Compiling ... Reading: /home/workspace/ByteMLPerf/bytedance/models/model_1/mxr1//model-static-batch-size-32.mxr Allocating params ... Running performance report ... main:@0 = check_context::migraphx::version_1::gpu::context -> float_type, {}, {}: 0.00169648ms, 0.0149049% main:@1 = hip::hip_allocate_memory[shape=float_type, {1819072}, {1},id=main:scratch] -> float_type, {1819072}, {1}: 0.00111702ms, 0.00981389% lyiDxfzZg3:0 = @param:lyiDxfzZg3:0 -> float_type, {32, 701}, {701, 1}: 0.00073322ms, 0.00644191% main:@3 = slice[axes={0, 1},starts={0, 541},ends={32, 573}](lyiDxfzZg3:0) -> float_type, {32, 32}, {701, 1}: 0.00121722ms, 0.0106942% uFSOc5Chp2:0 = @param:uFSOc5Chp2:0 -> float_type, {32, 481}, {481, 1}: 0.00052948ms, 0.00465189% main:@5 = slice[axes={0, 1},starts={0, 337},ends={32, 353}](uFSOc5Chp2:0) -> float_type, {32, 16}, {481, 1}: 0.00084506ms, 0.00742451% nv2tzWRf1I:0 = @param:nv2tzWRf1I:0 -> float_type, {32, 301}, {301, 1}: 0.00051562ms, 0.00453012% main:@7 = slice[axes={0, 1},starts={0, 285},ends={32, 301}](nv2tzWRf1I:0) -> float_type, {32, 16}, {301, 1}: 0.0006886ms, 0.00604989% n4iptOVFwE:0 = @param:n4iptOVFwE:0 -> float_type, {32, 113}, {113, 1}: 0.00045494ms, 0.003997% main:@9 = slice[axes={0, 1},starts={0, 105},ends={32, 113}](n4iptOVFwE:0) -> float_type, {32, 8}, {113, 1}: 0.00069398ms, 0.00609715% main:@10 = load[offset=399872,end=409088](main:@1) -> float_type, {32, 72}, {72, 1}: 0.00055732ms, 0.00489649% main:@11 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=576,local=256,](main:@3,main:@7,main:@5,main:@9,main:@10) -> float_type, {32, 72}, {72, 1}: 0.021948ms, 0.19283% main:@12 = hip::hip_copy_literal[id=main:@literal:150] -> float_type, {128, 16}, {16, 1}: 0.00100868ms, 0.00886204% main:@13 = load[offset=0,end=1280](main:@1) -> float_type, {32, 10, 1}, {10, 1, 1}: 0.00053918ms, 0.00473711% 0MZVmrKeC6:0 = @param:0MZVmrKeC6:0 -> int32_type, {32, 10}, {10, 1}: 0.0005557ms, 0.00488226% main:@15 = unsqueeze[axes={2},steps={}](0MZVmrKeC6:0) -> int32_type, {32, 10, 1}, {10, 1, 1}: 0.00086968ms, 0.00764081% main:@16 = gpu::code_object[code_object=13520,symbol_name=convert_kernel,global=160,local=256,](main:@15,main:@13) -> float_type, {32, 10, 1}, {10, 1, 1}: 0.0132208ms, 0.116155% main:@17 = multibroadcast[out_lens={32, 10, 16}](main:@16) -> float_type, {32, 10, 16}, {10, 1, 0}: 0.00082732ms, 0.00726865% main:@18 = load[offset=29696,end=50176](main:@1) -> float_type, {32, 10, 16}, {160, 16, 1}: 0.00050158ms, 0.00440677% WsiwEgbYlT:0 = @param:WsiwEgbYlT:0 -> float_type, {32, 10, 17}, {170, 17, 1}: 0.0004486ms, 0.0039413% main:@20 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 10, 17}](WsiwEgbYlT:0) -> float_type, {32, 10, 16}, {170, 17, 1}: 0.00077634ms, 0.00682075% main:@21 = gpu::code_object[code_object=13512,symbol_name=add_kernel,global=5120,local=256,](main:@20,main:@17,main:@18) -> float_type, {32, 10, 16}, {160, 16, 1}: 0.0148892ms, 0.130814% main:@22 = load[offset=0,end=29696](main:@1) -> float_type, {32, 232}, {232, 1}: 0.00061604ms, 0.00541239% main:@23 = reshape_lazy[dims={-1, 160}](main:@21) -> float_type, {32, 160}, {160, 1}: 0.0006448ms, 0.00566507% main:@24 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=3712,local=256,](main:@11,main:@23,main:@22) -> float_type, {32, 232}, {232, 1}: 0.0144163ms, 0.126658% main:@25 = hip::hip_copy_literal[id=main:@literal:157] -> float_type, {232, 128}, {128, 1}: 0.0008235ms, 0.00723509% main:@26 = load[offset=46080,end=62464](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00048452ms, 0.00425688% 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, {32, 128}, {128, 1}: 0.0295993ms, 0.260052% main:@28 = load[offset=29696,end=46080](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00063208ms, 0.00555331% main:@29 = gpu::code_object[code_object=13520,symbol_name=log_add_kernel,global=2048,local=256,](main:@27,main:@28) -> float_type, {32, 128}, {128, 1}: 0.0134212ms, 0.117916% main:@30 = hip::hip_copy_literal[id=main:@literal:121] -> float_type, {128, 16}, {16, 1}: 0.00081314ms, 0.00714407% main:@31 = load[offset=409088,end=411136](main:@1) -> float_type, {32, 16}, {16, 1}: 0.00049754ms, 0.00437127% 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, {32, 16}, {16, 1}: 0.0218541ms, 0.192005% YK10Jin6ca:0 = @param:YK10Jin6ca:0 -> int32_type, {32, 10}, {10, 1}: 0.00068634ms, 0.00603003% main:@34 = unsqueeze[axes={2},steps={}](YK10Jin6ca:0) -> int32_type, {32, 10, 1}, {10, 1, 1}: 0.0006254ms, 0.00549462% main:@35 = load[offset=29696,end=30976](main:@1) -> float_type, {32, 10, 1}, {10, 1, 1}: 0.00050818ms, 0.00446476% main:@36 = gpu::code_object[code_object=13520,symbol_name=convert_kernel,global=160,local=256,](main:@34,main:@35) -> float_type, {32, 10, 1}, {10, 1, 1}: 0.013653ms, 0.119952% main:@37 = load[offset=59392,end=79872](main:@1) -> float_type, {32, 10, 16}, {160, 16, 1}: 0.00062462ms, 0.00548777% main:@38 = multibroadcast[out_lens={32, 10, 16}](main:@36) -> float_type, {32, 10, 16}, {10, 1, 0}: 0.00055836ms, 0.00490563% rlh04eiWzL:0 = @param:rlh04eiWzL:0 -> float_type, {32, 10, 17}, {170, 17, 1}: 0.00046082ms, 0.00404866% main:@40 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 10, 17}](rlh04eiWzL:0) -> float_type, {32, 10, 16}, {170, 17, 1}: 0.00089492ms, 0.00786257% main:@41 = gpu::code_object[code_object=13512,symbol_name=pow_kernel,global=5120,local=256,](main:@40,main:@38,main:@37) -> float_type, {32, 10, 16}, {160, 16, 1}: 0.0161444ms, 0.141841% main:@42 = load[offset=29696,end=59392](main:@1) -> float_type, {32, 232}, {232, 1}: 0.00062712ms, 0.00550974% main:@43 = reshape_lazy[dims={-1, 160}](main:@41) -> float_type, {32, 160}, {160, 1}: 0.0005266ms, 0.00462659% main:@44 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=3712,local=256,](main:@11,main:@43,main:@42) -> float_type, {32, 232}, {232, 1}: 0.0144159ms, 0.126655% main:@45 = hip::hip_copy_literal[id=main:@literal:117] -> float_type, {232, 128}, {128, 1}: 0.00075878ms, 0.00666647% main:@46 = load[offset=75776,end=92160](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00045946ms, 0.00403671% 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, {32, 128}, {128, 1}: 0.0237854ms, 0.208973% main:@48 = load[offset=59392,end=75776](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00062052ms, 0.00545175% main:@49 = gpu::code_object[code_object=13520,symbol_name=rsqrt_pow_kernel,global=2048,local=256,](main:@47,main:@48) -> float_type, {32, 128}, {128, 1}: 0.0164516ms, 0.14454% main:@50 = load[offset=411136,end=413184](main:@1) -> float_type, {32, 16}, {16, 1}: 0.00060828ms, 0.00534421% 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, {32, 16}, {16, 1}: 0.0206139ms, 0.181109% uVm4kOBe8l:0 = @param:uVm4kOBe8l:0 -> int32_type, {32, 50}, {50, 1}: 0.00066608ms, 0.00585203% main:@53 = unsqueeze[axes={2},steps={}](uVm4kOBe8l:0) -> int32_type, {32, 50, 1}, {50, 1, 1}: 0.00057884ms, 0.00508556% main:@54 = load[offset=59392,end=65792](main:@1) -> float_type, {32, 50, 1}, {50, 1, 1}: 0.00050168ms, 0.00440765% main:@55 = gpu::code_object[code_object=13520,symbol_name=convert_kernel,global=800,local=256,](main:@53,main:@54) -> float_type, {32, 50, 1}, {50, 1, 1}: 0.0134397ms, 0.118078% main:@56 = multibroadcast[out_lens={32, 50, 16}](main:@55) -> float_type, {32, 50, 16}, {50, 1, 0}: 0.00069016ms, 0.00606359% FkwYpLOMG4:0 = @param:FkwYpLOMG4:0 -> float_type, {32, 50, 17}, {850, 17, 1}: 0.00046474ms, 0.0040831% main:@58 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 50, 17}](FkwYpLOMG4:0) -> float_type, {32, 50, 16}, {850, 17, 1}: 0.00084786ms, 0.00744911% main:@59 = load[offset=65792,end=168192](main:@1) -> float_type, {32, 50, 16}, {800, 16, 1}: 0.00051328ms, 0.00450956% main:@60 = gpu::code_object[code_object=13512,symbol_name=div_kernel,global=25600,local=256,](main:@58,main:@56,main:@59) -> float_type, {32, 50, 16}, {800, 16, 1}: 0.0176518ms, 0.155085% main:@61 = load[offset=176640,end=288256](main:@1) -> float_type, {32, 872}, {872, 1}: 0.00060354ms, 0.00530257% main:@62 = reshape_lazy[dims={-1, 800}](main:@60) -> float_type, {32, 800}, {800, 1}: 0.00050198ms, 0.00441028% main:@63 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=13952,local=256,](main:@11,main:@62,main:@61) -> float_type, {32, 872}, {872, 1}: 0.0160329ms, 0.140862% main:@64 = hip::hip_copy_literal[id=main:@literal:89] -> float_type, {872, 128}, {128, 1}: 0.0008818ms, 0.0077473% main:@65 = load[offset=75776,end=92160](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00050962ms, 0.00447741% 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, {32, 128}, {128, 1}: 0.0263456ms, 0.231467% main:@67 = load[offset=59392,end=75776](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00061234ms, 0.00537988% main:@68 = gpu::code_object[code_object=13520,symbol_name=neg_div_kernel,global=2048,local=256,](main:@66,main:@67) -> float_type, {32, 128}, {128, 1}: 0.0133096ms, 0.116935% main:@69 = hip::hip_copy_literal[id=main:@literal:100] -> float_type, {128, 16}, {16, 1}: 0.00077064ms, 0.00677067% main:@70 = load[offset=413184,end=415232](main:@1) -> float_type, {32, 16}, {16, 1}: 0.00047838ms, 0.00420294% 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, {32, 16}, {16, 1}: 0.0207129ms, 0.181979% fFjVmAbhCX:0 = @param:fFjVmAbhCX:0 -> int32_type, {32, 10}, {10, 1}: 0.00068954ms, 0.00605814% main:@73 = unsqueeze[axes={2},steps={}](fFjVmAbhCX:0) -> int32_type, {32, 10, 1}, {10, 1, 1}: 0.00056818ms, 0.0049919% main:@74 = load[offset=79872,end=81152](main:@1) -> float_type, {32, 10, 1}, {10, 1, 1}: 0.00049808ms, 0.00437602% main:@75 = gpu::code_object[code_object=13520,symbol_name=convert_kernel,global=160,local=256,](main:@73,main:@74) -> float_type, {32, 10, 1}, {10, 1, 1}: 0.01336ms, 0.117378% main:@76 = multibroadcast[out_lens={32, 10, 16}](main:@75) -> float_type, {32, 10, 16}, {10, 1, 0}: 0.00068174ms, 0.00598961% main:@77 = load[offset=59392,end=79872](main:@1) -> float_type, {32, 10, 16}, {160, 16, 1}: 0.00047652ms, 0.0041866% pjwGhTSAys:0 = @param:pjwGhTSAys:0 -> float_type, {32, 10, 17}, {170, 17, 1}: 0.00046062ms, 0.0040469% main:@79 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 10, 17}](pjwGhTSAys:0) -> float_type, {32, 10, 16}, {170, 17, 1}: 0.0008524ms, 0.007489% main:@80 = gpu::code_object[code_object=13512,symbol_name=pow_kernel,global=5120,local=256,](main:@79,main:@76,main:@77) -> float_type, {32, 10, 16}, {160, 16, 1}: 0.0160339ms, 0.14087% main:@81 = load[offset=146944,end=176640](main:@1) -> float_type, {32, 232}, {232, 1}: 0.00066414ms, 0.00583499% main:@82 = reshape_lazy[dims={-1, 160}](main:@80) -> float_type, {32, 160}, {160, 1}: 0.00049714ms, 0.00436776% main:@83 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=3712,local=256,](main:@11,main:@82,main:@81) -> float_type, {32, 232}, {232, 1}: 0.014389ms, 0.126418% main:@84 = hip::hip_copy_literal[id=main:@literal:130] -> float_type, {232, 128}, {128, 1}: 0.00079618ms, 0.00699506% main:@85 = load[offset=75776,end=92160](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00045204ms, 0.00397152% 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, {32, 128}, {128, 1}: 0.0239477ms, 0.210399% main:@87 = load[offset=59392,end=75776](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00063636ms, 0.00559092% main:@88 = gpu::code_object[code_object=13520,symbol_name=log_add_kernel,global=2048,local=256,](main:@86,main:@87) -> float_type, {32, 128}, {128, 1}: 0.0133223ms, 0.117047% main:@89 = load[offset=415232,end=417280](main:@1) -> float_type, {32, 16}, {16, 1}: 0.00060422ms, 0.00530854% 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, {32, 16}, {16, 1}: 0.0202482ms, 0.177896% v3TAQfOsga:0 = @param:v3TAQfOsga:0 -> int32_type, {32, 15}, {15, 1}: 0.00072032ms, 0.00632857% main:@92 = unsqueeze[axes={2},steps={}](v3TAQfOsga:0) -> int32_type, {32, 15, 1}, {15, 1, 1}: 0.0005382ms, 0.0047285% main:@93 = load[offset=59392,end=61312](main:@1) -> float_type, {32, 15, 1}, {15, 1, 1}: 0.00047926ms, 0.00421067% main:@94 = gpu::code_object[code_object=13520,symbol_name=convert_kernel,global=240,local=256,](main:@92,main:@93) -> float_type, {32, 15, 1}, {15, 1, 1}: 0.0135809ms, 0.119319% main:@95 = load[offset=107008,end=145408](main:@1) -> float_type, {32, 15, 20}, {300, 20, 1}: 0.00062352ms, 0.00547811% main:@96 = multibroadcast[out_lens={32, 15, 20}](main:@94) -> float_type, {32, 15, 20}, {15, 1, 0}: 0.00054586ms, 0.0047958% 1qyHnroO6I:0 = @param:1qyHnroO6I:0 -> float_type, {32, 15, 21}, {315, 21, 1}: 0.00052164ms, 0.00458301% main:@98 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 15, 21}](1qyHnroO6I:0) -> float_type, {32, 15, 20}, {315, 21, 1}: 0.0008466ms, 0.00743804% main:@99 = gpu::code_object[code_object=13512,symbol_name=add_kernel,global=9600,local=256,](main:@98,main:@96,main:@95) -> float_type, {32, 15, 20}, {300, 20, 1}: 0.0150496ms, 0.132222% main:@100 = load[offset=59392,end=107008](main:@1) -> float_type, {32, 372}, {372, 1}: 0.00061444ms, 0.00539833% main:@101 = reshape_lazy[dims={-1, 300}](main:@99) -> float_type, {32, 300}, {300, 1}: 0.00050076ms, 0.00439957% main:@102 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=5952,local=256,](main:@11,main:@101,main:@100) -> float_type, {32, 372}, {372, 1}: 0.0146439ms, 0.128658% main:@103 = hip::hip_copy_literal[id=main:@literal:105] -> float_type, {372, 128}, {128, 1}: 0.0007479ms, 0.00657088% main:@104 = load[offset=123392,end=139776](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00047618ms, 0.00418361% 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, {32, 128}, {128, 1}: 0.0275488ms, 0.242038% main:@106 = load[offset=107008,end=123392](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00064084ms, 0.00563028% main:@107 = gpu::code_object[code_object=13520,symbol_name=log_add_kernel,global=2048,local=256,](main:@105,main:@106) -> float_type, {32, 128}, {128, 1}: 0.0133372ms, 0.117178% main:@108 = load[offset=417280,end=419328](main:@1) -> float_type, {32, 16}, {16, 1}: 0.0006184ms, 0.00543312% 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, {32, 16}, {16, 1}: 0.0194417ms, 0.170811% main:@110 = load[offset=107008,end=108928](main:@1) -> float_type, {32, 15, 1}, {15, 1, 1}: 0.00064298ms, 0.00564908% kJg5ca1rqS:0 = @param:kJg5ca1rqS:0 -> int32_type, {32, 15}, {15, 1}: 0.00049506ms, 0.00434949% main:@112 = unsqueeze[axes={2},steps={}](kJg5ca1rqS:0) -> int32_type, {32, 15, 1}, {15, 1, 1}: 0.0005898ms, 0.00518185% main:@113 = gpu::code_object[code_object=13520,symbol_name=convert_kernel,global=240,local=256,](main:@112,main:@110) -> float_type, {32, 15, 1}, {15, 1, 1}: 0.0133348ms, 0.117157% main:@114 = load[offset=288256,end=342016](main:@1) -> float_type, {32, 15, 28}, {420, 28, 1}: 0.00063576ms, 0.00558564% main:@115 = multibroadcast[out_lens={32, 15, 28}](main:@113) -> float_type, {32, 15, 28}, {15, 1, 0}: 0.00053628ms, 0.00471164% fQkYLvqDjr:0 = @param:fQkYLvqDjr:0 -> float_type, {32, 15, 93}, {1395, 93, 1}: 0.00046422ms, 0.00407853% main:@117 = slice[axes={0, 1, 2},starts={0, 0, 65},ends={32, 15, 93}](fQkYLvqDjr:0) -> float_type, {32, 15, 28}, {1395, 93, 1}: 0.0007949ms, 0.00698381% main:@118 = gpu::code_object[code_object=13512,symbol_name=add_kernel,global=13440,local=256,](main:@117,main:@115,main:@114) -> float_type, {32, 15, 28}, {420, 28, 1}: 0.0158117ms, 0.138918% main:@119 = load[offset=995456,end=1058432](main:@1) -> float_type, {32, 492}, {492, 1}: 0.0006065ms, 0.00532857% main:@120 = reshape_lazy[dims={-1, 420}](main:@118) -> float_type, {32, 420}, {420, 1}: 0.00050226ms, 0.00441274% main:@121 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=7872,local=256,](main:@11,main:@120,main:@119) -> float_type, {32, 492}, {492, 1}: 0.0148987ms, 0.130896% main:@122 = hip::hip_copy_literal[id=main:@literal:127] -> float_type, {492, 128}, {128, 1}: 0.00079572ms, 0.00699102% main:@123 = load[offset=123392,end=139776](main:@1) -> float_type, {32, 128}, {128, 1}: 0.0004614ms, 0.00405376% 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, {32, 128}, {128, 1}: 0.0330876ms, 0.290701% main:@125 = load[offset=107008,end=123392](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00061018ms, 0.0053609% main:@126 = gpu::code_object[code_object=13520,symbol_name=rsqrt_pow_kernel,global=2048,local=256,](main:@124,main:@125) -> float_type, {32, 128}, {128, 1}: 0.016462ms, 0.144631% main:@127 = load[offset=419328,end=421376](main:@1) -> float_type, {32, 16}, {16, 1}: 0.000626ms, 0.0054999% 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, {32, 16}, {16, 1}: 0.0201888ms, 0.177374% Ouj93S4aWb:0 = @param:Ouj93S4aWb:0 -> int32_type, {32, 60}, {60, 1}: 0.00067516ms, 0.0059318% main:@130 = unsqueeze[axes={2},steps={}](Ouj93S4aWb:0) -> int32_type, {32, 60, 1}, {60, 1, 1}: 0.00055192ms, 0.00484905% main:@131 = load[offset=107008,end=114688](main:@1) -> float_type, {32, 60, 1}, {60, 1, 1}: 0.00046066ms, 0.00404726% main:@132 = gpu::code_object[code_object=13520,symbol_name=convert_kernel,global=960,local=256,](main:@130,main:@131) -> float_type, {32, 60, 1}, {60, 1, 1}: 0.0134904ms, 0.118523% main:@133 = multibroadcast[out_lens={32, 60, 32}](main:@132) -> float_type, {32, 60, 32}, {60, 1, 0}: 0.0006559ms, 0.00576259% fsm0Ck1QGl:0 = @param:fsm0Ck1QGl:0 -> float_type, {32, 60, 97}, {5820, 97, 1}: 0.00055084ms, 0.00483956% main:@135 = slice[axes={0, 1, 2},starts={0, 0, 65},ends={32, 60, 97}](fsm0Ck1QGl:0) -> float_type, {32, 60, 32}, {5820, 97, 1}: 0.0008156ms, 0.00716568% main:@136 = load[offset=421376,end=667136](main:@1) -> float_type, {32, 60, 32}, {1920, 32, 1}: 0.00045688ms, 0.00401405% main:@137 = gpu::code_object[code_object=13512,symbol_name=div_kernel,global=61440,local=256,](main:@135,main:@133,main:@136) -> float_type, {32, 60, 32}, {1920, 32, 1}: 0.0186239ms, 0.163626% main:@138 = reshape_lazy[dims={-1, 1920}](main:@137) -> float_type, {32, 1920}, {1920, 1}: 0.00066262ms, 0.00582163% main:@139 = load[offset=1249408,end=1504384](main:@1) -> float_type, {32, 1992}, {1992, 1}: 0.00046572ms, 0.00409171% main:@140 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=31872,local=256,](main:@11,main:@138,main:@139) -> float_type, {32, 1992}, {1992, 1}: 0.0186929ms, 0.164232% main:@141 = hip::hip_copy_literal[id=main:@literal:109] -> float_type, {1992, 128}, {128, 1}: 0.00074448ms, 0.00654083% main:@142 = load[offset=123392,end=139776](main:@1) -> float_type, {32, 128}, {128, 1}: 0.0004964ms, 0.00436126% 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, {32, 128}, {128, 1}: 0.0291291ms, 0.255921% main:@144 = load[offset=107008,end=123392](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00060998ms, 0.00535915% main:@145 = gpu::code_object[code_object=13520,symbol_name=neg_pow_kernel,global=2048,local=256,](main:@143,main:@144) -> float_type, {32, 128}, {128, 1}: 0.0163767ms, 0.143882% main:@146 = load[offset=421376,end=423424](main:@1) -> float_type, {32, 16}, {16, 1}: 0.00062162ms, 0.00546141% 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, {32, 16}, {16, 1}: 0.0194375ms, 0.170773% main:@148 = load[offset=107008,end=113408](main:@1) -> float_type, {32, 50, 1}, {50, 1, 1}: 0.0006596ms, 0.0057951% TWPHfwn7r5:0 = @param:TWPHfwn7r5:0 -> int32_type, {32, 50}, {50, 1}: 0.00055956ms, 0.00491617% main:@150 = unsqueeze[axes={2},steps={}](TWPHfwn7r5:0) -> int32_type, {32, 50, 1}, {50, 1, 1}: 0.00057644ms, 0.00506447% main:@151 = gpu::code_object[code_object=13520,symbol_name=convert_kernel,global=800,local=256,](main:@150,main:@148) -> float_type, {32, 50, 1}, {50, 1, 1}: 0.0132302ms, 0.116237% main:@152 = load[offset=423424,end=525824](main:@1) -> float_type, {32, 50, 16}, {800, 16, 1}: 0.00063562ms, 0.00558441% main:@153 = multibroadcast[out_lens={32, 50, 16}](main:@151) -> float_type, {32, 50, 16}, {50, 1, 0}: 0.00054234ms, 0.00476488% NoPO6DfgwG:0 = @param:NoPO6DfgwG:0 -> float_type, {32, 50, 17}, {850, 17, 1}: 0.00051524ms, 0.00452678% main:@155 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 50, 17}](NoPO6DfgwG:0) -> float_type, {32, 50, 16}, {850, 17, 1}: 0.00079472ms, 0.00698223% main:@156 = gpu::code_object[code_object=13512,symbol_name=sub_kernel,global=25600,local=256,](main:@155,main:@153,main:@152) -> float_type, {32, 50, 16}, {800, 16, 1}: 0.0170503ms, 0.1498% main:@157 = reshape_lazy[dims={-1, 800}](main:@156) -> float_type, {32, 800}, {800, 1}: 0.00066102ms, 0.00580757% main:@158 = load[offset=288256,end=399872](main:@1) -> float_type, {32, 872}, {872, 1}: 0.00045304ms, 0.00398031% main:@159 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=13952,local=256,](main:@11,main:@157,main:@158) -> float_type, {32, 872}, {872, 1}: 0.0160196ms, 0.140745% main:@160 = hip::hip_copy_literal[id=main:@literal:61] -> float_type, {872, 128}, {128, 1}: 0.00072994ms, 0.00641309% main:@161 = load[offset=123392,end=139776](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00048402ms, 0.00425249% 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, {32, 128}, {128, 1}: 0.0251315ms, 0.220799% main:@163 = load[offset=107008,end=123392](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00061366ms, 0.00539148% main:@164 = gpu::code_object[code_object=13520,symbol_name=tanh_div_kernel,global=2048,local=256,](main:@162,main:@163) -> float_type, {32, 128}, {128, 1}: 0.0141484ms, 0.124305% main:@165 = hip::hip_copy_literal[id=main:@literal:50] -> float_type, {128, 16}, {16, 1}: 0.00076896ms, 0.00675591% main:@166 = load[offset=423424,end=425472](main:@1) -> float_type, {32, 16}, {16, 1}: 0.0004718ms, 0.00414513% 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, {32, 16}, {16, 1}: 0.0204445ms, 0.179621% loNH8y6bTn:0 = @param:loNH8y6bTn:0 -> int32_type, {32, 15}, {15, 1}: 0.00068126ms, 0.0059854% main:@169 = unsqueeze[axes={2},steps={}](loNH8y6bTn:0) -> int32_type, {32, 15, 1}, {15, 1, 1}: 0.00055714ms, 0.00489491% main:@170 = load[offset=107008,end=108928](main:@1) -> float_type, {32, 15, 1}, {15, 1, 1}: 0.0004958ms, 0.00435599% main:@171 = gpu::code_object[code_object=13520,symbol_name=convert_kernel,global=240,local=256,](main:@169,main:@170) -> float_type, {32, 15, 1}, {15, 1, 1}: 0.0135355ms, 0.11892% main:@172 = load[offset=425472,end=456192](main:@1) -> float_type, {32, 15, 16}, {240, 16, 1}: 0.00064718ms, 0.00568598% main:@173 = multibroadcast[out_lens={32, 15, 16}](main:@171) -> float_type, {32, 15, 16}, {15, 1, 0}: 0.00054742ms, 0.00480951% 4Rg0WA19ot:0 = @param:4Rg0WA19ot:0 -> float_type, {32, 15, 17}, {255, 17, 1}: 0.00054318ms, 0.00477226% main:@175 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 15, 17}](4Rg0WA19ot:0) -> float_type, {32, 15, 16}, {255, 17, 1}: 0.0008196ms, 0.00720082% main:@176 = gpu::code_object[code_object=13512,symbol_name=add_kernel,global=7680,local=256,](main:@175,main:@173,main:@172) -> float_type, {32, 15, 16}, {240, 16, 1}: 0.0147016ms, 0.129165% main:@177 = reshape_lazy[dims={-1, 240}](main:@176) -> float_type, {32, 240}, {240, 1}: 0.00064424ms, 0.00566015% main:@178 = load[offset=107008,end=146944](main:@1) -> float_type, {32, 312}, {312, 1}: 0.00046718ms, 0.00410454% main:@179 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=4992,local=256,](main:@11,main:@177,main:@178) -> float_type, {32, 312}, {312, 1}: 0.0144448ms, 0.126908% main:@180 = hip::hip_copy_literal[id=main:@literal:139] -> float_type, {312, 128}, {128, 1}: 0.00079826ms, 0.00701333% main:@181 = load[offset=443904,end=460288](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00045358ms, 0.00398505% 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, {32, 128}, {128, 1}: 0.0260663ms, 0.229013% main:@183 = load[offset=427520,end=443904](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00061414ms, 0.0053957% main:@184 = gpu::code_object[code_object=13520,symbol_name=neg_pow_kernel,global=2048,local=256,](main:@182,main:@183) -> float_type, {32, 128}, {128, 1}: 0.0162366ms, 0.142651% main:@185 = hip::hip_copy_literal[id=main:@literal:91] -> float_type, {128, 16}, {16, 1}: 0.00076944ms, 0.00676013% main:@186 = load[offset=425472,end=427520](main:@1) -> float_type, {32, 16}, {16, 1}: 0.00047508ms, 0.00417395% 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, {32, 16}, {16, 1}: 0.0193025ms, 0.169588% vOI1hWzBDK:0 = @param:vOI1hWzBDK:0 -> int32_type, {32, 50}, {50, 1}: 0.00068804ms, 0.00604497% main:@189 = unsqueeze[axes={2},steps={}](vOI1hWzBDK:0) -> int32_type, {32, 50, 1}, {50, 1, 1}: 0.00056094ms, 0.00492829% main:@190 = load[offset=683520,end=689920](main:@1) -> float_type, {32, 50, 1}, {50, 1, 1}: 0.00046852ms, 0.00411631% main:@191 = gpu::code_object[code_object=13520,symbol_name=convert_kernel,global=800,local=256,](main:@189,main:@190) -> float_type, {32, 50, 1}, {50, 1, 1}: 0.0133713ms, 0.117477% main:@192 = load[offset=427520,end=683520](main:@1) -> float_type, {32, 50, 40}, {2000, 40, 1}: 0.00064228ms, 0.00564293% main:@193 = multibroadcast[out_lens={32, 50, 40}](main:@191) -> float_type, {32, 50, 40}, {50, 1, 0}: 0.00054968ms, 0.00482937% UrTfLOxlV3:0 = @param:UrTfLOxlV3:0 -> float_type, {32, 50, 41}, {2050, 41, 1}: 0.00047834ms, 0.00420259% main:@195 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 50, 41}](UrTfLOxlV3:0) -> float_type, {32, 50, 40}, {2050, 41, 1}: 0.00080672ms, 0.00708766% main:@196 = gpu::code_object[code_object=13512,symbol_name=sub_kernel,global=64000,local=256,](main:@195,main:@193,main:@192) -> float_type, {32, 50, 40}, {2000, 40, 1}: 0.0187088ms, 0.164371% main:@197 = load[offset=2017408,end=2282624](main:@1) -> float_type, {32, 2072}, {2072, 1}: 0.00062086ms, 0.00545474% main:@198 = reshape_lazy[dims={-1, 2000}](main:@196) -> float_type, {32, 2000}, {2000, 1}: 0.0004949ms, 0.00434808% main:@199 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=33152,local=256,](main:@11,main:@198,main:@197) -> float_type, {32, 2072}, {2072, 1}: 0.018578ms, 0.163222% main:@200 = hip::hip_copy_literal[id=main:@literal:85] -> float_type, {2072, 128}, {128, 1}: 0.00074164ms, 0.00651588% main:@201 = load[offset=445952,end=462336](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00046762ms, 0.0041084% 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, {32, 128}, {128, 1}: 0.0301405ms, 0.264808% main:@203 = load[offset=429568,end=445952](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00060816ms, 0.00534316% main:@204 = gpu::code_object[code_object=13520,symbol_name=log_add_kernel,global=2048,local=256,](main:@202,main:@203) -> float_type, {32, 128}, {128, 1}: 0.0132508ms, 0.116419% main:@205 = hip::hip_copy_literal[id=main:@literal:79] -> float_type, {128, 16}, {16, 1}: 0.00074316ms, 0.00652924% main:@206 = load[offset=427520,end=429568](main:@1) -> float_type, {32, 16}, {16, 1}: 0.00047472ms, 0.00417078% 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, {32, 16}, {16, 1}: 0.0204354ms, 0.179541% vyDGpmRhC3:0 = @param:vyDGpmRhC3:0 -> int32_type, {32, 60}, {60, 1}: 0.0006899ms, 0.00606131% main:@209 = unsqueeze[axes={2},steps={}](vyDGpmRhC3:0) -> int32_type, {32, 60, 1}, {60, 1, 1}: 0.00055178ms, 0.00484782% main:@210 = load[offset=552448,end=560128](main:@1) -> float_type, {32, 60, 1}, {60, 1, 1}: 0.00047214ms, 0.00414812% main:@211 = gpu::code_object[code_object=13520,symbol_name=convert_kernel,global=960,local=256,](main:@209,main:@210) -> float_type, {32, 60, 1}, {60, 1, 1}: 0.0133347ms, 0.117156% main:@212 = load[offset=429568,end=552448](main:@1) -> float_type, {32, 60, 16}, {960, 16, 1}: 0.00062194ms, 0.00546423% main:@213 = multibroadcast[out_lens={32, 60, 16}](main:@211) -> float_type, {32, 60, 16}, {60, 1, 0}: 0.00054366ms, 0.00477647% y5fPVEH2j4:0 = @param:y5fPVEH2j4:0 -> float_type, {32, 60, 17}, {1020, 17, 1}: 0.00053818ms, 0.00472833% main:@215 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 60, 17}](y5fPVEH2j4:0) -> float_type, {32, 60, 16}, {1020, 17, 1}: 0.00080634ms, 0.00708432% main:@216 = gpu::code_object[code_object=13512,symbol_name=add_kernel,global=30720,local=256,](main:@215,main:@213,main:@212) -> float_type, {32, 60, 16}, {960, 16, 1}: 0.018462ms, 0.162203% main:@217 = load[offset=1773696,end=1905792](main:@1) -> float_type, {32, 1032}, {1032, 1}: 0.00062524ms, 0.00549322% main:@218 = reshape_lazy[dims={-1, 960}](main:@216) -> float_type, {32, 960}, {960, 1}: 0.00052342ms, 0.00459865% main:@219 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=16512,local=256,](main:@11,main:@218,main:@217) -> float_type, {32, 1032}, {1032, 1}: 0.0161839ms, 0.142188% main:@220 = hip::hip_copy_literal[id=main:@literal:56] -> float_type, {1032, 128}, {128, 1}: 0.00076448ms, 0.00671655% main:@221 = load[offset=448000,end=464384](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00045426ms, 0.00399103% 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, {32, 128}, {128, 1}: 0.0252282ms, 0.221649% main:@223 = load[offset=431616,end=448000](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00062106ms, 0.00545649% main:@224 = gpu::code_object[code_object=13520,symbol_name=log_sub_kernel,global=2048,local=256,](main:@222,main:@223) -> float_type, {32, 128}, {128, 1}: 0.0132343ms, 0.116273% main:@225 = hip::hip_copy_literal[id=main:@literal:90] -> float_type, {128, 16}, {16, 1}: 0.00070458ms, 0.00619028% main:@226 = load[offset=429568,end=431616](main:@1) -> float_type, {32, 16}, {16, 1}: 0.00046006ms, 0.00404198% 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, {32, 16}, {16, 1}: 0.0197559ms, 0.173571% yDES0Cj7Td:0 = @param:yDES0Cj7Td:0 -> int32_type, {32, 10}, {10, 1}: 0.00067472ms, 0.00592794% main:@229 = unsqueeze[axes={2},steps={}](yDES0Cj7Td:0) -> int32_type, {32, 10, 1}, {10, 1, 1}: 0.00054522ms, 0.00479018% main:@230 = load[offset=452096,end=453376](main:@1) -> float_type, {32, 10, 1}, {10, 1, 1}: 0.00045816ms, 0.00402529% main:@231 = gpu::code_object[code_object=13520,symbol_name=convert_kernel,global=160,local=256,](main:@229,main:@230) -> float_type, {32, 10, 1}, {10, 1, 1}: 0.0132118ms, 0.116076% main:@232 = load[offset=431616,end=452096](main:@1) -> float_type, {32, 10, 16}, {160, 16, 1}: 0.00062104ms, 0.00545632% main:@233 = multibroadcast[out_lens={32, 10, 16}](main:@231) -> float_type, {32, 10, 16}, {10, 1, 0}: 0.0005451ms, 0.00478913% r1GsND0okc:0 = @param:r1GsND0okc:0 -> float_type, {32, 10, 17}, {170, 17, 1}: 0.00051052ms, 0.00448531% main:@235 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 10, 17}](r1GsND0okc:0) -> float_type, {32, 10, 16}, {170, 17, 1}: 0.00079442ms, 0.0069796% main:@236 = gpu::code_object[code_object=13512,symbol_name=sub_kernel,global=5120,local=256,](main:@235,main:@233,main:@232) -> float_type, {32, 10, 16}, {160, 16, 1}: 0.0144331ms, 0.126806% main:@237 = reshape_lazy[dims={-1, 160}](main:@236) -> float_type, {32, 160}, {160, 1}: 0.00065848ms, 0.00578526% main:@238 = load[offset=1089152,end=1118848](main:@1) -> float_type, {32, 232}, {232, 1}: 0.0004605ms, 0.00404585% main:@239 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=3712,local=256,](main:@11,main:@237,main:@238) -> float_type, {32, 232}, {232, 1}: 0.0144077ms, 0.126583% main:@240 = hip::hip_copy_literal[id=main:@literal:132] -> float_type, {232, 128}, {128, 1}: 0.0007483ms, 0.0065744% main:@241 = load[offset=431616,end=448000](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00046144ms, 0.00405411% 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, {32, 128}, {128, 1}: 0.0237123ms, 0.208331% main:@243 = hip::hip_copy_literal[id=main:@literal:104] -> float_type, {128, 16}, {16, 1}: 0.0007808ms, 0.00685993% main:@244 = load[offset=448000,end=464384](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00047816ms, 0.00420101% main:@245 = gpu::code_object[code_object=13520,symbol_name=neg_div_kernel,global=2048,local=256,](main:@242,main:@244) -> float_type, {32, 128}, {128, 1}: 0.0133377ms, 0.117182% main:@246 = load[offset=431616,end=433664](main:@1) -> float_type, {32, 16}, {16, 1}: 0.00062794ms, 0.00551694% 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, {32, 16}, {16, 1}: 0.0201443ms, 0.176983% main:@248 = load[offset=433664,end=440064](main:@1) -> float_type, {32, 50, 1}, {50, 1, 1}: 0.00064024ms, 0.00562501% KOJcBD1Cxa:0 = @param:KOJcBD1Cxa:0 -> int32_type, {32, 50}, {50, 1}: 0.00055436ms, 0.00487048% main:@250 = unsqueeze[axes={2},steps={}](KOJcBD1Cxa:0) -> int32_type, {32, 50, 1}, {50, 1, 1}: 0.0005871ms, 0.00515813% main:@251 = gpu::code_object[code_object=13520,symbol_name=convert_kernel,global=800,local=256,](main:@250,main:@248) -> float_type, {32, 50, 1}, {50, 1, 1}: 0.0132179ms, 0.116129% main:@252 = multibroadcast[out_lens={32, 50, 20}](main:@251) -> float_type, {32, 50, 20}, {50, 1, 0}: 0.00070012ms, 0.0061511% gkGLzseUSO:0 = @param:gkGLzseUSO:0 -> float_type, {32, 50, 21}, {1050, 21, 1}: 0.00054354ms, 0.00477542% main:@254 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 50, 21}](gkGLzseUSO:0) -> float_type, {32, 50, 20}, {1050, 21, 1}: 0.00085796ms, 0.00753784% main:@255 = load[offset=440064,end=568064](main:@1) -> float_type, {32, 50, 20}, {1000, 20, 1}: 0.00047186ms, 0.00414566% main:@256 = gpu::code_object[code_object=13512,symbol_name=div_kernel,global=32000,local=256,](main:@254,main:@252,main:@255) -> float_type, {32, 50, 20}, {1000, 20, 1}: 0.0183757ms, 0.161445% main:@257 = load[offset=1636480,end=1773696](main:@1) -> float_type, {32, 1072}, {1072, 1}: 0.0006011ms, 0.00528113% main:@258 = reshape_lazy[dims={-1, 1000}](main:@256) -> float_type, {32, 1000}, {1000, 1}: 0.0005008ms, 0.00439992% main:@259 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=17152,local=256,](main:@11,main:@258,main:@257) -> float_type, {32, 1072}, {1072, 1}: 0.016261ms, 0.142865% main:@260 = hip::hip_copy_literal[id=main:@literal:57] -> float_type, {1072, 128}, {128, 1}: 0.00077466ms, 0.00680599% main:@261 = load[offset=454144,end=470528](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00046516ms, 0.00408679% main:@262 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@259,main:@260,main:@261) -> float_type, {32, 128}, {128, 1}: 0.0261136ms, 0.229428% main:@263 = load[offset=437760,end=454144](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00061428ms, 0.00539693% main:@264 = gpu::code_object[code_object=13520,symbol_name=neg_add_kernel,global=2048,local=256,](main:@262,main:@263) -> float_type, {32, 128}, {128, 1}: 0.0131844ms, 0.115835% main:@265 = load[offset=435712,end=437760](main:@1) -> float_type, {32, 16}, {16, 1}: 0.0005941ms, 0.00521963% 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, {32, 16}, {16, 1}: 0.0198955ms, 0.174798% CedNyHXqE1:0 = @param:CedNyHXqE1:0 -> int32_type, {32, 50}, {50, 1}: 0.00069454ms, 0.00610207% main:@268 = unsqueeze[axes={2},steps={}](CedNyHXqE1:0) -> int32_type, {32, 50, 1}, {50, 1, 1}: 0.00055148ms, 0.00484518% main:@269 = load[offset=540160,end=546560](main:@1) -> float_type, {32, 50, 1}, {50, 1, 1}: 0.00046788ms, 0.00411069% main:@270 = gpu::code_object[code_object=13520,symbol_name=convert_kernel,global=800,local=256,](main:@268,main:@269) -> float_type, {32, 50, 1}, {50, 1, 1}: 0.013344ms, 0.117238% main:@271 = load[offset=437760,end=540160](main:@1) -> float_type, {32, 50, 16}, {800, 16, 1}: 0.00062772ms, 0.00551501% main:@272 = multibroadcast[out_lens={32, 50, 16}](main:@270) -> float_type, {32, 50, 16}, {50, 1, 0}: 0.00054574ms, 0.00479475% YsGAJ1iRPS:0 = @param:YsGAJ1iRPS:0 -> float_type, {32, 50, 17}, {850, 17, 1}: 0.00052576ms, 0.00461921% main:@274 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 50, 17}](YsGAJ1iRPS:0) -> float_type, {32, 50, 16}, {850, 17, 1}: 0.00083344ms, 0.00732242% main:@275 = gpu::code_object[code_object=13512,symbol_name=add_kernel,global=25600,local=256,](main:@274,main:@272,main:@271) -> float_type, {32, 50, 16}, {800, 16, 1}: 0.0172002ms, 0.151117% main:@276 = load[offset=1905792,end=2017408](main:@1) -> float_type, {32, 872}, {872, 1}: 0.00061982ms, 0.0054456% main:@277 = reshape_lazy[dims={-1, 800}](main:@275) -> float_type, {32, 800}, {800, 1}: 0.000522ms, 0.00458617% main:@278 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=13952,local=256,](main:@11,main:@277,main:@276) -> float_type, {32, 872}, {872, 1}: 0.0158172ms, 0.138966% main:@279 = hip::hip_copy_literal[id=main:@literal:159] -> float_type, {872, 128}, {128, 1}: 0.00078466ms, 0.00689385% main:@280 = load[offset=454144,end=470528](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00046734ms, 0.00410594% 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, {32, 128}, {128, 1}: 0.0250729ms, 0.220285% main:@282 = load[offset=437760,end=454144](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00060736ms, 0.00533613% main:@283 = gpu::code_object[code_object=13520,symbol_name=neg_add_kernel,global=2048,local=256,](main:@281,main:@282) -> float_type, {32, 128}, {128, 1}: 0.0132301ms, 0.116237% main:@284 = hip::hip_copy_literal[id=main:@literal:64] -> float_type, {128, 16}, {16, 1}: 0.00070538ms, 0.00619731% main:@285 = load[offset=433664,end=435712](main:@1) -> float_type, {32, 16}, {16, 1}: 0.00046192ms, 0.00405833% 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, {32, 16}, {16, 1}: 0.0204886ms, 0.180008% I5iNBsTtoe:0 = @param:I5iNBsTtoe:0 -> int32_type, {32, 60}, {60, 1}: 0.00067672ms, 0.00594551% main:@288 = unsqueeze[axes={2},steps={}](I5iNBsTtoe:0) -> int32_type, {32, 60, 1}, {60, 1, 1}: 0.00055272ms, 0.00485607% main:@289 = load[offset=437760,end=445440](main:@1) -> float_type, {32, 60, 1}, {60, 1, 1}: 0.000467ms, 0.00410296% main:@290 = gpu::code_object[code_object=13520,symbol_name=convert_kernel,global=960,local=256,](main:@288,main:@289) -> float_type, {32, 60, 1}, {60, 1, 1}: 0.0132471ms, 0.116386% main:@291 = multibroadcast[out_lens={32, 60, 16}](main:@290) -> float_type, {32, 60, 16}, {60, 1, 0}: 0.00065686ms, 0.00577102% S3KJ2tl1VG:0 = @param:S3KJ2tl1VG:0 -> float_type, {32, 60, 17}, {1020, 17, 1}: 0.00052794ms, 0.00463836% main:@293 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 60, 17}](S3KJ2tl1VG:0) -> float_type, {32, 60, 16}, {1020, 17, 1}: 0.00097164ms, 0.00853661% main:@294 = load[offset=445440,end=568320](main:@1) -> float_type, {32, 60, 16}, {960, 16, 1}: 0.00049ms, 0.00430503% main:@295 = gpu::code_object[code_object=13512,symbol_name=pow_kernel,global=30720,local=256,](main:@293,main:@291,main:@294) -> float_type, {32, 60, 16}, {960, 16, 1}: 0.0191987ms, 0.168675% main:@296 = load[offset=1504384,end=1636480](main:@1) -> float_type, {32, 1032}, {1032, 1}: 0.00060544ms, 0.00531926% main:@297 = reshape_lazy[dims={-1, 960}](main:@295) -> float_type, {32, 960}, {960, 1}: 0.00049456ms, 0.00434509% main:@298 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=16512,local=256,](main:@11,main:@297,main:@296) -> float_type, {32, 1032}, {1032, 1}: 0.0159793ms, 0.140391% main:@299 = hip::hip_copy_literal[id=main:@literal:151] -> float_type, {1032, 128}, {128, 1}: 0.00076582ms, 0.00672832% main:@300 = load[offset=454144,end=470528](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00047746ms, 0.00419486% main:@301 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@298,main:@299,main:@300) -> float_type, {32, 128}, {128, 1}: 0.0251847ms, 0.221267% main:@302 = load[offset=437760,end=454144](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00063506ms, 0.00557949% main:@303 = gpu::code_object[code_object=13520,symbol_name=neg_sub_kernel,global=2048,local=256,](main:@301,main:@302) -> float_type, {32, 128}, {128, 1}: 0.0132143ms, 0.116098% main:@304 = hip::hip_copy_literal[id=main:@literal:108] -> float_type, {128, 16}, {16, 1}: 0.00080316ms, 0.00705638% main:@305 = load[offset=399872,end=401920](main:@1) -> float_type, {32, 16}, {16, 1}: 0.000455ms, 0.00399753% main:@306 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@303,main:@304,main:@305) -> float_type, {32, 16}, {16, 1}: 0.0203701ms, 0.178967% main:@307 = load[offset=401920,end=406016](main:@1) -> float_type, {32, 32}, {32, 1}: 0.00063002ms, 0.00553521% main:@308 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=512,local=256,](main:@266,main:@167,main:@307) -> float_type, {32, 32}, {32, 1}: 0.0134177ms, 0.117885% main:@309 = load[offset=445952,end=456192](main:@1) -> float_type, {32, 80}, {80, 1}: 0.000623ms, 0.00547354% main:@310 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=512,local=256,](main:@128,main:@266,main:@167,main:@109,main:@187,main:@309) -> float_type, {32, 80}, {80, 1}: 0.0148775ms, 0.13071% main:@311 = load[offset=435712,end=445952](main:@1) -> float_type, {32, 80}, {80, 1}: 0.00064706ms, 0.00568492% main:@312 = gpu::code_object[code_object=13560,symbol_name=neg_rsqrt_tanh_log_concat_add_kernel,global=640,local=256,](main:@128,main:@308,main:@109,main:@187,main:@310,main:@311) -> float_type, {32, 80}, {80, 1}: 0.0158912ms, 0.139617% main:@313 = load[offset=401920,end=403968](main:@1) -> float_type, {32, 16}, {16, 1}: 0.00060208ms, 0.00528974% main:@314 = gpu::code_object[code_object=13520,symbol_name=rsqrt_sub_kernel,global=256,local=256,](main:@227,main:@313) -> float_type, {32, 16}, {16, 1}: 0.0129998ms, 0.114213% main:@315 = load[offset=403968,end=406016](main:@1) -> float_type, {32, 16}, {16, 1}: 0.00059794ms, 0.00525337% main:@316 = gpu::code_object[code_object=13520,symbol_name=tanh_sub_kernel,global=256,local=256,](main:@32,main:@315) -> float_type, {32, 16}, {16, 1}: 0.013693ms, 0.120304% main:@317 = load[offset=406016,end=408064](main:@1) -> float_type, {32, 16}, {16, 1}: 0.00059502ms, 0.00522771% main:@318 = gpu::code_object[code_object=13520,symbol_name=tanh_div_kernel,global=256,local=256,](main:@147,main:@317) -> float_type, {32, 16}, {16, 1}: 0.0140445ms, 0.123392% main:@319 = load[offset=408064,end=410112](main:@1) -> float_type, {32, 16}, {16, 1}: 0.00064054ms, 0.00562764% main:@320 = gpu::code_object[code_object=13520,symbol_name=tanh_sub_kernel,global=256,local=256,](main:@51,main:@319) -> float_type, {32, 16}, {16, 1}: 0.0136352ms, 0.119796% main:@321 = load[offset=410112,end=412160](main:@1) -> float_type, {32, 16}, {16, 1}: 0.00059322ms, 0.0052119% main:@322 = gpu::code_object[code_object=13520,symbol_name=rsqrt_sub_kernel,global=256,local=256,](main:@71,main:@321) -> float_type, {32, 16}, {16, 1}: 0.0129848ms, 0.114081% main:@323 = load[offset=412160,end=414208](main:@1) -> float_type, {32, 16}, {16, 1}: 0.00062452ms, 0.00548689% main:@324 = gpu::code_object[code_object=13520,symbol_name=neg_div_kernel,global=256,local=256,](main:@90,main:@323) -> float_type, {32, 16}, {16, 1}: 0.0129576ms, 0.113843% main:@325 = load[offset=414208,end=416256](main:@1) -> float_type, {32, 16}, {16, 1}: 0.00060224ms, 0.00529115% main:@326 = gpu::code_object[code_object=13520,symbol_name=tanh_add_kernel,global=256,local=256,](main:@247,main:@325) -> float_type, {32, 16}, {16, 1}: 0.0135609ms, 0.119143% main:@327 = load[offset=416256,end=418304](main:@1) -> float_type, {32, 16}, {16, 1}: 0.0006013ms, 0.00528289% main:@328 = gpu::code_object[code_object=13520,symbol_name=rsqrt_add_kernel,global=256,local=256,](main:@207,main:@327) -> float_type, {32, 16}, {16, 1}: 0.0129111ms, 0.113434% main:@329 = load[offset=418304,end=420352](main:@1) -> float_type, {32, 16}, {16, 1}: 0.0005975ms, 0.0052495% main:@330 = gpu::code_object[code_object=13520,symbol_name=rsqrt_pow_kernel,global=256,local=256,](main:@306,main:@329) -> float_type, {32, 16}, {16, 1}: 0.0162814ms, 0.143045% main:@331 = load[offset=399872,end=401920](main:@1) -> float_type, {32, 16}, {16, 1}: 0.00061262ms, 0.00538234% main:@332 = gpu::code_object[code_object=13520,symbol_name=neg_div_kernel,global=256,local=256,](main:@286,main:@331) -> float_type, {32, 16}, {16, 1}: 0.0131325ms, 0.115379% main:@333 = load[offset=1058432,end=1089152](main:@1) -> float_type, {32, 240}, {240, 1}: 0.00061066ms, 0.00536512% main:@334 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=698,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, {32, 240}, {240, 1}: 0.0191642ms, 0.168373% main:@335 = hip::hip_copy_literal[id=main:@literal:107] -> float_type, {128, 240}, {240, 1}: 0.00080214ms, 0.00704742% main:@336 = hip::hip_copy_literal[id=main:@literal:137] -> float_type, {128, 240}, {240, 1}: 0.00060586ms, 0.00532295% w27LgiJW6P:0 = @param:w27LgiJW6P:0 -> float_type, {32, 13}, {13, 1}: 0.00055142ms, 0.00484465% main:@338 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](w27LgiJW6P:0) -> float_type, {32, 4}, {13, 1}: 0.00084198ms, 0.00739745% 0cjwsL5ldh:0 = @param:0cjwsL5ldh:0 -> float_type, {32, 33}, {33, 1}: 0.00050338ms, 0.00442258% main:@340 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](0cjwsL5ldh:0) -> float_type, {32, 16}, {33, 1}: 0.0006629ms, 0.00582409% 0sDBrPmfS9:0 = @param:0sDBrPmfS9:0 -> float_type, {32, 13}, {13, 1}: 0.00049244ms, 0.00432647% main:@342 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](0sDBrPmfS9:0) -> float_type, {32, 4}, {13, 1}: 0.0008069ms, 0.00708924% whRnWGgf0Q:0 = @param:whRnWGgf0Q:0 -> float_type, {32, 13}, {13, 1}: 0.00051156ms, 0.00449445% main:@344 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](whRnWGgf0Q:0) -> float_type, {32, 4}, {13, 1}: 0.00061396ms, 0.00539411% 3mk8BKjDey:0 = @param:3mk8BKjDey:0 -> float_type, {32, 25}, {25, 1}: 0.00050464ms, 0.00443365% main:@346 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](3mk8BKjDey:0) -> float_type, {32, 16}, {25, 