davidquarel's picture
Upload folder using huggingface_hub
71e1228 verified
type train | step 10 | loss 183.8180 0.4146 298.7767 3.9381 593.5026 9.7626 751.7532 45.9789 3.4629 10.5665 13.6273 23.5429 | lr 1.3e-05 | norm 342.1078 | dt 0.118
type train | step 20 | loss 181.1809 0.4033 295.1017 3.8714 588.2693 9.6662 734.1727 45.5815 3.4528 10.5747 13.5244 23.5021 | lr 2.7e-05 | norm 338.4109 | dt 0.118
type train | step 30 | loss 179.1681 0.3884 291.5784 3.8100 575.4286 9.6432 731.4196 45.2702 3.4329 10.3882 13.5583 23.5265 | lr 4.0e-05 | norm 341.0122 | dt 0.118
type train | step 40 | loss 174.8164 0.3845 284.7663 3.6916 570.8116 9.6149 716.6218 43.4306 3.4475 10.4039 13.6778 23.3043 | lr 5.3e-05 | norm 321.5496 | dt 0.118
type train | step 50 | loss 170.5358 0.3460 281.8332 3.6316 551.1471 9.1877 704.5116 43.1092 3.3739 10.2791 13.4043 23.0739 | lr 6.7e-05 | norm 324.3975 | dt 0.118
type train | step 60 | loss 165.8851 0.3223 271.3100 3.4597 538.9424 8.9088 688.6050 40.7117 3.3561 10.1795 13.3460 22.7474 | lr 8.0e-05 | norm 317.7036 | dt 0.119
type train | step 70 | loss 159.6810 0.2965 263.8040 3.3046 522.1560 8.7233 667.6918 39.1574 3.3008 10.0645 13.2432 22.5155 | lr 9.3e-05 | norm 308.9028 | dt 0.118
type train | step 80 | loss 152.8544 0.2770 253.3631 3.1591 510.0374 8.3466 646.8739 36.8732 3.2737 10.0044 13.1267 22.2214 | lr 1.1e-04 | norm 291.6853 | dt 0.118
type train | step 90 | loss 145.8890 0.2434 242.3029 2.9738 483.5088 7.9773 625.3025 35.8444 3.1757 9.6979 12.9993 21.9151 | lr 1.2e-04 | norm 287.9713 | dt 0.118
type train | step 100 | loss 137.3826 0.2159 229.7944 2.7850 464.5647 7.4984 597.3239 33.0417 3.0858 9.5519 12.7245 21.3614 | lr 1.3e-04 | norm 280.0275 | dt 0.120
type train | step 110 | loss 127.8891 0.1912 216.4542 2.6168 438.4927 7.1490 567.9374 31.0237 2.9793 9.3889 12.6435 21.0110 | lr 1.5e-04 | norm 263.7258 | dt 0.119
type train | step 120 | loss 119.2628 0.1730 203.5401 2.4225 416.5339 6.8846 546.5914 28.4660 2.8994 9.2650 12.5791 20.5158 | lr 1.6e-04 | norm 244.6344 | dt 0.118
type train | step 130 | loss 110.6586 0.1460 190.2479 2.2566 390.8100 6.4815 516.4224 26.4368 2.7722 9.0349 12.3811 20.0489 | lr 1.7e-04 | norm 234.0981 | dt 0.119
type train | step 140 | loss 101.9535 0.1321 177.4156 2.0528 376.8900 6.1482 484.2498 23.9637 2.6821 8.8994 12.2053 19.4635 | lr 1.9e-04 | norm 217.9883 | dt 0.118
type train | step 150 | loss 93.8091 0.1083 162.6851 1.8545 340.5433 5.5412 453.9835 22.5163 2.5682 8.4879 11.8575 19.0879 | lr 2.0e-04 | norm 206.5607 | dt 0.119
type train | step 160 | loss 84.5473 0.0903 148.9559 1.6805 314.6523 5.1628 415.1609 20.1729 2.4327 8.2827 11.5624 18.3717 | lr 2.1e-04 | norm 192.2758 | dt 0.120
type train | step 170 | loss 76.4982 0.0747 136.5102 1.5196 285.5909 4.6006 380.2351 18.1331 2.3145 7.9412 11.1877 17.7497 | lr 2.3e-04 | norm 179.7324 | dt 0.121
type train | step 180 | loss 68.6922 0.0660 124.0968 1.3851 267.1680 4.3772 359.7953 16.1646 2.2387 7.8031 11.0905 17.1860 | lr 2.4e-04 | norm 161.0744 | dt 0.118
type train | step 190 | loss 60.5073 0.0541 110.9434 1.2333 239.2544 4.0009 330.8048 14.7540 2.1236 7.4655 10.8149 16.6186 | lr 2.5e-04 | norm 148.5903 | dt 0.120
type train | step 200 | loss 53.8581 0.0470 99.9773 1.1053 224.0415 3.6829 302.4424 13.1537 2.0289 7.2111 10.5439 15.9912 | lr 2.7e-04 | norm 134.1746 | dt 0.120
type train | step 210 | loss 47.0746 0.0380 89.1983 0.9729 199.4205 3.3140 275.2100 11.7877 1.9139 6.8761 10.2169 15.3379 | lr 2.8e-04 | norm 121.6171 | dt 0.119
type train | step 220 | loss 40.9939 0.0307 79.0003 0.8687 178.3986 3.0199 247.5034 10.4529 1.7883 6.5760 9.8710 14.5985 | lr 2.9e-04 | norm 110.1053 | dt 0.120
type train | step 230 | loss 35.2686 0.0255 69.7100 0.7745 159.5983 2.6467 222.2337 9.3333 1.6962 6.2673 9.4196 13.9271 | lr 3.1e-04 | norm 97.9092 | dt 0.122
type train | step 240 | loss 30.9708 0.0238 62.8715 0.7156 146.6446 2.4577 205.4835 8.7896 1.6431 6.0616 9.1970 13.4364 | lr 3.2e-04 | norm 85.2132 | dt 0.119
type train | step 250 | loss 26.7326 0.0208 54.7110 0.6310 128.4947 2.1941 182.2693 7.9794 1.5399 5.6937 8.8395 12.8161 | lr 3.3e-04 | norm 74.1240 | dt 0.120
type train | step 260 | loss 22.9430 0.0193 47.9269 0.5799 115.6187 1.9866 163.4063 7.3139 1.4694 5.4212 8.5000 12.1694 | lr 3.5e-04 | norm 63.8002 | dt 0.121
type train | step 270 | loss 20.1021 0.0182 43.0261 0.5296 103.6433 1.7916 146.7733 6.6306 1.4050 5.1062 8.1030 11.5474 | lr 3.6e-04 | norm 53.8914 | dt 0.122
type train | step 280 | loss 17.7343 0.0178 38.4394 0.4979 93.4553 1.6580 135.2379 6.3137 1.3356 4.8510 7.8518 11.1884 | lr 3.7e-04 | norm 45.3118 | dt 0.122
type train | step 290 | loss 15.4288 0.0162 34.2375 0.4656 82.4010 1.4818 119.4850 5.5574 1.2607 4.5343 7.3896 10.4931 | lr 3.9e-04 | norm 36.9143 | dt 0.122
type train | step 300 | loss 14.0034 0.0166 31.5172 0.4630 76.5560 1.3769 110.4684 5.0384 1.1956 4.3370 7.0771 10.0690 | lr 4.0e-04 | norm 30.5520 | dt 0.122
type train | step 310 | loss 12.8172 0.0169 28.8622 0.4456 69.7371 1.2957 101.4036 4.7181 1.1506 4.0721 6.7406 9.6928 | lr 4.1e-04 | norm 24.6368 | dt 0.124
type train | step 320 | loss 11.4434 0.0169 26.2164 0.4312 63.8540 1.2063 92.8901 4.3091 1.1080 3.8279 6.4165 9.2229 | lr 4.3e-04 | norm 19.2640 | dt 0.128
type train | step 330 | loss 10.1926 0.0166 24.2818 0.4328 59.1612 1.1447 86.2451 3.8824 1.0691 3.5848 6.1132 8.7633 | lr 4.4e-04 | norm 15.1760 | dt 0.127
type train | step 340 | loss 9.0720 0.0163 22.1443 0.4288 54.2905 1.0999 80.4030 3.6273 0.9878 3.3504 5.7922 8.3901 | lr 4.5e-04 | norm 12.8704 | dt 0.132
type train | step 350 | loss 8.2413 0.0171 20.5432 0.4294 51.0060 1.0694 76.1323 3.4656 0.9179 3.1691 5.5067 7.9792 | lr 4.7e-04 | norm 11.2556 | dt 0.132
type train | step 360 | loss 7.4620 0.0178 19.0219 0.4213 47.0881 1.0348 71.4545 3.2108 0.8518 2.9704 5.2492 7.6228 | lr 4.8e-04 | norm 10.0365 | dt 0.132
type train | step 370 | loss 6.6526 0.0191 17.7954 0.4218 44.5005 1.0499 68.4091 3.1355 0.7784 2.8265 5.0567 7.2580 | lr 4.9e-04 | norm 9.0370 | dt 0.133
type train | step 380 | loss 6.2852 0.0205 16.7045 0.4206 42.4912 1.0479 65.6709 3.1060 0.7477 2.6702 4.8291 7.0065 | lr 5.1e-04 | norm 8.0510 | dt 0.135
type train | step 390 | loss 5.3553 0.0192 15.2660 0.4244 39.2216 0.9844 62.4285 2.8900 0.6880 2.5113 4.4612 6.4669 | lr 5.2e-04 | norm 7.3346 | dt 0.137
type train | step 400 | loss 4.8512 0.0201 14.0875 0.4136 36.5590 0.9850 58.6659 2.8281 0.6392 2.3524 4.2404 6.1500 | lr 5.3e-04 | norm 6.7980 | dt 0.142
type train | step 410 | loss 4.3679 0.0211 13.3005 0.4275 34.8477 0.9817 56.7702 2.7824 0.6097 2.2470 4.0253 5.8175 | lr 5.5e-04 | norm 6.0295 | dt 0.143
type train | step 420 | loss 3.9954 0.0218 12.4372 0.4213 32.7939 0.9927 54.2152 2.7410 0.5710 2.1170 3.8190 5.5720 | lr 5.6e-04 | norm 6.0786 | dt 0.135
type train | step 430 | loss 3.4720 0.0212 11.6331 0.4366 31.0513 0.9764 52.6664 2.7247 0.5251 2.0165 3.5891 5.2008 | lr 5.7e-04 | norm 5.2532 | dt 0.139
type train | step 440 | loss 3.0908 0.0229 10.7369 0.4265 29.1497 0.9807 50.3615 2.7060 0.4862 1.8756 3.3735 4.9163 | lr 5.9e-04 | norm 4.9529 | dt 0.144
type train | step 450 | loss 2.8127 0.0228 10.1596 0.4407 27.9763 0.9714 48.8930 2.6851 0.4402 1.7817 3.1710 4.6046 | lr 6.0e-04 | norm 4.4355 | dt 0.136
type train | step 460 | loss 2.4742 0.0229 9.2590 0.4299 25.7537 0.9473 45.7536 2.6045 0.4056 1.6613 2.9504 4.3015 | lr 6.1e-04 | norm 4.4646 | dt 0.139
type train | step 470 | loss 2.3039 0.0251 8.9929 0.4476 25.3845 0.9481 45.4951 2.6165 0.3833 1.5759 2.7708 4.0718 | lr 6.3e-04 | norm 3.9706 | dt 0.140
type train | step 480 | loss 1.9999 0.0242 8.4305 0.4484 23.8921 0.9478 43.8088 2.6080 0.3489 1.4782 2.5909 3.8285 | lr 6.4e-04 | norm 3.7538 | dt 0.143
type train | step 490 | loss 1.7240 0.0242 7.8379 0.4492 22.5781 0.9568 42.5949 2.5757 0.3274 1.3898 2.4300 3.5887 | lr 6.5e-04 | norm 3.6307 | dt 0.149
type train | step 500 | loss 1.5530 0.0252 7.4849 0.4458 21.8061 0.9633 41.4013 2.6010 0.3108 1.3053 2.2544 3.3713 | lr 6.7e-04 | norm 3.4703 | dt 0.140
type train | step 510 | loss 1.3394 0.0246 7.0849 0.4488 21.0231 0.9552 40.8507 2.5690 0.2991 1.2221 2.0948 3.1422 | lr 6.8e-04 | norm 3.3100 | dt 0.140
type train | step 520 | loss 1.1636 0.0257 6.5411 0.4461 19.9100 0.9609 39.5784 2.5628 0.2828 1.1437 1.9394 2.9257 | lr 6.9e-04 | norm 3.4333 | dt 0.144
type train | step 530 | loss 1.0992 0.0270 6.4261 0.4596 19.5499 0.9722 39.1033 2.5940 0.2669 1.0749 1.8160 2.7936 | lr 7.1e-04 | norm 3.0833 | dt 0.143
type train | step 540 | loss 0.8718 0.0251 5.8496 0.4488 17.9474 0.9487 36.7381 2.4844 0.2466 0.9938 1.6560 2.5816 | lr 7.2e-04 | norm 2.8398 | dt 0.154
type train | step 550 | loss 0.7791 0.0248 5.6453 0.4599 17.4197 0.9685 36.6295 2.5053 0.2353 0.9267 1.5467 2.4198 | lr 7.3e-04 | norm 3.0306 | dt 0.150
type train | step 560 | loss 0.7678 0.0284 5.4386 0.4572 17.1760 0.9762 35.7572 2.5584 0.2124 0.8796 1.4259 2.3201 | lr 7.5e-04 | norm 3.2020 | dt 0.157
type train | step 570 | loss 0.6431 0.0264 5.1719 0.4642 16.7533 0.9712 35.9592 2.5344 0.2045 0.8096 1.3062 2.1531 | lr 7.6e-04 | norm 2.7309 | dt 0.148
type train | step 580 | loss 0.5691 0.0260 4.8115 0.4514 15.3041 0.9580 33.5255 2.4674 0.2022 0.7414 1.1808 2.0070 | lr 7.7e-04 | norm 3.0977 | dt 0.158
type train | step 590 | loss 0.5173 0.0273 4.6674 0.4713 15.4897 0.9789 34.1840 2.5118 0.2049 0.7117 1.1000 1.8893 | lr 7.9e-04 | norm 2.8225 | dt 0.151
type train | step 600 | loss 0.4327 0.0254 4.3483 0.4662 14.1878 0.9548 32.2283 2.4187 0.1945 0.6547 0.9831 1.7405 | lr 8.0e-04 | norm 2.4601 | dt 0.145
type train | step 610 | loss 0.4087 0.0266 4.1871 0.4817 14.2428 0.9764 32.5352 2.4468 0.1848 0.6182 0.9132 1.6613 | lr 8.1e-04 | norm 2.6531 | dt 0.158
type train | step 620 | loss 0.3762 0.0272 4.0029 0.4850 13.5826 0.9620 31.7321 2.4588 0.1791 0.5850 0.8268 1.5588 | lr 8.3e-04 | norm 2.2446 | dt 0.156
type train | step 630 | loss 0.3507 0.0271 3.9041 0.4802 13.4696 0.9868 31.8350 2.4706 0.1832 0.5471 0.7534 1.4605 | lr 8.4e-04 | norm 2.4570 | dt 0.147
type train | step 640 | loss 0.3345 0.0263 3.6032 0.4719 12.5710 0.9800 30.2745 2.4093 0.1905 0.5115 0.6747 1.3612 | lr 8.5e-04 | norm 2.4011 | dt 0.156
type train | step 650 | loss 0.3423 0.0284 3.6889 0.4900 13.1316 1.0371 31.3887 2.4880 0.2128 0.4837 0.6366 1.3427 | lr 8.7e-04 | norm 2.8802 | dt 0.157
type train | step 660 | loss 0.2950 0.0246 3.4106 0.4880 11.7266 0.9655 29.0053 2.3359 0.1772 0.4276 0.5483 1.2071 | lr 8.8e-04 | norm 2.4497 | dt 0.142
type train | step 670 | loss 0.2721 0.0238 3.2194 0.4823 11.4337 0.9882 29.2098 2.3214 0.1610 0.3991 0.4927 1.1333 | lr 8.9e-04 | norm 2.3711 | dt 0.155
type train | step 680 | loss 0.2585 0.0248 3.1271 0.4962 11.1346 1.0025 28.4550 2.3084 0.1643 0.3684 0.4461 1.0774 | lr 9.1e-04 | norm 1.9814 | dt 0.151
type train | step 690 | loss 0.2627 0.0270 3.2002 0.5040 11.8461 1.0685 30.2121 2.4214 0.1638 0.3449 0.4128 1.0585 | lr 9.2e-04 | norm 2.5055 | dt 0.147
type train | step 700 | loss 0.2528 0.0256 2.9773 0.4864 10.6863 1.0659 28.0479 2.3232 0.1450 0.3396 0.3631 0.9832 | lr 9.3e-04 | norm 2.3631 | dt 0.149
type train | step 710 | loss 0.2294 0.0260 2.8030 0.4785 10.2611 1.0567 27.4284 2.2615 0.1207 0.3129 0.3292 0.9189 | lr 9.5e-04 | norm 2.2704 | dt 0.144
type train | step 720 | loss 0.2407 0.0267 2.8422 0.5055 10.1723 1.0858 27.2449 2.2278 0.1050 0.3036 0.3001 0.8791 | lr 9.6e-04 | norm 2.2702 | dt 0.159
type train | step 730 | loss 0.2301 0.0295 2.7709 0.5027 10.0799 1.1237 27.0787 2.2826 0.1013 0.2970 0.2728 0.8419 | lr 9.7e-04 | norm 2.3749 | dt 0.153
type train | step 740 | loss 0.2143 0.0280 2.6354 0.5068 9.6429 1.1823 26.9313 2.2397 0.0926 0.2754 0.2510 0.7988 | lr 9.9e-04 | norm 2.1631 | dt 0.148
type train | step 750 | loss 0.2473 0.0321 2.8072 0.5111 10.2645 1.2578 27.8294 2.3038 0.1014 0.2774 0.2386 0.7790 | lr 1.0e-03 | norm 3.4264 | dt 0.162
type train | step 760 | loss 0.2152 0.0307 2.5607 0.4923 9.4775 1.2763 26.7318 2.2698 0.0877 0.2463 0.2128 0.7155 | lr 1.0e-03 | norm 2.5873 | dt 0.178
type train | step 770 | loss 0.1986 0.0293 2.4676 0.5031 9.1827 1.2911 26.6180 2.2668 0.0712 0.2307 0.1994 0.6605 | lr 1.0e-03 | norm 2.4458 | dt 0.155
type train | step 780 | loss 0.1958 0.0268 2.4547 0.5039 8.8486 1.2804 25.7196 2.1524 0.0699 0.2142 0.1740 0.6212 | lr 1.0e-03 | norm 2.4301 | dt 0.152
type train | step 790 | loss 0.1920 0.0313 2.4077 0.5222 9.0828 1.3757 26.1666 2.2512 0.0629 0.2027 0.1593 0.6152 | lr 1.0e-03 | norm 2.5061 | dt 0.139
type train | step 800 | loss 0.1869 0.0305 2.2986 0.5379 8.7307 1.3936 25.6288 2.2764 0.0582 0.1819 0.1465 0.5970 | lr 1.0e-03 | norm 2.7788 | dt 0.169
type train | step 810 | loss 0.1951 0.0308 2.3819 0.5375 8.9142 1.4688 26.6492 2.3001 0.0648 0.1890 0.1348 0.5794 | lr 1.0e-03 | norm 2.6305 | dt 0.161
type train | step 820 | loss 0.1853 0.0287 2.2528 0.5243 8.3992 1.5018 25.6356 2.2456 0.0667 0.1741 0.1156 0.5375 | lr 1.0e-03 | norm 2.3531 | dt 0.165
type train | step 830 | loss 0.1763 0.0294 2.1788 0.5165 8.1223 1.6009 25.0394 2.2185 0.0714 0.1600 0.1027 0.4972 | lr 1.0e-03 | norm 2.3745 | dt 0.162
type train | step 840 | loss 0.1687 0.0284 2.1510 0.5323 7.9045 1.5890 24.8428 2.1825 0.0798 0.1603 0.0889 0.4716 | lr 1.0e-03 | norm 2.2581 | dt 0.150
type train | step 850 | loss 0.1774 0.0311 2.2093 0.5536 8.3348 1.6481 25.3459 2.2441 0.0686 0.1512 0.0823 0.4762 | lr 1.0e-03 | norm 2.4799 | dt 0.171
type train | step 860 | loss 0.1682 0.0289 2.0431 0.5659 7.6589 1.7627 24.3002 2.2292 0.0821 0.1586 0.0747 0.4592 | lr 1.0e-03 | norm 2.3072 | dt 0.153
type train | step 870 | loss 0.1651 0.0297 2.0458 0.5602 7.6376 1.7779 24.4685 2.2460 0.0696 0.1503 0.0644 0.4308 | lr 1.0e-03 | norm 2.3316 | dt 0.168
type train | step 880 | loss 0.1710 0.0296 2.0241 0.5483 7.4949 1.7640 24.2728 2.2237 0.0671 0.1406 0.0564 0.4220 | lr 1.0e-03 | norm 2.5797 | dt 0.166
type train | step 890 | loss 0.1699 0.0300 2.0334 0.5610 7.6958 1.8521 24.7498 2.3360 0.0606 0.1397 0.0705 0.4060 | lr 1.0e-03 | norm 2.7186 | dt 0.163
type train | step 900 | loss 0.1609 0.0279 1.9374 0.5675 7.1858 1.8180 23.7561 2.2384 0.0522 0.1446 0.0630 0.3965 | lr 1.0e-03 | norm 2.7343 | dt 0.169
type train | step 910 | loss 0.1637 0.0289 2.0366 0.5645 7.5824 1.8131 24.1198 2.2853 0.0455 0.1395 0.0657 0.3909 | lr 1.0e-03 | norm 2.5918 | dt 0.162
type train | step 920 | loss 0.1616 0.0290 1.9014 0.5900 7.2884 1.8348 24.1580 2.2982 0.0508 0.1570 0.0580 0.3914 | lr 1.0e-03 | norm 2.9604 | dt 0.160
