synthos / trainer_state.json
tzar z
Sovereign Genesis - checkpoint-6500 - Weights Only Search
05c2caf verified
Raw History Blame Contribute Delete
199 kB
{ "best_global_step": null, "best_metric": null, "best_model_checkpoint": null, "epoch": 3.070382616910723, "eval_steps": 500, "global_step": 6500, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "entropy": 5.8557036250829695, "epoch": 0.004723665564478035, "grad_norm": 1.1635308265686035, "learning_rate": 3.6e-05, "loss": 9.016449737548829, "mean_token_accuracy": 0.030926086311228573, "num_tokens": 134075.0, "step": 10 }, { "entropy": 6.803842556476593, "epoch": 0.00944733112895607, "grad_norm": 0.6144386529922485, "learning_rate": 7.6e-05, "loss": 6.777875518798828, "mean_token_accuracy": 0.09398306957446038, "num_tokens": 266103.0, "step": 20 }, { "entropy": 5.3631006419658664, "epoch": 0.014170996693434105, "grad_norm": 0.3949488699436188, "learning_rate": 0.000116, "loss": 5.411441421508789, "mean_token_accuracy": 0.1942991679534316, "num_tokens": 406925.0, "step": 30 }, { "entropy": 4.53178850710392, "epoch": 0.01889466225791214, "grad_norm": 0.2796455919742584, "learning_rate": 0.00015600000000000002, "loss": 4.523899078369141, "mean_token_accuracy": 0.2943518474698067, "num_tokens": 549837.0, "step": 40 }, { "entropy": 3.7455823540687563, "epoch": 0.023618327822390175, "grad_norm": 0.32603198289871216, "learning_rate": 0.000196, "loss": 3.8331756591796875, "mean_token_accuracy": 0.35103216096758844, "num_tokens": 690318.0, "step": 50 }, { "entropy": 3.1043672770261765, "epoch": 0.02834199338686821, "grad_norm": 0.31668633222579956, "learning_rate": 0.00019999991038785437, "loss": 3.1435787200927736, "mean_token_accuracy": 0.42379639223217963, "num_tokens": 824090.0, "step": 60 }, { "entropy": 2.728278709948063, "epoch": 0.03306565895134624, "grad_norm": 0.34853872656822205, "learning_rate": 0.00019999960061768058, "loss": 2.771727180480957, "mean_token_accuracy": 0.48015250265598297, "num_tokens": 958789.0, "step": 70 }, { "entropy": 2.440419462323189, "epoch": 0.03778932451582428, "grad_norm": 0.278272420167923, "learning_rate": 0.00019999906958384112, "loss": 2.5111177444458006, "mean_token_accuracy": 0.5124442256987095, "num_tokens": 1101049.0, "step": 80 }, { "entropy": 2.25365926027298, "epoch": 0.042512990080302314, "grad_norm": 0.20885923504829407, "learning_rate": 0.00019999831728751097, "loss": 2.28131160736084, "mean_token_accuracy": 0.5472530923783779, "num_tokens": 1243306.0, "step": 90 }, { "entropy": 2.0528148651123046, "epoch": 0.04723665564478035, "grad_norm": 0.1696874350309372, "learning_rate": 0.0001999973437303547, "loss": 2.124062347412109, "mean_token_accuracy": 0.5843483690172434, "num_tokens": 1376760.0, "step": 100 }, { "entropy": 1.857421062886715, "epoch": 0.051960321209258385, "grad_norm": 0.16955959796905518, "learning_rate": 0.0001999961489145264, "loss": 1.8606906890869142, "mean_token_accuracy": 0.6146147038787604, "num_tokens": 1513709.0, "step": 110 }, { "entropy": 1.8045475505292417, "epoch": 0.05668398677373642, "grad_norm": 0.14853711426258087, "learning_rate": 0.00019999473284266982, "loss": 1.8547624588012694, "mean_token_accuracy": 0.6264711648225785, "num_tokens": 1652555.0, "step": 120 }, { "entropy": 1.5771285906434058, "epoch": 0.061407652338214457, "grad_norm": 0.1552414894104004, "learning_rate": 0.00019999309551791826, "loss": 1.6219715118408202, "mean_token_accuracy": 0.658776619285345, "num_tokens": 1786366.0, "step": 130 }, { "entropy": 1.5116388112306596, "epoch": 0.06613131790269249, "grad_norm": 0.15822529792785645, "learning_rate": 0.00019999123694389447, "loss": 1.5329365730285645, "mean_token_accuracy": 0.6778819467872381, "num_tokens": 1921087.0, "step": 140 }, { "entropy": 1.5189390406012535, "epoch": 0.07085498346717052, "grad_norm": 0.13936467468738556, "learning_rate": 0.0001999891571247108, "loss": 1.553725814819336, "mean_token_accuracy": 0.6716477766633033, "num_tokens": 2062223.0, "step": 150 }, { "entropy": 1.4038615018129348, "epoch": 0.07557864903164856, "grad_norm": 0.1497100293636322, "learning_rate": 0.00019998685606496924, "loss": 1.4266416549682617, "mean_token_accuracy": 0.698700550198555, "num_tokens": 2207314.0, "step": 160 }, { "entropy": 1.3689204968512059, "epoch": 0.08030231459612659, "grad_norm": 0.10406970977783203, "learning_rate": 0.00019998433376976118, "loss": 1.3864148139953614, "mean_token_accuracy": 0.7021394304931163, "num_tokens": 2356214.0, "step": 170 }, { "entropy": 1.374360391125083, "epoch": 0.08502598016060463, "grad_norm": 0.12781788408756256, "learning_rate": 0.00019998159024466755, "loss": 1.4181779861450194, "mean_token_accuracy": 0.6961348332464695, "num_tokens": 2497950.0, "step": 180 }, { "entropy": 1.2449575688689947, "epoch": 0.08974964572508266, "grad_norm": 0.11341937631368637, "learning_rate": 0.00019997862549575877, "loss": 1.2644970893859864, "mean_token_accuracy": 0.7251733422279358, "num_tokens": 2641877.0, "step": 190 }, { "entropy": 1.2353131141513587, "epoch": 0.0944733112895607, "grad_norm": 0.0990600511431694, "learning_rate": 0.00019997543952959487, "loss": 1.2454343795776368, "mean_token_accuracy": 0.7267829045653343, "num_tokens": 2773328.0, "step": 200 }, { "entropy": 1.3188120108097792, "epoch": 0.09919697685403873, "grad_norm": 0.11764945089817047, "learning_rate": 0.0001999720323532252, "loss": 1.311232852935791, "mean_token_accuracy": 0.711747232079506, "num_tokens": 2916683.0, "step": 210 }, { "entropy": 1.1950139990076423, "epoch": 0.10392064241851677, "grad_norm": 0.6117010712623596, "learning_rate": 0.00019996840397418862, "loss": 1.2524967193603516, "mean_token_accuracy": 0.7309570647776127, "num_tokens": 3056938.0, "step": 220 }, { "entropy": 1.3037370018661023, "epoch": 0.1086443079829948, "grad_norm": 0.1238599494099617, "learning_rate": 0.0001999645544005135, "loss": 1.3111289024353028, "mean_token_accuracy": 0.7115058593451977, "num_tokens": 3195628.0, "step": 230 }, { "entropy": 1.2695199608802796, "epoch": 0.11336797354747284, "grad_norm": 0.11078216880559921, "learning_rate": 0.00019996048364071754, "loss": 1.3134763717651368, "mean_token_accuracy": 0.7214690744876862, "num_tokens": 3331492.0, "step": 240 }, { "entropy": 1.2206239182502032, "epoch": 0.11809163911195088, "grad_norm": 0.10117341578006744, "learning_rate": 0.00019995619170380788, "loss": 1.251480770111084, "mean_token_accuracy": 0.7287674643099308, "num_tokens": 3480322.0, "step": 250 }, { "entropy": 1.2233508143574, "epoch": 0.12281530467642891, "grad_norm": 0.08893683552742004, "learning_rate": 0.00019995167859928107, "loss": 1.2418370246887207, "mean_token_accuracy": 0.7306129030883313, "num_tokens": 3626738.0, "step": 260 }, { "entropy": 1.1926977697759866, "epoch": 0.12753897024090693, "grad_norm": 0.2707843482494354, "learning_rate": 0.00019994694433712304, "loss": 1.2434834480285644, "mean_token_accuracy": 0.7310510978102684, "num_tokens": 3763571.0, "step": 270 }, { "entropy": 1.2124732583761215, "epoch": 0.13226263580538497, "grad_norm": 0.09966970235109329, "learning_rate": 0.000199941988927809, "loss": 1.235425090789795, "mean_token_accuracy": 0.7312963537871837, "num_tokens": 3902958.0, "step": 280 }, { "entropy": 1.1035200484097003, "epoch": 0.136986301369863, "grad_norm": 0.08873643726110458, "learning_rate": 0.0001999368123823035, "loss": 1.136031436920166, "mean_token_accuracy": 0.7455673456192017, "num_tokens": 4053435.0, "step": 290 }, { "entropy": 1.1018408123403787, "epoch": 0.14170996693434104, "grad_norm": 0.08802961558103561, "learning_rate": 0.00019993141471206046, "loss": 1.1224894523620605, "mean_token_accuracy": 0.752431484311819, "num_tokens": 4192717.0, "step": 300 }, { "entropy": 1.0678345449268818, "epoch": 0.14643363249881908, "grad_norm": 0.09011015295982361, "learning_rate": 0.00019992579592902296, "loss": 1.0647945404052734, "mean_token_accuracy": 0.7590971812605858, "num_tokens": 4335397.0, "step": 310 }, { "entropy": 1.123979426175356, "epoch": 0.1511572980632971, "grad_norm": 0.09066376835107803, "learning_rate": 0.0001999199560456234, "loss": 1.137593936920166, "mean_token_accuracy": 0.7412543177604676, "num_tokens": 4485396.0, "step": 320 }, { "entropy": 0.9984387885779142, "epoch": 0.15588096362777515, "grad_norm": 0.09314711391925812, "learning_rate": 0.00019991389507478333, "loss": 1.00946044921875, "mean_token_accuracy": 0.7691142901778221, "num_tokens": 4632249.0, "step": 330 }, { "entropy": 1.2071824323385953, "epoch": 0.16060462919225318, "grad_norm": 0.08062185347080231, "learning_rate": 0.0001999076130299136, "loss": 1.2388092041015626, "mean_token_accuracy": 0.7264691971242427, "num_tokens": 4779005.0, "step": 340 }, { "entropy": 1.0473538793623447, "epoch": 0.16532829475673122, "grad_norm": 0.08660266548395157, "learning_rate": 0.00019990110992491408, "loss": 1.0670055389404296, "mean_token_accuracy": 0.7600623242557049, "num_tokens": 4919578.0, "step": 350 }, { "entropy": 1.0980668909847737, "epoch": 0.17005196032120926, "grad_norm": 0.10104081779718399, "learning_rate": 0.00019989438577417388, "loss": 1.1178273200988769, "mean_token_accuracy": 0.7528468511998654, "num_tokens": 5055566.0, "step": 360 }, { "entropy": 1.0160497423261403, "epoch": 0.1747756258856873, "grad_norm": 0.07290609925985336, "learning_rate": 0.00019988744059257113, "loss": 1.0367406845092773, "mean_token_accuracy": 0.7629740111529827, "num_tokens": 5197101.0, "step": 370 }, { "entropy": 1.0185628725215792, "epoch": 0.17949929145016533, "grad_norm": 0.08332032710313797, "learning_rate": 0.00019988027439547308, "loss": 1.0431556701660156, "mean_token_accuracy": 0.7625533670186997, "num_tokens": 5339798.0, "step": 380 }, { "entropy": 1.038666208833456, "epoch": 0.18422295701464336, "grad_norm": 0.08882896602153778, "learning_rate": 0.00019987288719873595, "loss": 1.0364270210266113, "mean_token_accuracy": 0.7640068434178829, "num_tokens": 5478802.0, "step": 390 }, { "entropy": 1.0126425851136447, "epoch": 0.1889466225791214, "grad_norm": 0.08613762259483337, "learning_rate": 0.00019986527901870503, "loss": 1.037272834777832, "mean_token_accuracy": 0.7654325738549232, "num_tokens": 5620754.0, "step": 400 }, { "entropy": 1.0018428541719913, "epoch": 0.19367028814359943, "grad_norm": 0.08839229494333267, "learning_rate": 0.00019985744987221452, "loss": 1.0412793159484863, "mean_token_accuracy": 0.7647483617067337, "num_tokens": 5759105.0, "step": 410 }, { "entropy": 1.0421779200434684, "epoch": 0.19839395370807747, "grad_norm": 0.09663446247577667, "learning_rate": 0.00019984939977658753, "loss": 1.0763293266296388, "mean_token_accuracy": 0.7581371806561947, "num_tokens": 5904005.0, "step": 420 }, { "entropy": 1.0125879775732756, "epoch": 0.2031176192725555, "grad_norm": 0.08860418200492859, "learning_rate": 0.00019984112874963606, "loss": 1.0376442909240722, "mean_token_accuracy": 0.7651007890701294, "num_tokens": 6043524.0, "step": 430 }, { "entropy": 0.9748837495222687, "epoch": 0.20784128483703354, "grad_norm": 0.1348097175359726, "learning_rate": 0.00019983263680966099, "loss": 0.991539478302002, "mean_token_accuracy": 0.7708897858858108, "num_tokens": 6185284.0, "step": 440 }, { "entropy": 1.0165104916319252, "epoch": 0.21256495040151158, "grad_norm": 0.09242016822099686, "learning_rate": 0.00019982392397545194, "loss": 1.0488774299621582, "mean_token_accuracy": 0.7631760179996491, "num_tokens": 6326506.0, "step": 450 }, { "entropy": 0.9810311056673526, "epoch": 0.2172886159659896, "grad_norm": 0.08120118826627731, "learning_rate": 0.00019981499026628733, "loss": 1.0122363090515136, "mean_token_accuracy": 0.7749755322933197, "num_tokens": 6466442.0, "step": 460 }, { "entropy": 0.9703505631536246, "epoch": 0.22201228153046765, "grad_norm": 0.08855105191469193, "learning_rate": 0.0001998058357019343, "loss": 0.9949516296386719, "mean_token_accuracy": 0.7743386216461658, "num_tokens": 6604118.0, "step": 470 }, { "entropy": 0.9631600104272365, "epoch": 0.22673594709494568, "grad_norm": 0.07770537585020065, "learning_rate": 0.00019979646030264867, "loss": 0.9812436103820801, "mean_token_accuracy": 0.7728767231106758, "num_tokens": 6743426.0, "step": 480 }, { "entropy": 0.9502388194203377, "epoch": 0.23145961265942372, "grad_norm": 0.06688424199819565, "learning_rate": 0.00019978686408917486, "loss": 0.9885772705078125, "mean_token_accuracy": 0.7820859834551811, "num_tokens": 6883757.0, "step": 490 }, { "entropy": 0.978575604595244, "epoch": 0.23618327822390175, "grad_norm": 0.08154246211051941, "learning_rate": 0.00019977704708274587, "loss": 1.0107855796813965, "mean_token_accuracy": 0.7707071267068386, "num_tokens": 7018127.0, "step": 500 }, { "entropy": 0.9492714935913682, "epoch": 0.2409069437883798, "grad_norm": 0.07454526424407959, "learning_rate": 0.0001997670093050833, "loss": 0.9670686721801758, "mean_token_accuracy": 0.7780260860919952, "num_tokens": 7154868.0, "step": 510 }, { "entropy": 0.9543494507670403, "epoch": 0.24563060935285783, "grad_norm": 0.06816540658473969, "learning_rate": 0.00019975675077839715, "loss": 0.9841312408447266, "mean_token_accuracy": 0.775021756440401, "num_tokens": 7297804.0, "step": 520 }, { "entropy": 0.8711556019261479, "epoch": 0.25035427491733586, "grad_norm": 0.06323365122079849, "learning_rate": 0.0001997462715253859, "loss": 0.8774194717407227, "mean_token_accuracy": 0.7944991394877434, "num_tokens": 7430552.0, "step": 530 }, { "entropy": 1.0319619249552487, "epoch": 0.25507794048181387, "grad_norm": 0.08967553824186325, "learning_rate": 0.00019973557156923645, "loss": 1.0639806747436524, "mean_token_accuracy": 0.7608748763799668, "num_tokens": 7573555.0, "step": 540 }, { "entropy": 0.9270606219768525, "epoch": 0.25980160604629193, "grad_norm": 0.07305469363927841, "learning_rate": 0.00019972465093362403, "loss": 0.9450544357299805, "mean_token_accuracy": 0.781608372926712, "num_tokens": 7716356.0, "step": 550 }, { "entropy": 0.9537868164479733, "epoch": 0.26452527161076994, "grad_norm": 0.07966265827417374, "learning_rate": 0.00019971350964271208, "loss": 0.9807086944580078, "mean_token_accuracy": 0.7759939216077327, "num_tokens": 7855886.0, "step": 560 }, { "entropy": 0.9782497521489859, "epoch": 0.269248937175248, "grad_norm": 0.1003744974732399, "learning_rate": 0.00019970214772115235, "loss": 0.9984622001647949, "mean_token_accuracy": 0.7716974891722203, "num_tokens": 7989011.0, "step": 570 }, { "entropy": 0.9414560319855809, "epoch": 0.273972602739726, "grad_norm": 0.07943235337734222, "learning_rate": 0.00019969056519408477, "loss": 0.9575101852416992, "mean_token_accuracy": 0.7781248845160007, "num_tokens": 8131777.0, "step": 580 }, { "entropy": 0.904687955789268, "epoch": 0.2786962683042041, "grad_norm": 0.07007589191198349, "learning_rate": 0.00019967876208713733, "loss": 0.9075414657592773, "mean_token_accuracy": 0.785995677113533, "num_tokens": 8267959.0, "step": 590 }, { "entropy": 0.9708761606365442, "epoch": 0.2834199338686821, "grad_norm": 0.07442475110292435, "learning_rate": 0.00019966673842642613, "loss": 0.9890885353088379, "mean_token_accuracy": 0.7723605215549469, "num_tokens": 8408214.0, "step": 600 }, { "entropy": 0.8597494594752788, "epoch": 0.28814359943316015, "grad_norm": 0.09197904169559479, "learning_rate": 0.00019965449423855528, "loss": 0.8936526298522949, "mean_token_accuracy": 0.7920610755681992, "num_tokens": 8540631.0, "step": 610 }, { "entropy": 0.8835750674828887, "epoch": 0.29286726499763815, "grad_norm": 0.0747399628162384, "learning_rate": 0.00019964202955061685, "loss": 0.889069938659668, "mean_token_accuracy": 0.7892274640500545, "num_tokens": 8681445.0, "step": 620 }, { "entropy": 0.9012997291982174, "epoch": 0.2975909305621162, "grad_norm": 0.0739719420671463, "learning_rate": 0.0001996293443901907, "loss": 0.9057247161865234, "mean_token_accuracy": 0.7823390491306782, "num_tokens": 8821843.0, "step": 630 }, { "entropy": 0.8901631090790033, "epoch": 0.3023145961265942, "grad_norm": 0.08016190677881241, "learning_rate": 0.00019961643878534465, "loss": 0.8986432075500488, "mean_token_accuracy": 0.7896160095930099, "num_tokens": 8969571.0, "step": 640 }, { "entropy": 0.8565576003864408, "epoch": 0.3070382616910723, "grad_norm": 0.06093674898147583, "learning_rate": 0.00019960331276463425, "loss": 0.8730425834655762, "mean_token_accuracy": 0.7965864725410938, "num_tokens": 9105075.0, "step": 650 }, { "entropy": 0.8957660870626569, "epoch": 0.3117619272555503, "grad_norm": 0.06780891865491867, "learning_rate": 0.00019958996635710267, "loss": 0.9151910781860352, "mean_token_accuracy": 0.7878959491848946, "num_tokens": 9242267.0, "step": 660 }, { "entropy": 0.9176279336214066, "epoch": 0.31648559282002836, "grad_norm": 0.07680392265319824, "learning_rate": 0.00019957639959228082, "loss": 0.9226786613464355, "mean_token_accuracy": 0.7837329708039761, "num_tokens": 9391460.0, "step": 670 }, { "entropy": 0.8888965021818876, "epoch": 0.32120925838450637, "grad_norm": 0.0915764570236206, "learning_rate": 0.00019956261250018712, "loss": 0.92776460647583, "mean_token_accuracy": 0.7923628441989422, "num_tokens": 9529311.0, "step": 680 }, { "entropy": 0.8909371437504887, "epoch": 0.32593292394898443, "grad_norm": 0.0839536041021347, "learning_rate": 0.00019954860511132752, "loss": 0.9093255996704102, "mean_token_accuracy": 0.790923023968935, "num_tokens": 9673501.0, "step": 690 }, { "entropy": 0.8591799547895789, "epoch": 0.33065658951346244, "grad_norm": 0.08593319356441498, "learning_rate": 0.00019953437745669542, "loss": 0.8891085624694824, "mean_token_accuracy": 0.7998368456959725, "num_tokens": 9806954.0, "step": 700 }, { "entropy": 0.8408734206110239, "epoch": 0.3353802550779405, "grad_norm": 0.07651221007108688, "learning_rate": 0.00019951992956777158, "loss": 0.8647046089172363, "mean_token_accuracy": 0.7982326999306679, "num_tokens": 9948276.0, "step": 710 }, { "entropy": 0.8772768702358007, "epoch": 0.3401039206424185, "grad_norm": 0.10006756335496902, "learning_rate": 0.00019950526147652405, "loss": 0.9168255805969239, "mean_token_accuracy": 0.7887730821967125, "num_tokens": 10080105.0, "step": 720 }, { "entropy": 0.8855971502140164, "epoch": 0.3448275862068966, "grad_norm": 0.07733277976512909, "learning_rate": 0.0001994903732154081, "loss": 0.9009263992309571, "mean_token_accuracy": 0.7946964234113694, "num_tokens": 10217877.0, "step": 730 }, { "entropy": 0.8447553940117359, "epoch": 0.3495512517713746, "grad_norm": 0.08651448041200638, "learning_rate": 0.00019947526481736615, "loss": 0.8569165229797363, "mean_token_accuracy": 0.7986436150968075, "num_tokens": 10358817.0, "step": 740 }, { "entropy": 0.9258218709379434, "epoch": 0.35427491733585265, "grad_norm": 0.08018967509269714, "learning_rate": 0.00019945993631582777, "loss": 0.9246758460998535, "mean_token_accuracy": 0.7827687576413155, "num_tokens": 10498267.0, "step": 750 }, { "entropy": 0.8718725338578224, "epoch": 0.35899858290033065, "grad_norm": 0.07328029721975327, "learning_rate": 0.00019944438774470956, "loss": 0.8837751388549805, "mean_token_accuracy": 0.7905028581619262, "num_tokens": 10642335.0, "step": 760 }, { "entropy": 0.8753207590430975, "epoch": 0.3637222484648087, "grad_norm": 0.06343984603881836, "learning_rate": 0.00019942861913841483, "loss": 0.911489200592041, "mean_token_accuracy": 0.7933915212750435, "num_tokens": 10783674.0, "step": 770 }, { "entropy": 0.8820716785266995, "epoch": 0.3684459140292867, "grad_norm": 0.15441954135894775, "learning_rate": 0.0001994126305318341, "loss": 0.891299057006836, "mean_token_accuracy": 0.7918222531676292, "num_tokens": 10928117.0, "step": 780 }, { "entropy": 0.8966817736625672, "epoch": 0.3731695795937648, "grad_norm": 0.07478035986423492, "learning_rate": 0.00019939642196034436, "loss": 0.9111992835998535, "mean_token_accuracy": 0.7887391865253448, "num_tokens": 11071334.0, "step": 790 }, { "entropy": 0.8712172966450453, "epoch": 0.3778932451582428, "grad_norm": 0.06701675802469254, "learning_rate": 0.00019937999345980947, "loss": 0.8795169830322266, "mean_token_accuracy": 0.7929419346153737, "num_tokens": 11215748.0, "step": 800 }, { "entropy": 0.9084727305918932, "epoch": 0.3826169107227208, "grad_norm": 0.06413980573415756, "learning_rate": 0.0001993633450665799, "loss": 0.9320363998413086, "mean_token_accuracy": 0.7900137290358543, "num_tokens": 11352770.0, "step": 810 }, { "entropy": 0.7942084204405546, "epoch": 0.38734057628719887, "grad_norm": 0.07811287045478821, "learning_rate": 0.00019934647681749262, "loss": 0.8022373199462891, "mean_token_accuracy": 0.8104212641716003, "num_tokens": 11490862.0, "step": 820 }, { "entropy": 