| type train | step 10 | loss 165.1385 270.2136 572.4084 821.6742 | lr 1.3e-05 | norm 328.8093 | dt 0.025 | |
| type train | step 20 | loss 163.0963 266.0186 567.0890 816.2864 | lr 2.7e-05 | norm 330.9393 | dt 0.024 | |
| type train | step 30 | loss 160.7960 263.0918 558.5065 807.5233 | lr 4.0e-05 | norm 332.8567 | dt 0.025 | |
| type train | step 40 | loss 157.8691 256.7778 547.8269 786.3160 | lr 5.3e-05 | norm 309.0007 | dt 0.025 | |
| type train | step 50 | loss 152.8079 253.0090 537.9453 777.2868 | lr 6.7e-05 | norm 317.0534 | dt 0.025 | |
| type train | step 60 | loss 147.5021 244.9229 524.6030 753.7883 | lr 8.0e-05 | norm 307.9474 | dt 0.025 | |
| type train | step 70 | loss 142.2691 236.6360 511.6404 732.1572 | lr 9.3e-05 | norm 301.6382 | dt 0.025 | |
| type train | step 80 | loss 136.8360 225.1527 490.6563 704.2423 | lr 1.1e-04 | norm 278.2949 | dt 0.025 | |
| type train | step 90 | loss 129.5293 216.7067 473.0341 681.3131 | lr 1.2e-04 | norm 278.0518 | dt 0.025 | |
| type train | step 100 | loss 121.4600 203.8161 450.6063 650.6235 | lr 1.3e-04 | norm 271.5758 | dt 0.025 | |
| type train | step 110 | loss 112.2996 192.6810 427.5995 620.1606 | lr 1.5e-04 | norm 255.0939 | dt 0.025 | |
| type train | step 120 | loss 105.8364 179.8049 406.2076 584.6907 | lr 1.6e-04 | norm 234.0229 | dt 0.025 | |
| type train | step 130 | loss 98.0283 167.8793 382.9848 549.0172 | lr 1.7e-04 | norm 223.6146 | dt 0.025 | |
| type train | step 140 | loss 90.6825 154.9549 358.7951 516.9789 | lr 1.9e-04 | norm 205.8964 | dt 0.025 | |
| type train | step 150 | loss 81.9171 143.6754 332.3669 486.8011 | lr 2.0e-04 | norm 196.6172 | dt 0.025 | |
| type train | step 160 | loss 73.3901 130.9413 303.4353 444.7008 | lr 2.1e-04 | norm 182.2378 | dt 0.025 | |
| type train | step 170 | loss 65.9645 119.5788 280.9860 411.7304 | lr 2.3e-04 | norm 173.8460 | dt 0.025 | |
| type train | step 180 | loss 60.0250 108.9502 260.2289 380.9074 | lr 2.4e-04 | norm 152.5309 | dt 0.025 | |
| type train | step 190 | loss 52.3086 96.5664 237.1376 346.7883 | lr 2.5e-04 | norm 140.1891 | dt 0.025 | |
| type train | step 200 | loss 46.6304 87.1684 215.4709 319.3999 | lr 2.7e-04 | norm 125.2332 | dt 0.025 | |
| type train | step 210 | loss 40.3619 76.9613 194.3418 290.2514 | lr 2.8e-04 | norm 114.1173 | dt 0.025 | |
| type train | step 220 | loss 34.6424 68.7590 174.5483 258.4792 | lr 2.9e-04 | norm 102.4734 | dt 0.025 | |
| type train | step 230 | loss 30.1486 61.1061 156.2560 235.4085 | lr 3.1e-04 | norm 92.6648 | dt 0.025 | |
| type train | step 240 | loss 26.7249 55.5610 142.3275 216.8329 | lr 3.2e-04 | norm 80.0612 | dt 0.025 | |
| type train | step 250 | loss 23.2184 48.2219 125.5102 190.7561 | lr 3.3e-04 | norm 69.1847 | dt 0.025 | |
| type train | step 260 | loss 19.9679 42.8265 111.7102 171.8960 | lr 3.5e-04 | norm 58.9870 | dt 0.025 | |
| type train | step 270 | loss 17.8285 38.5623 101.4058 153.1217 | lr 3.6e-04 | norm 50.4383 | dt 0.025 | |
| type train | step 280 | loss 15.8238 34.9245 91.8012 139.9742 | lr 3.7e-04 | norm 42.1871 | dt 0.025 | |
| type train | step 290 | loss 13.9385 31.4421 80.9965 124.3892 | lr 3.9e-04 | norm 34.6667 | dt 0.025 | |
| type train | step 300 | loss 12.6789 29.3418 75.1400 113.7109 | lr 4.0e-04 | norm 28.6889 | dt 0.025 | |
| type train | step 310 | loss 11.6092 26.9576 68.7072 104.2443 | lr 4.1e-04 | norm 23.0038 | dt 0.025 | |
| type train | step 320 | loss 10.3727 24.8519 63.1290 96.1963 | lr 4.3e-04 | norm 18.4055 | dt 0.025 | |
| type train | step 330 | loss 9.3595 23.0765 58.8250 88.3715 | lr 4.4e-04 | norm 14.5114 | dt 0.025 | |
| type train | step 340 | loss 8.3622 21.1823 54.2628 82.1875 | lr 4.5e-04 | norm 12.3091 | dt 0.025 | |
| type train | step 350 | loss 7.6069 19.7164 50.6525 77.6228 | lr 4.7e-04 | norm 10.7173 | dt 0.025 | |
| type train | step 360 | loss 6.7946 18.1085 47.2662 72.9439 | lr 4.8e-04 | norm 9.4891 | dt 0.025 | |
| type train | step 370 | loss 6.0886 16.8680 44.5916 69.4364 | lr 4.9e-04 | norm 8.6261 | dt 0.025 | |
| type train | step 380 | loss 5.7135 15.9004 42.8735 67.2396 | lr 5.1e-04 | norm 7.7483 | dt 0.025 | |
| type train | step 390 | loss 4.8557 14.4815 39.5186 63.6469 | lr 5.2e-04 | norm 6.9354 | dt 0.026 | |
| type train | step 400 | loss 4.3964 13.4400 36.9178 59.4483 | lr 5.3e-04 | norm 6.3993 | dt 0.025 | |
| type train | step 410 | loss 3.9887 12.6107 35.1733 57.6062 | lr 5.5e-04 | norm 5.7077 | dt 0.025 | |
| type train | step 420 | loss 3.6124 11.7977 32.9769 55.2922 | lr 5.6e-04 | norm 5.7990 | dt 0.026 | |
| type train | step 430 | loss 3.1673 10.9804 31.4535 53.2687 | lr 5.7e-04 | norm 4.9609 | dt 0.025 | |
| type train | step 440 | loss 2.8001 10.0485 29.5011 51.0227 | lr 5.9e-04 | norm 4.6763 | dt 0.025 | |
| type train | step 450 | loss 2.5459 9.5637 28.4607 49.7129 | lr 6.0e-04 | norm 4.1491 | dt 0.025 | |
| type train | step 460 | loss 2.2041 8.7742 26.1134 46.3414 | lr 6.1e-04 | norm 4.1282 | dt 0.026 | |
| type train | step 470 | loss 2.0346 8.5131 25.7058 46.3008 | lr 6.3e-04 | norm 3.7111 | dt 0.026 | |
| type train | step 480 | loss 1.7395 7.9282 24.2272 44.5632 | lr 6.4e-04 | norm 3.4851 | dt 0.026 | |
| type train | step 490 | loss 1.5209 7.3677 22.9940 42.9800 | lr 6.5e-04 | norm 3.3488 | dt 0.026 | |
| type train | step 500 | loss 1.3636 7.0701 22.0606 41.9193 | lr 6.7e-04 | norm 3.2384 | dt 0.026 | |
| type train | step 510 | loss 1.1889 6.6789 21.4254 41.5707 | lr 6.8e-04 | norm 3.0236 | dt 0.026 | |
| type train | step 520 | loss 1.0289 6.1938 20.3961 40.1946 | lr 6.9e-04 | norm 3.1671 | dt 0.026 | |
| type train | step 530 | loss 0.9935 6.0423 19.9224 39.7254 | lr 7.1e-04 | norm 2.8277 | dt 0.026 | |
| type train | step 540 | loss 0.7917 5.5704 18.1786 37.2943 | lr 7.2e-04 | norm 2.6145 | dt 0.026 | |
| type train | step 550 | loss 0.7107 5.3478 17.8380 37.0811 | lr 7.3e-04 | norm 2.7307 | dt 0.026 | |
| type train | step 560 | loss 0.7130 5.1705 17.3733 36.4978 | lr 7.5e-04 | norm 2.9970 | dt 0.025 | |
| type train | step 570 | loss 0.6016 4.9042 17.1031 36.8298 | lr 7.6e-04 | norm 2.5208 | dt 0.026 | |
| type train | step 580 | loss 0.5516 4.5979 15.8052 33.9836 | lr 7.7e-04 | norm 2.9171 | dt 0.026 | |
| type train | step 590 | loss 0.5125 4.4266 15.7397 34.9871 | lr 7.9e-04 | norm 2.5681 | dt 0.026 | |
| type train | step 600 | loss 0.4316 4.1440 14.4388 32.9529 | lr 8.0e-04 | norm 2.2182 | dt 0.026 | |
| type train | step 610 | loss 0.4109 3.9861 14.4659 33.2300 | lr 8.1e-04 | norm 2.4214 | dt 0.027 | |
| type train | step 620 | loss 0.3815 3.8405 13.7613 32.5364 | lr 8.3e-04 | norm 2.0169 | dt 0.026 | |
| type train | step 630 | loss 0.3535 3.6983 13.7901 32.7935 | lr 8.4e-04 | norm 2.2334 | dt 0.027 | |
| type train | step 640 | loss 0.3384 3.4545 12.8687 30.9562 | lr 8.5e-04 | norm 2.1879 | dt 0.026 | |
| type train | step 650 | loss 0.3538 3.5087 13.3596 31.9760 | lr 8.7e-04 | norm 2.5553 | dt 0.026 | |
| type train | step 660 | loss 0.3032 3.2737 11.9051 29.7264 | lr 8.8e-04 | norm 2.1741 | dt 0.026 | |
| type train | step 670 | loss 0.2833 3.0281 11.6011 29.7092 | lr 8.9e-04 | norm 2.0794 | dt 0.026 | |
| type train | step 680 | loss 0.2701 2.9810 11.4060 29.1997 | lr 9.1e-04 | norm 1.7558 | dt 0.026 | |
| type train | step 690 | loss 0.2736 2.9943 12.1487 31.0265 | lr 9.2e-04 | norm 2.1771 | dt 0.026 | |
| type train | step 700 | loss 0.2595 2.7941 11.0060 28.6038 | lr 9.3e-04 | norm 2.0659 | dt 0.026 | |
| type train | step 710 | loss 0.2396 2.6249 10.4282 28.0704 | lr 9.5e-04 | norm 1.9216 | dt 0.027 | |
| type train | step 720 | loss 0.2474 2.6553 10.4821 27.8841 | lr 9.6e-04 | norm 1.9816 | dt 0.026 | |
| type train | step 730 | loss 0.2412 2.5772 10.3022 27.5018 | lr 9.7e-04 | norm 2.1148 | dt 0.026 | |
| type train | step 740 | loss 0.2219 2.4358 9.9118 27.6275 | lr 9.9e-04 | norm 1.8178 | dt 0.027 | |
| type train | step 750 | loss 0.2576 2.6055 10.5491 28.1582 | lr 1.0e-03 | norm 2.9292 | dt 0.026 | |
| type train | step 760 | loss 0.2197 2.3766 9.7639 27.3210 | lr 1.0e-03 | norm 2.2145 | dt 0.026 | |
| type train | step 770 | loss 0.2072 2.2662 9.5252 27.2014 | lr 1.0e-03 | norm 2.0254 | dt 0.026 | |
| type train | step 780 | loss 0.2031 2.2571 9.1705 26.3610 | lr 1.0e-03 | norm 2.0233 | dt 0.026 | |
| type train | step 790 | loss 0.1991 2.2210 9.2737 26.5751 | lr 1.0e-03 | norm 2.0667 | dt 0.026 | |
| type train | step 800 | loss 0.1930 2.0839 8.9798 26.4337 | lr 1.0e-03 | norm 2.2032 | dt 0.027 | |
| type train | step 810 | loss 0.2022 2.1660 9.2296 27.1243 | lr 1.0e-03 | norm 1.9982 | dt 0.026 | |
| type train | step 820 | loss 0.1921 2.0341 8.6821 26.0543 | lr 1.0e-03 | norm 1.9730 | dt 0.026 | |
| type train | step 830 | loss 0.1825 1.9648 8.5645 25.6894 | lr 1.0e-03 | norm 1.9061 | dt 0.026 | |
| type train | step 840 | loss 0.1733 1.9379 8.2640 25.4123 | lr 1.0e-03 | norm 1.8042 | dt 0.026 | |
| type train | step 850 | loss 0.1816 1.9837 8.5251 25.8336 | lr 1.0e-03 | norm 2.0445 | dt 0.026 | |
| type train | step 860 | loss 0.1730 1.8393 7.9306 24.9149 | lr 1.0e-03 | norm 1.7425 | dt 0.026 | |
| type train | step 870 | loss 0.1687 1.8073 7.9461 24.9986 | lr 1.0e-03 | norm 1.7133 | dt 0.026 | |
| type train | step 880 | loss 0.1732 1.7904 7.8238 24.7140 | lr 1.0e-03 | norm 1.6960 | dt 0.027 | |
| type train | step 890 | loss 0.1737 1.8036 8.0769 25.2210 | lr 1.0e-03 | norm 2.1105 | dt 0.027 | |
| type train | step 900 | loss 0.1639 1.7288 7.5433 24.3281 | lr 1.0e-03 | norm 1.9975 | dt 0.026 | |
| type train | step 910 | loss 0.1658 1.7870 7.7996 24.6590 | lr 1.0e-03 | norm 2.0305 | dt 0.027 | |
| type train | step 920 | loss 0.1637 1.6970 7.5930 24.8338 | lr 1.0e-03 | norm 2.0157 | dt 0.026 | |
| type train | step 930 | loss 0.1597 1.6539 7.4916 24.5246 | lr 1.0e-03 | norm 1.8745 | dt 0.026 | |
| type train | step 940 | loss 0.1593 1.6501 7.5990 24.7488 | lr 1.0e-03 | norm 1.5527 | dt 0.026 | |
| type train | step 950 | loss 0.1534 1.6182 7.4112 24.4319 | lr 1.0e-03 | norm 1.6904 | dt 0.027 | |
| type train | step 960 | loss 0.1538 1.6252 7.2769 24.0812 | lr 1.0e-03 | norm 2.0223 | dt 0.026 | |
| type train | step 970 | loss 0.1514 1.5446 7.0412 23.2963 | lr 1.0e-03 | norm 1.9348 | dt 0.026 | |
| type train | step 980 | loss 0.1499 1.5749 7.0243 24.0068 | lr 1.0e-03 | norm 1.9766 | dt 0.027 | |
| type train | step 990 | loss 0.1547 1.5937 7.2931 24.4444 | lr 1.0e-03 | norm 1.9051 | dt 0.026 | |
| type train | step 1000 | loss 0.1457 1.5152 7.1559 24.1883 | lr 1.0e-03 | norm 1.6060 | dt 0.026 | |
| type train | step 1010 | loss 0.1519 1.4931 6.8835 23.3158 | lr 1.0e-03 | norm 1.6552 | dt 0.026 | |
| type train | step 1020 | loss 0.1466 1.5125 6.9455 23.4447 | lr 1.0e-03 | norm 1.6694 | dt 0.026 | |
| type train | step 1030 | loss 0.1463 1.4932 6.8728 23.2036 | lr 1.0e-03 | norm 2.1455 | dt 0.029 | |
| type train | step 1040 | loss 0.1446 1.4881 6.8930 23.8392 | lr 1.0e-03 | norm 1.6853 | dt 0.029 | |
| type train | step 1050 | loss 0.1411 1.4387 6.6535 23.1715 | lr 1.0e-03 | norm 1.8199 | dt 0.028 | |
| type train | step 1060 | loss 0.1453 1.4588 6.8294 23.4503 | lr 1.0e-03 | norm 1.5351 | dt 0.027 | |
| type train | step 1070 | loss 0.1400 1.3788 6.4395 22.3954 | lr 1.0e-03 | norm 1.7496 | dt 0.027 | |
| type train | step 1080 | loss 0.1421 1.4325 6.6556 23.2528 | lr 1.0e-03 | norm 1.6429 | dt 0.035 | |
| type train | step 1090 | loss 0.1397 1.4080 6.7020 23.2236 | lr 1.0e-03 | norm 1.6664 | dt 0.026 | |
| type train | step 1100 | loss 0.1403 1.4044 6.6847 23.4283 | lr 1.0e-03 | norm 1.7520 | dt 0.026 | |
| type train | step 1110 | loss 0.1398 1.4011 6.5721 23.2076 | lr 1.0e-03 | norm 1.8002 | dt 0.026 | |
| type train | step 1120 | loss 0.1387 1.3818 6.7241 23.6714 | lr 1.0e-03 | norm 1.6650 | dt 0.026 | |
| type train | step 1130 | loss 0.1358 1.3517 6.6162 23.5787 | lr 1.0e-03 | norm 1.8641 | dt 0.026 | |
| type train | step 1140 | loss 0.1369 1.3865 6.5940 23.6261 | lr 1.0e-03 | norm 1.6908 | dt 0.026 | |
| type train | step 1150 | loss 0.1304 1.3261 6.3151 22.2271 | lr 1.0e-03 | norm 1.6046 | dt 0.026 | |
| type train | step 1160 | loss 0.1313 1.3489 6.3709 22.9001 | lr 1.0e-03 | norm 1.8888 | dt 0.026 | |
| type train | step 1170 | loss 0.1347 1.3547 6.3901 22.6852 | lr 1.0e-03 | norm 2.1202 | dt 0.026 | |
| type train | step 1180 | loss 0.1315 1.3279 6.6526 23.6912 | lr 1.0e-03 | norm 1.7143 | dt 0.026 | |
| type train | step 1190 | loss 0.1348 1.2996 6.2509 22.1545 | lr 1.0e-03 | norm 2.0113 | dt 0.026 | |
| type train | step 1200 | loss 0.1292 1.3264 6.3911 22.9994 | lr 1.0e-03 | norm 1.8157 | dt 0.026 | |
| type train | step 1210 | loss 0.1269 1.2688 6.1580 21.9254 | lr 1.0e-03 | norm 1.5827 | dt 0.026 | |
| type train | step 1220 | loss 0.1272 1.2810 6.4127 22.8977 | lr 1.0e-03 | norm 1.8684 | dt 0.028 | |
| type train | step 1230 | loss 0.1264 1.2709 6.1283 22.4981 | lr 1.0e-03 | norm 1.5098 | dt 0.026 | |
| type train | step 1240 | loss 0.1262 1.2812 6.3206 23.0331 | lr 1.0e-03 | norm 1.6844 | dt 0.026 | |
| type train | step 1250 | loss 0.1264 1.2207 6.0299 22.0720 | lr 1.0e-03 | norm 1.7052 | dt 0.027 | |
| type train | step 1260 | loss 0.1298 1.3155 6.3290 22.8424 | lr 1.0e-03 | norm 1.9428 | dt 0.026 | |
| type train | step 1270 | loss 0.1244 1.2293 5.9954 21.4596 | lr 1.0e-03 | norm 1.6419 | dt 0.027 | |
| type train | step 1280 | loss 0.1241 1.1964 6.0151 22.1390 | lr 1.0e-03 | norm 1.6917 | dt 0.026 | |
| type train | step 1290 | loss 0.1230 1.2128 5.8916 21.7072 | lr 1.0e-03 | norm 1.3979 | dt 0.026 | |
| type train | step 1300 | loss 0.1265 1.2680 6.4163 23.5105 | lr 1.0e-03 | norm 1.7171 | dt 0.026 | |
| type train | step 1310 | loss 0.1232 1.2028 5.9695 21.9812 | lr 1.0e-03 | norm 1.7213 | dt 0.026 | |
| type train | step 1320 | loss 0.1198 1.1595 5.7707 21.5102 | lr 1.0e-03 | norm 1.5406 | dt 0.026 | |
| type train | step 1330 | loss 0.1233 1.2016 6.0319 21.8333 | lr 1.0e-03 | norm 1.6312 | dt 0.026 | |
| type train | step 1340 | loss 0.1247 1.2040 5.8695 21.4616 | lr 1.0e-03 | norm 1.7372 | dt 0.026 | |
| type train | step 1350 | loss 0.1203 1.1670 5.7546 21.6908 | lr 1.0e-03 | norm 1.5204 | dt 0.026 | |
| type train | step 1360 | loss 0.1317 1.2473 6.0627 22.3057 | lr 1.0e-03 | norm 2.3767 | dt 0.026 | |
| type train | step 1370 | loss 0.1237 1.2033 5.9863 22.1185 | lr 1.0e-03 | norm 1.8920 | dt 0.026 | |
| type train | step 1380 | loss 0.1194 1.1632 5.8937 22.0040 | lr 1.0e-03 | norm 1.7189 | dt 0.027 | |
| type train | step 1390 | loss 0.1179 1.1611 5.8235 21.4412 | lr 1.0e-03 | norm 1.7479 | dt 0.027 | |
| type train | step 1400 | loss 0.1187 1.1669 5.8823 21.6764 | lr 1.0e-03 | norm 1.7711 | dt 0.027 | |
| type train | step 1410 | loss 0.1181 1.1341 5.7004 21.6349 | lr 1.0e-03 | norm 1.9526 | dt 0.026 | |
| type train | step 1420 | loss 0.1248 1.2066 6.0239 22.5222 | lr 1.0e-03 | norm 1.6986 | dt 0.026 | |
| type train | step 1430 | loss 0.1201 1.1321 5.7736 21.6482 | lr 1.0e-03 | norm 1.7226 | dt 0.026 | |
| type train | step 1440 | loss 0.1158 1.1270 5.7753 21.5888 | lr 1.0e-03 | norm 1.6448 | dt 0.026 | |
| type train | step 1450 | loss 0.1120 1.1127 5.7307 21.4037 | lr 1.0e-03 | norm 1.6072 | dt 0.027 | |
| type train | step 1460 | loss 0.1198 1.1707 5.8725 21.8626 | lr 1.0e-03 | norm 1.7951 | dt 0.026 | |
| type train | step 1470 | loss 0.1147 1.1022 5.5479 21.1201 | lr 1.0e-03 | norm 1.5490 | dt 0.028 | |
| type train | step 1480 | loss 0.1128 1.0966 5.6277 21.2932 | lr 1.0e-03 | norm 1.5236 | dt 0.027 | |
| type train | step 1490 | loss 0.1169 1.0982 5.5891 21.2182 | lr 1.0e-03 | norm 1.5146 | dt 0.026 | |
| type train | step 1500 | loss 0.1176 1.1251 5.8661 21.6501 | lr 1.0e-03 | norm 1.9210 | dt 0.027 | |
| type train | step 1510 | loss 0.1131 1.0849 5.5387 20.9497 | lr 1.0e-03 | norm 1.7862 | dt 0.027 | |
| type train | step 1520 | loss 0.1152 1.1178 5.6841 21.2342 | lr 1.0e-03 | norm 1.7847 | dt 0.026 | |
| type train | step 1530 | loss 0.1143 1.0992 5.6419 21.5566 | lr 1.0e-03 | norm 1.8392 | dt 0.027 | |
| type train | step 1540 | loss 0.1127 1.0728 5.6048 21.4201 | lr 1.0e-03 | norm 1.6977 | dt 0.027 | |
| type train | step 1550 | loss 0.1127 1.0775 5.7579 21.7380 | lr 1.0e-03 | norm 1.4410 | dt 0.028 | |
| type train | step 1560 | loss 0.1091 1.0615 5.6606 21.5432 | lr 1.0e-03 | norm 1.5395 | dt 0.027 | |
| type train | step 1570 | loss 0.1105 1.0756 5.6108 21.3327 | lr 1.0e-03 | norm 1.8617 | dt 0.027 | |
| type train | step 1580 | loss 0.1097 1.0279 5.3970 20.6618 | lr 1.0e-03 | norm 1.7776 | dt 0.026 | |
| type train | step 1590 | loss 0.1095 1.0643 5.4080 21.1733 | lr 1.0e-03 | norm 1.8055 | dt 0.026 | |
| type train | step 1600 | loss 0.1135 1.0843 5.6225 21.5358 | lr 1.0e-03 | norm 1.7471 | dt 0.027 | |
| type train | step 1610 | loss 0.1071 1.0403 5.6548 21.6265 | lr 1.0e-03 | norm 1.4551 | dt 0.027 | |
| type train | step 1620 | loss 0.1107 1.0375 5.4591 20.8138 | lr 1.0e-03 | norm 1.5478 | dt 0.026 | |
| type train | step 1630 | loss 0.1080 1.0536 5.5290 21.0002 | lr 1.0e-03 | norm 1.6034 | dt 0.027 | |
| type train | step 1640 | loss 0.1082 1.0624 5.4730 20.8332 | lr 1.0e-03 | norm 1.9253 | dt 0.027 | |
| type train | step 1650 | loss 0.1086 1.0510 5.5478 21.4375 | lr 1.0e-03 | norm 1.5738 | dt 0.026 | |
| type train | step 1660 | loss 0.1056 1.0212 5.3582 20.7518 | lr 1.0e-03 | norm 1.6928 | dt 0.027 | |
| type train | step 1670 | loss 0.1097 1.0426 5.5614 21.2529 | lr 9.9e-04 | norm 1.4242 | dt 0.026 | |
| type train | step 1680 | loss 0.1048 0.9967 5.2454 20.3233 | lr 9.9e-04 | norm 1.5946 | dt 0.026 | |
| type train | step 1690 | loss 0.1070 1.0298 5.4314 21.0025 | lr 9.9e-04 | norm 1.5415 | dt 0.026 | |
| type train | step 1700 | loss 0.1064 1.0296 5.5174 21.1010 | lr 9.9e-04 | norm 1.5728 | dt 0.026 | |
| type train | step 1710 | loss 0.1071 1.0335 5.5309 21.3624 | lr 9.9e-04 | norm 1.6325 | dt 0.027 | |
| type train | step 1720 | loss 0.1070 1.0269 5.4303 21.1157 | lr 9.9e-04 | norm 1.6732 | dt 0.027 | |
| type train | step 1730 | loss 0.1066 1.0294 5.6116 21.6718 | lr 9.9e-04 | norm 1.5229 | dt 0.027 | |
| type train | step 1740 | loss 0.1053 1.0061 5.5351 21.6837 | lr 9.9e-04 | norm 1.7467 | dt 0.027 | |
| type train | step 1750 | loss 0.1057 1.0324 5.5299 21.5855 | lr 9.9e-04 | norm 1.5528 | dt 0.027 | |
| type train | step 1760 | loss 0.1004 0.9873 5.2899 20.3241 | lr 9.9e-04 | norm 1.5197 | dt 0.027 | |
| type train | step 1770 | loss 0.1015 1.0095 5.3679 20.9531 | lr 9.9e-04 | norm 1.7566 | dt 0.027 | |
| type train | step 1780 | loss 0.1057 1.0262 5.3808 20.8243 | lr 9.9e-04 | norm 1.9688 | dt 0.028 | |
| type train | step 1790 | loss 0.1036 1.0087 5.6624 21.9378 | lr 9.9e-04 | norm 1.5807 | dt 0.027 | |
| type train | step 1800 | loss 0.1045 0.9882 5.2971 20.4102 | lr 9.9e-04 | norm 1.8153 | dt 0.027 | |
| type train | step 1810 | loss 0.1014 1.0088 5.4454 21.2443 | lr 9.9e-04 | norm 1.6855 | dt 0.026 | |
| type train | step 1820 | loss 0.0987 0.9692 5.2601 20.2369 | lr 9.9e-04 | norm 1.4601 | dt 0.027 | |
| type train | step 1830 | loss 0.0999 1.0019 5.5362 21.3493 | lr 9.9e-04 | norm 1.7669 | dt 0.028 | |
| type train | step 1840 | loss 0.1000 0.9875 5.3061 20.8368 | lr 9.9e-04 | norm 1.4527 | dt 0.026 | |
| type train | step 1850 | loss 0.1008 0.9924 5.4496 21.4343 | lr 9.9e-04 | norm 1.5866 | dt 0.026 | |
| type train | step 1860 | loss 0.1002 0.9476 5.2154 20.4854 | lr 9.9e-04 | norm 1.5748 | dt 0.027 | |
| type train | step 1870 | loss 0.1031 1.0367 5.4791 21.2105 | lr 9.9e-04 | norm 1.8220 | dt 0.027 | |
| type train | step 1880 | loss 0.0986 0.9652 5.2119 20.1225 | lr 9.9e-04 | norm 1.5538 | dt 0.026 | |
| type train | step 1890 | loss 0.0983 0.9477 5.2679 20.7477 | lr 9.9e-04 | norm 1.6048 | dt 0.027 | |
| type train | step 1900 | loss 0.0986 0.9530 5.1531 20.2456 | lr 9.9e-04 | norm 1.3310 | dt 0.026 | |
| type train | step 1910 | loss 0.1020 1.0092 5.6238 22.0082 | lr 9.9e-04 | norm 1.6405 | dt 0.033 | |
| type train | step 1920 | loss 0.0975 0.9537 5.2523 20.5363 | lr 9.9e-04 | norm 1.6020 | dt 0.027 | |
| type train | step 1930 | loss 0.0965 0.9229 5.0576 20.1481 | lr 9.9e-04 | norm 1.4377 | dt 0.027 | |
| type train | step 1940 | loss 0.0994 0.9621 5.3027 20.5026 | lr 9.9e-04 | norm 1.5186 | dt 0.026 | |
| type train | step 1950 | loss 0.1009 0.9673 5.1745 20.1139 | lr 9.9e-04 | norm 1.6299 | dt 0.027 | |
| type train | step 1960 | loss 0.0977 0.9380 5.1091 20.3134 | lr 9.9e-04 | norm 1.4258 | dt 0.027 | |
| type train | step 1970 | loss 0.1059 1.0012 5.3140 20.8917 | lr 9.9e-04 | norm 2.2642 | dt 0.026 | |
| type train | step 1980 | loss 0.0996 0.9736 5.3158 20.8329 | lr 9.9e-04 | norm 1.7452 | dt 0.027 | |
| type train | step 1990 | loss 0.0974 0.9352 5.2650 20.7196 | lr 9.9e-04 | norm 1.6040 | dt 0.027 | |
| type train | step 2000 | loss 0.0957 0.9462 5.1973 20.2038 | lr 9.9e-04 | norm 1.6235 | dt 0.027 | |
| type train | step 2010 | loss 0.0976 0.9528 5.2220 20.4228 | lr 9.9e-04 | norm 1.6267 | dt 0.026 | |
| type train | step 2020 | loss 0.0963 0.9220 5.1052 20.3498 | lr 9.9e-04 | norm 1.8404 | dt 0.027 | |
| type train | step 2030 | loss 0.1024 0.9894 5.3893 21.2001 | lr 9.9e-04 | norm 1.6476 | dt 0.026 | |
| type train | step 2040 | loss 0.0985 0.9320 5.1505 20.3558 | lr 9.9e-04 | norm 1.6094 | dt 0.026 | |
| type train | step 2050 | loss 0.0954 0.9261 5.2290 20.4708 | lr 9.9e-04 | norm 1.5417 | dt 0.028 | |
| type train | step 2060 | loss 0.0926 0.9249 5.1712 20.2750 | lr 9.9e-04 | norm 1.4770 | dt 0.026 | |
| type train | step 2070 | loss 0.0988 0.9707 5.2701 20.6631 | lr 9.9e-04 | norm 1.6615 | dt 0.026 | |
| type train | step 2080 | loss 0.0946 0.9168 5.0163 19.9201 | lr 9.9e-04 | norm 1.4287 | dt 0.027 | |
| type train | step 2090 | loss 0.0936 0.9177 5.0872 20.2043 | lr 9.9e-04 | norm 1.4651 | dt 0.027 | |
| type train | step 2100 | loss 0.0970 0.9219 5.0416 20.0309 | lr 9.9e-04 | norm 1.4218 | dt 0.026 | |
| type train | step 2110 | loss 0.0973 0.9453 5.3165 20.5381 | lr 9.9e-04 | norm 1.7966 | dt 0.026 | |
| type train | step 2120 | loss 0.0944 0.9176 5.0336 19.8887 | lr 9.9e-04 | norm 1.6872 | dt 0.027 | |
| type train | step 2130 | loss 0.0961 0.9453 5.1612 20.1580 | lr 9.9e-04 | norm 1.6858 | dt 0.026 | |
| type train | step 2140 | loss 0.0949 0.9318 5.1348 20.3688 | lr 9.9e-04 | norm 1.7102 | dt 0.027 | |
| type train | step 2150 | loss 0.0941 0.9080 5.0942 20.3775 | lr 9.9e-04 | norm 1.6295 | dt 0.027 | |
| type train | step 2160 | loss 0.0945 0.9239 5.2674 20.6863 | lr 9.9e-04 | norm 1.3852 | dt 0.027 | |
| type train | step 2170 | loss 0.0916 0.9117 5.1875 20.4736 | lr 9.9e-04 | norm 1.4653 | dt 0.027 | |
| type train | step 2180 | loss 0.0929 0.9260 5.1450 20.3727 | lr 9.9e-04 | norm 1.7889 | dt 0.026 | |
| type train | step 2190 | loss 0.0927 0.8824 4.9251 19.7178 | lr 9.9e-04 | norm 1.7193 | dt 0.027 | |
| type train | step 2200 | loss 0.0923 0.9124 4.9644 20.1150 | lr 9.9e-04 | norm 1.7031 | dt 0.027 | |
| type train | step 2210 | loss 0.0960 0.9346 5.1596 20.4639 | lr 9.9e-04 | norm 1.6587 | dt 0.026 | |
| type train | step 2220 | loss 0.0909 0.9057 5.2000 20.5912 | lr 9.9e-04 | norm 1.3944 | dt 0.026 | |
| type train | step 2230 | loss 0.0941 0.9091 5.0403 19.8358 | lr 9.9e-04 | norm 1.5084 | dt 0.027 | |
| type train | step 2240 | loss 0.0914 0.9239 5.1075 20.0611 | lr 9.9e-04 | norm 1.5437 | dt 0.026 | |
| type train | step 2250 | loss 0.0918 0.9242 5.0220 19.9530 | lr 9.9e-04 | norm 1.8187 | dt 0.026 | |
| type train | step 2260 | loss 0.0925 0.9206 5.1409 20.4014 | lr 9.9e-04 | norm 1.4972 | dt 0.026 | |
| type train | step 2270 | loss 0.0899 0.8948 4.9454 19.8079 | lr 9.9e-04 | norm 1.6050 | dt 0.026 | |
| type train | step 2280 | loss 0.0942 0.9237 5.1633 20.2833 | lr 9.9e-04 | norm 1.3847 | dt 0.026 | |
| type train | step 2290 | loss 0.0897 0.8789 4.8468 19.4097 | lr 9.9e-04 | norm 1.5297 | dt 0.026 | |
| type train | step 2300 | loss 0.0912 0.9074 5.0405 20.0400 | lr 9.9e-04 | norm 1.4912 | dt 0.026 | |
| type train | step 2310 | loss 0.0906 0.9097 5.1160 20.2753 | lr 9.9e-04 | norm 1.5138 | dt 0.026 | |
| type train | step 2320 | loss 0.0916 0.9190 5.1489 20.4062 | lr 9.9e-04 | norm 1.5858 | dt 0.026 | |
| type train | step 2330 | loss 0.0917 0.9079 5.0522 20.1934 | lr 9.9e-04 | norm 1.6189 | dt 0.026 | |
| type train | step 2340 | loss 0.0913 0.9224 5.2218 20.7745 | lr 9.8e-04 | norm 1.4747 | dt 0.026 | |
| type train | step 2350 | loss 0.0900 0.8971 5.1472 20.7043 | lr 9.8e-04 | norm 1.6445 | dt 0.026 | |
| type train | step 2360 | loss 0.0911 0.9172 5.1475 20.6582 | lr 9.8e-04 | norm 1.4921 | dt 0.026 | |
| type train | step 2370 | loss 0.0858 0.8824 4.9320 19.5445 | lr 9.8e-04 | norm 1.4648 | dt 0.026 | |
| type train | step 2380 | loss 0.0870 0.9046 5.0317 20.1019 | lr 9.8e-04 | norm 1.6719 | dt 0.026 | |
| type train | step 2390 | loss 0.0910 0.9209 5.0165 20.0116 | lr 9.8e-04 | norm 1.8855 | dt 0.026 | |
| type train | step 2400 | loss 0.0895 0.9121 5.2704 21.0164 | lr 9.8e-04 | norm 1.5217 | dt 0.026 | |
| type train | step 2410 | loss 0.0903 0.8908 4.9593 19.5807 | lr 9.8e-04 | norm 1.7036 | dt 0.026 | |
| type train | step 2420 | loss 0.0876 0.9104 5.1059 20.3880 | lr 9.8e-04 | norm 1.6318 | dt 0.026 | |
| type train | step 2430 | loss 0.0852 0.8780 4.9478 19.4855 | lr 9.8e-04 | norm 1.4331 | dt 0.026 | |
| type train | step 2440 | loss 0.0869 0.9099 5.2337 20.5040 | lr 9.8e-04 | norm 1.7151 | dt 0.029 | |
| type train | step 2450 | loss 0.0869 0.8952 4.9955 20.0050 | lr 9.8e-04 | norm 1.3810 | dt 0.026 | |
| type train | step 2460 | loss 0.0873 0.9063 5.1178 20.5693 | lr 9.8e-04 | norm 1.5095 | dt 0.026 | |
| type train | step 2470 | loss 0.0878 0.8639 4.9117 19.6125 | lr 9.8e-04 | norm 1.5219 | dt 0.026 | |
| type train | step 2480 | loss 0.0897 0.9347 5.1598 20.4611 | lr 9.8e-04 | norm 1.7396 | dt 0.026 | |
| type train | step 2490 | loss 0.0854 0.8832 4.9318 19.4357 | lr 9.8e-04 | norm 1.4755 | dt 0.026 | |
| type train | step 2500 | loss 0.0854 0.8727 4.9823 19.9930 | lr 9.8e-04 | norm 1.5577 | dt 0.026 | |
