astratool/test_summ
09
1{2 "best_global_step": null,3 "best_metric": null,4 "best_model_checkpoint": null,5 "epoch": 0.20882790413617155,6 "eval_steps": 500,7 "global_step": 21200,8 "is_hyper_param_search": false,9 "is_local_process_zero": true,10 "is_world_process_zero": true,11 "log_history": [12 {13 "epoch": 9.850372836611865e-05,14 "grad_norm": 13.80892276763916,15 "learning_rate": 9.999556733222353e-06,16 "loss": 1.3441,17 "mean_token_accuracy": 0.287836092710495,18 "num_tokens": 36184.0,19 "step": 1020 },21 {22 "epoch": 0.0001970074567322373,23 "grad_norm": 7.670838832855225,24 "learning_rate": 9.999064214580522e-06,25 "loss": 1.4053,26 "mean_token_accuracy": 0.2846388816833496,27 "num_tokens": 74933.0,28 "step": 2029 },30 {31 "epoch": 0.00029551118509835595,32 "grad_norm": 12.051396369934082,33 "learning_rate": 9.998571695938692e-06,34 "loss": 1.3428,35 "mean_token_accuracy": 0.26332987695932386,36 "num_tokens": 110369.0,37 "step": 3038 },39 {40 "epoch": 0.0003940149134644746,41 "grad_norm": 14.364080429077148,42 "learning_rate": 9.998079177296862e-06,43 "loss": 1.3506,44 "mean_token_accuracy": 0.2840617597103119,45 "num_tokens": 147242.0,46 "step": 4047 },48 {49 "epoch": 0.0004925186418305933,50 "grad_norm": 12.7632474899292,51 "learning_rate": 9.997586658655031e-06,52 "loss": 1.388,53 "mean_token_accuracy": 0.3153156369924545,54 "num_tokens": 181353.0,55 "step": 5056 },57 {58 "epoch": 0.0005910223701967119,59 "grad_norm": 7.527547359466553,60 "learning_rate": 9.997094140013201e-06,61 "loss": 1.357,62 "mean_token_accuracy": 0.3107276916503906,63 "num_tokens": 219950.0,64 "step": 6065 },66 {67 "epoch": 0.0006895260985628306,68 "grad_norm": 11.05453109741211,69 "learning_rate": 9.99660162137137e-06,70 "loss": 1.1579,71 "mean_token_accuracy": 0.2719868704676628,72 "num_tokens": 259916.0,73 "step": 7074 },75 {76 "epoch": 0.0007880298269289492,77 "grad_norm": 18.75937271118164,78 "learning_rate": 9.996109102729539e-06,79 "loss": 1.3622,80 "mean_token_accuracy": 0.28812679573893546,81 "num_tokens": 297028.0,82 "step": 8083 },84 {85 "epoch": 0.000886533555295068,86 "grad_norm": 25.195301055908203,87 "learning_rate": 9.995616584087708e-06,88 "loss": 1.3522,89 "mean_token_accuracy": 0.2844611883163452,90 "num_tokens": 332601.0,91 "step": 9092 },93 {94 "epoch": 0.0009850372836611866,95 "grad_norm": 12.858141899108887,96 "learning_rate": 9.995124065445878e-06,97 "loss": 1.2718,98 "mean_token_accuracy": 0.3379636367317289,99 "num_tokens": 382615.0,100 "step": 100101 },102 {103 "epoch": 0.0010835410120273053,104 "grad_norm": 12.052271842956543,105 "learning_rate": 9.994631546804048e-06,106 "loss": 1.2689,107 "mean_token_accuracy": 0.2996787801384926,108 "num_tokens": 424004.0,109 "step": 110110 },111 {112 "epoch": 0.0011820447403934238,113 "grad_norm": 17.62078094482422,114 "learning_rate": 9.994139028162216e-06,115 "loss": 1.1202,116 "mean_token_accuracy": 0.21718956530094147,117 "num_tokens": 464380.0,118 "step": 120119 },120 {121 "epoch": 0.0012805484687595425,122 "grad_norm": 1.7914295196533203,123 "learning_rate": 9.993646509520385e-06,124 "loss": 1.1385,125 "mean_token_accuracy": 0.21831632629036904,126 "num_tokens": 502361.0,127 "step": 130128 },129 {130 "epoch": 0.0013790521971256612,131 "grad_norm": 13.129693984985352,132 "learning_rate": 9.993153990878555e-06,133 "loss": 1.2664,134 "mean_token_accuracy": 0.28777758926153185,135 "num_tokens": 538812.0,136 "step": 140137 },138 {139 "epoch": 0.00147755592549178,140 "grad_norm": 8.33322811126709,141 "learning_rate": 9.992661472236726e-06,142 "loss": 1.2428,143 "mean_token_accuracy": 0.22953060865402222,144 "num_tokens": 574283.0,145 "step": 150146 },147 {148 "epoch": 0.0015760596538578985,149 "grad_norm": 4.951908111572266,150 "learning_rate": 9.992168953594894e-06,151 "loss": 1.2274,152 "mean_token_accuracy": 0.2668746553361416,153 "num_tokens": 