| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 141420, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07071135624381275, | |
| "grad_norm": 0.950070858001709, | |
| "learning_rate": 3.125e-05, | |
| "loss": 5.0021, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.1414227124876255, | |
| "grad_norm": 0.9159772992134094, | |
| "learning_rate": 6.25e-05, | |
| "loss": 3.5952, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.21213406873143828, | |
| "grad_norm": 0.8742101192474365, | |
| "learning_rate": 9.375e-05, | |
| "loss": 3.3236, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.282845424975251, | |
| "grad_norm": 0.7689530253410339, | |
| "learning_rate": 0.000125, | |
| "loss": 3.1344, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.3535567812190638, | |
| "grad_norm": 0.7027168273925781, | |
| "learning_rate": 0.00015625, | |
| "loss": 3.0058, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.42426813746287656, | |
| "grad_norm": 0.6757282018661499, | |
| "learning_rate": 0.0001875, | |
| "loss": 2.8925, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.4949794937066893, | |
| "grad_norm": 0.6485090851783752, | |
| "learning_rate": 0.00021875, | |
| "loss": 2.8129, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.565690849950502, | |
| "grad_norm": 0.6332668662071228, | |
| "learning_rate": 0.00025, | |
| "loss": 2.7427, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.6364022061943148, | |
| "grad_norm": 0.5585498213768005, | |
| "learning_rate": 0.00028125000000000003, | |
| "loss": 2.6942, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.7071135624381276, | |
| "grad_norm": 0.5789941549301147, | |
| "learning_rate": 0.0003125, | |
| "loss": 2.6592, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.7778249186819404, | |
| "grad_norm": 0.49083390831947327, | |
| "learning_rate": 0.00034371875, | |
| "loss": 2.6102, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.8485362749257531, | |
| "grad_norm": 0.5267419815063477, | |
| "learning_rate": 0.00037496875000000003, | |
| "loss": 2.5811, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.9192476311695659, | |
| "grad_norm": 0.46979865431785583, | |
| "learning_rate": 0.0004061875, | |
| "loss": 2.5689, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.9899589874133786, | |
| "grad_norm": 0.4309781491756439, | |
| "learning_rate": 0.00043740625, | |
| "loss": 2.5477, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.48551466175887975, | |
| "eval_loss": 2.7489795684814453, | |
| "eval_runtime": 122.3768, | |
| "eval_samples_per_second": 382.981, | |
| "eval_steps_per_second": 5.99, | |
| "step": 14142 | |
| }, | |
| { | |
| "epoch": 1.0606703436571914, | |
| "grad_norm": 0.3975456655025482, | |
| "learning_rate": 0.00046865625, | |
| "loss": 2.5021, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.131381699901004, | |
| "grad_norm": 0.40377160906791687, | |
| "learning_rate": 0.00049990625, | |
| "loss": 2.4895, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.2020930561448169, | |
| "grad_norm": 0.36566758155822754, | |
| "learning_rate": 0.000531125, | |
| "loss": 2.4727, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 1.2728044123886297, | |
| "grad_norm": 0.37737491726875305, | |
| "learning_rate": 0.0005623749999999999, | |
| "loss": 2.4652, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.3435157686324424, | |
| "grad_norm": 0.3293197751045227, | |
| "learning_rate": 0.000593625, | |