type train | step 930 | loss 0.1582 0.0293 1.8580 0.5723 7.1986 1.8449 23.8721 2.3379 0.0482 0.1624 0.0559 0.3654 | lr 1.0e-03 | norm 2.5048 | dt 0.145
type train | step 940 | loss 0.1572 0.0275 1.8848 0.5813 7.2682 1.8197 24.2516 2.3517 0.0456 0.1571 0.0577 0.3487 | lr 1.0e-03 | norm 2.5933 | dt 0.152
type train | step 950 | loss 0.1531 0.0273 1.8350 0.5728 7.0734 1.8041 23.8145 2.3510 0.0491 0.1670 0.0537 0.3233 | lr 1.0e-03 | norm 2.5081 | dt 0.146
type train | step 960 | loss 0.1530 0.0291 1.8493 0.5660 6.9847 1.9020 23.6426 2.3776 0.0429 0.1538 0.0540 0.3404 | lr 1.0e-03 | norm 2.7993 | dt 0.157
type train | step 970 | loss 0.1497 0.0290 1.7506 0.5854 6.7373 1.7938 22.9105 2.3941 0.0442 0.1547 0.0574 0.3258 | lr 1.0e-03 | norm 2.7751 | dt 0.146
type train | step 980 | loss 0.1497 0.0299 1.7720 0.5646 6.7466 1.8855 23.2679 2.4356 0.0465 0.1499 0.0626 0.3199 | lr 1.0e-03 | norm 2.5859 | dt 0.170
type train | step 990 | loss 0.1528 0.0309 1.7973 0.5669 6.9846 1.9481 23.7523 2.5801 0.0459 0.1408 0.0515 0.3183 | lr 1.0e-03 | norm 2.5671 | dt 0.166
type train | step 1000 | loss 0.1452 0.0283 1.7525 0.5432 6.8010 1.8354 23.6432 2.4749 0.0382 0.1319 0.0550 0.2899 | lr 1.0e-03 | norm 2.2313 | dt 0.160
type train | step 1010 | loss 0.1512 0.0291 1.7256 0.5376 6.5489 1.8635 22.7346 2.4801 0.0495 0.1264 0.0491 0.2841 | lr 1.0e-03 | norm 2.3661 | dt 0.167
type train | step 1020 | loss 0.1449 0.0288 1.7195 0.5483 6.5882 1.8668 22.9976 2.5711 0.0425 0.1273 0.0439 0.3117 | lr 1.0e-03 | norm 2.8790 | dt 0.177
type train | step 1030 | loss 0.1453 0.0303 1.7107 0.5716 6.6366 1.9395 22.7793 2.6219 0.0357 0.1292 0.0383 0.3064 | lr 1.0e-03 | norm 3.3492 | dt 0.149
type train | step 1040 | loss 0.1458 0.0286 1.6922 0.5742 6.5926 1.8934 23.1849 2.6775 0.0336 0.1365 0.0415 0.2914 | lr 1.0e-03 | norm 2.5018 | dt 0.143
type train | step 1050 | loss 0.1411 0.0291 1.6276 0.5714 6.3423 1.9756 22.6367 2.7186 0.0314 0.1271 0.0391 0.3032 | lr 1.0e-03 | norm 2.6342 | dt 0.149
type train | step 1060 | loss 0.1453 0.0287 1.6473 0.5739 6.5303 1.9749 23.0142 2.8495 0.0294 0.1457 0.0400 0.2982 | lr 1.0e-03 | norm 2.6839 | dt 0.157
type train | step 1070 | loss 0.1400 0.0290 1.5847 0.5846 6.1461 1.9503 21.9626 2.7648 0.0284 0.1354 0.0356 0.2862 | lr 1.0e-03 | norm 2.5754 | dt 0.144
type train | step 1080 | loss 0.1412 0.0305 1.6159 0.5973 6.3815 1.8546 22.7638 2.8145 0.0358 0.1300 0.0517 0.2899 | lr 9.9e-04 | norm 2.4803 | dt 0.160
type train | step 1090 | loss 0.1396 0.0295 1.6129 0.5890 6.4109 1.9002 22.8222 2.8500 0.0318 0.1318 0.0389 0.2829 | lr 9.9e-04 | norm 2.6591 | dt 0.167
type train | step 1100 | loss 0.1403 0.0294 1.5860 0.5753 6.3868 1.9417 22.8526 2.8189 0.0295 0.1333 0.0362 0.2833 | lr 9.9e-04 | norm 2.8626 | dt 0.153
type train | step 1110 | loss 0.1400 0.0289 1.5808 0.5606 6.2371 1.9324 22.6808 2.9841 0.0280 0.1235 0.0385 0.2942 | lr 9.9e-04 | norm 2.8656 | dt 0.168
type train | step 1120 | loss 0.1384 0.0307 1.5647 0.5847 6.4111 1.8800 23.1989 2.9532 0.0270 0.1197 0.0348 0.2812 | lr 9.9e-04 | norm 2.6484 | dt 0.149
type train | step 1130 | loss 0.1376 0.0287 1.5535 0.5667 6.3405 1.9268 22.9927 2.9543 0.0252 0.1138 0.0380 0.2728 | lr 9.9e-04 | norm 3.0529 | dt 0.183
type train | step 1140 | loss 0.1366 0.0313 1.5580 0.6292 6.3119 1.9586 23.1749 3.0506 0.0250 0.1268 0.0265 0.2766 | lr 9.9e-04 | norm 3.1424 | dt 0.156
type train | step 1150 | loss 0.1312 0.0282 1.5115 0.6223 6.0060 1.9538 21.8915 2.9706 0.0234 0.1289 0.0324 0.2944 | lr 9.9e-04 | norm 3.6670 | dt 0.165
type train | step 1160 | loss 0.1332 0.0289 1.5194 0.6069 6.0979 1.9781 22.3201 3.0050 0.0231 0.1285 0.0342 0.2791 | lr 9.9e-04 | norm 2.9473 | dt 0.165
type train | step 1170 | loss 0.1365 0.0317 1.5293 0.5820 6.1077 1.9575 22.4077 3.1683 0.0233 0.1291 0.0290 0.3037 | lr 9.9e-04 | norm 4.0862 | dt 0.160
type train | step 1180 | loss 0.1333 0.0302 1.5081 0.5774 6.3444 1.9527 23.2477 3.1709 0.0222 0.1247 0.0275 0.2931 | lr 9.9e-04 | norm 3.2224 | dt 0.156
type train | step 1190 | loss 0.1352 0.0284 1.4761 0.5518 5.9525 2.0049 21.7528 3.1658 0.0210 0.1200 0.0263 0.2942 | lr 9.9e-04 | norm 3.4072 | dt 0.160
type train | step 1200 | loss 0.1307 0.0340 1.4883 0.5819 6.0406 1.9942 22.5785 3.2758 0.0212 0.1129 0.0278 0.2803 | lr 9.9e-04 | norm 4.4203 | dt 0.170
type train | step 1210 | loss 0.1277 0.0290 1.4484 0.5570 5.8368 2.0281 21.5333 3.1293 0.0201 0.1125 0.0272 0.2881 | lr 9.9e-04 | norm 3.5143 | dt 0.167
type train | step 1220 | loss 0.1298 0.0304 1.4462 0.6008 6.0973 2.0813 22.5275 3.2128 0.0218 0.1064 0.0248 0.2808 | lr 9.9e-04 | norm 3.0990 | dt 0.167
type train | step 1230 | loss 0.1287 0.0302 1.4413 0.5916 5.9211 1.9645 22.1069 3.2210 0.0207 0.1115 0.0202 0.2786 | lr 9.9e-04 | norm 2.9249 | dt 0.153
type train | step 1240 | loss 0.1287 0.0311 1.4413 0.5990 6.0148 1.9837 22.4970 3.2251 0.0196 0.1099 0.0193 0.2947 | lr 9.9e-04 | norm 3.1744 | dt 0.169
type train | step 1250 | loss 0.1285 0.0295 1.3914 0.5593 5.7623 1.9738 21.4567 3.2493 0.0189 0.1132 0.0222 0.2770 | lr 9.9e-04 | norm 3.0160 | dt 0.160
type train | step 1260 | loss 0.1315 0.0319 1.4800 0.5933 6.0542 2.0102 22.5443 3.4504 0.0238 0.1051 0.0205 0.2927 | lr 9.9e-04 | norm 3.2401 | dt 0.172
type train | step 1270 | loss 0.1263 0.0279 1.4065 0.5999 5.7642 1.9328 21.1766 3.2721 0.0220 0.1145 0.0307 0.2982 | lr 9.9e-04 | norm 3.9363 | dt 0.158
type train | step 1280 | loss 0.1267 0.0285 1.3699 0.5804 5.7520 1.9808 21.7856 3.2868 0.0220 0.1194 0.0281 0.2958 | lr 9.9e-04 | norm 3.6093 | dt 0.166
type train | step 1290 | loss 0.1256 0.0298 1.3797 0.5729 5.6641 2.0001 21.1888 3.2376 0.0214 0.1043 0.0229 0.2843 | lr 9.9e-04 | norm 2.8620 | dt 0.158
type train | step 1300 | loss 0.1300 0.0345 1.4338 0.5813 6.1034 2.0097 22.9624 3.4287 0.0241 0.1091 0.0319 0.2804 | lr 9.9e-04 | norm 3.1904 | dt 0.156
type train | step 1310 | loss 0.1265 0.0314 1.3718 0.5921 5.6885 2.0148 21.5151 3.4208 0.0221 0.0993 0.0251 0.2821 | lr 9.8e-04 | norm 3.0375 | dt 0.164
type train | step 1320 | loss 0.1226 0.0293 1.3175 0.5368 5.5038 1.8835 21.1243 3.2518 0.0209 0.0969 0.0247 0.2861 | lr 9.8e-04 | norm 4.3645 | dt 0.170
type train | step 1330 | loss 0.1263 0.0292 1.3743 0.5930 5.7470 1.9308 21.3114 3.2768 0.0193 0.0952 0.0196 0.2731 | lr 9.8e-04 | norm 3.4259 | dt 0.164
type train | step 1340 | loss 0.1278 0.0332 1.3718 0.6201 5.6971 1.9705 21.1756 3.3029 0.0186 0.0902 0.0274 0.2764 | lr 9.8e-04 | norm 3.4353 | dt 0.159
type train | step 1350 | loss 0.1241 0.0318 1.3185 0.6405 5.5560 2.0864 21.1859 3.3339 0.0210 0.0979 0.0193 0.2628 | lr 9.8e-04 | norm 3.4311 | dt 0.154
type train | step 1360 | loss 0.1357 0.0335 1.4092 0.6600 5.7728 2.1457 21.9196 3.4305 0.0223 0.1139 0.0205 0.2880 | lr 9.8e-04 | norm 4.4697 | dt 0.159
type train | step 1370 | loss 0.1271 0.0321 1.3682 0.5974 5.7476 2.0966 21.5590 3.4343 0.0231 0.1002 0.0228 0.2962 | lr 9.8e-04 | norm 4.7526 | dt 0.166
type train | step 1380 | loss 0.1216 0.0303 1.3169 0.5740 5.5991 1.9858 21.6082 3.3125 0.0212 0.0950 0.0255 0.2782 | lr 9.8e-04 | norm 3.4931 | dt 0.158
type train | step 1390 | loss 0.1205 0.0291 1.3444 0.6324 5.5955 1.9509 20.9781 3.2834 0.0220 0.0823 0.0222 0.2765 | lr 9.8e-04 | norm 4.4385 | dt 0.165
type train | step 1400 | loss 0.1228 0.0323 1.3387 0.6152 5.6372 2.0684 21.4244 3.4253 0.0199 0.1143 0.0186 0.2840 | lr 9.8e-04 | norm 3.6743 | dt 0.169
type train | step 1410 | loss 0.1221 0.0318 1.2988 0.6406 5.5364 2.0084 21.0927 3.3845 0.0195 0.1138 0.0250 0.2937 | lr 9.8e-04 | norm 4.0860 | dt 0.164
type train | step 1420 | loss 0.1277 0.0329 1.3620 0.6231 5.7690 2.1117 22.0932 3.3403 0.0194 0.1195 0.0186 0.2914 | lr 9.8e-04 | norm 3.6370 | dt 0.162
type train | step 1430 | loss 0.1234 0.0327 1.3042 0.5731 5.6194 2.1109 21.3048 3.3781 0.0245 0.1287 0.0165 0.2812 | lr 9.8e-04 | norm 3.5372 | dt 0.151
type train | step 1440 | loss 0.1194 0.0296 1.2801 0.6355 5.4981 2.0030 21.1494 3.1911 0.0232 0.0995 0.0160 0.2793 | lr 9.8e-04 | norm 3.3933 | dt 0.152
type train | step 1450 | loss 0.1171 0.0300 1.2929 0.5853 5.4713 2.0025 20.9872 3.2851 0.0241 0.0945 0.0158 0.2667 | lr 9.8e-04 | norm 4.0307 | dt 0.162
type train | step 1460 | loss 0.1234 0.0327 1.3379 0.6086 5.7002 2.0773 21.4659 3.3914 0.0271 0.0895 0.0257 0.2741 | lr 9.8e-04 | norm 4.2683 | dt 0.170
type train | step 1470 | loss 0.1191 0.0305 1.2582 0.6503 5.3527 2.1132 20.6206 3.3060 0.0228 0.1001 0.0218 0.2761 | lr 9.7e-04 | norm 4.6720 | dt 0.158
type train | step 1480 | loss 0.1174 0.0298 1.2594 0.6613 5.4200 2.1254 20.9988 3.3247 0.0207 0.1160 0.0222 0.2685 | lr 9.7e-04 | norm 3.4133 | dt 0.154
type train | step 1490 | loss 0.1213 0.0291 1.2708 0.6113 5.3654 1.9700 20.7755 3.3401 0.0202 0.1364 0.0297 0.2679 | lr 9.7e-04 | norm 5.4592 | dt 0.154
type train | step 1500 | loss 0.1212 0.0304 1.2803 0.6318 5.5378 1.9542 21.3887 3.4024 0.0189 0.1161 0.0338 0.2711 | lr 9.7e-04 | norm 3.9460 | dt 0.163
type train | step 1510 | loss 0.1176 0.0288 1.2508 0.6657 5.2734 1.9619 20.6231 3.3050 0.0183 0.1213 0.0279 0.2791 | lr 9.7e-04 | norm 3.8110 | dt 0.147
type train | step 1520 | loss 0.1211 0.0302 1.3124 0.6396 5.5321 1.9714 20.9166 3.3438 0.0190 0.1220 0.0328 0.2770 | lr 9.7e-04 | norm 4.5113 | dt 0.149
type train | step 1530 | loss 0.1198 0.0310 1.2454 0.6422 5.4260 2.0412 20.9783 3.3941 0.0196 0.1214 0.0249 0.2818 | lr 9.7e-04 | norm 3.5865 | dt 0.148
type train | step 1540 | loss 0.1181 0.0332 1.2307 0.6101 5.4435 2.0130 20.9358 3.4211 0.0239 0.1126 0.0258 0.2774 | lr 9.7e-04 | norm 3.2817 | dt 0.158
type train | step 1550 | loss 0.1183 0.0300 1.2501 0.6241 5.5353 2.0881 21.3656 3.3535 0.0208 0.1124 0.0229 0.2645 | lr 9.7e-04 | norm 4.0002 | dt 0.173
type train | step 1560 | loss 0.1150 0.0305 1.2401 0.6284 5.4311 2.0967 21.0850 3.2376 0.0197 0.1272 0.0197 0.2588 | lr 9.7e-04 | norm 3.6704 | dt 0.170
type train | step 1570 | loss 0.1167 0.0298 1.2538 0.6180 5.4140 2.0155 20.9618 3.4101 0.0201 0.1239 0.0216 0.2640 | lr 9.7e-04 | norm 5.8878 | dt 0.144
type train | step 1580 | loss 0.1149 0.0308 1.2017 0.6295 5.2149 2.0178 20.3236 3.1828 0.0181 0.1076 0.0151 0.2824 | lr 9.7e-04 | norm 3.5438 | dt 0.152
type train | step 1590 | loss 0.1156 0.0318 1.2145 0.6396 5.2022 2.2012 20.6212 3.2165 0.0168 0.1047 0.0127 0.2827 | lr 9.7e-04 | norm 3.9175 | dt 0.149
type train | step 1600 | loss 0.1192 0.0336 1.2535 0.6319 5.4402 2.1517 21.1736 3.4245 0.0174 0.1198 0.0142 0.2888 | lr 9.7e-04 | norm 3.9197 | dt 0.150
type train | step 1610 | loss 0.1131 0.0311 1.2113 0.6378 5.4061 2.0702 21.2357 3.4799 0.0168 0.1284 0.0118 0.2602 | lr 9.6e-04 | norm 4.8819 | dt 0.160
type train | step 1620 | loss 0.1178 0.0305 1.2250 0.6626 5.2638 1.9849 20.4084 3.3047 0.0175 0.1159 0.0168 0.2811 | lr 9.6e-04 | norm 4.0427 | dt 0.168
type train | step 1630 | loss 0.1141 0.0326 1.2212 0.6139 5.3311 2.1047 20.6952 3.4446 0.0173 0.1474 0.0142 0.2596 | lr 9.6e-04 | norm 6.2942 | dt 0.166
type train | step 1640 | loss 0.1150 0.0323 1.2265 0.6101 5.2969 2.0687 20.5341 3.3071 0.0180 0.1190 0.0118 0.2946 | lr 9.6e-04 | norm 4.4476 | dt 0.144
type train | step 1650 | loss 0.1158 0.0311 1.2069 0.6367 5.3587 2.1157 20.9154 3.2723 0.0164 0.1281 0.0112 0.2734 | lr 9.6e-04 | norm 5.0082 | dt 0.158
type train | step 1660 | loss 0.1125 0.0317 1.1866 0.6152 5.1626 2.0884 20.3453 3.2236 0.0177 0.1148 0.0116 0.2679 | lr 9.6e-04 | norm 3.8196 | dt 0.166
type train | step 1670 | loss 0.1166 0.0318 1.2099 0.6073 5.3766 2.1344 20.8555 3.3196 0.0182 0.1135 0.0120 0.2635 | lr 9.6e-04 | norm 3.9559 | dt 0.162
type train | step 1680 | loss 0.1115 0.0315 1.1650 0.5994 5.0496 2.1312 19.9056 3.4309 0.0168 0.0970 0.0092 0.2815 | lr 9.6e-04 | norm 4.2854 | dt 0.168
type train | step 1690 | loss 0.1139 0.0330 1.1994 0.6350 5.2424 2.1082 20.7280 3.4040 0.0199 0.1075 0.0100 0.2735 | lr 9.6e-04 | norm 4.3113 | dt 0.162
type train | step 1700 | loss 0.1134 0.0305 1.2002 0.5999 5.3200 2.0953 20.8618 3.3034 0.0181 0.1121 0.0096 0.2874 | lr 9.6e-04 | norm 4.6850 | dt 0.152
type train | step 1710 | loss 0.1137 0.0294 1.1953 0.6156 5.3733 2.0637 20.9519 3.4463 0.0158 0.1097 0.0130 0.2917 | lr 9.6e-04 | norm 4.3697 | dt 0.163
type train | step 1720 | loss 0.1137 0.0294 1.1970 0.6140 5.2179 2.0749 20.6794 3.4374 0.0177 0.1019 0.0086 0.2934 | lr 9.5e-04 | norm 4.2557 | dt 0.166
type train | step 1730 | loss 0.1135 0.0305 1.1796 0.6289 5.4072 2.0657 21.2403 3.3843 0.0187 0.1082 0.0097 0.2654 | lr 9.5e-04 | norm 4.3736 | dt 0.148
type train | step 1740 | loss 0.1122 0.0305 1.1773 0.6220 5.3331 2.0888 21.1164 3.3675 0.0174 0.1004 0.0092 0.2723 | lr 9.5e-04 | norm 4.2435 | dt 0.147
type train | step 1750 | loss 0.1124 0.0334 1.1820 0.5873 5.3066 2.0882 21.3249 3.3117 0.0173 0.1069 0.0081 0.2742 | lr 9.5e-04 | norm 3.8985 | dt 0.151
type train | step 1760 | loss 0.1080 0.0311 1.1571 0.5843 5.1095 2.1416 20.2010 3.3671 0.0173 0.1071 0.0071 0.2801 | lr 9.5e-04 | norm 5.3036 | dt 0.146
type train | step 1770 | loss 0.1097 0.0309 1.1603 0.6164 5.1984 2.2004 20.6270 3.3438 0.0148 0.0957 0.0095 0.2743 | lr 9.5e-04 | norm 4.5342 | dt 0.167
type train | step 1780 | loss 0.1136 0.0330 1.1867 0.6173 5.2387 2.1717 20.7066 3.4224 0.0150 0.1117 0.0087 0.2873 | lr 9.5e-04 | norm 3.9402 | dt 0.165
type train | step 1790 | loss 0.1105 0.0317 1.1593 0.6280 5.4068 2.1319 21.5003 3.3679 0.0144 0.1017 0.0109 0.2936 | lr 9.5e-04 | norm 3.6540 | dt 0.162
type train | step 1800 | loss 0.1118 0.0317 1.1509 0.5996 5.1728 2.2233 20.1529 3.3409 0.0141 0.1052 0.0096 0.2964 | lr 9.5e-04 | norm 4.1322 | dt 0.145
type train | step 1810 | loss 0.1092 0.0317 1.1606 0.5823 5.1819 2.1500 20.8738 3.3327 0.0198 0.1161 0.0131 0.2776 | lr 9.5e-04 | norm 4.2224 | dt 0.163
type train | step 1820 | loss 0.1060 0.0297 1.1365 0.5743 5.0964 2.1480 20.0471 3.2695 0.0153 0.0986 0.0105 0.2963 | lr 9.5e-04 | norm 3.9246 | dt 0.164
type train | step 1830 | loss 0.1082 0.0304 1.1320 0.6217 5.3638 2.1215 21.0081 3.3746 0.0136 0.0899 0.0117 0.2864 | lr 9.4e-04 | norm 3.9794 | dt 0.162
type train | step 1840 | loss 0.1078 0.0312 1.1322 0.6115 5.1855 2.1302 20.6349 3.3802 0.0135 0.0916 0.0100 0.2849 | lr 9.4e-04 | norm 4.0231 | dt 0.158
type train | step 1850 | loss 0.1082 0.0310 1.1401 0.5996 5.2449 2.0784 20.9642 3.2969 0.0137 0.0991 0.0104 0.2967 | lr 9.4e-04 | norm 3.6725 | dt 0.157
type train | step 1860 | loss 0.1076 0.0306 1.0929 0.5976 5.0415 2.1187 20.0675 3.3070 0.0140 0.0957 0.0111 0.2787 | lr 9.4e-04 | norm 3.7541 | dt 0.153