1}: 0.00079314ms, 0.00696835% TxEdCcekGS:0 = @param:TxEdCcekGS:0 -> float_type, {32, 49}, {49, 1}: 0.00057674ms, 0.00506711% main:@348 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](TxEdCcekGS:0) -> float_type, {32, 8}, {49, 1}: 0.00064232ms, 0.00564328% LKAvFme9ZS:0 = @param:LKAvFme9ZS:0 -> float_type, {32, 49}, {49, 1}: 0.0004696ms, 0.0041258% main:@350 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](LKAvFme9ZS:0) -> float_type, {32, 8}, {49, 1}: 0.00095464ms, 0.00838725% KFhBukd7EV:0 = @param:KFhBukd7EV:0 -> float_type, {32, 13}, {13, 1}: 0.00052814ms, 0.00464012% main:@352 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](KFhBukd7EV:0) -> float_type, {32, 4}, {13, 1}: 0.00061404ms, 0.00539482% main:@353 = slice[axes={0, 1},starts={0, 257},ends={32, 289}](lyiDxfzZg3:0) -> float_type, {32, 32}, {701, 1}: 0.00060744ms, 0.00533683% m5MfQhR1Bp:0 = @param:m5MfQhR1Bp:0 -> float_type, {32, 13}, {13, 1}: 0.00046268ms, 0.004065% main:@355 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](m5MfQhR1Bp:0) -> float_type, {32, 4}, {13, 1}: 0.00084988ms, 0.00746686% main:@356 = slice[axes={0, 1},starts={0, 129},ends={32, 161}](nv2tzWRf1I:0) -> float_type, {32, 32}, {301, 1}: 0.00067918ms, 0.00596712% Qp3K1PHvte:0 = @param:Qp3K1PHvte:0 -> float_type, {32, 69}, {69, 1}: 0.00049766ms, 0.00437233% main:@358 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](Qp3K1PHvte:0) -> float_type, {32, 16}, {69, 1}: 0.0007039ms, 0.00618431% S5AzoJmqUC:0 = @param:S5AzoJmqUC:0 -> float_type, {32, 13}, {13, 1}: 0.0004515ms, 0.00396678% main:@360 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](S5AzoJmqUC:0) -> float_type, {32, 4}, {13, 1}: 0.0009041ms, 0.00794322% main:@361 = slice[axes={0, 1},starts={0, 129},ends={32, 161}](uFSOc5Chp2:0) -> float_type, {32, 32}, {481, 1}: 0.00075308ms, 0.00661639% Vi28Dy4Ll5:0 = @param:Vi28Dy4Ll5:0 -> float_type, {32, 13}, {13, 1}: 0.00051472ms, 0.00452221% main:@363 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](Vi28Dy4Ll5:0) -> float_type, {32, 4}, {13, 1}: 0.00069456ms, 0.00610225% mIR0ToEnMH:0 = @param:mIR0ToEnMH:0 -> float_type, {32, 13}, {13, 1}: 0.00051108ms, 0.00449023% main:@365 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](mIR0ToEnMH:0) -> float_type, {32, 4}, {13, 1}: 0.00084984ms, 0.0074665% cHzIDJm6us:0 = @param:cHzIDJm6us:0 -> float_type, {32, 33}, {33, 1}: 0.00049654ms, 0.00436249% main:@367 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](cHzIDJm6us:0) -> float_type, {32, 16}, {33, 1}: 0.0006969ms, 0.00612281% ifP6Ceko5T:0 = @param:ifP6Ceko5T:0 -> float_type, {32, 73}, {73, 1}: 0.00047918ms, 0.00420997% main:@369 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](ifP6Ceko5T:0) -> float_type, {32, 16}, {73, 1}: 0.00070296ms, 0.00617605% gU9SoiHyRO:0 = @param:gU9SoiHyRO:0 -> float_type, {32, 13}, {13, 1}: 0.0005216ms, 0.00458266% main:@371 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](gU9SoiHyRO:0) -> float_type, {32, 4}, {13, 1}: 0.00067208ms, 0.00590474% d1ExPaGqMN:0 = @param:d1ExPaGqMN:0 -> float_type, {32, 13}, {13, 1}: 0.00049986ms, 0.00439166% main:@373 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](d1ExPaGqMN:0) -> float_type, {32, 4}, {13, 1}: 0.00069368ms, 0.00609452% qeZ76Hc03v:0 = @param:qeZ76Hc03v:0 -> float_type, {32, 13}, {13, 1}: 0.00047168ms, 0.00414407% main:@375 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](qeZ76Hc03v:0) -> float_type, {32, 4}, {13, 1}: 0.0006468ms, 0.00568264% main:@376 = load[offset=399872,end=430080](main:@1) -> float_type, {32, 236}, {236, 1}: 0.000474ms, 0.00416446% main:@377 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=359,local=256,](main:@353,main:@356,main:@361,main:@369,main:@358,main:@350,main:@348,main:@367,main:@346,main:@340,main:@342,main:@371,main:@352,main:@375,main:@363,main:@355,main:@365,main:@373,main:@360,main:@344,main:@338,main:@376) -> float_type, {32, 236}, {236, 1}: 0.0361932ms, 0.317985% main:@378 = hip::hip_copy_literal[id=main:@literal:5] -> float_type, {236, 896}, {896, 1}: 0.00081386ms, 0.00715039% main:@379 = load[offset=1134720,end=1249408](main:@1) -> float_type, {32, 896}, {896, 1}: 0.00047984ms, 0.00421577% 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, {32, 896}, {896, 1}: 0.0275292ms, 0.241866% main:@381 = hip::hip_copy_literal[id=main:@literal:16] -> float_type, {128, 18, 18}, {324, 18, 1}: 0.0007879ms, 0.00692231% main:@382 = load[offset=2282624,end=2448512](main:@1) -> float_type, {128, 18, 18}, {324, 18, 1}: 0.0004852ms, 0.00426286% main:@383 = gpu::code_object[code_object=14560,symbol_name=softmax_kernel,global=147456,local=64,](main:@381,main:@382) -> float_type, {128, 18, 18}, {324, 18, 1}: 0.0207205ms, 0.182046% main:@384 = hip::hip_copy_literal[id=main:@literal:8] -> float_type, {128, 18, 64}, {1152, 64, 1}: 0.00071242ms, 0.00625916% main:@385 = load[offset=405632,end=995456](main:@1) -> float_type, {128, 18, 64}, {1152, 64, 1}: 0.00052936ms, 0.00465084% 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, {128, 18, 64}, {1152, 64, 1}: 0.0241322ms, 0.21202% main:@387 = load[offset=2282624,end=2376320](main:@1) -> float_type, {32, 732}, {732, 1}: 0.00063262ms, 0.00555806% main:@388 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=11712,local=256,](main:@121,main:@334,main:@387) -> float_type, {32, 732}, {732, 1}: 0.0156934ms, 0.137879% main:@389 = hip::hip_copy_literal[id=main:@literal:155] -> float_type, {732, 128}, {128, 1}: 0.00076578ms, 0.00672797% main:@390 = load[offset=2376320,end=2392704](main:@1) -> float_type, {32, 128}, {128, 1}: 0.0004711ms, 0.00413898% 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, {32, 128}, {128, 1}: 0.0384991ms, 0.338245% main:@392 = load[offset=2282624,end=2299008](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00063612ms, 0.00558881% main:@393 = gpu::code_object[code_object=13520,symbol_name=log_add_kernel,global=2048,local=256,](main:@391,main:@392) -> float_type, {32, 128}, {128, 1}: 0.0133373ms, 0.117179% main:@394 = load[offset=2376320,end=2407040](main:@1) -> float_type, {32, 240}, {240, 1}: 0.00061432ms, 0.00539728% 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, {32, 240}, {240, 1}: 0.022351ms, 0.196371% main:@396 = load[offset=2282624,end=2376320](main:@1) -> float_type, {32, 732}, {732, 1}: 0.00062808ms, 0.00551817% main:@397 = gpu::code_object[code_object=13560,symbol_name=noop_rsqrt_add_concat_noop_kernel,global=11712,local=256,](main:@121,main:@395,main:@334,main:@396) -> float_type, {32, 732}, {732, 1}: 0.0158275ms, 0.139057% main:@398 = hip::hip_copy_literal[id=main:@literal:118] -> float_type, {732, 256}, {256, 1}: 0.000795ms, 0.00698469% main:@399 = load[offset=995456,end=1028224](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00049928ms, 0.00438656% 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, {32, 256}, {256, 1}: 0.038293ms, 0.336434% main:@401 = load[offset=2282624,end=2424960](main:@1) -> float_type, {32, 1112}, {1112, 1}: 0.00061212ms, 0.00537795% main:@402 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=17792,local=256,](main:@63,main:@334,main:@401) -> float_type, {32, 1112}, {1112, 1}: 0.0161082ms, 0.141523% main:@403 = hip::hip_copy_literal[id=main:@literal:128] -> float_type, {1112, 128}, {128, 1}: 0.00071672ms, 0.00629694% main:@404 = load[offset=2424960,end=2441344](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00048336ms, 0.00424669% 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, {32, 128}, {128, 1}: 0.0253652ms, 0.222853% main:@406 = load[offset=1028224,end=1044608](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00062692ms, 0.00550798% main:@407 = gpu::code_object[code_object=13520,symbol_name=neg_add_kernel,global=2048,local=256,](main:@405,main:@406) -> float_type, {32, 128}, {128, 1}: 0.0132901ms, 0.116763% main:@408 = hip::hip_copy_literal[id=main:@literal:67] -> float_type, {128, 240}, {240, 1}: 0.00076152ms, 0.00669054% main:@409 = load[offset=2424960,end=2455680](main:@1) -> float_type, {32, 240}, {240, 1}: 0.0004596ms, 0.00403794% 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, {32, 240}, {240, 1}: 0.0215519ms, 0.18935% main:@411 = load[offset=2282624,end=2424960](main:@1) -> float_type, {32, 1112}, {1112, 1}: 0.00062084ms, 0.00545456% main:@412 = gpu::code_object[code_object=13552,symbol_name=noop_log_add_concat_noop_kernel,global=17792,local=256,](main:@63,main:@410,main:@334,main:@411) -> float_type, {32, 1112}, {1112, 1}: 0.016367ms, 0.143797% main:@413 = hip::hip_copy_literal[id=main:@literal:87] -> float_type, {1112, 256}, {256, 1}: 0.00074826ms, 0.00657404% main:@414 = load[offset=176640,end=209408](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00046838ms, 0.00411508% 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, {32, 256}, {256, 1}: 0.0316817ms, 0.278348% main:@416 = load[offset=2282624,end=2578560](main:@1) -> float_type, {32, 2312}, {2312, 1}: 0.00061344ms, 0.00538955% main:@417 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=36992,local=256,](main:@199,main:@334,main:@416) -> float_type, {32, 2312}, {2312, 1}: 0.0181541ms, 0.159498% main:@418 = hip::hip_copy_literal[id=main:@literal:53] -> float_type, {2312, 128}, {128, 1}: 0.00084216ms, 0.00739903% main:@419 = load[offset=209408,end=225792](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00045734ms, 0.00401809% 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, {32, 128}, {128, 1}: 0.0299363ms, 0.263013% main:@421 = load[offset=240128,end=256512](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00061242ms, 0.00538058% main:@422 = gpu::code_object[code_object=13520,symbol_name=rsqrt_add_kernel,global=2048,local=256,](main:@420,main:@421) -> float_type, {32, 128}, {128, 1}: 0.0131939ms, 0.115919% main:@423 = load[offset=209408,end=240128](main:@1) -> float_type, {32, 240}, {240, 1}: 0.00060448ms, 0.00531083% 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, {32, 240}, {240, 1}: 0.0217241ms, 0.190863% main:@425 = load[offset=2282624,end=2578560](main:@1) -> float_type, {32, 2312}, {2312, 1}: 0.00062144ms, 0.00545983% main:@426 = gpu::code_object[code_object=13552,symbol_name=noop_tanh_pow_concat_noop_kernel,global=36992,local=256,](main:@199,main:@424,main:@334,main:@425) -> float_type, {32, 2312}, {2312, 1}: 0.0183057ms, 0.16083% main:@427 = hip::hip_copy_literal[id=main:@literal:144] -> float_type, {2312, 256}, {256, 1}: 0.00079516ms, 0.0069861% main:@428 = load[offset=209408,end=242176](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00049068ms, 0.004311% 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, {32, 256}, {256, 1}: 0.0316663ms, 0.278213% main:@430 = load[offset=2017408,end=2303104](main:@1) -> float_type, {32, 2232}, {2232, 1}: 0.00061346ms, 0.00538972% main:@431 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=35712,local=256,](main:@140,main:@334,main:@430) -> float_type, {32, 2232}, {2232, 1}: 0.0184719ms, 0.16229% main:@432 = hip::hip_copy_literal[id=main:@literal:66] -> float_type, {2232, 128}, {128, 1}: 0.00080772ms, 0.00709645% main:@433 = load[offset=242176,end=258560](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00052408ms, 0.00460445% 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, {32, 128}, {128, 1}: 0.0292911ms, 0.257345% main:@435 = load[offset=1028224,end=1044608](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00061174ms, 0.00537461% main:@436 = gpu::code_object[code_object=13520,symbol_name=neg_div_kernel,global=2048,local=256,](main:@434,main:@435) -> float_type, {32, 128}, {128, 1}: 0.0132938ms, 0.116796% main:@437 = load[offset=242176,end=272896](main:@1) -> float_type, {32, 240}, {240, 1}: 0.00062388ms, 0.00548127% 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, {32, 240}, {240, 1}: 0.0206188ms, 0.181152% main:@439 = load[offset=2017408,end=2303104](main:@1) -> float_type, {32, 2232}, {2232, 1}: 0.0006306ms, 0.00554031% main:@440 = gpu::code_object[code_object=13560,symbol_name=noop_rsqrt_add_concat_noop_kernel,global=35712,local=256,](main:@140,main:@438,main:@334,main:@439) -> float_type, {32, 2232}, {2232, 1}: 0.018108ms, 0.159093% main:@441 = hip::hip_copy_literal[id=main:@literal:69] -> float_type, {2232, 256}, {256, 1}: 0.00079126ms, 0.00695183% main:@442 = load[offset=242176,end=274944](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00047192ms, 0.00414618% 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, {32, 256}, {256, 1}: 0.0301109ms, 0.264547% main:@444 = load[offset=1249408,end=1309824](main:@1) -> float_type, {32, 472}, {472, 1}: 0.0006225ms, 0.00546915% main:@445 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=7552,local=256,](main:@83,main:@334,main:@444) -> float_type, {32, 472}, {472, 1}: 0.0147937ms, 0.129974% main:@446 = hip::hip_copy_literal[id=main:@literal:119] -> float_type, {472, 128}, {128, 1}: 0.00072492ms, 0.00636898% main:@447 = load[offset=1309824,end=1326208](main:@1) -> float_type, {32, 128}, {128, 1}: 0.0004581ms, 0.00402476% 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, {32, 128}, {128, 1}: 0.0315142ms, 0.276877% main:@449 = load[offset=1028224,end=1044608](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00066074ms, 0.00580511% main:@450 = gpu::code_object[code_object=13520,symbol_name=log_add_kernel,global=2048,local=256,](main:@448,main:@449) -> float_type, {32, 128}, {128, 1}: 0.0134756ms, 0.118394% main:@451 = hip::hip_copy_literal[id=main:@literal:102] -> float_type, {128, 240}, {240, 1}: 0.00082386ms, 0.00723825% main:@452 = load[offset=1249408,end=1280128](main:@1) -> float_type, {32, 240}, {240, 1}: 0.0004828ms, 0.00424177% 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, {32, 240}, {240, 1}: 0.0202976ms, 0.17833% main:@454 = load[offset=1282176,end=1342592](main:@1) -> float_type, {32, 472}, {472, 1}: 0.00060814ms, 0.00534298% main:@455 = gpu::code_object[code_object=13552,symbol_name=noop_tanh_div_concat_noop_kernel,global=7552,local=256,](main:@83,main:@453,main:@334,main:@454) -> float_type, {32, 472}, {472, 1}: 0.0154482ms, 0.135725% main:@456 = hip::hip_copy_literal[id=main:@literal:80] -> float_type, {472, 256}, {256, 1}: 0.00076064ms, 0.00668281% main:@457 = load[offset=1249408,end=1282176](main:@1) -> float_type, {32, 256}, {256, 1}: 0.000499ms, 0.0043841% 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, {32, 256}, {256, 1}: 0.0310396ms, 0.272707% main:@459 = load[offset=1282176,end=1444992](main:@1) -> float_type, {32, 1272}, {1272, 1}: 0.00066438ms, 0.00583709% main:@460 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=20352,local=256,](main:@219,main:@334,main:@459) -> float_type, {32, 1272}, {1272, 1}: 0.0166469ms, 0.146256% main:@461 = hip::hip_copy_literal[id=main:@literal:59] -> float_type, {1272, 128}, {128, 1}: 0.00071034ms, 0.00624089% main:@462 = load[offset=1028224,end=1044608](main:@1) -> float_type, {32, 128}, {128, 1}: 0.0004669ms, 0.00410208% 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, {32, 128}, {128, 1}: 0.02611ms, 0.229397% main:@464 = load[offset=146944,end=163328](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00060202ms, 0.00528921% main:@465 = gpu::code_object[code_object=13520,symbol_name=rsqrt_add_kernel,global=2048,local=256,](main:@463,main:@464) -> float_type, {32, 128}, {128, 1}: 0.013295ms, 0.116807% main:@466 = hip::hip_copy_literal[id=main:@literal:39] -> float_type, {128, 240}, {240, 1}: 0.00078334ms, 0.00688225% main:@467 = load[offset=1282176,end=1312896](main:@1) -> float_type, {32, 240}, {240, 1}: 0.00045516ms, 0.00399893% 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, {32, 240}, {240, 1}: 0.0210462ms, 0.184907% main:@469 = load[offset=1314944,end=1477760](main:@1) -> float_type, {32, 1272}, {1272, 1}: 0.00060558ms, 0.00532049% main:@470 = gpu::code_object[code_object=13552,symbol_name=noop_log_sub_concat_noop_kernel,global=20352,local=256,](main:@219,main:@468,main:@334,main:@469) -> float_type, {32, 1272}, {1272, 1}: 0.0166907ms, 0.146641% main:@471 = hip::hip_copy_literal[id=main:@literal:40] -> float_type, {1272, 256}, {256, 1}: 0.00076504ms, 0.00672147% main:@472 = load[offset=1282176,end=1314944](main:@1) -> float_type, {32, 256}, {256, 1}: 0.0004843ms, 0.00425495% 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, {32, 256}, {256, 1}: 0.0305092ms, 0.268047% main:@474 = load[offset=1314944,end=1457280](main:@1) -> float_type, {32, 1112}, {1112, 1}: 0.00060822ms, 0.00534368% main:@475 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=17792,local=256,](main:@159,main:@334,main:@474) -> float_type, {32, 1112}, {1112, 1}: 0.0164578ms, 0.144595% main:@476 = hip::hip_copy_literal[id=main:@literal:54] -> float_type, {1112, 128}, {128, 1}: 0.00076808ms, 0.00674818% main:@477 = load[offset=1028224,end=1044608](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00046348ms, 0.00407203% 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, {32, 128}, {128, 1}: 0.0262221ms, 0.230382% main:@479 = load[offset=146944,end=163328](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00066222ms, 0.00581812% main:@480 = gpu::code_object[code_object=13520,symbol_name=log_pow_kernel,global=2048,local=256,](main:@478,main:@479) -> float_type, {32, 128}, {128, 1}: 0.0164952ms, 0.144923% main:@481 = load[offset=1457280,end=1488000](main:@1) -> float_type, {32, 240}, {240, 1}: 0.00060222ms, 0.00529097% 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, {32, 240}, {240, 1}: 0.0204147ms, 0.179359% main:@483 = load[offset=1314944,end=1457280](main:@1) -> float_type, {32, 1112}, {1112, 1}: 0.00060336ms, 0.00530099% main:@484 = gpu::code_object[code_object=13560,symbol_name=noop_rsqrt_add_concat_noop_kernel,global=17792,local=256,](main:@159,main:@482,main:@334,main:@483) -> float_type, {32, 1112}, {1112, 1}: 0.0165215ms, 0.145154% main:@485 = hip::hip_copy_literal[id=main:@literal:123] -> float_type, {1112, 256}, {256, 1}: 0.0007354ms, 0.00646106% main:@486 = load[offset=274944,end=307712](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00047438ms, 0.0041678% 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, {32, 256}, {256, 1}: 0.0300354ms, 0.263884% main:@488 = load[offset=307712,end=368128](main:@1) -> float_type, {32, 472}, {472, 1}: 0.00062878ms, 0.00552432% main:@489 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=7552,local=256,](main:@239,main:@334,main:@488) -> float_type, {32, 472}, {472, 1}: 0.0147836ms, 0.129886% main:@490 = hip::hip_copy_literal[id=main:@literal:99] -> float_type, {472, 128}, {128, 1}: 0.00075286ms, 0.00661446% main:@491 = load[offset=368128,end=384512](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00048142ms, 0.00422965% 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, {32, 128}, {128, 1}: 0.0315628ms, 0.277304% main:@493 = load[offset=146944,end=163328](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00061374ms, 0.00539218% main:@494 = gpu::code_object[code_object=13520,symbol_name=log_div_kernel,global=2048,local=256,](main:@492,main:@493) -> float_type, {32, 128}, {128, 1}: 0.01365ms, 0.119926% main:@495 = load[offset=307712,end=338432](main:@1) -> float_type, {32, 240}, {240, 1}: 0.00060546ms, 0.00531944% 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, {32, 240}, {240, 1}: 0.0206536ms, 0.181458% main:@497 = load[offset=340480,end=400896](main:@1) -> float_type, {32, 472}, {472, 1}: 0.00060878ms, 0.0053486% main:@498 = gpu::code_object[code_object=13560,symbol_name=noop_rsqrt_div_concat_noop_kernel,global=7552,local=256,](main:@239,main:@496,main:@334,main:@497) -> float_type, {32, 472}, {472, 1}: 0.0152448ms, 0.133937% main:@499 = hip::hip_copy_literal[id=main:@literal:63] -> float_type, {472, 256}, {256, 1}: 0.00071592ms, 0.00628991% main:@500 = load[offset=307712,end=340480](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00046272ms, 0.00406535% main:@501 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@498,main:@499,main:@500) -> float_type, {32, 256}, {256, 1}: 0.0307969ms, 0.270575% main:@502 = load[offset=1314944,end=1457280](main:@1) -> float_type, {32, 1112}, {1112, 1}: 0.0006127ms, 0.00538304% main:@503 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=17792,local=256,](main:@278,main:@334,main:@502) -> float_type, {32, 1112}, {1112, 1}: 0.0162738ms, 0.142978% main:@504 = hip::hip_copy_literal[id=main:@literal:135] -> float_type, {1112, 128}, {128, 1}: 0.00071794ms, 0.00630766% main:@505 = load[offset=340480,end=356864](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00048448ms, 0.00425653% 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, {32, 128}, {128, 1}: 0.026004ms, 0.228465% main:@507 = load[offset=146944,end=163328](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00061134ms, 0.0053711% main:@508 = gpu::code_object[code_object=13520,symbol_name=neg_pow_kernel,global=2048,local=256,](main:@506,main:@507) -> float_type, {32, 128}, {128, 1}: 0.0162753ms, 0.142991% main:@509 = hip::hip_copy_literal[id=main:@literal:74] -> float_type, {128, 240}, {240, 1}: 0.00080328ms, 0.00705744% main:@510 = load[offset=340480,end=371200](main:@1) -> float_type, {32, 240}, {240, 1}: 0.00045366ms, 0.00398576% 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, {32, 240}, {240, 1}: 0.020765ms, 0.182437% main:@512 = load[offset=1314944,end=1457280](main:@1) -> float_type, {32, 1112}, {1112, 1}: 0.00062174ms, 0.00546247% main:@513 = gpu::code_object[code_object=13552,symbol_name=noop_log_add_concat_noop_kernel,global=17792,local=256,](main:@278,main:@511,main:@334,main:@512) -> float_type, {32, 1112}, {1112, 1}: 0.0164711ms, 0.144712% main:@514 = hip::hip_copy_literal[id=main:@literal:55] -> float_type, {1112, 256}, {256, 1}: 0.0007728ms, 0.00678965% main:@515 = load[offset=340480,end=373248](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00050018ms, 0.00439447% 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, {32, 256}, {256, 1}: 0.0294981ms, 0.259163% main:@517 = load[offset=1314944,end=1482880](main:@1) -> float_type, {32, 1312}, {1312, 1}: 0.00062964ms, 0.00553188% main:@518 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=20992,local=256,](main:@259,main:@334,main:@517) -> float_type, {32, 1312}, {1312, 1}: 0.0166068ms, 0.145903% main:@519 = hip::hip_copy_literal[id=main:@literal:46] -> float_type, {1312, 128}, {128, 1}: 0.0008138ms, 0.00714986% main:@520 = load[offset=373248,end=389632](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00046602ms, 0.00409435% 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, {32, 128}, {128, 1}: 0.025285ms, 0.222148% main:@522 = load[offset=146944,end=163328](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00060934ms, 0.00535352% main:@523 = gpu::code_object[code_object=13520,symbol_name=tanh_add_kernel,global=2048,local=256,](main:@521,main:@522) -> float_type, {32, 128}, {128, 1}: 0.013897ms, 0.122096% main:@524 = load[offset=373248,end=403968](main:@1) -> float_type, {32, 240}, {240, 1}: 0.0006018ms, 0.00528728% 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, {32, 240}, {240, 1}: 0.020639ms, 0.181329% main:@526 = load[offset=1314944,end=1482880](main:@1) -> float_type, {32, 1312}, {1312, 1}: 0.00060682ms, 0.00533138% main:@527 = gpu::code_object[code_object=13552,symbol_name=noop_neg_div_concat_noop_kernel,global=20992,local=256,](main:@259,main:@525,main:@334,main:@526) -> float_type, {32, 1312}, {1312, 1}: 0.016893ms, 0.148418% main:@528 = hip::hip_copy_literal[id=main:@literal:60] -> float_type, {1312, 256}, {256, 1}: 0.00072496ms, 0.00636934% main:@529 = load[offset=1089152,end=1121920](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00046896ms, 0.00412018% 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, {32, 256}, {256, 1}: 0.0303872ms, 0.266975% main:@531 = load[offset=1314944,end=1375360](main:@1) -> float_type, {32, 472}, {472, 1}: 0.0006244ms, 0.00548584% main:@532 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=7552,local=256,](main:@44,main:@334,main:@531) -> float_type, {32, 472}, {472, 1}: 0.0147982ms, 0.130013% main:@533 = hip::hip_copy_literal[id=main:@literal:145] -> float_type, {472, 128}, {128, 1}: 0.0007537ms, 0.00662184% main:@534 = load[offset=373248,end=389632](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00045376ms, 0.00398663% 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, {32, 128}, {128, 1}: 0.031029ms, 0.272614% main:@536 = load[offset=146944,end=163328](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00062392ms, 0.00548162% main:@537 = gpu::code_object[code_object=13520,symbol_name=log_pow_kernel,global=2048,local=256,](main:@535,main:@536) -> float_type, {32, 128}, {128, 1}: 0.0165048ms, 0.145007% main:@538 = hip::hip_copy_literal[id=main:@literal:71] -> float_type, {128, 240}, {240, 1}: 0.00078186ms, 0.00686925% main:@539 = load[offset=373248,end=403968](main:@1) -> float_type, {32, 240}, {240, 1}: 0.0004578ms, 0.00402213% 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, {32, 240}, {240, 1}: 0.0203978ms, 0.179211% main:@541 = load[offset=1347712,end=1408128](main:@1) -> float_type, {32, 472}, {472, 1}: 0.00060846ms, 0.00534579% main:@542 = gpu::code_object[code_object=13552,symbol_name=noop_log_pow_concat_noop_kernel,global=7552,local=256,](main:@44,main:@540,main:@334,main:@541) -> float_type, {32, 472}, {472, 1}: 0.0170302ms, 0.149624% main:@543 = hip::hip_copy_literal[id=main:@literal:37] -> float_type, {472, 256}, {256, 1}: 0.00078238ms, 0.00687382% main:@544 = load[offset=1314944,end=1347712](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00049946ms, 0.00438814% 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, {32, 256}, {256, 1}: 0.0306475ms, 0.269262% main:@546 = load[offset=1347712,end=1408128](main:@1) -> float_type, {32, 472}, {472, 1}: 0.0006153ms, 0.00540589% main:@547 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=7552,local=256,](main:@24,main:@334,main:@546) -> float_type, {32, 472}, {472, 1}: 0.0147306ms, 0.12942% main:@548 = hip::hip_copy_literal[id=main:@literal:47] -> float_type, {472, 128}, {128, 1}: 0.00076406ms, 0.00671286% main:@549 = load[offset=146944,end=163328](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00050458ms, 0.00443313% 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, {32, 128}, {128, 1}: 0.0306189ms, 0.269011% main:@551 = load[offset=29696,end=46080](main:@1) -> float_type, {32, 128}, {128, 1}: 0.0006458ms, 0.00567385% main:@552 = gpu::code_object[code_object=13520,symbol_name=tanh_sub_kernel,global=2048,local=256,](main:@550,main:@551) -> float_type, {32, 128}, {128, 1}: 0.0138013ms, 0.121255% main:@553 = load[offset=373248,end=403968](main:@1) -> float_type, {32, 240}, {240, 1}: 0.00059738ms, 0.00524845% 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, {32, 240}, {240, 1}: 0.020337ms, 0.178677% main:@555 = load[offset=1347712,end=1408128](main:@1) -> float_type, {32, 472}, {472, 1}: 0.00061868ms, 0.00543558% main:@556 = gpu::code_object[code_object=13552,symbol_name=noop_tanh_div_concat_noop_kernel,global=7552,local=256,](main:@24,main:@554,main:@334,main:@555) -> float_type, {32, 472}, {472, 1}: 0.0154468ms, 0.135712% main:@557 = hip::hip_copy_literal[id=main:@literal:156] -> float_type, {472, 256}, {256, 1}: 0.00078354ms, 0.00688401% main:@558 = load[offset=0,end=32768](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00048888ms, 0.00429519% 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, {32, 256}, {256, 1}: 0.0308592ms, 0.271122% main:@560 = load[offset=1636480,end=1799296](main:@1) -> float_type, {32, 1272}, {1272, 1}: 0.00064016ms, 0.0056243% main:@561 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=20352,local=256,](main:@298,main:@334,main:@560) -> float_type, {32, 1272}, {1272, 1}: 0.0167203ms, 0.146901% main:@562 = hip::hip_copy_literal[id=main:@literal:136] -> float_type, {1272, 128}, {128, 1}: 0.00073416ms, 0.00645017% main:@563 = load[offset=146944,end=163328](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00049102ms, 0.00431399% 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, {32, 128}, {128, 1}: 0.025553ms, 0.224503% main:@565 = load[offset=32768,end=49152](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00061216ms, 0.0053783% main:@566 = gpu::code_object[code_object=13520,symbol_name=neg_pow_kernel,global=2048,local=256,](main:@564,main:@565) -> float_type, {32, 128}, {128, 1}: 0.0162702ms, 0.142946% main:@567 = hip::hip_copy_literal[id=main:@literal:78] -> float_type, {128, 240}, {240, 1}: 0.0007225ms, 0.00634772% main:@568 = load[offset=373248,end=403968](main:@1) -> float_type, {32, 240}, {240, 1}: 0.00049166ms, 0.00431961% 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, {32, 240}, {240, 1}: 0.0202526ms, 0.177934% main:@570 = load[offset=1636480,end=1799296](main:@1) -> float_type, {32, 1272}, {1272, 1}: 0.00061636ms, 0.0054152% main:@571 = gpu::code_object[code_object=13560,symbol_name=noop_rsqrt_div_concat_noop_kernel,global=20352,local=256,](main:@298,main:@569,main:@334,main:@570) -> float_type, {32, 1272}, {1272, 1}: 0.0171246ms, 0.150453% main:@572 = hip::hip_copy_literal[id=main:@literal:131] -> float_type, {1272, 256}, {256, 1}: 0.00079568ms, 0.00699067% main:@573 = load[offset=1347712,end=1380480](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00046432ms, 0.00407941% 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, {32, 256}, {256, 1}: 0.0305132ms, 0.268082% main:@575 = load[offset=1380480,end=1458816](main:@1) -> float_type, {32, 612}, {612, 1}: 0.00061428ms, 0.00539693% main:@576 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=9792,local=256,](main:@102,main:@334,main:@575) -> float_type, {32, 612}, {612, 1}: 0.0154264ms, 0.135533% main:@577 = hip::hip_copy_literal[id=main:@literal:125] -> float_type, {612, 128}, {128, 1}: 0.00078656ms, 0.00691054% main:@578 = load[offset=146944,end=163328](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00046248ms, 0.00406325% 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, {32, 128}, {128, 1}: 0.0346061ms, 0.304042% main:@580 = load[offset=32768,end=49152](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00061408ms, 0.00539517% main:@581 = gpu::code_object[code_object=13520,symbol_name=neg_pow_kernel,global=2048,local=256,](main:@579,main:@580) -> float_type, {32, 128}, {128, 1}: 0.016366ms, 0.143788% main:@582 = hip::hip_copy_literal[id=main:@literal:124] -> float_type, {128, 240}, {240, 1}: 0.00077268ms, 0.00678859% main:@583 = load[offset=373248,end=403968](main:@1) -> float_type, {32, 240}, {240, 1}: 0.00045284ms, 0.00397855% 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, {32, 240}, {240, 1}: 0.0203954ms, 0.179189% main:@585 = load[offset=1380480,end=1458816](main:@1) -> float_type, {32, 612}, {612, 1}: 0.00062446ms, 0.00548637% main:@586 = gpu::code_object[code_object=13560,symbol_name=noop_rsqrt_add_concat_noop_kernel,global=9792,local=256,](main:@102,main:@584,main:@334,main:@585) -> float_type, {32, 612}, {612, 1}: 0.0156449ms, 0.137452% main:@587 = hip::hip_copy_literal[id=main:@literal:103] -> float_type, {612, 256}, {256, 1}: 0.00070436ms, 0.00618835% main:@588 = load[offset=32768,end=65536](main:@1) -> float_type, {32, 256}, {256, 1}: 0.0004683ms, 0.00411438% 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, {32, 256}, {256, 1}: 0.0345826ms, 0.303835% main:@590 = load[offset=1380480,end=1451136](main:@1) -> float_type, {32, 552}, {552, 1}: 0.00061342ms, 0.00538937% main:@591 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=8832,local=256,](main:@179,main:@334,main:@590) -> float_type, {32, 552}, {552, 1}: 0.0152222ms, 0.133739% main:@592 = hip::hip_copy_literal[id=main:@literal:122] -> float_type, {552, 128}, {128, 1}: 0.0007172ms, 0.00630116% main:@593 = load[offset=65536,end=81920](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00050234ms, 0.00441345% 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, {32, 128}, {128, 1}: 0.0328077ms, 0.288241% main:@595 = load[offset=146944,end=163328](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00063372ms, 0.00556772% main:@596 = gpu::code_object[code_object=13520,symbol_name=neg_div_kernel,global=2048,local=256,](main:@594,main:@595) -> float_type, {32, 128}, {128, 1}: 0.0133542ms, 0.117327% main:@597 = load[offset=65536,end=96256](main:@1) -> float_type, {32, 240}, {240, 1}: 0.00059896ms, 0.00526233% 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, {32, 240}, {240, 1}: 0.020703ms, 0.181892% main:@599 = load[offset=1380480,end=1451136](main:@1) -> float_type, {32, 552}, {552, 1}: 0.00063014ms, 0.00553627% main:@600 = gpu::code_object[code_object=13560,symbol_name=noop_rsqrt_sub_concat_noop_kernel,global=8832,local=256,](main:@179,main:@598,main:@334,main:@599) -> float_type, {32, 552}, {552, 1}: 0.0153404ms, 0.134777% main:@601 = hip::hip_copy_literal[id=main:@literal:98] -> float_type, {552, 256}, {256, 1}: 0.00071324ms, 0.00626637% main:@602 = load[offset=65536,end=98304](main:@1) -> float_type, {32, 256}, {256, 1}: 0.0004666ms, 0.00409944% 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, {32, 256}, {256, 1}: 0.0331201ms, 0.290985% main:@604 = unsqueeze[axes={1},steps={}](main:@400) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.00084914ms, 0.00746035% main:@605 = unsqueeze[axes={1},steps={}](main:@415) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.00049924ms, 0.00438621% main:@606 = unsqueeze[axes={1},steps={}](main:@429) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.00050432ms, 0.00443084% main:@607 = unsqueeze[axes={1},steps={}](main:@443) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.00047796ms, 0.00419925% main:@608 = unsqueeze[axes={1},steps={}](main:@458) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.00049762ms, 0.00437198% main:@609 = unsqueeze[axes={1},steps={}](main:@473) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.00050114ms, 0.0044029% main:@610 = unsqueeze[axes={1},steps={}](main:@487) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.00048498ms, 0.00426093% main:@611 = unsqueeze[axes={1},steps={}](main:@501) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.00047892ms, 0.00420768% main:@612 = unsqueeze[axes={1},steps={}](main:@516) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.00046998ms, 0.00412914% main:@613 = unsqueeze[axes={1},steps={}](main:@530) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.00047154ms, 0.00414284% main:@614 = unsqueeze[axes={1},steps={}](main:@545) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.00047742ms, 0.00419451% main:@615 = unsqueeze[axes={1},steps={}](main:@559) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.00047582ms, 0.00418045% main:@616 = unsqueeze[axes={1},steps={}](main:@574) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.0004696ms, 0.0041258% main:@617 = unsqueeze[axes={1},steps={}](main:@589) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.00046698ms, 0.00410278% main:@618 = unsqueeze[axes={1},steps={}](main:@603) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.00048252ms, 0.00423931% main:@619 = load[offset=1380480,end=1872000](main:@1) -> float_type, {32, 15, 256}, {3840, 256, 1}: 0.00047516ms, 0.00417465% main:@620 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=4096,local=256,](main:@605,main:@612,main:@606,main:@607,main:@609,main:@616,main:@604,main:@613,main:@610,main:@617,main:@618,main:@614,main:@611,main:@608,main:@615,main:@619) -> float_type, {32, 15, 256}, {3840, 256, 1}: 0.025384ms, 0.223018% main:@621 = hip::hip_copy_literal[id=main:@literal:35] -> float_type, {256, 768}, {768, 1}: 0.0008123ms, 0.00713669% main:@622 = reshape_lazy[dims={480, 256}](main:@620) -> float_type, {480, 256}, {256, 1}: 0.00054954ms, 0.00482814% main:@623 = load[offset=4072192,end=5546752](main:@1) -> float_type, {480, 768}, {768, 1}: 0.00045192ms, 0.00397047% main:@624 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@622,main:@621,main:@623) -> float_type, {480, 768}, {768, 1}: 0.0346877ms, 0.304759% main:@625 = hip::hip_copy_literal[id=main:@literal:24] -> float_type, {130, 128}, {128, 1}: 0.00082354ms, 0.00723544% main:@626 = load[offset=0,end=16640](main:@1) -> float_type, {32, 130}, {130, 1}: 0.00047702ms, 0.00419099% 2CnSyUBzm0:0 = @param:2CnSyUBzm0:0 -> float_type, {32, 65}, {65, 1}: 0.00051746ms, 0.00454629% main:@628 = slice[axes={0, 1},starts={0, 53},ends={32, 57}](2CnSyUBzm0:0) -> float_type, {32, 4}, {65, 1}: 0.00071724ms, 0.00630151% 9YqhflbMKe:0 = @param:9YqhflbMKe:0 -> float_type, {32, 29}, {29, 1}: 0.00050478ms, 0.00443488% main:@630 = slice[axes={0, 1},starts={0, 21},ends={32, 29}](9YqhflbMKe:0) -> float_type, {32, 8}, {29, 1}: 0.0010676ms, 0.00937969% ZcWERoz0h6:0 = @param:ZcWERoz0h6:0 -> float_type, {32, 49}, {49, 1}: 0.0004994ms, 0.00438762% main:@632 = slice[axes={0, 1},starts={0, 37},ends={32, 41}](ZcWERoz0h6:0) -> float_type, {32, 4}, {49, 1}: 0.00065046ms, 0.0057148% ljyhc7HCLk:0 = @param:ljyhc7HCLk:0 -> float_type, {32, 21}, {21, 1}: 0.0005425ms, 0.00476628% main:@634 = slice[axes={0, 1},starts={0, 13},ends={32, 21}](ljyhc7HCLk:0) -> float_type, {32, 8}, {21, 1}: 0.00076242ms, 0.00669845% YbIVz8l1un:0 = @param:YbIVz8l1un:0 -> float_type, {32, 45}, {45, 1}: 0.00051562ms, 0.00453012% main:@636 = slice[axes={0, 1},starts={0, 29},ends={32, 37}](YbIVz8l1un:0) -> float_type, {32, 8}, {45, 1}: 0.000649ms, 0.00570197% FcmWLZhDgE:0 = @param:FcmWLZhDgE:0 -> float_type, {32, 57}, {57, 1}: 0.00051984ms, 0.0045672% main:@638 = slice[axes={0, 1},starts={0, 41},ends={32, 49}](FcmWLZhDgE:0) -> float_type, {32, 8}, {57, 1}: 0.00078522ms, 0.00689877% main:@639 = slice[axes={0, 1},starts={0, 53},ends={32, 61}](Qp3K1PHvte:0) -> float_type, {32, 8}, {69, 1}: 0.00060456ms, 0.00531153% bJsIAzhYc3:0 = @param:bJsIAzhYc3:0 -> float_type, {32, 37}, {37, 1}: 0.00044494ms, 0.00390914% main:@641 = slice[axes={0, 1},starts={0, 25},ends={32, 29}](bJsIAzhYc3:0) -> float_type, {32, 4}, {37, 1}: 0.00061554ms, 0.005408% main:@642 = slice[axes={0, 1},starts={0, 57},ends={32, 65}](ifP6Ceko5T:0) -> float_type, {32, 8}, {73, 1}: 0.00077496ms, 0.00680862% main:@643 = slice[axes={0, 1},starts={0, 217},ends={32, 233}](uFSOc5Chp2:0) -> float_type, {32, 16}, {481, 1}: 0.00065334ms, 0.0057401% main:@644 = slice[axes={0, 1},starts={0, 509},ends={32, 517}](lyiDxfzZg3:0) -> float_type, {32, 8}, {701, 1}: 0.00059004ms, 0.00518396% main:@645 = slice[axes={0, 1},starts={0, 197},ends={32, 237}](nv2tzWRf1I:0) -> float_type, {32, 40}, {301, 1}: 0.00059296ms, 0.00520961% isCn4mEOSt:0 = @param:isCn4mEOSt:0 -> float_type, {32, 49}, {49, 1}: 0.0005341ms, 0.00469248% main:@647 = slice[axes={0, 1},starts={0, 37},ends={32, 41}](isCn4mEOSt:0) -> float_type, {32, 4}, {49, 1}: 0.00072824ms, 0.00639815% w04Keh2FJu:0 = @param:w04Keh2FJu:0 -> float_type, {32, 13}, {13, 1}: 0.00050354ms, 0.00442399% main:@649 = slice[axes={0, 1},starts={0, 11},ends={32, 13}](w04Keh2FJu:0) -> float_type, {32, 2}, {13, 1}: 0.00072494ms, 0.00636916% main:@650 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=297,local=256,](main:@644,main:@645,main:@643,main:@642,main:@639,main:@628,main:@636,main:@647,main:@632,main:@649,main:@638,main:@641,main:@634,main:@630,main:@626) -> float_type, {32, 130}, {130, 1}: 0.0284578ms, 0.250024% main:@651 = load[offset=389248,end=405632](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00067986ms, 0.0059731% 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, {32, 128}, {128, 1}: 0.0225234ms, 0.197886% a5QhOKE1Yc:0 = @param:a5QhOKE1Yc:0 -> float_type, {32, 15, 9}, {135, 9, 1}: 0.00072388ms, 0.00635985% main:@654 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 15, 9}](a5QhOKE1Yc:0) -> float_type, {32, 15, 8}, {135, 9, 1}: 0.00069524ms, 0.00610822% main:@655 = load[offset=0,end=145920](main:@1) -> float_type, {32, 15, 76}, {1140, 76, 1}: 0.00048982ms, 0.00430345% W3KAtTdC92:0 = @param:W3KAtTdC92:0 -> float_type, {32, 15, 17}, {255, 17, 1}: 0.00054916ms, 0.0048248% main:@657 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 15, 17}](W3KAtTdC92:0) -> float_type, {32, 15, 16}, {255, 17, 1}: 0.0010134ms, 0.00890351% ZebaMGw6vA:0 = @param:ZebaMGw6vA:0 -> float_type, {32, 15, 9}, {135, 9, 1}: 0.00058096ms, 0.00510418% main:@659 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 15, 9}](ZebaMGw6vA:0) -> float_type, {32, 15, 8}, {135, 9, 1}: 0.00070076ms, 0.00615672% wcHWv4SNE2:0 = @param:wcHWv4SNE2:0 -> float_type, {32, 15, 9}, {135, 9, 1}: 0.00047976ms, 0.00421506% main:@661 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 15, 9}](wcHWv4SNE2:0) -> float_type, {32, 15, 8}, {135, 9, 1}: 0.00079222ms, 0.00696027% main:@662 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=6080,local=256,](main:@654,main:@98,main:@661,main:@657,main:@175,main:@659,main:@655) -> float_type, {32, 15, 76}, {1140, 76, 1}: 0.019738ms, 0.173414% main:@663 = hip::hip_copy_literal[id=main:@literal:26] -> float_type, {76, 64}, {64, 1}: 0.00075698ms, 0.00665066% main:@664 = reshape_lazy[dims={480, 76}](main:@662) -> float_type, {480, 76}, {76, 1}: 0.00054568ms, 0.00479422% main:@665 = load[offset=1573632,end=1696512](main:@1) -> float_type, {480, 64}, {64, 1}: 0.00048458ms, 0.00425741% 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, {480, 64}, {64, 1}: 0.0228248ms, 0.200534% 6DYAlXkjub:0 = @param:6DYAlXkjub:0 -> float_type, {32, 30, 17}, {510, 17, 1}: 0.00067816ms, 0.00595816% main:@668 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 30, 17}](6DYAlXkjub:0) -> float_type, {32, 30, 16}, {510, 17, 1}: 0.00067962ms, 0.00597099% 74jZTbEdLU:0 = @param:74jZTbEdLU:0 -> float_type, {32, 30, 9}, {270, 9, 1}: 0.00058002ms, 0.00509593% main:@670 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 30, 9}](74jZTbEdLU:0) -> float_type, {32, 30, 8}, {270, 9, 1}: 0.00077396ms, 0.00679984% Z9QLFGO2Su:0 = @param:Z9QLFGO2Su:0 -> float_type, {32, 30, 9}, {270, 9, 1}: 0.00057254ms, 0.00503021% main:@672 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 30, 9}](Z9QLFGO2Su:0) -> float_type, {32, 30, 8}, {270, 9, 1}: 0.00066408ms, 0.00583446% WixXdEln2I:0 = @param:WixXdEln2I:0 -> float_type, {32, 30, 17}, {510, 17, 1}: 0.00053166ms, 0.00467105% main:@674 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 30, 17}](WixXdEln2I:0) -> float_type, {32, 30, 16}, {510, 17, 1}: 0.00099852ms, 0.00877277% IYWQSCjynR:0 = @param:IYWQSCjynR:0 -> float_type, {32, 30, 9}, {270, 9, 1}: 0.00052142ms, 0.00458108% main:@676 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 30, 9}](IYWQSCjynR:0) -> float_type, {32, 30, 8}, {270, 9, 1}: 0.00065364ms, 0.00574273% VvUalZIQHK:0 = @param:VvUalZIQHK:0 -> float_type, {32, 30, 5}, {150, 5, 1}: 0.00047928ms, 0.00421085% main:@678 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 30, 5}](VvUalZIQHK:0) -> float_type, {32, 30, 4}, {150, 5, 1}: 0.0007725ms, 0.00678701% bX19AmYnBL:0 = @param:bX19AmYnBL:0 -> float_type, {32, 30, 9}, {270, 9, 1}: 0.00051456ms, 0.00452081% main:@680 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 30, 9}](bX19AmYnBL:0) -> float_type, {32, 30, 8}, {270, 9, 1}: 0.00065236ms, 0.00573149% main:@681 = load[offset=128128,end=389248](main:@1) -> float_type, {32, 30, 68}, {2040, 68, 1}: 0.00048116ms, 0.00422736% main:@682 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=9325,local=256,](main:@678,main:@674,main:@676,main:@668,main:@670,main:@680,main:@672,main:@681) -> float_type, {32, 30, 68}, {2040, 68, 1}: 0.0238722ms, 0.209736% main:@683 = hip::hip_copy_literal[id=main:@literal:25] -> float_type, {68, 64}, {64, 1}: 0.0007972ms, 0.00700402% main:@684 = load[offset=1696512,end=1942272](main:@1) -> float_type, {960, 64}, {64, 1}: 0.00046484ms, 0.00408398% main:@685 = reshape_lazy[dims={960, 68}](main:@682) -> float_type, {960, 68}, {68, 1}: 0.00052916ms, 0.00464908% main:@686 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@685,main:@683,main:@684) -> float_type, {960, 64}, {64, 1}: 0.0209678ms, 0.184218% yFeB2xg6sO:0 = @param:yFeB2xg6sO:0 -> float_type, {32, 15, 9}, {135, 9, 1}: 0.0006884ms, 0.00604813% main:@688 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 15, 9}](yFeB2xg6sO:0) -> float_type, {32, 15, 8}, {135, 9, 1}: 0.00069782ms, 0.00613089% T3LHlorDnM:0 = @param:T3LHlorDnM:0 -> float_type, {32, 15, 17}, {255, 17, 1}: 0.00050652ms, 0.00445017% main:@690 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 15, 17}](T3LHlorDnM:0) -> float_type, {32, 15, 16}, {255, 17, 1}: 0.00079474ms, 0.00698241% kw7Kzgp1JD:0 = @param:kw7Kzgp1JD:0 -> float_type, {32, 15, 21}, {315, 21, 1}: 0.00055206ms, 0.00485028% main:@692 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 15, 21}](kw7Kzgp1JD:0) -> float_type, {32, 15, 20}, {315, 21, 1}: 0.00088772ms, 0.00779931% hGtXM6ZJUv:0 = @param:hGtXM6ZJUv:0 -> float_type, {32, 15, 9}, {135, 9, 1}: 0.00048842ms, 0.00429115% main:@694 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 15, 9}](hGtXM6ZJUv:0) -> float_type, {32, 15, 8}, {135, 9, 1}: 0.00077158ms, 0.00677893% main:@695 = load[offset=1249408,end=1403008](main:@1) -> float_type, {32, 15, 80}, {1200, 80, 1}: 0.0004602ms, 0.00404321% main:@696 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=7680,local=256,](main:@694,main:@690,main:@117,main:@692,main:@688,main:@695) -> float_type, {32, 15, 80}, {1200, 