Compiling ... Reading: /home/workspace/ByteMLPerf/bytedance/models/model_1/mxr1//model-static-batch-size-64.mxr Allocating params ... Running performance report ... main:@0 = check_context::migraphx::version_1::gpu::context -> float_type, {}, {}: 0.00151968ms, 0.0130256% main:@1 = hip::hip_allocate_memory[shape=float_type, {3926656}, {1},id=main:scratch] -> float_type, {3926656}, {1}: 0.0010882ms, 0.0093273% lyiDxfzZg3:0 = @param:lyiDxfzZg3:0 -> float_type, {64, 701}, {701, 1}: 0.0007674ms, 0.00657762% main:@3 = slice[axes={0, 1},starts={0, 541},ends={64, 573}](lyiDxfzZg3:0) -> float_type, {64, 32}, {701, 1}: 0.00130888ms, 0.0112188% uFSOc5Chp2:0 = @param:uFSOc5Chp2:0 -> float_type, {64, 481}, {481, 1}: 0.00050172ms, 0.0043004% main:@5 = slice[axes={0, 1},starts={0, 337},ends={64, 353}](uFSOc5Chp2:0) -> float_type, {64, 16}, {481, 1}: 0.0008108ms, 0.00694962% nv2tzWRf1I:0 = @param:nv2tzWRf1I:0 -> float_type, {64, 301}, {301, 1}: 0.00050664ms, 0.00434257% main:@7 = slice[axes={0, 1},starts={0, 285},ends={64, 301}](nv2tzWRf1I:0) -> float_type, {64, 16}, {301, 1}: 0.0007287ms, 0.00624591% n4iptOVFwE:0 = @param:n4iptOVFwE:0 -> float_type, {64, 113}, {113, 1}: 0.00045402ms, 0.00389154% main:@9 = slice[axes={0, 1},starts={0, 105},ends={64, 113}](n4iptOVFwE:0) -> float_type, {64, 8}, {113, 1}: 0.00071422ms, 0.0061218% main:@10 = load[offset=799744,end=818176](main:@1) -> float_type, {64, 72}, {72, 1}: 0.0006005ms, 0.00514707% main:@11 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=1152,local=256,](main:@3,main:@7,main:@5,main:@9,main:@10) -> float_type, {64, 72}, {72, 1}: 0.0218606ms, 0.187374% main:@12 = hip::hip_copy_literal[id=main:@literal:150] -> float_type, {128, 16}, {16, 1}: 0.00090794ms, 0.00778223% 0MZVmrKeC6:0 = @param:0MZVmrKeC6:0 -> int32_type, {64, 10}, {10, 1}: 0.0005494ms, 0.00470908% main:@14 = unsqueeze[axes={2},steps={}](0MZVmrKeC6:0) -> int32_type, {64, 10, 1}, {10, 1, 1}: 0.00086602ms, 0.00742292% main:@15 = load[offset=0,end=2560](main:@1) -> float_type, {64, 10, 1}, {10, 1, 1}: 0.00049112ms, 0.00420954% main:@16 = gpu::code_object[code_object=13520,symbol_name=convert_kernel,global=320,local=256,](main:@14,main:@15) -> float_type, {64, 10, 1}, {10, 1, 1}: 0.0131541ms, 0.112747% main:@17 = multibroadcast[out_lens={64, 10, 16}](main:@16) -> float_type, {64, 10, 16}, {10, 1, 0}: 0.0008067ms, 0.00691447% WsiwEgbYlT:0 = @param:WsiwEgbYlT:0 -> float_type, {64, 10, 17}, {170, 17, 1}: 0.00045514ms, 0.00390114% main:@19 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={64, 10, 17}](WsiwEgbYlT:0) -> float_type, {64, 10, 16}, {170, 17, 1}: 0.00094302ms, 0.00808291% main:@20 = load[offset=59392,end=100352](main:@1) -> float_type, {64, 10, 16}, {160, 16, 1}: 0.0004937ms, 0.00423165% main:@21 = gpu::code_object[code_object=13512,symbol_name=add_kernel,global=10240,local=256,](main:@19,main:@17,main:@20) -> float_type, {64, 10, 16}, {160, 16, 1}: 0.0154476ms, 0.132406% main:@22 = load[offset=0,end=59392](main:@1) -> float_type, {64, 232}, {232, 1}: 0.0006379ms, 0.00546764% main:@23 = reshape_lazy[dims={-1, 160}](main:@21) -> float_type, {64, 160}, {160, 1}: 0.00069234ms, 0.00593426% main:@24 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=7424,local=256,](main:@11,main:@23,main:@22) -> float_type, {64, 232}, {232, 1}: 0.015246ms, 0.130678% main:@25 = hip::hip_copy_literal[id=main:@literal:157] -> float_type, {232, 128}, {128, 1}: 0.00087684ms, 0.00751566% main:@26 = load[offset=92160,end=124928](main:@1) -> float_type, {64, 128}, {128, 1}: 0.00047448ms, 0.00406691% main:@27 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@24,main:@25,main:@26) -> float_type, {64, 128}, {128, 1}: 0.0305088ms, 0.261501% main:@28 = load[offset=59392,end=92160](main:@1) -> float_type, {64, 128}, {128, 1}: 0.00063764ms, 0.00546541% main:@29 = gpu::code_object[code_object=13520,symbol_name=log_add_kernel,global=4096,local=256,](main:@27,main:@28) -> float_type, {64, 128}, {128, 1}: 0.0136203ms, 0.116744% main:@30 = hip::hip_copy_literal[id=main:@literal:121] -> float_type, {128, 16}, {16, 1}: 0.00078826ms, 0.00675642% main:@31 = load[offset=818176,end=822272](main:@1) -> float_type, {64, 16}, {16, 1}: 0.0004612ms, 0.00395309% main:@32 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@29,main:@30,main:@31) -> float_type, {64, 16}, {16, 1}: 0.0213949ms, 0.183383% main:@33 = load[offset=59392,end=61952](main:@1) -> float_type, {64, 10, 1}, {10, 1, 1}: 0.00066092ms, 0.00566495% YK10Jin6ca:0 = @param:YK10Jin6ca:0 -> int32_type, {64, 10}, {10, 1}: 0.00052154ms, 0.00447028% main:@35 = unsqueeze[axes={2},steps={}](YK10Jin6ca:0) -> int32_type, {64, 10, 1}, {10, 1, 1}: 0.00059736ms, 0.00512016% main:@36 = gpu::code_object[code_object=13520,symbol_name=convert_kernel,global=320,local=256,](main:@35,main:@33) -> float_type, {64, 10, 1}, {10, 1, 1}: 0.0137215ms, 0.117611% main:@37 = multibroadcast[out_lens={64, 10, 16}](main:@36) -> float_type, {64, 10, 16}, {10, 1, 0}: 0.00067058ms, 0.00574775% rlh04eiWzL:0 = @param:rlh04eiWzL:0 -> float_type, {64, 10, 17}, {170, 17, 1}: 0.00046882ms, 0.0040184% main:@39 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={64, 10, 17}](rlh04eiWzL:0) -> float_type, {64, 10, 16}, {170, 17, 1}: 0.0008134ms, 0.0069719% main:@40 = load[offset=118784,end=159744](main:@1) -> float_type, {64, 10, 16}, {160, 16, 1}: 0.00048814ms, 0.004184% main:@41 = gpu::code_object[code_object=13512,symbol_name=pow_kernel,global=10240,local=256,](main:@39,main:@37,main:@40) -> float_type, {64, 10, 16}, {160, 16, 1}: 0.0164755ms, 0.141217% main:@42 = load[offset=59392,end=118784](main:@1) -> float_type, {64, 232}, {232, 1}: 0.00060184ms, 0.00515856% main:@43 = reshape_lazy[dims={-1, 160}](main:@41) -> float_type, {64, 160}, {160, 1}: 0.00054836ms, 0.00470016% main:@44 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=7424,local=256,](main:@11,main:@43,main:@42) -> float_type, {64, 232}, {232, 1}: 0.0152555ms, 0.13076% main:@45 = hip::hip_copy_literal[id=main:@literal:117] -> float_type, {232, 128}, {128, 1}: 0.00081032ms, 0.0069455% main:@46 = load[offset=151552,end=184320](main:@1) -> float_type, {64, 128}, {128, 1}: 0.00048812ms, 0.00418383% main:@47 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@44,main:@45,main:@46) -> float_type, {64, 128}, {128, 1}: 0.0250475ms, 0.21469% main:@48 = load[offset=118784,end=151552](main:@1) -> float_type, {64, 128}, {128, 1}: 0.0006665ms, 0.00571278% main:@49 = gpu::code_object[code_object=13520,symbol_name=rsqrt_pow_kernel,global=4096,local=256,](main:@47,main:@48) -> float_type, {64, 128}, {128, 1}: 0.016567ms, 0.142% main:@50 = load[offset=822272,end=826368](main:@1) -> float_type, {64, 16}, {16, 1}: 0.0006404ms, 0.00548906% main:@51 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@49,main:@12,main:@50) -> float_type, {64, 16}, {16, 1}: 0.0195287ms, 0.167386% uVm4kOBe8l:0 = @param:uVm4kOBe8l:0 -> int32_type, {64, 50}, {50, 1}: 0.00067864ms, 0.00581683% main:@53 = unsqueeze[axes={2},steps={}](uVm4kOBe8l:0) -> int32_type, {64, 50, 1}, {50, 1, 1}: 0.00059156ms, 0.00507044% main:@54 = load[offset=118784,end=131584](main:@1) -> float_type, {64, 50, 1}, {50, 1, 1}: 0.00049784ms, 0.00426714% main:@55 = gpu::code_object[code_object=13520,symbol_name=convert_kernel,global=1600,local=256,](main:@53,main:@54) -> float_type, {64, 50, 1}, {50, 1, 1}: 0.0135163ms, 0.115852% main:@56 = multibroadcast[out_lens={64, 50, 16}](main:@55) -> float_type, {64, 50, 16}, {50, 1, 0}: 0.00068642ms, 0.00588352% FkwYpLOMG4:0 = @param:FkwYpLOMG4:0 -> float_type, {64, 50, 17}, {850, 17, 1}: 0.0004642ms, 0.0039788% main:@58 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={64, 50, 17}](FkwYpLOMG4:0) -> float_type, {64, 50, 16}, {850, 17, 1}: 0.00086692ms, 0.00743064% main:@59 = load[offset=131584,end=336384](main:@1) -> float_type, {64, 50, 16}, {800, 16, 1}: 0.00048518ms, 0.00415863% main:@60 = gpu::code_object[code_object=13512,symbol_name=div_kernel,global=51200,local=256,](main:@58,main:@56,main:@59) -> float_type, {64, 50, 16}, {800, 16, 1}: 0.0185558ms, 0.159048% main:@61 = load[offset=353280,end=576512](main:@1) -> float_type, {64, 872}, {872, 1}: 0.00060296ms, 0.00516816% main:@62 = reshape_lazy[dims={-1, 800}](main:@60) -> float_type, {64, 800}, {800, 1}: 0.0005553ms, 0.00475965% main:@63 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=27904,local=256,](main:@11,main:@62,main:@61) -> float_type, {64, 872}, {872, 1}: 0.0179247ms, 0.153638% main:@64 = hip::hip_copy_literal[id=main:@literal:89] -> float_type, {872, 128}, {128, 1}: 0.00078426ms, 0.00672213% main:@65 = load[offset=151552,end=184320](main:@1) -> float_type, {64, 128}, {128, 1}: 0.0004678ms, 0.00400966% main:@66 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@63,main:@64,main:@65) -> float_type, {64, 128}, {128, 1}: 0.0277884ms, 0.238183% main:@67 = load[offset=118784,end=151552](main:@1) -> float_type, {64, 128}, {128, 1}: 0.00061788ms, 0.00529604% main:@68 = gpu::code_object[code_object=13520,symbol_name=neg_div_kernel,global=4096,local=256,](main:@66,main:@67) -> float_type, {64, 128}, {128, 1}: 0.0135884ms, 0.116471% main:@69 = hip::hip_copy_literal[id=main:@literal:100] -> float_type, {128, 16}, {16, 1}: 0.00077972ms, 0.00668322% main:@70 = load[offset=826368,end=830464](main:@1) -> float_type, {64, 16}, {16, 1}: 0.0004704ms, 0.00403194% main:@71 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@68,main:@69,main:@70) -> float_type, {64, 16}, {16, 1}: 0.0199297ms, 0.170824% main:@72 = load[offset=118784,end=121344](main:@1) -> float_type, {64, 10, 1}, {10, 1, 1}: 0.00062122ms, 0.00532467% fFjVmAbhCX:0 = @param:fFjVmAbhCX:0 -> int32_type, {64, 10}, {10, 1}: 0.00050574ms, 0.00433485% main:@74 = unsqueeze[axes={2},steps={}](fFjVmAbhCX:0) -> int32_type, {64, 10, 1}, {10, 1, 1}: 0.0005526ms, 0.0047365% main:@75 = gpu::code_object[code_object=13520,symbol_name=convert_kernel,global=320,local=256,](main:@74,main:@72) -> float_type, {64, 10, 1}, {10, 1, 1}: 0.0135816ms, 0.116412% main:@76 = multibroadcast[out_lens={64, 10, 16}](main:@75) -> float_type, {64, 10, 16}, {10, 1, 0}: 0.00066552ms, 0.00570438% pjwGhTSAys:0 = @param:pjwGhTSAys:0 -> float_type, {64, 10, 17}, {170, 17, 1}: 0.00047176ms, 0.0040436% main:@78 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={64, 10, 17}](pjwGhTSAys:0) -> float_type, {64, 10, 16}, {170, 17, 1}: 0.0008765ms, 0.00751275% main:@79 = load[offset=121344,end=162304](main:@1) -> float_type, {64, 10, 16}, {160, 16, 1}: 0.00048708ms, 0.00417491% main:@80 = gpu::code_object[code_object=13512,symbol_name=pow_kernel,global=10240,local=256,](main:@78,main:@76,main:@79) -> float_type, {64, 10, 16}, {160, 16, 1}: 0.0161612ms, 0.138522% main:@81 = load[offset=293888,end=353280](main:@1) -> float_type, {64, 232}, {232, 1}: 0.00062238ms, 0.00533461% main:@82 = reshape_lazy[dims={-1, 160}](main:@80) -> float_type, {64, 160}, {160, 1}: 0.00054292ms, 0.00465353% main:@83 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=7424,local=256,](main:@11,main:@82,main:@81) -> float_type, {64, 232}, {232, 1}: 0.0149772ms, 0.128374% main:@84 = hip::hip_copy_literal[id=main:@literal:130] -> float_type, {232, 128}, {128, 1}: 0.00079796ms, 0.00683956% main:@85 = load[offset=151552,end=184320](main:@1) -> float_type, {64, 128}, {128, 1}: 0.0004629ms, 0.00396766% main:@86 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@83,main:@84,main:@85) -> float_type, {64, 128}, {128, 1}: 0.0249474ms, 0.213832% main:@87 = load[offset=118784,end=151552](main:@1) -> float_type, {64, 128}, {128, 1}: 0.00064926ms, 0.00556501% main:@88 = gpu::code_object[code_object=13520,symbol_name=log_add_kernel,global=4096,local=256,](main:@86,main:@87) -> float_type, {64, 128}, {128, 1}: 0.0135565ms, 0.116197% main:@89 = load[offset=830464,end=834560](main:@1) -> float_type, {64, 16}, {16, 1}: 0.0005938ms, 0.00508964% main:@90 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@88,main:@12,main:@89) -> float_type, {64, 16}, {16, 1}: 0.0193816ms, 0.166126% v3TAQfOsga:0 = @param:v3TAQfOsga:0 -> int32_type, {64, 15}, {15, 1}: 0.00070636ms, 0.00605443% main:@92 = unsqueeze[axes={2},steps={}](v3TAQfOsga:0) -> int32_type, {64, 15, 1}, {15, 1, 1}: 0.0005833ms, 0.00499964% main:@93 = load[offset=118784,end=122624](main:@1) -> float_type, {64, 15, 1}, {15, 1, 1}: 0.0004849ms, 0.00415623% main:@94 = gpu::code_object[code_object=13520,symbol_name=convert_kernel,global=480,local=256,](main:@92,main:@93) -> float_type, {64, 15, 1}, {15, 1, 1}: 0.0133774ms, 0.114661% main:@95 = load[offset=214016,end=290816](main:@1) -> float_type, {64, 15, 20}, {300, 20, 1}: 0.00063252ms, 0.00542152% main:@96 = multibroadcast[out_lens={64, 15, 20}](main:@94) -> float_type, {64, 15, 20}, {15, 1, 0}: 0.0005532ms, 0.00474165% 1qyHnroO6I:0 = @param:1qyHnroO6I:0 -> float_type, {64, 15, 21}, {315, 21, 1}: 0.00051582ms, 0.00442125% main:@98 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={64, 15, 21}](1qyHnroO6I:0) -> float_type, {64, 15, 20}, {315, 21, 1}: 0.00082342ms, 0.00705779% main:@99 = gpu::code_object[code_object=13512,symbol_name=add_kernel,global=19200,local=256,](main:@98,main:@96,main:@95) -> float_type, {64, 15, 20}, {300, 20, 1}: 0.0165169ms, 0.141571% main:@100 = load[offset=118784,end=214016](main:@1) -> float_type, {64, 372}, {372, 1}: 0.00063774ms, 0.00546627% main:@101 = reshape_lazy[dims={-1, 300}](main:@99) -> float_type, {64, 300}, {300, 1}: 0.00053032ms, 0.00454554% main:@102 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=11904,local=256,](main:@11,main:@101,main:@100) -> float_type, {64, 372}, {372, 1}: 0.0157031ms, 0.134596% main:@103 = hip::hip_copy_literal[id=main:@literal:105] -> float_type, {372, 128}, {128, 1}: 0.00074624ms, 0.00639625% main:@104 = load[offset=246784,end=279552](main:@1) -> float_type, {64, 128}, {128, 1}: 0.00046468ms, 0.00398291% main:@105 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@102,main:@103,main:@104) -> float_type, {64, 128}, {128, 1}: 0.0271694ms, 0.232878% main:@106 = load[offset=214016,end=246784](main:@1) -> float_type, {64, 128}, {128, 1}: 0.0006656ms, 0.00570506% main:@107 = gpu::code_object[code_object=13520,symbol_name=log_add_kernel,global=4096,local=256,](main:@105,main:@106) -> float_type, {64, 128}, {128, 1}: 0.0136032ms, 0.116597% main:@108 = load[offset=834560,end=838656](main:@1) -> float_type, {64, 16}, {16, 1}: 0.00060434ms, 0.00517998% main:@109 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@107,main:@12,main:@108) -> float_type, {64, 16}, {16, 1}: 0.019707ms, 0.168915% main:@110 = load[offset=214016,end=217856](main:@1) -> float_type, {64, 15, 1}, {15, 1, 1}: 0.00062808ms, 0.00538347% kJg5ca1rqS:0 = @param:kJg5ca1rqS:0 -> int32_type, {64, 15}, {15, 1}: 0.00049896ms, 0.00427674% main:@112 = unsqueeze[axes={2},steps={}](kJg5ca1rqS:0) -> int32_type, {64, 15, 1}, {15, 1, 1}: 0.00055148ms, 0.0047269% main:@113 = gpu::code_object[code_object=13520,symbol_name=convert_kernel,global=480,local=256,](main:@112,main:@110) -> float_type, {64, 15, 1}, {15, 1, 1}: 0.0135517ms, 0.116156% main:@114 = load[offset=576512,end=684032](main:@1) -> float_type, {64, 15, 28}, {420, 28, 1}: 0.0006162ms, 0.00528164% main:@115 = multibroadcast[out_lens={64, 15, 28}](main:@113) -> float_type, {64, 15, 28}, {15, 1, 0}: 0.00051232ms, 0.00439125% fQkYLvqDjr:0 = @param:fQkYLvqDjr:0 -> float_type, {64, 15, 93}, {1395, 93, 1}: 0.00046546ms, 0.0039896% main:@117 = slice[axes={0, 1, 2},starts={0, 0, 65},ends={64, 15, 93}](fQkYLvqDjr:0) -> float_type, {64, 15, 28}, {1395, 93, 1}: 0.00078564ms, 0.00673396% main:@118 = gpu::code_object[code_object=13512,symbol_name=add_kernel,global=26880,local=256,](main:@117,main:@115,main:@114) -> float_type, {64, 15, 28}, {420, 28, 1}: 0.0179514ms, 0.153867% main:@119 = reshape_lazy[dims={-1, 420}](main:@118) -> float_type, {64, 420}, {420, 1}: 0.00068592ms, 0.00587923% main:@120 = load[offset=1990912,end=2116864](main:@1) -> float_type, {64, 492}, {492, 1}: 0.00049562ms, 0.00424811% main:@121 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=15744,local=256,](main:@11,main:@119,main:@120) -> float_type, {64, 492}, {492, 1}: 0.0160887ms, 0.137902% main:@122 = hip::hip_copy_literal[id=main:@literal:127] -> float_type, {492, 128}, {128, 1}: 0.00074438ms, 0.00638031% main:@123 = load[offset=246784,end=279552](main:@1) -> float_type, {64, 128}, {128, 1}: 0.00046998ms, 0.00402834% main:@124 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@121,main:@122,main:@123) -> float_type, {64, 128}, {128, 1}: 0.0316001ms, 0.270854% main:@125 = load[offset=214016,end=246784](main:@1) -> float_type, {64, 128}, {128, 1}: 0.00061714ms, 0.0052897% main:@126 = gpu::code_object[code_object=13520,symbol_name=rsqrt_pow_kernel,global=4096,local=256,](main:@124,main:@125) -> float_type, {64, 128}, {128, 1}: 0.0165807ms, 0.142118% main:@127 = load[offset=838656,end=842752](main:@1) -> float_type, {64, 16}, {16, 1}: 0.00064874ms, 0.00556055% main:@128 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@126,main:@12,main:@127) -> float_type, {64, 16}, {16, 1}: 0.0193161ms, 0.165564% main:@129 = load[offset=214016,end=229376](main:@1) -> float_type, {64, 60, 1}, {60, 1, 1}: 0.00063566ms, 0.00544844% Ouj93S4aWb:0 = @param:Ouj93S4aWb:0 -> int32_type, {64, 60}, {60, 1}: 0.00049552ms, 0.00424725% main:@131 = unsqueeze[axes={2},steps={}](Ouj93S4aWb:0) -> int32_type, {64, 60, 1}, {60, 1, 1}: 0.00055184ms, 0.00472999% main:@132 = gpu::code_object[code_object=13520,symbol_name=convert_kernel,global=1920,local=256,](main:@131,main:@129) -> float_type, {64, 60, 1}, {60, 1, 1}: 0.0136264ms, 0.116796% main:@133 = load[offset=842752,end=1334272](main:@1) -> float_type, {64, 60, 32}, {1920, 32, 1}: 0.00064104ms, 0.00549455% main:@134 = multibroadcast[out_lens={64, 60, 32}](main:@132) -> float_type, {64, 60, 32}, {60, 1, 0}: 0.00050834ms, 0.00435714% fsm0Ck1QGl:0 = @param:fsm0Ck1QGl:0 -> float_type, {64, 60, 97}, {5820, 97, 1}: 0.0005567ms, 0.00477165% main:@136 = slice[axes={0, 1, 2},starts={0, 0, 65},ends={64, 60, 97}](fsm0Ck1QGl:0) -> float_type, {64, 60, 32}, {5820, 97, 1}: 0.00085288ms, 0.0073103% main:@137 = gpu::code_object[code_object=13512,symbol_name=div_kernel,global=122880,local=256,](main:@136,main:@134,main:@133) -> float_type, {64, 60, 32}, {1920, 32, 1}: 0.0187049ms, 0.160325% main:@138 = load[offset=2482432,end=2992384](main:@1) -> float_type, {64, 1992}, {1992, 1}: 0.00062498ms, 0.0053569% main:@139 = reshape_lazy[dims={-1, 1920}](main:@137) -> float_type, {64, 1920}, {1920, 1}: 0.00054134ms, 0.00463999% main:@140 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=63744,local=256,](main:@11,main:@139,main:@138) -> float_type, {64, 1992}, {1992, 1}: 0.0188853ms, 0.161872% main:@141 = hip::hip_copy_literal[id=main:@literal:109] -> float_type, {1992, 128}, {128, 1}: 0.0008019ms, 0.00687333% main:@142 = load[offset=246784,end=279552](main:@1) -> float_type, {64, 128}, {128, 1}: 0.00048822ms, 0.00418468% main:@143 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@140,main:@141,main:@142) -> float_type, {64, 128}, {128, 1}: 0.0302792ms, 0.259532% main:@144 = load[offset=214016,end=246784](main:@1) -> float_type, {64, 128}, {128, 1}: 0.00062352ms, 0.00534438% main:@145 = gpu::code_object[code_object=13520,symbol_name=neg_pow_kernel,global=4096,local=256,](main:@143,main:@144) -> float_type, {64, 128}, {128, 1}: 0.0164467ms, 0.140969% main:@146 = load[offset=842752,end=846848](main:@1) -> float_type, {64, 16}, {16, 1}: 0.00061944ms, 0.00530941% main:@147 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@145,main:@12,main:@146) -> float_type, {64, 16}, {16, 1}: 0.0195062ms, 0.167193% main:@148 = load[offset=214016,end=226816](main:@1) -> float_type, {64, 50, 1}, {50, 1, 1}: 0.00063372ms, 0.00543181% TWPHfwn7r5:0 = @param:TWPHfwn7r5:0 -> int32_type, {64, 50}, {50, 1}: 0.0005808ms, 0.00497822% main:@150 = unsqueeze[axes={2},steps={}](TWPHfwn7r5:0) -> int32_type, {64, 50, 1}, {50, 1, 1}: 0.00053752ms, 0.00460725% main:@151 = gpu::code_object[code_object=13520,symbol_name=convert_kernel,global=1600,local=256,](main:@150,main:@148) -> float_type, {64, 50, 1}, {50, 1, 1}: 0.0135927ms, 0.116507% main:@152 = load[offset=846848,end=1051648](main:@1) -> float_type, {64, 50, 16}, {800, 16, 1}: 0.0006349ms, 0.00544192% main:@153 = multibroadcast[out_lens={64, 50, 16}](main:@151) -> float_type, {64, 50, 16}, {50, 1, 0}: 0.00051528ms, 0.00441662% NoPO6DfgwG:0 = @param:NoPO6DfgwG:0 -> float_type, {64, 50, 17}, {850, 17, 1}: 0.00051856ms, 0.00444474% main:@155 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={64, 50, 17}](NoPO6DfgwG:0) -> float_type, {64, 50, 16}, {850, 17, 1}: 0.0008344ms, 0.0071519% main:@156 = gpu::code_object[code_object=13512,symbol_name=sub_kernel,global=51200,local=256,](main:@155,main:@153,main:@152) -> float_type, {64, 50, 16}, {800, 16, 1}: 0.0183793ms, 0.157535% main:@157 = load[offset=576512,end=799744](main:@1) -> float_type, {64, 872}, {872, 1}: 0.00060846ms, 0.0052153% main:@158 = reshape_lazy[dims={-1, 800}](main:@156) -> float_type, {64, 800}, {800, 1}: 0.00052316ms, 0.00448416% main:@159 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=27904,local=256,](main:@11,main:@158,main:@157) -> float_type, {64, 872}, {872, 1}: 0.017971ms, 0.154035% main:@160 = hip::hip_copy_literal[id=main:@literal:61] -> float_type, {872, 128}, {128, 1}: 0.00078434ms, 0.00672282% main:@161 = load[offset=246784,end=279552](main:@1) -> float_type, {64, 128}, {128, 1}: 0.00047636ms, 0.00408303% main:@162 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@159,main:@160,main:@161) -> float_type, {64, 128}, {128, 1}: 0.0254037ms, 0.217743% main:@163 = load[offset=214016,end=246784](main:@1) -> float_type, {64, 128}, {128, 1}: 0.00061612ms, 0.00528095% main:@164 = gpu::code_object[code_object=13520,symbol_name=tanh_div_kernel,global=4096,local=256,](main:@162,main:@163) -> float_type, {64, 128}, {128, 1}: 0.0142323ms, 0.121989% main:@165 = hip::hip_copy_literal[id=main:@literal:50] -> float_type, {128, 16}, {16, 1}: 0.00073372ms, 0.00628894% main:@166 = load[offset=846848,end=850944](main:@1) -> float_type, {64, 16}, {16, 1}: 0.0004603ms, 0.00394537% main:@167 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@164,main:@165,main:@166) -> float_type, {64, 16}, {16, 1}: 0.0195784ms, 0.167812% loNH8y6bTn:0 = @param:loNH8y6bTn:0 -> int32_type, {64, 15}, {15, 1}: 0.0006667ms, 0.00571449% main:@169 = unsqueeze[axes={2},steps={}](loNH8y6bTn:0) -> int32_type, {64, 15, 1}, {15, 1, 1}: 0.00057776ms, 0.00495216% main:@170 = load[offset=214016,end=217856](main:@1) -> float_type, {64, 15, 1}, {15, 1, 1}: 0.00050492ms, 0.00432782% main:@171 = gpu::code_object[code_object=13520,symbol_name=convert_kernel,global=480,local=256,](main:@169,main:@170) -> float_type, {64, 15, 1}, {15, 1, 1}: 0.013393ms, 0.114796% main:@172 = multibroadcast[out_lens={64, 15, 16}](main:@171) -> float_type, {64, 15, 16}, {15, 1, 0}: 0.00066304ms, 0.00568312% 4Rg0WA19ot:0 = @param:4Rg0WA19ot:0 -> float_type, {64, 15, 17}, {255, 17, 1}: 0.00055658ms, 0.00477062% main:@174 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={64, 15, 17}](4Rg0WA19ot:0) -> float_type, {64, 15, 16}, {255, 17, 1}: 0.00088134ms, 0.00755424% main:@175 = load[offset=850944,end=912384](main:@1) -> float_type, {64, 15, 16}, {240, 16, 1}: 0.0004787ms, 0.00410308% main:@176 = gpu::code_object[code_object=13512,symbol_name=add_kernel,global=15360,local=256,](main:@174,main:@172,main:@175) -> float_type, {64, 15, 16}, {240, 16, 1}: 0.0156555ms, 0.134188% main:@177 = reshape_lazy[dims={-1, 240}](main:@176) -> float_type, {64, 240}, {240, 1}: 0.0006826ms, 0.00585077% main:@178 = load[offset=214016,end=293888](main:@1) -> float_type, {64, 312}, {312, 1}: 0.00045808ms, 0.00392634% main:@179 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=9984,local=256,](main:@11,main:@177,main:@178) -> float_type, {64, 312}, {312, 1}: 0.0155721ms, 0.133473% main:@180 = hip::hip_copy_literal[id=main:@literal:139] -> float_type, {312, 128}, {128, 1}: 0.00082832ms, 0.00709978% main:@181 = load[offset=887808,end=920576](main:@1) -> float_type, {64, 128}, {128, 1}: 0.00050006ms, 0.00428617% main:@182 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@179,main:@180,main:@181) -> float_type, {64, 128}, {128, 1}: 0.0259111ms, 0.222092% main:@183 = load[offset=855040,end=887808](main:@1) -> float_type, {64, 128}, {128, 1}: 0.00062146ms, 0.00532672% main:@184 = gpu::code_object[code_object=13520,symbol_name=neg_pow_kernel,global=4096,local=256,](main:@182,main:@183) -> float_type, {64, 128}, {128, 1}: 0.016384ms, 0.140432% main:@185 = hip::hip_copy_literal[id=main:@literal:91] -> float_type, {128, 16}, {16, 1}: 0.00076784ms, 0.00658139% main:@186 = load[offset=850944,end=855040](main:@1) -> float_type, {64, 16}, {16, 1}: 0.00048058ms, 0.0041192% main:@187 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@184,main:@185,main:@186) -> float_type, {64, 16}, {16, 1}: 0.0194836ms, 0.167% vOI1hWzBDK:0 = @param:vOI1hWzBDK:0 -> int32_type, {64, 50}, {50, 1}: 0.00072056ms, 0.00617614% main:@189 = unsqueeze[axes={2},steps={}](vOI1hWzBDK:0) -> int32_type, {64, 50, 1}, {50, 1, 1}: 0.0005511ms, 0.00472365% main:@190 = load[offset=855040,end=867840](main:@1) -> float_type, {64, 50, 1}, {50, 1, 1}: 0.0005031ms, 0.00431222% main:@191 = gpu::code_object[code_object=13520,symbol_name=convert_kernel,global=1600,local=256,](main:@189,main:@190) -> float_type, {64, 50, 1}, {50, 1, 1}: 0.0137753ms, 0.118072% main:@192 = multibroadcast[out_lens={64, 50, 40}](main:@191) -> float_type, {64, 50, 40}, {50, 1, 0}: 0.00066618ms, 0.00571003% UrTfLOxlV3:0 = @param:UrTfLOxlV3:0 -> float_type, {64, 50, 41}, {2050, 41, 1}: 0.0005046ms, 0.00432508% main:@194 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={64, 50, 41}](UrTfLOxlV3:0) -> float_type, {64, 50, 40}, {2050, 41, 1}: 0.00097344ms, 0.00834365% main:@195 = load[offset=867840,end=1379840](main:@1) -> float_type, {64, 50, 40}, {2000, 40, 1}: 0.0004904ms, 0.00420337% main:@196 = gpu::code_object[code_object=13512,symbol_name=sub_kernel,global=128000,local=256,](main:@194,main:@192,main:@195) -> float_type, {64, 50, 40}, {2000, 40, 1}: 0.0186191ms, 0.15959% main:@197 = reshape_lazy[dims={-1, 2000}](main:@196) -> float_type, {64, 2000}, {2000, 1}: 0.00066114ms, 0.00566683% main:@198 = load[offset=4018432,end=4548864](main:@1) -> float_type, {64, 2072}, {2072, 1}: 0.00047826ms, 0.00409931% main:@199 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=66304,local=256,](main:@11,main:@197,main:@198) -> float_type, {64, 2072}, {2072, 1}: 0.0191698ms, 0.16431% main:@200 = hip::hip_copy_literal[id=main:@literal:85] -> float_type, {2072, 128}, {128, 1}: 0.00078494ms, 0.00672796% main:@201 = load[offset=891904,end=924672](main:@1) -> float_type, {64, 128}, {128, 1}: 0.0004686ms, 0.00401651% main:@202 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@199,main:@200,main:@201) -> float_type, {64, 128}, {128, 1}: 0.0307553ms, 0.263613% main:@203 = load[offset=859136,end=891904](main:@1) -> float_type, {64, 128}, {128, 1}: 0.00062626ms, 0.00536787% main:@204 = gpu::code_object[code_object=13520,symbol_name=log_add_kernel,global=4096,local=256,](main:@202,main:@203) -> float_type, {64, 128}, {128, 1}: 0.0134728ms, 0.11548% main:@205 = hip::hip_copy_literal[id=main:@literal:79] -> float_type, {128, 16}, {16, 1}: 0.00080738ms, 0.0069203% main:@206 = load[offset=855040,end=859136](main:@1) -> float_type, {64, 16}, {16, 1}: 0.0004689ms, 0.00401909% main:@207 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@204,main:@205,main:@206) -> float_type, {64, 16}, {16, 1}: 0.0194025ms, 0.166305% vyDGpmRhC3:0 = @param:vyDGpmRhC3:0 -> int32_type, {64, 60}, {60, 1}: 0.0006521ms, 0.00558935% main:@209 = unsqueeze[axes={2},steps={}](vyDGpmRhC3:0) -> int32_type, {64, 60, 1}, {60, 1, 1}: 0.0005493ms, 0.00470822% main:@210 = load[offset=859136,end=874496](main:@1) -> float_type, {64, 60, 1}, {60, 1, 1}: 0.00047954ms, 0.00411028% main:@211 = gpu::code_object[code_object=13520,symbol_name=convert_kernel,global=1920,local=256,](main:@209,main:@210) -> float_type, {64, 60, 1}, {60, 1, 1}: 0.0135637ms, 0.116258% main:@212 = multibroadcast[out_lens={64, 60, 16}](main:@211) -> float_type, {64, 60, 16}, {60, 1, 0}: 0.00065496ms, 0.00561386% y5fPVEH2j4:0 = @param:y5fPVEH2j4:0 -> float_type, {64, 60, 17}, {1020, 17, 1}: 0.0005656ms, 0.00484793% main:@214 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={64, 60, 17}](y5fPVEH2j4:0) -> float_type, {64, 60, 16}, {1020, 17, 1}: 0.00084946ms, 0.00728098% main:@215 = load[offset=874496,end=1120256](main:@1) -> float_type, {64, 60, 16}, {960, 16, 1}: 0.00046202ms, 0.00396012% main:@216 = gpu::code_object[code_object=13512,symbol_name=add_kernel,global=61440,local=256,](main:@214,main:@212,main:@215) -> float_type, {64, 60, 16}, {960, 16, 1}: 0.0183468ms, 0.157256% main:@217 = reshape_lazy[dims={-1, 960}](main:@216) -> float_type, {64, 960}, {960, 1}: 0.00067308ms, 0.00576918% main:@218 = load[offset=3531008,end=3795200](main:@1) -> float_type, {64, 1032}, {1032, 1}: 0.0004564ms, 0.00391194% main:@219 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=33024,local=256,](main:@11,main:@217,main:@218) -> float_type, {64, 1032}, {1032, 1}: 0.018288ms, 0.156752% main:@220 = hip::hip_copy_literal[id=main:@literal:56] -> float_type, {1032, 128}, {128, 1}: 0.0007735ms, 0.00662991% main:@221 = load[offset=896000,end=928768](main:@1) -> float_type, {64, 128}, {128, 1}: 0.0004721ms, 0.00404651% main:@222 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@219,main:@220,main:@221) -> float_type, {64, 128}, {128, 1}: 0.0260327ms, 0.223134% main:@223 = load[offset=863232,end=896000](main:@1) -> float_type, {64, 128}, {128, 1}: 0.00061708ms, 0.00528918% main:@224 = gpu::code_object[code_object=13520,symbol_name=log_sub_kernel,global=4096,local=256,](main:@222,main:@223) -> float_type, {64, 128}, {128, 1}: 0.0135264ms, 0.115939% main:@225 = hip::hip_copy_literal[id=main:@literal:90] -> float_type, {128, 16}, {16, 1}: 0.00074056ms, 0.00634757% main:@226 = load[offset=859136,end=863232](main:@1) -> float_type, {64, 16}, {16, 1}: 0.00050242ms, 0.0043064% main:@227 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@224,main:@225,main:@226) -> float_type, {64, 16}, {16, 1}: 0.0195742ms, 0.167776% yDES0Cj7Td:0 = @param:yDES0Cj7Td:0 -> int32_type, {64, 10}, {10, 1}: 0.00068058ms, 0.00583346% main:@229 = unsqueeze[axes={2},steps={}](yDES0Cj7Td:0) -> int32_type, {64, 10, 1}, {10, 1, 1}: 0.0005468ms, 0.00468679% main:@230 = load[offset=904192,end=906752](main:@1) -> float_type, {64, 10, 1}, {10, 1, 1}: 0.00049632ms, 0.00425411% main:@231 = gpu::code_object[code_object=13520,symbol_name=convert_kernel,global=320,local=256,](main:@229,main:@230) -> float_type, {64, 10, 1}, {10, 1, 1}: 0.0132021ms, 0.113159% main:@232 = load[offset=863232,end=904192](main:@1) -> float_type, {64, 10, 16}, {160, 16, 1}: 0.00064104ms, 0.00549455% main:@233 = multibroadcast[out_lens={64, 10, 16}](main:@231) -> float_type, {64, 10, 16}, {10, 1, 0}: 0.00053572ms, 0.00459182% r1GsND0okc:0 = @param:r1GsND0okc:0 -> float_type, {64, 10, 17}, {170, 17, 1}: 0.00046762ms, 0.00400811% main:@235 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={64, 10, 17}](r1GsND0okc:0) -> float_type, {64, 10, 16}, {170, 17, 1}: 0.0008385ms, 0.00718704% main:@236 = gpu::code_object[code_object=13512,symbol_name=sub_kernel,global=10240,local=256,](main:@235,main:@233,main:@232) -> float_type, {64, 10, 16}, {160, 16, 1}: 0.0150121ms, 0.128673% main:@237 = load[offset=2178304,end=2237696](main:@1) -> float_type, {64, 232}, {232, 1}: 0.00063132ms, 0.00541124% main:@238 = reshape_lazy[dims={-1, 160}](main:@236) -> float_type, {64, 160}, {160, 1}: 0.00053338ms, 0.00457176% main:@239 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=7424,local=256,](main:@11,main:@238,main:@237) -> float_type, {64, 232}, {232, 1}: 0.0148199ms, 0.127026% main:@240 = hip::hip_copy_literal[id=main:@literal:132] -> float_type, {232, 128}, {128, 1}: 0.0007984ms, 0.00684333% main:@241 = load[offset=863232,end=896000](main:@1) -> float_type, {64, 128}, {128, 1}: 0.00046568ms, 0.00399149% main:@242 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@239,main:@240,main:@241) -> float_type, {64, 128}, {128, 1}: 0.0251025ms, 0.215161% main:@243 = hip::hip_copy_literal[id=main:@literal:104] -> float_type, {128, 16}, {16, 1}: 0.0007742ms, 0.00663591% main:@244 = load[offset=896000,end=928768](main:@1) -> float_type, {64, 128}, {128, 1}: 0.00050018ms, 0.0042872% main:@245 = gpu::code_object[code_object=13520,symbol_name=neg_div_kernel,global=4096,local=256,](main:@242,main:@244) -> float_type, {64, 128}, {128, 1}: 0.0135247ms, 0.115924% main:@246 = load[offset=863232,end=867328](main:@1) -> float_type, {64, 16}, {16, 1}: 0.00064944ms, 0.00556655% main:@247 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@245,main:@243,main:@246) -> float_type, {64, 16}, {16, 1}: 0.0200615ms, 0.171953% KOJcBD1Cxa:0 = @param:KOJcBD1Cxa:0 -> int32_type, {64, 50}, {50, 1}: 0.0006492ms, 0.00556449% main:@249 = unsqueeze[axes={2},steps={}](KOJcBD1Cxa:0) -> int32_type, {64, 50, 1}, {50, 1, 1}: 0.00054628ms, 0.00468233% main:@250 = load[offset=1123328,end=1136128](main:@1) -> float_type, {64, 50, 1}, {50, 1, 1}: 0.00050386ms, 0.00431874% main:@251 = gpu::code_object[code_object=13520,symbol_name=convert_kernel,global=1600,local=256,](main:@249,main:@250) -> float_type, {64, 50, 1}, {50, 1, 1}: 0.0134232ms, 0.115054% main:@252 = load[offset=867328,end=1123328](main:@1) -> float_type, {64, 50, 20}, {1000, 20, 1}: 0.0006377ms, 0.00546592% main:@253 = multibroadcast[out_lens={64, 50, 20}](main:@251) -> float_type, {64, 50, 20}, {50, 1, 0}: 0.00053014ms, 0.00454399% gkGLzseUSO:0 = @param:gkGLzseUSO:0 -> float_type, {64, 50, 21}, {1050, 21, 1}: 0.00046034ms, 0.00394572% main:@255 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={64, 50, 21}](gkGLzseUSO:0) -> float_type, {64, 50, 20}, {1050, 21, 1}: 0.0009904ms, 0.00848902% main:@256 = gpu::code_object[code_object=13512,symbol_name=div_kernel,global=64000,local=256,](main:@255,main:@253,main:@252) -> float_type, {64, 50, 20}, {1000, 20, 1}: 0.0184132ms, 0.157825% main:@257 = load[offset=3256576,end=3531008](main:@1) -> float_type, {64, 1072}, {1072, 1}: 0.00059436ms, 0.00509444% main:@258 = reshape_lazy[dims={-1, 1000}](main:@256) -> float_type, {64, 1000}, {1000, 1}: 0.00054162ms, 0.00464239% main:@259 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=34304,local=256,](main:@11,main:@258,main:@257) -> float_type, {64, 1072}, {1072, 1}: 0.0183951ms, 0.15767% main:@260 = hip::hip_copy_literal[id=main:@literal:57] -> float_type, {1072, 128}, {128, 1}: 0.00076462ms, 0.00655379% main:@261 = load[offset=908288,end=941056](main:@1) -> float_type, {64, 128}, {128, 1}: 0.0004882ms, 0.00418451% 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, {64, 128}, {128, 1}: 0.0264932ms, 0.227081% main:@263 = load[offset=875520,end=908288](main:@1) -> float_type, {64, 128}, {128, 1}: 0.00062284ms, 0.00533855% main:@264 = gpu::code_object[code_object=13520,symbol_name=neg_add_kernel,global=4096,local=256,](main:@262,main:@263) -> float_type, {64, 128}, {128, 1}: 0.0133679ms, 0.11458% main:@265 = load[offset=871424,end=875520](main:@1) -> float_type, {64, 16}, {16, 1}: 0.0006032ms, 0.00517021% main:@266 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@264,main:@243,main:@265) -> float_type, {64, 16}, {16, 1}: 0.0194811ms, 0.166978% CedNyHXqE1:0 = @param:CedNyHXqE1:0 -> int32_type, {64, 50}, {50, 1}: 0.00068358ms, 0.00585917% main:@268 = unsqueeze[axes={2},steps={}](CedNyHXqE1:0) -> int32_type, {64, 50, 1}, {50, 1, 1}: 0.00054872ms, 0.00470325% main:@269 = load[offset=1080320,end=1093120](main:@1) -> float_type, {64, 50, 1}, {50, 1, 1}: 0.00048946ms, 0.00419531% main:@270 = gpu::code_object[code_object=13520,symbol_name=convert_kernel,global=1600,local=256,](main:@268,main:@269) -> float_type, {64, 50, 1}, {50, 1, 1}: 0.0135111ms, 0.115808% main:@271 = load[offset=875520,end=1080320](main:@1) -> float_type, {64, 50, 16}, {800, 16, 1}: 0.0006457ms, 0.00553449% main:@272 = multibroadcast[out_lens={64, 50, 16}](main:@270) -> float_type, {64, 50, 16}, {50, 1, 0}: 0.00053086ms, 0.00455016% YsGAJ1iRPS:0 = @param:YsGAJ1iRPS:0 -> float_type, {64, 50, 17}, {850, 17, 1}: 0.00052548ms, 0.00450405% main:@274 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={64, 50, 17}](YsGAJ1iRPS:0) -> float_type, {64, 50, 16}, {850, 17, 1}: 0.00084578ms, 0.00724944% main:@275 = gpu::code_object[code_object=13512,symbol_name=add_kernel,global=51200,local=256,](main:@274,main:@272,main:@271) -> float_type, {64, 50, 16}, {800, 16, 1}: 0.0183646ms, 0.157409% main:@276 = reshape_lazy[dims={-1, 800}](main:@275) -> float_type, {64, 800}, {800, 1}: 0.00067408ms, 0.00577775% main:@277 = load[offset=3795200,end=4018432](main:@1) -> float_type, {64, 872}, {872, 1}: 0.0004931ms, 0.00422651% main:@278 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=27904,local=256,](main:@11,main:@276,main:@277) -> float_type, {64, 872}, {872, 1}: 0.0177977ms, 0.15255% main:@279 = hip::hip_copy_literal[id=main:@literal:159] -> float_type, {872, 128}, {128, 1}: 0.00080928ms, 0.00693659% main:@280 = load[offset=908288,end=941056](main:@1) -> float_type, {64, 128}, {128, 1}: 0.00047416ms, 0.00406417% main:@281 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@278,main:@279,main:@280) -> float_type, {64, 128}, {128, 1}: 0.0246644ms, 0.211406% main:@282 = load[offset=875520,end=908288](main:@1) -> float_type, {64, 128}, {128, 1}: 0.00062252ms, 0.00533581% main:@283 = gpu::code_object[code_object=13520,symbol_name=neg_add_kernel,global=4096,local=256,](main:@281,main:@282) -> float_type, {64, 128}, {128, 1}: 0.0133006ms, 0.114003% main:@284 = hip::hip_copy_literal[id=main:@literal:64] -> float_type, {128, 16}, {16, 1}: 0.00071568ms, 0.00613431% main:@285 = load[offset=867328,end=871424](main:@1) -> float_type, {64, 16}, {16, 1}: 0.00050212ms, 0.00430382% main:@286 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@283,main:@284,main:@285) -> float_type, {64, 16}, {16, 1}: 0.0196837ms, 0.168715% main:@287 = load[offset=875520,end=890880](main:@1) -> float_type, {64, 60, 1}, {60, 1, 1}: 0.00063576ms, 0.00544929% I5iNBsTtoe:0 = @param:I5iNBsTtoe:0 -> int32_type, {64, 60}, {60, 1}: 0.0004996ms, 0.00428222% main:@289 = unsqueeze[axes={2},steps={}](I5iNBsTtoe:0) -> int32_type, {64, 60, 1}, {60, 1, 1}: 0.00054978ms, 0.00471233% main:@290 = gpu::code_object[code_object=13520,symbol_name=convert_kernel,global=1920,local=256,](main:@289,main:@287) -> float_type, {64, 60, 1}, {60, 1, 1}: 0.0134143ms, 0.114978% main:@291 = load[offset=890880,end=1136640](main:@1) -> float_type, {64, 60, 16}, {960, 16, 1}: 0.00063142ms, 0.00541209% main:@292 = multibroadcast[out_lens={64, 60, 16}](main:@290) -> float_type, {64, 60, 16}, {60, 1, 0}: 0.0005099ms, 0.00437051% S3KJ2tl1VG:0 = @param:S3KJ2tl1VG:0 -> float_type, {64, 60, 17}, {1020, 17, 1}: 0.0005158ms, 0.00442108% main:@294 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={64, 60, 17}](S3KJ2tl1VG:0) -> float_type, {64, 60, 16}, {1020, 17, 1}: 0.00078742ms, 0.00674922% main:@295 = gpu::code_object[code_object=13512,symbol_name=pow_kernel,global=61440,local=256,](main:@294,main:@292,main:@291) -> float_type, {64, 60, 16}, {960, 16, 1}: 0.0196161ms, 0.168136% main:@296 = load[offset=2992384,end=3256576](main:@1) -> float_type, {64, 1032}, {1032, 1}: 0.00060652ms, 0.00519867% main:@297 = reshape_lazy[dims={-1, 960}](main:@295) -> float_type, {64, 960}, {960, 1}: 0.00051372ms, 0.00440325% main:@298 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=33024,local=256,](main:@11,main:@297,main:@296) -> float_type, {64, 1032}, {1032, 1}: 0.0183996ms, 0.157709% main:@299 = hip::hip_copy_literal[id=main:@literal:151] -> float_type, {1032, 128}, {128, 1}: 0.00070516ms, 0.00604414% main:@300 = load[offset=908288,end=941056](main:@1) -> float_type, {64, 128}, {128, 1}: 0.0004778ms, 0.00409537% 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, {64, 128}, {128, 1}: 0.0263916ms, 0.22621% main:@302 = load[offset=875520,end=908288](main:@1) -> float_type, {64, 128}, {128, 1}: 0.00062728ms, 0.00537661% main:@303 = gpu::code_object[code_object=13520,symbol_name=neg_sub_kernel,global=4096,local=256,](main:@301,main:@302) -> float_type, {64, 128}, {128, 1}: 0.0133901ms, 0.114771% main:@304 = hip::hip_copy_literal[id=main:@literal:108] -> float_type, {128, 16}, {16, 1}: 0.0007478ms, 0.00640962% main:@305 = load[offset=799744,end=803840](main:@1) -> float_type, {64, 16}, {16, 1}: 0.00046478ms, 0.00398377% 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, {64, 16}, {16, 1}: 0.0197185ms, 0.169014% main:@307 = load[offset=803840,end=812032](main:@1) -> float_type, {64, 32}, {32, 1}: 0.0006161ms, 0.00528078% main:@308 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=1024,local=256,](main:@266,main:@167,main:@307) -> float_type, {64, 32}, {32, 1}: 0.0133968ms, 0.114828% main:@309 = load[offset=891904,end=912384](main:@1) -> float_type, {64, 80}, {80, 1}: 0.00061642ms, 0.00528353% main:@310 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=1024,local=256,](main:@128,main:@266,main:@167,main:@109,main:@187,main:@309) -> float_type, {64, 80}, {80, 1}: 0.0149151ms, 0.127842% main:@311 = load[offset=871424,end=891904](main:@1) -> float_type, {64, 80}, {80, 1}: 0.00061876ms, 0.00530358% main:@312 = gpu::code_object[code_object=13560,symbol_name=neg_rsqrt_tanh_log_concat_add_kernel,global=1280,local=256,](main:@128,main:@308,main:@109,main:@187,main:@310,main:@311) -> float_type, {64, 80}, {80, 1}: 0.016009ms, 0.137218% main:@313 = load[offset=803840,end=807936](main:@1) -> float_type, {64, 16}, {16, 1}: 0.00063512ms, 0.00544381% main:@314 = gpu::code_object[code_object=13520,symbol_name=rsqrt_sub_kernel,global=512,local=256,](main:@227,main:@313) -> float_type, {64, 16}, {16, 1}: 0.0129675ms, 0.111149% main:@315 = load[offset=807936,end=812032](main:@1) -> float_type, {64, 16}, {16, 1}: 0.00061656ms, 0.00528473% main:@316 = gpu::code_object[code_object=13520,symbol_name=tanh_sub_kernel,global=512,local=256,](main:@32,main:@315) -> float_type, {64, 16}, {16, 1}: 0.0135218ms, 0.115899% main:@317 = load[offset=812032,end=816128](main:@1) -> float_type, {64, 16}, {16, 1}: 0.00061344ms, 0.00525798% main:@318 = gpu::code_object[code_object=13520,symbol_name=tanh_div_kernel,global=512,local=256,](main:@147,main:@317) -> float_type, {64, 16}, {16, 1}: 0.0138309ms, 0.118549% main:@319 = load[offset=816128,end=820224](main:@1) -> float_type, {64, 16}, {16, 1}: 0.00063608ms, 0.00545204% main:@320 = gpu::code_object[code_object=13520,symbol_name=tanh_sub_kernel,global=512,local=256,](main:@51,main:@319) -> float_type, {64, 16}, {16, 1}: 0.0136472ms, 0.116975% main:@321 = load[offset=820224,end=824320](main:@1) -> float_type, {64, 16}, {16, 1}: 0.00064154ms, 0.00549884% main:@322 = gpu::code_object[code_object=13520,symbol_name=rsqrt_sub_kernel,global=512,local=256,](main:@71,main:@321) -> float_type, {64, 16}, {16, 1}: 0.0131712ms, 0.112895% main:@323 = load[offset=824320,end=828416](main:@1) -> float_type, {64, 16}, {16, 1}: 0.00062334ms, 0.00534284% main:@324 = gpu::code_object[code_object=13520,symbol_name=neg_div_kernel,global=512,local=256,](main:@90,main:@323) -> float_type, {64, 16}, {16, 1}: 0.012997ms, 0.111401% main:@325 = load[offset=828416,end=832512](main:@1) -> float_type, {64, 16}, {16, 1}: 0.00061092ms, 0.00523638% main:@326 = gpu::code_object[code_object=13520,symbol_name=tanh_add_kernel,global=512,local=256,](main:@247,main:@325) -> float_type, {64, 16}, {16, 1}: 0.0135006ms, 0.115718% main:@327 = load[offset=832512,end=836608](main:@1) -> float_type, {64, 16}, {16, 1}: 0.00061366ms, 0.00525987% main:@328 = gpu::code_object[code_object=13520,symbol_name=rsqrt_add_kernel,global=512,local=256,](main:@207,main:@327) -> float_type, {64, 16}, {16, 1}: 0.0128342ms, 0.110006% main:@329 = load[offset=836608,end=840704](main:@1) -> float_type, {64, 16}, {16, 1}: 0.00060178ms, 0.00515804% main:@330 = gpu::code_object[code_object=13520,symbol_name=rsqrt_pow_kernel,global=512,local=256,](main:@306,main:@329) -> float_type, {64, 16}, {16, 1}: 0.01606ms, 0.137655% main:@331 = load[offset=799744,end=803840](main:@1) -> float_type, {64, 16}, {16, 1}: 0.00063938ms, 0.00548032% main:@332 = gpu::code_object[code_object=13520,symbol_name=neg_div_kernel,global=512,local=256,](main:@286,main:@331) -> float_type, {64, 16}, {16, 1}: 0.012958ms, 0.111067% main:@333 = load[offset=2116864,end=2178304](main:@1) -> float_type, {64, 240}, {240, 1}: 0.00060306ms, 0.00516901% main:@334 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=1396,local=256,](main:@322,main:@332,main:@328,main:@318,main:@314,main:@330,main:@312,main:@320,main:@326,main:@324,main:@316,main:@333) -> float_type, {64, 240}, {240, 1}: 0.0193729ms, 0.166051% main:@335 = hip::hip_copy_literal[id=main:@literal:107] -> float_type, {128, 240}, {240, 1}: 0.00077494ms, 0.00664225% main:@336 = hip::hip_copy_literal[id=main:@literal:137] -> float_type, {128, 240}, {240, 1}: 0.0006279ms, 0.00538192% w27LgiJW6P:0 = @param:w27LgiJW6P:0 -> float_type, {64, 13}, {13, 1}: 0.00056512ms, 0.00484382% main:@338 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](w27LgiJW6P:0) -> float_type, {64, 4}, {13, 1}: 0.00085784ms, 0.00735281% 0cjwsL5ldh:0 = @param:0cjwsL5ldh:0 -> float_type, {64, 33}, {33, 1}: 0.00046508ms, 0.00398634% main:@340 = slice[axes={0, 1},starts={0, 1},ends={64, 17}](0cjwsL5ldh:0) -> float_type, {64, 16}, {33, 1}: 0.00064986ms, 0.00557015% 0sDBrPmfS9:0 = @param:0sDBrPmfS9:0 -> float_type, {64, 13}, {13, 1}: 0.00046796ms, 0.00401103% main:@342 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](0sDBrPmfS9:0) -> float_type, {64, 