0.7759390229359269, "epoch": 0.3920642418516769, "grad_norm": 0.063596211373806, "learning_rate": 0.0001993293887498711, "loss": 0.797119426727295, "mean_token_accuracy": 0.8127175860106945, "num_tokens": 11630700.0, "step": 830 }, { "entropy": 0.8476412085816264, "epoch": 0.39678790741615494, "grad_norm": 0.07755774259567261, "learning_rate": 0.00019931208090152514, "loss": 0.8529770851135254, "mean_token_accuracy": 0.8008555345237255, "num_tokens": 11766340.0, "step": 840 }, { "entropy": 0.6891022263094783, "epoch": 0.40151157298063295, "grad_norm": 0.06106919050216675, "learning_rate": 0.0001992945533107509, "loss": 0.7167798995971679, "mean_token_accuracy": 0.8321699120104313, "num_tokens": 11907211.0, "step": 850 }, { "entropy": 0.7868106795474887, "epoch": 0.406235238545111, "grad_norm": 0.06206279993057251, "learning_rate": 0.00019927680601633074, "loss": 0.7856165885925293, "mean_token_accuracy": 0.8127061523497104, "num_tokens": 12048793.0, "step": 860 }, { "entropy": 0.7624041633680463, "epoch": 0.410958904109589, "grad_norm": 0.06740711629390717, "learning_rate": 0.0001992588390575331, "loss": 0.7627418518066407, "mean_token_accuracy": 0.8209725596010685, "num_tokens": 12183561.0, "step": 870 }, { "entropy": 0.7676180090755225, "epoch": 0.4156825696740671, "grad_norm": 0.06678751856088638, "learning_rate": 0.00019924065247411252, "loss": 0.7924910545349121, "mean_token_accuracy": 0.817066914588213, "num_tokens": 12328771.0, "step": 880 }, { "entropy": 0.8627284154295921, "epoch": 0.4204062352385451, "grad_norm": 0.07592441886663437, "learning_rate": 0.00019922224630630945, "loss": 0.8818714141845703, "mean_token_accuracy": 0.7971808977425099, "num_tokens": 12469884.0, "step": 890 }, { "entropy": 0.7684948749840259, "epoch": 0.42512990080302315, "grad_norm": 0.059819918125867844, "learning_rate": 0.0001992036205948502, "loss": 0.7722855091094971, "mean_token_accuracy": 0.8185511201620101, "num_tokens": 12615473.0, "step": 900 }, { "entropy": 0.8133318793028593, "epoch": 0.42985356636750116, "grad_norm": 0.07291887700557709, "learning_rate": 0.00019918477538094688, "loss": 0.8389737129211425, "mean_token_accuracy": 0.8051427081227303, "num_tokens": 12750820.0, "step": 910 }, { "entropy": 0.8525479229167103, "epoch": 0.4345772319319792, "grad_norm": 0.07239335030317307, "learning_rate": 0.00019916571070629726, "loss": 0.8786557197570801, "mean_token_accuracy": 0.8011668443679809, "num_tokens": 12898942.0, "step": 920 }, { "entropy": 0.8252839187160135, "epoch": 0.43930089749645723, "grad_norm": 0.07384217530488968, "learning_rate": 0.00019914642661308474, "loss": 0.8349889755249024, "mean_token_accuracy": 0.8039939947426319, "num_tokens": 13039100.0, "step": 930 }, { "entropy": 0.7331045351922512, "epoch": 0.4440245630609353, "grad_norm": 0.059965118765830994, "learning_rate": 0.0001991269231439781, "loss": 0.7454654216766358, "mean_token_accuracy": 0.8229805640876293, "num_tokens": 13173602.0, "step": 940 }, { "entropy": 0.8036073945462704, "epoch": 0.4487482286254133, "grad_norm": 0.061711378395557404, "learning_rate": 0.0001991072003421317, "loss": 0.8097902297973633, "mean_token_accuracy": 0.809221301227808, "num_tokens": 13318862.0, "step": 950 }, { "entropy": 0.8632235093042254, "epoch": 0.45347189418989137, "grad_norm": 0.06610094010829926, "learning_rate": 0.00019908725825118503, "loss": 0.9003376007080078, "mean_token_accuracy": 0.794591860473156, "num_tokens": 13458323.0, "step": 960 }, { "entropy": 0.7831415619701148, "epoch": 0.4581955597543694, "grad_norm": 0.06393159925937653, "learning_rate": 0.00019906709691526296, "loss": 0.7798863410949707, "mean_token_accuracy": 0.815378001332283, "num_tokens": 13594585.0, "step": 970 }, { "entropy": 0.7741080436855554, "epoch": 0.46291922531884744, "grad_norm": 0.06187094748020172, "learning_rate": 0.00019904671637897527, "loss": 0.7788362979888916, "mean_token_accuracy": 0.8184373766183853, "num_tokens": 13737357.0, "step": 980 }, { "entropy": 0.7595177697017789, "epoch": 0.46764289088332545, "grad_norm": 0.05875694379210472, "learning_rate": 0.00019902611668741693, "loss": 0.753893518447876, "mean_token_accuracy": 0.8178720653057099, "num_tokens": 13878878.0, "step": 990 }, { "entropy": 0.825975145585835, "epoch": 0.4723665564478035, "grad_norm": 0.05704476311802864, "learning_rate": 0.00019900529788616773, "loss": 0.8415384292602539, "mean_token_accuracy": 0.8027308411896229, "num_tokens": 14017830.0, "step": 1000 }, { "entropy": 0.7697717467322945, "epoch": 0.4770902220122815, "grad_norm": 0.07954072952270508, "learning_rate": 0.00019898426002129232, "loss": 0.777623462677002, "mean_token_accuracy": 0.8132883630692959, "num_tokens": 14164135.0, "step": 1010 }, { "entropy": 0.8544552912935615, "epoch": 0.4818138875767596, "grad_norm": 0.0676150768995285, "learning_rate": 0.00019896300313934, "loss": 0.8684463500976562, "mean_token_accuracy": 0.7986083067953587, "num_tokens": 14307388.0, "step": 1020 }, { "entropy": 0.8102986900135875, "epoch": 0.4865375531412376, "grad_norm": 0.09053028374910355, "learning_rate": 0.00019894152728734476, "loss": 0.8150781631469727, "mean_token_accuracy": 0.8103755861520767, "num_tokens": 14449647.0, "step": 1030 }, { "entropy": 0.7788866188377142, "epoch": 0.49126121870571565, "grad_norm": 0.06917324662208557, "learning_rate": 0.00019891983251282502, "loss": 0.779150915145874, "mean_token_accuracy": 0.8139691516757012, "num_tokens": 14592748.0, "step": 1040 }, { "entropy": 0.8062027977779508, "epoch": 0.49598488427019366, "grad_norm": 0.07920317351818085, "learning_rate": 0.0001988979188637836, "loss": 0.8316155433654785, "mean_token_accuracy": 0.8057821422815323, "num_tokens": 14733460.0, "step": 1050 }, { "entropy": 0.7319147000089288, "epoch": 0.5007085498346717, "grad_norm": 0.06537997722625732, "learning_rate": 0.00019887578638870773, "loss": 0.7396903991699219, "mean_token_accuracy": 0.8268778897821903, "num_tokens": 14868589.0, "step": 1060 }, { "entropy": 0.8478265231475234, "epoch": 0.5054322153991497, "grad_norm": 0.06009751185774803, "learning_rate": 0.0001988534351365686, "loss": 0.8618332862854003, "mean_token_accuracy": 0.7993136338889599, "num_tokens": 15008474.0, "step": 1070 }, { "entropy": 0.7692794919013977, "epoch": 0.5101558809636277, "grad_norm": 0.07316073030233383, "learning_rate": 0.0001988308651568217, "loss": 0.7756832599639892, "mean_token_accuracy": 0.816083624213934, "num_tokens": 15145892.0, "step": 1080 }, { "entropy": 0.7628285551443696, "epoch": 0.5148795465281059, "grad_norm": 0.06232941523194313, "learning_rate": 0.00019880807649940636, "loss": 0.7790435314178467, "mean_token_accuracy": 0.8200066022574901, "num_tokens": 15290594.0, "step": 1090 }, { "entropy": 0.7464076656848192, "epoch": 0.5196032120925839, "grad_norm": 0.061241474002599716, "learning_rate": 0.0001987850692147458, "loss": 0.7722044944763183, "mean_token_accuracy": 0.8226780131459236, "num_tokens": 15422773.0, "step": 1100 }, { "entropy": 0.7531939372420311, "epoch": 0.5243268776570619, "grad_norm": 0.06392857432365417, "learning_rate": 0.00019876184335374702, "loss": 0.7565675735473633, "mean_token_accuracy": 0.8209423452615738, "num_tokens": 15564404.0, "step": 1110 }, { "entropy": 0.7380848005414009, "epoch": 0.5290505432215399, "grad_norm": 0.06649397313594818, "learning_rate": 0.0001987383989678006, "loss": 0.7445190906524658, "mean_token_accuracy": 0.822866752743721, "num_tokens": 15698232.0, "step": 1120 }, { "entropy": 0.7761778807267546, "epoch": 0.533774208786018, "grad_norm": 0.07237660884857178, "learning_rate": 0.00019871473610878057, "loss": 0.7909494400024414, "mean_token_accuracy": 0.814056969434023, "num_tokens": 15839627.0, "step": 1130 }, { "entropy": 0.7619022153317928, "epoch": 0.538497874350496, "grad_norm": 0.06544393301010132, "learning_rate": 0.00019869085482904458, "loss": 0.7763425350189209, "mean_token_accuracy": 0.8176195241510869, "num_tokens": 15976065.0, "step": 1140 }, { "entropy": 0.7591480510309339, "epoch": 0.543221539914974, "grad_norm": 0.07460027933120728, "learning_rate": 0.00019866675518143338, "loss": 0.777614688873291, "mean_token_accuracy": 0.8194335848093033, "num_tokens": 16123226.0, "step": 1150 }, { "entropy": 0.7916649857535958, "epoch": 0.547945205479452, "grad_norm": 0.07511059939861298, "learning_rate": 0.00019864243721927095, "loss": 0.7786167144775391, "mean_token_accuracy": 0.8125829711556435, "num_tokens": 16265911.0, "step": 1160 }, { "entropy": 0.7833957163617015, "epoch": 0.5526688710439301, "grad_norm": 0.051263704895973206, "learning_rate": 0.0001986179009963643, "loss": 0.788210678100586, "mean_token_accuracy": 0.8091006651520729, "num_tokens": 16408517.0, "step": 1170 }, { "entropy": 0.7674215922132135, "epoch": 0.5573925366084082, "grad_norm": 0.0641508400440216, "learning_rate": 0.00019859314656700338, "loss": 0.7742294788360595, "mean_token_accuracy": 0.8170669317245484, "num_tokens": 16548449.0, "step": 1180 }, { "entropy": 0.7492768192663789, "epoch": 0.5621162021728862, "grad_norm": 0.06696533411741257, "learning_rate": 0.00019856817398596101, "loss": 0.7694369792938233, "mean_token_accuracy": 0.8197654329240323, "num_tokens": 16684809.0, "step": 1190 }, { "entropy": 0.8142074134200812, "epoch": 0.5668398677373642, "grad_norm": 0.07176313549280167, "learning_rate": 0.0001985429833084927, "loss": 0.8397821426391602, "mean_token_accuracy": 0.8066790141165257, "num_tokens": 16833694.0, "step": 1200 }, { "entropy": 0.7840611487627029, "epoch": 0.5715635333018422, "grad_norm": 0.06795291602611542, "learning_rate": 0.00019851757459033635, "loss": 0.7850320339202881, "mean_token_accuracy": 0.8136720523238182, "num_tokens": 16977567.0, "step": 1210 }, { "entropy": 0.7338969888165593, "epoch": 0.5762871988663203, "grad_norm": 0.0653485506772995, "learning_rate": 0.00019849194788771257, "loss": 0.7483477592468262, "mean_token_accuracy": 0.8235686682164669, "num_tokens": 17121584.0, "step": 1220 }, { "entropy": 0.6994957081973553, "epoch": 0.5810108644307983, "grad_norm": 0.05542505159974098, "learning_rate": 0.0001984661032573241, "loss": 0.6853211402893067, "mean_token_accuracy": 0.8307100392878055, "num_tokens": 17255874.0, "step": 1230 }, { "entropy": 0.7630806406959891, "epoch": 0.5857345299952763, "grad_norm": 0.06317171454429626, "learning_rate": 0.00019844004075635603, "loss": 0.7817557334899903, "mean_token_accuracy": 0.8199850477278232, "num_tokens": 17391776.0, "step": 1240 }, { "entropy": 0.6812972750514745, "epoch": 0.5904581955597543, "grad_norm": 0.058689989149570465, "learning_rate": 0.0001984137604424753, "loss": 0.6721157550811767, "mean_token_accuracy": 0.8356978453695774, "num_tokens": 17529474.0, "step": 1250 }, { "entropy": 0.7690079558640719, "epoch": 0.5951818611242324, "grad_norm": 0.06261543184518814, "learning_rate": 0.00019838726237383109, "loss": 0.7748788833618164, "mean_token_accuracy": 0.8140993498265743, "num_tokens": 17668282.0, "step": 1260 }, { "entropy": 0.7554839655756951, "epoch": 0.5999055266887104, "grad_norm": 0.05502874031662941, "learning_rate": 0.00019836054660905407, "loss": 0.7733687400817871, "mean_token_accuracy": 0.8195707939565182, "num_tokens": 17814404.0, "step": 1270 }, { "entropy": 0.738018630258739, "epoch": 0.6046291922531885, "grad_norm": 0.0700552761554718, "learning_rate": 0.00019833361320725686, "loss": 0.749139404296875, "mean_token_accuracy": 0.821788027882576, "num_tokens": 17957161.0, "step": 1280 }, { "entropy": 0.7832638947293162, "epoch": 0.6093528578176665, "grad_norm": 0.06342653930187225, "learning_rate": 0.0001983064622280335, "loss": 0.8009039878845214, "mean_token_accuracy": 0.8136322379112244, "num_tokens": 18100446.0, "step": 1290 }, { "entropy": 0.7787608832120896, "epoch": 0.6140765233821446, "grad_norm": 0.06524264812469482, "learning_rate": 0.0001982790937314595, "loss": 0.7779227733612061, "mean_token_accuracy": 0.8171707160770894, "num_tokens": 18245060.0, "step": 1300 }, { "entropy": 0.7660725971683859, "epoch": 0.6188001889466226, "grad_norm": 0.07097435742616653, "learning_rate": 0.0001982515077780916, "loss": 0.7690092086791992, "mean_token_accuracy": 0.8177868269383908, "num_tokens": 18377118.0, "step": 1310 }, { "entropy": 0.7343099780380726, "epoch": 0.6235238545111006, "grad_norm": 0.06699824333190918, "learning_rate": 0.00019822370442896777, "loss": 0.7325379848480225, "mean_token_accuracy": 0.8246756240725517, "num_tokens": 18512990.0, "step": 1320 }, { "entropy": 0.7235783237963915, "epoch": 0.6282475200755786, "grad_norm": 0.061282556504011154, "learning_rate": 0.00019819568374560698, "loss": 0.7249025821685791, "mean_token_accuracy": 0.8252772688865662, "num_tokens": 18649517.0, "step": 1330 }, { "entropy": 0.7023943999782205, "epoch": 0.6329711856400567, "grad_norm": 0.06388921290636063, "learning_rate": 0.000198167445790009, "loss": 0.7134102344512939, "mean_token_accuracy": 0.8308245539665222, "num_tokens": 18786054.0, "step": 1340 }, { "entropy": 0.6977807080373168, "epoch": 0.6376948512045347, "grad_norm": 0.06597480922937393, "learning_rate": 0.0001981389906246545, "loss": 0.7127728462219238, "mean_token_accuracy": 0.8313981927931309, "num_tokens": 18923642.0, "step": 1350 }, { "entropy": 0.7465406663715839, "epoch": 0.6424185167690127, "grad_norm": 0.058440908789634705, "learning_rate": 0.00019811031831250462, "loss": 0.73866868019104, "mean_token_accuracy": 0.8232753865420819, "num_tokens": 19063072.0, "step": 1360 }, { "entropy": 0.7289104614406824, "epoch": 0.6471421823334907, "grad_norm": 0.061089277267456055, "learning_rate": 0.00019808142891700103, "loss": 0.725409746170044, "mean_token_accuracy": 0.8270080789923668, "num_tokens": 19197316.0, "step": 1370 }, { "entropy": 0.702583946660161, "epoch": 0.6518658478979689, "grad_norm": 0.06704925745725632, "learning_rate": 0.00019805232250206574, "loss": 0.7035065174102784, "mean_token_accuracy": 0.8279381163418293, "num_tokens": 19338724.0, "step": 1380 }, { "entropy": 0.758362696133554, "epoch": 0.6565895134624469, "grad_norm": 0.05272005498409271, "learning_rate": 0.0001980229991321009, "loss": 0.7745654582977295, "mean_token_accuracy": 0.8178007207810879, "num_tokens": 19474185.0, "step": 1390 }, { "entropy": 0.7246218353509903, "epoch": 0.6613131790269249, "grad_norm": 0.06560070067644119, "learning_rate": 0.00019799345887198873, "loss": 0.7173084735870361, "mean_token_accuracy": 0.824228448420763, "num_tokens": 19612375.0, "step": 1400 }, { "entropy": 0.692228338867426, "epoch": 0.6660368445914029, "grad_norm": 0.047808270901441574, "learning_rate": 0.0001979637017870914, "loss": 0.6872843265533447, "mean_token_accuracy": 0.8327514350414276, "num_tokens": 19755324.0, "step": 1410 }, { "entropy": 0.7644963745027781, "epoch": 0.670760510155881, "grad_norm": 0.06209803745150566, "learning_rate": 0.00019793372794325075, "loss": 0.7778499603271485, "mean_token_accuracy": 0.8190472275018692, "num_tokens": 19892730.0, "step": 1420 }, { "entropy": 0.7111064940690994, "epoch": 0.675484175720359, "grad_norm": 0.06112042814493179, "learning_rate": 0.00019790353740678832, "loss": 0.7190793514251709, "mean_token_accuracy": 0.8267201483249664, "num_tokens": 20033350.0, "step": 1430 }, { "entropy": 0.7761819878593087, "epoch": 0.680207841284837, "grad_norm": 0.06455013900995255, "learning_rate": 0.00019787313024450498, "loss": 0.7655410766601562, "mean_token_accuracy": 0.8162212684750557, "num_tokens": 20178654.0, "step": 1440 }, { "entropy": 0.750446498952806, "epoch": 0.684931506849315, "grad_norm": 0.05603810399770737, "learning_rate": 0.00019784250652368104, "loss": 0.7613818168640136, "mean_token_accuracy": 0.8208815030753612, "num_tokens": 20321652.0, "step": 1450 }, { "entropy": 0.7475989449769258, "epoch": 0.6896551724137931, "grad_norm": 0.07111338526010513, "learning_rate": 0.00019781166631207598, "loss": 0.7683018684387207, "mean_token_accuracy": 0.8204028435051441, "num_tokens": 20463869.0, "step": 1460 }, { "entropy": 0.6830848310142755, "epoch": 0.6943788379782712, "grad_norm": 0.060475390404462814, "learning_rate": 0.00019778060967792817, "loss": 0.6971904277801514, "mean_token_accuracy": 0.8328015118837356, "num_tokens": 20594850.0, "step": 1470 }, { "entropy": 0.6571184325963259, "epoch": 0.6991025035427492, "grad_norm": 0.06787912547588348, "learning_rate": 0.00019774933668995502, "loss": 0.657703161239624, "mean_token_accuracy": 0.841765958070755, "num_tokens": 20729161.0, "step": 1480 }, { "entropy": 0.6259467517957091, "epoch": 0.7038261691072272, "grad_norm": 0.05940692499279976, "learning_rate": 0.0001977178474173525, "loss": 0.6371507167816162, "mean_token_accuracy": 0.847675783932209, "num_tokens": 20868798.0, "step": 1490 }, { "entropy": 0.6716957721859217, "epoch": 0.7085498346717053, "grad_norm": 0.06572386622428894, "learning_rate": 0.00019768614192979523, "loss": 0.6638718605041504, "mean_token_accuracy": 0.8394592143595219, "num_tokens": 21009132.0, "step": 1500 }, { "entropy": 0.6677395252510905, "epoch": 0.7132735002361833, "grad_norm": 0.0641874372959137, "learning_rate": 0.00019765422029743626, "loss": 0.6707793235778808, "mean_token_accuracy": 0.8393042981624603, "num_tokens": 21137884.0, "step": 1510 }, { "entropy": 0.7278847776353359, "epoch": 0.7179971658006613, "grad_norm": 0.07689928263425827, "learning_rate": 0.00019762208259090674, "loss": 0.7351594924926758, "mean_token_accuracy": 0.8226647913455963, "num_tokens": 21269708.0, "step": 1520 }, { "entropy": 0.72459547650069, "epoch": 0.7227208313651393, "grad_norm": 0.06539573520421982, "learning_rate": 0.00019758972888131614, "loss": 0.7338285446166992, "mean_token_accuracy": 0.8271237276494503, "num_tokens": 21409844.0, "step": 1530 }, { "entropy": 0.6734345428645611, "epoch": 0.7274444969296174, "grad_norm": 0.06240133196115494, "learning_rate": 0.0001975571592402517, "loss": 0.6539097785949707, "mean_token_accuracy": 0.8374067522585392, "num_tokens": 21553377.0, "step": 1540 }, { "entropy": 0.7088769232854247, "epoch": 0.7321681624940954, "grad_norm": 0.05273374915122986, "learning_rate": 0.00019752437373977847, "loss": 0.7295046329498291, "mean_token_accuracy": 0.8294990807771683, "num_tokens": 21688735.0, "step": 1550 }, { "entropy": 0.675998998619616, "epoch": 0.7368918280585735, "grad_norm": 0.05165047198534012, "learning_rate": 0.0001974913724524391, "loss": 0.6759829044342041, "mean_token_accuracy": 0.8369193807244301, "num_tokens": 21824467.0, "step": 1560 }, { "entropy": 0.674262166209519, "epoch": 0.7416154936230515, "grad_norm": 0.05678558722138405, "learning_rate": 0.00019745815545125386, "loss": 0.6977355003356933, "mean_token_accuracy": 0.8329213678836822, "num_tokens": 21958794.0, "step": 1570 }, { "entropy": 0.7110530557110906, "epoch": 0.7463391591875296, "grad_norm": 0.05730152130126953, "learning_rate": 0.00019742472280972012, "loss": 0.7211824893951416, "mean_token_accuracy": 0.8288441769778728, "num_tokens": 22097721.0, "step": 1580 }, { "entropy": 0.6991312104277313, "epoch": 0.7510628247520076, "grad_norm": 0.05507818982005119, "learning_rate": 0.00019739107460181246, "loss": 0.6827840328216552, "mean_token_accuracy": 0.8327453970909119, "num_tokens": 22248392.0, "step": 1590 }, { "entropy": 0.6840994134545326, "epoch": 0.7557864903164856, "grad_norm": 0.054441455751657486, "learning_rate": 0.00019735721090198246, "loss": 0.6832675933837891, "mean_token_accuracy": 0.8338833898305893, "num_tokens": 22383385.0, "step": 1600 }, { "entropy": 0.6878544526174665, "epoch": 0.7605101558809636, "grad_norm": 0.05949411541223526, "learning_rate": 0.00019732313178515845, "loss": 0.6931551456451416, "mean_token_accuracy": 0.8354523934423923, "num_tokens": 22518485.0, "step": 1610 }, { "entropy": 0.6842782452702523, "epoch": 0.7652338214454416, "grad_norm": 0.06143569573760033, "learning_rate": 0.0001972888373267455, "loss": 0.6856177806854248, "mean_token_accuracy": 0.8309583082795143, "num_tokens": 22662002.0, "step": 1620 }, { "entropy": 0.7122562890872359, "epoch": 0.7699574870099197, "grad_norm": 0.07086401432752609, "learning_rate": 0.00019725432760262502, "loss": 0.7232859134674072, "mean_token_accuracy": 0.8299242489039897, "num_tokens": 22805194.0, "step": 1630 }, { "entropy": 0.6739442015066743, "epoch": 0.7746811525743977, "grad_norm": 0.06858492642641068, "learning_rate": 0.00019721960268915482, "loss": 0.6691046714782715, "mean_token_accuracy": 0.8398007422685623, "num_tokens": 22947066.0, "step": 1640 }, { "entropy": 0.6781346376985311, "epoch": 0.7794048181388757, "grad_norm": 0.06279728561639786, "learning_rate": 0.00019718466266316885, "loss": 0.6975903034210205, "mean_token_accuracy": 0.8357139840722084, "num_tokens": 