type train | step 1870 | loss 0.1105 0.0336 1.1693 0.6077 5.2470 2.1079 20.9753 3.4367 0.0144 0.0985 0.0098 0.2901 | lr 9.4e-04 | norm 3.8940 | dt 0.163
type train | step 1880 | loss 0.1057 0.0273 1.1198 0.6019 5.0852 2.0658 19.7647 3.2196 0.0135 0.1149 0.0116 0.2844 | lr 9.4e-04 | norm 3.3943 | dt 0.161
type train | step 1890 | loss 0.1059 0.0276 1.0931 0.5942 5.0952 2.0412 20.4724 3.1794 0.0128 0.1124 0.0089 0.2892 | lr 9.4e-04 | norm 4.1786 | dt 0.166
type train | step 1900 | loss 0.1052 0.0276 1.1000 0.6252 5.0302 2.1194 19.9343 3.2559 0.0128 0.1044 0.0085 0.3011 | lr 9.4e-04 | norm 5.0246 | dt 0.162
type train | step 1910 | loss 0.1100 0.0315 1.1433 0.6161 5.3907 2.1122 21.5760 3.4048 0.0131 0.1048 0.0092 0.2741 | lr 9.4e-04 | norm 4.0667 | dt 0.157
type train | step 1920 | loss 0.1057 0.0297 1.1069 0.6028 5.0664 2.1894 20.1270 3.2738 0.0127 0.1068 0.0074 0.2987 | lr 9.3e-04 | norm 3.9389 | dt 0.143
type train | step 1930 | loss 0.1034 0.0302 1.0585 0.6010 4.8874 2.0829 19.9014 3.1842 0.0122 0.1000 0.0084 0.2684 | lr 9.3e-04 | norm 4.1599 | dt 0.157
type train | step 1940 | loss 0.1067 0.0309 1.1138 0.5952 5.1357 2.1014 20.0839 3.2664 0.0125 0.0903 0.0095 0.2856 | lr 9.3e-04 | norm 4.0292 | dt 0.155
type train | step 1950 | loss 0.1081 0.0335 1.1078 0.6202 5.0332 2.1693 19.9505 3.3197 0.0125 0.0840 0.0093 0.3062 | lr 9.3e-04 | norm 4.5636 | dt 0.152
type train | step 1960 | loss 0.1045 0.0314 1.0656 0.6437 4.9419 2.2073 19.9766 3.4829 0.0117 0.0735 0.0092 0.2675 | lr 9.3e-04 | norm 5.9235 | dt 0.163
type train | step 1970 | loss 0.1142 0.0335 1.1378 0.6419 5.1342 2.1485 20.4675 3.4349 0.0124 0.0859 0.0088 0.2846 | lr 9.3e-04 | norm 4.6706 | dt 0.164
type train | step 1980 | loss 0.1073 0.0312 1.1136 0.6322 5.1709 2.1076 20.4939 3.3625 0.0120 0.0810 0.0094 0.3097 | lr 9.3e-04 | norm 3.9532 | dt 0.161
type train | step 1990 | loss 0.1038 0.0307 1.0743 0.5967 5.0274 2.0979 20.4124 3.3435 0.0118 0.0859 0.0068 0.3172 | lr 9.3e-04 | norm 4.7831 | dt 0.160
type train | step 2000 | loss 0.1020 0.0294 1.0970 0.5959 5.0487 2.0644 19.8839 3.1417 0.0114 0.0846 0.0066 0.3282 | lr 9.3e-04 | norm 5.0531 | dt 0.163
type train | step 2010 | loss 0.1043 0.0333 1.0819 0.6222 5.0458 2.1101 20.2851 3.1932 0.0123 0.0771 0.0086 0.3241 | lr 9.2e-04 | norm 3.7508 | dt 0.154
type train | step 2020 | loss 0.1032 0.0315 1.0569 0.6023 4.9553 2.1864 19.9414 3.2776 0.0112 0.0733 0.0100 0.3080 | lr 9.2e-04 | norm 3.8339 | dt 0.160
type train | step 2030 | loss 0.1090 0.0330 1.1170 0.6122 5.1934 2.1045 20.8576 3.3408 0.0119 0.0834 0.0117 0.3119 | lr 9.2e-04 | norm 3.7793 | dt 0.163
type train | step 2040 | loss 0.1051 0.0315 1.0710 0.6005 5.0605 2.1272 20.1881 3.2360 0.0136 0.0814 0.0094 0.3100 | lr 9.2e-04 | norm 4.1615 | dt 0.172
type train | step 2050 | loss 0.1020 0.0295 1.0652 0.5797 5.0071 2.1042 20.1184 3.2048 0.0125 0.0789 0.0078 0.3011 | lr 9.2e-04 | norm 4.2946 | dt 0.165
type train | step 2060 | loss 0.1001 0.0295 1.0603 0.6010 5.0015 2.0772 19.9621 3.2107 0.0122 0.0733 0.0072 0.3029 | lr 9.2e-04 | norm 3.9836 | dt 0.147
type train | step 2070 | loss 0.1056 0.0328 1.1080 0.6017 5.1530 2.1112 20.4970 3.2128 0.0127 0.0740 0.0065 0.3159 | lr 9.2e-04 | norm 3.7575 | dt 0.159
type train | step 2080 | loss 0.1021 0.0324 1.0365 0.6086 4.8662 2.1346 19.6530 3.2585 0.0125 0.0620 0.0070 0.3323 | lr 9.2e-04 | norm 3.8905 | dt 0.162
type train | step 2090 | loss 0.1011 0.0315 1.0415 0.6092 4.9351 2.0955 19.9807 3.3973 0.0118 0.0586 0.0068 0.3091 | lr 9.2e-04 | norm 4.1727 | dt 0.169
type train | step 2100 | loss 0.1044 0.0312 1.0492 0.5745 4.9042 2.1398 19.8242 3.4080 0.0115 0.0600 0.0068 0.2939 | lr 9.1e-04 | norm 7.1788 | dt 0.165
type train | step 2110 | loss 0.1038 0.0329 1.0701 0.5750 5.0708 2.1603 20.3448 3.3741 0.0118 0.0644 0.0080 0.3219 | lr 9.1e-04 | norm 3.8887 | dt 0.155
type train | step 2120 | loss 0.1009 0.0309 1.0405 0.5771 4.8598 2.1976 19.6771 3.1100 0.0112 0.0566 0.0093 0.3436 | lr 9.1e-04 | norm 3.6615 | dt 0.149
type train | step 2130 | loss 0.1031 0.0323 1.0825 0.6021 5.0430 2.1083 19.9954 3.1471 0.0109 0.0580 0.0069 0.3176 | lr 9.1e-04 | norm 3.6584 | dt 0.160
type train | step 2140 | loss 0.1029 0.0316 1.0364 0.5792 4.9619 2.1308 20.0384 3.2283 0.0112 0.0582 0.0094 0.3282 | lr 9.1e-04 | norm 3.7238 | dt 0.160
type train | step 2150 | loss 0.1014 0.0321 1.0261 0.6012 4.9795 2.1007 19.9506 3.1609 0.0111 0.0567 0.0071 0.3234 | lr 9.1e-04 | norm 5.2722 | dt 0.162
type train | step 2160 | loss 0.1016 0.0309 1.0498 0.6387 5.0884 2.1541 20.4352 3.1736 0.0117 0.0576 0.0079 0.3005 | lr 9.1e-04 | norm 5.0279 | dt 0.162
type train | step 2170 | loss 0.0982 0.0310 1.0421 0.5833 5.0154 2.1243 20.1994 3.2975 0.0128 0.0559 0.0098 0.2894 | lr 9.1e-04 | norm 3.9589 | dt 0.142
type train | step 2180 | loss 0.1000 0.0317 1.0391 0.6226 5.0248 2.1032 20.0937 3.2579 0.0157 0.0602 0.0061 0.3009 | lr 9.0e-04 | norm 4.4491 | dt 0.165
type train | step 2190 | loss 0.0988 0.0309 0.9993 0.5954 4.8044 2.1401 19.4386 3.2598 0.0196 0.0577 0.0071 0.3023 | lr 9.0e-04 | norm 4.1146 | dt 0.163
type train | step 2200 | loss 0.1000 0.0323 1.0074 0.5902 4.7975 2.2017 19.8201 3.4410 0.0238 0.0576 0.0065 0.3057 | lr 9.0e-04 | norm 5.0862 | dt 0.154
type train | step 2210 | loss 0.1033 0.0326 1.0554 0.6218 4.9892 2.2370 20.2349 3.3876 0.0196 0.0654 0.0078 0.3175 | lr 9.0e-04 | norm 4.1150 | dt 0.160
type train | step 2220 | loss 0.0981 0.0315 1.0177 0.5986 5.0267 2.1462 20.3958 3.1677 0.0256 0.0686 0.0069 0.3000 | lr 9.0e-04 | norm 3.9968 | dt 0.165
type train | step 2230 | loss 0.1021 0.0308 1.0414 0.6069 4.8727 2.1671 19.5839 3.1695 0.0225 0.0650 0.0076 0.2833 | lr 9.0e-04 | norm 4.0279 | dt 0.157
type train | step 2240 | loss 0.0993 0.0322 1.0289 0.6072 4.9657 2.2070 19.8479 3.1821 0.0268 0.0725 0.0096 0.2976 | lr 9.0e-04 | norm 3.7163 | dt 0.158
type train | step 2250 | loss 0.1002 0.0317 1.0506 0.5898 4.9040 2.2199 19.7593 3.1780 0.0379 0.0598 0.0103 0.3154 | lr 8.9e-04 | norm 4.1046 | dt 0.160
type train | step 2260 | loss 0.1015 0.0304 1.0232 0.5881 4.9941 2.2546 20.1889 3.3015 0.0324 0.0600 0.0100 0.3215 | lr 8.9e-04 | norm 4.0369 | dt 0.158
type train | step 2270 | loss 0.0987 0.0305 1.0072 0.5869 4.7782 2.2474 19.5918 3.2274 0.0295 0.0657 0.0088 0.3184 | lr 8.9e-04 | norm 4.1184 | dt 0.155
type train | step 2280 | loss 0.1029 0.0296 1.0318 0.5909 5.0031 2.2431 20.1184 3.2267 0.0306 0.0663 0.0133 0.2892 | lr 8.9e-04 | norm 4.2488 | dt 0.155
type train | step 2290 | loss 0.0988 0.0293 0.9987 0.5992 4.6956 2.1868 19.1430 3.2320 0.0312 0.0633 0.0124 0.3090 | lr 8.9e-04 | norm 5.1617 | dt 0.160
type train | step 2300 | loss 0.1008 0.0298 1.0253 0.6313 4.8667 2.0372 19.9742 3.1379 0.0271 0.0649 0.0269 0.3160 | lr 8.9e-04 | norm 3.9399 | dt 0.158
type train | step 2310 | loss 0.1008 0.0308 1.0263 0.6024 4.9583 2.1472 20.1153 3.2029 0.0284 0.0725 0.0180 0.3078 | lr 8.9e-04 | norm 3.7563 | dt 0.159
type train | step 2320 | loss 0.1015 0.0292 1.0268 0.5956 5.0313 2.0585 20.2082 3.3480 0.0253 0.0664 0.0284 0.3004 | lr 8.9e-04 | norm 4.8530 | dt 0.154
type train | step 2330 | loss 0.1019 0.0289 1.0223 0.6109 4.8579 2.1031 19.9544 3.2895 0.0255 0.0777 0.0180 0.3203 | lr 8.8e-04 | norm 4.0488 | dt 0.156
type train | step 2340 | loss 0.1010 0.0281 1.0169 0.5983 5.0506 2.0316 20.4904 3.2070 0.0234 0.0852 0.0144 0.3054 | lr 8.8e-04 | norm 4.1326 | dt 0.171
type train | step 2350 | loss 0.0996 0.0283 1.0199 0.6260 5.0138 2.0369 20.3346 3.3438 0.0261 0.0982 0.0146 0.2964 | lr 8.8e-04 | norm 4.6842 | dt 0.171
type train | step 2360 | loss 0.1004 0.0316 1.0236 0.6374 4.9801 2.1001 20.5428 3.2331 0.0293 0.0947 0.0145 0.2885 | lr 8.8e-04 | norm 4.5177 | dt 0.164
type train | step 2370 | loss 0.0975 0.0294 1.0021 0.5959 4.8029 2.0799 19.4461 3.1863 0.0272 0.0948 0.0139 0.3210 | lr 8.8e-04 | norm 3.8897 | dt 0.166
type train | step 2380 | loss 0.0981 0.0288 1.0040 0.6018 4.8810 2.1458 19.9171 3.2454 0.0267 0.0941 0.0107 0.3081 | lr 8.8e-04 | norm 4.1693 | dt 0.148
type train | step 2390 | loss 0.1025 0.0320 1.0273 0.6110 4.9052 2.0866 20.0014 3.2785 0.0270 0.0996 0.0117 0.2845 | lr 8.8e-04 | norm 4.2317 | dt 0.166
type train | step 2400 | loss 0.0999 0.0312 1.0144 0.6192 5.0867 2.0880 20.7832 3.3074 0.0248 0.0915 0.0112 0.2646 | lr 8.7e-04 | norm 4.3228 | dt 0.164
type train | step 2410 | loss 0.1011 0.0301 1.0101 0.6521 4.8586 2.1680 19.4971 3.2903 0.0229 0.1063 0.0108 0.3031 | lr 8.7e-04 | norm 4.1214 | dt 0.162
type train | step 2420 | loss 0.0994 0.0327 1.0143 0.6389 4.8837 2.1628 20.2638 3.3275 0.0222 0.1080 0.0108 0.3102 | lr 8.7e-04 | norm 3.8364 | dt 0.161
type train | step 2430 | loss 0.0965 0.0302 0.9874 0.6398 4.8169 2.1181 19.3404 3.1780 0.0223 0.1206 0.0087 0.3051 | lr 8.7e-04 | norm 3.6255 | dt 0.155
type train | step 2440 | loss 0.0984 0.0314 1.0041 0.6241 5.0620 2.1297 20.3704 3.1801 0.0255 0.1207 0.0104 0.3344 | lr 8.7e-04 | norm 3.8368 | dt 0.158
type train | step 2450 | loss 0.0983 0.0306 0.9976 0.6229 4.8883 2.1321 20.0173 3.2102 0.0240 0.1351 0.0100 0.3149 | lr 8.7e-04 | norm 3.3316 | dt 0.151
type train | step 2460 | loss 0.0991 0.0301 1.0065 0.6314 4.9609 2.0544 20.3356 3.3792 0.0193 0.1322 0.0082 0.3120 | lr 8.6e-04 | norm 3.9722 | dt 0.160
type train | step 2470 | loss 0.0981 0.0296 0.9779 0.5933 4.7674 2.0664 19.4709 3.3078 0.0197 0.1198 0.0091 0.2971 | lr 8.6e-04 | norm 4.1307 | dt 0.152
type train | step 2480 | loss 0.1012 0.0328 1.0334 0.6210 4.9574 2.2024 20.3399 3.4190 0.0198 0.1128 0.0082 0.3021 | lr 8.6e-04 | norm 5.2692 | dt 0.157
type train | step 2490 | loss 0.0968 0.0299 0.9918 0.6212 4.7993 2.1225 19.1551 3.4336 0.0200 0.1011 0.0104 0.3170 | lr 8.6e-04 | norm 7.0308 | dt 0.165
type train | step 2500 | loss 0.0962 0.0295 0.9767 0.6125 4.8359 2.0852 19.8097 3.2172 0.0178 0.1085 0.0091 0.3168 | lr 8.6e-04 | norm 5.7430 | dt 0.145
type train | step 2510 | loss 0.0968 0.0299 0.9810 0.5894 4.7769 2.1469 19.3467 3.1624 0.0181 0.1014 0.0096 0.3214 | lr 8.6e-04 | norm 3.7814 | dt 0.147
type train | step 2520 | loss 0.1017 0.0311 1.0351 0.6095 5.1075 2.0802 20.9270 3.2106 0.0201 0.1048 0.0102 0.3237 | lr 8.6e-04 | norm 3.7542 | dt 0.154
type train | step 2530 | loss 0.0971 0.0298 0.9958 0.5914 4.8147 2.0605 19.6264 3.2701 0.0229 0.0919 0.0088 0.3262 | lr 8.5e-04 | norm 8.5416 | dt 0.164
type train | step 2540 | loss 0.0952 0.0295 0.9502 0.6085 4.6632 2.0177 19.3270 3.0119 0.0230 0.0948 0.0082 0.3182 | lr 8.5e-04 | norm 3.5847 | dt 0.159
type train | step 2550 | loss 0.0985 0.0311 1.0047 0.5990 4.8918 2.0371 19.4960 3.1147 0.0247 0.0921 0.0085 0.3193 | lr 8.5e-04 | norm 4.4551 | dt 0.163
type train | step 2560 | loss 0.1004 0.0298 0.9984 0.6146 4.7510 2.1407 19.3454 3.2523 0.0225 0.0865 0.0076 0.3187 | lr 8.5e-04 | norm 3.8051 | dt 0.165
type train | step 2570 | loss 0.0976 0.0311 0.9570 0.5809 4.6999 2.1660 19.4569 3.2967 0.0204 0.0889 0.0069 0.3058 | lr 8.5e-04 | norm 4.2835 | dt 0.162
type train | step 2580 | loss 0.1064 0.0330 1.0215 0.6000 4.8573 2.1355 19.7478 3.2044 0.0209 0.0990 0.0071 0.3395 | lr 8.5e-04 | norm 4.7481 | dt 0.150
type train | step 2590 | loss 0.0992 0.0333 1.0076 0.6219 4.9182 2.0848 19.9749 3.2406 0.0226 0.0825 0.0066 0.3193 | lr 8.4e-04 | norm 5.1750 | dt 0.164
type train | step 2600 | loss 0.0961 0.0295 0.9662 0.6216 4.7903 2.0417 19.8984 3.1730 0.0219 0.0852 0.0074 0.3127 | lr 8.4e-04 | norm 5.3557 | dt 0.162
type train | step 2610 | loss 0.0954 0.0282 0.9922 0.5909 4.8335 2.0633 19.4222 3.0999 0.0206 0.0837 0.0065 0.3222 | lr 8.4e-04 | norm 4.2962 | dt 0.155
type train | step 2620 | loss 0.0978 0.0305 0.9791 0.6292 4.7810 2.0679 19.6738 3.3126 0.0203 0.0758 0.0083 0.3141 | lr 8.4e-04 | norm 4.0562 | dt 0.159
type train | step 2630 | loss 0.0967 0.0303 0.9595 0.5956 4.7385 2.1230 19.4576 3.2831 0.0172 0.0941 0.0071 0.3201 | lr 8.4e-04 | norm 5.0488 | dt 0.158
type train | step 2640 | loss 0.1024 0.0322 1.0142 0.6208 4.9365 2.1510 20.2793 3.2863 0.0192 0.0921 0.0068 0.3107 | lr 8.4e-04 | norm 4.4036 | dt 0.143
type train | step 2650 | loss 0.0988 0.0303 0.9805 0.6062 4.8038 2.1178 19.6600 3.2109 0.0198 0.0780 0.0073 0.3069 | lr 8.4e-04 | norm 4.8560 | dt 0.156
type train | step 2660 | loss 0.0961 0.0277 0.9713 0.6010 4.8049 2.1376 19.6592 3.1373 0.0186 0.0842 0.0064 0.3128 | lr 8.3e-04 | norm 3.9135 | dt 0.148
type train | step 2670 | loss 0.0940 0.0273 0.9669 0.5920 4.8027 2.0758 19.4809 3.0934 0.0164 0.0865 0.0057 0.3313 | lr 8.3e-04 | norm 4.3524 | dt 0.167
type train | step 2680 | loss 0.0995 0.0322 1.0109 0.6091 4.9130 2.1494 19.9758 3.1624 0.0174 0.0858 0.0059 0.3104 | lr 8.3e-04 | norm 4.2553 | dt 0.145
type train | step 2690 | loss 0.0965 0.0295 0.9453 0.5750 4.6769 2.1252 19.1912 3.2028 0.0158 0.0851 0.0066 0.3048 | lr 8.3e-04 | norm 4.0853 | dt 0.162
type train | step 2700 | loss 0.0949 0.0290 0.9547 0.6132 4.7160 2.1600 19.5345 3.2414 0.0172 0.0722 0.0060 0.2965 | lr 8.3e-04 | norm 3.8661 | dt 0.151
type train | step 2710 | loss 0.0988 0.0288 0.9669 0.6176 4.6940 2.1511 19.2573 3.0966 0.0160 0.0799 0.0064 0.2978 | lr 8.3e-04 | norm 3.9197 | dt 0.162
type train | step 2720 | loss 0.0983 0.0294 0.9830 0.5896 4.8774 2.1766 19.8251 3.3679 0.0159 0.0771 0.0074 0.3015 | lr 8.2e-04 | norm 4.1674 | dt 0.164
type train | step 2730 | loss 0.0953 0.0279 0.9545 0.5916 4.6833 2.1692 19.2373 3.1844 0.0176 0.0652 0.0064 0.3100 | lr 8.2e-04 | norm 4.1775 | dt 0.167
type train | step 2740 | loss 0.0982 0.0303 0.9970 0.6203 4.8335 2.2345 19.5034 3.2680 0.0155 0.0677 0.0067 0.3041 | lr 8.2e-04 | norm 4.1814 | dt 0.161
type train | step 2750 | loss 0.0974 0.0327 0.9565 0.6197 4.7705 2.1709 19.6511 3.2036 0.0175 0.0848 0.0084 0.2820 | lr 8.2e-04 | norm 4.1272 | dt 0.143
type train | step 2760 | loss 0.0966 0.0306 0.9460 0.5925 4.7653 2.1438 19.5329 3.2653 0.0173 0.0769 0.0103 0.2932 | lr 8.2e-04 | norm 5.4727 | dt 0.157
type train | step 2770 | loss 0.0964 0.0307 0.9743 0.5930 4.8928 2.1675 20.0108 3.1770 0.0170 0.0769 0.0093 0.3055 | lr 8.2e-04 | norm 3.8526 | dt 0.166
type train | step 2780 | loss 0.0937 0.0284 0.9611 0.6041 4.8306 2.1338 19.8332 3.0939 0.0167 0.0798 0.0082 0.2949 | lr 8.1e-04 | norm 4.2835 | dt 0.147
type train | step 2790 | loss 0.0952 0.0281 0.9567 0.6214 4.8459 2.1360 19.7013 3.1361 0.0161 0.0668 0.0067 0.3067 | lr 8.1e-04 | norm 3.8799 | dt 0.164
type train | step 2800 | loss 0.0943 0.0284 0.9279 0.5992 4.5943 2.1303 18.9463 3.0646 0.0151 0.0720 0.0066 0.3117 | lr 8.1e-04 | norm 3.5088 | dt 0.162