4}, {13, 1}: 0.00079278ms, 0.00679516% main:@343 = load[offset=799744,end=860160](main:@1) -> float_type, {64, 236}, {236, 1}: 0.00049164ms, 0.004214% whRnWGgf0Q:0 = @param:whRnWGgf0Q:0 -> float_type, {64, 13}, {13, 1}: 0.00051798ms, 0.00443977% main:@345 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](whRnWGgf0Q:0) -> float_type, {64, 4}, {13, 1}: 0.0008215ms, 0.00704133% 3mk8BKjDey:0 = @param:3mk8BKjDey:0 -> float_type, {64, 25}, {25, 1}: 0.0004878ms, 0.00418108% main:@347 = slice[axes={0, 1},starts={0, 1},ends={64, 17}](3mk8BKjDey:0) -> float_type, {64, 16}, {25, 1}: 0.00077546ms, 0.00664671% TxEdCcekGS:0 = @param:TxEdCcekGS:0 -> float_type, {64, 49}, {49, 1}: 0.00054526ms, 0.00467359% main:@349 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](TxEdCcekGS:0) -> float_type, {64, 8}, {49, 1}: 0.00063396ms, 0.00543387% LKAvFme9ZS:0 = @param:LKAvFme9ZS:0 -> float_type, {64, 49}, {49, 1}: 0.0004462ms, 0.00382452% main:@351 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](LKAvFme9ZS:0) -> float_type, {64, 8}, {49, 1}: 0.00077656ms, 0.00665613% KFhBukd7EV:0 = @param:KFhBukd7EV:0 -> float_type, {64, 13}, {13, 1}: 0.0005032ms, 0.00431308% main:@353 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](KFhBukd7EV:0) -> float_type, {64, 4}, {13, 1}: 0.00063252ms, 0.00542152% main:@354 = slice[axes={0, 1},starts={0, 257},ends={64, 289}](lyiDxfzZg3:0) -> float_type, {64, 32}, {701, 1}: 0.00063784ms, 0.00546712% m5MfQhR1Bp:0 = @param:m5MfQhR1Bp:0 -> float_type, {64, 13}, {13, 1}: 0.00049604ms, 0.00425171% main:@356 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](m5MfQhR1Bp:0) -> float_type, {64, 4}, {13, 1}: 0.00071856ms, 0.006159% main:@357 = slice[axes={0, 1},starts={0, 129},ends={64, 161}](nv2tzWRf1I:0) -> float_type, {64, 32}, {301, 1}: 0.00067572ms, 0.0057918% Qp3K1PHvte:0 = @param:Qp3K1PHvte:0 -> float_type, {64, 69}, {69, 1}: 0.00050332ms, 0.00431411% main:@359 = slice[axes={0, 1},starts={0, 1},ends={64, 17}](Qp3K1PHvte:0) -> float_type, {64, 16}, {69, 1}: 0.0007237ms, 0.00620305% S5AzoJmqUC:0 = @param:S5AzoJmqUC:0 -> float_type, {64, 13}, {13, 1}: 0.00046042ms, 0.0039464% main:@361 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](S5AzoJmqUC:0) -> float_type, {64, 4}, {13, 1}: 0.00075298ms, 0.00645402% main:@362 = slice[axes={0, 1},starts={0, 129},ends={64, 161}](uFSOc5Chp2:0) -> float_type, {64, 32}, {481, 1}: 0.00065208ms, 0.00558918% Vi28Dy4Ll5:0 = @param:Vi28Dy4Ll5:0 -> float_type, {64, 13}, {13, 1}: 0.00054134ms, 0.00463999% main:@364 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](Vi28Dy4Ll5:0) -> float_type, {64, 4}, {13, 1}: 0.00068876ms, 0.00590357% mIR0ToEnMH:0 = @param:mIR0ToEnMH:0 -> float_type, {64, 13}, {13, 1}: 0.00048942ms, 0.00419497% main:@366 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](mIR0ToEnMH:0) -> float_type, {64, 4}, {13, 1}: 0.0006994ms, 0.00599477% cHzIDJm6us:0 = @param:cHzIDJm6us:0 -> float_type, {64, 33}, {33, 1}: 0.00049682ms, 0.0042584% main:@368 = slice[axes={0, 1},starts={0, 1},ends={64, 17}](cHzIDJm6us:0) -> float_type, {64, 16}, {33, 1}: 0.00068914ms, 0.00590683% ifP6Ceko5T:0 = @param:ifP6Ceko5T:0 -> float_type, {64, 73}, {73, 1}: 0.00044838ms, 0.0038432% main:@370 = slice[axes={0, 1},starts={0, 1},ends={64, 17}](ifP6Ceko5T:0) -> float_type, {64, 16}, {73, 1}: 0.00069728ms, 0.0059766% gU9SoiHyRO:0 = @param:gU9SoiHyRO:0 -> float_type, {64, 13}, {13, 1}: 0.00057032ms, 0.00488839% main:@372 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](gU9SoiHyRO:0) -> float_type, {64, 4}, {13, 1}: 0.00070432ms, 0.00603694% d1ExPaGqMN:0 = @param:d1ExPaGqMN:0 -> float_type, {64, 13}, {13, 1}: 0.0005137ms, 0.00440308% main:@374 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](d1ExPaGqMN:0) -> float_type, {64, 4}, {13, 1}: 0.00071214ms, 0.00610397% qeZ76Hc03v:0 = @param:qeZ76Hc03v:0 -> float_type, {64, 13}, {13, 1}: 0.00045858ms, 0.00393063% main:@376 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](qeZ76Hc03v:0) -> float_type, {64, 4}, {13, 1}: 0.00085576ms, 0.00733498% main:@377 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=719,local=256,](main:@354,main:@357,main:@362,main:@370,main:@359,main:@351,main:@349,main:@368,main:@347,main:@340,main:@342,main:@372,main:@353,main:@376,main:@364,main:@356,main:@366,main:@374,main:@361,main:@345,main:@338,main:@343) -> float_type, {64, 236}, {236, 1}: 0.0360393ms, 0.308904% main:@378 = hip::hip_copy_literal[id=main:@literal:5] -> float_type, {236, 896}, {896, 1}: 0.0007811ms, 0.00669505% main:@379 = load[offset=2253056,end=2482432](main:@1) -> float_type, {64, 896}, {896, 1}: 0.00047108ms, 0.00403777% main:@380 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@377,main:@378,main:@379) -> float_type, {64, 896}, {896, 1}: 0.0284893ms, 0.244191% main:@381 = hip::hip_copy_literal[id=main:@literal:16] -> float_type, {256, 18, 18}, {324, 18, 1}: 0.00077508ms, 0.00664345% main:@382 = load[offset=4548864,end=4880640](main:@1) -> float_type, {256, 18, 18}, {324, 18, 1}: 0.0004943ms, 0.0042368% main:@383 = gpu::code_object[code_object=14560,symbol_name=softmax_kernel,global=294912,local=64,](main:@381,main:@382) -> float_type, {256, 18, 18}, {324, 18, 1}: 0.0295593ms, 0.253362% main:@384 = hip::hip_copy_literal[id=main:@literal:9] -> float_type, {256, 18, 64}, {1152, 64, 1}: 0.0007889ms, 0.0067619% main:@385 = load[offset=811264,end=1990912](main:@1) -> float_type, {256, 18, 64}, {1152, 64, 1}: 0.00048682ms, 0.00417268% main:@386 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@383,main:@384,main:@385) -> float_type, {256, 18, 64}, {1152, 64, 1}: 0.0269347ms, 0.230866% main:@387 = load[offset=4548864,end=4736256](main:@1) -> float_type, {64, 732}, {732, 1}: 0.00064754ms, 0.00555026% main:@388 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=23424,local=256,](main:@121,main:@334,main:@387) -> float_type, {64, 732}, {732, 1}: 0.0171144ms, 0.146693% main:@389 = hip::hip_copy_literal[id=main:@literal:155] -> float_type, {732, 128}, {128, 1}: 0.00089666ms, 0.00768555% main:@390 = load[offset=4736256,end=4769024](main:@1) -> float_type, {64, 128}, {128, 1}: 0.00048208ms, 0.00413206% main:@391 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@388,main:@389,main:@390) -> float_type, {64, 128}, {128, 1}: 0.0381981ms, 0.327407% main:@392 = load[offset=4548864,end=4581632](main:@1) -> float_type, {64, 128}, {128, 1}: 0.0006243ms, 0.00535107% main:@393 = gpu::code_object[code_object=13520,symbol_name=log_add_kernel,global=4096,local=256,](main:@391,main:@392) -> float_type, {64, 128}, {128, 1}: 0.0135114ms, 0.11581% main:@394 = load[offset=4736256,end=4797696](main:@1) -> float_type, {64, 240}, {240, 1}: 0.00062552ms, 0.00536152% main:@395 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@393,main:@336,main:@394) -> float_type, {64, 240}, {240, 1}: 0.0219ms, 0.187712% main:@396 = load[offset=4548864,end=4736256](main:@1) -> float_type, {64, 732}, {732, 1}: 0.00061622ms, 0.00528181% main:@397 = gpu::code_object[code_object=13560,symbol_name=noop_rsqrt_add_concat_noop_kernel,global=23424,local=256,](main:@121,main:@395,main:@334,main:@396) -> float_type, {64, 732}, {732, 1}: 0.0168568ms, 0.144485% main:@398 = hip::hip_copy_literal[id=main:@literal:118] -> float_type, {732, 256}, {256, 1}: 0.00084306ms, 0.00722613% main:@399 = load[offset=1990912,end=2056448](main:@1) -> float_type, {64, 256}, {256, 1}: 0.0004736ms, 0.00405937% main:@400 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@397,main:@398,main:@399) -> float_type, {64, 256}, {256, 1}: 0.0398564ms, 0.341622% main:@401 = load[offset=4548864,end=4833536](main:@1) -> float_type, {64, 1112}, {1112, 1}: 0.0006323ms, 0.00541964% main:@402 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=35584,local=256,](main:@63,main:@334,main:@401) -> float_type, {64, 1112}, {1112, 1}: 0.0181264ms, 0.155367% main:@403 = hip::hip_copy_literal[id=main:@literal:128] -> float_type, {1112, 128}, {128, 1}: 0.00079728ms, 0.00683373% main:@404 = load[offset=4833536,end=4866304](main:@1) -> float_type, {64, 128}, {128, 1}: 0.00047486ms, 0.00407017% main:@405 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@402,main:@403,main:@404) -> float_type, {64, 128}, {128, 1}: 0.0261313ms, 0.223979% main:@406 = load[offset=2056448,end=2089216](main:@1) -> float_type, {64, 128}, {128, 1}: 0.00064136ms, 0.00549729% main:@407 = gpu::code_object[code_object=13520,symbol_name=neg_add_kernel,global=4096,local=256,](main:@405,main:@406) -> float_type, {64, 128}, {128, 1}: 0.0134536ms, 0.115315% main:@408 = hip::hip_copy_literal[id=main:@literal:67] -> float_type, {128, 240}, {240, 1}: 0.00074662ms, 0.00639951% main:@409 = load[offset=4833536,end=4894976](main:@1) -> float_type, {64, 240}, {240, 1}: 0.0004935ms, 0.00422994% main:@410 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@407,main:@408,main:@409) -> float_type, {64, 240}, {240, 1}: 0.0199822ms, 0.171274% main:@411 = load[offset=4548864,end=4833536](main:@1) -> float_type, {64, 1112}, {1112, 1}: 0.00060716ms, 0.00520415% main:@412 = gpu::code_object[code_object=13552,symbol_name=noop_log_add_concat_noop_kernel,global=35584,local=256,](main:@63,main:@410,main:@334,main:@411) -> float_type, {64, 1112}, {1112, 1}: 0.0184521ms, 0.158158% main:@413 = hip::hip_copy_literal[id=main:@literal:87] -> float_type, {1112, 256}, {256, 1}: 0.00075512ms, 0.00647237% main:@414 = load[offset=353280,end=418816](main:@1) -> float_type, {64, 256}, {256, 1}: 0.00049238ms, 0.00422034% main:@415 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@412,main:@413,main:@414) -> float_type, {64, 256}, {256, 1}: 0.0308686ms, 0.264584% main:@416 = load[offset=4548864,end=5140736](main:@1) -> float_type, {64, 2312}, {2312, 1}: 0.00064954ms, 0.00556741% main:@417 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=73984,local=256,](main:@199,main:@334,main:@416) -> float_type, {64, 2312}, {2312, 1}: 0.0187234ms, 0.160484% main:@418 = hip::hip_copy_literal[id=main:@literal:53] -> float_type, {2312, 128}, {128, 1}: 0.00076644ms, 0.00656939% main:@419 = load[offset=418816,end=451584](main:@1) -> float_type, {64, 128}, {128, 1}: 0.00048414ms, 0.00414971% main:@420 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@417,main:@418,main:@419) -> float_type, {64, 128}, {128, 1}: 0.0309769ms, 0.265513% main:@421 = load[offset=480256,end=513024](main:@1) -> float_type, {64, 128}, {128, 1}: 0.00062784ms, 0.00538141% main:@422 = gpu::code_object[code_object=13520,symbol_name=rsqrt_add_kernel,global=4096,local=256,](main:@420,main:@421) -> float_type, {64, 128}, {128, 1}: 0.0134707ms, 0.115461% main:@423 = load[offset=418816,end=480256](main:@1) -> float_type, {64, 240}, {240, 1}: 0.000593ms, 0.00508279% main:@424 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@422,main:@335,main:@423) -> float_type, {64, 240}, {240, 1}: 0.0199525ms, 0.171019% main:@425 = load[offset=4548864,end=5140736](main:@1) -> float_type, {64, 2312}, {2312, 1}: 0.0006154ms, 0.00527478% main:@426 = gpu::code_object[code_object=13552,symbol_name=noop_tanh_pow_concat_noop_kernel,global=73984,local=256,](main:@199,main:@424,main:@334,main:@425) -> float_type, {64, 2312}, {2312, 1}: 0.0190054ms, 0.162901% main:@427 = hip::hip_copy_literal[id=main:@literal:144] -> float_type, {2312, 256}, {256, 1}: 0.00079304ms, 0.00679739% main:@428 = load[offset=418816,end=484352](main:@1) -> float_type, {64, 256}, {256, 1}: 0.00049032ms, 0.00420268% main:@429 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@426,main:@427,main:@428) -> float_type, {64, 256}, {256, 1}: 0.0349523ms, 0.299587% main:@430 = load[offset=4018432,end=4589824](main:@1) -> float_type, {64, 2232}, {2232, 1}: 0.00062336ms, 0.00534301% main:@431 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=71424,local=256,](main:@140,main:@334,main:@430) -> float_type, {64, 2232}, {2232, 1}: 0.0192038ms, 0.164602% main:@432 = hip::hip_copy_literal[id=main:@literal:66] -> float_type, {2232, 128}, {128, 1}: 0.00076038ms, 0.00651745% main:@433 = load[offset=484352,end=517120](main:@1) -> float_type, {64, 128}, {128, 1}: 0.0004934ms, 0.00422908% main:@434 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@431,main:@432,main:@433) -> float_type, {64, 128}, {128, 1}: 0.0302694ms, 0.259448% main:@435 = load[offset=2056448,end=2089216](main:@1) -> float_type, {64, 128}, {128, 1}: 0.00061444ms, 0.00526655% main:@436 = gpu::code_object[code_object=13520,symbol_name=neg_div_kernel,global=4096,local=256,](main:@434,main:@435) -> float_type, {64, 128}, {128, 1}: 0.0134955ms, 0.115674% main:@437 = load[offset=484352,end=545792](main:@1) -> float_type, {64, 240}, {240, 1}: 0.00061664ms, 0.00528541% main:@438 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@436,main:@335,main:@437) -> float_type, {64, 240}, {240, 1}: 0.0198707ms, 0.170318% main:@439 = load[offset=4018432,end=4589824](main:@1) -> float_type, {64, 2232}, {2232, 1}: 0.00061542ms, 0.00527495% main:@440 = gpu::code_object[code_object=13560,symbol_name=noop_rsqrt_add_concat_noop_kernel,global=71424,local=256,](main:@140,main:@438,main:@334,main:@439) -> float_type, {64, 2232}, {2232, 1}: 0.0189544ms, 0.162464% main:@441 = hip::hip_copy_literal[id=main:@literal:69] -> float_type, {2232, 256}, {256, 1}: 0.00071796ms, 0.00615386% main:@442 = load[offset=484352,end=549888](main:@1) -> float_type, {64, 256}, {256, 1}: 0.00048582ms, 0.00416411% main:@443 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@440,main:@441,main:@442) -> float_type, {64, 256}, {256, 1}: 0.0330049ms, 0.282895% main:@444 = load[offset=2482432,end=2603264](main:@1) -> float_type, {64, 472}, {472, 1}: 0.00061712ms, 0.00528952% main:@445 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=15104,local=256,](main:@83,main:@334,main:@444) -> float_type, {64, 472}, {472, 1}: 0.0158911ms, 0.136207% main:@446 = hip::hip_copy_literal[id=main:@literal:119] -> float_type, {472, 128}, {128, 1}: 0.0007923ms, 0.00679105% main:@447 = load[offset=2603264,end=2636032](main:@1) -> float_type, {64, 128}, {128, 1}: 0.00046586ms, 0.00399303% main:@448 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@445,main:@446,main:@447) -> float_type, {64, 128}, {128, 1}: 0.0315769ms, 0.270656% main:@449 = load[offset=2056448,end=2089216](main:@1) -> float_type, {64, 128}, {128, 1}: 0.0006582ms, 0.00564163% main:@450 = gpu::code_object[code_object=13520,symbol_name=log_add_kernel,global=4096,local=256,](main:@448,main:@449) -> float_type, {64, 128}, {128, 1}: 0.0136238ms, 0.116773% main:@451 = hip::hip_copy_literal[id=main:@literal:102] -> float_type, {128, 240}, {240, 1}: 0.00072124ms, 0.00618197% main:@452 = load[offset=2482432,end=2543872](main:@1) -> float_type, {64, 240}, {240, 1}: 0.00046404ms, 0.00397743% main:@453 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@450,main:@451,main:@452) -> float_type, {64, 240}, {240, 1}: 0.019598ms, 0.16798% main:@454 = load[offset=2547968,end=2668800](main:@1) -> float_type, {64, 472}, {472, 1}: 0.0006228ms, 0.00533821% main:@455 = gpu::code_object[code_object=13552,symbol_name=noop_tanh_div_concat_noop_kernel,global=15104,local=256,](main:@83,main:@453,main:@334,main:@454) -> float_type, {64, 472}, {472, 1}: 0.016508ms, 0.141495% main:@456 = hip::hip_copy_literal[id=main:@literal:80] -> float_type, {472, 256}, {256, 1}: 0.00077524ms, 0.00664482% main:@457 = load[offset=2482432,end=2547968](main:@1) -> float_type, {64, 256}, {256, 1}: 0.00048498ms, 0.00415691% main:@458 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@455,main:@456,main:@457) -> float_type, {64, 256}, {256, 1}: 0.0311131ms, 0.26668% main:@459 = load[offset=2547968,end=2873600](main:@1) -> float_type, {64, 1272}, {1272, 1}: 0.00063884ms, 0.00547569% main:@460 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=40704,local=256,](main:@219,main:@334,main:@459) -> float_type, {64, 1272}, {1272, 1}: 0.0184152ms, 0.157842% main:@461 = hip::hip_copy_literal[id=main:@literal:59] -> float_type, {1272, 128}, {128, 1}: 0.00073122ms, 0.00626751% main:@462 = load[offset=2056448,end=2089216](main:@1) -> float_type, {64, 128}, {128, 1}: 0.00051678ms, 0.00442948% main:@463 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@460,main:@461,main:@462) -> float_type, {64, 128}, {128, 1}: 0.0263258ms, 0.225646% main:@464 = load[offset=293888,end=326656](main:@1) -> float_type, {64, 128}, {128, 1}: 0.0006153ms, 0.00527393% main:@465 = gpu::code_object[code_object=13520,symbol_name=rsqrt_add_kernel,global=4096,local=256,](main:@463,main:@464) -> float_type, {64, 128}, {128, 1}: 0.0135714ms, 0.116325% main:@466 = hip::hip_copy_literal[id=main:@literal:39] -> float_type, {128, 240}, {240, 1}: 0.0007304ms, 0.00626048% main:@467 = load[offset=2547968,end=2609408](main:@1) -> float_type, {64, 240}, {240, 1}: 0.0004613ms, 0.00395394% main:@468 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@465,main:@466,main:@467) -> float_type, {64, 240}, {240, 1}: 0.0194991ms, 0.167133% main:@469 = load[offset=2613504,end=2939136](main:@1) -> float_type, {64, 1272}, {1272, 1}: 0.00062264ms, 0.00533684% main:@470 = gpu::code_object[code_object=13552,symbol_name=noop_log_sub_concat_noop_kernel,global=40704,local=256,](main:@219,main:@468,main:@334,main:@469) -> float_type, {64, 1272}, {1272, 1}: 0.0183162ms, 0.156994% main:@471 = hip::hip_copy_literal[id=main:@literal:40] -> float_type, {1272, 256}, {256, 1}: 0.00071898ms, 0.0061626% main:@472 = load[offset=2547968,end=2613504](main:@1) -> float_type, {64, 256}, {256, 1}: 0.00047282ms, 0.00405269% main:@473 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@470,main:@471,main:@472) -> float_type, {64, 256}, {256, 1}: 0.0291719ms, 0.250041% main:@474 = load[offset=2613504,end=2898176](main:@1) -> float_type, {64, 1112}, {1112, 1}: 0.00066042ms, 0.00566066% main:@475 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=35584,local=256,](main:@159,main:@334,main:@474) -> float_type, {64, 1112}, {1112, 1}: 0.0182402ms, 0.156342% main:@476 = hip::hip_copy_literal[id=main:@literal:54] -> float_type, {1112, 128}, {128, 1}: 0.00075464ms, 0.00646825% main:@477 = load[offset=2056448,end=2089216](main:@1) -> float_type, {64, 128}, {128, 1}: 0.00046158ms, 0.00395634% main:@478 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@475,main:@476,main:@477) -> float_type, {64, 128}, {128, 1}: 0.0265994ms, 0.227992% main:@479 = load[offset=293888,end=326656](main:@1) -> float_type, {64, 128}, {128, 1}: 0.00060624ms, 0.00519627% main:@480 = gpu::code_object[code_object=13520,symbol_name=log_pow_kernel,global=4096,local=256,](main:@478,main:@479) -> float_type, {64, 128}, {128, 1}: 0.0166373ms, 0.142604% main:@481 = load[offset=2898176,end=2959616](main:@1) -> float_type, {64, 240}, {240, 1}: 0.00061294ms, 0.0052537% main:@482 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@480,main:@336,main:@481) -> float_type, {64, 240}, {240, 1}: 0.0197631ms, 0.169396% main:@483 = load[offset=2613504,end=2898176](main:@1) -> float_type, {64, 1112}, {1112, 1}: 0.00062026ms, 0.00531644% main:@484 = gpu::code_object[code_object=13560,symbol_name=noop_rsqrt_add_concat_noop_kernel,global=35584,local=256,](main:@159,main:@482,main:@334,main:@483) -> float_type, {64, 1112}, {1112, 1}: 0.018422ms, 0.157901% main:@485 = hip::hip_copy_literal[id=main:@literal:123] -> float_type, {1112, 256}, {256, 1}: 0.00067566ms, 0.00579129% main:@486 = load[offset=549888,end=615424](main:@1) -> float_type, {64, 256}, {256, 1}: 0.00047562ms, 0.00407669% main:@487 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@484,main:@485,main:@486) -> float_type, {64, 256}, {256, 1}: 0.0286506ms, 0.245573% main:@488 = load[offset=615424,end=736256](main:@1) -> float_type, {64, 472}, {472, 1}: 0.0006744ms, 0.00578049% main:@489 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=15104,local=256,](main:@239,main:@334,main:@488) -> float_type, {64, 472}, {472, 1}: 0.0158476ms, 0.135835% main:@490 = hip::hip_copy_literal[id=main:@literal:99] -> float_type, {472, 128}, {128, 1}: 0.00072816ms, 0.00624128% main:@491 = load[offset=736256,end=769024](main:@1) -> float_type, {64, 128}, {128, 1}: 0.00048616ms, 0.00416703% main:@492 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@489,main:@490,main:@491) -> float_type, {64, 128}, {128, 1}: 0.0311413ms, 0.266921% main:@493 = load[offset=293888,end=326656](main:@1) -> float_type, {64, 128}, {128, 1}: 0.00062514ms, 0.00535827% main:@494 = gpu::code_object[code_object=13520,symbol_name=log_div_kernel,global=4096,local=256,](main:@492,main:@493) -> float_type, {64, 128}, {128, 1}: 0.0137658ms, 0.117991% main:@495 = load[offset=615424,end=676864](main:@1) -> float_type, {64, 240}, {240, 1}: 0.00058936ms, 0.00505159% main:@496 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@494,main:@335,main:@495) -> float_type, {64, 240}, {240, 1}: 0.0194345ms, 0.166579% main:@497 = load[offset=680960,end=801792](main:@1) -> float_type, {64, 472}, {472, 1}: 0.00063814ms, 0.00546969% main:@498 = gpu::code_object[code_object=13560,symbol_name=noop_rsqrt_div_concat_noop_kernel,global=15104,local=256,](main:@239,main:@496,main:@334,main:@497) -> float_type, {64, 472}, {472, 1}: 0.0162431ms, 0.139224% main:@499 = hip::hip_copy_literal[id=main:@literal:63] -> float_type, {472, 256}, {256, 1}: 0.00076652ms, 0.00657008% main:@500 = load[offset=615424,end=680960](main:@1) -> float_type, {64, 256}, {256, 1}: 0.00048876ms, 0.00418931% 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, {64, 256}, {256, 1}: 0.0313256ms, 0.268501% main:@502 = load[offset=2613504,end=2898176](main:@1) -> float_type, {64, 1112}, {1112, 1}: 0.00062918ms, 0.00539289% main:@503 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=35584,local=256,](main:@278,main:@334,main:@502) -> float_type, {64, 1112}, {1112, 1}: 0.0182033ms, 0.156026% main:@504 = hip::hip_copy_literal[id=main:@literal:135] -> float_type, {1112, 128}, {128, 1}: 0.00074182ms, 0.00635837% main:@505 = load[offset=680960,end=713728](main:@1) -> float_type, {64, 128}, {128, 1}: 0.00045604ms, 0.00390886% main:@506 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@503,main:@504,main:@505) -> float_type, {64, 128}, {128, 1}: 0.0259872ms, 0.222744% main:@507 = load[offset=293888,end=326656](main:@1) -> float_type, {64, 128}, {128, 1}: 0.00061622ms, 0.00528181% main:@508 = gpu::code_object[code_object=13520,symbol_name=neg_pow_kernel,global=4096,local=256,](main:@506,main:@507) -> float_type, {64, 128}, {128, 1}: 0.0164105ms, 0.140659% main:@509 = hip::hip_copy_literal[id=main:@literal:74] -> float_type, {128, 240}, {240, 1}: 0.000731ms, 0.00626563% main:@510 = load[offset=680960,end=742400](main:@1) -> float_type, {64, 240}, {240, 1}: 0.00047064ms, 0.004034% main:@511 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@508,main:@509,main:@510) -> float_type, {64, 240}, {240, 1}: 0.0198608ms, 0.170233% main:@512 = load[offset=2613504,end=2898176](main:@1) -> float_type, {64, 1112}, {1112, 1}: 0.00062114ms, 0.00532398% main:@513 = gpu::code_object[code_object=13552,symbol_name=noop_log_add_concat_noop_kernel,global=35584,local=256,](main:@278,main:@511,main:@334,main:@512) -> float_type, {64, 1112}, {1112, 1}: 0.0183191ms, 0.157019% main:@514 = hip::hip_copy_literal[id=main:@literal:55] -> float_type, {1112, 256}, {256, 1}: 0.00075882ms, 0.00650408% main:@515 = load[offset=680960,end=746496](main:@1) -> float_type, {64, 256}, {256, 1}: 0.0005039ms, 0.00431908% main:@516 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@513,main:@514,main:@515) -> float_type, {64, 256}, {256, 1}: 0.0285728ms, 0.244906% main:@517 = load[offset=2613504,end=2949376](main:@1) -> float_type, {64, 1312}, {1312, 1}: 0.00062236ms, 0.00533444% main:@518 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=41984,local=256,](main:@259,main:@334,main:@517) -> float_type, {64, 1312}, {1312, 1}: 0.01839ms, 0.157626% main:@519 = hip::hip_copy_literal[id=main:@literal:46] -> float_type, {1312, 128}, {128, 1}: 0.000751ms, 0.00643705% main:@520 = load[offset=746496,end=779264](main:@1) -> float_type, {64, 128}, {128, 1}: 0.00048352ms, 0.0041444% main:@521 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@518,main:@519,main:@520) -> float_type, {64, 128}, {128, 1}: 0.0263967ms, 0.226254% main:@522 = load[offset=293888,end=326656](main:@1) -> float_type, {64, 128}, {128, 1}: 0.0006445ms, 0.00552421% main:@523 = gpu::code_object[code_object=13520,symbol_name=tanh_add_kernel,global=4096,local=256,](main:@521,main:@522) -> float_type, {64, 128}, {128, 1}: 0.0140921ms, 0.120788% main:@524 = load[offset=746496,end=807936](main:@1) -> float_type, {64, 240}, {240, 1}: 0.00060922ms, 0.00522181% main:@525 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@523,main:@335,main:@524) -> float_type, {64, 240}, {240, 1}: 0.0197591ms, 0.169362% main:@526 = load[offset=2613504,end=2949376](main:@1) -> float_type, {64, 1312}, {1312, 1}: 0.0006175ms, 0.00529278% main:@527 = gpu::code_object[code_object=13552,symbol_name=noop_neg_div_concat_noop_kernel,global=41984,local=256,](main:@259,main:@525,main:@334,main:@526) -> float_type, {64, 1312}, {1312, 1}: 0.0185468ms, 0.15897% main:@528 = hip::hip_copy_literal[id=main:@literal:60] -> float_type, {1312, 256}, {256, 1}: 0.00072458ms, 0.0062106% main:@529 = load[offset=2178304,end=2243840](main:@1) -> float_type, {64, 256}, {256, 1}: 0.00049764ms, 0.00426543% main:@530 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@527,main:@528,main:@529) -> float_type, {64, 256}, {256, 1}: 0.0283272ms, 0.242801% main:@531 = load[offset=2613504,end=2734336](main:@1) -> float_type, {64, 472}, {472, 1}: 0.00062432ms, 0.00535124% main:@532 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=15104,local=256,](main:@44,main:@334,main:@531) -> float_type, {64, 472}, {472, 1}: 0.0158328ms, 0.135708% main:@533 = hip::hip_copy_literal[id=main:@literal:145] -> float_type, {472, 128}, {128, 1}: 0.0007667ms, 0.00657162% main:@534 = load[offset=746496,end=779264](main:@1) -> float_type, {64, 128}, {128, 1}: 0.00047728ms, 0.00409091% main:@535 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@532,main:@533,main:@534) -> float_type, {64, 128}, {128, 1}: 0.0309748ms, 0.265494% main:@536 = load[offset=293888,end=326656](main:@1) -> float_type, {64, 128}, {128, 1}: 0.00062606ms, 0.00536615% main:@537 = gpu::code_object[code_object=13520,symbol_name=log_pow_kernel,global=4096,local=256,](main:@535,main:@536) -> float_type, {64, 128}, {128, 1}: 0.016625ms, 0.142498% main:@538 = hip::hip_copy_literal[id=main:@literal:71] -> float_type, {128, 240}, {240, 1}: 0.00077944ms, 0.00668082% main:@539 = load[offset=746496,end=807936](main:@1) -> float_type, {64, 240}, {240, 1}: 0.00046828ms, 0.00401377% main:@540 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@537,main:@538,main:@539) -> float_type, {64, 240}, {240, 1}: 0.0199014ms, 0.170581% main:@541 = load[offset=2679040,end=2799872](main:@1) -> float_type, {64, 472}, {472, 1}: 0.00061662ms, 0.00528524% main:@542 = gpu::code_object[code_object=13552,symbol_name=noop_log_pow_concat_noop_kernel,global=15104,local=256,](main:@44,main:@540,main:@334,main:@541) -> float_type, {64, 472}, {472, 1}: 0.0175037ms, 0.15003% main:@543 = hip::hip_copy_literal[id=main:@literal:37] -> float_type, {472, 256}, {256, 1}: 0.00071736ms, 0.00614871% main:@544 = load[offset=2613504,end=2679040](main:@1) -> float_type, {64, 256}, {256, 1}: 0.00048464ms, 0.004154% main:@545 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@542,main:@543,main:@544) -> float_type, {64, 256}, {256, 1}: 0.0309618ms, 0.265383% main:@546 = load[offset=2679040,end=2799872](main:@1) -> float_type, {64, 472}, {472, 1}: 0.0006134ms, 0.00525764% main:@547 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=15104,local=256,](main:@24,main:@334,main:@546) -> float_type, {64, 472}, {472, 1}: 0.016016ms, 0.137278% main:@548 = hip::hip_copy_literal[id=main:@literal:47] -> float_type, {472, 128}, {128, 1}: 0.00077568ms, 0.00664859% main:@549 = load[offset=293888,end=326656](main:@1) -> float_type, {64, 128}, {128, 1}: 0.00046066ms, 0.00394846% main:@550 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@547,main:@548,main:@549) -> float_type, {64, 128}, {128, 1}: 0.0306312ms, 0.262549% main:@551 = load[offset=59392,end=92160](main:@1) -> float_type, {64, 128}, {128, 1}: 0.00060802ms, 0.00521153% main:@552 = gpu::code_object[code_object=13520,symbol_name=tanh_sub_kernel,global=4096,local=256,](main:@550,main:@551) -> float_type, {64, 128}, {128, 1}: 0.0138669ms, 0.118857% main:@553 = load[offset=746496,end=807936](main:@1) -> float_type, {64, 240}, {240, 1}: 0.00061906ms, 0.00530615% main:@554 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@552,main:@336,main:@553) -> float_type, {64, 240}, {240, 1}: 0.0197451ms, 0.169241% main:@555 = load[offset=2679040,end=2799872](main:@1) -> float_type, {64, 472}, {472, 1}: 0.00061324ms, 0.00525627% main:@556 = gpu::code_object[code_object=13552,symbol_name=noop_tanh_div_concat_noop_kernel,global=15104,local=256,](main:@24,main:@554,main:@334,main:@555) -> float_type, {64, 472}, {472, 1}: 0.0164015ms, 0.140582% main:@557 = hip::hip_copy_literal[id=main:@literal:156] -> float_type, {472, 256}, {256, 1}: 0.0007695ms, 0.00659562% main:@558 = load[offset=0,end=65536](main:@1) -> float_type, {64, 256}, {256, 1}: 0.00049612ms, 0.0042524% main:@559 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@556,main:@557,main:@558) -> float_type, {64, 256}, {256, 1}: 0.0307596ms, 0.26365% main:@560 = load[offset=3256576,end=3582208](main:@1) -> float_type, {64, 1272}, {1272, 1}: 0.00062094ms, 0.00532227% main:@561 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=40704,local=256,](main:@298,main:@334,main:@560) -> float_type, {64, 1272}, {1272, 1}: 0.0181579ms, 0.155637% main:@562 = hip::hip_copy_literal[id=main:@literal:136] -> float_type, {1272, 128}, {128, 1}: 0.00072196ms, 0.00618814% main:@563 = load[offset=293888,end=326656](main:@1) -> float_type, {64, 128}, {128, 1}: 0.00047426ms, 0.00406503% main:@564 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@561,main:@562,main:@563) -> float_type, {64, 128}, {128, 1}: 0.026228ms, 0.224808% main:@565 = load[offset=65536,end=98304](main:@1) -> float_type, {64, 128}, {128, 1}: 0.00061692ms, 0.00528781% main:@566 = gpu::code_object[code_object=13520,symbol_name=neg_pow_kernel,global=4096,local=256,](main:@564,main:@565) -> float_type, {64, 128}, {128, 1}: 0.0165013ms, 0.141438% main:@567 = hip::hip_copy_literal[id=main:@literal:78] -> float_type, {128, 240}, {240, 1}: 0.00075658ms, 0.00648488% main:@568 = load[offset=746496,end=807936](main:@1) -> float_type, {64, 240}, {240, 1}: 0.00047946ms, 0.0041096% main:@569 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@566,main:@567,main:@568) -> float_type, {64, 240}, {240, 1}: 0.0198915ms, 0.170496% main:@570 = load[offset=3256576,end=3582208](main:@1) -> float_type, {64, 1272}, {1272, 1}: 0.00061494ms, 0.00527084% main:@571 = gpu::code_object[code_object=13560,symbol_name=noop_rsqrt_div_concat_noop_kernel,global=40704,local=256,](main:@298,main:@569,main:@334,main:@570) -> float_type, {64, 1272}, {1272, 1}: 0.0184288ms, 0.157959% main:@572 = hip::hip_copy_literal[id=main:@literal:131] -> float_type, {1272, 256}, {256, 1}: 0.00075352ms, 0.00645865% main:@573 = load[offset=2679040,end=2744576](main:@1) -> float_type, {64, 256}, {256, 1}: 0.000464ms, 0.00397709% main:@574 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@571,main:@572,main:@573) -> float_type, {64, 256}, {256, 1}: 0.0283416ms, 0.242924% main:@575 = load[offset=2744576,end=2901248](main:@1) -> float_type, {64, 612}, {612, 1}: 0.00063088ms, 0.00540747% main:@576 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=19584,local=256,](main:@102,main:@334,main:@575) -> float_type, {64, 612}, {612, 1}: 0.0165253ms, 0.141643% main:@577 = hip::hip_copy_literal[id=main:@literal:125] -> float_type, {612, 128}, {128, 1}: 0.00071804ms, 0.00615454% main:@578 = load[offset=293888,end=326656](main:@1) -> float_type, {64, 128}, {128, 1}: 0.000507ms, 0.00434565% main:@579 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@576,main:@577,main:@578) -> float_type, {64, 128}, {128, 1}: 0.0339851ms, 0.291297% main:@580 = load[offset=65536,end=98304](main:@1) -> float_type, {64, 128}, {128, 1}: 0.00061222ms, 0.00524753% main:@581 = gpu::code_object[code_object=13520,symbol_name=neg_pow_kernel,global=4096,local=256,](main:@579,main:@580) -> float_type, {64, 128}, {128, 1}: 0.0164794ms, 0.14125% main:@582 = hip::hip_copy_literal[id=main:@literal:124] -> float_type, {128, 240}, {240, 1}: 0.00070198ms, 0.00601689% main:@583 = load[offset=746496,end=807936](main:@1) -> float_type, {64, 240}, {240, 1}: 0.00046356ms, 0.00397332% main:@584 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@581,main:@582,main:@583) -> float_type, {64, 240}, {240, 1}: 0.0198276ms, 0.169948% main:@585 = load[offset=2744576,end=2901248](main:@1) -> float_type, {64, 612}, {612, 1}: 0.00062014ms, 0.00531541% main:@586 = gpu::code_object[code_object=13560,symbol_name=noop_rsqrt_add_concat_noop_kernel,global=19584,local=256,](main:@102,main:@584,main:@334,main:@585) -> float_type, {64, 612}, {612, 1}: 0.0167227ms, 0.143335% main:@587 = hip::hip_copy_literal[id=main:@literal:103] -> float_type, {612, 256}, {256, 1}: 0.00069514ms, 0.00595826% main:@588 = load[offset=65536,end=131072](main:@1) -> float_type, {64, 256}, {256, 1}: 0.00045888ms, 0.0039332% main:@589 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@586,main:@587,main:@588) -> float_type, {64, 256}, {256, 1}: 0.034665ms, 0.297124% main:@590 = load[offset=2744576,end=2885888](main:@1) -> float_type, {64, 552}, {552, 1}: 0.0006418ms, 0.00550106% main:@591 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=17664,local=256,](main:@179,main:@334,main:@590) -> float_type, {64, 552}, {552, 1}: 0.0162687ms, 0.139444% main:@592 = hip::hip_copy_literal[id=main:@literal:122] -> float_type, {552, 128}, {128, 1}: 0.00078878ms, 0.00676088% main:@593 = load[offset=131072,end=163840](main:@1) -> float_type, {64, 128}, {128, 1}: 0.00045834ms, 0.00392857% main:@594 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@591,main:@592,main:@593) -> float_type, {64, 128}, {128, 1}: 0.0323611ms, 0.277377% main:@595 = load[offset=293888,end=326656](main:@1) -> float_type, {64, 128}, {128, 1}: 0.0006215ms, 0.00532707% main:@596 = gpu::code_object[code_object=13520,symbol_name=neg_div_kernel,global=4096,local=256,](main:@594,main:@595) -> float_type, {64, 128}, {128, 1}: 0.013629ms, 0.116819% main:@597 = load[offset=131072,end=192512](main:@1) -> float_type, {64, 240}, {240, 1}: 0.0006001ms, 0.00514364% main:@598 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@596,main:@336,main:@597) -> float_type, {64, 240}, {240, 1}: 0.0197047ms, 0.168895% main:@599 = load[offset=2744576,end=2885888](main:@1) -> float_type, {64, 552}, {552, 1}: 0.00064182ms, 0.00550124% main:@600 = gpu::code_object[code_object=13560,symbol_name=noop_rsqrt_sub_concat_noop_kernel,global=17664,local=256,](main:@179,main:@598,main:@334,main:@599) -> float_type, {64, 552}, {552, 1}: 0.0164511ms, 0.141007% main:@601 = hip::hip_copy_literal[id=main:@literal:98] -> float_type, {552, 256}, {256, 1}: 0.00075642ms, 0.00648351% main:@602 = load[offset=131072,end=196608](main:@1) -> float_type, {64, 256}, {256, 1}: 0.00046546ms, 0.0039896% main:@603 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@600,main:@601,main:@602) -> float_type, {64, 256}, {256, 1}: 0.0329406ms, 0.282344% main:@604 = load[offset=2744576,end=3727616](main:@1) -> float_type, {64, 15, 256}, {3840, 256, 1}: 0.00066148ms, 0.00566975% main:@605 = unsqueeze[axes={1},steps={}](main:@400) -> float_type, {64, 1, 256}, {256, 256, 1}: 0.00066112ms, 0.00566666% main:@606 = unsqueeze[axes={1},steps={}](main:@415) -> float_type, {64, 1, 256}, {256, 256, 1}: 0.0005393ms, 0.00462251% main:@607 = unsqueeze[axes={1},steps={}](main:@429) -> float_type, {64, 1, 256}, {256, 256, 1}: 0.00048368ms, 0.00414577% main:@608 = unsqueeze[axes={1},steps={}](main:@443) -> float_type, {64, 1, 256}, {256, 256, 1}: 0.00048072ms, 0.0041204% main:@609 = unsqueeze[axes={1},steps={}](main:@458) -> float_type, {64, 1, 256}, {256, 256, 1}: 0.00050668ms, 0.00434291% main:@610 = unsqueeze[axes={1},steps={}](main:@473) -> float_type, {64, 1, 256}, {256, 256, 1}: 0.00048106ms, 0.00412331% main:@611 = unsqueeze[axes={1},steps={}](main:@487) -> float_type, {64, 1, 256}, {256, 256, 1}: 0.0004865ms, 0.00416994% main:@612 = unsqueeze[axes={1},steps={}](main:@501) -> float_type, {64, 1, 256}, {256, 256, 1}: 0.00048096ms, 0.00412246% main:@613 = unsqueeze[axes={1},steps={}](main:@516) -> float_type, {64, 1, 256}, {256, 256, 1}: 0.00047416ms, 0.00406417% main:@614 = unsqueeze[axes={1},steps={}](main:@530) -> float_type, {64, 1, 256}, {256, 256, 1}: 0.00048344ms, 0.00414371% main:@615 = unsqueeze[axes={1},steps={}](main:@545) -> float_type, {64, 1, 256}, {256, 256, 1}: 0.00047556ms, 0.00407617% main:@616 = unsqueeze[axes={1},steps={}](main:@559) -> float_type, {64, 1, 256}, {256, 256, 1}: 0.00047144ms, 0.00404086% main:@617 = unsqueeze[axes={1},steps={}](main:@574) -> float_type, {64, 1, 256}, {256, 256, 1}: 0.00046906ms, 0.00402046% main:@618 = unsqueeze[axes={1},steps={}](main:@589) -> float_type, {64, 1, 256}, {256, 256, 1}: 0.00049046ms, 0.00420388% main:@619 = unsqueeze[axes={1},steps={}](main:@603) -> float_type, {64, 1, 256}, {256, 256, 1}: 0.00048722ms, 0.00417611% main:@620 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=8192,local=256,](main:@606,main:@613,main:@607,main:@608,main:@610,main:@617,main:@605,main:@614,main:@611,main:@618,main:@619,main:@615,main:@612,main:@609,main:@616,main:@604) -> float_type, {64, 15, 256}, {3840, 256, 1}: 0.0258202ms, 0.221313% main:@621 = hip::hip_copy_literal[id=main:@literal:35] -> float_type, {256, 768}, {768, 1}: 0.00074452ms, 0.00638151% main:@622 = load[offset=9298432,end=12247552](main:@1) -> float_type, {960, 768}, {768, 1}: 0.0005189ms, 0.00444765% main:@623 = reshape_lazy[dims={960, 256}](main:@620) -> float_type, {960, 256}, {256, 1}: 0.0005854ms, 0.00501764% main:@624 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@623,main:@621,main:@622) -> float_type, {960, 768}, {768, 1}: 0.0390279ms, 0.33452% main:@625 = hip::hip_copy_literal[id=main:@literal:25] -> float_type, {130, 128}, {128, 1}: 0.00075276ms, 0.00645214% 2CnSyUBzm0:0 = @param:2CnSyUBzm0:0 -> float_type, {64, 65}, {65, 1}: 0.00054054ms, 0.00463313% main:@627 = slice[axes={0, 1},starts={0, 53},ends={64, 57}](2CnSyUBzm0:0) -> float_type, {64, 4}, {65, 1}: 0.0007347ms, 0.00629734% 9YqhflbMKe:0 = @param:9YqhflbMKe:0 -> float_type, {64, 29}, {29, 1}: 0.0005115ms, 0.00438422% main:@629 = slice[axes={0, 1},starts={0, 21},ends={64, 29}](9YqhflbMKe:0) -> float_type, {64, 8}, {29, 1}: 0.00077898ms, 0.00667688% ZcWERoz0h6:0 = @param:ZcWERoz0h6:0 -> float_type, {64, 49}, {49, 1}: 0.00048726ms, 0.00417646% main:@631 = slice[axes={0, 1},starts={0, 37},ends={64, 41}](ZcWERoz0h6:0) -> float_type, {64, 4}, {49, 1}: 0.000627ms, 0.00537421% ljyhc7HCLk:0 = @param:ljyhc7HCLk:0 -> float_type, {64, 21}, {21, 1}: 0.00049968ms, 0.00428291% main:@633 = slice[axes={0, 1},starts={0, 13},ends={64, 21}](ljyhc7HCLk:0) -> float_type, {64, 8}, {21, 1}: 0.00075334ms, 0.00645711% YbIVz8l1un:0 = @param:YbIVz8l1un:0 -> float_type, {64, 45}, {45, 1}: 0.0005207ms, 0.00446308% main:@635 = slice[axes={0, 1},starts={0, 29},ends={64, 37}](YbIVz8l1un:0) -> float_type, {64, 8}, {45, 1}: 0.00090014ms, 0.00771538% FcmWLZhDgE:0 = @param:FcmWLZhDgE:0 -> float_type, {64, 57}, {57, 1}: 0.00052612ms, 0.00450954% main:@637 = slice[axes={0, 1},starts={0, 41},ends={64, 49}](FcmWLZhDgE:0) -> float_type, {64, 8}, {57, 1}: 0.00076476ms, 0.00655499% main:@638 = slice[axes={0, 1},starts={0, 53},ends={64, 61}](Qp3K1PHvte:0) -> float_type, {64, 8}, {69, 1}: 0.00061974ms, 0.00531198% bJsIAzhYc3:0 = @param:bJsIAzhYc3:0 -> float_type, {64, 37}, {37, 1}: 0.00048668ms, 0.00417148% main:@640 = slice[axes={0, 1},starts={0, 25},ends={64, 29}](bJsIAzhYc3:0) -> float_type, {64, 4}, {37, 1}: 0.0008594ms, 0.00736618% main:@641 = slice[axes={0, 1},starts={0, 57},ends={64, 65}](ifP6Ceko5T:0) -> float_type, {64, 8}, {73, 1}: 0.00072624ms, 0.00622483% main:@642 = slice[axes={0, 1},starts={0, 217},ends={64, 233}](uFSOc5Chp2:0) -> float_type, {64, 16}, {481, 1}: 0.00060754ms, 0.00520741% main:@643 = slice[axes={0, 1},starts={0, 509},ends={64, 517}](lyiDxfzZg3:0) -> float_type, {64, 8}, {701, 1}: 0.00060696ms, 0.00520244% main:@644 = slice[axes={0, 1},starts={0, 197},ends={64, 237}](nv2tzWRf1I:0) -> float_type, {64, 40}, {301, 1}: 0.00087662ms, 0.00751378% isCn4mEOSt:0 = @param:isCn4mEOSt:0 -> float_type, {64, 49}, {49, 1}: 0.00053348ms, 0.00457262% main:@646 = slice[axes={0, 1},starts={0, 37},ends={64, 41}](isCn4mEOSt:0) -> float_type, {64, 4}, {49, 1}: 0.0007333ms, 0.00628534% w04Keh2FJu:0 = @param:w04Keh2FJu:0 -> float_type, {64, 13}, {13, 1}: 0.00051838ms, 0.00444319% main:@648 = slice[axes={0, 1},starts={0, 11},ends={64, 13}](w04Keh2FJu:0) -> float_type, {64, 2}, {13, 1}: 0.00072212ms, 0.00618951% main:@649 = load[offset=0,end=33280](main:@1) -> float_type, {64, 130}, {130, 1}: 0.00050622ms, 0.00433897% main:@650 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=594,local=256,](main:@643,main:@644,main:@642,main:@641,main:@638,main:@627,main:@635,main:@646,main:@631,main:@648,main:@637,main:@640,main:@633,main:@629,main:@649) -> float_type, {64, 130}, {130, 1}: 0.0285529ms, 0.244736% main:@651 = load[offset=778496,end=811264](main:@1) -> float_type, {64, 128}, {128, 1}: 0.00066396ms, 0.005691% main:@652 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@650,main:@625,main:@651) -> float_type, {64, 128}, {128, 1}: 0.0228354ms, 0.19573% main:@653 = load[offset=0,end=291840](main:@1) -> float_type, {64, 15, 76}, {1140, 76, 1}: 0.00065116ms, 0.00558129% a5QhOKE1Yc:0 = @param:a5QhOKE1Yc:0 -> float_type, {64, 15, 9}, {135, 9, 1}: 0.00064068ms, 0.00549146% main:@655 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={64, 15, 9}](a5QhOKE1Yc:0) -> float_type, {64, 15, 8}, {135, 9, 1}: 0.00072108ms, 0.0061806% W3KAtTdC92:0 = @param:W3KAtTdC92:0 -> float_type, {64, 15, 17}, {255, 17, 1}: 0.00052422ms, 0.00449325% main:@657 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={64, 15, 17}](W3KAtTdC92:0) -> float_type, {64, 15, 16}, {255, 17, 1}: 0.0009981ms, 0.00855502% ZebaMGw6vA:0 = @param:ZebaMGw6vA:0 -> float_type, {64, 15, 9}, {135, 9, 1}: 0.000574ms, 0.00491993% main:@659 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={64, 15, 9}](ZebaMGw6vA:0) -> float_type, {64, 15, 8}, {135, 9, 1}: 0.0006377ms, 0.00546592% wcHWv4SNE2:0 = @param:wcHWv4SNE2:0 -> float_type, {64, 15, 9}, {135, 9, 1}: 0.0004746ms, 0.00406794% main:@661 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={64, 15, 9}](wcHWv4SNE2:0) -> float_type, {64, 15, 8}, {135, 9, 1}: 0.00082914ms, 0.00710681% main:@662 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=12160,local=256,](main:@655,main:@98,main:@661,main:@657,main:@174,main:@659,main:@653) -> float_type, {64, 15, 76}, {1140, 76, 1}: 0.0211195ms, 0.181022% main:@663 = hip::hip_copy_literal[id=main:@literal:27] -> float_type, {76, 64}, {64, 1}: 0.00079596ms, 0.00682242% main:@664 = reshape_lazy[dims={960, 76}](main:@662) -> float_type, {960, 76}, {76, 1}: 0.00057894ms, 0.00496227% main:@665 = load[offset=3147264,end=3393024](main:@1) -> float_type, {960, 64}, {64, 1}: 0.0004707ms, 0.00403451% main:@666 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@664,main:@663,main:@665) -> float_type, {960, 64}, {64, 1}: 0.0216651ms, 0.185699% 6DYAlXkjub:0 = @param:6DYAlXkjub:0 -> float_type, {64, 30, 17}, {510, 17, 1}: 0.00073852ms, 0.00633008% main:@668 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={64, 30, 17}](6DYAlXkjub:0) -> float_type, {64, 30, 16}, {510, 17, 1}: 0.00068866ms, 0.00590272% 74jZTbEdLU:0 = @param:74jZTbEdLU:0 -> float_type, {64, 30, 9}, {270, 9, 1}: 0.00052386ms, 0.00449016% main:@670 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={64, 30, 9}](74jZTbEdLU:0) -> float_type, {64, 30, 8}, {270, 9, 1}: 0.00079202ms, 0.00678865% Z9QLFGO2Su:0 = @param:Z9QLFGO2Su:0 -> float_type, {64, 30, 9}, {270, 9, 1}: 0.00058662ms, 0.0050281% main:@672 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={64, 30, 9}](Z9QLFGO2Su:0) -> float_type, {64, 30, 8}, {270, 9, 1}: 0.00064322ms, 0.00551324% WixXdEln2I:0 = @param:WixXdEln2I:0 -> float_type, {64, 30, 17}, {510, 17, 1}: 0.00052756ms, 0.00452188% main:@674 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={64, 30, 17}](WixXdEln2I:0) -> float_type, {64, 30, 16}, {510, 17, 1}: 0.00100462ms, 0.00861091% IYWQSCjynR:0 = @param:IYWQSCjynR:0 -> float_type, {64, 30, 9}, {270, 9, 1}: 0.00050288ms, 0.00431034% main:@676 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={64, 30, 9}](IYWQSCjynR:0) -> float_type, {64, 30, 8}, {270, 9, 1}: 0.00064608ms, 0.00553775% VvUalZIQHK:0 = @param:VvUalZIQHK:0 -> float_type, {64, 30, 5}, {150, 5, 1}: 0.00052464ms, 0.00449685% main:@678 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={64, 30, 5}](VvUalZIQHK:0) -> float_type, {64, 30, 4}, {150, 5, 1}: 0.0007483ms, 0.00641391% bX19AmYnBL:0 = @param:bX19AmYnBL:0 -> float_type, {64, 30, 9}, {270, 9, 1}: 0.00051472ms, 0.00441182% main:@680 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={64, 30, 9}](bX19AmYnBL:0) -> float_type, {64, 30, 8}, {270, 9, 1}: 0.00066446ms, 0.00569529% main:@681 = load[offset=256256,end=778496](main:@1) -> float_type, {64, 30, 68}, {2040, 68, 1}: 0.00051438ms, 0.00440891% main:@682 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=18651,local=256,](main:@678,main:@674,main:@676,main:@668,main:@670,main:@680,main:@672,main:@681) -> float_type, {64, 30, 68}, {2040, 68, 1}: 0.025128ms, 0.21538% main:@683 = hip::hip_copy_literal[id=main:@literal:26] -> float_type, {68, 64}, {64, 1}: 0.00081326ms, 0.0069707% main:@684 = reshape_lazy[dims={1920, 68}](main:@682) -> float_type, {1920, 68}, {68, 1}: 0.00052656ms, 0.00451331% main:@685 = load[offset=4547072,end=5038592](main:@1) -> float_type, {1920, 64}, {64, 1}: 0.00045978ms, 0.00394092% main:@686 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@684,main:@683,main:@685) -> float_type, {1920, 64}, {64, 1}: 0.0237503ms, 0.203571% main:@687 = load[offset=2482432,end=2789632](main:@1) -> float_type, {64, 15, 80}, {1200, 80, 1}: 0.00065396ms, 0.00560529% yFeB2xg6sO:0 = @param:yFeB2xg6sO:0 -> float_type, {64, 15, 9}, {135, 9, 1}: 0.00053756ms, 0.00460759% main:@689 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={64, 15, 9}](yFeB2xg6sO:0) -> float_type, {64, 15, 8}, {135, 9, 1}: 0.00070994ms, 0.00608511% T3LHlorDnM:0 = @param:T3LHlorDnM:0 -> float_type, {64, 15, 17}, {255, 17, 1}: 0.00051218ms, 0.00439005% main:@691 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={64, 15, 17}](T3LHlorDnM:0) -> float_type, {64, 15, 16}, {255, 17, 1}: 0.00075364ms, 0.00645968% kw7Kzgp1JD:0 = @param:kw7Kzgp1JD:0 -> float_type, {64, 15, 21}, {315, 21, 1}: 0.00057994ms, 0.00497084% main:@693 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={64, 15, 21}](kw7Kzgp1JD:0) -> float_type, {64, 15, 20}, {315, 21, 1}: 0.0006471ms, 0.00554649% hGtXM6ZJUv:0 = @param:hGtXM6ZJUv:0 -> float_type, {64, 15, 9}, {135, 9, 1}: 0.00053036ms, 0.00454588% main:@695 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={64, 15, 9}](hGtXM6ZJUv:0) -> float_type, {64, 15, 8}, {135, 9, 1}: 0.00080206ms, 0.0068747% main:@696 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=15360,local=256,](main:@695,main:@691,main:@117,main:@693,main:@689,main:@687) -> float_type, {64, 15, 80}, {1200, 80, 1}: 0.0197548ms, 0.169324% main:@697 = hip::hip_copy_literal[id=main:@literal:10] -> float_type, {80, 64}, {64, 1}: 0.00072568ms, 0.00622003% main:@698 = load[offset=2007296,end=2253056](main:@1) -> float_type, {960, 64}, {64, 1}: 0.00049128ms, 0.00421091% main:@699 = reshape_lazy[dims={960, 80}](main:@696) -> float_type, {960, 80}, {80, 1}: 0.00054952ms, 0.0047101% main:@700 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@699,main:@697,main:@698) -> float_type, {960, 64}, {64, 1}: 0.0203637ms, 0.174544% main:@701 = hip::hip_copy_literal[id=main:@literal:162] -> float_type, {256}, {1}: 0.00082184ms, 0.00704424% main:@702 = hip::hip_copy_literal[id=main:@literal:6] -> float_type, {64, 18, 256}, {4608, 256, 1}: 0.00065946ms, 0.00565243% main:@703 = load[offset=5038592,end=6218240](main:@1) -> float_type, {64, 18, 256}, {4608, 256, 1}: 0.00049006ms, 0.00420045% main:@704 = slice[axes={0},starts={192},ends={256}](main:@386) -> float_type, {64, 18, 64}, {1152, 64, 1}: 0.00077268ms, 0.00662288% main:@705 = slice[axes={0},starts={128},ends={192}](main:@386) -> float_type, {64, 18, 64}, {1152, 64, 1}: 0.00086006ms, 0.00737184% main:@706 = slice[axes={0},starts={64},ends={128}](main:@386) -> float_type, {64, 18, 64}, {1152, 64, 1}: 0.00060598ms, 0.00519404% main:@707 = slice[axes={0},starts={0},ends={64}](main:@386) -> float_type, {64, 18, 64}, {1152, 64, 1}: 0.00056354ms, 0.00483027% main:@708 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=73728,local=256,](main:@707,main:@706,main:@705,main:@704,main:@703) -> float_type, {64, 18, 256}, {4608, 256, 1}: 0.0202846ms, 0.173865% main:@709 = hip::hip_copy_literal[id=main:@literal:163] -> float_type, {256}, {1}: 0.00078794ms, 0.00675368% main:@710 = multibroadcast[out_lens={64, 18, 256}](main:@709) -> float_type, {64, 18, 256}, {0, 0, 1}: 0.00064694ms, 0.00554512% main:@711 = multibroadcast[out_lens={64, 18, 256}](main:@701) -> float_type, {64, 18, 256}, {0, 0, 1}: 0.00070258ms, 0.00602203% main:@712 = load[offset=811264,end=1990912](main:@1) -> float_type, {64, 18, 256}, {4608, 256, 1}: 0.00050126ms, 0.00429645% main:@713 = gpu::code_object[code_object=13624,symbol_name=mul_relu_reduce_prod_sub_mul_reduce_sum_add_add_add_add_pow_kernel,global=147456,local=128,](main:@708,main:@702,main:@711,main:@710,main:@712) -> float_type, {64, 18, 256}, {4608, 256, 1}: 0.0264946ms, 0.227093% main:@714 = hip::hip_copy_literal[id=main:@literal:20] -> float_type, {896}, {1}: 0.00072694ms, 0.00623083% main:@715 = load[offset=0,end=229376](main:@1) -> float_type, {64, 896}, {896, 1}: 0.0004588ms, 0.00393252% main:@716 = multibroadcast[out_lens={64, 896}](main:@714) -> float_type, {64, 896}, {0, 1}: 0.00056804ms, 0.00486885% main:@717 = gpu::code_object[code_object=13536,symbol_name=add_leaky_relu_kernel,global=28672,local=256,](main:@380,main:@716,main:@715) -> float_type, {64, 896}, {896, 1}: 0.0179531ms, 0.153882% main:@718 = hip::hip_copy_literal[id=main:@literal:146] -> float_type, {256, 512}, {512, 1}: 0.00072996ms, 0.00625671% main:@719 = hip::hip_copy_literal[id=main:@literal:154] -> float_type, {512, 256}, {256, 1}: 0.00058822ms, 0.00504181% main:@720 = hip::hip_copy_literal[id=main:@literal:164] -> float_type, {256, 512}, {512, 1}: 0.00066942ms, 0.0057378% 06AyS2QTzo:0 = @param:06AyS2QTzo:0 -> float_type, {64, 50, 9}, {450, 9, 1}: 0.0005374ms, 0.00460622% main:@722 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={64, 50, 9}](06AyS2QTzo:0) -> float_type, {64, 50, 8}, {450, 9, 1}: 0.00088616ms, 0.00759555% main:@723 = load[offset=3393024,end=4365824](main:@1) -> float_type, {64, 50, 76}, {3800, 76, 1}: 0.00048434ms, 0.00415143% Uc35C1q4oy:0 = @param:Uc35C1q4oy:0 -> float_type, {64, 50, 9}, {450, 9, 1}: 0.00052328ms, 0.00448519% main:@725 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={64, 50, 9}](Uc35C1q4oy:0) -> float_type, {64, 50, 8}, {450, 9, 1}: 0.00089606ms, 0.00768041% IFQJaVMDS5:0 = @param:IFQJaVMDS5:0 -> float_type, {64, 50, 17}, {850, 17, 1}: 0.00051606ms, 0.00442331% main:@727 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={64, 50, 17}](IFQJaVMDS5:0) -> float_type, {64, 50, 16}, {850, 17, 1}: 0.00084834ms, 0.00727138% r8QjI5SCed:0 = @param:r8QjI5SCed:0 -> float_type, {64, 50, 9}, {450, 9, 1}: 0.00049982ms, 0.00428411% main:@729 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={64, 50, 9}](r8QjI5SCed:0) -> float_type, {64, 50, 8}, {450, 9, 1}: 0.00063418ms, 0.00543575% main:@730 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=40533,local=256,](main:@255,main:@722,main:@727,main:@729,main:@155,main:@725,main:@723) -> float_type, {64, 50, 76}, {3800, 76, 1}: 0.0253718ms, 0.217469% main:@731 = hip::hip_copy_literal[id=main:@literal:7] -> float_type, {76, 128}, {128, 1}: 0.00078358ms, 0.0067163% main:@732 = load[offset=12247552,end=13885952](main:@1) -> float_type, {3200, 128}, {128, 1}: 0.00048132ms, 0.00412554% main:@733 = reshape_lazy[dims={3200, 76}](main:@730) -> float_type, {3200, 76}, {76, 1}: 0.00055268ms, 0.00473719% main:@734 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@733,main:@731,main:@732) -> float_type, {3200, 128}, {128, 1}: 0.0258012ms, 0.22115% main:@735 = load[offset=13885952,end=15114752](main:@1) -> float_type, {64, 50, 96}, {4800, 96, 1}: 0.00065502ms, 0.00561438% 0pnqc4Em1y:0 = @param:0pnqc4Em1y:0 -> float_type, {64, 50, 9}, {450, 9, 1}: 0.00053304ms, 0.00456885% main:@737 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={64, 50, 9}](0pnqc4Em1y:0) -> float_type, {64, 50, 8}, {450, 9, 1}: 0.00066012ms, 0.00565809% hKxguEQn94:0 = @param:hKxguEQn94:0 -> float_type, {64, 50, 9}, {450, 9, 1}: 0.00053204ms, 0.00456028% main:@739 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={64, 50, 9}](hKxguEQn94:0) -> float_type, {64, 50, 8}, {450, 9, 1}: 0.0007879ms, 0.00675333% D0gn7y6swk:0 = @param:D0gn7y6swk:0 -> float_type, {64, 50, 17}, {850, 17, 1}: 0.00049932ms, 0.00427982% main:@741 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={64, 50, 17}](D0gn7y6swk:0) -> float_type, {64, 50, 16}, {850, 17, 1}: 0.00065776ms, 0.00563786% SKG7aiAo3m:0 = @param:SKG7aiAo3m:0 -> float_type, {64, 50, 9}, {450, 9, 1}: 0.0004866ms, 0.0041708% main:@743 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={64, 50, 9}](SKG7aiAo3m:0) -> float_type, {64, 50, 8}, {450, 9, 1}: 0.00103576ms, 0.00887782% UpWkNPIeJO:0 = @param:UpWkNPIeJO:0 -> float_type, {64, 50, 17}, {850, 17, 1}: 0.0005068ms, 0.00434394% main:@745 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={64, 50, 17}](UpWkNPIeJO:0) -> float_type, {64, 50, 16}, {850, 17, 1}: 0.00064678ms, 0.00554375% nEuw2U7mFR:0 = @param:nEuw2U7mFR:0 -> float_type, {64, 50, 41}, {2050, 41, 1}: 0.0005908ms, 0.00506393% main:@747 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={64, 50, 41}](nEuw2U7mFR:0) -> float_type, {64, 50, 40}, {2050, 41, 1}: 0.00076056ms, 0.00651899% main:@748 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=51200,local=256,](main:@747,main:@739,main:@743,main:@741,main:@745,main:@737,main:@735) -> float_type, {64, 50, 96}, {4800, 96, 1}: 0.0265635ms, 0.227684% main:@749 = hip::hip_copy_literal[id=main:@literal:14] -> float_type, {96, 256}, {256, 1}: 0.00077846ms, 0.00667242% main:@750 = reshape_lazy[dims={3200, 96}](main:@748) -> float_type, {3200, 96}, {96, 1}: 0.00053102ms, 0.00455154% main:@751 = load[offset=6021632,end=9298432](main:@1) -> float_type, {3200, 256}, {256, 1}: 0.00045554ms, 0.00390457% main:@752 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@750,main:@749,main:@751) -> float_type, {3200, 256}, {256, 1}: 0.0309396ms, 0.265193% m2YscaoDjP:0 = @param:m2YscaoDjP:0 -> float_type, {64, 17}, {17, 1}: 0.00068208ms, 0.00584632% main:@754 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](m2YscaoDjP:0) -> float_type, {64, 4}, {17, 1}: 0.00071114ms, 0.0060954% 09zp4EDXTh:0 = @param:09zp4EDXTh:0 -> float_type, {64, 25}, {25, 1}: 0.00057834ms, 0.00495713% main:@756 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](09zp4EDXTh:0) -> float_type, {64, 8}, {25, 1}: 0.00080294ms, 0.00688225% HZ0l3MmcPY:0 = @param:HZ0l3MmcPY:0 -> float_type, {64, 17}, {17, 1}: 0.00049726ms, 0.00426217% main:@758 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](HZ0l3MmcPY:0) -> float_type, {64, 4}, {17, 1}: 0.00065618ms, 0.00562432% o74vNAnXxy:0 = @param:o74vNAnXxy:0 -> float_type, {64, 17}, {17, 1}: 0.0005025ms, 0.00430708% main:@760 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](o74vNAnXxy:0) -> float_type, {64, 4}, {17, 1}: 0.00077658ms, 0.00665631% 1PJsBhDaCK:0 = @param:1PJsBhDaCK:0 -> float_type, {64, 25}, {25, 1}: 0.00056082ms, 0.00480696% main:@762 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](1PJsBhDaCK:0) -> float_type, {64, 8}, {25, 1}: 0.00065912ms, 0.00564952% 1WNvHElUG7:0 = @param:1WNvHElUG7:0 -> float_type, {64, 25}, {25, 1}: 0.00053262ms, 0.00456525% main:@764 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](1WNvHElUG7:0) -> float_type, {64, 8}, {25, 1}: 0.00079294ms, 0.00679653% 1u5yIsHTXE:0 = @param:1u5yIsHTXE:0 -> float_type, {64, 17}, {17, 1}: 0.00046056ms, 0.0039476% main:@766 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](1u5yIsHTXE:0) -> float_type, {64, 4}, {17, 1}: 0.00089032ms, 0.00763121% 2lVXMZgJfO:0 = @param:2lVXMZgJfO:0 -> float_type, {64, 25}, {25, 1}: 0.00049284ms, 0.00422428% main:@768 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](2lVXMZgJfO:0) -> float_type, {64, 8}, {25, 1}: 0.00074652ms, 0.00639865% 3kdi0vsSZG:0 = @param:3kdi0vsSZG:0 -> float_type, {64, 25}, {25, 1}: 0.0005708ms, 0.0048925% main:@770 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](3kdi0vsSZG:0) -> float_type, {64, 8}, {25, 1}: 0.00061112ms, 0.0052381% Tg4u0GfnFb:0 = @param:Tg4u0GfnFb:0 -> float_type, {64, 25}, {25, 1}: 0.0005505ms, 0.0047185% main:@772 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](Tg4u0GfnFb:0) -> float_type, {64, 8}, {25, 1}: 0.00069348ms, 0.00594403% 5JHUPChra8:0 = @param:5JHUPChra8:0 -> float_type, {64, 17}, {17, 1}: 0.00053644ms, 0.00459799% main:@774 = slice[axes={0, 1},starts={0, 1},ends={64, 3}](5JHUPChra8:0) -> float_type, {64, 2}, {17, 1}: 0.00070942ms, 0.00608066% xymEBUMS25:0 = @param:xymEBUMS25:0 -> float_type, {64, 13}, {13, 1}: 0.00046892ms, 0.00401926% main:@776 = slice[axes={0, 1},starts={0, 1},ends={64, 3}](xymEBUMS25:0) -> float_type, {64, 2}, {13, 1}: 0.0006985ms, 0.00598706% zZHaVOckpw:0 = @param:zZHaVOckpw:0 -> float_type, {64, 13}, {13, 1}: 0.00052378ms, 0.00448948% main:@778 = slice[axes={0, 1},starts={0, 1},ends={64, 3}](zZHaVOckpw:0) -> float_type, {64, 2}, {13, 1}: 0.00070596ms, 0.006051% pH5srbufJ1:0 = @param:pH5srbufJ1:0 -> float_type, {64, 17}, {17, 1}: 0.00050136ms, 0.00429731% main:@780 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](pH5srbufJ1:0) -> float_type, {64, 4}, {17, 1}: 0.00069538ms, 0.00596032% main:@781 = load[offset=2253056,end=2302720](main:@1) -> float_type, {64, 194}, {194, 1}: 0.00048312ms, 0.00414097% msiPzSBKMj:0 = @param:msiPzSBKMj:0 -> float_type, {64, 17}, {17, 1}: 0.00048364ms, 0.00414543% main:@783 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](msiPzSBKMj:0) -> float_type, {64, 4}, {17, 1}: 0.0006752ms, 0.00578735% 92MwiALCHX:0 = @param:92MwiALCHX:0 -> float_type, {64, 25}, {25, 1}: 0.00050322ms, 0.00431325% main:@785 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](92MwiALCHX:0) -> float_type, {64, 8}, {25, 1}: 0.00065872ms, 0.00564609% ujOA5LQW7M:0 = @param:ujOA5LQW7M:0 -> float_type, {64, 13}, {13, 1}: 0.00051746ms, 0.00443531% main:@787 = slice[axes={0, 1},starts={0, 1},ends={64, 3}](ujOA5LQW7M:0) -> float_type, {64, 2}, {13, 1}: 0.0006857ms, 0.00587735% AVLcBpWeuo:0 = @param:AVLcBpWeuo:0 -> float_type, {64, 21}, {21, 1}: 0.0005668ms, 0.00485822% main:@789 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](AVLcBpWeuo:0) -> float_type, {64, 4}, {21, 1}: 0.00068678ms, 0.0058866% main:@790 = slice[axes={0, 1},starts={0, 17},ends={64, 21}](ifP6Ceko5T:0) -> float_type, {64, 4}, {73, 1}: 0.00066746ms, 0.005721% RpITWntAji:0 = @param:RpITWntAji:0 -> float_type, {64, 17}, {17, 1}: 0.00047652ms, 0.0040844% main:@792 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](RpITWntAji:0) -> float_type, {64, 8}, {17, 1}: 0.00067348ms, 0.0057726% IztrlRfWQU:0 = @param:IztrlRfWQU:0 -> float_type, {64, 13}, {13, 1}: 0.00050336ms, 0.00431445% main:@794 = slice[axes={0, 1},starts={0, 1},ends={64, 3}](IztrlRfWQU:0) -> float_type, {64, 2}, {13, 1}: 0.00069274ms, 0.00593769% JSWhLK4Agx:0 = @param:JSWhLK4Agx:0 -> float_type, {64, 25}, {25, 1}: 0.00048778ms, 0.00418091% main:@796 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](JSWhLK4Agx:0) -> float_type, {64, 8}, {25, 1}: 0.00065024ms, 0.00557341% KN13IWXZSu:0 = @param:KN13IWXZSu:0 -> float_type, {64, 13}, {13, 1}: 0.0005001ms, 0.00428651% main:@798 = slice[axes={0, 1},starts={0, 1},ends={64, 3}](KN13IWXZSu:0) -> float_type, {64, 2}, {13, 1}: 0.0009745ms, 0.00835274% NXOSdQVPs0:0 = @param:NXOSdQVPs0:0 -> float_type, {64, 25}, {25, 1}: 0.00054776ms, 0.00469502% main:@800 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](NXOSdQVPs0:0) -> float_type, {64, 8}, {25, 1}: 0.0006938ms, 0.00594677% ry2xad1vc9:0 = @param:ry2xad1vc9:0 -> float_type, {64, 25}, {25, 1}: 0.00048142ms, 0.0041264% main:@802 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](ry2xad1vc9:0) -> float_type, {64, 8}, {25, 1}: 0.00068026ms, 0.00583072% main:@803 = slice[axes={0, 1},starts={0, 17},ends={64, 25}](Qp3K1PHvte:0) -> float_type, {64, 8}, {69, 1}: 0.00092452ms, 0.00792434% RHo9Jz6KpI:0 = @param:RHo9Jz6KpI:0 -> float_type, {64, 21}, {21, 1}: 0.0004813ms, 0.00412537% main:@805 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](RHo9Jz6KpI:0) -> float_type, {64, 4}, {21, 1}: 0.00064854ms, 0.00555884% S5D6WGIXMd:0 = @param:S5D6WGIXMd:0 -> float_type, {64, 17}, {17, 1}: 0.00063024ms, 0.00540198% main:@807 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](S5D6WGIXMd:0) -> float_type, {64, 4}, {17, 1}: 0.00065198ms, 0.00558832% TidKGrnobI:0 = @param:TidKGrnobI:0 -> float_type, {64, 13}, {13, 1}: 0.00048906ms, 0.00419188% main:@809 = slice[axes={0, 1},starts={0, 1},ends={64, 3}](TidKGrnobI:0) -> float_type, {64, 2}, {13, 1}: 0.00066408ms, 0.00569203% XxirNgB98u:0 = @param:XxirNgB98u:0 -> float_type, {64, 13}, {13, 1}: 0.00049644ms, 0.00425514% main:@811 = slice[axes={0, 1},starts={0, 1},ends={64, 3}](XxirNgB98u:0) -> float_type, {64, 2}, {13, 1}: 0.00066362ms, 0.00568809% main:@812 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](YbIVz8l1un:0) -> float_type, {64, 8}, {45, 1}: 0.00063216ms, 0.00541844% main:@813 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](bJsIAzhYc3:0) -> float_type, {64, 8}, {37, 1}: 0.0009381ms, 0.00804074% tWcngAq7hV:0 = @param:tWcngAq7hV:0 -> float_type, {64, 13}, {13, 1}: 0.00048176ms, 0.00412931% main:@815 = slice[axes={0, 1},starts={0, 1},ends={64, 3}](tWcngAq7hV:0) -> float_type, {64, 2}, {13, 1}: 0.00068678ms, 0.0058866% nPUSZFm74Y:0 = @param:nPUSZFm74Y:0 -> float_type, {64, 13}, {13, 1}: 0.0005424ms, 0.00464908% main:@817 = slice[axes={0, 1},starts={0, 1},ends={64, 3}](nPUSZFm74Y:0) -> float_type, {64, 2}, {13, 1}: 0.00064982ms, 0.00556981% xT9FhjM1Uu:0 = @param:xT9FhjM1Uu:0 -> float_type, {64, 17}, {17, 1}: 0.00049792ms, 0.00426783% main:@819 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](xT9FhjM1Uu:0) -> float_type, {64, 4}, {17, 1}: 0.00069618ms, 0.00596717% main:@820 = slice[axes={0, 1},starts={0, 161},ends={64, 169}](uFSOc5Chp2:0) -> float_type, {64, 8}, {481, 1}: 0.0006172ms, 0.00529021% qd1komWTn4:0 = @param:qd1komWTn4:0 -> float_type, {64, 25}, {25, 1}: 0.00046806ms, 0.00401189% main:@822 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](qd1komWTn4:0) -> float_type, {64, 8}, {25, 1}: 0.00065984ms, 0.00565569% main:@823 = slice[axes={0, 1},starts={0, 161},ends={64, 163}](nv2tzWRf1I:0) -> float_type, {64, 2}, {301, 1}: 0.00091118ms, 0.00781% main:@824 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=326,local=256,](main:@774,main:@811,main:@815,main:@776,main:@794,main:@798,main:@817,main:@787,main:@778,main:@809,main:@823,main:@758,main:@780,main:@783,main:@766,main:@805,main:@807,main:@754,main:@789,main:@760,main:@819,main:@790,main:@756,main:@762,main:@803,main:@796,main:@822,main:@802,main:@768,main:@785,main:@772,main:@764,main:@800,main:@813,main:@770,main:@812,main:@792,main:@820,main:@781) -> float_type, {64, 194}, {194, 1}: 0.0504671ms, 0.432569% main:@825 = hip::hip_copy_literal[id=main:@literal:143] -> float_type, {194, 192}, {192, 1}: 0.0008173ms, 0.00700533% main:@826 = load[offset=2302720,end=2351872](main:@1) -> float_type, {64, 192}, {192, 1}: 0.00051402ms, 0.00440582% main:@827 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@824,main:@825,main:@826) -> float_type, {64, 192}, {192, 1}: 0.0255782ms, 0.219239% main:@828 = slice[axes={0, 1},starts={0, 1},ends={64, 17}](isCn4mEOSt:0) -> float_type, {64, 16}, {49, 1}: 0.00083334ms, 0.00714281% main:@829 = unsqueeze[axes={1},steps={}](main:@828) -> float_type, {64, 1, 16}, {49, 16, 1}: 0.00083084ms, 0.00712138% 6HSQqXdJzb:0 = @param:6HSQqXdJzb:0 -> float_type, {64, 45}, {45, 1}: 0.00057196ms, 0.00490244% main:@831 = slice[axes={0, 1},starts={0, 1},ends={64, 17}](6HSQqXdJzb:0) -> float_type, {64, 16}, {45, 1}: 0.00079878ms, 0.00684659% main:@832 = unsqueeze[axes={1},steps={}](main:@831) -> float_type, {64, 1, 16}, {45, 16, 1}: 0.0005779ms, 0.00495336% main:@833 = slice[axes={0, 1},starts={0, 1},ends={64, 17}](2CnSyUBzm0:0) -> float_type, {64, 16}, {65, 1}: 0.0006316ms, 0.00541364% main:@834 = unsqueeze[axes={1},steps={}](main:@833) -> float_type, {64, 1, 16}, {65, 16, 1}: 0.00057916ms, 0.00496416% ahAlEtBvzo:0 = @param:ahAlEtBvzo:0 -> float_type, {64, 41}, {41, 1}: 0.00052326ms, 0.00448502% main:@836 = slice[axes={0, 1},starts={0, 1},ends={64, 17}](ahAlEtBvzo:0) -> float_type, {64, 16}, {41, 1}: 0.00073716ms, 0.00631842% main:@837 = unsqueeze[axes={1},steps={}](main:@836) -> float_type, {64, 1, 16}, {41, 16, 1}: 0.00059264ms, 0.0050797% Dy3rwVmn1A:0 = @param:Dy3rwVmn1A:0 -> float_type, {64, 41}, {41, 1}: 0.00051784ms, 0.00443857% main:@839 = slice[axes={0, 1},starts={0, 1},ends={64, 17}](Dy3rwVmn1A:0) -> float_type, {64, 16}, {41, 1}: 0.00064052ms, 0.00549009% main:@840 = unsqueeze[axes={1},steps={}](main:@839) -> float_type, {64, 1, 16}, {41, 16, 1}: 0.00064134ms, 0.00549712% main:@841 = slice[axes={0, 1},starts={0, 1},ends={64, 17}](FcmWLZhDgE:0) -> float_type, {64, 16}, {57, 1}: 0.00075206ms, 0.00644614% main:@842 = unsqueeze[axes={1},steps={}](main:@841) -> float_type, {64, 1, 16}, {57, 16, 1}: 0.00071996ms, 0.006171% 1aXQU27NrD:0 = @param:1aXQU27NrD:0 -> float_type, {64, 49}, {49, 1}: 0.0004675ms, 0.00400709% main:@844 = slice[axes={0, 1},starts={0, 1},ends={64, 17}](1aXQU27NrD:0) -> float_type, {64, 16}, {49, 1}: 0.00060788ms, 0.00521033% main:@845 = unsqueeze[axes={1},steps={}](main:@844) -> float_type, {64, 1, 16}, {49, 16, 1}: 0.00056998ms, 0.00488547% X29iyIeu8L:0 = @param:X29iyIeu8L:0 -> float_type, {64, 45}, {45, 1}: 0.00055512ms, 0.0047581% main:@847 = slice[axes={0, 1},starts={0, 1},ends={64, 17}](X29iyIeu8L:0) -> float_type, {64, 16}, {45, 1}: 0.00079222ms, 0.00679036% main:@848 = unsqueeze[axes={1},steps={}](main:@847) -> float_type, {64, 1, 16}, {45, 16, 1}: 0.00056856ms, 0.0048733% main:@849 = slice[axes={0, 1},starts={0, 1},ends={64, 17}](ZcWERoz0h6:0) -> float_type, {64, 16}, {49, 1}: 0.00062086ms, 0.00532158% main:@850 = unsqueeze[axes={1},steps={}](main:@849) -> float_type, {64, 1, 16}, {49, 16, 1}: 0.000543ms, 0.00465422% mgnGNH4Mz3:0 = @param:mgnGNH4Mz3:0 -> float_type, {64, 49}, {49, 1}: 0.0005297ms, 0.00454022% main:@852 = slice[axes={0, 1},starts={0, 1},ends={64, 17}](mgnGNH4Mz3:0) -> float_type, {64, 16}, {49, 1}: 0.00072568ms, 0.00622003% main:@853 = unsqueeze[axes={1},steps={}](main:@852) -> float_type, {64, 1, 16}, {49, 16, 1}: 0.00058022ms, 0.00497324% zIoqYe9cRb:0 = @param:zIoqYe9cRb:0 -> float_type, {64, 45}, {45, 1}: 0.00050984ms, 0.00437% main:@855 = slice[axes={0, 1},starts={0, 1},ends={64, 17}](zIoqYe9cRb:0) -> float_type, {64, 16}, {45, 1}: 0.0006656ms, 0.00570506% main:@856 = unsqueeze[axes={1},steps={}](main:@855) -> float_type, {64, 1, 16}, {45, 16, 1}: 0.00087316ms, 0.00748412% dAFYcGrfEl:0 = @param:dAFYcGrfEl:0 -> float_type, {64, 45}, {45, 1}: 0.0004904ms, 0.00420337% main:@858 = slice[axes={0, 1},starts={0, 1},ends={64, 17}](dAFYcGrfEl:0) -> float_type, {64, 16}, {45, 1}: 0.00081224ms, 0.00696196% main:@859 = unsqueeze[axes={1},steps={}](main:@858) -> float_type, {64, 1, 16}, {45, 16, 1}: 0.00055604ms, 0.00476599% pYw7fMVoQg:0 = @param:pYw7fMVoQg:0 -> float_type, {64, 41}, {41, 1}: 0.00053244ms, 0.00456371% main:@861 = slice[axes={0, 1},starts={0, 1},ends={64, 17}](pYw7fMVoQg:0) -> float_type, {64, 16}, {41, 1}: 0.00070916ms, 0.00607843% main:@862 = unsqueeze[axes={1},steps={}](main:@861) -> float_type, {64, 1, 16}, {41, 16, 1}: 0.0005703ms, 0.00488822% main:@863 = reshape_lazy[dims={-1, 12, 16}](main:@827) -> float_type, {64, 12, 16}, {192, 16, 1}: 0.00051894ms, 0.00444799% main:@864 = load[offset=2375936,end=2478336](main:@1) -> float_type, {64, 25, 16}, {400, 16, 1}: 0.00047576ms, 0.00407788% main:@865 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=1828,local=256,](main:@834,main:@853,main:@845,main:@832,main:@856,main:@859,main:@848,main:@850,main:@837,main:@842,main:@862,main:@840,main:@829,main:@863,main:@864) -> float_type, {64, 25, 16}, {400, 16, 1}: 0.0290999ms, 0.249424% main:@866 = transpose[permutation={0, 2, 1}](main:@865) -> float_type, {64, 16, 25}, {400, 1, 16}: 0.00087618ms, 0.00751001% main:@867 = load[offset=2273536,end=2375936](main:@1) -> float_type, {64, 16, 25}, {400, 25, 1}: 0.00051032ms, 0.00437411% main:@868 = gpu::code_object[code_object=13528,symbol_name=contiguous_kernel,global=25600,local=256,](main:@866,main:@867) -> float_type, {64, 16, 25}, {400, 25, 1}: 0.0170112ms, 0.145809% main:@869 = hip::hip_copy_literal[id=main:@literal:115] -> float_type, {25, 5}, {5, 1}: 0.00076104ms, 0.00652311% main:@870 = reshape_lazy[dims={1024, 25}](main:@868) -> float_type, {1024, 25}, {25, 1}: 0.0005251ms, 0.00450079% main:@871 = load[offset=2253056,end=2273536](main:@1) -> float_type, {1024, 5}, {5, 1}: 0.00045664ms, 0.003914% main:@872 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@870,main:@869,main:@871) -> float_type, {1024, 5}, {5, 1}: 0.0212037ms, 0.181743% main:@873 = hip::hip_copy_literal[id=main:@literal:75] -> float_type, {160, 160}, {160, 1}: 0.0008084ms, 0.00692904% main:@874 = load[offset=2273536,end=2314496](main:@1) -> float_type, {64, 160}, {160, 1}: 0.00047212ms, 0.00404669% 4M92ryD8ux:0 = @param:4M92ryD8ux:0 -> float_type, {64, 9}, {9, 1}: 0.00050656ms, 0.00434188% main:@876 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](4M92ryD8ux:0) -> float_type, {64, 4}, {9, 1}: 0.00068922ms, 0.00590752% 4gFa5o2YsJ:0 = @param:4gFa5o2YsJ:0 -> float_type, {64, 9}, {9, 1}: 0.00047216ms, 0.00404703% main:@878 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](4gFa5o2YsJ:0) -> float_type, {64, 4}, {9, 1}: 0.00087418ms, 0.00749287% 6XMrIsBjLY:0 = @param:6XMrIsBjLY:0 -> float_type, {64, 13}, {13, 1}: 0.0004538ms, 0.00388966% main:@880 = slice[axes={0, 1},starts={0, 1},ends={64, 3}](6XMrIsBjLY:0) -> float_type, {64, 2}, {13, 1}: 0.00065694ms, 0.00563083% 72zWmh93Bo:0 = @param:72zWmh93Bo:0 -> float_type, {64, 17}, {17, 1}: 0.00051852ms, 0.00444439% main:@882 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](72zWmh93Bo:0) -> float_type, {64, 4}, {17, 1}: 0.00081042ms, 0.00694636% JjyOIxALzM:0 = @param:JjyOIxALzM:0 -> float_type, {64, 5}, {5, 1}: 0.00061836ms, 0.00530015% main:@884 = slice[axes={0, 1},starts={0, 1},ends={64, 3}](JjyOIxALzM:0) -> float_type, {64, 2}, {5, 1}: 0.0006293ms, 0.00539392% 8vM5p4RzXd:0 = @param:8vM5p4RzXd:0 -> float_type, {64, 13}, {13, 1}: 0.00063232ms, 0.00541981% main:@886 = slice[axes={0, 1},starts={0, 1},ends={64, 3}](8vM5p4RzXd:0) -> float_type, {64, 2}, {13, 1}: 0.00079364ms, 0.00680253% AB6lekmMxa:0 = @param:AB6lekmMxa:0 -> float_type, {64, 25}, {25, 1}: 0.00052332ms, 0.00448554% main:@888 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](AB6lekmMxa:0) -> float_type, {64, 8}, {25, 1}: 0.00065812ms, 0.00564095% 9dgeU4bKka:0 = @param:9dgeU4bKka:0 -> float_type, {64, 17}, {17, 1}: 0.0004815ms, 0.00412708% main:@890 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](9dgeU4bKka:0) -> float_type, {64, 4}, {17, 1}: 0.00077112ms, 0.00660951% EZBIAqn6LS:0 = @param:EZBIAqn6LS:0 -> float_type, {64, 13}, {13, 1}: 0.00050272ms, 0.00430897% main:@892 = slice[axes={0, 1},starts={0, 1},ends={64, 3}](EZBIAqn6LS:0) -> float_type, {64, 2}, {13, 1}: 0.0009173ms, 0.00786246% BYyjkPCX0m:0 = @param:BYyjkPCX0m:0 -> float_type, {64, 13}, {13, 1}: 0.0004942ms, 0.00423594% main:@894 = slice[axes={0, 1},starts={0, 1},ends={64, 3}](BYyjkPCX0m:0) -> float_type, {64, 2}, {13, 1}: 0.00073012ms, 0.00625808% EKpP4xMXu8:0 = @param:EKpP4xMXu8:0 -> float_type, {64, 17}, {17, 1}: 0.0005102ms, 0.00437308% main:@896 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](EKpP4xMXu8:0) -> float_type, {64, 4}, {17, 1}: 0.00069598ms, 0.00596546% EUciqFuptg:0 = @param:EUciqFuptg:0 -> float_type, {64, 17}, {17, 1}: 0.0004705ms, 0.0040328% main:@898 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](EUciqFuptg:0) -> float_type, {64, 4}, {17, 1}: 0.0007167ms, 0.00614306% nyLxrsTmOB:0 = @param:nyLxrsTmOB:0 -> float_type, {64, 17}, {17, 1}: 0.00051428ms, 0.00440805% main:@900 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](nyLxrsTmOB:0) -> float_type, {64, 8}, {17, 1}: 0.00068374ms, 0.00586055% Hnjq487SIY:0 = @param:Hnjq487SIY:0 -> float_type, {64, 13}, {13, 1}: 0.00048788ms, 0.00418177% main:@902 = slice[axes={0, 1},starts={0, 1},ends={64, 3}](Hnjq487SIY:0) -> float_type, {64, 2}, {13, 1}: 0.00069234ms, 0.00593426% IRem4MiFDP:0 = @param:IRem4MiFDP:0 -> float_type, {64, 5}, {5, 1}: 0.00051902ms, 0.00444868% main:@904 = slice[axes={0, 1},starts={0, 1},ends={64, 3}](IRem4MiFDP:0) -> float_type, {64, 2}, {5, 1}: 0.0007032ms, 0.00602734% KasFvpq4IG:0 = @param:KasFvpq4IG:0 -> float_type, {64, 5}, {5, 1}: 0.00054706ms, 0.00468902% main:@906 = slice[axes={0, 1},starts={0, 1},ends={64, 3}](KasFvpq4IG:0) -> float_type, {64, 2}, {5, 1}: 0.00065664ms, 0.00562826% JCVMzpWyri:0 = @param:JCVMzpWyri:0 -> float_type, {64, 17}, {17, 1}: 0.00045226ms, 0.00387646% main:@908 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](JCVMzpWyri:0) -> float_type, {64, 4}, {17, 1}: 0.0009759ms, 0.00836474% JSPsGophtx:0 = @param:JSPsGophtx:0 -> float_type, {64, 25}, {25, 1}: 0.00048702ms, 0.0041744% main:@910 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](JSPsGophtx:0) -> float_type, {64, 8}, {25, 1}: 0.00067904ms, 0.00582026% JaHTUxB6zL:0 = @param:JaHTUxB6zL:0 -> float_type, {64, 17}, {17, 1}: 0.00052318ms, 0.00448434% main:@912 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](JaHTUxB6zL:0) -> float_type, {64, 4}, {17, 1}: 0.00068554ms, 0.00587597% khJcIxW3zv:0 = @param:khJcIxW3zv:0 -> float_type, {64, 9}, {9, 1}: 0.00054208ms, 0.00464633% main:@914 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](khJcIxW3zv:0) -> float_type, {64, 4}, {9, 1}: 0.00073084ms, 0.00626425% LMobYAd5ru:0 = @param:LMobYAd5ru:0 -> float_type, {64, 17}, {17, 1}: 0.0006121ms, 0.0052465% main:@916 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](LMobYAd5ru:0) -> float_type, {64, 4}, {17, 1}: 0.00072012ms, 0.00617237% lQmygxs5aD:0 = @param:lQmygxs5aD:0 -> float_type, {64, 13}, {13, 1}: 0.00053468ms, 0.00458291% main:@918 = slice[axes={0, 1},starts={0, 1},ends={64, 3}](lQmygxs5aD:0) -> float_type, {64, 2}, {13, 1}: 0.00065412ms, 0.00560666% fMiQsYNnet:0 = @param:fMiQsYNnet:0 -> float_type, {64, 5}, {5, 1}: 0.00048098ms, 0.00412263% main:@920 = slice[axes={0, 1},starts={0, 1},ends={64, 3}](fMiQsYNnet:0) -> float_type, {64, 2}, {5, 1}: 0.0006421ms, 0.00550364% XPcHGKbop2:0 = @param:XPcHGKbop2:0 -> float_type, {64, 13}, {13, 1}: 0.00052018ms, 0.00445862% main:@922 = slice[axes={0, 1},starts={0, 1},ends={64, 3}](XPcHGKbop2:0) -> float_type, {64, 2}, {13, 1}: 0.00066542ms, 0.00570352% TSqGwrHeo4:0 = @param:TSqGwrHeo4:0 -> float_type, {64, 5}, {5, 1}: 0.00049154ms, 0.00421314% main:@924 = slice[axes={0, 1},starts={0, 1},ends={64, 3}](TSqGwrHeo4:0) -> float_type, {64, 2}, {5, 1}: 0.0009563ms, 0.00819674% Tej5M3O6As:0 = @param:Tej5M3O6As:0 -> float_type, {64, 17}, {17, 1}: 0.00046528ms, 0.00398806% main:@926 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](Tej5M3O6As:0) -> float_type, {64, 8}, {17, 1}: 0.00067154ms, 0.00575598% bTeX85HPmJ:0 = @param:bTeX85HPmJ:0 -> float_type, {64, 17}, {17, 1}: 0.00050586ms, 0.00433588% main:@928 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](bTeX85HPmJ:0) -> float_type, {64, 4}, {17, 1}: 0.00064936ms, 0.00556586% agDiJj8lLK:0 = @param:agDiJj8lLK:0 -> float_type, {64, 13}, {13, 1}: 0.00045774ms, 0.00392343% main:@930 = slice[axes={0, 1},starts={0, 1},ends={64, 3}](agDiJj8lLK:0) -> float_type, {64, 2}, {13, 1}: 0.00065108ms, 0.00558061% fZhaL2tCm6:0 = @param:fZhaL2tCm6:0 -> float_type, {64, 17}, {17, 1}: 0.00054944ms, 0.00470942% main:@932 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](fZhaL2tCm6:0) -> float_type, {64, 8}, {17, 1}: 0.000632ms, 0.00541707% cN2JSTnFOu:0 = @param:cN2JSTnFOu:0 -> float_type, {64, 9}, {9, 1}: 0.00048918ms, 0.00419291% main:@934 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](cN2JSTnFOu:0) -> float_type, {64, 4}, {9, 1}: 0.00065814ms, 0.00564112% omr2g4XaH5:0 = @param:omr2g4XaH5:0 -> float_type, {64, 17}, {17, 1}: 0.00052878ms, 0.00453234% main:@936 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](omr2g4XaH5:0) -> float_type, {64, 8}, {17, 1}: 0.0006869ms, 0.00588763% iCbkMrmXBh:0 = @param:iCbkMrmXBh:0 -> float_type, {64, 9}, {9, 1}: 0.00048722ms, 0.00417611% main:@938 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](iCbkMrmXBh:0) -> float_type, {64, 4}, {9, 1}: 0.0006383ms, 0.00547107% d8RoEzG52p:0 = @param:d8RoEzG52p:0 -> float_type, {64, 17}, {17, 1}: 0.00047262ms, 0.00405097% main:@940 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](d8RoEzG52p:0) -> float_type, {64, 8}, {17, 1}: 0.000938ms, 0.00803989% eQquakfWBY:0 = @param:eQquakfWBY:0 -> float_type, {64, 17}, {17, 1}: 0.00050258ms, 0.00430777% main:@942 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](eQquakfWBY:0) -> float_type, {64, 4}, {17, 1}: 0.00063746ms, 0.00546387% qzd23aNGIT:0 = @param:qzd23aNGIT:0 -> float_type, {64, 17}, {17, 1}: 0.00044548ms, 0.00381835% main:@944 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](qzd23aNGIT:0) -> float_type, {64, 4}, {17, 1}: 0.0006511ms, 0.00558078% Vbv2Om3ptd:0 = @param:Vbv2Om3ptd:0 -> float_type, {64, 17}, {17, 1}: 0.00045432ms, 0.00389412% main:@946 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](Vbv2Om3ptd:0) -> float_type, {64, 8}, {17, 1}: 0.0006632ms, 0.00568449% r1WGP7Xe9H:0 = @param:r1WGP7Xe9H:0 -> float_type, {64, 13}, {13, 1}: 0.00045448ms, 0.00389549% main:@948 = slice[axes={0, 1},starts={0, 1},ends={64, 3}](r1WGP7Xe9H:0) -> float_type, {64, 2}, {13, 1}: 0.00065978ms, 0.00565518% aqMQBgb3hf:0 = @param:aqMQBgb3hf:0 -> float_type, {64, 17}, {17, 1}: 0.0004753ms, 0.00407394% main:@950 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](aqMQBgb3hf:0) -> float_type, {64, 8}, {17, 1}: 0.00065068ms, 0.00557718% main:@951 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=269,local=256,](main:@884,main:@892,main:@904,main:@918,main:@924,main:@948,main:@930,main:@922,main:@886,main:@920,main:@894,main:@902,main:@906,main:@880,main:@934,main:@890,main:@876,main:@916,main:@914,main:@898,main:@928,main:@896,main:@942,main:@908,main:@882,main:@944,main:@938,main:@912,main:@878,main:@910,main:@932,main:@940,main:@926,main:@936,main:@888,main:@950,main:@946,main:@900,main:@874) -> float_type, {64, 160}, {160, 1}: 0.047237ms, 0.404883% main:@952 = load[offset=2359552,end=2400512](main:@1) -> float_type, {64, 160}, {160, 1}: 0.00068542ms, 0.00587495% main:@953 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@951,main:@873,main:@952) -> float_type, {64, 160}, {160, 1}: 0.0211933ms, 0.181655% main:@954 = load[offset=2273536,end=2359552](main:@1) -> float_type, {64, 21, 16}, {336, 16, 1}: 0.0006558ms, 0.00562106% 2RDXZrx9GA:0 = @param:2RDXZrx9GA:0 -> float_type, {64, 45}, {45, 1}: 0.00055432ms, 0.00475125% main:@956 = slice[axes={0, 1},starts={0, 1},ends={64, 17}](2RDXZrx9GA:0) -> float_type, {64, 16}, {45, 1}: 0.00068706ms, 0.005889% main:@957 = unsqueeze[axes={1},steps={}](main:@956) -> float_type, {64, 1, 16}, {45, 16, 1}: 0.00075262ms, 0.00645094% duyRoIEkH8:0 = @param:duyRoIEkH8:0 -> float_type, {64, 45}, {45, 1}: 0.0005176ms, 0.00443651% main:@959 = slice[axes={0, 1},starts={0, 1},ends={64, 17}](duyRoIEkH8:0) -> float_type, {64, 16}, {45, 1}: 0.00083802ms, 0.00718293% main:@960 = unsqueeze[axes={1},steps={}](main:@959) -> float_type, {64, 1, 16}, {45, 16, 1}: 0.00058572ms, 0.00502039% 2srtUfGvja:0 = @param:2srtUfGvja:0 -> float_type, {64, 41}, {41, 1}: 0.00055086ms, 0.00472159% main:@962 = slice[axes={0, 1},starts={0, 1},ends={64, 17}](2srtUfGvja:0) -> float_type, {64, 16}, {41, 1}: 0.00064424ms, 0.00552198% main:@963 = unsqueeze[axes={1},steps={}](main:@962) -> float_type, {64, 1, 16}, {41, 16, 1}: 0.00056548ms, 0.0048469% 3Laj620MSn:0 = @param:3Laj620MSn:0 -> float_type, {64, 45}, {45, 1}: 0.00053402ms, 0.00457725% main:@965 = slice[axes={0, 1},starts={0, 1},ends={64, 17}](3Laj620MSn:0) -> float_type, {64, 16}, {45, 1}: 0.00078466ms, 0.00672556% main:@966 = unsqueeze[axes={1},steps={}](main:@965) -> float_type, {64, 1, 16}, {45, 16, 1}: 0.00073582ms, 0.00630694% CHAYRkBvIo:0 = @param:CHAYRkBvIo:0 -> float_type, {64, 41}, {41, 1}: 0.00048846ms, 0.00418674% main:@968 = slice[axes={0, 1},starts={0, 1},ends={64, 17}](CHAYRkBvIo:0) -> float_type, {64, 16}, {41, 1}: 0.00066998ms, 0.0057426% main:@969 = unsqueeze[axes={1},steps={}](main:@968) -> float_type, {64, 1, 16}, {41, 16, 1}: 0.0005796ms, 0.00496793% jznORDJr0N:0 = @param:jznORDJr0N:0 -> float_type, {64, 45}, {45, 1}: 0.00050352ms, 0.00431582% main:@971 = slice[axes={0, 1},starts={0, 1},ends={64, 17}](jznORDJr0N:0) -> float_type, {64, 16}, {45, 1}: 0.00078032ms, 0.00668836% main:@972 = unsqueeze[axes={1},steps={}](main:@971) -> float_type, {64, 1, 16}, {45, 16, 1}: 0.00055126ms, 0.00472502% HRGzlnEkXQ:0 = @param:HRGzlnEkXQ:0 -> float_type, {64, 41}, {41, 1}: 0.00049392ms, 0.00423354% main:@974 = slice[axes={0, 1},starts={0, 1},ends={64, 17}](HRGzlnEkXQ:0) -> float_type, {64, 16}, {41, 1}: 0.00066152ms, 0.00567009% main:@975 = unsqueeze[axes={1},steps={}](main:@974) -> float_type, {64, 1, 16}, {41, 16, 1}: 0.00094566ms, 0.00810554% HzcyUlm8Yj:0 = @param:HzcyUlm8Yj:0 -> float_type, {64, 45}, {45, 1}: 0.00050588ms, 0.00433605% main:@977 = slice[axes={0, 1},starts={0, 1},ends={64, 17}](HzcyUlm8Yj:0) -> float_type, {64, 16}, {45, 1}: 0.00080658ms, 0.00691344% main:@978 = unsqueeze[axes={1},steps={}](main:@977) -> float_type, {64, 1, 16}, {45, 16, 1}: 0.00057324ms, 0.00491342% main:@979 = slice[axes={0, 1},starts={0, 289},ends={64, 305}](lyiDxfzZg3:0) -> float_type, {64, 16}, {701, 1}: 0.00059932ms, 0.00513696% main:@980 = unsqueeze[axes={1},steps={}](main:@979) -> float_type, {64, 1, 16}, {701, 16, 1}: 0.00090216ms, 0.00773269% VUqfdN5Fzv:0 = @param:VUqfdN5Fzv:0 -> float_type, {64, 45}, {45, 1}: 0.00050786ms, 0.00435302% main:@982 = slice[axes={0, 1},starts={0, 1},ends={64, 17}](VUqfdN5Fzv:0) -> float_type, {64, 16}, {45, 1}: 0.00074062ms, 0.00634808% main:@983 = unsqueeze[axes={1},steps={}](main:@982) -> float_type, {64, 1, 16}, {45, 16, 1}: 0.00058324ms, 0.00499913% eQI8ikEMgP:0 = @param:eQI8ikEMgP:0 -> float_type, {64, 41}, {41, 1}: 0.00049702ms, 0.00426011% main:@985 = slice[axes={0, 1},starts={0, 1},ends={64, 17}](eQI8ikEMgP:0) -> float_type, {64, 16}, {41, 1}: 0.00063174ms, 0.00541484% main:@986 = unsqueeze[axes={1},steps={}](main:@985) -> float_type, {64, 1, 16}, {41, 16, 1}: 0.00056846ms, 0.00487245% main:@987 = reshape_lazy[dims={-1, 10, 16}](main:@953) -> float_type, {64, 10, 16}, {160, 16, 1}: 0.0004766ms, 0.00408508% main:@988 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=1792,local=256,](main:@972,main:@975,main:@980,main:@957,main:@983,main:@969,main:@978,main:@963,main:@966,main:@960,main:@986,main:@987,main:@954) -> float_type, {64, 21, 16}, {336, 16, 1}: 0.0271338ms, 0.232572% main:@989 = load[offset=229376,end=256256](main:@1) -> float_type, {64, 21, 5}, {105, 5, 1}: 0.00066524ms, 0.00570198% main:@990 = reshape_lazy[dims={64, 16, 5}](main:@872) -> float_type, {64, 16, 5}, {80, 5, 1}: 0.00052796ms, 0.00452531% main:@991 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@988,main:@990,main:@989) -> float_type, {64, 21, 5}, {105, 5, 1}: 0.020533ms, 0.175995% XRvxjULC7M:0 = @param:XRvxjULC7M:0 -> float_type, {64, 60, 5}, {300, 5, 1}: 0.00071934ms, 0.00616568% main:@993 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={64, 60, 5}](XRvxjULC7M:0) -> float_type, {64, 60, 4}, {300, 5, 1}: 0.00073288ms, 0.00628174% main:@994 = load[offset=13885952,end=14930432](main:@1) -> float_type, {64, 60, 68}, {4080, 68, 1}: 0.00049408ms, 0.00423491% main:@995 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=65280,local=256,](main:@136,main:@214,main:@993,main:@294,main:@994) -> float_type, {64, 60, 68}, {4080, 68, 1}: 0.0214318ms, 0.183699% main:@996 = hip::hip_copy_literal[id=main:@literal:152] -> float_type, {68, 64}, {64, 1}: 0.00079378ms, 0.00680373% main:@997 = load[offset=3393024,end=4376064](main:@1) -> float_type, {3840, 64}, {64, 1}: 0.00050006ms, 0.00428617% main:@998 = reshape_lazy[dims={3840, 68}](main:@995) -> float_type, {3840, 68}, {68, 1}: 0.00055164ms, 0.00472828% main:@999 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@998,main:@996,main:@997) -> float_type, {3840, 64}, {64, 1}: 0.0238268ms, 0.204227% main:@1000 = slice[axes={0, 1},starts={0, 45},ends={64, 53}](2CnSyUBzm0:0) -> float_type, {64, 8}, {65, 1}: 0.000831ms, 0.00712276% main:@1001 = slice[axes={0, 1},starts={0, 13},ends={64, 21}](9YqhflbMKe:0) -> float_type, {64, 8}, {29, 1}: 0.00066296ms, 0.00568243% main:@1002 = slice[axes={0, 1},starts={0, 45},ends={64, 53}](Qp3K1PHvte:0) -> float_type, {64, 8}, {69, 1}: 0.00079548ms, 0.0068183% main:@1003 = slice[axes={0, 1},starts={0, 189},ends={64, 197}](nv2tzWRf1I:0) -> float_type, {64, 8}, {301, 1}: 0.00062602ms, 0.00536581% main:@1004 = slice[axes={0, 1},starts={0, 209},ends={64, 217}](uFSOc5Chp2:0) -> float_type, {64, 8}, {481, 1}: 0.00062302ms, 0.0053401% main:@1005 = slice[axes={0, 1},starts={0, 49},ends={64, 57}](ifP6Ceko5T:0) -> float_type, {64, 8}, {73, 1}: 0.00062892ms, 0.00539067% main:@1006 = load[offset=1990912,end=2003200](main:@1) -> float_type, {64, 48}, {48, 1}: 0.00052392ms, 0.00449068% main:@1007 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=512,local=256,](main:@1003,main:@1000,main:@1004,main:@1005,main:@1002,main:@1001,main:@1006) -> float_type, {64, 48}, {48, 1}: 0.015904ms, 0.136318% main:@1008 = hip::hip_copy_literal[id=main:@literal:160] -> float_type, {48, 64}, {64, 1}: 0.00077692ms, 0.00665922% main:@1009 = load[offset=2253056,end=2269440](main:@1) -> float_type, {64, 64}, {64, 1}: 0.00048328ms, 0.00414234% main:@1010 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1007,main:@1008,main:@1009) -> float_type, {64, 64}, {64, 1}: 0.0198232ms, 0.169911% main:@1011 = reshape_lazy[dims={64, 60, 64}](main:@999) -> float_type, {64, 60, 64}, {3840, 64, 1}: 0.00070988ms, 0.0060846% main:@1012 = transpose[permutation={0, 2, 1}](main:@1011) -> float_type, {64, 64, 60}, {3840, 1, 64}: 0.00056128ms, 0.0048109% main:@1013 = load[offset=1990912,end=2006272](main:@1) -> float_type, {64, 1, 60}, {60, 60, 1}: 0.00049934ms, 0.00428% main:@1014 = reshape_lazy[dims={64, 1, 64}](main:@1010) -> float_type, {64, 1, 64}, {64, 64, 1}: 0.00048154ms, 0.00412743% main:@1015 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1014,main:@1012,main:@1013) -> float_type, {64, 1, 60}, {60, 60, 1}: 0.023205ms, 0.198898% main:@1016 = load[offset=13885952,end=15176192](main:@1) -> float_type, {64, 60, 84}, {5040, 84, 1}: 0.0006788ms, 0.0058182% 4luvCw32jf:0 = @param:4luvCw32jf:0 -> float_type, {64, 60, 5}, {300, 5, 1}: 0.00051782ms, 0.00443839% main:@1018 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={64, 60, 5}](4luvCw32jf:0) -> float_type, {64, 60, 4}, {300, 5, 1}: 0.00070224ms, 0.00601911% F5awZuD1c2:0 = @param:F5awZuD1c2:0 -> float_type, {64, 60, 9}, {540, 9, 1}: 0.00052946ms, 0.00453816% main:@1020 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={64, 60, 9}](F5awZuD1c2:0) -> float_type, {64, 60, 8}, {540, 9, 1}: 0.00078964ms, 0.00676825% PnxiROp9zw:0 = @param:PnxiROp9zw:0 -> float_type, {64, 60, 5}, {300, 5, 1}: 0.00050318ms, 0.00431291% main:@1022 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={64, 60, 5}](PnxiROp9zw:0) -> float_type, {64, 60, 4}, {300, 5, 1}: 0.00066254ms, 0.00567883% main:@1023 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=46080,local=256,](main:@136,main:@214,main:@993,main:@294,main:@1022,main:@1020,main:@1018,main:@1016) -> float_type, {64, 60, 84}, {5040, 84, 1}: 0.0231203ms, 0.198171% main:@1024 = hip::hip_copy_literal[id=main:@literal:134] -> float_type, {84, 64}, {64, 1}: 0.00078046ms, 0.00668956% main:@1025 = reshape_lazy[dims={3840, 84}](main:@1023) -> float_type, {3840, 84}, {84, 1}: 0.00053326ms, 0.00457074% main:@1026 = load[offset=3393024,end=4376064](main:@1) -> float_type, {3840, 64}, {64, 1}: 0.0004811ms, 0.00412366% main:@1027 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1025,main:@1024,main:@1026) -> float_type, {3840, 64}, {64, 1}: 0.0227877ms, 0.19532% mt0cODHoal:0 = @param:mt0cODHoal:0 -> int32_type, {64, 60}, {60, 1}: 0.00068182ms, 0.00584409% main:@1029 = unsqueeze[axes={1},steps={}](mt0cODHoal:0) -> int32_type, {64, 1, 60}, {60, 60, 1}: 0.0005965ms, 0.00511278% main:@1030 = load[offset=2268416,end=2283776](main:@1) -> float_type, {64, 1, 60}, {60, 60, 1}: 0.00048766ms, 0.00417988% main:@1031 = gpu::code_object[code_object=13544,symbol_name=convert_add_add_sub_kernel,global=1920,local=256,](main:@1029,main:@1015,main:@1030) -> float_type, {64, 1, 60}, {60, 60, 1}: 0.0135793ms, 0.116392% main:@1032 = load[offset=2253056,end=2268416](main:@1) -> float_type, {64, 1, 60}, {60, 60, 1}: 0.00066572ms, 0.00570609% main:@1033 = gpu::code_object[code_object=14552,symbol_name=softmax_kernel,global=4096,local=64,](main:@1031,main:@1032) -> float_type, {64, 1, 60}, {60, 60, 1}: 0.0158923ms, 0.136218% main:@1034 = load[offset=1990912,end=2007296](main:@1) -> float_type, {64, 1, 64}, {64, 64, 1}: 0.00064ms, 0.00548564% main:@1035 = reshape_lazy[dims={64, 60, 64}](main:@1027) -> float_type, {64, 60, 64}, {3840, 64, 1}: 0.00052808ms, 0.00452634% main:@1036 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1033,main:@1035,main:@1034) -> float_type, {64, 1, 64}, {64, 64, 1}: 0.0223489ms, 0.191559% main:@1037 = slice[axes={1},starts={0},ends={256}](main:@624) -> float_type, {960, 256}, {768, 1}: 0.00083122ms, 0.00712464% main:@1038 = reshape_lazy[dims={64, 15, 256}](main:@1037) -> float_type, {64, 15, 256}, {11520, 768, 1}: 0.00065946ms, 0.00565243% main:@1039 = slice[axes={2},starts={0},ends={64}](main:@1038) -> float_type, {64, 15, 64}, {11520, 768, 1}: 0.00079626ms, 0.00682499% main:@1040 = unsqueeze[axes={0},steps={}](main:@1039) -> float_type, {1, 64, 15, 64}, {737280, 11520, 768, 1}: 0.00066996ms, 0.00574243% main:@1041 = slice[axes={2},starts={64},ends={128}](main:@1038) -> float_type, {64, 15, 64}, {11520, 768, 1}: 0.00067562ms, 0.00579095% main:@1042 = unsqueeze[axes={0},steps={}](main:@1041) -> float_type, {1, 64, 15, 64}, {737280, 11520, 768, 1}: 0.00099304ms, 0.00851165% main:@1043 = slice[axes={2},starts={128},ends={192}](main:@1038) -> float_type, {64, 15, 64}, {11520, 768, 1}: 0.00063472ms, 0.00544038% main:@1044 = unsqueeze[axes={0},steps={}](main:@1043) -> float_type, {1, 64, 15, 64}, {737280, 11520, 768, 1}: 0.0006397ms, 0.00548307% main:@1045 = slice[axes={2},starts={192},ends={256}](main:@1038) -> float_type, {64, 15, 64}, {11520, 768, 1}: 0.00101756ms, 0.00872182% main:@1046 = unsqueeze[axes={0},steps={}](main:@1045) -> float_type, {1, 64, 15, 64}, {737280, 11520, 768, 1}: 0.00059608ms, 0.00510918% main:@1047 = load[offset=3393024,end=4376064](main:@1) -> float_type, {4, 64, 15, 64}, {61440, 960, 64, 1}: 0.00051824ms, 0.00444199% main:@1048 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=30720,local=256,](main:@1040,main:@1042,main:@1044,main:@1046,main:@1047) -> float_type, {4, 64, 15, 64}, {61440, 960, 64, 1}: 0.0237986ms, 0.203985% main:@1049 = slice[axes={1},starts={512},ends={768}](main:@624) -> float_type, {960, 256}, {768, 1}: 0.00078236ms, 0.00670585% main:@1050 = reshape_lazy[dims={64, 15, 256}](main:@1049) -> float_type, {64, 15, 256}, {11520, 768, 1}: 0.00056638ms, 0.00485462% main:@1051 = slice[axes={2},starts={0},ends={64}](main:@1050) -> float_type, {64, 15, 64}, {11520, 768, 1}: 0.00078022ms, 0.00668751% main:@1052 = unsqueeze[axes={0},steps={}](main:@1051) -> float_type, {1, 64, 15, 64}, {737280, 11520, 768, 1}: 0.00091392ms, 0.00783349% main:@1053 = slice[axes={2},starts={64},ends={128}](main:@1050) -> float_type, {64, 15, 64}, {11520, 768, 1}: 0.0006069ms, 0.00520193% main:@1054 = unsqueeze[axes={0},steps={}](main:@1053) -> float_type, {1, 64, 15, 64}, {737280, 11520, 768, 1}: 0.00062622ms, 0.00536752% main:@1055 = slice[axes={2},starts={128},ends={192}](main:@1050) -> float_type, {64, 15, 64}, {11520, 768, 1}: 0.00097234ms, 0.00833422% main:@1056 = unsqueeze[axes={0},steps={}](main:@1055) -> float_type, {1, 64, 15, 64}, {737280, 11520, 768, 1}: 0.00059794ms, 0.00512513% main:@1057 = slice[axes={2},starts={192},ends={256}](main:@1050) -> float_type, {64, 15, 64}, {11520, 768, 1}: 0.00072278ms, 0.00619517% main:@1058 = unsqueeze[axes={0},steps={}](main:@1057) -> float_type, {1, 64, 15, 64}, {737280, 11520, 768, 1}: 0.0005877ms, 0.00503736% main:@1059 = load[offset=5038592,end=6021632](main:@1) -> float_type, {4, 64, 15, 64}, {61440, 960, 64, 1}: 0.00049802ms, 0.00426868% main:@1060 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=30720,local=256,](main:@1052,main:@1054,main:@1056,main:@1058,main:@1059) -> float_type, {4, 64, 15, 64}, {61440, 960, 64, 1}: 0.0238092ms, 0.204076% main:@1061 = transpose[permutation={0, 1, 3, 2}](main:@1060) -> float_type, {4, 64, 64, 15}, {61440, 960, 1, 64}: 0.00068068ms, 0.00583432% main:@1062 = load[offset=2253056,end=2483456](main:@1) -> float_type, {4, 64, 15, 15}, {14400, 225, 15, 1}: 0.00052064ms, 0.00446257% main:@1063 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1048,main:@1061,main:@1062) -> float_type, {4, 64, 15, 15}, {14400, 225, 15, 1}: 0.0207631ms, 0.177967% main:@1064 = load[offset=2483456,end=2713856](main:@1) -> float_type, {4, 64, 15, 15}, {14400, 225, 15, 1}: 0.00064318ms, 0.00551289% main:@1065 = gpu::code_object[code_object=13512,symbol_name=pow_kernel,global=28800,local=256,](main:@1063,main:@1064) -> float_type, {4, 64, 15, 15}, {14400, 225, 15, 1}: 0.0187962ms, 0.161108% main:@1066 = load[offset=2253056,end=2483456](main:@1) -> float_type, {4, 64, 15, 15}, {14400, 225, 15, 1}: 0.00065152ms, 0.00558438% main:@1067 = gpu::code_object[code_object=14576,symbol_name=softmax_kernel,global=245760,local=64,](main:@1065,main:@1066) -> float_type, {4, 64, 15, 15}, {14400, 225, 15, 1}: 0.0261918ms, 0.224498% main:@1068 = slice[axes={1},starts={256},ends={512}](main:@624) -> float_type, {960, 256}, {768, 1}: 0.0009303ms, 0.00797389% main:@1069 = reshape_lazy[dims={64, 15, 256}](main:@1068) -> float_type, {64, 15, 256}, {11520, 768, 1}: 0.00058092ms, 0.00497924% main:@1070 = slice[axes={2},starts={0},ends={64}](main:@1069) -> float_type, {64, 15, 64}, {11520, 768, 1}: 0.00100012ms, 0.00857234% main:@1071 = unsqueeze[axes={0},steps={}](main:@1070) -> float_type, {1, 64, 15, 64}, {737280, 11520, 768, 1}: 0.0006072ms, 0.0052045% main:@1072 = slice[axes={2},starts={64},ends={128}](main:@1069) -> float_type, {64, 15, 64}, {11520, 768, 1}: 0.00097152ms, 0.0083272% main:@1073 = unsqueeze[axes={0},steps={}](main:@1072) -> float_type, {1, 64, 15, 64}, {737280, 11520, 768, 1}: 0.00060036ms, 0.00514587% main:@1074 = slice[axes={2},starts={128},ends={192}](main:@1069) -> float_type, {64, 15, 64}, {11520, 768, 1}: 0.00074174ms, 0.00635768% main:@1075 = unsqueeze[axes={0},steps={}](main:@1074) -> float_type, {1, 64, 15, 64}, {737280, 11520, 768, 1}: 0.00091896ms, 0.00787669% main:@1076 = slice[axes={2},starts={192},ends={256}](main:@1069) -> float_type, {64, 15, 64}, {11520, 768, 1}: 0.00065848ms, 0.00564403% main:@1077 = unsqueeze[axes={0},steps={}](main:@1076) -> float_type, {1, 64, 15, 64}, {737280, 11520, 768, 1}: 0.0009127ms, 0.00782303% main:@1078 = load[offset=3393024,end=4376064](main:@1) -> float_type, {4, 64, 15, 64}, {61440, 960, 64, 1}: 0.00049242ms, 0.00422068% main:@1079 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=30720,local=256,](main:@1071,main:@1073,main:@1075,main:@1077,main:@1078) -> float_type, {4, 64, 15, 64}, {61440, 960, 64, 1}: 0.0239084ms, 0.204926% main:@1080 = load[offset=5038592,end=6021632](main:@1) -> float_type, {4, 64, 15, 64}, {61440, 960, 64, 1}: 0.00068514ms, 0.00587255% main:@1081 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1067,main:@1079,main:@1080) -> float_type, {4, 64, 15, 64}, {61440, 960, 64, 1}: 0.0207402ms, 0.177771% main:@1082 = slice[axes={1},starts={32},ends={96}](main:@652) -> float_type, {64, 64}, {128, 1}: 0.00078476ms, 0.00672642% main:@1083 = load[offset=2265856,end=2282240](main:@1) -> float_type, {64, 64}, {64, 1}: 0.00048234ms, 0.00413428% main:@1084 = gpu::code_object[code_object=13528,symbol_name=contiguous_kernel,global=2048,local=256,](main:@1082,main:@1083) -> float_type, {64, 64}, {64, 1}: 0.0133235ms, 0.1142% main:@1085 = slice[axes={1},starts={64},ends={128}](main:@734) -> float_type, {3200, 64}, {128, 1}: 0.00076982ms, 0.00659836% main:@1086 = reshape_lazy[dims={1, 64, 50, 64}](main:@1085) -> float_type, {1, 64, 50, 64}, {409600, 6400, 128, 1}: 0.00060806ms, 0.00521187% main:@1087 = transpose[permutation={0, 1, 3, 2}](main:@1086) -> float_type, {1, 64, 64, 50}, {409600, 6400, 1, 128}: 0.00059014ms, 0.00505827% main:@1088 = reshape_lazy[dims={1, 64, 1, 64}](main:@1084) -> float_type, {1, 64, 1, 64}, {4096, 64, 64, 1}: 0.0005036ms, 0.00431651% main:@1089 = load[offset=2253056,end=2265856](main:@1) -> float_type, {1, 64, 1, 50}, {3200, 50, 50, 1}: 0.00048488ms, 0.00415606% main:@1090 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1088,main:@1087,main:@1089) -> float_type, {1, 64, 1, 50}, {3200, 50, 50, 1}: 0.0243745ms, 0.208922% ObZQkfyaKG:0 = @param:ObZQkfyaKG:0 -> int32_type, {64, 50}, {50, 1}: 0.00068784ms, 0.00589569% main:@1092 = unsqueeze[axes={1},steps={}](ObZQkfyaKG:0) -> int32_type, {64, 1, 50}, {50, 50, 1}: 0.00060778ms, 0.00520947% main:@1093 = load[offset=2265856,end=2278656](main:@1) -> float_type, {64, 1, 50}, {50, 50, 1}: 0.00050188ms, 0.00430177% main:@1094 = gpu::code_object[code_object=13528,symbol_name=convert_add_kernel,global=1600,local=256,](main:@1092,main:@1093) -> float_type, {64, 1, 50}, {50, 50, 1}: 0.0133808ms, 0.114691% main:@1095 = multibroadcast[out_lens={1, 64, 1, 50}](main:@1094) -> float_type, {1, 64, 1, 50}, {0, 50, 50, 1}: 0.0007428ms, 0.00636677% main:@1096 = load[offset=2282240,end=2295040](main:@1) -> float_type, {1, 64, 1, 50}, {3200, 50, 50, 1}: 0.0005ms, 0.00428565% main:@1097 = gpu::code_object[code_object=13520,symbol_name=pow_pow_kernel,global=1600,local=256,](main:@1090,main:@1095,main:@1096) -> float_type, {1, 64, 1, 50}, {3200, 50, 50, 1}: 0.0179935ms, 0.154228% main:@1098 = load[offset=2269440,end=2282240](main:@1) -> float_type, {1, 64, 1, 50}, {3200, 50, 50, 1}: 0.00064498ms, 0.00552832% main:@1099 = gpu::code_object[code_object=14584,symbol_name=softmax_kernel,global=4096,local=64,](main:@1097,main:@1098) -> float_type, {1, 64, 1, 50}, {3200, 50, 50, 1}: 0.0159223ms, 0.136475% main:@1100 = slice[axes={1},starts={0},ends={64}](main:@734) -> float_type, {3200, 64}, {128, 1}: 0.00106518ms, 0.00912998% main:@1101 = load[offset=2253056,end=2269440](main:@1) -> float_type, {1, 64, 1, 64}, {4096, 64, 64, 1}: 0.00048372ms, 0.00414611% main:@1102 = reshape_lazy[dims={1, 64, 50, 64}](main:@1100) -> float_type, {1, 64, 50, 64}, {409600, 6400, 128, 1}: 0.00059534ms, 0.00510284% main:@1103 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1099,main:@1102,main:@1101) -> float_type, {1, 64, 1, 64}, {4096, 64, 64, 1}: 0.0218744ms, 0.187492% main:@1104 = hip::hip_copy_literal[id=main:@literal:24] -> float_type, {64, 64}, {64, 1}: 0.00083558ms, 0.00716201% main:@1105 = load[offset=2269440,end=2433280](main:@1) -> float_type, {64, 10, 64}, {640, 64, 1}: 0.00050654ms, 0.00434171% main:@1106 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=10240,local=256,](main:@39,main:@235,main:@78,main:@19,main:@1105) -> float_type, {64, 10, 64}, {640, 64, 1}: 0.0171721ms, 0.147187% main:@1107 = reshape_lazy[dims={640, 64}](main:@1106) -> float_type, {640, 64}, {64, 1}: 0.00068356ms, 0.005859% main:@1108 = load[offset=13714944,end=13878784](main:@1) -> float_type, {640, 64}, {64, 1}: 0.00051048ms, 0.00437548% main:@1109 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1107,main:@1104,main:@1108) -> float_type, {640, 64}, {64, 1}: 0.0195428ms, 0.167507% main:@1110 = hip::hip_copy_literal[id=main:@literal:116] -> float_type, {128, 3429}, {3429, 1}: 0.00075274ms, 0.00645197% main:@1111 = slice[axes={1},starts={128},ends={256}](main:@717) -> float_type, {64, 128}, {896, 1}: 0.0006426ms, 0.00550792% main:@1112 = load[offset=2269440,end=3147264](main:@1) -> float_type, {64, 3429}, {3429, 1}: 0.00047942ms, 0.00410926% main:@1113 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1111,main:@1110,main:@1112) -> float_type, {64, 3429}, {3429, 1}: 0.0279896ms, 0.239907% main:@1114 = hip::hip_copy_literal[id=main:@literal:21] -> float_type, {84, 256}, {256, 1}: 0.00078748ms, 0.00674973% main:@1115 = load[offset=3393024,end=4468224](main:@1) -> float_type, {64, 50, 84}, {4200, 84, 1}: 0.00052476ms, 0.00449788% pHflUqMEaX:0 = @param:pHflUqMEaX:0 -> float_type, {64, 50, 5}, {250, 5, 1}: 0.00056528ms, 0.00484519% main:@1117 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={64, 50, 5}](pHflUqMEaX:0) -> float_type, {64, 50, 4}, {250, 5, 1}: 0.00075864ms, 0.00650254% lDIuWZ8VFU:0 = @param:lDIuWZ8VFU:0 -> float_type, {64, 50, 5}, {250, 5, 1}: 0.00051692ms, 0.00443068% main:@1119 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={64, 50, 5}](lDIuWZ8VFU:0) -> float_type, {64, 50, 4}, {250, 5, 1}: 0.00074806ms, 0.00641185% waXyDlz530:0 = @param:waXyDlz530:0 -> float_type, {64, 50, 5}, {250, 5, 1}: 0.00050234ms, 0.00430571% main:@1121 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={64, 50, 5}](waXyDlz530:0) -> float_type, {64, 50, 4}, {250, 5, 1}: 0.00063328ms, 0.00542804% main:@1122 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=44800,local=256,](main:@1117,main:@58,main:@274,main:@1121,main:@1119,main:@194,main:@1115) -> float_type, {64, 50, 84}, {4200, 84, 1}: 0.022748ms, 0.19498% main:@1123 = load[offset=10044928,end=13321728](main:@1) -> float_type, {3200, 256}, {256, 1}: 0.00066154ms, 0.00567026% main:@1124 = reshape_lazy[dims={3200, 84}](main:@1122) -> float_type, {3200, 84}, {84, 1}: 0.00054244ms, 0.00464942% main:@1125 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1124,main:@1114,main:@1123) -> float_type, {3200, 256}, {256, 1}: 0.02999ms, 0.257053% main:@1126 = hip::hip_copy_literal[id=main:@literal:49] -> float_type, {256, 512}, {512, 1}: 0.00079422ms, 0.0068075% main:@1127 = slice[axes={0, 1, 2},starts={0, 9, 0},ends={64, 10, 256}](main:@713) -> float_type, {64, 1, 256}, {4608, 256, 1}: 0.0006862ms, 0.00588163% main:@1128 = squeeze[axes={1}](main:@1127) -> float_type, {64, 256}, {4608, 1}: 0.00089064ms, 0.00763395% main:@1129 = load[offset=3524096,end=3655168](main:@1) -> float_type, {64, 512}, {512, 1}: 0.00050018ms, 0.0042872% main:@1130 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1128,main:@1126,main:@1129) -> float_type, {64, 512}, {512, 1}: 0.027158ms, 0.232779% main:@1131 = load[offset=3393024,end=3524096](main:@1) -> float_type, {64, 512}, {512, 1}: 0.00065294ms, 0.00559655% main:@1132 = gpu::code_object[code_object=13512,symbol_name=relu_kernel,global=16384,local=256,](main:@1130,main:@1131) -> float_type, {64, 512}, {512, 1}: 0.0158583ms, 0.135927% main:@1133 = load[offset=9298432,end=9363968](main:@1) -> float_type, {64, 256}, {256, 1}: 0.0006154ms, 0.00527478% main:@1134 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1132,main:@719,main:@1133) -> float_type, {64, 256}, {256, 1}: 0.0314894ms, 0.269905% main:@1135 = load[offset=3393024,end=3524096](main:@1) -> float_type, {64, 512}, {512, 1}: 0.0006519ms, 0.00558763% main:@1136 = slice[axes={0, 1, 2},starts={0, 16, 0},ends={64, 17, 256}](main:@713) -> float_type, {64, 1, 256}, {4608, 256, 1}: 0.00067114ms, 0.00575255% main:@1137 = squeeze[axes={1}](main:@1136) -> float_type, {64, 256}, {4608, 1}: 0.00094256ms, 0.00807897% main:@1138 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1137,main:@718,main:@1135) -> float_type, {64, 512}, {512, 1}: 0.0240525ms, 0.206161% main:@1139 = load[offset=3524096,end=3655168](main:@1) -> float_type, {64, 512}, {512, 1}: 0.0006511ms, 0.00558078% main:@1140 = gpu::code_object[code_object=13512,symbol_name=relu_kernel,global=16384,local=256,](main:@1138,main:@1139) -> float_type, {64, 512}, {512, 1}: 0.0156193ms, 0.133878% main:@1141 = hip::hip_copy_literal[id=main:@literal:84] -> float_type, {512, 256}, {256, 1}: 0.00082252ms, 0.00705007% main:@1142 = load[offset=9363968,end=9429504](main:@1) -> float_type, {64, 256}, {256, 1}: 0.00049966ms, 0.00428274% main:@1143 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1140,main:@1141,main:@1142) -> float_type, {64, 256}, {256, 1}: 0.0303715ms, 0.260323% main:@1144 = hip::hip_copy_literal[id=main:@literal:142] -> float_type, {256, 512}, {512, 1}: 0.0007731ms, 0.00662648% main:@1145 = slice[axes={0, 1, 2},starts={0, 1, 0},ends={64, 2, 256}](main:@713) -> float_type, {64, 1, 256}, {4608, 256, 1}: 0.00071042ms, 0.00608923% main:@1146 = squeeze[axes={1}](main:@1145) -> float_type, {64, 256}, {4608, 1}: 0.0006085ms, 0.00521564% main:@1147 = load[offset=3393024,end=3524096](main:@1) -> float_type, {64, 512}, {512, 1}: 0.00048636ms, 0.00416874% main:@1148 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1146,main:@1144,main:@1147) -> float_type, {64, 512}, {512, 1}: 0.0238625ms, 0.204533% main:@1149 = hip::hip_copy_literal[id=main:@literal:120] -> float_type, {512, 256}, {256, 1}: 0.00085722ms, 0.0073475% main:@1150 = load[offset=3524096,end=3655168](main:@1) -> float_type, {64, 512}, {512, 1}: 0.00049908ms, 0.00427777% main:@1151 = gpu::code_object[code_object=13512,symbol_name=relu_kernel,global=16384,local=256,](main:@1148,main:@1150) -> float_type, {64, 512}, {512, 1}: 0.0155912ms, 0.133637% main:@1152 = load[offset=9429504,end=9495040](main:@1) -> float_type, {64, 256}, {256, 1}: 0.00059208ms, 0.0050749% main:@1153 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1151,main:@1149,main:@1152) -> float_type, {64, 256}, {256, 1}: 0.0303006ms, 0.259716% main:@1154 = hip::hip_copy_literal[id=main:@literal:62] -> float_type, {256, 512}, {512, 1}: 0.00083028ms, 0.00711658% main:@1155 = slice[axes={0, 1, 2},starts={0, 15, 0},ends={64, 16, 256}](main:@713) -> float_type, {64, 1, 256}, {4608, 256, 1}: 0.0006212ms, 0.0053245% main:@1156 = squeeze[axes={1}](main:@1155) -> float_type, {64, 256}, {4608, 1}: 0.00061922ms, 0.00530752% main:@1157 = load[offset=3524096,end=3655168](main:@1) -> float_type, {64, 512}, {512, 1}: 0.0004634ms, 0.00397194% main:@1158 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1156,main:@1154,main:@1157) -> float_type, {64, 512}, {512, 1}: 0.023753ms, 0.203595% main:@1159 = load[offset=3393024,end=3524096](main:@1) -> float_type, {64, 512}, {512, 1}: 0.0006749ms, 0.00578478% main:@1160 = gpu::code_object[code_object=13512,symbol_name=relu_kernel,global=16384,local=256,](main:@1158,main:@1159) -> float_type, {64, 512}, {512, 1}: 0.0158114ms, 0.135525% main:@1161 = hip::hip_copy_literal[id=main:@literal:95] -> float_type, {512, 256}, {256, 1}: 0.00077814ms, 0.00666968% main:@1162 = load[offset=9495040,end=9560576](main:@1) -> float_type, {64, 256}, {256, 1}: 0.00047892ms, 0.00410497% main:@1163 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1160,main:@1161,main:@1162) -> float_type, {64, 256}, {256, 1}: 0.030034ms, 0.257431% main:@1164 = hip::hip_copy_literal[id=main:@literal:138] -> float_type, {256, 512}, {512, 1}: 0.00075966ms, 0.00651128% main:@1165 = slice[axes={0, 1, 2},starts={0, 8, 0},ends={64, 9, 256}](main:@713) -> float_type, {64, 1, 256}, {4608, 256, 1}: 0.00062172ms, 0.00532895% main:@1166 = squeeze[axes={1}](main:@1165) -> float_type, {64, 256}, {4608, 1}: 0.00060404ms, 0.00517741% main:@1167 = load[offset=3524096,end=3655168](main:@1) -> float_type, {64, 512}, {512, 1}: 0.00045808ms, 0.00392634% main:@1168 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1166,main:@1164,main:@1167) -> float_type, {64, 512}, {512, 1}: 0.0235058ms, 0.201476% main:@1169 = load[offset=3393024,end=3524096](main:@1) -> float_type, {64, 512}, {512, 1}: 0.00063926ms, 0.00547929% main:@1170 = gpu::code_object[code_object=13512,symbol_name=relu_kernel,global=16384,local=256,](main:@1168,main:@1169) -> float_type, {64, 512}, {512, 1}: 0.0159576ms, 0.136778% main:@1171 = hip::hip_copy_literal[id=main:@literal:44] -> float_type, {512, 256}, {256, 1}: 0.00079342ms, 0.00680065% main:@1172 = load[offset=9560576,end=9626112](main:@1) -> float_type, {64, 256}, {256, 1}: 0.00046038ms, 0.00394606% main:@1173 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1170,main:@1171,main:@1172) -> float_type, {64, 256}, {256, 1}: 0.0302398ms, 0.259194% main:@1174 = hip::hip_copy_literal[id=main:@literal:48] -> float_type, {256, 512}, {512, 1}: 0.00079172ms, 0.00678608% main:@1175 = slice[axes={0, 1, 2},starts={0, 5, 0},ends={64, 6, 256}](main:@713) -> float_type, {64, 1, 256}, {4608, 256, 1}: 0.00062096ms, 0.00532244% main:@1176 = squeeze[axes={1}](main:@1175) -> float_type, {64, 256}, {4608, 1}: 0.00060374ms, 0.00517484% main:@1177 = load[offset=3524096,end=3655168](main:@1) -> float_type, {64, 512}, {512, 1}: 0.00049866ms, 0.00427417% main:@1178 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1176,main:@1174,main:@1177) -> float_type, {64, 512}, {512, 1}: 0.0235303ms, 0.201686% main:@1179 = load[offset=3393024,end=3524096](main:@1) -> float_type, {64, 512}, {512, 1}: 0.00066904ms, 0.00573455% main:@1180 = gpu::code_object[code_object=13512,symbol_name=relu_kernel,global=16384,local=256,](main:@1178,main:@1179) -> float_type, {64, 512}, {512, 1}: 0.0157994ms, 0.135422% main:@1181 = load[offset=9651712,end=9717248](main:@1) -> float_type, {64, 256}, {256, 1}: 0.00066542ms, 0.00570352% main:@1182 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1180,main:@719,main:@1181) -> float_type, {64, 256}, {256, 1}: 0.0300788ms, 0.257815% main:@1183 = slice[axes={0, 1, 2},starts={0, 2, 0},ends={64, 3, 256}](main:@713) -> float_type, {64, 1, 256}, {4608, 256, 1}: 0.0007984ms, 0.00684333% main:@1184 = squeeze[axes={1}](main:@1183) -> float_type, {64, 256}, {4608, 1}: 0.00091644ms, 0.00785509% main:@1185 = load[offset=3524096,end=3655168](main:@1) -> float_type, {64, 512}, {512, 1}: 0.0004732ms, 0.00405594% main:@1186 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1184,main:@720,main:@1185) -> float_type, {64, 512}, {512, 1}: 0.0235172ms, 0.201573% main:@1187 = load[offset=3393024,end=3524096](main:@1) -> float_type, {64, 512}, {512, 1}: 0.00062516ms, 0.00535844% main:@1188 = gpu::code_object[code_object=13512,symbol_name=relu_kernel,global=16384,local=256,](main:@1186,main:@1187) -> float_type, {64, 512}, {512, 1}: 0.0156844ms, 0.134435% main:@1189 = hip::hip_copy_literal[id=main:@literal:97] -> float_type, {512, 256}, {256, 1}: 0.00073358ms, 0.00628774% main:@1190 = load[offset=9717248,end=9782784](main:@1) -> float_type, {64, 256}, {256, 1}: 0.00048344ms, 0.00414371% main:@1191 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1188,main:@1189,main:@1190) -> float_type, {64, 256}, {256, 1}: 0.0303818ms, 0.260412% main:@1192 = slice[axes={0, 1, 2},starts={0, 0, 0},ends={64, 1, 256}](main:@713) -> float_type, {64, 1, 256}, {4608, 256, 1}: 0.00076494ms, 0.00655654% main:@1193 = squeeze[axes={1}](main:@1192) -> float_type, {64, 256}, {4608, 1}: 0.00059984ms, 0.00514141% main:@1194 = load[offset=3524096,end=3655168](main:@1) -> float_type, {64, 512}, {512, 1}: 0.00047462ms, 0.00406811% main:@1195 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1193,main:@720,main:@1194) -> float_type, {64, 512}, {512, 1}: 0.0236078ms, 0.20235% main:@1196 = load[offset=3393024,end=3524096](main:@1) -> float_type, {64, 512}, {512, 1}: 0.00062302ms, 0.0053401% main:@1197 = gpu::code_object[code_object=13512,symbol_name=relu_kernel,global=16384,local=256,](main:@1195,main:@1196) -> float_type, {64, 512}, {512, 1}: 0.0158654ms, 0.135987% main:@1198 = hip::hip_copy_literal[id=main:@literal:70] -> float_type, {512, 256}, {256, 1}: 0.00076032ms, 0.00651694% main:@1199 = load[offset=9782784,end=9848320](main:@1) -> float_type, {64, 256}, {256, 1}: 0.00048814ms, 0.004184% main:@1200 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1197,main:@1198,main:@1199) -> float_type, {64, 256}, {256, 1}: 0.0301228ms, 0.258192% main:@1201 = hip::hip_copy_literal[id=main:@literal:86] -> float_type, {256, 512}, {512, 1}: 0.00082498ms, 0.00707116% main:@1202 = load[offset=3393024,end=3524096](main:@1) -> float_type, {64, 512}, {512, 1}: 0.00046634ms, 0.00399714% main:@1203 = slice[axes={0, 1, 2},starts={0, 14, 0},ends={64, 15, 256}](main:@713) -> float_type, {64, 1, 256}, {4608, 256, 1}: 0.00064414ms, 0.00552112% main:@1204 = squeeze[axes={1}](main:@1203) -> float_type, {64, 256}, {4608, 1}: 0.00093952ms, 0.00805291% main:@1205 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1204,main:@1201,main:@1202) -> float_type, {64, 512}, {512, 1}: 0.0249637ms, 0.213971% main:@1206 = load[offset=3524096,end=3655168](main:@1) -> float_type, {64, 512}, {512, 1}: 0.00062628ms, 0.00536804% main:@1207 = gpu::code_object[code_object=13512,symbol_name=relu_kernel,global=16384,local=256,](main:@1205,main:@1206) -> float_type, {64, 512}, {512, 1}: 0.0156219ms, 0.1339% main:@1208 = load[offset=9848320,end=9913856](main:@1) -> float_type, {64, 256}, {256, 1}: 0.00065174ms, 0.00558626% main:@1209 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1207,main:@719,main:@1208) -> float_type, {64, 256}, {256, 1}: 0.0298345ms, 0.25572% main:@1210 = hip::hip_copy_literal[id=main:@literal:141] -> float_type, {256, 512}, {512, 1}: 0.00075952ms, 0.00651008% main:@1211 = slice[axes={0, 1, 2},starts={0, 12, 0},ends={64, 13, 256}](main:@713) -> float_type, {64, 1, 256}, {4608, 256, 1}: 0.0006144ms, 0.00526621% main:@1212 = squeeze[axes={1}](main:@1211) -> float_type, {64, 256}, {4608, 1}: 0.00057492ms, 0.00492782% main:@1213 = load[offset=3524096,end=3655168](main:@1) -> float_type, {64, 512}, {512, 1}: 0.0004634ms, 0.00397194% main:@1214 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1212,main:@1210,main:@1213) -> float_type, {64, 512}, {512, 1}: 0.0237616ms, 0.203668% main:@1215 = load[offset=3393024,end=3524096](main:@1) -> float_type, {64, 512}, {512, 1}: 0.00064856ms, 0.00555901% main:@1216 = gpu::code_object[code_object=13512,symbol_name=relu_kernel,global=16384,local=256,](main:@1214,main:@1215) -> float_type, {64, 512}, {512, 1}: 0.0158279ms, 0.135665% main:@1217 = hip::hip_copy_literal[id=main:@literal:113] -> float_type, {512, 256}, {256, 1}: 0.0007567ms, 0.00648591% main:@1218 = load[offset=9913856,end=9979392](main:@1) -> float_type, {64, 256}, {256, 1}: 0.0005185ms, 0.00444422% main:@1219 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1216,main:@1217,main:@1218) -> float_type, {64, 256}, {256, 1}: 0.0300337ms, 0.257428% main:@1220 = slice[axes={0, 1, 2},starts={0, 17, 0},ends={64, 18, 256}](main:@713) -> float_type, {64, 1, 256}, {4608, 256, 1}: 0.00077948ms, 0.00668116% main:@1221 = squeeze[axes={1}](main:@1220) -> float_type, {64, 256}, {4608, 1}: 0.00060418ms, 0.00517861% main:@1222 = load[offset=3524096,end=3655168](main:@1) -> float_type, {64, 512}, {512, 1}: 0.00046212ms, 0.00396097% main:@1223 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1221,main:@718,main:@1222) -> float_type, {64, 512}, {512, 1}: 0.0246937ms, 0.211657% main:@1224 = load[offset=3393024,end=3524096](main:@1) -> float_type, {64, 512}, {512, 1}: 0.0006176ms, 0.00529364% main:@1225 = gpu::code_object[code_object=13512,symbol_name=relu_kernel,global=16384,local=256,](main:@1223,main:@1224) -> float_type, {64, 512}, {512, 1}: 0.0157903ms, 0.135343% main:@1226 = hip::hip_copy_literal[id=main:@literal:96] -> float_type, {512, 256}, {256, 1}: 0.000811ms, 0.00695133% main:@1227 = load[offset=9979392,end=10044928](main:@1) -> float_type, {64, 256}, {256, 1}: 0.00047634ms, 0.00408286% main:@1228 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1225,main:@1226,main:@1227) -> float_type, {64, 256}, {256, 1}: 0.0300439ms, 0.257515% main:@1229 = slice[axes={0, 1, 2},starts={0, 10, 0},ends={64, 11, 256}](main:@713) -> float_type, {64, 1, 256}, {4608, 256, 1}: 0.00078422ms, 0.00672179% main:@1230 = squeeze[axes={1}](main:@1229) -> float_type, {64, 256}, {4608, 1}: 0.00091728ms, 0.00786229% main:@1231 = load[offset=3524096,end=3655168](main:@1) -> float_type, {64, 512}, {512, 1}: 0.00047136ms, 0.00404017% main:@1232 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1230,main:@718,main:@1231) -> float_type, {64, 512}, {512, 1}: 0.0235099ms, 0.201511% main:@1233 = load[offset=3393024,end=3524096](main:@1) -> float_type, {64, 512}, {512, 1}: 0.00060156ms, 0.00515616% main:@1234 = gpu::code_object[code_object=13512,symbol_name=relu_kernel,global=16384,local=256,](main:@1232,main:@1233) -> float_type, {64, 512}, {512, 1}: 0.0157788ms, 0.135245% main:@1235 = hip::hip_copy_literal[id=main:@literal:42] -> float_type, {512, 256}, {256, 1}: 0.00075618ms, 0.00648145% main:@1236 = load[offset=13321728,end=13387264](main:@1) -> float_type, {64, 256}, {256, 1}: 0.00049066ms, 0.0042056% main:@1237 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1234,main:@1235,main:@1236) -> float_type, {64, 256}, {256, 1}: 0.0306907ms, 0.26306% main:@1238 = load[offset=3393024,end=3524096](main:@1) -> float_type, {64, 512}, {512, 1}: 0.00064062ms, 0.00549095% main:@1239 = slice[axes={0, 1, 2},starts={0, 13, 0},ends={64, 14, 256}](main:@713) -> float_type, {64, 1, 256}, {4608, 256, 1}: 0.00063476ms, 0.00544072% main:@1240 = squeeze[axes={1}](main:@1239) -> float_type, {64, 256}, {4608, 1}: 0.00060102ms, 0.00515153% main:@1241 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1240,main:@720,main:@1238) -> float_type, {64, 512}, {512, 1}: 0.0245827ms, 0.210706% main:@1242 = load[offset=3524096,end=3655168](main:@1) -> float_type, {64, 512}, {512, 1}: 0.00061832ms, 0.00529981% main:@1243 = gpu::code_object[code_object=13512,symbol_name=relu_kernel,global=16384,local=256,](main:@1241,main:@1242) -> float_type, {64, 512}, {512, 1}: 0.0156822ms, 0.134417% main:@1244 = hip::hip_copy_literal[id=main:@literal:41] -> float_type, {512, 256}, {256, 1}: 0.00075068ms, 0.00643431% main:@1245 = load[offset=13387264,end=13452800](main:@1) -> float_type, {64, 256}, {256, 1}: 0.00049272ms, 0.00422325% main:@1246 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1243,main:@1244,main:@1245) -> float_type, {64, 256}, {256, 1}: 0.0299653ms, 0.256842% main:@1247 = hip::hip_copy_literal[id=main:@literal:112] -> float_type, {256, 512}, {512, 1}: 0.00081764ms, 0.00700824% main:@1248 = slice[axes={0, 1, 2},starts={0, 7, 0},ends={64, 8, 256}](main:@713) -> float_type, {64, 1, 256}, {4608, 256, 1}: 0.00062314ms, 0.00534112% main:@1249 = squeeze[axes={1}](main:@1248) -> float_type, {64, 256}, {4608, 1}: 0.00093164ms, 0.00798537% main:@1250 = load[offset=3524096,end=3655168](main:@1) -> float_type, {64, 512}, {512, 1}: 0.00045908ms, 0.00393492% main:@1251 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1249,main:@1247,main:@1250) -> float_type, {64, 512}, {512, 1}: 0.0234685ms, 0.201156% main:@1252 = load[offset=3393024,end=3524096](main:@1) -> float_type, {64, 512}, {512, 1}: 0.00062714ms, 0.00537541% main:@1253 = gpu::code_object[code_object=13512,symbol_name=relu_kernel,global=16384,local=256,](main:@1251,main:@1252) -> float_type, {64, 512}, {512, 1}: 0.0156959ms, 0.134534% main:@1254 = hip::hip_copy_literal[id=main:@literal:140] -> float_type, {512, 256}, {256, 1}: 0.00080576ms, 0.00690642% main:@1255 = load[offset=13452800,end=13518336](main:@1) -> float_type, {64, 256}, {256, 1}: 0.0004695ms, 0.00402423% main:@1256 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1253,main:@1254,main:@1255) -> float_type, {64, 256}, {256, 1}: 0.0301834ms, 0.258712% main:@1257 = hip::hip_copy_literal[id=main:@literal:153] -> float_type, {256, 512}, {512, 1}: 0.00073086ms, 0.00626443% main:@1258 = slice[axes={0, 1, 2},starts={0, 6, 0},ends={64, 7, 256}](main:@713) -> float_type, {64, 1, 256}, {4608, 256, 1}: 0.0009437ms, 0.00808874% main:@1259 = squeeze[axes={1}](main:@1258) -> float_type, {64, 256}, {4608, 1}: 0.00057992ms, 0.00497067% main:@1260 = load[offset=3524096,end=3655168](main:@1) -> float_type, {64, 512}, {512, 1}: 0.0004736ms, 0.00405937% main:@1261 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1259,main:@1257,main:@1260) -> float_type, {64, 512}, {512, 1}: 0.0249126ms, 0.213534% main:@1262 = load[offset=3393024,end=3524096](main:@1) -> float_type, {64, 512}, {512, 1}: 0.0006169ms, 0.00528764% main:@1263 = gpu::code_object[code_object=13512,symbol_name=relu_kernel,global=16384,local=256,](main:@1261,main:@1262) -> float_type, {64, 512}, {512, 1}: 0.0157616ms, 0.135097% main:@1264 = hip::hip_copy_literal[id=main:@literal:93] -> float_type, {512, 256}, {256, 1}: 0.0007098ms, 0.00608391% main:@1265 = load[offset=13518336,end=13583872](main:@1) -> float_type, {64, 256}, {256, 1}: 0.00047618ms, 0.00408148% main:@1266 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1263,main:@1264,main:@1265) -> float_type, {64, 256}, {256, 1}: 0.0301409ms, 0.258347% main:@1267 = hip::hip_copy_literal[id=main:@literal:68] -> float_type, {256, 512}, {512, 1}: 0.00074864ms, 0.00641682% main:@1268 = slice[axes={0, 1, 2},starts={0, 11, 0},ends={64, 12, 256}](main:@713) -> float_type, {64, 1, 256}, {4608, 256, 1}: 0.00064282ms, 0.00550981% main:@1269 = squeeze[axes={1}](main:@1268) -> float_type, {64, 256}, {4608, 1}: 0.00091214ms, 0.00781823% main:@1270 = load[offset=3524096,end=3655168](main:@1) -> float_type, {64, 512}, {512, 1}: 0.0004656ms, 0.0039908% main:@1271 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1269,main:@1267,main:@1270) -> float_type, {64, 512}, {512, 1}: 0.0235674ms, 0.202003% main:@1272 = load[offset=3393024,end=3524096](main:@1) -> float_type, {64, 512}, {512, 1}: 0.00060266ms, 0.00516558% main:@1273 = gpu::code_object[code_object=13512,symbol_name=relu_kernel,global=16384,local=256,](main:@1271,main:@1272) -> float_type, {64, 512}, {512, 1}: 0.0157542ms, 0.135034% main:@1274 = load[offset=13583872,end=13649408](main:@1) -> float_type, {64, 256}, {256, 1}: 0.00059682ms, 0.00511553% main:@1275 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1273,main:@719,main:@1274) -> float_type, {64, 256}, {256, 1}: 0.030006ms, 0.257191% main:@1276 = hip::hip_copy_literal[id=main:@literal:15] -> float_type, {82, 256}, {256, 1}: 0.0008154ms, 0.00698904% main:@1277 = slice[axes={0, 1},starts={0, 9},ends={64, 11}](w04Keh2FJu:0) -> float_type, {64, 2}, {13, 1}: 0.00083348ms, 0.00714401% main:@1278 = slice[axes={0, 1},starts={0, 41},ends={64, 45}](2CnSyUBzm0:0) -> float_type, {64, 4}, {65, 1}: 0.00065048ms, 0.00557546% main:@1279 = slice[axes={0, 1},starts={0, 5},ends={64, 13}](9YqhflbMKe:0) -> float_type, {64, 8}, {29, 1}: 0.00078634ms, 0.00673996% main:@1280 = slice[axes={0, 1},starts={0, 501},ends={64, 509}](lyiDxfzZg3:0) -> float_type, {64, 8}, {701, 1}: 0.0006091ms, 0.00522078% main:@1281 = slice[axes={0, 1},starts={0, 181},ends={64, 189}](nv2tzWRf1I:0) -> float_type, {64, 8}, {301, 1}: 0.00057602ms, 0.00493724% main:@1282 = slice[axes={0, 1},starts={0, 33},ends={64, 41}](FcmWLZhDgE:0) -> float_type, {64, 8}, {57, 1}: 0.0006019ms, 0.00515907% main:@1283 = slice[axes={0, 1},starts={0, 21},ends={64, 29}](YbIVz8l1un:0) -> float_type, {64, 8}, {45, 1}: 0.00076624ms, 0.00656768% main:@1284 = slice[axes={0, 1},starts={0, 41},ends={64, 49}](ifP6Ceko5T:0) -> float_type, {64, 8}, {73, 1}: 0.0005871ms, 0.00503221% main:@1285 = slice[axes={0, 1},starts={0, 37},ends={64, 45}](Qp3K1PHvte:0) -> float_type, {64, 8}, {69, 1}: 0.0005987ms, 0.00513164% main:@1286 = slice[axes={0, 1},starts={0, 5},ends={64, 13}](ljyhc7HCLk:0) -> float_type, {64, 8}, {21, 1}: 0.00059412ms, 0.00509238% main:@1287 = slice[axes={0, 1},starts={0, 33},ends={64, 37}](ZcWERoz0h6:0) -> float_type, {64, 4}, {49, 1}: 0.00073418ms, 0.00629288% main:@1288 = slice[axes={0, 1},starts={0, 21},ends={64, 25}](bJsIAzhYc3:0) -> float_type, {64, 4}, {37, 1}: 0.00059694ms, 0.00511656% main:@1289 = slice[axes={0, 1},starts={0, 33},ends={64, 37}](isCn4mEOSt:0) -> float_type, {64, 4}, {49, 1}: 0.00061042ms, 0.0052321% main:@1290 = load[offset=3393024,end=3414016](main:@1) -> float_type, {64, 82}, {82, 1}: 0.00048744ms, 0.004178% main:@1291 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=403,local=256,](main:@1280,main:@1281,main:@1284,main:@1285,main:@1278,main:@1283,main:@1289,main:@1287,main:@1277,main:@1282,main:@1288,main:@1286,main:@1279,main:@1290) -> float_type, {64, 82}, {82, 1}: 0.0278689ms, 0.238873% main:@1292 = load[offset=13944320,end=14009856](main:@1) -> float_type, {64, 256}, {256, 1}: 0.00063376ms, 0.00543215% main:@1293 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1291,main:@1276,main:@1292) -> float_type, {64, 256}, {256, 1}: 0.0207777ms, 0.178092% main:@1294 = load[offset=3393024,end=3524096](main:@1) -> float_type, {64, 512}, {512, 1}: 0.00066822ms, 0.00572752% main:@1295 = slice[axes={0, 1, 2},starts={0, 4, 0},ends={64, 5, 256}](main:@713) -> float_type, {64, 1, 256}, {4608, 256, 1}: 0.00061882ms, 0.0053041% main:@1296 = squeeze[axes={1}](main:@1295) -> float_type, {64, 256}, {4608, 1}: 0.00073578ms, 0.0063066% main:@1297 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1296,main:@720,main:@1294) -> float_type, {64, 512}, {512, 1}: 0.0235608ms, 0.201947% main:@1298 = load[offset=3524096,end=3655168](main:@1) -> float_type, {64, 512}, {512, 1}: 0.0006149ms, 0.0052705% main:@1299 = gpu::code_object[code_object=13512,symbol_name=relu_kernel,global=16384,local=256,](main:@1297,main:@1298) -> float_type, {64, 512}, {512, 1}: 0.0155543ms, 0.133321% main:@1300 = hip::hip_copy_literal[id=main:@literal:94] -> float_type, {512, 256}, {256, 1}: 0.00072856ms, 0.00624471% main:@1301 = load[offset=13649408,end=13714944](main:@1) -> float_type, {64, 256}, {256, 1}: 0.00047466ms, 0.00406846% main:@1302 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1299,main:@1300,main:@1301) -> float_type, {64, 256}, {256, 1}: 0.0310004ms, 0.265714% main:@1303 = hip::hip_copy_literal[id=main:@literal:149] -> float_type, {256, 512}, {512, 1}: 0.0007807ms, 0.00669162% main:@1304 = slice[axes={0, 1, 2},starts={0, 3, 0},ends={64, 4, 256}](main:@713) -> float_type, {64, 1, 256}, {4608, 256, 1}: 0.00066274ms, 0.00568055% main:@1305 = squeeze[axes={1}](main:@1304) -> float_type, {64, 256}, {4608, 1}: 0.00062788ms, 0.00538175% main:@1306 = load[offset=3524096,end=3655168](main:@1) -> float_type, {64, 512}, {512, 1}: 0.00045928ms, 0.00393663% main:@1307 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1305,main:@1303,main:@1306) -> float_type, {64, 512}, {512, 1}: 0.0249478ms, 0.213836% main:@1308 = load[offset=3393024,end=3524096](main:@1) -> float_type, {64, 512}, {512, 1}: 0.000623ms, 0.00533992% main:@1309 = gpu::code_object[code_object=13512,symbol_name=relu_kernel,global=16384,local=256,](main:@1307,main:@1308) -> float_type, {64, 512}, {512, 1}: 0.0158262ms, 0.135651% main:@1310 = load[offset=13878784,end=13944320](main:@1) -> float_type, {64, 256}, {256, 1}: 0.00061096ms, 0.00523673% main:@1311 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1309,main:@1149,main:@1310) -> float_type, {64, 256}, {256, 1}: 0.0299573ms, 0.256773% main:@1312 = load[offset=3442176,end=3491840](main:@1) -> float_type, {64, 194}, {194, 1}: 0.00062122ms, 0.00532467% main:@1313 = slice[axes={0, 1},starts={0, 9},ends={64, 17}](09zp4EDXTh:0) -> float_type, {64, 8}, {25, 1}: 0.00067192ms, 0.00575923% main:@1314 = slice[axes={0, 1},starts={0, 169},ends={64, 177}](uFSOc5Chp2:0) -> float_type, {64, 8}, {481, 1}: 0.0007383ms, 0.0063282% main:@1315 = slice[axes={0, 1},starts={0, 9},ends={64, 17}](1PJsBhDaCK:0) -> float_type, {64, 8}, {25, 1}: 0.0010837ms, 0.00928873% main:@1316 = slice[axes={0, 1},starts={0, 5},ends={64, 9}](1u5yIsHTXE:0) -> float_type, {64, 4}, {17, 1}: 0.00060118ms, 0.0051529% main:@1317 = slice[axes={0, 1},starts={0, 9},ends={64, 17}](2lVXMZgJfO:0) -> float_type, {64, 8}, {25, 1}: 0.00060928ms, 0.00522233% main:@1318 = slice[axes={0, 1},starts={0, 9},ends={64, 17}](3kdi0vsSZG:0) -> float_type, {64, 8}, {25, 1}: 0.00060268ms, 0.00516576% main:@1319 = slice[axes={0, 1},starts={0, 25},ends={64, 33}](Qp3K1PHvte:0) -> float_type, {64, 8}, {69, 1}: 0.00104512ms, 0.00895804% main:@1320 = slice[axes={0, 1},starts={0, 3},ends={64, 5}](5JHUPChra8:0) -> float_type, {64, 2}, {17, 1}: 0.00061684ms, 0.00528712% main:@1321 = slice[axes={0, 1},starts={0, 9},ends={64, 17}](RpITWntAji:0) -> float_type, {64, 8}, {17, 1}: 0.00062622ms, 0.00536752% main:@1322 = slice[axes={0, 1},starts={0, 9},ends={64, 17}](92MwiALCHX:0) -> float_type, {64, 8}, {25, 1}: 0.0006182ms, 0.00529878% main:@1323 = slice[axes={0, 1},starts={0, 5},ends={64, 9}](AVLcBpWeuo:0) -> float_type, {64, 4}, {21, 1}: 0.0010306ms, 0.00883359% main:@1324 = slice[axes={0, 1},starts={0, 9},ends={64, 17}](JSWhLK4Agx:0) -> float_type, {64, 8}, {25, 1}: 0.00060948ms, 0.00522404% main:@1325 = slice[axes={0, 1},starts={0, 5},ends={64, 9}](o74vNAnXxy:0) -> float_type, {64, 4}, {17, 1}: 0.00059032ms, 0.00505981% main:@1326 = slice[axes={0, 1},starts={0, 9},ends={64, 17}](Tg4u0GfnFb:0) -> float_type, {64, 8}, {25, 1}: 0.00062068ms, 0.00532004% main:@1327 = slice[axes={0, 1},starts={0, 3},ends={64, 5}](TidKGrnobI:0) -> float_type, {64, 2}, {13, 1}: 0.00108808ms, 0.00932627% main:@1328 = slice[axes={0, 1},starts={0, 5},ends={64, 9}](HZ0l3MmcPY:0) -> float_type, {64, 4}, {17, 1}: 0.0006209ms, 0.00532192% main:@1329 = slice[axes={0, 1},starts={0, 3},ends={64, 5}](IztrlRfWQU:0) -> float_type, {64, 2}, {13, 1}: 0.00058838ms, 0.00504319% main:@1330 = slice[axes={0, 1},starts={0, 3},ends={64, 5}](KN13IWXZSu:0) -> float_type, {64, 2}, {13, 1}: 0.0005918ms, 0.0050725% main:@1331 = slice[axes={0, 1},starts={0, 5},ends={64, 9}](RHo9Jz6KpI:0) -> float_type, {64, 4}, {21, 1}: 0.00100472ms, 0.00861176% main:@1332 = slice[axes={0, 1},starts={0, 9},ends={64, 17}](NXOSdQVPs0:0) -> float_type, {64, 8}, {25, 1}: 0.00059536ms, 0.00510301% main:@1333 = slice[axes={0, 1},starts={0, 9},ends={64, 17}](ry2xad1vc9:0) -> float_type, {64, 8}, {25, 1}: 0.00063206ms, 0.00541758% main:@1334 = slice[axes={0, 1},starts={0, 5},ends={64, 9}](S5D6WGIXMd:0) -> float_type, {64, 4}, {17, 1}: 0.00070544ms, 0.00604654% main:@1335 = slice[axes={0, 1},starts={0, 9},ends={64, 17}](1WNvHElUG7:0) -> float_type, {64, 8}, {25, 1}: 0.00097494ms, 0.00835651% main:@1336 = slice[axes={0, 1},starts={0, 9},ends={64, 17}](YbIVz8l1un:0) -> float_type, {64, 8}, {45, 1}: 0.00068064ms, 0.00583397% main:@1337 = slice[axes={0, 1},starts={0, 9},ends={64, 17}](bJsIAzhYc3:0) -> float_type, {64, 8}, {37, 1}: 0.00064262ms, 0.00550809% main:@1338 = slice[axes={0, 1},starts={0, 5},ends={64, 9}](pH5srbufJ1:0) -> float_type, {64, 4}, {17, 1}: 0.00072318ms, 0.0061986% main:@1339 = slice[axes={0, 1},starts={0, 3},ends={64, 5}](tWcngAq7hV:0) -> float_type, {64, 2}, {13, 1}: 0.00100108ms, 0.00858056% main:@1340 = slice[axes={0, 1},starts={0, 3},ends={64, 5}](xymEBUMS25:0) -> float_type, {64, 2}, {13, 1}: 0.00063332ms, 0.00542838% main:@1341 = slice[axes={0, 1},starts={0, 5},ends={64, 9}](msiPzSBKMj:0) -> float_type, {64, 4}, {17, 1}: 0.00062536ms, 0.00536015% main:@1342 = slice[axes={0, 1},starts={0, 5},ends={64, 9}](xT9FhjM1Uu:0) -> float_type, {64, 4}, {17, 1}: 0.00065026ms, 0.00557358% main:@1343 = slice[axes={0, 1},starts={0, 5},ends={64, 9}](m2YscaoDjP:0) -> float_type, {64, 4}, {17, 1}: 0.00097252ms, 0.00833577% main:@1344 = slice[axes={0, 1},starts={0, 3},ends={64, 5}](zZHaVOckpw:0) -> float_type, {64, 2}, {13, 1}: 0.0006479ms, 0.00555335% main:@1345 = slice[axes={0, 1},starts={0, 9},ends={64, 17}](qd1komWTn4:0) -> float_type, {64, 8}, {25, 1}: 0.0006404ms, 0.00548906% main:@1346 = slice[axes={0, 1},starts={0, 3},ends={64, 5}](ujOA5LQW7M:0) -> float_type, {64, 2}, {13, 1}: 0.00063862ms, 0.00547381% main:@1347 = slice[axes={0, 1},starts={0, 3},ends={64, 5}](nPUSZFm74Y:0) -> float_type, {64, 2}, {13, 1}: 0.00092052ms, 0.00789006% main:@1348 = slice[axes={0, 1},starts={0, 163},ends={64, 165}](nv2tzWRf1I:0) -> float_type, {64, 2}, {301, 1}: 0.00064078ms, 0.00549232% main:@1349 = slice[axes={0, 1},starts={0, 21},ends={64, 25}](ifP6Ceko5T:0) -> float_type, {64, 4}, {73, 1}: 0.00062002ms, 0.00531438% main:@1350 = slice[axes={0, 1},starts={0, 3},ends={64, 5}](XxirNgB98u:0) -> float_type, {64, 2}, {13, 1}: 0.00064346ms, 0.00551529% main:@1351 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=326,local=256,](main:@1320,main:@1350,main:@1339,main:@1340,main:@1329,main:@1330,main:@1347,main:@1346,main:@1344,main:@1327,main:@1348,main:@1328,main:@1338,main:@1341,main:@1316,main:@1331,main:@1334,main:@1343,main:@1323,main:@1325,main:@1342,main:@1349,main:@1313,main:@1315,main:@1319,main:@1324,main:@1345,main:@1333,main:@1317,main:@1322,main:@1326,main:@1335,main:@1332,main:@1337,main:@1318,main:@1336,main:@1321,main:@1314,main:@1312) -> float_type, {64, 194}, {194, 1}: 0.0478188ms, 0.40987% main:@1352 = hip::hip_copy_literal[id=main:@literal:106] -> float_type, {194, 192}, {192, 1}: 0.00084178ms, 0.00721515% main:@1353 = load[offset=3393024,end=3442176](main:@1) -> float_type, {64, 192}, {192, 1}: 0.00047528ms, 0.00407377% main:@1354 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1351,main:@1352,main:@1353) -> float_type, {64, 192}, {192, 1}: 0.0249851ms, 0.214155% vAbKepTjQg:0 = @param:vAbKepTjQg:0 -> float_type, {64, 5}, {5, 1}: 0.00077416ms, 0.00663556% main:@1356 = slice[axes={0, 1},starts={0, 3},ends={64, 5}](vAbKepTjQg:0) -> float_type, {64, 2}, {5, 1}: 0.00069312ms, 0.00594094% main:@1357 = load[offset=3442176,end=3485696](main:@1) -> float_type, {64, 170}, {170, 1}: 0.0005022ms, 0.00430451% lDhFNYzbxR:0 = @param:lDhFNYzbxR:0 -> float_type, {64, 17}, {17, 1}: 0.00063444ms, 0.00543798% main:@1359 = slice[axes={0, 1},starts={0, 9},ends={64, 17}](lDhFNYzbxR:0) -> float_type, {64, 8}, {17, 1}: 0.00083548ms, 0.00716116% 1QWFmJ96aB:0 = @param:1QWFmJ96aB:0 -> float_type, {64, 5}, {5, 1}: 0.0005105ms, 0.00437565% main:@1361 = slice[axes={0, 1},starts={0, 3},ends={64, 5}](1QWFmJ96aB:0) -> float_type, {64, 2}, {5, 1}: 0.00062398ms, 0.00534832% 3VED7WSJ1b:0 = @param:3VED7WSJ1b:0 -> float_type, {64, 5}, {5, 1}: 0.00053298ms, 0.00456834% main:@1363 = slice[axes={0, 1},starts={0, 3},ends={64, 5}](3VED7WSJ1b:0) -> float_type, {64, 2}, {5, 1}: 0.00074094ms, 0.00635082% gdck97AK4w:0 = @param:gdck97AK4w:0 -> float_type, {64, 17}, {17, 1}: 0.00050462ms, 0.00432525% main:@1365 = slice[axes={0, 1},starts={0, 5},ends={64, 9}](gdck97AK4w:0) -> float_type, {64, 4}, {17, 1}: 0.00062222ms, 0.00533324% tIGjbcEuDd:0 = @param:tIGjbcEuDd:0 -> float_type, {64, 9}, {9, 1}: 0.00048118ms, 0.00412434% main:@1367 = slice[axes={0, 1},starts={0, 5},ends={64, 9}](tIGjbcEuDd:0) -> float_type, {64, 4}, {9, 1}: 0.00073454ms, 0.00629597% NJBt1ubXFW:0 = @param:NJBt1ubXFW:0 -> float_type, {64, 5}, {5, 1}: 0.00053204ms, 0.00456028% main:@1369 = slice[axes={0, 1},starts={0, 3},ends={64, 5}](NJBt1ubXFW:0) -> float_type, {64, 2}, {5, 1}: 0.00076388ms, 0.00654745% zX6xKOWG9S:0 = @param:zX6xKOWG9S:0 -> float_type, {64, 5}, {5, 1}: 0.00050646ms, 0.00434102% main:@1371 = slice[axes={0, 1},starts={0, 3},ends={64, 5}](zX6xKOWG9S:0) -> float_type, {64, 2}, {5, 1}: 0.00078692ms, 0.00674493% 5yColzZ8K3:0 = @param:5yColzZ8K3:0 -> float_type, {64, 5}, {5, 1}: 0.0005021ms, 0.00430365% main:@1373 = slice[axes={0, 1},starts={0, 3},ends={64, 5}](5yColzZ8K3:0) -> float_type, {64, 2}, {5, 1}: 0.00059476ms, 0.00509787% IxpbBg7JTO:0 = @param:IxpbBg7JTO:0 -> float_type, {64, 5}, {5, 1}: 0.00054496ms, 0.00467102% main:@1375 = slice[axes={0, 1},starts={0, 3},ends={64, 5}](IxpbBg7JTO:0) -> float_type, {64, 2}, {5, 1}: 0.00075728ms, 0.00649088% 6Wh2Jm7GHC:0 = @param:6Wh2Jm7GHC:0 -> float_type, {64, 5}, {5, 1}: 0.00056682ms, 0.00485839% main:@1377 = slice[axes={0, 1},starts={0, 3},ends={64, 5}](6Wh2Jm7GHC:0) -> float_type, {64, 2}, {5, 1}: 0.00059688ms, 0.00511604% UMO0o14RuI:0 = @param:UMO0o14RuI:0 -> float_type, {64, 9}, {9, 1}: 0.00050808ms, 0.00435491% main:@1379 = slice[axes={0, 1},starts={0, 5},ends={64, 9}](UMO0o14RuI:0) -> float_type, {64, 4}, {9, 1}: 0.000802ms, 0.00687419% IMdFk7D16X:0 = @param:IMdFk7D16X:0 -> float_type, {64, 9}, {9, 1}: 0.00055532ms, 0.00475982% main:@1381 = slice[axes={0, 1},starts={0, 5},ends={64, 9}](IMdFk7D16X:0) -> float_type, {64, 4}, {9, 1}: 0.00063722ms, 0.00546181% Pu1O5qFxY6:0 = @param:Pu1O5qFxY6:0 -> float_type, {64, 17}, {17, 1}: 0.00056766ms, 0.00486559% main:@1383 = slice[axes={0, 1},starts={0, 5},ends={64, 9}](Pu1O5qFxY6:0) -> float_type, {64, 4}, {17, 1}: 0.00075326ms, 0.00645642% Cw59eOo7M1:0 = @param:Cw59eOo7M1:0 -> float_type, {64, 9}, {9, 1}: 0.00051686ms, 0.00443017% main:@1385 = slice[axes={0, 1},starts={0, 5},ends={64, 9}](Cw59eOo7M1:0) -> float_type, {64, 4}, {9, 1}: 0.00096832ms, 0.00829977% E1ZkVIXqS5:0 = @param:E1ZkVIXqS5:0 -> float_type, {64, 9}, {9, 1}: 0.00056514ms, 0.00484399% main:@1387 = slice[axes={0, 1},starts={0, 5},ends={64, 9}](E1ZkVIXqS5:0) -> float_type, {64, 4}, {9, 1}: 0.00073112ms, 0.00626665% EhUjmCTaGW:0 = @param:EhUjmCTaGW:0 -> float_type, {64, 9}, {9, 1}: 0.00049138ms, 0.00421177% main:@1389 = slice[axes={0, 1},starts={0, 5},ends={64, 9}](EhUjmCTaGW:0) -> float_type, {64, 4}, {9, 1}: 0.00060356ms, 0.0051733% l5WSX3VvMK:0 = @param:l5WSX3VvMK:0 -> float_type, {64, 25}, {25, 1}: 0.00049408ms, 0.00423491% main:@1391 = slice[axes={0, 1},starts={0, 9},ends={64, 17}](l5WSX3VvMK:0) -> float_type, {64, 8}, {25, 1}: 0.00064424ms, 0.00552198% FWhBxj4Tun:0 = @param:FWhBxj4Tun:0 -> float_type, {64, 9}, {9, 1}: 0.00050746ms, 0.0043496% main:@1393 = slice[axes={0, 1},starts={0, 5},ends={64, 9}](FWhBxj4Tun:0) -> float_type, {64, 4}, {9, 1}: 0.00073272ms, 0.00628037% HmATr3gKsW:0 = @param:HmATr3gKsW:0 -> float_type, {64, 9}, {9, 1}: 0.00052694ms, 0.00451656% main:@1395 = slice[axes={0, 1},starts={0, 5},ends={64, 9}](HmATr3gKsW:0) -> float_type, {64, 4}, {9, 1}: 0.00070064ms, 0.0060054% Ouh1XpPZaj:0 = @param:Ouh1XpPZaj:0 -> float_type, {64, 9}, {9, 1}: 0.0005164ms, 0.00442622% main:@1397 = slice[axes={0, 1},starts={0, 5},ends={64, 9}](Ouh1XpPZaj:0) -> float_type, {64, 4}, {9, 1}: 0.00068124ms, 0.00583912% J3H2EjAXt8:0 = @param:J3H2EjAXt8:0 -> float_type, {64, 5}, {5, 1}: 0.00051754ms, 0.00443599% main:@1399 = slice[axes={0, 1},starts={0, 3},ends={64, 5}](J3H2EjAXt8:0) -> float_type, {64, 2}, {5, 1}: 0.00068824ms, 0.00589912% NLt5JnRMjG:0 = @param:NLt5JnRMjG:0 -> float_type, {64, 5}, {5, 1}: 0.00056738ms, 0.00486319% main:@1401 = slice[axes={0, 1},starts={0, 3},ends={64, 5}](NLt5JnRMjG:0) -> float_type, {64, 2}, {5, 1}: 0.00101932ms, 0.0087369% KCNfS4GlpP:0 = @param:KCNfS4GlpP:0 -> float_type, {64, 9}, {9, 1}: 0.00046534ms, 0.00398857% main:@1403 = slice[axes={0, 1},starts={0, 5},ends={64, 9}](KCNfS4GlpP:0) -> float_type, {64, 4}, {9, 1}: 0.00070324ms, 0.00602769% O5aJeUQh2G:0 = @param:O5aJeUQh2G:0 -> float_type, {64, 17}, {17, 1}: 0.00050114ms, 0.00429542% main:@1405 = slice[axes={0, 1},starts={0, 9},ends={64, 17}](O5aJeUQh2G:0) -> float_type, {64, 8}, {17, 1}: 0.00065436ms, 0.00560872% nltbg9ZWQX:0 = @param:nltbg9ZWQX:0 -> float_type, {64, 17}, {17, 1}: 0.00060914ms, 0.00522113% main:@1407 = slice[axes={0, 1},starts={0, 5},ends={64, 9}](nltbg9ZWQX:0) -> float_type, {64, 4}, {17, 1}: 0.00070356ms, 0.00603043% Uc42tiuavR:0 = @param:Uc42tiuavR:0 -> float_type, {64, 9}, {9, 1}: 0.0005179ms, 0.00443908% main:@1409 = slice[axes={0, 1},starts={0, 5},ends={64, 9}](Uc42tiuavR:0) -> float_type, {64, 4}, {9, 1}: 0.00065564ms, 0.00561969% U6GqhP8k7I:0 = @param:U6GqhP8k7I:0 -> float_type, {64, 9}, {9, 1}: 0.00051676ms, 0.00442931% main:@1411 = slice[axes={0, 1},starts={0, 5},ends={64, 9}](U6GqhP8k7I:0) -> float_type, {64, 4}, {9, 1}: 0.00064712ms, 0.00554666% oQ5OcjZJe8:0 = @param:oQ5OcjZJe8:0 -> float_type, {64, 17}, {17, 1}: 0.00053318ms, 0.00457005% main:@1413 = slice[axes={0, 1},starts={0, 9},ends={64, 17}](oQ5OcjZJe8:0) -> float_type, {64, 8}, {17, 1}: 0.00070084ms, 0.00600711% yVxKAL4U5G:0 = @param:yVxKAL4U5G:0 -> float_type, {64, 9}, {9, 1}: 0.00054768ms, 0.00469433% main:@1415 = slice[axes={0, 1},starts={0, 5},ends={64, 9}](yVxKAL4U5G:0) -> float_type, {64, 4}, {9, 1}: 0.00064996ms, 0.00557101% ZRYNFfjSez:0 = @param:ZRYNFfjSez:0 -> float_type, {64, 33}, {33, 1}: 0.00054856ms, 0.00470188% main:@1417 = slice[axes={0, 1},starts={0, 9},ends={64, 17}](ZRYNFfjSez:0) -> float_type, {64, 8}, {33, 1}: 0.00099762ms, 0.00855091% ecPTQUALkq:0 = @param:ecPTQUALkq:0 -> float_type, {64, 5}, {5, 1}: 0.00046934ms, 0.00402286% main:@1419 = slice[axes={0, 1},starts={0, 3},ends={64, 5}](ecPTQUALkq:0) -> float_type, {64, 2}, {5, 1}: 0.00066924ms, 0.00573626% magwfJSnUu:0 = @param:magwfJSnUu:0 -> float_type, {64, 9}, {9, 1}: 0.00051556ms, 0.00441902% main:@1421 = slice[axes={0, 1},starts={0, 5},ends={64, 9}](magwfJSnUu:0) -> float_type, {64, 4}, {9, 1}: 0.00066772ms, 0.00572323% mNgQ4ohWqU:0 = @param:mNgQ4ohWqU:0 -> float_type, {64, 9}, {9, 1}: 0.00046784ms, 0.00401% main:@1423 = slice[axes={0, 1},starts={0, 5},ends={64, 9}](mNgQ4ohWqU:0) -> float_type, {64, 4}, {9, 1}: 0.00066028ms, 0.00565946% gsP96ZTbn0:0 = @param:gsP96ZTbn0:0 -> float_type, {64, 17}, {17, 1}: 0.00055432ms, 0.00475125% main:@1425 = slice[axes={0, 1},starts={0, 9},ends={64, 17}](gsP96ZTbn0:0) -> float_type, {64, 8}, {17, 1}: 0.0006373ms, 0.00546249% gqk5byC8aJ:0 = @param:gqk5byC8aJ:0 -> float_type, {64, 5}, {5, 1}: 0.00050888ms, 0.00436177% main:@1427 = slice[axes={0, 1},starts={0, 3},ends={64, 5}](gqk5byC8aJ:0) -> float_type, {64, 2}, {5, 1}: 0.00070746ms, 0.00606386% mTMtDEo3r8:0 = @param:mTMtDEo3r8:0 -> float_type, {64, 9}, {9, 1}: 0.0006434ms, 0.00551478% main:@1429 = slice[axes={0, 1},starts={0, 5},ends={64, 9}](mTMtDEo3r8:0) -> float_type, {64, 4}, {9, 1}: 0.00064958ms, 0.00556775% xVErSbJ0hg:0 = @param:xVErSbJ0hg:0 -> float_type, {64, 9}, {9, 1}: 0.0005423ms, 0.00464822% main:@1431 = slice[axes={0, 1},starts={0, 5},ends={64, 9}](xVErSbJ0hg:0) -> float_type, {64, 4}, {9, 1}: 0.0006598ms, 0.00565535% qfwYOHde3u:0 = @param:qfwYOHde3u:0 -> float_type, {64, 17}, {17, 1}: 0.00053348ms, 0.00457262% main:@1433 = slice[axes={0, 1},starts={0, 9},ends={64, 17}](qfwYOHde3u:0) -> float_type, {64, 8}, {17, 1}: 0.00103276ms, 0.0088521% pQsi1GNxvj:0 = @param:pQsi1GNxvj:0 -> float_type, {64, 5}, {5, 1}: 0.00057272ms, 0.00490896% main:@1435 = slice[axes={0, 1},starts={0, 3},ends={64, 5}](pQsi1GNxvj:0) -> float_type, {64, 2}, {5, 1}: 0.00066524ms, 0.00570198% l4bReWtsMB:0 = @param:l4bReWtsMB:0 -> float_type, {64, 25}, {25, 1}: 0.0005208ms, 0.00446394% main:@1437 = slice[axes={0, 1},starts={0, 9},ends={64, 17}](l4bReWtsMB:0) -> float_type, {64, 8}, {25, 1}: 0.00065612ms, 0.00562381% main:@1438 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=265,local=256,](main:@1377,main:@1419,main:@1369,main:@1356,main:@1371,main:@1435,main:@1401,main:@1363,main:@1361,main:@1399,main:@1427,main:@1375,main:@1373,main:@1389,main:@1411,main:@1387,main:@1385,main:@1403,main:@1409,main:@1381,main:@1423,main:@1429,main:@1415,main:@1365,main:@1383,main:@1367,main:@1431,main:@1379,main:@1397,main:@1407,main:@1395,main:@1421,main:@1393,main:@1433,main:@1359,main:@1425,main:@1405,main:@1417,main:@1437,main:@1391,main:@1413,main:@1357) -> float_type, {64, 170}, {170, 1}: 0.0479319ms, 0.410839% main:@1439 = hip::hip_copy_literal[id=main:@literal:73] -> float_type, {170, 160}, {160, 1}: 0.0008518ms, 0.00730104% main:@1440 = load[offset=3515904,end=3556864](main:@1) -> float_type, {64, 160}, {160, 1}: 0.00047796ms, 0.00409674% main:@1441 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1438,main:@1439,main:@1440) -> float_type, {64, 160}, {160, 1}: 0.0226299ms, 0.193968% main:@1442 = load[offset=3442176,end=3515904](main:@1) -> float_type, {64, 18, 16}, {288, 16, 1}: 0.00063978ms, 0.00548375% main:@1443 = slice[axes={0, 1},starts={0, 25},ends={64, 41}](TxEdCcekGS:0) -> float_type, {64, 16}, {49, 1}: 0.00067412ms, 0.00577809% main:@1444 = unsqueeze[axes={1},steps={}](main:@1443) -> float_type, {64, 1, 16}, {49, 16, 1}: 0.0008523ms, 0.00730532% 2Yy3zxNGBC:0 = @param:2Yy3zxNGBC:0 -> float_type, {64, 45}, {45, 1}: 0.00054684ms, 0.00468713% main:@1446 = slice[axes={0, 1},starts={0, 17},ends={64, 33}](2Yy3zxNGBC:0) -> float_type, {64, 16}, {45, 1}: 0.0008176ms, 0.0070079% main:@1447 = unsqueeze[axes={1},steps={}](main:@1446) -> float_type, {64, 1, 16}, {45, 16, 1}: 0.00059236ms, 0.0050773% main:@1448 = slice[axes={0, 1},starts={0, 25},ends={64, 41}](LKAvFme9ZS:0) -> float_type, {64, 16}, {49, 1}: 0.00065996ms, 0.00565672% main:@1449 = unsqueeze[axes={1},steps={}](main:@1448) -> float_type, {64, 1, 16}, {49, 16, 1}: 0.00057972ms, 0.00496896% WDLoyKsiA7:0 = @param:WDLoyKsiA7:0 -> float_type, {64, 33}, {33, 1}: 0.0005211ms, 0.00446651% main:@1451 = slice[axes={0, 1},starts={0, 17},ends={64, 33}](WDLoyKsiA7:0) -> float_type, {64, 16}, {33, 1}: 0.000763ms, 0.00653991% main:@1452 = unsqueeze[axes={1},steps={}](main:@1451) -> float_type, {64, 1, 16}, {33, 16, 1}: 0.000572ms, 0.00490279% z3lqgZM8WU:0 = @param:z3lqgZM8WU:0 -> float_type, {64, 41}, {41, 1}: 0.0004963ms, 0.00425394% main:@1454 = slice[axes={0, 1},starts={0, 17},ends={64, 33}](z3lqgZM8WU:0) -> float_type, {64, 16}, {41, 1}: 0.0006206ms, 0.00531935% main:@1455 = unsqueeze[axes={1},steps={}](main:@1454) -> float_type, {64, 1, 16}, {41, 16, 1}: 0.00057084ms, 0.00489284% pY0t1rI5WU:0 = @param:pY0t1rI5WU:0 -> float_type, {64, 41}, {41, 1}: 0.00062738ms, 0.00537747% main:@1457 = slice[axes={0, 1},starts={0, 17},ends={64, 33}](pY0t1rI5WU:0) -> float_type, {64, 16}, {41, 1}: 0.00080982ms, 0.00694122% main:@1458 = unsqueeze[axes={1},steps={}](main:@1457) -> float_type, {64, 1, 16}, {41, 16, 1}: 0.00057044ms, 0.00488942% sZdFJI9nwg:0 = @param:sZdFJI9nwg:0 -> float_type, {64, 41}, {41, 1}: 0.00061144ms, 0.00524084% main:@1460 = slice[axes={0, 1},starts={0, 17},ends={64, 33}](sZdFJI9nwg:0) -> float_type, {64, 16}, {41, 1}: 0.00061542ms, 0.00527495% main:@1461 = unsqueeze[axes={1},steps={}](main:@1460) -> float_type, {64, 1, 16}, {41, 16, 1}: 0.0005663ms, 0.00485393% VfFM90O7LN:0 = @param:VfFM90O7LN:0 -> float_type, {64, 41}, {41, 1}: 0.00056084ms, 0.00480713% main:@1463 = slice[axes={0, 1},starts={0, 17},ends={64, 33}](VfFM90O7LN:0) -> float_type, {64, 16}, {41, 1}: 0.00080246ms, 0.00687813% main:@1464 = unsqueeze[axes={1},steps={}](main:@1463) -> float_type, {64, 1, 16}, {41, 16, 1}: 0.00055626ms, 0.00476788% main:@1465 = reshape_lazy[dims={-1, 10, 16}](main:@1441) -> float_type, {64, 10, 16}, {160, 16, 1}: 0.0005326ms, 0.00456508% main:@1466 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=2048,local=256,](main:@1452,main:@1455,main:@1449,main:@1458,main:@1444,main:@1461,main:@1464,main:@1447,main:@1465,main:@1442) -> float_type, {64, 18, 16}, {288, 16, 1}: 0.0252267ms, 0.216226% main:@1467 = transpose[permutation={0, 2, 1}](main:@1466) -> float_type, {64, 16, 18}, {288, 1, 16}: 0.00073846ms, 0.00632957% main:@1468 = load[offset=3515904,end=3589632](main:@1) -> float_type, {64, 16, 18}, {288, 18, 1}: 0.00051906ms, 0.00444902% main:@1469 = gpu::code_object[code_object=13528,symbol_name=contiguous_kernel,global=18432,local=256,](main:@1467,main:@1468) -> float_type, {64, 16, 18}, {288, 18, 1}: 0.0158293ms, 0.135678% main:@1470 = hip::hip_copy_literal[id=main:@literal:133] -> float_type, {18, 4}, {4, 1}: 0.00074636ms, 0.00639728% main:@1471 = load[offset=3442176,end=3458560](main:@1) -> float_type, {1024, 4}, {4, 1}: 0.00046878ms, 0.00401806% main:@1472 = reshape_lazy[dims={1024, 18}](main:@1469) -> float_type, {1024, 18}, {18, 1}: 0.00056374ms, 0.00483199% main:@1473 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1472,main:@1470,main:@1471) -> float_type, {1024, 4}, {4, 1}: 0.020855ms, 0.178755% main:@1474 = load[offset=3458560,end=3560960](main:@1) -> float_type, {64, 25, 16}, {400, 16, 1}: 0.00063768ms, 0.00546575% main:@1475 = slice[axes={0, 1},starts={0, 17},ends={64, 33}](2CnSyUBzm0:0) -> float_type, {64, 16}, {65, 1}: 0.00064652ms, 0.00554152% main:@1476 = unsqueeze[axes={1},steps={}](main:@1475) -> float_type, {64, 1, 16}, {65, 16, 1}: 0.00098752ms, 0.00846434% main:@1477 = slice[axes={0, 1},starts={0, 17},ends={64, 33}](zIoqYe9cRb:0) -> float_type, {64, 16}, {45, 1}: 0.00076252ms, 0.00653579% main:@1478 = unsqueeze[axes={1},steps={}](main:@1477) -> float_type, {64, 1, 16}, {45, 16, 1}: 0.00057712ms, 0.00494667% main:@1479 = slice[axes={0, 1},starts={0, 17},ends={64, 33}](mgnGNH4Mz3:0) -> float_type, {64, 16}, {49, 1}: 0.00062696ms, 0.00537387% main:@1480 = unsqueeze[axes={1},steps={}](main:@1479) -> float_type, {64, 1, 16}, {49, 16, 1}: 0.00092572ms, 0.00793463% main:@1481 = slice[axes={0, 1},starts={0, 17},ends={64, 33}](6HSQqXdJzb:0) -> float_type, {64, 16}, {45, 1}: 0.0007776ms, 0.00666505% main:@1482 = unsqueeze[axes={1},steps={}](main:@1481) -> float_type, {64, 1, 16}, {45, 16, 1}: 0.00055874ms, 0.00478913% main:@1483 = slice[axes={0, 1},starts={0, 17},ends={64, 33}](isCn4mEOSt:0) -> float_type, {64, 16}, {49, 1}: 0.00061852ms, 0.00530152% main:@1484 = unsqueeze[axes={1},steps={}](main:@1483) -> float_type, {64, 1, 16}, {49, 16, 1}: 0.00090664ms, 0.00777109% main:@1485 = slice[axes={0, 1},starts={0, 17},ends={64, 33}](Dy3rwVmn1A:0) -> float_type, {64, 16}, {41, 1}: 0.00077682ms, 0.00665836% main:@1486 = unsqueeze[axes={1},steps={}](main:@1485) -> float_type, {64, 1, 16}, {41, 16, 1}: 0.0006419ms, 0.00550192% main:@1487 = slice[axes={0, 1},starts={0, 17},ends={64, 33}](FcmWLZhDgE:0) -> float_type, {64, 16}, {57, 1}: 0.00060922ms, 0.00522181% main:@1488 = unsqueeze[axes={1},steps={}](main:@1487) -> float_type, {64, 1, 16}, {57, 16, 1}: 0.00090188ms, 0.00773029% main:@1489 = slice[axes={0, 1},starts={0, 17},ends={64, 33}](ahAlEtBvzo:0) -> float_type, {64, 16}, {41, 1}: 0.00078752ms, 0.00675008% main:@1490 = unsqueeze[axes={1},steps={}](main:@1489) -> float_type, {64, 1, 16}, {41, 16, 1}: 0.00055632ms, 0.00476839% main:@1491 = slice[axes={0, 1},starts={0, 17},ends={64, 33}](dAFYcGrfEl:0) -> float_type, {64, 16}, {45, 1}: 0.00059922ms, 0.0051361% main:@1492 = unsqueeze[axes={1},steps={}](main:@1491) -> float_type, {64, 1, 16}, {45, 16, 1}: 0.00089092ms, 0.00763635% main:@1493 = slice[axes={0, 1},starts={0, 17},ends={64, 33}](X29iyIeu8L:0) -> float_type, {64, 16}, {45, 1}: 0.0007527ms, 0.00645162% main:@1494 = unsqueeze[axes={1},steps={}](main:@1493) -> float_type, {64, 1, 16}, {45, 16, 1}: 0.00055294ms, 0.00473942% main:@1495 = slice[axes={0, 1},starts={0, 17},ends={64, 33}](ZcWERoz0h6:0) -> float_type, {64, 16}, {49, 1}: 0.00061908ms, 0.00530632% main:@1496 = unsqueeze[axes={1},steps={}](main:@1495) -> float_type, {64, 1, 16}, {49, 16, 1}: 0.0009034ms, 0.00774332% main:@1497 = slice[axes={0, 1},starts={0, 17},ends={64, 33}](pYw7fMVoQg:0) -> float_type, {64, 16}, {41, 1}: 0.00069636ms, 0.00596872% main:@1498 = unsqueeze[axes={1},steps={}](main:@1497) -> float_type, {64, 1, 16}, {41, 16, 1}: 0.00055156ms, 0.00472759% main:@1499 = slice[axes={0, 1},starts={0, 17},ends={64, 33}](1aXQU27NrD:0) -> float_type, {64, 16}, {49, 1}: 0.00070264ms, 0.00602254% main:@1500 = unsqueeze[axes={1},steps={}](main:@1499) -> float_type, {64, 1, 16}, {49, 16, 1}: 0.00089414ms, 0.00766395% main:@1501 = reshape_lazy[dims={-1, 12, 16}](main:@1354) -> float_type, {64, 12, 16}, {192, 16, 1}: 0.0004993ms, 0.00427965% main:@1502 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=1828,local=256,](main:@1476,main:@1480,main:@1500,main:@1482,main:@1478,main:@1492,main:@1494,main:@1496,main:@1490,main:@1488,main:@1498,main:@1486,main:@1484,main:@1501,main:@1474) -> float_type, {64, 25, 16}, {400, 16, 1}: 0.0285654ms, 0.244843% main:@1503 = load[offset=9626112,end=9651712](main:@1) -> float_type, {64, 25, 4}, {100, 4, 1}: 0.00066322ms, 0.00568466% main:@1504 = reshape_lazy[dims={64, 16, 4}](main:@1473) -> float_type, {64, 16, 4}, {64, 4, 1}: 0.00052564ms, 0.00450542% main:@1505 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1502,main:@1504,main:@1503) -> float_type, {64, 25, 4}, {100, 4, 1}: 0.0199999ms, 0.171425% main:@1506 = hip::hip_copy_literal[id=main:@literal:2] -> float_type, {64, 4508}, {4508, 1}: 0.0008067ms, 0.00691447% main:@1507 = hip::hip_copy_literal[id=main:@literal:22] -> float_type, {128, 4508}, {4508, 1}: 0.00063492ms, 0.00544209% main:@1508 = slice[axes={1},starts={384},ends={512}](main:@717) -> float_type, {64, 128}, {896, 1}: 0.00063388ms, 0.00543318% main:@1509 = load[offset=3393024,end=4547072](main:@1) -> float_type, {64, 4508}, {4508, 1}: 0.00046276ms, 0.00396646% main:@1510 = gpu::gemm[alpha=1,beta=1,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1508,main:@1507,main:@1506,main:@1509) -> float_type, {64, 4508}, {4508, 1}: 0.0281552ms, 0.241327% main:@1511 = load[offset=14009856,end=15648256](main:@1) -> float_type, {2, 64, 50, 64}, {204800, 3200, 64, 1}: 0.00070544ms, 0.00604654% main:@1512 = slice[axes={1},starts={128},ends={256}](main:@752) -> float_type, {3200, 128}, {256, 1}: 0.00062166ms, 0.00532844% main:@1513 = reshape_lazy[dims={64, 50, 128}](main:@1512) -> float_type, {64, 50, 128}, {12800, 256, 1}: 0.00098038ms, 0.00840314% main:@1514 = slice[axes={2},starts={0},ends={64}](main:@1513) -> float_type, {64, 50, 64}, {12800, 256, 1}: 0.0007127ms, 0.00610877% main:@1515 = unsqueeze[axes={0},steps={}](main:@1514) -> float_type, {1, 64, 50, 64}, {819200, 12800, 256, 1}: 0.00101042ms, 0.00866062% main:@1516 = slice[axes={2},starts={64},ends={128}](main:@1513) -> float_type, {64, 50, 64}, {12800, 256, 1}: 0.0006268ms, 0.0053725% main:@1517 = unsqueeze[axes={0},steps={}](main:@1516) -> float_type, {1, 64, 50, 64}, {819200, 12800, 256, 1}: 0.00065154ms, 0.00558455% main:@1518 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=102400,local=256,](main:@1515,main:@1517,main:@1511) -> float_type, {2, 64, 50, 64}, {204800, 3200, 64, 1}: 0.0200931ms, 0.172224% main:@1519 = load[offset=15648256,end=15681024](main:@1) -> float_type, {2, 64, 1, 64}, {4096, 64, 64, 1}: 0.00067756ms, 0.00580757% main:@1520 = slice[axes={1},starts={128},ends={256}](main:@1293) -> float_type, {64, 128}, {256, 1}: 0.0006586ms, 0.00564506% main:@1521 = reshape_lazy[dims={64, 1, 128}](main:@1520) -> float_type, {64, 1, 128}, {256, 128, 1}: 0.00057544ms, 0.00493227% main:@1522 = slice[axes={2},starts={0},ends={64}](main:@1521) -> float_type, {64, 1, 64}, {256, 128, 1}: 0.00074944ms, 0.00642368% main:@1523 = unsqueeze[axes={0},steps={}](main:@1522) -> float_type, {1, 64, 1, 64}, {16384, 256, 128, 1}: 0.00062688ms, 0.00537318% main:@1524 = slice[axes={2},starts={64},ends={128}](main:@1521) -> float_type, {64, 1, 64}, {256, 128, 1}: 0.00067048ms, 0.00574689% main:@1525 = unsqueeze[axes={0},steps={}](main:@1524) -> float_type, {1, 64, 1, 64}, {16384, 256, 128, 1}: 0.00097886ms, 0.00839011% main:@1526 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=2048,local=256,](main:@1523,main:@1525,main:@1519) -> float_type, {2, 64, 1, 64}, {4096, 64, 64, 1}: 0.0137906ms, 0.118203% main:@1527 = transpose[permutation={0, 1, 3, 2}](main:@1518) -> float_type, {2, 64, 64, 50}, {204800, 3200, 1, 64}: 0.00069464ms, 0.00595397% main:@1528 = load[offset=15681024,end=15706624](main:@1) -> float_type, {2, 64, 1, 50}, {3200, 50, 50, 1}: 0.00047674ms, 0.00408628% main:@1529 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1526,main:@1527,main:@1528) -> float_type, {2, 64, 1, 50}, {3200, 50, 50, 1}: 0.0224582ms, 0.192496% main:@1530 = load[offset=14009856,end=14022656](main:@1) -> float_type, {64, 1, 50}, {50, 50, 1}: 0.00062874ms, 0.00538912% WjNeBZroCJ:0 = @param:WjNeBZroCJ:0 -> int32_type, {64, 50}, {50, 1}: 0.00054908ms, 0.00470633% main:@1532 = unsqueeze[axes={1},steps={}](WjNeBZroCJ:0) -> int32_type, {64, 1, 50}, {50, 50, 1}: 0.00060394ms, 0.00517656% main:@1533 = gpu::code_object[code_object=13536,symbol_name=convert_pow_add_kernel,global=1600,local=256,](main:@1532,main:@1530) -> float_type, {64, 1, 50}, {50, 50, 1}: 0.0129211ms, 0.110751% main:@1534 = load[offset=14035456,end=14061056](main:@1) -> float_type, {2, 64, 1, 50}, {3200, 50, 50, 1}: 0.0006881ms, 0.00589792% main:@1535 = multibroadcast[out_lens={2, 64, 1, 50}](main:@1533) -> float_type, {2, 64, 1, 50}, {0, 50, 50, 1}: 0.00059022ms, 0.00505896% main:@1536 = gpu::code_object[code_object=13520,symbol_name=pow_mul_kernel,global=3200,local=256,](main:@1529,main:@1535,main:@1534) -> float_type, {2, 64, 1, 50}, {3200, 50, 50, 1}: 0.0154345ms, 0.132294% main:@1537 = load[offset=14009856,end=14035456](main:@1) -> float_type, {2, 64, 1, 50}, {3200, 50, 50, 1}: 0.00063648ms, 0.00545547% main:@1538 = gpu::code_object[code_object=14584,symbol_name=softmax_kernel,global=8192,local=64,](main:@1536,main:@1537) -> float_type, {2, 64, 1, 50}, {3200, 50, 50, 1}: 0.0179338ms, 0.153716% main:@1539 = load[offset=14035456,end=15673856](main:@1) -> float_type, {2, 64, 50, 64}, {204800, 3200, 64, 1}: 0.00063066ms, 0.00540558% main:@1540 = slice[axes={1},starts={0},ends={128}](main:@752) -> float_type, {3200, 128}, {256, 1}: 0.00062298ms, 0.00533975% main:@1541 = reshape_lazy[dims={64, 50, 128}](main:@1540) -> float_type, {64, 50, 128}, {12800, 256, 1}: 0.00094696ms, 0.00811668% main:@1542 = slice[axes={2},starts={0},ends={64}](main:@1541) -> float_type, {64, 50, 64}, {12800, 256, 1}: 0.00065402ms, 0.00560581% main:@1543 = unsqueeze[axes={0},steps={}](main:@1542) -> float_type, {1, 64, 50, 64}, {819200, 12800, 256, 1}: 0.00094156ms, 0.0080704% main:@1544 = slice[axes={2},starts={64},ends={128}](main:@1541) -> float_type, {64, 50, 64}, {12800, 256, 1}: 0.000648ms, 0.00555421% main:@1545 = unsqueeze[axes={0},steps={}](main:@1544) -> float_type, {1, 64, 50, 64}, {819200, 12800, 256, 1}: 0.0006445ms, 0.00552421% main:@1546 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=102400,local=256,](main:@1543,main:@1545,main:@1539) -> float_type, {2, 64, 50, 64}, {204800, 3200, 64, 1}: 0.0197701ms, 0.169456% main:@1547 = load[offset=6059520,end=6092288](main:@1) -> float_type, {2, 64, 1, 64}, {4096, 64, 64, 1}: 0.0006609ms, 0.00566478% main:@1548 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1538,main:@1546,main:@1547) -> float_type, {2, 64, 1, 64}, {4096, 64, 64, 1}: 0.0204824ms, 0.175561% main:@1549 = slice[axes={0, 1},starts={0, 5},ends={64, 9}](LMobYAd5ru:0) -> float_type, {64, 4}, {17, 1}: 0.00084832ms, 0.00727121% main:@1550 = load[offset=6133248,end=6174208](main:@1) -> float_type, {64, 160}, {160, 1}: 0.00054662ms, 0.00468525% main:@1551 = slice[axes={0, 1},starts={0, 3},ends={64, 5}](KasFvpq4IG:0) -> float_type, {64, 2}, {5, 1}: 0.00063918ms, 0.00547861% main:@1552 = slice[axes={0, 1},starts={0, 5},ends={64, 9}](4M92ryD8ux:0) -> float_type, {64, 4}, {9, 1}: 0.00089566ms, 0.00767698% main:@1553 = slice[axes={0, 1},starts={0, 5},ends={64, 9}](4gFa5o2YsJ:0) -> float_type, {64, 4}, {9, 1}: 0.00067356ms, 0.00577329% main:@1554 = slice[axes={0, 1},starts={0, 3},ends={64, 5}](TSqGwrHeo4:0) -> float_type, {64, 2}, {5, 1}: 0.000608ms, 0.00521135% main:@1555 = slice[axes={0, 1},starts={0, 3},ends={64, 5}](6XMrIsBjLY:0) -> float_type, {64, 2}, {13, 1}: 0.00060404ms, 0.00517741% main:@1556 = slice[axes={0, 1},starts={0, 5},ends={64, 9}](72zWmh93Bo:0) -> float_type, {64, 4}, {17, 1}: 0.00074856ms, 0.00641614% main:@1557 = slice[axes={0, 1},starts={0, 5},ends={64, 9}](qzd23aNGIT:0) -> float_type, {64, 4}, {17, 1}: 0.00060806ms, 0.00521187% main:@1558 = slice[axes={0, 1},starts={0, 3},ends={64, 5}](8vM5p4RzXd:0) -> float_type, {64, 2}, {13, 1}: 0.00059784ms, 0.00512427% main:@1559 = slice[axes={0, 1},starts={0, 5},ends={64, 9}](9dgeU4bKka:0) -> float_type, {64, 4}, {17, 1}: 0.00063238ms, 0.00542032% main:@1560 = slice[axes={0, 1},starts={0, 9},ends={64, 17}](AB6lekmMxa:0) -> float_type, {64, 8}, {25, 1}: 0.00074874ms, 0.00641768% main:@1561 = slice[axes={0, 1},starts={0, 5},ends={64, 9}](eQquakfWBY:0) -> float_type, {64, 4}, {17, 1}: 0.00059896ms, 0.00513387% main:@1562 = slice[axes={0, 1},starts={0, 3},ends={64, 5}](BYyjkPCX0m:0) -> float_type, {64, 2}, {13, 1}: 0.00058674ms, 0.00502913% main:@1563 = slice[axes={0, 1},starts={0, 5},ends={64, 9}](bTeX85HPmJ:0) -> float_type, {64, 4}, {17, 1}: 0.00063074ms, 0.00540627% main:@1564 = slice[axes={0, 1},starts={0, 3},ends={64, 5}](IRem4MiFDP:0) -> float_type, {64, 2}, {5, 1}: 0.00077588ms, 0.00665031% main:@1565 = slice[axes={0, 1},starts={0, 3},ends={64, 5}](fMiQsYNnet:0) -> float_type, {64, 2}, {5, 1}: 0.0006108ms, 0.00523535% main:@1566 = slice[axes={0, 1},starts={0, 5},ends={64, 9}](EKpP4xMXu8:0) -> float_type, {64, 4}, {17, 1}: 0.00059984ms, 0.00514141% main:@1567 = slice[axes={0, 1},starts={0, 3},ends={64, 5}](EZBIAqn6LS:0) -> float_type, {64, 2}, {13, 1}: 0.00060524ms, 0.0051877% main:@1568 = slice[axes={0, 1},starts={0, 5},ends={64, 9}](EUciqFuptg:0) -> float_type, {64, 4}, {17, 1}: 0.00067002ms, 0.00574295% main:@1569 = slice[axes={0, 1},starts={0, 9},ends={64, 17}](Vbv2Om3ptd:0) -> float_type, {64, 8}, {17, 1}: 0.00059936ms, 0.0051373% main:@1570 = slice[axes={0, 1},starts={0, 3},ends={64, 5}](Hnjq487SIY:0) -> float_type, {64, 2}, {13, 1}: 0.00069376ms, 0.00594643% main:@1571 = slice[axes={0, 1},starts={0, 5},ends={64, 9}](JCVMzpWyri:0) -> float_type, {64, 4}, {17, 1}: 0.00065278ms, 0.00559518% main:@1572 = slice[axes={0, 1},starts={0, 9},ends={64, 17}](JSPsGophtx:0) -> float_type, {64, 8}, {25, 1}: 0.00067968ms, 0.00582575% main:@1573 = slice[axes={0, 1},starts={0, 5},ends={64, 9}](JaHTUxB6zL:0) -> float_type, {64, 4}, {17, 1}: 0.00064602ms, 0.00553724% main:@1574 = slice[axes={0, 1},starts={0, 3},ends={64, 5}](JjyOIxALzM:0) -> float_type, {64, 2}, {5, 1}: 0.00063764ms, 0.00546541% main:@1575 = slice[axes={0, 1},starts={0, 5},ends={64, 9}](khJcIxW3zv:0) -> float_type, {64, 4}, {9, 1}: 0.000628ms, 0.00538278% main:@1576 = slice[axes={0, 1},starts={0, 9},ends={64, 17}](Tej5M3O6As:0) -> float_type, {64, 8}, {17, 1}: 0.00066646ms, 0.00571243% main:@1577 = slice[axes={0, 1},starts={0, 3},ends={64, 5}](r1WGP7Xe9H:0) -> float_type, {64, 2}, {13, 1}: 0.00062416ms, 0.00534987% main:@1578 = slice[axes={0, 1},starts={0, 3},ends={64, 5}](XPcHGKbop2:0) -> float_type, {64, 2}, {13, 1}: 0.00065414ms, 0.00560683% main:@1579 = slice[axes={0, 1},starts={0, 3},ends={64, 5}](lQmygxs5aD:0) -> float_type, {64, 2}, {13, 1}: 0.00059664ms, 0.00511398% main:@1580 = slice[axes={0, 1},starts={0, 9},ends={64, 17}](d8RoEzG52p:0) -> float_type, {64, 8}, {17, 1}: 0.00063578ms, 0.00544947% main:@1581 = slice[axes={0, 1},starts={0, 9},ends={64, 17}](omr2g4XaH5:0) -> float_type, {64, 8}, {17, 1}: 0.00061226ms, 0.00524787% main:@1582 = slice[axes={0, 1},starts={0, 3},ends={64, 5}](agDiJj8lLK:0) -> float_type, {64, 2}, {13, 1}: 0.00065982ms, 0.00565552% main:@1583 = slice[axes={0, 1},starts={0, 9},ends={64, 17}](aqMQBgb3hf:0) -> float_type, {64, 8}, {17, 1}: 0.00060308ms, 0.00516918% main:@1584 = slice[axes={0, 1},starts={0, 5},ends={64, 9}](cN2JSTnFOu:0) -> float_type, {64, 4}, {9, 1}: 0.00065794ms, 0.00563941% main:@1585 = slice[axes={0, 1},starts={0, 5},ends={64, 9}](iCbkMrmXBh:0) -> float_type, {64, 4}, {9, 1}: 0.00064394ms, 0.00551941% main:@1586 = slice[axes={0, 1},starts={0, 9},ends={64, 17}](nyLxrsTmOB:0) -> float_type, {64, 8}, {17, 1}: 0.00063686ms, 0.00545872% main:@1587 = slice[axes={0, 1},starts={0, 9},ends={64, 17}](fZhaL2tCm6:0) -> float_type, {64, 8}, {17, 1}: 0.00060836ms, 0.00521444% main:@1588 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=269,local=256,](main:@1574,main:@1567,main:@1564,main:@1579,main:@1554,main:@1577,main:@1582,main:@1578,main:@1558,main:@1565,main:@1562,main:@1570,main:@1551,main:@1555,main:@1584,main:@1559,main:@1552,main:@1549,main:@1575,main:@1568,main:@1563,main:@1566,main:@1561,main:@1571,main:@1556,main:@1557,main:@1585,main:@1573,main:@1553,main:@1572,main:@1587,main:@1580,main:@1576,main:@1581,main:@1560,main:@1583,main:@1569,main:@1586,main:@1550) -> float_type, {64, 160}, {160, 1}: 0.0454336ms, 0.389425% main:@1589 = load[offset=6092288,end=6133248](main:@1) -> float_type, {64, 160}, {160, 1}: 0.00068202ms, 0.0058458% main:@1590 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1588,main:@873,main:@1589) -> float_type, {64, 160}, {160, 1}: 0.0210668ms, 0.18057% main:@1591 = load[offset=6133248,end=6176768](main:@1) -> float_type, {64, 170}, {170, 1}: 0.00062388ms, 0.00534747% main:@1592 = slice[axes={0, 1},starts={0, 1},ends={64, 3}](1QWFmJ96aB:0) -> float_type, {64, 2}, {5, 1}: 0.00065852ms, 0.00564438% main:@1593 = slice[axes={0, 1},starts={0, 1},ends={64, 3}](3VED7WSJ1b:0) -> float_type, {64, 2}, {5, 1}: 0.00065202ms, 0.00558866% main:@1594 = slice[axes={0, 1},starts={0, 1},ends={64, 3}](5yColzZ8K3:0) -> float_type, {64, 2}, {5, 1}: 0.0009555ms, 0.00818988% main:@1595 = slice[axes={0, 1},starts={0, 1},ends={64, 3}](6Wh2Jm7GHC:0) -> float_type, {64, 2}, {5, 1}: 0.00064298ms, 0.00551118% main:@1596 = slice[axes={0, 1},starts={0, 1},ends={64, 3}](zX6xKOWG9S:0) -> float_type, {64, 2}, {5, 1}: 0.00062794ms, 0.00538227% main:@1597 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](gdck97AK4w:0) -> float_type, {64, 4}, {17, 1}: 0.00059858ms, 0.00513061% main:@1598 = slice[axes={0, 1},starts={0, 1},ends={64, 3}](NLt5JnRMjG:0) -> float_type, {64, 2}, {5, 1}: 0.00090814ms, 0.00778395% main:@1599 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](Cw59eOo7M1:0) -> float_type, {64, 4}, {9, 1}: 0.00061898ms, 0.00530547% main:@1600 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](E1ZkVIXqS5:0) -> float_type, {64, 4}, {9, 1}: 0.00062444ms, 0.00535227% main:@1601 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](EhUjmCTaGW:0) -> float_type, {64, 4}, {9, 1}: 0.00060962ms, 0.00522524% main:@1602 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](Pu1O5qFxY6:0) -> float_type, {64, 4}, {17, 1}: 0.0009039ms, 0.0077476% main:@1603 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](FWhBxj4Tun:0) -> float_type, {64, 4}, {9, 1}: 0.00070356ms, 0.00603043% main:@1604 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](yVxKAL4U5G:0) -> float_type, {64, 4}, {9, 1}: 0.00056986ms, 0.00488444% main:@1605 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](HmATr3gKsW:0) -> float_type, {64, 4}, {9, 1}: 0.00059446ms, 0.0050953% main:@1606 = slice[axes={0, 1},starts={0, 1},ends={64, 3}](J3H2EjAXt8:0) -> float_type, {64, 2}, {5, 1}: 0.00088424ms, 0.00757909% main:@1607 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](IMdFk7D16X:0) -> float_type, {64, 4}, {9, 1}: 0.00060666ms, 0.00519987% main:@1608 = slice[axes={0, 1},starts={0, 1},ends={64, 3}](IxpbBg7JTO:0) -> float_type, {64, 2}, {5, 1}: 0.00061638ms, 0.00528318% main:@1609 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](ZRYNFfjSez:0) -> float_type, {64, 8}, {33, 1}: 0.00061142ms, 0.00524067% main:@1610 = slice[axes={0, 1},starts={0, 1},ends={64, 3}](vAbKepTjQg:0) -> float_type, {64, 2}, {5, 1}: 0.00081242ms, 0.0069635% main:@1611 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](KCNfS4GlpP:0) -> float_type, {64, 4}, {9, 1}: 0.00058952ms, 0.00505296% main:@1612 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](O5aJeUQh2G:0) -> float_type, {64, 8}, {17, 1}: 0.00058642ms, 0.00502639% main:@1613 = slice[axes={0, 1},starts={0, 1},ends={64, 3}](gqk5byC8aJ:0) -> float_type, {64, 2}, {5, 1}: 0.00059038ms, 0.00506033% main:@1614 = slice[axes={0, 1},starts={0, 1},ends={64, 3}](NJBt1ubXFW:0) -> float_type, {64, 2}, {5, 1}: 0.00073338ms, 0.00628603% main:@1615 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](Ouh1XpPZaj:0) -> float_type, {64, 4}, {9, 1}: 0.0005827ms, 0.0049945% main:@1616 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](gsP96ZTbn0:0) -> float_type, {64, 8}, {17, 1}: 0.0005942ms, 0.00509307% main:@1617 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](mTMtDEo3r8:0) -> float_type, {64, 4}, {9, 1}: 0.000575ms, 0.0049285% main:@1618 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](U6GqhP8k7I:0) -> float_type, {64, 4}, {9, 1}: 0.00073456ms, 0.00629614% main:@1619 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](UMO0o14RuI:0) -> float_type, {64, 4}, {9, 1}: 0.00059904ms, 0.00513456% main:@1620 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](Uc42tiuavR:0) -> float_type, {64, 4}, {9, 1}: 0.00062804ms, 0.00538312% main:@1621 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](oQ5OcjZJe8:0) -> float_type, {64, 8}, {17, 1}: 0.00062762ms, 0.00537952% main:@1622 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](qfwYOHde3u:0) -> float_type, {64, 8}, {17, 1}: 0.0007549ms, 0.00647048% main:@1623 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](mNgQ4ohWqU:0) -> float_type, {64, 4}, {9, 1}: 0.00062894ms, 0.00539084% main:@1624 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](nltbg9ZWQX:0) -> float_type, {64, 4}, {17, 1}: 0.00059954ms, 0.00513884% main:@1625 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](tIGjbcEuDd:0) -> float_type, {64, 4}, {9, 1}: 0.00058826ms, 0.00504216% main:@1626 = slice[axes={0, 1},starts={0, 1},ends={64, 3}](ecPTQUALkq:0) -> float_type, {64, 2}, {5, 1}: 0.000686ms, 0.00587992% main:@1627 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](magwfJSnUu:0) -> float_type, {64, 4}, {9, 1}: 0.00057752ms, 0.0049501% main:@1628 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](xVErSbJ0hg:0) -> float_type, {64, 4}, {9, 1}: 0.0007021ms, 0.00601791% main:@1629 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](l4bReWtsMB:0) -> float_type, {64, 8}, {25, 1}: 0.00064606ms, 0.00553758% main:@1630 = slice[axes={0, 1},starts={0, 1},ends={64, 3}](pQsi1GNxvj:0) -> float_type, {64, 2}, {5, 1}: 0.00066046ms, 0.00566101% main:@1631 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](lDhFNYzbxR:0) -> float_type, {64, 8}, {17, 1}: 0.00064142ms, 0.00549781% main:@1632 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](l5WSX3VvMK:0) -> float_type, {64, 8}, {25, 1}: 0.00063898ms, 0.00547689% main:@1633 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=265,local=256,](main:@1595,main:@1626,main:@1614,main:@1610,main:@1596,main:@1630,main:@1598,main:@1593,main:@1592,main:@1606,main:@1613,main:@1608,main:@1594,main:@1601,main:@1618,main:@1600,main:@1599,main:@1611,main:@1620,main:@1607,main:@1623,main:@1617,main:@1604,main:@1597,main:@1602,main:@1625,main:@1628,main:@1619,main:@1615,main:@1624,main:@1605,main:@1627,main:@1603,main:@1622,main:@1631,main:@1616,main:@1612,main:@1609,main:@1629,main:@1632,main:@1621,main:@1591) -> float_type, {64, 170}, {170, 1}: 0.044543ms, 0.381792% main:@1634 = hip::hip_copy_literal[id=main:@literal:101] -> float_type, {170, 160}, {160, 1}: 0.00088062ms, 0.00754806% main:@1635 = load[offset=6176768,end=6217728](main:@1) -> float_type, {64, 160}, {160, 1}: 0.0005087ms, 0.00436022% main:@1636 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1633,main:@1634,main:@1635) -> float_type, {64, 160}, {160, 1}: 0.022645ms, 0.194097% main:@1637 = slice[axes={0, 1},starts={0, 1},ends={64, 17}](2Yy3zxNGBC:0) -> float_type, {64, 16}, {45, 1}: 0.00085394ms, 0.00731938% main:@1638 = unsqueeze[axes={1},steps={}](main:@1637) -> float_type, {64, 1, 16}, {45, 16, 1}: 0.00078622ms, 0.00673893% main:@1639 = slice[axes={0, 1},starts={0, 1},ends={64, 17}](WDLoyKsiA7:0) -> float_type, {64, 16}, {33, 1}: 0.00077088ms, 0.00660745% main:@1640 = unsqueeze[axes={1},steps={}](main:@1639) -> float_type, {64, 1, 16}, {33, 16, 1}: 0.00058186ms, 0.0049873% main:@1641 = slice[axes={0, 1},starts={0, 9},ends={64, 25}](LKAvFme9ZS:0) -> float_type, {64, 16}, {49, 1}: 0.00062736ms, 0.0053773% main:@1642 = unsqueeze[axes={1},steps={}](main:@1641) -> float_type, {64, 1, 16}, {49, 16, 1}: 0.00073988ms, 0.00634174% main:@1643 = slice[axes={0, 1},starts={0, 1},ends={64, 17}](VfFM90O7LN:0) -> float_type, {64, 16}, {41, 1}: 0.00074166ms, 0.006357% main:@1644 = unsqueeze[axes={1},steps={}](main:@1643) -> float_type, {64, 1, 16}, {41, 16, 1}: 0.00058768ms, 0.00503719% main:@1645 = slice[axes={0, 1},starts={0, 9},ends={64, 25}](TxEdCcekGS:0) -> float_type, {64, 16}, {49, 1}: 0.00062016ms, 0.00531558% main:@1646 = unsqueeze[axes={1},steps={}](main:@1645) -> float_type, {64, 1, 16}, {49, 16, 1}: 0.00070128ms, 0.00601089% main:@1647 = slice[axes={0, 1},starts={0, 1},ends={64, 17}](z3lqgZM8WU:0) -> float_type, {64, 16}, {41, 1}: 0.00073568ms, 0.00630574% main:@1648 = unsqueeze[axes={1},steps={}](main:@1647) -> float_type, {64, 1, 16}, {41, 16, 1}: 0.00058462ms, 0.00501096% main:@1649 = slice[axes={0, 1},starts={0, 1},ends={64, 17}](pY0t1rI5WU:0) -> float_type, {64, 16}, {41, 1}: 0.00061618ms, 0.00528147% main:@1650 = unsqueeze[axes={1},steps={}](main:@1649) -> float_type, {64, 1, 16}, {41, 16, 1}: 0.00071428ms, 0.00612231% main:@1651 = slice[axes={0, 1},starts={0, 1},ends={64, 17}](sZdFJI9nwg:0) -> float_type, {64, 16}, {41, 1}: 0.00076876ms, 0.00658928% main:@1652 = unsqueeze[axes={1},steps={}](main:@1651) -> float_type, {64, 1, 16}, {41, 16, 1}: 0.00055038ms, 0.00471748% main:@1653 = load[offset=6217728,end=6291456](main:@1) -> float_type, {64, 18, 16}, {288, 16, 1}: 0.00051764ms, 0.00443685% main:@1654 = reshape_lazy[dims={-1, 10, 16}](main:@1636) -> float_type, {64, 10, 16}, {160, 16, 1}: 0.00052134ms, 0.00446857% main:@1655 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=2048,local=256,](main:@1640,main:@1648,main:@1642,main:@1650,main:@1646,main:@1652,main:@1644,main:@1638,main:@1654,main:@1653) -> float_type, {64, 18, 16}, {288, 16, 1}: 0.024619ms, 0.211017% main:@1656 = hip::hip_copy_literal[id=main:@literal:129] -> float_type, {18, 4}, {4, 1}: 0.00080262ms, 0.0068795% main:@1657 = load[offset=6133248,end=6206976](main:@1) -> float_type, {64, 16, 18}, {288, 18, 1}: 0.0005043ms, 0.00432251% main:@1658 = transpose[permutation={0, 2, 1}](main:@1655) -> float_type, {64, 16, 18}, {288, 1, 16}: 0.00054226ms, 0.00464788% main:@1659 = gpu::code_object[code_object=13528,symbol_name=contiguous_kernel,global=18432,local=256,](main:@1658,main:@1657) -> float_type, {64, 16, 18}, {288, 18, 1}: 0.0162548ms, 0.139325% main:@1660 = load[offset=6043136,end=6059520](main:@1) -> float_type, {1024, 4}, {4, 1}: 0.00061648ms, 0.00528404% main:@1661 = reshape_lazy[dims={1024, 18}](main:@1659) -> float_type, {1024, 18}, {18, 1}: 0.00054534ms, 0.00467428% main:@1662 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1661,main:@1656,main:@1660) -> float_type, {1024, 4}, {4, 1}: 0.0184467ms, 0.158112% main:@1663 = slice[axes={0, 1},starts={0, 17},ends={64, 33}](2RDXZrx9GA:0) -> float_type, {64, 16}, {45, 1}: 0.00084392ms, 0.0072335% main:@1664 = unsqueeze[axes={1},steps={}](main:@1663) -> float_type, {64, 1, 16}, {45, 16, 1}: 0.00062444ms, 0.00535227% main:@1665 = load[offset=6133248,end=6219264](main:@1) -> float_type, {64, 21, 16}, {336, 16, 1}: 0.00049258ms, 0.00422205% main:@1666 = slice[axes={0, 1},starts={0, 17},ends={64, 33}](2srtUfGvja:0) -> float_type, {64, 16}, {41, 1}: 0.00076996ms, 0.00659956% main:@1667 = unsqueeze[axes={1},steps={}](main:@1666) -> float_type, {64, 1, 16}, {41, 16, 1}: 0.00058842ms, 0.00504353% main:@1668 = slice[axes={0, 1},starts={0, 17},ends={64, 33}](3Laj620MSn:0) -> float_type, {64, 16}, {45, 1}: 0.00064416ms, 0.00552129% main:@1669 = unsqueeze[axes={1},steps={}](main:@1668) -> float_type, {64, 1, 16}, {45, 16, 1}: 0.00055366ms, 0.00474559% main:@1670 = slice[axes={0, 1},starts={0, 17},ends={64, 33}](jznORDJr0N:0) -> float_type, {64, 16}, {45, 1}: 0.00076448ms, 0.00655259% main:@1671 = unsqueeze[axes={1},steps={}](main:@1670) -> float_type, {64, 1, 16}, {45, 16, 1}: 0.00055914ms, 0.00479256% main:@1672 = slice[axes={0, 1},starts={0, 17},ends={64, 33}](HRGzlnEkXQ:0) -> float_type, {64, 16}, {41, 1}: 0.00064252ms, 0.00550724% main:@1673 = unsqueeze[axes={1},steps={}](main:@1672) -> float_type, {64, 1, 16}, {41, 16, 1}: 0.00057752ms, 0.0049501% main:@1674 = slice[axes={0, 1},starts={0, 305},ends={64, 321}](lyiDxfzZg3:0) -> float_type, {64, 16}, {701, 1}: 0.0007346ms, 0.00629648% main:@1675 = unsqueeze[axes={1},steps={}](main:@1674) -> float_type, {64, 1, 16}, {701, 16, 1}: 0.00055414ms, 0.0047497% main:@1676 = slice[axes={0, 1},starts={0, 17},ends={64, 33}](VUqfdN5Fzv:0) -> float_type, {64, 16}, {45, 1}: 0.00060854ms, 0.00521598% main:@1677 = unsqueeze[axes={1},steps={}](main:@1676) -> float_type, {64, 1, 16}, {45, 16, 1}: 0.00055966ms, 0.00479702% main:@1678 = slice[axes={0, 1},starts={0, 17},ends={64, 33}](CHAYRkBvIo:0) -> float_type, {64, 16}, {41, 1}: 0.00077478ms, 0.00664088% main:@1679 = unsqueeze[axes={1},steps={}](main:@1678) -> float_type, {64, 1, 16}, {41, 16, 1}: 0.0005691ms, 0.00487793% main:@1680 = slice[axes={0, 1},starts={0, 17},ends={64, 33}](HzcyUlm8Yj:0) -> float_type, {64, 16}, {45, 1}: 0.00062308ms, 0.00534061% main:@1681 = unsqueeze[axes={1},steps={}](main:@1680) -> float_type, {64, 1, 16}, {45, 16, 1}: 0.00054886ms, 0.00470445% main:@1682 = slice[axes={0, 1},starts={0, 17},ends={64, 33}](duyRoIEkH8:0) -> float_type, {64, 16}, {45, 1}: 0.0007042ms, 0.00603591% main:@1683 = unsqueeze[axes={1},steps={}](main:@1682) -> float_type, {64, 1, 16}, {45, 16, 1}: 0.00056494ms, 0.00484227% main:@1684 = slice[axes={0, 1},starts={0, 17},ends={64, 33}](eQI8ikEMgP:0) -> float_type, {64, 16}, {41, 1}: 0.00060594ms, 0.0051937% main:@1685 = unsqueeze[axes={1},steps={}](main:@1684) -> float_type, {64, 1, 16}, {41, 16, 1}: 0.0005449ms, 0.00467051% main:@1686 = reshape_lazy[dims={-1, 10, 16}](main:@1590) -> float_type, {64, 10, 16}, {160, 16, 1}: 0.00049628ms, 0.00425377% main:@1687 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=1792,local=256,](main:@1671,main:@1673,main:@1675,main:@1664,main:@1677,main:@1679,main:@1681,main:@1667,main:@1669,main:@1683,main:@1685,main:@1686,main:@1665) -> float_type, {64, 21, 16}, {336, 16, 1}: 0.0270848ms, 0.232152% main:@1688 = load[offset=6021632,end=6043136](main:@1) -> float_type, {64, 21, 4}, {84, 4, 1}: 0.0006589ms, 0.00564763% main:@1689 = reshape_lazy[dims={64, 16, 4}](main:@1662) -> float_type, {64, 16, 4}, {64, 4, 1}: 0.00051706ms, 0.00443188% main:@1690 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1687,main:@1689,main:@1688) -> float_type, {64, 21, 4}, {84, 4, 1}: 0.0174527ms, 0.149592% main:@1691 = load[offset=6092288,end=6114816](main:@1) -> float_type, {64, 88}, {88, 1}: 0.00065086ms, 0.00557872% main:@1692 = slice[axes={0, 1},starts={0, 241},ends={64, 273}](uFSOc5Chp2:0) -> float_type, {64, 32}, {481, 1}: 0.00068052ms, 0.00583295% main:@1693 = slice[axes={0, 1},starts={0, 237},ends={64, 253}](nv2tzWRf1I:0) -> float_type, {64, 16}, {301, 1}: 0.00060978ms, 0.00522661% main:@1694 = slice[axes={0, 1},starts={0, 517},ends={64, 525}](lyiDxfzZg3:0) -> float_type, {64, 8}, {701, 1}: 0.0007533ms, 0.00645677% main:@1695 = slice[axes={0, 1},starts={0, 1},ends={64, 33}](n4iptOVFwE:0) -> float_type, {64, 32}, {113, 1}: 0.00064978ms, 0.00556946% main:@1696 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=1408,local=256,](main:@1694,main:@1693,main:@1692,main:@1695,main:@1691) -> float_type, {64, 88}, {88, 1}: 0.0164235ms, 0.140771% main:@1697 = hip::hip_copy_literal[id=main:@literal:58] -> float_type, {88, 32}, {32, 1}: 0.00088874ms, 0.00761766% main:@1698 = load[offset=6043136,end=6051328](main:@1) -> float_type, {64, 32}, {32, 1}: 0.0004789ms, 0.0041048% main:@1699 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1696,main:@1697,main:@1698) -> float_type, {64, 32}, {32, 1}: 0.0205417ms, 0.176069% main:@1700 = slice[axes={1},starts={0},ends={32}](main:@1109) -> float_type, {640, 32}, {64, 1}: 0.0008008ms, 0.0068639% main:@1701 = reshape_lazy[dims={1, 64, 10, 32}](main:@1700) -> float_type, {1, 64, 10, 32}, {40960, 640, 64, 1}: 0.0007574ms, 0.00649191% main:@1702 = transpose[permutation={0, 1, 3, 2}](main:@1701) -> float_type, {1, 64, 32, 10}, {40960, 640, 1, 64}: 0.00059948ms, 0.00513833% main:@1703 = reshape_lazy[dims={1, 64, 1, 32}](main:@1699) -> float_type, {1, 64, 1, 32}, {2048, 32, 32, 1}: 0.00047928ms, 0.00410806% main:@1704 = load[offset=6051328,end=6053888](main:@1) -> float_type, {1, 64, 1, 10}, {640, 10, 10, 1}: 0.0005027ms, 0.0043088% main:@1705 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1703,main:@1702,main:@1704) -> float_type, {1, 64, 1, 10}, {640, 10, 10, 1}: 0.0201189ms, 0.172445% main:@1706 = unsqueeze[axes={1},steps={}](0MZVmrKeC6:0) -> int32_type, {64, 1, 10}, {10, 10, 1}: 0.00074656ms, 0.00639899% main:@1707 = load[offset=6043136,end=6045696](main:@1) -> float_type, {64, 1, 10}, {10, 10, 1}: 0.00051838ms, 0.00444319% main:@1708 = gpu::code_object[code_object=13536,symbol_name=convert_mul_add_kernel,global=320,local=256,](main:@1706,main:@1707) -> float_type, {64, 1, 10}, {10, 10, 1}: 0.0132086ms, 0.113215% main:@1709 = multibroadcast[out_lens={1, 64, 1, 10}](main:@1708) -> float_type, {1, 64, 1, 10}, {0, 10, 10, 1}: 0.00069726ms, 0.00597643% main:@1710 = load[offset=6045696,end=6048256](main:@1) -> float_type, {1, 64, 1, 10}, {640, 10, 10, 1}: 0.00051204ms, 0.00438885% main:@1711 = gpu::code_object[code_object=13520,symbol_name=pow_div_kernel,global=320,local=256,](main:@1705,main:@1709,main:@1710) -> float_type, {1, 64, 1, 10}, {640, 10, 10, 1}: 0.0155916ms, 0.13364% main:@1712 = load[offset=6051328,end=6053888](main:@1) -> float_type, {1, 64, 1, 10}, {640, 10, 10, 1}: 0.00065204ms, 0.00558883% main:@1713 = gpu::code_object[code_object=14576,symbol_name=softmax_kernel,global=4096,local=64,](main:@1711,main:@1712) -> float_type, {1, 64, 1, 10}, {640, 10, 10, 1}: 0.0156893ms, 0.134478% main:@1714 = slice[axes={1},starts={32},ends={64}](main:@1109) -> float_type, {640, 32}, {64, 1}: 0.00077396ms, 0.00663385% main:@1715 = load[offset=6043136,end=6051328](main:@1) -> float_type, {1, 64, 1, 32}, {2048, 32, 32, 1}: 0.0004683ms, 0.00401394% main:@1716 = reshape_lazy[dims={1, 64, 10, 32}](main:@1714) -> float_type, {1, 64, 10, 32}, {40960, 640, 64, 1}: 0.00061348ms, 0.00525833% main:@1717 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1713,main:@1716,main:@1715) -> float_type, {1, 64, 1, 32}, {2048, 32, 32, 1}: 0.0220725ms, 0.18919% main:@1718 = slice[axes={0, 1},starts={0, 525},ends={64, 533}](lyiDxfzZg3:0) -> float_type, {64, 8}, {701, 1}: 0.00084074ms, 0.00720624% main:@1719 = slice[axes={0, 1},starts={0, 41},ends={64, 73}](n4iptOVFwE:0) -> float_type, {64, 32}, {113, 1}: 0.00064918ms, 0.00556432% main:@1720 = slice[axes={0, 1},starts={0, 273},ends={64, 305}](uFSOc5Chp2:0) -> float_type, {64, 32}, {481, 1}: 0.00081524ms, 0.00698767% main:@1721 = slice[axes={0, 1},starts={0, 253},ends={64, 269}](nv2tzWRf1I:0) -> float_type, {64, 16}, {301, 1}: 0.00062944ms, 0.00539512% main:@1722 = load[offset=6100480,end=6123008](main:@1) -> float_type, {64, 88}, {88, 1}: 0.00053258ms, 0.00456491% main:@1723 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=1408,local=256,](main:@1718,main:@1721,main:@1720,main:@1719,main:@1722) -> float_type, {64, 88}, {88, 1}: 0.0160105ms, 0.137231% main:@1724 = hip::hip_copy_literal[id=main:@literal:51] -> float_type, {88, 32}, {32, 1}: 0.00086618ms, 0.00742429% main:@1725 = load[offset=6092288,end=6100480](main:@1) -> float_type, {64, 32}, {32, 1}: 0.00052292ms, 0.00448211% main:@1726 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1723,main:@1724,main:@1725) -> float_type, {64, 32}, {32, 1}: 0.0205763ms, 0.176366% main:@1727 = load[offset=6051328,end=6059008](main:@1) -> float_type, {1, 64, 1, 30}, {1920, 30, 30, 1}: 0.0006566ms, 0.00562792% main:@1728 = slice[axes={1},starts={32},ends={64}](main:@686) -> float_type, {1920, 32}, {64, 1}: 0.00066058ms, 0.00566203% main:@1729 = reshape_lazy[dims={1, 64, 30, 32}](main:@1728) -> float_type, {1, 64, 30, 32}, {122880, 1920, 64, 1}: 0.00094254ms, 0.0080788% main:@1730 = transpose[permutation={0, 1, 3, 2}](main:@1729) -> float_type, {1, 64, 32, 30}, {122880, 1920, 1, 64}: 0.00059136ms, 0.00506873% main:@1731 = reshape_lazy[dims={1, 64, 1, 32}](main:@1726) -> float_type, {1, 64, 1, 32}, {2048, 32, 32, 1}: 0.00050992ms, 0.00437068% main:@1732 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1731,main:@1730,main:@1727) -> float_type, {1, 64, 1, 30}, {1920, 30, 30, 1}: 0.0189532ms, 0.162454% VX8uEBNndW:0 = @param:VX8uEBNndW:0 -> int32_type, {64, 30}, {30, 1}: 0.00072932ms, 0.00625123% main:@1734 = unsqueeze[axes={1},steps={}](VX8uEBNndW:0) -> int32_type, {64, 1, 30}, {30, 30, 1}: 0.00060394ms, 0.00517656% main:@1735 = load[offset=6092288,end=6099968](main:@1) -> float_type, {64, 1, 30}, {30, 30, 1}: 0.00049474ms, 0.00424057% main:@1736 = gpu::code_object[code_object=13528,symbol_name=convert_add_kernel,global=960,local=256,](main:@1734,main:@1735) -> float_type, {64, 1, 30}, {30, 30, 1}: 0.0133036ms, 0.114029% main:@1737 = multibroadcast[out_lens={1, 64, 1, 30}](main:@1736) -> float_type, {1, 64, 1, 30}, {0, 30, 30, 1}: 0.00069096ms, 0.00592243% main:@1738 = load[offset=6099968,end=6107648](main:@1) -> float_type, {1, 64, 1, 30}, {1920, 30, 30, 1}: 0.0004715ms, 0.00404137% main:@1739 = gpu::code_object[code_object=13520,symbol_name=pow_sub_kernel,global=960,local=256,](main:@1732,main:@1737,main:@1738) -> float_type, {1, 64, 1, 30}, {1920, 30, 30, 1}: 0.0151512ms, 0.129866% main:@1740 = load[offset=6092288,end=6099968](main:@1) -> float_type, {1, 64, 1, 30}, {1920, 30, 30, 1}: 0.00062278ms, 0.00533804% main:@1741 = gpu::code_object[code_object=14584,symbol_name=softmax_kernel,global=4096,local=64,](main:@1739,main:@1740) -> float_type, {1, 64, 1, 30}, {1920, 30, 30, 1}: 0.0157989ms, 0.135418% main:@1742 = slice[axes={1},starts={0},ends={32}](main:@686) -> float_type, {1920, 32}, {64, 1}: 0.00078ms, 0.00668562% main:@1743 = reshape_lazy[dims={1, 64, 30, 32}](main:@1742) -> float_type, {1, 64, 30, 32}, {122880, 1920, 64, 1}: 0.0006062ms, 0.00519593% main:@1744 = load[offset=6051328,end=6059520](main:@1) -> float_type, {1, 64, 1, 32}, {2048, 32, 32, 1}: 0.00050418ms, 0.00432148% main:@1745 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1741,main:@1743,main:@1744) -> float_type, {1, 64, 1, 32}, {2048, 32, 32, 1}: 0.0189936ms, 0.1628% main:@1746 = slice[axes={1},starts={96},ends={128}](main:@652) -> float_type, {64, 32}, {128, 1}: 0.00078668ms, 0.00674288% main:@1747 = load[offset=4550912,end=4559104](main:@1) -> float_type, {64, 32}, {32, 1}: 0.00051898ms, 0.00444834% main:@1748 = gpu::code_object[code_object=13528,symbol_name=contiguous_kernel,global=1024,local=256,](main:@1746,main:@1747) -> float_type, {64, 32}, {32, 1}: 0.013078ms, 0.112095% main:@1749 = slice[axes={1},starts={0},ends={32}](main:@666) -> float_type, {960, 32}, {64, 1}: 0.00076538ms, 0.00656031% main:@1750 = reshape_lazy[dims={1, 64, 15, 32}](main:@1749) -> float_type, {1, 64, 15, 32}, {61440, 960, 64, 1}: 0.0005999ms, 0.00514193% main:@1751 = transpose[permutation={0, 1, 3, 2}](main:@1750) -> float_type, {1, 64, 32, 15}, {61440, 960, 1, 64}: 0.00057226ms, 0.00490502% main:@1752 = reshape_lazy[dims={1, 64, 1, 32}](main:@1748) -> float_type, {1, 64, 1, 32}, {2048, 32, 32, 1}: 0.00048992ms, 0.00419925% main:@1753 = load[offset=4547072,end=4550912](main:@1) -> float_type, {1, 64, 1, 15}, {960, 15, 15, 1}: 0.0004615ms, 0.00395566% main:@1754 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1752,main:@1751,main:@1753) -> float_type, {1, 64, 1, 15}, {960, 15, 15, 1}: 0.0210586ms, 0.1805% hzqb8Cpfc1:0 = @param:hzqb8Cpfc1:0 -> int32_type, {64, 15}, {15, 1}: 0.00074756ms, 0.00640757% main:@1756 = unsqueeze[axes={1},steps={}](hzqb8Cpfc1:0) -> int32_type, {64, 1, 15}, {15, 15, 1}: 0.00058374ms, 0.00500341% main:@1757 = load[offset=4554752,end=4558592](main:@1) -> float_type, {64, 1, 15}, {15, 15, 1}: 0.00050266ms, 0.00430845% main:@1758 = gpu::code_object[code_object=13536,symbol_name=convert_mul_add_kernel,global=480,local=256,](main:@1756,main:@1757) -> float_type, {64, 1, 15}, {15, 15, 1}: 0.0132237ms, 0.113344% main:@1759 = load[offset=4550912,end=4554752](main:@1) -> float_type, {1, 64, 1, 15}, {960, 15, 15, 1}: 0.000636ms, 0.00545135% main:@1760 = multibroadcast[out_lens={1, 64, 1, 15}](main:@1758) -> float_type, {1, 64, 1, 15}, {0, 15, 15, 1}: 0.00055668ms, 0.00477148% main:@1761 = gpu::code_object[code_object=13520,symbol_name=add_sub_kernel,global=480,local=256,](main:@1754,main:@1760,main:@1759) -> float_type, {1, 64, 1, 15}, {960, 15, 15, 1}: 0.0131382ms, 0.112612% main:@1762 = load[offset=4555264,end=4559104](main:@1) -> float_type, {1, 64, 1, 15}, {960, 15, 15, 1}: 0.00060772ms, 0.00520895% main:@1763 = gpu::code_object[code_object=14568,symbol_name=softmax_kernel,global=4096,local=64,](main:@1761,main:@1762) -> float_type, {1, 64, 1, 15}, {960, 15, 15, 1}: 0.0155469ms, 0.133257% main:@1764 = load[offset=4547072,end=4555264](main:@1) -> float_type, {1, 64, 1, 32}, {2048, 32, 32, 1}: 0.00064174ms, 0.00550055% main:@1765 = slice[axes={1},starts={32},ends={64}](main:@666) -> float_type, {960, 32}, {64, 1}: 0.00062788ms, 0.00538175% main:@1766 = reshape_lazy[dims={1, 64, 15, 32}](main:@1765) -> float_type, {1, 64, 15, 32}, {61440, 960, 64, 1}: 0.00060304ms, 0.00516884% main:@1767 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1763,main:@1766,main:@1764) -> float_type, {1, 64, 1, 32}, {2048, 32, 32, 1}: 0.0203397ms, 0.174338% main:@1768 = slice[axes={1},starts={0},ends={128}](main:@1293) -> float_type, {64, 128}, {256, 1}: 0.00074908ms, 0.00642059% main:@1769 = reshape_lazy[dims={64, 1, 128}](main:@1768) -> float_type, {64, 1, 128}, {256, 128, 1}: 0.00058218ms, 0.00499004% main:@1770 = slice[axes={2},starts={0},ends={64}](main:@1769) -> float_type, {64, 1, 64}, {256, 128, 1}: 0.00112974ms, 0.00968335% main:@1771 = unsqueeze[axes={0},steps={}](main:@1770) -> float_type, {1, 64, 1, 64}, {16384, 256, 128, 1}: 0.00065266ms, 0.00559415% main:@1772 = slice[axes={2},starts={64},ends={128}](main:@1769) -> float_type, {64, 1, 64}, {256, 128, 1}: 0.0007432ms, 0.0063702% main:@1773 = unsqueeze[axes={0},steps={}](main:@1772) -> float_type, {1, 64, 1, 64}, {16384, 256, 128, 1}: 0.00096836ms, 0.00830011% main:@1774 = load[offset=3147264,end=3180032](main:@1) -> float_type, {2, 64, 1, 64}, {4096, 64, 64, 1}: 0.00054596ms, 0.00467959% main:@1775 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=2048,local=256,](main:@1771,main:@1773,main:@1774) -> float_type, {2, 64, 1, 64}, {4096, 64, 64, 1}: 0.0139559ms, 0.11962% main:@1776 = load[offset=6092288,end=7730688](main:@1) -> float_type, {2, 64, 50, 64}, {204800, 3200, 64, 1}: 0.0006462ms, 0.00553878% main:@1777 = slice[axes={1},starts={128},ends={256}](main:@1125) -> float_type, {3200, 128}, {256, 1}: 0.0006001ms, 0.00514364% main:@1778 = reshape_lazy[dims={64, 50, 128}](main:@1777) -> float_type, {64, 50, 128}, {12800, 256, 1}: 0.0006176ms, 0.00529364% main:@1779 = slice[axes={2},starts={0},ends={64}](main:@1778) -> float_type, {64, 50, 64}, {12800, 256, 1}: 0.00068902ms, 0.0059058% main:@1780 = unsqueeze[axes={0},steps={}](main:@1779) -> float_type, {1, 64, 50, 64}, {819200, 12800, 256, 1}: 0.000645ms, 0.00552849% main:@1781 = slice[axes={2},starts={64},ends={128}](main:@1778) -> float_type, {64, 50, 64}, {12800, 256, 1}: 0.0010202ms, 0.00874445% main:@1782 = unsqueeze[axes={0},steps={}](main:@1781) -> float_type, {1, 64, 50, 64}, {819200, 12800, 256, 1}: 0.00060656ms, 0.00519901% main:@1783 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=102400,local=256,](main:@1780,main:@1782,main:@1776) -> float_type, {2, 64, 50, 64}, {204800, 3200, 64, 1}: 0.0199666ms, 0.17114% main:@1784 = load[offset=3180032,end=3205632](main:@1) -> float_type, {2, 64, 1, 50}, {3200, 50, 50, 1}: 0.00062466ms, 0.00535415% main:@1785 = transpose[permutation={0, 1, 3, 2}](main:@1783) -> float_type, {2, 64, 64, 50}, {204800, 3200, 1, 64}: 0.00055428ms, 0.0047509% main:@1786 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1775,main:@1785,main:@1784) -> float_type, {2, 64, 1, 50}, {3200, 50, 50, 1}: 0.0220649ms, 0.189125% main:@1787 = load[offset=3147264,end=3160064](main:@1) -> float_type, {64, 1, 50}, {50, 50, 1}: 0.0006351ms, 0.00544364% main:@1788 = unsqueeze[axes={1},steps={}](vOI1hWzBDK:0) -> int32_type, {64, 1, 50}, {50, 50, 1}: 0.00058788ms, 0.0050389% main:@1789 = gpu::code_object[code_object=13528,symbol_name=convert_mul_kernel,global=1600,local=256,](main:@1788,main:@1787) -> float_type, {64, 1, 50}, {50, 50, 1}: 0.0135243ms, 0.115921% main:@1790 = load[offset=3213824,end=3239424](main:@1) -> float_type, {2, 64, 1, 50}, {3200, 50, 50, 1}: 0.00067314ms, 0.00576969% main:@1791 = multibroadcast[out_lens={2, 64, 1, 50}](main:@1789) -> float_type, {2, 64, 1, 50}, {0, 50, 50, 1}: 0.00053298ms, 0.00456834% main:@1792 = gpu::code_object[code_object=13520,symbol_name=add_mul_kernel,global=3200,local=256,](main:@1786,main:@1791,main:@1790) -> float_type, {2, 64, 1, 50}, {3200, 50, 50, 1}: 0.0135501ms, 0.116142% main:@1793 = load[offset=3188224,end=3213824](main:@1) -> float_type, {2, 64, 1, 50}, {3200, 50, 50, 1}: 0.00063688ms, 0.00545889% main:@1794 = gpu::code_object[code_object=14584,symbol_name=softmax_kernel,global=8192,local=64,](main:@1792,main:@1793) -> float_type, {2, 64, 1, 50}, {3200, 50, 50, 1}: 0.01797ms, 0.154026% main:@1795 = slice[axes={1},starts={0},ends={128}](main:@1125) -> float_type, {3200, 128}, {256, 1}: 0.00073838ms, 0.00632888% main:@1796 = reshape_lazy[dims={64, 50, 128}](main:@1795) -> float_type, {64, 50, 128}, {12800, 256, 1}: 0.00059186ms, 0.00507301% main:@1797 = slice[axes={2},starts={0},ends={64}](main:@1796) -> float_type, {64, 50, 64}, {12800, 256, 1}: 0.0006874ms, 0.00589192% main:@1798 = unsqueeze[axes={0},steps={}](main:@1797) -> float_type, {1, 64, 50, 64}, {819200, 12800, 256, 1}: 0.0006229ms, 0.00533907% main:@1799 = slice[axes={2},starts={64},ends={128}](main:@1796) -> float_type, {64, 50, 64}, {12800, 256, 1}: 0.00062214ms, 0.00533255% main:@1800 = unsqueeze[axes={0},steps={}](main:@1799) -> float_type, {1, 64, 50, 64}, {819200, 12800, 256, 1}: 0.00096324ms, 0.00825623% main:@1801 = load[offset=6092288,end=7730688](main:@1) -> float_type, {2, 64, 50, 64}, {204800, 3200, 64, 1}: 0.00047856ms, 0.00410188% main:@1802 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=102400,local=256,](main:@1798,main:@1800,main:@1801) -> float_type, {2, 64, 50, 64}, {204800, 3200, 64, 1}: 0.0199732ms, 0.171196% main:@1803 = load[offset=3155456,end=3188224](main:@1) -> float_type, {2, 64, 1, 64}, {4096, 64, 64, 1}: 0.00062548ms, 0.00536118% main:@1804 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1794,main:@1802,main:@1803) -> float_type, {2, 64, 1, 64}, {4096, 64, 64, 1}: 0.01991ms, 0.170655% H6VFcpqaB0:0 = @param:H6VFcpqaB0:0 -> float_type, {64, 30, 9}, {270, 9, 1}: 0.00075302ms, 0.00645437% main:@1806 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={64, 30, 9}](H6VFcpqaB0:0) -> float_type, {64, 30, 8}, {270, 9, 1}: 0.00075064ms, 0.00643397% CYnW83Np9f:0 = @param:CYnW83Np9f:0 -> float_type, {64, 30, 17}, {510, 17, 1}: 0.00059392ms, 0.00509067% main:@1808 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={64, 30, 17}](CYnW83Np9f:0) -> float_type, {64, 30, 16}, {510, 17, 1}: 0.00077176ms, 0.00661499% hEoRIXiSF0:0 = @param:hEoRIXiSF0:0 -> float_type, {64, 30, 9}, {270, 9, 1}: 0.00058338ms, 0.00500033% main:@1810 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={64, 30, 9}](hEoRIXiSF0:0) -> float_type, {64, 30, 8}, {270, 9, 1}: 0.0006482ms, 0.00555592% 4RjhYVIfz6:0 = @param:4RjhYVIfz6:0 -> float_type, {64, 30, 9}, {270, 9, 1}: 0.00052364ms, 0.00448828% main:@1812 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={64, 30, 9}](4RjhYVIfz6:0) -> float_type, {64, 30, 8}, {270, 9, 1}: 0.00083576ms, 0.00716356% aHGpDUJj9f:0 = @param:aHGpDUJj9f:0 -> float_type, {64, 30, 21}, {630, 21, 1}: 0.00052002ms, 0.00445725% main:@1814 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={64, 30, 21}](aHGpDUJj9f:0) -> float_type, {64, 30, 20}, {630, 21, 1}: 0.00100486ms, 0.00861296% Hj2GwToc7U:0 = @param:Hj2GwToc7U:0 -> float_type, {64, 30, 9}, {270, 9, 1}: 0.0005098ms, 0.00436965% main:@1816 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={64, 30, 9}](Hj2GwToc7U:0) -> float_type, {64, 30, 8}, {270, 9, 1}: 0.00082204ms, 0.00704596% S4PeIw3K1X:0 = @param:S4PeIw3K1X:0 -> float_type, {64, 30, 21}, {630, 21, 1}: 0.00056588ms, 0.00485033% main:@1818 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={64, 30, 21}](S4PeIw3K1X:0) -> float_type, {64, 30, 20}, {630, 21, 1}: 0.00066424ms, 0.0056934% DW8PcmTVkC:0 = @param:DW8PcmTVkC:0 -> float_type, {64, 30, 9}, {270, 9, 1}: 0.00055386ms, 0.0047473% main:@1820 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={64, 30, 9}](DW8PcmTVkC:0) -> float_type, {64, 30, 8}, {270, 9, 1}: 0.00075166ms, 0.00644271% cEDTxpPUfb:0 = @param:cEDTxpPUfb:0 -> float_type, {64, 30, 9}, {270, 9, 1}: 0.0005412ms, 0.00463879% main:@1822 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={64, 30, 9}](cEDTxpPUfb:0) -> float_type, {64, 30, 8}, {270, 9, 1}: 0.00070746ms, 0.00606386% main:@1823 = load[offset=6092288,end=6891008](main:@1) -> float_type, {64, 30, 104}, {3120, 104, 1}: 0.00047758ms, 0.00409348% main:@1824 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=22186,local=256,](main:@1806,main:@1820,main:@1810,main:@1816,main:@1814,main:@1812,main:@1822,main:@1808,main:@1818,main:@1823) -> float_type, {64, 30, 104}, {3120, 104, 1}: 0.027762ms, 0.237956% main:@1825 = hip::hip_copy_literal[id=main:@literal:23] -> float_type, {104, 64}, {64, 1}: 0.00076396ms, 0.00654814% main:@1826 = load[offset=6891008,end=7382528](main:@1) -> float_type, {1920, 64}, {64, 1}: 0.00049364ms, 0.00423114% main:@1827 = reshape_lazy[dims={1920, 104}](main:@1824) -> float_type, {1920, 104}, {104, 1}: 0.00054288ms, 0.00465319% main:@1828 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1827,main:@1825,main:@1826) -> float_type, {1920, 64}, {64, 1}: 0.0253073ms, 0.216917% main:@1829 = slice[axes={0, 1},starts={0, 305},ends={64, 337}](uFSOc5Chp2:0) -> float_type, {64, 32}, {481, 1}: 0.00084734ms, 0.00726281% main:@1830 = slice[axes={0, 1},starts={0, 73},ends={64, 105}](n4iptOVFwE:0) -> float_type, {64, 32}, {113, 1}: 0.00061154ms, 0.0052417% main:@1831 = slice[axes={0, 1},starts={0, 533},ends={64, 541}](lyiDxfzZg3:0) -> float_type, {64, 8}, {701, 1}: 0.00074508ms, 0.00638631% main:@1832 = slice[axes={0, 1},starts={0, 269},ends={64, 285}](nv2tzWRf1I:0) -> float_type, {64, 16}, {301, 1}: 0.0006128ms, 0.0052525% main:@1833 = load[offset=3196416,end=3218944](main:@1) -> float_type, {64, 88}, {88, 1}: 0.00051392ms, 0.00440497% main:@1834 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=1408,local=256,](main:@1831,main:@1832,main:@1829,main:@1830,main:@1833) -> float_type, {64, 88}, {88, 1}: 0.0159593ms, 0.136792% main:@1835 = hip::hip_copy_literal[id=main:@literal:77] -> float_type, {88, 32}, {32, 1}: 0.0008132ms, 0.00697019% main:@1836 = load[offset=3188224,end=3196416](main:@1) -> float_type, {64, 32}, {32, 1}: 0.00050354ms, 0.004316% main:@1837 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1834,main:@1835,main:@1836) -> float_type, {64, 32}, {32, 1}: 0.0212389ms, 0.182045% main:@1838 = slice[axes={1},starts={32},ends={64}](main:@1828) -> float_type, {1920, 32}, {64, 1}: 0.00079096ms, 0.00677956% main:@1839 = reshape_lazy[dims={1, 64, 30, 32}](main:@1838) -> float_type, {1, 64, 30, 32}, {122880, 1920, 64, 1}: 0.00072166ms, 0.00618557% main:@1840 = transpose[permutation={0, 1, 3, 2}](main:@1839) -> float_type, {1, 64, 32, 30}, {122880, 1920, 1, 64}: 0.00058848ms, 0.00504404% main:@1841 = load[offset=3147264,end=3154944](main:@1) -> float_type, {1, 64, 1, 30}, {1920, 30, 30, 1}: 0.00051576ms, 0.00442074% main:@1842 = reshape_lazy[dims={1, 64, 1, 32}](main:@1837) -> float_type, {1, 64, 1, 32}, {2048, 32, 32, 1}: 0.0005062ms, 0.0043388% main:@1843 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1842,main:@1840,main:@1841) -> float_type, {1, 64, 1, 30}, {1920, 30, 30, 1}: 0.0179005ms, 0.15343% main:@1844 = load[offset=3188224,end=3195904](main:@1) -> float_type, {64, 1, 30}, {30, 30, 1}: 0.00064764ms, 0.00555112% LjbB2i31Wu:0 = @param:LjbB2i31Wu:0 -> int32_type, {64, 30}, {30, 1}: 0.0005688ms, 0.00487536% main:@1846 = unsqueeze[axes={1},steps={}](LjbB2i31Wu:0) -> int32_type, {64, 1, 30}, {30, 30, 1}: 0.00057264ms, 0.00490827% main:@1847 = gpu::code_object[code_object=13528,symbol_name=convert_add_kernel,global=960,local=256,](main:@1846,main:@1844) -> float_type, {64, 1, 30}, {30, 30, 1}: 0.0133846ms, 0.114724% main:@1848 = multibroadcast[out_lens={1, 64, 1, 30}](main:@1847) -> float_type, {1, 64, 1, 30}, {0, 30, 30, 1}: 0.00068112ms, 0.00583809% main:@1849 = load[offset=3195904,end=3203584](main:@1) -> float_type, {1, 64, 1, 30}, {1920, 30, 30, 1}: 0.0004854ms, 0.00416051% main:@1850 = gpu::code_object[code_object=13520,symbol_name=pow_pow_kernel,global=960,local=256,](main:@1843,main:@1848,main:@1849) -> float_type, {1, 64, 1, 30}, {1920, 30, 30, 1}: 0.0181793ms, 0.155821% main:@1851 = load[offset=3188224,end=3195904](main:@1) -> float_type, {1, 64, 1, 30}, {1920, 30, 30, 1}: 0.00062452ms, 0.00535295% main:@1852 = gpu::code_object[code_object=14584,symbol_name=softmax_kernel,global=4096,local=64,](main:@1850,main:@1851) -> float_type, {1, 64, 1, 30}, {1920, 30, 30, 1}: 0.0157968ms, 0.135399% main:@1853 = load[offset=3147264,end=3155456](main:@1) -> float_type, {1, 64, 1, 32}, {2048, 32, 32, 1}: 0.00066946ms, 0.00573815% main:@1854 = slice[axes={1},starts={0},ends={32}](main:@1828) -> float_type, {1920, 32}, {64, 1}: 0.00062346ms, 0.00534387% main:@1855 = reshape_lazy[dims={1, 64, 30, 32}](main:@1854) -> float_type, {1, 64, 30, 32}, {122880, 1920, 64, 1}: 0.000936ms, 0.00802274% main:@1856 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1852,main:@1855,main:@1853) -> float_type, {1, 64, 1, 32}, {2048, 32, 32, 1}: 0.019118ms, 0.163867% main:@1857 = slice[axes={1},starts={0},ends={32}](main:@652) -> float_type, {64, 32}, {128, 1}: 0.00074852ms, 0.00641579% main:@1858 = load[offset=3188224,end=3196416](main:@1) -> float_type, {64, 32}, {32, 1}: 0.0004849ms, 0.00415623% main:@1859 = gpu::code_object[code_object=13528,symbol_name=contiguous_kernel,global=1024,local=256,](main:@1857,main:@1858) -> float_type, {64, 32}, {32, 1}: 0.0130672ms, 0.112003% main:@1860 = load[offset=778496,end=782336](main:@1) -> float_type, {1, 64, 1, 15}, {960, 15, 15, 1}: 0.00062798ms, 0.00538261% main:@1861 = slice[axes={1},starts={0},ends={32}](main:@700) -> float_type, {960, 32}, {64, 1}: 0.00061734ms, 0.00529141% main:@1862 = reshape_lazy[dims={1, 64, 15, 32}](main:@1861) -> float_type, {1, 64, 15, 32}, {61440, 960, 64, 1}: 0.0005767ms, 0.00494307% main:@1863 = transpose[permutation={0, 1, 3, 2}](main:@1862) -> float_type, {1, 64, 32, 15}, {61440, 960, 1, 64}: 0.0005709ms, 0.00489336% main:@1864 = reshape_lazy[dims={1, 64, 1, 32}](main:@1859) -> float_type, {1, 64, 1, 32}, {2048, 32, 32, 1}: 0.00049876ms, 0.00427503% main:@1865 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1864,main:@1863,main:@1860) -> float_type, {1, 64, 1, 15}, {960, 15, 15, 1}: 0.0191432ms, 0.164082% main:@1866 = load[offset=782336,end=786176](main:@1) -> float_type, {64, 1, 15}, {15, 15, 1}: 0.00065534ms, 0.00561712% UaC5MZd8qb:0 = @param:UaC5MZd8qb:0 -> int32_type, {64, 15}, {15, 1}: 0.00057788ms, 0.00495319% main:@1868 = unsqueeze[axes={1},steps={}](UaC5MZd8qb:0) -> int32_type, {64, 1, 15}, {15, 15, 1}: 0.00058588ms, 0.00502176% main:@1869 = gpu::code_object[code_object=13536,symbol_name=convert_div_mul_kernel,global=480,local=256,](main:@1868,main:@1866) -> float_type, {64, 1, 15}, {15, 15, 1}: 0.0135535ms, 0.116172% main:@1870 = multibroadcast[out_lens={1, 64, 1, 15}](main:@1869) -> float_type, {1, 64, 1, 15}, {0, 15, 15, 1}: 0.0006891ms, 0.00590649% main:@1871 = load[offset=790528,end=794368](main:@1) -> float_type, {1, 64, 1, 15}, {960, 15, 15, 1}: 0.00051034ms, 0.00437428% main:@1872 = gpu::code_object[code_object=13520,symbol_name=add_pow_kernel,global=480,local=256,](main:@1865,main:@1870,main:@1871) -> float_type, {1, 64, 1, 15}, {960, 15, 15, 1}: 0.0163123ms, 0.139817% main:@1873 = load[offset=786688,end=790528](main:@1) -> float_type, {1, 64, 1, 15}, {960, 15, 15, 1}: 0.00062188ms, 0.00533032% main:@1874 = gpu::code_object[code_object=14568,symbol_name=softmax_kernel,global=4096,local=64,](main:@1872,main:@1873) -> float_type, {1, 64, 1, 15}, {960, 15, 15, 1}: 0.0155778ms, 0.133522% main:@1875 = slice[axes={1},starts={32},ends={64}](main:@700) -> float_type, {960, 32}, {64, 1}: 0.00073122ms, 0.00626751% main:@1876 = load[offset=778496,end=786688](main:@1) -> float_type, {1, 64, 1, 32}, {2048, 32, 32, 1}: 0.0004903ms, 0.00420251% main:@1877 = reshape_lazy[dims={1, 64, 15, 32}](main:@1875) -> float_type, {1, 64, 15, 32}, {61440, 960, 64, 1}: 0.00060572ms, 0.00519181% main:@1878 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1874,main:@1877,main:@1876) -> float_type, {1, 64, 1, 32}, {2048, 32, 32, 1}: 0.0196908ms, 0.168776% Kanv8c4Yl7:0 = @param:Kanv8c4Yl7:0 -> float_type, {64, 201}, {201, 1}: 0.0007198ms, 0.00616963% main:@1880 = slice[axes={0, 1},starts={0, 1},ends={64, 65}](Kanv8c4Yl7:0) -> float_type, {64, 64}, {201, 1}: 0.00068522ms, 0.00587323% main:@1881 = slice[axes={0, 1},starts={0, 193},ends={64, 257}](lyiDxfzZg3:0) -> float_type, {64, 64}, {701, 1}: 0.00107758ms, 0.00923627% main:@1882 = load[offset=2040064,end=2072832](main:@1) -> float_type, {64, 128}, {128, 1}: 0.00050568ms, 0.00433434% main:@1883 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=4096,local=256,](main:@1881,main:@1880,main:@1882) -> float_type, {64, 128}, {128, 1}: 0.0144041ms, 0.123462% main:@1884 = slice[axes={0, 1},starts={0, 1},ends={64, 65}](nv2tzWRf1I:0) -> float_type, {64, 64}, {301, 1}: 0.00077808ms, 0.00666916% main:@1885 = slice[axes={0, 1},starts={0, 65},ends={64, 129}](uFSOc5Chp2:0) -> float_type, {64, 64}, {481, 1}: 0.00064596ms, 0.00553672% main:@1886 = slice[axes={0, 1},starts={0, 1},ends={64, 65}](uFSOc5Chp2:0) -> float_type, {64, 64}, {481, 1}: 0.00092816ms, 0.00795554% main:@1887 = load[offset=2007296,end=2040064](main:@1) -> float_type, {64, 128}, {128, 1}: 0.00047728ms, 0.00409091% main:@1888 = gpu::code_object[code_object=13544,symbol_name=noop_add_concat_div_kernel,global=4096,local=256,](main:@1885,main:@1884,main:@1886,main:@1883,main:@1887) -> float_type, {64, 128}, {128, 1}: 0.0145569ms, 0.124772% main:@1889 = transpose[permutation={1, 2, 0, 3}](main:@1081) -> float_type, {64, 15, 4, 64}, {960, 64, 61440, 1}: 0.00071534ms, 0.0061314% main:@1890 = load[offset=2040064,end=2105600](main:@1) -> float_type, {64, 1, 4, 64}, {64, 64, 4096, 1}: 0.0004954ms, 0.00424623% main:@1891 = gpu::code_object[code_object=13528,symbol_name=reduce_prod_kernel,global=16384,local=256,](main:@1889,main:@1890) -> float_type, {64, 1, 4, 64}, {64, 64, 4096, 1}: 0.0156013ms, 0.133724% main:@1892 = unsqueeze[axes={1},steps={}](main:@1134) -> float_type, {64, 1, 256}, {256, 256, 1}: 0.00073136ms, 0.00626871% main:@1893 = unsqueeze[axes={1},steps={}](main:@1143) -> float_type, {64, 1, 256}, {256, 256, 1}: 0.00050332ms, 0.00431411% main:@1894 = unsqueeze[axes={1},steps={}](main:@1153) -> float_type, {64, 1, 256}, {256, 256, 1}: 0.00048254ms, 0.004136% main:@1895 = unsqueeze[axes={1},steps={}](main:@1163) -> float_type, {64, 1, 256}, {256, 256, 1}: 0.00049604ms, 0.00425171% main:@1896 = unsqueeze[axes={1},steps={}](main:@1173) -> float_type, {64, 1, 256}, {256, 256, 1}: 0.00049738ms, 0.0042632% main:@1897 = unsqueeze[axes={1},steps={}](main:@1182) -> float_type, {64, 1, 256}, {256, 256, 1}: 0.0004924ms, 0.00422051% main:@1898 = unsqueeze[axes={1},steps={}](main:@1191) -> float_type, {64, 1, 256}, {256, 256, 1}: 0.000466ms, 0.00399423% main:@1899 = unsqueeze[axes={1},steps={}](main:@1200) -> float_type, {64, 1, 256}, {256, 256, 1}: 0.00046302ms, 0.00396869% main:@1900 = unsqueeze[axes={1},steps={}](main:@1209) -> float_type, {64, 1, 256}, {256, 256, 1}: 0.0004692ms, 0.00402166% main:@1901 = unsqueeze[axes={1},steps={}](main:@1219) -> float_type, {64, 1, 256}, {256, 256, 1}: 0.00048408ms, 0.0041492% main:@1902 = unsqueeze[axes={1},steps={}](main:@1228) -> float_type, {64, 1, 256}, {256, 256, 1}: 0.00052108ms, 0.00446634% main:@1903 = unsqueeze[axes={1},steps={}](main:@1237) -> float_type, {64, 1, 256}, {256, 256, 1}: 0.0004636ms, 0.00397366% main:@1904 = unsqueeze[axes={1},steps={}](main:@1246) -> float_type, {64, 1, 256}, {256, 256, 1}: 0.00046276ms, 0.00396646% main:@1905 = unsqueeze[axes={1},steps={}](main:@1256) -> float_type, {64, 1, 256}, {256, 256, 1}: 0.00048582ms, 0.00416411% main:@1906 = unsqueeze[axes={1},steps={}](main:@1266) -> float_type, {64, 1, 256}, {256, 256, 1}: 0.00046628ms, 0.00399663% main:@1907 = unsqueeze[axes={1},steps={}](main:@1275) -> float_type, {64, 1, 256}, {256, 256, 1}: 0.00047508ms, 0.00407206% main:@1908 = unsqueeze[axes={1},steps={}](main:@1302) -> float_type, {64, 1, 256}, {256, 256, 1}: 0.00046176ms, 0.00395789% main:@1909 = unsqueeze[axes={1},steps={}](main:@1311) -> float_type, {64, 1, 256}, {256, 256, 1}: 0.00048316ms, 0.00414131% main:@1910 = load[offset=4555264,end=5734912](main:@1) -> float_type, {64, 18, 256}, {4608, 256, 1}: 0.00049644ms, 0.00425514% main:@1911 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=8192,local=256,](main:@1899,main:@1894,main:@1898,main:@1909,main:@1908,main:@1897,main:@1906,main:@1905,main:@1896,main:@1892,main:@1903,main:@1907,main:@1901,main:@1904,main:@1900,main:@1895,main:@1893,main:@1902,main:@1910) -> float_type, {64, 18, 256}, {4608, 256, 1}: 0.028033ms, 0.240279% main:@1912 = hip::hip_copy_literal[id=main:@literal:158] -> float_type, {1}, {0}: 0.00083198ms, 0.00713116% main:@1913 = multibroadcast[out_lens={64, 18, 1}](main:@1912) -> float_type, {64, 18, 1}, {0, 0, 0}: 0.00060444ms, 0.00518084% main:@1914 = multibroadcast[out_lens={64, 18, 256}](main:@1913) -> float_type, {64, 18, 256}, {0, 0, 0}: 0.00054076ms, 0.00463502% main:@1915 = load[offset=6891008,end=8070656](main:@1) -> float_type, {64, 18, 256}, {4608, 256, 1}: 0.00050854ms, 0.00435885% main:@1916 = gpu::code_object[code_object=17656,symbol_name=pow_reduce_sum_div_pow_add_kernel,global=147456,local=128,](main:@1911,main:@713,main:@1914,main:@711,main:@1915) -> float_type, {64, 18, 256}, {4608, 256, 1}: 0.0310975ms, 0.266546% main:@1917 = load[offset=786688,end=852224](main:@1) -> float_type, {64, 1, 256}, {256, 256, 1}: 0.00061966ms, 0.0053113% main:@1918 = gpu::code_object[code_object=13528,symbol_name=reduce_sum_kernel,global=16384,local=256,](main:@1916,main:@1917) -> float_type, {64, 1, 256}, {256, 256, 1}: 0.0157401ms, 0.134913% main:@1919 = load[offset=852224,end=878848](main:@1) -> float_type, {64, 1, 104}, {104, 104, 1}: 0.00065272ms, 0.00559466% main:@1920 = gpu::code_object[code_object=13528,symbol_name=reduce_mean_kernel,global=6656,local=256,](main:@1824,main:@1919) -> float_type, {64, 1, 104}, {104, 104, 1}: 0.0148606ms, 0.127375% main:@1921 = load[offset=878848,end=896256](main:@1) -> float_type, {64, 1, 68}, {68, 68, 1}: 0.00064818ms, 0.00555575% main:@1922 = gpu::code_object[code_object=13528,symbol_name=reduce_prod_kernel,global=4352,local=256,](main:@682,main:@1921) -> float_type, {64, 1, 68}, {68, 68, 1}: 0.0147573ms, 0.12649% main:@1923 = transpose[permutation={1, 2, 0, 3}](main:@1804) -> float_type, {64, 1, 2, 64}, {64, 64, 4096, 1}: 0.00070034ms, 0.00600283% main:@1924 = load[offset=256256,end=289024](main:@1) -> float_type, {64, 1, 2, 64}, {128, 128, 64, 1}: 0.00050012ms, 0.00428668% main:@1925 = gpu::code_object[code_object=13528,symbol_name=contiguous_kernel,global=4096,local=256,](main:@1923,main:@1924) -> float_type, {64, 1, 2, 64}, {128, 128, 64, 1}: 0.0133149ms, 0.114126% main:@1926 = load[offset=289024,end=354560](main:@1) -> float_type, {64, 1, 4, 64}, {256, 256, 64, 1}: 0.00063616ms, 0.00545272% main:@1927 = gpu::code_object[code_object=13528,symbol_name=contiguous_kernel,global=8192,local=256,](main:@1891,main:@1926) -> float_type, {64, 1, 4, 64}, {256, 256, 64, 1}: 0.0141696ms, 0.121452% main:@1928 = load[offset=354560,end=387328](main:@1) -> float_type, {64, 1, 2, 64}, {128, 128, 64, 1}: 0.00060366ms, 0.00517416% main:@1929 = transpose[permutation={1, 2, 0, 3}](main:@1548) -> float_type, {64, 1, 2, 64}, {64, 64, 4096, 1}: 0.00054936ms, 0.00470873% main:@1930 = gpu::code_object[code_object=13528,symbol_name=contiguous_kernel,global=4096,local=256,](main:@1929,main:@1928) -> float_type, {64, 1, 2, 64}, {128, 128, 64, 1}: 0.0132702ms, 0.113743% 01E2VTipRX:0 = @param:01E2VTipRX:0 -> float_type, {64, 9}, {9, 1}: 0.00070128ms, 0.00601089% main:@1932 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](01E2VTipRX:0) -> float_type, {64, 8}, {9, 1}: 0.00071504ms, 0.00612883% main:@1933 = transpose[permutation={1, 2, 0, 3}](main:@1103) -> float_type, {64, 1, 1, 64}, {64, 64, 4096, 1}: 0.00052762ms, 0.00452239% main:@1934 = squeeze[axes={1}](main:@1918) -> float_type, {64, 256}, {256, 1}: 0.00056816ms, 0.00486987% u2OJ7kfglD:0 = @param:u2OJ7kfglD:0 -> float_type, {64, 93}, {93, 1}: 0.00053748ms, 0.00460691% main:@1936 = slice[axes={0, 1},starts={0, 61},ends={64, 65}](u2OJ7kfglD:0) -> float_type, {64, 4}, {93, 1}: 0.0008035ms, 0.00688705% main:@1937 = transpose[permutation={1, 2, 0, 3}](main:@1717) -> float_type, {64, 1, 1, 32}, {32, 32, 2048, 1}: 0.0004968ms, 0.00425823% main:@1938 = transpose[permutation={1, 2, 0, 3}](main:@1745) -> float_type, {64, 1, 1, 32}, {32, 32, 2048, 1}: 0.00048388ms, 0.00414748% main:@1939 = slice[axes={0, 1},starts={0, 17},ends={64, 33}](0cjwsL5ldh:0) -> float_type, {64, 16}, {33, 1}: 0.00099504ms, 0.00852879% WycVU13dfZ:0 = @param:WycVU13dfZ:0 -> float_type, {64, 5}, {5, 1}: 0.00058472ms, 0.00501181% main:@1941 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](WycVU13dfZ:0) -> float_type, {64, 4}, {5, 1}: 0.00072554ms, 0.00621883% main:@1942 = transpose[permutation={1, 2, 0, 3}](main:@1767) -> float_type, {64, 1, 1, 32}, {32, 32, 2048, 1}: 0.00049514ms, 0.004244% main:@1943 = slice[axes={0, 1},starts={0, 5},ends={64, 13}](0sDBrPmfS9:0) -> float_type, {64, 8}, {13, 1}: 0.00066552ms, 0.00570438% main:@1944 = transpose[permutation={1, 2, 0, 3}](main:@1856) -> float_type, {64, 1, 1, 32}, {32, 32, 2048, 1}: 0.00048142ms, 0.0041264% main:@1945 = squeeze[axes={1}](main:@1920) -> float_type, {64, 104}, {104, 1}: 0.0004924ms, 0.00422051% 1f2xpcXgdE:0 = @param:1f2xpcXgdE:0 -> float_type, {64, 13}, {13, 1}: 0.00048234ms, 0.00413428% main:@1947 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](1f2xpcXgdE:0) -> float_type, {64, 4}, {13, 1}: 0.00082942ms, 0.00710921% vZDAiUHErW:0 = @param:vZDAiUHErW:0 -> float_type, {64, 17}, {17, 1}: 0.00053732ms, 0.00460553% main:@1949 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](vZDAiUHErW:0) -> float_type, {64, 8}, {17, 1}: 0.0008289ms, 0.00710476% main:@1950 = squeeze[axes={1}](main:@1922) -> float_type, {64, 68}, {68, 1}: 0.00049478ms, 0.00424091% main:@1951 = slice[axes={0, 1},starts={0, 33},ends={64, 41}](2CnSyUBzm0:0) -> float_type, {64, 8}, {65, 1}: 0.0007247ms, 0.00621163% main:@1952 = transpose[permutation={1, 2, 0, 3}](main:@1878) -> float_type, {64, 1, 1, 32}, {32, 32, 2048, 1}: 0.00047058ms, 0.00403349% main:@1953 = slice[axes={0, 1},starts={0, 5},ends={64, 13}](m5MfQhR1Bp:0) -> float_type, {64, 8}, {13, 1}: 0.00070054ms, 0.00600454% main:@1954 = slice[axes={0, 1},starts={0, 33},ends={64, 37}](2RDXZrx9GA:0) -> float_type, {64, 4}, {45, 1}: 0.00070338ms, 0.00602889% main:@1955 = slice[axes={0, 1},starts={0, 33},ends={64, 37}](2Yy3zxNGBC:0) -> float_type, {64, 4}, {45, 1}: 0.00067922ms, 0.0058218% main:@1956 = slice[axes={0, 1},starts={0, 17},ends={64, 25}](ZRYNFfjSez:0) -> float_type, {64, 8}, {33, 1}: 0.00067964ms, 0.0058254% BQHyb3DGCE:0 = @param:BQHyb3DGCE:0 -> float_type, {64, 9}, {9, 1}: 0.00050656ms, 0.00434188% main:@1958 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](BQHyb3DGCE:0) -> float_type, {64, 8}, {9, 1}: 0.00071824ms, 0.00615626% main:@1959 = slice[axes={0, 1},starts={0, 33},ends={64, 37}](3Laj620MSn:0) -> float_type, {64, 4}, {45, 1}: 0.00067446ms, 0.005781% RF5h8lytsu:0 = @param:RF5h8lytsu:0 -> float_type, {64, 9}, {9, 1}: 0.00054646ms, 0.00468388% main:@1961 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](RF5h8lytsu:0) -> float_type, {64, 8}, {9, 1}: 0.00073036ms, 0.00626014% EClzvVLqMP:0 = @param:EClzvVLqMP:0 -> float_type, {64, 9}, {9, 1}: 0.00059316ms, 0.00508416% main:@1963 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](EClzvVLqMP:0) -> float_type, {64, 8}, {9, 1}: 0.00070998ms, 0.00608546% main:@1964 = slice[axes={0, 1},starts={0, 17},ends={64, 25}](3mk8BKjDey:0) -> float_type, {64, 8}, {25, 1}: 0.000672ms, 0.00575992% MQb6fHVndu:0 = @param:MQb6fHVndu:0 -> float_type, {64, 9}, {9, 1}: 0.00062284ms, 0.00533855% main:@1966 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](MQb6fHVndu:0) -> float_type, {64, 8}, {9, 1}: 0.00079372ms, 0.00680322% MtvajoSCeD:0 = @param:MtvajoSCeD:0 -> float_type, {64, 9}, {9, 1}: 0.00059032ms, 0.00505981% main:@1968 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](MtvajoSCeD:0) -> float_type, {64, 8}, {9, 1}: 0.00070814ms, 0.00606969% 4U3hVYNn2F:0 = @param:4U3hVYNn2F:0 -> float_type, {64, 5}, {5, 1}: 0.00054842ms, 0.00470068% main:@1970 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](4U3hVYNn2F:0) -> float_type, {64, 4}, {5, 1}: 0.00075054ms, 0.00643311% 4dtfMgk7Ni:0 = @param:4dtfMgk7Ni:0 -> float_type, {64, 9}, {9, 1}: 0.0005555ms, 0.00476136% main:@1972 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](4dtfMgk7Ni:0) -> float_type, {64, 8}, {9, 1}: 0.00074176ms, 0.00635785% nP9Yv1iZ8f:0 = @param:nP9Yv1iZ8f:0 -> float_type, {64, 9}, {9, 1}: 0.00053876ms, 0.00461788% main:@1974 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](nP9Yv1iZ8f:0) -> float_type, {64, 8}, {9, 1}: 0.00075124ms, 0.00643911% T1FVGiDRMg:0 = @param:T1FVGiDRMg:0 -> float_type, {64, 5}, {5, 1}: 0.00048808ms, 0.00418348% main:@1976 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](T1FVGiDRMg:0) -> float_type, {64, 4}, {5, 1}: 0.0006531ms, 0.00559792% 50ovs1NnYh:0 = @param:50ovs1NnYh:0 -> float_type, {64, 9}, {9, 1}: 0.00067186ms, 0.00575872% main:@1978 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](50ovs1NnYh:0) -> float_type, {64, 8}, {9, 1}: 0.0007094ms, 0.00608049% main:@1979 = slice[axes={0, 1},starts={0, 5},ends={64, 9}](5JHUPChra8:0) -> float_type, {64, 4}, {17, 1}: 0.00070986ms, 0.00608443% 6eQEnYsRTi:0 = @param:6eQEnYsRTi:0 -> float_type, {64, 5}, {5, 1}: 0.00052822ms, 0.00452754% main:@1981 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](6eQEnYsRTi:0) -> float_type, {64, 4}, {5, 1}: 0.00071808ms, 0.00615488% 5bcvLlI6HE:0 = @param:5bcvLlI6HE:0 -> float_type, {64, 9}, {9, 1}: 0.00053634ms, 0.00459713% main:@1983 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](5bcvLlI6HE:0) -> float_type, {64, 8}, {9, 1}: 0.00072958ms, 0.00625345% 5cO3meIa4V:0 = @param:5cO3meIa4V:0 -> float_type, {64, 5}, {5, 1}: 0.00062506ms, 0.00535758% main:@1985 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](5cO3meIa4V:0) -> float_type, {64, 4}, {5, 1}: 0.00070666ms, 0.006057% Q2Om7B1nhU:0 = @param:Q2Om7B1nhU:0 -> float_type, {64, 9}, {9, 1}: 0.0006479ms, 0.00555335% main:@1987 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](Q2Om7B1nhU:0) -> float_type, {64, 8}, {9, 1}: 0.00068396ms, 0.00586243% main:@1988 = slice[axes={0, 1},starts={0, 33},ends={64, 37}](6HSQqXdJzb:0) -> float_type, {64, 4}, {45, 1}: 0.00071534ms, 0.0061314% main:@1989 = slice[axes={0, 1},starts={0, 5},ends={64, 13}](w27LgiJW6P:0) -> float_type, {64, 8}, {13, 1}: 0.00072342ms, 0.00620065% KgZXYMNz6P:0 = @param:KgZXYMNz6P:0 -> float_type, {64, 13}, {13, 1}: 0.0005197ms, 0.00445451% main:@1991 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](KgZXYMNz6P:0) -> float_type, {64, 4}, {13, 1}: 0.0007253ms, 0.00621677% 6hbLUlYs1H:0 = @param:6hbLUlYs1H:0 -> float_type, {64, 9}, {9, 1}: 0.00054526ms, 0.00467359% main:@1993 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](6hbLUlYs1H:0) -> float_type, {64, 8}, {9, 1}: 0.00073172ms, 0.0062718% y3bafuwnEk:0 = @param:y3bafuwnEk:0 -> float_type, {64, 5}, {5, 1}: 0.00061634ms, 0.00528284% main:@1995 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](y3bafuwnEk:0) -> float_type, {64, 4}, {5, 1}: 0.000665ms, 0.00569992% main:@1996 = slice[axes={0, 1},starts={0, 25},ends={64, 41}](ifP6Ceko5T:0) -> float_type, {64, 16}, {73, 1}: 0.00068202ms, 0.0058458% 89nc05gQza:0 = @param:89nc05gQza:0 -> float_type, {64, 9}, {9, 1}: 0.00055054ms, 0.00471885% main:@1998 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](89nc05gQza:0) -> float_type, {64, 8}, {9, 1}: 0.00068582ms, 0.00587837% 8KRQILP469:0 = @param:8KRQILP469:0 -> float_type, {64, 9}, {9, 1}: 0.00054958ms, 0.00471062% main:@2000 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](8KRQILP469:0) -> float_type, {64, 8}, {9, 1}: 0.00064532ms, 0.00553124% VP35W0GOCc:0 = @param:VP35W0GOCc:0 -> float_type, {64, 13}, {13, 1}: 0.00053134ms, 0.00455428% main:@2002 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](VP35W0GOCc:0) -> float_type, {64, 4}, {13, 1}: 0.00073686ms, 0.00631585% 8iOST90kWM:0 = @param:8iOST90kWM:0 -> float_type, {64, 9}, {9, 1}: 0.0005199ms, 0.00445622% main:@2004 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](8iOST90kWM:0) -> float_type, {64, 8}, {9, 1}: 0.00069666ms, 0.00597129% 8Dv01FAoNM:0 = @param:8Dv01FAoNM:0 -> float_type, {64, 17}, {17, 1}: 0.0005237ms, 0.00448879% main:@2006 = slice[axes={0, 1},starts={0, 1},ends={64, 17}](8Dv01FAoNM:0) -> float_type, {64, 16}, {17, 1}: 0.00066856ms, 0.00573043% main:@2007 = slice[axes={0, 1},starts={0, 33},ends={64, 37}](X29iyIeu8L:0) -> float_type, {64, 4}, {45, 1}: 0.00069568ms, 0.00596289% 985IRiY0X6:0 = @param:985IRiY0X6:0 -> float_type, {64, 57}, {57, 1}: 0.00048292ms, 0.00413926% main:@2009 = slice[axes={0, 1},starts={0, 9},ends={64, 17}](985IRiY0X6:0) -> float_type, {64, 8}, {57, 1}: 0.00065612ms, 0.00562381% main:@2010 = slice[axes={0, 1},starts={0, 41},ends={64, 49}](985IRiY0X6:0) -> float_type, {64, 8}, {57, 1}: 0.00064404ms, 0.00552026% main:@2011 = slice[axes={0, 1},starts={0, 17},ends={64, 25}](985IRiY0X6:0) -> float_type, {64, 8}, {57, 1}: 0.00065788ms, 0.00563889% main:@2012 = slice[axes={0, 1},starts={0, 49},ends={64, 57}](985IRiY0X6:0) -> float_type, {64, 8}, {57, 1}: 0.00064956ms, 0.00556758% 9N1KUX0DZ5:0 = @param:9N1KUX0DZ5:0 -> float_type, {64, 13}, {13, 1}: 0.00048828ms, 0.0041852% main:@2014 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](9N1KUX0DZ5:0) -> float_type, {64, 4}, {13, 1}: 0.00071612ms, 0.00613808% main:@2015 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](985IRiY0X6:0) -> float_type, {64, 8}, {57, 1}: 0.00067662ms, 0.00579952% main:@2016 = slice[axes={0, 1},starts={0, 33},ends={64, 41}](985IRiY0X6:0) -> float_type, {64, 8}, {57, 1}: 0.00069356ms, 0.00594472% main:@2017 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](9YqhflbMKe:0) -> float_type, {64, 4}, {29, 1}: 0.00065542ms, 0.00561781% main:@2018 = slice[axes={0, 1},starts={0, 33},ends={64, 37}](VUqfdN5Fzv:0) -> float_type, {64, 4}, {45, 1}: 0.00065474ms, 0.00561198% main:@2019 = slice[axes={0, 1},starts={0, 25},ends={64, 33}](985IRiY0X6:0) -> float_type, {64, 8}, {57, 1}: 0.0006545ms, 0.00560992% 9mfMQZIXVH:0 = @param:9mfMQZIXVH:0 -> float_type, {64, 9}, {9, 1}: 0.00052168ms, 0.00447148% main:@2021 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](9mfMQZIXVH:0) -> float_type, {64, 8}, {9, 1}: 0.00069816ms, 0.00598414% 9rDQhGZo18:0 = @param:9rDQhGZo18:0 -> float_type, {64, 9}, {9, 1}: 0.00052972ms, 0.00454039% main:@2023 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](9rDQhGZo18:0) -> float_type, {64, 8}, {9, 1}: 0.00067898ms, 0.00581975% main:@2024 = slice[axes={0, 1},starts={0, 17},ends={64, 25}](l4bReWtsMB:0) -> float_type, {64, 8}, {25, 1}: 0.0006795ms, 0.0058242% main:@2025 = slice[axes={0, 1},starts={0, 9},ends={64, 13}](AVLcBpWeuo:0) -> float_type, {64, 4}, {21, 1}: 0.00069814ms, 0.00598397% AfxiFXbSj3:0 = @param:AfxiFXbSj3:0 -> float_type, {64, 5}, {5, 1}: 0.00058552ms, 0.00501867% main:@2027 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](AfxiFXbSj3:0) -> float_type, {64, 4}, {5, 1}: 0.00068026ms, 0.00583072% AvlwbHMnuE:0 = @param:AvlwbHMnuE:0 -> float_type, {64, 17}, {17, 1}: 0.00053148ms, 0.00455548% main:@2029 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](AvlwbHMnuE:0) -> float_type, {64, 8}, {17, 1}: 0.00070784ms, 0.00606711% Nb9eDKFYZp:0 = @param:Nb9eDKFYZp:0 -> float_type, {64, 5}, {5, 1}: 0.0005623ms, 0.00481965% main:@2031 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](Nb9eDKFYZp:0) -> float_type, {64, 4}, {5, 1}: 0.00068514ms, 0.00587255% KH4GEDx8pO:0 = @param:KH4GEDx8pO:0 -> float_type, {64, 13}, {13, 1}: 0.00052702ms, 0.00451725% main:@2033 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](KH4GEDx8pO:0) -> float_type, {64, 4}, {13, 1}: 0.00068324ms, 0.00585626% CrB2gE9Adp:0 = @param:CrB2gE9Adp:0 -> float_type, {64, 13}, {13, 1}: 0.00053646ms, 0.00459816% main:@2035 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](CrB2gE9Adp:0) -> float_type, {64, 4}, {13, 1}: 0.00067572ms, 0.0057918% uhcrWfUe6M:0 = @param:uhcrWfUe6M:0 -> float_type, {64, 9}, {9, 1}: 0.00057348ms, 0.00491547% main:@2037 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](uhcrWfUe6M:0) -> float_type, {64, 8}, {9, 1}: 0.00072618ms, 0.00622431% main:@2038 = slice[axes={0, 1},starts={0, 33},ends={64, 41}](n4iptOVFwE:0) -> float_type, {64, 8}, {113, 1}: 0.00068632ms, 0.00588266% j25yTFI0eZ:0 = @param:j25yTFI0eZ:0 -> float_type, {64, 9}, {9, 1}: 0.00065962ms, 0.00565381% main:@2040 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](j25yTFI0eZ:0) -> float_type, {64, 8}, {9, 1}: 0.00067708ms, 0.00580346% dMO2bknDYA:0 = @param:dMO2bknDYA:0 -> float_type, {64, 481}, {481, 1}: 0.0006343ms, 0.00543678% main:@2042 = slice[axes={0, 1},starts={0, 233},ends={64, 241}](dMO2bknDYA:0) -> float_type, {64, 8}, {481, 1}: 0.00068472ms, 0.00586895% EJxze1pkRu:0 = @param:EJxze1pkRu:0 -> float_type, {64, 13}, {13, 1}: 0.00055194ms, 0.00473085% main:@2044 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](EJxze1pkRu:0) -> float_type, {64, 4}, {13, 1}: 0.00068402ms, 0.00586295% DilwT5jYf7:0 = @param:DilwT5jYf7:0 -> float_type, {64, 9}, {9, 1}: 0.00050956ms, 0.0043676% main:@2046 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](DilwT5jYf7:0) -> float_type, {64, 8}, {9, 1}: 0.00069282ms, 0.00593837% main:@2047 = slice[axes={0, 1},starts={0, 33},ends={64, 37}](Qp3K1PHvte:0) -> float_type, {64, 4}, {69, 1}: 0.00066352ms, 0.00568723% EyFb0qo6VX:0 = @param:EyFb0qo6VX:0 -> float_type, {64, 5}, {5, 1}: 0.00055458ms, 0.00475348% main:@2049 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](EyFb0qo6VX:0) -> float_type, {64, 4}, {5, 1}: 0.00066844ms, 0.0057294% main:@2050 = slice[axes={0, 1},starts={0, 5},ends={64, 13}](mIR0ToEnMH:0) -> float_type, {64, 8}, {13, 1}: 0.00068994ms, 0.00591369% F6DhiS9CN5:0 = @param:F6DhiS9CN5:0 -> float_type, {64, 9}, {9, 1}: 0.00057064ms, 0.00489113% main:@2052 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](F6DhiS9CN5:0) -> float_type, {64, 8}, {9, 1}: 0.00068374ms, 0.00586055% NKT10PBtly:0 = @param:NKT10PBtly:0 -> float_type, {64, 5}, {5, 1}: 0.00055274ms, 0.0047377% main:@2054 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](NKT10PBtly:0) -> float_type, {64, 4}, {5, 1}: 0.0006843ms, 0.00586535% zOu1aQKy4V:0 = @param:zOu1aQKy4V:0 -> float_type, {64, 9}, {9, 1}: 0.0005276ms, 0.00452222% main:@2056 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](zOu1aQKy4V:0) -> float_type, {64, 8}, {9, 1}: 0.00070136ms, 0.00601157% EsROupwkSG:0 = @param:EsROupwkSG:0 -> float_type, {64, 13}, {13, 1}: 0.00053172ms, 0.00455754% main:@2058 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](EsROupwkSG:0) -> float_type, {64, 4}, {13, 1}: 0.00066086ms, 0.00566443% FrTLGaKo7c:0 = @param:FrTLGaKo7c:0 -> float_type, {64, 13}, {13, 1}: 0.0005217ms, 0.00447165% main:@2060 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](FrTLGaKo7c:0) -> float_type, {64, 4}, {13, 1}: 0.00066244ms, 0.00567798% mDpoFQLqWK:0 = @param:mDpoFQLqWK:0 -> float_type, {64, 13}, {13, 1}: 0.00054728ms, 0.0046909% main:@2062 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](mDpoFQLqWK:0) -> float_type, {64, 4}, {13, 1}: 0.0006473ms, 0.00554821% HFesjVwyB6:0 = @param:HFesjVwyB6:0 -> float_type, {64, 9}, {9, 1}: 0.0005501ms, 0.00471508% main:@2064 