Invalid JSON:Unexpected token 'I', ..."ad_norm": Infinity,
"... is not valid JSON
{
"best_global_step": 78000,
"best_metric": 3.2710349559783936,
"best_model_checkpoint": "/scratch/cl5625/exceptions/models/high_2000_634/checkpoint-20000",
"epoch": 26.372523686477177,
"eval_steps": 1000,
"global_step": 98000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.013458225667527994,
"grad_norm": 1.26742684841156,
"learning_rate": 0.000294,
"loss": 8.4484,
"step": 50
},
{
"epoch": 0.026916451335055987,
"grad_norm": 1.158722996711731,
"learning_rate": 0.0005939999999999999,
"loss": 6.7858,
"step": 100
},
{
"epoch": 0.04037467700258398,
"grad_norm": 1.5116316080093384,
"learning_rate": 0.0005998020735155513,
"loss": 6.2921,
"step": 150
},
{
"epoch": 0.053832902670111975,
"grad_norm": 0.9892672300338745,
"learning_rate": 0.0005996001077150935,
"loss": 6.0248,
"step": 200
},
{
"epoch": 0.06729112833763996,
"grad_norm": 1.1211016178131104,
"learning_rate": 0.0005993981419146358,
"loss": 5.8647,
"step": 250
},
{
"epoch": 0.08074935400516796,
"grad_norm": 1.6260143518447876,
"learning_rate": 0.0005991961761141779,
"loss": 5.7301,
"step": 300
},
{
"epoch": 0.09420757967269595,
"grad_norm": 1.0242689847946167,
"learning_rate": 0.0005989942103137202,
"loss": 5.6218,
"step": 350
},
{
"epoch": 0.10766580534022395,
"grad_norm": 1.1451252698898315,
"learning_rate": 0.0005987922445132624,
"loss": 5.4895,
"step": 400
},
{
"epoch": 0.12112403100775193,
"grad_norm": 1.1215616464614868,
"learning_rate": 0.0005985902787128047,
"loss": 5.3563,
"step": 450
},
{
"epoch": 0.13458225667527993,
"grad_norm": 1.6647250652313232,
"learning_rate": 0.0005983883129123468,
"loss": 5.253,
"step": 500
},
{
"epoch": 0.14804048234280792,
"grad_norm": 1.2564589977264404,
"learning_rate": 0.000598186347111889,
"loss": 5.1438,
"step": 550
},
{
"epoch": 0.16149870801033592,
"grad_norm": 1.2288016080856323,
"learning_rate": 0.0005979843813114312,
"loss": 5.0377,
"step": 600
},
{
"epoch": 0.1749569336778639,
"grad_norm": 1.4052221775054932,
"learning_rate": 0.0005977824155109734,
"loss": 4.964,
"step": 650
},
{
"epoch": 0.1884151593453919,
"grad_norm": 1.148010492324829,
"learning_rate": 0.0005975804497105157,
"loss": 4.9008,
"step": 700
},
{
"epoch": 0.2018733850129199,
"grad_norm": 0.8613486886024475,
"learning_rate": 0.0005973784839100578,
"loss": 4.8446,
"step": 750
},
{
"epoch": 0.2153316106804479,
"grad_norm": 0.9448251724243164,
"learning_rate": 0.0005971765181096001,
"loss": 4.7833,
"step": 800
},
{
"epoch": 0.2287898363479759,
"grad_norm": 1.1187844276428223,
"learning_rate": 0.0005969745523091422,
"loss": 4.7343,
"step": 850
},
{
"epoch": 0.24224806201550386,
"grad_norm": 1.007033348083496,
"learning_rate": 0.0005967725865086845,
"loss": 4.6853,
"step": 900
},
{
"epoch": 0.2557062876830319,
"grad_norm": 1.0675344467163086,
"learning_rate": 0.0005965706207082267,
"loss": 4.6427,
"step": 950
},
{
"epoch": 0.26916451335055985,
"grad_norm": 0.7794160842895508,
"learning_rate": 0.0005963686549077689,
"loss": 4.6119,
"step": 1000
},
{
"epoch": 0.26916451335055985,
"eval_accuracy": 0.2648011784272542,
"eval_loss": 4.530973434448242,
"eval_runtime": 82.8695,
"eval_samples_per_second": 217.354,
"eval_steps_per_second": 13.588,
"step": 1000
},
{
"epoch": 0.2826227390180879,
"grad_norm": 0.9382805228233337,
"learning_rate": 0.0005961666891073111,
"loss": 4.5565,
"step": 1050
},
{
"epoch": 0.29608096468561584,
"grad_norm": 0.997184157371521,
"learning_rate": 0.0005959647233068533,
"loss": 4.5084,
"step": 1100
},
{
"epoch": 0.30953919035314387,
"grad_norm": 0.9038007855415344,
"learning_rate": 0.0005957627575063955,
"loss": 4.4788,
"step": 1150
},
{
"epoch": 0.32299741602067183,
"grad_norm": 0.8304411172866821,
"learning_rate": 0.0005955607917059378,
"loss": 4.4294,
"step": 1200
},
{
"epoch": 0.3364556416881998,
"grad_norm": 0.8128106594085693,
"learning_rate": 0.00059535882590548,
"loss": 4.3881,
"step": 1250
},
{
"epoch": 0.3499138673557278,
"grad_norm": 0.8509671688079834,
"learning_rate": 0.0005951568601050221,
"loss": 4.3664,
"step": 1300
},
{
"epoch": 0.3633720930232558,
"grad_norm": 0.8358233571052551,
"learning_rate": 0.0005949548943045644,
"loss": 4.3327,
"step": 1350
},
{
"epoch": 0.3768303186907838,
"grad_norm": 0.8237759470939636,
"learning_rate": 0.0005947529285041066,
"loss": 4.3039,
"step": 1400
},
{
"epoch": 0.3902885443583118,
"grad_norm": 0.7692011594772339,
"learning_rate": 0.0005945509627036488,
"loss": 4.2737,
"step": 1450
},
{
"epoch": 0.4037467700258398,
"grad_norm": 0.8751443028450012,
"learning_rate": 0.000594348996903191,
"loss": 4.2545,
"step": 1500
},
{
"epoch": 0.4172049956933678,
"grad_norm": 0.7871449589729309,
"learning_rate": 0.0005941470311027332,
"loss": 4.2336,
"step": 1550
},
{
"epoch": 0.4306632213608958,
"grad_norm": 0.7425031065940857,
"learning_rate": 0.0005939450653022754,
"loss": 4.2027,
"step": 1600
},
{
"epoch": 0.44412144702842377,
"grad_norm": 0.7063210010528564,
"learning_rate": 0.0005937430995018177,
"loss": 4.1993,
"step": 1650
},
{
"epoch": 0.4575796726959518,
"grad_norm": 0.6517214775085449,
"learning_rate": 0.0005935411337013598,
"loss": 4.1855,
"step": 1700
},
{
"epoch": 0.47103789836347976,
"grad_norm": 0.5370175838470459,
"learning_rate": 0.0005933391679009021,
"loss": 4.1539,
"step": 1750
},
{
"epoch": 0.4844961240310077,
"grad_norm": 0.7828596234321594,
"learning_rate": 0.0005931372021004442,
"loss": 4.1482,
"step": 1800
},
{
"epoch": 0.49795434969853575,
"grad_norm": 0.5271194577217102,
"learning_rate": 0.0005929352362999865,
"loss": 4.1161,
"step": 1850
},
{
"epoch": 0.5114125753660638,
"grad_norm": 0.6511763334274292,
"learning_rate": 0.0005927332704995287,
"loss": 4.1093,
"step": 1900
},
{
"epoch": 0.5248708010335917,
"grad_norm": 0.63717120885849,
"learning_rate": 0.0005925313046990709,
"loss": 4.1033,
"step": 1950
},
{
"epoch": 0.5383290267011197,
"grad_norm": 0.6466953158378601,
"learning_rate": 0.0005923293388986131,
"loss": 4.0754,
"step": 2000
},
{
"epoch": 0.5383290267011197,
"eval_accuracy": 0.31760615775610346,
"eval_loss": 4.02086877822876,
"eval_runtime": 83.8696,
"eval_samples_per_second": 214.762,
"eval_steps_per_second": 13.426,
"step": 2000
},
{
"epoch": 0.5517872523686477,
"grad_norm": 0.7861951589584351,
"learning_rate": 0.0005921273730981553,
"loss": 4.0694,
"step": 2050
},
{
"epoch": 0.5652454780361758,
"grad_norm": 0.6967383623123169,
"learning_rate": 0.0005919254072976975,
"loss": 4.0637,
"step": 2100
},
{
"epoch": 0.5787037037037037,
"grad_norm": 0.5648856163024902,
"learning_rate": 0.0005917234414972398,
"loss": 4.042,
"step": 2150
},
{
"epoch": 0.5921619293712317,
"grad_norm": 0.5680930614471436,
"learning_rate": 0.000591521475696782,
"loss": 4.0242,
"step": 2200
},
{
"epoch": 0.6056201550387597,
"grad_norm": 0.6702027320861816,
"learning_rate": 0.0005913195098963241,
"loss": 4.0204,
"step": 2250
},
{
"epoch": 0.6190783807062877,
"grad_norm": 0.52902752161026,
"learning_rate": 0.0005911175440958664,
"loss": 4.0118,
"step": 2300
},
{
"epoch": 0.6325366063738157,
"grad_norm": 0.6083616018295288,
"learning_rate": 0.0005909155782954086,
"loss": 3.9906,
"step": 2350
},
{
"epoch": 0.6459948320413437,
"grad_norm": 0.6891540288925171,
"learning_rate": 0.0005907136124949508,
"loss": 3.9988,
"step": 2400
},
{
"epoch": 0.6594530577088716,
"grad_norm": 0.7903435826301575,
"learning_rate": 0.000590511646694493,
"loss": 3.9788,
"step": 2450
},
{
"epoch": 0.6729112833763996,
"grad_norm": 0.6207765340805054,
"learning_rate": 0.0005903096808940352,
"loss": 3.9775,
"step": 2500
},
{
"epoch": 0.6863695090439277,
"grad_norm": 0.6579967141151428,
"learning_rate": 0.0005901077150935774,
"loss": 3.9634,
"step": 2550
},
{
"epoch": 0.6998277347114557,
"grad_norm": 0.5054481029510498,
"learning_rate": 0.0005899057492931197,
"loss": 3.9566,
"step": 2600
},
{
"epoch": 0.7132859603789836,
"grad_norm": 0.5562361478805542,
"learning_rate": 0.0005897037834926618,
"loss": 3.9488,
"step": 2650
},
{
"epoch": 0.7267441860465116,
"grad_norm": 0.6277523636817932,
"learning_rate": 0.0005895018176922041,
"loss": 3.9408,
"step": 2700
},
{
"epoch": 0.7402024117140397,
"grad_norm": 0.6413837671279907,
"learning_rate": 0.0005892998518917462,
"loss": 3.9268,
"step": 2750
},
{
"epoch": 0.7536606373815676,
"grad_norm": 0.5833922028541565,
"learning_rate": 0.0005890978860912885,
"loss": 3.9319,
"step": 2800
},
{
"epoch": 0.7671188630490956,
"grad_norm": 0.5738604068756104,
"learning_rate": 0.0005888959202908307,
"loss": 3.8917,
"step": 2850
},
{
"epoch": 0.7805770887166236,
"grad_norm": 0.6214295625686646,
"learning_rate": 0.0005886939544903729,
"loss": 3.9083,
"step": 2900
},
{
"epoch": 0.7940353143841516,
"grad_norm": 0.4847414493560791,
"learning_rate": 0.0005884919886899151,
"loss": 3.9243,
"step": 2950
},
{
"epoch": 0.8074935400516796,
"grad_norm": 0.5091471672058105,
"learning_rate": 0.0005882900228894573,
"loss": 3.8895,
"step": 3000
},
{
"epoch": 0.8074935400516796,
"eval_accuracy": 0.33418686303064754,
"eval_loss": 3.8370444774627686,
"eval_runtime": 83.6955,
"eval_samples_per_second": 215.209,
"eval_steps_per_second": 13.454,
"step": 3000
},
{
"epoch": 0.8209517657192076,
"grad_norm": 0.5805885791778564,
"learning_rate": 0.0005880880570889996,
"loss": 3.8729,
"step": 3050
},
{
"epoch": 0.8344099913867356,
"grad_norm": 0.51325523853302,
"learning_rate": 0.0005878860912885418,
"loss": 3.8711,
"step": 3100
},
{
"epoch": 0.8478682170542635,
"grad_norm": 0.5391101241111755,
"learning_rate": 0.000587684125488084,
"loss": 3.8707,
"step": 3150
},
{
"epoch": 0.8613264427217916,
"grad_norm": 0.594567060470581,
"learning_rate": 0.0005874821596876261,
"loss": 3.8685,
"step": 3200
},
{
"epoch": 0.8747846683893196,
"grad_norm": 0.5041269063949585,
"learning_rate": 0.0005872801938871684,
"loss": 3.8576,
"step": 3250
},
{
"epoch": 0.8882428940568475,
"grad_norm": 0.5653175115585327,
"learning_rate": 0.0005870782280867106,
"loss": 3.8611,
"step": 3300
},
{
"epoch": 0.9017011197243755,
"grad_norm": 0.4646280109882355,
"learning_rate": 0.0005868762622862528,
"loss": 3.8466,
"step": 3350
},
{
"epoch": 0.9151593453919036,
"grad_norm": 0.48554784059524536,
"learning_rate": 0.000586674296485795,
"loss": 3.8411,
"step": 3400
},
{
"epoch": 0.9286175710594315,
"grad_norm": 0.4847779870033264,
"learning_rate": 0.0005864723306853373,
"loss": 3.83,
"step": 3450
},
{
"epoch": 0.9420757967269595,
"grad_norm": 0.5213958621025085,
"learning_rate": 0.0005862703648848794,
"loss": 3.8296,
"step": 3500
},
{
"epoch": 0.9555340223944875,
"grad_norm": 0.5259765982627869,
"learning_rate": 0.0005860683990844217,
"loss": 3.8194,
"step": 3550
},
{
"epoch": 0.9689922480620154,
"grad_norm": 0.4607684016227722,
"learning_rate": 0.0005858664332839638,
"loss": 3.8207,
"step": 3600
},
{
"epoch": 0.9824504737295435,
"grad_norm": 0.4271778166294098,
"learning_rate": 0.0005856644674835061,
"loss": 3.8292,
"step": 3650
},
{
"epoch": 0.9959086993970715,
"grad_norm": 0.5108965039253235,
"learning_rate": 0.0005854625016830483,
"loss": 3.8086,
"step": 3700
},
{
"epoch": 1.009151593453919,
"grad_norm": 0.4472891390323639,
"learning_rate": 0.0005852605358825905,
"loss": 3.7699,
"step": 3750
},
{
"epoch": 1.022609819121447,
"grad_norm": 0.5941793322563171,
"learning_rate": 0.0005850585700821327,
"loss": 3.7369,
"step": 3800
},
{
"epoch": 1.036068044788975,
"grad_norm": 0.5670310258865356,
"learning_rate": 0.0005848566042816749,
"loss": 3.7476,
"step": 3850
},
{
"epoch": 1.049526270456503,
"grad_norm": 0.4884444773197174,
"learning_rate": 0.0005846546384812171,
"loss": 3.7429,
"step": 3900
},
{
"epoch": 1.062984496124031,
"grad_norm": 0.5299445986747742,
"learning_rate": 0.0005844526726807594,
"loss": 3.7333,
"step": 3950
},
{
"epoch": 1.076442721791559,
"grad_norm": 0.4771522581577301,
"learning_rate": 0.0005842507068803016,
"loss": 3.7314,
"step": 4000
},
{
"epoch": 1.076442721791559,
"eval_accuracy": 0.3445179838794142,
"eval_loss": 3.731550931930542,
"eval_runtime": 83.6352,
"eval_samples_per_second": 215.364,
"eval_steps_per_second": 13.463,
"step": 4000
},
{
"epoch": 1.089900947459087,
"grad_norm": 0.4465484023094177,
"learning_rate": 0.0005840487410798437,
"loss": 3.7332,
"step": 4050
},
{
"epoch": 1.103359173126615,
"grad_norm": 0.4768674671649933,
"learning_rate": 0.000583846775279386,
"loss": 3.7332,
"step": 4100
},
{
"epoch": 1.116817398794143,
"grad_norm": 0.45904719829559326,
"learning_rate": 0.0005836448094789282,
"loss": 3.7303,
"step": 4150
},
{
"epoch": 1.130275624461671,
"grad_norm": 0.42836010456085205,
"learning_rate": 0.0005834428436784704,
"loss": 3.7262,
"step": 4200
},
{
"epoch": 1.143733850129199,
"grad_norm": 0.46241292357444763,
"learning_rate": 0.0005832408778780126,
"loss": 3.7227,
"step": 4250
},
{
"epoch": 1.157192075796727,
"grad_norm": 0.4167355000972748,
"learning_rate": 0.0005830389120775548,
"loss": 3.7372,
"step": 4300
},
{
"epoch": 1.1706503014642549,
"grad_norm": 0.47947895526885986,
"learning_rate": 0.000582836946277097,
"loss": 3.7189,
"step": 4350
},
{
"epoch": 1.1841085271317828,
"grad_norm": 0.42984476685523987,
"learning_rate": 0.0005826349804766393,
"loss": 3.7039,
"step": 4400
},
{
"epoch": 1.197566752799311,
"grad_norm": 0.46569886803627014,
"learning_rate": 0.0005824330146761814,
"loss": 3.7191,
"step": 4450
},
{
"epoch": 1.211024978466839,
"grad_norm": 0.48893851041793823,
"learning_rate": 0.0005822310488757237,
"loss": 3.7167,
"step": 4500
},
{
"epoch": 1.224483204134367,
"grad_norm": 0.4580209255218506,
"learning_rate": 0.0005820290830752658,
"loss": 3.7044,
"step": 4550
},
{
"epoch": 1.237941429801895,
"grad_norm": 0.45199137926101685,
"learning_rate": 0.0005818271172748081,
"loss": 3.6876,
"step": 4600
},
{
"epoch": 1.251399655469423,
"grad_norm": 0.46312665939331055,
"learning_rate": 0.0005816251514743503,
"loss": 3.7082,
"step": 4650
},
{
"epoch": 1.2648578811369509,
"grad_norm": 0.43382787704467773,
"learning_rate": 0.0005814231856738925,
"loss": 3.6933,
"step": 4700
},
{
"epoch": 1.2783161068044788,
"grad_norm": 0.4415830969810486,
"learning_rate": 0.0005812212198734347,
"loss": 3.6978,
"step": 4750
},
{
"epoch": 1.291774332472007,
"grad_norm": 0.4108583927154541,
"learning_rate": 0.0005810192540729769,
"loss": 3.6998,
"step": 4800
},
{
"epoch": 1.3052325581395348,
"grad_norm": 0.457771360874176,
"learning_rate": 0.0005808172882725192,
"loss": 3.6897,
"step": 4850
},
{
"epoch": 1.318690783807063,
"grad_norm": 0.3960273861885071,
"learning_rate": 0.0005806153224720614,
"loss": 3.6867,
"step": 4900
},
{
"epoch": 1.332149009474591,
"grad_norm": 0.40958383679389954,
"learning_rate": 0.0005804133566716036,
"loss": 3.6898,
"step": 4950
},
{
"epoch": 1.3456072351421189,
"grad_norm": 0.47377800941467285,
"learning_rate": 0.0005802113908711457,
"loss": 3.6767,
"step": 5000
},
{
"epoch": 1.3456072351421189,
"eval_accuracy": 0.350994966173888,
"eval_loss": 3.6645967960357666,
"eval_runtime": 82.8324,
"eval_samples_per_second": 217.451,
"eval_steps_per_second": 13.594,
"step": 5000
},
{
"epoch": 1.3590654608096469,
"grad_norm": 0.45037412643432617,
"learning_rate": 0.000580009425070688,
"loss": 3.6829,
"step": 5050
},
{
"epoch": 1.3725236864771748,
"grad_norm": 0.429791659116745,
"learning_rate": 0.0005798074592702302,
"loss": 3.6884,
"step": 5100
},
{
"epoch": 1.3859819121447028,
"grad_norm": 0.47882944345474243,
"learning_rate": 0.0005796054934697724,
"loss": 3.6755,
"step": 5150
},
{
"epoch": 1.3994401378122308,
"grad_norm": 0.3995591998100281,
"learning_rate": 0.0005794035276693146,
"loss": 3.6671,
"step": 5200
},
{
