diff --git "a/lr1e-4_total_batch_size40960_seq_len512/log2.txt" "b/lr1e-4_total_batch_size40960_seq_len512/log2.txt" --- "a/lr1e-4_total_batch_size40960_seq_len512/log2.txt" +++ "b/lr1e-4_total_batch_size40960_seq_len512/log2.txt" @@ -10,1034 +10,1034 @@ max_steps: 1000 6 train 11.754723 (lr=2.4476e-06) (hash(x)=25129410) 7 train 11.757739 (lr=2.7972e-06) (hash(x)=26582391) 8 train 11.737188 (lr=3.1469e-06) (hash(x)=26472336) -9 train 11.743741 (lr=3.4965e-06) (hash(x)=27654289) -10 train 11.730678 (lr=3.8462e-06) (hash(x)=26559876) -11 train 11.731321 (lr=4.1958e-06) (hash(x)=21770329) -12 train 11.707128 (lr=4.5455e-06) (hash(x)=22723459) -13 train 11.695724 (lr=4.8951e-06) (hash(x)=30338342) -14 train 11.703027 (lr=5.2448e-06) (hash(x)=25678059) -15 train 11.684772 (lr=5.5944e-06) (hash(x)=26109335) -16 train 11.678885 (lr=5.9441e-06) (hash(x)=25044885) -17 train 11.671003 (lr=6.2937e-06) (hash(x)=28870690) -18 train 11.645406 (lr=6.6434e-06) (hash(x)=35279941) -19 train 11.638879 (lr=6.9930e-06) (hash(x)=25015923) -20 train 11.619357 (lr=7.3427e-06) (hash(x)=26740855) -21 train 11.615834 (lr=7.6923e-06) (hash(x)=22262151) -22 train 11.581081 (lr=8.0420e-06) (hash(x)=26621834) -23 train 11.559290 (lr=8.3916e-06) (hash(x)=23234741) -24 train 11.568782 (lr=8.7413e-06) (hash(x)=25470563) -25 train 11.542329 (lr=9.0909e-06) (hash(x)=21799102) -26 train 11.524164 (lr=9.4406e-06) (hash(x)=23275479) -27 train 11.509420 (lr=9.7902e-06) (hash(x)=22091862) -28 train 11.466881 (lr=1.0140e-05) (hash(x)=17430373) -29 train 11.452047 (lr=1.0490e-05) (hash(x)=22954861) -30 train 11.440687 (lr=1.0839e-05) (hash(x)=26671799) -31 train 11.428970 (lr=1.1189e-05) (hash(x)=26924723) -32 train 11.409254 (lr=1.1538e-05) (hash(x)=23557495) -33 train 11.348234 (lr=1.1888e-05) (hash(x)=27804380) -34 train 11.359153 (lr=1.2238e-05) (hash(x)=26654908) -35 train 11.311438 (lr=1.2587e-05) (hash(x)=23627518) -36 train 11.292107 (lr=1.2937e-05) (hash(x)=21889990) -37 train 11.270789 (lr=1.3287e-05) (hash(x)=22528001) -38 train 11.264645 (lr=1.3636e-05) (hash(x)=29248942) -39 train 11.186676 (lr=1.3986e-05) (hash(x)=25533417) -40 train 11.140759 (lr=1.4336e-05) (hash(x)=20612533) -41 train 11.126341 (lr=1.4685e-05) (hash(x)=24677740) -42 train 11.111518 (lr=1.5035e-05) (hash(x)=25347203) -43 train 11.053201 (lr=1.5385e-05) (hash(x)=28922363) -44 train 11.030757 (lr=1.5734e-05) (hash(x)=25631397) -45 train 10.987176 (lr=1.6084e-05) (hash(x)=26237983) -46 train 10.952265 (lr=1.6434e-05) (hash(x)=24742645) -47 train 10.945024 (lr=1.6783e-05) (hash(x)=26157660) -48 train 10.868473 (lr=1.7133e-05) (hash(x)=22582169) -49 train 10.837150 (lr=1.7483e-05) (hash(x)=26431960) -50 val loss 10.8427 -50 val perplexity 51159.7617 -50 train 10.849695 (lr=1.7832e-05) (hash(x)=25716176) -51 train 10.799539 (lr=1.8182e-05) (hash(x)=28019788) -52 train 10.774885 (lr=1.8531e-05) (hash(x)=20382963) -53 train 10.734505 (lr=1.8881e-05) (hash(x)=24489647) -54 train 10.719847 (lr=1.9231e-05) (hash(x)=25128502) -55 train 10.699546 (lr=1.9580e-05) (hash(x)=25231390) -56 train 10.619297 (lr=1.9930e-05) (hash(x)=22708977) -57 train 10.682159 (lr=2.0280e-05) (hash(x)=24364920) -58 train 10.638140 (lr=2.0629e-05) (hash(x)=24932925) -59 train 10.613176 (lr=2.0979e-05) (hash(x)=26913684) -60 train 10.540243 (lr=2.1329e-05) (hash(x)=18017792) -61 train 10.566302 (lr=2.1678e-05) (hash(x)=28613882) -62 train 10.566837 (lr=2.2028e-05) (hash(x)=22652243) -63 train 10.505860 (lr=2.2378e-05) (hash(x)=27194521) -64 train 10.498785 (lr=2.2727e-05) (hash(x)=24936836) -65 train 10.518402 (lr=2.3077e-05) (hash(x)=24765578) -66 train 10.546158 (lr=2.3427e-05) (hash(x)=23107416) -67 train 10.437639 (lr=2.3776e-05) (hash(x)=29314255) -68 train 10.427565 (lr=2.4126e-05) (hash(x)=24013079) -69 train 10.405937 (lr=2.4476e-05) (hash(x)=25322001) -70 train 10.500807 (lr=2.4825e-05) (hash(x)=26690440) -71 train 10.439799 (lr=2.5175e-05) (hash(x)=22598512) -72 train 10.416533 (lr=2.5524e-05) (hash(x)=18473911) -73 train 10.466362 (lr=2.5874e-05) (hash(x)=22322442) -74 train 10.445258 (lr=2.6224e-05) (hash(x)=26324153) -75 train 10.417565 (lr=2.6573e-05) (hash(x)=23446058) -76 train 10.463790 (lr=2.6923e-05) (hash(x)=24843741) -77 train 10.429681 (lr=2.7273e-05) (hash(x)=23197102) -78 train 10.503236 (lr=2.7622e-05) (hash(x)=23054940) -79 train 10.440751 (lr=2.7972e-05) (hash(x)=27140757) -80 train 10.326969 (lr=2.8322e-05) (hash(x)=25149419) -81 train 10.402832 (lr=2.8671e-05) (hash(x)=24727764) -82 train 10.274497 (lr=2.9021e-05) (hash(x)=21935931) -83 train 10.368818 (lr=2.9371e-05) (hash(x)=27606073) -84 train 10.395518 (lr=2.9720e-05) (hash(x)=22450341) -85 train 10.399356 (lr=3.0070e-05) (hash(x)=25945859) -86 train 10.353920 (lr=3.0420e-05) (hash(x)=21613707) -87 train 10.366814 (lr=3.0769e-05) (hash(x)=24396519) -88 train 10.361895 (lr=3.1119e-05) (hash(x)=29138828) -89 train 10.434487 (lr=3.1469e-05) (hash(x)=24685515) -90 train 10.327500 (lr=3.1818e-05) (hash(x)=22825749) -91 train 10.354466 (lr=3.2168e-05) (hash(x)=28767869) -92 train 10.339799 (lr=3.2517e-05) (hash(x)=21159060) -93 train 10.321786 (lr=3.2867e-05) (hash(x)=25045397) -94 train 10.399552 (lr=3.3217e-05) (hash(x)=25560928) -95 train 10.358747 (lr=3.3566e-05) (hash(x)=28959222) -96 train 10.308291 (lr=3.3916e-05) (hash(x)=21248405) -97 train 10.315656 (lr=3.4266e-05) (hash(x)=25830182) -98 train 10.263159 (lr=3.4615e-05) (hash(x)=25173113) -99 train 10.316659 (lr=3.4965e-05) (hash(x)=23486277) -100 val loss 10.3136 -100 val perplexity 30138.7266 -100 train 10.278738 (lr=3.5315e-05) (hash(x)=28916006) -101 train 10.347554 (lr=3.5664e-05) (hash(x)=25556929) -102 train 10.305278 (lr=3.6014e-05) (hash(x)=29648798) -103 train 10.257299 (lr=3.6364e-05) (hash(x)=24976217) -104 train 10.276778 (lr=3.6713e-05) (hash(x)=31019606) -105 train 10.283932 (lr=3.7063e-05) (hash(x)=28269421) -106 train 10.253519 (lr=3.7413e-05) (hash(x)=24865358) -107 train 10.158926 (lr=3.7762e-05) (hash(x)=26984272) -108 train 10.257380 (lr=3.8112e-05) (hash(x)=24544116) -109 train 10.240767 (lr=3.8462e-05) (hash(x)=22059850) -110 train 10.227790 (lr=3.8811e-05) (hash(x)=23926632) -111 train 10.198865 (lr=3.9161e-05) (hash(x)=27025333) -112 train 10.186539 (lr=3.9510e-05) (hash(x)=29120407) -113 train 10.234774 (lr=3.9860e-05) (hash(x)=23827393) -114 train 10.161518 (lr=4.0210e-05) (hash(x)=27230027) -115 train 10.172895 (lr=4.0559e-05) (hash(x)=24968260) -116 train 10.209167 (lr=4.0909e-05) (hash(x)=27544803) -117 train 10.185083 (lr=4.1259e-05) (hash(x)=26540663) -118 train 10.209141 (lr=4.1608e-05) (hash(x)=25856625) -119 train 10.159639 (lr=4.1958e-05) (hash(x)=23164356) -120 train 10.183915 (lr=4.2308e-05) (hash(x)=25646282) -121 train 10.201326 (lr=4.2657e-05) (hash(x)=23843233) -122 train 10.101642 (lr=4.3007e-05) (hash(x)=24506028) -123 train 10.141010 (lr=4.3357e-05) (hash(x)=22781277) -124 train 10.162203 (lr=4.3706e-05) (hash(x)=24948650) -125 train 10.117309 (lr=4.4056e-05) (hash(x)=26373991) -126 train 10.160105 (lr=4.4406e-05) (hash(x)=23402396) -127 train 10.119009 (lr=4.4755e-05) (hash(x)=21562133) -128 train 10.162827 (lr=4.5105e-05) (hash(x)=25009210) -129 train 10.060822 (lr=4.5455e-05) (hash(x)=27161811) -130 train 10.063899 (lr=4.5804e-05) (hash(x)=29515100) -131 train 10.056215 (lr=4.6154e-05) (hash(x)=25305929) -132 train 10.066180 (lr=4.6503e-05) (hash(x)=26808543) -133 train 10.030074 (lr=4.6853e-05) (hash(x)=25669509) -134 train 10.047958 (lr=4.7203e-05) (hash(x)=28435805) -135 train 10.047024 (lr=4.7552e-05) (hash(x)=26482758) -136 train 9.992270 (lr=4.7902e-05) (hash(x)=24929929) -137 train 10.049812 (lr=4.8252e-05) (hash(x)=25043037) -138 train 10.060753 (lr=4.8601e-05) (hash(x)=29151257) -139 train 10.021026 (lr=4.8951e-05) (hash(x)=26400365) -140 train 9.950485 (lr=4.9301e-05) (hash(x)=24098812) -141 train 10.005425 (lr=4.9650e-05) (hash(x)=20967666) -142 train 10.025003 (lr=5.0000e-05) (hash(x)=28667963) -143 train 10.000402 (lr=5.0350e-05) (hash(x)=27083417) -144 train 9.966627 (lr=5.0699e-05) (hash(x)=23649001) -145 train 9.948689 (lr=5.1049e-05) (hash(x)=31338300) -146 train 9.906151 (lr=5.1399e-05) (hash(x)=30603174) -147 train 9.904183 (lr=5.1748e-05) (hash(x)=24830965) -148 train 9.939874 (lr=5.2098e-05) (hash(x)=27269893) -149 train 9.906730 (lr=5.2448e-05) (hash(x)=24825041) -150 val loss 9.8926 -150 val perplexity 19783.4316 -150 train 9.884244 (lr=5.2797e-05) (hash(x)=22724871) -151 train 9.822634 (lr=5.3147e-05) (hash(x)=24683717) -152 train 9.867231 (lr=5.3497e-05) (hash(x)=26190337) -153 train 9.851982 (lr=5.3846e-05) (hash(x)=23086289) -154 train 9.881013 (lr=5.4196e-05) (hash(x)=23593875) -155 train 9.842882 (lr=5.4545e-05) (hash(x)=22414190) -156 train 9.833754 (lr=5.4895e-05) (hash(x)=26758114) -157 train 9.797348 (lr=5.5245e-05) (hash(x)=21563184) -158 train 9.810123 (lr=5.5594e-05) (hash(x)=19218470) -159 train 9.827530 (lr=5.5944e-05) (hash(x)=25588469) -160 train 9.773418 (lr=5.6294e-05) (hash(x)=30433767) -161 train 9.758258 (lr=5.6643e-05) (hash(x)=17851621) -162 train 9.797021 (lr=5.6993e-05) (hash(x)=25183690) -163 train 9.825255 (lr=5.7343e-05) (hash(x)=27040642) -164 train 9.700119 (lr=5.7692e-05) (hash(x)=22183303) -165 train 9.833939 (lr=5.8042e-05) (hash(x)=27537063) -166 train 9.701316 (lr=5.8392e-05) (hash(x)=19492832) -167 train 9.662508 (lr=5.8741e-05) (hash(x)=25951629) -168 train 9.630916 (lr=5.9091e-05) (hash(x)=22408682) -169 train 9.688861 (lr=5.9441e-05) (hash(x)=27633457) -170 train 9.684103 (lr=5.9790e-05) (hash(x)=27635080) -171 train 9.680268 (lr=6.0140e-05) (hash(x)=22287596) -172 train 9.690231 (lr=6.0490e-05) (hash(x)=35895440) -173 train 9.669487 (lr=6.0839e-05) (hash(x)=25888847) -174 train 9.591892 (lr=6.1189e-05) (hash(x)=26518800) -175 train 9.635827 (lr=6.1538e-05) (hash(x)=37519283) -176 train 9.680520 (lr=6.1888e-05) (hash(x)=25141945) -177 train 9.705601 (lr=6.2238e-05) (hash(x)=28527939) -178 train 9.587241 (lr=6.2587e-05) (hash(x)=21942471) -179 train 9.603369 (lr=6.2937e-05) (hash(x)=24955630) -180 train 9.544624 (lr=6.3287e-05) (hash(x)=25541230) -181 train 9.475161 (lr=6.3636e-05) (hash(x)=23501481) -182 train 9.496090 (lr=6.3986e-05) (hash(x)=22754988) -183 train 9.434362 (lr=6.4336e-05) (hash(x)=23475070) -184 train 9.505939 (lr=6.4685e-05) (hash(x)=23609959) -185 train 9.474903 (lr=6.5035e-05) (hash(x)=23441847) -186 train 9.463425 (lr=6.5385e-05) (hash(x)=23096292) -187 train 9.471959 (lr=6.5734e-05) (hash(x)=26171093) -188 train 9.478407 (lr=6.6084e-05) (hash(x)=20060838) -189 train 9.401480 (lr=6.6434e-05) (hash(x)=25451055) -190 train 9.306700 (lr=6.6783e-05) (hash(x)=24934792) -191 train 9.332983 (lr=6.7133e-05) (hash(x)=25367597) -192 train 9.398968 (lr=6.7483e-05) (hash(x)=25022148) -193 train 9.393510 (lr=6.7832e-05) (hash(x)=25431365) -194 train 9.383740 (lr=6.8182e-05) (hash(x)=29307788) -195 train 9.303251 (lr=6.8531e-05) (hash(x)=25651929) -196 train 9.343544 (lr=6.8881e-05) (hash(x)=26312387) -197 train 9.326360 (lr=6.9231e-05) (hash(x)=23804913) -198 train 9.381145 (lr=6.9580e-05) (hash(x)=26115519) -199 train 9.279168 (lr=6.9930e-05) (hash(x)=23210747) -200 val loss 9.2906 -200 val perplexity 10835.8887 -200 train 9.307692 (lr=7.0280e-05) (hash(x)=18463712) -201 train 9.248871 (lr=7.0629e-05) (hash(x)=23245699) -202 train 9.257250 (lr=7.0979e-05) (hash(x)=21770300) -203 train 9.407807 (lr=7.1329e-05) (hash(x)=25387532) -204 train 9.254056 (lr=7.1678e-05) (hash(x)=27718703) -205 train 9.333721 (lr=7.2028e-05) (hash(x)=27661052) -206 train 9.207791 (lr=7.2378e-05) (hash(x)=26538928) -207 train 9.226565 (lr=7.2727e-05) (hash(x)=21772654) -208 train 9.227448 (lr=7.3077e-05) (hash(x)=26263254) -209 train 9.214375 (lr=7.3427e-05) (hash(x)=23552356) -210 train 9.296247 (lr=7.3776e-05) (hash(x)=22707141) -211 train 9.288834 (lr=7.4126e-05) (hash(x)=34234481) -212 train 9.183087 (lr=7.4476e-05) (hash(x)=35008670) -213 train 9.190234 (lr=7.4825e-05) (hash(x)=25891543) -214 train 9.096066 (lr=7.5175e-05) (hash(x)=21971224) -215 train 9.123388 (lr=7.5524e-05) (hash(x)=27312186) -216 train 9.164174 (lr=7.5874e-05) (hash(x)=22163749) -217 train 9.005236 (lr=7.6224e-05) (hash(x)=25942888) -218 train 9.101052 (lr=7.6573e-05) (hash(x)=25493916) -219 train 9.198045 (lr=7.6923e-05) (hash(x)=24289951) -220 train 9.056392 (lr=7.7273e-05) (hash(x)=23648671) -221 train 9.034428 (lr=7.7622e-05) (hash(x)=24992055) -222 train 9.113957 (lr=7.7972e-05) (hash(x)=26005549) -223 train 9.034801 (lr=7.8322e-05) (hash(x)=21046554) -224 train 9.012487 (lr=7.8671e-05) (hash(x)=26321539) -225 train 9.029699 (lr=7.9021e-05) (hash(x)=27807872) -226 train 9.051228 (lr=7.9371e-05) (hash(x)=21555364) -227 train 9.027201 (lr=7.9720e-05) (hash(x)=26325324) -228 train 8.983374 (lr=8.0070e-05) (hash(x)=24956294) -229 train 8.982791 (lr=8.0420e-05) (hash(x)=24429389) -230 train 9.016384 (lr=8.0769e-05) (hash(x)=27929846) -231 train 8.987251 (lr=8.1119e-05) (hash(x)=25282897) -232 train 8.908170 (lr=8.1469e-05) (hash(x)=26802508) -233 train 8.965171 (lr=8.1818e-05) (hash(x)=28390416) -234 train 9.025198 (lr=8.2168e-05) (hash(x)=26292451) -235 train 8.994064 (lr=8.2517e-05) (hash(x)=22915344) -236 train 8.934509 (lr=8.2867e-05) (hash(x)=24744383) -237 train 8.837766 (lr=8.3217e-05) (hash(x)=26284678) -238 train 8.864168 (lr=8.3566e-05) (hash(x)=21654269) -239 train 9.119543 (lr=8.3916e-05) (hash(x)=29466268) -240 train 9.278403 (lr=8.4266e-05) (hash(x)=32616762) -241 train 9.244632 (lr=8.4615e-05) (hash(x)=32201138) -242 train 9.227572 (lr=8.4965e-05) (hash(x)=33683468) -243 train 8.983342 (lr=8.5315e-05) (hash(x)=27498125) -244 train 8.884466 (lr=8.5664e-05) (hash(x)=25164318) -245 train 8.854906 (lr=8.6014e-05) (hash(x)=25690221) -246 train 8.865967 (lr=8.6364e-05) (hash(x)=29031550) -247 train 8.893048 (lr=8.6713e-05) (hash(x)=24094093) -248 train 9.176089 (lr=8.7063e-05) (hash(x)=61455960) -249 train 10.917917 (lr=8.7413e-05) (hash(x)=61104710) -250 val