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](HFesjVwyB6:0) -> float_type, {64, 8}, {9, 1}: 0.00070282ms, 0.00602409% HKq2ZSlA3D:0 = @param:HKq2ZSlA3D:0 -> float_type, {64, 9}, {9, 1}: 0.00060768ms, 0.00520861% main:@2066 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](HKq2ZSlA3D:0) -> float_type, {64, 8}, {9, 1}: 0.0006675ms, 0.00572135% HLQJSVgFyi:0 = @param:HLQJSVgFyi:0 -> float_type, {64, 9}, {9, 1}: 0.0005484ms, 0.0047005% main:@2068 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](HLQJSVgFyi:0) -> float_type, {64, 8}, {9, 1}: 0.00067964ms, 0.0058254% cNgUKoOFB1:0 = @param:cNgUKoOFB1:0 -> float_type, {64, 9}, {9, 1}: 0.00057654ms, 0.0049417% main:@2070 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](cNgUKoOFB1:0) -> float_type, {64, 8}, {9, 1}: 0.00066928ms, 0.0057366% nIJL1FUMdk:0 = @param:nIJL1FUMdk:0 -> float_type, {64, 9}, {9, 1}: 0.0005819ms, 0.00498764% main:@2072 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](nIJL1FUMdk:0) -> float_type, {64, 8}, {9, 1}: 0.0006998ms, 0.0059982% main:@2073 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](ljyhc7HCLk:0) -> float_type, {64, 4}, {21, 1}: 0.00065482ms, 0.00561266% HyzMmXUNAP:0 = @param:HyzMmXUNAP:0 -> float_type, {64, 13}, {13, 1}: 0.00051224ms, 0.00439057% main:@2075 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](HyzMmXUNAP:0) -> float_type, {64, 4}, {13, 1}: 0.00065946ms, 0.00565243% main:@2076 = slice[axes={0, 1},starts={0, 33},ends={64, 37}](HzcyUlm8Yj:0) -> float_type, {64, 4}, {45, 1}: 0.00064656ms, 0.00554186% fvGJwyjlm4:0 = @param:fvGJwyjlm4:0 -> float_type, {64, 13}, {13, 1}: 0.0004969ms, 0.00425908% main:@2078 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](fvGJwyjlm4:0) -> float_type, {64, 4}, {13, 1}: 0.00068916ms, 0.005907% IQawVv9sqx:0 = @param:IQawVv9sqx:0 -> float_type, {64, 9}, {9, 1}: 0.00060134ms, 0.00515427% main:@2080 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](IQawVv9sqx:0) -> float_type, {64, 8}, {9, 1}: 0.00066336ms, 0.00568586% main:@2081 = slice[axes={0, 1},starts={0, 321},ends={64, 501}](lyiDxfzZg3:0) -> float_type, {64, 180}, {701, 1}: 0.00064968ms, 0.00556861% fKsuaR6OE2:0 = @param:fKsuaR6OE2:0 -> float_type, {64, 13}, {13, 1}: 0.00052676ms, 0.00451502% main:@2083 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](fKsuaR6OE2:0) -> float_type, {64, 4}, {13, 1}: 0.00072128ms, 0.00618231% main:@2084 = slice[axes={0, 1},starts={0, 9},ends={64, 17}](gdck97AK4w:0) -> float_type, {64, 8}, {17, 1}: 0.00066326ms, 0.00568501% JKWg9U1qmT:0 = @param:JKWg9U1qmT:0 -> float_type, {64, 13}, {13, 1}: 0.00055186ms, 0.00473016% main:@2086 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](JKWg9U1qmT:0) -> float_type, {64, 4}, {13, 1}: 0.0006487ms, 0.00556021% hADUwpE6P7:0 = @param:hADUwpE6P7:0 -> float_type, {64, 9}, {9, 1}: 0.0005119ms, 0.00438765% main:@2088 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](hADUwpE6P7:0) -> float_type, {64, 8}, {9, 1}: 0.00068024ms, 0.00583055% YMxBt9210u:0 = @param:YMxBt9210u:0 -> float_type, {64, 5}, {5, 1}: 0.00050422ms, 0.00432182% main:@2090 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](YMxBt9210u:0) -> float_type, {64, 4}, {5, 1}: 0.00067574ms, 0.00579197% KBG8U2uYOC:0 = @param:KBG8U2uYOC:0 -> float_type, {64, 9}, {9, 1}: 0.0006372ms, 0.00546164% main:@2092 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](KBG8U2uYOC:0) -> float_type, {64, 8}, {9, 1}: 0.00071386ms, 0.00611871% main:@2093 = slice[axes={0, 1},starts={0, 5},ends={64, 13}](KFhBukd7EV:0) -> float_type, {64, 8}, {13, 1}: 0.00068024ms, 0.00583055% KGspl1Fzg0:0 = @param:KGspl1Fzg0:0 -> float_type, {64, 9}, {9, 1}: 0.00057694ms, 0.00494513% main:@2095 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](KGspl1Fzg0:0) -> float_type, {64, 8}, {9, 1}: 0.00071246ms, 0.00610671% main:@2096 = slice[axes={0, 1},starts={0, 161},ends={64, 193}](Kanv8c4Yl7:0) -> float_type, {64, 32}, {201, 1}: 0.00066226ms, 0.00567643% XevLAmJgSy:0 = @param:XevLAmJgSy:0 -> float_type, {64, 5}, {5, 1}: 0.00055038ms, 0.00471748% main:@2098 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](XevLAmJgSy:0) -> float_type, {64, 4}, {5, 1}: 0.00067556ms, 0.00579043% KOdpo5kqsI:0 = @param:KOdpo5kqsI:0 -> float_type, {64, 93}, {93, 1}: 0.00055458ms, 0.00475348% main:@2100 = slice[axes={0, 1},starts={0, 33},ends={64, 49}](KOdpo5kqsI:0) -> float_type, {64, 16}, {93, 1}: 0.0006863ms, 0.00588249% main:@2101 = slice[axes={0, 1},starts={0, 129},ends={64, 161}](Kanv8c4Yl7:0) -> float_type, {64, 32}, {201, 1}: 0.0006724ms, 0.00576335% main:@2102 = slice[axes={0, 1},starts={0, 1},ends={64, 33}](KOdpo5kqsI:0) -> float_type, {64, 32}, {93, 1}: 0.00064408ms, 0.00552061% main:@2103 = slice[axes={0, 1},starts={0, 65},ends={64, 97}](Kanv8c4Yl7:0) -> float_type, {64, 32}, {201, 1}: 0.00066356ms, 0.00568758% main:@2104 = slice[axes={0, 1},starts={0, 97},ends={64, 129}](Kanv8c4Yl7:0) -> float_type, {64, 32}, {201, 1}: 0.00064686ms, 0.00554444% M4pau3Zg7l:0 = @param:M4pau3Zg7l:0 -> float_type, {64, 13}, {13, 1}: 0.00053304ms, 0.00456885% main:@2106 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](M4pau3Zg7l:0) -> float_type, {64, 4}, {13, 1}: 0.00072186ms, 0.00618728% Ya2E8PbliH:0 = @param:Ya2E8PbliH:0 -> float_type, {64, 9}, {9, 1}: 0.00055608ms, 0.00476633% main:@2108 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](Ya2E8PbliH:0) -> float_type, {64, 8}, {9, 1}: 0.00066686ms, 0.00571586% NFBzi76hoK:0 = @param:NFBzi76hoK:0 -> float_type, {64, 73}, {73, 1}: 0.00057482ms, 0.00492696% main:@2110 = slice[axes={0, 1},starts={0, 1},ends={64, 65}](NFBzi76hoK:0) -> float_type, {64, 64}, {73, 1}: 0.0006597ms, 0.00565449% main:@2111 = slice[axes={0, 1},starts={0, 33},ends={64, 37}](duyRoIEkH8:0) -> float_type, {64, 4}, {45, 1}: 0.00066256ms, 0.00567901% OfNJgEDtw2:0 = @param:OfNJgEDtw2:0 -> float_type, {64, 13}, {13, 1}: 0.0005565ms, 0.00476993% main:@2113 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](OfNJgEDtw2:0) -> float_type, {64, 4}, {13, 1}: 0.00067112ms, 0.00575238% main:@2114 = slice[axes={0, 1},starts={0, 9},ends={64, 17}](Pu1O5qFxY6:0) -> float_type, {64, 8}, {17, 1}: 0.000666ms, 0.00570849% PxhmZizCI5:0 = @param:PxhmZizCI5:0 -> float_type, {64, 13}, {13, 1}: 0.00050804ms, 0.00435457% main:@2116 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](PxhmZizCI5:0) -> float_type, {64, 4}, {13, 1}: 0.0007109ms, 0.00609334% QMjOWELyPk:0 = @param:QMjOWELyPk:0 -> float_type, {64, 9}, {9, 1}: 0.00049316ms, 0.00422703% main:@2118 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](QMjOWELyPk:0) -> float_type, {64, 8}, {9, 1}: 0.00068664ms, 0.0058854% mQnuwGeC8a:0 = @param:mQnuwGeC8a:0 -> float_type, {64, 9}, {9, 1}: 0.00055564ms, 0.00476256% main:@2120 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](mQnuwGeC8a:0) -> float_type, {64, 8}, {9, 1}: 0.000673ms, 0.00576849% main:@2121 = slice[axes={0, 1},starts={0, 9},ends={64, 13}](RHo9Jz6KpI:0) -> float_type, {64, 4}, {21, 1}: 0.00067734ms, 0.00580569% RNGK6g4bmu:0 = @param:RNGK6g4bmu:0 -> float_type, {64, 9}, {9, 1}: 0.00053576ms, 0.00459216% main:@2123 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](RNGK6g4bmu:0) -> float_type, {64, 8}, {9, 1}: 0.00064498ms, 0.00552832% V84toiBOPm:0 = @param:V84toiBOPm:0 -> float_type, {64, 9}, {9, 1}: 0.00051464ms, 0.00441114% main:@2125 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](V84toiBOPm:0) -> float_type, {64, 8}, {9, 1}: 0.00073844ms, 0.0063294% main:@2126 = slice[axes={0, 1},starts={0, 5},ends={64, 13}](S5AzoJmqUC:0) -> float_type, {64, 8}, {13, 1}: 0.00072314ms, 0.00619825% main:@2127 = slice[axes={0, 1},starts={0, 165},ends={64, 181}](nv2tzWRf1I:0) -> float_type, {64, 16}, {301, 1}: 0.0006769ms, 0.00580192% main:@2128 = slice[axes={0, 1},starts={0, 17},ends={64, 21}](YbIVz8l1un:0) -> float_type, {64, 4}, {45, 1}: 0.0006566ms, 0.00562792% main:@2129 = slice[axes={0, 1},starts={0, 17},ends={64, 21}](bJsIAzhYc3:0) -> float_type, {64, 4}, {37, 1}: 0.0006972ms, 0.00597592% main:@2130 = slice[axes={0, 1},starts={0, 53},ends={64, 57}](u2OJ7kfglD:0) -> float_type, {64, 4}, {93, 1}: 0.00067014ms, 0.00574398% TkHQvxKeD6:0 = @param:TkHQvxKeD6:0 -> float_type, {64, 9}, {9, 1}: 0.0004825ms, 0.00413566% main:@2132 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](TkHQvxKeD6:0) -> float_type, {64, 8}, {9, 1}: 0.00065996ms, 0.00565672% UTzDZk1RS4:0 = @param:UTzDZk1RS4:0 -> float_type, {64, 13}, {13, 1}: 0.00051328ms, 0.00439948% main:@2134 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](UTzDZk1RS4:0) -> float_type, {64, 4}, {13, 1}: 0.00066338ms, 0.00568603% main:@2135 = slice[axes={0, 1},starts={0, 5},ends={64, 13}](qeZ76Hc03v:0) -> float_type, {64, 8}, {13, 1}: 0.00071464ms, 0.0061254% bhuXRdz0E5:0 = @param:bhuXRdz0E5:0 -> float_type, {64, 9}, {9, 1}: 0.00064186ms, 0.00550158% main:@2137 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](bhuXRdz0E5:0) -> float_type, {64, 8}, {9, 1}: 0.00070112ms, 0.00600951% ToDLE1sRI6:0 = @param:ToDLE1sRI6:0 -> float_type, {64, 9}, {9, 1}: 0.00055656ms, 0.00477045% main:@2139 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](ToDLE1sRI6:0) -> float_type, {64, 8}, {9, 1}: 0.00069266ms, 0.005937% zvFYW45wMr:0 = @param:zvFYW45wMr:0 -> float_type, {64, 9}, {9, 1}: 0.00055588ms, 0.00476462% main:@2141 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](zvFYW45wMr:0) -> float_type, {64, 8}, {9, 1}: 0.00065526ms, 0.00561643% main:@2142 = slice[axes={0, 1},starts={0, 33},ends={64, 41}](1aXQU27NrD:0) -> float_type, {64, 8}, {49, 1}: 0.00069246ms, 0.00593529% uGCZaoYr3L:0 = @param:uGCZaoYr3L:0 -> float_type, {64, 13}, {13, 1}: 0.00049744ms, 0.00426371% main:@2144 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](uGCZaoYr3L:0) -> float_type, {64, 4}, {13, 1}: 0.0007006ms, 0.00600506% ihcEmsYo9g:0 = @param:ihcEmsYo9g:0 -> float_type, {64, 5}, {5, 1}: 0.00055158ms, 0.00472776% main:@2146 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](ihcEmsYo9g:0) -> float_type, {64, 4}, {5, 1}: 0.0006563ms, 0.00562535% vi7GCYp4FT:0 = @param:vi7GCYp4FT:0 -> float_type, {64, 9}, {9, 1}: 0.00052726ms, 0.00451931% main:@2148 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](vi7GCYp4FT:0) -> float_type, {64, 8}, {9, 1}: 0.0007121ms, 0.00610363% ihKZd8ANk7:0 = @param:ihKZd8ANk7:0 -> float_type, {64, 9}, {9, 1}: 0.00052924ms, 0.00453628% main:@2150 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](ihKZd8ANk7:0) -> float_type, {64, 8}, {9, 1}: 0.00062984ms, 0.00539855% kX6xhVY7bJ:0 = @param:kX6xhVY7bJ:0 -> float_type, {64, 13}, {13, 1}: 0.00055782ms, 0.00478125% main:@2152 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](kX6xhVY7bJ:0) -> float_type, {64, 4}, {13, 1}: 0.00065934ms, 0.00565141% ad9BJGx5jl:0 = @param:ad9BJGx5jl:0 -> float_type, {64, 13}, {13, 1}: 0.0005602ms, 0.00480165% main:@2154 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](ad9BJGx5jl:0) -> float_type, {64, 4}, {13, 1}: 0.0006494ms, 0.00556621% alvtrHJ95G:0 = @param:alvtrHJ95G:0 -> float_type, {64, 13}, {13, 1}: 0.00045988ms, 0.00394177% main:@2156 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](alvtrHJ95G:0) -> float_type, {64, 4}, {13, 1}: 0.00071132ms, 0.00609694% aocCEyfXIS:0 = @param:aocCEyfXIS:0 -> float_type, {64, 93}, {93, 1}: 0.00052916ms, 0.00453559% main:@2158 = slice[axes={0, 1},starts={0, 1},ends={64, 33}](aocCEyfXIS:0) -> float_type, {64, 32}, {93, 1}: 0.00063622ms, 0.00545324% main:@2159 = slice[axes={0, 1},starts={0, 33},ends={64, 49}](aocCEyfXIS:0) -> float_type, {64, 16}, {93, 1}: 0.00065922ms, 0.00565038% bsFqHW5i6g:0 = @param:bsFqHW5i6g:0 -> float_type, {64, 9}, {9, 1}: 0.00064566ms, 0.00553415% main:@2161 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](bsFqHW5i6g:0) -> float_type, {64, 8}, {9, 1}: 0.00065214ms, 0.00558969% main:@2162 = slice[axes={0, 1},starts={0, 17},ends={64, 33}](cHzIDJm6us:0) -> float_type, {64, 16}, {33, 1}: 0.0006678ms, 0.00572392% gwNpMszEo3:0 = @param:gwNpMszEo3:0 -> float_type, {64, 5}, {5, 1}: 0.00054052ms, 0.00463296% main:@2164 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](gwNpMszEo3:0) -> float_type, {64, 4}, {5, 1}: 0.0006726ms, 0.00576506% ceq9LkmQA5:0 = @param:ceq9LkmQA5:0 -> float_type, {64, 9}, {9, 1}: 0.00057074ms, 0.00489199% main:@2166 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](ceq9LkmQA5:0) -> float_type, {64, 8}, {9, 1}: 0.00069022ms, 0.00591609% main:@2167 = slice[axes={0, 1},starts={0, 33},ends={64, 37}](jznORDJr0N:0) -> float_type, {64, 4}, {45, 1}: 0.00068218ms, 0.00584717% main:@2168 = slice[axes={0, 1},starts={0, 5},ends={64, 13}](gU9SoiHyRO:0) -> float_type, {64, 8}, {13, 1}: 0.0007051ms, 0.00604363% main:@2169 = slice[axes={0, 1},starts={0, 65},ends={64, 129}](lyiDxfzZg3:0) -> float_type, {64, 64}, {701, 1}: 0.0006696ms, 0.00573935% main:@2170 = slice[axes={0, 1},starts={0, 33},ends={64, 37}](zIoqYe9cRb:0) -> float_type, {64, 4}, {45, 1}: 0.00066826ms, 0.00572786% gHPTxdcQWX:0 = @param:gHPTxdcQWX:0 -> float_type, {64, 9}, {9, 1}: 0.00047428ms, 0.0040652% main:@2172 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](gHPTxdcQWX:0) -> float_type, {64, 8}, {9, 1}: 0.00071744ms, 0.0061494% main:@2173 = slice[axes={0, 1},starts={0, 57},ends={64, 61}](u2OJ7kfglD:0) -> float_type, {64, 4}, {93, 1}: 0.00066704ms, 0.0057174% main:@2174 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](w04Keh2FJu:0) -> float_type, {64, 8}, {13, 1}: 0.00066688ms, 0.00571603% main:@2175 = slice[axes={0, 1},starts={0, 177},ends={64, 209}](uFSOc5Chp2:0) -> float_type, {64, 32}, {481, 1}: 0.000676ms, 0.0057942% sfqQKWXEgN:0 = @param:sfqQKWXEgN:0 -> float_type, {64, 5}, {5, 1}: 0.00049028ms, 0.00420234% main:@2177 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](sfqQKWXEgN:0) -> float_type, {64, 4}, {5, 1}: 0.00069106ms, 0.00592329% main:@2178 = slice[axes={0, 1},starts={0, 33},ends={64, 37}](dAFYcGrfEl:0) -> float_type, {64, 4}, {45, 1}: 0.00066804ms, 0.00572598% tZOFD39jRS:0 = @param:tZOFD39jRS:0 -> float_type, {64, 9}, {9, 1}: 0.000576ms, 0.00493707% main:@2180 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](tZOFD39jRS:0) -> float_type, {64, 8}, {9, 1}: 0.00067902ms, 0.00582009% vPaCFg3sAl:0 = @param:vPaCFg3sAl:0 -> float_type, {64, 13}, {13, 1}: 0.00051492ms, 0.00441354% main:@2182 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](vPaCFg3sAl:0) -> float_type, {64, 4}, {13, 1}: 0.00072458ms, 0.0062106% main:@2183 = slice[axes={0, 1},starts={0, 33},ends={64, 49}](u2OJ7kfglD:0) -> float_type, {64, 16}, {93, 1}: 0.00065008ms, 0.00557204% fpcFBPsTyk:0 = @param:fpcFBPsTyk:0 -> float_type, {64, 9}, {9, 1}: 0.0005301ms, 0.00454365% main:@2185 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](fpcFBPsTyk:0) -> float_type, {64, 8}, {9, 1}: 0.00068262ms, 0.00585095% main:@2186 = slice[axes={0, 1},starts={0, 129},ends={64, 193}](lyiDxfzZg3:0) -> float_type, {64, 64}, {701, 1}: 0.0006584ms, 0.00564335% uzaUY08Rsp:0 = @param:uzaUY08Rsp:0 -> float_type, {64, 9}, {9, 1}: 0.00051482ms, 0.00441268% main:@2188 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](uzaUY08Rsp:0) -> float_type, {64, 8}, {9, 1}: 0.00071138ms, 0.00609746% main:@2189 = slice[axes={0, 1},starts={0, 5},ends={64, 13}](whRnWGgf0Q:0) -> float_type, {64, 8}, {13, 1}: 0.00068636ms, 0.005883% sOz8pkBZ5n:0 = @param:sOz8pkBZ5n:0 -> float_type, {64, 5}, {5, 1}: 0.0005825ms, 0.00499279% main:@2191 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](sOz8pkBZ5n:0) -> float_type, {64, 4}, {5, 1}: 0.00071458ms, 0.00612488% sXCjeM94Zu:0 = @param:sXCjeM94Zu:0 -> float_type, {64, 5}, {5, 1}: 0.00049102ms, 0.00420868% main:@2193 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](sXCjeM94Zu:0) -> float_type, {64, 4}, {5, 1}: 0.00071256ms, 0.00610757% main:@2194 = slice[axes={0, 1},starts={0, 17},ends={64, 25}](l5WSX3VvMK:0) -> float_type, {64, 8}, {25, 1}: 0.00066598ms, 0.00570832% main:@2195 = slice[axes={0, 1},starts={0, 49},ends={64, 53}](u2OJ7kfglD:0) -> float_type, {64, 4}, {93, 1}: 0.0006345ms, 0.00543849% dXvEYVC4st:0 = @param:dXvEYVC4st:0 -> float_type, {64, 13}, {13, 1}: 0.00051306ms, 0.00439759% main:@2197 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](dXvEYVC4st:0) -> float_type, {64, 4}, {13, 1}: 0.00065662ms, 0.00562809% dz1saATctE:0 = @param:dz1saATctE:0 -> float_type, {64, 9}, {9, 1}: 0.00057374ms, 0.0049177% main:@2199 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](dz1saATctE:0) -> float_type, {64, 8}, {9, 1}: 0.00071266ms, 0.00610843% main:@2200 = slice[axes={0, 1},starts={0, 1},ends={64, 65}](lyiDxfzZg3:0) -> float_type, {64, 64}, {701, 1}: 0.00064634ms, 0.00553998% v4fNptDqcL:0 = @param:v4fNptDqcL:0 -> float_type, {64, 9}, {9, 1}: 0.00056018ms, 0.00480147% main:@2202 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](v4fNptDqcL:0) -> float_type, {64, 8}, {9, 1}: 0.0006575ms, 0.00563563% main:@2203 = slice[axes={0, 1},starts={0, 5},ends={64, 13}](d1ExPaGqMN:0) -> float_type, {64, 8}, {13, 1}: 0.0006584ms, 0.00564335% pIb4NdDSn5:0 = @param:pIb4NdDSn5:0 -> float_type, {64, 9}, {9, 1}: 0.00059194ms, 0.0050737% main:@2205 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](pIb4NdDSn5:0) -> float_type, {64, 8}, {9, 1}: 0.00071432ms, 0.00612266% dR9G1bayok:0 = @param:dR9G1bayok:0 -> float_type, {64, 13}, {13, 1}: 0.00052098ms, 0.00446548% main:@2207 = slice[axes={0, 1},starts={0, 1},ends={64, 5}](dR9G1bayok:0) -> float_type, {64, 4}, {13, 1}: 0.00069414ms, 0.00594969% fqAX8l29PG:0 = @param:fqAX8l29PG:0 -> float_type, {64, 9}, {9, 1}: 0.00046562ms, 0.00399097% main:@2209 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](fqAX8l29PG:0) -> float_type, {64, 8}, {9, 1}: 0.00068952ms, 0.00591009% main:@2210 = slice[axes={0, 1},starts={0, 33},ends={64, 41}](mgnGNH4Mz3:0) -> float_type, {64, 8}, {49, 1}: 0.00068572ms, 0.00587752% main:@2211 = slice[axes={0, 1},starts={0, 1},ends={64, 33}](u2OJ7kfglD:0) -> float_type, {64, 32}, {93, 1}: 0.00067308ms, 0.00576918% main:@2212 = slice[axes={0, 1},starts={0, 65},ends={64, 129}](nv2tzWRf1I:0) -> float_type, {64, 64}, {301, 1}: 0.00066784ms, 0.00572426% sjKQJGby31:0 = @param:sjKQJGby31:0 -> float_type, {64, 9}, {9, 1}: 0.00059642ms, 0.0051121% main:@2214 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](sjKQJGby31:0) -> float_type, {64, 8}, {9, 1}: 0.00071954ms, 0.0061674% main:@2215 = slice[axes={0, 1},starts={0, 5},ends={64, 13}](Vi28Dy4Ll5:0) -> float_type, {64, 8}, {13, 1}: 0.00068274ms, 0.00585197% aKmfBy1Mhi:0 = @param:aKmfBy1Mhi:0 -> float_type, {64, 9}, {9, 1}: 0.00048752ms, 0.00417868% uM7aDfZz1t:0 = @param:uM7aDfZz1t:0 -> float_type, {64, 9}, {9, 1}: 0.00056964ms, 0.00488256% Wiks0GPJMQ:0 = @param:Wiks0GPJMQ:0 -> float_type, {64, 9}, {9, 1}: 0.00047678ms, 0.00408663% main:@2219 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](Wiks0GPJMQ:0) -> float_type, {64, 8}, {9, 1}: 0.00078518ms, 0.00673002% main:@2220 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](aKmfBy1Mhi:0) -> float_type, {64, 8}, {9, 1}: 0.000686ms, 0.00587992% main:@2221 = slice[axes={0, 1},starts={0, 1},ends={64, 9}](uM7aDfZz1t:0) -> float_type, {64, 8}, {9, 1}: 0.00066374ms, 0.00568912% main:@2222 = reshape_lazy[dims={-1, 32}](main:@1952) -> float_type, {64, 32}, {32, 1}: 0.00051744ms, 0.00443514% main:@2223 = reshape_lazy[dims={-1, 32}](main:@1942) -> float_type, {64, 32}, {32, 1}: 0.00049218ms, 0.00421863% main:@2224 = reshape_lazy[dims={-1, 128}](main:@1925) -> float_type, {64, 128}, {128, 1}: 0.00048432ms, 0.00415126% main:@2225 = reshape_lazy[dims={-1, 32}](main:@1944) -> float_type, {64, 32}, {32, 1}: 0.00046486ms, 0.00398446% main:@2226 = load[offset=973056,end=1850880](main:@1) -> float_type, {64, 3429}, {3429, 1}: 0.00047936ms, 0.00410874% main:@2227 = reshape_lazy[dims={-1, 105}](main:@991) -> float_type, {64, 105}, {105, 1}: 0.00048494ms, 0.00415657% main:@2228 = reshape_lazy[dims={-1, 64}](main:@1036) -> float_type, {64, 64}, {64, 1}: 0.00047858ms, 0.00410206% main:@2229 = reshape_lazy[dims={-1, 256}](main:@1927) -> float_type, {64, 256}, {256, 1}: 0.00046496ms, 0.00398531% main:@2230 = reshape_lazy[dims={-1, 64}](main:@1933) -> float_type, {64, 64}, {64, 1}: 0.00046474ms, 0.00398343% main:@2231 = reshape_lazy[dims={-1, 100}](main:@1505) -> float_type, {64, 100}, {100, 1}: 0.00046684ms, 0.00400143% main:@2232 = reshape_lazy[dims={-1, 128}](main:@1930) -> float_type, {64, 128}, {128, 1}: 0.00045764ms, 0.00392257% main:@2233 = reshape_lazy[dims={-1, 84}](main:@1690) -> float_type, {64, 84}, {84, 1}: 0.00046038ms, 0.00394606% main:@2234 = reshape_lazy[dims={-1, 32}](main:@1937) -> float_type, {64, 32}, {32, 1}: 0.00045458ms, 0.00389634% main:@2235 = reshape_lazy[dims={-1, 32}](main:@1938) -> float_type, {64, 32}, {32, 1}: 0.00047174ms, 0.00404343% main:@2236 = 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=1148,local=256,](main:@2200,main:@1884,main:@2169,main:@2212,main:@2186,main:@1886,main:@1888,main:@2110,main:@2211,main:@2102,main:@2158,main:@2227,main:@2233,main:@2231,main:@2081,main:@2127,main:@2175,main:@1996,main:@2047,main:@1951,main:@2167,main:@1954,main:@2018,main:@2076,main:@1959,main:@2111,main:@1949,main:@2058,main:@2210,main:@2142,main:@1988,main:@2170,main:@2178,main:@2007,main:@2113,main:@2129,main:@2128,main:@2084,main:@2114,main:@1956,main:@1955,main:@2024,main:@2194,main:@2025,main:@2205,main:@2174,main:@2037,main:@2199,main:@2121,main:@2029,main:@1979,main:@1976,main:@1958,main:@2017,main:@2073,main:@2183,main:@2100,main:@2159,main:@2103,main:@2104,main:@2101,main:@2096,main:@2195,main:@2130,main:@2173,main:@1936,main:@2232,main:@2054,main:@2049,main:@2164,main:@2177,main:@2027,main:@1985,main:@2191,main:@2031,main:@1941,main:@2193,main:@2224,main:@2228,main:@2015,main:@2009,main:@2011,main:@2019,main:@2016,main:@2010,main:@2012,main:@2078,main:@2086,main:@2014,main:@2062,main:@2083,main:@2033,main:@1947,main:@2106,main:@1991,main:@2044,main:@2002,main:@2060,main:@2134,main:@2156,main:@2152,main:@2116,main:@2182,main:@2035,main:@2154,main:@2222,main:@2230,main:@2223,main:@2207,main:@2144,main:@2197,main:@2075,main:@1993,main:@2132,main:@2004,main:@1998,main:@2052,main:@2098,main:@2090,main:@1981,main:@1995,main:@1970,main:@2146,main:@2042,main:@2139,main:@1968,main:@2137,main:@2219,main:@2209,main:@2095,main:@2148,main:@2161,main:@1943,main:@2188,main:@1978,main:@1966,main:@1987,main:@1964,main:@2000,main:@2185,main:@1963,main:@2080,main:@2021,main:@2172,main:@2214,main:@1974,main:@2150,main:@1932,main:@2092,main:@2221,main:@2162,main:@2070,main:@2066,main:@2141,main:@2088,main:@2064,main:@2068,main:@2006,main:@1939,main:@2234,main:@2056,main:@2135,main:@2215,main:@1953,main:@2072,main:@2050,main:@2166,main:@2203,main:@2126,main:@2189,main:@2180,main:@1989,main:@2046,main:@2220,main:@2040,main:@2123,main:@2108,main:@2118,main:@2125,main:@2023,main:@2120,main:@2202,main:@1983,main:@1972,main:@1961,main:@2038,main:@2168,main:@2093,main:@2235,main:@1950,main:@2225,main:@1945,main:@2229,main:@1934,main:@1113,main:@2226) -> float_type, {64, 3429}, {3429, 1}: 0.231893ms, 1.98763% main:@2237 = hip::hip_copy_literal[id=main:@literal:29] -> float_type, {3429, 2905}, {2905, 1}: 0.00088256ms, 0.00756469% main:@2238 = hip::hip_copy_literal[id=main:@literal:8] -> float_type, {64, 2905}, {2905, 1}: 0.0007101ms, 0.00608649% main:@2239 = load[offset=229376,end=973056](main:@1) -> float_type, {64, 2905}, {2905, 1}: 0.00051362ms, 0.00440239% main:@2240 = gpu::gemm[alpha=1,beta=1,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@2236,main:@2237,main:@2238,main:@2239) -> float_type, {64, 2905}, {2905, 1}: 0.0958618ms, 0.821661% main:@2241 = hip::hip_copy_literal[id=main:@literal:4] -> float_type, {581, 581}, {581, 1}: 0.00081978ms, 0.00702659% main:@2242 = slice[axes={1},starts={1743},ends={2324}](main:@2240) -> float_type, {64, 581}, {2905, 1}: 0.00070056ms, 0.00600471% main:@2243 = load[offset=1121792,end=1270528](main:@1) -> float_type, {64, 581}, {581, 1}: 0.00053376ms, 0.00457502% main:@2244 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@2242,main:@2241,main:@2243) -> float_type, {64, 581}, {581, 1}: 0.0371829ms, 0.318706% main:@2245 = load[offset=973056,end=1121792](main:@1) -> float_type, {64, 581}, {581, 1}: 0.00065734ms, 0.00563426% main:@2246 = gpu::code_object[code_object=13544,symbol_name=pow_mul_add_neg_sub_kernel,global=37184,local=256,](main:@2242,main:@2244,main:@2245) -> float_type, {64, 581}, {581, 1}: 0.0189894ms, 0.162764% main:@2247 = slice[axes={1},starts={1162},ends={1743}](main:@2240) -> float_type, {64, 581}, {2905, 1}: 0.00078798ms, 0.00675402% main:@2248 = slice[axes={1},starts={0},ends={581}](main:@2240) -> float_type, {64, 581}, {2905, 1}: 0.00059944ms, 0.00513798% main:@2249 = load[offset=1121792,end=1270528](main:@1) -> float_type, {64, 581}, {581, 1}: 0.00051794ms, 0.00443942% main:@2250 = gpu::code_object[code_object=13544,symbol_name=pow_pow_sigmoid_add_kernel,global=37184,local=256,](main:@2247,main:@2248,main:@2249) -> float_type, {64, 581}, {581, 1}: 0.0197471ms, 0.169259% main:@2251 = slice[axes={1},starts={2324},ends={2905}](main:@2240) -> float_type, {64, 581}, {2905, 1}: 0.00074246ms, 0.00636385% main:@2252 = slice[axes={1},starts={581},ends={1162}](main:@2240) -> float_type, {64, 581}, {2905, 1}: 0.00058646ms, 0.00502673% main:@2253 = load[offset=1270528,end=1419264](main:@1) -> float_type, {64, 581}, {581, 1}: 0.00047506ms, 0.00407189% main:@2254 = gpu::code_object[code_object=13536,symbol_name=pow_mul_rsqrt_sub_kernel,global=37184,local=256,](main:@2251,main:@2252,main:@2253) -> float_type, {64, 581}, {581, 1}: 0.0189202ms, 0.162171% main:@2255 = load[offset=1419264,end=1419520](main:@1) -> float_type, {64, 1}, {1, 1}: 0.00061804ms, 0.00529741% main:@2256 = gpu::code_object[code_object=13552,symbol_name=reduce_mean_add_rsqrt_add_kernel,global=16384,local=256,](main:@2254,main:@2255) -> float_type, {64, 1}, {1, 1}: 0.0163126ms, 0.13982% main:@2257 = load[offset=1419520,end=1419776](main:@1) -> float_type, {64, 1}, {1, 1}: 0.00061036ms, 0.00523158% main:@2258 = gpu::code_object[code_object=13552,symbol_name=reduce_mean_add_neg_pow_kernel,global=16384,local=256,](main:@2246,main:@2257) -> float_type, {64, 1}, {1, 1}: 0.0163302ms, 0.139971% main:@2259 = load[offset=1419776,end=1420032](main:@1) -> float_type, {64, 1}, {1, 1}: 0.00061116ms, 0.00523844% main:@2260 = gpu::code_object[code_object=13552,symbol_name=reduce_prod_add_rsqrt_sub_kernel,global=16384,local=256,](main:@2250,main:@2259) -> float_type, {64, 1}, {1, 1}: 0.015756ms, 0.13505% main:@2261 = load[offset=2580736,end=3175680](main:@1) -> float_type, {64, 2324}, {2324, 1}: 0.00061988ms, 0.00531318% main:@2262 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=37184,local=256,](main:@2247,main:@2248,main:@2251,main:@2242,main:@2261) -> float_type, {64, 2324}, {2324, 1}: 0.0201025ms, 0.172305% main:@2263 = slice[axes={1},starts={0},ends={4070}](main:@1510) -> float_type, {64, 4070}, {4508, 1}: 0.00076636ms, 0.00656871% main:@2264 = load[offset=1538816,end=2580736](main:@1) -> float_type, {64, 4070}, {4070, 1}: 0.00049724ms, 0.004262% main:@2265 = gpu::code_object[code_object=21808,symbol_name=leaky_relu_noop_noop_noop_noop_noop_noop_concat_rsqrt_pow_kernel,global=37211,local=256,](main:@2262,main:@2250,main:@2254,main:@2246,main:@2260,main:@2256,main:@2258,main:@2263,main:@2264) -> float_type, {64, 4070}, {4070, 1}: 0.0368153ms, 0.315555% main:@2266 = hip::hip_copy_literal[id=main:@literal:1] -> float_type, {64, 2130}, {2130, 1}: 0.00074548ms, 0.00638974% main:@2267 = hip::hip_copy_literal[id=main:@literal:32] -> float_type, {4070, 2130}, {2130, 1}: 0.00056414ms, 0.00483542% main:@2268 = load[offset=993536,end=1538816](main:@1) -> float_type, {64, 2130}, {2130, 1}: 0.00046762ms, 0.00400811% main:@2269 = gpu::gemm[alpha=1,beta=1,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@2265,main:@2267,main:@2266,main:@2268) -> float_type, {64, 2130}, {2130, 1}: 0.0882033ms, 0.756018% main:@2270 = hip::hip_copy_literal[id=main:@literal:92] -> float_type, {426, 426}, {426, 1}: 0.00074334ms, 0.0063714% main:@2271 = slice[axes={1},starts={852},ends={1278}](main:@2269) -> float_type, {64, 426}, {2130, 1}: 0.00062108ms, 0.00532347% main:@2272 = load[offset=1647872,end=1756928](main:@1) -> float_type, {64, 426}, {426, 1}: 0.00047918ms, 0.0041072% 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, {64, 426}, {426, 1}: 0.0315771ms, 0.270657% main:@2274 = hip::hip_copy_literal[id=main:@literal:18] -> float_type, {64, 2}, {2, 1}: 0.00082564ms, 0.00707681% main:@2275 = hip::hip_copy_literal[id=main:@literal:114] -> float_type, {128, 2985}, {2985, 1}: 0.00059778ms, 0.00512376% main:@2276 = load[offset=229376,end=993536](main:@1) -> float_type, {64, 2985}, {2985, 1}: 0.00048242ms, 0.00413497% main:@2277 = slice[axes={1},starts={768},ends={896}](main:@717) -> float_type, {64, 128}, {896, 1}: 0.00060164ms, 0.00515684% main:@2278 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@2277,main:@2275,main:@2276) -> float_type, {64, 2985}, {2985, 1}: 0.0266309ms, 0.228262% main:@2279 = load[offset=1538816,end=1647872](main:@1) -> float_type, {64, 426}, {426, 1}: 0.00063828ms, 0.00547089% main:@2280 = gpu::code_object[code_object=13544,symbol_name=sub_sub_mul_rsqrt_sub_kernel,global=13632,local=256,](main:@2271,main:@2273,main:@2279) -> float_type, {64, 426}, {426, 1}: 0.0155464ms, 0.133253% main:@2281 = slice[axes={1},starts={1704},ends={2130}](main:@2269) -> float_type, {64, 426}, {2130, 1}: 0.00074004ms, 0.00634311% main:@2282 = slice[axes={1},starts={426},ends={852}](main:@2269) -> float_type, {64, 426}, {2130, 1}: 0.00058172ms, 0.0049861% main:@2283 = load[offset=1647872,end=1756928](main:@1) -> float_type, {64, 426}, {426, 1}: 0.00048558ms, 0.00416206% main:@2284 = gpu::code_object[code_object=13536,symbol_name=div_mul_neg_pow_kernel,global=13632,local=256,](main:@2282,main:@2281,main:@2283) -> float_type, {64, 426}, {426, 1}: 0.015979ms, 0.136961% main:@2285 = slice[axes={1},starts={1278},ends={1704}](main:@2269) -> float_type, {64, 426}, {2130, 1}: 0.00073722ms, 0.00631894% main:@2286 = slice[axes={1},starts={0},ends={426}](main:@2269) -> float_type, {64, 426}, {2130, 1}: 0.00057528ms, 0.0049309% main:@2287 = load[offset=1756928,end=1865984](main:@1) -> float_type, {64, 426}, {426, 1}: 0.0004658ms, 0.00399251% main:@2288 = gpu::code_object[code_object=13536,symbol_name=add_pow_log_pow_kernel,global=13632,local=256,](main:@2285,main:@2286,main:@2287) -> float_type, {64, 426}, {426, 1}: 0.0175951ms, 0.150813% main:@2289 = load[offset=1865984,end=2302208](main:@1) -> float_type, {64, 1704}, {1704, 1}: 0.00062666ms, 0.0053713% main:@2290 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=27264,local=256,](main:@2285,main:@2286,main:@2282,main:@2271,main:@2289) -> float_type, {64, 1704}, {1704, 1}: 0.0192664ms, 0.165139% main:@2291 = load[offset=993792,end=994048](main:@1) -> float_type, {64, 1}, {1, 1}: 0.00063024ms, 0.00540198% main:@2292 = gpu::code_object[code_object=13544,symbol_name=reduce_mean_mul_neg_kernel,global=16384,local=256,](main:@2284,main:@2291) -> float_type, {64, 1}, {1, 1}: 0.0153155ms, 0.131274% main:@2293 = load[offset=993536,end=993792](main:@1) -> float_type, {64, 1}, {1, 1}: 0.0006125ms, 0.00524993% main:@2294 = gpu::code_object[code_object=13560,symbol_name=reduce_mean_add_sigmoid_sub_kernel,global=16384,local=256,](main:@2288,main:@2293) -> float_type, {64, 1}, {1, 1}: 0.0154684ms, 0.132584% main:@2295 = load[offset=994560,end=994816](main:@1) -> float_type, {64, 1}, {1, 1}: 0.0005978ms, 0.00512393% main:@2296 = gpu::code_object[code_object=13544,symbol_name=reduce_prod_mul_log_kernel,global=16384,local=256,](main:@2280,main:@2295) -> float_type, {64, 1}, {1, 1}: 0.0154565ms, 0.132483% main:@2297 = load[offset=994048,end=994560](main:@1) -> float_type, {64, 2}, {2, 1}: 0.00061416ms, 0.00526415% main:@2298 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=64,local=256,](main:@2292,main:@2296,main:@2297) -> float_type, {64, 2}, {2, 1}: 0.0132826ms, 0.113849% main:@2299 = load[offset=2302208,end=3066368](main:@1) -> float_type, {64, 2985}, {2985, 1}: 0.00063754ms, 0.00546455% main:@2300 = gpu::code_object[code_object=21792,symbol_name=leaky_relu_noop_noop_noop_noop_add_concat_log_pow_kernel,global=31840,local=256,](main:@2290,main:@2288,main:@2284,main:@2280,main:@2294,main:@2274,main:@2298,main:@2278,main:@2299) -> float_type, {64, 2985}, {2985, 1}: 0.0317247ms, 0.271922% main:@2301 = hip::hip_copy_literal[id=main:@literal:19] -> float_type, {64, 856}, {856, 1}: 0.00079282ms, 0.0067955% main:@2302 = hip::hip_copy_literal[id=main:@literal:30] -> float_type, {2985, 856}, {856, 1}: 0.00061616ms, 0.0052813% main:@2303 = load[offset=613632,end=832768](main:@1) -> float_type, {64, 856}, {856, 1}: 0.00048106ms, 0.00412331% main:@2304 = gpu::gemm[alpha=1,beta=1,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@2300,main:@2302,main:@2301,main:@2303) -> float_type, {64, 856}, {856, 1}: 0.0663888ms, 0.569039% main:@2305 = hip::hip_copy_literal[id=main:@literal:161] -> float_type, {214, 214}, {214, 1}: 0.00082668ms, 0.00708573% main:@2306 = slice[axes={1},starts={642},ends={856}](main:@2304) -> float_type, {64, 214}, {856, 1}: 0.00063656ms, 0.00545615% main:@2307 = load[offset=887552,end=942336](main:@1) -> float_type, {64, 214}, {214, 1}: 0.0004632ms, 0.00397023% 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, {64, 214}, {214, 1}: 0.0238309ms, 0.204262% main:@2309 = hip::hip_copy_literal[id=main:@literal:43] -> float_type, {128, 1501}, {1501, 1}: 0.00080246ms, 0.00687813% main:@2310 = slice[axes={1},starts={0},ends={128}](main:@717) -> float_type, {64, 128}, {896, 1}: 0.00063472ms, 0.00544038% main:@2311 = load[offset=229376,end=613632](main:@1) -> float_type, {64, 1501}, {1501, 1}: 0.00049462ms, 0.00423954% 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, {64, 1501}, {1501, 1}: 0.023894ms, 0.204802% main:@2313 = hip::hip_copy_literal[id=main:@literal:148] -> float_type, {1}, {0}: 0.0008411ms, 0.00720933% main:@2314 = hip::hip_copy_literal[id=main:@literal:126] -> float_type, {1, 214}, {214, 1}: 0.00066732ms, 0.0057198% main:@2315 = load[offset=832768,end=887552](main:@1) -> float_type, {64, 214}, {214, 1}: 0.0004696ms, 0.00402509% main:@2316 = multibroadcast[out_lens={64, 214}](main:@2314) -> float_type, {64, 214}, {0, 1}: 0.00065984ms, 0.00565569% main:@2317 = gpu::code_object[code_object=17648,symbol_name=pow_sub_div_pow_log_div_kernel,global=6848,local=256,](main:@2308,main:@2316,main:@2306,main:@2315) -> float_type, {64, 214}, {214, 1}: 0.0206176ms, 0.17672% main:@2318 = slice[axes={1},starts={214},ends={428}](main:@2304) -> float_type, {64, 214}, {856, 1}: 0.00076212ms, 0.00653236% main:@2319 = multibroadcast[out_lens={214}](main:@2313) -> float_type, {214}, {0}: 0.00058188ms, 0.00498747% main:@2320 = multibroadcast[out_lens={64, 214}](main:@2319) -> float_type, {64, 214}, {0, 0}: 0.00069624ms, 0.00596769% main:@2321 = load[offset=887552,end=942336](main:@1) -> float_type, {64, 214}, {214, 1}: 0.00050924ms, 0.00436485% main:@2322 = gpu::code_object[code_object=13536,symbol_name=add_add_neg_add_kernel,global=6848,local=256,](main:@2318,main:@2320,main:@2321) -> float_type, {64, 214}, {214, 1}: 0.0145885ms, 0.125042% main:@2323 = slice[axes={1},starts={428},ends={642}](main:@2304) -> float_type, {64, 214}, {856, 1}: 0.00074554ms, 0.00639025% main:@2324 = slice[axes={1},starts={0},ends={214}](main:@2304) -> float_type, {64, 214}, {856, 1}: 0.00056694ms, 0.00485942% main:@2325 = load[offset=942336,end=997120](main:@1) -> float_type, {64, 214}, {214, 1}: 0.00046594ms, 0.00399371% main:@2326 = gpu::code_object[code_object=13536,symbol_name=add_add_rsqrt_pow_kernel,global=6848,local=256,](main:@2324,main:@2323,main:@2320,main:@2325) -> float_type, {64, 214}, {214, 1}: 0.0152778ms, 0.130951% main:@2327 = load[offset=997120,end=1216256](main:@1) -> float_type, {64, 856}, {856, 1}: 0.00062534ms, 0.00535998% main:@2328 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=13696,local=256,](main:@2318,main:@2318,main:@2324,main:@2306,main:@2327) -> float_type, {64, 856}, {856, 1}: 0.0169961ms, 0.145679% main:@2329 = load[offset=613632,end=613888](main:@1) -> float_type, {64, 1}, {1, 1}: 0.00063746ms, 0.00546387% main:@2330 = gpu::code_object[code_object=13552,symbol_name=reduce_prod_pow_neg_sub_kernel,global=8192,local=128,](main:@2317,main:@2329) -> float_type, {64, 1}, {1, 1}: 0.0162472ms, 0.139259% main:@2331 = load[offset=613888,end=614144](main:@1) -> float_type, {64, 1}, {1, 1}: 0.00060494ms, 0.00518513% main:@2332 = gpu::code_object[code_object=13552,symbol_name=reduce_prod_add_log_div_kernel,global=8192,local=128,](main:@2326,main:@2331) -> float_type, {64, 1}, {1, 1}: 0.0155127ms, 0.132964% main:@2333 = load[offset=614144,end=614400](main:@1) -> float_type, {64, 1}, {1, 1}: 0.00058818ms, 0.00504147% main:@2334 = gpu::code_object[code_object=13552,symbol_name=reduce_prod_mul_rsqrt_add_kernel,global=8192,local=128,](main:@2322,main:@2333) -> float_type, {64, 1}, {1, 1}: 0.0153602ms, 0.131657% main:@2335 = hip::hip_copy_literal[id=main:@literal:0] -> float_type, {64, 400}, {400, 1}: 0.00075532ms, 0.00647408% main:@2336 = hip::hip_copy_literal[id=main:@literal:33] -> float_type, {1501, 400}, {400, 1}: 0.00071792ms, 0.00615351% main:@2337 = load[offset=1216256,end=1600512](main:@1) -> float_type, {64, 1501}, {1501, 1}: 0.00046834ms, 0.00401429% main:@2338 = gpu::code_object[code_object=17720,symbol_name=leaky_relu_noop_noop_noop_noop_noop_noop_concat_log_add_pow_kernel,global=13723,local=256,](main:@2328,main:@2322,main:@2326,main:@2317,main:@2334,main:@2332,main:@2330,main:@2312,main:@2337) -> float_type, {64, 1501}, {1501, 1}: 0.0310333ms, 0.265996% main:@2339 = load[offset=409344,end=511744](main:@1) -> float_type, {64, 400}, {400, 1}: 0.0006209ms, 0.00532192% main:@2340 = gpu::gemm[alpha=1,beta=1,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@2338,main:@2336,main:@2335,main:@2339) -> float_type, {64, 400}, {400, 1}: 0.035376ms, 0.303219% main:@2341 = hip::hip_copy_literal[id=main:@literal:52] -> float_type, {128, 703}, {703, 1}: 0.00079116ms, 0.00678128% main:@2342 = load[offset=229376,end=409344](main:@1) -> float_type, {64, 703}, {703, 1}: 0.0005067ms, 0.00434308% main:@2343 = slice[axes={1},starts={256},ends={384}](main:@717) -> float_type, {64, 128}, {896, 1}: 0.0006288ms, 0.00538964% main:@2344 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@2343,main:@2341,main:@2342) -> float_type, {64, 703}, {703, 1}: 0.0231391ms, 0.198332% main:@2345 = hip::hip_copy_literal[id=main:@literal:3] -> float_type, {100, 100}, {100, 1}: 0.00070214ms, 0.00601826% main:@2346 = slice[axes={1},starts={200},ends={300}](main:@2340) -> float_type, {64, 100}, {400, 1}: 0.000625ms, 0.00535707% main:@2347 = load[offset=562944,end=588544](main:@1) -> float_type, {64, 100}, {100, 1}: 0.0004768ms, 0.0040868% main:@2348 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@2346,main:@2345,main:@2347) -> float_type, {64, 100}, {100, 1}: 0.022952ms, 0.196728% main:@2349 = slice[axes={1},starts={300},ends={400}](main:@2340) -> float_type, {64, 100}, {400, 1}: 0.0007274ms, 0.00623477% main:@2350 = slice[axes={1},starts={0},ends={100}](main:@2340) -> float_type, {64, 100}, {400, 1}: 0.00058972ms, 0.00505467% main:@2351 = load[offset=511744,end=537344](main:@1) -> float_type, {64, 100}, {100, 1}: 0.00047958ms, 0.00411063% main:@2352 = gpu::code_object[code_object=13536,symbol_name=add_add_rsqrt_add_kernel,global=3200,local=256,](main:@2349,main:@2350,main:@2351) -> float_type, {64, 100}, {100, 1}: 0.0144618ms, 0.123957% main:@2353 = load[offset=537344,end=562944](main:@1) -> float_type, {64, 100}, {100, 1}: 0.00063148ms, 0.00541261% main:@2354 = gpu::code_object[code_object=13544,symbol_name=sub_sub_pow_log_pow_kernel,global=3200,local=256,](main:@2346,main:@2348,main:@2353) -> float_type, {64, 100}, {100, 1}: 0.0171462ms, 0.146965% main:@2355 = slice[axes={1},starts={100},ends={200}](main:@2340) -> float_type, {64, 100}, {400, 1}: 0.00072906ms, 0.006249% main:@2356 = load[offset=562944,end=588544](main:@1) -> float_type, {64, 100}, {100, 1}: 0.00048322ms, 0.00414183% main:@2357 = gpu::code_object[code_object=13536,symbol_name=add_add_neg_add_kernel,global=3200,local=256,](main:@2355,main:@2350,main:@2356) -> float_type, {64, 100}, {100, 1}: 0.0141505ms, 0.121288% main:@2358 = hip::hip_copy_literal[id=main:@literal:31] -> float_type, {703, 212}, {212, 1}: 0.00073746ms, 0.006321% main:@2359 = hip::hip_copy_literal[id=main:@literal:34] -> float_type, {64, 212}, {212, 1}: 0.00063316ms, 0.00542701% main:@2360 = load[offset=588800,end=589056](main:@1) -> float_type, {64, 1}, {1, 1}: 0.00048772ms, 0.0041804% main:@2361 = gpu::code_object[code_object=14256,symbol_name=reduce_mean_add_sigmoid_kernel,global=4096,local=64,](main:@2352,main:@2360) -> float_type, {64, 1}, {1, 1}: 0.0153628ms, 0.13168% main:@2362 = load[offset=588544,end=588800](main:@1) -> float_type, {64, 1}, {1, 1}: 0.00059498ms, 0.00509976% main:@2363 = gpu::code_object[code_object=14256,symbol_name=reduce_mean_pow_log_pow_kernel,global=4096,local=64,](main:@2354,main:@2362) -> float_type, {64, 1}, {1, 1}: 0.0160703ms, 0.137744% main:@2364 = load[offset=691456,end=691712](main:@1) -> float_type, {64, 1}, {1, 1}: 0.00059858ms, 0.00513061% main:@2365 = gpu::code_object[code_object=14256,symbol_name=reduce_mean_mul_rsqrt_kernel,global=4096,local=64,](main:@2357,main:@2364) -> float_type, {64, 1}, {1, 1}: 0.0152524ms, 0.130733% main:@2366 = load[offset=589056,end=691456](main:@1) -> float_type, {64, 400}, {400, 1}: 0.0005915ms, 0.00506993% main:@2367 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=6400,local=256,](main:@2349,main:@2350,main:@2355,main:@2346,main:@2366) -> float_type, {64, 400}, {400, 1}: 0.0157417ms, 0.134927% main:@2368 = load[offset=409344,end=409856](main:@1) -> float_type, {64, 2}, {2, 1}: 0.00060344ms, 0.00517227% main:@2369 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=64,local=256,](main:@2361,main:@2365,main:@2368) -> float_type, {64, 2}, {2, 1}: 0.0131963ms, 0.11311% main:@2370 = load[offset=691456,end=871424](main:@1) -> float_type, {64, 703}, {703, 1}: 0.00063166ms, 0.00541415% main:@2371 = gpu::code_object[code_object=13600,symbol_name=leaky_relu_noop_noop_noop_add_noop_concat_log_div_kernel,global=7498,local=256,](main:@2367,main:@2352,main:@2357,main:@2354,main:@2274,main:@2369,main:@2363,main:@2344,main:@2370) -> float_type, {64, 703}, {703, 1}: 0.0229629ms, 0.196822% main:@2372 = load[offset=229376,end=283648](main:@1) -> float_type, {64, 212}, {212, 1}: 0.00060566ms, 0.0051913% main:@2373 = gpu::gemm[alpha=1,beta=1,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@2371,main:@2358,main:@2359,main:@2372) -> float_type, {64, 212}, {212, 1}: 0.0379628ms, 0.325391% main:@2374 = hip::hip_copy_literal[id=main:@literal:81] -> float_type, {128, 167}, {167, 1}: 0.00075126ms, 0.00643928% main:@2375 = slice[axes={1},starts={512},ends={640}](main:@717) -> float_type, {64, 128}, {896, 1}: 0.00062758ms, 0.00537918% main:@2376 = load[offset=283648,end=326400](main:@1) -> float_type, {64, 167}, {167, 1}: 0.00046962ms, 0.00402526% 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, {64, 167}, {167, 1}: 0.0222498ms, 0.19071% main:@2378 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](0cjwsL5ldh:0) -> float_type, {64, 1}, {33, 1}: 0.0008531ms, 0.00731218% main:@2379 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](gsP96ZTbn0:0) -> float_type, {64, 1}, {17, 1}: 0.00065796ms, 0.00563958% main:@2380 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](1PJsBhDaCK:0) -> float_type, {64, 1}, {25, 1}: 0.00076806ms, 0.00658328% main:@2381 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](1QWFmJ96aB:0) -> float_type, {64, 1}, {5, 1}: 0.00061568ms, 0.00527718% main:@2382 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](1WNvHElUG7:0) -> float_type, {64, 1}, {25, 1}: 0.00062378ms, 0.00534661% main:@2383 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](2CnSyUBzm0:0) -> float_type, {64, 1}, {65, 1}: 0.00061396ms, 0.00526244% main:@2384 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](2RDXZrx9GA:0) -> float_type, {64, 1}, {45, 1}: 0.00075202ms, 0.00644579% main:@2385 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](2Yy3zxNGBC:0) -> float_type, {64, 1}, {45, 1}: 0.0007122ms, 0.00610448% main:@2386 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](2lVXMZgJfO:0) -> float_type, {64, 1}, {25, 1}: 0.00075844ms, 0.00650082% main:@2387 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](2srtUfGvja:0) -> float_type, {64, 1}, {41, 1}: 0.00071582ms, 0.00613551% main:@2388 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](3VED7WSJ1b:0) -> float_type, {64, 1}, {5, 1}: 0.00084616ms, 0.0072527% main:@2389 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](gHPTxdcQWX:0) -> float_type, {64, 1}, {9, 1}: 0.00068788ms, 0.00589603% main:@2390 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](3kdi0vsSZG:0) -> float_type, {64, 1}, {25, 1}: 0.00075572ms, 0.00647751% main:@2391 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](z3lqgZM8WU:0) -> float_type, {64, 1}, {41, 1}: 0.0006672ms, 0.00571878% main:@2392 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](4M92ryD8ux:0) -> float_type, {64, 1}, {9, 1}: 0.00086332ms, 0.00739978% main:@2393 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](4gFa5o2YsJ:0) -> float_type, {64, 1}, {9, 1}: 0.00071064ms, 0.00609111% main:@2394 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](5yColzZ8K3:0) -> float_type, {64, 1}, {5, 1}: 0.00069752ms, 0.00597866% main:@2395 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](6HSQqXdJzb:0) -> float_type, {64, 1}, {45, 1}: 0.00067914ms, 0.00582112% main:@2396 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](6Wh2Jm7GHC:0) -> float_type, {64, 1}, {5, 1}: 0.00079802ms, 