"epoch": 1.412898363479759,
"grad_norm": 0.3892046809196472,
"learning_rate": 0.0005792015618688568,
"loss": 3.6648,
"step": 5250
},
{
"epoch": 1.4263565891472867,
"grad_norm": 0.4052110016345978,
"learning_rate": 0.000578999596068399,
"loss": 3.6546,
"step": 5300
},
{
"epoch": 1.4398148148148149,
"grad_norm": 0.4371936321258545,
"learning_rate": 0.0005787976302679413,
"loss": 3.6665,
"step": 5350
},
{
"epoch": 1.4532730404823428,
"grad_norm": 0.4228512942790985,
"learning_rate": 0.0005785956644674834,
"loss": 3.6691,
"step": 5400
},
{
"epoch": 1.4667312661498708,
"grad_norm": 0.39720264077186584,
"learning_rate": 0.0005783936986670257,
"loss": 3.6524,
"step": 5450
},
{
"epoch": 1.4801894918173988,
"grad_norm": 0.41909971833229065,
"learning_rate": 0.0005781917328665678,
"loss": 3.668,
"step": 5500
},
{
"epoch": 1.4936477174849268,
"grad_norm": 0.40924331545829773,
"learning_rate": 0.0005779897670661101,
"loss": 3.6612,
"step": 5550
},
{
"epoch": 1.507105943152455,
"grad_norm": 0.41536152362823486,
"learning_rate": 0.0005777878012656523,
"loss": 3.6536,
"step": 5600
},
{
"epoch": 1.5205641688199827,
"grad_norm": 0.41084960103034973,
"learning_rate": 0.0005775858354651945,
"loss": 3.6443,
"step": 5650
},
{
"epoch": 1.5340223944875109,
"grad_norm": 0.39396101236343384,
"learning_rate": 0.0005773838696647367,
"loss": 3.6459,
"step": 5700
},
{
"epoch": 1.5474806201550386,
"grad_norm": 0.41501590609550476,
"learning_rate": 0.000577181903864279,
"loss": 3.6501,
"step": 5750
},
{
"epoch": 1.5609388458225668,
"grad_norm": 0.4097847044467926,
"learning_rate": 0.0005769799380638212,
"loss": 3.6452,
"step": 5800
},
{
"epoch": 1.5743970714900948,
"grad_norm": 0.3934478461742401,
"learning_rate": 0.0005767779722633633,
"loss": 3.6376,
"step": 5850
},
{
"epoch": 1.5878552971576227,
"grad_norm": 0.44573262333869934,
"learning_rate": 0.0005765760064629056,
"loss": 3.6351,
"step": 5900
},
{
"epoch": 1.6013135228251507,
"grad_norm": 0.39932000637054443,
"learning_rate": 0.0005763740406624477,
"loss": 3.6476,
"step": 5950
},
{
"epoch": 1.6147717484926787,
"grad_norm": 0.39502787590026855,
"learning_rate": 0.00057617207486199,
"loss": 3.6455,
"step": 6000
},
{
"epoch": 1.6147717484926787,
"eval_accuracy": 0.3563786351608169,
"eval_loss": 3.6069247722625732,
"eval_runtime": 82.8031,
"eval_samples_per_second": 217.528,
"eval_steps_per_second": 13.599,
"step": 6000
},
{
"epoch": 1.6282299741602069,
"grad_norm": 0.3885681927204132,
"learning_rate": 0.0005759701090615322,
"loss": 3.6073,
"step": 6050
},
{
"epoch": 1.6416881998277346,
"grad_norm": 0.39660024642944336,
"learning_rate": 0.0005757681432610744,
"loss": 3.636,
"step": 6100
},
{
"epoch": 1.6551464254952628,
"grad_norm": 0.43337056040763855,
"learning_rate": 0.0005755661774606166,
"loss": 3.624,
"step": 6150
},
{
"epoch": 1.6686046511627906,
"grad_norm": 0.4270458519458771,
"learning_rate": 0.0005753642116601588,
"loss": 3.6209,
"step": 6200
},
{
"epoch": 1.6820628768303187,
"grad_norm": 0.37905168533325195,
"learning_rate": 0.000575162245859701,
"loss": 3.6177,
"step": 6250
},
{
"epoch": 1.6955211024978467,
"grad_norm": 0.4157800078392029,
"learning_rate": 0.0005749602800592433,
"loss": 3.6165,
"step": 6300
},
{
"epoch": 1.7089793281653747,
"grad_norm": 0.47244226932525635,
"learning_rate": 0.0005747583142587854,
"loss": 3.6105,
"step": 6350
},
{
"epoch": 1.7224375538329026,
"grad_norm": 0.3916477859020233,
"learning_rate": 0.0005745563484583277,
"loss": 3.6228,
"step": 6400
},
{
"epoch": 1.7358957795004306,
"grad_norm": 0.38795655965805054,
"learning_rate": 0.0005743543826578698,
"loss": 3.6216,
"step": 6450
},
{
"epoch": 1.7493540051679588,
"grad_norm": 0.414130836725235,
"learning_rate": 0.0005741524168574121,
"loss": 3.613,
"step": 6500
},
{
"epoch": 1.7628122308354865,
"grad_norm": 0.39510488510131836,
"learning_rate": 0.0005739504510569543,
"loss": 3.6139,
"step": 6550
},
{
"epoch": 1.7762704565030147,
"grad_norm": 0.43209174275398254,
"learning_rate": 0.0005737484852564966,
"loss": 3.6095,
"step": 6600
},
{
"epoch": 1.7897286821705425,
"grad_norm": 0.42028865218162537,
"learning_rate": 0.0005735465194560387,
"loss": 3.6099,
"step": 6650
},
{
"epoch": 1.8031869078380707,
"grad_norm": 0.35338959097862244,
"learning_rate": 0.000573344553655581,
"loss": 3.6131,
"step": 6700
},
{
"epoch": 1.8166451335055986,
"grad_norm": 0.3739859163761139,
"learning_rate": 0.0005731425878551232,
"loss": 3.6104,
"step": 6750
},
{
"epoch": 1.8301033591731266,
"grad_norm": 0.3555767834186554,
"learning_rate": 0.0005729406220546653,
"loss": 3.5971,
"step": 6800
},
{
"epoch": 1.8435615848406546,
"grad_norm": 0.3831947445869446,
"learning_rate": 0.0005727386562542076,
"loss": 3.596,
"step": 6850
},
{
"epoch": 1.8570198105081825,
"grad_norm": 0.3899814188480377,
"learning_rate": 0.0005725366904537497,
"loss": 3.6082,
"step": 6900
},
{
"epoch": 1.8704780361757107,
"grad_norm": 0.3782346546649933,
"learning_rate": 0.000572334724653292,
"loss": 3.6027,
"step": 6950
},
{
"epoch": 1.8839362618432385,
"grad_norm": 0.399288535118103,
"learning_rate": 0.0005721327588528342,
"loss": 3.5834,
"step": 7000
},
{
"epoch": 1.8839362618432385,
"eval_accuracy": 0.36115551145941843,
"eval_loss": 3.5635132789611816,
"eval_runtime": 82.8213,
"eval_samples_per_second": 217.48,
"eval_steps_per_second": 13.596,
"step": 7000
},
{
"epoch": 1.8973944875107667,
"grad_norm": 0.41457751393318176,
"learning_rate": 0.0005719307930523764,
"loss": 3.5887,
"step": 7050
},
{
"epoch": 1.9108527131782944,
"grad_norm": 0.33313989639282227,
"learning_rate": 0.0005717288272519186,
"loss": 3.592,
"step": 7100
},
{
"epoch": 1.9243109388458226,
"grad_norm": 0.40676578879356384,
"learning_rate": 0.0005715268614514608,
"loss": 3.5908,
"step": 7150
},
{
"epoch": 1.9377691645133506,
"grad_norm": 0.40036818385124207,
"learning_rate": 0.000571324895651003,
"loss": 3.5903,
"step": 7200
},
{
"epoch": 1.9512273901808785,
"grad_norm": 0.38596102595329285,
"learning_rate": 0.0005711229298505453,
"loss": 3.5864,
"step": 7250
},
{
"epoch": 1.9646856158484065,
"grad_norm": 0.35873594880104065,
"learning_rate": 0.0005709209640500874,
"loss": 3.5819,
"step": 7300
},
{
"epoch": 1.9781438415159345,
"grad_norm": 0.34385398030281067,
"learning_rate": 0.0005707189982496297,
"loss": 3.5866,
"step": 7350
},
{
"epoch": 1.9916020671834627,
"grad_norm": 0.3601487874984741,
"learning_rate": 0.0005705170324491719,
"loss": 3.5759,
"step": 7400
},
{
"epoch": 2.00484496124031,
"grad_norm": 0.4122074842453003,
"learning_rate": 0.0005703150666487141,
"loss": 3.5499,
"step": 7450
},
{
"epoch": 2.018303186907838,
"grad_norm": 0.362978458404541,
"learning_rate": 0.0005701131008482563,
"loss": 3.4967,
"step": 7500
},
{
"epoch": 2.0317614125753662,
"grad_norm": 0.36486706137657166,
"learning_rate": 0.0005699111350477986,
"loss": 3.4914,
"step": 7550
},
{
"epoch": 2.045219638242894,
"grad_norm": 0.4011174738407135,
"learning_rate": 0.0005697091692473408,
"loss": 3.4924,
"step": 7600
},
{
"epoch": 2.058677863910422,
"grad_norm": 0.3816431164741516,
"learning_rate": 0.000569507203446883,
"loss": 3.501,
"step": 7650
},
{
"epoch": 2.07213608957795,
"grad_norm": 0.3811775743961334,
"learning_rate": 0.0005693052376464252,
"loss": 3.4885,
"step": 7700
},
{
"epoch": 2.085594315245478,
"grad_norm": 0.46005651354789734,
"learning_rate": 0.0005691032718459673,
"loss": 3.4956,
"step": 7750
},
{
"epoch": 2.099052540913006,
"grad_norm": 0.40435007214546204,
"learning_rate": 0.0005689013060455096,
"loss": 3.5002,
"step": 7800
},
{
"epoch": 2.112510766580534,
"grad_norm": 0.3826838433742523,
"learning_rate": 0.0005686993402450518,
"loss": 3.5016,
"step": 7850
},
{
"epoch": 2.125968992248062,
"grad_norm": 0.3628646433353424,
"learning_rate": 0.000568497374444594,
"loss": 3.4906,
"step": 7900
},
{
"epoch": 2.13942721791559,
"grad_norm": 0.3847992718219757,
"learning_rate": 0.0005682954086441362,
"loss": 3.4973,
"step": 7950
},
{
"epoch": 2.152885443583118,
"grad_norm": 0.4251818358898163,
"learning_rate": 0.0005680934428436784,
"loss": 3.5054,
"step": 8000
},
{
"epoch": 2.152885443583118,
"eval_accuracy": 0.36444077507797584,
"eval_loss": 3.5345206260681152,
"eval_runtime": 82.8388,
"eval_samples_per_second": 217.434,
"eval_steps_per_second": 13.593,
"step": 8000
},
{
"epoch": 2.166343669250646,
"grad_norm": 0.35960686206817627,
"learning_rate": 0.0005678914770432206,
"loss": 3.5043,
"step": 8050
},
{
"epoch": 2.179801894918174,
"grad_norm": 0.3829933702945709,
"learning_rate": 0.0005676895112427629,
"loss": 3.4899,
"step": 8100
},
{
"epoch": 2.193260120585702,
"grad_norm": 0.37171506881713867,
"learning_rate": 0.000567487545442305,
"loss": 3.5007,
"step": 8150
},
{
"epoch": 2.20671834625323,
"grad_norm": 0.372265100479126,
"learning_rate": 0.0005672855796418473,
"loss": 3.5095,
"step": 8200
},
{
"epoch": 2.220176571920758,
"grad_norm": 0.3993186056613922,
"learning_rate": 0.0005670836138413894,
"loss": 3.5103,
"step": 8250
},
{
"epoch": 2.233634797588286,
"grad_norm": 0.3925275504589081,
"learning_rate": 0.0005668816480409317,
"loss": 3.4978,
"step": 8300
},
{
"epoch": 2.2470930232558137,
"grad_norm": 0.36354324221611023,
"learning_rate": 0.0005666796822404739,
"loss": 3.4996,
"step": 8350
},
{
"epoch": 2.260551248923342,
"grad_norm": 0.3708871006965637,
"learning_rate": 0.0005664777164400162,
"loss": 3.4949,
"step": 8400
},
{
"epoch": 2.27400947459087,
"grad_norm": 0.3904614746570587,
"learning_rate": 0.0005662757506395583,
"loss": 3.4986,
"step": 8450
},
{
"epoch": 2.287467700258398,
"grad_norm": 0.3834005892276764,
"learning_rate": 0.0005660737848391006,
"loss": 3.5061,
"step": 8500
},
{
"epoch": 2.300925925925926,
"grad_norm": 0.398995041847229,
"learning_rate": 0.0005658718190386428,
"loss": 3.5025,
"step": 8550
},
{
"epoch": 2.314384151593454,
"grad_norm": 0.38279715180397034,
"learning_rate": 0.000565669853238185,
"loss": 3.498,
"step": 8600
},
{
"epoch": 2.327842377260982,
"grad_norm": 0.394349604845047,
"learning_rate": 0.0005654678874377272,
"loss": 3.508,
"step": 8650
},
{
"epoch": 2.3413006029285097,
"grad_norm": 0.3909831643104553,
"learning_rate": 0.0005652659216372693,
"loss": 3.4965,
"step": 8700
},
{
"epoch": 2.354758828596038,
"grad_norm": 0.4046575427055359,
"learning_rate": 0.0005650639558368116,
"loss": 3.4994,
"step": 8750
},
{
"epoch": 2.3682170542635657,
"grad_norm": 0.37517574429512024,
"learning_rate": 0.0005648619900363538,
"loss": 3.4988,
"step": 8800
},
{
"epoch": 2.381675279931094,
"grad_norm": 0.4017314016819,
"learning_rate": 0.000564660024235896,
"loss": 3.493,
"step": 8850
},
{
"epoch": 2.395133505598622,
"grad_norm": 0.38650795817375183,
"learning_rate": 0.0005644580584354382,
"loss": 3.4845,
"step": 8900
},
{
"epoch": 2.40859173126615,
"grad_norm": 0.38423967361450195,
"learning_rate": 0.0005642560926349804,
"loss": 3.4889,
"step": 8950
},
{
"epoch": 2.422049956933678,
"grad_norm": 0.3676753342151642,
"learning_rate": 0.0005640541268345226,
"loss": 3.4959,
"step": 9000
},
{
"epoch": 2.422049956933678,
"eval_accuracy": 0.36706394475872356,
"eval_loss": 3.5067124366760254,
"eval_runtime": 82.7449,
"eval_samples_per_second": 217.681,
"eval_steps_per_second": 13.608,
"step": 9000
},
{
"epoch": 2.4355081826012057,
"grad_norm": 0.40596437454223633,
"learning_rate": 0.0005638521610340649,
"loss": 3.4883,
"step": 9050
},
{
"epoch": 2.448966408268734,
"grad_norm": 0.39690712094306946,
"learning_rate": 0.000563650195233607,
"loss": 3.481,
"step": 9100
},
{
"epoch": 2.4624246339362617,
"grad_norm": 0.39955735206604004,
"learning_rate": 0.0005634482294331493,
"loss": 3.4916,
"step": 9150
},
{
"epoch": 2.47588285960379,
"grad_norm": 0.35812297463417053,
"learning_rate": 0.0005632462636326914,
"loss": 3.491,
"step": 9200
},
{
"epoch": 2.4893410852713176,
"grad_norm": 0.3697480261325836,
"learning_rate": 0.0005630442978322337,
"loss": 3.4797,
"step": 9250
},
{
"epoch": 2.502799310938846,
"grad_norm": 0.36421218514442444,
"learning_rate": 0.0005628423320317759,
"loss": 3.497,
"step": 9300
},
{
"epoch": 2.516257536606374,
"grad_norm": 0.40997207164764404,
"learning_rate": 0.0005626403662313182,
"loss": 3.4898,
"step": 9350
},
{
"epoch": 2.5297157622739017,
"grad_norm": 0.38088980317115784,
"learning_rate": 0.0005624384004308603,
"loss": 3.4929,
"step": 9400
},
{
"epoch": 2.54317398794143,
"grad_norm": 0.36392611265182495,
"learning_rate": 0.0005622364346304026,
"loss": 3.4857,
"step": 9450
},
{
"epoch": 2.5566322136089576,
"grad_norm": 0.3923434317111969,
"learning_rate": 0.0005620344688299448,
"loss": 3.4931,
"step": 9500
},
{
"epoch": 2.570090439276486,
"grad_norm": 0.35099494457244873,
"learning_rate": 0.0005618325030294869,
"loss": 3.483,
"step": 9550
},
{
"epoch": 2.583548664944014,
"grad_norm": 0.3673897683620453,
"learning_rate": 0.0005616305372290292,
"loss": 3.4876,
"step": 9600
},
{
"epoch": 2.5970068906115418,
"grad_norm": 0.35084185004234314,
"learning_rate": 0.0005614285714285713,
"loss": 3.4847,
"step": 9650
},
{
"epoch": 2.6104651162790695,
"grad_norm": 0.36708134412765503,
"learning_rate": 0.0005612266056281136,
"loss": 3.4896,
"step": 9700
},
{
"epoch": 2.6239233419465977,
"grad_norm": 0.3375837802886963,
"learning_rate": 0.0005610246398276558,
"loss": 3.488,
"step": 9750
},
{
"epoch": 2.637381567614126,
"grad_norm": 0.363008588552475,
"learning_rate": 0.000560822674027198,
"loss": 3.4869,
"step": 9800
},
{
"epoch": 2.6508397932816536,
"grad_norm": 0.34566596150398254,
"learning_rate": 0.0005606207082267402,
"loss": 3.4793,
"step": 9850
},
{
"epoch": 2.664298018949182,
"grad_norm": 0.3520165681838989,
"learning_rate": 0.0005604187424262824,
"loss": 3.4799,
"step": 9900
},
{
"epoch": 2.6777562446167096,
"grad_norm": 0.34362173080444336,
"learning_rate": 0.0005602167766258246,
"loss": 3.4979,
"step": 9950
},
{
"epoch": 2.6912144702842378,
"grad_norm": 0.377301424741745,
"learning_rate": 0.0005600148108253669,
"loss": 3.4743,
"step": 10000
},
{
"epoch": 2.6912144702842378,
"eval_accuracy": 0.3697724022210894,
"eval_loss": 3.4816815853118896,
"eval_runtime": 82.7175,
"eval_samples_per_second": 217.753,
"eval_steps_per_second": 13.613,
"step": 10000
},
{
"epoch": 2.704672695951766,
"grad_norm": 0.3740174472332001,
"learning_rate": 0.000559812845024909,
"loss": 3.4764,
"step": 10050
},
{
"epoch": 2.7181309216192937,
"grad_norm": 0.3593759834766388,
"learning_rate": 0.0005596108792244513,
"loss": 3.4815,
"step": 10100
},
{
"epoch": 2.7315891472868215,
"grad_norm": 0.3728601932525635,
"learning_rate": 0.0005594089134239935,
"loss": 3.4768,
"step": 10150
},
{
"epoch": 2.7450473729543496,
"grad_norm": 0.3996655344963074,
"learning_rate": 0.0005592069476235358,
"loss": 3.4812,
"step": 10200
},
{
"epoch": 2.758505598621878,
"grad_norm": 0.34932830929756165,
"learning_rate": 0.0005590049818230779,
"loss": 3.4811,
"step": 10250
},
{
"epoch": 2.7719638242894056,
"grad_norm": 0.3714875280857086,
"learning_rate": 0.0005588030160226202,
"loss": 3.4646,
"step": 10300
},
{
"epoch": 2.7854220499569338,
"grad_norm": 0.3649805784225464,
"learning_rate": 0.0005586010502221623,
"loss": 3.4769,
"step": 10350
},
{
"epoch": 2.7988802756244615,
"grad_norm": 0.3443527817726135,
"learning_rate": 0.0005583990844217045,
"loss": 3.4677,
"step": 10400
},
{
"epoch": 2.8123385012919897,
"grad_norm": 0.3676701784133911,
"learning_rate": 0.0005581971186212468,
"loss": 3.4802,
"step": 10450
},
{
"epoch": 2.825796726959518,
"grad_norm": 0.3489024341106415,
"learning_rate": 0.0005579951528207889,
"loss": 3.4772,
"step": 10500
},
{
"epoch": 2.8392549526270456,
"grad_norm": 0.35671162605285645,
