anish12/gemma_Code_Instruct_Finetune_test_2500
05
1{2 "best_global_step": null,3 "best_metric": null,4 "best_model_checkpoint": null,5 "epoch": 1.6514701024116287,6 "eval_steps": 500,7 "global_step": 2500,8 "is_hyper_param_search": false,9 "is_local_process_zero": true,10 "is_world_process_zero": true,11 "log_history": [12 {13 "entropy": 6.753322017192841,14 "epoch": 0.006607201850016518,15 "grad_norm": 0.0,16 "learning_rate": 9e-07,17 "loss": 220.5158,18 "mean_token_accuracy": 0.021751992404460907,19 "num_tokens": 40960.0,20 "step": 1021 },22 {23 "entropy": 6.814661657810211,24 "epoch": 0.013214403700033036,25 "grad_norm": 0.0,26 "learning_rate": 1.9e-06,27 "loss": 38.3501,28 "mean_token_accuracy": 0.02807121314108372,29 "num_tokens": 81920.0,30 "step": 2031 },32 {33 "entropy": 6.945201528072357,34 "epoch": 0.019821605550049554,35 "grad_norm": 0.0,36 "learning_rate": 2.9e-06,37 "loss": 62.9007,38 "mean_token_accuracy": 0.01135626919567585,39 "num_tokens": 122880.0,40 "step": 3041 },42 {43 "entropy": 7.206499898433686,44 "epoch": 0.026428807400066073,45 "grad_norm": 0.0,46 "learning_rate": 3.9e-06,47 "loss": 49.7069,48 "mean_token_accuracy": 0.023292785882949828,49 "num_tokens": 163840.0,50 "step": 4051 },52 {53 "entropy": 2.605212825164199,54 "epoch": 0.03303600925008259,55 "grad_norm": 0.0,56 "learning_rate": 4.9000000000000005e-06,57 "loss": 214.3199,58 "mean_token_accuracy": 0.021621854975819588,59 "num_tokens": 204800.0,60 "step": 5061 },62 {63 "entropy": 0.06497852916363626,64 "epoch": 0.03964321110009911,65 "grad_norm": 0.0,66 "learning_rate": 5.9e-06,67 "loss": 11.6628,68 "mean_token_accuracy": 0.043425589054822925,69 "num_tokens": 245760.0,70 "step": 6071 },72 {73 "entropy": 0.021584587066899984,74 "epoch": 0.04625041295011562,75 "grad_norm": 19.475257873535156,76 "learning_rate": 6.900000000000001e-06,77 "loss": 4.9491,78 "mean_token_accuracy": 0.039490992575883864,79 "num_tokens": 286720.0,80 "step": 7081 },82 {83 "entropy": 0.03706419242080301,84 "epoch": 0.052857614800132145,85 "grad_norm": 65.02709197998047,86 "learning_rate": 7.9e-06,87 "loss": 11.027,88 "mean_token_accuracy": 0.07533193230628968,89 "num_tokens": 327680.0,90 "step": 8091 },92 {93 "entropy": 0.05241385023109615,94 "epoch": 0.05946481665014866,95 "grad_norm": 0.0,96 "learning_rate": 8.9e-06,97 "loss": 21.8544,98 "mean_token_accuracy": 0.012833333015441895,99 "num_tokens": 368640.0,100 "step": 90101 },102 {103 "entropy": 0.4075972191989422,104 "epoch": 0.06607201850016518,105 "grad_norm": 0.0,106 "learning_rate": 9.900000000000002e-06,107 "loss": 35.157,108 "mean_token_accuracy": 0.024832390993833543,109 "num_tokens": 409600.0,110 "step": 100111 },112 {113 "entropy": 1.835695518553257,114 "epoch": 0.07267922035018169,115 "grad_norm": 0.0,116 "learning_rate": 1.09e-05,117 "loss": 0.0,118 "mean_token_accuracy": 0.0,119 "num_tokens": 450560.0,120 "step": 110121 },122 {123 "entropy": 0.9127246515825391,124 "epoch": 0.07928642220019821,125 "grad_norm": 0.0,126 "learning_rate": 1.19e-05,127 "loss": 134.9978,128 "mean_token_accuracy": 0.07018110677599906,129 "num_tokens": 491520.0,130 "step": 120131 },132 {133 "entropy": 0.09644881915301085,134 "epoch": 0.08589362405021474,135 "grad_norm": 18.287517547607422,136 "learning_rate": 1.29e-05,137 "loss": 17.8291,138 "mean_token_accuracy": 0.11415426433086395,139 "num_tokens": 532480.0,140 "step": 130141 },142 {143 "entropy": 1.061981100589037,144 "epoch": 0.09250082590023125,145 "grad_norm": 0.0,146 "learning_rate": 1.3900000000000002e-05,147 "loss": 73.0922,148 "mean_token_accuracy": 0.0463849127292633,149 "num_tokens": 573440.0,150 "step": 140151 },152 {153 "entropy": 1.7951227528974414,154 "epoch": 0.09910802775024777,155 "grad_norm": 0.0,156 "learning_rate": 1.49e-05,157 "loss": 26.185,158 "mean_token_accuracy": 