| "loss": 2.4574, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 1.414227124876255, | |
| "grad_norm": 0.290075421333313, | |
| "learning_rate": 0.000624875, | |
| "loss": 2.4275, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 1.4849384811200679, | |
| "grad_norm": 0.2959868311882019, | |
| "learning_rate": 0.00065609375, | |
| "loss": 2.4386, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 1.5556498373638807, | |
| "grad_norm": 0.27695414423942566, | |
| "learning_rate": 0.00068734375, | |
| "loss": 2.4161, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 1.6263611936076934, | |
| "grad_norm": 0.2632512152194977, | |
| "learning_rate": 0.00071859375, | |
| "loss": 2.4201, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 1.697072549851506, | |
| "grad_norm": 0.2265060991048813, | |
| "learning_rate": 0.0007498125, | |
| "loss": 2.4058, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 1.7677839060953189, | |
| "grad_norm": 0.26139551401138306, | |
| "learning_rate": 0.0007810625, | |
| "loss": 2.3958, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 1.8384952623391317, | |
| "grad_norm": 0.2395378053188324, | |
| "learning_rate": 0.0008123125, | |
| "loss": 2.3868, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 1.9092066185829444, | |
| "grad_norm": 0.23772157728672028, | |
| "learning_rate": 0.00084353125, | |
| "loss": 2.3766, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 1.979917974826757, | |
| "grad_norm": 0.23179960250854492, | |
| "learning_rate": 0.00087478125, | |
| "loss": 2.377, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5044756048031578, | |
| "eval_loss": 2.580270290374756, | |
| "eval_runtime": 126.9742, | |
| "eval_samples_per_second": 369.114, | |
| "eval_steps_per_second": 5.773, | |
| "step": 28284 | |
| }, | |
| { | |
| "epoch": 2.05062933107057, | |
| "grad_norm": 0.25655558705329895, | |
| "learning_rate": 0.0009060312499999999, | |
| "loss": 2.3364, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 2.1213406873143827, | |
| "grad_norm": 0.22631210088729858, | |
| "learning_rate": 0.00093725, | |
| "loss": 2.3297, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 2.1920520435581956, | |
| "grad_norm": 0.2411614954471588, | |
| "learning_rate": 0.0009685000000000001, | |
| "loss": 2.3249, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 2.262763399802008, | |
| "grad_norm": 0.25610190629959106, | |
| "learning_rate": 0.00099975, | |
| "loss": 2.3166, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 2.333474756045821, | |
| "grad_norm": 0.2797723412513733, | |
| "learning_rate": 0.000990943154816304, | |
| "loss": 2.3086, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 2.4041861122896337, | |
| "grad_norm": 0.2100619226694107, | |
| "learning_rate": 0.0009818040577590935, | |
| "loss": 2.3117, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 2.4748974685334466, | |
| "grad_norm": 0.2163330614566803, | |
| "learning_rate": 0.0009726740997989398, | |
| "loss": 2.302, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 2.5456088247772595, | |
| "grad_norm": 0.22995002567768097, | |
| "learning_rate": 0.0009635350027417291, | |
| "loss": 2.2872, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 2.616320181021072, | |
| "grad_norm": 0.2004874050617218, | |
| "learning_rate": 0.0009544141838786328, | |
| "loss": 2.2843, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 2.6870315372648848, | |
| "grad_norm": 0.2153329849243164, | |
