{ "best_global_step": 104000, "best_metric": 3.2727417945861816, "best_model_checkpoint": "/scratch/cl5625/exceptions/models/high_2000_1208/checkpoint-20000", "epoch": 33.36929371231697, "eval_steps": 1000, "global_step": 124000, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.013458225667527994, "grad_norm": 0.8966911435127258, "learning_rate": 0.000294, "loss": 8.6065, "step": 50 }, { "epoch": 0.026916451335055987, "grad_norm": 1.2048732042312622, "learning_rate": 0.0005939999999999999, "loss": 6.8585, "step": 100 }, { "epoch": 0.04037467700258398, "grad_norm": 0.9153667688369751, "learning_rate": 0.0005998020735155513, "loss": 6.3263, "step": 150 }, { "epoch": 0.053832902670111975, "grad_norm": 1.4322563409805298, "learning_rate": 0.0005996001077150935, "loss": 6.0257, "step": 200 }, { "epoch": 0.06729112833763996, "grad_norm": 1.2138304710388184, "learning_rate": 0.0005993981419146358, "loss": 5.865, "step": 250 }, { "epoch": 0.08074935400516796, "grad_norm": 1.4077575206756592, "learning_rate": 0.0005991961761141779, "loss": 5.7384, "step": 300 }, { "epoch": 0.09420757967269595, "grad_norm": 1.1934620141983032, "learning_rate": 0.0005989942103137202, "loss": 5.6093, "step": 350 }, { "epoch": 0.10766580534022395, "grad_norm": 1.1857770681381226, "learning_rate": 0.0005987922445132624, "loss": 5.4578, "step": 400 }, { "epoch": 0.12112403100775193, "grad_norm": 1.8576644659042358, "learning_rate": 0.0005985902787128047, "loss": 5.3535, "step": 450 }, { "epoch": 0.13458225667527993, "grad_norm": 1.6441043615341187, "learning_rate": 0.0005983883129123468, "loss": 5.2506, "step": 500 }, { "epoch": 0.14804048234280792, "grad_norm": 1.3366986513137817, "learning_rate": 0.000598186347111889, "loss": 5.14, "step": 550 }, { "epoch": 0.16149870801033592, "grad_norm": 1.1094809770584106, "learning_rate": 0.0005979843813114312, "loss": 5.0689, "step": 600 }, { "epoch": 0.1749569336778639, "grad_norm": 1.1344627141952515, "learning_rate": 0.0005977824155109734, "loss": 4.9848, "step": 650 }, { "epoch": 0.1884151593453919, "grad_norm": 1.487619400024414, "learning_rate": 0.0005975804497105157, "loss": 4.9104, "step": 700 }, { "epoch": 0.2018733850129199, "grad_norm": 1.3195019960403442, "learning_rate": 0.0005973784839100578, "loss": 4.8522, "step": 750 }, { "epoch": 0.2153316106804479, "grad_norm": 1.1873126029968262, "learning_rate": 0.0005971765181096001, "loss": 4.7907, "step": 800 }, { "epoch": 0.2287898363479759, "grad_norm": 0.8753162026405334, "learning_rate": 0.0005969745523091422, "loss": 4.7346, "step": 850 }, { "epoch": 0.24224806201550386, "grad_norm": 1.0871766805648804, "learning_rate": 0.0005967725865086845, "loss": 4.6963, "step": 900 }, { "epoch": 0.2557062876830319, "grad_norm": 1.1119087934494019, "learning_rate": 0.0005965706207082267, "loss": 4.6261, "step": 950 }, { "epoch": 0.26916451335055985, "grad_norm": 1.053986668586731, "learning_rate": 0.0005963686549077689, "loss": 4.5929, "step": 1000 }, { "epoch": 0.26916451335055985, "eval_accuracy": 0.2670213769207134, "eval_loss": 4.52205228805542, "eval_runtime": 90.1745, "eval_samples_per_second": 199.746, "eval_steps_per_second": 12.487, "step": 1000 }, { "epoch": 0.2826227390180879, "grad_norm": 0.7153794169425964, "learning_rate": 0.0005961666891073111, "loss": 4.5503, "step": 1050 }, { "epoch": 0.29608096468561584, "grad_norm": 0.7922793030738831, "learning_rate": 0.0005959647233068533, "loss": 4.508, "step": 1100 }, { "epoch": 0.30953919035314387, "grad_norm": 0.9259533286094666, "learning_rate": 0.0005957627575063955, "loss": 4.4437, "step": 1150 }, { "epoch": 0.32299741602067183, "grad_norm": 0.8578700423240662, "learning_rate": 0.0005955607917059378, "loss": 4.4307, "step": 1200 }, { "epoch": 0.3364556416881998, "grad_norm": 0.851991593837738, "learning_rate": 0.00059535882590548, "loss": 4.397, "step": 1250 }, { "epoch": 0.3499138673557278, "grad_norm": 0.8914237022399902, "learning_rate": 0.0005951568601050221, "loss": 4.3653, "step": 1300 }, { "epoch": 0.3633720930232558, "grad_norm": 0.7544556856155396, "learning_rate": 0.0005949548943045644, "loss": 4.3258, "step": 1350 }, { "epoch": 0.3768303186907838, "grad_norm": 0.9407439231872559, "learning_rate": 0.0005947529285041066, "loss": 4.3163, "step": 1400 }, { "epoch": 0.3902885443583118, "grad_norm": 0.8338985443115234, "learning_rate": 0.0005945509627036488, "loss": 4.2708, "step": 1450 }, { "epoch": 0.4037467700258398, "grad_norm": 0.6592851877212524, "learning_rate": 0.000594348996903191, "loss": 4.2533, "step": 1500 }, { "epoch": 0.4172049956933678, "grad_norm": 0.9020254611968994, "learning_rate": 0.0005941470311027332, "loss": 4.2288, "step": 1550 }, { "epoch": 0.4306632213608958, "grad_norm": 0.6539929509162903, "learning_rate": 0.0005939450653022754, "loss": 4.2256, "step": 1600 }, { "epoch": 0.44412144702842377, "grad_norm": 0.6174359917640686, "learning_rate": 0.0005937430995018177, "loss": 4.1885, "step": 1650 }, { "epoch": 0.4575796726959518, "grad_norm": 0.8427085876464844, "learning_rate": 0.0005935411337013598, "loss": 4.173, "step": 1700 }, { "epoch": 0.47103789836347976, "grad_norm": 0.7832385301589966, "learning_rate": 0.0005933391679009021, "loss": 4.1612, "step": 1750 }, { "epoch": 0.4844961240310077, "grad_norm": 0.7887812852859497, "learning_rate": 0.0005931372021004442, "loss": 4.1244, "step": 1800 }, { "epoch": 0.49795434969853575, "grad_norm": 0.8242790102958679, "learning_rate": 0.0005929352362999865, "loss": 4.1335, "step": 1850 }, { "epoch": 0.5114125753660638, "grad_norm": 0.6264904737472534, "learning_rate": 0.0005927332704995287, "loss": 4.1285, "step": 1900 }, { "epoch": 0.5248708010335917, "grad_norm": 0.5883820652961731, "learning_rate": 0.0005925313046990709, "loss": 4.0959, "step": 1950 }, { "epoch": 0.5383290267011197, "grad_norm": 0.6146206855773926, "learning_rate": 0.0005923293388986131, "loss": 4.0854, "step": 2000 }, { "epoch": 0.5383290267011197, "eval_accuracy": 0.31779552922535226, "eval_loss": 4.020822048187256, "eval_runtime": 90.5526, "eval_samples_per_second": 198.912, "eval_steps_per_second": 12.435, "step": 2000 }, { "epoch": 0.5517872523686477, "grad_norm": 0.6017832159996033, "learning_rate": 0.0005921273730981553, "loss": 4.0682, "step": 2050 }, { "epoch": 0.5652454780361758, "grad_norm": 0.6516919732093811, "learning_rate": 0.0005919254072976975, "loss": 4.0639, "step": 2100 }, { "epoch": 0.5787037037037037, "grad_norm": 0.7468438744544983, "learning_rate": 0.0005917234414972398, "loss": 4.0659, "step": 2150 }, { "epoch": 0.5921619293712317, "grad_norm": 0.6093896627426147, "learning_rate": 0.000591521475696782, "loss": 4.0391, "step": 2200 }, { "epoch": 0.6056201550387597, "grad_norm": 0.5305492877960205, "learning_rate": 0.0005913195098963241, "loss": 4.0327, "step": 2250 }, { "epoch": 0.6190783807062877, "grad_norm": 0.5827754735946655, "learning_rate": 0.0005911175440958664, "loss": 4.018, "step": 2300 }, { "epoch": 0.6325366063738157, "grad_norm": 0.6132373213768005, "learning_rate": 0.0005909155782954086, "loss": 4.0198, "step": 2350 }, { "epoch": 0.6459948320413437, "grad_norm": 0.5304743051528931, "learning_rate": 0.0005907136124949508, "loss": 3.9967, "step": 2400 }, { "epoch": 0.6594530577088716, "grad_norm": 0.6900047659873962, "learning_rate": 0.000590511646694493, "loss": 3.9902, "step": 2450 }, { "epoch": 0.6729112833763996, "grad_norm": 0.612856388092041, "learning_rate": 0.0005903096808940352, "loss": 3.9779, "step": 2500 }, { "epoch": 0.6863695090439277, "grad_norm": 0.6652615666389465, "learning_rate": 0.0005901077150935774, "loss": 3.9835, "step": 2550 }, { "epoch": 0.6998277347114557, "grad_norm": 0.5629779100418091, "learning_rate": 0.0005899057492931197, "loss": 3.9563, "step": 2600 }, { "epoch": 0.7132859603789836, "grad_norm": 0.5258368253707886, "learning_rate": 0.0005897037834926618, "loss": 3.9506, "step": 2650 }, { "epoch": 0.7267441860465116, "grad_norm": 0.5157285928726196, "learning_rate": 0.0005895018176922041, "loss": 3.9386, "step": 2700 }, { "epoch": 0.7402024117140397, "grad_norm": 0.5549437403678894, "learning_rate": 0.0005892998518917462, "loss": 3.937, "step": 2750 }, { "epoch": 0.7536606373815676, "grad_norm": 0.6192649602890015, "learning_rate": 0.0005890978860912885, "loss": 3.9281, "step": 2800 }, { "epoch": 0.7671188630490956, "grad_norm": 0.5465920567512512, "learning_rate": 0.0005888959202908307, "loss": 3.9218, "step": 2850 }, { "epoch": 0.7805770887166236, "grad_norm": 0.4951372742652893, "learning_rate": 0.0005886939544903729, "loss": 3.9108, "step": 2900 }, { "epoch": 0.7940353143841516, "grad_norm": 0.6514776945114136, "learning_rate": 0.0005884919886899151, "loss": 3.9119, "step": 2950 }, { "epoch": 0.8074935400516796, "grad_norm": 0.5681062936782837, "learning_rate": 0.0005882900228894573, "loss": 3.8989, "step": 3000 }, { "epoch": 0.8074935400516796, "eval_accuracy": 0.3335286803796382, "eval_loss": 3.8467557430267334, "eval_runtime": 89.2556, "eval_samples_per_second": 201.803, "eval_steps_per_second": 12.615, "step": 3000 }, { "epoch": 0.8209517657192076, "grad_norm": 0.4961033761501312, "learning_rate": 0.0005880880570889996, "loss": 3.895, "step": 3050 }, { "epoch": 0.8344099913867356, "grad_norm": 0.5452906489372253, "learning_rate": 0.0005878860912885418, "loss": 3.8861, "step": 3100 }, { "epoch": 0.8478682170542635, "grad_norm": 0.5429787039756775, "learning_rate": 0.000587684125488084, "loss": 3.881, "step": 3150 }, { "epoch": 0.8613264427217916, "grad_norm": 0.46722593903541565, "learning_rate": 0.0005874821596876261, "loss": 3.8706, "step": 3200 }, { "epoch": 0.8747846683893196, "grad_norm": 0.5518487095832825, "learning_rate": 0.0005872801938871684, "loss": 3.8696, "step": 3250 }, { "epoch": 0.8882428940568475, "grad_norm": 0.4713578224182129, "learning_rate": 0.0005870782280867106, "loss": 3.8552, "step": 3300 }, { "epoch": 0.9017011197243755, "grad_norm": 0.4993533194065094, "learning_rate": 0.0005868762622862528, "loss": 3.8544, "step": 3350 }, { "epoch": 0.9151593453919036, "grad_norm": 0.498419851064682, "learning_rate": 0.000586674296485795, "loss": 3.8421, "step": 3400 }, { "epoch": 0.9286175710594315, "grad_norm": 0.5733362436294556, "learning_rate": 0.0005864723306853373, "loss": 3.8391, "step": 3450 }, { "epoch": 0.9420757967269595, "grad_norm": 0.4247209429740906, "learning_rate": 0.0005862703648848794, "loss": 3.8337, "step": 3500 }, { "epoch": 0.9555340223944875, "grad_norm": 0.43807166814804077, "learning_rate": 0.0005860683990844217, "loss": 3.8269, "step": 3550 }, { "epoch": 0.9689922480620154, "grad_norm": 0.5553157329559326, "learning_rate": 0.0005858664332839638, "loss": 3.8157, "step": 3600 }, { "epoch": 0.9824504737295435, "grad_norm": 0.463137149810791, "learning_rate": 0.0005856644674835061, "loss": 3.8169, "step": 3650 }, { "epoch": 0.9959086993970715, "grad_norm": 0.455089271068573, "learning_rate": 0.0005854625016830483, "loss": 3.8122, "step": 3700 }, { "epoch": 1.009151593453919, "grad_norm": 0.4528520703315735, "learning_rate": 0.0005852605358825905, "loss": 3.7817, "step": 3750 }, { "epoch": 1.022609819121447, "grad_norm": 0.4980250895023346, "learning_rate": 0.0005850585700821327, "loss": 3.7517, "step": 3800 }, { "epoch": 1.036068044788975, "grad_norm": 0.46913275122642517, "learning_rate": 0.0005848566042816749, "loss": 3.7477, "step": 3850 }, { "epoch": 1.049526270456503, "grad_norm": 0.4094890058040619, "learning_rate": 0.0005846546384812171, "loss": 3.7329, "step": 3900 }, { "epoch": 1.062984496124031, "grad_norm": 0.4407455325126648, "learning_rate": 0.0005844526726807594, "loss": 3.7414, "step": 3950 }, { "epoch": 1.076442721791559, "grad_norm": 0.44919925928115845, "learning_rate": 0.0005842507068803016, "loss": 3.7397, "step": 4000 }, { "epoch": 1.076442721791559, "eval_accuracy": 0.3441485845704951, "eval_loss": 3.736539363861084, "eval_runtime": 89.4001, "eval_samples_per_second": 201.476, "eval_steps_per_second": 12.595, "step": 4000 }, { "epoch": 1.089900947459087, "grad_norm": 0.4611460566520691, "learning_rate": 0.0005840487410798437, "loss": 3.7266, "step": 4050 }, { "epoch": 1.103359173126615, "grad_norm": 0.40715456008911133, "learning_rate": 0.000583846775279386, "loss": 3.7411, "step": 4100 }, { "epoch": 1.116817398794143, "grad_norm": 0.45220470428466797, "learning_rate": 0.0005836448094789282, "loss": 3.7195, "step": 4150 }, { "epoch": 1.130275624461671, "grad_norm": 0.46470025181770325, "learning_rate": 0.0005834428436784704, "loss": 3.7253, "step": 4200 }, { "epoch": 1.143733850129199, "grad_norm": 0.42880234122276306, "learning_rate": 0.0005832408778780126, "loss": 3.7232, "step": 4250 }, { "epoch": 1.157192075796727, "grad_norm": 0.45918765664100647, "learning_rate": 0.0005830389120775548, "loss": 3.7164, "step": 4300 }, { "epoch": 1.1706503014642549, "grad_norm": 0.4585005044937134, "learning_rate": 0.000582836946277097, "loss": 3.7264, "step": 4350 }, { "epoch": 1.1841085271317828, "grad_norm": 0.4340853989124298, "learning_rate": 0.0005826349804766393, "loss": 3.7137, "step": 4400 }, { "epoch": 1.197566752799311, "grad_norm": 0.45874127745628357, "learning_rate": 0.0005824330146761814, "loss": 3.7157, "step": 4450 }, { "epoch": 1.211024978466839, "grad_norm": 0.4334906339645386, "learning_rate": 0.0005822310488757237, "loss": 3.6989, "step": 4500 }, { "epoch": 1.224483204134367, "grad_norm": 0.45151418447494507, "learning_rate": 0.0005820290830752658, "loss": 3.7126, "step": 4550 }, { "epoch": 1.237941429801895, "grad_norm": 0.36777910590171814, "learning_rate": 0.0005818271172748081, "loss": 3.717, "step": 4600 }, { "epoch": 1.251399655469423, "grad_norm": 0.4326784014701843, "learning_rate": 0.0005816251514743503, "loss": 3.7005, "step": 4650 }, { "epoch": 1.2648578811369509, "grad_norm": 0.41935795545578003, "learning_rate": 0.0005814231856738925, "loss": 3.7017, "step": 4700 }, { "epoch": 1.2783161068044788, "grad_norm": 0.44362640380859375, "learning_rate": 0.0005812212198734347, "loss": 3.6986, "step": 4750 }, { "epoch": 1.291774332472007, "grad_norm": 0.4299990236759186, "learning_rate": 0.0005810192540729769, "loss": 3.6967, "step": 4800 }, { "epoch": 1.3052325581395348, "grad_norm": 0.442768394947052, "learning_rate": 0.0005808172882725192, "loss": 3.6887, "step": 4850 }, { "epoch": 1.318690783807063, "grad_norm": 0.42501920461654663, "learning_rate": 0.0005806153224720614, "loss": 3.6987, "step": 4900 }, { "epoch": 1.332149009474591, "grad_norm": 0.5787553191184998, "learning_rate": 0.0005804133566716036, "loss": 3.6963, "step": 4950 }, { "epoch": 1.3456072351421189, "grad_norm": 0.43269771337509155, "learning_rate": 0.0005802113908711457, "loss": 3.6972, "step": 5000 }, { "epoch": 1.3456072351421189, "eval_accuracy": 0.3507485551054679, "eval_loss": 3.6660983562469482, "eval_runtime": 89.5086, "eval_samples_per_second": 201.232, "eval_steps_per_second": 12.58, "step": 5000 }, { "epoch": 1.3590654608096469, "grad_norm": 0.40254127979278564, "learning_rate": 0.000580009425070688, "loss": 3.6895, "step": 5050 }, { "epoch": 1.3725236864771748, "grad_norm": 0.41719353199005127, "learning_rate": 0.0005798074592702302, "loss": 3.6812, "step": 5100 }, { "epoch": 1.3859819121447028, "grad_norm": 0.455530047416687, "learning_rate": 0.0005796054934697724, "loss": 3.679, "step": 5150 }, { "epoch": 1.3994401378122308, "grad_norm": 0.38591596484184265, "learning_rate": 0.0005794035276693146, "loss": 3.6641, "step": 5200 }, { "epoch": 1.412898363479759, "grad_norm": 0.46197277307510376, "learning_rate": 0.0005792015618688568, "loss": 3.6713, "step": 5250 }, { "epoch": 1.4263565891472867, "grad_norm": 0.45006898045539856, "learning_rate": 0.000578999596068399, "loss": 3.6638, "step": 5300 }, { "epoch": 1.4398148148148149, "grad_norm": 0.42323538661003113, "learning_rate": 0.0005787976302679413, "loss": 3.6613, "step": 5350 }, { "epoch": 1.4532730404823428, "grad_norm": 0.4862610697746277, "learning_rate": 0.0005785956644674834, "loss": 3.6683, "step": 5400 }, { "epoch": 1.4667312661498708, "grad_norm": 0.4114309847354889, "learning_rate": 0.0005783936986670257, "loss": 3.6687, "step": 5450 }, { "epoch": 1.4801894918173988, "grad_norm": 0.4172651171684265, "learning_rate": 0.0005781917328665678, "loss": 3.668, "step": 5500 }, { "epoch": 1.4936477174849268, "grad_norm": 0.3980918228626251, "learning_rate": 0.0005779897670661101, "loss": 3.659, "step": 5550 }, { "epoch": 1.507105943152455, "grad_norm": 0.43365082144737244, "learning_rate": 0.0005777878012656523, "loss": 3.6637, "step": 5600 }, { "epoch": 1.5205641688199827, "grad_norm": 0.39800316095352173, "learning_rate": 0.0005775858354651945, "loss": 3.6608, "step": 5650 }, { "epoch": 1.5340223944875109, "grad_norm": 0.4080260396003723, "learning_rate": 0.0005773838696647367, "loss": 3.6434, "step": 5700 }, { "epoch": 1.5474806201550386, "grad_norm": 0.3940686881542206, "learning_rate": 0.000577181903864279, "loss": 3.6455, "step": 5750 }, { "epoch": 1.5609388458225668, "grad_norm": 0.3766889274120331, "learning_rate": 0.0005769799380638212, "loss": 3.6487, "step": 5800 }, { "epoch": 1.5743970714900948, "grad_norm": 0.39470019936561584, "learning_rate": 0.0005767779722633633, "loss": 3.6451, "step": 5850 }, { "epoch": 1.5878552971576227, "grad_norm": 0.41684457659721375, "learning_rate": 0.0005765760064629056, "loss": 3.6579, "step": 5900 }, { "epoch": 1.6013135228251507, "grad_norm": 0.38935387134552, "learning_rate": 0.0005763740406624477, "loss": 3.6381, "step": 5950 }, { "epoch": 1.6147717484926787, "grad_norm": 0.38764357566833496, "learning_rate": 0.00057617207486199, "loss": 3.6419, "step": 6000 }, { "epoch": 1.6147717484926787, "eval_accuracy": 0.35599695875721904, "eval_loss": 3.6079609394073486, "eval_runtime": 89.4352, "eval_samples_per_second": 201.397, "eval_steps_per_second": 12.59, "step": 6000 }, { "epoch": 1.6282299741602069, "grad_norm": 0.44735968112945557, "learning_rate": 0.0005759701090615322, "loss": 3.6374, "step": 6050 }, { "epoch": 1.6416881998277346, "grad_norm": 0.4507206678390503, "learning_rate": 0.0005757681432610744, "loss": 3.6237, "step": 6100 }, { "epoch": 1.6551464254952628, "grad_norm": 0.3999354839324951, "learning_rate": 0.0005755661774606166, "loss": 3.6312, "step": 6150 }, { "epoch": 1.6686046511627906, "grad_norm": 0.4594976603984833, "learning_rate": 0.0005753642116601588, "loss": 3.6364, "step": 6200 }, { "epoch": 1.6820628768303187, "grad_norm": 0.40472447872161865, "learning_rate": 0.000575162245859701, "loss": 3.6215, "step": 6250 }, { "epoch": 1.6955211024978467, "grad_norm": 0.367737352848053, "learning_rate": 0.0005749602800592433, "loss": 3.6221, "step": 6300 }, { "epoch": 1.7089793281653747, "grad_norm": 0.36534759402275085, "learning_rate": 0.0005747583142587854, "loss": 3.6185, "step": 6350 }, { "epoch": 1.7224375538329026, "grad_norm": 0.3866477608680725, "learning_rate": 0.0005745563484583277, "loss": 3.6153, "step": 6400 }, { "epoch": 1.7358957795004306, "grad_norm": 0.4261724352836609, "learning_rate": 0.0005743543826578698, "loss": 3.6152, "step": 6450 }, { "epoch": 1.7493540051679588, "grad_norm": 0.3843180239200592, "learning_rate": 0.0005741524168574121, "loss": 3.6064, "step": 6500 }, { "epoch": 1.7628122308354865, "grad_norm": 0.4083123803138733, "learning_rate": 0.0005739504510569543, "loss": 3.6175, "step": 6550 }, { "epoch": 1.7762704565030147, "grad_norm": 0.38491323590278625, "learning_rate": 0.0005737484852564966, "loss": 3.6114, "step": 6600 }, { "epoch": 1.7897286821705425, "grad_norm": 0.41620877385139465, "learning_rate": 0.0005735465194560387, "loss": 3.6221, "step": 6650 }, { "epoch": 1.8031869078380707, "grad_norm": 0.334351509809494, "learning_rate": 0.000573344553655581, "loss": 3.5979, "step": 6700 }, { "epoch": 1.8166451335055986, "grad_norm": 0.45867523550987244, "learning_rate": 0.0005731425878551232, "loss": 3.6177, "step": 6750 }, { "epoch": 1.8301033591731266, "grad_norm": 0.3858473598957062, "learning_rate": 0.0005729406220546653, "loss": 3.6184, "step": 6800 }, { "epoch": 1.8435615848406546, "grad_norm": 0.34941157698631287, "learning_rate": 0.0005727386562542076, "loss": 3.6057, "step": 6850 }, { "epoch": 1.8570198105081825, "grad_norm": 0.364150732755661, "learning_rate": 0.0005725366904537497, "loss": 3.6044, "step": 6900 }, { "epoch": 1.8704780361757107, "grad_norm": 0.41074541211128235, "learning_rate": 0.000572334724653292, "loss": 3.6055, "step": 6950 }, { "epoch": 1.8839362618432385, "grad_norm": 0.3702612817287445, "learning_rate": 0.0005721327588528342, "loss": 3.5926, "step": 7000 }, { "epoch": 1.8839362618432385, "eval_accuracy": 0.36019105332257295, "eval_loss": 3.5668296813964844, "eval_runtime": 89.053, "eval_samples_per_second": 202.262, "eval_steps_per_second": 12.644, "step": 7000 }, { "epoch": 1.8973944875107667, "grad_norm": 0.36478501558303833, "learning_rate": 0.0005719307930523764, "loss": 3.5984, "step": 7050 }, { "epoch": 1.9108527131782944, "grad_norm": 0.40379798412323, "learning_rate": 0.0005717288272519186, "loss": 3.5937, "step": 7100 }, { "epoch": 1.9243109388458226, "grad_norm": 0.4081585109233856, "learning_rate": 0.0005715268614514608, "loss": 3.5918, "step": 7150 }, { "epoch": 1.9377691645133506, "grad_norm": 0.3517591655254364, "learning_rate": 0.000571324895651003, "loss": 3.5924, "step": 7200 }, { "epoch": 1.9512273901808785, "grad_norm": 0.37822774052619934, "learning_rate": 0.0005711229298505453, "loss": 3.5838, "step": 7250 }, { "epoch": 1.9646856158484065, "grad_norm": 0.3480222821235657, "learning_rate": 0.0005709209640500874, "loss": 3.5794, "step": 7300 }, { "epoch": 1.9781438415159345, "grad_norm": 0.3855062425136566, "learning_rate": 0.0005707189982496297, "loss": 3.5761, "step": 7350 }, { "epoch": 1.9916020671834627, "grad_norm": 0.36663126945495605, "learning_rate": 0.0005705170324491719, "loss": 3.5871, "step": 7400 }, { "epoch": 2.00484496124031, "grad_norm": 0.3746161460876465, "learning_rate": 0.0005703150666487141, "loss": 3.5418, "step": 7450 }, { "epoch": 2.018303186907838, "grad_norm": 0.3908737897872925, "learning_rate": 0.0005701131008482563, "loss": 3.493, "step": 7500 }, { "epoch": 2.0317614125753662, "grad_norm": 0.36380451917648315, "learning_rate": 0.0005699111350477986, "loss": 3.4989, "step": 7550 }, { "epoch": 2.045219638242894, "grad_norm": 0.39646074175834656, "learning_rate": 0.0005697091692473408, "loss": 3.4959, "step": 7600 }, { "epoch": 2.058677863910422, "grad_norm": 0.36035415530204773, "learning_rate": 0.000569507203446883, "loss": 3.4878, "step": 7650 }, { "epoch": 2.07213608957795, "grad_norm": 0.3832433819770813, "learning_rate": 0.0005693052376464252, "loss": 3.5133, "step": 7700 }, { "epoch": 2.085594315245478, "grad_norm": 0.3533892035484314, "learning_rate": 0.0005691032718459673, "loss": 3.5027, "step": 7750 }, { "epoch": 2.099052540913006, "grad_norm": 0.3850933313369751, "learning_rate": 0.0005689013060455096, "loss": 3.5023, "step": 7800 }, { "epoch": 2.112510766580534, "grad_norm": 0.3618728816509247, "learning_rate": 0.0005686993402450518, "loss": 3.5023, "step": 7850 }, { "epoch": 2.125968992248062, "grad_norm": 0.36867016553878784, "learning_rate": 0.000568497374444594, "loss": 3.5004, "step": 7900 }, { "epoch": 2.13942721791559, "grad_norm": 0.3630622625350952, "learning_rate": 0.0005682954086441362, "loss": 3.4875, "step": 7950 }, { "epoch": 2.152885443583118, "grad_norm": 0.4012206196784973, "learning_rate": 0.0005680934428436784, "loss": 3.5152, "step": 8000 }, { "epoch": 2.152885443583118, "eval_accuracy": 0.3640480166951104, "eval_loss": 3.5348846912384033, "eval_runtime": 89.1099, "eval_samples_per_second": 202.132, "eval_steps_per_second": 12.636, "step": 8000 }, { "epoch": 2.166343669250646, "grad_norm": 0.4097427427768707, "learning_rate": 0.0005678914770432206, "loss": 3.5051, "step": 8050 }, { "epoch": 2.179801894918174, "grad_norm": 0.37478891015052795, "learning_rate": 0.0005676895112427629, "loss": 3.5036, "step": 8100 }, { "epoch": 2.193260120585702, "grad_norm": 0.385329931974411, "learning_rate": 0.000567487545442305, "loss": 3.4956, "step": 8150 }, { "epoch": 2.20671834625323, "grad_norm": 0.40505632758140564, "learning_rate": 0.0005672855796418473, "loss": 3.5012, "step": 8200 }, { "epoch": 2.220176571920758, "grad_norm": 0.4362744390964508, "learning_rate": 0.0005670836138413894, "loss": 3.5054, "step": 8250 }, { "epoch": 2.233634797588286, "grad_norm": 0.337893545627594, "learning_rate": 0.0005668816480409317, "loss": 3.4965, "step": 8300 }, { "epoch": 2.2470930232558137, "grad_norm": 0.3597669005393982, "learning_rate": 0.0005666796822404739, "loss": 3.5006, "step": 8350 }, { "epoch": 2.260551248923342, "grad_norm": 0.3700796961784363, "learning_rate": 0.0005664777164400162, "loss": 3.4999, "step": 8400 }, { "epoch": 2.27400947459087, "grad_norm": 0.37950870394706726, "learning_rate": 0.0005662757506395583, "loss": 3.4976, "step": 8450 }, { "epoch": 2.287467700258398, "grad_norm": 0.37483659386634827, "learning_rate": 0.0005660737848391006, "loss": 3.5031, "step": 8500 }, { "epoch": 2.300925925925926, "grad_norm": 0.3778204023838043, "learning_rate": 0.0005658718190386428, "loss": 3.5025, "step": 8550 }, { "epoch": 2.314384151593454, "grad_norm": 0.39371633529663086, "learning_rate": 0.000565669853238185, "loss": 3.5017, "step": 8600 }, { "epoch": 2.327842377260982, "grad_norm": 0.38425666093826294, "learning_rate": 0.0005654678874377272, "loss": 3.5019, "step": 8650 }, { "epoch": 2.3413006029285097, "grad_norm": 0.34810662269592285, "learning_rate": 0.0005652659216372693, "loss": 3.509, "step": 8700 }, { "epoch": 2.354758828596038, "grad_norm": 0.3727887272834778, "learning_rate": 0.0005650639558368116, "loss": 3.4938, "step": 8750 }, { "epoch": 2.3682170542635657, "grad_norm": 0.34845060110092163, "learning_rate": 0.0005648619900363538, "loss": 3.5106, "step": 8800 }, { "epoch": 2.381675279931094, "grad_norm": 0.3758419454097748, "learning_rate": 0.000564660024235896, "loss": 3.4994, "step": 8850 }, { "epoch": 2.395133505598622, "grad_norm": 0.34191444516181946, "learning_rate": 0.0005644580584354382, "loss": 3.4965, "step": 8900 }, { "epoch": 2.40859173126615, "grad_norm": 0.36028486490249634, "learning_rate": 0.0005642560926349804, "loss": 3.4893, "step": 8950 }, { "epoch": 2.422049956933678, "grad_norm": 0.37876632809638977, "learning_rate": 0.0005640541268345226, "loss": 3.4969, "step": 9000 }, { "epoch": 2.422049956933678, "eval_accuracy": 0.36681166676010296, "eval_loss": 3.508786916732788, "eval_runtime": 89.0481, "eval_samples_per_second": 202.273, "eval_steps_per_second": 12.645, "step": 9000 }, { "epoch": 2.4355081826012057, "grad_norm": 0.35556432604789734, "learning_rate": 0.0005638521610340649, "loss": 3.5017, "step": 9050 }, { "epoch": 2.448966408268734, "grad_norm": 0.35798487067222595, "learning_rate": 0.000563650195233607, "loss": 3.4861, "step": 9100 }, { "epoch": 2.4624246339362617, "grad_norm": 0.3988047242164612, "learning_rate": 0.0005634482294331493, "loss": 3.4987, "step": 9150 }, { "epoch": 2.47588285960379, "grad_norm": 0.35068902373313904, "learning_rate": 0.0005632462636326914, "loss": 3.4886, "step": 9200 }, { "epoch": 2.4893410852713176, "grad_norm": 0.3887159526348114, "learning_rate": 0.0005630442978322337, "loss": 3.4944, "step": 9250 }, { "epoch": 2.502799310938846, "grad_norm": 0.3577377200126648, "learning_rate": 0.0005628423320317759, "loss": 3.4902, "step": 9300 }, { "epoch": 2.516257536606374, "grad_norm": 0.3493818938732147, "learning_rate": 0.0005626403662313182, "loss": 3.474, "step": 9350 }, { "epoch": 2.5297157622739017, "grad_norm": 0.35755589604377747, "learning_rate": 0.0005624384004308603, "loss": 3.4943, "step": 9400 }, { "epoch": 2.54317398794143, "grad_norm": 0.34538453817367554, "learning_rate": 0.0005622364346304026, "loss": 3.4928, "step": 9450 }, { "epoch": 2.5566322136089576, "grad_norm": 0.37912717461586, "learning_rate": 0.0005620344688299448, "loss": 3.4964, "step": 9500 }, { "epoch": 2.570090439276486, "grad_norm": 0.4231334328651428, "learning_rate": 0.0005618325030294869, "loss": 3.5014, "step": 9550 }, { "epoch": 2.583548664944014, "grad_norm": 0.35249027609825134, "learning_rate": 0.0005616305372290292, "loss": 3.4989, "step": 9600 }, { "epoch": 2.5970068906115418, "grad_norm": 0.3673330247402191, "learning_rate": 0.0005614285714285713, "loss": 3.4933, "step": 9650 }, { "epoch": 2.6104651162790695, "grad_norm": 0.36924853920936584, "learning_rate": 0.0005612266056281136, "loss": 3.4852, "step": 9700 }, { "epoch": 2.6239233419465977, "grad_norm": 0.37280353903770447, "learning_rate": 0.0005610246398276558, "loss": 3.4825, "step": 9750 }, { "epoch": 2.637381567614126, "grad_norm": 0.3669414818286896, "learning_rate": 0.000560822674027198, "loss": 3.4856, "step": 9800 }, { "epoch": 2.6508397932816536, "grad_norm": 0.33487293124198914, "learning_rate": 0.0005606207082267402, "loss": 3.4863, "step": 9850 }, { "epoch": 2.664298018949182, "grad_norm": 0.42584335803985596, "learning_rate": 0.0005604187424262824, "loss": 3.485, "step": 9900 }, { "epoch": 2.6777562446167096, "grad_norm": 0.3721439242362976, "learning_rate": 0.0005602167766258246, "loss": 3.473, "step": 9950 }, { "epoch": 2.6912144702842378, "grad_norm": 0.3362332880496979, "learning_rate": 0.0005600148108253669, "loss": 3.4725, "step": 10000 }, { "epoch": 2.6912144702842378, "eval_accuracy": 0.3691482260358717, "eval_loss": 3.4851438999176025, "eval_runtime": 88.9071, "eval_samples_per_second": 202.593, "eval_steps_per_second": 12.665, "step": 10000 }, { "epoch": 2.704672695951766, "grad_norm": 0.4206249713897705, "learning_rate": 0.000559812845024909, "loss": 3.4844, "step": 10050 }, { "epoch": 2.7181309216192937, "grad_norm": 0.3479953706264496, "learning_rate": 0.0005596108792244513, "loss": 3.4837, "step": 10100 }, { "epoch": 2.7315891472868215, "grad_norm": 0.379986047744751, "learning_rate": 0.0005594089134239935, "loss": 3.4661, "step": 10150 }, { "epoch": 2.7450473729543496, "grad_norm": 0.35398250818252563, "learning_rate": 0.0005592069476235358, "loss": 3.4803, "step": 10200 }, { "epoch": 2.758505598621878, "grad_norm": 0.3484811782836914, "learning_rate": 0.0005590049818230779, "loss": 3.4786, "step": 10250 }, { "epoch": 2.7719638242894056, "grad_norm": 0.3662926256656647, "learning_rate": 0.0005588030160226202, "loss": 3.4711, "step": 10300 }, { "epoch": 2.7854220499569338, "grad_norm": 0.3605268597602844, "learning_rate": 0.0005586010502221623, "loss": 3.4897, "step": 10350 }, { "epoch": 2.7988802756244615, "grad_norm": 0.34791073203086853, "learning_rate": 0.0005583990844217045, "loss": 3.4742, "step": 10400 }, { "epoch": 2.8123385012919897, "grad_norm": 0.34151312708854675, "learning_rate": 0.0005581971186212468, "loss": 3.4723, "step": 10450 }, { "epoch": 2.825796726959518, "grad_norm": 0.3514784574508667, "learning_rate": 0.0005579951528207889, "loss": 3.4734, "step": 10500 }, { "epoch": 2.8392549526270456, "grad_norm": 0.35521063208580017, "learning_rate": 0.0005577931870203312, "loss": 3.4809, "step": 10550 }, { "epoch": 2.8527131782945734, "grad_norm": 0.35965123772621155, "learning_rate": 0.0005575912212198733, "loss": 3.4683, "step": 10600 }, { "epoch": 2.8661714039621016, "grad_norm": 0.37720105051994324, "learning_rate": 0.0005573892554194156, "loss": 3.4777, "step": 10650 }, { "epoch": 2.8796296296296298, "grad_norm": 0.35280463099479675, "learning_rate": 0.0005571872896189578, "loss": 3.4709, "step": 10700 }, { "epoch": 2.8930878552971575, "grad_norm": 0.34837934374809265, "learning_rate": 0.0005569853238185, "loss": 3.4713, "step": 10750 }, { "epoch": 2.9065460809646857, "grad_norm": 0.3864405155181885, "learning_rate": 0.0005567833580180422, "loss": 3.4581, "step": 10800 }, { "epoch": 2.9200043066322134, "grad_norm": 0.3737727701663971, "learning_rate": 0.0005565813922175844, "loss": 3.4633, "step": 10850 }, { "epoch": 2.9334625322997416, "grad_norm": 0.3452976942062378, "learning_rate": 0.0005563794264171266, "loss": 3.4727, "step": 10900 }, { "epoch": 2.94692075796727, "grad_norm": 0.3639717400074005, "learning_rate": 0.0005561774606166689, "loss": 3.464, "step": 10950 }, { "epoch": 2.9603789836347976, "grad_norm": 0.42750686407089233, "learning_rate": 0.000555975494816211, "loss": 3.4718, "step": 11000 }, { "epoch": 2.9603789836347976, "eval_accuracy": 0.3713497372701739, "eval_loss": 3.46138334274292, "eval_runtime": 89.357, "eval_samples_per_second": 201.574, "eval_steps_per_second": 12.601, "step": 11000 }, { "epoch": 2.9738372093023253, "grad_norm": 0.34111878275871277, "learning_rate": 0.0005557735290157534, "loss": 3.4719, "step": 11050 }, { "epoch": 2.9872954349698535, "grad_norm": 0.3508542478084564, "learning_rate": 0.0005555715632152955, "loss": 3.4659, "step": 11100 }, { "epoch": 3.000538329026701, "grad_norm": 0.4099224805831909, "learning_rate": 0.0005553695974148378, "loss": 3.4607, "step": 11150 }, { "epoch": 3.013996554694229, "grad_norm": 0.3317889869213104, "learning_rate": 0.0005551676316143799, "loss": 3.3802, "step": 11200 }, { "epoch": 3.027454780361757, "grad_norm": 0.349567174911499, "learning_rate": 0.0005549656658139222, "loss": 3.3764, "step": 11250 }, { "epoch": 3.0409130060292853, "grad_norm": 0.38516244292259216, "learning_rate": 0.0005547637000134644, "loss": 3.3685, "step": 11300 }, { "epoch": 3.054371231696813, "grad_norm": 0.33860960602760315, "learning_rate": 0.0005545617342130065, "loss": 3.3929, "step": 11350 }, { "epoch": 3.067829457364341, "grad_norm": 0.37568598985671997, "learning_rate": 0.0005543597684125488, "loss": 3.3781, "step": 11400 }, { "epoch": 3.081287683031869, "grad_norm": 0.36252379417419434, "learning_rate": 0.0005541578026120909, "loss": 3.3823, "step": 11450 }, { "epoch": 3.094745908699397, "grad_norm": 0.3808792233467102, "learning_rate": 0.0005539558368116332, "loss": 3.3856, "step": 11500 }, { "epoch": 3.108204134366925, "grad_norm": 0.3400123715400696, "learning_rate": 0.0005537538710111754, "loss": 3.3834, "step": 11550 }, { "epoch": 3.121662360034453, "grad_norm": 0.365142822265625, "learning_rate": 0.0005535519052107176, "loss": 3.392, "step": 11600 }, { "epoch": 3.135120585701981, "grad_norm": 0.3488098382949829, "learning_rate": 0.0005533499394102598, "loss": 3.3778, "step": 11650 }, { "epoch": 3.148578811369509, "grad_norm": 0.31865015625953674, "learning_rate": 0.000553147973609802, "loss": 3.391, "step": 11700 }, { "epoch": 3.162037037037037, "grad_norm": 0.36687901616096497, "learning_rate": 0.0005529460078093442, "loss": 3.3963, "step": 11750 }, { "epoch": 3.175495262704565, "grad_norm": 0.3641960024833679, "learning_rate": 0.0005527440420088865, "loss": 3.3926, "step": 11800 }, { "epoch": 3.188953488372093, "grad_norm": 0.3725584149360657, "learning_rate": 0.0005525420762084286, "loss": 3.3953, "step": 11850 }, { "epoch": 3.202411714039621, "grad_norm": 0.40352463722229004, "learning_rate": 0.0005523401104079709, "loss": 3.396, "step": 11900 }, { "epoch": 3.215869939707149, "grad_norm": 0.37560948729515076, "learning_rate": 0.0005521381446075131, "loss": 3.3956, "step": 11950 }, { "epoch": 3.229328165374677, "grad_norm": 0.3621182143688202, "learning_rate": 0.0005519361788070554, "loss": 3.3896, "step": 12000 }, { "epoch": 3.229328165374677, "eval_accuracy": 0.3730891734277605, "eval_loss": 3.452282190322876, "eval_runtime": 89.6688, "eval_samples_per_second": 200.872, "eval_steps_per_second": 12.557, "step": 12000 }, { "epoch": 3.242786391042205, "grad_norm": 0.35463830828666687, "learning_rate": 0.0005517342130065975, "loss": 3.3955, "step": 12050 }, { "epoch": 3.2562446167097328, "grad_norm": 0.3728417158126831, "learning_rate": 0.0005515322472061398, "loss": 3.4032, "step": 12100 }, { "epoch": 3.269702842377261, "grad_norm": 0.3544614613056183, "learning_rate": 0.0005513302814056819, "loss": 3.3998, "step": 12150 }, { "epoch": 3.283161068044789, "grad_norm": 0.34991875290870667, "learning_rate": 0.0005511283156052242, "loss": 3.3919, "step": 12200 }, { "epoch": 3.296619293712317, "grad_norm": 0.3334318995475769, "learning_rate": 0.0005509263498047664, "loss": 3.3917, "step": 12250 }, { "epoch": 3.310077519379845, "grad_norm": 0.3585456609725952, "learning_rate": 0.0005507243840043085, "loss": 3.4012, "step": 12300 }, { "epoch": 3.323535745047373, "grad_norm": 0.36692893505096436, "learning_rate": 0.0005505224182038508, "loss": 3.3976, "step": 12350 }, { "epoch": 3.336993970714901, "grad_norm": 0.3426796793937683, "learning_rate": 0.0005503204524033929, "loss": 3.3981, "step": 12400 }, { "epoch": 3.3504521963824287, "grad_norm": 0.3477403521537781, "learning_rate": 0.0005501184866029352, "loss": 3.3912, "step": 12450 }, { "epoch": 3.363910422049957, "grad_norm": 0.35531580448150635, "learning_rate": 0.0005499165208024774, "loss": 3.3949, "step": 12500 }, { "epoch": 3.3773686477174847, "grad_norm": 0.334770143032074, "learning_rate": 0.0005497145550020196, "loss": 3.3899, "step": 12550 }, { "epoch": 3.390826873385013, "grad_norm": 0.35484474897384644, "learning_rate": 0.0005495125892015618, "loss": 3.3913, "step": 12600 }, { "epoch": 3.404285099052541, "grad_norm": 0.3469596207141876, "learning_rate": 0.000549310623401104, "loss": 3.3932, "step": 12650 }, { "epoch": 3.417743324720069, "grad_norm": 0.37239179015159607, "learning_rate": 0.0005491086576006462, "loss": 3.4103, "step": 12700 }, { "epoch": 3.431201550387597, "grad_norm": 0.3561572730541229, "learning_rate": 0.0005489066918001885, "loss": 3.3926, "step": 12750 }, { "epoch": 3.4446597760551247, "grad_norm": 0.3374239206314087, "learning_rate": 0.0005487047259997306, "loss": 3.4029, "step": 12800 }, { "epoch": 3.458118001722653, "grad_norm": 0.3496391773223877, "learning_rate": 0.0005485027601992729, "loss": 3.3964, "step": 12850 }, { "epoch": 3.471576227390181, "grad_norm": 0.3446263372898102, "learning_rate": 0.0005483007943988151, "loss": 3.3918, "step": 12900 }, { "epoch": 3.485034453057709, "grad_norm": 0.32869866490364075, "learning_rate": 0.0005480988285983574, "loss": 3.4067, "step": 12950 }, { "epoch": 3.4984926787252366, "grad_norm": 0.3769306242465973, "learning_rate": 0.0005478968627978995, "loss": 3.3969, "step": 13000 }, { "epoch": 3.4984926787252366, "eval_accuracy": 0.37452874426398924, "eval_loss": 3.4374747276306152, "eval_runtime": 88.8675, "eval_samples_per_second": 202.684, "eval_steps_per_second": 12.671, "step": 13000 }, { "epoch": 3.511950904392765, "grad_norm": 0.3424180746078491, "learning_rate": 0.0005476948969974418, "loss": 3.4045, "step": 13050 }, { "epoch": 3.525409130060293, "grad_norm": 0.3807520270347595, "learning_rate": 0.0005474929311969839, "loss": 3.3979, "step": 13100 }, { "epoch": 3.5388673557278207, "grad_norm": 0.4000932574272156, "learning_rate": 0.0005472909653965261, "loss": 3.4005, "step": 13150 }, { "epoch": 3.552325581395349, "grad_norm": 0.342479944229126, "learning_rate": 0.0005470889995960684, "loss": 3.3852, "step": 13200 }, { "epoch": 3.5657838070628767, "grad_norm": 0.3163456320762634, "learning_rate": 0.0005468870337956105, "loss": 3.3971, "step": 13250 }, { "epoch": 3.579242032730405, "grad_norm": 0.36020761728286743, "learning_rate": 0.0005466850679951528, "loss": 3.3911, "step": 13300 }, { "epoch": 3.592700258397933, "grad_norm": 0.34629690647125244, "learning_rate": 0.0005464831021946949, "loss": 3.3972, "step": 13350 }, { "epoch": 3.606158484065461, "grad_norm": 0.3748731315135956, "learning_rate": 0.0005462811363942372, "loss": 3.3925, "step": 13400 }, { "epoch": 3.6196167097329885, "grad_norm": 0.34598422050476074, "learning_rate": 0.0005460791705937794, "loss": 3.3977, "step": 13450 }, { "epoch": 3.6330749354005167, "grad_norm": 0.33285003900527954, "learning_rate": 0.0005458772047933216, "loss": 3.3862, "step": 13500 }, { "epoch": 3.646533161068045, "grad_norm": 0.36419209837913513, "learning_rate": 0.0005456752389928638, "loss": 3.3893, "step": 13550 }, { "epoch": 3.6599913867355727, "grad_norm": 0.3598211407661438, "learning_rate": 0.000545473273192406, "loss": 3.4047, "step": 13600 }, { "epoch": 3.673449612403101, "grad_norm": 0.3460333049297333, "learning_rate": 0.0005452713073919482, "loss": 3.395, "step": 13650 }, { "epoch": 3.6869078380706286, "grad_norm": 0.3223993182182312, "learning_rate": 0.0005450693415914905, "loss": 3.3951, "step": 13700 }, { "epoch": 3.700366063738157, "grad_norm": 0.367910236120224, "learning_rate": 0.0005448673757910327, "loss": 3.3963, "step": 13750 }, { "epoch": 3.713824289405685, "grad_norm": 0.31978780031204224, "learning_rate": 0.0005446654099905749, "loss": 3.3843, "step": 13800 }, { "epoch": 3.7272825150732127, "grad_norm": 0.37149783968925476, "learning_rate": 0.0005444634441901171, "loss": 3.3958, "step": 13850 }, { "epoch": 3.7407407407407405, "grad_norm": 0.35949620604515076, "learning_rate": 0.0005442614783896594, "loss": 3.3863, "step": 13900 }, { "epoch": 3.7541989664082687, "grad_norm": 0.355377197265625, "learning_rate": 0.0005440595125892015, "loss": 3.3976, "step": 13950 }, { "epoch": 3.767657192075797, "grad_norm": 0.32358214259147644, "learning_rate": 0.0005438575467887438, "loss": 3.3913, "step": 14000 }, { "epoch": 3.767657192075797, "eval_accuracy": 0.37618463098964683, "eval_loss": 3.420905590057373, "eval_runtime": 89.8357, "eval_samples_per_second": 200.499, "eval_steps_per_second": 12.534, "step": 14000 }, { "epoch": 3.7811154177433246, "grad_norm": 0.34888753294944763, "learning_rate": 0.0005436555809882859, "loss": 3.3917, "step": 14050 }, { "epoch": 3.794573643410853, "grad_norm": 0.3531152307987213, "learning_rate": 0.0005434536151878281, "loss": 3.3893, "step": 14100 }, { "epoch": 3.8080318690783805, "grad_norm": 0.3416253924369812, "learning_rate": 0.0005432516493873704, "loss": 3.3961, "step": 14150 }, { "epoch": 3.8214900947459087, "grad_norm": 0.3324134945869446, "learning_rate": 0.0005430496835869125, "loss": 3.387, "step": 14200 }, { "epoch": 3.834948320413437, "grad_norm": 0.3525480628013611, "learning_rate": 0.0005428477177864548, "loss": 3.3958, "step": 14250 }, { "epoch": 3.8484065460809647, "grad_norm": 0.3697124123573303, "learning_rate": 0.0005426457519859969, "loss": 3.4013, "step": 14300 }, { "epoch": 3.8618647717484924, "grad_norm": 0.3347870409488678, "learning_rate": 0.0005424437861855392, "loss": 3.3922, "step": 14350 }, { "epoch": 3.8753229974160206, "grad_norm": 0.34625548124313354, "learning_rate": 0.0005422418203850814, "loss": 3.3933, "step": 14400 }, { "epoch": 3.888781223083549, "grad_norm": 0.3729144334793091, "learning_rate": 0.0005420398545846236, "loss": 3.3915, "step": 14450 }, { "epoch": 3.9022394487510765, "grad_norm": 0.3475784361362457, "learning_rate": 0.0005418378887841658, "loss": 3.3944, "step": 14500 }, { "epoch": 3.9156976744186047, "grad_norm": 0.3474798798561096, "learning_rate": 0.000541635922983708, "loss": 3.4017, "step": 14550 }, { "epoch": 3.9291559000861325, "grad_norm": 0.3410467505455017, "learning_rate": 0.0005414339571832503, "loss": 3.386, "step": 14600 }, { "epoch": 3.9426141257536607, "grad_norm": 0.34732016921043396, "learning_rate": 0.0005412319913827925, "loss": 3.394, "step": 14650 }, { "epoch": 3.956072351421189, "grad_norm": 0.32139548659324646, "learning_rate": 0.0005410300255823347, "loss": 3.385, "step": 14700 }, { "epoch": 3.9695305770887166, "grad_norm": 0.328720360994339, "learning_rate": 0.0005408280597818769, "loss": 3.3803, "step": 14750 }, { "epoch": 3.9829888027562443, "grad_norm": 0.35579606890678406, "learning_rate": 0.0005406260939814191, "loss": 3.4056, "step": 14800 }, { "epoch": 3.9964470284237725, "grad_norm": 0.3351224660873413, "learning_rate": 0.0005404241281809614, "loss": 3.3863, "step": 14850 }, { "epoch": 4.00968992248062, "grad_norm": 0.3696632981300354, "learning_rate": 0.0005402221623805035, "loss": 3.3029, "step": 14900 }, { "epoch": 4.023148148148148, "grad_norm": 0.33017241954803467, "learning_rate": 0.0005400201965800457, "loss": 3.2878, "step": 14950 }, { "epoch": 4.036606373815676, "grad_norm": 0.34758812189102173, "learning_rate": 0.000539818230779588, "loss": 3.2989, "step": 15000 }, { "epoch": 4.036606373815676, "eval_accuracy": 0.3775983438742513, "eval_loss": 3.4153592586517334, "eval_runtime": 89.1199, "eval_samples_per_second": 202.11, "eval_steps_per_second": 12.635, "step": 15000 }, { "epoch": 4.050064599483204, "grad_norm": 0.33058854937553406, "learning_rate": 0.0005396162649791301, "loss": 3.3013, "step": 15050 }, { "epoch": 4.0635228251507325, "grad_norm": 0.35816290974617004, "learning_rate": 0.0005394142991786724, "loss": 3.2947, "step": 15100 }, { "epoch": 4.07698105081826, "grad_norm": 0.3572397530078888, "learning_rate": 0.0005392123333782145, "loss": 3.3099, "step": 15150 }, { "epoch": 4.090439276485788, "grad_norm": 0.36381787061691284, "learning_rate": 0.0005390103675777568, "loss": 3.3076, "step": 15200 }, { "epoch": 4.103897502153316, "grad_norm": 0.39142516255378723, "learning_rate": 0.000538808401777299, "loss": 3.3173, "step": 15250 }, { "epoch": 4.117355727820844, "grad_norm": 0.3246189057826996, "learning_rate": 0.0005386064359768412, "loss": 3.3056, "step": 15300 }, { "epoch": 4.1308139534883725, "grad_norm": 0.3524934649467468, "learning_rate": 0.0005384044701763834, "loss": 3.3093, "step": 15350 }, { "epoch": 4.1442721791559, "grad_norm": 0.3528584837913513, "learning_rate": 0.0005382025043759256, "loss": 3.2963, "step": 15400 }, { "epoch": 4.157730404823428, "grad_norm": 0.3665054142475128, "learning_rate": 0.0005380005385754678, "loss": 3.313, "step": 15450 }, { "epoch": 4.171188630490956, "grad_norm": 0.36652475595474243, "learning_rate": 0.0005377985727750101, "loss": 3.3135, "step": 15500 }, { "epoch": 4.184646856158484, "grad_norm": 0.3758777976036072, "learning_rate": 0.0005375966069745523, "loss": 3.3163, "step": 15550 }, { "epoch": 4.198105081826012, "grad_norm": 0.33912044763565063, "learning_rate": 0.0005373946411740945, "loss": 3.3075, "step": 15600 }, { "epoch": 4.21156330749354, "grad_norm": 0.332670122385025, "learning_rate": 0.0005371926753736367, "loss": 3.3225, "step": 15650 }, { "epoch": 4.225021533161068, "grad_norm": 0.3336251974105835, "learning_rate": 0.000536990709573179, "loss": 3.3212, "step": 15700 }, { "epoch": 4.238479758828596, "grad_norm": 0.35916876792907715, "learning_rate": 0.0005367887437727211, "loss": 3.3289, "step": 15750 }, { "epoch": 4.251937984496124, "grad_norm": 0.32652392983436584, "learning_rate": 0.0005365867779722634, "loss": 3.329, "step": 15800 }, { "epoch": 4.265396210163652, "grad_norm": 0.34005942940711975, "learning_rate": 0.0005363848121718055, "loss": 3.3379, "step": 15850 }, { "epoch": 4.27885443583118, "grad_norm": 0.3562110662460327, "learning_rate": 0.0005361828463713477, "loss": 3.3135, "step": 15900 }, { "epoch": 4.292312661498708, "grad_norm": 0.36522018909454346, "learning_rate": 0.00053598088057089, "loss": 3.3163, "step": 15950 }, { "epoch": 4.305770887166236, "grad_norm": 0.35446590185165405, "learning_rate": 0.0005357789147704321, "loss": 3.3334, "step": 16000 }, { "epoch": 4.305770887166236, "eval_accuracy": 0.37851575792263736, "eval_loss": 3.4047274589538574, "eval_runtime": 88.4624, "eval_samples_per_second": 203.612, "eval_steps_per_second": 12.729, "step": 16000 }, { "epoch": 4.319229112833764, "grad_norm": 0.3302324712276459, "learning_rate": 0.0005355769489699744, "loss": 3.3239, "step": 16050 }, { "epoch": 4.332687338501292, "grad_norm": 0.3508687913417816, "learning_rate": 0.0005353749831695165, "loss": 3.3308, "step": 16100 }, { "epoch": 4.34614556416882, "grad_norm": 0.34046798944473267, "learning_rate": 0.0005351730173690588, "loss": 3.3358, "step": 16150 }, { "epoch": 4.359603789836348, "grad_norm": 0.3654434382915497, "learning_rate": 0.000534971051568601, "loss": 3.3117, "step": 16200 }, { "epoch": 4.373062015503876, "grad_norm": 0.35325589776039124, "learning_rate": 0.0005347690857681432, "loss": 3.3336, "step": 16250 }, { "epoch": 4.386520241171404, "grad_norm": 0.3389158546924591, "learning_rate": 0.0005345671199676854, "loss": 3.325, "step": 16300 }, { "epoch": 4.399978466838932, "grad_norm": 0.3626989722251892, "learning_rate": 0.0005343651541672276, "loss": 3.3272, "step": 16350 }, { "epoch": 4.41343669250646, "grad_norm": 0.34801793098449707, "learning_rate": 0.0005341631883667699, "loss": 3.3257, "step": 16400 }, { "epoch": 4.426894918173988, "grad_norm": 0.321744441986084, "learning_rate": 0.0005339612225663121, "loss": 3.335, "step": 16450 }, { "epoch": 4.440353143841516, "grad_norm": 0.34780287742614746, "learning_rate": 0.0005337592567658543, "loss": 3.3402, "step": 16500 }, { "epoch": 4.453811369509044, "grad_norm": 0.36421075463294983, "learning_rate": 0.0005335572909653965, "loss": 3.3383, "step": 16550 }, { "epoch": 4.467269595176572, "grad_norm": 0.3883045017719269, "learning_rate": 0.0005333553251649387, "loss": 3.3352, "step": 16600 }, { "epoch": 4.4807278208441, "grad_norm": 0.3691977262496948, "learning_rate": 0.000533153359364481, "loss": 3.3244, "step": 16650 }, { "epoch": 4.4941860465116275, "grad_norm": 0.35588181018829346, "learning_rate": 0.0005329513935640231, "loss": 3.3301, "step": 16700 }, { "epoch": 4.507644272179156, "grad_norm": 0.33899396657943726, "learning_rate": 0.0005327494277635654, "loss": 3.3302, "step": 16750 }, { "epoch": 4.521102497846684, "grad_norm": 0.351391464471817, "learning_rate": 0.0005325474619631075, "loss": 3.3355, "step": 16800 }, { "epoch": 4.534560723514212, "grad_norm": 0.33637428283691406, "learning_rate": 0.0005323454961626497, "loss": 3.3384, "step": 16850 }, { "epoch": 4.54801894918174, "grad_norm": 0.34806662797927856, "learning_rate": 0.000532143530362192, "loss": 3.3453, "step": 16900 }, { "epoch": 4.5614771748492675, "grad_norm": 0.32941877841949463, "learning_rate": 0.0005319415645617341, "loss": 3.3331, "step": 16950 }, { "epoch": 4.574935400516796, "grad_norm": 0.3837328851222992, "learning_rate": 0.0005317395987612764, "loss": 3.3276, "step": 17000 }, { "epoch": 4.574935400516796, "eval_accuracy": 0.3797251060719251, "eval_loss": 3.393202543258667, "eval_runtime": 88.7513, "eval_samples_per_second": 202.949, "eval_steps_per_second": 12.687, "step": 17000 }, { "epoch": 4.588393626184324, "grad_norm": 0.34744006395339966, "learning_rate": 0.0005315376329608185, "loss": 3.3315, "step": 17050 }, { "epoch": 4.601851851851852, "grad_norm": 0.3779725432395935, "learning_rate": 0.0005313356671603608, "loss": 3.3307, "step": 17100 }, { "epoch": 4.61531007751938, "grad_norm": 0.3435402512550354, "learning_rate": 0.000531133701359903, "loss": 3.3437, "step": 17150 }, { "epoch": 4.628768303186908, "grad_norm": 0.35903772711753845, "learning_rate": 0.0005309317355594452, "loss": 3.3371, "step": 17200 }, { "epoch": 4.642226528854436, "grad_norm": 0.3584959805011749, "learning_rate": 0.0005307297697589874, "loss": 3.3397, "step": 17250 }, { "epoch": 4.655684754521964, "grad_norm": 0.3358038663864136, "learning_rate": 0.0005305278039585297, "loss": 3.3294, "step": 17300 }, { "epoch": 4.669142980189492, "grad_norm": 0.3338400721549988, "learning_rate": 0.0005303258381580719, "loss": 3.3274, "step": 17350 }, { "epoch": 4.682601205857019, "grad_norm": 0.3644915819168091, "learning_rate": 0.0005301238723576141, "loss": 3.3364, "step": 17400 }, { "epoch": 4.696059431524548, "grad_norm": 0.3266715407371521, "learning_rate": 0.0005299219065571563, "loss": 3.3418, "step": 17450 }, { "epoch": 4.709517657192076, "grad_norm": 0.3627817630767822, "learning_rate": 0.0005297199407566985, "loss": 3.3476, "step": 17500 }, { "epoch": 4.722975882859604, "grad_norm": 0.3106552064418793, "learning_rate": 0.0005295179749562407, "loss": 3.3472, "step": 17550 }, { "epoch": 4.736434108527131, "grad_norm": 0.36165159940719604, "learning_rate": 0.000529316009155783, "loss": 3.3336, "step": 17600 }, { "epoch": 4.7498923341946595, "grad_norm": 0.3304913341999054, "learning_rate": 0.0005291140433553251, "loss": 3.3463, "step": 17650 }, { "epoch": 4.763350559862188, "grad_norm": 0.3533960282802582, "learning_rate": 0.0005289120775548673, "loss": 3.3416, "step": 17700 }, { "epoch": 4.776808785529716, "grad_norm": 0.3458789587020874, "learning_rate": 0.0005287101117544095, "loss": 3.3445, "step": 17750 }, { "epoch": 4.790267011197244, "grad_norm": 0.32624074816703796, "learning_rate": 0.0005285081459539517, "loss": 3.3296, "step": 17800 }, { "epoch": 4.803725236864771, "grad_norm": 0.3615491986274719, "learning_rate": 0.000528306180153494, "loss": 3.3306, "step": 17850 }, { "epoch": 4.8171834625323, "grad_norm": 0.3486765921115875, "learning_rate": 0.0005281042143530361, "loss": 3.3317, "step": 17900 }, { "epoch": 4.830641688199828, "grad_norm": 0.3309303820133209, "learning_rate": 0.0005279022485525784, "loss": 3.3415, "step": 17950 }, { "epoch": 4.844099913867356, "grad_norm": 0.34980034828186035, "learning_rate": 0.0005277002827521205, "loss": 3.3278, "step": 18000 }, { "epoch": 4.844099913867356, "eval_accuracy": 0.3803980646952912, "eval_loss": 3.3822619915008545, "eval_runtime": 84.7527, "eval_samples_per_second": 212.524, "eval_steps_per_second": 13.286, "step": 18000 }, { "epoch": 4.857558139534884, "grad_norm": 0.31751900911331177, "learning_rate": 0.0005274983169516628, "loss": 3.3359, "step": 18050 }, { "epoch": 4.871016365202411, "grad_norm": 0.4001691937446594, "learning_rate": 0.000527296351151205, "loss": 3.3373, "step": 18100 }, { "epoch": 4.88447459086994, "grad_norm": 0.32267269492149353, "learning_rate": 0.0005270943853507472, "loss": 3.3312, "step": 18150 }, { "epoch": 4.897932816537468, "grad_norm": 0.3582165837287903, "learning_rate": 0.0005268924195502894, "loss": 3.3349, "step": 18200 }, { "epoch": 4.911391042204996, "grad_norm": 0.3568687438964844, "learning_rate": 0.0005266904537498317, "loss": 3.3309, "step": 18250 }, { "epoch": 4.924849267872523, "grad_norm": 0.3639770448207855, "learning_rate": 0.0005264884879493739, "loss": 3.3262, "step": 18300 }, { "epoch": 4.9383074935400515, "grad_norm": 0.321569561958313, "learning_rate": 0.0005262865221489161, "loss": 3.3273, "step": 18350 }, { "epoch": 4.95176571920758, "grad_norm": 0.3282260000705719, "learning_rate": 0.0005260845563484583, "loss": 3.3389, "step": 18400 }, { "epoch": 4.965223944875108, "grad_norm": 0.34112313389778137, "learning_rate": 0.0005258825905480005, "loss": 3.3316, "step": 18450 }, { "epoch": 4.978682170542635, "grad_norm": 0.3521008789539337, "learning_rate": 0.0005256806247475427, "loss": 3.3318, "step": 18500 }, { "epoch": 4.992140396210163, "grad_norm": 0.3335284888744354, "learning_rate": 0.000525478658947085, "loss": 3.3378, "step": 18550 }, { "epoch": 5.0053832902670115, "grad_norm": 0.34815871715545654, "learning_rate": 0.0005252766931466271, "loss": 3.3033, "step": 18600 }, { "epoch": 5.018841515934539, "grad_norm": 0.3716835677623749, "learning_rate": 0.0005250747273461693, "loss": 3.2465, "step": 18650 }, { "epoch": 5.032299741602067, "grad_norm": 0.34348708391189575, "learning_rate": 0.0005248727615457115, "loss": 3.22, "step": 18700 }, { "epoch": 5.045757967269595, "grad_norm": 0.34355980157852173, "learning_rate": 0.0005246707957452537, "loss": 3.2468, "step": 18750 }, { "epoch": 5.059216192937123, "grad_norm": 0.3393841087818146, "learning_rate": 0.000524468829944796, "loss": 3.2384, "step": 18800 }, { "epoch": 5.0726744186046515, "grad_norm": 0.37807780504226685, "learning_rate": 0.0005242668641443381, "loss": 3.2484, "step": 18850 }, { "epoch": 5.086132644272179, "grad_norm": 0.3223433494567871, "learning_rate": 0.0005240648983438804, "loss": 3.2499, "step": 18900 }, { "epoch": 5.099590869939707, "grad_norm": 0.3463059961795807, "learning_rate": 0.0005238629325434225, "loss": 3.2496, "step": 18950 }, { "epoch": 5.113049095607235, "grad_norm": 0.3511742055416107, "learning_rate": 0.0005236609667429648, "loss": 3.2588, "step": 19000 }, { "epoch": 5.113049095607235, "eval_accuracy": 0.38150713179689294, "eval_loss": 3.3810367584228516, "eval_runtime": 84.1776, "eval_samples_per_second": 213.976, "eval_steps_per_second": 13.376, "step": 19000 }, { "epoch": 5.126507321274763, "grad_norm": 0.34395796060562134, "learning_rate": 0.000523459000942507, "loss": 3.2508, "step": 19050 }, { "epoch": 5.139965546942291, "grad_norm": 0.3511422872543335, "learning_rate": 0.0005232570351420493, "loss": 3.2636, "step": 19100 }, { "epoch": 5.153423772609819, "grad_norm": 0.3584088087081909, "learning_rate": 0.0005230550693415915, "loss": 3.2743, "step": 19150 }, { "epoch": 5.166881998277347, "grad_norm": 0.36160993576049805, "learning_rate": 0.0005228531035411337, "loss": 3.2551, "step": 19200 }, { "epoch": 5.180340223944875, "grad_norm": 0.33585062623023987, "learning_rate": 0.0005226511377406759, "loss": 3.2666, "step": 19250 }, { "epoch": 5.1937984496124034, "grad_norm": 0.37425586581230164, "learning_rate": 0.0005224491719402181, "loss": 3.2686, "step": 19300 }, { "epoch": 5.207256675279931, "grad_norm": 0.3439202606678009, "learning_rate": 0.0005222472061397603, "loss": 3.2681, "step": 19350 }, { "epoch": 5.220714900947459, "grad_norm": 0.3334115743637085, "learning_rate": 0.0005220452403393026, "loss": 3.2665, "step": 19400 }, { "epoch": 5.234173126614987, "grad_norm": 0.3346470892429352, "learning_rate": 0.0005218432745388447, "loss": 3.2712, "step": 19450 }, { "epoch": 5.247631352282515, "grad_norm": 0.3484283983707428, "learning_rate": 0.000521641308738387, "loss": 3.268, "step": 19500 }, { "epoch": 5.2610895779500435, "grad_norm": 0.3476269841194153, "learning_rate": 0.0005214393429379291, "loss": 3.2694, "step": 19550 }, { "epoch": 5.274547803617571, "grad_norm": 0.33217039704322815, "learning_rate": 0.0005212373771374713, "loss": 3.279, "step": 19600 }, { "epoch": 5.288006029285099, "grad_norm": 0.3590564429759979, "learning_rate": 0.0005210354113370136, "loss": 3.2609, "step": 19650 }, { "epoch": 5.301464254952627, "grad_norm": 0.3503280580043793, "learning_rate": 0.0005208334455365557, "loss": 3.2769, "step": 19700 }, { "epoch": 5.314922480620155, "grad_norm": 0.3696460723876953, "learning_rate": 0.000520631479736098, "loss": 3.2856, "step": 19750 }, { "epoch": 5.328380706287683, "grad_norm": 0.35827770829200745, "learning_rate": 0.0005204295139356401, "loss": 3.2806, "step": 19800 }, { "epoch": 5.341838931955211, "grad_norm": 0.38634854555130005, "learning_rate": 0.0005202275481351824, "loss": 3.2765, "step": 19850 }, { "epoch": 5.355297157622739, "grad_norm": 0.33159148693084717, "learning_rate": 0.0005200255823347246, "loss": 3.2766, "step": 19900 }, { "epoch": 5.368755383290267, "grad_norm": 0.3398463726043701, "learning_rate": 0.0005198236165342669, "loss": 3.2782, "step": 19950 }, { "epoch": 5.3822136089577945, "grad_norm": 0.32563892006874084, "learning_rate": 0.000519621650733809, "loss": 3.2688, "step": 20000 }, { "epoch": 5.3822136089577945, "eval_accuracy": 0.38212424593649896, "eval_loss": 3.374769687652588, "eval_runtime": 84.557, "eval_samples_per_second": 213.016, "eval_steps_per_second": 13.316, "step": 20000 }, { "epoch": 5.395671834625323, "grad_norm": 0.3491804599761963, "learning_rate": 0.0005194196849333513, "loss": 3.2753, "step": 20050 }, { "epoch": 5.409130060292851, "grad_norm": 0.32255983352661133, "learning_rate": 0.0005192177191328935, "loss": 3.3006, "step": 20100 }, { "epoch": 5.422588285960379, "grad_norm": 0.3570587933063507, "learning_rate": 0.0005190157533324357, "loss": 3.2828, "step": 20150 }, { "epoch": 5.436046511627907, "grad_norm": 0.3464536964893341, "learning_rate": 0.0005188137875319779, "loss": 3.2947, "step": 20200 }, { "epoch": 5.449504737295435, "grad_norm": 0.33288997411727905, "learning_rate": 0.00051861182173152, "loss": 3.2925, "step": 20250 }, { "epoch": 5.462962962962963, "grad_norm": 0.3999335765838623, "learning_rate": 0.0005184098559310623, "loss": 3.2864, "step": 20300 }, { "epoch": 5.476421188630491, "grad_norm": 0.3677201569080353, "learning_rate": 0.0005182078901306046, "loss": 3.2725, "step": 20350 }, { "epoch": 5.489879414298019, "grad_norm": 0.3399883806705475, "learning_rate": 0.0005180059243301467, "loss": 3.2955, "step": 20400 }, { "epoch": 5.503337639965547, "grad_norm": 0.3565647304058075, "learning_rate": 0.000517803958529689, "loss": 3.2879, "step": 20450 }, { "epoch": 5.516795865633075, "grad_norm": 0.3631874918937683, "learning_rate": 0.0005176019927292311, "loss": 3.2881, "step": 20500 }, { "epoch": 5.530254091300603, "grad_norm": 0.35753047466278076, "learning_rate": 0.0005174000269287733, "loss": 3.2765, "step": 20550 }, { "epoch": 5.543712316968131, "grad_norm": 0.3494873344898224, "learning_rate": 0.0005171980611283156, "loss": 3.2798, "step": 20600 }, { "epoch": 5.557170542635659, "grad_norm": 0.3338465392589569, "learning_rate": 0.0005169960953278577, "loss": 3.276, "step": 20650 }, { "epoch": 5.5706287683031865, "grad_norm": 0.3469972610473633, "learning_rate": 0.0005167941295274, "loss": 3.2867, "step": 20700 }, { "epoch": 5.584086993970715, "grad_norm": 0.3427984118461609, "learning_rate": 0.0005165921637269421, "loss": 3.2872, "step": 20750 }, { "epoch": 5.597545219638243, "grad_norm": 0.34492751955986023, "learning_rate": 0.0005163901979264844, "loss": 3.2895, "step": 20800 }, { "epoch": 5.611003445305771, "grad_norm": 0.32378867268562317, "learning_rate": 0.0005161882321260266, "loss": 3.2828, "step": 20850 }, { "epoch": 5.624461670973298, "grad_norm": 0.33641505241394043, "learning_rate": 0.0005159862663255689, "loss": 3.2687, "step": 20900 }, { "epoch": 5.637919896640827, "grad_norm": 0.38284191489219666, "learning_rate": 0.000515784300525111, "loss": 3.2925, "step": 20950 }, { "epoch": 5.651378122308355, "grad_norm": 0.3551783263683319, "learning_rate": 0.0005155823347246533, "loss": 3.2819, "step": 21000 }, { "epoch": 5.651378122308355, "eval_accuracy": 0.3831818144285632, "eval_loss": 3.362936496734619, "eval_runtime": 84.2532, "eval_samples_per_second": 213.784, "eval_steps_per_second": 13.364, "step": 21000 }, { "epoch": 5.664836347975883, "grad_norm": 0.3598018288612366, "learning_rate": 0.0005153803689241955, "loss": 3.2921, "step": 21050 }, { "epoch": 5.678294573643411, "grad_norm": 0.3473359942436218, "learning_rate": 0.0005151784031237377, "loss": 3.2824, "step": 21100 }, { "epoch": 5.6917527993109385, "grad_norm": 0.33983397483825684, "learning_rate": 0.0005149764373232799, "loss": 3.2783, "step": 21150 }, { "epoch": 5.705211024978467, "grad_norm": 0.34451305866241455, "learning_rate": 0.000514774471522822, "loss": 3.2899, "step": 21200 }, { "epoch": 5.718669250645995, "grad_norm": 0.34196949005126953, "learning_rate": 0.0005145725057223643, "loss": 3.2842, "step": 21250 }, { "epoch": 5.732127476313523, "grad_norm": 0.3753472864627838, "learning_rate": 0.0005143705399219066, "loss": 3.2919, "step": 21300 }, { "epoch": 5.745585701981051, "grad_norm": 0.3294236660003662, "learning_rate": 0.0005141685741214487, "loss": 3.288, "step": 21350 }, { "epoch": 5.7590439276485785, "grad_norm": 0.32671308517456055, "learning_rate": 0.0005139666083209909, "loss": 3.2783, "step": 21400 }, { "epoch": 5.772502153316107, "grad_norm": 0.31655827164649963, "learning_rate": 0.0005137646425205331, "loss": 3.2926, "step": 21450 }, { "epoch": 5.785960378983635, "grad_norm": 0.34189605712890625, "learning_rate": 0.0005135626767200753, "loss": 3.2937, "step": 21500 }, { "epoch": 5.799418604651163, "grad_norm": 0.3352980315685272, "learning_rate": 0.0005133607109196176, "loss": 3.286, "step": 21550 }, { "epoch": 5.81287683031869, "grad_norm": 0.35656702518463135, "learning_rate": 0.0005131587451191597, "loss": 3.2827, "step": 21600 }, { "epoch": 5.826335055986219, "grad_norm": 0.32037410140037537, "learning_rate": 0.000512956779318702, "loss": 3.2897, "step": 21650 }, { "epoch": 5.839793281653747, "grad_norm": 0.33292341232299805, "learning_rate": 0.0005127548135182441, "loss": 3.2969, "step": 21700 }, { "epoch": 5.853251507321275, "grad_norm": 0.3731493353843689, "learning_rate": 0.0005125528477177865, "loss": 3.2891, "step": 21750 }, { "epoch": 5.866709732988802, "grad_norm": 0.3637171685695648, "learning_rate": 0.0005123508819173286, "loss": 3.2919, "step": 21800 }, { "epoch": 5.8801679586563305, "grad_norm": 0.3596208691596985, "learning_rate": 0.0005121489161168709, "loss": 3.2926, "step": 21850 }, { "epoch": 5.893626184323859, "grad_norm": 0.3490025997161865, "learning_rate": 0.000511946950316413, "loss": 3.2855, "step": 21900 }, { "epoch": 5.907084409991387, "grad_norm": 0.33612897992134094, "learning_rate": 0.0005117449845159553, "loss": 3.2957, "step": 21950 }, { "epoch": 5.920542635658915, "grad_norm": 0.331669420003891, "learning_rate": 0.0005115430187154975, "loss": 3.2922, "step": 22000 }, { "epoch": 5.920542635658915, "eval_accuracy": 0.38385922757300744, "eval_loss": 3.354418992996216, "eval_runtime": 84.493, "eval_samples_per_second": 213.178, "eval_steps_per_second": 13.327, "step": 22000 }, { "epoch": 5.934000861326442, "grad_norm": 0.3502340316772461, "learning_rate": 0.0005113410529150397, "loss": 3.2938, "step": 22050 }, { "epoch": 5.9474590869939705, "grad_norm": 0.3122784197330475, "learning_rate": 0.0005111390871145819, "loss": 3.2833, "step": 22100 }, { "epoch": 5.960917312661499, "grad_norm": 0.34915393590927124, "learning_rate": 0.000510937121314124, "loss": 3.2983, "step": 22150 }, { "epoch": 5.974375538329027, "grad_norm": 0.35112547874450684, "learning_rate": 0.0005107351555136663, "loss": 3.3069, "step": 22200 }, { "epoch": 5.987833763996555, "grad_norm": 0.3398374915122986, "learning_rate": 0.0005105331897132085, "loss": 3.2882, "step": 22250 }, { "epoch": 6.001076658053402, "grad_norm": 0.352754682302475, "learning_rate": 0.0005103312239127507, "loss": 3.2717, "step": 22300 }, { "epoch": 6.0145348837209305, "grad_norm": 0.38981321454048157, "learning_rate": 0.0005101292581122929, "loss": 3.1929, "step": 22350 }, { "epoch": 6.027993109388458, "grad_norm": 0.3339367210865021, "learning_rate": 0.0005099272923118351, "loss": 3.1948, "step": 22400 }, { "epoch": 6.041451335055986, "grad_norm": 0.37438926100730896, "learning_rate": 0.0005097253265113773, "loss": 3.1856, "step": 22450 }, { "epoch": 6.054909560723514, "grad_norm": 0.3462595045566559, "learning_rate": 0.0005095233607109196, "loss": 3.2029, "step": 22500 }, { "epoch": 6.068367786391042, "grad_norm": 0.3641872704029083, "learning_rate": 0.0005093213949104617, "loss": 3.197, "step": 22550 }, { "epoch": 6.0818260120585705, "grad_norm": 0.3788128197193146, "learning_rate": 0.000509119429110004, "loss": 3.2085, "step": 22600 }, { "epoch": 6.095284237726098, "grad_norm": 0.3462986648082733, "learning_rate": 0.0005089174633095462, "loss": 3.2136, "step": 22650 }, { "epoch": 6.108742463393626, "grad_norm": 0.34483978152275085, "learning_rate": 0.0005087154975090885, "loss": 3.2224, "step": 22700 }, { "epoch": 6.122200689061154, "grad_norm": 0.3590666353702545, "learning_rate": 0.0005085135317086306, "loss": 3.2083, "step": 22750 }, { "epoch": 6.135658914728682, "grad_norm": 0.3543896973133087, "learning_rate": 0.0005083115659081729, "loss": 3.224, "step": 22800 }, { "epoch": 6.149117140396211, "grad_norm": 0.32914939522743225, "learning_rate": 0.0005081096001077151, "loss": 3.212, "step": 22850 }, { "epoch": 6.162575366063738, "grad_norm": 0.3688567876815796, "learning_rate": 0.0005079076343072573, "loss": 3.2234, "step": 22900 }, { "epoch": 6.176033591731266, "grad_norm": 0.3502994179725647, "learning_rate": 0.0005077056685067995, "loss": 3.2192, "step": 22950 }, { "epoch": 6.189491817398794, "grad_norm": 0.3282775580883026, "learning_rate": 0.0005075037027063417, "loss": 3.2204, "step": 23000 }, { "epoch": 6.189491817398794, "eval_accuracy": 0.38417778015352233, "eval_loss": 3.3597404956817627, "eval_runtime": 87.7605, "eval_samples_per_second": 205.24, "eval_steps_per_second": 12.83, "step": 23000 }, { "epoch": 6.2029500430663225, "grad_norm": 0.3568970263004303, "learning_rate": 0.0005073017369058839, "loss": 3.2305, "step": 23050 }, { "epoch": 6.21640826873385, "grad_norm": 0.3501264452934265, "learning_rate": 0.0005070997711054262, "loss": 3.2408, "step": 23100 }, { "epoch": 6.229866494401378, "grad_norm": 0.38300803303718567, "learning_rate": 0.0005068978053049683, "loss": 3.2223, "step": 23150 }, { "epoch": 6.243324720068906, "grad_norm": 0.34357237815856934, "learning_rate": 0.0005066958395045105, "loss": 3.2282, "step": 23200 }, { "epoch": 6.256782945736434, "grad_norm": 0.35204562544822693, "learning_rate": 0.0005064938737040527, "loss": 3.2218, "step": 23250 }, { "epoch": 6.270241171403962, "grad_norm": 0.3680749237537384, "learning_rate": 0.0005062919079035949, "loss": 3.2297, "step": 23300 }, { "epoch": 6.28369939707149, "grad_norm": 0.3626226484775543, "learning_rate": 0.0005060899421031372, "loss": 3.2332, "step": 23350 }, { "epoch": 6.297157622739018, "grad_norm": 0.3031291961669922, "learning_rate": 0.0005058879763026793, "loss": 3.2339, "step": 23400 }, { "epoch": 6.310615848406546, "grad_norm": 0.3484286069869995, "learning_rate": 0.0005056860105022216, "loss": 3.2366, "step": 23450 }, { "epoch": 6.324074074074074, "grad_norm": 0.3633332848548889, "learning_rate": 0.0005054840447017637, "loss": 3.2369, "step": 23500 }, { "epoch": 6.337532299741602, "grad_norm": 0.384171724319458, "learning_rate": 0.0005052820789013061, "loss": 3.2423, "step": 23550 }, { "epoch": 6.35099052540913, "grad_norm": 0.3473833501338959, "learning_rate": 0.0005050801131008482, "loss": 3.2322, "step": 23600 }, { "epoch": 6.364448751076658, "grad_norm": 0.3337860107421875, "learning_rate": 0.0005048781473003905, "loss": 3.2337, "step": 23650 }, { "epoch": 6.377906976744186, "grad_norm": 0.32066184282302856, "learning_rate": 0.0005046761814999326, "loss": 3.2396, "step": 23700 }, { "epoch": 6.3913652024117145, "grad_norm": 0.3586727976799011, "learning_rate": 0.0005044742156994749, "loss": 3.2382, "step": 23750 }, { "epoch": 6.404823428079242, "grad_norm": 0.3122645318508148, "learning_rate": 0.0005042722498990171, "loss": 3.2338, "step": 23800 }, { "epoch": 6.41828165374677, "grad_norm": 0.3277284502983093, "learning_rate": 0.0005040702840985593, "loss": 3.2286, "step": 23850 }, { "epoch": 6.431739879414298, "grad_norm": 0.3412085175514221, "learning_rate": 0.0005038683182981015, "loss": 3.2421, "step": 23900 }, { "epoch": 6.445198105081826, "grad_norm": 0.37467247247695923, "learning_rate": 0.0005036663524976436, "loss": 3.2218, "step": 23950 }, { "epoch": 6.458656330749354, "grad_norm": 0.37306609749794006, "learning_rate": 0.0005034643866971859, "loss": 3.2478, "step": 24000 }, { "epoch": 6.458656330749354, "eval_accuracy": 0.38472438248386703, "eval_loss": 3.3520491123199463, "eval_runtime": 87.8028, "eval_samples_per_second": 205.141, "eval_steps_per_second": 12.824, "step": 24000 }, { "epoch": 6.472114556416882, "grad_norm": 0.3688383996486664, "learning_rate": 0.0005032624208967281, "loss": 3.2365, "step": 24050 }, { "epoch": 6.48557278208441, "grad_norm": 0.38653844594955444, "learning_rate": 0.0005030604550962703, "loss": 3.2543, "step": 24100 }, { "epoch": 6.499031007751938, "grad_norm": 0.3507419228553772, "learning_rate": 0.0005028584892958125, "loss": 3.2465, "step": 24150 }, { "epoch": 6.5124892334194655, "grad_norm": 0.33346864581108093, "learning_rate": 0.0005026565234953547, "loss": 3.2468, "step": 24200 }, { "epoch": 6.525947459086994, "grad_norm": 0.3589414954185486, "learning_rate": 0.0005024545576948969, "loss": 3.2423, "step": 24250 }, { "epoch": 6.539405684754522, "grad_norm": 0.33978089690208435, "learning_rate": 0.0005022525918944392, "loss": 3.2464, "step": 24300 }, { "epoch": 6.55286391042205, "grad_norm": 0.3351050317287445, "learning_rate": 0.0005020506260939813, "loss": 3.2504, "step": 24350 }, { "epoch": 6.566322136089578, "grad_norm": 0.360805869102478, "learning_rate": 0.0005018486602935236, "loss": 3.2588, "step": 24400 }, { "epoch": 6.579780361757106, "grad_norm": 0.35320815443992615, "learning_rate": 0.0005016466944930658, "loss": 3.2711, "step": 24450 }, { "epoch": 6.593238587424634, "grad_norm": 0.3797634541988373, "learning_rate": 0.0005014447286926081, "loss": 3.2452, "step": 24500 }, { "epoch": 6.606696813092162, "grad_norm": 0.33502110838890076, "learning_rate": 0.0005012427628921502, "loss": 3.2437, "step": 24550 }, { "epoch": 6.62015503875969, "grad_norm": 0.3418447971343994, "learning_rate": 0.0005010407970916925, "loss": 3.2457, "step": 24600 }, { "epoch": 6.633613264427218, "grad_norm": 0.33172404766082764, "learning_rate": 0.0005008388312912346, "loss": 3.2407, "step": 24650 }, { "epoch": 6.647071490094746, "grad_norm": 0.3561548888683319, "learning_rate": 0.0005006368654907769, "loss": 3.2516, "step": 24700 }, { "epoch": 6.660529715762274, "grad_norm": 0.3367748260498047, "learning_rate": 0.0005004348996903191, "loss": 3.2498, "step": 24750 }, { "epoch": 6.673987941429802, "grad_norm": 0.3726949691772461, "learning_rate": 0.0005002329338898613, "loss": 3.2442, "step": 24800 }, { "epoch": 6.68744616709733, "grad_norm": 0.33080199360847473, "learning_rate": 0.0005000309680894035, "loss": 3.2484, "step": 24850 }, { "epoch": 6.7009043927648575, "grad_norm": 0.3477340340614319, "learning_rate": 0.0004998290022889456, "loss": 3.2554, "step": 24900 }, { "epoch": 6.714362618432386, "grad_norm": 0.34718841314315796, "learning_rate": 0.0004996270364884879, "loss": 3.249, "step": 24950 }, { "epoch": 6.727820844099914, "grad_norm": 0.3280639946460724, "learning_rate": 0.0004994250706880301, "loss": 3.2492, "step": 25000 }, { "epoch": 6.727820844099914, "eval_accuracy": 0.385652552570954, "eval_loss": 3.342715263366699, "eval_runtime": 87.6454, "eval_samples_per_second": 205.51, "eval_steps_per_second": 12.847, "step": 25000 }, { "epoch": 6.741279069767442, "grad_norm": 0.33485645055770874, "learning_rate": 0.0004992231048875723, "loss": 3.2543, "step": 25050 }, { "epoch": 6.754737295434969, "grad_norm": 0.34157443046569824, "learning_rate": 0.0004990211390871145, "loss": 3.2516, "step": 25100 }, { "epoch": 6.768195521102498, "grad_norm": 0.33038467168807983, "learning_rate": 0.0004988191732866567, "loss": 3.2469, "step": 25150 }, { "epoch": 6.781653746770026, "grad_norm": 0.3318648636341095, "learning_rate": 0.0004986172074861989, "loss": 3.2579, "step": 25200 }, { "epoch": 6.795111972437554, "grad_norm": 0.34127184748649597, "learning_rate": 0.0004984152416857412, "loss": 3.2505, "step": 25250 }, { "epoch": 6.808570198105082, "grad_norm": 0.35870561003685, "learning_rate": 0.0004982132758852834, "loss": 3.2452, "step": 25300 }, { "epoch": 6.822028423772609, "grad_norm": 0.33743295073509216, "learning_rate": 0.0004980113100848256, "loss": 3.2506, "step": 25350 }, { "epoch": 6.835486649440138, "grad_norm": 0.3330428898334503, "learning_rate": 0.0004978093442843678, "loss": 3.2646, "step": 25400 }, { "epoch": 6.848944875107666, "grad_norm": 0.3399394452571869, "learning_rate": 0.0004976073784839101, "loss": 3.2571, "step": 25450 }, { "epoch": 6.862403100775194, "grad_norm": 0.35361889004707336, "learning_rate": 0.0004974054126834522, "loss": 3.2598, "step": 25500 }, { "epoch": 6.875861326442722, "grad_norm": 0.3305911421775818, "learning_rate": 0.0004972034468829945, "loss": 3.2477, "step": 25550 }, { "epoch": 6.8893195521102495, "grad_norm": 0.3251939117908478, "learning_rate": 0.0004970014810825366, "loss": 3.2501, "step": 25600 }, { "epoch": 6.902777777777778, "grad_norm": 0.3742504119873047, "learning_rate": 0.0004967995152820789, "loss": 3.2601, "step": 25650 }, { "epoch": 6.916236003445306, "grad_norm": 0.3559335172176361, "learning_rate": 0.0004965975494816211, "loss": 3.2633, "step": 25700 }, { "epoch": 6.929694229112834, "grad_norm": 0.31849339604377747, "learning_rate": 0.0004963955836811633, "loss": 3.2591, "step": 25750 }, { "epoch": 6.943152454780362, "grad_norm": 0.3387935161590576, "learning_rate": 0.0004961936178807055, "loss": 3.2542, "step": 25800 }, { "epoch": 6.9566106804478895, "grad_norm": 0.330844521522522, "learning_rate": 0.0004959916520802476, "loss": 3.2516, "step": 25850 }, { "epoch": 6.970068906115418, "grad_norm": 0.3462895154953003, "learning_rate": 0.0004957896862797899, "loss": 3.253, "step": 25900 }, { "epoch": 6.983527131782946, "grad_norm": 0.35414084792137146, "learning_rate": 0.0004955877204793321, "loss": 3.2574, "step": 25950 }, { "epoch": 6.996985357450473, "grad_norm": 0.32836857438087463, "learning_rate": 0.0004953857546788743, "loss": 3.264, "step": 26000 }, { "epoch": 6.996985357450473, "eval_accuracy": 0.38607888283218883, "eval_loss": 3.3374903202056885, "eval_runtime": 88.3286, "eval_samples_per_second": 203.92, "eval_steps_per_second": 12.748, "step": 26000 }, { "epoch": 7.010228251507321, "grad_norm": 0.34610164165496826, "learning_rate": 0.0004951837888784165, "loss": 3.1889, "step": 26050 }, { "epoch": 7.0236864771748495, "grad_norm": 0.35366278886795044, "learning_rate": 0.0004949818230779587, "loss": 3.1516, "step": 26100 }, { "epoch": 7.037144702842378, "grad_norm": 0.3489835262298584, "learning_rate": 0.0004947798572775009, "loss": 3.1586, "step": 26150 }, { "epoch": 7.050602928509905, "grad_norm": 0.37238720059394836, "learning_rate": 0.0004945778914770432, "loss": 3.1644, "step": 26200 }, { "epoch": 7.064061154177433, "grad_norm": 0.33292341232299805, "learning_rate": 0.0004943759256765854, "loss": 3.1729, "step": 26250 }, { "epoch": 7.077519379844961, "grad_norm": 0.3589879274368286, "learning_rate": 0.0004941739598761276, "loss": 3.1727, "step": 26300 }, { "epoch": 7.09097760551249, "grad_norm": 0.35007166862487793, "learning_rate": 0.0004939719940756698, "loss": 3.1729, "step": 26350 }, { "epoch": 7.104435831180017, "grad_norm": 0.36499452590942383, "learning_rate": 0.0004937700282752121, "loss": 3.1658, "step": 26400 }, { "epoch": 7.117894056847545, "grad_norm": 0.35566914081573486, "learning_rate": 0.0004935680624747542, "loss": 3.1732, "step": 26450 }, { "epoch": 7.131352282515073, "grad_norm": 0.38588660955429077, "learning_rate": 0.0004933660966742965, "loss": 3.1778, "step": 26500 }, { "epoch": 7.144810508182601, "grad_norm": 0.3605356514453888, "learning_rate": 0.0004931641308738386, "loss": 3.1752, "step": 26550 }, { "epoch": 7.15826873385013, "grad_norm": 0.37103271484375, "learning_rate": 0.0004929621650733809, "loss": 3.1925, "step": 26600 }, { "epoch": 7.171726959517657, "grad_norm": 0.3648276627063751, "learning_rate": 0.0004927601992729231, "loss": 3.1899, "step": 26650 }, { "epoch": 7.185185185185185, "grad_norm": 0.35377174615859985, "learning_rate": 0.0004925582334724652, "loss": 3.1915, "step": 26700 }, { "epoch": 7.198643410852713, "grad_norm": 0.3399772047996521, "learning_rate": 0.0004923562676720075, "loss": 3.1972, "step": 26750 }, { "epoch": 7.2121016365202415, "grad_norm": 0.35400402545928955, "learning_rate": 0.0004921543018715497, "loss": 3.199, "step": 26800 }, { "epoch": 7.225559862187769, "grad_norm": 0.3436429500579834, "learning_rate": 0.0004919523360710919, "loss": 3.1839, "step": 26850 }, { "epoch": 7.239018087855297, "grad_norm": 0.31973743438720703, "learning_rate": 0.0004917503702706341, "loss": 3.1854, "step": 26900 }, { "epoch": 7.252476313522825, "grad_norm": 0.3425561487674713, "learning_rate": 0.0004915484044701763, "loss": 3.1974, "step": 26950 }, { "epoch": 7.265934539190353, "grad_norm": 0.3549090027809143, "learning_rate": 0.0004913464386697185, "loss": 3.2051, "step": 27000 }, { "epoch": 7.265934539190353, "eval_accuracy": 0.38628944043827274, "eval_loss": 3.3425750732421875, "eval_runtime": 88.1866, "eval_samples_per_second": 204.249, "eval_steps_per_second": 12.768, "step": 27000 }, { "epoch": 7.279392764857882, "grad_norm": 0.34049826860427856, "learning_rate": 0.0004911444728692608, "loss": 3.1987, "step": 27050 }, { "epoch": 7.292850990525409, "grad_norm": 0.3492026627063751, "learning_rate": 0.000490942507068803, "loss": 3.1879, "step": 27100 }, { "epoch": 7.306309216192937, "grad_norm": 0.33971208333969116, "learning_rate": 0.0004907405412683452, "loss": 3.2011, "step": 27150 }, { "epoch": 7.319767441860465, "grad_norm": 0.345683217048645, "learning_rate": 0.0004905385754678874, "loss": 3.2004, "step": 27200 }, { "epoch": 7.333225667527993, "grad_norm": 0.37204256653785706, "learning_rate": 0.0004903366096674297, "loss": 3.1979, "step": 27250 }, { "epoch": 7.346683893195521, "grad_norm": 0.367231547832489, "learning_rate": 0.0004901346438669718, "loss": 3.2097, "step": 27300 }, { "epoch": 7.360142118863049, "grad_norm": 0.33875802159309387, "learning_rate": 0.0004899326780665141, "loss": 3.222, "step": 27350 }, { "epoch": 7.373600344530577, "grad_norm": 0.3583943843841553, "learning_rate": 0.0004897307122660562, "loss": 3.1964, "step": 27400 }, { "epoch": 7.387058570198105, "grad_norm": 0.34471067786216736, "learning_rate": 0.0004895287464655985, "loss": 3.2117, "step": 27450 }, { "epoch": 7.4005167958656335, "grad_norm": 0.335271954536438, "learning_rate": 0.0004893267806651407, "loss": 3.201, "step": 27500 }, { "epoch": 7.413975021533161, "grad_norm": 0.36515283584594727, "learning_rate": 0.0004891248148646829, "loss": 3.2106, "step": 27550 }, { "epoch": 7.427433247200689, "grad_norm": 0.3432753086090088, "learning_rate": 0.0004889228490642251, "loss": 3.2181, "step": 27600 }, { "epoch": 7.440891472868217, "grad_norm": 0.3460571765899658, "learning_rate": 0.0004887208832637672, "loss": 3.2091, "step": 27650 }, { "epoch": 7.454349698535745, "grad_norm": 0.366629034280777, "learning_rate": 0.0004885189174633095, "loss": 3.2156, "step": 27700 }, { "epoch": 7.467807924203273, "grad_norm": 0.3270829916000366, "learning_rate": 0.0004883169516628517, "loss": 3.2139, "step": 27750 }, { "epoch": 7.481266149870801, "grad_norm": 0.3518711030483246, "learning_rate": 0.0004881149858623939, "loss": 3.2114, "step": 27800 }, { "epoch": 7.494724375538329, "grad_norm": 0.34649714827537537, "learning_rate": 0.00048791302006193614, "loss": 3.2123, "step": 27850 }, { "epoch": 7.508182601205857, "grad_norm": 0.5178732872009277, "learning_rate": 0.00048771105426147833, "loss": 3.2167, "step": 27900 }, { "epoch": 7.521640826873385, "grad_norm": 0.3424026370048523, "learning_rate": 0.00048750908846102053, "loss": 3.2017, "step": 27950 }, { "epoch": 7.535099052540913, "grad_norm": 0.34950610995292664, "learning_rate": 0.0004873071226605628, "loss": 3.2193, "step": 28000 }, { "epoch": 7.535099052540913, "eval_accuracy": 0.38668426311139387, "eval_loss": 3.3392179012298584, "eval_runtime": 87.7547, "eval_samples_per_second": 205.254, "eval_steps_per_second": 12.831, "step": 28000 }, { "epoch": 7.548557278208441, "grad_norm": 0.34104108810424805, "learning_rate": 0.00048710515686010503, "loss": 3.2092, "step": 28050 }, { "epoch": 7.562015503875969, "grad_norm": 0.32871460914611816, "learning_rate": 0.0004869031910596472, "loss": 3.2222, "step": 28100 }, { "epoch": 7.575473729543497, "grad_norm": 0.34514591097831726, "learning_rate": 0.0004867012252591894, "loss": 3.2164, "step": 28150 }, { "epoch": 7.588931955211025, "grad_norm": 0.33675217628479004, "learning_rate": 0.0004864992594587316, "loss": 3.2146, "step": 28200 }, { "epoch": 7.602390180878553, "grad_norm": 0.35956332087516785, "learning_rate": 0.0004862972936582738, "loss": 3.2161, "step": 28250 }, { "epoch": 7.615848406546081, "grad_norm": 0.3696342706680298, "learning_rate": 0.00048609532785781606, "loss": 3.211, "step": 28300 }, { "epoch": 7.629306632213609, "grad_norm": 0.340660035610199, "learning_rate": 0.00048589336205735826, "loss": 3.2156, "step": 28350 }, { "epoch": 7.6427648578811365, "grad_norm": 0.34237396717071533, "learning_rate": 0.00048569139625690046, "loss": 3.2297, "step": 28400 }, { "epoch": 7.656223083548665, "grad_norm": 0.37521055340766907, "learning_rate": 0.00048548943045644265, "loss": 3.2151, "step": 28450 }, { "epoch": 7.669681309216193, "grad_norm": 0.3448121249675751, "learning_rate": 0.00048528746465598485, "loss": 3.201, "step": 28500 }, { "epoch": 7.683139534883721, "grad_norm": 0.34937021136283875, "learning_rate": 0.0004850854988555271, "loss": 3.2188, "step": 28550 }, { "epoch": 7.696597760551249, "grad_norm": 0.32854047417640686, "learning_rate": 0.0004848835330550693, "loss": 3.2189, "step": 28600 }, { "epoch": 7.7100559862187765, "grad_norm": 0.35428887605667114, "learning_rate": 0.0004846815672546115, "loss": 3.2181, "step": 28650 }, { "epoch": 7.723514211886305, "grad_norm": 0.34827983379364014, "learning_rate": 0.0004844796014541537, "loss": 3.2252, "step": 28700 }, { "epoch": 7.736972437553833, "grad_norm": 0.3507974147796631, "learning_rate": 0.0004842776356536959, "loss": 3.2141, "step": 28750 }, { "epoch": 7.750430663221361, "grad_norm": 0.3425598740577698, "learning_rate": 0.00048407566985323813, "loss": 3.2175, "step": 28800 }, { "epoch": 7.763888888888889, "grad_norm": 0.33663254976272583, "learning_rate": 0.0004838737040527804, "loss": 3.2409, "step": 28850 }, { "epoch": 7.777347114556417, "grad_norm": 0.33253276348114014, "learning_rate": 0.0004836717382523226, "loss": 3.2262, "step": 28900 }, { "epoch": 7.790805340223945, "grad_norm": 0.3696092665195465, "learning_rate": 0.0004834697724518648, "loss": 3.2065, "step": 28950 }, { "epoch": 7.804263565891473, "grad_norm": 0.32295504212379456, "learning_rate": 0.000483267806651407, "loss": 3.2262, "step": 29000 }, { "epoch": 7.804263565891473, "eval_accuracy": 0.3878248378011093, "eval_loss": 3.3295774459838867, "eval_runtime": 86.7214, "eval_samples_per_second": 207.7, "eval_steps_per_second": 12.984, "step": 29000 }, { "epoch": 7.817721791559001, "grad_norm": 0.3298531770706177, "learning_rate": 0.0004830658408509492, "loss": 3.2213, "step": 29050 }, { "epoch": 7.831180017226529, "grad_norm": 0.3383207619190216, "learning_rate": 0.0004828638750504914, "loss": 3.2347, "step": 29100 }, { "epoch": 7.844638242894057, "grad_norm": 0.36010101437568665, "learning_rate": 0.0004826619092500336, "loss": 3.2233, "step": 29150 }, { "epoch": 7.858096468561585, "grad_norm": 0.34656408429145813, "learning_rate": 0.0004824599434495758, "loss": 3.2259, "step": 29200 }, { "epoch": 7.871554694229113, "grad_norm": 0.3213461637496948, "learning_rate": 0.00048225797764911806, "loss": 3.2245, "step": 29250 }, { "epoch": 7.885012919896641, "grad_norm": 0.35127902030944824, "learning_rate": 0.00048205601184866026, "loss": 3.2149, "step": 29300 }, { "epoch": 7.8984711455641685, "grad_norm": 0.3342956602573395, "learning_rate": 0.00048185404604820245, "loss": 3.2277, "step": 29350 }, { "epoch": 7.911929371231697, "grad_norm": 0.359724760055542, "learning_rate": 0.00048165208024774465, "loss": 3.2216, "step": 29400 }, { "epoch": 7.925387596899225, "grad_norm": 0.3557015657424927, "learning_rate": 0.00048145011444728685, "loss": 3.2249, "step": 29450 }, { "epoch": 7.938845822566753, "grad_norm": 0.3411562740802765, "learning_rate": 0.0004812481486468291, "loss": 3.2336, "step": 29500 }, { "epoch": 7.95230404823428, "grad_norm": 0.3151187598705292, "learning_rate": 0.0004810461828463713, "loss": 3.2251, "step": 29550 }, { "epoch": 7.965762273901809, "grad_norm": 0.36215487122535706, "learning_rate": 0.0004808442170459135, "loss": 3.2345, "step": 29600 }, { "epoch": 7.979220499569337, "grad_norm": 0.3466815650463104, "learning_rate": 0.0004806422512454557, "loss": 3.2336, "step": 29650 }, { "epoch": 7.992678725236865, "grad_norm": 0.340318888425827, "learning_rate": 0.00048044028544499793, "loss": 3.2202, "step": 29700 }, { "epoch": 8.005921619293712, "grad_norm": 0.3526352643966675, "learning_rate": 0.0004802383196445402, "loss": 3.1826, "step": 29750 }, { "epoch": 8.01937984496124, "grad_norm": 0.3376880884170532, "learning_rate": 0.0004800363538440824, "loss": 3.1301, "step": 29800 }, { "epoch": 8.032838070628769, "grad_norm": 0.35756751894950867, "learning_rate": 0.0004798343880436246, "loss": 3.1278, "step": 29850 }, { "epoch": 8.046296296296296, "grad_norm": 0.3583872616291046, "learning_rate": 0.00047963242224316683, "loss": 3.1379, "step": 29900 }, { "epoch": 8.059754521963825, "grad_norm": 0.352107435464859, "learning_rate": 0.000479430456442709, "loss": 3.1389, "step": 29950 }, { "epoch": 8.073212747631352, "grad_norm": 0.32794874906539917, "learning_rate": 0.0004792284906422512, "loss": 3.1215, "step": 30000 }, { "epoch": 8.073212747631352, "eval_accuracy": 0.3876267745834154, "eval_loss": 3.33420991897583, "eval_runtime": 86.664, "eval_samples_per_second": 207.837, "eval_steps_per_second": 12.993, "step": 30000 }, { "epoch": 8.08667097329888, "grad_norm": 0.35623717308044434, "learning_rate": 0.0004790265248417934, "loss": 3.1381, "step": 30050 }, { "epoch": 8.100129198966409, "grad_norm": 0.34889814257621765, "learning_rate": 0.0004788245590413356, "loss": 3.1485, "step": 30100 }, { "epoch": 8.113587424633936, "grad_norm": 0.3647051155567169, "learning_rate": 0.00047862259324087786, "loss": 3.1463, "step": 30150 }, { "epoch": 8.127045650301465, "grad_norm": 0.3717019855976105, "learning_rate": 0.00047842062744042006, "loss": 3.1599, "step": 30200 }, { "epoch": 8.140503875968992, "grad_norm": 0.36561229825019836, "learning_rate": 0.00047821866163996225, "loss": 3.1543, "step": 30250 }, { "epoch": 8.15396210163652, "grad_norm": 0.37198320031166077, "learning_rate": 0.00047801669583950445, "loss": 3.1518, "step": 30300 }, { "epoch": 8.167420327304049, "grad_norm": 0.37193116545677185, "learning_rate": 0.00047781473003904665, "loss": 3.1594, "step": 30350 }, { "epoch": 8.180878552971576, "grad_norm": 0.3370911777019501, "learning_rate": 0.0004776127642385889, "loss": 3.1515, "step": 30400 }, { "epoch": 8.194336778639105, "grad_norm": 0.3725023865699768, "learning_rate": 0.0004774107984381311, "loss": 3.1578, "step": 30450 }, { "epoch": 8.207795004306632, "grad_norm": 0.3886215090751648, "learning_rate": 0.0004772088326376733, "loss": 3.1708, "step": 30500 }, { "epoch": 8.22125322997416, "grad_norm": 0.3475479185581207, "learning_rate": 0.0004770068668372155, "loss": 3.1582, "step": 30550 }, { "epoch": 8.234711455641689, "grad_norm": 0.36613041162490845, "learning_rate": 0.0004768049010367577, "loss": 3.1532, "step": 30600 }, { "epoch": 8.248169681309216, "grad_norm": 0.3678821325302124, "learning_rate": 0.0004766029352363, "loss": 3.1717, "step": 30650 }, { "epoch": 8.261627906976745, "grad_norm": 0.3694922924041748, "learning_rate": 0.0004764009694358422, "loss": 3.1564, "step": 30700 }, { "epoch": 8.275086132644272, "grad_norm": 0.39171549677848816, "learning_rate": 0.0004761990036353844, "loss": 3.1649, "step": 30750 }, { "epoch": 8.2885443583118, "grad_norm": 0.3456292450428009, "learning_rate": 0.0004759970378349266, "loss": 3.1692, "step": 30800 }, { "epoch": 8.302002583979329, "grad_norm": 0.37906062602996826, "learning_rate": 0.0004757950720344688, "loss": 3.1663, "step": 30850 }, { "epoch": 8.315460809646856, "grad_norm": 0.36503514647483826, "learning_rate": 0.000475593106234011, "loss": 3.1734, "step": 30900 }, { "epoch": 8.328919035314383, "grad_norm": 0.3400176167488098, "learning_rate": 0.0004753911404335532, "loss": 3.1824, "step": 30950 }, { "epoch": 8.342377260981912, "grad_norm": 0.33662930130958557, "learning_rate": 0.0004751891746330954, "loss": 3.1749, "step": 31000 }, { "epoch": 8.342377260981912, "eval_accuracy": 0.3880142092703581, "eval_loss": 3.3296737670898438, "eval_runtime": 87.4696, "eval_samples_per_second": 205.923, "eval_steps_per_second": 12.873, "step": 31000 }, { "epoch": 8.35583548664944, "grad_norm": 0.3648368716239929, "learning_rate": 0.0004749872088326376, "loss": 3.1802, "step": 31050 }, { "epoch": 8.369293712316969, "grad_norm": 0.3359640836715698, "learning_rate": 0.00047478524303217986, "loss": 3.1849, "step": 31100 }, { "epoch": 8.382751937984496, "grad_norm": 0.35760825872421265, "learning_rate": 0.00047458327723172206, "loss": 3.1772, "step": 31150 }, { "epoch": 8.396210163652023, "grad_norm": 0.382489413022995, "learning_rate": 0.00047438131143126425, "loss": 3.1804, "step": 31200 }, { "epoch": 8.409668389319553, "grad_norm": 0.3471967875957489, "learning_rate": 0.00047417934563080645, "loss": 3.1865, "step": 31250 }, { "epoch": 8.42312661498708, "grad_norm": 0.3476189374923706, "learning_rate": 0.00047397737983034864, "loss": 3.1876, "step": 31300 }, { "epoch": 8.436584840654609, "grad_norm": 0.35890787839889526, "learning_rate": 0.0004737754140298909, "loss": 3.183, "step": 31350 }, { "epoch": 8.450043066322136, "grad_norm": 0.33333414793014526, "learning_rate": 0.0004735734482294331, "loss": 3.1763, "step": 31400 }, { "epoch": 8.463501291989663, "grad_norm": 0.35873493552207947, "learning_rate": 0.0004733714824289753, "loss": 3.1922, "step": 31450 }, { "epoch": 8.476959517657193, "grad_norm": 0.35295209288597107, "learning_rate": 0.0004731695166285175, "loss": 3.1904, "step": 31500 }, { "epoch": 8.49041774332472, "grad_norm": 0.3632144629955292, "learning_rate": 0.0004729675508280598, "loss": 3.171, "step": 31550 }, { "epoch": 8.503875968992247, "grad_norm": 0.3484487235546112, "learning_rate": 0.000472765585027602, "loss": 3.1783, "step": 31600 }, { "epoch": 8.517334194659776, "grad_norm": 0.3353268504142761, "learning_rate": 0.0004725636192271442, "loss": 3.1844, "step": 31650 }, { "epoch": 8.530792420327304, "grad_norm": 0.3567664325237274, "learning_rate": 0.0004723616534266864, "loss": 3.1806, "step": 31700 }, { "epoch": 8.544250645994833, "grad_norm": 0.35602352023124695, "learning_rate": 0.0004721596876262286, "loss": 3.1921, "step": 31750 }, { "epoch": 8.55770887166236, "grad_norm": 0.34894222021102905, "learning_rate": 0.0004719577218257708, "loss": 3.1907, "step": 31800 }, { "epoch": 8.571167097329887, "grad_norm": 0.3618151843547821, "learning_rate": 0.000471755756025313, "loss": 3.1857, "step": 31850 }, { "epoch": 8.584625322997416, "grad_norm": 0.34802767634391785, "learning_rate": 0.0004715537902248552, "loss": 3.1868, "step": 31900 }, { "epoch": 8.598083548664944, "grad_norm": 0.3836779296398163, "learning_rate": 0.0004713518244243974, "loss": 3.196, "step": 31950 }, { "epoch": 8.611541774332473, "grad_norm": 0.3747783601284027, "learning_rate": 0.00047114985862393966, "loss": 3.1944, "step": 32000 }, { "epoch": 8.611541774332473, "eval_accuracy": 0.38849529754679746, "eval_loss": 3.32307767868042, "eval_runtime": 87.6755, "eval_samples_per_second": 205.439, "eval_steps_per_second": 12.843, "step": 32000 }, { "epoch": 8.625, "grad_norm": 0.33180832862854004, "learning_rate": 0.00047094789282348186, "loss": 3.192, "step": 32050 }, { "epoch": 8.638458225667527, "grad_norm": 0.3259965777397156, "learning_rate": 0.00047074592702302405, "loss": 3.1975, "step": 32100 }, { "epoch": 8.651916451335056, "grad_norm": 0.3438642919063568, "learning_rate": 0.00047054396122256625, "loss": 3.1899, "step": 32150 }, { "epoch": 8.665374677002584, "grad_norm": 0.4004620313644409, "learning_rate": 0.00047034199542210844, "loss": 3.1882, "step": 32200 }, { "epoch": 8.678832902670113, "grad_norm": 0.3403623402118683, "learning_rate": 0.0004701400296216507, "loss": 3.1806, "step": 32250 }, { "epoch": 8.69229112833764, "grad_norm": 0.3368901014328003, "learning_rate": 0.0004699380638211929, "loss": 3.1914, "step": 32300 }, { "epoch": 8.705749354005167, "grad_norm": 0.3399077355861664, "learning_rate": 0.0004697360980207351, "loss": 3.1939, "step": 32350 }, { "epoch": 8.719207579672696, "grad_norm": 0.36910849809646606, "learning_rate": 0.0004695341322202773, "loss": 3.1849, "step": 32400 }, { "epoch": 8.732665805340224, "grad_norm": 0.3569251596927643, "learning_rate": 0.0004693321664198196, "loss": 3.2088, "step": 32450 }, { "epoch": 8.746124031007753, "grad_norm": 0.33960893750190735, "learning_rate": 0.0004691302006193618, "loss": 3.1919, "step": 32500 }, { "epoch": 8.75958225667528, "grad_norm": 0.36516520380973816, "learning_rate": 0.000468928234818904, "loss": 3.1945, "step": 32550 }, { "epoch": 8.773040482342807, "grad_norm": 0.3720601499080658, "learning_rate": 0.0004687262690184462, "loss": 3.2066, "step": 32600 }, { "epoch": 8.786498708010337, "grad_norm": 0.3640294373035431, "learning_rate": 0.00046852430321798837, "loss": 3.1955, "step": 32650 }, { "epoch": 8.799956933677864, "grad_norm": 0.36675596237182617, "learning_rate": 0.0004683223374175306, "loss": 3.1951, "step": 32700 }, { "epoch": 8.813415159345391, "grad_norm": 0.34109437465667725, "learning_rate": 0.0004681203716170728, "loss": 3.198, "step": 32750 }, { "epoch": 8.82687338501292, "grad_norm": 0.38729915022850037, "learning_rate": 0.000467918405816615, "loss": 3.1938, "step": 32800 }, { "epoch": 8.840331610680447, "grad_norm": 0.3687329590320587, "learning_rate": 0.0004677164400161572, "loss": 3.1969, "step": 32850 }, { "epoch": 8.853789836347977, "grad_norm": 0.38186848163604736, "learning_rate": 0.0004675144742156994, "loss": 3.1982, "step": 32900 }, { "epoch": 8.867248062015504, "grad_norm": 0.3522675037384033, "learning_rate": 0.00046731250841524166, "loss": 3.193, "step": 32950 }, { "epoch": 8.880706287683031, "grad_norm": 0.3754071891307831, "learning_rate": 0.00046711054261478385, "loss": 3.1866, "step": 33000 }, { "epoch": 8.880706287683031, "eval_accuracy": 0.388980079816326, "eval_loss": 3.317448616027832, "eval_runtime": 88.2344, "eval_samples_per_second": 204.138, "eval_steps_per_second": 12.761, "step": 33000 }, { "epoch": 8.89416451335056, "grad_norm": 0.3508091866970062, "learning_rate": 0.00046690857681432605, "loss": 3.2062, "step": 33050 }, { "epoch": 8.907622739018088, "grad_norm": 0.3021308183670044, "learning_rate": 0.00046670661101386825, "loss": 3.2019, "step": 33100 }, { "epoch": 8.921080964685617, "grad_norm": 0.33504754304885864, "learning_rate": 0.00046650464521341044, "loss": 3.2104, "step": 33150 }, { "epoch": 8.934539190353144, "grad_norm": 0.32749006152153015, "learning_rate": 0.0004663026794129527, "loss": 3.2021, "step": 33200 }, { "epoch": 8.947997416020671, "grad_norm": 0.3676239252090454, "learning_rate": 0.0004661007136124949, "loss": 3.1997, "step": 33250 }, { "epoch": 8.9614556416882, "grad_norm": 0.3513721823692322, "learning_rate": 0.00046589874781203714, "loss": 3.2092, "step": 33300 }, { "epoch": 8.974913867355728, "grad_norm": 0.3342949450016022, "learning_rate": 0.0004656967820115794, "loss": 3.1986, "step": 33350 }, { "epoch": 8.988372093023255, "grad_norm": 0.3422151505947113, "learning_rate": 0.0004654948162111216, "loss": 3.204, "step": 33400 }, { "epoch": 9.001614987080103, "grad_norm": 0.3525255620479584, "learning_rate": 0.0004652928504106638, "loss": 3.1975, "step": 33450 }, { "epoch": 9.015073212747632, "grad_norm": 0.3380824625492096, "learning_rate": 0.000465090884610206, "loss": 3.0995, "step": 33500 }, { "epoch": 9.02853143841516, "grad_norm": 0.3548218309879303, "learning_rate": 0.00046488891880974817, "loss": 3.0869, "step": 33550 }, { "epoch": 9.041989664082687, "grad_norm": 0.3807477056980133, "learning_rate": 0.0004646869530092904, "loss": 3.1083, "step": 33600 }, { "epoch": 9.055447889750216, "grad_norm": 0.33358967304229736, "learning_rate": 0.0004644849872088326, "loss": 3.1118, "step": 33650 }, { "epoch": 9.068906115417743, "grad_norm": 0.3386094272136688, "learning_rate": 0.0004642830214083748, "loss": 3.1083, "step": 33700 }, { "epoch": 9.082364341085272, "grad_norm": 0.3573903441429138, "learning_rate": 0.000464081055607917, "loss": 3.1117, "step": 33750 }, { "epoch": 9.0958225667528, "grad_norm": 0.33198776841163635, "learning_rate": 0.0004638790898074592, "loss": 3.1315, "step": 33800 }, { "epoch": 9.109280792420327, "grad_norm": 0.35434484481811523, "learning_rate": 0.00046367712400700146, "loss": 3.1135, "step": 33850 }, { "epoch": 9.122739018087856, "grad_norm": 0.36608824133872986, "learning_rate": 0.00046347515820654365, "loss": 3.1209, "step": 33900 }, { "epoch": 9.136197243755383, "grad_norm": 0.3372901380062103, "learning_rate": 0.00046327319240608585, "loss": 3.1284, "step": 33950 }, { "epoch": 9.14965546942291, "grad_norm": 0.3538866639137268, "learning_rate": 0.00046307122660562805, "loss": 3.1258, "step": 34000 }, { "epoch": 9.14965546942291, "eval_accuracy": 0.3888806679434845, "eval_loss": 3.325159788131714, "eval_runtime": 87.3642, "eval_samples_per_second": 206.171, "eval_steps_per_second": 12.889, "step": 34000 }, { "epoch": 9.16311369509044, "grad_norm": 0.3983548879623413, "learning_rate": 0.00046286926080517024, "loss": 3.1304, "step": 34050 }, { "epoch": 9.176571920757967, "grad_norm": 0.36228519678115845, "learning_rate": 0.0004626672950047125, "loss": 3.1338, "step": 34100 }, { "epoch": 9.190030146425496, "grad_norm": 0.3778652846813202, "learning_rate": 0.0004624653292042547, "loss": 3.1291, "step": 34150 }, { "epoch": 9.203488372093023, "grad_norm": 0.36719369888305664, "learning_rate": 0.00046226336340379694, "loss": 3.1375, "step": 34200 }, { "epoch": 9.21694659776055, "grad_norm": 0.35282188653945923, "learning_rate": 0.00046206139760333913, "loss": 3.132, "step": 34250 }, { "epoch": 9.23040482342808, "grad_norm": 0.33646711707115173, "learning_rate": 0.0004618594318028814, "loss": 3.1454, "step": 34300 }, { "epoch": 9.243863049095607, "grad_norm": 0.3522685170173645, "learning_rate": 0.0004616574660024236, "loss": 3.1459, "step": 34350 }, { "epoch": 9.257321274763136, "grad_norm": 0.36446160078048706, "learning_rate": 0.0004614555002019658, "loss": 3.1453, "step": 34400 }, { "epoch": 9.270779500430663, "grad_norm": 0.3586257994174957, "learning_rate": 0.000461253534401508, "loss": 3.1475, "step": 34450 }, { "epoch": 9.28423772609819, "grad_norm": 0.4070618748664856, "learning_rate": 0.00046105156860105017, "loss": 3.1352, "step": 34500 }, { "epoch": 9.29769595176572, "grad_norm": 0.3892214894294739, "learning_rate": 0.0004608496028005924, "loss": 3.1476, "step": 34550 }, { "epoch": 9.311154177433247, "grad_norm": 0.3597724139690399, "learning_rate": 0.0004606476370001346, "loss": 3.1403, "step": 34600 }, { "epoch": 9.324612403100776, "grad_norm": 0.36641639471054077, "learning_rate": 0.0004604456711996768, "loss": 3.1488, "step": 34650 }, { "epoch": 9.338070628768303, "grad_norm": 0.373392790555954, "learning_rate": 0.000460243705399219, "loss": 3.1429, "step": 34700 }, { "epoch": 9.35152885443583, "grad_norm": 0.3421435058116913, "learning_rate": 0.0004600417395987612, "loss": 3.1677, "step": 34750 }, { "epoch": 9.36498708010336, "grad_norm": 0.3360295295715332, "learning_rate": 0.00045983977379830345, "loss": 3.1539, "step": 34800 }, { "epoch": 9.378445305770887, "grad_norm": 0.37752994894981384, "learning_rate": 0.00045963780799784565, "loss": 3.1446, "step": 34850 }, { "epoch": 9.391903531438416, "grad_norm": 0.3226233720779419, "learning_rate": 0.00045943584219738785, "loss": 3.1392, "step": 34900 }, { "epoch": 9.405361757105943, "grad_norm": 0.35676461458206177, "learning_rate": 0.00045923387639693004, "loss": 3.1527, "step": 34950 }, { "epoch": 9.41881998277347, "grad_norm": 0.3426799178123474, "learning_rate": 0.00045903191059647224, "loss": 3.157, "step": 35000 }, { "epoch": 9.41881998277347, "eval_accuracy": 0.3893731641397581, "eval_loss": 3.3207197189331055, "eval_runtime": 87.7887, "eval_samples_per_second": 205.174, "eval_steps_per_second": 12.826, "step": 35000 }, { "epoch": 9.432278208441, "grad_norm": 0.3415524363517761, "learning_rate": 0.0004588299447960145, "loss": 3.1635, "step": 35050 }, { "epoch": 9.445736434108527, "grad_norm": 0.3550635278224945, "learning_rate": 0.00045862797899555674, "loss": 3.1633, "step": 35100 }, { "epoch": 9.459194659776054, "grad_norm": 0.33996516466140747, "learning_rate": 0.00045842601319509894, "loss": 3.16, "step": 35150 }, { "epoch": 9.472652885443583, "grad_norm": 0.35329610109329224, "learning_rate": 0.00045822404739464113, "loss": 3.1616, "step": 35200 }, { "epoch": 9.48611111111111, "grad_norm": 0.33477357029914856, "learning_rate": 0.0004580220815941834, "loss": 3.1517, "step": 35250 }, { "epoch": 9.49956933677864, "grad_norm": 0.34619536995887756, "learning_rate": 0.0004578201157937256, "loss": 3.1549, "step": 35300 }, { "epoch": 9.513027562446167, "grad_norm": 0.3580850660800934, "learning_rate": 0.0004576181499932678, "loss": 3.1615, "step": 35350 }, { "epoch": 9.526485788113694, "grad_norm": 0.33334505558013916, "learning_rate": 0.00045741618419280997, "loss": 3.1563, "step": 35400 }, { "epoch": 9.539944013781223, "grad_norm": 0.3636830747127533, "learning_rate": 0.0004572142183923522, "loss": 3.1713, "step": 35450 }, { "epoch": 9.55340223944875, "grad_norm": 0.32588592171669006, "learning_rate": 0.0004570122525918944, "loss": 3.17, "step": 35500 }, { "epoch": 9.56686046511628, "grad_norm": 0.3375829756259918, "learning_rate": 0.0004568102867914366, "loss": 3.1677, "step": 35550 }, { "epoch": 9.580318690783807, "grad_norm": 0.36470961570739746, "learning_rate": 0.0004566083209909788, "loss": 3.1621, "step": 35600 }, { "epoch": 9.593776916451334, "grad_norm": 0.37833788990974426, "learning_rate": 0.000456406355190521, "loss": 3.166, "step": 35650 }, { "epoch": 9.607235142118864, "grad_norm": 0.3430384397506714, "learning_rate": 0.00045620438939006326, "loss": 3.172, "step": 35700 }, { "epoch": 9.62069336778639, "grad_norm": 0.3499787151813507, "learning_rate": 0.00045600242358960545, "loss": 3.1631, "step": 35750 }, { "epoch": 9.634151593453918, "grad_norm": 0.3460032343864441, "learning_rate": 0.00045580045778914765, "loss": 3.1615, "step": 35800 }, { "epoch": 9.647609819121447, "grad_norm": 0.3551907539367676, "learning_rate": 0.00045559849198868984, "loss": 3.167, "step": 35850 }, { "epoch": 9.661068044788975, "grad_norm": 0.3299188017845154, "learning_rate": 0.00045539652618823204, "loss": 3.1753, "step": 35900 }, { "epoch": 9.674526270456504, "grad_norm": 0.3797774612903595, "learning_rate": 0.0004551945603877743, "loss": 3.1723, "step": 35950 }, { "epoch": 9.687984496124031, "grad_norm": 0.345142126083374, "learning_rate": 0.00045499259458731654, "loss": 3.1741, "step": 36000 }, { "epoch": 9.687984496124031, "eval_accuracy": 0.3896962798882067, "eval_loss": 3.3145992755889893, "eval_runtime": 87.9367, "eval_samples_per_second": 204.829, "eval_steps_per_second": 12.805, "step": 36000 }, { "epoch": 9.701442721791558, "grad_norm": 0.3563687801361084, "learning_rate": 0.00045479062878685874, "loss": 3.1675, "step": 36050 }, { "epoch": 9.714900947459087, "grad_norm": 0.3506101071834564, "learning_rate": 0.00045458866298640093, "loss": 3.1727, "step": 36100 }, { "epoch": 9.728359173126615, "grad_norm": 0.3728286027908325, "learning_rate": 0.0004543866971859432, "loss": 3.1741, "step": 36150 }, { "epoch": 9.741817398794144, "grad_norm": 0.3504398465156555, "learning_rate": 0.0004541847313854854, "loss": 3.1642, "step": 36200 }, { "epoch": 9.755275624461671, "grad_norm": 0.34962764382362366, "learning_rate": 0.0004539827655850276, "loss": 3.176, "step": 36250 }, { "epoch": 9.768733850129198, "grad_norm": 0.3368580639362335, "learning_rate": 0.00045378079978456977, "loss": 3.1726, "step": 36300 }, { "epoch": 9.782192075796727, "grad_norm": 0.3463308811187744, "learning_rate": 0.00045357883398411197, "loss": 3.1703, "step": 36350 }, { "epoch": 9.795650301464255, "grad_norm": 0.34614890813827515, "learning_rate": 0.0004533768681836542, "loss": 3.169, "step": 36400 }, { "epoch": 9.809108527131784, "grad_norm": 0.3787212371826172, "learning_rate": 0.0004531749023831964, "loss": 3.1676, "step": 36450 }, { "epoch": 9.822566752799311, "grad_norm": 0.35479286313056946, "learning_rate": 0.0004529729365827386, "loss": 3.1922, "step": 36500 }, { "epoch": 9.836024978466838, "grad_norm": 0.3774144947528839, "learning_rate": 0.0004527709707822808, "loss": 3.1752, "step": 36550 }, { "epoch": 9.849483204134367, "grad_norm": 0.35198262333869934, "learning_rate": 0.000452569004981823, "loss": 3.17, "step": 36600 }, { "epoch": 9.862941429801895, "grad_norm": 0.3490941524505615, "learning_rate": 0.00045236703918136525, "loss": 3.1777, "step": 36650 }, { "epoch": 9.876399655469424, "grad_norm": 0.367546945810318, "learning_rate": 0.00045216507338090745, "loss": 3.1761, "step": 36700 }, { "epoch": 9.889857881136951, "grad_norm": 0.3417567312717438, "learning_rate": 0.00045196310758044964, "loss": 3.1851, "step": 36750 }, { "epoch": 9.903316106804478, "grad_norm": 0.3448176383972168, "learning_rate": 0.00045176114177999184, "loss": 3.1708, "step": 36800 }, { "epoch": 9.916774332472007, "grad_norm": 0.35334399342536926, "learning_rate": 0.00045155917597953404, "loss": 3.1809, "step": 36850 }, { "epoch": 9.930232558139535, "grad_norm": 0.36826086044311523, "learning_rate": 0.00045135721017907634, "loss": 3.1849, "step": 36900 }, { "epoch": 9.943690783807062, "grad_norm": 0.3243925869464874, "learning_rate": 0.00045115524437861854, "loss": 3.1786, "step": 36950 }, { "epoch": 9.957149009474591, "grad_norm": 0.35191184282302856, "learning_rate": 0.00045095327857816073, "loss": 3.1805, "step": 37000 }, { "epoch": 9.957149009474591, "eval_accuracy": 0.39071288851572317, "eval_loss": 3.3055007457733154, "eval_runtime": 88.172, "eval_samples_per_second": 204.283, "eval_steps_per_second": 12.77, "step": 37000 }, { "epoch": 9.970607235142118, "grad_norm": 0.38593989610671997, "learning_rate": 0.00045075131277770293, "loss": 3.1912, "step": 37050 }, { "epoch": 9.984065460809648, "grad_norm": 0.36267486214637756, "learning_rate": 0.0004505493469772452, "loss": 3.1848, "step": 37100 }, { "epoch": 9.997523686477175, "grad_norm": 0.3603392243385315, "learning_rate": 0.0004503473811767874, "loss": 3.1785, "step": 37150 }, { "epoch": 10.010766580534023, "grad_norm": 0.3506993353366852, "learning_rate": 0.00045014541537632957, "loss": 3.1058, "step": 37200 }, { "epoch": 10.02422480620155, "grad_norm": 0.348012238740921, "learning_rate": 0.00044994344957587177, "loss": 3.0732, "step": 37250 }, { "epoch": 10.037683031869078, "grad_norm": 0.34269097447395325, "learning_rate": 0.00044974148377541396, "loss": 3.0742, "step": 37300 }, { "epoch": 10.051141257536607, "grad_norm": 0.3394729495048523, "learning_rate": 0.0004495395179749562, "loss": 3.0842, "step": 37350 }, { "epoch": 10.064599483204134, "grad_norm": 0.3448140621185303, "learning_rate": 0.0004493375521744984, "loss": 3.0895, "step": 37400 }, { "epoch": 10.078057708871663, "grad_norm": 0.38341960310935974, "learning_rate": 0.0004491355863740406, "loss": 3.0889, "step": 37450 }, { "epoch": 10.09151593453919, "grad_norm": 0.3606523871421814, "learning_rate": 0.0004489336205735828, "loss": 3.0876, "step": 37500 }, { "epoch": 10.104974160206718, "grad_norm": 0.3559594750404358, "learning_rate": 0.00044873165477312505, "loss": 3.0941, "step": 37550 }, { "epoch": 10.118432385874247, "grad_norm": 0.35412347316741943, "learning_rate": 0.00044852968897266725, "loss": 3.1045, "step": 37600 }, { "epoch": 10.131890611541774, "grad_norm": 0.3366471529006958, "learning_rate": 0.00044832772317220945, "loss": 3.1012, "step": 37650 }, { "epoch": 10.145348837209303, "grad_norm": 0.4085768163204193, "learning_rate": 0.00044812575737175164, "loss": 3.1144, "step": 37700 }, { "epoch": 10.15880706287683, "grad_norm": 0.3462596833705902, "learning_rate": 0.00044792379157129384, "loss": 3.1093, "step": 37750 }, { "epoch": 10.172265288544358, "grad_norm": 0.40925300121307373, "learning_rate": 0.00044772182577083614, "loss": 3.1168, "step": 37800 }, { "epoch": 10.185723514211887, "grad_norm": 0.3763587772846222, "learning_rate": 0.00044751985997037834, "loss": 3.1081, "step": 37850 }, { "epoch": 10.199181739879414, "grad_norm": 0.35570356249809265, "learning_rate": 0.00044731789416992053, "loss": 3.1153, "step": 37900 }, { "epoch": 10.212639965546943, "grad_norm": 0.35475507378578186, "learning_rate": 0.00044711592836946273, "loss": 3.1183, "step": 37950 }, { "epoch": 10.22609819121447, "grad_norm": 0.35603535175323486, "learning_rate": 0.000446913962569005, "loss": 3.1222, "step": 38000 }, { "epoch": 10.22609819121447, "eval_accuracy": 0.3902001840043146, "eval_loss": 3.3156235218048096, "eval_runtime": 87.8345, "eval_samples_per_second": 205.068, "eval_steps_per_second": 12.82, "step": 38000 }, { "epoch": 10.239556416881998, "grad_norm": 0.3575981557369232, "learning_rate": 0.0004467119967685472, "loss": 3.1125, "step": 38050 }, { "epoch": 10.253014642549527, "grad_norm": 0.37977248430252075, "learning_rate": 0.00044651003096808937, "loss": 3.121, "step": 38100 }, { "epoch": 10.266472868217054, "grad_norm": 0.3563908040523529, "learning_rate": 0.00044630806516763157, "loss": 3.1301, "step": 38150 }, { "epoch": 10.279931093884581, "grad_norm": 0.38932687044143677, "learning_rate": 0.00044610609936717376, "loss": 3.1217, "step": 38200 }, { "epoch": 10.29338931955211, "grad_norm": 0.36320436000823975, "learning_rate": 0.000445904133566716, "loss": 3.1087, "step": 38250 }, { "epoch": 10.306847545219638, "grad_norm": 0.37047919631004333, "learning_rate": 0.0004457021677662582, "loss": 3.1263, "step": 38300 }, { "epoch": 10.320305770887167, "grad_norm": 0.3520812690258026, "learning_rate": 0.0004455002019658004, "loss": 3.1224, "step": 38350 }, { "epoch": 10.333763996554694, "grad_norm": 0.3531203269958496, "learning_rate": 0.0004452982361653426, "loss": 3.1409, "step": 38400 }, { "epoch": 10.347222222222221, "grad_norm": 0.36067909002304077, "learning_rate": 0.0004450962703648848, "loss": 3.1286, "step": 38450 }, { "epoch": 10.36068044788975, "grad_norm": 0.34704872965812683, "learning_rate": 0.00044489430456442705, "loss": 3.1307, "step": 38500 }, { "epoch": 10.374138673557278, "grad_norm": 0.377408504486084, "learning_rate": 0.00044469233876396925, "loss": 3.1355, "step": 38550 }, { "epoch": 10.387596899224807, "grad_norm": 0.3663313686847687, "learning_rate": 0.00044449037296351144, "loss": 3.1439, "step": 38600 }, { "epoch": 10.401055124892334, "grad_norm": 0.3542163074016571, "learning_rate": 0.0004442884071630537, "loss": 3.1279, "step": 38650 }, { "epoch": 10.414513350559861, "grad_norm": 0.33422955870628357, "learning_rate": 0.00044408644136259594, "loss": 3.1333, "step": 38700 }, { "epoch": 10.42797157622739, "grad_norm": 0.3569745421409607, "learning_rate": 0.00044388447556213814, "loss": 3.1405, "step": 38750 }, { "epoch": 10.441429801894918, "grad_norm": 0.37442442774772644, "learning_rate": 0.00044368250976168033, "loss": 3.1325, "step": 38800 }, { "epoch": 10.454888027562447, "grad_norm": 0.35368022322654724, "learning_rate": 0.00044348054396122253, "loss": 3.1391, "step": 38850 }, { "epoch": 10.468346253229974, "grad_norm": 0.3554447293281555, "learning_rate": 0.0004432785781607647, "loss": 3.1434, "step": 38900 }, { "epoch": 10.481804478897502, "grad_norm": 0.36405107378959656, "learning_rate": 0.000443076612360307, "loss": 3.1422, "step": 38950 }, { "epoch": 10.49526270456503, "grad_norm": 0.34430205821990967, "learning_rate": 0.0004428746465598492, "loss": 3.1334, "step": 39000 }, { "epoch": 10.49526270456503, "eval_accuracy": 0.3905743637748785, "eval_loss": 3.314378023147583, "eval_runtime": 88.0967, "eval_samples_per_second": 204.457, "eval_steps_per_second": 12.781, "step": 39000 }, { "epoch": 10.508720930232558, "grad_norm": 0.33116546273231506, "learning_rate": 0.00044267268075939137, "loss": 3.147, "step": 39050 }, { "epoch": 10.522179155900087, "grad_norm": 0.344468355178833, "learning_rate": 0.00044247071495893357, "loss": 3.1402, "step": 39100 }, { "epoch": 10.535637381567614, "grad_norm": 0.3650263845920563, "learning_rate": 0.00044226874915847576, "loss": 3.1404, "step": 39150 }, { "epoch": 10.549095607235142, "grad_norm": 0.3484593629837036, "learning_rate": 0.000442066783358018, "loss": 3.1412, "step": 39200 }, { "epoch": 10.56255383290267, "grad_norm": 0.351365864276886, "learning_rate": 0.0004418648175575602, "loss": 3.1512, "step": 39250 }, { "epoch": 10.576012058570198, "grad_norm": 0.3528461456298828, "learning_rate": 0.0004416628517571024, "loss": 3.1396, "step": 39300 }, { "epoch": 10.589470284237725, "grad_norm": 0.3310932517051697, "learning_rate": 0.0004414608859566446, "loss": 3.1524, "step": 39350 }, { "epoch": 10.602928509905254, "grad_norm": 0.3519033193588257, "learning_rate": 0.00044125892015618685, "loss": 3.1505, "step": 39400 }, { "epoch": 10.616386735572782, "grad_norm": 0.3464779257774353, "learning_rate": 0.00044105695435572905, "loss": 3.143, "step": 39450 }, { "epoch": 10.62984496124031, "grad_norm": 0.3933066427707672, "learning_rate": 0.00044085498855527124, "loss": 3.1493, "step": 39500 }, { "epoch": 10.643303186907838, "grad_norm": 0.3580518364906311, "learning_rate": 0.0004406530227548135, "loss": 3.1533, "step": 39550 }, { "epoch": 10.656761412575365, "grad_norm": 0.424843966960907, "learning_rate": 0.00044045105695435574, "loss": 3.1308, "step": 39600 }, { "epoch": 10.670219638242894, "grad_norm": 0.35034167766571045, "learning_rate": 0.00044024909115389794, "loss": 3.1618, "step": 39650 }, { "epoch": 10.683677863910422, "grad_norm": 0.34508782625198364, "learning_rate": 0.00044004712535344014, "loss": 3.1489, "step": 39700 }, { "epoch": 10.69713608957795, "grad_norm": 0.3473265469074249, "learning_rate": 0.00043984515955298233, "loss": 3.1419, "step": 39750 }, { "epoch": 10.710594315245478, "grad_norm": 0.3852803409099579, "learning_rate": 0.00043964319375252453, "loss": 3.1469, "step": 39800 }, { "epoch": 10.724052540913005, "grad_norm": 0.3702237904071808, "learning_rate": 0.0004394412279520668, "loss": 3.1483, "step": 39850 }, { "epoch": 10.737510766580534, "grad_norm": 0.3771539330482483, "learning_rate": 0.000439239262151609, "loss": 3.1567, "step": 39900 }, { "epoch": 10.750968992248062, "grad_norm": 0.35787418484687805, "learning_rate": 0.00043903729635115117, "loss": 3.1577, "step": 39950 }, { "epoch": 10.764427217915589, "grad_norm": 0.35247600078582764, "learning_rate": 0.00043883533055069337, "loss": 3.1497, "step": 40000 }, { "epoch": 10.764427217915589, "eval_accuracy": 0.39091518896078414, "eval_loss": 3.306220054626465, "eval_runtime": 88.1301, "eval_samples_per_second": 204.38, "eval_steps_per_second": 12.777, "step": 40000 }, { "epoch": 10.777885443583118, "grad_norm": 0.35555049777030945, "learning_rate": 0.00043863336475023556, "loss": 3.1444, "step": 40050 }, { "epoch": 10.791343669250645, "grad_norm": 0.3686332702636719, "learning_rate": 0.0004384313989497778, "loss": 3.1548, "step": 40100 }, { "epoch": 10.804801894918175, "grad_norm": 0.3736445903778076, "learning_rate": 0.00043822943314932, "loss": 3.1487, "step": 40150 }, { "epoch": 10.818260120585702, "grad_norm": 0.3347689211368561, "learning_rate": 0.0004380274673488622, "loss": 3.1531, "step": 40200 }, { "epoch": 10.83171834625323, "grad_norm": 0.34181126952171326, "learning_rate": 0.0004378255015484044, "loss": 3.1399, "step": 40250 }, { "epoch": 10.845176571920758, "grad_norm": 0.3598361909389496, "learning_rate": 0.0004376235357479466, "loss": 3.1542, "step": 40300 }, { "epoch": 10.858634797588286, "grad_norm": 0.34456607699394226, "learning_rate": 0.00043742156994748885, "loss": 3.1559, "step": 40350 }, { "epoch": 10.872093023255815, "grad_norm": 0.3753799498081207, "learning_rate": 0.00043721960414703104, "loss": 3.1669, "step": 40400 }, { "epoch": 10.885551248923342, "grad_norm": 0.36482125520706177, "learning_rate": 0.0004370176383465733, "loss": 3.1523, "step": 40450 }, { "epoch": 10.89900947459087, "grad_norm": 0.38292697072029114, "learning_rate": 0.0004368156725461155, "loss": 3.161, "step": 40500 }, { "epoch": 10.912467700258398, "grad_norm": 0.32300931215286255, "learning_rate": 0.00043661370674565774, "loss": 3.1547, "step": 40550 }, { "epoch": 10.925925925925926, "grad_norm": 0.35990387201309204, "learning_rate": 0.00043641174094519994, "loss": 3.1549, "step": 40600 }, { "epoch": 10.939384151593455, "grad_norm": 0.38015124201774597, "learning_rate": 0.00043620977514474213, "loss": 3.1514, "step": 40650 }, { "epoch": 10.952842377260982, "grad_norm": 0.3817886710166931, "learning_rate": 0.00043600780934428433, "loss": 3.1636, "step": 40700 }, { "epoch": 10.96630060292851, "grad_norm": 0.3476899564266205, "learning_rate": 0.0004358058435438265, "loss": 3.1509, "step": 40750 }, { "epoch": 10.979758828596038, "grad_norm": 0.34082624316215515, "learning_rate": 0.0004356038777433688, "loss": 3.1678, "step": 40800 }, { "epoch": 10.993217054263566, "grad_norm": 0.34645843505859375, "learning_rate": 0.00043540191194291097, "loss": 3.1662, "step": 40850 }, { "epoch": 11.006459948320414, "grad_norm": 0.3485479950904846, "learning_rate": 0.00043519994614245317, "loss": 3.116, "step": 40900 }, { "epoch": 11.019918173987941, "grad_norm": 0.3675366938114166, "learning_rate": 0.00043499798034199536, "loss": 3.0632, "step": 40950 }, { "epoch": 11.03337639965547, "grad_norm": 0.3343339264392853, "learning_rate": 0.00043479601454153756, "loss": 3.0698, "step": 41000 }, { "epoch": 11.03337639965547, "eval_accuracy": 0.3909793992524227, "eval_loss": 3.31091046333313, "eval_runtime": 84.2407, "eval_samples_per_second": 213.816, "eval_steps_per_second": 13.366, "step": 41000 }, { "epoch": 11.046834625322997, "grad_norm": 0.36293190717697144, "learning_rate": 0.0004345940487410798, "loss": 3.0706, "step": 41050 }, { "epoch": 11.060292850990525, "grad_norm": 0.3535679876804352, "learning_rate": 0.000434392082940622, "loss": 3.0633, "step": 41100 }, { "epoch": 11.073751076658054, "grad_norm": 0.33150696754455566, "learning_rate": 0.0004341901171401642, "loss": 3.0755, "step": 41150 }, { "epoch": 11.087209302325581, "grad_norm": 0.36507776379585266, "learning_rate": 0.0004339881513397064, "loss": 3.0763, "step": 41200 }, { "epoch": 11.10066752799311, "grad_norm": 0.36090418696403503, "learning_rate": 0.0004337861855392486, "loss": 3.0905, "step": 41250 }, { "epoch": 11.114125753660637, "grad_norm": 0.38765281438827515, "learning_rate": 0.00043358421973879084, "loss": 3.0822, "step": 41300 }, { "epoch": 11.127583979328165, "grad_norm": 0.33757099509239197, "learning_rate": 0.0004333822539383331, "loss": 3.0931, "step": 41350 }, { "epoch": 11.141042204995694, "grad_norm": 0.35109296441078186, "learning_rate": 0.0004331802881378753, "loss": 3.0877, "step": 41400 }, { "epoch": 11.154500430663221, "grad_norm": 0.3479686677455902, "learning_rate": 0.00043297832233741754, "loss": 3.0916, "step": 41450 }, { "epoch": 11.16795865633075, "grad_norm": 0.3615115284919739, "learning_rate": 0.00043277635653695974, "loss": 3.0859, "step": 41500 }, { "epoch": 11.181416881998278, "grad_norm": 0.37314116954803467, "learning_rate": 0.00043257439073650193, "loss": 3.1043, "step": 41550 }, { "epoch": 11.194875107665805, "grad_norm": 0.3375728726387024, "learning_rate": 0.00043237242493604413, "loss": 3.1011, "step": 41600 }, { "epoch": 11.208333333333334, "grad_norm": 0.3412526547908783, "learning_rate": 0.0004321704591355863, "loss": 3.0902, "step": 41650 }, { "epoch": 11.221791559000861, "grad_norm": 0.3929307162761688, "learning_rate": 0.0004319684933351286, "loss": 3.0889, "step": 41700 }, { "epoch": 11.235249784668389, "grad_norm": 0.36704355478286743, "learning_rate": 0.00043176652753467077, "loss": 3.0918, "step": 41750 }, { "epoch": 11.248708010335918, "grad_norm": 0.3677278161048889, "learning_rate": 0.00043156456173421297, "loss": 3.1011, "step": 41800 }, { "epoch": 11.262166236003445, "grad_norm": 0.37482133507728577, "learning_rate": 0.00043136259593375516, "loss": 3.0974, "step": 41850 }, { "epoch": 11.275624461670974, "grad_norm": 0.36803311109542847, "learning_rate": 0.00043116063013329736, "loss": 3.1111, "step": 41900 }, { "epoch": 11.289082687338501, "grad_norm": 0.4043966829776764, "learning_rate": 0.0004309586643328396, "loss": 3.1013, "step": 41950 }, { "epoch": 11.302540913006029, "grad_norm": 0.37155815958976746, "learning_rate": 0.0004307566985323818, "loss": 3.1033, "step": 42000 }, { "epoch": 11.302540913006029, "eval_accuracy": 0.39119245573618455, "eval_loss": 3.309682846069336, "eval_runtime": 84.3256, "eval_samples_per_second": 213.601, "eval_steps_per_second": 13.353, "step": 42000 }, { "epoch": 11.315999138673558, "grad_norm": 0.3757474422454834, "learning_rate": 0.000430554732731924, "loss": 3.1042, "step": 42050 }, { "epoch": 11.329457364341085, "grad_norm": 0.3633876144886017, "learning_rate": 0.0004303527669314662, "loss": 3.0951, "step": 42100 }, { "epoch": 11.342915590008614, "grad_norm": 0.34023556113243103, "learning_rate": 0.0004301508011310084, "loss": 3.102, "step": 42150 }, { "epoch": 11.356373815676141, "grad_norm": 0.3511642515659332, "learning_rate": 0.00042994883533055065, "loss": 3.103, "step": 42200 }, { "epoch": 11.369832041343669, "grad_norm": 0.37258774042129517, "learning_rate": 0.0004297468695300929, "loss": 3.1049, "step": 42250 }, { "epoch": 11.383290267011198, "grad_norm": 0.3569978177547455, "learning_rate": 0.0004295449037296351, "loss": 3.1066, "step": 42300 }, { "epoch": 11.396748492678725, "grad_norm": 0.36627477407455444, "learning_rate": 0.0004293429379291773, "loss": 3.1104, "step": 42350 }, { "epoch": 11.410206718346252, "grad_norm": 0.33913758397102356, "learning_rate": 0.00042914097212871954, "loss": 3.1198, "step": 42400 }, { "epoch": 11.423664944013781, "grad_norm": 0.3415125608444214, "learning_rate": 0.00042893900632826173, "loss": 3.1135, "step": 42450 }, { "epoch": 11.437123169681309, "grad_norm": 0.38717105984687805, "learning_rate": 0.00042873704052780393, "loss": 3.1106, "step": 42500 }, { "epoch": 11.450581395348838, "grad_norm": 0.3576693534851074, "learning_rate": 0.0004285350747273461, "loss": 3.108, "step": 42550 }, { "epoch": 11.464039621016365, "grad_norm": 0.3778170943260193, "learning_rate": 0.0004283331089268883, "loss": 3.1131, "step": 42600 }, { "epoch": 11.477497846683892, "grad_norm": 0.3914255201816559, "learning_rate": 0.00042813114312643057, "loss": 3.1215, "step": 42650 }, { "epoch": 11.490956072351421, "grad_norm": 0.3620469570159912, "learning_rate": 0.00042792917732597277, "loss": 3.1206, "step": 42700 }, { "epoch": 11.504414298018949, "grad_norm": 0.36856579780578613, "learning_rate": 0.00042772721152551496, "loss": 3.1164, "step": 42750 }, { "epoch": 11.517872523686478, "grad_norm": 0.35470420122146606, "learning_rate": 0.00042752524572505716, "loss": 3.1205, "step": 42800 }, { "epoch": 11.531330749354005, "grad_norm": 0.3629094362258911, "learning_rate": 0.00042732327992459936, "loss": 3.1215, "step": 42850 }, { "epoch": 11.544788975021532, "grad_norm": 0.3661877512931824, "learning_rate": 0.0004271213141241416, "loss": 3.1351, "step": 42900 }, { "epoch": 11.558247200689062, "grad_norm": 0.3606022000312805, "learning_rate": 0.0004269193483236838, "loss": 3.1315, "step": 42950 }, { "epoch": 11.571705426356589, "grad_norm": 0.3552758991718292, "learning_rate": 0.000426717382523226, "loss": 3.1317, "step": 43000 }, { "epoch": 11.571705426356589, "eval_accuracy": 0.3915242632330784, "eval_loss": 3.3031299114227295, "eval_runtime": 84.278, "eval_samples_per_second": 213.721, "eval_steps_per_second": 13.361, "step": 43000 }, { "epoch": 11.585163652024118, "grad_norm": 0.3457125127315521, "learning_rate": 0.0004265154167227682, "loss": 3.127, "step": 43050 }, { "epoch": 11.598621877691645, "grad_norm": 0.36335304379463196, "learning_rate": 0.0004263134509223105, "loss": 3.1305, "step": 43100 }, { "epoch": 11.612080103359173, "grad_norm": 0.3702634572982788, "learning_rate": 0.0004261114851218527, "loss": 3.1314, "step": 43150 }, { "epoch": 11.625538329026702, "grad_norm": 0.3538024425506592, "learning_rate": 0.0004259095193213949, "loss": 3.1253, "step": 43200 }, { "epoch": 11.638996554694229, "grad_norm": 0.3644096851348877, "learning_rate": 0.0004257075535209371, "loss": 3.1377, "step": 43250 }, { "epoch": 11.652454780361758, "grad_norm": 0.3496798574924469, "learning_rate": 0.00042550558772047934, "loss": 3.1354, "step": 43300 }, { "epoch": 11.665913006029285, "grad_norm": 0.3464163839817047, "learning_rate": 0.00042530362192002153, "loss": 3.1224, "step": 43350 }, { "epoch": 11.679371231696813, "grad_norm": 0.3821980953216553, "learning_rate": 0.00042510165611956373, "loss": 3.1225, "step": 43400 }, { "epoch": 11.692829457364342, "grad_norm": 0.3555978238582611, "learning_rate": 0.0004248996903191059, "loss": 3.1304, "step": 43450 }, { "epoch": 11.706287683031869, "grad_norm": 0.34647437930107117, "learning_rate": 0.0004246977245186481, "loss": 3.1398, "step": 43500 }, { "epoch": 11.719745908699396, "grad_norm": 0.3689185678958893, "learning_rate": 0.0004244957587181904, "loss": 3.1278, "step": 43550 }, { "epoch": 11.733204134366925, "grad_norm": 0.3581877648830414, "learning_rate": 0.00042429379291773257, "loss": 3.1262, "step": 43600 }, { "epoch": 11.746662360034453, "grad_norm": 0.37423980236053467, "learning_rate": 0.00042409182711727477, "loss": 3.1397, "step": 43650 }, { "epoch": 11.760120585701982, "grad_norm": 0.3919681906700134, "learning_rate": 0.00042388986131681696, "loss": 3.1285, "step": 43700 }, { "epoch": 11.773578811369509, "grad_norm": 0.348802387714386, "learning_rate": 0.00042368789551635916, "loss": 3.1447, "step": 43750 }, { "epoch": 11.787037037037036, "grad_norm": 0.3656277358531952, "learning_rate": 0.0004234859297159014, "loss": 3.1322, "step": 43800 }, { "epoch": 11.800495262704565, "grad_norm": 0.36101657152175903, "learning_rate": 0.0004232839639154436, "loss": 3.1468, "step": 43850 }, { "epoch": 11.813953488372093, "grad_norm": 0.35467052459716797, "learning_rate": 0.0004230819981149858, "loss": 3.1376, "step": 43900 }, { "epoch": 11.827411714039622, "grad_norm": 0.35919854044914246, "learning_rate": 0.000422880032314528, "loss": 3.1309, "step": 43950 }, { "epoch": 11.840869939707149, "grad_norm": 0.33611345291137695, "learning_rate": 0.0004226780665140703, "loss": 3.1351, "step": 44000 }, { "epoch": 11.840869939707149, "eval_accuracy": 0.3920807524272794, "eval_loss": 3.2992093563079834, "eval_runtime": 84.2871, "eval_samples_per_second": 213.698, "eval_steps_per_second": 13.359, "step": 44000 }, { "epoch": 11.854328165374676, "grad_norm": 0.36410412192344666, "learning_rate": 0.0004224761007136125, "loss": 3.1421, "step": 44050 }, { "epoch": 11.867786391042205, "grad_norm": 0.3480066657066345, "learning_rate": 0.0004222741349131547, "loss": 3.137, "step": 44100 }, { "epoch": 11.881244616709733, "grad_norm": 0.36214300990104675, "learning_rate": 0.0004220721691126969, "loss": 3.1478, "step": 44150 }, { "epoch": 11.89470284237726, "grad_norm": 0.3660474419593811, "learning_rate": 0.0004218702033122391, "loss": 3.1228, "step": 44200 }, { "epoch": 11.90816106804479, "grad_norm": 0.3615580201148987, "learning_rate": 0.00042166823751178134, "loss": 3.1344, "step": 44250 }, { "epoch": 11.921619293712316, "grad_norm": 0.3521454632282257, "learning_rate": 0.00042146627171132353, "loss": 3.1452, "step": 44300 }, { "epoch": 11.935077519379846, "grad_norm": 0.3320244252681732, "learning_rate": 0.00042126430591086573, "loss": 3.133, "step": 44350 }, { "epoch": 11.948535745047373, "grad_norm": 0.3705757260322571, "learning_rate": 0.0004210623401104079, "loss": 3.1322, "step": 44400 }, { "epoch": 11.9619939707149, "grad_norm": 0.34349524974823, "learning_rate": 0.0004208603743099501, "loss": 3.1307, "step": 44450 }, { "epoch": 11.97545219638243, "grad_norm": 0.3370743691921234, "learning_rate": 0.00042065840850949237, "loss": 3.1353, "step": 44500 }, { "epoch": 11.988910422049956, "grad_norm": 0.3711462616920471, "learning_rate": 0.00042045644270903457, "loss": 3.1402, "step": 44550 }, { "epoch": 12.002153316106805, "grad_norm": 0.3837710916996002, "learning_rate": 0.00042025447690857676, "loss": 3.1314, "step": 44600 }, { "epoch": 12.015611541774332, "grad_norm": 0.38589784502983093, "learning_rate": 0.00042005251110811896, "loss": 3.0303, "step": 44650 }, { "epoch": 12.029069767441861, "grad_norm": 0.3585261106491089, "learning_rate": 0.00041985054530766115, "loss": 3.0456, "step": 44700 }, { "epoch": 12.042527993109388, "grad_norm": 0.3696274161338806, "learning_rate": 0.0004196485795072034, "loss": 3.0529, "step": 44750 }, { "epoch": 12.055986218776916, "grad_norm": 0.3691428005695343, "learning_rate": 0.0004194466137067456, "loss": 3.0489, "step": 44800 }, { "epoch": 12.069444444444445, "grad_norm": 0.35204964876174927, "learning_rate": 0.0004192446479062878, "loss": 3.0474, "step": 44850 }, { "epoch": 12.082902670111972, "grad_norm": 0.38218945264816284, "learning_rate": 0.00041904268210583005, "loss": 3.0588, "step": 44900 }, { "epoch": 12.096360895779501, "grad_norm": 0.3627367913722992, "learning_rate": 0.0004188407163053723, "loss": 3.0597, "step": 44950 }, { "epoch": 12.109819121447028, "grad_norm": 0.3626028597354889, "learning_rate": 0.0004186387505049145, "loss": 3.0598, "step": 45000 }, { "epoch": 12.109819121447028, "eval_accuracy": 0.39129393189928174, "eval_loss": 3.307757616043091, "eval_runtime": 84.2296, "eval_samples_per_second": 213.844, "eval_steps_per_second": 13.368, "step": 45000 }, { "epoch": 12.123277347114556, "grad_norm": 0.39511799812316895, "learning_rate": 0.0004184367847044567, "loss": 3.0587, "step": 45050 }, { "epoch": 12.136735572782085, "grad_norm": 0.4090312719345093, "learning_rate": 0.0004182348189039989, "loss": 3.0599, "step": 45100 }, { "epoch": 12.150193798449612, "grad_norm": 0.39701277017593384, "learning_rate": 0.00041803285310354114, "loss": 3.0677, "step": 45150 }, { "epoch": 12.163652024117141, "grad_norm": 0.37128493189811707, "learning_rate": 0.00041783088730308333, "loss": 3.082, "step": 45200 }, { "epoch": 12.177110249784668, "grad_norm": 0.39566662907600403, "learning_rate": 0.00041762892150262553, "loss": 3.0731, "step": 45250 }, { "epoch": 12.190568475452196, "grad_norm": 0.32580286264419556, "learning_rate": 0.0004174269557021677, "loss": 3.0668, "step": 45300 }, { "epoch": 12.204026701119725, "grad_norm": 0.3477148115634918, "learning_rate": 0.0004172249899017099, "loss": 3.0767, "step": 45350 }, { "epoch": 12.217484926787252, "grad_norm": 0.3254866600036621, "learning_rate": 0.00041702302410125217, "loss": 3.0774, "step": 45400 }, { "epoch": 12.230943152454781, "grad_norm": 0.37678778171539307, "learning_rate": 0.00041682105830079437, "loss": 3.0851, "step": 45450 }, { "epoch": 12.244401378122308, "grad_norm": 0.38609495759010315, "learning_rate": 0.00041661909250033656, "loss": 3.0808, "step": 45500 }, { "epoch": 12.257859603789836, "grad_norm": 0.3586466610431671, "learning_rate": 0.00041641712669987876, "loss": 3.0884, "step": 45550 }, { "epoch": 12.271317829457365, "grad_norm": 0.37479081749916077, "learning_rate": 0.00041621516089942096, "loss": 3.0768, "step": 45600 }, { "epoch": 12.284776055124892, "grad_norm": 0.3411691188812256, "learning_rate": 0.0004160131950989632, "loss": 3.0837, "step": 45650 }, { "epoch": 12.298234280792421, "grad_norm": 0.36653265357017517, "learning_rate": 0.0004158112292985054, "loss": 3.0925, "step": 45700 }, { "epoch": 12.311692506459949, "grad_norm": 0.3570050001144409, "learning_rate": 0.0004156092634980476, "loss": 3.0891, "step": 45750 }, { "epoch": 12.325150732127476, "grad_norm": 0.37854400277137756, "learning_rate": 0.00041540729769758985, "loss": 3.0932, "step": 45800 }, { "epoch": 12.338608957795005, "grad_norm": 0.34331968426704407, "learning_rate": 0.0004152053318971321, "loss": 3.1047, "step": 45850 }, { "epoch": 12.352067183462532, "grad_norm": 0.38357841968536377, "learning_rate": 0.0004150033660966743, "loss": 3.0985, "step": 45900 }, { "epoch": 12.36552540913006, "grad_norm": 0.3992276191711426, "learning_rate": 0.0004148014002962165, "loss": 3.1048, "step": 45950 }, { "epoch": 12.378983634797589, "grad_norm": 0.4186655580997467, "learning_rate": 0.0004145994344957587, "loss": 3.0973, "step": 46000 }, { "epoch": 12.378983634797589, "eval_accuracy": 0.39201675942935194, "eval_loss": 3.3033642768859863, "eval_runtime": 84.2981, "eval_samples_per_second": 213.67, "eval_steps_per_second": 13.357, "step": 46000 }, { "epoch": 12.392441860465116, "grad_norm": 0.36977341771125793, "learning_rate": 0.0004143974686953009, "loss": 3.092, "step": 46050 }, { "epoch": 12.405900086132645, "grad_norm": 0.34746086597442627, "learning_rate": 0.00041419550289484313, "loss": 3.0894, "step": 46100 }, { "epoch": 12.419358311800172, "grad_norm": 0.37170976400375366, "learning_rate": 0.00041399353709438533, "loss": 3.0844, "step": 46150 }, { "epoch": 12.4328165374677, "grad_norm": 0.37848731875419617, "learning_rate": 0.0004137915712939275, "loss": 3.0849, "step": 46200 }, { "epoch": 12.446274763135229, "grad_norm": 0.3569401204586029, "learning_rate": 0.0004135896054934697, "loss": 3.1005, "step": 46250 }, { "epoch": 12.459732988802756, "grad_norm": 0.3888532519340515, "learning_rate": 0.0004133876396930119, "loss": 3.0948, "step": 46300 }, { "epoch": 12.473191214470285, "grad_norm": 0.3701384365558624, "learning_rate": 0.00041318567389255417, "loss": 3.1017, "step": 46350 }, { "epoch": 12.486649440137812, "grad_norm": 0.3526800572872162, "learning_rate": 0.00041298370809209636, "loss": 3.1008, "step": 46400 }, { "epoch": 12.50010766580534, "grad_norm": 0.3700055778026581, "learning_rate": 0.00041278174229163856, "loss": 3.1032, "step": 46450 }, { "epoch": 12.513565891472869, "grad_norm": 0.36047184467315674, "learning_rate": 0.00041257977649118076, "loss": 3.1158, "step": 46500 }, { "epoch": 12.527024117140396, "grad_norm": 0.359749436378479, "learning_rate": 0.00041237781069072295, "loss": 3.1017, "step": 46550 }, { "epoch": 12.540482342807923, "grad_norm": 0.35695669054985046, "learning_rate": 0.0004121758448902652, "loss": 3.0996, "step": 46600 }, { "epoch": 12.553940568475452, "grad_norm": 0.36963951587677, "learning_rate": 0.0004119738790898074, "loss": 3.1095, "step": 46650 }, { "epoch": 12.56739879414298, "grad_norm": 0.3756096661090851, "learning_rate": 0.00041177191328934965, "loss": 3.1052, "step": 46700 }, { "epoch": 12.580857019810509, "grad_norm": 0.3856138586997986, "learning_rate": 0.00041156994748889185, "loss": 3.1089, "step": 46750 }, { "epoch": 12.594315245478036, "grad_norm": 0.3549295663833618, "learning_rate": 0.0004113679816884341, "loss": 3.0999, "step": 46800 }, { "epoch": 12.607773471145563, "grad_norm": 0.36051085591316223, "learning_rate": 0.0004111660158879763, "loss": 3.1071, "step": 46850 }, { "epoch": 12.621231696813092, "grad_norm": 0.37265828251838684, "learning_rate": 0.0004109640500875185, "loss": 3.1097, "step": 46900 }, { "epoch": 12.63468992248062, "grad_norm": 0.3527121841907501, "learning_rate": 0.0004107620842870607, "loss": 3.1085, "step": 46950 }, { "epoch": 12.648148148148149, "grad_norm": 0.35205259919166565, "learning_rate": 0.0004105601184866029, "loss": 3.1139, "step": 47000 }, { "epoch": 12.648148148148149, "eval_accuracy": 0.39267146538098324, "eval_loss": 3.2952685356140137, "eval_runtime": 84.2558, "eval_samples_per_second": 213.778, "eval_steps_per_second": 13.364, "step": 47000 }, { "epoch": 12.661606373815676, "grad_norm": 0.36217617988586426, "learning_rate": 0.00041035815268614513, "loss": 3.1118, "step": 47050 }, { "epoch": 12.675064599483203, "grad_norm": 0.37881895899772644, "learning_rate": 0.0004101561868856873, "loss": 3.1218, "step": 47100 }, { "epoch": 12.688522825150732, "grad_norm": 0.3426401913166046, "learning_rate": 0.0004099542210852295, "loss": 3.1114, "step": 47150 }, { "epoch": 12.70198105081826, "grad_norm": 0.34635311365127563, "learning_rate": 0.0004097522552847717, "loss": 3.1118, "step": 47200 }, { "epoch": 12.715439276485789, "grad_norm": 0.34759101271629333, "learning_rate": 0.00040955028948431397, "loss": 3.1153, "step": 47250 }, { "epoch": 12.728897502153316, "grad_norm": 0.39212140440940857, "learning_rate": 0.00040934832368385616, "loss": 3.1197, "step": 47300 }, { "epoch": 12.742355727820843, "grad_norm": 0.34549346566200256, "learning_rate": 0.00040914635788339836, "loss": 3.125, "step": 47350 }, { "epoch": 12.755813953488373, "grad_norm": 0.33954575657844543, "learning_rate": 0.00040894439208294056, "loss": 3.1183, "step": 47400 }, { "epoch": 12.7692721791559, "grad_norm": 0.34983354806900024, "learning_rate": 0.00040874242628248275, "loss": 3.1252, "step": 47450 }, { "epoch": 12.782730404823429, "grad_norm": 0.3618473410606384, "learning_rate": 0.000408540460482025, "loss": 3.1039, "step": 47500 }, { "epoch": 12.796188630490956, "grad_norm": 0.3492349088191986, "learning_rate": 0.00040833849468156725, "loss": 3.1332, "step": 47550 }, { "epoch": 12.809646856158484, "grad_norm": 0.3821626603603363, "learning_rate": 0.00040813652888110945, "loss": 3.1107, "step": 47600 }, { "epoch": 12.823105081826013, "grad_norm": 0.3705442547798157, "learning_rate": 0.00040793456308065165, "loss": 3.1254, "step": 47650 }, { "epoch": 12.83656330749354, "grad_norm": 0.3588872253894806, "learning_rate": 0.0004077325972801939, "loss": 3.1226, "step": 47700 }, { "epoch": 12.850021533161069, "grad_norm": 0.37265342473983765, "learning_rate": 0.0004075306314797361, "loss": 3.1101, "step": 47750 }, { "epoch": 12.863479758828596, "grad_norm": 0.3652234673500061, "learning_rate": 0.0004073286656792783, "loss": 3.12, "step": 47800 }, { "epoch": 12.876937984496124, "grad_norm": 0.3789381980895996, "learning_rate": 0.0004071266998788205, "loss": 3.1207, "step": 47850 }, { "epoch": 12.890396210163653, "grad_norm": 0.34515896439552307, "learning_rate": 0.0004069247340783627, "loss": 3.1286, "step": 47900 }, { "epoch": 12.90385443583118, "grad_norm": 0.37213557958602905, "learning_rate": 0.00040672276827790493, "loss": 3.1279, "step": 47950 }, { "epoch": 12.917312661498707, "grad_norm": 0.3628597557544708, "learning_rate": 0.0004065208024774471, "loss": 3.1098, "step": 48000 }, { "epoch": 12.917312661498707, "eval_accuracy": 0.39299349466087624, "eval_loss": 3.292800188064575, "eval_runtime": 84.4074, "eval_samples_per_second": 213.394, "eval_steps_per_second": 13.34, "step": 48000 }, { "epoch": 12.930770887166236, "grad_norm": 0.35086748003959656, "learning_rate": 0.0004063188366769893, "loss": 3.1178, "step": 48050 }, { "epoch": 12.944229112833764, "grad_norm": 0.40841004252433777, "learning_rate": 0.0004061168708765315, "loss": 3.1226, "step": 48100 }, { "epoch": 12.957687338501293, "grad_norm": 0.3559737205505371, "learning_rate": 0.0004059149050760737, "loss": 3.1358, "step": 48150 }, { "epoch": 12.97114556416882, "grad_norm": 0.3919322192668915, "learning_rate": 0.00040571293927561597, "loss": 3.1335, "step": 48200 }, { "epoch": 12.984603789836347, "grad_norm": 0.35989293456077576, "learning_rate": 0.00040551097347515816, "loss": 3.1246, "step": 48250 }, { "epoch": 12.998062015503876, "grad_norm": 0.3764020800590515, "learning_rate": 0.00040530900767470036, "loss": 3.1221, "step": 48300 }, { "epoch": 13.011304909560723, "grad_norm": 0.3399815559387207, "learning_rate": 0.00040510704187424255, "loss": 3.0507, "step": 48350 }, { "epoch": 13.024763135228252, "grad_norm": 0.3811633884906769, "learning_rate": 0.00040490507607378475, "loss": 3.0216, "step": 48400 }, { "epoch": 13.038221360895779, "grad_norm": 0.35439440608024597, "learning_rate": 0.00040470311027332705, "loss": 3.0306, "step": 48450 }, { "epoch": 13.051679586563308, "grad_norm": 0.3569784164428711, "learning_rate": 0.00040450114447286925, "loss": 3.0333, "step": 48500 }, { "epoch": 13.065137812230835, "grad_norm": 0.37488240003585815, "learning_rate": 0.00040429917867241145, "loss": 3.0394, "step": 48550 }, { "epoch": 13.078596037898363, "grad_norm": 0.37975722551345825, "learning_rate": 0.00040409721287195364, "loss": 3.0402, "step": 48600 }, { "epoch": 13.092054263565892, "grad_norm": 0.37718072533607483, "learning_rate": 0.0004038952470714959, "loss": 3.0411, "step": 48650 }, { "epoch": 13.10551248923342, "grad_norm": 0.35373955965042114, "learning_rate": 0.0004036932812710381, "loss": 3.0521, "step": 48700 }, { "epoch": 13.118970714900948, "grad_norm": 0.39553093910217285, "learning_rate": 0.0004034913154705803, "loss": 3.045, "step": 48750 }, { "epoch": 13.132428940568476, "grad_norm": 0.3530154526233673, "learning_rate": 0.0004032893496701225, "loss": 3.0471, "step": 48800 }, { "epoch": 13.145887166236003, "grad_norm": 0.39944958686828613, "learning_rate": 0.0004030873838696647, "loss": 3.0552, "step": 48850 }, { "epoch": 13.159345391903532, "grad_norm": 0.3472399115562439, "learning_rate": 0.00040288541806920693, "loss": 3.0566, "step": 48900 }, { "epoch": 13.17280361757106, "grad_norm": 0.3732159733772278, "learning_rate": 0.0004026834522687491, "loss": 3.0581, "step": 48950 }, { "epoch": 13.186261843238588, "grad_norm": 0.3669090270996094, "learning_rate": 0.0004024814864682913, "loss": 3.0614, "step": 49000 }, { "epoch": 13.186261843238588, "eval_accuracy": 0.39207771031532357, "eval_loss": 3.3035924434661865, "eval_runtime": 84.3763, "eval_samples_per_second": 213.472, "eval_steps_per_second": 13.345, "step": 49000 }, { "epoch": 13.199720068906116, "grad_norm": 0.3668637275695801, "learning_rate": 0.0004022795206678335, "loss": 3.0574, "step": 49050 }, { "epoch": 13.213178294573643, "grad_norm": 0.3803250193595886, "learning_rate": 0.00040207755486737577, "loss": 3.0672, "step": 49100 }, { "epoch": 13.226636520241172, "grad_norm": 0.4007713794708252, "learning_rate": 0.00040187558906691796, "loss": 3.0597, "step": 49150 }, { "epoch": 13.2400947459087, "grad_norm": 0.37357866764068604, "learning_rate": 0.00040167362326646016, "loss": 3.0649, "step": 49200 }, { "epoch": 13.253552971576227, "grad_norm": 0.3848036825656891, "learning_rate": 0.00040147165746600235, "loss": 3.0663, "step": 49250 }, { "epoch": 13.267011197243756, "grad_norm": 0.3878563344478607, "learning_rate": 0.00040126969166554455, "loss": 3.068, "step": 49300 }, { "epoch": 13.280469422911283, "grad_norm": 0.3646444082260132, "learning_rate": 0.00040106772586508686, "loss": 3.0814, "step": 49350 }, { "epoch": 13.293927648578812, "grad_norm": 0.37095150351524353, "learning_rate": 0.00040086576006462905, "loss": 3.0718, "step": 49400 }, { "epoch": 13.30738587424634, "grad_norm": 0.38114747405052185, "learning_rate": 0.00040066379426417125, "loss": 3.0783, "step": 49450 }, { "epoch": 13.320844099913867, "grad_norm": 0.38861939311027527, "learning_rate": 0.00040046182846371344, "loss": 3.0669, "step": 49500 }, { "epoch": 13.334302325581396, "grad_norm": 0.3983568847179413, "learning_rate": 0.0004002598626632557, "loss": 3.0697, "step": 49550 }, { "epoch": 13.347760551248923, "grad_norm": 0.3889234662055969, "learning_rate": 0.0004000578968627979, "loss": 3.0612, "step": 49600 }, { "epoch": 13.361218776916452, "grad_norm": 0.35765254497528076, "learning_rate": 0.0003998559310623401, "loss": 3.0659, "step": 49650 }, { "epoch": 13.37467700258398, "grad_norm": 0.39406129717826843, "learning_rate": 0.0003996539652618823, "loss": 3.077, "step": 49700 }, { "epoch": 13.388135228251507, "grad_norm": 0.40474364161491394, "learning_rate": 0.0003994519994614245, "loss": 3.0741, "step": 49750 }, { "epoch": 13.401593453919036, "grad_norm": 0.38321611285209656, "learning_rate": 0.00039925003366096673, "loss": 3.0795, "step": 49800 }, { "epoch": 13.415051679586563, "grad_norm": 0.37154221534729004, "learning_rate": 0.0003990480678605089, "loss": 3.0756, "step": 49850 }, { "epoch": 13.428509905254092, "grad_norm": 0.38573193550109863, "learning_rate": 0.0003988461020600511, "loss": 3.074, "step": 49900 }, { "epoch": 13.44196813092162, "grad_norm": 0.3848041892051697, "learning_rate": 0.0003986441362595933, "loss": 3.0801, "step": 49950 }, { "epoch": 13.455426356589147, "grad_norm": 0.35528790950775146, "learning_rate": 0.0003984421704591355, "loss": 3.0893, "step": 50000 }, { "epoch": 13.455426356589147, "eval_accuracy": 0.39282367962562903, "eval_loss": 3.2962677478790283, "eval_runtime": 84.322, "eval_samples_per_second": 213.61, "eval_steps_per_second": 13.354, "step": 50000 }, { "epoch": 13.468884582256676, "grad_norm": 0.3837132453918457, "learning_rate": 0.00039824020465867776, "loss": 3.0822, "step": 50050 }, { "epoch": 13.482342807924203, "grad_norm": 0.3933063745498657, "learning_rate": 0.00039803823885821996, "loss": 3.0871, "step": 50100 }, { "epoch": 13.49580103359173, "grad_norm": 0.3604951798915863, "learning_rate": 0.00039783627305776216, "loss": 3.0849, "step": 50150 }, { "epoch": 13.50925925925926, "grad_norm": 0.3549090027809143, "learning_rate": 0.00039763430725730435, "loss": 3.0842, "step": 50200 }, { "epoch": 13.522717484926787, "grad_norm": 0.3664722442626953, "learning_rate": 0.00039743234145684666, "loss": 3.099, "step": 50250 }, { "epoch": 13.536175710594316, "grad_norm": 0.35317978262901306, "learning_rate": 0.00039723037565638885, "loss": 3.0917, "step": 50300 }, { "epoch": 13.549633936261843, "grad_norm": 0.3661048412322998, "learning_rate": 0.00039702840985593105, "loss": 3.0861, "step": 50350 }, { "epoch": 13.56309216192937, "grad_norm": 0.38903483748435974, "learning_rate": 0.00039682644405547324, "loss": 3.0975, "step": 50400 }, { "epoch": 13.5765503875969, "grad_norm": 0.3472830057144165, "learning_rate": 0.00039662447825501544, "loss": 3.0983, "step": 50450 }, { "epoch": 13.590008613264427, "grad_norm": 0.40977710485458374, "learning_rate": 0.0003964225124545577, "loss": 3.0923, "step": 50500 }, { "epoch": 13.603466838931956, "grad_norm": 0.37898826599121094, "learning_rate": 0.0003962205466540999, "loss": 3.1029, "step": 50550 }, { "epoch": 13.616925064599483, "grad_norm": 0.36426326632499695, "learning_rate": 0.0003960185808536421, "loss": 3.0972, "step": 50600 }, { "epoch": 13.63038329026701, "grad_norm": 0.3560165464878082, "learning_rate": 0.0003958166150531843, "loss": 3.1104, "step": 50650 }, { "epoch": 13.64384151593454, "grad_norm": 0.38337618112564087, "learning_rate": 0.0003956146492527265, "loss": 3.099, "step": 50700 }, { "epoch": 13.657299741602067, "grad_norm": 0.373165488243103, "learning_rate": 0.0003954126834522687, "loss": 3.0936, "step": 50750 }, { "epoch": 13.670757967269594, "grad_norm": 0.35410720109939575, "learning_rate": 0.0003952107176518109, "loss": 3.1071, "step": 50800 }, { "epoch": 13.684216192937123, "grad_norm": 0.4100600481033325, "learning_rate": 0.0003950087518513531, "loss": 3.0973, "step": 50850 }, { "epoch": 13.69767441860465, "grad_norm": 0.38382619619369507, "learning_rate": 0.0003948067860508953, "loss": 3.1021, "step": 50900 }, { "epoch": 13.71113264427218, "grad_norm": 0.36865368485450745, "learning_rate": 0.0003946048202504375, "loss": 3.0984, "step": 50950 }, { "epoch": 13.724590869939707, "grad_norm": 0.35743772983551025, "learning_rate": 0.00039440285444997976, "loss": 3.0893, "step": 51000 }, { "epoch": 13.724590869939707, "eval_accuracy": 0.39332182545839195, "eval_loss": 3.289945363998413, "eval_runtime": 84.2626, "eval_samples_per_second": 213.76, "eval_steps_per_second": 13.363, "step": 51000 }, { "epoch": 13.738049095607234, "grad_norm": 0.3683944344520569, "learning_rate": 0.00039420088864952196, "loss": 3.1162, "step": 51050 }, { "epoch": 13.751507321274763, "grad_norm": 0.36566293239593506, "learning_rate": 0.00039399892284906415, "loss": 3.0974, "step": 51100 }, { "epoch": 13.76496554694229, "grad_norm": 0.34378859400749207, "learning_rate": 0.00039379695704860646, "loss": 3.1029, "step": 51150 }, { "epoch": 13.77842377260982, "grad_norm": 0.409297913312912, "learning_rate": 0.00039359499124814865, "loss": 3.1031, "step": 51200 }, { "epoch": 13.791881998277347, "grad_norm": 0.3918900191783905, "learning_rate": 0.00039339302544769085, "loss": 3.1081, "step": 51250 }, { "epoch": 13.805340223944874, "grad_norm": 0.3867114782333374, "learning_rate": 0.00039319105964723305, "loss": 3.1026, "step": 51300 }, { "epoch": 13.818798449612403, "grad_norm": 0.38156697154045105, "learning_rate": 0.00039298909384677524, "loss": 3.1067, "step": 51350 }, { "epoch": 13.83225667527993, "grad_norm": 0.381967693567276, "learning_rate": 0.0003927871280463175, "loss": 3.1001, "step": 51400 }, { "epoch": 13.84571490094746, "grad_norm": 0.37300676107406616, "learning_rate": 0.0003925851622458597, "loss": 3.0924, "step": 51450 }, { "epoch": 13.859173126614987, "grad_norm": 0.36449840664863586, "learning_rate": 0.0003923831964454019, "loss": 3.1068, "step": 51500 }, { "epoch": 13.872631352282514, "grad_norm": 0.37150195240974426, "learning_rate": 0.0003921812306449441, "loss": 3.0981, "step": 51550 }, { "epoch": 13.886089577950044, "grad_norm": 0.36262959241867065, "learning_rate": 0.0003919792648444863, "loss": 3.0958, "step": 51600 }, { "epoch": 13.89954780361757, "grad_norm": 0.37757211923599243, "learning_rate": 0.0003917772990440285, "loss": 3.1027, "step": 51650 }, { "epoch": 13.9130060292851, "grad_norm": 0.3373667597770691, "learning_rate": 0.0003915753332435707, "loss": 3.1131, "step": 51700 }, { "epoch": 13.926464254952627, "grad_norm": 0.35641902685165405, "learning_rate": 0.0003913733674431129, "loss": 3.1163, "step": 51750 }, { "epoch": 13.939922480620154, "grad_norm": 0.35720914602279663, "learning_rate": 0.0003911714016426551, "loss": 3.0914, "step": 51800 }, { "epoch": 13.953380706287684, "grad_norm": 0.37049901485443115, "learning_rate": 0.0003909694358421973, "loss": 3.1012, "step": 51850 }, { "epoch": 13.96683893195521, "grad_norm": 0.3717890679836273, "learning_rate": 0.00039076747004173956, "loss": 3.1198, "step": 51900 }, { "epoch": 13.98029715762274, "grad_norm": 0.3679327070713043, "learning_rate": 0.00039056550424128176, "loss": 3.1053, "step": 51950 }, { "epoch": 13.993755383290267, "grad_norm": 0.36129581928253174, "learning_rate": 0.00039036353844082395, "loss": 3.108, "step": 52000 }, { "epoch": 13.993755383290267, "eval_accuracy": 0.3937257744673805, "eval_loss": 3.2866153717041016, "eval_runtime": 84.429, "eval_samples_per_second": 213.339, "eval_steps_per_second": 13.337, "step": 52000 }, { "epoch": 14.006998277347115, "grad_norm": 0.38459154963493347, "learning_rate": 0.0003901615726403662, "loss": 3.054, "step": 52050 }, { "epoch": 14.020456503014643, "grad_norm": 0.3999425768852234, "learning_rate": 0.00038995960683990845, "loss": 3.007, "step": 52100 }, { "epoch": 14.03391472868217, "grad_norm": 0.35180553793907166, "learning_rate": 0.00038975764103945065, "loss": 3.0228, "step": 52150 }, { "epoch": 14.047372954349699, "grad_norm": 0.407040536403656, "learning_rate": 0.00038955567523899285, "loss": 3.0114, "step": 52200 }, { "epoch": 14.060831180017226, "grad_norm": 0.36089593172073364, "learning_rate": 0.00038935370943853504, "loss": 3.0107, "step": 52250 }, { "epoch": 14.074289405684755, "grad_norm": 0.36350467801094055, "learning_rate": 0.00038915174363807724, "loss": 3.0155, "step": 52300 }, { "epoch": 14.087747631352283, "grad_norm": 0.36970725655555725, "learning_rate": 0.0003889497778376195, "loss": 3.0151, "step": 52350 }, { "epoch": 14.10120585701981, "grad_norm": 0.38831764459609985, "learning_rate": 0.0003887478120371617, "loss": 3.0335, "step": 52400 }, { "epoch": 14.114664082687339, "grad_norm": 0.38094422221183777, "learning_rate": 0.0003885458462367039, "loss": 3.031, "step": 52450 }, { "epoch": 14.128122308354866, "grad_norm": 0.3756919205188751, "learning_rate": 0.0003883438804362461, "loss": 3.0327, "step": 52500 }, { "epoch": 14.141580534022394, "grad_norm": 0.372146874666214, "learning_rate": 0.0003881419146357883, "loss": 3.0398, "step": 52550 }, { "epoch": 14.155038759689923, "grad_norm": 0.36529073119163513, "learning_rate": 0.0003879399488353305, "loss": 3.0487, "step": 52600 }, { "epoch": 14.16849698535745, "grad_norm": 0.3786604404449463, "learning_rate": 0.0003877379830348727, "loss": 3.0371, "step": 52650 }, { "epoch": 14.18195521102498, "grad_norm": 0.4139573276042938, "learning_rate": 0.0003875360172344149, "loss": 3.0559, "step": 52700 }, { "epoch": 14.195413436692506, "grad_norm": 0.35359033942222595, "learning_rate": 0.0003873340514339571, "loss": 3.0373, "step": 52750 }, { "epoch": 14.208871662360034, "grad_norm": 0.37699979543685913, "learning_rate": 0.0003871320856334993, "loss": 3.0482, "step": 52800 }, { "epoch": 14.222329888027563, "grad_norm": 0.3546140193939209, "learning_rate": 0.00038693011983304156, "loss": 3.0478, "step": 52850 }, { "epoch": 14.23578811369509, "grad_norm": 0.3665039837360382, "learning_rate": 0.0003867281540325838, "loss": 3.0516, "step": 52900 }, { "epoch": 14.24924633936262, "grad_norm": 0.3570476174354553, "learning_rate": 0.000386526188232126, "loss": 3.0523, "step": 52950 }, { "epoch": 14.262704565030146, "grad_norm": 0.40596699714660645, "learning_rate": 0.00038632422243166825, "loss": 3.0564, "step": 53000 }, { "epoch": 14.262704565030146, "eval_accuracy": 0.39316233187442334, "eval_loss": 3.299149513244629, "eval_runtime": 84.3356, "eval_samples_per_second": 213.575, "eval_steps_per_second": 13.351, "step": 53000 }, { "epoch": 14.276162790697674, "grad_norm": 0.36422741413116455, "learning_rate": 0.00038612225663121045, "loss": 3.0472, "step": 53050 }, { "epoch": 14.289621016365203, "grad_norm": 0.36780548095703125, "learning_rate": 0.00038592029083075265, "loss": 3.0387, "step": 53100 }, { "epoch": 14.30307924203273, "grad_norm": 0.3683931827545166, "learning_rate": 0.00038571832503029484, "loss": 3.0483, "step": 53150 }, { "epoch": 14.31653746770026, "grad_norm": 0.38417869806289673, "learning_rate": 0.00038551635922983704, "loss": 3.0559, "step": 53200 }, { "epoch": 14.329995693367787, "grad_norm": 0.39943137764930725, "learning_rate": 0.0003853143934293793, "loss": 3.0479, "step": 53250 }, { "epoch": 14.343453919035314, "grad_norm": 0.4140488803386688, "learning_rate": 0.0003851124276289215, "loss": 3.0612, "step": 53300 }, { "epoch": 14.356912144702843, "grad_norm": 0.3700057864189148, "learning_rate": 0.0003849104618284637, "loss": 3.0584, "step": 53350 }, { "epoch": 14.37037037037037, "grad_norm": 0.36353835463523865, "learning_rate": 0.0003847084960280059, "loss": 3.0683, "step": 53400 }, { "epoch": 14.383828596037898, "grad_norm": 0.3969934284687042, "learning_rate": 0.0003845065302275481, "loss": 3.0688, "step": 53450 }, { "epoch": 14.397286821705427, "grad_norm": 0.38807758688926697, "learning_rate": 0.0003843045644270903, "loss": 3.0707, "step": 53500 }, { "epoch": 14.410745047372954, "grad_norm": 0.36600351333618164, "learning_rate": 0.0003841025986266325, "loss": 3.08, "step": 53550 }, { "epoch": 14.424203273040483, "grad_norm": 0.36152181029319763, "learning_rate": 0.0003839006328261747, "loss": 3.0594, "step": 53600 }, { "epoch": 14.43766149870801, "grad_norm": 0.38738057017326355, "learning_rate": 0.0003836986670257169, "loss": 3.0623, "step": 53650 }, { "epoch": 14.451119724375538, "grad_norm": 0.3771466314792633, "learning_rate": 0.0003834967012252591, "loss": 3.0713, "step": 53700 }, { "epoch": 14.464577950043067, "grad_norm": 0.3636995851993561, "learning_rate": 0.00038329473542480136, "loss": 3.0727, "step": 53750 }, { "epoch": 14.478036175710594, "grad_norm": 0.3603183329105377, "learning_rate": 0.0003830927696243436, "loss": 3.0736, "step": 53800 }, { "epoch": 14.491494401378123, "grad_norm": 0.4068652093410492, "learning_rate": 0.0003828908038238858, "loss": 3.0775, "step": 53850 }, { "epoch": 14.50495262704565, "grad_norm": 0.3683650493621826, "learning_rate": 0.000382688838023428, "loss": 3.0743, "step": 53900 }, { "epoch": 14.518410852713178, "grad_norm": 0.34459036588668823, "learning_rate": 0.00038248687222297025, "loss": 3.0744, "step": 53950 }, { "epoch": 14.531869078380707, "grad_norm": 0.37111741304397583, "learning_rate": 0.00038228490642251245, "loss": 3.0737, "step": 54000 }, { "epoch": 14.531869078380707, "eval_accuracy": 0.39345415732846944, "eval_loss": 3.292693853378296, "eval_runtime": 84.4283, "eval_samples_per_second": 213.341, "eval_steps_per_second": 13.337, "step": 54000 }, { "epoch": 14.545327304048234, "grad_norm": 0.3831128478050232, "learning_rate": 0.00038208294062205464, "loss": 3.0687, "step": 54050 }, { "epoch": 14.558785529715763, "grad_norm": 0.3837164044380188, "learning_rate": 0.00038188097482159684, "loss": 3.0722, "step": 54100 }, { "epoch": 14.57224375538329, "grad_norm": 0.362723171710968, "learning_rate": 0.00038167900902113904, "loss": 3.0852, "step": 54150 }, { "epoch": 14.585701981050818, "grad_norm": 0.37301892042160034, "learning_rate": 0.0003814770432206813, "loss": 3.0733, "step": 54200 }, { "epoch": 14.599160206718347, "grad_norm": 0.3915317952632904, "learning_rate": 0.0003812750774202235, "loss": 3.0799, "step": 54250 }, { "epoch": 14.612618432385874, "grad_norm": 0.3872342109680176, "learning_rate": 0.0003810731116197657, "loss": 3.0755, "step": 54300 }, { "epoch": 14.626076658053403, "grad_norm": 0.367057740688324, "learning_rate": 0.0003808711458193079, "loss": 3.0825, "step": 54350 }, { "epoch": 14.63953488372093, "grad_norm": 0.42809247970581055, "learning_rate": 0.00038066918001885007, "loss": 3.0796, "step": 54400 }, { "epoch": 14.652993109388458, "grad_norm": 0.39945393800735474, "learning_rate": 0.0003804672142183923, "loss": 3.0829, "step": 54450 }, { "epoch": 14.666451335055987, "grad_norm": 0.3408666253089905, "learning_rate": 0.0003802652484179345, "loss": 3.0862, "step": 54500 }, { "epoch": 14.679909560723514, "grad_norm": 0.3764728307723999, "learning_rate": 0.0003800632826174767, "loss": 3.0856, "step": 54550 }, { "epoch": 14.693367786391041, "grad_norm": 0.3803330957889557, "learning_rate": 0.0003798613168170189, "loss": 3.0751, "step": 54600 }, { "epoch": 14.70682601205857, "grad_norm": 0.39075732231140137, "learning_rate": 0.0003796593510165611, "loss": 3.0867, "step": 54650 }, { "epoch": 14.720284237726098, "grad_norm": 0.3951171636581421, "learning_rate": 0.0003794573852161034, "loss": 3.0824, "step": 54700 }, { "epoch": 14.733742463393627, "grad_norm": 0.37119993567466736, "learning_rate": 0.0003792554194156456, "loss": 3.0845, "step": 54750 }, { "epoch": 14.747200689061154, "grad_norm": 0.37848755717277527, "learning_rate": 0.0003790534536151878, "loss": 3.0861, "step": 54800 }, { "epoch": 14.760658914728682, "grad_norm": 0.4070167541503906, "learning_rate": 0.00037885148781473005, "loss": 3.0856, "step": 54850 }, { "epoch": 14.77411714039621, "grad_norm": 0.3541718125343323, "learning_rate": 0.00037864952201427225, "loss": 3.0917, "step": 54900 }, { "epoch": 14.787575366063738, "grad_norm": 0.38448378443717957, "learning_rate": 0.00037844755621381444, "loss": 3.0865, "step": 54950 }, { "epoch": 14.801033591731267, "grad_norm": 0.36622485518455505, "learning_rate": 0.00037824559041335664, "loss": 3.0934, "step": 55000 }, { "epoch": 14.801033591731267, "eval_accuracy": 0.39399760890000274, "eval_loss": 3.2891037464141846, "eval_runtime": 84.0862, "eval_samples_per_second": 214.209, "eval_steps_per_second": 13.391, "step": 55000 }, { "epoch": 14.814491817398794, "grad_norm": 0.35305047035217285, "learning_rate": 0.00037804362461289884, "loss": 3.0932, "step": 55050 }, { "epoch": 14.827950043066322, "grad_norm": 0.3791576325893402, "learning_rate": 0.0003778416588124411, "loss": 3.0863, "step": 55100 }, { "epoch": 14.84140826873385, "grad_norm": 0.36797282099723816, "learning_rate": 0.0003776396930119833, "loss": 3.0889, "step": 55150 }, { "epoch": 14.854866494401378, "grad_norm": 0.3456723988056183, "learning_rate": 0.0003774377272115255, "loss": 3.0958, "step": 55200 }, { "epoch": 14.868324720068905, "grad_norm": 0.38494226336479187, "learning_rate": 0.0003772357614110677, "loss": 3.0857, "step": 55250 }, { "epoch": 14.881782945736434, "grad_norm": 0.35462987422943115, "learning_rate": 0.00037703379561060987, "loss": 3.0942, "step": 55300 }, { "epoch": 14.895241171403962, "grad_norm": 0.36904287338256836, "learning_rate": 0.0003768318298101521, "loss": 3.1013, "step": 55350 }, { "epoch": 14.90869939707149, "grad_norm": 0.36798912286758423, "learning_rate": 0.0003766298640096943, "loss": 3.093, "step": 55400 }, { "epoch": 14.922157622739018, "grad_norm": 0.3697493076324463, "learning_rate": 0.0003764278982092365, "loss": 3.0977, "step": 55450 }, { "epoch": 14.935615848406545, "grad_norm": 0.39716044068336487, "learning_rate": 0.0003762259324087787, "loss": 3.1009, "step": 55500 }, { "epoch": 14.949074074074074, "grad_norm": 0.39862293004989624, "learning_rate": 0.0003760239666083209, "loss": 3.0913, "step": 55550 }, { "epoch": 14.962532299741602, "grad_norm": 0.426544725894928, "learning_rate": 0.0003758220008078632, "loss": 3.089, "step": 55600 }, { "epoch": 14.97599052540913, "grad_norm": 0.3789792060852051, "learning_rate": 0.0003756200350074054, "loss": 3.0964, "step": 55650 }, { "epoch": 14.989448751076658, "grad_norm": 0.3739278316497803, "learning_rate": 0.0003754180692069476, "loss": 3.0979, "step": 55700 }, { "epoch": 15.002691645133506, "grad_norm": 0.4127860367298126, "learning_rate": 0.0003752161034064898, "loss": 3.0805, "step": 55750 }, { "epoch": 15.016149870801033, "grad_norm": 0.3938711881637573, "learning_rate": 0.00037501413760603205, "loss": 2.9973, "step": 55800 }, { "epoch": 15.02960809646856, "grad_norm": 0.3406141400337219, "learning_rate": 0.00037481217180557425, "loss": 2.9984, "step": 55850 }, { "epoch": 15.04306632213609, "grad_norm": 0.3878343105316162, "learning_rate": 0.00037461020600511644, "loss": 2.9946, "step": 55900 }, { "epoch": 15.056524547803617, "grad_norm": 0.42925557494163513, "learning_rate": 0.00037440824020465864, "loss": 3.0021, "step": 55950 }, { "epoch": 15.069982773471146, "grad_norm": 0.36447229981422424, "learning_rate": 0.00037420627440420083, "loss": 3.0086, "step": 56000 }, { "epoch": 15.069982773471146, "eval_accuracy": 0.39346197990207005, "eval_loss": 3.2978291511535645, "eval_runtime": 84.1272, "eval_samples_per_second": 214.104, "eval_steps_per_second": 13.384, "step": 56000 }, { "epoch": 15.083440999138674, "grad_norm": 0.4149514436721802, "learning_rate": 0.0003740043086037431, "loss": 3.0051, "step": 56050 }, { "epoch": 15.0968992248062, "grad_norm": 0.4040587246417999, "learning_rate": 0.0003738023428032853, "loss": 3.0164, "step": 56100 }, { "epoch": 15.11035745047373, "grad_norm": 0.41578409075737, "learning_rate": 0.0003736003770028275, "loss": 3.0111, "step": 56150 }, { "epoch": 15.123815676141257, "grad_norm": 0.4122296869754791, "learning_rate": 0.00037339841120236967, "loss": 3.0214, "step": 56200 }, { "epoch": 15.137273901808786, "grad_norm": 0.4029271900653839, "learning_rate": 0.00037319644540191187, "loss": 3.0259, "step": 56250 }, { "epoch": 15.150732127476314, "grad_norm": 0.3622616231441498, "learning_rate": 0.0003729944796014541, "loss": 3.0213, "step": 56300 }, { "epoch": 15.164190353143841, "grad_norm": 0.39520934224128723, "learning_rate": 0.0003727925138009963, "loss": 3.034, "step": 56350 }, { "epoch": 15.17764857881137, "grad_norm": 0.3711996376514435, "learning_rate": 0.0003725905480005385, "loss": 3.029, "step": 56400 }, { "epoch": 15.191106804478897, "grad_norm": 0.4158594608306885, "learning_rate": 0.0003723885822000807, "loss": 3.033, "step": 56450 }, { "epoch": 15.204565030146426, "grad_norm": 0.4163890480995178, "learning_rate": 0.000372186616399623, "loss": 3.023, "step": 56500 }, { "epoch": 15.218023255813954, "grad_norm": 0.4126832187175751, "learning_rate": 0.0003719846505991652, "loss": 3.0363, "step": 56550 }, { "epoch": 15.231481481481481, "grad_norm": 0.3895432651042938, "learning_rate": 0.0003717826847987074, "loss": 3.0311, "step": 56600 }, { "epoch": 15.24493970714901, "grad_norm": 0.3966125547885895, "learning_rate": 0.0003715807189982496, "loss": 3.0431, "step": 56650 }, { "epoch": 15.258397932816537, "grad_norm": 0.3741343021392822, "learning_rate": 0.00037137875319779185, "loss": 3.0379, "step": 56700 }, { "epoch": 15.271856158484065, "grad_norm": 0.37849706411361694, "learning_rate": 0.00037117678739733405, "loss": 3.0511, "step": 56750 }, { "epoch": 15.285314384151594, "grad_norm": 0.38693898916244507, "learning_rate": 0.00037097482159687624, "loss": 3.0524, "step": 56800 }, { "epoch": 15.298772609819121, "grad_norm": 0.3879419267177582, "learning_rate": 0.00037077285579641844, "loss": 3.0444, "step": 56850 }, { "epoch": 15.31223083548665, "grad_norm": 0.38886481523513794, "learning_rate": 0.00037057088999596063, "loss": 3.0338, "step": 56900 }, { "epoch": 15.325689061154177, "grad_norm": 0.36853650212287903, "learning_rate": 0.0003703689241955029, "loss": 3.0466, "step": 56950 }, { "epoch": 15.339147286821705, "grad_norm": 0.383457750082016, "learning_rate": 0.0003701669583950451, "loss": 3.047, "step": 57000 }, { "epoch": 15.339147286821705, "eval_accuracy": 0.39383724614118965, "eval_loss": 3.2929184436798096, "eval_runtime": 84.2703, "eval_samples_per_second": 213.741, "eval_steps_per_second": 13.362, "step": 57000 }, { "epoch": 15.352605512489234, "grad_norm": 0.3808010220527649, "learning_rate": 0.0003699649925945873, "loss": 3.0444, "step": 57050 }, { "epoch": 15.366063738156761, "grad_norm": 0.4095914363861084, "learning_rate": 0.0003697630267941295, "loss": 3.0514, "step": 57100 }, { "epoch": 15.37952196382429, "grad_norm": 0.3669900894165039, "learning_rate": 0.00036956106099367167, "loss": 3.0462, "step": 57150 }, { "epoch": 15.392980189491817, "grad_norm": 0.3716629147529602, "learning_rate": 0.0003693590951932139, "loss": 3.0458, "step": 57200 }, { "epoch": 15.406438415159345, "grad_norm": 0.38203051686286926, "learning_rate": 0.0003691571293927561, "loss": 3.0479, "step": 57250 }, { "epoch": 15.419896640826874, "grad_norm": 0.37157347798347473, "learning_rate": 0.0003689551635922983, "loss": 3.0497, "step": 57300 }, { "epoch": 15.433354866494401, "grad_norm": 0.36887258291244507, "learning_rate": 0.00036875319779184056, "loss": 3.0448, "step": 57350 }, { "epoch": 15.44681309216193, "grad_norm": 0.3627953827381134, "learning_rate": 0.0003685512319913828, "loss": 3.0498, "step": 57400 }, { "epoch": 15.460271317829458, "grad_norm": 0.3988632559776306, "learning_rate": 0.000368349266190925, "loss": 3.0679, "step": 57450 }, { "epoch": 15.473729543496985, "grad_norm": 0.41351136565208435, "learning_rate": 0.0003681473003904672, "loss": 3.045, "step": 57500 }, { "epoch": 15.487187769164514, "grad_norm": 0.3687242865562439, "learning_rate": 0.0003679453345900094, "loss": 3.0479, "step": 57550 }, { "epoch": 15.500645994832041, "grad_norm": 0.3737011253833771, "learning_rate": 0.0003677433687895516, "loss": 3.0599, "step": 57600 }, { "epoch": 15.514104220499568, "grad_norm": 0.4048716127872467, "learning_rate": 0.00036754140298909385, "loss": 3.0586, "step": 57650 }, { "epoch": 15.527562446167098, "grad_norm": 0.36465123295783997, "learning_rate": 0.00036733943718863604, "loss": 3.0744, "step": 57700 }, { "epoch": 15.541020671834625, "grad_norm": 0.3666316866874695, "learning_rate": 0.00036713747138817824, "loss": 3.0563, "step": 57750 }, { "epoch": 15.554478897502154, "grad_norm": 0.37599480152130127, "learning_rate": 0.00036693550558772044, "loss": 3.0602, "step": 57800 }, { "epoch": 15.567937123169681, "grad_norm": 0.38387683033943176, "learning_rate": 0.00036673353978726263, "loss": 3.0745, "step": 57850 }, { "epoch": 15.581395348837209, "grad_norm": 0.3832012116909027, "learning_rate": 0.0003665315739868049, "loss": 3.0711, "step": 57900 }, { "epoch": 15.594853574504738, "grad_norm": 0.3958407938480377, "learning_rate": 0.0003663296081863471, "loss": 3.0672, "step": 57950 }, { "epoch": 15.608311800172265, "grad_norm": 0.36599913239479065, "learning_rate": 0.0003661276423858893, "loss": 3.063, "step": 58000 }, { "epoch": 15.608311800172265, "eval_accuracy": 0.39374978542246025, "eval_loss": 3.2895333766937256, "eval_runtime": 84.3811, "eval_samples_per_second": 213.46, "eval_steps_per_second": 13.344, "step": 58000 }, { "epoch": 15.621770025839794, "grad_norm": 0.39657357335090637, "learning_rate": 0.00036592567658543147, "loss": 3.0706, "step": 58050 }, { "epoch": 15.635228251507321, "grad_norm": 0.36927181482315063, "learning_rate": 0.00036572371078497367, "loss": 3.0651, "step": 58100 }, { "epoch": 15.648686477174849, "grad_norm": 0.38070324063301086, "learning_rate": 0.0003655217449845159, "loss": 3.0712, "step": 58150 }, { "epoch": 15.662144702842378, "grad_norm": 0.36729058623313904, "learning_rate": 0.0003653197791840581, "loss": 3.0727, "step": 58200 }, { "epoch": 15.675602928509905, "grad_norm": 0.3709213137626648, "learning_rate": 0.00036511781338360036, "loss": 3.0636, "step": 58250 }, { "epoch": 15.689061154177434, "grad_norm": 0.34084853529930115, "learning_rate": 0.00036491584758314256, "loss": 3.0752, "step": 58300 }, { "epoch": 15.702519379844961, "grad_norm": 0.4009411036968231, "learning_rate": 0.0003647138817826848, "loss": 3.0757, "step": 58350 }, { "epoch": 15.715977605512489, "grad_norm": 0.394245445728302, "learning_rate": 0.000364511915982227, "loss": 3.0708, "step": 58400 }, { "epoch": 15.729435831180018, "grad_norm": 0.3695777356624603, "learning_rate": 0.0003643099501817692, "loss": 3.0635, "step": 58450 }, { "epoch": 15.742894056847545, "grad_norm": 0.3706803023815155, "learning_rate": 0.0003641079843813114, "loss": 3.078, "step": 58500 }, { "epoch": 15.756352282515074, "grad_norm": 0.38254979252815247, "learning_rate": 0.0003639060185808536, "loss": 3.0782, "step": 58550 }, { "epoch": 15.769810508182601, "grad_norm": 0.38933348655700684, "learning_rate": 0.00036370405278039584, "loss": 3.0668, "step": 58600 }, { "epoch": 15.783268733850129, "grad_norm": 0.36985987424850464, "learning_rate": 0.00036350208697993804, "loss": 3.0669, "step": 58650 }, { "epoch": 15.796726959517658, "grad_norm": 0.40725845098495483, "learning_rate": 0.00036330012117948024, "loss": 3.0812, "step": 58700 }, { "epoch": 15.810185185185185, "grad_norm": 0.3564852476119995, "learning_rate": 0.00036309815537902243, "loss": 3.0722, "step": 58750 }, { "epoch": 15.823643410852712, "grad_norm": 0.3941057026386261, "learning_rate": 0.0003628961895785647, "loss": 3.0807, "step": 58800 }, { "epoch": 15.837101636520241, "grad_norm": 0.41038039326667786, "learning_rate": 0.0003626942237781069, "loss": 3.0783, "step": 58850 }, { "epoch": 15.850559862187769, "grad_norm": 0.3650042712688446, "learning_rate": 0.0003624922579776491, "loss": 3.0741, "step": 58900 }, { "epoch": 15.864018087855298, "grad_norm": 0.38325804471969604, "learning_rate": 0.00036229029217719127, "loss": 3.0692, "step": 58950 }, { "epoch": 15.877476313522825, "grad_norm": 0.38196754455566406, "learning_rate": 0.00036208832637673347, "loss": 3.0813, "step": 59000 }, { "epoch": 15.877476313522825, "eval_accuracy": 0.3946187429732646, "eval_loss": 3.2819015979766846, "eval_runtime": 84.302, "eval_samples_per_second": 213.66, "eval_steps_per_second": 13.357, "step": 59000 }, { "epoch": 15.890934539190352, "grad_norm": 0.3796331286430359, "learning_rate": 0.0003618863605762757, "loss": 3.0757, "step": 59050 }, { "epoch": 15.904392764857882, "grad_norm": 0.38216689229011536, "learning_rate": 0.0003616843947758179, "loss": 3.0867, "step": 59100 }, { "epoch": 15.917850990525409, "grad_norm": 0.3886622190475464, "learning_rate": 0.00036148242897536016, "loss": 3.0757, "step": 59150 }, { "epoch": 15.931309216192938, "grad_norm": 0.37623468041419983, "learning_rate": 0.00036128046317490236, "loss": 3.0783, "step": 59200 }, { "epoch": 15.944767441860465, "grad_norm": 0.36371952295303345, "learning_rate": 0.0003610784973744446, "loss": 3.0838, "step": 59250 }, { "epoch": 15.958225667527993, "grad_norm": 0.3721584975719452, "learning_rate": 0.0003608765315739868, "loss": 3.0743, "step": 59300 }, { "epoch": 15.971683893195522, "grad_norm": 0.41483673453330994, "learning_rate": 0.000360674565773529, "loss": 3.088, "step": 59350 }, { "epoch": 15.985142118863049, "grad_norm": 0.39113107323646545, "learning_rate": 0.0003604725999730712, "loss": 3.0794, "step": 59400 }, { "epoch": 15.998600344530576, "grad_norm": 0.3843020796775818, "learning_rate": 0.0003602706341726134, "loss": 3.0755, "step": 59450 }, { "epoch": 16.011843238587424, "grad_norm": 0.3941507935523987, "learning_rate": 0.00036006866837215564, "loss": 2.9901, "step": 59500 }, { "epoch": 16.02530146425495, "grad_norm": 0.4168238043785095, "learning_rate": 0.00035986670257169784, "loss": 2.9901, "step": 59550 }, { "epoch": 16.03875968992248, "grad_norm": 0.4108584225177765, "learning_rate": 0.00035966473677124004, "loss": 2.9998, "step": 59600 }, { "epoch": 16.05221791559001, "grad_norm": 0.3931739032268524, "learning_rate": 0.00035946277097078223, "loss": 2.9857, "step": 59650 }, { "epoch": 16.065676141257537, "grad_norm": 0.41490697860717773, "learning_rate": 0.00035926080517032443, "loss": 2.9893, "step": 59700 }, { "epoch": 16.079134366925064, "grad_norm": 0.3986837565898895, "learning_rate": 0.0003590588393698667, "loss": 3.006, "step": 59750 }, { "epoch": 16.09259259259259, "grad_norm": 0.4169183373451233, "learning_rate": 0.0003588568735694089, "loss": 3.0026, "step": 59800 }, { "epoch": 16.10605081826012, "grad_norm": 0.3991803228855133, "learning_rate": 0.00035865490776895107, "loss": 2.9958, "step": 59850 }, { "epoch": 16.11950904392765, "grad_norm": 0.438391774892807, "learning_rate": 0.00035845294196849327, "loss": 3.0077, "step": 59900 }, { "epoch": 16.132967269595177, "grad_norm": 0.3962257504463196, "learning_rate": 0.00035825097616803546, "loss": 3.0028, "step": 59950 }, { "epoch": 16.146425495262704, "grad_norm": 0.3833325505256653, "learning_rate": 0.0003580490103675777, "loss": 3.0017, "step": 60000 }, { "epoch": 16.146425495262704, "eval_accuracy": 0.3941621002393273, "eval_loss": 3.294508934020996, "eval_runtime": 84.2172, "eval_samples_per_second": 213.876, "eval_steps_per_second": 13.37, "step": 60000 }, { "epoch": 16.15988372093023, "grad_norm": 0.3867778480052948, "learning_rate": 0.00035784704456711996, "loss": 3.0079, "step": 60050 }, { "epoch": 16.17334194659776, "grad_norm": 0.40779435634613037, "learning_rate": 0.00035764507876666216, "loss": 3.0106, "step": 60100 }, { "epoch": 16.18680017226529, "grad_norm": 0.40491026639938354, "learning_rate": 0.00035744311296620436, "loss": 3.0132, "step": 60150 }, { "epoch": 16.200258397932817, "grad_norm": 0.37903210520744324, "learning_rate": 0.0003572411471657466, "loss": 3.0205, "step": 60200 }, { "epoch": 16.213716623600344, "grad_norm": 0.41829851269721985, "learning_rate": 0.0003570391813652888, "loss": 3.0283, "step": 60250 }, { "epoch": 16.227174849267872, "grad_norm": 0.3932286500930786, "learning_rate": 0.000356837215564831, "loss": 3.0208, "step": 60300 }, { "epoch": 16.2406330749354, "grad_norm": 0.4128912389278412, "learning_rate": 0.0003566352497643732, "loss": 3.0246, "step": 60350 }, { "epoch": 16.25409130060293, "grad_norm": 0.3961387872695923, "learning_rate": 0.0003564332839639154, "loss": 3.0255, "step": 60400 }, { "epoch": 16.267549526270457, "grad_norm": 0.40615180134773254, "learning_rate": 0.00035623131816345764, "loss": 3.0274, "step": 60450 }, { "epoch": 16.281007751937985, "grad_norm": 0.38457584381103516, "learning_rate": 0.00035602935236299984, "loss": 3.0314, "step": 60500 }, { "epoch": 16.294465977605512, "grad_norm": 0.4016266167163849, "learning_rate": 0.00035582738656254203, "loss": 3.0388, "step": 60550 }, { "epoch": 16.30792420327304, "grad_norm": 0.4087277054786682, "learning_rate": 0.00035562542076208423, "loss": 3.0362, "step": 60600 }, { "epoch": 16.32138242894057, "grad_norm": 0.3765519857406616, "learning_rate": 0.0003554234549616264, "loss": 3.0337, "step": 60650 }, { "epoch": 16.334840654608097, "grad_norm": 0.3905170261859894, "learning_rate": 0.0003552214891611687, "loss": 3.0391, "step": 60700 }, { "epoch": 16.348298880275625, "grad_norm": 0.404747873544693, "learning_rate": 0.00035501952336071087, "loss": 3.0303, "step": 60750 }, { "epoch": 16.361757105943152, "grad_norm": 0.41902950406074524, "learning_rate": 0.00035481755756025307, "loss": 3.0401, "step": 60800 }, { "epoch": 16.37521533161068, "grad_norm": 0.3708718419075012, "learning_rate": 0.00035461559175979526, "loss": 3.0386, "step": 60850 }, { "epoch": 16.38867355727821, "grad_norm": 0.38788917660713196, "learning_rate": 0.0003544136259593375, "loss": 3.0364, "step": 60900 }, { "epoch": 16.402131782945737, "grad_norm": 0.4070287048816681, "learning_rate": 0.00035421166015887977, "loss": 3.0424, "step": 60950 }, { "epoch": 16.415590008613265, "grad_norm": 0.3842083513736725, "learning_rate": 0.00035400969435842196, "loss": 3.038, "step": 61000 }, { "epoch": 16.415590008613265, "eval_accuracy": 0.3939595825005552, "eval_loss": 3.2929747104644775, "eval_runtime": 84.1825, "eval_samples_per_second": 213.964, "eval_steps_per_second": 13.376, "step": 61000 }, { "epoch": 16.429048234280792, "grad_norm": 0.3588254749774933, "learning_rate": 0.00035380772855796416, "loss": 3.0327, "step": 61050 }, { "epoch": 16.44250645994832, "grad_norm": 0.4259098470211029, "learning_rate": 0.0003536057627575064, "loss": 3.0472, "step": 61100 }, { "epoch": 16.45596468561585, "grad_norm": 0.38360491394996643, "learning_rate": 0.0003534037969570486, "loss": 3.0426, "step": 61150 }, { "epoch": 16.469422911283377, "grad_norm": 0.40665701031684875, "learning_rate": 0.0003532018311565908, "loss": 3.0399, "step": 61200 }, { "epoch": 16.482881136950905, "grad_norm": 0.39631187915802, "learning_rate": 0.000352999865356133, "loss": 3.0323, "step": 61250 }, { "epoch": 16.496339362618432, "grad_norm": 0.3738632798194885, "learning_rate": 0.0003527978995556752, "loss": 3.0373, "step": 61300 }, { "epoch": 16.50979758828596, "grad_norm": 0.4072265028953552, "learning_rate": 0.00035259593375521744, "loss": 3.0502, "step": 61350 }, { "epoch": 16.52325581395349, "grad_norm": 0.39553919434547424, "learning_rate": 0.00035239396795475964, "loss": 3.0483, "step": 61400 }, { "epoch": 16.536714039621017, "grad_norm": 0.4070907235145569, "learning_rate": 0.00035219200215430183, "loss": 3.0439, "step": 61450 }, { "epoch": 16.550172265288545, "grad_norm": 0.4034261405467987, "learning_rate": 0.00035199003635384403, "loss": 3.047, "step": 61500 }, { "epoch": 16.563630490956072, "grad_norm": 0.3746528923511505, "learning_rate": 0.0003517880705533862, "loss": 3.0452, "step": 61550 }, { "epoch": 16.5770887166236, "grad_norm": 0.37463000416755676, "learning_rate": 0.0003515861047529285, "loss": 3.0557, "step": 61600 }, { "epoch": 16.590546942291127, "grad_norm": 0.3937366008758545, "learning_rate": 0.0003513841389524707, "loss": 3.0455, "step": 61650 }, { "epoch": 16.604005167958658, "grad_norm": 0.3887689411640167, "learning_rate": 0.00035118217315201287, "loss": 3.0628, "step": 61700 }, { "epoch": 16.617463393626185, "grad_norm": 0.3895772695541382, "learning_rate": 0.00035098020735155507, "loss": 3.0501, "step": 61750 }, { "epoch": 16.630921619293712, "grad_norm": 0.3983083665370941, "learning_rate": 0.00035077824155109737, "loss": 3.0507, "step": 61800 }, { "epoch": 16.64437984496124, "grad_norm": 0.40093982219696045, "learning_rate": 0.00035057627575063957, "loss": 3.0563, "step": 61850 }, { "epoch": 16.657838070628767, "grad_norm": 0.38137128949165344, "learning_rate": 0.00035037430995018176, "loss": 3.0613, "step": 61900 }, { "epoch": 16.671296296296298, "grad_norm": 0.3852699398994446, "learning_rate": 0.00035017234414972396, "loss": 3.0527, "step": 61950 }, { "epoch": 16.684754521963825, "grad_norm": 0.39291974902153015, "learning_rate": 0.00034997037834926615, "loss": 3.0502, "step": 62000 }, { "epoch": 16.684754521963825, "eval_accuracy": 0.39450640212461097, "eval_loss": 3.2875256538391113, "eval_runtime": 84.1718, "eval_samples_per_second": 213.991, "eval_steps_per_second": 13.377, "step": 62000 }, { "epoch": 16.698212747631352, "grad_norm": 0.3874092102050781, "learning_rate": 0.0003497684125488084, "loss": 3.061, "step": 62050 }, { "epoch": 16.71167097329888, "grad_norm": 0.40655258297920227, "learning_rate": 0.0003495664467483506, "loss": 3.0631, "step": 62100 }, { "epoch": 16.725129198966407, "grad_norm": 0.3778446614742279, "learning_rate": 0.0003493644809478928, "loss": 3.0609, "step": 62150 }, { "epoch": 16.738587424633938, "grad_norm": 0.36551788449287415, "learning_rate": 0.000349162515147435, "loss": 3.0705, "step": 62200 }, { "epoch": 16.752045650301465, "grad_norm": 0.4111713469028473, "learning_rate": 0.0003489605493469772, "loss": 3.0618, "step": 62250 }, { "epoch": 16.765503875968992, "grad_norm": 0.4054068326950073, "learning_rate": 0.00034875858354651944, "loss": 3.0539, "step": 62300 }, { "epoch": 16.77896210163652, "grad_norm": 0.39005810022354126, "learning_rate": 0.00034855661774606164, "loss": 3.049, "step": 62350 }, { "epoch": 16.792420327304047, "grad_norm": 0.3782937228679657, "learning_rate": 0.00034835465194560383, "loss": 3.0594, "step": 62400 }, { "epoch": 16.805878552971578, "grad_norm": 0.3960946798324585, "learning_rate": 0.00034815268614514603, "loss": 3.0617, "step": 62450 }, { "epoch": 16.819336778639105, "grad_norm": 0.3699169158935547, "learning_rate": 0.0003479507203446882, "loss": 3.0675, "step": 62500 }, { "epoch": 16.832795004306632, "grad_norm": 0.38070520758628845, "learning_rate": 0.0003477487545442305, "loss": 3.0642, "step": 62550 }, { "epoch": 16.84625322997416, "grad_norm": 0.3870474100112915, "learning_rate": 0.00034754678874377267, "loss": 3.0738, "step": 62600 }, { "epoch": 16.859711455641687, "grad_norm": 0.38762134313583374, "learning_rate": 0.00034734482294331487, "loss": 3.0511, "step": 62650 }, { "epoch": 16.873169681309218, "grad_norm": 0.3905826807022095, "learning_rate": 0.00034714285714285717, "loss": 3.0606, "step": 62700 }, { "epoch": 16.886627906976745, "grad_norm": 0.3941311538219452, "learning_rate": 0.00034694089134239937, "loss": 3.0604, "step": 62750 }, { "epoch": 16.900086132644272, "grad_norm": 0.3685128390789032, "learning_rate": 0.00034673892554194156, "loss": 3.0612, "step": 62800 }, { "epoch": 16.9135443583118, "grad_norm": 0.403667688369751, "learning_rate": 0.00034653695974148376, "loss": 3.0636, "step": 62850 }, { "epoch": 16.927002583979327, "grad_norm": 0.41439446806907654, "learning_rate": 0.00034633499394102595, "loss": 3.076, "step": 62900 }, { "epoch": 16.940460809646858, "grad_norm": 0.4269420802593231, "learning_rate": 0.0003461330281405682, "loss": 3.064, "step": 62950 }, { "epoch": 16.953919035314385, "grad_norm": 0.4000409245491028, "learning_rate": 0.0003459310623401104, "loss": 3.0795, "step": 63000 }, { "epoch": 16.953919035314385, "eval_accuracy": 0.39524346239276015, "eval_loss": 3.2783825397491455, "eval_runtime": 84.6772, "eval_samples_per_second": 212.714, "eval_steps_per_second": 13.298, "step": 63000 }, { "epoch": 16.967377260981912, "grad_norm": 0.4015685021877289, "learning_rate": 0.0003457290965396526, "loss": 3.0622, "step": 63050 }, { "epoch": 16.98083548664944, "grad_norm": 0.40094172954559326, "learning_rate": 0.0003455271307391948, "loss": 3.0641, "step": 63100 }, { "epoch": 16.994293712316967, "grad_norm": 0.40834230184555054, "learning_rate": 0.000345325164938737, "loss": 3.0729, "step": 63150 }, { "epoch": 17.007536606373815, "grad_norm": 0.43471765518188477, "learning_rate": 0.00034512319913827924, "loss": 3.0287, "step": 63200 }, { "epoch": 17.020994832041342, "grad_norm": 0.40283820033073425, "learning_rate": 0.00034492123333782144, "loss": 2.9724, "step": 63250 }, { "epoch": 17.034453057708873, "grad_norm": 0.39246246218681335, "learning_rate": 0.00034471926753736363, "loss": 2.9799, "step": 63300 }, { "epoch": 17.0479112833764, "grad_norm": 0.38070741295814514, "learning_rate": 0.00034451730173690583, "loss": 2.9707, "step": 63350 }, { "epoch": 17.061369509043928, "grad_norm": 0.3874499797821045, "learning_rate": 0.000344315335936448, "loss": 2.9896, "step": 63400 }, { "epoch": 17.074827734711455, "grad_norm": 0.411945641040802, "learning_rate": 0.0003441133701359903, "loss": 2.9876, "step": 63450 }, { "epoch": 17.088285960378983, "grad_norm": 0.38966816663742065, "learning_rate": 0.00034391140433553247, "loss": 2.9961, "step": 63500 }, { "epoch": 17.101744186046513, "grad_norm": 0.3948839604854584, "learning_rate": 0.00034370943853507467, "loss": 2.9908, "step": 63550 }, { "epoch": 17.11520241171404, "grad_norm": 0.44139349460601807, "learning_rate": 0.0003435074727346169, "loss": 3.0079, "step": 63600 }, { "epoch": 17.128660637381568, "grad_norm": 0.40904638171195984, "learning_rate": 0.00034330550693415917, "loss": 2.9926, "step": 63650 }, { "epoch": 17.142118863049095, "grad_norm": 0.37809765338897705, "learning_rate": 0.00034310354113370136, "loss": 2.9871, "step": 63700 }, { "epoch": 17.155577088716623, "grad_norm": 0.4005953371524811, "learning_rate": 0.00034290157533324356, "loss": 2.9962, "step": 63750 }, { "epoch": 17.16903531438415, "grad_norm": 0.3982625901699066, "learning_rate": 0.00034269960953278576, "loss": 3.0034, "step": 63800 }, { "epoch": 17.18249354005168, "grad_norm": 0.37310752272605896, "learning_rate": 0.00034249764373232795, "loss": 2.9922, "step": 63850 }, { "epoch": 17.195951765719208, "grad_norm": 0.35717523097991943, "learning_rate": 0.0003422956779318702, "loss": 3.0183, "step": 63900 }, { "epoch": 17.209409991386735, "grad_norm": 0.41922739148139954, "learning_rate": 0.0003420937121314124, "loss": 3.0097, "step": 63950 }, { "epoch": 17.222868217054263, "grad_norm": 0.4463370740413666, "learning_rate": 0.0003418917463309546, "loss": 3.0119, "step": 64000 }, { "epoch": 17.222868217054263, "eval_accuracy": 0.39430344979841664, "eval_loss": 3.293164014816284, "eval_runtime": 84.1962, "eval_samples_per_second": 213.929, "eval_steps_per_second": 13.374, "step": 64000 }, { "epoch": 17.23632644272179, "grad_norm": 0.4146665334701538, "learning_rate": 0.0003416897805304968, "loss": 3.0098, "step": 64050 }, { "epoch": 17.24978466838932, "grad_norm": 0.39131778478622437, "learning_rate": 0.000341487814730039, "loss": 3.0131, "step": 64100 }, { "epoch": 17.263242894056848, "grad_norm": 0.3792233467102051, "learning_rate": 0.00034128584892958124, "loss": 3.0138, "step": 64150 }, { "epoch": 17.276701119724375, "grad_norm": 0.38367384672164917, "learning_rate": 0.00034108388312912343, "loss": 3.0241, "step": 64200 }, { "epoch": 17.290159345391903, "grad_norm": 0.40969136357307434, "learning_rate": 0.00034088191732866563, "loss": 3.0249, "step": 64250 }, { "epoch": 17.30361757105943, "grad_norm": 0.41047418117523193, "learning_rate": 0.0003406799515282078, "loss": 3.0121, "step": 64300 }, { "epoch": 17.31707579672696, "grad_norm": 0.4021008610725403, "learning_rate": 0.00034047798572775, "loss": 3.0179, "step": 64350 }, { "epoch": 17.330534022394488, "grad_norm": 0.41244468092918396, "learning_rate": 0.00034027601992729227, "loss": 3.025, "step": 64400 }, { "epoch": 17.343992248062015, "grad_norm": 0.41508427262306213, "learning_rate": 0.00034007405412683447, "loss": 3.0094, "step": 64450 }, { "epoch": 17.357450473729543, "grad_norm": 0.3844899833202362, "learning_rate": 0.0003398720883263767, "loss": 3.0283, "step": 64500 }, { "epoch": 17.37090869939707, "grad_norm": 0.3954097628593445, "learning_rate": 0.00033967012252591897, "loss": 3.029, "step": 64550 }, { "epoch": 17.3843669250646, "grad_norm": 0.4179207682609558, "learning_rate": 0.00033946815672546116, "loss": 3.025, "step": 64600 }, { "epoch": 17.397825150732128, "grad_norm": 0.45622459053993225, "learning_rate": 0.00033926619092500336, "loss": 3.0297, "step": 64650 }, { "epoch": 17.411283376399656, "grad_norm": 0.383299857378006, "learning_rate": 0.00033906422512454556, "loss": 3.0269, "step": 64700 }, { "epoch": 17.424741602067183, "grad_norm": 0.42966628074645996, "learning_rate": 0.00033886225932408775, "loss": 3.0307, "step": 64750 }, { "epoch": 17.43819982773471, "grad_norm": 0.41395410895347595, "learning_rate": 0.00033866029352363, "loss": 3.0371, "step": 64800 }, { "epoch": 17.45165805340224, "grad_norm": 0.3621746897697449, "learning_rate": 0.0003384583277231722, "loss": 3.0224, "step": 64850 }, { "epoch": 17.46511627906977, "grad_norm": 0.4138607382774353, "learning_rate": 0.0003382563619227144, "loss": 3.0178, "step": 64900 }, { "epoch": 17.478574504737296, "grad_norm": 0.39790982007980347, "learning_rate": 0.0003380543961222566, "loss": 3.0318, "step": 64950 }, { "epoch": 17.492032730404823, "grad_norm": 0.4042600989341736, "learning_rate": 0.0003378524303217988, "loss": 3.0361, "step": 65000 }, { "epoch": 17.492032730404823, "eval_accuracy": 0.39438504358694554, "eval_loss": 3.2896738052368164, "eval_runtime": 84.2441, "eval_samples_per_second": 213.807, "eval_steps_per_second": 13.366, "step": 65000 }, { "epoch": 17.50549095607235, "grad_norm": 0.4193043112754822, "learning_rate": 0.00033765046452134104, "loss": 3.0426, "step": 65050 }, { "epoch": 17.51894918173988, "grad_norm": 0.40196123719215393, "learning_rate": 0.00033744849872088323, "loss": 3.0348, "step": 65100 }, { "epoch": 17.53240740740741, "grad_norm": 0.40320277214050293, "learning_rate": 0.00033724653292042543, "loss": 3.0389, "step": 65150 }, { "epoch": 17.545865633074936, "grad_norm": 0.3964909613132477, "learning_rate": 0.0003370445671199676, "loss": 3.0265, "step": 65200 }, { "epoch": 17.559323858742463, "grad_norm": 0.40705111622810364, "learning_rate": 0.0003368426013195098, "loss": 3.0402, "step": 65250 }, { "epoch": 17.57278208440999, "grad_norm": 0.38822999596595764, "learning_rate": 0.00033664063551905207, "loss": 3.0294, "step": 65300 }, { "epoch": 17.58624031007752, "grad_norm": 0.3868526518344879, "learning_rate": 0.00033643866971859427, "loss": 3.0417, "step": 65350 }, { "epoch": 17.59969853574505, "grad_norm": 0.3836658000946045, "learning_rate": 0.0003362367039181365, "loss": 3.0319, "step": 65400 }, { "epoch": 17.613156761412576, "grad_norm": 0.39039117097854614, "learning_rate": 0.0003360347381176787, "loss": 3.0439, "step": 65450 }, { "epoch": 17.626614987080103, "grad_norm": 0.3786974251270294, "learning_rate": 0.00033583277231722097, "loss": 3.0384, "step": 65500 }, { "epoch": 17.64007321274763, "grad_norm": 0.38470643758773804, "learning_rate": 0.00033563080651676316, "loss": 3.0402, "step": 65550 }, { "epoch": 17.653531438415158, "grad_norm": 0.3940637707710266, "learning_rate": 0.00033542884071630536, "loss": 3.0493, "step": 65600 }, { "epoch": 17.66698966408269, "grad_norm": 0.3953503370285034, "learning_rate": 0.00033522687491584755, "loss": 3.0447, "step": 65650 }, { "epoch": 17.680447889750216, "grad_norm": 0.38392022252082825, "learning_rate": 0.00033502490911538975, "loss": 3.0463, "step": 65700 }, { "epoch": 17.693906115417743, "grad_norm": 0.41958823800086975, "learning_rate": 0.000334822943314932, "loss": 3.0397, "step": 65750 }, { "epoch": 17.70736434108527, "grad_norm": 0.40301191806793213, "learning_rate": 0.0003346209775144742, "loss": 3.057, "step": 65800 }, { "epoch": 17.720822566752798, "grad_norm": 0.4074375033378601, "learning_rate": 0.0003344190117140164, "loss": 3.0343, "step": 65850 }, { "epoch": 17.73428079242033, "grad_norm": 0.4151737689971924, "learning_rate": 0.0003342170459135586, "loss": 3.0405, "step": 65900 }, { "epoch": 17.747739018087856, "grad_norm": 0.43711110949516296, "learning_rate": 0.0003340150801131008, "loss": 3.0309, "step": 65950 }, { "epoch": 17.761197243755383, "grad_norm": 0.3818199634552002, "learning_rate": 0.00033381311431264303, "loss": 3.0541, "step": 66000 }, { "epoch": 17.761197243755383, "eval_accuracy": 0.39482212988688126, "eval_loss": 3.2832953929901123, "eval_runtime": 84.4291, "eval_samples_per_second": 213.339, "eval_steps_per_second": 13.337, "step": 66000 }, { "epoch": 17.77465546942291, "grad_norm": 0.3675343990325928, "learning_rate": 0.00033361114851218523, "loss": 3.0484, "step": 66050 }, { "epoch": 17.788113695090438, "grad_norm": 0.40552109479904175, "learning_rate": 0.0003334091827117274, "loss": 3.0509, "step": 66100 }, { "epoch": 17.80157192075797, "grad_norm": 0.38828209042549133, "learning_rate": 0.0003332072169112696, "loss": 3.0509, "step": 66150 }, { "epoch": 17.815030146425496, "grad_norm": 0.40719807147979736, "learning_rate": 0.0003330052511108118, "loss": 3.0486, "step": 66200 }, { "epoch": 17.828488372093023, "grad_norm": 0.4311251938343048, "learning_rate": 0.0003328032853103541, "loss": 3.05, "step": 66250 }, { "epoch": 17.84194659776055, "grad_norm": 0.39757847785949707, "learning_rate": 0.0003326013195098963, "loss": 3.045, "step": 66300 }, { "epoch": 17.855404823428078, "grad_norm": 0.38673529028892517, "learning_rate": 0.0003323993537094385, "loss": 3.0503, "step": 66350 }, { "epoch": 17.86886304909561, "grad_norm": 0.4126622676849365, "learning_rate": 0.00033219738790898077, "loss": 3.0606, "step": 66400 }, { "epoch": 17.882321274763136, "grad_norm": 0.41488194465637207, "learning_rate": 0.00033199542210852296, "loss": 3.0484, "step": 66450 }, { "epoch": 17.895779500430663, "grad_norm": 0.39812037348747253, "learning_rate": 0.00033179345630806516, "loss": 3.0533, "step": 66500 }, { "epoch": 17.90923772609819, "grad_norm": 0.40347662568092346, "learning_rate": 0.00033159149050760735, "loss": 3.0444, "step": 66550 }, { "epoch": 17.922695951765718, "grad_norm": 0.38835588097572327, "learning_rate": 0.00033138952470714955, "loss": 3.0562, "step": 66600 }, { "epoch": 17.93615417743325, "grad_norm": 0.38559824228286743, "learning_rate": 0.0003311875589066918, "loss": 3.0597, "step": 66650 }, { "epoch": 17.949612403100776, "grad_norm": 0.39280909299850464, "learning_rate": 0.000330985593106234, "loss": 3.0506, "step": 66700 }, { "epoch": 17.963070628768303, "grad_norm": 0.4078952968120575, "learning_rate": 0.0003307836273057762, "loss": 3.0493, "step": 66750 }, { "epoch": 17.97652885443583, "grad_norm": 0.3855295479297638, "learning_rate": 0.0003305816615053184, "loss": 3.0515, "step": 66800 }, { "epoch": 17.989987080103358, "grad_norm": 0.3947485387325287, "learning_rate": 0.0003303796957048606, "loss": 3.0457, "step": 66850 }, { "epoch": 18.003229974160206, "grad_norm": 0.3720707297325134, "learning_rate": 0.00033017772990440284, "loss": 3.0407, "step": 66900 }, { "epoch": 18.016688199827733, "grad_norm": 0.39281925559043884, "learning_rate": 0.00032997576410394503, "loss": 2.9672, "step": 66950 }, { "epoch": 18.030146425495264, "grad_norm": 0.40296903252601624, "learning_rate": 0.00032977379830348723, "loss": 2.9625, "step": 67000 }, { "epoch": 18.030146425495264, "eval_accuracy": 0.3947613962946207, "eval_loss": 3.2923240661621094, "eval_runtime": 84.3224, "eval_samples_per_second": 213.609, "eval_steps_per_second": 13.354, "step": 67000 }, { "epoch": 18.04360465116279, "grad_norm": 0.39170727133750916, "learning_rate": 0.0003295718325030294, "loss": 2.9558, "step": 67050 }, { "epoch": 18.05706287683032, "grad_norm": 0.42225125432014465, "learning_rate": 0.0003293698667025716, "loss": 2.9637, "step": 67100 }, { "epoch": 18.070521102497846, "grad_norm": 0.38866063952445984, "learning_rate": 0.0003291679009021139, "loss": 2.979, "step": 67150 }, { "epoch": 18.083979328165373, "grad_norm": 0.4313035309314728, "learning_rate": 0.0003289659351016561, "loss": 2.9725, "step": 67200 }, { "epoch": 18.097437553832904, "grad_norm": 0.43296217918395996, "learning_rate": 0.0003287639693011983, "loss": 2.9914, "step": 67250 }, { "epoch": 18.11089577950043, "grad_norm": 0.39378491044044495, "learning_rate": 0.0003285620035007405, "loss": 2.9781, "step": 67300 }, { "epoch": 18.12435400516796, "grad_norm": 0.4113011062145233, "learning_rate": 0.00032836003770028276, "loss": 2.9787, "step": 67350 }, { "epoch": 18.137812230835486, "grad_norm": 0.4032084345817566, "learning_rate": 0.00032815807189982496, "loss": 2.9797, "step": 67400 }, { "epoch": 18.151270456503013, "grad_norm": 0.405706524848938, "learning_rate": 0.00032795610609936715, "loss": 2.9983, "step": 67450 }, { "epoch": 18.164728682170544, "grad_norm": 0.40440014004707336, "learning_rate": 0.00032775414029890935, "loss": 2.9931, "step": 67500 }, { "epoch": 18.17818690783807, "grad_norm": 0.4321618378162384, "learning_rate": 0.00032755217449845155, "loss": 2.9937, "step": 67550 }, { "epoch": 18.1916451335056, "grad_norm": 0.42441773414611816, "learning_rate": 0.0003273502086979938, "loss": 2.99, "step": 67600 }, { "epoch": 18.205103359173126, "grad_norm": 0.38702791929244995, "learning_rate": 0.000327148242897536, "loss": 2.991, "step": 67650 }, { "epoch": 18.218561584840653, "grad_norm": 0.4204156696796417, "learning_rate": 0.0003269462770970782, "loss": 2.9891, "step": 67700 }, { "epoch": 18.232019810508184, "grad_norm": 0.4095945656299591, "learning_rate": 0.0003267443112966204, "loss": 3.0104, "step": 67750 }, { "epoch": 18.24547803617571, "grad_norm": 0.395253449678421, "learning_rate": 0.0003265423454961626, "loss": 3.0061, "step": 67800 }, { "epoch": 18.25893626184324, "grad_norm": 0.40738797187805176, "learning_rate": 0.00032634037969570483, "loss": 2.9976, "step": 67850 }, { "epoch": 18.272394487510766, "grad_norm": 0.4268544912338257, "learning_rate": 0.00032613841389524703, "loss": 3.0001, "step": 67900 }, { "epoch": 18.285852713178294, "grad_norm": 0.40578269958496094, "learning_rate": 0.0003259364480947892, "loss": 3.0027, "step": 67950 }, { "epoch": 18.29931093884582, "grad_norm": 0.3866397440433502, "learning_rate": 0.0003257344822943314, "loss": 3.0122, "step": 68000 }, { "epoch": 18.29931093884582, "eval_accuracy": 0.3947345605212963, "eval_loss": 3.292531967163086, "eval_runtime": 84.2705, "eval_samples_per_second": 213.74, "eval_steps_per_second": 13.362, "step": 68000 }, { "epoch": 18.31276916451335, "grad_norm": 0.434059739112854, "learning_rate": 0.0003255325164938737, "loss": 3.0145, "step": 68050 }, { "epoch": 18.32622739018088, "grad_norm": 0.40752777457237244, "learning_rate": 0.0003253305506934159, "loss": 3.0132, "step": 68100 }, { "epoch": 18.339685615848406, "grad_norm": 0.3976019620895386, "learning_rate": 0.0003251285848929581, "loss": 3.0111, "step": 68150 }, { "epoch": 18.353143841515934, "grad_norm": 0.4011041820049286, "learning_rate": 0.0003249266190925003, "loss": 3.0114, "step": 68200 }, { "epoch": 18.36660206718346, "grad_norm": 0.42141130566596985, "learning_rate": 0.0003247246532920425, "loss": 3.0157, "step": 68250 }, { "epoch": 18.38006029285099, "grad_norm": 0.44433796405792236, "learning_rate": 0.00032452268749158476, "loss": 3.0075, "step": 68300 }, { "epoch": 18.39351851851852, "grad_norm": 0.39202117919921875, "learning_rate": 0.00032432072169112696, "loss": 3.0201, "step": 68350 }, { "epoch": 18.406976744186046, "grad_norm": 0.41736364364624023, "learning_rate": 0.00032411875589066915, "loss": 3.0086, "step": 68400 }, { "epoch": 18.420434969853574, "grad_norm": 0.39685291051864624, "learning_rate": 0.00032391679009021135, "loss": 3.016, "step": 68450 }, { "epoch": 18.4338931955211, "grad_norm": 0.43723633885383606, "learning_rate": 0.0003237148242897536, "loss": 3.0146, "step": 68500 }, { "epoch": 18.447351421188632, "grad_norm": 0.42096295952796936, "learning_rate": 0.0003235128584892958, "loss": 3.0179, "step": 68550 }, { "epoch": 18.46080964685616, "grad_norm": 0.41369014978408813, "learning_rate": 0.000323310892688838, "loss": 3.0155, "step": 68600 }, { "epoch": 18.474267872523686, "grad_norm": 0.41110730171203613, "learning_rate": 0.0003231089268883802, "loss": 3.015, "step": 68650 }, { "epoch": 18.487726098191214, "grad_norm": 0.39935365319252014, "learning_rate": 0.0003229069610879224, "loss": 3.0315, "step": 68700 }, { "epoch": 18.50118432385874, "grad_norm": 0.3989832401275635, "learning_rate": 0.00032270499528746463, "loss": 3.0217, "step": 68750 }, { "epoch": 18.514642549526272, "grad_norm": 0.4078866243362427, "learning_rate": 0.00032250302948700683, "loss": 3.0264, "step": 68800 }, { "epoch": 18.5281007751938, "grad_norm": 0.4013637900352478, "learning_rate": 0.000322301063686549, "loss": 3.0267, "step": 68850 }, { "epoch": 18.541559000861326, "grad_norm": 0.38905027508735657, "learning_rate": 0.0003220990978860912, "loss": 3.0212, "step": 68900 }, { "epoch": 18.555017226528854, "grad_norm": 0.4122450351715088, "learning_rate": 0.0003218971320856335, "loss": 3.0263, "step": 68950 }, { "epoch": 18.56847545219638, "grad_norm": 0.40349528193473816, "learning_rate": 0.0003216951662851757, "loss": 3.0234, "step": 69000 }, { "epoch": 18.56847545219638, "eval_accuracy": 0.39492480116538964, "eval_loss": 3.2897045612335205, "eval_runtime": 84.2999, "eval_samples_per_second": 213.666, "eval_steps_per_second": 13.357, "step": 69000 }, { "epoch": 18.581933677863912, "grad_norm": 0.3932398557662964, "learning_rate": 0.0003214932004847179, "loss": 3.0295, "step": 69050 }, { "epoch": 18.59539190353144, "grad_norm": 0.4231753349304199, "learning_rate": 0.0003212912346842601, "loss": 3.022, "step": 69100 }, { "epoch": 18.608850129198967, "grad_norm": 0.41071921586990356, "learning_rate": 0.0003210892688838023, "loss": 3.0148, "step": 69150 }, { "epoch": 18.622308354866494, "grad_norm": 0.4104444980621338, "learning_rate": 0.00032088730308334456, "loss": 3.0233, "step": 69200 }, { "epoch": 18.63576658053402, "grad_norm": 0.4138637185096741, "learning_rate": 0.00032068533728288676, "loss": 3.0323, "step": 69250 }, { "epoch": 18.649224806201552, "grad_norm": 0.3882623016834259, "learning_rate": 0.00032048337148242895, "loss": 3.0267, "step": 69300 }, { "epoch": 18.66268303186908, "grad_norm": 0.41541221737861633, "learning_rate": 0.00032028140568197115, "loss": 3.0318, "step": 69350 }, { "epoch": 18.676141257536607, "grad_norm": 0.4507473111152649, "learning_rate": 0.00032007943988151334, "loss": 3.0323, "step": 69400 }, { "epoch": 18.689599483204134, "grad_norm": 0.4193718433380127, "learning_rate": 0.0003198774740810556, "loss": 3.0344, "step": 69450 }, { "epoch": 18.70305770887166, "grad_norm": 0.42218124866485596, "learning_rate": 0.0003196755082805978, "loss": 3.0344, "step": 69500 }, { "epoch": 18.716515934539192, "grad_norm": 0.3834584653377533, "learning_rate": 0.00031947354248014, "loss": 3.0258, "step": 69550 }, { "epoch": 18.72997416020672, "grad_norm": 0.410442978143692, "learning_rate": 0.0003192715766796822, "loss": 3.0256, "step": 69600 }, { "epoch": 18.743432385874247, "grad_norm": 0.42289379239082336, "learning_rate": 0.0003190696108792244, "loss": 3.0309, "step": 69650 }, { "epoch": 18.756890611541774, "grad_norm": 0.4159040153026581, "learning_rate": 0.00031886764507876663, "loss": 3.0447, "step": 69700 }, { "epoch": 18.7703488372093, "grad_norm": 0.39476460218429565, "learning_rate": 0.0003186656792783088, "loss": 3.0321, "step": 69750 }, { "epoch": 18.783807062876832, "grad_norm": 0.4341265559196472, "learning_rate": 0.000318463713477851, "loss": 3.0538, "step": 69800 }, { "epoch": 18.79726528854436, "grad_norm": 0.3935094475746155, "learning_rate": 0.00031826174767739327, "loss": 3.0307, "step": 69850 }, { "epoch": 18.810723514211887, "grad_norm": 0.41315150260925293, "learning_rate": 0.0003180597818769355, "loss": 3.0372, "step": 69900 }, { "epoch": 18.824181739879414, "grad_norm": 0.43439584970474243, "learning_rate": 0.0003178578160764777, "loss": 3.0453, "step": 69950 }, { "epoch": 18.83763996554694, "grad_norm": 0.3953023850917816, "learning_rate": 0.0003176558502760199, "loss": 3.0303, "step": 70000 }, { "epoch": 18.83763996554694, "eval_accuracy": 0.3956587106747274, "eval_loss": 3.2794418334960938, "eval_runtime": 84.2367, "eval_samples_per_second": 213.826, "eval_steps_per_second": 13.367, "step": 70000 }, { "epoch": 18.85109819121447, "grad_norm": 0.3870137631893158, "learning_rate": 0.0003174538844755621, "loss": 3.0376, "step": 70050 }, { "epoch": 18.864556416882, "grad_norm": 0.4227631688117981, "learning_rate": 0.0003172519186751043, "loss": 3.041, "step": 70100 }, { "epoch": 18.878014642549527, "grad_norm": 0.3939836919307709, "learning_rate": 0.00031704995287464656, "loss": 3.0363, "step": 70150 }, { "epoch": 18.891472868217054, "grad_norm": 0.40550845861434937, "learning_rate": 0.00031684798707418875, "loss": 3.0482, "step": 70200 }, { "epoch": 18.90493109388458, "grad_norm": 0.3825607895851135, "learning_rate": 0.00031664602127373095, "loss": 3.0388, "step": 70250 }, { "epoch": 18.91838931955211, "grad_norm": 0.39878538250923157, "learning_rate": 0.00031644405547327315, "loss": 3.0499, "step": 70300 }, { "epoch": 18.93184754521964, "grad_norm": 0.42560845613479614, "learning_rate": 0.00031624208967281534, "loss": 3.0404, "step": 70350 }, { "epoch": 18.945305770887167, "grad_norm": 0.39959248900413513, "learning_rate": 0.0003160401238723576, "loss": 3.0438, "step": 70400 }, { "epoch": 18.958763996554694, "grad_norm": 0.3780984580516815, "learning_rate": 0.0003158381580718998, "loss": 3.0429, "step": 70450 }, { "epoch": 18.97222222222222, "grad_norm": 0.4092783033847809, "learning_rate": 0.000315636192271442, "loss": 3.053, "step": 70500 }, { "epoch": 18.98568044788975, "grad_norm": 0.3804001808166504, "learning_rate": 0.0003154342264709842, "loss": 3.0393, "step": 70550 }, { "epoch": 18.99913867355728, "grad_norm": 0.44005870819091797, "learning_rate": 0.00031523226067052643, "loss": 3.0403, "step": 70600 }, { "epoch": 19.012381567614124, "grad_norm": 0.39213648438453674, "learning_rate": 0.0003150302948700686, "loss": 2.9572, "step": 70650 }, { "epoch": 19.025839793281655, "grad_norm": 0.40990149974823, "learning_rate": 0.0003148283290696108, "loss": 2.9514, "step": 70700 }, { "epoch": 19.039298018949182, "grad_norm": 0.43813619017601013, "learning_rate": 0.0003146263632691531, "loss": 2.956, "step": 70750 }, { "epoch": 19.05275624461671, "grad_norm": 0.4364962577819824, "learning_rate": 0.0003144243974686953, "loss": 2.9542, "step": 70800 }, { "epoch": 19.066214470284237, "grad_norm": 0.44929248094558716, "learning_rate": 0.0003142224316682375, "loss": 2.9621, "step": 70850 }, { "epoch": 19.079672695951764, "grad_norm": 0.4099480211734772, "learning_rate": 0.0003140204658677797, "loss": 2.9712, "step": 70900 }, { "epoch": 19.093130921619295, "grad_norm": 0.393709659576416, "learning_rate": 0.0003138185000673219, "loss": 2.9694, "step": 70950 }, { "epoch": 19.106589147286822, "grad_norm": 0.40786850452423096, "learning_rate": 0.0003136165342668641, "loss": 2.9725, "step": 71000 }, { "epoch": 19.106589147286822, "eval_accuracy": 0.3949505504701584, "eval_loss": 3.2920520305633545, "eval_runtime": 84.1457, "eval_samples_per_second": 214.057, "eval_steps_per_second": 13.382, "step": 71000 }, { "epoch": 19.12004737295435, "grad_norm": 0.4190482199192047, "learning_rate": 0.00031341456846640636, "loss": 2.9667, "step": 71050 }, { "epoch": 19.133505598621877, "grad_norm": 0.4100850820541382, "learning_rate": 0.00031321260266594855, "loss": 2.9747, "step": 71100 }, { "epoch": 19.146963824289404, "grad_norm": 0.4308258295059204, "learning_rate": 0.00031301063686549075, "loss": 2.9763, "step": 71150 }, { "epoch": 19.160422049956935, "grad_norm": 0.41372618079185486, "learning_rate": 0.00031280867106503295, "loss": 2.9849, "step": 71200 }, { "epoch": 19.173880275624462, "grad_norm": 0.40586134791374207, "learning_rate": 0.00031260670526457514, "loss": 2.9846, "step": 71250 }, { "epoch": 19.18733850129199, "grad_norm": 0.41523006558418274, "learning_rate": 0.0003124047394641174, "loss": 2.9732, "step": 71300 }, { "epoch": 19.200796726959517, "grad_norm": 0.4177076518535614, "learning_rate": 0.0003122027736636596, "loss": 2.9882, "step": 71350 }, { "epoch": 19.214254952627044, "grad_norm": 0.4282616674900055, "learning_rate": 0.0003120008078632018, "loss": 2.9876, "step": 71400 }, { "epoch": 19.227713178294575, "grad_norm": 0.40223127603530884, "learning_rate": 0.000311798842062744, "loss": 2.993, "step": 71450 }, { "epoch": 19.241171403962102, "grad_norm": 0.4155823886394501, "learning_rate": 0.0003115968762622862, "loss": 2.9904, "step": 71500 }, { "epoch": 19.25462962962963, "grad_norm": 0.4424571096897125, "learning_rate": 0.00031139491046182843, "loss": 2.9923, "step": 71550 }, { "epoch": 19.268087855297157, "grad_norm": 0.42144522070884705, "learning_rate": 0.0003111929446613707, "loss": 2.9912, "step": 71600 }, { "epoch": 19.281546080964684, "grad_norm": 0.4306216835975647, "learning_rate": 0.0003109909788609129, "loss": 2.9929, "step": 71650 }, { "epoch": 19.295004306632215, "grad_norm": 0.43049681186676025, "learning_rate": 0.00031078901306045507, "loss": 3.0012, "step": 71700 }, { "epoch": 19.308462532299743, "grad_norm": 0.4196663498878479, "learning_rate": 0.0003105870472599973, "loss": 2.9855, "step": 71750 }, { "epoch": 19.32192075796727, "grad_norm": 0.4187632203102112, "learning_rate": 0.0003103850814595395, "loss": 3.0017, "step": 71800 }, { "epoch": 19.335378983634797, "grad_norm": 0.3931756615638733, "learning_rate": 0.0003101831156590817, "loss": 3.0001, "step": 71850 }, { "epoch": 19.348837209302324, "grad_norm": 0.4297729432582855, "learning_rate": 0.0003099811498586239, "loss": 3.0043, "step": 71900 }, { "epoch": 19.362295434969855, "grad_norm": 0.4008636772632599, "learning_rate": 0.0003097791840581661, "loss": 3.0034, "step": 71950 }, { "epoch": 19.375753660637383, "grad_norm": 0.4368478059768677, "learning_rate": 0.00030957721825770836, "loss": 3.0149, "step": 72000 }, { "epoch": 19.375753660637383, "eval_accuracy": 0.39520391493733464, "eval_loss": 3.289865493774414, "eval_runtime": 84.3044, "eval_samples_per_second": 213.654, "eval_steps_per_second": 13.356, "step": 72000 }, { "epoch": 19.38921188630491, "grad_norm": 0.4222554564476013, "learning_rate": 0.00030937525245725055, "loss": 3.004, "step": 72050 }, { "epoch": 19.402670111972437, "grad_norm": 0.4352582097053528, "learning_rate": 0.00030917328665679275, "loss": 3.0041, "step": 72100 }, { "epoch": 19.416128337639964, "grad_norm": 0.41281789541244507, "learning_rate": 0.00030897132085633494, "loss": 3.001, "step": 72150 }, { "epoch": 19.429586563307495, "grad_norm": 0.4322177469730377, "learning_rate": 0.00030876935505587714, "loss": 3.0007, "step": 72200 }, { "epoch": 19.443044788975023, "grad_norm": 0.4109971225261688, "learning_rate": 0.0003085673892554194, "loss": 2.9968, "step": 72250 }, { "epoch": 19.45650301464255, "grad_norm": 0.4037865400314331, "learning_rate": 0.0003083654234549616, "loss": 3.0081, "step": 72300 }, { "epoch": 19.469961240310077, "grad_norm": 0.41936466097831726, "learning_rate": 0.0003081634576545038, "loss": 3.0122, "step": 72350 }, { "epoch": 19.483419465977605, "grad_norm": 0.41000697016716003, "learning_rate": 0.000307961491854046, "loss": 3.0012, "step": 72400 }, { "epoch": 19.496877691645132, "grad_norm": 0.40724891424179077, "learning_rate": 0.00030775952605358823, "loss": 3.0076, "step": 72450 }, { "epoch": 19.510335917312663, "grad_norm": 0.4325968325138092, "learning_rate": 0.0003075575602531305, "loss": 3.0025, "step": 72500 }, { "epoch": 19.52379414298019, "grad_norm": 0.4281940162181854, "learning_rate": 0.0003073555944526727, "loss": 3.0107, "step": 72550 }, { "epoch": 19.537252368647717, "grad_norm": 0.42506474256515503, "learning_rate": 0.00030715362865221487, "loss": 3.0072, "step": 72600 }, { "epoch": 19.550710594315245, "grad_norm": 0.43724924325942993, "learning_rate": 0.0003069516628517571, "loss": 3.0121, "step": 72650 }, { "epoch": 19.564168819982772, "grad_norm": 0.4503478705883026, "learning_rate": 0.0003067496970512993, "loss": 3.0069, "step": 72700 }, { "epoch": 19.577627045650303, "grad_norm": 0.4112701416015625, "learning_rate": 0.0003065477312508415, "loss": 3.0038, "step": 72750 }, { "epoch": 19.59108527131783, "grad_norm": 0.428783655166626, "learning_rate": 0.0003063457654503837, "loss": 3.0042, "step": 72800 }, { "epoch": 19.604543496985357, "grad_norm": 0.4140222668647766, "learning_rate": 0.0003061437996499259, "loss": 3.0208, "step": 72850 }, { "epoch": 19.618001722652885, "grad_norm": 0.39155247807502747, "learning_rate": 0.00030594183384946816, "loss": 3.0183, "step": 72900 }, { "epoch": 19.631459948320412, "grad_norm": 0.44719892740249634, "learning_rate": 0.00030573986804901035, "loss": 3.0273, "step": 72950 }, { "epoch": 19.644918173987943, "grad_norm": 0.42863699793815613, "learning_rate": 0.00030553790224855255, "loss": 3.0114, "step": 73000 }, { "epoch": 19.644918173987943, "eval_accuracy": 0.3954941106885473, "eval_loss": 3.282496690750122, "eval_runtime": 84.0925, "eval_samples_per_second": 214.193, "eval_steps_per_second": 13.39, "step": 73000 }, { "epoch": 19.65837639965547, "grad_norm": 0.40204405784606934, "learning_rate": 0.00030533593644809474, "loss": 3.0191, "step": 73050 }, { "epoch": 19.671834625322997, "grad_norm": 0.5403961539268494, "learning_rate": 0.00030513397064763694, "loss": 3.0225, "step": 73100 }, { "epoch": 19.685292850990525, "grad_norm": 0.3981429636478424, "learning_rate": 0.0003049320048471792, "loss": 3.017, "step": 73150 }, { "epoch": 19.698751076658052, "grad_norm": 0.41193249821662903, "learning_rate": 0.0003047300390467214, "loss": 3.0141, "step": 73200 }, { "epoch": 19.712209302325583, "grad_norm": 0.4144224226474762, "learning_rate": 0.0003045280732462636, "loss": 3.0261, "step": 73250 }, { "epoch": 19.72566752799311, "grad_norm": 0.41177770495414734, "learning_rate": 0.0003043261074458058, "loss": 3.0294, "step": 73300 }, { "epoch": 19.739125753660637, "grad_norm": 0.41590163111686707, "learning_rate": 0.000304124141645348, "loss": 3.0213, "step": 73350 }, { "epoch": 19.752583979328165, "grad_norm": 0.41663703322410583, "learning_rate": 0.0003039221758448903, "loss": 3.0266, "step": 73400 }, { "epoch": 19.766042204995692, "grad_norm": 0.4177576005458832, "learning_rate": 0.0003037202100444325, "loss": 3.0302, "step": 73450 }, { "epoch": 19.779500430663223, "grad_norm": 0.4391637444496155, "learning_rate": 0.00030351824424397467, "loss": 3.0236, "step": 73500 }, { "epoch": 19.79295865633075, "grad_norm": 0.42288199067115784, "learning_rate": 0.00030331627844351687, "loss": 3.0299, "step": 73550 }, { "epoch": 19.806416881998278, "grad_norm": 0.4287734925746918, "learning_rate": 0.0003031143126430591, "loss": 3.0197, "step": 73600 }, { "epoch": 19.819875107665805, "grad_norm": 0.4228683114051819, "learning_rate": 0.0003029123468426013, "loss": 3.0321, "step": 73650 }, { "epoch": 19.833333333333332, "grad_norm": 0.41015538573265076, "learning_rate": 0.0003027103810421435, "loss": 3.0173, "step": 73700 }, { "epoch": 19.846791559000863, "grad_norm": 0.3986530005931854, "learning_rate": 0.0003025084152416857, "loss": 3.0236, "step": 73750 }, { "epoch": 19.86024978466839, "grad_norm": 0.42268338799476624, "learning_rate": 0.0003023064494412279, "loss": 3.0255, "step": 73800 }, { "epoch": 19.873708010335918, "grad_norm": 0.4277915954589844, "learning_rate": 0.00030210448364077015, "loss": 3.0403, "step": 73850 }, { "epoch": 19.887166236003445, "grad_norm": 0.4032815992832184, "learning_rate": 0.00030190251784031235, "loss": 3.035, "step": 73900 }, { "epoch": 19.900624461670972, "grad_norm": 0.40952691435813904, "learning_rate": 0.00030170055203985454, "loss": 3.0269, "step": 73950 }, { "epoch": 19.9140826873385, "grad_norm": 0.42225638031959534, "learning_rate": 0.00030149858623939674, "loss": 3.0345, "step": 74000 }, { "epoch": 19.9140826873385, "eval_accuracy": 0.39611828687376494, "eval_loss": 3.2770566940307617, "eval_runtime": 84.1856, "eval_samples_per_second": 213.956, "eval_steps_per_second": 13.375, "step": 74000 }, { "epoch": 19.92754091300603, "grad_norm": 0.4844612181186676, "learning_rate": 0.00030129662043893894, "loss": 3.0382, "step": 74050 }, { "epoch": 19.940999138673558, "grad_norm": 0.46322786808013916, "learning_rate": 0.0003010946546384812, "loss": 3.0347, "step": 74100 }, { "epoch": 19.954457364341085, "grad_norm": 0.38965100049972534, "learning_rate": 0.0003008926888380234, "loss": 3.0335, "step": 74150 }, { "epoch": 19.967915590008612, "grad_norm": 0.40263253450393677, "learning_rate": 0.0003006907230375656, "loss": 3.0308, "step": 74200 }, { "epoch": 19.98137381567614, "grad_norm": 0.42235392332077026, "learning_rate": 0.0003004887572371078, "loss": 3.0277, "step": 74250 }, { "epoch": 19.99483204134367, "grad_norm": 0.4125952422618866, "learning_rate": 0.0003002867914366501, "loss": 3.0265, "step": 74300 }, { "epoch": 20.00807493540052, "grad_norm": 0.4285377860069275, "learning_rate": 0.0003000848256361923, "loss": 2.9725, "step": 74350 }, { "epoch": 20.021533161068046, "grad_norm": 0.42794254422187805, "learning_rate": 0.00029988285983573447, "loss": 2.9399, "step": 74400 }, { "epoch": 20.034991386735573, "grad_norm": 0.42025497555732727, "learning_rate": 0.00029968089403527667, "loss": 2.9645, "step": 74450 }, { "epoch": 20.0484496124031, "grad_norm": 0.4233931303024292, "learning_rate": 0.0002994789282348189, "loss": 2.9521, "step": 74500 }, { "epoch": 20.061907838070628, "grad_norm": 0.4307868480682373, "learning_rate": 0.0002992769624343611, "loss": 2.9538, "step": 74550 }, { "epoch": 20.075366063738155, "grad_norm": 0.40296119451522827, "learning_rate": 0.0002990749966339033, "loss": 2.9526, "step": 74600 }, { "epoch": 20.088824289405686, "grad_norm": 0.4423048794269562, "learning_rate": 0.0002988730308334455, "loss": 2.9594, "step": 74650 }, { "epoch": 20.102282515073213, "grad_norm": 0.40786945819854736, "learning_rate": 0.0002986710650329877, "loss": 2.9683, "step": 74700 }, { "epoch": 20.11574074074074, "grad_norm": 0.4542340040206909, "learning_rate": 0.00029846909923252995, "loss": 2.9531, "step": 74750 }, { "epoch": 20.129198966408268, "grad_norm": 0.39865925908088684, "learning_rate": 0.00029826713343207215, "loss": 2.9618, "step": 74800 }, { "epoch": 20.142657192075795, "grad_norm": 0.4336836040019989, "learning_rate": 0.00029806516763161435, "loss": 2.9662, "step": 74850 }, { "epoch": 20.156115417743326, "grad_norm": 0.434036523103714, "learning_rate": 0.00029786320183115654, "loss": 2.9768, "step": 74900 }, { "epoch": 20.169573643410853, "grad_norm": 0.4439670145511627, "learning_rate": 0.0002976612360306988, "loss": 2.9656, "step": 74950 }, { "epoch": 20.18303186907838, "grad_norm": 0.4146876633167267, "learning_rate": 0.000297459270230241, "loss": 2.9698, "step": 75000 }, { "epoch": 20.18303186907838, "eval_accuracy": 0.39523020747638127, "eval_loss": 3.2923033237457275, "eval_runtime": 84.1703, "eval_samples_per_second": 213.995, "eval_steps_per_second": 13.378, "step": 75000 }, { "epoch": 20.196490094745908, "grad_norm": 0.40108510851860046, "learning_rate": 0.0002972573044297832, "loss": 2.9759, "step": 75050 }, { "epoch": 20.209948320413435, "grad_norm": 0.4314274191856384, "learning_rate": 0.00029705533862932543, "loss": 2.9713, "step": 75100 }, { "epoch": 20.223406546080966, "grad_norm": 0.42117205262184143, "learning_rate": 0.00029685337282886763, "loss": 2.9687, "step": 75150 }, { "epoch": 20.236864771748493, "grad_norm": 0.41496866941452026, "learning_rate": 0.0002966514070284098, "loss": 2.9855, "step": 75200 }, { "epoch": 20.25032299741602, "grad_norm": 0.4257923364639282, "learning_rate": 0.000296449441227952, "loss": 2.9734, "step": 75250 }, { "epoch": 20.263781223083548, "grad_norm": 0.44260990619659424, "learning_rate": 0.0002962474754274942, "loss": 2.9859, "step": 75300 }, { "epoch": 20.277239448751075, "grad_norm": 0.4316161274909973, "learning_rate": 0.00029604550962703647, "loss": 2.9723, "step": 75350 }, { "epoch": 20.290697674418606, "grad_norm": 0.4551236927509308, "learning_rate": 0.00029584354382657867, "loss": 2.9815, "step": 75400 }, { "epoch": 20.304155900086133, "grad_norm": 0.4505977928638458, "learning_rate": 0.0002956415780261209, "loss": 2.9736, "step": 75450 }, { "epoch": 20.31761412575366, "grad_norm": 0.417833536863327, "learning_rate": 0.0002954396122256631, "loss": 2.9779, "step": 75500 }, { "epoch": 20.331072351421188, "grad_norm": 0.4278238117694855, "learning_rate": 0.0002952376464252053, "loss": 2.9783, "step": 75550 }, { "epoch": 20.344530577088715, "grad_norm": 0.4067491292953491, "learning_rate": 0.0002950356806247475, "loss": 2.9846, "step": 75600 }, { "epoch": 20.357988802756246, "grad_norm": 0.46258091926574707, "learning_rate": 0.0002948337148242897, "loss": 2.9934, "step": 75650 }, { "epoch": 20.371447028423773, "grad_norm": 0.46640563011169434, "learning_rate": 0.00029463174902383195, "loss": 2.9984, "step": 75700 }, { "epoch": 20.3849052540913, "grad_norm": 0.4281347393989563, "learning_rate": 0.00029442978322337415, "loss": 2.9832, "step": 75750 }, { "epoch": 20.398363479758828, "grad_norm": 0.40202075242996216, "learning_rate": 0.00029422781742291634, "loss": 2.9862, "step": 75800 }, { "epoch": 20.411821705426355, "grad_norm": 0.43491074442863464, "learning_rate": 0.0002940258516224586, "loss": 3.0028, "step": 75850 }, { "epoch": 20.425279931093886, "grad_norm": 0.4118116796016693, "learning_rate": 0.0002938238858220008, "loss": 2.9898, "step": 75900 }, { "epoch": 20.438738156761413, "grad_norm": 0.4505225718021393, "learning_rate": 0.000293621920021543, "loss": 2.9878, "step": 75950 }, { "epoch": 20.45219638242894, "grad_norm": 0.4443102777004242, "learning_rate": 0.0002934199542210852, "loss": 3.0003, "step": 76000 }, { "epoch": 20.45219638242894, "eval_accuracy": 0.39568022275212916, "eval_loss": 3.2855124473571777, "eval_runtime": 84.2204, "eval_samples_per_second": 213.867, "eval_steps_per_second": 13.37, "step": 76000 }, { "epoch": 20.465654608096468, "grad_norm": 0.4270143210887909, "learning_rate": 0.00029321798842062743, "loss": 3.0053, "step": 76050 }, { "epoch": 20.479112833763995, "grad_norm": 0.44338861107826233, "learning_rate": 0.00029301602262016963, "loss": 3.0084, "step": 76100 }, { "epoch": 20.492571059431526, "grad_norm": 0.4215359091758728, "learning_rate": 0.0002928140568197118, "loss": 3.0041, "step": 76150 }, { "epoch": 20.506029285099054, "grad_norm": 0.4466627240180969, "learning_rate": 0.000292612091019254, "loss": 3.001, "step": 76200 }, { "epoch": 20.51948751076658, "grad_norm": 0.4711362421512604, "learning_rate": 0.0002924101252187962, "loss": 2.9919, "step": 76250 }, { "epoch": 20.532945736434108, "grad_norm": 0.4064037501811981, "learning_rate": 0.00029220815941833847, "loss": 2.9949, "step": 76300 }, { "epoch": 20.546403962101635, "grad_norm": 0.45425111055374146, "learning_rate": 0.0002920061936178807, "loss": 3.0046, "step": 76350 }, { "epoch": 20.559862187769163, "grad_norm": 0.4240078032016754, "learning_rate": 0.0002918042278174229, "loss": 3.0034, "step": 76400 }, { "epoch": 20.573320413436694, "grad_norm": 0.41471272706985474, "learning_rate": 0.0002916022620169651, "loss": 3.0037, "step": 76450 }, { "epoch": 20.58677863910422, "grad_norm": 0.44904863834381104, "learning_rate": 0.0002914002962165073, "loss": 3.0002, "step": 76500 }, { "epoch": 20.600236864771748, "grad_norm": 0.40947166085243225, "learning_rate": 0.0002911983304160495, "loss": 3.0096, "step": 76550 }, { "epoch": 20.613695090439276, "grad_norm": 0.45504602789878845, "learning_rate": 0.00029099636461559175, "loss": 3.0149, "step": 76600 }, { "epoch": 20.627153316106803, "grad_norm": 0.44130855798721313, "learning_rate": 0.00029079439881513395, "loss": 3.0, "step": 76650 }, { "epoch": 20.640611541774334, "grad_norm": 0.41574594378471375, "learning_rate": 0.00029059243301467614, "loss": 3.0135, "step": 76700 }, { "epoch": 20.65406976744186, "grad_norm": 0.4264014661312103, "learning_rate": 0.0002903904672142184, "loss": 3.0095, "step": 76750 }, { "epoch": 20.66752799310939, "grad_norm": 0.3829843997955322, "learning_rate": 0.0002901885014137606, "loss": 3.0188, "step": 76800 }, { "epoch": 20.680986218776916, "grad_norm": 0.41488978266716003, "learning_rate": 0.0002899865356133028, "loss": 3.0143, "step": 76850 }, { "epoch": 20.694444444444443, "grad_norm": 0.44957494735717773, "learning_rate": 0.000289784569812845, "loss": 3.0058, "step": 76900 }, { "epoch": 20.707902670111974, "grad_norm": 0.447897732257843, "learning_rate": 0.00028958260401238723, "loss": 3.0046, "step": 76950 }, { "epoch": 20.7213608957795, "grad_norm": 0.4053957164287567, "learning_rate": 0.00028938063821192943, "loss": 3.0066, "step": 77000 }, { "epoch": 20.7213608957795, "eval_accuracy": 0.3959690060942194, "eval_loss": 3.284266710281372, "eval_runtime": 84.0458, "eval_samples_per_second": 214.312, "eval_steps_per_second": 13.397, "step": 77000 }, { "epoch": 20.73481912144703, "grad_norm": 0.42543306946754456, "learning_rate": 0.0002891786724114716, "loss": 3.0144, "step": 77050 }, { "epoch": 20.748277347114556, "grad_norm": 0.43379610776901245, "learning_rate": 0.0002889767066110138, "loss": 3.0165, "step": 77100 }, { "epoch": 20.761735572782083, "grad_norm": 0.4360213279724121, "learning_rate": 0.000288774740810556, "loss": 3.0053, "step": 77150 }, { "epoch": 20.775193798449614, "grad_norm": 0.4409184455871582, "learning_rate": 0.00028857277501009827, "loss": 3.0098, "step": 77200 }, { "epoch": 20.78865202411714, "grad_norm": 0.4014228582382202, "learning_rate": 0.00028837080920964046, "loss": 3.0204, "step": 77250 }, { "epoch": 20.80211024978467, "grad_norm": 0.44066837430000305, "learning_rate": 0.0002881688434091827, "loss": 3.0154, "step": 77300 }, { "epoch": 20.815568475452196, "grad_norm": 0.4298141598701477, "learning_rate": 0.0002879668776087249, "loss": 3.0089, "step": 77350 }, { "epoch": 20.829026701119723, "grad_norm": 0.45948871970176697, "learning_rate": 0.0002877649118082671, "loss": 3.0072, "step": 77400 }, { "epoch": 20.842484926787254, "grad_norm": 0.409786194562912, "learning_rate": 0.0002875629460078093, "loss": 3.0263, "step": 77450 }, { "epoch": 20.85594315245478, "grad_norm": 0.4201638996601105, "learning_rate": 0.0002873609802073515, "loss": 3.0099, "step": 77500 }, { "epoch": 20.86940137812231, "grad_norm": 0.4363999664783478, "learning_rate": 0.00028715901440689375, "loss": 3.0135, "step": 77550 }, { "epoch": 20.882859603789836, "grad_norm": 0.43155109882354736, "learning_rate": 0.00028695704860643594, "loss": 3.0108, "step": 77600 }, { "epoch": 20.896317829457363, "grad_norm": 0.4437941014766693, "learning_rate": 0.0002867550828059782, "loss": 3.0232, "step": 77650 }, { "epoch": 20.909776055124894, "grad_norm": 0.41121140122413635, "learning_rate": 0.0002865531170055204, "loss": 3.0258, "step": 77700 }, { "epoch": 20.92323428079242, "grad_norm": 0.42540350556373596, "learning_rate": 0.0002863511512050626, "loss": 3.0168, "step": 77750 }, { "epoch": 20.93669250645995, "grad_norm": 0.41012293100357056, "learning_rate": 0.0002861491854046048, "loss": 3.0203, "step": 77800 }, { "epoch": 20.950150732127476, "grad_norm": 0.4128982126712799, "learning_rate": 0.000285947219604147, "loss": 3.0204, "step": 77850 }, { "epoch": 20.963608957795003, "grad_norm": 0.452108234167099, "learning_rate": 0.00028574525380368923, "loss": 3.0198, "step": 77900 }, { "epoch": 20.977067183462534, "grad_norm": 0.4156314432621002, "learning_rate": 0.0002855432880032314, "loss": 3.0175, "step": 77950 }, { "epoch": 20.99052540913006, "grad_norm": 0.4258723855018616, "learning_rate": 0.0002853413222027736, "loss": 3.0236, "step": 78000 }, { "epoch": 20.99052540913006, "eval_accuracy": 0.39619499155379345, "eval_loss": 3.275212287902832, "eval_runtime": 84.0307, "eval_samples_per_second": 214.35, "eval_steps_per_second": 13.4, "step": 78000 }, { "epoch": 21.00376830318691, "grad_norm": 0.4432719051837921, "learning_rate": 0.00028513935640231587, "loss": 2.9986, "step": 78050 }, { "epoch": 21.017226528854437, "grad_norm": 0.422293484210968, "learning_rate": 0.00028493739060185807, "loss": 2.9247, "step": 78100 }, { "epoch": 21.030684754521964, "grad_norm": 0.44783154129981995, "learning_rate": 0.00028473542480140026, "loss": 2.9395, "step": 78150 }, { "epoch": 21.04414298018949, "grad_norm": 0.4324184060096741, "learning_rate": 0.0002845334590009425, "loss": 2.9322, "step": 78200 }, { "epoch": 21.05760120585702, "grad_norm": 0.4230210781097412, "learning_rate": 0.0002843314932004847, "loss": 2.938, "step": 78250 }, { "epoch": 21.07105943152455, "grad_norm": 0.4204503893852234, "learning_rate": 0.0002841295274000269, "loss": 2.9474, "step": 78300 }, { "epoch": 21.084517657192077, "grad_norm": 0.413437157869339, "learning_rate": 0.0002839275615995691, "loss": 2.9412, "step": 78350 }, { "epoch": 21.097975882859604, "grad_norm": 0.44208431243896484, "learning_rate": 0.0002837255957991113, "loss": 2.9442, "step": 78400 }, { "epoch": 21.11143410852713, "grad_norm": 0.4027329087257385, "learning_rate": 0.00028352362999865355, "loss": 2.9592, "step": 78450 }, { "epoch": 21.12489233419466, "grad_norm": 0.4688771665096283, "learning_rate": 0.00028332166419819574, "loss": 2.9498, "step": 78500 }, { "epoch": 21.13835055986219, "grad_norm": 0.4691401720046997, "learning_rate": 0.000283119698397738, "loss": 2.9517, "step": 78550 }, { "epoch": 21.151808785529717, "grad_norm": 0.4198927879333496, "learning_rate": 0.0002829177325972802, "loss": 2.9617, "step": 78600 }, { "epoch": 21.165267011197244, "grad_norm": 0.4209950566291809, "learning_rate": 0.0002827157667968224, "loss": 2.9547, "step": 78650 }, { "epoch": 21.17872523686477, "grad_norm": 0.44256526231765747, "learning_rate": 0.0002825138009963646, "loss": 2.959, "step": 78700 }, { "epoch": 21.1921834625323, "grad_norm": 0.4478328227996826, "learning_rate": 0.0002823118351959068, "loss": 2.9712, "step": 78750 }, { "epoch": 21.205641688199826, "grad_norm": 0.4415542483329773, "learning_rate": 0.00028210986939544903, "loss": 2.9695, "step": 78800 }, { "epoch": 21.219099913867357, "grad_norm": 0.42408132553100586, "learning_rate": 0.0002819079035949912, "loss": 2.9576, "step": 78850 }, { "epoch": 21.232558139534884, "grad_norm": 0.4532260596752167, "learning_rate": 0.0002817059377945334, "loss": 2.9654, "step": 78900 }, { "epoch": 21.24601636520241, "grad_norm": 0.4424733519554138, "learning_rate": 0.00028150397199407567, "loss": 2.971, "step": 78950 }, { "epoch": 21.25947459086994, "grad_norm": 0.42319878935813904, "learning_rate": 0.00028130200619361787, "loss": 2.9793, "step": 79000 }, { "epoch": 21.25947459086994, "eval_accuracy": 0.3956429568806705, "eval_loss": 3.291125774383545, "eval_runtime": 84.0955, "eval_samples_per_second": 214.185, "eval_steps_per_second": 13.39, "step": 79000 }, { "epoch": 21.272932816537466, "grad_norm": 0.43081074953079224, "learning_rate": 0.00028110004039316006, "loss": 2.9673, "step": 79050 }, { "epoch": 21.286391042204997, "grad_norm": 0.4247548282146454, "learning_rate": 0.00028089807459270226, "loss": 2.9688, "step": 79100 }, { "epoch": 21.299849267872524, "grad_norm": 0.46288377046585083, "learning_rate": 0.0002806961087922445, "loss": 2.9804, "step": 79150 }, { "epoch": 21.31330749354005, "grad_norm": 0.4464956521987915, "learning_rate": 0.0002804941429917867, "loss": 2.9755, "step": 79200 }, { "epoch": 21.32676571920758, "grad_norm": 0.42706120014190674, "learning_rate": 0.0002802921771913289, "loss": 2.9889, "step": 79250 }, { "epoch": 21.340223944875106, "grad_norm": 0.4685516655445099, "learning_rate": 0.0002800902113908711, "loss": 2.9751, "step": 79300 }, { "epoch": 21.353682170542637, "grad_norm": 0.44617944955825806, "learning_rate": 0.0002798882455904133, "loss": 2.9713, "step": 79350 }, { "epoch": 21.367140396210164, "grad_norm": 0.4019935429096222, "learning_rate": 0.00027968627978995555, "loss": 2.9746, "step": 79400 }, { "epoch": 21.38059862187769, "grad_norm": 0.41742581129074097, "learning_rate": 0.00027948431398949774, "loss": 2.9702, "step": 79450 }, { "epoch": 21.39405684754522, "grad_norm": 0.44034504890441895, "learning_rate": 0.00027928234818904, "loss": 2.9848, "step": 79500 }, { "epoch": 21.407515073212746, "grad_norm": 0.44808515906333923, "learning_rate": 0.0002790803823885822, "loss": 2.9756, "step": 79550 }, { "epoch": 21.420973298880277, "grad_norm": 0.4280387759208679, "learning_rate": 0.0002788784165881244, "loss": 2.9905, "step": 79600 }, { "epoch": 21.434431524547804, "grad_norm": 0.42210879921913147, "learning_rate": 0.0002786764507876666, "loss": 2.9842, "step": 79650 }, { "epoch": 21.44788975021533, "grad_norm": 0.44455796480178833, "learning_rate": 0.0002784744849872088, "loss": 2.9805, "step": 79700 }, { "epoch": 21.46134797588286, "grad_norm": 0.45095279812812805, "learning_rate": 0.000278272519186751, "loss": 2.9814, "step": 79750 }, { "epoch": 21.474806201550386, "grad_norm": 0.4286750257015228, "learning_rate": 0.0002780705533862932, "loss": 2.9958, "step": 79800 }, { "epoch": 21.488264427217917, "grad_norm": 0.43087276816368103, "learning_rate": 0.0002778685875858355, "loss": 2.9874, "step": 79850 }, { "epoch": 21.501722652885444, "grad_norm": 0.4437640309333801, "learning_rate": 0.00027766662178537767, "loss": 3.0018, "step": 79900 }, { "epoch": 21.51518087855297, "grad_norm": 0.44071564078330994, "learning_rate": 0.00027746465598491987, "loss": 2.9924, "step": 79950 }, { "epoch": 21.5286391042205, "grad_norm": 0.41708236932754517, "learning_rate": 0.00027726269018446206, "loss": 2.9953, "step": 80000 }, { "epoch": 21.5286391042205, "eval_accuracy": 0.39576029548468017, "eval_loss": 3.2859139442443848, "eval_runtime": 84.2568, "eval_samples_per_second": 213.775, "eval_steps_per_second": 13.364, "step": 80000 }, { "epoch": 21.542097329888026, "grad_norm": 0.4567016661167145, "learning_rate": 0.00027706072438400426, "loss": 2.982, "step": 80050 }, { "epoch": 21.555555555555557, "grad_norm": 0.43089812994003296, "learning_rate": 0.0002768587585835465, "loss": 2.9855, "step": 80100 }, { "epoch": 21.569013781223084, "grad_norm": 0.414765328168869, "learning_rate": 0.0002766567927830887, "loss": 2.9885, "step": 80150 }, { "epoch": 21.58247200689061, "grad_norm": 0.429155170917511, "learning_rate": 0.0002764548269826309, "loss": 2.9986, "step": 80200 }, { "epoch": 21.59593023255814, "grad_norm": 0.46186089515686035, "learning_rate": 0.0002762528611821731, "loss": 2.988, "step": 80250 }, { "epoch": 21.609388458225666, "grad_norm": 0.44444066286087036, "learning_rate": 0.00027605089538171535, "loss": 2.9902, "step": 80300 }, { "epoch": 21.622846683893197, "grad_norm": 0.42528036236763, "learning_rate": 0.00027584892958125754, "loss": 2.9959, "step": 80350 }, { "epoch": 21.636304909560724, "grad_norm": 0.45731401443481445, "learning_rate": 0.0002756469637807998, "loss": 3.0027, "step": 80400 }, { "epoch": 21.649763135228252, "grad_norm": 0.42883455753326416, "learning_rate": 0.000275444997980342, "loss": 2.9955, "step": 80450 }, { "epoch": 21.66322136089578, "grad_norm": 0.40275973081588745, "learning_rate": 0.0002752430321798842, "loss": 2.9981, "step": 80500 }, { "epoch": 21.676679586563306, "grad_norm": 0.4312342405319214, "learning_rate": 0.0002750410663794264, "loss": 3.0058, "step": 80550 }, { "epoch": 21.690137812230837, "grad_norm": 0.43648454546928406, "learning_rate": 0.0002748391005789686, "loss": 3.0002, "step": 80600 }, { "epoch": 21.703596037898365, "grad_norm": 0.4063485264778137, "learning_rate": 0.00027463713477851083, "loss": 3.0025, "step": 80650 }, { "epoch": 21.717054263565892, "grad_norm": 0.44078052043914795, "learning_rate": 0.000274435168978053, "loss": 2.9997, "step": 80700 }, { "epoch": 21.73051248923342, "grad_norm": 0.4490727186203003, "learning_rate": 0.0002742332031775953, "loss": 2.9849, "step": 80750 }, { "epoch": 21.743970714900946, "grad_norm": 0.4792262613773346, "learning_rate": 0.00027403123737713747, "loss": 3.015, "step": 80800 }, { "epoch": 21.757428940568474, "grad_norm": 0.4362357258796692, "learning_rate": 0.00027382927157667967, "loss": 3.0016, "step": 80850 }, { "epoch": 21.770887166236005, "grad_norm": 0.4123806059360504, "learning_rate": 0.00027362730577622186, "loss": 3.0035, "step": 80900 }, { "epoch": 21.784345391903532, "grad_norm": 0.41003212332725525, "learning_rate": 0.00027342533997576406, "loss": 2.9887, "step": 80950 }, { "epoch": 21.79780361757106, "grad_norm": 0.42923009395599365, "learning_rate": 0.0002732233741753063, "loss": 3.0028, "step": 81000 }, { "epoch": 21.79780361757106, "eval_accuracy": 0.39638946942525377, "eval_loss": 3.2749459743499756, "eval_runtime": 83.8283, "eval_samples_per_second": 214.868, "eval_steps_per_second": 13.432, "step": 81000 }, { "epoch": 21.811261843238587, "grad_norm": 0.44958174228668213, "learning_rate": 0.0002730214083748485, "loss": 3.0057, "step": 81050 }, { "epoch": 21.824720068906114, "grad_norm": 0.437704473733902, "learning_rate": 0.0002728194425743907, "loss": 3.0011, "step": 81100 }, { "epoch": 21.838178294573645, "grad_norm": 0.4207727313041687, "learning_rate": 0.0002726174767739329, "loss": 2.9998, "step": 81150 }, { "epoch": 21.851636520241172, "grad_norm": 0.42127111554145813, "learning_rate": 0.00027241551097347515, "loss": 3.0176, "step": 81200 }, { "epoch": 21.8650947459087, "grad_norm": 0.4241941273212433, "learning_rate": 0.00027221354517301734, "loss": 2.9997, "step": 81250 }, { "epoch": 21.878552971576227, "grad_norm": 0.4211195111274719, "learning_rate": 0.00027201157937255954, "loss": 3.0019, "step": 81300 }, { "epoch": 21.892011197243754, "grad_norm": 0.4051482081413269, "learning_rate": 0.0002718096135721018, "loss": 3.0036, "step": 81350 }, { "epoch": 21.905469422911285, "grad_norm": 0.44209906458854675, "learning_rate": 0.000271607647771644, "loss": 3.0129, "step": 81400 }, { "epoch": 21.918927648578812, "grad_norm": 0.40235477685928345, "learning_rate": 0.0002714056819711862, "loss": 3.017, "step": 81450 }, { "epoch": 21.93238587424634, "grad_norm": 0.42072221636772156, "learning_rate": 0.0002712037161707284, "loss": 3.0024, "step": 81500 }, { "epoch": 21.945844099913867, "grad_norm": 0.4275611340999603, "learning_rate": 0.0002710017503702706, "loss": 3.0145, "step": 81550 }, { "epoch": 21.959302325581394, "grad_norm": 0.43438443541526794, "learning_rate": 0.0002707997845698128, "loss": 3.0087, "step": 81600 }, { "epoch": 21.972760551248925, "grad_norm": 0.44001317024230957, "learning_rate": 0.000270597818769355, "loss": 3.0132, "step": 81650 }, { "epoch": 21.986218776916452, "grad_norm": 0.41912901401519775, "learning_rate": 0.00027039585296889727, "loss": 3.0129, "step": 81700 }, { "epoch": 21.99967700258398, "grad_norm": 0.42187735438346863, "learning_rate": 0.00027019388716843947, "loss": 3.0144, "step": 81750 }, { "epoch": 22.012919896640827, "grad_norm": 0.4827214479446411, "learning_rate": 0.00026999192136798166, "loss": 2.9178, "step": 81800 }, { "epoch": 22.026378122308355, "grad_norm": 0.43587005138397217, "learning_rate": 0.00026978995556752386, "loss": 2.9323, "step": 81850 }, { "epoch": 22.039836347975882, "grad_norm": 0.44450515508651733, "learning_rate": 0.00026958798976706606, "loss": 2.9233, "step": 81900 }, { "epoch": 22.05329457364341, "grad_norm": 0.43394994735717773, "learning_rate": 0.0002693860239666083, "loss": 2.9288, "step": 81950 }, { "epoch": 22.06675279931094, "grad_norm": 0.44612663984298706, "learning_rate": 0.0002691840581661505, "loss": 2.9377, "step": 82000 }, { "epoch": 22.06675279931094, "eval_accuracy": 0.396058422456349, "eval_loss": 3.289057493209839, "eval_runtime": 83.9224, "eval_samples_per_second": 214.627, "eval_steps_per_second": 13.417, "step": 82000 }, { "epoch": 22.080211024978468, "grad_norm": 0.4203513562679291, "learning_rate": 0.0002689820923656927, "loss": 2.9275, "step": 82050 }, { "epoch": 22.093669250645995, "grad_norm": 0.45312684774398804, "learning_rate": 0.00026878012656523495, "loss": 2.9374, "step": 82100 }, { "epoch": 22.107127476313522, "grad_norm": 0.47145548462867737, "learning_rate": 0.00026857816076477714, "loss": 2.9412, "step": 82150 }, { "epoch": 22.12058570198105, "grad_norm": 0.43565475940704346, "learning_rate": 0.00026837619496431934, "loss": 2.9443, "step": 82200 }, { "epoch": 22.13404392764858, "grad_norm": 0.42642873525619507, "learning_rate": 0.0002681742291638616, "loss": 2.9447, "step": 82250 }, { "epoch": 22.147502153316108, "grad_norm": 0.47161009907722473, "learning_rate": 0.0002679722633634038, "loss": 2.9448, "step": 82300 }, { "epoch": 22.160960378983635, "grad_norm": 0.47496211528778076, "learning_rate": 0.000267770297562946, "loss": 2.9489, "step": 82350 }, { "epoch": 22.174418604651162, "grad_norm": 0.4288054406642914, "learning_rate": 0.0002675683317624882, "loss": 2.9493, "step": 82400 }, { "epoch": 22.18787683031869, "grad_norm": 0.4706723093986511, "learning_rate": 0.0002673663659620304, "loss": 2.9505, "step": 82450 }, { "epoch": 22.20133505598622, "grad_norm": 0.45602041482925415, "learning_rate": 0.0002671644001615726, "loss": 2.9579, "step": 82500 }, { "epoch": 22.214793281653748, "grad_norm": 0.4584351181983948, "learning_rate": 0.0002669624343611148, "loss": 2.9577, "step": 82550 }, { "epoch": 22.228251507321275, "grad_norm": 0.42825227975845337, "learning_rate": 0.00026676046856065707, "loss": 2.9568, "step": 82600 }, { "epoch": 22.241709732988802, "grad_norm": 0.46178528666496277, "learning_rate": 0.00026655850276019927, "loss": 2.9482, "step": 82650 }, { "epoch": 22.25516795865633, "grad_norm": 0.41845497488975525, "learning_rate": 0.00026635653695974146, "loss": 2.9635, "step": 82700 }, { "epoch": 22.26862618432386, "grad_norm": 0.4651488661766052, "learning_rate": 0.00026615457115928366, "loss": 2.9652, "step": 82750 }, { "epoch": 22.282084409991388, "grad_norm": 0.43473976850509644, "learning_rate": 0.00026595260535882586, "loss": 2.9566, "step": 82800 }, { "epoch": 22.295542635658915, "grad_norm": 0.4451102316379547, "learning_rate": 0.0002657506395583681, "loss": 2.9629, "step": 82850 }, { "epoch": 22.309000861326442, "grad_norm": 0.4579492509365082, "learning_rate": 0.0002655486737579103, "loss": 2.9704, "step": 82900 }, { "epoch": 22.32245908699397, "grad_norm": 0.4861108362674713, "learning_rate": 0.00026534670795745255, "loss": 2.9632, "step": 82950 }, { "epoch": 22.3359173126615, "grad_norm": 0.4306095838546753, "learning_rate": 0.00026514474215699475, "loss": 2.9726, "step": 83000 }, { "epoch": 22.3359173126615, "eval_accuracy": 0.3960007309760442, "eval_loss": 3.2885091304779053, "eval_runtime": 83.9431, "eval_samples_per_second": 214.574, "eval_steps_per_second": 13.414, "step": 83000 }, { "epoch": 22.349375538329028, "grad_norm": 0.46351316571235657, "learning_rate": 0.00026494277635653694, "loss": 2.9677, "step": 83050 }, { "epoch": 22.362833763996555, "grad_norm": 0.46026599407196045, "learning_rate": 0.00026474081055607914, "loss": 2.9621, "step": 83100 }, { "epoch": 22.376291989664082, "grad_norm": 0.4538576900959015, "learning_rate": 0.00026453884475562134, "loss": 2.963, "step": 83150 }, { "epoch": 22.38975021533161, "grad_norm": 0.42874839901924133, "learning_rate": 0.0002643368789551636, "loss": 2.9723, "step": 83200 }, { "epoch": 22.403208440999137, "grad_norm": 0.4932001829147339, "learning_rate": 0.0002641349131547058, "loss": 2.9793, "step": 83250 }, { "epoch": 22.416666666666668, "grad_norm": 0.4467307925224304, "learning_rate": 0.000263932947354248, "loss": 2.9627, "step": 83300 }, { "epoch": 22.430124892334195, "grad_norm": 0.4314446747303009, "learning_rate": 0.0002637309815537902, "loss": 2.9787, "step": 83350 }, { "epoch": 22.443583118001722, "grad_norm": 0.43153485655784607, "learning_rate": 0.0002635290157533324, "loss": 2.9745, "step": 83400 }, { "epoch": 22.45704134366925, "grad_norm": 0.4590122401714325, "learning_rate": 0.0002633270499528746, "loss": 2.9702, "step": 83450 }, { "epoch": 22.470499569336777, "grad_norm": 0.43995824456214905, "learning_rate": 0.0002631250841524168, "loss": 2.9735, "step": 83500 }, { "epoch": 22.483957795004308, "grad_norm": 0.45518141984939575, "learning_rate": 0.00026292311835195907, "loss": 2.9699, "step": 83550 }, { "epoch": 22.497416020671835, "grad_norm": 0.459585577249527, "learning_rate": 0.00026272115255150126, "loss": 2.9735, "step": 83600 }, { "epoch": 22.510874246339363, "grad_norm": 0.4053054749965668, "learning_rate": 0.00026251918675104346, "loss": 2.9722, "step": 83650 }, { "epoch": 22.52433247200689, "grad_norm": 0.42346060276031494, "learning_rate": 0.00026231722095058566, "loss": 2.9775, "step": 83700 }, { "epoch": 22.537790697674417, "grad_norm": 0.4662436246871948, "learning_rate": 0.00026211525515012785, "loss": 2.9793, "step": 83750 }, { "epoch": 22.551248923341948, "grad_norm": 0.4733358919620514, "learning_rate": 0.0002619132893496701, "loss": 2.9769, "step": 83800 }, { "epoch": 22.564707149009475, "grad_norm": 0.4621383547782898, "learning_rate": 0.0002617113235492123, "loss": 2.9854, "step": 83850 }, { "epoch": 22.578165374677003, "grad_norm": 0.4373529851436615, "learning_rate": 0.00026150935774875455, "loss": 2.9861, "step": 83900 }, { "epoch": 22.59162360034453, "grad_norm": 0.43990325927734375, "learning_rate": 0.00026130739194829675, "loss": 2.9911, "step": 83950 }, { "epoch": 22.605081826012057, "grad_norm": 0.4210509657859802, "learning_rate": 0.00026110542614783894, "loss": 2.9868, "step": 84000 }, { "epoch": 22.605081826012057, "eval_accuracy": 0.3961254575662322, "eval_loss": 3.286008596420288, "eval_runtime": 83.9197, "eval_samples_per_second": 214.634, "eval_steps_per_second": 13.418, "step": 84000 }, { "epoch": 22.618540051679588, "grad_norm": 0.4237697422504425, "learning_rate": 0.00026090346034738114, "loss": 2.9899, "step": 84050 }, { "epoch": 22.631998277347115, "grad_norm": 0.4568379819393158, "learning_rate": 0.0002607014945469234, "loss": 2.9839, "step": 84100 }, { "epoch": 22.645456503014643, "grad_norm": 0.4833359718322754, "learning_rate": 0.0002604995287464656, "loss": 2.9809, "step": 84150 }, { "epoch": 22.65891472868217, "grad_norm": 0.4397493302822113, "learning_rate": 0.0002602975629460078, "loss": 2.9883, "step": 84200 }, { "epoch": 22.672372954349697, "grad_norm": 0.4514474868774414, "learning_rate": 0.00026009559714555, "loss": 2.9968, "step": 84250 }, { "epoch": 22.685831180017228, "grad_norm": 0.4431244730949402, "learning_rate": 0.0002598936313450922, "loss": 2.9908, "step": 84300 }, { "epoch": 22.699289405684755, "grad_norm": 0.4466392695903778, "learning_rate": 0.0002596916655446344, "loss": 2.981, "step": 84350 }, { "epoch": 22.712747631352283, "grad_norm": 0.47167009115219116, "learning_rate": 0.0002594896997441766, "loss": 2.9881, "step": 84400 }, { "epoch": 22.72620585701981, "grad_norm": 0.4247466027736664, "learning_rate": 0.00025928773394371887, "loss": 2.9811, "step": 84450 }, { "epoch": 22.739664082687337, "grad_norm": 0.4402214586734772, "learning_rate": 0.00025908576814326107, "loss": 2.9915, "step": 84500 }, { "epoch": 22.753122308354868, "grad_norm": 0.4316474199295044, "learning_rate": 0.00025888380234280326, "loss": 2.9842, "step": 84550 }, { "epoch": 22.766580534022395, "grad_norm": 0.4481562674045563, "learning_rate": 0.00025868183654234546, "loss": 2.9933, "step": 84600 }, { "epoch": 22.780038759689923, "grad_norm": 0.4497462213039398, "learning_rate": 0.00025847987074188765, "loss": 3.0006, "step": 84650 }, { "epoch": 22.79349698535745, "grad_norm": 0.4655749499797821, "learning_rate": 0.0002582779049414299, "loss": 2.997, "step": 84700 }, { "epoch": 22.806955211024977, "grad_norm": 0.4840506315231323, "learning_rate": 0.0002580759391409721, "loss": 3.0028, "step": 84750 }, { "epoch": 22.820413436692505, "grad_norm": 0.4458298981189728, "learning_rate": 0.00025787397334051435, "loss": 2.9878, "step": 84800 }, { "epoch": 22.833871662360036, "grad_norm": 0.4646628797054291, "learning_rate": 0.00025767200754005655, "loss": 3.0075, "step": 84850 }, { "epoch": 22.847329888027563, "grad_norm": 0.49665722250938416, "learning_rate": 0.00025747004173959874, "loss": 3.0067, "step": 84900 }, { "epoch": 22.86078811369509, "grad_norm": 0.43514347076416016, "learning_rate": 0.00025726807593914094, "loss": 2.9904, "step": 84950 }, { "epoch": 22.874246339362617, "grad_norm": 0.4649242162704468, "learning_rate": 0.00025706611013868313, "loss": 3.0054, "step": 85000 }, { "epoch": 22.874246339362617, "eval_accuracy": 0.39701668772242726, "eval_loss": 3.2736573219299316, "eval_runtime": 83.9771, "eval_samples_per_second": 214.487, "eval_steps_per_second": 13.408, "step": 85000 }, { "epoch": 22.887704565030145, "grad_norm": 0.4361061751842499, "learning_rate": 0.0002568641443382254, "loss": 2.9948, "step": 85050 }, { "epoch": 22.901162790697676, "grad_norm": 0.41159138083457947, "learning_rate": 0.0002566621785377676, "loss": 2.9974, "step": 85100 }, { "epoch": 22.914621016365203, "grad_norm": 0.4136468768119812, "learning_rate": 0.0002564602127373098, "loss": 2.9977, "step": 85150 }, { "epoch": 22.92807924203273, "grad_norm": 0.45003828406333923, "learning_rate": 0.00025625824693685203, "loss": 2.9851, "step": 85200 }, { "epoch": 22.941537467700257, "grad_norm": 0.42788249254226685, "learning_rate": 0.0002560562811363942, "loss": 2.9985, "step": 85250 }, { "epoch": 22.954995693367785, "grad_norm": 0.43006718158721924, "learning_rate": 0.0002558543153359364, "loss": 3.0012, "step": 85300 }, { "epoch": 22.968453919035316, "grad_norm": 0.42816489934921265, "learning_rate": 0.0002556523495354786, "loss": 2.9948, "step": 85350 }, { "epoch": 22.981912144702843, "grad_norm": 0.42106497287750244, "learning_rate": 0.00025545038373502087, "loss": 2.9958, "step": 85400 }, { "epoch": 22.99537037037037, "grad_norm": 0.4418798089027405, "learning_rate": 0.00025524841793456306, "loss": 3.0094, "step": 85450 }, { "epoch": 23.00861326442722, "grad_norm": 0.4254322052001953, "learning_rate": 0.00025504645213410526, "loss": 2.9523, "step": 85500 }, { "epoch": 23.022071490094746, "grad_norm": 0.4377330541610718, "learning_rate": 0.00025484448633364745, "loss": 2.9239, "step": 85550 }, { "epoch": 23.035529715762273, "grad_norm": 0.4505152404308319, "learning_rate": 0.00025464252053318965, "loss": 2.9205, "step": 85600 }, { "epoch": 23.0489879414298, "grad_norm": 0.45015689730644226, "learning_rate": 0.0002544405547327319, "loss": 2.9157, "step": 85650 }, { "epoch": 23.06244616709733, "grad_norm": 0.4604092240333557, "learning_rate": 0.0002542385889322741, "loss": 2.919, "step": 85700 }, { "epoch": 23.07590439276486, "grad_norm": 0.4214881360530853, "learning_rate": 0.00025403662313181635, "loss": 2.9218, "step": 85750 }, { "epoch": 23.089362618432386, "grad_norm": 0.46785667538642883, "learning_rate": 0.00025383465733135854, "loss": 2.9308, "step": 85800 }, { "epoch": 23.102820844099913, "grad_norm": 0.45559147000312805, "learning_rate": 0.00025363269153090074, "loss": 2.9302, "step": 85850 }, { "epoch": 23.11627906976744, "grad_norm": 0.4463777244091034, "learning_rate": 0.00025343072573044294, "loss": 2.929, "step": 85900 }, { "epoch": 23.12973729543497, "grad_norm": 0.463544636964798, "learning_rate": 0.00025322875992998513, "loss": 2.9306, "step": 85950 }, { "epoch": 23.1431955211025, "grad_norm": 0.4276140630245209, "learning_rate": 0.0002530267941295274, "loss": 2.9233, "step": 86000 }, { "epoch": 23.1431955211025, "eval_accuracy": 0.39601811447293456, "eval_loss": 3.291121482849121, "eval_runtime": 84.0617, "eval_samples_per_second": 214.271, "eval_steps_per_second": 13.395, "step": 86000 }, { "epoch": 23.156653746770026, "grad_norm": 0.4453248083591461, "learning_rate": 0.0002528248283290696, "loss": 2.9369, "step": 86050 }, { "epoch": 23.170111972437553, "grad_norm": 0.4330044984817505, "learning_rate": 0.00025262286252861183, "loss": 2.9344, "step": 86100 }, { "epoch": 23.18357019810508, "grad_norm": 0.43674394488334656, "learning_rate": 0.000252420896728154, "loss": 2.9474, "step": 86150 }, { "epoch": 23.19702842377261, "grad_norm": 0.48593881726264954, "learning_rate": 0.0002522189309276962, "loss": 2.9449, "step": 86200 }, { "epoch": 23.21048664944014, "grad_norm": 0.4298238754272461, "learning_rate": 0.0002520169651272384, "loss": 2.9502, "step": 86250 }, { "epoch": 23.223944875107666, "grad_norm": 0.5104933381080627, "learning_rate": 0.00025181499932678067, "loss": 2.943, "step": 86300 }, { "epoch": 23.237403100775193, "grad_norm": 0.4299788773059845, "learning_rate": 0.00025161303352632286, "loss": 2.9565, "step": 86350 }, { "epoch": 23.25086132644272, "grad_norm": 0.4533037841320038, "learning_rate": 0.00025141106772586506, "loss": 2.9529, "step": 86400 }, { "epoch": 23.26431955211025, "grad_norm": 0.5035787224769592, "learning_rate": 0.00025120910192540726, "loss": 2.9517, "step": 86450 }, { "epoch": 23.27777777777778, "grad_norm": 0.46547257900238037, "learning_rate": 0.00025100713612494945, "loss": 2.9443, "step": 86500 }, { "epoch": 23.291236003445306, "grad_norm": 0.42908090353012085, "learning_rate": 0.0002508051703244917, "loss": 2.9455, "step": 86550 }, { "epoch": 23.304694229112833, "grad_norm": 0.47686538100242615, "learning_rate": 0.0002506032045240339, "loss": 2.9493, "step": 86600 }, { "epoch": 23.31815245478036, "grad_norm": 0.44427627325057983, "learning_rate": 0.00025040123872357615, "loss": 2.9707, "step": 86650 }, { "epoch": 23.33161068044789, "grad_norm": 0.4498743712902069, "learning_rate": 0.00025019927292311834, "loss": 2.958, "step": 86700 }, { "epoch": 23.34506890611542, "grad_norm": 0.45148012042045593, "learning_rate": 0.00024999730712266054, "loss": 2.9652, "step": 86750 }, { "epoch": 23.358527131782946, "grad_norm": 0.4607517421245575, "learning_rate": 0.00024979534132220274, "loss": 2.9589, "step": 86800 }, { "epoch": 23.371985357450473, "grad_norm": 0.42376628518104553, "learning_rate": 0.00024959337552174493, "loss": 2.9705, "step": 86850 }, { "epoch": 23.385443583118, "grad_norm": 0.47415992617607117, "learning_rate": 0.0002493914097212872, "loss": 2.9589, "step": 86900 }, { "epoch": 23.39890180878553, "grad_norm": 0.4472445249557495, "learning_rate": 0.0002491894439208294, "loss": 2.9609, "step": 86950 }, { "epoch": 23.41236003445306, "grad_norm": 0.46697506308555603, "learning_rate": 0.00024898747812037163, "loss": 2.9528, "step": 87000 }, { "epoch": 23.41236003445306, "eval_accuracy": 0.3964992027493739, "eval_loss": 3.283453941345215, "eval_runtime": 83.9935, "eval_samples_per_second": 214.445, "eval_steps_per_second": 13.406, "step": 87000 }, { "epoch": 23.425818260120586, "grad_norm": 0.5904593467712402, "learning_rate": 0.0002487855123199138, "loss": 2.967, "step": 87050 }, { "epoch": 23.439276485788113, "grad_norm": 0.49468109011650085, "learning_rate": 0.000248583546519456, "loss": 2.9732, "step": 87100 }, { "epoch": 23.45273471145564, "grad_norm": 0.45308589935302734, "learning_rate": 0.0002483815807189982, "loss": 2.9798, "step": 87150 }, { "epoch": 23.466192937123168, "grad_norm": 0.4565460979938507, "learning_rate": 0.0002481796149185404, "loss": 2.9742, "step": 87200 }, { "epoch": 23.4796511627907, "grad_norm": 0.4237934350967407, "learning_rate": 0.00024797764911808266, "loss": 2.9548, "step": 87250 }, { "epoch": 23.493109388458226, "grad_norm": 0.426654577255249, "learning_rate": 0.00024777568331762486, "loss": 2.963, "step": 87300 }, { "epoch": 23.506567614125753, "grad_norm": 0.447750449180603, "learning_rate": 0.00024757371751716706, "loss": 2.9707, "step": 87350 }, { "epoch": 23.52002583979328, "grad_norm": 0.4716593027114868, "learning_rate": 0.0002473717517167093, "loss": 2.9643, "step": 87400 }, { "epoch": 23.533484065460808, "grad_norm": 0.43913519382476807, "learning_rate": 0.0002471697859162515, "loss": 2.9654, "step": 87450 }, { "epoch": 23.54694229112834, "grad_norm": 0.4803863763809204, "learning_rate": 0.0002469678201157937, "loss": 2.9741, "step": 87500 }, { "epoch": 23.560400516795866, "grad_norm": 0.4569952189922333, "learning_rate": 0.0002467658543153359, "loss": 2.9743, "step": 87550 }, { "epoch": 23.573858742463393, "grad_norm": 0.4601038694381714, "learning_rate": 0.00024656388851487815, "loss": 2.9795, "step": 87600 }, { "epoch": 23.58731696813092, "grad_norm": 0.44096481800079346, "learning_rate": 0.00024636192271442034, "loss": 2.9774, "step": 87650 }, { "epoch": 23.600775193798448, "grad_norm": 0.45582062005996704, "learning_rate": 0.00024615995691396254, "loss": 2.9792, "step": 87700 }, { "epoch": 23.61423341946598, "grad_norm": 0.4384517967700958, "learning_rate": 0.00024595799111350473, "loss": 2.9735, "step": 87750 }, { "epoch": 23.627691645133506, "grad_norm": 0.4709102511405945, "learning_rate": 0.00024575602531304693, "loss": 2.9772, "step": 87800 }, { "epoch": 23.641149870801033, "grad_norm": 0.4474182426929474, "learning_rate": 0.0002455540595125892, "loss": 2.9737, "step": 87850 }, { "epoch": 23.65460809646856, "grad_norm": 0.47512778639793396, "learning_rate": 0.00024535209371213143, "loss": 2.969, "step": 87900 }, { "epoch": 23.668066322136088, "grad_norm": 0.4888509213924408, "learning_rate": 0.0002451501279116736, "loss": 2.9792, "step": 87950 }, { "epoch": 23.68152454780362, "grad_norm": 0.47287872433662415, "learning_rate": 0.0002449481621112158, "loss": 2.97, "step": 88000 }, { "epoch": 23.68152454780362, "eval_accuracy": 0.39661110901060526, "eval_loss": 3.2801294326782227, "eval_runtime": 84.0084, "eval_samples_per_second": 214.407, "eval_steps_per_second": 13.403, "step": 88000 }, { "epoch": 23.694982773471146, "grad_norm": 0.47039493918418884, "learning_rate": 0.000244746196310758, "loss": 2.9744, "step": 88050 }, { "epoch": 23.708440999138674, "grad_norm": 0.43563172221183777, "learning_rate": 0.0002445442305103002, "loss": 2.9885, "step": 88100 }, { "epoch": 23.7218992248062, "grad_norm": 0.42644500732421875, "learning_rate": 0.00024434226470984246, "loss": 2.9807, "step": 88150 }, { "epoch": 23.735357450473728, "grad_norm": 0.452154278755188, "learning_rate": 0.00024414029890938463, "loss": 2.9846, "step": 88200 }, { "epoch": 23.74881567614126, "grad_norm": 0.47990021109580994, "learning_rate": 0.00024393833310892686, "loss": 2.985, "step": 88250 }, { "epoch": 23.762273901808786, "grad_norm": 0.4605526626110077, "learning_rate": 0.0002437363673084691, "loss": 2.9819, "step": 88300 }, { "epoch": 23.775732127476314, "grad_norm": 0.43846940994262695, "learning_rate": 0.0002435344015080113, "loss": 2.965, "step": 88350 }, { "epoch": 23.78919035314384, "grad_norm": 0.4395309090614319, "learning_rate": 0.0002433324357075535, "loss": 2.9785, "step": 88400 }, { "epoch": 23.802648578811368, "grad_norm": 0.4506181478500366, "learning_rate": 0.00024313046990709572, "loss": 2.9777, "step": 88450 }, { "epoch": 23.8161068044789, "grad_norm": 0.4906696379184723, "learning_rate": 0.00024292850410663792, "loss": 2.9772, "step": 88500 }, { "epoch": 23.829565030146426, "grad_norm": 0.47423872351646423, "learning_rate": 0.00024272653830618014, "loss": 2.9695, "step": 88550 }, { "epoch": 23.843023255813954, "grad_norm": 0.4664286971092224, "learning_rate": 0.00024252457250572234, "loss": 2.9769, "step": 88600 }, { "epoch": 23.85648148148148, "grad_norm": 0.44155094027519226, "learning_rate": 0.00024232260670526453, "loss": 2.9766, "step": 88650 }, { "epoch": 23.86993970714901, "grad_norm": 0.45096275210380554, "learning_rate": 0.00024212064090480676, "loss": 2.9964, "step": 88700 }, { "epoch": 23.88339793281654, "grad_norm": 0.43338635563850403, "learning_rate": 0.00024191867510434898, "loss": 2.9943, "step": 88750 }, { "epoch": 23.896856158484066, "grad_norm": 0.45911747217178345, "learning_rate": 0.0002417167093038912, "loss": 2.9875, "step": 88800 }, { "epoch": 23.910314384151594, "grad_norm": 0.4572448134422302, "learning_rate": 0.0002415147435034334, "loss": 2.9924, "step": 88850 }, { "epoch": 23.92377260981912, "grad_norm": 0.4544650614261627, "learning_rate": 0.00024131277770297562, "loss": 2.9838, "step": 88900 }, { "epoch": 23.93723083548665, "grad_norm": 0.46896883845329285, "learning_rate": 0.00024111081190251782, "loss": 2.9849, "step": 88950 }, { "epoch": 23.95068906115418, "grad_norm": 0.4377521574497223, "learning_rate": 0.00024090884610206002, "loss": 2.9728, "step": 89000 }, { "epoch": 23.95068906115418, "eval_accuracy": 0.3970318982822063, "eval_loss": 3.2736828327178955, "eval_runtime": 84.1787, "eval_samples_per_second": 213.973, "eval_steps_per_second": 13.376, "step": 89000 }, { "epoch": 23.964147286821706, "grad_norm": 0.459255188703537, "learning_rate": 0.00024070688030160224, "loss": 2.9961, "step": 89050 }, { "epoch": 23.977605512489234, "grad_norm": 0.45383593440055847, "learning_rate": 0.00024050491450114443, "loss": 2.989, "step": 89100 }, { "epoch": 23.99106373815676, "grad_norm": 0.4712170660495758, "learning_rate": 0.00024030294870068666, "loss": 2.9929, "step": 89150 }, { "epoch": 24.00430663221361, "grad_norm": 0.47109752893447876, "learning_rate": 0.00024010098290022888, "loss": 2.9677, "step": 89200 }, { "epoch": 24.017764857881136, "grad_norm": 0.4482331871986389, "learning_rate": 0.0002398990170997711, "loss": 2.9114, "step": 89250 }, { "epoch": 24.031223083548664, "grad_norm": 0.4347875416278839, "learning_rate": 0.0002396970512993133, "loss": 2.8935, "step": 89300 }, { "epoch": 24.044681309216195, "grad_norm": 0.46514439582824707, "learning_rate": 0.00023949508549885552, "loss": 2.9106, "step": 89350 }, { "epoch": 24.058139534883722, "grad_norm": 0.47808149456977844, "learning_rate": 0.00023929311969839772, "loss": 2.917, "step": 89400 }, { "epoch": 24.07159776055125, "grad_norm": 0.49696797132492065, "learning_rate": 0.00023909115389793992, "loss": 2.9212, "step": 89450 }, { "epoch": 24.085055986218777, "grad_norm": 0.47918155789375305, "learning_rate": 0.00023888918809748214, "loss": 2.9234, "step": 89500 }, { "epoch": 24.098514211886304, "grad_norm": 0.4478085935115814, "learning_rate": 0.00023868722229702433, "loss": 2.9265, "step": 89550 }, { "epoch": 24.11197243755383, "grad_norm": 0.45360133051872253, "learning_rate": 0.00023848525649656656, "loss": 2.9209, "step": 89600 }, { "epoch": 24.125430663221362, "grad_norm": 0.4664364755153656, "learning_rate": 0.00023828329069610878, "loss": 2.9289, "step": 89650 }, { "epoch": 24.13888888888889, "grad_norm": 0.48554185032844543, "learning_rate": 0.000238081324895651, "loss": 2.9267, "step": 89700 }, { "epoch": 24.152347114556417, "grad_norm": 0.47263389825820923, "learning_rate": 0.0002378793590951932, "loss": 2.933, "step": 89750 }, { "epoch": 24.165805340223944, "grad_norm": 0.48952844738960266, "learning_rate": 0.0002376773932947354, "loss": 2.924, "step": 89800 }, { "epoch": 24.17926356589147, "grad_norm": 0.44809359312057495, "learning_rate": 0.00023747542749427762, "loss": 2.9298, "step": 89850 }, { "epoch": 24.192721791559002, "grad_norm": 0.4537670314311981, "learning_rate": 0.00023727346169381982, "loss": 2.9255, "step": 89900 }, { "epoch": 24.20618001722653, "grad_norm": 0.4854421317577362, "learning_rate": 0.00023707149589336204, "loss": 2.9313, "step": 89950 }, { "epoch": 24.219638242894057, "grad_norm": 0.46356984972953796, "learning_rate": 0.00023686953009290424, "loss": 2.9435, "step": 90000 }, { "epoch": 24.219638242894057, "eval_accuracy": 0.39640793939069974, "eval_loss": 3.288700819015503, "eval_runtime": 84.1466, "eval_samples_per_second": 214.055, "eval_steps_per_second": 13.381, "step": 90000 }, { "epoch": 24.233096468561584, "grad_norm": 0.5046513080596924, "learning_rate": 0.00023666756429244643, "loss": 2.9484, "step": 90050 }, { "epoch": 24.24655469422911, "grad_norm": 0.4581146538257599, "learning_rate": 0.00023646559849198868, "loss": 2.9311, "step": 90100 }, { "epoch": 24.260012919896642, "grad_norm": 0.42044317722320557, "learning_rate": 0.00023626363269153088, "loss": 2.9344, "step": 90150 }, { "epoch": 24.27347114556417, "grad_norm": 0.4418257474899292, "learning_rate": 0.0002360616668910731, "loss": 2.9471, "step": 90200 }, { "epoch": 24.286929371231697, "grad_norm": 0.4563981294631958, "learning_rate": 0.0002358597010906153, "loss": 2.9365, "step": 90250 }, { "epoch": 24.300387596899224, "grad_norm": 0.4627529978752136, "learning_rate": 0.00023565773529015752, "loss": 2.9524, "step": 90300 }, { "epoch": 24.31384582256675, "grad_norm": 0.44930917024612427, "learning_rate": 0.00023545576948969972, "loss": 2.9427, "step": 90350 }, { "epoch": 24.327304048234282, "grad_norm": 0.4439551830291748, "learning_rate": 0.00023525380368924194, "loss": 2.9419, "step": 90400 }, { "epoch": 24.34076227390181, "grad_norm": 0.4861012101173401, "learning_rate": 0.00023505183788878414, "loss": 2.9401, "step": 90450 }, { "epoch": 24.354220499569337, "grad_norm": 0.43457016348838806, "learning_rate": 0.00023484987208832633, "loss": 2.9452, "step": 90500 }, { "epoch": 24.367678725236864, "grad_norm": 0.4539009928703308, "learning_rate": 0.00023464790628786858, "loss": 2.9355, "step": 90550 }, { "epoch": 24.38113695090439, "grad_norm": 0.495311439037323, "learning_rate": 0.00023444594048741078, "loss": 2.9517, "step": 90600 }, { "epoch": 24.394595176571922, "grad_norm": 0.4873776137828827, "learning_rate": 0.000234243974686953, "loss": 2.9379, "step": 90650 }, { "epoch": 24.40805340223945, "grad_norm": 0.4435269236564636, "learning_rate": 0.0002340420088864952, "loss": 2.9579, "step": 90700 }, { "epoch": 24.421511627906977, "grad_norm": 0.4554521143436432, "learning_rate": 0.00023384004308603742, "loss": 2.9542, "step": 90750 }, { "epoch": 24.434969853574504, "grad_norm": 0.44386592507362366, "learning_rate": 0.00023363807728557962, "loss": 2.96, "step": 90800 }, { "epoch": 24.44842807924203, "grad_norm": 0.436383992433548, "learning_rate": 0.0002334361114851218, "loss": 2.9601, "step": 90850 }, { "epoch": 24.461886304909562, "grad_norm": 0.4544837772846222, "learning_rate": 0.00023323414568466404, "loss": 2.9581, "step": 90900 }, { "epoch": 24.47534453057709, "grad_norm": 0.44542551040649414, "learning_rate": 0.00023303217988420623, "loss": 2.9575, "step": 90950 }, { "epoch": 24.488802756244617, "grad_norm": 0.4572954773902893, "learning_rate": 0.00023283021408374848, "loss": 2.9571, "step": 91000 }, { "epoch": 24.488802756244617, "eval_accuracy": 0.39685578172933633, "eval_loss": 3.2813351154327393, "eval_runtime": 84.0012, "eval_samples_per_second": 214.425, "eval_steps_per_second": 13.405, "step": 91000 }, { "epoch": 24.502260981912144, "grad_norm": 0.4547075033187866, "learning_rate": 0.00023262824828329068, "loss": 2.9532, "step": 91050 }, { "epoch": 24.51571920757967, "grad_norm": 0.4635523855686188, "learning_rate": 0.0002324262824828329, "loss": 2.9631, "step": 91100 }, { "epoch": 24.529177433247202, "grad_norm": 0.49691566824913025, "learning_rate": 0.0002322243166823751, "loss": 2.9491, "step": 91150 }, { "epoch": 24.54263565891473, "grad_norm": 0.4702029526233673, "learning_rate": 0.0002320223508819173, "loss": 2.9545, "step": 91200 }, { "epoch": 24.556093884582257, "grad_norm": 0.42786115407943726, "learning_rate": 0.00023182038508145952, "loss": 2.967, "step": 91250 }, { "epoch": 24.569552110249784, "grad_norm": 0.4416402578353882, "learning_rate": 0.0002316184192810017, "loss": 2.9609, "step": 91300 }, { "epoch": 24.58301033591731, "grad_norm": 0.465877890586853, "learning_rate": 0.00023141645348054394, "loss": 2.9648, "step": 91350 }, { "epoch": 24.596468561584842, "grad_norm": 0.44854775071144104, "learning_rate": 0.00023121448768008613, "loss": 2.9509, "step": 91400 }, { "epoch": 24.60992678725237, "grad_norm": 0.456569641828537, "learning_rate": 0.00023101252187962838, "loss": 2.9626, "step": 91450 }, { "epoch": 24.623385012919897, "grad_norm": 0.4613625109195709, "learning_rate": 0.00023081055607917058, "loss": 2.9674, "step": 91500 }, { "epoch": 24.636843238587424, "grad_norm": 0.47509756684303284, "learning_rate": 0.0002306085902787128, "loss": 2.9574, "step": 91550 }, { "epoch": 24.65030146425495, "grad_norm": 0.480103999376297, "learning_rate": 0.000230406624478255, "loss": 2.9611, "step": 91600 }, { "epoch": 24.66375968992248, "grad_norm": 0.44871267676353455, "learning_rate": 0.0002302046586777972, "loss": 2.9734, "step": 91650 }, { "epoch": 24.67721791559001, "grad_norm": 0.4531683921813965, "learning_rate": 0.00023000269287733942, "loss": 2.9583, "step": 91700 }, { "epoch": 24.690676141257537, "grad_norm": 0.4490005671977997, "learning_rate": 0.00022980072707688161, "loss": 2.9628, "step": 91750 }, { "epoch": 24.704134366925064, "grad_norm": 0.546238362789154, "learning_rate": 0.00022959876127642384, "loss": 2.973, "step": 91800 }, { "epoch": 24.71759259259259, "grad_norm": 0.43302252888679504, "learning_rate": 0.00022939679547596603, "loss": 2.9686, "step": 91850 }, { "epoch": 24.73105081826012, "grad_norm": 0.47013354301452637, "learning_rate": 0.00022919482967550828, "loss": 2.9526, "step": 91900 }, { "epoch": 24.74450904392765, "grad_norm": 0.4680594801902771, "learning_rate": 0.00022899286387505048, "loss": 2.9645, "step": 91950 }, { "epoch": 24.757967269595177, "grad_norm": 0.5106493234634399, "learning_rate": 0.00022879089807459268, "loss": 2.9761, "step": 92000 }, { "epoch": 24.757967269595177, "eval_accuracy": 0.3969161893810302, "eval_loss": 3.279923439025879, "eval_runtime": 83.9591, "eval_samples_per_second": 214.533, "eval_steps_per_second": 13.411, "step": 92000 }, { "epoch": 24.771425495262704, "grad_norm": 0.45333370566368103, "learning_rate": 0.0002285889322741349, "loss": 2.9641, "step": 92050 }, { "epoch": 24.78488372093023, "grad_norm": 0.48366883397102356, "learning_rate": 0.0002283869664736771, "loss": 2.967, "step": 92100 }, { "epoch": 24.79834194659776, "grad_norm": 0.46147432923316956, "learning_rate": 0.00022818500067321932, "loss": 2.9815, "step": 92150 }, { "epoch": 24.81180017226529, "grad_norm": 0.4639972150325775, "learning_rate": 0.00022798303487276151, "loss": 2.9669, "step": 92200 }, { "epoch": 24.825258397932817, "grad_norm": 0.45889705419540405, "learning_rate": 0.0002277810690723037, "loss": 2.9811, "step": 92250 }, { "epoch": 24.838716623600344, "grad_norm": 0.4609624445438385, "learning_rate": 0.00022757910327184596, "loss": 2.979, "step": 92300 }, { "epoch": 24.852174849267872, "grad_norm": 0.45697692036628723, "learning_rate": 0.00022737713747138818, "loss": 2.9789, "step": 92350 }, { "epoch": 24.8656330749354, "grad_norm": 0.45627814531326294, "learning_rate": 0.00022717517167093038, "loss": 2.9801, "step": 92400 }, { "epoch": 24.87909130060293, "grad_norm": 0.4848513901233673, "learning_rate": 0.00022697320587047258, "loss": 2.9808, "step": 92450 }, { "epoch": 24.892549526270457, "grad_norm": 0.44247519969940186, "learning_rate": 0.0002267712400700148, "loss": 2.9763, "step": 92500 }, { "epoch": 24.906007751937985, "grad_norm": 0.45784133672714233, "learning_rate": 0.000226569274269557, "loss": 2.9812, "step": 92550 }, { "epoch": 24.919465977605512, "grad_norm": 0.45247259736061096, "learning_rate": 0.00022636730846909922, "loss": 2.9914, "step": 92600 }, { "epoch": 24.93292420327304, "grad_norm": 0.48047441244125366, "learning_rate": 0.00022616534266864141, "loss": 2.983, "step": 92650 }, { "epoch": 24.94638242894057, "grad_norm": 0.5039317011833191, "learning_rate": 0.0002259633768681836, "loss": 2.9788, "step": 92700 }, { "epoch": 24.959840654608097, "grad_norm": 0.4445413053035736, "learning_rate": 0.00022576141106772586, "loss": 2.9792, "step": 92750 }, { "epoch": 24.973298880275625, "grad_norm": 0.4654720425605774, "learning_rate": 0.00022555944526726806, "loss": 2.9884, "step": 92800 }, { "epoch": 24.986757105943152, "grad_norm": 0.4949401021003723, "learning_rate": 0.00022535747946681028, "loss": 2.9895, "step": 92850 }, { "epoch": 25.0, "grad_norm": 0.9609937071800232, "learning_rate": 0.00022515551366635248, "loss": 2.9747, "step": 92900 }, { "epoch": 25.013458225667527, "grad_norm": 0.4609181582927704, "learning_rate": 0.0002249535478658947, "loss": 2.8994, "step": 92950 }, { "epoch": 25.026916451335055, "grad_norm": 0.4751029312610626, "learning_rate": 0.0002247515820654369, "loss": 2.8999, "step": 93000 }, { "epoch": 25.026916451335055, "eval_accuracy": 0.3965985059753598, "eval_loss": 3.2856178283691406, "eval_runtime": 83.9914, "eval_samples_per_second": 214.45, "eval_steps_per_second": 13.406, "step": 93000 }, { "epoch": 25.040374677002585, "grad_norm": 0.4698486924171448, "learning_rate": 0.0002245496162649791, "loss": 2.8946, "step": 93050 }, { "epoch": 25.053832902670113, "grad_norm": 0.4572475254535675, "learning_rate": 0.00022434765046452131, "loss": 2.9131, "step": 93100 }, { "epoch": 25.06729112833764, "grad_norm": 0.4726247191429138, "learning_rate": 0.0002241456846640635, "loss": 2.9103, "step": 93150 }, { "epoch": 25.080749354005167, "grad_norm": 0.4614805579185486, "learning_rate": 0.00022394371886360576, "loss": 2.9139, "step": 93200 }, { "epoch": 25.094207579672695, "grad_norm": 0.4564429819583893, "learning_rate": 0.00022374175306314796, "loss": 2.911, "step": 93250 }, { "epoch": 25.107665805340226, "grad_norm": 0.4605506658554077, "learning_rate": 0.00022353978726269018, "loss": 2.916, "step": 93300 }, { "epoch": 25.121124031007753, "grad_norm": 0.4785519540309906, "learning_rate": 0.00022333782146223238, "loss": 2.9163, "step": 93350 }, { "epoch": 25.13458225667528, "grad_norm": 0.46773386001586914, "learning_rate": 0.0002231358556617746, "loss": 2.9313, "step": 93400 }, { "epoch": 25.148040482342807, "grad_norm": 0.44259631633758545, "learning_rate": 0.0002229338898613168, "loss": 2.9122, "step": 93450 }, { "epoch": 25.161498708010335, "grad_norm": 0.4517664909362793, "learning_rate": 0.000222731924060859, "loss": 2.923, "step": 93500 }, { "epoch": 25.174956933677866, "grad_norm": 0.46426427364349365, "learning_rate": 0.00022252995826040122, "loss": 2.9165, "step": 93550 }, { "epoch": 25.188415159345393, "grad_norm": 0.48092758655548096, "learning_rate": 0.0002223279924599434, "loss": 2.91, "step": 93600 }, { "epoch": 25.20187338501292, "grad_norm": 0.4678271412849426, "learning_rate": 0.00022212602665948566, "loss": 2.9233, "step": 93650 }, { "epoch": 25.215331610680447, "grad_norm": 0.45270681381225586, "learning_rate": 0.00022192406085902786, "loss": 2.9271, "step": 93700 }, { "epoch": 25.228789836347975, "grad_norm": 0.5142789483070374, "learning_rate": 0.00022172209505857008, "loss": 2.9075, "step": 93750 }, { "epoch": 25.242248062015506, "grad_norm": 0.5167204141616821, "learning_rate": 0.00022152012925811228, "loss": 2.9228, "step": 93800 }, { "epoch": 25.255706287683033, "grad_norm": 0.4759088158607483, "learning_rate": 0.00022131816345765447, "loss": 2.9355, "step": 93850 }, { "epoch": 25.26916451335056, "grad_norm": 0.4936668872833252, "learning_rate": 0.0002211161976571967, "loss": 2.9284, "step": 93900 }, { "epoch": 25.282622739018088, "grad_norm": 0.48341456055641174, "learning_rate": 0.0002209142318567389, "loss": 2.9428, "step": 93950 }, { "epoch": 25.296080964685615, "grad_norm": 0.4980866611003876, "learning_rate": 0.00022071226605628112, "loss": 2.9318, "step": 94000 }, { "epoch": 25.296080964685615, "eval_accuracy": 0.3968106932842771, "eval_loss": 3.2875349521636963, "eval_runtime": 85.4169, "eval_samples_per_second": 210.872, "eval_steps_per_second": 13.182, "step": 94000 }, { "epoch": 25.309539190353142, "grad_norm": 0.47112831473350525, "learning_rate": 0.0002205103002558233, "loss": 2.9312, "step": 94050 }, { "epoch": 25.322997416020673, "grad_norm": 0.4677145779132843, "learning_rate": 0.00022030833445536556, "loss": 2.925, "step": 94100 }, { "epoch": 25.3364556416882, "grad_norm": 0.47816231846809387, "learning_rate": 0.00022010636865490776, "loss": 2.9364, "step": 94150 }, { "epoch": 25.349913867355728, "grad_norm": 0.5191966891288757, "learning_rate": 0.00021990440285444998, "loss": 2.9393, "step": 94200 }, { "epoch": 25.363372093023255, "grad_norm": 0.49411892890930176, "learning_rate": 0.00021970243705399218, "loss": 2.9502, "step": 94250 }, { "epoch": 25.376830318690782, "grad_norm": 0.4911155700683594, "learning_rate": 0.00021950047125353437, "loss": 2.9462, "step": 94300 }, { "epoch": 25.390288544358313, "grad_norm": 0.47045525908470154, "learning_rate": 0.0002192985054530766, "loss": 2.9318, "step": 94350 }, { "epoch": 25.40374677002584, "grad_norm": 0.47992491722106934, "learning_rate": 0.0002190965396526188, "loss": 2.9352, "step": 94400 }, { "epoch": 25.417204995693368, "grad_norm": 0.45442041754722595, "learning_rate": 0.00021889457385216102, "loss": 2.9349, "step": 94450 }, { "epoch": 25.430663221360895, "grad_norm": 0.481173038482666, "learning_rate": 0.0002186926080517032, "loss": 2.9338, "step": 94500 }, { "epoch": 25.444121447028422, "grad_norm": 0.49459052085876465, "learning_rate": 0.00021849064225124546, "loss": 2.9449, "step": 94550 }, { "epoch": 25.457579672695953, "grad_norm": 0.4651554524898529, "learning_rate": 0.00021828867645078766, "loss": 2.9481, "step": 94600 }, { "epoch": 25.47103789836348, "grad_norm": 0.45628347992897034, "learning_rate": 0.00021808671065032985, "loss": 2.9565, "step": 94650 }, { "epoch": 25.484496124031008, "grad_norm": 0.45765358209609985, "learning_rate": 0.00021788474484987208, "loss": 2.9551, "step": 94700 }, { "epoch": 25.497954349698535, "grad_norm": 0.4496651589870453, "learning_rate": 0.00021768277904941427, "loss": 2.9388, "step": 94750 }, { "epoch": 25.511412575366062, "grad_norm": 0.46032917499542236, "learning_rate": 0.0002174808132489565, "loss": 2.9494, "step": 94800 }, { "epoch": 25.524870801033593, "grad_norm": 0.4763024151325226, "learning_rate": 0.0002172788474484987, "loss": 2.9531, "step": 94850 }, { "epoch": 25.53832902670112, "grad_norm": 0.48358646035194397, "learning_rate": 0.0002170768816480409, "loss": 2.9531, "step": 94900 }, { "epoch": 25.551787252368648, "grad_norm": 0.4779926836490631, "learning_rate": 0.0002168749158475831, "loss": 2.9441, "step": 94950 }, { "epoch": 25.565245478036175, "grad_norm": 0.4780193567276001, "learning_rate": 0.00021667295004712534, "loss": 2.9437, "step": 95000 }, { "epoch": 25.565245478036175, "eval_accuracy": 0.39687327387308224, "eval_loss": 3.283360242843628, "eval_runtime": 83.7458, "eval_samples_per_second": 215.079, "eval_steps_per_second": 13.445, "step": 95000 }, { "epoch": 25.578703703703702, "grad_norm": 0.488501638174057, "learning_rate": 0.00021647098424666756, "loss": 2.9528, "step": 95050 }, { "epoch": 25.592161929371233, "grad_norm": 0.44475841522216797, "learning_rate": 0.00021626901844620976, "loss": 2.9496, "step": 95100 }, { "epoch": 25.60562015503876, "grad_norm": 0.480144739151001, "learning_rate": 0.00021606705264575198, "loss": 2.9611, "step": 95150 }, { "epoch": 25.619078380706288, "grad_norm": 0.44786661863327026, "learning_rate": 0.00021586508684529417, "loss": 2.9509, "step": 95200 }, { "epoch": 25.632536606373815, "grad_norm": 0.45520010590553284, "learning_rate": 0.0002156631210448364, "loss": 2.9417, "step": 95250 }, { "epoch": 25.645994832041342, "grad_norm": 0.4738970994949341, "learning_rate": 0.0002154611552443786, "loss": 2.9596, "step": 95300 }, { "epoch": 25.659453057708873, "grad_norm": 0.4593546986579895, "learning_rate": 0.0002152591894439208, "loss": 2.9519, "step": 95350 }, { "epoch": 25.6729112833764, "grad_norm": 0.46724069118499756, "learning_rate": 0.000215057223643463, "loss": 2.9491, "step": 95400 }, { "epoch": 25.686369509043928, "grad_norm": 0.48032504320144653, "learning_rate": 0.00021485525784300524, "loss": 2.9718, "step": 95450 }, { "epoch": 25.699827734711455, "grad_norm": 0.46497252583503723, "learning_rate": 0.00021465329204254746, "loss": 2.9629, "step": 95500 }, { "epoch": 25.713285960378983, "grad_norm": 0.453347772359848, "learning_rate": 0.00021445132624208966, "loss": 2.9635, "step": 95550 }, { "epoch": 25.72674418604651, "grad_norm": 0.46574610471725464, "learning_rate": 0.00021424936044163188, "loss": 2.9652, "step": 95600 }, { "epoch": 25.74020241171404, "grad_norm": 0.47337067127227783, "learning_rate": 0.00021404739464117407, "loss": 2.9635, "step": 95650 }, { "epoch": 25.753660637381568, "grad_norm": 0.45252326130867004, "learning_rate": 0.00021384542884071627, "loss": 2.9683, "step": 95700 }, { "epoch": 25.767118863049095, "grad_norm": 0.4818001985549927, "learning_rate": 0.0002136434630402585, "loss": 2.9698, "step": 95750 }, { "epoch": 25.780577088716623, "grad_norm": 0.46499624848365784, "learning_rate": 0.0002134414972398007, "loss": 2.9622, "step": 95800 }, { "epoch": 25.794035314384153, "grad_norm": 0.4897556006908417, "learning_rate": 0.0002132395314393429, "loss": 2.965, "step": 95850 }, { "epoch": 25.80749354005168, "grad_norm": 0.4888896644115448, "learning_rate": 0.00021303756563888514, "loss": 2.9696, "step": 95900 }, { "epoch": 25.820951765719208, "grad_norm": 0.46453386545181274, "learning_rate": 0.00021283559983842736, "loss": 2.9519, "step": 95950 }, { "epoch": 25.834409991386735, "grad_norm": 0.4988659918308258, "learning_rate": 0.00021263363403796956, "loss": 2.9606, "step": 96000 }, { "epoch": 25.834409991386735, "eval_accuracy": 0.3976640056878802, "eval_loss": 3.2750282287597656, "eval_runtime": 83.8338, "eval_samples_per_second": 214.854, "eval_steps_per_second": 13.431, "step": 96000 }, { "epoch": 25.847868217054263, "grad_norm": 0.5047129392623901, "learning_rate": 0.00021243166823751175, "loss": 2.959, "step": 96050 }, { "epoch": 25.86132644272179, "grad_norm": 0.4605954885482788, "learning_rate": 0.00021222970243705398, "loss": 2.9571, "step": 96100 }, { "epoch": 25.87478466838932, "grad_norm": 0.4731179475784302, "learning_rate": 0.00021202773663659617, "loss": 2.9732, "step": 96150 }, { "epoch": 25.888242894056848, "grad_norm": 0.49822744727134705, "learning_rate": 0.0002118257708361384, "loss": 2.9623, "step": 96200 }, { "epoch": 25.901701119724375, "grad_norm": 0.46090516448020935, "learning_rate": 0.0002116238050356806, "loss": 2.9764, "step": 96250 }, { "epoch": 25.915159345391903, "grad_norm": 0.4802016019821167, "learning_rate": 0.00021142183923522281, "loss": 2.9673, "step": 96300 }, { "epoch": 25.92861757105943, "grad_norm": 0.477112740278244, "learning_rate": 0.00021121987343476504, "loss": 2.9716, "step": 96350 }, { "epoch": 25.94207579672696, "grad_norm": 0.4376496970653534, "learning_rate": 0.00021101790763430726, "loss": 2.9672, "step": 96400 }, { "epoch": 25.955534022394488, "grad_norm": 0.45158451795578003, "learning_rate": 0.00021081594183384946, "loss": 2.9593, "step": 96450 }, { "epoch": 25.968992248062015, "grad_norm": 0.5054965615272522, "learning_rate": 0.00021061397603339165, "loss": 2.968, "step": 96500 }, { "epoch": 25.982450473729543, "grad_norm": 0.46955764293670654, "learning_rate": 0.00021041201023293388, "loss": 2.9688, "step": 96550 }, { "epoch": 25.99590869939707, "grad_norm": 0.47011783719062805, "learning_rate": 0.00021021004443247607, "loss": 2.9778, "step": 96600 }, { "epoch": 26.009151593453918, "grad_norm": 0.45663025975227356, "learning_rate": 0.0002100080786320183, "loss": 2.9147, "step": 96650 }, { "epoch": 26.022609819121445, "grad_norm": 0.4516238272190094, "learning_rate": 0.0002098061128315605, "loss": 2.8801, "step": 96700 }, { "epoch": 26.036068044788976, "grad_norm": 0.5051943063735962, "learning_rate": 0.00020960414703110274, "loss": 2.8837, "step": 96750 }, { "epoch": 26.049526270456504, "grad_norm": 0.4774962365627289, "learning_rate": 0.00020940218123064494, "loss": 2.896, "step": 96800 }, { "epoch": 26.06298449612403, "grad_norm": 0.4575546681880951, "learning_rate": 0.00020920021543018713, "loss": 2.9035, "step": 96850 }, { "epoch": 26.076442721791558, "grad_norm": 0.4669003188610077, "learning_rate": 0.00020899824962972936, "loss": 2.8937, "step": 96900 }, { "epoch": 26.089900947459086, "grad_norm": 0.47874999046325684, "learning_rate": 0.00020879628382927155, "loss": 2.9065, "step": 96950 }, { "epoch": 26.103359173126616, "grad_norm": 0.45106983184814453, "learning_rate": 0.00020859431802881378, "loss": 2.9163, "step": 97000 }, { "epoch": 26.103359173126616, "eval_accuracy": 0.39677027665400716, "eval_loss": 3.2907752990722656, "eval_runtime": 83.7697, "eval_samples_per_second": 215.018, "eval_steps_per_second": 13.442, "step": 97000 }, { "epoch": 26.116817398794144, "grad_norm": 0.46213534474372864, "learning_rate": 0.00020839235222835597, "loss": 2.9136, "step": 97050 }, { "epoch": 26.13027562446167, "grad_norm": 0.47658196091651917, "learning_rate": 0.00020819038642789817, "loss": 2.908, "step": 97100 }, { "epoch": 26.1437338501292, "grad_norm": 0.4760020971298218, "learning_rate": 0.0002079884206274404, "loss": 2.9077, "step": 97150 }, { "epoch": 26.157192075796726, "grad_norm": 0.5179885625839233, "learning_rate": 0.00020778645482698264, "loss": 2.9048, "step": 97200 }, { "epoch": 26.170650301464256, "grad_norm": 0.46104031801223755, "learning_rate": 0.00020758448902652484, "loss": 2.9132, "step": 97250 }, { "epoch": 26.184108527131784, "grad_norm": 0.48797598481178284, "learning_rate": 0.00020738252322606703, "loss": 2.927, "step": 97300 }, { "epoch": 26.19756675279931, "grad_norm": 0.47429800033569336, "learning_rate": 0.00020718055742560926, "loss": 2.9094, "step": 97350 }, { "epoch": 26.21102497846684, "grad_norm": 0.4832933843135834, "learning_rate": 0.00020697859162515145, "loss": 2.9134, "step": 97400 }, { "epoch": 26.224483204134366, "grad_norm": 0.46741122007369995, "learning_rate": 0.00020677662582469368, "loss": 2.9204, "step": 97450 }, { "epoch": 26.237941429801896, "grad_norm": 0.44947007298469543, "learning_rate": 0.00020657466002423587, "loss": 2.9185, "step": 97500 }, { "epoch": 26.251399655469424, "grad_norm": 0.5112219452857971, "learning_rate": 0.00020637269422377807, "loss": 2.9169, "step": 97550 }, { "epoch": 26.26485788113695, "grad_norm": 0.47085320949554443, "learning_rate": 0.0002061707284233203, "loss": 2.9181, "step": 97600 }, { "epoch": 26.27831610680448, "grad_norm": 0.46664902567863464, "learning_rate": 0.00020596876262286251, "loss": 2.9291, "step": 97650 }, { "epoch": 26.291774332472006, "grad_norm": 0.46194952726364136, "learning_rate": 0.00020576679682240474, "loss": 2.9217, "step": 97700 }, { "epoch": 26.305232558139537, "grad_norm": 0.46810612082481384, "learning_rate": 0.00020556483102194693, "loss": 2.9271, "step": 97750 }, { "epoch": 26.318690783807064, "grad_norm": 0.47608909010887146, "learning_rate": 0.00020536286522148916, "loss": 2.9223, "step": 97800 }, { "epoch": 26.33214900947459, "grad_norm": 0.487276166677475, "learning_rate": 0.00020516089942103135, "loss": 2.9229, "step": 97850 }, { "epoch": 26.34560723514212, "grad_norm": 0.4970484972000122, "learning_rate": 0.00020495893362057355, "loss": 2.9228, "step": 97900 }, { "epoch": 26.359065460809646, "grad_norm": 0.47998833656311035, "learning_rate": 0.00020475696782011577, "loss": 2.9395, "step": 97950 }, { "epoch": 26.372523686477177, "grad_norm": 0.468665212392807, "learning_rate": 0.00020455500201965797, "loss": 2.9334, "step": 98000 }, { "epoch": 26.372523686477177, "eval_accuracy": 0.39712066276320246, "eval_loss": 3.2837681770324707, "eval_runtime": 83.6485, "eval_samples_per_second": 215.33, "eval_steps_per_second": 13.461, "step": 98000 }, { "epoch": 26.385981912144704, "grad_norm": 0.4812726080417633, "learning_rate": 0.0002043530362192002, "loss": 2.9203, "step": 98050 }, { "epoch": 26.39944013781223, "grad_norm": 0.47966867685317993, "learning_rate": 0.00020415107041874242, "loss": 2.9386, "step": 98100 }, { "epoch": 26.41289836347976, "grad_norm": 0.4546014666557312, "learning_rate": 0.00020394910461828464, "loss": 2.9366, "step": 98150 }, { "epoch": 26.426356589147286, "grad_norm": 0.4557799994945526, "learning_rate": 0.00020374713881782683, "loss": 2.9384, "step": 98200 }, { "epoch": 26.439814814814813, "grad_norm": 0.5014848709106445, "learning_rate": 0.00020354517301736906, "loss": 2.9375, "step": 98250 }, { "epoch": 26.453273040482344, "grad_norm": 0.49908778071403503, "learning_rate": 0.00020334320721691125, "loss": 2.9293, "step": 98300 }, { "epoch": 26.46673126614987, "grad_norm": 0.49170541763305664, "learning_rate": 0.00020314124141645345, "loss": 2.9444, "step": 98350 }, { "epoch": 26.4801894918174, "grad_norm": 0.4941914975643158, "learning_rate": 0.00020293927561599567, "loss": 2.9318, "step": 98400 }, { "epoch": 26.493647717484926, "grad_norm": 0.4966713488101959, "learning_rate": 0.00020273730981553787, "loss": 2.9447, "step": 98450 }, { "epoch": 26.507105943152453, "grad_norm": 0.5141641497612, "learning_rate": 0.0002025353440150801, "loss": 2.9441, "step": 98500 }, { "epoch": 26.520564168819984, "grad_norm": 0.4713413715362549, "learning_rate": 0.00020233337821462232, "loss": 2.9434, "step": 98550 }, { "epoch": 26.53402239448751, "grad_norm": 0.4796540439128876, "learning_rate": 0.00020213141241416454, "loss": 2.9414, "step": 98600 }, { "epoch": 26.54748062015504, "grad_norm": 0.4777824580669403, "learning_rate": 0.00020192944661370674, "loss": 2.9441, "step": 98650 }, { "epoch": 26.560938845822566, "grad_norm": 0.48075199127197266, "learning_rate": 0.00020172748081324893, "loss": 2.9466, "step": 98700 }, { "epoch": 26.574397071490093, "grad_norm": 0.4639039933681488, "learning_rate": 0.00020152551501279115, "loss": 2.9362, "step": 98750 }, { "epoch": 26.587855297157624, "grad_norm": 0.48601824045181274, "learning_rate": 0.00020132354921233335, "loss": 2.9426, "step": 98800 }, { "epoch": 26.60131352282515, "grad_norm": 0.4678082764148712, "learning_rate": 0.00020112158341187557, "loss": 2.9364, "step": 98850 }, { "epoch": 26.61477174849268, "grad_norm": 0.4817858338356018, "learning_rate": 0.00020091961761141777, "loss": 2.94, "step": 98900 }, { "epoch": 26.628229974160206, "grad_norm": 0.46154728531837463, "learning_rate": 0.00020071765181095997, "loss": 2.9263, "step": 98950 }, { "epoch": 26.641688199827733, "grad_norm": 0.45043495297431946, "learning_rate": 0.00020051568601050222, "loss": 2.9322, "step": 99000 }, { "epoch": 26.641688199827733, "eval_accuracy": 0.3972260502130999, "eval_loss": 3.278003454208374, "eval_runtime": 84.0483, "eval_samples_per_second": 214.305, "eval_steps_per_second": 13.397, "step": 99000 }, { "epoch": 26.655146425495264, "grad_norm": 0.5009729266166687, "learning_rate": 0.00020031372021004444, "loss": 2.9577, "step": 99050 }, { "epoch": 26.66860465116279, "grad_norm": 0.44591593742370605, "learning_rate": 0.00020011175440958664, "loss": 2.9448, "step": 99100 }, { "epoch": 26.68206287683032, "grad_norm": 0.4795645773410797, "learning_rate": 0.00019990978860912883, "loss": 2.9498, "step": 99150 }, { "epoch": 26.695521102497846, "grad_norm": 0.5004894733428955, "learning_rate": 0.00019970782280867105, "loss": 2.9477, "step": 99200 }, { "epoch": 26.708979328165373, "grad_norm": 0.47407266497612, "learning_rate": 0.00019950585700821325, "loss": 2.9451, "step": 99250 }, { "epoch": 26.722437553832904, "grad_norm": 0.4809727668762207, "learning_rate": 0.00019930389120775547, "loss": 2.9432, "step": 99300 }, { "epoch": 26.73589577950043, "grad_norm": 0.4729549288749695, "learning_rate": 0.00019910192540729767, "loss": 2.9483, "step": 99350 }, { "epoch": 26.74935400516796, "grad_norm": 0.489544540643692, "learning_rate": 0.00019889995960683987, "loss": 2.9468, "step": 99400 }, { "epoch": 26.762812230835486, "grad_norm": 0.4922120273113251, "learning_rate": 0.00019869799380638212, "loss": 2.9615, "step": 99450 }, { "epoch": 26.776270456503013, "grad_norm": 0.47596344351768494, "learning_rate": 0.0001984960280059243, "loss": 2.9431, "step": 99500 }, { "epoch": 26.789728682170544, "grad_norm": 0.5055460333824158, "learning_rate": 0.00019829406220546654, "loss": 2.9551, "step": 99550 }, { "epoch": 26.80318690783807, "grad_norm": 0.4794882833957672, "learning_rate": 0.00019809209640500873, "loss": 2.9575, "step": 99600 }, { "epoch": 26.8166451335056, "grad_norm": 0.4809652864933014, "learning_rate": 0.00019789013060455096, "loss": 2.9513, "step": 99650 }, { "epoch": 26.830103359173126, "grad_norm": 0.44904300570487976, "learning_rate": 0.00019768816480409315, "loss": 2.9546, "step": 99700 }, { "epoch": 26.843561584840653, "grad_norm": 0.5154889225959778, "learning_rate": 0.00019748619900363535, "loss": 2.9653, "step": 99750 }, { "epoch": 26.857019810508184, "grad_norm": 0.528717577457428, "learning_rate": 0.00019728423320317757, "loss": 2.9559, "step": 99800 }, { "epoch": 26.87047803617571, "grad_norm": 0.4803217053413391, "learning_rate": 0.00019708226740271977, "loss": 2.9515, "step": 99850 }, { "epoch": 26.88393626184324, "grad_norm": 0.4940747320652008, "learning_rate": 0.00019688030160226202, "loss": 2.9521, "step": 99900 }, { "epoch": 26.897394487510766, "grad_norm": 0.46168291568756104, "learning_rate": 0.0001966783358018042, "loss": 2.9596, "step": 99950 }, { "epoch": 26.910852713178294, "grad_norm": 0.5004278421401978, "learning_rate": 0.00019647637000134644, "loss": 2.9657, "step": 100000 }, { "epoch": 26.910852713178294, "eval_accuracy": 0.3976188085959654, "eval_loss": 3.275502920150757, "eval_runtime": 84.0403, "eval_samples_per_second": 214.326, "eval_steps_per_second": 13.398, "step": 100000 }, { "epoch": 26.92431093884582, "grad_norm": 0.4664307236671448, "learning_rate": 0.00019627440420088863, "loss": 2.9633, "step": 100050 }, { "epoch": 26.93776916451335, "grad_norm": 0.492509126663208, "learning_rate": 0.00019607243840043086, "loss": 2.9557, "step": 100100 }, { "epoch": 26.95122739018088, "grad_norm": 0.5034955739974976, "learning_rate": 0.00019587047259997305, "loss": 2.9548, "step": 100150 }, { "epoch": 26.964685615848406, "grad_norm": 0.48850440979003906, "learning_rate": 0.00019566850679951525, "loss": 2.9565, "step": 100200 }, { "epoch": 26.978143841515934, "grad_norm": 0.45618629455566406, "learning_rate": 0.00019546654099905747, "loss": 2.9572, "step": 100250 }, { "epoch": 26.99160206718346, "grad_norm": 0.4687550365924835, "learning_rate": 0.00019526457519859967, "loss": 2.9539, "step": 100300 }, { "epoch": 27.00484496124031, "grad_norm": 0.5012471675872803, "learning_rate": 0.00019506260939814192, "loss": 2.9226, "step": 100350 }, { "epoch": 27.01830318690784, "grad_norm": 0.4801206886768341, "learning_rate": 0.0001948606435976841, "loss": 2.8866, "step": 100400 }, { "epoch": 27.031761412575367, "grad_norm": 0.49274516105651855, "learning_rate": 0.00019465867779722634, "loss": 2.8789, "step": 100450 }, { "epoch": 27.045219638242894, "grad_norm": 0.46303367614746094, "learning_rate": 0.00019445671199676853, "loss": 2.8867, "step": 100500 }, { "epoch": 27.05867786391042, "grad_norm": 0.47992032766342163, "learning_rate": 0.00019425474619631073, "loss": 2.8882, "step": 100550 }, { "epoch": 27.07213608957795, "grad_norm": 0.48844286799430847, "learning_rate": 0.00019405278039585295, "loss": 2.8891, "step": 100600 }, { "epoch": 27.085594315245476, "grad_norm": 0.47748664021492004, "learning_rate": 0.00019385081459539515, "loss": 2.8819, "step": 100650 }, { "epoch": 27.099052540913007, "grad_norm": 0.49191465973854065, "learning_rate": 0.00019364884879493737, "loss": 2.8856, "step": 100700 }, { "epoch": 27.112510766580534, "grad_norm": 0.49035146832466125, "learning_rate": 0.00019344688299447957, "loss": 2.9045, "step": 100750 }, { "epoch": 27.125968992248062, "grad_norm": 0.492626428604126, "learning_rate": 0.00019324491719402182, "loss": 2.907, "step": 100800 }, { "epoch": 27.13942721791559, "grad_norm": 0.4873746931552887, "learning_rate": 0.00019304295139356401, "loss": 2.9093, "step": 100850 }, { "epoch": 27.152885443583116, "grad_norm": 0.48530182242393494, "learning_rate": 0.0001928409855931062, "loss": 2.9079, "step": 100900 }, { "epoch": 27.166343669250647, "grad_norm": 0.5125190019607544, "learning_rate": 0.00019263901979264843, "loss": 2.8998, "step": 100950 }, { "epoch": 27.179801894918175, "grad_norm": 0.46221446990966797, "learning_rate": 0.00019243705399219063, "loss": 2.9095, "step": 101000 }, { "epoch": 27.179801894918175, "eval_accuracy": 0.39690793222000725, "eval_loss": 3.28844952583313, "eval_runtime": 84.0707, "eval_samples_per_second": 214.248, "eval_steps_per_second": 13.393, "step": 101000 }, { "epoch": 27.193260120585702, "grad_norm": 0.49618521332740784, "learning_rate": 0.00019223508819173285, "loss": 2.908, "step": 101050 }, { "epoch": 27.20671834625323, "grad_norm": 0.47813424468040466, "learning_rate": 0.00019203312239127505, "loss": 2.8946, "step": 101100 }, { "epoch": 27.220176571920756, "grad_norm": 0.4884779453277588, "learning_rate": 0.00019183115659081727, "loss": 2.9174, "step": 101150 }, { "epoch": 27.233634797588287, "grad_norm": 0.5239505767822266, "learning_rate": 0.00019162919079035947, "loss": 2.9165, "step": 101200 }, { "epoch": 27.247093023255815, "grad_norm": 0.4790191352367401, "learning_rate": 0.00019142722498990172, "loss": 2.9176, "step": 101250 }, { "epoch": 27.260551248923342, "grad_norm": 0.5082780122756958, "learning_rate": 0.00019122525918944391, "loss": 2.9076, "step": 101300 }, { "epoch": 27.27400947459087, "grad_norm": 0.4764711260795593, "learning_rate": 0.0001910232933889861, "loss": 2.9146, "step": 101350 }, { "epoch": 27.287467700258397, "grad_norm": 0.46437740325927734, "learning_rate": 0.00019082132758852833, "loss": 2.9212, "step": 101400 }, { "epoch": 27.300925925925927, "grad_norm": 0.4945138394832611, "learning_rate": 0.00019061936178807053, "loss": 2.9105, "step": 101450 }, { "epoch": 27.314384151593455, "grad_norm": 0.5206960439682007, "learning_rate": 0.00019041739598761275, "loss": 2.9113, "step": 101500 }, { "epoch": 27.327842377260982, "grad_norm": 0.5159957408905029, "learning_rate": 0.00019021543018715495, "loss": 2.9172, "step": 101550 }, { "epoch": 27.34130060292851, "grad_norm": Infinity, "learning_rate": 0.00019001346438669714, "loss": 2.9163, "step": 101600 }, { "epoch": 27.354758828596037, "grad_norm": 0.48938295245170593, "learning_rate": 0.0001898114985862394, "loss": 2.9308, "step": 101650 }, { "epoch": 27.368217054263567, "grad_norm": 0.49014759063720703, "learning_rate": 0.0001896095327857816, "loss": 2.9108, "step": 101700 }, { "epoch": 27.381675279931095, "grad_norm": 0.4875050187110901, "learning_rate": 0.00018940756698532381, "loss": 2.9145, "step": 101750 }, { "epoch": 27.395133505598622, "grad_norm": 0.4899139106273651, "learning_rate": 0.000189205601184866, "loss": 2.9245, "step": 101800 }, { "epoch": 27.40859173126615, "grad_norm": 0.4790636897087097, "learning_rate": 0.00018900363538440823, "loss": 2.9169, "step": 101850 }, { "epoch": 27.422049956933677, "grad_norm": 0.4836386740207672, "learning_rate": 0.00018880166958395043, "loss": 2.9226, "step": 101900 }, { "epoch": 27.435508182601207, "grad_norm": 0.5054500699043274, "learning_rate": 0.00018859970378349263, "loss": 2.9227, "step": 101950 }, { "epoch": 27.448966408268735, "grad_norm": 0.4658251702785492, "learning_rate": 0.00018839773798303485, "loss": 2.9236, "step": 102000 }, { "epoch": 27.448966408268735, "eval_accuracy": 0.39734056399886486, "eval_loss": 3.286876678466797, "eval_runtime": 83.971, "eval_samples_per_second": 214.503, "eval_steps_per_second": 13.409, "step": 102000 }, { "epoch": 27.462424633936262, "grad_norm": 0.4930401146411896, "learning_rate": 0.00018819577218257705, "loss": 2.9215, "step": 102050 }, { "epoch": 27.47588285960379, "grad_norm": 0.4859257638454437, "learning_rate": 0.0001879938063821193, "loss": 2.9259, "step": 102100 }, { "epoch": 27.489341085271317, "grad_norm": 0.46256932616233826, "learning_rate": 0.0001877918405816615, "loss": 2.9334, "step": 102150 }, { "epoch": 27.502799310938848, "grad_norm": 0.48986995220184326, "learning_rate": 0.00018758987478120371, "loss": 2.9305, "step": 102200 }, { "epoch": 27.516257536606375, "grad_norm": 0.4924154281616211, "learning_rate": 0.0001873879089807459, "loss": 2.9334, "step": 102250 }, { "epoch": 27.529715762273902, "grad_norm": 0.47396549582481384, "learning_rate": 0.00018718594318028813, "loss": 2.9282, "step": 102300 }, { "epoch": 27.54317398794143, "grad_norm": 0.4847225844860077, "learning_rate": 0.00018698397737983033, "loss": 2.9254, "step": 102350 }, { "epoch": 27.556632213608957, "grad_norm": 0.4604421555995941, "learning_rate": 0.00018678201157937253, "loss": 2.9322, "step": 102400 }, { "epoch": 27.570090439276484, "grad_norm": 0.4923710823059082, "learning_rate": 0.00018658004577891475, "loss": 2.9295, "step": 102450 }, { "epoch": 27.583548664944015, "grad_norm": 0.5104304552078247, "learning_rate": 0.00018637807997845695, "loss": 2.9257, "step": 102500 }, { "epoch": 27.597006890611542, "grad_norm": 0.4732654392719269, "learning_rate": 0.0001861761141779992, "loss": 2.9429, "step": 102550 }, { "epoch": 27.61046511627907, "grad_norm": 0.47959786653518677, "learning_rate": 0.0001859741483775414, "loss": 2.9296, "step": 102600 }, { "epoch": 27.623923341946597, "grad_norm": 0.5053572058677673, "learning_rate": 0.00018577218257708362, "loss": 2.9232, "step": 102650 }, { "epoch": 27.637381567614124, "grad_norm": 0.47885841131210327, "learning_rate": 0.0001855702167766258, "loss": 2.9288, "step": 102700 }, { "epoch": 27.650839793281655, "grad_norm": 0.5167924761772156, "learning_rate": 0.000185368250976168, "loss": 2.929, "step": 102750 }, { "epoch": 27.664298018949182, "grad_norm": 0.48342621326446533, "learning_rate": 0.00018516628517571023, "loss": 2.938, "step": 102800 }, { "epoch": 27.67775624461671, "grad_norm": 0.5054475665092468, "learning_rate": 0.00018496431937525243, "loss": 2.9331, "step": 102850 }, { "epoch": 27.691214470284237, "grad_norm": 0.46684083342552185, "learning_rate": 0.00018476235357479465, "loss": 2.9428, "step": 102900 }, { "epoch": 27.704672695951764, "grad_norm": 0.5017584562301636, "learning_rate": 0.00018456038777433685, "loss": 2.9348, "step": 102950 }, { "epoch": 27.718130921619295, "grad_norm": 0.46952277421951294, "learning_rate": 0.0001843584219738791, "loss": 2.9426, "step": 103000 }, { "epoch": 27.718130921619295, "eval_accuracy": 0.39760642285443104, "eval_loss": 3.2794814109802246, "eval_runtime": 83.8754, "eval_samples_per_second": 214.747, "eval_steps_per_second": 13.425, "step": 103000 }, { "epoch": 27.731589147286822, "grad_norm": 0.491432785987854, "learning_rate": 0.0001841564561734213, "loss": 2.948, "step": 103050 }, { "epoch": 27.74504737295435, "grad_norm": 0.4925087094306946, "learning_rate": 0.00018395449037296352, "loss": 2.9392, "step": 103100 }, { "epoch": 27.758505598621877, "grad_norm": 0.4647904932498932, "learning_rate": 0.0001837525245725057, "loss": 2.953, "step": 103150 }, { "epoch": 27.771963824289404, "grad_norm": 0.5009186863899231, "learning_rate": 0.0001835505587720479, "loss": 2.9375, "step": 103200 }, { "epoch": 27.785422049956935, "grad_norm": 0.5034534931182861, "learning_rate": 0.00018334859297159013, "loss": 2.9452, "step": 103250 }, { "epoch": 27.798880275624462, "grad_norm": 0.47611668705940247, "learning_rate": 0.00018314662717113233, "loss": 2.9399, "step": 103300 }, { "epoch": 27.81233850129199, "grad_norm": 0.506561815738678, "learning_rate": 0.00018294466137067455, "loss": 2.951, "step": 103350 }, { "epoch": 27.825796726959517, "grad_norm": 0.4973388612270355, "learning_rate": 0.00018274269557021675, "loss": 2.9287, "step": 103400 }, { "epoch": 27.839254952627044, "grad_norm": 0.4861920475959778, "learning_rate": 0.000182540729769759, "loss": 2.9454, "step": 103450 }, { "epoch": 27.852713178294575, "grad_norm": 0.5094921588897705, "learning_rate": 0.0001823387639693012, "loss": 2.9481, "step": 103500 }, { "epoch": 27.866171403962102, "grad_norm": 0.4648674428462982, "learning_rate": 0.0001821367981688434, "loss": 2.9432, "step": 103550 }, { "epoch": 27.87962962962963, "grad_norm": 0.5445858240127563, "learning_rate": 0.0001819348323683856, "loss": 2.948, "step": 103600 }, { "epoch": 27.893087855297157, "grad_norm": 0.48193544149398804, "learning_rate": 0.0001817328665679278, "loss": 2.942, "step": 103650 }, { "epoch": 27.906546080964684, "grad_norm": 0.5002299547195435, "learning_rate": 0.00018153090076747003, "loss": 2.9535, "step": 103700 }, { "epoch": 27.920004306632215, "grad_norm": 0.5245981216430664, "learning_rate": 0.00018132893496701223, "loss": 2.9465, "step": 103750 }, { "epoch": 27.933462532299743, "grad_norm": 0.5068734288215637, "learning_rate": 0.00018112696916655442, "loss": 2.9405, "step": 103800 }, { "epoch": 27.94692075796727, "grad_norm": 0.4961998462677002, "learning_rate": 0.00018092500336609665, "loss": 2.9508, "step": 103850 }, { "epoch": 27.960378983634797, "grad_norm": 0.46381625533103943, "learning_rate": 0.0001807230375656389, "loss": 2.94, "step": 103900 }, { "epoch": 27.973837209302324, "grad_norm": 0.4833906292915344, "learning_rate": 0.0001805210717651811, "loss": 2.95, "step": 103950 }, { "epoch": 27.987295434969855, "grad_norm": 0.4873865246772766, "learning_rate": 0.0001803191059647233, "loss": 2.963, "step": 104000 }, { "epoch": 27.987295434969855, "eval_accuracy": 0.39772006746535143, "eval_loss": 3.2727417945861816, "eval_runtime": 83.8715, "eval_samples_per_second": 214.757, "eval_steps_per_second": 13.425, "step": 104000 }, { "epoch": 28.0005383290267, "grad_norm": 0.4941236972808838, "learning_rate": 0.0001801171401642655, "loss": 2.9422, "step": 104050 }, { "epoch": 28.01399655469423, "grad_norm": 0.49093982577323914, "learning_rate": 0.0001799151743638077, "loss": 2.8689, "step": 104100 }, { "epoch": 28.027454780361758, "grad_norm": 0.5120478272438049, "learning_rate": 0.00017971320856334993, "loss": 2.8857, "step": 104150 }, { "epoch": 28.040913006029285, "grad_norm": 0.4717957377433777, "learning_rate": 0.00017951124276289213, "loss": 2.8768, "step": 104200 }, { "epoch": 28.054371231696813, "grad_norm": 0.4936118423938751, "learning_rate": 0.00017930927696243432, "loss": 2.8843, "step": 104250 }, { "epoch": 28.06782945736434, "grad_norm": 0.4972836971282959, "learning_rate": 0.00017910731116197655, "loss": 2.8788, "step": 104300 }, { "epoch": 28.08128768303187, "grad_norm": 0.4746985137462616, "learning_rate": 0.00017890534536151877, "loss": 2.8813, "step": 104350 }, { "epoch": 28.094745908699398, "grad_norm": 0.5059629082679749, "learning_rate": 0.000178703379561061, "loss": 2.8828, "step": 104400 }, { "epoch": 28.108204134366925, "grad_norm": 0.4806491434574127, "learning_rate": 0.0001785014137606032, "loss": 2.89, "step": 104450 }, { "epoch": 28.121662360034453, "grad_norm": 0.5036650896072388, "learning_rate": 0.0001782994479601454, "loss": 2.8802, "step": 104500 }, { "epoch": 28.13512058570198, "grad_norm": 0.4994170367717743, "learning_rate": 0.0001780974821596876, "loss": 2.8938, "step": 104550 }, { "epoch": 28.14857881136951, "grad_norm": 0.5043787956237793, "learning_rate": 0.0001778955163592298, "loss": 2.8922, "step": 104600 }, { "epoch": 28.162037037037038, "grad_norm": 0.47212567925453186, "learning_rate": 0.00017769355055877203, "loss": 2.8962, "step": 104650 }, { "epoch": 28.175495262704565, "grad_norm": 0.4917432367801666, "learning_rate": 0.00017749158475831422, "loss": 2.8879, "step": 104700 }, { "epoch": 28.188953488372093, "grad_norm": 0.48865607380867004, "learning_rate": 0.00017728961895785645, "loss": 2.898, "step": 104750 }, { "epoch": 28.20241171403962, "grad_norm": 0.505818784236908, "learning_rate": 0.00017708765315739867, "loss": 2.8974, "step": 104800 }, { "epoch": 28.215869939707147, "grad_norm": 0.5183230638504028, "learning_rate": 0.0001768856873569409, "loss": 2.8967, "step": 104850 }, { "epoch": 28.229328165374678, "grad_norm": 0.48350223898887634, "learning_rate": 0.0001766837215564831, "loss": 2.9063, "step": 104900 }, { "epoch": 28.242786391042205, "grad_norm": 0.4778302311897278, "learning_rate": 0.0001764817557560253, "loss": 2.8915, "step": 104950 }, { "epoch": 28.256244616709733, "grad_norm": 0.524963915348053, "learning_rate": 0.0001762797899555675, "loss": 2.9024, "step": 105000 }, { "epoch": 28.256244616709733, "eval_accuracy": 0.3972076888945095, "eval_loss": 3.289083242416382, "eval_runtime": 84.2414, "eval_samples_per_second": 213.814, "eval_steps_per_second": 13.366, "step": 105000 }, { "epoch": 28.26970284237726, "grad_norm": 0.49806711077690125, "learning_rate": 0.0001760778241551097, "loss": 2.9048, "step": 105050 }, { "epoch": 28.283161068044787, "grad_norm": 0.4823664724826813, "learning_rate": 0.00017587585835465193, "loss": 2.91, "step": 105100 }, { "epoch": 28.296619293712318, "grad_norm": 0.4940940737724304, "learning_rate": 0.00017567389255419412, "loss": 2.9071, "step": 105150 }, { "epoch": 28.310077519379846, "grad_norm": 0.4807303845882416, "learning_rate": 0.00017547192675373635, "loss": 2.9051, "step": 105200 }, { "epoch": 28.323535745047373, "grad_norm": 0.47984981536865234, "learning_rate": 0.00017526996095327857, "loss": 2.9115, "step": 105250 }, { "epoch": 28.3369939707149, "grad_norm": 0.48672279715538025, "learning_rate": 0.0001750679951528208, "loss": 2.9045, "step": 105300 }, { "epoch": 28.350452196382427, "grad_norm": 0.49630510807037354, "learning_rate": 0.000174866029352363, "loss": 2.9019, "step": 105350 }, { "epoch": 28.36391042204996, "grad_norm": 0.4995165467262268, "learning_rate": 0.0001746640635519052, "loss": 2.9166, "step": 105400 }, { "epoch": 28.377368647717486, "grad_norm": 0.4939239025115967, "learning_rate": 0.0001744620977514474, "loss": 2.8999, "step": 105450 }, { "epoch": 28.390826873385013, "grad_norm": 0.5187941193580627, "learning_rate": 0.0001742601319509896, "loss": 2.9106, "step": 105500 }, { "epoch": 28.40428509905254, "grad_norm": 0.5010408759117126, "learning_rate": 0.00017405816615053183, "loss": 2.9143, "step": 105550 }, { "epoch": 28.417743324720067, "grad_norm": 0.4850703775882721, "learning_rate": 0.00017385620035007403, "loss": 2.9171, "step": 105600 }, { "epoch": 28.4312015503876, "grad_norm": 0.49331873655319214, "learning_rate": 0.00017365423454961622, "loss": 2.9133, "step": 105650 }, { "epoch": 28.444659776055126, "grad_norm": 0.48752540349960327, "learning_rate": 0.00017345226874915847, "loss": 2.9162, "step": 105700 }, { "epoch": 28.458118001722653, "grad_norm": 0.48763734102249146, "learning_rate": 0.00017325030294870067, "loss": 2.9166, "step": 105750 }, { "epoch": 28.47157622739018, "grad_norm": 0.5082513689994812, "learning_rate": 0.0001730483371482429, "loss": 2.911, "step": 105800 }, { "epoch": 28.485034453057708, "grad_norm": 0.511461615562439, "learning_rate": 0.0001728463713477851, "loss": 2.9104, "step": 105850 }, { "epoch": 28.49849267872524, "grad_norm": 0.4683901071548462, "learning_rate": 0.0001726444055473273, "loss": 2.9134, "step": 105900 }, { "epoch": 28.511950904392766, "grad_norm": 0.46261781454086304, "learning_rate": 0.0001724424397468695, "loss": 2.9188, "step": 105950 }, { "epoch": 28.525409130060293, "grad_norm": 0.5006155371665955, "learning_rate": 0.00017224047394641173, "loss": 2.9245, "step": 106000 }, { "epoch": 28.525409130060293, "eval_accuracy": 0.3973929317832469, "eval_loss": 3.2823522090911865, "eval_runtime": 84.2348, "eval_samples_per_second": 213.831, "eval_steps_per_second": 13.367, "step": 106000 }, { "epoch": 28.53886735572782, "grad_norm": 0.49079206585884094, "learning_rate": 0.00017203850814595393, "loss": 2.9187, "step": 106050 }, { "epoch": 28.552325581395348, "grad_norm": 0.4927675127983093, "learning_rate": 0.00017183654234549612, "loss": 2.928, "step": 106100 }, { "epoch": 28.56578380706288, "grad_norm": 0.4958413541316986, "learning_rate": 0.00017163457654503837, "loss": 2.9321, "step": 106150 }, { "epoch": 28.579242032730406, "grad_norm": 0.5253115296363831, "learning_rate": 0.00017143261074458057, "loss": 2.9212, "step": 106200 }, { "epoch": 28.592700258397933, "grad_norm": 0.4913957417011261, "learning_rate": 0.0001712306449441228, "loss": 2.9217, "step": 106250 }, { "epoch": 28.60615848406546, "grad_norm": 0.48931920528411865, "learning_rate": 0.000171028679143665, "loss": 2.9303, "step": 106300 }, { "epoch": 28.619616709732988, "grad_norm": 0.49885618686676025, "learning_rate": 0.0001708267133432072, "loss": 2.9186, "step": 106350 }, { "epoch": 28.63307493540052, "grad_norm": 0.49400222301483154, "learning_rate": 0.0001706247475427494, "loss": 2.9239, "step": 106400 }, { "epoch": 28.646533161068046, "grad_norm": 0.4754203259944916, "learning_rate": 0.0001704227817422916, "loss": 2.9284, "step": 106450 }, { "epoch": 28.659991386735573, "grad_norm": 0.515365719795227, "learning_rate": 0.00017022081594183383, "loss": 2.9208, "step": 106500 }, { "epoch": 28.6734496124031, "grad_norm": 0.4889947772026062, "learning_rate": 0.00017001885014137605, "loss": 2.9298, "step": 106550 }, { "epoch": 28.686907838070628, "grad_norm": 0.507577121257782, "learning_rate": 0.00016981688434091827, "loss": 2.9336, "step": 106600 }, { "epoch": 28.70036606373816, "grad_norm": 0.4832504391670227, "learning_rate": 0.00016961491854046047, "loss": 2.9211, "step": 106650 }, { "epoch": 28.713824289405686, "grad_norm": 0.48080530762672424, "learning_rate": 0.0001694129527400027, "loss": 2.9281, "step": 106700 }, { "epoch": 28.727282515073213, "grad_norm": 0.48052066564559937, "learning_rate": 0.0001692109869395449, "loss": 2.9356, "step": 106750 }, { "epoch": 28.74074074074074, "grad_norm": 0.4884222149848938, "learning_rate": 0.00016900902113908708, "loss": 2.9386, "step": 106800 }, { "epoch": 28.754198966408268, "grad_norm": 0.4646649658679962, "learning_rate": 0.0001688070553386293, "loss": 2.929, "step": 106850 }, { "epoch": 28.767657192075795, "grad_norm": 0.49423521757125854, "learning_rate": 0.0001686050895381715, "loss": 2.9336, "step": 106900 }, { "epoch": 28.781115417743326, "grad_norm": 0.47869858145713806, "learning_rate": 0.00016840312373771373, "loss": 2.9254, "step": 106950 }, { "epoch": 28.794573643410853, "grad_norm": 0.48208820819854736, "learning_rate": 0.00016820115793725595, "loss": 2.9363, "step": 107000 }, { "epoch": 28.794573643410853, "eval_accuracy": 0.3979914673105514, "eval_loss": 3.278270721435547, "eval_runtime": 84.5483, "eval_samples_per_second": 213.038, "eval_steps_per_second": 13.318, "step": 107000 }, { "epoch": 28.80803186907838, "grad_norm": 0.5072889924049377, "learning_rate": 0.00016799919213679817, "loss": 2.9372, "step": 107050 }, { "epoch": 28.821490094745908, "grad_norm": 0.4976736307144165, "learning_rate": 0.00016779722633634037, "loss": 2.9245, "step": 107100 }, { "epoch": 28.834948320413435, "grad_norm": 0.46875888109207153, "learning_rate": 0.0001675952605358826, "loss": 2.9375, "step": 107150 }, { "epoch": 28.848406546080966, "grad_norm": 0.5036534667015076, "learning_rate": 0.0001673932947354248, "loss": 2.9346, "step": 107200 }, { "epoch": 28.861864771748493, "grad_norm": 0.5005677938461304, "learning_rate": 0.00016719132893496698, "loss": 2.9357, "step": 107250 }, { "epoch": 28.87532299741602, "grad_norm": 0.47183048725128174, "learning_rate": 0.0001669893631345092, "loss": 2.9305, "step": 107300 }, { "epoch": 28.888781223083548, "grad_norm": 0.47492411732673645, "learning_rate": 0.0001667873973340514, "loss": 2.9351, "step": 107350 }, { "epoch": 28.902239448751075, "grad_norm": 0.5148754715919495, "learning_rate": 0.00016658543153359363, "loss": 2.9298, "step": 107400 }, { "epoch": 28.915697674418606, "grad_norm": 0.48464518785476685, "learning_rate": 0.00016638346573313585, "loss": 2.9515, "step": 107450 }, { "epoch": 28.929155900086133, "grad_norm": 0.5281347036361694, "learning_rate": 0.00016618149993267807, "loss": 2.9466, "step": 107500 }, { "epoch": 28.94261412575366, "grad_norm": 0.5068103075027466, "learning_rate": 0.00016597953413222027, "loss": 2.9318, "step": 107550 }, { "epoch": 28.956072351421188, "grad_norm": 0.4937821626663208, "learning_rate": 0.00016577756833176247, "loss": 2.9267, "step": 107600 }, { "epoch": 28.969530577088715, "grad_norm": 0.47072193026542664, "learning_rate": 0.0001655756025313047, "loss": 2.939, "step": 107650 }, { "epoch": 28.982988802756246, "grad_norm": 0.482469767332077, "learning_rate": 0.00016537363673084688, "loss": 2.9345, "step": 107700 }, { "epoch": 28.996447028423773, "grad_norm": 0.4808470606803894, "learning_rate": 0.0001651716709303891, "loss": 2.9333, "step": 107750 }, { "epoch": 29.00968992248062, "grad_norm": 0.5030806660652161, "learning_rate": 0.0001649697051299313, "loss": 2.89, "step": 107800 }, { "epoch": 29.02314814814815, "grad_norm": 0.5034016966819763, "learning_rate": 0.0001647677393294735, "loss": 2.8709, "step": 107850 }, { "epoch": 29.036606373815676, "grad_norm": 0.4961067736148834, "learning_rate": 0.00016456577352901575, "loss": 2.8661, "step": 107900 }, { "epoch": 29.050064599483203, "grad_norm": 0.5241340398788452, "learning_rate": 0.00016436380772855797, "loss": 2.8781, "step": 107950 }, { "epoch": 29.06352282515073, "grad_norm": 0.5237793922424316, "learning_rate": 0.00016416184192810017, "loss": 2.8756, "step": 108000 }, { "epoch": 29.06352282515073, "eval_accuracy": 0.3974687672884309, "eval_loss": 3.2890923023223877, "eval_runtime": 84.0275, "eval_samples_per_second": 214.358, "eval_steps_per_second": 13.4, "step": 108000 }, { "epoch": 29.07698105081826, "grad_norm": 0.4945203363895416, "learning_rate": 0.00016395987612764237, "loss": 2.8869, "step": 108050 }, { "epoch": 29.09043927648579, "grad_norm": 0.5141903758049011, "learning_rate": 0.0001637579103271846, "loss": 2.8779, "step": 108100 }, { "epoch": 29.103897502153316, "grad_norm": 0.5113682746887207, "learning_rate": 0.00016355594452672679, "loss": 2.8839, "step": 108150 }, { "epoch": 29.117355727820843, "grad_norm": 0.4902428388595581, "learning_rate": 0.000163353978726269, "loss": 2.8777, "step": 108200 }, { "epoch": 29.13081395348837, "grad_norm": 0.5147128105163574, "learning_rate": 0.0001631520129258112, "loss": 2.8808, "step": 108250 }, { "epoch": 29.1442721791559, "grad_norm": 0.5210860967636108, "learning_rate": 0.0001629500471253534, "loss": 2.8876, "step": 108300 }, { "epoch": 29.15773040482343, "grad_norm": 0.4885682463645935, "learning_rate": 0.00016274808132489565, "loss": 2.8836, "step": 108350 }, { "epoch": 29.171188630490956, "grad_norm": 0.46739473938941956, "learning_rate": 0.00016254611552443785, "loss": 2.8854, "step": 108400 }, { "epoch": 29.184646856158484, "grad_norm": 0.5281350016593933, "learning_rate": 0.00016234414972398007, "loss": 2.8739, "step": 108450 }, { "epoch": 29.19810508182601, "grad_norm": 0.516180694103241, "learning_rate": 0.00016214218392352227, "loss": 2.8952, "step": 108500 }, { "epoch": 29.21156330749354, "grad_norm": 0.5106911063194275, "learning_rate": 0.0001619402181230645, "loss": 2.886, "step": 108550 }, { "epoch": 29.22502153316107, "grad_norm": 0.5003798604011536, "learning_rate": 0.00016173825232260669, "loss": 2.8872, "step": 108600 }, { "epoch": 29.238479758828596, "grad_norm": 0.49094250798225403, "learning_rate": 0.00016153628652214888, "loss": 2.8918, "step": 108650 }, { "epoch": 29.251937984496124, "grad_norm": 0.48865559697151184, "learning_rate": 0.0001613343207216911, "loss": 2.895, "step": 108700 }, { "epoch": 29.26539621016365, "grad_norm": 0.46626487374305725, "learning_rate": 0.0001611323549212333, "loss": 2.9045, "step": 108750 }, { "epoch": 29.27885443583118, "grad_norm": 0.457952082157135, "learning_rate": 0.00016093038912077555, "loss": 2.9005, "step": 108800 }, { "epoch": 29.29231266149871, "grad_norm": 0.49217286705970764, "learning_rate": 0.00016072842332031775, "loss": 2.9044, "step": 108850 }, { "epoch": 29.305770887166236, "grad_norm": 0.4805237054824829, "learning_rate": 0.00016052645751985997, "loss": 2.8909, "step": 108900 }, { "epoch": 29.319229112833764, "grad_norm": 0.4833526909351349, "learning_rate": 0.00016032449171940217, "loss": 2.8915, "step": 108950 }, { "epoch": 29.33268733850129, "grad_norm": 0.48899632692337036, "learning_rate": 0.0001601225259189444, "loss": 2.8986, "step": 109000 }, { "epoch": 29.33268733850129, "eval_accuracy": 0.39724647582194605, "eval_loss": 3.2852847576141357, "eval_runtime": 84.1715, "eval_samples_per_second": 213.992, "eval_steps_per_second": 13.377, "step": 109000 }, { "epoch": 29.34614556416882, "grad_norm": 0.4914514124393463, "learning_rate": 0.00015992056011848659, "loss": 2.8909, "step": 109050 }, { "epoch": 29.35960378983635, "grad_norm": 0.519985020160675, "learning_rate": 0.00015971859431802878, "loss": 2.9079, "step": 109100 }, { "epoch": 29.373062015503876, "grad_norm": 0.5080176591873169, "learning_rate": 0.000159516628517571, "loss": 2.8982, "step": 109150 }, { "epoch": 29.386520241171404, "grad_norm": 0.49518075585365295, "learning_rate": 0.0001593146627171132, "loss": 2.9058, "step": 109200 }, { "epoch": 29.39997846683893, "grad_norm": 0.5173744559288025, "learning_rate": 0.00015911269691665545, "loss": 2.9055, "step": 109250 }, { "epoch": 29.41343669250646, "grad_norm": 0.5047296285629272, "learning_rate": 0.00015891073111619765, "loss": 2.9116, "step": 109300 }, { "epoch": 29.42689491817399, "grad_norm": 0.5084612369537354, "learning_rate": 0.00015870876531573987, "loss": 2.9139, "step": 109350 }, { "epoch": 29.440353143841516, "grad_norm": 0.4801039397716522, "learning_rate": 0.00015850679951528207, "loss": 2.905, "step": 109400 }, { "epoch": 29.453811369509044, "grad_norm": 0.5018198490142822, "learning_rate": 0.00015830483371482426, "loss": 2.9154, "step": 109450 }, { "epoch": 29.46726959517657, "grad_norm": 0.4790503680706024, "learning_rate": 0.00015810286791436649, "loss": 2.9117, "step": 109500 }, { "epoch": 29.4807278208441, "grad_norm": 0.5301246047019958, "learning_rate": 0.00015790090211390868, "loss": 2.9013, "step": 109550 }, { "epoch": 29.49418604651163, "grad_norm": 0.5275840163230896, "learning_rate": 0.0001576989363134509, "loss": 2.9033, "step": 109600 }, { "epoch": 29.507644272179157, "grad_norm": 0.49783873558044434, "learning_rate": 0.0001574969705129931, "loss": 2.9009, "step": 109650 }, { "epoch": 29.521102497846684, "grad_norm": 0.4843440055847168, "learning_rate": 0.00015729500471253535, "loss": 2.9055, "step": 109700 }, { "epoch": 29.53456072351421, "grad_norm": 0.5092455148696899, "learning_rate": 0.00015709303891207755, "loss": 2.923, "step": 109750 }, { "epoch": 29.54801894918174, "grad_norm": 0.4970625638961792, "learning_rate": 0.00015689107311161977, "loss": 2.9098, "step": 109800 }, { "epoch": 29.56147717484927, "grad_norm": 0.5609508156776428, "learning_rate": 0.00015668910731116197, "loss": 2.9101, "step": 109850 }, { "epoch": 29.574935400516797, "grad_norm": 0.49250948429107666, "learning_rate": 0.00015648714151070416, "loss": 2.918, "step": 109900 }, { "epoch": 29.588393626184324, "grad_norm": 0.4990371763706207, "learning_rate": 0.0001562851757102464, "loss": 2.9178, "step": 109950 }, { "epoch": 29.60185185185185, "grad_norm": 0.5044280886650085, "learning_rate": 0.00015608320990978858, "loss": 2.9028, "step": 110000 }, { "epoch": 29.60185185185185, "eval_accuracy": 0.39783610230709426, "eval_loss": 3.280951976776123, "eval_runtime": 84.2722, "eval_samples_per_second": 213.736, "eval_steps_per_second": 13.361, "step": 110000 }, { "epoch": 29.61531007751938, "grad_norm": 0.5069197416305542, "learning_rate": 0.0001558812441093308, "loss": 2.9157, "step": 110050 }, { "epoch": 29.62876830318691, "grad_norm": 0.5038443207740784, "learning_rate": 0.000155679278308873, "loss": 2.9223, "step": 110100 }, { "epoch": 29.642226528854437, "grad_norm": 0.48724475502967834, "learning_rate": 0.00015547731250841525, "loss": 2.9161, "step": 110150 }, { "epoch": 29.655684754521964, "grad_norm": 0.5094838738441467, "learning_rate": 0.00015527534670795745, "loss": 2.9074, "step": 110200 }, { "epoch": 29.66914298018949, "grad_norm": 0.47840720415115356, "learning_rate": 0.00015507338090749964, "loss": 2.9163, "step": 110250 }, { "epoch": 29.68260120585702, "grad_norm": 0.4904327392578125, "learning_rate": 0.00015487141510704187, "loss": 2.9138, "step": 110300 }, { "epoch": 29.69605943152455, "grad_norm": 0.47974318265914917, "learning_rate": 0.00015466944930658406, "loss": 2.9062, "step": 110350 }, { "epoch": 29.709517657192077, "grad_norm": 0.49263542890548706, "learning_rate": 0.0001544674835061263, "loss": 2.9131, "step": 110400 }, { "epoch": 29.722975882859604, "grad_norm": 0.5041084289550781, "learning_rate": 0.00015426551770566848, "loss": 2.9162, "step": 110450 }, { "epoch": 29.73643410852713, "grad_norm": 0.4936414957046509, "learning_rate": 0.00015406355190521068, "loss": 2.9207, "step": 110500 }, { "epoch": 29.74989233419466, "grad_norm": 0.49592894315719604, "learning_rate": 0.0001538615861047529, "loss": 2.9263, "step": 110550 }, { "epoch": 29.76335055986219, "grad_norm": 0.5023185610771179, "learning_rate": 0.00015365962030429513, "loss": 2.9115, "step": 110600 }, { "epoch": 29.776808785529717, "grad_norm": 0.4932989776134491, "learning_rate": 0.00015345765450383735, "loss": 2.9321, "step": 110650 }, { "epoch": 29.790267011197244, "grad_norm": 0.5014770030975342, "learning_rate": 0.00015325568870337955, "loss": 2.9144, "step": 110700 }, { "epoch": 29.80372523686477, "grad_norm": 0.5211611390113831, "learning_rate": 0.00015305372290292177, "loss": 2.9225, "step": 110750 }, { "epoch": 29.8171834625323, "grad_norm": 0.4848087728023529, "learning_rate": 0.00015285175710246396, "loss": 2.9207, "step": 110800 }, { "epoch": 29.830641688199826, "grad_norm": 0.5076679587364197, "learning_rate": 0.0001526497913020062, "loss": 2.9128, "step": 110850 }, { "epoch": 29.844099913867357, "grad_norm": 0.4708443582057953, "learning_rate": 0.00015244782550154838, "loss": 2.9207, "step": 110900 }, { "epoch": 29.857558139534884, "grad_norm": 0.4948192834854126, "learning_rate": 0.00015224585970109058, "loss": 2.9277, "step": 110950 }, { "epoch": 29.87101636520241, "grad_norm": 0.5008178353309631, "learning_rate": 0.00015204389390063283, "loss": 2.9246, "step": 111000 }, { "epoch": 29.87101636520241, "eval_accuracy": 0.3981930072276234, "eval_loss": 3.2754876613616943, "eval_runtime": 84.2371, "eval_samples_per_second": 213.825, "eval_steps_per_second": 13.367, "step": 111000 }, { "epoch": 29.88447459086994, "grad_norm": 0.5242527723312378, "learning_rate": 0.00015184192810017503, "loss": 2.9257, "step": 111050 }, { "epoch": 29.897932816537466, "grad_norm": Infinity, "learning_rate": 0.00015163996229971725, "loss": 2.9271, "step": 111100 }, { "epoch": 29.911391042204997, "grad_norm": 0.5184350609779358, "learning_rate": 0.00015143799649925945, "loss": 2.9267, "step": 111150 }, { "epoch": 29.924849267872524, "grad_norm": 0.540956437587738, "learning_rate": 0.00015123603069880167, "loss": 2.9265, "step": 111200 }, { "epoch": 29.93830749354005, "grad_norm": 0.5077881217002869, "learning_rate": 0.00015103406489834386, "loss": 2.9124, "step": 111250 }, { "epoch": 29.95176571920758, "grad_norm": 0.4702873229980469, "learning_rate": 0.00015083209909788606, "loss": 2.9345, "step": 111300 }, { "epoch": 29.965223944875106, "grad_norm": 0.48382219672203064, "learning_rate": 0.00015063013329742828, "loss": 2.9227, "step": 111350 }, { "epoch": 29.978682170542637, "grad_norm": 0.4831846058368683, "learning_rate": 0.00015042816749697048, "loss": 2.9184, "step": 111400 }, { "epoch": 29.992140396210164, "grad_norm": 0.4870881736278534, "learning_rate": 0.00015022620169651273, "loss": 2.9305, "step": 111450 }, { "epoch": 30.005383290267012, "grad_norm": 0.5071820020675659, "learning_rate": 0.00015002423589605493, "loss": 2.8904, "step": 111500 }, { "epoch": 30.01884151593454, "grad_norm": 0.5172536373138428, "learning_rate": 0.00014982227009559712, "loss": 2.8505, "step": 111550 }, { "epoch": 30.032299741602067, "grad_norm": 0.5182663202285767, "learning_rate": 0.00014962030429513935, "loss": 2.8645, "step": 111600 }, { "epoch": 30.045757967269594, "grad_norm": 0.526090681552887, "learning_rate": 0.00014941833849468154, "loss": 2.8609, "step": 111650 }, { "epoch": 30.05921619293712, "grad_norm": 0.485584557056427, "learning_rate": 0.00014921637269422377, "loss": 2.8625, "step": 111700 }, { "epoch": 30.072674418604652, "grad_norm": 0.5050146579742432, "learning_rate": 0.00014901440689376596, "loss": 2.8641, "step": 111750 }, { "epoch": 30.08613264427218, "grad_norm": 0.4776037335395813, "learning_rate": 0.00014881244109330818, "loss": 2.8723, "step": 111800 }, { "epoch": 30.099590869939707, "grad_norm": 0.5164368152618408, "learning_rate": 0.0001486104752928504, "loss": 2.8731, "step": 111850 }, { "epoch": 30.113049095607234, "grad_norm": 0.5111197829246521, "learning_rate": 0.0001484085094923926, "loss": 2.8712, "step": 111900 }, { "epoch": 30.12650732127476, "grad_norm": 0.5170813798904419, "learning_rate": 0.00014820654369193483, "loss": 2.8597, "step": 111950 }, { "epoch": 30.139965546942292, "grad_norm": 0.5497874021530151, "learning_rate": 0.00014800457789147702, "loss": 2.8615, "step": 112000 }, { "epoch": 30.139965546942292, "eval_accuracy": 0.3975050553381894, "eval_loss": 3.2868740558624268, "eval_runtime": 84.2131, "eval_samples_per_second": 213.886, "eval_steps_per_second": 13.371, "step": 112000 }, { "epoch": 30.15342377260982, "grad_norm": 0.47900107502937317, "learning_rate": 0.00014780261209101925, "loss": 2.8826, "step": 112050 }, { "epoch": 30.166881998277347, "grad_norm": 0.5066820979118347, "learning_rate": 0.00014760064629056144, "loss": 2.8738, "step": 112100 }, { "epoch": 30.180340223944874, "grad_norm": 0.5189939737319946, "learning_rate": 0.00014739868049010367, "loss": 2.8772, "step": 112150 }, { "epoch": 30.1937984496124, "grad_norm": 0.5010428428649902, "learning_rate": 0.00014719671468964586, "loss": 2.8769, "step": 112200 }, { "epoch": 30.207256675279933, "grad_norm": 0.5018505454063416, "learning_rate": 0.00014699474888918808, "loss": 2.8781, "step": 112250 }, { "epoch": 30.22071490094746, "grad_norm": 0.5155407190322876, "learning_rate": 0.0001467927830887303, "loss": 2.871, "step": 112300 }, { "epoch": 30.234173126614987, "grad_norm": 0.49259355664253235, "learning_rate": 0.0001465908172882725, "loss": 2.8838, "step": 112350 }, { "epoch": 30.247631352282514, "grad_norm": 0.4987989664077759, "learning_rate": 0.0001463888514878147, "loss": 2.8695, "step": 112400 }, { "epoch": 30.26108957795004, "grad_norm": 0.4886040985584259, "learning_rate": 0.00014618688568735692, "loss": 2.895, "step": 112450 }, { "epoch": 30.274547803617573, "grad_norm": 0.5024857521057129, "learning_rate": 0.00014598491988689915, "loss": 2.878, "step": 112500 }, { "epoch": 30.2880060292851, "grad_norm": 0.5005546808242798, "learning_rate": 0.00014578295408644134, "loss": 2.8905, "step": 112550 }, { "epoch": 30.301464254952627, "grad_norm": 0.49909210205078125, "learning_rate": 0.00014558098828598357, "loss": 2.888, "step": 112600 }, { "epoch": 30.314922480620154, "grad_norm": 0.5091086626052856, "learning_rate": 0.00014537902248552576, "loss": 2.8973, "step": 112650 }, { "epoch": 30.328380706287682, "grad_norm": 0.5131893754005432, "learning_rate": 0.00014517705668506799, "loss": 2.894, "step": 112700 }, { "epoch": 30.341838931955213, "grad_norm": 0.4898597300052643, "learning_rate": 0.0001449750908846102, "loss": 2.8916, "step": 112750 }, { "epoch": 30.35529715762274, "grad_norm": 0.4944542646408081, "learning_rate": 0.0001447731250841524, "loss": 2.8834, "step": 112800 }, { "epoch": 30.368755383290267, "grad_norm": 0.5082573890686035, "learning_rate": 0.0001445711592836946, "loss": 2.8951, "step": 112850 }, { "epoch": 30.382213608957795, "grad_norm": 0.49803459644317627, "learning_rate": 0.00014436919348323682, "loss": 2.9, "step": 112900 }, { "epoch": 30.395671834625322, "grad_norm": 0.4845969080924988, "learning_rate": 0.00014416722768277905, "loss": 2.8976, "step": 112950 }, { "epoch": 30.409130060292853, "grad_norm": 0.4821034073829651, "learning_rate": 0.00014396526188232124, "loss": 2.8963, "step": 113000 }, { "epoch": 30.409130060292853, "eval_accuracy": 0.3978802129304534, "eval_loss": 3.284929037094116, "eval_runtime": 84.2378, "eval_samples_per_second": 213.823, "eval_steps_per_second": 13.367, "step": 113000 }, { "epoch": 30.42258828596038, "grad_norm": 0.5150105953216553, "learning_rate": 0.00014376329608186347, "loss": 2.8899, "step": 113050 }, { "epoch": 30.436046511627907, "grad_norm": 0.5295490026473999, "learning_rate": 0.00014356133028140566, "loss": 2.8918, "step": 113100 }, { "epoch": 30.449504737295435, "grad_norm": 0.5084060430526733, "learning_rate": 0.00014335936448094789, "loss": 2.8944, "step": 113150 }, { "epoch": 30.462962962962962, "grad_norm": 0.4947968125343323, "learning_rate": 0.00014315739868049008, "loss": 2.9049, "step": 113200 }, { "epoch": 30.47642118863049, "grad_norm": 0.5141030550003052, "learning_rate": 0.0001429554328800323, "loss": 2.8873, "step": 113250 }, { "epoch": 30.48987941429802, "grad_norm": 0.5051911473274231, "learning_rate": 0.0001427534670795745, "loss": 2.8991, "step": 113300 }, { "epoch": 30.503337639965547, "grad_norm": 0.5075148940086365, "learning_rate": 0.00014255150127911672, "loss": 2.8984, "step": 113350 }, { "epoch": 30.516795865633075, "grad_norm": 0.5077381134033203, "learning_rate": 0.00014234953547865895, "loss": 2.9003, "step": 113400 }, { "epoch": 30.530254091300602, "grad_norm": 0.501668393611908, "learning_rate": 0.00014214756967820114, "loss": 2.8974, "step": 113450 }, { "epoch": 30.54371231696813, "grad_norm": 0.509915292263031, "learning_rate": 0.00014194560387774334, "loss": 2.9069, "step": 113500 }, { "epoch": 30.55717054263566, "grad_norm": 0.5007650852203369, "learning_rate": 0.00014174363807728556, "loss": 2.9017, "step": 113550 }, { "epoch": 30.570628768303187, "grad_norm": 0.5104435682296753, "learning_rate": 0.00014154167227682779, "loss": 2.8916, "step": 113600 }, { "epoch": 30.584086993970715, "grad_norm": 0.5037990212440491, "learning_rate": 0.00014133970647636998, "loss": 2.9098, "step": 113650 }, { "epoch": 30.597545219638242, "grad_norm": 0.5024341940879822, "learning_rate": 0.0001411377406759122, "loss": 2.9025, "step": 113700 }, { "epoch": 30.61100344530577, "grad_norm": 0.5257053375244141, "learning_rate": 0.0001409357748754544, "loss": 2.8987, "step": 113750 }, { "epoch": 30.6244616709733, "grad_norm": 0.46181410551071167, "learning_rate": 0.00014073380907499662, "loss": 2.9109, "step": 113800 }, { "epoch": 30.637919896640827, "grad_norm": 0.5127111673355103, "learning_rate": 0.00014053184327453885, "loss": 2.9069, "step": 113850 }, { "epoch": 30.651378122308355, "grad_norm": 0.4897187352180481, "learning_rate": 0.00014032987747408104, "loss": 2.9048, "step": 113900 }, { "epoch": 30.664836347975882, "grad_norm": 0.5005426406860352, "learning_rate": 0.00014012791167362324, "loss": 2.9008, "step": 113950 }, { "epoch": 30.67829457364341, "grad_norm": 0.5037729740142822, "learning_rate": 0.00013992594587316546, "loss": 2.918, "step": 114000 }, { "epoch": 30.67829457364341, "eval_accuracy": 0.3980335136436548, "eval_loss": 3.278815984725952, "eval_runtime": 84.3786, "eval_samples_per_second": 213.466, "eval_steps_per_second": 13.345, "step": 114000 }, { "epoch": 30.69175279931094, "grad_norm": 0.523506224155426, "learning_rate": 0.0001397239800727077, "loss": 2.9053, "step": 114050 }, { "epoch": 30.705211024978468, "grad_norm": 0.4961335062980652, "learning_rate": 0.00013952201427224988, "loss": 2.905, "step": 114100 }, { "epoch": 30.718669250645995, "grad_norm": 0.5052204728126526, "learning_rate": 0.0001393200484717921, "loss": 2.9071, "step": 114150 }, { "epoch": 30.732127476313522, "grad_norm": 0.5077371001243591, "learning_rate": 0.0001391180826713343, "loss": 2.9145, "step": 114200 }, { "epoch": 30.74558570198105, "grad_norm": 0.5354120135307312, "learning_rate": 0.00013891611687087653, "loss": 2.9087, "step": 114250 }, { "epoch": 30.75904392764858, "grad_norm": 0.5553905367851257, "learning_rate": 0.00013871415107041872, "loss": 2.9169, "step": 114300 }, { "epoch": 30.772502153316108, "grad_norm": 0.5040802359580994, "learning_rate": 0.00013851218526996094, "loss": 2.9133, "step": 114350 }, { "epoch": 30.785960378983635, "grad_norm": 0.49840259552001953, "learning_rate": 0.00013831021946950314, "loss": 2.9128, "step": 114400 }, { "epoch": 30.799418604651162, "grad_norm": 0.5111094117164612, "learning_rate": 0.00013810825366904536, "loss": 2.9148, "step": 114450 }, { "epoch": 30.81287683031869, "grad_norm": 0.5349999070167542, "learning_rate": 0.0001379062878685876, "loss": 2.9149, "step": 114500 }, { "epoch": 30.82633505598622, "grad_norm": 0.4961295425891876, "learning_rate": 0.00013770432206812978, "loss": 2.9145, "step": 114550 }, { "epoch": 30.839793281653748, "grad_norm": 0.5046606659889221, "learning_rate": 0.00013750235626767198, "loss": 2.9172, "step": 114600 }, { "epoch": 30.853251507321275, "grad_norm": 0.5328389406204224, "learning_rate": 0.0001373003904672142, "loss": 2.9154, "step": 114650 }, { "epoch": 30.866709732988802, "grad_norm": 0.4973182678222656, "learning_rate": 0.00013709842466675643, "loss": 2.9158, "step": 114700 }, { "epoch": 30.88016795865633, "grad_norm": 0.5038853287696838, "learning_rate": 0.00013689645886629862, "loss": 2.9113, "step": 114750 }, { "epoch": 30.89362618432386, "grad_norm": 0.5285792946815491, "learning_rate": 0.00013669449306584084, "loss": 2.91, "step": 114800 }, { "epoch": 30.907084409991388, "grad_norm": 0.502298891544342, "learning_rate": 0.00013649252726538304, "loss": 2.9152, "step": 114850 }, { "epoch": 30.920542635658915, "grad_norm": 0.5127186179161072, "learning_rate": 0.00013629056146492526, "loss": 2.9186, "step": 114900 }, { "epoch": 30.934000861326442, "grad_norm": 0.5167381763458252, "learning_rate": 0.0001360885956644675, "loss": 2.9176, "step": 114950 }, { "epoch": 30.94745908699397, "grad_norm": 0.5052900910377502, "learning_rate": 0.00013588662986400968, "loss": 2.9152, "step": 115000 }, { "epoch": 30.94745908699397, "eval_accuracy": 0.39822255917233695, "eval_loss": 3.2754769325256348, "eval_runtime": 84.3993, "eval_samples_per_second": 213.414, "eval_steps_per_second": 13.341, "step": 115000 }, { "epoch": 30.9609173126615, "grad_norm": 0.501179039478302, "learning_rate": 0.00013568466406355188, "loss": 2.9281, "step": 115050 }, { "epoch": 30.974375538329028, "grad_norm": 0.48699599504470825, "learning_rate": 0.0001354826982630941, "loss": 2.9207, "step": 115100 }, { "epoch": 30.987833763996555, "grad_norm": 0.5128811597824097, "learning_rate": 0.00013528073246263633, "loss": 2.9114, "step": 115150 }, { "epoch": 31.001076658053403, "grad_norm": 0.5467365384101868, "learning_rate": 0.00013507876666217852, "loss": 2.9059, "step": 115200 }, { "epoch": 31.01453488372093, "grad_norm": 0.5502208471298218, "learning_rate": 0.00013487680086172075, "loss": 2.849, "step": 115250 }, { "epoch": 31.027993109388458, "grad_norm": 0.5167126655578613, "learning_rate": 0.00013467483506126294, "loss": 2.8426, "step": 115300 }, { "epoch": 31.041451335055985, "grad_norm": 0.5184963941574097, "learning_rate": 0.00013447286926080516, "loss": 2.8467, "step": 115350 }, { "epoch": 31.054909560723516, "grad_norm": 0.5206332206726074, "learning_rate": 0.00013427090346034736, "loss": 2.8499, "step": 115400 }, { "epoch": 31.068367786391043, "grad_norm": 0.5013174414634705, "learning_rate": 0.00013406893765988958, "loss": 2.8558, "step": 115450 }, { "epoch": 31.08182601205857, "grad_norm": 0.5105743408203125, "learning_rate": 0.00013386697185943178, "loss": 2.86, "step": 115500 }, { "epoch": 31.095284237726098, "grad_norm": 0.511475145816803, "learning_rate": 0.000133665006058974, "loss": 2.8515, "step": 115550 }, { "epoch": 31.108742463393625, "grad_norm": 0.5447846055030823, "learning_rate": 0.00013346304025851623, "loss": 2.8609, "step": 115600 }, { "epoch": 31.122200689061152, "grad_norm": 0.5339901447296143, "learning_rate": 0.00013326107445805842, "loss": 2.8659, "step": 115650 }, { "epoch": 31.135658914728683, "grad_norm": 0.49852365255355835, "learning_rate": 0.00013305910865760065, "loss": 2.8626, "step": 115700 }, { "epoch": 31.14911714039621, "grad_norm": 0.5074359774589539, "learning_rate": 0.00013285714285714284, "loss": 2.8682, "step": 115750 }, { "epoch": 31.162575366063738, "grad_norm": 0.4895899295806885, "learning_rate": 0.00013265517705668506, "loss": 2.8582, "step": 115800 }, { "epoch": 31.176033591731265, "grad_norm": 0.5205885171890259, "learning_rate": 0.00013245321125622726, "loss": 2.8695, "step": 115850 }, { "epoch": 31.189491817398793, "grad_norm": 0.47917288541793823, "learning_rate": 0.00013225124545576948, "loss": 2.8703, "step": 115900 }, { "epoch": 31.202950043066323, "grad_norm": 0.5210146307945251, "learning_rate": 0.00013204927965531168, "loss": 2.8712, "step": 115950 }, { "epoch": 31.21640826873385, "grad_norm": 0.5375768542289734, "learning_rate": 0.0001318473138548539, "loss": 2.8768, "step": 116000 }, { "epoch": 31.21640826873385, "eval_accuracy": 0.39771593888483997, "eval_loss": 3.28580641746521, "eval_runtime": 84.3527, "eval_samples_per_second": 213.532, "eval_steps_per_second": 13.349, "step": 116000 }, { "epoch": 31.229866494401378, "grad_norm": 0.5357069969177246, "learning_rate": 0.00013164534805439613, "loss": 2.885, "step": 116050 }, { "epoch": 31.243324720068905, "grad_norm": 0.5172284245491028, "learning_rate": 0.00013144338225393832, "loss": 2.8737, "step": 116100 }, { "epoch": 31.256782945736433, "grad_norm": 0.5309697985649109, "learning_rate": 0.00013124141645348052, "loss": 2.876, "step": 116150 }, { "epoch": 31.270241171403963, "grad_norm": 0.5113084316253662, "learning_rate": 0.00013103945065302274, "loss": 2.8656, "step": 116200 }, { "epoch": 31.28369939707149, "grad_norm": 0.5179013013839722, "learning_rate": 0.00013083748485256497, "loss": 2.8736, "step": 116250 }, { "epoch": 31.297157622739018, "grad_norm": 0.516130268573761, "learning_rate": 0.00013063551905210716, "loss": 2.8765, "step": 116300 }, { "epoch": 31.310615848406545, "grad_norm": 0.5475269556045532, "learning_rate": 0.00013043355325164938, "loss": 2.8789, "step": 116350 }, { "epoch": 31.324074074074073, "grad_norm": 0.5318151712417603, "learning_rate": 0.00013023158745119158, "loss": 2.8932, "step": 116400 }, { "epoch": 31.337532299741603, "grad_norm": 0.5311176776885986, "learning_rate": 0.00013002962165073378, "loss": 2.8654, "step": 116450 }, { "epoch": 31.35099052540913, "grad_norm": 0.5081773400306702, "learning_rate": 0.000129827655850276, "loss": 2.8748, "step": 116500 }, { "epoch": 31.364448751076658, "grad_norm": 0.5100054740905762, "learning_rate": 0.00012962569004981822, "loss": 2.8837, "step": 116550 }, { "epoch": 31.377906976744185, "grad_norm": 0.5171526670455933, "learning_rate": 0.00012942372424936042, "loss": 2.88, "step": 116600 }, { "epoch": 31.391365202411713, "grad_norm": 0.5279509425163269, "learning_rate": 0.00012922175844890264, "loss": 2.892, "step": 116650 }, { "epoch": 31.404823428079244, "grad_norm": 0.5070922374725342, "learning_rate": 0.00012901979264844487, "loss": 2.88, "step": 116700 }, { "epoch": 31.41828165374677, "grad_norm": 0.47861653566360474, "learning_rate": 0.00012881782684798706, "loss": 2.8934, "step": 116750 }, { "epoch": 31.431739879414298, "grad_norm": 0.490649938583374, "learning_rate": 0.00012861586104752928, "loss": 2.888, "step": 116800 }, { "epoch": 31.445198105081825, "grad_norm": 0.5430272817611694, "learning_rate": 0.00012841389524707148, "loss": 2.8829, "step": 116850 }, { "epoch": 31.458656330749353, "grad_norm": 0.5027969479560852, "learning_rate": 0.0001282119294466137, "loss": 2.8944, "step": 116900 }, { "epoch": 31.472114556416884, "grad_norm": 0.5254911780357361, "learning_rate": 0.0001280099636461559, "loss": 2.8932, "step": 116950 }, { "epoch": 31.48557278208441, "grad_norm": 0.4988119900226593, "learning_rate": 0.00012780799784569812, "loss": 2.8868, "step": 117000 }, { "epoch": 31.48557278208441, "eval_accuracy": 0.39804426968235573, "eval_loss": 3.2833847999572754, "eval_runtime": 84.2746, "eval_samples_per_second": 213.73, "eval_steps_per_second": 13.361, "step": 117000 }, { "epoch": 31.499031007751938, "grad_norm": 0.5362693071365356, "learning_rate": 0.00012760603204524032, "loss": 2.8915, "step": 117050 }, { "epoch": 31.512489233419466, "grad_norm": 0.5097172260284424, "learning_rate": 0.00012740406624478254, "loss": 2.8897, "step": 117100 }, { "epoch": 31.525947459086993, "grad_norm": 0.502714216709137, "learning_rate": 0.00012720210044432477, "loss": 2.9004, "step": 117150 }, { "epoch": 31.539405684754524, "grad_norm": 0.5185081362724304, "learning_rate": 0.00012700013464386696, "loss": 2.8897, "step": 117200 }, { "epoch": 31.55286391042205, "grad_norm": 0.5322837233543396, "learning_rate": 0.00012679816884340916, "loss": 2.8887, "step": 117250 }, { "epoch": 31.56632213608958, "grad_norm": 0.482925146818161, "learning_rate": 0.00012659620304295138, "loss": 2.8929, "step": 117300 }, { "epoch": 31.579780361757106, "grad_norm": 0.4954151213169098, "learning_rate": 0.0001263942372424936, "loss": 2.8746, "step": 117350 }, { "epoch": 31.593238587424633, "grad_norm": 0.49335721135139465, "learning_rate": 0.0001261922714420358, "loss": 2.8896, "step": 117400 }, { "epoch": 31.606696813092164, "grad_norm": 0.5141397714614868, "learning_rate": 0.00012599030564157802, "loss": 2.8872, "step": 117450 }, { "epoch": 31.62015503875969, "grad_norm": 0.5417701601982117, "learning_rate": 0.00012578833984112022, "loss": 2.8953, "step": 117500 }, { "epoch": 31.63361326442722, "grad_norm": 0.529873251914978, "learning_rate": 0.00012558637404066244, "loss": 2.9028, "step": 117550 }, { "epoch": 31.647071490094746, "grad_norm": 0.510009765625, "learning_rate": 0.00012538440824020467, "loss": 2.894, "step": 117600 }, { "epoch": 31.660529715762273, "grad_norm": 0.544947624206543, "learning_rate": 0.00012518244243974686, "loss": 2.9001, "step": 117650 }, { "epoch": 31.6739879414298, "grad_norm": 0.527076244354248, "learning_rate": 0.00012498047663928906, "loss": 2.8974, "step": 117700 }, { "epoch": 31.68744616709733, "grad_norm": 0.5229929089546204, "learning_rate": 0.00012477851083883128, "loss": 2.8987, "step": 117750 }, { "epoch": 31.70090439276486, "grad_norm": 0.5023860931396484, "learning_rate": 0.0001245765450383735, "loss": 2.9021, "step": 117800 }, { "epoch": 31.714362618432386, "grad_norm": 0.5246515274047852, "learning_rate": 0.0001243745792379157, "loss": 2.8945, "step": 117850 }, { "epoch": 31.727820844099913, "grad_norm": 0.5159626007080078, "learning_rate": 0.00012417261343745792, "loss": 2.8991, "step": 117900 }, { "epoch": 31.74127906976744, "grad_norm": 0.5257189273834229, "learning_rate": 0.00012397064763700012, "loss": 2.9054, "step": 117950 }, { "epoch": 31.75473729543497, "grad_norm": 0.5196221470832825, "learning_rate": 0.00012376868183654232, "loss": 2.8962, "step": 118000 }, { "epoch": 31.75473729543497, "eval_accuracy": 0.39826341039002916, "eval_loss": 3.279503583908081, "eval_runtime": 84.0907, "eval_samples_per_second": 214.197, "eval_steps_per_second": 13.39, "step": 118000 }, { "epoch": 31.7681955211025, "grad_norm": 0.5018131732940674, "learning_rate": 0.00012356671603608454, "loss": 2.8978, "step": 118050 }, { "epoch": 31.781653746770026, "grad_norm": 0.47433024644851685, "learning_rate": 0.00012336475023562676, "loss": 2.8956, "step": 118100 }, { "epoch": 31.795111972437553, "grad_norm": 0.5425954461097717, "learning_rate": 0.00012316278443516896, "loss": 2.9015, "step": 118150 }, { "epoch": 31.80857019810508, "grad_norm": 0.4845956265926361, "learning_rate": 0.00012296081863471118, "loss": 2.9057, "step": 118200 }, { "epoch": 31.82202842377261, "grad_norm": 0.5126825571060181, "learning_rate": 0.0001227588528342534, "loss": 2.8968, "step": 118250 }, { "epoch": 31.83548664944014, "grad_norm": 0.5188189148902893, "learning_rate": 0.0001225568870337956, "loss": 2.9073, "step": 118300 }, { "epoch": 31.848944875107666, "grad_norm": 0.5191381573677063, "learning_rate": 0.0001223549212333378, "loss": 2.9082, "step": 118350 }, { "epoch": 31.862403100775193, "grad_norm": 0.4905800223350525, "learning_rate": 0.00012215295543288002, "loss": 2.9009, "step": 118400 }, { "epoch": 31.87586132644272, "grad_norm": 0.49753281474113464, "learning_rate": 0.00012195098963242223, "loss": 2.9032, "step": 118450 }, { "epoch": 31.88931955211025, "grad_norm": 0.5176593661308289, "learning_rate": 0.00012174902383196445, "loss": 2.9051, "step": 118500 }, { "epoch": 31.90277777777778, "grad_norm": 0.49819475412368774, "learning_rate": 0.00012154705803150665, "loss": 2.8974, "step": 118550 }, { "epoch": 31.916236003445306, "grad_norm": 0.5007299184799194, "learning_rate": 0.00012134509223104886, "loss": 2.9181, "step": 118600 }, { "epoch": 31.929694229112833, "grad_norm": 0.5035253167152405, "learning_rate": 0.00012114312643059107, "loss": 2.9012, "step": 118650 }, { "epoch": 31.94315245478036, "grad_norm": 0.6211641430854797, "learning_rate": 0.00012094116063013329, "loss": 2.9118, "step": 118700 }, { "epoch": 31.95661068044789, "grad_norm": 0.5061435103416443, "learning_rate": 0.0001207391948296755, "loss": 2.9048, "step": 118750 }, { "epoch": 31.97006890611542, "grad_norm": 0.4980998635292053, "learning_rate": 0.00012053722902921771, "loss": 2.9065, "step": 118800 }, { "epoch": 31.983527131782946, "grad_norm": 0.5427234768867493, "learning_rate": 0.00012033526322875992, "loss": 2.9085, "step": 118850 }, { "epoch": 31.996985357450473, "grad_norm": 0.4822652339935303, "learning_rate": 0.00012013329742830212, "loss": 2.9102, "step": 118900 }, { "epoch": 32.01022825150732, "grad_norm": 0.4894477128982544, "learning_rate": 0.00011993133162784434, "loss": 2.8644, "step": 118950 }, { "epoch": 32.02368647717485, "grad_norm": 0.526383638381958, "learning_rate": 0.00011972936582738655, "loss": 2.847, "step": 119000 }, { "epoch": 32.02368647717485, "eval_accuracy": 0.39809392129534865, "eval_loss": 3.2846009731292725, "eval_runtime": 84.1418, "eval_samples_per_second": 214.067, "eval_steps_per_second": 13.382, "step": 119000 }, { "epoch": 32.037144702842376, "grad_norm": 0.5376343131065369, "learning_rate": 0.00011952740002692876, "loss": 2.8419, "step": 119050 }, { "epoch": 32.0506029285099, "grad_norm": 0.5181412100791931, "learning_rate": 0.00011932543422647097, "loss": 2.8521, "step": 119100 }, { "epoch": 32.06406115417743, "grad_norm": 0.48993155360221863, "learning_rate": 0.00011912346842601319, "loss": 2.8479, "step": 119150 }, { "epoch": 32.07751937984496, "grad_norm": 0.5353118777275085, "learning_rate": 0.0001189215026255554, "loss": 2.8574, "step": 119200 }, { "epoch": 32.09097760551249, "grad_norm": 0.5269080400466919, "learning_rate": 0.00011871953682509761, "loss": 2.8595, "step": 119250 }, { "epoch": 32.10443583118002, "grad_norm": 0.5100624561309814, "learning_rate": 0.00011851757102463981, "loss": 2.8484, "step": 119300 }, { "epoch": 32.11789405684755, "grad_norm": 0.5078878998756409, "learning_rate": 0.00011831560522418203, "loss": 2.8521, "step": 119350 }, { "epoch": 32.131352282515074, "grad_norm": 0.5233724117279053, "learning_rate": 0.00011811363942372424, "loss": 2.8533, "step": 119400 }, { "epoch": 32.1448105081826, "grad_norm": 0.4920700192451477, "learning_rate": 0.00011791167362326645, "loss": 2.8464, "step": 119450 }, { "epoch": 32.15826873385013, "grad_norm": 0.5175674557685852, "learning_rate": 0.00011770970782280866, "loss": 2.8637, "step": 119500 }, { "epoch": 32.171726959517656, "grad_norm": 0.5312711596488953, "learning_rate": 0.00011750774202235087, "loss": 2.8582, "step": 119550 }, { "epoch": 32.18518518518518, "grad_norm": 0.4959535002708435, "learning_rate": 0.00011730577622189309, "loss": 2.8523, "step": 119600 }, { "epoch": 32.19864341085271, "grad_norm": 0.5239533185958862, "learning_rate": 0.00011710381042143529, "loss": 2.8761, "step": 119650 }, { "epoch": 32.21210163652024, "grad_norm": 0.5062153339385986, "learning_rate": 0.0001169018446209775, "loss": 2.8616, "step": 119700 }, { "epoch": 32.22555986218777, "grad_norm": 0.530886709690094, "learning_rate": 0.00011669987882051971, "loss": 2.8666, "step": 119750 }, { "epoch": 32.2390180878553, "grad_norm": 0.5261471271514893, "learning_rate": 0.00011649791302006193, "loss": 2.8676, "step": 119800 }, { "epoch": 32.25247631352283, "grad_norm": 0.5268979668617249, "learning_rate": 0.00011629594721960414, "loss": 2.8731, "step": 119850 }, { "epoch": 32.265934539190354, "grad_norm": 0.49872177839279175, "learning_rate": 0.00011609398141914635, "loss": 2.8635, "step": 119900 }, { "epoch": 32.27939276485788, "grad_norm": 0.5125359892845154, "learning_rate": 0.00011589201561868856, "loss": 2.8658, "step": 119950 }, { "epoch": 32.29285099052541, "grad_norm": 0.510800302028656, "learning_rate": 0.00011569004981823076, "loss": 2.872, "step": 120000 }, { "epoch": 32.29285099052541, "eval_accuracy": 0.3980136312690865, "eval_loss": 3.2843637466430664, "eval_runtime": 83.3793, "eval_samples_per_second": 216.025, "eval_steps_per_second": 13.505, "step": 120000 }, { "epoch": 32.306309216192936, "grad_norm": 0.5112922787666321, "learning_rate": 0.00011548808401777298, "loss": 2.8678, "step": 120050 }, { "epoch": 32.31976744186046, "grad_norm": 0.4980347454547882, "learning_rate": 0.00011528611821731519, "loss": 2.8694, "step": 120100 }, { "epoch": 32.33322566752799, "grad_norm": 0.5203003287315369, "learning_rate": 0.0001150841524168574, "loss": 2.8707, "step": 120150 }, { "epoch": 32.34668389319552, "grad_norm": 0.4873710572719574, "learning_rate": 0.00011488218661639961, "loss": 2.8717, "step": 120200 }, { "epoch": 32.36014211886305, "grad_norm": 0.5280223488807678, "learning_rate": 0.00011468022081594183, "loss": 2.8711, "step": 120250 }, { "epoch": 32.37360034453058, "grad_norm": 0.4919604957103729, "learning_rate": 0.00011447825501548404, "loss": 2.8794, "step": 120300 }, { "epoch": 32.38705857019811, "grad_norm": 0.502147376537323, "learning_rate": 0.00011427628921502625, "loss": 2.8699, "step": 120350 }, { "epoch": 32.400516795865634, "grad_norm": 0.5419111251831055, "learning_rate": 0.00011407432341456845, "loss": 2.877, "step": 120400 }, { "epoch": 32.41397502153316, "grad_norm": 0.5109226703643799, "learning_rate": 0.00011387235761411066, "loss": 2.8717, "step": 120450 }, { "epoch": 32.42743324720069, "grad_norm": 0.5274528861045837, "learning_rate": 0.00011367039181365288, "loss": 2.8865, "step": 120500 }, { "epoch": 32.440891472868216, "grad_norm": 0.5329537391662598, "learning_rate": 0.00011346842601319509, "loss": 2.8804, "step": 120550 }, { "epoch": 32.454349698535744, "grad_norm": 0.5048822164535522, "learning_rate": 0.0001132664602127373, "loss": 2.8793, "step": 120600 }, { "epoch": 32.46780792420327, "grad_norm": 0.49886929988861084, "learning_rate": 0.00011306449441227951, "loss": 2.8818, "step": 120650 }, { "epoch": 32.4812661498708, "grad_norm": 0.5073612928390503, "learning_rate": 0.00011286252861182173, "loss": 2.8853, "step": 120700 }, { "epoch": 32.494724375538325, "grad_norm": 0.5252017378807068, "learning_rate": 0.00011266056281136394, "loss": 2.8712, "step": 120750 }, { "epoch": 32.50818260120586, "grad_norm": 0.49827513098716736, "learning_rate": 0.00011245859701090614, "loss": 2.879, "step": 120800 }, { "epoch": 32.52164082687339, "grad_norm": 0.5000596046447754, "learning_rate": 0.00011225663121044835, "loss": 2.8784, "step": 120850 }, { "epoch": 32.535099052540914, "grad_norm": 0.5500352382659912, "learning_rate": 0.00011205466540999056, "loss": 2.8802, "step": 120900 }, { "epoch": 32.54855727820844, "grad_norm": 0.5427979826927185, "learning_rate": 0.00011185269960953278, "loss": 2.8815, "step": 120950 }, { "epoch": 32.56201550387597, "grad_norm": 0.5111308097839355, "learning_rate": 0.00011165073380907499, "loss": 2.8846, "step": 121000 }, { "epoch": 32.56201550387597, "eval_accuracy": 0.3983339221992905, "eval_loss": 3.2795493602752686, "eval_runtime": 84.28, "eval_samples_per_second": 213.716, "eval_steps_per_second": 13.36, "step": 121000 }, { "epoch": 32.575473729543496, "grad_norm": 0.5127388834953308, "learning_rate": 0.0001114487680086172, "loss": 2.8879, "step": 121050 }, { "epoch": 32.588931955211024, "grad_norm": 0.5021010637283325, "learning_rate": 0.0001112468022081594, "loss": 2.8825, "step": 121100 }, { "epoch": 32.60239018087855, "grad_norm": 0.5289945602416992, "learning_rate": 0.00011104483640770163, "loss": 2.8859, "step": 121150 }, { "epoch": 32.61584840654608, "grad_norm": 0.5273314118385315, "learning_rate": 0.00011084287060724383, "loss": 2.8894, "step": 121200 }, { "epoch": 32.629306632213606, "grad_norm": 0.5341081023216248, "learning_rate": 0.00011064090480678604, "loss": 2.8809, "step": 121250 }, { "epoch": 32.64276485788114, "grad_norm": 0.507257878780365, "learning_rate": 0.00011043893900632825, "loss": 2.8815, "step": 121300 }, { "epoch": 32.65622308354867, "grad_norm": 0.5302804708480835, "learning_rate": 0.00011023697320587046, "loss": 2.8826, "step": 121350 }, { "epoch": 32.669681309216195, "grad_norm": 0.5019773840904236, "learning_rate": 0.00011003500740541268, "loss": 2.8825, "step": 121400 }, { "epoch": 32.68313953488372, "grad_norm": 0.5342096090316772, "learning_rate": 0.00010983304160495489, "loss": 2.8835, "step": 121450 }, { "epoch": 32.69659776055125, "grad_norm": 0.5053223371505737, "learning_rate": 0.00010963107580449709, "loss": 2.8795, "step": 121500 }, { "epoch": 32.71005598621878, "grad_norm": 0.5185446739196777, "learning_rate": 0.0001094291100040393, "loss": 2.8884, "step": 121550 }, { "epoch": 32.723514211886304, "grad_norm": 0.553935170173645, "learning_rate": 0.00010922714420358152, "loss": 2.8833, "step": 121600 }, { "epoch": 32.73697243755383, "grad_norm": 0.535712480545044, "learning_rate": 0.00010902517840312373, "loss": 2.8855, "step": 121650 }, { "epoch": 32.75043066322136, "grad_norm": 0.5135592222213745, "learning_rate": 0.00010882321260266594, "loss": 2.8896, "step": 121700 }, { "epoch": 32.763888888888886, "grad_norm": 0.4964793920516968, "learning_rate": 0.00010862124680220815, "loss": 2.8966, "step": 121750 }, { "epoch": 32.77734711455642, "grad_norm": 0.5169598460197449, "learning_rate": 0.00010841928100175037, "loss": 2.9057, "step": 121800 }, { "epoch": 32.79080534022395, "grad_norm": 0.5452906489372253, "learning_rate": 0.00010821731520129258, "loss": 2.8853, "step": 121850 }, { "epoch": 32.804263565891475, "grad_norm": 0.5410115718841553, "learning_rate": 0.00010801534940083478, "loss": 2.8926, "step": 121900 }, { "epoch": 32.817721791559, "grad_norm": 0.5537077784538269, "learning_rate": 0.00010781338360037699, "loss": 2.8905, "step": 121950 }, { "epoch": 32.83118001722653, "grad_norm": 0.5235624313354492, "learning_rate": 0.0001076114177999192, "loss": 2.8901, "step": 122000 }, { "epoch": 32.83118001722653, "eval_accuracy": 0.3983279466222345, "eval_loss": 3.277186393737793, "eval_runtime": 84.0356, "eval_samples_per_second": 214.338, "eval_steps_per_second": 13.399, "step": 122000 }, { "epoch": 32.84463824289406, "grad_norm": 0.5131085515022278, "learning_rate": 0.00010740945199946142, "loss": 2.8901, "step": 122050 }, { "epoch": 32.858096468561584, "grad_norm": 0.5139198303222656, "learning_rate": 0.00010720748619900363, "loss": 2.8771, "step": 122100 }, { "epoch": 32.87155469422911, "grad_norm": 0.5392455458641052, "learning_rate": 0.00010700552039854584, "loss": 2.8877, "step": 122150 }, { "epoch": 32.88501291989664, "grad_norm": 0.49859973788261414, "learning_rate": 0.00010680355459808805, "loss": 2.9026, "step": 122200 }, { "epoch": 32.898471145564166, "grad_norm": 0.5468138456344604, "learning_rate": 0.00010660158879763027, "loss": 2.8863, "step": 122250 }, { "epoch": 32.9119293712317, "grad_norm": 0.5251834988594055, "learning_rate": 0.00010639962299717247, "loss": 2.8912, "step": 122300 }, { "epoch": 32.92538759689923, "grad_norm": 0.5111126899719238, "learning_rate": 0.00010619765719671468, "loss": 2.8838, "step": 122350 }, { "epoch": 32.938845822566755, "grad_norm": 0.5119268298149109, "learning_rate": 0.00010599569139625689, "loss": 2.8871, "step": 122400 }, { "epoch": 32.95230404823428, "grad_norm": 0.5077241659164429, "learning_rate": 0.0001057937255957991, "loss": 2.8932, "step": 122450 }, { "epoch": 32.96576227390181, "grad_norm": 0.5331259369850159, "learning_rate": 0.00010559175979534132, "loss": 2.8953, "step": 122500 }, { "epoch": 32.97922049956934, "grad_norm": 0.4993937611579895, "learning_rate": 0.00010538979399488353, "loss": 2.8972, "step": 122550 }, { "epoch": 32.992678725236864, "grad_norm": 0.5301192402839661, "learning_rate": 0.00010518782819442573, "loss": 2.894, "step": 122600 }, { "epoch": 33.005921619293716, "grad_norm": 0.5341933369636536, "learning_rate": 0.00010498586239396794, "loss": 2.8617, "step": 122650 }, { "epoch": 33.01937984496124, "grad_norm": 0.5181511044502258, "learning_rate": 0.00010478389659351016, "loss": 2.8456, "step": 122700 }, { "epoch": 33.03283807062877, "grad_norm": 0.5161030888557434, "learning_rate": 0.00010458193079305237, "loss": 2.8368, "step": 122750 }, { "epoch": 33.0462962962963, "grad_norm": 0.5211310386657715, "learning_rate": 0.00010437996499259458, "loss": 2.834, "step": 122800 }, { "epoch": 33.059754521963825, "grad_norm": 0.5173688530921936, "learning_rate": 0.00010417799919213679, "loss": 2.8438, "step": 122850 }, { "epoch": 33.07321274763135, "grad_norm": 0.5314739346504211, "learning_rate": 0.000103976033391679, "loss": 2.8433, "step": 122900 }, { "epoch": 33.08667097329888, "grad_norm": 0.5037255883216858, "learning_rate": 0.00010377406759122122, "loss": 2.8509, "step": 122950 }, { "epoch": 33.10012919896641, "grad_norm": 0.5201254487037659, "learning_rate": 0.00010357210179076342, "loss": 2.8568, "step": 123000 }, { "epoch": 33.10012919896641, "eval_accuracy": 0.39814400749576384, "eval_loss": 3.285991907119751, "eval_runtime": 84.4772, "eval_samples_per_second": 213.217, "eval_steps_per_second": 13.329, "step": 123000 }, { "epoch": 33.113587424633934, "grad_norm": 0.5263310074806213, "learning_rate": 0.00010337013599030563, "loss": 2.8527, "step": 123050 }, { "epoch": 33.12704565030146, "grad_norm": 0.5239158868789673, "learning_rate": 0.00010316817018984784, "loss": 2.8515, "step": 123100 }, { "epoch": 33.14050387596899, "grad_norm": 0.5073361396789551, "learning_rate": 0.00010296620438939006, "loss": 2.8486, "step": 123150 }, { "epoch": 33.15396210163652, "grad_norm": 0.5051706433296204, "learning_rate": 0.00010276423858893227, "loss": 2.8489, "step": 123200 }, { "epoch": 33.16742032730405, "grad_norm": 0.49514684081077576, "learning_rate": 0.00010256227278847448, "loss": 2.859, "step": 123250 }, { "epoch": 33.18087855297158, "grad_norm": 0.5307257175445557, "learning_rate": 0.00010236030698801669, "loss": 2.8379, "step": 123300 }, { "epoch": 33.194336778639105, "grad_norm": 0.5356868505477905, "learning_rate": 0.00010215834118755888, "loss": 2.8595, "step": 123350 }, { "epoch": 33.20779500430663, "grad_norm": 0.5151222944259644, "learning_rate": 0.00010195637538710111, "loss": 2.8509, "step": 123400 }, { "epoch": 33.22125322997416, "grad_norm": 0.547874391078949, "learning_rate": 0.00010175440958664332, "loss": 2.843, "step": 123450 }, { "epoch": 33.23471145564169, "grad_norm": 0.5329242944717407, "learning_rate": 0.00010155244378618553, "loss": 2.86, "step": 123500 }, { "epoch": 33.248169681309214, "grad_norm": 0.5350729823112488, "learning_rate": 0.00010135047798572774, "loss": 2.8592, "step": 123550 }, { "epoch": 33.26162790697674, "grad_norm": 0.5196654200553894, "learning_rate": 0.00010114851218526996, "loss": 2.8569, "step": 123600 }, { "epoch": 33.27508613264427, "grad_norm": 0.5498469471931458, "learning_rate": 0.00010094654638481217, "loss": 2.847, "step": 123650 }, { "epoch": 33.2885443583118, "grad_norm": 0.5427135229110718, "learning_rate": 0.00010074458058435438, "loss": 2.8574, "step": 123700 }, { "epoch": 33.30200258397933, "grad_norm": 0.5531679391860962, "learning_rate": 0.00010054261478389658, "loss": 2.8702, "step": 123750 }, { "epoch": 33.31546080964686, "grad_norm": 0.4978818893432617, "learning_rate": 0.0001003406489834388, "loss": 2.8651, "step": 123800 }, { "epoch": 33.328919035314385, "grad_norm": 0.5036234855651855, "learning_rate": 0.00010013868318298101, "loss": 2.8549, "step": 123850 }, { "epoch": 33.34237726098191, "grad_norm": 0.5336086750030518, "learning_rate": 9.993671738252322e-05, "loss": 2.8537, "step": 123900 }, { "epoch": 33.35583548664944, "grad_norm": 0.5408747792243958, "learning_rate": 9.973475158206543e-05, "loss": 2.861, "step": 123950 }, { "epoch": 33.36929371231697, "grad_norm": 0.4914780855178833, "learning_rate": 9.953278578160764e-05, "loss": 2.8786, "step": 124000 }, { "epoch": 33.36929371231697, "eval_accuracy": 0.398215062539303, "eval_loss": 3.284298896789551, "eval_runtime": 84.0215, "eval_samples_per_second": 214.374, "eval_steps_per_second": 13.401, "step": 124000 }, { "epoch": 33.36929371231697, "step": 124000, "total_flos": 2.591465108668416e+18, "train_loss": 3.122740319282778, "train_runtime": 84389.1333, "train_samples_per_second": 140.879, "train_steps_per_second": 1.761 } ], "logging_steps": 50, "max_steps": 148640, "num_input_tokens_seen": 0, "num_train_epochs": 40, "save_steps": 10000, "stateful_callbacks": { "EarlyStoppingCallback": { "args": { "early_stopping_patience": 20, "early_stopping_threshold": 0.0 }, "attributes": { "early_stopping_patience_counter": 16 } }, "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": false }, "attributes": {} } }, "total_flos": 2.591465108668416e+18, "train_batch_size": 16, "trial_name": null, "trial_params": null }