| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 0.52, |
| "eval_steps": 1000, |
| "global_step": 26000, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.002, |
| "grad_norm": 0.7236698865890503, |
| "learning_rate": 0.000396, |
| "loss": 8.129180297851562, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.004, |
| "grad_norm": 0.6699316501617432, |
| "learning_rate": 0.0005999985696164921, |
| "loss": 6.683881225585938, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.006, |
| "grad_norm": 0.4809923470020294, |
| "learning_rate": 0.0005999867739541219, |
| "loss": 6.335140991210937, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.008, |
| "grad_norm": 0.6197829246520996, |
| "learning_rate": 0.0005999630639489827, |
| "loss": 6.119940185546875, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.01, |
| "grad_norm": 0.5310641527175903, |
| "learning_rate": 0.0005999274405427468, |
| "loss": 5.908866577148437, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.012, |
| "grad_norm": 0.4188089668750763, |
| "learning_rate": 0.0005998799051502414, |
| "loss": 5.712981567382813, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.014, |
| "grad_norm": 0.43575364351272583, |
| "learning_rate": 0.0005998204596593927, |
| "loss": 5.54151123046875, |
| "step": 700 |
| }, |
| { |
| "epoch": 0.016, |
| "grad_norm": 0.3631345331668854, |
| "learning_rate": 0.0005997491064311517, |
| "loss": 5.377118530273438, |
| "step": 800 |
| }, |
| { |
| "epoch": 0.018, |
| "grad_norm": 0.4753637909889221, |
| "learning_rate": 0.000599665848299399, |
| "loss": 5.2575732421875, |
| "step": 900 |
| }, |
| { |
| "epoch": 0.02, |
| "grad_norm": 0.3885057270526886, |
| "learning_rate": 0.0005995706885708337, |
| "loss": 5.168424072265625, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.02, |
| "eval_accuracy": 0.21422370900467913, |
| "eval_loss": 5.0658464431762695, |
| "eval_runtime": 5.7327, |
| "eval_samples_per_second": 338.583, |
| "eval_steps_per_second": 21.281, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.022, |
| "grad_norm": 0.35898318886756897, |
| "learning_rate": 0.0005994636310248412, |
| "loss": 5.058907165527343, |
| "step": 1100 |
| }, |
| { |
| "epoch": 0.024, |
| "grad_norm": 0.36867955327033997, |
| "learning_rate": 0.000599344679913343, |
| "loss": 4.899845275878906, |
| "step": 1200 |
| }, |
| { |
| "epoch": 0.026, |
| "grad_norm": 0.3753994405269623, |
| "learning_rate": 0.0005992138399606286, |
| "loss": 4.828855285644531, |
| "step": 1300 |
| }, |
| { |
| "epoch": 0.028, |
| "grad_norm": 0.3426174521446228, |
| "learning_rate": 0.0005990711163631674, |
| "loss": 4.73110595703125, |
| "step": 1400 |
| }, |
| { |
| "epoch": 0.03, |
| "grad_norm": 0.34207016229629517, |
| "learning_rate": 0.000598916514789402, |
| "loss": 4.640643005371094, |
| "step": 1500 |
| }, |
| { |
| "epoch": 0.032, |
| "grad_norm": 0.40568745136260986, |
| "learning_rate": 0.0005987500413795236, |
| "loss": 4.591531677246094, |
| "step": 1600 |
| }, |
| { |
| "epoch": 0.034, |
| "grad_norm": 0.31946808099746704, |
| "learning_rate": 0.0005985717027452278, |
| "loss": 4.533291015625, |
| "step": 1700 |
| }, |
| { |
| "epoch": 0.036, |
| "grad_norm": 0.30501991510391235, |
| "learning_rate": 0.0005983815059694525, |
| "loss": 4.492686157226562, |
| "step": 1800 |
| }, |
| { |
| "epoch": 0.038, |
| "grad_norm": 0.33989349007606506, |
| "learning_rate": 0.0005981794586060958, |
| "loss": 4.41447509765625, |
| "step": 1900 |
| }, |
| { |
| "epoch": 0.04, |
| "grad_norm": 0.3303995430469513, |
| "learning_rate": 0.0005979655686797167, |
| "loss": 4.396521911621094, |
| "step": 2000 |
| }, |
| { |
| "epoch": 0.04, |
| "eval_accuracy": 0.2874403514237522, |
| "eval_loss": 4.30245304107666, |
| "eval_runtime": 5.7068, |
| "eval_samples_per_second": 340.121, |
| "eval_steps_per_second": 21.378, |
| "step": 2000 |
| }, |
| { |
| "epoch": 0.042, |
| "grad_norm": 0.46283307671546936, |
| "learning_rate": 0.0005977398446852161, |
| "loss": 4.379761352539062, |
| "step": 2100 |
| }, |
| { |
| "epoch": 0.044, |
| "grad_norm": 0.3387432396411896, |
| "learning_rate": 0.0005975022955874992, |
| "loss": 4.3154080200195315, |
| "step": 2200 |
| }, |
| { |
| "epoch": 0.046, |
| "grad_norm": 0.2719409465789795, |
| "learning_rate": 0.0005972529308211201, |
| "loss": 4.275622863769531, |
| "step": 2300 |
| }, |
| { |
| "epoch": 0.048, |
| "grad_norm": 0.2836616039276123, |
| "learning_rate": 0.0005969917602899061, |
| "loss": 4.258504333496094, |
| "step": 2400 |
| }, |
| { |
| "epoch": 0.05, |
| "grad_norm": 0.32210105657577515, |
| "learning_rate": 0.0005967187943665656, |
| "loss": 4.207368469238281, |
| "step": 2500 |
| }, |
| { |
| "epoch": 0.052, |
| "grad_norm": 0.30050167441368103, |
| "learning_rate": 0.0005964340438922749, |
| "loss": 4.2094793701171875, |
| "step": 2600 |
| }, |
| { |
| "epoch": 0.054, |
| "grad_norm": 0.3175508379936218, |
| "learning_rate": 0.0005961375201762486, |
| "loss": 4.171758117675782, |
| "step": 2700 |
| }, |
| { |
| "epoch": 0.056, |
| "grad_norm": 0.285259872674942, |
| "learning_rate": 0.0005958292349952897, |
| "loss": 4.1502682495117185, |
| "step": 2800 |
| }, |
| { |
| "epoch": 0.058, |
| "grad_norm": 0.2848680913448334, |
| "learning_rate": 0.0005955092005933225, |
| "loss": 4.128558959960937, |
| "step": 2900 |
| }, |
| { |
| "epoch": 0.06, |
| "grad_norm": 0.2977091073989868, |
| "learning_rate": 0.000595177429680906, |
| "loss": 4.134291076660157, |
| "step": 3000 |
| }, |
| { |
| "epoch": 0.06, |
| "eval_accuracy": 0.3091028793639367, |
| "eval_loss": 4.053634166717529, |
| "eval_runtime": 6.2016, |
| "eval_samples_per_second": 312.982, |
| "eval_steps_per_second": 19.672, |
| "step": 3000 |
| }, |
| { |
| "epoch": 0.062, |
| "grad_norm": 0.2866455018520355, |
| "learning_rate": 0.0005948339354347288, |
| "loss": 4.118316040039063, |
| "step": 3100 |
| }, |
| { |
| "epoch": 0.064, |
| "grad_norm": 0.2707529664039612, |
| "learning_rate": 0.0005944787314970865, |
| "loss": 4.104917602539063, |
| "step": 3200 |
| }, |
| { |
| "epoch": 0.066, |
| "grad_norm": 0.3138796091079712, |