80, 1}: 0.0183811ms, 0.161492% main:@697 = hip::hip_copy_literal[id=main:@literal:14] -> float_type, {80, 64}, {64, 1}: 0.00076436ms, 0.0067155% main:@698 = load[offset=1003648,end=1126528](main:@1) -> float_type, {480, 64}, {64, 1}: 0.00047836ms, 0.00420276% main:@699 = reshape_lazy[dims={480, 80}](main:@696) -> float_type, {480, 80}, {80, 1}: 0.00052788ms, 0.00463784% 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, {480, 64}, {64, 1}: 0.0197646ms, 0.173648% main:@701 = hip::hip_copy_literal[id=main:@literal:162] -> float_type, {256}, {1}: 0.00073646ms, 0.00647037% main:@702 = slice[axes={0},starts={96},ends={128}](main:@386) -> float_type, {32, 18, 64}, {1152, 64, 1}: 0.00067656ms, 0.0059441% main:@703 = slice[axes={0},starts={64},ends={96}](main:@386) -> float_type, {32, 18, 64}, {1152, 64, 1}: 0.00068276ms, 0.00599858% main:@704 = slice[axes={0},starts={32},ends={64}](main:@386) -> float_type, {32, 18, 64}, {1152, 64, 1}: 0.00085606ms, 0.00752115% main:@705 = slice[axes={0},starts={0},ends={32}](main:@386) -> float_type, {32, 18, 64}, {1152, 64, 1}: 0.00056452ms, 0.00495975% main:@706 = load[offset=1942272,end=2532096](main:@1) -> float_type, {32, 18, 256}, {4608, 256, 1}: 0.00048658ms, 0.00427498% main:@707 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=36864,local=256,](main:@705,main:@704,main:@703,main:@702,main:@706) -> float_type, {32, 18, 256}, {4608, 256, 1}: 0.0198047ms, 0.174% main:@708 = hip::hip_copy_literal[id=main:@literal:6] -> float_type, {32, 18, 256}, {4608, 256, 1}: 0.00077954ms, 0.00684886% main:@709 = hip::hip_copy_literal[id=main:@literal:163] -> float_type, {256}, {1}: 0.0006143ms, 0.0053971% main:@710 = multibroadcast[out_lens={32, 18, 256}](main:@709) -> float_type, {32, 18, 256}, {0, 0, 1}: 0.00066192ms, 0.00581548% main:@711 = multibroadcast[out_lens={32, 18, 256}](main:@701) -> float_type, {32, 18, 256}, {0, 0, 1}: 0.00067394ms, 0.00592109% main:@712 = load[offset=413824,end=1003648](main:@1) -> float_type, {32, 18, 256}, {4608, 256, 1}: 0.00047632ms, 0.00418484% 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=73728,local=128,](main:@707,main:@708,main:@711,main:@710,main:@712) -> float_type, {32, 18, 256}, {4608, 256, 1}: 0.0226368ms, 0.198882% main:@714 = hip::hip_copy_literal[id=main:@literal:146] -> float_type, {256, 512}, {512, 1}: 0.0007597ms, 0.00667455% main:@715 = hip::hip_copy_literal[id=main:@literal:154] -> float_type, {512, 256}, {256, 1}: 0.00063032ms, 0.00553785% main:@716 = hip::hip_copy_literal[id=main:@literal:164] -> float_type, {256, 512}, {512, 1}: 0.00065558ms, 0.00575978% 06AyS2QTzo:0 = @param:06AyS2QTzo:0 -> float_type, {32, 50, 9}, {450, 9, 1}: 0.00054826ms, 0.00481689% main:@718 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 50, 9}](06AyS2QTzo:0) -> float_type, {32, 50, 8}, {450, 9, 1}: 0.00110534ms, 0.00971127% Uc35C1q4oy:0 = @param:Uc35C1q4oy:0 -> float_type, {32, 50, 9}, {450, 9, 1}: 0.00052634ms, 0.00462431% main:@720 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 50, 9}](Uc35C1q4oy:0) -> float_type, {32, 50, 8}, {450, 9, 1}: 0.00066486ms, 0.00584131% IFQJaVMDS5:0 = @param:IFQJaVMDS5:0 -> float_type, {32, 50, 17}, {850, 17, 1}: 0.00052372ms, 0.00460129% main:@722 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 50, 17}](IFQJaVMDS5:0) -> float_type, {32, 50, 16}, {850, 17, 1}: 0.0008159ms, 0.00716831% r8QjI5SCed:0 = @param:r8QjI5SCed:0 -> float_type, {32, 50, 9}, {450, 9, 1}: 0.0004949ms, 0.00434808% main:@724 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 50, 9}](r8QjI5SCed:0) -> float_type, {32, 50, 8}, {450, 9, 1}: 0.00063268ms, 0.00555858% main:@725 = load[offset=1942272,end=2428672](main:@1) -> float_type, {32, 50, 76}, {3800, 76, 1}: 0.00046856ms, 0.00411666% main:@726 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=20266,local=256,](main:@254,main:@718,main:@722,main:@724,main:@155,main:@720,main:@725) -> float_type, {32, 50, 76}, {3800, 76, 1}: 0.0242768ms, 0.21329% main:@727 = hip::hip_copy_literal[id=main:@literal:7] -> float_type, {76, 128}, {128, 1}: 0.00074132ms, 0.00651307% main:@728 = reshape_lazy[dims={1600, 76}](main:@726) -> float_type, {1600, 76}, {76, 1}: 0.00053074ms, 0.00466296% main:@729 = load[offset=5546752,end=6365952](main:@1) -> float_type, {1600, 128}, {128, 1}: 0.00045918ms, 0.00403425% main:@730 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@728,main:@727,main:@729) -> float_type, {1600, 128}, {128, 1}: 0.0239948ms, 0.210813% 0pnqc4Em1y:0 = @param:0pnqc4Em1y:0 -> float_type, {32, 50, 9}, {450, 9, 1}: 0.00068266ms, 0.0059977% main:@732 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 50, 9}](0pnqc4Em1y:0) -> float_type, {32, 50, 8}, {450, 9, 1}: 0.0007218ms, 0.00634157% hKxguEQn94:0 = @param:hKxguEQn94:0 -> float_type, {32, 50, 9}, {450, 9, 1}: 0.00053852ms, 0.00473132% main:@734 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 50, 9}](hKxguEQn94:0) -> float_type, {32, 50, 8}, {450, 9, 1}: 0.00081168ms, 0.00713124% D0gn7y6swk:0 = @param:D0gn7y6swk:0 -> float_type, {32, 50, 17}, {850, 17, 1}: 0.00051178ms, 0.00449638% main:@736 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 50, 17}](D0gn7y6swk:0) -> float_type, {32, 50, 16}, {850, 17, 1}: 0.00088922ms, 0.00781249% SKG7aiAo3m:0 = @param:SKG7aiAo3m:0 -> float_type, {32, 50, 9}, {450, 9, 1}: 0.00054104ms, 0.00475346% main:@738 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 50, 9}](SKG7aiAo3m:0) -> float_type, {32, 50, 8}, {450, 9, 1}: 0.00078126ms, 0.00686398% UpWkNPIeJO:0 = @param:UpWkNPIeJO:0 -> float_type, {32, 50, 17}, {850, 17, 1}: 0.00050764ms, 0.00446001% main:@740 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 50, 17}](UpWkNPIeJO:0) -> float_type, {32, 50, 16}, {850, 17, 1}: 0.00064162ms, 0.00563713% nEuw2U7mFR:0 = @param:nEuw2U7mFR:0 -> float_type, {32, 50, 41}, {2050, 41, 1}: 0.00057162ms, 0.00502213% main:@742 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 50, 41}](nEuw2U7mFR:0) -> float_type, {32, 50, 40}, {2050, 41, 1}: 0.00079458ms, 0.006981% main:@743 = load[offset=6365952,end=6980352](main:@1) -> float_type, {32, 50, 96}, {4800, 96, 1}: 0.00045728ms, 0.00401756% main:@744 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=25600,local=256,](main:@742,main:@734,main:@738,main:@736,main:@740,main:@732,main:@743) -> float_type, {32, 50, 96}, {4800, 96, 1}: 0.0244118ms, 0.214476% main:@745 = hip::hip_copy_literal[id=main:@literal:11] -> float_type, {96, 256}, {256, 1}: 0.00080464ms, 0.00706939% main:@746 = load[offset=2433792,end=4072192](main:@1) -> float_type, {1600, 256}, {256, 1}: 0.00049554ms, 0.0043537% main:@747 = reshape_lazy[dims={1600, 96}](main:@744) -> float_type, {1600, 96}, {96, 1}: 0.00049656ms, 0.00436266% main:@748 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@747,main:@745,main:@746) -> float_type, {1600, 256}, {256, 1}: 0.0260647ms, 0.228999% m2YscaoDjP:0 = @param:m2YscaoDjP:0 -> float_type, {32, 17}, {17, 1}: 0.00067532ms, 0.00593321% main:@750 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](m2YscaoDjP:0) -> float_type, {32, 4}, {17, 1}: 0.00075162ms, 0.00660356% 09zp4EDXTh:0 = @param:09zp4EDXTh:0 -> float_type, {32, 25}, {25, 1}: 0.00055972ms, 0.00491757% main:@752 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](09zp4EDXTh:0) -> float_type, {32, 8}, {25, 1}: 0.00075858ms, 0.00666471% HZ0l3MmcPY:0 = @param:HZ0l3MmcPY:0 -> float_type, {32, 17}, {17, 1}: 0.00053286ms, 0.00468159% main:@754 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](HZ0l3MmcPY:0) -> float_type, {32, 4}, {17, 1}: 0.00088624ms, 0.00778631% o74vNAnXxy:0 = @param:o74vNAnXxy:0 -> float_type, {32, 17}, {17, 1}: 0.00052154ms, 0.00458213% main:@756 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](o74vNAnXxy:0) -> float_type, {32, 4}, {17, 1}: 0.00074722ms, 0.00656491% 1PJsBhDaCK:0 = @param:1PJsBhDaCK:0 -> float_type, {32, 25}, {25, 1}: 0.00058104ms, 0.00510489% main:@758 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](1PJsBhDaCK:0) -> float_type, {32, 8}, {25, 1}: 0.0006213ms, 0.0054586% 1WNvHElUG7:0 = @param:1WNvHElUG7:0 -> float_type, {32, 25}, {25, 1}: 0.00049914ms, 0.00438533% main:@760 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](1WNvHElUG7:0) -> float_type, {32, 8}, {25, 1}: 0.00081584ms, 0.00716779% 1u5yIsHTXE:0 = @param:1u5yIsHTXE:0 -> float_type, {32, 17}, {17, 1}: 0.00052948ms, 0.00465189% main:@762 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](1u5yIsHTXE:0) -> float_type, {32, 4}, {17, 1}: 0.00062088ms, 0.00545491% 2lVXMZgJfO:0 = @param:2lVXMZgJfO:0 -> float_type, {32, 25}, {25, 1}: 0.00051804ms, 0.00455138% main:@764 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](2lVXMZgJfO:0) -> float_type, {32, 8}, {25, 1}: 0.00077062ms, 0.00677049% 3kdi0vsSZG:0 = @param:3kdi0vsSZG:0 -> float_type, {32, 25}, {25, 1}: 0.00051386ms, 0.00451466% main:@766 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](3kdi0vsSZG:0) -> float_type, {32, 8}, {25, 1}: 0.00064244ms, 0.00564433% Tg4u0GfnFb:0 = @param:Tg4u0GfnFb:0 -> float_type, {32, 25}, {25, 1}: 0.00046684ms, 0.00410155% main:@768 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](Tg4u0GfnFb:0) -> float_type, {32, 8}, {25, 1}: 0.00072332ms, 0.00635493% 5JHUPChra8:0 = @param:5JHUPChra8:0 -> float_type, {32, 17}, {17, 1}: 0.0005189ms, 0.00455894% main:@770 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](5JHUPChra8:0) -> float_type, {32, 2}, {17, 1}: 0.00095392ms, 0.00838093% xymEBUMS25:0 = @param:xymEBUMS25:0 -> float_type, {32, 13}, {13, 1}: 0.00047702ms, 0.00419099% main:@772 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](xymEBUMS25:0) -> float_type, {32, 2}, {13, 1}: 0.00071794ms, 0.00630766% zZHaVOckpw:0 = @param:zZHaVOckpw:0 -> float_type, {32, 13}, {13, 1}: 0.00052548ms, 0.00461675% main:@774 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](zZHaVOckpw:0) -> float_type, {32, 2}, {13, 1}: 0.0007029ms, 0.00617552% pH5srbufJ1:0 = @param:pH5srbufJ1:0 -> float_type, {32, 17}, {17, 1}: 0.00049586ms, 0.00435651% main:@776 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](pH5srbufJ1:0) -> float_type, {32, 4}, {17, 1}: 0.00069538ms, 0.00610945% msiPzSBKMj:0 = @param:msiPzSBKMj:0 -> float_type, {32, 17}, {17, 1}: 0.00048478ms, 0.00425917% main:@778 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](msiPzSBKMj:0) -> float_type, {32, 4}, {17, 1}: 0.00068386ms, 0.00600824% 92MwiALCHX:0 = @param:92MwiALCHX:0 -> float_type, {32, 25}, {25, 1}: 0.0005127ms, 0.00450447% main:@780 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](92MwiALCHX:0) -> float_type, {32, 8}, {25, 1}: 0.00066842ms, 0.00587259% ujOA5LQW7M:0 = @param:ujOA5LQW7M:0 -> float_type, {32, 13}, {13, 1}: 0.00050096ms, 0.00440132% main:@782 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](ujOA5LQW7M:0) -> float_type, {32, 2}, {13, 1}: 0.00065288ms, 0.00573606% AVLcBpWeuo:0 = @param:AVLcBpWeuo:0 -> float_type, {32, 21}, {21, 1}: 0.00055868ms, 0.00490844% main:@784 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](AVLcBpWeuo:0) -> float_type, {32, 4}, {21, 1}: 0.00069842ms, 0.00613616% main:@785 = slice[axes={0, 1},starts={0, 17},ends={32, 21}](ifP6Ceko5T:0) -> float_type, {32, 4}, {73, 1}: 0.0006328ms, 0.00555964% RpITWntAji:0 = @param:RpITWntAji:0 -> float_type, {32, 17}, {17, 1}: 0.00049704ms, 0.00436688% main:@787 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](RpITWntAji:0) -> float_type, {32, 8}, {17, 1}: 0.0006767ms, 0.00594533% IztrlRfWQU:0 = @param:IztrlRfWQU:0 -> float_type, {32, 13}, {13, 1}: 0.00053262ms, 0.00467948% main:@789 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](IztrlRfWQU:0) -> float_type, {32, 2}, {13, 1}: 0.00067022ms, 0.0058884% JSWhLK4Agx:0 = @param:JSWhLK4Agx:0 -> float_type, {32, 25}, {25, 1}: 0.0004868ms, 0.00427692% main:@791 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](JSWhLK4Agx:0) -> float_type, {32, 8}, {25, 1}: 0.00094618ms, 0.00831293% KN13IWXZSu:0 = @param:KN13IWXZSu:0 -> float_type, {32, 13}, {13, 1}: 0.00050824ms, 0.00446528% main:@793 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](KN13IWXZSu:0) -> float_type, {32, 2}, {13, 1}: 0.00066648ms, 0.00585554% NXOSdQVPs0:0 = @param:NXOSdQVPs0:0 -> float_type, {32, 25}, {25, 1}: 0.0005796ms, 0.00509224% main:@795 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](NXOSdQVPs0:0) -> float_type, {32, 8}, {25, 1}: 0.0006345ms, 0.00557457% ry2xad1vc9:0 = @param:ry2xad1vc9:0 -> float_type, {32, 25}, {25, 1}: 0.00044796ms, 0.00393568% main:@797 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](ry2xad1vc9:0) -> float_type, {32, 8}, {25, 1}: 0.00064572ms, 0.00567315% main:@798 = slice[axes={0, 1},starts={0, 17},ends={32, 25}](Qp3K1PHvte:0) -> float_type, {32, 8}, {69, 1}: 0.00069472ms, 0.00610365% RHo9Jz6KpI:0 = @param:RHo9Jz6KpI:0 -> float_type, {32, 21}, {21, 1}: 0.00048266ms, 0.00424054% main:@800 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](RHo9Jz6KpI:0) -> float_type, {32, 4}, {21, 1}: 0.00067478ms, 0.00592847% S5D6WGIXMd:0 = @param:S5D6WGIXMd:0 -> float_type, {32, 17}, {17, 1}: 0.00063808ms, 0.00560603% main:@802 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](S5D6WGIXMd:0) -> float_type, {32, 4}, {17, 1}: 0.00067656ms, 0.0059441% TidKGrnobI:0 = @param:TidKGrnobI:0 -> float_type, {32, 13}, {13, 1}: 0.00049452ms, 0.00434474% main:@804 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](TidKGrnobI:0) -> float_type, {32, 2}, {13, 1}: 0.00067244ms, 0.00590791% XxirNgB98u:0 = @param:XxirNgB98u:0 -> float_type, {32, 13}, {13, 1}: 0.00049164ms, 0.00431944% main:@806 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](XxirNgB98u:0) -> float_type, {32, 2}, {13, 1}: 0.00067634ms, 0.00594217% main:@807 = load[offset=24576,end=49408](main:@1) -> float_type, {32, 194}, {194, 1}: 0.00047076ms, 0.00413599% main:@808 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](YbIVz8l1un:0) -> float_type, {32, 8}, {45, 1}: 0.0006546ms, 0.00575117% main:@809 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](bJsIAzhYc3:0) -> float_type, {32, 8}, {37, 1}: 0.00067422ms, 0.00592355% tWcngAq7hV:0 = @param:tWcngAq7hV:0 -> float_type, {32, 13}, {13, 1}: 0.00047548ms, 0.00417746% main:@811 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](tWcngAq7hV:0) -> float_type, {32, 2}, {13, 1}: 0.00069634ms, 0.00611789% nPUSZFm74Y:0 = @param:nPUSZFm74Y:0 -> float_type, {32, 13}, {13, 1}: 0.00054924ms, 0.0048255% main:@813 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](nPUSZFm74Y:0) -> float_type, {32, 2}, {13, 1}: 0.00063306ms, 0.00556192% xT9FhjM1Uu:0 = @param:xT9FhjM1Uu:0 -> float_type, {32, 17}, {17, 1}: 0.00049676ms, 0.00436442% main:@815 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](xT9FhjM1Uu:0) -> float_type, {32, 4}, {17, 1}: 0.00064492ms, 0.00566612% main:@816 = slice[axes={0, 1},starts={0, 161},ends={32, 169}](uFSOc5Chp2:0) -> float_type, {32, 8}, {481, 1}: 0.00072404ms, 0.00636125% qd1komWTn4:0 = @param:qd1komWTn4:0 -> float_type, {32, 25}, {25, 1}: 0.00045538ms, 0.00400087% main:@818 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](qd1komWTn4:0) -> float_type, {32, 8}, {25, 1}: 0.00066442ms, 0.00583745% main:@819 = slice[axes={0, 1},starts={0, 161},ends={32, 163}](nv2tzWRf1I:0) -> float_type, {32, 2}, {301, 1}: 0.00063802ms, 0.0056055% main:@820 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=163,local=256,](main:@770,main:@806,main:@811,main:@772,main:@789,main:@793,main:@813,main:@782,main:@774,main:@804,main:@819,main:@754,main:@776,main:@778,main:@762,main:@800,main:@802,main:@750,main:@784,main:@756,main:@815,main:@785,main:@752,main:@758,main:@798,main:@791,main:@818,main:@797,main:@764,main:@780,main:@768,main:@760,main:@795,main:@809,main:@766,main:@808,main:@787,main:@816,main:@807) -> float_type, {32, 194}, {194, 1}: 0.050423ms, 0.443006% main:@821 = hip::hip_copy_literal[id=main:@literal:143] -> float_type, {194, 192}, {192, 1}: 0.0007865ms, 0.00691001% main:@822 = load[offset=0,end=24576](main:@1) -> float_type, {32, 192}, {192, 1}: 0.00049494ms, 0.00434843% main:@823 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@820,main:@821,main:@822) -> float_type, {32, 192}, {192, 1}: 0.0247295ms, 0.217268% main:@824 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](isCn4mEOSt:0) -> float_type, {32, 16}, {49, 1}: 0.00085152ms, 0.00748126% main:@825 = unsqueeze[axes={1},steps={}](main:@824) -> float_type, {32, 1, 16}, {49, 16, 1}: 0.00083344ms, 0.00732242% 6HSQqXdJzb:0 = @param:6HSQqXdJzb:0 -> float_type, {32, 45}, {45, 1}: 0.00055662ms, 0.00489034% main:@827 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](6HSQqXdJzb:0) -> float_type, {32, 16}, {45, 1}: 0.00079626ms, 0.00699576% main:@828 = unsqueeze[axes={1},steps={}](main:@827) -> float_type, {32, 1, 16}, {45, 16, 1}: 0.00057444ms, 0.0050469% main:@829 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](2CnSyUBzm0:0) -> float_type, {32, 16}, {65, 1}: 0.0006244ms, 0.00548584% main:@830 = unsqueeze[axes={1},steps={}](main:@829) -> float_type, {32, 1, 16}, {65, 16, 1}: 0.00057332ms, 0.00503706% ahAlEtBvzo:0 = @param:ahAlEtBvzo:0 -> float_type, {32, 41}, {41, 1}: 0.00054576ms, 0.00479492% main:@832 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](ahAlEtBvzo:0) -> float_type, {32, 16}, {41, 1}: 0.00073698ms, 0.00647494% main:@833 = unsqueeze[axes={1},steps={}](main:@832) -> float_type, {32, 1, 16}, {41, 16, 1}: 0.00055964ms, 0.00491687% Dy3rwVmn1A:0 = @param:Dy3rwVmn1A:0 -> float_type, {32, 41}, {41, 1}: 0.00051222ms, 0.00450025% main:@835 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](Dy3rwVmn1A:0) -> float_type, {32, 16}, {41, 1}: 0.00063008ms, 0.00553574% main:@836 = unsqueeze[axes={1},steps={}](main:@835) -> float_type, {32, 1, 16}, {41, 16, 1}: 0.00057306ms, 0.00503478% main:@837 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](FcmWLZhDgE:0) -> float_type, {32, 16}, {57, 1}: 0.00075922ms, 0.00667034% main:@838 = unsqueeze[axes={1},steps={}](main:@837) -> float_type, {32, 1, 16}, {57, 16, 1}: 0.00071428ms, 0.0062755% 1aXQU27NrD:0 = @param:1aXQU27NrD:0 -> float_type, {32, 49}, {49, 1}: 0.00046442ms, 0.00408029% main:@840 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](1aXQU27NrD:0) -> float_type, {32, 16}, {49, 1}: 0.00063464ms, 0.0055758% main:@841 = unsqueeze[axes={1},steps={}](main:@840) -> float_type, {32, 1, 16}, {49, 16, 1}: 0.00054804ms, 0.00481496% X29iyIeu8L:0 = @param:X29iyIeu8L:0 -> float_type, {32, 45}, {45, 1}: 0.00055284ms, 0.00485713% main:@843 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](X29iyIeu8L:0) -> float_type, {32, 16}, {45, 1}: 0.0007486ms, 0.00657703% main:@844 = unsqueeze[axes={1},steps={}](main:@843) -> float_type, {32, 1, 16}, {45, 16, 1}: 0.0005541ms, 0.0048682% main:@845 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](ZcWERoz0h6:0) -> float_type, {32, 16}, {49, 1}: 0.00058868ms, 0.00517201% main:@846 = unsqueeze[axes={1},steps={}](main:@845) -> float_type, {32, 1, 16}, {49, 16, 1}: 0.0005743ms, 0.00504567% mgnGNH4Mz3:0 = @param:mgnGNH4Mz3:0 -> float_type, {32, 49}, {49, 1}: 0.00058526ms, 0.00514196% main:@848 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](mgnGNH4Mz3:0) -> float_type, {32, 16}, {49, 1}: 0.00077116ms, 0.00677524% main:@849 = unsqueeze[axes={1},steps={}](main:@848) -> float_type, {32, 1, 16}, {49, 16, 1}: 0.00087008ms, 0.00764433% zIoqYe9cRb:0 = @param:zIoqYe9cRb:0 -> float_type, {32, 45}, {45, 1}: 0.00054442ms, 0.00478315% main:@851 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](zIoqYe9cRb:0) -> float_type, {32, 16}, {45, 1}: 0.00059398ms, 0.00521858% main:@852 = unsqueeze[axes={1},steps={}](main:@851) -> float_type, {32, 1, 16}, {45, 16, 1}: 0.00056294ms, 0.00494586% dAFYcGrfEl:0 = @param:dAFYcGrfEl:0 -> float_type, {32, 45}, {45, 1}: 0.00046088ms, 0.00404919% main:@854 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](dAFYcGrfEl:0) -> float_type, {32, 16}, {45, 1}: 0.0007479ms, 0.00657088% main:@855 = unsqueeze[axes={1},steps={}](main:@854) -> float_type, {32, 1, 16}, {45, 16, 1}: 0.00057322ms, 0.00503618% pYw7fMVoQg:0 = @param:pYw7fMVoQg:0 -> float_type, {32, 41}, {41, 1}: 0.00058138ms, 0.00510787% main:@857 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](pYw7fMVoQg:0) -> float_type, {32, 16}, {41, 1}: 0.00074146ms, 0.0065143% main:@858 = unsqueeze[axes={1},steps={}](main:@857) -> float_type, {32, 1, 16}, {41, 16, 1}: 0.00056568ms, 0.00496994% main:@859 = load[offset=24576,end=75776](main:@1) -> float_type, {32, 25, 16}, {400, 16, 1}: 0.000489ms, 0.00429624% main:@860 = reshape_lazy[dims={-1, 12, 16}](main:@823) -> float_type, {32, 12, 16}, {192, 16, 1}: 0.00049332ms, 0.0043342% main:@861 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=914,local=256,](main:@830,main:@849,main:@841,main:@828,main:@852,main:@855,main:@844,main:@846,main:@833,main:@838,main:@858,main:@836,main:@825,main:@860,main:@859) -> float_type, {32, 25, 16}, {400, 16, 1}: 0.0289265ms, 0.254141% main:@862 = hip::hip_copy_literal[id=main:@literal:115] -> float_type, {25, 5}, {5, 1}: 0.00081198ms, 0.00713387% main:@863 = transpose[permutation={0, 2, 1}](main:@861) -> float_type, {32, 16, 25}, {400, 1, 16}: 0.00072494ms, 0.00636916% main:@864 = load[offset=75776,end=126976](main:@1) -> float_type, {32, 16, 25}, {400, 25, 1}: 0.0005095ms, 0.00447635% main:@865 = gpu::code_object[code_object=13528,symbol_name=contiguous_kernel,global=12800,local=256,](main:@863,main:@864) -> float_type, {32, 16, 25}, {400, 25, 1}: 0.0150464ms, 0.132194% main:@866 = load[offset=0,end=10240](main:@1) -> float_type, {512, 5}, {5, 1}: 0.0006071ms, 0.00533384% main:@867 = reshape_lazy[dims={512, 25}](main:@865) -> float_type, {512, 25}, {25, 1}: 0.00050698ms, 0.00445421% main:@868 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@867,main:@862,main:@866) -> float_type, {512, 5}, {5, 1}: 0.0211957ms, 0.18622% main:@869 = hip::hip_copy_literal[id=main:@literal:75] -> float_type, {160, 160}, {160, 1}: 0.00079698ms, 0.00700209% 4M92ryD8ux:0 = @param:4M92ryD8ux:0 -> float_type, {32, 9}, {9, 1}: 0.00059836ms, 0.00525706% main:@871 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](4M92ryD8ux:0) -> float_type, {32, 4}, {9, 1}: 0.00069666ms, 0.0061207% 4gFa5o2YsJ:0 = @param:4gFa5o2YsJ:0 -> float_type, {32, 9}, {9, 1}: 0.00053064ms, 0.00466208% main:@873 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](4gFa5o2YsJ:0) -> float_type, {32, 4}, {9, 1}: 0.00074892ms, 0.00657984% 6XMrIsBjLY:0 = @param:6XMrIsBjLY:0 -> float_type, {32, 13}, {13, 1}: 0.00049896ms, 0.00438375% main:@875 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](6XMrIsBjLY:0) -> float_type, {32, 2}, {13, 1}: 0.00064364ms, 0.00565488% 72zWmh93Bo:0 = @param:72zWmh93Bo:0 -> float_type, {32, 17}, {17, 1}: 0.00052998ms, 0.00465629% main:@877 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](72zWmh93Bo:0) -> float_type, {32, 4}, {17, 1}: 0.00074864ms, 0.00657738% JjyOIxALzM:0 = @param:JjyOIxALzM:0 -> float_type, {32, 5}, {5, 1}: 0.00058358ms, 0.0051272% main:@879 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](JjyOIxALzM:0) -> float_type, {32, 2}, {5, 1}: 0.0006568ms, 0.0057705% 8vM5p4RzXd:0 = @param:8vM5p4RzXd:0 -> float_type, {32, 13}, {13, 1}: 0.0006167ms, 0.00541819% main:@881 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](8vM5p4RzXd:0) -> float_type, {32, 2}, {13, 1}: 0.00072694ms, 0.00638673% AB6lekmMxa:0 = @param:AB6lekmMxa:0 -> float_type, {32, 25}, {25, 1}: 0.00046148ms, 0.00405446% main:@883 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](AB6lekmMxa:0) -> float_type, {32, 8}, {25, 1}: 0.00060292ms, 0.00529712% 9dgeU4bKka:0 = @param:9dgeU4bKka:0 -> float_type, {32, 17}, {17, 1}: 0.00043836ms, 0.00385133% main:@885 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](9dgeU4bKka:0) -> float_type, {32, 4}, {17, 1}: 0.00105444ms, 0.00926407% EZBIAqn6LS:0 = @param:EZBIAqn6LS:0 -> float_type, {32, 13}, {13, 1}: 0.00052706ms, 0.00463063% main:@887 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](EZBIAqn6LS:0) -> float_type, {32, 2}, {13, 1}: 0.0006302ms, 0.0055368% BYyjkPCX0m:0 = @param:BYyjkPCX0m:0 -> float_type, {32, 13}, {13, 1}: 0.00047812ms, 0.00420066% main:@889 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](BYyjkPCX0m:0) -> float_type, {32, 2}, {13, 1}: 0.00071464ms, 0.00627867% EKpP4xMXu8:0 = @param:EKpP4xMXu8:0 -> float_type, {32, 17}, {17, 1}: 0.00048884ms, 0.00429484% main:@891 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](EKpP4xMXu8:0) -> float_type, {32, 4}, {17, 1}: 0.00070074ms, 0.00615654% EUciqFuptg:0 = @param:EUciqFuptg:0 -> float_type, {32, 17}, {17, 1}: 0.00059226ms, 0.00520346% main:@893 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](EUciqFuptg:0) -> float_type, {32, 4}, {17, 1}: 0.00067564ms, 0.00593602% nyLxrsTmOB:0 = @param:nyLxrsTmOB:0 -> float_type, {32, 17}, {17, 1}: 0.00049542ms, 0.00435265% main:@895 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](nyLxrsTmOB:0) -> float_type, {32, 8}, {17, 1}: 0.00066492ms, 0.00584184% Hnjq487SIY:0 = @param:Hnjq487SIY:0 -> float_type, {32, 13}, {13, 1}: 0.00054282ms, 0.00476909% main:@897 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](Hnjq487SIY:0) -> float_type, {32, 2}, {13, 1}: 0.00068108ms, 0.00598382% IRem4MiFDP:0 = @param:IRem4MiFDP:0 -> float_type, {32, 5}, {5, 1}: 0.0004962ms, 0.0043595% main:@899 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](IRem4MiFDP:0) -> float_type, {32, 2}, {5, 1}: 0.0006628ms, 0.00582321% KasFvpq4IG:0 = @param:KasFvpq4IG:0 -> float_type, {32, 5}, {5, 1}: 0.00054706ms, 0.00480635% main:@901 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](KasFvpq4IG:0) -> float_type, {32, 2}, {5, 1}: 0.00094576ms, 0.00830924% JCVMzpWyri:0 = @param:JCVMzpWyri:0 -> float_type, {32, 17}, {17, 1}: 0.00046888ms, 0.00411947% main:@903 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](JCVMzpWyri:0) -> float_type, {32, 4}, {17, 1}: 0.00068416ms, 0.00601088% JSPsGophtx:0 = @param:JSPsGophtx:0 -> float_type, {32, 25}, {25, 1}: 0.00048626ms, 0.00427217% main:@905 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](JSPsGophtx:0) -> float_type, {32, 8}, {25, 1}: 0.00068378ms, 0.00600754% JaHTUxB6zL:0 = @param:JaHTUxB6zL:0 -> float_type, {32, 17}, {17, 1}: 0.00052096ms, 0.00457704% main:@907 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](JaHTUxB6zL:0) -> float_type, {32, 4}, {17, 1}: 0.00065556ms, 0.0057596% khJcIxW3zv:0 = @param:khJcIxW3zv:0 -> float_type, {32, 9}, {9, 1}: 0.00054518ms, 0.00478983% main:@909 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](khJcIxW3zv:0) -> float_type, {32, 4}, {9, 1}: 0.00064636ms, 0.00567877% LMobYAd5ru:0 = @param:LMobYAd5ru:0 -> float_type, {32, 17}, {17, 1}: 0.00062ms, 0.00544718% main:@911 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](LMobYAd5ru:0) -> float_type, {32, 4}, {17, 1}: 0.00064724ms, 0.00568651% lQmygxs5aD:0 = @param:lQmygxs5aD:0 -> float_type, {32, 13}, {13, 1}: 0.00052996ms, 0.00465611% main:@913 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](lQmygxs5aD:0) -> float_type, {32, 2}, {13, 1}: 0.000683ms, 0.00600068% fMiQsYNnet:0 = @param:fMiQsYNnet:0 -> float_type, {32, 5}, {5, 1}: 0.00049242ms, 0.00432629% main:@915 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](fMiQsYNnet:0) -> float_type, {32, 2}, {5, 1}: 0.00065194ms, 0.0057278% XPcHGKbop2:0 = @param:XPcHGKbop2:0 -> float_type, {32, 13}, {13, 1}: 0.00053082ms, 0.00466367% main:@917 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](XPcHGKbop2:0) -> float_type, {32, 2}, {13, 1}: 0.00095324ms, 0.00837495% TSqGwrHeo4:0 = @param:TSqGwrHeo4:0 -> float_type, {32, 5}, {5, 1}: 0.00051712ms, 0.0045433% main:@919 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](TSqGwrHeo4:0) -> float_type, {32, 2}, {5, 1}: 0.00065094ms, 0.00571901% Tej5M3O6As:0 = @param:Tej5M3O6As:0 -> float_type, {32, 17}, {17, 1}: 0.00048346ms, 0.00424757% main:@921 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](Tej5M3O6As:0) -> float_type, {32, 8}, {17, 1}: 0.00063026ms, 0.00553732% bTeX85HPmJ:0 = @param:bTeX85HPmJ:0 -> float_type, {32, 17}, {17, 1}: 0.00050108ms, 0.00440238% main:@923 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](bTeX85HPmJ:0) -> float_type, {32, 4}, {17, 1}: 0.0006693ms, 0.00588032% agDiJj8lLK:0 = @param:agDiJj8lLK:0 -> float_type, {32, 13}, {13, 1}: 0.00045214ms, 0.0039724% main:@925 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](agDiJj8lLK:0) -> float_type, {32, 2}, {13, 1}: 0.00066476ms, 0.00584043% fZhaL2tCm6:0 = @param:fZhaL2tCm6:0 -> float_type, {32, 17}, {17, 1}: 0.00050868ms, 0.00446915% main:@927 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](fZhaL2tCm6:0) -> float_type, {32, 8}, {17, 1}: 0.0006804ms, 0.00597784% cN2JSTnFOu:0 = @param:cN2JSTnFOu:0 -> float_type, {32, 9}, {9, 1}: 0.0004733ms, 0.00415831% main:@929 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](cN2JSTnFOu:0) -> float_type, {32, 4}, {9, 1}: 0.0006425ms, 0.00564486% omr2g4XaH5:0 = @param:omr2g4XaH5:0 -> float_type, {32, 17}, {17, 1}: 0.0005302ms, 0.00465822% main:@931 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](omr2g4XaH5:0) -> float_type, {32, 8}, {17, 1}: 0.00065566ms, 0.00576048% iCbkMrmXBh:0 = @param:iCbkMrmXBh:0 -> float_type, {32, 9}, {9, 1}: 0.00051178ms, 0.00449638% main:@933 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](iCbkMrmXBh:0) -> float_type, {32, 4}, {9, 1}: 0.00092292ms, 0.00810857% d8RoEzG52p:0 = @param:d8RoEzG52p:0 -> float_type, {32, 17}, {17, 1}: 0.00047624ms, 0.00418414% main:@935 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](d8RoEzG52p:0) -> float_type, {32, 8}, {17, 1}: 0.00065182ms, 0.00572674% eQquakfWBY:0 = @param:eQquakfWBY:0 -> float_type, {32, 17}, {17, 1}: 0.00047136ms, 0.00414126% main:@937 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](eQquakfWBY:0) -> float_type, {32, 4}, {17, 1}: 0.00067114ms, 0.00589649% qzd23aNGIT:0 = @param:qzd23aNGIT:0 -> float_type, {32, 17}, {17, 1}: 0.00046112ms, 0.0040513% main:@939 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](qzd23aNGIT:0) -> float_type, {32, 4}, {17, 1}: 0.00066968ms, 0.00588366% Vbv2Om3ptd:0 = @param:Vbv2Om3ptd:0 -> float_type, {32, 17}, {17, 1}: 0.00048072ms, 0.0042235% main:@941 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](Vbv2Om3ptd:0) -> float_type, {32, 8}, {17, 1}: 0.00068656ms, 0.00603196% r1WGP7Xe9H:0 = @param:r1WGP7Xe9H:0 -> float_type, {32, 13}, {13, 1}: 0.00045148ms, 0.0039666% main:@943 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](r1WGP7Xe9H:0) -> float_type, {32, 2}, {13, 1}: 0.00064362ms, 0.0056547% aqMQBgb3hf:0 = @param:aqMQBgb3hf:0 -> float_type, {32, 17}, {17, 1}: 0.00046594ms, 0.00409364% main:@945 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](aqMQBgb3hf:0) -> float_type, {32, 8}, {17, 1}: 0.00064266ms, 0.00564627% main:@946 = load[offset=30720,end=51200](main:@1) -> float_type, {32, 160}, {160, 1}: 0.00047958ms, 0.00421348% main:@947 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=134,local=256,](main:@879,main:@887,main:@899,main:@913,main:@919,main:@943,main:@925,main:@917,main:@881,main:@915,main:@889,main:@897,main:@901,main:@875,main:@929,main:@885,main:@871,main:@911,main:@909,main:@893,main:@923,main:@891,main:@937,main:@903,main:@877,main:@939,main:@933,main:@907,main:@873,main:@905,main:@927,main:@935,main:@921,main:@931,main:@883,main:@945,main:@941,main:@895,main:@946) -> float_type, {32, 160}, {160, 1}: 0.0475979ms, 0.418184% main:@948 = load[offset=10240,end=30720](main:@1) -> float_type, {32, 160}, {160, 1}: 0.00070922ms, 0.00623105% main:@949 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@947,main:@869,main:@948) -> float_type, {32, 160}, {160, 1}: 0.0211072ms, 0.185443% 2RDXZrx9GA:0 = @param:2RDXZrx9GA:0 -> float_type, {32, 45}, {45, 1}: 0.00070188ms, 0.00616656% main:@951 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](2RDXZrx9GA:0) -> float_type, {32, 16}, {45, 1}: 0.0006979ms, 0.00613159% main:@952 = unsqueeze[axes={1},steps={}](main:@951) -> float_type, {32, 1, 16}, {45, 16, 1}: 0.0007521ms, 0.00660778% duyRoIEkH8:0 = @param:duyRoIEkH8:0 -> float_type, {32, 45}, {45, 1}: 0.0005253ms, 0.00461517% main:@954 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](duyRoIEkH8:0) -> float_type, {32, 16}, {45, 1}: 0.00077474ms, 0.00680669% main:@955 = unsqueeze[axes={1},steps={}](main:@954) -> float_type, {32, 1, 16}, {45, 16, 1}: 0.0005699ms, 0.00500701% 2srtUfGvja:0 = @param:2srtUfGvja:0 -> float_type, {32, 41}, {41, 1}: 0.00054712ms, 0.00480687% main:@957 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](2srtUfGvja:0) -> float_type, {32, 16}, {41, 1}: 0.00069404ms, 0.00609768% main:@958 = unsqueeze[axes={1},steps={}](main:@957) -> float_type, {32, 1, 16}, {41, 16, 1}: 0.00057392ms, 0.00504233% 3Laj620MSn:0 = @param:3Laj620MSn:0 -> float_type, {32, 45}, {45, 1}: 0.00051142ms, 0.00449322% main:@960 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](3Laj620MSn:0) -> float_type, {32, 16}, {45, 1}: 0.00073486ms, 0.00645632% main:@961 = unsqueeze[axes={1},steps={}](main:@960) -> float_type, {32, 1, 16}, {45, 16, 1}: 0.00070794ms, 0.0062198% CHAYRkBvIo:0 = @param:CHAYRkBvIo:0 -> float_type, {32, 41}, {41, 1}: 0.00045268ms, 0.00397714% main:@963 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](CHAYRkBvIo:0) -> float_type, {32, 16}, {41, 1}: 0.00059554ms, 0.00523228% main:@964 = unsqueeze[axes={1},steps={}](main:@963) -> float_type, {32, 1, 16}, {41, 16, 1}: 0.00054434ms, 0.00478245% jznORDJr0N:0 = @param:jznORDJr0N:0 -> float_type, {32, 45}, {45, 1}: 0.0005595ms, 0.00491564% main:@966 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](jznORDJr0N:0) -> float_type, {32, 16}, {45, 1}: 0.0007538ms, 0.00662272% main:@967 = unsqueeze[axes={1},steps={}](main:@966) -> float_type, {32, 1, 16}, {45, 16, 1}: 0.00054406ms, 0.00477999% HRGzlnEkXQ:0 = @param:HRGzlnEkXQ:0 -> float_type, {32, 41}, {41, 1}: 0.00049418ms, 0.00434175% main:@969 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](HRGzlnEkXQ:0) -> float_type, {32, 16}, {41, 1}: 0.00062234ms, 0.00546774% main:@970 = unsqueeze[axes={1},steps={}](main:@969) -> float_type, {32, 1, 16}, {41, 16, 1}: 0.00071646ms, 0.00629466% HzcyUlm8Yj:0 = @param:HzcyUlm8Yj:0 -> float_type, {32, 45}, {45, 1}: 0.00051044ms, 0.00448461% main:@972 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](HzcyUlm8Yj:0) -> float_type, {32, 16}, {45, 1}: 0.00074086ms, 0.00650903% main:@973 = unsqueeze[axes={1},steps={}](main:@972) -> float_type, {32, 1, 16}, {45, 16, 1}: 0.00089382ms, 0.0078529% main:@974 = slice[axes={0, 1},starts={0, 289},ends={32, 305}](lyiDxfzZg3:0) -> float_type, {32, 16}, {701, 1}: 0.00061594ms, 0.00541151% main:@975 = unsqueeze[axes={1},steps={}](main:@974) -> float_type, {32, 1, 16}, {701, 16, 1}: 0.00061712ms, 0.00542188% VUqfdN5Fzv:0 = @param:VUqfdN5Fzv:0 -> float_type, {32, 45}, {45, 1}: 0.00051868ms, 0.00455701% main:@977 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](VUqfdN5Fzv:0) -> float_type, {32, 16}, {45, 1}: 0.00073496ms, 0.00645719% main:@978 = unsqueeze[axes={1},steps={}](main:@977) -> float_type, {32, 1, 16}, {45, 16, 1}: 0.0009025ms, 0.00792916% eQI8ikEMgP:0 = @param:eQI8ikEMgP:0 -> float_type, {32, 41}, {41, 1}: 0.0005054ms, 0.00444033% main:@980 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](eQI8ikEMgP:0) -> float_type, {32, 16}, {41, 1}: 0.00061506ms, 0.00540378% main:@981 = unsqueeze[axes={1},steps={}](main:@980) -> float_type, {32, 1, 16}, {41, 16, 1}: 0.0005944ms, 0.00522227% main:@982 = reshape_lazy[dims={-1, 10, 16}](main:@949) -> float_type, {32, 10, 16}, {160, 16, 1}: 0.00051186ms, 0.00449709% main:@983 = load[offset=30720,end=73728](main:@1) -> float_type, {32, 21, 16}, {336, 16, 1}: 0.00048644ms, 0.00427375% main:@984 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=896,local=256,](main:@967,main:@970,main:@975,main:@952,main:@978,main:@964,main:@973,main:@958,main:@961,main:@955,main:@981,main:@982,main:@983) -> float_type, {32, 21, 16}, {336, 16, 1}: 0.0272533ms, 0.239442% main:@985 = load[offset=114688,end=128128](main:@1) -> float_type, {32, 21, 5}, {105, 5, 1}: 0.00063804ms, 0.00560568% main:@986 = reshape_lazy[dims={32, 16, 5}](main:@868) -> float_type, {32, 16, 5}, {80, 5, 1}: 0.00051082ms, 0.00448795% main:@987 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@984,main:@986,main:@985) -> float_type, {32, 21, 5}, {105, 5, 1}: 0.0195082ms, 0.171394% XRvxjULC7M:0 = @param:XRvxjULC7M:0 -> float_type, {32, 60, 5}, {300, 5, 1}: 0.00068708ms, 0.00603653% main:@989 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 60, 5}](XRvxjULC7M:0) -> float_type, {32, 60, 4}, {300, 5, 1}: 0.00069684ms, 0.00612228% main:@990 = load[offset=6365952,end=6888192](main:@1) -> float_type, {32, 60, 68}, {4080, 68, 1}: 0.00050114ms, 0.0044029% main:@991 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=32640,local=256,](main:@135,main:@215,main:@989,main:@293,main:@990) -> float_type, {32, 60, 68}, {4080, 68, 1}: 0.0205423ms, 0.18048% main:@992 = hip::hip_copy_literal[id=main:@literal:152] -> float_type, {68, 64}, {64, 1}: 0.00083418ms, 0.00732892% main:@993 = load[offset=1942272,end=2433792](main:@1) -> float_type, {1920, 64}, {64, 1}: 0.00048788ms, 0.0042864% main:@994 = reshape_lazy[dims={1920, 68}](main:@991) -> float_type, {1920, 68}, {68, 1}: 0.0004984ms, 0.00437883% main:@995 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@994,main:@992,main:@993) -> float_type, {1920, 64}, {64, 1}: 0.023891ms, 0.209901% main:@996 = load[offset=0,end=6144](main:@1) -> float_type, {32, 48}, {48, 1}: 0.00061478ms, 0.00540132% main:@997 = slice[axes={0, 1},starts={0, 45},ends={32, 53}](2CnSyUBzm0:0) -> float_type, {32, 8}, {65, 1}: 0.00071068ms, 0.00624388% main:@998 = slice[axes={0, 1},starts={0, 13},ends={32, 21}](9YqhflbMKe:0) -> float_type, {32, 8}, {29, 1}: 0.0009451ms, 0.00830344% main:@999 = slice[axes={0, 1},starts={0, 45},ends={32, 53}](Qp3K1PHvte:0) -> float_type, {32, 8}, {69, 1}: 0.00076662ms, 0.00673535% main:@1000 = slice[axes={0, 1},starts={0, 189},ends={32, 197}](nv2tzWRf1I:0) -> float_type, {32, 8}, {301, 1}: 0.00063008ms, 0.00553574% main:@1001 = slice[axes={0, 1},starts={0, 209},ends={32, 217}](uFSOc5Chp2:0) -> float_type, {32, 8}, {481, 1}: 0.00064322ms, 0.00565119% main:@1002 = slice[axes={0, 1},starts={0, 49},ends={32, 57}](ifP6Ceko5T:0) -> float_type, {32, 8}, {73, 1}: 0.00089358ms, 0.00785079% main:@1003 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=256,local=256,](main:@1000,main:@997,main:@1001,main:@1002,main:@999,main:@998,main:@996) -> float_type, {32, 48}, {48, 1}: 0.0161302ms, 0.141716% main:@1004 = hip::hip_copy_literal[id=main:@literal:160] -> float_type, {48, 64}, {64, 1}: 0.00080124ms, 0.00703951% main:@1005 = load[offset=7680,end=15872](main:@1) -> float_type, {32, 64}, {64, 1}: 0.00049778ms, 0.00437338% main:@1006 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1003,main:@1004,main:@1005) -> float_type, {32, 64}, {64, 1}: 0.0203663ms, 0.178934% main:@1007 = reshape_lazy[dims={32, 60, 64}](main:@995) -> float_type, {32, 60, 64}, {3840, 64, 1}: 0.00070718ms, 0.00621312% main:@1008 = transpose[permutation={0, 2, 1}](main:@1007) -> float_type, {32, 64, 60}, {3840, 1, 64}: 0.00054758ms, 0.00481092% main:@1009 = load[offset=0,end=7680](main:@1) -> float_type, {32, 1, 60}, {60, 60, 1}: 0.00050254ms, 0.0044152% main:@1010 = reshape_lazy[dims={32, 1, 64}](main:@1006) -> float_type, {32, 1, 64}, {64, 64, 1}: 0.0004739ms, 0.00416358% main:@1011 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1010,main:@1008,main:@1009) -> float_type, {32, 1, 60}, {60, 60, 1}: 0.0232722ms, 0.204464% main:@1012 = load[offset=6365952,end=7011072](main:@1) -> float_type, {32, 60, 84}, {5040, 84, 1}: 0.00066512ms, 0.0058436% 4luvCw32jf:0 = @param:4luvCw32jf:0 -> float_type, {32, 60, 5}, {300, 5, 1}: 0.00052192ms, 0.00458547% main:@1014 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 60, 5}](4luvCw32jf:0) -> float_type, {32, 60, 4}, {300, 5, 1}: 0.0007325ms, 0.00643558% F5awZuD1c2:0 = @param:F5awZuD1c2:0 -> float_type, {32, 60, 9}, {540, 9, 1}: 0.00052748ms, 0.00463432% main:@1016 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 60, 9}](F5awZuD1c2:0) -> float_type, {32, 60, 8}, {540, 9, 1}: 0.00107888ms, 0.0094788% PnxiROp9zw:0 = @param:PnxiROp9zw:0 -> float_type, {32, 60, 5}, {300, 5, 1}: 0.00050138ms, 0.00440501% main:@1018 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 60, 5}](PnxiROp9zw:0) -> float_type, {32, 60, 4}, {300, 5, 1}: 0.0006956ms, 0.00611139% main:@1019 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=23040,local=256,](main:@135,main:@215,main:@989,main:@293,main:@1018,main:@1016,main:@1014,main:@1012) -> float_type, {32, 60, 84}, {5040, 84, 1}: 0.0218802ms, 0.192235% main:@1020 = hip::hip_copy_literal[id=main:@literal:134] -> float_type, {84, 64}, {64, 1}: 0.00084582ms, 0.00743118% main:@1021 = reshape_lazy[dims={1920, 84}](main:@1019) -> float_type, {1920, 84}, {84, 1}: 0.00054596ms, 0.00479668% main:@1022 = load[offset=1942272,end=2433792](main:@1) -> float_type, {1920, 64}, {64, 1}: 0.000481ms, 0.00422596% main:@1023 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1021,main:@1020,main:@1022) -> float_type, {1920, 64}, {64, 1}: 0.0216594ms, 0.190295% mt0cODHoal:0 = @param:mt0cODHoal:0 -> int32_type, {32, 60}, {60, 1}: 0.00067524ms, 0.00593251% main:@1025 = unsqueeze[axes={1},steps={}](mt0cODHoal:0) -> int32_type, {32, 1, 60}, {60, 60, 1}: 0.000625ms, 0.00549111% main:@1026 = load[offset=7680,end=15360](main:@1) -> float_type, {32, 1, 60}, {60, 60, 1}: 0.0004916ms, 0.00431909% main:@1027 = gpu::code_object[code_object=13544,symbol_name=convert_add_add_sub_kernel,global=960,local=256,](main:@1025,main:@1011,main:@1026) -> float_type, {32, 1, 60}, {60, 60, 1}: 0.0134677ms, 0.118324% main:@1028 = load[offset=0,end=7680](main:@1) -> float_type, {32, 1, 60}, {60, 60, 1}: 0.00066842ms, 0.00587259% main:@1029 = gpu::code_object[code_object=14552,symbol_name=softmax_kernel,global=2048,local=64,](main:@1027,main:@1028) -> float_type, {32, 1, 60}, {60, 60, 1}: 0.0148199ms, 0.130204% main:@1030 = reshape_lazy[dims={32, 60, 64}](main:@1023) -> float_type, {32, 60, 64}, {3840, 64, 1}: 0.00067572ms, 0.00593672% main:@1031 = load[offset=405632,end=413824](main:@1) -> float_type, {32, 1, 64}, {64, 64, 1}: 0.0004991ms, 0.00438498% main:@1032 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1029,main:@1030,main:@1031) -> float_type, {32, 1, 64}, {64, 64, 1}: 0.0225975ms, 0.198536% main:@1033 = slice[axes={1},starts={0},ends={256}](main:@624) -> float_type, {480, 256}, {768, 1}: 0.00082432ms, 0.00724229% main:@1034 = reshape_lazy[dims={32, 15, 256}](main:@1033) -> float_type, {32, 15, 256}, {11520, 768, 1}: 0.0006376ms, 0.00560181% main:@1035 = slice[axes={2},starts={0},ends={64}](main:@1034) -> float_type, {32, 15, 64}, {11520, 768, 1}: 0.00078646ms, 0.00690966% main:@1036 = unsqueeze[axes={0},steps={}](main:@1035) -> float_type, {1, 32, 15, 64}, {368640, 11520, 768, 1}: 0.00074984ms, 0.00658793% main:@1037 = slice[axes={2},starts={64},ends={128}](main:@1034) -> float_type, {32, 15, 64}, {11520, 768, 1}: 0.00101056ms, 0.00887855% main:@1038 = unsqueeze[axes={0},steps={}](main:@1037) -> float_type, {1, 32, 15, 64}, {368640, 11520, 768, 1}: 0.0006189ms, 0.00543752% main:@1039 = slice[axes={2},starts={128},ends={192}](main:@1034) -> float_type, {32, 15, 64}, {11520, 768, 1}: 0.00074036ms, 0.00650464% main:@1040 = unsqueeze[axes={0},steps={}](main:@1039) -> float_type, {1, 32, 15, 64}, {368640, 11520, 768, 1}: 0.00060132ms, 0.00528306% main:@1041 = slice[axes={2},starts={192},ends={256}](main:@1034) -> float_type, {32, 15, 64}, {11520, 768, 1}: 0.00069296ms, 0.00608819% main:@1042 = unsqueeze[axes={0},steps={}](main:@1041) -> float_type, {1, 32, 15, 64}, {368640, 11520, 768, 1}: 0.00097064ms, 0.00852783% main:@1043 = load[offset=1942272,end=2433792](main:@1) -> float_type, {4, 32, 15, 64}, {30720, 960, 64, 1}: 0.00048798ms, 0.00428728% main:@1044 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=15360,local=256,](main:@1036,main:@1038,main:@1040,main:@1042,main:@1043) -> float_type, {4, 32, 15, 64}, {30720, 960, 64, 1}: 0.0211548ms, 0.185862% main:@1045 = slice[axes={1},starts={512},ends={768}](main:@624) -> float_type, {480, 256}, {768, 1}: 0.00076588ms, 0.00672885% main:@1046 = reshape_lazy[dims={32, 15, 256}](main:@1045) -> float_type, {32, 15, 256}, {11520, 768, 1}: 0.00054122ms, 0.00475504% main:@1047 = slice[axes={2},starts={0},ends={64}](main:@1046) -> float_type, {32, 15, 64}, {11520, 768, 1}: 0.00082916ms, 0.00728481% main:@1048 = unsqueeze[axes={0},steps={}](main:@1047) -> float_type, {1, 32, 15, 64}, {368640, 11520, 768, 1}: 0.00056442ms, 0.00495887% main:@1049 = slice[axes={2},starts={64},ends={128}](main:@1046) -> float_type, {32, 15, 64}, {11520, 768, 1}: 0.00070156ms, 0.00616375% main:@1050 = unsqueeze[axes={0},steps={}](main:@1049) -> float_type, {1, 32, 15, 64}, {368640, 11520, 768, 1}: 0.0005936ms, 0.00521524% main:@1051 = slice[axes={2},starts={128},ends={192}](main:@1046) -> float_type, {32, 15, 64}, {11520, 768, 1}: 0.00068168ms, 0.00598909% main:@1052 = unsqueeze[axes={0},steps={}](main:@1051) -> float_type, {1, 32, 15, 64}, {368640, 11520, 768, 1}: 