23081033.0, "step": 1650 }, { "entropy": 0.6545021275058389, "epoch": 0.7841284837033538, "grad_norm": 0.05892058089375496, "learning_rate": 0.00019714950760197695, "loss": 0.6395064353942871, "mean_token_accuracy": 0.8424684174358845, "num_tokens": 23219586.0, "step": 1660 }, { "entropy": 0.7294390849769116, "epoch": 0.7888521492678319, "grad_norm": 0.07074124366044998, "learning_rate": 0.00019711413758336486, "loss": 0.7317322731018067, "mean_token_accuracy": 0.8234610050916672, "num_tokens": 23366178.0, "step": 1670 }, { "entropy": 0.6562610348686576, "epoch": 0.7935758148323099, "grad_norm": 0.0659153088927269, "learning_rate": 0.00019707855268559382, "loss": 0.6403413772583008, "mean_token_accuracy": 0.8414281204342842, "num_tokens": 23501225.0, "step": 1680 }, { "entropy": 0.6413846110925079, "epoch": 0.7982994803967879, "grad_norm": 0.06126869469881058, "learning_rate": 0.0001970427529874007, "loss": 0.6617055416107178, "mean_token_accuracy": 0.8455805391073227, "num_tokens": 23641082.0, "step": 1690 }, { "entropy": 0.6908547822386026, "epoch": 0.8030231459612659, "grad_norm": 0.06157476082444191, "learning_rate": 0.00019700673856799745, "loss": 0.6820013046264648, "mean_token_accuracy": 0.8359040506184101, "num_tokens": 23779296.0, "step": 1700 }, { "entropy": 0.7308671203441918, "epoch": 0.807746811525744, "grad_norm": 0.057691268622875214, "learning_rate": 0.00019697050950707125, "loss": 0.729799222946167, "mean_token_accuracy": 0.8249626256525516, "num_tokens": 23917061.0, "step": 1710 }, { "entropy": 0.7200676096603275, "epoch": 0.812470477090222, "grad_norm": 0.06415704637765884, "learning_rate": 0.00019693406588478416, "loss": 0.7374712944030761, "mean_token_accuracy": 0.8268370419740677, "num_tokens": 24056729.0, "step": 1720 }, { "entropy": 0.6000354990363121, "epoch": 0.8171941426547, "grad_norm": 0.0773177444934845, "learning_rate": 0.000196897407781773, "loss": 0.5927223205566406, "mean_token_accuracy": 0.8535788625478744, "num_tokens": 24199200.0, "step": 1730 }, { "entropy": 0.6869307642802596, "epoch": 0.821917808219178, "grad_norm": 0.07565486431121826, "learning_rate": 0.00019686053527914917, "loss": 0.6846215724945068, "mean_token_accuracy": 0.8362017579376697, "num_tokens": 24335670.0, "step": 1740 }, { "entropy": 0.6294289611279964, "epoch": 0.8266414737836562, "grad_norm": 0.060879867523908615, "learning_rate": 0.00019682344845849841, "loss": 0.62716646194458, "mean_token_accuracy": 0.8449809499084949, "num_tokens": 24469821.0, "step": 1750 }, { "entropy": 0.6230787862092256, "epoch": 0.8313651393481342, "grad_norm": 0.05589767172932625, "learning_rate": 0.00019678614740188075, "loss": 0.6157899379730225, "mean_token_accuracy": 0.8494812779128551, "num_tokens": 24616195.0, "step": 1760 }, { "entropy": 0.6489476334303618, "epoch": 0.8360888049126122, "grad_norm": 0.05557194724678993, "learning_rate": 0.00019674863219183014, "loss": 0.6374253749847412, "mean_token_accuracy": 0.8444698095321655, "num_tokens": 24754455.0, "step": 1770 }, { "entropy": 0.6818723721429706, "epoch": 0.8408124704770902, "grad_norm": 0.052851974964141846, "learning_rate": 0.00019671090291135455, "loss": 0.6785852909088135, "mean_token_accuracy": 0.8359914973378182, "num_tokens": 24896122.0, "step": 1780 }, { "entropy": 0.6280808305367828, "epoch": 0.8455361360415683, "grad_norm": 0.06030328571796417, "learning_rate": 0.0001966729596439354, "loss": 0.6169336795806885, "mean_token_accuracy": 0.8470665298402309, "num_tokens": 25031694.0, "step": 1790 }, { "entropy": 0.6297697514295578, "epoch": 0.8502598016060463, "grad_norm": 0.05448286235332489, "learning_rate": 0.00019663480247352773, "loss": 0.6322630882263184, "mean_token_accuracy": 0.8457629255950451, "num_tokens": 25173070.0, "step": 1800 }, { "entropy": 0.6825473893433809, "epoch": 0.8549834671705243, "grad_norm": 0.05518447235226631, "learning_rate": 0.00019659643148455983, "loss": 0.6699731826782227, "mean_token_accuracy": 0.8336914747953414, "num_tokens": 25313372.0, "step": 1810 }, { "entropy": 0.6175509238615632, "epoch": 0.8597071327350023, "grad_norm": 0.06463871151208878, "learning_rate": 0.0001965578467619331, "loss": 0.6056600570678711, "mean_token_accuracy": 0.8523232497274875, "num_tokens": 25451221.0, "step": 1820 }, { "entropy": 0.723086348734796, "epoch": 0.8644307982994804, "grad_norm": 0.06886560469865799, "learning_rate": 0.00019651904839102186, "loss": 0.7144670009613037, "mean_token_accuracy": 0.8268191002309322, "num_tokens": 25583322.0, "step": 1830 }, { "entropy": 0.683582342043519, "epoch": 0.8691544638639584, "grad_norm": 0.05712641775608063, "learning_rate": 0.0001964800364576731, "loss": 0.674677848815918, "mean_token_accuracy": 0.8341485545039177, "num_tokens": 25720258.0, "step": 1840 }, { "entropy": 0.6558683659881354, "epoch": 0.8738781294284365, "grad_norm": 0.06446284800767899, "learning_rate": 0.00019644081104820647, "loss": 0.6572765350341797, "mean_token_accuracy": 0.8389321446418763, "num_tokens": 25855817.0, "step": 1850 }, { "entropy": 0.6757289810106158, "epoch": 0.8786017949929145, "grad_norm": 0.05949859693646431, "learning_rate": 0.00019640137224941382, "loss": 0.6753401279449462, "mean_token_accuracy": 0.8351259626448154, "num_tokens": 25995714.0, "step": 1860 }, { "entropy": 0.5843123767524958, "epoch": 0.8833254605573926, "grad_norm": 0.05486496165394783, "learning_rate": 0.0001963617201485593, "loss": 0.5740822315216064, "mean_token_accuracy": 0.8587191238999367, "num_tokens": 26129624.0, "step": 1870 }, { "entropy": 0.647276826389134, "epoch": 0.8880491261218706, "grad_norm": 0.05606291443109512, "learning_rate": 0.0001963218548333789, "loss": 0.6594659328460694, "mean_token_accuracy": 0.8412343457341194, "num_tokens": 26264668.0, "step": 1880 }, { "entropy": 0.6084568424150347, "epoch": 0.8927727916863486, "grad_norm": 0.07708149403333664, "learning_rate": 0.00019628177639208048, "loss": 0.6106575965881348, "mean_token_accuracy": 0.8510070964694023, "num_tokens": 26399298.0, "step": 1890 }, { "entropy": 0.6972724292427301, "epoch": 0.8974964572508266, "grad_norm": 0.06351560354232788, "learning_rate": 0.0001962414849133434, "loss": 0.6829310894012451, "mean_token_accuracy": 0.8365243338048458, "num_tokens": 26541607.0, "step": 1900 }, { "entropy": 0.6462207734584808, "epoch": 0.9022201228153047, "grad_norm": 0.05115530267357826, "learning_rate": 0.00019620098048631838, "loss": 0.6354568004608154, "mean_token_accuracy": 0.8417663842439651, "num_tokens": 26685875.0, "step": 1910 }, { "entropy": 0.6453853419050575, "epoch": 0.9069437883797827, "grad_norm": 0.07085788249969482, "learning_rate": 0.0001961602632006274, "loss": 0.6471972465515137, "mean_token_accuracy": 0.8423703745007515, "num_tokens": 26826404.0, "step": 1920 }, { "entropy": 0.6378903038799763, "epoch": 0.9116674539442607, "grad_norm": 0.06412967294454575, "learning_rate": 0.0001961193331463634, "loss": 0.6362363338470459, "mean_token_accuracy": 0.8454982757568359, "num_tokens": 26960578.0, "step": 1930 }, { "entropy": 0.7276632092893124, "epoch": 0.9163911195087387, "grad_norm": 0.0602058544754982, "learning_rate": 0.00019607819041409007, "loss": 0.7387227535247802, "mean_token_accuracy": 0.8244086593389511, "num_tokens": 27094998.0, "step": 1940 }, { "entropy": 0.6283036121167243, "epoch": 0.9211147850732169, "grad_norm": 0.05576204136013985, "learning_rate": 0.00019603683509484163, "loss": 0.6084622383117676, "mean_token_accuracy": 0.8472853414714336, "num_tokens": 27230442.0, "step": 1950 }, { "entropy": 0.6102050228044391, "epoch": 0.9258384506376949, "grad_norm": 0.05982797220349312, "learning_rate": 0.0001959952672801228, "loss": 0.6122698307037353, "mean_token_accuracy": 0.8533632613718509, "num_tokens": 27366971.0, "step": 1960 }, { "entropy": 0.6490558795630932, "epoch": 0.9305621162021729, "grad_norm": 0.06649584323167801, "learning_rate": 0.0001959534870619084, "loss": 0.663581371307373, "mean_token_accuracy": 0.8419334851205349, "num_tokens": 27504324.0, "step": 1970 }, { "entropy": 0.7232493607327342, "epoch": 0.9352857817666509, "grad_norm": 0.06305187940597534, "learning_rate": 0.00019591149453264324, "loss": 0.7096786022186279, "mean_token_accuracy": 0.825618925690651, "num_tokens": 27640987.0, "step": 1980 }, { "entropy": 0.650670844130218, "epoch": 0.940009447331129, "grad_norm": 0.06343328952789307, "learning_rate": 0.00019586928978524188, "loss": 0.6513476848602295, "mean_token_accuracy": 0.8433810010552406, "num_tokens": 27775300.0, "step": 1990 }, { "entropy": 0.6628386480733752, "epoch": 0.944733112895607, "grad_norm": 0.06584423035383224, "learning_rate": 0.0001958268729130885, "loss": 0.6749490737915039, "mean_token_accuracy": 0.8359682351350785, "num_tokens": 27913726.0, "step": 2000 }, { "entropy": 0.6213206199929118, "epoch": 0.949456778460085, "grad_norm": 0.058901596814394, "learning_rate": 0.00019578424401003663, "loss": 0.6026661396026611, "mean_token_accuracy": 0.8501860998570919, "num_tokens": 28058288.0, "step": 2010 }, { "entropy": 0.6488163938745857, "epoch": 0.954180444024563, "grad_norm": 0.07867100089788437, "learning_rate": 0.00019574140317040882, "loss": 0.6409980773925781, "mean_token_accuracy": 0.8418553799390793, "num_tokens": 28199386.0, "step": 2020 }, { "entropy": 0.6404497560113669, "epoch": 0.958904109589041, "grad_norm": 0.058080337941646576, "learning_rate": 0.00019569835048899678, "loss": 0.6465488910675049, "mean_token_accuracy": 0.8448394730687141, "num_tokens": 28336266.0, "step": 2030 }, { "entropy": 0.6589969670400023, "epoch": 0.9636277751535192, "grad_norm": 0.06629297137260437, "learning_rate": 0.00019565508606106074, "loss": 0.6600945472717286, "mean_token_accuracy": 0.8386769525706768, "num_tokens": 28475569.0, "step": 2040 }, { "entropy": 0.6279681742191314, "epoch": 0.9683514407179972, "grad_norm": 0.06361208856105804, "learning_rate": 0.00019561160998232956, "loss": 0.6077560901641845, "mean_token_accuracy": 0.8486055627465248, "num_tokens": 28606376.0, "step": 2050 }, { "entropy": 0.7186167985200882, "epoch": 0.9730751062824752, "grad_norm": 0.05606354400515556, "learning_rate": 0.00019556792234900043, "loss": 0.723689079284668, "mean_token_accuracy": 0.8276458971202374, "num_tokens": 28754004.0, "step": 2060 }, { "entropy": 0.6525811120867729, "epoch": 0.9777987718469532, "grad_norm": 0.06746236234903336, "learning_rate": 0.00019552402325773854, "loss": 0.6400609016418457, "mean_token_accuracy": 0.8430978931486607, "num_tokens": 28902382.0, "step": 2070 }, { "entropy": 0.6609141793102026, "epoch": 0.9825224374114313, "grad_norm": 0.06037747114896774, "learning_rate": 0.00019547991280567705, "loss": 0.6507028579711914, "mean_token_accuracy": 0.8434766083955765, "num_tokens": 29033278.0, "step": 2080 }, { "entropy": 0.6118150997906924, "epoch": 0.9872461029759093, "grad_norm": 0.06339357048273087, "learning_rate": 0.00019543559109041678, "loss": 0.6195655822753906, "mean_token_accuracy": 0.8527800843119622, "num_tokens": 29169177.0, "step": 2090 }, { "entropy": 0.6759829174727201, "epoch": 0.9919697685403873, "grad_norm": 0.0663270652294159, "learning_rate": 0.0001953910582100259, "loss": 0.6755183219909668, "mean_token_accuracy": 0.840421637147665, "num_tokens": 29307425.0, "step": 2100 }, { "entropy": 0.6058363078162075, "epoch": 0.9966934341048653, "grad_norm": 0.05744867026805878, "learning_rate": 0.00019534631426303999, "loss": 0.5953300952911377, "mean_token_accuracy": 0.8538600817322731, "num_tokens": 29448102.0, "step": 2110 }, { "entropy": 0.6707561606541276, "epoch": 1.0014170996693434, "grad_norm": 0.06992816179990768, "learning_rate": 0.00019530135934846138, "loss": 0.6712033748626709, "mean_token_accuracy": 0.8415774501860142, "num_tokens": 29587134.0, "step": 2120 }, { "entropy": 0.5769699368625879, "epoch": 1.0061407652338215, "grad_norm": 0.04780084639787674, "learning_rate": 0.00019525619356575952, "loss": 0.5691622734069824, "mean_token_accuracy": 0.8600594893097877, "num_tokens": 29728250.0, "step": 2130 }, { "entropy": 0.5800661744549871, "epoch": 1.0108644307982995, "grad_norm": 0.05516349524259567, "learning_rate": 0.00019521081701487022, "loss": 0.5810145854949951, "mean_token_accuracy": 0.8577248364686966, "num_tokens": 29873598.0, "step": 2140 }, { "entropy": 0.5718592911958694, "epoch": 1.0155880963627775, "grad_norm": 0.05691181495785713, "learning_rate": 0.00019516522979619568, "loss": 0.5542007446289062, "mean_token_accuracy": 0.8590722389519214, "num_tokens": 30014444.0, "step": 2150 }, { "entropy": 0.5361275443807244, "epoch": 1.0203117619272555, "grad_norm": 0.0631425529718399, "learning_rate": 0.00019511943201060426, "loss": 0.5191277027130127, "mean_token_accuracy": 0.8707845136523247, "num_tokens": 30151147.0, "step": 2160 }, { "entropy": 0.614743078686297, "epoch": 1.0250354274917335, "grad_norm": 0.06928510218858719, "learning_rate": 0.00019507342375943028, "loss": 0.60310378074646, "mean_token_accuracy": 0.8502935551106929, "num_tokens": 30286217.0, "step": 2170 }, { "entropy": 0.5105130102485418, "epoch": 1.0297590930562117, "grad_norm": 0.06384274363517761, "learning_rate": 0.0001950272051444736, "loss": 0.508014726638794, "mean_token_accuracy": 0.8759646862745285, "num_tokens": 30428446.0, "step": 2180 }, { "entropy": 0.5641612399369478, "epoch": 1.0344827586206897, "grad_norm": 0.06349515169858932, "learning_rate": 0.00019498077626799968, "loss": 0.5558900833129883, "mean_token_accuracy": 0.8626515477895736, "num_tokens": 30573670.0, "step": 2190 }, { "entropy": 0.6112361473962664, "epoch": 1.0392064241851677, "grad_norm": 0.06788723915815353, "learning_rate": 0.00019493413723273918, "loss": 0.6215075969696044, "mean_token_accuracy": 0.8535605676472187, "num_tokens": 30705296.0, "step": 2200 }, { "entropy": 0.5486962845548987, "epoch": 1.0439300897496457, "grad_norm": 0.06855203211307526, "learning_rate": 0.00019488728814188772, "loss": 0.5528660297393799, "mean_token_accuracy": 0.8610604539513588, "num_tokens": 30842845.0, "step": 2210 }, { "entropy": 0.5727268299087882, "epoch": 1.0486537553141237, "grad_norm": 0.07690499722957611, "learning_rate": 0.00019484022909910573, "loss": 0.5606137275695801, "mean_token_accuracy": 0.8585053950548172, "num_tokens": 30987383.0, "step": 2220 }, { "entropy": 0.6143275920301676, "epoch": 1.0533774208786018, "grad_norm": 0.058543384075164795, "learning_rate": 0.0001947929602085182, "loss": 0.6057152271270752, "mean_token_accuracy": 0.852120365202427, "num_tokens": 31125195.0, "step": 2230 }, { "entropy": 0.559723743982613, "epoch": 1.0581010864430798, "grad_norm": 0.06465470790863037, "learning_rate": 0.00019474548157471443, "loss": 0.5535128593444825, "mean_token_accuracy": 0.864807965606451, "num_tokens": 31269029.0, "step": 2240 }, { "entropy": 0.6108309054747224, "epoch": 1.0628247520075578, "grad_norm": 0.06008979305624962, "learning_rate": 0.0001946977933027478, "loss": 0.6124683856964112, "mean_token_accuracy": 0.8495584823191166, "num_tokens": 31409970.0, "step": 2250 }, { "entropy": 0.6022991143167019, "epoch": 1.067548417572036, "grad_norm": 0.06370719522237778, "learning_rate": 0.00019464989549813555, "loss": 0.5946242809295654, "mean_token_accuracy": 0.8521519854664803, "num_tokens": 31548666.0, "step": 2260 }, { "entropy": 0.5398570565506816, "epoch": 1.072272083136514, "grad_norm": 0.06874193251132965, "learning_rate": 0.00019460178826685853, "loss": 0.5321762084960937, "mean_token_accuracy": 0.8678795844316483, "num_tokens": 31693005.0, "step": 2270 }, { "entropy": 0.5729977376759052, "epoch": 1.076995748700992, "grad_norm": 0.0696796253323555, "learning_rate": 0.00019455347171536103, "loss": 0.5660377979278565, "mean_token_accuracy": 0.863119576126337, "num_tokens": 31829248.0, "step": 2280 }, { "entropy": 0.5336826818063856, "epoch": 1.08171941426547, "grad_norm": 0.04893127456307411, "learning_rate": 0.0001945049459505504, "loss": 0.5218155860900879, "mean_token_accuracy": 0.8675340116024017, "num_tokens": 31961785.0, "step": 2290 }, { "entropy": 0.5194626787677408, "epoch": 1.086443079829948, "grad_norm": 0.0646308958530426, "learning_rate": 0.00019445621107979698, "loss": 0.510835075378418, "mean_token_accuracy": 0.8696671769022941, "num_tokens": 32100659.0, "step": 2300 }, { "entropy": 0.5810475660488009, "epoch": 1.091166745394426, "grad_norm": 0.06206967309117317, "learning_rate": 0.00019440726721093373, "loss": 0.5790423393249512, "mean_token_accuracy": 0.8569555424153805, "num_tokens": 32238717.0, "step": 2310 }, { "entropy": 0.5714470906183123, "epoch": 1.095890410958904, "grad_norm": 0.06636273115873337, "learning_rate": 0.00019435811445225616, "loss": 0.5673160076141357, "mean_token_accuracy": 0.8606886953115463, "num_tokens": 32382320.0, "step": 2320 }, { "entropy": 0.5273692019283771, "epoch": 1.100614076523382, "grad_norm": 0.06179823353886604, "learning_rate": 0.00019430875291252182, "loss": 0.512396764755249, "mean_token_accuracy": 0.8682898908853531, "num_tokens": 32525305.0, "step": 2330 }, { "entropy": 0.5568313036113978, "epoch": 1.10533774208786, "grad_norm": 0.06668296456336975, "learning_rate": 0.00019425918270095033, "loss": 0.5360489368438721, "mean_token_accuracy": 0.8645501948893071, "num_tokens": 32662894.0, "step": 2340 }, { "entropy": 0.5756445590406656, "epoch": 1.1100614076523383, "grad_norm": 0.07136117666959763, "learning_rate": 0.000194209403927223, "loss": 0.5787982940673828, "mean_token_accuracy": 0.8601546064019203, "num_tokens": 32809006.0, "step": 2350 }, { "entropy": 0.532723237015307, "epoch": 1.1147850732168163, "grad_norm": 0.06656378507614136, "learning_rate": 0.0001941594167014826, "loss": 0.5200218200683594, "mean_token_accuracy": 0.8685436010360718, "num_tokens": 32952389.0, "step": 2360 }, { "entropy": 0.5847611540928483, "epoch": 1.1195087387812943, "grad_norm": 0.051534589380025864, "learning_rate": 0.00019410922113433313, "loss": 0.5709563732147217, "mean_token_accuracy": 0.8572433613240719, "num_tokens": 33091437.0, "step": 2370 }, { "entropy": 0.5530805753543973, "epoch": 1.1242324043457723, "grad_norm": 0.0659450888633728, "learning_rate": 0.00019405881733683958, "loss": 0.5454719066619873, "mean_token_accuracy": 0.8633896291255951, "num_tokens": 33231652.0, "step": 2380 }, { "entropy": 0.5709455469623208, "epoch": 1.1289560699102503, "grad_norm": 0.055913567543029785, "learning_rate": 0.00019400820542052775, "loss": 0.5664645671844483, "mean_token_accuracy": 0.862431301176548, "num_tokens": 33375871.0, "step": 2390 }, { "entropy": 0.5522926053032279, "epoch": 1.1336797354747283, "grad_norm": 0.06643116474151611, "learning_rate": 0.00019395738549738377, "loss": 0.5359935760498047, "mean_token_accuracy": 0.8657760076224804, "num_tokens": 33512934.0, "step": 2400 }, { "entropy": 0.5144829558208585, "epoch": 1.1384034010392063, "grad_norm": 0.06935354322195053, "learning_rate": 0.00019390635767985422, "loss": 0.5021259307861328, "mean_token_accuracy": 0.8736603081226348, "num_tokens": 33655842.0, "step": 2410 }, { "entropy": 0.5451503895223141, "epoch": 1.1431270666036846, "grad_norm": 0.05893075093626976, "learning_rate": 0.00019385512208084548, "loss": 0.5468964099884033, "mean_token_accuracy": 0.8673000067472458, "num_tokens": 33790736.0, "step": 2420 }, { "entropy": 0.5215972164645791, "epoch": 1.1478507321681626, "grad_norm": 0.0698215439915657, "learning_rate": 0.0001938036788137238, "loss": 0.5161412239074707, "mean_token_accuracy": 0.8676531538367271, "num_tokens": 33939085.0, "step": 2430 }, { "entropy": 0.5718188390135766, "epoch": 1.1525743977326406, "grad_norm": 0.07106591761112213, "learning_rate": 0.00019375202799231493, "loss": 0.5472604274749756, "mean_token_accuracy": 0.8642895676195621, "num_tokens": 34079164.0, "step": 2440 }, { "entropy": 0.5533800795674324, "epoch": 1.1572980632971186, "grad_norm": 0.06189075484871864, "learning_rate": 0.00019370016973090378, "loss": 0.5506540775299072, "mean_token_accuracy": 0.8638756394386291, "num_tokens": 34213555.0, "step": 2450 }, { "entropy": 0.5459094783291221, "epoch": 1.1620217288615966, "grad_norm": 0.05126383900642395, "learning_rate": 0.00019364810414423427, "loss": 0.5373441696166992, "mean_token_accuracy": 0.8651516288518906, "num_tokens": 34363172.0, "step": 2460 }, { "entropy": 0.5104342142120004, "epoch": 1.1667453944260746, "grad_norm": 0.06708457320928574, "learning_rate": 0.00019359583134750916, "loss": 0.4922313690185547, "mean_token_accuracy": 0.8784280359745026, "num_tokens": 34511555.0, "step": 2470 }, { "entropy": 0.5967657320201397, "epoch": 1.1714690599905526, "grad_norm": 0.058943428099155426, "learning_rate": 0.00019354335145638959, "loss": 