| "learning_rate": 0.0005941118319753391, |
| "loss": 4.084369506835937, |
| "step": 3300 |
| }, |
| { |
| "epoch": 0.068, |
| "grad_norm": 0.2631393373012543, |
| "learning_rate": 0.0005937332514413514, |
| "loss": 4.044774475097657, |
| "step": 3400 |
| }, |
| { |
| "epoch": 0.07, |
| "grad_norm": 0.27509307861328125, |
| "learning_rate": 0.0005933430049309136, |
| "loss": 4.025779418945312, |
| "step": 3500 |
| }, |
| { |
| "epoch": 0.072, |
| "grad_norm": 0.33931243419647217, |
| "learning_rate": 0.0005929411079431449, |
| "loss": 4.004125366210937, |
| "step": 3600 |
| }, |
| { |
| "epoch": 0.074, |
| "grad_norm": 0.2669001519680023, |
| "learning_rate": 0.000592527576439877, |
| "loss": 4.012510070800781, |
| "step": 3700 |
| }, |
| { |
| "epoch": 0.076, |
| "grad_norm": 0.2727104425430298, |
| "learning_rate": 0.000592102426845021, |
| "loss": 3.9971853637695314, |
| "step": 3800 |
| }, |
| { |
| "epoch": 0.078, |
| "grad_norm": 0.25884559750556946, |
| "learning_rate": 0.0005916656760439145, |
| "loss": 3.9807003784179686, |
| "step": 3900 |
| }, |
| { |
| "epoch": 0.08, |
| "grad_norm": 0.3023005425930023, |
| "learning_rate": 0.0005912173413826517, |
| "loss": 3.9553176879882814, |
| "step": 4000 |
| }, |
| { |
| "epoch": 0.08, |
| "eval_accuracy": 0.32060359872601835, |
| "eval_loss": 3.923754930496216, |
| "eval_runtime": 6.2736, |
| "eval_samples_per_second": 309.393, |
| "eval_steps_per_second": 19.447, |
| "step": 4000 |
| }, |
| { |
| "epoch": 0.082, |
| "grad_norm": 0.2674967348575592, |
| "learning_rate": 0.0005907574406673935, |
| "loss": 3.9582391357421876, |
| "step": 4100 |
| }, |
| { |
| "epoch": 0.084, |
| "grad_norm": 0.25437116622924805, |
| "learning_rate": 0.0005902859921636611, |
| "loss": 3.942552185058594, |
| "step": 4200 |
| }, |
| { |
| "epoch": 0.086, |
| "grad_norm": 0.25480520725250244, |
| "learning_rate": 0.0005898030145956102, |
| "loss": 3.9550421142578127, |
| "step": 4300 |
| }, |
| { |
| "epoch": 0.088, |
| "grad_norm": 0.25695815682411194, |
| "learning_rate": 0.0005893085271452874, |
| "loss": 3.9156878662109373, |
| "step": 4400 |
| }, |
| { |
| "epoch": 0.09, |
| "grad_norm": 0.25535979866981506, |
| "learning_rate": 0.0005888025494518687, |
| "loss": 3.9294302368164065, |
| "step": 4500 |
| }, |
| { |
| "epoch": 0.092, |
| "grad_norm": 0.24557402729988098, |
| "learning_rate": 0.0005882851016108786, |
| "loss": 3.927276306152344, |
| "step": 4600 |
| }, |
| { |
| "epoch": 0.094, |
| "grad_norm": 0.26716557145118713, |
| "learning_rate": 0.0005877562041733932, |
| "loss": 3.904244384765625, |
| "step": 4700 |
| }, |
| { |
| "epoch": 0.096, |
| "grad_norm": 0.2525787651538849, |
| "learning_rate": 0.0005872158781452231, |
| "loss": 3.8892666625976564, |
| "step": 4800 |
| }, |
| { |
| "epoch": 0.098, |
| "grad_norm": 0.3125140964984894, |
| "learning_rate": 0.0005866641449860792, |
| "loss": 3.887486572265625, |
| "step": 4900 |
| }, |
| { |
| "epoch": 0.1, |
| "grad_norm": 0.2451365739107132, |
| "learning_rate": 0.0005861010266087211, |
| "loss": 3.875788269042969, |
| "step": 5000 |
| }, |
| { |
| "epoch": 0.1, |
| "eval_accuracy": 0.32875099183244255, |
| "eval_loss": 3.831427812576294, |
| "eval_runtime": 5.9329, |
| "eval_samples_per_second": 327.159, |
| "eval_steps_per_second": 20.563, |
| "step": 5000 |
| }, |
| { |
| "epoch": 0.102, |
| "grad_norm": 0.28080520033836365, |
| "learning_rate": 0.0005855265453780858, |
| "loss": 3.8711001586914064, |
| "step": 5100 |
| }, |
| { |
| "epoch": 0.104, |
| "grad_norm": 0.25133365392684937, |
| "learning_rate": 0.0005849407241104003, |
| "loss": 3.8725830078125, |
| "step": 5200 |
| }, |
| { |
| "epoch": 0.106, |
| "grad_norm": 0.27164554595947266, |
| "learning_rate": 0.000584343586072275, |
| "loss": 3.847073974609375, |
| "step": 5300 |
| }, |
| { |
| "epoch": 0.108, |
| "grad_norm": 0.2697848081588745, |
| "learning_rate": 0.0005837351549797795, |
| "loss": 3.79802978515625, |
| "step": 5400 |
| }, |
| { |
| "epoch": 0.11, |
| "grad_norm": 0.27253684401512146, |
| "learning_rate": 0.0005831154549975012, |
| "loss": 3.812421875, |
| "step": 5500 |
| }, |
| { |
| "epoch": 0.112, |
| "grad_norm": 0.27660876512527466, |
| "learning_rate": 0.0005824845107375853, |
| "loss": 3.7727542114257813, |
| "step": 5600 |
| }, |
| { |
| "epoch": 0.114, |
| "grad_norm": 0.2439262717962265, |
| "learning_rate": 0.0005818423472587571, |
| "loss": 3.7664410400390627, |
| "step": 5700 |
| }, |
| { |
| "epoch": 0.116, |
| "grad_norm": 0.24943126738071442, |
| "learning_rate": 0.0005811889900653269, |
| "loss": 3.7883651733398436, |
| "step": 5800 |
| }, |
| { |
| "epoch": 0.118, |
| "grad_norm": 0.2556191682815552, |
| "learning_rate": 0.0005805244651061771, |
| "loss": 3.7454019165039063, |
| "step": 5900 |
| }, |
| { |
| "epoch": 0.12, |
| "grad_norm": 0.27899155020713806, |
| "learning_rate": 0.0005798487987737322, |
| "loss": 3.767793884277344, |
| "step": 6000 |
| }, |
| { |
| "epoch": 0.12, |
| "eval_accuracy": 0.33610391076885543, |
| "eval_loss": 3.7605059146881104, |
| "eval_runtime": 5.697, |
| "eval_samples_per_second": 340.705, |
| "eval_steps_per_second": 21.415, |
| "step": 6000 |
| }, |
| { |
| "epoch": 0.122, |
| "grad_norm": 0.25070950388908386, |
| "learning_rate": 0.0005791620179029091, |
| "loss": 3.7485830688476565, |
| "step": 6100 |
| }, |
| { |
| "epoch": 0.124, |
| "grad_norm": 0.23322179913520813, |
| "learning_rate": 0.0005784641497700528, |
| "loss": 3.7345367431640626, |
| "step": 6200 |
| }, |
| { |
| "epoch": 0.126, |
| "grad_norm": 0.27075648307800293, |
| "learning_rate": 0.0005777552220918525, |
| "loss": 3.747908935546875, |
| "step": 6300 |
| }, |
| { |
| "epoch": 0.128, |
| "grad_norm": 0.22065384685993195, |
| "learning_rate": 0.0005770352630242405, |
| "loss": 3.7215350341796873, |
| "step": 6400 |
| }, |
| { |
| "epoch": 0.13, |
| "grad_norm": 0.2387268990278244, |
| "learning_rate": 0.0005763043011612747, |
| "loss": 3.7065789794921873, |
| "step": 6500 |
| }, |
| { |
| "epoch": 0.132, |
| "grad_norm": 0.2440839409828186, |
| "learning_rate": 0.0005755623655340019, |
