diff --git "a/lr1e-4_total_batch_size5120_seq_len512/log2.txt" "b/lr1e-4_total_batch_size5120_seq_len512/log2.txt" --- "a/lr1e-4_total_batch_size5120_seq_len512/log2.txt" +++ "b/lr1e-4_total_batch_size5120_seq_len512/log2.txt" @@ -21,1023 +21,1023 @@ max_steps: 1000 17 train 11.743484 (lr=6.2937e-06) (hash(x)=25588236) 18 train 11.715164 (lr=6.6434e-06) (hash(x)=23770034) 19 train 11.749471 (lr=6.9930e-06) (hash(x)=24011372) -20 train 11.657701 (lr=7.3427e-06) (hash(x)=25441898) -21 train 11.647016 (lr=7.6923e-06) (hash(x)=28375581) -22 train 11.656828 (lr=8.0420e-06) (hash(x)=24046679) -23 train 11.682503 (lr=8.3916e-06) (hash(x)=24611628) -24 train 11.669896 (lr=8.7413e-06) (hash(x)=26169030) -25 train 11.623303 (lr=9.0909e-06) (hash(x)=30298407) -26 train 11.645949 (lr=9.4406e-06) (hash(x)=23711112) -27 train 11.582213 (lr=9.7902e-06) (hash(x)=19245352) -28 train 11.576291 (lr=1.0140e-05) (hash(x)=21529136) -29 train 11.548808 (lr=1.0490e-05) (hash(x)=28936608) -30 train 11.553458 (lr=1.0839e-05) (hash(x)=24339013) -31 train 11.519302 (lr=1.1189e-05) (hash(x)=25767553) -32 train 11.518982 (lr=1.1538e-05) (hash(x)=26439905) -33 train 11.510749 (lr=1.1888e-05) (hash(x)=31093473) -34 train 11.484274 (lr=1.2238e-05) (hash(x)=25450374) -35 train 11.525412 (lr=1.2587e-05) (hash(x)=24809873) -36 train 11.515468 (lr=1.2937e-05) (hash(x)=23253252) -37 train 11.522291 (lr=1.3287e-05) (hash(x)=27852919) -38 train 11.403470 (lr=1.3636e-05) (hash(x)=23327497) -39 train 11.454610 (lr=1.3986e-05) (hash(x)=22512166) -40 train 11.466125 (lr=1.4336e-05) (hash(x)=22859419) -41 train 11.378222 (lr=1.4685e-05) (hash(x)=27620338) -42 train 11.366343 (lr=1.5035e-05) (hash(x)=26397837) -43 train 11.382690 (lr=1.5385e-05) (hash(x)=28092148) -44 train 11.338550 (lr=1.5734e-05) (hash(x)=24662703) -45 train 11.330945 (lr=1.6084e-05) (hash(x)=27938767) -46 train 11.357602 (lr=1.6434e-05) (hash(x)=26037988) -47 train 11.262557 (lr=1.6783e-05) (hash(x)=24732833) -48 train 11.281154 (lr=1.7133e-05) (hash(x)=25259526) -49 train 11.278589 (lr=1.7483e-05) (hash(x)=23200230) -50 val loss 11.2359 -50 val perplexity 75800.6250 -50 train 11.234625 (lr=1.7832e-05) (hash(x)=26721357) -51 train 11.228384 (lr=1.8182e-05) (hash(x)=22694718) -52 train 11.160281 (lr=1.8531e-05) (hash(x)=28066766) -53 train 11.160612 (lr=1.8881e-05) (hash(x)=23125151) -54 train 11.182727 (lr=1.9231e-05) (hash(x)=27193725) -55 train 11.089060 (lr=1.9580e-05) (hash(x)=25129410) -56 train 11.132565 (lr=1.9930e-05) (hash(x)=24263988) -57 train 11.002010 (lr=2.0280e-05) (hash(x)=23059154) -58 train 11.053906 (lr=2.0629e-05) (hash(x)=26063864) -59 train 11.107506 (lr=2.0979e-05) (hash(x)=27858570) -60 train 10.951129 (lr=2.1329e-05) (hash(x)=23874620) -61 train 10.948944 (lr=2.1678e-05) (hash(x)=22402617) -62 train 10.912158 (lr=2.2028e-05) (hash(x)=23600822) -63 train 10.787768 (lr=2.2378e-05) (hash(x)=26582391) -64 train 10.748692 (lr=2.2727e-05) (hash(x)=23225283) -65 train 10.844752 (lr=2.3077e-05) (hash(x)=26075451) -66 train 10.828836 (lr=2.3427e-05) (hash(x)=24723419) -67 train 10.745433 (lr=2.3776e-05) (hash(x)=27279806) -68 train 10.664749 (lr=2.4126e-05) (hash(x)=25870391) -69 train 10.711833 (lr=2.4476e-05) (hash(x)=26188136) -70 train 10.638556 (lr=2.4825e-05) (hash(x)=30373443) -71 train 10.702901 (lr=2.5175e-05) (hash(x)=26472336) -72 train 10.620476 (lr=2.5524e-05) (hash(x)=26651572) -73 train 10.583684 (lr=2.5874e-05) (hash(x)=26376212) -74 train 10.570152 (lr=2.6224e-05) (hash(x)=26733350) -75 train 10.645628 (lr=2.6573e-05) (hash(x)=28301589) -76 train 10.578505 (lr=2.6923e-05) (hash(x)=27599559) -77 train 10.459722 (lr=2.7273e-05) (hash(x)=28035221) -78 train 10.530845 (lr=2.7622e-05) (hash(x)=25016783) -79 train 10.551449 (lr=2.7972e-05) (hash(x)=27654289) -80 train 10.576966 (lr=2.8322e-05) (hash(x)=24597558) -81 train 10.529563 (lr=2.8671e-05) (hash(x)=21560904) -82 train 10.360724 (lr=2.9021e-05) (hash(x)=21983837) -83 train 10.394977 (lr=2.9371e-05) (hash(x)=24995715) -84 train 10.435885 (lr=2.9720e-05) (hash(x)=29876413) -85 train 10.391919 (lr=3.0070e-05) (hash(x)=23792508) -86 train 10.326373 (lr=3.0420e-05) (hash(x)=25509120) -87 train 10.464686 (lr=3.0769e-05) (hash(x)=26559876) -88 train 10.385931 (lr=3.1119e-05) (hash(x)=23569647) -89 train 10.412681 (lr=3.1469e-05) (hash(x)=25758852) -90 train 10.320277 (lr=3.1818e-05) (hash(x)=25706298) -91 train 10.193954 (lr=3.2168e-05) (hash(x)=28364895) -92 train 10.256680 (lr=3.2517e-05) (hash(x)=25304663) -93 train 10.331723 (lr=3.2867e-05) (hash(x)=25269299) -94 train 10.480817 (lr=3.3217e-05) (hash(x)=25870566) -95 train 10.295540 (lr=3.3566e-05) (hash(x)=21770329) -96 train 10.483542 (lr=3.3916e-05) (hash(x)=27595900) -97 train 10.496296 (lr=3.4266e-05) (hash(x)=24785397) -98 train 10.260822 (lr=3.4615e-05) (hash(x)=21521480) -99 train 10.297720 (lr=3.4965e-05) (hash(x)=24628606) -100 val loss 10.2878 -100 val perplexity 29372.0234 -100 train 10.325157 (lr=3.5315e-05) (hash(x)=24670150) -101 train 10.317250 (lr=3.5664e-05) (hash(x)=23181910) -102 train 10.178874 (lr=3.6014e-05) (hash(x)=22714991) -103 train 10.296162 (lr=3.6364e-05) (hash(x)=22723459) -104 train 10.220614 (lr=3.6713e-05) (hash(x)=21524316) -105 train 10.211928 (lr=3.7063e-05) (hash(x)=25506632) -106 train 10.036073 (lr=3.7413e-05) (hash(x)=21675672) -107 train 10.159115 (lr=3.7762e-05) (hash(x)=22897919) -108 train 10.187289 (lr=3.8112e-05) (hash(x)=23321631) -109 train 10.211846 (lr=3.8462e-05) (hash(x)=26546719) -110 train 10.698538 (lr=3.8811e-05) (hash(x)=31962348) -111 train 10.206660 (lr=3.9161e-05) (hash(x)=30338342) -112 train 10.179773 (lr=3.9510e-05) (hash(x)=23724471) -113 train 10.172065 (lr=3.9860e-05) (hash(x)=29175888) -114 train 10.253031 (lr=4.0210e-05) (hash(x)=23256716) -115 train 10.213995 (lr=4.0559e-05) (hash(x)=27063280) -116 train 10.171045 (lr=4.0909e-05) (hash(x)=31057659) -117 train 10.198005 (lr=4.1259e-05) (hash(x)=32915097) -118 train 10.142936 (lr=4.1608e-05) (hash(x)=28842717) -119 train 10.127883 (lr=4.1958e-05) (hash(x)=25678059) -120 train 10.128162 (lr=4.2308e-05) (hash(x)=21593510) -121 train 10.012947 (lr=4.2657e-05) (hash(x)=20083773) -122 train 10.051068 (lr=4.3007e-05) (hash(x)=23002820) -123 train 10.029854 (lr=4.3357e-05) (hash(x)=21853028) -124 train 10.057321 (lr=4.3706e-05) (hash(x)=26985625) -125 train 10.003461 (lr=4.4056e-05) (hash(x)=21808483) -126 train 10.118939 (lr=4.4406e-05) (hash(x)=28873251) -127 train 10.069919 (lr=4.4755e-05) (hash(x)=26109335) -128 train 10.072942 (lr=4.5105e-05) (hash(x)=26334674) -129 train 10.042968 (lr=4.5455e-05) (hash(x)=24916754) -130 train 10.090995 (lr=4.5804e-05) (hash(x)=25449624) -131 train 10.158465 (lr=4.6154e-05) (hash(x)=25334848) -132 train 10.189995 (lr=4.6503e-05) (hash(x)=27484863) -133 train 10.022834 (lr=4.6853e-05) (hash(x)=24917705) -134 train 9.871879 (lr=4.7203e-05) (hash(x)=25143449) -135 train 9.928664 (lr=4.7552e-05) (hash(x)=25044885) -136 train 10.136272 (lr=4.7902e-05) (hash(x)=27821028) -137 train 10.067722 (lr=4.8252e-05) (hash(x)=28747022) -138 train 10.019983 (lr=4.8601e-05) (hash(x)=27182888) -139 train 9.895723 (lr=4.8951e-05) (hash(x)=23678349) -140 train 10.033727 (lr=4.9301e-05) (hash(x)=23593235) -141 train 10.125788 (lr=4.9650e-05) (hash(x)=28529813) -142 train 10.189143 (lr=5.0000e-05) (hash(x)=32074661) -143 train 9.918561 (lr=5.0350e-05) (hash(x)=28870690) -144 train 9.986192 (lr=5.0699e-05) (hash(x)=27307705) -145 train 9.964223 (lr=5.1049e-05) (hash(x)=25044834) -146 train 10.023608 (lr=5.1399e-05) (hash(x)=23712023) -147 train 10.148318 (lr=5.1748e-05) (hash(x)=32982615) -148 train 10.025879 (lr=5.2098e-05) (hash(x)=30113660) -149 train 9.870357 (lr=5.2448e-05) (hash(x)=20970960) -150 val loss 9.8995 -150 val perplexity 19920.4062 -150 train 9.871345 (lr=5.2797e-05) (hash(x)=23132684) -151 train 10.465314 (lr=5.3147e-05) (hash(x)=35279941) -152 train 10.177967 (lr=5.3497e-05) (hash(x)=31227444) -153 train 9.844146 (lr=5.3846e-05) (hash(x)=25529472) -154 train 9.841098 (lr=5.4196e-05) (hash(x)=24350409) -155 train 10.059849 (lr=5.4545e-05) (hash(x)=26400041) -156 train 9.841269 (lr=5.4895e-05) (hash(x)=25262621) -157 train 9.829726 (lr=5.5245e-05) (hash(x)=24656138) -158 train 9.856341 (lr=5.5594e-05) (hash(x)=26803414) -159 train 9.813565 (lr=5.5944e-05) (hash(x)=25015923) -160 train 9.711070 (lr=5.6294e-05) (hash(x)=23581172) -161 train 9.776576 (lr=5.6643e-05) (hash(x)=22924885) -162 train 9.736098 (lr=5.6993e-05) (hash(x)=23414296) -163 train 9.950458 (lr=5.7343e-05) (hash(x)=24853586) -164 train 9.947077 (lr=5.7692e-05) (hash(x)=25000130) -165 train 9.820336 (lr=5.8042e-05) (hash(x)=27004780) -166 train 9.781570 (lr=5.8392e-05) (hash(x)=26148573) -167 train 9.748810 (lr=5.8741e-05) (hash(x)=26740855) -168 train 9.546291 (lr=5.9091e-05) (hash(x)=20965419) -169 train 9.673761 (lr=5.9441e-05) (hash(x)=23950114) -170 train 9.675754 (lr=5.9790e-05) (hash(x)=24951982) -171 train 9.751177 (lr=6.0140e-05) (hash(x)=24584116) -172 train 9.847892 (lr=6.0490e-05) (hash(x)=24378759) -173 train 9.671325 (lr=6.0839e-05) (hash(x)=25718516) -174 train 9.744838 (lr=6.1189e-05) (hash(x)=28424396) -175 train 9.651832 (lr=6.1538e-05) (hash(x)=22262151) -176 train 9.805852 (lr=6.1888e-05) (hash(x)=26438412) -177 train 9.664753 (lr=6.2238e-05) (hash(x)=23025303) -178 train 9.560608 (lr=6.2587e-05) (hash(x)=24190770) -179 train 9.625232 (lr=6.2937e-05) (hash(x)=26627860) -180 train 9.662766 (lr=6.3287e-05) (hash(x)=23663439) -181 train 9.466022 (lr=6.3636e-05) (hash(x)=23700532) -182 train 9.537752 (lr=6.3986e-05) (hash(x)=23075676) -183 train 9.570643 (lr=6.4336e-05) (hash(x)=26621834) -184 train 9.575083 (lr=6.4685e-05) (hash(x)=29426269) -185 train 9.570395 (lr=6.5035e-05) (hash(x)=21821465) -186 train 9.594153 (lr=6.5385e-05) (hash(x)=26506130) -187 train 9.712993 (lr=6.5734e-05) (hash(x)=26930630) -188 train 9.567522 (lr=6.6084e-05) (hash(x)=22993793) -189 train 9.571979 (lr=6.6434e-05) (hash(x)=19557946) -190 train 9.679810 (lr=6.6783e-05) (hash(x)=23572891) -191 train 9.472575 (lr=6.7133e-05) (hash(x)=23234741) -192 train 9.576774 (lr=6.7483e-05) (hash(x)=25547951) -193 train 9.577289 (lr=6.7832e-05) (hash(x)=26713563) -194 train 9.555125 (lr=6.8182e-05) (hash(x)=25913622) -195 train 9.611600 (lr=6.8531e-05) (hash(x)=28102443) -196 train 9.545808 (lr=6.8881e-05) (hash(x)=23093351) -197 train 9.826259 (lr=6.9231e-05) (hash(x)=31689122) -198 train 9.761626 (lr=6.9580e-05) (hash(x)=32252517) -199 train 9.535264 (lr=6.9930e-05) (hash(x)=25470563) -200 val loss 9.5248 -200 val perplexity 13694.7764 -200 train 9.604159 (lr=7.0280e-05) (hash(x)=25597614) -201 train 9.595760 (lr=7.0629e-05) (hash(x)=23757479) -202 train 9.419785 (lr=7.0979e-05) (hash(x)=25140048) -203 train 9.507541 (lr=7.1329e-05) (hash(x)=28282861) -204 train 9.427428 (lr=7.1678e-05) (hash(x)=24754885) -205 train 9.579193 (lr=7.2028e-05) (hash(x)=26731964) -206 train 9.612823 (lr=7.2378e-05) (hash(x)=26660561) -207 train 9.460600 (lr=7.2727e-05) (hash(x)=21799102) -208 train 9.445565 (lr=7.3077e-05) (hash(x)=27025986) -209 train 9.366461 (lr=7.3427e-05) (hash(x)=24672077) -210 train 9.407405 (lr=7.3776e-05) (hash(x)=25322984) -211 train 9.360783 (lr=7.4126e-05) (hash(x)=23471769) -212 train 9.127860 (lr=7.4476e-05) (hash(x)=20766491) -213 train 9.645239 (lr=7.4825e-05) (hash(x)=24058931) -214 train 9.333905 (lr=7.5175e-05) (hash(x)=23872843) -215 train 9.339537 (lr=7.5524e-05) (hash(x)=23275479) -216 train 9.452003 (lr=7.5874e-05) (hash(x)=24914695) -217 train 9.349166 (lr=7.6224e-05) (hash(x)=24364396) -218 train 9.338374 (lr=7.6573e-05) (hash(x)=27986474) -219 train 9.414897 (lr=7.6923e-05) (hash(x)=24473581) -220 train 9.345592 (lr=7.7273e-05) (hash(x)=22974689) -221 train 9.374198 (lr=7.7622e-05) (hash(x)=23774644) -222 train 9.422793 (lr=7.7972e-05) (hash(x)=23245327) -223 train 9.256053 (lr=7.8322e-05) (hash(x)=22091862) -224 train 9.176079 (lr=7.8671e-05) (hash(x)=24362839) -225 train 9.110301 (lr=7.9021e-05) (hash(x)=25482303) -226 train 9.321770 (lr=7.9371e-05) (hash(x)=24911853) -227 train 9.270859 (lr=7.9720e-05) (hash(x)=26018202) -228 train 9.169250 (lr=8.0070e-05) (hash(x)=26124495) -229 train 9.211135 (lr=8.0420e-05) (hash(x)=24560096) -230 train 9.186211 (lr=8.0769e-05) (hash(x)=24695331) -231 train 9.132274 (lr=8.1119e-05) (hash(x)=17430373) -232 train 9.085945 (lr=8.1469e-05) (hash(x)=21813345) -233 train 8.806046 (lr=8.1818e-05) (hash(x)=20098681) -234 train 8.950845 (lr=8.2168e-05) (hash(x)=25095928) -235 train 9.060810 (lr=8.2517e-05) (hash(x)=24078083) -236 train 9.132156 (lr=8.2867e-05) (hash(x)=22901505) -237 train 9.080871 (lr=8.3217e-05) (hash(x)=26595592) -238 train 9.063289 (lr=8.3566e-05) (hash(x)=27663196) -239 train 8.955091 (lr=8.3916e-05) (hash(x)=22954861) -240 train 9.127967 (lr=8.4266e-05) (hash(x)=30159234) -241 train 9.088605 (lr=8.4615e-05) (hash(x)=29294271) -242 train 9.086323 (lr=8.4965e-05) (hash(x)=23728322) -243 train 9.099353 (lr=8.5315e-05) (hash(x)=28695016) -244 train 9.150974 (lr=8.5664e-05) (hash(x)=26702728) -245 train 8.961917 (lr=8.6014e-05) (hash(x)=27100115) -246 train 8.964152 (lr=8.6364e-05) (hash(x)=25082752) -247 train 9.220180 (lr=8.6713e-05) (hash(x)=26671799) -248 train 9.064783 (lr=8.7063e-05) (hash(x)=23718946) -249 train 9.081349 (lr=8.7413e-05) (hash(x)=28137394) -250 val loss 8.9596 -250 val perplexity 7782.2695 -250 train 9.077417 (lr=8.7762e-05) (hash(x)=23893495) -251 train 8.992588 (lr=8.8112e-05) (hash(x)=23166092) -252 train 8.871581 (lr=8.8462e-05) (hash(x)=25907665) -253 train 8.828327 (lr=8.8811e-05) (hash(x)=25885986) -254 train 8.896298 (lr=8.9161e-05) (hash(x)=25262712) -255 train 8.848207 (lr=8.9510e-05) (hash(x)=26924723) -256 train 8.993649 (lr=8.9860e-05) (hash(x)=28744736) -257 train 8.847756 (lr=9.0210e-05) (hash(x)=26140590) -258 train 8.974219 (lr=9.0559e-05) (hash(x)=25780449) -259 train 8.848479 (lr=9.0909e-05) (hash(x)=25138659) -260 train 8.953708 (lr=9.1259e-05) (hash(x)=27244046) -261 train 8.914029 (lr=9.1608e-05) (hash(x)=27224685) -262 train 8.985224 (lr=9.1958e-05) (hash(x)=28274477) -263 train 8.736698 (lr=9.2308e-05) (hash(x)=23557495) -264 train 8.816499 (lr=9.2657e-05) (hash(x)=24680596) -265 train 8.736740 (lr=9.3007e-05) (hash(x)=23928957) -266 train 8.672368 (lr=9.3357e-05) (hash(x)=23761390) -267 train 8.739343 (lr=9.3706e-05) (hash(x)=25288123) -268 train 8.795394 (lr=9.4056e-05) (hash(x)=28705502) -269 train 8.516548 (lr=9.4406e-05) (hash(x)=23246294) -270 train 8.496886 (lr=9.4755e-05) (hash(x)=28639079) -271 train 8.448086 (lr=9.5105e-05) (hash(x)=27804380) -272 train 8.651340 (lr=9.5455e-05) (hash(x)=24172235) -273 train 8.714636 (lr=9.5804e-05) (hash(x)=23089140) -274 train 8.884443 (lr=9.6154e-05) (hash(x)=27163701) -275 train 8.918172 (lr=9.6503e-05) (hash(x)=26993263) -276 train 8.678073 (lr=9.6853e-05) (hash(x)=28224233) -277 train 8.952127 (lr=9.7203e-05) (hash(x)=27397203) -278 train 8.753987 (lr=9.7552e-05) (hash(x)=27923882) -279 train 8.683087 (lr=9.7902e-05) (hash(x)=26654908) -280 train 8.695821 (lr=9.8252e-05) (hash(x)=24213147) -281 train 8.468570 (lr=9.8601e-05) (hash(x)=21965022) -282 train 8.531305 (lr=9.8951e-05) (hash(x)=25465685) -283 train 8.645694 (lr=9.9301e-05) (hash(x)=27347722) -284 train 8.627509 (lr=9.9650e-05) (hash(x)=26732050) -285 train 8.698267 (lr=1.0000e-04) (hash(x)=28314127) -286 train 8.538644 (lr=1.0000e-04) (hash(x)=21471186) -287 train 8.501497 (lr=1.0000e-04) (hash(x)=23627518) -288 train 8.473663 (lr=9.9998e-05) (hash(x)=20870353) -289 train 8.446799 (lr=9.9996e-05) (hash(x)=25024764) -290 train 8.277788 (lr=9.9993e-05) (hash(x)=20683822) -291 train 8.557170 (lr=9.9989e-05) (hash(x)=21768671) -292 train 8.475009 (lr=9.9984e-05) (hash(x)=25557309) -293 train 8.638965 (lr=9.9979e-05) (hash(x)=25076667) -294 train 8.358286 (lr=9.9972e-05) (hash(x)=23765822) -295 train 8.385844 (lr=9.9965e-05) (hash(x)=21889990) -296 train 8.436145 (lr=9.9956e-05) (hash(x)=26339893) -297 train 8.245259 (lr=9.9947e-05) (hash(x)=20932794) -298 train 8.344548 (lr=9.9937e-05) (hash(x)=21750070) -299 train 8.487418 (lr=9.9926e-05) (hash(x)=23665838) -300 val loss 8.4177 -300 val perplexity 4526.3418 -300 train 9.056525 (lr=9.9915e-05) (hash(x)=32888061) -301 train 8.848121 (lr=9.9902e-05) (hash(x)=30223582) -302 train 8.573112 (lr=9.9889e-05) (hash(x)=26908418) -303 train 8.121158 (lr=9.9874e-05) (hash(x)=22528001) -304 train 8.542716 (lr=9.9859e-05) (hash(x)=27452187) -305 train 8.435355 (lr=9.9843e-05) (hash(x)=25181641) -306 train 8.427194 (lr=9.9826e-05) (hash(x)=25546593) -307 train 8.348720 (lr=9.9808e-05) (hash(x)=22487328) -308 train 8.511443 (lr=9.9789e-05) (hash(x)=27804274) -309 train 8.556011 (lr=9.9770e-05) (hash(x)=26544630) -310 train 8.695384 (lr=9.9749e-05) (hash(x)=27738934) -311 train 8.456397 (lr=9.9728e-05) (hash(x)=29248942) -312 train 8.310699 (lr=9.9706e-05) (hash(x)=25103452) -313 train 8.501703 (lr=9.9683e-05) (hash(x)=25052066) -314 train 8.388857 (lr=9.9659e-05) (hash(x)=24481302) -315 train 8.291567 (lr=9.9634e-05) (hash(x)=23543273) -316 train 8.296641 (lr=9.9609e-05) (hash(x)=25608244) -317 train 8.470215 (lr=9.9582e-05) (hash(x)=27451288) -318 train 8.156415 (lr=9.9555e-05) (hash(x)=22806491) -319 train 8.321669 (lr=9.9526e-05) (hash(x)=25533417) -320 train 8.294555 (lr=9.9497e-05) (hash(x)=24557997) -321 train 8.315181 (lr=9.9467e-05) (hash(x)=24432899) -322 train 8.417811 (lr=9.9437e-05) (hash(x)=27583287) -323 train 8.258604 (lr=9.9405e-05) (hash(x)=25552036) -324 train 8.139359 (lr=9.9372e-05) (hash(x)=24201868) -325 train 8.454796 (lr=9.9339e-05) (hash(x)=28149782) -326 train 8.278832 (lr=9.9305e-05) (hash(x)=25529698) -327 train 8.080646 (lr=9.9270e-05) (hash(x)=20612533) -328 train 8.121573 (lr=9.9234e-05) (hash(x)=20699000) -329 train 8.043709 (lr=9.9197e-05) (hash(x)=19774173) -330 train 8.133275 (lr=9.9159e-05) (hash(x)=21681646) -331 train 7.999214 (lr=9.9121e-05) (hash(x)=20216795) -332 train 8.467620 (lr=9.9081e-05) (hash(x)=27697998) -333 train 8.301514 (lr=9.9041e-05) (hash(x)=25896435) -334 train 8.185733 (lr=9.9000e-05) (hash(x)=21585310) -335 train 8.216571 (lr=9.8958e-05) (hash(x)=24677740) -336 train 8.182680 (lr=9.8915e-05) (hash(x)=22027900) -337 train 8.120502 (lr=9.8872e-05) (hash(x)=21835643) -338 train 8.044324 (lr=9.8827e-05) (hash(x)=20558462) -339 train 8.071731 (lr=9.8782e-05) (hash(x)=16722715) -340 train 8.144255 (lr=9.8736e-05) (hash(x)=20972655) -341 train 8.406032 (lr=9.8689e-05) (hash(x)=26303975) -342 train 8.172811 (lr=9.8641e-05) (hash(x)=22938170) -343 train 8.181975 (lr=9.8592e-05) (hash(x)=25347203) -344 train 8.225675 (lr=9.8543e-05) (hash(x)=27398686) -345 train 8.164271 (lr=9.8492e-05) (hash(x)=25973417) -346 train 8.167585 (lr=9.8441e-05) (hash(x)=26918389) -347 train 8.122058 (lr=9.8389e-05) (hash(x)=24272489) -348 train 8.096065 (lr=9.8336e-05) (hash(x)=25593714) -349 train 8.164804 (lr=9.8282e-05) (hash(x)=29260846) -350 val loss 8.1481 -350 val perplexity 3456.6528 -350 train 8.266328 (lr=9.8228e-05) (hash(x)=27951602) -351 train 8.332129 (lr=9.8172e-05) (hash(x)=28922363) -352 train 8.225944 (lr=9.8116e-05) (hash(x)=27210734) -353 train 8.224375 (lr=9.8059e-05) (hash(x)=26322572) -354 train 8.169836 (lr=9.8001e-05) (hash(x)=27084665) -355 train 8.109509 (lr=9.7942e-05) (hash(x)=25510798) -356 train 8.182357 (lr=9.7882e-05) (hash(x)=24970921) -357 train 8.264482 (lr=9.7822e-05) (hash(x)=24138948) -358 train 8.162926 (lr=9.7761e-05) (hash(x)=24790211) -359 train 8.128478 (lr=9.7699e-05) (hash(x)=25631397) -360 train 8.054370 (lr=9.7636e-05) (hash(x)=23226625) -361 train 8.045502 (lr=9.7572e-05) (hash(x)=24001903) -362 train 8.083979 (lr=9.7507e-05) (hash(x)=24587948) -363 train 7.828613 (lr=9.7442e-05) (hash(x)=21333676) -364 train 7.725572 (lr=9.7376e-05) (hash(x)=23673779) -365 train 8.865267 (lr=9.7309e-05) (hash(x)=30770484) -366 train 8.171538 (lr=9.7241e-05) (hash(x)=26564899) -367 train 8.145097 (lr=9.7172e-05) (hash(x)=26237983) -368 train 7.899007 (lr=9.7103e-05) (hash(x)=23764356) -369 train 8.164248 (lr=9.7032e-05) (hash(x)=26205744) -370 train 8.006961 (lr=9.6961e-05) (hash(x)=19208770) -371 train 8.097083 (lr=9.6889e-05) (hash(x)=25976502) -372 train 7.928078 (lr=9.6817e-05) (hash(x)=23983933) -373 train 7.974586 (lr=9.6743e-05) (hash(x)=24080636) -374 train 8.095126 (lr=9.6669e-05) (hash(x)=24404047) -375 train 8.077460 (lr=9.6593e-05) (hash(x)=24742645) -376 train 8.054320 (lr=9.6518e-05) (hash(x)=24159600) -377 train 8.243296 (lr=9.6441e-05) (hash(x)=28677257) -378 train 8.082477 (lr=9.6363e-05) (hash(x)=25604111) -379 train 7.968868 (lr=9.6285e-05) (hash(x)=27086333) -380 train 7.977652 (lr=9.6206e-05) (hash(x)=25188207) -381 train 8.109197 (lr=9.6126e-05) (hash(x)=27855233) -382 train 7.794618 (lr=9.6045e-05) (hash(x)=19470039) -383 train 8.050038 (lr=9.5963e-05) (hash(x)=26157660) -384 train 7.978778 (lr=9.5881e-05) (hash(x)=25291570) -385 train 7.947104 (lr=9.5798e-05) (hash(x)=25046062) -386 train 8.197604 (lr=9.5714e-05) (hash(x)=27020337) -387 train 7.844612 (lr=9.5629e-05) (hash(x)=23616370) -388 train 7.588484 (lr=9.5544e-05) (hash(x)=19113218) -389 train 7.907253 (lr=9.5457e-05) (hash(x)=24302232) -390 train 7.906401 (lr=9.5370e-05) (hash(x)=22188949) -391 train 7.708833 (lr=9.5282e-05) (hash(x)=22582169) -392 train 7.944036 (lr=9.5194e-05) (hash(x)=24700570) -393 train 7.811014 (lr=9.5104e-05) (hash(x)=22773833) -394 train 7.711431 (lr=9.5014e-05) (hash(x)=21875928) -395 train 7.945917 (lr=9.4923e-05) (hash(x)=26233189) -396 train 7.800710 (lr=9.4831e-05) (hash(x)=24321467) -397 train 8.018307 (lr=9.4739e-05) (hash(x)=26431507) -398 train 8.109365 (lr=9.4646e-05) (hash(x)=28690877) -399 train 8.068495 (lr=9.4551e-05) (hash(x)=26431960) -400 val loss 7.9516 -400 val perplexity 2840.1738 -400 train 7.932497 (lr=9.4457e-05) (hash(x)=24580300) -401 train 7.918422 (lr=9.4361e-05) (hash(x)=25112360) -402 train 7.957445 (lr=9.4265e-05) (hash(x)=27597243) -403 train 8.387721 (lr=9.4168e-05) (hash(x)=30707498) -404 train 8.018311 (lr=9.4070e-05) (hash(x)=28485465) -405 train 7.759810 (lr=9.3971e-05) (hash(x)=22586447) -406 train 7.913458 (lr=9.3872e-05) (hash(x)=23175270) -407 train 7.953313 (lr=9.3772e-05) (hash(x)=25716176) -408 train 8.028157 (lr=9.3671e-05) (hash(x)=26861373) -409 train 8.156397 (lr=9.3569e-05) (hash(x)=25118971) -410 train 7.659088 (lr=9.3467e-05) (hash(x)=19829066) -411 train 7.802091 (lr=9.3364e-05) (hash(x)=26256420) -412 train 8.017024 (lr=9.3260e-05) (hash(x)=27796153) -413 train 7.781218 (lr=9.3155e-05) (hash(x)=22633318) -414 train 7.823593 (lr=9.3050e-05) (hash(x)=22589383) -415 train 8.207155 (lr=9.2944e-05) (hash(x)=28019788) -416 train 8.138515 (lr=9.2837e-05) (hash(x)=28970440) -417 train 7.796050 (lr=9.2729e-05) (hash(x)=27396089) -418 train 7.830936 (lr=9.2621e-05) (hash(x)=21183513) -419 train 7.847047 (lr=9.2512e-05) (hash(x)=23510110) -420 train 8.045158 (lr=9.2402e-05) (hash(x)=28833467) -421 train 7.915658 (lr=9.2292e-05) (hash(x)=23646926) -422 train 7.881616 (lr=9.2181e-05) (hash(x)=24697272) -423 train 7.656999 (lr=9.2069e-05) (hash(x)=20382963) -424 train 7.710730 (lr=9.1956e-05) (hash(x)=23467595) -425 train 7.831064 (lr=9.1843e-05) (hash(x)=24304768) -426 train 7.756701 (lr=9.1729e-05) (hash(x)=21392328) -427 train 7.812335 (lr=9.1614e-05) (hash(x)=25339466) -428 train 7.718620 (lr=9.1499e-05) (hash(x)=22092542) -429 train 7.827802 (lr=9.1382e-05) (hash(x)=22088696) -430 train 7.739713 (lr=9.1266e-05) (hash(x)=22184471) -431 train 7.843549 (lr=9.1148e-05) (hash(x)=24489647) -432 train 7.869493 (lr=9.1030e-05) (hash(x)=26794132) -433 train 7.680702 (lr=9.0911e-05) (hash(x)=22940357) -434 train 7.871318 (lr=9.0791e-05) (hash(x)=23719522) -435 train 7.681929 (lr=9.0671e-05) (hash(x)=22927699) -436 train 7.856727 (lr=9.0550e-05) (hash(x)=26068576) -437 train 7.992668 (lr=9.0428e-05) (hash(x)=27631132) -438 train 7.863513 (lr=9.0306e-05) (hash(x)=26739991) -439 train 7.912216 (lr=9.0182e-05) (hash(x)=25128502) -440 train 8.011527 (lr=9.0059e-05) (hash(x)=25657260) -441 train 7.745845 (lr=8.9934e-05) (hash(x)=23576982) -442 train 7.939442 (lr=8.9809e-05) (hash(x)=27117886) -443 train 7.928978 (lr=8.9683e-05) (hash(x)=25808969) -444 train 7.857088 (lr=8.9557e-05) (hash(x)=24738238) -445 train 7.818183 (lr=8.9430e-05) (hash(x)=23429962) -446 train 7.939254 (lr=8.9302e-05) (hash(x)=25075165) -447 train 7.791748 (lr=8.9173e-05) (hash(x)=25231390) -448 train 7.608206 (lr=8.9044e-05) (hash(x)=22055054) -449 train 7.839536 (lr=8.8914e-05) (hash(x)=25395441) -450 val loss 7.8375 -450 val perplexity 2533.8687 -450 train 7.875294 (lr=8.8784e-05) (hash(x)=25863209) -451 train 7.587674 (lr=8.8653e-05) (hash(x)=21154388) -452 train 7.572248 (lr=8.8521e-05) (hash(x)=21600876) -453 train 7.700493 (lr=8.8388e-05) (hash(x)=24278611) -454 train 7.586690 (lr=8.8255e-05) (hash(x)=23221720) -455 train 7.770946 (lr=8.8122e-05) (hash(x)=22708977) -456 train 8.341041 (lr=8.7987e-05) (hash(x)=23637758) -457 train 8.048926 (lr=8.7852e-05) (hash(x)=28228490) -458 train 8.071681 (lr=8.7717e-05) (hash(x)=28638071) -459 train 7.924314 (lr=8.7580e-05) (hash(x)=27258353) -460 train 7.894745 (lr=8.7444e-05) (hash(x)=26604728) -461 train 7.778495 (lr=8.7306e-05) (hash(x)=23252199) -462 train 7.895186 (lr=8.7168e-05) (hash(x)=26441427) -463 train 7.665602 (lr=8.7029e-05) (hash(x)=24364920) -464 train 7.837111 (lr=8.6890e-05) (hash(x)=25623792) -465 train 7.738840 (lr=8.6750e-05) (hash(x)=23283905) -466 train 8.109988 (lr=8.6609e-05) (hash(x)=26025267) -467 train 7.872530 (lr=8.6468e-05) (hash(x)=27243972) -468 train 8.290081 (lr=8.6326e-05) (hash(x)=30449945) -469 train 8.003036 (lr=8.6184e-05) (hash(x)=28113043) -470 train 7.780609 (lr=8.6041e-05) (hash(x)=25182521) -471 train 7.817902 (lr=8.5897e-05) (hash(x)=24932925) -472 train 7.644380 (lr=8.5753e-05) (hash(x)=20353098) -473 train 7.578041 (lr=8.5608e-05) (hash(x)=19001259) -474 train 8.087781 (lr=8.5462e-05) (hash(x)=27585685) -475 train 7.830338 (lr=8.5316e-05) (hash(x)=26371091) -476 train 7.654768 (lr=8.5170e-05) (hash(x)=24891798) -477 train 7.709324 (lr=8.5022e-05) (hash(x)=24258817) -478 train 7.795334 (lr=8.4875e-05) (hash(x)=24330263) -479 train 7.856529 (lr=8.4726e-05) (hash(x)=26913684) -480 train 7.841763 (lr=8.4577e-05) (hash(x)=26338455) -481 train 8.063830 (lr=8.4428e-05) (hash(x)=27753043) -482 train 7.799088 (lr=8.4278e-05) (hash(x)=26123289) -483 train 7.901208 (lr=8.4127e-05) (hash(x)=29239611) -484 train 7.814688 (lr=8.3976e-05) (hash(x)=26553003) -485 train 7.785366 (lr=8.3824e-05) (hash(x)=22984557) -486 train 7.324088 (lr=8.3672e-05) (hash(x)=16947491) -487 train 7.355917 (lr=8.3519e-05) (hash(x)=18017792) -488 train 7.525796 (lr=8.3366e-05) (hash(x)=19918608) -489 train 7.947198 (lr=8.3212e-05) (hash(x)=23374526) -490 train 7.826778 (lr=8.3057e-05) (hash(x)=25009505) -491 train 7.776108 (lr=8.2902e-05) (hash(x)=27574089) -492 train 7.910175 (lr=8.2746e-05) (hash(x)=24122664) -493 train 7.790071 (lr=8.2590e-05) (hash(x)=26154906) -494 train 7.772087 (lr=8.2434e-05) (hash(x)=25192767) -495 train 7.894056 (lr=8.2276e-05) (hash(x)=28613882) -496 train 7.769869 (lr=8.2119e-05) (hash(x)=23547219) -497 train 7.776222 (lr=8.1960e-05) (hash(x)=25272182) -498 train 7.646951 (lr=8.1801e-05) (hash(x)=24992761) -499 train 7.917989 (lr=8.1642e-05) (hash(x)=26981914) -500 val loss 7.7764 -500 val perplexity 2383.6409 -500 train 7.699880 (lr=8.1482e-05) (hash(x)=22051933) -501 train 7.886009 (lr=8.1322e-05) (hash(x)=24232348) -502 train 7.874481 (lr=8.1161e-05) (hash(x)=23158331) -503 train 7.679631 (lr=8.1000e-05) (hash(x)=22652243) -504 train 7.721078 (lr=8.0838e-05) (hash(x)=23805602) -505 train 8.005980 (lr=8.0676e-05) (hash(x)=25411991) -506 train 7.322259 (lr=8.0513e-05) (hash(x)=18827215) -507 train 7.044342 (lr=8.0349e-05) (hash(x)=15446025) -508 train 7.382634 (lr=8.0186e-05) (hash(x)=20516263) -509 train 7.870894 (lr=8.0021e-05) (hash(x)=27846176) -510 train 7.601924 (lr=7.9856e-05) (hash(x)=23342449) -511 train 7.827890 (lr=7.9691e-05) (hash(x)=27194521) -512 train 7.650694 (lr=7.9525e-05) (hash(x)=23008284) -513 train 8.081702 (lr=7.9359e-05) (hash(x)=29430001) -514 train 7.434021 (lr=7.9192e-05) (hash(x)=22579319) -515 train 7.500001 (lr=7.9025e-05) (hash(x)=25264518) -516 train 7.819285 (lr=7.8858e-05) (hash(x)=25359075) -517 train 7.784191 (lr=7.8689e-05) (hash(x)=25568956) -518 train 8.069611 (lr=7.8521e-05) (hash(x)=32004108) -519 train 7.727131 (lr=7.8352e-05) (hash(x)=24936836) -520 train 7.885005 (lr=7.8182e-05) (hash(x)=27263338) -521 train 7.929016 (lr=7.8012e-05) (hash(x)=27452099) -522 train 7.817587 (lr=7.7842e-05) (hash(x)=25965406) -523 train 8.010417 (lr=7.7671e-05) (hash(x)=28197282) -524 train 7.704326 (lr=7.7500e-05) (hash(x)=22466209) -525 train 7.758869 (lr=7.7328e-05) (hash(x)=22931889) -526 train 7.903917 (lr=7.7156e-05) (hash(x)=26903920) -527 train 7.798918 (lr=7.6984e-05) (hash(x)=24765578) -528 train 7.867216 (lr=7.6811e-05) (hash(x)=27811359) -529 train 7.680454 (lr=7.6637e-05) (hash(x)=25078649) -530 train 7.852741 (lr=7.6463e-05) (hash(x)=25572416) -531 train 7.930327 (lr=7.6289e-05) (hash(x)=27448185) -532 train 7.998192 (lr=7.6115e-05) (hash(x)=25923719) -533 train 7.787836 (lr=7.5940e-05) (hash(x)=24804856) -534 train 7.936095 (lr=7.5764e-05) (hash(x)=23207829) -535 train 7.811322 (lr=7.5588e-05) (hash(x)=23107416) -536 train 7.666179 (lr=7.5412e-05) (hash(x)=26739531) -537 train 7.660806 (lr=7.5235e-05) (hash(x)=24960796) -538 train 7.804595 (lr=7.5058e-05) (hash(x)=24667802) -539 train 7.913820 (lr=7.4881e-05) (hash(x)=26755138) -540 train 7.780447 (lr=7.4703e-05) (hash(x)=25537132) -541 train 7.680098 (lr=7.4525e-05) (hash(x)=24542526) -542 train 7.506296 (lr=7.4346e-05) (hash(x)=21296355) -543 train 8.003779 (lr=7.4167e-05) (hash(x)=29314255) -544 train 7.773300 (lr=7.3988e-05) (hash(x)=26001799) -545 train 7.819228 (lr=7.3808e-05) (hash(x)=27347755) -546 train 7.799649 (lr=7.3628e-05) (hash(x)=25107798) -547 train 7.606548 (lr=7.3448e-05) (hash(x)=22112669) -548 train 7.595827 (lr=7.3267e-05) (hash(x)=21897967) -549 train 7.688321 (lr=7.3086e-05) (hash(x)=25161929) -550 val loss 7.7535 -550 val perplexity 2329.8088 -550 train 7.656169 (lr=7.2904e-05) (hash(x)=27465106) -551 train 7.768997 (lr=7.2723e-05) (hash(x)=24013079) -552 train 7.697281 (lr=7.2540e-05) (hash(x)=23142015) -553 train 7.756181 (lr=7.2358e-05) (hash(x)=26768629) -554 train 7.774313 (lr=7.2175e-05) (hash(x)=26393383) -555 train 7.589027 (lr=7.1992e-05) (hash(x)=22537194) -556 train 7.841833 (lr=7.1808e-05) (hash(x)=24046036) -557 train 7.793911 (lr=7.1624e-05) (hash(x)=24974360) -558 train 8.006940 (lr=7.1440e-05) (hash(x)=28379928) -559 train 7.764841 (lr=7.1256e-05) (hash(x)=25322001) -560 train 7.861572 (lr=7.1071e-05) (hash(x)=26622031) -561 train 7.510855 (lr=7.0886e-05) (hash(x)=20562247) -562 train 7.904912 (lr=7.0701e-05) (hash(x)=27381885) -563 train 8.056027 (lr=7.0515e-05) (hash(x)=27028126) -564 train 7.863399 (lr=7.0329e-05) (hash(x)=28882928) -565 train 7.805505 (lr=7.0143e-05) (hash(x)=25666355) -566 train 7.898448 (lr=6.9956e-05) (hash(x)=24330810) -567 train 7.811962 (lr=6.9769e-05) (hash(x)=26690440) -568 train 7.675626 (lr=6.9582e-05) (hash(x)=22923592) -569 train 7.805740 (lr=6.9394e-05) (hash(x)=27348418) -570 train 7.830551 (lr=6.9207e-05) (hash(x)=28849848) -571 train 7.883322 (lr=6.9019e-05) (hash(x)=26967331) -572 train 7.620302 (lr=6.8830e-05) (hash(x)=22831467) -573 train 7.725523 (lr=6.8642e-05) (hash(x)=24765121) -574 train 7.720890 (lr=6.8453e-05) (hash(x)=24331857) -575 train 7.549108 (lr=6.8264e-05) (hash(x)=22598512) -576 train 7.742252 (lr=6.8075e-05) (hash(x)=25149353) -577 train 7.620255 (lr=6.7885e-05) (hash(x)=23725598) -578 train 7.853076 (lr=6.7695e-05) (hash(x)=26449557) -579 train 7.865144 (lr=6.7505e-05) (hash(x)=24697985) -580 train 7.835664 (lr=6.7315e-05) (hash(x)=26923059) -581 train 7.661486 (lr=6.7124e-05) (hash(x)=25201962) -582 train 7.332506 (lr=6.6933e-05) (hash(x)=20931520) -583 train 7.386402 (lr=6.6742e-05) (hash(x)=18473911) -584 train 7.551183 (lr=6.6551e-05) (hash(x)=21306267) -585 train 7.822694 (lr=6.6360e-05) (hash(x)=25982840) -586 train 7.708252 (lr=6.6168e-05) (hash(x)=25364874) -587 train 7.611151 (lr=6.5976e-05) (hash(x)=23172124) -588 train 7.894548 (lr=6.5784e-05) (hash(x)=27876897) -589 train 8.491273 (lr=6.5592e-05) (hash(x)=34646114) -590 train 8.830293 (lr=6.5399e-05) (hash(x)=35153576) -591 train 7.644284 (lr=6.5206e-05) (hash(x)=22322442) -592 train 7.912062 (lr=6.5013e-05) (hash(x)=27907331) -593 train 7.819154 (lr=6.4820e-05) (hash(x)=26211794) -594 train 8.042597 (lr=6.4627e-05) (hash(x)=29291512) -595 train 8.290645 (lr=6.4433e-05) (hash(x)=29659121) -596 train 8.093459 (lr=6.4240e-05) (hash(x)=29674399) -597 train 7.713676 (lr=6.4046e-05) (hash(x)=23538306) -598 train 7.842334 (lr=6.3852e-05) (hash(x)=21991524) -599 train 7.792507 (lr=6.3658e-05) (hash(x)=26324153) -600 val loss 7.7565 -600 val perplexity 2336.6699 -600 train 7.745420 (lr=6.3463e-05) (hash(x)=23712082) -601 train 7.777402 (lr=6.3269e-05) (hash(x)=24910403) -602 train 7.742374 (lr=6.3074e-05) (hash(x)=26737205) -603 train 7.898977 (lr=6.2879e-05) (hash(x)=26939970) -604 train 7.764542 (lr=6.2684e-05) (hash(x)=27651943) -605 train 8.055849 (lr=6.2489e-05) (hash(x)=27515446) -606 train 7.935135 (lr=6.2294e-05) (hash(x)=26753129) -607 train 7.646013 (lr=6.2098e-05) (hash(x)=23446058) -608 train 7.926376 (lr=6.1903e-05) (hash(x)=27587849) -609 train 7.827156 (lr=6.1707e-05) (hash(x)=25308253) -610 train 7.841098 (lr=6.1511e-05) (hash(x)=26615098) -611 train 7.860497 (lr=6.1315e-05) (hash(x)=29981801) -612 train 7.963656 (lr=6.1119e-05) (hash(x)=29592345) -613 train 7.973532 (lr=6.0923e-05) (hash(x)=23470413) -614 train 7.877948 (lr=6.0726e-05) (hash(x)=24742370) -615 train 7.889167 (lr=6.0530e-05) (hash(x)=24843741) -616 train 7.780878 (lr=6.0333e-05) (hash(x)=25192548) -617 train 7.854270 (lr=6.0137e-05) (hash(x)=27176996) -618 train 7.716012 (lr=5.9940e-05) (hash(x)=23964552) -619 train 7.691032 (lr=5.9743e-05) (hash(x)=22855363) -620 train 7.661778 (lr=5.9546e-05) (hash(x)=26332996) -621 train 7.504344 (lr=5.9349e-05) (hash(x)=22960957) -622 train 7.838528 (lr=5.9152e-05) (hash(x)=22752597) -623 train 7.738686 (lr=5.8955e-05) (hash(x)=23197102) -624 train 8.208415 (lr=5.8758e-05) (hash(x)=27383319) -625 train 7.760849 (lr=5.8560e-05) (hash(x)=26132276) -626 train 7.624730 (lr=5.8363e-05) (hash(x)=19810497) -627 train 7.732967 (lr=5.8165e-05) (hash(x)=25704919) -628 train 7.875746 (lr=5.7968e-05) (hash(x)=27174264) -629 train 8.015245 (lr=5.7770e-05) (hash(x)=22280814) -630 train 8.756621 (lr=5.7573e-05) (hash(x)=24279448) -631 train 8.197635 (lr=5.7375e-05) (hash(x)=23054940) -632 train 8.315286 (lr=5.7177e-05) (hash(x)=23942400) -633 train 8.591723 (lr=5.6979e-05) (hash(x)=24712416) -634 train 8.146629 (lr=5.6782e-05) (hash(x)=24564658) -635 train 7.778137 (lr=5.6584e-05) (hash(x)=24909904) -636 train 7.848792 (lr=5.6386e-05) (hash(x)=23049534) -637 train 7.891622 (lr=5.6188e-05) (hash(x)=24321591) -638 train 7.681496 (lr=5.5990e-05) (hash(x)=26153298) -639 train 7.755660 (lr=5.5792e-05) (hash(x)=27140757) -640 train 7.539018 (lr=5.5594e-05) (hash(x)=25115907) -641 train 7.632585 (lr=5.5396e-05) (hash(x)=26563770) -642 train 7.789254 (lr=5.5198e-05) (hash(x)=28089252) -643 train 7.699501 (lr=5.5000e-05) (hash(x)=25993110) -644 train 7.693474 (lr=5.4802e-05) (hash(x)=25150008) -645 train 7.895142 (lr=5.4604e-05) (hash(x)=28520222) -646 train 7.548975 (lr=5.4406e-05) (hash(x)=21349943) -647 train 7.640263 (lr=5.4208e-05) (hash(x)=25149419) -648 train 7.842599 (lr=5.4010e-05) (hash(x)=25730641) -649 train 7.931845 (lr=5.3812e-05) (hash(x)=26112813) -650 val loss 7.7322 -650 val perplexity 2280.5588 -650 train 7.774684 (lr=5.3614e-05) (hash(x)=25907805) -651 train 7.802969 (lr=5.3416e-05) (hash(x)=27623643) -652 train 7.763165 (lr=5.3218e-05) (hash(x)=26484959) -653 train 7.813635 (lr=5.3021e-05) (hash(x)=29199854) -654 train 7.791019 (lr=5.2823e-05) (hash(x)=28369628) -655 train 7.679792 (lr=5.2625e-05) (hash(x)=24727764) -656 train 7.429103 (lr=5.2427e-05) (hash(x)=22610673) -657 train 7.354550 (lr=5.2230e-05) (hash(x)=22667179) -658 train 7.165549 (lr=5.2032e-05) (hash(x)=18477300) -659 train 7.566417 (lr=5.1835e-05) (hash(x)=23155773) -660 train 7.364282 (lr=5.1637e-05) (hash(x)=19461032) -661 train 7.647500 (lr=5.1440e-05) (hash(x)=23453788) -662 train 7.735754 (lr=5.1242e-05) (hash(x)=24543466) -663 train 7.527348 (lr=5.1045e-05) (hash(x)=21935931) -664 train 7.395804 (lr=5.0848e-05) (hash(x)=19910292) -665 train 7.658514 (lr=5.0651e-05) (hash(x)=24481079) -666 train 7.387647 (lr=5.0454e-05) (hash(x)=18922411) -667 train 7.397889 (lr=5.0257e-05) (hash(x)=20054917) -668 train 7.641472 (lr=5.0060e-05) (hash(x)=24850470) -669 train 7.664930 (lr=4.9863e-05) (hash(x)=25907741) -670 train 7.808061 (lr=4.9667e-05) (hash(x)=26873522) -671 train 7.834194 (lr=4.9470e-05) (hash(x)=27606073) -672 train 7.938076 (lr=4.9274e-05) (hash(x)=26209645) -673 train 7.774406 (lr=4.9077e-05) (hash(x)=25202001) -674 train 7.837802 (lr=4.8881e-05) (hash(x)=25569462) -675 train 7.808832 (lr=4.8685e-05) (hash(x)=26534487) -676 train 7.793518 (lr=4.8489e-05) (hash(x)=26455057) -677 train 7.647308 (lr=4.8293e-05) (hash(x)=24095850) -678 train 7.880449 (lr=4.8097e-05) (hash(x)=25287752) -679 train 7.402772 (lr=4.7902e-05) (hash(x)=22450341) -680 train 8.110153 (lr=4.7706e-05) (hash(x)=29004853) -681 train 7.840163 (lr=4.7511e-05) (hash(x)=27993763) -682 train 7.656087 (lr=4.7316e-05) (hash(x)=26382658) -683 train 7.812973 (lr=4.7121e-05) (hash(x)=25013073) -684 train 7.865118 (lr=4.6926e-05) (hash(x)=30595809) -685 train 7.954965 (lr=4.6731e-05) (hash(x)=30934371) -686 train 8.497666 (lr=4.6537e-05) (hash(x)=33060834) -687 train 7.754931 (lr=4.6342e-05) (hash(x)=25945859) -688 train 7.596353 (lr=4.6148e-05) (hash(x)=23375678) -689 train 7.658917 (lr=4.5954e-05) (hash(x)=25218689) -690 train 7.777696 (lr=4.5760e-05) (hash(x)=28127397) -691 train 7.718316 (lr=4.5567e-05) (hash(x)=24418091) -692 train 7.632825 (lr=4.5373e-05) (hash(x)=22761099) -693 train 7.628966 (lr=4.5180e-05) (hash(x)=24615466) -694 train 7.784940 (lr=4.4987e-05) (hash(x)=24699240) -695 train 7.683035 (lr=4.4794e-05) (hash(x)=21613707) -696 train 7.737260 (lr=4.4601e-05) (hash(x)=24977554) -697 train 7.710114 (lr=4.4408e-05) (hash(x)=24348175) -698 train 7.867577 (lr=4.4216e-05) (hash(x)=25102767) -699 train 7.815587 (lr=4.4024e-05) (hash(x)=26386157) -700 val loss 7.7096 -700 val perplexity 2229.6958 -700 train 7.749883 (lr=4.3832e-05) (hash(x)=26423460) -701 train 7.673131 (lr=4.3640e-05) (hash(x)=25771047) -702 train 7.738013 (lr=4.3449e-05) (hash(x)=26999875) -703 train 7.658440 (lr=4.3258e-05) (hash(x)=24396519) -704 train 7.690834 (lr=4.3067e-05) (hash(x)=22588122) -705 train 7.986687 (lr=4.2876e-05) (hash(x)=25142399) -706 train 7.733136 (lr=4.2685e-05) (hash(x)=20440214) -707 train 7.592997 (lr=4.2495e-05) (hash(x)=23265507) -708 train 7.662392 (lr=4.2305e-05) (hash(x)=24563470) -709 train 7.597501 (lr=4.2115e-05) (hash(x)=22514858) -710 train 7.942938 (lr=4.1925e-05) (hash(x)=26691212) -711 train 7.983129 (lr=4.1736e-05) (hash(x)=29138828) -712 train 7.971571 (lr=4.1547e-05) (hash(x)=28028528) -713 train 7.784318 (lr=4.1358e-05) (hash(x)=20531210) -714 train 7.689323 (lr=4.1170e-05) (hash(x)=25075352) -715 train 7.742566 (lr=4.0981e-05) (hash(x)=24265353) -716 train 7.835346 (lr=4.0793e-05) (hash(x)=24635726) -717 train 7.707597 (lr=4.0606e-05) (hash(x)=24999726) -718 train 7.904259 (lr=4.0418e-05) (hash(x)=27412910) -719 train 7.693563 (lr=4.0231e-05) (hash(x)=24685515) -720 train 7.758164 (lr=4.0044e-05) (hash(x)=23780329) -721 train 7.571144 (lr=3.9857e-05) (hash(x)=25071701) -722 train 7.729990 (lr=3.9671e-05) (hash(x)=23767130) -723 train 7.679067 (lr=3.9485e-05) (hash(x)=24876269) -724 train 7.810070 (lr=3.9299e-05) (hash(x)=26405773) -725 train 8.191543 (lr=3.9114e-05) (hash(x)=31733180) -726 train 7.435280 (lr=3.8929e-05) (hash(x)=21337509) -727 train 7.519728 (lr=3.8744e-05) (hash(x)=22825749) -728 train 7.878829 (lr=3.8560e-05) (hash(x)=28638695) -729 train 7.767526 (lr=3.8376e-05) (hash(x)=26393943) -730 train 7.934808 (lr=3.8192e-05) (hash(x)=27563583) -731 train 7.461130 (lr=3.8008e-05) (hash(x)=21239652) -732 train 7.611682 (lr=3.7825e-05) (hash(x)=23986428) -733 train 7.494781 (lr=3.7642e-05) (hash(x)=24943881) -734 train 8.056351 (lr=3.7460e-05) (hash(x)=29691448) -735 train 7.975795 (lr=3.7277e-05) (hash(x)=28767869) -736 train 7.706277 (lr=3.7096e-05) (hash(x)=23628188) -737 train 8.012710 (lr=3.6914e-05) (hash(x)=29341482) -738 train 7.808498 (lr=3.6733e-05) (hash(x)=30336570) -739 train 7.771975 (lr=3.6552e-05) (hash(x)=25614301) -740 train 7.669013 (lr=3.6372e-05) (hash(x)=24160500) -741 train 7.860786 (lr=3.6192e-05) (hash(x)=26030058) -742 train 8.215158 (lr=3.6012e-05) (hash(x)=29243936) -743 train 7.504323 (lr=3.5833e-05) (hash(x)=21159060) -744 train 7.622685 (lr=3.5654e-05) (hash(x)=23701853) -745 train 7.808171 (lr=3.5475e-05) (hash(x)=24629937) -746 train 7.696424 (lr=3.5297e-05) (hash(x)=25110108) -747 train 7.764526 (lr=3.5119e-05) (hash(x)=26751788) -748 train 7.795796 (lr=3.4942e-05) (hash(x)=26430427) -749 train 7.640619 (lr=3.4765e-05) (hash(x)=26012353) -750 val loss 7.7044 -750 val perplexity 