"learning_rate": 0.0005577931870203312,
"loss": 3.4853,
"step": 10550
},
{
"epoch": 2.8527131782945734,
"grad_norm": 0.3905632495880127,
"learning_rate": 0.0005575912212198733,
"loss": 3.4749,
"step": 10600
},
{
"epoch": 2.8661714039621016,
"grad_norm": 0.35995879769325256,
"learning_rate": 0.0005573892554194156,
"loss": 3.4718,
"step": 10650
},
{
"epoch": 2.8796296296296298,
"grad_norm": 0.387455552816391,
"learning_rate": 0.0005571872896189578,
"loss": 3.4766,
"step": 10700
},
{
"epoch": 2.8930878552971575,
"grad_norm": 0.3779407739639282,
"learning_rate": 0.0005569853238185,
"loss": 3.4616,
"step": 10750
},
{
"epoch": 2.9065460809646857,
"grad_norm": 0.3573041558265686,
"learning_rate": 0.0005567833580180422,
"loss": 3.4643,
"step": 10800
},
{
"epoch": 2.9200043066322134,
"grad_norm": 0.38502898812294006,
"learning_rate": 0.0005565813922175844,
"loss": 3.4588,
"step": 10850
},
{
"epoch": 2.9334625322997416,
"grad_norm": 0.3809449374675751,
"learning_rate": 0.0005563794264171266,
"loss": 3.4651,
"step": 10900
},
{
"epoch": 2.94692075796727,
"grad_norm": 0.34850195050239563,
"learning_rate": 0.0005561774606166689,
"loss": 3.4523,
"step": 10950
},
{
"epoch": 2.9603789836347976,
"grad_norm": 0.3215171992778778,
"learning_rate": 0.000555975494816211,
"loss": 3.4678,
"step": 11000
},
{
"epoch": 2.9603789836347976,
"eval_accuracy": 0.3720947287587792,
"eval_loss": 3.4595441818237305,
"eval_runtime": 82.9189,
"eval_samples_per_second": 217.224,
"eval_steps_per_second": 13.58,
"step": 11000
},
{
"epoch": 2.9738372093023253,
"grad_norm": 0.3726615607738495,
"learning_rate": 0.0005557735290157534,
"loss": 3.4657,
"step": 11050
},
{
"epoch": 2.9872954349698535,
"grad_norm": 0.3686408996582031,
"learning_rate": 0.0005555715632152955,
"loss": 3.4611,
"step": 11100
},
{
"epoch": 3.000538329026701,
"grad_norm": 0.40089622139930725,
"learning_rate": 0.0005553695974148378,
"loss": 3.4624,
"step": 11150
},
{
"epoch": 3.013996554694229,
"grad_norm": 0.36176449060440063,
"learning_rate": 0.0005551676316143799,
"loss": 3.3777,
"step": 11200
},
{
"epoch": 3.027454780361757,
"grad_norm": 0.3481358289718628,
"learning_rate": 0.0005549656658139222,
"loss": 3.3579,
"step": 11250
},
{
"epoch": 3.0409130060292853,
"grad_norm": 0.4106177091598511,
"learning_rate": 0.0005547637000134644,
"loss": 3.3734,
"step": 11300
},
{
"epoch": 3.054371231696813,
"grad_norm": 0.372576504945755,
"learning_rate": 0.0005545617342130065,
"loss": 3.3767,
"step": 11350
},
{
"epoch": 3.067829457364341,
"grad_norm": 0.4045032560825348,
"learning_rate": 0.0005543597684125488,
"loss": 3.3723,
"step": 11400
},
{
"epoch": 3.081287683031869,
"grad_norm": 0.339962899684906,
"learning_rate": 0.0005541578026120909,
"loss": 3.3753,
"step": 11450
},
{
"epoch": 3.094745908699397,
"grad_norm": 0.4003972113132477,
"learning_rate": 0.0005539558368116332,
"loss": 3.3782,
"step": 11500
},
{
"epoch": 3.108204134366925,
"grad_norm": 0.3368057608604431,
"learning_rate": 0.0005537538710111754,
"loss": 3.3898,
"step": 11550
},
{
"epoch": 3.121662360034453,
"grad_norm": 0.36275339126586914,
"learning_rate": 0.0005535519052107176,
"loss": 3.3905,
"step": 11600
},
{
"epoch": 3.135120585701981,
"grad_norm": 0.3708229959011078,
"learning_rate": 0.0005533499394102598,
"loss": 3.3845,
"step": 11650
},
{
"epoch": 3.148578811369509,
"grad_norm": 0.37828299403190613,
"learning_rate": 0.000553147973609802,
"loss": 3.3919,
"step": 11700
},
{
"epoch": 3.162037037037037,
"grad_norm": 0.36038196086883545,
"learning_rate": 0.0005529460078093442,
"loss": 3.385,
"step": 11750
},
{
"epoch": 3.175495262704565,
"grad_norm": 0.35368356108665466,
"learning_rate": 0.0005527440420088865,
"loss": 3.3906,
"step": 11800
},
{
"epoch": 3.188953488372093,
"grad_norm": 0.3683849573135376,
"learning_rate": 0.0005525420762084286,
"loss": 3.3897,
"step": 11850
},
{
"epoch": 3.202411714039621,
"grad_norm": 0.35951370000839233,
"learning_rate": 0.0005523401104079709,
"loss": 3.3879,
"step": 11900
},
{
"epoch": 3.215869939707149,
"grad_norm": 0.34719353914260864,
"learning_rate": 0.0005521381446075131,
"loss": 3.4005,
"step": 11950
},
{
"epoch": 3.229328165374677,
"grad_norm": 0.38465508818626404,
"learning_rate": 0.0005519361788070554,
"loss": 3.3961,
"step": 12000
},
{
"epoch": 3.229328165374677,
"eval_accuracy": 0.37356211319003246,
"eval_loss": 3.448230743408203,
"eval_runtime": 83.0076,
"eval_samples_per_second": 216.992,
"eval_steps_per_second": 13.565,
"step": 12000
},
{
"epoch": 3.242786391042205,
"grad_norm": 0.37311500310897827,
"learning_rate": 0.0005517342130065975,
"loss": 3.3838,
"step": 12050
},
{
"epoch": 3.2562446167097328,
"grad_norm": 0.34380656480789185,
"learning_rate": 0.0005515322472061398,
"loss": 3.389,
"step": 12100
},
{
"epoch": 3.269702842377261,
"grad_norm": 0.3579881489276886,
"learning_rate": 0.0005513302814056819,
"loss": 3.4034,
"step": 12150
},
{
"epoch": 3.283161068044789,
"grad_norm": 0.35201603174209595,
"learning_rate": 0.0005511283156052242,
"loss": 3.3858,
"step": 12200
},
{
"epoch": 3.296619293712317,
"grad_norm": 0.3588719666004181,
"learning_rate": 0.0005509263498047664,
"loss": 3.3931,
"step": 12250
},
{
"epoch": 3.310077519379845,
"grad_norm": 0.3580819070339203,
"learning_rate": 0.0005507243840043085,
"loss": 3.3901,
"step": 12300
},
{
"epoch": 3.323535745047373,
"grad_norm": 0.3629094362258911,
"learning_rate": 0.0005505224182038508,
"loss": 3.387,
"step": 12350
},
{
"epoch": 3.336993970714901,
"grad_norm": 0.3542305827140808,
"learning_rate": 0.0005503204524033929,
"loss": 3.402,
"step": 12400
},
{
"epoch": 3.3504521963824287,
"grad_norm": 0.35322701930999756,
"learning_rate": 0.0005501184866029352,
"loss": 3.4089,
"step": 12450
},
{
"epoch": 3.363910422049957,
"grad_norm": 0.36537405848503113,
"learning_rate": 0.0005499165208024774,
"loss": 3.3972,
"step": 12500
},
{
"epoch": 3.3773686477174847,
"grad_norm": 0.3306311070919037,
"learning_rate": 0.0005497145550020196,
"loss": 3.3995,
"step": 12550
},
{
"epoch": 3.390826873385013,
"grad_norm": 0.3468082845211029,
"learning_rate": 0.0005495125892015618,
"loss": 3.4029,
"step": 12600
},
{
"epoch": 3.404285099052541,
"grad_norm": 0.3479803800582886,
"learning_rate": 0.000549310623401104,
"loss": 3.3892,
"step": 12650
},
{
"epoch": 3.417743324720069,
"grad_norm": 0.3454436957836151,
"learning_rate": 0.0005491086576006462,
"loss": 3.3924,
"step": 12700
},
{
"epoch": 3.431201550387597,
"grad_norm": 0.32811784744262695,
"learning_rate": 0.0005489066918001885,
"loss": 3.385,
"step": 12750
},
{
"epoch": 3.4446597760551247,
"grad_norm": 0.4041195511817932,
"learning_rate": 0.0005487047259997306,
"loss": 3.3943,
"step": 12800
},
{
"epoch": 3.458118001722653,
"grad_norm": 0.33224380016326904,
"learning_rate": 0.0005485027601992729,
"loss": 3.3811,
"step": 12850
},
{
"epoch": 3.471576227390181,
"grad_norm": 0.3803240656852722,
"learning_rate": 0.0005483007943988151,
"loss": 3.3927,
"step": 12900
},
{
"epoch": 3.485034453057709,
"grad_norm": 0.34609562158584595,
"learning_rate": 0.0005480988285983574,
"loss": 3.3977,
"step": 12950
},
{
"epoch": 3.4984926787252366,
"grad_norm": 0.3851400911808014,
"learning_rate": 0.0005478968627978995,
"loss": 3.3878,
"step": 13000
},
{
"epoch": 3.4984926787252366,
"eval_accuracy": 0.37542464623497357,
"eval_loss": 3.4328770637512207,
"eval_runtime": 82.8867,
"eval_samples_per_second": 217.309,
"eval_steps_per_second": 13.585,
"step": 13000
},
{
"epoch": 3.511950904392765,
"grad_norm": 0.3414784073829651,
"learning_rate": 0.0005476948969974418,
"loss": 3.3884,
"step": 13050
},
{
"epoch": 3.525409130060293,
"grad_norm": 0.3504672348499298,
"learning_rate": 0.0005474929311969839,
"loss": 3.3964,
"step": 13100
},
{
"epoch": 3.5388673557278207,
"grad_norm": 0.35418644547462463,
"learning_rate": 0.0005472909653965261,
"loss": 3.398,
"step": 13150
},
{
"epoch": 3.552325581395349,
"grad_norm": 0.337748646736145,
"learning_rate": 0.0005470889995960684,
"loss": 3.3928,
"step": 13200
},
{
"epoch": 3.5657838070628767,
"grad_norm": 0.3459875285625458,
"learning_rate": 0.0005468870337956105,
"loss": 3.398,
"step": 13250
},
{
"epoch": 3.579242032730405,
"grad_norm": 0.3401067852973938,
"learning_rate": 0.0005466850679951528,
"loss": 3.3979,
"step": 13300
},
{
"epoch": 3.592700258397933,
"grad_norm": 0.362411230802536,
"learning_rate": 0.0005464831021946949,
"loss": 3.3869,
"step": 13350
},
{
"epoch": 3.606158484065461,
"grad_norm": 0.3600657880306244,
"learning_rate": 0.0005462811363942372,
"loss": 3.4032,
"step": 13400
},
{
"epoch": 3.6196167097329885,
"grad_norm": 0.35751640796661377,
"learning_rate": 0.0005460791705937794,
"loss": 3.385,
"step": 13450
},
{
"epoch": 3.6330749354005167,
"grad_norm": 0.38260284066200256,
"learning_rate": 0.0005458772047933216,
"loss": 3.3917,
"step": 13500
},
{
"epoch": 3.646533161068045,
"grad_norm": 0.34781214594841003,
"learning_rate": 0.0005456752389928638,
"loss": 3.3891,
"step": 13550
},
{
"epoch": 3.6599913867355727,
"grad_norm": 0.34643206000328064,
"learning_rate": 0.000545473273192406,
"loss": 3.3952,
"step": 13600
},
{
"epoch": 3.673449612403101,
"grad_norm": 0.3336317837238312,
"learning_rate": 0.0005452713073919482,
"loss": 3.3829,
"step": 13650
},
{
"epoch": 3.6869078380706286,
"grad_norm": 0.355820894241333,
"learning_rate": 0.0005450693415914905,
"loss": 3.3957,
"step": 13700
},
{
"epoch": 3.700366063738157,
"grad_norm": 0.36233624815940857,
"learning_rate": 0.0005448673757910327,
"loss": 3.3846,
"step": 13750
},
{
"epoch": 3.713824289405685,
"grad_norm": 0.39441055059432983,
"learning_rate": 0.0005446654099905749,
"loss": 3.3923,
"step": 13800
},
{
"epoch": 3.7272825150732127,
"grad_norm": 0.3557799160480499,
"learning_rate": 0.0005444634441901171,
"loss": 3.4041,
"step": 13850
},
{
"epoch": 3.7407407407407405,
"grad_norm": 0.3422970771789551,
"learning_rate": 0.0005442614783896594,
"loss": 3.4028,
"step": 13900
},
{
"epoch": 3.7541989664082687,
"grad_norm": 0.35169336199760437,
"learning_rate": 0.0005440595125892015,
"loss": 3.3893,
"step": 13950
},
{
"epoch": 3.767657192075797,
"grad_norm": 0.308647483587265,
"learning_rate": 0.0005438575467887438,
"loss": 3.3992,
"step": 14000
},
{
"epoch": 3.767657192075797,
"eval_accuracy": 0.3768375985915891,
"eval_loss": 3.416151285171509,
"eval_runtime": 83.1015,
"eval_samples_per_second": 216.747,
"eval_steps_per_second": 13.55,
"step": 14000
},
{
"epoch": 3.7811154177433246,
"grad_norm": 0.35147786140441895,
"learning_rate": 0.0005436555809882859,
"loss": 3.3928,
"step": 14050
},
{
"epoch": 3.794573643410853,
"grad_norm": 0.36079832911491394,
"learning_rate": 0.0005434536151878281,
"loss": 3.3905,
"step": 14100
},
{
"epoch": 3.8080318690783805,
"grad_norm": 0.3377189636230469,
"learning_rate": 0.0005432516493873704,
"loss": 3.3878,
"step": 14150
},
{
"epoch": 3.8214900947459087,
"grad_norm": 0.3728129267692566,
"learning_rate": 0.0005430496835869125,
"loss": 3.3897,
"step": 14200
},
{
"epoch": 3.834948320413437,
"grad_norm": 0.36678987741470337,
"learning_rate": 0.0005428477177864548,
"loss": 3.3851,
"step": 14250
},
{
"epoch": 3.8484065460809647,
"grad_norm": 0.3631365895271301,
"learning_rate": 0.0005426457519859969,
"loss": 3.3857,
"step": 14300
},
{
"epoch": 3.8618647717484924,
"grad_norm": 0.3270263075828552,
"learning_rate": 0.0005424437861855392,
"loss": 3.389,
"step": 14350
},
{
"epoch": 3.8753229974160206,
"grad_norm": 0.33803483843803406,
"learning_rate": 0.0005422418203850814,
"loss": 3.3978,
"step": 14400
},
{
"epoch": 3.888781223083549,
"grad_norm": 0.3551308512687683,
"learning_rate": 0.0005420398545846236,
"loss": 3.3832,
"step": 14450
},
{
"epoch": 3.9022394487510765,
"grad_norm": 0.34793177247047424,
"learning_rate": 0.0005418378887841658,
"loss": 3.3931,
"step": 14500
},
{
"epoch": 3.9156976744186047,
"grad_norm": 0.36282774806022644,
"learning_rate": 0.000541635922983708,
"loss": 3.397,
"step": 14550
},
{
"epoch": 3.9291559000861325,
"grad_norm": 0.36660727858543396,
"learning_rate": 0.0005414339571832503,
"loss": 3.3846,
"step": 14600
},
{
"epoch": 3.9426141257536607,
"grad_norm": 0.32748737931251526,
"learning_rate": 0.0005412319913827925,
"loss": 3.387,
"step": 14650
},
{
"epoch": 3.956072351421189,
"grad_norm": 0.3460530638694763,
"learning_rate": 0.0005410300255823347,
"loss": 3.3965,
"step": 14700
},
{
"epoch": 3.9695305770887166,
"grad_norm": 0.33843541145324707,
"learning_rate": 0.0005408280597818769,
"loss": 3.3764,
"step": 14750
},
{
"epoch": 3.9829888027562443,
"grad_norm": 0.3329560160636902,
"learning_rate": 0.0005406260939814191,
"loss": 3.375,
"step": 14800
},
{
"epoch": 3.9964470284237725,
"grad_norm": 0.3391987383365631,
"learning_rate": 0.0005404241281809614,
"loss": 3.3885,
"step": 14850
},
{
"epoch": 4.00968992248062,
"grad_norm": 0.3517279326915741,
"learning_rate": 0.0005402221623805035,
"loss": 3.3073,
"step": 14900
},
{
"epoch": 4.023148148148148,
"grad_norm": 0.3302111327648163,
"learning_rate": 0.0005400201965800457,
"loss": 3.2819,
"step": 14950
},
{
"epoch": 4.036606373815676,
"grad_norm": 0.3727094829082489,
"learning_rate": 0.000539818230779588,
"loss": 3.2876,
"step": 15000
},
{
"epoch": 4.036606373815676,
"eval_accuracy": 0.37796361460265887,
"eval_loss": 3.409923553466797,
"eval_runtime": 83.0189,
"eval_samples_per_second": 216.963,
"eval_steps_per_second": 13.563,
"step": 15000
},
{
"epoch": 4.050064599483204,
"grad_norm": 0.33077356219291687,
"learning_rate": 0.0005396162649791301,
"loss": 3.3023,
"step": 15050
},
{
"epoch": 4.0635228251507325,
"grad_norm": 0.3349703550338745,
"learning_rate": 0.0005394142991786724,
"loss": 3.2951,
"step": 15100
},
{
"epoch": 4.07698105081826,
"grad_norm": 0.4006906747817993,
"learning_rate": 0.0005392123333782145,
"loss": 3.3135,
"step": 15150
},
{
"epoch": 4.090439276485788,
"grad_norm": 0.3686450123786926,
"learning_rate": 0.0005390103675777568,
"loss": 3.3036,
"step": 15200
},
{
"epoch": 4.103897502153316,
"grad_norm": 0.3732355535030365,
"learning_rate": 0.000538808401777299,
"loss": 3.2979,
"step": 15250
},
{
"epoch": 4.117355727820844,
"grad_norm": 0.34645920991897583,
"learning_rate": 0.0005386064359768412,
"loss": 3.3003,
"step": 15300
},
{
"epoch": 4.1308139534883725,
"grad_norm": 0.3556557297706604,
"learning_rate": 0.0005384044701763834,
"loss": 3.3105,
"step": 15350
},
{
"epoch": 4.1442721791559,
"grad_norm": 0.3306709825992584,
"learning_rate": 0.0005382025043759256,
"loss": 3.3084,
"step": 15400
},
{
"epoch": 4.157730404823428,
"grad_norm": 0.33862653374671936,
"learning_rate": 0.0005380005385754678,
"loss": 3.3163,
"step": 15450
},
{
"epoch": 4.171188630490956,
"grad_norm": 0.35408005118370056,
"learning_rate": 0.0005377985727750101,
"loss": 3.3127,
"step": 15500
},
{
"epoch": 4.184646856158484,
"grad_norm": 0.33308085799217224,
"learning_rate": 0.0005375966069745523,
"loss": 3.3215,
"step": 15550
},
{
"epoch": 4.198105081826012,
"grad_norm": 0.38570937514305115,
"learning_rate": 0.0005373946411740945,
"loss": 3.3054,
"step": 15600
},
{
"epoch": 4.21156330749354,
"grad_norm": 0.342415988445282,
"learning_rate": 0.0005371926753736367,
"loss": 3.3119,
"step": 15650
},
{
"epoch": 4.225021533161068,
"grad_norm": 0.3338956832885742,
"learning_rate": 0.000536990709573179,
"loss": 3.3078,
"step": 15700
},
{
"epoch": 4.238479758828596,
"grad_norm": 0.3540484309196472,
"learning_rate": 0.0005367887437727211,
"loss": 3.3256,
"step": 15750
},
{
"epoch": 4.251937984496124,
"grad_norm": 0.3811263144016266,
"learning_rate": 0.0005365867779722634,
"loss": 3.3148,
"step": 15800
},
{
"epoch": 4.265396210163652,
"grad_norm": 0.3342931866645813,