| "loss": 3.7222140502929686, |
| "step": 6600 |
| }, |
| { |
| "epoch": 0.134, |
| "grad_norm": 0.7614469528198242, |
| "learning_rate": 0.000574809485609306, |
| "loss": 3.7069766235351564, |
| "step": 6700 |
| }, |
| { |
| "epoch": 0.136, |
| "grad_norm": 0.25670167803764343, |
| "learning_rate": 0.0005740456912887371, |
| "loss": 3.7104458618164062, |
| "step": 6800 |
| }, |
| { |
| "epoch": 0.138, |
| "grad_norm": 0.25214189291000366, |
| "learning_rate": 0.0005732710129073234, |
| "loss": 3.7040945434570314, |
| "step": 6900 |
| }, |
| { |
| "epoch": 0.14, |
| "grad_norm": 0.2262343317270279, |
| "learning_rate": 0.0005724854812323674, |
| "loss": 3.686309814453125, |
| "step": 7000 |
| }, |
| { |
| "epoch": 0.14, |
| "eval_accuracy": 0.3409998074307532, |
| "eval_loss": 3.705655097961426, |
| "eval_runtime": 6.3038, |
| "eval_samples_per_second": 307.912, |
| "eval_steps_per_second": 19.354, |
| "step": 7000 |
| }, |
| { |
| "epoch": 0.142, |
| "grad_norm": 0.2256837785243988, |
| "learning_rate": 0.0005716891274622232, |
| "loss": 3.6851046752929686, |
| "step": 7100 |
| }, |
| { |
| "epoch": 0.144, |
| "grad_norm": 0.22516745328903198, |
| "learning_rate": 0.0005708819832250578, |
| "loss": 3.6776641845703124, |
| "step": 7200 |
| }, |
| { |
| "epoch": 0.146, |
| "grad_norm": 0.28322234749794006, |
| "learning_rate": 0.000570064080577595, |
| "loss": 3.6858819580078124, |
| "step": 7300 |
| }, |
| { |
| "epoch": 0.148, |
| "grad_norm": 0.2589821219444275, |
| "learning_rate": 0.0005692354520038415, |
| "loss": 3.683369140625, |
| "step": 7400 |
| }, |
| { |
| "epoch": 0.15, |
| "grad_norm": 0.25471949577331543, |
| "learning_rate": 0.0005683961304137982, |
| "loss": 3.6476519775390623, |
| "step": 7500 |
| }, |
| { |
| "epoch": 0.152, |
| "grad_norm": 0.24686427414417267, |
| "learning_rate": 0.0005675461491421516, |
| "loss": 3.66244873046875, |
| "step": 7600 |
| }, |
| { |
| "epoch": 0.154, |
| "grad_norm": 0.23603662848472595, |
| "learning_rate": 0.0005666855419469506, |
| "loss": 3.6586270141601562, |
| "step": 7700 |
| }, |
| { |
| "epoch": 0.156, |
| "grad_norm": 0.23954837024211884, |
| "learning_rate": 0.0005658143430082659, |
| "loss": 3.6526480102539063, |
| "step": 7800 |
| }, |
| { |
| "epoch": 0.158, |
| "grad_norm": 0.24939873814582825, |
| "learning_rate": 0.0005649325869268321, |
| "loss": 3.6602520751953125, |
| "step": 7900 |
| }, |
| { |
| "epoch": 0.16, |
| "grad_norm": 0.22606751322746277, |
| "learning_rate": 0.0005640403087226739, |
| "loss": 3.6500091552734375, |
| "step": 8000 |
| }, |
| { |
| "epoch": 0.16, |
| "eval_accuracy": 0.34401134847875336, |
| "eval_loss": 3.6704835891723633, |
| "eval_runtime": 5.9263, |
| "eval_samples_per_second": 327.521, |
| "eval_steps_per_second": 20.586, |
| "step": 8000 |
| }, |
| { |
| "epoch": 0.162, |
| "grad_norm": 0.23801666498184204, |
| "learning_rate": 0.0005631375438337147, |
| "loss": 3.632862548828125, |
| "step": 8100 |
| }, |
| { |
| "epoch": 0.164, |
| "grad_norm": 0.23611237108707428, |
| "learning_rate": 0.0005622243281143698, |
| "loss": 3.6399630737304687, |
| "step": 8200 |
| }, |
| { |
| "epoch": 0.166, |
| "grad_norm": 0.23898784816265106, |
| "learning_rate": 0.0005613006978341218, |
| "loss": 3.6880291748046874, |
| "step": 8300 |
| }, |
| { |
| "epoch": 0.168, |
| "grad_norm": 0.29460519552230835, |
| "learning_rate": 0.0005603666896760804, |
| "loss": 3.7291055297851563, |
| "step": 8400 |
| }, |
| { |
| "epoch": 0.17, |
| "grad_norm": 0.2292584925889969, |
| "learning_rate": 0.0005594223407355256, |
| "loss": 3.7337985229492188, |
| "step": 8500 |
| }, |
| { |
| "epoch": 0.172, |
| "grad_norm": 0.2524070739746094, |
| "learning_rate": 0.0005584676885184341, |
| "loss": 3.7386798095703124, |
| "step": 8600 |
| }, |
| { |
| "epoch": 0.174, |
| "grad_norm": 0.25900545716285706, |
| "learning_rate": 0.0005575027709399899, |
| "loss": 3.7031732177734376, |
| "step": 8700 |
| }, |
| { |
| "epoch": 0.176, |
| "grad_norm": 0.22311732172966003, |
| "learning_rate": 0.0005565276263230785, |
| "loss": 3.71237060546875, |
| "step": 8800 |
| }, |
| { |
| "epoch": 0.178, |
| "grad_norm": 0.23083606362342834, |
| "learning_rate": 0.0005555422933967647, |
| "loss": 3.6988143920898438, |
| "step": 8900 |
| }, |
| { |
| "epoch": 0.18, |
| "grad_norm": 0.23646661639213562, |
| "learning_rate": 0.0005545468112947544, |
| "loss": 3.692724609375, |
| "step": 9000 |
| }, |
| { |
| "epoch": 0.18, |
| "eval_accuracy": 0.3474917099443364, |
| "eval_loss": 3.6365442276000977, |
| "eval_runtime": 5.728, |
| "eval_samples_per_second": 338.859, |
| "eval_steps_per_second": 21.299, |
| "step": 9000 |
| }, |
| { |
| "epoch": 0.182, |
| "grad_norm": 0.2665349841117859, |
| "learning_rate": 0.0005535412195538409, |
| "loss": 3.683785400390625, |
| "step": 9100 |
| }, |
| { |
| "epoch": 0.184, |
| "grad_norm": 0.22817476093769073, |
| "learning_rate": 0.0005525255581123336, |
| "loss": 3.6991543579101562, |
| "step": 9200 |
| }, |
| { |
| "epoch": 0.186, |
| "grad_norm": 0.2592155635356903, |
| "learning_rate": 0.000551499867308473, |
| "loss": 3.6590890502929687, |
| "step": 9300 |
| }, |
| { |
| "epoch": 0.188, |
| "grad_norm": 0.24686357378959656, |
| "learning_rate": 0.0005504641878788275, |
| "loss": 3.6509600830078126, |
| "step": 9400 |
| }, |
| { |
| "epoch": 0.19, |
| "grad_norm": 0.22637774050235748, |
| "learning_rate": 0.0005494185609566764, |
| "loss": 3.666563720703125, |
| "step": 9500 |
| }, |
| { |
| "epoch": 0.192, |
| "grad_norm": 0.21554476022720337, |
| "learning_rate": 0.0005483630280703757, |
| "loss": 3.6956219482421875, |
| "step": 9600 |
| }, |
| { |
| "epoch": 0.194, |
| "grad_norm": 0.2128387987613678, |
| "learning_rate": 0.0005472976311417091, |
| "loss": 3.661249084472656, |
| "step": 9700 |
| }, |
| { |
| "epoch": 0.196, |
| "grad_norm": 0.21794714033603668, |
| "learning_rate": 0.0005462224124842225, |
| "loss": 3.6454119873046875, |
| "step": 9800 |
| }, |
| { |
| "epoch": 0.198, |
| "grad_norm": 0.23011332750320435, |
| "learning_rate": 0.0005451374148015441, |
| "loss": 3.6472012329101564, |
| "step": 9900 |