type train | step 2810 | loss 0.0953 0.0298 0.9388 0.6111 4.6246 2.2449 19.4386 3.2203 0.0143 0.0664 0.0063 0.3098 | lr 8.1e-04 | norm 4.5189 | dt 0.149
type train | step 2820 | loss 0.0979 0.0300 0.9782 0.5960 4.7871 2.2159 19.7979 3.2837 0.0136 0.0713 0.0073 0.3070 | lr 8.1e-04 | norm 3.5667 | dt 0.147
type train | step 2830 | loss 0.0933 0.0280 0.9503 0.6128 4.8513 2.1389 19.9881 3.1118 0.0132 0.0607 0.0066 0.2889 | lr 8.1e-04 | norm 3.8589 | dt 0.146
type train | step 2840 | loss 0.0963 0.0279 0.9702 0.6108 4.7134 2.1258 19.1932 3.1260 0.0139 0.0629 0.0070 0.2962 | lr 8.0e-04 | norm 4.0280 | dt 0.152
type train | step 2850 | loss 0.0944 0.0291 0.9605 0.6162 4.7971 2.1712 19.5020 3.1848 0.0139 0.0515 0.0078 0.2935 | lr 8.0e-04 | norm 3.9524 | dt 0.146
type train | step 2860 | loss 0.0948 0.0305 0.9789 0.5900 4.7265 2.2156 19.3405 3.1709 0.0153 0.0523 0.0077 0.2922 | lr 8.0e-04 | norm 4.4107 | dt 0.166
type train | step 2870 | loss 0.0957 0.0290 0.9608 0.5931 4.8337 2.1826 19.8254 3.2333 0.0129 0.0558 0.0067 0.2968 | lr 8.0e-04 | norm 3.8939 | dt 0.142
type train | step 2880 | loss 0.0926 0.0290 0.9415 0.6054 4.6191 2.1919 19.2453 3.1294 0.0132 0.0529 0.0081 0.2876 | lr 8.0e-04 | norm 4.4593 | dt 0.168
type train | step 2890 | loss 0.0968 0.0295 0.9700 0.5866 4.8304 2.2103 19.7609 3.2628 0.0132 0.0509 0.0101 0.2865 | lr 7.9e-04 | norm 3.8996 | dt 0.144
type train | step 2900 | loss 0.0928 0.0300 0.9372 0.6322 4.5469 2.1664 18.8221 3.1251 0.0136 0.0527 0.0133 0.2709 | lr 7.9e-04 | norm 3.8360 | dt 0.159
type train | step 2910 | loss 0.0952 0.0302 0.9647 0.6476 4.7069 2.1853 19.5786 3.1499 0.0131 0.0624 0.0157 0.2738 | lr 7.9e-04 | norm 4.9672 | dt 0.152
type train | step 2920 | loss 0.0945 0.0322 0.9631 0.6483 4.7836 2.1458 19.7540 3.2273 0.0145 0.0571 0.0221 0.2941 | lr 7.9e-04 | norm 5.3068 | dt 0.162
type train | step 2930 | loss 0.0953 0.0303 0.9689 0.6309 4.8625 2.1781 19.8373 3.0979 0.0162 0.0479 0.0227 0.2729 | lr 7.9e-04 | norm 3.6928 | dt 0.160
type train | step 2940 | loss 0.0953 0.0310 0.9583 0.5967 4.7092 2.1856 19.6124 3.3467 0.0160 0.0528 0.0235 0.2840 | lr 7.9e-04 | norm 5.1276 | dt 0.143
type train | step 2950 | loss 0.0944 0.0301 0.9581 0.6165 4.9003 2.1908 20.2049 3.2227 0.0151 0.0498 0.0249 0.2540 | lr 7.8e-04 | norm 4.1099 | dt 0.161
type train | step 2960 | loss 0.0933 0.0292 0.9611 0.6559 4.8680 2.1699 20.0486 3.1237 0.0154 0.0583 0.0325 0.2884 | lr 7.8e-04 | norm 4.4623 | dt 0.157
type train | step 2970 | loss 0.0945 0.0316 0.9673 0.6369 4.8373 2.1508 20.2354 3.3198 0.0142 0.0560 0.0363 0.2509 | lr 7.8e-04 | norm 3.4690 | dt 0.149
type train | step 2980 | loss 0.0913 0.0288 0.9368 0.6369 4.6553 2.1049 19.1373 3.2195 0.0131 0.0552 0.0344 0.2697 | lr 7.8e-04 | norm 3.4601 | dt 0.145
type train | step 2990 | loss 0.0917 0.0284 0.9511 0.6279 4.7416 2.1330 19.5788 3.1826 0.0125 0.0556 0.0293 0.2706 | lr 7.8e-04 | norm 3.9988 | dt 0.161
type train | step 3000 | loss 0.0963 0.0318 0.9615 0.6095 4.7347 2.0917 19.6946 3.3481 0.0131 0.0580 0.0204 0.2887 | lr 7.8e-04 | norm 5.3264 | dt 0.159
type train | step 3010 | loss 0.0938 0.0294 0.9579 0.6316 4.9343 2.0761 20.5021 3.1950 0.0123 0.0535 0.0135 0.2877 | lr 7.7e-04 | norm 4.6291 | dt 0.170
type train | step 3020 | loss 0.0941 0.0282 0.9517 0.6014 4.7242 2.1727 19.1861 3.1621 0.0122 0.0537 0.0096 0.2941 | lr 7.7e-04 | norm 4.1541 | dt 0.155
type train | step 3030 | loss 0.0931 0.0318 0.9624 0.6231 4.7481 2.1248 20.0101 3.1858 0.0126 0.0489 0.0104 0.2780 | lr 7.7e-04 | norm 4.8543 | dt 0.164
type train | step 3040 | loss 0.0898 0.0284 0.9255 0.5913 4.6859 2.1844 19.0608 2.9755 0.0128 0.0592 0.0103 0.2772 | lr 7.7e-04 | norm 4.3677 | dt 0.164
type train | step 3050 | loss 0.0923 0.0295 0.9466 0.6353 4.9216 2.1063 20.0790 3.1798 0.0120 0.0596 0.0154 0.2801 | lr 7.7e-04 | norm 5.5407 | dt 0.168
type train | step 3060 | loss 0.0918 0.0292 0.9464 0.6137 4.7554 2.1603 19.7611 3.1376 0.0129 0.0583 0.0113 0.2655 | lr 7.6e-04 | norm 4.1454 | dt 0.149
type train | step 3070 | loss 0.0922 0.0295 0.9538 0.6080 4.8323 2.1224 20.0389 3.0907 0.0121 0.0579 0.0102 0.2888 | lr 7.6e-04 | norm 4.5095 | dt 0.160
type train | step 3080 | loss 0.0921 0.0293 0.9227 0.6412 4.6391 2.1741 19.1783 3.0987 0.0116 0.0471 0.0114 0.2949 | lr 7.6e-04 | norm 4.0189 | dt 0.162
type train | step 3090 | loss 0.0951 0.0329 0.9772 0.6303 4.8107 2.2115 20.0457 3.2420 0.0125 0.0460 0.0122 0.2793 | lr 7.6e-04 | norm 5.1346 | dt 0.149
type train | step 3100 | loss 0.0911 0.0289 0.9344 0.5928 4.6795 2.1451 18.9134 3.0828 0.0122 0.0468 0.0125 0.2914 | lr 7.6e-04 | norm 3.5603 | dt 0.157
type train | step 3110 | loss 0.0902 0.0277 0.9264 0.5982 4.7174 2.1266 19.5597 3.0523 0.0123 0.0477 0.0149 0.2891 | lr 7.5e-04 | norm 3.8313 | dt 0.158
type train | step 3120 | loss 0.0911 0.0284 0.9299 0.6007 4.6543 2.1195 19.0685 3.1443 0.0120 0.0600 0.0154 0.2878 | lr 7.5e-04 | norm 3.4515 | dt 0.161
type train | step 3130 | loss 0.0949 0.0311 0.9731 0.6059 4.9836 2.1126 20.6366 3.1318 0.0117 0.0609 0.0182 0.2850 | lr 7.5e-04 | norm 3.7804 | dt 0.161
type train | step 3140 | loss 0.0911 0.0297 0.9412 0.6554 4.6852 2.1569 19.4005 3.0936 0.0114 0.0498 0.0199 0.2841 | lr 7.5e-04 | norm 3.6711 | dt 0.155
type train | step 3150 | loss 0.0892 0.0303 0.9034 0.6528 4.5571 2.1307 19.0736 3.1229 0.0117 0.0469 0.0290 0.2642 | lr 7.5e-04 | norm 4.2732 | dt 0.152
type train | step 3160 | loss 0.0925 0.0300 0.9494 0.5895 4.7652 2.1591 19.2762 3.1222 0.0130 0.0525 0.0383 0.2718 | lr 7.5e-04 | norm 3.7668 | dt 0.154
type train | step 3170 | loss 0.0943 0.0314 0.9526 0.6457 4.6050 2.1858 19.0706 3.2446 0.0126 0.0524 0.0316 0.2825 | lr 7.4e-04 | norm 4.9355 | dt 0.146
type train | step 3180 | loss 0.0913 0.0290 0.9141 0.6449 4.5914 2.1559 19.2151 3.1583 0.0130 0.0597 0.0246 0.2757 | lr 7.4e-04 | norm 3.5648 | dt 0.165
type train | step 3190 | loss 0.0993 0.0328 0.9614 0.6024 4.7188 2.1093 19.4133 3.3033 0.0123 0.0615 0.0160 0.2700 | lr 7.4e-04 | norm 5.5719 | dt 0.166
type train | step 3200 | loss 0.0932 0.0312 0.9567 0.5982 4.7903 2.1981 19.7013 3.1919 0.0124 0.0574 0.0101 0.2816 | lr 7.4e-04 | norm 4.2067 | dt 0.162
type train | step 3210 | loss 0.0902 0.0305 0.9220 0.6199 4.6836 2.1552 19.6845 3.0824 0.0127 0.0499 0.0141 0.2758 | lr 7.4e-04 | norm 3.9334 | dt 0.167
type train | step 3220 | loss 0.0895 0.0281 0.9407 0.6050 4.7230 2.0886 19.1711 3.0479 0.0164 0.0609 0.0198 0.2758 | lr 7.3e-04 | norm 4.0996 | dt 0.149
type train | step 3230 | loss 0.0918 0.0318 0.9374 0.6265 4.6551 2.1591 19.4376 3.1712 0.0143 0.0542 0.0202 0.2746 | lr 7.3e-04 | norm 3.6369 | dt 0.158
type train | step 3240 | loss 0.0906 0.0296 0.9204 0.5949 4.6312 2.1142 19.2616 3.1181 0.0139 0.0519 0.0156 0.2705 | lr 7.3e-04 | norm 4.1281 | dt 0.156
type train | step 3250 | loss 0.0965 0.0318 0.9689 0.6157 4.8158 2.1899 20.0430 3.2119 0.0151 0.0508 0.0156 0.2857 | lr 7.3e-04 | norm 3.7310 | dt 0.148
type train | step 3260 | loss 0.0934 0.0294 0.9451 0.5737 4.6633 2.1788 19.3587 3.1525 0.0148 0.0542 0.0145 0.2638 | lr 7.3e-04 | norm 4.0155 | dt 0.144
type train | step 3270 | loss 0.0904 0.0295 0.9303 0.6139 4.6995 2.1249 19.4521 3.1181 0.0134 0.0537 0.0172 0.2697 | lr 7.2e-04 | norm 4.5706 | dt 0.155
type train | step 3280 | loss 0.0886 0.0275 0.9223 0.5866 4.6939 2.1077 19.2889 3.0534 0.0141 0.0569 0.0207 0.2592 | lr 7.2e-04 | norm 3.7551 | dt 0.150
type train | step 3290 | loss 0.0939 0.0311 0.9709 0.6492 4.7828 2.1592 19.7196 3.2499 0.0135 0.0542 0.0215 0.2669 | lr 7.2e-04 | norm 3.8592 | dt 0.150
type train | step 3300 | loss 0.0912 0.0289 0.9133 0.6205 4.5652 2.1445 18.9466 3.1293 0.0135 0.0557 0.0238 0.2820 | lr 7.2e-04 | norm 3.6094 | dt 0.147
type train | step 3310 | loss 0.0897 0.0300 0.9146 0.6407 4.6068 2.1668 19.3271 3.1655 0.0136 0.0604 0.0290 0.2557 | lr 7.2e-04 | norm 3.8140 | dt 0.153
type train | step 3320 | loss 0.0932 0.0314 0.9326 0.6252 4.5702 2.0933 19.0114 3.1850 0.0158 0.0526 0.0289 0.2545 | lr 7.1e-04 | norm 4.0936 | dt 0.151
type train | step 3330 | loss 0.0937 0.0323 0.9422 0.6383 4.7661 2.1748 19.6168 3.2265 0.0146 0.0524 0.0283 0.2723 | lr 7.1e-04 | norm 4.2259 | dt 0.164
type train | step 3340 | loss 0.0903 0.0305 0.9163 0.6217 4.5719 2.1735 19.0297 3.1790 0.0166 0.0551 0.0309 0.2802 | lr 7.1e-04 | norm 5.3507 | dt 0.160
type train | step 3350 | loss 0.0925 0.0329 0.9583 0.6328 4.7121 2.1459 19.2858 3.2185 0.0141 0.0585 0.0315 0.2575 | lr 7.1e-04 | norm 4.3353 | dt 0.146
type train | step 3360 | loss 0.0922 0.0319 0.9209 0.6230 4.6691 2.1463 19.4256 3.1722 0.0151 0.0504 0.0321 0.2670 | lr 7.1e-04 | norm 4.3440 | dt 0.163
type train | step 3370 | loss 0.0916 0.0320 0.9121 0.6243 4.6523 2.1211 19.3299 3.2162 0.0138 0.0513 0.0352 0.2649 | lr 7.0e-04 | norm 5.0998 | dt 0.160
type train | step 3380 | loss 0.0913 0.0301 0.9360 0.6296 4.7927 2.1325 19.8343 3.1365 0.0168 0.0462 0.0356 0.2541 | lr 7.0e-04 | norm 3.7315 | dt 0.147
type train | step 3390 | loss 0.0891 0.0300 0.9290 0.6442 4.7435 2.0661 19.6694 3.1318 0.0180 0.0561 0.0340 0.2455 | lr 7.0e-04 | norm 3.9115 | dt 0.163
type train | step 3400 | loss 0.0903 0.0311 0.9259 0.6489 4.7511 2.1843 19.5382 3.0539 0.0177 0.0483 0.0282 0.2627 | lr 7.0e-04 | norm 4.2024 | dt 0.165
type train | step 3410 | loss 0.0894 0.0299 0.8926 0.6341 4.5001 2.0978 18.7265 3.0157 0.0205 0.0462 0.0352 0.2642 | lr 7.0e-04 | norm 4.0480 | dt 0.165
type train | step 3420 | loss 0.0899 0.0301 0.9044 0.6217 4.5413 2.2428 19.2855 3.2077 0.0174 0.0474 0.0295 0.2571 | lr 7.0e-04 | norm 4.8072 | dt 0.158
type train | step 3430 | loss 0.0930 0.0322 0.9406 0.6800 4.6643 2.2340 19.6016 3.3151 0.0209 0.0469 0.0292 0.2573 | lr 6.9e-04 | norm 5.7447 | dt 0.151
type train | step 3440 | loss 0.0887 0.0296 0.9216 0.6808 4.7526 2.0780 19.7893 3.0676 0.0206 0.0491 0.0229 0.2496 | lr 6.9e-04 | norm 3.7389 | dt 0.148
type train | step 3450 | loss 0.0921 0.0293 0.9401 0.6585 4.6302 2.0921 19.0273 3.0148 0.0199 0.0415 0.0216 0.2519 | lr 6.9e-04 | norm 3.8843 | dt 0.144
type train | step 3460 | loss 0.0901 0.0296 0.9235 0.6583 4.7030 2.1774 19.3155 3.1015 0.0181 0.0459 0.0231 0.2535 | lr 6.9e-04 | norm 4.0427 | dt 0.149
type train | step 3470 | loss 0.0904 0.0294 0.9447 0.6438 4.6286 2.1454 19.1067 3.0729 0.0172 0.0422 0.0247 0.2597 | lr 6.9e-04 | norm 3.5590 | dt 0.166
type train | step 3480 | loss 0.0913 0.0285 0.9287 0.6479 4.7502 2.1398 19.6029 3.1168 0.0167 0.0426 0.0249 0.2653 | lr 6.8e-04 | norm 3.3605 | dt 0.166
type train | step 3490 | loss 0.0886 0.0292 0.9089 0.6434 4.5397 2.1670 19.1059 3.2032 0.0161 0.0441 0.0290 0.2506 | lr 6.8e-04 | norm 5.0191 | dt 0.166
type train | step 3500 | loss 0.0927 0.0288 0.9366 0.6610 4.7218 2.1561 19.6230 3.2035 0.0167 0.0396 0.0321 0.2465 | lr 6.8e-04 | norm 3.4736 | dt 0.159
type train | step 3510 | loss 0.0884 0.0287 0.9066 0.6143 4.4695 2.1149 18.6637 3.1783 0.0165 0.0415 0.0300 0.2456 | lr 6.8e-04 | norm 4.2443 | dt 0.166
type train | step 3520 | loss 0.0912 0.0301 0.9367 0.6336 4.6217 2.0826 19.4565 3.1418 0.0173 0.0405 0.0320 0.2459 | lr 6.8e-04 | norm 4.1091 | dt 0.160
type train | step 3530 | loss 0.0904 0.0296 0.9293 0.6687 4.6857 2.1196 19.5592 3.1595 0.0184 0.0407 0.0360 0.2492 | lr 6.7e-04 | norm 3.7096 | dt 0.159
type train | step 3540 | loss 0.0911 0.0272 0.9339 0.6295 4.7651 2.1206 19.6756 3.1716 0.0163 0.0398 0.0326 0.2456 | lr 6.7e-04 | norm 3.7118 | dt 0.167
type train | step 3550 | loss 0.0922 0.0290 0.9257 0.6488 4.6251 2.1257 19.5348 3.2501 0.0166 0.0396 0.0335 0.2518 | lr 6.7e-04 | norm 4.0840 | dt 0.155
type train | step 3560 | loss 0.0906 0.0294 0.9331 0.6331 4.8182 2.1175 20.0666 3.2118 0.0146 0.0378 0.0342 0.2343 | lr 6.7e-04 | norm 3.9293 | dt 0.154
type train | step 3570 | loss 0.0900 0.0280 0.9327 0.6697 4.7777 2.1085 19.9003 3.2349 0.0155 0.0377 0.0499 0.2383 | lr 6.6e-04 | norm 4.1349 | dt 0.146
type train | step 3580 | loss 0.0903 0.0306 0.9362 0.6691 4.7533 2.1207 20.1168 3.2347 0.0146 0.0383 0.0568 0.2333 | lr 6.6e-04 | norm 4.1814 | dt 0.167
type train | step 3590 | loss 0.0877 0.0271 0.9075 0.6767 4.5653 2.1041 19.0277 3.1907 0.0148 0.0388 0.0604 0.2428 | lr 6.6e-04 | norm 4.1102 | dt 0.167
type train | step 3600 | loss 0.0884 0.0263 0.9259 0.7028 4.6515 2.1329 19.4538 3.1731 0.0155 0.0435 0.0600 0.2290 | lr 6.6e-04 | norm 4.3364 | dt 0.164
type train | step 3610 | loss 0.0926 0.0299 0.9262 0.7146 4.6386 2.1575 19.5358 3.3027 0.0151 0.0449 0.0608 0.2403 | lr 6.6e-04 | norm 5.8790 | dt 0.161
type train | step 3620 | loss 0.0901 0.0287 0.9280 0.6900 4.8356 2.1308 20.3937 3.2233 0.0176 0.0461 0.0627 0.2363 | lr 6.5e-04 | norm 3.6458 | dt 0.146
type train | step 3630 | loss 0.0912 0.0285 0.9274 0.6976 4.6475 2.1122 19.0312 3.1290 0.0173 0.0444 0.0651 0.2417 | lr 6.5e-04 | norm 4.3481 | dt 0.160
type train | step 3640 | loss 0.0901 0.0311 0.9349 0.7274 4.6606 2.1016 19.8718 3.2353 0.0163 0.0463 0.0528 0.2223 | lr 6.5e-04 | norm 5.7941 | dt 0.151
type train | step 3650 | loss 0.0869 0.0283 0.8975 0.6961 4.6103 2.1237 18.9626 2.9788 0.0162 0.0451 0.0464 0.2259 | lr 6.5e-04 | norm 3.3515 | dt 0.164
type train | step 3660 | loss 0.0891 0.0288 0.9189 0.6764 4.8311 2.1205 19.9729 3.1163 0.0157 0.0472 0.0383 0.2311 | lr 6.5e-04 | norm 3.8520 | dt 0.150
type train | step 3670 | loss 0.0888 0.0273 0.9207 0.7242 4.6662 2.0903 19.6005 3.2008 0.0166 0.0419 0.0402 0.2214 | lr 6.4e-04 | norm 3.6227 | dt 0.169
type train | step 3680 | loss 0.0890 0.0292 0.9256 0.6917 4.7511 2.0484 19.8956 3.2110 0.0154 0.0437 0.0398 0.2378 | lr 6.4e-04 | norm 3.9350 | dt 0.149
type train | step 3690 | loss 0.0889 0.0295 0.8976 0.6683 4.5723 2.1389 19.0405 3.1133 0.0148 0.0403 0.0395 0.2452 | lr 6.4e-04 | norm 3.8336 | dt 0.146
type train | step 3700 | loss 0.0922 0.0322 0.9462 0.7126 4.7236 2.1305 19.9254 3.2469 0.0160 0.0467 0.0449 0.2270 | lr 6.4e-04 | norm 4.0542 | dt 0.167
type train | step 3710 | loss 0.0879 0.0274 0.9065 0.7120 4.6000 2.0734 18.8410 3.1708 0.0144 0.0481 0.0475 0.2212 | lr 6.4e-04 | norm 6.0667 | dt 0.149
type train | step 3720 | loss 0.0871 0.0278 0.8980 0.7401 4.6485 2.1523 19.4564 3.1722 0.0136 0.0540 0.0483 0.2341 | lr 6.3e-04 | norm 4.0234 | dt 0.148
type train | step 3730 | loss 0.0883 0.0293 0.9039 0.7039 4.5897 2.1125 18.9893 3.2179 0.0125 0.0545 0.0594 0.2252 | lr 6.3e-04 | norm 4.1268 | dt 0.149
type train | step 3740 | loss 0.0918 0.0305 0.9469 0.6884 4.9059 2.1104 20.5213 3.2062 0.0141 0.0568 0.0706 0.2248 | lr 6.3e-04 | norm 3.9416 | dt 0.153
