ASLP-lab/CN-MultiDialect-ASR
18203
1{2 "best_global_step": 12000,3 "best_metric": 0.23060274,4 "best_model_checkpoint": "/home/work_nfs23/sywang/code/ms-swift/examples/qwen3asr/exp3/exp_stage6_zh_en/v7-20260708-235534/checkpoint-12000",5 "epoch": 0.9931102973123953,6 "eval_steps": 500,7 "global_step": 16000,8 "is_hyper_param_search": false,9 "is_local_process_zero": true,10 "is_world_process_zero": true,11 "log_history": [12 {13 "epoch": 6.481931615621455e-05,14 "grad_norm": 2.655649185180664,15 "learning_rate": 1.295336787564767e-08,16 "loss": 0.3341776132583618,17 "step": 1,18 "token_acc": 0.911737684143830219 },20 {21 "epoch": 0.00032409658078107277,22 "grad_norm": 2.3714418411254883,23 "learning_rate": 6.476683937823834e-08,24 "loss": 0.32723408937454224,25 "step": 5,26 "token_acc": 0.913336488096209427 },28 {29 "epoch": 0.0006481931615621455,30 "grad_norm": 2.3853228092193604,31 "learning_rate": 1.2953367875647668e-07,32 "loss": 0.31995859146118166,33 "step": 10,34 "token_acc": 0.914780224707585535 },36 {37 "epoch": 0.0009722897423432182,38 "grad_norm": 2.6294362545013428,39 "learning_rate": 1.9430051813471504e-07,40 "loss": 0.3260782241821289,41 "step": 15,42 "token_acc": 0.913507559565177543 },44 {45 "epoch": 0.001296386323124291,46 "grad_norm": 2.2157599925994873,47 "learning_rate": 2.5906735751295336e-07,48 "loss": 0.3227314233779907,49 "step": 20,50 "token_acc": 0.914481796876513251 },52 {53 "epoch": 0.0016204829039053639,54 "grad_norm": 2.3223657608032227,55 "learning_rate": 3.238341968911917e-07,56 "loss": 0.3281126022338867,57 "step": 25,58 "token_acc": 0.913487614080834559 },60 {61 "epoch": 0.0019445794846864365,62 "grad_norm": 1.8758041858673096,63 "learning_rate": 3.886010362694301e-07,64 "loss": 0.3164376258850098,65 "step": 30,66 "token_acc": 0.915510251812582867 },68 {69 "epoch": 0.0022686760654675093,70 "grad_norm": 1.8605087995529175,71 "learning_rate": 4.5336787564766845e-07,72 "loss": 0.3189058303833008,73 "step": 35,74 "token_acc": 0.915336589784233275 },76 {77 "epoch": 0.002592772646248582,78 "grad_norm": 1.5158028602600098,79 "learning_rate": 5.181347150259067e-07,80 "loss": 0.31297638416290285,81 "step": 40,82 "token_acc": 0.91654775634134183 },84 {85 "epoch": 0.002916869227029655,86 "grad_norm": 1.369957685470581,87 "learning_rate": 5.829015544041451e-07,88 "loss": 0.3067481994628906,89 "step": 45,90 "token_acc": 0.91667262434763491 },92 {93 "epoch": 0.0032409658078107278,94 "grad_norm": 1.3565272092819214,95 "learning_rate": 6.476683937823834e-07,96 "loss": 0.30818629264831543,97 "step": 50,98 "token_acc": 0.917218412288899799 },100 {101 "epoch": 0.0035650623885918,102 "grad_norm": 1.7045643329620361,103 "learning_rate": 7.124352331606219e-07,104 "loss": 0.31780500411987306,105 "step": 55,106 "token_acc": 0.9151205018157808107 },108 {109 "epoch": 0.003889158969372873,110 "grad_norm": 1.3580641746520996,111 "learning_rate": 7.772020725388602e-07,112 "loss": 0.30206310749053955,113 "step": 60,114 "token_acc": 0.9192551764849309115 },116 {117 "epoch": 0.004213255550153946,118 "grad_norm": 1.372952938079834,119 "learning_rate": 8.419689119170985e-07,120 "loss": 0.3010685443878174,121 "step": 65,122 "token_acc": 0.9178735707714548123 },124 {125 "epoch": 0.004537352130935019,126 "grad_norm": 1.3031820058822632,127 "learning_rate": 9.067357512953369e-07,128 "loss": 0.29661927223205564,129 "step": 70,130 "token_acc": 0.9191524095598799131 },132 {133 "epoch": 0.004861448711716091,134 "grad_norm": 1.260411024093628,135 "learning_rate": 9.715025906735752e-07,136 "loss": 0.3009307861328125,137 "step": 75,138 "token_acc": 0.9196397599272189139 },140 {141 "epoch": 0.005185545292497164,142 "grad_norm": 1.2549397945404053,143 "learning_rate": 1.0362694300518134e-06,144 "loss": 0.2917083740234375,145 "step": 80,146 "token_acc": 0.9206776871569922147 },148 {149 "epoch": 0.005509641873278237,150 "grad_norm": 