0.043432857096195224,159 "num_tokens": 614400.0,160 "step": 150161 },162 {163 "entropy": 0.100301666976884,164 "epoch": 0.10571522960026429,165 "grad_norm": 0.0,166 "learning_rate": 1.59e-05,167 "loss": 45.3161,168 "mean_token_accuracy": 0.08393882662057876,169 "num_tokens": 655360.0,170 "step": 160171 },172 {173 "entropy": 0.7640071153640747,174 "epoch": 0.11232243145028081,175 "grad_norm": 0.0,176 "learning_rate": 1.69e-05,177 "loss": 61.7114,178 "mean_token_accuracy": 0.04098484814167023,179 "num_tokens": 696320.0,180 "step": 170181 },182 {183 "entropy": 0.3028068076006093,184 "epoch": 0.11892963330029732,185 "grad_norm": 0.0,186 "learning_rate": 1.79e-05,187 "loss": 57.9743,188 "mean_token_accuracy": 0.07434001564979553,189 "num_tokens": 737280.0,190 "step": 180191 },192 {193 "entropy": 0.02740152989870239,194 "epoch": 0.12553683515031383,195 "grad_norm": 30.224332809448242,196 "learning_rate": 1.8900000000000002e-05,197 "loss": 43.8902,198 "mean_token_accuracy": 0.06686745882034302,199 "num_tokens": 778240.0,200 "step": 190201 },202 {203 "entropy": 0.00011176439667224259,204 "epoch": 0.13214403700033037,205 "grad_norm": 31.883712768554688,206 "learning_rate": 1.9900000000000003e-05,207 "loss": 0.0793,208 "mean_token_accuracy": 0.02380952388048172,209 "num_tokens": 819200.0,210 "step": 200211 },212 {213 "entropy": 0.005680033719772837,214 "epoch": 0.13875123885034688,215 "grad_norm": 33.85360336303711,216 "learning_rate": 2.09e-05,217 "loss": 0.2071,218 "mean_token_accuracy": 0.015425531566143036,219 "num_tokens": 860160.0,220 "step": 210221 },222 {223 "entropy": 0.0075609647078181295,224 "epoch": 0.14535844070036338,225 "grad_norm": 19.486099243164062,226 "learning_rate": 2.19e-05,227 "loss": 1.201,228 "mean_token_accuracy": 0.07949066758155823,229 "num_tokens": 901120.0,230 "step": 220231 },232 {233 "entropy": 0.011990599616638065,234 "epoch": 0.15196564255037992,235 "grad_norm": 0.0,236 "learning_rate": 2.29e-05,237 "loss": 2.0033,238 "mean_token_accuracy": 0.02794431447982788,239 "num_tokens": 942080.0,240 "step": 230241 },242 {243 "entropy": 0.0031902942908345723,244 "epoch": 0.15857284440039643,245 "grad_norm": 0.0,246 "learning_rate": 2.39e-05,247 "loss": 1.1575,248 "mean_token_accuracy": 0.017129629850387573,249 "num_tokens": 983040.0,250 "step": 240251 },252 {253 "entropy": 0.14381448374479078,254 "epoch": 0.16518004625041294,255 "grad_norm": 0.0,256 "learning_rate": 2.4900000000000002e-05,257 "loss": 0.7004,258 "mean_token_accuracy": 0.021666666865348815,259 "num_tokens": 1024000.0,260 "step": 250261 },262 {263 "entropy": 3.897232323416836,264 "epoch": 0.17178724810042947,265 "grad_norm": 0.0,266 "learning_rate": 2.5900000000000003e-05,267 "loss": 0.812,268 "mean_token_accuracy": 0.039259061217308044,269 "num_tokens": 1064960.0,270 "step": 260271 },272 {273 "entropy": 0.019721664518976723,274 "epoch": 0.17839444995044598,275 "grad_norm": 0.0,276 "learning_rate": 2.6900000000000003e-05,277 "loss": 2.7005,278 "mean_token_accuracy": 0.05297277569770813,279 "num_tokens": 1105920.0,280 "step": 270281 },282 {283 "entropy": 0.02900617156851304,284 "epoch": 0.1850016518004625,285 "grad_norm": 18.309810638427734,286 "learning_rate": 2.7900000000000004e-05,287 "loss": 0.6091,288 "mean_token_accuracy": 0.07364537045359612,289 "num_tokens": 1146880.0,290 "step": 280291 },292 {293 "entropy": 0.03415291602525096,294 "epoch": 0.19160885365047903,295 "grad_norm": 0.0,296 "learning_rate": 2.8899999999999998e-05,297 "loss": 19.6735,298 "mean_token_accuracy": 0.10901817381381988,299 "num_tokens": 1187840.0,300 "step": 290301 },302 {303 "entropy": 0.03262247524494468,304 "epoch": 0.19821605550049554,305 "grad_norm": 0.0,306 "learning_rate": 2.9900000000000002e-05,307 "loss": 10.0336,308 "mean_token_accuracy": 0.06002416908740997,309 "num_tokens": 1228800.0,310 "step": 300311 },312 {313 "entropy": 0.003781934377029783,314 "epoch": 0.20482325735051204,315 "grad_norm": 89.78754425048828,316 "learning_rate": 3.09e-05,317 "loss": 7.6655,318 "mean_token_accuracy": 0.07553683370351791,319 "num_tokens": 1269760.0,320 "step": 310321 },322 {323 "entropy": 0.02375032416614431,324 "epoch": 0.21143045920052858,325 "grad_norm": 0.0,326 "learning_rate": 3.19e-05,327 "loss": 55.6063,328 "mean_token_accuracy": 0.08025674372911454,329 "num_tokens": 1310720.0,330 "step": 320331 },332 {333 "entropy": 0.0065040366395965775,334 "epoch": 0.2180376610505451,335 "grad_norm": 0.0,336 "learning_rate": 3.29e-05,337 "loss": 10.669,338 "mean_token_accuracy": 0.01593567281961441,339 "num_tokens": 1351680.0,340 "step": 330341 },342 {343 "entropy": 0.003754374942396499,344 "epoch": 0.22464486290056163,345 "grad_norm": 0.0,346 "learning_rate": 3.3900000000000004e-05,347 "loss": 0.3125,348 "mean_token_accuracy": 0.042081044614315034,349 "num_tokens": 1392640.0,350 "step": 340351 },352 {353 "entropy": 0.0002896140339728959,354 "epoch": 0.23125206475057813,355 "grad_norm": 16.06511878967285,356 "learning_rate": 3.49e-05,357 "loss": 0.0652,358 "mean_token_accuracy": 0.02388888895511627,359 "num_tokens": 1433600.0,360 "step": 350361 },362 {363 "entropy": 0.0001928646220630714,364 "epoch": 0.23785926660059464,365 "grad_norm": 0.0,366 "learning_rate": 3.59e-05,367 "loss": 1.1458,368 "mean_token_accuracy": 0.02368421107530594,369 "num_tokens": 1474560.0,370 "step": 360371 },372 {373 "entropy": 0.0003226411399264606,374 "epoch": 0.24446646845061118,375 "grad_norm": 0.0,376 "learning_rate": 3.69e-05,377 "loss": 4.4569,378 "mean_token_accuracy": 0.04751228392124176,379 "num_tokens": 1515520.0,380 "step": 370381 },382 {383 "entropy": 0.030229031606691593,384 "epoch": 0.25107367030062766,385 "grad_norm": 0.0,386 "learning_rate": 3.79e-05,387 "loss": 4.4461,388 "mean_token_accuracy": 0.06874987185001373,389 "num_tokens": 1556480.0,390 "step": 380391 },392 {393 "entropy": 0.014421097982523178,394 "epoch": 0.2576808721506442,395 "grad_norm": 0.0,396 "learning_rate": 3.8900000000000004e-05,397 "loss": 89.155,398 "mean_token_accuracy": 0.07968828082084656,399 "num_tokens": 1597440.0,400 "step": 390401 },402 {403 "entropy": 0.018559856562546172,404 "epoch": 0.26428807400066073,405 "grad_norm": 0.0,406 "learning_rate": 3.99e-05,407 "loss": 1.1971,408 "mean_token_accuracy": 0.04879816025495529,409 "num_tokens": 1638400.0,410 "step": 400411 },412 {413 "entropy": 0.0025903429913086027,414 "epoch": 0.27089527585067724,415 "grad_norm": 0.0,416 "learning_rate": 4.09e-05,417 "loss": 46.7922,418 "mean_token_accuracy": 0.06377801150083542,419 "num_tokens": 1679360.0,420 "step": 410421 },422 {423 "entropy": 0.0014362634548334085,424 "epoch": 0.27750247770069375,425 "grad_norm": 0.0,426 "learning_rate": 4.19e-05,427 "loss": 0.2403,428 "mean_token_accuracy": 0.019583334028720856,429 "num_tokens": 1720320.0,430 "step": 420431 },432 {433 "entropy": 0.01779812537132557,434 "epoch": 0.28410967955071026,435 "grad_norm": 0.0,436 "learning_rate": 4.29e-05,437 "loss": 3.4174,438 "mean_token_accuracy": 0.09800765216350556,439 "num_tokens": 1761280.0,440 "step": 430441 },442 {443 "entropy": 0.00543891376372585,444 "epoch": 0.29071688140072677,445 "grad_norm": 0.0,446 "learning_rate": 4.39e-05,447 "loss": 1.7927,448 "mean_token_accuracy": 0.10844782143831252,449 "num_tokens": 1802240.0,450 "step": 440451 },452 {453 "entropy": 0.008688661311566648,454 "epoch": 0.29732408325074333,455 "grad_norm": 11.81262493133545,456 "learning_rate": 4.49e-05,457 "loss": 19.4753,458 "mean_token_accuracy": 0.03172447681427002,459 "num_tokens": 1843200.0,460 "step": 450461 },462 {463 "entropy": 0.03670089549409568,464 "epoch": 0.30393128510075984,465 "grad_norm": 0.0,466 "learning_rate": 