| }, |
| { |
| "epoch": 0.2, |
| "grad_norm": 0.21280302107334137, |
| "learning_rate": 0.0005440426811856883, |
| "loss": 3.6411611938476565, |
| "step": 10000 |
| }, |
| { |
| "epoch": 0.2, |
| "eval_accuracy": 0.3510436547424966, |
| "eval_loss": 3.6030726432800293, |
| "eval_runtime": 5.6604, |
| "eval_samples_per_second": 342.906, |
| "eval_steps_per_second": 21.553, |
| "step": 10000 |
| }, |
| { |
| "epoch": 0.202, |
| "grad_norm": 0.24885770678520203, |
| "learning_rate": 0.0005429382551153432, |
| "loss": 3.639742431640625, |
| "step": 10100 |
| }, |
| { |
| "epoch": 0.204, |
| "grad_norm": 0.22552932798862457, |
| "learning_rate": 0.0005418241804541457, |
| "loss": 3.6341900634765625, |
| "step": 10200 |
| }, |
| { |
| "epoch": 0.206, |
| "grad_norm": 0.24933825433254242, |
| "learning_rate": 0.0005407005014489372, |
| "loss": 3.6378778076171874, |
| "step": 10300 |
| }, |
| { |
| "epoch": 0.208, |
| "grad_norm": 0.24248844385147095, |
| "learning_rate": 0.0005395672627280085, |
| "loss": 3.630330810546875, |
| "step": 10400 |
| }, |
| { |
| "epoch": 0.21, |
| "grad_norm": 0.20998498797416687, |
| "learning_rate": 0.0005384245092993256, |
| "loss": 3.6162420654296876, |
| "step": 10500 |
| }, |
| { |
| "epoch": 0.212, |
| "grad_norm": 0.23281893134117126, |
| "learning_rate": 0.0005372722865487427, |
| "loss": 3.6330096435546877, |
| "step": 10600 |
| }, |
| { |
| "epoch": 0.214, |
| "grad_norm": 0.2150452584028244, |
| "learning_rate": 0.0005361106402382001, |
| "loss": 3.625997314453125, |
| "step": 10700 |
| }, |
| { |
| "epoch": 0.216, |
| "grad_norm": 0.2066139578819275, |
| "learning_rate": 0.0005349396165039064, |
| "loss": 3.594141540527344, |
| "step": 10800 |
| }, |
| { |
| "epoch": 0.218, |
| "grad_norm": 0.23046144843101501, |
| "learning_rate": 0.0005337592618545055, |
| "loss": 3.598072204589844, |
| "step": 10900 |
| }, |
| { |
| "epoch": 0.22, |
| "grad_norm": 0.2208685725927353, |
| "learning_rate": 0.0005325696231692306, |
| "loss": 3.6059771728515626, |
| "step": 11000 |
| }, |
| { |
| "epoch": 0.22, |
| "eval_accuracy": 0.3537567638687666, |
| "eval_loss": 3.57378888130188, |
| "eval_runtime": 6.1179, |
| "eval_samples_per_second": 317.265, |
| "eval_steps_per_second": 19.941, |
| "step": 11000 |
| }, |
| { |
| "epoch": 0.222, |
| "grad_norm": 0.22270554304122925, |
| "learning_rate": 0.0005313707476960418, |
| "loss": 3.616676940917969, |
| "step": 11100 |
| }, |
| { |
| "epoch": 0.224, |
| "grad_norm": 0.20033609867095947, |
| "learning_rate": 0.0005301626830497491, |
| "loss": 3.6163037109375, |
| "step": 11200 |
| }, |
| { |
| "epoch": 0.226, |
| "grad_norm": 0.2057274878025055, |
| "learning_rate": 0.0005289454772101221, |
| "loss": 3.593800354003906, |
| "step": 11300 |
| }, |
| { |
| "epoch": 0.228, |
| "grad_norm": 0.20418649911880493, |
| "learning_rate": 0.000527719178519984, |
| "loss": 3.606725769042969, |
| "step": 11400 |
| }, |
| { |
| "epoch": 0.23, |
| "grad_norm": 0.20258976519107819, |
| "learning_rate": 0.000526483835683292, |
| "loss": 3.594194030761719, |
| "step": 11500 |
| }, |
| { |
| "epoch": 0.232, |
| "grad_norm": 0.2198181301355362, |
| "learning_rate": 0.0005252394977632023, |
| "loss": 3.596621398925781, |
| "step": 11600 |
| }, |
| { |
| "epoch": 0.234, |
| "grad_norm": 0.2138666957616806, |
| "learning_rate": 0.000523986214180122, |
| "loss": 3.5838372802734373, |
| "step": 11700 |
| }, |
| { |
| "epoch": 0.236, |
| "grad_norm": 0.22914420068264008, |
| "learning_rate": 0.0005227240347097463, |
| "loss": 3.559365539550781, |
| "step": 11800 |
| }, |
| { |
| "epoch": 0.238, |
| "grad_norm": 0.23261770606040955, |
| "learning_rate": 0.0005214530094810812, |
| "loss": 3.5759823608398436, |
| "step": 11900 |
| }, |
| { |
| "epoch": 0.24, |
| "grad_norm": 0.21132150292396545, |
| "learning_rate": 0.0005201731889744533, |
| "loss": 3.56010009765625, |
| "step": 12000 |
| }, |
| { |
| "epoch": 0.24, |
| "eval_accuracy": 0.35579436830733646, |
| "eval_loss": 3.5506765842437744, |
| "eval_runtime": 5.6754, |
| "eval_samples_per_second": 342.004, |
| "eval_steps_per_second": 21.496, |
| "step": 12000 |
| }, |
| { |
| "epoch": 0.242, |
| "grad_norm": 0.19794468581676483, |
| "learning_rate": 0.0005188846240195041, |
| "loss": 3.563157958984375, |
| "step": 12100 |
| }, |
| { |
| "epoch": 0.244, |
| "grad_norm": 0.22670620679855347, |
| "learning_rate": 0.0005175873657931715, |
| "loss": 3.573887939453125, |
| "step": 12200 |
| }, |
| { |
| "epoch": 0.246, |
| "grad_norm": 0.2241670936346054, |
| "learning_rate": 0.0005162814658176576, |
| "loss": 3.614027099609375, |
| "step": 12300 |
| }, |
| { |
| "epoch": 0.248, |
| "grad_norm": 0.23707276582717896, |
| "learning_rate": 0.000514966975958382, |
| "loss": 3.628121643066406, |
| "step": 12400 |
| }, |
| { |
| "epoch": 0.25, |
| "grad_norm": 0.22935494780540466, |
| "learning_rate": 0.0005136439484219223, |
| "loss": 3.6195343017578123, |
| "step": 12500 |
| }, |
| { |
| "epoch": 0.252, |
| "grad_norm": 0.21692432463169098, |
| "learning_rate": 0.0005123124357539403, |
| "loss": 3.62494140625, |
| "step": 12600 |
| }, |
| { |
| "epoch": 0.254, |
| "grad_norm": 0.1959449201822281, |
| "learning_rate": 0.0005109724908370951, |
| "loss": 3.619110412597656, |
| "step": 12700 |
| }, |
| { |
| "epoch": 0.256, |
| "grad_norm": 0.22109989821910858, |
| "learning_rate": 0.000509624166888943, |
| "loss": 3.609862060546875, |
| "step": 12800 |
| }, |
| { |
| "epoch": 0.258, |
| "grad_norm": 0.20637010037899017, |
| "learning_rate": 0.0005082675174598238, |
| "loss": 3.6038763427734377, |
| "step": 12900 |
| }, |
| { |
| "epoch": 0.26, |
| "grad_norm": 0.21297694742679596, |
| "learning_rate": 0.000506902596430734, |
| "loss": 3.627516174316406, |
| "step": 13000 |
| }, |
| { |
| "epoch": 0.26, |
| "eval_accuracy": 0.35723510890244603, |
| "eval_loss": 3.5337636470794678, |
| "eval_runtime": 5.6987, |
| "eval_samples_per_second": 340.603, |
| "eval_steps_per_second": 21.408, |
| "step": 13000 |
| }, |
| { |
| "epoch": 0.262, |
| "grad_norm": 0.2433704286813736, |
| "learning_rate": 0.0005055294580111867, |