"learning_rate": 0.0005363848121718055,
"loss": 3.3176,
"step": 15850
},
{
"epoch": 4.27885443583118,
"grad_norm": 0.3282807171344757,
"learning_rate": 0.0005361828463713477,
"loss": 3.3182,
"step": 15900
},
{
"epoch": 4.292312661498708,
"grad_norm": 0.38345375657081604,
"learning_rate": 0.00053598088057089,
"loss": 3.3277,
"step": 15950
},
{
"epoch": 4.305770887166236,
"grad_norm": 0.3509785532951355,
"learning_rate": 0.0005357789147704321,
"loss": 3.3279,
"step": 16000
},
{
"epoch": 4.305770887166236,
"eval_accuracy": 0.37911483668421964,
"eval_loss": 3.401001214981079,
"eval_runtime": 82.8356,
"eval_samples_per_second": 217.443,
"eval_steps_per_second": 13.593,
"step": 16000
},
{
"epoch": 4.319229112833764,
"grad_norm": 0.36147937178611755,
"learning_rate": 0.0005355769489699744,
"loss": 3.336,
"step": 16050
},
{
"epoch": 4.332687338501292,
"grad_norm": 0.36040619015693665,
"learning_rate": 0.0005353749831695165,
"loss": 3.3274,
"step": 16100
},
{
"epoch": 4.34614556416882,
"grad_norm": 0.32985731959342957,
"learning_rate": 0.0005351730173690588,
"loss": 3.3176,
"step": 16150
},
{
"epoch": 4.359603789836348,
"grad_norm": 0.3309231102466583,
"learning_rate": 0.000534971051568601,
"loss": 3.3288,
"step": 16200
},
{
"epoch": 4.373062015503876,
"grad_norm": 0.34250351786613464,
"learning_rate": 0.0005347690857681432,
"loss": 3.3268,
"step": 16250
},
{
"epoch": 4.386520241171404,
"grad_norm": 0.36243459582328796,
"learning_rate": 0.0005345671199676854,
"loss": 3.3376,
"step": 16300
},
{
"epoch": 4.399978466838932,
"grad_norm": 0.33996084332466125,
"learning_rate": 0.0005343651541672276,
"loss": 3.3321,
"step": 16350
},
{
"epoch": 4.41343669250646,
"grad_norm": 0.3623185157775879,
"learning_rate": 0.0005341631883667699,
"loss": 3.3353,
"step": 16400
},
{
"epoch": 4.426894918173988,
"grad_norm": 0.33067142963409424,
"learning_rate": 0.0005339612225663121,
"loss": 3.3258,
"step": 16450
},
{
"epoch": 4.440353143841516,
"grad_norm": 0.34787866473197937,
"learning_rate": 0.0005337592567658543,
"loss": 3.3372,
"step": 16500
},
{
"epoch": 4.453811369509044,
"grad_norm": 0.3300538957118988,
"learning_rate": 0.0005335572909653965,
"loss": 3.3338,
"step": 16550
},
{
"epoch": 4.467269595176572,
"grad_norm": 0.3577615022659302,
"learning_rate": 0.0005333553251649387,
"loss": 3.3274,
"step": 16600
},
{
"epoch": 4.4807278208441,
"grad_norm": Infinity,
"learning_rate": 0.000533153359364481,
"loss": 3.332,
"step": 16650
},
{
"epoch": 4.4941860465116275,
"grad_norm": 0.31497806310653687,
"learning_rate": 0.0005329513935640231,
"loss": 3.325,
"step": 16700
},
{
"epoch": 4.507644272179156,
"grad_norm": 0.32799437642097473,
"learning_rate": 0.0005327494277635654,
"loss": 3.3236,
"step": 16750
},
{
"epoch": 4.521102497846684,
"grad_norm": 0.33367159962654114,
"learning_rate": 0.0005325474619631075,
"loss": 3.3347,
"step": 16800
},
{
"epoch": 4.534560723514212,
"grad_norm": 0.34321150183677673,
"learning_rate": 0.0005323454961626497,
"loss": 3.3302,
"step": 16850
},
{
"epoch": 4.54801894918174,
"grad_norm": 0.3412811756134033,
"learning_rate": 0.000532143530362192,
"loss": 3.3247,
"step": 16900
},
{
"epoch": 4.5614771748492675,
"grad_norm": 0.33372411131858826,
"learning_rate": 0.0005319415645617341,
"loss": 3.3292,
"step": 16950
},
{
"epoch": 4.574935400516796,
"grad_norm": 0.32199379801750183,
"learning_rate": 0.0005317395987612764,
"loss": 3.3261,
"step": 17000
},
{
"epoch": 4.574935400516796,
"eval_accuracy": 0.38016881983005024,
"eval_loss": 3.3901989459991455,
"eval_runtime": 83.0783,
"eval_samples_per_second": 216.807,
"eval_steps_per_second": 13.553,
"step": 17000
},
{
"epoch": 4.588393626184324,
"grad_norm": 0.3851243555545807,
"learning_rate": 0.0005315376329608185,
"loss": 3.327,
"step": 17050
},
{
"epoch": 4.601851851851852,
"grad_norm": 0.34229084849357605,
"learning_rate": 0.0005313356671603608,
"loss": 3.3348,
"step": 17100
},
{
"epoch": 4.61531007751938,
"grad_norm": 0.3780457377433777,
"learning_rate": 0.000531133701359903,
"loss": 3.3342,
"step": 17150
},
{
"epoch": 4.628768303186908,
"grad_norm": 0.3306736648082733,
"learning_rate": 0.0005309317355594452,
"loss": 3.3299,
"step": 17200
},
{
"epoch": 4.642226528854436,
"grad_norm": 0.3649004399776459,
"learning_rate": 0.0005307297697589874,
"loss": 3.3277,
"step": 17250
},
{
"epoch": 4.655684754521964,
"grad_norm": 0.3345679044723511,
"learning_rate": 0.0005305278039585297,
"loss": 3.3354,
"step": 17300
},
{
"epoch": 4.669142980189492,
"grad_norm": 0.3180847764015198,
"learning_rate": 0.0005303258381580719,
"loss": 3.336,
"step": 17350
},
{
"epoch": 4.682601205857019,
"grad_norm": 0.34038251638412476,
"learning_rate": 0.0005301238723576141,
"loss": 3.3339,
"step": 17400
},
{
"epoch": 4.696059431524548,
"grad_norm": 0.32033297419548035,
"learning_rate": 0.0005299219065571563,
"loss": 3.3349,
"step": 17450
},
{
"epoch": 4.709517657192076,
"grad_norm": 0.3335103392601013,
"learning_rate": 0.0005297199407566985,
"loss": 3.3396,
"step": 17500
},
{
"epoch": 4.722975882859604,
"grad_norm": 0.34451714158058167,
"learning_rate": 0.0005295179749562407,
"loss": 3.32,
"step": 17550
},
{
"epoch": 4.736434108527131,
"grad_norm": 0.31060317158699036,
"learning_rate": 0.000529316009155783,
"loss": 3.3269,
"step": 17600
},
{
"epoch": 4.7498923341946595,
"grad_norm": 0.34606438875198364,
"learning_rate": 0.0005291140433553251,
"loss": 3.3277,
"step": 17650
},
{
"epoch": 4.763350559862188,
"grad_norm": 0.3411105275154114,
"learning_rate": 0.0005289120775548673,
"loss": 3.3377,
"step": 17700
},
{
"epoch": 4.776808785529716,
"grad_norm": 0.36869412660598755,
"learning_rate": 0.0005287101117544095,
"loss": 3.3392,
"step": 17750
},
{
"epoch": 4.790267011197244,
"grad_norm": 0.3445747494697571,
"learning_rate": 0.0005285081459539517,
"loss": 3.3426,
"step": 17800
},
{
"epoch": 4.803725236864771,
"grad_norm": 0.33783137798309326,
"learning_rate": 0.000528306180153494,
"loss": 3.329,
"step": 17850
},
{
"epoch": 4.8171834625323,
"grad_norm": 0.31216782331466675,
"learning_rate": 0.0005281042143530361,
"loss": 3.3344,
"step": 17900
},
{
"epoch": 4.830641688199828,
"grad_norm": 0.360944002866745,
"learning_rate": 0.0005279022485525784,
"loss": 3.3305,
"step": 17950
},
{
"epoch": 4.844099913867356,
"grad_norm": 0.33704376220703125,
"learning_rate": 0.0005277002827521205,
"loss": 3.3371,
"step": 18000
},
{
"epoch": 4.844099913867356,
"eval_accuracy": 0.3811343644354514,
"eval_loss": 3.378103733062744,
"eval_runtime": 82.7799,
"eval_samples_per_second": 217.589,
"eval_steps_per_second": 13.602,
"step": 18000
},
{
"epoch": 4.857558139534884,
"grad_norm": 0.3187216520309448,
"learning_rate": 0.0005274983169516628,
"loss": 3.3309,
"step": 18050
},
{
"epoch": 4.871016365202411,
"grad_norm": 0.3320375978946686,
"learning_rate": 0.000527296351151205,
"loss": 3.3341,
"step": 18100
},
{
"epoch": 4.88447459086994,
"grad_norm": 0.36164674162864685,
"learning_rate": 0.0005270943853507472,
"loss": 3.3401,
"step": 18150
},
{
"epoch": 4.897932816537468,
"grad_norm": 0.351095050573349,
"learning_rate": 0.0005268924195502894,
"loss": 3.3205,
"step": 18200
},
{
"epoch": 4.911391042204996,
"grad_norm": 0.34818631410598755,
"learning_rate": 0.0005266904537498317,
"loss": 3.3326,
"step": 18250
},
{
"epoch": 4.924849267872523,
"grad_norm": 0.3314363956451416,
"learning_rate": 0.0005264884879493739,
"loss": 3.33,
"step": 18300
},
{
"epoch": 4.9383074935400515,
"grad_norm": 0.3220027983188629,
"learning_rate": 0.0005262865221489161,
"loss": 3.3279,
"step": 18350
},
{
"epoch": 4.95176571920758,
"grad_norm": 0.3427307605743408,
"learning_rate": 0.0005260845563484583,
"loss": 3.3364,
"step": 18400
},
{
"epoch": 4.965223944875108,
"grad_norm": 0.38316917419433594,
"learning_rate": 0.0005258825905480005,
"loss": 3.3295,
"step": 18450
},
{
"epoch": 4.978682170542635,
"grad_norm": 0.3396889865398407,
"learning_rate": 0.0005256806247475427,
"loss": 3.3235,
"step": 18500
},
{
"epoch": 4.992140396210163,
"grad_norm": 0.34338682889938354,
"learning_rate": 0.000525478658947085,
"loss": 3.3487,
"step": 18550
},
{
"epoch": 5.0053832902670115,
"grad_norm": 0.34141039848327637,
"learning_rate": 0.0005252766931466271,
"loss": 3.2943,
"step": 18600
},
{
"epoch": 5.018841515934539,
"grad_norm": 0.34381935000419617,
"learning_rate": 0.0005250747273461693,
"loss": 3.2327,
"step": 18650
},
{
"epoch": 5.032299741602067,
"grad_norm": 0.3238722085952759,
"learning_rate": 0.0005248727615457115,
"loss": 3.2372,
"step": 18700
},
{
"epoch": 5.045757967269595,
"grad_norm": 0.3302757740020752,
"learning_rate": 0.0005246707957452537,
"loss": 3.2414,
"step": 18750
},
{
"epoch": 5.059216192937123,
"grad_norm": 0.32701438665390015,
"learning_rate": 0.000524468829944796,
"loss": 3.2382,
"step": 18800
},
{
"epoch": 5.0726744186046515,
"grad_norm": 0.35157033801078796,
"learning_rate": 0.0005242668641443381,
"loss": 3.2562,
"step": 18850
},
{
"epoch": 5.086132644272179,
"grad_norm": 0.3694489598274231,
"learning_rate": 0.0005240648983438804,
"loss": 3.2462,
"step": 18900
},
{
"epoch": 5.099590869939707,
"grad_norm": 0.3730630576610565,
"learning_rate": 0.0005238629325434225,
"loss": 3.254,
"step": 18950
},
{
"epoch": 5.113049095607235,
"grad_norm": 0.36214181780815125,
"learning_rate": 0.0005236609667429648,
"loss": 3.2554,
"step": 19000
},
{
"epoch": 5.113049095607235,
"eval_accuracy": 0.38188793902564633,
"eval_loss": 3.378777265548706,
"eval_runtime": 82.9376,
"eval_samples_per_second": 217.175,
"eval_steps_per_second": 13.576,
"step": 19000
},
{
"epoch": 5.126507321274763,
"grad_norm": 0.324454128742218,
"learning_rate": 0.000523459000942507,
"loss": 3.2507,
"step": 19050
},
{
"epoch": 5.139965546942291,
"grad_norm": 0.32653293013572693,
"learning_rate": 0.0005232570351420493,
"loss": 3.2566,
"step": 19100
},
{
"epoch": 5.153423772609819,
"grad_norm": 0.3454332947731018,
"learning_rate": 0.0005230550693415915,
"loss": 3.2618,
"step": 19150
},
{
"epoch": 5.166881998277347,
"grad_norm": 0.33300575613975525,
"learning_rate": 0.0005228531035411337,
"loss": 3.2572,
"step": 19200
},
{
"epoch": 5.180340223944875,
"grad_norm": 0.3484051525592804,
"learning_rate": 0.0005226511377406759,
"loss": 3.2567,
"step": 19250
},
{
"epoch": 5.1937984496124034,
"grad_norm": 0.3399806320667267,
"learning_rate": 0.0005224491719402181,
"loss": 3.2681,
"step": 19300
},
{
"epoch": 5.207256675279931,
"grad_norm": 0.3478161692619324,
"learning_rate": 0.0005222472061397603,
"loss": 3.2655,
"step": 19350
},
{
"epoch": 5.220714900947459,
"grad_norm": 0.36208802461624146,
"learning_rate": 0.0005220452403393026,
"loss": 3.2621,
"step": 19400
},
{
"epoch": 5.234173126614987,
"grad_norm": 0.3394205868244171,
"learning_rate": 0.0005218432745388447,
"loss": 3.2673,
"step": 19450
},
{
"epoch": 5.247631352282515,
"grad_norm": 0.3359786868095398,
"learning_rate": 0.000521641308738387,
"loss": 3.2614,
"step": 19500
},
{
"epoch": 5.2610895779500435,
"grad_norm": 0.3136228919029236,
"learning_rate": 0.0005214393429379291,
"loss": 3.2696,
"step": 19550
},
{
"epoch": 5.274547803617571,
"grad_norm": 0.37108007073402405,
"learning_rate": 0.0005212373771374713,
"loss": 3.2662,
"step": 19600
},
{
"epoch": 5.288006029285099,
"grad_norm": 0.33170050382614136,
"learning_rate": 0.0005210354113370136,
"loss": 3.2764,
"step": 19650
},
{
"epoch": 5.301464254952627,
"grad_norm": 0.36899083852767944,
"learning_rate": 0.0005208334455365557,
"loss": 3.2668,
"step": 19700
},
{
"epoch": 5.314922480620155,
"grad_norm": 0.366172194480896,
"learning_rate": 0.000520631479736098,
"loss": 3.2791,
"step": 19750
},
{
"epoch": 5.328380706287683,
"grad_norm": 0.3375883996486664,
"learning_rate": 0.0005204295139356401,
"loss": 3.2791,
"step": 19800
},
{
"epoch": 5.341838931955211,
"grad_norm": 0.3462762236595154,
"learning_rate": 0.0005202275481351824,
"loss": 3.2756,
"step": 19850
},
{
"epoch": 5.355297157622739,
"grad_norm": 0.3278405964374542,
"learning_rate": 0.0005200255823347246,
"loss": 3.2797,
"step": 19900
},
{
"epoch": 5.368755383290267,
"grad_norm": 0.3242386281490326,
"learning_rate": 0.0005198236165342669,
"loss": 3.2693,
"step": 19950
},
{
"epoch": 5.3822136089577945,
"grad_norm": 0.3416351079940796,
"learning_rate": 0.000519621650733809,
"loss": 3.2715,
"step": 20000
},
{
"epoch": 5.3822136089577945,
"eval_accuracy": 0.3827995947906875,
"eval_loss": 3.3710803985595703,
"eval_runtime": 83.119,
"eval_samples_per_second": 216.701,
"eval_steps_per_second": 13.547,
"step": 20000
},
{
"epoch": 5.395671834625323,
"grad_norm": 0.363948792219162,
"learning_rate": 0.0005194196849333513,
"loss": 3.2757,
"step": 20050
},
{
"epoch": 5.409130060292851,
"grad_norm": 0.329689085483551,
"learning_rate": 0.0005192177191328935,
"loss": 3.2833,
"step": 20100
},
{
"epoch": 5.422588285960379,
"grad_norm": 0.3331378996372223,
"learning_rate": 0.0005190157533324357,
"loss": 3.284,
"step": 20150
},
{
"epoch": 5.436046511627907,
"grad_norm": 0.3332585394382477,
"learning_rate": 0.0005188137875319779,
"loss": 3.2738,
"step": 20200
},
{
"epoch": 5.449504737295435,
"grad_norm": 0.33604896068573,
"learning_rate": 0.00051861182173152,
"loss": 3.274,
"step": 20250
},
{
"epoch": 5.462962962962963,
"grad_norm": 0.35873791575431824,
"learning_rate": 0.0005184098559310623,
"loss": 3.2766,
"step": 20300
},
{
"epoch": 5.476421188630491,
"grad_norm": 0.35145166516304016,
"learning_rate": 0.0005182078901306046,
"loss": 3.2781,
"step": 20350
},
{
"epoch": 5.489879414298019,
"grad_norm": 0.32983332872390747,
"learning_rate": 0.0005180059243301467,
"loss": 3.2794,
"step": 20400
},
{
"epoch": 5.503337639965547,
"grad_norm": 0.31841978430747986,
"learning_rate": 0.000517803958529689,
"loss": 3.2822,
"step": 20450
},
{
"epoch": 5.516795865633075,
"grad_norm": 0.33397188782691956,
"learning_rate": 0.0005176019927292311,
"loss": 3.2899,
"step": 20500
},
{
"epoch": 5.530254091300603,
"grad_norm": 0.3516835868358612,
"learning_rate": 0.0005174000269287733,
"loss": 3.2734,
"step": 20550
},
{
"epoch": 5.543712316968131,
"grad_norm": 0.3212178945541382,
"learning_rate": 0.0005171980611283156,
"loss": 3.2799,
"step": 20600
},
{
"epoch": 5.557170542635659,
"grad_norm": 0.325201153755188,
"learning_rate": 0.0005169960953278577,
"loss": 3.2734,
"step": 20650
},
{
"epoch": 5.5706287683031865,
"grad_norm": 0.33020320534706116,
"learning_rate": 0.0005167941295274,
"loss": 3.2771,
"step": 20700
},
{
"epoch": 5.584086993970715,
"grad_norm": 0.3616073429584503,
"learning_rate": 0.0005165921637269421,
"loss": 3.269,
"step": 20750
},
{
"epoch": 5.597545219638243,
"grad_norm": 0.3325035572052002,
"learning_rate": 0.0005163901979264844,
"loss": 3.2864,
"step": 20800
},
{
"epoch": 5.611003445305771,
"grad_norm": 0.3346830904483795,
"learning_rate": 0.0005161882321260266,
"loss": 3.2817,
"step": 20850
},
{
"epoch": 5.624461670973298,
"grad_norm": 0.3304823637008667,
"learning_rate": 0.0005159862663255689,
"loss": 3.2812,
"step": 20900
},
{
"epoch": 5.637919896640827,
"grad_norm": 0.34642040729522705,
"learning_rate": 0.000515784300525111,
"loss": 3.2861,
"step": 20950
},
{
"epoch": 5.651378122308355,
"grad_norm": 0.35256847739219666,
"learning_rate": 0.0005155823347246533,
"loss": 3.2811,
"step": 21000
},
{
"epoch": 5.651378122308355,
"eval_accuracy": 0.3835094933449455,
"eval_loss": 3.362375497817993,
"eval_runtime": 83.0951,
"eval_samples_per_second": 216.764,
"eval_steps_per_second": 13.551,
"step": 21000
},
{
"epoch": 5.664836347975883,
"grad_norm": 0.3544245660305023,
"learning_rate": 0.0005153803689241955,
"loss": 3.2824,
"step": 21050
},
{
"epoch": 5.678294573643411,