0.0009648ms, 0.00847652% main:@1053 = slice[axes={2},starts={192},ends={256}](main:@1046) -> float_type, {32, 15, 64}, {11520, 768, 1}: 0.00066888ms, 0.00587663% main:@1054 = unsqueeze[axes={0},steps={}](main:@1053) -> float_type, {1, 32, 15, 64}, {368640, 11520, 768, 1}: 0.00064274ms, 0.00564697% main:@1055 = load[offset=6365952,end=6857472](main:@1) -> float_type, {4, 32, 15, 64}, {30720, 960, 64, 1}: 0.00051726ms, 0.00454453% main:@1056 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=15360,local=256,](main:@1048,main:@1050,main:@1052,main:@1054,main:@1055) -> float_type, {4, 32, 15, 64}, {30720, 960, 64, 1}: 0.0208433ms, 0.183125% main:@1057 = load[offset=1249408,end=1364608](main:@1) -> float_type, {4, 32, 15, 15}, {7200, 225, 15, 1}: 0.00065446ms, 0.00574994% main:@1058 = transpose[permutation={0, 1, 3, 2}](main:@1056) -> float_type, {4, 32, 64, 15}, {30720, 960, 1, 64}: 0.00053078ms, 0.00466331% main:@1059 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1044,main:@1058,main:@1057) -> float_type, {4, 32, 15, 15}, {7200, 225, 15, 1}: 0.0221063ms, 0.194221% main:@1060 = load[offset=1364608,end=1479808](main:@1) -> float_type, {4, 32, 15, 15}, {7200, 225, 15, 1}: 0.00065104ms, 0.00571989% main:@1061 = gpu::code_object[code_object=13512,symbol_name=pow_kernel,global=14400,local=256,](main:@1059,main:@1060) -> float_type, {4, 32, 15, 15}, {7200, 225, 15, 1}: 0.0161653ms, 0.142024% main:@1062 = load[offset=1249408,end=1364608](main:@1) -> float_type, {4, 32, 15, 15}, {7200, 225, 15, 1}: 0.00063972ms, 0.00562044% main:@1063 = gpu::code_object[code_object=14576,symbol_name=softmax_kernel,global=122880,local=64,](main:@1061,main:@1062) -> float_type, {4, 32, 15, 15}, {7200, 225, 15, 1}: 0.0192089ms, 0.168765% main:@1064 = slice[axes={1},starts={256},ends={512}](main:@624) -> float_type, {480, 256}, {768, 1}: 0.0008295ms, 0.0072878% main:@1065 = reshape_lazy[dims={32, 15, 256}](main:@1064) -> float_type, {32, 15, 256}, {11520, 768, 1}: 0.00060022ms, 0.0052734% main:@1066 = slice[axes={2},starts={0},ends={64}](main:@1065) -> float_type, {32, 15, 64}, {11520, 768, 1}: 0.00069706ms, 0.00612421% main:@1067 = unsqueeze[axes={0},steps={}](main:@1066) -> float_type, {1, 32, 15, 64}, {368640, 11520, 768, 1}: 0.00092856ms, 0.00815812% main:@1068 = slice[axes={2},starts={64},ends={128}](main:@1065) -> float_type, {32, 15, 64}, {11520, 768, 1}: 0.00066106ms, 0.00580792% main:@1069 = unsqueeze[axes={0},steps={}](main:@1068) -> float_type, {1, 32, 15, 64}, {368640, 11520, 768, 1}: 0.00094864ms, 0.00833454% main:@1070 = slice[axes={2},starts={128},ends={192}](main:@1065) -> float_type, {32, 15, 64}, {11520, 768, 1}: 0.00065176ms, 0.00572622% main:@1071 = unsqueeze[axes={0},steps={}](main:@1070) -> float_type, {1, 32, 15, 64}, {368640, 11520, 768, 1}: 0.00065612ms, 0.00576452% main:@1072 = slice[axes={2},starts={192},ends={256}](main:@1065) -> float_type, {32, 15, 64}, {11520, 768, 1}: 0.00099796ms, 0.00876785% main:@1073 = unsqueeze[axes={0},steps={}](main:@1072) -> float_type, {1, 32, 15, 64}, {368640, 11520, 768, 1}: 0.00056646ms, 0.00497679% main:@1074 = load[offset=6365952,end=6857472](main:@1) -> float_type, {4, 32, 15, 64}, {30720, 960, 64, 1}: 0.00048568ms, 0.00426708% main:@1075 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=15360,local=256,](main:@1067,main:@1069,main:@1071,main:@1073,main:@1074) -> float_type, {4, 32, 15, 64}, {30720, 960, 64, 1}: 0.0206986ms, 0.181853% main:@1076 = load[offset=1942272,end=2433792](main:@1) -> float_type, {4, 32, 15, 64}, {30720, 960, 64, 1}: 0.00062622ms, 0.00550183% main:@1077 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1063,main:@1075,main:@1076) -> float_type, {4, 32, 15, 64}, {30720, 960, 64, 1}: 0.0198739ms, 0.174607% main:@1078 = slice[axes={1},starts={32},ends={96}](main:@652) -> float_type, {32, 64}, {128, 1}: 0.00075518ms, 0.00663484% main:@1079 = load[offset=6400,end=14592](main:@1) -> float_type, {32, 64}, {64, 1}: 0.00047842ms, 0.00420329% main:@1080 = gpu::code_object[code_object=13528,symbol_name=contiguous_kernel,global=1024,local=256,](main:@1078,main:@1079) -> float_type, {32, 64}, {64, 1}: 0.0130357ms, 0.114529% main:@1081 = slice[axes={1},starts={64},ends={128}](main:@730) -> float_type, {1600, 64}, {128, 1}: 0.00074732ms, 0.00656579% main:@1082 = reshape_lazy[dims={1, 32, 50, 64}](main:@1081) -> float_type, {1, 32, 50, 64}, {204800, 6400, 128, 1}: 0.00056462ms, 0.00496062% main:@1083 = transpose[permutation={0, 1, 3, 2}](main:@1082) -> float_type, {1, 32, 64, 50}, {204800, 6400, 1, 128}: 0.00092596ms, 0.00813528% main:@1084 = load[offset=0,end=6400](main:@1) -> float_type, {1, 32, 1, 50}, {1600, 50, 50, 1}: 0.00050844ms, 0.00446704% main:@1085 = reshape_lazy[dims={1, 32, 1, 64}](main:@1080) -> float_type, {1, 32, 1, 64}, {2048, 64, 64, 1}: 0.0004881ms, 0.00428834% main:@1086 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1085,main:@1083,main:@1084) -> float_type, {1, 32, 1, 50}, {1600, 50, 50, 1}: 0.0241224ms, 0.211934% main:@1087 = load[offset=6400,end=12800](main:@1) -> float_type, {32, 1, 50}, {50, 50, 1}: 0.000644ms, 0.00565804% ObZQkfyaKG:0 = @param:ObZQkfyaKG:0 -> int32_type, {32, 50}, {50, 1}: 0.00054016ms, 0.00474572% main:@1089 = unsqueeze[axes={1},steps={}](ObZQkfyaKG:0) -> int32_type, {32, 1, 50}, {50, 50, 1}: 0.00059532ms, 0.00523035% main:@1090 = gpu::code_object[code_object=13528,symbol_name=convert_add_kernel,global=800,local=256,](main:@1089,main:@1087) -> float_type, {32, 1, 50}, {50, 50, 1}: 0.0134632ms, 0.118284% main:@1091 = multibroadcast[out_lens={1, 32, 1, 50}](main:@1090) -> float_type, {1, 32, 1, 50}, {0, 50, 50, 1}: 0.00072902ms, 0.00640501% main:@1092 = load[offset=12800,end=19200](main:@1) -> float_type, {1, 32, 1, 50}, {1600, 50, 50, 1}: 0.00049476ms, 0.00434685% main:@1093 = gpu::code_object[code_object=13520,symbol_name=pow_pow_kernel,global=800,local=256,](main:@1086,main:@1091,main:@1092) -> float_type, {1, 32, 1, 50}, {1600, 50, 50, 1}: 0.0178816ms, 0.157104% main:@1094 = load[offset=0,end=6400](main:@1) -> float_type, {1, 32, 1, 50}, {1600, 50, 50, 1}: 0.00063896ms, 0.00561376% main:@1095 = gpu::code_object[code_object=14584,symbol_name=softmax_kernel,global=2048,local=64,](main:@1093,main:@1094) -> float_type, {1, 32, 1, 50}, {1600, 50, 50, 1}: 0.014741ms, 0.129511% main:@1096 = slice[axes={1},starts={0},ends={64}](main:@730) -> float_type, {1600, 64}, {128, 1}: 0.00072772ms, 0.00639358% main:@1097 = load[offset=1126528,end=1134720](main:@1) -> float_type, {1, 32, 1, 64}, {2048, 64, 64, 1}: 0.00047262ms, 0.00415233% main:@1098 = reshape_lazy[dims={1, 32, 50, 64}](main:@1096) -> float_type, {1, 32, 50, 64}, {204800, 6400, 128, 1}: 0.0009378ms, 0.0082393% main:@1099 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1095,main:@1098,main:@1097) -> float_type, {1, 32, 1, 64}, {2048, 64, 64, 1}: 0.0219034ms, 0.192439% main:@1100 = hip::hip_copy_literal[id=main:@literal:23] -> float_type, {64, 64}, {64, 1}: 0.00080774ms, 0.00709662% main:@1101 = load[offset=0,end=81920](main:@1) -> float_type, {32, 10, 64}, {640, 64, 1}: 0.0005103ms, 0.00448338% main:@1102 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=5120,local=256,](main:@40,main:@235,main:@79,main:@20,main:@1101) -> float_type, {32, 10, 64}, {640, 64, 1}: 0.0168631ms, 0.148155% main:@1103 = load[offset=6247680,end=6329600](main:@1) -> float_type, {320, 64}, {64, 1}: 0.000638ms, 0.00560532% main:@1104 = reshape_lazy[dims={320, 64}](main:@1102) -> float_type, {320, 64}, {64, 1}: 0.00054562ms, 0.00479369% main:@1105 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1104,main:@1100,main:@1103) -> float_type, {320, 64}, {64, 1}: 0.0193642ms, 0.17013% main:@1106 = hip::hip_copy_literal[id=main:@literal:19] -> float_type, {896}, {1}: 0.0008348ms, 0.00733437% main:@1107 = multibroadcast[out_lens={32, 896}](main:@1106) -> float_type, {32, 896}, {0, 1}: 0.00058764ms, 0.00516287% main:@1108 = load[offset=0,end=114688](main:@1) -> float_type, {32, 896}, {896, 1}: 0.00053258ms, 0.00467913% main:@1109 = gpu::code_object[code_object=13536,symbol_name=add_leaky_relu_kernel,global=14336,local=256,](main:@380,main:@1107,main:@1108) -> float_type, {32, 896}, {896, 1}: 0.0155531ms, 0.136646% main:@1110 = hip::hip_copy_literal[id=main:@literal:116] -> float_type, {128, 3429}, {3429, 1}: 0.0007719ms, 0.00678174% main:@1111 = slice[axes={1},starts={128},ends={256}](main:@1109) -> float_type, {32, 128}, {896, 1}: 0.00064268ms, 0.00564644% main:@1112 = load[offset=1134720,end=1573632](main:@1) -> float_type, {32, 3429}, {3429, 1}: 0.00049268ms, 0.00432858% 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, {32, 3429}, {3429, 1}: 0.0228264ms, 0.200547% main:@1114 = load[offset=6329600,end=6867200](main:@1) -> float_type, {32, 50, 84}, {4200, 84, 1}: 0.00066872ms, 0.00587522% pHflUqMEaX:0 = @param:pHflUqMEaX:0 -> float_type, {32, 50, 5}, {250, 5, 1}: 0.0005416ms, 0.00475838% main:@1116 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 50, 5}](pHflUqMEaX:0) -> float_type, {32, 50, 4}, {250, 5, 1}: 0.00069712ms, 0.00612474% lDIuWZ8VFU:0 = @param:lDIuWZ8VFU:0 -> float_type, {32, 50, 5}, {250, 5, 1}: 0.0005257ms, 0.00461868% main:@1118 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 50, 5}](lDIuWZ8VFU:0) -> float_type, {32, 50, 4}, {250, 5, 1}: 0.00080636ms, 0.0070845% waXyDlz530:0 = @param:waXyDlz530:0 -> float_type, {32, 50, 5}, {250, 5, 1}: 0.00050296ms, 0.00441889% main:@1120 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 50, 5}](waXyDlz530:0) -> float_type, {32, 50, 4}, {250, 5, 1}: 0.0006657ms, 0.00584869% main:@1121 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=22400,local=256,](main:@1116,main:@58,main:@274,main:@1120,main:@1118,main:@195,main:@1114) -> float_type, {32, 50, 84}, {4200, 84, 1}: 0.0214091ms, 0.188095% main:@1122 = hip::hip_copy_literal[id=main:@literal:13] -> float_type, {84, 256}, {256, 1}: 0.00085478ms, 0.00750991% main:@1123 = load[offset=4445440,end=6083840](main:@1) -> float_type, {1600, 256}, {256, 1}: 0.00050248ms, 0.00441468% main:@1124 = reshape_lazy[dims={1600, 84}](main:@1121) -> float_type, {1600, 84}, {84, 1}: 0.0005266ms, 0.00462659% main:@1125 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1124,main:@1122,main:@1123) -> float_type, {1600, 256}, {256, 1}: 0.0248936ms, 0.21871% main:@1126 = hip::hip_copy_literal[id=main:@literal:49] -> float_type, {256, 512}, {512, 1}: 0.00076144ms, 0.00668984% main:@1127 = load[offset=4072192,end=4137728](main:@1) -> float_type, {32, 512}, {512, 1}: 0.000493ms, 0.00433139% main:@1128 = slice[axes={0, 1, 2},starts={0, 9, 0},ends={32, 10, 256}](main:@713) -> float_type, {32, 1, 256}, {4608, 256, 1}: 0.00066222ms, 0.00581812% main:@1129 = squeeze[axes={1}](main:@1128) -> float_type, {32, 256}, {4608, 1}: 0.00076376ms, 0.00671022% main:@1130 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1129,main:@1126,main:@1127) -> float_type, {32, 512}, {512, 1}: 0.0262015ms, 0.2302% main:@1131 = load[offset=4137728,end=4203264](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00066646ms, 0.00585537% main:@1132 = gpu::code_object[code_object=13512,symbol_name=relu_kernel,global=8192,local=256,](main:@1130,main:@1131) -> float_type, {32, 512}, {512, 1}: 0.0145224ms, 0.127591% main:@1133 = load[offset=4072192,end=4104960](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00062212ms, 0.00546581% main:@1134 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1132,main:@715,main:@1133) -> float_type, {32, 256}, {256, 1}: 0.0303985ms, 0.267075% main:@1135 = slice[axes={0, 1, 2},starts={0, 16, 0},ends={32, 17, 256}](main:@713) -> float_type, {32, 1, 256}, {4608, 256, 1}: 0.00077058ms, 0.00677014% main:@1136 = squeeze[axes={1}](main:@1135) -> float_type, {32, 256}, {4608, 1}: 0.00068986ms, 0.00606096% main:@1137 = load[offset=4203264,end=4268800](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00049838ms, 0.00437865% main:@1138 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1136,main:@714,main:@1137) -> float_type, {32, 512}, {512, 1}: 0.0238806ms, 0.209809% main:@1139 = load[offset=4137728,end=4203264](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00063824ms, 0.00560743% main:@1140 = gpu::code_object[code_object=13512,symbol_name=relu_kernel,global=8192,local=256,](main:@1138,main:@1139) -> float_type, {32, 512}, {512, 1}: 0.0145201ms, 0.127571% main:@1141 = hip::hip_copy_literal[id=main:@literal:84] -> float_type, {512, 256}, {256, 1}: 0.00077192ms, 0.00678192% main:@1142 = load[offset=4104960,end=4137728](main:@1) -> float_type, {32, 256}, {256, 1}: 0.0004797ms, 0.00421454% 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, {32, 256}, {256, 1}: 0.0304201ms, 0.267264% main:@1144 = hip::hip_copy_literal[id=main:@literal:142] -> float_type, {256, 512}, {512, 1}: 0.00073952ms, 0.00649726% main:@1145 = slice[axes={0, 1, 2},starts={0, 1, 0},ends={32, 2, 256}](main:@713) -> float_type, {32, 1, 256}, {4608, 256, 1}: 0.00066514ms, 0.00584377% main:@1146 = squeeze[axes={1}](main:@1145) -> float_type, {32, 256}, {4608, 1}: 0.00059588ms, 0.00523527% main:@1147 = load[offset=4137728,end=4203264](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00048678ms, 0.00427674% 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, {32, 512}, {512, 1}: 0.0241519ms, 0.212193% main:@1149 = hip::hip_copy_literal[id=main:@literal:120] -> float_type, {512, 256}, {256, 1}: 0.00085994ms, 0.00755524% main:@1150 = load[offset=4203264,end=4268800](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00048782ms, 0.00428588% main:@1151 = gpu::code_object[code_object=13512,symbol_name=relu_kernel,global=8192,local=256,](main:@1148,main:@1150) -> float_type, {32, 512}, {512, 1}: 0.0143481ms, 0.126059% main:@1152 = load[offset=4137728,end=4170496](main:@1) -> float_type, {32, 256}, {256, 1}: 0.0006028ms, 0.00529607% 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, {32, 256}, {256, 1}: 0.0302023ms, 0.265351% main:@1154 = hip::hip_copy_literal[id=main:@literal:62] -> float_type, {256, 512}, {512, 1}: 0.00085406ms, 0.00750358% main:@1155 = slice[axes={0, 1, 2},starts={0, 15, 0},ends={32, 16, 256}](main:@713) -> float_type, {32, 1, 256}, {4608, 256, 1}: 0.00064594ms, 0.00567508% main:@1156 = squeeze[axes={1}](main:@1155) -> float_type, {32, 256}, {4608, 1}: 0.00061214ms, 0.00537812% main:@1157 = load[offset=4268800,end=4334336](main:@1) -> float_type, {32, 512}, {512, 1}: 0.0004639ms, 0.00407572% 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, {32, 512}, {512, 1}: 0.0233929ms, 0.205525% main:@1159 = load[offset=4203264,end=4268800](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00062178ms, 0.00546282% main:@1160 = gpu::code_object[code_object=13512,symbol_name=relu_kernel,global=8192,local=256,](main:@1158,main:@1159) -> float_type, {32, 512}, {512, 1}: 0.0142469ms, 0.12517% main:@1161 = hip::hip_copy_literal[id=main:@literal:95] -> float_type, {512, 256}, {256, 1}: 0.00076472ms, 0.00671866% main:@1162 = load[offset=4170496,end=4203264](main:@1) -> float_type, {32, 256}, {256, 1}: 0.0004697ms, 0.00412668% 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, {32, 256}, {256, 1}: 0.0302546ms, 0.26581% main:@1164 = hip::hip_copy_literal[id=main:@literal:138] -> float_type, {256, 512}, {512, 1}: 0.0007996ms, 0.00702511% main:@1165 = slice[axes={0, 1, 2},starts={0, 8, 0},ends={32, 9, 256}](main:@713) -> float_type, {32, 1, 256}, {4608, 256, 1}: 0.00061944ms, 0.00544226% main:@1166 = squeeze[axes={1}](main:@1165) -> float_type, {32, 256}, {4608, 1}: 0.00057546ms, 0.00505586% main:@1167 = load[offset=4301568,end=4367104](main:@1) -> float_type, {32, 512}, {512, 1}: 0.0004694ms, 0.00412404% 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, {32, 512}, {512, 1}: 0.0238676ms, 0.209695% main:@1169 = load[offset=4236032,end=4301568](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00061872ms, 0.00543594% main:@1170 = gpu::code_object[code_object=13512,symbol_name=relu_kernel,global=8192,local=256,](main:@1168,main:@1169) -> float_type, {32, 512}, {512, 1}: 0.0142927ms, 0.125573% main:@1171 = hip::hip_copy_literal[id=main:@literal:44] -> float_type, {512, 256}, {256, 1}: 0.00073842ms, 0.00648759% main:@1172 = load[offset=4203264,end=4236032](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00048252ms, 0.00423931% 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, {32, 256}, {256, 1}: 0.0301328ms, 0.26474% main:@1174 = hip::hip_copy_literal[id=main:@literal:48] -> float_type, {256, 512}, {512, 1}: 0.0008059ms, 0.00708046% main:@1175 = slice[axes={0, 1, 2},starts={0, 5, 0},ends={32, 6, 256}](main:@713) -> float_type, {32, 1, 256}, {4608, 256, 1}: 0.0006256ms, 0.00549638% main:@1176 = squeeze[axes={1}](main:@1175) -> float_type, {32, 256}, {4608, 1}: 0.00060092ms, 0.00527955% main:@1177 = load[offset=4347136,end=4412672](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00046358ms, 0.00407291% 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, {32, 512}, {512, 1}: 0.0231835ms, 0.203685% main:@1179 = load[offset=4281600,end=4347136](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00061778ms, 0.00542768% main:@1180 = gpu::code_object[code_object=13512,symbol_name=relu_kernel,global=8192,local=256,](main:@1178,main:@1179) -> float_type, {32, 512}, {512, 1}: 0.0143568ms, 0.126136% main:@1181 = load[offset=4248832,end=4281600](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00063348ms, 0.00556561% main:@1182 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1180,main:@715,main:@1181) -> float_type, {32, 256}, {256, 1}: 0.0302864ms, 0.26609% main:@1183 = slice[axes={0, 1, 2},starts={0, 2, 0},ends={32, 3, 256}](main:@713) -> float_type, {32, 1, 256}, {4608, 256, 1}: 0.00075538ms, 0.0066366% main:@1184 = squeeze[axes={1}](main:@1183) -> float_type, {32, 256}, {4608, 1}: 0.00091968ms, 0.0080801% main:@1185 = load[offset=4379904,end=4445440](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00047604ms, 0.00418238% main:@1186 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1184,main:@716,main:@1185) -> float_type, {32, 512}, {512, 1}: 0.0237702ms, 0.20884% main:@1187 = load[offset=4314368,end=4379904](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00061492ms, 0.00540255% main:@1188 = gpu::code_object[code_object=13512,symbol_name=relu_kernel,global=8192,local=256,](main:@1186,main:@1187) -> float_type, {32, 512}, {512, 1}: 0.0142844ms, 0.1255% main:@1189 = hip::hip_copy_literal[id=main:@literal:97] -> float_type, {512, 256}, {256, 1}: 0.0007727ms, 0.00678877% main:@1190 = load[offset=4281600,end=4314368](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00046882ms, 0.00411895% 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, {32, 256}, {256, 1}: 0.0303835ms, 0.266943% main:@1192 = slice[axes={0, 1, 2},starts={0, 0, 0},ends={32, 1, 256}](main:@713) -> float_type, {32, 1, 256}, {4608, 256, 1}: 0.00074724ms, 0.00656508% main:@1193 = squeeze[axes={1}](main:@1192) -> float_type, {32, 256}, {4608, 1}: 0.00061472ms, 0.00540079% main:@1194 = load[offset=6083840,end=6149376](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00046756ms, 0.00410788% main:@1195 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1193,main:@716,main:@1194) -> float_type, {32, 512}, {512, 1}: 0.0231487ms, 0.203379% main:@1196 = load[offset=4347136,end=4412672](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00060534ms, 0.00531838% main:@1197 = gpu::code_object[code_object=13512,symbol_name=relu_kernel,global=8192,local=256,](main:@1195,main:@1196) -> float_type, {32, 512}, {512, 1}: 0.0143179ms, 0.125794% main:@1198 = hip::hip_copy_literal[id=main:@literal:70] -> float_type, {512, 256}, {256, 1}: 0.00078078ms, 0.00685976% main:@1199 = load[offset=4314368,end=4347136](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00050168ms, 0.00440765% 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, {32, 256}, {256, 1}: 0.0307334ms, 0.270017% main:@1201 = hip::hip_copy_literal[id=main:@literal:86] -> float_type, {256, 512}, {512, 1}: 0.00076256ms, 0.00669968% main:@1202 = slice[axes={0, 1, 2},starts={0, 14, 0},ends={32, 15, 256}](main:@713) -> float_type, {32, 1, 256}, {4608, 256, 1}: 0.00061388ms, 0.00539341% main:@1203 = squeeze[axes={1}](main:@1202) -> float_type, {32, 256}, {4608, 1}: 0.00090368ms, 0.00793953% main:@1204 = load[offset=6083840,end=6149376](main:@1) -> float_type, {32, 512}, {512, 1}: 0.0004688ms, 0.00411877% main:@1205 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1203,main:@1201,main:@1204) -> float_type, {32, 512}, {512, 1}: 0.0239177ms, 0.210135% main:@1206 = load[offset=4379904,end=4445440](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00061108ms, 0.00536881% main:@1207 = gpu::code_object[code_object=13512,symbol_name=relu_kernel,global=8192,local=256,](main:@1205,main:@1206) -> float_type, {32, 512}, {512, 1}: 0.0144304ms, 0.126782% main:@1208 = load[offset=4347136,end=4379904](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00060036ms, 0.00527463% main:@1209 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1207,main:@715,main:@1208) -> float_type, {32, 256}, {256, 1}: 0.0302838ms, 0.266067% main:@1210 = hip::hip_copy_literal[id=main:@literal:141] -> float_type, {256, 512}, {512, 1}: 0.00079552ms, 0.00698926% main:@1211 = slice[axes={0, 1, 2},starts={0, 12, 0},ends={32, 13, 256}](main:@713) -> float_type, {32, 1, 256}, {4608, 256, 1}: 0.00071288ms, 0.0062632% main:@1212 = squeeze[axes={1}](main:@1211) -> float_type, {32, 256}, {4608, 1}: 0.0005692ms, 0.00500086% main:@1213 = load[offset=4379904,end=4445440](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00047766ms, 0.00419661% 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, {32, 512}, {512, 1}: 0.0233276ms, 0.204951% main:@1215 = load[offset=6083840,end=6149376](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00062782ms, 0.00551589% main:@1216 = gpu::code_object[code_object=13512,symbol_name=relu_kernel,global=8192,local=256,](main:@1214,main:@1215) -> float_type, {32, 512}, {512, 1}: 0.0143177ms, 0.125792% main:@1217 = hip::hip_copy_literal[id=main:@literal:113] -> float_type, {512, 256}, {256, 1}: 0.00075136ms, 0.00660128% main:@1218 = load[offset=4379904,end=4412672](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00046126ms, 0.00405253% 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, {32, 256}, {256, 1}: 0.0303576ms, 0.266715% main:@1220 = load[offset=6083840,end=6149376](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00061296ms, 0.00538533% main:@1221 = slice[axes={0, 1, 2},starts={0, 17, 0},ends={32, 18, 256}](main:@713) -> float_type, {32, 1, 256}, {4608, 256, 1}: 0.00061156ms, 0.00537303% main:@1222 = squeeze[axes={1}](main:@1221) -> float_type, {32, 256}, {4608, 1}: 0.00058996ms, 0.00518326% main:@1223 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1222,main:@714,main:@1220) -> float_type, {32, 512}, {512, 1}: 0.0235608ms, 0.207% main:@1224 = load[offset=6149376,end=6214912](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00061472ms, 0.00540079% main:@1225 = gpu::code_object[code_object=13512,symbol_name=relu_kernel,global=8192,local=256,](main:@1223,main:@1224) -> float_type, {32, 512}, {512, 1}: 0.0144163ms, 0.126658% main:@1226 = hip::hip_copy_literal[id=main:@literal:96] -> float_type, {512, 256}, {256, 1}: 0.0008082ms, 0.00710066% main:@1227 = load[offset=4412672,end=4445440](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00045494ms, 0.003997% 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, {32, 256}, {256, 1}: 0.0302352ms, 0.26564% main:@1229 = slice[axes={0, 1, 2},starts={0, 10, 0},ends={32, 11, 256}](main:@713) -> float_type, {32, 1, 256}, {4608, 256, 1}: 0.00075876ms, 0.0066663% main:@1230 = squeeze[axes={1}](main:@1229) -> float_type, {32, 256}, {4608, 1}: 0.0009298ms, 0.00816901% main:@1231 = load[offset=6182144,end=6247680](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00047036ms, 0.00413248% main:@1232 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1230,main:@714,main:@1231) -> float_type, {32, 512}, {512, 1}: 0.0230184ms, 0.202235% main:@1233 = load[offset=6116608,end=6182144](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00064146ms, 0.00563572% main:@1234 = gpu::code_object[code_object=13512,symbol_name=relu_kernel,global=8192,local=256,](main:@1232,main:@1233) -> float_type, {32, 512}, {512, 1}: 0.0143615ms, 0.126177% main:@1235 = hip::hip_copy_literal[id=main:@literal:42] -> float_type, {512, 256}, {256, 1}: 0.00073314ms, 0.0064412% main:@1236 = load[offset=6083840,end=6116608](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00046408ms, 0.0040773% 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, {32, 256}, {256, 1}: 0.0304586ms, 0.267603% main:@1238 = load[offset=6116608,end=6182144](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00061526ms, 0.00540554% main:@1239 = slice[axes={0, 1, 2},starts={0, 13, 0},ends={32, 14, 256}](main:@713) -> float_type, {32, 1, 256}, {4608, 256, 1}: 0.00063378ms, 0.00556825% main:@1240 = squeeze[axes={1}](main:@1239) -> float_type, {32, 256}, {4608, 1}: 0.00059588ms, 0.00523527% main:@1241 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1240,main:@716,main:@1238) -> float_type, {32, 512}, {512, 1}: 0.0236266ms, 0.207578% main:@1242 = load[offset=6182144,end=6247680](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00061816ms, 0.00543102% main:@1243 = gpu::code_object[code_object=13512,symbol_name=relu_kernel,global=8192,local=256,](main:@1241,main:@1242) -> float_type, {32, 512}, {512, 1}: 0.0143018ms, 0.125652% main:@1244 = hip::hip_copy_literal[id=main:@literal:41] -> float_type, {512, 256}, {256, 1}: 0.00075908ms, 0.00666911% main:@1245 = load[offset=6116608,end=6149376](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00048686ms, 0.00427744% 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, {32, 256}, {256, 1}: 0.0299953ms, 0.263532% main:@1247 = hip::hip_copy_literal[id=main:@literal:112] -> float_type, {256, 512}, {512, 1}: 0.00078222ms, 0.00687241% main:@1248 = slice[axes={0, 1, 2},starts={0, 7, 0},ends={32, 8, 256}](main:@713) -> float_type, {32, 1, 256}, {4608, 256, 1}: 0.00059834ms, 0.00525688% main:@1249 = squeeze[axes={1}](main:@1248) -> float_type, {32, 256}, {4608, 1}: 0.0008995ms, 0.00790281% main:@1250 = load[offset=6329600,end=6395136](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00044984ms, 0.00395219% 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, {32, 512}, {512, 1}: 0.023768ms, 0.20882% main:@1252 = load[offset=6182144,end=6247680](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00061762ms, 0.00542627% main:@1253 = gpu::code_object[code_object=13512,symbol_name=relu_kernel,global=8192,local=256,](main:@1251,main:@1252) -> float_type, {32, 512}, {512, 1}: 0.0142336ms, 0.125053% main:@1254 = hip::hip_copy_literal[id=main:@literal:140] -> float_type, {512, 256}, {256, 1}: 0.0007551ms, 0.00663414% main:@1255 = load[offset=6149376,end=6182144](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00045808ms, 0.00402459% 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, {32, 256}, {256, 1}: 0.0305658ms, 0.268544% main:@1257 = hip::hip_copy_literal[id=main:@literal:153] -> float_type, {256, 512}, {512, 1}: 0.00072014ms, 0.00632699% main:@1258 = slice[axes={0, 1, 2},starts={0, 6, 0},ends={32, 7, 256}](main:@713) -> float_type, {32, 1, 256}, {4608, 256, 1}: 0.00095266ms, 0.00836986% main:@1259 = squeeze[axes={1}](main:@1258) -> float_type, {32, 256}, {4608, 1}: 0.00059502ms, 0.00522771% main:@1260 = load[offset=6182144,end=6247680](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00045814ms, 0.00402512% 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, {32, 512}, {512, 1}: 0.0239556ms, 0.210468% main:@1262 = load[offset=6329600,end=6395136](main:@1) -> float_type, {32, 512}, {512, 1}: 0.0006274ms, 0.0055122% main:@1263 = gpu::code_object[code_object=13512,symbol_name=relu_kernel,global=8192,local=256,](main:@1261,main:@1262) -> float_type, {32, 512}, {512, 1}: 0.0142658ms, 0.125336% main:@1264 = hip::hip_copy_literal[id=main:@literal:93] -> float_type, {512, 256}, {256, 1}: 0.0007376ms, 0.00648039% main:@1265 = load[offset=6182144,end=6214912](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00047486ms, 0.00417201% 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, {32, 256}, {256, 1}: 0.0299967ms, 0.263544% main:@1267 = hip::hip_copy_literal[id=main:@literal:68] -> float_type, {256, 512}, {512, 1}: 0.0007832ms, 0.00688102% main:@1268 = load[offset=6329600,end=6395136](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00047858ms, 0.0042047% main:@1269 = slice[axes={0, 1, 2},starts={0, 11, 0},ends={32, 12, 256}](main:@713) -> float_type, {32, 1, 256}, {4608, 256, 1}: 0.00063234ms, 0.0055556% main:@1270 = squeeze[axes={1}](main:@1269) -> float_type, {32, 256}, {4608, 1}: 0.0009051ms, 0.00795201% main:@1271 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1270,main:@1267,main:@1268) -> float_type, {32, 512}, {512, 1}: 0.0234762ms, 0.206257% main:@1272 = load[offset=6395136,end=6460672](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00061048ms, 0.00536354% main:@1273 = gpu::code_object[code_object=13512,symbol_name=relu_kernel,global=8192,local=256,](main:@1271,main:@1272) -> float_type, {32, 512}, {512, 1}: 0.0143326ms, 0.125923% main:@1274 = load[offset=6214912,end=6247680](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00059836ms, 0.00525706% main:@1275 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1273,main:@715,main:@1274) -> float_type, {32, 256}, {256, 1}: 0.030185ms, 0.265198% main:@1276 = slice[axes={0, 1},starts={0, 9},ends={32, 11}](w04Keh2FJu:0) -> float_type, {32, 2}, {13, 1}: 0.00084706ms, 0.00744208% main:@1277 = slice[axes={0, 1},starts={0, 41},ends={32, 45}](2CnSyUBzm0:0) -> float_type, {32, 4}, {65, 1}: 0.0009517ms, 0.00836142% main:@1278 = slice[axes={0, 1},starts={0, 5},ends={32, 13}](9YqhflbMKe:0) -> float_type, {32, 8}, {29, 1}: 0.0008245ms, 0.00724387% main:@1279 = slice[axes={0, 1},starts={0, 501},ends={32, 509}](lyiDxfzZg3:0) -> float_type, {32, 8}, {701, 1}: 0.00062088ms, 0.00545491% main:@1280 = slice[axes={0, 1},starts={0, 181},ends={32, 189}](nv2tzWRf1I:0) -> float_type, {32, 8}, {301, 1}: 0.00059874ms, 0.0052604% main:@1281 = slice[axes={0, 1},starts={0, 33},ends={32, 41}](FcmWLZhDgE:0) -> float_type, {32, 8}, {57, 1}: 0.00094372ms, 0.00829131% main:@1282 = slice[axes={0, 1},starts={0, 21},ends={32, 29}](YbIVz8l1un:0) -> float_type, {32, 8}, {45, 1}: 0.00076742ms, 0.00674238% main:@1283 = slice[axes={0, 1},starts={0, 41},ends={32, 49}](ifP6Ceko5T:0) -> float_type, {32, 8}, {73, 1}: 0.00059238ms, 0.00520452% main:@1284 = slice[axes={0, 1},starts={0, 37},ends={32, 45}](Qp3K1PHvte:0) -> float_type, {32, 8}, {69, 1}: 0.00063744ms, 0.0056004% main:@1285 = slice[axes={0, 1},starts={0, 5},ends={32, 13}](ljyhc7HCLk:0) -> float_type, {32, 8}, {21, 1}: 0.00094378ms, 0.00829184% main:@1286 = slice[axes={0, 1},starts={0, 33},ends={32, 37}](ZcWERoz0h6:0) -> float_type, {32, 4}, {49, 1}: 0.000755ms, 0.00663326% main:@1287 = slice[axes={0, 1},starts={0, 21},ends={32, 25}](bJsIAzhYc3:0) -> float_type, {32, 4}, {37, 1}: 0.00059098ms, 0.00519222% main:@1288 = slice[axes={0, 1},starts={0, 33},ends={32, 37}](isCn4mEOSt:0) -> float_type, {32, 4}, {49, 1}: 0.00058438ms, 0.00513423% main:@1289 = load[offset=4236032,end=4246528](main:@1) -> float_type, {32, 82}, {82, 1}: 0.00047836ms, 0.00420276% main:@1290 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=201,local=256,](main:@1279,main:@1280,main:@1283,main:@1284,main:@1277,main:@1282,main:@1288,main:@1286,main:@1276,main:@1281,main:@1287,main:@1285,main:@1278,main:@1289) -> float_type, {32, 82}, {82, 1}: 0.0280569ms, 0.246501% main:@1291 = hip::hip_copy_literal[id=main:@literal:12] -> float_type, {82, 256}, {256, 1}: 0.00079662ms, 0.00699892% main:@1292 = load[offset=6395136,end=6427904](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00049258ms, 0.0043277% main:@1293 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1290,main:@1291,main:@1292) -> float_type, {32, 256}, {256, 1}: 0.0218059ms, 0.191582% main:@1294 = slice[axes={0, 1, 2},starts={0, 4, 0},ends={32, 5, 256}](main:@713) -> float_type, {32, 1, 256}, {4608, 256, 1}: 0.00075878ms, 0.00666647% main:@1295 = squeeze[axes={1}](main:@1294) -> float_type, {32, 256}, {4608, 1}: 0.00072522ms, 0.00637162% main:@1296 = load[offset=6329600,end=6395136](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00048072ms, 0.0042235% main:@1297 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1295,main:@716,main:@1296) -> float_type, {32, 512}, {512, 1}: 0.0234787ms, 0.206279% main:@1298 = load[offset=6427904,end=6493440](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00061476ms, 0.00540114% main:@1299 = gpu::code_object[code_object=13512,symbol_name=relu_kernel,global=8192,local=256,](main:@1297,main:@1298) -> float_type, {32, 512}, {512, 1}: 0.014344ms, 0.126023% main:@1300 = hip::hip_copy_literal[id=main:@literal:94] -> float_type, {512, 256}, {256, 1}: 0.00070784ms, 0.00621892% main:@1301 = load[offset=6329600,end=6362368](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00046696ms, 0.00410261% 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, {32, 256}, {256, 1}: 0.0309927ms, 0.272295% main:@1303 = hip::hip_copy_literal[id=main:@literal:149] -> float_type, {256, 512}, {512, 1}: 0.00075418ms, 0.00662606% main:@1304 = slice[axes={0, 1, 2},starts={0, 3, 0},ends={32, 4, 256}](main:@713) -> float_type, {32, 1, 256}, {4608, 256, 1}: 0.00067668ms, 0.00594516% main:@1305 = squeeze[axes={1}](main:@1304) -> float_type, {32, 256}, {4608, 1}: 0.0005958ms, 0.00523457% main:@1306 = load[offset=6493440,end=6558976](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00045984ms, 0.00404005% 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, {32, 512}, {512, 1}: 0.0242147ms, 0.212745% main:@1308 = load[offset=6427904,end=6493440](main:@1) -> float_type, {32, 512}, {512, 1}: 0.00061862ms, 0.00543506% main:@1309 = gpu::code_object[code_object=13512,symbol_name=relu_kernel,global=8192,local=256,](main:@1307,main:@1308) -> float_type, {32, 512}, {512, 1}: 0.0143294ms, 0.125895% main:@1310 = load[offset=6362368,end=6395136](main:@1) -> float_type, {32, 256}, {256, 1}: 0.0005989ms, 0.0052618% 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, {32, 256}, {256, 1}: 0.0299253ms, 0.262917% main:@1312 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](09zp4EDXTh:0) -> float_type, {32, 8}, {25, 1}: 0.00083738ms, 0.00735703% main:@1313 = slice[axes={0, 1},starts={0, 169},ends={32, 177}](uFSOc5Chp2:0) -> float_type, {32, 8}, {481, 1}: 0.00068572ms, 0.00602458% main:@1314 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](1PJsBhDaCK:0) -> float_type, {32, 8}, {25, 1}: 0.0007605ms, 0.00668158% main:@1315 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](1u5yIsHTXE:0) -> float_type, {32, 4}, {17, 1}: 0.00064798ms, 0.00569301% main:@1316 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](2lVXMZgJfO:0) -> float_type, {32, 8}, {25, 1}: 0.0006423ms, 0.0056431% main:@1317 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](3kdi0vsSZG:0) -> float_type, {32, 8}, {25, 1}: 0.0006371ms, 0.00559742% main:@1318 = slice[axes={0, 1},starts={0, 25},ends={32, 33}](Qp3K1PHvte:0) -> float_type, {32, 8}, {69, 1}: 0.0007509ms, 0.00659724% main:@1319 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](5JHUPChra8:0) -> float_type, {32, 2}, {17, 1}: 0.00060276ms, 0.00529571% main:@1320 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](RpITWntAji:0) -> float_type, {32, 8}, {17, 1}: 0.00058936ms, 0.00517798% main:@1321 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](92MwiALCHX:0) -> float_type, {32, 8}, {25, 1}: 0.00062206ms, 0.00546528% main:@1322 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](AVLcBpWeuo:0) -> float_type, {32, 4}, {21, 1}: 0.00079326ms, 0.0069694% main:@1323 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](JSWhLK4Agx:0) -> float_type, {32, 8}, {25, 1}: 0.00059918ms, 0.00526426% main:@1324 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](o74vNAnXxy:0) -> float_type, {32, 4}, {17, 1}: 0.00058944ms, 0.00517869% main:@1325 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](Tg4u0GfnFb:0) -> float_type, {32, 8}, {25, 1}: 0.00058236ms, 0.00511648% main:@1326 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](TidKGrnobI:0) -> float_type, {32, 2}, {13, 1}: 0.00074652ms, 0.00655876% main:@1327 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](HZ0l3MmcPY:0) -> float_type, {32, 4}, {17, 1}: 0.00066168ms, 0.00581337% main:@1328 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](IztrlRfWQU:0) -> float_type, {32, 2}, {13, 1}: 0.00062124ms, 0.00545808% main:@1329 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](KN13IWXZSu:0) -> float_type, {32, 2}, {13, 1}: 0.00059348ms, 0.00521418% main:@1330 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](RHo9Jz6KpI:0) -> float_type, {32, 4}, {21, 1}: 0.00070726ms, 0.00621383% main:@1331 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](NXOSdQVPs0:0) -> float_type, {32, 8}, {25, 1}: 0.00060812ms, 0.00534281% main:@1332 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](ry2xad1vc9:0) -> float_type, {32, 8}, {25, 1}: 0.00060234ms, 0.00529202% main:@1333 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](S5D6WGIXMd:0) -> float_type, {32, 4}, {17, 1}: 0.00075846ms, 0.00666366% main:@1334 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](1WNvHElUG7:0) -> float_type, {32, 8}, {25, 1}: 0.00064918ms, 0.00570355% main:@1335 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](YbIVz8l1un:0) -> float_type, {32, 8}, {45, 1}: 0.00065896ms, 0.00578947% main:@1336 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](bJsIAzhYc3:0) -> float_type, {32, 8}, {37, 1}: 0.00063978ms, 0.00562096% main:@1337 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](pH5srbufJ1:0) -> float_type, {32, 4}, {17, 1}: 0.0006616ms, 0.00581267% main:@1338 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](tWcngAq7hV:0) -> float_type, {32, 2}, {13, 1}: 0.00066644ms, 0.00585519% main:@1339 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](xymEBUMS25:0) -> float_type, {32, 2}, {13, 1}: 0.00065786ms, 0.00577981% main:@1340 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](msiPzSBKMj:0) -> float_type, {32, 4}, {17, 1}: 0.0006434ms, 0.00565277% main:@1341 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](xT9FhjM1Uu:0) -> float_type, {32, 4}, {17, 1}: 0.00062982ms, 0.00553346% main:@1342 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](m2YscaoDjP:0) -> float_type, {32, 4}, {17, 1}: 0.00061078ms, 0.00536618% main:@1343 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](zZHaVOckpw:0) -> float_type, {32, 2}, {13, 1}: 0.00062802ms, 0.00551764% main:@1344 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](qd1komWTn4:0) -> float_type, {32, 8}, {25, 1}: 0.00061844ms, 0.00543348% main:@1345 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](ujOA5LQW7M:0) -> float_type, {32, 2}, {13, 1}: 0.0006594ms, 0.00579334% main:@1346 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](nPUSZFm74Y:0) -> float_type, {32, 2}, {13, 1}: 0.00060374ms, 0.00530432% main:@1347 = slice[axes={0, 1},starts={0, 163},ends={32, 165}](nv2tzWRf1I:0) -> float_type, {32, 2}, {301, 1}: 0.00064608ms, 0.00567631% main:@1348 = slice[axes={0, 1},starts={0, 21},ends={32, 25}](ifP6Ceko5T:0) -> float_type, {32, 4}, {73, 1}: 0.00063268ms, 0.00555858% main:@1349 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](XxirNgB98u:0) -> float_type, {32, 2}, {13, 1}: 0.00063402ms, 0.00557036% main:@1350 = load[offset=6452480,end=6477312](main:@1) -> float_type, {32, 194}, {194, 1}: 0.00048642ms, 0.00427358% main:@1351 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=163,local=256,](main:@1319,main:@1349,main:@1338,main:@1339,main:@1328,main:@1329,main:@1346,main:@1345,main:@1343,main:@1326,main:@1347,main:@1327,main:@1337,main:@1340,main:@1315,main:@1330,main:@1333,main:@1342,main:@1322,main:@1324,main:@1341,main:@1348,main:@1312,main:@1314,main:@1318,main:@1323,main:@1344,main:@1332,main:@1316,main:@1321,main:@1325,main:@1334,main:@1331,main:@1336,main:@1317,main:@1335,main:@1320,main:@1313,main:@1350) -> float_type, {32, 194}, {194, 1}: 0.0481461ms, 0.423% main:@1352 = hip::hip_copy_literal[id=main:@literal:106] -> float_type, {194, 192}, {192, 1}: 0.00089196ms, 0.00783656% main:@1353 = load[offset=6427904,end=6452480](main:@1) -> float_type, {32, 192}, {192, 1}: 0.00047292ms, 0.00415497% 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, {32, 192}, {192, 1}: 0.0232435ms, 0.204212% main:@1355 = load[offset=6452480,end=6474240](main:@1) -> float_type, {32, 170}, {170, 1}: 0.00063512ms, 0.00558002% vAbKepTjQg:0 = @param:vAbKepTjQg:0 -> float_type, {32, 5}, {5, 1}: 0.00061702ms, 0.005421% main:@1357 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](vAbKepTjQg:0) -> float_type, {32, 2}, {5, 1}: 0.00069106ms, 0.0060715% lDhFNYzbxR:0 = @param:lDhFNYzbxR:0 -> float_type, {32, 17}, {17, 1}: 0.00060644ms, 0.00532805% main:@1359 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](lDhFNYzbxR:0) -> float_type, {32, 8}, {17, 1}: 0.0007622ms, 0.00669652% 1QWFmJ96aB:0 = @param:1QWFmJ96aB:0 -> float_type, {32, 5}, {5, 1}: 0.00053386ms, 0.00469037% main:@1361 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](1QWFmJ96aB:0) -> float_type, {32, 2}, {5, 1}: 0.00063906ms, 0.00561464% 3VED7WSJ1b:0 = @param:3VED7WSJ1b:0 -> float_type, {32, 5}, {5, 1}: 0.00055024ms, 0.00483429% main:@1363 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](3VED7WSJ1b:0) -> float_type, {32, 2}, {5, 1}: 0.00076928ms, 0.00675872% gdck97AK4w:0 = @param:gdck97AK4w:0 -> float_type, {32, 17}, {17, 1}: 0.00054578ms, 0.0047951% main:@1365 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](gdck97AK4w:0) -> float_type, {32, 4}, {17, 1}: 0.00060902ms, 0.00535071% tIGjbcEuDd:0 = @param:tIGjbcEuDd:0 -> float_type, {32, 9}, {9, 1}: 0.00046836ms, 0.00411491% main:@1367 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](tIGjbcEuDd:0) -> float_type, {32, 4}, {9, 1}: 0.00075364ms, 0.00662131% NJBt1ubXFW:0 = @param:NJBt1ubXFW:0 -> float_type, {32, 5}, {5, 1}: 0.0005337ms, 0.00468897% main:@1369 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](NJBt1ubXFW:0) -> float_type, {32, 2}, {5, 1}: 0.00078186ms, 0.00686925% zX6xKOWG9S:0 = @param:zX6xKOWG9S:0 -> float_type, {32, 5}, {5, 1}: 0.00050832ms, 0.00446599% main:@1371 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](zX6xKOWG9S:0) -> float_type, {32, 2}, {5, 1}: 0.0007648ms, 0.00671936% 5yColzZ8K3:0 = @param:5yColzZ8K3:0 -> float_type, {32, 5}, {5, 1}: 0.00056144ms, 0.00493269% main:@1373 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](5yColzZ8K3:0) -> float_type, {32, 2}, {5, 1}: 0.0006107ms, 0.00536547% IxpbBg7JTO:0 = @param:IxpbBg7JTO:0 -> float_type, {32, 5}, {5, 1}: 0.000529ms, 0.00464768% main:@1375 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](IxpbBg7JTO:0) -> float_type, {32, 2}, {5, 1}: 0.00075876ms, 0.0066663% 6Wh2Jm7GHC:0 = @param:6Wh2Jm7GHC:0 -> float_type, {32, 5}, {5, 1}: 0.00055168ms, 0.00484694% main:@1377 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](6Wh2Jm7GHC:0) -> float_type, {32, 2}, {5, 1}: 0.00066018ms, 0.00580019% UMO0o14RuI:0 = @param:UMO0o14RuI:0 -> float_type, {32, 9}, {9, 1}: 0.00054304ms, 0.00477103% main:@1379 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](UMO0o14RuI:0) -> float_type, {32, 4}, {9, 1}: 0.00075042ms, 0.00659302% IMdFk7D16X:0 = @param:IMdFk7D16X:0 -> float_type, {32, 9}, {9, 1}: 0.0005589ms, 0.00491037% main:@1381 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](IMdFk7D16X:0) -> float_type, {32, 4}, {9, 1}: 0.00061118ms, 0.00536969% Pu1O5qFxY6:0 = @param:Pu1O5qFxY6:0 -> float_type, {32, 17}, {17, 1}: 0.00052356ms, 0.00459988% main:@1383 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](Pu1O5qFxY6:0) -> float_type, {32, 4}, {17, 1}: 0.0007612ms, 0.00668773% Cw59eOo7M1:0 = @param:Cw59eOo7M1:0 -> float_type, {32, 9}, {9, 1}: 0.0005656ms, 0.00496923% main:@1385 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](Cw59eOo7M1:0) -> float_type, {32, 4}, {9, 1}: 0.00093368ms, 0.0082031% E1ZkVIXqS5:0 = @param:E1ZkVIXqS5:0 -> float_type, {32, 9}, {9, 1}: 0.00051048ms, 0.00448496% main:@1387 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](E1ZkVIXqS5:0) -> float_type, {32, 4}, {9, 1}: 0.00072122ms, 0.00633648% EhUjmCTaGW:0 = @param:EhUjmCTaGW:0 -> float_type, {32, 9}, {9, 1}: 0.00047832ms, 0.00420241% main:@1389 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](EhUjmCTaGW:0) -> float_type, {32, 4}, {9, 1}: 0.00062098ms, 0.00545579% l5WSX3VvMK:0 = @param:l5WSX3VvMK:0 -> float_type, {32, 25}, {25, 1}: 0.00050836ms, 0.00446634% main:@1391 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](l5WSX3VvMK:0) -> float_type, {32, 8}, {25, 1}: 0.00064436ms, 0.0056612% FWhBxj4Tun:0 = @param:FWhBxj4Tun:0 -> float_type, {32, 9}, {9, 1}: 0.00052492ms, 0.00461183% main:@1393 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](FWhBxj4Tun:0) -> float_type, {32, 