613007.0,154 "step": 160155 },156 {157 "epoch": 0.0016745633822240172,158 "grad_norm": 9.569474220275879,159 "learning_rate": 9.991676434953064e-06,160 "loss": 1.4752,161 "mean_token_accuracy": 0.29456586241722105,162 "num_tokens": 657188.0,163 "step": 170164 },165 {166 "epoch": 0.001773067110590136,167 "grad_norm": 12.480253219604492,168 "learning_rate": 9.991183916311234e-06,169 "loss": 1.3493,170 "mean_token_accuracy": 0.27840689569711685,171 "num_tokens": 698027.0,172 "step": 180173 },174 {175 "epoch": 0.0018715708389562544,176 "grad_norm": 15.09508228302002,177 "learning_rate": 9.990691397669403e-06,178 "loss": 1.5722,179 "mean_token_accuracy": 0.2553876742720604,180 "num_tokens": 733782.0,181 "step": 190182 },183 {184 "epoch": 0.001970074567322373,185 "grad_norm": 9.428617477416992,186 "learning_rate": 9.990198879027571e-06,187 "loss": 1.3645,188 "mean_token_accuracy": 0.25801796838641167,189 "num_tokens": 769584.0,190 "step": 200191 },192 {193 "epoch": 0.0020685782956884917,194 "grad_norm": 2.836794376373291,195 "learning_rate": 9.989706360385741e-06,196 "loss": 1.3846,197 "mean_token_accuracy": 0.36412381678819655,198 "num_tokens": 817707.0,199 "step": 210200 },201 {202 "epoch": 0.0021670820240546106,203 "grad_norm": 8.119157791137695,204 "learning_rate": 9.98921384174391e-06,205 "loss": 1.3178,206 "mean_token_accuracy": 0.27328921034932135,207 "num_tokens": 858667.0,208 "step": 220209 },210 {211 "epoch": 0.002265585752420729,212 "grad_norm": 11.941020011901855,213 "learning_rate": 9.98872132310208e-06,214 "loss": 1.3766,215 "mean_token_accuracy": 0.3305239647626877,216 "num_tokens": 900153.0,217 "step": 230218 },219 {220 "epoch": 0.0023640894807868476,221 "grad_norm": 6.975298881530762,222 "learning_rate": 9.98822880446025e-06,223 "loss": 1.1101,224 "mean_token_accuracy": 0.3095137126743793,225 "num_tokens": 937428.0,226 "step": 240227 },228 {229 "epoch": 0.0024625932091529665,230 "grad_norm": 12.61508846282959,231 "learning_rate": 9.98773628581842e-06,232 "loss": 1.1659,233 "mean_token_accuracy": 0.29779959172010423,234 "num_tokens": 978168.0,235 "step": 250236 },237 {238 "epoch": 0.002561096937519085,239 "grad_norm": 8.673341751098633,240 "learning_rate": 9.98724376717659e-06,241 "loss": 1.2544,242 "mean_token_accuracy": 0.2580150574445724,243 "num_tokens": 1014256.0,244 "step": 260245 },246 {247 "epoch": 0.0026596006658852036,248 "grad_norm": 10.724403381347656,249 "learning_rate": 9.986751248534757e-06,250 "loss": 1.3806,251 "mean_token_accuracy": 0.2694324948824942,252 "num_tokens": 1052196.0,253 "step": 270254 },255 {256 "epoch": 0.0027581043942513225,257 "grad_norm": 7.510775089263916,258 "learning_rate": 9.986258729892927e-06,259 "loss": 1.2511,260 "mean_token_accuracy": 0.25950719621032475,261 "num_tokens": 1094703.0,262 "step": 280263 },264 {265 "epoch": 0.002856608122617441,266 "grad_norm": 20.335159301757812,267 "learning_rate": 9.985766211251097e-06,268 "loss": 1.5055,269 "mean_token_accuracy": 0.3105102404952049,270 "num_tokens": 1129205.0,271 "step": 290272 },273 {274 "epoch": 0.00295511185098356,275 "grad_norm": 8.025446891784668,276 "learning_rate": 9.985273692609266e-06,277 "loss": 1.3469,278 "mean_token_accuracy": 0.2759637013077736,279 "num_tokens": 1167154.0,280 "step": 300281 },282 {283 "epoch": 0.0030536155793496784,284 "grad_norm": 8.353572845458984,285 "learning_rate": 9.984781173967434e-06,286 "loss": 1.4438,287 "mean_token_accuracy": 0.3341895163059235,288 "num_tokens": 1202601.0,289 "step": 310290 },291 {292 "epoch": 0.003152119307715797,293 "grad_norm": 7.285520076751709,294 "learning_rate": 9.984288655325604e-06,295 "loss": 1.2048,296 "mean_token_accuracy": 0.2948574051260948,297 "num_tokens": 1238969.0,298 "step": 320299 },300 {301 "epoch": 0.003250623036081916,302 "grad_norm": 6.152414321899414,303 "learning_rate": 9.983796136683774e-06,304 "loss": 1.2602,305 "mean_token_accuracy": 0.2921844758093357,306 "num_tokens": 1282134.0,307 "step": 330308 },309 {310 "epoch": 0.0033491267644480344,311 "grad_norm": 10.239402770996094,312 "learning_rate": 9.983303618041943e-06,313 "loss": 1.4673,314 "mean_token_accuracy": 0.3393166959285736,315 "num_tokens": 1322079.0,316 "step": 340317 },318 {319 "epoch": 0.003447630492814153,320 "grad_norm": 12.494967460632324,321 "learning_rate": 9.982811099400113e-06,322 "loss": 1.3341,323 "mean_token_accuracy": 0.26767021380364897,324 "num_tokens": 1358597.0,325 "step": 350326 },327 {328 "epoch": 0.003546134221180272,329 "grad_norm": 12.256545066833496,330 "learning_rate": 9.982318580758283e-06,331 "loss": 1.432,332 "mean_token_accuracy": 0.3195224553346634,333 "num_tokens": 1397372.0,334 "step": 360335 },336 {337 "epoch": 0.0036446379495463904,338 "grad_norm": 8.974289894104004,339 "learning_rate": 9.981826062116452e-06,340 "loss": 1.5435,341 "mean_token_accuracy": 0.2692535579204559,342 "num_tokens": 1431942.0,343 "step": 370344 },345 {346 "epoch": 0.003743141677912509,347 "grad_norm": 18.903564453125,348 "learning_rate": 9.981333543474622e-06,349 "loss": 1.3261,350 "mean_token_accuracy": 0.3077515698969364,351 "num_tokens": 1472380.0,352 "step": 380353 },354 {355 "epoch": 0.003841645406278628,356 "grad_norm": 6.635251998901367,357 "learning_rate": 9.98084102483279e-06,358 "loss": 1.1202,359 "mean_token_accuracy": 0.23408697955310345,360 "num_tokens": 1509098.0,361 "step": 390362 },363 {364 "epoch": 0.003940149134644746,365 "grad_norm": 14.887081146240234,366 "learning_rate": 9.98034850619096e-06,367 "loss": 1.3723,368 "mean_token_accuracy": 0.22168226912617683,369 "num_tokens": 1544216.0,370 "step": 400371 },372 {373 "epoch": 0.004038652863010865,374 "grad_norm": 10.794672012329102,375 "learning_rate": 9.97985598754913e-06,376 "loss": 1.4108,377 "mean_token_accuracy": 0.3656989440321922,378 "num_tokens": 1583648.0,379 "step": 410380 },381 {382 "epoch": 0.004137156591376983,383 "grad_norm": 70.41635131835938,384 "learning_rate": 9.979363468907299e-06,385 "loss": 1.1106,386 "mean_token_accuracy": 0.23466279059648515,387 "num_tokens": 1623687.0,388 "step": 420389 },390 {391 "epoch": 0.004235660319743102,392 "grad_norm": 7.897043228149414,393 "learning_rate": 9.978870950265469e-06,394 "loss": 1.4343,395 "mean_token_accuracy": 0.2966897919774055,396 "num_tokens": 1659415.0,397 "step": 430398 },399 {400 "epoch": 0.004334164048109221,401 "grad_norm": 10.63398551940918,402 "learning_rate": 9.978378431623638e-06,403 "loss": 1.2922,404 "mean_token_accuracy": 0.24354422688484192,405 "num_tokens": 1694361.0,406 "step": 440407 },408 {409 "epoch": 0.004432667776475339,410 "grad_norm": 11.589272499084473,411 "learning_rate": 9.977885912981808e-06,412 "loss": 1.5229,413 "mean_token_accuracy": 0.28925437927246095,414 "num_tokens": 1733256.0,415 "step": 450416 },417 {418 "epoch": 0.004531171504841458,419 "grad_norm": 8.766853332519531,420 "learning_rate": 9.977393394339978e-06,421 "loss": 1.4083,422 "mean_token_accuracy": 0.32210655957460405,423 "num_tokens": 1775735.0,424 "step": 460425 },426 {427 "epoch": 0.004629675233207577,428 "grad_norm": 13.658514022827148,429 "learning_rate": 9.976900875698146e-06,430 "loss": 1.2992,431 "mean_token_accuracy": 0.2186996452510357,432 "num_tokens": 1812083.0,433 "step": 470434 },435 {436 "epoch": 0.004728178961573695,437 "grad_norm": 15.588377952575684,438 "learning_rate": 9.976408357056315e-06,439 "loss": 1.2571,440 "mean_token_accuracy": 0.2709283992648125,441 "num_tokens": 1849403.0,442 "step": 480443 },444 {445 "epoch": 0.004826682689939814,446 "grad_norm": 7.753566741943359,447 "learning_rate": 9.975915838414485e-06,448 "loss": 1.2993,449 "mean_token_accuracy": 0.31253829821944235,450 "num_tokens": 1886632.0,451 "step": 490452 },453 {454 "epoch": 0.004925186418305933,455 "grad_norm": 11.971506118774414,456 "learning_rate": 9.975423319772655e-06,457 "loss": 1.352,458 "mean_token_accuracy": 0.2921320587396622,459 "num_tokens": 1922678.0,460 "step": 500461 },462 {463 "epoch": 0.005023690146672051,464 "grad_norm": 0.804463267326355,465 "learning_rate": 9.974930801130823e-06,466 "loss": 1.2679,467 "mean_token_accuracy": 0.291484996303916,468 "num_tokens": 1971776.0,469 "step": 510470 },471 {472 "epoch": 0.00512219387503817,473 "grad_norm": 10.847861289978027,474 "learning_rate": 9.974438282488992e-06,475 "loss": 1.3743,476 "mean_token_accuracy": 0.31155874878168105,477 "num_tokens": 2007163.0,478 "step": 520479 },480 {481 "epoch": 0.005220697603404289,482 "grad_norm": 8.9625825881958,483 "learning_rate": 9.973945763847162e-06,484 "loss": 0.9257,485 "mean_token_accuracy": 0.1884881116449833,486 "num_tokens": 2044217.0,487 "step": 530488 },489 {490 "epoch": 0.005319201331770407,491 "grad_norm": 11.655451774597168,492 "learning_rate": 9.973453245205332e-06,493 "loss": 1.4013,494 "mean_token_accuracy": 0.2849789783358574,495 "num_tokens": 2082121.0,496 "step": 540497 },498 {499 "epoch": 0.005417705060136526,500 "grad_norm": 10.9754056930542,501 "learning_rate": 9.972960726563501e-06,502 "loss": 1.3047,503 "mean_token_accuracy": 0.30112670212984083,504 "num_tokens": 2117744.0,505 "step": 550506 },507 {508 "epoch": 0.005516208788502645,509 "grad_norm": 13.801913261413574,510 "learning_rate": 9.972468207921671e-06,511 "loss": 1.5113,512 "mean_token_accuracy": 0.3287483021616936,513 "num_tokens": 2155805.0,514 "step": 560515 },516 {517 "epoch": 0.005614712516868763,518 "grad_norm": 5.056882381439209,519 "learning_rate": 9.97197568927984e-06,520 "loss": 1.1955,521 "mean_token_accuracy": 0.2325913280248642,522 "num_tokens": 2194559.0,523 "step": 570524 },525 {526 "epoch": 0.005713216245234882,527 "grad_norm": 6.317229747772217,528 "learning_rate": 9.97148317063801e-06,529 "loss": 1.3083,530 "mean_token_accuracy": 0.2851055234670639,531 "num_tokens": 2229409.0,532 "step": 580533 },534 {535 "epoch": 0.005811719973601001,536 "grad_norm": 9.799811363220215,537 "learning_rate": 9.970990651996178e-06,538 "loss": 1.3791,539 "mean_token_accuracy": 0.2873207736760378,540 "num_tokens": 2265482.0,541 "step": 590542 },543 {544 "epoch": 0.00591022370196712,545 "grad_norm": 12.647117614746094,546 "learning_rate": 9.970498133354348e-06,547 "loss": 1.4659,548 "mean_token_accuracy": 0.28821760416030884,549 "num_tokens": 2303177.0,550 "step": 600551 },552 {553 "epoch": 0.006008727430333238,554 "grad_norm": 7.061530113220215,555 "learning_rate": 9.970005614712518e-06,556 "loss": 1.1316,557 "mean_token_accuracy": 0.2284083217382431,558 "num_tokens": 2341123.0,559 "step": 610560 },561 {562 "epoch": 0.006107231158699357,563 "grad_norm": 8.10919189453125,564 "learning_rate": 9.969513096070688e-06,565 "loss": 0.9361,566 "mean_token_accuracy": 0.19455431252717972,567 "num_tokens": 2379263.0,568 "step": 620569 },570 {571 "epoch": 0.006205734887065476,572 "grad_norm": 5.393375396728516,573 "learning_rate": 9.969020577428856e-06,574 "loss": 1.4136,575 "mean_token_accuracy": 0.294439297914505,576 "num_tokens": 2417198.0,577 "step": 630578 },579 {580 "epoch": 0.006304238615431594,581 "grad_norm": 11.77549934387207,582 "learning_rate": 9.968528058787025e-06,583 "loss": 1.2196,584 "mean_token_accuracy": 0.2791116774082184,585 "num_tokens": 2452615.0,586 "step": 640587 },588 {589 "epoch": 0.006402742343797713,590 "grad_norm": 19.36256980895996,591 "learning_rate": 9.968035540145195e-06,592 "loss": 1.1888,593 "mean_token_accuracy": 0.24847980588674545,594 "num_tokens": 2489754.0,595 "step": 650596 },597 {598 "epoch": 0.006501246072163832,599 "grad_norm": 12.459017753601074,600 "learning_rate": 9.967543021503365e-06,601 "loss": 1.1242,602 "mean_token_accuracy": 0.272831392288208,603 "num_tokens": 2524523.0,604 "step": 660605 },606 {607 "epoch": 