"grad_norm": 0.3260439336299896,
"learning_rate": 0.0005151784031237377,
"loss": 3.2829,
"step": 21100
},
{
"epoch": 5.6917527993109385,
"grad_norm": 0.3130982518196106,
"learning_rate": 0.0005149764373232799,
"loss": 3.2909,
"step": 21150
},
{
"epoch": 5.705211024978467,
"grad_norm": 0.3578866422176361,
"learning_rate": 0.000514774471522822,
"loss": 3.2825,
"step": 21200
},
{
"epoch": 5.718669250645995,
"grad_norm": 0.3706607222557068,
"learning_rate": 0.0005145725057223643,
"loss": 3.2924,
"step": 21250
},
{
"epoch": 5.732127476313523,
"grad_norm": 0.3390621840953827,
"learning_rate": 0.0005143705399219066,
"loss": 3.2894,
"step": 21300
},
{
"epoch": 5.745585701981051,
"grad_norm": 0.3412586748600006,
"learning_rate": 0.0005141685741214487,
"loss": 3.2808,
"step": 21350
},
{
"epoch": 5.7590439276485785,
"grad_norm": 0.31726738810539246,
"learning_rate": 0.0005139666083209909,
"loss": 3.2843,
"step": 21400
},
{
"epoch": 5.772502153316107,
"grad_norm": 0.3289768397808075,
"learning_rate": 0.0005137646425205331,
"loss": 3.2988,
"step": 21450
},
{
"epoch": 5.785960378983635,
"grad_norm": 0.3447217345237732,
"learning_rate": 0.0005135626767200753,
"loss": 3.2907,
"step": 21500
},
{
"epoch": 5.799418604651163,
"grad_norm": 0.33407557010650635,
"learning_rate": 0.0005133607109196176,
"loss": 3.2807,
"step": 21550
},
{
"epoch": 5.81287683031869,
"grad_norm": 0.3523687422275543,
"learning_rate": 0.0005131587451191597,
"loss": 3.2769,
"step": 21600
},
{
"epoch": 5.826335055986219,
"grad_norm": 0.3190504312515259,
"learning_rate": 0.000512956779318702,
"loss": 3.2864,
"step": 21650
},
{
"epoch": 5.839793281653747,
"grad_norm": 0.348217248916626,
"learning_rate": 0.0005127548135182441,
"loss": 3.2857,
"step": 21700
},
{
"epoch": 5.853251507321275,
"grad_norm": 0.35180917382240295,
"learning_rate": 0.0005125528477177865,
"loss": 3.2955,
"step": 21750
},
{
"epoch": 5.866709732988802,
"grad_norm": 0.33137446641921997,
"learning_rate": 0.0005123508819173286,
"loss": 3.2981,
"step": 21800
},
{
"epoch": 5.8801679586563305,
"grad_norm": 0.32640713453292847,
"learning_rate": 0.0005121489161168709,
"loss": 3.2822,
"step": 21850
},
{
"epoch": 5.893626184323859,
"grad_norm": 0.365726113319397,
"learning_rate": 0.000511946950316413,
"loss": 3.282,
"step": 21900
},
{
"epoch": 5.907084409991387,
"grad_norm": 0.32220879197120667,
"learning_rate": 0.0005117449845159553,
"loss": 3.296,
"step": 21950
},
{
"epoch": 5.920542635658915,
"grad_norm": 0.3470675051212311,
"learning_rate": 0.0005115430187154975,
"loss": 3.2944,
"step": 22000
},
{
"epoch": 5.920542635658915,
"eval_accuracy": 0.3844178897043198,
"eval_loss": 3.351707696914673,
"eval_runtime": 83.1103,
"eval_samples_per_second": 216.724,
"eval_steps_per_second": 13.548,
"step": 22000
},
{
"epoch": 5.934000861326442,
"grad_norm": 0.3670094609260559,
"learning_rate": 0.0005113410529150397,
"loss": 3.2952,
"step": 22050
},
{
"epoch": 5.9474590869939705,
"grad_norm": 0.30900025367736816,
"learning_rate": 0.0005111390871145819,
"loss": 3.2879,
"step": 22100
},
{
"epoch": 5.960917312661499,
"grad_norm": 0.3392358422279358,
"learning_rate": 0.000510937121314124,
"loss": 3.2929,
"step": 22150
},
{
"epoch": 5.974375538329027,
"grad_norm": 0.33311930298805237,
"learning_rate": 0.0005107351555136663,
"loss": 3.2973,
"step": 22200
},
{
"epoch": 5.987833763996555,
"grad_norm": 0.3448430299758911,
"learning_rate": 0.0005105331897132085,
"loss": 3.2883,
"step": 22250
},
{
"epoch": 6.001076658053402,
"grad_norm": 0.33929625153541565,
"learning_rate": 0.0005103312239127507,
"loss": 3.2842,
"step": 22300
},
{
"epoch": 6.0145348837209305,
"grad_norm": 0.3301530182361603,
"learning_rate": 0.0005101292581122929,
"loss": 3.1882,
"step": 22350
},
{
"epoch": 6.027993109388458,
"grad_norm": 0.36186957359313965,
"learning_rate": 0.0005099272923118351,
"loss": 3.1836,
"step": 22400
},
{
"epoch": 6.041451335055986,
"grad_norm": 0.33315980434417725,
"learning_rate": 0.0005097253265113773,
"loss": 3.2008,
"step": 22450
},
{
"epoch": 6.054909560723514,
"grad_norm": 0.3193178176879883,
"learning_rate": 0.0005095233607109196,
"loss": 3.2039,
"step": 22500
},
{
"epoch": 6.068367786391042,
"grad_norm": 0.35974186658859253,
"learning_rate": 0.0005093213949104617,
"loss": 3.1965,
"step": 22550
},
{
"epoch": 6.0818260120585705,
"grad_norm": 0.3507646918296814,
"learning_rate": 0.000509119429110004,
"loss": 3.2123,
"step": 22600
},
{
"epoch": 6.095284237726098,
"grad_norm": 0.32048681378364563,
"learning_rate": 0.0005089174633095462,
"loss": 3.1991,
"step": 22650
},
{
"epoch": 6.108742463393626,
"grad_norm": 0.3518259823322296,
"learning_rate": 0.0005087154975090885,
"loss": 3.2052,
"step": 22700
},
{
"epoch": 6.122200689061154,
"grad_norm": 0.36318284273147583,
"learning_rate": 0.0005085135317086306,
"loss": 3.2092,
"step": 22750
},
{
"epoch": 6.135658914728682,
"grad_norm": 0.35523420572280884,
"learning_rate": 0.0005083115659081729,
"loss": 3.2063,
"step": 22800
},
{
"epoch": 6.149117140396211,
"grad_norm": 0.34889161586761475,
"learning_rate": 0.0005081096001077151,
"loss": 3.2211,
"step": 22850
},
{
"epoch": 6.162575366063738,
"grad_norm": 0.3717955946922302,
"learning_rate": 0.0005079076343072573,
"loss": 3.215,
"step": 22900
},
{
"epoch": 6.176033591731266,
"grad_norm": 0.3229867219924927,
"learning_rate": 0.0005077056685067995,
"loss": 3.2147,
"step": 22950
},
{
"epoch": 6.189491817398794,
"grad_norm": 0.34120723605155945,
"learning_rate": 0.0005075037027063417,
"loss": 3.2169,
"step": 23000
},
{
"epoch": 6.189491817398794,
"eval_accuracy": 0.3850950855550529,
"eval_loss": 3.354823112487793,
"eval_runtime": 82.8063,
"eval_samples_per_second": 217.52,
"eval_steps_per_second": 13.598,
"step": 23000
},
{
"epoch": 6.2029500430663225,
"grad_norm": 0.354176789522171,
"learning_rate": 0.0005073017369058839,
"loss": 3.2236,
"step": 23050
},
{
"epoch": 6.21640826873385,
"grad_norm": 0.3649733364582062,
"learning_rate": 0.0005070997711054262,
"loss": 3.2226,
"step": 23100
},
{
"epoch": 6.229866494401378,
"grad_norm": 0.3236199617385864,
"learning_rate": 0.0005068978053049683,
"loss": 3.2222,
"step": 23150
},
{
"epoch": 6.243324720068906,
"grad_norm": 0.359089195728302,
"learning_rate": 0.0005066958395045105,
"loss": 3.2299,
"step": 23200
},
{
"epoch": 6.256782945736434,
"grad_norm": 0.31758424639701843,
"learning_rate": 0.0005064938737040527,
"loss": 3.2261,
"step": 23250
},
{
"epoch": 6.270241171403962,
"grad_norm": 0.3593178689479828,
"learning_rate": 0.0005062919079035949,
"loss": 3.2276,
"step": 23300
},
{
"epoch": 6.28369939707149,
"grad_norm": 0.3542214334011078,
"learning_rate": 0.0005060899421031372,
"loss": 3.2285,
"step": 23350
},
{
"epoch": 6.297157622739018,
"grad_norm": 0.3509684205055237,
"learning_rate": 0.0005058879763026793,
"loss": 3.2272,
"step": 23400
},
{
"epoch": 6.310615848406546,
"grad_norm": 0.34707286953926086,
"learning_rate": 0.0005056860105022216,
"loss": 3.2346,
"step": 23450
},
{
"epoch": 6.324074074074074,
"grad_norm": 0.3352578580379486,
"learning_rate": 0.0005054840447017637,
"loss": 3.2147,
"step": 23500
},
{
"epoch": 6.337532299741602,
"grad_norm": 0.3522467613220215,
"learning_rate": 0.0005052820789013061,
"loss": 3.2303,
"step": 23550
},
{
"epoch": 6.35099052540913,
"grad_norm": 0.34847375750541687,
"learning_rate": 0.0005050801131008482,
"loss": 3.2276,
"step": 23600
},
{
"epoch": 6.364448751076658,
"grad_norm": 0.3291275203227997,
"learning_rate": 0.0005048781473003905,
"loss": 3.2327,
"step": 23650
},
{
"epoch": 6.377906976744186,
"grad_norm": 0.34598827362060547,
"learning_rate": 0.0005046761814999326,
"loss": 3.2327,
"step": 23700
},
{
"epoch": 6.3913652024117145,
"grad_norm": 0.35077574849128723,
"learning_rate": 0.0005044742156994749,
"loss": 3.2378,
"step": 23750
},
{
"epoch": 6.404823428079242,
"grad_norm": 0.3228260278701782,
"learning_rate": 0.0005042722498990171,
"loss": 3.2194,
"step": 23800
},
{
"epoch": 6.41828165374677,
"grad_norm": 0.3222607672214508,
"learning_rate": 0.0005040702840985593,
"loss": 3.232,
"step": 23850
},
{
"epoch": 6.431739879414298,
"grad_norm": 0.3459157943725586,
"learning_rate": 0.0005038683182981015,
"loss": 3.2413,
"step": 23900
},
{
"epoch": 6.445198105081826,
"grad_norm": 0.35747694969177246,
"learning_rate": 0.0005036663524976436,
"loss": 3.2377,
"step": 23950
},
{
"epoch": 6.458656330749354,
"grad_norm": 0.35461848974227905,
"learning_rate": 0.0005034643866971859,
"loss": 3.2437,
"step": 24000
},
{
"epoch": 6.458656330749354,
"eval_accuracy": 0.38544090849631446,
"eval_loss": 3.349268913269043,
"eval_runtime": 82.6729,
"eval_samples_per_second": 217.871,
"eval_steps_per_second": 13.62,
"step": 24000
},
{
"epoch": 6.472114556416882,
"grad_norm": 0.3157093822956085,
"learning_rate": 0.0005032624208967281,
"loss": 3.2397,
"step": 24050
},
{
"epoch": 6.48557278208441,
"grad_norm": 0.32771405577659607,
"learning_rate": 0.0005030604550962703,
"loss": 3.2375,
"step": 24100
},
{
"epoch": 6.499031007751938,
"grad_norm": 0.33250144124031067,
"learning_rate": 0.0005028584892958125,
"loss": 3.2381,
"step": 24150
},
{
"epoch": 6.5124892334194655,
"grad_norm": 0.3377365469932556,
"learning_rate": 0.0005026565234953547,
"loss": 3.2447,
"step": 24200
},
{
"epoch": 6.525947459086994,
"grad_norm": 0.35289251804351807,
"learning_rate": 0.0005024545576948969,
"loss": 3.2405,
"step": 24250
},
{
"epoch": 6.539405684754522,
"grad_norm": 0.3252585530281067,
"learning_rate": 0.0005022525918944392,
"loss": 3.26,
"step": 24300
},
{
"epoch": 6.55286391042205,
"grad_norm": 0.34997978806495667,
"learning_rate": 0.0005020506260939813,
"loss": 3.2403,
"step": 24350
},
{
"epoch": 6.566322136089578,
"grad_norm": 0.32051241397857666,
"learning_rate": 0.0005018486602935236,
"loss": 3.2586,
"step": 24400
},
{
"epoch": 6.579780361757106,
"grad_norm": 0.3474780023097992,
"learning_rate": 0.0005016466944930658,
"loss": 3.2473,
"step": 24450
},
{
"epoch": 6.593238587424634,
"grad_norm": 0.3350488543510437,
"learning_rate": 0.0005014447286926081,
"loss": 3.2291,
"step": 24500
},
{
"epoch": 6.606696813092162,
"grad_norm": 0.3512289226055145,
"learning_rate": 0.0005012427628921502,
"loss": 3.2501,
"step": 24550
},
{
"epoch": 6.62015503875969,
"grad_norm": 0.351237952709198,
"learning_rate": 0.0005010407970916925,
"loss": 3.2543,
"step": 24600
},
{
"epoch": 6.633613264427218,
"grad_norm": 0.3321438133716583,
"learning_rate": 0.0005008388312912346,
"loss": 3.2453,
"step": 24650
},
{
"epoch": 6.647071490094746,
"grad_norm": 0.3477047383785248,
"learning_rate": 0.0005006368654907769,
"loss": 3.2424,
"step": 24700
},
{
"epoch": 6.660529715762274,
"grad_norm": 0.34080350399017334,
"learning_rate": 0.0005004348996903191,
"loss": 3.2597,
"step": 24750
},
{
"epoch": 6.673987941429802,
"grad_norm": 0.35621824860572815,
"learning_rate": 0.0005002329338898613,
"loss": 3.2574,
"step": 24800
},
{
"epoch": 6.68744616709733,
"grad_norm": 0.3517288267612457,
"learning_rate": 0.0005000309680894035,
"loss": 3.2518,
"step": 24850
},
{
"epoch": 6.7009043927648575,
"grad_norm": 0.37831586599349976,
"learning_rate": 0.0004998290022889456,
"loss": 3.254,
"step": 24900
},
{
"epoch": 6.714362618432386,
"grad_norm": 0.33792027831077576,
"learning_rate": 0.0004996270364884879,
"loss": 3.245,
"step": 24950
},
{
"epoch": 6.727820844099914,
"grad_norm": 0.3736184537410736,
"learning_rate": 0.0004994250706880301,
"loss": 3.2561,
"step": 25000
},
{
"epoch": 6.727820844099914,
"eval_accuracy": 0.38640362828347097,
"eval_loss": 3.3389387130737305,
"eval_runtime": 82.8152,
"eval_samples_per_second": 217.496,
"eval_steps_per_second": 13.597,
"step": 25000
},
{
"epoch": 6.741279069767442,
"grad_norm": 0.3358832001686096,
"learning_rate": 0.0004992231048875723,
"loss": 3.2547,
"step": 25050
},
{
"epoch": 6.754737295434969,
"grad_norm": 0.3319481909275055,
"learning_rate": 0.0004990211390871145,
"loss": 3.2547,
"step": 25100
},
{
"epoch": 6.768195521102498,
"grad_norm": 0.3206111192703247,
"learning_rate": 0.0004988191732866567,
"loss": 3.2623,
"step": 25150
},
{
"epoch": 6.781653746770026,
"grad_norm": 0.3242841362953186,
"learning_rate": 0.0004986172074861989,
"loss": 3.2551,
"step": 25200
},
{
"epoch": 6.795111972437554,
"grad_norm": 0.33223605155944824,
"learning_rate": 0.0004984152416857412,
"loss": 3.2508,
"step": 25250
},
{
"epoch": 6.808570198105082,
"grad_norm": 0.3294408321380615,
"learning_rate": 0.0004982132758852834,
"loss": 3.2488,
"step": 25300
},
{
"epoch": 6.822028423772609,
"grad_norm": 0.32208919525146484,
"learning_rate": 0.0004980113100848256,
"loss": 3.2535,
"step": 25350
},
{
"epoch": 6.835486649440138,
"grad_norm": 0.3115708827972412,
"learning_rate": 0.0004978093442843678,
"loss": 3.2498,
"step": 25400
},
{
"epoch": 6.848944875107666,
"grad_norm": 0.3404845893383026,
"learning_rate": 0.0004976073784839101,
"loss": 3.2518,
"step": 25450
},
{
"epoch": 6.862403100775194,
"grad_norm": 0.3291415274143219,
"learning_rate": 0.0004974054126834522,
"loss": 3.2623,
"step": 25500
},
{
"epoch": 6.875861326442722,
"grad_norm": 0.3349682688713074,
"learning_rate": 0.0004972034468829945,
"loss": 3.2609,
"step": 25550
},
{
"epoch": 6.8893195521102495,
"grad_norm": 0.3402669429779053,
"learning_rate": 0.0004970014810825366,
"loss": 3.2538,
"step": 25600
},
{
"epoch": 6.902777777777778,
"grad_norm": 0.32534581422805786,
"learning_rate": 0.0004967995152820789,
"loss": 3.2462,
"step": 25650
},
{
"epoch": 6.916236003445306,
"grad_norm": 0.32399308681488037,
"learning_rate": 0.0004965975494816211,
"loss": 3.2551,
"step": 25700
},
{
"epoch": 6.929694229112834,
"grad_norm": 0.3327549695968628,
"learning_rate": 0.0004963955836811633,
"loss": 3.2568,
"step": 25750
},
{
"epoch": 6.943152454780362,
"grad_norm": 0.3390246629714966,
"learning_rate": 0.0004961936178807055,
"loss": 3.2489,
"step": 25800
},
{
"epoch": 6.9566106804478895,
"grad_norm": 0.32658979296684265,
"learning_rate": 0.0004959916520802476,
"loss": 3.2484,
"step": 25850
},
{
"epoch": 6.970068906115418,
"grad_norm": 0.35533612966537476,
"learning_rate": 0.0004957896862797899,
"loss": 3.256,
"step": 25900
},
{
"epoch": 6.983527131782946,
"grad_norm": 0.3066417872905731,
"learning_rate": 0.0004955877204793321,
"loss": 3.2483,
"step": 25950
},
{
"epoch": 6.996985357450473,
"grad_norm": 0.33188706636428833,
"learning_rate": 0.0004953857546788743,
"loss": 3.2573,
"step": 26000
},
{
"epoch": 6.996985357450473,
"eval_accuracy": 0.3870749572040036,
"eval_loss": 3.331630229949951,
"eval_runtime": 82.8971,
"eval_samples_per_second": 217.281,
"eval_steps_per_second": 13.583,
"step": 26000
},
{
"epoch": 7.010228251507321,
"grad_norm": 0.35253655910491943,
"learning_rate": 0.0004951837888784165,
"loss": 3.1749,
"step": 26050
},
{
"epoch": 7.0236864771748495,
"grad_norm": 0.3584299385547638,
"learning_rate": 0.0004949818230779587,
"loss": 3.1631,
"step": 26100
},
{
"epoch": 7.037144702842378,
"grad_norm": 0.3117905855178833,
"learning_rate": 0.0004947798572775009,
"loss": 3.1539,
"step": 26150
},
{
"epoch": 7.050602928509905,
"grad_norm": 0.33536961674690247,
"learning_rate": 0.0004945778914770432,
"loss": 3.1587,
"step": 26200
},
{
"epoch": 7.064061154177433,
"grad_norm": 0.3320726752281189,
"learning_rate": 0.0004943759256765854,
"loss": 3.1566,
"step": 26250
},
{
"epoch": 7.077519379844961,
"grad_norm": 0.3623173236846924,
"learning_rate": 0.0004941739598761276,
"loss": 3.18,
"step": 26300
},
{
"epoch": 7.09097760551249,
"grad_norm": 0.3439716398715973,
"learning_rate": 0.0004939719940756698,
"loss": 3.1685,
"step": 26350
},
{
"epoch": 7.104435831180017,