| type train | step 2510 | loss 0.0864 0.8763 4.8703 19.4832 | lr 9.8e-04 | norm 1.2773 | dt 0.026 | |
| type train | step 2520 | loss 0.0893 0.9319 5.3165 21.1550 | lr 9.8e-04 | norm 1.5769 | dt 0.026 | |
| type train | step 2530 | loss 0.0851 0.8724 4.9475 19.7500 | lr 9.8e-04 | norm 1.5299 | dt 0.026 | |
| type train | step 2540 | loss 0.0841 0.8510 4.7954 19.3913 | lr 9.8e-04 | norm 1.3783 | dt 0.026 | |
| type train | step 2550 | loss 0.0868 0.8854 5.0246 19.8161 | lr 9.8e-04 | norm 1.4555 | dt 0.026 | |
| type train | step 2560 | loss 0.0880 0.8937 4.8747 19.3329 | lr 9.8e-04 | norm 1.5411 | dt 0.026 | |
| type train | step 2570 | loss 0.0858 0.8705 4.8298 19.5821 | lr 9.8e-04 | norm 1.3689 | dt 0.027 | |
| type train | step 2580 | loss 0.0925 0.9200 5.0197 20.1412 | lr 9.8e-04 | norm 2.1563 | dt 0.026 | |
| type train | step 2590 | loss 0.0873 0.9039 5.0546 20.0681 | lr 9.8e-04 | norm 1.6886 | dt 0.026 | |
| type train | step 2600 | loss 0.0852 0.8686 5.0042 19.9740 | lr 9.8e-04 | norm 1.5454 | dt 0.026 | |
| type train | step 2610 | loss 0.0838 0.8789 4.9445 19.5303 | lr 9.8e-04 | norm 1.5679 | dt 0.026 | |
| type train | step 2620 | loss 0.0856 0.8859 4.9472 19.7144 | lr 9.8e-04 | norm 1.5724 | dt 0.026 | |
| type train | step 2630 | loss 0.0842 0.8617 4.8482 19.6483 | lr 9.8e-04 | norm 1.7705 | dt 0.026 | |
| type train | step 2640 | loss 0.0896 0.9209 5.1294 20.4280 | lr 9.8e-04 | norm 1.5611 | dt 0.026 | |
| type train | step 2650 | loss 0.0863 0.8673 4.9015 19.6567 | lr 9.8e-04 | norm 1.5452 | dt 0.026 | |
| type train | step 2660 | loss 0.0836 0.8656 4.9861 19.7403 | lr 9.8e-04 | norm 1.4715 | dt 0.026 | |
| type train | step 2670 | loss 0.0814 0.8671 4.9352 19.6332 | lr 9.8e-04 | norm 1.4104 | dt 0.027 | |
| type train | step 2680 | loss 0.0866 0.9061 5.0215 20.0253 | lr 9.8e-04 | norm 1.5986 | dt 0.026 | |
| type train | step 2690 | loss 0.0830 0.8571 4.7875 19.2144 | lr 9.8e-04 | norm 1.3854 | dt 0.026 | |
| type train | step 2700 | loss 0.0821 0.8599 4.8771 19.5560 | lr 9.8e-04 | norm 1.4377 | dt 0.026 | |
| type train | step 2710 | loss 0.0858 0.8638 4.8062 19.3414 | lr 9.8e-04 | norm 1.3656 | dt 0.028 | |
| type train | step 2720 | loss 0.0855 0.8845 5.0554 19.8741 | lr 9.8e-04 | norm 1.7429 | dt 0.026 | |
| type train | step 2730 | loss 0.0829 0.8654 4.8239 19.2883 | lr 9.8e-04 | norm 1.6156 | dt 0.026 | |
| type train | step 2740 | loss 0.0849 0.8881 4.9426 19.5471 | lr 9.8e-04 | norm 1.6239 | dt 0.027 | |
| type train | step 2750 | loss 0.0839 0.8721 4.9049 19.7321 | lr 9.8e-04 | norm 1.6375 | dt 0.027 | |
| type train | step 2760 | loss 0.0831 0.8548 4.8752 19.7070 | lr 9.8e-04 | norm 1.5788 | dt 0.026 | |
| type train | step 2770 | loss 0.0834 0.8710 5.0364 20.0589 | lr 9.8e-04 | norm 1.3224 | dt 0.026 | |
| type train | step 2780 | loss 0.0809 0.8566 4.9745 19.8272 | lr 9.8e-04 | norm 1.3919 | dt 0.026 | |
| type train | step 2790 | loss 0.0817 0.8766 4.9547 19.7718 | lr 9.8e-04 | norm 1.7208 | dt 0.026 | |
| type train | step 2800 | loss 0.0818 0.8333 4.7244 19.1403 | lr 9.8e-04 | norm 1.6535 | dt 0.026 | |
| type train | step 2810 | loss 0.0814 0.8612 4.7768 19.4537 | lr 9.7e-04 | norm 1.6346 | dt 0.026 | |
| type train | step 2820 | loss 0.0845 0.8811 4.9490 19.8867 | lr 9.7e-04 | norm 1.5957 | dt 0.026 | |
| type train | step 2830 | loss 0.0802 0.8574 4.9877 19.9720 | lr 9.7e-04 | norm 1.3449 | dt 0.026 | |
| type train | step 2840 | loss 0.0830 0.8564 4.8516 19.1975 | lr 9.7e-04 | norm 1.4545 | dt 0.027 | |
| type train | step 2850 | loss 0.0810 0.8739 4.9187 19.5319 | lr 9.7e-04 | norm 1.4968 | dt 0.026 | |
| type train | step 2860 | loss 0.0811 0.8741 4.8291 19.3868 | lr 9.7e-04 | norm 1.7688 | dt 0.026 | |
| type train | step 2870 | loss 0.0817 0.8692 4.9478 19.7758 | lr 9.7e-04 | norm 1.4576 | dt 0.026 | |
| type train | step 2880 | loss 0.0800 0.8442 4.7603 19.2442 | lr 9.7e-04 | norm 1.5669 | dt 0.026 | |
| type train | step 2890 | loss 0.0830 0.8766 4.9689 19.6805 | lr 9.7e-04 | norm 1.3428 | dt 0.026 | |
| type train | step 2900 | loss 0.0796 0.8338 4.6796 18.8139 | lr 9.7e-04 | norm 1.4776 | dt 0.026 | |
| type train | step 2910 | loss 0.0811 0.8588 4.8599 19.5068 | lr 9.7e-04 | norm 1.4369 | dt 0.026 | |
| type train | step 2920 | loss 0.0804 0.8623 4.9394 19.7084 | lr 9.7e-04 | norm 1.4624 | dt 0.026 | |
| type train | step 2930 | loss 0.0815 0.8705 4.9825 19.8284 | lr 9.7e-04 | norm 1.5287 | dt 0.026 | |
| type train | step 2940 | loss 0.0816 0.8623 4.8472 19.6251 | lr 9.7e-04 | norm 1.5581 | dt 0.026 | |
| type train | step 2950 | loss 0.0809 0.8753 5.0402 20.2366 | lr 9.7e-04 | norm 1.4120 | dt 0.026 | |
| type train | step 2960 | loss 0.0801 0.8543 4.9825 20.1078 | lr 9.7e-04 | norm 1.5744 | dt 0.026 | |
| type train | step 2970 | loss 0.0807 0.8708 4.9708 20.1460 | lr 9.7e-04 | norm 1.4252 | dt 0.026 | |
| type train | step 2980 | loss 0.0767 0.8417 4.7667 19.0074 | lr 9.7e-04 | norm 1.3842 | dt 0.026 | |
| type train | step 2990 | loss 0.0773 0.8599 4.8763 19.5828 | lr 9.7e-04 | norm 1.6253 | dt 0.026 | |
| type train | step 3000 | loss 0.0812 0.8731 4.8700 19.4906 | lr 9.7e-04 | norm 1.8407 | dt 0.026 | |
| type train | step 3010 | loss 0.0796 0.8663 5.0619 20.4867 | lr 9.7e-04 | norm 1.4719 | dt 0.026 | |
| type train | step 3020 | loss 0.0800 0.8492 4.7919 19.0495 | lr 9.7e-04 | norm 1.6438 | dt 0.026 | |
| type train | step 3030 | loss 0.0781 0.8658 4.9280 19.8943 | lr 9.7e-04 | norm 1.5675 | dt 0.026 | |
| type train | step 3040 | loss 0.0758 0.8410 4.8034 18.9889 | lr 9.7e-04 | norm 1.3317 | dt 0.026 | |
| type train | step 3050 | loss 0.0775 0.8686 5.0956 19.9920 | lr 9.7e-04 | norm 1.6820 | dt 0.029 | |
| type train | step 3060 | loss 0.0777 0.8571 4.8538 19.5000 | lr 9.7e-04 | norm 1.3329 | dt 0.026 | |
| type train | step 3070 | loss 0.0779 0.8655 4.9511 20.0690 | lr 9.7e-04 | norm 1.4266 | dt 0.026 | |
| type train | step 3080 | loss 0.0778 0.8280 4.7619 19.1125 | lr 9.7e-04 | norm 1.4442 | dt 0.026 | |
| type train | step 3090 | loss 0.0797 0.8885 4.9832 19.9500 | lr 9.7e-04 | norm 1.6586 | dt 0.026 | |
| type train | step 3100 | loss 0.0758 0.8430 4.7870 18.9518 | lr 9.7e-04 | norm 1.4314 | dt 0.026 | |
| type train | step 3110 | loss 0.0758 0.8331 4.8336 19.5156 | lr 9.7e-04 | norm 1.4611 | dt 0.026 | |
| type train | step 3120 | loss 0.0771 0.8352 4.7298 18.9707 | lr 9.7e-04 | norm 1.2035 | dt 0.026 | |
| type train | step 3130 | loss 0.0799 0.8916 5.1546 20.6402 | lr 9.7e-04 | norm 1.5345 | dt 0.026 | |
| type train | step 3140 | loss 0.0754 0.8377 4.7978 19.2487 | lr 9.7e-04 | norm 1.4598 | dt 0.026 | |
| type train | step 3150 | loss 0.0748 0.8160 4.6619 18.8974 | lr 9.7e-04 | norm 1.3386 | dt 0.026 | |
| type train | step 3160 | loss 0.0774 0.8507 4.8673 19.3476 | lr 9.7e-04 | norm 1.4265 | dt 0.025 | |
| type train | step 3170 | loss 0.0788 0.8539 4.7364 18.8671 | lr 9.7e-04 | norm 1.5050 | dt 0.026 | |
| type train | step 3180 | loss 0.0767 0.8358 4.6982 19.0723 | lr 9.7e-04 | norm 1.3205 | dt 0.026 | |
| type train | step 3190 | loss 0.0823 0.8739 4.8639 19.6198 | lr 9.6e-04 | norm 2.0987 | dt 0.026 | |
| type train | step 3200 | loss 0.0772 0.8667 4.8987 19.5769 | lr 9.6e-04 | norm 1.6150 | dt 0.027 | |
| type train | step 3210 | loss 0.0762 0.8336 4.8522 19.4688 | lr 9.6e-04 | norm 1.4885 | dt 0.026 | |
| type train | step 3220 | loss 0.0744 0.8428 4.7930 19.1026 | lr 9.6e-04 | norm 1.5011 | dt 0.026 | |
| type train | step 3230 | loss 0.0765 0.8478 4.8080 19.2651 | lr 9.6e-04 | norm 1.5298 | dt 0.026 | |
| type train | step 3240 | loss 0.0754 0.8316 4.7144 19.1808 | lr 9.6e-04 | norm 1.7375 | dt 0.026 | |
| type train | step 3250 | loss 0.0800 0.8791 4.9702 19.9348 | lr 9.6e-04 | norm 1.5032 | dt 0.026 | |
| type train | step 3260 | loss 0.0770 0.8361 4.7611 19.1917 | lr 9.6e-04 | norm 1.4781 | dt 0.026 | |
| type train | step 3270 | loss 0.0749 0.8317 4.8385 19.2632 | lr 9.6e-04 | norm 1.4257 | dt 0.025 | |
| type train | step 3280 | loss 0.0731 0.8370 4.7974 19.1777 | lr 9.6e-04 | norm 1.3657 | dt 0.026 | |
| type train | step 3290 | loss 0.0777 0.8681 4.8795 19.5955 | lr 9.6e-04 | norm 1.5514 | dt 0.026 | |
| type train | step 3300 | loss 0.0747 0.8255 4.6489 18.7415 | lr 9.6e-04 | norm 1.3289 | dt 0.026 | |
| type train | step 3310 | loss 0.0734 0.8295 4.7631 19.1078 | lr 9.6e-04 | norm 1.3811 | dt 0.026 | |
| type train | step 3320 | loss 0.0765 0.8323 4.6668 18.8995 | lr 9.6e-04 | norm 1.3340 | dt 0.026 | |
| type train | step 3330 | loss 0.0769 0.8520 4.9105 19.4483 | lr 9.6e-04 | norm 1.6655 | dt 0.026 | |
| type train | step 3340 | loss 0.0741 0.8357 4.6939 18.8570 | lr 9.6e-04 | norm 1.5486 | dt 0.026 | |
| type train | step 3350 | loss 0.0766 0.8561 4.8091 19.1038 | lr 9.6e-04 | norm 1.5689 | dt 0.026 | |
| type train | step 3360 | loss 0.0757 0.8380 4.7703 19.2713 | lr 9.6e-04 | norm 1.5744 | dt 0.027 | |
| type train | step 3370 | loss 0.0751 0.8248 4.7418 19.2604 | lr 9.6e-04 | norm 1.5531 | dt 0.026 | |
| type train | step 3380 | loss 0.0751 0.8430 4.8964 19.6378 | lr 9.6e-04 | norm 1.2816 | dt 0.026 | |
| type train | step 3390 | loss 0.0732 0.8291 4.8470 19.4250 | lr 9.6e-04 | norm 1.3587 | dt 0.026 | |
| type train | step 3400 | loss 0.0732 0.8464 4.8251 19.3160 | lr 9.6e-04 | norm 1.6457 | dt 0.026 | |
| type train | step 3410 | loss 0.0735 0.8076 4.5982 18.6778 | lr 9.6e-04 | norm 1.6067 | dt 0.026 | |
| type train | step 3420 | loss 0.0735 0.8310 4.6554 19.0357 | lr 9.6e-04 | norm 1.5797 | dt 0.026 | |
| type train | step 3430 | loss 0.0759 0.8507 4.8222 19.4565 | lr 9.6e-04 | norm 1.5573 | dt 0.026 | |
| type train | step 3440 | loss 0.0727 0.8290 4.8521 19.5237 | lr 9.6e-04 | norm 1.2952 | dt 0.027 | |
| type train | step 3450 | loss 0.0745 0.8317 4.7393 18.7954 | lr 9.6e-04 | norm 1.3903 | dt 0.026 | |
| type train | step 3460 | loss 0.0729 0.8487 4.7914 19.0782 | lr 9.6e-04 | norm 1.4289 | dt 0.027 | |
| type train | step 3470 | loss 0.0731 0.8488 4.7047 18.9360 | lr 9.6e-04 | norm 1.7006 | dt 0.027 | |
| type train | step 3480 | loss 0.0737 0.8404 4.8212 19.3643 | lr 9.6e-04 | norm 1.3915 | dt 0.026 | |
| type train | step 3490 | loss 0.0727 0.8209 4.6469 18.8430 | lr 9.6e-04 | norm 1.5222 | dt 0.026 | |
| type train | step 3500 | loss 0.0747 0.8502 4.8415 19.2494 | lr 9.6e-04 | norm 1.2988 | dt 0.026 | |
| type train | step 3510 | loss 0.0719 0.8092 4.5573 18.3822 | lr 9.6e-04 | norm 1.4362 | dt 0.026 | |
| type train | step 3520 | loss 0.0739 0.8328 4.7444 19.1276 | lr 9.5e-04 | norm 1.3999 | dt 0.025 | |
| type train | step 3530 | loss 0.0727 0.8386 4.8233 19.2810 | lr 9.5e-04 | norm 1.4111 | dt 0.026 | |
| type train | step 3540 | loss 0.0737 0.8448 4.8655 19.3957 | lr 9.5e-04 | norm 1.4583 | dt 0.026 | |
| type train | step 3550 | loss 0.0743 0.8366 4.7403 19.1902 | lr 9.5e-04 | norm 1.5204 | dt 0.026 | |
| type train | step 3560 | loss 0.0736 0.8468 4.9165 19.8274 | lr 9.5e-04 | norm 1.3700 | dt 0.026 | |
| type train | step 3570 | loss 0.0729 0.8338 4.8755 19.7000 | lr 9.5e-04 | norm 1.5297 | dt 0.026 | |
| type train | step 3580 | loss 0.0735 0.8466 4.8500 19.7616 | lr 9.5e-04 | norm 1.3771 | dt 0.026 | |
| type train | step 3590 | loss 0.0694 0.8218 4.6601 18.5886 | lr 9.5e-04 | norm 1.3293 | dt 0.026 | |
| type train | step 3600 | loss 0.0698 0.8390 4.7713 19.1675 | lr 9.5e-04 | norm 1.5789 | dt 0.026 | |
| type train | step 3610 | loss 0.0739 0.8450 4.7571 19.0538 | lr 9.5e-04 | norm 1.7729 | dt 0.026 | |
| type train | step 3620 | loss 0.0721 0.8443 4.9488 20.0694 | lr 9.5e-04 | norm 1.4309 | dt 0.026 | |
| type train | step 3630 | loss 0.0727 0.8246 4.6794 18.6539 | lr 9.5e-04 | norm 1.6134 | dt 0.026 | |
| type train | step 3640 | loss 0.0715 0.8388 4.8070 19.5221 | lr 9.5e-04 | norm 1.5189 | dt 0.026 | |
| type train | step 3650 | loss 0.0692 0.8201 4.6918 18.6026 | lr 9.5e-04 | norm 1.2927 | dt 0.026 | |
| type train | step 3660 | loss 0.0709 0.8443 4.9992 19.6912 | lr 9.5e-04 | norm 1.6511 | dt 0.028 | |
| type train | step 3670 | loss 0.0711 0.8360 4.7618 19.1239 | lr 9.5e-04 | norm 1.2979 | dt 0.026 | |
| type train | step 3680 | loss 0.0713 0.8429 4.8483 19.6728 | lr 9.5e-04 | norm 1.4140 | dt 0.026 | |
| type train | step 3690 | loss 0.0709 0.8067 4.6586 18.7251 | lr 9.5e-04 | norm 1.4277 | dt 0.026 | |
| type train | step 3700 | loss 0.0732 0.8631 4.8661 19.5659 | lr 9.5e-04 | norm 1.6175 | dt 0.026 | |
| type train | step 3710 | loss 0.0691 0.8214 4.6884 18.6027 | lr 9.5e-04 | norm 1.3960 | dt 0.026 | |
| type train | step 3720 | loss 0.0691 0.8118 4.7342 19.1467 | lr 9.5e-04 | norm 1.4230 | dt 0.025 | |
| type train | step 3730 | loss 0.0705 0.8159 4.6390 18.5881 | lr 9.5e-04 | norm 1.1817 | dt 0.026 | |
| type train | step 3740 | loss 0.0733 0.8689 5.0429 20.2612 | lr 9.5e-04 | norm 1.4718 | dt 0.026 | |
| type train | step 3750 | loss 0.0689 0.8168 4.6973 18.8398 | lr 9.5e-04 | norm 1.4254 | dt 0.026 | |
| type train | step 3760 | loss 0.0688 0.7967 4.5744 18.5165 | lr 9.5e-04 | norm 1.2880 | dt 0.026 | |
| type train | step 3770 | loss 0.0709 0.8306 4.7753 18.9817 | lr 9.5e-04 | norm 1.3990 | dt 0.026 | |
| type train | step 3780 | loss 0.0721 0.8345 4.6487 18.4719 | lr 9.5e-04 | norm 1.4632 | dt 0.026 | |
| type train | step 3790 | loss 0.0703 0.8170 4.6169 18.6887 | lr 9.5e-04 | norm 1.3022 | dt 0.026 | |
| type train | step 3800 | loss 0.0754 0.8479 4.7440 19.2037 | lr 9.5e-04 | norm 2.0348 | dt 0.026 | |
| type train | step 3810 | loss 0.0711 0.8448 4.8075 19.1856 | lr 9.5e-04 | norm 1.5883 | dt 0.026 | |
| type train | step 3820 | loss 0.0704 0.8147 4.7607 19.0785 | lr 9.4e-04 | norm 1.4354 | dt 0.026 | |
| type train | step 3830 | loss 0.0685 0.8223 4.7093 18.7591 | lr 9.4e-04 | norm 1.4860 | dt 0.025 | |
| type train | step 3840 | loss 0.0703 0.8295 4.7179 18.8920 | lr 9.4e-04 | norm 1.4922 | dt 0.026 | |
| type train | step 3850 | loss 0.0694 0.8126 4.6251 18.8222 | lr 9.4e-04 | norm 1.6934 | dt 0.026 | |
| type train | step 3860 | loss 0.0735 0.8542 4.8744 19.5768 | lr 9.4e-04 | norm 1.4505 | dt 0.026 | |
| type train | step 3870 | loss 0.0710 0.8174 4.6680 18.8216 | lr 9.4e-04 | norm 1.4172 | dt 0.026 | |
| type train | step 3880 | loss 0.0690 0.8126 4.7503 18.8930 | lr 9.4e-04 | norm 1.3734 | dt 0.026 | |
| type train | step 3890 | loss 0.0674 0.8196 4.7143 18.8196 | lr 9.4e-04 | norm 1.3205 | dt 0.026 | |
| type train | step 3900 | loss 0.0713 0.8481 4.8029 19.2158 | lr 9.4e-04 | norm 1.5091 | dt 0.026 | |
| type train | step 3910 | loss 0.0690 0.8100 4.5637 18.4600 | lr 9.4e-04 | norm 1.3004 | dt 0.026 | |
| type train | step 3920 | loss 0.0678 0.8119 4.6964 18.7713 | lr 9.4e-04 | norm 1.3391 | dt 0.026 | |
| type train | step 3930 | loss 0.0706 0.8098 4.5838 18.5570 | lr 9.4e-04 | norm 1.2707 | dt 0.026 | |
| type train | step 3940 | loss 0.0713 0.8323 4.8143 19.1252 | lr 9.4e-04 | norm 1.6239 | dt 0.026 | |
| type train | step 3950 | loss 0.0684 0.8208 4.6224 18.5086 | lr 9.4e-04 | norm 1.5200 | dt 0.026 | |
| type train | step 3960 | loss 0.0703 0.8379 4.7315 18.7254 | lr 9.4e-04 | norm 1.5356 | dt 0.026 | |
| type train | step 3970 | loss 0.0701 0.8241 4.6867 18.9538 | lr 9.4e-04 | norm 1.5365 | dt 0.026 | |
| type train | step 3980 | loss 0.0696 0.8083 4.6740 18.9051 | lr 9.4e-04 | norm 1.5045 | dt 0.026 | |
| type train | step 3990 | loss 0.0694 0.8243 4.8074 19.3013 | lr 9.4e-04 | norm 1.2534 | dt 0.026 | |
| type train | step 4000 | loss 0.0678 0.8126 4.7656 19.0962 | lr 9.4e-04 | norm 1.2967 | dt 0.026 | |
| type train | step 4010 | loss 0.0674 0.8277 4.7490 18.9767 | lr 9.4e-04 | norm 1.6055 | dt 0.025 | |
| type train | step 4020 | loss 0.0678 0.7943 4.5172 18.3139 | lr 9.4e-04 | norm 1.5696 | dt 0.025 | |
| type train | step 4030 | loss 0.0678 0.8143 4.5825 18.7507 | lr 9.4e-04 | norm 1.5437 | dt 0.027 | |
| type train | step 4040 | loss 0.0708 0.8348 4.7477 19.0910 | lr 9.4e-04 | norm 1.5295 | dt 0.026 | |
| type train | step 4050 | loss 0.0675 0.8118 4.7794 19.2177 | lr 9.4e-04 | norm 1.2799 | dt 0.026 | |
| type train | step 4060 | loss 0.0690 0.8152 4.6586 18.5065 | lr 9.4e-04 | norm 1.3502 | dt 0.026 | |
| type train | step 4070 | loss 0.0674 0.8337 4.7160 18.7327 | lr 9.4e-04 | norm 1.3796 | dt 0.026 | |
| type train | step 4080 | loss 0.0677 0.8378 4.6410 18.5834 | lr 9.4e-04 | norm 1.6538 | dt 0.026 | |
| type train | step 4090 | loss 0.0681 0.8261 4.7405 19.0503 | lr 9.3e-04 | norm 1.3403 | dt 0.026 | |
| type train | step 4100 | loss 0.0675 0.8097 4.5822 18.5919 | lr 9.3e-04 | norm 1.5298 | dt 0.026 | |
| type train | step 4110 | loss 0.0695 0.8320 4.7717 18.9638 | lr 9.3e-04 | norm 1.3091 | dt 0.026 | |
| type train | step 4120 | loss 0.0666 0.7940 4.4825 18.0466 | lr 9.3e-04 | norm 1.3978 | dt 0.026 | |
| type train | step 4130 | loss 0.0688 0.8191 4.6608 18.8435 | lr 9.3e-04 | norm 1.3734 | dt 0.026 | |
| type train | step 4140 | loss 0.0675 0.8251 4.7545 19.0023 | lr 9.3e-04 | norm 1.3819 | dt 0.025 | |
| type train | step 4150 | loss 0.0684 0.8336 4.8007 19.1064 | lr 9.3e-04 | norm 1.4216 | dt 0.026 | |
| type train | step 4160 | loss 0.0691 0.8205 4.6815 18.9428 | lr 9.3e-04 | norm 1.5013 | dt 0.026 | |
| type train | step 4170 | loss 0.0683 0.8300 4.8476 19.5425 | lr 9.3e-04 | norm 1.3512 | dt 0.026 | |
| type train | step 4180 | loss 0.0680 0.8191 4.7986 19.3725 | lr 9.3e-04 | norm 1.5139 | dt 0.026 | |
| type train | step 4190 | loss 0.0685 0.8348 4.7826 19.5298 | lr 9.3e-04 | norm 1.3388 | dt 0.026 | |
| type train | step 4200 | loss 0.0645 0.8097 4.5947 18.3254 | lr 9.3e-04 | norm 1.3226 | dt 0.026 | |
| type train | step 4210 | loss 0.0648 0.8259 4.7014 18.9060 | lr 9.3e-04 | norm 1.5539 | dt 0.026 | |
| type train | step 4220 | loss 0.0688 0.8264 4.6937 18.7757 | lr 9.3e-04 | norm 1.7503 | dt 0.026 | |
| type train | step 4230 | loss 0.0670 0.8263 4.8919 19.7953 | lr 9.3e-04 | norm 1.4059 | dt 0.026 | |
| type train | step 4240 | loss 0.0675 0.8104 4.6193 18.3602 | lr 9.3e-04 | norm 1.5768 | dt 0.026 | |
| type train | step 4250 | loss 0.0665 0.8264 4.7431 19.2780 | lr 9.3e-04 | norm 1.4929 | dt 0.026 | |
| type train | step 4260 | loss 0.0646 0.8076 4.6270 18.3581 | lr 9.3e-04 | norm 1.2664 | dt 0.026 | |
| type train | step 4270 | loss 0.0658 0.8292 4.9403 19.4325 | lr 9.3e-04 | norm 1.6082 | dt 0.028 | |
| type train | step 4280 | loss 0.0664 0.8233 4.7040 18.8688 | lr 9.3e-04 | norm 1.2719 | dt 0.026 | |
| type train | step 4290 | loss 0.0664 0.8282 4.7898 19.4015 | lr 9.3e-04 | norm 1.3803 | dt 0.026 | |
| type train | step 4300 | loss 0.0659 0.7944 4.5892 18.4430 | lr 9.3e-04 | norm 1.3822 | dt 0.026 | |
| type train | step 4310 | loss 0.0682 0.8477 4.8051 19.3109 | lr 9.3e-04 | norm 1.6002 | dt 0.026 | |
| type train | step 4320 | loss 0.0647 0.8087 4.6258 18.3655 | lr 9.3e-04 | norm 1.3689 | dt 0.026 | |
| type train | step 4330 | loss 0.0645 0.8009 4.6784 18.9314 | lr 9.3e-04 | norm 1.4086 | dt 0.026 | |
| type train | step 4340 | loss 0.0658 0.8053 4.5792 18.3483 | lr 9.2e-04 | norm 1.1592 | dt 0.026 | |
| type train | step 4350 | loss 0.0684 0.8508 4.9793 20.0053 | lr 9.2e-04 | norm 1.4375 | dt 0.025 | |
| type train | step 4360 | loss 0.0644 0.8021 4.6349 18.6030 | lr 9.2e-04 | norm 1.4125 | dt 0.026 | |
| type train | step 4370 | loss 0.0643 0.7849 4.5190 18.3049 | lr 9.2e-04 | norm 1.2783 | dt 0.026 | |
| type train | step 4380 | loss 0.0664 0.8184 4.7148 18.7702 | lr 9.2e-04 | norm 1.3819 | dt 0.026 | |
| type train | step 4390 | loss 0.0673 0.8218 4.5979 18.2741 | lr 9.2e-04 | norm 1.4297 | dt 0.026 | |
| type train | step 4400 | loss 0.0655 0.8074 4.5640 18.4633 | lr 9.2e-04 | norm 1.2716 | dt 0.026 | |
| type train | step 4410 | loss 0.0704 0.8308 4.6620 18.9248 | lr 9.2e-04 | norm 2.0027 | dt 0.026 | |
| type train | step 4420 | loss 0.0664 0.8309 4.7490 18.9064 | lr 9.2e-04 | norm 1.5534 | dt 0.026 | |
| type train | step 4430 | loss 0.0659 0.8051 4.6994 18.8773 | lr 9.2e-04 | norm 1.4079 | dt 0.026 | |
| type train | step 4440 | loss 0.0638 0.8107 4.6580 18.5270 | lr 9.2e-04 | norm 1.4509 | dt 0.026 | |
| type train | step 4450 | loss 0.0658 0.8190 4.6670 18.6604 | lr 9.2e-04 | norm 1.4677 | dt 0.026 | |
| type train | step 4460 | loss 0.0653 0.8045 4.5839 18.5979 | lr 9.2e-04 | norm 1.6731 | dt 0.026 | |
| type train | step 4470 | loss 0.0687 0.8406 4.8146 19.3516 | lr 9.2e-04 | norm 1.4156 | dt 0.026 | |
| type train | step 4480 | loss 0.0666 0.8058 4.6079 18.5978 | lr 9.2e-04 | norm 1.4065 | dt 0.026 | |
| type train | step 4490 | loss 0.0648 0.7994 4.6958 18.6978 | lr 9.2e-04 | norm 1.3399 | dt 0.026 | |
| type train | step 4500 | loss 0.0632 0.8113 4.6583 18.6173 | lr 9.2e-04 | norm 1.3010 | dt 0.026 | |
| type train | step 4510 | loss 0.0666 0.8362 4.7567 18.9939 | lr 9.2e-04 | norm 1.4769 | dt 0.026 | |
| type train | step 4520 | loss 0.0645 0.7981 4.5110 18.2710 | lr 9.2e-04 | norm 1.2864 | dt 0.026 | |
| type train | step 4530 | loss 0.0639 0.8041 4.6445 18.5659 | lr 9.2e-04 | norm 1.3250 | dt 0.026 | |
| type train | step 4540 | loss 0.0661 0.7974 4.5302 18.3584 | lr 9.2e-04 | norm 1.2337 | dt 0.025 | |
| type train | step 4550 | loss 0.0675 0.8203 4.7556 18.9457 | lr 9.2e-04 | norm 1.6024 | dt 0.026 | |
| type train | step 4560 | loss 0.0644 0.8105 4.5606 18.3187 | lr 9.2e-04 | norm 1.4801 | dt 0.025 | |
| type train | step 4570 | loss 0.0659 0.8272 4.6812 18.5411 | lr 9.2e-04 | norm 1.5096 | dt 0.026 | |
| type train | step 4580 | loss 0.0658 0.8116 4.6422 18.7577 | lr 9.1e-04 | norm 1.5159 | dt 0.026 | |
| type train | step 4590 | loss 0.0655 0.8019 4.6321 18.7098 | lr 9.1e-04 | norm 1.4680 | dt 0.026 | |
| type train | step 4600 | loss 0.0657 0.8137 4.7536 19.1210 | lr 9.1e-04 | norm 1.2472 | dt 0.026 | |
| type train | step 4610 | loss 0.0637 0.8017 4.7091 18.8993 | lr 9.1e-04 | norm 1.2717 | dt 0.026 | |
| type train | step 4620 | loss 0.0635 0.8162 4.6961 18.7586 | lr 9.1e-04 | norm 1.5569 | dt 0.026 | |
| type train | step 4630 | loss 0.0638 0.7847 4.4705 18.1308 | lr 9.1e-04 | norm 1.5260 | dt 0.026 | |
| type train | step 4640 | loss 0.0639 0.8070 4.5449 18.5718 | lr 9.1e-04 | norm 1.5255 | dt 0.026 | |
| type train | step 4650 | loss 0.0671 0.8249 4.6983 18.9334 | lr 9.1e-04 | norm 1.5324 | dt 0.026 | |
| type train | step 4660 | loss 0.0639 0.8016 4.7244 19.0368 | lr 9.1e-04 | norm 1.2562 | dt 0.026 | |
| type train | step 4670 | loss 0.0654 0.8054 4.6010 18.3381 | lr 9.1e-04 | norm 1.3342 | dt 0.026 | |
| type train | step 4680 | loss 0.0635 0.8215 4.6669 18.5562 | lr 9.1e-04 | norm 1.3283 | dt 0.026 | |
| type train | step 4690 | loss 0.0638 0.8253 4.5905 18.4149 | lr 9.1e-04 | norm 1.6157 | dt 0.026 | |
| type train | step 4700 | loss 0.0642 0.8182 4.6940 18.8554 | lr 9.1e-04 | norm 1.3227 | dt 0.026 | |
| type train | step 4710 | loss 0.0640 0.8026 4.5409 18.4241 | lr 9.1e-04 | norm 1.4824 | dt 0.026 | |
| type train | step 4720 | loss 0.0655 0.8223 4.7179 18.8016 | lr 9.1e-04 | norm 1.2523 | dt 0.026 | |
| type train | step 4730 | loss 0.0628 0.7855 4.4322 17.8538 | lr 9.1e-04 | norm 1.3539 | dt 0.026 | |
| type train | step 4740 | loss 0.0650 0.8075 4.6136 18.6870 | lr 9.1e-04 | norm 1.3245 | dt 0.026 | |
| type train | step 4750 | loss 0.0639 0.8153 4.7111 18.8588 | lr 9.1e-04 | norm 1.3694 | dt 0.025 | |
| type train | step 4760 | loss 0.0646 0.8243 4.7572 18.9522 | lr 9.1e-04 | norm 1.4000 | dt 0.025 | |
| type train | step 4770 | loss 0.0660 0.8150 4.6409 18.8086 | lr 9.1e-04 | norm 1.4997 | dt 0.026 | |
| type train | step 4780 | loss 0.0648 0.8216 4.8060 19.3732 | lr 9.1e-04 | norm 1.3437 | dt 0.026 | |
| type train | step 4790 | loss 0.0643 0.8120 4.7451 19.1726 | lr 9.1e-04 | norm 1.4745 | dt 0.026 | |
| type train | step 4800 | loss 0.0651 0.8282 4.7336 19.3700 | lr 9.1e-04 | norm 1.3413 | dt 0.026 | |
| type train | step 4810 | loss 0.0612 0.8047 4.5515 18.1978 | lr 9.0e-04 | norm 1.2985 | dt 0.026 | |
| type train | step 4820 | loss 0.0613 0.8191 4.6545 18.7589 | lr 9.0e-04 | norm 1.5225 | dt 0.026 | |
| type train | step 4830 | loss 0.0657 0.8188 4.6586 18.6173 | lr 9.0e-04 | norm 1.7294 | dt 0.025 | |
| type train | step 4840 | loss 0.0638 0.8220 4.8425 19.6830 | lr 9.0e-04 | norm 1.4017 | dt 0.026 | |
| type train | step 4850 | loss 0.0637 0.8053 4.5670 18.1820 | lr 9.0e-04 | norm 1.5586 | dt 0.026 | |
| type train | step 4860 | loss 0.0634 0.8207 4.6986 19.1238 | lr 9.0e-04 | norm 1.4820 | dt 0.026 | |
| type train | step 4870 | loss 0.0613 0.8060 4.5891 18.2066 | lr 9.0e-04 | norm 1.2690 | dt 0.026 | |
| type train | step 4880 | loss 0.0618 0.8253 4.8988 19.2764 | lr 9.0e-04 | norm 1.5730 | dt 0.027 | |
| type train | step 4890 | loss 0.0628 0.8202 4.6516 18.7260 | lr 9.0e-04 | norm 1.2590 | dt 0.026 | |
| type train | step 4900 | loss 0.0628 0.8223 4.7432 19.2773 | lr 9.0e-04 | norm 1.3702 | dt 0.026 | |
| type train | step 4910 | loss 0.0628 0.7909 4.5399 18.2717 | lr 9.0e-04 | norm 1.3481 | dt 0.027 | |
| type train | step 4920 | loss 0.0649 0.8379 4.7695 19.1330 | lr 9.0e-04 | norm 1.5922 | dt 0.026 | |
| type train | step 4930 | loss 0.0611 0.8053 4.5874 18.2201 | lr 9.0e-04 | norm 1.3487 | dt 0.026 | |
| type train | step 4940 | loss 0.0608 0.7980 4.6327 18.7944 | lr 9.0e-04 | norm 1.3784 | dt 0.025 | |
| type train | step 4950 | loss 0.0624 0.8013 4.5401 18.1981 | lr 9.0e-04 | norm 1.1354 | dt 0.026 | |
| type train | step 4960 | loss 0.0648 0.8469 4.9348 19.8723 | lr 9.0e-04 | norm 1.3922 | dt 0.026 | |
| type train | step 4970 | loss 0.0612 0.7976 4.5941 18.4624 | lr 9.0e-04 | norm 1.3592 | dt 0.026 | |
| type train | step 4980 | loss 0.0611 0.7779 4.4732 18.1917 | lr 9.0e-04 | norm 1.2513 | dt 0.025 | |
| type train | step 4990 | loss 0.0634 0.8130 4.6729 18.6285 | lr 9.0e-04 | norm 1.3510 | dt 0.026 | |
| type train | step 5000 | loss 0.0636 0.8166 4.5488 18.1927 | lr 9.0e-04 | norm 1.3902 | dt 0.026 | |
| type train | step 5010 | loss 0.0624 0.8009 4.5239 18.3359 | lr 9.0e-04 | norm 1.2426 | dt 0.026 | |
