10}, {10, 10, 1}: 0.0007715ms, 0.00680243% main:@1521 = load[offset=4500992,end=4502272](main:@1) -> float_type, {32, 1, 10}, {10, 10, 1}: 0.00049462ms, 0.00436114% main:@1522 = gpu::code_object[code_object=13536,symbol_name=convert_mul_add_kernel,global=160,local=256,](main:@1520,main:@1521) -> float_type, {32, 1, 10}, {10, 10, 1}: 0.0132392ms, 0.116732% main:@1523 = load[offset=4499712,end=4500992](main:@1) -> float_type, {1, 32, 1, 10}, {320, 10, 10, 1}: 0.00065156ms, 0.0057449% main:@1524 = multibroadcast[out_lens={1, 32, 1, 10}](main:@1522) -> float_type, {1, 32, 1, 10}, {0, 10, 10, 1}: 0.00061266ms, 0.00540191% main:@1525 = gpu::code_object[code_object=13520,symbol_name=pow_div_kernel,global=160,local=256,](main:@1519,main:@1524,main:@1523) -> float_type, {1, 32, 1, 10}, {320, 10, 10, 1}: 0.0157823ms, 0.139155% main:@1526 = load[offset=4749568,end=4750848](main:@1) -> float_type, {1, 32, 1, 10}, {320, 10, 10, 1}: 0.00063598ms, 0.00560753% main:@1527 = gpu::code_object[code_object=14576,symbol_name=softmax_kernel,global=2048,local=64,](main:@1525,main:@1526) -> float_type, {1, 32, 1, 10}, {320, 10, 10, 1}: 0.0145572ms, 0.128353% main:@1528 = slice[axes={1},starts={32},ends={64}](main:@615) -> float_type, {320, 32}, {64, 1}: 0.0007484ms, 0.00659875% main:@1529 = reshape_lazy[dims={1, 32, 10, 32}](main:@1528) -> float_type, {1, 32, 10, 32}, {20480, 640, 64, 1}: 0.0005972ms, 0.0052656% main:@1530 = load[offset=4499712,end=4503808](main:@1) -> float_type, {1, 32, 1, 32}, {1024, 32, 32, 1}: 0.00047876ms, 0.0042213% main:@1531 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1527,main:@1529,main:@1530) -> float_type, {1, 32, 1, 32}, {1024, 32, 32, 1}: 0.0206604ms, 0.182165% IYWQSCjynR:0 = @param:IYWQSCjynR:0 -> float_type, {32, 30, 9}, {270, 9, 1}: 0.00074754ms, 0.00659117% main:@1533 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 30, 9}](IYWQSCjynR:0) -> float_type, {32, 30, 8}, {270, 9, 1}: 0.00068248ms, 0.00601752% 74jZTbEdLU:0 = @param:74jZTbEdLU:0 -> float_type, {32, 30, 9}, {270, 9, 1}: 0.0005581ms, 0.00492085% main:@1535 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 30, 9}](74jZTbEdLU:0) -> float_type, {32, 30, 8}, {270, 9, 1}: 0.00078312ms, 0.00690488% Z9QLFGO2Su:0 = @param:Z9QLFGO2Su:0 -> float_type, {32, 30, 9}, {270, 9, 1}: 0.00050412ms, 0.0044449% main:@1537 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 30, 9}](Z9QLFGO2Su:0) -> float_type, {32, 30, 8}, {270, 9, 1}: 0.00094756ms, 0.00835477% WixXdEln2I:0 = @param:WixXdEln2I:0 -> float_type, {32, 30, 17}, {510, 17, 1}: 0.00053178ms, 0.00468878% main:@1539 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 30, 17}](WixXdEln2I:0) -> float_type, {32, 30, 16}, {510, 17, 1}: 0.0008001ms, 0.0070546% main:@1540 = load[offset=4749568,end=5010688](main:@1) -> float_type, {32, 30, 68}, {2040, 68, 1}: 0.00048134ms, 0.00424404% 6DYAlXkjub:0 = @param:6DYAlXkjub:0 -> float_type, {32, 30, 17}, {510, 17, 1}: 0.00057082ms, 0.005033% main:@1542 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 30, 17}](6DYAlXkjub:0) -> float_type, {32, 30, 16}, {510, 17, 1}: 0.00063854ms, 0.0056301% VvUalZIQHK:0 = @param:VvUalZIQHK:0 -> float_type, {32, 30, 5}, {150, 5, 1}: 0.00056478ms, 0.00497975% main:@1544 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 30, 5}](VvUalZIQHK:0) -> float_type, {32, 30, 4}, {150, 5, 1}: 0.00078124ms, 0.00688831% bX19AmYnBL:0 = @param:bX19AmYnBL:0 -> float_type, {32, 30, 9}, {270, 9, 1}: 0.0005434ms, 0.00479124% main:@1546 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 30, 9}](bX19AmYnBL:0) -> float_type, {32, 30, 8}, {270, 9, 1}: 0.00063212ms, 0.00557349% main:@1547 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=9325,local=256,](main:@1544,main:@1539,main:@1533,main:@1542,main:@1535,main:@1546,main:@1537,main:@1540) -> float_type, {32, 30, 68}, {2040, 68, 1}: 0.0232592ms, 0.20508% main:@1548 = hip::hip_copy_literal[id=main:@literal:28] -> float_type, {68, 64}, {64, 1}: 0.0007543ms, 0.00665077% main:@1549 = load[offset=5010688,end=5256448](main:@1) -> float_type, {960, 64}, {64, 1}: 0.00046352ms, 0.00408692% main:@1550 = reshape_lazy[dims={960, 68}](main:@1547) -> float_type, {960, 68}, {68, 1}: 0.0005471ms, 0.00482386% main:@1551 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1550,main:@1548,main:@1549) -> float_type, {960, 64}, {64, 1}: 0.0212698ms, 0.187539% main:@1552 = reshape_lazy[dims={1, 32, 1, 32}](main:@1277) -> float_type, {1, 32, 1, 32}, {1024, 32, 32, 1}: 0.00069674ms, 0.00614326% main:@1553 = load[offset=3751680,end=3755520](main:@1) -> float_type, {1, 32, 1, 30}, {960, 30, 30, 1}: 0.00050762ms, 0.00447576% main:@1554 = slice[axes={1},starts={32},ends={64}](main:@1551) -> float_type, {960, 32}, {64, 1}: 0.00062764ms, 0.00553399% main:@1555 = reshape_lazy[dims={1, 32, 30, 32}](main:@1554) -> float_type, {1, 32, 30, 32}, {61440, 1920, 64, 1}: 0.00058794ms, 0.00518395% main:@1556 = transpose[permutation={0, 1, 3, 2}](main:@1555) -> float_type, {1, 32, 32, 30}, {61440, 1920, 1, 64}: 0.0006062ms, 0.00534495% main:@1557 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1552,main:@1556,main:@1553) -> float_type, {1, 32, 1, 30}, {960, 30, 30, 1}: 0.0191152ms, 0.168542% VX8uEBNndW:0 = @param:VX8uEBNndW:0 -> int32_type, {32, 30}, {30, 1}: 0.00067598ms, 0.00596021% main:@1559 = unsqueeze[axes={1},steps={}](VX8uEBNndW:0) -> int32_type, {32, 1, 30}, {30, 30, 1}: 0.0005761ms, 0.00507956% main:@1560 = load[offset=3755520,end=3759360](main:@1) -> float_type, {32, 1, 30}, {30, 30, 1}: 0.00046596ms, 0.00410844% main:@1561 = gpu::code_object[code_object=13528,symbol_name=convert_add_kernel,global=480,local=256,](main:@1559,main:@1560) -> float_type, {32, 1, 30}, {30, 30, 1}: 0.0133686ms, 0.117873% main:@1562 = multibroadcast[out_lens={1, 32, 1, 30}](main:@1561) -> float_type, {1, 32, 1, 30}, {0, 30, 30, 1}: 0.0007257ms, 0.0063986% main:@1563 = load[offset=3759616,end=3763456](main:@1) -> float_type, {1, 32, 1, 30}, {960, 30, 30, 1}: 0.00051328ms, 0.00452566% main:@1564 = gpu::code_object[code_object=13520,symbol_name=pow_sub_kernel,global=480,local=256,](main:@1557,main:@1562,main:@1563) -> float_type, {1, 32, 1, 30}, {960, 30, 30, 1}: 0.0153168ms, 0.13505% main:@1565 = load[offset=3755776,end=3759616](main:@1) -> float_type, {1, 32, 1, 30}, {960, 30, 30, 1}: 0.0006259ms, 0.00551865% main:@1566 = gpu::code_object[code_object=14584,symbol_name=softmax_kernel,global=2048,local=64,](main:@1564,main:@1565) -> float_type, {1, 32, 1, 30}, {960, 30, 30, 1}: 0.0147876ms, 0.130385% main:@1567 = load[offset=3751680,end=3755776](main:@1) -> float_type, {1, 32, 1, 32}, {1024, 32, 32, 1}: 0.00066898ms, 0.00589849% main:@1568 = slice[axes={1},starts={0},ends={32}](main:@1551) -> float_type, {960, 32}, {64, 1}: 0.00066084ms, 0.00582672% main:@1569 = reshape_lazy[dims={1, 32, 30, 32}](main:@1568) -> float_type, {1, 32, 30, 32}, {61440, 1920, 64, 1}: 0.00057934ms, 0.00510812% main:@1570 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1566,main:@1569,main:@1567) -> float_type, {1, 32, 1, 32}, {1024, 32, 32, 1}: 0.0183109ms, 0.16145% main:@1571 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](EhUjmCTaGW:0) -> float_type, {32, 4}, {9, 1}: 0.00083094ms, 0.00732652% main:@1572 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](l5WSX3VvMK:0) -> float_type, {32, 8}, {25, 1}: 0.00062906ms, 0.00554651% main:@1573 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](FWhBxj4Tun:0) -> float_type, {32, 4}, {9, 1}: 0.0010485ms, 0.00924478% main:@1574 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](Ouh1XpPZaj:0) -> float_type, {32, 4}, {9, 1}: 0.00063686ms, 0.00561529% main:@1575 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](HmATr3gKsW:0) -> float_type, {32, 4}, {9, 1}: 0.00064068ms, 0.00564897% main:@1576 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](J3H2EjAXt8:0) -> float_type, {32, 2}, {5, 1}: 0.00060896ms, 0.00536929% main:@1577 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](NLt5JnRMjG:0) -> float_type, {32, 2}, {5, 1}: 0.00103798ms, 0.00915202% main:@1578 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](KCNfS4GlpP:0) -> float_type, {32, 4}, {9, 1}: 0.00061542ms, 0.00542625% main:@1579 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](O5aJeUQh2G:0) -> float_type, {32, 8}, {17, 1}: 0.00060032ms, 0.00529311% main:@1580 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](nltbg9ZWQX:0) -> float_type, {32, 4}, {17, 1}: 0.00060186ms, 0.00530669% main:@1581 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](U6GqhP8k7I:0) -> float_type, {32, 4}, {9, 1}: 0.00099372ms, 0.00876177% main:@1582 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](IxpbBg7JTO:0) -> float_type, {32, 2}, {5, 1}: 0.00060132ms, 0.00530193% main:@1583 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](5yColzZ8K3:0) -> float_type, {32, 2}, {5, 1}: 0.00058586ms, 0.00516561% main:@1584 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](6Wh2Jm7GHC:0) -> float_type, {32, 2}, {5, 1}: 0.00061724ms, 0.00544229% main:@1585 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](UMO0o14RuI:0) -> float_type, {32, 4}, {9, 1}: 0.00102582ms, 0.0090448% main:@1586 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](Pu1O5qFxY6:0) -> float_type, {32, 4}, {17, 1}: 0.00059684ms, 0.00526242% main:@1587 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](IMdFk7D16X:0) -> float_type, {32, 4}, {9, 1}: 0.00058858ms, 0.00518959% main:@1588 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](Cw59eOo7M1:0) -> float_type, {32, 4}, {9, 1}: 0.0005764ms, 0.0050822% main:@1589 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](E1ZkVIXqS5:0) -> float_type, {32, 4}, {9, 1}: 0.00102226ms, 0.00901341% main:@1590 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](vAbKepTjQg:0) -> float_type, {32, 2}, {5, 1}: 0.00065218ms, 0.00575037% main:@1591 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](lDhFNYzbxR:0) -> float_type, {32, 8}, {17, 1}: 0.0006856ms, 0.00604503% main:@1592 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](ecPTQUALkq:0) -> float_type, {32, 2}, {5, 1}: 0.00065632ms, 0.00578687% main:@1593 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](magwfJSnUu:0) -> float_type, {32, 4}, {9, 1}: 0.00095486ms, 0.00841914% main:@1594 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](mNgQ4ohWqU:0) -> float_type, {32, 4}, {9, 1}: 0.00063856ms, 0.00563028% main:@1595 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](gsP96ZTbn0:0) -> float_type, {32, 8}, {17, 1}: 0.0006564ms, 0.00578757% main:@1596 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](gqk5byC8aJ:0) -> float_type, {32, 2}, {5, 1}: 0.00063292ms, 0.00558055% main:@1597 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](mTMtDEo3r8:0) -> float_type, {32, 4}, {9, 1}: 0.0009598ms, 0.00846269% main:@1598 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](xVErSbJ0hg:0) -> float_type, {32, 4}, {9, 1}: 0.00064828ms, 0.00571598% main:@1599 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](qfwYOHde3u:0) -> float_type, {32, 8}, {17, 1}: 0.00063622ms, 0.00560964% main:@1600 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](pQsi1GNxvj:0) -> float_type, {32, 2}, {5, 1}: 0.00059962ms, 0.00528694% main:@1601 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](1QWFmJ96aB:0) -> float_type, {32, 2}, {5, 1}: 0.00093578ms, 0.00825091% main:@1602 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](3VED7WSJ1b:0) -> float_type, {32, 2}, {5, 1}: 0.00063024ms, 0.00555692% main:@1603 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](gdck97AK4w:0) -> float_type, {32, 4}, {17, 1}: 0.00063914ms, 0.00563539% main:@1604 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](tIGjbcEuDd:0) -> float_type, {32, 4}, {9, 1}: 0.00060856ms, 0.00536576% main:@1605 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](NJBt1ubXFW:0) -> float_type, {32, 2}, {5, 1}: 0.00092608ms, 0.00816538% main:@1606 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](zX6xKOWG9S:0) -> float_type, {32, 2}, {5, 1}: 0.00063794ms, 0.00562481% main:@1607 = load[offset=3776256,end=3798016](main:@1) -> float_type, {32, 170}, {170, 1}: 0.00045812ms, 0.00403931% main:@1608 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](Uc42tiuavR:0) -> float_type, {32, 4}, {9, 1}: 0.0006514ms, 0.00574349% main:@1609 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](l4bReWtsMB:0) -> float_type, {32, 8}, {25, 1}: 0.00062866ms, 0.00554299% main:@1610 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](yVxKAL4U5G:0) -> float_type, {32, 4}, {9, 1}: 0.00063256ms, 0.00557737% main:@1611 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](ZRYNFfjSez:0) -> float_type, {32, 8}, {33, 1}: 0.00064028ms, 0.00564544% main:@1612 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](oQ5OcjZJe8:0) -> float_type, {32, 8}, {17, 1}: 0.00064516ms, 0.00568847% main:@1613 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=132,local=256,](main:@1584,main:@1592,main:@1605,main:@1590,main:@1606,main:@1600,main:@1577,main:@1602,main:@1601,main:@1576,main:@1596,main:@1582,main:@1583,main:@1571,main:@1581,main:@1589,main:@1588,main:@1578,main:@1608,main:@1587,main:@1594,main:@1597,main:@1610,main:@1603,main:@1586,main:@1604,main:@1598,main:@1585,main:@1574,main:@1580,main:@1575,main:@1593,main:@1573,main:@1599,main:@1591,main:@1595,main:@1579,main:@1611,main:@1609,main:@1572,main:@1612,main:@1607) -> float_type, {32, 170}, {170, 1}: 0.0444536ms, 0.391954% main:@1614 = hip::hip_copy_literal[id=main:@literal:73] -> float_type, {170, 160}, {160, 1}: 0.00085602ms, 0.00754765% main:@1615 = load[offset=3755776,end=3776256](main:@1) -> float_type, {32, 160}, {160, 1}: 0.00047904ms, 0.00422376% main:@1616 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1613,main:@1614,main:@1615) -> float_type, {32, 160}, {160, 1}: 0.0221509ms, 0.195308% main:@1617 = slice[axes={0, 1},starts={0, 25},ends={32, 41}](LKAvFme9ZS:0) -> float_type, {32, 16}, {49, 1}: 0.00079694ms, 0.00702673% main:@1618 = unsqueeze[axes={1},steps={}](main:@1617) -> float_type, {32, 1, 16}, {49, 16, 1}: 0.00079312ms, 0.00699305% main:@1619 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](WDLoyKsiA7:0) -> float_type, {32, 16}, {33, 1}: 0.00079316ms, 0.00699341% main:@1620 = unsqueeze[axes={1},steps={}](main:@1619) -> float_type, {32, 1, 16}, {33, 16, 1}: 0.0006001ms, 0.00529117% main:@1621 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](z3lqgZM8WU:0) -> float_type, {32, 16}, {41, 1}: 0.00060394ms, 0.00532503% main:@1622 = unsqueeze[axes={1},steps={}](main:@1621) -> float_type, {32, 1, 16}, {41, 16, 1}: 0.00072178ms, 0.00636404% main:@1623 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](pY0t1rI5WU:0) -> float_type, {32, 16}, {41, 1}: 0.00075932ms, 0.00669503% main:@1624 = unsqueeze[axes={1},steps={}](main:@1623) -> float_type, {32, 1, 16}, {41, 16, 1}: 0.00057784ms, 0.0050949% main:@1625 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](sZdFJI9nwg:0) -> float_type, {32, 16}, {41, 1}: 0.00059258ms, 0.00522486% main:@1626 = unsqueeze[axes={1},steps={}](main:@1625) -> float_type, {32, 1, 16}, {41, 16, 1}: 0.00071534ms, 0.00630726% main:@1627 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](VfFM90O7LN:0) -> float_type, {32, 16}, {41, 1}: 0.00073628ms, 0.00649189% main:@1628 = unsqueeze[axes={1},steps={}](main:@1627) -> float_type, {32, 1, 16}, {41, 16, 1}: 0.00056382ms, 0.00497128% main:@1629 = reshape_lazy[dims={-1, 10, 16}](main:@1616) -> float_type, {32, 10, 16}, {160, 16, 1}: 0.00050546ms, 0.00445671% main:@1630 = load[offset=3776256,end=3813120](main:@1) -> float_type, {32, 18, 16}, {288, 16, 1}: 0.00050336ms, 0.0044382% main:@1631 = slice[axes={0, 1},starts={0, 25},ends={32, 41}](TxEdCcekGS:0) -> float_type, {32, 16}, {49, 1}: 0.00063748ms, 0.00562075% main:@1632 = unsqueeze[axes={1},steps={}](main:@1631) -> float_type, {32, 1, 16}, {49, 16, 1}: 0.00056126ms, 0.00494871% main:@1633 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](2Yy3zxNGBC:0) -> float_type, {32, 16}, {45, 1}: 0.00073666ms, 0.00649524% main:@1634 = unsqueeze[axes={1},steps={}](main:@1633) -> float_type, {32, 1, 16}, {45, 16, 1}: 0.00056822ms, 0.00501008% main:@1635 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=1024,local=256,](main:@1620,main:@1622,main:@1618,main:@1624,main:@1632,main:@1626,main:@1628,main:@1634,main:@1629,main:@1630) -> float_type, {32, 18, 16}, {288, 16, 1}: 0.0245849ms, 0.216769% main:@1636 = hip::hip_copy_literal[id=main:@literal:133] -> float_type, {18, 4}, {4, 1}: 0.00080576ms, 0.0071045% main:@1637 = transpose[permutation={0, 2, 1}](main:@1635) -> float_type, {32, 16, 18}, {288, 1, 16}: 0.00058564ms, 0.00516367% main:@1638 = load[offset=5010688,end=5047552](main:@1) -> float_type, {32, 16, 18}, {288, 18, 1}: 0.00051044ms, 0.00450062% main:@1639 = gpu::code_object[code_object=13528,symbol_name=contiguous_kernel,global=9216,local=256,](main:@1637,main:@1638) -> float_type, {32, 16, 18}, {288, 18, 1}: 0.0145607ms, 0.128384% main:@1640 = reshape_lazy[dims={512, 18}](main:@1639) -> float_type, {512, 18}, {18, 1}: 0.0006829ms, 0.00602123% main:@1641 = load[offset=3759872,end=3768064](main:@1) -> float_type, {512, 4}, {4, 1}: 0.00048566ms, 0.00428213% main:@1642 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1640,main:@1636,main:@1641) -> float_type, {512, 4}, {4, 1}: 0.0195067ms, 0.171993% main:@1643 = slice[axes={1},starts={96},ends={128}](main:@643) -> float_type, {32, 32}, {128, 1}: 0.00079474ms, 0.00700734% main:@1644 = load[offset=3768064,end=3772160](main:@1) -> float_type, {32, 32}, {32, 1}: 0.000474ms, 0.00417933% main:@1645 = gpu::code_object[code_object=13528,symbol_name=contiguous_kernel,global=512,local=256,](main:@1643,main:@1644) -> float_type, {32, 32}, {32, 1}: 0.0129534ms, 0.114212% main:@1646 = load[offset=3755776,end=3757696](main:@1) -> float_type, {1, 32, 1, 15}, {480, 15, 15, 1}: 0.00063752ms, 0.00562111% main:@1647 = reshape_lazy[dims={1, 32, 1, 32}](main:@1645) -> float_type, {1, 32, 1, 32}, {1024, 32, 32, 1}: 0.00054032ms, 0.00476408% main:@1648 = slice[axes={1},starts={0},ends={32}](main:@553) -> float_type, {480, 32}, {64, 1}: 0.00061522ms, 0.00542448% main:@1649 = reshape_lazy[dims={1, 32, 15, 32}](main:@1648) -> float_type, {1, 32, 15, 32}, {30720, 960, 64, 1}: 0.00057306ms, 0.00505275% main:@1650 = transpose[permutation={0, 1, 3, 2}](main:@1649) -> float_type, {1, 32, 32, 15}, {30720, 960, 1, 64}: 0.00057494ms, 0.00506933% main:@1651 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1647,main:@1650,main:@1646) -> float_type, {1, 32, 1, 15}, {480, 15, 15, 1}: 0.0213885ms, 0.188586% hzqb8Cpfc1:0 = @param:hzqb8Cpfc1:0 -> int32_type, {32, 15}, {15, 1}: 0.0007261ms, 0.00640213% main:@1653 = unsqueeze[axes={1},steps={}](hzqb8Cpfc1:0) -> int32_type, {32, 1, 15}, {15, 15, 1}: 0.00057568ms, 0.00507585% main:@1654 = load[offset=3757696,end=3759616](main:@1) -> float_type, {32, 1, 15}, {15, 15, 1}: 0.00050436ms, 0.00444701% main:@1655 = gpu::code_object[code_object=13536,symbol_name=convert_mul_add_kernel,global=240,local=256,](main:@1653,main:@1654) -> float_type, {32, 1, 15}, {15, 15, 1}: 0.0132383ms, 0.116724% main:@1656 = multibroadcast[out_lens={1, 32, 1, 15}](main:@1655) -> float_type, {1, 32, 1, 15}, {0, 15, 15, 1}: 0.00070606ms, 0.00622543% main:@1657 = load[offset=3769984,end=3771904](main:@1) -> float_type, {1, 32, 1, 15}, {480, 15, 15, 1}: 0.00051648ms, 0.00455388% main:@1658 = gpu::code_object[code_object=13520,symbol_name=add_sub_kernel,global=240,local=256,](main:@1651,main:@1656,main:@1657) -> float_type, {1, 32, 1, 15}, {480, 15, 15, 1}: 0.0131557ms, 0.115996% main:@1659 = load[offset=3768064,end=3769984](main:@1) -> float_type, {1, 32, 1, 15}, {480, 15, 15, 1}: 0.00063962ms, 0.00563962% main:@1660 = gpu::code_object[code_object=14568,symbol_name=softmax_kernel,global=2048,local=64,](main:@1658,main:@1659) -> float_type, {1, 32, 1, 15}, {480, 15, 15, 1}: 0.0145007ms, 0.127855% main:@1661 = slice[axes={1},starts={32},ends={64}](main:@553) -> float_type, {480, 32}, {64, 1}: 0.00077368ms, 0.00682165% main:@1662 = load[offset=3755776,end=3759872](main:@1) -> float_type, {1, 32, 1, 32}, {1024, 32, 32, 1}: 0.00050812ms, 0.00448017% main:@1663 = reshape_lazy[dims={1, 32, 15, 32}](main:@1661) -> float_type, {1, 32, 15, 32}, {30720, 960, 64, 1}: 0.00091748ms, 0.00808955% main:@1664 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1660,main:@1663,main:@1662) -> float_type, {1, 32, 1, 32}, {1024, 32, 32, 1}: 0.0186955ms, 0.164841% main:@1665 = slice[axes={1},starts={0},ends={128}](main:@661) -> float_type, {32, 128}, {256, 1}: 0.00073062ms, 0.00644198% main:@1666 = reshape_lazy[dims={32, 1, 128}](main:@1665) -> float_type, {32, 1, 128}, {256, 128, 1}: 0.0005806ms, 0.00511923% main:@1667 = slice[axes={2},starts={0},ends={64}](main:@1666) -> float_type, {32, 