"grad_norm": 0.34991180896759033,
"learning_rate": 0.0004937700282752121,
"loss": 3.1648,
"step": 26400
},
{
"epoch": 7.117894056847545,
"grad_norm": 0.3564968407154083,
"learning_rate": 0.0004935680624747542,
"loss": 3.1616,
"step": 26450
},
{
"epoch": 7.131352282515073,
"grad_norm": 0.3411671221256256,
"learning_rate": 0.0004933660966742965,
"loss": 3.173,
"step": 26500
},
{
"epoch": 7.144810508182601,
"grad_norm": 0.33032065629959106,
"learning_rate": 0.0004931641308738386,
"loss": 3.1859,
"step": 26550
},
{
"epoch": 7.15826873385013,
"grad_norm": 0.3649570941925049,
"learning_rate": 0.0004929621650733809,
"loss": 3.1791,
"step": 26600
},
{
"epoch": 7.171726959517657,
"grad_norm": 0.3800533413887024,
"learning_rate": 0.0004927601992729231,
"loss": 3.182,
"step": 26650
},
{
"epoch": 7.185185185185185,
"grad_norm": 0.33807170391082764,
"learning_rate": 0.0004925582334724652,
"loss": 3.1907,
"step": 26700
},
{
"epoch": 7.198643410852713,
"grad_norm": 0.33888861536979675,
"learning_rate": 0.0004923562676720075,
"loss": 3.1878,
"step": 26750
},
{
"epoch": 7.2121016365202415,
"grad_norm": 0.3583280146121979,
"learning_rate": 0.0004921543018715497,
"loss": 3.1849,
"step": 26800
},
{
"epoch": 7.225559862187769,
"grad_norm": 0.33588823676109314,
"learning_rate": 0.0004919523360710919,
"loss": 3.1933,
"step": 26850
},
{
"epoch": 7.239018087855297,
"grad_norm": 0.3458443880081177,
"learning_rate": 0.0004917503702706341,
"loss": 3.1998,
"step": 26900
},
{
"epoch": 7.252476313522825,
"grad_norm": 0.3643375039100647,
"learning_rate": 0.0004915484044701763,
"loss": 3.1857,
"step": 26950
},
{
"epoch": 7.265934539190353,
"grad_norm": 0.3227783143520355,
"learning_rate": 0.0004913464386697185,
"loss": 3.1941,
"step": 27000
},
{
"epoch": 7.265934539190353,
"eval_accuracy": 0.386897754182578,
"eval_loss": 3.340879440307617,
"eval_runtime": 82.7527,
"eval_samples_per_second": 217.661,
"eval_steps_per_second": 13.607,
"step": 27000
},
{
"epoch": 7.279392764857882,
"grad_norm": 0.3403446674346924,
"learning_rate": 0.0004911444728692608,
"loss": 3.1955,
"step": 27050
},
{
"epoch": 7.292850990525409,
"grad_norm": 0.32488778233528137,
"learning_rate": 0.000490942507068803,
"loss": 3.1895,
"step": 27100
},
{
"epoch": 7.306309216192937,
"grad_norm": 0.33461448550224304,
"learning_rate": 0.0004907405412683452,
"loss": 3.2078,
"step": 27150
},
{
"epoch": 7.319767441860465,
"grad_norm": 0.36523109674453735,
"learning_rate": 0.0004905385754678874,
"loss": 3.1978,
"step": 27200
},
{
"epoch": 7.333225667527993,
"grad_norm": 0.36637189984321594,
"learning_rate": 0.0004903366096674297,
"loss": 3.2034,
"step": 27250
},
{
"epoch": 7.346683893195521,
"grad_norm": 0.33307406306266785,
"learning_rate": 0.0004901346438669718,
"loss": 3.198,
"step": 27300
},
{
"epoch": 7.360142118863049,
"grad_norm": 0.3306449055671692,
"learning_rate": 0.0004899326780665141,
"loss": 3.2053,
"step": 27350
},
{
"epoch": 7.373600344530577,
"grad_norm": 0.33870238065719604,
"learning_rate": 0.0004897307122660562,
"loss": 3.2052,
"step": 27400
},
{
"epoch": 7.387058570198105,
"grad_norm": 0.35059431195259094,
"learning_rate": 0.0004895287464655985,
"loss": 3.2115,
"step": 27450
},
{
"epoch": 7.4005167958656335,
"grad_norm": 0.32307863235473633,
"learning_rate": 0.0004893267806651407,
"loss": 3.2043,
"step": 27500
},
{
"epoch": 7.413975021533161,
"grad_norm": 0.33382049202919006,
"learning_rate": 0.0004891248148646829,
"loss": 3.2005,
"step": 27550
},
{
"epoch": 7.427433247200689,
"grad_norm": 0.3406108021736145,
"learning_rate": 0.0004889228490642251,
"loss": 3.2031,
"step": 27600
},
{
"epoch": 7.440891472868217,
"grad_norm": 0.3531828224658966,
"learning_rate": 0.0004887208832637672,
"loss": 3.2108,
"step": 27650
},
{
"epoch": 7.454349698535745,
"grad_norm": 0.3409297466278076,
"learning_rate": 0.0004885189174633095,
"loss": 3.2143,
"step": 27700
},
{
"epoch": 7.467807924203273,
"grad_norm": 0.3473402261734009,
"learning_rate": 0.0004883169516628517,
"loss": 3.2069,
"step": 27750
},
{
"epoch": 7.481266149870801,
"grad_norm": 0.36594653129577637,
"learning_rate": 0.0004881149858623939,
"loss": 3.2003,
"step": 27800
},
{
"epoch": 7.494724375538329,
"grad_norm": 0.32872921228408813,
"learning_rate": 0.00048791302006193614,
"loss": 3.2108,
"step": 27850
},
{
"epoch": 7.508182601205857,
"grad_norm": 0.3425224721431732,
"learning_rate": 0.00048771105426147833,
"loss": 3.2107,
"step": 27900
},
{
"epoch": 7.521640826873385,
"grad_norm": 0.31066015362739563,
"learning_rate": 0.00048750908846102053,
"loss": 3.2056,
"step": 27950
},
{
"epoch": 7.535099052540913,
"grad_norm": 0.3437727391719818,
"learning_rate": 0.0004873071226605628,
"loss": 3.2031,
"step": 28000
},
{
"epoch": 7.535099052540913,
"eval_accuracy": 0.3872242379835491,
"eval_loss": 3.334498405456543,
"eval_runtime": 82.6982,
"eval_samples_per_second": 217.804,
"eval_steps_per_second": 13.616,
"step": 28000
},
{
"epoch": 7.548557278208441,
"grad_norm": 0.39486926794052124,
"learning_rate": 0.00048710515686010503,
"loss": 3.2147,
"step": 28050
},
{
"epoch": 7.562015503875969,
"grad_norm": 0.33266329765319824,
"learning_rate": 0.0004869031910596472,
"loss": 3.2087,
"step": 28100
},
{
"epoch": 7.575473729543497,
"grad_norm": 0.3571857511997223,
"learning_rate": 0.0004867012252591894,
"loss": 3.2204,
"step": 28150
},
{
"epoch": 7.588931955211025,
"grad_norm": 0.3355523347854614,
"learning_rate": 0.0004864992594587316,
"loss": 3.2047,
"step": 28200
},
{
"epoch": 7.602390180878553,
"grad_norm": 0.3486146032810211,
"learning_rate": 0.0004862972936582738,
"loss": 3.2099,
"step": 28250
},
{
"epoch": 7.615848406546081,
"grad_norm": 0.3364751935005188,
"learning_rate": 0.00048609532785781606,
"loss": 3.2113,
"step": 28300
},
{
"epoch": 7.629306632213609,
"grad_norm": 0.35110846161842346,
"learning_rate": 0.00048589336205735826,
"loss": 3.2216,
"step": 28350
},
{
"epoch": 7.6427648578811365,
"grad_norm": 0.3510671854019165,
"learning_rate": 0.00048569139625690046,
"loss": 3.2076,
"step": 28400
},
{
"epoch": 7.656223083548665,
"grad_norm": 0.34089210629463196,
"learning_rate": 0.00048548943045644265,
"loss": 3.2249,
"step": 28450
},
{
"epoch": 7.669681309216193,
"grad_norm": 0.32025688886642456,
"learning_rate": 0.00048528746465598485,
"loss": 3.2187,
"step": 28500
},
{
"epoch": 7.683139534883721,
"grad_norm": 0.3627590239048004,
"learning_rate": 0.0004850854988555271,
"loss": 3.225,
"step": 28550
},
{
"epoch": 7.696597760551249,
"grad_norm": 0.3553415834903717,
"learning_rate": 0.0004848835330550693,
"loss": 3.2304,
"step": 28600
},
{
"epoch": 7.7100559862187765,
"grad_norm": 0.3471336364746094,
"learning_rate": 0.0004846815672546115,
"loss": 3.2075,
"step": 28650
},
{
"epoch": 7.723514211886305,
"grad_norm": 0.32429608702659607,
"learning_rate": 0.0004844796014541537,
"loss": 3.2206,
"step": 28700
},
{
"epoch": 7.736972437553833,
"grad_norm": 0.3769390881061554,
"learning_rate": 0.0004842776356536959,
"loss": 3.2225,
"step": 28750
},
{
"epoch": 7.750430663221361,
"grad_norm": 0.3422105610370636,
"learning_rate": 0.00048407566985323813,
"loss": 3.2133,
"step": 28800
},
{
"epoch": 7.763888888888889,
"grad_norm": 0.33347204327583313,
"learning_rate": 0.0004838737040527804,
"loss": 3.2149,
"step": 28850
},
{
"epoch": 7.777347114556417,
"grad_norm": 0.34581631422042847,
"learning_rate": 0.0004836717382523226,
"loss": 3.2263,
"step": 28900
},
{
"epoch": 7.790805340223945,
"grad_norm": 0.3858388066291809,
"learning_rate": 0.0004834697724518648,
"loss": 3.2224,
"step": 28950
},
{
"epoch": 7.804263565891473,
"grad_norm": 0.3505910038948059,
"learning_rate": 0.000483267806651407,
"loss": 3.2186,
"step": 29000
},
{
"epoch": 7.804263565891473,
"eval_accuracy": 0.38842576355923625,
"eval_loss": 3.322479724884033,
"eval_runtime": 82.6576,
"eval_samples_per_second": 217.911,
"eval_steps_per_second": 13.622,
"step": 29000
},
{
"epoch": 7.817721791559001,
"grad_norm": 0.3376924395561218,
"learning_rate": 0.0004830658408509492,
"loss": 3.2282,
"step": 29050
},
{
"epoch": 7.831180017226529,
"grad_norm": 0.3375740647315979,
"learning_rate": 0.0004828638750504914,
"loss": 3.2171,
"step": 29100
},
{
"epoch": 7.844638242894057,
"grad_norm": 0.3212764263153076,
"learning_rate": 0.0004826619092500336,
"loss": 3.2158,
"step": 29150
},
{
"epoch": 7.858096468561585,
"grad_norm": 0.3387219309806824,
"learning_rate": 0.0004824599434495758,
"loss": 3.2302,
"step": 29200
},
{
"epoch": 7.871554694229113,
"grad_norm": 0.34973713755607605,
"learning_rate": 0.00048225797764911806,
"loss": 3.2351,
"step": 29250
},
{
"epoch": 7.885012919896641,
"grad_norm": 0.3481665849685669,
"learning_rate": 0.00048205601184866026,
"loss": 3.2244,
"step": 29300
},
{
"epoch": 7.8984711455641685,
"grad_norm": 0.32571884989738464,
"learning_rate": 0.00048185404604820245,
"loss": 3.2164,
"step": 29350
},
{
"epoch": 7.911929371231697,
"grad_norm": 0.3369390070438385,
"learning_rate": 0.00048165208024774465,
"loss": 3.2271,
"step": 29400
},
{
"epoch": 7.925387596899225,
"grad_norm": 0.3274584412574768,
"learning_rate": 0.00048145011444728685,
"loss": 3.2163,
"step": 29450
},
{
"epoch": 7.938845822566753,
"grad_norm": 0.34484657645225525,
"learning_rate": 0.0004812481486468291,
"loss": 3.2214,
"step": 29500
},
{
"epoch": 7.95230404823428,
"grad_norm": 0.3353259265422821,
"learning_rate": 0.0004810461828463713,
"loss": 3.2342,
"step": 29550
},
{
"epoch": 7.965762273901809,
"grad_norm": 0.34124571084976196,
"learning_rate": 0.0004808442170459135,
"loss": 3.2196,
"step": 29600
},
{
"epoch": 7.979220499569337,
"grad_norm": 0.34763258695602417,
"learning_rate": 0.0004806422512454557,
"loss": 3.233,
"step": 29650
},
{
"epoch": 7.992678725236865,
"grad_norm": 0.3418041467666626,
"learning_rate": 0.00048044028544499793,
"loss": 3.2198,
"step": 29700
},
{
"epoch": 8.005921619293712,
"grad_norm": 0.3311484158039093,
"learning_rate": 0.0004802383196445402,
"loss": 3.1747,
"step": 29750
},
{
"epoch": 8.01937984496124,
"grad_norm": 0.3664562702178955,
"learning_rate": 0.0004800363538440824,
"loss": 3.1189,
"step": 29800
},
{
"epoch": 8.032838070628769,
"grad_norm": 0.3370324671268463,
"learning_rate": 0.0004798343880436246,
"loss": 3.1187,
"step": 29850
},
{
"epoch": 8.046296296296296,
"grad_norm": 0.33492693305015564,
"learning_rate": 0.00047963242224316683,
"loss": 3.1183,
"step": 29900
},
{
"epoch": 8.059754521963825,
"grad_norm": 0.3898618519306183,
"learning_rate": 0.000479430456442709,
"loss": 3.134,
"step": 29950
},
{
"epoch": 8.073212747631352,
"grad_norm": 0.33758506178855896,
"learning_rate": 0.0004792284906422512,
"loss": 3.149,
"step": 30000
},
{
"epoch": 8.073212747631352,
"eval_accuracy": 0.38790567106164925,
"eval_loss": 3.330857038497925,
"eval_runtime": 82.8519,
"eval_samples_per_second": 217.4,
"eval_steps_per_second": 13.591,
"step": 30000
},
{
"epoch": 8.08667097329888,
"grad_norm": 0.34538140892982483,
"learning_rate": 0.0004790265248417934,
"loss": 3.1408,
"step": 30050
},
{
"epoch": 8.100129198966409,
"grad_norm": 0.3376050889492035,
"learning_rate": 0.0004788245590413356,
"loss": 3.1406,
"step": 30100
},
{
"epoch": 8.113587424633936,
"grad_norm": 0.3391457796096802,
"learning_rate": 0.00047862259324087786,
"loss": 3.1441,
"step": 30150
},
{
"epoch": 8.127045650301465,
"grad_norm": 0.31127339601516724,
"learning_rate": 0.00047842062744042006,
"loss": 3.1539,
"step": 30200
},
{
"epoch": 8.140503875968992,
"grad_norm": 0.35242050886154175,
"learning_rate": 0.00047821866163996225,
"loss": 3.156,
"step": 30250
},
{
"epoch": 8.15396210163652,
"grad_norm": 0.3241572082042694,
"learning_rate": 0.00047801669583950445,
"loss": 3.1499,
"step": 30300
},
{
"epoch": 8.167420327304049,
"grad_norm": 0.3501478135585785,
"learning_rate": 0.00047781473003904665,
"loss": 3.1565,
"step": 30350
},
{
"epoch": 8.180878552971576,
"grad_norm": 0.35967209935188293,
"learning_rate": 0.0004776127642385889,
"loss": 3.1548,
"step": 30400
},
{
"epoch": 8.194336778639105,
"grad_norm": 0.3533713221549988,
"learning_rate": 0.0004774107984381311,
"loss": 3.1471,
"step": 30450
},
{
"epoch": 8.207795004306632,
"grad_norm": 0.32562148571014404,
"learning_rate": 0.0004772088326376733,
"loss": 3.1535,
"step": 30500
},
{
"epoch": 8.22125322997416,
"grad_norm": 0.3363795578479767,
"learning_rate": 0.0004770068668372155,
"loss": 3.1602,
"step": 30550
},
{
"epoch": 8.234711455641689,
"grad_norm": 0.48085713386535645,
"learning_rate": 0.0004768049010367577,
"loss": 3.1638,
"step": 30600
},
{
"epoch": 8.248169681309216,
"grad_norm": 0.3464566469192505,
"learning_rate": 0.0004766029352363,
"loss": 3.1633,
"step": 30650
},
{
"epoch": 8.261627906976745,
"grad_norm": 0.337944358587265,
"learning_rate": 0.0004764009694358422,
"loss": 3.1666,
"step": 30700
},
{
"epoch": 8.275086132644272,
"grad_norm": 0.32248654961586,
"learning_rate": 0.0004761990036353844,
"loss": 3.1641,
"step": 30750
},
{
"epoch": 8.2885443583118,
"grad_norm": 0.33475276827812195,
"learning_rate": 0.0004759970378349266,
"loss": 3.1617,
"step": 30800
},
{
"epoch": 8.302002583979329,
"grad_norm": 0.35940995812416077,
"learning_rate": 0.0004757950720344688,
"loss": 3.1671,
"step": 30850
},
{
"epoch": 8.315460809646856,
"grad_norm": 0.3745073974132538,
"learning_rate": 0.000475593106234011,
"loss": 3.1796,
"step": 30900
},
{
"epoch": 8.328919035314383,
"grad_norm": 0.3514716327190399,
"learning_rate": 0.0004753911404335532,
"loss": 3.1621,
"step": 30950
},
{
"epoch": 8.342377260981912,
"grad_norm": 0.33975857496261597,
"learning_rate": 0.0004751891746330954,
"loss": 3.1658,
"step": 31000
},
{
"epoch": 8.342377260981912,
"eval_accuracy": 0.3882654008004231,
"eval_loss": 3.32810378074646,
"eval_runtime": 82.907,
"eval_samples_per_second": 217.256,
"eval_steps_per_second": 13.581,
"step": 31000
},
{
"epoch": 8.35583548664944,
"grad_norm": 0.3387977182865143,
"learning_rate": 0.0004749872088326376,
"loss": 3.176,
"step": 31050
},
{
"epoch": 8.369293712316969,
"grad_norm": 0.3448338508605957,
"learning_rate": 0.00047478524303217986,
"loss": 3.1835,
"step": 31100
},
{
"epoch": 8.382751937984496,
"grad_norm": 0.34663844108581543,
"learning_rate": 0.00047458327723172206,
"loss": 3.1711,
"step": 31150
},
{
"epoch": 8.396210163652023,
"grad_norm": 0.3234356939792633,
"learning_rate": 0.00047438131143126425,
"loss": 3.1775,
"step": 31200
},
{
"epoch": 8.409668389319553,
"grad_norm": 0.3639753460884094,
"learning_rate": 0.00047417934563080645,
"loss": 3.1772,
"step": 31250
},
{
"epoch": 8.42312661498708,
"grad_norm": 0.33908578753471375,
"learning_rate": 0.00047397737983034864,
"loss": 3.1786,
"step": 31300
},
{
"epoch": 8.436584840654609,
"grad_norm": 0.35978230834007263,
"learning_rate": 0.0004737754140298909,
"loss": 3.1795,
"step": 31350
},
{
"epoch": 8.450043066322136,
"grad_norm": 0.35240867733955383,
"learning_rate": 0.0004735734482294331,
"loss": 3.1807,
"step": 31400
},
{
"epoch": 8.463501291989663,
"grad_norm": 0.3479791581630707,
"learning_rate": 0.0004733714824289753,
"loss": 3.1775,
"step": 31450
},
{
"epoch": 8.476959517657193,
"grad_norm": 0.3236258029937744,
"learning_rate": 0.0004731695166285175,
"loss": 3.182,
"step": 31500
},
{
"epoch": 8.49041774332472,
"grad_norm": 0.35990241169929504,
"learning_rate": 0.0004729675508280598,
"loss": 3.1796,
"step": 31550
},
{
"epoch": 8.503875968992247,
"grad_norm": 0.3206709027290344,
"learning_rate": 0.000472765585027602,
"loss": 3.1819,
"step": 31600
},
{
"epoch": 8.517334194659776,
"grad_norm": 0.3248264491558075,
"learning_rate": 0.0004725636192271442,
"loss": 3.1909,
"step": 31650
},
{
"epoch": 8.530792420327304,