loss 8.8090 -250 val perplexity 6694.3369 -250 train 8.819347 (lr=8.7762e-05) (hash(x)=20881126) -251 train 8.852946 (lr=8.8112e-05) (hash(x)=26920243) -252 train 8.754860 (lr=8.8462e-05) (hash(x)=21970343) -253 train 8.772523 (lr=8.8811e-05) (hash(x)=31676352) -254 train 8.780556 (lr=8.9161e-05) (hash(x)=25600562) -255 train 8.780517 (lr=8.9510e-05) (hash(x)=22022924) -256 train 8.734633 (lr=8.9860e-05) (hash(x)=22777085) -257 train 8.809948 (lr=9.0210e-05) (hash(x)=21727744) -258 train 8.832301 (lr=9.0559e-05) (hash(x)=20701998) -259 train 8.801223 (lr=9.0909e-05) (hash(x)=20863982) -260 train 8.732726 (lr=9.1259e-05) (hash(x)=25084926) -261 train 8.731787 (lr=9.1608e-05) (hash(x)=24412897) -262 train 8.662621 (lr=9.1958e-05) (hash(x)=25434913) -263 train 8.694664 (lr=9.2308e-05) (hash(x)=26302544) -264 train 8.731335 (lr=9.2657e-05) (hash(x)=34976197) -265 train 8.662732 (lr=9.3007e-05) (hash(x)=23870805) -266 train 8.664909 (lr=9.3357e-05) (hash(x)=25041103) -267 train 8.684845 (lr=9.3706e-05) (hash(x)=26041745) -268 train 8.644638 (lr=9.4056e-05) (hash(x)=24068619) -269 train 8.633161 (lr=9.4406e-05) (hash(x)=25441262) -270 train 8.645530 (lr=9.4755e-05) (hash(x)=29181867) -271 train 8.656697 (lr=9.5105e-05) (hash(x)=27093757) -272 train 8.553125 (lr=9.5455e-05) (hash(x)=25150275) -273 train 8.626158 (lr=9.5804e-05) (hash(x)=33457102) -274 train 8.531854 (lr=9.6154e-05) (hash(x)=25755322) -275 train 8.554464 (lr=9.6503e-05) (hash(x)=23630794) -276 train 8.472621 (lr=9.6853e-05) (hash(x)=25167987) -277 train 8.582054 (lr=9.7203e-05) (hash(x)=24043224) -278 train 8.552578 (lr=9.7552e-05) (hash(x)=24665411) -279 train 8.544823 (lr=9.7902e-05) (hash(x)=17321657) -280 train 8.504908 (lr=9.8252e-05) (hash(x)=27932591) -281 train 8.608794 (lr=9.8601e-05) (hash(x)=29447954) -282 train 8.534290 (lr=9.8951e-05) (hash(x)=24621816) -283 train 8.502698 (lr=9.9301e-05) (hash(x)=24682294) -284 train 8.497841 (lr=9.9650e-05) (hash(x)=28057095) -285 train 8.522227 (lr=1.0000e-04) (hash(x)=24547536) -286 train 8.350670 (lr=1.0000e-04) (hash(x)=18987545) -287 train 8.436469 (lr=1.0000e-04) (hash(x)=26745063) -288 train 8.365735 (lr=9.9998e-05) (hash(x)=21447406) -289 train 8.309195 (lr=9.9996e-05) (hash(x)=24650655) -290 train 8.513535 (lr=9.9993e-05) (hash(x)=26519976) -291 train 8.778757 (lr=9.9989e-05) (hash(x)=20106514) -292 train 8.315075 (lr=9.9984e-05) (hash(x)=20117808) -293 train 8.413671 (lr=9.9979e-05) (hash(x)=25510334) -294 train 8.340737 (lr=9.9972e-05) (hash(x)=25749120) -295 train 8.399283 (lr=9.9965e-05) (hash(x)=24357241) -296 train 8.379560 (lr=9.9956e-05) (hash(x)=25584945) -297 train 8.299529 (lr=9.9947e-05) (hash(x)=25198897) -298 train 8.308249 (lr=9.9937e-05) (hash(x)=23475891) -299 train 8.287750 (lr=9.9926e-05) (hash(x)=23644804) -300 val loss 8.3086 -300 val perplexity 4058.6750 -300 train 8.435748 (lr=9.9915e-05) (hash(x)=25629584) -301 train 8.480757 (lr=9.9902e-05) (hash(x)=29797280) -302 train 8.497992 (lr=9.9889e-05) (hash(x)=30026394) -303 train 8.419273 (lr=9.9874e-05) (hash(x)=26594177) -304 train 8.305301 (lr=9.9859e-05) (hash(x)=23766256) -305 train 8.259278 (lr=9.9843e-05) (hash(x)=24250150) -306 train 8.253291 (lr=9.9826e-05) (hash(x)=26108817) -307 train 8.200242 (lr=9.9808e-05) (hash(x)=18582688) -308 train 8.176954 (lr=9.9789e-05) (hash(x)=23272705) -309 train 8.152193 (lr=9.9770e-05) (hash(x)=24793412) -310 train 8.166468 (lr=9.9749e-05) (hash(x)=19280989) -311 train 8.157917 (lr=9.9728e-05) (hash(x)=22751150) -312 train 8.245569 (lr=9.9706e-05) (hash(x)=23537942) -313 train 8.102750 (lr=9.9683e-05) (hash(x)=23748375) -314 train 8.187785 (lr=9.9659e-05) (hash(x)=27413825) -315 train 8.191016 (lr=9.9634e-05) (hash(x)=21682445) -316 train 8.125028 (lr=9.9609e-05) (hash(x)=22040184) -317 train 8.069901 (lr=9.9582e-05) (hash(x)=24463229) -318 train 8.183447 (lr=9.9555e-05) (hash(x)=24091954) -319 train 8.102878 (lr=9.9526e-05) (hash(x)=26670219) -320 train 8.085333 (lr=9.9497e-05) (hash(x)=25558986) -321 train 8.049417 (lr=9.9467e-05) (hash(x)=19218943) -322 train 8.143736 (lr=9.9437e-05) (hash(x)=24551402) -323 train 8.042617 (lr=9.9405e-05) (hash(x)=25229281) -324 train 7.878533 (lr=9.9372e-05) (hash(x)=21251127) -325 train 8.064113 (lr=9.9339e-05) (hash(x)=25800857) -326 train 7.988725 (lr=9.9305e-05) (hash(x)=25836189) -327 train 8.010557 (lr=9.9270e-05) (hash(x)=25999255) -328 train 8.048934 (lr=9.9234e-05) (hash(x)=20149394) -329 train 7.997222 (lr=9.9197e-05) (hash(x)=21999338) -330 train 8.070961 (lr=9.9159e-05) (hash(x)=27611302) -331 train 8.037538 (lr=9.9121e-05) (hash(x)=22778356) -332 train 7.991700 (lr=9.9081e-05) (hash(x)=24195959) -333 train 8.208155 (lr=9.9041e-05) (hash(x)=19127465) -334 train 8.072520 (lr=9.9000e-05) (hash(x)=22426274) -335 train 7.929836 (lr=9.8958e-05) (hash(x)=23785205) -336 train 8.038789 (lr=9.8915e-05) (hash(x)=28505676) -337 train 7.982605 (lr=9.8872e-05) (hash(x)=27706477) -338 train 8.036742 (lr=9.8827e-05) (hash(x)=28684175) -339 train 8.014040 (lr=9.8782e-05) (hash(x)=28003600) -340 train 8.014636 (lr=9.8736e-05) (hash(x)=27586845) -341 train 7.955000 (lr=9.8689e-05) (hash(x)=21176405) -342 train 7.877286 (lr=9.8641e-05) (hash(x)=25888433) -343 train 8.016254 (lr=9.8592e-05) (hash(x)=27916982) -344 train 7.874392 (lr=9.8543e-05) (hash(x)=24334708) -345 train 7.941211 (lr=9.8492e-05) (hash(x)=25617709) -346 train 8.036167 (lr=9.8441e-05) (hash(x)=25152362) -347 train 7.900301 (lr=9.8389e-05) (hash(x)=24276512) -348 train 7.813753 (lr=9.8336e-05) (hash(x)=26776133) -349 train 7.775318 (lr=9.8282e-05) (hash(x)=20254159) -350 val loss 7.9409 -350 val perplexity 2809.8557 -350 train 7.814607 (lr=9.8228e-05) (hash(x)=27873855) -351 train 7.906783 (lr=9.8172e-05) (hash(x)=25113614) -352 train 7.965298 (lr=9.8116e-05) (hash(x)=26192886) -353 train 7.856303 (lr=9.8059e-05) (hash(x)=24614912) -354 train 7.846335 (lr=9.8001e-05) (hash(x)=25397093) -355 train 7.875048 (lr=9.7942e-05) (hash(x)=19811802) -356 train 7.864684 (lr=9.7882e-05) (hash(x)=25579655) -357 train 7.962258 (lr=9.7822e-05) (hash(x)=26096514) -358 train 7.858101 (lr=9.7761e-05) (hash(x)=22589633) -359 train 7.864480 (lr=9.7699e-05) (hash(x)=26757147) -360 train 7.956186 (lr=9.7636e-05) (hash(x)=24963730) -361 train 7.854289 (lr=9.7572e-05) (hash(x)=27703124) -362 train 7.743626 (lr=9.7507e-05) (hash(x)=24781713) -363 train 7.891638 (lr=9.7442e-05) (hash(x)=24720476) -364 train 7.875727 (lr=9.7376e-05) (hash(x)=23481301) -365 train 7.793710 (lr=9.7309e-05) (hash(x)=25641033) -366 train 7.846750 (lr=9.7241e-05) (hash(x)=27369437) -367 train 7.884250 (lr=9.7172e-05) (hash(x)=24760900) -368 train 7.763644 (lr=9.7103e-05) (hash(x)=21692546) -369 train 7.829751 (lr=9.7032e-05) (hash(x)=22720391) -370 train 7.782937 (lr=9.6961e-05) (hash(x)=24376979) -371 train 7.795093 (lr=9.6889e-05) (hash(x)=22244509) -372 train 7.745054 (lr=9.6817e-05) (hash(x)=21831960) -373 train 7.725542 (lr=9.6743e-05) (hash(x)=24921535) -374 train 7.737894 (lr=9.6669e-05) (hash(x)=25381251) -375 train 7.736143 (lr=9.6593e-05) (hash(x)=25687352) -376 train 7.763109 (lr=9.6518e-05) (hash(x)=25221654) -377 train 7.808217 (lr=9.6441e-05) (hash(x)=24469863) -378 train 7.841477 (lr=9.6363e-05) (hash(x)=25291413) -379 train 7.748312 (lr=9.6285e-05) (hash(x)=25744274) -380 train 7.685605 (lr=9.6206e-05) (hash(x)=24246963) -381 train 7.710941 (lr=9.6126e-05) (hash(x)=19724058) -382 train 7.667651 (lr=9.6045e-05) (hash(x)=20705649) -383 train 7.874396 (lr=9.5963e-05) (hash(x)=24938685) -384 train 7.897980 (lr=9.5881e-05) (hash(x)=21707000) -385 train 7.711274 (lr=9.5798e-05) (hash(x)=24948980) -386 train 7.939587 (lr=9.5714e-05) (hash(x)=30773958) -387 train 8.002645 (lr=9.5629e-05) (hash(x)=26769046) -388 train 7.721035 (lr=9.5544e-05) (hash(x)=18849656) -389 train 8.200633 (lr=9.5457e-05) (hash(x)=24624950) -390 train 7.816331 (lr=9.5370e-05) (hash(x)=27299605) -391 train 7.807438 (lr=9.5282e-05) (hash(x)=24113445) -392 train 7.825818 (lr=9.5194e-05) (hash(x)=32644465) -393 train 7.902504 (lr=9.5104e-05) (hash(x)=25829219) -394 train 7.739758 (lr=9.5014e-05) (hash(x)=25194550) -395 train 7.775651 (lr=9.4923e-05) (hash(x)=25941804) -396 train 7.771498 (lr=9.4831e-05) (hash(x)=24075727) -397 train 7.813376 (lr=9.4739e-05) (hash(x)=23104539) -398 train 7.750863 (lr=9.4646e-05) (hash(x)=23018983) -399 train 7.701500 (lr=9.4551e-05) (hash(x)=24422929) -400 val loss 7.7568 -400 val perplexity 2337.3364 -400 train 7.736512 (lr=9.4457e-05) (hash(x)=23389696) -401 train 7.697756 (lr=9.4361e-05) (hash(x)=29102969) -402 train 7.838785 (lr=9.4265e-05) (hash(x)=28359859) -403 train 7.818171 (lr=9.4168e-05) (hash(x)=24630955) -404 train 7.868329 (lr=9.4070e-05) (hash(x)=22969113) -405 train 7.691509 (lr=9.3971e-05) (hash(x)=27004415) -406 train 7.626100 (lr=9.3872e-05) (hash(x)=22265063) -407 train 7.663659 (lr=9.3772e-05) (hash(x)=26905990) -408 train 7.703139 (lr=9.3671e-05) (hash(x)=24047679) -409 train 7.832115 (lr=9.3569e-05) (hash(x)=26905582) -410 train 7.611836 (lr=9.3467e-05) (hash(x)=29018970) -411 train 7.613941 (lr=9.3364e-05) (hash(x)=24503315) -412 train 7.623048 (lr=9.3260e-05) (hash(x)=22607537) -413 train 8.289001 (lr=9.3155e-05) (hash(x)=26690225) -414 train 7.682450 (lr=9.3050e-05) (hash(x)=24226255) -415 train 7.630458 (lr=9.2944e-05) (hash(x)=23622702) -416 train 7.686643 (lr=9.2837e-05) (hash(x)=23910425) -417 train 7.893850 (lr=9.2729e-05) (hash(x)=27213318) -418 train 7.703747 (lr=9.2621e-05) (hash(x)=26701577) -419 train 7.645214 (lr=9.2512e-05) (hash(x)=26216794) -420 train 7.847490 (lr=9.2402e-05) (hash(x)=27579623) -421 train 7.690247 (lr=9.2292e-05) (hash(x)=26086198) -422 train 7.662221 (lr=9.2181e-05) (hash(x)=24863441) -423 train 7.781695 (lr=9.2069e-05) (hash(x)=25236385) -424 train 7.803724 (lr=9.1956e-05) (hash(x)=25330803) -425 train 7.693798 (lr=9.1843e-05) (hash(x)=22508532) -426 train 7.672997 (lr=9.1729e-05) (hash(x)=21799261) -427 train 7.764224 (lr=9.1614e-05) (hash(x)=26311168) -428 train 7.560589 (lr=9.1499e-05) (hash(x)=20977430) -429 train 7.622502 (lr=9.1382e-05) (hash(x)=26941617) -430 train 7.873599 (lr=9.1266e-05) (hash(x)=30008957) -431 train 7.739056 (lr=9.1148e-05) (hash(x)=27125962) -432 train 7.598950 (lr=9.1030e-05) (hash(x)=25888358) -433 train 7.733124 (lr=9.0911e-05) (hash(x)=21725719) -434 train 7.584195 (lr=9.0791e-05) (hash(x)=21669704) -435 train 7.572590 (lr=9.0671e-05) (hash(x)=23166993) -436 train 7.638998 (lr=9.0550e-05) (hash(x)=26235974) -437 train 7.788588 (lr=9.0428e-05) (hash(x)=25426430) -438 train 7.534440 (lr=9.0306e-05) (hash(x)=22966314) -439 train 7.683634 (lr=9.0182e-05) (hash(x)=23448505) -440 train 7.550514 (lr=9.0059e-05) (hash(x)=26330693) -441 train 7.503012 (lr=8.9934e-05) (hash(x)=29550596) -442 train 7.569022 (lr=8.9809e-05) (hash(x)=22124892) -443 train 7.593273 (lr=8.9683e-05) (hash(x)=25607640) -444 train 7.440102 (lr=8.9557e-05) (hash(x)=23795894) -445 train 7.334437 (lr=8.9430e-05) (hash(x)=24165449) -446 train 7.597899 (lr=8.9302e-05) (hash(x)=24697756) -447 train 7.681522 (lr=8.9173e-05) (hash(x)=27218899) -448 train 7.567974 (lr=8.9044e-05) (hash(x)=24748522) -449 train 7.648011 (lr=8.8914e-05) (hash(x)=24965317) -450 val loss 7.6558 -450 val perplexity 2112.8896 -450 train 7.441152 (lr=8.8784e-05) (hash(x)=22031210) -451 train 7.567151 (lr=8.8653e-05) (hash(x)=25877990) -452 train 7.608446 (lr=8.8521e-05) (hash(x)=24382381) -453 train 7.512006 (lr=8.8388e-05) (hash(x)=24551999) -454 train 7.613674 (lr=8.8255e-05) (hash(x)=25765516) -455 train 7.634459 (lr=8.8122e-05) (hash(x)=25204247) -456 train 7.681192 (lr=8.7987e-05) (hash(x)=26321813) -457 train 7.611825 (lr=8.7852e-05) (hash(x)=25528323) -458 train 7.602286 (lr=8.7717e-05) (hash(x)=25114729) -459 train 7.576212 (lr=8.7580e-05) (hash(x)=31341964) -460 train 7.621799 (lr=8.7444e-05) (hash(x)=26326024) -461 train 7.538342 (lr=8.7306e-05) (hash(x)=27273311) -462 train 7.527748 (lr=8.7168e-05) (hash(x)=17819313) -463 train 7.676411 (lr=8.7029e-05) (hash(x)=23769243) -464 train 7.633408 (lr=8.6890e-05) (hash(x)=23788738) -465 train 7.584522 (lr=8.6750e-05) (hash(x)=24818744) -466 train 7.514266 (lr=8.6609e-05) (hash(x)=23472795) -467 train 7.910563 (lr=8.6468e-05) (hash(x)=24991832) -468 train 7.652891 (lr=8.6326e-05) (hash(x)=22424066) -469 train 7.544734 (lr=8.6184e-05) (hash(x)=26474878) -470 train 7.638586 (lr=8.6041e-05) (hash(x)=22704349) -471 train 7.575641 (lr=8.5897e-05) (hash(x)=22293673) -472 train 7.600972 (lr=8.5753e-05) (hash(x)=21785487) -473 train 7.602975 (lr=8.5608e-05) (hash(x)=25251063) -474 train 7.506220 (lr=8.5462e-05) (hash(x)=22722211) -475 train 7.570879 (lr=8.5316e-05) (hash(x)=24522269) -476 train 7.521543 (lr=8.5170e-05) (hash(x)=23329475) -477 train 7.545805 (lr=8.5022e-05) (hash(x)=24735165) -478 train 7.364789 (lr=8.4875e-05) (hash(x)=27002892) -479 train 7.408906 (lr=8.4726e-05) (hash(x)=23499349) -480 train 7.739799 (lr=8.4577e-05) (hash(x)=22461589) -481 train 7.506948 (lr=8.4428e-05) (hash(x)=26415040) -482 train 7.734367 (lr=8.4278e-05) (hash(x)=26349465) -483 train 7.407123 (lr=8.4127e-05) (hash(x)=21596677) -484 train 7.458468 (lr=8.3976e-05) (hash(x)=24117012) -485 train 7.531465 (lr=8.3824e-05) (hash(x)=24105761) -486 train 7.468420 (lr=8.3672e-05) (hash(x)=25725049) -487 train 7.322726 (lr=8.3519e-05) (hash(x)=22892776) -488 train 7.429528 (lr=8.3366e-05) (hash(x)=21630383) -489 train 7.487536 (lr=8.3212e-05) (hash(x)=24727094) -490 train 7.420637 (lr=8.3057e-05) (hash(x)=22835000) -491 train 7.591608 (lr=8.2902e-05) (hash(x)=22839049) -492 train 7.435911 (lr=8.2746e-05) (hash(x)=20654136) -493 train 7.659186 (lr=8.2590e-05) (hash(x)=26776000) -494 train 7.273840 (lr=8.2434e-05) (hash(x)=22560658) -495 train 7.715094 (lr=8.2276e-05) (hash(x)=25995933) -496 train 7.653127 (lr=8.2119e-05) (hash(x)=24635394) -497 train 7.734672 (lr=8.1960e-05) (hash(x)=27537228) -498 train 7.632674 (lr=8.1801e-05) (hash(x)=26496730) -499 