1.2358394861221313,151 "learning_rate": 1.1010362694300518e-06,152 "loss": 0.29244284629821776,153 "step": 85,154 "token_acc": 0.9203724671065474155 },156 {157 "epoch": 0.00583373845405931,158 "grad_norm": 1.2670187950134277,159 "learning_rate": 1.1658031088082903e-06,160 "loss": 0.29021873474121096,161 "step": 90,162 "token_acc": 0.9205806383595583163 },164 {165 "epoch": 0.006157835034840383,166 "grad_norm": 1.128826379776001,167 "learning_rate": 1.2305699481865285e-06,168 "loss": 0.2890714168548584,169 "step": 95,170 "token_acc": 0.9221680580499583171 },172 {173 "epoch": 0.0064819316156214555,174 "grad_norm": 1.2712926864624023,175 "learning_rate": 1.2953367875647669e-06,176 "loss": 0.288629150390625,177 "step": 100,178 "token_acc": 0.9215318116919947179 },180 {181 "epoch": 0.006806028196402528,182 "grad_norm": 1.3695846796035767,183 "learning_rate": 1.3601036269430053e-06,184 "loss": 0.2901022911071777,185 "step": 105,186 "token_acc": 0.9207788849772262187 },188 {189 "epoch": 0.0071301247771836,190 "grad_norm": 1.181462049484253,191 "learning_rate": 1.4248704663212437e-06,192 "loss": 0.2921628475189209,193 "step": 110,194 "token_acc": 0.9201108559458211195 },196 {197 "epoch": 0.007454221357964673,198 "grad_norm": 1.1615545749664307,199 "learning_rate": 1.489637305699482e-06,200 "loss": 0.27958681583404543,201 "step": 115,202 "token_acc": 0.9226613535406324203 },204 {205 "epoch": 0.007778317938745746,206 "grad_norm": 1.1601577997207642,207 "learning_rate": 1.5544041450777204e-06,208 "loss": 0.2812881231307983,209 "step": 120,210 "token_acc": 0.9219507427677873211 },212 {213 "epoch": 0.00810241451952682,214 "grad_norm": 1.1718589067459106,215 "learning_rate": 1.6191709844559586e-06,216 "loss": 0.28042919635772706,217 "step": 125,218 "token_acc": 0.922784509384652219 },220 {221 "epoch": 0.008426511100307892,222 "grad_norm": 1.1261916160583496,223 "learning_rate": 1.683937823834197e-06,224 "loss": 0.28291802406311034,225 "step": 130,226 "token_acc": 0.9216872217187438227 },228 {229 "epoch": 0.008750607681088965,230 "grad_norm": 1.2394858598709106,231 "learning_rate": 1.7487046632124354e-06,232 "loss": 0.2855110168457031,233 "step": 135,234 "token_acc": 0.9219139435628654235 },236 {237 "epoch": 0.009074704261870037,238 "grad_norm": 1.1633332967758179,239 "learning_rate": 1.8134715025906738e-06,240 "loss": 0.2800710916519165,241 "step": 140,242 "token_acc": 0.9219525269495525243 },244 {245 "epoch": 0.00939880084265111,246 "grad_norm": 1.0819878578186035,247 "learning_rate": 1.878238341968912e-06,248 "loss": 0.2783178806304932,249 "step": 145,250 "token_acc": 0.9228247096782621251 },252 {253 "epoch": 0.009722897423432183,254 "grad_norm": 1.2350913286209106,255 "learning_rate": 1.9430051813471504e-06,256 "loss": 0.2704397439956665,257 "step": 150,258 "token_acc": 0.9243719866405021259 },260 {261 "epoch": 0.010046994004213255,262 "grad_norm": 1.2952532768249512,263 "learning_rate": 2.0077720207253886e-06,264 "loss": 0.2756261110305786,265 "step": 155,266 "token_acc": 0.923563500377939267 },268 {269 "epoch": 0.010371090584994328,270 "grad_norm": 2.835852861404419,271 "learning_rate": 2.072538860103627e-06,272 "loss": 0.27072920799255373,273 "step": 160,274 "token_acc": 0.9240289571712448275 },276 {277 "epoch": 0.0106951871657754,278 "grad_norm": 1.2432365417480469,279 "learning_rate": 2.1373056994818655e-06,280 "loss": 0.28088808059692383,281 "step": 165,282 "token_acc": 0.9218431443392088283 },284 {285 "epoch": 0.011019283746556474,286 "grad_norm": 1.1758121252059937,287 "learning_rate": 2.2020725388601037e-06,288 "loss": 0.2745814323425293,289 "step": 170,290 "token_acc": 0.923520129422373291 },292 {293 "epoch": 0.011343380327337546,294 "grad_norm": 1.1752557754516602,295 "learning_rate": 2.2668393782383423e-06,296 "loss": 0.284597110748291,297 "step": 175,298 "token_acc": 0.921214273500901299 },300 {301 "epoch": 0.01166747690811862,302 "grad_norm": 1.1469591856002808,303 "learning_rate": 2.3316062176165805e-06,304 "loss": 0.27476656436920166,305 "step": 180,306 "token_acc": 0.9236426183753611307 },308 {309 "epoch": 0.011991573488899692,310 "grad_norm": 1.1166425943374634,311 "learning_rate": 2.3963730569948187e-06,312 "loss": 0.2817973613739014,313 "step": 185,314 "token_acc": 0.9223100209377249315 },316 {317 "epoch": 0.012315670069680765,318 "grad_norm": 1.1793495416641235,319 "learning_rate": 2.461139896373057e-06,320 "loss": 0.27724857330322267,321 "step": 190,322 "token_acc": 0.9227060065421657323 },324 {325 "epoch": 0.012639766650461837,326 "grad_norm": 1.1739124059677124,327 "learning_rate": 2.5259067357512956e-06,328 "loss": 0.28834104537963867,329 "step": 195,330 "token_acc": 0.9206041147010362331 },332 {333 "epoch": 0.012963863231242911,334 "grad_norm": 1.135248064994812,335 "learning_rate": 2.5906735751295338e-06,336 "loss": 0.27553977966308596,337 "step": 200,338 "token_acc": 0.9237138096229952339 },340 {341 "epoch": 0.013287959812023983,342 "grad_norm": 1.1080269813537598,343 "learning_rate": 2.6554404145077724e-06,344 "loss": 0.2846039295196533,345 "step": 205,346 "token_acc": 0.9222145902722878347 },348 {349 "epoch": 0.013612056392805057,350 "grad_norm": 1.1565662622451782,351 "learning_rate": 2.7202072538860106e-06,352 "loss": 0.2811276912689209,353 "step": 210,354 "token_acc": 0.9220826230889215355 },356 {357 "epoch": 0.013936152973586129,358 "grad_norm": 1.1716532707214355,359 "learning_rate": 2.784974093264249e-06,360 "loss": 0.28411569595336916,361 "step": 215,362 "token_acc": 0.9210638619159232363 },364 {365 "epoch": 0.0142602495543672,366 "grad_norm": 1.2090009450912476,367 "learning_rate": 2.8497409326424875e-06,368 "loss": 0.28354754447937014,369 "step": 220,370 "token_acc": 0.9223695272586173371 },372 {373 "epoch": 0.014584346135148274,374 "grad_norm": 1.1573808193206787,375 "learning_rate": 2.9145077720207252e-06,376 "loss": 0.27943172454833987,377 "step": 225,378 "token_acc": 0.922367671423052379 },380 {381 "epoch": 0.014908442715929346,382 "grad_norm": 1.1708178520202637,383 "learning_rate": 2.979274611398964e-06,384 "loss": 0.26848132610321046,385 "step": 230,386 "token_acc": 0.9253435183804218387 },388 {389 "epoch": 0.01523253929671042,390 "grad_norm": 1.145649790763855,391 "learning_rate": 3.0440414507772025e-06,392 "loss": 0.26862215995788574,393 "step": 235,394 "token_acc": 0.9247367940405213395 },396 {397 "epoch": 0.015556635877491492,398 "grad_norm": 1.1238332986831665,399 "learning_rate": 3.1088082901554407e-06,400 "loss": 0.27870104312896726,401 "step": 240,402 "token_acc": 0.9225761980010014403 },404 {405 "epoch": 0.015880732458272564,406 "grad_norm": 1.142770528793335,407 "learning_rate": 3.1735751295336793e-06,408 "loss": 0.271391487121582,409 "step": 245,410 "token_acc": 0.9252171658239337411 },412 {413 "epoch": 0.01620482903905364,414 "grad_norm": 1.0345611572265625,415 "learning_rate": 3.238341968911917e-06,416 "loss": 0.2781627655029297,417 "step": 250,418 "token_acc": 0.9220097608841605419 },420 {421 "epoch": 0.01652892561983471,422 "grad_norm": 1.0694793462753296,423 "learning_rate": 3.3031088082901557e-06,424 "loss": 0.26626498699188234,425 "step": 255,426 "token_acc": 0.9258351506262423427 },428 {429 "epoch": 0.016853022200615783,430 "grad_norm": 1.2326105833053589,431 "learning_rate": 3.367875647668394e-06,432 "loss": 0.2813969612121582,433 "step": 260,434 "token_acc": 0.9220959919142596435 },436 {437 "epoch": 0.017177118781396855,438 "grad_norm": 1.1490870714187622,439 "learning_rate": 3.4326424870466326e-06,440 "loss": 0.27063512802124023,441 "step": 265,442 "token_acc": 0.925166912735017443 },444 {445 "epoch": 0.01750121536217793,446 "grad_norm": 1.1896758079528809,447 "learning_rate": 3.497409326424871e-06,448 "loss": 0.27055246829986573,449 "step": 270,450 "token_acc": 0.9244674824013597451 },452 {453 "epoch": 0.017825311942959002,454 "grad_norm": 1.175060510635376,455 "learning_rate": 3.562176165803109e-06,456 "loss": 0.27529966831207275,457 "step": 275,458 "token_acc": 0.9235797078850252459 },460 {461 "epoch": 0.018149408523740074,462 "grad_norm": 1.1656664609909058,463 "learning_rate": 3.6269430051813476e-06,464 "loss": 0.2720619201660156,465 "step": 280,466 "token_acc": 0.924282242288921467 },468 {469 "epoch": 0.018473505104521146,470 "grad_norm": 1.2035537958145142,471 "learning_rate": 3.6917098445595854e-06,472 "loss": 0.27737472057342527,473 "step": 285,474 "token_acc": 0.9232464193903782475 },476 {477 "epoch": 0.01879760168530222,478 "grad_norm": 1.0576199293136597,479 "learning_rate": 3.756476683937824e-06,480 "loss": 0.2786442756652832,481 "step": 290,482 "token_acc": 0.922734228213919483 },484 {485 "epoch": 0.019121698266083294,486 "grad_norm": 1.101805567741394,487 "learning_rate": 3.821243523316063e-06,488 "loss": 0.26576249599456786,489 "step": 295,490 "token_acc": 0.9256528679087294491 },492 {493 "epoch": 0.019445794846864366,494 "grad_norm": 1.164818525314331,495 "learning_rate": 3.886010362694301e-06,496 "loss": 0.27654869556427003,497 "step": 300,498 "token_acc": 0.9231146575969651499 },500 {501 "epoch": 0.019769891427645438,502 "grad_norm": 1.1586562395095825,503 "learning_rate": 3.9507772020725395e-06,504 "loss": 0.276000452041626,505 "step": 305,506 "token_acc": 0.9234045303260413507 },508 {509 "epoch": 0.02009398800842651,510 "grad_norm": 1.2469451427459717,511 "learning_rate": 4.015544041450777e-06,512 "loss": 0.28005189895629884,513 "step": 310,514 "token_acc": 0.9224248584478456515 },516 {517 "epoch": 0.020418084589207585,518 "grad_norm": 1.088731050491333,519 "learning_rate": 4.080310880829016e-06,520 "loss": 0.27149338722229005,521 "step": 315,522 "token_acc": 0.9231371230103549523 },524 {525 "epoch": 0.020742181169988657,526 "grad_norm": 1.0491572618484497,527 "learning_rate": 4.145077720207254e-06,528 "loss": 0.26023097038269044,529 "step": 320,530 "token_acc": 0.9270244697348017531 },532 {533 "epoch": 0.02106627775076973,534 "grad_norm": 1.1924079656600952,535 "learning_rate": 4.209844559585492e-06,536 "loss": 0.27563767433166503,537 "step": 325,538 "token_acc": 0.922792435810702539 },540 {541 "epoch": 0.0213903743315508,542 "grad_norm": 1.0052635669708252,543 "learning_rate": 4.274611398963731e-06,544 "loss": 0.2668116569519043,545 "step": 330,546 "token_acc": 0.9254766824471007547 },548 {549 "epoch": 0.021714470912331876,550 "grad_norm": 1.1143276691436768,551 "learning_rate": 4.339378238341969e-06,552 "loss": 0.27575368881225587,553 "step": 335,554 "token_acc": 0.9234813273435626555 },556 {557 "epoch": 0.02203856749311295,558 "grad_norm": 1.1198474168777466,559 "learning_rate": 4.404145077720207e-06,560 "loss": 0.2739673137664795,561 "step": 340,562 "token_acc": 0.9234454717323182563 },564 {565 "epoch": 0.02236266407389402,566 "grad_norm": 1.1597661972045898,567 "learning_rate": 4.468911917098446e-06,568 "loss": 0.2812596559524536,569 "step": 345,570 "token_acc": 0.9220368755769822571 },572 {573 "epoch": 0.022686760654675092,574 "grad_norm": 1.1597846746444702,575 "learning_rate": 4.533678756476685e-06,576 "loss": 0.2782421112060547,577 "step": 350,578 "token_acc": 0.9231371002202281579 },580 {581 "epoch": 0.023010857235456168,582 "grad_norm": 1.1377077102661133,583 "learning_rate": 4.598445595854923e-06,584 "loss": 0.27525331974029543,585 "step": 355,586 "token_acc": 0.9234653879321608587 },588 {589 "epoch": 0.02333495381623724,590 "grad_norm": 1.1639304161071777,591 "learning_rate": 4.663212435233161e-06,592 "loss": 0.27627902030944823,593 "step": 360,594 "token_acc": 0.9239800719987891595 },596 {597 "epoch": 0.02365905039701831,598 "grad_norm": 1.1461061239242554,599 "learning_rate": 4.7279792746114e-06,600 "loss": 0.26558914184570315,601 "step": 365,602 "token_acc": 