2218.0164 -750 train 7.803008 (lr=3.4588e-05) (hash(x)=22735910) -751 train 7.584431 (lr=3.4412e-05) (hash(x)=25045397) -752 train 7.321576 (lr=3.4236e-05) (hash(x)=21554427) -753 train 7.430646 (lr=3.4060e-05) (hash(x)=23751143) -754 train 8.193354 (lr=3.3885e-05) (hash(x)=28602273) -755 train 8.229067 (lr=3.3711e-05) (hash(x)=29989709) -756 train 7.614389 (lr=3.3537e-05) (hash(x)=22331648) -757 train 8.069057 (lr=3.3363e-05) (hash(x)=31017246) -758 train 7.985886 (lr=3.3189e-05) (hash(x)=29709045) -759 train 8.030248 (lr=3.3016e-05) (hash(x)=25560928) -760 train 7.709583 (lr=3.2844e-05) (hash(x)=25075464) -761 train 7.791104 (lr=3.2672e-05) (hash(x)=27352253) -762 train 8.008784 (lr=3.2500e-05) (hash(x)=28187891) -763 train 7.859994 (lr=3.2329e-05) (hash(x)=26062687) -764 train 7.896323 (lr=3.2158e-05) (hash(x)=27427811) -765 train 7.936597 (lr=3.1988e-05) (hash(x)=27614522) -766 train 7.707165 (lr=3.1818e-05) (hash(x)=26129544) -767 train 8.244055 (lr=3.1648e-05) (hash(x)=28959222) -768 train 7.840927 (lr=3.1479e-05) (hash(x)=26860067) -769 train 7.615459 (lr=3.1311e-05) (hash(x)=25122598) -770 train 7.681770 (lr=3.1142e-05) (hash(x)=25245030) -771 train 7.701392 (lr=3.0975e-05) (hash(x)=25434884) -772 train 7.806191 (lr=3.0808e-05) (hash(x)=27732790) -773 train 7.828601 (lr=3.0641e-05) (hash(x)=27824438) -774 train 7.914417 (lr=3.0475e-05) (hash(x)=27201953) -775 train 7.619573 (lr=3.0309e-05) (hash(x)=21248405) -776 train 7.466390 (lr=3.0144e-05) (hash(x)=22805934) -777 train 7.723545 (lr=2.9979e-05) (hash(x)=26482588) -778 train 7.778482 (lr=2.9814e-05) (hash(x)=24153691) -779 train 7.888611 (lr=2.9651e-05) (hash(x)=25044192) -780 train 7.856115 (lr=2.9487e-05) (hash(x)=25910078) -781 train 7.843599 (lr=2.9324e-05) (hash(x)=28645524) -782 train 7.646734 (lr=2.9162e-05) (hash(x)=24368498) -783 train 7.721490 (lr=2.9000e-05) (hash(x)=25830182) -784 train 7.795751 (lr=2.8839e-05) (hash(x)=29181807) -785 train 7.720297 (lr=2.8678e-05) (hash(x)=25585137) -786 train 7.678884 (lr=2.8518e-05) (hash(x)=24798246) -787 train 7.786583 (lr=2.8358e-05) (hash(x)=26621419) -788 train 7.339437 (lr=2.8199e-05) (hash(x)=21446891) -789 train 7.298126 (lr=2.8040e-05) (hash(x)=22165286) -790 train 7.519704 (lr=2.7881e-05) (hash(x)=23477219) -791 train 7.816527 (lr=2.7724e-05) (hash(x)=25173113) -792 train 7.708726 (lr=2.7566e-05) (hash(x)=25853788) -793 train 7.799372 (lr=2.7410e-05) (hash(x)=27267091) -794 train 7.624678 (lr=2.7254e-05) (hash(x)=23743694) -795 train 7.601518 (lr=2.7098e-05) (hash(x)=24400133) -796 train 7.679440 (lr=2.6943e-05) (hash(x)=23663639) -797 train 7.548856 (lr=2.6788e-05) (hash(x)=23103223) -798 train 8.093730 (lr=2.6634e-05) (hash(x)=28748411) -799 train 7.416397 (lr=2.6481e-05) (hash(x)=23486277) -800 val loss 7.6914 -800 val perplexity 2189.4199 -800 train 7.660384 (lr=2.6328e-05) (hash(x)=25678518) -801 train 7.607830 (lr=2.6176e-05) (hash(x)=23421286) -802 train 7.766420 (lr=2.6024e-05) (hash(x)=26054104) -803 train 7.768556 (lr=2.5873e-05) (hash(x)=25978130) -804 train 7.806517 (lr=2.5722e-05) (hash(x)=26006525) -805 train 7.724640 (lr=2.5572e-05) (hash(x)=25769432) -806 train 7.523845 (lr=2.5423e-05) (hash(x)=22430795) -807 train 7.963420 (lr=2.5274e-05) (hash(x)=28916006) -808 train 7.709450 (lr=2.5125e-05) (hash(x)=25166800) -809 train 7.811880 (lr=2.4978e-05) (hash(x)=24226056) -810 train 7.665109 (lr=2.4830e-05) (hash(x)=25057934) -811 train 8.437477 (lr=2.4684e-05) (hash(x)=33195100) -812 train 7.844055 (lr=2.4538e-05) (hash(x)=26312888) -813 train 8.010243 (lr=2.4392e-05) (hash(x)=27730410) -814 train 7.878589 (lr=2.4247e-05) (hash(x)=27372474) -815 train 7.748260 (lr=2.4103e-05) (hash(x)=25556929) -816 train 7.801756 (lr=2.3959e-05) (hash(x)=26909985) -817 train 7.609512 (lr=2.3816e-05) (hash(x)=25991247) -818 train 7.820467 (lr=2.3674e-05) (hash(x)=27438141) -819 train 7.972879 (lr=2.3532e-05) (hash(x)=29536986) -820 train 7.705176 (lr=2.3391e-05) (hash(x)=24478391) -821 train 7.720511 (lr=2.3250e-05) (hash(x)=26125216) -822 train 7.743051 (lr=2.3110e-05) (hash(x)=26422130) -823 train 7.959709 (lr=2.2971e-05) (hash(x)=29648798) -824 train 7.605270 (lr=2.2832e-05) (hash(x)=21247770) -825 train 7.725322 (lr=2.2694e-05) (hash(x)=23195388) -826 train 7.692420 (lr=2.2556e-05) (hash(x)=25796725) -827 train 7.734476 (lr=2.2420e-05) (hash(x)=23124767) -828 train 7.787076 (lr=2.2283e-05) (hash(x)=25233464) -829 train 7.761665 (lr=2.2148e-05) (hash(x)=25713275) -830 train 7.695826 (lr=2.2013e-05) (hash(x)=25550167) -831 train 7.589298 (lr=2.1878e-05) (hash(x)=24976217) -832 train 7.899483 (lr=2.1745e-05) (hash(x)=28536827) -833 train 7.793140 (lr=2.1612e-05) (hash(x)=27500801) -834 train 7.587474 (lr=2.1479e-05) (hash(x)=25545765) -835 train 7.639930 (lr=2.1347e-05) (hash(x)=23632825) -836 train 7.677104 (lr=2.1216e-05) (hash(x)=25708009) -837 train 7.739574 (lr=2.1086e-05) (hash(x)=24456276) -838 train 7.835100 (lr=2.0956e-05) (hash(x)=29189855) -839 train 8.083582 (lr=2.0827e-05) (hash(x)=31019606) -840 train 7.918494 (lr=2.0698e-05) (hash(x)=26328013) -841 train 7.841841 (lr=2.0570e-05) (hash(x)=25027904) -842 train 7.626923 (lr=2.0443e-05) (hash(x)=23734189) -843 train 7.924020 (lr=2.0317e-05) (hash(x)=28236580) -844 train 7.680163 (lr=2.0191e-05) (hash(x)=26509780) -845 train 7.746202 (lr=2.0066e-05) (hash(x)=25386473) -846 train 7.721813 (lr=1.9941e-05) (hash(x)=24052671) -847 train 7.900289 (lr=1.9818e-05) (hash(x)=28269421) -848 train 7.310966 (lr=1.9694e-05) (hash(x)=22251724) -849 train 7.593678 (lr=1.9572e-05) (hash(x)=24308447) -850 val loss 7.6914 -850 val perplexity 2189.4326 -850 train 7.554504 (lr=1.9450e-05) (hash(x)=24242830) -851 train 7.756790 (lr=1.9329e-05) (hash(x)=25563279) -852 train 7.756253 (lr=1.9209e-05) (hash(x)=26354481) -853 train 7.852019 (lr=1.9089e-05) (hash(x)=26152637) -854 train 7.950653 (lr=1.8970e-05) (hash(x)=28051025) -855 train 7.733476 (lr=1.8852e-05) (hash(x)=24865358) -856 train 7.615482 (lr=1.8734e-05) (hash(x)=24288911) -857 train 7.517526 (lr=1.8618e-05) (hash(x)=22230964) -858 train 7.426700 (lr=1.8501e-05) (hash(x)=21303832) -859 train 7.576074 (lr=1.8386e-05) (hash(x)=22155546) -860 train 7.700043 (lr=1.8271e-05) (hash(x)=25296428) -861 train 7.844306 (lr=1.8157e-05) (hash(x)=29142319) -862 train 7.675201 (lr=1.8044e-05) (hash(x)=25545430) -863 train 7.498362 (lr=1.7931e-05) (hash(x)=26984272) -864 train 7.700910 (lr=1.7819e-05) (hash(x)=25429005) -865 train 7.766165 (lr=1.7708e-05) (hash(x)=27077032) -866 train 7.713490 (lr=1.7598e-05) (hash(x)=26494424) -867 train 7.692723 (lr=1.7488e-05) (hash(x)=23193673) -868 train 7.745712 (lr=1.7379e-05) (hash(x)=25075134) -869 train 7.808554 (lr=1.7271e-05) (hash(x)=27112558) -870 train 7.936128 (lr=1.7163e-05) (hash(x)=27436608) -871 train 7.646001 (lr=1.7056e-05) (hash(x)=24544116) -872 train 7.967008 (lr=1.6950e-05) (hash(x)=31632686) -873 train 7.660706 (lr=1.6845e-05) (hash(x)=25890184) -874 train 7.740621 (lr=1.6740e-05) (hash(x)=22887555) -875 train 7.642976 (lr=1.6636e-05) (hash(x)=24547533) -876 train 7.771511 (lr=1.6533e-05) (hash(x)=26553496) -877 train 7.859274 (lr=1.6431e-05) (hash(x)=27467688) -878 train 7.626846 (lr=1.6329e-05) (hash(x)=24766934) -879 train 7.525403 (lr=1.6228e-05) (hash(x)=22059850) -880 train 7.879493 (lr=1.6128e-05) (hash(x)=22871702) -881 train 7.713406 (lr=1.6029e-05) (hash(x)=23893130) -882 train 7.784138 (lr=1.5930e-05) (hash(x)=25125691) -883 train 7.720818 (lr=1.5832e-05) (hash(x)=25994573) -884 train 7.758055 (lr=1.5735e-05) (hash(x)=26076345) -885 train 7.714443 (lr=1.5639e-05) (hash(x)=26577783) -886 train 7.827947 (lr=1.5543e-05) (hash(x)=27395225) -887 train 7.655860 (lr=1.5449e-05) (hash(x)=23926632) -888 train 7.500920 (lr=1.5354e-05) (hash(x)=21737239) -889 train 7.761072 (lr=1.5261e-05) (hash(x)=23574207) -890 train 7.485734 (lr=1.5169e-05) (hash(x)=24365231) -891 train 7.678573 (lr=1.5077e-05) (hash(x)=27111369) -892 train 7.903385 (lr=1.4986e-05) (hash(x)=27290015) -893 train 7.668022 (lr=1.4896e-05) (hash(x)=23979820) -894 train 7.784252 (lr=1.4806e-05) (hash(x)=26450121) -895 train 7.666934 (lr=1.4718e-05) (hash(x)=27025333) -896 train 7.670912 (lr=1.4630e-05) (hash(x)=23624605) -897 train 7.454749 (lr=1.4543e-05) (hash(x)=22846386) -898 train 7.581425 (lr=1.4456e-05) (hash(x)=22970561) -899 train 7.386701 (lr=1.4371e-05) (hash(x)=16908068) -900 val loss 7.6757 -900 val perplexity 2155.3518 -900 train 7.636807 (lr=1.4286e-05) (hash(x)=24661446) -901 train 7.722177 (lr=1.4202e-05) (hash(x)=25664727) -902 train 7.718565 (lr=1.4119e-05) (hash(x)=25667011) -903 train 7.781650 (lr=1.4037e-05) (hash(x)=29120407) -904 train 7.774096 (lr=1.3955e-05) (hash(x)=23385735) -905 train 7.830679 (lr=1.3874e-05) (hash(x)=25564213) -906 train 7.906770 (lr=1.3794e-05) (hash(x)=25413898) -907 train 7.867984 (lr=1.3715e-05) (hash(x)=27092710) -908 train 7.647148 (lr=1.3637e-05) (hash(x)=25789923) -909 train 7.761417 (lr=1.3559e-05) (hash(x)=28533197) -910 train 7.598724 (lr=1.3482e-05) (hash(x)=22982996) -911 train 7.678322 (lr=1.3407e-05) (hash(x)=23827393) -912 train 7.302722 (lr=1.3331e-05) (hash(x)=21242640) -913 train 7.677151 (lr=1.3257e-05) (hash(x)=24154233) -914 train 7.692226 (lr=1.3183e-05) (hash(x)=24331967) -915 train 7.922234 (lr=1.3111e-05) (hash(x)=32812727) -916 train 7.584315 (lr=1.3039e-05) (hash(x)=23572994) -917 train 7.633311 (lr=1.2968e-05) (hash(x)=26305435) -918 train 7.672638 (lr=1.2897e-05) (hash(x)=26268355) -919 train 7.839767 (lr=1.2828e-05) (hash(x)=27230027) -920 train 7.662517 (lr=1.2759e-05) (hash(x)=23885377) -921 train 7.556822 (lr=1.2691e-05) (hash(x)=23532437) -922 train 7.673448 (lr=1.2624e-05) (hash(x)=25577034) -923 train 7.799294 (lr=1.2558e-05) (hash(x)=25703381) -924 train 7.796239 (lr=1.2493e-05) (hash(x)=27113866) -925 train 7.629724 (lr=1.2428e-05) (hash(x)=26961429) -926 train 7.443880 (lr=1.2364e-05) (hash(x)=21355372) -927 train 7.616615 (lr=1.2301e-05) (hash(x)=24968260) -928 train 7.599031 (lr=1.2239e-05) (hash(x)=25357517) -929 train 7.664371 (lr=1.2178e-05) (hash(x)=24854265) -930 train 7.424274 (lr=1.2118e-05) (hash(x)=21102770) -931 train 7.745500 (lr=1.2058e-05) (hash(x)=25676468) -932 train 7.651903 (lr=1.1999e-05) (hash(x)=22809869) -933 train 7.749649 (lr=1.1941e-05) (hash(x)=25503865) -934 train 7.624060 (lr=1.1884e-05) (hash(x)=24853995) -935 train 7.787876 (lr=1.1828e-05) (hash(x)=27544803) -936 train 7.819981 (lr=1.1772e-05) (hash(x)=25981933) -937 train 7.535001 (lr=1.1718e-05) (hash(x)=24658683) -938 train 7.732615 (lr=1.1664e-05) (hash(x)=23855201) -939 train 7.659743 (lr=1.1611e-05) (hash(x)=24331407) -940 train 7.858729 (lr=1.1559e-05) (hash(x)=29265551) -941 train 7.576809 (lr=1.1508e-05) (hash(x)=21892556) -942 train 7.583662 (lr=1.1457e-05) (hash(x)=27183405) -943 train 7.583963 (lr=1.1408e-05) (hash(x)=26540663) -944 train 7.535895 (lr=1.1359e-05) (hash(x)=25718393) -945 train 7.737484 (lr=1.1311e-05) (hash(x)=26819462) -946 train 7.726364 (lr=1.1264e-05) (hash(x)=27427540) -947 train 7.716547 (lr=1.1218e-05) (hash(x)=25532657) -948 train 7.941617 (lr=1.1173e-05) (hash(x)=27641372) -949 train 7.651660 (lr=1.1128e-05) (hash(x)=26515570) -950 val loss 7.6738 -950 val perplexity 2151.1648 -950 train 7.892084 (lr=1.1085e-05) (hash(x)=26911957) -951 train 7.981761 (lr=1.1042e-05) (hash(x)=25856625) -952 train 7.573560 (lr=1.1000e-05) (hash(x)=25219129) -953 train 7.528502 (lr=1.0959e-05) (hash(x)=25260471) -954 train 7.901083 (lr=1.0919e-05) (hash(x)=29373370) -955 train 7.464719 (lr=1.0879e-05) (hash(x)=23437426) -956 train 7.782407 (lr=1.0841e-05) (hash(x)=23769521) -957 train 7.802667 (lr=1.0803e-05) (hash(x)=25961833) -958 train 7.529685 (lr=1.0766e-05) (hash(x)=23582666) -959 train 7.592771 (lr=1.0730e-05) (hash(x)=23164356) -960 train 7.588975 (lr=1.0695e-05) (hash(x)=24443114) -961 train 7.643714 (lr=1.0661e-05) (hash(x)=25052665) -962 train 7.780682 (lr=1.0628e-05) (hash(x)=27802272) -963 train 7.640738 (lr=1.0595e-05) (hash(x)=25957896) -964 train 7.693431 (lr=1.0563e-05) (hash(x)=26737251) -965 train 7.792041 (lr=1.0533e-05) (hash(x)=24723263) -966 train 7.731790 (lr=1.0503e-05) (hash(x)=24707011) -967 train 7.714994 (lr=1.0474e-05) (hash(x)=25646282) -968 train 7.721770 (lr=1.0445e-05) (hash(x)=27544665) -969 train 7.683399 (lr=1.0418e-05) (hash(x)=25851993) -970 train 7.940343 (lr=1.0391e-05) (hash(x)=29059700) -971 train 7.612316 (lr=1.0366e-05) (hash(x)=21513584) -972 train 7.462730 (lr=1.0341e-05) (hash(x)=23151267) -973 train 7.645348 (lr=1.0317e-05) (hash(x)=26017176) -974 train 7.770299 (lr=1.0294e-05) (hash(x)=26979518) -975 train 7.717550 (lr=1.0272e-05) (hash(x)=23843233) -976 train 7.856168 (lr=1.0251e-05) (hash(x)=24193010) -977 train 7.593259 (lr=1.0230e-05) (hash(x)=21476847) -978 train 7.483603 (lr=1.0211e-05) (hash(x)=21366504) -979 train 7.585090 (lr=1.0192e-05) (hash(x)=23226697) -980 train 7.378542 (lr=1.0174e-05) (hash(x)=19961773) -981 train 7.228125 (lr=1.0157e-05) (hash(x)=19772969) -982 train 7.529696 (lr=1.0141e-05) (hash(x)=23110142) -983 train 7.766515 (lr=1.0126e-05) (hash(x)=24506028) -984 train 7.744504 (lr=1.0111e-05) (hash(x)=25480731) -985 train 7.592827 (lr=1.0098e-05) (hash(x)=21077417) -986 train 7.583657 (lr=1.0085e-05) (hash(x)=23686713) -987 train 7.680549 (lr=1.0074e-05) (hash(x)=26024321) -988 train 7.749382 (lr=1.0063e-05) (hash(x)=27424109) -989 train 7.830625 (lr=1.0053e-05) (hash(x)=27786174) -990 train 7.652845 (lr=1.0044e-05) (hash(x)=25232502) -991 train 7.644319 (lr=1.0035e-05) (hash(x)=22781277) -992 train 7.693330 (lr=1.0028e-05) (hash(x)=26184527) -993 train 7.705282 (lr=1.0021e-05) (hash(x)=24459895) -994 train 7.658930 (lr=1.0016e-05) (hash(x)=25244624) -995 train 7.817449 (lr=1.0011e-05) (hash(x)=24451843) -996 train 7.781450 (lr=1.0007e-05) (hash(x)=22129897) -997 