"grad_norm": 0.3305676281452179,
"learning_rate": 0.0004723616534266864,
"loss": 3.1779,
"step": 31700
},
{
"epoch": 8.544250645994833,
"grad_norm": 0.3449922502040863,
"learning_rate": 0.0004721596876262286,
"loss": 3.1762,
"step": 31750
},
{
"epoch": 8.55770887166236,
"grad_norm": 0.3667323887348175,
"learning_rate": 0.0004719577218257708,
"loss": 3.1902,
"step": 31800
},
{
"epoch": 8.571167097329887,
"grad_norm": 0.3351379334926605,
"learning_rate": 0.000471755756025313,
"loss": 3.1828,
"step": 31850
},
{
"epoch": 8.584625322997416,
"grad_norm": 0.3552919626235962,
"learning_rate": 0.0004715537902248552,
"loss": 3.1877,
"step": 31900
},
{
"epoch": 8.598083548664944,
"grad_norm": 0.3373323082923889,
"learning_rate": 0.0004713518244243974,
"loss": 3.1865,
"step": 31950
},
{
"epoch": 8.611541774332473,
"grad_norm": 0.3515632450580597,
"learning_rate": 0.00047114985862393966,
"loss": 3.1912,
"step": 32000
},
{
"epoch": 8.611541774332473,
"eval_accuracy": 0.38909709247976887,
"eval_loss": 3.3180899620056152,
"eval_runtime": 82.9693,
"eval_samples_per_second": 217.092,
"eval_steps_per_second": 13.571,
"step": 32000
},
{
"epoch": 8.625,
"grad_norm": 0.36250558495521545,
"learning_rate": 0.00047094789282348186,
"loss": 3.1785,
"step": 32050
},
{
"epoch": 8.638458225667527,
"grad_norm": 0.34723225235939026,
"learning_rate": 0.00047074592702302405,
"loss": 3.2016,
"step": 32100
},
{
"epoch": 8.651916451335056,
"grad_norm": 0.353263795375824,
"learning_rate": 0.00047054396122256625,
"loss": 3.1893,
"step": 32150
},
{
"epoch": 8.665374677002584,
"grad_norm": 0.3421275019645691,
"learning_rate": 0.00047034199542210844,
"loss": 3.2,
"step": 32200
},
{
"epoch": 8.678832902670113,
"grad_norm": 0.36237993836402893,
"learning_rate": 0.0004701400296216507,
"loss": 3.1906,
"step": 32250
},
{
"epoch": 8.69229112833764,
"grad_norm": 0.3536783456802368,
"learning_rate": 0.0004699380638211929,
"loss": 3.196,
"step": 32300
},
{
"epoch": 8.705749354005167,
"grad_norm": 0.34335219860076904,
"learning_rate": 0.0004697360980207351,
"loss": 3.1898,
"step": 32350
},
{
"epoch": 8.719207579672696,
"grad_norm": 0.3655270040035248,
"learning_rate": 0.0004695341322202773,
"loss": 3.193,
"step": 32400
},
{
"epoch": 8.732665805340224,
"grad_norm": 0.34032607078552246,
"learning_rate": 0.0004693321664198196,
"loss": 3.1907,
"step": 32450
},
{
"epoch": 8.746124031007753,
"grad_norm": 0.3389233648777008,
"learning_rate": 0.0004691302006193618,
"loss": 3.191,
"step": 32500
},
{
"epoch": 8.75958225667528,
"grad_norm": 0.31701526045799255,
"learning_rate": 0.000468928234818904,
"loss": 3.1914,
"step": 32550
},
{
"epoch": 8.773040482342807,
"grad_norm": 0.3224322497844696,
"learning_rate": 0.0004687262690184462,
"loss": 3.1989,
"step": 32600
},
{
"epoch": 8.786498708010337,
"grad_norm": 0.37413617968559265,
"learning_rate": 0.00046852430321798837,
"loss": 3.1916,
"step": 32650
},
{
"epoch": 8.799956933677864,
"grad_norm": 0.36142292618751526,
"learning_rate": 0.0004683223374175306,
"loss": 3.1976,
"step": 32700
},
{
"epoch": 8.813415159345391,
"grad_norm": 0.33773449063301086,
"learning_rate": 0.0004681203716170728,
"loss": 3.2034,
"step": 32750
},
{
"epoch": 8.82687338501292,
"grad_norm": 0.33591005206108093,
"learning_rate": 0.000467918405816615,
"loss": 3.1985,
"step": 32800
},
{
"epoch": 8.840331610680447,
"grad_norm": 0.32911378145217896,
"learning_rate": 0.0004677164400161572,
"loss": 3.2054,
"step": 32850
},
{
"epoch": 8.853789836347977,
"grad_norm": 0.3356723487377167,
"learning_rate": 0.0004675144742156994,
"loss": 3.1904,
"step": 32900
},
{
"epoch": 8.867248062015504,
"grad_norm": 0.3588046133518219,
"learning_rate": 0.00046731250841524166,
"loss": 3.1878,
"step": 32950
},
{
"epoch": 8.880706287683031,
"grad_norm": 0.3451225459575653,
"learning_rate": 0.00046711054261478385,
"loss": 3.1972,
"step": 33000
},
{
"epoch": 8.880706287683031,
"eval_accuracy": 0.38944606617984184,
"eval_loss": 3.314375877380371,
"eval_runtime": 82.7475,
"eval_samples_per_second": 217.674,
"eval_steps_per_second": 13.608,
"step": 33000
},
{
"epoch": 8.89416451335056,
"grad_norm": 0.3362581729888916,
"learning_rate": 0.00046690857681432605,
"loss": 3.1881,
"step": 33050
},
{
"epoch": 8.907622739018088,
"grad_norm": 0.3405635952949524,
"learning_rate": 0.00046670661101386825,
"loss": 3.1947,
"step": 33100
},
{
"epoch": 8.921080964685617,
"grad_norm": 0.34576869010925293,
"learning_rate": 0.00046650464521341044,
"loss": 3.1969,
"step": 33150
},
{
"epoch": 8.934539190353144,
"grad_norm": 0.3643607497215271,
"learning_rate": 0.0004663026794129527,
"loss": 3.2137,
"step": 33200
},
{
"epoch": 8.947997416020671,
"grad_norm": 0.3174397945404053,
"learning_rate": 0.0004661007136124949,
"loss": 3.2024,
"step": 33250
},
{
"epoch": 8.9614556416882,
"grad_norm": 0.3258208632469177,
"learning_rate": 0.00046589874781203714,
"loss": 3.204,
"step": 33300
},
{
"epoch": 8.974913867355728,
"grad_norm": 0.3384305238723755,
"learning_rate": 0.0004656967820115794,
"loss": 3.2048,
"step": 33350
},
{
"epoch": 8.988372093023255,
"grad_norm": 0.3482845723628998,
"learning_rate": 0.0004654948162111216,
"loss": 3.1992,
"step": 33400
},
{
"epoch": 9.001614987080103,
"grad_norm": 0.3722672164440155,
"learning_rate": 0.0004652928504106638,
"loss": 3.1969,
"step": 33450
},
{
"epoch": 9.015073212747632,
"grad_norm": 0.3285539448261261,
"learning_rate": 0.000465090884610206,
"loss": 3.0978,
"step": 33500
},
{
"epoch": 9.02853143841516,
"grad_norm": 0.34922707080841064,
"learning_rate": 0.00046488891880974817,
"loss": 3.0943,
"step": 33550
},
{
"epoch": 9.041989664082687,
"grad_norm": 0.3537651300430298,
"learning_rate": 0.0004646869530092904,
"loss": 3.0935,
"step": 33600
},
{
"epoch": 9.055447889750216,
"grad_norm": 0.35505229234695435,
"learning_rate": 0.0004644849872088326,
"loss": 3.1107,
"step": 33650
},
{
"epoch": 9.068906115417743,
"grad_norm": 0.3516994118690491,
"learning_rate": 0.0004642830214083748,
"loss": 3.1112,
"step": 33700
},
{
"epoch": 9.082364341085272,
"grad_norm": 0.3468421995639801,
"learning_rate": 0.000464081055607917,
"loss": 3.1108,
"step": 33750
},
{
"epoch": 9.0958225667528,
"grad_norm": 0.3428479731082916,
"learning_rate": 0.0004638790898074592,
"loss": 3.1048,
"step": 33800
},
{
"epoch": 9.109280792420327,
"grad_norm": 0.3707702159881592,
"learning_rate": 0.00046367712400700146,
"loss": 3.1115,
"step": 33850
},
{
"epoch": 9.122739018087856,
"grad_norm": 0.3596304953098297,
"learning_rate": 0.00046347515820654365,
"loss": 3.121,
"step": 33900
},
{
"epoch": 9.136197243755383,
"grad_norm": 0.3731954097747803,
"learning_rate": 0.00046327319240608585,
"loss": 3.1271,
"step": 33950
},
{
"epoch": 9.14965546942291,
"grad_norm": 0.3276529014110565,
"learning_rate": 0.00046307122660562805,
"loss": 3.1215,
"step": 34000
},
{
"epoch": 9.14965546942291,
"eval_accuracy": 0.3893910908709262,
"eval_loss": 3.3236443996429443,
"eval_runtime": 82.8723,
"eval_samples_per_second": 217.346,
"eval_steps_per_second": 13.587,
"step": 34000
},
{
"epoch": 9.16311369509044,
"grad_norm": 0.3522314131259918,
"learning_rate": 0.00046286926080517024,
"loss": 3.1298,
"step": 34050
},
{
"epoch": 9.176571920757967,
"grad_norm": 0.3590625524520874,
"learning_rate": 0.0004626672950047125,
"loss": 3.1201,
"step": 34100
},
{
"epoch": 9.190030146425496,
"grad_norm": 0.34925782680511475,
"learning_rate": 0.0004624653292042547,
"loss": 3.1308,
"step": 34150
},
{
"epoch": 9.203488372093023,
"grad_norm": 0.353376567363739,
"learning_rate": 0.00046226336340379694,
"loss": 3.1351,
"step": 34200
},
{
"epoch": 9.21694659776055,
"grad_norm": 0.33763816952705383,
"learning_rate": 0.00046206139760333913,
"loss": 3.1355,
"step": 34250
},
{
"epoch": 9.23040482342808,
"grad_norm": 0.3503144681453705,
"learning_rate": 0.0004618594318028814,
"loss": 3.1288,
"step": 34300
},
{
"epoch": 9.243863049095607,
"grad_norm": 0.347207248210907,
"learning_rate": 0.0004616574660024236,
"loss": 3.1372,
"step": 34350
},
{
"epoch": 9.257321274763136,
"grad_norm": 0.3392535150051117,
"learning_rate": 0.0004614555002019658,
"loss": 3.1428,
"step": 34400
},
{
"epoch": 9.270779500430663,
"grad_norm": 0.3617876470088959,
"learning_rate": 0.000461253534401508,
"loss": 3.1319,
"step": 34450
},
{
"epoch": 9.28423772609819,
"grad_norm": 0.3634539842605591,
"learning_rate": 0.00046105156860105017,
"loss": 3.149,
"step": 34500
},
{
"epoch": 9.29769595176572,
"grad_norm": 0.3327468931674957,
"learning_rate": 0.0004608496028005924,
"loss": 3.1465,
"step": 34550
},
{
"epoch": 9.311154177433247,
"grad_norm": 0.35100021958351135,
"learning_rate": 0.0004606476370001346,
"loss": 3.1493,
"step": 34600
},
{
"epoch": 9.324612403100776,
"grad_norm": 0.33665528893470764,
"learning_rate": 0.0004604456711996768,
"loss": 3.1573,
"step": 34650
},
{
"epoch": 9.338070628768303,
"grad_norm": 0.35000231862068176,
"learning_rate": 0.000460243705399219,
"loss": 3.1417,
"step": 34700
},
{
"epoch": 9.35152885443583,
"grad_norm": 0.358193039894104,
"learning_rate": 0.0004600417395987612,
"loss": 3.1419,
"step": 34750
},
{
"epoch": 9.36498708010336,
"grad_norm": 0.34701231122016907,
"learning_rate": 0.00045983977379830345,
"loss": 3.1437,
"step": 34800
},
{
"epoch": 9.378445305770887,
"grad_norm": 0.35321080684661865,
"learning_rate": 0.00045963780799784565,
"loss": 3.1468,
"step": 34850
},
{
"epoch": 9.391903531438416,
"grad_norm": 0.34045690298080444,
"learning_rate": 0.00045943584219738785,
"loss": 3.1561,
"step": 34900
},
{
"epoch": 9.405361757105943,
"grad_norm": 0.3426523208618164,
"learning_rate": 0.00045923387639693004,
"loss": 3.1539,
"step": 34950
},
{
"epoch": 9.41881998277347,
"grad_norm": 0.34726616740226746,
"learning_rate": 0.00045903191059647224,
"loss": 3.1541,
"step": 35000
},
{
"epoch": 9.41881998277347,
"eval_accuracy": 0.3896481493311917,
"eval_loss": 3.318563938140869,
"eval_runtime": 82.9403,
"eval_samples_per_second": 217.168,
"eval_steps_per_second": 13.576,
"step": 35000
},
{
"epoch": 9.432278208441,
"grad_norm": 0.34399011731147766,
"learning_rate": 0.0004588299447960145,
"loss": 3.1556,
"step": 35050
},
{
"epoch": 9.445736434108527,
"grad_norm": 0.3531264364719391,
"learning_rate": 0.00045862797899555674,
"loss": 3.1629,
"step": 35100
},
{
"epoch": 9.459194659776054,
"grad_norm": 0.35482361912727356,
"learning_rate": 0.00045842601319509894,
"loss": 3.1473,
"step": 35150
},
{
"epoch": 9.472652885443583,
"grad_norm": 0.33983859419822693,
"learning_rate": 0.00045822404739464113,
"loss": 3.1545,
"step": 35200
},
{
"epoch": 9.48611111111111,
"grad_norm": 0.34992870688438416,
"learning_rate": 0.0004580220815941834,
"loss": 3.167,
"step": 35250
},
{
"epoch": 9.49956933677864,
"grad_norm": 0.33600035309791565,
"learning_rate": 0.0004578201157937256,
"loss": 3.1537,
"step": 35300
},
{
"epoch": 9.513027562446167,
"grad_norm": 0.3373678922653198,
"learning_rate": 0.0004576181499932678,
"loss": 3.1577,
"step": 35350
},
{
"epoch": 9.526485788113694,
"grad_norm": 0.338767409324646,
"learning_rate": 0.00045741618419280997,
"loss": 3.1566,
"step": 35400
},
{
"epoch": 9.539944013781223,
"grad_norm": 0.36286288499832153,
"learning_rate": 0.0004572142183923522,
"loss": 3.1706,
"step": 35450
},
{
"epoch": 9.55340223944875,
"grad_norm": 0.3332397937774658,
"learning_rate": 0.0004570122525918944,
"loss": 3.1704,
"step": 35500
},
{
"epoch": 9.56686046511628,
"grad_norm": 0.34046825766563416,
"learning_rate": 0.0004568102867914366,
"loss": 3.1648,
"step": 35550
},
{
"epoch": 9.580318690783807,
"grad_norm": 0.3536531627178192,
"learning_rate": 0.0004566083209909788,
"loss": 3.1648,
"step": 35600
},
{
"epoch": 9.593776916451334,
"grad_norm": 0.3309786021709442,
"learning_rate": 0.000456406355190521,
"loss": 3.1673,
"step": 35650
},
{
"epoch": 9.607235142118864,
"grad_norm": 0.33974435925483704,
"learning_rate": 0.00045620438939006326,
"loss": 3.1639,
"step": 35700
},
{
"epoch": 9.62069336778639,
"grad_norm": 0.3273696303367615,
"learning_rate": 0.00045600242358960545,
"loss": 3.167,
"step": 35750
},
{
"epoch": 9.634151593453918,
"grad_norm": 0.3411939740180969,
"learning_rate": 0.00045580045778914765,
"loss": 3.1563,
"step": 35800
},
{
"epoch": 9.647609819121447,
"grad_norm": 0.3397693932056427,
"learning_rate": 0.00045559849198868984,
"loss": 3.1586,
"step": 35850
},
{
"epoch": 9.661068044788975,
"grad_norm": 0.3585307002067566,
"learning_rate": 0.00045539652618823204,
"loss": 3.1671,
"step": 35900
},
{
"epoch": 9.674526270456504,
"grad_norm": 0.33368492126464844,
"learning_rate": 0.0004551945603877743,
"loss": 3.1664,
"step": 35950
},
{
"epoch": 9.687984496124031,
"grad_norm": 0.34546005725860596,
"learning_rate": 0.00045499259458731654,
"loss": 3.1716,
"step": 36000
},
{
"epoch": 9.687984496124031,
"eval_accuracy": 0.39044713830701255,
"eval_loss": 3.309567928314209,
"eval_runtime": 82.8105,
"eval_samples_per_second": 217.509,
"eval_steps_per_second": 13.597,
"step": 36000
},
{
"epoch": 9.701442721791558,
"grad_norm": 0.33162131905555725,
"learning_rate": 0.00045479062878685874,
"loss": 3.1648,
"step": 36050
},
{
"epoch": 9.714900947459087,
"grad_norm": 0.3669474422931671,
"learning_rate": 0.00045458866298640093,
"loss": 3.1775,
"step": 36100
},
{
"epoch": 9.728359173126615,
"grad_norm": 0.3566083014011383,
"learning_rate": 0.0004543866971859432,
"loss": 3.1657,
"step": 36150
},
{
"epoch": 9.741817398794144,
"grad_norm": 0.332046240568161,
"learning_rate": 0.0004541847313854854,
"loss": 3.1742,
"step": 36200
},
{
"epoch": 9.755275624461671,
"grad_norm": 0.33060288429260254,
"learning_rate": 0.0004539827655850276,
"loss": 3.1663,
"step": 36250
},
{
"epoch": 9.768733850129198,
"grad_norm": 0.34776023030281067,
"learning_rate": 0.00045378079978456977,
"loss": 3.1825,
"step": 36300
},
{
"epoch": 9.782192075796727,
"grad_norm": 0.34656497836112976,
"learning_rate": 0.00045357883398411197,
"loss": 3.1883,
"step": 36350
},
{
"epoch": 9.795650301464255,
"grad_norm": 0.32794493436813354,
"learning_rate": 0.0004533768681836542,
"loss": 3.1813,
"step": 36400
},
{
"epoch": 9.809108527131784,
"grad_norm": 0.3393864333629608,
"learning_rate": 0.0004531749023831964,
"loss": 3.1759,
"step": 36450
},
{
"epoch": 9.822566752799311,
"grad_norm": 0.3410272002220154,
"learning_rate": 0.0004529729365827386,
"loss": 3.1715,
"step": 36500
},
{
"epoch": 9.836024978466838,
"grad_norm": 0.322590708732605,
"learning_rate": 0.0004527709707822808,
"loss": 3.172,
"step": 36550
},
{
"epoch": 9.849483204134367,
"grad_norm": 0.3111822307109833,
"learning_rate": 0.000452569004981823,
"loss": 3.1772,
"step": 36600
},
{
"epoch": 9.862941429801895,
"grad_norm": 0.340970516204834,
"learning_rate": 0.00045236703918136525,
"loss": 3.1627,
"step": 36650
},
{
"epoch": 9.876399655469424,
"grad_norm": 0.35041001439094543,
"learning_rate": 0.00045216507338090745,
"loss": 3.1777,
"step": 36700
},
{
"epoch": 9.889857881136951,
"grad_norm": 0.3762494623661041,
"learning_rate": 0.00045196310758044964,
"loss": 3.181,
"step": 36750
},
{
"epoch": 9.903316106804478,
"grad_norm": 0.3204391598701477,
"learning_rate": 0.00045176114177999184,
"loss": 3.187,
"step": 36800
},
{
"epoch": 9.916774332472007,
"grad_norm": 0.3289971649646759,
"learning_rate": 0.00045155917597953404,
"loss": 3.178,
"step": 36850
},
{
"epoch": 9.930232558139535,
"grad_norm": 0.335293173789978,
"learning_rate": 0.00045135721017907634,
"loss": 3.1753,
"step": 36900
},
{
"epoch": 9.943690783807062,
"grad_norm": 0.3508659899234772,
"learning_rate": 0.00045115524437861854,
"loss": 3.1786,
"step": 36950
},
{
"epoch": 9.957149009474591,