4}, {9, 1}: 0.00072774ms, 0.00639376% HmATr3gKsW:0 = @param:HmATr3gKsW:0 -> float_type, {32, 9}, {9, 1}: 0.00053486ms, 0.00469916% main:@1395 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](HmATr3gKsW:0) -> float_type, {32, 4}, {9, 1}: 0.00069044ms, 0.00606605% Ouh1XpPZaj:0 = @param:Ouh1XpPZaj:0 -> float_type, {32, 9}, {9, 1}: 0.00052628ms, 0.00462378% main:@1397 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](Ouh1XpPZaj:0) -> float_type, {32, 4}, {9, 1}: 0.00070172ms, 0.00616515% J3H2EjAXt8:0 = @param:J3H2EjAXt8:0 -> float_type, {32, 5}, {5, 1}: 0.00055622ms, 0.00488682% main:@1399 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](J3H2EjAXt8:0) -> float_type, {32, 2}, {5, 1}: 0.00070992ms, 0.0062372% NLt5JnRMjG:0 = @param:NLt5JnRMjG:0 -> float_type, {32, 5}, {5, 1}: 0.00054276ms, 0.00476857% main:@1401 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](NLt5JnRMjG:0) -> float_type, {32, 2}, {5, 1}: 0.0010336ms, 0.00908098% KCNfS4GlpP:0 = @param:KCNfS4GlpP:0 -> float_type, {32, 9}, {9, 1}: 0.000465ms, 0.00408539% main:@1403 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](KCNfS4GlpP:0) -> float_type, {32, 4}, {9, 1}: 0.0007052ms, 0.00619573% O5aJeUQh2G:0 = @param:O5aJeUQh2G:0 -> float_type, {32, 17}, {17, 1}: 0.00056244ms, 0.00494147% main:@1405 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](O5aJeUQh2G:0) -> float_type, {32, 8}, {17, 1}: 0.00064706ms, 0.00568492% nltbg9ZWQX:0 = @param:nltbg9ZWQX:0 -> float_type, {32, 17}, {17, 1}: 0.00060802ms, 0.00534193% main:@1407 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](nltbg9ZWQX:0) -> float_type, {32, 4}, {17, 1}: 0.00071794ms, 0.00630766% Uc42tiuavR:0 = @param:Uc42tiuavR:0 -> float_type, {32, 9}, {9, 1}: 0.00052466ms, 0.00460955% main:@1409 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](Uc42tiuavR:0) -> float_type, {32, 4}, {9, 1}: 0.00064616ms, 0.00567702% U6GqhP8k7I:0 = @param:U6GqhP8k7I:0 -> float_type, {32, 9}, {9, 1}: 0.00055136ms, 0.00484413% main:@1411 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](U6GqhP8k7I:0) -> float_type, {32, 4}, {9, 1}: 0.00065032ms, 0.00571357% oQ5OcjZJe8:0 = @param:oQ5OcjZJe8:0 -> float_type, {32, 17}, {17, 1}: 0.00056426ms, 0.00495746% main:@1413 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](oQ5OcjZJe8:0) -> float_type, {32, 8}, {17, 1}: 0.00068456ms, 0.00601439% yVxKAL4U5G:0 = @param:yVxKAL4U5G:0 -> float_type, {32, 9}, {9, 1}: 0.00059728ms, 0.00524757% main:@1415 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](yVxKAL4U5G:0) -> float_type, {32, 4}, {9, 1}: 0.00064698ms, 0.00568422% ZRYNFfjSez:0 = @param:ZRYNFfjSez:0 -> float_type, {32, 33}, {33, 1}: 0.00051976ms, 0.00456649% main:@1417 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](ZRYNFfjSez:0) -> float_type, {32, 8}, {33, 1}: 0.00097564ms, 0.00857175% ecPTQUALkq:0 = @param:ecPTQUALkq:0 -> float_type, {32, 5}, {5, 1}: 0.0005056ms, 0.00444209% main:@1419 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](ecPTQUALkq:0) -> float_type, {32, 2}, {5, 1}: 0.00065142ms, 0.00572323% magwfJSnUu:0 = @param:magwfJSnUu:0 -> float_type, {32, 9}, {9, 1}: 0.00052842ms, 0.00464258% main:@1421 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](magwfJSnUu:0) -> float_type, {32, 4}, {9, 1}: 0.0006435ms, 0.00565365% mNgQ4ohWqU:0 = @param:mNgQ4ohWqU:0 -> float_type, {32, 9}, {9, 1}: 0.00045768ms, 0.00402107% main:@1423 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](mNgQ4ohWqU:0) -> float_type, {32, 4}, {9, 1}: 0.00065134ms, 0.00572253% gsP96ZTbn0:0 = @param:gsP96ZTbn0:0 -> float_type, {32, 17}, {17, 1}: 0.00051758ms, 0.00454734% main:@1425 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](gsP96ZTbn0:0) -> float_type, {32, 8}, {17, 1}: 0.0006809ms, 0.00598223% gqk5byC8aJ:0 = @param:gqk5byC8aJ:0 -> float_type, {32, 5}, {5, 1}: 0.00051236ms, 0.00450148% main:@1427 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](gqk5byC8aJ:0) -> float_type, {32, 2}, {5, 1}: 0.0006826ms, 0.00599717% mTMtDEo3r8:0 = @param:mTMtDEo3r8:0 -> float_type, {32, 9}, {9, 1}: 0.00060602ms, 0.00532436% main:@1429 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](mTMtDEo3r8:0) -> float_type, {32, 4}, {9, 1}: 0.00068086ms, 0.00598188% xVErSbJ0hg:0 = @param:xVErSbJ0hg:0 -> float_type, {32, 9}, {9, 1}: 0.00055362ms, 0.00486398% main:@1431 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](xVErSbJ0hg:0) -> float_type, {32, 4}, {9, 1}: 0.00065276ms, 0.005735% qfwYOHde3u:0 = @param:qfwYOHde3u:0 -> float_type, {32, 17}, {17, 1}: 0.00051758ms, 0.00454734% main:@1433 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](qfwYOHde3u:0) -> float_type, {32, 8}, {17, 1}: 0.001006ms, 0.00883849% pQsi1GNxvj:0 = @param:pQsi1GNxvj:0 -> float_type, {32, 5}, {5, 1}: 0.00055652ms, 0.00488946% main:@1435 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](pQsi1GNxvj:0) -> float_type, {32, 2}, {5, 1}: 0.00070358ms, 0.0061815% l4bReWtsMB:0 = @param:l4bReWtsMB:0 -> float_type, {32, 25}, {25, 1}: 0.0004783ms, 0.00420224% main:@1437 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](l4bReWtsMB:0) -> float_type, {32, 8}, {25, 1}: 0.00063752ms, 0.00560111% main:@1438 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=132,local=256,](main:@1377,main:@1419,main:@1369,main:@1357,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:@1355) -> float_type, {32, 170}, {170, 1}: 0.0478635ms, 0.420518% main:@1439 = hip::hip_copy_literal[id=main:@literal:73] -> float_type, {170, 160}, {160, 1}: 0.00082886ms, 0.00728218% main:@1440 = load[offset=6489344,end=6509824](main:@1) -> float_type, {32, 160}, {160, 1}: 0.0005163ms, 0.0045361% 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, {32, 160}, {160, 1}: 0.0236747ms, 0.208% main:@1442 = load[offset=6452480,end=6489344](main:@1) -> float_type, {32, 18, 16}, {288, 16, 1}: 0.000633ms, 0.0055614% main:@1443 = slice[axes={0, 1},starts={0, 25},ends={32, 41}](TxEdCcekGS:0) -> float_type, {32, 16}, {49, 1}: 0.00070322ms, 0.00617833% main:@1444 = unsqueeze[axes={1},steps={}](main:@1443) -> float_type, {32, 1, 16}, {49, 16, 1}: 0.0008651ms, 0.00760057% 2Yy3zxNGBC:0 = @param:2Yy3zxNGBC:0 -> float_type, {32, 45}, {45, 1}: 0.00050548ms, 0.00444103% main:@1446 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](2Yy3zxNGBC:0) -> float_type, {32, 16}, {45, 1}: 0.00075836ms, 0.00666278% main:@1447 = unsqueeze[axes={1},steps={}](main:@1446) -> float_type, {32, 1, 16}, {45, 16, 1}: 0.00059466ms, 0.00522455% main:@1448 = slice[axes={0, 1},starts={0, 25},ends={32, 41}](LKAvFme9ZS:0) -> float_type, {32, 16}, {49, 1}: 0.00062804ms, 0.00551782% main:@1449 = unsqueeze[axes={1},steps={}](main:@1448) -> float_type, {32, 1, 16}, {49, 16, 1}: 0.00056966ms, 0.00500491% WDLoyKsiA7:0 = @param:WDLoyKsiA7:0 -> float_type, {32, 33}, {33, 1}: 0.00057832ms, 0.00508099% main:@1451 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](WDLoyKsiA7:0) -> float_type, {32, 16}, {33, 1}: 0.00075544ms, 0.00663713% main:@1452 = unsqueeze[axes={1},steps={}](main:@1451) -> float_type, {32, 1, 16}, {33, 16, 1}: 0.00055616ms, 0.0048863% z3lqgZM8WU:0 = @param:z3lqgZM8WU:0 -> float_type, {32, 41}, {41, 1}: 0.00050716ms, 0.00445579% main:@1454 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](z3lqgZM8WU:0) -> float_type, {32, 16}, {41, 1}: 0.000627ms, 0.00550868% main:@1455 = unsqueeze[axes={1},steps={}](main:@1454) -> float_type, {32, 1, 16}, {41, 16, 1}: 0.00054876ms, 0.00482128% pY0t1rI5WU:0 = @param:pY0t1rI5WU:0 -> float_type, {32, 41}, {41, 1}: 0.0005194ms, 0.00456333% main:@1457 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](pY0t1rI5WU:0) -> float_type, {32, 16}, {41, 1}: 0.00073616ms, 0.00646774% main:@1458 = unsqueeze[axes={1},steps={}](main:@1457) -> float_type, {32, 1, 16}, {41, 16, 1}: 0.00056964ms, 0.00500473% sZdFJI9nwg:0 = @param:sZdFJI9nwg:0 -> float_type, {32, 41}, {41, 1}: 0.00056408ms, 0.00495588% main:@1460 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](sZdFJI9nwg:0) -> float_type, {32, 16}, {41, 1}: 0.00064304ms, 0.00564961% main:@1461 = unsqueeze[axes={1},steps={}](main:@1460) -> float_type, {32, 1, 16}, {41, 16, 1}: 0.00054532ms, 0.00479106% VfFM90O7LN:0 = @param:VfFM90O7LN:0 -> float_type, {32, 41}, {41, 1}: 0.0005883ms, 0.00516867% main:@1463 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](VfFM90O7LN:0) -> float_type, {32, 16}, {41, 1}: 0.00078276ms, 0.00687715% main:@1464 = unsqueeze[axes={1},steps={}](main:@1463) -> float_type, {32, 1, 16}, {41, 16, 1}: 0.00057782ms, 0.0050766% main:@1465 = reshape_lazy[dims={-1, 10, 16}](main:@1441) -> float_type, {32, 10, 16}, {160, 16, 1}: 0.00056074ms, 0.00492654% main:@1466 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=1024,local=256,](main:@1452,main:@1455,main:@1449,main:@1458,main:@1444,main:@1461,main:@1464,main:@1447,main:@1465,main:@1442) -> float_type, {32, 18, 16}, {288, 16, 1}: 0.0251395ms, 0.22087% main:@1467 = hip::hip_copy_literal[id=main:@literal:133] -> float_type, {18, 4}, {4, 1}: 0.0007881ms, 0.00692407% main:@1468 = transpose[permutation={0, 2, 1}](main:@1466) -> float_type, {32, 16, 18}, {288, 1, 16}: 0.00058516ms, 0.00514108% main:@1469 = load[offset=6489344,end=6526208](main:@1) -> float_type, {32, 16, 18}, {288, 18, 1}: 0.0005064ms, 0.00444912% main:@1470 = gpu::code_object[code_object=13528,symbol_name=contiguous_kernel,global=9216,local=256,](main:@1468,main:@1469) -> float_type, {32, 16, 18}, {288, 18, 1}: 0.0146545ms, 0.128751% main:@1471 = load[offset=6452480,end=6460672](main:@1) -> float_type, {512, 4}, {4, 1}: 0.00060666ms, 0.00532998% main:@1472 = reshape_lazy[dims={512, 18}](main:@1470) -> float_type, {512, 18}, {18, 1}: 0.00053844ms, 0.00473061% main:@1473 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1472,main:@1467,main:@1471) -> float_type, {512, 4}, {4, 1}: 0.0206456ms, 0.181387% main:@1474 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](2CnSyUBzm0:0) -> float_type, {32, 16}, {65, 1}: 0.00082514ms, 0.00724949% main:@1475 = unsqueeze[axes={1},steps={}](main:@1474) -> float_type, {32, 1, 16}, {65, 16, 1}: 0.00100362ms, 0.00881758% main:@1476 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](zIoqYe9cRb:0) -> float_type, {32, 16}, {45, 1}: 0.00076522ms, 0.00672305% main:@1477 = unsqueeze[axes={1},steps={}](main:@1476) -> float_type, {32, 1, 16}, {45, 16, 1}: 0.00055666ms, 0.00489069% main:@1478 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](mgnGNH4Mz3:0) -> float_type, {32, 16}, {49, 1}: 0.00063464ms, 0.0055758% main:@1479 = unsqueeze[axes={1},steps={}](main:@1478) -> float_type, {32, 1, 16}, {49, 16, 1}: 0.0009749ms, 0.00856525% main:@1480 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](6HSQqXdJzb:0) -> float_type, {32, 16}, {45, 1}: 0.00075788ms, 0.00665856% main:@1481 = unsqueeze[axes={1},steps={}](main:@1480) -> float_type, {32, 1, 16}, {45, 16, 1}: 0.00055286ms, 0.0048573% main:@1482 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](isCn4mEOSt:0) -> float_type, {32, 16}, {49, 1}: 0.00061954ms, 0.00544314% main:@1483 = unsqueeze[axes={1},steps={}](main:@1482) -> float_type, {32, 1, 16}, {49, 16, 1}: 0.00092888ms, 0.00816093% main:@1484 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](Dy3rwVmn1A:0) -> float_type, {32, 16}, {41, 1}: 0.00075372ms, 0.00662201% main:@1485 = unsqueeze[axes={1},steps={}](main:@1484) -> float_type, {32, 1, 16}, {41, 16, 1}: 0.00055816ms, 0.00490387% main:@1486 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](FcmWLZhDgE:0) -> float_type, {32, 16}, {57, 1}: 0.0006194ms, 0.00544191% main:@1487 = unsqueeze[axes={1},steps={}](main:@1486) -> float_type, {32, 1, 16}, {57, 16, 1}: 0.00091842ms, 0.00806903% main:@1488 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](ahAlEtBvzo:0) -> float_type, {32, 16}, {41, 1}: 0.00078352ms, 0.00688383% main:@1489 = unsqueeze[axes={1},steps={}](main:@1488) -> float_type, {32, 1, 16}, {41, 16, 1}: 0.0005671ms, 0.00498241% main:@1490 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](dAFYcGrfEl:0) -> float_type, {32, 16}, {45, 1}: 0.00060948ms, 0.00535475% main:@1491 = unsqueeze[axes={1},steps={}](main:@1490) -> float_type, {32, 1, 16}, {45, 16, 1}: 0.00091286ms, 0.00802018% main:@1492 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](X29iyIeu8L:0) -> float_type, {32, 16}, {45, 1}: 0.00072866ms, 0.00640184% main:@1493 = unsqueeze[axes={1},steps={}](main:@1492) -> float_type, {32, 1, 16}, {45, 16, 1}: 0.00055404ms, 0.00486767% main:@1494 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](ZcWERoz0h6:0) -> float_type, {32, 16}, {49, 1}: 0.00062112ms, 0.00545702% main:@1495 = unsqueeze[axes={1},steps={}](main:@1494) -> float_type, {32, 1, 16}, {49, 16, 1}: 0.00094716ms, 0.00832154% main:@1496 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](pYw7fMVoQg:0) -> float_type, {32, 16}, {41, 1}: 0.00069806ms, 0.006133% main:@1497 = unsqueeze[axes={1},steps={}](main:@1496) -> float_type, {32, 1, 16}, {41, 16, 1}: 0.00056746ms, 0.00498558% main:@1498 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](1aXQU27NrD:0) -> float_type, {32, 16}, {49, 1}: 0.00068602ms, 0.00602722% main:@1499 = unsqueeze[axes={1},steps={}](main:@1498) -> float_type, {32, 1, 16}, {49, 16, 1}: 0.0009062ms, 0.00796167% main:@1500 = load[offset=6460672,end=6511872](main:@1) -> float_type, {32, 25, 16}, {400, 16, 1}: 0.00050574ms, 0.00444332% main:@1501 = reshape_lazy[dims={-1, 12, 16}](main:@1354) -> float_type, {32, 12, 16}, {192, 16, 1}: 0.00048876ms, 0.00429414% main:@1502 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=914,local=256,](main:@1475,main:@1479,main:@1499,main:@1481,main:@1477,main:@1491,main:@1493,main:@1495,main:@1489,main:@1487,main:@1497,main:@1485,main:@1483,main:@1501,main:@1500) -> float_type, {32, 25, 16}, {400, 16, 1}: 0.0281801ms, 0.247584% main:@1503 = load[offset=4236032,end=4248832](main:@1) -> float_type, {32, 25, 4}, {100, 4, 1}: 0.00065928ms, 0.00579229% main:@1504 = reshape_lazy[dims={32, 16, 4}](main:@1473) -> float_type, {32, 16, 4}, {64, 4, 1}: 0.00050286ms, 0.00441802% 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, {32, 25, 4}, {100, 4, 1}: 0.0200486ms, 0.176143% main:@1506 = slice[axes={1},starts={128},ends={256}](main:@748) -> float_type, {1600, 128}, {256, 1}: 0.0007807ms, 0.00685906% main:@1507 = reshape_lazy[dims={32, 50, 128}](main:@1506) -> float_type, {32, 50, 128}, {12800, 256, 1}: 0.00075504ms, 0.00663361% main:@1508 = slice[axes={2},starts={0},ends={64}](main:@1507) -> float_type, {32, 50, 64}, {12800, 256, 1}: 0.00070962ms, 0.00623456% main:@1509 = unsqueeze[axes={0},steps={}](main:@1508) -> float_type, {1, 32, 50, 64}, {409600, 12800, 256, 1}: 0.00100954ms, 0.00886959% main:@1510 = slice[axes={2},starts={64},ends={128}](main:@1507) -> float_type, {32, 50, 64}, {12800, 256, 1}: 0.00067002ms, 0.00588665% main:@1511 = unsqueeze[axes={0},steps={}](main:@1510) -> float_type, {1, 32, 50, 64}, {409600, 12800, 256, 1}: 0.0009632ms, 0.00846246% main:@1512 = load[offset=6427904,end=7247104](main:@1) -> float_type, {2, 32, 50, 64}, {102400, 3200, 64, 1}: 0.00047716ms, 0.00419222% main:@1513 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=51200,local=256,](main:@1509,main:@1511,main:@1512) -> float_type, {2, 32, 50, 64}, {102400, 3200, 64, 1}: 0.0195332ms, 0.171614% main:@1514 = load[offset=7247104,end=7263488](main:@1) -> float_type, {2, 32, 1, 64}, {2048, 64, 64, 1}: 0.00064156ms, 0.0056366% main:@1515 = slice[axes={1},starts={128},ends={256}](main:@1293) -> float_type, {32, 128}, {256, 1}: 0.00060448ms, 0.00531083% main:@1516 = reshape_lazy[dims={32, 1, 128}](main:@1515) -> float_type, {32, 1, 128}, {256, 128, 1}: 0.00066032ms, 0.00580142% main:@1517 = slice[axes={2},starts={0},ends={64}](main:@1516) -> float_type, {32, 1, 64}, {256, 128, 1}: 0.00101616ms, 0.00892775% main:@1518 = unsqueeze[axes={0},steps={}](main:@1517) -> float_type, {1, 32, 1, 64}, {8192, 256, 128, 1}: 0.0006494ms, 0.00570548% main:@1519 = slice[axes={2},starts={64},ends={128}](main:@1516) -> float_type, {32, 1, 64}, {256, 128, 1}: 0.00072558ms, 0.00637478% main:@1520 = unsqueeze[axes={0},steps={}](main:@1519) -> float_type, {1, 32, 1, 64}, {8192, 256, 128, 1}: 0.00063834ms, 0.00560831% main:@1521 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=1024,local=256,](main:@1518,main:@1520,main:@1514) -> float_type, {2, 32, 1, 64}, {2048, 64, 64, 1}: 0.0135496ms, 0.119044% main:@1522 = transpose[permutation={0, 1, 3, 2}](main:@1513) -> float_type, {2, 32, 64, 50}, {102400, 3200, 1, 64}: 0.0006921ms, 0.00608064% main:@1523 = load[offset=7263488,end=7276288](main:@1) -> float_type, {2, 32, 1, 50}, {1600, 50, 50, 1}: 0.00049834ms, 0.0043783% main:@1524 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1521,main:@1522,main:@1523) -> float_type, {2, 32, 1, 50}, {1600, 50, 50, 1}: 0.0221706ms, 0.194786% main:@1525 = load[offset=6427904,end=6434304](main:@1) -> float_type, {32, 1, 50}, {50, 50, 1}: 0.00065652ms, 0.00576804% WjNeBZroCJ:0 = @param:WjNeBZroCJ:0 -> int32_type, {32, 50}, {50, 1}: 0.0005704ms, 0.00501141% main:@1527 = unsqueeze[axes={1},steps={}](WjNeBZroCJ:0) -> int32_type, {32, 1, 50}, {50, 50, 1}: 0.000614ms, 0.00539447% main:@1528 = gpu::code_object[code_object=13536,symbol_name=convert_pow_add_kernel,global=800,local=256,](main:@1527,main:@1525) -> float_type, {32, 1, 50}, {50, 50, 1}: 0.0127999ms, 0.112457% main:@1529 = multibroadcast[out_lens={2, 32, 1, 50}](main:@1528) -> float_type, {2, 32, 1, 50}, {0, 50, 50, 1}: 0.00072826ms, 0.00639833% main:@1530 = load[offset=6440704,end=6453504](main:@1) -> float_type, {2, 32, 1, 50}, {1600, 50, 50, 1}: 0.00048216ms, 0.00423615% main:@1531 = gpu::code_object[code_object=13520,symbol_name=pow_mul_kernel,global=1600,local=256,](main:@1524,main:@1529,main:@1530) -> float_type, {2, 32, 1, 50}, {1600, 50, 50, 1}: 0.01531ms, 0.13451% main:@1532 = load[offset=6427904,end=6440704](main:@1) -> float_type, {2, 32, 1, 50}, {1600, 50, 50, 1}: 0.00061932ms, 0.00544121% main:@1533 = gpu::code_object[code_object=14584,symbol_name=softmax_kernel,global=4096,local=64,](main:@1531,main:@1532) -> float_type, {2, 32, 1, 50}, {1600, 50, 50, 1}: 0.0158747ms, 0.139472% main:@1534 = slice[axes={1},starts={0},ends={128}](main:@748) -> float_type, {1600, 128}, {256, 1}: 0.00074062ms, 0.00650692% main:@1535 = reshape_lazy[dims={32, 50, 128}](main:@1534) -> float_type, {32, 50, 128}, {12800, 256, 1}: 0.00061928ms, 0.00544086% main:@1536 = slice[axes={2},starts={0},ends={64}](main:@1535) -> float_type, {32, 50, 64}, {12800, 256, 1}: 0.00071514ms, 0.00628306% main:@1537 = unsqueeze[axes={0},steps={}](main:@1536) -> float_type, {1, 32, 50, 64}, {409600, 12800, 256, 1}: 0.0009679ms, 0.00850375% main:@1538 = slice[axes={2},starts={64},ends={128}](main:@1535) -> float_type, {32, 50, 64}, {12800, 256, 1}: 0.00063974ms, 0.00562061% main:@1539 = unsqueeze[axes={0},steps={}](main:@1538) -> float_type, {1, 32, 50, 64}, {409600, 12800, 256, 1}: 0.00093764ms, 0.00823789% main:@1540 = load[offset=6440704,end=7259904](main:@1) -> float_type, {2, 32, 50, 64}, {102400, 3200, 64, 1}: 0.00045656ms, 0.00401123% main:@1541 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=51200,local=256,](main:@1537,main:@1539,main:@1540) -> float_type, {2, 32, 50, 64}, {102400, 3200, 64, 1}: 0.0194218ms, 0.170636% main:@1542 = load[offset=2452736,end=2469120](main:@1) -> float_type, {2, 32, 1, 64}, {2048, 64, 64, 1}: 0.00063802ms, 0.0056055% main:@1543 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1533,main:@1541,main:@1542) -> float_type, {2, 32, 1, 64}, {2048, 64, 64, 1}: 0.0206448ms, 0.181381% main:@1544 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](LMobYAd5ru:0) -> float_type, {32, 4}, {17, 1}: 0.000861ms, 0.00756455% main:@1545 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](KasFvpq4IG:0) -> float_type, {32, 2}, {5, 1}: 0.00064666ms, 0.00568141% main:@1546 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](4M92ryD8ux:0) -> float_type, {32, 4}, {9, 1}: 0.00076906ms, 0.00675679% main:@1547 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](4gFa5o2YsJ:0) -> float_type, {32, 4}, {9, 1}: 0.0009548ms, 0.00838866% main:@1548 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](TSqGwrHeo4:0) -> float_type, {32, 2}, {5, 1}: 0.0006578ms, 0.00577928% main:@1549 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](6XMrIsBjLY:0) -> float_type, {32, 2}, {13, 1}: 0.00061382ms, 0.00539288% main:@1550 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](72zWmh93Bo:0) -> float_type, {32, 4}, {17, 1}: 0.00076464ms, 0.00671796% main:@1551 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](qzd23aNGIT:0) -> float_type, {32, 4}, {17, 1}: 0.00096074ms, 0.00844085% main:@1552 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](8vM5p4RzXd:0) -> float_type, {32, 2}, {13, 1}: 0.00061494ms, 0.00540272% main:@1553 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](9dgeU4bKka:0) -> float_type, {32, 4}, {17, 1}: 0.00063038ms, 0.00553838% main:@1554 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](AB6lekmMxa:0) -> float_type, {32, 8}, {25, 1}: 0.00080212ms, 0.00704725% main:@1555 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](eQquakfWBY:0) -> float_type, {32, 4}, {17, 1}: 0.0009326ms, 0.00819361% main:@1556 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](BYyjkPCX0m:0) -> float_type, {32, 2}, {13, 1}: 0.00062966ms, 0.00553205% main:@1557 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](bTeX85HPmJ:0) -> float_type, {32, 4}, {17, 1}: 0.00064262ms, 0.00564592% main:@1558 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](IRem4MiFDP:0) -> float_type, {32, 2}, {5, 1}: 0.00075528ms, 0.00663572% main:@1559 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](fMiQsYNnet:0) -> float_type, {32, 2}, {5, 1}: 0.00094052ms, 0.0082632% main:@1560 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](EKpP4xMXu8:0) -> float_type, {32, 4}, {17, 1}: 0.00059902ms, 0.00526286% main:@1561 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](EZBIAqn6LS:0) -> float_type, {32, 2}, {13, 1}: 0.00059362ms, 0.00521541% main:@1562 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](EUciqFuptg:0) -> float_type, {32, 4}, {17, 1}: 0.0006867ms, 0.00603319% main:@1563 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](Vbv2Om3ptd:0) -> float_type, {32, 8}, {17, 1}: 0.00100582ms, 0.00883691% main:@1564 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](Hnjq487SIY:0) -> float_type, {32, 2}, {13, 1}: 0.00067244ms, 0.00590791% main:@1565 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](JCVMzpWyri:0) -> float_type, {32, 4}, {17, 1}: 0.00062496ms, 0.00549076% main:@1566 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](JSPsGophtx:0) -> float_type, {32, 8}, {25, 1}: 0.0006371ms, 0.00559742% main:@1567 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](JaHTUxB6zL:0) -> float_type, {32, 4}, {17, 1}: 0.00094952ms, 0.00834227% main:@1568 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](JjyOIxALzM:0) -> float_type, {32, 2}, {5, 1}: 0.00062836ms, 0.00552063% main:@1569 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](khJcIxW3zv:0) -> float_type, {32, 4}, {9, 1}: 0.00063398ms, 0.00557001% main:@1570 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](Tej5M3O6As:0) -> float_type, {32, 8}, {17, 1}: 0.00066144ms, 0.00581126% main:@1571 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](r1WGP7Xe9H:0) -> float_type, {32, 2}, {13, 1}: 0.00097314ms, 0.00854979% main:@1572 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](XPcHGKbop2:0) -> float_type, {32, 2}, {13, 1}: 0.00064666ms, 0.00568141% main:@1573 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](lQmygxs5aD:0) -> float_type, {32, 2}, {13, 1}: 0.0006437ms, 0.0056554% main:@1574 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](d8RoEzG52p:0) -> float_type, {32, 8}, {17, 1}: 0.00062706ms, 0.00550921% main:@1575 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](omr2g4XaH5:0) -> float_type, {32, 8}, {17, 1}: 0.0009804ms, 0.00861357% main:@1576 = slice[axes={0, 1},starts={0, 3},ends={32, 5}](agDiJj8lLK:0) -> float_type, {32, 2}, {13, 1}: 0.00064308ms, 0.00564996% main:@1577 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](aqMQBgb3hf:0) -> float_type, {32, 8}, {17, 1}: 0.00066294ms, 0.00582444% main:@1578 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](cN2JSTnFOu:0) -> float_type, {32, 4}, {9, 1}: 0.00064482ms, 0.00566524% main:@1579 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](iCbkMrmXBh:0) -> float_type, {32, 4}, {9, 1}: 0.0009321ms, 0.00818922% main:@1580 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](nyLxrsTmOB:0) -> float_type, {32, 8}, {17, 1}: 0.00066462ms, 0.0058392% main:@1581 = load[offset=2489600,end=2510080](main:@1) -> float_type, {32, 160}, {160, 1}: 0.00051612ms, 0.00453451% main:@1582 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](fZhaL2tCm6:0) -> float_type, {32, 8}, {17, 1}: 0.0006701ms, 0.00588735% main:@1583 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=134,local=256,](main:@1568,main:@1561,main:@1558,main:@1573,main:@1548,main:@1571,main:@1576,main:@1572,main:@1552,main:@1559,main:@1556,main:@1564,main:@1545,main:@1549,main:@1578,main:@1553,main:@1546,main:@1544,main:@1569,main:@1562,main:@1557,main:@1560,main:@1555,main:@1565,main:@1550,main:@1551,main:@1579,main:@1567,main:@1547,main:@1566,main:@1582,main:@1574,main:@1570,main:@1575,main:@1554,main:@1577,main:@1563,main:@1580,main:@1581) -> float_type, {32, 160}, {160, 1}: 0.0452765ms, 0.397789% main:@1584 = load[offset=2469120,end=2489600](main:@1) -> float_type, {32, 160}, {160, 1}: 0.00069178ms, 0.00607782% main:@1585 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1583,main:@869,main:@1584) -> float_type, {32, 160}, {160, 1}: 0.0216225ms, 0.189971% main:@1586 = load[offset=2489600,end=2511360](main:@1) -> float_type, {32, 170}, {170, 1}: 0.00064968ms, 0.00570794% main:@1587 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](1QWFmJ96aB:0) -> float_type, {32, 2}, {5, 1}: 0.00068958ms, 0.0060585% main:@1588 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](3VED7WSJ1b:0) -> float_type, {32, 2}, {5, 1}: 0.00064522ms, 0.00566876% main:@1589 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](5yColzZ8K3:0) -> float_type, {32, 2}, {5, 1}: 0.00092656ms, 0.00814055% main:@1590 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](6Wh2Jm7GHC:0) -> float_type, {32, 2}, {5, 1}: 0.00063588ms, 0.0055867% main:@1591 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](zX6xKOWG9S:0) -> float_type, {32, 2}, {5, 1}: 0.0006347ms, 0.00557633% main:@1592 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](gdck97AK4w:0) -> float_type, {32, 4}, {17, 1}: 0.00063992ms, 0.00562219% main:@1593 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](NLt5JnRMjG:0) -> float_type, {32, 2}, {5, 1}: 0.00088346ms, 0.00776188% main:@1594 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](Cw59eOo7M1:0) -> float_type, {32, 4}, {9, 1}: 0.00058886ms, 0.00517359% main:@1595 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](E1ZkVIXqS5:0) -> float_type, {32, 4}, {9, 1}: 0.00063556ms, 0.00558389% main:@1596 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](EhUjmCTaGW:0) -> float_type, {32, 4}, {9, 1}: 0.0005946ms, 0.00522402% main:@1597 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](Pu1O5qFxY6:0) -> float_type, {32, 4}, {17, 1}: 0.00087204ms, 0.00766155% main:@1598 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](FWhBxj4Tun:0) -> float_type, {32, 4}, {9, 1}: 0.00061264ms, 0.00538252% main:@1599 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](yVxKAL4U5G:0) -> float_type, {32, 4}, {9, 1}: 0.0005975ms, 0.0052495% main:@1600 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](HmATr3gKsW:0) -> float_type, {32, 4}, {9, 1}: 0.00061842ms, 0.0054333% main:@1601 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](J3H2EjAXt8:0) -> float_type, {32, 2}, {5, 1}: 0.00087996ms, 0.00773113% main:@1602 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](IMdFk7D16X:0) -> float_type, {32, 4}, {9, 1}: 0.00059412ms, 0.00521981% main:@1603 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](IxpbBg7JTO:0) -> float_type, {32, 2}, {5, 1}: 0.00059426ms, 0.00522104% main:@1604 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](ZRYNFfjSez:0) -> float_type, {32, 8}, {33, 1}: 0.00061404ms, 0.00539482% main:@1605 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](vAbKepTjQg:0) -> float_type, {32, 2}, {5, 1}: 0.00079782ms, 0.00700947% main:@1606 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](KCNfS4GlpP:0) -> float_type, {32, 4}, {9, 1}: 0.00063292ms, 0.00556069% main:@1607 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](O5aJeUQh2G:0) -> float_type, {32, 8}, {17, 1}: 0.00058746ms, 0.00516129% main:@1608 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](gqk5byC8aJ:0) -> float_type, {32, 2}, {5, 1}: 0.00060286ms, 0.00529659% main:@1609 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](NJBt1ubXFW:0) -> float_type, {32, 2}, {5, 1}: 0.00074722ms, 0.00656491% main:@1610 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](Ouh1XpPZaj:0) -> float_type, {32, 4}, {9, 1}: 0.00061986ms, 0.00544595% main:@1611 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](gsP96ZTbn0:0) -> float_type, {32, 8}, {17, 1}: 0.00062634ms, 0.00550288% main:@1612 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](mTMtDEo3r8:0) -> float_type, {32, 4}, {9, 1}: 0.00064972ms, 0.00570829% main:@1613 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](U6GqhP8k7I:0) -> float_type, {32, 4}, {9, 1}: 0.00073046ms, 0.00641766% main:@1614 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](UMO0o14RuI:0) -> float_type, {32, 4}, {9, 1}: 0.0005734ms, 0.00503776% main:@1615 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](Uc42tiuavR:0) -> float_type, {32, 4}, {9, 1}: 0.00060028ms, 0.00527393% main:@1616 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](oQ5OcjZJe8:0) -> float_type, {32, 8}, {17, 1}: 0.00060116ms, 0.00528166% main:@1617 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](qfwYOHde3u:0) -> float_type, {32, 8}, {17, 1}: 0.00079378ms, 0.00697397% main:@1618 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](mNgQ4ohWqU:0) -> float_type, {32, 4}, {9, 1}: 0.00062594ms, 0.00549937% main:@1619 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](nltbg9ZWQX:0) -> float_type, {32, 4}, {17, 1}: 0.0005989ms, 0.0052618% main:@1620 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](tIGjbcEuDd:0) -> float_type, {32, 4}, {9, 1}: 0.00059866ms, 0.00525969% main:@1621 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](ecPTQUALkq:0) -> float_type, {32, 2}, {5, 1}: 0.0006907ms, 0.00606834% main:@1622 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](magwfJSnUu:0) -> float_type, {32, 4}, {9, 1}: 0.0006093ms, 0.00535317% main:@1623 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](xVErSbJ0hg:0) -> float_type, {32, 4}, {9, 1}: 0.0006898ms, 0.00606043% main:@1624 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](l4bReWtsMB:0) -> float_type, {32, 8}, {25, 1}: 0.00064676ms, 0.00568229% main:@1625 = slice[axes={0, 1},starts={0, 1},ends={32, 3}](pQsi1GNxvj:0) -> float_type, {32, 2}, {5, 1}: 0.00066448ms, 0.00583797% main:@1626 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](lDhFNYzbxR:0) -> float_type, {32, 8}, {17, 1}: 0.00063596ms, 0.0055874% main:@1627 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](l5WSX3VvMK:0) -> float_type, {32, 8}, {25, 1}: 0.00065224ms, 0.00573043% main:@1628 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=132,local=256,](main:@1590,main:@1621,main:@1609,main:@1605,main:@1591,main:@1625,main:@1593,main:@1588,main:@1587,main:@1601,main:@1608,main:@1603,main:@1589,main:@1596,main:@1613,main:@1595,main:@1594,main:@1606,main:@1615,main:@1602,main:@1618,main:@1612,main:@1599,main:@1592,main:@1597,main:@1620,main:@1623,main:@1614,main:@1610,main:@1619,main:@1600,main:@1622,main:@1598,main:@1617,main:@1626,main:@1611,main:@1607,main:@1604,main:@1624,main:@1627,main:@1616,main:@1586) -> float_type, {32, 170}, {170, 1}: 0.0441368ms, 0.387776% main:@1629 = hip::hip_copy_literal[id=main:@literal:101] -> float_type, {170, 160}, {160, 1}: 0.00092804ms, 0.00815355% main:@1630 = load[offset=2526464,end=2546944](main:@1) -> float_type, {32, 160}, {160, 1}: 0.00049228ms, 0.00432506% main:@1631 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1628,main:@1629,main:@1630) -> float_type, {32, 160}, {160, 1}: 0.0226631ms, 0.199113% main:@1632 = load[offset=2489600,end=2526464](main:@1) -> float_type, {32, 18, 16}, {288, 16, 1}: 0.0006468ms, 0.00568264% main:@1633 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](2Yy3zxNGBC:0) -> float_type, {32, 16}, {45, 1}: 0.0006533ms, 0.00573975% main:@1634 = unsqueeze[axes={1},steps={}](main:@1633) -> float_type, {32, 1, 16}, {45, 16, 1}: 0.00078508ms, 0.00689754% main:@1635 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](WDLoyKsiA7:0) -> float_type, {32, 16}, {33, 1}: 0.00077778ms, 0.0068334% main:@1636 = unsqueeze[axes={1},steps={}](main:@1635) -> float_type, {32, 1, 16}, {33, 16, 1}: 0.00057612ms, 0.00506166% main:@1637 = slice[axes={0, 1},starts={0, 9},ends={32, 25}](LKAvFme9ZS:0) -> float_type, {32, 16}, {49, 1}: 0.00061832ms, 0.00543242% main:@1638 = unsqueeze[axes={1},steps={}](main:@1637) -> float_type, {32, 1, 16}, {49, 16, 1}: 0.0007253ms, 0.00637232% main:@1639 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](VfFM90O7LN:0) -> float_type, {32, 16}, {41, 1}: 0.00075016ms, 0.00659074% main:@1640 = unsqueeze[axes={1},steps={}](main:@1639) -> float_type, {32, 1, 16}, {41, 16, 1}: 0.00054976ms, 0.00483007% main:@1641 = slice[axes={0, 1},starts={0, 9},ends={32, 25}](TxEdCcekGS:0) -> float_type, {32, 16}, {49, 1}: 0.00060194ms, 0.00528851% main:@1642 = unsqueeze[axes={1},steps={}](main:@1641) -> float_type, {32, 1, 16}, {49, 16, 1}: 0.00071508ms, 0.00628253% main:@1643 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](z3lqgZM8WU:0) -> float_type, {32, 16}, {41, 1}: 0.00075056ms, 0.00659425% main:@1644 = unsqueeze[axes={1},steps={}](main:@1643) -> float_type, {32, 1, 16}, {41, 16, 1}: 0.00054926ms, 0.00482568% main:@1645 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](pY0t1rI5WU:0) -> float_type, {32, 16}, {41, 1}: 0.00061118ms, 0.00536969% main:@1646 = unsqueeze[axes={1},steps={}](main:@1645) -> float_type, {32, 1, 16}, {41, 16, 1}: 0.00069374ms, 0.00609504% main:@1647 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](sZdFJI9nwg:0) -> float_type, {32, 16}, {41, 1}: 0.00076012ms, 0.00667824% main:@1648 = unsqueeze[axes={1},steps={}](main:@1647) -> float_type, {32, 1, 16}, {41, 16, 1}: 0.00056662ms, 0.0049782% main:@1649 = reshape_lazy[dims={-1, 10, 16}](main:@1631) -> float_type, {32, 10, 16}, {160, 16, 1}: 0.00052388ms, 0.00460269% main:@1650 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=1024,local=256,](main:@1636,main:@1644,main:@1638,main:@1646,main:@1642,main:@1648,main:@1640,main:@1634,main:@1649,main:@1632) -> float_type, {32, 18, 16}, {288, 16, 1}: 0.0243274ms, 0.213735% main:@1651 = transpose[permutation={0, 2, 1}](main:@1650) -> float_type, {32, 16, 18}, {288, 1, 16}: 0.00070386ms, 0.00618396% main:@1652 = load[offset=2526464,end=2563328](main:@1) -> float_type, {32, 16, 18}, {288, 18, 1}: 0.00049376ms, 0.00433806% main:@1653 = gpu::code_object[code_object=13528,symbol_name=contiguous_kernel,global=9216,local=256,](main:@1651,main:@1652) -> float_type, {32, 16, 18}, {288, 18, 1}: 0.014525ms, 0.127613% main:@1654 = hip::hip_copy_literal[id=main:@literal:129] -> float_type, {18, 4}, {4, 1}: 0.00075428ms, 0.00662693% main:@1655 = load[offset=2444544,end=2452736](main:@1) -> float_type, {512, 4}, {4, 1}: 0.00048254ms, 0.00423949% main:@1656 = reshape_lazy[dims={512, 18}](main:@1653) -> float_type, {512, 18}, {18, 1}: 0.00053866ms, 0.00473255% main:@1657 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1656,main:@1654,main:@1655) -> float_type, {512, 4}, {4, 1}: 0.0196352ms, 0.172511% main:@1658 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](2RDXZrx9GA:0) -> float_type, {32, 16}, {45, 1}: 0.0008367ms, 0.00735106% main:@1659 = unsqueeze[axes={1},steps={}](main:@1658) -> float_type, {32, 1, 16}, {45, 16, 1}: 0.00066384ms, 0.00583235% main:@1660 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](2srtUfGvja:0) -> float_type, {32, 16}, {41, 1}: 0.0007783ms, 0.00683797% main:@1661 = unsqueeze[axes={1},steps={}](main:@1660) -> float_type, {32, 1, 16}, {41, 16, 1}: 0.00056088ms, 0.00492777% main:@1662 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](3Laj620MSn:0) -> float_type, {32, 16}, {45, 1}: 0.00062094ms, 0.00545544% main:@1663 = unsqueeze[axes={1},steps={}](main:@1662) -> float_type, {32, 1, 16}, {45, 16, 1}: 0.0005789ms, 0.00508609% main:@1664 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](jznORDJr0N:0) -> float_type, {32, 16}, {45, 1}: 0.00074566ms, 0.0065512% main:@1665 = unsqueeze[axes={1},steps={}](main:@1664) -> float_type, {32, 1, 16}, {45, 16, 1}: 0.0005606ms, 0.00492531% main:@1666 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](HRGzlnEkXQ:0) -> float_type, {32, 16}, {41, 1}: 0.00063742ms, 0.00560023% main:@1667 = unsqueeze[axes={1},steps={}](main:@1666) -> float_type, {32, 1, 16}, {41, 16, 1}: 0.0005581ms, 0.00490334% main:@1668 = slice[axes={0, 1},starts={0, 305},ends={32, 321}](lyiDxfzZg3:0) -> float_type, {32, 16}, {701, 1}: 0.00074818ms, 0.00657334% main:@1669 = unsqueeze[axes={1},steps={}](main:@1668) -> float_type, {32, 1, 16}, {701, 16, 1}: 0.00055606ms, 0.00488542% main:@1670 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](VUqfdN5Fzv:0) -> float_type, {32, 16}, {45, 1}: 0.0006215ms, 0.00546036% main:@1671 = unsqueeze[axes={1},steps={}](main:@1670) -> float_type, {32, 1, 16}, {45, 16, 1}: 0.00055038ms, 0.00483552% main:@1672 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](CHAYRkBvIo:0) -> float_type, {32, 16}, {41, 1}: 0.0007808ms, 0.00685993% main:@1673 = unsqueeze[axes={1},steps={}](main:@1672) -> float_type, {32, 1, 16}, {41, 16, 1}: 0.00058608ms, 0.00514917% main:@1674 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](HzcyUlm8Yj:0) -> float_type, {32, 16}, {45, 1}: 0.00063076ms, 0.00554172% main:@1675 = unsqueeze[axes={1},steps={}](main:@1674) -> float_type, {32, 1, 16}, {45, 16, 1}: 0.00055158ms, 0.00484606% main:@1676 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](duyRoIEkH8:0) -> float_type, {32, 16}, {45, 1}: 0.0006737ms, 0.00591898% main:@1677 = unsqueeze[axes={1},steps={}](main:@1676) -> float_type, {32, 1, 16}, {45, 16, 1}: 0.0005505ms, 0.00483657% main:@1678 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](eQI8ikEMgP:0) -> float_type, {32, 16}, {41, 1}: 0.00064314ms, 0.00565048% main:@1679 = unsqueeze[axes={1},steps={}](main:@1678) -> float_type, {32, 1, 16}, {41, 16, 1}: 0.00056752ms, 0.0049861% main:@1680 = reshape_lazy[dims={-1, 10, 16}](main:@1585) -> float_type, {32, 10, 16}, {160, 16, 1}: 0.0004964ms, 0.00436126% main:@1681 = load[offset=2489600,end=2532608](main:@1) -> float_type, {32, 21, 16}, {336, 16, 1}: 0.00049892ms, 0.0043834% main:@1682 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=896,local=256,](main:@1665,main:@1667,main:@1669,main:@1659,main:@1671,main:@1673,main:@1675,main:@1661,main:@1663,main:@1677,main:@1679,main:@1680,main:@1681) -> float_type, {32, 21, 16}, {336, 16, 1}: 0.0265684ms, 0.233424% main:@1683 = load[offset=2433792,end=2444544](main:@1) -> float_type, {32, 21, 4}, {84, 4, 1}: 0.00067126ms, 0.00589754% main:@1684 = reshape_lazy[dims={32, 16, 4}](main:@1657) -> float_type, {32, 16, 4}, {64, 4, 1}: 0.000516ms, 0.00453346% main:@1685 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1682,main:@1684,main:@1683) -> float_type, {32, 21, 4}, {84, 4, 1}: 0.0184199ms, 0.161833% main:@1686 = slice[axes={0, 1},starts={0, 241},ends={32, 273}](uFSOc5Chp2:0) -> float_type, {32, 32}, {481, 1}: 0.00083508ms, 0.00733683% main:@1687 = slice[axes={0, 1},starts={0, 237},ends={32, 253}](nv2tzWRf1I:0) -> float_type, {32, 16}, {301, 1}: 0.00061156ms, 0.00537303% main:@1688 = load[offset=2469120,end=2480384](main:@1) -> float_type, {32, 88}, {88, 1}: 0.00048332ms, 0.00424634% main:@1689 = slice[axes={0, 1},starts={0, 517},ends={32, 525}](lyiDxfzZg3:0) -> float_type, {32, 8}, {701, 1}: 0.000778ms, 0.00683533% main:@1690 = slice[axes={0, 1},starts={0, 1},ends={32, 33}](n4iptOVFwE:0) -> float_type, {32, 32}, {113, 1}: 0.00064698ms, 0.00568422% main:@1691 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=704,local=256,](main:@1689,main:@1687,main:@1686,main:@1690,main:@1688) -> float_type, {32, 88}, {88, 1}: 0.0159997ms, 0.14057% main:@1692 = hip::hip_copy_literal[id=main:@literal:58] -> float_type, {88, 32}, {32, 1}: 0.00077462ms, 0.00680564% main:@1693 = load[offset=2445824,end=2449920](main:@1) -> float_type, {32, 32}, {32, 1}: 0.00049692ms, 0.00436583% main:@1694 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1691,main:@1692,main:@1693) -> float_type, {32, 32}, {32, 1}: 0.020561ms, 0.180644% main:@1695 = load[offset=2444544,end=2445824](main:@1) -> float_type, {1, 32, 1, 10}, {320, 10, 10, 1}: 0.00066226ms, 0.00581847% main:@1696 = slice[axes={1},starts={0},ends={32}](main:@1105) -> float_type, {320, 32}, {64, 1}: 0.00065858ms, 0.00578614% main:@1697 = reshape_lazy[dims={1, 32, 10, 32}](main:@1696) -> float_type, {1, 32, 10, 32}, {20480, 640, 64, 1}: 0.0006822ms, 0.00599366% main:@1698 = transpose[permutation={0, 1, 3, 2}](main:@1697) -> float_type, {1, 32, 32, 10}, {20480, 640, 1, 64}: 0.00062938ms, 0.00552959% main:@1699 = reshape_lazy[dims={1, 32, 1, 32}](main:@1694) -> float_type, {1, 32, 1, 32}, {1024, 32, 32, 1}: 0.0004978ms, 0.00437356% main:@1700 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1699,main:@1698,main:@1695) -> float_type, {1, 32, 1, 10}, {320, 10, 10, 1}: 0.0213834ms, 0.18787% main:@1701 = unsqueeze[axes={1},steps={}](0MZVmrKeC6:0) -> int32_type, {32, 1, 10}, {10, 10, 1}: 0.00077054ms, 0.00676979% main:@1702 = load[offset=2445824,end=2447104](main:@1) -> float_type, {32, 1, 10}, {10, 10, 1}: 0.00049964ms, 0.00438972% main:@1703 = gpu::code_object[code_object=13536,symbol_name=convert_mul_add_kernel,global=160,local=256,](main:@1701,main:@1702) -> float_type, {32, 1, 10}, {10, 10, 1}: 0.0132926ms, 0.116786% main:@1704 = multibroadcast[out_lens={1, 32, 1, 10}](main:@1703) -> float_type, {1, 32, 1, 10}, {0, 10, 10, 1}: 0.00072362ms, 0.00635756% main:@1705 = load[offset=2447104,end=2448384](main:@1) -> float_type, {1, 32, 1, 10}, {320, 10, 10, 1}: 0.00049418ms, 0.00434175% main:@1706 = gpu::code_object[code_object=13520,symbol_name=pow_div_kernel,global=160,local=256,](main:@1700,main:@1704,main:@1705) -> float_type, {1, 32, 1, 10}, {320, 10, 10, 1}: 0.0158243ms, 0.139029% main:@1707 = load[offset=2448640,end=2449920](main:@1) -> float_type, {1, 32, 1, 10}, {320, 10, 10, 1}: 0.00062254ms, 0.0054695% main:@1708 = gpu::code_object[code_object=14576,symbol_name=softmax_kernel,global=2048,local=64,](main:@1706,main:@1707) -> float_type, {1, 32, 1, 10}, {320, 10, 10, 1}: 0.0145315ms, 0.127671% main:@1709 = load[offset=2444544,end=2448640](main:@1) -> float_type, {1, 32, 1, 32}, {1024, 32, 32, 1}: 0.00064342ms, 0.00565294% main:@1710 = slice[axes={1},starts={32},ends={64}](main:@1105) -> float_type, {320, 32}, {64, 1}: 0.00064838ms, 0.00569652% main:@1711 = reshape_lazy[dims={1, 32, 10, 32}](main:@1710) -> float_type, {1, 32, 10, 32}, {20480, 640, 64, 1}: 0.00097484ms, 0.00856473% main:@1712 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1708,main:@1711,main:@1709) -> float_type, {1, 32, 1, 32}, {1024, 32, 32, 1}: 0.0207551ms, 0.18235% main:@1713 = load[offset=2469120,end=2480384](main:@1) -> float_type, {32, 88}, {88, 1}: 0.00061874ms, 0.00543611% main:@1714 = slice[axes={0, 1},starts={0, 525},ends={32, 533}](lyiDxfzZg3:0) -> float_type, {32, 8}, {701, 1}: 0.0006751ms, 0.00593128% main:@1715 = slice[axes={0, 1},starts={0, 41},ends={32, 73}](n4iptOVFwE:0) -> float_type, {32, 32}, {113, 1}: 0.00062492ms, 0.00549041% main:@1716 = slice[axes={0, 1},starts={0, 273},ends={32, 305}](uFSOc5Chp2:0) -> float_type, {32, 32}, {481, 1}: 0.00080224ms, 0.0070483% main:@1717 = slice[axes={0, 1},starts={0, 253},ends={32, 269}](nv2tzWRf1I:0) -> float_type, {32, 16}, {301, 1}: 0.0009503ms, 0.00834912% main:@1718 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=704,local=256,](main:@1714,main:@1717,main:@1716,main:@1715,main:@1713) -> float_type, {32, 88}, {88, 1}: 0.0159438ms, 0.140079% main:@1719 = hip::hip_copy_literal[id=main:@literal:51] -> float_type, {88, 32}, {32, 1}: 0.00081342ms, 0.00714653% main:@1720 = load[offset=2480384,end=2484480](main:@1) -> float_type, {32, 32}, {32, 1}: 0.00050446ms, 0.00443207% main:@1721 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1718,main:@1719,main:@1720) -> float_type, {32, 32}, {32, 1}: 0.0203648ms, 0.178921% main:@1722 = slice[axes={1},starts={32},ends={64}](main:@686) -> float_type, {960, 32}, {64, 1}: 0.00079868ms, 0.00701702% main:@1723 = reshape_lazy[dims={1, 32, 30, 32}](main:@1722) -> float_type, {1, 32, 30, 32}, {61440, 