4.5900000000000004e-05,467 "loss": 9.5863,468 "mean_token_accuracy": 0.09990399181842805,469 "num_tokens": 1884160.0,470 "step": 460471 },472 {473 "entropy": 0.0009635578760935459,474 "epoch": 0.31053848695077635,475 "grad_norm": 0.0,476 "learning_rate": 4.69e-05,477 "loss": 0.1027,478 "mean_token_accuracy": 0.047025862336158755,479 "num_tokens": 1925120.0,480 "step": 470481 },482 {483 "entropy": 0.03435224668937735,484 "epoch": 0.31714568880079286,485 "grad_norm": 0.0,486 "learning_rate": 4.79e-05,487 "loss": 9.3884,488 "mean_token_accuracy": 0.12481869906187057,489 "num_tokens": 1966080.0,490 "step": 480491 },492 {493 "entropy": 0.02036606671208574,494 "epoch": 0.32375289065080937,495 "grad_norm": 12.76618766784668,496 "learning_rate": 4.89e-05,497 "loss": 8.3534,498 "mean_token_accuracy": 0.025242918729782106,499 "num_tokens": 2007040.0,500 "step": 490501 },502 {503 "entropy": 0.005103755064374127,504 "epoch": 0.3303600925008259,505 "grad_norm": 10.067124366760254,506 "learning_rate": 4.99e-05,507 "loss": 0.996,508 "mean_token_accuracy": 0.03732142895460129,509 "num_tokens": 2048000.0,510 "step": 500511 },512 {513 "entropy": 0.02930727633283823,514 "epoch": 0.33696729435084244,515 "grad_norm": 13.147978782653809,516 "learning_rate": 4.999843636237961e-05,517 "loss": 2.6708,518 "mean_token_accuracy": 0.16797256022691726,519 "num_tokens": 2088960.0,520 "step": 510521 },522 {523 "entropy": 0.01542719653371023,524 "epoch": 0.34357449620085895,525 "grad_norm": 0.0,526 "learning_rate": 4.999303144641334e-05,527 "loss": 0.7947,528 "mean_token_accuracy": 0.029625000059604646,529 "num_tokens": 2129920.0,530 "step": 520531 },532 {533 "entropy": 0.011237168186926282,534 "epoch": 0.35018169805087546,535 "grad_norm": 0.0,536 "learning_rate": 4.9983766782431473e-05,537 "loss": 1.8663,538 "mean_token_accuracy": 0.015829145908355713,539 "num_tokens": 2170880.0,540 "step": 530541 },542 {543 "entropy": 0.021366176847368478,544 "epoch": 0.35678889990089196,545 "grad_norm": 0.0,546 "learning_rate": 4.997064380120359e-05,547 "loss": 0.7902,548 "mean_token_accuracy": 0.034427966177463534,549 "num_tokens": 2211840.0,550 "step": 540551 },552 {553 "entropy": 0.013086076875333674,554 "epoch": 0.3633961017509085,555 "grad_norm": 0.0,556 "learning_rate": 4.995366452935061e-05,557 "loss": 4.0451,558 "mean_token_accuracy": 0.03658333420753479,559 "num_tokens": 2252800.0,560 "step": 550561 },562 {563 "entropy": 0.15343078810255975,564 "epoch": 0.370003303600925,565 "grad_norm": 21.48503303527832,566 "learning_rate": 4.993283158903187e-05,567 "loss": 108.0904,568 "mean_token_accuracy": 0.03694373369216919,569 "num_tokens": 2293760.0,570 "step": 560571 },572 {573 "entropy": 0.08353904361720196,574 "epoch": 0.37661050545094155,575 "grad_norm": 0.0,576 "learning_rate": 4.9908148197540174e-05,577 "loss": 3.3117,578 "mean_token_accuracy": 0.05191558599472046,579 "num_tokens": 2334720.0,580 "step": 570581 },582 {583 "entropy": 0.0030798995423538143,584 "epoch": 0.38321770730095805,585 "grad_norm": 15.358031272888184,586 "learning_rate": 4.987961816680492e-05,587 "loss": 107.2357,588 "mean_token_accuracy": 0.040966108441352844,589 "num_tokens": 2375680.0,590 "step": 580591 },592 {593 "entropy": 0.0050722503918223085,594 "epoch": 0.38982490915097456,595 "grad_norm": 0.0,596 "learning_rate": 4.984724590280343e-05,597 "loss": 1.1264,598 "mean_token_accuracy": 0.01942567527294159,599 "num_tokens": 2416640.0,600 "step": 590601 },602 {603 "entropy": 0.025612023619032698,604 "epoch": 0.39643211100099107,605 "grad_norm": 0.0,606 "learning_rate": 4.981103640488048e-05,607 "loss": 1.0595,608 "mean_token_accuracy": 0.05857066810131073,609 "num_tokens": 2457600.0,610 "step": 600611 },612 {613 "entropy": 0.02604469430661993,614 "epoch": 0.4030393128510076,615 "grad_norm": 0.0,616 "learning_rate": 