| "learning_rate": 0.0009452750868214221, | |
| "loss": 2.2788, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 2.7577428935086976, | |
| "grad_norm": 0.22890245914459229, | |
| "learning_rate": 0.0009361359897642113, | |
| "loss": 2.2711, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 2.82845424975251, | |
| "grad_norm": 0.2471633404493332, | |
| "learning_rate": 0.0009269968927070006, | |
| "loss": 2.2673, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 2.899165605996323, | |
| "grad_norm": 0.20824675261974335, | |
| "learning_rate": 0.0009178577956497898, | |
| "loss": 2.2589, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 2.9698769622401358, | |
| "grad_norm": 0.27108854055404663, | |
| "learning_rate": 0.0009087278376896363, | |
| "loss": 2.2625, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.5164308635329491, | |
| "eval_loss": 2.4760255813598633, | |
| "eval_runtime": 124.5514, | |
| "eval_samples_per_second": 376.295, | |
| "eval_steps_per_second": 5.885, | |
| "step": 42426 | |
| }, | |
| { | |
| "epoch": 3.0405883184839486, | |
| "grad_norm": 0.2760034501552582, | |
| "learning_rate": 0.0008995887406324256, | |
| "loss": 2.22, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 3.1112996747277615, | |
| "grad_norm": 0.24826580286026, | |
| "learning_rate": 0.0008904587826722719, | |
| "loss": 2.1969, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 3.182011030971574, | |
| "grad_norm": 0.1961706429719925, | |
| "learning_rate": 0.0008813196856150612, | |
| "loss": 2.1929, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 3.2527223872153868, | |
| "grad_norm": 0.2030291110277176, | |
| "learning_rate": 0.0008721805885578505, | |
| "loss": 2.1974, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 3.3234337434591996, | |
| "grad_norm": 0.24897335469722748, | |
| "learning_rate": 0.0008630414915006397, | |
| "loss": 2.1993, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 3.3941450997030125, | |
| "grad_norm": 0.2421874701976776, | |
| "learning_rate": 0.0008539115335404863, | |
| "loss": 2.186, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 3.464856455946825, | |
| "grad_norm": 0.2960880398750305, | |
| "learning_rate": 0.0008447724364832756, | |
| "loss": 2.1877, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 3.5355678121906378, | |
| "grad_norm": 0.20504000782966614, | |
| "learning_rate": 0.0008356424785231219, | |
| "loss": 2.1816, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 3.6062791684344506, | |
| "grad_norm": 0.23933938145637512, | |
| "learning_rate": 0.0008265033814659112, | |
| "loss": 2.1807, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 3.6769905246782635, | |
| "grad_norm": 0.23281992971897125, | |
| "learning_rate": 0.0008173734235057576, | |
| "loss": 2.1804, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 3.747701880922076, | |
| "grad_norm": 0.20451070368289948, | |
| "learning_rate": 0.0008082343264485469, | |
| "loss": 2.1807, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 3.8184132371658888, | |
| "grad_norm": 0.23853066563606262, | |
| "learning_rate": 0.0007990952293913362, | |
| "loss": 2.1767, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 3.8891245934097016, | |
| "grad_norm": 0.24050584435462952, | |
| "learning_rate": 0.0007899652714311825, | |
| "loss": 2.1667, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 3.9598359496535145, | |
| "grad_norm": 0.21722733974456787, | |
| "learning_rate": 0.0007808261743739718, | |