0.00659974980052995,608 "grad_norm": 17.65422821044922,609 "learning_rate": 9.967050502861534e-06,610 "loss": 1.4547,611 "mean_token_accuracy": 0.3275242060422897,612 "num_tokens": 2563531.0,613 "step": 670614 },615 {616 "epoch": 0.006698253528896069,617 "grad_norm": 10.63198184967041,618 "learning_rate": 9.966557984219704e-06,619 "loss": 1.3766,620 "mean_token_accuracy": 0.2821400508284569,621 "num_tokens": 2598161.0,622 "step": 680623 },624 {625 "epoch": 0.006796757257262188,626 "grad_norm": 28.730846405029297,627 "learning_rate": 9.966065465577874e-06,628 "loss": 1.5358,629 "mean_token_accuracy": 0.32285157442092893,630 "num_tokens": 2632555.0,631 "step": 690632 },633 {634 "epoch": 0.006895260985628306,635 "grad_norm": 6.657609462738037,636 "learning_rate": 9.965572946936042e-06,637 "loss": 1.3745,638 "mean_token_accuracy": 0.34917974919080735,639 "num_tokens": 2670096.0,640 "step": 700641 },642 {643 "epoch": 0.006993764713994425,644 "grad_norm": 30.633703231811523,645 "learning_rate": 9.965080428294211e-06,646 "loss": 1.3555,647 "mean_token_accuracy": 0.3193323977291584,648 "num_tokens": 2703786.0,649 "step": 710650 },651 {652 "epoch": 0.007092268442360544,653 "grad_norm": 7.800205707550049,654 "learning_rate": 9.964587909652381e-06,655 "loss": 1.1842,656 "mean_token_accuracy": 0.25614908784627916,657 "num_tokens": 2740345.0,658 "step": 720659 },660 {661 "epoch": 0.007190772170726662,662 "grad_norm": 0.728541374206543,663 "learning_rate": 9.96409539101055e-06,664 "loss": 1.1098,665 "mean_token_accuracy": 0.23233613818883897,666 "num_tokens": 2782060.0,667 "step": 730668 },669 {670 "epoch": 0.007289275899092781,671 "grad_norm": 8.87043571472168,672 "learning_rate": 9.96360287236872e-06,673 "loss": 1.3956,674 "mean_token_accuracy": 0.26713471710681913,675 "num_tokens": 2817938.0,676 "step": 740677 },678 {679 "epoch": 0.0073877796274589,680 "grad_norm": 7.2556047439575195,681 "learning_rate": 9.96311035372689e-06,682 "loss": 1.1512,683 "mean_token_accuracy": 0.2411678448319435,684 "num_tokens": 2851198.0,685 "step": 750686 },687 {688 "epoch": 0.007486283355825018,689 "grad_norm": 12.792369842529297,690 "learning_rate": 9.96261783508506e-06,691 "loss": 1.1543,692 "mean_token_accuracy": 0.23933272659778596,693 "num_tokens": 2888736.0,694 "step": 760695 },696 {697 "epoch": 0.007584787084191137,698 "grad_norm": 15.33095932006836,699 "learning_rate": 9.96212531644323e-06,700 "loss": 1.2687,701 "mean_token_accuracy": 0.26932157687842845,702 "num_tokens": 2925690.0,703 "step": 770704 },705 {706 "epoch": 0.007683290812557256,707 "grad_norm": 8.721963882446289,708 "learning_rate": 9.961632797801397e-06,709 "loss": 1.314,710 "mean_token_accuracy": 0.28175555542111397,711 "num_tokens": 2962167.0,712 "step": 780713 },714 {715 "epoch": 0.007781794540923374,716 "grad_norm": 16.84264373779297,717 "learning_rate": 9.961140279159567e-06,718 "loss": 1.4362,719 "mean_token_accuracy": 0.3012395367026329,720 "num_tokens": 2997717.0,721 "step": 790722 },723 {724 "epoch": 0.007880298269289493,725 "grad_norm": 14.453497886657715,726 "learning_rate": 9.960647760517737e-06,727 "loss": 1.3445,728 "mean_token_accuracy": 0.24600916355848312,729 "num_tokens": 3032512.0,730 "step": 800731 },732 {733 "epoch": 0.007978801997655612,734 "grad_norm": 8.368857383728027,735 "learning_rate": 9.960155241875906e-06,736 "loss": 1.1994,737 "mean_token_accuracy": 0.24641867950558663,738 "num_tokens": 3070387.0,739 "step": 810740 },741 {742 "epoch": 0.00807730572602173,743 "grad_norm": 10.384809494018555,744 "learning_rate": 9.959662723234074e-06,745 "loss": 1.2123,746 "mean_token_accuracy": 0.24860502257943154,747 "num_tokens": 3106333.0,748 "step": 820749 },750 {751 "epoch": 0.00817580945438785,752 "grad_norm": 8.294401168823242,753 "learning_rate": 9.959170204592244e-06,754 "loss": 1.2495,755 "mean_token_accuracy": 0.29464175552129745,756 "num_tokens": 