4.977099526497631e-05,617 "loss": 2.9458,618 "mean_token_accuracy": 0.11472017914056779,619 "num_tokens": 2498560.0,620 "step": 610621 },622 {623 "entropy": 1.34335226020159,624 "epoch": 0.4096465147010241,625 "grad_norm": 0.0,626 "learning_rate": 4.972712866676297e-05,627 "loss": 0.8969,628 "mean_token_accuracy": 0.0374501496553421,629 "num_tokens": 2539520.0,630 "step": 620631 },632 {633 "entropy": 0.017200125091973145,634 "epoch": 0.41625371655104065,635 "grad_norm": 6.167868614196777,636 "learning_rate": 4.96794433846894e-05,637 "loss": 0.7561,638 "mean_token_accuracy": 0.04967625737190247,639 "num_tokens": 2580480.0,640 "step": 630641 },642 {643 "entropy": 0.013742945558520164,644 "epoch": 0.42286091840105716,645 "grad_norm": 4.658291816711426,646 "learning_rate": 4.962794678293523e-05,647 "loss": 3.9976,648 "mean_token_accuracy": 0.09124306589365005,649 "num_tokens": 2621440.0,650 "step": 640651 },652 {653 "entropy": 0.018405587278946686,654 "epoch": 0.42946812025107367,655 "grad_norm": 0.0,656 "learning_rate": 4.957264681427348e-05,657 "loss": 1.2244,658 "mean_token_accuracy": 0.030132341384887695,659 "num_tokens": 2662400.0,660 "step": 650661 },662 {663 "entropy": 0.007059359534223398,664 "epoch": 0.4360753221010902,665 "grad_norm": 41.432003021240234,666 "learning_rate": 4.9513552018842426e-05,667 "loss": 5.2841,668 "mean_token_accuracy": 0.10671066045761109,669 "num_tokens": 2703360.0,670 "step": 660671 },672 {673 "entropy": 0.004513938081906041,674 "epoch": 0.4426825239511067,675 "grad_norm": 11.688268661499023,676 "learning_rate": 4.9450671522826694e-05,677 "loss": 0.4656,678 "mean_token_accuracy": 0.0375,679 "num_tokens": 2744320.0,680 "step": 670681 },682 {683 "entropy": 6.520482699785834e-12,684 "epoch": 0.44928972580112325,685 "grad_norm": 0.0,686 "learning_rate": 4.938401503704789e-05,687 "loss": 0.0,688 "mean_token_accuracy": 0.0,689 "num_tokens": 2785280.0,690 "step": 680691 },692 {693 "entropy": 0.019155432335637948,694 "epoch": 0.45589692765113976,695 "grad_norm": 7.324256420135498,696 "learning_rate": 4.9313592855464916e-05,697 "loss": 8.6945,698 "mean_token_accuracy": 0.07001543641090394,699 "num_tokens": 2826240.0,700 "step": 690701 },702 {703 "entropy": 0.0031058518021977906,704 "epoch": 0.46250412950115627,705 "grad_norm": 0.0,706 "learning_rate": 4.923941585358426e-05,707 "loss": 24.0465,708 "mean_token_accuracy": 0.039618226885795596,709 "num_tokens": 2867200.0,710 "step": 700711 },712 {713 "entropy": 0.011669352487973095,714 "epoch": 0.4691113313511728,715 "grad_norm": 0.0,716 "learning_rate": 4.9161495486780456e-05,717 "loss": 3.7727,718 "mean_token_accuracy": 0.14685782939195632,719 "num_tokens": 2908160.0,720 "step": 710721 },722 {723 "entropy": 0.011068030085762338,724 "epoch": 0.4757185332011893,725 "grad_norm": 0.0,726 "learning_rate": 4.907984378852699e-05,727 "loss": 4.0753,728 "mean_token_accuracy": 0.05300231724977493,729 "num_tokens": 2949120.0,730 "step": 720731 },732 {733 "entropy": 0.019895143689083482,734 "epoch": 0.4823257350512058,735 "grad_norm": 0.0,736 "learning_rate": 4.8994473368537916e-05,737 "loss": 3.4842,738 "mean_token_accuracy": 0.06887777000665665,739 "num_tokens": 2990080.0,740 "step": 730741 },742 {743 "entropy": 0.00012087312168218567,744 "epoch": 0.48893293690122236,745 "grad_norm": 0.0,746 "learning_rate": 4.8905397410820554e-05,747 "loss": 0.4551,748 "mean_token_accuracy": 0.02083333283662796,749 "num_tokens": 3031040.0,750 "step": 740751 },752 {753 "entropy": 0.011406292311664856,754 "epoch": 0.49554013875123887,755 "grad_norm": 0.0,756 "learning_rate": 4.881262967163939e-05,757 "loss": 25.7439,758 "mean_token_accuracy": 0.08733686208724975,759 "num_tokens": 3072000.0,760 "step": 750761 },762 {763 "entropy": 0.007139437921159697,764 "epoch": 0.5021473406012553,765 "grad_norm": 0.0,766 "learning_rate": 4.8716184477391675e-05,767 "loss": 3.6028,768 "mean_token_accuracy": 0.03767942488193512,769 "num_tokens": 3112960.0,770 "step": 760771 },772 {773 "entropy": 0.011418092579742734,774 "epoch": 0.5087545424512719,775 "grad_norm": 0.0,776 "learning_rate": 4.8616076722394974e-05,777 "loss": 3.0005,778 "mean_token_accuracy": 0.07940062433481217,779 "num_tokens": 3153920.0,780 "step": 770781 },782 {783 "entropy": 0.006054546175801079,784 "epoch": 0.5153617443012884,785 "grad_norm": 17.855485916137695,786 "learning_rate": 4.8512321866586985e-05,787 "loss": 0.2468,788 "mean_token_accuracy": 0.03400537669658661,789 "num_tokens": 3194880.0,790 "step": 780791 },792 {793 "entropy": 0.02499405129653951,794 "epoch": 0.521968946151305,795 "grad_norm": 0.0,796 "learning_rate": 4.840493593313798e-05,797 "loss": 7.3568,798 "mean_token_accuracy": 0.06639731675386429,799 "num_tokens": 3235840.0,800 "step": 790801 },802 {803 "entropy": 0.030509774422012016,804 "epoch": 0.5285761480013215,805 "grad_norm": 8.936971664428711,806 "learning_rate": 4.829393550597633e-05,807 "loss": 2.2173,808 "mean_token_accuracy": 0.0885118618607521,809 "num_tokens": 3276800.0,810 "step": 800811 },812 {813 "entropy": 0.008201998914637442,814 "epoch": 0.535183349851338,815 "grad_norm": 0.0,816 "learning_rate": 4.8179337727227424e-05,817 "loss": 10.5952,818 "mean_token_accuracy": 0.03566812574863434,819 "num_tokens": 3317760.0,820 "step": 810821 },822 {823 "entropy": 0.0163260028579316,824 "epoch": 0.5417905517013545,825 "grad_norm": 8.472152709960938,826 "learning_rate": 4.806116029456631e-05,827 "loss": 9.1567,828 "mean_token_accuracy": 0.02788277491927147,829 "num_tokens": 3358720.0,830 "step": 820831 },832 {833 "entropy": 0.014577616745373233,834 "epoch": 0.548397753551371,835 "grad_norm": 0.0,836 "learning_rate": 4.7939421458484604e-05,837 "loss": 14.4947,838 "mean_token_accuracy": 0.038144654035568236,839 "num_tokens": 3399680.0,840 "step": 830841 },842 {843 "entropy": 0.060555978957563636,844 "epoch": 0.5550049554013875,845 "grad_norm": 5.755632400512695,846 "learning_rate": 4.781414001947205e-05,847 "loss": 17.9944,848 "mean_token_accuracy": 0.08551393002271652,849 "num_tokens": 3440640.0,850 "step": 840851 },852 {853 "entropy": 0.01132769331743475,854 "epoch": 0.561612157251404,855 "grad_norm": 0.0,856 "learning_rate": 4.768533532511306e-05,857 "loss": 54.7683,858 "mean_token_accuracy": 0.034085105359554294,859 "num_tokens": 3481600.0,860 "step": 850861 },862 {863 "entropy": 0.03579027488522115,864 "epoch": 0.5682193591014205,865 "grad_norm": 1.5527058839797974,866 "learning_rate": 4.755302726709882e-05,867 "loss": 48.52,868 "mean_token_accuracy": 0.06784629076719284,869 "num_tokens": 3522560.0,870 "step": 860871 },872 {873 "entropy": 0.025357943994458766,874 "epoch": 0.574826560951437,875 "grad_norm": 0.0,876 "learning_rate": 4.741723627815536e-05,877 "loss": 8.7313,878 "mean_token_accuracy": 0.06684041172266006,879 "num_tokens": 3563520.0,880 "step": 870881 },882 {883 "entropy": 0.003138951886649011,884 "epoch": 0.5814337628014535,885 "grad_norm": 0.7629010677337646,886 "learning_rate": 4.7277983328888084e-05,887 "loss": 4.1401,888 "mean_token_accuracy": 0.0589192196726799,889 "num_tokens": 3604480.0,890 "step": 880891 },892 {893 "entropy": 0.008384830836621403,894 "epoch": 0.5880409646514702,895 "grad_norm": 0.0,896 "learning_rate": 4.7135289924543197e-05,897 "loss": 4.0731,898 "mean_token_accuracy": 0.032692307233810426,899 "num_tokens": 3645440.0,900 "step": 890901 },902 {903 "entropy": 0.006141025204078688,904 "epoch": 0.5946481665014867,905 "grad_norm": 7.293692588806152,906 "learning_rate": 4.6989178101686584e-05,907 "loss": 0.9345,908 "mean_token_accuracy": 0.06051216870546341,909 "num_tokens": 3686400.0,910 "step": 900911 },912 {913 "entropy": 