train 7.620546 (lr=1.0004e-05) (hash(x)=21116390) -998 train 7.482774 (lr=1.0002e-05) (hash(x)=20650070) -999 val loss 7.6691 -999 val perplexity 2141.1951 -999 train 7.806221 (lr=1.0000e-05) (hash(x)=24948650) +20 train 11.657885 (lr=7.3427e-06) (hash(x)=25441898) +21 train 11.650809 (lr=7.6923e-06) (hash(x)=28375581) +22 train 11.681827 (lr=8.0420e-06) (hash(x)=24046679) +23 train 11.661906 (lr=8.3916e-06) (hash(x)=24611628) +24 train 11.662688 (lr=8.7413e-06) (hash(x)=26169030) +25 train 11.630103 (lr=9.0909e-06) (hash(x)=30298407) +26 train 11.633911 (lr=9.4406e-06) (hash(x)=23711112) +27 train 11.604510 (lr=9.7902e-06) (hash(x)=19245352) +28 train 11.580858 (lr=1.0140e-05) (hash(x)=21529136) +29 train 11.574893 (lr=1.0490e-05) (hash(x)=28936608) +30 train 11.549212 (lr=1.0839e-05) (hash(x)=24339013) +31 train 11.521532 (lr=1.1189e-05) (hash(x)=25767553) +32 train 11.517820 (lr=1.1538e-05) (hash(x)=26439905) +33 train 11.536746 (lr=1.1888e-05) (hash(x)=31093473) +34 train 11.459716 (lr=1.2238e-05) (hash(x)=25450374) +35 train 11.518849 (lr=1.2587e-05) (hash(x)=24809873) +36 train 11.487486 (lr=1.2937e-05) (hash(x)=23253252) +37 train 11.512972 (lr=1.3287e-05) (hash(x)=27852919) +38 train 11.407732 (lr=1.3636e-05) (hash(x)=23327497) +39 train 11.417821 (lr=1.3986e-05) (hash(x)=22512166) +40 train 11.431041 (lr=1.4336e-05) (hash(x)=22859419) +41 train 11.394411 (lr=1.4685e-05) (hash(x)=27620338) +42 train 11.357187 (lr=1.5035e-05) (hash(x)=26397837) +43 train 11.365531 (lr=1.5385e-05) (hash(x)=28092148) +44 train 11.318680 (lr=1.5734e-05) (hash(x)=24662703) +45 train 11.324164 (lr=1.6084e-05) (hash(x)=27938767) +46 train 11.332822 (lr=1.6434e-05) (hash(x)=26037988) +47 train 11.229604 (lr=1.6783e-05) (hash(x)=24732833) +48 train 11.247222 (lr=1.7133e-05) (hash(x)=25259526) +49 train 11.246371 (lr=1.7483e-05) (hash(x)=23200230) +50 val loss 11.2055 +50 val perplexity 73535.9844 +50 train 11.193517 (lr=1.7832e-05) (hash(x)=26721357) +51 train 11.204566 (lr=1.8182e-05) (hash(x)=22694718) +52 train 11.152442 (lr=1.8531e-05) (hash(x)=28066766) +53 train 11.117385 (lr=1.8881e-05) (hash(x)=23125151) +54 train 11.113418 (lr=1.9231e-05) (hash(x)=27193725) +55 train 11.061242 (lr=1.9580e-05) (hash(x)=25129410) +56 train 11.079288 (lr=1.9930e-05) (hash(x)=24263988) +57 train 10.957766 (lr=2.0280e-05) (hash(x)=23059154) +58 train 10.997256 (lr=2.0629e-05) (hash(x)=26063864) +59 train 11.061813 (lr=2.0979e-05) (hash(x)=27858570) +60 train 10.902617 (lr=2.1329e-05) (hash(x)=23874620) +61 train 10.886213 (lr=2.1678e-05) (hash(x)=22402617) +62 train 10.856033 (lr=2.2028e-05) (hash(x)=23600822) +63 train 10.730399 (lr=2.2378e-05) (hash(x)=26582391) +64 train 10.721463 (lr=2.2727e-05) (hash(x)=23225283) +65 train 10.820321 (lr=2.3077e-05) (hash(x)=26075451) +66 train 10.795652 (lr=2.3427e-05) (hash(x)=24723419) +67 train 10.695048 (lr=2.3776e-05) (hash(x)=27279806) +68 train 10.610442 (lr=2.4126e-05) (hash(x)=25870391) +69 train 10.655809 (lr=2.4476e-05) (hash(x)=26188136) +70 train 10.589160 (lr=2.4825e-05) (hash(x)=30373443) +71 train 10.663558 (lr=2.5175e-05) (hash(x)=26472336) +72 train 10.578518 (lr=2.5524e-05) (hash(x)=26651572) +73 train 10.568738 (lr=2.5874e-05) (hash(x)=26376212) +74 train 10.542186 (lr=2.6224e-05) (hash(x)=26733350) +75 train 10.632166 (lr=2.6573e-05) (hash(x)=28301589) +76 train 10.549161 (lr=2.6923e-05) (hash(x)=27599559) +77 train 10.429263 (lr=2.7273e-05) (hash(x)=28035221) +78 train 10.495929 (lr=2.7622e-05) (hash(x)=25016783) +79 train 10.529335 (lr=2.7972e-05) (hash(x)=27654289) +80 train 10.549395 (lr=2.8322e-05) (hash(x)=24597558) +81 train 10.512833 (lr=2.8671e-05) (hash(x)=21560904) +82 train 10.334525 (lr=2.9021e-05) (hash(x)=21983837) +83 train 10.373830 (lr=2.9371e-05) (hash(x)=24995715) +84 train 10.426661 (lr=2.9720e-05) (hash(x)=29876413) +85 train 10.370793 (lr=3.0070e-05) (hash(x)=23792508) +86 train 10.308570 (lr=3.0420e-05) (hash(x)=25509120) +87 train 10.449273 (lr=3.0769e-05) (hash(x)=26559876) +88 train 10.371466 (lr=3.1119e-05) (hash(x)=23569647) +89 train 10.403503 (lr=3.1469e-05) (hash(x)=25758852) +90 train 10.293134 (lr=3.1818e-05) (hash(x)=25706298) +91 train 10.180539 (lr=3.2168e-05) (hash(x)=28364895) +92 train 10.237029 (lr=3.2517e-05) (hash(x)=25304663) +93 train 10.320305 (lr=3.2867e-05) (hash(x)=25269299) +94 train 10.469955 (lr=3.3217e-05) (hash(x)=25870566) +95 train 10.285203 (lr=3.3566e-05) (hash(x)=21770329) +96 train 10.463753 (lr=3.3916e-05) (hash(x)=27595900) +97 train 10.483836 (lr=3.4266e-05) (hash(x)=24785397) +98 train 10.255240 (lr=3.4615e-05) (hash(x)=21521480) +99 train 10.286616 (lr=3.4965e-05) (hash(x)=24628606) +100 val loss 10.2756 +100 val perplexity 29016.9863 +100 train 10.313794 (lr=3.5315e-05) (hash(x)=24670150) +101 train 10.297599 (lr=3.5664e-05) (hash(x)=23181910) +102 train 10.168429 (lr=3.6014e-05) (hash(x)=22714991) +103 train 10.286510 (lr=3.6364e-05) (hash(x)=22723459) +104 train 10.212409 (lr=3.6713e-05) (hash(x)=21524316) +105 train 10.204615 (lr=3.7063e-05) (hash(x)=25506632) +106 train 10.034871 (lr=3.7413e-05) (hash(x)=21675672) +107 train 10.159293 (lr=3.7762e-05) (hash(x)=22897919) +108 train 10.189796 (lr=3.8112e-05) (hash(x)=23321631) +109 train 10.215676 (lr=3.8462e-05) (hash(x)=26546719) +110 train 10.692629 (lr=3.8811e-05) (hash(x)=31962348) +111 train 10.212363 (lr=3.9161e-05) (hash(x)=30338342) +112 train 10.173563 (lr=3.9510e-05) (hash(x)=23724471) +113 train 10.161604 (lr=3.9860e-05) (hash(x)=29175888) +114 train 10.249930 (lr=4.0210e-05) (hash(x)=23256716) +115 train 10.213691 (lr=4.0559e-05) (hash(x)=27063280) +116 train 10.163449 (lr=4.0909e-05) (hash(x)=31057659) +117 train 10.180331 (lr=4.1259e-05) (hash(x)=32915097) +118 train 10.128627 (lr=4.1608e-05) (hash(x)=28842717) +119 train 10.130229 (lr=4.1958e-05) (hash(x)=25678059) +120 train 10.121390 (lr=4.2308e-05) (hash(x)=21593510) +121 train 10.000831 (lr=4.2657e-05) (hash(x)=20083773) +122 train 10.052983 (lr=4.3007e-05) (hash(x)=23002820) +123 train 10.032634 (lr=4.3357e-05) (hash(x)=21853028) +124 train 10.059019 (lr=4.3706e-05) (hash(x)=26985625) +125 train 10.001665 (lr=4.4056e-05) (hash(x)=21808483) +126 train 10.120544 (lr=4.4406e-05) (hash(x)=28873251) +127 train 10.065219 (lr=4.4755e-05) (hash(x)=26109335) +128 train 10.067626 (lr=4.5105e-05) (hash(x)=26334674) +129 train 10.042725 (lr=4.5455e-05) (hash(x)=24916754) +130 train 10.088353 (lr=4.5804e-05) (hash(x)=25449624) +131 train 10.153441 (lr=4.6154e-05) (hash(x)=25334848) +132 train 10.189923 (lr=4.6503e-05) (hash(x)=27484863) +133 train 10.016569 (lr=4.6853e-05) (hash(x)=24917705) +134 train 9.866614 (lr=4.7203e-05) (hash(x)=25143449) +135 train 9.922203 (lr=4.7552e-05) (hash(x)=25044885) +136 train 10.134152 (lr=4.7902e-05) (hash(x)=27821028) +137 train 10.067044 (lr=4.8252e-05) (hash(x)=28747022) +138 train 10.014403 (lr=4.8601e-05) (hash(x)=27182888) +139 train 9.893394 (lr=4.8951e-05) (hash(x)=23678349) +140 train 10.030443 (lr=4.9301e-05) (hash(x)=23593235) +141 train 10.135492 (lr=4.9650e-05) (hash(x)=28529813) +142 train 10.191312 (lr=5.0000e-05) (hash(x)=32074661) +143 train 9.916725 (lr=5.0350e-05) (hash(x)=28870690) +144 train 9.993833 (lr=5.0699e-05) (hash(x)=27307705) +145 train 9.961196 (lr=5.1049e-05) (hash(x)=25044834) +146 train 10.023426 (lr=5.1399e-05) (hash(x)=23712023) +147 train 10.143596 (lr=5.1748e-05) (hash(x)=32982615) +148 train 10.022178 (lr=5.2098e-05) (hash(x)=30113660) +149 train 9.869081 (lr=5.2448e-05) (hash(x)=20970960) +150 val loss 9.8965 +150 val perplexity 19860.4062 +150 train 9.871131 (lr=5.2797e-05) (hash(x)=23132684) +151 train 10.464212 (lr=5.3147e-05) (hash(x)=35279941) +152 train 10.171824 (lr=5.3497e-05) (hash(x)=31227444) +153 train 9.836967 (lr=5.3846e-05) (hash(x)=25529472) +154 train 9.840961 (lr=5.4196e-05) (hash(x)=24350409) +155 train 10.055727 (lr=5.4545e-05) (hash(x)=26400041) +156 train 9.834991 (lr=5.4895e-05) (hash(x)=25262621) +157 train 9.823700 (lr=5.5245e-05) (hash(x)=24656138) +158 train 9.850169 (lr=5.5594e-05) (hash(x)=26803414) +159 train 9.806029 (lr=5.5944e-05) (hash(x)=25015923) +160 train 9.707911 (lr=5.6294e-05) (hash(x)=23581172) +161 train 9.772491 (lr=5.6643e-05) (hash(x)=22924885) +162 train 9.728972 (lr=5.6993e-05) (hash(x)=23414296) +163 train 9.945615 (lr=5.7343e-05) (hash(x)=24853586) +164 train 9.940111 (lr=5.7692e-05) (hash(x)=25000130) +165 train 9.805405 (lr=5.8042e-05) (hash(x)=27004780) +166 train 9.771125 (lr=5.8392e-05) (hash(x)=26148573) +167 train 9.737063 (lr=5.8741e-05) (hash(x)=26740855) +168 train 9.537338 (lr=5.9091e-05) (hash(x)=20965419) +169 train 9.657731 (lr=5.9441e-05) (hash(x)=23950114) +170 train 9.657863 (lr=5.9790e-05) (hash(x)=24951982) +171 train 9.726392 (lr=6.0140e-05) (hash(x)=24584116) +172 train 9.817248 (lr=6.0490e-05) (hash(x)=24378759) +173 train 9.653930 (lr=6.0839e-05) (hash(x)=25718516) +174 train 9.727645 (lr=6.1189e-05) (hash(x)=28424396) +175 train 9.639813 (lr=6.1538e-05) (hash(x)=22262151) +176 train 9.792890 (lr=6.1888e-05) (hash(x)=26438412) +177 train 9.654679 (lr=6.2238e-05) (hash(x)=23025303) +178 train 9.546203 (lr=6.2587e-05) (hash(x)=24190770) +179 train 9.615925 (lr=6.2937e-05) (hash(x)=26627860) +180 train 9.652865 (lr=6.3287e-05) (hash(x)=23663439) +181 train 9.455048 (lr=6.3636e-05) (hash(x)=23700532) +182 train 9.525774 (lr=6.3986e-05) (hash(x)=23075676) +183 train 9.561461 (lr=6.4336e-05) (hash(x)=26621834) +184 train 9.564555 (lr=6.4685e-05) (hash(x)=29426269) +185 train 9.552994 (lr=6.5035e-05) (hash(x)=21821465) +186 train 9.580278 (lr=6.5385e-05) (hash(x)=26506130) +187 train 9.703278 (lr=6.5734e-05) (hash(x)=26930630) +188 train 9.546063 (lr=6.6084e-05) (hash(x)=22993793) +189 train 9.552161 (lr=6.6434e-05) (hash(x)=19557946) +190 train 9.661199 (lr=6.6783e-05) (hash(x)=23572891) +191 train 9.456439 (lr=6.7133e-05) (hash(x)=23234741) +192 train 9.554379 (lr=6.7483e-05) (hash(x)=25547951) +193 train 9.557416 (lr=6.7832e-05) (hash(x)=26713563) +194 train 9.539050 (lr=6.8182e-05) (hash(x)=25913622) +195 train 9.599185 (lr=6.8531e-05) (hash(x)=28102443) +196 train 9.532004 (lr=6.8881e-05) (hash(x)=23093351) +197 train 9.814094 (lr=6.9231e-05) (hash(x)=31689122) +198 train 9.745892 (lr=6.9580e-05) (hash(x)=32252517) +199 train 9.519612 (lr=6.9930e-05) (hash(x)=25470563) +200 val loss 9.5080 +200 val perplexity 13467.3848 +200 train 9.596503 (lr=7.0280e-05) (hash(x)=25597614) +201 train 9.571892 (lr=7.0629e-05) (hash(x)=23757479) +202 train 9.396857 (lr=7.0979e-05) (hash(x)=25140048) +203 train 9.497906 (lr=7.1329e-05) (hash(x)=28282861) +204 train 9.420123 (lr=7.1678e-05) (hash(x)=24754885) +205 train 9.565269 (lr=7.2028e-05) (hash(x)=26731964) +206 train 9.599589 (lr=7.2378e-05) (hash(x)=26660561) +207 train 9.460540 (lr=7.2727e-05) (hash(x)=21799102) +208 train 9.443197 (lr=7.3077e-05) (hash(x)=27025986) +209 train 9.359263 (lr=7.3427e-05) (hash(x)=24672077) +210 train 9.404801 (lr=7.3776e-05) (hash(x)=25322984) +211 train 9.360745 (lr=7.4126e-05) (hash(x)=23471769) +212 train 9.131750 (lr=7.4476e-05) (hash(x)=20766491) +213 train 9.636192 (lr=7.4825e-05) (hash(x)=24058931) +214 train 9.331761 (lr=7.5175e-05) (hash(x)=23872843) +215 train 9.350955 (lr=7.5524e-05) (hash(x)=23275479) +216 train 9.462272 (lr=7.5874e-05) (hash(x)=24914695) +217 train 9.364687 (lr=7.6224e-05) (hash(x)=24364396) +218 train 9.346486 (lr=7.6573e-05) (hash(x)=27986474) +219 train 9.421395 (lr=7.6923e-05) (hash(x)=24473581) +220 train 9.371788 (lr=7.7273e-05) (hash(x)=22974689) +221 train 9.391391 (lr=7.7622e-05) (hash(x)=23774644) +222 train 9.445415 (lr=7.7972e-05) (hash(x)=23245327) +223 train 9.273964 (lr=7.8322e-05) (hash(x)=22091862) +224 train 9.197638 (lr=7.8671e-05) (hash(x)=24362839) +225 train 9.138532 (lr=7.9021e-05) (hash(x)=25482303) +226 train 9.354041 (lr=7.9371e-05) (hash(x)=24911853) +227 train 9.310000 (lr=7.9720e-05) (hash(x)=26018202) +228 train 9.205912 (lr=8.0070e-05) (hash(x)=26124495) +229 train 9.262061 (lr=8.0420e-05) (hash(x)=24560096) +230 train 9.229075 (lr=8.0769e-05) (hash(x)=24695331) +231 train 9.175804 (lr=8.1119e-05) (hash(x)=17430373) +232 train 9.148310 (lr=8.1469e-05) (hash(x)=21813345) +233 train 8.870238 (lr=8.1818e-05) (hash(x)=20098681) +234 train 9.025661 (lr=8.2168e-05) (hash(x)=25095928) +235 train 9.109674 (lr=8.2517e-05) (hash(x)=24078083) +236 train 9.188262 (lr=8.2867e-05) (hash(x)=22901505) +237 train 9.148584 (lr=8.3217e-05) (hash(x)=26595592) +238 train 9.136312 (lr=8.3566e-05) (hash(x)=27663196) +239 train 9.043187 (lr=8.3916e-05) (hash(x)=22954861) +240 train 9.217849 (lr=8.4266e-05) (hash(x)=30159234) +241 train 9.180590 (lr=8.4615e-05) (hash(x)=29294271) +242 train 9.173262 (lr=8.4965e-05) (hash(x)=23728322) +243 train 9.191935 (lr=8.5315e-05) (hash(x)=28695016) +244 train 9.236748 (lr=8.5664e-05) (hash(x)=26702728) +245 train 9.060979 (lr=8.6014e-05) (hash(x)=27100115) +246 train 9.057665 (lr=8.6364e-05) (hash(x)=25082752) +247 train 9.314529 (lr=8.6713e-05) (hash(x)=26671799) +248 train 9.172324 (lr=8.7063e-05) (hash(x)=23718946) +249 train 9.192331 (lr=8.7413e-05) (hash(x)=28137394) +250 val loss 9.0734 +250 val perplexity 8720.5176 +250 train 9.171360 (lr=8.7762e-05) (hash(x)=23893495) +251 train 9.095279 (lr=8.8112e-05) (hash(x)=23166092) +252 train 8.984591 (lr=8.8462e-05) (hash(x)=25907665) +253 train 8.945840 (lr=8.8811e-05) (hash(x)=25885986) +254 train 9.017692 (lr=8.9161e-05) (hash(x)=25262712) +255 train 8.983135 (lr=8.9510e-05) (hash(x)=26924723) +256 train 9.125360 (lr=8.9860e-05) (hash(x)=28744736) +257 train 8.984564 (lr=9.0210e-05) (hash(x)=26140590) +258 train 9.106241 (lr=9.0559e-05) (hash(x)=25780449) +259 train 8.999628 (lr=9.0909e-05) (hash(x)=25138659) +260 train 9.080793 (lr=9.1259e-05) (hash(x)=27244046) +261 train 9.067635 (lr=9.1608e-05) (hash(x)=27224685) +262 train 9.120746 (lr=9.1958e-05) (hash(x)=28274477) +263 train 8.912524 (lr=9.2308e-05) (hash(x)=23557495) +264 train 8.972316 (lr=9.2657e-05) (hash(x)=24680596) +265 train 8.896482 (lr=9.3007e-05) (hash(x)=23928957) +266 train 8.838000 (lr=9.3357e-05) (hash(x)=23761390) +267 train 8.934711 (lr=9.3706e-05) (hash(x)=25288123) +268 train 8.972369 (lr=9.4056e-05) (hash(x)=28705502) +269 train 8.714244 (lr=9.4406e-05) (hash(x)=23246294) +270 train 8.685000 (lr=9.4755e-05) (hash(x)=28639079) +271 train 8.644155 (lr=9.5105e-05) (hash(x)=27804380) +272 train 8.859668 (lr=9.5455e-05) (hash(x)=24172235) +273 train 8.915594 (lr=9.5804e-05) (hash(x)=23089140) +274 train 9.074924 (lr=9.6154e-05) (hash(x)=27163701) +275 train 9.101553 (lr=9.6503e-05) (hash(x)=26993263) +276 train 8.900540 (lr=9.6853e-05) (hash(x)=28224233) +277 train 9.174178 (lr=9.7203e-05) (hash(x)=27397203) +278 train 8.960447 (lr=9.7552e-05) (hash(x)=27923882) +279 train 8.918372 (lr=9.7902e-05) (hash(x)=26654908) +280 train 8.938707 (lr=9.8252e-05) (hash(x)=24213147) +281 train 8.717601 (lr=9.8601e-05) (hash(x)=21965022) +282 train 8.797449 (lr=9.8951e-05) (hash(x)=25465685) +283 train 8.872131 (lr=9.9301e-05) (hash(x)=27347722) +284 train 8.873816 (lr=9.9650e-05) (hash(x)=26732050) +285 train 8.963028 (lr=1.0000e-04) (hash(x)=28314127) +286 train 8.839798 (lr=1.0000e-04) (hash(x)=21471186) +287 train 8.796781 (lr=1.0000e-04) (hash(x)=23627518) +288 train 8.780684 (lr=9.9998e-05) (hash(x)=20870353) +289 train 8.746551 (lr=9.9996e-05) (hash(x)=25024764) +290 train 8.602712 (lr=9.9993e-05) (hash(x)=20683822) +291 train 8.871664 (lr=9.9989e-05) (hash(x)=21768671) +292 train 8.780704 (lr=9.9984e-05) (hash(x)=25557309) +293 train 8.952842 (lr=9.9979e-05) (hash(x)=25076667) +294 train 8.682808 (lr=9.9972e-05) (hash(x)=23765822) +295 train 8.710036 (lr=9.9965e-05) (hash(x)=21889990) +296 train 8.737759 (lr=9.9956e-05) (hash(x)=26339893) +297 train 8.589422 (lr=9.9947e-05) (hash(x)=20932794) +298 train 8.656519 (lr=9.9937e-05) (hash(x)=21750070) +299 train 8.807393 (lr=9.9926e-05) (hash(x)=23665838) +300 val loss 8.7288 +300 val perplexity 6178.0903 +300 train 9.315824 (lr=9.9915e-05) (hash(x)=32888061) +301 train 9.129425 (lr=9.9902e-05) (hash(x)=30223582) +302 train 8.863655 (lr=9.9889e-05) (hash(x)=26908418) +303 train 8.482724 (lr=9.9874e-05) (hash(x)=22528001) +304 train 8.836436 (lr=9.9859e-05) (hash(x)=27452187) +305 train 8.757055 (lr=9.9843e-05) (hash(x)=25181641) +306 train 8.729921 (lr=9.9826e-05) (hash(x)=25546593) +307 train 8.650927 (lr=9.9808e-05) (hash(x)=22487328) +308 train 8.797595 (lr=9.9789e-05) (hash(x)=27804274) +309 train 8.846377 (lr=9.9770e-05) (hash(x)=26544630) +310 train 8.954944 (lr=9.9749e-05) (hash(x)=27738934) +311 train 8.756655 (lr=9.9728e-05) (hash(x)=29248942) +312 train 8.629038 (lr=9.9706e-05) (hash(x)=25103452) +313 train 8.801126 (lr=9.9683e-05) (hash(x)=25052066) +314 train 8.687764 (lr=9.9659e-05) (hash(x)=24481302) +315 train 8.575988 (lr=9.9634e-05) (hash(x)=23543273) +316 train 8.601743 (lr=9.9609e-05) (hash(x)=25608244) +317 train 8.754558 (lr=9.9582e-05) (hash(x)=27451288) +318 train 8.449246 (lr=9.9555e-05) (hash(x)=22806491) +319 train 8.626355 (lr=9.9526e-05) (hash(x)=25533417) +320 train 8.585757 (lr=9.9497e-05) (hash(x)=24557997) +321 train 8.595964 (lr=9.9467e-05) (hash(x)=24432899) +322 train 8.659699 (lr=9.9437e-05) (hash(x)=27583287) +323 train 8.480232 (lr=9.9405e-05) (hash(x)=25552036) +324 train 8.409078 (lr=9.9372e-05) (hash(x)=24201868) +325 train 8.693912 (lr=9.9339e-05) (hash(x)=28149782) +326 train 8.555322 (lr=9.9305e-05) (hash(x)=25529698) +327 train 8.395670 (lr=9.9270e-05) (hash(x)=20612533) +328 train 8.418468 (lr=9.9234e-05) (hash(x)=20699000) +329 train 8.335983 (lr=9.9197e-05) (hash(x)=19774173) +330 train 8.423822 (lr=9.9159e-05) (hash(x)=21681646) +331 train 8.278747 (lr=9.9121e-05) (hash(x)=20216795) +332 train 8.699717 (lr=9.9081e-05) (hash(x)=27697998) +333 train 8.576868 (lr=9.9041e-05) (hash(x)=25896435) +334 train 8.477157 (lr=9.9000e-05) (hash(x)=21585310) +335 train 8.483977 (lr=9.8958e-05) (hash(x)=24677740) +336 train 8.477096 (lr=9.8915e-05) (hash(x)=22027900) +337 train 8.401347 (lr=9.8872e-05) (hash(x)=21835643) +338 train 8.318322 (lr=9.8827e-05) (hash(x)=20558462) +339 train 8.353884 (lr=9.8782e-05) (hash(x)=16722715) +340 train 8.395968 (lr=9.8736e-05) (hash(x)=20972655) +341 train 8.621846 (lr=9.8689e-05) (hash(x)=26303975) +342 train 8.405646 (lr=9.8641e-05) (hash(x)=22938170) +343 train 8.395881 (lr=9.8592e-05) (hash(x)=25347203) +344 train 8.451096 (lr=9.8543e-05) (hash(x)=27398686) +345 train 8.399959 (lr=9.8492e-05) (hash(x)=25973417) +346 train 8.388673 (lr=9.8441e-05) (hash(x)=26918389) +347 train 8.342374 (lr=9.8389e-05) (hash(x)=24272489) +348 train 8.319698 (lr=9.8336e-05) (hash(x)=25593714) +349 train 8.342719 (lr=9.8282e-05) (hash(x)=29260846) +350 val loss 8.3584 +350 val perplexity 4265.9062 +350 train 8.453647 (lr=9.8228e-05) (hash(x)=27951602) +351 train 8.512511 (lr=9.8172e-05) (hash(x)=28922363) +352 train 8.419683 (lr=9.8116e-05) (hash(x)=27210734) +353 train 8.381358 (lr=9.8059e-05) (hash(x)=26322572) +354 train 8.335694 (lr=9.8001e-05) (hash(x)=27084665) +355 train 8.295417 (lr=9.7942e-05) (hash(x)=25510798) +356 train 8.358284 (lr=9.7882e-05) (hash(x)=24970921) +357 train 8.440063 (lr=9.7822e-05) (hash(x)=24138948) +358 train 8.332622 (lr=9.7761e-05) (hash(x)=24790211) +359 train 8.320186 (lr=9.7699e-05) (hash(x)=25631397) +360 train 8.242071 (lr=9.7636e-05) (hash(x)=23226625) +361 train 8.215034 (lr=9.7572e-05) (hash(x)=24001903) +362 train 8.275225 (lr=9.7507e-05) (hash(x)=24587948) +363 train 8.023534 (lr=9.7442e-05) (hash(x)=21333676) +364 train 7.936651 (lr=9.7376e-05) (hash(x)=23673779) +365 train 8.978670 (lr=9.7309e-05) (hash(x)=30770484) +366 train 8.327465 (lr=9.7241e-05) (hash(x)=26564899) +367 train 8.288922 (lr=9.7172e-05) (hash(x)=26237983) +368 train 8.099959 (lr=9.7103e-05) (hash(x)=23764356) +369 train 8.334517 (lr=9.7032e-05) (hash(x)=26205744) +370 train 8.221969 (lr=9.6961e-05) (hash(x)=19208770) +371 train 8.265100 (lr=9.6889e-05) (hash(x)=25976502) +372 train 8.101439 (lr=9.6817e-05) (hash(x)=23983933) +373 train 8.162311 (lr=9.6743e-05) (hash(x)=24080636) +374 train 8.297702 (lr=9.6669e-05) (hash(x)=24404047) +375 train 8.274593 (lr=9.6593e-05) (hash(x)=24742645) +376 train 8.226458 (lr=9.6518e-05) (hash(x)=24159600) +377 train 8.393997 (lr=9.6441e-05) (hash(x)=28677257) +378 train 8.245221 (lr=9.6363e-05) (hash(x)=25604111) +379 train 8.125875 (lr=9.6285e-05) (hash(x)=27086333) +380 train 8.151988 (lr=9.6206e-05) (hash(x)=25188207) +381 train 8.247045 (lr=9.6126e-05) (hash(x)=27855233) +382 train 8.012374 (lr=9.6045e-05) (hash(x)=19470039) +383 train 8.216577 (lr=9.5963e-05) (hash(x)=26157660) +384 train 8.172297 (lr=9.5881e-05) (hash(x)=25291570) +385 train 8.145168 (lr=9.5798e-05) (hash(x)=25046062) +386 train 8.367648 (lr=9.5714e-05) (hash(x)=27020337) +387 train 8.027189 (lr=9.5629e-05) (hash(x)=23616370) +388 train 7.782655 (lr=9.5544e-05) (hash(x)=19113218) +389 train 8.081535 (lr=9.5457e-05) (hash(x)=24302232) +390 train 8.089665 (lr=9.5370e-05) (hash(x)=22188949) +391 train 7.913142 (lr=9.5282e-05) (hash(x)=22582169) +392 train 8.162911 (lr=9.5194e-05) (hash(x)=24700570) +393 train 7.989130 (lr=9.5104e-05) (hash(x)=22773833) +394 train 7.901718 (lr=9.5014e-05) (hash(x)=21875928) +395 train 8.083673 (lr=9.4923e-05) (hash(x)=26233189) +396 train 7.965098 (lr=9.4831e-05) (hash(x)=24321467) +397 train 8.168046 (lr=9.4739e-05) (hash(x)=26431507) +398 train 8.276227 (lr=9.4646e-05) (hash(x)=28690877) +399 train 8.230265 (lr=9.4551e-05) (hash(x)=26431960) +400 val loss 8.1384 +400 val perplexity 3423.2729 +400 train 8.131128 (lr=9.4457e-05) (hash(x)=24580300) +401 train 8.095754 (lr=9.4361e-05) (hash(x)=25112360) +402 train 8.130211 (lr=9.4265e-05) (hash(x)=27597243) +403 train 8.531963 (lr=9.4168e-05) (hash(x)=30707498) +404 train 8.140757 (lr=9.4070e-05) (hash(x)=28485465) +405 train 7.969414 (lr=9.3971e-05) (hash(x)=22586447) +406 train 8.134711 (lr=9.3872e-05) (hash(x)=23175270) +407 train 8.132045 (lr=9.3772e-05) (hash(x)=25716176) +408 train 8.189781 (lr=9.3671e-05) (hash(x)=26861373) +409 train 8.339179 (lr=9.3569e-05) (hash(x)=25118971) +410 train 7.899127 (lr=9.3467e-05) (hash(x)=19829066) +411 train 7.966927 (lr=9.3364e-05) (hash(x)=26256420) +412 train 8.185081 (lr=9.3260e-05) (hash(x)=27796153) +413 train 7.989813 (lr=9.3155e-05) (hash(x)=22633318) +414 train 8.024325 (lr=9.3050e-05) (hash(x)=22589383) +415 train 8.353332 (lr=9.2944e-05) (hash(x)=28019788) +416 train 8.300772 (lr=9.2837e-05) (hash(x)=28970440) +417 train 7.982526 (lr=9.2729e-05) (hash(x)=27396089) +418 train 8.049418 (lr=9.2621e-05) (hash(x)=21183513) +419 train 8.045957 (lr=9.2512e-05) (hash(x)=23510110) +420 train 8.224435 (lr=9.2402e-05) (hash(x)=28833467) +421 train 8.118170 (lr=9.2292e-05) (hash(x)=23646926) +422 train 8.075900 (lr=9.2181e-05) (hash(x)=24697272) +423 train 7.874082 (lr=9.2069e-05) (hash(x)=20382963) +424 train 7.904645 (lr=9.1956e-05) (hash(x)=23467595) +425 train 8.025537 (lr=9.1843e-05) (hash(x)=24304768) +426 train 7.937191 (lr=9.1729e-05) (hash(x)=21392328) +427 train 7.987301 (lr=9.1614e-05) (hash(x)=25339466) +428 train 7.896797 (lr=9.1499e-05) (hash(x)=22092542) +429 train 8.040436 (lr=9.1382e-05) (hash(x)=22088696) +430 train 7.974613 (lr=9.1266e-05) (hash(x)=22184471) +431 train 8.060342 (lr=9.1148e-05) (hash(x)=24489647) +432 train 8.045403 (lr=9.1030e-05) (hash(x)=26794132) +433 train 7.867191 (lr=9.0911e-05) (hash(x)=22940357) +434 train 8.055971 (lr=9.0791e-05) (hash(x)=23719522) +435 train 7.855799 (lr=9.0671e-05) (hash(x)=22927699) +436 train 7.986556 (lr=9.0550e-05) (hash(x)=26068576) +437 train 8.125354 (lr=9.0428e-05) (hash(x)=27631132) +438 train 8.006819 (lr=9.0306e-05) (hash(x)=26739991) +439 train 8.079972 (lr=9.0182e-05) (hash(x)=25128502) +440 train 8.166019 (lr=9.0059e-05) (hash(x)=25657260) +441 train 7.940502 (lr=8.9934e-05) (hash(x)=23576982) +442 train 8.066111 (lr=8.9809e-05) (hash(x)=27117886) +443 train 8.122828 (lr=8.9683e-05) (hash(x)=25808969) +444 train 8.047095 (lr=8.9557e-05) (hash(x)=24738238) +445 train 7.968239 (lr=8.9430e-05) (hash(x)=23429962) +446 train 8.079332 (lr=8.9302e-05) (hash(x)=25075165) +447 train 7.924945 (lr=8.9173e-05) (hash(x)=25231390) +448 train 7.761856 (lr=8.9044e-05) (hash(x)=22055054) +449 train 7.997369 (lr=8.8914e-05) (hash(x)=25395441) +450 val loss 7.9922 +450 val perplexity 2957.8967 +450 train 8.028117 (lr=8.8784e-05) (hash(x)=25863209) +451 train 7.736910 (lr=8.8653e-05) (hash(x)=21154388) +452 train 7.728127 (lr=8.8521e-05) (hash(x)=21600876) +453 train 7.854784 (lr=8.8388e-05) (hash(x)=24278611) +454 train 7.776032 (lr=8.8255e-05) (hash(x)=23221720) +455 train 7.931270 (lr=8.8122e-05) (hash(x)=22708977) +456 train 8.489234 (lr=8.7987e-05) (hash(x)=23637758) +457 train 8.176578 (lr=8.7852e-05) (hash(x)=28228490) +458 train 8.209311 (lr=8.7717e-05) (hash(x)=28638071) +459 train 8.063590 (lr=8.7580e-05) (hash(x)=27258353) +460 train 8.010899 (lr=8.7444e-05) (hash(x)=26604728) +461 train 7.943528 (lr=8.7306e-05) (hash(x)=23252199) +462 train 8.028507 (lr=8.7168e-05) (hash(x)=26441427) +463 train 7.798856 (lr=8.7029e-05) (hash(x)=24364920) +464 train 7.984162 (lr=8.6890e-05) (hash(x)=25623792) +465 train 7.886867 (lr=8.6750e-05) (hash(x)=23283905) +466 train 8.215955 (lr=8.6609e-05) (hash(x)=26025267) +467 train 7.990715 (lr=8.6468e-05) (hash(x)=27243972) +468 train 8.402939 (lr=8.6326e-05) (hash(x)=30449945) +469 train 8.111275 (lr=8.6184e-05) (hash(x)=28113043) +470 train 7.908231 (lr=8.6041e-05) (hash(x)=25182521) +471 train 7.925531 (lr=8.5897e-05) (hash(x)=24932925) +472 train 7.807649 (lr=8.5753e-05) (hash(x)=20353098) +473 train 7.742064 (lr=8.5608e-05) (hash(x)=19001259) +474 train 8.211596 (lr=8.5462e-05) (hash(x)=27585685) +475 train 7.970085 (lr=8.5316e-05) (hash(x)=26371091) +476 train 7.783611 (lr=8.5170e-05) (hash(x)=24891798) +477 train 7.844266 (lr=8.5022e-05) (hash(x)=24258817) +478 train 7.923096 (lr=8.4875e-05) (hash(x)=24330263) +479 train 7.979390 (lr=8.4726e-05) (hash(x)=26913684) +480 train 7.946094 (lr=8.4577e-05) (hash(x)=26338455) +481 train 8.168668 (lr=8.4428e-05) (hash(x)=27753043) +482 train 7.908193 (lr=8.4278e-05) (hash(x)=26123289) +483 train 8.023372 (lr=8.4127e-05) (hash(x)=29239611) +484 train 7.935291 (lr=8.3976e-05) (hash(x)=26553003) +485 train 7.919011 (lr=8.3824e-05) (hash(x)=22984557) +486 train 7.507417 (lr=8.3672e-05) (hash(x)=16947491) +487 train 7.491841 (lr=8.3519e-05) (hash(x)=18017792) +488 train 7.641500 (lr=8.3366e-05) (hash(x)=19918608) +489 train 8.019450 (lr=8.3212e-05) (hash(x)=23374526) +490 train 7.927112 (lr=8.3057e-05) (hash(x)=25009505) +491 train 7.885498 (lr=8.2902e-05) (hash(x)=27574089) +492 train 8.016278 (lr=8.2746e-05) (hash(x)=24122664) +493 train 7.904282 (lr=8.2590e-05) (hash(x)=26154906) +494 train 7.857390 (lr=8.2434e-05) (hash(x)=25192767) +495 train 7.969627 (lr=8.2276e-05) (hash(x)=28613882) +496 train 7.881052 (lr=8.2119e-05) (hash(x)=23547219) +497 train 7.871373 (lr=8.1960e-05) (hash(x)=25272182) +498 train 7.752473 (lr=8.1801e-05) (hash(x)=24992761) +499 train 8.016029 (lr=8.1642e-05) (hash(x)=26981914) +500 val loss 7.8733 +500 val perplexity 2626.1943 +500 train 7.783356 (lr=8.1482e-05) (hash(x)=22051933) +501 train 7.988568 (lr=8.1322e-05) (hash(x)=24232348) +502 train 7.978487 (lr=8.1161e-05) (hash(x)=23158331) +503 train 7.777245 (lr=8.1000e-05) (hash(x)=22652243) +504 train 7.818677 (lr=8.0838e-05) (hash(x)=23805602) +505 train 8.084427 (lr=8.0676e-05) (hash(x)=25411991) +506 train 7.453513 (lr=8.0513e-05) (hash(x)=18827215) +507 train 7.165349 (lr=8.0349e-05) (hash(x)=15446025) +508 train 7.501356 (lr=8.0186e-05) (hash(x)=20516263) +509 train 7.968163 (lr=8.0021e-05) (hash(x)=27846176) +510 train 7.680051 (lr=7.9856e-05) (hash(x)=23342449) +511 train 7.917188 (lr=7.9691e-05) (hash(x)=27194521) +512 train 7.749077 (lr=7.9525e-05) (hash(x)=23008284) +513 train 8.137077 (lr=7.9359e-05) (hash(x)=29430001) +514 train 7.539665 (lr=7.9192e-05) (hash(x)=22579319) +515 train 7.601607 (lr=7.9025e-05) (hash(x)=25264518) +516 train 7.888702 (lr=7.8858e-05) (hash(x)=25359075) +517 train 7.872169 (lr=7.8689e-05) (hash(x)=25568956) +518 train 8.134679 (lr=7.8521e-05) (hash(x)=32004108) +519 train 7.812849 (lr=7.8352e-05) (hash(x)=24936836) +520 train 7.941904 (lr=7.8182e-05) (hash(x)=27263338) +521 train 7.975307 (lr=7.8012e-05) (hash(x)=27452099) +522 train 7.871152 (lr=7.7842e-05) (hash(x)=25965406) +523 train 8.059302 (lr=7.7671e-05) (hash(x)=28197282) +524 train 7.798197 (lr=7.7500e-05) (hash(x)=22466209) +525 train 7.834746 (lr=7.7328e-05) (hash(x)=22931889) +526 train 7.953221 (lr=7.7156e-05) (hash(x)=26903920) +527 train 7.854158 (lr=7.6984e-05) (hash(x)=24765578) +528 train 7.946190 (lr=7.6811e-05) (hash(x)=27811359) +529 train 7.726755 (lr=7.6637e-05) (hash(x)=25078649) +530 train 7.910915 (lr=7.6463e-05) (hash(x)=25572416) +531 train 7.985146 (lr=7.6289e-05) (hash(x)=27448185) +532 train 8.055017 (lr=7.6115e-05) (hash(x)=25923719) +533 train 7.854694 (lr=7.5940e-05) (hash(x)=24804856) +534 train 7.987044 (lr=7.5764e-05) (hash(x)=23207829) +535 train 7.867074 (lr=7.5588e-05) (hash(x)=23107416) +536 train 7.747884 (lr=7.5412e-05) (hash(x)=26739531) +537 train 7.753458 (lr=7.5235e-05) (hash(x)=24960796) +538 train 7.875792 (lr=7.5058e-05) (hash(x)=24667802) +539 train 7.986289 (lr=7.4881e-05) (hash(x)=26755138) +540 train 7.831022 (lr=7.4703e-05) (hash(x)=25537132) +541 train 7.768228 (lr=7.4525e-05) (hash(x)=24542526) +542 train 7.589191 (lr=7.4346e-05) (hash(x)=21296355) +543 train 8.026015 (lr=7.4167e-05) (hash(x)=29314255) +544 train 7.836854 (lr=7.3988e-05) (hash(x)=26001799) +545 train 7.890655 (lr=7.3808e-05) (hash(x)=27347755) +546 train 7.848896 (lr=7.3628e-05) (hash(x)=25107798) +547 train 7.670450 (lr=7.3448e-05) (hash(x)=22112669) +548 train 7.663572 (lr=7.3267e-05) (hash(x)=21897967) +549 train 7.768433 (lr=7.3086e-05) (hash(x)=25161929) +550 val loss 7.8174 +550 val perplexity 2483.3381 +550 train 7.731859 (lr=7.2904e-05) (hash(x)=27465106) +551 train 7.841363 (lr=7.2723e-05) (hash(x)=24013079) +552 train 7.760349 (lr=7.2540e-05) (hash(x)=23142015) +553 train 7.803506 (lr=7.2358e-05) (hash(x)=26768629) +554 train 7.847647 (lr=7.2175e-05) (hash(x)=26393383) +555 train 7.599916 (lr=7.1992e-05) (hash(x)=22537194) +556 train 7.850276 (lr=7.1808e-05) (hash(x)=24046036) +557 train 7.816450 (lr=7.1624e-05) (hash(x)=24974360) +558 train 8.047343 (lr=7.1440e-05) (hash(x)=28379928) +559 train 7.835066 (lr=7.1256e-05) (hash(x)=25322001) +560 train 7.919348 (lr=7.1071e-05) (hash(x)=26622031) +561 train 7.561779 (lr=7.0886e-05) (hash(x)=20562247) +562 train 7.957068 (lr=7.0701e-05) (hash(x)=27381885) +563 train 8.109596 (lr=7.0515e-05) (hash(x)=27028126) +564 train 7.922890 (lr=7.0329e-05) (hash(x)=28882928) +565 train 7.868667 (lr=7.0143e-05) (hash(x)=25666355) +566 train 7.941362 (lr=6.9956e-05) (hash(x)=24330810) +567 train 7.866081 (lr=6.9769e-05) (hash(x)=26690440) +568 train 7.733994 (lr=6.9582e-05) (hash(x)=22923592) +569 train 7.857698 (lr=6.9394e-05) (hash(x)=27348418) +570 train 7.885051 (lr=6.9207e-05) (hash(x)=28849848) +571 train 7.956679 (lr=6.9019e-05) (hash(x)=26967331) +572 train 7.692927 (lr=6.8830e-05) (hash(x)=22831467) +573 train 7.794309 (lr=6.8642e-05) (hash(x)=24765121) +574 train 7.778310 (lr=6.8453e-05) (hash(x)=24331857) +575 train 7.607823 (lr=6.8264e-05) (hash(x)=22598512) +576 train 7.795448 (lr=6.8075e-05) (hash(x)=25149353) +577 train 7.676713 (lr=6.7885e-05) (hash(x)=23725598) +578 train 7.906079 (lr=6.7695e-05) (hash(x)=26449557) +579 train 7.897598 (lr=6.7505e-05) (hash(x)=24697985) +580 train 7.909112 (lr=6.7315e-05) (hash(x)=26923059) +581 train 7.731369 (lr=6.7124e-05) (hash(x)=25201962) +582 train 7.406195 (lr=6.6933e-05) (hash(x)=20931520) +583 train 7.458481 (lr=6.6742e-05) (hash(x)=18473911) +584 train 7.605979 (lr=6.6551e-05) (hash(x)=21306267) +585 train 7.885912 (lr=6.6360e-05) (hash(x)=25982840) +586 train 7.771975 (lr=6.6168e-05) (hash(x)=25364874) +587 train 7.679974 (lr=6.5976e-05) (hash(x)=23172124) +588 train 7.930065 (lr=6.5784e-05) (hash(x)=27876897) +589 train 8.523589 (lr=6.5592e-05) (hash(x)=34646114) +590 train 8.861931 (lr=6.5399e-05) (hash(x)=35153576) +591 train 7.691928 (lr=6.5206e-05) (hash(x)=22322442) +592 train 7.969649 (lr=6.5013e-05) (hash(x)=27907331) +593 train 7.878075 (lr=6.4820e-05) (hash(x)=26211794) +594 train 8.082365 (lr=6.4627e-05) (hash(x)=29291512) +595 train 8.333148 (lr=6.4433e-05) (hash(x)=29659121) +596 train 8.117756 (lr=6.4240e-05) (hash(x)=29674399) +597 train 7.782672 (lr=6.4046e-05) (hash(x)=23538306) +598 train 7.891967 (lr=6.3852e-05) (hash(x)=21991524) +599 train 7.847769 (lr=6.3658e-05) (hash(x)=26324153) +600 val loss 7.8193 +600 val perplexity 2488.2786 +600 train 7.817334 (lr=6.3463e-05) (hash(x)=23712082) +601 train 7.840537 (lr=6.3269e-05) (hash(x)=24910403) +602 train 7.800004 (lr=6.3074e-05) (hash(x)=26737205) +603 train 7.947093 (lr=6.2879e-05) (hash(x)=26939970) +604 train 7.811485 (lr=6.2684e-05) (hash(x)=27651943) +605 train 8.105732 (lr=6.2489e-05) (hash(x)=27515446) +606 train 7.979828 (lr=6.2294e-05) (hash(x)=26753129) +607 train 7.705410 (lr=6.2098e-05) (hash(x)=23446058) +608 train 7.975355 (lr=6.1903e-05) (hash(x)=27587849) +609 train 7.883319 (lr=6.1707e-05) (hash(x)=25308253) +610 train 7.888650 (lr=6.1511e-05) (hash(x)=26615098) +611 train 7.911775 (lr=6.1315e-05) (hash(x)=29981801) +612 train 8.003817 (lr=6.1119e-05) (hash(x)=29592345) +613 train 8.023692 (lr=6.0923e-05) (hash(x)=23470413) +614 train 7.926247 (lr=6.0726e-05) (hash(x)=24742370) +615 train 7.905869 (lr=6.0530e-05) (hash(x)=24843741) +616 train 7.832581 (lr=6.0333e-05) (hash(x)=25192548) +617 train 7.918457 (lr=6.0137e-05) (hash(x)=27176996) +618 train 7.769153 (lr=5.9940e-05) (hash(x)=23964552) +619 train 7.757080 (lr=5.9743e-05) (hash(x)=22855363) +620 train 7.713716 (lr=5.9546e-05) (hash(x)=26332996) +621 train 7.561992 (lr=5.9349e-05) (hash(x)=22960957) +622 train 7.884956 (lr=5.9152e-05) (hash(x)=22752597) +623 train 7.786986 (lr=5.8955e-05) (hash(x)=23197102) +624 train 8.231909 (lr=5.8758e-05) (hash(x)=27383319) +625 train 7.831959 (lr=5.8560e-05) (hash(x)=26132276) +626 train 7.700591 (lr=5.8363e-05) (hash(x)=19810497) +627 train 7.799007 (lr=5.8165e-05) (hash(x)=25704919) +628 train 7.916129 (lr=5.7968e-05) (hash(x)=27174264) +629 train 8.025556 (lr=5.7770e-05) (hash(x)=22280814) +630 train 8.735373 (lr=5.7573e-05) (hash(x)=24279448) +631 train 8.208377 (lr=5.7375e-05) (hash(x)=23054940) +632 train 8.304580 (lr=5.7177e-05) (hash(x)=23942400) +633 train 8.580954 (lr=5.6979e-05) (hash(x)=24712416) +634 train 8.168818 (lr=5.6782e-05) (hash(x)=24564658) +635 train 7.830007 (lr=5.6584e-05) (hash(x)=24909904) +636 train 7.885648 (lr=5.6386e-05) (hash(x)=23049534) +637 train 7.924608 (lr=5.6188e-05) (hash(x)=24321591) +638 train 7.720563 (lr=5.5990e-05) (hash(x)=26153298) +639 train 7.791595 (lr=5.5792e-05) (hash(x)=27140757) +640 train 7.574867 (lr=5.5594e-05) (hash(x)=25115907) +641 train 7.653893 (lr=5.5396e-05) (hash(x)=26563770) +642 train 7.826556 (lr=5.5198e-05) (hash(x)=28089252) +643 train 7.752496 (lr=5.5000e-05) (hash(x)=25993110) +644 train 7.730805 (lr=5.4802e-05) (hash(x)=25150008) +645 train 7.924302 (lr=5.4604e-05) (hash(x)=28520222) +646 train 7.589480 (lr=5.4406e-05) (hash(x)=21349943) +647 train 7.656353 (lr=5.4208e-05) (hash(x)=25149419) +648 train 7.879052 (lr=5.4010e-05) (hash(x)=25730641) +649 train 7.975147 (lr=5.3812e-05) (hash(x)=26112813) +650 val loss 7.7714 +650 val perplexity 2371.7510 +650 train 7.800326 (lr=5.3614e-05) (hash(x)=25907805) +651 train 7.841703 (lr=5.3416e-05) (hash(x)=27623643) +652 train 7.811011 (lr=5.3218e-05) (hash(x)=26484959) +653 train 7.877267 (lr=5.3021e-05) (hash(x)=29199854) +654 train 7.866321 (lr=5.2823e-05) (hash(x)=28369628) +655 train 7.744977 (lr=5.2625e-05) (hash(x)=24727764) +656 train 7.458787 (lr=5.2427e-05) (hash(x)=22610673) +657 train 7.372344 (lr=5.2230e-05) (hash(x)=22667179) +658 train 7.200322 (lr=5.2032e-05) (hash(x)=18477300) +659 train 7.620136 (lr=5.1835e-05) (hash(x)=23155773) +660 train 7.419064 (lr=5.1637e-05) (hash(x)=19461032) +661 train 7.697927 (lr=5.1440e-05) (hash(x)=23453788) +662 train 7.788021 (lr=5.1242e-05) (hash(x)=24543466) +663 train 7.593221 (lr=5.1045e-05) (hash(x)=21935931) +664 train 7.468285 (lr=5.0848e-05) (hash(x)=19910292) +665 train 7.702228 (lr=5.0651e-05) (hash(x)=24481079) +666 train 7.442731 (lr=5.0454e-05) (hash(x)=18922411) +667 train 7.459817 (lr=5.0257e-05) (hash(x)=20054917) +668 train 7.704576 (lr=5.0060e-05) (hash(x)=24850470) +669 train 7.707784 (lr=4.9863e-05) (hash(x)=25907741) +670 train 7.853311 (lr=4.9667e-05) (hash(x)=26873522) +671 train 7.875531 (lr=4.9470e-05) (hash(x)=27606073) +672 train 7.969500 (lr=4.9274e-05) (hash(x)=26209645) +673 train 7.826120 (lr=4.9077e-05) (hash(x)=25202001) +674 train 7.875925 (lr=4.8881e-05) (hash(x)=25569462) +675 train 7.839563 (lr=4.8685e-05) (hash(x)=26534487) +676 train 7.829555 (lr=4.8489e-05) (hash(x)=26455057) +677 train 7.702078 (lr=4.8293e-05) (hash(x)=24095850) +678 train 7.912537 (lr=4.8097e-05) (hash(x)=25287752) +679 train 7.441053 (lr=4.7902e-05) (hash(x)=22450341) +680 train 8.134704 (lr=4.7706e-05) (hash(x)=29004853) +681 train 7.887872 (lr=4.7511e-05) (hash(x)=27993763) +682 train 7.693124 (lr=4.7316e-05) (hash(x)=26382658) +683 train 7.850461 (lr=4.7121e-05) (hash(x)=25013073) +684 train 7.893066 (lr=4.6926e-05) (hash(x)=30595809) +685 train 8.007025 (lr=4.6731e-05) (hash(x)=30934371) +686 train 8.514073 (lr=4.6537e-05) (hash(x)=33060834) +687 train 7.800910 (lr=4.6342e-05) (hash(x)=25945859) +688 train 7.644035 (lr=4.6148e-05) (hash(x)=23375678) +689 train 7.701813 (lr=4.5954e-05) (hash(x)=25218689) +690 train 7.828738 (lr=4.5760e-05) (hash(x)=28127397) +691 train 7.754880 (lr=4.5567e-05) (hash(x)=24418091) +692 train 7.679464 (lr=4.5373e-05) (hash(x)=22761099) +693 train 7.666978 (lr=4.5180e-05) (hash(x)=24615466) +694 train 7.825765 (lr=4.4987e-05) (hash(x)=24699240) +695 train 7.719493 (lr=4.4794e-05) (hash(x)=21613707) +696 train 7.750179 (lr=4.4601e-05) (hash(x)=24977554) +697 train 7.744665 (lr=4.4408e-05) (hash(x)=24348175) +698 train 7.863518 (lr=4.4216e-05) (hash(x)=25102767) +699 train 7.827512 (lr=4.4024e-05) (hash(x)=26386157) +700 val loss 7.7499 +700 val perplexity 2321.3655 +700 train 7.787271 (lr=4.3832e-05) (hash(x)=26423460) +701 train 7.719305 (lr=4.3640e-05) (hash(x)=25771047) +702 train 7.783035 (lr=4.3449e-05) (hash(x)=26999875) +703 train 7.702168 (lr=4.3258e-05) (hash(x)=24396519) +704 train 7.732126 (lr=4.3067e-05) (hash(x)=22588122) +705 train 8.020427 (lr=4.2876e-05) (hash(x)=25142399) +706 train 7.753125 (lr=4.2685e-05) (hash(x)=20440214) +707 train 7.629843 (lr=4.2495e-05) (hash(x)=23265507) +708 train 7.709021 (lr=4.2305e-05) (hash(x)=24563470) +709 train 7.637650 (lr=4.2115e-05) (hash(x)=22514858) +710 train 7.971630 (lr=4.1925e-05) (hash(x)=26691212) +711 train 8.013155 (lr=4.1736e-05) (hash(x)=29138828) +712 train 7.984135 (lr=4.1547e-05) (hash(x)=28028528) +713 train 7.797307 (lr=4.1358e-05) (hash(x)=20531210) +714 train 7.721904 (lr=4.1170e-05) (hash(x)=25075352) +715 train 7.781747 (lr=4.0981e-05) (hash(x)=24265353) +716 train 7.866539 (lr=4.0793e-05) (hash(x)=24635726) +717 train 7.723149 (lr=4.0606e-05) (hash(x)=24999726) +718 train 7.921286 (lr=4.0418e-05) (hash(x)=27412910) +719 train 7.725751 (lr=4.0231e-05) (hash(x)=24685515) +720 train 7.791103 (lr=4.0044e-05) (hash(x)=23780329) +721 train 7.620557 (lr=3.9857e-05) (hash(x)=25071701) +722 train 7.758102 (lr=3.9671e-05) (hash(x)=23767130) +723 train 7.716944 (lr=3.9485e-05) (hash(x)=24876269) +724 train 7.835083 (lr=3.9299e-05) (hash(x)=26405773) +725 train 8.199580 (lr=3.9114e-05) (hash(x)=31733180) +726 train 7.487280 (lr=3.8929e-05) (hash(x)=21337509) +727 train 7.566322 (lr=3.8744e-05) (hash(x)=22825749) +728 train 7.910472 (lr=3.8560e-05) (hash(x)=28638695) +729 train 7.801475 (lr=3.8376e-05) (hash(x)=26393943) +730 train 7.962570 (lr=3.8192e-05) (hash(x)=27563583) +731 train 7.525585 (lr=3.8008e-05) (hash(x)=21239652) +732 train 7.655273 (lr=3.7825e-05) (hash(x)=23986428) +733 train 7.543106 (lr=3.7642e-05) (hash(x)=24943881) +734 train 8.073760 (lr=3.7460e-05) (hash(x)=29691448) +735 train 8.001682 (lr=3.7277e-05) (hash(x)=28767869) +736 train 7.738295 (lr=3.7096e-05) (hash(x)=23628188) +737 train 8.041641 (lr=3.6914e-05) (hash(x)=29341482) +738 train 7.851739 (lr=3.6733e-05) (hash(x)=30336570) +739 train 7.790923 (lr=3.6552e-05) (hash(x)=25614301) +740 train 7.693447 (lr=3.6372e-05) (hash(x)=24160500) +741 train 7.868117 (lr=3.6192e-05) (hash(x)=26030058) +742 train 8.210966 (lr=3.6012e-05) (hash(x)=29243936) +743 train 7.531816 (lr=3.5833e-05) (hash(x)=21159060) +744 train 7.645225 (lr=3.5654e-05) (hash(x)=23701853) +745 train 7.834294 (lr=3.5475e-05) (hash(x)=24629937) +746 train 7.727683 (lr=3.5297e-05) (hash(x)=25110108) +747 train 7.789520 (lr=3.5119e-05) (hash(x)=26751788) +748 train 7.833237 (lr=3.4942e-05) (hash(x)=26430427) +749 train 7.668863 (lr=3.4765e-05) (hash(x)=26012353) +750 val loss 7.7434 +750 val perplexity 2306.2227 +750 train 7.839715 (lr=3.4588e-05) (hash(x)=22735910) +751 train 7.613408 (lr=3.4412e-05) (hash(x)=25045397) +752 train 7.367143 (lr=3.4236e-05) (hash(x)=21554427) +753 train 7.464153 (lr=3.4060e-05) (hash(x)=23751143) +754 