| type train | step 5020 | loss 0.0667 0.8235 4.6045 18.7964 | lr 9.0e-04 | norm 1.9755 | dt 0.026 | |
| type train | step 5030 | loss 0.0631 0.8271 4.6984 18.7817 | lr 8.9e-04 | norm 1.5271 | dt 0.026 | |
| type train | step 5040 | loss 0.0626 0.7994 4.6557 18.7632 | lr 8.9e-04 | norm 1.3824 | dt 0.026 | |
| type train | step 5050 | loss 0.0607 0.8081 4.6240 18.3727 | lr 8.9e-04 | norm 1.4456 | dt 0.026 | |
| type train | step 5060 | loss 0.0623 0.8120 4.6197 18.5435 | lr 8.9e-04 | norm 1.4312 | dt 0.026 | |
| type train | step 5070 | loss 0.0620 0.7999 4.5544 18.4402 | lr 8.9e-04 | norm 1.6400 | dt 0.025 | |
| type train | step 5080 | loss 0.0653 0.8328 4.7774 19.2453 | lr 8.9e-04 | norm 1.4059 | dt 0.026 | |
| type train | step 5090 | loss 0.0633 0.8018 4.5551 18.4816 | lr 8.9e-04 | norm 1.3781 | dt 0.026 | |
| type train | step 5100 | loss 0.0619 0.7952 4.6514 18.5993 | lr 8.9e-04 | norm 1.3250 | dt 0.026 | |
| type train | step 5110 | loss 0.0603 0.8072 4.6170 18.4749 | lr 8.9e-04 | norm 1.3042 | dt 0.026 | |
| type train | step 5120 | loss 0.0630 0.8311 4.7161 18.8859 | lr 8.9e-04 | norm 1.4710 | dt 0.026 | |
| type train | step 5130 | loss 0.0611 0.7928 4.4708 18.1212 | lr 8.9e-04 | norm 1.2578 | dt 0.026 | |
| type train | step 5140 | loss 0.0607 0.8003 4.6042 18.4540 | lr 8.9e-04 | norm 1.3086 | dt 0.026 | |
| type train | step 5150 | loss 0.0632 0.7928 4.4907 18.2696 | lr 8.9e-04 | norm 1.2354 | dt 0.026 | |
| type train | step 5160 | loss 0.0639 0.8118 4.7075 18.8363 | lr 8.9e-04 | norm 1.5687 | dt 0.025 | |
| type train | step 5170 | loss 0.0612 0.8035 4.5191 18.1975 | lr 8.9e-04 | norm 1.4522 | dt 0.026 | |
| type train | step 5180 | loss 0.0626 0.8222 4.6403 18.4617 | lr 8.9e-04 | norm 1.4964 | dt 0.026 | |
| type train | step 5190 | loss 0.0625 0.8043 4.6024 18.6149 | lr 8.9e-04 | norm 1.4977 | dt 0.026 | |
| type train | step 5200 | loss 0.0622 0.7947 4.5920 18.6123 | lr 8.9e-04 | norm 1.4536 | dt 0.026 | |
| type train | step 5210 | loss 0.0626 0.8085 4.6964 19.0256 | lr 8.9e-04 | norm 1.2141 | dt 0.026 | |
| type train | step 5220 | loss 0.0607 0.7968 4.6662 18.8020 | lr 8.9e-04 | norm 1.2393 | dt 0.026 | |
| type train | step 5230 | loss 0.0604 0.8126 4.6487 18.6558 | lr 8.8e-04 | norm 1.5327 | dt 0.026 | |
| type train | step 5240 | loss 0.0606 0.7792 4.4362 18.0093 | lr 8.8e-04 | norm 1.4903 | dt 0.026 | |
| type train | step 5250 | loss 0.0606 0.8019 4.5074 18.4532 | lr 8.8e-04 | norm 1.4997 | dt 0.026 | |
| type train | step 5260 | loss 0.0635 0.8157 4.6532 18.8319 | lr 8.8e-04 | norm 1.4969 | dt 0.025 | |
| type train | step 5270 | loss 0.0607 0.7962 4.6850 18.9530 | lr 8.8e-04 | norm 1.2286 | dt 0.026 | |
| type train | step 5280 | loss 0.0622 0.7995 4.5749 18.2078 | lr 8.8e-04 | norm 1.3036 | dt 0.026 | |
| type train | step 5290 | loss 0.0608 0.8186 4.6255 18.4431 | lr 8.8e-04 | norm 1.3102 | dt 0.026 | |
| type train | step 5300 | loss 0.0607 0.8172 4.5563 18.3025 | lr 8.8e-04 | norm 1.5944 | dt 0.026 | |
| type train | step 5310 | loss 0.0610 0.8128 4.6602 18.7190 | lr 8.8e-04 | norm 1.2798 | dt 0.025 | |
| type train | step 5320 | loss 0.0608 0.7935 4.5047 18.3327 | lr 8.8e-04 | norm 1.4655 | dt 0.026 | |
| type train | step 5330 | loss 0.0626 0.8161 4.6738 18.6868 | lr 8.8e-04 | norm 1.2283 | dt 0.026 | |
| type train | step 5340 | loss 0.0598 0.7784 4.4033 17.7405 | lr 8.8e-04 | norm 1.3358 | dt 0.026 | |
| type train | step 5350 | loss 0.0622 0.7982 4.5827 18.5883 | lr 8.8e-04 | norm 1.3071 | dt 0.026 | |
| type train | step 5360 | loss 0.0605 0.8065 4.6743 18.7366 | lr 8.8e-04 | norm 1.3441 | dt 0.025 | |
| type train | step 5370 | loss 0.0614 0.8156 4.7223 18.8543 | lr 8.8e-04 | norm 1.3689 | dt 0.026 | |
| type train | step 5380 | loss 0.0628 0.8049 4.6099 18.7100 | lr 8.8e-04 | norm 1.4699 | dt 0.026 | |
| type train | step 5390 | loss 0.0613 0.8114 4.7612 19.2508 | lr 8.8e-04 | norm 1.3126 | dt 0.026 | |
| type train | step 5400 | loss 0.0612 0.8044 4.7069 19.0391 | lr 8.8e-04 | norm 1.4429 | dt 0.026 | |
| type train | step 5410 | loss 0.0622 0.8188 4.6943 19.2666 | lr 8.8e-04 | norm 1.3121 | dt 0.026 | |
| type train | step 5420 | loss 0.0582 0.7956 4.5209 18.0952 | lr 8.8e-04 | norm 1.2781 | dt 0.026 | |
| type train | step 5430 | loss 0.0585 0.8098 4.6035 18.6418 | lr 8.8e-04 | norm 1.4903 | dt 0.026 | |
| type train | step 5440 | loss 0.0625 0.8064 4.6195 18.5103 | lr 8.7e-04 | norm 1.7223 | dt 0.026 | |
| type train | step 5450 | loss 0.0605 0.8106 4.8018 19.5793 | lr 8.7e-04 | norm 1.3730 | dt 0.026 | |
| type train | step 5460 | loss 0.0605 0.7955 4.5349 18.0738 | lr 8.7e-04 | norm 1.5176 | dt 0.026 | |
| type train | step 5470 | loss 0.0600 0.8099 4.6638 19.0338 | lr 8.7e-04 | norm 1.4395 | dt 0.026 | |
| type train | step 5480 | loss 0.0584 0.7963 4.5446 18.1127 | lr 8.7e-04 | norm 1.2727 | dt 0.026 | |
| type train | step 5490 | loss 0.0589 0.8145 4.8591 19.1903 | lr 8.7e-04 | norm 1.5627 | dt 0.029 | |
| type train | step 5500 | loss 0.0598 0.8088 4.6164 18.6180 | lr 8.7e-04 | norm 1.2388 | dt 0.026 | |
| type train | step 5510 | loss 0.0598 0.8140 4.7023 19.1765 | lr 8.7e-04 | norm 1.3370 | dt 0.026 | |
| type train | step 5520 | loss 0.0598 0.7814 4.5013 18.1702 | lr 8.7e-04 | norm 1.3238 | dt 0.025 | |
| type train | step 5530 | loss 0.0620 0.8261 4.7277 18.9936 | lr 8.7e-04 | norm 1.5679 | dt 0.026 | |
| type train | step 5540 | loss 0.0583 0.7954 4.5483 18.1145 | lr 8.7e-04 | norm 1.3200 | dt 0.026 | |
| type train | step 5550 | loss 0.0579 0.7882 4.5984 18.6834 | lr 8.7e-04 | norm 1.4023 | dt 0.026 | |
| type train | step 5560 | loss 0.0592 0.7920 4.5043 18.0903 | lr 8.7e-04 | norm 1.1203 | dt 0.026 | |
| type train | step 5570 | loss 0.0616 0.8349 4.8979 19.7775 | lr 8.7e-04 | norm 1.3630 | dt 0.025 | |
| type train | step 5580 | loss 0.0581 0.7892 4.5665 18.3573 | lr 8.7e-04 | norm 1.3386 | dt 0.025 | |
| type train | step 5590 | loss 0.0581 0.7701 4.4408 18.0844 | lr 8.7e-04 | norm 1.2406 | dt 0.026 | |
| type train | step 5600 | loss 0.0599 0.8060 4.6382 18.5226 | lr 8.7e-04 | norm 1.3286 | dt 0.026 | |
| type train | step 5610 | loss 0.0606 0.8058 4.5144 18.1164 | lr 8.7e-04 | norm 1.3808 | dt 0.026 | |
| type train | step 5620 | loss 0.0594 0.7906 4.4886 18.2460 | lr 8.7e-04 | norm 1.2217 | dt 0.026 | |
| type train | step 5630 | loss 0.0634 0.8099 4.5639 18.6923 | lr 8.6e-04 | norm 1.9359 | dt 0.026 | |
| type train | step 5640 | loss 0.0601 0.8153 4.6583 18.6972 | lr 8.6e-04 | norm 1.5106 | dt 0.026 | |
| type train | step 5650 | loss 0.0596 0.7912 4.6149 18.6720 | lr 8.6e-04 | norm 1.3623 | dt 0.026 | |
| type train | step 5660 | loss 0.0577 0.7994 4.5913 18.2787 | lr 8.6e-04 | norm 1.4372 | dt 0.026 | |
| type train | step 5670 | loss 0.0591 0.8017 4.5822 18.4611 | lr 8.6e-04 | norm 1.4154 | dt 0.026 | |
| type train | step 5680 | loss 0.0588 0.7895 4.5289 18.3371 | lr 8.6e-04 | norm 1.6147 | dt 0.026 | |
| type train | step 5690 | loss 0.0620 0.8218 4.7412 19.1643 | lr 8.6e-04 | norm 1.3870 | dt 0.026 | |
| type train | step 5700 | loss 0.0603 0.7924 4.5144 18.3801 | lr 8.6e-04 | norm 1.3421 | dt 0.026 | |
| type train | step 5710 | loss 0.0590 0.7838 4.6125 18.5227 | lr 8.6e-04 | norm 1.3063 | dt 0.026 | |
| type train | step 5720 | loss 0.0569 0.7979 4.5824 18.3794 | lr 8.6e-04 | norm 1.2752 | dt 0.026 | |
| type train | step 5730 | loss 0.0599 0.8212 4.6942 18.7984 | lr 8.6e-04 | norm 1.4366 | dt 0.026 | |
| type train | step 5740 | loss 0.0583 0.7836 4.4504 18.0283 | lr 8.6e-04 | norm 1.2369 | dt 0.026 | |
| type train | step 5750 | loss 0.0579 0.7899 4.5749 18.3898 | lr 8.6e-04 | norm 1.2748 | dt 0.026 | |
| type train | step 5760 | loss 0.0602 0.7823 4.4485 18.1639 | lr 8.6e-04 | norm 1.2120 | dt 0.025 | |
| type train | step 5770 | loss 0.0610 0.7998 4.6618 18.7453 | lr 8.6e-04 | norm 1.5790 | dt 0.025 | |
| type train | step 5780 | loss 0.0582 0.7913 4.4867 18.1220 | lr 8.6e-04 | norm 1.4350 | dt 0.026 | |
| type train | step 5790 | loss 0.0597 0.8111 4.6150 18.3790 | lr 8.6e-04 | norm 1.4740 | dt 0.026 | |
| type train | step 5800 | loss 0.0592 0.7948 4.5730 18.5271 | lr 8.6e-04 | norm 1.4736 | dt 0.026 | |
| type train | step 5810 | loss 0.0594 0.7836 4.5675 18.5231 | lr 8.6e-04 | norm 1.4454 | dt 0.026 | |
| type train | step 5820 | loss 0.0594 0.7995 4.6564 18.9358 | lr 8.5e-04 | norm 1.1820 | dt 0.026 | |
| type train | step 5830 | loss 0.0576 0.7893 4.6242 18.7029 | lr 8.5e-04 | norm 1.2108 | dt 0.026 | |
| type train | step 5840 | loss 0.0571 0.8010 4.6146 18.5499 | lr 8.5e-04 | norm 1.5152 | dt 0.025 | |
| type train | step 5850 | loss 0.0578 0.7689 4.4084 17.9323 | lr 8.5e-04 | norm 1.4625 | dt 0.026 | |
| type train | step 5860 | loss 0.0577 0.7941 4.4747 18.3682 | lr 8.5e-04 | norm 1.4785 | dt 0.026 | |
| type train | step 5870 | loss 0.0604 0.8035 4.6134 18.7597 | lr 8.5e-04 | norm 1.4745 | dt 0.026 | |
| type train | step 5880 | loss 0.0578 0.7863 4.6564 18.8640 | lr 8.5e-04 | norm 1.2064 | dt 0.026 | |
| type train | step 5890 | loss 0.0593 0.7882 4.5376 18.1344 | lr 8.5e-04 | norm 1.2954 | dt 0.026 | |
| type train | step 5900 | loss 0.0579 0.8068 4.5932 18.3598 | lr 8.5e-04 | norm 1.2911 | dt 0.026 | |
| type train | step 5910 | loss 0.0579 0.8047 4.5259 18.2241 | lr 8.5e-04 | norm 1.5757 | dt 0.026 | |
| type train | step 5920 | loss 0.0582 0.8048 4.6355 18.6451 | lr 8.5e-04 | norm 1.2578 | dt 0.026 | |
| type train | step 5930 | loss 0.0579 0.7852 4.4711 18.2441 | lr 8.5e-04 | norm 1.4452 | dt 0.026 | |
| type train | step 5940 | loss 0.0596 0.8065 4.6338 18.6069 | lr 8.5e-04 | norm 1.2179 | dt 0.026 | |
| type train | step 5950 | loss 0.0570 0.7687 4.3723 17.6338 | lr 8.5e-04 | norm 1.3310 | dt 0.026 | |
| type train | step 5960 | loss 0.0594 0.7872 4.5458 18.5121 | lr 8.5e-04 | norm 1.2967 | dt 0.025 | |
| type train | step 5970 | loss 0.0577 0.7951 4.6390 18.6547 | lr 8.5e-04 | norm 1.2986 | dt 0.026 | |
| type train | step 5980 | loss 0.0585 0.8059 4.6951 18.7839 | lr 8.5e-04 | norm 1.3561 | dt 0.026 | |
| type train | step 5990 | loss 0.0601 0.7948 4.5806 18.6298 | lr 8.5e-04 | norm 1.4380 | dt 0.026 | |
| type train | step 6000 | loss 0.0588 0.8019 4.7262 19.1706 | lr 8.4e-04 | norm 1.2981 | dt 0.027 | |
| type train | step 6010 | loss 0.0587 0.7942 4.6735 18.9510 | lr 8.4e-04 | norm 1.4438 | dt 0.025 | |
| type train | step 6020 | loss 0.0592 0.8073 4.6605 19.1889 | lr 8.4e-04 | norm 1.2960 | dt 0.026 | |
| type train | step 6030 | loss 0.0555 0.7864 4.4925 17.9979 | lr 8.4e-04 | norm 1.2526 | dt 0.026 | |
| type train | step 6040 | loss 0.0556 0.8008 4.5706 18.5715 | lr 8.4e-04 | norm 1.4747 | dt 0.025 | |
| type train | step 6050 | loss 0.0596 0.7961 4.5780 18.4257 | lr 8.4e-04 | norm 1.6742 | dt 0.026 | |
| type train | step 6060 | loss 0.0578 0.8001 4.7667 19.4842 | lr 8.4e-04 | norm 1.3500 | dt 0.026 | |
| type train | step 6070 | loss 0.0580 0.7837 4.5192 18.0160 | lr 8.4e-04 | norm 1.5094 | dt 0.026 | |
| type train | step 6080 | loss 0.0576 0.7977 4.6349 18.9810 | lr 8.4e-04 | norm 1.4260 | dt 0.026 | |
| type train | step 6090 | loss 0.0559 0.7859 4.5134 18.0642 | lr 8.4e-04 | norm 1.2309 | dt 0.026 | |
| type train | step 6100 | loss 0.0563 0.8044 4.8273 19.1345 | lr 8.4e-04 | norm 1.5476 | dt 0.027 | |
| type train | step 6110 | loss 0.0572 0.7976 4.5837 18.5520 | lr 8.4e-04 | norm 1.2098 | dt 0.026 | |
| type train | step 6120 | loss 0.0572 0.8032 4.6645 19.1107 | lr 8.4e-04 | norm 1.3258 | dt 0.026 | |
| type train | step 6130 | loss 0.0573 0.7732 4.4693 18.0912 | lr 8.4e-04 | norm 1.3225 | dt 0.026 | |
| type train | step 6140 | loss 0.0592 0.8151 4.6857 18.9279 | lr 8.4e-04 | norm 1.5515 | dt 0.026 | |
| type train | step 6150 | loss 0.0559 0.7856 4.5142 18.0301 | lr 8.4e-04 | norm 1.3196 | dt 0.026 | |
| type train | step 6160 | loss 0.0554 0.7802 4.5713 18.6152 | lr 8.4e-04 | norm 1.3604 | dt 0.026 | |
| type train | step 6170 | loss 0.0567 0.7833 4.4780 18.0323 | lr 8.4e-04 | norm 1.0892 | dt 0.025 | |
| type train | step 6180 | loss 0.0591 0.8229 4.8596 19.7122 | lr 8.3e-04 | norm 1.3432 | dt 0.026 | |
| type train | step 6190 | loss 0.0558 0.7789 4.5377 18.2812 | lr 8.3e-04 | norm 1.3303 | dt 0.026 | |
| type train | step 6200 | loss 0.0559 0.7596 4.4172 18.0143 | lr 8.3e-04 | norm 1.2248 | dt 0.026 | |
| type train | step 6210 | loss 0.0577 0.7976 4.6086 18.4546 | lr 8.3e-04 | norm 1.3212 | dt 0.026 | |
| type train | step 6220 | loss 0.0581 0.7950 4.4846 18.0696 | lr 8.3e-04 | norm 1.3574 | dt 0.026 | |
| type train | step 6230 | loss 0.0568 0.7808 4.4592 18.1851 | lr 8.3e-04 | norm 1.2051 | dt 0.026 | |
| type train | step 6240 | loss 0.0608 0.7981 4.5243 18.6361 | lr 8.3e-04 | norm 1.9061 | dt 0.026 | |
| type train | step 6250 | loss 0.0576 0.8042 4.6267 18.6188 | lr 8.3e-04 | norm 1.4920 | dt 0.026 | |
| type train | step 6260 | loss 0.0574 0.7812 4.5921 18.6408 | lr 8.3e-04 | norm 1.3545 | dt 0.026 | |
| type train | step 6270 | loss 0.0553 0.7902 4.5629 18.1936 | lr 8.3e-04 | norm 1.4072 | dt 0.026 | |
| type train | step 6280 | loss 0.0567 0.7923 4.5473 18.3779 | lr 8.3e-04 | norm 1.3974 | dt 0.025 | |
| type train | step 6290 | loss 0.0564 0.7796 4.5084 18.2762 | lr 8.3e-04 | norm 1.6072 | dt 0.026 | |
| type train | step 6300 | loss 0.0596 0.8108 4.7021 19.1073 | lr 8.3e-04 | norm 1.3628 | dt 0.026 | |
| type train | step 6310 | loss 0.0579 0.7855 4.4763 18.3331 | lr 8.3e-04 | norm 1.3213 | dt 0.026 | |
| type train | step 6320 | loss 0.0566 0.7732 4.5792 18.5003 | lr 8.3e-04 | norm 1.2892 | dt 0.026 | |
| type train | step 6330 | loss 0.0546 0.7892 4.5590 18.3333 | lr 8.3e-04 | norm 1.2656 | dt 0.026 | |
| type train | step 6340 | loss 0.0575 0.8123 4.6656 18.7441 | lr 8.3e-04 | norm 1.4214 | dt 0.026 | |
| type train | step 6350 | loss 0.0560 0.7728 4.4216 17.9727 | lr 8.2e-04 | norm 1.2157 | dt 0.026 | |
| type train | step 6360 | loss 0.0558 0.7800 4.5345 18.3369 | lr 8.2e-04 | norm 1.2547 | dt 0.026 | |
| type train | step 6370 | loss 0.0579 0.7721 4.4187 18.1063 | lr 8.2e-04 | norm 1.1907 | dt 0.026 | |
| type train | step 6380 | loss 0.0586 0.7878 4.6267 18.6742 | lr 8.2e-04 | norm 1.5276 | dt 0.026 | |
| type train | step 6390 | loss 0.0559 0.7806 4.4593 18.0658 | lr 8.2e-04 | norm 1.4126 | dt 0.026 | |
| type train | step 6400 | loss 0.0575 0.8005 4.5830 18.3168 | lr 8.2e-04 | norm 1.4413 | dt 0.026 | |
| type train | step 6410 | loss 0.0570 0.7859 4.5381 18.4534 | lr 8.2e-04 | norm 1.4637 | dt 0.026 | |
| type train | step 6420 | loss 0.0572 0.7722 4.5426 18.4598 | lr 8.2e-04 | norm 1.4263 | dt 0.026 | |
| type train | step 6430 | loss 0.0574 0.7907 4.6309 18.8849 | lr 8.2e-04 | norm 1.1851 | dt 0.026 | |
| type train | step 6440 | loss 0.0555 0.7796 4.5960 18.6353 | lr 8.2e-04 | norm 1.2038 | dt 0.025 | |
| type train | step 6450 | loss 0.0552 0.7905 4.5832 18.4840 | lr 8.2e-04 | norm 1.4743 | dt 0.026 | |
| type train | step 6460 | loss 0.0557 0.7580 4.3854 17.8724 | lr 8.2e-04 | norm 1.4396 | dt 0.026 | |
| type train | step 6470 | loss 0.0555 0.7829 4.4451 18.3148 | lr 8.2e-04 | norm 1.4590 | dt 0.026 | |
| type train | step 6480 | loss 0.0581 0.7939 4.5762 18.7027 | lr 8.2e-04 | norm 1.4642 | dt 0.026 | |
| type train | step 6490 | loss 0.0559 0.7766 4.6286 18.8132 | lr 8.2e-04 | norm 1.1998 | dt 0.026 | |
| type train | step 6500 | loss 0.0570 0.7779 4.5069 18.0782 | lr 8.2e-04 | norm 1.2830 | dt 0.026 | |
| type train | step 6510 | loss 0.0560 0.7959 4.5653 18.2903 | lr 8.2e-04 | norm 1.2798 | dt 0.027 | |
| type train | step 6520 | loss 0.0555 0.7940 4.4932 18.1749 | lr 8.1e-04 | norm 1.5307 | dt 0.026 | |
| type train | step 6530 | loss 0.0561 0.7939 4.6070 18.5948 | lr 8.1e-04 | norm 1.2271 | dt 0.026 | |
| type train | step 6540 | loss 0.0560 0.7760 4.4444 18.1901 | lr 8.1e-04 | norm 1.4491 | dt 0.026 | |
| type train | step 6550 | loss 0.0575 0.7977 4.6050 18.5510 | lr 8.1e-04 | norm 1.1951 | dt 0.026 | |
| type train | step 6560 | loss 0.0548 0.7580 4.3446 17.5839 | lr 8.1e-04 | norm 1.3079 | dt 0.026 | |
| type train | step 6570 | loss 0.0573 0.7758 4.5213 18.4523 | lr 8.1e-04 | norm 1.2852 | dt 0.026 | |
| type train | step 6580 | loss 0.0556 0.7863 4.6054 18.6114 | lr 8.1e-04 | norm 1.2728 | dt 0.026 | |
| type train | step 6590 | loss 0.0565 0.7955 4.6751 18.7199 | lr 8.1e-04 | norm 1.3642 | dt 0.025 | |
| type train | step 6600 | loss 0.0578 0.7842 4.5528 18.5682 | lr 8.1e-04 | norm 1.4397 | dt 0.026 | |
| type train | step 6610 | loss 0.0565 0.7907 4.7031 19.1343 | lr 8.1e-04 | norm 1.2681 | dt 0.026 | |
| type train | step 6620 | loss 0.0567 0.7838 4.6410 18.8992 | lr 8.1e-04 | norm 1.4202 | dt 0.025 | |
| type train | step 6630 | loss 0.0570 0.7969 4.6396 19.1256 | lr 8.1e-04 | norm 1.2779 | dt 0.026 | |
| type train | step 6640 | loss 0.0536 0.7759 4.4675 17.9514 | lr 8.1e-04 | norm 1.2190 | dt 0.026 | |
| type train | step 6650 | loss 0.0536 0.7927 4.5420 18.5242 | lr 8.1e-04 | norm 1.4542 | dt 0.026 | |
| type train | step 6660 | loss 0.0576 0.7871 4.5498 18.3798 | lr 8.1e-04 | norm 1.6609 | dt 0.026 | |
| type train | step 6670 | loss 0.0556 0.7903 4.7370 19.4488 | lr 8.1e-04 | norm 1.3284 | dt 0.025 | |
| type train | step 6680 | loss 0.0559 0.7747 4.4844 17.9490 | lr 8.1e-04 | norm 1.5047 | dt 0.025 | |
| type train | step 6690 | loss 0.0557 0.7872 4.6120 18.9104 | lr 8.0e-04 | norm 1.4090 | dt 0.026 | |
| type train | step 6700 | loss 0.0540 0.7775 4.4872 18.0161 | lr 8.0e-04 | norm 1.2202 | dt 0.026 | |
| type train | step 6710 | loss 0.0545 0.7946 4.7851 19.0937 | lr 8.0e-04 | norm 1.5451 | dt 0.027 | |
| type train | step 6720 | loss 0.0552 0.7884 4.5597 18.4869 | lr 8.0e-04 | norm 1.1980 | dt 0.026 | |
| type train | step 6730 | loss 0.0552 0.7958 4.6354 19.0448 | lr 8.0e-04 | norm 1.2957 | dt 0.026 | |
| type train | step 6740 | loss 0.0556 0.7629 4.4406 18.0363 | lr 8.0e-04 | norm 1.3114 | dt 0.026 | |
| type train | step 6750 | loss 0.0572 0.8053 4.6591 18.8584 | lr 8.0e-04 | norm 1.5542 | dt 0.026 | |
| type train | step 6760 | loss 0.0538 0.7764 4.4886 17.9728 | lr 8.0e-04 | norm 1.2881 | dt 0.025 | |
| type train | step 6770 | loss 0.0537 0.7724 4.5444 18.5797 | lr 8.0e-04 | norm 1.3484 | dt 0.026 | |
| type train | step 6780 | loss 0.0548 0.7747 4.4568 17.9865 | lr 8.0e-04 | norm 1.0897 | dt 0.026 | |
| type train | step 6790 | loss 0.0573 0.8132 4.8338 19.6393 | lr 8.0e-04 | norm 1.3217 | dt 0.026 | |
| type train | step 6800 | loss 0.0541 0.7726 4.5123 18.2256 | lr 8.0e-04 | norm 1.3276 | dt 0.026 | |
| type train | step 6810 | loss 0.0539 0.7516 4.3936 17.9732 | lr 8.0e-04 | norm 1.1985 | dt 0.026 | |
| type train | step 6820 | loss 0.0560 0.7894 4.5831 18.4118 | lr 8.0e-04 | norm 1.3148 | dt 0.025 | |
| type train | step 6830 | loss 0.0562 0.7855 4.4626 18.0532 | lr 8.0e-04 | norm 1.3430 | dt 0.026 | |
| type train | step 6840 | loss 0.0550 0.7720 4.4340 18.1354 | lr 8.0e-04 | norm 1.1871 | dt 0.026 | |
| type train | step 6850 | loss 0.0589 0.7878 4.5043 18.5804 | lr 7.9e-04 | norm 1.8927 | dt 0.026 | |
| type train | step 6860 | loss 0.0559 0.7952 4.5979 18.5923 | lr 7.9e-04 | norm 1.4775 | dt 0.026 | |
| type train | step 6870 | loss 0.0555 0.7737 4.5682 18.5959 | lr 7.9e-04 | norm 1.3337 | dt 0.025 | |
| type train | step 6880 | loss 0.0535 0.7824 4.5459 18.1356 | lr 7.9e-04 | norm 1.4010 | dt 0.026 | |
| type train | step 6890 | loss 0.0551 0.7831 4.5212 18.3483 | lr 7.9e-04 | norm 1.3905 | dt 0.026 | |
| type train | step 6900 | loss 0.0547 0.7706 4.4901 18.2284 | lr 7.9e-04 | norm 1.5841 | dt 0.026 | |
| type train | step 6910 | loss 0.0576 0.8018 4.6811 19.0327 | lr 7.9e-04 | norm 1.3445 | dt 0.026 | |
| type train | step 6920 | loss 0.0559 0.7775 4.4481 18.3040 | lr 7.9e-04 | norm 1.3052 | dt 0.026 | |
| type train | step 6930 | loss 0.0546 0.7656 4.5562 18.4446 | lr 7.9e-04 | norm 1.2721 | dt 0.026 | |
| type train | step 6940 | loss 0.0529 0.7831 4.5359 18.2902 | lr 7.9e-04 | norm 1.2426 | dt 0.026 | |
| type train | step 6950 | loss 0.0557 0.8040 4.6404 18.7272 | lr 7.9e-04 | norm 1.4158 | dt 0.026 | |
| type train | step 6960 | loss 0.0542 0.7645 4.4048 17.9120 | lr 7.9e-04 | norm 1.1970 | dt 0.026 | |
| type train | step 6970 | loss 0.0537 0.7729 4.5068 18.3036 | lr 7.9e-04 | norm 1.2472 | dt 0.026 | |
| type train | step 6980 | loss 0.0558 0.7660 4.4006 18.0739 | lr 7.9e-04 | norm 1.1675 | dt 0.025 | |
| type train | step 6990 | loss 0.0567 0.7807 4.6048 18.6378 | lr 7.9e-04 | norm 1.5213 | dt 0.026 | |
| type train | step 7000 | loss 0.0541 0.7743 4.4357 18.0375 | lr 7.9e-04 | norm 1.3817 | dt 0.026 | |
| type train | step 7010 | loss 0.0560 0.7946 4.5617 18.2679 | lr 7.8e-04 | norm 1.4392 | dt 0.025 | |
| type train | step 7020 | loss 0.0554 0.7793 4.5187 18.4165 | lr 7.8e-04 | norm 1.4543 | dt 0.026 | |
| type train | step 7030 | loss 0.0556 0.7649 4.5289 18.3976 | lr 7.8e-04 | norm 1.4165 | dt 0.026 | |
| type train | step 7040 | loss 0.0558 0.7844 4.6107 18.8385 | lr 7.8e-04 | norm 1.1730 | dt 0.026 | |
| type train | step 7050 | loss 0.0538 0.7744 4.5710 18.6175 | lr 7.8e-04 | norm 1.1853 | dt 0.026 | |
| type train | step 7060 | loss 0.0535 0.7865 4.5560 18.4708 | lr 7.8e-04 | norm 1.4639 | dt 0.025 | |
| type train | step 7070 | loss 0.0541 0.7538 4.3614 17.8345 | lr 7.8e-04 | norm 1.4256 | dt 0.026 | |
| type train | step 7080 | loss 0.0540 0.7787 4.4222 18.2904 | lr 7.8e-04 | norm 1.4404 | dt 0.026 | |
| type train | step 7090 | loss 0.0565 0.7856 4.5486 18.6427 | lr 7.8e-04 | norm 1.4329 | dt 0.026 | |
| type train | step 7100 | loss 0.0545 0.7704 4.6085 18.7795 | lr 7.8e-04 | norm 1.1873 | dt 0.026 | |
| type train | step 7110 | loss 0.0555 0.7741 4.4802 18.0499 | lr 7.8e-04 | norm 1.2573 | dt 0.026 | |
| type train | step 7120 | loss 0.0543 0.7888 4.5404 18.2528 | lr 7.8e-04 | norm 1.2445 | dt 0.026 | |
| type train | step 7130 | loss 0.0540 0.7891 4.4708 18.1498 | lr 7.8e-04 | norm 1.5219 | dt 0.026 | |
| type train | step 7140 | loss 0.0544 0.7875 4.5857 18.5662 | lr 7.8e-04 | norm 1.2094 | dt 0.026 | |
| type train | step 7150 | loss 0.0544 0.7700 4.4228 18.1498 | lr 7.8e-04 | norm 1.4307 | dt 0.026 | |
| type train | step 7160 | loss 0.0558 0.7913 4.5866 18.5072 | lr 7.8e-04 | norm 1.1680 | dt 0.025 | |
| type train | step 7170 | loss 0.0535 0.7536 4.3193 17.5671 | lr 7.7e-04 | norm 1.2986 | dt 0.026 | |
| type train | step 7180 | loss 0.0558 0.7716 4.5015 18.4035 | lr 7.7e-04 | norm 1.2565 | dt 0.026 | |
| type train | step 7190 | loss 0.0542 0.7832 4.5800 18.5865 | lr 7.7e-04 | norm 1.2866 | dt 0.026 | |
| type train | step 7200 | loss 0.0547 0.7901 4.6496 18.6771 | lr 7.7e-04 | norm 1.3282 | dt 0.026 | |
| type train | step 7210 | loss 0.0566 0.7793 4.5243 18.5048 | lr 7.7e-04 | norm 1.4266 | dt 0.026 | |
| type train | step 7220 | loss 0.0552 0.7854 4.6852 19.1123 | lr 7.7e-04 | norm 1.2678 | dt 0.026 | |
| type train | step 7230 | loss 0.0549 0.7786 4.6160 18.8827 | lr 7.7e-04 | norm 1.4115 | dt 0.026 | |
| type train | step 7240 | loss 0.0556 0.7908 4.6163 19.0673 | lr 7.7e-04 | norm 1.2798 | dt 0.026 | |
| type train | step 7250 | loss 0.0521 0.7711 4.4473 17.9078 | lr 7.7e-04 | norm 1.2221 | dt 0.025 | |
| type train | step 7260 | loss 0.0521 0.7876 4.5191 18.4920 | lr 7.7e-04 | norm 1.4343 | dt 0.026 | |
| type train | step 7270 | loss 0.0560 0.7797 4.5237 18.3381 | lr 7.7e-04 | norm 1.6432 | dt 0.025 | |
| type train | step 7280 | loss 0.0540 0.7860 4.7176 19.4151 | lr 7.7e-04 | norm 1.3204 | dt 0.026 | |
| type train | step 7290 | loss 0.0545 0.7707 4.4652 17.9263 | lr 7.7e-04 | norm 1.4909 | dt 0.026 | |
| type train | step 7300 | loss 0.0545 0.7812 4.5833 18.8633 | lr 7.7e-04 | norm 1.4118 | dt 0.025 | |
| type train | step 7310 | loss 0.0526 0.7725 4.4702 18.0178 | lr 7.7e-04 | norm 1.2066 | dt 0.026 | |
| type train | step 7320 | loss 0.0532 0.7906 4.7567 19.1048 | lr 7.7e-04 | norm 1.5433 | dt 0.027 | |
| type train | step 7330 | loss 0.0538 0.7819 4.5343 18.4402 | lr 7.6e-04 | norm 1.1703 | dt 0.026 | |
| type train | step 7340 | loss 0.0538 0.7924 4.6208 19.0052 | lr 7.6e-04 | norm 1.3009 | dt 0.026 | |
| type train | step 7350 | loss 0.0541 0.7585 4.4182 17.9888 | lr 7.6e-04 | norm 1.3015 | dt 0.026 | |
| type train | step 7360 | loss 0.0556 0.7987 4.6341 18.7984 | lr 7.6e-04 | norm 1.5386 | dt 0.026 | |
| type train | step 7370 | loss 0.0525 0.7729 4.4670 17.9433 | lr 7.6e-04 | norm 1.2883 | dt 0.025 | |
| type train | step 7380 | loss 0.0521 0.7683 4.5250 18.5657 | lr 7.6e-04 | norm 1.3192 | dt 0.026 | |
| type train | step 7390 | loss 0.0532 0.7709 4.4383 17.9471 | lr 7.6e-04 | norm 1.0718 | dt 0.026 | |
| type train | step 7400 | loss 0.0556 0.8084 4.8155 19.5793 | lr 7.6e-04 | norm 1.3174 | dt 0.026 | |
| type train | step 7410 | loss 0.0525 0.7683 4.4919 18.2081 | lr 7.6e-04 | norm 1.3109 | dt 0.026 | |
| type train | step 7420 | loss 0.0527 0.7462 4.3685 17.9415 | lr 7.6e-04 | norm 1.1927 | dt 0.026 | |
| type train | step 7430 | loss 0.0545 0.7855 4.5577 18.3779 | lr 7.6e-04 | norm 1.2901 | dt 0.026 | |
| type train | step 7440 | loss 0.0547 0.7800 4.4291 18.0230 | lr 7.6e-04 | norm 1.3238 | dt 0.026 | |
| type train | step 7450 | loss 0.0535 0.7680 4.4121 18.1040 | lr 7.6e-04 | norm 1.1775 | dt 0.026 | |
| type train | step 7460 | loss 0.0576 0.7830 4.4816 18.5275 | lr 7.6e-04 | norm 1.8753 | dt 0.026 | |
| type train | step 7470 | loss 0.0546 0.7904 4.5756 18.5812 | lr 7.6e-04 | norm 1.4777 | dt 0.026 | |
| type train | step 7480 | loss 0.0544 0.7686 4.5458 18.5589 | lr 7.5e-04 | norm 1.3288 | dt 0.026 | |
| type train | step 7490 | loss 0.0519 0.7772 4.5280 18.1109 | lr 7.5e-04 | norm 1.3934 | dt 0.026 | |
| type train | step 7500 | loss 0.0537 0.7786 4.4878 18.3367 | lr 7.5e-04 | norm 1.3748 | dt 0.026 | |
| type train | step 7510 | loss 0.0532 0.7663 4.4703 18.2095 | lr 7.5e-04 | norm 1.5943 | dt 0.026 | |
| type train | step 7520 | loss 0.0565 0.7977 4.6624 18.9756 | lr 7.5e-04 | norm 1.3500 | dt 0.025 | |
| type train | step 7530 | loss 0.0548 0.7726 4.4283 18.2698 | lr 7.5e-04 | norm 1.3024 | dt 0.026 | |