type train | step 3750 | loss 0.0882 0.0275 0.9148 0.6153 4.6086 2.1262 19.2764 3.2226 0.0152 0.0598 0.0687 0.2409 | lr 6.3e-04 | norm 4.0764 | dt 0.145
type train | step 3760 | loss 0.0864 0.0273 0.8782 0.6510 4.4799 2.0418 18.9546 3.0825 0.0138 0.0526 0.0689 0.2322 | lr 6.3e-04 | norm 4.2578 | dt 0.160
type train | step 3770 | loss 0.0894 0.0268 0.9205 0.7065 4.6915 2.0599 19.1986 3.2047 0.0138 0.0587 0.0688 0.2158 | lr 6.2e-04 | norm 4.5260 | dt 0.157
type train | step 3780 | loss 0.0915 0.0294 0.9250 0.7154 4.5188 2.1014 18.9241 3.2772 0.0137 0.0597 0.0715 0.2265 | lr 6.2e-04 | norm 3.3357 | dt 0.162
type train | step 3790 | loss 0.0884 0.0267 0.8956 0.6950 4.5299 2.1101 19.1546 3.3080 0.0135 0.0617 0.0717 0.2263 | lr 6.2e-04 | norm 5.6792 | dt 0.164
type train | step 3800 | loss 0.0963 0.0311 0.9265 0.6862 4.6353 2.1414 19.3526 3.3474 0.0137 0.0582 0.0733 0.2317 | lr 6.2e-04 | norm 4.7366 | dt 0.162
type train | step 3810 | loss 0.0909 0.0302 0.9319 0.6671 4.7041 2.0767 19.6148 3.2931 0.0148 0.0593 0.0771 0.2420 | lr 6.2e-04 | norm 4.3181 | dt 0.157
type train | step 3820 | loss 0.0876 0.0275 0.8979 0.6681 4.6112 2.0020 19.5603 3.1009 0.0131 0.0554 0.0840 0.2393 | lr 6.1e-04 | norm 4.0685 | dt 0.147
type train | step 3830 | loss 0.0868 0.0261 0.9173 0.6596 4.6444 2.0279 19.0443 3.0540 0.0136 0.0524 0.0731 0.2406 | lr 6.1e-04 | norm 3.9033 | dt 0.157
type train | step 3840 | loss 0.0888 0.0291 0.9099 0.6935 4.5739 2.1052 19.2657 3.2076 0.0133 0.0498 0.0626 0.2352 | lr 6.1e-04 | norm 3.7854 | dt 0.149
type train | step 3850 | loss 0.0875 0.0294 0.8993 0.6656 4.5713 2.0963 19.2075 3.2648 0.0135 0.0516 0.0667 0.2326 | lr 6.1e-04 | norm 3.5919 | dt 0.164
type train | step 3860 | loss 0.0935 0.0317 0.9445 0.6783 4.7317 2.0722 19.9917 3.1965 0.0136 0.0549 0.0696 0.2262 | lr 6.1e-04 | norm 3.6878 | dt 0.149
type train | step 3870 | loss 0.0900 0.0326 0.9247 0.6722 4.5795 2.0395 19.2189 3.1358 0.0139 0.0535 0.0721 0.2246 | lr 6.0e-04 | norm 3.4456 | dt 0.149
type train | step 3880 | loss 0.0872 0.0301 0.9053 0.6628 4.6290 2.0310 19.3616 3.0135 0.0135 0.0598 0.0685 0.2307 | lr 6.0e-04 | norm 3.8875 | dt 0.148
type train | step 3890 | loss 0.0857 0.0290 0.9020 0.6534 4.6311 2.0622 19.2276 3.0412 0.0137 0.0531 0.0596 0.2447 | lr 6.0e-04 | norm 4.2601 | dt 0.166
type train | step 3900 | loss 0.0906 0.0323 0.9417 0.6924 4.7016 2.0927 19.5689 3.1718 0.0139 0.0604 0.0566 0.2262 | lr 6.0e-04 | norm 4.4331 | dt 0.162
type train | step 3910 | loss 0.0880 0.0303 0.8912 0.6791 4.5106 2.0907 18.8500 3.0583 0.0133 0.0566 0.0610 0.2255 | lr 5.9e-04 | norm 3.8424 | dt 0.151
type train | step 3920 | loss 0.0868 0.0310 0.8924 0.6461 4.5317 2.0520 19.2561 3.1731 0.0134 0.0566 0.0591 0.2170 | lr 5.9e-04 | norm 3.5348 | dt 0.145
type train | step 3930 | loss 0.0900 0.0314 0.9135 0.6373 4.5027 2.0555 18.9451 3.0791 0.0129 0.0543 0.0560 0.2338 | lr 5.9e-04 | norm 5.2293 | dt 0.161
type train | step 3940 | loss 0.0911 0.0314 0.9173 0.6509 4.6847 2.0885 19.5141 3.2103 0.0127 0.0530 0.0585 0.2230 | lr 5.9e-04 | norm 3.8877 | dt 0.152
type train | step 3950 | loss 0.0872 0.0300 0.8946 0.6403 4.5007 2.1063 18.9365 3.1133 0.0127 0.0530 0.0656 0.2266 | lr 5.9e-04 | norm 3.9791 | dt 0.166
type train | step 3960 | loss 0.0895 0.0306 0.9348 0.6588 4.6363 2.1182 19.1689 3.1615 0.0124 0.0528 0.0792 0.2268 | lr 5.8e-04 | norm 5.1202 | dt 0.159
type train | step 3970 | loss 0.0892 0.0311 0.9012 0.6737 4.6097 2.1179 19.3181 3.2489 0.0120 0.0534 0.0747 0.2277 | lr 5.8e-04 | norm 6.2777 | dt 0.146
type train | step 3980 | loss 0.0890 0.0310 0.8928 0.6458 4.5829 2.0869 19.2218 3.2359 0.0115 0.0534 0.0619 0.2204 | lr 5.8e-04 | norm 4.0161 | dt 0.160
type train | step 3990 | loss 0.0885 0.0293 0.9126 0.6425 4.7299 2.0477 19.7610 3.1999 0.0115 0.0539 0.0462 0.2176 | lr 5.8e-04 | norm 4.8213 | dt 0.163
type train | step 4000 | loss 0.0864 0.0294 0.9101 0.6864 4.6912 2.0823 19.5666 3.1736 0.0114 0.0536 0.0381 0.2173 | lr 5.8e-04 | norm 3.5549 | dt 0.143
type train | step 4010 | loss 0.0874 0.0287 0.9037 0.6802 4.6741 2.0561 19.4716 3.2118 0.0120 0.0661 0.0442 0.2223 | lr 5.7e-04 | norm 5.2906 | dt 0.166
type train | step 4020 | loss 0.0868 0.0280 0.8677 0.6906 4.4260 2.0610 18.6324 3.1273 0.0115 0.0680 0.0408 0.2183 | lr 5.7e-04 | norm 3.9182 | dt 0.165
type train | step 4030 | loss 0.0871 0.0295 0.8877 0.6633 4.4881 2.1727 19.2248 3.2254 0.0112 0.0747 0.0366 0.2288 | lr 5.7e-04 | norm 6.3475 | dt 0.164
type train | step 4040 | loss 0.0902 0.0308 0.9201 0.6907 4.5904 2.2052 19.5746 3.2460 0.0115 0.0738 0.0425 0.2247 | lr 5.7e-04 | norm 4.0788 | dt 0.146
type train | step 4050 | loss 0.0861 0.0283 0.9013 0.6485 4.6980 2.0275 19.7484 3.1673 0.0108 0.0718 0.0467 0.2081 | lr 5.7e-04 | norm 3.7711 | dt 0.160
type train | step 4060 | loss 0.0895 0.0278 0.9189 0.6599 4.5726 2.0886 18.9856 3.1261 0.0116 0.0698 0.0502 0.2185 | lr 5.6e-04 | norm 3.9164 | dt 0.147
type train | step 4070 | loss 0.0879 0.0296 0.9030 0.6661 4.6266 2.1223 19.2632 3.1320 0.0113 0.0731 0.0534 0.2037 | lr 5.6e-04 | norm 4.1200 | dt 0.168
type train | step 4080 | loss 0.0874 0.0299 0.9262 0.6706 4.5614 2.1572 19.0193 3.1461 0.0116 0.0856 0.0593 0.2197 | lr 5.6e-04 | norm 3.9946 | dt 0.148
type train | step 4090 | loss 0.0884 0.0294 0.9090 0.6755 4.6847 2.1201 19.5312 3.2282 0.0112 0.0826 0.0728 0.2138 | lr 5.6e-04 | norm 4.3356 | dt 0.164
type train | step 4100 | loss 0.0862 0.0295 0.8929 0.6923 4.4845 2.1437 19.0780 3.3481 0.0113 0.0850 0.0842 0.2043 | lr 5.6e-04 | norm 4.4186 | dt 0.160
type train | step 4110 | loss 0.0895 0.0293 0.9168 0.6855 4.6607 2.1263 19.5640 3.1509 0.0113 0.0816 0.0817 0.2185 | lr 5.5e-04 | norm 3.4689 | dt 0.160
type train | step 4120 | loss 0.0860 0.0287 0.8914 0.6417 4.4110 2.0756 18.5857 3.1502 0.0112 0.0881 0.0814 0.2115 | lr 5.5e-04 | norm 3.8394 | dt 0.154
type train | step 4130 | loss 0.0890 0.0322 0.9192 0.6520 4.5525 2.0295 19.3884 3.1633 0.0111 0.0810 0.0757 0.2168 | lr 5.5e-04 | norm 3.9000 | dt 0.162
type train | step 4140 | loss 0.0879 0.0298 0.9114 0.6447 4.6266 2.0911 19.4907 3.2320 0.0116 0.0840 0.0648 0.2132 | lr 5.5e-04 | norm 4.1311 | dt 0.165
type train | step 4150 | loss 0.0884 0.0295 0.9180 0.6393 4.7077 2.1192 19.6514 3.1890 0.0107 0.0797 0.0599 0.2182 | lr 5.4e-04 | norm 5.0042 | dt 0.165
type train | step 4160 | loss 0.0896 0.0301 0.9097 0.6352 4.5611 2.1208 19.4890 3.2190 0.0109 0.0767 0.0655 0.2270 | lr 5.4e-04 | norm 3.7658 | dt 0.167
type train | step 4170 | loss 0.0882 0.0296 0.9154 0.6339 4.7586 2.0277 20.0078 3.2276 0.0104 0.0732 0.0651 0.2152 | lr 5.4e-04 | norm 5.0990 | dt 0.149
type train | step 4180 | loss 0.0879 0.0295 0.9148 0.6374 4.7276 2.0093 19.8513 3.1207 0.0108 0.0740 0.0628 0.2204 | lr 5.4e-04 | norm 3.7541 | dt 0.148
type train | step 4190 | loss 0.0880 0.0328 0.9198 0.6333 4.6896 2.0802 20.0261 3.1690 0.0109 0.0740 0.0606 0.2246 | lr 5.4e-04 | norm 3.5448 | dt 0.160
type train | step 4200 | loss 0.0849 0.0297 0.8916 0.6192 4.5149 2.0398 18.9632 3.1616 0.0104 0.0748 0.0645 0.2269 | lr 5.3e-04 | norm 4.8539 | dt 0.145
type train | step 4210 | loss 0.0859 0.0296 0.9100 0.6395 4.5918 2.1318 19.4538 3.1995 0.0105 0.0677 0.0619 0.2183 | lr 5.3e-04 | norm 4.7586 | dt 0.169
type train | step 4220 | loss 0.0901 0.0315 0.9130 0.6279 4.5737 2.0493 19.4827 3.2521 0.0108 0.0750 0.0683 0.2228 | lr 5.3e-04 | norm 4.4945 | dt 0.162
type train | step 4230 | loss 0.0873 0.0297 0.9142 0.6193 4.7733 1.9919 20.3481 3.1351 0.0101 0.0658 0.0789 0.2291 | lr 5.3e-04 | norm 3.9378 | dt 0.159
type train | step 4240 | loss 0.0889 0.0286 0.9132 0.5917 4.5838 2.0417 18.9892 3.0174 0.0100 0.0688 0.0766 0.2334 | lr 5.3e-04 | norm 4.7204 | dt 0.153
type train | step 4250 | loss 0.0882 0.0309 0.9197 0.6427 4.6037 2.0318 19.8131 3.1420 0.0104 0.0723 0.0723 0.2311 | lr 5.2e-04 | norm 4.1488 | dt 0.163
type train | step 4260 | loss 0.0857 0.0277 0.8858 0.6180 4.5613 2.0424 18.9590 3.0759 0.0101 0.0726 0.0681 0.2225 | lr 5.2e-04 | norm 3.8720 | dt 0.153
type train | step 4270 | loss 0.0874 0.0287 0.9093 0.6121 4.7755 2.0763 19.9242 3.0859 0.0108 0.0826 0.0631 0.2121 | lr 5.2e-04 | norm 3.9088 | dt 0.161
type train | step 4280 | loss 0.0879 0.0292 0.9080 0.6235 4.6185 2.0390 19.5462 3.0022 0.0101 0.0782 0.0605 0.2219 | lr 5.2e-04 | norm 3.5251 | dt 0.164
type train | step 4290 | loss 0.0876 0.0283 0.9100 0.5948 4.6891 2.0573 19.8801 3.1034 0.0102 0.0841 0.0589 0.2143 | lr 5.2e-04 | norm 6.7100 | dt 0.165
type train | step 4300 | loss 0.0882 0.0281 0.8861 0.5840 4.5216 2.0529 18.9959 3.1345 0.0100 0.0766 0.0591 0.2132 | lr 5.1e-04 | norm 6.0410 | dt 0.154
type train | step 4310 | loss 0.0906 0.0319 0.9332 0.5955 4.6594 2.1188 19.8497 3.1438 0.0100 0.0821 0.0618 0.2250 | lr 5.1e-04 | norm 4.9157 | dt 0.143
type train | step 4320 | loss 0.0869 0.0286 0.8979 0.5901 4.5435 2.0235 18.8563 3.0243 0.0092 0.0727 0.0632 0.2239 | lr 5.1e-04 | norm 4.0429 | dt 0.165
type train | step 4330 | loss 0.0860 0.0272 0.8870 0.5720 4.5927 2.0515 19.4072 3.0764 0.0099 0.0751 0.0672 0.2136 | lr 5.1e-04 | norm 3.4898 | dt 0.162
type train | step 4340 | loss 0.0869 0.0281 0.8939 0.5902 4.5440 2.0623 18.9869 2.9411 0.0098 0.0712 0.0697 0.2200 | lr 5.1e-04 | norm 3.7487 | dt 0.165
type train | step 4350 | loss 0.0905 0.0290 0.9364 0.6028 4.8557 2.0351 20.5012 3.1829 0.0105 0.0718 0.0721 0.2129 | lr 5.0e-04 | norm 3.8659 | dt 0.146
type train | step 4360 | loss 0.0866 0.0287 0.9057 0.5708 4.5695 2.1042 19.2789 3.1714 0.0098 0.0723 0.0793 0.2129 | lr 5.0e-04 | norm 5.8055 | dt 0.159
type train | step 4370 | loss 0.0847 0.0287 0.8680 0.5450 4.4333 2.0266 18.9284 3.0322 0.0092 0.0749 0.0815 0.2200 | lr 5.0e-04 | norm 3.8408 | dt 0.149
type train | step 4380 | loss 0.0878 0.0279 0.9131 0.5732 4.6450 2.0659 19.1981 3.0194 0.0094 0.0700 0.0860 0.2154 | lr 5.0e-04 | norm 3.7414 | dt 0.148
type train | step 4390 | loss 0.0893 0.0294 0.9168 0.5808 4.4743 2.1080 18.8714 3.0917 0.0099 0.0674 0.0884 0.2227 | lr 4.9e-04 | norm 3.9457 | dt 0.147
type train | step 4400 | loss 0.0867 0.0293 0.8884 0.5993 4.4860 2.0662 19.1956 3.1222 0.0099 0.0694 0.0862 0.2196 | lr 4.9e-04 | norm 3.6627 | dt 0.166
type train | step 4410 | loss 0.0937 0.0319 0.9089 0.6192 4.5709 2.0462 19.3553 3.1693 0.0106 0.0766 0.0694 0.2132 | lr 4.9e-04 | norm 4.1024 | dt 0.163
type train | step 4420 | loss 0.0879 0.0307 0.9204 0.6484 4.6488 2.0198 19.5943 3.2240 0.0097 0.0732 0.0627 0.2171 | lr 4.9e-04 | norm 4.3746 | dt 0.145
type train | step 4430 | loss 0.0855 0.0286 0.8901 0.6210 4.5628 2.0252 19.5445 3.1414 0.0093 0.0857 0.0674 0.2211 | lr 4.9e-04 | norm 3.7377 | dt 0.161
type train | step 4440 | loss 0.0846 0.0262 0.9114 0.5661 4.5996 2.0050 19.0699 3.0424 0.0092 0.0792 0.0749 0.2176 | lr 4.8e-04 | norm 3.9705 | dt 0.164
type train | step 4450 | loss 0.0866 0.0288 0.9036 0.6115 4.5279 2.0045 19.2588 3.2082 0.0099 0.0880 0.0815 0.2148 | lr 4.8e-04 | norm 3.7119 | dt 0.159
type train | step 4460 | loss 0.0856 0.0282 0.8964 0.6337 4.5357 2.0124 19.2369 3.2502 0.0099 0.1029 0.0777 0.2123 | lr 4.8e-04 | norm 4.8907 | dt 0.165
type train | step 4470 | loss 0.0910 0.0296 0.9348 0.5952 4.6808 2.0314 19.9806 3.2272 0.0093 0.1031 0.0740 0.2244 | lr 4.8e-04 | norm 3.7885 | dt 0.166
type train | step 4480 | loss 0.0874 0.0291 0.9145 0.5701 4.5253 2.0356 19.1771 3.2095 0.0093 0.0974 0.0690 0.2126 | lr 4.8e-04 | norm 5.0547 | dt 0.157
type train | step 4490 | loss 0.0849 0.0266 0.8981 0.5471 4.5810 2.0798 19.3436 3.2122 0.0095 0.0944 0.0701 0.2120 | lr 4.7e-04 | norm 4.2423 | dt 0.157
type train | step 4500 | loss 0.0832 0.0259 0.8942 0.5531 4.5926 2.0458 19.2331 3.0990 0.0088 0.0983 0.0772 0.2134 | lr 4.7e-04 | norm 3.8256 | dt 0.147
type train | step 4510 | loss 0.0877 0.0299 0.9364 0.5990 4.6644 2.0956 19.5741 3.2416 0.0092 0.0997 0.0759 0.2136 | lr 4.7e-04 | norm 4.1111 | dt 0.159
type train | step 4520 | loss 0.0855 0.0285 0.8831 0.5694 4.4699 2.0535 18.8494 3.1348 0.0096 0.0986 0.0759 0.2123 | lr 4.7e-04 | norm 3.9612 | dt 0.164
type train | step 4530 | loss 0.0841 0.0320 0.8875 0.5634 4.4888 2.0313 19.2484 3.0775 0.0089 0.0974 0.0777 0.2045 | lr 4.7e-04 | norm 3.9157 | dt 0.172
type train | step 4540 | loss 0.0874 0.0297 0.9056 0.5438 4.4581 1.9808 18.9253 3.1733 0.0096 0.0981 0.0732 0.2129 | lr 4.6e-04 | norm 3.8182 | dt 0.161
type train | step 4550 | loss 0.0882 0.0312 0.9112 0.5513 4.6333 2.0750 19.5222 3.2125 0.0092 0.0969 0.0681 0.2097 | lr 4.6e-04 | norm 4.1777 | dt 0.144
type train | step 4560 | loss 0.0844 0.0284 0.8911 0.5485 4.4532 2.0656 18.9096 3.1160 0.0088 0.0982 0.0713 0.2063 | lr 4.6e-04 | norm 3.8209 | dt 0.149
type train | step 4570 | loss 0.0864 0.0301 0.9309 0.5650 4.5947 2.0936 19.1874 3.1696 0.0090 0.0933 0.0727 0.2035 | lr 4.6e-04 | norm 4.0893 | dt 0.171
type train | step 4580 | loss 0.0866 0.0306 0.8978 0.5780 4.5619 2.0665 19.3099 3.1025 0.0091 0.0949 0.0715 0.2250 | lr 4.6e-04 | norm 3.9868 | dt 0.158
type train | step 4590 | loss 0.0860 0.0294 0.8939 0.5507 4.5420 2.0613 19.2451 3.1386 0.0087 0.0945 0.0787 0.2008 | lr 4.5e-04 | norm 3.7569 | dt 0.156
type train | step 4600 | loss 0.0857 0.0283 0.9102 0.5658 4.6887 2.0087 19.7493 3.1840 0.0087 0.0943 0.0870 0.1855 | lr 4.5e-04 | norm 4.6404 | dt 0.161
type train | step 4610 | loss 0.0833 0.0275 0.9093 0.5465 4.6533 2.0220 19.5818 3.1186 0.0088 0.0899 0.0799 0.2029 | lr 4.5e-04 | norm 5.6135 | dt 0.146
type train | step 4620 | loss 0.0844 0.0282 0.9055 0.5439 4.6256 2.0266 19.4453 3.0898 0.0088 0.0863 0.0781 0.2078 | lr 4.5e-04 | norm 5.0541 | dt 0.161
type train | step 4630 | loss 0.0842 0.0277 0.8691 0.5440 4.3826 2.0098 18.6574 3.0593 0.0088 0.0828 0.0828 0.2107 | lr 4.5e-04 | norm 4.3113 | dt 0.165
type train | step 4640 | loss 0.0841 0.0281 0.8869 0.5465 4.4457 2.0928 19.2422 3.0866 0.0087 0.0838 0.0840 0.2213 | lr 4.4e-04 | norm 4.1760 | dt 0.166
type train | step 4650 | loss 0.0866 0.0306 0.9181 0.5485 4.5348 2.1377 19.6102 3.2102 0.0091 0.0814 0.0834 0.2028 | lr 4.4e-04 | norm 3.6248 | dt 0.152
type train | step 4660 | loss 0.0831 0.0289 0.9004 0.5598 4.6668 1.9753 19.7775 3.1081 0.0086 0.0835 0.0817 0.1939 | lr 4.4e-04 | norm 3.7025 | dt 0.146
type train | step 4670 | loss 0.0858 0.0291 0.9195 0.5324 4.5424 2.0266 19.0064 3.1846 0.0089 0.0811 0.0842 0.1991 | lr 4.4e-04 | norm 5.2653 | dt 0.148
type train | step 4680 | loss 0.0843 0.0296 0.9032 0.5425 4.5816 2.0523 19.2528 3.1336 0.0084 0.0840 0.0867 0.2137 | lr 4.4e-04 | norm 4.3646 | dt 0.161