train 7.370257 (lr=8.1642e-05) (hash(x)=16533310) -500 val loss 7.6008 -500 val perplexity 1999.7646 -500 train 7.491931 (lr=8.1482e-05) (hash(x)=24881176) -501 train 7.699295 (lr=8.1322e-05) (hash(x)=25593613) -502 train 7.456593 (lr=8.1161e-05) (hash(x)=22886646) -503 train 7.718415 (lr=8.1000e-05) (hash(x)=21407676) -504 train 7.588889 (lr=8.0838e-05) (hash(x)=21720304) -505 train 7.715669 (lr=8.0676e-05) (hash(x)=27263416) -506 train 7.482646 (lr=8.0513e-05) (hash(x)=25840801) -507 train 7.624564 (lr=8.0349e-05) (hash(x)=31424568) -508 train 7.625317 (lr=8.0186e-05) (hash(x)=23403065) -509 train 7.449756 (lr=8.0021e-05) (hash(x)=22656033) -510 train 7.584896 (lr=7.9856e-05) (hash(x)=26869295) -511 train 7.835832 (lr=7.9691e-05) (hash(x)=25504151) -512 train 7.613889 (lr=7.9525e-05) (hash(x)=25781518) -513 train 7.579289 (lr=7.9359e-05) (hash(x)=26432850) -514 train 7.452091 (lr=7.9192e-05) (hash(x)=20155894) -515 train 7.367572 (lr=7.9025e-05) (hash(x)=22270169) -516 train 7.516262 (lr=7.8858e-05) (hash(x)=22108461) -517 train 7.582701 (lr=7.8689e-05) (hash(x)=24802848) -518 train 7.664511 (lr=7.8521e-05) (hash(x)=25086604) -519 train 7.665092 (lr=7.8352e-05) (hash(x)=26983186) -520 train 7.527274 (lr=7.8182e-05) (hash(x)=27208507) -521 train 7.712989 (lr=7.8012e-05) (hash(x)=23862845) -522 train 7.528232 (lr=7.7842e-05) (hash(x)=22664494) -523 train 7.505093 (lr=7.7671e-05) (hash(x)=24092927) -524 train 7.492690 (lr=7.7500e-05) (hash(x)=27213812) -525 train 7.467358 (lr=7.7328e-05) (hash(x)=23929955) -526 train 7.472635 (lr=7.7156e-05) (hash(x)=23025790) -527 train 7.418207 (lr=7.6984e-05) (hash(x)=20452272) -528 train 7.614613 (lr=7.6811e-05) (hash(x)=26473464) -529 train 7.583344 (lr=7.6637e-05) (hash(x)=28276603) -530 train 7.414617 (lr=7.6463e-05) (hash(x)=21480367) -531 train 7.611022 (lr=7.6289e-05) (hash(x)=24658642) -532 train 7.503337 (lr=7.6115e-05) (hash(x)=26854801) -533 train 7.397410 (lr=7.5940e-05) (hash(x)=17862226) -534 train 7.505068 (lr=7.5764e-05) (hash(x)=28519182) -535 train 7.485750 (lr=7.5588e-05) (hash(x)=25193694) -536 train 7.473051 (lr=7.5412e-05) (hash(x)=25642620) -537 train 7.547242 (lr=7.5235e-05) (hash(x)=25765909) -538 train 7.373703 (lr=7.5058e-05) (hash(x)=20750531) -539 train 7.479123 (lr=7.4881e-05) (hash(x)=24917098) -540 train 7.434493 (lr=7.4703e-05) (hash(x)=24499766) -541 train 7.482337 (lr=7.4525e-05) (hash(x)=25067358) -542 train 7.601202 (lr=7.4346e-05) (hash(x)=27479796) -543 train 7.478824 (lr=7.4167e-05) (hash(x)=23136080) -544 train 7.554965 (lr=7.3988e-05) (hash(x)=29810074) -545 train 7.448411 (lr=7.3808e-05) (hash(x)=24765539) -546 train 7.487386 (lr=7.3628e-05) (hash(x)=22216309) -547 train 7.414191 (lr=7.3448e-05) (hash(x)=20613861) -548 train 7.385593 (lr=7.3267e-05) (hash(x)=23462447) -549 train 7.452051 (lr=7.3086e-05) (hash(x)=26739455) -550 val loss 7.5604 -550 val perplexity 1920.5990 -550 train 7.514684 (lr=7.2904e-05) (hash(x)=23222996) -551 train 7.536718 (lr=7.2723e-05) (hash(x)=26489285) -552 train 7.348818 (lr=7.2540e-05) (hash(x)=25185942) -553 train 7.680166 (lr=7.2358e-05) (hash(x)=27476158) -554 train 7.436522 (lr=7.2175e-05) (hash(x)=23653175) -555 train 7.416105 (lr=7.1992e-05) (hash(x)=22405076) -556 train 7.441228 (lr=7.1808e-05) (hash(x)=25673545) -557 train 7.533946 (lr=7.1624e-05) (hash(x)=22675428) -558 train 7.641112 (lr=7.1440e-05) (hash(x)=28168456) -559 train 7.756699 (lr=7.1256e-05) (hash(x)=26072586) -560 train 7.549207 (lr=7.1071e-05) (hash(x)=24974331) -561 train 7.352679 (lr=7.0886e-05) (hash(x)=21951984) -562 train 7.707144 (lr=7.0701e-05) (hash(x)=25403751) -563 train 7.493234 (lr=7.0515e-05) (hash(x)=26396012) -564 train 7.467531 (lr=7.0329e-05) (hash(x)=24827190) -565 train 7.585389 (lr=7.0143e-05) (hash(x)=24847458) -566 train 7.553688 (lr=6.9956e-05) (hash(x)=22159088) -567 train 7.492432 (lr=6.9769e-05) (hash(x)=22140034) -568 train 7.537660 (lr=6.9582e-05) (hash(x)=19044797) -569 train 7.436035 (lr=6.9394e-05) (hash(x)=23932794) -570 train 7.545462 (lr=6.9207e-05) (hash(x)=25256890) -571 train 7.619053 (lr=6.9019e-05) (hash(x)=24161711) -572 train 7.531468 (lr=6.8830e-05) (hash(x)=22036817) -573 train 7.371822 (lr=6.8642e-05) (hash(x)=26663208) -574 train 7.441002 (lr=6.8453e-05) (hash(x)=24072213) -575 train 7.432795 (lr=6.8264e-05) (hash(x)=22766044) -576 train 7.551762 (lr=6.8075e-05) (hash(x)=27871153) -577 train 7.525133 (lr=6.7885e-05) (hash(x)=24037280) -578 train 7.630750 (lr=6.7695e-05) (hash(x)=24121850) -579 train 7.472374 (lr=6.7505e-05) (hash(x)=24380031) -580 train 7.429338 (lr=6.7315e-05) (hash(x)=23055215) -581 train 7.702837 (lr=6.7124e-05) (hash(x)=25329724) -582 train 7.414389 (lr=6.6933e-05) (hash(x)=25662408) -583 train 7.511388 (lr=6.6742e-05) (hash(x)=26346814) -584 train 7.619717 (lr=6.6551e-05) (hash(x)=25283256) -585 train 7.421248 (lr=6.6360e-05) (hash(x)=24021771) -586 train 7.440869 (lr=6.6168e-05) (hash(x)=23576840) -587 train 7.605403 (lr=6.5976e-05) (hash(x)=26414858) -588 train 7.356205 (lr=6.5784e-05) (hash(x)=20099261) -589 train 7.365086 (lr=6.5592e-05) (hash(x)=22453718) -590 train 7.329267 (lr=6.5399e-05) (hash(x)=22529445) -591 train 7.415786 (lr=6.5206e-05) (hash(x)=24484543) -592 train 7.343019 (lr=6.5013e-05) (hash(x)=24043998) -593 train 7.474518 (lr=6.4820e-05) (hash(x)=24301906) -594 train 7.408827 (lr=6.4627e-05) (hash(x)=26328881) -595 train 7.460378 (lr=6.4433e-05) (hash(x)=23101508) -596 train 7.438038 (lr=6.4240e-05) (hash(x)=23527176) -597 train 7.950079 (lr=6.4046e-05) (hash(x)=22705673) -598 train 7.496606 (lr=6.3852e-05) (hash(x)=28664796) -599 train 7.621476 (lr=6.3658e-05) (hash(x)=27102890) -600 val loss 7.5413 -600 val perplexity 1884.3082 -600 train 7.697159 (lr=6.3463e-05) (hash(x)=26787259) -601 train 7.742362 (lr=6.3269e-05) (hash(x)=23396088) -602 train 7.458788 (lr=6.3074e-05) (hash(x)=24911831) -603 train 7.484278 (lr=6.2879e-05) (hash(x)=23453491) -604 train 7.530580 (lr=6.2684e-05) (hash(x)=23453396) -605 train 7.621758 (lr=6.2489e-05) (hash(x)=23869612) -606 train 7.710625 (lr=6.2294e-05) (hash(x)=26581590) -607 train 7.530530 (lr=6.2098e-05) (hash(x)=28056342) -608 train 7.592166 (lr=6.1903e-05) (hash(x)=25900866) -609 train 7.621538 (lr=6.1707e-05) (hash(x)=24549177) -610 train 7.532023 (lr=6.1511e-05) (hash(x)=28239583) -611 train 7.558032 (lr=6.1315e-05) (hash(x)=23223554) -612 train 7.627146 (lr=6.1119e-05) (hash(x)=24793480) -613 train 7.514314 (lr=6.0923e-05) (hash(x)=25415570) -614 train 7.585463 (lr=6.0726e-05) (hash(x)=24425760) -615 train 7.648799 (lr=6.0530e-05) (hash(x)=23707134) -616 train 7.598625 (lr=6.0333e-05) (hash(x)=28321697) -617 train 7.526546 (lr=6.0137e-05) (hash(x)=23434031) -618 train 7.507131 (lr=5.9940e-05) (hash(x)=27003481) -619 train 7.496655 (lr=5.9743e-05) (hash(x)=22236893) -620 train 7.521164 (lr=5.9546e-05) (hash(x)=26857458) -621 train 7.518344 (lr=5.9349e-05) (hash(x)=26413122) -622 train 7.823038 (lr=5.9152e-05) (hash(x)=27696537) -623 train 7.550300 (lr=5.8955e-05) (hash(x)=21950234) -624 train 7.573440 (lr=5.8758e-05) (hash(x)=24051952) -625 train 7.582304 (lr=5.8560e-05) (hash(x)=24029939) -626 train 7.670248 (lr=5.8363e-05) (hash(x)=24393346) -627 train 7.632165 (lr=5.8165e-05) (hash(x)=26985352) -628 train 7.384880 (lr=5.7968e-05) (hash(x)=19493265) -629 train 7.458791 (lr=5.7770e-05) (hash(x)=23944010) -630 train 7.673358 (lr=5.7573e-05) (hash(x)=21517435) -631 train 7.438535 (lr=5.7375e-05) (hash(x)=25025868) -632 train 7.457140 (lr=5.7177e-05) (hash(x)=22940997) -633 train 7.595205 (lr=5.6979e-05) (hash(x)=25864840) -634 train 7.494565 (lr=5.6782e-05) (hash(x)=24682151) -635 train 7.417654 (lr=5.6584e-05) (hash(x)=27055330) -636 train 7.468966 (lr=5.6386e-05) (hash(x)=22705710) -637 train 7.574765 (lr=5.6188e-05) (hash(x)=26775238) -638 train 7.712748 (lr=5.5990e-05) (hash(x)=21734331) -639 train 7.437728 (lr=5.5792e-05) (hash(x)=25693442) -640 train 7.551642 (lr=5.5594e-05) (hash(x)=25290454) -641 train 7.511855 (lr=5.5396e-05) (hash(x)=25699860) -642 train 7.678161 (lr=5.5198e-05) (hash(x)=19919115) -643 train 7.760191 (lr=5.5000e-05) (hash(x)=25586967) -644 train 7.481709 (lr=5.4802e-05) (hash(x)=25373506) -645 train 7.493378 (lr=5.4604e-05) (hash(x)=26496148) -646 train 7.566517 (lr=5.4406e-05) (hash(x)=27395083) -647 train 7.766306 (lr=5.4208e-05) (hash(x)=27389709) -648 train 7.630821 (lr=5.4010e-05) (hash(x)=29172968) -649 train 7.560219 (lr=5.3812e-05) (hash(x)=25464920) -650 val loss 7.5098 -650 val perplexity 1825.8177 -650 train 7.589197 (lr=5.3614e-05) (hash(x)=25992320) -651 train 7.574807 (lr=5.3416e-05) (hash(x)=24949509) -652 train 7.610744 (lr=5.3218e-05) (hash(x)=25615151) -653 train 7.565460 (lr=5.3021e-05) (hash(x)=27923548) -654 train 7.601680 (lr=5.2823e-05) (hash(x)=26387848) -655 train 7.543574 (lr=5.2625e-05) (hash(x)=22863518) -656 train 7.487079 (lr=5.2427e-05) (hash(x)=20869581) -657 train 7.649981 (lr=5.2230e-05) (hash(x)=18133606) -658 train 7.532894 (lr=5.2032e-05) (hash(x)=28736122) -659 train 7.510650 (lr=5.1835e-05) (hash(x)=25528329) -660 train 7.575254 (lr=5.1637e-05) (hash(x)=24020765) -661 train 7.608615 (lr=5.1440e-05) (hash(x)=27792245) -662 train 7.624794 (lr=5.1242e-05) (hash(x)=28287635) -663 train 7.358210 (lr=5.1045e-05) (hash(x)=20951206) -664 train 7.445130 (lr=5.0848e-05) (hash(x)=22356452) -665 train 7.516540 (lr=5.0651e-05) (hash(x)=24385170) -666 train 7.592459 (lr=5.0454e-05) (hash(x)=23455854) -667 train 7.571509 (lr=5.0257e-05) (hash(x)=21589405) -668 train 7.553639 (lr=5.0060e-05) (hash(x)=24698150) -669 train 7.531401 (lr=4.9863e-05) (hash(x)=28010836) -670 train 7.657823 (lr=4.9667e-05) (hash(x)=25427381) -671 train 7.617446 (lr=4.9470e-05) (hash(x)=27478557) -672 train 7.556267 (lr=4.9274e-05) (hash(x)=23865569) -673 train 7.630557 (lr=4.9077e-05) (hash(x)=23407132) -674 train 7.486390 (lr=4.8881e-05) (hash(x)=20253664) -675 train 7.456762 (lr=4.8685e-05) (hash(x)=21087928) -676 train 7.571109 (lr=4.8489e-05) (hash(x)=25957589) -677 train 7.497773 (lr=4.8293e-05) (hash(x)=27008610) -678 train 7.566462 (lr=4.8097e-05) (hash(x)=27580265) -679 train 7.608348 (lr=4.7902e-05) (hash(x)=24244798) -680 train 7.478534 (lr=4.7706e-05) (hash(x)=24535871) -681 train 7.584311 (lr=4.7511e-05) (hash(x)=27558681) -682 train 7.467221 (lr=4.7316e-05) (hash(x)=21152793) -683 train 7.441523 (lr=4.7121e-05) (hash(x)=28147005) -684 train 7.536576 (lr=4.6926e-05) (hash(x)=21933180) -685 train 7.504664 (lr=4.6731e-05) (hash(x)=24735895) -686 train 7.555192 (lr=4.6537e-05) (hash(x)=26209948) -687 train 7.538921 (lr=4.6342e-05) (hash(x)=23358313) -688 train 7.475364 (lr=4.6148e-05) (hash(x)=22398001) -689 train 7.383994 (lr=4.5954e-05) (hash(x)=23724957) -690 train 7.434212 (lr=4.5760e-05) (hash(x)=23083514) -691 train 7.312852 (lr=4.5567e-05) (hash(x)=16016924) -692 train 7.591164 (lr=4.5373e-05) (hash(x)=27173674) -693 train 7.392171 (lr=4.5180e-05) (hash(x)=28056250) -694 train 7.498176 (lr=4.4987e-05) (hash(x)=22233617) -695 train 7.591100 (lr=4.4794e-05) (hash(x)=23777908) -696 train 7.566017 (lr=4.4601e-05) (hash(x)=27842849) -697 train 7.549255 (lr=4.4408e-05) (hash(x)=23711679) -698 train 7.398949 (lr=4.4216e-05) (hash(x)=17204843) -699 train 7.702005 (lr=4.4024e-05) (hash(x)=23909800) -700 val loss 7.4818 -700 val perplexity 1775.3721 -700 train 7.556087 (lr=4.3832e-05) (hash(x)=25033286) -701 train 7.843888 (lr=4.3640e-05) (hash(x)=23608371) -702 train 7.637945 (lr=4.3449e-05) (hash(x)=23654625) -703 train 7.452461 (lr=4.3258e-05) (hash(x)=22633313) -704 train 7.515621 (lr=4.3067e-05) (hash(x)=24267011) -705 train 7.466049 (lr=4.2876e-05) (hash(x)=23449416) -706 train 7.712853 (lr=4.2685e-05) (hash(x)=28527906) -707 train 7.405141 (lr=4.2495e-05) (hash(x)=21189075) -708 train 7.472712 (lr=4.2305e-05) (hash(x)=25278645) -709 train 7.998730 (lr=4.2115e-05) (hash(x)=23654290) -710 train 7.454517 (lr=4.1925e-05) (hash(x)=24571139) -711 train 7.553723 (lr=4.1736e-05) (hash(x)=26203172) -712 train 7.482317 (lr=4.1547e-05) (hash(x)=19623503) -713 train 7.469617 (lr=4.1358e-05) (hash(x)=23970545) -714 train 7.649938 (lr=4.1170e-05) (hash(x)=25724900) -715 train 7.563676 (lr=4.0981e-05) (hash(x)=23245421) -716 train 7.509876 (lr=4.0793e-05) (hash(x)=24087537) -717 train 7.436679 (lr=4.0606e-05) (hash(x)=25549646) -718 train 7.509693 (lr=4.0418e-05) (hash(x)=26406563) -719 train 7.550564 (lr=4.0231e-05) (hash(x)=23972429) -720 train 7.605820 (lr=4.0044e-05) (hash(x)=26378040) -721 train 7.458829 (lr=3.9857e-05) (hash(x)=24618531) -722 train 7.473231 (lr=3.9671e-05) (hash(x)=20135714) -723 train 7.614961 (lr=3.9485e-05) (hash(x)=26490981) -724 train 7.627093 (lr=3.9299e-05) (hash(x)=23927567) -725 train 7.468428 (lr=3.9114e-05) (hash(x)=27943609) -726 train 7.522170 (lr=3.8929e-05) (hash(x)=22962561) -727 train 7.559854 (lr=3.8744e-05) (hash(x)=24772257) -728 train 7.483731 (lr=3.8560e-05) (hash(x)=25539761) -729 train 7.647997 (lr=3.8376e-05) (hash(x)=23316262) -730 train 7.415804 (lr=3.8192e-05) (hash(x)=23857558) -731 train 7.597335 (lr=3.8008e-05) (hash(x)=30252690) -732 train 7.665937 (lr=3.7825e-05) (hash(x)=30590162) -733 train 7.552160 (lr=3.7642e-05) (hash(x)=25015721) -734 train 7.633540 (lr=3.7460e-05) (hash(x)=28639679) -735 train 7.505615 (lr=3.7277e-05) (hash(x)=26199058) -736 train 7.564039 (lr=3.7096e-05) (hash(x)=28907461) -737 train 7.546357 (lr=3.6914e-05) (hash(x)=26466915) -738 train 7.621041 (lr=3.6733e-05) (hash(x)=20332613) -739 train 7.506954 (lr=3.6552e-05) (hash(x)=26766470) -740 train 7.450099 (lr=3.6372e-05) (hash(x)=26722145) -741 train 7.500107 (lr=3.6192e-05) (hash(x)=21222489) -742 train 7.634332 (lr=3.6012e-05) (hash(x)=24662435) -743 train 7.442200 (lr=3.5833e-05) (hash(x)=24308927) -744 train 7.402232 (lr=3.5654e-05) (hash(x)=28442502) -745 train 7.444362 (lr=3.5475e-05) (hash(x)=22810435) -746 train 7.578400 (lr=3.5297e-05) (hash(x)=30486518) -747 train 7.559855 (lr=3.5119e-05) (hash(x)=22688712) -748 