0.6136114597320557, "mean_token_accuracy": 0.8560904346406459, "num_tokens": 34646708.0, "step": 2480 }, { "entropy": 0.5279749985784292, "epoch": 1.1761927255550306, "grad_norm": 0.0692218765616417, "learning_rate": 0.00019349066458699494, "loss": 0.5430376529693604, "mean_token_accuracy": 0.8691238626837731, "num_tokens": 34783441.0, "step": 2490 }, { "entropy": 0.5345676735043525, "epoch": 1.1809163911195086, "grad_norm": 0.06222553178668022, "learning_rate": 0.00019343777085590257, "loss": 0.5202937126159668, "mean_token_accuracy": 0.8695143103599549, "num_tokens": 34916658.0, "step": 2500 }, { "entropy": 0.5780558319762349, "epoch": 1.1856400566839869, "grad_norm": 0.057331833988428116, "learning_rate": 0.00019338467038014764, "loss": 0.5633671760559082, "mean_token_accuracy": 0.8644892483949661, "num_tokens": 35060956.0, "step": 2510 }, { "entropy": 0.5132569143548608, "epoch": 1.1903637222484649, "grad_norm": 0.0568612702190876, "learning_rate": 0.00019333136327722252, "loss": 0.5132866382598877, "mean_token_accuracy": 0.8704852610826492, "num_tokens": 35200975.0, "step": 2520 }, { "entropy": 0.5895077161490917, "epoch": 1.1950873878129429, "grad_norm": 0.07144573330879211, "learning_rate": 0.00019327784966507707, "loss": 0.5915816783905029, "mean_token_accuracy": 0.8547378368675709, "num_tokens": 35342784.0, "step": 2530 }, { "entropy": 0.48900156300514935, "epoch": 1.1998110533774209, "grad_norm": 0.07097811996936798, "learning_rate": 0.00019322412966211788, "loss": 0.47861623764038086, "mean_token_accuracy": 0.8806159406900406, "num_tokens": 35475904.0, "step": 2540 }, { "entropy": 0.5424768718890846, "epoch": 1.204534718941899, "grad_norm": 0.05827151611447334, "learning_rate": 0.00019317020338720825, "loss": 0.5270711421966553, "mean_token_accuracy": 0.8665493495762349, "num_tokens": 35618384.0, "step": 2550 }, { "entropy": 0.5501299062743783, "epoch": 1.209258384506377, "grad_norm": 0.07674292474985123, "learning_rate": 0.000193116070959668, "loss": 0.5388836860656738, "mean_token_accuracy": 0.8634671896696091, "num_tokens": 35756183.0, "step": 2560 }, { "entropy": 0.5613229911774397, "epoch": 1.213982050070855, "grad_norm": 0.07786308974027634, "learning_rate": 0.0001930617324992729, "loss": 0.5495596408843995, "mean_token_accuracy": 0.8650944642722607, "num_tokens": 35895269.0, "step": 2570 }, { "entropy": 0.5857510989531874, "epoch": 1.2187057156353331, "grad_norm": 0.06801457703113556, "learning_rate": 0.00019300718812625479, "loss": 0.5794153690338135, "mean_token_accuracy": 0.8595944873988628, "num_tokens": 36042942.0, "step": 2580 }, { "entropy": 0.47505145743489263, "epoch": 1.2234293811998112, "grad_norm": 0.06459733843803406, "learning_rate": 0.00019295243796130097, "loss": 0.46596531867980956, "mean_token_accuracy": 0.879586087167263, "num_tokens": 36183298.0, "step": 2590 }, { "entropy": 0.5188927501440048, "epoch": 1.2281530467642892, "grad_norm": 0.06884653866291046, "learning_rate": 0.00019289748212555414, "loss": 0.5044856071472168, "mean_token_accuracy": 0.8714294105768203, "num_tokens": 36321511.0, "step": 2600 }, { "entropy": 0.5761474084109068, "epoch": 1.2328767123287672, "grad_norm": 0.07391291856765747, "learning_rate": 0.00019284232074061214, "loss": 0.5720541954040528, "mean_token_accuracy": 0.8583825193345547, "num_tokens": 36459611.0, "step": 2610 }, { "entropy": 0.531072998046875, "epoch": 1.2376003778932452, "grad_norm": 0.07800538092851639, "learning_rate": 0.0001927869539285275, "loss": 0.5167572975158692, "mean_token_accuracy": 0.8735670812427998, "num_tokens": 36600442.0, "step": 2620 }, { "entropy": 0.5576183291152119, "epoch": 1.2423240434577232, "grad_norm": 0.0661255493760109, "learning_rate": 0.00019273138181180737, "loss": 0.5394849300384521, "mean_token_accuracy": 0.8654220625758171, "num_tokens": 36744616.0, "step": 2630 }, { "entropy": 0.5643019031733274, "epoch": 1.2470477090222012, "grad_norm": 0.07070118188858032, "learning_rate": 0.00019267560451341313, "loss": 0.5709936141967773, "mean_token_accuracy": 0.8607587151229381, "num_tokens": 36885224.0, "step": 2640 }, { "entropy": 0.49656188022345304, "epoch": 1.2517713745866792, "grad_norm": 0.05357030779123306, "learning_rate": 0.0001926196221567602, "loss": 0.4949103832244873, "mean_token_accuracy": 0.8784331850707531, "num_tokens": 37021048.0, "step": 2650 }, { "entropy": 0.5479259440675378, "epoch": 1.2564950401511572, "grad_norm": 0.055224936455488205, "learning_rate": 0.00019256343486571771, "loss": 0.540528678894043, "mean_token_accuracy": 0.8648318573832512, "num_tokens": 37150043.0, "step": 2660 }, { "entropy": 0.5205691078677773, "epoch": 1.2612187057156352, "grad_norm": 0.06225382909178734, "learning_rate": 0.00019250704276460812, "loss": 0.5114886283874511, "mean_token_accuracy": 0.8723285183310509, "num_tokens": 37286490.0, "step": 2670 }, { "entropy": 0.526428890787065, "epoch": 1.2659423712801134, "grad_norm": 0.07201260328292847, "learning_rate": 0.00019245044597820726, "loss": 0.5282705783843994, "mean_token_accuracy": 0.869306180626154, "num_tokens": 37430286.0, "step": 2680 }, { "entropy": 0.5554586604237557, "epoch": 1.2706660368445915, "grad_norm": 0.07201368361711502, "learning_rate": 0.00019239364463174367, "loss": 0.5310179710388183, "mean_token_accuracy": 0.8649682223796844, "num_tokens": 37573458.0, "step": 2690 }, { "entropy": 0.5209276856854558, "epoch": 1.2753897024090695, "grad_norm": 0.06191640719771385, "learning_rate": 0.00019233663885089865, "loss": 0.5082337379455566, "mean_token_accuracy": 0.8736192300915718, "num_tokens": 37714516.0, "step": 2700 }, { "entropy": 0.5643254201859236, "epoch": 1.2801133679735475, "grad_norm": 0.06433577090501785, "learning_rate": 0.0001922794287618058, "loss": 0.5594176292419434, "mean_token_accuracy": 0.8620877884328365, "num_tokens": 37854181.0, "step": 2710 }, { "entropy": 0.5228854121640325, "epoch": 1.2848370335380255, "grad_norm": 0.056869398802518845, "learning_rate": 0.0001922220144910507, "loss": 0.5108164787292481, "mean_token_accuracy": 0.8693295270204544, "num_tokens": 37995429.0, "step": 2720 }, { "entropy": 0.5236076887696981, "epoch": 1.2895606991025035, "grad_norm": 0.07903646677732468, "learning_rate": 0.0001921643961656708, "loss": 0.5172504425048828, "mean_token_accuracy": 0.8720158644020557, "num_tokens": 38137137.0, "step": 2730 }, { "entropy": 0.5044055638834835, "epoch": 1.2942843646669817, "grad_norm": 0.06947501003742218, "learning_rate": 0.00019210657391315508, "loss": 0.503062391281128, "mean_token_accuracy": 0.8754982620477676, "num_tokens": 38268165.0, "step": 2740 }, { "entropy": 0.5331376258283853, "epoch": 1.2990080302314597, "grad_norm": 0.07525797188282013, "learning_rate": 0.00019204854786144363, "loss": 0.5150176525115967, "mean_token_accuracy": 0.8699349507689476, "num_tokens": 38400090.0, "step": 2750 }, { "entropy": 0.6020360633730888, "epoch": 1.3037316957959377, "grad_norm": 0.06762351095676422, "learning_rate": 0.00019199031813892757, "loss": 0.5924798965454101, "mean_token_accuracy": 0.85457673817873, "num_tokens": 38536283.0, "step": 2760 }, { "entropy": 0.5128505306318403, "epoch": 1.3084553613604157, "grad_norm": 0.07126962393522263, "learning_rate": 0.00019193188487444857, "loss": 0.5211223602294922, "mean_token_accuracy": 0.8695483796298504, "num_tokens": 38670762.0, "step": 2770 }, { "entropy": 0.4951697552576661, "epoch": 1.3131790269248937, "grad_norm": 0.06562139838933945, "learning_rate": 0.00019187324819729882, "loss": 0.4793668746948242, "mean_token_accuracy": 0.8797172121703625, "num_tokens": 38806779.0, "step": 2780 }, { "entropy": 0.549693813174963, "epoch": 1.3179026924893718, "grad_norm": 0.07003356516361237, "learning_rate": 0.0001918144082372204, "loss": 0.5359210014343262, "mean_token_accuracy": 0.8648910485208035, "num_tokens": 38954406.0, "step": 2790 }, { "entropy": 0.5086442392319441, "epoch": 1.3226263580538498, "grad_norm": 0.0650491937994957, "learning_rate": 0.00019175536512440537, "loss": 0.5049900531768798, "mean_token_accuracy": 0.8776748277246952, "num_tokens": 39084233.0, "step": 2800 }, { "entropy": 0.4928426995873451, "epoch": 1.3273500236183278, "grad_norm": 0.06262879073619843, "learning_rate": 0.00019169611898949517, "loss": 0.4780909538269043, "mean_token_accuracy": 0.8800336785614491, "num_tokens": 39219447.0, "step": 2810 }, { "entropy": 0.5353525076061487, "epoch": 1.3320736891828058, "grad_norm": 0.06984758377075195, "learning_rate": 0.00019163666996358046, "loss": 0.5310516357421875, "mean_token_accuracy": 0.8697080694139003, "num_tokens": 39363717.0, "step": 2820 }, { "entropy": 0.5275251345708967, "epoch": 1.3367973547472838, "grad_norm": 0.09146755188703537, "learning_rate": 0.00019157701817820087, "loss": 0.5290673732757568, "mean_token_accuracy": 0.8689099937677384, "num_tokens": 39503384.0, "step": 2830 }, { "entropy": 0.5574715923517942, "epoch": 1.3415210203117618, "grad_norm": 0.07356512546539307, "learning_rate": 0.00019151716376534475, "loss": 0.5338760852813721, "mean_token_accuracy": 0.8645951241254807, "num_tokens": 39644932.0, "step": 2840 }, { "entropy": 0.5245337123051286, "epoch": 1.34624468587624, "grad_norm": 0.058665141463279724, "learning_rate": 0.00019145710685744858, "loss": 0.5056859016418457, "mean_token_accuracy": 0.871939604729414, "num_tokens": 39788930.0, "step": 2850 }, { "entropy": 0.5012327982112765, "epoch": 1.350968351440718, "grad_norm": 0.05447728931903839, "learning_rate": 0.00019139684758739706, "loss": 0.48896212577819825, "mean_token_accuracy": 0.877267012745142, "num_tokens": 39935030.0, "step": 2860 }, { "entropy": 0.5086068087257445, "epoch": 1.355692017005196, "grad_norm": 0.07317899167537689, "learning_rate": 0.0001913363860885226, "loss": 0.49643878936767577, "mean_token_accuracy": 0.8740778245031834, "num_tokens": 40077034.0, "step": 2870 }, { "entropy": 0.4870829051360488, "epoch": 1.360415682569674, "grad_norm": 0.07304569333791733, "learning_rate": 0.00019127572249460502, "loss": 0.4915927886962891, "mean_token_accuracy": 0.8796208336949348, "num_tokens": 40214131.0, "step": 2880 }, { "entropy": 0.4913898404687643, "epoch": 1.365139348134152, "grad_norm": 0.06799453496932983, "learning_rate": 0.0001912148569398714, "loss": 0.47771739959716797, "mean_token_accuracy": 0.8788500644266606, "num_tokens": 40352766.0, "step": 2890 }, { "entropy": 0.4784863717854023, "epoch": 1.36986301369863, "grad_norm": 0.08276737481355667, "learning_rate": 0.00019115378955899558, "loss": 0.4630417346954346, "mean_token_accuracy": 0.8837412089109421, "num_tokens": 40492488.0, "step": 2900 }, { "entropy": 0.5437351418659091, "epoch": 1.3745866792631083, "grad_norm": 0.06532137840986252, "learning_rate": 0.0001910925204870981, "loss": 0.5189805507659913, "mean_token_accuracy": 0.8683592341840267, "num_tokens": 40627891.0, "step": 2910 }, { "entropy": 0.49372219778597354, "epoch": 1.3793103448275863, "grad_norm": 0.06270914524793625, "learning_rate": 0.0001910310498597456, "loss": 0.4856466293334961, "mean_token_accuracy": 0.8784211046993733, "num_tokens": 40767077.0, "step": 2920 }, { "entropy": 0.5272138256579637, "epoch": 1.3840340103920643, "grad_norm": 0.0641990453004837, "learning_rate": 0.00019096937781295085, "loss": 0.5140623569488525, "mean_token_accuracy": 0.8715197399258614, "num_tokens": 40906006.0, "step": 2930 }, { "entropy": 0.5239824512973428, "epoch": 1.3887576759565423, "grad_norm": 0.07271122187376022, "learning_rate": 0.00019090750448317223, "loss": 0.5042078971862793, "mean_token_accuracy": 0.8712317928671837, "num_tokens": 41039686.0, "step": 2940 }, { "entropy": 0.5243585165590048, "epoch": 1.3934813415210203, "grad_norm": 0.07042913883924484, "learning_rate": 0.00019084543000731345, "loss": 0.5045619487762452, "mean_token_accuracy": 0.8734136171638965, "num_tokens": 41185421.0, "step": 2950 }, { "entropy": 0.5187683286145329, "epoch": 1.3982050070854983, "grad_norm": 0.06604288518428802, "learning_rate": 0.00019078315452272334, "loss": 0.5099934101104736, "mean_token_accuracy": 0.8721372559666634, "num_tokens": 41323693.0, "step": 2960 }, { "entropy": 0.5132398827001452, "epoch": 1.4029286726499763, "grad_norm": 0.06239806115627289, "learning_rate": 0.00019072067816719549, "loss": 0.5069376468658447, "mean_token_accuracy": 0.8741942852735519, "num_tokens": 41461381.0, "step": 2970 }, { "entropy": 0.5633071081712842, "epoch": 1.4076523382144543, "grad_norm": 0.07068385928869247, "learning_rate": 0.00019065800107896788, "loss": 0.5480347633361816, "mean_token_accuracy": 0.8667258083820343, "num_tokens": 41605603.0, "step": 2980 }, { "entropy": 0.4838137686252594, "epoch": 1.4123760037789324, "grad_norm": 0.06998838484287262, "learning_rate": 0.0001905951233967227, "loss": 0.48467297554016114, "mean_token_accuracy": 0.8761447243392467, "num_tokens": 41741505.0, "step": 2990 }, { "entropy": 0.5486295022070408, "epoch": 1.4170996693434104, "grad_norm": 0.05402971804141998, "learning_rate": 0.00019053204525958602, "loss": 0.5331336975097656, "mean_token_accuracy": 0.8677968755364418, "num_tokens": 41871433.0, "step": 3000 }, { "entropy": 0.47180736158043146, "epoch": 1.4218233349078886, "grad_norm": 0.0684208869934082, "learning_rate": 0.0001904687668071273, "loss": 0.4505447864532471, "mean_token_accuracy": 0.8834298044443131, "num_tokens": 42015316.0, "step": 3010 }, { "entropy": 0.5403513418510556, "epoch": 1.4265470004723666, "grad_norm": 0.05338026583194733, "learning_rate": 0.00019040528817935944, "loss": 0.5551342964172363, "mean_token_accuracy": 0.8633888840675354, "num_tokens": 42153151.0, "step": 3020 }, { "entropy": 0.5293775004334748, "epoch": 1.4312706660368446, "grad_norm": 0.06934598088264465, "learning_rate": 0.00019034160951673804, "loss": 0.5058796882629395, "mean_token_accuracy": 0.8730652637779712, "num_tokens": 42293214.0, "step": 3030 }, { "entropy": 0.5343153689056634, "epoch": 1.4359943316013226, "grad_norm": 0.061721064150333405, "learning_rate": 0.00019027773096016147, "loss": 0.5258009433746338, "mean_token_accuracy": 0.8706065185368062, "num_tokens": 42433348.0, "step": 3040 }, { "entropy": 0.5479841718450189, "epoch": 1.4407179971658006, "grad_norm": 0.05038747936487198, "learning_rate": 0.00019021365265097027, "loss": 0.5306890964508056, "mean_token_accuracy": 0.8660327635705471, "num_tokens": 42575411.0, "step": 3050 }, { "entropy": 0.5482240395620466, "epoch": 1.4454416627302786, "grad_norm": 0.07062775641679764, "learning_rate": 0.00019014937473094706, "loss": 0.5523650646209717, "mean_token_accuracy": 0.8641908958554267, "num_tokens": 42708786.0, "step": 3060 }, { "entropy": 0.494945321790874, "epoch": 1.4501653282947569, "grad_norm": 0.07384450733661652, "learning_rate": 0.00019008489734231613, "loss": 0.49300556182861327, "mean_token_accuracy": 0.8807456009089947, "num_tokens": 42839904.0, "step": 3070 }, { "entropy": 0.5401355775073171, "epoch": 1.4548889938592349, "grad_norm": 0.06551490724086761, "learning_rate": 0.00019002022062774302, "loss": 0.5346176624298096, "mean_token_accuracy": 0.870025560259819, "num_tokens": 42981810.0, "step": 3080 }, { "entropy": 0.5414276383817196, "epoch": 1.4596126594237129, "grad_norm": 0.06896756589412689, "learning_rate": 0.00018995534473033437, "loss": 0.5193339347839355, "mean_token_accuracy": 0.8681399069726468, "num_tokens": 43125106.0, "step": 3090 }, { "entropy": 0.5742442956194281, "epoch": 1.4643363249881909, "grad_norm": 0.06355969607830048, "learning_rate": 0.0001898902697936376, "loss": 0.5853897094726562, "mean_token_accuracy": 0.8584777720272541, "num_tokens": 43267336.0, "step": 3100 }, { "entropy": 0.49505286756902933, "epoch": 1.469059990552669, "grad_norm": 0.061108529567718506, "learning_rate": 0.0001898249959616404, "loss": 0.47298288345336914, "mean_token_accuracy": 0.8794999420642853, "num_tokens": 43409209.0, "step": 3110 }, { "entropy": 0.5030368724837899, "epoch": 1.473783656117147, "grad_norm": 0.07009301334619522, "learning_rate": 0.0001897595233787707, "loss": 0.48371176719665526, "mean_token_accuracy": 0.8776668392121791, "num_tokens": 43547233.0, "step": 3120 }, { "entropy": 0.4946042187511921, "epoch": 1.478507321681625, "grad_norm": 0.055976901203393936, "learning_rate": 0.00018969385218989607, "loss": 0.48110108375549315, "mean_token_accuracy": 0.8766251891851425, "num_tokens": 43690554.0, "step": 3130 }, { "entropy": 0.5155046440660953, "epoch": 1.483230987246103, "grad_norm": 0.06256508082151413, "learning_rate": 0.00018962798254032354, "loss": 0.49575061798095704, "mean_token_accuracy": 0.8766602538526058, "num_tokens": 43835981.0, "step": 3140 }, { "entropy": 0.5539642686024309, "epoch": 1.487954652810581, "grad_norm": 0.068775475025177, "learning_rate": 0.00018956191457579938, "loss": 0.5431741714477539, "mean_token_accuracy": 0.8645640470087528, "num_tokens": 43977358.0, "step": 3150 }, { "entropy": 0.5029204994440079, "epoch": 1.492678318375059, "grad_norm": 0.07776771485805511, "learning_rate": 0.00018949564844250846, "loss": 0.49552211761474607, "mean_token_accuracy": 0.8764367088675499, "num_tokens": 44126230.0, "step": 3160 }, { "entropy": 0.5319400677457452, "epoch": 1.4974019839395372, "grad_norm": 0.06949610263109207, "learning_rate": 0.00018942918428707433, "loss": 0.5000473976135253, "mean_token_accuracy": 0.8752985373139381, "num_tokens": 44267939.0, "step": 3170 }, { "entropy": 0.4848332326859236, "epoch": 1.5021256495040152, "grad_norm": 0.07205299288034439, "learning_rate": 0.0001893625222565586, "loss": 0.487746524810791, "mean_token_accuracy": 0.8753070525825024, "num_tokens": 44413254.0, "step": 3180 }, { "entropy": 0.5334670664742589, "epoch": 1.5068493150684932, "grad_norm": 0.06868889182806015, "learning_rate": 0.00018929566249846068, "loss": 0.5071711540222168, "mean_token_accuracy": 0.870256232470274, "num_tokens": 44557768.0, "step": 3190 }, { "entropy": 0.4812082316726446, "epoch": 1.5115729806329712, "grad_norm": 0.06909193843603134, "learning_rate": 0.0001892286051607175, "loss": 0.46418137550354005, "mean_token_accuracy": 0.881144455075264, "num_tokens": 44699319.0, "step": 3200 }, { "entropy": 0.5083514029160142, "epoch": 1.5162966461974492, "grad_norm": 0.06829367578029633, "learning_rate": 0.00018916135039170323, "loss": 0.4949441909790039, "mean_token_accuracy": 0.8760163858532906, "num_tokens": 44834568.0, "step": 3210 }, { "entropy": 0.4428059158846736, "epoch": 1.5210203117619272, "grad_norm": 0.07612506300210953, "learning_rate": 0.0001890938983402288, "loss": 0.42614378929138186, "mean_token_accuracy": 0.8896358072757721, "num_tokens": 44974071.0, "step": 3220 }, { "entropy": 0.48266046196222306, "epoch": 1.5257439773264054, "grad_norm": 0.07178372889757156, "learning_rate": 0.00018902624915554165, "loss": 0.4783198833465576, "mean_token_accuracy": 0.8790047898888588, "num_tokens": 45113823.0, "step": 3230 }, { "entropy": 0.5540031120181084, "epoch": 1.5304676428908834, "grad_norm": 0.06889275461435318, "learning_rate": 0.0001889584029873255, "loss": 0.534952449798584, "mean_token_accuracy": 0.8661975428462029, "num_tokens": 45255853.0, "step": 3240 }, { "entropy": 0.5310316779650748, "epoch": 1.5351913084553614, "grad_norm": 0.07437170296907425, "learning_rate": 0.00018889035998569985, "loss": 0.5156550884246827, "mean_token_accuracy": 0.8707649812102318, "num_tokens": 45399198.0, "step": 3250 }, { "entropy": 0.5045477893203497, "epoch": 1.5399149740198395, "grad_norm": 0.07557965815067291, "learning_rate": 0.00018882212030121973, "loss": 0.49659004211425783, "mean_token_accuracy": 0.8758061565458775, "num_tokens": 45543272.0, "step": 3260 }, { "entropy": 0.4774668848142028, "epoch": 1.5446386395843175, "grad_norm": 0.05187695845961571, "learning_rate": 0.0001887536840848754, "loss": 0.4510806083679199, "mean_token_accuracy": 0.8840405188500882, "num_tokens": 45685360.0, "step": 3270 }, { "entropy": 0.5271233171224594, "epoch": 1.5493623051487955, "grad_norm": 0.06921805441379547, "learning_rate": 0.00018868505148809193, "loss": 0.5342346668243408, "mean_token_accuracy": 0.8704235851764679, "num_tokens": 45821887.0, "step": 3280 }, { "entropy": 0.48698661047965286, "epoch": 1.5540859707132735, "grad_norm": 0.06425916403532028, "learning_rate": 0.0001886162226627289, "loss": 0.47869176864624025, "mean_token_accuracy": 0.880816039443016, "num_tokens": 45961940.0, "step": 3290 }, { "entropy": 0.5056364919990302, "epoch": 1.5588096362777515, "grad_norm": 0.07033051550388336, "learning_rate": 0.00018854719776108012, "loss": 0.48381242752075193, "mean_token_accuracy": 0.8775354057550431, "num_tokens": 46099097.0, "step": 3300 }, { "entropy": 0.47637706957757475, "epoch": 1.5635333018422295, "grad_norm": 