| type train | step 7540 | loss 0.0536 0.7605 4.5320 18.4108 | lr 7.5e-04 | norm 1.2648 | dt 0.026 | |
| type train | step 7550 | loss 0.0517 0.7776 4.5109 18.2657 | lr 7.5e-04 | norm 1.2327 | dt 0.026 | |
| type train | step 7560 | loss 0.0544 0.8015 4.6088 18.6948 | lr 7.5e-04 | norm 1.3947 | dt 0.026 | |
| type train | step 7570 | loss 0.0531 0.7610 4.3804 17.8728 | lr 7.5e-04 | norm 1.1782 | dt 0.026 | |
| type train | step 7580 | loss 0.0525 0.7684 4.4830 18.2706 | lr 7.5e-04 | norm 1.2229 | dt 0.026 | |
| type train | step 7590 | loss 0.0545 0.7617 4.3780 18.0424 | lr 7.5e-04 | norm 1.1613 | dt 0.026 | |
| type train | step 7600 | loss 0.0554 0.7775 4.5787 18.6030 | lr 7.5e-04 | norm 1.5007 | dt 0.026 | |
| type train | step 7610 | loss 0.0531 0.7699 4.4118 18.0113 | lr 7.5e-04 | norm 1.3784 | dt 0.026 | |
| type train | step 7620 | loss 0.0547 0.7906 4.5332 18.2395 | lr 7.5e-04 | norm 1.4291 | dt 0.026 | |
| type train | step 7630 | loss 0.0540 0.7724 4.4955 18.3802 | lr 7.4e-04 | norm 1.4364 | dt 0.026 | |
| type train | step 7640 | loss 0.0544 0.7612 4.5007 18.3547 | lr 7.4e-04 | norm 1.4200 | dt 0.026 | |
| type train | step 7650 | loss 0.0545 0.7802 4.5925 18.7897 | lr 7.4e-04 | norm 1.1651 | dt 0.026 | |
| type train | step 7660 | loss 0.0528 0.7711 4.5448 18.5929 | lr 7.4e-04 | norm 1.1705 | dt 0.026 | |
| type train | step 7670 | loss 0.0522 0.7835 4.5342 18.4453 | lr 7.4e-04 | norm 1.4477 | dt 0.026 | |
| type train | step 7680 | loss 0.0530 0.7496 4.3337 17.7891 | lr 7.4e-04 | norm 1.4154 | dt 0.026 | |
| type train | step 7690 | loss 0.0529 0.7736 4.3957 18.2688 | lr 7.4e-04 | norm 1.4400 | dt 0.026 | |
| type train | step 7700 | loss 0.0553 0.7807 4.5241 18.6217 | lr 7.4e-04 | norm 1.4204 | dt 0.026 | |
| type train | step 7710 | loss 0.0531 0.7660 4.5861 18.7361 | lr 7.4e-04 | norm 1.1807 | dt 0.025 | |
| type train | step 7720 | loss 0.0544 0.7697 4.4582 18.0237 | lr 7.4e-04 | norm 1.2670 | dt 0.026 | |
| type train | step 7730 | loss 0.0531 0.7825 4.5163 18.2237 | lr 7.4e-04 | norm 1.2196 | dt 0.025 | |
| type train | step 7740 | loss 0.0532 0.7836 4.4431 18.1264 | lr 7.4e-04 | norm 1.5042 | dt 0.026 | |
| type train | step 7750 | loss 0.0534 0.7827 4.5652 18.5449 | lr 7.4e-04 | norm 1.2024 | dt 0.026 | |
| type train | step 7760 | loss 0.0534 0.7663 4.4019 18.1087 | lr 7.4e-04 | norm 1.4296 | dt 0.026 | |
| type train | step 7770 | loss 0.0547 0.7866 4.5683 18.4799 | lr 7.4e-04 | norm 1.1691 | dt 0.026 | |
| type train | step 7780 | loss 0.0524 0.7519 4.2960 17.5533 | lr 7.3e-04 | norm 1.2918 | dt 0.026 | |
| type train | step 7790 | loss 0.0546 0.7663 4.4831 18.3623 | lr 7.3e-04 | norm 1.2259 | dt 0.026 | |
| type train | step 7800 | loss 0.0529 0.7782 4.5533 18.5577 | lr 7.3e-04 | norm 1.2429 | dt 0.026 | |
| type train | step 7810 | loss 0.0542 0.7867 4.6277 18.6493 | lr 7.3e-04 | norm 1.3473 | dt 0.026 | |
| type train | step 7820 | loss 0.0553 0.7767 4.5014 18.4603 | lr 7.3e-04 | norm 1.4131 | dt 0.025 | |
| type train | step 7830 | loss 0.0544 0.7817 4.6576 19.0815 | lr 7.3e-04 | norm 1.2762 | dt 0.025 | |
| type train | step 7840 | loss 0.0542 0.7757 4.5956 18.8744 | lr 7.3e-04 | norm 1.4006 | dt 0.026 | |
| type train | step 7850 | loss 0.0543 0.7854 4.5883 19.0090 | lr 7.3e-04 | norm 1.2650 | dt 0.026 | |
| type train | step 7860 | loss 0.0509 0.7666 4.4254 17.8941 | lr 7.3e-04 | norm 1.1929 | dt 0.026 | |
| type train | step 7870 | loss 0.0510 0.7845 4.4941 18.4805 | lr 7.3e-04 | norm 1.4231 | dt 0.026 | |
| type train | step 7880 | loss 0.0549 0.7754 4.4980 18.3053 | lr 7.3e-04 | norm 1.6251 | dt 0.026 | |
| type train | step 7890 | loss 0.0529 0.7826 4.7045 19.3598 | lr 7.3e-04 | norm 1.2985 | dt 0.026 | |
| type train | step 7900 | loss 0.0533 0.7677 4.4492 17.9204 | lr 7.3e-04 | norm 1.4938 | dt 0.026 | |
| type train | step 7910 | loss 0.0534 0.7781 4.5573 18.8158 | lr 7.3e-04 | norm 1.3926 | dt 0.026 | |
| type train | step 7920 | loss 0.0516 0.7689 4.4515 18.0049 | lr 7.3e-04 | norm 1.1920 | dt 0.026 | |
| type train | step 7930 | loss 0.0518 0.7865 4.7362 19.1031 | lr 7.2e-04 | norm 1.5261 | dt 0.027 | |
| type train | step 7940 | loss 0.0524 0.7774 4.5171 18.4116 | lr 7.2e-04 | norm 1.1533 | dt 0.026 | |
| type train | step 7950 | loss 0.0526 0.7895 4.6062 18.9614 | lr 7.2e-04 | norm 1.2826 | dt 0.026 | |
| type train | step 7960 | loss 0.0530 0.7550 4.4019 17.9673 | lr 7.2e-04 | norm 1.2836 | dt 0.026 | |
| type train | step 7970 | loss 0.0544 0.7950 4.6106 18.7623 | lr 7.2e-04 | norm 1.5341 | dt 0.026 | |
| type train | step 7980 | loss 0.0513 0.7702 4.4450 17.9044 | lr 7.2e-04 | norm 1.2661 | dt 0.026 | |
| type train | step 7990 | loss 0.0513 0.7655 4.5041 18.5462 | lr 7.2e-04 | norm 1.3115 | dt 0.026 | |
| type train | step 8000 | loss 0.0522 0.7674 4.4194 17.9190 | lr 7.2e-04 | norm 1.0591 | dt 0.026 | |
| type train | step 8010 | loss 0.0544 0.8078 4.7965 19.5438 | lr 7.2e-04 | norm 1.3022 | dt 0.028 | |
| type train | step 8020 | loss 0.0516 0.7655 4.4754 18.2137 | lr 7.2e-04 | norm 1.3036 | dt 0.027 | |
| type train | step 8030 | loss 0.0516 0.7424 4.3496 17.9247 | lr 7.2e-04 | norm 1.1795 | dt 0.026 | |
| type train | step 8040 | loss 0.0533 0.7835 4.5398 18.3574 | lr 7.2e-04 | norm 1.2692 | dt 0.026 | |
| type train | step 8050 | loss 0.0536 0.7771 4.4031 17.9839 | lr 7.2e-04 | norm 1.3102 | dt 0.026 | |
| type train | step 8060 | loss 0.0521 0.7650 4.3938 18.0929 | lr 7.2e-04 | norm 1.1607 | dt 0.026 | |
| type train | step 8070 | loss 0.0565 0.7826 4.4603 18.4818 | lr 7.2e-04 | norm 1.8725 | dt 0.026 | |
| type train | step 8080 | loss 0.0537 0.7879 4.5512 18.5702 | lr 7.1e-04 | norm 1.4596 | dt 0.026 | |
| type train | step 8090 | loss 0.0531 0.7664 4.5261 18.5359 | lr 7.1e-04 | norm 1.3095 | dt 0.026 | |
| type train | step 8100 | loss 0.0510 0.7766 4.5060 18.0853 | lr 7.1e-04 | norm 1.4006 | dt 0.026 | |
| type train | step 8110 | loss 0.0525 0.7744 4.4665 18.3255 | lr 7.1e-04 | norm 1.3715 | dt 0.026 | |
| type train | step 8120 | loss 0.0521 0.7629 4.4493 18.1949 | lr 7.1e-04 | norm 1.5885 | dt 0.026 | |
| type train | step 8130 | loss 0.0552 0.7958 4.6428 18.9327 | lr 7.1e-04 | norm 1.3400 | dt 0.026 | |
| type train | step 8140 | loss 0.0536 0.7696 4.4123 18.2532 | lr 7.1e-04 | norm 1.2845 | dt 0.026 | |
| type train | step 8150 | loss 0.0526 0.7579 4.5161 18.3693 | lr 7.1e-04 | norm 1.2623 | dt 0.026 | |
| type train | step 8160 | loss 0.0505 0.7742 4.4893 18.2308 | lr 7.1e-04 | norm 1.2334 | dt 0.026 | |
| type train | step 8170 | loss 0.0531 0.8005 4.5801 18.6751 | lr 7.1e-04 | norm 1.3781 | dt 0.026 | |
| type train | step 8180 | loss 0.0519 0.7579 4.3574 17.8640 | lr 7.1e-04 | norm 1.1613 | dt 0.026 | |
| type train | step 8190 | loss 0.0516 0.7665 4.4564 18.2384 | lr 7.1e-04 | norm 1.2274 | dt 0.026 | |
| type train | step 8200 | loss 0.0535 0.7584 4.3614 18.0159 | lr 7.1e-04 | norm 1.1640 | dt 0.025 | |
| type train | step 8210 | loss 0.0544 0.7737 4.5613 18.5797 | lr 7.1e-04 | norm 1.4937 | dt 0.025 | |
| type train | step 8220 | loss 0.0519 0.7671 4.3876 17.9921 | lr 7.1e-04 | norm 1.3584 | dt 0.026 | |
| type train | step 8230 | loss 0.0534 0.7885 4.5081 18.2003 | lr 7.0e-04 | norm 1.4120 | dt 0.026 | |
| type train | step 8240 | loss 0.0528 0.7702 4.4708 18.3556 | lr 7.0e-04 | norm 1.4152 | dt 0.026 | |
| type train | step 8250 | loss 0.0534 0.7595 4.4736 18.3206 | lr 7.0e-04 | norm 1.4109 | dt 0.025 | |
| type train | step 8260 | loss 0.0535 0.7772 4.5741 18.7767 | lr 7.0e-04 | norm 1.1496 | dt 0.025 | |
| type train | step 8270 | loss 0.0516 0.7699 4.5284 18.5886 | lr 7.0e-04 | norm 1.1665 | dt 0.026 | |
| type train | step 8280 | loss 0.0511 0.7811 4.5109 18.4123 | lr 7.0e-04 | norm 1.4438 | dt 0.026 | |
| type train | step 8290 | loss 0.0517 0.7481 4.3120 17.7620 | lr 7.0e-04 | norm 1.4130 | dt 0.026 | |
| type train | step 8300 | loss 0.0516 0.7717 4.3743 18.2351 | lr 7.0e-04 | norm 1.4236 | dt 0.026 | |
| type train | step 8310 | loss 0.0543 0.7768 4.4965 18.6033 | lr 7.0e-04 | norm 1.4231 | dt 0.026 | |
| type train | step 8320 | loss 0.0520 0.7642 4.5727 18.7169 | lr 7.0e-04 | norm 1.1812 | dt 0.026 | |
| type train | step 8330 | loss 0.0533 0.7679 4.4368 18.0086 | lr 7.0e-04 | norm 1.2578 | dt 0.026 | |
| type train | step 8340 | loss 0.0519 0.7815 4.4916 18.2091 | lr 7.0e-04 | norm 1.2079 | dt 0.026 | |
| type train | step 8350 | loss 0.0520 0.7818 4.4198 18.1073 | lr 7.0e-04 | norm 1.4863 | dt 0.025 | |
| type train | step 8360 | loss 0.0521 0.7793 4.5447 18.5026 | lr 7.0e-04 | norm 1.1803 | dt 0.026 | |
| type train | step 8370 | loss 0.0523 0.7647 4.3848 18.0701 | lr 6.9e-04 | norm 1.4225 | dt 0.026 | |
| type train | step 8380 | loss 0.0537 0.7837 4.5528 18.4653 | lr 6.9e-04 | norm 1.1692 | dt 0.026 | |
| type train | step 8390 | loss 0.0514 0.7502 4.2804 17.5371 | lr 6.9e-04 | norm 1.2860 | dt 0.026 | |
| type train | step 8400 | loss 0.0535 0.7642 4.4619 18.3330 | lr 6.9e-04 | norm 1.2157 | dt 0.026 | |
| type train | step 8410 | loss 0.0519 0.7763 4.5346 18.5125 | lr 6.9e-04 | norm 1.2455 | dt 0.026 | |
| type train | step 8420 | loss 0.0522 0.7843 4.6072 18.6032 | lr 6.9e-04 | norm 1.3109 | dt 0.026 | |
| type train | step 8430 | loss 0.0542 0.7745 4.4784 18.4284 | lr 6.9e-04 | norm 1.4124 | dt 0.026 | |
| type train | step 8440 | loss 0.0528 0.7788 4.6464 19.0697 | lr 6.9e-04 | norm 1.2502 | dt 0.026 | |
| type train | step 8450 | loss 0.0528 0.7740 4.5753 18.8498 | lr 6.9e-04 | norm 1.4237 | dt 0.026 | |
| type train | step 8460 | loss 0.0532 0.7819 4.5652 18.9769 | lr 6.9e-04 | norm 1.2640 | dt 0.025 | |
| type train | step 8470 | loss 0.0498 0.7634 4.4050 17.8544 | lr 6.9e-04 | norm 1.1880 | dt 0.026 | |
| type train | step 8480 | loss 0.0500 0.7817 4.4764 18.4662 | lr 6.9e-04 | norm 1.4082 | dt 0.026 | |
| type train | step 8490 | loss 0.0537 0.7727 4.4721 18.2908 | lr 6.9e-04 | norm 1.6019 | dt 0.026 | |
| type train | step 8500 | loss 0.0518 0.7797 4.6802 19.3446 | lr 6.9e-04 | norm 1.2825 | dt 0.025 | |
| type train | step 8510 | loss 0.0522 0.7659 4.4331 17.8969 | lr 6.8e-04 | norm 1.4917 | dt 0.026 | |
| type train | step 8520 | loss 0.0522 0.7768 4.5319 18.7835 | lr 6.8e-04 | norm 1.4071 | dt 0.026 | |
| type train | step 8530 | loss 0.0505 0.7667 4.4338 17.9752 | lr 6.8e-04 | norm 1.1810 | dt 0.026 | |
| type train | step 8540 | loss 0.0508 0.7849 4.7151 19.0797 | lr 6.8e-04 | norm 1.5173 | dt 0.031 | |
| type train | step 8550 | loss 0.0512 0.7747 4.5032 18.3951 | lr 6.8e-04 | norm 1.1514 | dt 0.026 | |
| type train | step 8560 | loss 0.0516 0.7855 4.5880 18.9422 | lr 6.8e-04 | norm 1.2575 | dt 0.025 | |
| type train | step 8570 | loss 0.0520 0.7536 4.3842 17.9311 | lr 6.8e-04 | norm 1.2871 | dt 0.026 | |
| type train | step 8580 | loss 0.0533 0.7921 4.5917 18.7404 | lr 6.8e-04 | norm 1.5295 | dt 0.026 | |
| type train | step 8590 | loss 0.0502 0.7677 4.4238 17.8708 | lr 6.8e-04 | norm 1.2629 | dt 0.026 | |
| type train | step 8600 | loss 0.0501 0.7621 4.4873 18.5193 | lr 6.8e-04 | norm 1.2806 | dt 0.026 | |
| type train | step 8610 | loss 0.0509 0.7654 4.4055 17.8886 | lr 6.8e-04 | norm 1.0372 | dt 0.026 | |
| type train | step 8620 | loss 0.0535 0.8070 4.7772 19.5147 | lr 6.8e-04 | norm 1.3140 | dt 0.025 | |
| type train | step 8630 | loss 0.0506 0.7652 4.4573 18.1904 | lr 6.8e-04 | norm 1.3105 | dt 0.026 | |
| type train | step 8640 | loss 0.0505 0.7395 4.3347 17.9071 | lr 6.8e-04 | norm 1.1762 | dt 0.026 | |
| type train | step 8650 | loss 0.0523 0.7818 4.5224 18.3213 | lr 6.8e-04 | norm 1.2567 | dt 0.026 | |
| type train | step 8660 | loss 0.0527 0.7774 4.3845 17.9672 | lr 6.7e-04 | norm 1.3021 | dt 0.026 | |
| type train | step 8670 | loss 0.0511 0.7638 4.3744 18.0654 | lr 6.7e-04 | norm 1.1588 | dt 0.026 | |
| type train | step 8680 | loss 0.0553 0.7829 4.4430 18.4694 | lr 6.7e-04 | norm 1.8777 | dt 0.026 | |
| type train | step 8690 | loss 0.0525 0.7865 4.5352 18.5437 | lr 6.7e-04 | norm 1.4645 | dt 0.026 | |
| type train | step 8700 | loss 0.0519 0.7629 4.5074 18.4863 | lr 6.7e-04 | norm 1.3092 | dt 0.026 | |
| type train | step 8710 | loss 0.0497 0.7747 4.4837 18.0480 | lr 6.7e-04 | norm 1.3741 | dt 0.026 | |
| type train | step 8720 | loss 0.0513 0.7727 4.4448 18.2904 | lr 6.7e-04 | norm 1.3576 | dt 0.026 | |
| type train | step 8730 | loss 0.0511 0.7628 4.4280 18.1795 | lr 6.7e-04 | norm 1.5811 | dt 0.026 | |
| type train | step 8740 | loss 0.0539 0.7947 4.6261 18.9224 | lr 6.7e-04 | norm 1.3397 | dt 0.025 | |
| type train | step 8750 | loss 0.0525 0.7681 4.3998 18.2315 | lr 6.7e-04 | norm 1.2668 | dt 0.026 | |
| type train | step 8760 | loss 0.0515 0.7561 4.4995 18.3359 | lr 6.7e-04 | norm 1.2489 | dt 0.026 | |
| type train | step 8770 | loss 0.0495 0.7744 4.4724 18.1979 | lr 6.7e-04 | norm 1.2276 | dt 0.026 | |
| type train | step 8780 | loss 0.0522 0.7995 4.5680 18.6511 | lr 6.7e-04 | norm 1.3813 | dt 0.026 | |
| type train | step 8790 | loss 0.0507 0.7561 4.3412 17.8279 | lr 6.7e-04 | norm 1.1561 | dt 0.025 | |
| type train | step 8800 | loss 0.0506 0.7632 4.4419 18.2159 | lr 6.6e-04 | norm 1.2036 | dt 0.026 | |
| type train | step 8810 | loss 0.0524 0.7564 4.3452 17.9882 | lr 6.6e-04 | norm 1.1420 | dt 0.026 | |
| type train | step 8820 | loss 0.0530 0.7730 4.5449 18.5268 | lr 6.6e-04 | norm 1.4822 | dt 0.027 | |
| type train | step 8830 | loss 0.0506 0.7664 4.3669 17.9756 | lr 6.6e-04 | norm 1.3468 | dt 0.026 | |
| type train | step 8840 | loss 0.0522 0.7891 4.4931 18.1628 | lr 6.6e-04 | norm 1.4110 | dt 0.026 | |
| type train | step 8850 | loss 0.0518 0.7686 4.4494 18.3332 | lr 6.6e-04 | norm 1.4409 | dt 0.026 | |
| type train | step 8860 | loss 0.0522 0.7570 4.4550 18.2839 | lr 6.6e-04 | norm 1.4029 | dt 0.026 | |
| type train | step 8870 | loss 0.0524 0.7746 4.5581 18.7490 | lr 6.6e-04 | norm 1.1429 | dt 0.026 | |
| type train | step 8880 | loss 0.0504 0.7709 4.5127 18.5609 | lr 6.6e-04 | norm 1.1568 | dt 0.025 | |
| type train | step 8890 | loss 0.0500 0.7807 4.4890 18.3613 | lr 6.6e-04 | norm 1.4093 | dt 0.026 | |
| type train | step 8900 | loss 0.0506 0.7481 4.2967 17.7245 | lr 6.6e-04 | norm 1.4002 | dt 0.026 | |
| type train | step 8910 | loss 0.0507 0.7704 4.3625 18.2221 | lr 6.6e-04 | norm 1.4257 | dt 0.026 | |
| type train | step 8920 | loss 0.0532 0.7750 4.4780 18.5760 | lr 6.6e-04 | norm 1.3919 | dt 0.026 | |
| type train | step 8930 | loss 0.0508 0.7625 4.5557 18.6898 | lr 6.6e-04 | norm 1.1868 | dt 0.026 | |
| type train | step 8940 | loss 0.0521 0.7687 4.4243 17.9776 | lr 6.5e-04 | norm 1.2466 | dt 0.026 | |
| type train | step 8950 | loss 0.0506 0.7785 4.4710 18.1721 | lr 6.5e-04 | norm 1.1949 | dt 0.026 | |
| type train | step 8960 | loss 0.0509 0.7832 4.4000 18.0923 | lr 6.5e-04 | norm 1.4864 | dt 0.025 | |
| type train | step 8970 | loss 0.0507 0.7786 4.5303 18.4906 | lr 6.5e-04 | norm 1.1825 | dt 0.025 | |
| type train | step 8980 | loss 0.0510 0.7639 4.3681 18.0361 | lr 6.5e-04 | norm 1.4155 | dt 0.026 | |
| type train | step 8990 | loss 0.0527 0.7843 4.5385 18.4433 | lr 6.5e-04 | norm 1.1674 | dt 0.026 | |
| type train | step 9000 | loss 0.0502 0.7508 4.2640 17.5222 | lr 6.5e-04 | norm 1.2835 | dt 0.026 | |
| type train | step 9010 | loss 0.0525 0.7630 4.4433 18.3105 | lr 6.5e-04 | norm 1.2199 | dt 0.026 | |
| type train | step 9020 | loss 0.0506 0.7761 4.5172 18.4728 | lr 6.5e-04 | norm 1.2114 | dt 0.026 | |
| type train | step 9030 | loss 0.0510 0.7815 4.5864 18.5700 | lr 6.5e-04 | norm 1.3100 | dt 0.027 | |
| type train | step 9040 | loss 0.0532 0.7730 4.4618 18.3925 | lr 6.5e-04 | norm 1.3983 | dt 0.026 | |
| type train | step 9050 | loss 0.0514 0.7774 4.6298 19.0562 | lr 6.5e-04 | norm 1.2411 | dt 0.026 | |
| type train | step 9060 | loss 0.0517 0.7738 4.5595 18.7951 | lr 6.5e-04 | norm 1.4010 | dt 0.026 | |
| type train | step 9070 | loss 0.0519 0.7793 4.5478 18.9556 | lr 6.5e-04 | norm 1.2596 | dt 0.026 | |
| type train | step 9080 | loss 0.0490 0.7651 4.3931 17.8218 | lr 6.4e-04 | norm 1.1983 | dt 0.026 | |
| type train | step 9090 | loss 0.0490 0.7827 4.4541 18.4397 | lr 6.4e-04 | norm 1.4077 | dt 0.026 | |
| type train | step 9100 | loss 0.0524 0.7706 4.4487 18.2760 | lr 6.4e-04 | norm 1.5977 | dt 0.026 | |
| type train | step 9110 | loss 0.0507 0.7767 4.6656 19.3194 | lr 6.4e-04 | norm 1.2865 | dt 0.026 | |
| type train | step 9120 | loss 0.0511 0.7656 4.4192 17.8811 | lr 6.4e-04 | norm 1.4911 | dt 0.026 | |
| type train | step 9130 | loss 0.0511 0.7743 4.5149 18.7513 | lr 6.4e-04 | norm 1.3903 | dt 0.026 | |
| type train | step 9140 | loss 0.0494 0.7654 4.4174 17.9620 | lr 6.4e-04 | norm 1.1866 | dt 0.026 | |
| type train | step 9150 | loss 0.0496 0.7799 4.7005 19.0691 | lr 6.4e-04 | norm 1.5070 | dt 0.027 | |
| type train | step 9160 | loss 0.0501 0.7718 4.4865 18.3705 | lr 6.4e-04 | norm 1.1385 | dt 0.025 | |
| type train | step 9170 | loss 0.0505 0.7826 4.5733 18.9184 | lr 6.4e-04 | norm 1.2481 | dt 0.026 | |
| type train | step 9180 | loss 0.0509 0.7512 4.3736 17.8998 | lr 6.4e-04 | norm 1.2904 | dt 0.026 | |
| type train | step 9190 | loss 0.0521 0.7914 4.5755 18.7153 | lr 6.4e-04 | norm 1.5294 | dt 0.026 | |
| type train | step 9200 | loss 0.0491 0.7668 4.4063 17.8372 | lr 6.4e-04 | norm 1.2500 | dt 0.026 | |
| type train | step 9210 | loss 0.0494 0.7595 4.4721 18.5048 | lr 6.4e-04 | norm 1.2741 | dt 0.025 | |
| type train | step 9220 | loss 0.0497 0.7631 4.3886 17.8666 | lr 6.3e-04 | norm 1.0277 | dt 0.026 | |
| type train | step 9230 | loss 0.0523 0.8025 4.7579 19.4938 | lr 6.3e-04 | norm 1.2951 | dt 0.026 | |
| type train | step 9240 | loss 0.0496 0.7634 4.4429 18.1812 | lr 6.3e-04 | norm 1.3194 | dt 0.026 | |
| type train | step 9250 | loss 0.0494 0.7363 4.3251 17.8825 | lr 6.3e-04 | norm 1.1571 | dt 0.026 | |
| type train | step 9260 | loss 0.0511 0.7806 4.5038 18.2944 | lr 6.3e-04 | norm 1.2579 | dt 0.026 | |
| type train | step 9270 | loss 0.0514 0.7756 4.3671 17.9345 | lr 6.3e-04 | norm 1.2833 | dt 0.026 | |
| type train | step 9280 | loss 0.0500 0.7618 4.3562 18.0526 | lr 6.3e-04 | norm 1.1376 | dt 0.026 | |
| type train | step 9290 | loss 0.0541 0.7801 4.4228 18.4502 | lr 6.3e-04 | norm 1.8861 | dt 0.026 | |
| type train | step 9300 | loss 0.0513 0.7848 4.5226 18.5221 | lr 6.3e-04 | norm 1.4510 | dt 0.026 | |
| type train | step 9310 | loss 0.0508 0.7578 4.4944 18.4661 | lr 6.3e-04 | norm 1.3049 | dt 0.026 | |
| type train | step 9320 | loss 0.0484 0.7725 4.4667 18.0318 | lr 6.3e-04 | norm 1.3616 | dt 0.026 | |
| type train | step 9330 | loss 0.0503 0.7724 4.4270 18.2757 | lr 6.3e-04 | norm 1.3640 | dt 0.026 | |
| type train | step 9340 | loss 0.0500 0.7590 4.4102 18.1663 | lr 6.3e-04 | norm 1.5800 | dt 0.026 | |
| type train | step 9350 | loss 0.0527 0.7892 4.6098 18.8984 | lr 6.2e-04 | norm 1.3271 | dt 0.026 | |
| type train | step 9360 | loss 0.0515 0.7618 4.3920 18.2224 | lr 6.2e-04 | norm 1.2785 | dt 0.026 | |
| type train | step 9370 | loss 0.0501 0.7522 4.4919 18.3039 | lr 6.2e-04 | norm 1.2459 | dt 0.026 | |
| type train | step 9380 | loss 0.0484 0.7711 4.4546 18.1665 | lr 6.2e-04 | norm 1.2105 | dt 0.026 | |
| type train | step 9390 | loss 0.0511 0.7970 4.5538 18.6261 | lr 6.2e-04 | norm 1.3867 | dt 0.026 | |
| type train | step 9400 | loss 0.0497 0.7535 4.3268 17.8065 | lr 6.2e-04 | norm 1.1514 | dt 0.026 | |
| type train | step 9410 | loss 0.0489 0.7587 4.4249 18.1991 | lr 6.2e-04 | norm 1.1884 | dt 0.026 | |
| type train | step 9420 | loss 0.0512 0.7522 4.3367 17.9894 | lr 6.2e-04 | norm 1.1467 | dt 0.026 | |
| type train | step 9430 | loss 0.0522 0.7689 4.5343 18.4777 | lr 6.2e-04 | norm 1.4854 | dt 0.026 | |
| type train | step 9440 | loss 0.0496 0.7617 4.3480 17.9549 | lr 6.2e-04 | norm 1.3326 | dt 0.026 | |
| type train | step 9450 | loss 0.0510 0.7864 4.4758 18.1428 | lr 6.2e-04 | norm 1.4085 | dt 0.026 | |
| type train | step 9460 | loss 0.0507 0.7643 4.4287 18.2978 | lr 6.2e-04 | norm 1.4056 | dt 0.025 | |
| type train | step 9470 | loss 0.0510 0.7512 4.4385 18.2642 | lr 6.2e-04 | norm 1.3874 | dt 0.026 | |
| type train | step 9480 | loss 0.0512 0.7671 4.5457 18.7290 | lr 6.2e-04 | norm 1.1241 | dt 0.026 | |
| type train | step 9490 | loss 0.0495 0.7662 4.4989 18.5439 | lr 6.1e-04 | norm 1.1411 | dt 0.026 | |
| type train | step 9500 | loss 0.0490 0.7764 4.4733 18.3247 | lr 6.1e-04 | norm 1.4077 | dt 0.026 | |
| type train | step 9510 | loss 0.0494 0.7455 4.2863 17.6992 | lr 6.1e-04 | norm 1.3850 | dt 0.026 | |
| type train | step 9520 | loss 0.0497 0.7648 4.3421 18.1875 | lr 6.1e-04 | norm 1.4011 | dt 0.026 | |
| type train | step 9530 | loss 0.0520 0.7681 4.4631 18.5529 | lr 6.1e-04 | norm 1.3894 | dt 0.026 | |
| type train | step 9540 | loss 0.0497 0.7523 4.5468 18.6838 | lr 6.1e-04 | norm 1.1714 | dt 0.026 | |
| type train | step 9550 | loss 0.0511 0.7648 4.4121 17.9524 | lr 6.1e-04 | norm 1.2547 | dt 0.026 | |
| type train | step 9560 | loss 0.0499 0.7717 4.4567 18.1543 | lr 6.1e-04 | norm 1.1836 | dt 0.026 | |
| type train | step 9570 | loss 0.0499 0.7802 4.3843 18.0698 | lr 6.1e-04 | norm 1.4671 | dt 0.026 | |
| type train | step 9580 | loss 0.0497 0.7729 4.5145 18.4779 | lr 6.1e-04 | norm 1.1770 | dt 0.025 | |
| type train | step 9590 | loss 0.0500 0.7561 4.3529 18.0091 | lr 6.1e-04 | norm 1.4184 | dt 0.026 | |
| type train | step 9600 | loss 0.0516 0.7760 4.5242 18.4332 | lr 6.1e-04 | norm 1.1422 | dt 0.026 | |
| type train | step 9610 | loss 0.0494 0.7485 4.2444 17.5096 | lr 6.1e-04 | norm 1.2876 | dt 0.026 | |
| type train | step 9620 | loss 0.0516 0.7555 4.4261 18.2924 | lr 6.1e-04 | norm 1.2037 | dt 0.026 | |
| type train | step 9630 | loss 0.0499 0.7701 4.5048 18.4522 | lr 6.0e-04 | norm 1.2023 | dt 0.026 | |
| type train | step 9640 | loss 0.0501 0.7759 4.5705 18.5506 | lr 6.0e-04 | norm 1.2994 | dt 0.026 | |
| type train | step 9650 | loss 0.0521 0.7667 4.4522 18.3693 | lr 6.0e-04 | norm 1.3940 | dt 0.025 | |
| type train | step 9660 | loss 0.0504 0.7709 4.6097 19.0374 | lr 6.0e-04 | norm 1.2285 | dt 0.026 | |
| type train | step 9670 | loss 0.0506 0.7687 4.5448 18.7621 | lr 6.0e-04 | norm 1.3889 | dt 0.026 | |
| type train | step 9680 | loss 0.0511 0.7704 4.5307 18.9302 | lr 6.0e-04 | norm 1.2563 | dt 0.026 | |
| type train | step 9690 | loss 0.0480 0.7584 4.3795 17.8000 | lr 6.0e-04 | norm 1.1784 | dt 0.026 | |
| type train | step 9700 | loss 0.0480 0.7749 4.4370 18.4275 | lr 6.0e-04 | norm 1.3722 | dt 0.025 | |
| type train | step 9710 | loss 0.0516 0.7645 4.4279 18.2371 | lr 6.0e-04 | norm 1.5950 | dt 0.026 | |
| type train | step 9720 | loss 0.0498 0.7675 4.6510 19.3139 | lr 6.0e-04 | norm 1.2731 | dt 0.026 | |
| type train | step 9730 | loss 0.0502 0.7625 4.4048 17.8607 | lr 6.0e-04 | norm 1.4964 | dt 0.026 | |
| type train | step 9740 | loss 0.0499 0.7673 4.5013 18.7229 | lr 6.0e-04 | norm 1.3805 | dt 0.026 | |
| type train | step 9750 | loss 0.0487 0.7603 4.4045 17.9324 | lr 6.0e-04 | norm 1.1874 | dt 0.026 | |
| type train | step 9760 | loss 0.0488 0.7751 4.6853 19.0482 | lr 6.0e-04 | norm 1.4876 | dt 0.027 | |
| type train | step 9770 | loss 0.0492 0.7637 4.4732 18.3426 | lr 5.9e-04 | norm 1.1285 | dt 0.026 | |
| type train | step 9780 | loss 0.0497 0.7743 4.5619 18.9064 | lr 5.9e-04 | norm 1.2421 | dt 0.026 | |
| type train | step 9790 | loss 0.0500 0.7452 4.3535 17.8826 | lr 5.9e-04 | norm 1.2822 | dt 0.026 | |
| type train | step 9800 | loss 0.0511 0.7860 4.5571 18.6861 | lr 5.9e-04 | norm 1.5161 | dt 0.026 | |
| type train | step 9810 | loss 0.0482 0.7613 4.3974 17.7931 | lr 5.9e-04 | norm 1.2549 | dt 0.026 | |
| type train | step 9820 | loss 0.0483 0.7528 4.4553 18.4738 | lr 5.9e-04 | norm 1.2468 | dt 0.026 | |
| type train | step 9830 | loss 0.0490 0.7576 4.3766 17.8466 | lr 5.9e-04 | norm 1.0318 | dt 0.026 | |
| type train | step 9840 | loss 0.0518 0.7965 4.7384 19.4606 | lr 5.9e-04 | norm 1.3093 | dt 0.026 | |
| type train | step 9850 | loss 0.0488 0.7594 4.4269 18.1620 | lr 5.9e-04 | norm 1.3189 | dt 0.026 | |
| type train | step 9860 | loss 0.0484 0.7332 4.3143 17.8649 | lr 5.9e-04 | norm 1.1417 | dt 0.026 | |
| type train | step 9870 | loss 0.0503 0.7757 4.4861 18.2695 | lr 5.9e-04 | norm 1.2667 | dt 0.026 | |
| type train | step 9880 | loss 0.0507 0.7710 4.3491 17.9203 | lr 5.9e-04 | norm 1.2872 | dt 0.025 | |
| type train | step 9890 | loss 0.0491 0.7572 4.3383 18.0293 | lr 5.9e-04 | norm 1.1523 | dt 0.026 | |
| type train | step 9900 | loss 0.0532 0.7735 4.4079 18.4209 | lr 5.8e-04 | norm 1.8871 | dt 0.026 | |
| type train | step 9910 | loss 0.0503 0.7810 4.5057 18.5116 | lr 5.8e-04 | norm 1.4444 | dt 0.026 | |
| type train | step 9920 | loss 0.0499 0.7521 4.4790 18.4352 | lr 5.8e-04 | norm 1.2882 | dt 0.025 | |
| type train | step 9930 | loss 0.0476 0.7700 4.4524 18.0045 | lr 5.8e-04 | norm 1.3612 | dt 0.026 | |
| type train | step 9940 | loss 0.0496 0.7697 4.4097 18.2448 | lr 5.8e-04 | norm 1.3598 | dt 0.026 | |
| type train | step 9950 | loss 0.0494 0.7554 4.3969 18.1444 | lr 5.8e-04 | norm 1.5727 | dt 0.026 | |
| type train | step 9960 | loss 0.0522 0.7843 4.5943 18.8637 | lr 5.8e-04 | norm 1.3397 | dt 0.026 | |
| type train | step 9970 | loss 0.0506 0.7581 4.3793 18.2008 | lr 5.8e-04 | norm 1.2688 | dt 0.026 | |
| type train | step 9980 | loss 0.0493 0.7465 4.4758 18.2960 | lr 5.8e-04 | norm 1.2312 | dt 0.025 | |
| type train | step 9990 | loss 0.0475 0.7658 4.4428 18.1249 | lr 5.8e-04 | norm 1.1948 | dt 0.025 | |
| type train | step 10000 | loss 0.0503 0.7919 4.5393 18.6130 | lr 5.8e-04 | norm 1.3873 | dt 0.026 | |
| type train | step 10010 | loss 0.0491 0.7488 4.3143 17.7718 | lr 5.8e-04 | norm 1.1449 | dt 0.026 | |
| type train | step 10020 | loss 0.0484 0.7532 4.4063 18.1735 | lr 5.8e-04 | norm 1.1922 | dt 0.026 | |
| type train | step 10030 | loss 0.0505 0.7481 4.3229 17.9701 | lr 5.8e-04 | norm 1.1212 | dt 0.026 | |
| type train | step 10040 | loss 0.0511 0.7642 4.5166 18.4524 | lr 5.7e-04 | norm 1.4629 | dt 0.026 | |