1, 64}, {256, 128, 1}: 0.00112002ms, 0.00987538% main:@1668 = unsqueeze[axes={0},steps={}](main:@1667) -> float_type, {1, 32, 1, 64}, {8192, 256, 128, 1}: 0.00067956ms, 0.00599178% main:@1669 = slice[axes={2},starts={64},ends={128}](main:@1666) -> float_type, {32, 1, 64}, {256, 128, 1}: 0.00070818ms, 0.00624413% main:@1670 = unsqueeze[axes={0},steps={}](main:@1669) -> float_type, {1, 32, 1, 64}, {8192, 256, 128, 1}: 0.00063188ms, 0.00557138% main:@1671 = load[offset=1371264,end=1387648](main:@1) -> float_type, {2, 32, 1, 64}, {2048, 64, 64, 1}: 0.0004918ms, 0.00433627% main:@1672 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=1024,local=256,](main:@1668,main:@1670,main:@1671) -> float_type, {2, 32, 1, 64}, {2048, 64, 64, 1}: 0.0136748ms, 0.120573% main:@1673 = slice[axes={1},starts={128},ends={256}](main:@609) -> float_type, {1600, 128}, {256, 1}: 0.00076712ms, 0.00676381% main:@1674 = reshape_lazy[dims={32, 50, 128}](main:@1673) -> float_type, {32, 50, 128}, {12800, 256, 1}: 0.00061284ms, 0.0054035% main:@1675 = slice[axes={2},starts={0},ends={64}](main:@1674) -> float_type, {32, 50, 64}, {12800, 256, 1}: 0.00071044ms, 0.00626405% main:@1676 = unsqueeze[axes={0},steps={}](main:@1675) -> float_type, {1, 32, 50, 64}, {409600, 12800, 256, 1}: 0.000649ms, 0.00572233% main:@1677 = slice[axes={2},starts={64},ends={128}](main:@1674) -> float_type, {32, 50, 64}, {12800, 256, 1}: 0.00100578ms, 0.00886811% main:@1678 = unsqueeze[axes={0},steps={}](main:@1677) -> float_type, {1, 32, 50, 64}, {409600, 12800, 256, 1}: 0.00058522ms, 0.00515997% main:@1679 = load[offset=5010688,end=5829888](main:@1) -> float_type, {2, 32, 50, 64}, {102400, 3200, 64, 1}: 0.0004878ms, 0.004301% main:@1680 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=51200,local=256,](main:@1676,main:@1678,main:@1679) -> float_type, {2, 32, 50, 64}, {102400, 3200, 64, 1}: 0.0190292ms, 0.167783% main:@1681 = transpose[permutation={0, 1, 3, 2}](main:@1680) -> float_type, {2, 32, 64, 50}, {102400, 3200, 1, 64}: 0.00070672ms, 0.00623125% main:@1682 = load[offset=1387648,end=1400448](main:@1) -> float_type, {2, 32, 1, 50}, {1600, 50, 50, 1}: 0.00049652ms, 0.00437789% main:@1683 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1672,main:@1681,main:@1682) -> float_type, {2, 32, 1, 50}, {1600, 50, 50, 1}: 0.0219059ms, 0.193147% main:@1684 = unsqueeze[axes={1},steps={}](vOI1hWzBDK:0) -> int32_type, {32, 1, 50}, {50, 50, 1}: 0.00074364ms, 0.00655678% main:@1685 = load[offset=1400448,end=1406848](main:@1) -> float_type, {32, 1, 50}, {50, 50, 1}: 0.0005227ms, 0.00460872% main:@1686 = gpu::code_object[code_object=13528,symbol_name=convert_mul_kernel,global=800,local=256,](main:@1684,main:@1685) -> float_type, {32, 1, 50}, {50, 50, 1}: 0.0133155ms, 0.117405% main:@1687 = load[offset=1371264,end=1384064](main:@1) -> float_type, {2, 32, 1, 50}, {1600, 50, 50, 1}: 0.0006628ms, 0.005844% main:@1688 = multibroadcast[out_lens={2, 32, 1, 50}](main:@1686) -> float_type, {2, 32, 1, 50}, {0, 50, 50, 1}: 0.0005933ms, 0.00523121% main:@1689 = gpu::code_object[code_object=13520,symbol_name=add_mul_kernel,global=1600,local=256,](main:@1683,main:@1688,main:@1687) -> float_type, {2, 32, 1, 50}, {1600, 50, 50, 1}: 0.0132674ms, 0.11698% main:@1690 = load[offset=1391744,end=1404544](main:@1) -> float_type, {2, 32, 1, 50}, {1600, 50, 50, 1}: 0.00064594ms, 0.00569535% main:@1691 = gpu::code_object[code_object=14584,symbol_name=softmax_kernel,global=4096,local=64,](main:@1689,main:@1690) -> float_type, {2, 32, 1, 50}, {1600, 50, 50, 1}: 0.0158544ms, 0.139791% main:@1692 = slice[axes={1},starts={0},ends={128}](main:@609) -> float_type, {1600, 128}, {256, 1}: 0.00105614ms, 0.00931214% main:@1693 = reshape_lazy[dims={32, 50, 128}](main:@1692) -> float_type, {32, 50, 128}, {12800, 256, 1}: 0.00058412ms, 0.00515027% main:@1694 = slice[axes={2},starts={0},ends={64}](main:@1693) -> float_type, {32, 50, 64}, {12800, 256, 1}: 0.00103754ms, 0.00914814% main:@1695 = unsqueeze[axes={0},steps={}](main:@1694) -> float_type, {1, 32, 50, 64}, {409600, 12800, 256, 1}: 0.0006092ms, 0.0053714% main:@1696 = slice[axes={2},starts={64},ends={128}](main:@1693) -> float_type, {32, 50, 64}, {12800, 256, 1}: 0.00070952ms, 0.00625594% main:@1697 = unsqueeze[axes={0},steps={}](main:@1696) -> float_type, {1, 32, 50, 64}, {409600, 12800, 256, 1}: 0.00094214ms, 0.00830698% main:@1698 = load[offset=5010688,end=5829888](main:@1) -> float_type, {2, 32, 50, 64}, {102400, 3200, 64, 1}: 0.000535ms, 0.00471717% main:@1699 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=51200,local=256,](main:@1695,main:@1697,main:@1698) -> float_type, {2, 32, 50, 64}, {102400, 3200, 64, 1}: 0.0191008ms, 0.168414% main:@1700 = load[offset=1375360,end=1391744](main:@1) -> float_type, {2, 32, 1, 64}, {2048, 64, 64, 1}: 0.00063022ms, 0.00555674% main:@1701 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1691,main:@1699,main:@1700) -> float_type, {2, 32, 1, 64}, {2048, 64, 64, 1}: 0.0195419ms, 0.172303% main:@1702 = slice[axes={1},starts={32},ends={64}](main:@525) -> float_type, {960, 32}, {64, 1}: 0.0007756ms, 0.00683858% main:@1703 = reshape_lazy[dims={1, 32, 30, 32}](main:@1702) -> float_type, {1, 32, 30, 32}, {61440, 1920, 64, 1}: 0.00095702ms, 0.00843818% main:@1704 = transpose[permutation={0, 1, 3, 2}](main:@1703) -> float_type, {1, 32, 32, 30}, {61440, 1920, 1, 64}: 0.00059578ms, 0.00525308% main:@1705 = reshape_lazy[dims={1, 32, 1, 32}](main:@1268) -> float_type, {1, 32, 1, 32}, {1024, 32, 32, 1}: 0.0005102ms, 0.00449851% main:@1706 = load[offset=1371264,end=1375104](main:@1) -> float_type, {1, 32, 1, 30}, {960, 30, 30, 1}: 0.00050722ms, 0.00447223% main:@1707 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1705,main:@1704,main:@1706) -> float_type, {1, 32, 1, 30}, {960, 30, 30, 1}: 0.0187362ms, 0.1652% LjbB2i31Wu:0 = @param:LjbB2i31Wu:0 -> int32_type, {32, 30}, {30, 1}: 0.00071074ms, 0.0062667% main:@1709 = unsqueeze[axes={1},steps={}](LjbB2i31Wu:0) -> int32_type, {32, 1, 30}, {30, 30, 1}: 0.0005655ms, 0.00498609% main:@1710 = load[offset=1391744,end=1395584](main:@1) -> float_type, {32, 1, 30}, {30, 30, 1}: 0.00047952ms, 0.004228% main:@1711 = gpu::code_object[code_object=13528,symbol_name=convert_add_kernel,global=480,local=256,](main:@1709,main:@1710) -> float_type, {32, 1, 30}, {30, 30, 1}: 0.0132831ms, 0.117119% main:@1712 = load[offset=1395584,end=1399424](main:@1) -> float_type, {1, 32, 1, 30}, {960, 30, 30, 1}: 0.00067154ms, 0.00592106% main:@1713 = multibroadcast[out_lens={1, 32, 1, 30}](main:@1711) -> float_type, {1, 32, 1, 30}, {0, 30, 30, 1}: 0.00056358ms, 0.00496917% main:@1714 = gpu::code_object[code_object=13520,symbol_name=pow_pow_kernel,global=480,local=256,](main:@1707,main:@1713,main:@1712) -> float_type, {1, 32, 1, 30}, {960, 30, 30, 1}: 0.0181483ms, 0.160016% main:@1715 = load[offset=1391744,end=1395584](main:@1) -> float_type, {1, 32, 1, 30}, {960, 30, 30, 1}: 0.00061844ms, 0.00545287% main:@1716 = gpu::code_object[code_object=14584,symbol_name=softmax_kernel,global=2048,local=64,](main:@1714,main:@1715) -> float_type, {1, 32, 1, 30}, {960, 30, 30, 1}: 0.0147673ms, 0.130205% main:@1717 = slice[axes={1},starts={0},ends={32}](main:@525) -> float_type, {960, 32}, {64, 1}: 0.00074748ms, 0.00659064% main:@1718 = reshape_lazy[dims={1, 32, 30, 32}](main:@1717) -> float_type, {1, 32, 30, 32}, {61440, 1920, 64, 1}: 0.00058148ms, 0.00512699% main:@1719 = load[offset=1371264,end=1375360](main:@1) -> float_type, {1, 32, 1, 32}, {1024, 32, 32, 1}: 0.00049978ms, 0.00440663% main:@1720 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1716,main:@1718,main:@1719) -> float_type, {1, 32, 1, 32}, {1024, 32, 32, 1}: 0.0179101ms, 0.157916% main:@1721 = slice[axes={1},starts={0},ends={32}](main:@643) -> float_type, {32, 32}, {128, 1}: 0.00072858ms, 0.00642399% main:@1722 = load[offset=529408,end=533504](main:@1) -> float_type, {32, 32}, {32, 1}: 0.00049924ms, 0.00440187% main:@1723 = gpu::code_object[code_object=13528,symbol_name=contiguous_kernel,global=512,local=256,](main:@1721,main:@1722) -> float_type, {32, 32}, {32, 1}: 0.013141ms, 0.115866% main:@1724 = slice[axes={1},starts={0},ends={32}](main:@792) -> float_type, {480, 32}, {64, 1}: 0.00074548ms, 0.006573% main:@1725 = reshape_lazy[dims={1, 32, 15, 32}](main:@1724) -> float_type, {1, 32, 15, 32}, {30720, 960, 64, 1}: 0.00059072ms, 0.00520846% main:@1726 = transpose[permutation={0, 1, 3, 2}](main:@1725) -> float_type, {1, 32, 32, 15}, {30720, 960, 1, 64}: 0.00058654ms, 0.00517161% main:@1727 = reshape_lazy[dims={1, 32, 1, 32}](main:@1723) -> float_type, {1, 32, 1, 32}, {1024, 32, 32, 1}: 0.00050192ms, 0.0044255% main:@1728 = load[offset=527488,end=529408](main:@1) -> float_type, {1, 32, 1, 15}, {480, 15, 15, 1}: 0.00048254ms, 0.00425462% main:@1729 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1727,main:@1726,main:@1728) -> float_type, {1, 32, 1, 15}, {480, 15, 15, 1}: 0.0198652ms, 0.175155% UaC5MZd8qb:0 = @param:UaC5MZd8qb:0 -> int32_type, {32, 15}, {15, 1}: 0.0007072ms, 0.00623548% main:@1731 = unsqueeze[axes={1},steps={}](UaC5MZd8qb:0) -> int32_type, {32, 1, 15}, {15, 15, 1}: 0.00056816ms, 0.00500955% main:@1732 = load[offset=531328,end=533248](main:@1) -> float_type, {32, 1, 15}, {15, 15, 1}: 0.00049942ms, 0.00440346% main:@1733 = gpu::code_object[code_object=13536,symbol_name=convert_div_mul_kernel,global=240,local=256,](main:@1731,main:@1732) -> float_type, {32, 1, 15}, {15, 15, 1}: 0.0135083ms, 0.119105% main:@1734 = load[offset=529408,end=531328](main:@1) -> float_type, {1, 32, 1, 15}, {480, 15, 15, 1}: 0.00062594ms, 0.005519% main:@1735 = multibroadcast[out_lens={1, 32, 1, 15}](main:@1733) -> float_type, {1, 32, 1, 15}, {0, 15, 15, 1}: 0.00052804ms, 0.0046558% main:@1736 = gpu::code_object[code_object=13520,symbol_name=add_pow_kernel,global=240,local=256,](main:@1729,main:@1735,main:@1734) -> float_type, {1, 32, 1, 15}, {480, 15, 15, 1}: 0.0164328ms, 0.14489% main:@1737 = load[offset=531584,end=533504](main:@1) -> float_type, {1, 32, 1, 15}, {480, 15, 15, 1}: 0.00062314ms, 0.00549432% main:@1738 = gpu::code_object[code_object=14568,symbol_name=softmax_kernel,global=2048,local=64,](main:@1736,main:@1737) -> float_type, {1, 32, 1, 15}, {480, 15, 15, 1}: 0.0145038ms, 0.127882% main:@1739 = slice[axes={1},starts={32},ends={64}](main:@792) -> float_type, {480, 32}, {64, 1}: 0.00072478ms, 0.00639049% main:@1740 = load[offset=527488,end=531584](main:@1) -> float_type, {1, 32, 1, 32}, {1024, 32, 32, 1}: 0.00050362ms, 0.00444049% main:@1741 = reshape_lazy[dims={1, 32, 15, 32}](main:@1739) -> float_type, {1, 32, 15, 32}, {30720, 960, 64, 1}: 0.00059512ms, 0.00524726% main:@1742 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1738,main:@1741,main:@1740) -> float_type, {1, 32, 1, 32}, {1024, 32, 32, 1}: 0.0190139ms, 0.167648% main:@1743 = load[offset=531584,end=597120](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00064824ms, 0.00571563% main:@1744 = slice[axes={0, 1, 2},starts={0, 13, 0},ends={32, 14, 256}](main:@778) -> float_type, {32, 1, 256}, {4608, 256, 1}: 0.00062958ms, 0.0055511% main:@1745 = squeeze[axes={1}](main:@1744) -> float_type, {32, 256}, {4608, 1}: 0.0006713ms, 0.00591895% main:@1746 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1745,main:@759,main:@1743) -> float_type, {32, 512}, {512, 1}: 0.0246422ms, 0.217274% main:@1747 = hip::hip_copy_literal[id=main:@literal:41] -> float_type, {512, 256}, {256, 1}: 0.0008747ms, 0.00771236% main:@1748 = load[offset=597120,end=662656](main:@1) -> float_type, {32, 512}, {512, 1}: 0.0004783ms, 0.00421724% main:@1749 = gpu::code_object[code_object=13512,symbol_name=relu_kernel,global=8192,local=256,](main:@1746,main:@1748) -> float_type, {32, 512}, {512, 1}: 0.0144036ms, 0.126998% main:@1750 = load[offset=544384,end=577152](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00062588ms, 0.00551847% main:@1751 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1749,main:@1747,main:@1750) -> float_type, {32, 256}, {256, 1}: 0.0308233ms, 0.271773% main:@1752 = hip::hip_copy_literal[id=main:@literal:112] -> float_type, {256, 512}, {512, 1}: 0.00074334ms, 0.00655414% main:@1753 = load[offset=577152,end=642688](main:@1) -> float_type, {32, 512}, {512, 1}: 0.0005064ms, 0.004465% main:@1754 = slice[axes={0, 1, 2},starts={0, 7, 0},ends={32, 8, 256}](main:@778) -> float_type, {32, 1, 256}, {4608, 256, 1}: 0.00062774ms, 0.00553487% main:@1755 = squeeze[axes={1}](main:@1754) -> float_type, {32, 256}, {4608, 1}: 0.0009469ms, 0.00834895% main:@1756 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1755,main:@1752,main:@1753) -> float_type, {32, 512}, {512, 1}: 0.0234766ms, 0.206997% main:@1757 = hip::hip_copy_literal[id=main:@literal:140] -> float_type, {512, 256}, {256, 1}: 0.00075542ms, 0.00666065% main:@1758 = load[offset=642688,end=708224](main:@1) -> float_type, {32, 512}, {512, 1}: 0.0005017ms, 0.00442356% main:@1759 = gpu::code_object[code_object=13512,symbol_name=relu_kernel,global=8192,local=256,](main:@1756,main:@1758) -> float_type, {32, 512}, {512, 1}: 0.0143816ms, 0.126805% main:@1760 = load[offset=577152,end=609920](main:@1) -> float_type, {32, 256}, {256, 1}: 0.0006131ms, 0.00540579% main:@1761 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1759,main:@1757,main:@1760) -> float_type, {32, 256}, {256, 1}: 0.0303155ms, 0.267296% main:@1762 = hip::hip_copy_literal[id=main:@literal:153] -> float_type, {256, 512}, {512, 1}: 0.00079518ms, 0.00701122% main:@1763 = load[offset=609920,end=675456](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00048142ms, 0.00424475% main:@1764 = slice[axes={0, 1, 2},starts={0, 6, 0},ends={32, 7, 256}](main:@778) -> float_type, {32, 1, 256}, {4608, 256, 1}: 0.00060786ms, 0.00535959% main:@1765 = squeeze[axes={1}](main:@1764) -> float_type, {32, 256}, {4608, 1}: 0.00061178ms, 0.00539415% main:@1766 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1765,main:@1762,main:@1763) -> float_type, {32, 512}, {512, 1}: 0.0237743ms, 0.209622% main:@1767 = hip::hip_copy_literal[id=main:@literal:93] -> float_type, {512, 256}, {256, 1}: 0.00072382ms, 0.00638203% main:@1768 = load[offset=675456,end=740992](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00047882ms, 0.00422182% main:@1769 = gpu::code_object[code_object=13512,symbol_name=relu_kernel,global=8192,local=256,](main:@1766,main:@1768) -> float_type, {32, 512}, {512, 1}: 0.0143511ms, 0.126536% main:@1770 = load[offset=609920,end=642688](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00060896ms, 0.00536929% main:@1771 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1769,main:@1767,main:@1770) -> float_type, {32, 256}, {256, 1}: 0.0300791ms, 0.265211% main:@1772 = hip::hip_copy_literal[id=main:@literal:68] -> float_type, {256, 512}, {512, 1}: 0.00081358ms, 0.00717345% main:@1773 = load[offset=642688,end=708224](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00048084ms, 0.00423964% main:@1774 = slice[axes={0, 1, 2},starts={0, 11, 0},ends={32, 12, 256}](main:@778) -> float_type, {32, 1, 256}, {4608, 256, 1}: 0.00060424ms, 0.00532767% main:@1775 = squeeze[axes={1}](main:@1774) -> float_type, {32, 256}, {4608, 1}: 0.00091912ms, 0.00810401% main:@1776 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1775,main:@1772,main:@1773) -> float_type, {32, 512}, {512, 1}: 0.0232048ms, 0.2046% main:@1777 = load[offset=1391744,end=1457280](main:@1) -> float_type, {32, 512}, {512, 1}: 0.0006402ms, 0.00564474% main:@1778 = gpu::code_object[code_object=13512,symbol_name=relu_kernel,global=8192,local=256,](main:@1776,main:@1777) -> float_type, {32, 512}, {512, 1}: 0.0143275ms, 0.126327% main:@1779 = load[offset=642688,end=675456](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00061886ms, 0.00545658% main:@1780 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1778,main:@768,main:@1779) -> float_type, {32, 256}, {256, 1}: 0.0300272ms, 0.264754% main:@1781 = load[offset=675456,end=740992](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00063216ms, 0.00557385% main:@1782 = slice[axes={0, 1, 2},starts={0, 4, 0},ends={32, 5, 256}](main:@778) -> float_type, {32, 1, 256}, {4608, 256, 1}: 0.0005885ms, 0.00518889% main:@1783 = squeeze[axes={1}](main:@1782) -> float_type, {32, 256}, {4608, 1}: 0.00061108ms, 0.00538798% main:@1784 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1783,main:@759,main:@1781) -> float_type, {32, 512}, {512, 1}: 0.0231967ms, 0.204528% main:@1785 = hip::hip_copy_literal[id=main:@literal:94] -> float_type, {512, 256}, {256, 1}: 0.00074616ms, 0.006579% main:@1786 = load[offset=1391744,end=1457280](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00046936ms, 0.00413841% main:@1787 = gpu::code_object[code_object=13512,symbol_name=relu_kernel,global=8192,local=256,](main:@1784,main:@1786) -> float_type, {32, 512}, {512, 1}: 0.0143421ms, 0.126457% main:@1788 = load[offset=675456,end=708224](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00065766ms, 0.00579868% main:@1789 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1787,main:@1785,main:@1788) -> float_type, {32, 256}, {256, 1}: 0.0306519ms, 0.270263% main:@1790 = hip::hip_copy_literal[id=main:@literal:149] -> float_type, {256, 512}, {512, 1}: 0.00080046ms, 0.00705777% main:@1791 = load[offset=1391744,end=1457280](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00051062ms, 0.00450221% main:@1792 = slice[axes={0, 1, 2},starts={0, 3, 0},ends={32, 4, 256}](main:@778) -> float_type, {32, 1, 256}, {4608, 256, 1}: 0.00059492ms, 0.0052455% main:@1793 = squeeze[axes={1}](main:@1792) -> float_type, {32, 256}, {4608, 1}: 0.00065884ms, 0.00580909% main:@1794 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1793,main:@1790,main:@1791) -> float_type, {32, 512}, {512, 1}: 0.0238774ms, 0.210531% main:@1795 = load[offset=1457280,end=1522816](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00060968ms, 0.00537564% main:@1796 = gpu::code_object[code_object=13512,symbol_name=relu_kernel,global=8192,local=256,](main:@1794,main:@1795) -> float_type, {32, 512}, {512, 1}: 0.0143439ms, 0.126472% main:@1797 = load[offset=708224,end=740992](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00061646ms, 0.00543542% main:@1798 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1796,main:@1171,main:@1797) -> float_type, {32, 256}, {256, 1}: 0.029998ms, 0.264497% main:@1799 = hip::hip_copy_literal[id=main:@literal:106] -> float_type, {194, 192}, {192, 1}: 0.00078304ms, 0.00690418% main:@1800 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](HZ0l3MmcPY:0) -> float_type, {32, 4}, {17, 1}: 0.00072754ms, 0.00641483% main:@1801 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](IztrlRfWQU:0) -> float_type, {32, 2}, {13, 1}: 0.00076306ms, 0.00672801% main:@1802 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](KN13IWXZSu:0) -> float_type, {32, 2}, {13, 1}: 0.00081112ms, 0.00715176% main:@1803 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](RHo9Jz6KpI:0) -> float_type, {32, 4}, {21, 1}: 0.0006689ms, 0.00589779% main:@1804 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](NXOSdQVPs0:0) -> float_type, {32, 8}, {25, 1}: 0.00062846ms, 0.00554122% main:@1805 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](ry2xad1vc9:0) -> float_type, {32, 8}, {25, 1}: 0.0006198ms, 0.00546487% main:@1806 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](S5D6WGIXMd:0) -> float_type, {32, 4}, {17, 1}: 0.0007643ms, 0.00673894% main:@1807 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](5JHUPChra8:0) -> float_type, {32, 2}, {17, 1}: 0.00061624ms, 0.00543348% main:@1808 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](RpITWntAji:0) -> float_type, {32, 8}, {17, 1}: 0.00060732ms, 0.00535483% main:@1809 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](92MwiALCHX:0) -> float_type, {32, 8}, {25, 1}: 0.00062754ms, 0.00553311% main:@1810 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](JSWhLK4Agx:0) -> float_type, {32, 8}, {25, 1}: 0.00075206ms, 0.00663102% main:@1811 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](AVLcBpWeuo:0) -> float_type, {32, 4}, {21, 1}: 0.00064274ms, 0.00566713% main:@1812 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](o74vNAnXxy:0) -> float_type, {32, 4}, {17, 1}: 0.00063276ms, 0.00557914% main:@1813 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](Tg4u0GfnFb:0) -> float_type, {32, 8}, {25, 1}: 0.00060074ms, 0.00529681% main:@1814 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](TidKGrnobI:0) -> float_type, {32, 2}, {13, 1}: 0.00081236ms, 0.0071627% main:@1815 = slice[axes={0, 1},starts={0, 169},ends={32, 177}](uFSOc5Chp2:0) -> float_type, {32, 8}, {481, 1}: 0.00063096ms, 0.00556327% main:@1816 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](xymEBUMS25:0) -> float_type, {32, 2}, {13, 1}: 