| "loss": 3.600766296386719, |
| "step": 13100 |
| }, |
| { |
| "epoch": 0.264, |
| "grad_norm": 0.20984166860580444, |
| "learning_rate": 0.0005041481567370593, |
| "loss": 3.57820556640625, |
| "step": 13200 |
| }, |
| { |
| "epoch": 0.266, |
| "grad_norm": 0.22343853116035461, |
| "learning_rate": 0.0005027587474684261, |
| "loss": 3.603807373046875, |
| "step": 13300 |
| }, |
| { |
| "epoch": 0.268, |
| "grad_norm": 0.2158922404050827, |
| "learning_rate": 0.0005013612853873809, |
| "loss": 3.5946688842773438, |
| "step": 13400 |
| }, |
| { |
| "epoch": 0.27, |
| "grad_norm": 0.21097783744335175, |
| "learning_rate": 0.0004999558259958449, |
| "loss": 3.589884033203125, |
| "step": 13500 |
| }, |
| { |
| "epoch": 0.272, |
| "grad_norm": 0.28342482447624207, |
| "learning_rate": 0.0004985424251133622, |
| "loss": 3.5845135498046874, |
| "step": 13600 |
| }, |
| { |
| "epoch": 0.274, |
| "grad_norm": 0.22716180980205536, |
| "learning_rate": 0.0004971211388748826, |
| "loss": 3.601893310546875, |
| "step": 13700 |
| }, |
| { |
| "epoch": 0.276, |
| "grad_norm": 0.2315467894077301, |
| "learning_rate": 0.000495692023728533, |
| "loss": 3.583456726074219, |
| "step": 13800 |
| }, |
| { |
| "epoch": 0.278, |
| "grad_norm": 0.20143155753612518, |
| "learning_rate": 0.0004942551364333746, |
| "loss": 3.5840383911132814, |
| "step": 13900 |
| }, |
| { |
| "epoch": 0.28, |
| "grad_norm": 0.2158096581697464, |
| "learning_rate": 0.0004928105340571493, |
| "loss": 3.5770263671875, |
| "step": 14000 |
| }, |
| { |
| "epoch": 0.28, |
| "eval_accuracy": 0.3599905630986912, |
| "eval_loss": 3.508211851119995, |
| "eval_runtime": 9.9993, |
| "eval_samples_per_second": 194.114, |
| "eval_steps_per_second": 12.201, |
| "step": 14000 |
| }, |
| { |
| "epoch": 0.282, |
| "grad_norm": 0.22074639797210693, |
| "learning_rate": 0.0004913582739740128, |
| "loss": 3.578910217285156, |
| "step": 14100 |
| }, |
| { |
| "epoch": 0.284, |
| "grad_norm": 0.20357950031757355, |
| "learning_rate": 0.000489898413862256, |
| "loss": 3.5391220092773437, |
| "step": 14200 |
| }, |
| { |
| "epoch": 0.286, |
| "grad_norm": 0.2000010460615158, |
| "learning_rate": 0.0004884310117020142, |
| "loss": 3.558287353515625, |
| "step": 14300 |
| }, |
| { |
| "epoch": 0.288, |
| "grad_norm": 0.20651493966579437, |
| "learning_rate": 0.00048695612577296456, |
| "loss": 3.5417572021484376, |
| "step": 14400 |
| }, |
| { |
| "epoch": 0.29, |
| "grad_norm": 0.18496066331863403, |
| "learning_rate": 0.0004854738146520113, |
| "loss": 3.5497479248046875, |
| "step": 14500 |
| }, |
| { |
| "epoch": 0.292, |
| "grad_norm": 0.23755651712417603, |
| "learning_rate": 0.000483984137210959, |
| "loss": 3.5615032958984374, |
| "step": 14600 |
| }, |
| { |
| "epoch": 0.294, |
| "grad_norm": 0.20592571794986725, |
| "learning_rate": 0.0004824871526141749, |
| "loss": 3.5748583984375, |
| "step": 14700 |
| }, |
| { |
| "epoch": 0.296, |
| "grad_norm": 0.2018384486436844, |
| "learning_rate": 0.00048098292031623904, |
| "loss": 3.575079040527344, |
| "step": 14800 |
| }, |
| { |
| "epoch": 0.298, |
| "grad_norm": 0.19982250034809113, |
| "learning_rate": 0.00047947150005958237, |
| "loss": 3.56544921875, |
| "step": 14900 |
| }, |
| { |
| "epoch": 0.3, |
| "grad_norm": 0.2055491805076599, |
| "learning_rate": 0.00047795295187211473, |
| "loss": 3.5550732421875, |
| "step": 15000 |
| }, |
| { |
| "epoch": 0.3, |
| "eval_accuracy": 0.3619394445335035, |
| "eval_loss": 3.490281105041504, |
| "eval_runtime": 5.758, |
| "eval_samples_per_second": 337.096, |
| "eval_steps_per_second": 21.188, |
| "step": 15000 |
| }, |
| { |
| "epoch": 0.302, |
| "grad_norm": 0.18866001069545746, |
| "learning_rate": 0.00047642733606484064, |
| "loss": 3.544368896484375, |
| "step": 15100 |
| }, |
| { |
| "epoch": 0.304, |
| "grad_norm": 0.19551938772201538, |
| "learning_rate": 0.00047489471322946335, |
| "loss": 3.5462738037109376, |
| "step": 15200 |
| }, |
| { |
| "epoch": 0.306, |
| "grad_norm": 0.20952892303466797, |
| "learning_rate": 0.00047335514423597913, |
| "loss": 3.547786865234375, |
| "step": 15300 |
| }, |
| { |
| "epoch": 0.308, |
| "grad_norm": 0.1951807737350464, |
| "learning_rate": 0.0004718086902302595, |
| "loss": 3.5607293701171874, |
| "step": 15400 |
| }, |
| { |
| "epoch": 0.31, |
| "grad_norm": 0.19585217535495758, |
| "learning_rate": 0.00047025541263162244, |
| "loss": 3.5599630737304686, |
| "step": 15500 |
| }, |
| { |
| "epoch": 0.312, |
| "grad_norm": 0.2213257998228073, |
| "learning_rate": 0.00046869537313039343, |
| "loss": 3.5485220336914063, |
| "step": 15600 |
| }, |
| { |
| "epoch": 0.314, |
| "grad_norm": 0.28433483839035034, |
| "learning_rate": 0.0004671286336854554, |
| "loss": 3.558450012207031, |
| "step": 15700 |
| }, |
| { |
| "epoch": 0.316, |
| "grad_norm": 0.23959818482398987, |
| "learning_rate": 0.00046555525652178736, |
| "loss": 3.5676364135742187, |
| "step": 15800 |
| }, |
| { |
| "epoch": 0.318, |
| "grad_norm": 0.20079217851161957, |
| "learning_rate": 0.0004639753041279938, |
| "loss": 3.552779541015625, |
| "step": 15900 |
| }, |
| { |
| "epoch": 0.32, |
| "grad_norm": 0.20217208564281464, |
| "learning_rate": 0.00046238883925382235, |
| "loss": 3.537972717285156, |
| "step": 16000 |
| }, |
| { |
| "epoch": 0.32, |
| "eval_accuracy": 0.363115024333292, |
| "eval_loss": 3.475611448287964, |
| "eval_runtime": 5.721, |
| "eval_samples_per_second": 339.274, |
| "eval_steps_per_second": 21.325, |
| "step": 16000 |
| }, |
| { |
| "epoch": 0.322, |
| "grad_norm": 0.22858326137065887, |
| "learning_rate": 0.00046079592490767184, |
| "loss": 3.5346307373046875, |
| "step": 16100 |
| }, |
| { |
| "epoch": 0.324, |
| "grad_norm": 0.19820822775363922, |
| "learning_rate": 0.0004591966243540899, |
| "loss": 3.559911804199219, |
| "step": 16200 |
| }, |
| { |
| "epoch": 0.326, |
| "grad_norm": 0.19681668281555176, |
| "learning_rate": 0.0004575910011112601, |
| "loss": 3.5361669921875, |
| "step": 16300 |
| }, |
| { |
| "epoch": 0.328, |
| "grad_norm": 0.19931091368198395, |
| "learning_rate": 0.0004559791189484793, |
| "loss": 3.5331732177734376, |