| "loss": 2.1715, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.523546219218542, | |
| "eval_loss": 2.4206314086914062, | |
| "eval_runtime": 124.5767, | |
| "eval_samples_per_second": 376.218, | |
| "eval_steps_per_second": 5.884, | |
| "step": 56568 | |
| }, | |
| { | |
| "epoch": 4.030547305897327, | |
| "grad_norm": 0.2417266070842743, | |
| "learning_rate": 0.0007716962164138184, | |
| "loss": 2.1342, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 4.10125866214114, | |
| "grad_norm": 0.22849377989768982, | |
| "learning_rate": 0.0007625571193566076, | |
| "loss": 2.1082, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 4.171970018384952, | |
| "grad_norm": 0.226350799202919, | |
| "learning_rate": 0.000753427161396454, | |
| "loss": 2.1144, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 4.2426813746287655, | |
| "grad_norm": 0.207255020737648, | |
| "learning_rate": 0.0007442880643392433, | |
| "loss": 2.1116, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 4.313392730872578, | |
| "grad_norm": 0.21048206090927124, | |
| "learning_rate": 0.0007351581063790898, | |
| "loss": 2.1092, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 4.384104087116391, | |
| "grad_norm": 0.26391103863716125, | |
| "learning_rate": 0.000726019009321879, | |
| "loss": 2.1105, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 4.454815443360204, | |
| "grad_norm": 0.22511842846870422, | |
| "learning_rate": 0.0007168799122646683, | |
| "loss": 2.1111, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 4.525526799604016, | |
| "grad_norm": 0.264876127243042, | |
| "learning_rate": 0.0007077499543045147, | |
| "loss": 2.1069, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 4.596238155847829, | |
| "grad_norm": 0.20152664184570312, | |
| "learning_rate": 0.0006986199963443612, | |
| "loss": 2.1107, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 4.666949512091642, | |
| "grad_norm": 0.2202775925397873, | |
| "learning_rate": 0.0006894808992871504, | |
| "loss": 2.1192, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 4.737660868335455, | |
| "grad_norm": 0.2462519109249115, | |
| "learning_rate": 0.0006803418022299397, | |
| "loss": 2.1094, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 4.8083722245792675, | |
| "grad_norm": 0.24858810007572174, | |
| "learning_rate": 0.000671202705172729, | |
| "loss": 2.116, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 4.87908358082308, | |
| "grad_norm": 0.25709378719329834, | |
| "learning_rate": 0.0006620727472125753, | |
| "loss": 2.1125, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 4.949794937066893, | |
| "grad_norm": 0.22404509782791138, | |
| "learning_rate": 0.0006529336501553646, | |
| "loss": 2.0996, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.5277767446216196, | |
| "eval_loss": 2.388011932373047, | |
| "eval_runtime": 124.6122, | |
| "eval_samples_per_second": 376.111, | |
| "eval_steps_per_second": 5.882, | |
| "step": 70710 | |
| }, | |
| { | |
| "epoch": 5.020506293310706, | |
| "grad_norm": 0.21308551728725433, | |
| "learning_rate": 0.0006438036921952112, | |
| "loss": 2.0836, | |
| "step": 71000 | |
| }, | |
| { | |
| "epoch": 5.091217649554518, | |
| "grad_norm": 0.26624125242233276, | |
| "learning_rate": 0.0006346645951380004, | |
| "loss": 2.0401, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 5.161929005798331, | |
| "grad_norm": 0.24497570097446442, | |
| "learning_rate": 0.0006255254980807897, | |