0.00060108ms, 0.00529981% main:@1817 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](msiPzSBKMj:0) -> float_type, {32, 4}, {17, 1}: 0.00061688ms, 0.00543912% main:@1818 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](xT9FhjM1Uu:0) -> float_type, {32, 4}, {17, 1}: 0.00068418ms, 0.00603251% main:@1819 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](m2YscaoDjP:0) -> float_type, {32, 4}, {17, 1}: 0.0005904ms, 0.00520564% main:@1820 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](tWcngAq7hV:0) -> float_type, {32, 2}, {13, 1}: 0.00066672ms, 0.00587857% main:@1821 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](zZHaVOckpw:0) -> float_type, {32, 2}, {13, 1}: 0.00068492ms, 0.00603904% main:@1822 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](ujOA5LQW7M:0) -> float_type, {32, 2}, {13, 1}: 0.00065382ms, 0.00576483% main:@1823 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](qd1komWTn4:0) -> float_type, {32, 8}, {25, 1}: 0.00065656ms, 0.00578898% main:@1824 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](nPUSZFm74Y:0) -> float_type, {32, 2}, {13, 1}: 0.00062934ms, 0.00554898% main:@1825 = slice[axes={0, 1},starts={0, 21},ends={32, 25}](ifP6Ceko5T:0) -> float_type, {32, 4}, {73, 1}: 0.00064686ms, 0.00570346% main:@1826 = slice[axes={0, 1},starts={0, 163},ends={32, 165}](nv2tzWRf1I:0) -> float_type, {32, 2}, {301, 1}: 0.00060574ms, 0.0053409% main:@1827 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](XxirNgB98u:0) -> float_type, {32, 2}, {13, 1}: 0.00065002ms, 0.00573132% main:@1828 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](1PJsBhDaCK:0) -> float_type, {32, 8}, {25, 1}: 0.00066156ms, 0.00583307% main:@1829 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](1WNvHElUG7:0) -> float_type, {32, 8}, {25, 1}: 0.00064108ms, 0.00565249% main:@1830 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](1u5yIsHTXE:0) -> float_type, {32, 4}, {17, 1}: 0.00064924ms, 0.00572444% main:@1831 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](2lVXMZgJfO:0) -> float_type, {32, 8}, {25, 1}: 0.0006469ms, 0.00570381% main:@1832 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](3kdi0vsSZG:0) -> float_type, {32, 8}, {25, 1}: 0.0006336ms, 0.00558654% main:@1833 = slice[axes={0, 1},starts={0, 25},ends={32, 33}](Qp3K1PHvte:0) -> float_type, {32, 8}, {69, 1}: 0.00070002ms, 0.00617218% main:@1834 = load[offset=1391744,end=1416576](main:@1) -> float_type, {32, 194}, {194, 1}: 0.00047606ms, 0.00419749% main:@1835 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](YbIVz8l1un:0) -> float_type, {32, 8}, {45, 1}: 0.00067934ms, 0.00598984% main:@1836 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](bJsIAzhYc3:0) -> float_type, {32, 8}, {37, 1}: 0.0006529ms, 0.00575671% main:@1837 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](pH5srbufJ1:0) -> float_type, {32, 4}, {17, 1}: 0.00062954ms, 0.00555074% main:@1838 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](09zp4EDXTh:0) -> float_type, {32, 8}, {25, 1}: 0.0009539ms, 0.00841067% main:@1839 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=163,local=256,](main:@1807,main:@1827,main:@1820,main:@1816,main:@1801,main:@1802,main:@1824,main:@1822,main:@1821,main:@1814,main:@1826,main:@1800,main:@1837,main:@1817,main:@1830,main:@1803,main:@1806,main:@1819,main:@1811,main:@1812,main:@1818,main:@1825,main:@1838,main:@1828,main:@1833,main:@1810,main:@1823,main:@1805,main:@1831,main:@1809,main:@1813,main:@1829,main:@1804,main:@1836,main:@1832,main:@1835,main:@1808,main:@1815,main:@1834) -> float_type, {32, 194}, {194, 1}: 0.0486989ms, 0.429385% main:@1840 = load[offset=740992,end=765568](main:@1) -> float_type, {32, 192}, {192, 1}: 0.00066828ms, 0.00589232% main:@1841 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1839,main:@1799,main:@1840) -> float_type, {32, 192}, {192, 1}: 0.0233728ms, 0.206082% main:@1842 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](Dy3rwVmn1A:0) -> float_type, {32, 16}, {41, 1}: 0.00083472ms, 0.00735985% main:@1843 = unsqueeze[axes={1},steps={}](main:@1842) -> float_type, {32, 1, 16}, {41, 16, 1}: 0.00071644ms, 0.00631695% main:@1844 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](FcmWLZhDgE:0) -> float_type, {32, 16}, {57, 1}: 0.0008267ms, 0.00728913% main:@1845 = unsqueeze[axes={1},steps={}](main:@1844) -> float_type, {32, 1, 16}, {57, 16, 1}: 0.0005876ms, 0.00518095% main:@1846 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](ahAlEtBvzo:0) -> float_type, {32, 16}, {41, 1}: 0.00062672ms, 0.00552588% main:@1847 = unsqueeze[axes={1},steps={}](main:@1846) -> float_type, {32, 1, 16}, {41, 16, 1}: 0.00055958ms, 0.0049339% main:@1848 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](dAFYcGrfEl:0) -> float_type, {32, 16}, {45, 1}: 0.00075872ms, 0.00668974% main:@1849 = unsqueeze[axes={1},steps={}](main:@1848) -> float_type, {32, 1, 16}, {45, 16, 1}: 0.00058694ms, 0.00517513% main:@1850 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](X29iyIeu8L:0) -> float_type, {32, 16}, {45, 1}: 0.00065142ms, 0.00574366% main:@1851 = unsqueeze[axes={1},steps={}](main:@1850) -> float_type, {32, 1, 16}, {45, 16, 1}: 0.00056854ms, 0.0050129% main:@1852 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](ZcWERoz0h6:0) -> float_type, {32, 16}, {49, 1}: 0.00075362ms, 0.00664478% main:@1853 = unsqueeze[axes={1},steps={}](main:@1852) -> float_type, {32, 1, 16}, {49, 16, 1}: 0.00056928ms, 0.00501942% main:@1854 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](pYw7fMVoQg:0) -> float_type, {32, 16}, {41, 1}: 0.00064622ms, 0.00569781% main:@1855 = unsqueeze[axes={1},steps={}](main:@1854) -> float_type, {32, 1, 16}, {41, 16, 1}: 0.00055624ms, 0.00490445% main:@1856 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](mgnGNH4Mz3:0) -> float_type, {32, 16}, {49, 1}: 0.0007618ms, 0.0067169% main:@1857 = unsqueeze[axes={1},steps={}](main:@1856) -> float_type, {32, 1, 16}, {49, 16, 1}: 0.0005567ms, 0.0049085% main:@1858 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](6HSQqXdJzb:0) -> float_type, {32, 16}, {45, 1}: 0.000623ms, 0.00549308% main:@1859 = unsqueeze[axes={1},steps={}](main:@1858) -> float_type, {32, 1, 16}, {45, 16, 1}: 0.00055846ms, 0.00492402% main:@1860 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](isCn4mEOSt:0) -> float_type, {32, 16}, {49, 1}: 0.00078892ms, 0.00695602% main:@1861 = unsqueeze[axes={1},steps={}](main:@1860) -> float_type, {32, 1, 16}, {49, 16, 1}: 0.00056606ms, 0.00499103% main:@1862 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](2CnSyUBzm0:0) -> float_type, {32, 16}, {65, 1}: 0.00064372ms, 0.00567577% main:@1863 = unsqueeze[axes={1},steps={}](main:@1862) -> float_type, {32, 1, 16}, {65, 16, 1}: 0.0009227ms, 0.00813558% main:@1864 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](zIoqYe9cRb:0) -> float_type, {32, 16}, {45, 1}: 0.00075758ms, 0.00667969% main:@1865 = unsqueeze[axes={1},steps={}](main:@1864) -> float_type, {32, 1, 16}, {45, 16, 1}: 0.00056934ms, 0.00501995% main:@1866 = load[offset=1391744,end=1442944](main:@1) -> float_type, {32, 25, 16}, {400, 16, 1}: 0.00049658ms, 0.00437842% main:@1867 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](1aXQU27NrD:0) -> float_type, {32, 16}, {49, 1}: 0.00064186ms, 0.00565937% main:@1868 = unsqueeze[axes={1},steps={}](main:@1867) -> float_type, {32, 1, 16}, {49, 16, 1}: 0.00058682ms, 0.00517408% main:@1869 = reshape_lazy[dims={-1, 12, 16}](main:@1841) -> float_type, {32, 12, 16}, {192, 16, 1}: 0.00053462ms, 0.00471382% main:@1870 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=914,local=256,](main:@1863,main:@1857,main:@1868,main:@1859,main:@1865,main:@1849,main:@1851,main:@1853,main:@1847,main:@1845,main:@1855,main:@1843,main:@1861,main:@1869,main:@1866) -> float_type, {32, 25, 16}, {400, 16, 1}: 0.0284046ms, 0.250447% main:@1871 = reshape_lazy[dims={32, 16, 4}](main:@1642) -> float_type, {32, 16, 4}, {64, 4, 1}: 0.00069764ms, 0.00615119% main:@1872 = load[offset=531584,end=544384](main:@1) -> float_type, {32, 25, 4}, {100, 4, 1}: 0.0005203ms, 0.00458756% main:@1873 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1870,main:@1871,main:@1872) -> float_type, {32, 25, 4}, {100, 4, 1}: 0.0197984ms, 0.174565% main:@1874 = transpose[permutation={1, 2, 0, 3}](main:@1121) -> float_type, {32, 15, 4, 64}, {960, 64, 30720, 1}: 0.0007528ms, 0.00663755% main:@1875 = load[offset=1391744,end=1424512](main:@1) -> float_type, {32, 1, 4, 64}, {64, 64, 2048, 1}: 0.00050302ms, 0.0044352% main:@1876 = gpu::code_object[code_object=13528,symbol_name=reduce_prod_kernel,global=8192,local=256,](main:@1874,main:@1875) -> float_type, {32, 1, 4, 64}, {64, 64, 2048, 1}: 0.0145614ms, 0.12839% main:@1877 = unsqueeze[axes={1},steps={}](main:@1156) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.00073004ms, 0.00643687% main:@1878 = unsqueeze[axes={1},steps={}](main:@1175) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.00050494ms, 0.00445213% main:@1879 = unsqueeze[axes={1},steps={}](main:@1165) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.000494ms, 0.00435567% main:@1880 = unsqueeze[axes={1},steps={}](main:@1185) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.00047956ms, 0.00422835% main:@1881 = unsqueeze[axes={1},steps={}](main:@1195) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.00048534ms, 0.00427931% main:@1882 = unsqueeze[axes={1},steps={}](main:@1204) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.00050564ms, 0.0044583% main:@1883 = unsqueeze[axes={1},steps={}](main:@1213) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.00048782ms, 0.00430118% main:@1884 = unsqueeze[axes={1},steps={}](main:@1222) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.00049284ms, 0.00434544% main:@1885 = unsqueeze[axes={1},steps={}](main:@1231) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.00049246ms, 0.00434209% main:@1886 = unsqueeze[axes={1},steps={}](main:@1250) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.00047338ms, 0.00417386% main:@1887 = unsqueeze[axes={1},steps={}](main:@1241) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.00048304ms, 0.00425903% main:@1888 = unsqueeze[axes={1},steps={}](main:@1259) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.0004919ms, 0.00433715% main:@1889 = unsqueeze[axes={1},steps={}](main:@1751) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.0004862ms, 0.0042869% main:@1890 = load[offset=1424512,end=2014336](main:@1) -> float_type, {32, 18, 256}, {4608, 256, 1}: 0.00048564ms, 0.00428196% main:@1891 = unsqueeze[axes={1},steps={}](main:@1761) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.00049478ms, 0.00436255% main:@1892 = unsqueeze[axes={1},steps={}](main:@1771) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.00047942ms, 0.00422712% main:@1893 = unsqueeze[axes={1},steps={}](main:@1780) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.00047112ms, 0.00415393% main:@1894 = unsqueeze[axes={1},steps={}](main:@1789) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.00047128ms, 0.00415534% main:@1895 = unsqueeze[axes={1},steps={}](main:@1798) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.0004709ms, 0.00415199% main:@1896 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=4096,local=256,](main:@1884,main:@1878,main:@1883,main:@1895,main:@1894,main:@1882,main:@1892,main:@1891,main:@1881,main:@1877,main:@1888,main:@1893,main:@1887,main:@1889,main:@1885,main:@1880,main:@1879,main:@1886,main:@1890) -> float_type, {32, 18, 256}, {4608, 256, 1}: 0.0273465ms, 0.241118% main:@1897 = hip::hip_copy_literal[id=main:@literal:158] -> float_type, {1}, {0}: 0.0008479ms, 0.00747606% main:@1898 = multibroadcast[out_lens={32, 18, 1}](main:@1897) -> float_type, {32, 18, 1}, {0, 0, 0}: 0.00065456ms, 0.00577135% main:@1899 = multibroadcast[out_lens={32, 18, 256}](main:@1898) -> float_type, {32, 18, 256}, {0, 0, 0}: 0.00058742ms, 0.00517937% main:@1900 = load[offset=2014336,end=2604160](main:@1) -> float_type, {32, 18, 256}, {4608, 256, 1}: 0.00052602ms, 0.00463799% main:@1901 = gpu::code_object[code_object=17656,symbol_name=pow_reduce_sum_div_pow_add_kernel,global=73728,local=128,](main:@1896,main:@778,main:@1899,main:@776,main:@1900) -> float_type, {32, 18, 256}, {4608, 256, 1}: 0.0245347ms, 0.216326% main:@1902 = load[offset=544384,end=577152](main:@1) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.00064552ms, 0.00569164% main:@1903 = gpu::code_object[code_object=13528,symbol_name=reduce_sum_kernel,global=8192,local=256,](main:@1901,main:@1902) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.0145063ms, 0.127904% main:@1904 = transpose[permutation={1, 2, 0, 3}](main:@1701) -> float_type, {32, 1, 2, 64}, {64, 64, 2048, 1}: 0.00069814ms, 0.0061556% main:@1905 = load[offset=577152,end=593536](main:@1) -> float_type, {32, 1, 2, 64}, {128, 128, 64, 1}: 0.0005059ms, 0.00446059% main:@1906 = gpu::code_object[code_object=13528,symbol_name=contiguous_kernel,global=2048,local=256,](main:@1904,main:@1905) -> float_type, {32, 1, 2, 64}, {128, 128, 64, 1}: 0.0130811ms, 0.115338% main:@1907 = load[offset=593536,end=606848](main:@1) -> float_type, {32, 1, 104}, {104, 104, 1}: 0.0006252ms, 0.00551248% main:@1908 = gpu::code_object[code_object=13528,symbol_name=reduce_mean_kernel,global=3328,local=256,](main:@521,main:@1907) -> float_type, {32, 1, 104}, {104, 104, 1}: 0.0147549ms, 0.130096% main:@1909 = load[offset=114688,end=123392](main:@1) -> float_type, {32, 1, 68}, {68, 68, 1}: 0.00064334ms, 0.00567242% main:@1910 = gpu::code_object[code_object=13528,symbol_name=reduce_prod_kernel,global=2176,local=256,](main:@1547,main:@1909) -> float_type, {32, 1, 68}, {68, 68, 1}: 0.0146437ms, 0.129116% Kanv8c4Yl7:0 = @param:Kanv8c4Yl7:0 -> float_type, {32, 201}, {201, 1}: 0.00071804ms, 0.00633106% main:@1912 = slice[axes={0, 1},starts={0, 1},ends={32, 65}](Kanv8c4Yl7:0) -> float_type, {32, 64}, {201, 1}: 0.00070356ms, 0.00620339% main:@1913 = slice[axes={0, 1},starts={0, 193},ends={32, 257}](lyiDxfzZg3:0) -> float_type, {32, 64}, {701, 1}: 0.00077664ms, 0.00684775% main:@1914 = load[offset=139776,end=156160](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00051368ms, 0.00452919% main:@1915 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=2048,local=256,](main:@1913,main:@1912,main:@1914) -> float_type, {32, 128}, {128, 1}: 0.0140243ms, 0.123655% main:@1916 = slice[axes={0, 1},starts={0, 1},ends={32, 65}](nv2tzWRf1I:0) -> float_type, {32, 64}, {301, 1}: 0.00079788ms, 0.00703502% main:@1917 = slice[axes={0, 1},starts={0, 65},ends={32, 129}](uFSOc5Chp2:0) -> float_type, {32, 64}, {481, 1}: 0.00064372ms, 0.00567577% main:@1918 = load[offset=123392,end=139776](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00046214ms, 0.00407475% main:@1919 = slice[axes={0, 1},starts={0, 1},ends={32, 65}](uFSOc5Chp2:0) -> float_type, {32, 64}, {481, 1}: 0.00095544ms, 0.00842425% main:@1920 = gpu::code_object[code_object=13544,symbol_name=noop_add_concat_div_kernel,global=2048,local=256,](main:@1917,main:@1916,main:@1919,main:@1915,main:@1918) -> float_type, {32, 128}, {128, 1}: 0.0145417ms, 0.128216% main:@1921 = load[offset=139776,end=172544](main:@1) -> float_type, {32, 1, 4, 64}, {256, 256, 64, 1}: 0.00066738ms, 0.00588439% main:@1922 = gpu::code_object[code_object=13528,symbol_name=contiguous_kernel,global=4096,local=256,](main:@1876,main:@1921) -> float_type, {32, 1, 4, 64}, {256, 256, 64, 1}: 0.0133329ms, 0.117558% main:@1923 = transpose[permutation={1, 2, 0, 3}](main:@1315) -> float_type, {32, 1, 2, 64}, {64, 64, 2048, 1}: 0.0006972ms, 0.00614731% main:@1924 = load[offset=172544,end=188928](main:@1) -> float_type, {32, 1, 2, 64}, {128, 128, 64, 1}: 0.00049758ms, 0.00438723% main:@1925 = gpu::code_object[code_object=13528,symbol_name=contiguous_kernel,global=2048,local=256,](main:@1923,main:@1924) -> float_type, {32, 1, 2, 64}, {128, 128, 64, 1}: 0.0130571ms, 0.115126% main:@1926 = slice[axes={0, 1},starts={0, 33},ends={32, 37}](Qp3K1PHvte:0) -> float_type, {32, 4}, {69, 1}: 0.00113978ms, 0.0100496% main:@1927 = slice[axes={0, 1},starts={0, 5},ends={32, 13}](mIR0ToEnMH:0) -> float_type, {32, 8}, {13, 1}: 0.0006633ms, 0.00584841% EsROupwkSG:0 = @param:EsROupwkSG:0 -> float_type, {32, 13}, {13, 1}: 0.00055088ms, 0.00485719% main:@1929 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](EsROupwkSG:0) -> float_type, {32, 4}, {13, 1}: 0.00075472ms, 0.00665447% EyFb0qo6VX:0 = @param:EyFb0qo6VX:0 -> float_type, {32, 5}, {5, 1}: 0.00057584ms, 0.00507726% main:@1931 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](EyFb0qo6VX:0) -> float_type, {32, 4}, {5, 1}: 0.00094444ms, 0.00832726% F6DhiS9CN5:0 = @param:F6DhiS9CN5:0 -> float_type, {32, 9}, {9, 1}: 0.00055966ms, 0.0049346% main:@1933 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](F6DhiS9CN5:0) -> float_type, {32, 8}, {9, 1}: 0.00078836ms, 0.00695108% NKT10PBtly:0 = @param:NKT10PBtly:0 -> float_type, {32, 5}, {5, 1}: 0.00052854ms, 0.00466021% main:@1935 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](NKT10PBtly:0) -> float_type, {32, 4}, {5, 1}: 0.00061496ms, 0.00542219% zOu1aQKy4V:0 = @param:zOu1aQKy4V:0 -> float_type, {32, 9}, {9, 1}: 0.0005223ms, 0.00460519% main:@1937 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](zOu1aQKy4V:0) -> float_type, {32, 8}, {9, 1}: 0.00063696ms, 0.00561617% FrTLGaKo7c:0 = @param:FrTLGaKo7c:0 -> float_type, {32, 13}, {13, 1}: 0.00051296ms, 0.00452284% main:@1939 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](FrTLGaKo7c:0) -> float_type, {32, 4}, {13, 1}: 0.0007343ms, 0.00647443% mDpoFQLqWK:0 = @param:mDpoFQLqWK:0 -> float_type, {32, 13}, {13, 1}: 0.00065128ms, 0.00574243% main:@1941 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](mDpoFQLqWK:0) -> float_type, {32, 4}, {13, 1}: 0.00069274ms, 0.00610799% HFesjVwyB6:0 = @param:HFesjVwyB6:0 -> float_type, {32, 9}, {9, 1}: 0.00055748ms, 0.00491538% main:@1943 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](HFesjVwyB6:0) -> float_type, {32, 8}, {9, 1}: 0.0007118ms, 0.00627604% HKq2ZSlA3D:0 = @param:HKq2ZSlA3D:0 -> float_type, {32, 9}, {9, 1}: 0.00048258ms, 0.00425498% main:@1945 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](HKq2ZSlA3D:0) -> float_type, {32, 8}, {9, 1}: 0.00071406ms, 0.00629597% HLQJSVgFyi:0 = @param:HLQJSVgFyi:0 -> float_type, {32, 9}, {9, 1}: 0.00053852ms, 0.00474821% main:@1947 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](HLQJSVgFyi:0) -> float_type, {32, 8}, {9, 1}: 0.00096354ms, 0.00849567% cNgUKoOFB1:0 = @param:cNgUKoOFB1:0 -> float_type, {32, 9}, {9, 1}: 0.00058934ms, 0.0051963% main:@1949 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](cNgUKoOFB1:0) -> float_type, {32, 8}, {9, 1}: 0.000699ms, 0.00616318% nIJL1FUMdk:0 = @param:nIJL1FUMdk:0 -> float_type, {32, 9}, {9, 1}: 0.00056344ms, 0.00496793% main:@1951 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](nIJL1FUMdk:0) -> float_type, {32, 8}, {9, 1}: 0.00068288ms, 0.00602105% main:@1952 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](ljyhc7HCLk:0) -> float_type, {32, 4}, {21, 1}: 0.00084342ms, 0.00743656% HyzMmXUNAP:0 = @param:HyzMmXUNAP:0 -> float_type, {32, 13}, {13, 1}: 0.00049254ms, 0.0043428% main:@1954 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](HyzMmXUNAP:0) -> float_type, {32, 4}, {13, 1}: 0.00070972ms, 0.0062577% main:@1955 = slice[axes={0, 1},starts={0, 33},ends={32, 37}](HzcyUlm8Yj:0) -> float_type, {32, 4}, {45, 1}: 0.00071804ms, 0.00633106% fvGJwyjlm4:0 = @param:fvGJwyjlm4:0 -> float_type, {32, 13}, {13, 1}: 0.00048976ms, 0.00431828% main:@1957 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](fvGJwyjlm4:0) -> float_type, {32, 4}, {13, 1}: 0.00075474ms, 0.00665465% IQawVv9sqx:0 = @param:IQawVv9sqx:0 -> float_type, {32, 9}, {9, 1}: 0.00057374ms, 0.00505875% main:@1959 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](IQawVv9sqx:0) -> float_type, {32, 8}, {9, 1}: 0.00079148ms, 0.00697859% main:@1960 = slice[axes={0, 1},starts={0, 321},ends={32, 501}](lyiDxfzZg3:0) -> float_type, {32, 180}, {701, 1}: 0.00071496ms, 0.00630391% fKsuaR6OE2:0 = @param:fKsuaR6OE2:0 -> float_type, {32, 13}, {13, 1}: 0.00052632ms, 0.00464064% main:@1962 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](fKsuaR6OE2:0) -> float_type, {32, 4}, {13, 1}: 0.00073224ms, 0.00645627% main:@1963 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](gdck97AK4w:0) -> float_type, {32, 8}, {17, 1}: 0.00067186ms, 0.00592389% JKWg9U1qmT:0 = @param:JKWg9U1qmT:0 -> float_type, {32, 13}, {13, 1}: 0.000489ms, 0.00431158% main:@1965 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](JKWg9U1qmT:0) -> float_type, {32, 4}, {13, 1}: 0.00079636ms, 0.00702162% hADUwpE6P7:0 = @param:hADUwpE6P7:0 -> float_type, {32, 9}, {9, 1}: 0.00053ms, 0.00467309% main:@1967 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](hADUwpE6P7:0) -> float_type, {32, 