1920, 64, 1}: 0.00063934ms, 0.0056171% main:@1724 = transpose[permutation={0, 1, 3, 2}](main:@1723) -> float_type, {1, 32, 32, 30}, {61440, 1920, 1, 64}: 0.00059648ms, 0.00524054% main:@1725 = load[offset=2448640,end=2452480](main:@1) -> float_type, {1, 32, 1, 30}, {960, 30, 30, 1}: 0.00052936ms, 0.00465084% main:@1726 = reshape_lazy[dims={1, 32, 1, 32}](main:@1721) -> float_type, {1, 32, 1, 32}, {1024, 32, 32, 1}: 0.00051802ms, 0.00455121% main:@1727 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1726,main:@1724,main:@1725) -> float_type, {1, 32, 1, 30}, {960, 30, 30, 1}: 0.0199797ms, 0.175537% VX8uEBNndW:0 = @param:VX8uEBNndW:0 -> int32_type, {32, 30}, {30, 1}: 0.00072376ms, 0.00635879% main:@1729 = unsqueeze[axes={1},steps={}](VX8uEBNndW:0) -> int32_type, {32, 1, 30}, {30, 30, 1}: 0.00059742ms, 0.0052488% main:@1730 = load[offset=2469120,end=2472960](main:@1) -> float_type, {32, 1, 30}, {30, 30, 1}: 0.00049126ms, 0.0043161% main:@1731 = gpu::code_object[code_object=13528,symbol_name=convert_add_kernel,global=480,local=256,](main:@1729,main:@1730) -> float_type, {32, 1, 30}, {30, 30, 1}: 0.0132745ms, 0.116627% main:@1732 = load[offset=2472960,end=2476800](main:@1) -> float_type, {1, 32, 1, 30}, {960, 30, 30, 1}: 0.00064826ms, 0.00569547% main:@1733 = multibroadcast[out_lens={1, 32, 1, 30}](main:@1731) -> float_type, {1, 32, 1, 30}, {0, 30, 30, 1}: 0.00055328ms, 0.00486099% main:@1734 = gpu::code_object[code_object=13520,symbol_name=pow_sub_kernel,global=480,local=256,](main:@1727,main:@1733,main:@1732) -> float_type, {1, 32, 1, 30}, {960, 30, 30, 1}: 0.0151598ms, 0.133191% main:@1735 = load[offset=2469120,end=2472960](main:@1) -> float_type, {1, 32, 1, 30}, {960, 30, 30, 1}: 0.00061822ms, 0.00543154% main:@1736 = gpu::code_object[code_object=14584,symbol_name=softmax_kernel,global=2048,local=64,](main:@1734,main:@1735) -> float_type, {1, 32, 1, 30}, {960, 30, 30, 1}: 0.0146989ms, 0.129141% main:@1737 = slice[axes={1},starts={0},ends={32}](main:@686) -> float_type, {960, 32}, {64, 1}: 0.00108308ms, 0.0095157% main:@1738 = reshape_lazy[dims={1, 32, 30, 32}](main:@1737) -> float_type, {1, 32, 30, 32}, {61440, 1920, 64, 1}: 0.00059184ms, 0.00519977% main:@1739 = load[offset=2448640,end=2452736](main:@1) -> float_type, {1, 32, 1, 32}, {1024, 32, 32, 1}: 0.00049294ms, 0.00433086% main:@1740 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1736,main:@1738,main:@1739) -> float_type, {1, 32, 1, 32}, {1024, 32, 32, 1}: 0.0183868ms, 0.161543% main:@1741 = slice[axes={1},starts={96},ends={128}](main:@652) -> float_type, {32, 32}, {128, 1}: 0.00074524ms, 0.00654751% main:@1742 = load[offset=1698432,end=1702528](main:@1) -> float_type, {32, 32}, {32, 1}: 0.0004852ms, 0.00426286% main:@1743 = gpu::code_object[code_object=13528,symbol_name=contiguous_kernel,global=512,local=256,](main:@1741,main:@1742) -> float_type, {32, 32}, {32, 1}: 0.013093ms, 0.115032% main:@1744 = slice[axes={1},starts={0},ends={32}](main:@666) -> float_type, {480, 32}, {64, 1}: 0.00077292ms, 0.0067907% main:@1745 = reshape_lazy[dims={1, 32, 15, 32}](main:@1744) -> float_type, {1, 32, 15, 32}, {30720, 960, 64, 1}: 0.00060042ms, 0.00527516% main:@1746 = transpose[permutation={0, 1, 3, 2}](main:@1745) -> float_type, {1, 32, 32, 15}, {30720, 960, 1, 64}: 0.00093496ms, 0.00821435% main:@1747 = reshape_lazy[dims={1, 32, 1, 32}](main:@1743) -> float_type, {1, 32, 1, 32}, {1024, 32, 32, 1}: 0.00050266ms, 0.00441626% main:@1748 = load[offset=1696512,end=1698432](main:@1) -> float_type, {1, 32, 1, 15}, {480, 15, 15, 1}: 0.00047358ms, 0.00416077% main:@1749 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1747,main:@1746,main:@1748) -> float_type, {1, 32, 1, 15}, {480, 15, 15, 1}: 0.021551ms, 0.189342% hzqb8Cpfc1:0 = @param:hzqb8Cpfc1:0 -> int32_type, {32, 15}, {15, 1}: 0.00068494ms, 0.00601773% main:@1751 = unsqueeze[axes={1},steps={}](hzqb8Cpfc1:0) -> int32_type, {32, 1, 15}, {15, 15, 1}: 0.0005884ms, 0.00516955% main:@1752 = load[offset=1698432,end=1700352](main:@1) -> float_type, {32, 1, 15}, {15, 15, 1}: 0.00049746ms, 0.00437057% main:@1753 = gpu::code_object[code_object=13536,symbol_name=convert_mul_add_kernel,global=240,local=256,](main:@1751,main:@1752) -> float_type, {32, 1, 15}, {15, 15, 1}: 0.0132294ms, 0.11623% main:@1754 = multibroadcast[out_lens={1, 32, 1, 15}](main:@1753) -> float_type, {1, 32, 1, 15}, {0, 15, 15, 1}: 0.0006903ms, 0.00606482% main:@1755 = load[offset=1702528,end=1704448](main:@1) -> float_type, {1, 32, 1, 15}, {480, 15, 15, 1}: 0.00049652ms, 0.00436231% main:@1756 = gpu::code_object[code_object=13520,symbol_name=add_sub_kernel,global=240,local=256,](main:@1749,main:@1754,main:@1755) -> float_type, {1, 32, 1, 15}, {480, 15, 15, 1}: 0.0132317ms, 0.116251% main:@1757 = load[offset=1700608,end=1702528](main:@1) -> float_type, {1, 32, 1, 15}, {480, 15, 15, 1}: 0.00062692ms, 0.00550798% main:@1758 = gpu::code_object[code_object=14568,symbol_name=softmax_kernel,global=2048,local=64,](main:@1756,main:@1757) -> float_type, {1, 32, 1, 15}, {480, 15, 15, 1}: 0.0143782ms, 0.126324% main:@1759 = slice[axes={1},starts={32},ends={64}](main:@666) -> float_type, {480, 32}, {64, 1}: 0.00073288ms, 0.00643892% main:@1760 = reshape_lazy[dims={1, 32, 15, 32}](main:@1759) -> float_type, {1, 32, 15, 32}, {30720, 960, 64, 1}: 0.0009411ms, 0.00826829% main:@1761 = load[offset=1696512,end=1700608](main:@1) -> float_type, {1, 32, 1, 32}, {1024, 32, 32, 1}: 0.00048924ms, 0.00429835% main:@1762 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1758,main:@1760,main:@1761) -> float_type, {1, 32, 1, 32}, {1024, 32, 32, 1}: 0.0187326ms, 0.16458% main:@1763 = slice[axes={1},starts={0},ends={128}](main:@1293) -> float_type, {32, 128}, {256, 1}: 0.00073614ms, 0.00646756% main:@1764 = reshape_lazy[dims={32, 1, 128}](main:@1763) -> float_type, {32, 1, 128}, {256, 128, 1}: 0.0005693ms, 0.00500174% main:@1765 = slice[axes={2},starts={0},ends={64}](main:@1764) -> float_type, {32, 1, 64}, {256, 128, 1}: 0.00104754ms, 0.00920345% main:@1766 = unsqueeze[axes={0},steps={}](main:@1765) -> float_type, {1, 32, 1, 64}, {8192, 256, 128, 1}: 0.00065602ms, 0.00576364% main:@1767 = slice[axes={2},starts={64},ends={128}](main:@1764) -> float_type, {32, 1, 64}, {256, 128, 1}: 0.00101128ms, 0.00888488% main:@1768 = unsqueeze[axes={0},steps={}](main:@1767) -> float_type, {1, 32, 1, 64}, {8192, 256, 128, 1}: 0.0005957ms, 0.00523369% main:@1769 = load[offset=1573632,end=1590016](main:@1) -> float_type, {2, 32, 1, 64}, {2048, 64, 64, 1}: 0.00047952ms, 0.00421296% main:@1770 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=1024,local=256,](main:@1766,main:@1768,main:@1769) -> float_type, {2, 32, 1, 64}, {2048, 64, 64, 1}: 0.0136646ms, 0.120054% main:@1771 = slice[axes={1},starts={128},ends={256}](main:@1125) -> float_type, {1600, 128}, {256, 1}: 0.00072608ms, 0.00637918% main:@1772 = reshape_lazy[dims={32, 50, 128}](main:@1771) -> float_type, {32, 50, 128}, {12800, 256, 1}: 0.00089102ms, 0.0078283% main:@1773 = slice[axes={2},starts={0},ends={64}](main:@1772) -> float_type, {32, 50, 64}, {12800, 256, 1}: 0.00071336ms, 0.00626742% main:@1774 = unsqueeze[axes={0},steps={}](main:@1773) -> float_type, {1, 32, 50, 64}, {409600, 12800, 256, 1}: 0.00103426ms, 0.00908678% main:@1775 = slice[axes={2},starts={64},ends={128}](main:@1772) -> float_type, {32, 50, 64}, {12800, 256, 1}: 0.0006509ms, 0.00571866% main:@1776 = unsqueeze[axes={0},steps={}](main:@1775) -> float_type, {1, 32, 50, 64}, {409600, 12800, 256, 1}: 0.00067536ms, 0.00593356% main:@1777 = load[offset=2469120,end=3288320](main:@1) -> float_type, {2, 32, 50, 64}, {102400, 3200, 64, 1}: 0.0004728ms, 0.00415391% main:@1778 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=51200,local=256,](main:@1774,main:@1776,main:@1777) -> float_type, {2, 32, 50, 64}, {102400, 3200, 64, 1}: 0.0193415ms, 0.16993% main:@1779 = transpose[permutation={0, 1, 3, 2}](main:@1778) -> float_type, {2, 32, 64, 50}, {102400, 3200, 1, 64}: 0.0006997ms, 0.00614741% main:@1780 = load[offset=1590016,end=1602816](main:@1) -> float_type, {2, 32, 1, 50}, {1600, 50, 50, 1}: 0.00051478ms, 0.00452274% main:@1781 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1770,main:@1779,main:@1780) -> float_type, {2, 32, 1, 50}, {1600, 50, 50, 1}: 0.0223694ms, 0.196533% main:@1782 = load[offset=1573632,end=1580032](main:@1) -> float_type, {32, 1, 50}, {50, 50, 1}: 0.00067ms, 0.00588647% main:@1783 = unsqueeze[axes={1},steps={}](vOI1hWzBDK:0) -> int32_type, {32, 1, 50}, {50, 50, 1}: 0.00059544ms, 0.0052314% main:@1784 = gpu::code_object[code_object=13528,symbol_name=convert_mul_kernel,global=800,local=256,](main:@1783,main:@1782) -> float_type, {32, 1, 50}, {50, 50, 1}: 0.0132867ms, 0.116734% main:@1785 = multibroadcast[out_lens={2, 32, 1, 50}](main:@1784) -> float_type, {2, 32, 1, 50}, {0, 50, 50, 1}: 0.00067494ms, 0.00592987% main:@1786 = load[offset=1606912,end=1619712](main:@1) -> float_type, {2, 32, 1, 50}, {1600, 50, 50, 1}: 0.00049426ms, 0.00434246% main:@1787 = gpu::code_object[code_object=13520,symbol_name=add_mul_kernel,global=1600,local=256,](main:@1781,main:@1785,main:@1786) -> float_type, {2, 32, 1, 50}, {1600, 50, 50, 1}: 0.0133234ms, 0.117056% main:@1788 = load[offset=1594112,end=1606912](main:@1) -> float_type, {2, 32, 1, 50}, {1600, 50, 50, 1}: 0.00062904ms, 0.0055266% main:@1789 = gpu::code_object[code_object=14584,symbol_name=softmax_kernel,global=4096,local=64,](main:@1787,main:@1788) -> float_type, {2, 32, 1, 50}, {1600, 50, 50, 1}: 0.0157903ms, 0.13873% main:@1790 = load[offset=2469120,end=3288320](main:@1) -> float_type, {2, 32, 50, 64}, {102400, 3200, 64, 1}: 0.00062024ms, 0.00544929% main:@1791 = slice[axes={1},starts={0},ends={128}](main:@1125) -> float_type, {1600, 128}, {256, 1}: 0.00058762ms, 0.0051627% main:@1792 = reshape_lazy[dims={32, 50, 128}](main:@1791) -> float_type, {32, 50, 128}, {12800, 256, 1}: 0.000622ms, 0.00546475% main:@1793 = slice[axes={2},starts={0},ends={64}](main:@1792) -> float_type, {32, 50, 64}, {12800, 256, 1}: 0.00102472ms, 0.00900296% main:@1794 = unsqueeze[axes={0},steps={}](main:@1793) -> float_type, {1, 32, 50, 64}, {409600, 12800, 256, 1}: 0.0005975ms, 0.0052495% main:@1795 = slice[axes={2},starts={64},ends={128}](main:@1792) -> float_type, {32, 50, 64}, {12800, 256, 1}: 0.00101734ms, 0.00893812% main:@1796 = unsqueeze[axes={0},steps={}](main:@1795) -> float_type, {1, 32, 50, 64}, {409600, 12800, 256, 1}: 0.00057006ms, 0.00500842% main:@1797 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=51200,local=256,](main:@1794,main:@1796,main:@1790) -> float_type, {2, 32, 50, 64}, {102400, 3200, 64, 1}: 0.0190142ms, 0.167054% main:@1798 = load[offset=1577728,end=1594112](main:@1) -> float_type, {2, 32, 1, 64}, {2048, 64, 64, 1}: 0.00066622ms, 0.00585326% main:@1799 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1789,main:@1797,main:@1798) -> float_type, {2, 32, 1, 64}, {2048, 64, 64, 1}: 0.0194145ms, 0.170572% H6VFcpqaB0:0 = @param:H6VFcpqaB0:0 -> float_type, {32, 30, 9}, {270, 9, 1}: 0.00076466ms, 0.00671813% main:@1801 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 30, 9}](H6VFcpqaB0:0) -> float_type, {32, 30, 8}, {270, 9, 1}: 0.00074526ms, 0.00654769% CYnW83Np9f:0 = @param:CYnW83Np9f:0 -> float_type, {32, 30, 17}, {510, 17, 1}: 0.00057214ms, 0.00502669% main:@1803 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 30, 17}](CYnW83Np9f:0) -> float_type, {32, 30, 16}, {510, 17, 1}: 0.00077414ms, 0.00680142% hEoRIXiSF0:0 = @param:hEoRIXiSF0:0 -> float_type, {32, 30, 9}, {270, 9, 1}: 0.00054058ms, 0.00474941% main:@1805 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 30, 9}](hEoRIXiSF0:0) -> float_type, {32, 30, 8}, {270, 9, 1}: 0.0006365ms, 0.00559215% 4RjhYVIfz6:0 = @param:4RjhYVIfz6:0 -> float_type, {32, 30, 9}, {270, 9, 1}: 0.00056118ms, 0.0049304% main:@1807 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 30, 9}](4RjhYVIfz6:0) -> float_type, {32, 30, 8}, {270, 9, 1}: 0.00080298ms, 0.0070548% aHGpDUJj9f:0 = @param:aHGpDUJj9f:0 -> float_type, {32, 30, 21}, {630, 21, 1}: 0.00054434ms, 0.00478245% main:@1809 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 30, 21}](aHGpDUJj9f:0) -> float_type, {32, 30, 20}, {630, 21, 1}: 0.00064322ms, 0.00565119% Hj2GwToc7U:0 = @param:Hj2GwToc7U:0 -> float_type, {32, 30, 9}, {270, 9, 1}: 0.000548ms, 0.00481461% main:@1811 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 30, 9}](Hj2GwToc7U:0) -> float_type, {32, 30, 8}, {270, 9, 1}: 0.000829ms, 0.00728341% S4PeIw3K1X:0 = @param:S4PeIw3K1X:0 -> float_type, {32, 30, 21}, {630, 21, 1}: 0.00055292ms, 0.00485783% main:@1813 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 30, 21}](S4PeIw3K1X:0) -> float_type, {32, 30, 20}, {630, 21, 1}: 0.00093452ms, 0.00821048% DW8PcmTVkC:0 = @param:DW8PcmTVkC:0 -> float_type, {32, 30, 9}, {270, 9, 1}: 0.00052774ms, 0.00463661% main:@1815 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 30, 9}](DW8PcmTVkC:0) -> float_type, {32, 30, 8}, {270, 9, 1}: 0.0007037ms, 0.00618255% cEDTxpPUfb:0 = @param:cEDTxpPUfb:0 -> float_type, {32, 30, 9}, {270, 9, 1}: 0.00056514ms, 0.00496519% main:@1817 = slice[axes={0, 1, 2},starts={0, 0, 1},ends={32, 30, 9}](cEDTxpPUfb:0) -> float_type, {32, 30, 8}, {270, 9, 1}: 0.00069852ms, 0.00613704% main:@1818 = load[offset=2469120,end=2868480](main:@1) -> float_type, {32, 30, 104}, {3120, 104, 1}: 0.00049242ms, 0.00432629% main:@1819 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=11093,local=256,](main:@1801,main:@1815,main:@1805,main:@1811,main:@1809,main:@1807,main:@1817,main:@1803,main:@1813,main:@1818) -> float_type, {32, 30, 104}, {3120, 104, 1}: 0.0263289ms, 0.23132% main:@1820 = hip::hip_copy_literal[id=main:@literal:21] -> float_type, {104, 64}, {64, 1}: 0.00076006ms, 0.00667772% main:@1821 = reshape_lazy[dims={960, 104}](main:@1819) -> float_type, {960, 104}, {104, 1}: 0.00055122ms, 0.0048429% main:@1822 = load[offset=2868480,end=3114240](main:@1) -> float_type, {960, 64}, {64, 1}: 0.00047752ms, 0.00419538% main:@1823 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1821,main:@1820,main:@1822) -> float_type, {960, 64}, {64, 1}: 0.0238593ms, 0.209623% main:@1824 = slice[axes={0, 1},starts={0, 305},ends={32, 337}](uFSOc5Chp2:0) -> float_type, {32, 32}, {481, 1}: 0.00085436ms, 0.00750622% main:@1825 = slice[axes={0, 1},starts={0, 73},ends={32, 105}](n4iptOVFwE:0) -> float_type, {32, 32}, {113, 1}: 0.00063264ms, 0.00555823% main:@1826 = slice[axes={0, 1},starts={0, 533},ends={32, 541}](lyiDxfzZg3:0) -> float_type, {32, 8}, {701, 1}: 0.00077734ms, 0.00682953% main:@1827 = slice[axes={0, 1},starts={0, 269},ends={32, 285}](nv2tzWRf1I:0) -> float_type, {32, 16}, {301, 1}: 0.00059982ms, 0.00526988% main:@1828 = load[offset=1598208,end=1609472](main:@1) -> float_type, {32, 88}, {88, 1}: 0.00047392ms, 0.00416375% main:@1829 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=704,local=256,](main:@1826,main:@1827,main:@1824,main:@1825,main:@1828) -> float_type, {32, 88}, {88, 1}: 0.0158019ms, 0.138832% main:@1830 = hip::hip_copy_literal[id=main:@literal:77] -> float_type, {88, 32}, {32, 1}: 0.00082176ms, 0.0072198% main:@1831 = load[offset=1594112,end=1598208](main:@1) -> float_type, {32, 32}, {32, 1}: 0.00048914ms, 0.00429747% main:@1832 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1829,main:@1830,main:@1831) -> float_type, {32, 32}, {32, 1}: 0.0208429ms, 0.183121% main:@1833 = slice[axes={1},starts={32},ends={64}](main:@1823) -> float_type, {960, 32}, {64, 1}: 0.00076718ms, 0.00674027% main:@1834 = reshape_lazy[dims={1, 32, 30, 32}](main:@1833) -> float_type, {1, 32, 30, 32}, {61440, 1920, 64, 1}: 0.00068016ms, 0.00597573% main:@1835 = transpose[permutation={0, 1, 3, 2}](main:@1834) -> float_type, {1, 32, 32, 30}, {61440, 1920, 1, 64}: 0.00060044ms, 0.00527533% main:@1836 = load[offset=1573632,end=1577472](main:@1) -> float_type, {1, 32, 1, 30}, {960, 30, 30, 1}: 0.00047948ms, 0.0042126% main:@1837 = reshape_lazy[dims={1, 32, 1, 32}](main:@1832) -> float_type, {1, 32, 1, 32}, {1024, 32, 32, 1}: 0.00050046ms, 0.00439693% main:@1838 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1837,main:@1835,main:@1836) -> float_type, {1, 32, 1, 30}, {960, 30, 30, 1}: 0.0186566ms, 0.163913% LjbB2i31Wu:0 = @param:LjbB2i31Wu:0 -> int32_type, {32, 30}, {30, 1}: 0.0007126ms, 0.00626074% main:@1840 = unsqueeze[axes={1},steps={}](LjbB2i31Wu:0) -> int32_type, {32, 1, 30}, {30, 30, 1}: 0.00058806ms, 0.00516656% main:@1841 = load[offset=1594112,end=1597952](main:@1) -> float_type, {32, 1, 30}, {30, 30, 1}: 0.00046864ms, 0.00411737% main:@1842 = gpu::code_object[code_object=13528,symbol_name=convert_add_kernel,global=480,local=256,](main:@1840,main:@1841) -> float_type, {32, 1, 30}, {30, 30, 1}: 0.0132669ms, 0.11656% main:@1843 = multibroadcast[out_lens={1, 32, 1, 30}](main:@1842) -> float_type, {1, 32, 1, 30}, {0, 30, 30, 1}: 0.00069376ms, 0.00609522% main:@1844 = load[offset=1597952,end=1601792](main:@1) -> float_type, {1, 32, 1, 30}, {960, 30, 30, 1}: 0.00048504ms, 0.00426145% main:@1845 = gpu::code_object[code_object=13520,symbol_name=pow_pow_kernel,global=480,local=256,](main:@1838,main:@1843,main:@1844) -> float_type, {1, 32, 1, 30}, {960, 30, 30, 1}: 0.0182518ms, 0.160356% main:@1846 = load[offset=1594112,end=1597952](main:@1) -> float_type, {1, 32, 1, 30}, {960, 30, 30, 1}: 0.00064286ms, 0.00564802% main:@1847 = gpu::code_object[code_object=14584,symbol_name=softmax_kernel,global=2048,local=64,](main:@1845,main:@1846) -> float_type, {1, 32, 1, 30}, {960, 30, 30, 1}: 0.0146845ms, 0.129015% main:@1848 = load[offset=1573632,end=1577728](main:@1) -> float_type, {1, 32, 1, 32}, {1024, 32, 32, 1}: 0.00064166ms, 0.00563748% main:@1849 = slice[axes={1},starts={0},ends={32}](main:@1823) -> float_type, {960, 32}, {64, 1}: 0.00059146ms, 0.00519643% main:@1850 = reshape_lazy[dims={1, 32, 30, 32}](main:@1849) -> float_type, {1, 32, 30, 32}, {61440, 1920, 64, 1}: 0.00059982ms, 0.00526988% main:@1851 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1847,main:@1850,main:@1848) -> float_type, {1, 32, 1, 32}, {1024, 32, 32, 1}: 0.0185002ms, 0.162539% main:@1852 = slice[axes={1},starts={0},ends={32}](main:@652) -> float_type, {32, 32}, {128, 1}: 0.00077132ms, 0.00677664% main:@1853 = load[offset=1594112,end=1598208](main:@1) -> float_type, {32, 32}, {32, 1}: 0.00049746ms, 0.00437057% main:@1854 = gpu::code_object[code_object=13528,symbol_name=contiguous_kernel,global=512,local=256,](main:@1852,main:@1853) -> float_type, {32, 32}, {32, 1}: 0.0129479ms, 0.113757% main:@1855 = load[offset=389248,end=391168](main:@1) -> float_type, {1, 32, 1, 15}, {480, 15, 15, 1}: 0.000662ms, 0.00581618% main:@1856 = slice[axes={1},starts={0},ends={32}](main:@700) -> float_type, {480, 32}, {64, 1}: 0.00089578ms, 0.00787012% main:@1857 = reshape_lazy[dims={1, 32, 15, 32}](main:@1856) -> float_type, {1, 32, 15, 32}, {30720, 960, 64, 1}: 0.00061244ms, 0.00538076% main:@1858 = transpose[permutation={0, 1, 3, 2}](main:@1857) -> float_type, {1, 32, 32, 15}, {30720, 960, 1, 64}: 0.000586ms, 0.00514846% main:@1859 = reshape_lazy[dims={1, 32, 1, 32}](main:@1854) -> float_type, {1, 32, 1, 32}, {1024, 32, 32, 1}: 0.00049334ms, 0.00433437% main:@1860 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1859,main:@1858,main:@1855) -> float_type, {1, 32, 1, 15}, {480, 15, 15, 1}: 0.0203617ms, 0.178893% UaC5MZd8qb:0 = @param:UaC5MZd8qb:0 -> int32_type, {32, 15}, {15, 1}: 0.00069222ms, 0.00608169% main:@1862 = unsqueeze[axes={1},steps={}](UaC5MZd8qb:0) -> int32_type, {32, 1, 15}, {15, 15, 1}: 0.00059354ms, 0.00521471% main:@1863 = load[offset=391168,end=393088](main:@1) -> float_type, {32, 1, 15}, {15, 15, 1}: 0.00047922ms, 0.00421032% main:@1864 = gpu::code_object[code_object=13536,symbol_name=convert_div_mul_kernel,global=240,local=256,](main:@1862,main:@1863) -> float_type, {32, 1, 15}, {15, 15, 1}: 0.0136679ms, 0.120083% main:@1865 = multibroadcast[out_lens={1, 32, 1, 15}](main:@1864) -> float_type, {1, 32, 1, 15}, {0, 15, 15, 1}: 0.00065678ms, 0.00577032% main:@1866 = load[offset=395264,end=397184](main:@1) -> float_type, {1, 32, 1, 15}, {480, 15, 15, 1}: 0.00049994ms, 0.00439236% main:@1867 = gpu::code_object[code_object=13520,symbol_name=add_pow_kernel,global=240,local=256,](main:@1860,main:@1865,main:@1866) -> float_type, {1, 32, 1, 15}, {480, 15, 15, 1}: 0.0162148ms, 0.142459% main:@1868 = load[offset=393344,end=395264](main:@1) -> float_type, {1, 32, 1, 15}, {480, 15, 15, 1}: 0.00060544ms, 0.00531926% main:@1869 = gpu::code_object[code_object=14568,symbol_name=softmax_kernel,global=2048,local=64,](main:@1867,main:@1868) -> float_type, {1, 32, 1, 15}, {480, 15, 15, 1}: 0.0143892ms, 0.12642% main:@1870 = slice[axes={1},starts={32},ends={64}](main:@700) -> float_type, {480, 32}, {64, 1}: 0.00070488ms, 0.00619292% main:@1871 = reshape_lazy[dims={1, 32, 15, 32}](main:@1870) -> float_type, {1, 32, 15, 32}, {30720, 960, 64, 1}: 0.00061166ms, 0.00537391% main:@1872 = load[offset=389248,end=393344](main:@1) -> float_type, {1, 32, 1, 32}, {1024, 32, 32, 1}: 0.00052076ms, 0.00457528% main:@1873 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@1869,main:@1871,main:@1872) -> float_type, {1, 32, 1, 32}, {1024, 32, 32, 1}: 0.0192196ms, 0.168859% Kanv8c4Yl7:0 = @param:Kanv8c4Yl7:0 -> float_type, {32, 201}, {201, 1}: 0.00072266ms, 0.00634913% main:@1875 = slice[axes={0, 1},starts={0, 1},ends={32, 65}](Kanv8c4Yl7:0) -> float_type, {32, 64}, {201, 1}: 0.00068648ms, 0.00603126% main:@1876 = slice[axes={0, 1},starts={0, 193},ends={32, 257}](lyiDxfzZg3:0) -> float_type, {32, 64}, {701, 1}: 0.00077446ms, 0.00680423% main:@1877 = load[offset=1020032,end=1036416](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00046912ms, 0.00412158% main:@1878 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=2048,local=256,](main:@1876,main:@1875,main:@1877) -> float_type, {32, 128}, {128, 1}: 0.0143028ms, 0.125661% main:@1879 = slice[axes={0, 1},starts={0, 1},ends={32, 65}](nv2tzWRf1I:0) -> float_type, {32, 64}, {301, 1}: 0.0008027ms, 0.00705234% main:@1880 = slice[axes={0, 1},starts={0, 65},ends={32, 129}](uFSOc5Chp2:0) -> float_type, {32, 64}, {481, 1}: 0.00099496ms, 0.0087415% main:@1881 = slice[axes={0, 1},starts={0, 1},ends={32, 65}](uFSOc5Chp2:0) -> float_type, {32, 64}, {481, 1}: 0.0006202ms, 0.00544894% main:@1882 = load[offset=1003648,end=1020032](main:@1) -> float_type, {32, 128}, {128, 1}: 0.00047734ms, 0.0041938% main:@1883 = gpu::code_object[code_object=13544,symbol_name=noop_add_concat_div_kernel,global=2048,local=256,](main:@1880,main:@1879,main:@1881,main:@1878,main:@1882) -> float_type, {32, 128}, {128, 1}: 0.0148759ms, 0.130696% main:@1884 = transpose[permutation={1, 2, 0, 3}](main:@1077) -> float_type, {32, 15, 4, 64}, {960, 64, 30720, 1}: 0.00071296ms, 0.00626391% main:@1885 = load[offset=1020032,end=1052800](main:@1) -> float_type, {32, 1, 4, 64}, {64, 64, 2048, 1}: 0.00050654ms, 0.00445035% main:@1886 = gpu::code_object[code_object=13528,symbol_name=reduce_prod_kernel,global=8192,local=256,](main:@1884,main:@1885) -> float_type, {32, 1, 4, 64}, {64, 64, 2048, 1}: 0.0144741ms, 0.127166% main:@1887 = unsqueeze[axes={1},steps={}](main:@1134) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.00071912ms, 0.00631803% main:@1888 = unsqueeze[axes={1},steps={}](main:@1143) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.00050538ms, 0.00444016% main:@1889 = unsqueeze[axes={1},steps={}](main:@1153) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.000491ms, 0.00431382% main:@1890 = unsqueeze[axes={1},steps={}](main:@1163) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.00047702ms, 0.00419099% main:@1891 = unsqueeze[axes={1},steps={}](main:@1173) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.00047962ms, 0.00421383% main:@1892 = unsqueeze[axes={1},steps={}](main:@1182) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.0004927ms, 0.00432875% main:@1893 = unsqueeze[axes={1},steps={}](main:@1191) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.00048922ms, 0.00429818% main:@1894 = unsqueeze[axes={1},steps={}](main:@1200) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.00046636ms, 0.00409733% main:@1895 = unsqueeze[axes={1},steps={}](main:@1209) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.00047434ms, 0.00416744% main:@1896 = unsqueeze[axes={1},steps={}](main:@1219) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.00046724ms, 0.00410507% main:@1897 = unsqueeze[axes={1},steps={}](main:@1228) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.00047958ms, 0.00421348% main:@1898 = unsqueeze[axes={1},steps={}](main:@1237) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.00049012ms, 0.00430608% main:@1899 = unsqueeze[axes={1},steps={}](main:@1246) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.00048446ms, 0.00425636% main:@1900 = unsqueeze[axes={1},steps={}](main:@1256) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.00046424ms, 0.00407871% main:@1901 = unsqueeze[axes={1},steps={}](main:@1266) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.00046696ms, 0.00410261% main:@1902 = unsqueeze[axes={1},steps={}](main:@1275) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.00049636ms, 0.00436091% main:@1903 = unsqueeze[axes={1},steps={}](main:@1302) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.00046468ms, 0.00408257% main:@1904 = unsqueeze[axes={1},steps={}](main:@1311) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.00047414ms, 0.00416569% main:@1905 = load[offset=1700608,end=2290432](main:@1) -> float_type, {32, 18, 256}, {4608, 256, 1}: 0.00047738ms, 0.00419415% main:@1906 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=4096,local=256,](main:@1894,main:@1889,main:@1893,main:@1904,main:@1903,main:@1892,main:@1901,main:@1900,main:@1891,main:@1887,main:@1898,main:@1902,main:@1896,main:@1899,main:@1895,main:@1890,main:@1888,main:@1897,main:@1905) -> float_type, {32, 18, 256}, {4608, 256, 1}: 0.0275428ms, 0.241985% main:@1907 = hip::hip_copy_literal[id=main:@literal:158] -> float_type, {1}, {0}: 0.00090406ms, 0.00794287% main:@1908 = multibroadcast[out_lens={32, 18, 1}](main:@1907) -> float_type, {32, 18, 1}, {0, 0, 0}: 0.0006123ms, 0.00537953% main:@1909 = multibroadcast[out_lens={32, 18, 256}](main:@1908) -> float_type, {32, 18, 256}, {0, 0, 0}: 0.0005442ms, 0.00478122% main:@1910 = load[offset=2868480,end=3458304](main:@1) -> float_type, {32, 18, 256}, {4608, 256, 1}: 0.00046642ms, 0.00409786% main:@1911 = gpu::code_object[code_object=17656,symbol_name=pow_reduce_sum_div_pow_add_kernel,global=73728,local=128,](main:@1906,main:@713,main:@1909,main:@711,main:@1910) -> float_type, {32, 18, 256}, {4608, 256, 1}: 0.0248742ms, 0.218539% main:@1912 = load[offset=413824,end=446592](main:@1) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.00060906ms, 0.00535106% main:@1913 = gpu::code_object[code_object=13528,symbol_name=reduce_sum_kernel,global=8192,local=256,](main:@1911,main:@1912) -> float_type, {32, 1, 256}, {256, 256, 1}: 0.0146871ms, 0.129038% main:@1914 = load[offset=446592,end=459904](main:@1) -> float_type, {32, 1, 104}, {104, 104, 1}: 0.00061876ms, 0.00543629% main:@1915 = gpu::code_object[code_object=13528,symbol_name=reduce_mean_kernel,global=3328,local=256,](main:@1819,main:@1914) -> float_type, {32, 1, 104}, {104, 104, 1}: 0.0145483ms, 0.127818% main:@1916 = load[offset=393344,end=402048](main:@1) -> float_type, {32, 1, 68}, {68, 68, 1}: 0.0006426ms, 0.00564574% main:@1917 = gpu::code_object[code_object=13528,symbol_name=reduce_prod_kernel,global=2176,local=256,](main:@682,main:@1916) -> float_type, {32, 1, 68}, {68, 68, 1}: 0.0145658ms, 0.127972% main:@1918 = transpose[permutation={1, 2, 0, 3}](main:@1543) -> float_type, {32, 1, 2, 64}, {64, 64, 2048, 1}: 0.00069376ms, 0.00609522% main:@1919 = load[offset=128128,end=144512](main:@1) -> float_type, {32, 1, 2, 64}, {128, 128, 64, 1}: 0.00051664ms, 0.00453908% main:@1920 = gpu::code_object[code_object=13528,symbol_name=contiguous_kernel,global=2048,local=256,](main:@1918,main:@1919) -> float_type, {32, 1, 2, 64}, {128, 128, 64, 1}: 0.0132768ms, 0.116647% main:@1921 = load[offset=144512,end=177280](main:@1) -> float_type, {32, 1, 4, 64}, {256, 256, 64, 1}: 0.00063098ms, 0.00554365% main:@1922 = gpu::code_object[code_object=13528,symbol_name=contiguous_kernel,global=4096,local=256,](main:@1886,main:@1921) -> float_type, {32, 1, 4, 64}, {256, 256, 64, 1}: 0.0133291ms, 0.117106% main:@1923 = transpose[permutation={1, 2, 0, 3}](main:@1799) -> float_type, {32, 1, 2, 64}, {64, 64, 2048, 1}: 0.00069096ms, 0.00607062% main:@1924 = load[offset=177280,end=193664](main:@1) -> float_type, {32, 1, 2, 64}, {128, 128, 64, 1}: 0.00048908ms, 0.00429695% main:@1925 = gpu::code_object[code_object=13528,symbol_name=contiguous_kernel,global=2048,local=256,](main:@1923,main:@1924) -> float_type, {32, 1, 2, 64}, {128, 128, 64, 1}: 0.0131969ms, 0.115945% 01E2VTipRX:0 = @param:01E2VTipRX:0 -> float_type, {32, 9}, {9, 1}: 0.00072618ms, 0.00638005% main:@1927 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](01E2VTipRX:0) -> float_type, {32, 8}, {9, 1}: 0.00070042ms, 0.00615373% main:@1928 = transpose[permutation={1, 2, 0, 3}](main:@1099) -> float_type, {32, 1, 1, 64}, {64, 64, 2048, 1}: 0.00050826ms, 0.00446546% main:@1929 = squeeze[axes={1}](main:@1913) -> float_type, {32, 256}, {256, 1}: 0.00057028ms, 0.00501035% u2OJ7kfglD:0 = @param:u2OJ7kfglD:0 -> float_type, {32, 93}, {93, 1}: 0.00053734ms, 0.00472095% main:@1931 = slice[axes={0, 1},starts={0, 61},ends={32, 65}](u2OJ7kfglD:0) -> float_type, {32, 4}, {93, 1}: 0.0008134ms, 0.00714635% main:@1932 = transpose[permutation={1, 2, 0, 3}](main:@1712) -> float_type, {32, 1, 1, 32}, {32, 32, 1024, 1}: 0.00048602ms, 0.00427006% main:@1933 = transpose[permutation={1, 2, 0, 3}](main:@1740) -> float_type, {32, 1, 1, 32}, {32, 32, 1024, 1}: 0.00050174ms, 0.00440818% main:@1934 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](0cjwsL5ldh:0) -> float_type, {32, 16}, {33, 1}: 0.00069066ms, 0.00606798% WycVU13dfZ:0 = @param:WycVU13dfZ:0 -> float_type, {32, 5}, {5, 1}: 0.00052404ms, 0.0046041% main:@1936 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](WycVU13dfZ:0) -> float_type, {32, 4}, {5, 1}: 0.00075006ms, 0.00658986% main:@1937 = transpose[permutation={1, 2, 0, 3}](main:@1762) -> float_type, {32, 1, 1, 32}, {32, 32, 1024, 1}: 0.00049088ms, 0.00431276% main:@1938 = slice[axes={0, 1},starts={0, 5},ends={32, 13}](0sDBrPmfS9:0) -> float_type, {32, 8}, {13, 1}: 0.00096874ms, 0.00851113% main:@1939 = transpose[permutation={1, 2, 0, 3}](main:@1851) -> float_type, {32, 1, 1, 32}, {32, 32, 1024, 1}: 0.00048866ms, 0.00429326% main:@1940 = squeeze[axes={1}](main:@1915) -> float_type, {32, 104}, {104, 1}: 0.00048558ms, 0.0042662% 1f2xpcXgdE:0 = @param:1f2xpcXgdE:0 -> float_type, {32, 13}, {13, 1}: 0.00049448ms, 0.00434439% main:@1942 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](1f2xpcXgdE:0) -> float_type, {32, 4}, {13, 1}: 0.00074866ms, 0.00657756% vZDAiUHErW:0 = @param:vZDAiUHErW:0 -> float_type, {32, 17}, {17, 1}: 0.0005713ms, 0.00501931% main:@1944 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](vZDAiUHErW:0) -> float_type, {32, 8}, {17, 1}: 0.00102016ms, 0.0089629% main:@1945 = squeeze[axes={1}](main:@1917) -> float_type, {32, 68}, {68, 1}: 0.00048944ms, 0.00430011% main:@1946 = slice[axes={0, 1},starts={0, 33},ends={32, 41}](2CnSyUBzm0:0) -> float_type, {32, 8}, {65, 1}: 0.00066568ms, 0.00584852% main:@1947 = transpose[permutation={1, 2, 0, 3}](main:@1873) -> float_type, {32, 1, 1, 32}, {32, 32, 1024, 1}: 0.0004703ms, 0.00413195% main:@1948 = slice[axes={0, 1},starts={0, 5},ends={32, 13}](m5MfQhR1Bp:0) -> float_type, {32, 8}, {13, 1}: 0.00067164ms, 0.00590088% main:@1949 = slice[axes={0, 1},starts={0, 33},ends={32, 37}](2RDXZrx9GA:0) -> float_type, {32, 4}, {45, 1}: 0.00080538ms, 0.00707589% main:@1950 = slice[axes={0, 1},starts={0, 33},ends={32, 37}](2Yy3zxNGBC:0) -> float_type, {32, 4}, {45, 1}: 0.00064938ms, 0.00570531% main:@1951 = slice[axes={0, 1},starts={0, 17},ends={32, 25}](ZRYNFfjSez:0) -> float_type, {32, 8}, {33, 1}: 0.00076718ms, 0.00674027% BQHyb3DGCE:0 = @param:BQHyb3DGCE:0 -> float_type, {32, 9}, {9, 1}: 0.0005351ms, 0.00470127% main:@1953 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](BQHyb3DGCE:0) -> float_type, {32, 8}, {9, 1}: 0.0007416ms, 0.00651553% main:@1954 = slice[axes={0, 1},starts={0, 33},ends={32, 37}](3Laj620MSn:0) -> float_type, {32, 4}, {45, 1}: 0.0006672ms, 0.00586187% RF5h8lytsu:0 = @param:RF5h8lytsu:0 -> float_type, {32, 9}, {9, 1}: 0.0004936ms, 0.00433666% main:@1956 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](RF5h8lytsu:0) -> float_type, {32, 8}, {9, 1}: 0.00075336ms, 0.00661885% EClzvVLqMP:0 = @param:EClzvVLqMP:0 -> float_type, {32, 9}, {9, 1}: 0.00057114ms, 0.00501791% main:@1958 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](EClzvVLqMP:0) -> float_type, {32, 8}, {9, 1}: 0.00071008ms, 0.0062386% main:@1959 = slice[axes={0, 1},starts={0, 17},ends={32, 25}](3mk8BKjDey:0) -> float_type, {32, 8}, {25, 1}: 0.0006915ms, 0.00607536% MQb6fHVndu:0 = @param:MQb6fHVndu:0 -> float_type, {32, 9}, {9, 1}: 0.0005198ms, 0.00456685% main:@1961 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](MQb6fHVndu:0) -> float_type, {32, 8}, {9, 1}: 0.00072712ms, 0.00638831% MtvajoSCeD:0 = @param:MtvajoSCeD:0 -> float_type, {32, 9}, {9, 1}: 0.0005467ms, 0.00480318% main:@1963 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](MtvajoSCeD:0) -> float_type, {32, 8}, {9, 1}: 0.00073208ms, 0.00643189% 4U3hVYNn2F:0 = @param:4U3hVYNn2F:0 -> float_type, {32, 5}, {5, 1}: 0.00057352ms, 0.00503882% main:@1965 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](4U3hVYNn2F:0) -> float_type, {32, 4}, {5, 1}: 0.00077196ms, 0.00678227% 4dtfMgk7Ni:0 = @param:4dtfMgk7Ni:0 -> float_type, {32, 9}, {9, 1}: 0.00056588ms, 0.00497169% main:@1967 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](4dtfMgk7Ni:0) -> float_type, {32, 8}, {9, 1}: 0.00068334ms, 0.00600367% nP9Yv1iZ8f:0 = @param:nP9Yv1iZ8f:0 -> float_type, {32, 9}, {9, 1}: 0.00055118ms, 0.00484254% main:@1969 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](nP9Yv1iZ8f:0) -> float_type, {32, 8}, {9, 1}: 0.00074538ms, 0.00654874% T1FVGiDRMg:0 = @param:T1FVGiDRMg:0 -> float_type, {32, 5}, {5, 1}: 0.0005002ms, 0.00439465% main:@1971 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](T1FVGiDRMg:0) -> float_type, {32, 4}, {5, 1}: 0.00066894ms, 0.00587716% 50ovs1NnYh:0 = @param:50ovs1NnYh:0 -> float_type, {32, 9}, {9, 1}: 0.00061122ms, 0.00537004% main:@1973 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](50ovs1NnYh:0) -> float_type, {32, 8}, {9, 1}: 0.00077194ms, 0.00678209% main:@1974 = slice[axes={0, 1},starts={0, 5},ends={32, 9}](5JHUPChra8:0) -> float_type, {32, 4}, {17, 1}: 0.0007082ms, 0.00622209% 6eQEnYsRTi:0 = @param:6eQEnYsRTi:0 -> float_type, {32, 5}, {5, 1}: 0.00054386ms, 0.00477823% main:@1976 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](6eQEnYsRTi:0) -> float_type, {32, 4}, {5, 1}: 0.0006774ms, 0.00595148% 5bcvLlI6HE:0 = @param:5bcvLlI6HE:0 -> float_type, {32, 9}, {9, 1}: 0.00052272ms, 0.0045925% main:@1978 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](5bcvLlI6HE:0) -> float_type, {32, 8}, {9, 1}: 0.00070356ms, 0.00618132% 5cO3meIa4V:0 = @param:5cO3meIa4V:0 -> float_type, {32, 5}, {5, 1}: 0.00062934ms, 0.00552924% main:@1980 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](5cO3meIa4V:0) -> float_type, {32, 4}, {5, 1}: 0.00065274ms, 0.00573483% Q2Om7B1nhU:0 = @param:Q2Om7B1nhU:0 -> float_type, {32, 9}, {9, 1}: 0.0006105ms, 0.00536372% main:@1982 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](Q2Om7B1nhU:0) -> float_type, {32, 8}, {9, 1}: 0.00071344ms, 0.00626812% main:@1983 = slice[axes={0, 1},starts={0, 33},ends={32, 37}](6HSQqXdJzb:0) -> float_type, {32, 4}, {45, 1}: 0.0006815ms, 0.00598751% main:@1984 = slice[axes={0, 1},starts={0, 5},ends={32, 13}](w27LgiJW6P:0) -> float_type, {32, 8}, {13, 1}: 0.00073976ms, 0.00649937% KgZXYMNz6P:0 = @param:KgZXYMNz6P:0 -> float_type, {32, 13}, {13, 1}: 0.00054862ms, 0.00482005% main:@1986 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](KgZXYMNz6P:0) -> float_type, {32, 4}, {13, 1}: 0.00067176ms, 0.00590193% 6hbLUlYs1H:0 = @param:6hbLUlYs1H:0 -> float_type, {32, 9}, {9, 1}: 0.00052858ms, 0.00464399% main:@1988 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](6hbLUlYs1H:0) -> float_type, {32, 8}, {9, 1}: 0.00067562ms, 0.00593585% y3bafuwnEk:0 = @param:y3bafuwnEk:0 -> float_type, {32, 5}, {5, 1}: 0.00059896ms, 0.00526233% main:@1990 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](y3bafuwnEk:0) -> float_type, {32, 4}, {5, 1}: 0.00064156ms, 0.0056366% main:@1991 = slice[axes={0, 1},starts={0, 25},ends={32, 41}](ifP6Ceko5T:0) -> float_type, {32, 16}, {73, 1}: 0.00065238ms, 0.00573166% 89nc05gQza:0 = @param:89nc05gQza:0 -> float_type, {32, 9}, {9, 1}: 0.0005271ms, 0.00463098% main:@1993 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](89nc05gQza:0) -> float_type, {32, 8}, {9, 1}: 0.00070772ms, 0.00621787% 8KRQILP469:0 = @param:8KRQILP469:0 -> float_type, {32, 9}, {9, 1}: 0.0005025ms, 0.00441485% main:@1995 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](8KRQILP469:0) -> float_type, {32, 8}, {9, 1}: 0.00067178ms, 0.00590211% VP35W0GOCc:0 = @param:VP35W0GOCc:0 -> float_type, {32, 13}, {13, 1}: 0.00055784ms, 0.00490106% main:@1997 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](VP35W0GOCc:0) -> float_type, {32, 4}, {13, 1}: 0.00069176ms, 0.00607765% 8iOST90kWM:0 = @param:8iOST90kWM:0 -> float_type, {32, 9}, {9, 1}: 0.00053356ms, 0.00468774% main:@1999 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](8iOST90kWM:0) -> float_type, {32, 8}, {9, 1}: 0.0006912ms, 0.00607273% 8Dv01FAoNM:0 = @param:8Dv01FAoNM:0 -> float_type, {32, 17}, {17, 1}: 0.00054676ms, 0.00480371% main:@2001 = slice[axes={0, 1},starts={0, 1},ends={32, 17}](8Dv01FAoNM:0) -> float_type, {32, 16}, {17, 1}: 0.00068914ms, 0.00605463% main:@2002 = slice[axes={0, 1},starts={0, 33},ends={32, 37}](X29iyIeu8L:0) -> float_type, {32, 4}, {45, 1}: 0.00066396ms, 0.0058334% 985IRiY0X6:0 = @param:985IRiY0X6:0 -> float_type, {32, 57}, {57, 1}: 0.0005607ms, 0.00492618% main:@2004 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](985IRiY0X6:0) -> float_type, {32, 8}, {57, 1}: 0.00068878ms, 0.00605147% main:@2005 = slice[axes={0, 1},starts={0, 41},ends={32, 49}](985IRiY0X6:0) -> float_type, {32, 8}, {57, 1}: 0.0006677ms, 0.00586626% main:@2006 = slice[axes={0, 1},starts={0, 17},ends={32, 25}](985IRiY0X6:0) -> float_type, {32, 8}, {57, 1}: 0.00065658ms, 0.00576856% main:@2007 = slice[axes={0, 1},starts={0, 49},ends={32, 57}](985IRiY0X6:0) -> float_type, {32, 8}, {57, 1}: 0.00067478ms, 0.00592847% 9N1KUX0DZ5:0 = @param:9N1KUX0DZ5:0 -> float_type, {32, 13}, {13, 1}: 0.00052708ms, 0.00463081% main:@2009 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](9N1KUX0DZ5:0) -> float_type, {32, 4}, {13, 1}: 0.0006997ms, 0.00614741% main:@2010 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](985IRiY0X6:0) -> float_type, {32, 8}, {57, 1}: 0.000722ms, 0.00634333% main:@2011 = slice[axes={0, 1},starts={0, 33},ends={32, 41}](985IRiY0X6:0) -> float_type, {32, 8}, {57, 1}: 0.00069394ms, 0.0060968% main:@2012 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](9YqhflbMKe:0) -> float_type, {32, 4}, {29, 1}: 0.00068506ms, 0.00601878% main:@2013 = slice[axes={0, 1},starts={0, 33},ends={32, 37}](VUqfdN5Fzv:0) -> float_type, {32, 4}, {45, 1}: 0.0006598ms, 0.00579685% main:@2014 = slice[axes={0, 1},starts={0, 25},ends={32, 33}](985IRiY0X6:0) -> float_type, {32, 8}, {57, 1}: 0.00065614ms, 0.0057647% 9mfMQZIXVH:0 = @param:9mfMQZIXVH:0 -> float_type, {32, 9}, {9, 1}: 0.0005236ms, 0.00460023% main:@2016 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](9mfMQZIXVH:0) -> float_type, {32, 8}, {9, 1}: 0.0006828ms, 0.00599893% 9rDQhGZo18:0 = @param:9rDQhGZo18:0 -> float_type, {32, 9}, {9, 1}: 0.00061154ms, 0.00537285% main:@2018 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](9rDQhGZo18:0) -> float_type, {32, 8}, {9, 1}: 0.0007301ms, 0.00641449% main:@2019 = slice[axes={0, 1},starts={0, 17},ends={32, 25}](l4bReWtsMB:0) -> float_type, {32, 8}, {25, 1}: 0.00068176ms, 0.00598979% main:@2020 = slice[axes={0, 1},starts={0, 9},ends={32, 13}](AVLcBpWeuo:0) -> float_type, {32, 4}, {21, 1}: 0.00069864ms, 0.00613809% AfxiFXbSj3:0 = @param:AfxiFXbSj3:0 -> float_type, {32, 5}, {5, 1}: 0.0005725ms, 0.00502986% main:@2022 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](AfxiFXbSj3:0) -> float_type, {32, 4}, {5, 1}: 0.00069014ms, 0.00606342% AvlwbHMnuE:0 = @param:AvlwbHMnuE:0 -> float_type, {32, 17}, {17, 1}: 0.00056434ms, 0.00495816% main:@2024 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](AvlwbHMnuE:0) -> float_type, {32, 8}, {17, 1}: 0.00068134ms, 0.0059861% Nb9eDKFYZp:0 = @param:Nb9eDKFYZp:0 -> float_type, {32, 5}, {5, 1}: 0.00051428ms, 0.00451835% main:@2026 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](Nb9eDKFYZp:0) -> float_type, {32, 4}, {5, 1}: 0.00064712ms, 0.00568545% KH4GEDx8pO:0 = @param:KH4GEDx8pO:0 -> float_type, {32, 13}, {13, 1}: 0.00055186ms, 0.00484852% main:@2028 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](KH4GEDx8pO:0) -> float_type, {32, 4}, {13, 1}: 0.0006997ms, 0.00614741% CrB2gE9Adp:0 = @param:CrB2gE9Adp:0 -> float_type, {32, 13}, {13, 1}: 0.000523ms, 0.00459496% main:@2030 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](CrB2gE9Adp:0) -> float_type, {32, 4}, {13, 1}: 0.0006814ms, 0.00598663% uhcrWfUe6M:0 = @param:uhcrWfUe6M:0 -> float_type, {32, 9}, {9, 1}: 0.0005349ms, 0.00469951% main:@2032 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](uhcrWfUe6M:0) -> float_type, {32, 8}, {9, 1}: 0.0007106ms, 0.00624317% main:@2033 = slice[axes={0, 1},starts={0, 33},ends={32, 41}](n4iptOVFwE:0) -> float_type, {32, 8}, {113, 1}: 0.00069512ms, 0.00610717% j25yTFI0eZ:0 = @param:j25yTFI0eZ:0 -> float_type, {32, 9}, {9, 1}: 0.00064708ms, 0.0056851% main:@2035 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](j25yTFI0eZ:0) -> float_type, {32, 8}, {9, 1}: 0.00066974ms, 0.00588419% dMO2bknDYA:0 = @param:dMO2bknDYA:0 -> float_type, {32, 481}, {481, 1}: 0.00065218ms, 0.00572991% main:@2037 = slice[axes={0, 1},starts={0, 233},ends={32, 241}](dMO2bknDYA:0) -> float_type, {32, 8}, {481, 1}: 0.00073386ms, 0.00644753% EJxze1pkRu:0 = @param:EJxze1pkRu:0 -> float_type, {32, 13}, {13, 1}: 0.00053946ms, 0.00473957% main:@2039 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](EJxze1pkRu:0) -> float_type, {32, 4}, {13, 1}: 0.00067604ms, 0.00593954% DilwT5jYf7:0 = @param:DilwT5jYf7:0 -> float_type, {32, 9}, {9, 1}: 0.0005071ms, 0.00445527% main:@2041 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](DilwT5jYf7:0) -> float_type, {32, 8}, {9, 1}: 0.0006641ms, 0.00583463% main:@2042 = slice[axes={0, 1},starts={0, 33},ends={32, 37}](Qp3K1PHvte:0) -> float_type, {32, 4}, {69, 1}: 0.00068072ms, 0.00598065% EyFb0qo6VX:0 = @param:EyFb0qo6VX:0 -> float_type, {32, 5}, {5, 1}: 0.00056704ms, 0.00498189% main:@2044 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](EyFb0qo6VX:0) -> float_type, {32, 4}, {5, 1}: 0.00070944ms, 0.00623298% main:@2045 = slice[axes={0, 1},starts={0, 5},ends={32, 13}](mIR0ToEnMH:0) -> float_type, {32, 8}, {13, 1}: 0.0006952ms, 0.00610787% F6DhiS9CN5:0 = @param:F6DhiS9CN5:0 -> float_type, {32, 9}, {9, 1}: 0.00056988ms, 0.00500684% main:@2047 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](F6DhiS9CN5:0) -> float_type, {32, 8}, {9, 1}: 0.00070012ms, 0.0061511% NKT10PBtly:0 = @param:NKT10PBtly:0 -> float_type, {32, 5}, {5, 1}: 0.00056196ms, 0.00493725% main:@2049 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](NKT10PBtly:0) -> float_type, {32, 4}, {5, 1}: 0.00065846ms, 0.00578508% zOu1aQKy4V:0 = @param:zOu1aQKy4V:0 -> float_type, {32, 9}, {9, 1}: 0.00052984ms, 0.00465506% main:@2051 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](zOu1aQKy4V:0) -> float_type, {32, 8}, {9, 1}: 0.000696ms, 0.0061149% EsROupwkSG:0 = @param:EsROupwkSG:0 -> float_type, {32, 13}, {13, 1}: 0.00056222ms, 0.00493954% main:@2053 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](EsROupwkSG:0) -> float_type, {32, 4}, {13, 1}: 0.00068288ms, 0.00599963% FrTLGaKo7c:0 = @param:FrTLGaKo7c:0 -> float_type, {32, 13}, {13, 1}: 0.00057626ms, 0.00506289% main:@2055 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](FrTLGaKo7c:0) -> float_type, {32, 4}, {13, 1}: 0.00068882ms, 0.00605182% mDpoFQLqWK:0 = @param:mDpoFQLqWK:0 -> float_type, {32, 13}, {13, 1}: 0.00057216ms, 0.00502687% main:@2057 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](mDpoFQLqWK:0) -> float_type, {32, 4}, {13, 1}: 0.00068258ms, 0.00599699% HFesjVwyB6:0 = @param:HFesjVwyB6:0 -> float_type, {32, 9}, {9, 1}: 0.00056204ms, 0.00493796% main:@2059 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](HFesjVwyB6:0) -> float_type, {32, 8}, {9, 1}: 0.00065544ms, 0.00575855% HKq2ZSlA3D:0 = @param:HKq2ZSlA3D:0 -> float_type, {32, 9}, {9, 1}: 0.00065382ms, 0.00574432% main:@2061 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](HKq2ZSlA3D:0) -> float_type, {32, 8}, {9, 1}: 0.00066422ms, 0.00583569% HLQJSVgFyi:0 = @param:HLQJSVgFyi:0 -> float_type, {32, 9}, {9, 1}: 0.00054008ms, 0.00474502% main:@2063 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](HLQJSVgFyi:0) -> float_type, {32, 8}, {9, 1}: 0.0007061ms, 0.00620364% cNgUKoOFB1:0 = @param:cNgUKoOFB1:0 -> float_type, {32, 9}, {9, 1}: 0.00057914ms, 0.00508819% main:@2065 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](cNgUKoOFB1:0) -> float_type, {32, 8}, {9, 1}: 0.00064012ms, 0.00562395% nIJL1FUMdk:0 = @param:nIJL1FUMdk:0 -> float_type, {32, 9}, {9, 1}: 0.00057368ms, 0.00504022% main:@2067 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](nIJL1FUMdk:0) -> float_type, {32, 8}, {9, 1}: 0.00065082ms, 0.00571796% main:@2068 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](ljyhc7HCLk:0) -> float_type, {32, 4}, {21, 1}: 0.00068186ms, 0.00599067% HyzMmXUNAP:0 = @param:HyzMmXUNAP:0 -> float_type, {32, 13}, {13, 1}: 0.00050012ms, 0.00439394% main:@2070 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](HyzMmXUNAP:0) -> float_type, {32, 4}, {13, 1}: 0.0006434ms, 0.00565277% main:@2071 = slice[axes={0, 1},starts={0, 33},ends={32, 37}](HzcyUlm8Yj:0) -> float_type, {32, 4}, {45, 1}: 0.00064942ms, 0.00570566% fvGJwyjlm4:0 = @param:fvGJwyjlm4:0 -> float_type, {32, 13}, {13, 1}: 0.00050942ms, 0.00447565% main:@2073 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](fvGJwyjlm4:0) -> float_type, {32, 4}, {13, 1}: 0.00072272ms, 0.00634966% IQawVv9sqx:0 = @param:IQawVv9sqx:0 -> float_type, {32, 9}, {9, 1}: 0.00056452ms, 0.00495975% main:@2075 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](IQawVv9sqx:0) -> float_type, {32, 8}, {9, 1}: 0.00067546ms, 0.00593444% main:@2076 = slice[axes={0, 1},starts={0, 321},ends={32, 501}](lyiDxfzZg3:0) -> float_type, {32, 180}, {701, 1}: 0.00066278ms, 0.00582304% fKsuaR6OE2:0 = @param:fKsuaR6OE2:0 -> float_type, {32, 13}, {13, 1}: 0.0005885ms, 0.00517043% main:@2078 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](fKsuaR6OE2:0) -> float_type, {32, 4}, {13, 1}: 0.00068436ms, 0.00601263% main:@2079 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](gdck97AK4w:0) -> float_type, {32, 8}, {17, 1}: 0.0006858ms, 0.00602529% JKWg9U1qmT:0 = @param:JKWg9U1qmT:0 -> float_type, {32, 13}, {13, 1}: 0.00054172ms, 0.00475943% main:@2081 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](JKWg9U1qmT:0) -> float_type, {32, 4}, {13, 1}: 0.00065404ms, 0.00574625% hADUwpE6P7:0 = @param:hADUwpE6P7:0 -> float_type, {32, 9}, {9, 1}: 0.00051714ms, 0.00454348% main:@2083 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](hADUwpE6P7:0) -> float_type, {32, 8}, {9, 1}: 0.00069292ms, 0.00608784% YMxBt9210u:0 = @param:YMxBt9210u:0 -> float_type, {32, 5}, {5, 1}: 0.00059846ms, 0.00525794% main:@2085 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](YMxBt9210u:0) -> float_type, {32, 4}, {5, 1}: 0.00063894ms, 0.00561358% KBG8U2uYOC:0 = @param:KBG8U2uYOC:0 -> float_type, {32, 9}, {9, 1}: 0.0006281ms, 0.00551835% main:@2087 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](KBG8U2uYOC:0) -> float_type, {32, 8}, {9, 1}: 0.00066476ms, 0.00584043% main:@2088 = slice[axes={0, 1},starts={0, 5},ends={32, 13}](KFhBukd7EV:0) -> float_type, {32, 8}, {13, 1}: 0.00071214ms, 0.0062567% KGspl1Fzg0:0 = @param:KGspl1Fzg0:0 -> float_type, {32, 9}, {9, 1}: 0.00053258ms, 0.00467913% main:@2090 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](KGspl1Fzg0:0) -> float_type, {32, 8}, {9, 1}: 0.0006679ms, 0.00586802% main:@2091 = slice[axes={0, 1},starts={0, 161},ends={32, 193}](Kanv8c4Yl7:0) -> float_type, {32, 32}, {201, 1}: 0.00064844ms, 0.00569705% XevLAmJgSy:0 = @param:XevLAmJgSy:0 -> float_type, {32, 5}, {5, 1}: 0.00050834ms, 0.00446616% main:@2093 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](XevLAmJgSy:0) -> float_type, {32, 4}, {5, 1}: 0.0006971ms, 0.00612456% KOdpo5kqsI:0 = @param:KOdpo5kqsI:0 -> float_type, {32, 93}, {93, 1}: 0.00053118ms, 0.00466683% main:@2095 = slice[axes={0, 1},starts={0, 33},ends={32, 49}](KOdpo5kqsI:0) -> float_type, {32, 16}, {93, 1}: 0.00064734ms, 0.00568738% main:@2096 = slice[axes={0, 1},starts={0, 129},ends={32, 161}](Kanv8c4Yl7:0) -> float_type, {32, 32}, {201, 1}: 0.00065108ms, 0.00572024% main:@2097 = slice[axes={0, 1},starts={0, 1},ends={32, 33}](KOdpo5kqsI:0) -> float_type, {32, 32}, {93, 1}: 0.0006468ms, 0.00568264% main:@2098 = slice[axes={0, 1},starts={0, 65},ends={32, 97}](Kanv8c4Yl7:0) -> float_type, {32, 32}, {201, 1}: 0.00064674ms, 0.00568211% main:@2099 = slice[axes={0, 1},starts={0, 97},ends={32, 129}](Kanv8c4Yl7:0) -> float_type, {32, 32}, {201, 1}: 0.00063172ms, 0.00555015% M4pau3Zg7l:0 = @param:M4pau3Zg7l:0 -> float_type, {32, 13}, {13, 1}: 0.00056346ms, 0.00495043% main:@2101 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](M4pau3Zg7l:0) -> float_type, {32, 4}, {13, 1}: 0.00067848ms, 0.00596097% Ya2E8PbliH:0 = @param:Ya2E8PbliH:0 -> float_type, {32, 9}, {9, 1}: 0.0005656ms, 0.00496923% main:@2103 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](Ya2E8PbliH:0) -> float_type, {32, 8}, {9, 1}: 0.00068642ms, 0.00603073% NFBzi76hoK:0 = @param:NFBzi76hoK:0 -> float_type, {32, 73}, {73, 1}: 0.0005778ms, 0.00507642% main:@2105 = slice[axes={0, 1},starts={0, 1},ends={32, 65}](NFBzi76hoK:0) -> float_type, {32, 64}, {73, 1}: 0.0006909ms, 0.00607009% main:@2106 = slice[axes={0, 1},starts={0, 33},ends={32, 37}](duyRoIEkH8:0) -> float_type, {32, 4}, {45, 1}: 0.00068442ms, 0.00601316% OfNJgEDtw2:0 = @param:OfNJgEDtw2:0 -> float_type, {32, 13}, {13, 1}: 0.0005903ms, 0.00518624% main:@2108 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](OfNJgEDtw2:0) -> float_type, {32, 4}, {13, 1}: 0.00064336ms, 0.00565242% main:@2109 = slice[axes={0, 1},starts={0, 9},ends={32, 17}](Pu1O5qFxY6:0) -> float_type, {32, 8}, {17, 1}: 0.00064104ms, 0.00563203% PxhmZizCI5:0 = @param:PxhmZizCI5:0 -> float_type, {32, 13}, {13, 1}: 0.00050756ms, 0.00445931% main:@2111 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](PxhmZizCI5:0) -> float_type, {32, 4}, {13, 1}: 0.00067878ms, 0.00596361% QMjOWELyPk:0 = @param:QMjOWELyPk:0 -> float_type, {32, 9}, {9, 1}: 0.0005125ms, 0.00450271% main:@2113 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](QMjOWELyPk:0) -> float_type, {32, 8}, {9, 1}: 0.00066278ms, 0.00582304% mQnuwGeC8a:0 = @param:mQnuwGeC8a:0 -> float_type, {32, 9}, {9, 1}: 0.0005481ms, 0.00481548% main:@2115 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](mQnuwGeC8a:0) -> float_type, {32, 8}, {9, 1}: 0.00071044ms, 0.00624177% main:@2116 = slice[axes={0, 1},starts={0, 9},ends={32, 13}](RHo9Jz6KpI:0) -> float_type, {32, 4}, {21, 1}: 0.00072262ms, 0.00634878% RNGK6g4bmu:0 = @param:RNGK6g4bmu:0 -> float_type, {32, 9}, {9, 1}: 0.0005599ms, 0.00491916% main:@2118 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](RNGK6g4bmu:0) -> float_type, {32, 8}, {9, 1}: 0.00068886ms, 0.00605217% V84toiBOPm:0 = @param:V84toiBOPm:0 -> float_type, {32, 9}, {9, 1}: 0.00057142ms, 0.00502037% main:@2120 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](V84toiBOPm:0) -> float_type, {32, 8}, {9, 1}: 0.00067442ms, 0.0059253% main:@2121 = slice[axes={0, 1},starts={0, 5},ends={32, 13}](S5AzoJmqUC:0) -> float_type, {32, 8}, {13, 1}: 0.00070544ms, 0.00619784% main:@2122 = slice[axes={0, 1},starts={0, 165},ends={32, 181}](nv2tzWRf1I:0) -> float_type, {32, 16}, {301, 1}: 0.00066132ms, 0.00581021% main:@2123 = slice[axes={0, 1},starts={0, 17},ends={32, 21}](YbIVz8l1un:0) -> float_type, {32, 4}, {45, 1}: 0.0006776ms, 0.00595324% main:@2124 = slice[axes={0, 1},starts={0, 17},ends={32, 21}](bJsIAzhYc3:0) -> float_type, {32, 4}, {37, 1}: 0.00066884ms, 0.00587628% main:@2125 = slice[axes={0, 1},starts={0, 53},ends={32, 57}](u2OJ7kfglD:0) -> float_type, {32, 4}, {93, 1}: 0.00065812ms, 0.00578209% TkHQvxKeD6:0 = @param:TkHQvxKeD6:0 -> float_type, {32, 9}, {9, 1}: 0.0005445ms, 0.00478385% main:@2127 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](TkHQvxKeD6:0) -> float_type, {32, 8}, {9, 1}: 0.00067242ms, 0.00590773% UTzDZk1RS4:0 = @param:UTzDZk1RS4:0 -> float_type, {32, 13}, {13, 1}: 0.00053088ms, 0.00466419% main:@2129 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](UTzDZk1RS4:0) -> float_type, {32, 4}, {13, 1}: 0.0006988ms, 0.0061395% main:@2130 = slice[axes={0, 1},starts={0, 5},ends={32, 13}](qeZ76Hc03v:0) -> float_type, {32, 8}, {13, 1}: 0.00067112ms, 0.00589631% bhuXRdz0E5:0 = @param:bhuXRdz0E5:0 -> float_type, {32, 9}, {9, 1}: 0.00062512ms, 0.00549216% main:@2132 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](bhuXRdz0E5:0) -> float_type, {32, 8}, {9, 1}: 0.00066936ms, 0.00588085% ToDLE1sRI6:0 = @param:ToDLE1sRI6:0 -> float_type, {32, 9}, {9, 1}: 0.0005247ms, 0.0046099% main:@2134 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](ToDLE1sRI6:0) -> float_type, {32, 8}, {9, 1}: 0.00066508ms, 0.00584324% zvFYW45wMr:0 = @param:zvFYW45wMr:0 -> float_type, {32, 9}, {9, 1}: 0.00055286ms, 0.0048573% main:@2136 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](zvFYW45wMr:0) -> float_type, {32, 8}, {9, 1}: 0.00072296ms, 0.00635176% main:@2137 = slice[axes={0, 1},starts={0, 33},ends={32, 41}](1aXQU27NrD:0) -> float_type, {32, 8}, {49, 1}: 0.00065542ms, 0.00575837% uGCZaoYr3L:0 = @param:uGCZaoYr3L:0 -> float_type, {32, 13}, {13, 1}: 0.00047754ms, 0.00419556% main:@2139 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](uGCZaoYr3L:0) -> float_type, {32, 4}, {13, 1}: 0.00072546ms, 0.00637373% ihcEmsYo9g:0 = @param:ihcEmsYo9g:0 -> float_type, {32, 5}, {5, 1}: 0.00054994ms, 0.00483165% main:@2141 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](ihcEmsYo9g:0) -> float_type, {32, 4}, {5, 1}: 0.0006637ms, 0.00583112% vi7GCYp4FT:0 = @param:vi7GCYp4FT:0 -> float_type, {32, 9}, {9, 1}: 0.00054238ms, 0.00476523% main:@2143 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](vi7GCYp4FT:0) -> float_type, {32, 8}, {9, 1}: 0.00068274ms, 0.0059984% ihKZd8ANk7:0 = @param:ihKZd8ANk7:0 -> float_type, {32, 9}, {9, 1}: 0.00054492ms, 0.00478754% main:@2145 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](ihKZd8ANk7:0) -> float_type, {32, 8}, {9, 1}: 0.00066286ms, 0.00582374% kX6xhVY7bJ:0 = @param:kX6xhVY7bJ:0 -> float_type, {32, 13}, {13, 1}: 0.00056458ms, 0.00496027% main:@2147 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](kX6xhVY7bJ:0) -> float_type, {32, 4}, {13, 1}: 0.00071326ms, 0.00626654% ad9BJGx5jl:0 = @param:ad9BJGx5jl:0 -> float_type, {32, 13}, {13, 1}: 0.00062306ms, 0.00547407% main:@2149 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](ad9BJGx5jl:0) -> float_type, {32, 4}, {13, 1}: 0.000648ms, 0.00569318% alvtrHJ95G:0 = @param:alvtrHJ95G:0 -> float_type, {32, 13}, {13, 1}: 0.00048346ms, 0.00424757% main:@2151 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](alvtrHJ95G:0) -> float_type, {32, 4}, {13, 1}: 0.00065814ms, 0.00578227% aocCEyfXIS:0 = @param:aocCEyfXIS:0 -> float_type, {32, 93}, {93, 1}: 0.00053114ms, 0.00466648% main:@2153 = slice[axes={0, 1},starts={0, 1},ends={32, 33}](aocCEyfXIS:0) -> float_type, {32, 32}, {93, 1}: 0.00063548ms, 0.00558318% main:@2154 = slice[axes={0, 1},starts={0, 33},ends={32, 49}](aocCEyfXIS:0) -> float_type, {32, 16}, {93, 1}: 0.00065274ms, 0.00573483% bsFqHW5i6g:0 = @param:bsFqHW5i6g:0 -> float_type, {32, 9}, {9, 1}: 0.00061232ms, 0.00537971% main:@2156 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](bsFqHW5i6g:0) -> float_type, {32, 8}, {9, 1}: 0.00069926ms, 0.00614354% main:@2157 = slice[axes={0, 1},starts={0, 17},ends={32, 33}](cHzIDJm6us:0) -> float_type, {32, 16}, {33, 1}: 0.00072508ms, 0.00637039% gwNpMszEo3:0 = @param:gwNpMszEo3:0 -> float_type, {32, 5}, {5, 1}: 0.00054942ms, 0.00482708% main:@2159 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](gwNpMszEo3:0) -> float_type, {32, 4}, {5, 1}: 0.00068088ms, 0.00598206% ceq9LkmQA5:0 = @param:ceq9LkmQA5:0 -> float_type, {32, 9}, {9, 1}: 0.0005209ms, 0.00457651% main:@2161 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](ceq9LkmQA5:0) -> float_type, {32, 8}, {9, 1}: 0.00067952ms, 0.00597011% main:@2162 = slice[axes={0, 1},starts={0, 33},ends={32, 37}](jznORDJr0N:0) -> float_type, {32, 4}, {45, 1}: 0.0006636ms, 0.00583024% main:@2163 = slice[axes={0, 1},starts={0, 5},ends={32, 13}](gU9SoiHyRO:0) -> float_type, {32, 8}, {13, 1}: 0.00066932ms, 0.0058805% main:@2164 = slice[axes={0, 1},starts={0, 65},ends={32, 129}](lyiDxfzZg3:0) -> float_type, {32, 64}, {701, 1}: 0.00068592ms, 0.00602634% main:@2165 = slice[axes={0, 1},starts={0, 33},ends={32, 37}](zIoqYe9cRb:0) -> float_type, {32, 4}, {45, 1}: 0.00064798ms, 0.00569301% gHPTxdcQWX:0 = @param:gHPTxdcQWX:0 -> float_type, {32, 9}, {9, 1}: 0.0005067ms, 0.00445175% main:@2167 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](gHPTxdcQWX:0) -> float_type, {32, 8}, {9, 1}: 0.00067772ms, 0.0059543% main:@2168 = slice[axes={0, 1},starts={0, 57},ends={32, 61}](u2OJ7kfglD:0) -> float_type, {32, 4}, {93, 1}: 0.00064946ms, 0.00570601% main:@2169 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](w04Keh2FJu:0) -> float_type, {32, 8}, {13, 1}: 0.00067104ms, 0.00589561% main:@2170 = slice[axes={0, 1},starts={0, 177},ends={32, 209}](uFSOc5Chp2:0) -> float_type, {32, 32}, {481, 1}: 0.00065158ms, 0.00572464% sfqQKWXEgN:0 = @param:sfqQKWXEgN:0 -> float_type, {32, 5}, {5, 1}: 0.0005287ms, 0.00464504% main:@2172 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](sfqQKWXEgN:0) -> float_type, {32, 4}, {5, 1}: 0.00075212ms, 0.00660796% main:@2173 = slice[axes={0, 1},starts={0, 33},ends={32, 37}](dAFYcGrfEl:0) -> float_type, {32, 4}, {45, 1}: 0.000669ms, 0.00587768% tZOFD39jRS:0 = @param:tZOFD39jRS:0 -> float_type, {32, 9}, {9, 1}: 0.00060556ms, 0.00532031% main:@2175 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](tZOFD39jRS:0) -> float_type, {32, 8}, {9, 1}: 0.00069408ms, 0.00609803% vPaCFg3sAl:0 = @param:vPaCFg3sAl:0 -> float_type, {32, 13}, {13, 1}: 0.00051542ms, 0.00452836% main:@2177 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](vPaCFg3sAl:0) -> float_type, {32, 4}, {13, 1}: 0.0006798ms, 0.00597257% main:@2178 = slice[axes={0, 1},starts={0, 33},ends={32, 49}](u2OJ7kfglD:0) -> float_type, {32, 16}, {93, 1}: 0.00065376ms, 0.00574379% fpcFBPsTyk:0 = @param:fpcFBPsTyk:0 -> float_type, {32, 9}, {9, 1}: 0.0005701ms, 0.00500877% main:@2180 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](fpcFBPsTyk:0) -> float_type, {32, 8}, {9, 1}: 0.00066492ms, 0.00584184% main:@2181 = slice[axes={0, 1},starts={0, 129},ends={32, 193}](lyiDxfzZg3:0) -> float_type, {32, 64}, {701, 1}: 0.00065308ms, 0.00573781% uzaUY08Rsp:0 = @param:uzaUY08Rsp:0 -> float_type, {32, 9}, {9, 1}: 0.0005288ms, 0.00464592% main:@2183 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](uzaUY08Rsp:0) -> float_type, {32, 8}, {9, 1}: 0.0007159ms, 0.00628974% main:@2184 = slice[axes={0, 1},starts={0, 5},ends={32, 13}](whRnWGgf0Q:0) -> float_type, {32, 8}, {13, 1}: 0.00071226ms, 0.00625776% sOz8pkBZ5n:0 = @param:sOz8pkBZ5n:0 -> float_type, {32, 5}, {5, 1}: 0.00056236ms, 0.00494077% main:@2186 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](sOz8pkBZ5n:0) -> float_type, {32, 4}, {5, 1}: 0.0006866ms, 0.00603231% sXCjeM94Zu:0 = @param:sXCjeM94Zu:0 -> float_type, {32, 5}, {5, 1}: 0.00053042ms, 0.00466015% main:@2188 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](sXCjeM94Zu:0) -> float_type, {32, 4}, {5, 1}: 0.0006841ms, 0.00601035% main:@2189 = slice[axes={0, 1},starts={0, 17},ends={32, 25}](l5WSX3VvMK:0) -> float_type, {32, 8}, {25, 1}: 0.00064776ms, 0.00569107% main:@2190 = slice[axes={0, 1},starts={0, 49},ends={32, 53}](u2OJ7kfglD:0) -> float_type, {32, 4}, {93, 1}: 0.00066068ms, 0.00580459% dXvEYVC4st:0 = @param:dXvEYVC4st:0 -> float_type, {32, 13}, {13, 1}: 0.00048704ms, 0.00427902% main:@2192 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](dXvEYVC4st:0) -> float_type, {32, 4}, {13, 1}: 0.0006747ms, 0.00592776% dz1saATctE:0 = @param:dz1saATctE:0 -> float_type, {32, 9}, {9, 1}: 0.00054382ms, 0.00477788% main:@2194 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](dz1saATctE:0) -> float_type, {32, 8}, {9, 1}: 0.00071042ms, 0.00624159% main:@2195 = slice[axes={0, 1},starts={0, 1},ends={32, 65}](lyiDxfzZg3:0) -> float_type, {32, 64}, {701, 1}: 0.00067802ms, 0.00595693% v4fNptDqcL:0 = @param:v4fNptDqcL:0 -> float_type, {32, 9}, {9, 1}: 0.00056944ms, 0.00500297% main:@2197 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](v4fNptDqcL:0) -> float_type, {32, 8}, {9, 1}: 0.00068688ms, 0.00603477% main:@2198 = slice[axes={0, 1},starts={0, 5},ends={32, 13}](d1ExPaGqMN:0) -> float_type, {32, 8}, {13, 1}: 0.00068396ms, 0.00600912% pIb4NdDSn5:0 = @param:pIb4NdDSn5:0 -> float_type, {32, 9}, {9, 1}: 0.00056244ms, 0.00494147% main:@2200 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](pIb4NdDSn5:0) -> float_type, {32, 8}, {9, 1}: 0.00068552ms, 0.00602282% dR9G1bayok:0 = @param:dR9G1bayok:0 -> float_type, {32, 13}, {13, 1}: 0.00053464ms, 0.00469723% main:@2202 = slice[axes={0, 1},starts={0, 1},ends={32, 5}](dR9G1bayok:0) -> float_type, {32, 4}, {13, 1}: 0.00064228ms, 0.00564293% fqAX8l29PG:0 = @param:fqAX8l29PG:0 -> float_type, {32, 9}, {9, 1}: 0.0005168ms, 0.00454049% main:@2204 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](fqAX8l29PG:0) -> float_type, {32, 8}, {9, 1}: 0.00072584ms, 0.00637707% main:@2205 = slice[axes={0, 1},starts={0, 33},ends={32, 41}](mgnGNH4Mz3:0) -> float_type, {32, 8}, {49, 1}: 0.00068022ms, 0.00597626% main:@2206 = slice[axes={0, 1},starts={0, 1},ends={32, 33}](u2OJ7kfglD:0) -> float_type, {32, 32}, {93, 1}: 0.00066996ms, 0.00588612% main:@2207 = slice[axes={0, 1},starts={0, 65},ends={32, 129}](nv2tzWRf1I:0) -> float_type, {32, 64}, {301, 1}: 0.00065344ms, 0.00574098% sjKQJGby31:0 = @param:sjKQJGby31:0 -> float_type, {32, 9}, {9, 1}: 0.00051576ms, 0.00453135% main:@2209 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](sjKQJGby31:0) -> float_type, {32, 8}, {9, 1}: 0.00068286ms, 0.00599945% main:@2210 = slice[axes={0, 1},starts={0, 5},ends={32, 13}](Vi28Dy4Ll5:0) -> float_type, {32, 8}, {13, 1}: 0.0006436ms, 0.00565453% aKmfBy1Mhi:0 = @param:aKmfBy1Mhi:0 -> float_type, {32, 9}, {9, 1}: 0.00050658ms, 0.0044507% uM7aDfZz1t:0 = @param:uM7aDfZz1t:0 -> float_type, {32, 9}, {9, 1}: 0.0005724ms, 0.00502898% Wiks0GPJMQ:0 = @param:Wiks0GPJMQ:0 -> float_type, {32, 9}, {9, 1}: 0.00048958ms, 0.00430134% main:@2214 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](Wiks0GPJMQ:0) -> float_type, {32, 8}, {9, 1}: 0.0008031ms, 0.00705586% main:@2215 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](aKmfBy1Mhi:0) -> float_type, {32, 8}, {9, 1}: 0.00067ms, 0.00588647% main:@2216 = slice[axes={0, 1},starts={0, 1},ends={32, 9}](uM7aDfZz1t:0) -> float_type, {32, 8}, {9, 1}: 0.0006654ms, 0.00584606% main:@2217 = reshape_lazy[dims={-1, 32}](main:@1939) -> float_type, {32, 32}, {32, 1}: 0.00052442ms, 0.00460744% main:@2218 = reshape_lazy[dims={-1, 128}](main:@1920) -> float_type, {32, 128}, {128, 1}: 0.0004956ms, 0.00435423% main:@2219 = reshape_lazy[dims={-1, 84}](main:@1685) -> float_type, {32, 84}, {84, 1}: 0.00048672ms, 0.00427621% main:@2220 = reshape_lazy[dims={-1, 32}](main:@1932) -> float_type, {32, 32}, {32, 1}: 0.00048398ms, 0.00425214% main:@2221 = load[offset=1700608,end=2139520](main:@1) -> float_type, {32, 3429}, {3429, 1}: 0.00047406ms, 0.00416498% main:@2222 = reshape_lazy[dims={-1, 32}](main:@1947) -> float_type, {32, 32}, {32, 1}: 0.0004879ms, 0.00428658% main:@2223 = reshape_lazy[dims={-1, 105}](main:@987) -> float_type, {32, 105}, {105, 1}: 0.00046602ms, 0.00409435% main:@2224 = reshape_lazy[dims={-1, 64}](main:@1032) -> float_type, {32, 64}, {64, 1}: 0.00046296ms, 0.00406746% main:@2225 = reshape_lazy[dims={-1, 256}](main:@1922) -> float_type, {32, 256}, {256, 1}: 0.00047552ms, 0.00417781% main:@2226 = reshape_lazy[dims={-1, 100}](main:@1505) -> float_type, {32, 100}, {100, 1}: 0.0004789ms, 0.00420751% main:@2227 = reshape_lazy[dims={-1, 64}](main:@1928) -> float_type, {32, 64}, {64, 1}: 0.00046058ms, 0.00404655% main:@2228 = reshape_lazy[dims={-1, 32}](main:@1933) -> float_type, {32, 32}, {32, 1}: 0.00045728ms, 0.00401756% main:@2229 = reshape_lazy[dims={-1, 32}](main:@1937) -> float_type, {32, 32}, {32, 1}: 0.00045798ms, 0.00402371% main:@2230 = reshape_lazy[dims={-1, 128}](main:@1925) -> float_type, {32, 128}, {128, 1}: 0.00046696ms, 0.00410261% main:@2231 = gpu::code_object[code_object=72784,symbol_name=pow_div_add_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_noop_concat_log_add_kernel,global=574,local=256,](main:@2195,main:@1879,main:@2164,main:@2207,main:@2181,main:@1881,main:@1883,main:@2105,main:@2206,main:@2097,main:@2153,main:@2223,main:@2219,main:@2226,main:@2076,main:@2122,main:@2170,main:@1991,main:@2042,main:@1946,main:@2162,main:@1949,main:@2013,main:@2071,main:@1954,main:@2106,main:@1944,main:@2053,main:@2205,main:@2137,main:@1983,main:@2165,main:@2173,main:@2002,main:@2108,main:@2124,main:@2123,main:@2079,main:@2109,main:@1951,main:@1950,main:@2019,main:@2189,main:@2020,main:@2200,main:@2169,main:@2032,main:@2194,main:@2116,main:@2024,main:@1974,main:@1971,main:@1953,main:@2012,main:@2068,main:@2178,main:@2095,main:@2154,main:@2098,main:@2099,main:@2096,main:@2091,main:@2190,main:@2125,main:@2168,main:@1931,main:@2218,main:@2049,main:@2044,main:@2159,main:@2172,main:@2022,main:@1980,main:@2186,main:@2026,main:@1936,main:@2188,main:@2230,main:@2224,main:@2010,main:@2004,main:@2006,main:@2014,main:@2011,main:@2005,main:@2007,main:@2073,main:@2081,main:@2009,main:@2057,main:@2078,main:@2028,main:@1942,main:@2101,main:@1986,main:@2039,main:@1997,main:@2055,main:@2129,main:@2151,main:@2147,main:@2111,main:@2177,main:@2030,main:@2149,main:@2222,main:@2227,main:@2229,main:@2202,main:@2139,main:@2192,main:@2070,main:@1988,main:@2127,main:@1999,main:@1993,main:@2047,main:@2093,main:@2085,main:@1976,main:@1990,main:@1965,main:@2141,main:@2037,main:@2134,main:@1963,main:@2132,main:@2214,main:@2204,main:@2090,main:@2143,main:@2156,main:@1938,main:@2183,main:@1973,main:@1961,main:@1982,main:@1959,main:@1995,main:@2180,main:@1958,main:@2075,main:@2016,main:@2167,main:@2209,main:@1969,main:@2145,main:@1927,main:@2087,main:@2216,main:@2157,main:@2065,main:@2061,main:@2136,main:@2083,main:@2059,main:@2063,main:@2001,main:@1934,main:@2220,main:@2051,main:@2130,main:@2210,main:@1948,main:@2067,main:@2045,main:@2161,main:@2198,main:@2121,main:@2184,main:@2175,main:@1984,main:@2041,main:@2215,main:@2035,main:@2118,main:@2103,main:@2113,main:@2120,main:@2018,main:@2115,main:@2197,main:@1978,main:@1967,main:@1956,main:@2033,main:@2163,main:@2088,main:@2228,main:@1945,main:@2217,main:@1940,main:@2225,main:@1929,main:@1113,main:@2221) -> float_type, {32, 3429}, {3429, 1}: 0.231511ms, 2.034% main:@2232 = hip::hip_copy_literal[id=main:@literal:28] -> float_type, {3429, 2905}, {2905, 1}: 0.0008477ms, 0.0074477% main:@2233 = hip::hip_copy_literal[id=main:@literal:30] -> float_type, {32, 2905}, {2905, 1}: 0.00077202ms, 0.00678279% main:@2234 = load[offset=691712,end=1063552](main:@1) -> float_type, {32, 2905}, {2905, 1}: 0.0005293ms, 0.00465031% main:@2235 = gpu::gemm[alpha=1,beta=1,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@2231,main:@2232,main:@2233,main:@2234) -> float_type, {32, 2905}, {2905, 1}: 0.0842752ms, 0.740423% main:@2236 = hip::hip_copy_literal[id=main:@literal:4] -> float_type, {581, 581}, {581, 1}: 0.00084308ms, 0.00740711% main:@2237 = slice[axes={1},starts={1743},ends={2324}](main:@2235) -> float_type, {32, 581}, {2905, 1}: 0.00066452ms, 0.00583832% main:@2238 = load[offset=1137920,end=1212288](main:@1) -> float_type, {32, 581}, {581, 1}: 0.00050074ms, 0.00439939% main:@2239 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@2237,main:@2236,main:@2238) -> float_type, {32, 581}, {581, 1}: 0.0361612ms, 0.317704% main:@2240 = hip::hip_copy_literal[id=main:@literal:2] -> float_type, {32, 4508}, {4508, 1}: 0.00084754ms, 0.0074463% main:@2241 = hip::hip_copy_literal[id=main:@literal:20] -> float_type, {128, 4508}, {4508, 1}: 0.00061966ms, 0.00544419% main:@2242 = slice[axes={1},starts={384},ends={512}](main:@1109) -> float_type, {32, 128}, {896, 1}: 0.0006842ms, 0.00601123% main:@2243 = load[offset=114688,end=691712](main:@1) -> float_type, {32, 4508}, {4508, 1}: 0.00046854ms, 0.00411649% main:@2244 = gpu::gemm[alpha=1,beta=1,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@2242,main:@2241,main:@2240,main:@2243) -> float_type, {32, 4508}, {4508, 1}: 0.0267969ms, 0.235432% main:@2245 = load[offset=1063552,end=1137920](main:@1) -> float_type, {32, 581}, {581, 1}: 0.00065112ms, 0.00572059% main:@2246 = gpu::code_object[code_object=13544,symbol_name=pow_mul_add_neg_sub_kernel,global=18592,local=256,](main:@2237,main:@2239,main:@2245) -> float_type, {32, 581}, {581, 1}: 0.0171976ms, 0.151094% main:@2247 = slice[axes={1},starts={1162},ends={1743}](main:@2235) -> float_type, {32, 581}, {2905, 1}: 0.00074272ms, 0.00652537% main:@2248 = slice[axes={1},starts={0},ends={581}](main:@2235) -> float_type, {32, 581}, {2905, 1}: 0.0005767ms, 0.00506676% main:@2249 = load[offset=1137920,end=1212288](main:@1) -> float_type, {32, 581}, {581, 1}: 0.00050452ms, 0.0044326% main:@2250 = gpu::code_object[code_object=13544,symbol_name=pow_pow_sigmoid_add_kernel,global=18592,local=256,](main:@2247,main:@2248,main:@2249) -> float_type, {32, 581}, {581, 1}: 0.0178393ms, 0.156732% main:@2251 = slice[axes={1},starts={2324},ends={2905}](main:@2235) -> float_type, {32, 581}, {2905, 1}: 0.00073344ms, 0.00644384% main:@2252 = slice[axes={1},starts={581},ends={1162}](main:@2235) -> float_type, {32, 581}, {2905, 1}: 0.0005559ms, 0.00488401% main:@2253 = load[offset=1212288,end=1286656](main:@1) -> float_type, {32, 581}, {581, 1}: 0.00045996ms, 0.00404111% main:@2254 = gpu::code_object[code_object=13536,symbol_name=pow_mul_rsqrt_sub_kernel,global=18592,local=256,](main:@2251,main:@2252,main:@2253) -> float_type, {32, 581}, {581, 1}: 0.0170131ms, 0.149473% main:@2255 = load[offset=1286656,end=1286784](main:@1) -> float_type, {32, 1}, {1, 1}: 0.0006505ms, 0.00571515% main:@2256 = gpu::code_object[code_object=13552,symbol_name=reduce_mean_add_rsqrt_add_kernel,global=8192,local=256,](main:@2254,main:@2255) -> float_type, {32, 1}, {1, 1}: 0.0151047ms, 0.132706% main:@2257 = load[offset=1286784,end=1286912](main:@1) -> float_type, {32, 1}, {1, 1}: 0.00060156ms, 0.00528517% main:@2258 = gpu::code_object[code_object=13552,symbol_name=reduce_mean_add_neg_pow_kernel,global=8192,local=256,](main:@2246,main:@2257) -> float_type, {32, 1}, {1, 1}: 0.015347ms, 0.134835% main:@2259 = load[offset=1286912,end=1287040](main:@1) -> float_type, {32, 1}, {1, 1}: 0.00062634ms, 0.00550288% main:@2260 = gpu::code_object[code_object=13552,symbol_name=reduce_prod_add_rsqrt_sub_kernel,global=8192,local=256,](main:@2250,main:@2259) -> float_type, {32, 1}, {1, 1}: 0.0147131ms, 0.129266% main:@2261 = load[offset=1867392,end=2164864](main:@1) -> float_type, {32, 2324}, {2324, 1}: 0.000612ms, 0.00537689% main:@2262 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=18592,local=256,](main:@2247,main:@2248,main:@2251,main:@2237,main:@2261) -> float_type, {32, 2324}, {2324, 1}: 0.0177783ms, 0.156196% main:@2263 = slice[axes={1},starts={0},ends={4070}](main:@2244) -> float_type, {32, 4070}, {4508, 1}: 0.00069ms, 0.00606219% main:@2264 = load[offset=1346432,end=1867392](main:@1) -> float_type, {32, 4070}, {4070, 1}: 0.0004717ms, 0.00414425% main:@2265 = gpu::code_object[code_object=21808,symbol_name=leaky_relu_noop_noop_noop_noop_noop_noop_concat_rsqrt_pow_kernel,global=18605,local=256,](main:@2262,main:@2250,main:@2254,main:@2246,main:@2260,main:@2256,main:@2258,main:@2263,main:@2264) -> float_type, {32, 4070}, {4070, 1}: 0.0360556ms, 0.316776% main:@2266 = hip::hip_copy_literal[id=main:@literal:1] -> float_type, {32, 2130}, {2130, 1}: 0.00080856ms, 0.00710383% main:@2267 = hip::hip_copy_literal[id=main:@literal:31] -> float_type, {4070, 2130}, {2130, 1}: 0.0005929ms, 0.00520909% main:@2268 = load[offset=1073792,end=1346432](main:@1) -> float_type, {32, 2130}, {2130, 1}: 0.00051266ms, 0.00450412% 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, {32, 2130}, {2130, 1}: 0.0755241ms, 0.663538% main:@2270 = hip::hip_copy_literal[id=main:@literal:92] -> float_type, {426, 426}, {426, 1}: 0.00075468ms, 0.00663045% main:@2271 = slice[axes={1},starts={852},ends={1278}](main:@2269) -> float_type, {32, 426}, {2130, 1}: 0.0006129ms, 0.0053848% main:@2272 = load[offset=1400960,end=1455488](main:@1) -> float_type, {32, 426}, {426, 1}: 0.0004798ms, 0.00421542% main:@2273 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@2271,main:@2270,main:@2272) -> float_type, {32, 426}, {426, 1}: 0.0303257ms, 0.266435% main:@2274 = hip::hip_copy_literal[id=main:@literal:114] -> float_type, {128, 2985}, {2985, 1}: 0.00080402ms, 0.00706394% main:@2275 = slice[axes={1},starts={768},ends={896}](main:@1109) -> float_type, {32, 128}, {896, 1}: 0.00060942ms, 0.00535423% main:@2276 = load[offset=691712,end=1073792](main:@1) -> float_type, {32, 2985}, {2985, 1}: 0.00050512ms, 0.00443787% main:@2277 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@2275,main:@2274,main:@2276) -> float_type, {32, 2985}, {2985, 1}: 0.0231605ms, 0.203483% main:@2278 = hip::hip_copy_literal[id=main:@literal:17] -> float_type, {32, 2}, {2, 1}: 0.0007764ms, 0.00682128% main:@2279 = load[offset=1346432,end=1400960](main:@1) -> float_type, {32, 426}, {426, 1}: 0.00052444ms, 0.00460761% main:@2280 = gpu::code_object[code_object=13544,symbol_name=sub_sub_mul_rsqrt_sub_kernel,global=6816,local=256,](main:@2271,main:@2273,main:@2279) -> float_type, {32, 426}, {426, 1}: 0.0144784ms, 0.127204% main:@2281 = slice[axes={1},starts={1704},ends={2130}](main:@2269) -> float_type, {32, 426}, {2130, 1}: 0.00071736ms, 0.00630256% main:@2282 = slice[axes={1},starts={426},ends={852}](main:@2269) -> float_type, {32, 426}, {2130, 1}: 0.00058006ms, 0.00509628% main:@2283 = load[offset=1400960,end=1455488](main:@1) -> float_type, {32, 426}, {426, 1}: 0.00050676ms, 0.00445228% main:@2284 = gpu::code_object[code_object=13536,symbol_name=div_mul_neg_pow_kernel,global=6816,local=256,](main:@2282,main:@2281,main:@2283) -> float_type, {32, 426}, {426, 1}: 0.0152286ms, 0.133795% main:@2285 = slice[axes={1},starts={1278},ends={1704}](main:@2269) -> float_type, {32, 426}, {2130, 1}: 0.00070786ms, 0.0062191% main:@2286 = slice[axes={1},starts={0},ends={426}](main:@2269) -> float_type, {32, 426}, {2130, 1}: 0.00055326ms, 0.00486082% main:@2287 = load[offset=1455488,end=1510016](main:@1) -> float_type, {32, 426}, {426, 1}: 0.0004583ms, 0.00402652% main:@2288 = gpu::code_object[code_object=13536,symbol_name=add_pow_log_pow_kernel,global=6816,local=256,](main:@2285,main:@2286,main:@2287) -> float_type, {32, 426}, {426, 1}: 0.0175068ms, 0.153811% main:@2289 = load[offset=1510016,end=1728128](main:@1) -> float_type, {32, 1704}, {1704, 1}: 0.00059944ms, 0.00526655% main:@2290 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=13632,local=256,](main:@2285,main:@2286,main:@2282,main:@2271,main:@2289) -> float_type, {32, 1704}, {1704, 1}: 0.0167146ms, 0.14685% main:@2291 = load[offset=1073920,end=1074048](main:@1) -> float_type, {32, 1}, {1, 1}: 0.00061482ms, 0.00540167% main:@2292 = gpu::code_object[code_object=13544,symbol_name=reduce_mean_mul_neg_kernel,global=8192,local=256,](main:@2284,main:@2291) -> float_type, {32, 1}, {1, 1}: 0.0143064ms, 0.125693% main:@2293 = load[offset=1073792,end=1073920](main:@1) -> float_type, {32, 1}, {1, 1}: 0.00060114ms, 0.00528148% main:@2294 = gpu::code_object[code_object=13560,symbol_name=reduce_mean_add_sigmoid_sub_kernel,global=8192,local=256,](main:@2288,main:@2293) -> float_type, {32, 1}, {1, 1}: 0.0144356ms, 0.126828% main:@2295 = load[offset=1074304,end=1074432](main:@1) -> float_type, {32, 1}, {1, 1}: 0.0006068ms, 0.00533121% main:@2296 = gpu::code_object[code_object=13544,symbol_name=reduce_prod_mul_log_kernel,global=8192,local=256,](main:@2280,main:@2295) -> float_type, {32, 1}, {1, 1}: 0.0142658ms, 0.125336% main:@2297 = load[offset=1074048,end=1074304](main:@1) -> float_type, {32, 2}, {2, 1}: 0.00061314ms, 0.00538691% main:@2298 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=32,local=256,](main:@2292,main:@2296,main:@2297) -> float_type, {32, 2}, {2, 1}: 0.0131102ms, 0.115183% main:@2299 = load[offset=1728128,end=2110208](main:@1) -> float_type, {32, 2985}, {2985, 1}: 0.00062864ms, 0.00552309% main:@2300 = gpu::code_object[code_object=21792,symbol_name=leaky_relu_noop_noop_noop_noop_add_concat_log_pow_kernel,global=15920,local=256,](main:@2290,main:@2288,main:@2284,main:@2280,main:@2294,main:@2278,main:@2298,main:@2277,main:@2299) -> float_type, {32, 2985}, {2985, 1}: 0.0330968ms, 0.290781% main:@2301 = hip::hip_copy_literal[id=main:@literal:18] -> float_type, {32, 856}, {856, 1}: 0.0007553ms, 0.0066359% main:@2302 = hip::hip_copy_literal[id=main:@literal:29] -> float_type, {2985, 856}, {856, 1}: 0.000631ms, 0.00554382% main:@2303 = load[offset=883840,end=993408](main:@1) -> float_type, {32, 856}, {856, 1}: 0.0004978ms, 0.00437356% 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, {32, 856}, {856, 1}: 0.0653826ms, 0.574437% main:@2305 = hip::hip_copy_literal[id=main:@literal:161] -> float_type, {214, 214}, {214, 1}: 0.00076952ms, 0.00676083% main:@2306 = slice[axes={1},starts={642},ends={856}](main:@2304) -> float_type, {32, 214}, {856, 1}: 0.00059252ms, 0.00520575% main:@2307 = load[offset=1020800,end=1048192](main:@1) -> float_type, {32, 214}, {214, 1}: 0.00048658ms, 0.00427498% main:@2308 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@2306,main:@2305,main:@2307) -> float_type, {32, 214}, {214, 1}: 0.0261318ms, 0.229588% main:@2309 = hip::hip_copy_literal[id=main:@literal:43] -> float_type, {128, 1501}, {1501, 1}: 0.00078624ms, 0.00690773% main:@2310 = slice[axes={1},starts={0},ends={128}](main:@1109) -> float_type, {32, 128}, {896, 1}: 0.00061412ms, 0.00539552% main:@2311 = load[offset=691712,end=883840](main:@1) -> float_type, {32, 1501}, {1501, 1}: 0.00046078ms, 0.00404831% main:@2312 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@2310,main:@2309,main:@2311) -> float_type, {32, 1501}, {1501, 1}: 0.0235059ms, 0.206518% main:@2313 = hip::hip_copy_literal[id=main:@literal:148] -> float_type, {1}, {0}: 0.00079756ms, 0.00700718% main:@2314 = hip::hip_copy_literal[id=main:@literal:126] -> float_type, {1, 214}, {214, 1}: 0.00060326ms, 0.00530011% main:@2315 = multibroadcast[out_lens={32, 214}](main:@2314) -> float_type, {32, 214}, {0, 1}: 0.00069248ms, 0.00608397% main:@2316 = load[offset=993408,end=1020800](main:@1) -> float_type, {32, 214}, {214, 1}: 0.00050556ms, 0.00444174% main:@2317 = gpu::code_object[code_object=17648,symbol_name=pow_sub_div_pow_log_div_kernel,global=3424,local=256,](main:@2308,main:@2315,main:@2306,main:@2316) -> float_type, {32, 214}, {214, 1}: 0.0203936ms, 0.179174% main:@2318 = slice[axes={1},starts={214},ends={428}](main:@2304) -> float_type, {32, 214}, {856, 1}: 0.00072898ms, 0.00640465% main:@2319 = multibroadcast[out_lens={214}](main:@2313) -> float_type, {214}, {0}: 0.00056038ms, 0.00492337% main:@2320 = multibroadcast[out_lens={32, 214}](main:@2319) -> float_type, {32, 214}, {0, 0}: 0.00067546ms, 0.00593444% main:@2321 = load[offset=1020800,end=1048192](main:@1) -> float_type, {32, 214}, {214, 1}: 0.00048366ms, 0.00424933% main:@2322 = gpu::code_object[code_object=13536,symbol_name=add_add_neg_add_kernel,global=3424,local=256,](main:@2318,main:@2320,main:@2321) -> float_type, {32, 214}, {214, 1}: 0.0146189ms, 0.128438% main:@2323 = slice[axes={1},starts={428},ends={642}](main:@2304) -> float_type, {32, 214}, {856, 1}: 0.00072528ms, 0.00637215% main:@2324 = slice[axes={1},starts={0},ends={214}](main:@2304) -> float_type, {32, 214}, {856, 1}: 0.00059748ms, 0.00524933% main:@2325 = load[offset=1048192,end=1075584](main:@1) -> float_type, {32, 214}, {214, 1}: 0.00045148ms, 0.0039666% main:@2326 = gpu::code_object[code_object=13536,symbol_name=add_add_rsqrt_pow_kernel,global=3424,local=256,](main:@2324,main:@2323,main:@2320,main:@2325) -> float_type, {32, 214}, {214, 1}: 0.0151244ms, 0.13288% main:@2327 = load[offset=1075584,end=1185152](main:@1) -> float_type, {32, 856}, {856, 1}: 0.00062214ms, 0.00546598% main:@2328 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=6848,local=256,](main:@2318,main:@2318,main:@2324,main:@2306,main:@2327) -> float_type, {32, 856}, {856, 1}: 0.0157561ms, 0.138429% main:@2329 = load[offset=883840,end=883968](main:@1) -> float_type, {32, 1}, {1, 1}: 0.00060644ms, 0.00532805% main:@2330 = gpu::code_object[code_object=13552,symbol_name=reduce_prod_pow_neg_sub_kernel,global=4096,local=128,](main:@2317,main:@2329) -> float_type, {32, 1}, {1, 1}: 0.0149391ms, 0.131251% main:@2331 = load[offset=883968,end=884096](main:@1) -> float_type, {32, 1}, {1, 1}: 0.0006086ms, 0.00534702% main:@2332 = gpu::code_object[code_object=13552,symbol_name=reduce_prod_add_log_div_kernel,global=4096,local=128,](main:@2326,main:@2331) -> float_type, {32, 1}, {1, 1}: 0.014472ms, 0.127148% main:@2333 = load[offset=884096,end=884224](main:@1) -> float_type, {32, 1}, {1, 1}: 0.0006055ms, 0.00531979% main:@2334 = gpu::code_object[code_object=13552,symbol_name=reduce_prod_mul_rsqrt_add_kernel,global=4096,local=128,](main:@2322,main:@2333) -> float_type, {32, 1}, {1, 1}: 0.0143466ms, 0.126046% main:@2335 = hip::hip_copy_literal[id=main:@literal:0] -> float_type, {32, 400}, {400, 1}: 0.00076338ms, 0.00670689% main:@2336 = hip::hip_copy_literal[id=main:@literal:32] -> float_type, {1501, 400}, {400, 1}: 0.00055914ms, 0.00491248% main:@2337 = load[offset=1185152,end=1377280](main:@1) -> float_type, {32, 1501}, {1501, 1}: 0.00048726ms, 0.00428096% main:@2338 = gpu::code_object[code_object=17720,symbol_name=leaky_relu_noop_noop_noop_noop_noop_noop_concat_log_add_pow_kernel,global=6861,local=256,](main:@2328,main:@2322,main:@2326,main:@2317,main:@2334,main:@2332,main:@2330,main:@2312,main:@2337) -> float_type, {32, 1501}, {1501, 1}: 0.0320615ms, 0.281685% main:@2339 = load[offset=781696,end=832896](main:@1) -> float_type, {32, 400}, {400, 1}: 0.0005977ms, 0.00525126% 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, {32, 400}, {400, 1}: 0.031521ms, 0.276936% main:@2341 = hip::hip_copy_literal[id=main:@literal:52] -> float_type, {128, 703}, {703, 1}: 0.0008ms, 0.00702862% main:@2342 = slice[axes={1},starts={256},ends={384}](main:@1109) -> float_type, {32, 128}, {896, 1}: 0.00071752ms, 0.00630397% main:@2343 = load[offset=691712,end=781696](main:@1) -> float_type, {32, 703}, {703, 1}: 0.00045516ms, 0.00399893% main:@2344 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@2342,main:@2341,main:@2343) -> float_type, {32, 703}, {703, 1}: 0.0212454ms, 0.186658% main:@2345 = hip::hip_copy_literal[id=main:@literal:3] -> float_type, {100, 100}, {100, 1}: 0.0007584ms, 0.00666313% main:@2346 = slice[axes={1},starts={200},ends={300}](main:@2340) -> float_type, {32, 100}, {400, 1}: 0.0005832ms, 0.00512386% main:@2347 = load[offset=858496,end=871296](main:@1) -> float_type, {32, 100}, {100, 1}: 0.00049162ms, 0.00431926% 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, {32, 100}, {100, 1}: 0.0224631ms, 0.197355% main:@2349 = slice[axes={1},starts={300},ends={400}](main:@2340) -> float_type, {32, 100}, {400, 1}: 0.00074046ms, 0.00650552% main:@2350 = slice[axes={1},starts={0},ends={100}](main:@2340) -> float_type, {32, 100}, {400, 1}: 0.00058998ms, 0.00518343% main:@2351 = load[offset=832896,end=845696](main:@1) -> float_type, {32, 100}, {100, 1}: 0.00050008ms, 0.00439359% main:@2352 = gpu::code_object[code_object=13536,symbol_name=add_add_rsqrt_add_kernel,global=1600,local=256,](main:@2349,main:@2350,main:@2351) -> float_type, {32, 100}, {100, 1}: 0.0142525ms, 0.125219% main:@2353 = load[offset=845696,end=858496](main:@1) -> float_type, {32, 100}, {100, 1}: 0.00061756ms, 0.00542574% main:@2354 = gpu::code_object[code_object=13544,symbol_name=sub_sub_pow_log_pow_kernel,global=1600,local=256,](main:@2346,main:@2348,main:@2353) -> float_type, {32, 100}, {100, 1}: 0.0173261ms, 0.152223% main:@2355 = slice[axes={1},starts={100},ends={200}](main:@2340) -> float_type, {32, 100}, {400, 1}: 0.00070952ms, 0.00623368% main:@2356 = load[offset=858496,end=871296](main:@1) -> float_type, {32, 100}, {100, 1}: 0.00047416ms, 0.00416586% main:@2357 = gpu::code_object[code_object=13536,symbol_name=add_add_neg_add_kernel,global=1600,local=256,](main:@2355,main:@2350,main:@2356) -> float_type, {32, 100}, {100, 1}: 0.014158ms, 0.124389% main:@2358 = hip::hip_copy_literal[id=main:@literal:22] -> float_type, {703, 212}, {212, 1}: 0.00075106ms, 0.00659864% main:@2359 = hip::hip_copy_literal[id=main:@literal:33] -> float_type, {32, 212}, {212, 1}: 0.00071056ms, 0.00624282% main:@2360 = load[offset=871424,end=871552](main:@1) -> float_type, {32, 1}, {1, 1}: 0.00047572ms, 0.00417957% main:@2361 = gpu::code_object[code_object=14256,symbol_name=reduce_mean_add_sigmoid_kernel,global=2048,local=64,](main:@2352,main:@2360) -> float_type, {32, 1}, {1, 1}: 0.0144028ms, 0.126539% main:@2362 = load[offset=871296,end=871424](main:@1) -> float_type, {32, 1}, {1, 1}: 0.00062336ms, 0.0054767% main:@2363 = gpu::code_object[code_object=14256,symbol_name=reduce_mean_pow_log_pow_kernel,global=2048,local=64,](main:@2354,main:@2362) -> float_type, {32, 1}, {1, 1}: 0.0151859ms, 0.13342% main:@2364 = load[offset=922752,end=922880](main:@1) -> float_type, {32, 1}, {1, 1}: 0.00060334ms, 0.00530081% main:@2365 = gpu::code_object[code_object=14256,symbol_name=reduce_mean_mul_rsqrt_kernel,global=2048,local=64,](main:@2357,main:@2364) -> float_type, {32, 1}, {1, 1}: 0.014338ms, 0.125971% main:@2366 = load[offset=871552,end=922752](main:@1) -> float_type, {32, 400}, {400, 1}: 0.00060096ms, 0.0052799% main:@2367 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=3200,local=256,](main:@2349,main:@2350,main:@2355,main:@2346,main:@2366) -> float_type, {32, 400}, {400, 1}: 0.0156056ms, 0.137107% main:@2368 = load[offset=781696,end=781952](main:@1) -> float_type, {32, 2}, {2, 1}: 0.0006194ms, 0.00544191% main:@2369 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=32,local=256,](main:@2361,main:@2365,main:@2368) -> float_type, {32, 2}, {2, 1}: 0.0131322ms, 0.115376% main:@2370 = load[offset=922752,end=1012736](main:@1) -> float_type, {32, 703}, {703, 1}: 0.00060592ms, 0.00532348% main:@2371 = gpu::code_object[code_object=13600,symbol_name=leaky_relu_noop_noop_noop_add_noop_concat_log_div_kernel,global=3749,local=256,](main:@2367,main:@2352,main:@2357,main:@2354,main:@2278,main:@2369,main:@2363,main:@2344,main:@2370) -> float_type, {32, 703}, {703, 1}: 0.0221243ms, 0.194379% main:@2372 = load[offset=691712,end=718848](main:@1) -> float_type, {32, 212}, {212, 1}: 0.00060552ms, 0.00531996% 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, {32, 212}, {212, 1}: 0.0370514ms, 0.325525% main:@2374 = hip::hip_copy_literal[id=main:@literal:81] -> float_type, {128, 167}, {167, 1}: 0.00075894ms, 0.00666788% main:@2375 = slice[axes={1},starts={512},ends={640}](main:@1109) -> float_type, {32, 128}, {896, 1}: 0.00061382ms, 0.00539288% main:@2376 = load[offset=718848,end=740224](main:@1) -> float_type, {32, 167}, {167, 1}: 0.00047334ms, 0.00415866% main:@2377 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@2375,main:@2374,main:@2376) -> float_type, {32, 167}, {167, 1}: 0.0220979ms, 0.194147% main:@2378 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](0cjwsL5ldh:0) -> float_type, {32, 1}, {33, 1}: 0.00083776ms, 0.00736037% main:@2379 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](gsP96ZTbn0:0) -> float_type, {32, 1}, {17, 1}: 0.0006846ms, 0.00601474% main:@2380 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](1PJsBhDaCK:0) -> float_type, {32, 1}, {25, 1}: 0.00077546ms, 0.00681302% main:@2381 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](1QWFmJ96aB:0) -> float_type, {32, 1}, {5, 1}: 0.0006091ms, 0.00535142% main:@2382 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](1WNvHElUG7:0) -> float_type, {32, 1}, {25, 1}: 0.00073488ms, 0.00645649% main:@2383 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](2CnSyUBzm0:0) -> float_type, {32, 1}, {65, 1}: 0.00071002ms, 0.00623808% main:@2384 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](2RDXZrx9GA:0) -> float_type, {32, 1}, {45, 1}: 0.00084164ms, 0.00739446% main:@2385 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](2Yy3zxNGBC:0) -> float_type, {32, 1}, {45, 1}: 0.00067422ms, 0.00592355% main:@2386 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](2lVXMZgJfO:0) -> float_type, {32, 1}, {25, 1}: 0.00070316ms, 0.00617781% main:@2387 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](2srtUfGvja:0) -> float_type, {32, 1}, {41, 1}: 0.00068526ms, 0.00602054% main:@2388 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](3VED7WSJ1b:0) -> float_type, {32, 1}, {5, 1}: 0.00079676ms, 0.00700015% main:@2389 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](gHPTxdcQWX:0) -> float_type, {32, 1}, {9, 1}: 0.00065728ms, 0.00577471% main:@2390 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](3kdi0vsSZG:0) -> float_type, {32, 1}, {25, 1}: 0.00071932ms, 0.00631978% main:@2391 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](z3lqgZM8WU:0) -> float_type, {32, 1}, {41, 1}: 0.00067866ms, 0.00596255% main:@2392 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](4M92ryD8ux:0) -> float_type, {32, 1}, {9, 1}: 0.00081384ms, 0.00715022% main:@2393 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](4gFa5o2YsJ:0) -> float_type, {32, 1}, {9, 1}: 0.00067662ms, 0.00594463% main:@2394 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](5yColzZ8K3:0) -> float_type, {32, 1}, {5, 1}: 0.00072292ms, 0.00635141% main:@2395 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](6HSQqXdJzb:0) -> float_type, {32, 1}, {45, 1}: 0.00067434ms, 0.0059246% main:@2396 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](6Wh2Jm7GHC:0) -> float_type, {32, 1}, {5, 1}: 0.00080944ms, 0.00711156% main:@2397 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](6XMrIsBjLY:0) -> float_type, {32, 1}, {13, 1}: 0.00070616ms, 0.00620416% main:@2398 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](72zWmh93Bo:0) -> float_type, {32, 1}, {17, 1}: 0.00071084ms, 0.00624528% main:@2399 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](AB6lekmMxa:0) -> float_type, {32, 1}, {25, 1}: 0.0006706ms, 0.00589174% main:@2400 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](ifP6Ceko5T:0) -> float_type, {32, 1}, {73, 1}: 0.00077744ms, 0.00683041% main:@2401 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](89nc05gQza:0) -> float_type, {32, 1}, {9, 1}: 0.00065662ms, 0.00576892% main:@2402 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](8Dv01FAoNM:0) -> float_type, {32, 1}, {17, 1}: 0.000672ms, 0.00590404% main:@2403 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](UMO0o14RuI:0) -> float_type, {32, 1}, {9, 1}: 0.00068356ms, 0.0060056% main:@2404 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](8iOST90kWM:0) -> float_type, {32, 1}, {9, 1}: 0.00079318ms, 0.0069687% main:@2405 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](8vM5p4RzXd:0) -> float_type, {32, 1}, {13, 1}: 0.0007027ms, 0.00617376% main:@2406 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](9dgeU4bKka:0) -> float_type, {32, 1}, {17, 1}: 0.0006881ms, 0.00604549% main:@2407 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](9mfMQZIXVH:0) -> float_type, {32, 1}, {9, 1}: 0.00071678ms, 0.00629747% main:@2408 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](EhUjmCTaGW:0) -> float_type, {32, 1}, {9, 1}: 0.0008081ms, 0.00709979% main:@2409 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](AvlwbHMnuE:0) -> float_type, {32, 1}, {17, 1}: 0.00069438ms, 0.00610067% main:@2410 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](AVLcBpWeuo:0) -> float_type, {32, 1}, {21, 1}: 0.00068596ms, 0.00602669% main:@2411 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](Cw59eOo7M1:0) -> float_type, {32, 1}, {9, 1}: 0.00067908ms, 0.00596624% main:@2412 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](nyLxrsTmOB:0) -> float_type, {32, 1}, {17, 1}: 0.00080092ms, 0.0070367% main:@2413 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](m2YscaoDjP:0) -> float_type, {32, 1}, {17, 1}: 0.00070886ms, 0.00622788% main:@2414 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](BYyjkPCX0m:0) -> float_type, {32, 1}, {13, 1}: 0.00070016ms, 0.00615145% main:@2415 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](oQ5OcjZJe8:0) -> float_type, {32, 1}, {17, 1}: 0.00066302ms, 0.00582515% main:@2416 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](CHAYRkBvIo:0) -> float_type, {32, 1}, {41, 1}: 0.00080306ms, 0.00705551% main:@2417 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](isCn4mEOSt:0) -> float_type, {32, 1}, {49, 1}: 0.00070308ms, 0.0061771% main:@2418 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](r1WGP7Xe9H:0) -> float_type, {32, 1}, {13, 1}: 0.00072092ms, 0.00633384% main:@2419 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](qfwYOHde3u:0) -> float_type, {32, 1}, {17, 1}: 0.00068414ms, 0.0060107% main:@2420 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](tIGjbcEuDd:0) -> float_type, {32, 1}, {9, 1}: 0.00084376ms, 0.00741309% main:@2421 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](vZDAiUHErW:0) -> float_type, {32, 1}, {17, 1}: 0.00064738ms, 0.00568774% main:@2422 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](Dy3rwVmn1A:0) -> float_type, {32, 1}, {41, 1}: 0.00068694ms, 0.0060353% main:@2423 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](E1ZkVIXqS5:0) -> float_type, {32, 1}, {9, 1}: 0.00069952ms, 0.00614583% main:@2424 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](qzd23aNGIT:0) -> float_type, {32, 1}, {17, 1}: 0.0008273ms, 0.00726847% main:@2425 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](EUciqFuptg:0) -> float_type, {32, 1}, {17, 1}: 0.0006941ms, 0.00609821% main:@2426 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](EKpP4xMXu8:0) -> float_type, {32, 1}, {17, 1}: 0.00068398ms, 0.00600929% main:@2427 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](EZBIAqn6LS:0) -> float_type, {32, 1}, {13, 1}: 0.00071988ms, 0.0063247% main:@2428 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](EsROupwkSG:0) -> float_type, {32, 1}, {13, 1}: 0.00079786ms, 0.00700982% main:@2429 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](nP9Yv1iZ8f:0) -> float_type, {32, 1}, {9, 1}: 0.00068174ms, 0.00598961% main:@2430 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](pQsi1GNxvj:0) -> float_type, {32, 1}, {5, 1}: 0.00069464ms, 0.00610295% main:@2431 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](F6DhiS9CN5:0) -> float_type, {32, 1}, {9, 1}: 0.00067304ms, 0.00591318% main:@2432 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](FWhBxj4Tun:0) -> float_type, {32, 1}, {9, 1}: 0.0008003ms, 0.00703126% main:@2433 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](eQquakfWBY:0) -> float_type, {32, 1}, {17, 1}: 0.0007166ms, 0.00629589% main:@2434 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](FcmWLZhDgE:0) -> float_type, {32, 1}, {57, 1}: 0.00068798ms, 0.00604444% main:@2435 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](JCVMzpWyri:0) -> float_type, {32, 1}, {17, 1}: 0.00072298ms, 0.00635194% main:@2436 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](uFSOc5Chp2:0) -> float_type, {32, 1}, {481, 1}: 0.00084892ms, 0.00745842% main:@2437 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](HRGzlnEkXQ:0) -> float_type, {32, 1}, {41, 1}: 0.00068786ms, 0.00604338% main:@2438 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](HZ0l3MmcPY:0) -> float_type, {32, 1}, {17, 1}: 0.00070418ms, 0.00618677% main:@2439 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](YbIVz8l1un:0) -> float_type, {32, 1}, {45, 1}: 0.00066808ms, 0.0058696% main:@2440 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](OfNJgEDtw2:0) -> float_type, {32, 1}, {13, 1}: 0.00082066ms, 0.00721013% main:@2441 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](HmATr3gKsW:0) -> float_type, {32, 1}, {9, 1}: 0.00072462ms, 0.00636635% main:@2442 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](Hnjq487SIY:0) -> float_type, {32, 1}, {13, 1}: 0.00070718ms, 0.00621312% main:@2443 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](omr2g4XaH5:0) -> float_type, {32, 1}, {17, 1}: 0.00065802ms, 0.00578122% main:@2444 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](NJBt1ubXFW:0) -> float_type, {32, 1}, {5, 1}: 0.00082464ms, 0.0072451% main:@2445 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](mgnGNH4Mz3:0) -> float_type, {32, 1}, {49, 1}: 0.00068474ms, 0.00601597% main:@2446 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](IMdFk7D16X:0) -> float_type, {32, 1}, {9, 1}: 0.00067826ms, 0.00595904% main:@2447 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](jznORDJr0N:0) -> float_type, {32, 1}, {45, 1}: 0.00068524ms, 0.00602036% main:@2448 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](mNgQ4ohWqU:0) -> float_type, {32, 1}, {9, 1}: 0.00080668ms, 0.00708731% main:@2449 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](IxpbBg7JTO:0) -> float_type, {32, 1}, {5, 1}: 0.00065592ms, 0.00576277% main:@2450 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](J3H2EjAXt8:0) -> float_type, {32, 1}, {5, 1}: 0.00067514ms, 0.00593163% main:@2451 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](sZdFJI9nwg:0) -> float_type, {32, 1}, {41, 1}: 0.00067482ms, 0.00592882% main:@2452 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](JSWhLK4Agx:0) -> float_type, {32, 1}, {25, 1}: 0.00082498ms, 0.00724809% main:@2453 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](TkHQvxKeD6:0) -> float_type, {32, 1}, {9, 1}: 0.00068296ms, 0.00600033% main:@2454 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](JSPsGophtx:0) -> float_type, {32, 1}, {25, 1}: 0.00068064ms, 0.00597995% main:@2455 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](JaHTUxB6zL:0) -> float_type, {32, 1}, {17, 1}: 0.00068816ms, 0.00604602% main:@2456 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](U6GqhP8k7I:0) -> float_type, {32, 1}, {9, 1}: 0.00083656ms, 0.00734983% main:@2457 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](KCNfS4GlpP:0) -> float_type, {32, 1}, {9, 1}: 0.0006882ms, 0.00604637% main:@2458 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](eQI8ikEMgP:0) -> float_type, {32, 1}, {41, 1}: 0.00072516ms, 0.00637109% main:@2459 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](LKAvFme9ZS:0) -> float_type, {32, 1}, {49, 1}: 0.00070198ms, 0.00616744% main:@2460 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](pY0t1rI5WU:0) -> float_type, {32, 1}, {41, 1}: 0.00082624ms, 0.00725916% main:@2461 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](LMobYAd5ru:0) -> float_type, {32, 1}, {17, 1}: 0.00070352ms, 0.00618097% main:@2462 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](iCbkMrmXBh:0) -> float_type, {32, 1}, {9, 1}: 0.00066976ms, 0.00588436% main:@2463 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](NXOSdQVPs0:0) -> float_type, {32, 1}, {25, 1}: 0.00067016ms, 0.00588788% main:@2464 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](ecPTQUALkq:0) -> float_type, {32, 1}, {5, 1}: 0.00079976ms, 0.00702651% main:@2465 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](O5aJeUQh2G:0) -> float_type, {32, 1}, {17, 1}: 0.0006857ms, 0.00602441% main:@2466 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](gdck97AK4w:0) -> float_type, {32, 1}, {17, 1}: 0.00069314ms, 0.00608977% main:@2467 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](cHzIDJm6us:0) -> float_type, {32, 1}, {33, 1}: 0.0007184ms, 0.0063117% main:@2468 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](Ouh1XpPZaj:0) -> float_type, {32, 1}, {9, 1}: 0.00081118ms, 0.00712685% main:@2469 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](lyiDxfzZg3:0) -> float_type, {32, 1}, {701, 1}: 0.00065714ms, 0.00577348% main:@2470 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](Pu1O5qFxY6:0) -> float_type, {32, 1}, {17, 1}: 0.0006811ms, 0.00598399% main:@2471 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](ry2xad1vc9:0) -> float_type, {32, 1}, {25, 1}: 0.0006946ms, 0.0061026% main:@2472 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](zX6xKOWG9S:0) -> float_type, {32, 1}, {5, 1}: 0.00082948ms, 0.00728763% main:@2473 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](qd1komWTn4:0) -> float_type, {32, 1}, {25, 1}: 0.0007133ms, 0.00626689% main:@2474 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](Qp3K1PHvte:0) -> float_type, {32, 1}, {69, 1}: 0.00065938ms, 0.00579316% main:@2475 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](dz1saATctE:0) -> float_type, {32, 1}, {9, 1}: 0.0006799ms, 0.00597345% main:@2476 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](RpITWntAji:0) -> float_type, {32, 1}, {17, 1}: 0.0008472ms, 0.00744331% main:@2477 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](nltbg9ZWQX:0) -> float_type, {32, 1}, {17, 1}: 0.00066864ms, 0.00587452% main:@2478 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](uhcrWfUe6M:0) -> float_type, {32, 1}, {9, 1}: 0.00068652ms, 0.00603161% main:@2479 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](NLt5JnRMjG:0) -> float_type, {32, 1}, {5, 1}: 0.00064078ms, 0.00562975% main:@2480 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](TSqGwrHeo4:0) -> float_type, {32, 1}, {5, 1}: 0.0008383ms, 0.00736512% main:@2481 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](Tej5M3O6As:0) -> float_type, {32, 1}, {17, 1}: 0.00068658ms, 0.00603214% main:@2482 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](khJcIxW3zv:0) -> float_type, {32, 1}, {9, 1}: 0.00068736ms, 0.00603899% main:@2483 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](Tg4u0GfnFb:0) -> float_type, {32, 1}, {25, 1}: 0.00070466ms, 0.00619098% main:@2484 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](TxEdCcekGS:0) -> float_type, {32, 1}, {49, 1}: 0.00078966ms, 0.00693778% main:@2485 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](XPcHGKbop2:0) -> float_type, {32, 1}, {13, 1}: 0.00073562ms, 0.00646299% main:@2486 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](Uc42tiuavR:0) -> float_type, {32, 1}, {9, 1}: 0.00071612ms, 0.00629167% main:@2487 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](pIb4NdDSn5:0) -> float_type, {32, 1}, {9, 1}: 0.000682ms, 0.0059919% main:@2488 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](IRem4MiFDP:0) -> float_type, {32, 1}, {5, 1}: 0.00079726ms, 0.00700455% main:@2489 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](bTeX85HPmJ:0) -> float_type, {32, 1}, {17, 1}: 0.00074084ms, 0.00650885% main:@2490 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](duyRoIEkH8:0) -> float_type, {32, 1}, {45, 1}: 0.00067498ms, 0.00593022% main:@2491 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](d8RoEzG52p:0) -> float_type, {32, 1}, {17, 1}: 0.00067624ms, 0.00594129% main:@2492 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](X29iyIeu8L:0) -> float_type, {32, 1}, {45, 1}: 0.00080342ms, 0.00705867% main:@2493 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](zIoqYe9cRb:0) -> float_type, {32, 1}, {45, 1}: 0.00070184ms, 0.00616621% main:@2494 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](ZRYNFfjSez:0) -> float_type, {32, 1}, {33, 1}: 0.00065162ms, 0.00572499% main:@2495 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](ZcWERoz0h6:0) -> float_type, {32, 1}, {49, 1}: 0.00070642ms, 0.00620645% main:@2496 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](nv2tzWRf1I:0) -> float_type, {32, 1}, {301, 1}: 0.00079224ms, 0.00696044% main:@2497 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](ahAlEtBvzo:0) -> float_type, {32, 1}, {41, 1}: 0.00074752ms, 0.00656754% main:@2498 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](WDLoyKsiA7:0) -> float_type, {32, 1}, {33, 1}: 0.00067728ms, 0.00595043% main:@2499 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](aqMQBgb3hf:0) -> float_type, {32, 1}, {17, 1}: 0.00068292ms, 0.00599998% main:@2500 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](yVxKAL4U5G:0) -> float_type, {32, 1}, {9, 1}: 0.0008054ms, 0.00707606% main:@2501 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](bJsIAzhYc3:0) -> float_type, {32, 1}, {37, 1}: 0.00069536ms, 0.00610928% main:@2502 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](cN2JSTnFOu:0) -> float_type, {32, 1}, {9, 1}: 0.00074128ms, 0.00651272% main:@2503 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](vAbKepTjQg:0) -> float_type, {32, 1}, {5, 1}: 0.00065994ms, 0.00579808% main:@2504 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](fZhaL2tCm6:0) -> float_type, {32, 1}, {17, 1}: 0.00080332ms, 0.00705779% main:@2505 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](o74vNAnXxy:0) -> float_type, {32, 1}, {17, 1}: 0.00070766ms, 0.00621734% main:@2506 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](l5WSX3VvMK:0) -> float_type, {32, 1}, {25, 1}: 0.00067298ms, 0.00591265% main:@2507 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](pH5srbufJ1:0) -> float_type, {32, 1}, {17, 1}: 0.00072932ms, 0.00640764% main:@2508 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](lDhFNYzbxR:0) -> float_type, {32, 1}, {17, 1}: 0.00085834ms, 0.00754118% main:@2509 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](magwfJSnUu:0) -> float_type, {32, 1}, {9, 1}: 0.00067296ms, 0.00591248% main:@2510 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](w04Keh2FJu:0) -> float_type, {32, 1}, {13, 1}: 0.00071242ms, 0.00625916% main:@2511 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](fMiQsYNnet:0) -> float_type, {32, 1}, {5, 1}: 0.00070646ms, 0.0062068% main:@2512 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](gqk5byC8aJ:0) -> float_type, {32, 1}, {5, 1}: 0.0008401ms, 0.00738093% main:@2513 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](lQmygxs5aD:0) -> float_type, {32, 1}, {13, 1}: 0.00072726ms, 0.00638954% main:@2514 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](sjKQJGby31:0) -> float_type, {32, 1}, {9, 1}: 0.0006825ms, 0.00599629% main:@2515 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](zZHaVOckpw:0) -> float_type, {32, 1}, {13, 1}: 0.00071042ms, 0.00624159% main:@2516 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](l4bReWtsMB:0) -> float_type, {32, 1}, {25, 1}: 0.00084312ms, 0.00740746% main:@2517 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](tWcngAq7hV:0) -> float_type, {32, 1}, {13, 1}: 0.00074312ms, 0.00652889% main:@2518 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](xVErSbJ0hg:0) -> float_type, {32, 1}, {9, 1}: 0.00066924ms, 0.00587979% main:@2519 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](mTMtDEo3r8:0) -> float_type, {32, 1}, {9, 1}: 0.00066712ms, 0.00586117% main:@2520 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](VUqfdN5Fzv:0) -> float_type, {32, 1}, {45, 1}: 0.00075354ms, 0.00662043% main:@2521 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](Vbv2Om3ptd:0) -> float_type, {32, 1}, {17, 1}: 0.00067918ms, 0.00596712% ebEFYOy65u:0 = @param:ebEFYOy65u:0 -> float_type, {32, 1}, {1, 1}: 0.00072044ms, 0.00632962% iUEBSr6gOG:0 = @param:iUEBSr6gOG:0 -> float_type, {32, 1}, {1, 1}: 0.00051852ms, 0.0045556% qxzgXpiLdl:0 = @param:qxzgXpiLdl:0 -> float_type, {32, 1}, {1, 1}: 0.00064824ms, 0.00569529% 9NcxXsmyqf:0 = @param:9NcxXsmyqf:0 -> float_type, {32, 1}, {1, 1}: 0.00055644ms, 0.00488876% WqeN5KGIv9:0 = @param:WqeN5KGIv9:0 -> float_type, {32, 1}, {1, 1}: 0.00051874ms, 0.00455753% 7XiFOysrpI:0 = @param:7XiFOysrpI:0 -> float_type, {32, 1}, {1, 1}: 0.00055938ms, 0.00491459% vtIF8HqRG6:0 = @param:vtIF8HqRG6:0 -> float_type, {32, 1}, {1, 1}: 0.00049638ms, 0.00436108% NG35MnhTIF:0 = @param:NG35MnhTIF:0 -> float_type, {32, 1}, {1, 1}: 0.0005612ms, 0.00493058% lqOmNvz5fj:0 = @param:lqOmNvz5fj:0 -> float_type, {32, 1}, {1, 1}: 0.00054574ms, 0.00479475% XPpbueLMOZ:0 = @param:XPpbueLMOZ:0 -> float_type, {32, 1}, {1, 1}: 0.00065052ms, 0.00571532% bJxY1IhSVn:0 = @param:bJxY1IhSVn:0 -> float_type, {32, 1}, {1, 1}: 0.00053586ms, 0.00470795% iE352qtAOx:0 = @param:iE352qtAOx:0 -> float_type, {32, 1}, {1, 1}: 0.00050322ms, 0.00442118% c2MuYjWE3S:0 = @param:c2MuYjWE3S:0 -> float_type, {32, 1}, {1, 1}: 0.0005172ms, 0.004544% 7tFB8mLWUr:0 = @param:7tFB8mLWUr:0 -> float_type, {32, 1}, {1, 1}: 0.0005534ms, 0.00486205% FTIMrcfako:0 = @param:FTIMrcfako:0 -> float_type, {32, 1}, {1, 1}: 0.0007221ms, 0.00634421% xGCyW9icVP:0 = @param:xGCyW9icVP:0 -> float_type, {32, 1}, {1, 1}: 0.00047034ms, 0.0041323% lhbKMrO1my:0 = @param:lhbKMrO1my:0 -> float_type, {32, 1}, {1, 1}: 0.00050532ms, 0.00443963% StHFqj5edV:0 = @param:StHFqj5edV:0 -> float_type, {32, 1}, {1, 1}: 0.00048234ms, 0.00423773% ZyQ5FrpI6t:0 = @param:ZyQ5FrpI6t:0 -> float_type, {32, 1}, {1, 1}: 0.00047664ms, 0.00418765% mxUaAp0CV8:0 = @param:mxUaAp0CV8:0 -> float_type, {32, 1}, {1, 1}: 0.00057652ms, 0.00506518% vF46yONhri:0 = @param:vF46yONhri:0 -> float_type, {32, 1}, {1, 1}: 0.0006073ms, 0.0053356% URVDgNGoIx:0 = @param:URVDgNGoIx:0 -> float_type, {32, 1}, {1, 1}: 0.00050346ms, 0.00442329% RPGvVrZMak:0 = @param:RPGvVrZMak:0 -> float_type, {32, 1}, {1, 1}: 0.00047768ms, 0.00419679% main:@2545 = load[offset=740224,end=761600](main:@1) -> float_type, {32, 167}, {167, 1}: 0.00050036ms, 0.00439605% main:@2546 = gpu::code_object[code_object=25808,symbol_name=concat_kernel,global=32,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, {32, 167}, {167, 1}: 0.134156ms, 1.17866% main:@2547 = load[offset=794368,end=815744](main:@1) -> float_type, {32, 167}, {167, 1}: 0.00073024ms, 0.00641572% main:@2548 = gpu::code_object[code_object=13520,symbol_name=rsqrt_add_kernel,global=2672,local=256,](main:@2377,main:@2546,main:@2547) -> float_type, {32, 167}, {167, 1}: 0.0137141ms, 0.120489% main:@2549 = hip::hip_copy_literal[id=main:@literal:45] -> float_type, {167, 256}, {256, 1}: 0.00081224ms, 0.00713616% main:@2550 = load[offset=761600,end=794368](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00048174ms, 0.00423246% 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, {32, 256}, {256, 1}: 0.0238203ms, 0.20928% main:@2552 = hip::hip_copy_literal[id=main:@literal:38] -> float_type, {128, 256}, {256, 1}: 0.00079878ms, 0.0070179% main:@2553 = slice[axes={1},starts={640},ends={768}](main:@1109) -> float_type, {32, 128}, {896, 1}: 0.00069896ms, 0.00614091% main:@2554 = load[offset=794368,end=827136](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00048236ms, 0.00423791% 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, {32, 256}, {256, 1}: 0.0205098ms, 0.180195% main:@2556 = hip::hip_copy_literal[id=main:@literal:9] -> float_type, {32, 256}, {256, 1}: 0.00078916ms, 0.00693338% main:@2557 = load[offset=8192,end=40960](main:@1) -> float_type, {32, 256}, {256, 1}: 0.00046022ms, 0.00404339% main:@2558 = gpu::code_object[code_object=13552,symbol_name=add_leaky_relu_tanh_pow_kernel,global=4096,local=256,](main:@2551,main:@2556,main:@2555,main:@2557) -> float_type, {32, 256}, {256, 1}: 0.0177837ms, 0.156243% main:@2559 = hip::hip_copy_literal[id=main:@literal:76] -> float_type, {256, 64}, {64, 1}: 0.00076542ms, 0.00672481% main:@2560 = load[offset=0,end=8192](main:@1) -> float_type, {32, 64}, {64, 1}: 0.0004876ms, 0.00428394% main:@2561 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@2558,main:@2559,main:@2560) -> float_type, {32, 64}, {64, 1}: 0.0258853ms, 0.227422% main:@2562 = hip::hip_copy_literal[id=main:@literal:111] -> float_type, {53, 53}, {53, 1}: 0.00081966ms, 0.00720135% main:@2563 = slice[axes={1},starts={53},ends={106}](main:@2373) -> float_type, {32, 53}, {212, 1}: 0.0006394ms, 0.00561763% main:@2564 = load[offset=21760,end=28544](main:@1) -> float_type, {32, 53}, {53, 1}: 0.00046342ms, 0.0040715% main:@2565 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@2563,main:@2562,main:@2564) -> float_type, {32, 53}, {53, 1}: 0.0206835ms, 0.18172% main:@2566 = hip::hip_copy_literal[id=main:@literal:82] -> float_type, {1, 53}, {53, 1}: 0.00082012ms, 0.00720539% main:@2567 = slice[axes={1},starts={159},ends={212}](main:@2373) -> float_type, {32, 53}, {212, 1}: 0.00057678ms, 0.00506746% main:@2568 = load[offset=8192,end=14976](main:@1) -> float_type, {32, 53}, {53, 1}: 0.0004865ms, 0.00427428% main:@2569 = gpu::code_object[code_object=13536,symbol_name=div_pow_log_add_kernel,global=1696,local=256,](main:@2567,main:@2568) -> float_type, {32, 53}, {53, 1}: 0.015497ms, 0.136153% main:@2570 = multibroadcast[out_lens={32, 53}](main:@2566) -> float_type, {32, 53}, {0, 1}: 0.00079414ms, 0.00697714% main:@2571 = load[offset=14976,end=21760](main:@1) -> float_type, {32, 53}, {53, 1}: 0.00053198ms, 0.00467386% main:@2572 = gpu::code_object[code_object=13552,symbol_name=pow_sub_pow_pow_log_add_kernel,global=1696,local=256,](main:@2565,main:@2570,main:@2563,main:@2571) -> float_type, {32, 53}, {53, 1}: 0.0180569ms, 0.158644% main:@2573 = slice[axes={1},starts={106},ends={159}](main:@2373) -> float_type, {32, 53}, {212, 1}: 0.00072192ms, 0.00634263% main:@2574 = slice[axes={1},starts={0},ends={53}](main:@2373) -> float_type, {32, 53}, {212, 1}: 0.00055718ms, 0.00489526% main:@2575 = load[offset=21760,end=28544](main:@1) -> float_type, {32, 53}, {53, 1}: 0.00045646ms, 0.00401036% main:@2576 = gpu::code_object[code_object=13536,symbol_name=add_add_neg_div_kernel,global=1696,local=256,](main:@2573,main:@2574,main:@2575) -> float_type, {32, 53}, {53, 1}: 0.0141871ms, 0.124645% main:@2577 = hip::hip_copy_literal[id=main:@literal:83] -> float_type, {374, 212}, {212, 1}: 0.00070904ms, 0.00622947% main:@2578 = load[offset=55808,end=55936](main:@1) -> float_type, {32, 1}, {1, 1}: 0.00048712ms, 0.00427973% main:@2579 = gpu::code_object[code_object=13544,symbol_name=reduce_mean_add_neg_kernel,global=2048,local=64,](main:@2569,main:@2578) -> float_type, {32, 1}, {1, 1}: 0.0142596ms, 0.125282% main:@2580 = load[offset=28544,end=55680](main:@1) -> float_type, {32, 212}, {212, 1}: 0.00062002ms, 0.00544736% main:@2581 = gpu::code_object[code_object=13528,symbol_name=leaky_relu_kernel,global=3392,local=256,](main:@2373,main:@2580) -> float_type, {32, 212}, {212, 1}: 0.013278ms, 0.116658% main:@2582 = load[offset=56192,end=56320](main:@1) -> float_type, {32, 1}, {1, 1}: 0.00059724ms, 0.00524722% main:@2583 = gpu::code_object[code_object=13544,symbol_name=reduce_prod_add_log_kernel,global=2048,local=64,](main:@2576,main:@2582) -> float_type, {32, 1}, {1, 1}: 0.014086ms, 0.123756% main:@2584 = load[offset=55680,end=55808](main:@1) -> float_type, {32, 1}, {1, 1}: 0.00060514ms, 0.00531662% main:@2585 = gpu::code_object[code_object=13552,symbol_name=reduce_prod_pow_rsqrt_div_kernel,global=2048,local=64,](main:@2572,main:@2584) -> float_type, {32, 1}, {1, 1}: 0.0149171ms, 0.131058% main:@2586 = load[offset=55936,end=56192](main:@1) -> float_type, {32, 2}, {2, 1}: 0.00059758ms, 0.0052502% main:@2587 = gpu::code_object[code_object=13520,symbol_name=concat_kernel,global=32,local=256,](main:@2583,main:@2579,main:@2586) -> float_type, {32, 2}, {2, 1}: 0.0130952ms, 0.115052% main:@2588 = slice[axes={1},starts={106},ends={159}](main:@2581) -> float_type, {32, 53}, {212, 1}: 0.00071028ms, 0.00624036% main:@2589 = slice[axes={1},starts={159},ends={212}](main:@2581) -> float_type, {32, 53}, {212, 1}: 0.00054724ms, 0.00480793% main:@2590 = slice[axes={1},starts={53},ends={106}](main:@2581) -> float_type, {32, 53}, {212, 1}: 0.0006657ms, 0.00584869% main:@2591 = slice[axes={1},starts={0},ends={53}](main:@2581) -> float_type, {32, 53}, {212, 1}: 0.00053622ms, 0.00471111% main:@2592 = slice[axes={1},starts={4070},ends={4444}](main:@2244) -> float_type, {32, 374}, {4508, 1}: 0.00055556ms, 0.00488103% main:@2593 = load[offset=56192,end=104064](main:@1) -> float_type, {32, 374}, {374, 1}: 0.00045624ms, 0.00400842% main:@2594 = gpu::code_object[code_object=21816,symbol_name=noop_noop_noop_noop_noop_noop_noop_sub_noop_concat_neg_pow_kernel,global=1329,local=256,](main:@2588,main:@2591,main:@2589,main:@2590,main:@2576,main:@2569,main:@2572,main:@2278,main:@2587,main:@2585,main:@2592,main:@2593) -> float_type, {32, 374}, {374, 1}: 0.0372681ms, 0.327429% main:@2595 = load[offset=8192,end=35328](main:@1) -> float_type, {32, 212}, {212, 1}: 0.00067092ms, 0.00589455% main:@2596 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@2594,main:@2577,main:@2595) -> float_type, {32, 212}, {212, 1}: 0.0282353ms, 0.248069% main:@2597 = hip::hip_copy_literal[id=main:@literal:15] -> float_type, {277, 160}, {160, 1}: 0.00081642ms, 0.00717288% main:@2598 = load[offset=35328,end=35456](main:@1) -> float_type, {32, 1}, {1, 1}: 0.0004928ms, 0.00432963% main:@2599 = gpu::code_object[code_object=13528,symbol_name=reduce_prod_kernel,global=6144,local=192,](main:@2546,main:@2598) -> float_type, {32, 1}, {1, 1}: 0.0143095ms, 0.12572% main:@2600 = hip::hip_copy_literal[id=main:@literal:27] -> float_type, {32, 64}, {64, 1}: 0.00077008ms, 0.00676575% main:@2601 = hip::hip_copy_literal[id=main:@literal:34] -> float_type, {32, 212}, {212, 1}: 0.00065408ms, 0.0057466% main:@2602 = slice[axes={1},starts={4444},ends={4508}](main:@2244) -> float_type, {32, 64}, {4508, 1}: 0.00074444ms, 0.00654048% main:@2603 = load[offset=41088,end=76544](main:@1) -> float_type, {32, 277}, {277, 1}: 0.00045746ms, 0.00401914% main:@2604 = gpu::code_object[code_object=13600,symbol_name=rsqrt_add_sub_mul_add_leaky_relu_noop_concat_noop_kernel,global=2954,local=256,](main:@2602,main:@2561,main:@2600,main:@2596,main:@2601,main:@2599,main:@2603) -> float_type, {32, 277}, {277, 1}: 0.018109ms, 0.159101% main:@2605 = load[offset=20608,end=41088](main:@1) -> float_type, {32, 160}, {160, 1}: 0.00063ms, 0.00553504% main:@2606 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@2604,main:@2597,main:@2605) -> float_type, {32, 160}, {160, 1}: 0.0256192ms, 0.225084% main:@2607 = hip::hip_copy_literal[id=main:@literal:10] -> float_type, {160}, {1}: 0.00076562ms, 0.00672657% main:@2608 = multibroadcast[out_lens={32, 160}](main:@2607) -> float_type, {32, 160}, {0, 1}: 0.00056856ms, 0.00499524% main:@2609 = load[offset=128,end=20608](main:@1) -> float_type, {32, 160}, {160, 1}: 0.00045896ms, 0.00403232% main:@2610 = gpu::code_object[code_object=13536,symbol_name=add_leaky_relu_kernel,global=2560,local=256,](main:@2606,main:@2608,main:@2609) -> float_type, {32, 160}, {160, 1}: 0.0138447ms, 0.121636% main:@2611 = hip::hip_copy_literal[id=main:@literal:36] -> float_type, {64, 329}, {329, 1}: 0.000778ms, 0.00683533% main:@2612 = slice[axes={1},starts={80},ends={144}](main:@2610) -> float_type, {32, 64}, {160, 1}: 0.00062506ms, 0.00549164% main:@2613 = load[offset=20608,end=62720](main:@1) -> float_type, {32, 329}, {329, 1}: 0.00048424ms, 0.00425442% 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, {32, 329}, {329, 1}: 0.0199989ms, 0.175706% main:@2615 = hip::hip_copy_literal[id=main:@literal:88] -> float_type, {1, 329}, {329, 1}: 0.00081344ms, 0.0071467% main:@2616 = hip::hip_copy_literal[id=main:@literal:65] -> float_type, {16, 1}, {1, 1}: 0.00061354ms, 0.00539042% main:@2617 = slice[axes={1},starts={144},ends={160}](main:@2610) -> float_type, {32, 16}, {160, 1}: 0.00056066ms, 0.00492583% main:@2618 = load[offset=0,end=128](main:@1) -> float_type, {32, 1}, {1, 1}: 0.00049972ms, 0.00439043% main:@2619 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@2617,main:@2616,main:@2618) -> float_type, {32, 1}, {1, 1}: 0.0218675ms, 0.192123% main:@2620 = load[offset=62720,end=104832](main:@1) -> float_type, {32, 329}, {329, 1}: 0.00061156ms, 0.00537303% main:@2621 = gpu::code_object[code_object=13520,symbol_name=softmax_kernel,global=8192,local=256,](main:@2614,main:@2620) -> float_type, {32, 329}, {329, 1}: 0.0152655ms, 0.134119% main:@2622 = multibroadcast[out_lens={32, 329}](main:@2615) -> float_type, {32, 329}, {0, 1}: 0.00073572ms, 0.00646387% main:@2623 = load[offset=20608,end=20736](main:@1) -> float_type, {32, 1}, {1, 1}: 0.00046768ms, 0.00410893% main:@2624 = gpu::code_object[code_object=13536,symbol_name=add_reduce_prod_kernel,global=8192,local=256,](main:@2622,main:@2621,main:@2623) -> float_type, {32, 1}, {1, 1}: 0.0151305ms, 0.132933% main:@2625 = squeeze[axes={1}](main:@2624) -> float_type, {32}, {1}: 0.00073632ms, 0.00646914% main:#output_1 = @param:main:#output_1 -> float_type, {32}, {1}: 0.00077654ms, 0.00682251% main:@2627 = gpu::code_object[code_object=13512,symbol_name=mul_kernel,global=16,local=256,](main:@2625,main:#output_1) -> float_type, {32}, {1}: 0.0129465ms, 0.113745% main:@2628 = hip::hip_copy_literal[id=main:@literal:110] -> float_type, {64, 500}, {500, 1}: 0.00083936ms, 0.00737443% main:@2629 = slice[axes={1},starts={16},ends={80}](main:@2610) -> float_type, {32, 64}, {160, 1}: 0.0006044ms, 0.00531012% main:@2630 = load[offset=20992,end=84992](main:@1) -> float_type, {32, 500}, {500, 1}: 0.00046212ms, 0.00406008% main:@2631 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@2629,main:@2628,main:@2630) -> float_type, {32, 500}, {500, 1}: 0.0199117ms, 0.17494% main:@2632 = hip::hip_copy_literal[id=main:@literal:147] -> float_type, {16, 3}, {3, 1}: 0.00074704ms, 0.00656333% main:@2633 = slice[axes={1},starts={0},ends={16}](main:@2610) -> float_type, {32, 16}, {160, 1}: 0.00060034ms, 0.00527445% main:@2634 = load[offset=20608,end=20992](main:@1) -> float_type, {32, 3}, {3, 1}: 0.00045342ms, 0.00398365% main:@2635 = gpu::gemm[alpha=1,beta=0,int8_x4_format=0,compute_fp32=1,trans_batch=0](main:@2633,main:@2632,main:@2634) -> float_type, {32, 3}, {3, 1}: 0.0199336ms, 0.175132% main:@2636 = hip::hip_copy_literal[id=main:@literal:72] -> float_type, {1, 500}, {500, 1}: 0.00081874ms, 0.00719327% main:#output_2 = @param:main:#output_2 -> float_type, {32}, {1}: 0.00071976ms, 0.00632365% main:@2638 = gpu::code_object[code_object=13512,symbol_name=add_kernel,global=16,local=256,](main:@2627,main:#output_2) -> float_type, {32}, {1}: 0.0133871ms, 0.117616% main:#output_3 = @param:main:#output_3 -> float_type, {32}, {1}: 0.00076584ms, 0.0067285% main:@2640 = gpu::code_object[code_object=13520,symbol_name=mul_neg_kernel,global=16,local=256,](main:@2627,main:#output_3) -> float_type, {32}, {1}: 0.0132762ms, 0.116642% main:@2641 = load[offset=84992,end=148992](main:@1) -> float_type, {32, 500}, {500, 1}: 0.00063498ms, 0.00557879% main:@2642 = gpu::code_object[code_object=13520,symbol_name=softmax_kernel,global=8192,local=256,](main:@2631,main:@2641) -> float_type, {32, 500}, {500, 1}: 0.0152629ms, 0.134097% main:@2643 = multibroadcast[out_lens={32, 500}](main:@2636) -> float_type, {32, 500}, {0, 1}: 0.00071578ms, 0.00628868% main:@2644 = load[offset=128,end=256](main:@1) -> float_type, {32, 1}, {1, 1}: 0.00045644ms, 0.00401018% main:@2645 = gpu::code_object[code_object=13536,symbol_name=add_reduce_mean_kernel,global=8192,local=256,](main:@2643,main:@2642,main:@2644) -> float_type, {32, 1}, {1, 1}: 0.0147629ms, 0.129704% main:@2646 = squeeze[axes={1}](main:@2645) -> float_type, {32}, {1}: 0.00068218ms, 0.00599348% main:#output_4 = @param:main:#output_4 -> float_type, {32}, {1}: 0.00063384ms, 0.00556878% main:@2648 = gpu::code_object[code_object=13512,symbol_name=mul_kernel,global=16,local=256,](main:@2646,main:#output_4) -> float_type, {32}, {1}: 0.0130463ms, 0.114622% main:@2649 = load[offset=128,end=512](main:@1) -> float_type, {32, 3}, {3, 1}: 0.00059712ms, 0.00524616% main:@2650 = gpu::code_object[code_object=14512,symbol_name=softmax_kernel,global=2048,local=64,](main:@2635,main:@2649) -> float_type, {32, 3}, {3, 1}: 0.0141751ms, 0.124539% main:@2651 = squeeze[axes={1}](main:@2619) -> float_type, {32}, {1}: 0.00068976ms, 0.00606008% 86KRkhYbW4:0 = @param:86KRkhYbW4:0 -> float_type, {32}, {1}: 0.00062326ms, 0.00547582% main:#output_0 = @param:main:#output_0 -> float_type, {32}, {1}: 0.00060736ms, 0.00533613% main:@2654 = gpu::code_object[code_object=13536,symbol_name=sigmoid_mul_tanh_kernel,global=16,local=256,](86KRkhYbW4:0,main:@2651,main:#output_0) -> float_type, {32}, {1}: 0.0147252ms, 0.129373% main:@2655 = slice[axes={0, 1},starts={0, 0},ends={32, 1}](main:@2650) -> float_type, {32, 1}, {3, 1}: 0.00078952ms, 0.00693655% main:#output_5 = @param:main:#output_5 -> float_type, {32, 1}, {1, 1}: 0.0006385ms, 0.00560972% main:@2657 = gpu::code_object[code_object=13528,symbol_name=contiguous_kernel,global=32,local=256,](main:@2655,main:#output_5) -> float_type, {32, 1}, {1, 1}: 0.0130394ms, 0.114561% main:@2658 = squeeze[axes={1}](main:@2657) -> float_type, {32}, {1}: 0.00073496ms, 0.00645719% main:@2659 = hip::sync_stream(main:@2654,main:@2627,main:@2638,main:@2640,main:@2648,main:@2658) -> float_type, {32}, {1}: 0.0010593ms, 0.00930677% main:@2660 = @return(main:@2659,main:@2627,main:@2638,main:@2640,main:@2648,main:@2658) Summary: gpu::gemm: 4.795ms / 187 = 0.0256417ms, 42.1278% gpu::code_object::concat_kernel: 1.85721ms / 84 = 0.0221096ms, 16.317% slice: 0.610525ms / 859 = 0.000710739ms, 5.36393% gpu::code_object::relu_kernel: 0.258222ms / 18 = 0.0143457ms, 2.26868% load: 0.256201ms / 468 = 0.000547438ms, 2.25093% 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: 0.231511ms / 1 = 0.231511ms, 2.034% gpu::code_object::softmax_kernel: 0.218541ms / 14 = 0.0156101ms, 1.92005% @param: 0.209296ms / 383 = 0.000546465ms, 1.83883% gpu::code_object::convert_kernel: 0.200572ms / 15 = 0.0133715ms, 1.76218% gpu::code_object::contiguous_kernel: 0.136145ms / 10 = 0.0136145ms, 1.19614% hip::hip_copy_literal: 0.12708ms / 165 = 0.00077018ms, 1.11649% gpu::code_object::add_kernel: 0.109502ms / 7 = 0.0156431ms, 0.962057% unsqueeze: 0.0906772ms / 145 = 0.00062536ms, 0.796669% gpu::code_object::neg_pow_kernel: 0.0815248ms / 5 = 0.016305ms, 0.716259% gpu::code_object::log_add_kernel: 0.0801445ms / 6 = 0.0133574ms, 0.704132% gpu::code_object::neg_div_kernel: 0.0793854ms / 6 = 0.0132309ms, 0.697462% gpu::code_object::pow_kernel: 0.0675423ms / 4 = 0.0168856ms, 0.593411% gpu::code_object::noop_rsqrt_add_concat_noop_kernel: 0.0661019ms / 4 = 0.0165255ms, 0.580756% gpu::code_object::div_kernel: 0.0546515ms / 3 = 0.0182172ms, 0.480155% gpu::code_object::rsqrt_add_kernel: 0.0531141ms / 4 = 0.0132785ms, 0.466649% gpu::code_object::sub_kernel: 0.0501922ms / 3 = 0.0167307ms, 0.440977% gpu::code_object::rsqrt_pow_kernel: 0.049195ms / 3 = 0.0163983ms, 0.432216% reshape_lazy: 0.0457659ms / 81 = 0.000565011ms, 0.402089% gpu::code_object::reduce_prod_kernel: 0.0433494ms / 3 = 0.0144498ms, 0.380858% gpu::code_object::tanh_sub_kernel: 0.0411295ms / 3 = 0.0137098ms, 0.361354% gpu::code_object::convert_add_kernel: 0.0400045ms / 3 = 0.0133348ms, 0.351471% gpu::code_object::neg_add_kernel: 0.0397046ms / 3 = 0.0132349ms, 0.348835% gpu::code_object::noop_noop_noop_noop_noop_noop_noop_sub_noop_concat_neg_pow_kernel: 0.0372681ms / 1 = 0.0372681ms, 0.327429% gpu::code_object::pow_pow_kernel: 0.0361334ms / 2 = 0.0180667ms, 0.31746% gpu::code_object::leaky_relu_noop_noop_noop_noop_noop_noop_concat_rsqrt_pow_kernel: 0.0360556ms / 1 = 0.0360556ms, 0.316776% gpu::code_object::leaky_relu_noop_noop_noop_noop_add_concat_log_pow_kernel: 0.0330968ms / 1 = 0.0330968ms, 0.290781% gpu::code_object::log_pow_kernel: 0.0329999ms / 2 = 0.0165ms, 0.28993% gpu::code_object::noop_log_add_concat_noop_kernel: 0.0328382ms / 2 = 0.0164191ms, 0.288509% gpu::code_object::noop_rsqrt_div_concat_noop_kernel: 0.0323694ms / 2 = 0.0161847ms, 0.28439% gpu::code_object::leaky_relu_noop_noop_noop_noop_noop_noop_concat_log_add_pow_kernel: 0.0320615ms / 1 = 0.0320615ms, 0.281685% gpu::code_object::noop_tanh_div_concat_noop_kernel: 0.0308951ms / 2 = 0.0154475ms, 0.271437% gpu::code_object::add_leaky_relu_kernel: 0.0293978ms / 2 = 0.0146989ms, 0.258283% gpu::code_object::add_add_neg_add_kernel: 0.0287769ms / 2 = 0.0143884ms, 0.252827% gpu::code_object::tanh_div_kernel: 0.028193ms / 2 = 0.0140965ms, 0.247697% gpu::code_object::tanh_add_kernel: 0.0274579ms / 2 = 0.013729ms, 0.241239% gpu::code_object::convert_mul_add_kernel: 0.026522ms / 2 = 0.013261ms, 0.233016% gpu::code_object::mul_kernel: 0.0259928ms / 2 = 0.0129964ms, 0.228367% gpu::code_object::rsqrt_sub_kernel: 0.0259845ms / 2 = 0.0129923ms, 0.228294% gpu::code_object::pow_reduce_sum_div_pow_add_kernel: 0.0248742ms / 1 = 0.0248742ms, 0.218539% gpu::code_object::mul_relu_reduce_prod_sub_mul_reduce_sum_add_add_add_add_pow_kernel: 0.0226368ms / 1 = 0.0226368ms, 0.198882% multibroadcast: 0.022399ms / 35 = 0.000639972ms, 0.196793% gpu::code_object::leaky_relu_noop_noop_noop_add_noop_concat_log_div_kernel: 0.0221243ms / 1 = 0.0221243ms, 0.194379% gpu::code_object::pow_sub_div_pow_log_div_kernel: 0.0203936ms / 1 = 0.0203936ms, 0.179174% gpu::code_object::noop_tanh_pow_concat_noop_kernel: 0.0183057ms / 1 = 0.0183057ms, 0.16083% gpu::code_object::rsqrt_add_sub_mul_add_leaky_relu_noop_concat_noop_kernel: 0.018109ms / 1 = 0.018109ms, 0.159101% gpu::code_object::pow_sub_pow_pow_log_add_kernel: 0.0180569ms / 1 = 0.0180569ms, 0.158644% gpu::code_object::pow_pow_sigmoid_add_kernel: 0.0178393ms / 1 = 0.0178393ms, 0.156732% gpu::code_object::add_leaky_relu_tanh_pow_kernel: 0.0177837ms / 1 = 0.0177837ms, 0.156243% gpu::code_object::add_pow_log_pow_kernel: 0.0175068ms / 1 = 0.0175068ms, 0.153811% gpu::code_object::sub_sub_pow_log_pow_kernel: 0.0173261ms / 1 = 0.0173261ms, 0.152223% gpu::code_object::pow_mul_add_neg_sub_kernel: 0.0171976ms / 1 = 0.0171976ms, 0.151094% squeeze: 0.0170701ms / 25 = 0.000682804ms, 0.149974% gpu::code_object::noop_log_pow_concat_noop_kernel: 0.0170302ms / 1 = 0.0170302ms, 0.149624% gpu::code_object::pow_mul_rsqrt_sub_kernel: 0.0170131ms / 1 = 0.0170131ms, 0.149473% gpu::code_object::noop_neg_div_concat_noop_kernel: 0.016893ms / 1 = 0.016893ms, 0.148418% gpu::code_object::noop_log_sub_concat_noop_kernel: 0.0166907ms / 1 = 0.0166907ms, 0.146641% gpu::code_object::add_pow_kernel: 0.0162148ms / 1 = 0.0162148ms, 0.142459% gpu::code_object::neg_rsqrt_tanh_log_concat_add_kernel: 0.0158912ms / 1 = 0.0158912ms, 0.139617% gpu::code_object::pow_div_kernel: 0.0158243ms / 1 = 0.0158243ms, 0.139029% gpu::code_object::div_pow_log_add_kernel: 0.015497ms / 1 = 0.015497ms, 0.136153% gpu::code_object::reduce_mean_add_neg_pow_kernel: 0.015347ms / 1 = 0.015347ms, 0.134835% gpu::code_object::noop_rsqrt_sub_concat_noop_kernel: 0.0153404ms / 1 = 0.0153404ms, 0.134777% gpu::code_object::pow_mul_kernel: 0.01531ms / 1 = 0.01531ms, 0.13451% gpu::code_object::div_mul_neg_pow_kernel: 0.0152286ms / 1 = 0.0152286ms, 0.133795% gpu::code_object::reduce_mean_pow_log_pow_kernel: 0.0151859ms / 1 = 0.0151859ms, 0.13342% gpu::code_object::pow_sub_kernel: 0.0151598ms / 1 = 0.0151598ms, 0.133191% gpu::code_object::add_reduce_prod_kernel: 0.0151305ms / 1 = 0.0151305ms, 0.132933% gpu::code_object::add_add_rsqrt_pow_kernel: 0.0151244ms / 1 = 0.0151244ms, 0.13288% gpu::code_object::reduce_mean_add_rsqrt_add_kernel: 0.0151047ms / 1 = 0.0151047ms, 0.132706% gpu::code_object::reduce_prod_pow_neg_sub_kernel: 0.0149391ms / 1 = 0.0149391ms, 0.131251% gpu::code_object::reduce_prod_pow_rsqrt_div_kernel: 0.0149171ms / 1 = 0.0149171ms, 0.131058% gpu::code_object::noop_add_concat_div_kernel: 0.0148759ms / 1 = 0.0148759ms, 0.130696% gpu::code_object::add_reduce_mean_kernel: 0.0147629ms / 1 = 0.0147629ms, 0.129704% gpu::code_object::sigmoid_mul_tanh_kernel: 0.0147252ms / 1 = 0.0147252ms, 0.129373% gpu::code_object::reduce_prod_add_rsqrt_sub_kernel: 0.0147131ms / 1 = 0.0147131ms, 0.129266% gpu::code_object::reduce_sum_kernel: 0.0146871ms / 1 = 0.0146871ms, 0.129038% gpu::code_object::reduce_mean_kernel: 0.0145483ms / 1 = 0.0145483ms, 0.127818% gpu::code_object::sub_sub_mul_rsqrt_sub_kernel: 0.0144784ms / 1 = 0.0144784ms, 0.127204% gpu::code_object::reduce_prod_add_log_div_kernel: 0.014472ms / 1 = 0.014472ms, 0.127148% gpu::code_object::reduce_mean_add_sigmoid_sub_kernel: 0.0144356ms / 1 = 0.0144356ms, 0.126828% gpu::code_object::reduce_mean_add_sigmoid_kernel: 0.0144028ms / 1 = 0.0144028ms, 0.126539% gpu::code_object::reduce_prod_mul_rsqrt_add_kernel: 0.0143466ms / 1 = 0.0143466ms, 0.126046% gpu::code_object::reduce_mean_mul_rsqrt_kernel: 0.014338ms / 1 = 0.014338ms, 0.125971% gpu::code_object::reduce_mean_mul_neg_kernel: 0.0143064ms / 1 = 0.0143064ms, 0.125693% gpu::code_object::reduce_prod_mul_log_kernel: 0.0142658ms / 1 = 0.0142658ms, 0.125336% gpu::code_object::reduce_mean_add_neg_kernel: 0.0142596ms / 1 = 0.0142596ms, 0.125282% gpu::code_object::add_add_rsqrt_add_kernel: 0.0142525ms / 1 = 0.0142525ms, 0.125219% gpu::code_object::add_add_neg_div_kernel: 0.0141871ms / 1 = 0.0141871ms, 0.124645% gpu::code_object::reduce_prod_add_log_kernel: 0.014086ms / 1 = 0.014086ms, 0.123756% transpose: 0.0138009ms / 22 = 0.000627313ms, 0.121251% gpu::code_object::convert_div_mul_kernel: 0.0136679ms / 1 = 0.0136679ms, 0.120083% gpu::code_object::log_div_kernel: 0.01365ms / 1 = 0.01365ms, 0.119926% gpu::code_object::convert_add_add_sub_kernel: 0.0134677ms / 1 = 0.0134677ms, 0.118324% gpu::code_object::add_mul_kernel: 0.0133234ms / 1 = 0.0133234ms, 0.117056% gpu::code_object::convert_mul_kernel: 0.0132867ms / 1 = 0.0132867ms, 0.116734% gpu::code_object::leaky_relu_kernel: 0.013278ms / 1 = 0.013278ms, 0.116658% gpu::code_object::mul_neg_kernel: 0.0132762ms / 1 = 0.0132762ms, 0.116642% gpu::code_object::log_sub_kernel: 0.0132343ms / 1 = 0.0132343ms, 0.116273% gpu::code_object::add_sub_kernel: 0.0132317ms / 1 = 0.0132317ms, 0.116251% gpu::code_object::neg_sub_kernel: 0.0132143ms / 1 = 0.0132143ms, 0.116098% gpu::code_object::convert_pow_add_kernel: 0.0127999ms / 1 = 0.0127999ms, 0.112457% check_context::migraphx::version_1::gpu::context: 0.00169648ms / 1 = 0.00169648ms, 0.0149049% hip::hip_allocate_memory: 0.00111702ms / 1 = 0.00111702ms, 0.00981389% hip::sync_stream: 0.0010593ms / 1 = 0.0010593ms, 0.00930677% Batch size: 32 Rate: 6545.67/sec Total time: 4.88873ms Total instructions time: 11.382ms Overhead time: 0.830466ms, -6.49331ms Overhead: 17%, -133%