| "step": 16400 |
| }, |
| { |
| "epoch": 0.33, |
| "grad_norm": 0.20607522130012512, |
| "learning_rate": 0.0004543610418836253, |
| "loss": 3.5401296997070313, |
| "step": 16500 |
| }, |
| { |
| "epoch": 0.332, |
| "grad_norm": 0.21663250029087067, |
| "learning_rate": 0.0004527368341806138, |
| "loss": 3.530547790527344, |
| "step": 16600 |
| }, |
| { |
| "epoch": 0.334, |
| "grad_norm": 0.219991534948349, |
| "learning_rate": 0.00045110656034684634, |
| "loss": 3.535267639160156, |
| "step": 16700 |
| }, |
| { |
| "epoch": 0.336, |
| "grad_norm": 0.1956709921360016, |
| "learning_rate": 0.00044947028513064836, |
| "loss": 3.533216552734375, |
| "step": 16800 |
| }, |
| { |
| "epoch": 0.338, |
| "grad_norm": 0.20111192762851715, |
| "learning_rate": 0.0004478280735186978, |
| "loss": 3.5435382080078126, |
| "step": 16900 |
| }, |
| { |
| "epoch": 0.34, |
| "grad_norm": 0.1945837289094925, |
| "learning_rate": 0.0004461799907334432, |
| "loss": 3.509302978515625, |
| "step": 17000 |
| }, |
| { |
| "epoch": 0.34, |
| "eval_accuracy": 0.3652453846394267, |
| "eval_loss": 3.4567112922668457, |
| "eval_runtime": 5.9273, |
| "eval_samples_per_second": 327.467, |
| "eval_steps_per_second": 20.583, |
| "step": 17000 |
| }, |
| { |
| "epoch": 0.342, |
| "grad_norm": 0.22662605345249176, |
| "learning_rate": 0.0004445261022305145, |
| "loss": 3.5362484741210936, |
| "step": 17100 |
| }, |
| { |
| "epoch": 0.344, |
| "grad_norm": 0.19569633901119232, |
| "learning_rate": 0.00044286647369612284, |
| "loss": 3.4980740356445312, |
| "step": 17200 |
| }, |
| { |
| "epoch": 0.346, |
| "grad_norm": 0.1999066174030304, |
| "learning_rate": 0.0004412011710444514, |
| "loss": 3.5234307861328125, |
| "step": 17300 |
| }, |
| { |
| "epoch": 0.348, |
| "grad_norm": 0.209122896194458, |
| "learning_rate": 0.00043953026041503805, |
| "loss": 3.517156677246094, |
| "step": 17400 |
| }, |
| { |
| "epoch": 0.35, |
| "grad_norm": 0.2141777127981186, |
| "learning_rate": 0.0004378538081701487, |
| "loss": 3.5367440795898437, |
| "step": 17500 |
| }, |
| { |
| "epoch": 0.352, |
| "grad_norm": 0.20216333866119385, |
| "learning_rate": 0.0004361718808921406, |
| "loss": 3.5334283447265626, |
| "step": 17600 |
| }, |
| { |
| "epoch": 0.354, |
| "grad_norm": 0.24760980904102325, |
| "learning_rate": 0.0004344845453808194, |
| "loss": 3.516435546875, |
| "step": 17700 |
| }, |
| { |
| "epoch": 0.356, |
| "grad_norm": 0.19526250660419464, |
| "learning_rate": 0.0004327918686507849, |
| "loss": 3.526764221191406, |
| "step": 17800 |
| }, |
| { |
| "epoch": 0.358, |
| "grad_norm": 0.2374846190214157, |
| "learning_rate": 0.00043109391792877017, |
| "loss": 3.5132098388671875, |
| "step": 17900 |
| }, |
| { |
| "epoch": 0.36, |
| "grad_norm": 0.20901496708393097, |
| "learning_rate": 0.00042939076065097133, |
| "loss": 3.5150552368164063, |
| "step": 18000 |
| }, |
| { |
| "epoch": 0.36, |
| "eval_accuracy": 0.36633224143545756, |
| "eval_loss": 3.4449119567871094, |
| "eval_runtime": 6.3421, |
| "eval_samples_per_second": 306.051, |
| "eval_steps_per_second": 19.237, |
| "step": 18000 |
| }, |
| { |
| "epoch": 0.362, |
| "grad_norm": 0.2838000953197479, |
| "learning_rate": 0.0004276824644603691, |
| "loss": 3.5090277099609377, |
| "step": 18100 |
| }, |
| { |
| "epoch": 0.364, |
| "grad_norm": 0.6395978331565857, |
| "learning_rate": 0.0004259690972040426, |
| "loss": 3.5145654296875, |
| "step": 18200 |
| }, |
| { |
| "epoch": 0.366, |
| "grad_norm": 0.1982821524143219, |
| "learning_rate": 0.0004242507269304744, |
| "loss": 3.504261474609375, |
| "step": 18300 |
| }, |
| { |
| "epoch": 0.368, |
| "grad_norm": 0.22082549333572388, |
| "learning_rate": 0.0004225274218868482, |
| "loss": 3.496566162109375, |
| "step": 18400 |
| }, |
| { |
| "epoch": 0.37, |
| "grad_norm": 0.2385374903678894, |
| "learning_rate": 0.0004207992505163379, |
| "loss": 3.514676208496094, |
| "step": 18500 |
| }, |
| { |
| "epoch": 0.372, |
| "grad_norm": 0.18938113749027252, |
| "learning_rate": 0.00041906628145538987, |
| "loss": 3.5016729736328127, |
| "step": 18600 |
| }, |
| { |
| "epoch": 0.374, |
| "grad_norm": 0.17800572514533997, |
| "learning_rate": 0.0004173285835309965, |
| "loss": 3.471422119140625, |
| "step": 18700 |
| }, |
| { |
| "epoch": 0.376, |
| "grad_norm": 0.19377586245536804, |
| "learning_rate": 0.0004155862257579626, |
| "loss": 3.5025115966796876, |
| "step": 18800 |
| }, |
| { |
| "epoch": 0.378, |
| "grad_norm": 0.21612316370010376, |
| "learning_rate": 0.00041383927733616477, |
| "loss": 3.4965274047851564, |
| "step": 18900 |
| }, |
| { |
| "epoch": 0.38, |
| "grad_norm": 0.6052186489105225, |
| "learning_rate": 0.000412087807647803, |
| "loss": 3.4777236938476563, |
| "step": 19000 |
| }, |
| { |
| "epoch": 0.38, |
| "eval_accuracy": 0.3683073364850164, |
| "eval_loss": 3.4274120330810547, |
| "eval_runtime": 6.5604, |
| "eval_samples_per_second": 295.864, |
| "eval_steps_per_second": 18.596, |
| "step": 19000 |
| }, |
| { |
| "epoch": 0.382, |
| "grad_norm": 0.19751021265983582, |
| "learning_rate": 0.0004103318862546447, |
| "loss": 3.4927557373046874, |
| "step": 19100 |
| }, |
| { |
| "epoch": 0.384, |
| "grad_norm": 0.295538067817688, |
| "learning_rate": 0.0004085715828952621, |
| "loss": 3.4930191040039062, |
| "step": 19200 |
| }, |
| { |
| "epoch": 0.386, |
| "grad_norm": 0.18652473390102386, |
| "learning_rate": 0.00040680696748226293, |
| "loss": 3.4893539428710936, |
| "step": 19300 |
| }, |
| { |
| "epoch": 0.388, |
| "grad_norm": 0.19307976961135864, |
| "learning_rate": 0.00040503811009951325, |
| "loss": 3.484967041015625, |
| "step": 19400 |
| }, |
| { |
| "epoch": 0.39, |
| "grad_norm": 0.20981250703334808, |
| "learning_rate": 0.0004032650809993542, |
| "loss": 3.4780728149414064, |
| "step": 19500 |
| }, |
| { |
| "epoch": 0.392, |
| "grad_norm": 0.31385481357574463, |
| "learning_rate": 0.00040148795059981155, |
| "loss": 3.496181640625, |
| "step": 19600 |
| }, |
| { |
| "epoch": 0.394, |
| "grad_norm": 0.1942823827266693, |
| "learning_rate": 0.0003997067894817995, |
| "loss": 3.487859802246094, |