0.06629955768585205, "learning_rate": 0.00018847797693587323, "loss": 0.44725780487060546, "mean_token_accuracy": 0.8840812586247921, "num_tokens": 46245914.0, "step": 3310 }, { "entropy": 0.5118729459121824, "epoch": 1.5682569674067075, "grad_norm": 0.0727744773030281, "learning_rate": 0.0001884085603402694, "loss": 0.5179985046386719, "mean_token_accuracy": 0.8718493245542049, "num_tokens": 46381834.0, "step": 3320 }, { "entropy": 0.4960333503782749, "epoch": 1.5729806329711855, "grad_norm": 0.09013896435499191, "learning_rate": 0.00018833894812786292, "loss": 0.4809901237487793, "mean_token_accuracy": 0.8768595829606056, "num_tokens": 46517761.0, "step": 3330 }, { "entropy": 0.488414510153234, "epoch": 1.5777042985356635, "grad_norm": 0.07220719009637833, "learning_rate": 0.00018826914045268095, "loss": 0.46564831733703616, "mean_token_accuracy": 0.8816016264259815, "num_tokens": 46659649.0, "step": 3340 }, { "entropy": 0.47431168239563704, "epoch": 1.5824279641001417, "grad_norm": 0.06044609844684601, "learning_rate": 0.00018819913746918307, "loss": 0.45941834449768065, "mean_token_accuracy": 0.8830493785440922, "num_tokens": 46800020.0, "step": 3350 }, { "entropy": 0.5145462546497583, "epoch": 1.5871516296646198, "grad_norm": 0.06516613066196442, "learning_rate": 0.00018812893933226117, "loss": 0.5173689842224121, "mean_token_accuracy": 0.8722536742687226, "num_tokens": 46945890.0, "step": 3360 }, { "entropy": 0.46284727547317744, "epoch": 1.5918752952290978, "grad_norm": 0.0704319179058075, "learning_rate": 0.00018805854619723876, "loss": 0.45035176277160643, "mean_token_accuracy": 0.8863627515733242, "num_tokens": 47085983.0, "step": 3370 }, { "entropy": 0.47932797148823736, "epoch": 1.5965989607935758, "grad_norm": 0.06302246451377869, "learning_rate": 0.0001879879582198709, "loss": 0.4641761302947998, "mean_token_accuracy": 0.884004057943821, "num_tokens": 47219446.0, "step": 3380 }, { "entropy": 0.49306914713233707, "epoch": 1.601322626358054, "grad_norm": 0.09234043210744858, "learning_rate": 0.00018791717555634386, "loss": 0.49100627899169924, "mean_token_accuracy": 0.8744032070040703, "num_tokens": 47363147.0, "step": 3390 }, { "entropy": 0.5001145921647548, "epoch": 1.606046291922532, "grad_norm": 0.06382778286933899, "learning_rate": 0.00018784619836327443, "loss": 0.47536821365356446, "mean_token_accuracy": 0.8797802373766899, "num_tokens": 47506259.0, "step": 3400 }, { "entropy": 0.5013880920596421, "epoch": 1.61076995748701, "grad_norm": 0.05685625970363617, "learning_rate": 0.0001877750267977101, "loss": 0.4933168411254883, "mean_token_accuracy": 0.8751931264996529, "num_tokens": 47646962.0, "step": 3410 }, { "entropy": 0.48260427080094814, "epoch": 1.615493623051488, "grad_norm": 0.05712353065609932, "learning_rate": 0.0001877036610171282, "loss": 0.4721681118011475, "mean_token_accuracy": 0.8826416373252869, "num_tokens": 47788841.0, "step": 3420 }, { "entropy": 0.4521892111748457, "epoch": 1.620217288615966, "grad_norm": 0.06335393339395523, "learning_rate": 0.00018763210117943603, "loss": 0.44668188095092776, "mean_token_accuracy": 0.8898941166698933, "num_tokens": 47926380.0, "step": 3430 }, { "entropy": 0.480171043984592, "epoch": 1.624940954180444, "grad_norm": 0.09493982046842575, "learning_rate": 0.00018756034744297, "loss": 0.4756453037261963, "mean_token_accuracy": 0.8842961005866528, "num_tokens": 48071607.0, "step": 3440 }, { "entropy": 0.5601227654144167, "epoch": 1.629664619744922, "grad_norm": 0.13700838387012482, "learning_rate": 0.00018748839996649583, "loss": 0.5653737545013428, "mean_token_accuracy": 0.8629151083528995, "num_tokens": 48214472.0, "step": 3450 }, { "entropy": 0.5410040715709329, "epoch": 1.6343882853094, "grad_norm": 0.2977743148803711, "learning_rate": 0.00018741625890920772, "loss": 0.5802890300750733, "mean_token_accuracy": 0.8620551273226738, "num_tokens": 48354771.0, "step": 3460 }, { "entropy": 0.5264073548838496, "epoch": 1.639111950873878, "grad_norm": 0.07925760746002197, "learning_rate": 0.00018734392443072823, "loss": 0.5069135189056396, "mean_token_accuracy": 0.8724097192287446, "num_tokens": 48491163.0, "step": 3470 }, { "entropy": 0.4907891957089305, "epoch": 1.643835616438356, "grad_norm": 0.061221349984407425, "learning_rate": 0.00018727139669110797, "loss": 0.46407413482666016, "mean_token_accuracy": 0.8809487789869308, "num_tokens": 48637400.0, "step": 3480 }, { "entropy": 0.4608781412243843, "epoch": 1.648559282002834, "grad_norm": 0.0795220136642456, "learning_rate": 0.00018719867585082506, "loss": 0.44829983711242677, "mean_token_accuracy": 0.8841659784317016, "num_tokens": 48778644.0, "step": 3490 }, { "entropy": 0.48218404334038495, "epoch": 1.653282947567312, "grad_norm": 0.07065624743700027, "learning_rate": 0.00018712576207078496, "loss": 0.4626615524291992, "mean_token_accuracy": 0.8831795863807201, "num_tokens": 48926082.0, "step": 3500 }, { "entropy": 0.5286676125600934, "epoch": 1.6580066131317903, "grad_norm": 0.08479559421539307, "learning_rate": 0.00018705265551232004, "loss": 0.5079413890838623, "mean_token_accuracy": 0.8700848676264286, "num_tokens": 49071371.0, "step": 3510 }, { "entropy": 0.5223700776696205, "epoch": 1.6627302786962683, "grad_norm": 0.0745716392993927, "learning_rate": 0.00018697935633718918, "loss": 0.5000300884246827, "mean_token_accuracy": 0.8746945030987263, "num_tokens": 49205482.0, "step": 3520 }, { "entropy": 0.4563616782426834, "epoch": 1.6674539442607463, "grad_norm": 0.06475856900215149, "learning_rate": 0.0001869058647075775, "loss": 0.4412206172943115, "mean_token_accuracy": 0.8899600148200989, "num_tokens": 49343976.0, "step": 3530 }, { "entropy": 0.42491830242797735, "epoch": 1.6721776098252243, "grad_norm": 0.07917920500040054, "learning_rate": 0.00018683218078609587, "loss": 0.4145365238189697, "mean_token_accuracy": 0.891805000603199, "num_tokens": 49481604.0, "step": 3540 }, { "entropy": 0.4897112654522061, "epoch": 1.6769012753897026, "grad_norm": 0.06935668736696243, "learning_rate": 0.00018675830473578072, "loss": 0.46971497535705564, "mean_token_accuracy": 0.8802993208169937, "num_tokens": 49614451.0, "step": 3550 }, { "entropy": 0.5100322388112545, "epoch": 1.6816249409541806, "grad_norm": 0.06888169795274734, "learning_rate": 0.0001866842367200935, "loss": 0.5125060558319092, "mean_token_accuracy": 0.8740247212350368, "num_tokens": 49754058.0, "step": 3560 }, { "entropy": 0.49550435338169335, "epoch": 1.6863486065186586, "grad_norm": 0.06417348980903625, "learning_rate": 0.00018660997690292058, "loss": 0.4704303741455078, "mean_token_accuracy": 0.8816084496676921, "num_tokens": 49895130.0, "step": 3570 }, { "entropy": 0.5129264136776328, "epoch": 1.6910722720831366, "grad_norm": 0.07550029456615448, "learning_rate": 0.00018653552544857255, "loss": 0.5099283218383789, "mean_token_accuracy": 0.8740891300141811, "num_tokens": 50042421.0, "step": 3580 }, { "entropy": 0.522415111027658, "epoch": 1.6957959376476146, "grad_norm": 0.07896347343921661, "learning_rate": 0.00018646088252178409, "loss": 0.5285237789154053, "mean_token_accuracy": 0.8709682621061802, "num_tokens": 50178627.0, "step": 3590 }, { "entropy": 0.5415722474455833, "epoch": 1.7005196032120926, "grad_norm": 0.07830450683832169, "learning_rate": 0.0001863860482877135, "loss": 0.5372303485870361, "mean_token_accuracy": 0.8707671351730824, "num_tokens": 50320773.0, "step": 3600 }, { "entropy": 0.49348885137587784, "epoch": 1.7052432687765706, "grad_norm": 0.09282948821783066, "learning_rate": 0.00018631102291194246, "loss": 0.48967490196228025, "mean_token_accuracy": 0.8795689433813095, "num_tokens": 50459443.0, "step": 3610 }, { "entropy": 0.49875389523804187, "epoch": 1.7099669343410486, "grad_norm": 0.06347214430570602, "learning_rate": 0.00018623580656047553, "loss": 0.46561007499694823, "mean_token_accuracy": 0.879444283246994, "num_tokens": 50598648.0, "step": 3620 }, { "entropy": 0.49584178645163773, "epoch": 1.7146905999055266, "grad_norm": 0.07094968855381012, "learning_rate": 0.00018616039939973978, "loss": 0.4932081699371338, "mean_token_accuracy": 0.875949389487505, "num_tokens": 50745863.0, "step": 3630 }, { "entropy": 0.45952941924333573, "epoch": 1.7194142654700046, "grad_norm": 0.061786849051713943, "learning_rate": 0.0001860848015965846, "loss": 0.4315204620361328, "mean_token_accuracy": 0.889579851925373, "num_tokens": 50888365.0, "step": 3640 }, { "entropy": 0.5100168336182833, "epoch": 1.7241379310344827, "grad_norm": 0.07355868816375732, "learning_rate": 0.00018600901331828108, "loss": 0.4933262825012207, "mean_token_accuracy": 0.8758195139467716, "num_tokens": 51028272.0, "step": 3650 }, { "entropy": 0.4843357313424349, "epoch": 1.7288615965989607, "grad_norm": 0.0832064226269722, "learning_rate": 0.0001859330347325218, "loss": 0.46827449798583987, "mean_token_accuracy": 0.8799601912498474, "num_tokens": 51161711.0, "step": 3660 }, { "entropy": 0.47955589797347786, "epoch": 1.7335852621634387, "grad_norm": 0.06273242831230164, "learning_rate": 0.00018585686600742043, "loss": 0.46647076606750487, "mean_token_accuracy": 0.8856062985956669, "num_tokens": 51303784.0, "step": 3670 }, { "entropy": 0.4714097175747156, "epoch": 1.738308927727917, "grad_norm": 0.060745831578969955, "learning_rate": 0.00018578050731151143, "loss": 0.45835061073303224, "mean_token_accuracy": 0.88508874848485, "num_tokens": 51445840.0, "step": 3680 }, { "entropy": 0.4783116027712822, "epoch": 1.743032593292395, "grad_norm": 0.07195068895816803, "learning_rate": 0.00018570395881374946, "loss": 0.4727510929107666, "mean_token_accuracy": 0.8812548741698265, "num_tokens": 51592134.0, "step": 3690 }, { "entropy": 0.48504022927954793, "epoch": 1.747756258856873, "grad_norm": 0.07684134691953659, "learning_rate": 0.00018562722068350923, "loss": 0.4808910369873047, "mean_token_accuracy": 0.8813853487372398, "num_tokens": 51740003.0, "step": 3700 }, { "entropy": 0.5327062288299202, "epoch": 1.752479924421351, "grad_norm": 0.07457150518894196, "learning_rate": 0.000185550293090585, "loss": 0.5147149562835693, "mean_token_accuracy": 0.870915912091732, "num_tokens": 51876396.0, "step": 3710 }, { "entropy": 0.5028876660391688, "epoch": 1.7572035899858292, "grad_norm": 0.09055812656879425, "learning_rate": 0.00018547317620519027, "loss": 0.4847529411315918, "mean_token_accuracy": 0.8766453407704831, "num_tokens": 52018513.0, "step": 3720 }, { "entropy": 0.44939340762794017, "epoch": 1.7619272555503072, "grad_norm": 0.07099083065986633, "learning_rate": 0.00018539587019795738, "loss": 0.4363709926605225, "mean_token_accuracy": 0.8911823235452175, "num_tokens": 52153059.0, "step": 3730 }, { "entropy": 0.49043529834598304, "epoch": 1.7666509211147852, "grad_norm": 0.0828876793384552, "learning_rate": 0.00018531837523993712, "loss": 0.4825395107269287, "mean_token_accuracy": 0.8788322381675243, "num_tokens": 52295004.0, "step": 3740 }, { "entropy": 0.4715043371543288, "epoch": 1.7713745866792632, "grad_norm": 0.06490743905305862, "learning_rate": 0.00018524069150259833, "loss": 0.4430831432342529, "mean_token_accuracy": 0.88459402769804, "num_tokens": 52432676.0, "step": 3750 }, { "entropy": 0.5162357835099101, "epoch": 1.7760982522437412, "grad_norm": 0.07117088884115219, "learning_rate": 0.00018516281915782762, "loss": 0.5103609085083007, "mean_token_accuracy": 0.8724541738629341, "num_tokens": 52575356.0, "step": 3760 }, { "entropy": 0.4511245688423514, "epoch": 1.7808219178082192, "grad_norm": 0.0680798590183258, "learning_rate": 0.00018508475837792886, "loss": 0.42701416015625, "mean_token_accuracy": 0.889444986730814, "num_tokens": 52717023.0, "step": 3770 }, { "entropy": 0.4780085539445281, "epoch": 1.7855455833726972, "grad_norm": 0.08125492185354233, "learning_rate": 0.0001850065093356229, "loss": 0.4679851055145264, "mean_token_accuracy": 0.881610245257616, "num_tokens": 52855010.0, "step": 3780 }, { "entropy": 0.4280642387457192, "epoch": 1.7902692489371752, "grad_norm": 0.05039487034082413, "learning_rate": 0.0001849280722040471, "loss": 0.3990738868713379, "mean_token_accuracy": 0.8957292176783085, "num_tokens": 52997980.0, "step": 3790 }, { "entropy": 0.49470355790108445, "epoch": 1.7949929145016532, "grad_norm": 0.07032407820224762, "learning_rate": 0.00018484944715675504, "loss": 0.48061184883117675, "mean_token_accuracy": 0.879206795245409, "num_tokens": 53132237.0, "step": 3800 }, { "entropy": 0.42146230600774287, "epoch": 1.7997165800661312, "grad_norm": 0.06520136445760727, "learning_rate": 0.0001847706343677161, "loss": 0.40451779365539553, "mean_token_accuracy": 0.8958424828946591, "num_tokens": 53267497.0, "step": 3810 }, { "entropy": 0.44999085199087857, "epoch": 1.8044402456306092, "grad_norm": 0.06635092198848724, "learning_rate": 0.00018469163401131503, "loss": 0.43325080871582033, "mean_token_accuracy": 0.8897008091211319, "num_tokens": 53401000.0, "step": 3820 }, { "entropy": 0.4340560780838132, "epoch": 1.8091639111950872, "grad_norm": 0.09158578515052795, "learning_rate": 0.0001846124462623516, "loss": 0.4233553886413574, "mean_token_accuracy": 0.8928570128977299, "num_tokens": 53532302.0, "step": 3830 }, { "entropy": 0.475423426553607, "epoch": 1.8138875767595655, "grad_norm": 0.05990416929125786, "learning_rate": 0.00018453307129604024, "loss": 0.4380009651184082, "mean_token_accuracy": 0.8848062686622142, "num_tokens": 53676184.0, "step": 3840 }, { "entropy": 0.42227997723966837, "epoch": 1.8186112423240435, "grad_norm": 0.06041109934449196, "learning_rate": 0.0001844535092880096, "loss": 0.41153559684753416, "mean_token_accuracy": 0.8953505516052246, "num_tokens": 53806899.0, "step": 3850 }, { "entropy": 0.5063869832083583, "epoch": 1.8233349078885215, "grad_norm": 0.07307017594575882, "learning_rate": 0.00018437376041430218, "loss": 0.49002633094787595, "mean_token_accuracy": 0.8778577506542206, "num_tokens": 53944404.0, "step": 3860 }, { "entropy": 0.461343062762171, "epoch": 1.8280585734529995, "grad_norm": 0.05279884859919548, "learning_rate": 0.00018429382485137403, "loss": 0.4516768932342529, "mean_token_accuracy": 0.88583979383111, "num_tokens": 54084982.0, "step": 3870 }, { "entropy": 0.44929216261953114, "epoch": 1.8327822390174777, "grad_norm": 0.07792633771896362, "learning_rate": 0.00018421370277609412, "loss": 0.4260584831237793, "mean_token_accuracy": 0.891007500141859, "num_tokens": 54221997.0, "step": 3880 }, { "entropy": 0.44946251530200243, "epoch": 1.8375059045819557, "grad_norm": 0.05967279151082039, "learning_rate": 0.0001841333943657443, "loss": 0.4332094669342041, "mean_token_accuracy": 0.890493854880333, "num_tokens": 54350421.0, "step": 3890 }, { "entropy": 0.46414444241672753, "epoch": 1.8422295701464337, "grad_norm": 0.06685349345207214, "learning_rate": 0.00018405289979801857, "loss": 0.44049849510192873, "mean_token_accuracy": 0.8869092471897602, "num_tokens": 54482540.0, "step": 3900 }, { "entropy": 0.4221767971292138, "epoch": 1.8469532357109117, "grad_norm": 0.06851854920387268, "learning_rate": 0.00018397221925102282, "loss": 0.3982303380966187, "mean_token_accuracy": 0.8984388276934624, "num_tokens": 54623646.0, "step": 3910 }, { "entropy": 0.4745147889479995, "epoch": 1.8516769012753898, "grad_norm": 0.05917099863290787, "learning_rate": 0.00018389135290327455, "loss": 0.45580105781555175, "mean_token_accuracy": 0.8864144250750542, "num_tokens": 54766252.0, "step": 3920 }, { "entropy": 0.45762479957193136, "epoch": 1.8564005668398678, "grad_norm": 0.07429071515798569, "learning_rate": 0.00018381030093370235, "loss": 0.4452804088592529, "mean_token_accuracy": 0.8842066012322902, "num_tokens": 54902400.0, "step": 3930 }, { "entropy": 0.443438520655036, "epoch": 1.8611242324043458, "grad_norm": 0.07402551919221878, "learning_rate": 0.0001837290635216454, "loss": 0.42833647727966306, "mean_token_accuracy": 0.8897256970405578, "num_tokens": 55038647.0, "step": 3940 }, { "entropy": 0.46854693777859213, "epoch": 1.8658478979688238, "grad_norm": 0.05710764229297638, "learning_rate": 0.00018364764084685342, "loss": 0.4533882141113281, "mean_token_accuracy": 0.8868382841348648, "num_tokens": 55184375.0, "step": 3950 }, { "entropy": 0.4432526580989361, "epoch": 1.8705715635333018, "grad_norm": 0.08100740611553192, "learning_rate": 0.00018356603308948582, "loss": 0.4315582275390625, "mean_token_accuracy": 0.888525103777647, "num_tokens": 55319442.0, "step": 3960 }, { "entropy": 0.4862626751884818, "epoch": 1.8752952290977798, "grad_norm": 0.06960438936948776, "learning_rate": 0.00018348424043011167, "loss": 0.46373467445373534, "mean_token_accuracy": 0.884248336404562, "num_tokens": 55457273.0, "step": 3970 }, { "entropy": 0.43588207568973303, "epoch": 1.8800188946622578, "grad_norm": 0.06006895750761032, "learning_rate": 0.00018340226304970915, "loss": 0.4303433418273926, "mean_token_accuracy": 0.8929056175053119, "num_tokens": 55597070.0, "step": 3980 }, { "entropy": 0.44345300886780026, "epoch": 1.8847425602267358, "grad_norm": 0.06901763379573822, "learning_rate": 0.00018332010112966513, "loss": 0.41420421600341795, "mean_token_accuracy": 0.8946427717804909, "num_tokens": 55738796.0, "step": 3990 }, { "entropy": 0.47705135978758334, "epoch": 1.8894662257912138, "grad_norm": 0.07143493741750717, "learning_rate": 0.00018323775485177478, "loss": 0.46305241584777834, "mean_token_accuracy": 0.8818318158388138, "num_tokens": 55880056.0, "step": 4000 }, { "entropy": 0.433615381270647, "epoch": 1.894189891355692, "grad_norm": 0.06972270458936691, "learning_rate": 0.00018315522439824128, "loss": 0.40766301155090334, "mean_token_accuracy": 0.8946473956108093, "num_tokens": 56020972.0, "step": 4010 }, { "entropy": 0.42825274504721167, "epoch": 1.89891355692017, "grad_norm": 0.06997814774513245, "learning_rate": 0.00018307250995167523, "loss": 0.4165190696716309, "mean_token_accuracy": 0.8947152942419052, "num_tokens": 56162117.0, "step": 4020 }, { "entropy": 0.4143740843050182, "epoch": 1.903637222484648, "grad_norm": 0.07156620919704437, "learning_rate": 0.00018298961169509436, "loss": 0.39263908863067626, "mean_token_accuracy": 0.8984962649643421, "num_tokens": 56300379.0, "step": 4030 }, { "entropy": 0.43085772367194297, "epoch": 1.908360888049126, "grad_norm": 0.07594224065542221, "learning_rate": 0.00018290652981192316, "loss": 0.4042006015777588, "mean_token_accuracy": 0.8954192057251931, "num_tokens": 56440683.0, "step": 4040 }, { "entropy": 0.45114052202552557, "epoch": 1.9130845536136043, "grad_norm": 0.05773433670401573, "learning_rate": 0.00018282326448599236, "loss": 0.4327089309692383, "mean_token_accuracy": 0.8901154600083828, "num_tokens": 56566577.0, "step": 4050 }, { "entropy": 0.46935334652662275, "epoch": 1.9178082191780823, "grad_norm": 0.06875091791152954, "learning_rate": 0.00018273981590153862, "loss": 0.4578728675842285, "mean_token_accuracy": 0.8841269321739673, "num_tokens": 56714174.0, "step": 4060 }, { "entropy": 0.43663804326206446, "epoch": 1.9225318847425603, "grad_norm": 0.06521524488925934, "learning_rate": 0.00018265618424320406, "loss": 0.41753034591674804, "mean_token_accuracy": 0.8942579530179501, "num_tokens": 56845313.0, "step": 4070 }, { "entropy": 0.42922870479524133, "epoch": 1.9272555503070383, "grad_norm": 0.06244893744587898, "learning_rate": 0.00018257236969603592, "loss": 0.3928467035293579, "mean_token_accuracy": 0.8967472240328789, "num_tokens": 56980079.0, "step": 4080 }, { "entropy": 0.472935585398227, "epoch": 1.9319792158715163, "grad_norm": 0.06986632198095322, "learning_rate": 0.00018248837244548606, "loss": 0.44841680526733396, "mean_token_accuracy": 0.8853162541985512, "num_tokens": 57115505.0, "step": 4090 }, { "entropy": 0.4164626685902476, "epoch": 1.9367028814359943, "grad_norm": 0.07339894771575928, "learning_rate": 0.00018240419267741062, "loss": 0.40720267295837403, "mean_token_accuracy": 0.8971424035727977, "num_tokens": 57249678.0, "step": 4100 }, { "entropy": 0.4143263744190335, "epoch": 1.9414265470004723, "grad_norm": 0.08470109850168228, "learning_rate": 0.0001823198305780696, "loss": 0.3952558755874634, "mean_token_accuracy": 0.8978372812271118, "num_tokens": 57386637.0, "step": 4110 }, { "entropy": 0.464755230396986, "epoch": 1.9461502125649504, "grad_norm": 0.09255680441856384, "learning_rate": 0.00018223528633412638, "loss": 0.4513087749481201, "mean_token_accuracy": 0.8865956492722035, "num_tokens": 57524084.0, "step": 4120 }, { "entropy": 0.44751606732606886, "epoch": 1.9508738781294284, "grad_norm": 0.06314191967248917, "learning_rate": 0.00018215056013264747, "loss": 0.4226852893829346, "mean_token_accuracy": 