| type train | step 10050 | loss 0.0487 0.7572 4.3315 17.9200 | lr 5.7e-04 | norm 1.3339 | dt 0.026 | |
| type train | step 10060 | loss 0.0502 0.7814 4.4600 18.1262 | lr 5.7e-04 | norm 1.3985 | dt 0.026 | |
| type train | step 10070 | loss 0.0502 0.7610 4.4124 18.2680 | lr 5.7e-04 | norm 1.4115 | dt 0.026 | |
| type train | step 10080 | loss 0.0499 0.7473 4.4269 18.2382 | lr 5.7e-04 | norm 1.3882 | dt 0.026 | |
| type train | step 10090 | loss 0.0506 0.7643 4.5273 18.7031 | lr 5.7e-04 | norm 1.1413 | dt 0.026 | |
| type train | step 10100 | loss 0.0486 0.7622 4.4855 18.5200 | lr 5.7e-04 | norm 1.1423 | dt 0.026 | |
| type train | step 10110 | loss 0.0483 0.7715 4.4572 18.2928 | lr 5.7e-04 | norm 1.3953 | dt 0.026 | |
| type train | step 10120 | loss 0.0485 0.7415 4.2725 17.6772 | lr 5.7e-04 | norm 1.3706 | dt 0.026 | |
| type train | step 10130 | loss 0.0489 0.7610 4.3254 18.1684 | lr 5.7e-04 | norm 1.3953 | dt 0.026 | |
| type train | step 10140 | loss 0.0511 0.7637 4.4497 18.5071 | lr 5.7e-04 | norm 1.3901 | dt 0.026 | |
| type train | step 10150 | loss 0.0488 0.7484 4.5300 18.6542 | lr 5.7e-04 | norm 1.1662 | dt 0.026 | |
| type train | step 10160 | loss 0.0504 0.7622 4.3937 17.9304 | lr 5.7e-04 | norm 1.2598 | dt 0.026 | |
| type train | step 10170 | loss 0.0489 0.7688 4.4430 18.1234 | lr 5.7e-04 | norm 1.1760 | dt 0.026 | |
| type train | step 10180 | loss 0.0489 0.7779 4.3710 18.0458 | lr 5.6e-04 | norm 1.4414 | dt 0.026 | |
| type train | step 10190 | loss 0.0489 0.7688 4.5022 18.4560 | lr 5.6e-04 | norm 1.1694 | dt 0.027 | |
| type train | step 10200 | loss 0.0493 0.7526 4.3377 17.9984 | lr 5.6e-04 | norm 1.4162 | dt 0.026 | |
| type train | step 10210 | loss 0.0507 0.7735 4.5026 18.4105 | lr 5.6e-04 | norm 1.1386 | dt 0.026 | |
| type train | step 10220 | loss 0.0484 0.7443 4.2305 17.4920 | lr 5.6e-04 | norm 1.2818 | dt 0.026 | |
| type train | step 10230 | loss 0.0507 0.7513 4.4142 18.2721 | lr 5.6e-04 | norm 1.1962 | dt 0.026 | |
| type train | step 10240 | loss 0.0490 0.7678 4.4961 18.4205 | lr 5.6e-04 | norm 1.1946 | dt 0.026 | |
| type train | step 10250 | loss 0.0490 0.7722 4.5559 18.5172 | lr 5.6e-04 | norm 1.2826 | dt 0.026 | |
| type train | step 10260 | loss 0.0512 0.7644 4.4325 18.3553 | lr 5.6e-04 | norm 1.3829 | dt 0.025 | |
| type train | step 10270 | loss 0.0495 0.7668 4.5937 19.0200 | lr 5.6e-04 | norm 1.2224 | dt 0.026 | |
| type train | step 10280 | loss 0.0499 0.7651 4.5327 18.7352 | lr 5.6e-04 | norm 1.3895 | dt 0.026 | |
| type train | step 10290 | loss 0.0502 0.7670 4.5154 18.9084 | lr 5.6e-04 | norm 1.2543 | dt 0.026 | |
| type train | step 10300 | loss 0.0470 0.7548 4.3617 17.7774 | lr 5.6e-04 | norm 1.1612 | dt 0.026 | |
| type train | step 10310 | loss 0.0471 0.7716 4.4267 18.4037 | lr 5.5e-04 | norm 1.3634 | dt 0.025 | |
| type train | step 10320 | loss 0.0504 0.7607 4.4110 18.2121 | lr 5.5e-04 | norm 1.5800 | dt 0.026 | |
| type train | step 10330 | loss 0.0488 0.7649 4.6327 19.3166 | lr 5.5e-04 | norm 1.2580 | dt 0.026 | |
| type train | step 10340 | loss 0.0492 0.7606 4.3940 17.8369 | lr 5.5e-04 | norm 1.4876 | dt 0.026 | |
| type train | step 10350 | loss 0.0492 0.7642 4.4901 18.7092 | lr 5.5e-04 | norm 1.3773 | dt 0.026 | |
| type train | step 10360 | loss 0.0476 0.7570 4.3901 17.9041 | lr 5.5e-04 | norm 1.1736 | dt 0.026 | |
| type train | step 10370 | loss 0.0481 0.7726 4.6705 19.0416 | lr 5.5e-04 | norm 1.5067 | dt 0.028 | |
| type train | step 10380 | loss 0.0487 0.7599 4.4625 18.3166 | lr 5.5e-04 | norm 1.1559 | dt 0.026 | |
| type train | step 10390 | loss 0.0488 0.7707 4.5500 18.8889 | lr 5.5e-04 | norm 1.2261 | dt 0.026 | |
| type train | step 10400 | loss 0.0488 0.7410 4.3369 17.8712 | lr 5.5e-04 | norm 1.2569 | dt 0.026 | |
| type train | step 10410 | loss 0.0503 0.7830 4.5424 18.6619 | lr 5.5e-04 | norm 1.5127 | dt 0.026 | |
| type train | step 10420 | loss 0.0473 0.7586 4.3858 17.7552 | lr 5.5e-04 | norm 1.2507 | dt 0.026 | |
| type train | step 10430 | loss 0.0475 0.7502 4.4373 18.4631 | lr 5.5e-04 | norm 1.2432 | dt 0.026 | |
| type train | step 10440 | loss 0.0480 0.7545 4.3670 17.8222 | lr 5.5e-04 | norm 1.0219 | dt 0.025 | |
| type train | step 10450 | loss 0.0510 0.7930 4.7250 19.4412 | lr 5.4e-04 | norm 1.2887 | dt 0.026 | |
| type train | step 10460 | loss 0.0479 0.7564 4.4154 18.1495 | lr 5.4e-04 | norm 1.3046 | dt 0.026 | |
| type train | step 10470 | loss 0.0477 0.7298 4.3016 17.8407 | lr 5.4e-04 | norm 1.1338 | dt 0.026 | |
| type train | step 10480 | loss 0.0492 0.7716 4.4688 18.2365 | lr 5.4e-04 | norm 1.2415 | dt 0.026 | |
| type train | step 10490 | loss 0.0497 0.7687 4.3400 17.9018 | lr 5.4e-04 | norm 1.2610 | dt 0.025 | |
| type train | step 10500 | loss 0.0482 0.7550 4.3240 18.0058 | lr 5.4e-04 | norm 1.1255 | dt 0.025 | |
| type train | step 10510 | loss 0.0524 0.7700 4.3917 18.3971 | lr 5.4e-04 | norm 1.8729 | dt 0.026 | |
| type train | step 10520 | loss 0.0494 0.7771 4.4950 18.4995 | lr 5.4e-04 | norm 1.4267 | dt 0.026 | |
| type train | step 10530 | loss 0.0493 0.7491 4.4649 18.4095 | lr 5.4e-04 | norm 1.2714 | dt 0.026 | |
| type train | step 10540 | loss 0.0469 0.7685 4.4385 17.9765 | lr 5.4e-04 | norm 1.3428 | dt 0.026 | |
| type train | step 10550 | loss 0.0486 0.7666 4.3936 18.2189 | lr 5.4e-04 | norm 1.3440 | dt 0.026 | |
| type train | step 10560 | loss 0.0483 0.7531 4.3832 18.1248 | lr 5.4e-04 | norm 1.5626 | dt 0.027 | |
| type train | step 10570 | loss 0.0512 0.7793 4.5812 18.8475 | lr 5.4e-04 | norm 1.3159 | dt 0.026 | |
| type train | step 10580 | loss 0.0497 0.7550 4.3698 18.1878 | lr 5.3e-04 | norm 1.2598 | dt 0.026 | |
| type train | step 10590 | loss 0.0486 0.7446 4.4623 18.2785 | lr 5.3e-04 | norm 1.2197 | dt 0.025 | |
| type train | step 10600 | loss 0.0469 0.7614 4.4241 18.0952 | lr 5.3e-04 | norm 1.1936 | dt 0.026 | |
| type train | step 10610 | loss 0.0490 0.7897 4.5271 18.5978 | lr 5.3e-04 | norm 1.3648 | dt 0.025 | |
| type train | step 10620 | loss 0.0480 0.7458 4.3019 17.7425 | lr 5.3e-04 | norm 1.1309 | dt 0.026 | |
| type train | step 10630 | loss 0.0474 0.7488 4.3939 18.1600 | lr 5.3e-04 | norm 1.1779 | dt 0.026 | |
| type train | step 10640 | loss 0.0498 0.7449 4.3107 17.9640 | lr 5.3e-04 | norm 1.1357 | dt 0.026 | |
| type train | step 10650 | loss 0.0503 0.7617 4.5019 18.4264 | lr 5.3e-04 | norm 1.4500 | dt 0.026 | |
| type train | step 10660 | loss 0.0477 0.7529 4.3138 17.8985 | lr 5.3e-04 | norm 1.3234 | dt 0.026 | |
| type train | step 10670 | loss 0.0493 0.7788 4.4493 18.0964 | lr 5.3e-04 | norm 1.3924 | dt 0.026 | |
| type train | step 10680 | loss 0.0491 0.7579 4.4021 18.2233 | lr 5.3e-04 | norm 1.3979 | dt 0.026 | |
| type train | step 10690 | loss 0.0490 0.7429 4.4147 18.2233 | lr 5.3e-04 | norm 1.3930 | dt 0.026 | |
| type train | step 10700 | loss 0.0496 0.7603 4.5111 18.6883 | lr 5.3e-04 | norm 1.1124 | dt 0.026 | |
| type train | step 10710 | loss 0.0478 0.7596 4.4733 18.4965 | lr 5.3e-04 | norm 1.1258 | dt 0.026 | |
| type train | step 10720 | loss 0.0476 0.7680 4.4422 18.2668 | lr 5.2e-04 | norm 1.3771 | dt 0.026 | |
| type train | step 10730 | loss 0.0479 0.7392 4.2586 17.6541 | lr 5.2e-04 | norm 1.3651 | dt 0.026 | |
| type train | step 10740 | loss 0.0482 0.7593 4.3144 18.1477 | lr 5.2e-04 | norm 1.3809 | dt 0.026 | |
| type train | step 10750 | loss 0.0503 0.7605 4.4368 18.4866 | lr 5.2e-04 | norm 1.3768 | dt 0.026 | |
| type train | step 10760 | loss 0.0480 0.7453 4.5171 18.6392 | lr 5.2e-04 | norm 1.1626 | dt 0.025 | |
| type train | step 10770 | loss 0.0495 0.7583 4.3821 17.8972 | lr 5.2e-04 | norm 1.2459 | dt 0.026 | |
| type train | step 10780 | loss 0.0481 0.7650 4.4293 18.1125 | lr 5.2e-04 | norm 1.1570 | dt 0.026 | |
| type train | step 10790 | loss 0.0483 0.7753 4.3573 18.0284 | lr 5.2e-04 | norm 1.4381 | dt 0.026 | |
| type train | step 10800 | loss 0.0481 0.7657 4.4912 18.4334 | lr 5.2e-04 | norm 1.1576 | dt 0.026 | |
| type train | step 10810 | loss 0.0484 0.7495 4.3243 17.9815 | lr 5.2e-04 | norm 1.3821 | dt 0.026 | |
| type train | step 10820 | loss 0.0499 0.7701 4.4926 18.3999 | lr 5.2e-04 | norm 1.1232 | dt 0.026 | |
| type train | step 10830 | loss 0.0475 0.7408 4.2174 17.4563 | lr 5.2e-04 | norm 1.2710 | dt 0.026 | |
| type train | step 10840 | loss 0.0499 0.7478 4.4044 18.2487 | lr 5.2e-04 | norm 1.1795 | dt 0.026 | |
| type train | step 10850 | loss 0.0484 0.7653 4.4868 18.3845 | lr 5.2e-04 | norm 1.1792 | dt 0.026 | |
| type train | step 10860 | loss 0.0482 0.7708 4.5401 18.4939 | lr 5.1e-04 | norm 1.2779 | dt 0.025 | |
| type train | step 10870 | loss 0.0503 0.7604 4.4199 18.3393 | lr 5.1e-04 | norm 1.3704 | dt 0.026 | |
| type train | step 10880 | loss 0.0485 0.7643 4.5864 19.0045 | lr 5.1e-04 | norm 1.2120 | dt 0.026 | |
| type train | step 10890 | loss 0.0490 0.7613 4.5201 18.7179 | lr 5.1e-04 | norm 1.3825 | dt 0.026 | |
| type train | step 10900 | loss 0.0494 0.7639 4.5054 18.8956 | lr 5.1e-04 | norm 1.2542 | dt 0.026 | |
| type train | step 10910 | loss 0.0463 0.7513 4.3509 17.7651 | lr 5.1e-04 | norm 1.1546 | dt 0.026 | |
| type train | step 10920 | loss 0.0462 0.7701 4.4159 18.3759 | lr 5.1e-04 | norm 1.3445 | dt 0.027 | |
| type train | step 10930 | loss 0.0496 0.7573 4.3980 18.1936 | lr 5.1e-04 | norm 1.5710 | dt 0.026 | |
| type train | step 10940 | loss 0.0480 0.7613 4.6186 19.3053 | lr 5.1e-04 | norm 1.2544 | dt 0.026 | |
| type train | step 10950 | loss 0.0484 0.7584 4.3851 17.8162 | lr 5.1e-04 | norm 1.4979 | dt 0.026 | |
| type train | step 10960 | loss 0.0484 0.7626 4.4859 18.6865 | lr 5.1e-04 | norm 1.3693 | dt 0.026 | |
| type train | step 10970 | loss 0.0470 0.7552 4.3770 17.8887 | lr 5.1e-04 | norm 1.1734 | dt 0.026 | |
| type train | step 10980 | loss 0.0470 0.7712 4.6611 19.0357 | lr 5.1e-04 | norm 1.4762 | dt 0.027 | |
| type train | step 10990 | loss 0.0476 0.7578 4.4507 18.3022 | lr 5.0e-04 | norm 1.1290 | dt 0.026 | |
| type train | step 11000 | loss 0.0479 0.7666 4.5410 18.8842 | lr 5.0e-04 | norm 1.2196 | dt 0.026 | |
| type train | step 11010 | loss 0.0480 0.7393 4.3264 17.8533 | lr 5.0e-04 | norm 1.2642 | dt 0.026 | |
| type train | step 11020 | loss 0.0496 0.7827 4.5356 18.6506 | lr 5.0e-04 | norm 1.4992 | dt 0.026 | |
| type train | step 11030 | loss 0.0464 0.7569 4.3790 17.7351 | lr 5.0e-04 | norm 1.2408 | dt 0.026 | |
| type train | step 11040 | loss 0.0464 0.7478 4.4255 18.4486 | lr 5.0e-04 | norm 1.2100 | dt 0.026 | |
| type train | step 11050 | loss 0.0472 0.7520 4.3567 17.7999 | lr 5.0e-04 | norm 1.0028 | dt 0.026 | |
| type train | step 11060 | loss 0.0498 0.7897 4.7120 19.4267 | lr 5.0e-04 | norm 1.2761 | dt 0.026 | |
| type train | step 11070 | loss 0.0471 0.7547 4.4045 18.1281 | lr 5.0e-04 | norm 1.2898 | dt 0.025 | |
| type train | step 11080 | loss 0.0467 0.7272 4.2913 17.8142 | lr 5.0e-04 | norm 1.1213 | dt 0.027 | |
| type train | step 11090 | loss 0.0484 0.7687 4.4638 18.2135 | lr 5.0e-04 | norm 1.2314 | dt 0.026 | |
| type train | step 11100 | loss 0.0488 0.7682 4.3273 17.8810 | lr 5.0e-04 | norm 1.2657 | dt 0.026 | |
| type train | step 11110 | loss 0.0473 0.7536 4.3140 17.9971 | lr 5.0e-04 | norm 1.1225 | dt 0.026 | |
| type train | step 11120 | loss 0.0514 0.7647 4.3828 18.3837 | lr 5.0e-04 | norm 1.8574 | dt 0.026 | |
| type train | step 11130 | loss 0.0486 0.7754 4.4860 18.4684 | lr 4.9e-04 | norm 1.4226 | dt 0.025 | |
| type train | step 11140 | loss 0.0484 0.7465 4.4558 18.3873 | lr 4.9e-04 | norm 1.2541 | dt 0.026 | |
| type train | step 11150 | loss 0.0460 0.7649 4.4275 17.9578 | lr 4.9e-04 | norm 1.3232 | dt 0.026 | |
| type train | step 11160 | loss 0.0477 0.7660 4.3817 18.1962 | lr 4.9e-04 | norm 1.3453 | dt 0.026 | |
| type train | step 11170 | loss 0.0475 0.7520 4.3707 18.0948 | lr 4.9e-04 | norm 1.5556 | dt 0.026 | |
| type train | step 11180 | loss 0.0503 0.7771 4.5705 18.8365 | lr 4.9e-04 | norm 1.3159 | dt 0.026 | |
| type train | step 11190 | loss 0.0489 0.7541 4.3626 18.1804 | lr 4.9e-04 | norm 1.2528 | dt 0.026 | |
| type train | step 11200 | loss 0.0478 0.7418 4.4524 18.2559 | lr 4.9e-04 | norm 1.2244 | dt 0.025 | |
| type train | step 11210 | loss 0.0460 0.7583 4.4118 18.0765 | lr 4.9e-04 | norm 1.1722 | dt 0.026 | |
| type train | step 11220 | loss 0.0484 0.7896 4.5242 18.5718 | lr 4.9e-04 | norm 1.3780 | dt 0.026 | |
| type train | step 11230 | loss 0.0471 0.7448 4.2911 17.7256 | lr 4.9e-04 | norm 1.1246 | dt 0.026 | |
| type train | step 11240 | loss 0.0466 0.7468 4.3833 18.1597 | lr 4.9e-04 | norm 1.1784 | dt 0.027 | |
| type train | step 11250 | loss 0.0488 0.7433 4.3052 17.9602 | lr 4.9e-04 | norm 1.1197 | dt 0.026 | |
| type train | step 11260 | loss 0.0494 0.7599 4.4900 18.4170 | lr 4.9e-04 | norm 1.4233 | dt 0.025 | |
| type train | step 11270 | loss 0.0470 0.7500 4.3016 17.8708 | lr 4.8e-04 | norm 1.3036 | dt 0.026 | |
| type train | step 11280 | loss 0.0483 0.7778 4.4432 18.0721 | lr 4.8e-04 | norm 1.3810 | dt 0.026 | |
| type train | step 11290 | loss 0.0484 0.7558 4.3903 18.1821 | lr 4.8e-04 | norm 1.3925 | dt 0.026 | |
| type train | step 11300 | loss 0.0481 0.7410 4.4076 18.2112 | lr 4.8e-04 | norm 1.3802 | dt 0.026 | |
| type train | step 11310 | loss 0.0491 0.7584 4.5010 18.6811 | lr 4.8e-04 | norm 1.1175 | dt 0.025 | |
| type train | step 11320 | loss 0.0471 0.7581 4.4617 18.4885 | lr 4.8e-04 | norm 1.1196 | dt 0.026 | |
| type train | step 11330 | loss 0.0467 0.7650 4.4294 18.2507 | lr 4.8e-04 | norm 1.3602 | dt 0.026 | |
| type train | step 11340 | loss 0.0470 0.7379 4.2471 17.6461 | lr 4.8e-04 | norm 1.3492 | dt 0.026 | |
| type train | step 11350 | loss 0.0472 0.7580 4.3063 18.1254 | lr 4.8e-04 | norm 1.3596 | dt 0.026 | |
| type train | step 11360 | loss 0.0491 0.7572 4.4274 18.4718 | lr 4.8e-04 | norm 1.3583 | dt 0.026 | |
| type train | step 11370 | loss 0.0469 0.7425 4.5073 18.6393 | lr 4.8e-04 | norm 1.1435 | dt 0.026 | |
| type train | step 11380 | loss 0.0485 0.7563 4.3703 17.8810 | lr 4.8e-04 | norm 1.2597 | dt 0.026 | |
| type train | step 11390 | loss 0.0470 0.7616 4.4189 18.0942 | lr 4.8e-04 | norm 1.1288 | dt 0.026 | |
| type train | step 11400 | loss 0.0469 0.7738 4.3510 18.0106 | lr 4.8e-04 | norm 1.4123 | dt 0.026 | |
| type train | step 11410 | loss 0.0469 0.7650 4.4826 18.4254 | lr 4.7e-04 | norm 1.1605 | dt 0.026 | |
| type train | step 11420 | loss 0.0473 0.7477 4.3163 17.9771 | lr 4.7e-04 | norm 1.3778 | dt 0.026 | |
| type train | step 11430 | loss 0.0485 0.7679 4.4818 18.3973 | lr 4.7e-04 | norm 1.1022 | dt 0.025 | |
| type train | step 11440 | loss 0.0465 0.7393 4.2062 17.4502 | lr 4.7e-04 | norm 1.2752 | dt 0.026 | |
| type train | step 11450 | loss 0.0485 0.7456 4.3965 18.2433 | lr 4.7e-04 | norm 1.1687 | dt 0.026 | |
| type train | step 11460 | loss 0.0471 0.7633 4.4766 18.3647 | lr 4.7e-04 | norm 1.1584 | dt 0.026 | |
| type train | step 11470 | loss 0.0468 0.7688 4.5270 18.4820 | lr 4.7e-04 | norm 1.2604 | dt 0.026 | |
| type train | step 11480 | loss 0.0490 0.7579 4.4115 18.3339 | lr 4.7e-04 | norm 1.3582 | dt 0.026 | |
| type train | step 11490 | loss 0.0475 0.7628 4.5768 18.9914 | lr 4.7e-04 | norm 1.2064 | dt 0.025 | |
| type train | step 11500 | loss 0.0477 0.7592 4.5173 18.6987 | lr 4.7e-04 | norm 1.3777 | dt 0.026 | |
| type train | step 11510 | loss 0.0483 0.7631 4.5008 18.8725 | lr 4.7e-04 | norm 1.2438 | dt 0.025 | |
| type train | step 11520 | loss 0.0453 0.7503 4.3416 17.7522 | lr 4.7e-04 | norm 1.1639 | dt 0.025 | |
| type train | step 11530 | loss 0.0452 0.7685 4.4089 18.3512 | lr 4.7e-04 | norm 1.3403 | dt 0.026 | |
| type train | step 11540 | loss 0.0483 0.7540 4.3895 18.1905 | lr 4.6e-04 | norm 1.5598 | dt 0.026 | |
| type train | step 11550 | loss 0.0466 0.7588 4.6102 19.2944 | lr 4.6e-04 | norm 1.2359 | dt 0.026 | |
| type train | step 11560 | loss 0.0472 0.7557 4.3766 17.8056 | lr 4.6e-04 | norm 1.5014 | dt 0.026 | |
| type train | step 11570 | loss 0.0471 0.7604 4.4797 18.6783 | lr 4.6e-04 | norm 1.3420 | dt 0.026 | |
| type train | step 11580 | loss 0.0457 0.7533 4.3724 17.8750 | lr 4.6e-04 | norm 1.1592 | dt 0.026 | |
| type train | step 11590 | loss 0.0459 0.7695 4.6530 19.0182 | lr 4.6e-04 | norm 1.4651 | dt 0.028 | |
| type train | step 11600 | loss 0.0464 0.7561 4.4417 18.3020 | lr 4.6e-04 | norm 1.1201 | dt 0.026 | |
| type train | step 11610 | loss 0.0469 0.7638 4.5304 18.8735 | lr 4.6e-04 | norm 1.2130 | dt 0.026 | |
| type train | step 11620 | loss 0.0468 0.7363 4.3128 17.8489 | lr 4.6e-04 | norm 1.2503 | dt 0.025 | |
| type train | step 11630 | loss 0.0481 0.7799 4.5312 18.6371 | lr 4.6e-04 | norm 1.4938 | dt 0.025 | |
| type train | step 11640 | loss 0.0452 0.7551 4.3756 17.7085 | lr 4.6e-04 | norm 1.2384 | dt 0.025 | |
| type train | step 11650 | loss 0.0453 0.7456 4.4156 18.4368 | lr 4.6e-04 | norm 1.1991 | dt 0.025 | |
| type train | step 11660 | loss 0.0460 0.7502 4.3514 17.7915 | lr 4.6e-04 | norm 1.0097 | dt 0.026 | |
| type train | step 11670 | loss 0.0487 0.7868 4.7005 19.4134 | lr 4.6e-04 | norm 1.2728 | dt 0.026 | |
| type train | step 11680 | loss 0.0459 0.7529 4.3957 18.1143 | lr 4.5e-04 | norm 1.3011 | dt 0.026 | |
| type train | step 11690 | loss 0.0457 0.7245 4.2852 17.7976 | lr 4.5e-04 | norm 1.1083 | dt 0.027 | |
| type train | step 11700 | loss 0.0471 0.7672 4.4546 18.1876 | lr 4.5e-04 | norm 1.2215 | dt 0.026 | |
| type train | step 11710 | loss 0.0475 0.7649 4.3214 17.8566 | lr 4.5e-04 | norm 1.2472 | dt 0.026 | |
| type train | step 11720 | loss 0.0461 0.7515 4.3065 17.9947 | lr 4.5e-04 | norm 1.1192 | dt 0.026 | |
| type train | step 11730 | loss 0.0499 0.7613 4.3730 18.3766 | lr 4.5e-04 | norm 1.8542 | dt 0.026 | |
| type train | step 11740 | loss 0.0473 0.7751 4.4802 18.4602 | lr 4.5e-04 | norm 1.4149 | dt 0.026 | |
| type train | step 11750 | loss 0.0471 0.7442 4.4476 18.3637 | lr 4.5e-04 | norm 1.2347 | dt 0.026 | |
| type train | step 11760 | loss 0.0450 0.7621 4.4162 17.9348 | lr 4.5e-04 | norm 1.3219 | dt 0.026 | |
| type train | step 11770 | loss 0.0464 0.7643 4.3760 18.1925 | lr 4.5e-04 | norm 1.3218 | dt 0.026 | |
| type train | step 11780 | loss 0.0464 0.7500 4.3613 18.0725 | lr 4.5e-04 | norm 1.5448 | dt 0.026 | |
| type train | step 11790 | loss 0.0487 0.7744 4.5579 18.8235 | lr 4.5e-04 | norm 1.2987 | dt 0.026 | |
| type train | step 11800 | loss 0.0477 0.7520 4.3524 18.1755 | lr 4.5e-04 | norm 1.2450 | dt 0.025 | |
| type train | step 11810 | loss 0.0465 0.7404 4.4442 18.2469 | lr 4.5e-04 | norm 1.1961 | dt 0.026 | |
| type train | step 11820 | loss 0.0450 0.7555 4.4009 18.0651 | lr 4.4e-04 | norm 1.1624 | dt 0.026 | |
| type train | step 11830 | loss 0.0470 0.7891 4.5197 18.5557 | lr 4.4e-04 | norm 1.3657 | dt 0.025 | |
| type train | step 11840 | loss 0.0461 0.7433 4.2815 17.7172 | lr 4.4e-04 | norm 1.1154 | dt 0.026 | |
| type train | step 11850 | loss 0.0454 0.7435 4.3763 18.1609 | lr 4.4e-04 | norm 1.1659 | dt 0.025 | |
| type train | step 11860 | loss 0.0475 0.7410 4.2992 17.9490 | lr 4.4e-04 | norm 1.1158 | dt 0.026 | |
| type train | step 11870 | loss 0.0481 0.7547 4.4817 18.4065 | lr 4.4e-04 | norm 1.4028 | dt 0.026 | |
| type train | step 11880 | loss 0.0458 0.7455 4.2904 17.8542 | lr 4.4e-04 | norm 1.2827 | dt 0.026 | |
| type train | step 11890 | loss 0.0471 0.7722 4.4402 18.0464 | lr 4.4e-04 | norm 1.3620 | dt 0.026 | |
| type train | step 11900 | loss 0.0470 0.7504 4.3822 18.1633 | lr 4.4e-04 | norm 1.3781 | dt 0.026 | |
| type train | step 11910 | loss 0.0471 0.7348 4.4001 18.2066 | lr 4.4e-04 | norm 1.3780 | dt 0.026 | |
| type train | step 11920 | loss 0.0478 0.7510 4.4925 18.6717 | lr 4.4e-04 | norm 1.1038 | dt 0.026 | |
| type train | step 11930 | loss 0.0459 0.7504 4.4545 18.4675 | lr 4.4e-04 | norm 1.1096 | dt 0.026 | |
| type train | step 11940 | loss 0.0455 0.7577 4.4228 18.2283 | lr 4.4e-04 | norm 1.3389 | dt 0.027 | |
| type train | step 11950 | loss 0.0461 0.7314 4.2375 17.6418 | lr 4.4e-04 | norm 1.3398 | dt 0.026 | |
| type train | step 11960 | loss 0.0461 0.7518 4.2992 18.1084 | lr 4.3e-04 | norm 1.3479 | dt 0.026 | |
| type train | step 11970 | loss 0.0481 0.7500 4.4192 18.4588 | lr 4.3e-04 | norm 1.3534 | dt 0.026 | |
| type train | step 11980 | loss 0.0460 0.7354 4.5029 18.6243 | lr 4.3e-04 | norm 1.1359 | dt 0.026 | |
| type train | step 11990 | loss 0.0476 0.7512 4.3652 17.8714 | lr 4.3e-04 | norm 1.2708 | dt 0.026 | |
| type train | step 12000 | loss 0.0458 0.7542 4.4112 18.0811 | lr 4.3e-04 | norm 1.0979 | dt 0.025 | |
| type train | step 12010 | loss 0.0461 0.7672 4.3446 17.9986 | lr 4.3e-04 | norm 1.4048 | dt 0.026 | |
| type train | step 12020 | loss 0.0461 0.7587 4.4770 18.4032 | lr 4.3e-04 | norm 1.1501 | dt 0.026 | |
| type train | step 12030 | loss 0.0467 0.7423 4.3099 17.9755 | lr 4.3e-04 | norm 1.3722 | dt 0.026 | |
| type train | step 12040 | loss 0.0477 0.7609 4.4722 18.3893 | lr 4.3e-04 | norm 1.0870 | dt 0.025 | |
| type train | step 12050 | loss 0.0457 0.7341 4.1991 17.4459 | lr 4.3e-04 | norm 1.2836 | dt 0.026 | |
| type train | step 12060 | loss 0.0476 0.7391 4.3946 18.2361 | lr 4.3e-04 | norm 1.1568 | dt 0.026 | |
| type train | step 12070 | loss 0.0465 0.7560 4.4726 18.3446 | lr 4.3e-04 | norm 1.1599 | dt 0.026 | |
| type train | step 12080 | loss 0.0462 0.7617 4.5172 18.4803 | lr 4.3e-04 | norm 1.2442 | dt 0.026 | |
| type train | step 12090 | loss 0.0482 0.7511 4.4040 18.3295 | lr 4.3e-04 | norm 1.3466 | dt 0.026 | |
| type train | step 12100 | loss 0.0466 0.7563 4.5694 18.9859 | lr 4.2e-04 | norm 1.1997 | dt 0.026 | |
| type train | step 12110 | loss 0.0471 0.7529 4.5104 18.6954 | lr 4.2e-04 | norm 1.3807 | dt 0.026 | |
| type train | step 12120 | loss 0.0475 0.7555 4.4945 18.8715 | lr 4.2e-04 | norm 1.2338 | dt 0.026 | |
| type train | step 12130 | loss 0.0446 0.7421 4.3344 17.7467 | lr 4.2e-04 | norm 1.1497 | dt 0.026 | |
| type train | step 12140 | loss 0.0445 0.7602 4.3988 18.3303 | lr 4.2e-04 | norm 1.3087 | dt 0.025 | |
| type train | step 12150 | loss 0.0476 0.7473 4.3809 18.1785 | lr 4.2e-04 | norm 1.5347 | dt 0.026 | |
| type train | step 12160 | loss 0.0459 0.7530 4.6007 19.2893 | lr 4.2e-04 | norm 1.2268 | dt 0.026 | |
| type train | step 12170 | loss 0.0468 0.7490 4.3716 17.7916 | lr 4.2e-04 | norm 1.5164 | dt 0.025 | |
| type train | step 12180 | loss 0.0466 0.7539 4.4760 18.6554 | lr 4.2e-04 | norm 1.3440 | dt 0.026 | |
| type train | step 12190 | loss 0.0451 0.7466 4.3700 17.8614 | lr 4.2e-04 | norm 1.1605 | dt 0.026 | |
| type train | step 12200 | loss 0.0452 0.7611 4.6462 19.0061 | lr 4.2e-04 | norm 1.4405 | dt 0.028 | |
| type train | step 12210 | loss 0.0456 0.7487 4.4379 18.2960 | lr 4.2e-04 | norm 1.1171 | dt 0.026 | |
| type train | step 12220 | loss 0.0463 0.7570 4.5259 18.8526 | lr 4.2e-04 | norm 1.1987 | dt 0.026 | |
| type train | step 12230 | loss 0.0461 0.7296 4.3066 17.8343 | lr 4.2e-04 | norm 1.2470 | dt 0.025 | |
| type train | step 12240 | loss 0.0475 0.7740 4.5267 18.6421 | lr 4.2e-04 | norm 1.4897 | dt 0.026 | |
| type train | step 12250 | loss 0.0445 0.7477 4.3686 17.6973 | lr 4.1e-04 | norm 1.2344 | dt 0.026 | |
| type train | step 12260 | loss 0.0446 0.7391 4.4115 18.4238 | lr 4.1e-04 | norm 1.1817 | dt 0.026 | |
| type train | step 12270 | loss 0.0455 0.7442 4.3454 17.7960 | lr 4.1e-04 | norm 1.0206 | dt 0.026 | |
| type train | step 12280 | loss 0.0480 0.7794 4.6937 19.3942 | lr 4.1e-04 | norm 1.2611 | dt 0.025 | |
| type train | step 12290 | loss 0.0453 0.7464 4.3884 18.0934 | lr 4.1e-04 | norm 1.2840 | dt 0.026 | |
| type train | step 12300 | loss 0.0449 0.7187 4.2835 17.7830 | lr 4.1e-04 | norm 1.0888 | dt 0.026 | |
| type train | step 12310 | loss 0.0463 0.7594 4.4482 18.1773 | lr 4.1e-04 | norm 1.2104 | dt 0.025 | |
| type train | step 12320 | loss 0.0469 0.7586 4.3156 17.8484 | lr 4.1e-04 | norm 1.2417 | dt 0.026 | |
| type train | step 12330 | loss 0.0455 0.7449 4.3011 17.9930 | lr 4.1e-04 | norm 1.1134 | dt 0.026 | |
| type train | step 12340 | loss 0.0494 0.7527 4.3678 18.3603 | lr 4.1e-04 | norm 1.8416 | dt 0.026 | |
| type train | step 12350 | loss 0.0465 0.7681 4.4704 18.4438 | lr 4.1e-04 | norm 1.3883 | dt 0.026 | |
| type train | step 12360 | loss 0.0466 0.7383 4.4398 18.3411 | lr 4.1e-04 | norm 1.2199 | dt 0.026 | |
| type train | step 12370 | loss 0.0444 0.7537 4.4100 17.9176 | lr 4.1e-04 | norm 1.2875 | dt 0.025 | |
| type train | step 12380 | loss 0.0459 0.7565 4.3741 18.1771 | lr 4.1e-04 | norm 1.3186 | dt 0.025 | |
| type train | step 12390 | loss 0.0457 0.7433 4.3571 18.0587 | lr 4.0e-04 | norm 1.5327 | dt 0.026 | |
| type train | step 12400 | loss 0.0482 0.7675 4.5546 18.8093 | lr 4.0e-04 | norm 1.3004 | dt 0.026 | |
| type train | step 12410 | loss 0.0471 0.7457 4.3559 18.1525 | lr 4.0e-04 | norm 1.2341 | dt 0.026 | |
| type train | step 12420 | loss 0.0457 0.7329 4.4384 18.2365 | lr 4.0e-04 | norm 1.1741 | dt 0.026 | |
| type train | step 12430 | loss 0.0443 0.7496 4.3928 18.0579 | lr 4.0e-04 | norm 1.1323 | dt 0.026 | |
| type train | step 12440 | loss 0.0463 0.7810 4.5236 18.5524 | lr 4.0e-04 | norm 1.3521 | dt 0.026 | |
| type train | step 12450 | loss 0.0455 0.7366 4.2794 17.7078 | lr 4.0e-04 | norm 1.1188 | dt 0.026 | |
| type train | step 12460 | loss 0.0450 0.7351 4.3746 18.1423 | lr 4.0e-04 | norm 1.1572 | dt 0.027 | |
| type train | step 12470 | loss 0.0470 0.7361 4.2957 17.9363 | lr 4.0e-04 | norm 1.0963 | dt 0.026 | |
| type train | step 12480 | loss 0.0476 0.7502 4.4740 18.3904 | lr 4.0e-04 | norm 1.3857 | dt 0.025 | |
| type train | step 12490 | loss 0.0453 0.7418 4.2847 17.8407 | lr 4.0e-04 | norm 1.2752 | dt 0.026 | |
| type train | step 12500 | loss 0.0465 0.7707 4.4373 18.0310 | lr 4.0e-04 | norm 1.3777 | dt 0.026 | |
| type train | step 12510 | loss 0.0464 0.7477 4.3757 18.1496 | lr 4.0e-04 | norm 1.3773 | dt 0.026 | |
| type train | step 12520 | loss 0.0465 0.7318 4.3940 18.1882 | lr 4.0e-04 | norm 1.3718 | dt 0.025 | |
| type train | step 12530 | loss 0.0471 0.7480 4.4850 18.6678 | lr 3.9e-04 | norm 1.0800 | dt 0.025 | |