0.057043678060654204,914 "epoch": 0.6012553683515032,915 "grad_norm": 4.391047954559326,916 "learning_rate": 4.683967042480064e-05,917 "loss": 3.9633,918 "mean_token_accuracy": 0.12203554809093475,919 "num_tokens": 3727360.0,920 "step": 910921 },922 {923 "entropy": 0.012308508742717095,924 "epoch": 0.6078625702015197,925 "grad_norm": 2.2332639694213867,926 "learning_rate": 4.668678998279961e-05,927 "loss": 0.6798,928 "mean_token_accuracy": 0.10872009098529815,929 "num_tokens": 3768320.0,930 "step": 920931 },932 {933 "entropy": 0.06679710658499971,934 "epoch": 0.6144697720515362,935 "grad_norm": 0.0,936 "learning_rate": 4.653056038546381e-05,937 "loss": 1.5921,938 "mean_token_accuracy": 0.020000000298023225,939 "num_tokens": 3809280.0,940 "step": 930941 },942 {943 "entropy": 0.41163066651206465,944 "epoch": 0.6210769739015527,945 "grad_norm": 0.0,946 "learning_rate": 4.637100575979361e-05,947 "loss": 1.009,948 "mean_token_accuracy": 0.029701614379882814,949 "num_tokens": 3850240.0,950 "step": 940951 },952 {953 "entropy": 0.005564717383629158,954 "epoch": 0.6276841757515692,955 "grad_norm": 8.103035926818848,956 "learning_rate": 4.620815074628333e-05,957 "loss": 0.4482,958 "mean_token_accuracy": 0.060583002865314484,959 "num_tokens": 3891200.0,960 "step": 950961 },962 {963 "entropy": 0.006938851803111312,964 "epoch": 0.6342913776015857,965 "grad_norm": 0.0,966 "learning_rate": 4.6042020495116005e-05,967 "loss": 12.1827,968 "mean_token_accuracy": 0.03860159516334534,969 "num_tokens": 3932160.0,970 "step": 960971 },972 {973 "entropy": 0.025441451920087132,974 "epoch": 0.6408985794516022,975 "grad_norm": 0.0,976 "learning_rate": 4.587264066227933e-05,977 "loss": 4.7399,978 "mean_token_accuracy": 0.045529181510210036,979 "num_tokens": 3973120.0,980 "step": 970981 },982 {983 "entropy": 0.0037336295128966414,984 "epoch": 0.6475057813016187,985 "grad_norm": 0.0,986 "learning_rate": 4.570003740560352e-05,987 "loss": 9.9644,988 "mean_token_accuracy": 0.032598039507865904,989 "num_tokens": 4014080.0,990 "step": 980991 },992 {993 "entropy": 0.02236287235959935,994 "epoch": 0.6541129831516352,995 "grad_norm": 0.0,996 "learning_rate": 4.5524237380721755e-05,997 "loss": 25.0121,998 "mean_token_accuracy": 0.09488918632268906,999 "num_tokens": 4055040.0,1000 "step": 9901001 },1002 {1003 "entropy": 0.004253433111853155,1004 "epoch": 0.6607201850016517,1005 "grad_norm": 0.0,1006 "learning_rate": 4.5345267736953566e-05,1007 "loss": 2.2839,1008 "mean_token_accuracy": 0.061769942939281466,1009 "num_tokens": 4096000.0,1010 "step": 10001011 },1012 {1013 "entropy": 0.004871105551137589,1014 "epoch": 0.6673273868516684,1015 "grad_norm": 0.0,1016 "learning_rate": 4.516315611311215e-05,1017 "loss": 0.4531,1018 "mean_token_accuracy": 0.03642857223749161,1019 "num_tokens": 4136960.0,1020 "step": 10101021 },1022 {1023 "entropy": 0.025206556524062763,1024 "epoch": 0.6739345887016849,1025 "grad_norm": 0.0,1026 "learning_rate": 4.4977930633236e-05,1027 "loss": 1.3895,1028 "mean_token_accuracy": 0.0927792876958847,1029 "num_tokens": 4177920.0,1030 "step": 10201031 },1032 {1033 "entropy": 0.006291561650141375,1034 "epoch": 0.6805417905517014,1035 "grad_norm": 0.0,1036 "learning_rate": 4.478961990224567e-05,1037 "loss": 3.6759,1038 "mean_token_accuracy": 0.06052185446023941,1039 "num_tokens": 4218880.0,1040 "step": 10301041 },1042 {1043 "entropy": 0.0068054500185098735,1044 "epoch": 0.6871489924017179,1045 "grad_norm": 7.908809185028076,1046 "learning_rate": 4.4598253001526165e-05,1047 "loss": 0.8344,1048 "mean_token_accuracy": 0.04019886404275894,1049 "num_tokens": 4259840.0,1050 "step": 10401051 },1052 {1053 "entropy": 0.02280626448646217,1054 "epoch": 0.6937561942517344,1055 "grad_norm": 0.0,1056 "learning_rate": 4.440385948443586e-05,1057 "loss": 56.0997,1058 "mean_token_accuracy": 