train 7.473432 (lr=3.4942e-05) (hash(x)=20712962) -749 train 7.337163 (lr=3.4765e-05) (hash(x)=27490086) -750 val loss 7.4596 -750 val perplexity 1736.4541 -750 train 7.334404 (lr=3.4588e-05) (hash(x)=23973908) -751 train 7.488537 (lr=3.4412e-05) (hash(x)=29132401) -752 train 7.490381 (lr=3.4236e-05) (hash(x)=27386700) -753 train 7.518881 (lr=3.4060e-05) (hash(x)=24268614) -754 train 7.570349 (lr=3.3885e-05) (hash(x)=27378387) -755 train 7.373429 (lr=3.3711e-05) (hash(x)=22529400) -756 train 7.452330 (lr=3.3537e-05) (hash(x)=27586989) -757 train 7.381913 (lr=3.3363e-05) (hash(x)=22336349) -758 train 7.420479 (lr=3.3189e-05) (hash(x)=26147854) -759 train 7.513012 (lr=3.3016e-05) (hash(x)=25346788) -760 train 7.601246 (lr=3.2844e-05) (hash(x)=25956855) -761 train 7.531219 (lr=3.2672e-05) (hash(x)=24367780) -762 train 7.424407 (lr=3.2500e-05) (hash(x)=26084462) -763 train 7.501590 (lr=3.2329e-05) (hash(x)=26702459) -764 train 7.437404 (lr=3.2158e-05) (hash(x)=25717976) -765 train 7.375256 (lr=3.1988e-05) (hash(x)=22141366) -766 train 7.605688 (lr=3.1818e-05) (hash(x)=24652423) -767 train 7.527318 (lr=3.1648e-05) (hash(x)=30911699) -768 train 7.515924 (lr=3.1479e-05) (hash(x)=22076954) -769 train 7.697556 (lr=3.1311e-05) (hash(x)=21586769) -770 train 7.514791 (lr=3.1142e-05) (hash(x)=21994337) -771 train 7.473822 (lr=3.0975e-05) (hash(x)=25454145) -772 train 7.521131 (lr=3.0808e-05) (hash(x)=23239550) -773 train 7.570200 (lr=3.0641e-05) (hash(x)=23565926) -774 train 7.426509 (lr=3.0475e-05) (hash(x)=24705563) -775 train 7.541328 (lr=3.0309e-05) (hash(x)=27671095) -776 train 7.654326 (lr=3.0144e-05) (hash(x)=26254303) -777 train 7.393283 (lr=2.9979e-05) (hash(x)=27454020) -778 train 7.391985 (lr=2.9814e-05) (hash(x)=24394565) -779 train 7.349934 (lr=2.9651e-05) (hash(x)=21780199) -780 train 7.433042 (lr=2.9487e-05) (hash(x)=28296896) -781 train 7.569510 (lr=2.9324e-05) (hash(x)=26766303) -782 train 7.406379 (lr=2.9162e-05) (hash(x)=21599973) -783 train 7.731707 (lr=2.9000e-05) (hash(x)=24969360) -784 train 7.696099 (lr=2.8839e-05) (hash(x)=24382006) -785 train 7.370741 (lr=2.8678e-05) (hash(x)=22009299) -786 train 7.222163 (lr=2.8518e-05) (hash(x)=22376569) -787 train 7.393046 (lr=2.8358e-05) (hash(x)=28166843) -788 train 7.669737 (lr=2.8199e-05) (hash(x)=28230119) -789 train 7.611578 (lr=2.8040e-05) (hash(x)=26207465) -790 train 7.688897 (lr=2.7881e-05) (hash(x)=24748118) -791 train 7.610233 (lr=2.7724e-05) (hash(x)=27033802) -792 train 7.528696 (lr=2.7566e-05) (hash(x)=28788047) -793 train 7.397250 (lr=2.7410e-05) (hash(x)=22185099) -794 train 7.595845 (lr=2.7254e-05) (hash(x)=21585136) -795 train 7.406788 (lr=2.7098e-05) (hash(x)=27691968) -796 train 7.453660 (lr=2.6943e-05) (hash(x)=23706103) -797 train 7.289941 (lr=2.6788e-05) (hash(x)=27918030) -798 train 7.509144 (lr=2.6634e-05) (hash(x)=23127886) -799 train 7.527495 (lr=2.6481e-05) (hash(x)=26197070) -800 val loss 7.4417 -800 val perplexity 1705.7286 -800 train 7.574115 (lr=2.6328e-05) (hash(x)=26463005) -801 train 7.469114 (lr=2.6176e-05) (hash(x)=19160134) -802 train 7.417971 (lr=2.6024e-05) (hash(x)=23699136) -803 train 7.488408 (lr=2.5873e-05) (hash(x)=25353051) -804 train 7.415915 (lr=2.5722e-05) (hash(x)=21558484) -805 train 7.333173 (lr=2.5572e-05) (hash(x)=22828380) -806 train 7.464387 (lr=2.5423e-05) (hash(x)=24623811) -807 train 7.636195 (lr=2.5274e-05) (hash(x)=25410299) -808 train 7.492069 (lr=2.5125e-05) (hash(x)=26046091) -809 train 7.611971 (lr=2.4978e-05) (hash(x)=29887482) -810 train 7.349257 (lr=2.4830e-05) (hash(x)=21809706) -811 train 7.472101 (lr=2.4684e-05) (hash(x)=23712872) -812 train 7.361542 (lr=2.4538e-05) (hash(x)=25875103) -813 train 7.333164 (lr=2.4392e-05) (hash(x)=23736441) -814 train 7.424818 (lr=2.4247e-05) (hash(x)=25093883) -815 train 7.363343 (lr=2.4103e-05) (hash(x)=23692325) -816 train 7.262480 (lr=2.3959e-05) (hash(x)=23061082) -817 train 7.393563 (lr=2.3816e-05) (hash(x)=26094253) -818 train 7.550554 (lr=2.3674e-05) (hash(x)=27327886) -819 train 7.422655 (lr=2.3532e-05) (hash(x)=26186143) -820 train 7.419140 (lr=2.3391e-05) (hash(x)=25336043) -821 train 7.460618 (lr=2.3250e-05) (hash(x)=22811468) -822 train 7.441939 (lr=2.3110e-05) (hash(x)=27925831) -823 train 7.546475 (lr=2.2971e-05) (hash(x)=35528108) -824 train 7.537710 (lr=2.2832e-05) (hash(x)=24949421) -825 train 7.297977 (lr=2.2694e-05) (hash(x)=22845735) -826 train 7.411054 (lr=2.2556e-05) (hash(x)=28138281) -827 train 7.482001 (lr=2.2420e-05) (hash(x)=29038360) -828 train 7.425698 (lr=2.2283e-05) (hash(x)=22827903) -829 train 7.514131 (lr=2.2148e-05) (hash(x)=24647015) -830 train 7.399959 (lr=2.2013e-05) (hash(x)=20477804) -831 train 7.367899 (lr=2.1878e-05) (hash(x)=26692640) -832 train 7.446416 (lr=2.1745e-05) (hash(x)=27388794) -833 train 7.734144 (lr=2.1612e-05) (hash(x)=25265875) -834 train 7.370404 (lr=2.1479e-05) (hash(x)=21841888) -835 train 7.475332 (lr=2.1347e-05) (hash(x)=19809196) -836 train 7.534215 (lr=2.1216e-05) (hash(x)=23007945) -837 train 7.462128 (lr=2.1086e-05) (hash(x)=24079409) -838 train 7.383613 (lr=2.0956e-05) (hash(x)=25874533) -839 train 7.305530 (lr=2.0827e-05) (hash(x)=23660106) -840 train 7.794976 (lr=2.0698e-05) (hash(x)=41821329) -841 train 7.407612 (lr=2.0570e-05) (hash(x)=25763899) -842 train 7.245990 (lr=2.0443e-05) (hash(x)=28584111) -843 train 7.734368 (lr=2.0317e-05) (hash(x)=25559887) -844 train 7.494929 (lr=2.0191e-05) (hash(x)=26886377) -845 train 7.471811 (lr=2.0066e-05) (hash(x)=26461895) -846 train 7.517651 (lr=1.9941e-05) (hash(x)=23018827) -847 train 7.321105 (lr=1.9818e-05) (hash(x)=24375035) -848 train 7.491956 (lr=1.9694e-05) (hash(x)=26431670) -849 train 7.375950 (lr=1.9572e-05) (hash(x)=25033511) -850 val loss 7.4396 -850 val perplexity 1702.0132 -850 train 7.488484 (lr=1.9450e-05) (hash(x)=25602577) -851 train 7.400078 (lr=1.9329e-05) (hash(x)=23316582) -852 train 7.535356 (lr=1.9209e-05) (hash(x)=22470095) -853 train 7.406911 (lr=1.9089e-05) (hash(x)=25089202) -854 train 7.337296 (lr=1.8970e-05) (hash(x)=26243247) -855 train 7.435785 (lr=1.8852e-05) (hash(x)=22231144) -856 train 7.346595 (lr=1.8734e-05) (hash(x)=27646564) -857 train 7.491600 (lr=1.8618e-05) (hash(x)=23889875) -858 train 7.445905 (lr=1.8501e-05) (hash(x)=25255240) -859 train 7.527608 (lr=1.8386e-05) (hash(x)=23883299) -860 train 7.501969 (lr=1.8271e-05) (hash(x)=18463841) -861 train 7.533608 (lr=1.8157e-05) (hash(x)=30657762) -862 train 7.416885 (lr=1.8044e-05) (hash(x)=25288079) -863 train 7.461645 (lr=1.7931e-05) (hash(x)=28979310) -864 train 7.647059 (lr=1.7819e-05) (hash(x)=30254991) -865 train 7.492545 (lr=1.7708e-05) (hash(x)=27274653) -866 train 7.472941 (lr=1.7598e-05) (hash(x)=23026066) -867 train 7.428191 (lr=1.7488e-05) (hash(x)=14001963) -868 train 7.326510 (lr=1.7379e-05) (hash(x)=21988054) -869 train 7.420466 (lr=1.7271e-05) (hash(x)=26241099) -870 train 7.513637 (lr=1.7163e-05) (hash(x)=26209722) -871 train 7.371310 (lr=1.7056e-05) (hash(x)=26928312) -872 train 7.457225 (lr=1.6950e-05) (hash(x)=23563272) -873 train 7.384061 (lr=1.6845e-05) (hash(x)=25704828) -874 train 7.650157 (lr=1.6740e-05) (hash(x)=22498992) -875 train 7.784369 (lr=1.6636e-05) (hash(x)=22895918) -876 train 7.341893 (lr=1.6533e-05) (hash(x)=24404283) -877 train 7.376393 (lr=1.6431e-05) (hash(x)=26663932) -878 train 7.514564 (lr=1.6329e-05) (hash(x)=22805316) -879 train 7.509393 (lr=1.6228e-05) (hash(x)=23344466) -880 train 7.453814 (lr=1.6128e-05) (hash(x)=20966940) -881 train 7.451263 (lr=1.6029e-05) (hash(x)=27422664) -882 train 7.477421 (lr=1.5930e-05) (hash(x)=22116510) -883 train 7.340514 (lr=1.5832e-05) (hash(x)=27269633) -884 train 7.450755 (lr=1.5735e-05) (hash(x)=27157711) -885 train 7.491467 (lr=1.5639e-05) (hash(x)=22774645) -886 train 7.409890 (lr=1.5543e-05) (hash(x)=22143146) -887 train 7.402446 (lr=1.5449e-05) (hash(x)=22356438) -888 train 7.170747 (lr=1.5354e-05) (hash(x)=23172158) -889 train 7.429227 (lr=1.5261e-05) (hash(x)=26971690) -890 train 7.466310 (lr=1.5169e-05) (hash(x)=24851922) -891 train 7.281016 (lr=1.5077e-05) (hash(x)=21017641) -892 train 7.436307 (lr=1.4986e-05) (hash(x)=28186957) -893 train 7.455210 (lr=1.4896e-05) (hash(x)=25661082) -894 train 7.351072 (lr=1.4806e-05) (hash(x)=25462781) -895 train 7.305271 (lr=1.4718e-05) (hash(x)=21774374) -896 train 7.305682 (lr=1.4630e-05) (hash(x)=23657055) -897 train 7.333868 (lr=1.4543e-05) (hash(x)=23604050) -898 train 7.771733 (lr=1.4456e-05) (hash(x)=31534071) -899 train 7.578480 (lr=1.4371e-05) (hash(x)=24183732) -900 val loss 7.4296 -900 val perplexity 1685.2150 -900 train 7.407505 (lr=1.4286e-05) (hash(x)=23384280) -901 train 7.293602 (lr=1.4202e-05) (hash(x)=21975928) -902 train 7.294423 (lr=1.4119e-05) (hash(x)=25281674) -903 train 7.279311 (lr=1.4037e-05) (hash(x)=22947712) -904 train 7.354813 (lr=1.3955e-05) (hash(x)=24853977) -905 train 7.403245 (lr=1.3874e-05) (hash(x)=23696470) -906 train 7.413362 (lr=1.3794e-05) (hash(x)=27489711) -907 train 7.342287 (lr=1.3715e-05) (hash(x)=28050163) -908 train 7.214733 (lr=1.3637e-05) (hash(x)=26996807) -909 train 7.621967 (lr=1.3559e-05) (hash(x)=28626495) -910 train 7.336179 (lr=1.3482e-05) (hash(x)=24783242) -911 train 7.337069 (lr=1.3407e-05) (hash(x)=25832479) -912 train 7.224257 (lr=1.3331e-05) (hash(x)=23522137) -913 train 7.514649 (lr=1.3257e-05) (hash(x)=26731973) -914 train 7.595162 (lr=1.3183e-05) (hash(x)=26019142) -915 train 7.370749 (lr=1.3111e-05) (hash(x)=25081877) -916 train 7.487756 (lr=1.3039e-05) (hash(x)=23843571) -917 train 7.493926 (lr=1.2968e-05) (hash(x)=26805739) -918 train 7.396144 (lr=1.2897e-05) (hash(x)=27169927) -919 train 7.439053 (lr=1.2828e-05) (hash(x)=26545491) -920 train 7.450454 (lr=1.2759e-05) (hash(x)=25179497) -921 train 7.389654 (lr=1.2691e-05) (hash(x)=27085003) -922 train 7.410111 (lr=1.2624e-05) (hash(x)=24558202) -923 train 7.564204 (lr=1.2558e-05) (hash(x)=25317684) -924 train 7.422526 (lr=1.2493e-05) (hash(x)=27104868) -925 train 7.461666 (lr=1.2428e-05) (hash(x)=24806274) -926 train 7.518903 (lr=1.2364e-05) (hash(x)=24341937) -927 train 7.316999 (lr=1.2301e-05) (hash(x)=22736558) -928 train 7.333353 (lr=1.2239e-05) (hash(x)=23843744) -929 train 7.460651 (lr=1.2178e-05) (hash(x)=27218468) -930 train 7.399772 (lr=1.2118e-05) (hash(x)=26042053) -931 train 7.419421 (lr=1.2058e-05) (hash(x)=25776082) -932 train 7.299749 (lr=1.1999e-05) (hash(x)=21594447) -933 train 7.338398 (lr=1.1941e-05) (hash(x)=21674469) -934 train 7.281021 (lr=1.1884e-05) (hash(x)=26213966) -935 train 7.290705 (lr=1.1828e-05) (hash(x)=23770702) -936 train 7.407078 (lr=1.1772e-05) (hash(x)=27829878) -937 train 7.501351 (lr=1.1718e-05) (hash(x)=29177278) -938 train 7.426019 (lr=1.1664e-05) (hash(x)=26020089) -939 train 7.322884 (lr=1.1611e-05) (hash(x)=24676475) -940 train 7.393018 (lr=1.1559e-05) (hash(x)=24056174) -941 train 7.414608 (lr=1.1508e-05) (hash(x)=21887514) -942 train 7.415479 (lr=1.1457e-05) (hash(x)=24196100) -943 train 7.372238 (lr=1.1408e-05) (hash(x)=25700093) -944 train 7.545765 (lr=1.1359e-05) (hash(x)=24365834) -945 train 7.332805 (lr=1.1311e-05) (hash(x)=25631786) -946 train 7.464033 (lr=1.1264e-05) (hash(x)=24452345) -947 train 7.171823 (lr=1.1218e-05) (hash(x)=21465330) -948 train 7.298168 (lr=1.1173e-05) (hash(x)=20427493) -949 train 7.335891 (lr=1.1128e-05) (hash(x)=25439720) -950 val loss 7.4199 -950 val perplexity 1668.8562 -950 train 7.398656 (lr=1.1085e-05) (hash(x)=21652988) -951 train 7.339907 (lr=1.1042e-05) (hash(x)=22546779) -952 train 7.209517 (lr=1.1000e-05) (hash(x)=20855736) -953 train 7.397276 (lr=1.0959e-05) (hash(x)=27725873) -954 train 7.464233 (lr=1.0919e-05) (hash(x)=24582989) -955 train 7.338234 (lr=1.0879e-05) (hash(x)=24822372) -956 train 7.390271 (lr=1.0841e-05) (hash(x)=25624588) -957 train 7.347393 (lr=1.0803e-05) (hash(x)=23566489) -958 train 7.524933 (lr=1.0766e-05) (hash(x)=28117387) -959 train 7.290871 (lr=1.0730e-05) (hash(x)=26874283) -960 train 7.297306 (lr=1.0695e-05) (hash(x)=28570123) -961 train 7.451760 (lr=1.0661e-05) (hash(x)=24807017) -962 train 7.323246 (lr=1.0628e-05) (hash(x)=25189993) -963 train 7.455842 (lr=1.0595e-05) (hash(x)=29202226) -964 train 7.645636 (lr=1.0563e-05) (hash(x)=26060713) -965 train 7.524036 (lr=1.0533e-05) (hash(x)=23053285) -966 train 7.500039 (lr=1.0503e-05) (hash(x)=21968481) -967 train 7.385129 (lr=1.0474e-05) (hash(x)=24492698) -968 train 7.363263 (lr=1.0445e-05) (hash(x)=25760806) -969 train 7.384084 (lr=1.0418e-05) (hash(x)=27417392) -970 train 7.385349 (lr=1.0391e-05) (hash(x)=26767105) -971 train 7.450591 (lr=1.0366e-05) (hash(x)=25117678) -972 train 7.330565 (lr=1.0341e-05) (hash(x)=24233758) -973 train 7.392282 (lr=1.0317e-05) (hash(x)=29907634) -974 train 7.174641 (lr=1.0294e-05) (hash(x)=23552436) -975 train 7.544744 (lr=1.0272e-05) (hash(x)=28610406) -976 train 7.364544 (lr=1.0251e-05) (hash(x)=23858212) -977 train 7.265794 (lr=1.0230e-05) (hash(x)=20567764) -978 train 7.275741 (lr=1.0211e-05) (hash(x)=29631106) -979 train 7.253947 (lr=1.0192e-05) (hash(x)=19843564) -980 train 7.392068 (lr=1.0174e-05) (hash(x)=22589891) -981 train 7.177075 (lr=1.0157e-05) (hash(x)=21183943) -982 train 7.337965 (lr=1.0141e-05) (hash(x)=26152241) -983 train 7.350290 (lr=1.0126e-05) (hash(x)=22700785) -984 train 7.585400 (lr=1.0111e-05) (hash(x)=29611280) -985 train 7.703375 (lr=1.0098e-05) (hash(x)=25972823) -986 train 7.394578 (lr=1.0085e-05) (hash(x)=23447782) -987 train 7.338687 (lr=1.0074e-05) (hash(x)=22578453) -988 train 7.334603 (lr=1.0063e-05) (hash(x)=25474622) -989 train 7.404408 (lr=1.0053e-05) (hash(x)=22965762) -990 train 7.445079 (lr=1.0044e-05) (hash(x)=26116321) -991 train 7.362096 (lr=1.0035e-05) (hash(x)=25934549) -992 train 7.276827 (lr=1.0028e-05) (hash(x)=23937495) -993 train 7.459218 (lr=1.0021e-05) (hash(x)=26286442) -994 train 7.714988 (lr=1.0016e-05) (hash(x)=32737711) -995 train 7.286125 (lr=1.0011e-05) (hash(x)=20563613) -996 train 7.547499 (lr=1.0007e-05) (hash(x)=26234770) -997 