| type train | step 12540 | loss 0.0454 0.7485 4.4474 18.4617 | lr 3.9e-04 | norm 1.0904 | dt 0.026 | |
| type train | step 12550 | loss 0.0449 0.7538 4.4150 18.2051 | lr 3.9e-04 | norm 1.3161 | dt 0.026 | |
| type train | step 12560 | loss 0.0455 0.7298 4.2321 17.6402 | lr 3.9e-04 | norm 1.3348 | dt 0.026 | |
| type train | step 12570 | loss 0.0455 0.7498 4.2936 18.0911 | lr 3.9e-04 | norm 1.3392 | dt 0.026 | |
| type train | step 12580 | loss 0.0475 0.7482 4.4171 18.4405 | lr 3.9e-04 | norm 1.3498 | dt 0.025 | |
| type train | step 12590 | loss 0.0455 0.7328 4.5007 18.6308 | lr 3.9e-04 | norm 1.1193 | dt 0.026 | |
| type train | step 12600 | loss 0.0468 0.7503 4.3601 17.8593 | lr 3.9e-04 | norm 1.2577 | dt 0.026 | |
| type train | step 12610 | loss 0.0455 0.7525 4.4067 18.0749 | lr 3.9e-04 | norm 1.1001 | dt 0.026 | |
| type train | step 12620 | loss 0.0456 0.7647 4.3390 17.9914 | lr 3.9e-04 | norm 1.3917 | dt 0.026 | |
| type train | step 12630 | loss 0.0454 0.7568 4.4714 18.3959 | lr 3.9e-04 | norm 1.1332 | dt 0.026 | |
| type train | step 12640 | loss 0.0461 0.7397 4.3087 17.9620 | lr 3.9e-04 | norm 1.3661 | dt 0.025 | |
| type train | step 12650 | loss 0.0472 0.7586 4.4716 18.3786 | lr 3.9e-04 | norm 1.0774 | dt 0.025 | |
| type train | step 12660 | loss 0.0451 0.7325 4.1958 17.4322 | lr 3.9e-04 | norm 1.2697 | dt 0.026 | |
| type train | step 12670 | loss 0.0472 0.7365 4.3948 18.2322 | lr 3.9e-04 | norm 1.1678 | dt 0.026 | |
| type train | step 12680 | loss 0.0459 0.7534 4.4706 18.3350 | lr 3.8e-04 | norm 1.1523 | dt 0.026 | |
| type train | step 12690 | loss 0.0455 0.7594 4.5071 18.4693 | lr 3.8e-04 | norm 1.2446 | dt 0.026 | |
| type train | step 12700 | loss 0.0476 0.7488 4.3986 18.3152 | lr 3.8e-04 | norm 1.3370 | dt 0.025 | |
| type train | step 12710 | loss 0.0461 0.7537 4.5681 18.9705 | lr 3.8e-04 | norm 1.1882 | dt 0.026 | |
| type train | step 12720 | loss 0.0465 0.7516 4.5057 18.6693 | lr 3.8e-04 | norm 1.3750 | dt 0.026 | |
| type train | step 12730 | loss 0.0470 0.7529 4.4984 18.8604 | lr 3.8e-04 | norm 1.2330 | dt 0.026 | |
| type train | step 12740 | loss 0.0441 0.7399 4.3251 17.7313 | lr 3.8e-04 | norm 1.1280 | dt 0.026 | |
| type train | step 12750 | loss 0.0439 0.7578 4.3925 18.3149 | lr 3.8e-04 | norm 1.2849 | dt 0.025 | |
| type train | step 12760 | loss 0.0470 0.7440 4.3725 18.1573 | lr 3.8e-04 | norm 1.5204 | dt 0.025 | |
| type train | step 12770 | loss 0.0454 0.7509 4.5952 19.2868 | lr 3.8e-04 | norm 1.2217 | dt 0.026 | |
| type train | step 12780 | loss 0.0459 0.7471 4.3708 17.7770 | lr 3.8e-04 | norm 1.4931 | dt 0.026 | |
| type train | step 12790 | loss 0.0461 0.7530 4.4742 18.6521 | lr 3.8e-04 | norm 1.3374 | dt 0.025 | |
| type train | step 12800 | loss 0.0444 0.7459 4.3671 17.8584 | lr 3.8e-04 | norm 1.1539 | dt 0.026 | |
| type train | step 12810 | loss 0.0446 0.7594 4.6415 18.9946 | lr 3.8e-04 | norm 1.4238 | dt 0.027 | |
| type train | step 12820 | loss 0.0450 0.7466 4.4346 18.2914 | lr 3.8e-04 | norm 1.1064 | dt 0.026 | |
| type train | step 12830 | loss 0.0458 0.7541 4.5248 18.8365 | lr 3.7e-04 | norm 1.2015 | dt 0.026 | |
| type train | step 12840 | loss 0.0456 0.7284 4.3038 17.8119 | lr 3.7e-04 | norm 1.2356 | dt 0.026 | |
| type train | step 12850 | loss 0.0470 0.7708 4.5216 18.6350 | lr 3.7e-04 | norm 1.4765 | dt 0.026 | |
| type train | step 12860 | loss 0.0440 0.7466 4.3614 17.6940 | lr 3.7e-04 | norm 1.2223 | dt 0.025 | |
| type train | step 12870 | loss 0.0443 0.7370 4.4041 18.4096 | lr 3.7e-04 | norm 1.1727 | dt 0.026 | |
| type train | step 12880 | loss 0.0450 0.7436 4.3465 17.7914 | lr 3.7e-04 | norm 1.0026 | dt 0.026 | |
| type train | step 12890 | loss 0.0475 0.7773 4.6907 19.3778 | lr 3.7e-04 | norm 1.2549 | dt 0.025 | |
| type train | step 12900 | loss 0.0448 0.7445 4.3857 18.0820 | lr 3.7e-04 | norm 1.2852 | dt 0.026 | |
| type train | step 12910 | loss 0.0445 0.7168 4.2778 17.7698 | lr 3.7e-04 | norm 1.0736 | dt 0.026 | |
| type train | step 12920 | loss 0.0457 0.7572 4.4454 18.1629 | lr 3.7e-04 | norm 1.1894 | dt 0.026 | |
| type train | step 12930 | loss 0.0465 0.7574 4.3119 17.8349 | lr 3.7e-04 | norm 1.2398 | dt 0.026 | |
| type train | step 12940 | loss 0.0451 0.7428 4.3000 17.9813 | lr 3.7e-04 | norm 1.1051 | dt 0.026 | |
| type train | step 12950 | loss 0.0488 0.7496 4.3623 18.3455 | lr 3.7e-04 | norm 1.8360 | dt 0.026 | |
| type train | step 12960 | loss 0.0461 0.7644 4.4652 18.4302 | lr 3.7e-04 | norm 1.3717 | dt 0.026 | |
| type train | step 12970 | loss 0.0462 0.7356 4.4373 18.3259 | lr 3.7e-04 | norm 1.1965 | dt 0.026 | |
| type train | step 12980 | loss 0.0439 0.7512 4.4018 17.9071 | lr 3.6e-04 | norm 1.2816 | dt 0.025 | |
| type train | step 12990 | loss 0.0454 0.7554 4.3701 18.1617 | lr 3.6e-04 | norm 1.3177 | dt 0.026 | |
| type train | step 13000 | loss 0.0453 0.7434 4.3521 18.0534 | lr 3.6e-04 | norm 1.5223 | dt 0.026 | |
| type train | step 13010 | loss 0.0477 0.7657 4.5519 18.7917 | lr 3.6e-04 | norm 1.2937 | dt 0.025 | |
| type train | step 13020 | loss 0.0470 0.7433 4.3546 18.1504 | lr 3.6e-04 | norm 1.2510 | dt 0.025 | |
| type train | step 13030 | loss 0.0453 0.7308 4.4313 18.2166 | lr 3.6e-04 | norm 1.1566 | dt 0.026 | |
| type train | step 13040 | loss 0.0439 0.7470 4.3895 18.0409 | lr 3.6e-04 | norm 1.1267 | dt 0.026 | |
| type train | step 13050 | loss 0.0459 0.7797 4.5238 18.5372 | lr 3.6e-04 | norm 1.3551 | dt 0.026 | |
| type train | step 13060 | loss 0.0450 0.7358 4.2749 17.6932 | lr 3.6e-04 | norm 1.1072 | dt 0.026 | |
| type train | step 13070 | loss 0.0446 0.7328 4.3745 18.1365 | lr 3.6e-04 | norm 1.1456 | dt 0.026 | |
| type train | step 13080 | loss 0.0465 0.7340 4.2937 17.9295 | lr 3.6e-04 | norm 1.0784 | dt 0.026 | |
| type train | step 13090 | loss 0.0470 0.7473 4.4687 18.3746 | lr 3.6e-04 | norm 1.3572 | dt 0.026 | |
| type train | step 13100 | loss 0.0450 0.7403 4.2769 17.8246 | lr 3.6e-04 | norm 1.2568 | dt 0.026 | |
| type train | step 13110 | loss 0.0460 0.7693 4.4352 18.0156 | lr 3.6e-04 | norm 1.3553 | dt 0.026 | |
| type train | step 13120 | loss 0.0462 0.7459 4.3664 18.1397 | lr 3.6e-04 | norm 1.3690 | dt 0.026 | |
| type train | step 13130 | loss 0.0461 0.7310 4.3907 18.1672 | lr 3.5e-04 | norm 1.3590 | dt 0.026 | |
| type train | step 13140 | loss 0.0467 0.7454 4.4809 18.6657 | lr 3.5e-04 | norm 1.0714 | dt 0.026 | |
| type train | step 13150 | loss 0.0450 0.7465 4.4440 18.4572 | lr 3.5e-04 | norm 1.0792 | dt 0.026 | |
| type train | step 13160 | loss 0.0446 0.7523 4.4102 18.2012 | lr 3.5e-04 | norm 1.3003 | dt 0.026 | |
| type train | step 13170 | loss 0.0451 0.7272 4.2278 17.6354 | lr 3.5e-04 | norm 1.3181 | dt 0.026 | |
| type train | step 13180 | loss 0.0452 0.7464 4.2872 18.0779 | lr 3.5e-04 | norm 1.3329 | dt 0.026 | |
| type train | step 13190 | loss 0.0471 0.7458 4.4138 18.4380 | lr 3.5e-04 | norm 1.3359 | dt 0.026 | |
| type train | step 13200 | loss 0.0450 0.7318 4.4997 18.6238 | lr 3.5e-04 | norm 1.1340 | dt 0.026 | |
| type train | step 13210 | loss 0.0465 0.7490 4.3551 17.8508 | lr 3.5e-04 | norm 1.2586 | dt 0.026 | |
| type train | step 13220 | loss 0.0451 0.7499 4.4013 18.0684 | lr 3.5e-04 | norm 1.0684 | dt 0.026 | |
| type train | step 13230 | loss 0.0452 0.7620 4.3349 17.9749 | lr 3.5e-04 | norm 1.3723 | dt 0.026 | |
| type train | step 13240 | loss 0.0453 0.7558 4.4645 18.3920 | lr 3.5e-04 | norm 1.1285 | dt 0.026 | |
| type train | step 13250 | loss 0.0459 0.7383 4.3010 17.9506 | lr 3.5e-04 | norm 1.3726 | dt 0.026 | |
| type train | step 13260 | loss 0.0467 0.7560 4.4669 18.3734 | lr 3.5e-04 | norm 1.0756 | dt 0.026 | |
| type train | step 13270 | loss 0.0450 0.7318 4.1918 17.4168 | lr 3.5e-04 | norm 1.2793 | dt 0.025 | |
| type train | step 13280 | loss 0.0467 0.7347 4.3935 18.2221 | lr 3.4e-04 | norm 1.1492 | dt 0.026 | |
| type train | step 13290 | loss 0.0456 0.7517 4.4631 18.3216 | lr 3.4e-04 | norm 1.1329 | dt 0.026 | |
| type train | step 13300 | loss 0.0452 0.7587 4.5038 18.4656 | lr 3.4e-04 | norm 1.2415 | dt 0.025 | |
| type train | step 13310 | loss 0.0473 0.7474 4.3902 18.3017 | lr 3.4e-04 | norm 1.3449 | dt 0.026 | |
| type train | step 13320 | loss 0.0459 0.7515 4.5641 18.9649 | lr 3.4e-04 | norm 1.1901 | dt 0.027 | |
| type train | step 13330 | loss 0.0461 0.7503 4.4992 18.6557 | lr 3.4e-04 | norm 1.3752 | dt 0.026 | |
| type train | step 13340 | loss 0.0466 0.7521 4.4955 18.8604 | lr 3.4e-04 | norm 1.2198 | dt 0.026 | |
| type train | step 13350 | loss 0.0438 0.7399 4.3186 17.7321 | lr 3.4e-04 | norm 1.1304 | dt 0.026 | |
| type train | step 13360 | loss 0.0434 0.7550 4.3866 18.3024 | lr 3.4e-04 | norm 1.2657 | dt 0.026 | |
| type train | step 13370 | loss 0.0467 0.7431 4.3667 18.1356 | lr 3.4e-04 | norm 1.5026 | dt 0.026 | |
| type train | step 13380 | loss 0.0451 0.7492 4.5894 19.2677 | lr 3.4e-04 | norm 1.2108 | dt 0.025 | |
| type train | step 13390 | loss 0.0455 0.7463 4.3638 17.7723 | lr 3.4e-04 | norm 1.4903 | dt 0.025 | |
| type train | step 13400 | loss 0.0458 0.7515 4.4715 18.6432 | lr 3.4e-04 | norm 1.3360 | dt 0.026 | |
| type train | step 13410 | loss 0.0441 0.7441 4.3607 17.8444 | lr 3.4e-04 | norm 1.1424 | dt 0.026 | |
| type train | step 13420 | loss 0.0443 0.7579 4.6367 18.9849 | lr 3.4e-04 | norm 1.4258 | dt 0.027 | |
| type train | step 13430 | loss 0.0449 0.7454 4.4307 18.2991 | lr 3.3e-04 | norm 1.1091 | dt 0.026 | |
| type train | step 13440 | loss 0.0454 0.7515 4.5193 18.8171 | lr 3.3e-04 | norm 1.1801 | dt 0.025 | |
| type train | step 13450 | loss 0.0454 0.7280 4.2986 17.8033 | lr 3.3e-04 | norm 1.2206 | dt 0.026 | |
| type train | step 13460 | loss 0.0466 0.7689 4.5155 18.6270 | lr 3.3e-04 | norm 1.4593 | dt 0.026 | |
| type train | step 13470 | loss 0.0436 0.7456 4.3509 17.6916 | lr 3.3e-04 | norm 1.2215 | dt 0.026 | |
| type train | step 13480 | loss 0.0438 0.7360 4.4000 18.3988 | lr 3.3e-04 | norm 1.1511 | dt 0.026 | |
| type train | step 13490 | loss 0.0446 0.7426 4.3446 17.7765 | lr 3.3e-04 | norm 0.9943 | dt 0.026 | |
| type train | step 13500 | loss 0.0472 0.7758 4.6868 19.3549 | lr 3.3e-04 | norm 1.2567 | dt 0.026 | |
| type train | step 13510 | loss 0.0443 0.7431 4.3825 18.0705 | lr 3.3e-04 | norm 1.2528 | dt 0.026 | |
| type train | step 13520 | loss 0.0441 0.7164 4.2723 17.7528 | lr 3.3e-04 | norm 1.0555 | dt 0.026 | |
| type train | step 13530 | loss 0.0454 0.7559 4.4418 18.1552 | lr 3.3e-04 | norm 1.1863 | dt 0.026 | |
| type train | step 13540 | loss 0.0459 0.7550 4.3113 17.8257 | lr 3.3e-04 | norm 1.2279 | dt 0.026 | |
| type train | step 13550 | loss 0.0444 0.7415 4.3003 17.9759 | lr 3.3e-04 | norm 1.0943 | dt 0.025 | |
| type train | step 13560 | loss 0.0483 0.7483 4.3597 18.3255 | lr 3.3e-04 | norm 1.8311 | dt 0.026 | |
| type train | step 13570 | loss 0.0457 0.7627 4.4593 18.4190 | lr 3.3e-04 | norm 1.3492 | dt 0.026 | |
| type train | step 13580 | loss 0.0457 0.7349 4.4329 18.3137 | lr 3.3e-04 | norm 1.1707 | dt 0.026 | |
| type train | step 13590 | loss 0.0436 0.7495 4.3973 17.9035 | lr 3.2e-04 | norm 1.2688 | dt 0.026 | |
| type train | step 13600 | loss 0.0452 0.7530 4.3678 18.1475 | lr 3.2e-04 | norm 1.3152 | dt 0.026 | |
| type train | step 13610 | loss 0.0450 0.7425 4.3505 18.0453 | lr 3.2e-04 | norm 1.5093 | dt 0.025 | |
| type train | step 13620 | loss 0.0474 0.7648 4.5471 18.7704 | lr 3.2e-04 | norm 1.2917 | dt 0.026 | |
| type train | step 13630 | loss 0.0465 0.7412 4.3502 18.1449 | lr 3.2e-04 | norm 1.2261 | dt 0.026 | |
| type train | step 13640 | loss 0.0450 0.7299 4.4294 18.2138 | lr 3.2e-04 | norm 1.1379 | dt 0.026 | |
| type train | step 13650 | loss 0.0436 0.7451 4.3864 18.0345 | lr 3.2e-04 | norm 1.1073 | dt 0.026 | |
| type train | step 13660 | loss 0.0455 0.7780 4.5273 18.5339 | lr 3.2e-04 | norm 1.3508 | dt 0.026 | |
| type train | step 13670 | loss 0.0446 0.7344 4.2721 17.6821 | lr 3.2e-04 | norm 1.1052 | dt 0.026 | |
| type train | step 13680 | loss 0.0443 0.7302 4.3712 18.1354 | lr 3.2e-04 | norm 1.1418 | dt 0.027 | |
| type train | step 13690 | loss 0.0460 0.7324 4.2929 17.9207 | lr 3.2e-04 | norm 1.0695 | dt 0.026 | |
| type train | step 13700 | loss 0.0465 0.7457 4.4654 18.3503 | lr 3.2e-04 | norm 1.3206 | dt 0.026 | |
| type train | step 13710 | loss 0.0447 0.7394 4.2753 17.8255 | lr 3.2e-04 | norm 1.2621 | dt 0.026 | |
| type train | step 13720 | loss 0.0455 0.7691 4.4310 18.0057 | lr 3.2e-04 | norm 1.3506 | dt 0.026 | |
| type train | step 13730 | loss 0.0457 0.7440 4.3626 18.1254 | lr 3.2e-04 | norm 1.3621 | dt 0.026 | |
| type train | step 13740 | loss 0.0455 0.7297 4.3915 18.1512 | lr 3.2e-04 | norm 1.3505 | dt 0.026 | |
| type train | step 13750 | loss 0.0465 0.7440 4.4772 18.6532 | lr 3.1e-04 | norm 1.0686 | dt 0.026 | |
| type train | step 13760 | loss 0.0447 0.7456 4.4439 18.4538 | lr 3.1e-04 | norm 1.0761 | dt 0.026 | |
| type train | step 13770 | loss 0.0442 0.7506 4.4051 18.1960 | lr 3.1e-04 | norm 1.2874 | dt 0.026 | |
| type train | step 13780 | loss 0.0446 0.7265 4.2252 17.6262 | lr 3.1e-04 | norm 1.3010 | dt 0.026 | |
| type train | step 13790 | loss 0.0449 0.7443 4.2855 18.0616 | lr 3.1e-04 | norm 1.3393 | dt 0.026 | |
| type train | step 13800 | loss 0.0467 0.7440 4.4125 18.4214 | lr 3.1e-04 | norm 1.3282 | dt 0.026 | |
| type train | step 13810 | loss 0.0447 0.7298 4.5008 18.6099 | lr 3.1e-04 | norm 1.1305 | dt 0.025 | |
| type train | step 13820 | loss 0.0462 0.7490 4.3536 17.8435 | lr 3.1e-04 | norm 1.2619 | dt 0.026 | |
| type train | step 13830 | loss 0.0447 0.7483 4.3970 18.0662 | lr 3.1e-04 | norm 1.0669 | dt 0.026 | |
| type train | step 13840 | loss 0.0449 0.7600 4.3300 17.9741 | lr 3.1e-04 | norm 1.3671 | dt 0.026 | |
| type train | step 13850 | loss 0.0449 0.7549 4.4613 18.3839 | lr 3.1e-04 | norm 1.1156 | dt 0.026 | |
| type train | step 13860 | loss 0.0454 0.7374 4.2986 17.9476 | lr 3.1e-04 | norm 1.3622 | dt 0.026 | |
| type train | step 13870 | loss 0.0465 0.7533 4.4627 18.3703 | lr 3.1e-04 | norm 1.0637 | dt 0.026 | |
| type train | step 13880 | loss 0.0447 0.7308 4.1932 17.4181 | lr 3.1e-04 | norm 1.2732 | dt 0.026 | |
| type train | step 13890 | loss 0.0465 0.7330 4.3919 18.2119 | lr 3.1e-04 | norm 1.1469 | dt 0.026 | |
| type train | step 13900 | loss 0.0452 0.7502 4.4600 18.3168 | lr 3.1e-04 | norm 1.1226 | dt 0.026 | |
| type train | step 13910 | loss 0.0448 0.7575 4.5012 18.4635 | lr 3.0e-04 | norm 1.2281 | dt 0.026 | |
| type train | step 13920 | loss 0.0469 0.7451 4.3873 18.2927 | lr 3.0e-04 | norm 1.3315 | dt 0.026 | |
| type train | step 13930 | loss 0.0458 0.7499 4.5592 18.9551 | lr 3.0e-04 | norm 1.1941 | dt 0.026 | |
| type train | step 13940 | loss 0.0458 0.7485 4.4975 18.6421 | lr 3.0e-04 | norm 1.3632 | dt 0.026 | |
| type train | step 13950 | loss 0.0461 0.7510 4.4920 18.8556 | lr 3.0e-04 | norm 1.2064 | dt 0.026 | |
| type train | step 13960 | loss 0.0435 0.7378 4.3116 17.7186 | lr 3.0e-04 | norm 1.1158 | dt 0.026 | |
| type train | step 13970 | loss 0.0434 0.7530 4.3826 18.2943 | lr 3.0e-04 | norm 1.2679 | dt 0.027 | |
| type train | step 13980 | loss 0.0464 0.7420 4.3633 18.1232 | lr 3.0e-04 | norm 1.4939 | dt 0.026 | |
| type train | step 13990 | loss 0.0451 0.7481 4.5895 19.2569 | lr 3.0e-04 | norm 1.2191 | dt 0.026 | |
| type train | step 14000 | loss 0.0452 0.7453 4.3591 17.7687 | lr 3.0e-04 | norm 1.4762 | dt 0.026 | |
| type train | step 14010 | loss 0.0452 0.7501 4.4696 18.6501 | lr 3.0e-04 | norm 1.3067 | dt 0.026 | |
| type train | step 14020 | loss 0.0438 0.7427 4.3580 17.8424 | lr 3.0e-04 | norm 1.1353 | dt 0.026 | |
| type train | step 14030 | loss 0.0439 0.7575 4.6355 18.9731 | lr 3.0e-04 | norm 1.4195 | dt 0.028 | |
| type train | step 14040 | loss 0.0444 0.7437 4.4286 18.2961 | lr 3.0e-04 | norm 1.0999 | dt 0.026 | |
| type train | step 14050 | loss 0.0451 0.7513 4.5208 18.8124 | lr 3.0e-04 | norm 1.1930 | dt 0.026 | |
| type train | step 14060 | loss 0.0450 0.7266 4.2956 17.7919 | lr 3.0e-04 | norm 1.2122 | dt 0.026 | |
| type train | step 14070 | loss 0.0464 0.7676 4.5109 18.6359 | lr 2.9e-04 | norm 1.4461 | dt 0.026 | |
| type train | step 14080 | loss 0.0433 0.7447 4.3449 17.6936 | lr 2.9e-04 | norm 1.2124 | dt 0.026 | |
| type train | step 14090 | loss 0.0435 0.7357 4.3966 18.3954 | lr 2.9e-04 | norm 1.1435 | dt 0.026 | |
| type train | step 14100 | loss 0.0443 0.7422 4.3391 17.7743 | lr 2.9e-04 | norm 0.9870 | dt 0.025 | |
| type train | step 14110 | loss 0.0470 0.7753 4.6834 19.3388 | lr 2.9e-04 | norm 1.2516 | dt 0.026 | |
| type train | step 14120 | loss 0.0440 0.7429 4.3794 18.0692 | lr 2.9e-04 | norm 1.2464 | dt 0.026 | |
| type train | step 14130 | loss 0.0439 0.7162 4.2680 17.7411 | lr 2.9e-04 | norm 1.0451 | dt 0.026 | |
| type train | step 14140 | loss 0.0451 0.7544 4.4385 18.1483 | lr 2.9e-04 | norm 1.1726 | dt 0.026 | |
| type train | step 14150 | loss 0.0458 0.7537 4.3086 17.8202 | lr 2.9e-04 | norm 1.2239 | dt 0.026 | |
| type train | step 14160 | loss 0.0441 0.7414 4.2970 17.9636 | lr 2.9e-04 | norm 1.0854 | dt 0.026 | |
| type train | step 14170 | loss 0.0481 0.7469 4.3578 18.3171 | lr 2.9e-04 | norm 1.8157 | dt 0.027 | |
| type train | step 14180 | loss 0.0454 0.7615 4.4536 18.4199 | lr 2.9e-04 | norm 1.3331 | dt 0.026 | |
| type train | step 14190 | loss 0.0454 0.7340 4.4311 18.3107 | lr 2.9e-04 | norm 1.1609 | dt 0.026 | |
| type train | step 14200 | loss 0.0434 0.7472 4.3955 17.8974 | lr 2.9e-04 | norm 1.2563 | dt 0.026 | |
| type train | step 14210 | loss 0.0449 0.7514 4.3665 18.1410 | lr 2.9e-04 | norm 1.3049 | dt 0.026 | |
| type train | step 14220 | loss 0.0445 0.7426 4.3521 18.0333 | lr 2.9e-04 | norm 1.4889 | dt 0.026 | |
| type train | step 14230 | loss 0.0472 0.7641 4.5406 18.7656 | lr 2.9e-04 | norm 1.2950 | dt 0.026 | |
| type train | step 14240 | loss 0.0462 0.7399 4.3442 18.1347 | lr 2.8e-04 | norm 1.2199 | dt 0.026 | |
| type train | step 14250 | loss 0.0450 0.7298 4.4283 18.2048 | lr 2.8e-04 | norm 1.1341 | dt 0.026 | |
| type train | step 14260 | loss 0.0436 0.7441 4.3832 18.0383 | lr 2.8e-04 | norm 1.0992 | dt 0.025 | |
| type train | step 14270 | loss 0.0453 0.7763 4.5257 18.5348 | lr 2.8e-04 | norm 1.3518 | dt 0.025 | |
| type train | step 14280 | loss 0.0443 0.7340 4.2701 17.6643 | lr 2.8e-04 | norm 1.0931 | dt 0.026 | |
| type train | step 14290 | loss 0.0442 0.7298 4.3665 18.1334 | lr 2.8e-04 | norm 1.1420 | dt 0.026 | |
| type train | step 14300 | loss 0.0460 0.7322 4.2928 17.9146 | lr 2.8e-04 | norm 1.0682 | dt 0.026 | |
| type train | step 14310 | loss 0.0466 0.7449 4.4605 18.3400 | lr 2.8e-04 | norm 1.3239 | dt 0.026 | |
| type train | step 14320 | loss 0.0446 0.7391 4.2739 17.8161 | lr 2.8e-04 | norm 1.2470 | dt 0.026 | |
| type train | step 14330 | loss 0.0454 0.7678 4.4250 18.0008 | lr 2.8e-04 | norm 1.3523 | dt 0.026 | |
| type train | step 14340 | loss 0.0453 0.7430 4.3614 18.1149 | lr 2.8e-04 | norm 1.3584 | dt 0.026 | |
| type train | step 14350 | loss 0.0456 0.7295 4.3895 18.1486 | lr 2.8e-04 | norm 1.3590 | dt 0.026 | |
| type train | step 14360 | loss 0.0461 0.7425 4.4736 18.6577 | lr 2.8e-04 | norm 1.0526 | dt 0.026 | |
| type train | step 14370 | loss 0.0446 0.7439 4.4447 18.4480 | lr 2.8e-04 | norm 1.0645 | dt 0.026 | |
| type train | step 14380 | loss 0.0441 0.7495 4.4022 18.1858 | lr 2.8e-04 | norm 1.2827 | dt 0.026 | |
| type train | step 14390 | loss 0.0444 0.7248 4.2218 17.6207 | lr 2.8e-04 | norm 1.3039 | dt 0.026 | |
| type train | step 14400 | loss 0.0446 0.7442 4.2793 18.0570 | lr 2.8e-04 | norm 1.3278 | dt 0.026 | |
| type train | step 14410 | loss 0.0465 0.7429 4.4101 18.4002 | lr 2.7e-04 | norm 1.3257 | dt 0.026 | |
| type train | step 14420 | loss 0.0446 0.7289 4.4998 18.6045 | lr 2.7e-04 | norm 1.1410 | dt 0.026 | |
| type train | step 14430 | loss 0.0461 0.7493 4.3517 17.8376 | lr 2.7e-04 | norm 1.2656 | dt 0.026 | |
| type train | step 14440 | loss 0.0446 0.7477 4.3934 18.0604 | lr 2.7e-04 | norm 1.0519 | dt 0.026 | |
| type train | step 14450 | loss 0.0446 0.7595 4.3291 17.9688 | lr 2.7e-04 | norm 1.3794 | dt 0.026 | |
| type train | step 14460 | loss 0.0446 0.7542 4.4541 18.3718 | lr 2.7e-04 | norm 1.1065 | dt 0.026 | |
| type train | step 14470 | loss 0.0451 0.7367 4.2942 17.9389 | lr 2.7e-04 | norm 1.3461 | dt 0.026 | |
| type train | step 14480 | loss 0.0463 0.7526 4.4603 18.3728 | lr 2.7e-04 | norm 1.0585 | dt 0.026 | |
| type train | step 14490 | loss 0.0444 0.7299 4.1919 17.4075 | lr 2.7e-04 | norm 1.2538 | dt 0.026 | |
| type train | step 14500 | loss 0.0463 0.7326 4.3887 18.2112 | lr 2.7e-04 | norm 1.1561 | dt 0.026 | |
| type train | step 14510 | loss 0.0450 0.7497 4.4554 18.3132 | lr 2.7e-04 | norm 1.1245 | dt 0.025 | |
| type train | step 14520 | loss 0.0446 0.7568 4.4936 18.4537 | lr 2.7e-04 | norm 1.2151 | dt 0.025 | |
| type train | step 14530 | loss 0.0467 0.7446 4.3858 18.2953 | lr 2.7e-04 | norm 1.3247 | dt 0.026 | |
| type train | step 14540 | loss 0.0455 0.7485 4.5590 18.9659 | lr 2.7e-04 | norm 1.1858 | dt 0.025 | |
| type train | step 14550 | loss 0.0457 0.7485 4.4927 18.6395 | lr 2.7e-04 | norm 1.3752 | dt 0.026 | |
| type train | step 14560 | loss 0.0461 0.7511 4.4923 18.8523 | lr 2.7e-04 | norm 1.2137 | dt 0.026 | |
| type train | step 14570 | loss 0.0432 0.7377 4.3109 17.7183 | lr 2.7e-04 | norm 1.1136 | dt 0.026 | |
| type train | step 14580 | loss 0.0430 0.7528 4.3795 18.2795 | lr 2.6e-04 | norm 1.2499 | dt 0.026 | |
| type train | step 14590 | loss 0.0463 0.7412 4.3575 18.1113 | lr 2.6e-04 | norm 1.4858 | dt 0.026 | |
| type train | step 14600 | loss 0.0447 0.7475 4.5855 19.2478 | lr 2.6e-04 | norm 1.2089 | dt 0.026 | |
| type train | step 14610 | loss 0.0450 0.7450 4.3562 17.7633 | lr 2.6e-04 | norm 1.4822 | dt 0.026 | |
| type train | step 14620 | loss 0.0452 0.7498 4.4708 18.6425 | lr 2.6e-04 | norm 1.3059 | dt 0.026 | |
| type train | step 14630 | loss 0.0436 0.7423 4.3573 17.8324 | lr 2.6e-04 | norm 1.1292 | dt 0.026 | |
| type train | step 14640 | loss 0.0440 0.7567 4.6360 18.9700 | lr 2.6e-04 | norm 1.4130 | dt 0.027 | |
| type train | step 14650 | loss 0.0441 0.7434 4.4272 18.2950 | lr 2.6e-04 | norm 1.0863 | dt 0.026 | |
| type train | step 14660 | loss 0.0449 0.7512 4.5181 18.8059 | lr 2.6e-04 | norm 1.1823 | dt 0.026 | |
| type train | step 14670 | loss 0.0448 0.7265 4.2922 17.7847 | lr 2.6e-04 | norm 1.2035 | dt 0.026 | |
| type train | step 14680 | loss 0.0462 0.7683 4.5065 18.6269 | lr 2.6e-04 | norm 1.4297 | dt 0.026 | |
| type train | step 14690 | loss 0.0432 0.7438 4.3414 17.6978 | lr 2.6e-04 | norm 1.2052 | dt 0.026 | |
| type train | step 14700 | loss 0.0434 0.7345 4.3939 18.4035 | lr 2.6e-04 | norm 1.1253 | dt 0.025 | |
| type train | step 14710 | loss 0.0442 0.7426 4.3344 17.7761 | lr 2.6e-04 | norm 0.9824 | dt 0.025 | |
| type train | step 14720 | loss 0.0468 0.7749 4.6814 19.3181 | lr 2.6e-04 | norm 1.2560 | dt 0.026 | |
| type train | step 14730 | loss 0.0438 0.7426 4.3762 18.0669 | lr 2.6e-04 | norm 1.2377 | dt 0.025 | |
| type train | step 14740 | loss 0.0436 0.7155 4.2683 17.7393 | lr 2.6e-04 | norm 1.0357 | dt 0.026 | |
| type train | step 14750 | loss 0.0451 0.7546 4.4361 18.1497 | lr 2.6e-04 | norm 1.1738 | dt 0.026 | |
| type train | step 14760 | loss 0.0456 0.7537 4.3079 17.8103 | lr 2.5e-04 | norm 1.2131 | dt 0.026 | |
| type train | step 14770 | loss 0.0440 0.7411 4.2905 17.9747 | lr 2.5e-04 | norm 1.0851 | dt 0.026 | |
| type train | step 14780 | loss 0.0478 0.7458 4.3522 18.3175 | lr 2.5e-04 | norm 1.8086 | dt 0.026 | |
| type train | step 14790 | loss 0.0452 0.7614 4.4528 18.4182 | lr 2.5e-04 | norm 1.3323 | dt 0.026 | |
| type train | step 14800 | loss 0.0452 0.7328 4.4298 18.2994 | lr 2.5e-04 | norm 1.1358 | dt 0.026 | |
| type train | step 14810 | loss 0.0434 0.7470 4.3929 17.8975 | lr 2.5e-04 | norm 1.2491 | dt 0.026 | |
| type train | step 14820 | loss 0.0448 0.7524 4.3653 18.1327 | lr 2.5e-04 | norm 1.3051 | dt 0.026 | |
| type train | step 14830 | loss 0.0444 0.7430 4.3513 18.0331 | lr 2.5e-04 | norm 1.4780 | dt 0.026 | |
| type train | step 14840 | loss 0.0469 0.7636 4.5415 18.7599 | lr 2.5e-04 | norm 1.2934 | dt 0.026 | |
| type train | step 14850 | loss 0.0459 0.7387 4.3425 18.1339 | lr 2.5e-04 | norm 1.2133 | dt 0.026 | |
| type train | step 14860 | loss 0.0447 0.7292 4.4259 18.1992 | lr 2.5e-04 | norm 1.1164 | dt 0.026 | |
| type train | step 14870 | loss 0.0434 0.7430 4.3815 18.0274 | lr 2.5e-04 | norm 1.0875 | dt 0.026 | |
| type train | step 14880 | loss 0.0453 0.7763 4.5270 18.5295 | lr 2.5e-04 | norm 1.3563 | dt 0.026 | |
| type train | step 14890 | loss 0.0443 0.7338 4.2712 17.6545 | lr 2.5e-04 | norm 1.0975 | dt 0.026 | |
| type train | step 14900 | loss 0.0439 0.7301 4.3636 18.1317 | lr 2.5e-04 | norm 1.1399 | dt 0.026 | |
| type train | step 14910 | loss 0.0457 0.7318 4.2927 17.9140 | lr 2.5e-04 | norm 1.0628 | dt 0.026 | |
| type train | step 14920 | loss 0.0462 0.7433 4.4547 18.3204 | lr 2.5e-04 | norm 1.2888 | dt 0.026 | |
| type train | step 14930 | loss 0.0443 0.7390 4.2709 17.8062 | lr 2.5e-04 | norm 1.2365 | dt 0.026 | |
| type train | step 14940 | loss 0.0452 0.7675 4.4275 17.9910 | lr 2.4e-04 | norm 1.3602 | dt 0.026 | |
| type train | step 14950 | loss 0.0452 0.7426 4.3605 18.1073 | lr 2.4e-04 | norm 1.3625 | dt 0.026 | |
| type train | step 14960 | loss 0.0452 0.7288 4.3884 18.1426 | lr 2.4e-04 | norm 1.3382 | dt 0.026 | |
| type train | step 14970 | loss 0.0458 0.7418 4.4723 18.6550 | lr 2.4e-04 | norm 1.0495 | dt 0.026 | |
| type train | step 14980 | loss 0.0443 0.7444 4.4430 18.4437 | lr 2.4e-04 | norm 1.0615 | dt 0.025 | |
| type train | step 14990 | loss 0.0438 0.7485 4.3987 18.1833 | lr 2.4e-04 | norm 1.2673 | dt 0.026 | |
| type train | step 15000 | loss 0.0442 0.7245 4.2220 17.6196 | lr 2.4e-04 | norm 1.3017 | dt 0.026 | |
| type train | step 15010 | loss 0.0444 0.7438 4.2777 18.0515 | lr 2.4e-04 | norm 1.3374 | dt 0.026 | |
| type train | step 15020 | loss 0.0462 0.7414 4.4079 18.3930 | lr 2.4e-04 | norm 1.3082 | dt 0.026 | |