0.05091123506426811,1059 "num_tokens": 4300800.0,1060 "step": 10501061 },1062 {1063 "entropy": 0.009011537862716069,1064 "epoch": 0.7003633961017509,1065 "grad_norm": 0.0,1066 "learning_rate": 4.420646937174249e-05,1067 "loss": 4.2313,1068 "mean_token_accuracy": 0.10524300783872605,1069 "num_tokens": 4341760.0,1070 "step": 10601071 },1072 {1073 "entropy": 0.0018482632265659049,1074 "epoch": 0.7069705979517674,1075 "grad_norm": 20.32362937927246,1076 "learning_rate": 4.400611314698693e-05,1077 "loss": 0.0597,1078 "mean_token_accuracy": 0.04590909034013748,1079 "num_tokens": 4382720.0,1080 "step": 10701081 },1082 {1083 "entropy": 0.017142248089658096,1084 "epoch": 0.7135777998017839,1085 "grad_norm": 0.0,1086 "learning_rate": 4.3802821751775545e-05,1087 "loss": 3.2115,1088 "mean_token_accuracy": 0.03341048508882523,1089 "num_tokens": 4423680.0,1090 "step": 10801091 },1092 {1093 "entropy": 0.009434876564773731,1094 "epoch": 0.7201850016518004,1095 "grad_norm": 0.0,1096 "learning_rate": 4.35966265810018e-05,1097 "loss": 0.5134,1098 "mean_token_accuracy": 0.05730164796113968,1099 "num_tokens": 4464640.0,1100 "step": 10901101 },1102 {1103 "entropy": 0.01804768085712567,1104 "epoch": 0.726792203501817,1105 "grad_norm": 6.497892379760742,1106 "learning_rate": 4.338755947799779e-05,1107 "loss": 0.3714,1108 "mean_token_accuracy": 0.030211375653743745,1109 "num_tokens": 4505600.0,1110 "step": 11001111 },1112 {1113 "entropy": 0.010639433527830987,1114 "epoch": 0.7333994053518335,1115 "grad_norm": 0.0,1116 "learning_rate": 4.317565272961668e-05,1117 "loss": 5.4826,1118 "mean_token_accuracy": 0.03939753919839859,1119 "num_tokens": 4546560.0,1120 "step": 11101121 },1122 {1123 "entropy": 0.009574767109006644,1124 "epoch": 0.74000660720185,1125 "grad_norm": 1.7740790843963623,1126 "learning_rate": 4.296093906124648e-05,1127 "loss": 0.0149,1128 "mean_token_accuracy": 0.024193547666072845,1129 "num_tokens": 4587520.0,1130 "step": 11201131 },1132 {1133 "entropy": 0.016046867379918693,1134 "epoch": 0.7466138090518666,1135 "grad_norm": 5.654629230499268,1136 "learning_rate": 4.274345163175617e-05,1137 "loss": 0.1119,1138 "mean_token_accuracy": 0.020338982343673706,1139 "num_tokens": 4628480.0,1140 "step": 11301141 },1142 {1143 "entropy": 0.011864034581230953,1144 "epoch": 0.7532210109018831,1145 "grad_norm": 0.0,1146 "learning_rate": 4.252322402837491e-05,1147 "loss": 2.6892,1148 "mean_token_accuracy": 0.06476814448833465,1149 "num_tokens": 4669440.0,1150 "step": 11401151 },1152 {1153 "entropy": 0.016540668293600902,1154 "epoch": 0.7598282127518996,1155 "grad_norm": 16.53849220275879,1156 "learning_rate": 4.2300290261505036e-05,1157 "loss": 30.6023,1158 "mean_token_accuracy": 0.0714763343334198,1159 "num_tokens": 4710400.0,1160 "step": 11501161 },1162 {1163 "entropy": 0.0230761248501949,1164 "epoch": 0.7664354146019161,1165 "grad_norm": 0.0,1166 "learning_rate": 4.2074684759469746e-05,1167 "loss": 2.7597,1168 "mean_token_accuracy": 0.03069286197423935,1169 "num_tokens": 4751360.0,1170 "step": 11601171 },1172 {1173 "entropy": 0.011810722067821189,1174 "epoch": 0.7730426164519326,1175 "grad_norm": 0.0,1176 "learning_rate": 4.1846442363196216e-05,1177 "loss": 0.9111,1178 "mean_token_accuracy": 0.10935835093259812,1179 "num_tokens": 4792320.0,1180 "step": 11701181 },1182 {1183 "entropy": 0.014862862655547815,1184 "epoch": 0.7796498183019491,1185 "grad_norm": 0.0,1186 "learning_rate": 4.161559832083505e-05,1187 "loss": 6.2322,1188 "mean_token_accuracy": 0.07727646380662918,1189 "num_tokens": 4833280.0,1190 "step": 11801191 },1192 {1193 "entropy": 0.03382376935035154,1194 "epoch": 0.7862570201519656,1195 "grad_norm": 0.0,1196 "learning_rate": 4.138218828231674e-05,1197 "loss": 6.8809,1198 "mean_token_accuracy": 0.08848222196102143,1199 "num_tokens": 4874240.0,1200 "step": 1190