train 7.480315 (lr=1.0004e-05) (hash(x)=26179422) -998 train 7.395674 (lr=1.0002e-05) (hash(x)=23289651) -999 val loss 7.4181 -999 val perplexity 1665.8024 -999 train 7.777097 (lr=1.0000e-05) (hash(x)=36436617) +9 train 11.743742 (lr=3.4965e-06) (hash(x)=27654289) +10 train 11.729365 (lr=3.8462e-06) (hash(x)=26559876) +11 train 11.734584 (lr=4.1958e-06) (hash(x)=21770329) +12 train 11.707083 (lr=4.5455e-06) (hash(x)=22723459) +13 train 11.702140 (lr=4.8951e-06) (hash(x)=30338342) +14 train 11.689697 (lr=5.2448e-06) (hash(x)=25678059) +15 train 11.683121 (lr=5.5944e-06) (hash(x)=26109335) +16 train 11.672155 (lr=5.9441e-06) (hash(x)=25044885) +17 train 11.675049 (lr=6.2937e-06) (hash(x)=28870690) +18 train 11.651052 (lr=6.6434e-06) (hash(x)=35279941) +19 train 11.634657 (lr=6.9930e-06) (hash(x)=25015923) +20 train 11.628777 (lr=7.3427e-06) (hash(x)=26740855) +21 train 11.602887 (lr=7.6923e-06) (hash(x)=22262151) +22 train 11.583412 (lr=8.0420e-06) (hash(x)=26621834) +23 train 11.563446 (lr=8.3916e-06) (hash(x)=23234741) +24 train 11.570676 (lr=8.7413e-06) (hash(x)=25470563) +25 train 11.539077 (lr=9.0909e-06) (hash(x)=21799102) +26 train 11.525577 (lr=9.4406e-06) (hash(x)=23275479) +27 train 11.506817 (lr=9.7902e-06) (hash(x)=22091862) +28 train 11.476108 (lr=1.0140e-05) (hash(x)=17430373) +29 train 11.449938 (lr=1.0490e-05) (hash(x)=22954861) +30 train 11.437654 (lr=1.0839e-05) (hash(x)=26671799) +31 train 11.417167 (lr=1.1189e-05) (hash(x)=26924723) +32 train 11.396602 (lr=1.1538e-05) (hash(x)=23557495) +33 train 11.348690 (lr=1.1888e-05) (hash(x)=27804380) +34 train 11.352073 (lr=1.2238e-05) (hash(x)=26654908) +35 train 11.308615 (lr=1.2587e-05) (hash(x)=23627518) +36 train 11.280223 (lr=1.2937e-05) (hash(x)=21889990) +37 train 11.249405 (lr=1.3287e-05) (hash(x)=22528001) +38 train 11.253233 (lr=1.3636e-05) (hash(x)=29248942) +39 train 11.171371 (lr=1.3986e-05) (hash(x)=25533417) +40 train 11.126245 (lr=1.4336e-05) (hash(x)=20612533) +41 train 11.120057 (lr=1.4685e-05) (hash(x)=24677740) +42 train 11.103706 (lr=1.5035e-05) (hash(x)=25347203) +43 train 11.047005 (lr=1.5385e-05) (hash(x)=28922363) +44 train 11.018970 (lr=1.5734e-05) (hash(x)=25631397) +45 train 10.983917 (lr=1.6084e-05) (hash(x)=26237983) +46 train 10.948134 (lr=1.6434e-05) (hash(x)=24742645) +47 train 10.933228 (lr=1.6783e-05) (hash(x)=26157660) +48 train 10.860299 (lr=1.7133e-05) (hash(x)=22582169) +49 train 10.839372 (lr=1.7483e-05) (hash(x)=26431960) +50 val loss 10.8336 +50 val perplexity 50694.5820 +50 train 10.838658 (lr=1.7832e-05) (hash(x)=25716176) +51 train 10.797357 (lr=1.8182e-05) (hash(x)=28019788) +52 train 10.777383 (lr=1.8531e-05) (hash(x)=20382963) +53 train 10.732782 (lr=1.8881e-05) (hash(x)=24489647) +54 train 10.716461 (lr=1.9231e-05) (hash(x)=25128502) +55 train 10.691587 (lr=1.9580e-05) (hash(x)=25231390) +56 train 10.612194 (lr=1.9930e-05) (hash(x)=22708977) +57 train 10.675596 (lr=2.0280e-05) (hash(x)=24364920) +58 train 10.635049 (lr=2.0629e-05) (hash(x)=24932925) +59 train 10.613993 (lr=2.0979e-05) (hash(x)=26913684) +60 train 10.531951 (lr=2.1329e-05) (hash(x)=18017792) +61 train 10.562325 (lr=2.1678e-05) (hash(x)=28613882) +62 train 10.560807 (lr=2.2028e-05) (hash(x)=22652243) +63 train 10.499845 (lr=2.2378e-05) (hash(x)=27194521) +64 train 10.500281 (lr=2.2727e-05) (hash(x)=24936836) +65 train 10.515687 (lr=2.3077e-05) (hash(x)=24765578) +66 train 10.541842 (lr=2.3427e-05) (hash(x)=23107416) +67 train 10.439338 (lr=2.3776e-05) (hash(x)=29314255) +68 train 10.427672 (lr=2.4126e-05) (hash(x)=24013079) +69 train 10.401395 (lr=2.4476e-05) (hash(x)=25322001) +70 train 10.500172 (lr=2.4825e-05) (hash(x)=26690440) +71 train 10.439803 (lr=2.5175e-05) (hash(x)=22598512) +72 train 10.414017 (lr=2.5524e-05) (hash(x)=18473911) +73 train 10.465598 (lr=2.5874e-05) (hash(x)=22322442) +74 train 10.444426 (lr=2.6224e-05) (hash(x)=26324153) +75 train 10.417928 (lr=2.6573e-05) (hash(x)=23446058) +76 train 10.463191 (lr=2.6923e-05) (hash(x)=24843741) +77 train 10.425342 (lr=2.7273e-05) (hash(x)=23197102) +78 train 10.501419 (lr=2.7622e-05) (hash(x)=23054940) +79 train 10.438265 (lr=2.7972e-05) (hash(x)=27140757) +80 train 10.330658 (lr=2.8322e-05) (hash(x)=25149419) +81 train 10.402567 (lr=2.8671e-05) (hash(x)=24727764) +82 train 10.275390 (lr=2.9021e-05) (hash(x)=21935931) +83 train 10.368217 (lr=2.9371e-05) (hash(x)=27606073) +84 train 10.394423 (lr=2.9720e-05) (hash(x)=22450341) +85 train 10.398794 (lr=3.0070e-05) (hash(x)=25945859) +86 train 10.353902 (lr=3.0420e-05) (hash(x)=21613707) +87 train 10.366550 (lr=3.0769e-05) (hash(x)=24396519) +88 train 10.361097 (lr=3.1119e-05) (hash(x)=29138828) +89 train 10.432784 (lr=3.1469e-05) (hash(x)=24685515) +90 train 10.327359 (lr=3.1818e-05) (hash(x)=22825749) +91 train 10.354750 (lr=3.2168e-05) (hash(x)=28767869) +92 train 10.338910 (lr=3.2517e-05) (hash(x)=21159060) +93 train 10.320039 (lr=3.2867e-05) (hash(x)=25045397) +94 train 10.397500 (lr=3.3217e-05) (hash(x)=25560928) +95 train 10.356876 (lr=3.3566e-05) (hash(x)=28959222) +96 train 10.307255 (lr=3.3916e-05) (hash(x)=21248405) +97 train 10.316563 (lr=3.4266e-05) (hash(x)=25830182) +98 train 10.263371 (lr=3.4615e-05) (hash(x)=25173113) +99 train 10.314532 (lr=3.4965e-05) (hash(x)=23486277) +100 val loss 10.3125 +100 val perplexity 30107.5000 +100 train 10.279247 (lr=3.5315e-05) (hash(x)=28916006) +101 train 10.347860 (lr=3.5664e-05) (hash(x)=25556929) +102 train 10.303473 (lr=3.6014e-05) (hash(x)=29648798) +103 train 10.256695 (lr=3.6364e-05) (hash(x)=24976217) +104 train 10.276460 (lr=3.6713e-05) (hash(x)=31019606) +105 train 10.283141 (lr=3.7063e-05) (hash(x)=28269421) +106 train 10.251169 (lr=3.7413e-05) (hash(x)=24865358) +107 train 10.157942 (lr=3.7762e-05) (hash(x)=26984272) +108 train 10.254987 (lr=3.8112e-05) (hash(x)=24544116) +109 train 10.238841 (lr=3.8462e-05) (hash(x)=22059850) +110 train 10.225416 (lr=3.8811e-05) (hash(x)=23926632) +111 train 10.197513 (lr=3.9161e-05) (hash(x)=27025333) +112 train 10.183633 (lr=3.9510e-05) (hash(x)=29120407) +113 train 10.227939 (lr=3.9860e-05) (hash(x)=23827393) +114 train 10.157598 (lr=4.0210e-05) (hash(x)=27230027) +115 train 10.168883 (lr=4.0559e-05) (hash(x)=24968260) +116 train 10.204412 (lr=4.0909e-05) (hash(x)=27544803) +117 train 10.179665 (lr=4.1259e-05) (hash(x)=26540663) +118 train 10.204772 (lr=4.1608e-05) (hash(x)=25856625) +119 train 10.153449 (lr=4.1958e-05) (hash(x)=23164356) +120 train 10.178568 (lr=4.2308e-05) (hash(x)=25646282) +121 train 10.195406 (lr=4.2657e-05) (hash(x)=23843233) +122 train 10.094124 (lr=4.3007e-05) (hash(x)=24506028) +123 train 10.132697 (lr=4.3357e-05) (hash(x)=22781277) +124 train 10.154956 (lr=4.3706e-05) (hash(x)=24948650) +125 train 10.109292 (lr=4.4056e-05) (hash(x)=26373991) +126 train 10.152582 (lr=4.4406e-05) (hash(x)=23402396) +127 train 10.111260 (lr=4.4755e-05) (hash(x)=21562133) +128 train 10.155614 (lr=4.5105e-05) (hash(x)=25009210) +129 train 10.050844 (lr=4.5455e-05) (hash(x)=27161811) +130 train 10.054814 (lr=4.5804e-05) (hash(x)=29515100) +131 train 10.045181 (lr=4.6154e-05) (hash(x)=25305929) +132 train 10.056855 (lr=4.6503e-05) (hash(x)=26808543) +133 train 10.018795 (lr=4.6853e-05) (hash(x)=25669509) +134 train 10.037328 (lr=4.7203e-05) (hash(x)=28435805) +135 train 10.033699 (lr=4.7552e-05) (hash(x)=26482758) +136 train 9.980275 (lr=4.7902e-05) (hash(x)=24929929) +137 train 10.035925 (lr=4.8252e-05) (hash(x)=25043037) +138 train 10.046122 (lr=4.8601e-05) (hash(x)=29151257) +139 train 10.005488 (lr=4.8951e-05) (hash(x)=26400365) +140 train 9.933040 (lr=4.9301e-05) (hash(x)=24098812) +141 train 9.986092 (lr=4.9650e-05) (hash(x)=20967666) +142 train 10.007580 (lr=5.0000e-05) (hash(x)=28667963) +143 train 9.980922 (lr=5.0350e-05) (hash(x)=27083417) +144 train 9.945218 (lr=5.0699e-05) (hash(x)=23649001) +145 train 9.928185 (lr=5.1049e-05) (hash(x)=31338300) +146 train 9.882592 (lr=5.1399e-05) (hash(x)=30603174) +147 train 9.879318 (lr=5.1748e-05) (hash(x)=24830965) +148 train 9.913800 (lr=5.2098e-05) (hash(x)=27269893) +149 train 9.878550 (lr=5.2448e-05) (hash(x)=24825041) +150 val loss 9.8636 +150 val perplexity 19217.0664 +150 train 9.852741 (lr=5.2797e-05) (hash(x)=22724871) +151 train 9.792242 (lr=5.3147e-05) (hash(x)=24683717) +152 train 9.836679 (lr=5.3497e-05) (hash(x)=26190337) +153 train 9.818953 (lr=5.3846e-05) (hash(x)=23086289) +154 train 9.847963 (lr=5.4196e-05) (hash(x)=23593875) +155 train 9.805938 (lr=5.4545e-05) (hash(x)=22414190) +156 train 9.797126 (lr=5.4895e-05) (hash(x)=26758114) +157 train 9.761087 (lr=5.5245e-05) (hash(x)=21563184) +158 train 9.772513 (lr=5.5594e-05) (hash(x)=19218470) +159 train 9.788898 (lr=5.5944e-05) (hash(x)=25588469) +160 train 9.733636 (lr=5.6294e-05) (hash(x)=30433767) +161 train 9.720930 (lr=5.6643e-05) (hash(x)=17851621) +162 train 9.758940 (lr=5.6993e-05) (hash(x)=25183690) +163 train 9.787950 (lr=5.7343e-05) (hash(x)=27040642) +164 train 9.658743 (lr=5.7692e-05) (hash(x)=22183303) +165 train 9.796076 (lr=5.8042e-05) (hash(x)=27537063) +166 train 9.657582 (lr=5.8392e-05) (hash(x)=19492832) +167 train 9.618973 (lr=5.8741e-05) (hash(x)=25951629) +168 train 9.588369 (lr=5.9091e-05) (hash(x)=22408682) +169 train 9.645965 (lr=5.9441e-05) (hash(x)=27633457) +170 train 9.639729 (lr=5.9790e-05) (hash(x)=27635080) +171 train 9.637705 (lr=6.0140e-05) (hash(x)=22287596) +172 train 9.648672 (lr=6.0490e-05) (hash(x)=35895440) +173 train 9.626386 (lr=6.0839e-05) (hash(x)=25888847) +174 train 9.547156 (lr=6.1189e-05) (hash(x)=26518800) +175 train 9.592853 (lr=6.1538e-05) (hash(x)=37519283) +176 train 9.636898 (lr=6.1888e-05) (hash(x)=25141945) +177 train 9.659312 (lr=6.2238e-05) (hash(x)=28527939) +178 train 9.542221 (lr=6.2587e-05) (hash(x)=21942471) +179 train 9.562410 (lr=6.2937e-05) (hash(x)=24955630) +180 train 9.497692 (lr=6.3287e-05) (hash(x)=25541230) +181 train 9.429600 (lr=6.3636e-05) (hash(x)=23501481) +182 train 9.452502 (lr=6.3986e-05) (hash(x)=22754988) +183 train 9.389763 (lr=6.4336e-05) (hash(x)=23475070) +184 train 9.463747 (lr=6.4685e-05) (hash(x)=23609959) +185 train 9.430327 (lr=6.5035e-05) (hash(x)=23441847) +186 train 9.422603 (lr=6.5385e-05) (hash(x)=23096292) +187 train 9.430068 (lr=6.5734e-05) (hash(x)=26171093) +188 train 9.439085 (lr=6.6084e-05) (hash(x)=20060838) +189 train 9.360768 (lr=6.6434e-05) (hash(x)=25451055) +190 train 9.266185 (lr=6.6783e-05) (hash(x)=24934792) +191 train 9.295485 (lr=6.7133e-05) (hash(x)=25367597) +192 train 9.363704 (lr=6.7483e-05) (hash(x)=25022148) +193 train 9.354053 (lr=6.7832e-05) (hash(x)=25431365) +194 train 9.344863 (lr=6.8182e-05) (hash(x)=29307788) +195 train 9.262985 (lr=6.8531e-05) (hash(x)=25651929) +196 train 9.310635 (lr=6.8881e-05) (hash(x)=26312387) +197 train 9.292030 (lr=6.9231e-05) (hash(x)=23804913) +198 train 9.347159 (lr=6.9580e-05) (hash(x)=26115519) +199 train 9.245741 (lr=6.9930e-05) (hash(x)=23210747) +200 val loss 9.2580 +200 val perplexity 10488.6201 +200 train 9.279895 (lr=7.0280e-05) (hash(x)=18463712) +201 train 9.215055 (lr=7.0629e-05) (hash(x)=23245699) +202 train 9.225723 (lr=7.0979e-05) (hash(x)=21770300) +203 train 9.378979 (lr=7.1329e-05) (hash(x)=25387532) +204 train 9.224954 (lr=7.1678e-05) (hash(x)=27718703) +205 train 9.307646 (lr=7.2028e-05) (hash(x)=27661052) +206 train 9.181190 (lr=7.2378e-05) (hash(x)=26538928) +207 train 9.198976 (lr=7.2727e-05) (hash(x)=21772654) +208 train 9.200007 (lr=7.3077e-05) (hash(x)=26263254) +209 train 9.184479 (lr=7.3427e-05) (hash(x)=23552356) +210 train 9.266952 (lr=7.3776e-05) (hash(x)=22707141) +211 train 9.264257 (lr=7.4126e-05) (hash(x)=34234481) +212 train 9.154451 (lr=7.4476e-05) (hash(x)=35008670) +213 train 9.162195 (lr=7.4825e-05) (hash(x)=25891543) +214 train 9.064454 (lr=7.5175e-05) (hash(x)=21971224) +215 train 9.091294 (lr=7.5524e-05) (hash(x)=27312186) +216 train 9.136991 (lr=7.5874e-05) (hash(x)=22163749) +217 train 8.972001 (lr=7.6224e-05) (hash(x)=25942888) +218 train 9.068890 (lr=7.6573e-05) (hash(x)=25493916) +219 train 9.166369 (lr=7.6923e-05) (hash(x)=24289951) +220 train 9.022631 (lr=7.7273e-05) (hash(x)=23648671) +221 train 8.999970 (lr=7.7622e-05) (hash(x)=24992055) +222 train 9.077555 (lr=7.7972e-05) (hash(x)=26005549) +223 train 8.998693 (lr=7.8322e-05) (hash(x)=21046554) +224 train 8.974440 (lr=7.8671e-05) (hash(x)=26321539) +225 train 8.989400 (lr=7.9021e-05) (hash(x)=27807872) +226 train 9.006127 (lr=7.9371e-05) (hash(x)=21555364) +227 train 8.986025 (lr=7.9720e-05) (hash(x)=26325324) +228 train 8.940698 (lr=8.0070e-05) (hash(x)=24956294) +229 train 8.942142 (lr=8.0420e-05) (hash(x)=24429389) +230 train 8.974105 (lr=8.0769e-05) (hash(x)=27929846) +231 train 8.942498 (lr=8.1119e-05) (hash(x)=25282897) +232 train 8.857235 (lr=8.1469e-05) (hash(x)=26802508) +233 train 8.916639 (lr=8.1818e-05) (hash(x)=28390416) +234 train 8.982015 (lr=8.2168e-05) (hash(x)=26292451) +235 train 8.944869 (lr=8.2517e-05) (hash(x)=22915344) +236 train 8.878548 (lr=8.2867e-05) (hash(x)=24744383) +237 train 8.785595 (lr=8.3217e-05) (hash(x)=26284678) +238 train 8.809217 (lr=8.3566e-05) (hash(x)=21654269) +239 train 9.069222 (lr=8.3916e-05) (hash(x)=29466268) +240 train 9.226203 (lr=8.4266e-05) (hash(x)=32616762) +241 train 9.191886 (lr=8.4615e-05) (hash(x)=32201138) +242 train 9.177925 (lr=8.4965e-05) (hash(x)=33683468) +243 train 8.931837 (lr=8.5315e-05) (hash(x)=27498125) +244 train 8.831232 (lr=8.5664e-05) (hash(x)=25164318) +245 train 8.799298 (lr=8.6014e-05) (hash(x)=25690221) +246 train 8.807261 (lr=8.6364e-05) (hash(x)=29031550) +247 train 8.834379 (lr=8.6713e-05) (hash(x)=24094093) +248 train 9.099588 (lr=8.7063e-05) (hash(x)=61455960) +249 train 10.771669 (lr=8.7413e-05) (hash(x)=61104710) +250 val loss 8.7522 +250 val perplexity 6324.4492 +250 train 8.755510 (lr=8.7762e-05) (hash(x)=20881126) +251 train 8.795639 (lr=8.8112e-05) (hash(x)=26920243) +252 train 8.691395 (lr=8.8462e-05) (hash(x)=21970343) +253 train 