| "loss": 2.0422, | |
| "step": 73000 | |
| }, | |
| { | |
| "epoch": 5.232640362042144, | |
| "grad_norm": 0.2567467987537384, | |
| "learning_rate": 0.0006163955401206361, | |
| "loss": 2.044, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 5.303351718285957, | |
| "grad_norm": 0.23071114718914032, | |
| "learning_rate": 0.0006072564430634253, | |
| "loss": 2.0465, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 5.3740630745297695, | |
| "grad_norm": 0.25491389632225037, | |
| "learning_rate": 0.0005981264851032718, | |
| "loss": 2.0482, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 5.444774430773582, | |
| "grad_norm": 0.2559095621109009, | |
| "learning_rate": 0.0005889873880460611, | |
| "loss": 2.0496, | |
| "step": 77000 | |
| }, | |
| { | |
| "epoch": 5.515485787017395, | |
| "grad_norm": 0.22284868359565735, | |
| "learning_rate": 0.0005798574300859076, | |
| "loss": 2.0525, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 5.586197143261208, | |
| "grad_norm": 0.23331965506076813, | |
| "learning_rate": 0.0005707183330286967, | |
| "loss": 2.0649, | |
| "step": 79000 | |
| }, | |
| { | |
| "epoch": 5.65690849950502, | |
| "grad_norm": 0.2457083910703659, | |
| "learning_rate": 0.0005615883750685432, | |
| "loss": 2.0515, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 5.727619855748833, | |
| "grad_norm": 0.2747284173965454, | |
| "learning_rate": 0.0005524492780113325, | |
| "loss": 2.0636, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 5.798331211992646, | |
| "grad_norm": 0.2696532607078552, | |
| "learning_rate": 0.0005433101809541218, | |
| "loss": 2.0548, | |
| "step": 82000 | |
| }, | |
| { | |
| "epoch": 5.869042568236459, | |
| "grad_norm": 0.2052222490310669, | |
| "learning_rate": 0.0005341802229939682, | |
| "loss": 2.0486, | |
| "step": 83000 | |
| }, | |
| { | |
| "epoch": 5.9397539244802715, | |
| "grad_norm": 0.2313115894794464, | |
| "learning_rate": 0.0005250411259367574, | |
| "loss": 2.0456, | |
| "step": 84000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.5305995578608221, | |
| "eval_loss": 2.372159957885742, | |
| "eval_runtime": 124.7467, | |
| "eval_samples_per_second": 375.705, | |
| "eval_steps_per_second": 5.876, | |
| "step": 84852 | |
| }, | |
| { | |
| "epoch": 6.010465280724084, | |
| "grad_norm": 0.2167889028787613, | |
| "learning_rate": 0.0005159020288795467, | |
| "loss": 2.0537, | |
| "step": 85000 | |
| }, | |
| { | |
| "epoch": 6.081176636967897, | |
| "grad_norm": 0.2274264246225357, | |
| "learning_rate": 0.0005067720709193931, | |
| "loss": 1.9868, | |
| "step": 86000 | |
| }, | |
| { | |
| "epoch": 6.15188799321171, | |
| "grad_norm": 0.21367891132831573, | |
| "learning_rate": 0.0004976329738621824, | |
| "loss": 1.9966, | |
| "step": 87000 | |
| }, | |
| { | |
| "epoch": 6.222599349455523, | |
| "grad_norm": 0.23541270196437836, | |
| "learning_rate": 0.0004885030159020289, | |
| "loss": 2.0013, | |
| "step": 88000 | |
| }, | |
| { | |
| "epoch": 6.293310705699335, | |
| "grad_norm": 0.24054056406021118, | |
| "learning_rate": 0.0004793639188448181, | |
| "loss": 1.9875, | |
| "step": 89000 | |
| }, | |
| { | |
| "epoch": 6.364022061943148, | |
| "grad_norm": 0.23533108830451965, | |
| "learning_rate": 0.00047023396088466456, | |
| "loss": 1.9967, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 6.434733418186961, | |
| "grad_norm": 0.2795858383178711, | |
| "learning_rate": 0.00046110400292451105, | |
| "loss": 2.0019, | |