type train | step 4690 | loss 0.0841 0.0305 0.9255 0.5348 4.5217 2.0564 19.0582 3.0513 0.0086 0.0807 0.0869 0.2128 | lr 4.3e-04 | norm 4.2044 | dt 0.164
type train | step 4700 | loss 0.0845 0.0293 0.9082 0.5300 4.6442 2.0933 19.5341 3.1124 0.0085 0.0797 0.0832 0.2033 | lr 4.3e-04 | norm 3.9751 | dt 0.154
type train | step 4710 | loss 0.0826 0.0295 0.8945 0.5223 4.4467 2.0598 19.0635 3.1042 0.0087 0.0751 0.0840 0.2185 | lr 4.3e-04 | norm 4.0514 | dt 0.164
type train | step 4720 | loss 0.0861 0.0288 0.9122 0.5303 4.6278 2.0682 19.6113 3.1153 0.0087 0.0765 0.0833 0.2011 | lr 4.3e-04 | norm 3.6917 | dt 0.162
type train | step 4730 | loss 0.0820 0.0290 0.8937 0.5013 4.3774 2.0093 18.6299 3.0096 0.0085 0.0701 0.0790 0.2036 | lr 4.2e-04 | norm 5.0389 | dt 0.163
type train | step 4740 | loss 0.0848 0.0310 0.9124 0.5261 4.5074 2.0360 19.3892 3.0334 0.0085 0.0738 0.0885 0.2145 | lr 4.2e-04 | norm 3.5124 | dt 0.162
type train | step 4750 | loss 0.0840 0.0295 0.9098 0.5174 4.5944 2.0256 19.5578 3.0538 0.0085 0.0685 0.0899 0.2093 | lr 4.2e-04 | norm 4.4095 | dt 0.149
type train | step 4760 | loss 0.0844 0.0301 0.9181 0.5120 4.6715 2.0429 19.6826 3.1154 0.0083 0.0681 0.0819 0.2069 | lr 4.2e-04 | norm 5.1478 | dt 0.155
type train | step 4770 | loss 0.0851 0.0299 0.9053 0.5143 4.5230 2.0452 19.5030 3.1224 0.0087 0.0620 0.0774 0.2103 | lr 4.2e-04 | norm 3.8700 | dt 0.153
type train | step 4780 | loss 0.0840 0.0297 0.9134 0.5045 4.7247 1.9934 20.0450 3.0600 0.0083 0.0653 0.0794 0.2003 | lr 4.1e-04 | norm 4.0563 | dt 0.169
type train | step 4790 | loss 0.0835 0.0291 0.9127 0.5330 4.6996 2.0110 19.8668 3.0572 0.0086 0.0627 0.0837 0.2094 | lr 4.1e-04 | norm 3.9830 | dt 0.165
type train | step 4800 | loss 0.0842 0.0314 0.9210 0.5514 4.6494 2.0364 20.0197 3.1209 0.0086 0.0635 0.0821 0.2170 | lr 4.1e-04 | norm 4.2732 | dt 0.164
type train | step 4810 | loss 0.0813 0.0288 0.8887 0.5089 4.4892 2.0021 19.0067 3.0202 0.0083 0.0622 0.0810 0.2124 | lr 4.1e-04 | norm 4.3616 | dt 0.162
type train | step 4820 | loss 0.0821 0.0289 0.9065 0.5329 4.5572 2.0493 19.4786 3.0556 0.0082 0.0591 0.0802 0.2111 | lr 4.1e-04 | norm 4.0459 | dt 0.166
type train | step 4830 | loss 0.0859 0.0324 0.9087 0.5237 4.5416 2.0309 19.4831 3.2238 0.0088 0.0563 0.0791 0.2040 | lr 4.0e-04 | norm 4.3056 | dt 0.163
type train | step 4840 | loss 0.0829 0.0293 0.9133 0.5321 4.7422 1.9684 20.4194 3.1704 0.0085 0.0623 0.0778 0.2095 | lr 4.0e-04 | norm 4.9121 | dt 0.152
type train | step 4850 | loss 0.0839 0.0293 0.9155 0.5241 4.5641 2.0655 19.0679 3.0414 0.0083 0.0555 0.0830 0.2220 | lr 4.0e-04 | norm 4.2302 | dt 0.157
type train | step 4860 | loss 0.0834 0.0313 0.9183 0.5311 4.5727 2.0208 19.8538 3.0913 0.0084 0.0588 0.0800 0.2148 | lr 4.0e-04 | norm 3.9335 | dt 0.165
type train | step 4870 | loss 0.0806 0.0280 0.8853 0.5579 4.5317 1.9696 19.0059 3.0222 0.0084 0.0543 0.0816 0.2097 | lr 4.0e-04 | norm 4.1547 | dt 0.167
type train | step 4880 | loss 0.0824 0.0293 0.9076 0.5586 4.7529 2.0052 19.9944 3.0267 0.0081 0.0559 0.0835 0.2028 | lr 4.0e-04 | norm 3.8909 | dt 0.162
type train | step 4890 | loss 0.0820 0.0280 0.9056 0.5388 4.5843 1.9720 19.5669 3.0632 0.0081 0.0551 0.0824 0.2101 | lr 3.9e-04 | norm 4.0073 | dt 0.144
type train | step 4900 | loss 0.0822 0.0287 0.9081 0.5430 4.6633 1.9866 19.9366 3.0699 0.0082 0.0553 0.0808 0.2053 | lr 3.9e-04 | norm 4.1542 | dt 0.162
type train | step 4910 | loss 0.0823 0.0278 0.8873 0.5211 4.4965 2.0063 19.0300 3.1065 0.0080 0.0534 0.0807 0.2010 | lr 3.9e-04 | norm 3.9682 | dt 0.145
type train | step 4920 | loss 0.0850 0.0326 0.9317 0.5235 4.6185 2.0432 19.8469 3.1658 0.0084 0.0533 0.0859 0.2112 | lr 3.9e-04 | norm 3.8487 | dt 0.166
type train | step 4930 | loss 0.0817 0.0273 0.8977 0.5296 4.5189 1.9884 18.9477 3.1080 0.0079 0.0519 0.0832 0.2005 | lr 3.9e-04 | norm 4.8302 | dt 0.163
type train | step 4940 | loss 0.0808 0.0267 0.8840 0.5043 4.5673 2.0254 19.5029 3.1000 0.0078 0.0502 0.0793 0.1895 | lr 3.8e-04 | norm 3.6586 | dt 0.151
type train | step 4950 | loss 0.0817 0.0264 0.8912 0.5227 4.5125 2.0224 19.0624 3.1669 0.0080 0.0486 0.0795 0.1991 | lr 3.8e-04 | norm 3.6352 | dt 0.148
type train | step 4960 | loss 0.0850 0.0281 0.9334 0.5205 4.8220 2.0082 20.5575 3.1705 0.0083 0.0497 0.0833 0.1989 | lr 3.8e-04 | norm 4.2412 | dt 0.169
type train | step 4970 | loss 0.0815 0.0288 0.9070 0.5159 4.5427 2.0278 19.3663 3.1131 0.0078 0.0481 0.0805 0.2007 | lr 3.8e-04 | norm 4.1905 | dt 0.162
type train | step 4980 | loss 0.0797 0.0287 0.8658 0.5193 4.3984 1.9899 18.9736 3.1442 0.0078 0.0488 0.0813 0.1871 | lr 3.8e-04 | norm 4.3906 | dt 0.161
type train | step 4990 | loss 0.0831 0.0290 0.9133 0.5179 4.6173 2.0018 19.2730 3.0892 0.0080 0.0453 0.0818 0.2053 | lr 3.7e-04 | norm 3.8540 | dt 0.159
type train | step 5000 | loss 0.0846 0.0306 0.9135 0.5405 4.4492 2.0448 18.9280 3.2029 0.0083 0.0472 0.0821 0.2193 | lr 3.7e-04 | norm 3.9384 | dt 0.160
type train | step 5010 | loss 0.0820 0.0303 0.8864 0.5322 4.4516 2.0879 19.2350 3.1396 0.0082 0.0477 0.0827 0.2105 | lr 3.7e-04 | norm 3.8805 | dt 0.154
type train | step 5020 | loss 0.0890 0.0318 0.9041 0.5458 4.5409 2.0550 19.4054 3.1904 0.0087 0.0471 0.0812 0.1982 | lr 3.7e-04 | norm 4.4802 | dt 0.144
type train | step 5030 | loss 0.0833 0.0304 0.9195 0.5134 4.6154 2.0200 19.6583 3.2149 0.0081 0.0471 0.0782 0.1951 | lr 3.7e-04 | norm 4.2470 | dt 0.164
type train | step 5040 | loss 0.0812 0.0289 0.8851 0.5231 4.5346 1.9914 19.6163 3.1655 0.0078 0.0460 0.0815 0.1922 | lr 3.6e-04 | norm 3.6966 | dt 0.156
type train | step 5050 | loss 0.0806 0.0282 0.9118 0.5178 4.5754 1.9586 19.1234 3.0813 0.0078 0.0438 0.0793 0.2059 | lr 3.6e-04 | norm 3.8369 | dt 0.148
type train | step 5060 | loss 0.0824 0.0311 0.9029 0.5580 4.4947 2.1022 19.3078 3.3017 0.0082 0.0476 0.0771 0.2018 | lr 3.6e-04 | norm 4.3018 | dt 0.169
type train | step 5070 | loss 0.0813 0.0299 0.8911 0.5200 4.5146 2.0283 19.2732 3.3961 0.0081 0.0445 0.0778 0.2001 | lr 3.6e-04 | norm 4.8329 | dt 0.144
type train | step 5080 | loss 0.0863 0.0304 0.9314 0.5373 4.6533 2.0447 20.0331 3.3592 0.0084 0.0492 0.0771 0.2095 | lr 3.6e-04 | norm 4.4918 | dt 0.146
type train | step 5090 | loss 0.0833 0.0304 0.9115 0.5239 4.5029 2.0301 19.2606 3.2686 0.0082 0.0477 0.0714 0.2087 | lr 3.5e-04 | norm 4.3125 | dt 0.158
type train | step 5100 | loss 0.0810 0.0283 0.8942 0.5075 4.5531 2.0405 19.4160 3.2391 0.0082 0.0486 0.0714 0.1969 | lr 3.5e-04 | norm 3.9503 | dt 0.168
type train | step 5110 | loss 0.0794 0.0274 0.8925 0.5010 4.5705 1.9977 19.2909 3.1649 0.0079 0.0551 0.0707 0.1967 | lr 3.5e-04 | norm 4.3525 | dt 0.169
type train | step 5120 | loss 0.0837 0.0310 0.9322 0.5354 4.6358 2.0908 19.6364 3.3304 0.0080 0.0527 0.0669 0.1964 | lr 3.5e-04 | norm 4.2095 | dt 0.165
type train | step 5130 | loss 0.0818 0.0288 0.8789 0.5314 4.4395 2.0601 18.8970 3.1619 0.0079 0.0564 0.0678 0.2014 | lr 3.5e-04 | norm 3.9498 | dt 0.159
type train | step 5140 | loss 0.0805 0.0298 0.8839 0.5389 4.4562 2.0949 19.2836 3.3224 0.0078 0.0553 0.0603 0.1982 | lr 3.5e-04 | norm 4.2213 | dt 0.162
type train | step 5150 | loss 0.0835 0.0298 0.8993 0.5154 4.4353 2.0254 18.9942 3.2472 0.0081 0.0548 0.0608 0.2034 | lr 3.4e-04 | norm 5.2076 | dt 0.152
type train | step 5160 | loss 0.0841 0.0307 0.9032 0.5264 4.5991 2.1120 19.5801 3.3569 0.0081 0.0553 0.0570 0.2017 | lr 3.4e-04 | norm 4.7069 | dt 0.163
type train | step 5170 | loss 0.0811 0.0284 0.8875 0.5156 4.4238 2.1291 18.9776 3.3121 0.0078 0.0490 0.0466 0.1959 | lr 3.4e-04 | norm 4.7200 | dt 0.155
type train | step 5180 | loss 0.0824 0.0308 0.9230 0.5445 4.5644 2.1128 19.2600 3.2752 0.0076 0.0524 0.0471 0.1913 | lr 3.4e-04 | norm 4.9506 | dt 0.164
type train | step 5190 | loss 0.0828 0.0306 0.8960 0.5532 4.5297 2.0416 19.3454 3.1567 0.0081 0.0505 0.0421 0.1967 | lr 3.4e-04 | norm 4.0138 | dt 0.162
type train | step 5200 | loss 0.0826 0.0298 0.8877 0.5265 4.5147 2.0498 19.2781 3.2755 0.0080 0.0489 0.0372 0.2010 | lr 3.3e-04 | norm 4.5007 | dt 0.161
type train | step 5210 | loss 0.0822 0.0297 0.9045 0.5488 4.6718 1.9681 19.8366 3.2659 0.0077 0.0544 0.0321 0.1982 | lr 3.3e-04 | norm 4.9051 | dt 0.163
type train | step 5220 | loss 0.0799 0.0288 0.8993 0.5364 4.6282 2.0030 19.6527 3.2958 0.0078 0.0481 0.0219 0.1925 | lr 3.3e-04 | norm 3.9997 | dt 0.161
type train | step 5230 | loss 0.0812 0.0295 0.8974 0.5309 4.5992 1.9963 19.5073 3.2750 0.0077 0.0509 0.0174 0.1916 | lr 3.3e-04 | norm 4.4014 | dt 0.168
type train | step 5240 | loss 0.0811 0.0293 0.8628 0.5166 4.3552 1.9754 18.7364 3.2761 0.0077 0.0483 0.0155 0.1952 | lr 3.3e-04 | norm 4.2665 | dt 0.164
type train | step 5250 | loss 0.0809 0.0300 0.8818 0.5329 4.4114 2.0194 19.2787 3.2559 0.0078 0.0497 0.0155 0.1959 | lr 3.3e-04 | norm 4.1406 | dt 0.165
type train | step 5260 | loss 0.0830 0.0315 0.9082 0.5308 4.5079 2.0731 19.6632 3.5045 0.0079 0.0509 0.0168 0.1941 | lr 3.2e-04 | norm 5.1819 | dt 0.171
type train | step 5270 | loss 0.0799 0.0287 0.8889 0.5123 4.6422 1.9985 19.8529 3.3014 0.0075 0.0516 0.0190 0.1816 | lr 3.2e-04 | norm 4.1625 | dt 0.165
type train | step 5280 | loss 0.0824 0.0297 0.9124 0.5286 4.5207 2.0036 19.0820 3.2099 0.0077 0.0505 0.0147 0.1944 | lr 3.2e-04 | norm 4.1287 | dt 0.171
type train | step 5290 | loss 0.0814 0.0299 0.8927 0.5369 4.5554 2.0544 19.3185 3.3020 0.0077 0.0493 0.0161 0.1884 | lr 3.2e-04 | norm 4.8790 | dt 0.158
type train | step 5300 | loss 0.0809 0.0303 0.9174 0.5236 4.4889 2.1160 19.1327 3.2314 0.0077 0.0499 0.0164 0.1875 | lr 3.2e-04 | norm 4.5112 | dt 0.166
type train | step 5310 | loss 0.0816 0.0291 0.9011 0.5692 4.6202 2.0504 19.5969 3.2320 0.0076 0.0531 0.0144 0.1868 | lr 3.1e-04 | norm 4.1120 | dt 0.167
type train | step 5320 | loss 0.0795 0.0299 0.8864 0.5431 4.4210 2.0565 19.1048 3.3537 0.0075 0.0495 0.0136 0.1879 | lr 3.1e-04 | norm 4.6867 | dt 0.166
type train | step 5330 | loss 0.0830 0.0296 0.9050 0.5452 4.5981 2.0780 19.6748 3.3497 0.0077 0.0504 0.0152 0.1741 | lr 3.1e-04 | norm 4.6326 | dt 0.146
type train | step 5340 | loss 0.0790 0.0301 0.8858 0.5127 4.3509 2.0510 18.6645 3.1921 0.0076 0.0480 0.0140 0.1842 | lr 3.1e-04 | norm 4.3379 | dt 0.160
type train | step 5350 | loss 0.0820 0.0311 0.9009 0.5359 4.4852 2.0024 19.4503 3.2718 0.0078 0.0496 0.0153 0.1865 | lr 3.1e-04 | norm 4.5173 | dt 0.161
type train | step 5360 | loss 0.0812 0.0298 0.9019 0.5323 4.5705 1.9840 19.6045 3.3449 0.0076 0.0480 0.0143 0.1851 | lr 3.1e-04 | norm 4.0174 | dt 0.163
type train | step 5370 | loss 0.0813 0.0296 0.9091 0.5453 4.6505 2.0634 19.7129 3.2971 0.0076 0.0468 0.0134 0.1890 | lr 3.0e-04 | norm 5.1463 | dt 0.160
type train | step 5380 | loss 0.0825 0.0299 0.8983 0.5490 4.4974 2.0546 19.5554 3.4036 0.0076 0.0449 0.0146 0.1847 | lr 3.0e-04 | norm 5.1997 | dt 0.151
type train | step 5390 | loss 0.0814 0.0295 0.9044 0.5426 4.7004 2.0387 20.1274 3.3764 0.0074 0.0425 0.0139 0.1814 | lr 3.0e-04 | norm 4.9221 | dt 0.143
type train | step 5400 | loss 0.0809 0.0287 0.9041 0.5391 4.6753 2.0386 19.9343 3.3131 0.0077 0.0437 0.0146 0.1906 | lr 3.0e-04 | norm 4.6378 | dt 0.156
type train | step 5410 | loss 0.0816 0.0310 0.9132 0.5388 4.6282 2.1026 20.0693 3.3743 0.0076 0.0441 0.0154 0.1905 | lr 3.0e-04 | norm 4.5183 | dt 0.142
type train | step 5420 | loss 0.0787 0.0284 0.8811 0.5220 4.4628 2.0456 19.0612 3.2524 0.0074 0.0429 0.0164 0.1895 | lr 2.9e-04 | norm 4.6553 | dt 0.159
type train | step 5430 | loss 0.0792 0.0286 0.8949 0.5370 4.5282 2.1276 19.5316 3.3562 0.0075 0.0429 0.0167 0.1985 | lr 2.9e-04 | norm 4.7257 | dt 0.152
type train | step 5440 | loss 0.0833 0.0299 0.8985 0.5480 4.5099 2.0742 19.5123 3.3436 0.0078 0.0421 0.0198 0.1896 | lr 2.9e-04 | norm 4.4064 | dt 0.155
type train | step 5450 | loss 0.0804 0.0298 0.9098 0.5475 4.7128 1.9826 20.4506 3.3898 0.0074 0.0410 0.0190 0.1874 | lr 2.9e-04 | norm 5.5097 | dt 0.162
type train | step 5460 | loss 0.0815 0.0287 0.9053 0.5214 4.5418 2.0169 19.1326 3.2399 0.0075 0.0415 0.0182 0.1984 | lr 2.9e-04 | norm 4.5953 | dt 0.152
type train | step 5470 | loss 0.0807 0.0305 0.9119 0.5247 4.5513 2.0480 19.9143 3.4610 0.0076 0.0429 0.0196 0.1892 | lr 2.9e-04 | norm 4.9180 | dt 0.158
type train | step 5480 | loss 0.0782 0.0275 0.8812 0.5340 4.5061 2.0671 19.0856 3.2586 0.0073 0.0413 0.0169 0.1876 | lr 2.8e-04 | norm 4.1165 | dt 0.159
type train | step 5490 | loss 0.0797 0.0289 0.9001 0.5501 4.7222 2.0579 20.0456 3.3155 0.0074 0.0404 0.0195 0.1884 | lr 2.8e-04 | norm 4.4035 | dt 0.167
type train | step 5500 | loss 0.0799 0.0292 0.8970 0.5570 4.5625 2.0097 19.6119 3.3114 0.0074 0.0408 0.0215 0.1892 | lr 2.8e-04 | norm 5.8269 | dt 0.146
type train | step 5510 | loss 0.0799 0.0280 0.9079 0.5428 4.6450 2.0015 20.0099 3.2684 0.0073 0.0416 0.0177 0.1869 | lr 2.8e-04 | norm 4.7344 | dt 0.151
type train | step 5520 | loss 0.0799 0.0275 0.8793 0.5372 4.4736 2.0537 19.0899 3.2945 0.0073 0.0393 0.0181 0.1923 | lr 2.8e-04 | norm 4.8103 | dt 0.166
type train | step 5530 | loss 0.0826 0.0312 0.9248 0.5576 4.5896 2.1196 19.8874 3.4309 0.0076 0.0413 0.0174 0.1878 | lr 2.8e-04 | norm 4.4445 | dt 0.164
type train | step 5540 | loss 0.0792 0.0255 0.8934 0.5302 4.4974 2.0152 18.9989 3.2294 0.0074 0.0394 0.0138 0.1847 | lr 2.7e-04 | norm 4.3278 | dt 0.146
type train | step 5550 | loss 0.0785 0.0249 0.8774 0.5211 4.5434 2.0058 19.5798 3.3222 0.0073 0.0384 0.0184 0.1831 | lr 2.7e-04 | norm 4.5923 | dt 0.163
type train | step 5560 | loss 0.0792 0.0258 0.8868 0.5136 4.4944 2.0284 19.1451 3.1960 0.0071 0.0397 0.0196 0.1811 | lr 2.7e-04 | norm 4.0722 | dt 0.168
type train | step 5570 | loss 0.0826 0.0278 0.9273 0.5476 4.7950 1.9930 20.6275 3.3631 0.0075 0.0407 0.0176 0.1822 | lr 2.7e-04 | norm 4.4138 | dt 0.164
type train | step 5580 | loss 0.0791 0.0276 0.9001 0.5085 4.5163 1.9993 19.4149 3.3506 0.0072 0.0386 0.0122 0.1936 | lr 2.7e-04 | norm 4.3774 | dt 0.164
type train | step 5590 | loss 0.0773 0.0289 0.8631 0.4988 4.3740 1.9606 19.0388 3.2644 0.0074 0.0370 0.0117 0.1916 | lr 2.7e-04 | norm 4.0667 | dt 0.145
type train | step 5600 | loss 0.0806 0.0289 0.9086 0.5220 4.5866 1.9979 19.3246 3.2166 0.0073 0.0393 0.0144 0.1898 | lr 2.6e-04 | norm 5.1188 | dt 0.162
type train | step 5610 | loss 0.0823 0.0323 0.9066 0.5389 4.4225 2.0312 19.0043 3.3458 0.0075 0.0387 0.0203 0.1885 | lr 2.6e-04 | norm 3.7832 | dt 0.152
type train | step 5620 | loss 0.0799 0.0296 0.8793 0.5141 4.4286 2.0338 19.3185 3.2503 0.0074 0.0379 0.0267 0.1897 | lr 2.6e-04 | norm 4.6503 | dt 0.164