8.710552 (lr=8.8811e-05) (hash(x)=31676352) +254 train 8.714910 (lr=8.9161e-05) (hash(x)=25600562) +255 train 8.714534 (lr=8.9510e-05) (hash(x)=22022924) +256 train 8.666109 (lr=8.9860e-05) (hash(x)=22777085) +257 train 8.741806 (lr=9.0210e-05) (hash(x)=21727744) +258 train 8.761035 (lr=9.0559e-05) (hash(x)=20701998) +259 train 8.735220 (lr=9.0909e-05) (hash(x)=20863982) +260 train 8.662899 (lr=9.1259e-05) (hash(x)=25084926) +261 train 8.658141 (lr=9.1608e-05) (hash(x)=24412897) +262 train 8.592793 (lr=9.1958e-05) (hash(x)=25434913) +263 train 8.625324 (lr=9.2308e-05) (hash(x)=26302544) +264 train 8.662353 (lr=9.2657e-05) (hash(x)=34976197) +265 train 8.590086 (lr=9.3007e-05) (hash(x)=23870805) +266 train 8.592322 (lr=9.3357e-05) (hash(x)=25041103) +267 train 8.615456 (lr=9.3706e-05) (hash(x)=26041745) +268 train 8.574131 (lr=9.4056e-05) (hash(x)=24068619) +269 train 8.562845 (lr=9.4406e-05) (hash(x)=25441262) +270 train 8.577830 (lr=9.4755e-05) (hash(x)=29181867) +271 train 8.583162 (lr=9.5105e-05) (hash(x)=27093757) +272 train 8.484027 (lr=9.5455e-05) (hash(x)=25150275) +273 train 8.559868 (lr=9.5804e-05) (hash(x)=33457102) +274 train 8.461896 (lr=9.6154e-05) (hash(x)=25755322) +275 train 8.491273 (lr=9.6503e-05) (hash(x)=23630794) +276 train 8.404172 (lr=9.6853e-05) (hash(x)=25167987) +277 train 8.519456 (lr=9.7203e-05) (hash(x)=24043224) +278 train 8.488585 (lr=9.7552e-05) (hash(x)=24665411) +279 train 8.472032 (lr=9.7902e-05) (hash(x)=17321657) +280 train 8.437852 (lr=9.8252e-05) (hash(x)=27932591) +281 train 8.554876 (lr=9.8601e-05) (hash(x)=29447954) +282 train 8.471408 (lr=9.8951e-05) (hash(x)=24621816) +283 train 8.437234 (lr=9.9301e-05) (hash(x)=24682294) +284 train 8.438483 (lr=9.9650e-05) (hash(x)=28057095) +285 train 8.469583 (lr=1.0000e-04) (hash(x)=24547536) +286 train 8.299122 (lr=1.0000e-04) (hash(x)=18987545) +287 train 8.387889 (lr=1.0000e-04) (hash(x)=26745063) +288 train 8.312629 (lr=9.9998e-05) (hash(x)=21447406) +289 train 8.264685 (lr=9.9996e-05) (hash(x)=24650655) +290 train 8.468912 (lr=9.9993e-05) (hash(x)=26519976) +291 train 8.749260 (lr=9.9989e-05) (hash(x)=20106514) +292 train 8.278829 (lr=9.9984e-05) (hash(x)=20117808) +293 train 8.374424 (lr=9.9979e-05) (hash(x)=25510334) +294 train 8.298206 (lr=9.9972e-05) (hash(x)=25749120) +295 train 8.369887 (lr=9.9965e-05) (hash(x)=24357241) +296 train 8.349216 (lr=9.9956e-05) (hash(x)=25584945) +297 train 8.273057 (lr=9.9947e-05) (hash(x)=25198897) +298 train 8.289836 (lr=9.9937e-05) (hash(x)=23475891) +299 train 8.262544 (lr=9.9926e-05) (hash(x)=23644804) +300 val loss 8.2903 +300 val perplexity 3985.0120 +300 train 8.417461 (lr=9.9915e-05) (hash(x)=25629584) +301 train 8.474089 (lr=9.9902e-05) (hash(x)=29797280) +302 train 8.494260 (lr=9.9889e-05) (hash(x)=30026394) +303 train 8.428800 (lr=9.9874e-05) (hash(x)=26594177) +304 train 8.300892 (lr=9.9859e-05) (hash(x)=23766256) +305 train 8.252919 (lr=9.9843e-05) (hash(x)=24250150) +306 train 8.248328 (lr=9.9826e-05) (hash(x)=26108817) +307 train 8.202334 (lr=9.9808e-05) (hash(x)=18582688) +308 train 8.174821 (lr=9.9789e-05) (hash(x)=23272705) +309 train 8.151085 (lr=9.9770e-05) (hash(x)=24793412) +310 train 8.172600 (lr=9.9749e-05) (hash(x)=19280989) +311 train 8.164104 (lr=9.9728e-05) (hash(x)=22751150) +312 train 8.257322 (lr=9.9706e-05) (hash(x)=23537942) +313 train 8.111483 (lr=9.9683e-05) (hash(x)=23748375) +314 train 8.204211 (lr=9.9659e-05) (hash(x)=27413825) +315 train 8.212543 (lr=9.9634e-05) (hash(x)=21682445) +316 train 8.144847 (lr=9.9609e-05) (hash(x)=22040184) +317 train 8.089615 (lr=9.9582e-05) (hash(x)=24463229) +318 train 8.204624 (lr=9.9555e-05) (hash(x)=24091954) +319 train 8.131656 (lr=9.9526e-05) (hash(x)=26670219) +320 train 8.112108 (lr=9.9497e-05) (hash(x)=25558986) +321 train 8.080044 (lr=9.9467e-05) (hash(x)=19218943) +322 train 8.170919 (lr=9.9437e-05) (hash(x)=24551402) +323 train 8.064628 (lr=9.9405e-05) (hash(x)=25229281) +324 train 7.909699 (lr=9.9372e-05) (hash(x)=21251127) +325 train 8.086199 (lr=9.9339e-05) (hash(x)=25800857) +326 train 8.013880 (lr=9.9305e-05) (hash(x)=25836189) +327 train 8.033986 (lr=9.9270e-05) (hash(x)=25999255) +328 train 8.077474 (lr=9.9234e-05) (hash(x)=20149394) +329 train 8.029878 (lr=9.9197e-05) (hash(x)=21999338) +330 train 8.102167 (lr=9.9159e-05) (hash(x)=27611302) +331 train 8.062190 (lr=9.9121e-05) (hash(x)=22778356) +332 train 8.025073 (lr=9.9081e-05) (hash(x)=24195959) +333 train 8.237892 (lr=9.9041e-05) (hash(x)=19127465) +334 train 8.100348 (lr=9.9000e-05) (hash(x)=22426274) +335 train 7.960400 (lr=9.8958e-05) (hash(x)=23785205) +336 train 8.068583 (lr=9.8915e-05) (hash(x)=28505676) +337 train 8.010908 (lr=9.8872e-05) (hash(x)=27706477) +338 train 8.067745 (lr=9.8827e-05) (hash(x)=28684175) +339 train 8.039546 (lr=9.8782e-05) (hash(x)=28003600) +340 train 8.049531 (lr=9.8736e-05) (hash(x)=27586845) +341 train 7.989234 (lr=9.8689e-05) (hash(x)=21176405) +342 train 7.911770 (lr=9.8641e-05) (hash(x)=25888433) +343 train 8.051623 (lr=9.8592e-05) (hash(x)=27916982) +344 train 7.913222 (lr=9.8543e-05) (hash(x)=24334708) +345 train 7.973339 (lr=9.8492e-05) (hash(x)=25617709) +346 train 8.064034 (lr=9.8441e-05) (hash(x)=25152362) +347 train 7.936830 (lr=9.8389e-05) (hash(x)=24276512) +348 train 7.855710 (lr=9.8336e-05) (hash(x)=26776133) +349 train 7.821504 (lr=9.8282e-05) (hash(x)=20254159) +350 val loss 7.9798 +350 val perplexity 2921.4268 +350 train 7.863307 (lr=9.8228e-05) (hash(x)=27873855) +351 train 7.943535 (lr=9.8172e-05) (hash(x)=25113614) +352 train 8.002220 (lr=9.8116e-05) (hash(x)=26192886) +353 train 7.892105 (lr=9.8059e-05) (hash(x)=24614912) +354 train 7.882319 (lr=9.8001e-05) (hash(x)=25397093) +355 train 7.913244 (lr=9.7942e-05) (hash(x)=19811802) +356 train 7.902263 (lr=9.7882e-05) (hash(x)=25579655) +357 train 8.000493 (lr=9.7822e-05) (hash(x)=26096514) +358 train 7.894327 (lr=9.7761e-05) (hash(x)=22589633) +359 train 7.899044 (lr=9.7699e-05) (hash(x)=26757147) +360 train 7.991504 (lr=9.7636e-05) (hash(x)=24963730) +361 train 7.889130 (lr=9.7572e-05) (hash(x)=27703124) +362 train 7.784777 (lr=9.7507e-05) (hash(x)=24781713) +363 train 7.931467 (lr=9.7442e-05) (hash(x)=24720476) +364 train 7.904047 (lr=9.7376e-05) (hash(x)=23481301) +365 train 7.831268 (lr=9.7309e-05) (hash(x)=25641033) +366 train 7.874544 (lr=9.7241e-05) (hash(x)=27369437) +367 train 7.910106 (lr=9.7172e-05) (hash(x)=24760900) +368 train 7.811197 (lr=9.7103e-05) (hash(x)=21692546) +369 train 7.871605 (lr=9.7032e-05) (hash(x)=22720391) +370 train 7.823040 (lr=9.6961e-05) (hash(x)=24376979) +371 train 7.841172 (lr=9.6889e-05) (hash(x)=22244509) +372 train 7.782748 (lr=9.6817e-05) (hash(x)=21831960) +373 train 7.763693 (lr=9.6743e-05) (hash(x)=24921535) +374 train 7.779957 (lr=9.6669e-05) (hash(x)=25381251) +375 train 7.777678 (lr=9.6593e-05) (hash(x)=25687352) +376 train 7.813241 (lr=9.6518e-05) (hash(x)=25221654) +377 train 7.852357 (lr=9.6441e-05) (hash(x)=24469863) +378 train 7.881749 (lr=9.6363e-05) (hash(x)=25291413) +379 train 7.791284 (lr=9.6285e-05) (hash(x)=25744274) +380 train 7.722918 (lr=9.6206e-05) (hash(x)=24246963) +381 train 7.758779 (lr=9.6126e-05) (hash(x)=19724058) +382 train 7.723138 (lr=9.6045e-05) (hash(x)=20705649) +383 train 7.925657 (lr=9.5963e-05) (hash(x)=24938685) +384 train 7.953248 (lr=9.5881e-05) (hash(x)=21707000) +385 train 7.764791 (lr=9.5798e-05) (hash(x)=24948980) +386 train 7.979391 (lr=9.5714e-05) (hash(x)=30773958) +387 train 8.039646 (lr=9.5629e-05) (hash(x)=26769046) +388 train 7.766995 (lr=9.5544e-05) (hash(x)=18849656) +389 train 8.235681 (lr=9.5457e-05) (hash(x)=24624950) +390 train 7.865127 (lr=9.5370e-05) (hash(x)=27299605) +391 train 7.862686 (lr=9.5282e-05) (hash(x)=24113445) +392 train 7.881818 (lr=9.5194e-05) (hash(x)=32644465) +393 train 7.942194 (lr=9.5104e-05) (hash(x)=25829219) +394 train 7.794140 (lr=9.5014e-05) (hash(x)=25194550) +395 train 7.829473 (lr=9.4923e-05) (hash(x)=25941804) +396 train 7.823082 (lr=9.4831e-05) (hash(x)=24075727) +397 train 7.874372 (lr=9.4739e-05) (hash(x)=23104539) +398 train 7.803888 (lr=9.4646e-05) (hash(x)=23018983) +399 train 7.762291 (lr=9.4551e-05) (hash(x)=24422929) +400 val loss 7.8144 +400 val perplexity 2476.0286 +400 train 7.793239 (lr=9.4457e-05) (hash(x)=23389696) +401 train 7.751203 (lr=9.4361e-05) (hash(x)=29102969) +402 train 7.896787 (lr=9.4265e-05) (hash(x)=28359859) +403 train 7.866409 (lr=9.4168e-05) (hash(x)=24630955) +404 train 7.920580 (lr=9.4070e-05) (hash(x)=22969113) +405 train 7.762050 (lr=9.3971e-05) (hash(x)=27004415) +406 train 7.696747 (lr=9.3872e-05) (hash(x)=22265063) +407 train 7.722599 (lr=9.3772e-05) (hash(x)=26905990) +408 train 7.759955 (lr=9.3671e-05) (hash(x)=24047679) +409 train 7.883364 (lr=9.3569e-05) (hash(x)=26905582) +410 train 7.673956 (lr=9.3467e-05) (hash(x)=29018970) +411 train 7.681662 (lr=9.3364e-05) (hash(x)=24503315) +412 train 7.696468 (lr=9.3260e-05) (hash(x)=22607537) +413 train 8.317210 (lr=9.3155e-05) (hash(x)=26690225) +414 train 7.748966 (lr=9.3050e-05) (hash(x)=24226255) +415 train 7.694291 (lr=9.2944e-05) (hash(x)=23622702) +416 train 7.742213 (lr=9.2837e-05) (hash(x)=23910425) +417 train 7.950181 (lr=9.2729e-05) (hash(x)=27213318) +418 train 7.766903 (lr=9.2621e-05) (hash(x)=26701577) +419 train 7.709290 (lr=9.2512e-05) (hash(x)=26216794) +420 train 7.911399 (lr=9.2402e-05) (hash(x)=27579623) +421 train 7.749998 (lr=9.2292e-05) (hash(x)=26086198) +422 train 7.725486 (lr=9.2181e-05) (hash(x)=24863441) +423 train 7.834483 (lr=9.2069e-05) (hash(x)=25236385) +424 train 7.847645 (lr=9.1956e-05) (hash(x)=25330803) +425 train 7.751805 (lr=9.1843e-05) (hash(x)=22508532) +426 train 7.735059 (lr=9.1729e-05) (hash(x)=21799261) +427 train 7.833676 (lr=9.1614e-05) (hash(x)=26311168) +428 train 7.628454 (lr=9.1499e-05) (hash(x)=20977430) +429 train 7.690459 (lr=9.1382e-05) (hash(x)=26941617) +430 train 7.933640 (lr=9.1266e-05) (hash(x)=30008957) +431 train 7.799057 (lr=9.1148e-05) (hash(x)=27125962) +432 train 7.666130 (lr=9.1030e-05) (hash(x)=25888358) +433 train 7.801087 (lr=9.0911e-05) (hash(x)=21725719) +434 train 7.651401 (lr=9.0791e-05) (hash(x)=21669704) +435 train 7.641684 (lr=9.0671e-05) (hash(x)=23166993) +436 train 7.700220 (lr=9.0550e-05) (hash(x)=26235974) +437 train 7.834143 (lr=9.0428e-05) (hash(x)=25426430) +438 train 7.600670 (lr=9.0306e-05) (hash(x)=22966314) +439 train 7.743892 (lr=9.0182e-05) (hash(x)=23448505) +440 train 7.617342 (lr=9.0059e-05) (hash(x)=26330693) +441 train 7.571769 (lr=8.9934e-05) (hash(x)=29550596) +442 train 7.633179 (lr=8.9809e-05) (hash(x)=22124892) +443 train 7.662175 (lr=8.9683e-05) (hash(x)=25607640) +444 train 7.503279 (lr=8.9557e-05) (hash(x)=23795894) +445 train 7.405691 (lr=8.9430e-05) (hash(x)=24165449) +446 train 7.661191 (lr=8.9302e-05) (hash(x)=24697756) +447 train 7.731255 (lr=8.9173e-05) (hash(x)=27218899) +448 train 7.628002 (lr=8.9044e-05) (hash(x)=24748522) +449 train 7.710495 (lr=8.8914e-05) (hash(x)=24965317) +450 val loss 7.7181 +450 val perplexity 2248.6724 +450 train 7.509076 (lr=8.8784e-05) (hash(x)=22031210) +451 train 7.634946 (lr=8.8653e-05) (hash(x)=25877990) +452 train 7.670715 (lr=8.8521e-05) (hash(x)=24382381) +453 train 7.578864 (lr=8.8388e-05) (hash(x)=24551999) +454 train 7.685044 (lr=8.8255e-05) (hash(x)=25765516) +455 train 7.698905 (lr=8.8122e-05) (hash(x)=25204247) +456 train 7.742649 (lr=8.7987e-05) (hash(x)=26321813) +457 train 7.680390 (lr=8.7852e-05) (hash(x)=25528323) +458 train 7.668320 (lr=8.7717e-05) (hash(x)=25114729) +459 train 7.641239 (lr=8.7580e-05) (hash(x)=31341964) +460 train 7.687690 (lr=8.7444e-05) (hash(x)=26326024) +461 train 7.608315 (lr=8.7306e-05) (hash(x)=27273311) +462 train 7.594098 (lr=8.7168e-05) (hash(x)=17819313) +463 train 7.735268 (lr=8.7029e-05) (hash(x)=23769243) +464 train 7.701671 (lr=8.6890e-05) (hash(x)=23788738) +465 train 7.648674 (lr=8.6750e-05) (hash(x)=24818744) +466 train 7.581030 (lr=8.6609e-05) (hash(x)=23472795) +467 train 7.971263 (lr=8.6468e-05) (hash(x)=24991832) +468 train 7.717703 (lr=8.6326e-05) (hash(x)=22424066) +469 train 7.612124 (lr=8.6184e-05) (hash(x)=26474878) +470 train 7.694763 (lr=8.6041e-05) (hash(x)=22704349) +471 train 7.642395 (lr=8.5897e-05) (hash(x)=22293673) +472 train 7.674755 (lr=8.5753e-05) (hash(x)=21785487) +473 train 7.673493 (lr=8.5608e-05) (hash(x)=25251063) +474 train 7.568631 (lr=8.5462e-05) (hash(x)=22722211) +475 train 7.639103 (lr=8.5316e-05) (hash(x)=24522269) +476 train 7.589877 (lr=8.5170e-05) (hash(x)=23329475) +477 train 7.608531 (lr=8.5022e-05) (hash(x)=24735165) +478 train 7.437623 (lr=8.4875e-05) (hash(x)=27002892) +479 train 7.471248 (lr=8.4726e-05) (hash(x)=23499349) +480 train 7.796839 (lr=8.4577e-05) (hash(x)=22461589) +481 train 7.574415 (lr=8.4428e-05) (hash(x)=26415040) +482 train 7.796886 (lr=8.4278e-05) (hash(x)=26349465) +483 train 7.476307 (lr=8.4127e-05) (hash(x)=21596677) +484 train 7.520811 (lr=8.3976e-05) (hash(x)=24117012) +485 train 7.590433 (lr=8.3824e-05) (hash(x)=24105761) +486 train 7.538509 (lr=8.3672e-05) (hash(x)=25725049) +487 train 7.401945 (lr=8.3519e-05) (hash(x)=22892776) +488 train 7.498937 (lr=8.3366e-05) (hash(x)=21630383) +489 train 7.549215 (lr=8.3212e-05) (hash(x)=24727094) +490 train 7.480649 (lr=8.3057e-05) (hash(x)=22835000) +491 train 7.657239 (lr=8.2902e-05) (hash(x)=22839049) +492 train 7.518077 (lr=8.2746e-05) (hash(x)=20654136) +493 train 7.728491 (lr=8.2590e-05) (hash(x)=26776000) +494 train 7.352789 (lr=8.2434e-05) (hash(x)=22560658) +495 train 7.771667 (lr=8.2276e-05) (hash(x)=25995933) +496 train 7.698167 (lr=8.2119e-05) (hash(x)=24635394) +497 train 7.801898 (lr=8.1960e-05) (hash(x)=27537228) +498 train 7.700186 (lr=8.1801e-05) (hash(x)=26496730) +499 train 7.444531 (lr=8.1642e-05) (hash(x)=16533310) +500 val loss 7.6562 +500 val perplexity 2113.8096 +500 train 7.549390 (lr=8.1482e-05) (hash(x)=24881176) +501 train 7.764247 (lr=8.1322e-05) (hash(x)=25593613) +502 