3142933.0,757 "step": 830758 },759 {760 "epoch": 0.008274313182753967,761 "grad_norm": 6.105022430419922,762 "learning_rate": 9.958677685950414e-06,763 "loss": 1.2152,764 "mean_token_accuracy": 0.2727929688990116,765 "num_tokens": 3183607.0,766 "step": 840767 },768 {769 "epoch": 0.008372816911120086,770 "grad_norm": 6.826088905334473,771 "learning_rate": 9.958185167308583e-06,772 "loss": 1.1244,773 "mean_token_accuracy": 0.25121626555919646,774 "num_tokens": 3218883.0,775 "step": 850776 },777 {778 "epoch": 0.008471320639486205,779 "grad_norm": 6.379848480224609,780 "learning_rate": 9.957692648666753e-06,781 "loss": 1.3046,782 "mean_token_accuracy": 0.2697582277469337,783 "num_tokens": 3262073.0,784 "step": 860785 },786 {787 "epoch": 0.008569824367852323,788 "grad_norm": 12.71320915222168,789 "learning_rate": 9.957200130024923e-06,790 "loss": 1.3316,791 "mean_token_accuracy": 0.295292516797781,792 "num_tokens": 3298099.0,793 "step": 870794 },795 {796 "epoch": 0.008668328096218442,797 "grad_norm": 14.52015209197998,798 "learning_rate": 9.956707611383092e-06,799 "loss": 1.6632,800 "mean_token_accuracy": 0.37425926327705383,801 "num_tokens": 3335198.0,802 "step": 880803 },804 {805 "epoch": 0.008766831824584561,806 "grad_norm": 7.63631010055542,807 "learning_rate": 9.956215092741262e-06,808 "loss": 1.2336,809 "mean_token_accuracy": 0.2655528262257576,810 "num_tokens": 3370590.0,811 "step": 890812 },813 {814 "epoch": 0.008865335552950679,815 "grad_norm": 3.4239141941070557,816 "learning_rate": 9.95572257409943e-06,817 "loss": 1.4429,818 "mean_token_accuracy": 0.3862385615706444,819 "num_tokens": 3412432.0,820 "step": 900821 },822 {823 "epoch": 0.008963839281316797,824 "grad_norm": 8.51039981842041,825 "learning_rate": 9.9552300554576e-06,826 "loss": 1.2388,827 "mean_token_accuracy": 0.35555260330438615,828 "num_tokens": 3447198.0,829 "step": 910830 },831 {832 "epoch": 0.009062343009682916,833 "grad_norm": 1.027651071548462,834 "learning_rate": 9.95473753681577e-06,835 "loss": 1.3361,836 "mean_token_accuracy": 0.2850140180438757,837 "num_tokens": 3488184.0,838 "step": 920839 },840 {841 "epoch": 0.009160846738049035,842 "grad_norm": 21.422971725463867,843 "learning_rate": 9.954245018173939e-06,844 "loss": 1.2932,845 "mean_token_accuracy": 0.2794327441602945,846 "num_tokens": 3526308.0,847 "step": 930848 },849 {850 "epoch": 0.009259350466415154,851 "grad_norm": 13.776223182678223,852 "learning_rate": 9.953752499532107e-06,853 "loss": 1.187,854 "mean_token_accuracy": 0.2856494963169098,855 "num_tokens": 3566052.0,856 "step": 940857 },858 {859 "epoch": 0.009357854194781273,860 "grad_norm": 1.5548714399337769,861 "learning_rate": 9.953259980890277e-06,862 "loss": 1.3606,863 "mean_token_accuracy": 0.2707196932286024,864 "num_tokens": 3607576.0,865 "step": 950866 },867 {868 "epoch": 0.00945635792314739,869 "grad_norm": 7.968335151672363,870 "learning_rate": 9.952767462248446e-06,871 "loss": 1.3907,872 "mean_token_accuracy": 0.3057244636118412,873 "num_tokens": 3644951.0,874 "step": 960875 },876 {877 "epoch": 0.00955486165151351,878 "grad_norm": 12.585286140441895,879 "learning_rate": 9.952274943606616e-06,880 "loss": 1.4397,881 "mean_token_accuracy": 0.306073372066021,882 "num_tokens": 3678537.0,883 "step": 970884 },885 {886 "epoch": 0.009653365379879628,887 "grad_norm": 12.399375915527344,888 "learning_rate": 9.951782424964786e-06,889 "loss": 1.1529,890 "mean_token_accuracy": 0.24479719083756207,891 "num_tokens": 3719074.0,892 "step": 980893 },894 {895 "epoch": 0.009751869108245747,896 "grad_norm": 9.571209907531738,897 "learning_rate": 9.951289906322955e-06,898 "loss": 1.1512,899 "mean_token_accuracy": 0.28902862444519994,900 "num_tokens": 3756663.0,901 "step": 990902 },903 {904 "epoch": 0.009850372836611866,905 "grad_norm": 3.028090238571167,906 "learning_rate": 9.950797387681125e-06,907 "loss": 1.3075,908 "mean_token_accuracy": 0.3854308143258095,909 "num_tokens": 3805581.0,910 "step": 1000911 },912 {913 "epoch": 0.009948876564977985,914 "grad_norm": 15.040968894958496,915 "learning_rate": 9.950304869039295e-06,916 "loss": 1.5773,917 "mean_token_accuracy": 0.3517814606428146,918 "num_tokens": 3840646.0,919 "step": 1010920 },921 {922 "epoch": 0.010047380293344102,923 "grad_norm": 1.7873173952102661,924 "learning_rate": 9.949812350397463e-06,925 "loss": 1.2817,926 "mean_token_accuracy": 0.2500580433756113,927 "num_tokens": 3879331.0,928 "step": 1020929 },930 {931 "epoch": 0.010145884021710221,932 "grad_norm": 7.113542556762695,933 "learning_rate": 9.949319831755632e-06,934 "loss": 1.0864,935 "mean_token_accuracy": 0.24927083402872086,936 "num_tokens": 3915701.0,937 "step": 1030938 },939 {940 "epoch": 0.01024438775007634,941 "grad_norm": 12.868072509765625,942 "learning_rate": 9.948827313113802e-06,943 "loss": 1.3924,944 "mean_token_accuracy": 0.36228409707546233,945 "num_tokens": 3953147.0,946 "step": 1040947 },948 {949 "epoch": 0.01034289147844246,950 "grad_norm": 12.542919158935547,951 "learning_rate": 9.948334794471972e-06,952 "loss": 1.2732,953 "mean_token_accuracy": 0.26514705270528793,954 "num_tokens": 3989771.0,955 "step": 1050956 },957 {958 "epoch": 0.010441395206808578,959 "grad_norm": 8.249795913696289,960 "learning_rate": 9.947842275830141e-06,961 "loss": 1.2483,962 "mean_token_accuracy": 0.2580167345702648,963 "num_tokens": 4034248.0,964 "step": 1060965 },966 {967 "epoch": 0.010539898935174697,968 "grad_norm": 11.386072158813477,969 "learning_rate": 9.947349757188311e-06,970 "loss": 1.3869,971 "mean_token_accuracy": 0.32561925053596497,972 "num_tokens": 4068719.0,973 "step": 1070974 },975 {976 "epoch": 0.010638402663540814,977 "grad_norm": 6.921145915985107,978 "learning_rate": 9.94685723854648e-06,979 "loss": 1.1179,980 "mean_token_accuracy": 0.2305791638791561,981 "num_tokens": 4103703.0,982 "step": 1080983 },984 {985 "epoch": 0.010736906391906933,986 "grad_norm": 6.271225452423096,987 "learning_rate": 9.946364719904649e-06,988 "loss": 1.43,989 "mean_token_accuracy": 0.31061185747385023,990 "num_tokens": 4146702.0,991 "step": 1090992 },993 {994 "epoch": 0.010835410120273052,995 "grad_norm": 9.407719612121582,996 "learning_rate": 9.945872201262818e-06,997 "loss": 1.4472,998 "mean_token_accuracy": 0.30268693789839746,999 "num_tokens": 4182020.0,1000 "step": 11001001 },1002 {1003 "epoch": 0.010933913848639171,1004 "grad_norm": 9.300726890563965,1005 "learning_rate": 9.945379682620988e-06,1006 "loss": 1.262,1007 "mean_token_accuracy": 0.23403102159500122,1008 "num_tokens": 4216975.0,1009 "step": 11101010 },1011 {1012 "epoch": 0.01103241757700529,1013 "grad_norm": 10.964862823486328,1014 "learning_rate": 9.944887163979158e-06,1015 "loss": 1.3479,1016 "mean_token_accuracy": 0.2730256706476212,1017 "num_tokens": 4258216.0,1018 "step": 11201019 },1020 {1021 "epoch": 0.011130921305371409,1022 "grad_norm": 9.638137817382812,1023 "learning_rate": 9.944394645337326e-06,1024 "loss": 1.2431,1025 "mean_token_accuracy": 0.26802519112825396,1026 "num_tokens": 4294499.0,1027 "step": 11301028 },1029 {1030 "epoch": 0.011229425033737526,1031 "grad_norm": 15.855734825134277,1032 "learning_rate": 9.943902126695495e-06,1033 "loss": 1.3849,1034 "mean_token_accuracy": 0.31018309742212297,1035 "num_tokens": 4329986.0,1036 "step": 11401037 },1038 {1039 "epoch": 0.011327928762103645,1040 "grad_norm": 9.584433555603027,1041 "learning_rate": 9.943409608053665e-06,1042 "loss": 1.2969,1043 "mean_token_accuracy": 0.23223277740180492,1044 "num_tokens": 4364286.0,1045 "step": 11501046 },1047 {1048 "epoch": 0.011426432490469764,1049 "grad_norm": 8.683856010437012,1050 "learning_rate": 9.942917089411835e-06,1051 "loss": 1.2686,1052 "mean_token_accuracy": 0.23580990321934223,1053 "num_tokens": 4400284.0,1054 "step": 11601055 },1056 {1057 "epoch": 0.011524936218835883,1058 "grad_norm": 21.125038146972656,1059 "learning_rate": 9.942424570770004e-06,1060 "loss": 1.2729,1061 "mean_token_accuracy": 0.2784357912838459,1062 "num_tokens": 4436224.0,1063 "step": 11701064 },1065 {1066 "epoch": 0.011623439947202002,1067 "grad_norm": 11.860121726989746,1068 "learning_rate": 9.941932052128174e-06,1069 "loss": 1.3153,1070 "mean_token_accuracy": 0.24891949743032454,1071 "num_tokens": 4470181.0,1072 "step": 11801073 },1074 {1075 "epoch": 0.01172194367556812,1076 "grad_norm": 11.686821937561035,1077 "learning_rate": 9.941439533486344e-06,1078 "loss": 1.0919,1079 "mean_token_accuracy": 0.27350951805710794,1080 "num_tokens": 4526046.0,1081 "step": 11901082 },1083 {1084 "epoch": 0.01182044740393424,1085 "grad_norm": 12.6265230178833,1086 "learning_rate": 9.940947014844514e-06,1087 "loss": 1.0927,1088 "mean_token_accuracy": 0.2677152242511511,1089 "num_tokens": 4570578.0,1090 "step": 12001091 },1092 {1093 "epoch": 0.011918951132300357,1094 "grad_norm": 6.5598039627075195,1095 "learning_rate": 9.940454496202681e-06,1096 "loss": 1.4089,1097 "mean_token_accuracy": 0.32097743451595306,1098 "num_tokens": 4613970.0,1099 "step": 12101100 },1101 {1102 "epoch": 0.012017454860666476,1103 "grad_norm": 1.3488277196884155,1104 "learning_rate": 9.939961977560851e-06,1105 "loss": 0.8991,1106 "mean_token_accuracy": 0.24842826463282108,1107 "num_tokens": 4663299.0,1108 "step": 12201109 },1110 {1111 "epoch": 0.012115958589032595,1112 "grad_norm": 8.285106658935547,1113 "learning_rate": 9.93946945891902e-06,1114 "loss": 1.3577,1115 "mean_token_accuracy": 0.33487610667943957,1116 "num_tokens": 4697957.0,1117 "step": 12301118 },1119 {1120 "epoch": 0.012214462317398714,1121 "grad_norm": 14.000260353088379,1122 "learning_rate": 9.93897694027719e-06,1123 "loss": 1.2255,1124 "mean_token_accuracy": 0.25658253878355025,1125 "num_tokens": 4732944.0,1126 "step": 12401127 },1128 {1129 "epoch": 0.012312966045764833,1130 "grad_norm": 24.141399383544922,1131 "learning_rate": 9.938484421635358e-06,1132 "loss": 1.2675,1133 "mean_token_accuracy": 0.2739123493432999,1134 "num_tokens": 4771206.0,1135 "step": 12501136 },1137 {1138 "epoch": 0.012411469774130952,1139 "grad_norm": 6.599194526672363,1140 "learning_rate": 9.937991902993528e-06,1141 "loss": 1.3218,1142 "mean_token_accuracy": 0.24770741313695907,1143 "num_tokens": 4806076.0,1144 "step": 12601145 },1146 {1147 "epoch": 0.012509973502497069,1148 "grad_norm": 7.697349548339844,1149 "learning_rate": 9.9374993843517e-06,1150 "loss": 1.4019,1151 "mean_token_accuracy": 0.288208170235157,1152 "num_tokens": 4839326.0,1153 "step": 12701154 },1155 {1156 "epoch": 0.012608477230863188,1157 "grad_norm": 18.712209701538086,1158 "learning_rate": 9.93700686570987e-06,1159 "loss": 1.1979,1160 "mean_token_accuracy": 0.22817165050655602,1161 "num_tokens": 4880644.0,1162 "step": 12801163 },1164 {1165 "epoch": 0.012706980959229307,1166 "grad_norm": 10.353955268859863,1167 "learning_rate": 9.936514347068037e-06,1168 "loss": 1.2524,1169 "mean_token_accuracy": 0.27737057507038115,1170 "num_tokens": 4918863.0,1171 "step": 12901172 },1173 {1174 "epoch": 0.012805484687595426,1175 "grad_norm": 11.71806812286377,1176 "learning_rate": 9.936021828426207e-06,1177 "loss": 1.1717,1178 "mean_token_accuracy": 0.26971509531140325,1179 "num_tokens": 4960740.0,1180 "step": 13001181 },1182 {1183 "epoch": 0.012903988415961545,1184 "grad_norm": 2.740062713623047,1185 "learning_rate": 9.935529309784377e-06,1186 "loss": 1.2208,1187 "mean_token_accuracy": 0.28350216168910264,1188 "num_tokens": 5006035.0,1189 "step": 13101190 },1191 {1192 "epoch": 0.013002492144327664,1193 "grad_norm": 7.498782634735107,1194 "learning_rate": 9.935036791142546e-06,1195 "loss": 1.2331,1196 "mean_token_accuracy": 0.2826507996767759,1197 "num_tokens": 5054647.0,1198 "step": 13201199 },1200 {