train 8.234746 (lr=3.3885e-05) (hash(x)=28602273) +755 train 8.276996 (lr=3.3711e-05) (hash(x)=29989709) +756 train 7.662009 (lr=3.3537e-05) (hash(x)=22331648) +757 train 8.098683 (lr=3.3363e-05) (hash(x)=31017246) +758 train 8.010327 (lr=3.3189e-05) (hash(x)=29709045) +759 train 8.030388 (lr=3.3016e-05) (hash(x)=25560928) +760 train 7.742786 (lr=3.2844e-05) (hash(x)=25075464) +761 train 7.827994 (lr=3.2672e-05) (hash(x)=27352253) +762 train 8.047679 (lr=3.2500e-05) (hash(x)=28187891) +763 train 7.877928 (lr=3.2329e-05) (hash(x)=26062687) +764 train 7.926925 (lr=3.2158e-05) (hash(x)=27427811) +765 train 7.977590 (lr=3.1988e-05) (hash(x)=27614522) +766 train 7.741851 (lr=3.1818e-05) (hash(x)=26129544) +767 train 8.235064 (lr=3.1648e-05) (hash(x)=28959222) +768 train 7.849905 (lr=3.1479e-05) (hash(x)=26860067) +769 train 7.649452 (lr=3.1311e-05) (hash(x)=25122598) +770 train 7.741111 (lr=3.1142e-05) (hash(x)=25245030) +771 train 7.740355 (lr=3.0975e-05) (hash(x)=25434884) +772 train 7.841005 (lr=3.0808e-05) (hash(x)=27732790) +773 train 7.863842 (lr=3.0641e-05) (hash(x)=27824438) +774 train 7.936987 (lr=3.0475e-05) (hash(x)=27201953) +775 train 7.651192 (lr=3.0309e-05) (hash(x)=21248405) +776 train 7.505951 (lr=3.0144e-05) (hash(x)=22805934) +777 train 7.764906 (lr=2.9979e-05) (hash(x)=26482588) +778 train 7.813489 (lr=2.9814e-05) (hash(x)=24153691) +779 train 7.946086 (lr=2.9651e-05) (hash(x)=25044192) +780 train 7.876826 (lr=2.9487e-05) (hash(x)=25910078) +781 train 7.879182 (lr=2.9324e-05) (hash(x)=28645524) +782 train 7.694258 (lr=2.9162e-05) (hash(x)=24368498) +783 train 7.774515 (lr=2.9000e-05) (hash(x)=25830182) +784 train 7.849136 (lr=2.8839e-05) (hash(x)=29181807) +785 train 7.758777 (lr=2.8678e-05) (hash(x)=25585137) +786 train 7.728965 (lr=2.8518e-05) (hash(x)=24798246) +787 train 7.831422 (lr=2.8358e-05) (hash(x)=26621419) +788 train 7.395689 (lr=2.8199e-05) (hash(x)=21446891) +789 train 7.342617 (lr=2.8040e-05) (hash(x)=22165286) +790 train 7.565431 (lr=2.7881e-05) (hash(x)=23477219) +791 train 7.841313 (lr=2.7724e-05) (hash(x)=25173113) +792 train 7.730171 (lr=2.7566e-05) (hash(x)=25853788) +793 train 7.835870 (lr=2.7410e-05) (hash(x)=27267091) +794 train 7.647830 (lr=2.7254e-05) (hash(x)=23743694) +795 train 7.639535 (lr=2.7098e-05) (hash(x)=24400133) +796 train 7.712384 (lr=2.6943e-05) (hash(x)=23663639) +797 train 7.589183 (lr=2.6788e-05) (hash(x)=23103223) +798 train 8.118289 (lr=2.6634e-05) (hash(x)=28748411) +799 train 7.457672 (lr=2.6481e-05) (hash(x)=23486277) +800 val loss 7.7216 +800 val perplexity 2256.4963 +800 train 7.690685 (lr=2.6328e-05) (hash(x)=25678518) +801 train 7.637488 (lr=2.6176e-05) (hash(x)=23421286) +802 train 7.804496 (lr=2.6024e-05) (hash(x)=26054104) +803 train 7.816794 (lr=2.5873e-05) (hash(x)=25978130) +804 train 7.838860 (lr=2.5722e-05) (hash(x)=26006525) +805 train 7.758560 (lr=2.5572e-05) (hash(x)=25769432) +806 train 7.573352 (lr=2.5423e-05) (hash(x)=22430795) +807 train 8.008194 (lr=2.5274e-05) (hash(x)=28916006) +808 train 7.749007 (lr=2.5125e-05) (hash(x)=25166800) +809 train 7.848123 (lr=2.4978e-05) (hash(x)=24226056) +810 train 7.712584 (lr=2.4830e-05) (hash(x)=25057934) +811 train 8.433208 (lr=2.4684e-05) (hash(x)=33195100) +812 train 7.881772 (lr=2.4538e-05) (hash(x)=26312888) +813 train 8.057498 (lr=2.4392e-05) (hash(x)=27730410) +814 train 7.931027 (lr=2.4247e-05) (hash(x)=27372474) +815 train 7.782936 (lr=2.4103e-05) (hash(x)=25556929) +816 train 7.832053 (lr=2.3959e-05) (hash(x)=26909985) +817 train 7.654505 (lr=2.3816e-05) (hash(x)=25991247) +818 train 7.855127 (lr=2.3674e-05) (hash(x)=27438141) +819 train 8.025977 (lr=2.3532e-05) (hash(x)=29536986) +820 train 7.742598 (lr=2.3391e-05) (hash(x)=24478391) +821 train 7.740557 (lr=2.3250e-05) (hash(x)=26125216) +822 train 7.779691 (lr=2.3110e-05) (hash(x)=26422130) +823 train 7.985797 (lr=2.2971e-05) (hash(x)=29648798) +824 train 7.620800 (lr=2.2832e-05) (hash(x)=21247770) +825 train 7.758401 (lr=2.2694e-05) (hash(x)=23195388) +826 train 7.717407 (lr=2.2556e-05) (hash(x)=25796725) +827 train 7.762967 (lr=2.2420e-05) (hash(x)=23124767) +828 train 7.808054 (lr=2.2283e-05) (hash(x)=25233464) +829 train 7.781078 (lr=2.2148e-05) (hash(x)=25713275) +830 train 7.725812 (lr=2.2013e-05) (hash(x)=25550167) +831 train 7.632356 (lr=2.1878e-05) (hash(x)=24976217) +832 train 7.930112 (lr=2.1745e-05) (hash(x)=28536827) +833 train 7.817445 (lr=2.1612e-05) (hash(x)=27500801) +834 train 7.637158 (lr=2.1479e-05) (hash(x)=25545765) +835 train 7.684369 (lr=2.1347e-05) (hash(x)=23632825) +836 train 7.726700 (lr=2.1216e-05) (hash(x)=25708009) +837 train 7.775418 (lr=2.1086e-05) (hash(x)=24456276) +838 train 7.876058 (lr=2.0956e-05) (hash(x)=29189855) +839 train 8.083389 (lr=2.0827e-05) (hash(x)=31019606) +840 train 7.940263 (lr=2.0698e-05) (hash(x)=26328013) +841 train 7.873227 (lr=2.0570e-05) (hash(x)=25027904) +842 train 7.659144 (lr=2.0443e-05) (hash(x)=23734189) +843 train 7.976398 (lr=2.0317e-05) (hash(x)=28236580) +844 train 7.718137 (lr=2.0191e-05) (hash(x)=26509780) +845 train 7.772357 (lr=2.0066e-05) (hash(x)=25386473) +846 train 7.757452 (lr=1.9941e-05) (hash(x)=24052671) +847 train 7.940616 (lr=1.9818e-05) (hash(x)=28269421) +848 train 7.365494 (lr=1.9694e-05) (hash(x)=22251724) +849 train 7.625635 (lr=1.9572e-05) (hash(x)=24308447) +850 val loss 7.7320 +850 val perplexity 2280.0859 +850 train 7.606493 (lr=1.9450e-05) (hash(x)=24242830) +851 train 7.777753 (lr=1.9329e-05) (hash(x)=25563279) +852 train 7.796778 (lr=1.9209e-05) (hash(x)=26354481) +853 train 7.878568 (lr=1.9089e-05) (hash(x)=26152637) +854 train 7.986140 (lr=1.8970e-05) (hash(x)=28051025) +855 train 7.762688 (lr=1.8852e-05) (hash(x)=24865358) +856 train 7.648006 (lr=1.8734e-05) (hash(x)=24288911) +857 train 7.552729 (lr=1.8618e-05) (hash(x)=22230964) +858 train 7.466060 (lr=1.8501e-05) (hash(x)=21303832) +859 train 7.615192 (lr=1.8386e-05) (hash(x)=22155546) +860 train 7.729790 (lr=1.8271e-05) (hash(x)=25296428) +861 train 7.881613 (lr=1.8157e-05) (hash(x)=29142319) +862 train 7.724305 (lr=1.8044e-05) (hash(x)=25545430) +863 train 7.549764 (lr=1.7931e-05) (hash(x)=26984272) +864 train 7.730348 (lr=1.7819e-05) (hash(x)=25429005) +865 train 7.791476 (lr=1.7708e-05) (hash(x)=27077032) +866 train 7.741840 (lr=1.7598e-05) (hash(x)=26494424) +867 train 7.723072 (lr=1.7488e-05) (hash(x)=23193673) +868 train 7.778479 (lr=1.7379e-05) (hash(x)=25075134) +869 train 7.835979 (lr=1.7271e-05) (hash(x)=27112558) +870 train 7.966924 (lr=1.7163e-05) (hash(x)=27436608) +871 train 7.675697 (lr=1.7056e-05) (hash(x)=24544116) +872 train 8.011573 (lr=1.6950e-05) (hash(x)=31632686) +873 train 7.693210 (lr=1.6845e-05) (hash(x)=25890184) +874 train 7.782525 (lr=1.6740e-05) (hash(x)=22887555) +875 train 7.673402 (lr=1.6636e-05) (hash(x)=24547533) +876 train 7.801500 (lr=1.6533e-05) (hash(x)=26553496) +877 train 7.894932 (lr=1.6431e-05) (hash(x)=27467688) +878 train 7.642717 (lr=1.6329e-05) (hash(x)=24766934) +879 train 7.557905 (lr=1.6228e-05) (hash(x)=22059850) +880 train 7.918352 (lr=1.6128e-05) (hash(x)=22871702) +881 train 7.741314 (lr=1.6029e-05) (hash(x)=23893130) +882 train 7.812656 (lr=1.5930e-05) (hash(x)=25125691) +883 train 7.753754 (lr=1.5832e-05) (hash(x)=25994573) +884 train 7.798635 (lr=1.5735e-05) (hash(x)=26076345) +885 train 7.740541 (lr=1.5639e-05) (hash(x)=26577783) +886 train 7.848183 (lr=1.5543e-05) (hash(x)=27395225) +887 train 7.683504 (lr=1.5449e-05) (hash(x)=23926632) +888 train 7.549201 (lr=1.5354e-05) (hash(x)=21737239) +889 train 7.799978 (lr=1.5261e-05) (hash(x)=23574207) +890 train 7.512405 (lr=1.5169e-05) (hash(x)=24365231) +891 train 7.699380 (lr=1.5077e-05) (hash(x)=27111369) +892 train 7.924301 (lr=1.4986e-05) (hash(x)=27290015) +893 train 7.714011 (lr=1.4896e-05) (hash(x)=23979820) +894 train 7.817277 (lr=1.4806e-05) (hash(x)=26450121) +895 train 7.709533 (lr=1.4718e-05) (hash(x)=27025333) +896 train 7.688375 (lr=1.4630e-05) (hash(x)=23624605) +897 train 7.487306 (lr=1.4543e-05) (hash(x)=22846386) +898 train 7.620155 (lr=1.4456e-05) (hash(x)=22970561) +899 train 7.428380 (lr=1.4371e-05) (hash(x)=16908068) +900 val loss 7.7131 +900 val perplexity 2237.4890 +900 train 7.671172 (lr=1.4286e-05) (hash(x)=24661446) +901 train 7.749739 (lr=1.4202e-05) (hash(x)=25664727) +902 train 7.758882 (lr=1.4119e-05) (hash(x)=25667011) +903 train 7.821552 (lr=1.4037e-05) (hash(x)=29120407) +904 train 7.800092 (lr=1.3955e-05) (hash(x)=23385735) +905 train 7.854821 (lr=1.3874e-05) (hash(x)=25564213) +906 train 7.928378 (lr=1.3794e-05) (hash(x)=25413898) +907 train 7.893826 (lr=1.3715e-05) (hash(x)=27092710) +908 train 7.700174 (lr=1.3637e-05) (hash(x)=25789923) +909 train 7.805463 (lr=1.3559e-05) (hash(x)=28533197) +910 train 7.628718 (lr=1.3482e-05) (hash(x)=22982996) +911 train 7.715003 (lr=1.3407e-05) (hash(x)=23827393) +912 train 7.352606 (lr=1.3331e-05) (hash(x)=21242640) +913 train 7.713574 (lr=1.3257e-05) (hash(x)=24154233) +914 train 7.720458 (lr=1.3183e-05) (hash(x)=24331967) +915 train 7.941110 (lr=1.3111e-05) (hash(x)=32812727) +916 train 7.627933 (lr=1.3039e-05) (hash(x)=23572994) +917 train 7.686458 (lr=1.2968e-05) (hash(x)=26305435) +918 train 7.690072 (lr=1.2897e-05) (hash(x)=26268355) +919 train 7.874236 (lr=1.2828e-05) (hash(x)=27230027) +920 train 7.698805 (lr=1.2759e-05) (hash(x)=23885377) +921 train 7.596417 (lr=1.2691e-05) (hash(x)=23532437) +922 train 7.716924 (lr=1.2624e-05) (hash(x)=25577034) +923 train 7.834340 (lr=1.2558e-05) (hash(x)=25703381) +924 train 7.829640 (lr=1.2493e-05) (hash(x)=27113866) +925 train 7.668754 (lr=1.2428e-05) (hash(x)=26961429) +926 train 7.476012 (lr=1.2364e-05) (hash(x)=21355372) +927 train 7.664171 (lr=1.2301e-05) (hash(x)=24968260) +928 train 7.627688 (lr=1.2239e-05) (hash(x)=25357517) +929 train 7.683904 (lr=1.2178e-05) (hash(x)=24854265) +930 train 7.450107 (lr=1.2118e-05) (hash(x)=21102770) +931 train 7.773852 (lr=1.2058e-05) (hash(x)=25676468) +932 train 7.697301 (lr=1.1999e-05) (hash(x)=22809869) +933 train 7.792577 (lr=1.1941e-05) (hash(x)=25503865) +934 train 7.650753 (lr=1.1884e-05) (hash(x)=24853995) +935 train 7.819744 (lr=1.1828e-05) (hash(x)=27544803) +936 train 7.852772 (lr=1.1772e-05) (hash(x)=25981933) +937 train 7.573801 (lr=1.1718e-05) (hash(x)=24658683) +938 train 7.755859 (lr=1.1664e-05) (hash(x)=23855201) +939 train 7.710023 (lr=1.1611e-05) (hash(x)=24331407) +940 train 7.912000 (lr=1.1559e-05) (hash(x)=29265551) +941 train 7.590758 (lr=1.1508e-05) (hash(x)=21892556) +942 train 7.624099 (lr=1.1457e-05) (hash(x)=27183405) +943 train 7.631451 (lr=1.1408e-05) (hash(x)=26540663) +944 train 7.579454 (lr=1.1359e-05) (hash(x)=25718393) +945 train 7.758939 (lr=1.1311e-05) (hash(x)=26819462) +946 train 7.761447 (lr=1.1264e-05) (hash(x)=27427540) +947 train 7.732384 (lr=1.1218e-05) (hash(x)=25532657) +948 train 7.962454 (lr=1.1173e-05) (hash(x)=27641372) +949 train 7.692699 (lr=1.1128e-05) (hash(x)=26515570) +950 val loss 7.7071 +950 val perplexity 2224.1306 +950 train 7.901220 (lr=1.1085e-05) (hash(x)=26911957) +951 train 8.002204 (lr=1.1042e-05) (hash(x)=25856625) +952 train 7.608528 (lr=1.1000e-05) (hash(x)=25219129) +953 train 7.561040 (lr=1.0959e-05) (hash(x)=25260471) +954 train 7.933850 (lr=1.0919e-05) (hash(x)=29373370) +955 train 7.499208 (lr=1.0879e-05) (hash(x)=23437426) +956 train 7.794917 (lr=1.0841e-05) (hash(x)=23769521) +957 train 7.834813 (lr=1.0803e-05) (hash(x)=25961833) +958 train 7.557636 (lr=1.0766e-05) (hash(x)=23582666) +959 train 7.631147 (lr=1.0730e-05) (hash(x)=23164356) +960 train 7.622867 (lr=1.0695e-05) (hash(x)=24443114) +961 train 7.669073 (lr=1.0661e-05) (hash(x)=25052665) +962 train 7.819618 (lr=1.0628e-05) (hash(x)=27802272) +963 train 7.682610 (lr=1.0595e-05) (hash(x)=25957896) +964 train 7.739763 (lr=1.0563e-05) (hash(x)=26737251) +965 train 7.819353 (lr=1.0533e-05) (hash(x)=24723263) +966 train 7.767580 (lr=1.0503e-05) (hash(x)=24707011) +967 train 7.741700 (lr=1.0474e-05) (hash(x)=25646282) +968 train 7.773565 (lr=1.0445e-05) (hash(x)=27544665) +969 train 7.719388 (lr=1.0418e-05) (hash(x)=25851993) +970 train 7.970669 (lr=1.0391e-05) (hash(x)=29059700) +971 train 7.649812 (lr=1.0366e-05) (hash(x)=21513584) +972 train 7.502570 (lr=1.0341e-05) (hash(x)=23151267) +973 train 7.682794 (lr=1.0317e-05) (hash(x)=26017176) +974 train 7.797186 (lr=1.0294e-05) (hash(x)=26979518) +975 train 7.749977 (lr=1.0272e-05) (hash(x)=23843233) +976 train 7.899625 (lr=1.0251e-05) (hash(x)=24193010) +977 train 7.643147 (lr=1.0230e-05) (hash(x)=21476847) +978 train 7.528501 (lr=1.0211e-05) (hash(x)=21366504) +979 train 7.623815 (lr=1.0192e-05) (hash(x)=23226697) +980 train 7.412387 (lr=1.0174e-05) (hash(x)=19961773) +981 train 7.264277 (lr=1.0157e-05) (hash(x)=19772969) +982 train 7.553925 (lr=1.0141e-05) (hash(x)=23110142) +983 train 7.800233 (lr=1.0126e-05) (hash(x)=24506028) +984 train 7.785161 (lr=1.0111e-05) (hash(x)=25480731) +985 train 7.631804 (lr=1.0098e-05) (hash(x)=21077417) +986 train 7.624639 (lr=1.0085e-05) (hash(x)=23686713) +987 train 7.715808 (lr=1.0074e-05) (hash(x)=26024321) +988 train 7.783865 (lr=1.0063e-05) (hash(x)=27424109) +989 train 7.845912 (lr=1.0053e-05) (hash(x)=27786174) +990 train 7.690274 (lr=1.0044e-05) (hash(x)=25232502) +991 train 7.688741 (lr=1.0035e-05) (hash(x)=22781277) +992 train 7.728753 (lr=1.0028e-05) (hash(x)=26184527) +993 train 7.747872 (lr=1.0021e-05) (hash(x)=24459895) +994 train 7.693103 (lr=1.0016e-05) (hash(x)=25244624) +995 train 7.847892 (lr=1.0011e-05) (hash(x)=24451843) +996 train 7.803341 (lr=1.0007e-05) (hash(x)=22129897) +997 train 7.665625 (lr=1.0004e-05) (hash(x)=21116390) +998 train 7.514467 (lr=1.0002e-05) (hash(x)=20650070) +999 val loss 7.7025 +999 val perplexity 2213.9304 +999 train 7.819237 (lr=1.0000e-05) (hash(x)=24948650)