0.9252026481090521603 },604 {605 "epoch": 0.023983146977799383,606 "grad_norm": 1.254964828491211,607 "learning_rate": 4.7927461139896375e-06,608 "loss": 0.27816622257232665,609 "step": 370,610 "token_acc": 0.9229658348728891611 },612 {613 "epoch": 0.024307243558580455,614 "grad_norm": 1.1927502155303955,615 "learning_rate": 4.857512953367876e-06,616 "loss": 0.26456561088562014,617 "step": 375,618 "token_acc": 0.9256379821958457619 },620 {621 "epoch": 0.02463134013936153,622 "grad_norm": 1.0993506908416748,623 "learning_rate": 4.922279792746114e-06,624 "loss": 0.26846771240234374,625 "step": 380,626 "token_acc": 0.9255846499947682627 },628 {629 "epoch": 0.024955436720142603,630 "grad_norm": 1.1774364709854126,631 "learning_rate": 4.9870466321243525e-06,632 "loss": 0.2800713539123535,633 "step": 385,634 "token_acc": 0.923156113937995635 },636 {637 "epoch": 0.025279533300923675,638 "grad_norm": 1.2764763832092285,639 "learning_rate": 5.051813471502591e-06,640 "loss": 0.270487642288208,641 "step": 390,642 "token_acc": 0.9245638849599246643 },644 {645 "epoch": 0.025603629881704747,646 "grad_norm": 1.1404081583023071,647 "learning_rate": 5.11658031088083e-06,648 "loss": 0.27359600067138673,649 "step": 395,650 "token_acc": 0.9244739005075082651 },652 {653 "epoch": 0.025927726462485822,654 "grad_norm": 1.1143617630004883,655 "learning_rate": 5.1813471502590676e-06,656 "loss": 0.27301363945007323,657 "step": 400,658 "token_acc": 0.9238943472935289659 },660 {661 "epoch": 0.026251823043266894,662 "grad_norm": 1.5213898420333862,663 "learning_rate": 5.246113989637306e-06,664 "loss": 0.27214756011962893,665 "step": 405,666 "token_acc": 0.9242776311816523667 },668 {669 "epoch": 0.026575919624047966,670 "grad_norm": 1.1557198762893677,671 "learning_rate": 5.310880829015545e-06,672 "loss": 0.2753096103668213,673 "step": 410,674 "token_acc": 0.923890271693801675 },676 {677 "epoch": 0.026900016204829038,678 "grad_norm": 1.197363018989563,679 "learning_rate": 5.3756476683937834e-06,680 "loss": 0.27670223712921144,681 "step": 415,682 "token_acc": 0.9229302265035148683 },684 {685 "epoch": 0.027224112785610113,686 "grad_norm": 1.1271995306015015,687 "learning_rate": 5.440414507772021e-06,688 "loss": 0.2729542016983032,689 "step": 420,690 "token_acc": 0.9231763534323003691 },692 {693 "epoch": 0.027548209366391185,694 "grad_norm": 1.1053199768066406,695 "learning_rate": 5.505181347150259e-06,696 "loss": 0.27408714294433595,697 "step": 425,698 "token_acc": 0.923493480330874699 },700 {701 "epoch": 0.027872305947172257,702 "grad_norm": 1.1583583354949951,703 "learning_rate": 5.569948186528498e-06,704 "loss": 0.2778322696685791,705 "step": 430,706 "token_acc": 0.9223960251465049707 },708 {709 "epoch": 0.02819640252795333,710 "grad_norm": 1.1535427570343018,711 "learning_rate": 5.634715025906736e-06,712 "loss": 0.27780320644378664,713 "step": 435,714 "token_acc": 0.9218625346193012715 },716 {717 "epoch": 0.0285204991087344,718 "grad_norm": 1.1105202436447144,719 "learning_rate": 5.699481865284975e-06,720 "loss": 0.27113909721374513,721 "step": 440,722 "token_acc": 0.9243662988143548723 },724 {725 "epoch": 0.028844595689515477,726 "grad_norm": 1.1289681196212769,727 "learning_rate": 5.7642487046632135e-06,728 "loss": 0.2736364841461182,729 "step": 445,730 "token_acc": 0.9236354858492413731 },732 {733 "epoch": 0.02916869227029655,734 "grad_norm": 1.119524359703064,735 "learning_rate": 5.8290155440414505e-06,736 "loss": 0.2711230754852295,737 "step": 450,738 "token_acc": 0.9240434261931405739 },740 {741 "epoch": 0.02949278885107762,742 "grad_norm": 1.166691780090332,743 "learning_rate": 5.893782383419689e-06,744 "loss": 0.27475461959838865,745 "step": 455,746 "token_acc": 0.923216417669646747 },748 {749 "epoch": 0.029816885431858692,750 "grad_norm": 1.1108464002609253,751 "learning_rate": 5.958549222797928e-06,752 "loss": 0.27015066146850586,753 "step": 460,754 "token_acc": 0.9245998342566673755 },756 {757 "epoch": 0.030140982012639768,758 "grad_norm": 1.3036308288574219,759 "learning_rate": 6.023316062176166e-06,760 "loss": 0.27171549797058103,761 "step": 465,762 "token_acc": 0.9233259233259233763 },764 {765 "epoch": 0.03046507859342084,766 "grad_norm": 1.4210838079452515,767 "learning_rate": 6.088082901554405e-06,768 "loss": 0.27597966194152834,769 "step": 470,770 "token_acc": 0.9232971630278468771 },772 {773 "epoch": 0.030789175174201912,774 "grad_norm": 1.2602447271347046,775 "learning_rate": 6.152849740932643e-06,776 "loss": 0.28009696006774903,777 "step": 475,778 "token_acc": 0.921947459863359779 },780 {781 "epoch": 0.031113271754982984,782 "grad_norm": 1.1573154926300049,783 "learning_rate": 6.217616580310881e-06,784 "loss": 0.27212891578674314,785 "step": 480,786 "token_acc": 0.9239911824538681787 },788 {789 "epoch": 0.031437368335764056,790 "grad_norm": 1.5468491315841675,791 "learning_rate": 6.28238341968912e-06,792 "loss": 0.26889379024505616,793 "step": 485,794 "token_acc": 0.9242921558007661795 },796 {797 "epoch": 0.03176146491654513,798 "grad_norm": 1.0774251222610474,799 "learning_rate": 6.347150259067359e-06,800 "loss": 0.27394418716430663,801 "step": 490,802 "token_acc": 0.9239502943100065803 },804 {805 "epoch": 0.03208556149732621,806 "grad_norm": 1.0900217294692993,807 "learning_rate": 6.411917098445596e-06,808 "loss": 0.2677357912063599,809 "step": 495,810 "token_acc": 0.9250740075064756811 },812 {813 "epoch": 0.03240965807810728,814 "grad_norm": 1.1952015161514282,815 "learning_rate": 6.476683937823834e-06,816 "loss": 0.2757922649383545,817 "step": 500,818 "token_acc": 0.9237246194149515819 },820 {821 "epoch": 0.03240965807810728,822 "eval_loss": 0.2766605019569397,823 "eval_runtime": 118.2624,824 "eval_samples_per_second": 169.116,825 "eval_steps_per_second": 0.228,826 "eval_token_acc": 0.9228264253046955,827 "step": 500828 },829 {830 "epoch": 0.03273375465888835,831 "grad_norm": 1.072628378868103,832 "learning_rate": 6.541450777202073e-06,833 "loss": 0.27219414710998535,834 "step": 505,835 "token_acc": 0.923241036144262836 },837 {838 "epoch": 0.03305785123966942,839 "grad_norm": 1.123671531677246,840 "learning_rate": 6.6062176165803115e-06,841 "loss": 0.27762365341186523,842 "step": 510,843 "token_acc": 0.9229576107284878844 },845 {846 "epoch": 0.033381947820450494,847 "grad_norm": 1.149427890777588,848 "learning_rate": 6.67098445595855e-06,849 "loss": 0.277001690864563,850 "step": 515,851 "token_acc": 0.9234441979821897852 },853 {854 "epoch": 0.033706044401231566,855 "grad_norm": 1.1308335065841675,856 "learning_rate": 6.735751295336788e-06,857 "loss": 0.27818913459777833,858 "step": 520,859 "token_acc": 0.9231610048227251860 },861 {862 "epoch": 0.03403014098201264,863 "grad_norm": 1.2418826818466187,864 "learning_rate": 6.8005181347150265e-06,865 "loss": 0.27416648864746096,866 "step": 525,867 "token_acc": 0.9227283971513013868 },869 {870 "epoch": 0.03435423756279371,871 "grad_norm": 1.12945556640625,872 "learning_rate": 6.865284974093265e-06,873 "loss": 0.2727360248565674,874 "step": 530,875 "token_acc": 0.9247170911664296876 },877 {878 "epoch": 0.03467833414357478,879 "grad_norm": 1.153743028640747,880 "learning_rate": 6.930051813471503e-06,881 "loss": 0.2731607913970947,882 "step": 535,883 "token_acc": 0.9244839180524422884 },885 {886 "epoch": 0.03500243072435586,887 "grad_norm": 1.1583285331726074,888 "learning_rate": 6.994818652849742e-06,889 "loss": 0.27802448272705077,890 "step": 540,891 "token_acc": 0.922633087074326892 },893 {894 "epoch": 0.03532652730513693,895 "grad_norm": 1.2018134593963623,896 "learning_rate": 7.059585492227979e-06,897 "loss": 0.2711876392364502,898 "step": 545,899 "token_acc": 0.9240181425935071900 },901 {902 "epoch": 0.035650623885918005,903 "grad_norm": 1.1137855052947998,904 "learning_rate": 7.124352331606218e-06,905 "loss": 0.2710531711578369,906 "step": 550,907 "token_acc": 0.9236695779893513908 },909 {910 "epoch": 0.03597472046669908,911 "grad_norm": 1.1690529584884644,912 "learning_rate": 7.189119170984457e-06,913 "loss": 0.27385501861572265,914 "step": 555,915 "token_acc": 0.9241576534326812916 },917 {918 "epoch": 0.03629881704748015,919 "grad_norm": 1.1739192008972168,920 "learning_rate": 7.253886010362695e-06,921 "loss": 0.2713189601898193,922 "step": 560,923 "token_acc": 0.9240261137814765924 },925 {926 "epoch": 0.03662291362826122,927 "grad_norm": 1.574720859527588,928 "learning_rate": 7.318652849740934e-06,929 "loss": 0.27800450325012205,930 "step": 565,931 "token_acc": 0.9226385003529504932 },933 {934 "epoch": 0.03694701020904229,935 "grad_norm": 1.0243511199951172,936 "learning_rate": 7.383419689119171e-06,937 "loss": 0.27382216453552244,938 "step": 570,939 "token_acc": 0.9235590504565607940 },941 {942 "epoch": 0.037271106789823365,943 "grad_norm": 1.138895869255066,944 "learning_rate": 7.4481865284974095e-06,945 "loss": 0.2831470012664795,946 "step": 575,947 "token_acc": 0.9223049276273151948 },949 {950 "epoch": 0.03759520337060444,951 "grad_norm": 1.067686676979065,952 "learning_rate": 7.512953367875648e-06,953 "loss": 0.27533273696899413,954 "step": 580,955 "token_acc": 0.9240167802831672956 },957 {958 "epoch": 0.037919299951385516,959 "grad_norm": 1.2463687658309937,960 "learning_rate": 7.577720207253887e-06,961 "loss": 0.2789109230041504,962 "step": 585,963 "token_acc": 0.9221528491266894964 },965 {966 "epoch": 0.03824339653216659,967 "grad_norm": 1.0746787786483765,968 "learning_rate": 7.642487046632126e-06,969 "loss": 0.27881345748901365,970 "step": 590,971 "token_acc": 0.9218981710331191972 },973 {974 "epoch": 0.03856749311294766,975 "grad_norm": 1.2029082775115967,976 "learning_rate": 7.707253886010362e-06,977 "loss": 0.2803633451461792,978 "step": 595,979 "token_acc": 0.9225640759572481980 },981 {982 "epoch": 0.03889158969372873,983 "grad_norm": 1.1455609798431396,984 "learning_rate": 7.772020725388602e-06,985 "loss": 0.28123066425323484,986 "step": 600,987 "token_acc": 0.9223912180928449988 },989 {990 "epoch": 0.0392156862745098,991 "grad_norm": 1.1479970216751099,992 "learning_rate": 7.83678756476684e-06,993 "loss": 0.2779627799987793,994 "step": 605,995 "token_acc": 0.9228941389389806996 },997 {998 "epoch": 0.039539782855290875,999 "grad_norm": 1.1086708307266235,1000 "learning_rate": 7.901554404145079e-06,1001 "loss": 0.27979559898376466,1002 "step": 610,1003 "token_acc": 0.92243475547117791004 },1005 {1006 "epoch": 0.03986387943607195,1007 "grad_norm": 1.1686094999313354,1008 "learning_rate": 7.966321243523317e-06,1009 "loss": 0.2715947151184082,1010 "step": 615,1011 "token_acc": 0.92362656975046241012 },1013 {1014 "epoch": 0.04018797601685302,1015 "grad_norm": 1.0965114831924438,1016 "learning_rate": 8.031088082901555e-06,1017 "loss": 0.2718233823776245,1018 "step": 620,1019 "token_acc": 0.92355543447238011020 },1021 {1022 "epoch": 0.0405120725976341,1023 "grad_norm": 1.1116286516189575,1024 "learning_rate": 8.095854922279794e-06,1025 "loss": 0.26779100894927976,1026 "step": 625,1027 "token_acc": 0.92564862475952791028 },1029 {1030 "epoch": 0.04083616917841517,1031 "grad_norm": 1.1061184406280518,1032 "learning_rate": 8.160621761658032e-06,1033 "loss": 0.27161588668823244,1034 "step": 630,1035 "token_acc": 0.92313902272052081036 },1037 {1038 "epoch": 0.04116026575919624,1039 "grad_norm": 1.137664556503296,1040 "learning_rate": 8.22538860103627e-06,1041 "loss": 0.27380824089050293,1042 "step": 635,1043 "token_acc": 0.92298453685632791044 },1045 {1046 "epoch": 0.041484362339977314,1047 "grad_norm": 1.1426571607589722,1048 "learning_rate": 8.290155440414507e-06,1049 "loss": 0.2724440574645996,1050 "step": 640,1051 "token_acc": 0.92432930776322591052 },1053 {1054 "epoch": 0.041808458920758386,1055 "grad_norm": 