Methmani/Image_Classification_fashion-mnist
136
1{2 "best_metric": 0.9086,3 "best_model_checkpoint": "./results/checkpoint-4680",4 "epoch": 9.9984,5 "global_step": 4680,6 "is_hyper_param_search": false,7 "is_local_process_zero": true,8 "is_world_process_zero": true,9 "log_history": [10 {11 "epoch": 0.02,12 "learning_rate": 1.0683760683760685e-06,13 "loss": 2.3089,14 "step": 1015 },16 {17 "epoch": 0.04,18 "learning_rate": 2.136752136752137e-06,19 "loss": 2.2658,20 "step": 2021 },22 {23 "epoch": 0.06,24 "learning_rate": 3.205128205128205e-06,25 "loss": 2.1767,26 "step": 3027 },28 {29 "epoch": 0.09,30 "learning_rate": 4.273504273504274e-06,31 "loss": 2.0595,32 "step": 4033 },34 {35 "epoch": 0.11,36 "learning_rate": 5.341880341880342e-06,37 "loss": 1.8958,38 "step": 5039 },40 {41 "epoch": 0.13,42 "learning_rate": 6.41025641025641e-06,43 "loss": 1.753,44 "step": 6045 },46 {47 "epoch": 0.15,48 "learning_rate": 7.478632478632479e-06,49 "loss": 1.4999,50 "step": 7051 },52 {53 "epoch": 0.17,54 "learning_rate": 8.547008547008548e-06,55 "loss": 1.2355,56 "step": 8057 },58 {59 "epoch": 0.19,60 "learning_rate": 9.615384615384616e-06,61 "loss": 1.0098,62 "step": 9063 },64 {65 "epoch": 0.21,66 "learning_rate": 1.0683760683760684e-05,67 "loss": 0.7874,68 "step": 10069 },70 {71 "epoch": 0.23,72 "learning_rate": 1.1752136752136752e-05,73 "loss": 0.7019,74 "step": 11075 },76 {77 "epoch": 0.26,78 "learning_rate": 1.282051282051282e-05,79 "loss": 0.6812,80 "step": 12081 },82 {83 "epoch": 0.28,84 "learning_rate": 1.388888888888889e-05,85 "loss": 0.5752,86 "step": 13087 },88 {89 "epoch": 0.3,90 "learning_rate": 1.4957264957264958e-05,91 "loss": 0.6161,92 "step": 14093 },94 {95 "epoch": 0.32,96 "learning_rate": 1.602564102564103e-05,97 "loss": 0.5919,98 "step": 15099 },100 {101 "epoch": 0.34,102 "learning_rate": 1.7094017094017095e-05,103 "loss": 0.5563,104 "step": 160105 },106 {107 "epoch": 0.36,108 "learning_rate": 1.8162393162393162e-05,109 "loss": 0.5661,110 "step": 170111 },112 {113 "epoch": 0.38,114 "learning_rate": 1.923076923076923e-05,115 "loss": 0.5844,116 "step": 180117 },118 {119 "epoch": 0.41,120 "learning_rate": 2.02991452991453e-05,121 "loss": 0.5954,122 "step": 190123 },124 {125 "epoch": 0.43,126 "learning_rate": 2.1367521367521368e-05,127 "loss": 0.5298,128 "step": 200129 },130 {131 "epoch": 0.45,132 "learning_rate": 2.2435897435897437e-05,133 "loss": 0.5206,134 "step": 210135 },136 {137 "epoch": 0.47,138 "learning_rate": 2.3504273504273504e-05,139 "loss": 0.5294,140 "step": 220141 },142 {143 "epoch": 0.49,144 "learning_rate": 2.4572649572649573e-05,145 "loss": 0.5235,146 "step": 230147 },148 {149 "epoch": 0.51,150 "learning_rate": 2.564102564102564e-05,151 "loss": 0.4706,152 "step": 240153 },154 {155 "epoch": 0.53,156 "learning_rate": 2.670940170940171e-05,157 "loss": 0.5064,158 "step": 250159 },160 {161 "epoch": 0.55,162 "learning_rate": 2.777777777777778e-05,163 "loss": 0.4844,164 "step": 260165 },166 {167 "epoch": 0.58,168 "learning_rate": 2.8846153846153845e-05,169 "loss": 0.481,170 "step": 270171 },172 {173 "epoch": 0.6,174 "learning_rate": 2.9914529914529915e-05,175 "loss": 0.4895,176 "step": 280177 },178 {179 "epoch": 0.62,180 "learning_rate": 3.098290598290599e-05,181 "loss": 0.5137,182 "step": 290183 },184 {185 "epoch": 0.64,186 "learning_rate": 3.205128205128206e-05,187 "loss": 0.4769,188 "step": 300189 },190 {191 "epoch": 0.66,192 "learning_rate": 3.311965811965812e-05,193 "loss": 0.5685,194 "step": 310195 },196 {197 "epoch": 0.68,198 "learning_rate": 3.418803418803419e-05,199 "loss": 0.4679,200 "step": 320201 },202 {203 "epoch": 0.7,204 "learning_rate": 3.525641025641026e-05,205 "loss": 0.4553,206 "step": 330207 },208 {209 "epoch": 0.73,210 "learning_rate": 3.6324786324786323e-05,211 "loss": 0.4842,212 "step": 340213 },214 {215 "epoch": 0.75,216 "learning_rate": 3.739316239316239e-05,217 "loss": 0.5318,218 "step": 350219 },220 {221 "epoch": 0.77,222 "learning_rate": 3.846153846153846e-05,223 "loss": 0.4787,224 "step": 360225 },226 {227 "epoch": 0.79,228 "learning_rate": 3.952991452991453e-05,229 "loss": 0.4367,230 "step": 370231 },232 {233 "epoch": 0.81,234 "learning_rate": 4.05982905982906e-05,235 "loss": 0.467,236 "step": 380237 },238 {239 "epoch": 0.83,240 "learning_rate": 4.166666666666667e-05,241 "loss": 0.4721,242 "step": 390243 },244 {245 "epoch": 0.85,246 "learning_rate": 4.2735042735042735e-05,247 "loss": 0.4258,248 "step": 400249 },250 {251 "epoch": 0.87,252 "learning_rate": 4.3803418803418805e-05,253 "loss": 0.4811,254 "step": 410255 },256 {257 "epoch": 0.9,258 "learning_rate": 4.4871794871794874e-05,259 "loss": 0.4632,260 "step": 420261 },262 {263 "epoch": 0.92,264 "learning_rate": 4.594017094017094e-05,265 "loss": 0.4413,266 "step": 430267 },268 {269 "epoch": 0.94,270 "learning_rate": 4.700854700854701e-05,271 "loss": 0.4483,272 "step": 440273 },274 {275 "epoch": 0.96,276 "learning_rate": 4.8076923076923084e-05,277 "loss": 0.4529,278 "step": 450279 },280 {281 "epoch": 0.98,282 "learning_rate": 4.9145299145299147e-05,283 "loss": 0.4201,284 "step": 460285 },286 {287 "epoch": 1.0,288 "eval_accuracy": 0.8306,289 "eval_loss": 0.4693240821361542,290 "eval_runtime": 47.1099,291 "eval_samples_per_second": 212.27,292 "eval_steps_per_second": 6.644,293 "step": 468294 },295 {296 "epoch": 1.0,297 "learning_rate": 4.997625830959164e-05,298 "loss": 0.4572,299 "step": 470300 },301 {302 "epoch": 1.03,303 "learning_rate": 4.985754985754986e-05,304 "loss": 0.4523,305 "step": 480306 },307 {308 "epoch": 1.05,309 "learning_rate": 4.9738841405508076e-05,310 "loss": 0.4367,311 "step": 490312 },313 {314 "epoch": 1.07,315 "learning_rate": 4.962013295346629e-05,316 "loss": 0.4355,317 "step": 500318 },319 {320 "epoch": 1.09,321 "learning_rate": 4.95014245014245e-05,322 "loss": 0.4326,323 "step": 510324 },325 {326 "epoch": 1.11,327 "learning_rate": 4.938271604938271e-05,328 "loss": 0.4255,329 "step": 520330 },331 {332 "epoch": 1.13,333 "learning_rate": 4.926400759734093e-05,334 "loss": 0.4132,335 "step": 530336 },337 {338 "epoch": 1.15,339 "learning_rate": 4.9145299145299147e-05,340 "loss": 0.4226,341 "step": 540342 },343 {344 "epoch": 1.17,345 "learning_rate": 4.902659069325736e-05,346 "loss": 0.4443,347 "step": 550348 },349 {350 "epoch": 1.2,351 "learning_rate": 4.890788224121557e-05,352 "loss": 0.4309,353 "step": 560354 },355 {356 "epoch": 1.22,357 "learning_rate": 4.878917378917379e-05,358 "loss": 0.4062,359 "step": 570360 },361 {362 "epoch": 1.24,363 "learning_rate": 4.867046533713201e-05,364 "loss": 0.406,365 "step": 580366 },367 {368 "epoch": 1.26,369 "learning_rate": 4.8551756885090224e-05,370 "loss": 0.4252,371 "step": 590372 },373 {374 "epoch": 1.28,375 "learning_rate": 4.8433048433048433e-05,376 "loss": 0.4016,377 "step": 600378 },379 {380 "epoch": 1.3,381 "learning_rate": 4.831433998100665e-05,382 "loss": 0.411,383 "step": 610384 },385 {386 "epoch": 1.32,387 "learning_rate": 4.819563152896486e-05,388 "loss": 0.4265,389 "step": 620390 },391 {392 "epoch": 1.35,393 "learning_rate": 4.8076923076923084e-05,394 "loss": 0.4245,395 "step": 630396 },397 {398 "epoch": 1.37,399 "learning_rate": 4.7958214624881294e-05,400 "loss": 0.3896,401 "step": 640402 },403 {404 "epoch": 1.39,405 "learning_rate": 4.783950617283951e-05,406 "loss": 0.3938,407 "step": 650408 },409 {410 "epoch": 1.41,411 "learning_rate": 4.772079772079772e-05,412 "loss": 0.4191,413 "step": 660414 },415 {416 "epoch": 1.43,417 "learning_rate": 4.760208926875594e-05,418 "loss": 0.4494,419 "step": 670420 },421 {422 "epoch": 1.45,423 "learning_rate": 4.7483380816714154e-05,424 "loss": 0.4286,425 "step": 680426 },427 {428 "epoch": 1.47,429 "learning_rate": 4.736467236467237e-05,430 "loss": 0.3869,431 "step": 690432 },433 {434 "epoch": 1.49,435 "learning_rate": 4.724596391263058e-05,436 "loss": 0.365,437 "step": 700438 },439 {440 "epoch": 1.52,441 "learning_rate": 4.71272554605888e-05,442 "loss": 0.4056,443 "step": 710444 },445 {446 "epoch": 1.54,447 "learning_rate": 4.700854700854701e-05,448 "loss": 0.4182,449 "step": 720450 },451 {452 "epoch": 1.56,453 "learning_rate": 4.688983855650523e-05,454 "loss": 0.3884,455 "step": 730456 },457 {458 "epoch": 1.58,459 "learning_rate": 4.677113010446344e-05,460 "loss": 0.3964,461 "step": 740462 },463 {464 "epoch": 1.6,465 "learning_rate": 4.665242165242166e-05,466 "loss": 0.3833,467 "step": 750468 },469 {470 "epoch": 1.62,471 "learning_rate": 4.653371320037987e-05,472 "loss": 0.397,473 "step": 760474 },475 {476 "epoch": 1.64,477 "learning_rate": 4.6415004748338084e-05,478 "loss": 0.3863,479 "step": 770480 },481 {482 "epoch": 1.67,483 "learning_rate": 4.62962962962963e-05,484 "loss": 0.3379,485 "step": 780486 },487 {488 "epoch": 1.69,489 "learning_rate": 4.617758784425451e-05,490 "loss": 0.3488,491 "step": 790492 },493 {494 "epoch": 1.71,495 "learning_rate": 4.605887939221273e-05,496 "loss": 0.4028,497 "step": 800498 },499 {500 "epoch": 1.73,501 "learning_rate": 4.594017094017094e-05,502 "loss": 0.3853,503 "step": 810504 },505 {506 "epoch": 1.75,507 "learning_rate": 4.582146248812916e-05,508 "loss": 0.3498,509 "step": 820510 },511 {512 "epoch": 1.77,513 "learning_rate": 4.570275403608737e-05,514 "loss": 0.3778,515 "step": 830516 },517 {518 "epoch": 1.79,519 "learning_rate": 4.558404558404559e-05,520 "loss": 0.4053,521 "step": 840522 },523 {524 "epoch": 1.81,525 "learning_rate": 4.54653371320038e-05,526 "loss": 0.3552,527 "step": 850528 },529 {530 "epoch": 1.84,531 "learning_rate": 4.5346628679962014e-05,532 "loss": 0.3976,533 "step": 860534 },535 {536 "epoch": 1.86,537 "learning_rate": 4.522792022792023e-05,538 "loss": 0.3513,539 "step": 870540 },541 {542 "epoch": 1.88,543 "learning_rate": 4.510921177587845e-05,544 "loss": 0.3319,545 "step": 880546 },547 {548 "epoch": 1.9,549 "learning_rate": 4.499050332383666e-05,550 "loss": 0.3563,551 "step": 890552 },553 {554 "epoch": 1.92,555 "learning_rate": 4.4871794871794874e-05,556 "loss": 0.3692,557 "step": 900558 },559 {560 "epoch": 1.94,561 "learning_rate": 4.4753086419753084e-05,562 "loss": 0.362,563 "step": 910564 },565 {566 "epoch": 1.96,567 "learning_rate": 4.463437796771131e-05,568 "loss": 0.3649,569 "step": 920570 },571 {572 "epoch": 1.99,573 "learning_rate": 4.451566951566952e-05,574 "loss": 0.3531,575 "step": 930576 },577 {578 "epoch": 2.0,579 "eval_accuracy": 0.8684,580 "eval_loss": 0.3599550426006317,581 "eval_runtime": 47.3884,582 "eval_samples_per_second": 211.022,583 "eval_steps_per_second": 6.605,584 "step": 936585 },586 {587 "epoch": 2.01,588 "learning_rate": 4.4396961063627735e-05,589 "loss": 0.4153,590 "step": 940591 },592 {593 "epoch": 2.03,594 "learning_rate": 4.4278252611585945e-05,595 "loss": 0.3254,596 "step": 950597 },598 {599 "epoch": 2.05,600 "learning_rate": 4.415954415954416e-05,601 "loss": 0.3572,602 "step": 960603 },604 {605 "epoch": 2.07,606 "learning_rate": 4.404083570750238e-05,607 "loss": 0.3116,608 "step": 970609 },610 {611 "epoch": 2.09,612 "learning_rate": 4.3922127255460595e-05,613 "loss": 0.3853,614 "step": 980615 },616 {617 "epoch": 2.12,618 "learning_rate": 4.3803418803418805e-05,619 "loss": 0.3605,620 "step": 990621 },622 {623 "epoch": 2.14,624 "learning_rate": 4.368471035137702e-05,625 "loss": 0.3495,626 "step": 1000627 },628 {629 "epoch": 2.16,630 "learning_rate": 4.356600189933523e-05,631 "loss": 0.3736,632 "step": 1010633 },634 {635 "epoch": 2.18,636 "learning_rate": 4.344729344729345e-05,637 "loss": 0.3913,638 "step": 1020639 },640 {641 "epoch": 2.2,642 "learning_rate": 4.3328584995251665e-05,643 "loss": 0.3568,644 "step": 1030645 },646 {647 "epoch": 2.22,648 "learning_rate": 4.3209876543209875e-05,649 "loss": 0.3522,650 "step": 1040651 },652 {653 "epoch": 2.24,654 "learning_rate": 4.309116809116809e-05,655 "loss": 0.3462,656 "step": 1050657 },658 {659 "epoch": 2.26,660 "learning_rate": 4.297245963912631e-05,661 "loss": 0.34,662 "step": 1060663 },664 {665 "epoch": 2.29,666 "learning_rate": 4.2853751187084525e-05,667 "loss": 0.3718,668 "step": 1070669 },670 {671 "epoch": 2.31,672 "learning_rate": 4.2735042735042735e-05,673 "loss": 0.3828,674 "step": 1080675 },676 {677 "epoch": 2.33,678 "learning_rate": 4.261633428300095e-05,679 "loss": 0.3283,680 "step": 1090681 },682 {683 "epoch": 2.35,684 "learning_rate": 4.249762583095916e-05,685 "loss": 0.4015,686 "step": 1100687 },688 {689 "epoch": 2.37,690 "learning_rate": 4.2378917378917385e-05,691 "loss": 0.3469,692 "step": 1110693 },694 {695 "epoch": 2.39,696 "learning_rate": 4.2260208926875595e-05,697 "loss": 0.3562,698 "step": 1120699 },700 {701 "epoch": 2.41,702 "learning_rate": 4.214150047483381e-05,703 "loss": 0.3386,704 "step": 1130705 },706 {707 "epoch": 2.44,708 "learning_rate": 4.202279202279202e-05,709 "loss": 0.3469,710 "step": 1140711 },712 {713 "epoch": 2.46,714 "learning_rate": 4.190408357075024e-05,715 "loss": 0.3164,716 "step": 1150717 },718 {719 "epoch": 2.48,720 "learning_rate": 4.1785375118708455e-05,721 "loss": 0.3688,722 "step": 1160723 },724 {725 "epoch": 2.5,726 "learning_rate": 4.166666666666667e-05,727 "loss": 0.342,728 "step": 1170729 },730 {731 "epoch": 2.52,732 "learning_rate": 4.154795821462488e-05,733 "loss": 0.3089,734 "step": 1180735 },736 {737 "epoch": 2.54,738 "learning_rate": 4.14292497625831e-05,739 "loss": 0.3229,740 "step": 1190741 },742 {743 "epoch": 2.56,744 "learning_rate": 4.131054131054131e-05,745 "loss": 0.3834,746 "step": 1200747 },748 {749 "epoch": 2.58,750 "learning_rate": 4.119183285849953e-05,751 "loss": 0.3545,752 "step": 1210753 },754 {755 "epoch": 2.61,756 "learning_rate": 4.107312440645774e-05,757 "loss": 0.3156,758 "step": 1220759 },760 {761 "epoch": 2.63,762 "learning_rate": 4.095441595441596e-05,763 "loss": 0.3213,764 "step": 1230765 },766 {767 "epoch": 2.65,768 "learning_rate": 4.083570750237417e-05,769 "loss": 0.3634,770 "step": 1240771 },772 {773 "epoch": 2.67,774 "learning_rate": 4.0716999050332386e-05,775 "loss": 0.3813,776 "step": 1250777 },778 {779 "epoch": 2.69,780 "learning_rate": 4.05982905982906e-05,781 "loss": 0.3542,782 "step": 1260783 },784 {785 "epoch": 2.71,786 "learning_rate": 4.047958214624881e-05,787 "loss": 0.3294,788 "step": 1270789 },790 {791 "epoch": 2.73,792 "learning_rate": 4.036087369420703e-05,793 "loss": 0.3166,794 "step": 1280795 },796 {797 "epoch": 2.76,798 "learning_rate": 4.024216524216524e-05,799 "loss": 0.3437,800 "step": 1290801 },802 {803 "epoch": 2.78,804 "learning_rate": 4.012345679012346e-05,805 "loss": 0.3741,806 "step": 1300807 },808 {809 "epoch": 2.8,810 "learning_rate": 4.000474833808167e-05,811 "loss": 0.3283,812 "step": 1310813 },814 {815 "epoch": 2.82,816 "learning_rate": 3.988603988603989e-05,817 "loss": 0.3057,818 "step": 1320819 },820 {821 "epoch": 2.84,822 "learning_rate": 3.97673314339981e-05,823 "loss": 0.3714,824 "step": 1330825 },826 {827 "epoch": 2.86,828 "learning_rate": 3.9648622981956316e-05,829 "loss": 0.3622,830 "step": 1340831 },832 {833 "epoch": 2.88,834 "learning_rate": 3.952991452991453e-05,835 "loss": 0.3114,836 "step": 1350837 },838 {839 "epoch": 2.9,840 "learning_rate": 3.941120607787275e-05,841 "loss": 0.303,842 "step": 1360843 },844 {845 "epoch": 2.93,846 "learning_rate": 3.929249762583096e-05,847 "loss": 0.3716,848 "step": 1370849 },850 {851 "epoch": 2.95,852 "learning_rate": 3.9173789173789176e-05,853 "loss": 0.302,854 "step": 1380855 },856 {857 "epoch": 2.97,858 "learning_rate": 3.9055080721747386e-05,859 "loss": 0.3431,860 "step": 1390861 },862 {863 "epoch": 2.99,864 "learning_rate": 3.893637226970561e-05,865 "loss": 0.3058,866 "step": 1400867 },868 {869 "epoch": 3.0,870 "eval_accuracy": 0.869,871 "eval_loss": 0.38148394227027893,872 "eval_runtime": 47.5558,873 "eval_samples_per_second": 210.279,874 "eval_steps_per_second": 6.582,875 "step": 1404876 },877 {878 "epoch": 3.01,879 "learning_rate": 3.881766381766382e-05,880 "loss": 0.3738,881 "step": 1410882 },883 {884 "epoch": 3.03,885 "learning_rate": 3.8698955365622036e-05,886 "loss": 0.3314,887 "step": 1420888 },889 {890 "epoch": 3.06,891 "learning_rate": 3.8580246913580246e-05,892 "loss": 0.324,893 "step": 1430894 },895 {896 "epoch": 3.08,897 "learning_rate": 3.846153846153846e-05,898 "loss": 0.3121,899 "step": 1440900 },901 {902 "epoch": 3.1,903 "learning_rate": 3.834283000949668e-05,904 "loss": 0.3253,905 "step": 1450906 },907 {908 "epoch": 3.12,909 "learning_rate": 3.8224121557454896e-05,910 "loss": 0.2757,911 "step": 1460912 },913 {914 "epoch": 3.14,915 "learning_rate": 3.8105413105413106e-05,916 "loss": 0.3142,917 "step": 1470918 },919 {920 "epoch": 3.16,921 "learning_rate": 3.798670465337132e-05,922 "loss": 0.3552,923 "step": 1480924 },925 {926 "epoch": 3.18,927 "learning_rate": 3.786799620132954e-05,928 "loss": 0.2739,929 "step": 1490930 },931 {932 "epoch": 3.2,933 "learning_rate": 3.774928774928775e-05,934 "loss": 0.3667,935 "step": 1500936 },937 {938 "epoch": 3.23,939 "learning_rate": 3.7630579297245966e-05,940 "loss": 0.3323,941 "step": 1510942 },943 {944 "epoch": 3.25,945 "learning_rate": 3.7511870845204176e-05,946 "loss": 0.3214,947 "step": 1520948 },949 {950 "epoch": 3.27,951 "learning_rate": 3.739316239316239e-05,952 "loss": 0.3369,953 "step": 1530954 },955 {956 "epoch": 3.29,957 "learning_rate": 3.727445394112061e-05,958 "loss": 0.3375,959 "step": 1540960 },961 {962 "epoch": 3.31,963 "learning_rate": 3.7155745489078827e-05,964 "loss": 0.3019,965 "step": 1550966 },967 {968 "epoch": 3.33,969 "learning_rate": 3.7037037037037037e-05,970 "loss": 0.3012,971 "step": 1560972 },973 {974 "epoch": 3.35,975 "learning_rate": 3.691832858499525e-05,976 "loss": 0.3172,977 "step": 1570978 },979 {980 "epoch": 3.38,981 "learning_rate": 3.679962013295346e-05,982 "loss": 0.2997,983 "step": 1580984 },985 {986 "epoch": 3.4,987 "learning_rate": 3.668091168091169e-05,988 "loss": 0.3282,989 "step": 1590990 },991 {992 "epoch": 3.42,993 "learning_rate": 3.65622032288699e-05,994 "loss": 0.3156,995 "step": 1600996 },997 {998 "epoch": 3.44,999 "learning_rate": 3.6443494776828113e-05,1000 "loss": 0.3159,1001 "step": 16101002 },1003 {1004 "epoch": 3.46,1005 "learning_rate": 3.6324786324786323e-05,1006 "loss": 0.3228,1007 "step": 16201008 },1009 {1010 "epoch": 3.48,1011 "learning_rate": 3.620607787274454e-05,1012 "loss": 0.3314,1013 "step": 16301014 },1015 {1016 "epoch": 3.5,1017 "learning_rate": 3.608736942070276e-05,1018 "loss": 0.3157,1019 "step": 16401020 },1021 {1022 "epoch": 3.52,1023 "learning_rate": 3.5968660968660974e-05,1024 "loss": 0.3322,1025 "step": 16501026 },1027 {1028 "epoch": 3.55,1029 "learning_rate": 3.5849952516619184e-05,1030 "loss": 0.3038,1031 "step": 16601032 },1033 {1034 "epoch": 3.57,1035 "learning_rate": 3.57312440645774e-05,1036 "loss": 0.3524,1037 "step": 16701038 },1039 {1040 "epoch": 3.59,1041 "learning_rate": 3.561253561253561e-05,1042 "loss": 0.3029,1043 "step": 16801044 },1045 {1046 "epoch": 3.61,1047 "learning_rate": 3.5493827160493834e-05,1048 "loss": 0.304,1049 "step": 16901050 },1051 {1052 "epoch": 3.63,1053 "learning_rate": 3.5375118708452044e-05,1054 "loss": 0.2968,1055 "step": 17001056 },1057 {1058 "epoch": 3.65,1059 "learning_rate": 3.525641025641026e-05,1060 "loss": 0.3294,1061 "step": 17101062 },1063 {1064 "epoch": 3.67,1065 "learning_rate": 3.513770180436847e-05,1066 "loss": 0.2952,1067 "step": 17201068 },1069 {1070 "epoch": 3.7,1071 "learning_rate": 3.501899335232669e-05,1072 "loss": 0.3556,1073 "step": 17301074 },1075 {1076 "epoch": 3.72,1077 "learning_rate": 3.4900284900284904e-05,1078 "loss": 0.2956,1079 "step": 17401080 },1081 {1082 "epoch": 3.74,1083 "learning_rate": 3.4781576448243114e-05,1084 "loss": 0.3118,1085 "step": 17501086 },1087 {1088 "epoch": 3.76,1089 "learning_rate": 3.466286799620133e-05,1090 "loss": 0.3057,1091 "step": 17601092 },1093 {1094 "epoch": 3.78,1095 "learning_rate": 3.454415954415954e-05,1096 "loss": 0.2832,1097 "step": 17701098 },1099 {1100 "epoch": 3.8,1101 "learning_rate": 3.4425451092117764e-05,1102 "loss": 0.3224,1103 "step": 17801104 },1105 {1106 "epoch": 3.82,1107 "learning_rate": 3.4306742640075974e-05,1108 "loss": 0.3237,1109 "step": 17901110 },1111 {1112 "epoch": 3.84,1113 "learning_rate": 3.418803418803419e-05,1114 "loss": 0.2871,1115 "step": 18001116 },1117 {1118 "epoch": 3.87,1119 "learning_rate": 3.40693257359924e-05,1120 "loss": 0.3196,1121 "step": 18101122 },1123 {1124 "epoch": 3.89,1125 "learning_rate": 3.395061728395062e-05,1126 "loss": 0.3489,1127 "step": 18201128 },1129 {1130 "epoch": 3.91,1131 "learning_rate": 3.3831908831908834e-05,1132 "loss": 0.3108,1133 "step": 18301134 },1135 {1136 "epoch": 3.93,1137 "learning_rate": 3.371320037986705e-05,1138 "loss": 0.3285,1139 "step": 18401140 },1141 {1142 "epoch": 3.95,1143 "learning_rate": 3.359449192782526e-05,1144 "loss": 0.2986,1145 "step": 18501146 },1147 {1148 "epoch": 3.97,1149 "learning_rate": 3.347578347578348e-05,1150 "loss": 0.301,1151 "step": 18601152 },1153 {1154 "epoch": 3.99,1155 "learning_rate": 3.335707502374169e-05,1156 "loss": 0.2819,1157 "step": 18701158 },1159 {1160 "epoch": 4.0,1161 "eval_accuracy": 0.8852,1162 "eval_loss": 0.32173556089401245,1163 "eval_runtime": 47.4022,1164 "eval_samples_per_second": 210.961,1165 "eval_steps_per_second": 6.603,1166 "step": 18721167 },1168 {1169 "epoch": 4.02,1170 "learning_rate": 3.323836657169991e-05,1171 "loss": 0.309,1172 "step": 18801173 },1174 {1175 "epoch": 4.04,1176 "learning_rate": 3.311965811965812e-05,1177 "loss": 0.3257,1178 "step": 18901179 },1180 {1181 "epoch": 4.06,1182 "learning_rate": 3.300094966761634e-05,1183 "loss": 0.2827,1184 "step": 19001185 },1186 {1187 "epoch": 4.08,1188 "learning_rate": 3.288224121557455e-05,1189 "loss": 0.3234,1190 "step": 19101191 },1192 {1193 "epoch": 4.1,1194 "learning_rate": 3.2763532763532764e-05,1195 "loss": 0.3303,1196 "step": 19201197 },1198 {1199 "epoch": 4.12,1200 "learning_rate": 3.264482431149098e-05,