0.00684007% main:@2397 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](6XMrIsBjLY:0) -> float_type, {64, 1}, {13, 1}: 0.0007078ms, 0.00606677% main:@2398 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](72zWmh93Bo:0) -> float_type, {64, 1}, {17, 1}: 0.00066686ms, 0.00571586% main:@2399 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](AB6lekmMxa:0) -> float_type, {64, 1}, {25, 1}: 0.00069644ms, 0.0059694% main:@2400 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](ifP6Ceko5T:0) -> float_type, {64, 1}, {73, 1}: 0.0007997ms, 0.00685447% main:@2401 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](89nc05gQza:0) -> float_type, {64, 1}, {9, 1}: 0.00070828ms, 0.00607089% main:@2402 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](8Dv01FAoNM:0) -> float_type, {64, 1}, {17, 1}: 0.00066428ms, 0.00569375% main:@2403 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](UMO0o14RuI:0) -> float_type, {64, 1}, {9, 1}: 0.00068296ms, 0.00585386% main:@2404 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](8iOST90kWM:0) -> float_type, {64, 1}, {9, 1}: 0.00078738ms, 0.00674888% main:@2405 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](8vM5p4RzXd:0) -> float_type, {64, 1}, {13, 1}: 0.00069248ms, 0.00593546% main:@2406 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](9dgeU4bKka:0) -> float_type, {64, 1}, {17, 1}: 0.00068358ms, 0.00585917% main:@2407 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](9mfMQZIXVH:0) -> float_type, {64, 1}, {9, 1}: 0.00069084ms, 0.0059214% main:@2408 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](EhUjmCTaGW:0) -> float_type, {64, 1}, {9, 1}: 0.00079574ms, 0.00682053% main:@2409 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](AvlwbHMnuE:0) -> float_type, {64, 1}, {17, 1}: 0.00065686ms, 0.00563015% main:@2410 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](AVLcBpWeuo:0) -> float_type, {64, 1}, {21, 1}: 0.00070188ms, 0.00601603% main:@2411 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](Cw59eOo7M1:0) -> float_type, {64, 1}, {9, 1}: 0.00072444ms, 0.0062094% main:@2412 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](nyLxrsTmOB:0) -> float_type, {64, 1}, {17, 1}: 0.0007796ms, 0.00668219% main:@2413 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](m2YscaoDjP:0) -> float_type, {64, 1}, {17, 1}: 0.00070744ms, 0.00606369% main:@2414 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](BYyjkPCX0m:0) -> float_type, {64, 1}, {13, 1}: 0.00068002ms, 0.00582866% main:@2415 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](oQ5OcjZJe8:0) -> float_type, {64, 1}, {17, 1}: 0.00066892ms, 0.00573352% main:@2416 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](CHAYRkBvIo:0) -> float_type, {64, 1}, {41, 1}: 0.00080288ms, 0.00688173% main:@2417 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](isCn4mEOSt:0) -> float_type, {64, 1}, {49, 1}: 0.00069128ms, 0.00592517% main:@2418 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](r1WGP7Xe9H:0) -> float_type, {64, 1}, {13, 1}: 0.00067304ms, 0.00576883% main:@2419 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](qfwYOHde3u:0) -> float_type, {64, 1}, {17, 1}: 0.00068268ms, 0.00585146% main:@2420 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](tIGjbcEuDd:0) -> float_type, {64, 1}, {9, 1}: 0.00084996ms, 0.00728527% main:@2421 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](vZDAiUHErW:0) -> float_type, {64, 1}, {17, 1}: 0.00069302ms, 0.00594009% main:@2422 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](Dy3rwVmn1A:0) -> float_type, {64, 1}, {41, 1}: 0.00068636ms, 0.005883% main:@2423 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](E1ZkVIXqS5:0) -> float_type, {64, 1}, {9, 1}: 0.00065302ms, 0.00559723% main:@2424 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](qzd23aNGIT:0) -> float_type, {64, 1}, {17, 1}: 0.0008162ms, 0.0069959% main:@2425 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](EUciqFuptg:0) -> float_type, {64, 1}, {17, 1}: 0.00066042ms, 0.00566066% main:@2426 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](EKpP4xMXu8:0) -> float_type, {64, 1}, {17, 1}: 0.0006624ms, 0.00567763% main:@2427 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](EZBIAqn6LS:0) -> float_type, {64, 1}, {13, 1}: 0.0006746ms, 0.0057822% main:@2428 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](EsROupwkSG:0) -> float_type, {64, 1}, {13, 1}: 0.00078264ms, 0.00670825% main:@2429 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](nP9Yv1iZ8f:0) -> float_type, {64, 1}, {9, 1}: 0.00070396ms, 0.00603386% main:@2430 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](pQsi1GNxvj:0) -> float_type, {64, 1}, {5, 1}: 0.00067042ms, 0.00574638% main:@2431 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](F6DhiS9CN5:0) -> float_type, {64, 1}, {9, 1}: 0.0006558ms, 0.00562106% main:@2432 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](FWhBxj4Tun:0) -> float_type, {64, 1}, {9, 1}: 0.00078672ms, 0.00674322% main:@2433 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](eQquakfWBY:0) -> float_type, {64, 1}, {17, 1}: 0.00069058ms, 0.00591917% main:@2434 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](FcmWLZhDgE:0) -> float_type, {64, 1}, {57, 1}: 0.00072768ms, 0.00623717% main:@2435 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](JCVMzpWyri:0) -> float_type, {64, 1}, {17, 1}: 0.00066458ms, 0.00569632% main:@2436 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](uFSOc5Chp2:0) -> float_type, {64, 1}, {481, 1}: 0.00077598ms, 0.00665116% main:@2437 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](HRGzlnEkXQ:0) -> float_type, {64, 1}, {41, 1}: 0.00067648ms, 0.00579832% main:@2438 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](HZ0l3MmcPY:0) -> float_type, {64, 1}, {17, 1}: 0.00068114ms, 0.00583826% main:@2439 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](YbIVz8l1un:0) -> float_type, {64, 1}, {45, 1}: 0.00067684ms, 0.0058014% main:@2440 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](OfNJgEDtw2:0) -> float_type, {64, 1}, {13, 1}: 0.00075694ms, 0.00648797% main:@2441 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](HmATr3gKsW:0) -> float_type, {64, 1}, {9, 1}: 0.00066726ms, 0.00571929% main:@2442 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](Hnjq487SIY:0) -> float_type, {64, 1}, {13, 1}: 0.00066446ms, 0.00569529% main:@2443 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](omr2g4XaH5:0) -> float_type, {64, 1}, {17, 1}: 0.00066122ms, 0.00566752% main:@2444 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](NJBt1ubXFW:0) -> float_type, {64, 1}, {5, 1}: 0.00078684ms, 0.00674425% main:@2445 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](mgnGNH4Mz3:0) -> float_type, {64, 1}, {49, 1}: 0.000697ms, 0.0059742% main:@2446 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](IMdFk7D16X:0) -> float_type, {64, 1}, {9, 1}: 0.00070074ms, 0.00600626% main:@2447 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](jznORDJr0N:0) -> float_type, {64, 1}, {45, 1}: 0.0006399ms, 0.00548478% main:@2448 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](mNgQ4ohWqU:0) -> float_type, {64, 1}, {9, 1}: 0.00077936ms, 0.00668013% main:@2449 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](IxpbBg7JTO:0) -> float_type, {64, 1}, {5, 1}: 0.0006851ms, 0.0058722% main:@2450 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](J3H2EjAXt8:0) -> float_type, {64, 1}, {5, 1}: 0.00066006ms, 0.00565758% main:@2451 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](sZdFJI9nwg:0) -> float_type, {64, 1}, {41, 1}: 0.00065542ms, 0.00561781% main:@2452 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](JSWhLK4Agx:0) -> float_type, {64, 1}, {25, 1}: 0.0008039ms, 0.00689047% main:@2453 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](TkHQvxKeD6:0) -> float_type, {64, 1}, {9, 1}: 0.00065002ms, 0.00557152% main:@2454 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](JSPsGophtx:0) -> float_type, {64, 1}, {25, 1}: 0.0007325ms, 0.00627848% main:@2455 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](JaHTUxB6zL:0) -> float_type, {64, 1}, {17, 1}: 0.0006577ms, 0.00563735% main:@2456 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](U6GqhP8k7I:0) -> float_type, {64, 1}, {9, 1}: 0.00082938ms, 0.00710887% main:@2457 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](KCNfS4GlpP:0) -> float_type, {64, 1}, {9, 1}: 0.00068516ms, 0.00587272% main:@2458 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](eQI8ikEMgP:0) -> float_type, {64, 1}, {41, 1}: 0.0006546ms, 0.00561078% main:@2459 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](LKAvFme9ZS:0) -> float_type, {64, 1}, {49, 1}: 0.00069896ms, 0.005991% main:@2460 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](pY0t1rI5WU:0) -> float_type, {64, 1}, {41, 1}: 0.00079188ms, 0.00678745% main:@2461 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](LMobYAd5ru:0) -> float_type, {64, 1}, {17, 1}: 0.0006743ms, 0.00577963% main:@2462 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](iCbkMrmXBh:0) -> float_type, {64, 1}, {9, 1}: 0.00067948ms, 0.00582403% main:@2463 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](NXOSdQVPs0:0) -> float_type, {64, 1}, {25, 1}: 0.00067134ms, 0.00575426% main:@2464 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](ecPTQUALkq:0) -> float_type, {64, 1}, {5, 1}: 0.00081368ms, 0.0069743% main:@2465 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](O5aJeUQh2G:0) -> float_type, {64, 1}, {17, 1}: 0.00067678ms, 0.00580089% main:@2466 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](gdck97AK4w:0) -> float_type, {64, 1}, {17, 1}: 0.00066032ms, 0.00565981% main:@2467 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](cHzIDJm6us:0) -> float_type, {64, 1}, {33, 1}: 0.00066654ms, 0.00571312% main:@2468 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](Ouh1XpPZaj:0) -> float_type, {64, 1}, {9, 1}: 0.00080378ms, 0.00688945% main:@2469 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](lyiDxfzZg3:0) -> float_type, {64, 1}, {701, 1}: 0.00064352ms, 0.00551581% main:@2470 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](Pu1O5qFxY6:0) -> float_type, {64, 1}, {17, 1}: 0.00065242ms, 0.00559209% main:@2471 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](ry2xad1vc9:0) -> float_type, {64, 1}, {25, 1}: 0.00067158ms, 0.00575632% main:@2472 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](zX6xKOWG9S:0) -> float_type, {64, 1}, {5, 1}: 0.00080704ms, 0.00691739% main:@2473 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](qd1komWTn4:0) -> float_type, {64, 1}, {25, 1}: 0.00070844ms, 0.00607226% main:@2474 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](Qp3K1PHvte:0) -> float_type, {64, 1}, {69, 1}: 0.00061764ms, 0.00529398% main:@2475 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](dz1saATctE:0) -> float_type, {64, 1}, {9, 1}: 0.00066116ms, 0.00566701% main:@2476 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](RpITWntAji:0) -> float_type, {64, 1}, {17, 1}: 0.0008056ms, 0.00690504% main:@2477 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](nltbg9ZWQX:0) -> float_type, {64, 1}, {17, 1}: 0.00065658ms, 0.00562775% main:@2478 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](uhcrWfUe6M:0) -> float_type, {64, 1}, {9, 1}: 0.00064288ms, 0.00551032% main:@2479 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](NLt5JnRMjG:0) -> float_type, {64, 1}, {5, 1}: 0.00065652ms, 0.00562723% main:@2480 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](TSqGwrHeo4:0) -> float_type, {64, 1}, {5, 1}: 0.00079756ms, 0.00683613% main:@2481 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](Tej5M3O6As:0) -> float_type, {64, 1}, {17, 1}: 0.0007098ms, 0.00608391% main:@2482 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](khJcIxW3zv:0) -> float_type, {64, 1}, {9, 1}: 0.00066444ms, 0.00569512% main:@2483 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](Tg4u0GfnFb:0) -> float_type, {64, 1}, {25, 1}: 0.00070888ms, 0.00607603% main:@2484 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](TxEdCcekGS:0) -> float_type, {64, 1}, {49, 1}: 0.00083126ms, 0.00712498% main:@2485 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](XPcHGKbop2:0) -> float_type, {64, 1}, {13, 1}: 0.00068792ms, 0.00589637% main:@2486 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](Uc42tiuavR:0) -> float_type, {64, 1}, {9, 1}: 0.00066176ms, 0.00567215% main:@2487 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](pIb4NdDSn5:0) -> float_type, {64, 1}, {9, 1}: 0.0006756ms, 0.00579077% main:@2488 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](IRem4MiFDP:0) -> float_type, {64, 1}, {5, 1}: 0.0008013ms, 0.00686819% main:@2489 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](bTeX85HPmJ:0) -> float_type, {64, 1}, {17, 1}: 0.00066862ms, 0.00573095% main:@2490 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](duyRoIEkH8:0) -> float_type, {64, 1}, {45, 1}: 0.00068686ms, 0.00588729% main:@2491 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](d8RoEzG52p:0) -> float_type, {64, 1}, {17, 1}: 0.00067076ms, 0.00574929% main:@2492 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](X29iyIeu8L:0) -> float_type, {64, 1}, {45, 1}: 0.0007791ms, 0.00667791% main:@2493 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](zIoqYe9cRb:0) -> float_type, {64, 1}, {45, 1}: 0.00067154ms, 0.00575598% main:@2494 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](ZRYNFfjSez:0) -> float_type, {64, 1}, {33, 1}: 0.00068178ms, 0.00584375% main:@2495 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](ZcWERoz0h6:0) -> float_type, {64, 1}, {49, 1}: 0.00067174ms, 0.00575769% main:@2496 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](nv2tzWRf1I:0) -> float_type, {64, 1}, {301, 1}: 0.00080834ms, 0.00692853% main:@2497 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](ahAlEtBvzo:0) -> float_type, {64, 1}, {41, 1}: 0.00068524ms, 0.0058734% main:@2498 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](WDLoyKsiA7:0) -> float_type, {64, 1}, {33, 1}: 0.00063204ms, 0.00541741% main:@2499 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](aqMQBgb3hf:0) -> float_type, {64, 1}, {17, 1}: 0.00067472ms, 0.00578323% main:@2500 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](yVxKAL4U5G:0) -> float_type, {64, 1}, {9, 1}: 0.00081074ms, 0.0069491% main:@2501 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](bJsIAzhYc3:0) -> float_type, {64, 1}, {37, 1}: 0.00066738ms, 0.00572032% main:@2502 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](cN2JSTnFOu:0) -> float_type, {64, 1}, {9, 1}: 0.00070668ms, 0.00605717% main:@2503 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](vAbKepTjQg:0) -> float_type, {64, 1}, {5, 1}: 0.00066598ms, 0.00570832% main:@2504 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](fZhaL2tCm6:0) -> float_type, {64, 1}, {17, 1}: 0.0007919ms, 0.00678762% main:@2505 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](o74vNAnXxy:0) -> float_type, {64, 1}, {17, 1}: 0.00071144ms, 0.00609797% main:@2506 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](l5WSX3VvMK:0) -> float_type, {64, 1}, {25, 1}: 0.00065796ms, 0.00563958% main:@2507 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](pH5srbufJ1:0) -> float_type, {64, 1}, {17, 1}: 0.0006904ms, 0.00591763% main:@2508 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](lDhFNYzbxR:0) -> float_type, {64, 1}, {17, 1}: 0.00081066ms, 0.00694842% main:@2509 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](magwfJSnUu:0) -> float_type, {64, 1}, {9, 1}: 0.0006909ms, 0.00592192% main:@2510 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](w04Keh2FJu:0) -> float_type, {64, 1}, {13, 1}: 0.00065666ms, 0.00562843% main:@2511 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](fMiQsYNnet:0) -> float_type, {64, 1}, {5, 1}: 0.0006571ms, 0.00563221% main:@2512 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](gqk5byC8aJ:0) -> float_type, {64, 1}, {5, 1}: 0.00080302ms, 0.00688293% main:@2513 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](lQmygxs5aD:0) -> float_type, {64, 1}, {13, 1}: 0.0006851ms, 0.0058722% main:@2514 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](sjKQJGby31:0) -> float_type, {64, 1}, {9, 1}: 0.00067906ms, 0.00582043% main:@2515 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](zZHaVOckpw:0) -> float_type, {64, 1}, {13, 1}: 0.00066352ms, 0.00568723% main:@2516 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](l4bReWtsMB:0) -> float_type, {64, 1}, {25, 1}: 0.0007908ms, 0.00677819% main:@2517 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](tWcngAq7hV:0) -> float_type, {64, 1}, {13, 1}: 0.0007247ms, 0.00621163% main:@2518 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](xVErSbJ0hg:0) -> float_type, {64, 1}, {9, 1}: 0.00066282ms, 0.00568123% main:@2519 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](mTMtDEo3r8:0) -> float_type, {64, 1}, {9, 1}: 0.00070382ms, 0.00603266% main:@2520 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](VUqfdN5Fzv:0) -> float_type, {64, 1}, {45, 1}: 0.00071672ms, 0.00614323% main:@2521 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](Vbv2Om3ptd:0) -> float_type, {64, 1}, {17, 1}: 0.0007401ms, 0.00634362% ebEFYOy65u:0 = @param:ebEFYOy65u:0 -> float_type, {64, 1}, {1, 1}: 0.00060862ms, 0.00521667% iUEBSr6gOG:0 = @param:iUEBSr6gOG:0 -> float_type, {64, 1}, {1, 1}: 0.0005568ms, 0.0047725% qxzgXpiLdl:0 = @param:qxzgXpiLdl:0 -> float_type, {64, 1}, {1, 1}: 0.00060936ms, 0.00522301% 9NcxXsmyqf:0 = @param:9NcxXsmyqf:0 -> float_type, {64, 1}, {1, 1}: 0.00051812ms, 0.00444097% WqeN5KGIv9:0 = @param:WqeN5KGIv9:0 -> float_type, {64, 1}, {1, 1}: 0.00051344ms, 0.00440085% 7XiFOysrpI:0 = @param:7XiFOysrpI:0 -> float_type, {64, 1}, {1, 1}: 0.0004907ms, 0.00420594% vtIF8HqRG6:0 = @param:vtIF8HqRG6:0 -> float_type, {64, 1}, {1, 1}: 0.00050478ms, 0.00432662% NG35MnhTIF:0 = @param:NG35MnhTIF:0 -> float_type, {64, 1}, {1, 1}: 0.00067154ms, 0.00575598% lqOmNvz5fj:0 = @param:lqOmNvz5fj:0 -> float_type, {64, 1}, {1, 1}: 0.00050758ms, 0.00435062% XPpbueLMOZ:0 = @param:XPpbueLMOZ:0 -> float_type, {64, 1}, {1, 1}: 0.0005792ms, 0.0049645% bJxY1IhSVn:0 = @param:bJxY1IhSVn:0 -> float_type, {64, 1}, {1, 1}: 0.00048864ms, 0.00418828% iE352qtAOx:0 = @param:iE352qtAOx:0 -> float_type, {64, 1}, {1, 1}: 0.00052406ms, 0.00449188% c2MuYjWE3S:0 = @param:c2MuYjWE3S:0 -> float_type, {64, 1}, {1, 1}: 0.0005325ms, 0.00456422% 7tFB8mLWUr:0 = @param:7tFB8mLWUr:0 -> float_type, {64, 1}, {1, 1}: 0.00050658ms, 0.00434205% FTIMrcfako:0 = @param:FTIMrcfako:0 -> float_type, {64, 1}, {1, 1}: 0.00067184ms, 0.00575855% xGCyW9icVP:0 = @param:xGCyW9icVP:0 -> float_type, {64, 1}, {1, 1}: 0.0004974ms, 0.00426337% lhbKMrO1my:0 = @param:lhbKMrO1my:0 -> float_type, {64, 1}, {1, 1}: 0.00051464ms, 0.00441114% StHFqj5edV:0 = @param:StHFqj5edV:0 -> float_type, {64, 1}, {1, 1}: 0.00049228ms, 0.00421948% ZyQ5FrpI6t:0 = @param:ZyQ5FrpI6t:0 -> float_type, {64, 1}, {1, 1}: 0.00054496ms, 0.00467102% mxUaAp0CV8:0 = @param:mxUaAp0CV8:0 -> float_type, {64, 1}, {1, 1}: 0.0006287ms, 0.00538878% vF46yONhri:0 = @param:vF46yONhri:0 -> float_type, {64, 1}, {1, 1}: 0.00056658ms, 0.00485633% URVDgNGoIx:0 = @param:URVDgNGoIx:0 -> float_type, {64, 1}, {1, 1}: 0.00048642ms, 0.00416926% RPGvVrZMak:0 = @param:RPGvVrZMak:0 -> float_type, {64, 1}, {1, 1}: 0.00048848ms, 0.00418691% main:@2545 = load[offset=326400,end=369152](main:@1) -> float_type, {64, 167}, {167, 1}: 0.00048562ms, 0.0041624% main:@2546 = gpu::code_object[code_object=25808,symbol_name=concat_kernel,global=64,local=256,](main:@2469,main:@2496,main:@2391,main:@2460,main:@2385,main:@2459,main:@2484,main:@2399,main:@2513,main:@2383,main:@2421,main:@2425,main:@2416,main:@2428,iE352qtAOx:0,main:@2392,RPGvVrZMak:0,URVDgNGoIx:0,vF46yONhri:0,main:@2481,main:@2480,bJxY1IhSVn:0,main:@2502,XPpbueLMOZ:0,lqOmNvz5fj:0,main:@2462,main:@2412,main:@2499,main:@2488,NG35MnhTIF:0,main:@2393,vtIF8HqRG6:0,7XiFOysrpI:0,WqeN5KGIv9:0,mxUaAp0CV8:0,ZyQ5FrpI6t:0,main:@2511,main:@2491,main:@2504,main:@2521,main:@2437,main:@2414,main:@2424,main:@2443,main:@2482,9NcxXsmyqf:0,main:@2442,main:@2398,main:@2418,main:@2485,qxzgXpiLdl:0,StHFqj5edV:0,lhbKMrO1my:0,xGCyW9icVP:0,iUEBSr6gOG:0,ebEFYOy65u:0,main:@2436,main:@2445,main:@2400,main:@2474,main:@2417,main:@2422,main:@2440,main:@2493,main:@2476,main:@2495,main:@2483,main:@2395,main:@2497,main:@2478,main:@2475,main:@2501,main:@2471,main:@2452,main:@2382,main:@2473,main:@2410,main:@2487,main:@2510,main:@2505,main:@2438,main:@2517,main:@2439,main:@2386,main:@2390,main:@2409,main:@2492,main:@2380,main:@2463,main:@2413,main:@2507,main:@2515,main:@2509,main:@2518,main:@2516,main:@2441,main:@2415,main:@2449,main:@2465,main:@2468,main:@2506,main:@2500,main:@2498,FTIMrcfako:0,main:@2466,main:@2420,7tFB8mLWUr:0,main:@2403,main:@2470,main:@2519,c2MuYjWE3S:0,main:@2448,main:@2434,main:@2520,main:@2447,main:@2384,main:@2387,main:@2458,main:@2490,main:@2472,main:@2486,main:@2381,main:@2411,main:@2408,main:@2396,main:@2432,main:@2450,main:@2444,main:@2512,main:@2479,main:@2388,main:@2430,main:@2464,main:@2456,main:@2503,main:@2394,main:@2454,main:@2406,main:@2427,main:@2433,main:@2489,main:@2426,main:@2405,main:@2461,main:@2397,main:@2455,main:@2435,main:@2508,main:@2419,main:@2379,main:@2446,main:@2457,main:@2423,main:@2494,main:@2477,main:@2451,main:@2407,main:@2389,main:@2514,main:@2467,main:@2429,main:@2402,main:@2378,main:@2453,main:@2404,main:@2401,main:@2431,main:@2545) -> float_type, {64, 167}, {167, 1}: 0.140265ms, 1.20226% main:@2547 = load[offset=434688,end=477440](main:@1) -> float_type, {64, 167}, {167, 1}: 0.00074752ms, 0.00640722% main:@2548 = gpu::code_object[code_object=13520,symbol_name=rsqrt_add_kernel,global=5344,local=256,](main:@2377,main:@2546,main:@2547) -> float_type, {64, 167}, {167, 1}: 0.0139206ms, 0.119318% main:@2549 = hip::hip_copy_literal[id=main:@literal:45] -> float_type, {167, 256}, {256, 1}: 0.00080402ms, 0.0068915% main:@2550 = load[offset=369152,end=434688](main:@1) -> float_type, {64, 256}, {256, 1}: 0.00049118ms, 0.00421005% main:@2551 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@2548,main:@2549,main:@2550) -> float_type, {64, 256}, {256, 1}: 0.0256179ms, 0.219579% main:@2552 = hip::hip_copy_literal[id=main:@literal:38] -> float_type, {128, 256}, {256, 1}: 0.00083632ms, 0.00716836% main:@2553 = slice[axes={1},starts={640},ends={768}](main:@717) -> float_type, {64, 128}, {896, 1}: 0.00066856ms, 0.00573043% main:@2554 = load[offset=434688,end=500224](main:@1) -> float_type, {64, 256}, {256, 1}: 0.00050134ms, 0.00429714% main:@2555 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@2553,main:@2552,main:@2554) -> float_type, {64, 256}, {256, 1}: 0.0204831ms, 0.175567% main:@2556 = hip::hip_copy_literal[id=main:@literal:12] -> float_type, {64, 256}, {256, 1}: 0.00084828ms, 0.00727087% main:@2557 = load[offset=16384,end=81920](main:@1) -> float_type, {64, 256}, {256, 1}: 0.00048416ms, 0.00414988% main:@2558 = gpu::code_object[code_object=13552,symbol_name=add_leaky_relu_tanh_pow_kernel,global=8192,local=256,](main:@2551,main:@2556,main:@2555,main:@2557) -> float_type, {64, 256}, {256, 1}: 0.0180344ms, 0.154579% main:@2559 = hip::hip_copy_literal[id=main:@literal:76] -> float_type, {256, 64}, {64, 1}: 0.00081422ms, 0.00697893% main:@2560 = load[offset=0,end=16384](main:@1) -> float_type, {64, 64}, {64, 1}: 0.00047448ms, 0.00406691% 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, {64, 64}, {64, 1}: 0.0244016ms, 0.209154% main:@2562 = hip::hip_copy_literal[id=main:@literal:111] -> float_type, {53, 53}, {53, 1}: 0.00088076ms, 0.00754926% main:@2563 = slice[axes={1},starts={53},ends={106}](main:@2373) -> float_type, {64, 53}, {212, 1}: 0.00060202ms, 0.0051601% main:@2564 = load[offset=43520,end=57088](main:@1) -> float_type, {64, 53}, {53, 1}: 0.00047082ms, 0.00403554% 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, {64, 53}, {53, 1}: 0.0212255ms, 0.18193% main:@2566 = hip::hip_copy_literal[id=main:@literal:82] -> float_type, {1, 53}, {53, 1}: 0.000795ms, 0.00681419% main:@2567 = slice[axes={1},starts={159},ends={212}](main:@2373) -> float_type, {64, 53}, {212, 1}: 0.00063662ms, 0.00545667% main:@2568 = load[offset=16384,end=29952](main:@1) -> float_type, {64, 53}, {53, 1}: 0.00050842ms, 0.00435782% main:@2569 = gpu::code_object[code_object=13536,symbol_name=div_pow_log_add_kernel,global=3392,local=256,](main:@2567,main:@2568) -> float_type, {64, 53}, {53, 1}: 0.0156048ms, 0.133753% main:@2570 = multibroadcast[out_lens={64, 53}](main:@2566) -> float_type, {64, 53}, {0, 1}: 0.0007703ms, 0.00660248% main:@2571 = load[offset=29952,end=43520](main:@1) -> float_type, {64, 53}, {53, 1}: 0.00050302ms, 0.00431154% main:@2572 = gpu::code_object[code_object=13552,symbol_name=pow_sub_pow_pow_log_add_kernel,global=3392,local=256,](main:@2565,main:@2570,main:@2563,main:@2571) -> float_type, {64, 53}, {53, 1}: 0.0182614ms, 0.156524% main:@2573 = slice[axes={1},starts={106},ends={159}](main:@2373) -> float_type, {64, 53}, {212, 1}: 0.0007501ms, 0.00642934% main:@2574 = slice[axes={1},starts={0},ends={53}](main:@2373) -> float_type, {64, 53}, {212, 1}: 0.00055522ms, 0.00475896% main:@2575 = load[offset=43520,end=57088](main:@1) -> float_type, {64, 53}, {53, 1}: 0.00045604ms, 0.00390886% main:@2576 = gpu::code_object[code_object=13536,symbol_name=add_add_neg_div_kernel,global=3392,local=256,](main:@2573,main:@2574,main:@2575) -> float_type, {64, 53}, {53, 1}: 0.014175ms, 0.121499% main:@2577 = hip::hip_copy_literal[id=main:@literal:83] -> float_type, {374, 212}, {212, 1}: 0.00077334ms, 0.00662853% main:@2578 = load[offset=111616,end=111872](main:@1) -> float_type, {64, 1}, {1, 1}: 0.00047008ms, 0.0040292% main:@2579 = gpu::code_object[code_object=13544,symbol_name=reduce_mean_add_neg_kernel,global=4096,local=64,](main:@2569,main:@2578) -> float_type, {64, 1}, {1, 1}: 0.0153025ms, 0.131162% main:@2580 = load[offset=57088,end=111360](main:@1) -> float_type, {64, 212}, {212, 1}: 0.00060858ms, 0.00521633% main:@2581 = gpu::code_object[code_object=13528,symbol_name=leaky_relu_kernel,global=6784,local=256,](main:@2373,main:@2580) -> float_type, {64, 212}, {212, 1}: 0.0138746ms, 0.118924% main:@2582 = load[offset=112384,end=112640](main:@1) -> float_type, {64, 1}, {1, 1}: 0.00058536ms, 0.0050173% main:@2583 = gpu::code_object[code_object=13544,symbol_name=reduce_prod_add_log_kernel,global=4096,local=64,](main:@2576,main:@2582) -> float_type, {64, 1}, {1, 1}: 0.0152335ms, 0.130571% main:@2584 = load[offset=111360,end=111616](main:@1) -> float_type, {64, 1}, {1, 1}: 0.00058302ms, 0.00499724% main:@2585 = gpu::code_object[code_object=13552,symbol_name=reduce_prod_pow_rsqrt_div_kernel,global=4096,local=64,](main:@2572,main:@2584) -> float_type, {64, 1}, {1, 1}: 0.0158067ms, 0.135484% main:@2586 = load[offset=111872,end=112384](main:@1) -> float_type, {64, 2}, {2, 1}: 0.00062492ms, 0.00535638% main:@2587 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=64,local=256,](main:@2583,main:@2579,main:@2586) -> float_type, {64, 2}, {2, 1}: 0.0132627ms, 0.113678% main:@2588 = slice[axes={1},starts={106},ends={159}](main:@2581) -> float_type, {64, 53}, {212, 1}: 0.00073104ms, 0.00626597% main:@2589 = slice[axes={1},starts={159},ends={212}](main:@2581) -> float_type, {64, 53}, {212, 1}: 0.00054644ms, 0.0046837% main:@2590 = slice[axes={1},starts={53},ends={106}](main:@2581) -> float_type, {64, 53}, {212, 1}: 0.00068498ms, 0.00587117% main:@2591 = slice[axes={1},starts={0},ends={53}](main:@2581) -> float_type, {64, 53}, {212, 1}: 0.00053732ms, 0.00460553% main:@2592 = slice[axes={1},starts={4070},ends={4444}](main:@1510) -> float_type, {64, 374}, {4508, 1}: 0.00054726ms, 0.00469073% main:@2593 = load[offset=112384,end=208128](main:@1) -> float_type, {64, 374}, {374, 1}: 0.00045842ms, 0.00392926% main:@2594 = gpu::code_object[code_object=21816,symbol_name=noop_noop_noop_noop_noop_noop_noop_sub_noop_concat_neg_pow_kernel,global=2659,local=256,](main:@2588,main:@2591,main:@2589,main:@2590,main:@2576,main:@2569,main:@2572,main:@2274,main:@2587,main:@2585,main:@2592,main:@2593) -> float_type, {64, 374}, {374, 1}: 0.03664ms, 0.314052% main:@2595 = load[offset=16384,end=70656](main:@1) -> float_type, {64, 212}, {212, 1}: 0.00061652ms, 0.00528438% 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, {64, 212}, {212, 1}: 0.0281279ms, 0.241093% main:@2597 = hip::hip_copy_literal[id=main:@literal:17] -> float_type, {64, 212}, {212, 1}: 0.00076386ms, 0.00654728% main:@2598 = load[offset=70656,end=70912](main:@1) -> float_type, {64, 1}, {1, 1}: 0.00047794ms, 0.00409657% main:@2599 = gpu::code_object[code_object=13528,symbol_name=reduce_prod_kernel,global=12288,local=192,](main:@2546,main:@2598) -> float_type, {64, 1}, {1, 1}: 0.0155148ms, 0.132982% main:@2600 = hip::hip_copy_literal[id=main:@literal:28] -> float_type, {64, 64}, {64, 1}: 0.00073156ms, 0.00627043% main:@2601 = hip::hip_copy_literal[id=main:@literal:11] -> float_type, {277, 160}, {160, 1}: 0.00061748ms, 0.00529261% main:@2602 = load[offset=82176,end=153088](main:@1) -> float_type, {64, 277}, {277, 1}: 0.00045754ms, 0.00392172% main:@2603 = slice[axes={1},starts={4444},ends={4508}](main:@1510) -> float_type, {64, 64}, {4508, 1}: 0.00063714ms, 0.00546112% main:@2604 = gpu::code_object[code_object=13600,symbol_name=rsqrt_add_sub_mul_add_leaky_relu_noop_concat_noop_kernel,global=5909,local=256,](main:@2603,main:@2561,main:@2600,main:@2596,main:@2597,main:@2599,main:@2602) -> float_type, {64, 277}, {277, 1}: 0.0181651ms, 0.155698% main:@2605 = load[offset=41216,end=82176](main:@1) -> float_type, {64, 160}, {160, 1}: 0.0006366ms, 0.00545649% main:@2606 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@2604,main:@2601,main:@2605) -> float_type, {64, 160}, {160, 1}: 0.0254832ms, 0.218425% main:@2607 = hip::hip_copy_literal[id=main:@literal:13] -> float_type, {160}, {1}: 0.00079574ms, 0.00682053% main:@2608 = load[offset=256,end=41216](main:@1) -> float_type, {64, 160}, {160, 1}: 0.00049142ms, 0.00421211% main:@2609 = multibroadcast[out_lens={64, 160}](main:@2607) -> float_type, {64, 160}, {0, 1}: 0.0006019ms, 0.00515907% main:@2610 = gpu::code_object[code_object=13536,symbol_name=add_leaky_relu_kernel,global=5120,local=256,](main:@2606,main:@2609,main:@2608) -> float_type, {64, 160}, {160, 1}: 0.0140076ms, 0.120063% main:@2611 = hip::hip_copy_literal[id=main:@literal:36] -> float_type, {64, 329}, {329, 1}: 0.00078678ms, 0.00674373% main:@2612 = slice[axes={1},starts={80},ends={144}](main:@2610) -> float_type, {64, 64}, {160, 1}: 0.0005792ms, 0.0049645% main:@2613 = load[offset=41216,end=125440](main:@1) -> float_type, {64, 329}, {329, 1}: 0.00048358ms, 0.00414491% main:@2614 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@2612,main:@2611,main:@2613) -> float_type, {64, 329}, {329, 1}: 0.0199466ms, 0.170968% main:@2615 = hip::hip_copy_literal[id=main:@literal:88] -> float_type, {1, 329}, {329, 1}: 0.00080686ms, 0.00691584% main:@2616 = hip::hip_copy_literal[id=main:@literal:65] -> float_type, {16, 1}, {1, 1}: 0.00064898ms, 0.00556261% main:@2617 = slice[axes={1},starts={144},ends={160}](main:@2610) -> float_type, {64, 16}, {160, 1}: 0.0006279ms, 0.00538192% main:@2618 = load[offset=0,end=256](main:@1) -> float_type, {64, 1}, {1, 1}: 0.000492ms, 0.00421708% 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, {64, 1}, {1, 1}: 0.0218604ms, 0.187372% main:@2620 = load[offset=125440,end=209664](main:@1) -> float_type, {64, 329}, {329, 1}: 0.00064634ms, 0.00553998% main:@2621 = gpu::code_object[code_object=13520,symbol_name=softmax_kernel,global=16384,local=256,](main:@2614,main:@2620) -> float_type, {64, 329}, {329, 1}: 0.0164424ms, 0.140933% main:@2622 = multibroadcast[out_lens={64, 329}](main:@2615) -> float_type, {64, 329}, {0, 1}: 0.0007263ms, 0.00622534% main:@2623 = load[offset=41216,end=41472](main:@1) -> float_type, {64, 1}, {1, 1}: 0.00049614ms, 0.00425257% main:@2624 = gpu::code_object[code_object=13536,symbol_name=add_reduce_prod_kernel,global=16384,local=256,](main:@2622,main:@2621,main:@2623) -> float_type, {64, 1}, {1, 1}: 0.0161675ms, 0.138577% main:@2625 = squeeze[axes={1}](main:@2624) -> float_type, {64}, {1}: 0.0007247ms, 0.00621163% main:#output_1 = @param:main:#output_1 -> float_type, {64}, {1}: 0.00077376ms, 0.00663213% main:@2627 = gpu::code_object[code_object=13512,symbol_name=mul_kernel,global=32,local=256,](main:@2625,main:#output_1) -> float_type, {64}, {1}: 0.0131574ms, 0.112776% main:@2628 = hip::hip_copy_literal[id=main:@literal:110] -> float_type, {64, 500}, {500, 1}: 0.00079026ms, 0.00677356% main:@2629 = slice[axes={1},starts={16},ends={80}](main:@2610) -> float_type, {64, 64}, {160, 1}: 0.00058698ms, 0.00503119% main:@2630 = load[offset=41984,end=169984](main:@1) -> float_type, {64, 500}, {500, 1}: 0.00045476ms, 0.00389789% 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, {64, 500}, {500, 1}: 0.0196411ms, 0.16835% main:@2632 = hip::hip_copy_literal[id=main:@literal:147] -> float_type, {16, 3}, {3, 1}: 0.0007143ms, 0.00612248% main:@2633 = slice[axes={1},starts={0},ends={16}](main:@2610) -> float_type, {64, 16}, {160, 1}: 0.00060308ms, 0.00516918% main:@2634 = load[offset=41216,end=41984](main:@1) -> float_type, {64, 3}, {3, 1}: 0.0004725ms, 0.00404994% 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, {64, 3}, {3, 1}: 0.0196894ms, 0.168764% main:@2636 = hip::hip_copy_literal[id=main:@literal:72] -> float_type, {1, 500}, {500, 1}: 0.0007922ms, 0.00679019% main:#output_2 = @param:main:#output_2 -> float_type, {64}, {1}: 0.0006147ms, 0.00526878% main:@2638 = gpu::code_object[code_object=13512,symbol_name=add_kernel,global=32,local=256,](main:@2627,main:#output_2) -> float_type, {64}, {1}: 0.0132389ms, 0.113474% main:#output_3 = @param:main:#output_3 -> float_type, {64}, {1}: 0.00074674ms, 0.00640054% main:@2640 = gpu::code_object[code_object=13520,symbol_name=mul_neg_kernel,global=32,local=256,](main:@2627,main:#output_3) -> float_type, {64}, {1}: 0.013293ms, 0.113938% main:@2641 = load[offset=169984,end=297984](main:@1) -> float_type, {64, 500}, {500, 1}: 0.00063182ms, 0.00541552% main:@2642 = gpu::code_object[code_object=13520,symbol_name=softmax_kernel,global=16384,local=256,](main:@2631,main:@2641) -> float_type, {64, 500}, {500, 1}: 0.0164324ms, 0.140847% main:@2643 = multibroadcast[out_lens={64, 500}](main:@2636) -> float_type, {64, 500}, {0, 1}: 0.00070242ms, 0.00602066% main:@2644 = load[offset=256,end=512](main:@1) -> float_type, {64, 1}, {1, 1}: 0.00047216ms, 0.00404703% main:@2645 = gpu::code_object[code_object=13536,symbol_name=add_reduce_mean_kernel,global=16384,local=256,](main:@2643,main:@2642,main:@2644) -> float_type, {64, 1}, {1, 1}: 0.0157599ms, 0.135083% main:@2646 = squeeze[axes={1}](main:@2645) -> float_type, {64}, {1}: 0.0006655ms, 0.0057042% main:#output_4 = @param:main:#output_4 -> float_type, {64}, {1}: 0.00062166ms, 0.00532844% main:@2648 = gpu::code_object[code_object=13512,symbol_name=mul_kernel,global=32,local=256,](main:@2646,main:#output_4) -> float_type, {64}, {1}: 0.0131998ms, 0.11314% main:@2649 = load[offset=256,end=1024](main:@1) -> float_type, {64, 3}, {3, 1}: 0.00059482ms, 0.00509838% main:@2650 = gpu::code_object[code_object=14512,symbol_name=softmax_kernel,global=4096,local=64,](main:@2635,main:@2649) -> float_type, {64, 3}, {3, 1}: 0.015192ms, 0.130216% main:@2651 = slice[axes={0, 1},starts={0, 0},ends={64, 1}](main:@2650) -> float_type, {64, 1}, {3, 1}: 0.00077822ms, 0.00667036% main:#output_5 = @param:main:#output_5 -> float_type, {64, 1}, {1, 1}: 0.00068554ms, 0.00587597% main:@2653 = gpu::code_object[code_object=13528,symbol_name=contiguous_kernel,global=64,local=256,](main:@2651,main:#output_5) -> float_type, {64, 1}, {1, 1}: 0.0130647ms, 0.111981% main:@2654 = squeeze[axes={1}](main:@2619) -> float_type, {64}, {1}: 0.00067508ms, 0.00578632% 86KRkhYbW4:0 = @param:86KRkhYbW4:0 -> float_type, {64}, {1}: 0.00057796ms, 0.00495387% main:#output_0 = @param:main:#output_0 -> float_type, {64}, {1}: 0.00059046ms, 0.00506101% main:@2657 = gpu::code_object[code_object=13536,symbol_name=sigmoid_mul_tanh_kernel,global=32,local=256,](86KRkhYbW4:0,main:@2654,main:#output_0) -> float_type, {64}, {1}: 0.0147998ms, 0.126853% main:@2658 = squeeze[axes={1}](main:@2653) -> float_type, {64}, {1}: 0.00072572ms, 0.00622037% main:@2659 = hip::sync_stream(main:@2657,main:@2627,main:@2638,main:@2640,main:@2648,main:@2658) -> float_type, {64}, {1}: 0.00112526ms, 0.00964495% main:@2660 = @return(main:@2659,main:@2627,main:@2638,main:@2640,main:@2648,main:@2658) Summary: gpu::gemm: 4.85795ms / 187 = 0.0259783ms, 41.639% gpu::code_object::concat_kernel: 1.93681ms / 84 = 0.0230572ms, 16.601% slice: 0.608141ms / 859 = 0.000707964ms, 5.21256% gpu::code_object::relu_kernel: 0.28348ms / 18 = 0.0157489ms, 2.4298% load: 0.260486ms / 468 = 0.000556595ms, 2.23271% gpu::code_object::softmax_kernel: 0.249946ms / 14 = 0.0178533ms, 2.14237% 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: 0.231893ms / 1 = 0.231893ms, 1.98763% @param: 0.206998ms / 383 = 0.000540465ms, 1.77424% gpu::code_object::convert_kernel: 0.202404ms / 15 = 0.0134936ms, 1.73487% gpu::code_object::contiguous_kernel: 0.142383ms / 10 = 0.0142383ms, 1.22041% hip::hip_copy_literal: 0.126636ms / 165 = 0.000767489ms, 1.08543% gpu::code_object::add_kernel: 0.115522ms / 7 = 0.0165031ms, 0.990172% unsqueeze: 0.0902891ms / 145 = 0.000622684ms, 0.773896% gpu::code_object::neg_pow_kernel: 0.0822219ms / 5 = 0.0164444ms, 0.704749% gpu::code_object::log_add_kernel: 0.0813879ms / 6 = 0.0135647ms, 0.697601% gpu::code_object::neg_div_kernel: 0.0801926ms / 6 = 0.0133654ms, 0.687355% gpu::code_object::pow_kernel: 0.0710491ms / 4 = 0.0177623ms, 0.608983% gpu::code_object::noop_rsqrt_add_concat_noop_kernel: 0.070956ms / 4 = 0.017739ms, 0.608185% gpu::code_object::div_kernel: 0.0556739ms / 3 = 0.018558ms, 0.477198% gpu::code_object::rsqrt_add_kernel: 0.0537969ms / 4 = 0.0134492ms, 0.46111% gpu::code_object::sub_kernel: 0.0520105ms / 3 = 0.0173368ms, 0.445798% gpu::code_object::rsqrt_pow_kernel: 0.0492077ms / 3 = 0.0164026ms, 0.421774% reshape_lazy: 0.0464647ms / 81 = 0.000573638ms, 0.398263% gpu::code_object::reduce_prod_kernel: 0.0458735ms / 3 = 0.0152912ms, 0.393196% gpu::code_object::tanh_sub_kernel: 0.0410359ms / 3 = 0.0136786ms, 0.351731% gpu::code_object::neg_add_kernel: 0.040122ms / 3 = 0.013374ms, 0.343898% gpu::code_object::convert_add_kernel: 0.040069ms / 3 = 0.0133563ms, 0.343443% gpu::code_object::leaky_relu_noop_noop_noop_noop_noop_noop_concat_rsqrt_pow_kernel: 0.0368153ms / 1 = 0.0368153ms, 0.315555% gpu::code_object::noop_log_add_concat_noop_kernel: 0.0367712ms / 2 = 0.0183856ms, 0.315177% gpu::code_object::noop_noop_noop_noop_noop_noop_noop_sub_noop_concat_neg_pow_kernel: 0.03664ms / 1 = 0.03664ms, 0.314052% gpu::code_object::pow_pow_kernel: 0.0361729ms / 2 = 0.0180864ms, 0.310049% gpu::code_object::noop_rsqrt_div_concat_noop_kernel: 0.0346719ms / 2 = 0.0173359ms, 0.297183% gpu::code_object::log_pow_kernel: 0.0332624ms / 2 = 0.0166312ms, 0.285102% gpu::code_object::noop_tanh_div_concat_noop_kernel: 0.0329094ms / 2 = 0.0164547ms, 0.282077% gpu::code_object::add_leaky_relu_kernel: 0.0319607ms / 2 = 0.0159804ms, 0.273945% gpu::code_object::leaky_relu_noop_noop_noop_noop_add_concat_log_pow_kernel: 0.0317247ms / 1 = 0.0317247ms, 0.271922% gpu::code_object::pow_reduce_sum_div_pow_add_kernel: 0.0310975ms / 1 = 0.0310975ms, 0.266546% gpu::code_object::leaky_relu_noop_noop_noop_noop_noop_noop_concat_log_add_pow_kernel: 0.0310333ms / 1 = 0.0310333ms, 0.265996% gpu::code_object::add_add_neg_add_kernel: 0.0287389ms / 2 = 0.0143695ms, 0.24633% gpu::code_object::tanh_div_kernel: 0.0280632ms / 2 = 0.0140316ms, 0.240538% gpu::code_object::tanh_add_kernel: 0.0275927ms / 2 = 0.0137964ms, 0.236506% gpu::code_object::mul_relu_reduce_prod_sub_mul_reduce_sum_add_add_add_add_pow_kernel: 0.0264946ms / 1 = 0.0264946ms, 0.227093% gpu::code_object::convert_mul_add_kernel: 0.0264323ms / 2 = 0.0132161ms, 0.226559% gpu::code_object::mul_kernel: 0.0263572ms / 2 = 0.0131786ms, 0.225916% gpu::code_object::rsqrt_sub_kernel: 0.0261388ms / 2 = 0.0130694ms, 0.224043% gpu::code_object::leaky_relu_noop_noop_noop_add_noop_concat_log_div_kernel: 0.0229629ms / 1 = 0.0229629ms, 0.196822% multibroadcast: 0.0219919ms / 35 = 0.000628341ms, 0.188499% gpu::code_object::pow_sub_div_pow_log_div_kernel: 0.0206176ms / 1 = 0.0206176ms, 0.17672% gpu::code_object::pow_pow_sigmoid_add_kernel: 0.0197471ms / 1 = 0.0197471ms, 0.169259% gpu::code_object::noop_tanh_pow_concat_noop_kernel: 0.0190054ms / 1 = 0.0190054ms, 0.162901% gpu::code_object::pow_mul_add_neg_sub_kernel: 0.0189894ms / 1 = 0.0189894ms, 0.162764% gpu::code_object::pow_mul_rsqrt_sub_kernel: 0.0189202ms / 1 = 0.0189202ms, 0.162171% gpu::code_object::noop_neg_div_concat_noop_kernel: 0.0185468ms / 1 = 0.0185468ms, 0.15897% gpu::code_object::noop_log_sub_concat_noop_kernel: 0.0183162ms / 1 = 0.0183162ms, 0.156994% gpu::code_object::pow_sub_pow_pow_log_add_kernel: 0.0182614ms / 1 = 0.0182614ms, 0.156524% gpu::code_object::rsqrt_add_sub_mul_add_leaky_relu_noop_concat_noop_kernel: 0.0181651ms / 1 = 0.0181651ms, 0.155698% gpu::code_object::add_leaky_relu_tanh_pow_kernel: 0.0180344ms / 1 = 0.0180344ms, 0.154579% gpu::code_object::add_pow_log_pow_kernel: 0.0175951ms / 1 = 0.0175951ms, 0.150813% squeeze: 0.0175556ms / 25 = 0.000702224ms, 0.150474% gpu::code_object::noop_log_pow_concat_noop_kernel: 0.0175037ms / 1 = 0.0175037ms, 0.15003% gpu::code_object::sub_sub_pow_log_pow_kernel: 0.0171462ms / 1 = 0.0171462ms, 0.146965% gpu::code_object::noop_rsqrt_sub_concat_noop_kernel: 0.0164511ms / 1 = 0.0164511ms, 0.141007% gpu::code_object::reduce_mean_add_neg_pow_kernel: 0.0163302ms / 1 = 0.0163302ms, 0.139971% gpu::code_object::reduce_mean_add_rsqrt_add_kernel: 0.0163126ms / 1 = 0.0163126ms, 0.13982% gpu::code_object::add_pow_kernel: 0.0163123ms / 1 = 0.0163123ms, 0.139817% gpu::code_object::reduce_prod_pow_neg_sub_kernel: 0.0162472ms / 1 = 0.0162472ms, 0.139259% gpu::code_object::add_reduce_prod_kernel: 0.0161675ms / 1 = 0.0161675ms, 0.138577% gpu::code_object::reduce_mean_pow_log_pow_kernel: 0.0160703ms / 1 = 0.0160703ms, 0.137744% gpu::code_object::neg_rsqrt_tanh_log_concat_add_kernel: 0.016009ms / 1 = 0.016009ms, 0.137218% gpu::code_object::div_mul_neg_pow_kernel: 0.015979ms / 1 = 0.015979ms, 0.136961% gpu::code_object::reduce_prod_pow_rsqrt_div_kernel: 0.0158067ms / 1 = 0.0158067ms, 0.135484% gpu::code_object::add_reduce_mean_kernel: 0.0157599ms / 1 = 0.0157599ms, 0.135083% gpu::code_object::reduce_prod_add_rsqrt_sub_kernel: 0.015756ms / 1 = 0.015756ms, 0.13505% gpu::code_object::reduce_sum_kernel: 0.0157401ms / 1 = 0.0157401ms, 0.134913% gpu::code_object::div_pow_log_add_kernel: 0.0156048ms / 1 = 0.0156048ms, 0.133753% gpu::code_object::pow_div_kernel: 0.0155916ms / 1 = 0.0155916ms, 0.13364% gpu::code_object::sub_sub_mul_rsqrt_sub_kernel: 0.0155464ms / 1 = 0.0155464ms, 0.133253% gpu::code_object::reduce_prod_add_log_div_kernel: 0.0155127ms / 1 = 0.0155127ms, 0.132964% gpu::code_object::reduce_mean_add_sigmoid_sub_kernel: 0.0154684ms / 1 = 0.0154684ms, 0.132584% gpu::code_object::reduce_prod_mul_log_kernel: 0.0154565ms / 1 = 0.0154565ms, 0.132483% gpu::code_object::pow_mul_kernel: 0.0154345ms / 1 = 0.0154345ms, 0.132294% gpu::code_object::reduce_mean_add_sigmoid_kernel: 0.0153628ms / 1 = 0.0153628ms, 0.13168% gpu::code_object::reduce_prod_mul_rsqrt_add_kernel: 0.0153602ms / 1 = 0.0153602ms, 0.131657% gpu::code_object::reduce_mean_mul_neg_kernel: 0.0153155ms / 1 = 0.0153155ms, 0.131274% gpu::code_object::reduce_mean_add_neg_kernel: 0.0153025ms / 1 = 0.0153025ms, 0.131162% gpu::code_object::add_add_rsqrt_pow_kernel: 0.0152778ms / 1 = 0.0152778ms, 0.130951% gpu::code_object::reduce_mean_mul_rsqrt_kernel: 0.0152524ms / 1 = 0.0152524ms, 0.130733% gpu::code_object::reduce_prod_add_log_kernel: 0.0152335ms / 1 = 0.0152335ms, 0.130571% gpu::code_object::pow_sub_kernel: 0.0151512ms / 1 = 0.0151512ms, 0.129866% gpu::code_object::reduce_mean_kernel: 0.0148606ms / 1 = 0.0148606ms, 0.127375% gpu::code_object::sigmoid_mul_tanh_kernel: 0.0147998ms / 1 = 0.0147998ms, 0.126853% gpu::code_object::noop_add_concat_div_kernel: 0.0145569ms / 1 = 0.0145569ms, 0.124772% gpu::code_object::add_add_rsqrt_add_kernel: 0.0144618ms / 1 = 0.0144618ms, 0.123957% gpu::code_object::add_add_neg_div_kernel: 0.014175ms / 1 = 0.014175ms, 0.121499% gpu::code_object::leaky_relu_kernel: 0.0138746ms / 1 = 0.0138746ms, 0.118924% gpu::code_object::log_div_kernel: 0.0137658ms / 1 = 0.0137658ms, 0.117991% gpu::code_object::convert_add_add_sub_kernel: 0.0135793ms / 1 = 0.0135793ms, 0.116392% gpu::code_object::convert_div_mul_kernel: 0.0135535ms / 1 = 0.0135535ms, 0.116172% gpu::code_object::add_mul_kernel: 0.0135501ms / 1 = 0.0135501ms, 0.116142% gpu::code_object::log_sub_kernel: 0.0135264ms / 1 = 0.0135264ms, 0.115939% gpu::code_object::convert_mul_kernel: 0.0135243ms / 1 = 0.0135243ms, 0.115921% gpu::code_object::neg_sub_kernel: 0.0133901ms / 1 = 0.0133901ms, 0.114771% gpu::code_object::mul_neg_kernel: 0.013293ms / 1 = 0.013293ms, 0.113938% gpu::code_object::add_sub_kernel: 0.0131382ms / 1 = 0.0131382ms, 0.112612% transpose: 0.0130809ms / 22 = 0.000594585ms, 0.11212% gpu::code_object::convert_pow_add_kernel: 0.0129211ms / 1 = 0.0129211ms, 0.110751% check_context::migraphx::version_1::gpu::context: 0.00151968ms / 1 = 0.00151968ms, 0.0130256% hip::sync_stream: 0.00112526ms / 1 = 0.00112526ms, 0.00964495% hip::hip_allocate_memory: 0.0010882ms / 1 = 0.0010882ms, 0.0093273% Batch size: 64 Rate: 12341.6/sec Total time: 5.1857ms Total instructions time: 11.6668ms Overhead time: 0.817418ms, -6.48113ms Overhead: 16%, -125%