8}, {9, 1}: 0.00069704ms, 0.0061459% YMxBt9210u:0 = @param:YMxBt9210u:0 -> float_type, {32, 5}, {5, 1}: 0.0005073ms, 0.00447294% main:@1969 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](YMxBt9210u:0) -> float_type, {32, 4}, {5, 1}: 0.00072988ms, 0.00643546% KBG8U2uYOC:0 = @param:KBG8U2uYOC:0 -> float_type, {32, 9}, {9, 1}: 0.00055546ms, 0.00489757% main:@1971 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](KBG8U2uYOC:0) -> float_type, {32, 8}, {9, 1}: 0.00068456ms, 0.00603586% XevLAmJgSy:0 = @param:XevLAmJgSy:0 -> float_type, {32, 5}, {5, 1}: 0.0005664ms, 0.00499403% main:@1973 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](XevLAmJgSy:0) -> float_type, {32, 4}, {5, 1}: 0.00075872ms, 0.00668974% KOdpo5kqsI:0 = @param:KOdpo5kqsI:0 -> float_type, {32, 93}, {93, 1}: 0.00057444ms, 0.00506492% main:@1975 = slice[axes={0, 1},starts={0, 1},ends={32, 33}](KOdpo5kqsI:0) -> float_type, {32, 32}, {93, 1}: 0.00070078ms, 0.00617888% KGspl1Fzg0:0 = @param:KGspl1Fzg0:0 -> float_type, {32, 9}, {9, 1}: 0.00055778ms, 0.00491803% main:@1977 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](KGspl1Fzg0:0) -> float_type, {32, 8}, {9, 1}: 0.0007131ms, 0.00628751% main:@1978 = slice[axes={0, 1},starts={0, 33},ends={32, 49}](KOdpo5kqsI:0) -> float_type, {32, 16}, {93, 1}: 0.00068658ms, 0.00605367% main:@1979 = slice[axes={0, 1},starts={0, 65},ends={32, 97}](Kanv8c4Yl7:0) -> float_type, {32, 32}, {201, 1}: 0.00065916ms, 0.00581191% main:@1980 = slice[axes={0, 1},starts={0, 129},ends={32, 161}](Kanv8c4Yl7:0) -> float_type, {32, 32}, {201, 1}: 0.00068996ms, 0.00608348% main:@1981 = slice[axes={0, 1},starts={0, 5},ends={32, 13}](KFhBukd7EV:0) -> float_type, {32, 8}, {13, 1}: 0.00067002ms, 0.00590766% main:@1982 = slice[axes={0, 1},starts={0, 97},ends={32, 129}](Kanv8c4Yl7:0) -> float_type, {32, 32}, {201, 1}: 0.00065236ms, 0.00575195% main:@1983 = slice[axes={0, 1},starts={0, 161},ends={32, 193}](Kanv8c4Yl7:0) -> float_type, {32, 32}, {201, 1}: 0.0006554ms, 0.00577876% Ya2E8PbliH:0 = @param:Ya2E8PbliH:0 -> float_type, {32, 9}, {9, 1}: 0.00053258ms, 0.00469583% main:@1985 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](Ya2E8PbliH:0) -> float_type, {32, 8}, {9, 1}: 0.0006871ms, 0.00605826% M4pau3Zg7l:0 = @param:M4pau3Zg7l:0 -> float_type, {32, 13}, {13, 1}: 0.0005725ms, 0.00504781% main:@1987 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](M4pau3Zg7l:0) -> float_type, {32, 4}, {13, 1}: 0.00072574ms, 0.00639895% NFBzi76hoK:0 = @param:NFBzi76hoK:0 -> float_type, {32, 73}, {73, 1}: 0.00061418ms, 0.00541531% main:@1989 = slice[axes={0, 1},starts={0, 1},ends={32, 65}](NFBzi76hoK:0) -> float_type, {32, 64}, {73, 1}: 0.00066722ms, 0.00588297% main:@1990 = load[offset=781440,end=1220352](main:@1) -> float_type, {32, 3429}, {3429, 1}: 0.00046568ms, 0.00410597% main:@1991 = slice[axes={0, 1},starts={0, 33},ends={32, 37}](duyRoIEkH8:0) -> float_type, {32, 4}, {45, 1}: 0.0007119ms, 0.00627692% OfNJgEDtw2:0 = @param:OfNJgEDtw2:0 -> float_type, {32, 13}, {13, 1}: 0.00060186ms, 0.00530669% main:@1993 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](OfNJgEDtw2:0) -> float_type, {32, 4}, {13, 1}: 0.00069004ms, 0.00608418% PxhmZizCI5:0 = @param:PxhmZizCI5:0 -> float_type, {32, 13}, {13, 1}: 0.00058106ms, 0.00512329% main:@1995 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](PxhmZizCI5:0) -> float_type, {32, 4}, {13, 1}: 0.00069702ms, 0.00614573% main:@1996 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](Pu1O5qFxY6:0) -> float_type, {32, 8}, {17, 1}: 0.0006551ms, 0.00577611% main:@1997 = slice[axes={0, 1},starts={0, 9},ends={32, 13}](RHo9Jz6KpI:0) -> float_type, {32, 4}, {21, 1}: 0.00064884ms, 0.00572092% RNGK6g4bmu:0 = @param:RNGK6g4bmu:0 -> float_type, {32, 9}, {9, 1}: 0.00054866ms, 0.00483761% main:@1999 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](RNGK6g4bmu:0) -> float_type, {32, 8}, {9, 1}: 0.00071438ms, 0.00629879% mQnuwGeC8a:0 = @param:mQnuwGeC8a:0 -> float_type, {32, 9}, {9, 1}: 0.00057556ms, 0.0050748% main:@2001 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](mQnuwGeC8a:0) -> float_type, {32, 8}, {9, 1}: 0.00072282ms, 0.00637321% V84toiBOPm:0 = @param:V84toiBOPm:0 -> float_type, {32, 9}, {9, 1}: 0.0005305ms, 0.00467749% main:@2003 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](V84toiBOPm:0) -> float_type, {32, 8}, {9, 1}: 0.00067714ms, 0.00597044% QMjOWELyPk:0 = @param:QMjOWELyPk:0 -> float_type, {32, 9}, {9, 1}: 0.00052884ms, 0.00466286% main:@2005 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](QMjOWELyPk:0) -> float_type, {32, 8}, {9, 1}: 0.00067524ms, 0.00595369% main:@2006 = slice[axes={0, 1},starts={0, 5},ends={32, 13}](S5AzoJmqUC:0) -> float_type, {32, 8}, {13, 1}: 0.00067668ms, 0.00596639% main:@2007 = slice[axes={0, 1},starts={0, 165},ends={32, 181}](nv2tzWRf1I:0) -> float_type, {32, 16}, {301, 1}: 0.00066652ms, 0.0058768% main:@2008 = slice[axes={0, 1},starts={0, 5},ends={32, 13}](qeZ76Hc03v:0) -> float_type, {32, 8}, {13, 1}: 0.00065746ms, 0.00579692% u2OJ7kfglD:0 = @param:u2OJ7kfglD:0 -> float_type, {32, 93}, {93, 1}: 0.00055156ms, 0.00486318% main:@2010 = slice[axes={0, 1},starts={0, 53},ends={32, 57}](u2OJ7kfglD:0) -> float_type, {32, 4}, {93, 1}: 0.0006884ms, 0.00606972% main:@2011 = slice[axes={0, 1},starts={0, 17},ends={32, 21}](bJsIAzhYc3:0) -> float_type, {32, 4}, {37, 1}: 0.00067382ms, 0.00594117% main:@2012 = slice[axes={0, 1},starts={0, 17},ends={32, 21}](YbIVz8l1un:0) -> float_type, {32, 4}, {45, 1}: 0.00062734ms, 0.00553135% TkHQvxKeD6:0 = @param:TkHQvxKeD6:0 -> float_type, {32, 9}, {9, 1}: 0.00055286ms, 0.00487465% main:@2014 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](TkHQvxKeD6:0) -> float_type, {32, 8}, {9, 1}: 0.00072686ms, 0.00640883% ToDLE1sRI6:0 = @param:ToDLE1sRI6:0 -> float_type, {32, 9}, {9, 1}: 0.0005439ms, 0.00479564% main:@2016 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](ToDLE1sRI6:0) -> float_type, {32, 8}, {9, 1}: 0.00070184ms, 0.00618822% UTzDZk1RS4:0 = @param:UTzDZk1RS4:0 -> float_type, {32, 13}, {13, 1}: 0.0005151ms, 0.00454171% main:@2018 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](UTzDZk1RS4:0) -> float_type, {32, 4}, {13, 1}: 0.00070392ms, 0.00620656% 01E2VTipRX:0 = @param:01E2VTipRX:0 -> float_type, {32, 9}, {9, 1}: 0.00051446ms, 0.00453607% main:@2020 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](01E2VTipRX:0) -> float_type, {32, 8}, {9, 1}: 0.00068752ms, 0.00606196% main:@2021 = transpose[permutation={1, 2, 0, 3}](main:@1143) -> float_type, {32, 1, 1, 64}, {64, 64, 2048, 1}: 0.0005251ms, 0.00462988% main:@2022 = squeeze[axes={1}](main:@1903) -> float_type, {32, 256}, {256, 1}: 0.00055642ms, 0.00490604% dz1saATctE:0 = @param:dz1saATctE:0 -> float_type, {32, 9}, {9, 1}: 0.00061412ms, 0.00541478% uhcrWfUe6M:0 = @param:uhcrWfUe6M:0 -> float_type, {32, 9}, {9, 1}: 0.00051378ms, 0.00453007% vZDAiUHErW:0 = @param:vZDAiUHErW:0 -> float_type, {32, 17}, {17, 1}: 0.00052956ms, 0.00466921% 4dtfMgk7Ni:0 = @param:4dtfMgk7Ni:0 -> float_type, {32, 9}, {9, 1}: 0.00052214ms, 0.00460378% main:@2027 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](4dtfMgk7Ni:0) -> float_type, {32, 8}, {9, 1}: 0.00085326ms, 0.00752332% nP9Yv1iZ8f:0 = @param:nP9Yv1iZ8f:0 -> float_type, {32, 9}, {9, 1}: 0.00052044ms, 0.00458879% main:@2029 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](nP9Yv1iZ8f:0) -> float_type, {32, 8}, {9, 1}: 0.00066864ms, 0.0058955% T1FVGiDRMg:0 = @param:T1FVGiDRMg:0 -> float_type, {32, 5}, {5, 1}: 0.00052094ms, 0.0045932% main:@2031 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](T1FVGiDRMg:0) -> float_type, {32, 4}, {5, 1}: 0.0007319ms, 0.00645327% 50ovs1NnYh:0 = @param:50ovs1NnYh:0 -> float_type, {32, 9}, {9, 1}: 0.0005983ms, 0.0052753% main:@2033 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](50ovs1NnYh:0) -> float_type, {32, 8}, {9, 1}: 0.0006667ms, 0.00587839% main:@2034 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](5JHUPChra8:0) -> float_type, {32, 4}, {17, 1}: 0.00069282ms, 0.00610869% 6eQEnYsRTi:0 = @param:6eQEnYsRTi:0 -> float_type, {32, 5}, {5, 1}: 0.0005788ms, 0.00510336% main:@2036 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](6eQEnYsRTi:0) -> float_type, {32, 4}, {5, 1}: 0.00072516ms, 0.00639384% 5bcvLlI6HE:0 = @param:5bcvLlI6HE:0 -> float_type, {32, 9}, {9, 1}: 0.00056214ms, 0.00495647% main:@2038 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](5bcvLlI6HE:0) -> float_type, {32, 8}, {9, 1}: 0.00064028ms, 0.00564544% 5cO3meIa4V:0 = @param:5cO3meIa4V:0 -> float_type, {32, 5}, {5, 1}: 0.00050492ms, 0.00445195% main:@2040 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](5cO3meIa4V:0) -> float_type, {32, 4}, {5, 1}: 0.00069838ms, 0.00615772% Q2Om7B1nhU:0 = @param:Q2Om7B1nhU:0 -> float_type, {32, 9}, {9, 1}: 0.00059084ms, 0.00520952% main:@2042 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](Q2Om7B1nhU:0) -> float_type, {32, 8}, {9, 1}: 0.00067174ms, 0.00592283% main:@2043 = reshape_lazy[dims={-1, 128}](main:@1906) -> float_type, {32, 128}, {128, 1}: 0.00053234ms, 0.00469372% main:@2044 = slice[axes={0, 1},starts={0, 33},ends={32, 37}](6HSQqXdJzb:0) -> float_type, {32, 4}, {45, 1}: 0.00070156ms, 0.00618576% main:@2045 = slice[axes={0, 1},starts={0, 5},ends={32, 13}](w27LgiJW6P:0) -> float_type, {32, 8}, {13, 1}: 0.00071956ms, 0.00634446% KgZXYMNz6P:0 = @param:KgZXYMNz6P:0 -> float_type, {32, 13}, {13, 1}: 0.00052338ms, 0.00461472% main:@2047 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](KgZXYMNz6P:0) -> float_type, {32, 4}, {13, 1}: 0.00064838ms, 0.00571686% 6hbLUlYs1H:0 = @param:6hbLUlYs1H:0 -> float_type, {32, 9}, {9, 1}: 0.00054856ms, 0.00483673% main:@2049 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](6hbLUlYs1H:0) -> float_type, {32, 8}, {9, 1}: 0.00068266ms, 0.00601911% main:@2050 = slice[axes={0, 1},starts={0, 25},ends={32, 41}](ifP6Ceko5T:0) -> float_type, {32, 16}, {73, 1}: 0.00067162ms, 0.00592177% y3bafuwnEk:0 = @param:y3bafuwnEk:0 -> float_type, {32, 5}, {5, 1}: 0.00057408ms, 0.00506175% main:@2052 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](y3bafuwnEk:0) -> float_type, {32, 4}, {5, 1}: 0.00066912ms, 0.00589973% 8Dv01FAoNM:0 = @param:8Dv01FAoNM:0 -> float_type, {32, 17}, {17, 1}: 0.00050472ms, 0.00445019% main:@2054 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](8Dv01FAoNM:0) -> float_type, {32, 16}, {17, 1}: 0.00072586ms, 0.00640001% 89nc05gQza:0 = @param:89nc05gQza:0 -> float_type, {32, 9}, {9, 1}: 0.00055388ms, 0.00488364% main:@2056 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](89nc05gQza:0) -> float_type, {32, 8}, {9, 1}: 0.00067156ms, 0.00592124% VP35W0GOCc:0 = @param:VP35W0GOCc:0 -> float_type, {32, 13}, {13, 1}: 0.00050028ms, 0.00441104% main:@2058 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](VP35W0GOCc:0) -> float_type, {32, 4}, {13, 1}: 0.00066612ms, 0.00587328% 8KRQILP469:0 = @param:8KRQILP469:0 -> float_type, {32, 9}, {9, 1}: 0.0005694ms, 0.00502048% main:@2060 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](8KRQILP469:0) -> float_type, {32, 8}, {9, 1}: 0.00064196ms, 0.00566025% 8iOST90kWM:0 = @param:8iOST90kWM:0 -> float_type, {32, 9}, {9, 1}: 0.0005385ms, 0.00474803% main:@2062 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](8iOST90kWM:0) -> float_type, {32, 8}, {9, 1}: 0.00070026ms, 0.00617429% main:@2063 = slice[axes={0, 1},starts={0, 33},ends={32, 37}](X29iyIeu8L:0) -> float_type, {32, 4}, {45, 1}: 0.00066748ms, 0.00588527% 985IRiY0X6:0 = @param:985IRiY0X6:0 -> float_type, {32, 57}, {57, 1}: 0.00049178ms, 0.0043361% main:@2065 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](985IRiY0X6:0) -> float_type, {32, 8}, {57, 1}: 0.0006611ms, 0.00582901% main:@2066 = slice[axes={0, 1},starts={0, 49},ends={32, 57}](985IRiY0X6:0) -> float_type, {32, 8}, {57, 1}: 0.00068302ms, 0.00602229% main:@2067 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](985IRiY0X6:0) -> float_type, {32, 8}, {57, 1}: 0.00066834ms, 0.00589285% main:@2068 = slice[axes={0, 1},starts={0, 41},ends={32, 49}](985IRiY0X6:0) -> float_type, {32, 8}, {57, 1}: 0.00065304ms, 0.00575795% main:@2069 = slice[axes={0, 1},starts={0, 33},ends={32, 41}](985IRiY0X6:0) -> float_type, {32, 8}, {57, 1}: 0.00063862ms, 0.0056308% main:@2070 = slice[axes={0, 1},starts={0, 17},ends={32, 25}](985IRiY0X6:0) -> float_type, {32, 8}, {57, 1}: 0.00065484ms, 0.00577382% 9N1KUX0DZ5:0 = @param:9N1KUX0DZ5:0 -> float_type, {32, 13}, {13, 1}: 0.00050694ms, 0.00446976% main:@2072 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](9N1KUX0DZ5:0) -> float_type, {32, 4}, {13, 1}: 0.00069082ms, 0.00609106% main:@2073 = slice[axes={0, 1},starts={0, 25},ends={32, 33}](985IRiY0X6:0) -> float_type, {32, 8}, {57, 1}: 0.000668ms, 0.00588985% main:@2074 = slice[axes={0, 1},starts={0, 33},ends={32, 37}](VUqfdN5Fzv:0) -> float_type, {32, 4}, {45, 1}: 0.00070292ms, 0.00619775% main:@2075 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](9YqhflbMKe:0) -> float_type, {32, 4}, {29, 1}: 0.00066952ms, 0.00590325% main:@2076 = slice[axes={0, 1},starts={0, 17},ends={32, 25}](l4bReWtsMB:0) -> float_type, {32, 8}, {25, 1}: 0.00065284ms, 0.00575618% 9mfMQZIXVH:0 = @param:9mfMQZIXVH:0 -> float_type, {32, 9}, {9, 1}: 0.0005743ms, 0.00506369% main:@2078 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](9mfMQZIXVH:0) -> float_type, {32, 8}, {9, 1}: 0.00067234ms, 0.00592812% main:@2079 = slice[axes={0, 1},starts={0, 9},ends={32, 13}](AVLcBpWeuo:0) -> float_type, {32, 4}, {21, 1}: 0.00065628ms, 0.00578652% 9rDQhGZo18:0 = @param:9rDQhGZo18:0 -> float_type, {32, 9}, {9, 1}: 0.00055162ms, 0.00486371% main:@2081 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](9rDQhGZo18:0) -> float_type, {32, 8}, {9, 1}: 0.00074366ms, 0.00655696% AfxiFXbSj3:0 = @param:AfxiFXbSj3:0 -> float_type, {32, 5}, {5, 1}: 0.0005938ms, 0.00523562% main:@2083 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](AfxiFXbSj3:0) -> float_type, {32, 4}, {5, 1}: 0.0006986ms, 0.00615966% Nb9eDKFYZp:0 = @param:Nb9eDKFYZp:0 -> float_type, {32, 5}, {5, 1}: 0.0005236ms, 0.00461666% main:@2085 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](Nb9eDKFYZp:0) -> float_type, {32, 4}, {5, 1}: 0.0006893ms, 0.00607766% AvlwbHMnuE:0 = @param:AvlwbHMnuE:0 -> float_type, {32, 17}, {17, 1}: 0.00057572ms, 0.00507621% main:@2087 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](AvlwbHMnuE:0) -> float_type, {32, 8}, {17, 1}: 0.00067958ms, 0.00599195% KH4GEDx8pO:0 = @param:KH4GEDx8pO:0 -> float_type, {32, 13}, {13, 1}: 0.00065262ms, 0.00575424% main:@2089 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](KH4GEDx8pO:0) -> float_type, {32, 4}, {13, 1}: 0.0007117ms, 0.00627516% CrB2gE9Adp:0 = @param:CrB2gE9Adp:0 -> float_type, {32, 13}, {13, 1}: 0.0005578ms, 0.0049182% main:@2091 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](CrB2gE9Adp:0) -> float_type, {32, 4}, {13, 1}: 0.00070118ms, 0.00618241% main:@2092 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](uhcrWfUe6M:0) -> float_type, {32, 8}, {9, 1}: 0.00064798ms, 0.00571333% main:@2093 = slice[axes={0, 1},starts={0, 33},ends={32, 41}](n4iptOVFwE:0) -> float_type, {32, 8}, {113, 1}: 0.00062344ms, 0.00549696% j25yTFI0eZ:0 = @param:j25yTFI0eZ:0 -> float_type, {32, 9}, {9, 1}: 0.00061682ms, 0.00543859% main:@2095 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](j25yTFI0eZ:0) -> float_type, {32, 8}, {9, 1}: 0.00070438ms, 0.00621062% dMO2bknDYA:0 = @param:dMO2bknDYA:0 -> float_type, {32, 481}, {481, 1}: 0.00050726ms, 0.00447258% main:@2097 = slice[axes={0, 1},starts={0, 233},ends={32, 241}](dMO2bknDYA:0) -> float_type, {32, 8}, {481, 1}: 0.00067964ms, 0.00599248% DilwT5jYf7:0 = @param:DilwT5jYf7:0 -> float_type, {32, 9}, {9, 1}: 0.00051766ms, 0.00456428% main:@2099 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](DilwT5jYf7:0) -> float_type, {32, 8}, {9, 1}: 0.00071226ms, 0.0062801% EJxze1pkRu:0 = @param:EJxze1pkRu:0 -> float_type, {32, 13}, {13, 1}: 0.0005121ms, 0.00451526% main:@2101 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](EJxze1pkRu:0) -> float_type, {32, 4}, {13, 1}: 0.00067476ms, 0.00594946% main:@2102 = slice[axes={0, 1},starts={0, 61},ends={32, 65}](u2OJ7kfglD:0) -> float_type, {32, 4}, {93, 1}: 0.00065384ms, 0.005765% main:@2103 = transpose[permutation={1, 2, 0, 3}](main:@1531) -> float_type, {32, 1, 1, 32}, {32, 32, 1024, 1}: 0.00051212ms, 0.00451544% main:@2104 = transpose[permutation={1, 2, 0, 3}](main:@1570) -> float_type, {32, 1, 1, 32}, {32, 32, 1024, 1}: 0.00048852ms, 0.00430735% main:@2105 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](0cjwsL5ldh:0) -> float_type, {32, 16}, {33, 1}: 0.00067476ms, 0.00594946% WycVU13dfZ:0 = @param:WycVU13dfZ:0 -> float_type, {32, 5}, {5, 1}: 0.00051128ms, 0.00450803% main:@2107 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](WycVU13dfZ:0) -> float_type, {32, 4}, {5, 1}: 0.00068382ms, 0.00602934% main:@2108 = transpose[permutation={1, 2, 0, 3}](main:@1664) -> float_type, {32, 1, 1, 32}, {32, 32, 1024, 1}: 0.00049442ms, 0.00435937% main:@2109 = slice[axes={0, 1},starts={0, 5},ends={32, 13}](0sDBrPmfS9:0) -> float_type, {32, 8}, {13, 1}: 0.000704ms, 0.00620727% main:@2110 = transpose[permutation={1, 2, 0, 3}](main:@1720) -> float_type, {32, 1, 1, 32}, {32, 32, 1024, 1}: 0.0004715ms, 0.00415728% sfqQKWXEgN:0 = @param:sfqQKWXEgN:0 -> float_type, {32, 5}, {5, 1}: 0.00052032ms, 0.00458774% main:@2112 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](sfqQKWXEgN:0) -> float_type, {32, 4}, {5, 1}: 0.00065036ms, 0.00573432% tZOFD39jRS:0 = @param:tZOFD39jRS:0 -> float_type, {32, 9}, {9, 1}: 0.00054116ms, 0.00477149% main:@2114 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](tZOFD39jRS:0) -> float_type, {32, 8}, {9, 1}: 0.00073876ms, 0.00651375% main:@2115 = slice[axes={0, 1},starts={0, 57},ends={32, 61}](u2OJ7kfglD:0) -> float_type, {32, 4}, {93, 1}: 0.00067506ms, 0.0059521% main:@2116 = slice[axes={0, 1},starts={0, 33},ends={32, 37}](zIoqYe9cRb:0) -> float_type, {32, 4}, {45, 1}: 0.00064166ms, 0.00565761% main:@2117 = slice[axes={0, 1},starts={0, 33},ends={32, 37}](dAFYcGrfEl:0) -> float_type, {32, 4}, {45, 1}: 0.0006393ms, 0.0056368% vPaCFg3sAl:0 = @param:vPaCFg3sAl:0 -> float_type, {32, 13}, {13, 1}: 0.0005883ms, 0.00518713% main:@2119 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](vPaCFg3sAl:0) -> float_type, {32, 4}, {13, 1}: 0.00070514ms, 0.00621732% main:@2120 = slice[axes={0, 1},starts={0, 65},ends={32, 129}](nv2tzWRf1I:0) -> float_type, {32, 64}, {301, 1}: 0.00066512ms, 0.00586446% main:@2121 = slice[axes={0, 1},starts={0, 33},ends={32, 49}](u2OJ7kfglD:0) -> float_type, {32, 16}, {93, 1}: 0.00065144ms, 0.00574384% fpcFBPsTyk:0 = @param:fpcFBPsTyk:0 -> float_type, {32, 9}, {9, 1}: 0.00052838ms, 0.0046588% main:@2123 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](fpcFBPsTyk:0) -> float_type, {32, 8}, {9, 1}: 0.00071914ms, 0.00634076% main:@2124 = slice[axes={0, 1},starts={0, 129},ends={32, 193}](lyiDxfzZg3:0) -> float_type, {32, 64}, {701, 1}: 0.000657ms, 0.00579286% uzaUY08Rsp:0 = @param:uzaUY08Rsp:0 -> float_type, {32, 9}, {9, 1}: 0.00046228ms, 0.00407599% main:@2126 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](uzaUY08Rsp:0) -> float_type, {32, 8}, {9, 1}: 0.00067908ms, 0.00598755% main:@2127 = slice[axes={0, 1},starts={0, 5},ends={32, 13}](whRnWGgf0Q:0) -> float_type, {32, 8}, {13, 1}: 0.00067322ms, 0.00593588% sXCjeM94Zu:0 = @param:sXCjeM94Zu:0 -> float_type, {32, 5}, {5, 1}: 0.00055346ms, 0.00487994% main:@2129 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](sXCjeM94Zu:0) -> float_type, {32, 4}, {5, 1}: 0.00074396ms, 0.0065596% main:@2130 = slice[axes={0, 1},starts={0, 17},ends={32, 25}](l5WSX3VvMK:0) -> float_type, {32, 8}, {25, 1}: 0.00066622ms, 0.00587416% main:@2131 = slice[axes={0, 1},starts={0, 49},ends={32, 53}](u2OJ7kfglD:0) -> float_type, {32, 4}, {93, 1}: 0.00066084ms, 0.00582672% dXvEYVC4st:0 = @param:dXvEYVC4st:0 -> float_type, {32, 13}, {13, 1}: 0.000507ms, 0.00447029% main:@2133 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](dXvEYVC4st:0) -> float_type, {32, 4}, {13, 1}: 0.0006823ms, 0.00601594% main:@2134 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](dz1saATctE:0) -> float_type, {32, 8}, {9, 1}: 0.00065178ms, 0.00574684% sOz8pkBZ5n:0 = @param:sOz8pkBZ5n:0 -> float_type, {32, 5}, {5, 1}: 0.00057416ms, 0.00506245% main:@2136 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](sOz8pkBZ5n:0) -> float_type, {32, 4}, {5, 1}: 0.00069498ms, 0.00612774% main:@2137 = slice[axes={0, 1},starts={0, 1},ends={32, 65}](lyiDxfzZg3:0) -> float_type, {32, 64}, {701, 1}: 0.00065146ms, 0.00574402% pIb4NdDSn5:0 = @param:pIb4NdDSn5:0 -> float_type, {32, 9}, {9, 1}: 0.00059702ms, 0.00526401% main:@2139 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](pIb4NdDSn5:0) -> float_type, {32, 