| "step": 19700 |
| }, |
| { |
| "epoch": 0.396, |
| "grad_norm": 0.1936560422182083, |
| "learning_rate": 0.0003979216683863172, |
| "loss": 3.479795227050781, |
| "step": 19800 |
| }, |
| { |
| "epoch": 0.398, |
| "grad_norm": 0.2430066615343094, |
| "learning_rate": 0.00039613265821163883, |
| "loss": 3.4654913330078125, |
| "step": 19900 |
| }, |
| { |
| "epoch": 0.4, |
| "grad_norm": 0.20538495481014252, |
| "learning_rate": 0.0003943398300104985, |
| "loss": 3.4708984375, |
| "step": 20000 |
| }, |
| { |
| "epoch": 0.4, |
| "eval_accuracy": 0.36907962990408844, |
| "eval_loss": 3.415916919708252, |
| "eval_runtime": 5.8841, |
| "eval_samples_per_second": 329.872, |
| "eval_steps_per_second": 20.734, |
| "step": 20000 |
| }, |
| { |
| "epoch": 0.402, |
| "grad_norm": 0.20392677187919617, |
| "learning_rate": 0.0003925432549872678, |
| "loss": 3.467202453613281, |
| "step": 20100 |
| }, |
| { |
| "epoch": 0.404, |
| "grad_norm": 0.21144644916057587, |
| "learning_rate": 0.0003907430044951278, |
| "loss": 3.463043212890625, |
| "step": 20200 |
| }, |
| { |
| "epoch": 0.406, |
| "grad_norm": 0.1932135373353958, |
| "learning_rate": 0.00038893915003323544, |
| "loss": 3.484923095703125, |
| "step": 20300 |
| }, |
| { |
| "epoch": 0.408, |
| "grad_norm": 0.20958249270915985, |
| "learning_rate": 0.00038713176324388387, |
| "loss": 3.483151550292969, |
| "step": 20400 |
| }, |
| { |
| "epoch": 0.41, |
| "grad_norm": 0.4303552210330963, |
| "learning_rate": 0.00038532091590965674, |
| "loss": 3.477305603027344, |
| "step": 20500 |
| }, |
| { |
| "epoch": 0.412, |
| "grad_norm": 0.20084762573242188, |
| "learning_rate": 0.0003835066799505776, |
| "loss": 3.47593994140625, |
| "step": 20600 |
| }, |
| { |
| "epoch": 0.414, |
| "grad_norm": 0.18663835525512695, |
| "learning_rate": 0.0003816891274212534, |
| "loss": 3.465316162109375, |
| "step": 20700 |
| }, |
| { |
| "epoch": 0.416, |
| "grad_norm": 0.1979358196258545, |
| "learning_rate": 0.00037986833050801256, |
| "loss": 3.465248718261719, |
| "step": 20800 |
| }, |
| { |
| "epoch": 0.418, |
| "grad_norm": 0.20419807732105255, |
| "learning_rate": 0.0003780443615260386, |
| "loss": 3.4529428100585937, |
| "step": 20900 |
| }, |
| { |
| "epoch": 0.42, |
| "grad_norm": 0.18685045838356018, |
| "learning_rate": 0.0003762172929164974, |
| "loss": 3.4456027221679686, |
| "step": 21000 |
| }, |
| { |
| "epoch": 0.42, |
| "eval_accuracy": 0.3708611474909034, |
| "eval_loss": 3.402836799621582, |
| "eval_runtime": 5.7802, |
| "eval_samples_per_second": 335.801, |
| "eval_steps_per_second": 21.106, |
| "step": 21000 |
| }, |
| { |
| "epoch": 0.422, |
| "grad_norm": 0.18441364169120789, |
| "learning_rate": 0.0003743871972436601, |
| "loss": 3.4683151245117188, |
| "step": 21100 |
| }, |
| { |
| "epoch": 0.424, |
| "grad_norm": 0.18499158322811127, |
| "learning_rate": 0.0003725541471920217, |
| "loss": 3.440291748046875, |
| "step": 21200 |
| }, |
| { |
| "epoch": 0.426, |
| "grad_norm": 0.19171582162380219, |
| "learning_rate": 0.00037071821556341393, |
| "loss": 3.4521221923828125, |
| "step": 21300 |
| }, |
| { |
| "epoch": 0.428, |
| "grad_norm": 0.2012597918510437, |
| "learning_rate": 0.0003688794752741139, |
| "loss": 3.4446597290039063, |
| "step": 21400 |
| }, |
| { |
| "epoch": 0.43, |
| "grad_norm": 0.20872637629508972, |
| "learning_rate": 0.000367037999351948, |
| "loss": 3.4504824829101564, |
| "step": 21500 |
| }, |
| { |
| "epoch": 0.432, |
| "grad_norm": 0.19605720043182373, |
| "learning_rate": 0.0003651938609333918, |
| "loss": 3.432515869140625, |
| "step": 21600 |
| }, |
| { |
| "epoch": 0.434, |
| "grad_norm": 0.21077193319797516, |
| "learning_rate": 0.00036334713326066496, |
| "loss": 3.430672607421875, |
| "step": 21700 |
| }, |
| { |
| "epoch": 0.436, |
| "grad_norm": 0.20861658453941345, |
| "learning_rate": 0.00036149788967882265, |
| "loss": 3.434753112792969, |
| "step": 21800 |
| }, |
| { |
| "epoch": 0.438, |
| "grad_norm": 0.18320897221565247, |
| "learning_rate": 0.00035964620363284266, |
| "loss": 3.4054229736328123, |
| "step": 21900 |
| }, |
| { |
| "epoch": 0.44, |
| "grad_norm": 0.20277726650238037, |
| "learning_rate": 0.00035779214866470794, |
| "loss": 3.412484130859375, |
| "step": 22000 |
| }, |
| { |
| "epoch": 0.44, |
| "eval_accuracy": 0.37140054302511166, |
| "eval_loss": 3.3950016498565674, |
| "eval_runtime": 6.2447, |
| "eval_samples_per_second": 310.823, |
| "eval_steps_per_second": 19.537, |
| "step": 22000 |
| }, |
| { |
| "epoch": 0.442, |
| "grad_norm": 0.1982349008321762, |
| "learning_rate": 0.0003559357984104863, |
| "loss": 3.4337942504882815, |
| "step": 22100 |
| }, |
| { |
| "epoch": 0.444, |
| "grad_norm": 0.19478100538253784, |
| "learning_rate": 0.00035407722659740545, |
| "loss": 3.401671142578125, |
| "step": 22200 |
| }, |
| { |
| "epoch": 0.446, |
| "grad_norm": 0.20211219787597656, |
| "learning_rate": 0.0003522165070409252, |
| "loss": 3.416873779296875, |
| "step": 22300 |
| }, |
| { |
| "epoch": 0.448, |
| "grad_norm": 0.23854756355285645, |
| "learning_rate": 0.0003503537136418055, |
| "loss": 3.445356750488281, |
| "step": 22400 |
| }, |
| { |
| "epoch": 0.45, |
| "grad_norm": 0.1983487904071808, |
| "learning_rate": 0.00034848892038317174, |
| "loss": 3.4137481689453124, |
| "step": 22500 |
| }, |
| { |
| "epoch": 0.452, |
| "grad_norm": 0.20169700682163239, |
| "learning_rate": 0.00034662220132757596, |
| "loss": 3.4005947875976563, |
| "step": 22600 |
| }, |
| { |
| "epoch": 0.454, |
| "grad_norm": 0.19934488832950592, |
| "learning_rate": 0.00034475363061405555, |
| "loss": 3.4056222534179685, |
| "step": 22700 |
| }, |
| { |
| "epoch": 0.456, |
| "grad_norm": 0.212734192609787, |
| "learning_rate": 0.00034288328245518906, |
| "loss": 3.399494323730469, |
| "step": 22800 |
| }, |
| { |
| "epoch": 0.458, |
| "grad_norm": 0.2278880625963211, |
| "learning_rate": 0.00034101123113414807, |
| "loss": 3.403822326660156, |
| "step": 22900 |
| }, |
| { |
| "epoch": 0.46, |
| "grad_norm": 0.4838724434375763, |
| "learning_rate": 0.0003391375510017477, |
| "loss": 3.447413024902344, |