| type train | step 15030 | loss 0.0441 0.7281 4.4994 18.6012 | lr 2.4e-04 | norm 1.1377 | dt 0.026 | |
| type train | step 15040 | loss 0.0458 0.7496 4.3523 17.8318 | lr 2.4e-04 | norm 1.2600 | dt 0.026 | |
| type train | step 15050 | loss 0.0443 0.7472 4.3905 18.0646 | lr 2.4e-04 | norm 1.0397 | dt 0.026 | |
| type train | step 15060 | loss 0.0444 0.7589 4.3282 17.9633 | lr 2.4e-04 | norm 1.3689 | dt 0.026 | |
| type train | step 15070 | loss 0.0443 0.7533 4.4527 18.3609 | lr 2.4e-04 | norm 1.1057 | dt 0.026 | |
| type train | step 15080 | loss 0.0450 0.7363 4.2932 17.9346 | lr 2.4e-04 | norm 1.3477 | dt 0.026 | |
| type train | step 15090 | loss 0.0459 0.7520 4.4588 18.3773 | lr 2.4e-04 | norm 1.0576 | dt 0.025 | |
| type train | step 15100 | loss 0.0442 0.7301 4.1912 17.3950 | lr 2.4e-04 | norm 1.2524 | dt 0.026 | |
| type train | step 15110 | loss 0.0461 0.7327 4.3908 18.2122 | lr 2.4e-04 | norm 1.1426 | dt 0.026 | |
| type train | step 15120 | loss 0.0448 0.7488 4.4515 18.3085 | lr 2.4e-04 | norm 1.1207 | dt 0.026 | |
| type train | step 15130 | loss 0.0442 0.7564 4.4925 18.4559 | lr 2.3e-04 | norm 1.2051 | dt 0.025 | |
| type train | step 15140 | loss 0.0467 0.7444 4.3806 18.2899 | lr 2.3e-04 | norm 1.3211 | dt 0.025 | |
| type train | step 15150 | loss 0.0452 0.7481 4.5570 18.9617 | lr 2.3e-04 | norm 1.1872 | dt 0.026 | |
| type train | step 15160 | loss 0.0453 0.7477 4.4879 18.6315 | lr 2.3e-04 | norm 1.3673 | dt 0.026 | |
| type train | step 15170 | loss 0.0459 0.7509 4.4886 18.8466 | lr 2.3e-04 | norm 1.2009 | dt 0.026 | |
| type train | step 15180 | loss 0.0430 0.7376 4.3123 17.7098 | lr 2.3e-04 | norm 1.1099 | dt 0.027 | |
| type train | step 15190 | loss 0.0427 0.7525 4.3770 18.2744 | lr 2.3e-04 | norm 1.2343 | dt 0.026 | |
| type train | step 15200 | loss 0.0462 0.7411 4.3519 18.0981 | lr 2.3e-04 | norm 1.4787 | dt 0.026 | |
| type train | step 15210 | loss 0.0445 0.7466 4.5845 19.2556 | lr 2.3e-04 | norm 1.2091 | dt 0.025 | |
| type train | step 15220 | loss 0.0448 0.7443 4.3508 17.7624 | lr 2.3e-04 | norm 1.4752 | dt 0.026 | |
| type train | step 15230 | loss 0.0450 0.7497 4.4679 18.6417 | lr 2.3e-04 | norm 1.2964 | dt 0.026 | |
| type train | step 15240 | loss 0.0435 0.7422 4.3578 17.8264 | lr 2.3e-04 | norm 1.1363 | dt 0.026 | |
| type train | step 15250 | loss 0.0435 0.7558 4.6334 18.9647 | lr 2.3e-04 | norm 1.3934 | dt 0.029 | |
| type train | step 15260 | loss 0.0438 0.7430 4.4237 18.2939 | lr 2.3e-04 | norm 1.0810 | dt 0.027 | |
| type train | step 15270 | loss 0.0446 0.7504 4.5149 18.7938 | lr 2.3e-04 | norm 1.1742 | dt 0.026 | |
| type train | step 15280 | loss 0.0447 0.7261 4.2909 17.7745 | lr 2.3e-04 | norm 1.1978 | dt 0.026 | |
| type train | step 15290 | loss 0.0460 0.7671 4.5052 18.6191 | lr 2.3e-04 | norm 1.4147 | dt 0.026 | |
| type train | step 15300 | loss 0.0430 0.7439 4.3392 17.7050 | lr 2.3e-04 | norm 1.2037 | dt 0.025 | |
| type train | step 15310 | loss 0.0434 0.7348 4.3946 18.4023 | lr 2.3e-04 | norm 1.1119 | dt 0.026 | |
| type train | step 15320 | loss 0.0440 0.7422 4.3322 17.7739 | lr 2.2e-04 | norm 0.9803 | dt 0.026 | |
| type train | step 15330 | loss 0.0463 0.7754 4.6773 19.3074 | lr 2.2e-04 | norm 1.2427 | dt 0.025 | |
| type train | step 15340 | loss 0.0436 0.7429 4.3706 18.0617 | lr 2.2e-04 | norm 1.2204 | dt 0.026 | |
| type train | step 15350 | loss 0.0434 0.7147 4.2672 17.7303 | lr 2.2e-04 | norm 1.0380 | dt 0.026 | |
| type train | step 15360 | loss 0.0448 0.7542 4.4340 18.1396 | lr 2.2e-04 | norm 1.1667 | dt 0.025 | |
| type train | step 15370 | loss 0.0454 0.7532 4.3102 17.8090 | lr 2.2e-04 | norm 1.2095 | dt 0.026 | |
| type train | step 15380 | loss 0.0437 0.7408 4.2876 17.9692 | lr 2.2e-04 | norm 1.0875 | dt 0.026 | |
| type train | step 15390 | loss 0.0476 0.7454 4.3506 18.3038 | lr 2.2e-04 | norm 1.8027 | dt 0.026 | |
| type train | step 15400 | loss 0.0448 0.7603 4.4521 18.4200 | lr 2.2e-04 | norm 1.3107 | dt 0.026 | |
| type train | step 15410 | loss 0.0451 0.7324 4.4299 18.3029 | lr 2.2e-04 | norm 1.1281 | dt 0.026 | |
| type train | step 15420 | loss 0.0431 0.7461 4.3918 17.8927 | lr 2.2e-04 | norm 1.2399 | dt 0.026 | |
| type train | step 15430 | loss 0.0446 0.7511 4.3667 18.1299 | lr 2.2e-04 | norm 1.3054 | dt 0.026 | |
| type train | step 15440 | loss 0.0442 0.7432 4.3491 18.0281 | lr 2.2e-04 | norm 1.4693 | dt 0.026 | |
| type train | step 15450 | loss 0.0466 0.7628 4.5358 18.7588 | lr 2.2e-04 | norm 1.2836 | dt 0.026 | |
| type train | step 15460 | loss 0.0456 0.7380 4.3389 18.1256 | lr 2.2e-04 | norm 1.2001 | dt 0.026 | |
| type train | step 15470 | loss 0.0444 0.7286 4.4241 18.1972 | lr 2.2e-04 | norm 1.1056 | dt 0.026 | |
| type train | step 15480 | loss 0.0431 0.7427 4.3775 18.0211 | lr 2.2e-04 | norm 1.0862 | dt 0.026 | |
| type train | step 15490 | loss 0.0449 0.7756 4.5260 18.5192 | lr 2.2e-04 | norm 1.3582 | dt 0.026 | |
| type train | step 15500 | loss 0.0440 0.7339 4.2703 17.6609 | lr 2.2e-04 | norm 1.1131 | dt 0.026 | |
| type train | step 15510 | loss 0.0436 0.7295 4.3598 18.1346 | lr 2.2e-04 | norm 1.1267 | dt 0.025 | |
| type train | step 15520 | loss 0.0454 0.7309 4.2923 17.9096 | lr 2.2e-04 | norm 1.0577 | dt 0.026 | |
| type train | step 15530 | loss 0.0460 0.7420 4.4536 18.3231 | lr 2.1e-04 | norm 1.2880 | dt 0.026 | |
| type train | step 15540 | loss 0.0442 0.7373 4.2683 17.7997 | lr 2.1e-04 | norm 1.2307 | dt 0.026 | |
| type train | step 15550 | loss 0.0449 0.7670 4.4257 17.9781 | lr 2.1e-04 | norm 1.3600 | dt 0.027 | |
| type train | step 15560 | loss 0.0450 0.7416 4.3555 18.1054 | lr 2.1e-04 | norm 1.3611 | dt 0.026 | |
| type train | step 15570 | loss 0.0450 0.7286 4.3850 18.1440 | lr 2.1e-04 | norm 1.3295 | dt 0.026 | |
| type train | step 15580 | loss 0.0456 0.7411 4.4744 18.6519 | lr 2.1e-04 | norm 1.0567 | dt 0.026 | |
| type train | step 15590 | loss 0.0441 0.7434 4.4504 18.4416 | lr 2.1e-04 | norm 1.0589 | dt 0.026 | |
| type train | step 15600 | loss 0.0437 0.7473 4.3960 18.1832 | lr 2.1e-04 | norm 1.2703 | dt 0.026 | |
| type train | step 15610 | loss 0.0441 0.7235 4.2183 17.6225 | lr 2.1e-04 | norm 1.2976 | dt 0.026 | |
| type train | step 15620 | loss 0.0442 0.7439 4.2766 18.0518 | lr 2.1e-04 | norm 1.3379 | dt 0.025 | |
| type train | step 15630 | loss 0.0460 0.7416 4.4046 18.3812 | lr 2.1e-04 | norm 1.3012 | dt 0.026 | |
| type train | step 15640 | loss 0.0439 0.7281 4.4971 18.6018 | lr 2.1e-04 | norm 1.1447 | dt 0.026 | |
| type train | step 15650 | loss 0.0456 0.7488 4.3511 17.8323 | lr 2.1e-04 | norm 1.2627 | dt 0.026 | |
| type train | step 15660 | loss 0.0441 0.7468 4.3847 18.0683 | lr 2.1e-04 | norm 1.0299 | dt 0.025 | |
| type train | step 15670 | loss 0.0440 0.7582 4.3261 17.9615 | lr 2.1e-04 | norm 1.3637 | dt 0.026 | |
| type train | step 15680 | loss 0.0441 0.7529 4.4496 18.3650 | lr 2.1e-04 | norm 1.1045 | dt 0.026 | |
| type train | step 15690 | loss 0.0445 0.7362 4.2902 17.9276 | lr 2.1e-04 | norm 1.3354 | dt 0.026 | |
| type train | step 15700 | loss 0.0457 0.7518 4.4542 18.3763 | lr 2.1e-04 | norm 1.0588 | dt 0.026 | |
| type train | step 15710 | loss 0.0440 0.7290 4.1891 17.3987 | lr 2.1e-04 | norm 1.2504 | dt 0.025 | |
| type train | step 15720 | loss 0.0459 0.7333 4.3920 18.2084 | lr 2.1e-04 | norm 1.1407 | dt 0.026 | |
| type train | step 15730 | loss 0.0443 0.7481 4.4484 18.3100 | lr 2.0e-04 | norm 1.0990 | dt 0.026 | |
| type train | step 15740 | loss 0.0441 0.7553 4.4883 18.4564 | lr 2.0e-04 | norm 1.2092 | dt 0.025 | |
| type train | step 15750 | loss 0.0462 0.7438 4.3777 18.2910 | lr 2.0e-04 | norm 1.3064 | dt 0.026 | |
| type train | step 15760 | loss 0.0450 0.7480 4.5574 18.9634 | lr 2.0e-04 | norm 1.1960 | dt 0.026 | |
| type train | step 15770 | loss 0.0452 0.7469 4.4863 18.6364 | lr 2.0e-04 | norm 1.3718 | dt 0.026 | |
| type train | step 15780 | loss 0.0456 0.7504 4.4898 18.8477 | lr 2.0e-04 | norm 1.1869 | dt 0.026 | |
| type train | step 15790 | loss 0.0428 0.7365 4.3105 17.7026 | lr 2.0e-04 | norm 1.1049 | dt 0.025 | |
| type train | step 15800 | loss 0.0426 0.7518 4.3706 18.2739 | lr 2.0e-04 | norm 1.2239 | dt 0.026 | |
| type train | step 15810 | loss 0.0458 0.7405 4.3461 18.0915 | lr 2.0e-04 | norm 1.4551 | dt 0.026 | |
| type train | step 15820 | loss 0.0442 0.7462 4.5825 19.2585 | lr 2.0e-04 | norm 1.2152 | dt 0.026 | |
| type train | step 15830 | loss 0.0445 0.7440 4.3449 17.7612 | lr 2.0e-04 | norm 1.4543 | dt 0.026 | |
| type train | step 15840 | loss 0.0447 0.7492 4.4683 18.6442 | lr 2.0e-04 | norm 1.2897 | dt 0.026 | |
| type train | step 15850 | loss 0.0431 0.7409 4.3555 17.8169 | lr 2.0e-04 | norm 1.1204 | dt 0.026 | |
| type train | step 15860 | loss 0.0434 0.7548 4.6318 18.9709 | lr 2.0e-04 | norm 1.3873 | dt 0.027 | |
| type train | step 15870 | loss 0.0436 0.7432 4.4218 18.2975 | lr 2.0e-04 | norm 1.0538 | dt 0.026 | |
| type train | step 15880 | loss 0.0444 0.7496 4.5129 18.7950 | lr 2.0e-04 | norm 1.1729 | dt 0.026 | |
| type train | step 15890 | loss 0.0445 0.7257 4.2859 17.7860 | lr 2.0e-04 | norm 1.1991 | dt 0.026 | |
| type train | step 15900 | loss 0.0457 0.7661 4.5020 18.6151 | lr 2.0e-04 | norm 1.3972 | dt 0.026 | |
| type train | step 15910 | loss 0.0428 0.7425 4.3383 17.7029 | lr 2.0e-04 | norm 1.1965 | dt 0.026 | |
| type train | step 15920 | loss 0.0430 0.7335 4.3931 18.4114 | lr 2.0e-04 | norm 1.0844 | dt 0.026 | |
| type train | step 15930 | loss 0.0437 0.7422 4.3295 17.7813 | lr 2.0e-04 | norm 0.9605 | dt 0.026 | |
| type train | step 15940 | loss 0.0462 0.7750 4.6760 19.2968 | lr 2.0e-04 | norm 1.2447 | dt 0.026 | |
| type train | step 15950 | loss 0.0433 0.7420 4.3666 18.0595 | lr 1.9e-04 | norm 1.1957 | dt 0.026 | |
| type train | step 15960 | loss 0.0432 0.7144 4.2650 17.7379 | lr 1.9e-04 | norm 1.0208 | dt 0.026 | |
| type train | step 15970 | loss 0.0446 0.7534 4.4319 18.1400 | lr 1.9e-04 | norm 1.1552 | dt 0.026 | |
| type train | step 15980 | loss 0.0452 0.7524 4.3092 17.7995 | lr 1.9e-04 | norm 1.1994 | dt 0.026 | |
| type train | step 15990 | loss 0.0435 0.7404 4.2853 17.9768 | lr 1.9e-04 | norm 1.0781 | dt 0.026 | |
| type train | step 16000 | loss 0.0474 0.7457 4.3447 18.3040 | lr 1.9e-04 | norm 1.8052 | dt 0.026 | |
| type train | step 16010 | loss 0.0446 0.7586 4.4501 18.4088 | lr 1.9e-04 | norm 1.2942 | dt 0.026 | |
| type train | step 16020 | loss 0.0449 0.7325 4.4305 18.3013 | lr 1.9e-04 | norm 1.1179 | dt 0.026 | |
| type train | step 16030 | loss 0.0429 0.7445 4.3899 17.8894 | lr 1.9e-04 | norm 1.2279 | dt 0.026 | |
| type train | step 16040 | loss 0.0444 0.7510 4.3638 18.1276 | lr 1.9e-04 | norm 1.2978 | dt 0.026 | |
| type train | step 16050 | loss 0.0439 0.7433 4.3482 18.0262 | lr 1.9e-04 | norm 1.4589 | dt 0.026 | |
| type train | step 16060 | loss 0.0464 0.7624 4.5310 18.7530 | lr 1.9e-04 | norm 1.2830 | dt 0.026 | |
| type train | step 16070 | loss 0.0454 0.7368 4.3392 18.1370 | lr 1.9e-04 | norm 1.1953 | dt 0.026 | |
| type train | step 16080 | loss 0.0442 0.7284 4.4247 18.1873 | lr 1.9e-04 | norm 1.0950 | dt 0.026 | |
| type train | step 16090 | loss 0.0430 0.7414 4.3745 18.0139 | lr 1.9e-04 | norm 1.0638 | dt 0.025 | |
| type train | step 16100 | loss 0.0447 0.7745 4.5247 18.5202 | lr 1.9e-04 | norm 1.3542 | dt 0.026 | |
| type train | step 16110 | loss 0.0438 0.7338 4.2698 17.6666 | lr 1.9e-04 | norm 1.0971 | dt 0.026 | |
| type train | step 16120 | loss 0.0435 0.7297 4.3572 18.1346 | lr 1.9e-04 | norm 1.1225 | dt 0.025 | |
| type train | step 16130 | loss 0.0453 0.7311 4.2940 17.9099 | lr 1.9e-04 | norm 1.0578 | dt 0.026 | |
| type train | step 16140 | loss 0.0458 0.7414 4.4539 18.3174 | lr 1.9e-04 | norm 1.2688 | dt 0.026 | |
| type train | step 16150 | loss 0.0441 0.7371 4.2625 17.7987 | lr 1.9e-04 | norm 1.2164 | dt 0.026 | |
| type train | step 16160 | loss 0.0446 0.7662 4.4208 17.9842 | lr 1.9e-04 | norm 1.3546 | dt 0.026 | |
| type train | step 16170 | loss 0.0448 0.7421 4.3532 18.1042 | lr 1.9e-04 | norm 1.3498 | dt 0.026 | |
| type train | step 16180 | loss 0.0448 0.7287 4.3820 18.1434 | lr 1.8e-04 | norm 1.3273 | dt 0.026 | |
| type train | step 16190 | loss 0.0454 0.7402 4.4723 18.6483 | lr 1.8e-04 | norm 1.0524 | dt 0.025 | |
| type train | step 16200 | loss 0.0439 0.7430 4.4474 18.4442 | lr 1.8e-04 | norm 1.0571 | dt 0.026 | |
| type train | step 16210 | loss 0.0435 0.7466 4.3920 18.1778 | lr 1.8e-04 | norm 1.2560 | dt 0.026 | |
| type train | step 16220 | loss 0.0438 0.7232 4.2165 17.6208 | lr 1.8e-04 | norm 1.2847 | dt 0.026 | |
| type train | step 16230 | loss 0.0439 0.7440 4.2767 18.0527 | lr 1.8e-04 | norm 1.3339 | dt 0.026 | |
| type train | step 16240 | loss 0.0458 0.7417 4.4056 18.3794 | lr 1.8e-04 | norm 1.2910 | dt 0.026 | |
| type train | step 16250 | loss 0.0437 0.7275 4.4963 18.6013 | lr 1.8e-04 | norm 1.1582 | dt 0.026 | |
| type train | step 16260 | loss 0.0454 0.7487 4.3484 17.8313 | lr 1.8e-04 | norm 1.2616 | dt 0.026 | |
| type train | step 16270 | loss 0.0439 0.7459 4.3869 18.0662 | lr 1.8e-04 | norm 1.0105 | dt 0.026 | |
| type train | step 16280 | loss 0.0438 0.7573 4.3229 17.9564 | lr 1.8e-04 | norm 1.3622 | dt 0.025 | |
| type train | step 16290 | loss 0.0438 0.7526 4.4496 18.3548 | lr 1.8e-04 | norm 1.0832 | dt 0.026 | |
| type train | step 16300 | loss 0.0443 0.7356 4.2835 17.9225 | lr 1.8e-04 | norm 1.3233 | dt 0.026 | |
| type train | step 16310 | loss 0.0457 0.7514 4.4533 18.3696 | lr 1.8e-04 | norm 1.0711 | dt 0.026 | |
| type train | step 16320 | loss 0.0437 0.7285 4.1857 17.3927 | lr 1.8e-04 | norm 1.2361 | dt 0.026 | |
| type train | step 16330 | loss 0.0456 0.7331 4.3889 18.2106 | lr 1.8e-04 | norm 1.1389 | dt 0.026 | |
| type train | step 16340 | loss 0.0442 0.7473 4.4423 18.3064 | lr 1.8e-04 | norm 1.0987 | dt 0.026 | |
| type train | step 16350 | loss 0.0440 0.7554 4.4860 18.4518 | lr 1.8e-04 | norm 1.2154 | dt 0.025 | |
| type train | step 16360 | loss 0.0460 0.7433 4.3738 18.2820 | lr 1.8e-04 | norm 1.2880 | dt 0.026 | |
| type train | step 16370 | loss 0.0448 0.7475 4.5562 18.9670 | lr 1.8e-04 | norm 1.1963 | dt 0.026 | |
| type train | step 16380 | loss 0.0447 0.7469 4.4836 18.6393 | lr 1.8e-04 | norm 1.3644 | dt 0.026 | |
| type train | step 16390 | loss 0.0455 0.7502 4.4919 18.8478 | lr 1.8e-04 | norm 1.1823 | dt 0.026 | |
| type train | step 16400 | loss 0.0425 0.7356 4.3102 17.6983 | lr 1.8e-04 | norm 1.0987 | dt 0.025 | |
| type train | step 16410 | loss 0.0424 0.7507 4.3689 18.2666 | lr 1.8e-04 | norm 1.2193 | dt 0.026 | |
| type train | step 16420 | loss 0.0455 0.7399 4.3440 18.0884 | lr 1.7e-04 | norm 1.4455 | dt 0.026 | |
| type train | step 16430 | loss 0.0440 0.7459 4.5816 19.2536 | lr 1.7e-04 | norm 1.2110 | dt 0.026 | |
| type train | step 16440 | loss 0.0444 0.7434 4.3405 17.7629 | lr 1.7e-04 | norm 1.4476 | dt 0.026 | |
| type train | step 16450 | loss 0.0446 0.7488 4.4671 18.6325 | lr 1.7e-04 | norm 1.2725 | dt 0.025 | |
| type train | step 16460 | loss 0.0428 0.7404 4.3540 17.8160 | lr 1.7e-04 | norm 1.1208 | dt 0.025 | |
| type train | step 16470 | loss 0.0432 0.7545 4.6292 18.9699 | lr 1.7e-04 | norm 1.3816 | dt 0.027 | |
| type train | step 16480 | loss 0.0433 0.7420 4.4214 18.3027 | lr 1.7e-04 | norm 1.0428 | dt 0.026 | |
| type train | step 16490 | loss 0.0442 0.7494 4.5122 18.7882 | lr 1.7e-04 | norm 1.1774 | dt 0.026 | |
| type train | step 16500 | loss 0.0442 0.7248 4.2842 17.7836 | lr 1.7e-04 | norm 1.1851 | dt 0.025 | |
| type train | step 16510 | loss 0.0456 0.7660 4.4993 18.6117 | lr 1.7e-04 | norm 1.3823 | dt 0.026 | |
| type train | step 16520 | loss 0.0426 0.7420 4.3396 17.7066 | lr 1.7e-04 | norm 1.1929 | dt 0.026 | |
| type train | step 16530 | loss 0.0431 0.7335 4.3903 18.4067 | lr 1.7e-04 | norm 1.0831 | dt 0.026 | |
| type train | step 16540 | loss 0.0437 0.7428 4.3283 17.7836 | lr 1.7e-04 | norm 0.9746 | dt 0.026 | |
| type train | step 16550 | loss 0.0460 0.7745 4.6739 19.2963 | lr 1.7e-04 | norm 1.2491 | dt 0.026 | |
| type train | step 16560 | loss 0.0432 0.7409 4.3631 18.0557 | lr 1.7e-04 | norm 1.1869 | dt 0.025 | |
| type train | step 16570 | loss 0.0429 0.7139 4.2668 17.7343 | lr 1.7e-04 | norm 1.0136 | dt 0.026 | |
| type train | step 16580 | loss 0.0446 0.7530 4.4290 18.1475 | lr 1.7e-04 | norm 1.1613 | dt 0.026 | |
| type train | step 16590 | loss 0.0451 0.7524 4.3068 17.7967 | lr 1.7e-04 | norm 1.2137 | dt 0.025 | |
| type train | step 16600 | loss 0.0435 0.7401 4.2819 17.9738 | lr 1.7e-04 | norm 1.0919 | dt 0.026 | |
| type train | step 16610 | loss 0.0475 0.7455 4.3416 18.3054 | lr 1.7e-04 | norm 1.8013 | dt 0.026 | |
| type train | step 16620 | loss 0.0444 0.7582 4.4465 18.4058 | lr 1.7e-04 | norm 1.2814 | dt 0.026 | |
| type train | step 16630 | loss 0.0449 0.7327 4.4313 18.2953 | lr 1.7e-04 | norm 1.1235 | dt 0.026 | |
| type train | step 16640 | loss 0.0428 0.7434 4.3879 17.8825 | lr 1.7e-04 | norm 1.2226 | dt 0.026 | |
| type train | step 16650 | loss 0.0443 0.7506 4.3648 18.1331 | lr 1.7e-04 | norm 1.3087 | dt 0.026 | |
| type train | step 16660 | loss 0.0437 0.7435 4.3493 18.0354 | lr 1.7e-04 | norm 1.4496 | dt 0.026 | |
| type train | step 16670 | loss 0.0463 0.7619 4.5267 18.7631 | lr 1.6e-04 | norm 1.2800 | dt 0.026 | |
| type train | step 16680 | loss 0.0452 0.7357 4.3375 18.1395 | lr 1.6e-04 | norm 1.1915 | dt 0.026 | |
| type train | step 16690 | loss 0.0441 0.7280 4.4227 18.1933 | lr 1.6e-04 | norm 1.0927 | dt 0.026 | |
| type train | step 16700 | loss 0.0428 0.7410 4.3725 18.0137 | lr 1.6e-04 | norm 1.0596 | dt 0.026 | |
| type train | step 16710 | loss 0.0445 0.7746 4.5237 18.5140 | lr 1.6e-04 | norm 1.3590 | dt 0.026 | |
| type train | step 16720 | loss 0.0436 0.7340 4.2713 17.6659 | lr 1.6e-04 | norm 1.1000 | dt 0.026 | |
| type train | step 16730 | loss 0.0434 0.7294 4.3553 18.1313 | lr 1.6e-04 | norm 1.1264 | dt 0.026 | |
| type train | step 16740 | loss 0.0453 0.7302 4.2943 17.9122 | lr 1.6e-04 | norm 1.0615 | dt 0.026 | |
| type train | step 16750 | loss 0.0457 0.7408 4.4497 18.3133 | lr 1.6e-04 | norm 1.2658 | dt 0.026 | |
| type train | step 16760 | loss 0.0440 0.7364 4.2598 17.7911 | lr 1.6e-04 | norm 1.2135 | dt 0.026 | |
| type train | step 16770 | loss 0.0443 0.7664 4.4212 17.9818 | lr 1.6e-04 | norm 1.3636 | dt 0.026 | |
| type train | step 16780 | loss 0.0447 0.7420 4.3520 18.1069 | lr 1.6e-04 | norm 1.3502 | dt 0.026 | |
| type train | step 16790 | loss 0.0446 0.7282 4.3799 18.1437 | lr 1.6e-04 | norm 1.3293 | dt 0.025 | |
| type train | step 16800 | loss 0.0454 0.7394 4.4715 18.6567 | lr 1.6e-04 | norm 1.0589 | dt 0.026 | |
| type train | step 16810 | loss 0.0440 0.7429 4.4480 18.4377 | lr 1.6e-04 | norm 1.0654 | dt 0.026 | |
| type train | step 16820 | loss 0.0434 0.7462 4.3893 18.1704 | lr 1.6e-04 | norm 1.2480 | dt 0.026 | |
| type train | step 16830 | loss 0.0437 0.7228 4.2135 17.6184 | lr 1.6e-04 | norm 1.2883 | dt 0.026 | |
| type train | step 16840 | loss 0.0438 0.7442 4.2758 18.0436 | lr 1.6e-04 | norm 1.3352 | dt 0.026 | |
| type train | step 16850 | loss 0.0455 0.7411 4.4041 18.3720 | lr 1.6e-04 | norm 1.2875 | dt 0.026 | |
| type train | step 16860 | loss 0.0437 0.7269 4.4938 18.5933 | lr 1.6e-04 | norm 1.1623 | dt 0.026 | |
| type train | step 16870 | loss 0.0452 0.7487 4.3443 17.8346 | lr 1.6e-04 | norm 1.2580 | dt 0.026 | |
| type train | step 16880 | loss 0.0439 0.7457 4.3830 18.0692 | lr 1.6e-04 | norm 1.0275 | dt 0.025 | |
| type train | step 16890 | loss 0.0438 0.7567 4.3224 17.9535 | lr 1.6e-04 | norm 1.3588 | dt 0.026 | |
| type train | step 16900 | loss 0.0439 0.7523 4.4459 18.3580 | lr 1.6e-04 | norm 1.1001 | dt 0.026 | |
| type train | step 16910 | loss 0.0441 0.7355 4.2809 17.9294 | lr 1.6e-04 | norm 1.3157 | dt 0.026 | |
| type train | step 16920 | loss 0.0453 0.7501 4.4529 18.3676 | lr 1.6e-04 | norm 1.0453 | dt 0.026 | |
| type train | step 16930 | loss 0.0437 0.7290 4.1821 17.3836 | lr 1.6e-04 | norm 1.2266 | dt 0.026 | |
| type train | step 16940 | loss 0.0455 0.7329 4.3881 18.2134 | lr 1.5e-04 | norm 1.1369 | dt 0.026 | |
| type train | step 16950 | loss 0.0442 0.7462 4.4393 18.3009 | lr 1.5e-04 | norm 1.1028 | dt 0.026 | |
| type train | step 16960 | loss 0.0439 0.7548 4.4866 18.4514 | lr 1.5e-04 | norm 1.2099 | dt 0.026 | |
| type train | step 16970 | loss 0.0461 0.7424 4.3720 18.2762 | lr 1.5e-04 | norm 1.2799 | dt 0.026 | |
| type train | step 16980 | loss 0.0448 0.7476 4.5586 18.9731 | lr 1.5e-04 | norm 1.2069 | dt 0.026 | |
| type train | step 16990 | loss 0.0446 0.7461 4.4831 18.6460 | lr 1.5e-04 | norm 1.3592 | dt 0.026 | |
| type train | step 17000 | loss 0.0452 0.7491 4.4893 18.8470 | lr 1.5e-04 | norm 1.1730 | dt 0.026 | |
| type train | step 17010 | loss 0.0425 0.7355 4.3103 17.6909 | lr 1.5e-04 | norm 1.1093 | dt 0.026 | |
| type train | step 17020 | loss 0.0425 0.7501 4.3655 18.2645 | lr 1.5e-04 | norm 1.2298 | dt 0.026 | |
| type train | step 17030 | loss 0.0455 0.7394 4.3423 18.0910 | lr 1.5e-04 | norm 1.4325 | dt 0.026 | |
| type train | step 17040 | loss 0.0440 0.7453 4.5802 19.2556 | lr 1.5e-04 | norm 1.2201 | dt 0.025 | |
| type train | step 17050 | loss 0.0443 0.7424 4.3365 17.7591 | lr 1.5e-04 | norm 1.4450 | dt 0.026 | |
| type train | step 17060 | loss 0.0447 0.7478 4.4659 18.6332 | lr 1.5e-04 | norm 1.2741 | dt 0.026 | |
| type train | step 17070 | loss 0.0428 0.7402 4.3552 17.8260 | lr 1.5e-04 | norm 1.1232 | dt 0.026 | |
| type train | step 17080 | loss 0.0433 0.7541 4.6255 18.9597 | lr 1.5e-04 | norm 1.3867 | dt 0.027 | |
| type train | step 17090 | loss 0.0434 0.7412 4.4165 18.3001 | lr 1.5e-04 | norm 1.0304 | dt 0.026 | |
| type train | step 17100 | loss 0.0440 0.7489 4.5103 18.7939 | lr 1.5e-04 | norm 1.1695 | dt 0.026 | |
| type train | step 17110 | loss 0.0442 0.7237 4.2798 17.7892 | lr 1.5e-04 | norm 1.1876 | dt 0.026 | |
| type train | step 17120 | loss 0.0456 0.7651 4.4984 18.6127 | lr 1.5e-04 | norm 1.3867 | dt 0.026 | |
| type train | step 17130 | loss 0.0426 0.7411 4.3401 17.7073 | lr 1.5e-04 | norm 1.1920 | dt 0.026 | |
| type train | step 17140 | loss 0.0429 0.7327 4.3881 18.4104 | lr 1.5e-04 | norm 1.0730 | dt 0.026 | |
| type train | step 17150 | loss 0.0436 0.7420 4.3269 17.7826 | lr 1.5e-04 | norm 0.9678 | dt 0.026 | |
| type train | step 17160 | loss 0.0460 0.7733 4.6714 19.2947 | lr 1.5e-04 | norm 1.2565 | dt 0.026 | |
| type train | step 17170 | loss 0.0431 0.7407 4.3595 18.0611 | lr 1.5e-04 | norm 1.1823 | dt 0.026 | |
| type train | step 17180 | loss 0.0430 0.7127 4.2662 17.7416 | lr 1.5e-04 | norm 1.0204 | dt 0.026 | |
| type train | step 17190 | loss 0.0445 0.7520 4.4261 18.1506 | lr 1.5e-04 | norm 1.1535 | dt 0.026 | |
| type train | step 17200 | loss 0.0449 0.7521 4.3060 17.7970 | lr 1.5e-04 | norm 1.2023 | dt 0.026 | |
| type train | step 17210 | loss 0.0435 0.7408 4.2807 17.9782 | lr 1.5e-04 | norm 1.1014 | dt 0.026 | |
| type train | step 17220 | loss 0.0471 0.7446 4.3384 18.3004 | lr 1.5e-04 | norm 1.7876 | dt 0.025 | |
| type train | step 17230 | loss 0.0442 0.7568 4.4424 18.4005 | lr 1.5e-04 | norm 1.2623 | dt 0.026 | |
| type train | step 17240 | loss 0.0447 0.7320 4.4282 18.2942 | lr 1.4e-04 | norm 1.1214 | dt 0.026 | |
| type train | step 17250 | loss 0.0426 0.7430 4.3868 17.8798 | lr 1.4e-04 | norm 1.2095 | dt 0.026 | |
| type train | step 17260 | loss 0.0442 0.7497 4.3624 18.1364 | lr 1.4e-04 | norm 1.2961 | dt 0.026 | |
| type train | step 17270 | loss 0.0437 0.7437 4.3458 18.0404 | lr 1.4e-04 | norm 1.4473 | dt 0.026 | |
| type train | step 17280 | loss 0.0465 0.7614 4.5258 18.7616 | lr 1.4e-04 | norm 1.2836 | dt 0.026 | |
| type train | step 17290 | loss 0.0451 0.7351 4.3321 18.1343 | lr 1.4e-04 | norm 1.1948 | dt 0.026 | |
| type train | step 17300 | loss 0.0438 0.7269 4.4223 18.1954 | lr 1.4e-04 | norm 1.0821 | dt 0.026 | |
| type train | step 17310 | loss 0.0426 0.7402 4.3682 18.0178 | lr 1.4e-04 | norm 1.0519 | dt 0.026 | |
| type train | step 17320 | loss 0.0445 0.7736 4.5235 18.5082 | lr 1.4e-04 | norm 1.3570 | dt 0.026 | |
| type train | step 17330 | loss 0.0437 0.7345 4.2683 17.6627 | lr 1.4e-04 | norm 1.1008 | dt 0.026 | |
| type train | step 17340 | loss 0.0432 0.7289 4.3527 18.1309 | lr 1.4e-04 | norm 1.1129 | dt 0.026 | |
| type train | step 17350 | loss 0.0450 0.7298 4.2933 17.9068 | lr 1.4e-04 | norm 1.0586 | dt 0.026 | |
| type train | step 17360 | loss 0.0456 0.7400 4.4487 18.3177 | lr 1.4e-04 | norm 1.2635 | dt 0.026 | |
| type train | step 17370 | loss 0.0437 0.7360 4.2576 17.7835 | lr 1.4e-04 | norm 1.2067 | dt 0.026 | |
| type train | step 17380 | loss 0.0443 0.7660 4.4199 17.9746 | lr 1.4e-04 | norm 1.3619 | dt 0.026 | |
| type train | step 17390 | loss 0.0447 0.7417 4.3481 18.0980 | lr 1.4e-04 | norm 1.3482 | dt 0.025 | |
| type train | step 17400 | loss 0.0445 0.7278 4.3780 18.1505 | lr 1.4e-04 | norm 1.3166 | dt 0.026 | |
| type train | step 17410 | loss 0.0453 0.7382 4.4715 18.6561 | lr 1.4e-04 | norm 1.0583 | dt 0.026 | |
| type train | step 17420 | loss 0.0438 0.7426 4.4474 18.4395 | lr 1.4e-04 | norm 1.0715 | dt 0.026 | |
| type train | step 17430 | loss 0.0434 0.7459 4.3866 18.1646 | lr 1.4e-04 | norm 1.2464 | dt 0.026 | |
| type train | step 17440 | loss 0.0436 0.7224 4.2110 17.6262 | lr 1.4e-04 | norm 1.2887 | dt 0.026 | |
| type train | step 17450 | loss 0.0437 0.7436 4.2743 18.0448 | lr 1.4e-04 | norm 1.3324 | dt 0.027 | |
| type train | step 17460 | loss 0.0454 0.7410 4.4000 18.3685 | lr 1.4e-04 | norm 1.2751 | dt 0.026 | |
| type train | step 17470 | loss 0.0434 0.7268 4.4932 18.5990 | lr 1.4e-04 | norm 1.1645 | dt 0.026 | |
| type train | step 17480 | loss 0.0452 0.7480 4.3448 17.8384 | lr 1.4e-04 | norm 1.2648 | dt 0.026 | |
| type train | step 17490 | loss 0.0437 0.7452 4.3823 18.0734 | lr 1.4e-04 | norm 1.0163 | dt 0.026 | |
| type train | step 17500 | loss 0.0435 0.7568 4.3211 17.9489 | lr 1.4e-04 | norm 1.3607 | dt 0.026 | |
| type train | step 17510 | loss 0.0436 0.7520 4.4453 18.3655 | lr 1.4e-04 | norm 1.0874 | dt 0.026 | |