train 7.518763 (lr=8.1161e-05) (hash(x)=22886646) +503 train 7.773330 (lr=8.1000e-05) (hash(x)=21407676) +504 train 7.653380 (lr=8.0838e-05) (hash(x)=21720304) +505 train 7.778703 (lr=8.0676e-05) (hash(x)=27263416) +506 train 7.537158 (lr=8.0513e-05) (hash(x)=25840801) +507 train 7.677763 (lr=8.0349e-05) (hash(x)=31424568) +508 train 7.680216 (lr=8.0186e-05) (hash(x)=23403065) +509 train 7.512791 (lr=8.0021e-05) (hash(x)=22656033) +510 train 7.647969 (lr=7.9856e-05) (hash(x)=26869295) +511 train 7.881473 (lr=7.9691e-05) (hash(x)=25504151) +512 train 7.686346 (lr=7.9525e-05) (hash(x)=25781518) +513 train 7.650986 (lr=7.9359e-05) (hash(x)=26432850) +514 train 7.521093 (lr=7.9192e-05) (hash(x)=20155894) +515 train 7.439140 (lr=7.9025e-05) (hash(x)=22270169) +516 train 7.575593 (lr=7.8858e-05) (hash(x)=22108461) +517 train 7.636905 (lr=7.8689e-05) (hash(x)=24802848) +518 train 7.722232 (lr=7.8521e-05) (hash(x)=25086604) +519 train 7.728895 (lr=7.8352e-05) (hash(x)=26983186) +520 train 7.595819 (lr=7.8182e-05) (hash(x)=27208507) +521 train 7.765323 (lr=7.8012e-05) (hash(x)=23862845) +522 train 7.587440 (lr=7.7842e-05) (hash(x)=22664494) +523 train 7.566633 (lr=7.7671e-05) (hash(x)=24092927) +524 train 7.545974 (lr=7.7500e-05) (hash(x)=27213812) +525 train 7.530200 (lr=7.7328e-05) (hash(x)=23929955) +526 train 7.534419 (lr=7.7156e-05) (hash(x)=23025790) +527 train 7.477386 (lr=7.6984e-05) (hash(x)=20452272) +528 train 7.661662 (lr=7.6811e-05) (hash(x)=26473464) +529 train 7.643034 (lr=7.6637e-05) (hash(x)=28276603) +530 train 7.477668 (lr=7.6463e-05) (hash(x)=21480367) +531 train 7.669715 (lr=7.6289e-05) (hash(x)=24658642) +532 train 7.532758 (lr=7.6115e-05) (hash(x)=26854801) +533 train 7.459313 (lr=7.5940e-05) (hash(x)=17862226) +534 train 7.563970 (lr=7.5764e-05) (hash(x)=28519182) +535 train 7.532729 (lr=7.5588e-05) (hash(x)=25193694) +536 train 7.517937 (lr=7.5412e-05) (hash(x)=25642620) +537 train 7.580407 (lr=7.5235e-05) (hash(x)=25765909) +538 train 7.426261 (lr=7.5058e-05) (hash(x)=20750531) +539 train 7.528997 (lr=7.4881e-05) (hash(x)=24917098) +540 train 7.487857 (lr=7.4703e-05) (hash(x)=24499766) +541 train 7.542637 (lr=7.4525e-05) (hash(x)=25067358) +542 train 7.654972 (lr=7.4346e-05) (hash(x)=27479796) +543 train 7.523879 (lr=7.4167e-05) (hash(x)=23136080) +544 train 7.605150 (lr=7.3988e-05) (hash(x)=29810074) +545 train 7.514591 (lr=7.3808e-05) (hash(x)=24765539) +546 train 7.548844 (lr=7.3628e-05) (hash(x)=22216309) +547 train 7.473950 (lr=7.3448e-05) (hash(x)=20613861) +548 train 7.451754 (lr=7.3267e-05) (hash(x)=23462447) +549 train 7.514870 (lr=7.3086e-05) (hash(x)=26739455) +550 val loss 7.6070 +550 val perplexity 2012.2100 +550 train 7.575506 (lr=7.2904e-05) (hash(x)=23222996) +551 train 7.571495 (lr=7.2723e-05) (hash(x)=26489285) +552 train 7.411801 (lr=7.2540e-05) (hash(x)=25185942) +553 train 7.723765 (lr=7.2358e-05) (hash(x)=27476158) +554 train 7.494531 (lr=7.2175e-05) (hash(x)=23653175) +555 train 7.470922 (lr=7.1992e-05) (hash(x)=22405076) +556 train 7.490283 (lr=7.1808e-05) (hash(x)=25673545) +557 train 7.579535 (lr=7.1624e-05) (hash(x)=22675428) +558 train 7.687963 (lr=7.1440e-05) (hash(x)=28168456) +559 train 7.797359 (lr=7.1256e-05) (hash(x)=26072586) +560 train 7.597937 (lr=7.1071e-05) (hash(x)=24974331) +561 train 7.425802 (lr=7.0886e-05) (hash(x)=21951984) +562 train 7.760365 (lr=7.0701e-05) (hash(x)=25403751) +563 train 7.556926 (lr=7.0515e-05) (hash(x)=26396012) +564 train 7.536009 (lr=7.0329e-05) (hash(x)=24827190) +565 train 7.639754 (lr=7.0143e-05) (hash(x)=24847458) +566 train 7.602737 (lr=6.9956e-05) (hash(x)=22159088) +567 train 7.544682 (lr=6.9769e-05) (hash(x)=22140034) +568 train 7.597380 (lr=6.9582e-05) (hash(x)=19044797) +569 train 7.493758 (lr=6.9394e-05) (hash(x)=23932794) +570 train 7.602577 (lr=6.9207e-05) (hash(x)=25256890) +571 train 7.671770 (lr=6.9019e-05) (hash(x)=24161711) +572 train 7.586296 (lr=6.8830e-05) (hash(x)=22036817) +573 train 7.439018 (lr=6.8642e-05) (hash(x)=26663208) +574 train 7.501388 (lr=6.8453e-05) (hash(x)=24072213) +575 train 7.488218 (lr=6.8264e-05) (hash(x)=22766044) +576 train 7.613396 (lr=6.8075e-05) (hash(x)=27871153) +577 train 7.571980 (lr=6.7885e-05) (hash(x)=24037280) +578 train 7.672565 (lr=6.7695e-05) (hash(x)=24121850) +579 train 7.537667 (lr=6.7505e-05) (hash(x)=24380031) +580 train 7.489305 (lr=6.7315e-05) (hash(x)=23055215) +581 train 7.763681 (lr=6.7124e-05) (hash(x)=25329724) +582 train 7.471230 (lr=6.6933e-05) (hash(x)=25662408) +583 train 7.564167 (lr=6.6742e-05) (hash(x)=26346814) +584 train 7.678435 (lr=6.6551e-05) (hash(x)=25283256) +585 train 7.477571 (lr=6.6360e-05) (hash(x)=24021771) +586 train 7.497192 (lr=6.6168e-05) (hash(x)=23576840) +587 train 7.651349 (lr=6.5976e-05) (hash(x)=26414858) +588 train 7.415617 (lr=6.5784e-05) (hash(x)=20099261) +589 train 7.415029 (lr=6.5592e-05) (hash(x)=22453718) +590 train 7.383235 (lr=6.5399e-05) (hash(x)=22529445) +591 train 7.466390 (lr=6.5206e-05) (hash(x)=24484543) +592 train 7.389916 (lr=6.5013e-05) (hash(x)=24043998) +593 train 7.529675 (lr=6.4820e-05) (hash(x)=24301906) +594 train 7.471732 (lr=6.4627e-05) (hash(x)=26328881) +595 train 7.517441 (lr=6.4433e-05) (hash(x)=23101508) +596 train 7.489532 (lr=6.4240e-05) (hash(x)=23527176) +597 train 7.992914 (lr=6.4046e-05) (hash(x)=22705673) +598 train 7.546836 (lr=6.3852e-05) (hash(x)=28664796) +599 train 7.656979 (lr=6.3658e-05) (hash(x)=27102890) +600 val loss 7.5802 +600 val perplexity 1959.1005 +600 train 7.731663 (lr=6.3463e-05) (hash(x)=26787259) +601 train 7.780401 (lr=6.3269e-05) (hash(x)=23396088) +602 train 7.505399 (lr=6.3074e-05) (hash(x)=24911831) +603 train 7.520839 (lr=6.2879e-05) (hash(x)=23453491) +604 train 7.568392 (lr=6.2684e-05) (hash(x)=23453396) +605 train 7.660997 (lr=6.2489e-05) (hash(x)=23869612) +606 train 7.745296 (lr=6.2294e-05) (hash(x)=26581590) +607 train 7.568105 (lr=6.2098e-05) (hash(x)=28056342) +608 train 7.628171 (lr=6.1903e-05) (hash(x)=25900866) +609 train 7.649824 (lr=6.1707e-05) (hash(x)=24549177) +610 train 7.576864 (lr=6.1511e-05) (hash(x)=28239583) +611 train 7.598382 (lr=6.1315e-05) (hash(x)=23223554) +612 train 7.662986 (lr=6.1119e-05) (hash(x)=24793480) +613 train 7.566033 (lr=6.0923e-05) (hash(x)=25415570) +614 train 7.632292 (lr=6.0726e-05) (hash(x)=24425760) +615 train 7.684543 (lr=6.0530e-05) (hash(x)=23707134) +616 train 7.631091 (lr=6.0333e-05) (hash(x)=28321697) +617 train 7.566113 (lr=6.0137e-05) (hash(x)=23434031) +618 train 7.550931 (lr=5.9940e-05) (hash(x)=27003481) +619 train 7.531819 (lr=5.9743e-05) (hash(x)=22236893) +620 train 7.575702 (lr=5.9546e-05) (hash(x)=26857458) +621 train 7.558805 (lr=5.9349e-05) (hash(x)=26413122) +622 train 7.851124 (lr=5.9152e-05) (hash(x)=27696537) +623 train 7.592858 (lr=5.8955e-05) (hash(x)=21950234) +624 train 7.615797 (lr=5.8758e-05) (hash(x)=24051952) +625 train 7.631971 (lr=5.8560e-05) (hash(x)=24029939) +626 train 7.719388 (lr=5.8363e-05) (hash(x)=24393346) +627 train 7.670229 (lr=5.8165e-05) (hash(x)=26985352) +628 train 7.408197 (lr=5.7968e-05) (hash(x)=19493265) +629 train 7.492118 (lr=5.7770e-05) (hash(x)=23944010) +630 train 7.712536 (lr=5.7573e-05) (hash(x)=21517435) +631 train 7.481009 (lr=5.7375e-05) (hash(x)=25025868) +632 train 7.502947 (lr=5.7177e-05) (hash(x)=22940997) +633 train 7.637960 (lr=5.6979e-05) (hash(x)=25864840) +634 train 7.540421 (lr=5.6782e-05) (hash(x)=24682151) +635 train 7.458861 (lr=5.6584e-05) (hash(x)=27055330) +636 train 7.509686 (lr=5.6386e-05) (hash(x)=22705710) +637 train 7.607580 (lr=5.6188e-05) (hash(x)=26775238) +638 train 7.739300 (lr=5.5990e-05) (hash(x)=21734331) +639 train 7.483868 (lr=5.5792e-05) (hash(x)=25693442) +640 train 7.581870 (lr=5.5594e-05) (hash(x)=25290454) +641 train 7.546197 (lr=5.5396e-05) (hash(x)=25699860) +642 train 7.715387 (lr=5.5198e-05) (hash(x)=19919115) +643 train 7.785877 (lr=5.5000e-05) (hash(x)=25586967) +644 train 7.526165 (lr=5.4802e-05) (hash(x)=25373506) +645 train 7.538226 (lr=5.4604e-05) (hash(x)=26496148) +646 train 7.603675 (lr=5.4406e-05) (hash(x)=27395083) +647 train 7.801353 (lr=5.4208e-05) (hash(x)=27389709) +648 train 7.671622 (lr=5.4010e-05) (hash(x)=29172968) +649 train 7.602849 (lr=5.3812e-05) (hash(x)=25464920) +650 val loss 7.5588 +650 val perplexity 1917.4703 +650 train 7.640497 (lr=5.3614e-05) (hash(x)=25992320) +651 train 7.613410 (lr=5.3416e-05) (hash(x)=24949509) +652 train 7.636610 (lr=5.3218e-05) (hash(x)=25615151) +653 train 7.607439 (lr=5.3021e-05) (hash(x)=27923548) +654 train 7.638077 (lr=5.2823e-05) (hash(x)=26387848) +655 train 7.583149 (lr=5.2625e-05) (hash(x)=22863518) +656 train 7.528802 (lr=5.2427e-05) (hash(x)=20869581) +657 train 7.684580 (lr=5.2230e-05) (hash(x)=18133606) +658 train 7.569495 (lr=5.2032e-05) (hash(x)=28736122) +659 train 7.554872 (lr=5.1835e-05) (hash(x)=25528329) +660 train 7.612813 (lr=5.1637e-05) (hash(x)=24020765) +661 train 7.658054 (lr=5.1440e-05) (hash(x)=27792245) +662 train 7.673942 (lr=5.1242e-05) (hash(x)=28287635) +663 train 7.397913 (lr=5.1045e-05) (hash(x)=20951206) +664 train 7.480437 (lr=5.0848e-05) (hash(x)=22356452) +665 train 7.553826 (lr=5.0651e-05) (hash(x)=24385170) +666 train 7.636360 (lr=5.0454e-05) (hash(x)=23455854) +667 train 7.618863 (lr=5.0257e-05) (hash(x)=21589405) +668 train 7.590744 (lr=5.0060e-05) (hash(x)=24698150) +669 train 7.575311 (lr=4.9863e-05) (hash(x)=28010836) +670 train 7.696945 (lr=4.9667e-05) (hash(x)=25427381) +671 train 7.660029 (lr=4.9470e-05) (hash(x)=27478557) +672 train 7.605933 (lr=4.9274e-05) (hash(x)=23865569) +673 train 7.673905 (lr=4.9077e-05) (hash(x)=23407132) +674 train 7.531689 (lr=4.8881e-05) (hash(x)=20253664) +675 train 7.499235 (lr=4.8685e-05) (hash(x)=21087928) +676 train 7.615239 (lr=4.8489e-05) (hash(x)=25957589) +677 train 7.536899 (lr=4.8293e-05) (hash(x)=27008610) +678 train 7.612415 (lr=4.8097e-05) (hash(x)=27580265) +679 train 7.651302 (lr=4.7902e-05) (hash(x)=24244798) +680 train 7.528227 (lr=4.7706e-05) (hash(x)=24535871) +681 train 7.629283 (lr=4.7511e-05) (hash(x)=27558681) +682 train 7.498579 (lr=4.7316e-05) (hash(x)=21152793) +683 train 7.482566 (lr=4.7121e-05) (hash(x)=28147005) +684 train 7.573699 (lr=4.6926e-05) (hash(x)=21933180) +685 train 7.550570 (lr=4.6731e-05) (hash(x)=24735895) +686 train 7.598121 (lr=4.6537e-05) (hash(x)=26209948) +687 train 7.578415 (lr=4.6342e-05) (hash(x)=23358313) +688 train 7.526731 (lr=4.6148e-05) (hash(x)=22398001) +689 train 7.451158 (lr=4.5954e-05) (hash(x)=23724957) +690 train 7.497720 (lr=4.5760e-05) (hash(x)=23083514) +691 train 7.368492 (lr=4.5567e-05) (hash(x)=16016924) +692 train 7.632649 (lr=4.5373e-05) (hash(x)=27173674) +693 train 7.435435 (lr=4.5180e-05) (hash(x)=28056250) +694 train 7.552862 (lr=4.4987e-05) (hash(x)=22233617) +695 train 7.635375 (lr=4.4794e-05) (hash(x)=23777908) +696 train 7.604048 (lr=4.4601e-05) (hash(x)=27842849) +697 train 7.595302 (lr=4.4408e-05) (hash(x)=23711679) +698 train 7.437038 (lr=4.4216e-05) (hash(x)=17204843) +699 train 7.746481 (lr=4.4024e-05) (hash(x)=23909800) +700 val loss 7.5281 +700 val perplexity 1859.5602 +700 train 7.600750 (lr=4.3832e-05) (hash(x)=25033286) +701 train 7.892196 (lr=4.3640e-05) (hash(x)=23608371) +702 train 7.663329 (lr=4.3449e-05) (hash(x)=23654625) +703 train 7.506060 (lr=4.3258e-05) (hash(x)=22633313) +704 train 7.558197 (lr=4.3067e-05) (hash(x)=24267011) +705 train 7.513043 (lr=4.2876e-05) (hash(x)=23449416) +706 train 7.763673 (lr=4.2685e-05) (hash(x)=28527906) +707 train 7.454723 (lr=4.2495e-05) (hash(x)=21189075) +708 train 7.500453 (lr=4.2305e-05) (hash(x)=25278645) +709 train 8.016804 (lr=4.2115e-05) (hash(x)=23654290) +710 train 7.504948 (lr=4.1925e-05) (hash(x)=24571139) +711 train 7.596189 (lr=4.1736e-05) (hash(x)=26203172) +712 train 7.532542 (lr=4.1547e-05) (hash(x)=19623503) +713 train 7.517661 (lr=4.1358e-05) (hash(x)=23970545) +714 train 7.683751 (lr=4.1170e-05) (hash(x)=25724900) +715 train 7.608215 (lr=4.0981e-05) (hash(x)=23245421) +716 train 7.562238 (lr=4.0793e-05) (hash(x)=24087537) +717 train 7.482961 (lr=4.0606e-05) (hash(x)=25549646) +718 train 7.565393 (lr=4.0418e-05) (hash(x)=26406563) +719 train 7.613027 (lr=4.0231e-05) (hash(x)=23972429) +720 train 7.663969 (lr=4.0044e-05) (hash(x)=26378040) +721 train 7.517647 (lr=3.9857e-05) (hash(x)=24618531) +722 train 7.526717 (lr=3.9671e-05) (hash(x)=20135714) +723 train 7.653274 (lr=3.9485e-05) (hash(x)=26490981) +724 train 7.675335 (lr=3.9299e-05) (hash(x)=23927567) +725 train 7.512706 (lr=3.9114e-05) (hash(x)=27943609) +726 train 7.572027 (lr=3.8929e-05) (hash(x)=22962561) +727 train 7.617037 (lr=3.8744e-05) (hash(x)=24772257) +728 train 7.548615 (lr=3.8560e-05) (hash(x)=25539761) +729 train 7.689032 (lr=3.8376e-05) (hash(x)=23316262) +730 train 7.464641 (lr=3.8192e-05) (hash(x)=23857558) +731 train 7.651764 (lr=3.8008e-05) (hash(x)=30252690) +732 train 7.707368 (lr=3.7825e-05) (hash(x)=30590162) +733 train 7.592540 (lr=3.7642e-05) (hash(x)=25015721) +734 train 7.674330 (lr=3.7460e-05) (hash(x)=28639679) +735 train 7.555845 (lr=3.7277e-05) (hash(x)=26199058) +736 train 7.612779 (lr=3.7096e-05) (hash(x)=28907461) +737 train 7.588128 (lr=3.6914e-05) (hash(x)=26466915) +738 train 7.668214 (lr=3.6733e-05) (hash(x)=20332613) +739 train 7.548901 (lr=3.6552e-05) (hash(x)=26766470) +740 train 7.496449 (lr=3.6372e-05) (hash(x)=26722145) +741 train 7.550056 (lr=3.6192e-05) (hash(x)=21222489) +742 train 7.684075 (lr=3.6012e-05) (hash(x)=24662435) +743 train 7.496283 (lr=3.5833e-05) (hash(x)=24308927) +744 train 7.456376 (lr=3.5654e-05) (hash(x)=28442502) +745 train 7.501629 (lr=3.5475e-05) (hash(x)=22810435) +746 train 7.623215 (lr=3.5297e-05) (hash(x)=30486518) +747 train 7.601018 (lr=3.5119e-05) (hash(x)=22688712) +748 train 7.531012 (lr=3.4942e-05) (hash(x)=20712962) +749 train 7.396847 (lr=3.4765e-05) (hash(x)=27490086) +750 val loss 7.5112 +750 val perplexity 1828.3844 +750 train 7.394656 (lr=3.4588e-05) (hash(x)=23973908) +751 