| "step": 23000 |
| }, |
| { |
| "epoch": 0.46, |
| "eval_accuracy": 0.3657575583429366, |
| "eval_loss": 3.438328742980957, |
| "eval_runtime": 6.1806, |
| "eval_samples_per_second": 314.047, |
| "eval_steps_per_second": 19.739, |
| "step": 23000 |
| }, |
| { |
| "epoch": 0.462, |
| "grad_norm": 0.4579889178276062, |
| "learning_rate": 0.00033726231647349304, |
| "loss": 3.4946719360351564, |
| "step": 23100 |
| }, |
| { |
| "epoch": 0.464, |
| "grad_norm": 0.724980890750885, |
| "learning_rate": 0.00033538560202662403, |
| "loss": 3.561537170410156, |
| "step": 23200 |
| }, |
| { |
| "epoch": 0.466, |
| "grad_norm": 1.940724492073059, |
| "learning_rate": 0.00033350748219715745, |
| "loss": 3.64361572265625, |
| "step": 23300 |
| }, |
| { |
| "epoch": 0.468, |
| "grad_norm": 2.4526569843292236, |
| "learning_rate": 0.00033162803157692656, |
| "loss": 3.734945983886719, |
| "step": 23400 |
| }, |
| { |
| "epoch": 0.47, |
| "grad_norm": 2.5834765434265137, |
| "learning_rate": 0.0003297473248106184, |
| "loss": 3.86871826171875, |
| "step": 23500 |
| }, |
| { |
| "epoch": 0.472, |
| "grad_norm": 4.770744323730469, |
| "learning_rate": 0.0003278654365928096, |
| "loss": 3.974892883300781, |
| "step": 23600 |
| }, |
| { |
| "epoch": 0.474, |
| "grad_norm": 21.92555046081543, |
| "learning_rate": 0.0003259824416649996, |
| "loss": 4.260992431640625, |
| "step": 23700 |
| }, |
| { |
| "epoch": 0.476, |
| "grad_norm": 3082.5625, |
| "learning_rate": 0.00032409841481264223, |
| "loss": 5.928079833984375, |
| "step": 23800 |
| }, |
| { |
| "epoch": 0.478, |
| "grad_norm": 3319.344482421875, |
| "learning_rate": 0.0003222134308621751, |
| "loss": 9.030023803710938, |
| "step": 23900 |
| }, |
| { |
| "epoch": 0.48, |
| "grad_norm": 322.2440490722656, |
| "learning_rate": 0.00032032756467804865, |
| "loss": 8.4674365234375, |
| "step": 24000 |
| }, |
| { |
| "epoch": 0.48, |
| "eval_accuracy": 0.05309769310108071, |
| "eval_loss": 7.991781711578369, |
| "eval_runtime": 5.9674, |
| "eval_samples_per_second": 325.269, |
| "eval_steps_per_second": 20.445, |
| "step": 24000 |
| }, |
| { |
| "epoch": 0.482, |
| "grad_norm": 338.5876770019531, |
| "learning_rate": 0.0003184408911597519, |
| "loss": 7.839384155273438, |
| "step": 24100 |
| }, |
| { |
| "epoch": 0.484, |
| "grad_norm": 79.88603210449219, |
| "learning_rate": 0.0003165534852388384, |
| "loss": 7.44333251953125, |
| "step": 24200 |
| }, |
| { |
| "epoch": 0.486, |
| "grad_norm": 32.77883529663086, |
| "learning_rate": 0.0003146654218759499, |
| "loss": 7.235584106445312, |
| "step": 24300 |
| }, |
| { |
| "epoch": 0.488, |
| "grad_norm": 27.993099212646484, |
| "learning_rate": 0.0003127767760578392, |
| "loss": 7.177199096679687, |
| "step": 24400 |
| }, |
| { |
| "epoch": 0.49, |
| "grad_norm": 15.215092658996582, |
| "learning_rate": 0.000310887622794392, |
| "loss": 7.115394287109375, |
| "step": 24500 |
| }, |
| { |
| "epoch": 0.492, |
| "grad_norm": 12.316529273986816, |
| "learning_rate": 0.00030899803711564806, |
| "loss": 6.990827026367188, |
| "step": 24600 |
| }, |
| { |
| "epoch": 0.494, |
| "grad_norm": 2.696770191192627, |
| "learning_rate": 0.0003071080940688207, |
| "loss": 7.0589227294921875, |
| "step": 24700 |
| }, |
| { |
| "epoch": 0.496, |
| "grad_norm": 48.15589141845703, |
| "learning_rate": 0.0003052178687153168, |
| "loss": 7.067171020507812, |
| "step": 24800 |
| }, |
| { |
| "epoch": 0.498, |
| "grad_norm": 28.01230239868164, |
| "learning_rate": 0.0003033274361277553, |
| "loss": 7.030769653320313, |
| "step": 24900 |
| }, |
| { |
| "epoch": 0.5, |
| "grad_norm": 69.79560852050781, |
| "learning_rate": 0.0003014368713869857, |
| "loss": 6.974198608398438, |
| "step": 25000 |
| }, |
| { |
| "epoch": 0.5, |
| "eval_accuracy": 0.09623723724632026, |
| "eval_loss": 6.917368412017822, |
| "eval_runtime": 6.0867, |
| "eval_samples_per_second": 318.891, |
| "eval_steps_per_second": 20.044, |
| "step": 25000 |
| }, |
| { |
| "epoch": 0.502, |
| "grad_norm": 42.4299201965332, |
| "learning_rate": 0.00029954624957910595, |
| "loss": 6.993025512695312, |
| "step": 25100 |
| }, |
| { |
| "epoch": 0.504, |
| "grad_norm": 20.528335571289062, |
| "learning_rate": 0.0002976556457924811, |
| "loss": 7.020432739257813, |
| "step": 25200 |
| }, |
| { |
| "epoch": 0.506, |
| "grad_norm": 42.26169204711914, |
| "learning_rate": 0.0002957651351147596, |
| "loss": 6.934894409179687, |
| "step": 25300 |
| }, |
| { |
| "epoch": 0.508, |
| "grad_norm": 3.344475507736206, |
| "learning_rate": 0.0002938747926298927, |
| "loss": 6.941484375, |
| "step": 25400 |
| }, |
| { |
| "epoch": 0.51, |
| "grad_norm": 64.28266906738281, |
| "learning_rate": 0.0002919846934151513, |
| "loss": 7.314700317382813, |
| "step": 25500 |
| }, |
| { |
| "epoch": 0.512, |
| "grad_norm": 29.82670783996582, |
| "learning_rate": 0.00029009491253814457, |
| "loss": 7.119804077148437, |
| "step": 25600 |
| }, |
| { |
| "epoch": 0.514, |
| "grad_norm": 59.44496536254883, |
| "learning_rate": 0.0002882055250538384, |
| "loss": 7.006104736328125, |
| "step": 25700 |
| }, |
| { |
| "epoch": 0.516, |
| "grad_norm": 42.77421569824219, |
| "learning_rate": 0.00028631660600157493, |
| "loss": 6.991202392578125, |
| "step": 25800 |
| }, |
| { |
| "epoch": 0.518, |
| "grad_norm": 14.600953102111816, |
| "learning_rate": 0.0002844282304020917, |
| "loss": 6.969422607421875, |
| "step": 25900 |
| }, |
| { |
| "epoch": 0.52, |
| "grad_norm": 77.17425537109375, |
| "learning_rate": 0.00028254047325454234, |
| "loss": 7.046774291992188, |
| "step": 26000 |
| }, |
| { |
| "epoch": 0.52, |
| "eval_accuracy": 0.06708366478432748, |
| "eval_loss": 7.517740249633789, |
| "eval_runtime": 5.7572, |
| "eval_samples_per_second": 337.143, |
| "eval_steps_per_second": 21.191, |
| "step": 26000 |
| } |
| ], |
| "logging_steps": 100, |
| "max_steps": 50000, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 9223372036854775807, |
| "save_steps": 2000, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 8.1523113984e+17, |
| "train_batch_size": 120, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