8}, {9, 1}: 0.0007139ms, 0.00629456% main:@2140 = slice[axes={0, 1},starts={0, 5},ends={32, 13}](d1ExPaGqMN:0) -> float_type, {32, 8}, {13, 1}: 0.0006789ms, 0.00598596% v4fNptDqcL:0 = @param:v4fNptDqcL:0 -> float_type, {32, 9}, {9, 1}: 0.00053098ms, 0.00468173% main:@2142 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](v4fNptDqcL:0) -> float_type, {32, 8}, {9, 1}: 0.0007714ms, 0.00680154% dR9G1bayok:0 = @param:dR9G1bayok:0 -> float_type, {32, 13}, {13, 1}: 0.00051884ms, 0.00457469% main:@2144 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](dR9G1bayok:0) -> float_type, {32, 4}, {13, 1}: 0.0006869ms, 0.0060565% main:@2145 = slice[axes={0, 1},starts={0, 33},ends={32, 41}](mgnGNH4Mz3:0) -> float_type, {32, 8}, {49, 1}: 0.00064974ms, 0.00572885% fqAX8l29PG:0 = @param:fqAX8l29PG:0 -> float_type, {32, 9}, {9, 1}: 0.00049696ms, 0.00438177% main:@2147 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](fqAX8l29PG:0) -> float_type, {32, 8}, {9, 1}: 0.00068816ms, 0.00606761% main:@2148 = slice[axes={0, 1},starts={0, 1},ends={32, 33}](u2OJ7kfglD:0) -> float_type, {32, 32}, {93, 1}: 0.0006623ms, 0.00583959% Wiks0GPJMQ:0 = @param:Wiks0GPJMQ:0 -> float_type, {32, 9}, {9, 1}: 0.0005226ms, 0.00460784% main:@2150 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](Wiks0GPJMQ:0) -> float_type, {32, 8}, {9, 1}: 0.00070306ms, 0.00619898% aKmfBy1Mhi:0 = @param:aKmfBy1Mhi:0 -> float_type, {32, 9}, {9, 1}: 0.00054644ms, 0.00481804% main:@2152 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](aKmfBy1Mhi:0) -> float_type, {32, 8}, {9, 1}: 0.0007066ms, 0.00623019% uM7aDfZz1t:0 = @param:uM7aDfZz1t:0 -> float_type, {32, 9}, {9, 1}: 0.00055262ms, 0.00487253% main:@2154 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](uM7aDfZz1t:0) -> float_type, {32, 8}, {9, 1}: 0.00068012ms, 0.00599672% main:@2155 = squeeze[axes={1}](main:@1908) -> float_type, {32, 104}, {104, 1}: 0.00052014ms, 0.00458615% 1f2xpcXgdE:0 = @param:1f2xpcXgdE:0 -> float_type, {32, 13}, {13, 1}: 0.00053514ms, 0.00471841% main:@2157 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](1f2xpcXgdE:0) -> float_type, {32, 4}, {13, 1}: 0.00068018ms, 0.00599725% main:@2158 = squeeze[axes={1}](main:@1910) -> float_type, {32, 68}, {68, 1}: 0.00050514ms, 0.00445389% main:@2159 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](vZDAiUHErW:0) -> float_type, {32, 8}, {17, 1}: 0.00065778ms, 0.00579974% main:@2160 = slice[axes={0, 1},starts={0, 33},ends={32, 41}](2CnSyUBzm0:0) -> float_type, {32, 8}, {65, 1}: 0.00065198ms, 0.0057486% main:@2161 = transpose[permutation={1, 2, 0, 3}](main:@1742) -> float_type, {32, 1, 1, 32}, {32, 32, 1024, 1}: 0.00047968ms, 0.00422941% main:@2162 = slice[axes={0, 1},starts={0, 5},ends={32, 13}](m5MfQhR1Bp:0) -> float_type, {32, 8}, {13, 1}: 0.00065628ms, 0.00578652% main:@2163 = slice[axes={0, 1},starts={0, 33},ends={32, 37}](2RDXZrx9GA:0) -> float_type, {32, 4}, {45, 1}: 0.0006782ms, 0.00597979% main:@2164 = slice[axes={0, 1},starts={0, 33},ends={32, 37}](2Yy3zxNGBC:0) -> float_type, {32, 4}, {45, 1}: 0.0006481ms, 0.00571439% main:@2165 = slice[axes={0, 1},starts={0, 17},ends={32, 25}](ZRYNFfjSez:0) -> float_type, {32, 8}, {33, 1}: 0.0006758ms, 0.00595863% BQHyb3DGCE:0 = @param:BQHyb3DGCE:0 -> float_type, {32, 9}, {9, 1}: 0.00055986ms, 0.00493637% main:@2167 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](BQHyb3DGCE:0) -> float_type, {32, 8}, {9, 1}: 0.00070268ms, 0.00619563% main:@2168 = slice[axes={0, 1},starts={0, 33},ends={32, 37}](3Laj620MSn:0) -> float_type, {32, 4}, {45, 1}: 0.00064908ms, 0.00572303% RF5h8lytsu:0 = @param:RF5h8lytsu:0 -> float_type, {32, 9}, {9, 1}: 0.00049202ms, 0.00433821% main:@2170 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](RF5h8lytsu:0) -> float_type, {32, 8}, {9, 1}: 0.00066344ms, 0.00584965% EClzvVLqMP:0 = @param:EClzvVLqMP:0 -> float_type, {32, 9}, {9, 1}: 0.00050846ms, 0.00448317% main:@2172 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](EClzvVLqMP:0) -> float_type, {32, 8}, {9, 1}: 0.00069726ms, 0.00614784% MtvajoSCeD:0 = @param:MtvajoSCeD:0 -> float_type, {32, 9}, {9, 1}: 0.00051562ms, 0.0045463% main:@2174 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](MtvajoSCeD:0) -> float_type, {32, 8}, {9, 1}: 0.00070194ms, 0.00618911% main:@2175 = slice[axes={0, 1},starts={0, 17},ends={32, 25}](3mk8BKjDey:0) -> float_type, {32, 8}, {25, 1}: 0.00066912ms, 0.00589973% MQb6fHVndu:0 = @param:MQb6fHVndu:0 -> float_type, {32, 9}, {9, 1}: 0.00058966ms, 0.00519912% main:@2177 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](MQb6fHVndu:0) -> float_type, {32, 8}, {9, 1}: 0.00071644ms, 0.00631695% 4U3hVYNn2F:0 = @param:4U3hVYNn2F:0 -> float_type, {32, 5}, {5, 1}: 0.00050414ms, 0.00444508% main:@2179 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](4U3hVYNn2F:0) -> float_type, {32, 4}, {5, 1}: 0.00070268ms, 0.00619563% ceq9LkmQA5:0 = @param:ceq9LkmQA5:0 -> float_type, {32, 9}, {9, 1}: 0.00049288ms, 0.00434579% zvFYW45wMr:0 = @param:zvFYW45wMr:0 -> float_type, {32, 9}, {9, 1}: 0.00054676ms, 0.00482086% ihKZd8ANk7:0 = @param:ihKZd8ANk7:0 -> float_type, {32, 9}, {9, 1}: 0.00050548ms, 0.00445689% bsFqHW5i6g:0 = @param:bsFqHW5i6g:0 -> float_type, {32, 9}, {9, 1}: 0.00063788ms, 0.00562428% vi7GCYp4FT:0 = @param:vi7GCYp4FT:0 -> float_type, {32, 9}, {9, 1}: 0.00053338ms, 0.00470289% bhuXRdz0E5:0 = @param:bhuXRdz0E5:0 -> float_type, {32, 9}, {9, 1}: 0.00061918ms, 0.0054594% ihcEmsYo9g:0 = @param:ihcEmsYo9g:0 -> float_type, {32, 5}, {5, 1}: 0.0005103ms, 0.00449939% sjKQJGby31:0 = @param:sjKQJGby31:0 -> float_type, {32, 9}, {9, 1}: 0.00055292ms, 0.00487518% uGCZaoYr3L:0 = @param:uGCZaoYr3L:0 -> float_type, {32, 13}, {13, 1}: 0.00049906ms, 0.00440028% ad9BJGx5jl:0 = @param:ad9BJGx5jl:0 -> float_type, {32, 13}, {13, 1}: 0.00055748ms, 0.00491538% kX6xhVY7bJ:0 = @param:kX6xhVY7bJ:0 -> float_type, {32, 13}, {13, 1}: 0.00060046ms, 0.00529434% alvtrHJ95G:0 = @param:alvtrHJ95G:0 -> float_type, {32, 13}, {13, 1}: 0.00051402ms, 0.00453219% gHPTxdcQWX:0 = @param:gHPTxdcQWX:0 -> float_type, {32, 9}, {9, 1}: 0.00050372ms, 0.00444137% main:@2193 = reshape_lazy[dims={-1, 128}](main:@1925) -> float_type, {32, 128}, {128, 1}: 0.00058056ms, 0.00511888% main:@2194 = reshape_lazy[dims={-1, 84}](main:@1504) -> float_type, {32, 84}, {84, 1}: 0.00050358ms, 0.00444014% main:@2195 = reshape_lazy[dims={-1, 64}](main:@2021) -> float_type, {32, 64}, {64, 1}: 0.00047956ms, 0.00422835% main:@2196 = reshape_lazy[dims={-1, 105}](main:@1031) -> float_type, {32, 105}, {105, 1}: 0.00048948ms, 0.00431582% main:@2197 = reshape_lazy[dims={-1, 32}](main:@2161) -> float_type, {32, 32}, {32, 1}: 0.00046972ms, 0.00414159% main:@2198 = reshape_lazy[dims={-1, 256}](main:@1922) -> float_type, {32, 256}, {256, 1}: 0.00046784ms, 0.00412501% main:@2199 = reshape_lazy[dims={-1, 64}](main:@1076) -> float_type, {32, 64}, {64, 1}: 0.00049104ms, 0.00432957% main:@2200 = reshape_lazy[dims={-1, 32}](main:@2104) -> float_type, {32, 32}, {32, 1}: 0.00048314ms, 0.00425991% main:@2201 = reshape_lazy[dims={-1, 100}](main:@1873) -> float_type, {32, 100}, {100, 1}: 0.0004629ms, 0.00408146% main:@2202 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](sjKQJGby31:0) -> float_type, {32, 8}, {9, 1}: 0.00074826ms, 0.00659752% main:@2203 = slice[axes={0, 1},starts={0, 5},ends={32, 13}](Vi28Dy4Ll5:0) -> float_type, {32, 8}, {13, 1}: 0.00070566ms, 0.00622191% main:@2204 = slice[axes={0, 1},starts={0, 33},ends={32, 41}](1aXQU27NrD:0) -> float_type, {32, 8}, {49, 1}: 0.0006576ms, 0.00579815% main:@2205 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](uGCZaoYr3L:0) -> float_type, {32, 4}, {13, 1}: 0.00067918ms, 0.00598843% main:@2206 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](ihcEmsYo9g:0) -> float_type, {32, 4}, {5, 1}: 0.00063614ms, 0.00560894% main:@2207 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](vi7GCYp4FT:0) -> float_type, {32, 8}, {9, 1}: 0.00062786ms, 0.00553593% main:@2208 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](ihKZd8ANk7:0) -> float_type, {32, 8}, {9, 1}: 0.00064524ms, 0.00568917% main:@2209 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](kX6xhVY7bJ:0) -> float_type, {32, 4}, {13, 1}: 0.00061652ms, 0.00543595% main:@2210 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](ad9BJGx5jl:0) -> float_type, {32, 4}, {13, 1}: 0.0006892ms, 0.00607678% aocCEyfXIS:0 = @param:aocCEyfXIS:0 -> float_type, {32, 93}, {93, 1}: 0.00052536ms, 0.00463217% main:@2212 = slice[axes={0, 1},starts={0, 1},ends={32, 33}](aocCEyfXIS:0) -> float_type, {32, 32}, {93, 1}: 0.00071858ms, 0.00633582% main:@2213 = slice[axes={0, 1},starts={0, 33},ends={32, 49}](aocCEyfXIS:0) -> float_type, {32, 16}, {93, 1}: 0.00064748ms, 0.00570892% main:@2214 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](alvtrHJ95G:0) -> float_type, {32, 4}, {13, 1}: 0.00064384ms, 0.00567683% main:@2215 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](bsFqHW5i6g:0) -> float_type, {32, 8}, {9, 1}: 0.00065712ms, 0.00579392% main:@2216 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](ceq9LkmQA5:0) -> float_type, {32, 8}, {9, 1}: 0.00070256ms, 0.00619457% main:@2217 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](cHzIDJm6us:0) -> float_type, {32, 16}, {33, 1}: 0.00068446ms, 0.00603498% gwNpMszEo3:0 = @param:gwNpMszEo3:0 -> float_type, {32, 5}, {5, 1}: 0.00052344ms, 0.00461525% main:@2219 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](gwNpMszEo3:0) -> float_type, {32, 4}, {5, 1}: 0.00066566ms, 0.00586922% main:@2220 = slice[axes={0, 1},starts={0, 33},ends={32, 37}](jznORDJr0N:0) -> float_type, {32, 4}, {45, 1}: 0.00063698ms, 0.00561634% main:@2221 = slice[axes={0, 1},starts={0, 5},ends={32, 13}](gU9SoiHyRO:0) -> float_type, {32, 8}, {13, 1}: 0.00063814ms, 0.00562657% main:@2222 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](gHPTxdcQWX:0) -> float_type, {32, 8}, {9, 1}: 0.00065876ms, 0.00580838% main:@2223 = slice[axes={0, 1},starts={0, 177},ends={32, 209}](uFSOc5Chp2:0) -> float_type, {32, 32}, {481, 1}: 0.0006323ms, 0.00557508% main:@2224 = slice[axes={0, 1},starts={0, 65},ends={32, 129}](lyiDxfzZg3:0) -> float_type, {32, 64}, {701, 1}: 0.0006262ms, 0.0055213% main:@2225 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](w04Keh2FJu:0) -> float_type, {32, 8}, {13, 1}: 0.00068164ms, 0.00601012% main:@2226 = reshape_lazy[dims={-1, 32}](main:@2108) -> float_type, {32, 32}, {32, 1}: 0.00048862ms, 0.00430823% main:@2227 = reshape_lazy[dims={-1, 32}](main:@2110) -> float_type, {32, 32}, {32, 1}: 0.00047428ms, 0.0041818% main:@2228 = reshape_lazy[dims={-1, 32}](main:@2103) -> float_type, {32, 32}, {32, 1}: 0.00047096ms, 0.00415252% main:@2229 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](zvFYW45wMr:0) -> float_type, {32, 8}, {9, 1}: 0.0006725ms, 0.00592953% main:@2230 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](bhuXRdz0E5:0) -> float_type, {32, 8}, {9, 1}: 0.00067416ms, 0.00594417% main:@2231 = gpu::code_object[code_object=72784,symbol_name=pow_div_add_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_concat_log_add_kernel,global=574,local=256,](main:@2137,main:@1916,main:@2224,main:@2120,main:@2124,main:@1919,main:@1920,main:@1989,main:@2148,main:@1975,main:@2212,main:@2196,main:@2194,main:@2201,main:@1960,main:@2007,main:@2223,main:@2050,main:@1926,main:@2160,main:@2220,main:@2163,main:@2074,main:@1955,main:@2168,main:@1991,main:@2159,main:@1929,main:@2145,main:@2204,main:@2044,main:@2116,main:@2117,main:@2063,main:@1993,main:@2011,main:@2012,main:@1963,main:@1996,main:@2165,main:@2164,main:@2076,main:@2130,main:@2079,main:@2139,main:@2225,main:@2092,main:@2134,main:@1997,main:@2087,main:@2034,main:@2031,main:@2167,main:@2075,main:@1952,main:@2121,main:@1978,main:@2213,main:@1979,main:@1982,main:@1980,main:@1983,main:@2131,main:@2010,main:@2115,main:@2102,main:@2193,main:@1935,main:@1931,main:@2219,main:@2112,main:@2083,main:@2040,main:@2136,main:@2085,main:@2107,main:@2129,main:@2043,main:@2199,main:@2065,main:@2067,main:@2070,main:@2073,main:@2069,main:@2068,main:@2066,main:@1957,main:@1965,main:@2072,main:@1941,main:@1962,main:@2089,main:@2157,main:@1987,main:@2047,main:@2101,main:@2058,main:@1939,main:@2018,main:@2214,main:@2209,main:@1995,main:@2119,main:@2091,main:@2210,main:@2197,main:@2195,main:@2226,main:@2144,main:@2205,main:@2133,main:@1954,main:@2049,main:@2014,main:@2062,main:@2056,main:@1933,main:@1973,main:@1969,main:@2036,main:@2052,main:@2179,main:@2206,main:@2097,main:@2016,main:@2174,main:@2230,main:@2150,main:@2147,main:@1977,main:@2207,main:@2215,main:@2109,main:@2126,main:@2033,main:@2177,main:@2042,main:@2175,main:@2060,main:@2123,main:@2172,main:@1959,main:@2078,main:@2222,main:@2202,main:@2029,main:@2208,main:@2020,main:@1971,main:@2154,main:@2217,main:@1949,main:@1945,main:@2229,main:@1967,main:@1943,main:@1947,main:@2054,main:@2105,main:@2228,main:@1937,main:@2008,main:@2203,main:@2162,main:@1951,main:@1927,main:@2216,main:@2140,main:@2006,main:@2127,main:@2114,main:@2045,main:@2099,main:@2152,main:@2095,main:@1999,main:@1985,main:@2005,main:@2003,main:@2081,main:@2001,main:@2142,main:@2038,main:@2027,main:@2170,main:@2093,main:@2221,main:@1981,main:@2200,main:@2158,main:@2227,main:@2155,main:@2198,main:@2022,main:@1147,main:@1990) -> float_type, {32, 3429}, {3429, 1}: 0.230915ms, 2.03602% main:@2232 = hip::hip_copy_literal[id=main:@literal:6] -> float_type, {3429, 2905}, {2905, 1}: 0.00093444ms, 0.00823909% main:@2233 = hip::hip_copy_literal[id=main:@literal:20] -> float_type, {32, 2905}, {2905, 1}: 0.00062482ms, 0.00550913% main:@2234 = load[offset=1220352,end=1592192](main:@1) -> float_type, {32, 2905}, {2905, 1}: 0.0004857ms, 0.00428249% main:@2235 = gpu::gemm[alpha=1,beta=1,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@2231,main:@2232,main:@2233,main:@2234) -> float_type, {32, 2905}, {2905, 1}: 0.0836124ms, 0.737222% main:@2236 = hip::hip_copy_literal[id=main:@literal:4] -> float_type, {581, 581}, {581, 1}: 0.00080308ms, 0.00708087% main:@2237 = slice[axes={1},starts={1743},ends={2324}](main:@2235) -> float_type, {32, 581}, {2905, 1}: 0.00070172ms, 0.00618717% main:@2238 = load[offset=766080,end=840448](main:@1) -> float_type, {32, 581}, {581, 1}: 0.00050286ms, 0.00443379% main:@2239 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@2237,main:@2236,main:@2238) -> float_type, {32, 581}, {581, 1}: 0.0348164ms, 0.306981% main:@2240 = hip::hip_copy_literal[id=main:@literal:5] -> float_type, {128, 4508}, {4508, 1}: 0.00084498ms, 0.00745031% main:@2241 = hip::hip_copy_literal[id=main:@literal:2] -> float_type, {32, 4508}, {4508, 1}: 0.0005994ms, 0.005285% main:@2242 = slice[axes={1},starts={384},ends={512}](main:@765) -> float_type, {32, 128}, {896, 1}: 0.000619ms, 0.00545781% main:@2243 = load[offset=114688,end=691712](main:@1) -> float_type, {32, 4508}, {4508, 1}: 0.00049024ms, 0.00432252% main:@2244 = gpu::gemm[alpha=1,beta=1,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@2242,main:@2240,main:@2241,main:@2243) -> float_type, {32, 4508}, {4508, 1}: 0.027039ms, 0.238407% main:@2245 = load[offset=691712,end=766080](main:@1) -> float_type, {32, 581}, {581, 1}: 0.0006325ms, 0.00557684% main:@2246 = gpu::code_object[code_object=13544,symbol_name=pow_mul_add_neg_sub_kernel,global=18592,local=256,](main:@2237,main:@2239,main:@2245) -> float_type, {32, 581}, {581, 1}: 0.0170592ms, 0.150413% main:@2247 = slice[axes={1},starts={1162},ends={1743}](main:@2235) -> float_type, {32, 581}, {2905, 1}: 0.0007497ms, 0.00661021% main:@2248 = slice[axes={1},starts={0},ends={581}](main:@2235) -> float_type, {32, 581}, {2905, 1}: 0.00057614ms, 0.00507991% main:@2249 = load[offset=766080,end=840448](main:@1) -> float_type, {32, 581}, {581, 1}: 0.00047682ms, 0.00420419% main:@2250 = gpu::code_object[code_object=13544,symbol_name=pow_pow_sigmoid_add_kernel,global=18592,local=256,](main:@2247,main:@2248,main:@2249) -> float_type, {32, 581}, {581, 1}: 0.0178653ms, 0.157521% main:@2251 = slice[axes={1},starts={2324},ends={2905}](main:@2235) -> float_type, {32, 581}, {2905, 1}: 0.00075202ms, 0.00663067% main:@2252 = slice[axes={1},starts={581},ends={1162}](main:@2235) -> float_type, {32, 581}, {2905, 1}: 0.00059022ms, 0.00520405% main:@2253 = load[offset=840448,end=914816](main:@1) -> float_type, {32, 581}, {581, 1}: 0.00047028ms, 0.00414653% main:@2254 = gpu::code_object[code_object=13536,symbol_name=pow_mul_rsqrt_sub_kernel,global=18592,local=256,](main:@2251,main:@2252,main:@2253) -> float_type, {32, 581}, {581, 1}: 0.0168807ms, 0.148839% main:@2255 = load[offset=914816,end=914944](main:@1) -> float_type, {32, 1}, {1, 1}: 0.00061978ms, 0.00546469% main:@2256 = gpu::code_object[code_object=13552,symbol_name=reduce_mean_add_rsqrt_add_kernel,global=8192,local=256,](main:@2254,main:@2255) -> float_type, {32, 1}, {1, 1}: 0.01523ms, 0.134285% main:@2257 = load[offset=914944,end=915072](main:@1) -> float_type, {32, 1}, {1, 1}: 0.00061564ms, 0.00542819% main:@2258 = gpu::code_object[code_object=13552,symbol_name=reduce_mean_add_neg_pow_kernel,global=8192,local=256,](main:@2246,main:@2257) -> float_type, {32, 1}, {1, 1}: 0.0152754ms, 0.134686% main:@2259 = load[offset=915072,end=915200](main:@1) -> float_type, {32, 1}, {1, 1}: 0.00063884ms, 0.00563274% main:@2260 = gpu::code_object[code_object=13552,symbol_name=reduce_prod_add_rsqrt_sub_kernel,global=8192,local=256,](main:@2250,main:@2259) -> float_type, {32, 1}, {1, 1}: 0.0147993ms, 0.130488% main:@2261 = load[offset=915200,end=1212672](main:@1) -> float_type, {32, 2324}, {2324, 1}: 0.00061124ms, 0.00538939% main:@2262 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=18592,local=256,](main:@2247,main:@2248,main:@2251,main:@2237,main:@2261) -> float_type, {32, 2324}, {2324, 1}: 0.0178427ms, 0.157322% main:@2263 = slice[axes={1},starts={0},ends={4070}](main:@2244) -> float_type, {32, 4070}, {4508, 1}: 0.0007364ms, 0.00649294% main:@2264 = load[offset=1346432,end=1867392](main:@1) -> float_type, {32, 4070}, {4070, 1}: 0.0004666ms, 0.00411408% main:@2265 = gpu::code_object[code_object=21808,symbol_name=leaky_relu_noop_noop_noop_noop_noop_noop_concat_rsqrt_pow_kernel,global=18605,local=256,](main:@2262,main:@2250,main:@2254,main:@2246,main:@2260,main:@2256,main:@2258,main:@2263,main:@2264) -> float_type, {32, 4070}, {4070, 1}: 0.0357872ms, 0.315541% main:@2266 = hip::hip_copy_literal[id=main:@literal:24] -> float_type, {4070, 2130}, {2130, 1}: 0.00070716ms, 0.00623513% main:@2267 = hip::hip_copy_literal[id=main:@literal:1] -> float_type, {32, 2130}, {2130, 1}: 0.0006567ms, 0.00579022% main:@2268 = load[offset=1073792,end=1346432](main:@1) -> float_type, {32, 2130}, {2130, 1}: 0.00046232ms, 0.00407634% main:@2269 = gpu::gemm[alpha=1,beta=1,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@2265,main:@2266,main:@2267,main:@2268) -> float_type, {32, 2130}, {2130, 1}: 0.075137ms, 0.662494% main:@2270 = hip::hip_copy_literal[id=main:@literal:114] -> float_type, {128, 2985}, {2985, 1}: 0.00072446ms, 0.00638767% main:@2271 = slice[axes={1},starts={768},ends={896}](main:@765) -> float_type, {32, 128}, {896, 1}: 0.00058112ms, 0.00512382% main:@2272 = load[offset=691712,end=1073792](main:@1) -> float_type, {32, 2985}, {2985, 1}: 0.00048178ms, 0.00424792% main:@2273 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@2271,main:@2270,main:@2272) -> float_type, {32, 2985}, {2985, 1}: 0.0230191ms, 0.202962% main:@2274 = hip::hip_copy_literal[id=main:@literal:92] -> float_type, {426, 426}, {426, 1}: 0.00078626ms, 0.00693257% main:@2275 = slice[axes={1},starts={852},ends={1278}](main:@2269) -> float_type, {32, 426}, {2130, 1}: 0.00061824ms, 0.00545111% main:@2276 = load[offset=1400960,end=1455488](main:@1) -> float_type, {32, 426}, {426, 1}: 0.00047672ms, 0.00420331% main:@2277 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@2275,main:@2274,main:@2276) -> float_type, {32, 426}, {426, 1}: 0.0297592ms, 0.262391% main:@2278 = hip::hip_copy_literal[id=main:@literal:30] -> float_type, {32, 2}, {2, 1}: 0.00078228ms, 0.00689748% main:@2279 = load[offset=1346432,end=1400960](main:@1) -> float_type, {32, 426}, {426, 1}: 0.00050616ms, 0.00446289% main:@2280 = gpu::code_object[code_object=13544,symbol_name=sub_sub_mul_rsqrt_sub_kernel,global=6816,local=256,](main:@2275,main:@2277,main:@2279) -> float_type, {32, 426}, {426, 1}: 0.0145575ms, 0.128356% main:@2281 = slice[axes={1},starts={426},ends={852}](main:@2269) -> float_type, {32, 426}, {2130, 1}: 0.00072068ms, 0.00635434% main:@2282 = load[offset=1400960,end=1455488](main:@1) -> float_type, {32, 426}, {426, 1}: 0.00046752ms, 0.00412219% main:@2283 = slice[axes={1},starts={1704},ends={2130}](main:@2269) -> float_type, {32, 426}, {2130, 1}: 0.00057716ms, 0.0050889% main:@2284 = gpu::code_object[code_object=13536,symbol_name=div_mul_neg_pow_kernel,global=6816,local=256,](main:@2281,main:@2283,main:@2282) -> float_type, {32, 