type train | step 5630 | loss 0.0863 0.0320 0.8975 0.5352 4.5112 2.0116 19.4734 3.2955 0.0077 0.0388 0.0244 0.1856 | lr 2.6e-04 | norm 5.0670 | dt 0.159
type train | step 5640 | loss 0.0810 0.0304 0.9108 0.5228 4.5812 1.9886 19.7116 3.4680 0.0075 0.0374 0.0214 0.1812 | lr 2.6e-04 | norm 5.9211 | dt 0.161
type train | step 5650 | loss 0.0791 0.0294 0.8786 0.5279 4.5111 1.9621 19.6871 3.2788 0.0074 0.0408 0.0192 0.1812 | lr 2.6e-04 | norm 4.3506 | dt 0.166
type train | step 5660 | loss 0.0788 0.0269 0.9025 0.5059 4.5491 1.8963 19.2000 3.2290 0.0072 0.0414 0.0193 0.1890 | lr 2.6e-04 | norm 3.6964 | dt 0.163
type train | step 5670 | loss 0.0805 0.0296 0.8952 0.5590 4.4687 2.0087 19.3943 3.4506 0.0074 0.0408 0.0172 0.1939 | lr 2.5e-04 | norm 3.9619 | dt 0.162
type train | step 5680 | loss 0.0792 0.0294 0.8870 0.5405 4.4927 1.9630 19.3381 3.3190 0.0073 0.0423 0.0159 0.2044 | lr 2.5e-04 | norm 4.0906 | dt 0.166
type train | step 5690 | loss 0.0841 0.0293 0.9253 0.5724 4.6279 2.0292 20.0833 3.3989 0.0075 0.0481 0.0161 0.2030 | lr 2.5e-04 | norm 4.1253 | dt 0.158
type train | step 5700 | loss 0.0811 0.0288 0.9027 0.5537 4.4812 1.9757 19.3346 3.3885 0.0075 0.0498 0.0159 0.2045 | lr 2.5e-04 | norm 4.1930 | dt 0.164
type train | step 5710 | loss 0.0788 0.0272 0.8845 0.5355 4.5328 1.9984 19.4740 3.2602 0.0070 0.0476 0.0141 0.2003 | lr 2.5e-04 | norm 4.3367 | dt 0.156
type train | step 5720 | loss 0.0778 0.0261 0.8875 0.5413 4.5452 1.9534 19.3541 3.2910 0.0071 0.0470 0.0157 0.2099 | lr 2.5e-04 | norm 4.9597 | dt 0.145
type train | step 5730 | loss 0.0817 0.0296 0.9250 0.5688 4.6153 2.0315 19.7015 3.4059 0.0073 0.0492 0.0123 0.2108 | lr 2.4e-04 | norm 3.9099 | dt 0.161
type train | step 5740 | loss 0.0800 0.0282 0.8740 0.5533 4.4158 2.0660 18.9701 3.2745 0.0072 0.0462 0.0140 0.2100 | lr 2.4e-04 | norm 4.3622 | dt 0.165
type train | step 5750 | loss 0.0785 0.0282 0.8758 0.5390 4.4311 2.0330 19.3392 3.3048 0.0074 0.0437 0.0173 0.2042 | lr 2.4e-04 | norm 3.6574 | dt 0.161
type train | step 5760 | loss 0.0816 0.0290 0.8936 0.5367 4.4142 2.0398 19.0710 3.2426 0.0073 0.0446 0.0153 0.2051 | lr 2.4e-04 | norm 3.7756 | dt 0.146
type train | step 5770 | loss 0.0820 0.0303 0.8976 0.5285 4.5730 2.1507 19.6350 3.3716 0.0073 0.0479 0.0134 0.2028 | lr 2.4e-04 | norm 3.9871 | dt 0.163
type train | step 5780 | loss 0.0794 0.0274 0.8814 0.5411 4.3982 2.1108 19.0267 3.2974 0.0073 0.0444 0.0146 0.2052 | lr 2.4e-04 | norm 4.7586 | dt 0.167
type train | step 5790 | loss 0.0807 0.0282 0.9162 0.5468 4.5461 2.1232 19.3426 3.4132 0.0071 0.0474 0.0135 0.2069 | lr 2.4e-04 | norm 4.2589 | dt 0.157
type train | step 5800 | loss 0.0810 0.0284 0.8893 0.5473 4.5028 2.0701 19.3979 3.3388 0.0074 0.0445 0.0136 0.2111 | lr 2.3e-04 | norm 4.3572 | dt 0.162
type train | step 5810 | loss 0.0808 0.0289 0.8811 0.5210 4.4813 2.0224 19.3368 3.3684 0.0071 0.0440 0.0164 0.2092 | lr 2.3e-04 | norm 4.1356 | dt 0.162
type train | step 5820 | loss 0.0803 0.0271 0.8991 0.5427 4.6436 1.9926 19.8954 3.3625 0.0072 0.0443 0.0158 0.2032 | lr 2.3e-04 | norm 4.5991 | dt 0.152
type train | step 5830 | loss 0.0780 0.0277 0.8908 0.5276 4.6053 1.9673 19.7235 3.3502 0.0071 0.0435 0.0150 0.2059 | lr 2.3e-04 | norm 4.0154 | dt 0.160
type train | step 5840 | loss 0.0794 0.0280 0.8911 0.5428 4.5701 1.9808 19.5417 3.3188 0.0071 0.0446 0.0155 0.2053 | lr 2.3e-04 | norm 4.0428 | dt 0.158
type train | step 5850 | loss 0.0792 0.0273 0.8569 0.5312 4.3313 2.0193 18.7972 3.2924 0.0072 0.0456 0.0119 0.2010 | lr 2.3e-04 | norm 4.1637 | dt 0.156
type train | step 5860 | loss 0.0791 0.0280 0.8779 0.5410 4.3914 2.1062 19.3410 3.3892 0.0072 0.0487 0.0135 0.2049 | lr 2.2e-04 | norm 5.1953 | dt 0.148
type train | step 5870 | loss 0.0812 0.0290 0.9036 0.5327 4.4810 2.1095 19.7076 3.4192 0.0073 0.0453 0.0158 0.2126 | lr 2.2e-04 | norm 4.2992 | dt 0.165
type train | step 5880 | loss 0.0781 0.0271 0.8855 0.5398 4.6233 2.0002 19.9192 3.2559 0.0069 0.0486 0.0136 0.2024 | lr 2.2e-04 | norm 4.2020 | dt 0.162
type train | step 5890 | loss 0.0807 0.0268 0.9103 0.5050 4.4995 2.0163 19.1487 3.3440 0.0071 0.0449 0.0157 0.2024 | lr 2.2e-04 | norm 4.1889 | dt 0.158
type train | step 5900 | loss 0.0795 0.0278 0.8896 0.5296 4.5354 2.0456 19.3929 3.3275 0.0071 0.0475 0.0167 0.1979 | lr 2.2e-04 | norm 4.0929 | dt 0.154
type train | step 5910 | loss 0.0794 0.0288 0.9113 0.5148 4.4667 2.1079 19.1921 3.2854 0.0072 0.0473 0.0152 0.2055 | lr 2.2e-04 | norm 4.7243 | dt 0.155
type train | step 5920 | loss 0.0798 0.0278 0.8964 0.5331 4.5970 2.0447 19.6694 3.3976 0.0072 0.0457 0.0148 0.2049 | lr 2.2e-04 | norm 4.3942 | dt 0.147
type train | step 5930 | loss 0.0781 0.0278 0.8824 0.5249 4.3958 2.0901 19.1648 3.3650 0.0071 0.0433 0.0173 0.2077 | lr 2.1e-04 | norm 4.6722 | dt 0.159
type train | step 5940 | loss 0.0812 0.0271 0.9007 0.5370 4.5787 2.0593 19.7357 3.3180 0.0071 0.0462 0.0153 0.2057 | lr 2.1e-04 | norm 4.2286 | dt 0.162
type train | step 5950 | loss 0.0774 0.0283 0.8827 0.5195 4.3310 2.0266 18.7159 3.3631 0.0072 0.0476 0.0148 0.2102 | lr 2.1e-04 | norm 4.1287 | dt 0.151
type train | step 5960 | loss 0.0801 0.0311 0.8954 0.5478 4.4710 1.9724 19.5184 3.3685 0.0073 0.0483 0.0162 0.2141 | lr 2.1e-04 | norm 4.3572 | dt 0.160
type train | step 5970 | loss 0.0795 0.0281 0.8981 0.5372 4.5485 2.0996 19.6650 3.3329 0.0071 0.0459 0.0143 0.2205 | lr 2.1e-04 | norm 3.4686 | dt 0.147
type train | step 5980 | loss 0.0800 0.0296 0.9059 0.5383 4.6291 2.1121 19.7847 3.3041 0.0071 0.0487 0.0132 0.2135 | lr 2.1e-04 | norm 4.4120 | dt 0.156
type train | step 5990 | loss 0.0810 0.0294 0.8955 0.5529 4.4751 2.0490 19.6145 3.3533 0.0072 0.0451 0.0160 0.2128 | lr 2.1e-04 | norm 4.7650 | dt 0.158
type train | step 6000 | loss 0.0796 0.0279 0.9036 0.5584 4.6835 1.9828 20.1752 3.3733 0.0070 0.0464 0.0138 0.2209 | lr 2.1e-04 | norm 4.6633 | dt 0.147
type train | step 6010 | loss 0.0791 0.0290 0.9030 0.5437 4.6512 2.0062 19.9974 3.3471 0.0070 0.0491 0.0134 0.2223 | lr 2.0e-04 | norm 4.2801 | dt 0.166
type train | step 6020 | loss 0.0800 0.0312 0.9124 0.5476 4.6034 2.0768 20.1434 3.4819 0.0072 0.0507 0.0146 0.2269 | lr 2.0e-04 | norm 3.8199 | dt 0.167
type train | step 6030 | loss 0.0770 0.0288 0.8793 0.5209 4.4435 2.0486 19.1300 3.4568 0.0070 0.0473 0.0138 0.2254 | lr 2.0e-04 | norm 3.8736 | dt 0.166
type train | step 6040 | loss 0.0776 0.0292 0.8936 0.5442 4.5044 2.0646 19.6024 3.4276 0.0072 0.0463 0.0135 0.2286 | lr 2.0e-04 | norm 4.3385 | dt 0.160
type train | step 6050 | loss 0.0820 0.0313 0.8937 0.5512 4.4839 2.0418 19.5791 3.4811 0.0074 0.0486 0.0148 0.2375 | lr 2.0e-04 | norm 4.2349 | dt 0.166
type train | step 6060 | loss 0.0789 0.0298 0.9054 0.5591 4.6934 2.0111 20.4980 3.4426 0.0072 0.0477 0.0133 0.2396 | lr 2.0e-04 | norm 4.4178 | dt 0.146
type train | step 6070 | loss 0.0800 0.0288 0.9051 0.5312 4.5162 2.0427 19.1730 3.3411 0.0071 0.0498 0.0136 0.2442 | lr 2.0e-04 | norm 4.4266 | dt 0.147
type train | step 6080 | loss 0.0792 0.0315 0.9101 0.5455 4.5315 2.0200 19.9531 3.4495 0.0072 0.0466 0.0156 0.2413 | lr 1.9e-04 | norm 4.0003 | dt 0.147
type train | step 6090 | loss 0.0764 0.0278 0.8813 0.5574 4.4837 2.0723 19.1601 3.3881 0.0069 0.0488 0.0136 0.2457 | lr 1.9e-04 | norm 4.1703 | dt 0.165
type train | step 6100 | loss 0.0782 0.0292 0.9017 0.5383 4.7092 2.1039 20.1275 3.4164 0.0071 0.0446 0.0143 0.2486 | lr 1.9e-04 | norm 4.9286 | dt 0.144
type train | step 6110 | loss 0.0784 0.0289 0.8969 0.5609 4.5430 2.0436 19.6774 3.4858 0.0070 0.0475 0.0144 0.2468 | lr 1.9e-04 | norm 5.6476 | dt 0.158
type train | step 6120 | loss 0.0786 0.0286 0.9074 0.5583 4.6258 2.0111 20.0716 3.3650 0.0070 0.0497 0.0130 0.2546 | lr 1.9e-04 | norm 5.2231 | dt 0.158
type train | step 6130 | loss 0.0785 0.0289 0.8793 0.5491 4.4577 2.0716 19.1533 3.2585 0.0070 0.0470 0.0132 0.2474 | lr 1.9e-04 | norm 4.3869 | dt 0.160
type train | step 6140 | loss 0.0813 0.0326 0.9225 0.5469 4.5672 2.1296 19.9276 3.3737 0.0075 0.0489 0.0141 0.2438 | lr 1.9e-04 | norm 3.7323 | dt 0.158
type train | step 6150 | loss 0.0778 0.0284 0.8958 0.5549 4.4812 2.0888 19.0001 3.2702 0.0070 0.0455 0.0145 0.2478 | lr 1.9e-04 | norm 4.1239 | dt 0.146
type train | step 6160 | loss 0.0770 0.0278 0.8807 0.5434 4.5284 2.0530 19.6172 3.3086 0.0069 0.0508 0.0153 0.2388 | lr 1.8e-04 | norm 4.0435 | dt 0.152
type train | step 6170 | loss 0.0780 0.0287 0.8878 0.5342 4.4809 2.0201 19.2087 3.2932 0.0070 0.0564 0.0135 0.2412 | lr 1.8e-04 | norm 3.8791 | dt 0.166
type train | step 6180 | loss 0.0813 0.0308 0.9309 0.5556 4.7769 1.9809 20.6759 3.4255 0.0071 0.0585 0.0128 0.2422 | lr 1.8e-04 | norm 4.0058 | dt 0.164
type train | step 6190 | loss 0.0775 0.0287 0.9018 0.5311 4.4889 2.0245 19.4724 3.2766 0.0070 0.0527 0.0138 0.2410 | lr 1.8e-04 | norm 3.9965 | dt 0.159
type train | step 6200 | loss 0.0760 0.0298 0.8666 0.5306 4.3556 1.9576 19.0891 3.1995 0.0069 0.0522 0.0158 0.2305 | lr 1.8e-04 | norm 3.5207 | dt 0.163
type train | step 6210 | loss 0.0792 0.0289 0.9096 0.5277 4.5687 2.0389 19.3625 3.2441 0.0070 0.0515 0.0131 0.2314 | lr 1.8e-04 | norm 3.7567 | dt 0.164
type train | step 6220 | loss 0.0808 0.0308 0.9109 0.5289 4.4091 2.1285 19.0614 3.3947 0.0071 0.0497 0.0137 0.2282 | lr 1.8e-04 | norm 4.3548 | dt 0.156
type train | step 6230 | loss 0.0784 0.0294 0.8859 0.5365 4.4155 2.0680 19.3740 3.2518 0.0071 0.0561 0.0148 0.2263 | lr 1.8e-04 | norm 4.0169 | dt 0.169
type train | step 6240 | loss 0.0850 0.0314 0.9017 0.5346 4.4930 2.0701 19.5197 3.2934 0.0072 0.0563 0.0120 0.2207 | lr 1.8e-04 | norm 4.2225 | dt 0.146
type train | step 6250 | loss 0.0796 0.0293 0.9149 0.5202 4.5607 2.0886 19.7756 3.2882 0.0071 0.0596 0.0134 0.2239 | lr 1.7e-04 | norm 4.6989 | dt 0.157
type train | step 6260 | loss 0.0780 0.0292 0.8804 0.5182 4.4986 2.0198 19.7618 3.2825 0.0069 0.0557 0.0163 0.2146 | lr 1.7e-04 | norm 4.4324 | dt 0.148
type train | step 6270 | loss 0.0774 0.0282 0.9024 0.5200 4.5332 2.0037 19.2419 3.2085 0.0067 0.0560 0.0145 0.2181 | lr 1.7e-04 | norm 4.0515 | dt 0.169
type train | step 6280 | loss 0.0790 0.0296 0.8975 0.5458 4.4563 2.0934 19.4639 3.3550 0.0071 0.0573 0.0138 0.2178 | lr 1.7e-04 | norm 4.3486 | dt 0.147
type train | step 6290 | loss 0.0777 0.0292 0.8910 0.5449 4.4715 2.0631 19.3631 3.2866 0.0071 0.0522 0.0135 0.2235 | lr 1.7e-04 | norm 4.3713 | dt 0.172
type train | step 6300 | loss 0.0827 0.0303 0.9269 0.5388 4.6113 2.0467 20.1331 3.4098 0.0071 0.0512 0.0148 0.2240 | lr 1.7e-04 | norm 4.5710 | dt 0.161
type train | step 6310 | loss 0.0797 0.0304 0.9023 0.5181 4.4664 2.0493 19.3710 3.4044 0.0072 0.0533 0.0130 0.2227 | lr 1.7e-04 | norm 4.5820 | dt 0.146
type train | step 6320 | loss 0.0775 0.0294 0.8884 0.5225 4.5150 2.0262 19.5177 3.3277 0.0070 0.0517 0.0114 0.2318 | lr 1.7e-04 | norm 4.4257 | dt 0.156
type train | step 6330 | loss 0.0764 0.0281 0.8869 0.5249 4.5247 1.9572 19.3861 3.1440 0.0068 0.0513 0.0121 0.2278 | lr 1.7e-04 | norm 4.1987 | dt 0.147
type train | step 6340 | loss 0.0804 0.0306 0.9277 0.5468 4.6006 2.1038 19.7541 3.3773 0.0070 0.0537 0.0124 0.2228 | lr 1.6e-04 | norm 4.2987 | dt 0.149
type train | step 6350 | loss 0.0786 0.0302 0.8812 0.5308 4.3975 2.0383 19.0121 3.4482 0.0070 0.0522 0.0140 0.2287 | lr 1.6e-04 | norm 3.9846 | dt 0.159
type train | step 6360 | loss 0.0771 0.0306 0.8817 0.5335 4.4175 2.0346 19.3606 3.3410 0.0070 0.0518 0.0145 0.2226 | lr 1.6e-04 | norm 3.8326 | dt 0.173
type train | step 6370 | loss 0.0803 0.0302 0.8989 0.5098 4.3980 2.0150 19.1195 3.2721 0.0070 0.0530 0.0136 0.2236 | lr 1.6e-04 | norm 3.7975 | dt 0.145
type train | step 6380 | loss 0.0805 0.0306 0.8995 0.5303 4.5570 2.0399 19.6773 3.2642 0.0070 0.0512 0.0127 0.2285 | lr 1.6e-04 | norm 3.5784 | dt 0.147
type train | step 6390 | loss 0.0780 0.0291 0.8868 0.5107 4.3752 2.0600 19.0598 3.2319 0.0068 0.0506 0.0143 0.2228 | lr 1.6e-04 | norm 3.7272 | dt 0.161
type train | step 6400 | loss 0.0793 0.0305 0.9238 0.5371 4.5272 2.0351 19.3834 3.3000 0.0069 0.0532 0.0161 0.2200 | lr 1.6e-04 | norm 4.1392 | dt 0.164
type train | step 6410 | loss 0.0796 0.0294 0.8956 0.5357 4.4793 2.0935 19.4399 3.3161 0.0073 0.0528 0.0157 0.2231 | lr 1.6e-04 | norm 4.1456 | dt 0.157
type train | step 6420 | loss 0.0793 0.0293 0.8884 0.5307 4.4647 2.0490 19.3865 3.3119 0.0070 0.0511 0.0141 0.2282 | lr 1.6e-04 | norm 4.2651 | dt 0.159
type train | step 6430 | loss 0.0790 0.0286 0.9046 0.5285 4.6292 2.0221 19.9469 3.3434 0.0069 0.0564 0.0130 0.2171 | lr 1.5e-04 | norm 3.7115 | dt 0.163
type train | step 6440 | loss 0.0767 0.0287 0.8970 0.5362 4.5952 2.0585 19.7940 3.4428 0.0069 0.0538 0.0137 0.2190 | lr 1.5e-04 | norm 4.5281 | dt 0.166
type train | step 6450 | loss 0.0783 0.0290 0.8965 0.5252 4.5459 2.0348 19.5478 3.3992 0.0069 0.0521 0.0150 0.2231 | lr 1.5e-04 | norm 5.3140 | dt 0.158
type train | step 6460 | loss 0.0778 0.0288 0.8634 0.5231 4.3116 2.0370 18.8523 3.2375 0.0068 0.0511 0.0147 0.2247 | lr 1.5e-04 | norm 4.1711 | dt 0.160
type train | step 6470 | loss 0.0777 0.0294 0.8841 0.5234 4.3716 2.1008 19.4017 3.3570 0.0070 0.0513 0.0149 0.2237 | lr 1.5e-04 | norm 4.2774 | dt 0.153
type train | step 6480 | loss 0.0799 0.0316 0.9110 0.5213 4.4632 2.0589 19.7664 3.4648 0.0071 0.0515 0.0151 0.2276 | lr 1.5e-04 | norm 4.2708 | dt 0.163
type train | step 6490 | loss 0.0769 0.0292 0.8958 0.5194 4.6093 1.9945 19.9878 3.2974 0.0067 0.0529 0.0129 0.2190 | lr 1.5e-04 | norm 4.4373 | dt 0.165
type train | step 6500 | loss 0.0795 0.0286 0.9194 0.5027 4.4859 2.0265 19.2092 3.3131 0.0069 0.0504 0.0130 0.2174 | lr 1.5e-04 | norm 4.0955 | dt 0.164
type train | step 6510 | loss 0.0782 0.0298 0.8984 0.5249 4.5174 2.0881 19.4421 3.3421 0.0070 0.0495 0.0136 0.2181 | lr 1.5e-04 | norm 4.4803 | dt 0.159
type train | step 6520 | loss 0.0782 0.0304 0.9163 0.4999 4.4485 2.0985 19.2108 3.3407 0.0069 0.0499 0.0140 0.2188 | lr 1.5e-04 | norm 4.4988 | dt 0.154
type train | step 6530 | loss 0.0784 0.0288 0.9014 0.5237 4.5805 2.0356 19.7262 3.3247 0.0069 0.0498 0.0127 0.2192 | lr 1.5e-04 | norm 4.2051 | dt 0.146
type train | step 6540 | loss 0.0768 0.0288 0.8886 0.5074 4.3732 2.1183 19.2180 3.4410 0.0069 0.0491 0.0136 0.2242 | lr 1.4e-04 | norm 4.5295 | dt 0.147
type train | step 6550 | loss 0.0799 0.0296 0.9084 0.5246 4.5632 2.0999 19.7710 3.4176 0.0068 0.0523 0.0138 0.2277 | lr 1.4e-04 | norm 3.6475 | dt 0.154
type train | step 6560 | loss 0.0763 0.0297 0.8931 0.5025 4.3172 2.0844 18.7682 3.3512 0.0069 0.0496 0.0160 0.2343 | lr 1.4e-04 | norm 4.3393 | dt 0.175