1.2707536220550537,1056 "learning_rate": 8.354922279792747e-06,1057 "loss": 0.27617788314819336,1058 "step": 645,1059 "token_acc": 0.92319824187188941060 },1061 {1062 "epoch": 0.04213255550153946,1063 "grad_norm": 1.0595289468765259,1064 "learning_rate": 8.419689119170985e-06,1065 "loss": 0.27484869956970215,1066 "step": 650,1067 "token_acc": 0.92367666232073011068 },1069 {1070 "epoch": 0.04245665208232053,1071 "grad_norm": 1.2787915468215942,1072 "learning_rate": 8.484455958549224e-06,1073 "loss": 0.2772678375244141,1074 "step": 655,1075 "token_acc": 0.92256569599915021076 },1077 {1078 "epoch": 0.0427807486631016,1079 "grad_norm": 1.0741102695465088,1080 "learning_rate": 8.549222797927462e-06,1081 "loss": 0.2790708541870117,1082 "step": 660,1083 "token_acc": 0.92256152302492611084 },1085 {1086 "epoch": 0.043104845243882674,1087 "grad_norm": 1.0884015560150146,1088 "learning_rate": 8.6139896373057e-06,1089 "loss": 0.2731776237487793,1090 "step": 665,1091 "token_acc": 0.9237144277129821092 },1093 {1094 "epoch": 0.04342894182466375,1095 "grad_norm": 1.1796871423721313,1096 "learning_rate": 8.678756476683938e-06,1097 "loss": 0.26450159549713137,1098 "step": 670,1099 "token_acc": 0.92603454965114981100 },1101 {1102 "epoch": 0.043753038405444825,1103 "grad_norm": 1.1302798986434937,1104 "learning_rate": 8.743523316062177e-06,1105 "loss": 0.286497974395752,1106 "step": 675,1107 "token_acc": 0.92126078064084831108 },1109 {1110 "epoch": 0.0440771349862259,1111 "grad_norm": 1.0766465663909912,1112 "learning_rate": 8.808290155440415e-06,1113 "loss": 0.2754509925842285,1114 "step": 680,1115 "token_acc": 0.92239066447865381116 },1117 {1118 "epoch": 0.04440123156700697,1119 "grad_norm": 1.1511085033416748,1120 "learning_rate": 8.873056994818654e-06,1121 "loss": 0.2762892723083496,1122 "step": 685,1123 "token_acc": 0.92427732049151371124 },1125 {1126 "epoch": 0.04472532814778804,1127 "grad_norm": 1.2351347208023071,1128 "learning_rate": 8.937823834196892e-06,1129 "loss": 0.2697291851043701,1130 "step": 690,1131 "token_acc": 0.92414564352202091132 },1133 {1134 "epoch": 0.04504942472856911,1135 "grad_norm": 1.1533805131912231,1136 "learning_rate": 9.00259067357513e-06,1137 "loss": 0.27661752700805664,1138 "step": 695,1139 "token_acc": 0.92343764266986681140 },1141 {1142 "epoch": 0.045373521309350184,1143 "grad_norm": 1.4566006660461426,1144 "learning_rate": 9.06735751295337e-06,1145 "loss": 0.2665151596069336,1146 "step": 700,1147 "token_acc": 0.92564789292253961148 },1149 {1150 "epoch": 0.045697617890131256,1151 "grad_norm": 1.2104111909866333,1152 "learning_rate": 9.132124352331607e-06,1153 "loss": 0.2740550756454468,1154 "step": 705,1155 "token_acc": 0.92293905637791981156 },1157 {1158 "epoch": 0.046021714470912335,1159 "grad_norm": 1.082419753074646,1160 "learning_rate": 9.196891191709847e-06,1161 "loss": 0.2736359119415283,1162 "step": 710,1163 "token_acc": 0.92428698899500971164 },1165 {1166 "epoch": 0.04634581105169341,1167 "grad_norm": 1.2044905424118042,1168 "learning_rate": 9.261658031088083e-06,1169 "loss": 0.27272777557373046,1170 "step": 715,1171 "token_acc": 0.92376381654998641172 },1173 {1174 "epoch": 0.04666990763247448,1175 "grad_norm": 1.1270380020141602,1176 "learning_rate": 9.326424870466322e-06,1177 "loss": 0.27639906406402587,1178 "step": 720,1179 "token_acc": 0.92292218242226031180 },1181 {1182 "epoch": 0.04699400421325555,1183 "grad_norm": 1.140891432762146,1184 "learning_rate": 9.39119170984456e-06,1185 "loss": 0.2769869327545166,1186 "step": 725,1187 "token_acc": 0.92316430448967731188 },1189 {1190 "epoch": 0.04731810079403662,1191 "grad_norm": 1.2938241958618164,1192 "learning_rate": 9.4559585492228e-06,1193 "loss": 0.2770379543304443,1194 "step": 730,1195 "token_acc": 0.92274794426376881196 },1197 {1198 "epoch": 0.047642197374817695,1199 "grad_norm": 1.1942880153656006,1200 "learning_rate": 9.520725388601037e-06,