train 7.556612 (lr=3.4412e-05) (hash(x)=29132401) +752 train 7.534591 (lr=3.4236e-05) (hash(x)=27386700) +753 train 7.570761 (lr=3.4060e-05) (hash(x)=24268614) +754 train 7.622978 (lr=3.3885e-05) (hash(x)=27378387) +755 train 7.424912 (lr=3.3711e-05) (hash(x)=22529400) +756 train 7.504970 (lr=3.3537e-05) (hash(x)=27586989) +757 train 7.434928 (lr=3.3363e-05) (hash(x)=22336349) +758 train 7.471633 (lr=3.3189e-05) (hash(x)=26147854) +759 train 7.561799 (lr=3.3016e-05) (hash(x)=25346788) +760 train 7.646678 (lr=3.2844e-05) (hash(x)=25956855) +761 train 7.581672 (lr=3.2672e-05) (hash(x)=24367780) +762 train 7.474728 (lr=3.2500e-05) (hash(x)=26084462) +763 train 7.550223 (lr=3.2329e-05) (hash(x)=26702459) +764 train 7.486784 (lr=3.2158e-05) (hash(x)=25717976) +765 train 7.436249 (lr=3.1988e-05) (hash(x)=22141366) +766 train 7.659554 (lr=3.1818e-05) (hash(x)=24652423) +767 train 7.572710 (lr=3.1648e-05) (hash(x)=30911699) +768 train 7.564221 (lr=3.1479e-05) (hash(x)=22076954) +769 train 7.744118 (lr=3.1311e-05) (hash(x)=21586769) +770 train 7.568444 (lr=3.1142e-05) (hash(x)=21994337) +771 train 7.523945 (lr=3.0975e-05) (hash(x)=25454145) +772 train 7.574931 (lr=3.0808e-05) (hash(x)=23239550) +773 train 7.623880 (lr=3.0641e-05) (hash(x)=23565926) +774 train 7.485872 (lr=3.0475e-05) (hash(x)=24705563) +775 train 7.592660 (lr=3.0309e-05) (hash(x)=27671095) +776 train 7.702354 (lr=3.0144e-05) (hash(x)=26254303) +777 train 7.448067 (lr=2.9979e-05) (hash(x)=27454020) +778 train 7.447711 (lr=2.9814e-05) (hash(x)=24394565) +779 train 7.405477 (lr=2.9651e-05) (hash(x)=21780199) +780 train 7.494816 (lr=2.9487e-05) (hash(x)=28296896) +781 train 7.624675 (lr=2.9324e-05) (hash(x)=26766303) +782 train 7.462801 (lr=2.9162e-05) (hash(x)=21599973) +783 train 7.776432 (lr=2.9000e-05) (hash(x)=24969360) +784 train 7.752959 (lr=2.8839e-05) (hash(x)=24382006) +785 train 7.425342 (lr=2.8678e-05) (hash(x)=22009299) +786 train 7.280449 (lr=2.8518e-05) (hash(x)=22376569) +787 train 7.450193 (lr=2.8358e-05) (hash(x)=28166843) +788 train 7.716834 (lr=2.8199e-05) (hash(x)=28230119) +789 train 7.662913 (lr=2.8040e-05) (hash(x)=26207465) +790 train 7.735008 (lr=2.7881e-05) (hash(x)=24748118) +791 train 7.666214 (lr=2.7724e-05) (hash(x)=27033802) +792 train 7.585179 (lr=2.7566e-05) (hash(x)=28788047) +793 train 7.453538 (lr=2.7410e-05) (hash(x)=22185099) +794 train 7.649105 (lr=2.7254e-05) (hash(x)=21585136) +795 train 7.461902 (lr=2.7098e-05) (hash(x)=27691968) +796 train 7.511041 (lr=2.6943e-05) (hash(x)=23706103) +797 train 7.347975 (lr=2.6788e-05) (hash(x)=27918030) +798 train 7.567157 (lr=2.6634e-05) (hash(x)=23127886) +799 train 7.582398 (lr=2.6481e-05) (hash(x)=26197070) +800 val loss 7.4981 +800 val perplexity 1804.6023 +800 train 7.633736 (lr=2.6328e-05) (hash(x)=26463005) +801 train 7.523831 (lr=2.6176e-05) (hash(x)=19160134) +802 train 7.476406 (lr=2.6024e-05) (hash(x)=23699136) +803 train 7.538136 (lr=2.5873e-05) (hash(x)=25353051) +804 train 7.475466 (lr=2.5722e-05) (hash(x)=21558484) +805 train 7.391947 (lr=2.5572e-05) (hash(x)=22828380) +806 train 7.518044 (lr=2.5423e-05) (hash(x)=24623811) +807 train 7.686347 (lr=2.5274e-05) (hash(x)=25410299) +808 train 7.545372 (lr=2.5125e-05) (hash(x)=26046091) +809 train 7.666407 (lr=2.4978e-05) (hash(x)=29887482) +810 train 7.410117 (lr=2.4830e-05) (hash(x)=21809706) +811 train 7.532879 (lr=2.4684e-05) (hash(x)=23712872) +812 train 7.432785 (lr=2.4538e-05) (hash(x)=25875103) +813 train 7.402735 (lr=2.4392e-05) (hash(x)=23736441) +814 train 7.483343 (lr=2.4247e-05) (hash(x)=25093883) +815 train 7.430235 (lr=2.4103e-05) (hash(x)=23692325) +816 train 7.321647 (lr=2.3959e-05) (hash(x)=23061082) +817 train 7.461235 (lr=2.3816e-05) (hash(x)=26094253) +818 train 7.607501 (lr=2.3674e-05) (hash(x)=27327886) +819 train 7.472259 (lr=2.3532e-05) (hash(x)=26186143) +820 train 7.467237 (lr=2.3391e-05) (hash(x)=25336043) +821 train 7.515358 (lr=2.3250e-05) (hash(x)=22811468) +822 train 7.499232 (lr=2.3110e-05) (hash(x)=27925831) +823 train 7.595544 (lr=2.2971e-05) (hash(x)=35528108) +824 train 7.594224 (lr=2.2832e-05) (hash(x)=24949421) +825 train 7.355662 (lr=2.2694e-05) (hash(x)=22845735) +826 train 7.465000 (lr=2.2556e-05) (hash(x)=28138281) +827 train 7.534105 (lr=2.2420e-05) (hash(x)=29038360) +828 train 7.484027 (lr=2.2283e-05) (hash(x)=22827903) +829 train 7.565742 (lr=2.2148e-05) (hash(x)=24647015) +830 train 7.462306 (lr=2.2013e-05) (hash(x)=20477804) +831 train 7.439011 (lr=2.1878e-05) (hash(x)=26692640) +832 train 7.509223 (lr=2.1745e-05) (hash(x)=27388794) +833 train 7.775448 (lr=2.1612e-05) (hash(x)=25265875) +834 train 7.430226 (lr=2.1479e-05) (hash(x)=21841888) +835 train 7.519884 (lr=2.1347e-05) (hash(x)=19809196) +836 train 7.588273 (lr=2.1216e-05) (hash(x)=23007945) +837 train 7.521264 (lr=2.1086e-05) (hash(x)=24079409) +838 train 7.428468 (lr=2.0956e-05) (hash(x)=25874533) +839 train 7.377499 (lr=2.0827e-05) (hash(x)=23660106) +840 train 7.840339 (lr=2.0698e-05) (hash(x)=41821329) +841 train 7.458899 (lr=2.0570e-05) (hash(x)=25763899) +842 train 7.304698 (lr=2.0443e-05) (hash(x)=28584111) +843 train 7.781738 (lr=2.0317e-05) (hash(x)=25559887) +844 train 7.557550 (lr=2.0191e-05) (hash(x)=26886377) +845 train 7.525082 (lr=2.0066e-05) (hash(x)=26461895) +846 train 7.575584 (lr=1.9941e-05) (hash(x)=23018827) +847 train 7.379812 (lr=1.9818e-05) (hash(x)=24375035) +848 train 7.550056 (lr=1.9694e-05) (hash(x)=26431670) +849 train 7.437725 (lr=1.9572e-05) (hash(x)=25033511) +850 val loss 7.4947 +850 val perplexity 1798.4790 +850 train 7.550976 (lr=1.9450e-05) (hash(x)=25602577) +851 train 7.463595 (lr=1.9329e-05) (hash(x)=23316582) +852 train 7.593568 (lr=1.9209e-05) (hash(x)=22470095) +853 train 7.470088 (lr=1.9089e-05) (hash(x)=25089202) +854 train 7.407629 (lr=1.8970e-05) (hash(x)=26243247) +855 train 7.487153 (lr=1.8852e-05) (hash(x)=22231144) +856 train 7.406380 (lr=1.8734e-05) (hash(x)=27646564) +857 train 7.553175 (lr=1.8618e-05) (hash(x)=23889875) +858 train 7.499738 (lr=1.8501e-05) (hash(x)=25255240) +859 train 7.581257 (lr=1.8386e-05) (hash(x)=23883299) +860 train 7.559627 (lr=1.8271e-05) (hash(x)=18463841) +861 train 7.582145 (lr=1.8157e-05) (hash(x)=30657762) +862 train 7.478958 (lr=1.8044e-05) (hash(x)=25288079) +863 train 7.506331 (lr=1.7931e-05) (hash(x)=28979310) +864 train 7.693963 (lr=1.7819e-05) (hash(x)=30254991) +865 train 7.546370 (lr=1.7708e-05) (hash(x)=27274653) +866 train 7.530149 (lr=1.7598e-05) (hash(x)=23026066) +867 train 7.496278 (lr=1.7488e-05) (hash(x)=14001963) +868 train 7.395147 (lr=1.7379e-05) (hash(x)=21988054) +869 train 7.477046 (lr=1.7271e-05) (hash(x)=26241099) +870 train 7.574476 (lr=1.7163e-05) (hash(x)=26209722) +871 train 7.433781 (lr=1.7056e-05) (hash(x)=26928312) +872 train 7.509105 (lr=1.6950e-05) (hash(x)=23563272) +873 train 7.453948 (lr=1.6845e-05) (hash(x)=25704828) +874 train 7.707575 (lr=1.6740e-05) (hash(x)=22498992) +875 train 7.848406 (lr=1.6636e-05) (hash(x)=22895918) +876 train 7.393834 (lr=1.6533e-05) (hash(x)=24404283) +877 train 7.442524 (lr=1.6431e-05) (hash(x)=26663932) +878 train 7.579708 (lr=1.6329e-05) (hash(x)=22805316) +879 train 7.560781 (lr=1.6228e-05) (hash(x)=23344466) +880 train 7.508679 (lr=1.6128e-05) (hash(x)=20966940) +881 train 7.510818 (lr=1.6029e-05) (hash(x)=27422664) +882 train 7.546381 (lr=1.5930e-05) (hash(x)=22116510) +883 train 7.408511 (lr=1.5832e-05) (hash(x)=27269633) +884 train 7.516807 (lr=1.5735e-05) (hash(x)=27157711) +885 train 7.543061 (lr=1.5639e-05) (hash(x)=22774645) +886 train 7.467173 (lr=1.5543e-05) (hash(x)=22143146) +887 train 7.468454 (lr=1.5449e-05) (hash(x)=22356438) +888 train 7.249228 (lr=1.5354e-05) (hash(x)=23172158) +889 train 7.490241 (lr=1.5261e-05) (hash(x)=26971690) +890 train 7.528177 (lr=1.5169e-05) (hash(x)=24851922) +891 train 7.351926 (lr=1.5077e-05) (hash(x)=21017641) +892 train 7.494220 (lr=1.4986e-05) (hash(x)=28186957) +893 train 7.508732 (lr=1.4896e-05) (hash(x)=25661082) +894 train 7.413259 (lr=1.4806e-05) (hash(x)=25462781) +895 train 7.374049 (lr=1.4718e-05) (hash(x)=21774374) +896 train 7.370977 (lr=1.4630e-05) (hash(x)=23657055) +897 train 7.392092 (lr=1.4543e-05) (hash(x)=23604050) +898 train 7.823567 (lr=1.4456e-05) (hash(x)=31534071) +899 train 7.636145 (lr=1.4371e-05) (hash(x)=24183732) +900 val loss 7.4803 +900 val perplexity 1772.8552 +900 train 7.470290 (lr=1.4286e-05) (hash(x)=23384280) +901 train 7.353648 (lr=1.4202e-05) (hash(x)=21975928) +902 train 7.359590 (lr=1.4119e-05) (hash(x)=25281674) +903 train 7.339865 (lr=1.4037e-05) (hash(x)=22947712) +904 train 7.420816 (lr=1.3955e-05) (hash(x)=24853977) +905 train 7.471038 (lr=1.3874e-05) (hash(x)=23696470) +906 train 7.472367 (lr=1.3794e-05) (hash(x)=27489711) +907 train 7.405140 (lr=1.3715e-05) (hash(x)=28050163) +908 train 7.288162 (lr=1.3637e-05) (hash(x)=26996807) +909 train 7.674575 (lr=1.3559e-05) (hash(x)=28626495) +910 train 7.399714 (lr=1.3482e-05) (hash(x)=24783242) +911 train 7.411022 (lr=1.3407e-05) (hash(x)=25832479) +912 train 7.299203 (lr=1.3331e-05) (hash(x)=23522137) +913 train 7.589725 (lr=1.3257e-05) (hash(x)=26731973) +914 train 7.651633 (lr=1.3183e-05) (hash(x)=26019142) +915 train 7.434741 (lr=1.3111e-05) (hash(x)=25081877) +916 train 7.545449 (lr=1.3039e-05) (hash(x)=23843571) +917 train 7.555920 (lr=1.2968e-05) (hash(x)=26805739) +918 train 7.467187 (lr=1.2897e-05) (hash(x)=27169927) +919 train 7.499229 (lr=1.2828e-05) (hash(x)=26545491) +920 train 7.507201 (lr=1.2759e-05) (hash(x)=25179497) +921 train 7.454947 (lr=1.2691e-05) (hash(x)=27085003) +922 train 7.477243 (lr=1.2624e-05) (hash(x)=24558202) +923 train 7.618608 (lr=1.2558e-05) (hash(x)=25317684) +924 train 7.490183 (lr=1.2493e-05) (hash(x)=27104868) +925 train 7.526923 (lr=1.2428e-05) (hash(x)=24806274) +926 train 7.576830 (lr=1.2364e-05) (hash(x)=24341937) +927 train 7.372710 (lr=1.2301e-05) (hash(x)=22736558) +928 train 7.395321 (lr=1.2239e-05) (hash(x)=23843744) +929 train 7.522630 (lr=1.2178e-05) (hash(x)=27218468) +930 train 7.466071 (lr=1.2118e-05) (hash(x)=26042053) +931 train 7.478209 (lr=1.2058e-05) (hash(x)=25776082) +932 train 7.364209 (lr=1.1999e-05) (hash(x)=21594447) +933 train 7.402997 (lr=1.1941e-05) (hash(x)=21674469) +934 train 7.343635 (lr=1.1884e-05) (hash(x)=26213966) +935 train 7.347954 (lr=1.1828e-05) (hash(x)=23770702) +936 train 7.456875 (lr=1.1772e-05) (hash(x)=27829878) +937 train 7.566255 (lr=1.1718e-05) (hash(x)=29177278) +938 train 7.490334 (lr=1.1664e-05) (hash(x)=26020089) +939 train 7.389599 (lr=1.1611e-05) (hash(x)=24676475) +940 train 7.451327 (lr=1.1559e-05) (hash(x)=24056174) +941 train 7.474665 (lr=1.1508e-05) (hash(x)=21887514) +942 train 7.474421 (lr=1.1457e-05) (hash(x)=24196100) +943 train 7.435189 (lr=1.1408e-05) (hash(x)=25700093) +944 train 7.592031 (lr=1.1359e-05) (hash(x)=24365834) +945 train 7.390735 (lr=1.1311e-05) (hash(x)=25631786) +946 train 7.524990 (lr=1.1264e-05) (hash(x)=24452345) +947 train 7.242242 (lr=1.1218e-05) (hash(x)=21465330) +948 train 7.371133 (lr=1.1173e-05) (hash(x)=20427493) +949 train 7.398861 (lr=1.1128e-05) (hash(x)=25439720) +950 val loss 7.4757 +950 val perplexity 1764.6733 +950 train 7.469860 (lr=1.1085e-05) (hash(x)=21652988) +951 train 7.404196 (lr=1.1042e-05) (hash(x)=22546779) +952 train 7.270616 (lr=1.1000e-05) (hash(x)=20855736) +953 train 7.456549 (lr=1.0959e-05) (hash(x)=27725873) +954 train 7.521078 (lr=1.0919e-05) (hash(x)=24582989) +955 train 7.396343 (lr=1.0879e-05) (hash(x)=24822372) +956 train 7.454220 (lr=1.0841e-05) (hash(x)=25624588) +957 train 7.410018 (lr=1.0803e-05) (hash(x)=23566489) +958 train 7.591918 (lr=1.0766e-05) (hash(x)=28117387) +959 train 7.358937 (lr=1.0730e-05) (hash(x)=26874283) +960 train 7.364769 (lr=1.0695e-05) (hash(x)=28570123) +961 train 7.510401 (lr=1.0661e-05) (hash(x)=24807017) +962 train 7.394339 (lr=1.0628e-05) (hash(x)=25189993) +963 train 7.518966 (lr=1.0595e-05) (hash(x)=29202226) +964 train 7.691831 (lr=1.0563e-05) (hash(x)=26060713) +965 train 7.576899 (lr=1.0533e-05) (hash(x)=23053285) +966 train 7.558528 (lr=1.0503e-05) (hash(x)=21968481) +967 train 7.451611 (lr=1.0474e-05) (hash(x)=24492698) +968 train 7.420343 (lr=1.0445e-05) (hash(x)=25760806) +969 train 7.442885 (lr=1.0418e-05) (hash(x)=27417392) +970 train 7.445331 (lr=1.0391e-05) (hash(x)=26767105) +971 train 7.514004 (lr=1.0366e-05) (hash(x)=25117678) +972 train 7.393990 (lr=1.0341e-05) (hash(x)=24233758) +973 train 7.455315 (lr=1.0317e-05) (hash(x)=29907634) +974 train 7.247953 (lr=1.0294e-05) (hash(x)=23552436) +975 train 7.599018 (lr=1.0272e-05) (hash(x)=28610406) +976 train 7.435342 (lr=1.0251e-05) (hash(x)=23858212) +977 train 7.330421 (lr=1.0230e-05) (hash(x)=20567764) +978 train 7.342462 (lr=1.0211e-05) (hash(x)=29631106) +979 train 7.317904 (lr=1.0192e-05) (hash(x)=19843564) +980 train 7.446301 (lr=1.0174e-05) (hash(x)=22589891) +981 train 7.231230 (lr=1.0157e-05) (hash(x)=21183943) +982 train 7.406676 (lr=1.0141e-05) (hash(x)=26152241) +983 train 7.414310 (lr=1.0126e-05) (hash(x)=22700785) +984 train 7.642218 (lr=1.0111e-05) (hash(x)=29611280) +985 train 7.762672 (lr=1.0098e-05) (hash(x)=25972823) +986 train 7.457224 (lr=1.0085e-05) (hash(x)=23447782) +987 train 7.403764 (lr=1.0074e-05) (hash(x)=22578453) +988 train 7.396955 (lr=1.0063e-05) (hash(x)=25474622) +989 train 7.467656 (lr=1.0053e-05) (hash(x)=22965762) +990 train 7.501054 (lr=1.0044e-05) (hash(x)=26116321) +991 train 7.415164 (lr=1.0035e-05) (hash(x)=25934549) +992 train 7.334317 (lr=1.0028e-05) (hash(x)=23937495) +993 train 7.505708 (lr=1.0021e-05) (hash(x)=26286442) +994 train 7.755474 (lr=1.0016e-05) (hash(x)=32737711) +995 train 7.351036 (lr=1.0011e-05) (hash(x)=20563613) +996 train 7.603413 (lr=1.0007e-05) (hash(x)=26234770) +997 train 7.534561 (lr=1.0004e-05) (hash(x)=26179422) +998 train 7.468211 (lr=1.0002e-05) (hash(x)=23289651) +999 val loss 7.4747 +999 val perplexity 1762.8718 +999 train 7.832581 (lr=1.0000e-05) (hash(x)=36436617)