| "step": 91000 | |
| }, | |
| { | |
| "epoch": 6.5054447744307735, | |
| "grad_norm": 0.2619366943836212, | |
| "learning_rate": 0.00045196490586730035, | |
| "loss": 2.0028, | |
| "step": 92000 | |
| }, | |
| { | |
| "epoch": 6.576156130674587, | |
| "grad_norm": 0.24517033994197845, | |
| "learning_rate": 0.00044282580881008953, | |
| "loss": 2.0044, | |
| "step": 93000 | |
| }, | |
| { | |
| "epoch": 6.646867486918399, | |
| "grad_norm": 0.2473345845937729, | |
| "learning_rate": 0.0004336867117528788, | |
| "loss": 1.9935, | |
| "step": 94000 | |
| }, | |
| { | |
| "epoch": 6.717578843162212, | |
| "grad_norm": 0.229476198554039, | |
| "learning_rate": 0.00042454761469566806, | |
| "loss": 1.9943, | |
| "step": 95000 | |
| }, | |
| { | |
| "epoch": 6.788290199406025, | |
| "grad_norm": 0.2603880763053894, | |
| "learning_rate": 0.0004154176567355145, | |
| "loss": 2.0073, | |
| "step": 96000 | |
| }, | |
| { | |
| "epoch": 6.859001555649837, | |
| "grad_norm": 0.2473394274711609, | |
| "learning_rate": 0.0004062785596783038, | |
| "loss": 2.012, | |
| "step": 97000 | |
| }, | |
| { | |
| "epoch": 6.92971291189365, | |
| "grad_norm": 0.24815410375595093, | |
| "learning_rate": 0.00039714860171815024, | |
| "loss": 1.9983, | |
| "step": 98000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.5327383571510881, | |
| "eval_loss": 2.359189748764038, | |
| "eval_runtime": 124.2745, | |
| "eval_samples_per_second": 377.133, | |
| "eval_steps_per_second": 5.898, | |
| "step": 98994 | |
| }, | |
| { | |
| "epoch": 7.000424268137463, | |
| "grad_norm": 0.2351510226726532, | |
| "learning_rate": 0.0003880095046609395, | |
| "loss": 2.0039, | |
| "step": 99000 | |
| }, | |
| { | |
| "epoch": 7.0711356243812755, | |
| "grad_norm": 0.27717456221580505, | |
| "learning_rate": 0.0003788704076037288, | |
| "loss": 1.9358, | |
| "step": 100000 | |
| }, | |
| { | |
| "epoch": 7.141846980625088, | |
| "grad_norm": 0.2275402843952179, | |
| "learning_rate": 0.0003697495887406324, | |
| "loss": 1.9391, | |
| "step": 101000 | |
| }, | |
| { | |
| "epoch": 7.212558336868901, | |
| "grad_norm": 0.3047560453414917, | |
| "learning_rate": 0.0003606104916834217, | |
| "loss": 1.9472, | |
| "step": 102000 | |
| }, | |
| { | |
| "epoch": 7.283269693112714, | |
| "grad_norm": 0.2828271687030792, | |
| "learning_rate": 0.0003514713946262109, | |
| "loss": 1.9415, | |
| "step": 103000 | |
| }, | |
| { | |
| "epoch": 7.353981049356527, | |
| "grad_norm": 0.27506691217422485, | |
| "learning_rate": 0.0003423322975690002, | |
| "loss": 1.9467, | |
| "step": 104000 | |
| }, | |
| { | |
| "epoch": 7.424692405600339, | |
| "grad_norm": 0.23766624927520752, | |
| "learning_rate": 0.0003332023396088467, | |
| "loss": 1.9559, | |
| "step": 105000 | |
| }, | |
| { | |
| "epoch": 7.495403761844152, | |
| "grad_norm": 0.25875958800315857, | |
| "learning_rate": 0.00032406324255163587, | |
| "loss": 1.9569, | |
| "step": 106000 | |
| }, | |
| { | |
| "epoch": 7.566115118087965, | |
| "grad_norm": 0.24315999448299408, | |
| "learning_rate": 0.00031493328459148237, | |
| "loss": 1.9567, | |
| "step": 107000 | |
| }, | |
| { | |
| "epoch": 7.6368264743317775, | |
| "grad_norm": 0.29540637135505676, | |
| "learning_rate": 0.0003057941875342716, | |
| "loss": 1.9567, | |
| "step": 108000 | |
| }, | |
| { | |
| "epoch": 7.707537830575591, | |
| "grad_norm": 0.2695215940475464, | |
| "learning_rate": 0.00029665509047706085, | |
| "loss": 1.9605, | |
| "step": 109000 | |