0.8908874273300171, "num_tokens": 57665039.0, "step": 4130 }, { "entropy": 0.4326350511983037, "epoch": 1.9555975436939064, "grad_norm": 0.07483214884996414, "learning_rate": 0.00018206565216110184, "loss": 0.4215069770812988, "mean_token_accuracy": 0.8924913898110389, "num_tokens": 57803864.0, "step": 4140 }, { "entropy": 0.47436560206115247, "epoch": 1.9603212092583844, "grad_norm": 0.07787442952394485, "learning_rate": 0.00018198056260736075, "loss": 0.4476649284362793, "mean_token_accuracy": 0.8846142441034317, "num_tokens": 57942973.0, "step": 4150 }, { "entropy": 0.49699554201215507, "epoch": 1.9650448748228624, "grad_norm": 0.08379582315683365, "learning_rate": 0.00018189529165969724, "loss": 0.481109619140625, "mean_token_accuracy": 0.8770826883614063, "num_tokens": 58082147.0, "step": 4160 }, { "entropy": 0.44130736850202085, "epoch": 1.9697685403873406, "grad_norm": 0.06495905667543411, "learning_rate": 0.0001818098395067857, "loss": 0.4062828063964844, "mean_token_accuracy": 0.8939802438020706, "num_tokens": 58219628.0, "step": 4170 }, { "entropy": 0.44190878346562384, "epoch": 1.9744922059518186, "grad_norm": 0.07276763021945953, "learning_rate": 0.00018172420633770142, "loss": 0.4200998306274414, "mean_token_accuracy": 0.8907590121030807, "num_tokens": 58357238.0, "step": 4180 }, { "entropy": 0.40038548652082684, "epoch": 1.9792158715162966, "grad_norm": 0.06182929873466492, "learning_rate": 0.00018163839234192026, "loss": 0.37431883811950684, "mean_token_accuracy": 0.9021176740527153, "num_tokens": 58493334.0, "step": 4190 }, { "entropy": 0.4119089724496007, "epoch": 1.9839395370807746, "grad_norm": 0.059533506631851196, "learning_rate": 0.00018155239770931817, "loss": 0.3976623058319092, "mean_token_accuracy": 0.900549802929163, "num_tokens": 58631751.0, "step": 4200 }, { "entropy": 0.39649777598679065, "epoch": 1.9886632026452529, "grad_norm": 0.0839238315820694, "learning_rate": 0.00018146622263017078, "loss": 0.3782901048660278, "mean_token_accuracy": 0.9035607136785984, "num_tokens": 58767668.0, "step": 4210 }, { "entropy": 0.45503481905907395, "epoch": 1.9933868682097309, "grad_norm": 0.07671350240707397, "learning_rate": 0.000181379867295153, "loss": 0.4480896949768066, "mean_token_accuracy": 0.8880835019052029, "num_tokens": 58907627.0, "step": 4220 }, { "entropy": 0.4405150419101119, "epoch": 1.9981105337742089, "grad_norm": 0.06963629275560379, "learning_rate": 0.00018129333189533856, "loss": 0.43314685821533205, "mean_token_accuracy": 0.8928300783038139, "num_tokens": 59036976.0, "step": 4230 }, { "entropy": 0.3951432419940829, "epoch": 2.002834199338687, "grad_norm": 0.07430269569158554, "learning_rate": 0.00018120661662219964, "loss": 0.36306726932525635, "mean_token_accuracy": 0.9068064488470554, "num_tokens": 59170221.0, "step": 4240 }, { "entropy": 0.41677092798054216, "epoch": 2.007557864903165, "grad_norm": 0.07779572904109955, "learning_rate": 0.00018111972166760638, "loss": 0.3878666400909424, "mean_token_accuracy": 0.9012737199664116, "num_tokens": 59309539.0, "step": 4250 }, { "entropy": 0.3765132769942284, "epoch": 2.012281530467643, "grad_norm": 0.06441673636436462, "learning_rate": 0.0001810326472238265, "loss": 0.3350167989730835, "mean_token_accuracy": 0.9144553929567337, "num_tokens": 59452502.0, "step": 4260 }, { "entropy": 0.3678074230439961, "epoch": 2.017005196032121, "grad_norm": 0.06515892595052719, "learning_rate": 0.00018094539348352493, "loss": 0.33828184604644773, "mean_token_accuracy": 0.9135874055325985, "num_tokens": 59592150.0, "step": 4270 }, { "entropy": 0.41742901131510735, "epoch": 2.021728861596599, "grad_norm": 0.06118781492114067, "learning_rate": 0.00018085796063976322, "loss": 0.3819070100784302, "mean_token_accuracy": 0.9015013135969638, "num_tokens": 59733287.0, "step": 4280 }, { "entropy": 0.41190977245569227, "epoch": 2.026452527161077, "grad_norm": 0.07450436800718307, "learning_rate": 0.00018077034888599927, "loss": 0.40396270751953123, "mean_token_accuracy": 0.9035611592233181, "num_tokens": 59874970.0, "step": 4290 }, { "entropy": 0.36769567579030993, "epoch": 2.031176192725555, "grad_norm": 0.06857893615961075, "learning_rate": 0.00018068255841608682, "loss": 0.3381620407104492, "mean_token_accuracy": 0.911721583455801, "num_tokens": 60008166.0, "step": 4300 }, { "entropy": 0.39516677148640156, "epoch": 2.035899858290033, "grad_norm": 0.06760476529598236, "learning_rate": 0.00018059458942427505, "loss": 0.36032567024230955, "mean_token_accuracy": 0.9051235631108284, "num_tokens": 60147507.0, "step": 4310 }, { "entropy": 0.3761145258322358, "epoch": 2.040623523854511, "grad_norm": 0.06463611125946045, "learning_rate": 0.0001805064421052082, "loss": 0.3462586164474487, "mean_token_accuracy": 0.9122118346393109, "num_tokens": 60280780.0, "step": 4320 }, { "entropy": 0.3850742572918534, "epoch": 2.045347189418989, "grad_norm": 0.07972639799118042, "learning_rate": 0.000180418116653925, "loss": 0.3521642446517944, "mean_token_accuracy": 0.9066629782319069, "num_tokens": 60423923.0, "step": 4330 }, { "entropy": 0.37431473657488823, "epoch": 2.050070854983467, "grad_norm": 0.08885598182678223, "learning_rate": 0.00018032961326585833, "loss": 0.34549927711486816, "mean_token_accuracy": 0.9116423606872559, "num_tokens": 60566373.0, "step": 4340 }, { "entropy": 0.369992808252573, "epoch": 2.0547945205479454, "grad_norm": 0.07883164286613464, "learning_rate": 0.0001802409321368349, "loss": 0.33596875667572024, "mean_token_accuracy": 0.9116500176489353, "num_tokens": 60702266.0, "step": 4350 }, { "entropy": 0.3654436276294291, "epoch": 2.0595181861124234, "grad_norm": 0.06069241836667061, "learning_rate": 0.00018015207346307454, "loss": 0.33056566715240476, "mean_token_accuracy": 0.9144408896565437, "num_tokens": 60843867.0, "step": 4360 }, { "entropy": 0.40022768583148716, "epoch": 2.0642418516769014, "grad_norm": 0.07794971764087677, "learning_rate": 0.00018006303744119001, "loss": 0.38457231521606444, "mean_token_accuracy": 0.9037668913602829, "num_tokens": 60985829.0, "step": 4370 }, { "entropy": 0.38590763676911594, "epoch": 2.0689655172413794, "grad_norm": 0.07632618397474289, "learning_rate": 0.0001799738242681865, "loss": 0.3478914976119995, "mean_token_accuracy": 0.9088833943009377, "num_tokens": 61121224.0, "step": 4380 }, { "entropy": 0.3829372115433216, "epoch": 2.0736891828058575, "grad_norm": 0.08546989411115646, "learning_rate": 0.0001798844341414611, "loss": 0.3539114952087402, "mean_token_accuracy": 0.9087465710937976, "num_tokens": 61260945.0, "step": 4390 }, { "entropy": 0.3747755914926529, "epoch": 2.0784128483703355, "grad_norm": 0.07021734863519669, "learning_rate": 0.00017979486725880247, "loss": 0.3456892490386963, "mean_token_accuracy": 0.909780490398407, "num_tokens": 61396568.0, "step": 4400 }, { "entropy": 0.36171059496700764, "epoch": 2.0831365139348135, "grad_norm": 0.058188583701848984, "learning_rate": 0.00017970512381839038, "loss": 0.3306448936462402, "mean_token_accuracy": 0.9147086225450038, "num_tokens": 61536701.0, "step": 4410 }, { "entropy": 0.3334600352682173, "epoch": 2.0878601794992915, "grad_norm": 0.07211539894342422, "learning_rate": 0.00017961520401879523, "loss": 0.3133976697921753, "mean_token_accuracy": 0.9215232580900192, "num_tokens": 61678793.0, "step": 4420 }, { "entropy": 0.39225977901369335, "epoch": 2.0925838450637695, "grad_norm": 0.08037585020065308, "learning_rate": 0.00017952510805897769, "loss": 0.34722707271575926, "mean_token_accuracy": 0.9084143519401551, "num_tokens": 61825552.0, "step": 4430 }, { "entropy": 0.372490400634706, "epoch": 2.0973075106282475, "grad_norm": 0.07935888320207596, "learning_rate": 0.00017943483613828815, "loss": 0.3376767635345459, "mean_token_accuracy": 0.9104174576699734, "num_tokens": 61962290.0, "step": 4440 }, { "entropy": 0.3751124802976847, "epoch": 2.1020311761927255, "grad_norm": 0.08084080368280411, "learning_rate": 0.00017934438845646638, "loss": 0.34236645698547363, "mean_token_accuracy": 0.9100891225039959, "num_tokens": 62106492.0, "step": 4450 }, { "entropy": 0.34409331558272244, "epoch": 2.1067548417572035, "grad_norm": 0.05427849665284157, "learning_rate": 0.00017925376521364105, "loss": 0.30626444816589354, "mean_token_accuracy": 0.918618057668209, "num_tokens": 62249470.0, "step": 4460 }, { "entropy": 0.3816100746393204, "epoch": 2.1114785073216815, "grad_norm": 0.07898930460214615, "learning_rate": 0.00017916296661032926, "loss": 0.35970211029052734, "mean_token_accuracy": 0.9055430836975574, "num_tokens": 62389863.0, "step": 4470 }, { "entropy": 0.3693529287353158, "epoch": 2.1162021728861595, "grad_norm": 0.08355957269668579, "learning_rate": 0.00017907199284743612, "loss": 0.3422150850296021, "mean_token_accuracy": 0.9115280970931053, "num_tokens": 62523161.0, "step": 4480 }, { "entropy": 0.3890496199950576, "epoch": 2.1209258384506375, "grad_norm": 0.05940314382314682, "learning_rate": 0.00017898084412625433, "loss": 0.3440500974655151, "mean_token_accuracy": 0.9117121279239655, "num_tokens": 62676856.0, "step": 4490 }, { "entropy": 0.38828163743019106, "epoch": 2.1256495040151155, "grad_norm": 0.07935027778148651, "learning_rate": 0.00017888952064846366, "loss": 0.3778949737548828, "mean_token_accuracy": 0.9049822241067886, "num_tokens": 62812870.0, "step": 4500 }, { "entropy": 0.3644891042262316, "epoch": 2.1303731695795936, "grad_norm": 0.07077588140964508, "learning_rate": 0.00017879802261613068, "loss": 0.32347466945648196, "mean_token_accuracy": 0.9153170421719551, "num_tokens": 62950453.0, "step": 4510 }, { "entropy": 0.3753189181908965, "epoch": 2.135096835144072, "grad_norm": 0.07934729754924774, "learning_rate": 0.00017870635023170802, "loss": 0.35052378177642823, "mean_token_accuracy": 0.9095086582005024, "num_tokens": 63086813.0, "step": 4520 }, { "entropy": 0.344531574100256, "epoch": 2.13982050070855, "grad_norm": 0.06132613494992256, "learning_rate": 0.0001786145036980342, "loss": 0.31072587966918946, "mean_token_accuracy": 0.9187924608588218, "num_tokens": 63222413.0, "step": 4530 }, { "entropy": 0.3367819916456938, "epoch": 2.144544166273028, "grad_norm": 0.07123655825853348, "learning_rate": 0.000178522483218333, "loss": 0.3007503509521484, "mean_token_accuracy": 0.9209801577031612, "num_tokens": 63360839.0, "step": 4540 }, { "entropy": 0.37355535849928856, "epoch": 2.149267831837506, "grad_norm": 0.07354898750782013, "learning_rate": 0.00017843028899621315, "loss": 0.3450171947479248, "mean_token_accuracy": 0.9104658022522927, "num_tokens": 63502321.0, "step": 4550 }, { "entropy": 0.3626279551535845, "epoch": 2.153991497401984, "grad_norm": 0.07379408180713654, "learning_rate": 0.00017833792123566778, "loss": 0.3310466051101685, "mean_token_accuracy": 0.9146706216037274, "num_tokens": 63646863.0, "step": 4560 }, { "entropy": 0.3701190158724785, "epoch": 2.158715162966462, "grad_norm": 0.06171635165810585, "learning_rate": 0.00017824538014107403, "loss": 0.3444772005081177, "mean_token_accuracy": 0.9130326814949512, "num_tokens": 63788236.0, "step": 4570 }, { "entropy": 0.3738078149035573, "epoch": 2.16343882853094, "grad_norm": 0.05464085564017296, "learning_rate": 0.00017815266591719245, "loss": 0.35484113693237307, "mean_token_accuracy": 0.9113606236875057, "num_tokens": 63929307.0, "step": 4580 }, { "entropy": 0.3798240454867482, "epoch": 2.168162494095418, "grad_norm": 0.08261693269014359, "learning_rate": 0.00017805977876916685, "loss": 0.34636573791503905, "mean_token_accuracy": 0.9096949256956577, "num_tokens": 64065655.0, "step": 4590 }, { "entropy": 0.3755222076550126, "epoch": 2.172886159659896, "grad_norm": 0.06726128607988358, "learning_rate": 0.0001779667189025235, "loss": 0.341137433052063, "mean_token_accuracy": 0.912428331375122, "num_tokens": 64202678.0, "step": 4600 }, { "entropy": 0.32989654783159494, "epoch": 2.177609825224374, "grad_norm": 0.0664137676358223, "learning_rate": 0.00017787348652317092, "loss": 0.29708251953125, "mean_token_accuracy": 0.922088798135519, "num_tokens": 64343569.0, "step": 4610 }, { "entropy": 0.3934741189703345, "epoch": 2.182333490788852, "grad_norm": 0.08352284133434296, "learning_rate": 0.00017778008183739932, "loss": 0.35492236614227296, "mean_token_accuracy": 0.9071983322501183, "num_tokens": 64487045.0, "step": 4620 }, { "entropy": 0.34776588194072244, "epoch": 2.18705715635333, "grad_norm": 0.08185728639364243, "learning_rate": 0.0001776865050518801, "loss": 0.3108149528503418, "mean_token_accuracy": 0.9175512231886387, "num_tokens": 64628896.0, "step": 4630 }, { "entropy": 0.3408706252463162, "epoch": 2.191780821917808, "grad_norm": 0.07090574502944946, "learning_rate": 0.00017759275637366554, "loss": 0.3044494390487671, "mean_token_accuracy": 0.9201794132590294, "num_tokens": 64770622.0, "step": 4640 }, { "entropy": 0.38265586011111735, "epoch": 2.196504487482286, "grad_norm": 0.0640915110707283, "learning_rate": 0.0001774988360101883, "loss": 0.3519306182861328, "mean_token_accuracy": 0.9097047552466393, "num_tokens": 64914683.0, "step": 4650 }, { "entropy": 0.3646923953667283, "epoch": 2.201228153046764, "grad_norm": 0.05122111737728119, "learning_rate": 0.00017740474416926073, "loss": 0.34943673610687254, "mean_token_accuracy": 0.912996233254671, "num_tokens": 65046171.0, "step": 4660 }, { "entropy": 0.3709303382784128, "epoch": 2.205951818611242, "grad_norm": 0.07036547362804413, "learning_rate": 0.0001773104810590748, "loss": 0.3317831516265869, "mean_token_accuracy": 0.9141067914664746, "num_tokens": 65187824.0, "step": 4670 }, { "entropy": 0.37796775698661805, "epoch": 2.21067548417572, "grad_norm": 0.06316936761140823, "learning_rate": 0.00017721604688820133, "loss": 0.34329471588134763, "mean_token_accuracy": 0.9125158511102199, "num_tokens": 65326878.0, "step": 4680 }, { "entropy": 0.36794558390974996, "epoch": 2.2153991497401986, "grad_norm": 0.07478772103786469, "learning_rate": 0.00017712144186558972, "loss": 0.34176435470581057, "mean_token_accuracy": 0.9126577869057655, "num_tokens": 65472098.0, "step": 4690 }, { "entropy": 0.3847030168399215, "epoch": 2.2201228153046766, "grad_norm": 0.053054895251989365, "learning_rate": 0.00017702666620056726, "loss": 0.3531605005264282, "mean_token_accuracy": 0.9097660399973393, "num_tokens": 65607840.0, "step": 4700 }, { "entropy": 0.3529300346970558, "epoch": 2.2248464808691546, "grad_norm": 0.07372882217168808, "learning_rate": 0.00017693172010283892, "loss": 0.3232670783996582, "mean_token_accuracy": 0.9168261006474495, "num_tokens": 65740150.0, "step": 4710 }, { "entropy": 0.3606896920129657, "epoch": 2.2295701464336326, "grad_norm": 0.06723502278327942, "learning_rate": 0.00017683660378248678, "loss": 0.32462348937988283, "mean_token_accuracy": 0.9145806074142456, "num_tokens": 65877155.0, "step": 4720 }, { "entropy": 0.3608708480373025, "epoch": 2.2342938119981106, "grad_norm": 0.07947982847690582, "learning_rate": 0.0001767413174499695, "loss": 0.337389874458313, "mean_token_accuracy": 0.9128516048192978, "num_tokens": 66020895.0, "step": 4730 }, { "entropy": 0.3662800621241331, "epoch": 2.2390174775625886, "grad_norm": 0.07876905053853989, "learning_rate": 0.000176645861316122, "loss": 0.33322911262512206, "mean_token_accuracy": 0.9142575584352016, "num_tokens": 66167145.0, "step": 4740 }, { "entropy": 0.35156021658331155, "epoch": 2.2437411431270666, "grad_norm": 0.07309263199567795, "learning_rate": 0.00017655023559215484, "loss": 0.3214474439620972, "mean_token_accuracy": 0.9149869471788407, "num_tokens": 66306859.0, "step": 4750 }, { "entropy": 0.39834485165774824, "epoch": 2.2484648086915446, "grad_norm": 0.08221690356731415, "learning_rate": 0.00017645444048965388, "loss": 0.35496699810028076, "mean_token_accuracy": 0.9076141364872455, "num_tokens": 66445152.0, "step": 4760 }, { "entropy": 0.3521524195559323, "epoch": 2.2531884742560226, "grad_norm": 0.08107949793338776, "learning_rate": 0.00017635847622057965, "loss": 0.32071890830993655, "mean_token_accuracy": 0.9162487290799618, "num_tokens": 66578229.0, "step": 4770 }, { "entropy": 0.3437899772077799, "epoch": 2.2579121398205007, "grad_norm": 0.0685676783323288, "learning_rate": 0.00017626234299726715, "loss": 0.30377700328826907, "mean_token_accuracy": 0.9203708678483963, "num_tokens": 66719507.0, "step": 4780 }, { "entropy": 0.3620788909494877, "epoch": 2.2626358053849787, "grad_norm": 0.08134099096059799, "learning_rate": 0.00017616604103242502, "loss": 0.33079211711883544, "mean_token_accuracy": 0.9154773041605949, "num_tokens": 66861243.0, "step": 4790 }, { "entropy": 0.3565347112715244, "epoch": 2.2673594709494567, "grad_norm": 0.058874648064374924, "learning_rate": 0.00017606957053913544, "loss": 0.32880067825317383, "mean_token_accuracy": 0.9145106978714466, "num_tokens": 66999811.0, "step": 4800 }, { "entropy": 0.36329170744866135, "epoch": 2.2720831365139347, "grad_norm": 0.07846896350383759, "learning_rate": 0.00017597293173085343, "loss": 0.3312691688537598, "mean_token_accuracy": 0.9129531547427178, "num_tokens": 67131276.0, "step": 4810 }, { "entropy": 0.35137215442955494, "epoch": 2.2768068020784127, "grad_norm": 0.08317407965660095, "learning_rate": 0.00017587612482140635, "loss": 0.3243239164352417, "mean_token_accuracy": 0.9175820186734199, "num_tokens": 67266674.0, "step": 4820 }, { "entropy": 0.39138262122869494, "epoch": 2.2815304676428907, "grad_norm": 0.07166536897420883, "learning_rate": 0.00017577915002499358, "loss": 0.36102991104125975, "mean_token_accuracy": 0.9072135731577873, "num_tokens": 67410593.0, "step": 4830 }, { "entropy": 0.3518522471189499, "epoch": 2.286254133207369, "grad_norm": 0.07341722398996353, "learning_rate": 0.000175682007556186, "loss": 0.3146039009094238, "mean_token_accuracy": 0.9182766884565353, "num_tokens": 67552721.0, "step": 4840 }, { "entropy": 0.35175929674878714, "epoch": 2.2909777987718467, "grad_norm": 0.06283942610025406, "learning_rate": 0.00017558469762992549, "loss": 0.30020389556884763, "mean_token_accuracy": 0.9206199988722801, "num_tokens": 67688880.0, "step": 4850 }, { "entropy": 0.36785907223820685, "epoch": 2.295701464336325, "grad_norm": 0.07479214668273926, "learning_rate": 0.00017548722046152433, "loss": 0.3509315252304077, "mean_token_accuracy": 0.9113536290824413, "num_tokens": 67821502.0, "step": 4860 }, { "entropy": 0.36734033189713955, "epoch": 2.300425129900803, "grad_norm": 0.07622663676738739, "learning_rate": 0.000175389576266665, "loss": 0.33639395236968994, "mean_token_accuracy": 0.9118621356785297, "num_tokens": 67952323.0, "step": 4870 }, { "entropy": 0.38060117568820717, "epoch": 2.305148795465281, "grad_norm": 0.06672433763742447, "learning_rate": 0.00017529176526139944, "loss": 0.3438107013702393, "mean_token_accuracy": 0.9105553299188613, "num_tokens": 68089099.0, "step": 4880 }, { "entropy": 0.3727780839428306, "epoch": 2.309872461029759, "grad_norm": 0.07081436365842819, "learning_rate": 0.00017519378766214878, "loss": 0.3472867965698242, "mean_token_accuracy": 0.9104488752782345, "num_tokens": 68227401.0, "step": 4890 }, { "entropy": 0.3444125018082559, "epoch": 2.314596126594237, "grad_norm": 0.07869017869234085, "learning_rate": 0.0001750956436857027, "loss": 0.3122610092163086, "mean_token_accuracy": 0.9187357731163501, "num_tokens": 68372503.0, "step": 4900 }, { "entropy": 0.38529773708432913, "epoch": 2.319319792158715, "grad_norm": 0.06087201088666916, "learning_rate": 0.00017499733354921898, "loss": 0.35583522319793703, "mean_token_accuracy": 0.9096518360078335, "num_tokens": 68523575.0, "step": 4910 }, { "entropy": 0.329516320489347, "epoch": 2.324043457723193, "grad_norm": 0.06620240211486816, "learning_rate": 0.00017489885747022318, "loss": 0.2991222381591797, "mean_token_accuracy": 0.923500907421112, "num_tokens": 68662446.0, "step": 4920 }, { "entropy": 0.3849239324219525, "epoch": 2.328767123287671, "grad_norm": 0.08974209427833557, "learning_rate": 0.00017480021566660788, "loss": 0.3524339199066162, "mean_token_accuracy": 0.9093660332262516, "num_tokens": 68810406.0, "step": 4930 }, { "entropy": 0.3772364977747202, "epoch": 2.3334907888521492, "grad_norm": 0.07428587973117828, "learning_rate": 0.00017470140835663248, "loss": 0.35045742988586426, "mean_token_accuracy": 0.9099268049001694, "num_tokens": 68951698.0, "step": 4940 }, { "entropy": 0.3782703010365367, "epoch": 2.3382144544166272, "grad_norm": 0.08954386413097382, "learning_rate": 0.0001746024357589225, "loss": 0.3453653573989868, "mean_token_accuracy": 0.9120040215551853, "num_tokens": 69086778.0, "step": 4950 }, { "entropy": 0.3368293071165681, "epoch": 2.3429381199811052, "grad_norm": 