426}, {426, 1}: 0.0151397ms, 0.133489% main:@2285 = slice[axes={1},starts={0},ends={426}](main:@2269) -> float_type, {32, 426}, {2130, 1}: 0.00072044ms, 0.00635222% main:@2286 = load[offset=1455488,end=1510016](main:@1) -> float_type, {32, 426}, {426, 1}: 0.00049678ms, 0.00438018% main:@2287 = slice[axes={1},starts={1278},ends={1704}](main:@2269) -> float_type, {32, 426}, {2130, 1}: 0.00056786ms, 0.0050069% main:@2288 = gpu::code_object[code_object=13536,symbol_name=add_pow_log_pow_kernel,global=6816,local=256,](main:@2287,main:@2285,main:@2286) -> float_type, {32, 426}, {426, 1}: 0.017394ms, 0.153365% main:@2289 = load[offset=1510144,end=1510272](main:@1) -> float_type, {32, 1}, {1, 1}: 0.00060886ms, 0.00536841% main:@2290 = gpu::code_object[code_object=13544,symbol_name=reduce_mean_mul_neg_kernel,global=8192,local=256,](main:@2284,main:@2289) -> float_type, {32, 1}, {1, 1}: 0.0142881ms, 0.12598% main:@2291 = load[offset=1510016,end=1510144](main:@1) -> float_type, {32, 1}, {1, 1}: 0.00063124ms, 0.00556573% main:@2292 = gpu::code_object[code_object=13560,symbol_name=reduce_mean_add_sigmoid_sub_kernel,global=8192,local=256,](main:@2288,main:@2291) -> float_type, {32, 1}, {1, 1}: 0.0144587ms, 0.127484% main:@2293 = load[offset=1510272,end=1510400](main:@1) -> float_type, {32, 1}, {1, 1}: 0.00060776ms, 0.00535871% main:@2294 = gpu::code_object[code_object=13544,symbol_name=reduce_prod_mul_log_kernel,global=8192,local=256,](main:@2280,main:@2293) -> float_type, {32, 1}, {1, 1}: 0.0142364ms, 0.125525% main:@2295 = load[offset=1892224,end=2110336](main:@1) -> float_type, {32, 1704}, {1704, 1}: 0.00063726ms, 0.00561881% main:@2296 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=13632,local=256,](main:@2287,main:@2285,main:@2281,main:@2275,main:@2295) -> float_type, {32, 1704}, {1704, 1}: 0.016658ms, 0.146876% main:@2297 = load[offset=1073792,end=1074048](main:@1) -> float_type, {32, 2}, {2, 1}: 0.00062586ms, 0.0055183% main:@2298 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=32,local=256,](main:@2290,main:@2294,main:@2297) -> float_type, {32, 2}, {2, 1}: 0.0132643ms, 0.116953% main:@2299 = load[offset=1510144,end=1892224](main:@1) -> float_type, {32, 2985}, {2985, 1}: 0.00059138ms, 0.00521428% main:@2300 = gpu::code_object[code_object=21792,symbol_name=leaky_relu_noop_noop_noop_noop_add_concat_log_pow_kernel,global=15920,local=256,](main:@2296,main:@2288,main:@2284,main:@2280,main:@2292,main:@2278,main:@2298,main:@2273,main:@2299) -> float_type, {32, 2985}, {2985, 1}: 0.0332408ms, 0.293089% main:@2301 = hip::hip_copy_literal[id=main:@literal:27] -> float_type, {2985, 856}, {856, 1}: 0.0007404ms, 0.00652821% main:@2302 = hip::hip_copy_literal[id=main:@literal:29] -> float_type, {32, 856}, {856, 1}: 0.00064896ms, 0.00572197% main:@2303 = load[offset=691712,end=801280](main:@1) -> float_type, {32, 856}, {856, 1}: 0.00047402ms, 0.0041795% main:@2304 = gpu::gemm[alpha=1,beta=1,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@2300,main:@2301,main:@2302,main:@2303) -> float_type, {32, 856}, {856, 1}: 0.0650268ms, 0.573351% main:@2305 = hip::hip_copy_literal[id=main:@literal:43] -> float_type, {128, 1501}, {1501, 1}: 0.00076154ms, 0.00671461% main:@2306 = slice[axes={1},starts={0},ends={128}](main:@765) -> float_type, {32, 128}, {896, 1}: 0.00058914ms, 0.00519453% main:@2307 = load[offset=801280,end=993408](main:@1) -> float_type, {32, 1501}, {1501, 1}: 0.00048304ms, 0.00425903% main:@2308 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@2306,main:@2305,main:@2307) -> float_type, {32, 1501}, {1501, 1}: 0.0242198ms, 0.21355% main:@2309 = hip::hip_copy_literal[id=main:@literal:161] -> float_type, {214, 214}, {214, 1}: 0.0007695ms, 0.00678479% main:@2310 = slice[axes={1},starts={642},ends={856}](main:@2304) -> float_type, {32, 214}, {856, 1}: 0.00056686ms, 0.00499809% main:@2311 = load[offset=1020800,end=1048192](main:@1) -> float_type, {32, 214}, {214, 1}: 0.00048256ms, 0.0042548% main:@2312 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@2310,main:@2309,main:@2311) -> float_type, {32, 214}, {214, 1}: 0.0258378ms, 0.227816% main:@2313 = hip::hip_copy_literal[id=main:@literal:126] -> float_type, {1, 214}, {214, 1}: 0.00079204ms, 0.00698353% main:@2314 = load[offset=993408,end=1020800](main:@1) -> float_type, {32, 214}, {214, 1}: 0.00049738ms, 0.00438547% main:@2315 = multibroadcast[out_lens={32, 214}](main:@2313) -> float_type, {32, 214}, {0, 1}: 0.00073146ms, 0.00644939% main:@2316 = gpu::code_object[code_object=17648,symbol_name=pow_sub_div_pow_log_div_kernel,global=3424,local=256,](main:@2312,main:@2315,main:@2310,main:@2314) -> float_type, {32, 214}, {214, 1}: 0.0200982ms, 0.177209% main:@2317 = hip::hip_copy_literal[id=main:@literal:148] -> float_type, {1}, {0}: 0.00077276ms, 0.00681354% main:@2318 = multibroadcast[out_lens={214}](main:@2317) -> float_type, {214}, {0}: 0.00058664ms, 0.00517249% main:@2319 = load[offset=1020800,end=1048192](main:@1) -> float_type, {32, 214}, {214, 1}: 0.00048506ms, 0.00427684% main:@2320 = slice[axes={1},starts={428},ends={642}](main:@2304) -> float_type, {32, 214}, {856, 1}: 0.00058922ms, 0.00519524% main:@2321 = slice[axes={1},starts={0},ends={214}](main:@2304) -> float_type, {32, 214}, {856, 1}: 0.00057056ms, 0.00503071% main:@2322 = multibroadcast[out_lens={32, 214}](main:@2318) -> float_type, {32, 214}, {0, 0}: 0.00068464ms, 0.00603657% main:@2323 = gpu::code_object[code_object=13536,symbol_name=add_add_rsqrt_pow_kernel,global=3424,local=256,](main:@2321,main:@2320,main:@2322,main:@2319) -> float_type, {32, 214}, {214, 1}: 0.0151904ms, 0.133936% main:@2324 = slice[axes={1},starts={214},ends={428}](main:@2304) -> float_type, {32, 214}, {856, 1}: 0.0007673ms, 0.00676539% main:@2325 = load[offset=1048192,end=1075584](main:@1) -> float_type, {32, 214}, {214, 1}: 0.0004666ms, 0.00411408% main:@2326 = gpu::code_object[code_object=13536,symbol_name=add_add_neg_add_kernel,global=3424,local=256,](main:@2324,main:@2322,main:@2325) -> float_type, {32, 214}, {214, 1}: 0.0141775ms, 0.125005% main:@2327 = hip::hip_copy_literal[id=main:@literal:0] -> float_type, {32, 400}, {400, 1}: 0.0007097ms, 0.00625753% main:@2328 = hip::hip_copy_literal[id=main:@literal:17] -> float_type, {1501, 400}, {400, 1}: 0.00060248ms, 0.00531215% main:@2329 = load[offset=1075584,end=1075712](main:@1) -> float_type, {32, 1}, {1, 1}: 0.00045736ms, 0.00403261% main:@2330 = gpu::code_object[code_object=13552,symbol_name=reduce_prod_pow_neg_sub_kernel,global=4096,local=128,](main:@2316,main:@2329) -> float_type, {32, 1}, {1, 1}: 0.0150184ms, 0.132419% main:@2331 = load[offset=1075712,end=1075840](main:@1) -> float_type, {32, 1}, {1, 1}: 0.00061644ms, 0.00543524% main:@2332 = gpu::code_object[code_object=13552,symbol_name=reduce_prod_add_log_div_kernel,global=4096,local=128,](main:@2323,main:@2331) -> float_type, {32, 1}, {1, 1}: 0.0143905ms, 0.126883% main:@2333 = load[offset=1075840,end=1075968](main:@1) -> float_type, {32, 1}, {1, 1}: 0.0006229ms, 0.0054922% main:@2334 = gpu::code_object[code_object=13552,symbol_name=reduce_prod_mul_rsqrt_add_kernel,global=4096,local=128,](main:@2326,main:@2333) -> float_type, {32, 1}, {1, 1}: 0.0142605ms, 0.125737% main:@2335 = load[offset=1268096,end=1377664](main:@1) -> float_type, {32, 856}, {856, 1}: 0.00060692ms, 0.0053513% main:@2336 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=6848,local=256,](main:@2324,main:@2324,main:@2321,main:@2310,main:@2335) -> float_type, {32, 856}, {856, 1}: 0.0156751ms, 0.13821% main:@2337 = load[offset=1075968,end=1268096](main:@1) -> float_type, {32, 1501}, {1501, 1}: 0.0006168ms, 0.00543841% main:@2338 = gpu::code_object[code_object=17720,symbol_name=leaky_relu_noop_noop_noop_noop_noop_noop_concat_log_add_pow_kernel,global=6861,local=256,](main:@2336,main:@2326,main:@2323,main:@2316,main:@2334,main:@2332,main:@2330,main:@2308,main:@2337) -> float_type, {32, 1501}, {1501, 1}: 0.0320755ms, 0.282815% main:@2339 = load[offset=781696,end=832896](main:@1) -> float_type, {32, 400}, {400, 1}: 0.0006064ms, 0.00534672% main:@2340 = gpu::gemm[alpha=1,beta=1,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@2338,main:@2328,main:@2327,main:@2339) -> float_type, {32, 400}, {400, 1}: 0.0313461ms, 0.276383% main:@2341 = hip::hip_copy_literal[id=main:@literal:52] -> float_type, {128, 703}, {703, 1}: 0.00080178ms, 0.00706941% main:@2342 = slice[axes={1},starts={256},ends={384}](main:@765) -> float_type, {32, 128}, {896, 1}: 0.00059882ms, 0.00527988% main:@2343 = load[offset=691712,end=781696](main:@1) -> float_type, {32, 703}, {703, 1}: 0.00045824ms, 0.00404037% main:@2344 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@2342,main:@2341,main:@2343) -> float_type, {32, 703}, {703, 1}: 0.0209684ms, 0.184882% main:@2345 = hip::hip_copy_literal[id=main:@literal:3] -> float_type, {100, 100}, {100, 1}: 0.00076726ms, 0.00676504% main:@2346 = load[offset=858496,end=871296](main:@1) -> float_type, {32, 100}, {100, 1}: 0.00047284ms, 0.0041691% main:@2347 = slice[axes={1},starts={200},ends={300}](main:@2340) -> float_type, {32, 100}, {400, 1}: 0.00057416ms, 0.00506245% main:@2348 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@2347,main:@2345,main:@2346) -> float_type, {32, 100}, {100, 1}: 0.0220558ms, 0.194469% main:@2349 = slice[axes={1},starts={300},ends={400}](main:@2340) -> float_type, {32, 100}, {400, 1}: 0.00073704ms, 0.00649859% main:@2350 = slice[axes={1},starts={0},ends={100}](main:@2340) -> float_type, {32, 100}, {400, 1}: 0.00056908ms, 0.00501766% main:@2351 = load[offset=832896,end=845696](main:@1) -> float_type, {32, 100}, {100, 1}: 0.0004912ms, 0.00433098% main:@2352 = gpu::code_object[code_object=13536,symbol_name=add_add_rsqrt_add_kernel,global=1600,local=256,](main:@2349,main:@2350,main:@2351) -> float_type, {32, 100}, {100, 1}: 0.0143294ms, 0.126344% main:@2353 = load[offset=845696,end=858496](main:@1) -> float_type, {32, 100}, {100, 1}: 0.00060738ms, 0.00535536% main:@2354 = gpu::code_object[code_object=13544,symbol_name=sub_sub_pow_log_pow_kernel,global=1600,local=256,](main:@2347,main:@2348,main:@2353) -> float_type, {32, 100}, {100, 1}: 0.0172712ms, 0.152282% main:@2355 = slice[axes={1},starts={100},ends={200}](main:@2340) -> float_type, {32, 100}, {400, 1}: 0.00070978ms, 0.00625823% main:@2356 = load[offset=858496,end=871296](main:@1) -> float_type, {32, 100}, {100, 1}: 0.0004767ms, 0.00420313% main:@2357 = gpu::code_object[code_object=13536,symbol_name=add_add_neg_add_kernel,global=1600,local=256,](main:@2355,main:@2350,main:@2356) -> float_type, {32, 100}, {100, 1}: 0.0142239ms, 0.125415% main:@2358 = hip::hip_copy_literal[id=main:@literal:9] -> float_type, {32, 212}, {212, 1}: 0.000738ms, 0.00650705% main:@2359 = hip::hip_copy_literal[id=main:@literal:31] -> float_type, {703, 212}, {212, 1}: 0.00057248ms, 0.00504764% main:@2360 = load[offset=871424,end=871552](main:@1) -> float_type, {32, 1}, {1, 1}: 0.00047394ms, 0.0041788% main:@2361 = gpu::code_object[code_object=14256,symbol_name=reduce_mean_add_sigmoid_kernel,global=2048,local=64,](main:@2352,main:@2360) -> float_type, {32, 1}, {1, 1}: 0.0144008ms, 0.126974% main:@2362 = load[offset=871296,end=871424](main:@1) -> float_type, {32, 1}, {1, 1}: 0.0006188ms, 0.00545605% main:@2363 = gpu::code_object[code_object=14256,symbol_name=reduce_mean_pow_log_pow_kernel,global=2048,local=64,](main:@2354,main:@2362) -> float_type, {32, 1}, {1, 1}: 0.0150965ms, 0.133108% main:@2364 = load[offset=922752,end=922880](main:@1) -> float_type, {32, 1}, {1, 1}: 0.00060372ms, 0.00532309% main:@2365 = gpu::code_object[code_object=14256,symbol_name=reduce_mean_mul_rsqrt_kernel,global=2048,local=64,](main:@2357,main:@2364) -> float_type, {32, 1}, {1, 1}: 0.0142049ms, 0.125246% main:@2366 = load[offset=871552,end=922752](main:@1) -> float_type, {32, 400}, {400, 1}: 0.00060326ms, 0.00531903% main:@2367 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=3200,local=256,](main:@2349,main:@2350,main:@2355,main:@2347,main:@2366) -> float_type, {32, 400}, {400, 1}: 0.0155152ms, 0.136799% main:@2368 = load[offset=781696,end=781952](main:@1) -> float_type, {32, 2}, {2, 1}: 0.00059564ms, 0.00525184% main:@2369 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=32,local=256,](main:@2361,main:@2365,main:@2368) -> float_type, {32, 2}, {2, 1}: 0.0131239ms, 0.115715% main:@2370 = load[offset=922752,end=1012736](main:@1) -> float_type, {32, 703}, {703, 1}: 0.00060282ms, 0.00531515% main:@2371 = gpu::code_object[code_object=13600,symbol_name=leaky_relu_noop_noop_noop_add_noop_concat_log_div_kernel,global=3749,local=256,](main:@2367,main:@2352,main:@2357,main:@2354,main:@2278,main:@2369,main:@2363,main:@2344,main:@2370) -> float_type, {32, 703}, {703, 1}: 0.0220279ms, 0.194224% main:@2372 = load[offset=691712,end=718848](main:@1) -> float_type, {32, 212}, {212, 1}: 0.0006028ms, 0.00531497% main:@2373 = gpu::gemm[alpha=1,beta=1,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@2371,main:@2359,main:@2358,main:@2372) -> float_type, {32, 212}, {212, 1}: 0.0371398ms, 0.327467% main:@2374 = hip::hip_copy_literal[id=main:@literal:38] -> float_type, {128, 256}, {256, 1}: 0.00077296ms, 0.0068153% main:@2375 = slice[axes={1},starts={640},ends={768}](main:@765) -> float_type, {32, 128}, {896, 1}: 0.00059302ms, 0.00522874% main:@2376 = load[offset=718848,end=751616](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00047054ms, 0.00414882% main:@2377 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@2375,main:@2374,main:@2376) -> float_type, {32, 256}, {256, 1}: 0.0206289ms, 0.181888% main:@2378 = hip::hip_copy_literal[id=main:@literal:81] -> float_type, {128, 167}, {167, 1}: 0.00077912ms, 0.00686961% main:@2379 = slice[axes={1},starts={512},ends={640}](main:@765) -> float_type, {32, 128}, {896, 1}: 0.0006204ms, 0.00547016% main:@2380 = load[offset=751616,end=772992](main:@1) -> float_type, {32, 167}, {167, 1}: 0.0004661ms, 0.00410967% main:@2381 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@2379,main:@2378,main:@2380) -> float_type, {32, 167}, {167, 1}: 0.0215609ms, 0.190106% main:@2382 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](nP9Yv1iZ8f:0) -> float_type, {32, 1}, {9, 1}: 0.0008558ms, 0.00754571% main:@2383 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](pQsi1GNxvj:0) -> float_type, {32, 1}, {5, 1}: 0.00075512ms, 0.006658% main:@2384 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](EsROupwkSG:0) -> float_type, {32, 1}, {13, 1}: 0.00084746ms, 0.00747218% main:@2385 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](EKpP4xMXu8:0) -> float_type, {32, 1}, {17, 1}: 0.00067076ms, 0.00591419% main:@2386 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](eQquakfWBY:0) -> float_type, {32, 1}, {17, 1}: 0.00066564ms, 0.00586904% main:@2387 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](F6DhiS9CN5:0) -> float_type, {32, 1}, {9, 1}: 0.00069072ms, 0.00609018% main:@2388 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](EUciqFuptg:0) -> float_type, {32, 1}, {17, 1}: 0.00083312ms, 0.00734574% main:@2389 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](JCVMzpWyri:0) -> float_type, {32, 1}, {17, 1}: 0.00066838ms, 0.0058932% main:@2390 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](FWhBxj4Tun:0) -> float_type, {32, 1}, {9, 1}: 0.0006407ms, 0.00564914% main:@2391 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](FcmWLZhDgE:0) -> float_type, {32, 1}, {57, 1}: 0.00065848ms, 0.00580591% main:@2392 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](uFSOc5Chp2:0) -> float_type, {32, 1}, {481, 1}: 0.00082792ms, 0.00729989% main:@2393 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](HRGzlnEkXQ:0) -> float_type, {32, 1}, {41, 1}: 0.00067572ms, 0.00595792% main:@2394 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](HZ0l3MmcPY:0) -> float_type, {32, 1}, {17, 1}: 0.00064294ms, 0.00566889% main:@2395 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](YbIVz8l1un:0) -> float_type, {32, 1}, {45, 1}: 0.00063302ms, 0.00558143% main:@2396 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](OfNJgEDtw2:0) -> float_type, {32, 1}, {13, 1}: 0.00079868ms, 0.00704208% main:@2397 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](HmATr3gKsW:0) -> float_type, {32, 1}, {9, 1}: 0.00064972ms, 0.00572867% main:@2398 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](Hnjq487SIY:0) -> float_type, {32, 1}, {13, 1}: 0.00069024ms, 0.00608595% main:@2399 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](omr2g4XaH5:0) -> float_type, {32, 1}, {17, 1}: 0.00068268ms, 0.00601929% main:@2400 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](NJBt1ubXFW:0) -> float_type, {32, 1}, {5, 1}: 0.00076726ms, 0.00676504% main:@2401 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](mgnGNH4Mz3:0) -> float_type, {32, 1}, {49, 1}: 0.00063478ms, 0.00559695% main:@2402 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](IMdFk7D16X:0) -> float_type, {32, 1}, {9, 1}: 0.00063128ms, 0.00556609% main:@2403 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](IxpbBg7JTO:0) -> float_type, {32, 1}, {5, 1}: 0.00063728ms, 0.00561899% main:@2404 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](jznORDJr0N:0) -> float_type, {32, 1}, {45, 1}: 0.00076902ms, 0.00678056% main:@2405 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](mNgQ4ohWqU:0) -> float_type, {32, 1}, {9, 1}: 0.0006555ms, 0.00577964% main:@2406 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](J3H2EjAXt8:0) -> float_type, {32, 1}, {5, 1}: 0.00066842ms, 0.00589356% main:@2407 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](JSPsGophtx:0) -> float_type, {32, 1}, {25, 1}: 0.00066938ms, 0.00590202% main:@2408 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](TkHQvxKeD6:0) -> float_type, {32, 1}, {9, 1}: 0.00080296ms, 0.00707981% main:@2409 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](sZdFJI9nwg:0) -> float_type, {32, 1}, {41, 1}: 0.00067916ms, 0.00598825% main:@2410 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](JaHTUxB6zL:0) -> float_type, {32, 1}, {17, 1}: 0.00066444ms, 0.00585846% main:@2411 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](JSWhLK4Agx:0) -> float_type, {32, 1}, {25, 1}: 0.00066866ms, 0.00589567% main:@2412 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](U6GqhP8k7I:0) -> float_type, {32, 1}, {9, 1}: 0.00079432ms, 0.00700363% main:@2413 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](KCNfS4GlpP:0) -> float_type, {32, 1}, {9, 1}: 0.0006393ms, 0.0056368% main:@2414 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](eQI8ikEMgP:0) -> float_type, {32, 1}, {41, 1}: 0.00066914ms, 0.0058999% main:@2415 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](LKAvFme9ZS:0) -> float_type, {32, 1}, {49, 1}: 0.00059638ms, 0.00525837% main:@2416 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](pY0t1rI5WU:0) -> float_type, {32, 1}, {41, 1}: 0.0007821ms, 0.00689589% main:@2417 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](LMobYAd5ru:0) -> float_type, {32, 1}, {17, 1}: 0.00070932ms, 0.00625418% main:@2418 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](iCbkMrmXBh:0) -> float_type, {32, 1}, {9, 1}: 0.00063922ms, 0.00563609% main:@2419 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](NLt5JnRMjG:0) -> float_type, {32, 1}, {5, 1}: 0.00065324ms, 0.00575971% main:@2420 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](NXOSdQVPs0:0) -> float_type, {32, 1}, {25, 1}: 0.00080716ms, 0.00711685% main:@2421 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](ecPTQUALkq:0) -> float_type, {32, 1}, {5, 1}: 0.00063172ms, 0.00556997% main:@2422 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](O5aJeUQh2G:0) -> float_type, {32, 1}, {17, 1}: 0.00067392ms, 0.00594205% main:@2423 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](cHzIDJm6us:0) -> float_type, {32, 1}, {33, 1}: 0.0006212ms, 0.00547721% main:@2424 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](gdck97AK4w:0) -> float_type, {32, 1}, {17, 1}: 0.00076826ms, 0.00677386% main:@2425 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](Ouh1XpPZaj:0) -> float_type, {32, 1}, {9, 1}: 0.00064088ms, 0.00565073% main:@2426 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](lyiDxfzZg3:0) -> float_type, {32, 1}, {701, 1}: 0.00061762ms, 0.00544564% main:@2427 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](Pu1O5qFxY6:0) -> float_type, {32, 1}, {17, 1}: 0.00064476ms, 0.00568494% main:@2428 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](ry2xad1vc9:0) -> float_type, {32, 1}, {25, 1}: 0.0007861ms, 0.00693116% main:@2429 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](qd1komWTn4:0) -> float_type, {32, 1}, {25, 1}: 0.00061744ms, 0.00544406% main:@2430 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](Qp3K1PHvte:0) -> float_type, {32, 1}, {69, 1}: 0.00063054ms, 0.00555956% main:@2431 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](dz1saATctE:0) -> float_type, {32, 1}, {9, 1}: 0.00063468ms, 0.00559607% main:@2432 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](nltbg9ZWQX:0) -> float_type, {32, 1}, {17, 1}: 0.00078908ms, 0.00695743% main:@2433 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](RpITWntAji:0) -> float_type, {32, 1}, {17, 1}: 0.00064906ms, 0.00572286% main:@2434 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](uhcrWfUe6M:0) -> float_type, {32, 1}, {9, 1}: 0.00060316ms, 0.00531815% main:@2435 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](XPcHGKbop2:0) -> float_type, {32, 1}, {13, 1}: 0.00063034ms, 0.0055578% main:@2436 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](Tej5M3O6As:0) -> float_type, {32, 1}, {17, 1}: 0.00077452ms, 0.00682905% main:@2437 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](Tg4u0GfnFb:0) -> float_type, {32, 1}, {25, 1}: 0.0006456ms, 0.00569235% main:@2438 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](TSqGwrHeo4:0) -> float_type, {32, 1}, {5, 1}: 0.00063362ms, 0.00558672% main:@2439 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](khJcIxW3zv:0) -> float_type, {32, 1}, {9, 1}: 0.00064632ms, 0.0056987% main:@2440 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](Uc42tiuavR:0) -> float_type, {32, 1}, {9, 1}: 0.00075844ms, 0.00668727% main:@2441 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](pIb4NdDSn5:0) -> float_type, {32, 1}, {9, 1}: 0.00062282ms, 0.00549149% ebEFYOy65u:0 = @param:ebEFYOy65u:0 -> float_type, {32, 1}, {1, 1}: 0.00064674ms, 0.0057024% iUEBSr6gOG:0 = @param:iUEBSr6gOG:0 -> float_type, {32, 1}, {1, 1}: 0.00050002ms, 0.00440875% qxzgXpiLdl:0 = @param:qxzgXpiLdl:0 -> float_type, {32, 1}, {1, 1}: 0.00064828ms, 0.00571598% 9NcxXsmyqf:0 = @param:9NcxXsmyqf:0 -> float_type, {32, 1}, {1, 1}: 0.00056186ms, 0.004954% WqeN5KGIv9:0 = @param:WqeN5KGIv9:0 -> float_type, {32, 1}, {1, 1}: 0.00051652ms, 0.00455423% 7XiFOysrpI:0 = @param:7XiFOysrpI:0 -> float_type, {32, 1}, {1, 1}: 0.00057394ms, 0.00506051% vtIF8HqRG6:0 = @param:vtIF8HqRG6:0 -> float_type, {32, 1}, {1, 1}: 0.00060542ms, 0.00533808% NG35MnhTIF:0 = @param:NG35MnhTIF:0 -> float_type, {32, 1}, {1, 1}: 0.0005698ms, 0.00502401% lqOmNvz5fj:0 = @param:lqOmNvz5fj:0 -> float_type, {32, 1}, {1, 1}: 0.00057606ms, 0.0050792% XPpbueLMOZ:0 = @param:XPpbueLMOZ:0 -> float_type, {32, 1}, {1, 1}: 0.00057844ms, 0.00510019% bJxY1IhSVn:0 = @param:bJxY1IhSVn:0 -> float_type, {32, 1}, {1, 1}: 0.00052484ms, 0.00462759% iE352qtAOx:0 = @param:iE352qtAOx:0 -> float_type, {32, 1}, {1, 1}: 0.00058556ms, 0.00516297% main:@2454 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](4gFa5o2YsJ:0) -> float_type, {32, 1}, {9, 1}: 0.00082478ms, 0.0072722% main:@2455 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](5yColzZ8K3:0) -> float_type, {32, 1}, {5, 1}: 0.0006377ms, 0.00562269% main:@2456 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](6HSQqXdJzb:0) -> float_type, {32, 1}, {45, 1}: 0.00060374ms, 0.00532326% main:@2457 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](6Wh2Jm7GHC:0) -> float_type, {32, 1}, {5, 1}: 0.00063516ms, 0.0056003% main:@2458 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](72zWmh93Bo:0) -> float_type, {32, 1}, {17, 1}: 0.00079064ms, 0.00697119% main:@2459 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](6XMrIsBjLY:0) -> float_type, {32, 1}, {13, 1}: 0.00067702ms, 0.00596938% main:@2460 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](ifP6Ceko5T:0) -> float_type, {32, 1}, {73, 1}: 0.00062198ms, 0.00548409% main:@2461 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](AB6lekmMxa:0) -> float_type, {32, 1}, {25, 1}: 0.00066768ms, 0.00588703% main:@2462 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](89nc05gQza:0) -> float_type, {32, 1}, {9, 1}: 0.00078298ms, 0.00690365% main:@2463 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](8Dv01FAoNM:0) -> float_type, {32, 1}, {17, 1}: 0.00063944ms, 0.00563803% main:@2464 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](8iOST90kWM:0) -> float_type, {32, 1}, {9, 1}: 0.0006025ms, 0.00531233% main:@2465 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](8vM5p4RzXd:0) -> float_type, {32, 1}, {13, 1}: 0.00064112ms, 0.00565285% main:@2466 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](UMO0o14RuI:0) -> float_type, {32, 1}, {9, 1}: 0.00080902ms, 0.00713325% main:@2467 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](9dgeU4bKka:0) -> float_type, {32, 1}, {17, 1}: 0.0006873ms, 0.00606002% main:@2468 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](9mfMQZIXVH:0) -> float_type, {32, 1}, {9, 1}: 0.00065694ms, 0.00579233% main:@2469 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](AvlwbHMnuE:0) -> float_type, {32, 1}, {17, 1}: 0.00065298ms, 0.00575742% main:@2470 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](EhUjmCTaGW:0) -> float_type, {32, 1}, {9, 1}: 0.00079172ms, 0.00698071% main:@2471 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](Cw59eOo7M1:0) -> float_type, {32, 1}, {9, 1}: 0.00067482ms, 0.00594999% main:@2472 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](AVLcBpWeuo:0) -> float_type, {32, 1}, {21, 1}: 0.00061158ms, 0.00539239% main:@2473 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](nyLxrsTmOB:0) -> float_type, {32, 1}, {17, 1}: 0.00066558ms, 0.00586851% main:@2474 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](m2YscaoDjP:0) -> float_type, {32, 1}, {17, 1}: 0.00080334ms, 0.00708316% main:@2475 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](BYyjkPCX0m:0) -> float_type, {32, 1}, {13, 1}: 0.0006566ms, 0.00578934% main:@2476 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](oQ5OcjZJe8:0) -> float_type, {32, 1}, {17, 1}: 0.00063394ms, 0.00558954% main:@2477 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](qfwYOHde3u:0) -> float_type, {32, 1}, {17, 1}: 0.00067ms, 0.00590749% main:@2478 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](isCn4mEOSt:0) -> float_type, {32, 1}, {49, 1}: 0.0007773ms, 0.00685357% main:@2479 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](r1WGP7Xe9H:0) -> float_type, {32, 1}, {13, 1}: 0.00061538ms, 0.00542589% main:@2480 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](CHAYRkBvIo:0) -> float_type, {32, 1}, {41, 1}: 0.00064658ms, 0.00570099% main:@2481 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](tIGjbcEuDd:0) -> float_type, {32, 1}, {9, 1}: 0.00065638ms, 0.0057874% main:@2482 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](vZDAiUHErW:0) -> float_type, {32, 1}, {17, 1}: 0.00079122ms, 0.0069763% main:@2483 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](Dy3rwVmn1A:0) -> float_type, {32, 1}, {41, 1}: 0.00062994ms, 0.00555427% main:@2484 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](E1ZkVIXqS5:0) -> float_type, {32, 1}, {9, 1}: 0.0006729ms, 0.00593306% main:@2485 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](qzd23aNGIT:0) -> float_type, {32, 1}, {17, 1}: 0.00067292ms, 0.00593323% main:@2486 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](EZBIAqn6LS:0) -> float_type, {32, 1}, {13, 1}: 0.00079174ms, 0.00698089% xGCyW9icVP:0 = @param:xGCyW9icVP:0 -> float_type, {32, 1}, {1, 1}: 0.00055128ms, 0.00486072% lhbKMrO1my:0 = @param:lhbKMrO1my:0 -> float_type, {32, 1}, {1, 1}: 0.0005565ms, 0.00490674% StHFqj5edV:0 = @param:StHFqj5edV:0 -> float_type, {32, 1}, {1, 1}: 0.00050888ms, 0.00448687% main:@2490 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](0cjwsL5ldh:0) -> float_type, {32, 1}, {33, 1}: 0.0007936ms, 0.00699729% main:@2491 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](l5WSX3VvMK:0) -> float_type, {32, 1}, {25, 1}: 0.00067642ms, 0.00596409% main:@2492 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](pH5srbufJ1:0) -> float_type, {32, 1}, {17, 1}: 0.00066336ms, 0.00584894% main:@2493 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](lDhFNYzbxR:0) -> float_type, {32, 1}, {17, 1}: 0.00066024ms, 0.00582143% main:@2494 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](magwfJSnUu:0) -> float_type, {32, 1}, {9, 1}: 0.00082102ms, 0.00723905% main:@2495 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](w04Keh2FJu:0) -> float_type, {32, 1}, {13, 1}: 0.00067992ms, 0.00599495% main:@2496 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](xVErSbJ0hg:0) -> float_type, {32, 1}, {9, 1}: 0.0006782ms, 0.00597979% main:@2497 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](fMiQsYNnet:0) -> float_type, {32, 1}, {5, 1}: 0.00066022ms, 0.00582125% main:@2498 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](gqk5byC8aJ:0) -> float_type, {32, 1}, {5, 1}: 0.00084246ms, 0.00742809% main:@2499 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](lQmygxs5aD:0) -> float_type, {32, 1}, {13, 1}: 0.00066386ms, 0.00585335% main:@2500 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](sjKQJGby31:0) -> float_type, {32, 1}, {9, 1}: 0.00065994ms, 0.00581879% main:@2501 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](l4bReWtsMB:0) -> float_type, {32, 1}, {25, 1}: 0.0006683ms, 0.0058925% main:@2502 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](tWcngAq7hV:0) -> float_type, {32, 1}, {13, 1}: 0.00082304ms, 0.00725686% main:@2503 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](zZHaVOckpw:0) -> float_type, {32, 1}, {13, 1}: 0.00070568ms, 0.00622208% main:@2504 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](mTMtDEo3r8:0) -> float_type, {32, 1}, {9, 1}: 0.00067918ms, 0.00598843% main:@2505 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](WDLoyKsiA7:0) -> float_type, {32, 1}, {33, 1}: 0.00068642ms, 0.00605226% c2MuYjWE3S:0 = @param:c2MuYjWE3S:0 -> float_type, {32, 1}, {1, 1}: 0.00050428ms, 0.00444631% 7tFB8mLWUr:0 = @param:7tFB8mLWUr:0 -> float_type, {32, 1}, {1, 1}: 0.00050318ms, 0.00443661% FTIMrcfako:0 = @param:FTIMrcfako:0 -> float_type, {32, 1}, {1, 1}: 0.00067148ms, 0.00592054% main:@2509 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](gsP96ZTbn0:0) -> float_type, {32, 1}, {17, 1}: 0.00083438ms, 0.00735685% main:@2510 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](1PJsBhDaCK:0) -> float_type, {32, 1}, {25, 1}: 0.00065576ms, 0.00578193% main:@2511 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](1QWFmJ96aB:0) -> float_type, {32, 1}, {5, 1}: 0.000641ms, 0.00565179% main:@2512 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](2CnSyUBzm0:0) -> float_type, {32, 1}, {65, 1}: 0.00063706ms, 0.00561705% main:@2513 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](2RDXZrx9GA:0) -> float_type, {32, 1}, {45, 1}: 0.00079454ms, 0.00700557% main:@2514 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](2Yy3zxNGBC:0) -> float_type, {32, 1}, {45, 1}: 0.0006156ms, 0.00542783% main:@2515 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](2lVXMZgJfO:0) -> float_type, {32, 1}, {25, 1}: 0.00062806ms, 0.0055377% main:@2516 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](2srtUfGvja:0) -> float_type, {32, 1}, {41, 1}: 0.00064728ms, 0.00570716% main:@2517 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](3VED7WSJ1b:0) -> float_type, {32, 1}, {5, 1}: 0.00080838ms, 0.0071276% main:@2518 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](gHPTxdcQWX:0) -> float_type, {32, 1}, {9, 1}: 0.00063348ms, 0.00558548% main:@2519 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](3kdi0vsSZG:0) -> float_type, {32, 1}, {25, 1}: 0.0006368ms, 0.00561476% main:@2520 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](z3lqgZM8WU:0) -> float_type, {32, 1}, {41, 1}: 0.0006644ms, 0.00585811% main:@2521 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](4M92ryD8ux:0) -> float_type, {32, 1}, {9, 1}: 0.00083288ms, 0.00734362% main:@2522 = load[offset=0,end=21376](main:@1) -> float_type, {32, 167}, {167, 1}: 0.00046704ms, 0.00411796% main:@2523 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](VUqfdN5Fzv:0) -> float_type, {32, 1}, {45, 1}: 0.00067342ms, 0.00593764% main:@2524 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](Vbv2Om3ptd:0) -> float_type, {32, 1}, {17, 1}: 0.0006557ms, 0.0057814% main:@2525 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](zX6xKOWG9S:0) -> float_type, {32, 1}, {5, 1}: 0.00070406ms, 0.0062078% main:@2526 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](1WNvHElUG7:0) -> float_type, {32, 1}, {25, 1}: 0.00074564ms, 0.00657442% main:@2527 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](IRem4MiFDP:0) -> float_type, {32, 1}, {5, 1}: 0.0007023ms, 0.00619228% main:@2528 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](bTeX85HPmJ:0) -> float_type, {32, 1}, {17, 1}: 0.0006825ms, 0.0060177% main:@2529 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](duyRoIEkH8:0) -> float_type, {32, 1}, {45, 1}: 0.00066354ms, 0.00585053% main:@2530 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](d8RoEzG52p:0) -> float_type, {32, 1}, {17, 1}: 0.00076284ms, 0.00672607% main:@2531 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](zIoqYe9cRb:0) -> float_type, {32, 1}, {45, 1}: 0.00066782ms, 0.00588827% main:@2532 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](X29iyIeu8L:0) -> float_type, {32, 1}, {45, 1}: 0.00066724ms, 0.00588315% main:@2533 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](ZRYNFfjSez:0) -> float_type, {32, 1}, {33, 1}: 0.0006808ms, 0.00600271% main:@2534 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](ZcWERoz0h6:0) -> float_type, {32, 1}, {49, 1}: 0.0007727ms, 0.00681301% main:@2535 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](nv2tzWRf1I:0) -> float_type, {32, 1}, {301, 1}: 0.0006497ms, 0.0057285% main:@2536 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](TxEdCcekGS:0) -> float_type, {32, 1}, {49, 1}: 0.00069458ms, 0.00612421% main:@2537 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](yVxKAL4U5G:0) -> float_type, {32, 1}, {9, 1}: 0.00069096ms, 0.00609229% main:@2538 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](aqMQBgb3hf:0) -> float_type, {32, 1}, {17, 1}: 0.00075624ms, 0.00666788% main:@2539 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](ahAlEtBvzo:0) -> float_type, {32, 1}, {41, 1}: 0.00067544ms, 0.00595545% main:@2540 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](bJsIAzhYc3:0) -> float_type, {32, 1}, {37, 1}: 0.00067554ms, 0.00595633% main:@2541 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](cN2JSTnFOu:0) -> float_type, {32, 1}, {9, 1}: 0.00066842ms, 0.00589356% main:@2542 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](vAbKepTjQg:0) -> float_type, {32, 1}, {5, 1}: 0.00077518ms, 0.00683487% main:@2543 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](fZhaL2tCm6:0) -> float_type, {32, 1}, {17, 1}: 0.00065964ms, 0.00581614% main:@2544 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](o74vNAnXxy:0) -> float_type, {32, 1}, {17, 1}: 0.00070052ms, 0.00617659% ZyQ5FrpI6t:0 = @param:ZyQ5FrpI6t:0 -> float_type, {32, 1}, {1, 1}: 0.00050752ms, 0.00447488% mxUaAp0CV8:0 = @param:mxUaAp0CV8:0 -> float_type, {32, 1}, {1, 1}: 0.00061886ms, 0.00545658% vF46yONhri:0 = @param:vF46yONhri:0 -> float_type, {32, 1}, {1, 1}: 0.00059086ms, 0.0052097% URVDgNGoIx:0 = @param:URVDgNGoIx:0 -> float_type, {32, 1}, {1, 1}: 0.0004971ms, 0.004383% RPGvVrZMak:0 = @param:RPGvVrZMak:0 -> float_type, {32, 1}, {1, 1}: 0.00048862ms, 0.00430823% main:@2550 = gpu::code_object[code_object=25808,symbol_name=concat_kernel,global=32,local=256,](main:@2426,main:@2535,main:@2520,main:@2416,main:@2514,main:@2415,main:@2536,main:@2461,main:@2499,main:@2512,main:@2482,main:@2388,main:@2480,main:@2384,iE352qtAOx:0,main:@2521,RPGvVrZMak:0,URVDgNGoIx:0,vF46yONhri:0,main:@2436,main:@2438,bJxY1IhSVn:0,main:@2541,XPpbueLMOZ:0,lqOmNvz5fj:0,main:@2418,main:@2473,main:@2538,main:@2527,NG35MnhTIF:0,main:@2454,vtIF8HqRG6:0,7XiFOysrpI:0,WqeN5KGIv9:0,mxUaAp0CV8:0,ZyQ5FrpI6t:0,main:@2497,main:@2530,main:@2543,main:@2524,main:@2393,main:@2475,main:@2485,main:@2399,main:@2439,9NcxXsmyqf:0,main:@2398,main:@2458,main:@2479,main:@2435,qxzgXpiLdl:0,StHFqj5edV:0,lhbKMrO1my:0,xGCyW9icVP:0,iUEBSr6gOG:0,ebEFYOy65u:0,main:@2392,main:@2401,main:@2460,main:@2430,main:@2478,main:@2483,main:@2396,main:@2531,main:@2433,main:@2534,main:@2437,main:@2456,main:@2539,main:@2434,main:@2431,main:@2540,main:@2428,main:@2411,main:@2526,main:@2429,main:@2472,main:@2441,main:@2495,main:@2544,main:@2394,main:@2502,main:@2395,main:@2515,main:@2519,main:@2469,main:@2532,main:@2510,main:@2420,main:@2474,main:@2492,main:@2503,main:@2494,main:@2496,main:@2501,main:@2397,main:@2476,main:@2403,main:@2422,main:@2425,main:@2491,main:@2537,main:@2505,FTIMrcfako:0,main:@2424,main:@2481,7tFB8mLWUr:0,main:@2466,main:@2427,main:@2504,c2MuYjWE3S:0,main:@2405,main:@2391,main:@2523,main:@2404,main:@2513,main:@2516,main:@2414,main:@2529,main:@2525,main:@2440,main:@2511,main:@2471,main:@2470,main:@2457,main:@2390,main:@2406,main:@2400,main:@2498,main:@2419,main:@2517,main:@2383,main:@2421,main:@2412,main:@2542,main:@2455,main:@2407,main:@2467,main:@2486,main:@2386,main:@2528,main:@2385,main:@2465,main:@2417,main:@2459,main:@2410,main:@2389,main:@2493,main:@2477,main:@2509,main:@2402,main:@2413,main:@2484,main:@2533,main:@2432,main:@2409,main:@2468,main:@2518,main:@2500,main:@2423,main:@2382,main:@2463,main:@2490,main:@2408,main:@2464,main:@2462,main:@2387,main:@2522) -> float_type, {32, 167}, {167, 1}: 0.133455ms, 1.17669% main:@2551 = load[offset=21376,end=42752](main:@1) -> float_type, {32, 167}, {167, 1}: 0.00069496ms, 0.00612756% main:@2552 = gpu::code_object[code_object=13520,symbol_name=rsqrt_add_kernel,global=2672,local=256,](main:@2381,main:@2550,main:@2551) -> float_type, {32, 167}, {167, 1}: 0.0135341ms, 0.119332% main:@2553 = hip::hip_copy_literal[id=main:@literal:45] -> float_type, {167, 256}, {256, 1}: 0.00078032ms, 0.00688019% main:@2554 = load[offset=62336,end=95104](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00045652ms, 0.0040252% main:@2555 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@2552,main:@2553,main:@2554) -> float_type, {32, 256}, {256, 1}: 0.024311ms, 0.214353% main:@2556 = hip::hip_copy_literal[id=main:@literal:22] -> float_type, {32, 256}, {256, 1}: 0.00083282ms, 0.00734309% main:@2557 = load[offset=29568,end=62336](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00046948ms, 0.00413947% main:@2558 = gpu::code_object[code_object=13552,symbol_name=add_leaky_relu_tanh_pow_kernel,global=4096,local=256,](main:@2555,main:@2556,main:@2377,main:@2557) -> float_type, {32, 256}, {256, 1}: 0.0178923ms, 0.157759% main:@2559 = hip::hip_copy_literal[id=main:@literal:76] -> float_type, {256, 64}, {64, 1}: 0.00077852ms, 0.00686432% main:@2560 = load[offset=21376,end=29568](main:@1) -> float_type, {32, 64}, {64, 1}: 0.00047934ms, 0.00422641% main:@2561 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@2558,main:@2559,main:@2560) -> float_type, {32, 64}, {64, 1}: 0.0254114ms, 0.224056% main:@2562 = hip::hip_copy_literal[id=main:@literal:111] -> float_type, {53, 53}, {53, 1}: 0.0008517ms, 0.00750956% main:@2563 = slice[axes={1},starts={53},ends={106}](main:@2373) -> float_type, {32, 53}, {212, 1}: 0.00063232ms, 0.00557526% main:@2564 = load[offset=43136,end=49920](main:@1) -> float_type, {32, 53}, {53, 1}: 0.00047156ms, 0.00415781% main:@2565 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@2563,main:@2562,main:@2564) -> float_type, {32, 53}, {53, 1}: 0.0209588ms, 0.184796% main:@2566 = hip::hip_copy_literal[id=main:@literal:82] -> float_type, {1, 53}, {53, 1}: 0.0007494ms, 0.00660757% main:@2567 = slice[axes={1},starts={159},ends={212}](main:@2373) -> float_type, {32, 53}, {212, 1}: 0.00058262ms, 0.00513704% main:@2568 = load[offset=29568,end=36352](main:@1) -> float_type, {32, 53}, {53, 1}: 0.00047574ms, 0.00419467% main:@2569 = gpu::code_object[code_object=13536,symbol_name=div_pow_log_add_kernel,global=1696,local=256,](main:@2567,main:@2568) -> float_type, {32, 53}, {53, 1}: 0.0155859ms, 0.137423% main:@2570 = multibroadcast[out_lens={32, 53}](main:@2566) -> float_type, {32, 53}, {0, 1}: 0.0007922ms, 0.00698494% main:@2571 = load[offset=36352,end=43136](main:@1) -> float_type, {32, 53}, {53, 1}: 0.00047524ms, 0.00419026% main:@2572 = gpu::code_object[code_object=13552,symbol_name=pow_sub_pow_pow_log_add_kernel,global=1696,local=256,](main:@2565,main:@2570,main:@2563,main:@2571) -> float_type, {32, 53}, {53, 1}: 0.0180381ms, 0.159044% main:@2573 = load[offset=43136,end=49920](main:@1) -> float_type, {32, 53}, {53, 1}: 0.00062194ms, 0.00548373% main:@2574 = slice[axes={1},starts={106},ends={159}](main:@2373) -> float_type, {32, 53}, {212, 1}: 0.00056808ms, 