| }, | |
| { | |
| "epoch": 7.778249186819403, | |
| "grad_norm": 0.2626267075538635, | |
| "learning_rate": 0.00028751599341985014, | |
| "loss": 1.9527, | |
| "step": 110000 | |
| }, | |
| { | |
| "epoch": 7.848960543063216, | |
| "grad_norm": 0.25401443243026733, | |
| "learning_rate": 0.0002783768963626394, | |
| "loss": 1.966, | |
| "step": 111000 | |
| }, | |
| { | |
| "epoch": 7.919671899307029, | |
| "grad_norm": 0.22462651133537292, | |
| "learning_rate": 0.0002692377993054286, | |
| "loss": 1.9584, | |
| "step": 112000 | |
| }, | |
| { | |
| "epoch": 7.990383255550841, | |
| "grad_norm": 0.28454989194869995, | |
| "learning_rate": 0.0002601078413452751, | |
| "loss": 1.9482, | |
| "step": 113000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.5338266026714983, | |
| "eval_loss": 2.357896327972412, | |
| "eval_runtime": 124.302, | |
| "eval_samples_per_second": 377.05, | |
| "eval_steps_per_second": 5.897, | |
| "step": 113136 | |
| }, | |
| { | |
| "epoch": 8.061094611794655, | |
| "grad_norm": 0.26728782057762146, | |
| "learning_rate": 0.0002509687442880643, | |
| "loss": 1.8937, | |
| "step": 114000 | |
| }, | |
| { | |
| "epoch": 8.131805968038467, | |
| "grad_norm": 0.24568845331668854, | |
| "learning_rate": 0.00024183878632791082, | |
| "loss": 1.8983, | |
| "step": 115000 | |
| }, | |
| { | |
| "epoch": 8.20251732428228, | |
| "grad_norm": 0.2506534457206726, | |
| "learning_rate": 0.0002327088283677573, | |
| "loss": 1.9067, | |
| "step": 116000 | |
| }, | |
| { | |
| "epoch": 8.273228680526092, | |
| "grad_norm": 0.31150582432746887, | |
| "learning_rate": 0.00022356973131054653, | |
| "loss": 1.9085, | |
| "step": 117000 | |
| }, | |
| { | |
| "epoch": 8.343940036769904, | |
| "grad_norm": 0.2992372512817383, | |
| "learning_rate": 0.00021443063425333577, | |
| "loss": 1.9085, | |
| "step": 118000 | |
| }, | |
| { | |
| "epoch": 8.414651393013719, | |
| "grad_norm": 0.23084251582622528, | |
| "learning_rate": 0.00020529153719612504, | |
| "loss": 1.91, | |
| "step": 119000 | |
| }, | |
| { | |
| "epoch": 8.485362749257531, | |
| "grad_norm": 0.2754100561141968, | |
| "learning_rate": 0.0001961615792359715, | |
| "loss": 1.9076, | |
| "step": 120000 | |
| }, | |
| { | |
| "epoch": 8.556074105501343, | |
| "grad_norm": 0.25915420055389404, | |
| "learning_rate": 0.00018702248217876074, | |
| "loss": 1.915, | |
| "step": 121000 | |
| }, | |
| { | |
| "epoch": 8.626785461745156, | |
| "grad_norm": 0.26031365990638733, | |
| "learning_rate": 0.00017788338512155, | |
| "loss": 1.9159, | |
| "step": 122000 | |
| }, | |
| { | |
| "epoch": 8.697496817988968, | |
| "grad_norm": 0.2626364231109619, | |
| "learning_rate": 0.00016875342716139648, | |
| "loss": 1.905, | |
| "step": 123000 | |
| }, | |
| { | |
| "epoch": 8.768208174232782, | |
| "grad_norm": 0.30089327692985535, | |
| "learning_rate": 0.00015961433010418572, | |
| "loss": 1.9046, | |
| "step": 124000 | |
| }, | |
| { | |
| "epoch": 8.838919530476595, | |
| "grad_norm": 0.2982795536518097, | |
| "learning_rate": 0.00015047523304697496, | |
| "loss": 1.9052, | |
| "step": 125000 | |
| }, | |
| { | |
| "epoch": 8.909630886720407, | |
| "grad_norm": 0.2462824285030365, | |
| "learning_rate": 0.00014134527508682143, | |
| "loss": 1.907, | |
| "step": 126000 | |
| }, | |
| { | |
| "epoch": 8.98034224296422, | |
| "grad_norm": 0.2993851900100708, | |
| "learning_rate": 0.0001322061780296107, | |
| "loss": 1.9061, | |