type train | step 6570 | loss 0.0789 0.0312 0.9044 0.5209 4.4568 2.0368 19.5562 3.3633 0.0070 0.0502 0.0169 0.2327 | lr 1.4e-04 | norm 4.7288 | dt 0.153
type train | step 6580 | loss 0.0780 0.0288 0.9071 0.5198 4.5292 2.0234 19.6976 3.3696 0.0068 0.0486 0.0148 0.2332 | lr 1.4e-04 | norm 3.7923 | dt 0.150
type train | step 6590 | loss 0.0787 0.0298 0.9151 0.5191 4.6101 2.0567 19.8413 3.3154 0.0068 0.0492 0.0149 0.2297 | lr 1.4e-04 | norm 4.1149 | dt 0.160
type train | step 6600 | loss 0.0795 0.0293 0.9028 0.5090 4.4561 2.0945 19.6687 3.3888 0.0070 0.0486 0.0157 0.2324 | lr 1.4e-04 | norm 3.8530 | dt 0.168
type train | step 6610 | loss 0.0786 0.0284 0.9095 0.5160 4.6657 1.9577 20.2241 3.3327 0.0067 0.0488 0.0158 0.2272 | lr 1.4e-04 | norm 4.1240 | dt 0.167
type train | step 6620 | loss 0.0780 0.0288 0.9102 0.5093 4.6338 2.0368 20.0611 3.3491 0.0069 0.0500 0.0157 0.2204 | lr 1.4e-04 | norm 4.5663 | dt 0.164
type train | step 6630 | loss 0.0790 0.0309 0.9219 0.5082 4.5858 2.0099 20.1872 3.3515 0.0070 0.0496 0.0167 0.2254 | lr 1.4e-04 | norm 3.9677 | dt 0.167
type train | step 6640 | loss 0.0760 0.0284 0.8867 0.5020 4.4280 2.0452 19.1809 3.2833 0.0067 0.0487 0.0133 0.2277 | lr 1.4e-04 | norm 4.1966 | dt 0.167
type train | step 6650 | loss 0.0765 0.0287 0.9056 0.5146 4.4888 2.0983 19.6439 3.3531 0.0068 0.0475 0.0133 0.2245 | lr 1.3e-04 | norm 4.1460 | dt 0.161
type train | step 6660 | loss 0.0807 0.0318 0.8992 0.5124 4.4639 2.1004 19.6063 3.4316 0.0070 0.0463 0.0147 0.2263 | lr 1.3e-04 | norm 4.2752 | dt 0.162
type train | step 6670 | loss 0.0779 0.0293 0.9115 0.5183 4.6774 2.0338 20.5300 3.3094 0.0069 0.0483 0.0130 0.2256 | lr 1.3e-04 | norm 4.2267 | dt 0.146
type train | step 6680 | loss 0.0789 0.0290 0.9117 0.4941 4.4944 2.0568 19.2274 3.2864 0.0067 0.0493 0.0133 0.2275 | lr 1.3e-04 | norm 4.9991 | dt 0.157
type train | step 6690 | loss 0.0783 0.0306 0.9152 0.5174 4.5141 2.0638 20.0057 3.4101 0.0069 0.0489 0.0164 0.2166 | lr 1.3e-04 | norm 5.1360 | dt 0.159
type train | step 6700 | loss 0.0754 0.0276 0.8883 0.5052 4.4688 2.0587 19.2044 3.2903 0.0066 0.0487 0.0152 0.2182 | lr 1.3e-04 | norm 4.3524 | dt 0.164
type train | step 6710 | loss 0.0771 0.0291 0.9065 0.5115 4.6951 2.0559 20.1826 3.3090 0.0068 0.0485 0.0146 0.2184 | lr 1.3e-04 | norm 4.7684 | dt 0.153
type train | step 6720 | loss 0.0771 0.0292 0.9029 0.5184 4.5274 1.9687 19.7193 3.3836 0.0068 0.0452 0.0151 0.2230 | lr 1.3e-04 | norm 4.3382 | dt 0.159
type train | step 6730 | loss 0.0775 0.0286 0.9114 0.5178 4.6077 2.0066 20.1145 3.4176 0.0066 0.0473 0.0139 0.2243 | lr 1.3e-04 | norm 4.2538 | dt 0.147
type train | step 6740 | loss 0.0776 0.0287 0.8858 0.4911 4.4434 1.9666 19.2048 3.3208 0.0067 0.0465 0.0147 0.2296 | lr 1.3e-04 | norm 4.3465 | dt 0.164
type train | step 6750 | loss 0.0806 0.0320 0.9255 0.5076 4.5505 2.0731 19.9616 3.4593 0.0070 0.0482 0.0155 0.2321 | lr 1.3e-04 | norm 5.0023 | dt 0.145
type train | step 6760 | loss 0.0769 0.0272 0.9014 0.4956 4.4655 2.0468 19.0397 3.3199 0.0067 0.0470 0.0133 0.2319 | lr 1.3e-04 | norm 4.0184 | dt 0.147
type train | step 6770 | loss 0.0759 0.0280 0.8874 0.4967 4.5154 2.0707 19.6666 3.3781 0.0065 0.0455 0.0149 0.2273 | lr 1.3e-04 | norm 3.6624 | dt 0.169
type train | step 6780 | loss 0.0769 0.0283 0.8942 0.4945 4.4635 2.0199 19.2403 3.3603 0.0067 0.0437 0.0149 0.2369 | lr 1.3e-04 | norm 4.2427 | dt 0.156
type train | step 6790 | loss 0.0802 0.0300 0.9368 0.5054 4.7619 2.0211 20.7288 3.4890 0.0069 0.0438 0.0138 0.2328 | lr 1.2e-04 | norm 4.4626 | dt 0.149
type train | step 6800 | loss 0.0766 0.0290 0.9030 0.4837 4.4721 2.0619 19.5273 3.4278 0.0068 0.0458 0.0139 0.2318 | lr 1.2e-04 | norm 5.7139 | dt 0.168
type train | step 6810 | loss 0.0750 0.0288 0.8701 0.4826 4.3428 1.9739 19.1364 3.3259 0.0066 0.0461 0.0162 0.2389 | lr 1.2e-04 | norm 4.6662 | dt 0.158
type train | step 6820 | loss 0.0784 0.0294 0.9144 0.4870 4.5533 1.9706 19.3901 3.3364 0.0068 0.0453 0.0142 0.2410 | lr 1.2e-04 | norm 4.8766 | dt 0.155
type train | step 6830 | loss 0.0799 0.0303 0.9146 0.5131 4.3932 2.0362 19.0830 3.4469 0.0069 0.0453 0.0134 0.2429 | lr 1.2e-04 | norm 3.9077 | dt 0.171
type train | step 6840 | loss 0.0776 0.0290 0.8918 0.5086 4.3989 2.0691 19.4092 3.3635 0.0068 0.0440 0.0143 0.2380 | lr 1.2e-04 | norm 3.8875 | dt 0.160
type train | step 6850 | loss 0.0841 0.0315 0.9041 0.5160 4.4766 2.0955 19.5853 3.3140 0.0069 0.0438 0.0128 0.2373 | lr 1.2e-04 | norm 4.6866 | dt 0.159
type train | step 6860 | loss 0.0789 0.0301 0.9172 0.4921 4.5451 2.0392 19.8268 3.4159 0.0068 0.0448 0.0139 0.2419 | lr 1.2e-04 | norm 4.3171 | dt 0.169
type train | step 6870 | loss 0.0771 0.0283 0.8859 0.5014 4.4864 2.0003 19.8198 3.3197 0.0067 0.0468 0.0151 0.2379 | lr 1.2e-04 | norm 4.0958 | dt 0.164
type train | step 6880 | loss 0.0768 0.0276 0.9079 0.4921 4.5189 1.9493 19.2769 3.2712 0.0065 0.0446 0.0139 0.2409 | lr 1.2e-04 | norm 3.7774 | dt 0.151
type train | step 6890 | loss 0.0782 0.0305 0.9027 0.5113 4.4392 2.0176 19.4821 3.4083 0.0068 0.0454 0.0119 0.2375 | lr 1.2e-04 | norm 3.9822 | dt 0.167
type train | step 6900 | loss 0.0768 0.0294 0.8965 0.5069 4.4539 2.0426 19.3822 3.4317 0.0068 0.0438 0.0121 0.2330 | lr 1.2e-04 | norm 4.0820 | dt 0.166
type train | step 6910 | loss 0.0818 0.0303 0.9333 0.5039 4.5999 2.0431 20.1974 3.4465 0.0068 0.0449 0.0137 0.2390 | lr 1.2e-04 | norm 5.2270 | dt 0.161
type train | step 6920 | loss 0.0789 0.0294 0.9077 0.5042 4.4518 2.0214 19.4048 3.3589 0.0068 0.0459 0.0138 0.2392 | lr 1.2e-04 | norm 4.2157 | dt 0.153
type train | step 6930 | loss 0.0767 0.0283 0.8939 0.4959 4.5054 2.0387 19.5730 3.2577 0.0067 0.0452 0.0123 0.2348 | lr 1.2e-04 | norm 3.6796 | dt 0.165
type train | step 6940 | loss 0.0757 0.0275 0.8921 0.4862 4.5099 1.9854 19.4267 3.2644 0.0065 0.0456 0.0127 0.2378 | lr 1.2e-04 | norm 3.8949 | dt 0.157
type train | step 6950 | loss 0.0797 0.0307 0.9328 0.5090 4.5895 2.1317 19.7607 3.3485 0.0067 0.0426 0.0117 0.2346 | lr 1.1e-04 | norm 4.5982 | dt 0.151
type train | step 6960 | loss 0.0778 0.0289 0.8869 0.4953 4.3837 2.0556 19.0518 3.2883 0.0067 0.0426 0.0135 0.2332 | lr 1.1e-04 | norm 4.4838 | dt 0.147
type train | step 6970 | loss 0.0764 0.0290 0.8859 0.5041 4.4045 2.0333 19.3969 3.3008 0.0067 0.0433 0.0142 0.2242 | lr 1.1e-04 | norm 4.2099 | dt 0.164
type train | step 6980 | loss 0.0795 0.0297 0.9052 0.4923 4.3866 2.0414 19.1671 3.2845 0.0067 0.0447 0.0149 0.2271 | lr 1.1e-04 | norm 3.9495 | dt 0.146
type train | step 6990 | loss 0.0796 0.0304 0.9048 0.4998 4.5427 2.0233 19.7178 3.4204 0.0068 0.0449 0.0152 0.2280 | lr 1.1e-04 | norm 3.6960 | dt 0.155
type train | step 7000 | loss 0.0773 0.0285 0.8908 0.4972 4.3579 2.0535 19.0923 3.3335 0.0066 0.0442 0.0149 0.2236 | lr 1.1e-04 | norm 3.7873 | dt 0.166
type train | step 7010 | loss 0.0788 0.0296 0.9282 0.5048 4.5161 2.0493 19.4117 3.2415 0.0067 0.0428 0.0147 0.2248 | lr 1.1e-04 | norm 5.1590 | dt 0.151
type train | step 7020 | loss 0.0787 0.0299 0.9007 0.5224 4.4635 2.0715 19.4796 3.2831 0.0069 0.0446 0.0144 0.2264 | lr 1.1e-04 | norm 4.6706 | dt 0.155
type train | step 7030 | loss 0.0786 0.0294 0.8927 0.5045 4.4475 2.0170 19.4264 3.3801 0.0068 0.0467 0.0153 0.2306 | lr 1.1e-04 | norm 4.7406 | dt 0.168
type train | step 7040 | loss 0.0784 0.0291 0.9105 0.5080 4.6182 2.0157 19.9822 3.3375 0.0066 0.0468 0.0137 0.2256 | lr 1.1e-04 | norm 4.2634 | dt 0.158
type train | step 7050 | loss 0.0760 0.0286 0.9005 0.5038 4.5837 1.9920 19.8342 3.3368 0.0066 0.0453 0.0124 0.2250 | lr 1.1e-04 | norm 4.1152 | dt 0.167
type train | step 7060 | loss 0.0772 0.0286 0.8980 0.4997 4.5358 2.0327 19.5855 3.3500 0.0067 0.0454 0.0122 0.2230 | lr 1.1e-04 | norm 4.4953 | dt 0.168
type train | step 7070 | loss 0.0772 0.0284 0.8671 0.5030 4.3016 2.0126 18.8924 3.3364 0.0066 0.0435 0.0120 0.2263 | lr 1.1e-04 | norm 3.9654 | dt 0.165
type train | step 7080 | loss 0.0770 0.0297 0.8887 0.5004 4.3615 2.0594 19.4474 3.3654 0.0067 0.0448 0.0121 0.2254 | lr 1.1e-04 | norm 4.0842 | dt 0.164
type train | step 7090 | loss 0.0790 0.0308 0.9191 0.5091 4.4479 2.0744 19.7987 3.4537 0.0069 0.0460 0.0137 0.2337 | lr 1.1e-04 | norm 4.3755 | dt 0.147
type train | step 7100 | loss 0.0761 0.0279 0.8997 0.5044 4.6023 1.9870 20.0239 3.3252 0.0067 0.0440 0.0139 0.2241 | lr 1.1e-04 | norm 5.3063 | dt 0.167
type train | step 7110 | loss 0.0789 0.0286 0.9258 0.4849 4.4744 1.9964 19.2533 3.3459 0.0066 0.0449 0.0153 0.2233 | lr 1.1e-04 | norm 4.6399 | dt 0.159
type train | step 7120 | loss 0.0771 0.0295 0.9034 0.5033 4.5027 2.0230 19.4731 3.3211 0.0066 0.0453 0.0134 0.2305 | lr 1.1e-04 | norm 4.2150 | dt 0.162
type train | step 7130 | loss 0.0775 0.0290 0.9187 0.5014 4.4372 2.0818 19.2490 3.3674 0.0067 0.0443 0.0118 0.2328 | lr 1.1e-04 | norm 4.4505 | dt 0.157
type train | step 7140 | loss 0.0778 0.0282 0.9045 0.5110 4.5674 2.0032 19.7611 3.4402 0.0067 0.0446 0.0118 0.2283 | lr 1.1e-04 | norm 3.9343 | dt 0.152
type train | step 7150 | loss 0.0763 0.0282 0.8973 0.4875 4.3626 2.1229 19.2423 3.4072 0.0066 0.0442 0.0130 0.2265 | lr 1.1e-04 | norm 4.5560 | dt 0.145
type train | step 7160 | loss 0.0793 0.0288 0.9157 0.5041 4.5540 2.1136 19.8105 3.3522 0.0066 0.0438 0.0138 0.2263 | lr 1.1e-04 | norm 4.2551 | dt 0.166
type train | step 7170 | loss 0.0758 0.0282 0.8964 0.4729 4.3050 1.9936 18.8089 3.3155 0.0067 0.0433 0.0155 0.2257 | lr 1.1e-04 | norm 4.9785 | dt 0.148
type train | step 7180 | loss 0.0782 0.0304 0.9107 0.4935 4.4492 2.0349 19.5997 3.3530 0.0067 0.0442 0.0149 0.2246 | lr 1.0e-04 | norm 3.7744 | dt 0.165
type train | step 7190 | loss 0.0774 0.0286 0.9085 0.4980 4.5196 2.1184 19.7318 3.3715 0.0066 0.0437 0.0117 0.2277 | lr 1.0e-04 | norm 4.1796 | dt 0.157
type train | step 7200 | loss 0.0781 0.0287 0.9203 0.4980 4.5999 2.0969 19.8606 3.3195 0.0066 0.0423 0.0123 0.2258 | lr 1.0e-04 | norm 3.9880 | dt 0.159
type train | step 7210 | loss 0.0790 0.0299 0.9102 0.5004 4.4429 2.0376 19.7188 3.4689 0.0066 0.0438 0.0137 0.2232 | lr 1.0e-04 | norm 4.3897 | dt 0.165
type train | step 7220 | loss 0.0780 0.0279 0.9150 0.4925 4.6577 2.0104 20.2715 3.3447 0.0065 0.0444 0.0134 0.2222 | lr 1.0e-04 | norm 4.7193 | dt 0.150
type train | step 7230 | loss 0.0775 0.0279 0.9154 0.4899 4.6214 2.0432 20.0900 3.4253 0.0066 0.0438 0.0125 0.2220 | lr 1.0e-04 | norm 4.6372 | dt 0.163
type train | step 7240 | loss 0.0784 0.0309 0.9260 0.5035 4.5803 2.0829 20.2262 3.4213 0.0067 0.0444 0.0135 0.2236 | lr 1.0e-04 | norm 4.2346 | dt 0.159
type train | step 7250 | loss 0.0753 0.0275 0.8896 0.4989 4.4193 2.0519 19.2112 3.3617 0.0065 0.0429 0.0139 0.2263 | lr 1.0e-04 | norm 4.2141 | dt 0.159
type train | step 7260 | loss 0.0759 0.0275 0.9080 0.5059 4.4796 2.0465 19.6770 3.3392 0.0065 0.0434 0.0131 0.2257 | lr 1.0e-04 | norm 4.5065 | dt 0.157
type train | step 7270 | loss 0.0802 0.0306 0.9026 0.5062 4.4539 2.0704 19.6403 3.4248 0.0067 0.0430 0.0130 0.2258 | lr 1.0e-04 | norm 5.1624 | dt 0.159
type train | step 7280 | loss 0.0773 0.0283 0.9154 0.5068 4.6683 2.0020 20.5661 3.4152 0.0065 0.0443 0.0124 0.2237 | lr 1.0e-04 | norm 4.9950 | dt 0.144
type train | step 7290 | loss 0.0787 0.0275 0.9152 0.4835 4.4827 2.0569 19.2667 3.3839 0.0066 0.0434 0.0127 0.2272 | lr 1.0e-04 | norm 5.2949 | dt 0.144
type train | step 7300 | loss 0.0777 0.0321 0.9200 0.5039 4.5059 2.0383 20.0388 3.3988 0.0066 0.0439 0.0147 0.2200 | lr 1.0e-04 | norm 4.5519 | dt 0.153
type train | step 7310 | loss 0.0749 0.0281 0.8942 0.4913 4.4582 2.0243 19.2320 3.2407 0.0064 0.0426 0.0125 0.2205 | lr 1.0e-04 | norm 4.1368 | dt 0.166
type train | step 7320 | loss 0.0764 0.0291 0.9094 0.5024 4.6847 2.0824 20.2163 3.3603 0.0066 0.0425 0.0119 0.2207 | lr 1.0e-04 | norm 3.6399 | dt 0.168
type train | step 7330 | loss 0.0766 0.0286 0.9065 0.5051 4.5187 2.0195 19.7636 3.3060 0.0065 0.0422 0.0137 0.2213 | lr 1.0e-04 | norm 4.0012 | dt 0.161
type train | step 7340 | loss 0.0768 0.0286 0.9159 0.5013 4.6014 2.0353 20.1478 3.3750 0.0064 0.0426 0.0136 0.2238 | lr 1.0e-04 | norm 4.6623 | dt 0.148
type train | step 7350 | loss 0.0772 0.0274 0.8888 0.4825 4.4325 2.0574 19.2281 3.3239 0.0065 0.0424 0.0152 0.2236 | lr 1.0e-04 | norm 4.4155 | dt 0.165
type train | step 7360 | loss 0.0800 0.0313 0.9287 0.4953 4.5423 2.1122 20.0030 3.4925 0.0068 0.0427 0.0170 0.2232 | lr 1.0e-04 | norm 4.2640 | dt 0.159
type train | step 7370 | loss 0.0764 0.0266 0.9052 0.4894 4.4568 2.0626 19.0652 3.3451 0.0065 0.0413 0.0149 0.2262 | lr 1.0e-04 | norm 4.0792 | dt 0.161
type train | step 7380 | loss 0.0754 0.0274 0.8904 0.4838 4.5082 2.0007 19.6929 3.3790 0.0064 0.0408 0.0170 0.2238 | lr 1.0e-04 | norm 4.4508 | dt 0.161
type train | step 7390 | loss 0.0764 0.0276 0.8995 0.4779 4.4540 2.0440 19.2687 3.3114 0.0065 0.0400 0.0169 0.2285 | lr 1.0e-04 | norm 4.2042 | dt 0.153
type train | step 7400 | loss 0.0797 0.0300 0.9413 0.4935 4.7551 2.0638 20.7565 3.3885 0.0068 0.0407 0.0172 0.2261 | lr 1.0e-04 | norm 4.8839 | dt 0.169
type train | step 7410 | loss 0.0761 0.0283 0.9068 0.4793 4.4622 2.0329 19.5583 3.3792 0.0066 0.0414 0.0174 0.2262 | lr 1.0e-04 | norm 4.2323 | dt 0.160
type train | step 7420 | loss 0.0747 0.0305 0.8726 0.4731 4.3365 1.9807 19.1723 3.3159 0.0064 0.0417 0.0189 0.2268 | lr 1.0e-04 | norm 4.0382 | dt 0.149
type train | step 7430 | loss 0.0778 0.0288 0.9185 0.4812 4.5467 1.9955 19.4173 3.3693 0.0065 0.0398 0.0185 0.2322 | lr 1.0e-04 | norm 4.4789 | dt 0.168
type train | step 7440 | loss 0.0793 0.0307 0.9157 0.5036 4.3892 2.0615 19.1098 3.4695 0.0068 0.0401 0.0168 0.2300 | lr 1.0e-04 | norm 3.7305 | dt 0.158
type train | step 7450 | loss 0.0771 0.0283 0.8970 0.4957 4.3928 2.0668 19.4491 3.3318 0.0067 0.0396 0.0170 0.2361 | lr 1.0e-04 | norm 4.0865 | dt 0.160
type train | step 7460 | loss 0.0836 0.0321 0.9091 0.5066 4.4695 2.0890 19.6173 3.3930 0.0068 0.0413 0.0162 0.2350 | lr 1.0e-04 | norm 4.6009 | dt 0.162
type train | step 7470 | loss 0.0784 0.0297 0.9213 0.4871 4.5336 2.0286 19.8389 3.3479 0.0067 0.0401 0.0186 0.2348 | lr 1.0e-04 | norm 4.8387 | dt 0.154
type train | step 7480 | loss 0.0766 0.0289 0.8906 0.4885 4.4816 1.9859 19.8555 3.2808 0.0065 0.0394 0.0168 0.2310 | lr 1.0e-04 | norm 5.3804 | dt 0.150
type train | step 7490 | loss 0.0761 0.0277 0.9091 0.4722 4.5084 1.9765 19.3035 3.3259 0.0063 0.0400 0.0153 0.2287 | lr 1.0e-04 | norm 4.2328 | dt 0.167
type train | step 7500 | loss 0.0775 0.0299 0.9048 0.5000 4.4346 2.0082 19.4988 3.3226 0.0066 0.0395 0.0173 0.2260 | lr 1.0e-04 | norm 4.4218 | dt 0.162