0.00500884% main:@2575 = slice[axes={1},starts={0},ends={53}](main:@2373) -> float_type, {32, 53}, {212, 1}: 0.00055852ms, 0.00492455% main:@2576 = gpu::code_object[code_object=13536,symbol_name=add_add_neg_div_kernel,global=1696,local=256,](main:@2574,main:@2575,main:@2573) -> float_type, {32, 53}, {53, 1}: 0.0142028ms, 0.125228% main:@2577 = hip::hip_copy_literal[id=main:@literal:83] -> float_type, {374, 212}, {212, 1}: 0.0006943ms, 0.00612174% main:@2578 = load[offset=77184,end=77312](main:@1) -> float_type, {32, 1}, {1, 1}: 0.00045956ms, 0.00405201% main:@2579 = gpu::code_object[code_object=13544,symbol_name=reduce_mean_add_neg_kernel,global=2048,local=64,](main:@2569,main:@2578) -> float_type, {32, 1}, {1, 1}: 0.0142553ms, 0.125691% main:@2580 = load[offset=49920,end=77056](main:@1) -> float_type, {32, 212}, {212, 1}: 0.00060332ms, 0.00531956% main:@2581 = gpu::code_object[code_object=13528,symbol_name=leaky_relu_kernel,global=3392,local=256,](main:@2373,main:@2580) -> float_type, {32, 212}, {212, 1}: 0.0132868ms, 0.117152% main:@2582 = load[offset=77568,end=77696](main:@1) -> float_type, {32, 1}, {1, 1}: 0.00059686ms, 0.0052626% main:@2583 = gpu::code_object[code_object=13544,symbol_name=reduce_prod_add_log_kernel,global=2048,local=64,](main:@2576,main:@2582) -> float_type, {32, 1}, {1, 1}: 0.0139922ms, 0.123372% main:@2584 = load[offset=77056,end=77184](main:@1) -> float_type, {32, 1}, {1, 1}: 0.00059104ms, 0.00521128% main:@2585 = gpu::code_object[code_object=13552,symbol_name=reduce_prod_pow_rsqrt_div_kernel,global=2048,local=64,](main:@2572,main:@2584) -> float_type, {32, 1}, {1, 1}: 0.014818ms, 0.130653% main:@2586 = load[offset=77312,end=77568](main:@1) -> float_type, {32, 2}, {2, 1}: 0.00060118ms, 0.00530069% main:@2587 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=32,local=256,](main:@2583,main:@2579,main:@2586) -> float_type, {32, 2}, {2, 1}: 0.0133462ms, 0.117676% main:@2588 = slice[axes={1},starts={4070},ends={4444}](main:@2244) -> float_type, {32, 374}, {4508, 1}: 0.00071462ms, 0.00630091% main:@2589 = slice[axes={1},starts={106},ends={159}](main:@2581) -> float_type, {32, 53}, {212, 1}: 0.00054584ms, 0.00481275% main:@2590 = slice[axes={1},starts={159},ends={212}](main:@2581) -> float_type, {32, 53}, {212, 1}: 0.00067526ms, 0.00595386% main:@2591 = slice[axes={1},starts={53},ends={106}](main:@2581) -> float_type, {32, 53}, {212, 1}: 0.00055842ms, 0.00492367% main:@2592 = slice[axes={1},starts={0},ends={53}](main:@2581) -> float_type, {32, 53}, {212, 1}: 0.00054448ms, 0.00480076% main:@2593 = load[offset=691712,end=739584](main:@1) -> float_type, {32, 374}, {374, 1}: 0.0004533ms, 0.00399681% main:@2594 = gpu::code_object[code_object=21816,symbol_name=noop_noop_noop_noop_noop_noop_noop_sub_noop_concat_neg_pow_kernel,global=1329,local=256,](main:@2589,main:@2592,main:@2590,main:@2591,main:@2576,main:@2569,main:@2572,main:@2278,main:@2587,main:@2585,main:@2588,main:@2593) -> float_type, {32, 374}, {374, 1}: 0.037038ms, 0.32657% main:@2595 = load[offset=29568,end=56704](main:@1) -> float_type, {32, 212}, {212, 1}: 0.00062244ms, 0.00548814% main:@2596 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@2594,main:@2577,main:@2595) -> float_type, {32, 212}, {212, 1}: 0.027943ms, 0.246377% main:@2597 = hip::hip_copy_literal[id=main:@literal:16] -> float_type, {32, 64}, {64, 1}: 0.00084084ms, 0.00741381% main:@2598 = load[offset=56704,end=56832](main:@1) -> float_type, {32, 1}, {1, 1}: 0.00046788ms, 0.00412537% main:@2599 = gpu::code_object[code_object=13528,symbol_name=reduce_prod_kernel,global=6144,local=192,](main:@2550,main:@2598) -> float_type, {32, 1}, {1, 1}: 0.014474ms, 0.12762% main:@2600 = hip::hip_copy_literal[id=main:@literal:33] -> float_type, {277, 160}, {160, 1}: 0.00078072ms, 0.00688372% main:@2601 = hip::hip_copy_literal[id=main:@literal:32] -> float_type, {32, 212}, {212, 1}: 0.00064316ms, 0.00567083% main:@2602 = slice[axes={1},starts={4444},ends={4508}](main:@2244) -> float_type, {32, 64}, {4508, 1}: 0.00071216ms, 0.00627922% main:@2603 = load[offset=56832,end=92288](main:@1) -> float_type, {32, 277}, {277, 1}: 0.0005654ms, 0.00498521% main:@2604 = gpu::code_object[code_object=13600,symbol_name=rsqrt_add_sub_mul_add_leaky_relu_noop_concat_noop_kernel,global=2954,local=256,](main:@2602,main:@2561,main:@2597,main:@2596,main:@2601,main:@2599,main:@2603) -> float_type, {32, 277}, {277, 1}: 0.0180774ms, 0.159391% main:@2605 = load[offset=20480,end=40960](main:@1) -> float_type, {32, 160}, {160, 1}: 0.00063238ms, 0.00557579% main:@2606 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@2604,main:@2600,main:@2605) -> float_type, {32, 160}, {160, 1}: 0.0255358ms, 0.225153% main:@2607 = hip::hip_copy_literal[id=main:@literal:34] -> float_type, {160}, {1}: 0.00073432ms, 0.00647461% main:@2608 = multibroadcast[out_lens={32, 160}](main:@2607) -> float_type, {32, 160}, {0, 1}: 0.00061068ms, 0.00538445% main:@2609 = load[offset=0,end=20480](main:@1) -> float_type, {32, 160}, {160, 1}: 0.0004863ms, 0.00428778% main:@2610 = gpu::code_object[code_object=13536,symbol_name=add_leaky_relu_kernel,global=2560,local=256,](main:@2606,main:@2608,main:@2609) -> float_type, {32, 160}, {160, 1}: 0.0136396ms, 0.120262% main:@2611 = hip::hip_copy_literal[id=main:@literal:36] -> float_type, {64, 329}, {329, 1}: 0.00073738ms, 0.00650159% main:@2612 = load[offset=20608,end=62720](main:@1) -> float_type, {32, 329}, {329, 1}: 0.00046144ms, 0.00406858% main:@2613 = slice[axes={1},starts={80},ends={144}](main:@2610) -> float_type, {32, 64}, {160, 1}: 0.00057182ms, 0.00504182% main:@2614 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@2613,main:@2611,main:@2612) -> float_type, {32, 329}, {329, 1}: 0.0199416ms, 0.175828% main:@2615 = hip::hip_copy_literal[id=main:@literal:88] -> float_type, {1, 329}, {329, 1}: 0.00079592ms, 0.00701774% main:@2616 = hip::hip_copy_literal[id=main:@literal:65] -> float_type, {16, 1}, {1, 1}: 0.00058114ms, 0.005124% main:@2617 = slice[axes={1},starts={144},ends={160}](main:@2610) -> float_type, {32, 16}, {160, 1}: 0.0005543ms, 0.00488734% main:@2618 = load[offset=20480,end=20608](main:@1) -> float_type, {32, 1}, {1, 1}: 0.0004633ms, 0.00408498% main:@2619 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@2617,main:@2616,main:@2618) -> float_type, {32, 1}, {1, 1}: 0.0216139ms, 0.190573% main:@2620 = load[offset=62720,end=104832](main:@1) -> float_type, {32, 329}, {329, 1}: 0.00062526ms, 0.00551301% main:@2621 = gpu::code_object[code_object=13520,symbol_name=softmax_kernel,global=8192,local=256,](main:@2614,main:@2620) -> float_type, {32, 329}, {329, 1}: 0.0154518ms, 0.136241% main:@2622 = multibroadcast[out_lens={32, 329}](main:@2615) -> float_type, {32, 329}, {0, 1}: 0.0007282ms, 0.00642064% main:@2623 = load[offset=20608,end=20736](main:@1) -> float_type, {32, 1}, {1, 1}: 0.00048548ms, 0.00428055% main:@2624 = gpu::code_object[code_object=13536,symbol_name=add_reduce_prod_kernel,global=8192,local=256,](main:@2622,main:@2621,main:@2623) -> float_type, {32, 1}, {1, 1}: 0.0150797ms, 0.13296% main:#output_1 = @param:main:#output_1 -> float_type, {32}, {1}: 0.00090628ms, 0.0079908% main:@2626 = squeeze[axes={1}](main:@2624) -> float_type, {32}, {1}: 0.00058308ms, 0.0051411% main:@2627 = gpu::code_object[code_object=13512,symbol_name=mul_kernel,global=16,local=256,](main:@2626,main:#output_1) -> float_type, {32}, {1}: 0.0132478ms, 0.116808% main:@2628 = hip::hip_copy_literal[id=main:@literal:110] -> float_type, {64, 500}, {500, 1}: 0.0007592ms, 0.00669398% main:@2629 = slice[axes={1},starts={16},ends={80}](main:@2610) -> float_type, {32, 64}, {160, 1}: 0.00059012ms, 0.00520317% main:@2630 = load[offset=20992,end=84992](main:@1) -> float_type, {32, 500}, {500, 1}: 0.00046662ms, 0.00411426% main:@2631 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@2629,main:@2628,main:@2630) -> float_type, {32, 500}, {500, 1}: 0.0198147ms, 0.174709% main:@2632 = hip::hip_copy_literal[id=main:@literal:147] -> float_type, {16, 3}, {3, 1}: 0.00070636ms, 0.00622808% main:@2633 = slice[axes={1},starts={0},ends={16}](main:@2610) -> float_type, {32, 16}, {160, 1}: 0.00061468ms, 0.00541972% main:@2634 = load[offset=20608,end=20992](main:@1) -> float_type, {32, 3}, {3, 1}: 0.0004735ms, 0.00417492% main:@2635 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@2633,main:@2632,main:@2634) -> float_type, {32, 3}, {3, 1}: 0.0198387ms, 0.174921% main:@2636 = hip::hip_copy_literal[id=main:@literal:72] -> float_type, {1, 500}, {500, 1}: 0.00082176ms, 0.00724558% main:@2637 = squeeze[axes={1}](main:@2619) -> float_type, {32}, {1}: 0.0005688ms, 0.00501519% main:#output_0 = @param:main:#output_0 -> float_type, {32}, {1}: 0.0006173ms, 0.00544282% 86KRkhYbW4:0 = @param:86KRkhYbW4:0 -> float_type, {32}, {1}: 0.00052114ms, 0.00459497% main:@2640 = gpu::code_object[code_object=13536,symbol_name=sigmoid_mul_tanh_kernel,global=16,local=256,](86KRkhYbW4:0,main:@2637,main:#output_0) -> float_type, {32}, {1}: 0.0148143ms, 0.13062% main:#output_2 = @param:main:#output_2 -> float_type, {32}, {1}: 0.00075444ms, 0.00665201% main:@2642 = gpu::code_object[code_object=13512,symbol_name=add_kernel,global=16,local=256,](main:@2627,main:#output_2) -> float_type, {32}, {1}: 0.01335ms, 0.117709% main:#output_3 = @param:main:#output_3 -> float_type, {32}, {1}: 0.00072634ms, 0.00640424% main:@2644 = gpu::code_object[code_object=13520,symbol_name=mul_neg_kernel,global=16,local=256,](main:@2627,main:#output_3) -> float_type, {32}, {1}: 0.0131717ms, 0.116136% main:@2645 = load[offset=84992,end=148992](main:@1) -> float_type, {32, 500}, {500, 1}: 0.00063068ms, 0.0055608% main:@2646 = gpu::code_object[code_object=13520,symbol_name=softmax_kernel,global=8192,local=256,](main:@2631,main:@2645) -> float_type, {32, 500}, {500, 1}: 0.0151268ms, 0.133375% main:@2647 = multibroadcast[out_lens={32, 500}](main:@2636) -> float_type, {32, 500}, {0, 1}: 0.00071172ms, 0.00627534% main:@2648 = load[offset=0,end=128](main:@1) -> float_type, {32, 1}, {1, 1}: 0.00047202ms, 0.00416187% main:@2649 = gpu::code_object[code_object=13536,symbol_name=add_reduce_mean_kernel,global=8192,local=256,](main:@2647,main:@2646,main:@2648) -> float_type, {32, 1}, {1, 1}: 0.0147762ms, 0.130284% main:#output_4 = @param:main:#output_4 -> float_type, {32}, {1}: 0.0007297ms, 0.00643387% main:@2651 = squeeze[axes={1}](main:@2649) -> float_type, {32}, {1}: 0.00050854ms, 0.00448387% main:@2652 = gpu::code_object[code_object=13512,symbol_name=mul_kernel,global=16,local=256,](main:@2651,main:#output_4) -> float_type, {32}, {1}: 0.013005ms, 0.114667% main:@2653 = load[offset=0,end=384](main:@1) -> float_type, {32, 3}, {3, 1}: 0.00062102ms, 0.00547562% main:@2654 = gpu::code_object[code_object=14512,symbol_name=softmax_kernel,global=2048,local=64,](main:@2635,main:@2653) -> float_type, {32, 3}, {3, 1}: 0.0144228ms, 0.127168% main:@2655 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](main:@2654) -> float_type, {32, 1}, {3, 1}: 0.00079338ms, 0.00699535% main:#output_5 = @param:main:#output_5 -> float_type, {32, 1}, {1, 1}: 0.00065666ms, 0.00578987% main:@2657 = gpu::code_object[code_object=13528,symbol_name=contiguous_kernel,global=32,local=256,](main:@2655,main:#output_5) -> float_type, {32, 1}, {1, 1}: 0.013151ms, 0.115954% main:@2658 = squeeze[axes={1}](main:@2657) -> float_type, {32}, {1}: 0.00087184ms, 0.00768714% main:@2659 = hip::sync_stream(main:@2640,main:@2627,main:@2642,main:@2644,main:@2652,main:@2658) -> float_type, {32}, {1}: 0.00117186ms, 0.0103325% main:@2660 = @return(main:@2659,main:@2627,main:@2642,main:@2644,main:@2652,main:@2658) Summary: gpu::gemm: 4.7761ms / 187 = 0.0255406ms, 42.1115% gpu::code_object::concat_kernel: 1.84751ms / 84 = 0.0219942ms, 16.2898% slice: 0.604983ms / 859 = 0.000704287ms, 5.33422% gpu::code_object::relu_kernel: 0.258735ms / 18 = 0.0143742ms, 2.28131% load: 0.255684ms / 468 = 0.000546334ms, 2.2544% gpu::code_object::pow_div_add_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_concat_log_add_kernel: 0.230915ms / 1 = 0.230915ms, 2.03602% gpu::code_object::softmax_kernel: 0.21932ms / 14 = 0.0156657ms, 1.93378% @param: 0.209775ms / 383 = 0.000547716ms, 1.84962% gpu::code_object::convert_kernel: 0.200377ms / 15 = 0.0133584ms, 1.76675% gpu::code_object::contiguous_kernel: 0.135934ms / 10 = 0.0135934ms, 1.19855% hip::hip_copy_literal: 0.126417ms / 165 = 0.000766162ms, 1.11463% gpu::code_object::add_kernel: 0.108618ms / 7 = 0.0155168ms, 0.9577% unsqueeze: 0.0887336ms / 145 = 0.000611956ms, 0.782377% gpu::code_object::neg_pow_kernel: 0.0816322ms / 5 = 0.0163264ms, 0.719763% gpu::code_object::log_add_kernel: 0.0797937ms / 6 = 0.0132989ms, 0.703552% gpu::code_object::neg_div_kernel: 0.0792737ms / 6 = 0.0132123ms, 0.698968% gpu::code_object::pow_kernel: 0.0669181ms / 4 = 0.0167295ms, 0.590027% gpu::code_object::noop_rsqrt_add_concat_noop_kernel: 0.066082ms / 4 = 0.0165205ms, 0.582654% gpu::code_object::div_kernel: 0.054007ms / 3 = 0.0180023ms, 0.476188% gpu::code_object::rsqrt_add_kernel: 0.0530193ms / 4 = 0.0132548ms, 0.467479% gpu::code_object::sub_kernel: 0.0505533ms / 3 = 0.0168511ms, 0.445736% gpu::code_object::rsqrt_pow_kernel: 0.0488774ms / 3 = 0.0162925ms, 0.430959% reshape_lazy: 0.0470188ms / 81 = 0.000580479ms, 0.414571% gpu::code_object::reduce_prod_kernel: 0.0436792ms / 3 = 0.0145597ms, 0.385125% gpu::code_object::tanh_sub_kernel: 0.0410102ms / 3 = 0.0136701ms, 0.361593% gpu::code_object::convert_add_kernel: 0.0398892ms / 3 = 0.0132964ms, 0.351709% gpu::code_object::neg_add_kernel: 0.0395731ms / 3 = 0.013191ms, 0.348921% gpu::code_object::noop_noop_noop_noop_noop_noop_noop_sub_noop_concat_neg_pow_kernel: 0.037038ms / 1 = 0.037038ms, 0.32657% gpu::code_object::pow_pow_kernel: 0.0358886ms / 2 = 0.0179443ms, 0.316435% gpu::code_object::leaky_relu_noop_noop_noop_noop_noop_noop_concat_rsqrt_pow_kernel: 0.0357872ms / 1 = 0.0357872ms, 0.315541% gpu::code_object::leaky_relu_noop_noop_noop_noop_add_concat_log_pow_kernel: 0.0332408ms / 1 = 0.0332408ms, 0.293089% gpu::code_object::log_pow_kernel: 0.0330992ms / 2 = 0.0165496ms, 0.29184% gpu::code_object::noop_log_add_concat_noop_kernel: 0.0329563ms / 2 = 0.0164781ms, 0.29058% gpu::code_object::leaky_relu_noop_noop_noop_noop_noop_noop_concat_log_add_pow_kernel: 0.0320755ms / 1 = 0.0320755ms, 0.282815% gpu::code_object::noop_rsqrt_div_concat_noop_kernel: 0.0319444ms / 2 = 0.0159722ms, 0.281659% gpu::code_object::noop_tanh_div_concat_noop_kernel: 0.0308563ms / 2 = 0.0154281ms, 0.272064% gpu::code_object::add_leaky_relu_kernel: 0.0292154ms / 2 = 0.0146077ms, 0.257596% gpu::code_object::add_add_neg_add_kernel: 0.0284014ms / 2 = 0.0142007ms, 0.25042% gpu::code_object::tanh_div_kernel: 0.0281397ms / 2 = 0.0140699ms, 0.248112% gpu::code_object::tanh_add_kernel: 0.0275807ms / 2 = 0.0137904ms, 0.243183% gpu::code_object::convert_mul_add_kernel: 0.0264776ms / 2 = 0.0132388ms, 0.233457% gpu::code_object::mul_kernel: 0.0262529ms / 2 = 0.0131264ms, 0.231475% gpu::code_object::rsqrt_sub_kernel: 0.0259761ms / 2 = 0.0129881ms, 0.229035% gpu::code_object::pow_reduce_sum_div_pow_add_kernel: 0.0245347ms / 1 = 0.0245347ms, 0.216326% multibroadcast: 0.0234863ms / 35 = 0.000671038ms, 0.207082% gpu::code_object::mul_relu_reduce_prod_sub_mul_reduce_sum_add_add_add_add_pow_kernel: 0.0220324ms / 1 = 0.0220324ms, 0.194262% gpu::code_object::leaky_relu_noop_noop_noop_add_noop_concat_log_div_kernel: 0.0220279ms / 1 = 0.0220279ms, 0.194224% gpu::code_object::pow_sub_div_pow_log_div_kernel: 0.0200982ms / 1 = 0.0200982ms, 0.177209% gpu::code_object::noop_tanh_pow_concat_noop_kernel: 0.0184563ms / 1 = 0.0184563ms, 0.162732% gpu::code_object::rsqrt_add_sub_mul_add_leaky_relu_noop_concat_noop_kernel: 0.0180774ms / 1 = 0.0180774ms, 0.159391% gpu::code_object::pow_sub_pow_pow_log_add_kernel: 0.0180381ms / 1 = 0.0180381ms, 0.159044% gpu::code_object::add_leaky_relu_tanh_pow_kernel: 0.0178923ms / 1 = 0.0178923ms, 0.157759% gpu::code_object::pow_pow_sigmoid_add_kernel: 0.0178653ms / 1 = 0.0178653ms, 0.157521% gpu::code_object::add_pow_log_pow_kernel: 0.017394ms / 1 = 0.017394ms, 0.153365% gpu::code_object::sub_sub_pow_log_pow_kernel: 0.0172712ms / 1 = 0.0172712ms, 0.152282% gpu::code_object::noop_log_pow_concat_noop_kernel: 0.0171939ms / 1 = 0.0171939ms, 0.151601% squeeze: 0.0171533ms / 25 = 0.000686134ms, 0.151243% gpu::code_object::pow_mul_add_neg_sub_kernel: 0.0170592ms / 1 = 0.0170592ms, 0.150413% gpu::code_object::pow_mul_rsqrt_sub_kernel: 0.0168807ms / 1 = 0.0168807ms, 0.148839% gpu::code_object::noop_neg_div_concat_noop_kernel: 0.0165859ms / 1 = 0.0165859ms, 0.14624% gpu::code_object::noop_log_sub_concat_noop_kernel: 0.0164404ms / 1 = 0.0164404ms, 0.144958% gpu::code_object::add_pow_kernel: 0.0164328ms / 1 = 0.0164328ms, 0.14489% gpu::code_object::neg_rsqrt_tanh_log_concat_add_kernel: 0.0158978ms / 1 = 0.0158978ms, 0.140174% gpu::code_object::pow_div_kernel: 0.0157823ms / 1 = 0.0157823ms, 0.139155% gpu::code_object::div_pow_log_add_kernel: 0.0155859ms / 1 = 0.0155859ms, 0.137423% gpu::code_object::pow_mul_kernel: 0.0154184ms / 1 = 0.0154184ms, 0.135946% gpu::code_object::pow_sub_kernel: 0.0153168ms / 1 = 0.0153168ms, 0.13505% gpu::code_object::noop_rsqrt_sub_concat_noop_kernel: 0.0152904ms / 1 = 0.0152904ms, 0.134818% gpu::code_object::reduce_mean_add_neg_pow_kernel: 0.0152754ms / 1 = 0.0152754ms, 0.134686% gpu::code_object::reduce_mean_add_rsqrt_add_kernel: 0.01523ms / 1 = 0.01523ms, 0.134285% gpu::code_object::add_add_rsqrt_pow_kernel: 0.0151904ms / 1 = 0.0151904ms, 0.133936% gpu::code_object::div_mul_neg_pow_kernel: 0.0151397ms / 1 = 0.0151397ms, 0.133489% gpu::code_object::reduce_mean_pow_log_pow_kernel: 0.0150965ms / 1 = 0.0150965ms, 0.133108% gpu::code_object::add_reduce_prod_kernel: 0.0150797ms / 1 = 0.0150797ms, 0.13296% gpu::code_object::reduce_prod_pow_neg_sub_kernel: 0.0150184ms / 1 = 0.0150184ms, 0.132419% gpu::code_object::reduce_prod_pow_rsqrt_div_kernel: 0.014818ms / 1 = 0.014818ms, 0.130653% gpu::code_object::sigmoid_mul_tanh_kernel: 0.0148143ms / 1 = 0.0148143ms, 0.13062% gpu::code_object::reduce_prod_add_rsqrt_sub_kernel: 0.0147993ms / 1 = 0.0147993ms, 0.130488% gpu::code_object::add_reduce_mean_kernel: 0.0147762ms / 1 = 0.0147762ms, 0.130284% gpu::code_object::reduce_mean_kernel: 0.0147549ms / 1 = 0.0147549ms, 0.130096% gpu::code_object::sub_sub_mul_rsqrt_sub_kernel: 0.0145575ms / 1 = 0.0145575ms, 0.128356% gpu::code_object::noop_add_concat_div_kernel: 0.0145417ms / 1 = 0.0145417ms, 0.128216% gpu::code_object::reduce_sum_kernel: 0.0145063ms / 1 = 0.0145063ms, 0.127904% gpu::code_object::reduce_mean_add_sigmoid_sub_kernel: 0.0144587ms / 1 = 0.0144587ms, 0.127484% gpu::code_object::reduce_mean_add_sigmoid_kernel: 0.0144008ms / 1 = 0.0144008ms, 0.126974% gpu::code_object::reduce_prod_add_log_div_kernel: 0.0143905ms / 1 = 0.0143905ms, 0.126883% gpu::code_object::add_add_rsqrt_add_kernel: 0.0143294ms / 1 = 0.0143294ms, 0.126344% gpu::code_object::reduce_mean_mul_neg_kernel: 0.0142881ms / 1 = 0.0142881ms, 0.12598% gpu::code_object::reduce_prod_mul_rsqrt_add_kernel: 0.0142605ms / 1 = 0.0142605ms, 0.125737% gpu::code_object::reduce_mean_add_neg_kernel: 0.0142553ms / 1 = 0.0142553ms, 0.125691% gpu::code_object::reduce_prod_mul_log_kernel: 0.0142364ms / 1 = 0.0142364ms, 0.125525% gpu::code_object::reduce_mean_mul_rsqrt_kernel: 0.0142049ms / 1 = 0.0142049ms, 0.125246% gpu::code_object::add_add_neg_div_kernel: 0.0142028ms / 1 = 0.0142028ms, 0.125228% gpu::code_object::reduce_prod_add_log_kernel: 0.0139922ms / 1 = 0.0139922ms, 0.123372% gpu::code_object::convert_div_mul_kernel: 0.0135083ms / 1 = 0.0135083ms, 0.119105% gpu::code_object::log_div_kernel: 0.0134851ms / 1 = 0.0134851ms, 0.1189% gpu::code_object::convert_add_add_sub_kernel: 0.0134834ms / 1 = 0.0134834ms, 0.118885% transpose: 0.0134084ms / 22 = 0.000609472ms, 0.118224% gpu::code_object::log_sub_kernel: 0.0133835ms / 1 = 0.0133835ms, 0.118005% gpu::code_object::convert_mul_kernel: 0.0133155ms / 1 = 0.0133155ms, 0.117405% gpu::code_object::leaky_relu_kernel: 0.0132868ms / 1 = 0.0132868ms, 0.117152% gpu::code_object::add_mul_kernel: 0.0132674ms / 1 = 0.0132674ms, 0.11698% gpu::code_object::mul_neg_kernel: 0.0131717ms / 1 = 0.0131717ms, 0.116136% gpu::code_object::neg_sub_kernel: 0.0131596ms / 1 = 0.0131596ms, 0.11603% gpu::code_object::add_sub_kernel: 0.0131557ms / 1 = 0.0131557ms, 0.115996% gpu::code_object::convert_pow_add_kernel: 0.0128615ms / 1 = 0.0128615ms, 0.113402% check_context::migraphx::version_1::gpu::context: 0.00160664ms / 1 = 0.00160664ms, 0.014166% hip::sync_stream: 0.00117186ms / 1 = 0.00117186ms, 0.0103325% hip::hip_allocate_memory: 0.00108476ms / 1 = 0.00108476ms, 0.00956449% Batch size: 32 Rate: 6606.78/sec Total time: 4.84351ms Total instructions time: 11.3415ms Overhead time: 0.817697ms, -6.49803ms Overhead: 17%, -134%