0.06562545150518417, "learning_rate": 0.0001745032980924692, "loss": 0.30047316551208497, "mean_token_accuracy": 0.9200038239359856, "num_tokens": 69223879.0, "step": 4960 }, { "entropy": 0.33657576572149994, "epoch": 2.3476617855455832, "grad_norm": 0.0857587456703186, "learning_rate": 0.00017440399557662912, "loss": 0.2978757619857788, "mean_token_accuracy": 0.9229826048016548, "num_tokens": 69361696.0, "step": 4970 }, { "entropy": 0.35648726038634776, "epoch": 2.3523854511100613, "grad_norm": 0.06771888583898544, "learning_rate": 0.0001743045284311235, "loss": 0.3391841173171997, "mean_token_accuracy": 0.9125791676342487, "num_tokens": 69497700.0, "step": 4980 }, { "entropy": 0.377038012444973, "epoch": 2.3571091166745393, "grad_norm": 0.05512234941124916, "learning_rate": 0.0001742048968760379, "loss": 0.329133939743042, "mean_token_accuracy": 0.913535013794899, "num_tokens": 69638549.0, "step": 4990 }, { "entropy": 0.34691223232075574, "epoch": 2.3618327822390173, "grad_norm": 0.07337567210197449, "learning_rate": 0.0001741051011318216, "loss": 0.326473593711853, "mean_token_accuracy": 0.9155340820550919, "num_tokens": 69777845.0, "step": 5000 }, { "entropy": 0.3717569367960095, "epoch": 2.3665564478034957, "grad_norm": 0.06603619456291199, "learning_rate": 0.00017400514141928722, "loss": 0.32091450691223145, "mean_token_accuracy": 0.9150391787290573, "num_tokens": 69915165.0, "step": 5010 }, { "entropy": 0.3680802728049457, "epoch": 2.3712801133679737, "grad_norm": 0.09668844193220139, "learning_rate": 0.00017390501795961017, "loss": 0.34298737049102784, "mean_token_accuracy": 0.9123553864657878, "num_tokens": 70056805.0, "step": 5020 }, { "entropy": 0.35295952931046487, "epoch": 2.3760037789324517, "grad_norm": 0.06308913230895996, "learning_rate": 0.00017380473097432814, "loss": 0.3172333240509033, "mean_token_accuracy": 0.9175053678452969, "num_tokens": 70198087.0, "step": 5030 }, { "entropy": 0.36961612179875375, "epoch": 2.3807274444969297, "grad_norm": 0.0788240060210228, "learning_rate": 0.0001737042806853407, "loss": 0.3349175930023193, "mean_token_accuracy": 0.9132102608680726, "num_tokens": 70335404.0, "step": 5040 }, { "entropy": 0.3711566047742963, "epoch": 2.3854511100614078, "grad_norm": 0.07890364527702332, "learning_rate": 0.00017360366731490877, "loss": 0.3386693000793457, "mean_token_accuracy": 0.9134459957480431, "num_tokens": 70477392.0, "step": 5050 }, { "entropy": 0.35097449999302627, "epoch": 2.3901747756258858, "grad_norm": 0.06498616188764572, "learning_rate": 0.00017350289108565404, "loss": 0.3155566930770874, "mean_token_accuracy": 0.9186516053974628, "num_tokens": 70613392.0, "step": 5060 }, { "entropy": 0.3575643849559128, "epoch": 2.3948984411903638, "grad_norm": 0.07433422654867172, "learning_rate": 0.00017340195222055854, "loss": 0.3291230440139771, "mean_token_accuracy": 0.9150498561561108, "num_tokens": 70761404.0, "step": 5070 }, { "entropy": 0.3681374185718596, "epoch": 2.3996221067548418, "grad_norm": 0.0771808996796608, "learning_rate": 0.00017330085094296422, "loss": 0.3228792190551758, "mean_token_accuracy": 0.9155564263463021, "num_tokens": 70902964.0, "step": 5080 }, { "entropy": 0.3403599739074707, "epoch": 2.40434577231932, "grad_norm": 0.061653293669223785, "learning_rate": 0.0001731995874765724, "loss": 0.3083616256713867, "mean_token_accuracy": 0.918988062441349, "num_tokens": 71042549.0, "step": 5090 }, { "entropy": 0.36283190539106724, "epoch": 2.409069437883798, "grad_norm": 0.08139196038246155, "learning_rate": 0.00017309816204544317, "loss": 0.3336886644363403, "mean_token_accuracy": 0.9147178888320923, "num_tokens": 71180384.0, "step": 5100 }, { "entropy": 0.34851250536739825, "epoch": 2.413793103448276, "grad_norm": 0.06935746222734451, "learning_rate": 0.0001729965748739951, "loss": 0.317404580116272, "mean_token_accuracy": 0.9191467575728893, "num_tokens": 71318781.0, "step": 5110 }, { "entropy": 0.35370422853156924, "epoch": 2.418516769012754, "grad_norm": 0.0793750062584877, "learning_rate": 0.00017289482618700462, "loss": 0.30998620986938474, "mean_token_accuracy": 0.917963957041502, "num_tokens": 71451944.0, "step": 5120 }, { "entropy": 0.3418350827880204, "epoch": 2.423240434577232, "grad_norm": 0.07875291258096695, "learning_rate": 0.00017279291620960543, "loss": 0.3185269832611084, "mean_token_accuracy": 0.9166646711528301, "num_tokens": 71593532.0, "step": 5130 }, { "entropy": 0.36766334623098373, "epoch": 2.42796410014171, "grad_norm": 0.0857739970088005, "learning_rate": 0.0001726908451672882, "loss": 0.32974236011505126, "mean_token_accuracy": 0.9142980873584747, "num_tokens": 71730615.0, "step": 5140 }, { "entropy": 0.3898682286031544, "epoch": 2.432687765706188, "grad_norm": 0.08818437159061432, "learning_rate": 0.0001725886132859, "loss": 0.36099967956542967, "mean_token_accuracy": 0.9105031929910183, "num_tokens": 71871723.0, "step": 5150 }, { "entropy": 0.3895622887648642, "epoch": 2.4374114312706663, "grad_norm": 0.0812084749341011, "learning_rate": 0.0001724862207916437, "loss": 0.362603497505188, "mean_token_accuracy": 0.9069844730198383, "num_tokens": 72012227.0, "step": 5160 }, { "entropy": 0.3616925870999694, "epoch": 2.442135096835144, "grad_norm": 0.06379372626543045, "learning_rate": 0.0001723836679110776, "loss": 0.32681663036346437, "mean_token_accuracy": 0.9173739403486252, "num_tokens": 72155955.0, "step": 5170 }, { "entropy": 0.34351073727011683, "epoch": 2.4468587623996223, "grad_norm": 0.0815335288643837, "learning_rate": 0.00017228095487111487, "loss": 0.31263697147369385, "mean_token_accuracy": 0.9207109160721302, "num_tokens": 72296105.0, "step": 5180 }, { "entropy": 0.367705793119967, "epoch": 2.4515824279641003, "grad_norm": 0.07983652502298355, "learning_rate": 0.00017217808189902306, "loss": 0.34312000274658205, "mean_token_accuracy": 0.9133535690605641, "num_tokens": 72427639.0, "step": 5190 }, { "entropy": 0.3358869240619242, "epoch": 2.4563060935285783, "grad_norm": 0.07612093538045883, "learning_rate": 0.0001720750492224236, "loss": 0.28635563850402834, "mean_token_accuracy": 0.9230332836508751, "num_tokens": 72576973.0, "step": 5200 }, { "entropy": 0.3458953768014908, "epoch": 2.4610297590930563, "grad_norm": 0.07325061410665512, "learning_rate": 0.00017197185706929128, "loss": 0.30686020851135254, "mean_token_accuracy": 0.9192182451486588, "num_tokens": 72720494.0, "step": 5210 }, { "entropy": 0.3425209357403219, "epoch": 2.4657534246575343, "grad_norm": 0.06628164649009705, "learning_rate": 0.00017186850566795373, "loss": 0.3105634689331055, "mean_token_accuracy": 0.9185604088008403, "num_tokens": 72862211.0, "step": 5220 }, { "entropy": 0.3979892102070153, "epoch": 2.4704770902220123, "grad_norm": 0.07804671674966812, "learning_rate": 0.00017176499524709096, "loss": 0.3673126220703125, "mean_token_accuracy": 0.9062325954437256, "num_tokens": 73005958.0, "step": 5230 }, { "entropy": 0.3278886295855045, "epoch": 2.4752007557864903, "grad_norm": 0.07158853858709335, "learning_rate": 0.00017166132603573485, "loss": 0.28293752670288086, "mean_token_accuracy": 0.9247597858309746, "num_tokens": 73150446.0, "step": 5240 }, { "entropy": 0.33812597766518593, "epoch": 2.4799244213509684, "grad_norm": 0.06800093501806259, "learning_rate": 0.00017155749826326864, "loss": 0.30004396438598635, "mean_token_accuracy": 0.9229036450386048, "num_tokens": 73290485.0, "step": 5250 }, { "entropy": 0.32100099362432954, "epoch": 2.4846480869154464, "grad_norm": 0.06439868360757828, "learning_rate": 0.00017145351215942632, "loss": 0.2929414987564087, "mean_token_accuracy": 0.9239821515977382, "num_tokens": 73431354.0, "step": 5260 }, { "entropy": 0.35717023378238083, "epoch": 2.4893717524799244, "grad_norm": 0.06693439930677414, "learning_rate": 0.0001713493679542923, "loss": 0.32613224983215333, "mean_token_accuracy": 0.9171266071498394, "num_tokens": 73570275.0, "step": 5270 }, { "entropy": 0.3441743317991495, "epoch": 2.4940954180444024, "grad_norm": 0.0831870511174202, "learning_rate": 0.00017124506587830083, "loss": 0.31802730560302733, "mean_token_accuracy": 0.9191960535943509, "num_tokens": 73701904.0, "step": 5280 }, { "entropy": 0.3742942173033953, "epoch": 2.4988190836088804, "grad_norm": 0.08274650573730469, "learning_rate": 0.00017114060616223536, "loss": 0.3411454916000366, "mean_token_accuracy": 0.9124328427016735, "num_tokens": 73836277.0, "step": 5290 }, { "entropy": 0.34059149408712985, "epoch": 2.5035427491733584, "grad_norm": 0.0695531889796257, "learning_rate": 0.00017103598903722827, "loss": 0.3107386350631714, "mean_token_accuracy": 0.9196738213300705, "num_tokens": 73977700.0, "step": 5300 }, { "entropy": 0.3607602405361831, "epoch": 2.5082664147378364, "grad_norm": 0.08559349924325943, "learning_rate": 0.00017093121473476015, "loss": 0.3217759370803833, "mean_token_accuracy": 0.9155218534171581, "num_tokens": 74120486.0, "step": 5310 }, { "entropy": 0.3522417671047151, "epoch": 2.5129900803023144, "grad_norm": 0.06628310680389404, "learning_rate": 0.00017082628348665934, "loss": 0.3156892776489258, "mean_token_accuracy": 0.9205105111002923, "num_tokens": 74257237.0, "step": 5320 }, { "entropy": 0.37255304008722306, "epoch": 2.517713745866793, "grad_norm": 0.07685009390115738, "learning_rate": 0.00017072119552510157, "loss": 0.3570106029510498, "mean_token_accuracy": 0.9101258270442486, "num_tokens": 74394555.0, "step": 5330 }, { "entropy": 0.3417865173891187, "epoch": 2.5224374114312704, "grad_norm": 0.06182515248656273, "learning_rate": 0.0001706159510826092, "loss": 0.30006103515625, "mean_token_accuracy": 0.9223997719585896, "num_tokens": 74535607.0, "step": 5340 }, { "entropy": 0.3596333073452115, "epoch": 2.527161076995749, "grad_norm": 0.08518214523792267, "learning_rate": 0.00017051055039205086, "loss": 0.3307522773742676, "mean_token_accuracy": 0.9149652808904648, "num_tokens": 74674135.0, "step": 5350 }, { "entropy": 0.3565245810896158, "epoch": 2.531884742560227, "grad_norm": 0.07785555720329285, "learning_rate": 0.00017040499368664095, "loss": 0.3165716648101807, "mean_token_accuracy": 0.916690094769001, "num_tokens": 74807533.0, "step": 5360 }, { "entropy": 0.37183555755764247, "epoch": 2.536608408124705, "grad_norm": 0.0807003527879715, "learning_rate": 0.00017029928119993904, "loss": 0.34868154525756834, "mean_token_accuracy": 0.9119903385639191, "num_tokens": 74948566.0, "step": 5370 }, { "entropy": 0.3737936247140169, "epoch": 2.541332073689183, "grad_norm": 0.057071540504693985, "learning_rate": 0.00017019341316584934, "loss": 0.3513514995574951, "mean_token_accuracy": 0.9121352203190327, "num_tokens": 75087320.0, "step": 5380 }, { "entropy": 0.3535418752580881, "epoch": 2.546055739253661, "grad_norm": 0.07253091782331467, "learning_rate": 0.0001700873898186203, "loss": 0.3125487804412842, "mean_token_accuracy": 0.919706829637289, "num_tokens": 75225167.0, "step": 5390 }, { "entropy": 0.3378803776577115, "epoch": 2.550779404818139, "grad_norm": 0.08137330412864685, "learning_rate": 0.00016998121139284405, "loss": 0.30165958404541016, "mean_token_accuracy": 0.9213343054056168, "num_tokens": 75364306.0, "step": 5400 }, { "entropy": 0.3566731529310346, "epoch": 2.555503070382617, "grad_norm": 0.07202183455228806, "learning_rate": 0.00016987487812345574, "loss": 0.3214185476303101, "mean_token_accuracy": 0.9166874349117279, "num_tokens": 75504073.0, "step": 5410 }, { "entropy": 0.35006280736997725, "epoch": 2.560226735947095, "grad_norm": 0.07671020925045013, "learning_rate": 0.0001697683902457332, "loss": 0.32263405323028566, "mean_token_accuracy": 0.9176624983549118, "num_tokens": 75651688.0, "step": 5420 }, { "entropy": 0.3596495855599642, "epoch": 2.564950401511573, "grad_norm": 0.057611897587776184, "learning_rate": 0.0001696617479952964, "loss": 0.31757681369781493, "mean_token_accuracy": 0.9163362935185433, "num_tokens": 75792454.0, "step": 5430 }, { "entropy": 0.3428363500162959, "epoch": 2.569674067076051, "grad_norm": 0.06537020951509476, "learning_rate": 0.00016955495160810683, "loss": 0.3163311004638672, "mean_token_accuracy": 0.9194522321224212, "num_tokens": 75928302.0, "step": 5440 }, { "entropy": 0.33568228837102654, "epoch": 2.574397732640529, "grad_norm": 0.08116862922906876, "learning_rate": 0.000169448001320467, "loss": 0.2961144924163818, "mean_token_accuracy": 0.9231125563383102, "num_tokens": 76071503.0, "step": 5450 }, { "entropy": 0.334136747661978, "epoch": 2.579121398205007, "grad_norm": 0.0786835104227066, "learning_rate": 0.00016934089736902002, "loss": 0.3097526550292969, "mean_token_accuracy": 0.9204525671899318, "num_tokens": 76205030.0, "step": 5460 }, { "entropy": 0.3176261980086565, "epoch": 2.583845063769485, "grad_norm": 0.05822472646832466, "learning_rate": 0.0001692336399907489, "loss": 0.2668977975845337, "mean_token_accuracy": 0.9287205383181572, "num_tokens": 76346970.0, "step": 5470 }, { "entropy": 0.33487942814826965, "epoch": 2.5885687293339634, "grad_norm": 0.08300083875656128, "learning_rate": 0.00016912622942297627, "loss": 0.303263258934021, "mean_token_accuracy": 0.9212611213326454, "num_tokens": 76488488.0, "step": 5480 }, { "entropy": 0.33707099203020335, "epoch": 2.593292394898441, "grad_norm": 0.08029662817716599, "learning_rate": 0.00016901866590336362, "loss": 0.2941617965698242, "mean_token_accuracy": 0.9243578106164932, "num_tokens": 76628143.0, "step": 5490 }, { "entropy": 0.3697511538863182, "epoch": 2.5980160604629194, "grad_norm": 0.07310616225004196, "learning_rate": 0.00016891094966991083, "loss": 0.33811867237091064, "mean_token_accuracy": 0.9146531477570534, "num_tokens": 76770868.0, "step": 5500 }, { "entropy": 0.3343955673277378, "epoch": 2.602739726027397, "grad_norm": 0.06992948055267334, "learning_rate": 0.0001688030809609558, "loss": 0.2992079257965088, "mean_token_accuracy": 0.9212706468999385, "num_tokens": 76907759.0, "step": 5510 }, { "entropy": 0.34800674095749856, "epoch": 2.6074633915918755, "grad_norm": 0.05579889938235283, "learning_rate": 0.00016869506001517378, "loss": 0.3101692199707031, "mean_token_accuracy": 0.9179186180233956, "num_tokens": 77052000.0, "step": 5520 }, { "entropy": 0.3438716458156705, "epoch": 2.6121870571563535, "grad_norm": 0.0798083171248436, "learning_rate": 0.0001685868870715767, "loss": 0.313541579246521, "mean_token_accuracy": 0.9183024413883686, "num_tokens": 77191827.0, "step": 5530 }, { "entropy": 0.3275704473257065, "epoch": 2.6169107227208315, "grad_norm": 0.06428426504135132, "learning_rate": 0.00016847856236951305, "loss": 0.2806210517883301, "mean_token_accuracy": 0.9272417724132538, "num_tokens": 77328333.0, "step": 5540 }, { "entropy": 0.31136206425726415, "epoch": 2.6216343882853095, "grad_norm": 0.0643233060836792, "learning_rate": 0.0001683700861486669, "loss": 0.2782828569412231, "mean_token_accuracy": 0.9277472048997879, "num_tokens": 77462896.0, "step": 5550 }, { "entropy": 0.35034023597836494, "epoch": 2.6263580538497875, "grad_norm": 0.07152332365512848, "learning_rate": 0.00016826145864905781, "loss": 0.31547298431396487, "mean_token_accuracy": 0.9181152805685997, "num_tokens": 77599830.0, "step": 5560 }, { "entropy": 0.36181382462382317, "epoch": 2.6310817194142655, "grad_norm": 0.08249314874410629, "learning_rate": 0.0001681526801110398, "loss": 0.33376924991607665, "mean_token_accuracy": 0.913762079179287, "num_tokens": 77747327.0, "step": 5570 }, { "entropy": 0.3237987017259002, "epoch": 2.6358053849787435, "grad_norm": 0.06599178910255432, "learning_rate": 0.00016804375077530127, "loss": 0.2831145763397217, "mean_token_accuracy": 0.9258603036403656, "num_tokens": 77891525.0, "step": 5580 }, { "entropy": 0.34556327275931836, "epoch": 2.6405290505432215, "grad_norm": 0.0647219717502594, "learning_rate": 0.0001679346708828642, "loss": 0.3150164604187012, "mean_token_accuracy": 0.9196638904511929, "num_tokens": 78023048.0, "step": 5590 }, { "entropy": 0.32787940595299003, "epoch": 2.6452527161076995, "grad_norm": 0.07890662550926208, "learning_rate": 0.00016782544067508372, "loss": 0.2941134452819824, "mean_token_accuracy": 0.9236494019627571, "num_tokens": 78157202.0, "step": 5600 }, { "entropy": 0.32976763527840375, "epoch": 2.6499763816721775, "grad_norm": 0.06924191862344742, "learning_rate": 0.00016771606039364757, "loss": 0.3007648468017578, "mean_token_accuracy": 0.9245947524905205, "num_tokens": 78303678.0, "step": 5610 }, { "entropy": 0.3603557227179408, "epoch": 2.6547000472366555, "grad_norm": 0.08832179009914398, "learning_rate": 0.00016760653028057555, "loss": 0.3240990161895752, "mean_token_accuracy": 0.9171132750809192, "num_tokens": 78444983.0, "step": 5620 }, { "entropy": 0.32523603141307833, "epoch": 2.6594237128011335, "grad_norm": 0.07937014847993851, "learning_rate": 0.00016749685057821897, "loss": 0.31206109523773196, "mean_token_accuracy": 0.923108734190464, "num_tokens": 78580540.0, "step": 5630 }, { "entropy": 0.37154110576957466, "epoch": 2.6641473783656116, "grad_norm": 0.07532130181789398, "learning_rate": 0.00016738702152926008, "loss": 0.3190698385238647, "mean_token_accuracy": 0.9166950397193432, "num_tokens": 78720040.0, "step": 5640 }, { "entropy": 0.3364192970097065, "epoch": 2.66887104393009, "grad_norm": 0.05883653834462166, "learning_rate": 0.0001672770433767117, "loss": 0.3031845808029175, "mean_token_accuracy": 0.9220205329358577, "num_tokens": 78862789.0, "step": 5650 }, { "entropy": 0.3114610935561359, "epoch": 2.6735947094945676, "grad_norm": 0.0733380988240242, "learning_rate": 0.00016716691636391645, "loss": 0.2725443601608276, "mean_token_accuracy": 0.9293257258832455, "num_tokens": 78998375.0, "step": 5660 }, { "entropy": 0.31692771119996904, "epoch": 2.678318375059046, "grad_norm": 0.08704158663749695, "learning_rate": 0.00016705664073454637, "loss": 0.2870356321334839, "mean_token_accuracy": 0.9263888739049435, "num_tokens": 79145229.0, "step": 5670 }, { "entropy": 0.35461486913263796, "epoch": 2.6830420406235236, "grad_norm": 0.06502511352300644, "learning_rate": 0.0001669462167326023, "loss": 0.3148838996887207, "mean_token_accuracy": 0.9205213010311126, "num_tokens": 79284364.0, "step": 5680 }, { "entropy": 0.3435555054806173, "epoch": 2.687765706188002, "grad_norm": 0.06909587979316711, "learning_rate": 0.00016683564460241352, "loss": 0.3052964210510254, "mean_token_accuracy": 0.9217824183404446, "num_tokens": 79426301.0, "step": 5690 }, { "entropy": 0.328545256704092, "epoch": 2.69248937175248, "grad_norm": 0.08267869055271149, "learning_rate": 0.00016672492458863687, "loss": 0.29160315990448, "mean_token_accuracy": 0.9249641202390194, "num_tokens": 79572106.0, "step": 5700 }, { "entropy": 0.3531115956604481, "epoch": 2.697213037316958, "grad_norm": 0.06687217205762863, "learning_rate": 0.0001666140569362565, "loss": 0.3208428621292114, "mean_token_accuracy": 0.9184792272746563, "num_tokens": 79708770.0, "step": 5710 }, { "entropy": 0.3515950422734022, "epoch": 2.701936702881436, "grad_norm": 0.07523484528064728, "learning_rate": 0.00016650304189058326, "loss": 0.3100206613540649, "mean_token_accuracy": 0.9201281875371933, "num_tokens": 79848258.0, "step": 5720 }, { "entropy": 0.3321186739951372, "epoch": 2.706660368445914, "grad_norm": 0.08257053792476654, "learning_rate": 0.00016639187969725403, "loss": 0.29955267906188965, "mean_token_accuracy": 0.9246600136160851, "num_tokens": 79985549.0, "step": 5730 }, { "entropy": 0.3408555801026523, "epoch": 2.711384034010392, "grad_norm": 0.06815972179174423, "learning_rate": 0.00016628057060223136, "loss": 0.3028057098388672, "mean_token_accuracy": 0.9233085177838802, "num_tokens": 80126601.0, "step": 5740 }, { "entropy": 0.33446909915655854, "epoch": 2.71610769957487, "grad_norm": 0.06723276525735855, "learning_rate": 0.00016616911485180285, "loss": 0.3122803449630737, "mean_token_accuracy": 0.9200693279504776, "num_tokens": 80262647.0, "step": 5750 }, { "entropy": 0.3261042980477214, "epoch": 2.720831365139348, "grad_norm": 0.07140873372554779, "learning_rate": 0.00016605751269258053, "loss": 0.2946842432022095, "mean_token_accuracy": 0.9244419567286968, "num_tokens": 80398721.0, "step": 5760 }, { "entropy": 0.33038164898753164, "epoch": 2.725555030703826, "grad_norm": 0.07432924956083298, "learning_rate": 0.0001659457643715004, "loss": 0.2941530466079712, "mean_token_accuracy": 0.9233614422380925, "num_tokens": 80536181.0, "step": 5770 }, { "entropy": 0.32198099195957186, "epoch": 2.730278696268304, "grad_norm": 0.08363515883684158, "learning_rate": 0.00016583387013582193, "loss": 0.2979699373245239, "mean_token_accuracy": 0.9252390690147877, "num_tokens": 80675463.0, "step": 5780 }, { "entropy": 0.3356750601902604, "epoch": 2.735002361832782, "grad_norm": 0.07102660834789276, "learning_rate": 0.00016572183023312736, "loss": 0.28064689636230467, "mean_token_accuracy": 0.9246694676578044, "num_tokens": 80813075.0, "step": 5790 }, { "entropy": 0.32597459070384505, "epoch": 2.73972602739726, "grad_norm": 0.062226396054029465, "learning_rate": 0.00016560964491132128, "loss": 0.28543217182159425, "mean_token_accuracy": 0.9277141757309437, "num_tokens": 80957704.0, "step": 5800 }, { "entropy": 0.3176982102915645, "epoch": 2.744449692961738, "grad_norm": 0.07835379987955093, "learning_rate": 0.00016549731441863004, "loss": 0.2929314136505127, "mean_token_accuracy": 0.9269766561686993, "num_tokens": 81097003.0, "step": 5810 }, { "entropy": 0.32748577343299984, "epoch": 2.7491733585262166, "grad_norm": 0.08433811366558075, "learning_rate": 0.0001653848390036012, "loss": 0.29446067810058596, "mean_token_accuracy": 0.9267704904079437, "num_tokens": 81229688.0, "step": 5820 }, { "entropy": 0.3164313483983278, "epoch": 2.753897024090694, "grad_norm": 0.10078940540552139, "learning_rate": 0.00016527221891510295, "loss": 0.27254998683929443, "mean_token_accuracy": 0.9284701183438301, "num_tokens": 81374123.0, "step": 5830 }, { "entropy": 0.33126611560583114, "epoch": 2.7586206896551726, "grad_norm": 0.05966135486960411, "learning_rate": 0.0001651594544023237, "loss": 0.29353697299957277, "mean_token_accuracy": 0.9236054413020611, "num_tokens": 81506062.0, "step": 5840 }, { "entropy": 0.32021211497485635, "epoch": 2.7633443552196506, "grad_norm": 0.07689578831195831, "learning_rate": 0.00016504654571477124, "loss": 0.28202998638153076, "mean_token_accuracy": 0.9264591842889786, "num_tokens": 81640701.0, "step": 5850 }, { "entropy": 0.31721112634986637, "epoch": 2.7680680207841286, "grad_norm": 0.08030835539102554, "learning_rate": 0.00016493349310227258, "loss": 0.2831667423248291, "mean_token_accuracy": 0.9279636651277542, "num_tokens": 81780366.0, "step": 5860 }, { "entropy": 0.3528039675205946, "epoch": 2.7727916863486066, "grad_norm": 0.07704772055149078, "learning_rate": 0.00016482029681497297, "loss": 0.3269792079925537, "mean_token_accuracy": 0.9188001587986946, "num_tokens": 81921173.0, "step": 5870 }, { "entropy": 0.3146613361313939, "epoch": 2.7775153519130846, "grad_norm": 0.07407624274492264, "learning_rate": 0.00016470695710333572, "loss": 0.27025275230407714, "mean_token_accuracy": 0.9304538980126381, "num_tokens": 82072181.0, "step": 5880 }, { "entropy": 0.3299731435254216, "epoch": 2.7822390174775626, "grad_norm": 0.05570664256811142, "learning_rate": 0.00016459347421814144, "loss": 0.2936347007751465, "mean_token_accuracy": 0.9222227200865746, "num_tokens": 82212340.0, "step": 5890 }, { "entropy": 0.36671034097671507, "epoch": 2.7869626830420406, "grad_norm": 0.07951661944389343, "learning_rate": 0.00016447984841048752, "loss": 0.31798481941223145, "mean_token_accuracy": 0.9174000978469848, "num_tokens": 82354350.0, "step": 5900 }, { "entropy": 0.33117240741848947, "epoch": 2.7916863486065187, "grad_norm": 0.07602746039628983, "learning_rate": 0.00016436607993178753, "loss": 0.3032093524932861, "mean_token_accuracy": 0.9241811975836753, "num_tokens": 82488366.0, "step": 5910 }, { "entropy": 0.326247438788414, "epoch": 2.7964100141709967, "grad_norm": 0.07182633131742477, "learning_rate": 0.00016425216903377083, "loss": 0.2878411769866943, "mean_token_accuracy": 0.9265529617667199, "num_tokens": 82630446.0, "step": 5920 }, { "entropy": 0.33973691463470457, "epoch": 2.8011336797354747, "grad_norm": 0.08628895878791809, "learning_rate": 0.00016413811596848187, "loss": 0.293949031829834, "mean_token_accuracy": 0.9222980067133904, "num_tokens": 82771428.0, "step": 5930 }, { "entropy": 0.3236493097618222, "epoch": 2.8058573452999527, "grad_norm": 0.0650419294834137, "learning_rate": 0.00016402392098827963, "loss": 0.2848384380340576, "mean_token_accuracy": 0.9256645426154136, "num_tokens": 82912863.0, "step": 5940 }, { "entropy": 0.31968826381489635, "epoch": 2.8105810108644307, "grad_norm": 0.08068515360355377, "learning_rate": 0.00016390958434583713, "loss": 0.28550896644592283, "mean_token_accuracy": 0.9248417422175408, "num_tokens": 83054406.0, "step": 5950 }, { "entropy": 0.3312150011770427, "epoch": 2.8153046764289087, "grad_norm": 0.05488315224647522, "learning_rate": 0.00016379510629414083, "loss": 0.2847553491592407, "mean_token_accuracy": 0.9271492786705494, "num_tokens": 83206344.0, "step": 5960 }, { "entropy": 0.3419117519631982, "epoch": 2.820028341993387, "grad_norm": 0.07625547051429749, "learning_rate": 0.00016368048708649004, "loss": 0.31244821548461915, "mean_token_accuracy": 0.9216959983110428, "num_tokens": 83347742.0, "step": 5970 }, { "entropy": 0.33581029921770095, "epoch": 2.8247520075578647, "grad_norm": 0.08543124049901962, "learning_rate": 0.0001635657269764965, "loss": 0.288441801071167, "mean_token_accuracy": 0.9250920355319977, "num_tokens": 83488902.0, "step": 5980 }, { "entropy": 0.2990567754022777, "epoch": 2.829475673122343, "grad_norm": 0.061445217579603195, "learning_rate": 0.00016345082621808358, "loss": 0.2649609804153442, "mean_token_accuracy": 0.9306038342416286, "num_tokens": 83627444.0, "step": 5990 }, { "entropy": 0.3461949028074741, "epoch": 2.8341993386868207, "grad_norm": 0.08085981756448746, "learning_rate": 0.00016333578506548595, "loss": 0.30558631420135496, "mean_token_accuracy": 0.9227724611759186, "num_tokens": 83767681.0, "step": 6000 }, { "entropy": 0.33384765721857546, "epoch": 2.838923004251299, "grad_norm": 0.0700036883354187, "learning_rate": 0.0001632206037732489, "loss": 0.2932481288909912, "mean_token_accuracy": 0.924432634562254, "num_tokens": 83909784.0, "step": 6010 }, { "entropy": 0.3035454021766782, "epoch": 2.843646669815777, "grad_norm": 0.059564728289842606, "learning_rate": 0.0001631052825962278, "loss": 0.276937460899353, "mean_token_accuracy": 0.9300597168505191, "num_tokens": 84050579.0, "step": 6020 }, { "entropy": 0.3417934877797961, "epoch": 2.848370335380255, "grad_norm": 0.08422790467739105, "learning_rate": 0.00016298982178958755, "loss": 0.30077276229858396, "mean_token_accuracy": 0.9234837457537651, "num_tokens": 84182932.0, "step": 6030 }, { "entropy": 0.3213076775893569, "epoch": 2.853094000944733, "grad_norm": 0.07368471473455429, "learning_rate": 0.00016287422160880195, "loss": 0.28111531734466555, "mean_token_accuracy": 0.9266740024089813, "num_tokens": 84324453.0, "step": 6040 }, { "entropy": 0.3168294267728925, "epoch": 2.857817666509211, "grad_norm": 0.07869284600019455, "learning_rate": 0.00016275848230965325, "loss": 0.2750498056411743, "mean_token_accuracy": 0.9285811208188534, "num_tokens": 84463560.0, "step": 6050 }, { "entropy": 0.3199483294971287, "epoch": 2.862541332073689, "grad_norm": 0.07346589118242264, "learning_rate": 0.00016264260414823145, "loss": 0.27833597660064696, "mean_token_accuracy": 0.9264893747866154, "num_tokens": 84604085.0, "step": 6060 }, { "entropy": 0.31892386665567757, "epoch": 2.8672649976381672, "grad_norm": 0.0658896416425705, "learning_rate": 0.0001625265873809339, "loss": 0.272175931930542, "mean_token_accuracy": 0.9286142311990261, "num_tokens": 84737551.0, "step": 6070 }, { "entropy": 0.3409381164237857, "epoch": 2.8719886632026452, "grad_norm": 0.05578494071960449, "learning_rate": 0.0001624104322644645, "loss": 0.3090266227722168, "mean_token_accuracy": 0.9216448545455933, "num_tokens": 84875724.0, "step": 6080 }, { "entropy": 0.3283074229024351, "epoch": 2.8767123287671232, "grad_norm": 0.09109394997358322, "learning_rate": 0.00016229413905583342, "loss": 0.2874417304992676, "mean_token_accuracy": 0.9258060708642006, "num_tokens": 85013623.0, "step": 6090 }, { "entropy": 0.2992855066433549, "epoch": 2.8814359943316012, "grad_norm": 0.0764261856675148, "learning_rate": 0.00016217770801235624, "loss": 0.2696357727050781, "mean_token_accuracy": 0.929871492832899, "num_tokens": 85147931.0, "step": 6100 }, { "entropy": 0.29618923347443343, "epoch": 2.8861596598960793, "grad_norm": 0.06750992685556412, "learning_rate": 0.00016206113939165365, "loss": 0.2536092519760132, "mean_token_accuracy": 0.9344585418701172, "num_tokens": 85284684.0, "step": 6110 }, { "entropy": 0.2895301982760429, "epoch": 2.8908833254605573, "grad_norm": 0.07175379991531372, "learning_rate": 0.00016194443345165062, "loss": 0.25012803077697754, "mean_token_accuracy": 0.9349378451704979, "num_tokens": 85425297.0, "step": 6120 }, { "entropy": 0.28265368901193144, "epoch": 2.8956069910250353, "grad_norm": 0.06473183631896973, "learning_rate": 0.00016182759045057607, "loss": 0.2529444217681885, "mean_token_accuracy": 0.9354198552668095, "num_tokens": 85567865.0, "step": 6130 }, { "entropy": 0.3008987005800009, "epoch": 2.9003306565895137, "grad_norm": 0.07270916551351547, "learning_rate": 0.00016171061064696208, "loss": 0.2582638025283813, "mean_token_accuracy": 0.9329213611781597, "num_tokens": 85710103.0, "step": 6140 }, { "entropy": 0.3001433851197362, "epoch": 2.9050543221539913, "grad_norm": 0.0838063433766365, "learning_rate": 0.00016159349429964358, "loss": 0.2681830644607544, "mean_token_accuracy": 0.933569273352623, "num_tokens": 85852779.0, "step": 6150 }, { "entropy": 0.3165293637663126, "epoch": 2.9097779877184697, "grad_norm": 0.06948206573724747, "learning_rate": 0.00016147624166775744, "loss": 0.2737043619155884, "mean_token_accuracy": 0.9284098230302333, "num_tokens": 85994124.0, "step": 6160 }, { "entropy": 0.32346988916397096, "epoch": 2.9145016532829473, "grad_norm": 0.08191309869289398, "learning_rate": 0.0001613588530107422, "loss": 0.275394868850708, "mean_token_accuracy": 0.9277596913278103, "num_tokens": 86130678.0, "step": 6170 }, { "entropy": 0.288114645332098, "epoch": 2.9192253188474258, "grad_norm": 0.06179473176598549, "learning_rate": 0.00016124132858833735, "loss": 0.25613381862640383, "mean_token_accuracy": 0.9330091632902622, "num_tokens": 86264282.0, "step": 6180 }, { "entropy": 0.3138307509943843, "epoch": 2.9239489844119038, "grad_norm": 0.08615489304065704, "learning_rate": 0.00016112366866058282, "loss": 0.27022275924682615, "mean_token_accuracy": 0.9309975825250149, "num_tokens": 86395752.0, "step": 6190 }, { "entropy": 0.3056884821504354, "epoch": 2.9286726499763818, "grad_norm": 0.054475221782922745, "learning_rate": 0.0001610058734878182, "loss": 0.2727755308151245, "mean_token_accuracy": 0.9301555290818214, "num_tokens": 86535633.0, "step": 6200 }, { "entropy": 0.28974386136978864, "epoch": 2.9333963155408598, "grad_norm": 0.07793719321489334, "learning_rate": 0.00016088794333068252, "loss": 0.2515974521636963, "mean_token_accuracy": 0.9339422516524791, "num_tokens": 86679523.0, "step": 6210 }, { "entropy": 0.3044612675905228, "epoch": 2.938119981105338, "grad_norm": 0.07805371284484863, "learning_rate": 0.00016076987845011344, "loss": 0.26934287548065183, "mean_token_accuracy": 0.9311383672058582, "num_tokens": 86818047.0, "step": 6220 }, { "entropy": 0.34343143105506896, "epoch": 2.942843646669816, "grad_norm": 0.07105180621147156, "learning_rate": 0.00016065167910734666, "loss": 0.3158273458480835, "mean_token_accuracy": 0.9232825860381126, "num_tokens": 86960750.0, "step": 6230 }, { "entropy": 0.3021335743367672, "epoch": 2.947567312234294, "grad_norm": 0.07468222081661224, "learning_rate": 0.00016053334556391535, "loss": 0.25798649787902833, "mean_token_accuracy": 0.9327261500060559, "num_tokens": 87101702.0, "step": 6240 }, { "entropy": 0.2945020263083279, "epoch": 2.952290977798772, "grad_norm": 0.05098683387041092, "learning_rate": 0.00016041487808164982, "loss": 0.2529811143875122, "mean_token_accuracy": 0.9338640421628952, "num_tokens": 87237214.0, "step": 6250 }, { "entropy": 0.30731857661157846, "epoch": 2.95701464336325, "grad_norm": 0.07433168590068817, "learning_rate": 0.0001602962769226765, "loss": 0.2620837926864624, "mean_token_accuracy": 0.9326300747692585, "num_tokens": 87378353.0, "step": 6260 }, { "entropy": 0.3260970426723361, "epoch": 2.961738308927728, "grad_norm": 0.06732673197984695, "learning_rate": 0.0001601775423494178, "loss": 0.292596435546875, "mean_token_accuracy": 0.9259773470461369, "num_tokens": 87517051.0, "step": 6270 }, { "entropy": 0.31896489188075067, "epoch": 2.966461974492206, "grad_norm": 0.09348985552787781, "learning_rate": 0.00016005867462459117, "loss": 0.28274738788604736, "mean_token_accuracy": 0.9286909833550453, "num_tokens": 87653424.0, "step": 6280 }, { "entropy": 0.32120455633848904, "epoch": 2.971185640056684, "grad_norm": 0.07910658419132233, "learning_rate": 0.00015993967401120877, "loss": 0.27887530326843263, "mean_token_accuracy": 0.9292895056307315, "num_tokens": 87792261.0, "step": 6290 }, { "entropy": 0.3166860887780786, "epoch": 2.975909305621162, "grad_norm": 0.07976067811250687, "learning_rate": 0.00015982054077257678, "loss": 0.2847684621810913, "mean_token_accuracy": 0.9279345542192459, "num_tokens": 87924254.0, "step": 6300 }, { "entropy": 0.30990776466205716, "epoch": 2.9806329711856403, "grad_norm": 0.06893040984869003, "learning_rate": 0.00015970127517229483, "loss": 0.2684711217880249, "mean_token_accuracy": 0.930588485300541, "num_tokens": 88064452.0, "step": 6310 }, { "entropy": 0.29596504513174293, "epoch": 2.985356636750118, "grad_norm": 0.08254679292440414, "learning_rate": 0.00015958187747425537, "loss": 0.2469400405883789, "mean_token_accuracy": 0.9356708757579326, "num_tokens": 88197423.0, "step": 6320 }, { "entropy": 0.3303051806986332, "epoch": 2.9900803023145963, "grad_norm": 0.05904023349285126, "learning_rate": 0.00015946234794264323, "loss": 0.2958174705505371, "mean_token_accuracy": 0.926980547606945, "num_tokens": 88340784.0, "step": 6330 }, { "entropy": 0.31454561464488506, "epoch": 2.9948039678790743, "grad_norm": 0.07458409667015076, "learning_rate": 0.00015934268684193484, "loss": 0.27035863399505616, "mean_token_accuracy": 0.9297575414180755, "num_tokens": 88480345.0, "step": 6340 }, { "entropy": 0.29515541810542345, "epoch": 2.9995276334435523, "grad_norm": 0.07899592071771622, "learning_rate": 0.00015922289443689785, "loss": 0.25933036804199217, "mean_token_accuracy": 0.932505565136671, "num_tokens": 88619762.0, "step": 6350 }, { "entropy": 0.29944039676338435, "epoch": 3.0042512990080303, "grad_norm": 0.07667111605405807, "learning_rate": 0.00015910297099259034, "loss": 0.2365949869155884, "mean_token_accuracy": 0.9397827386856079, "num_tokens": 88754353.0, "step": 6360 }, { "entropy": 0.27041827561333776, "epoch": 3.0089749645725083, "grad_norm": 0.09656394273042679, "learning_rate": 0.00015898291677436042, "loss": 0.22616770267486572, "mean_token_accuracy": 0.9424898579716683, "num_tokens": 88890997.0, "step": 6370 }, { "entropy": 0.28427893491461875, "epoch": 3.0136986301369864, "grad_norm": 0.06355519592761993, "learning_rate": 0.00015886273204784548, "loss": 0.2329385757446289, "mean_token_accuracy": 0.9410013981163502, "num_tokens": 89032170.0, "step": 6380 }, { "entropy": 0.2778877759352326, "epoch": 3.0184222957014644, "grad_norm": 0.07877601683139801, "learning_rate": 0.0001587424170789717, "loss": 0.2319040298461914, "mean_token_accuracy": 0.9415215879678727, "num_tokens": 89178899.0, "step": 6390 }, { "entropy": 0.2784922879189253, "epoch": 3.0231459612659424, "grad_norm": 0.06589914858341217, "learning_rate": 0.0001586219721339535, "loss": 0.2364208459854126, "mean_token_accuracy": 0.9419798776507378, "num_tokens": 89325229.0, "step": 6400 }, { "entropy": 0.29138854378834367, "epoch": 3.0278696268304204, "grad_norm": 0.06874479353427887, "learning_rate": 0.00015850139747929283, "loss": 0.22848091125488282, "mean_token_accuracy": 0.9409735448658466, "num_tokens": 89467865.0, "step": 6410 }, { "entropy": 0.2573707280680537, "epoch": 3.0325932923948984, "grad_norm": 0.05927377566695213, "learning_rate": 0.00015838069338177863, "loss": 0.20626981258392335, "mean_token_accuracy": 0.9469245828688144, "num_tokens": 89601896.0, "step": 6420 }, { "entropy": 0.24344939803704618, "epoch": 3.0373169579593764, "grad_norm": 0.06410811841487885, "learning_rate": 0.00015825986010848634, "loss": 0.19441936016082764, "mean_token_accuracy": 0.9505223080515861, "num_tokens": 89742477.0, "step": 6430 }, { "entropy": 0.25771507173776625, "epoch": 3.0420406235238544, "grad_norm": 0.053573086857795715, "learning_rate": 0.00015813889792677706, "loss": 0.21072278022766114, "mean_token_accuracy": 0.9468448996543884, "num_tokens": 89872459.0, "step": 6440 }, { "entropy": 0.24889703365042806, "epoch": 3.0467642890883324, "grad_norm": 0.07980092614889145, "learning_rate": 0.0001580178071042973, "loss": 0.2015373945236206, "mean_token_accuracy": 0.9471607655286789, "num_tokens": 90011569.0, "step": 6450 }, { "entropy": 0.2616048602387309, "epoch": 3.0514879546528104, "grad_norm": 0.06496746838092804, "learning_rate": 0.00015789658790897807, "loss": 0.21244680881500244, "mean_token_accuracy": 0.9469091944396496, "num_tokens": 90150606.0, "step": 6460 }, { "entropy": 0.266824578307569, "epoch": 3.0562116202172884, "grad_norm": 0.06718926131725311, "learning_rate": 0.0001577752406090345, "loss": 0.22119715213775634, "mean_token_accuracy": 0.9423008799552918, "num_tokens": 90276269.0, "step": 6470 }, { "entropy": 0.27743001822382213, "epoch": 3.0609352857817664, "grad_norm": 0.07867974787950516, "learning_rate": 0.0001576537654729652, "loss": 0.2343437910079956, "mean_token_accuracy": 0.9408437065780163, "num_tokens": 90412699.0, "step": 6480 }, { "entropy": 0.26375128515064716, "epoch": 3.065658951346245, "grad_norm": 0.05987164005637169, "learning_rate": 0.0001575321627695515, "loss": 0.2072808027267456, "mean_token_accuracy": 0.9460294567048549, "num_tokens": 90547399.0, "step": 6490 }, { "entropy": 0.2649696357548237, "epoch": 3.070382616910723, "grad_norm": 0.06754233688116074, "learning_rate": 0.00015741043276785709, "loss": 0.21425178050994872, "mean_token_accuracy": 0.9447133392095566, "num_tokens": 90687331.0, "step": 6500 } ], "logging_steps": 10, "max_steps": 21170, "num_input_tokens_seen": 0, "num_train_epochs": 10, "save_steps": 500, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": false }, "attributes": {} } }, "total_flos": 1.3672583194431283e+18, "train_batch_size": 4, "trial_name": null, "trial_params": null }
{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 3.070382616910723,
"eval_steps": 500,
"global_step": 6500,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"entropy": 5.8557036250829695,
"epoch": 0.004723665564478035,
"grad_norm": 1.1635308265686035,
"learning_rate": 3.6e-05,
"loss": 9.016449737548829,
"mean_token_accuracy": 0.030926086311228573,
"num_tokens": 134075.0,
"step": 10
},
{
"entropy": 6.803842556476593,
"epoch": 0.00944733112895607,
"grad_norm": 0.6144386529922485,
"learning_rate": 7.6e-05,
"loss": 6.777875518798828,
"mean_token_accuracy": 0.09398306957446038,
"num_tokens": 266103.0,
"step": 20
},
{
"entropy": 5.3631006419658664,
"epoch": 0.014170996693434105,
"grad_norm": 0.3949488699436188,
"learning_rate": 0.000116,
"loss": 5.411441421508789,
"mean_token_accuracy": 0.1942991679534316,
"num_tokens": 406925.0,
"step": 30
},
{
"entropy": 4.53178850710392,
"epoch": 0.01889466225791214,
"grad_norm": 0.2796455919742584,
"learning_rate": 0.00015600000000000002,
"loss": 4.523899078369141,
"mean_token_accuracy": 0.2943518474698067,
"num_tokens": 549837.0,
"step": 40
},
{
"entropy": 3.7455823540687563,
"epoch": 0.023618327822390175,
"grad_norm": 0.32603198289871216,
"learning_rate": 0.000196,
"loss": 3.8331756591796875,
"mean_token_accuracy": 0.35103216096758844,
"num_tokens": 690318.0,
"step": 50
},
{
"entropy": 3.1043672770261765,
"epoch": 0.02834199338686821,
"grad_norm": 0.31668633222579956,
"learning_rate": 0.00019999991038785437,
"loss": 3.1435787200927736,
"mean_token_accuracy": 0.42379639223217963,
"num_tokens": 824090.0,
"step": 60
},
{
"entropy": 2.728278709948063,
"epoch": 0.03306565895134624,
"grad_norm": 0.34853872656822205,
"learning_rate": 0.00019999960061768058,
"loss": 2.771727180480957,
"mean_token_accuracy": 0.48015250265598297,
"num_tokens": 958789.0,
"step": 70
},
{
"entropy": 2.440419462323189,
"epoch": 0.03778932451582428,
"grad_norm": 0.278272420167923,
"learning_rate": 0.00019999906958384112,
"loss": 2.5111177444458006,
"mean_token_accuracy": 0.5124442256987095,
"num_tokens": 1101049.0,
"step": 80
},
{
"entropy": 2.25365926027298,
"epoch": 0.042512990080302314,
"grad_norm": 0.20885923504829407,
"learning_rate": 0.00019999831728751097,
"loss": 2.28131160736084,
"mean_token_accuracy": 0.5472530923783779,
"num_tokens": 1243306.0,
"step": 90