| "step": 127000 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.5346066724693145, | |
| "eval_loss": 2.3604698181152344, | |
| "eval_runtime": 124.2295, | |
| "eval_samples_per_second": 377.27, | |
| "eval_steps_per_second": 5.9, | |
| "step": 127278 | |
| }, | |
| { | |
| "epoch": 9.051053599208032, | |
| "grad_norm": 0.29052260518074036, | |
| "learning_rate": 0.00012307622006945714, | |
| "loss": 1.8738, | |
| "step": 128000 | |
| }, | |
| { | |
| "epoch": 9.121764955451846, | |
| "grad_norm": 0.2522347569465637, | |
| "learning_rate": 0.0001139462621093036, | |
| "loss": 1.8601, | |
| "step": 129000 | |
| }, | |
| { | |
| "epoch": 9.192476311695659, | |
| "grad_norm": 0.30291104316711426, | |
| "learning_rate": 0.00010480716505209286, | |
| "loss": 1.8793, | |
| "step": 130000 | |
| }, | |
| { | |
| "epoch": 9.263187667939471, | |
| "grad_norm": 0.26850393414497375, | |
| "learning_rate": 9.566806799488211e-05, | |
| "loss": 1.8676, | |
| "step": 131000 | |
| }, | |
| { | |
| "epoch": 9.333899024183284, | |
| "grad_norm": 0.270274817943573, | |
| "learning_rate": 8.652897093767135e-05, | |
| "loss": 1.8669, | |
| "step": 132000 | |
| }, | |
| { | |
| "epoch": 9.404610380427096, | |
| "grad_norm": 0.2623472809791565, | |
| "learning_rate": 7.739901297751782e-05, | |
| "loss": 1.8651, | |
| "step": 133000 | |
| }, | |
| { | |
| "epoch": 9.47532173667091, | |
| "grad_norm": 0.2898847758769989, | |
| "learning_rate": 6.825991592030707e-05, | |
| "loss": 1.8697, | |
| "step": 134000 | |
| }, | |
| { | |
| "epoch": 9.546033092914723, | |
| "grad_norm": 0.30749163031578064, | |
| "learning_rate": 5.912081886309633e-05, | |
| "loss": 1.8635, | |
| "step": 135000 | |
| }, | |
| { | |
| "epoch": 9.616744449158535, | |
| "grad_norm": 0.255743145942688, | |
| "learning_rate": 4.9981721805885585e-05, | |
| "loss": 1.8662, | |
| "step": 136000 | |
| }, | |
| { | |
| "epoch": 9.687455805402347, | |
| "grad_norm": 0.30720534920692444, | |
| "learning_rate": 4.085176384573204e-05, | |
| "loss": 1.8582, | |
| "step": 137000 | |
| }, | |
| { | |
| "epoch": 9.75816716164616, | |
| "grad_norm": 0.2820815145969391, | |
| "learning_rate": 3.17126667885213e-05, | |
| "loss": 1.8694, | |
| "step": 138000 | |
| }, | |
| { | |
| "epoch": 9.828878517889972, | |
| "grad_norm": 0.30231404304504395, | |
| "learning_rate": 2.2582708828367758e-05, | |
| "loss": 1.8654, | |
| "step": 139000 | |
| }, | |
| { | |
| "epoch": 9.899589874133786, | |
| "grad_norm": 0.292241632938385, | |
| "learning_rate": 1.344361177115701e-05, | |
| "loss": 1.8634, | |
| "step": 140000 | |
| }, | |
| { | |
| "epoch": 9.970301230377599, | |
| "grad_norm": 0.25611528754234314, | |
| "learning_rate": 4.313653811003472e-06, | |
| "loss": 1.8692, | |
| "step": 141000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.534805757971141, | |
| "eval_loss": 2.368875503540039, | |
| "eval_runtime": 124.6443, | |
| "eval_samples_per_second": 376.014, | |
| "eval_steps_per_second": 5.881, | |
| "step": 141420 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 141420, | |
| "total_flos": 5.9582336320512e+17, | |
| "train_loss": 2.1796733167279037, | |
| "train_runtime": 31005.8847, | |
| "train_samples_per_second": 145.948, | |
| "train_steps_per_second": 4.561 | |
| } | |
| ], | |
| "logging_steps": 1000, | |
| "max_steps": 141420, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 5000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5.9582336320512e+17, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |