NbAiLab/vit-front-page-384-complete-v2
013
1{2 "best_metric": 0.14763057231903076,3 "best_model_checkpoint": "./vit-front-page-384-complete-v2/checkpoint-28000",4 "epoch": 63.99933554817276,5 "global_step": 30080,6 "is_hyper_param_search": false,7 "is_local_process_zero": true,8 "is_world_process_zero": true,9 "log_history": [10 {11 "epoch": 0.21,12 "learning_rate": 7e-08,13 "loss": 1.0485,14 "step": 10015 },16 {17 "epoch": 0.43,18 "learning_rate": 1.4e-07,19 "loss": 0.9833,20 "step": 20021 },22 {23 "epoch": 0.43,24 "eval_accuracy": 0.6194995827128573,25 "eval_f1": 0.2849637284042793,26 "eval_loss": 0.9414144158363342,27 "eval_matthews_correlation": -0.046627386105974944,28 "eval_precision": 0.28223649662424866,29 "eval_recall": 0.3194839105028776,30 "eval_runtime": 1120.0285,31 "eval_samples_per_second": 54.56,32 "eval_steps_per_second": 1.705,33 "step": 20034 },35 {36 "epoch": 0.64,37 "learning_rate": 2.0999999999999997e-07,38 "loss": 0.9031,39 "step": 30040 },41 {42 "epoch": 0.85,43 "learning_rate": 2.8e-07,44 "loss": 0.8378,45 "step": 40046 },47 {48 "epoch": 0.85,49 "eval_accuracy": 0.6742542015087793,50 "eval_f1": 0.27318592449369794,51 "eval_loss": 0.8087042570114136,52 "eval_matthews_correlation": -0.008566496106974774,53 "eval_precision": 0.33248655167577795,54 "eval_recall": 0.3329403574518804,55 "eval_runtime": 1146.8267,56 "eval_samples_per_second": 53.285,57 "eval_steps_per_second": 1.665,58 "step": 40059 },60 {61 "epoch": 1.06,62 "learning_rate": 3.5e-07,63 "loss": 0.7846,64 "step": 50065 },66 {67 "epoch": 1.28,68 "learning_rate": 4.1999999999999995e-07,69 "loss": 0.725,70 "step": 60071 },72 {73 "epoch": 1.28,74 "eval_accuracy": 0.7299906724050467,75 "eval_f1": 0.45416361692283386,76 "eval_loss": 0.6847068071365356,77 "eval_matthews_correlation": 0.3478679528690636,78 "eval_precision": 0.7356178375809365,79 "eval_recall": 0.4490403016153819,80 "eval_runtime": 1122.3931,81 "eval_samples_per_second": 54.445,82 "eval_steps_per_second": 1.702,83 "step": 60084 },85 {86 "epoch": 1.49,87 "learning_rate": 4.9e-07,88 "loss": 0.6452,89 "step": 70090 },91 {92 "epoch": 1.7,93 "learning_rate": 5.6e-07,94 "loss": 0.5769,95 "step": 80096 },97 {98 "epoch": 1.7,99 "eval_accuracy": 0.7984093995974406,100 "eval_f1": 0.6063804641618133,101 "eval_loss": 0.5433278679847717,102 "eval_matthews_correlation": 0.554774970478316,103 "eval_precision": 0.782372439805869,104 "eval_recall": 0.6092247400862097,105 "eval_runtime": 1136.2592,106 "eval_samples_per_second": 53.781,107 "eval_steps_per_second": 1.681,108 "step": 800109 },110 {111 "epoch": 1.91,112 "learning_rate": 6.3e-07,113 "loss": 0.5208,114 "step": 900115 },116 {117 "epoch": 2.13,118 "learning_rate": 7e-07,119 "loss": 0.4706,120 "step": 1000121 },122 {123 "epoch": 2.13,124 "eval_accuracy": 0.8339688098316124,125 "eval_f1": 0.6871318497564141,126 "eval_loss": 0.4425855875015259,127 "eval_matthews_correlation": 0.6440133138789277,128 "eval_precision": 0.8172733311232007,129 "eval_recall": 0.6865053904162712,130 "eval_runtime": 1131.2502,131 "eval_samples_per_second": 54.019,132 "eval_steps_per_second": 1.688,133 "step": 1000134 },135 {136 "epoch": 2.34,137 "learning_rate": 6.975928473177441e-07,138 "loss": 0.4247,139 "step": 1100140 },141 {142 "epoch": 2.55,143 "learning_rate": 6.951856946354883e-07,144 "loss": 0.3902,145 "step": 1200146 },147 {148 "epoch": 2.55,149 "eval_accuracy": 0.8571732478031059,150 "eval_f1": 0.7517270150716192,151 "eval_loss": 0.3730458915233612,152 "eval_matthews_correlation": 0.6959063906454223,153 "eval_precision": 0.8447385961080253,154 "eval_recall": 0.7349271721409717,155 "eval_runtime": 1143.842,156 "eval_samples_per_second": 53.424,157 "eval_steps_per_second": 1.67,158 "step": 1200159 },160 {161 "epoch": 2.77,162 "learning_rate": 6.927785419532324e-07,163 "loss": 0.3604,164 "step": 1300165 },166 {167 "epoch": 2.98,168 "learning_rate": 6.903713892709766e-07,169 "loss": 0.3392,170 "step": 1400171 },172 {173 "epoch": 2.98,174 "eval_accuracy": 0.8786430803973229,175 "eval_f1": 0.8069413230330182,176 "eval_loss": 0.32644152641296387,177 "eval_matthews_correlation": 0.7428749706793645,178 "eval_precision": 0.8632726130534206,179 "eval_recall": 0.7855049543193654,180 "eval_runtime": 1136.2051,181 "eval_samples_per_second": 53.783,182 "eval_steps_per_second": 1.681,183 "step": 1400184 },185 {186 "epoch": 3.19,187 "learning_rate": 6.879642365887207e-07,188 "loss": 0.3188,189 "step": 1500190 },191 {192 "epoch": 3.4,193 "learning_rate": 6.855570839064649e-07,194 "loss": 0.3044,195 "step": 1600196 },197 {198 "epoch": 3.4,199 "eval_accuracy": 0.8894925461061383,200 "eval_f1": 0.8329212820595306,201 "eval_loss": 0.2972641587257385,202 "eval_matthews_correlation": 0.7664388748302527,203 "eval_precision": 0.8723203701164862,204 "eval_recall": 0.8120687720492296,205 "eval_runtime": 1138.2797,206 "eval_samples_per_second": 53.685,207 "eval_steps_per_second": 1.678,208 "step": 1600209 },210 {211 "epoch": 3.62,212 "learning_rate": 6.83149931224209e-07,213 "loss": 0.2943,214 "step": 1700215 },216 {217 "epoch": 3.83,218 "learning_rate": 6.807427785419532e-07,219 "loss": 0.2795,220 "step": 1800221 },222 {223 "epoch": 3.83,224 "eval_accuracy": 0.8961691403884862,225 "eval_f1": 0.8432760143718706,226 "eval_loss": 0.2766903340816498,227 "eval_matthews_correlation": 0.781173318686851,228 "eval_precision": 0.8834852690618357,229 "eval_recall": 0.8221927104856026,230 "eval_runtime": 1109.6741,231 "eval_samples_per_second": 55.069,232 "eval_steps_per_second": 1.721,233 "step": 1800234 },235 {236 "epoch": 4.04,237 "learning_rate": 6.783356258596973e-07,238 "loss": 0.2738,239 "step": 1900240 },241 {242 "epoch": 4.26,243 "learning_rate": 6.759284731774416e-07,244 "loss": 0.262,245 "step": 2000246 },247 {248 "epoch": 4.26,249 "eval_accuracy": 0.9019784319821957,250 "eval_f1": 0.8550525638971442,251 "eval_loss": 0.26178261637687683,252 "eval_matthews_correlation": 0.7937491549035616,253 "eval_precision": 0.8893566012052573,254 "eval_recall": 0.8349244764526037,255 "eval_runtime": 1138.053,256 "eval_samples_per_second": 53.696,257 "eval_steps_per_second": 1.678,258 "step": 2000259 },260 {261 "epoch": 4.47,262 "learning_rate": 6.735213204951857e-07,263 "loss": 0.2567,264 "step": 2100265 },266 {267 "epoch": 4.68,268 "learning_rate": 6.711141678129298e-07,269 "loss": 0.2497,270 "step": 2200271 },272 {273 "epoch": 4.68,274 "eval_accuracy": 0.9061185750053183,275 "eval_f1": 0.8647877107942263,276 "eval_loss": 0.25028473138809204,277 "eval_matthews_correlation": 0.8031001323768835,278 "eval_precision": 0.888845641786976,279 "eval_recall": 0.8490665078134735,280 "eval_runtime": 1128.8955,281 "eval_samples_per_second": 54.132,282 "eval_steps_per_second": 1.692,283 "step": 2200284 },285 {286 "epoch": 4.89,287 "learning_rate": 6.68707015130674e-07,288 "loss": 0.2444,289 "step": 2300290 },291 {292 "epoch": 5.11,293 "learning_rate": 6.662998624484181e-07,294 "loss": 0.2414,295 "step": 2400296 },297 {298 "epoch": 5.11,299 "eval_accuracy": 0.908589569457854,300 "eval_f1": 0.8705010060852946,301 "eval_loss": 0.24158228933811188,302 "eval_matthews_correlation": 0.808821805927799,303 "eval_precision": 0.888714743861985,304 "eval_recall": 0.8577216170630284,305 "eval_runtime": 1121.8959,306 "eval_samples_per_second": 54.469,307 "eval_steps_per_second": 1.702,308 "step": 2400309 },310 {311 "epoch": 5.32,312 "learning_rate": 6.638927097661623e-07,313 "loss": 0.2321,314 "step": 2500315 },316 {317 "epoch": 5.53,318 "learning_rate": 6.614855570839064e-07,319 "loss": 0.2356,320 "step": 2600321 },322 {323 "epoch": 5.53,324 "eval_accuracy": 0.9110441997087172,325 "eval_f1": 0.8747151973222319,326 "eval_loss": 0.23398438096046448,327 "eval_matthews_correlation": 0.8139518481821442,328 "eval_precision": 0.8928668344520597,329 "eval_recall": 0.861416198724987,330 "eval_runtime": 1133.6213,331 "eval_samples_per_second": 53.906,332 "eval_steps_per_second": 1.685,333 "step": 2600334 },335 {336 "epoch": 5.74,337 "learning_rate": 6.590784044016506e-07,338 "loss": 0.226,339 "step": 2700340 },341 {342 "epoch": 5.96,343 "learning_rate": 6.566712517193947e-07,344 "loss": 0.2251,345 "step": 2800346 },347 {348 "epoch": 5.96,349 "eval_accuracy": 0.9136297435729598,350 "eval_f1": 0.8786518904171384,351 "eval_loss": 0.22776289284229279,352 "eval_matthews_correlation": 0.8196091308971525,353 "eval_precision": 0.8955149466291931,354 "eval_recall": 0.8663159550061658,355 "eval_runtime": 1128.8199,356 "eval_samples_per_second": 54.135,357 "eval_steps_per_second": 1.692,358 "step": 2800359 },360 {361 "epoch": 6.17,362 "learning_rate": 6.542640990371389e-07,363 "loss": 0.2216,364 "step": 2900365 },366 {367 "epoch": 6.38,368 "learning_rate": 6.51856946354883e-07,369 "loss": 0.217,370 "step": 3000371 },372 {373 "epoch": 6.38,374 "eval_accuracy": 0.915577083571978,375 "eval_f1": 0.8816198142871025,376 "eval_loss": 0.22265483438968658,377 "eval_matthews_correlation": 0.8236394377439027,378 "eval_precision": 0.8990606065527942,379 "eval_recall": 0.8686464677513945,380 "eval_runtime": 1127.8916,381 "eval_samples_per_second": 54.18,382 "eval_steps_per_second": 1.693,383 "step": 3000384 },385 {386 "epoch": 6.6,387 "learning_rate": 6.494497936726272e-07,388 "loss": 0.2175,389 "step": 3100390 },391 {392 "epoch": 6.81,393 "learning_rate": 6.470426409903714e-07,394 "loss": 0.2133,395 "step": 3200396 },397 {398 "epoch": 6.81,399 "eval_accuracy": 0.9171153185291856,400 "eval_f1": 0.8827723003585022,401 "eval_loss": 0.21786576509475708,402 "eval_matthews_correlation": 0.8266852106519758,403 "eval_precision": 0.9034666000752295,404 "eval_recall": 0.8679497941871164,405 "eval_runtime": 1129.6487,406 "eval_samples_per_second": 54.096,407 "eval_steps_per_second": 1.691,408 "step": 3200409 },410 {411 "epoch": 7.02,412 "learning_rate": 6.446354883081155e-07,413 "loss": 0.2117,414 "step": 3300415 },416 {417 "epoch": 7.23,418 "learning_rate": 6.422283356258597e-07,419 "loss": 0.2098,420 "step": 3400421 },422 {423 "epoch": 7.23,424 "eval_accuracy": 0.9182935410495999,425 "eval_f1": 0.884903972857821,426 "eval_loss": 0.21342970430850983,427 "eval_matthews_correlation": 0.8292988605784972,428 "eval_precision": 0.9040958524512837,429 "eval_recall": 0.870991653583178,430 "eval_runtime": 1152.1924,431 "eval_samples_per_second": 53.037,432 "eval_steps_per_second": 1.658,433 "step": 3400434 },435 {436 "epoch": 7.45,437 "learning_rate": 6.398211829436038e-07,438 "loss": 0.2021,439 "step": 3500440 },441 {442 "epoch": 7.66,443 "learning_rate": 6.37414030261348e-07,444 "loss": 0.2062,445 "step": 3600446 },447 {448 "epoch": 7.66,449 "eval_accuracy": 0.9201754242419283,450 "eval_f1": 0.8875922877918782,451 "eval_loss": 0.20977966487407684,452 "eval_matthews_correlation": 0.833123364179002,453 "eval_precision": 0.9084794166339746,454 "eval_recall": 0.872252852044762,455 "eval_runtime": 1135.8312,456 "eval_samples_per_second": 53.801,457 "eval_steps_per_second": 1.682,458 "step": 3600459 },460 {461 "epoch": 7.87,462 "learning_rate": 6.350068775790921e-07,463 "loss": 0.2033,464 "step": 3700465 },466 {467 "epoch": 8.09,468 "learning_rate": 6.325997248968363e-07,469 "loss": 0.1982,470 "step": 3800471 },472 {473 "epoch": 8.09,474 "eval_accuracy": 0.9213863751656876,475 "eval_f1": 0.88930772340783,476 "eval_loss": 0.20641230046749115,477 "eval_matthews_correlation": 0.8357746010338397,478 "eval_precision": 0.9095846117319936,479 "eval_recall": 0.8745329047658825,480 "eval_runtime": 1143.3313,481 "eval_samples_per_second": 53.448,482 "eval_steps_per_second": 1.671,483 "step": 3800484 },485 {486 "epoch": 8.3,487 "learning_rate": 6.301925722145805e-07,488 "loss": 0.196,489 "step": 3900490 },491 {492 "epoch": 8.51,493 "learning_rate": 6.277854195323246e-07,494 "loss": 0.1983,495 "step": 4000496 },497 {498 "epoch": 8.51,499 "eval_accuracy": 0.9222700420559983,500 "eval_f1": 0.8900453287552592,501 "eval_loss": 0.20386044681072235,502 "eval_matthews_correlation": 0.837555384879415,503 "eval_precision": 0.9123527733555781,504 "eval_recall": 0.8740593137023058,505 "eval_runtime": 1129.2304,506 "eval_samples_per_second": 54.116,507 "eval_steps_per_second": 1.691,508 "step": 4000509 },510 {511 "epoch": 8.72,512 "learning_rate": 6.253782668500687e-07,513 "loss": 0.1965,514 "step": 4100515 },516 {517 "epoch": 8.94,518 "learning_rate": 6.229711141678129e-07,519 "loss": 0.1936,520 "step": 4200521 },522 {523 "epoch": 8.94,524 "eval_accuracy": 0.9233828077697229,525 "eval_f1": 0.8932485448536257,526 "eval_loss": 0.20058760046958923,527 "eval_matthews_correlation": 0.8401209500114148,528 "eval_precision": 0.9107366418558472,529 "eval_recall": 0.8798434624360084,530 "eval_runtime": 1120.0565,531 "eval_samples_per_second": 54.559,532 "eval_steps_per_second": 1.705,533 "step": 4200534 },535 {536 "epoch": 9.15,537 "learning_rate": 6.205880330123796e-07,538 "loss": 0.1922,539 "step": 4300540 },541 {542 "epoch": 9.36,543 "learning_rate": 6.181808803301237e-07,544 "loss": 0.1953,545 "step": 4400546 },547 {548 "epoch": 9.36,549 "eval_accuracy": 0.9238246412148783,550 "eval_f1": 0.8951433768565843,551 "eval_loss": 0.19883869588375092,552 "eval_matthews_correlation": 0.8415676876590389,553 "eval_precision": 0.9072146230906633,554 "eval_recall": 0.885557435823166,555 "eval_runtime": 1137.9471,556 "eval_samples_per_second": 53.701,557 "eval_steps_per_second": 1.678,558 "step": 4400559 },560 {561 "epoch": 9.57,562 "learning_rate": 6.15773727647868e-07,563 "loss": 0.1843,564 "step": 4500565 },566 {567 "epoch": 9.79,568 "learning_rate": 6.133665749656121e-07,569 "loss": 0.1852,570 "step": 4600571 },572 {573 "epoch": 9.79,574 "eval_accuracy": 0.9253792403737584,575 "eval_f1": 0.8946482066160663,576 "eval_loss": 0.19628000259399414,577 "eval_matthews_correlation": 0.8441339816426117,578 "eval_precision": 0.9173420280370865,579 "eval_recall": 0.8782309033246154,580 "eval_runtime": 1124.1932,581 "eval_samples_per_second": 54.358,582 "eval_steps_per_second": 1.699,583 "step": 4600584 },585 {586 "epoch": 10.0,587 "learning_rate": 6.109594222833563e-07,588 "loss": 0.1845,589 "step": 4700590 },591 {592 "epoch": 10.21,593 "learning_rate": 6.085522696011004e-07,594 "loss": 0.1839,595 "step": 4800596 },597 {598 "epoch": 10.21,599 "eval_accuracy": 0.9258374380205862,600 "eval_f1": 0.8972498157769592,601 "eval_loss": 0.1933354139328003,602 "eval_matthews_correlation": 0.8454266170310101,603 "eval_precision": 0.9128401019975159,604 "eval_recall": 0.8849382250200719,605 "eval_runtime": 1122.9259,606 "eval_samples_per_second": 54.419,607 "eval_steps_per_second": 1.701,608 "step": 4800609 },610 {611 "epoch": 10.43,612 "learning_rate": 6.061451169188445e-07,613 "loss": 0.1845,614 "step": 4900615 },616 {617 "epoch": 10.64,618 "learning_rate": 6.037379642365887e-07,619 "loss": 0.1791,620 "step": 5000621 },622 {623 "epoch": 10.64,624 "eval_accuracy": 0.9267047407092245,625 "eval_f1": 0.8983700490176202,626 "eval_loss": 0.19171090424060822,627 "eval_matthews_correlation": 0.8472176589659597,628 "eval_precision": 0.9145474859208962,629 "eval_recall": 0.8856752741366947,630 "eval_runtime": 1134.0907,631 "eval_samples_per_second": 53.884,632 "eval_steps_per_second": 1.684,633 "step": 5000634 },635 {636 "epoch": 10.85,637 "learning_rate": 6.013308115543328e-07,638 "loss": 0.182,639 "step": 5100640 },641 {642 "epoch": 11.06,643 "learning_rate": 5.98923658872077e-07,644 "loss": 0.18,645 "step": 5200646 },647 {648 "epoch": 11.06,649 "eval_accuracy": 0.9271138457510351,650 "eval_f1": 0.8995050329011628,651 "eval_loss": 0.1895856410264969,652 "eval_matthews_correlation": 0.848294574832317,653 "eval_precision": 0.9131718148389393,654 "eval_recall": 0.8885987433408609,655 "eval_runtime": 1126.9856,656 "eval_samples_per_second": 54.223,657 "eval_steps_per_second": 1.695,658 "step": 5200659 },660 {661 "epoch": 11.28,662 "learning_rate": 5.965165061898212e-07,663 "loss": 0.1763,664 "step": 5300665 },666 {667 "epoch": 11.49,668 "learning_rate": 5.941093535075653e-07,669 "loss": 0.1782,670 "step": 5400671 },672 {673 "epoch": 11.49,674 "eval_accuracy": 0.9275556791961904,675 "eval_f1": 0.9002696190246419,676 "eval_loss": 0.18787473440170288,677 "eval_matthews_correlation": 0.84921482512736,678 "eval_precision": 0.9139059903894043,679 "eval_recall": 0.88930836639531,680 "eval_runtime": 1128.4943,681 "eval_samples_per_second": 54.151,682 "eval_steps_per_second": 1.693,683 "step": 5400684 },685 {686 "epoch": 11.7,687 "learning_rate": 5.917022008253094e-07,688 "loss": 0.1788,689 "step": 5500690 },691 {692 "epoch": 11.91,693 "learning_rate": 5.892950481430536e-07,694 "loss": 0.1712,695 "step": 5600696 },697 {698 "epoch": 11.91,699 "eval_accuracy": 0.9287175375149324,700 "eval_f1": 0.9011964058613507,701 "eval_loss": 0.1861330270767212,702 "eval_matthews_correlation": 0.8514678062818634,703 "eval_precision": 0.9173976756930257,704 "eval_recall": 0.8884787278098573,705 "eval_runtime": 1120.0822,706 "eval_samples_per_second": 54.558,707 "eval_steps_per_second": 1.705,708 "step": 5600709 },710 {711 "epoch": 12.13,712 "learning_rate": 5.868878954607978e-07,713 "loss": 0.172,714 "step": 5700715 },716 {717 "epoch": 12.34,718 "learning_rate": 5.84480742778542e-07,719 "loss": 0.1737,720 "step": 5800721 },722 {723 "epoch": 12.34,724 "eval_accuracy": 0.9296993896152776,725 "eval_f1": 0.9010626619953944,726 "eval_loss": 0.18573108315467834,727 "eval_matthews_correlation": 0.8533347114503049,728 "eval_precision": 0.9231173958750271,729 "eval_recall": 0.8849159091550628,730 "eval_runtime": 1156.0243,731 "eval_samples_per_second": 52.861,732 "eval_steps_per_second": 1.652,733 "step": 5800734 },735 {736 "epoch": 12.55,737 "learning_rate": 5.820735900962861e-07,738 "loss": 0.1726,739 "step": 5900740 },741 {742 "epoch": 12.77,743 "learning_rate": 5.796664374140302e-07,744 "loss": 0.1696,745 "step": 6000746 },747 {748 "epoch": 12.77,749 "eval_accuracy": 0.930026673648726,750 "eval_f1": 0.9031918377265686,751 "eval_loss": 0.1828816682100296,752 "eval_matthews_correlation": 0.8542309647721633,753 "eval_precision": 0.9192960205580971,754 "eval_recall": 0.8904958440603054,755 "eval_runtime": 1124.8361,756 "eval_samples_per_second": 54.327,757 "eval_steps_per_second": 1.698,758 "step": 6000759 },760 {761 "epoch": 12.98,762 "learning_rate": 5.772592847317744e-07,763 "loss": 0.1682,764 "step": 6100765 },766 {767 "epoch": 13.19,768 "learning_rate": 5.748521320495185e-07,769 "loss": 0.1698,770 "step": 6200771 },772 {773 "epoch": 13.19,774 "eval_accuracy": 0.9302885008754848,775 "eval_f1": 0.9048186507788922,776 "eval_loss": 0.18205700814723969,777 "eval_matthews_correlation": 0.8551184049776112,778 "eval_precision": 0.9164794554225942,779 "eval_recall": 0.8951361110635792,780 "eval_runtime": 1130.5739,781 "eval_samples_per_second": 54.051,782 "eval_steps_per_second": 1.689,783 "step": 6200784 },785 {786 "epoch": 13.4,787 "learning_rate": 5.724449793672627e-07,788 "loss": 0.1682,789 "step": 6300790 },791 {792 "epoch": 13.62,793 "learning_rate": 5.700378266850069e-07,794 "loss": 0.1667,795 "step": 6400796 },797 {798 "epoch": 13.62,799 "eval_accuracy": 0.9303048650771573,800 "eval_f1": 0.9056480109419961,801 "eval_loss": 0.18090908229351044,802 "eval_matthews_correlation": 0.855428359187355,803 "eval_precision": 0.9147426706946558,804 "eval_recall": 0.8978667208863417,805 "eval_runtime": 1113.9498,806 "eval_samples_per_second": 54.858,807 "eval_steps_per_second": 1.715,808 "step": 6400809 },810 {811 "epoch": 13.83,812 "learning_rate": 5.67630674002751e-07,813 "loss": 0.1673,814 "step": 6500815 },816 {817 "epoch": 14.04,818 "learning_rate": 5.652235213204951e-07,819 "loss": 0.1651,820 "step": 6600821 },822 {823 "epoch": 14.04,824 "eval_accuracy": 0.9307794269256574,825 "eval_f1": 0.90568642496187,826 "eval_loss": 0.17903146147727966,827 "eval_matthews_correlation": 0.8562905390552094,828 "eval_precision": 0.9160847482436384,829 "eval_recall": 0.8970366589757913,830 "eval_runtime": 1123.7851,831 "eval_samples_per_second": 54.378,832 "eval_steps_per_second": 1.7,833 "step": 6600834 },835 {836 "epoch": 14.26,837 "learning_rate": 5.628163686382393e-07,838 "loss": 0.1638,839 "step": 6700840 },841 {842 "epoch": 14.47,843 "learning_rate": 5.604092159559834e-07,844 "loss": 0.1614,845 "step": 6800846 },847 {848 "epoch": 14.47,849 "eval_accuracy": 0.9319412852443993,850 "eval_f1": 0.9062846542418327,851 "eval_loss": 0.17761486768722534,852 "eval_matthews_correlation": 0.8583632583058451,853 "eval_precision": 0.9210761772993599,854 "eval_recall": 0.8945216174285454,855 "eval_runtime": 1135.9567,856 "eval_samples_per_second": 53.795,857 "eval_steps_per_second": 1.681,858 "step": 6800859 },860 {861 "epoch": 14.68,862 "learning_rate": 5.580261348005502e-07,863 "loss": 0.1629,864 "step": 6900865 },866 {867 "epoch": 14.89,868 "learning_rate": 5.556189821182944e-07,869 "loss": 0.16,870 "step": 7000871 },872 {873 "epoch": 14.89,874 "eval_accuracy": 0.9323667544878823,875 "eval_f1": 0.9057483267194343,876 "eval_loss": 0.177241712808609,877 "eval_matthews_correlation": 0.8590673404186833,878 "eval_precision": 0.9249219537323391,879 "eval_recall": 0.8912560253961389,880 "eval_runtime": 1147.9689,881 "eval_samples_per_second": 53.232,882 "eval_steps_per_second": 1.664,883 "step": 7000884 },885 {886 "epoch": 15.11,887 "learning_rate": 5.532118294360385e-07,888 "loss": 0.1622,889 "step": 7100890 },891 {892 "epoch": 15.32,893 "learning_rate": 5.508046767537827e-07,894 "loss": 0.1591,895 "step": 7200896 },897 {898 "epoch": 15.32,899 "eval_accuracy": 0.9334467917982622,900 "eval_f1": 0.90852409435722,901 "eval_loss": 0.17550112307071686,902 "eval_matthews_correlation": 0.8615051567396046,903 "eval_precision": 0.9232173166653612,904 "eval_recall": 0.8966351633696954,905 "eval_runtime": 1144.6322,906 "eval_samples_per_second": 53.387,907 "eval_steps_per_second": 1.669,908 "step": 7200909 },910 {911 "epoch": 15.53,912 "learning_rate": 5.483975240715268e-07,913 "loss": 0.16,914 "step": 7300915 },916 {917 "epoch": 15.74,918 "learning_rate": 5.45990371389271e-07,919 "loss": 0.1601,920 "step": 7400921 },922 {923 "epoch": 15.74,924 "eval_accuracy": 0.9333322423865552,925 "eval_f1": 0.9088120224095171,926 "eval_loss": 0.17453713715076447,927 "eval_matthews_correlation": 0.8613787118816909,928 "eval_precision": 0.9218305477922321,929 "eval_recall": 0.8980828107442319,930 "eval_runtime": 1143.4792,931 "eval_samples_per_second": 53.441,932 "eval_steps_per_second": 1.67,933 "step": 7400934 },935 {936 "epoch": 15.96,937 "learning_rate": 5.435832187070151e-07,938 "loss": 0.1584,939 "step": 7500940 },941 {942 "epoch": 16.17,943 "learning_rate": 5.411760660247592e-07,944 "loss": 0.1558,945 "step": 7600946 },947 {948 "epoch": 16.17,949 "eval_accuracy": 0.9338395326384002,950 "eval_f1": 0.9096298761981343,951 "eval_loss": 0.17321471869945526,952 "eval_matthews_correlation": 0.8625748894777745,953 "eval_precision": 0.9213568402254939,954 "eval_recall": 0.8999913944640027,955 "eval_runtime": 1116.2047,956 "eval_samples_per_second": 54.747,957 "eval_steps_per_second": 1.711,958 "step": 7600959 },960 {961 "epoch": 16.38,962 "learning_rate": 5.387689133425034e-07,963 "loss": 0.1544,964 "step": 7700965 },966 {967 "epoch": 16.6,968 "learning_rate": 5.363617606602475e-07,969 "loss": 0.1552,970 "step": 7800971 },972 {973 "epoch": 16.6,974 "eval_accuracy": 0.9334631559999346,975 "eval_f1": 0.9101663960284817,976 "eval_loss": 0.17314866185188293,977 "eval_matthews_correlation": 0.8621017254091837,978 "eval_precision": 0.9185625631174649,979 "eval_recall": 0.9029180840503908,980 "eval_runtime": 1134.8054,981 "eval_samples_per_second": 53.85,982 "eval_steps_per_second": 1.683,983 "step": 7800984 },985 {986 "epoch": 16.81,987 "learning_rate": 5.339546079779917e-07,988 "loss": 0.1571,989 "step": 7900990 },991 {992 "epoch": 17.02,993 "learning_rate": 5.315474552957359e-07,994 "loss": 0.1526,995 "step": 8000996 },997 {998 "epoch": 17.02,999 "eval_accuracy": 0.9340849956634866,1000 "eval_f1": 0.9110344713781423,1001 "eval_loss": 0.17203205823898315,1002 "eval_matthews_correlation": 0.8634918058087261,1003 "eval_precision": 0.9187082054219985,1004 "eval_recall": 0.9044161914490357,1005 "eval_runtime": 1153.6876,1006 "eval_samples_per_second": 52.968,1007 "eval_steps_per_second": 1.656,1008 "step": 80001009 },1010 {1011 "epoch": 17.23,1012 "learning_rate": 5.2914030261348e-07,1013 "loss": 0.1485,1014 "step": 81001015 },1016 {1017 "epoch": 17.45,1018 "learning_rate": 5.267331499312242e-07,1019 "loss": 0.1532,1020 "step": 82001021 },1022 {1023 "epoch": 17.45,1024 "eval_accuracy": 0.9346741069236937,1025 "eval_f1": 0.9117218687868994,1026 "eval_loss": 0.17076529562473297,1027 "eval_matthews_correlation": 0.8646272156128906,1028 "eval_precision": 0.9201726149656478,1029 "eval_recall": 0.9044721170341236,1030 "eval_runtime": 1135.1604,1031 "eval_samples_per_second": 53.833,1032 "eval_steps_per_second": 1.683,1033 "step": 82001034 },1035 {1036 "epoch": 17.66,1037 "learning_rate": 5.243259972489684e-07,1038 "loss": 0.1551,1039 "step": 83001040 },1041 {1042 "epoch": 17.87,1043 "learning_rate": 5.219188445667125e-07,1044 "loss": 0.1539,1045 "step": 84001046 },1047 {1048 "epoch": 17.87,1049 "eval_accuracy": 0.9351159403688492,1050 "eval_f1": 0.911610683009681,1051 "eval_loss": 0.1695307195186615,1052 "eval_matthews_correlation": 0.865289074877206,1053 "eval_precision": 0.9226978459827032,1054 "eval_recall": 0.9023554489680277,1055 "eval_runtime": 1122.959,1056 "eval_samples_per_second": 54.418,1057 "eval_steps_per_second": 1.701,1058 "step": 84001059 },1060 {1061 "epoch": 18.09,1062 "learning_rate": 5.195116918844567e-07,1063 "loss": 0.1477,1064 "step": 85001065 },1066 {1067 "epoch": 18.3,1068 "learning_rate": 5.171045392022008e-07,1069 "loss": 0.1483,1070 "step": 86001071 },1072 {1073 "epoch": 18.3,1074 "eval_accuracy": 0.9356723232257115,1075 "eval_f1": 0.9116987398295661,1076 "eval_loss": 0.16896899044513702,1077 "eval_matthews_correlation": 0.8662695435076357,1078 "eval_precision": 0.9255532080817271,1079 "eval_recall": 0.9005543243181938,1080 "eval_runtime": 1135.8593,1081 "eval_samples_per_second": 53.8,1082 "eval_steps_per_second": 1.682,1083 "step": 86001084 },1085 {1086 "epoch": 18.51,1087 "learning_rate": 5.146973865199449e-07,1088 "loss": 0.1489,1089 "step": 87001090 },1091 {1092 "epoch": 18.72,1093 "learning_rate": 5.122902338376891e-07,1094 "loss": 0.1469,1095 "step": 88001096 },1097 {1098 "epoch": 18.72,1099 "eval_accuracy": 0.9358032368390908,1100 "eval_f1": 0.9135814294302561,1101 "eval_loss": 0.16858318448066711,1102 "eval_matthews_correlation": 0.8670812623876053,1103 "eval_precision": 0.9210624088966272,1104 "eval_recall": 0.907023142033475,1105 "eval_runtime": 1122.9881,1106 "eval_samples_per_second": 54.416,1107 "eval_steps_per_second": 1.701,1108 "step": 88001109 },1110 {1111 "epoch": 18.94,1112 "learning_rate": 5.099071526822558e-07,1113 "loss": 0.1507,1114 "step": 89001115 },1116 {1117 "epoch": 19.15,1118 "learning_rate": 5.075e-07,1119 "loss": 0.1461,1120 "step": 90001121 },1122 {1123 "epoch": 19.15,1124 "eval_accuracy": 0.9359177862507978,1125 "eval_f1": 0.9125886359052826,1126 "eval_loss": 0.16734813153743744,1127 "eval_matthews_correlation": 0.8668498402291797,1128 "eval_precision": 0.9249962574783411,1129 "eval_recall": 0.9022824765643582,1130 "eval_runtime": 1124.669,1131 "eval_samples_per_second": 54.335,1132 "eval_steps_per_second": 1.698,1133 "step": 90001134 },1135 {1136 "epoch": 19.36,1137 "learning_rate": 5.050928473177441e-07,1138 "loss": 0.1453,1139 "step": 91001140 },1141 {1142 "epoch": 19.57,1143 "learning_rate": 5.026856946354882e-07,1144 "loss": 0.1487,1145 "step": 92001146 },1147 {1148 "epoch": 19.57,1149 "eval_accuracy": 0.936588718519367,1150 "eval_f1": 0.9144201514784519,1151 "eval_loss": 0.16664335131645203,1152 "eval_matthews_correlation": 0.8685701549851123,1153 "eval_precision": 0.9231788402692221,1154 "eval_recall": 0.9068124874405709,1155 "eval_runtime": 1132.0219,1156 "eval_samples_per_second": 53.982,1157 "eval_steps_per_second": 1.687,1158 "step": 92001159 },1160 {1161 "epoch": 19.79,1162 "learning_rate": 5.00302613480055e-07,1163 "loss": 0.1499,1164 "step": 93001165 },1166 {1167 "epoch": 20.0,1168 "learning_rate": 4.978954607977991e-07,1169 "loss": 0.1433,1170 "step": 94001171 },1172 {1173 "epoch": 20.0,1174 "eval_accuracy": 0.9369978235611776,1175 "eval_f1": 0.91467275197079,1176 "eval_loss": 0.16577279567718506,1177 "eval_matthews_correlation": 0.8693728793586808,1178 "eval_precision": 0.924099137774855,1179 "eval_recall": 0.9066503727608389,1180 "eval_runtime": 1144.9998,1181 "eval_samples_per_second": 53.37,1182 "eval_steps_per_second": 1.668,1183 "step": 94001184 },1185 {1186 "epoch": 20.21,1187 "learning_rate": 4.954883081155432e-07,1188 "loss": 0.144,1189 "step": 95001190 },1191 {1192 "epoch": 20.43,1193 "learning_rate": 4.930811554332874e-07,1194 "loss": 0.1437,1195 "step": 96001196 },1197 {1198 "epoch": 20.43,1199 "eval_accuracy": 0.9370796445695396,1200 "eval_f1": 0.9143506341308217,