| type train | step 17520 | loss 0.0442 0.7351 4.2801 17.9304 | lr 1.4e-04 | norm 1.3032 | dt 0.025 | |
| type train | step 17530 | loss 0.0454 0.7500 4.4522 18.3677 | lr 1.4e-04 | norm 1.0632 | dt 0.025 | |
| type train | step 17540 | loss 0.0435 0.7279 4.1812 17.3817 | lr 1.4e-04 | norm 1.2227 | dt 0.026 | |
| type train | step 17550 | loss 0.0453 0.7323 4.3877 18.2163 | lr 1.4e-04 | norm 1.1347 | dt 0.026 | |
| type train | step 17560 | loss 0.0439 0.7466 4.4367 18.2998 | lr 1.4e-04 | norm 1.0894 | dt 0.026 | |
| type train | step 17570 | loss 0.0438 0.7547 4.4839 18.4532 | lr 1.3e-04 | norm 1.2077 | dt 0.025 | |
| type train | step 17580 | loss 0.0457 0.7423 4.3691 18.2686 | lr 1.3e-04 | norm 1.2633 | dt 0.026 | |
| type train | step 17590 | loss 0.0446 0.7475 4.5547 18.9663 | lr 1.3e-04 | norm 1.2025 | dt 0.025 | |
| type train | step 17600 | loss 0.0445 0.7458 4.4823 18.6531 | lr 1.3e-04 | norm 1.3637 | dt 0.026 | |
| type train | step 17610 | loss 0.0453 0.7494 4.4874 18.8480 | lr 1.3e-04 | norm 1.1833 | dt 0.026 | |
| type train | step 17620 | loss 0.0423 0.7357 4.3071 17.6894 | lr 1.3e-04 | norm 1.1034 | dt 0.026 | |
| type train | step 17630 | loss 0.0424 0.7500 4.3661 18.2683 | lr 1.3e-04 | norm 1.2353 | dt 0.026 | |
| type train | step 17640 | loss 0.0451 0.7384 4.3392 18.0911 | lr 1.3e-04 | norm 1.4224 | dt 0.026 | |
| type train | step 17650 | loss 0.0438 0.7449 4.5783 19.2508 | lr 1.3e-04 | norm 1.2156 | dt 0.026 | |
| type train | step 17660 | loss 0.0441 0.7415 4.3329 17.7659 | lr 1.3e-04 | norm 1.4299 | dt 0.026 | |
| type train | step 17670 | loss 0.0446 0.7470 4.4656 18.6318 | lr 1.3e-04 | norm 1.2678 | dt 0.026 | |
| type train | step 17680 | loss 0.0427 0.7397 4.3523 17.8269 | lr 1.3e-04 | norm 1.1175 | dt 0.026 | |
| type train | step 17690 | loss 0.0429 0.7537 4.6223 18.9636 | lr 1.3e-04 | norm 1.3724 | dt 0.029 | |
| type train | step 17700 | loss 0.0431 0.7410 4.4133 18.3043 | lr 1.3e-04 | norm 1.0192 | dt 0.026 | |
| type train | step 17710 | loss 0.0438 0.7487 4.5111 18.7908 | lr 1.3e-04 | norm 1.1575 | dt 0.026 | |
| type train | step 17720 | loss 0.0438 0.7232 4.2792 17.7930 | lr 1.3e-04 | norm 1.1690 | dt 0.026 | |
| type train | step 17730 | loss 0.0454 0.7648 4.4981 18.6056 | lr 1.3e-04 | norm 1.3765 | dt 0.026 | |
| type train | step 17740 | loss 0.0424 0.7406 4.3402 17.7008 | lr 1.3e-04 | norm 1.1910 | dt 0.026 | |
| type train | step 17750 | loss 0.0426 0.7330 4.3854 18.4086 | lr 1.3e-04 | norm 1.0566 | dt 0.026 | |
| type train | step 17760 | loss 0.0432 0.7424 4.3260 17.7833 | lr 1.3e-04 | norm 0.9627 | dt 0.026 | |
| type train | step 17770 | loss 0.0461 0.7733 4.6696 19.2896 | lr 1.3e-04 | norm 1.2545 | dt 0.026 | |
| type train | step 17780 | loss 0.0428 0.7395 4.3569 18.0552 | lr 1.3e-04 | norm 1.1611 | dt 0.026 | |
| type train | step 17790 | loss 0.0427 0.7125 4.2652 17.7422 | lr 1.3e-04 | norm 1.0102 | dt 0.026 | |
| type train | step 17800 | loss 0.0442 0.7513 4.4265 18.1522 | lr 1.3e-04 | norm 1.1408 | dt 0.026 | |
| type train | step 17810 | loss 0.0447 0.7523 4.3062 17.7962 | lr 1.3e-04 | norm 1.1988 | dt 0.025 | |
| type train | step 17820 | loss 0.0429 0.7401 4.2792 17.9819 | lr 1.3e-04 | norm 1.0817 | dt 0.026 | |
| type train | step 17830 | loss 0.0470 0.7446 4.3357 18.2964 | lr 1.3e-04 | norm 1.7772 | dt 0.026 | |
| type train | step 17840 | loss 0.0441 0.7559 4.4409 18.3961 | lr 1.3e-04 | norm 1.2543 | dt 0.025 | |
| type train | step 17850 | loss 0.0450 0.7315 4.4312 18.2956 | lr 1.3e-04 | norm 1.1325 | dt 0.026 | |
| type train | step 17860 | loss 0.0425 0.7421 4.3849 17.8866 | lr 1.3e-04 | norm 1.2069 | dt 0.025 | |
| type train | step 17870 | loss 0.0440 0.7496 4.3600 18.1422 | lr 1.3e-04 | norm 1.2983 | dt 0.026 | |
| type train | step 17880 | loss 0.0434 0.7429 4.3426 18.0425 | lr 1.3e-04 | norm 1.4399 | dt 0.026 | |
| type train | step 17890 | loss 0.0461 0.7621 4.5251 18.7654 | lr 1.3e-04 | norm 1.2665 | dt 0.026 | |
| type train | step 17900 | loss 0.0451 0.7346 4.3311 18.1393 | lr 1.3e-04 | norm 1.1967 | dt 0.026 | |
| type train | step 17910 | loss 0.0440 0.7269 4.4207 18.1956 | lr 1.3e-04 | norm 1.0939 | dt 0.026 | |
| type train | step 17920 | loss 0.0425 0.7402 4.3669 18.0126 | lr 1.3e-04 | norm 1.0419 | dt 0.026 | |
| type train | step 17930 | loss 0.0444 0.7739 4.5213 18.5007 | lr 1.3e-04 | norm 1.3534 | dt 0.026 | |
| type train | step 17940 | loss 0.0433 0.7342 4.2661 17.6727 | lr 1.3e-04 | norm 1.0947 | dt 0.026 | |
| type train | step 17950 | loss 0.0432 0.7287 4.3513 18.1262 | lr 1.2e-04 | norm 1.1081 | dt 0.026 | |
| type train | step 17960 | loss 0.0449 0.7296 4.2936 17.9073 | lr 1.2e-04 | norm 1.0524 | dt 0.026 | |
| type train | step 17970 | loss 0.0454 0.7400 4.4481 18.3195 | lr 1.2e-04 | norm 1.2546 | dt 0.026 | |
| type train | step 17980 | loss 0.0435 0.7359 4.2541 17.7761 | lr 1.2e-04 | norm 1.1918 | dt 0.025 | |
| type train | step 17990 | loss 0.0442 0.7658 4.4186 17.9780 | lr 1.2e-04 | norm 1.3627 | dt 0.026 | |
| type train | step 18000 | loss 0.0445 0.7415 4.3476 18.1006 | lr 1.2e-04 | norm 1.3381 | dt 0.026 | |
| type train | step 18010 | loss 0.0443 0.7279 4.3764 18.1450 | lr 1.2e-04 | norm 1.3169 | dt 0.026 | |
| type train | step 18020 | loss 0.0449 0.7390 4.4703 18.6599 | lr 1.2e-04 | norm 1.0454 | dt 0.025 | |
| type train | step 18030 | loss 0.0434 0.7426 4.4443 18.4457 | lr 1.2e-04 | norm 1.0536 | dt 0.026 | |
| type train | step 18040 | loss 0.0430 0.7455 4.3870 18.1595 | lr 1.2e-04 | norm 1.2183 | dt 0.026 | |
| type train | step 18050 | loss 0.0434 0.7222 4.2127 17.6318 | lr 1.2e-04 | norm 1.2773 | dt 0.025 | |
| type train | step 18060 | loss 0.0435 0.7434 4.2737 18.0426 | lr 1.2e-04 | norm 1.3355 | dt 0.026 | |
| type train | step 18070 | loss 0.0454 0.7409 4.3988 18.3674 | lr 1.2e-04 | norm 1.2745 | dt 0.026 | |
| type train | step 18080 | loss 0.0436 0.7269 4.4930 18.5986 | lr 1.2e-04 | norm 1.1686 | dt 0.026 | |
| type train | step 18090 | loss 0.0448 0.7483 4.3438 17.8420 | lr 1.2e-04 | norm 1.2494 | dt 0.026 | |
| type train | step 18100 | loss 0.0434 0.7453 4.3781 18.0681 | lr 1.2e-04 | norm 0.9937 | dt 0.026 | |
| type train | step 18110 | loss 0.0433 0.7568 4.3204 17.9484 | lr 1.2e-04 | norm 1.3518 | dt 0.026 | |
| type train | step 18120 | loss 0.0436 0.7516 4.4454 18.3702 | lr 1.2e-04 | norm 1.0968 | dt 0.026 | |
| type train | step 18130 | loss 0.0439 0.7350 4.2784 17.9380 | lr 1.2e-04 | norm 1.2947 | dt 0.025 | |
| type train | step 18140 | loss 0.0450 0.7493 4.4513 18.3716 | lr 1.2e-04 | norm 1.0472 | dt 0.026 | |
| type train | step 18150 | loss 0.0432 0.7279 4.1805 17.3838 | lr 1.2e-04 | norm 1.2102 | dt 0.026 | |
| type train | step 18160 | loss 0.0451 0.7328 4.3875 18.2198 | lr 1.2e-04 | norm 1.1426 | dt 0.026 | |
| type train | step 18170 | loss 0.0437 0.7457 4.4337 18.3062 | lr 1.2e-04 | norm 1.0855 | dt 0.026 | |
| type train | step 18180 | loss 0.0436 0.7548 4.4814 18.4456 | lr 1.2e-04 | norm 1.2066 | dt 0.026 | |
| type train | step 18190 | loss 0.0457 0.7422 4.3657 18.2713 | lr 1.2e-04 | norm 1.2648 | dt 0.025 | |
| type train | step 18200 | loss 0.0446 0.7469 4.5540 18.9671 | lr 1.2e-04 | norm 1.2002 | dt 0.026 | |
| type train | step 18210 | loss 0.0444 0.7453 4.4835 18.6623 | lr 1.2e-04 | norm 1.3638 | dt 0.026 | |
| type train | step 18220 | loss 0.0450 0.7488 4.4849 18.8506 | lr 1.2e-04 | norm 1.1625 | dt 0.026 | |
| type train | step 18230 | loss 0.0423 0.7350 4.3072 17.6869 | lr 1.2e-04 | norm 1.1128 | dt 0.026 | |
| type train | step 18240 | loss 0.0420 0.7499 4.3630 18.2680 | lr 1.2e-04 | norm 1.2107 | dt 0.026 | |
| type train | step 18250 | loss 0.0449 0.7384 4.3384 18.0841 | lr 1.2e-04 | norm 1.4116 | dt 0.026 | |
| type train | step 18260 | loss 0.0437 0.7448 4.5775 19.2532 | lr 1.2e-04 | norm 1.2105 | dt 0.026 | |
| type train | step 18270 | loss 0.0441 0.7404 4.3304 17.7675 | lr 1.2e-04 | norm 1.4197 | dt 0.026 | |
| type train | step 18280 | loss 0.0443 0.7470 4.4645 18.6312 | lr 1.2e-04 | norm 1.2565 | dt 0.026 | |
| type train | step 18290 | loss 0.0425 0.7396 4.3509 17.8302 | lr 1.2e-04 | norm 1.1224 | dt 0.026 | |
| type train | step 18300 | loss 0.0429 0.7536 4.6207 18.9712 | lr 1.2e-04 | norm 1.3783 | dt 0.028 | |
| type train | step 18310 | loss 0.0430 0.7410 4.4104 18.3016 | lr 1.2e-04 | norm 1.0139 | dt 0.026 | |
| type train | step 18320 | loss 0.0439 0.7491 4.5088 18.7836 | lr 1.2e-04 | norm 1.1640 | dt 0.026 | |
| type train | step 18330 | loss 0.0437 0.7233 4.2784 17.7955 | lr 1.2e-04 | norm 1.1701 | dt 0.026 | |
| type train | step 18340 | loss 0.0455 0.7635 4.4978 18.6073 | lr 1.2e-04 | norm 1.3714 | dt 0.026 | |
| type train | step 18350 | loss 0.0425 0.7407 4.3395 17.7074 | lr 1.2e-04 | norm 1.1972 | dt 0.027 | |
| type train | step 18360 | loss 0.0426 0.7321 4.3860 18.4135 | lr 1.2e-04 | norm 1.0678 | dt 0.026 | |
| type train | step 18370 | loss 0.0431 0.7423 4.3260 17.7884 | lr 1.2e-04 | norm 0.9530 | dt 0.026 | |
| type train | step 18380 | loss 0.0457 0.7731 4.6715 19.2828 | lr 1.2e-04 | norm 1.2431 | dt 0.026 | |
| type train | step 18390 | loss 0.0435 0.7392 4.3545 18.0592 | lr 1.2e-04 | norm 1.1889 | dt 0.026 | |
| type train | step 18400 | loss 0.0427 0.7122 4.2627 17.7382 | lr 1.2e-04 | norm 1.0189 | dt 0.026 | |
| type train | step 18410 | loss 0.0442 0.7513 4.4259 18.1530 | lr 1.2e-04 | norm 1.1426 | dt 0.026 | |
| type train | step 18420 | loss 0.0446 0.7519 4.3075 17.7940 | lr 1.1e-04 | norm 1.1987 | dt 0.026 | |
| type train | step 18430 | loss 0.0430 0.7398 4.2781 17.9867 | lr 1.1e-04 | norm 1.0903 | dt 0.026 | |
| type train | step 18440 | loss 0.0468 0.7445 4.3338 18.2983 | lr 1.1e-04 | norm 1.7713 | dt 0.026 | |
| type train | step 18450 | loss 0.0441 0.7552 4.4383 18.4009 | lr 1.1e-04 | norm 1.2488 | dt 0.026 | |
| type train | step 18460 | loss 0.0444 0.7303 4.4297 18.2909 | lr 1.1e-04 | norm 1.1162 | dt 0.026 | |
| type train | step 18470 | loss 0.0425 0.7427 4.3840 17.8858 | lr 1.1e-04 | norm 1.2097 | dt 0.026 | |
| type train | step 18480 | loss 0.0439 0.7493 4.3603 18.1452 | lr 1.1e-04 | norm 1.3037 | dt 0.026 | |
| type train | step 18490 | loss 0.0433 0.7431 4.3402 18.0462 | lr 1.1e-04 | norm 1.4405 | dt 0.026 | |
| type train | step 18500 | loss 0.0460 0.7618 4.5227 18.7682 | lr 1.1e-04 | norm 1.2567 | dt 0.025 | |
| type train | step 18510 | loss 0.0450 0.7349 4.3317 18.1375 | lr 1.1e-04 | norm 1.2026 | dt 0.026 | |
| type train | step 18520 | loss 0.0437 0.7264 4.4191 18.1944 | lr 1.1e-04 | norm 1.0859 | dt 0.026 | |
| type train | step 18530 | loss 0.0426 0.7402 4.3644 18.0073 | lr 1.1e-04 | norm 1.0513 | dt 0.026 | |
| type train | step 18540 | loss 0.0442 0.7732 4.5226 18.5009 | lr 1.1e-04 | norm 1.3653 | dt 0.026 | |
| type train | step 18550 | loss 0.0433 0.7341 4.2648 17.6751 | lr 1.1e-04 | norm 1.1013 | dt 0.026 | |
| type train | step 18560 | loss 0.0430 0.7286 4.3501 18.1239 | lr 1.1e-04 | norm 1.1025 | dt 0.026 | |
| type train | step 18570 | loss 0.0449 0.7299 4.2919 17.9064 | lr 1.1e-04 | norm 1.0565 | dt 0.026 | |
| type train | step 18580 | loss 0.0454 0.7394 4.4476 18.3216 | lr 1.1e-04 | norm 1.2534 | dt 0.026 | |
| type train | step 18590 | loss 0.0434 0.7354 4.2536 17.7730 | lr 1.1e-04 | norm 1.1774 | dt 0.026 | |
| type train | step 18600 | loss 0.0441 0.7657 4.4182 17.9818 | lr 1.1e-04 | norm 1.3606 | dt 0.026 | |
| type train | step 18610 | loss 0.0443 0.7418 4.3475 18.1022 | lr 1.1e-04 | norm 1.3406 | dt 0.026 | |
| type train | step 18620 | loss 0.0442 0.7277 4.3766 18.1407 | lr 1.1e-04 | norm 1.3160 | dt 0.026 | |
| type train | step 18630 | loss 0.0450 0.7390 4.4711 18.6596 | lr 1.1e-04 | norm 1.0527 | dt 0.026 | |
| type train | step 18640 | loss 0.0434 0.7426 4.4470 18.4485 | lr 1.1e-04 | norm 1.0574 | dt 0.025 | |
| type train | step 18650 | loss 0.0429 0.7453 4.3854 18.1601 | lr 1.1e-04 | norm 1.2134 | dt 0.026 | |
| type train | step 18660 | loss 0.0432 0.7223 4.2141 17.6331 | lr 1.1e-04 | norm 1.2788 | dt 0.026 | |
| type train | step 18670 | loss 0.0435 0.7430 4.2731 18.0432 | lr 1.1e-04 | norm 1.3407 | dt 0.027 | |
| type train | step 18680 | loss 0.0452 0.7405 4.3986 18.3686 | lr 1.1e-04 | norm 1.2742 | dt 0.026 | |
| type train | step 18690 | loss 0.0433 0.7272 4.4967 18.5982 | lr 1.1e-04 | norm 1.1648 | dt 0.026 | |
| type train | step 18700 | loss 0.0447 0.7484 4.3441 17.8364 | lr 1.1e-04 | norm 1.2511 | dt 0.026 | |
| type train | step 18710 | loss 0.0434 0.7445 4.3780 18.0684 | lr 1.1e-04 | norm 0.9929 | dt 0.026 | |
| type train | step 18720 | loss 0.0433 0.7562 4.3202 17.9466 | lr 1.1e-04 | norm 1.3512 | dt 0.026 | |
| type train | step 18730 | loss 0.0434 0.7514 4.4448 18.3716 | lr 1.1e-04 | norm 1.0932 | dt 0.026 | |
| type train | step 18740 | loss 0.0438 0.7354 4.2775 17.9383 | lr 1.1e-04 | norm 1.2884 | dt 0.026 | |
| type train | step 18750 | loss 0.0449 0.7500 4.4529 18.3783 | lr 1.1e-04 | norm 1.0492 | dt 0.025 | |
| type train | step 18760 | loss 0.0432 0.7278 4.1789 17.3877 | lr 1.1e-04 | norm 1.2062 | dt 0.026 | |
| type train | step 18770 | loss 0.0452 0.7323 4.3875 18.2178 | lr 1.1e-04 | norm 1.1479 | dt 0.026 | |
| type train | step 18780 | loss 0.0437 0.7458 4.4332 18.3098 | lr 1.1e-04 | norm 1.0875 | dt 0.026 | |
| type train | step 18790 | loss 0.0435 0.7542 4.4818 18.4471 | lr 1.1e-04 | norm 1.2149 | dt 0.026 | |
| type train | step 18800 | loss 0.0456 0.7419 4.3623 18.2679 | lr 1.1e-04 | norm 1.2533 | dt 0.026 | |
| type train | step 18810 | loss 0.0445 0.7472 4.5553 18.9659 | lr 1.1e-04 | norm 1.2034 | dt 0.026 | |
| type train | step 18820 | loss 0.0443 0.7452 4.4816 18.6609 | lr 1.1e-04 | norm 1.3653 | dt 0.026 | |
| type train | step 18830 | loss 0.0449 0.7488 4.4855 18.8583 | lr 1.1e-04 | norm 1.1701 | dt 0.026 | |
| type train | step 18840 | loss 0.0422 0.7345 4.3078 17.6899 | lr 1.1e-04 | norm 1.1078 | dt 0.026 | |
| type train | step 18850 | loss 0.0421 0.7495 4.3630 18.2668 | lr 1.1e-04 | norm 1.2238 | dt 0.026 | |
| type train | step 18860 | loss 0.0448 0.7379 4.3372 18.0844 | lr 1.1e-04 | norm 1.4056 | dt 0.026 | |
| type train | step 18870 | loss 0.0435 0.7449 4.5755 19.2513 | lr 1.1e-04 | norm 1.2099 | dt 0.025 | |
| type train | step 18880 | loss 0.0440 0.7402 4.3297 17.7669 | lr 1.1e-04 | norm 1.4122 | dt 0.026 | |
| type train | step 18890 | loss 0.0441 0.7470 4.4642 18.6288 | lr 1.1e-04 | norm 1.2469 | dt 0.026 | |
| type train | step 18900 | loss 0.0426 0.7392 4.3514 17.8312 | lr 1.1e-04 | norm 1.1243 | dt 0.026 | |
| type train | step 18910 | loss 0.0427 0.7525 4.6212 18.9677 | lr 1.1e-04 | norm 1.3690 | dt 0.027 | |
| type train | step 18920 | loss 0.0430 0.7409 4.4086 18.3107 | lr 1.1e-04 | norm 1.0095 | dt 0.026 | |
| type train | step 18930 | loss 0.0437 0.7485 4.5073 18.7903 | lr 1.1e-04 | norm 1.1642 | dt 0.026 | |
| type train | step 18940 | loss 0.0438 0.7236 4.2752 17.8004 | lr 1.1e-04 | norm 1.1655 | dt 0.026 | |
| type train | step 18950 | loss 0.0453 0.7635 4.4984 18.6092 | lr 1.1e-04 | norm 1.3666 | dt 0.026 | |
| type train | step 18960 | loss 0.0423 0.7403 4.3387 17.7098 | lr 1.1e-04 | norm 1.1878 | dt 0.026 | |
| type train | step 18970 | loss 0.0425 0.7314 4.3858 18.4104 | lr 1.1e-04 | norm 1.0443 | dt 0.026 | |
| type train | step 18980 | loss 0.0432 0.7428 4.3249 17.7886 | lr 1.1e-04 | norm 0.9714 | dt 0.026 | |
| type train | step 18990 | loss 0.0455 0.7732 4.6670 19.2867 | lr 1.1e-04 | norm 1.2501 | dt 0.026 | |
| type train | step 19000 | loss 0.0428 0.7386 4.3513 18.0631 | lr 1.1e-04 | norm 1.1575 | dt 0.026 | |
| type train | step 19010 | loss 0.0426 0.7119 4.2637 17.7432 | lr 1.1e-04 | norm 1.0181 | dt 0.026 | |
| type train | step 19020 | loss 0.0440 0.7509 4.4254 18.1574 | lr 1.1e-04 | norm 1.1311 | dt 0.026 | |
| type train | step 19030 | loss 0.0446 0.7521 4.3091 17.8017 | lr 1.1e-04 | norm 1.2152 | dt 0.026 | |
| type train | step 19040 | loss 0.0429 0.7400 4.2782 17.9888 | lr 1.1e-04 | norm 1.0888 | dt 0.026 | |
| type train | step 19050 | loss 0.0468 0.7436 4.3338 18.2993 | lr 1.1e-04 | norm 1.7794 | dt 0.026 | |
| type train | step 19060 | loss 0.0440 0.7550 4.4378 18.4024 | lr 1.1e-04 | norm 1.2509 | dt 0.026 | |
| type train | step 19070 | loss 0.0444 0.7314 4.4295 18.2874 | lr 1.1e-04 | norm 1.1326 | dt 0.026 | |
| type train | step 19080 | loss 0.0423 0.7419 4.3806 17.8870 | lr 1.1e-04 | norm 1.2004 | dt 0.026 | |
| type train | step 19090 | loss 0.0438 0.7490 4.3593 18.1453 | lr 1.0e-04 | norm 1.2989 | dt 0.025 | |
| type train | step 19100 | loss 0.0433 0.7437 4.3424 18.0481 | lr 1.0e-04 | norm 1.4382 | dt 0.026 | |
| type train | step 19110 | loss 0.0459 0.7616 4.5236 18.7638 | lr 1.0e-04 | norm 1.2519 | dt 0.026 | |
| type train | step 19120 | loss 0.0449 0.7348 4.3319 18.1441 | lr 1.0e-04 | norm 1.1970 | dt 0.026 | |
| type train | step 19130 | loss 0.0436 0.7263 4.4185 18.1962 | lr 1.0e-04 | norm 1.0785 | dt 0.025 | |
| type train | step 19140 | loss 0.0423 0.7397 4.3625 18.0071 | lr 1.0e-04 | norm 1.0398 | dt 0.026 | |
| type train | step 19150 | loss 0.0440 0.7726 4.5225 18.5014 | lr 1.0e-04 | norm 1.3606 | dt 0.026 | |
| type train | step 19160 | loss 0.0433 0.7342 4.2636 17.6797 | lr 1.0e-04 | norm 1.0987 | dt 0.026 | |
| type train | step 19170 | loss 0.0428 0.7291 4.3484 18.1253 | lr 1.0e-04 | norm 1.1019 | dt 0.026 | |
| type train | step 19180 | loss 0.0449 0.7292 4.2919 17.9040 | lr 1.0e-04 | norm 1.0602 | dt 0.025 | |
| type train | step 19190 | loss 0.0452 0.7388 4.4471 18.3254 | lr 1.0e-04 | norm 1.2478 | dt 0.026 | |
| type train | step 19200 | loss 0.0434 0.7345 4.2539 17.7678 | lr 1.0e-04 | norm 1.1744 | dt 0.026 | |
| type train | step 19210 | loss 0.0440 0.7648 4.4173 17.9840 | lr 1.0e-04 | norm 1.3652 | dt 0.026 | |
| type train | step 19220 | loss 0.0442 0.7411 4.3492 18.1052 | lr 1.0e-04 | norm 1.3393 | dt 0.026 | |
| type train | step 19230 | loss 0.0442 0.7275 4.3758 18.1396 | lr 1.0e-04 | norm 1.3093 | dt 0.026 | |
| type train | step 19240 | loss 0.0448 0.7385 4.4703 18.6606 | lr 1.0e-04 | norm 1.0386 | dt 0.026 | |
| type train | step 19250 | loss 0.0433 0.7428 4.4465 18.4401 | lr 1.0e-04 | norm 1.0479 | dt 0.026 | |
| type train | step 19260 | loss 0.0429 0.7448 4.3827 18.1606 | lr 1.0e-04 | norm 1.2118 | dt 0.026 | |
| type train | step 19270 | loss 0.0430 0.7214 4.2144 17.6302 | lr 1.0e-04 | norm 1.2743 | dt 0.026 | |
| type train | step 19280 | loss 0.0433 0.7436 4.2739 18.0439 | lr 1.0e-04 | norm 1.3438 | dt 0.027 | |
| type train | step 19290 | loss 0.0452 0.7403 4.3970 18.3683 | lr 1.0e-04 | norm 1.2768 | dt 0.026 | |
| type train | step 19300 | loss 0.0433 0.7269 4.4938 18.5993 | lr 1.0e-04 | norm 1.1683 | dt 0.026 | |
| type train | step 19310 | loss 0.0447 0.7479 4.3421 17.8343 | lr 1.0e-04 | norm 1.2519 | dt 0.025 | |
| type train | step 19320 | loss 0.0434 0.7447 4.3758 18.0657 | lr 1.0e-04 | norm 0.9941 | dt 0.026 | |
| type train | step 19330 | loss 0.0431 0.7560 4.3188 17.9489 | lr 1.0e-04 | norm 1.3542 | dt 0.026 | |
| type train | step 19340 | loss 0.0433 0.7521 4.4434 18.3675 | lr 1.0e-04 | norm 1.0951 | dt 0.026 | |
| type train | step 19350 | loss 0.0439 0.7355 4.2748 17.9423 | lr 1.0e-04 | norm 1.2891 | dt 0.026 | |
| type train | step 19360 | loss 0.0449 0.7492 4.4505 18.3845 | lr 1.0e-04 | norm 1.0474 | dt 0.026 | |
| type train | step 19370 | loss 0.0431 0.7276 4.1799 17.3817 | lr 1.0e-04 | norm 1.2025 | dt 0.026 | |
| type train | step 19380 | loss 0.0451 0.7323 4.3863 18.2148 | lr 1.0e-04 | norm 1.1481 | dt 0.026 | |
| type train | step 19390 | loss 0.0435 0.7455 4.4311 18.3117 | lr 1.0e-04 | norm 1.0829 | dt 0.026 | |
| type train | step 19400 | loss 0.0434 0.7541 4.4796 18.4507 | lr 1.0e-04 | norm 1.2174 | dt 0.026 | |
| type train | step 19410 | loss 0.0455 0.7420 4.3612 18.2653 | lr 1.0e-04 | norm 1.2421 | dt 0.026 | |
| type train | step 19420 | loss 0.0444 0.7476 4.5541 18.9729 | lr 1.0e-04 | norm 1.2000 | dt 0.026 | |
| type train | step 19430 | loss 0.0442 0.7447 4.4804 18.6613 | lr 1.0e-04 | norm 1.3619 | dt 0.026 | |
| type train | step 19440 | loss 0.0448 0.7481 4.4845 18.8607 | lr 1.0e-04 | norm 1.1572 | dt 0.026 | |
| type train | step 19450 | loss 0.0422 0.7348 4.3045 17.6916 | lr 1.0e-04 | norm 1.1057 | dt 0.026 | |
| type train | step 19460 | loss 0.0420 0.7498 4.3611 18.2685 | lr 1.0e-04 | norm 1.2187 | dt 0.026 | |
| type train | step 19470 | loss 0.0448 0.7378 4.3330 18.0813 | lr 1.0e-04 | norm 1.3966 | dt 0.026 | |
| type train | step 19480 | loss 0.0434 0.7447 4.5723 19.2404 | lr 1.0e-04 | norm 1.2014 | dt 0.026 | |
| type train | step 19490 | loss 0.0440 0.7395 4.3317 17.7640 | lr 1.0e-04 | norm 1.4098 | dt 0.026 | |
| type train | step 19500 | loss 0.0441 0.7467 4.4621 18.6277 | lr 1.0e-04 | norm 1.2514 | dt 0.026 | |
| type train | step 19510 | loss 0.0424 0.7392 4.3483 17.8320 | lr 1.0e-04 | norm 1.1157 | dt 0.026 | |
| type train | step 19520 | loss 0.0428 0.7528 4.6204 18.9732 | lr 1.0e-04 | norm 1.3740 | dt 0.027 | |
| type train | step 19530 | loss 0.0429 0.7410 4.4052 18.3110 | lr 1.0e-04 | norm 1.0050 | dt 0.026 | |
| type train | step 19540 | loss 0.0436 0.7483 4.5067 18.7914 | lr 1.0e-04 | norm 1.1588 | dt 0.026 | |
| type train | step 19550 | loss 0.0437 0.7230 4.2743 17.8038 | lr 1.0e-04 | norm 1.1596 | dt 0.026 | |
| type train | step 19560 | loss 0.0452 0.7634 4.4983 18.6156 | lr 1.0e-04 | norm 1.3602 | dt 0.026 | |
| type train | step 19570 | loss 0.0423 0.7404 4.3370 17.7105 | lr 1.0e-04 | norm 1.1875 | dt 0.026 | |
| type train | step 19580 | loss 0.0424 0.7314 4.3856 18.4134 | lr 1.0e-04 | norm 1.0429 | dt 0.026 | |
| type train | step 19590 | loss 0.0430 0.7423 4.3252 17.7828 | lr 1.0e-04 | norm 0.9553 | dt 0.026 | |
| type train | step 19600 | loss 0.0454 0.7727 4.6668 19.2843 | lr 1.0e-04 | norm 1.2325 | dt 0.026 | |
| type train | step 19610 | loss 0.0425 0.7387 4.3492 18.0647 | lr 1.0e-04 | norm 1.1478 | dt 0.026 | |
| type train | step 19620 | loss 0.0424 0.7115 4.2625 17.7442 | lr 1.0e-04 | norm 1.0217 | dt 0.026 | |
| type train | step 19630 | loss 0.0438 0.7507 4.4237 18.1632 | lr 1.0e-04 | norm 1.1345 | dt 0.025 | |
| type train | step 19640 | loss 0.0445 0.7517 4.3068 17.8030 | lr 1.0e-04 | norm 1.2047 | dt 0.025 | |
| type train | step 19650 | loss 0.0427 0.7404 4.2785 17.9865 | lr 1.0e-04 | norm 1.0908 | dt 0.025 | |
| type train | step 19660 | loss 0.0467 0.7441 4.3316 18.2998 | lr 1.0e-04 | norm 1.7695 | dt 0.026 | |
| type train | step 19670 | loss 0.0438 0.7552 4.4354 18.4034 | lr 1.0e-04 | norm 1.2413 | dt 0.026 | |
| type train | step 19680 | loss 0.0443 0.7306 4.4282 18.2934 | lr 1.0e-04 | norm 1.1208 | dt 0.026 | |
| type train | step 19690 | loss 0.0422 0.7415 4.3793 17.8922 | lr 1.0e-04 | norm 1.2046 | dt 0.026 | |
| type train | step 19700 | loss 0.0436 0.7490 4.3594 18.1480 | lr 1.0e-04 | norm 1.3007 | dt 0.026 | |
| type train | step 19710 | loss 0.0431 0.7433 4.3376 18.0549 | lr 1.0e-04 | norm 1.4382 | dt 0.026 | |
| type train | step 19720 | loss 0.0458 0.7614 4.5224 18.7650 | lr 1.0e-04 | norm 1.2553 | dt 0.026 | |
| type train | step 19730 | loss 0.0447 0.7342 4.3320 18.1476 | lr 1.0e-04 | norm 1.1960 | dt 0.026 | |
| type train | step 19740 | loss 0.0435 0.7266 4.4194 18.1941 | lr 1.0e-04 | norm 1.0860 | dt 0.026 | |
| type train | step 19750 | loss 0.0421 0.7396 4.3624 18.0100 | lr 1.0e-04 | norm 1.0371 | dt 0.026 | |
| type train | step 19760 | loss 0.0441 0.7731 4.5230 18.5009 | lr 1.0e-04 | norm 1.3747 | dt 0.027 | |
| type train | step 19770 | loss 0.0432 0.7348 4.2622 17.6796 | lr 1.0e-04 | norm 1.1130 | dt 0.026 | |
| type train | step 19780 | loss 0.0428 0.7292 4.3445 18.1195 | lr 1.0e-04 | norm 1.1030 | dt 0.026 | |
| type train | step 19790 | loss 0.0447 0.7292 4.2912 17.9076 | lr 1.0e-04 | norm 1.0560 | dt 0.026 | |
| type train | step 19800 | loss 0.0452 0.7388 4.4464 18.3257 | lr 1.0e-04 | norm 1.2616 | dt 0.026 | |
| type train | step 19810 | loss 0.0434 0.7345 4.2512 17.7663 | lr 1.0e-04 | norm 1.1864 | dt 0.026 | |
| type train | step 19820 | loss 0.0438 0.7648 4.4178 17.9820 | lr 1.0e-04 | norm 1.3676 | dt 0.026 | |
| type train | step 19830 | loss 0.0442 0.7422 4.3479 18.1092 | lr 1.0e-04 | norm 1.3510 | dt 0.026 | |
| type train | step 19840 | loss 0.0440 0.7278 4.3721 18.1355 | lr 1.0e-04 | norm 1.3084 | dt 0.025 | |
| type train | step 19850 | loss 0.0446 0.7384 4.4701 18.6655 | lr 1.0e-04 | norm 1.0469 | dt 0.026 | |
| type train | step 19860 | loss 0.0434 0.7423 4.4459 18.4385 | lr 1.0e-04 | norm 1.0650 | dt 0.026 | |
| type train | step 19870 | loss 0.0427 0.7442 4.3827 18.1585 | lr 1.0e-04 | norm 1.2083 | dt 0.026 | |
| type train | step 19880 | loss 0.0430 0.7211 4.2135 17.6322 | lr 1.0e-04 | norm 1.2769 | dt 0.026 | |
| type train | step 19890 | loss 0.0432 0.7437 4.2757 18.0442 | lr 1.0e-04 | norm 1.3573 | dt 0.026 | |
| type train | step 19900 | loss 0.0450 0.7403 4.3953 18.3732 | lr 1.0e-04 | norm 1.2712 | dt 0.026 | |
| type train | step 19910 | loss 0.0430 0.7271 4.4947 18.6012 | lr 1.0e-04 | norm 1.1685 | dt 0.026 | |
| type train | step 19920 | loss 0.0445 0.7479 4.3403 17.8377 | lr 1.0e-04 | norm 1.2505 | dt 0.026 | |
| type train | step 19930 | loss 0.0432 0.7444 4.3737 18.0688 | lr 1.0e-04 | norm 0.9922 | dt 0.026 | |
| type train | step 19940 | loss 0.0430 0.7563 4.3181 17.9478 | lr 1.0e-04 | norm 1.3590 | dt 0.025 | |
| type train | step 19950 | loss 0.0431 0.7519 4.4427 18.3698 | lr 1.0e-04 | norm 1.1003 | dt 0.026 | |
| type train | step 19960 | loss 0.0437 0.7344 4.2736 17.9442 | lr 1.0e-04 | norm 1.2810 | dt 0.026 | |
| type train | step 19970 | loss 0.0447 0.7493 4.4505 18.3883 | lr 1.0e-04 | norm 1.0479 | dt 0.025 | |
| type train | step 19980 | loss 0.0430 0.7270 4.1780 17.3854 | lr 1.0e-04 | norm 1.2024 | dt 0.026 | |
| type train | step 19990 | loss 0.0449 0.7322 4.3860 18.2135 | lr 1.0e-04 | norm 1.1542 | dt 0.026 | |
| type train | step 20000 | loss 0.0434 0.7454 4.4314 18.3147 | lr 1.0e-04 | norm 1.0831 | dt 0.026 | |