"grad_norm": 0.3186536729335785,
"learning_rate": 0.00045095327857816073,
"loss": 3.1763,
"step": 37000
},
{
"epoch": 9.957149009474591,
"eval_accuracy": 0.3909272487617518,
"eval_loss": 3.302363157272339,
"eval_runtime": 82.7481,
"eval_samples_per_second": 217.673,
"eval_steps_per_second": 13.608,
"step": 37000
},
{
"epoch": 9.970607235142118,
"grad_norm": 0.3364878296852112,
"learning_rate": 0.00045075131277770293,
"loss": 3.1763,
"step": 37050
},
{
"epoch": 9.984065460809648,
"grad_norm": 0.3121895492076874,
"learning_rate": 0.0004505493469772452,
"loss": 3.1863,
"step": 37100
},
{
"epoch": 9.997523686477175,
"grad_norm": 0.34475064277648926,
"learning_rate": 0.0004503473811767874,
"loss": 3.1734,
"step": 37150
},
{
"epoch": 10.010766580534023,
"grad_norm": 0.34318941831588745,
"learning_rate": 0.00045014541537632957,
"loss": 3.0841,
"step": 37200
},
{
"epoch": 10.02422480620155,
"grad_norm": 0.3646996319293976,
"learning_rate": 0.00044994344957587177,
"loss": 3.0728,
"step": 37250
},
{
"epoch": 10.037683031869078,
"grad_norm": 0.35196927189826965,
"learning_rate": 0.00044974148377541396,
"loss": 3.0809,
"step": 37300
},
{
"epoch": 10.051141257536607,
"grad_norm": 0.35955750942230225,
"learning_rate": 0.0004495395179749562,
"loss": 3.0753,
"step": 37350
},
{
"epoch": 10.064599483204134,
"grad_norm": 0.36974605917930603,
"learning_rate": 0.0004493375521744984,
"loss": 3.0905,
"step": 37400
},
{
"epoch": 10.078057708871663,
"grad_norm": 0.36986610293388367,
"learning_rate": 0.0004491355863740406,
"loss": 3.1015,
"step": 37450
},
{
"epoch": 10.09151593453919,
"grad_norm": 0.37228667736053467,
"learning_rate": 0.0004489336205735828,
"loss": 3.0933,
"step": 37500
},
{
"epoch": 10.104974160206718,
"grad_norm": 0.3396550416946411,
"learning_rate": 0.00044873165477312505,
"loss": 3.1091,
"step": 37550
},
{
"epoch": 10.118432385874247,
"grad_norm": 0.34802475571632385,
"learning_rate": 0.00044852968897266725,
"loss": 3.1092,
"step": 37600
},
{
"epoch": 10.131890611541774,
"grad_norm": 0.3393282890319824,
"learning_rate": 0.00044832772317220945,
"loss": 3.109,
"step": 37650
},
{
"epoch": 10.145348837209303,
"grad_norm": 0.3503858149051666,
"learning_rate": 0.00044812575737175164,
"loss": 3.1161,
"step": 37700
},
{
"epoch": 10.15880706287683,
"grad_norm": 0.32870590686798096,
"learning_rate": 0.00044792379157129384,
"loss": 3.1062,
"step": 37750
},
{
"epoch": 10.172265288544358,
"grad_norm": 0.3703332841396332,
"learning_rate": 0.00044772182577083614,
"loss": 3.0957,
"step": 37800
},
{
"epoch": 10.185723514211887,
"grad_norm": 0.33780622482299805,
"learning_rate": 0.00044751985997037834,
"loss": 3.1039,
"step": 37850
},
{
"epoch": 10.199181739879414,
"grad_norm": 0.33272576332092285,
"learning_rate": 0.00044731789416992053,
"loss": 3.1089,
"step": 37900
},
{
"epoch": 10.212639965546943,
"grad_norm": 0.3478260934352875,
"learning_rate": 0.00044711592836946273,
"loss": 3.1104,
"step": 37950
},
{
"epoch": 10.22609819121447,
"grad_norm": 0.3370334804058075,
"learning_rate": 0.000446913962569005,
"loss": 3.1198,
"step": 38000
},
{
"epoch": 10.22609819121447,
"eval_accuracy": 0.39075862884191576,
"eval_loss": 3.313037872314453,
"eval_runtime": 82.9638,
"eval_samples_per_second": 217.107,
"eval_steps_per_second": 13.572,
"step": 38000
},
{
"epoch": 10.239556416881998,
"grad_norm": 0.35148367285728455,
"learning_rate": 0.0004467119967685472,
"loss": 3.1108,
"step": 38050
},
{
"epoch": 10.253014642549527,
"grad_norm": 0.3399736285209656,
"learning_rate": 0.00044651003096808937,
"loss": 3.1131,
"step": 38100
},
{
"epoch": 10.266472868217054,
"grad_norm": 0.3101232051849365,
"learning_rate": 0.00044630806516763157,
"loss": 3.1262,
"step": 38150
},
{
"epoch": 10.279931093884581,
"grad_norm": 0.3368324637413025,
"learning_rate": 0.00044610609936717376,
"loss": 3.1267,
"step": 38200
},
{
"epoch": 10.29338931955211,
"grad_norm": 0.3308991491794586,
"learning_rate": 0.000445904133566716,
"loss": 3.1136,
"step": 38250
},
{
"epoch": 10.306847545219638,
"grad_norm": 0.35431280732154846,
"learning_rate": 0.0004457021677662582,
"loss": 3.1243,
"step": 38300
},
{
"epoch": 10.320305770887167,
"grad_norm": 0.37211817502975464,
"learning_rate": 0.0004455002019658004,
"loss": 3.1135,
"step": 38350
},
{
"epoch": 10.333763996554694,
"grad_norm": 0.3618927597999573,
"learning_rate": 0.0004452982361653426,
"loss": 3.1335,
"step": 38400
},
{
"epoch": 10.347222222222221,
"grad_norm": 0.3535788357257843,
"learning_rate": 0.0004450962703648848,
"loss": 3.131,
"step": 38450
},
{
"epoch": 10.36068044788975,
"grad_norm": 0.35206955671310425,
"learning_rate": 0.00044489430456442705,
"loss": 3.1197,
"step": 38500
},
{
"epoch": 10.374138673557278,
"grad_norm": 0.36820894479751587,
"learning_rate": 0.00044469233876396925,
"loss": 3.1139,
"step": 38550
},
{
"epoch": 10.387596899224807,
"grad_norm": 0.34299564361572266,
"learning_rate": 0.00044449037296351144,
"loss": 3.1352,
"step": 38600
},
{
"epoch": 10.401055124892334,
"grad_norm": 0.3333747386932373,
"learning_rate": 0.0004442884071630537,
"loss": 3.1293,
"step": 38650
},
{
"epoch": 10.414513350559861,
"grad_norm": 0.31069228053092957,
"learning_rate": 0.00044408644136259594,
"loss": 3.1378,
"step": 38700
},
{
"epoch": 10.42797157622739,
"grad_norm": 0.34444460272789,
"learning_rate": 0.00044388447556213814,
"loss": 3.1333,
"step": 38750
},
{
"epoch": 10.441429801894918,
"grad_norm": 0.35531967878341675,
"learning_rate": 0.00044368250976168033,
"loss": 3.1249,
"step": 38800
},
{
"epoch": 10.454888027562447,
"grad_norm": 0.3596295714378357,
"learning_rate": 0.00044348054396122253,
"loss": 3.1299,
"step": 38850
},
{
"epoch": 10.468346253229974,
"grad_norm": 0.3504672944545746,
"learning_rate": 0.0004432785781607647,
"loss": 3.1348,
"step": 38900
},
{
"epoch": 10.481804478897502,
"grad_norm": 0.3549274802207947,
"learning_rate": 0.000443076612360307,
"loss": 3.1449,
"step": 38950
},
{
"epoch": 10.49526270456503,
"grad_norm": 0.3380185067653656,
"learning_rate": 0.0004428746465598492,
"loss": 3.123,
"step": 39000
},
{
"epoch": 10.49526270456503,
"eval_accuracy": 0.39101927264841485,
"eval_loss": 3.306272268295288,
"eval_runtime": 82.8378,
"eval_samples_per_second": 217.437,
"eval_steps_per_second": 13.593,
"step": 39000
},
{
"epoch": 10.508720930232558,
"grad_norm": 0.3382558226585388,
"learning_rate": 0.00044267268075939137,
"loss": 3.1348,
"step": 39050
},
{
"epoch": 10.522179155900087,
"grad_norm": 0.3373892307281494,
"learning_rate": 0.00044247071495893357,
"loss": 3.1429,
"step": 39100
},
{
"epoch": 10.535637381567614,
"grad_norm": 0.35838979482650757,
"learning_rate": 0.00044226874915847576,
"loss": 3.1441,
"step": 39150
},
{
"epoch": 10.549095607235142,
"grad_norm": 0.3495882749557495,
"learning_rate": 0.000442066783358018,
"loss": 3.1375,
"step": 39200
},
{
"epoch": 10.56255383290267,
"grad_norm": 0.352803498506546,
"learning_rate": 0.0004418648175575602,
"loss": 3.1425,
"step": 39250
},
{
"epoch": 10.576012058570198,
"grad_norm": 0.34696337580680847,
"learning_rate": 0.0004416628517571024,
"loss": 3.1525,
"step": 39300
},
{
"epoch": 10.589470284237725,
"grad_norm": 0.3385176360607147,
"learning_rate": 0.0004414608859566446,
"loss": 3.1492,
"step": 39350
},
{
"epoch": 10.602928509905254,
"grad_norm": 0.34304797649383545,
"learning_rate": 0.00044125892015618685,
"loss": 3.1365,
"step": 39400
},
{
"epoch": 10.616386735572782,
"grad_norm": 0.32966360449790955,
"learning_rate": 0.00044105695435572905,
"loss": 3.1466,
"step": 39450
},
{
"epoch": 10.62984496124031,
"grad_norm": 0.3779182732105255,
"learning_rate": 0.00044085498855527124,
"loss": 3.1443,
"step": 39500
},
{
"epoch": 10.643303186907838,
"grad_norm": 0.3469052314758301,
"learning_rate": 0.0004406530227548135,
"loss": 3.1548,
"step": 39550
},
{
"epoch": 10.656761412575365,
"grad_norm": 0.35312896966934204,
"learning_rate": 0.00044045105695435574,
"loss": 3.1479,
"step": 39600
},
{
"epoch": 10.670219638242894,
"grad_norm": 0.3488466143608093,
"learning_rate": 0.00044024909115389794,
"loss": 3.1633,
"step": 39650
},
{
"epoch": 10.683677863910422,
"grad_norm": 0.35734888911247253,
"learning_rate": 0.00044004712535344014,
"loss": 3.1414,
"step": 39700
},
{
"epoch": 10.69713608957795,
"grad_norm": 0.31818750500679016,
"learning_rate": 0.00043984515955298233,
"loss": 3.1447,
"step": 39750
},
{
"epoch": 10.710594315245478,
"grad_norm": 0.33872801065444946,
"learning_rate": 0.00043964319375252453,
"loss": 3.1523,
"step": 39800
},
{
"epoch": 10.724052540913005,
"grad_norm": 0.3522094786167145,
"learning_rate": 0.0004394412279520668,
"loss": 3.1499,
"step": 39850
},
{
"epoch": 10.737510766580534,
"grad_norm": 0.3578832149505615,
"learning_rate": 0.000439239262151609,
"loss": 3.146,
"step": 39900
},
{
"epoch": 10.750968992248062,
"grad_norm": 0.31521254777908325,
"learning_rate": 0.00043903729635115117,
"loss": 3.1513,
"step": 39950
},
{
"epoch": 10.764427217915589,
"grad_norm": 0.37201380729675293,
"learning_rate": 0.00043883533055069337,
"loss": 3.1361,
"step": 40000
},
{
"epoch": 10.764427217915589,
"eval_accuracy": 0.39140964080045787,
"eval_loss": 3.3027617931365967,
"eval_runtime": 82.8483,
"eval_samples_per_second": 217.41,
"eval_steps_per_second": 13.591,
"step": 40000
},
{
"epoch": 10.777885443583118,
"grad_norm": 0.3271602988243103,
"learning_rate": 0.00043863336475023556,
"loss": 3.1424,
"step": 40050
},
{
"epoch": 10.791343669250645,
"grad_norm": 0.3698268234729767,
"learning_rate": 0.0004384313989497778,
"loss": 3.1458,
"step": 40100
},
{
"epoch": 10.804801894918175,
"grad_norm": 0.36806368827819824,
"learning_rate": 0.00043822943314932,
"loss": 3.1553,
"step": 40150
},
{
"epoch": 10.818260120585702,
"grad_norm": 0.35377177596092224,
"learning_rate": 0.0004380274673488622,
"loss": 3.1628,
"step": 40200
},
{
"epoch": 10.83171834625323,
"grad_norm": 0.35258588194847107,
"learning_rate": 0.0004378255015484044,
"loss": 3.1551,
"step": 40250
},
{
"epoch": 10.845176571920758,
"grad_norm": 0.3363988399505615,
"learning_rate": 0.0004376235357479466,
"loss": 3.1605,
"step": 40300
},
{
"epoch": 10.858634797588286,
"grad_norm": 0.3378644287586212,
"learning_rate": 0.00043742156994748885,
"loss": 3.1622,
"step": 40350
},
{
"epoch": 10.872093023255815,
"grad_norm": 0.3653063476085663,
"learning_rate": 0.00043721960414703104,
"loss": 3.1627,
"step": 40400
},
{
"epoch": 10.885551248923342,
"grad_norm": 0.3499191701412201,
"learning_rate": 0.0004370176383465733,
"loss": 3.156,
"step": 40450
},
{
"epoch": 10.89900947459087,
"grad_norm": 0.3448904752731323,
"learning_rate": 0.0004368156725461155,
"loss": 3.1485,
"step": 40500
},
{
"epoch": 10.912467700258398,
"grad_norm": 0.36490872502326965,
"learning_rate": 0.00043661370674565774,
"loss": 3.1507,
"step": 40550
},
{
"epoch": 10.925925925925926,
"grad_norm": 0.3395492434501648,
"learning_rate": 0.00043641174094519994,
"loss": 3.153,
"step": 40600
},
{
"epoch": 10.939384151593455,
"grad_norm": 0.3651244342327118,
"learning_rate": 0.00043620977514474213,
"loss": 3.1623,
"step": 40650
},
{
"epoch": 10.952842377260982,
"grad_norm": 0.3531707227230072,
"learning_rate": 0.00043600780934428433,
"loss": 3.1413,
"step": 40700
},
{
"epoch": 10.96630060292851,
"grad_norm": 0.3748357594013214,
"learning_rate": 0.0004358058435438265,
"loss": 3.158,
"step": 40750
},
{
"epoch": 10.979758828596038,
"grad_norm": 0.3381361663341522,
"learning_rate": 0.0004356038777433688,
"loss": 3.1536,
"step": 40800
},
{
"epoch": 10.993217054263566,
"grad_norm": 0.3460608720779419,
"learning_rate": 0.00043540191194291097,
"loss": 3.1636,
"step": 40850
},
{
"epoch": 11.006459948320414,
"grad_norm": 0.3407193124294281,
"learning_rate": 0.00043519994614245317,
"loss": 3.114,
"step": 40900
},
{
"epoch": 11.019918173987941,
"grad_norm": 0.3596533536911011,
"learning_rate": 0.00043499798034199536,
"loss": 3.0561,
"step": 40950
},
{
"epoch": 11.03337639965547,
"grad_norm": 0.34655600786209106,
"learning_rate": 0.00043479601454153756,
"loss": 3.0629,
"step": 41000
},
{
"epoch": 11.03337639965547,
"eval_accuracy": 0.39149753610660953,
"eval_loss": 3.308816909790039,
"eval_runtime": 82.7392,
"eval_samples_per_second": 217.696,
"eval_steps_per_second": 13.609,
"step": 41000
},
{
"epoch": 11.046834625322997,
"grad_norm": 0.3394680321216583,
"learning_rate": 0.0004345940487410798,
"loss": 3.0624,
"step": 41050
},
{
"epoch": 11.060292850990525,
"grad_norm": 0.3626207411289215,
"learning_rate": 0.000434392082940622,
"loss": 3.064,
"step": 41100
},
{
"epoch": 11.073751076658054,
"grad_norm": 0.36266371607780457,
"learning_rate": 0.0004341901171401642,
"loss": 3.0769,
"step": 41150
},
{
"epoch": 11.087209302325581,
"grad_norm": 0.3615586459636688,
"learning_rate": 0.0004339881513397064,
"loss": 3.0719,
"step": 41200
},
{
"epoch": 11.10066752799311,
"grad_norm": 0.3421686589717865,
"learning_rate": 0.0004337861855392486,
"loss": 3.075,
"step": 41250
},
{
"epoch": 11.114125753660637,
"grad_norm": 0.36166146397590637,
"learning_rate": 0.00043358421973879084,
"loss": 3.0721,
"step": 41300
},
{
"epoch": 11.127583979328165,
"grad_norm": 0.3366292417049408,
"learning_rate": 0.0004333822539383331,
"loss": 3.0831,
"step": 41350
},
{
"epoch": 11.141042204995694,
"grad_norm": 0.34206289052963257,
"learning_rate": 0.0004331802881378753,
"loss": 3.0791,
"step": 41400
},
{
"epoch": 11.154500430663221,
"grad_norm": 0.3510792553424835,
"learning_rate": 0.00043297832233741754,
"loss": 3.0935,
"step": 41450
},
{
"epoch": 11.16795865633075,
"grad_norm": 0.3992671072483063,
"learning_rate": 0.00043277635653695974,
"loss": 3.0833,
"step": 41500
},
{
"epoch": 11.181416881998278,
"grad_norm": 0.3501514494419098,
"learning_rate": 0.00043257439073650193,
"loss": 3.097,
"step": 41550
},
{
"epoch": 11.194875107665805,
"grad_norm": 0.33838045597076416,
"learning_rate": 0.00043237242493604413,
"loss": 3.0872,
"step": 41600
},
{
"epoch": 11.208333333333334,
"grad_norm": 0.3412204086780548,
"learning_rate": 0.0004321704591355863,
"loss": 3.0981,
"step": 41650
},
{
"epoch": 11.221791559000861,
"grad_norm": 0.3283551037311554,
"learning_rate": 0.0004319684933351286,
"loss": 3.0979,
"step": 41700
},
{
"epoch": 11.235249784668389,
"grad_norm": 0.34778767824172974,
"learning_rate": 0.00043176652753467077,
"loss": 3.0971,
"step": 41750
},
{
"epoch": 11.248708010335918,
"grad_norm": 0.39209914207458496,
"learning_rate": 0.00043156456173421297,
"loss": 3.1069,
"step": 41800
},
{
"epoch": 11.262166236003445,
"grad_norm": 0.3595452606678009,
"learning_rate": 0.00043136259593375516,
"loss": 3.1021,
"step": 41850
},
{
"epoch": 11.275624461670974,
"grad_norm": 0.33872005343437195,
"learning_rate": 0.00043116063013329736,
"loss": 3.1054,
"step": 41900
},
{
"epoch": 11.289082687338501,
"grad_norm": 0.3570093810558319,
"learning_rate": 0.0004309586643328396,
"loss": 3.0961,
"step": 41950
},
{
"epoch": 11.302540913006029,
"grad_norm": 0.34249362349510193,
"learning_rate": 0.0004307566985323818,
"loss": 3.0983,
"step": 42000
},
{
"epoch": 11.302540913006029,
"eval_accuracy": 0.3915293696352899,
"eval_loss": 3.3084728717803955,
"eval_runtime": 82.7153,
"eval_samples_per_second": 217.759,
"eval_steps_per_second": 13.613,
"step": 42000
},
{
"epoch": 11.315999138673558,
"grad_norm": 0.37006908655166626,
"learning_rate": 0.000430554732731924,
"loss": 3.1002,
"step": 42050
},
{
"epoch": 11.329457364341085,
"grad_norm": 0.33864548802375793,
"learning_rate": 0.0004303527669314662,
"loss": 3.0924,
"step": 42100
},
{
"epoch": 11.342915590008614,
"grad_norm": 0.35206928849220276,
"learning_rate": 0.0004301508011310084,
"loss": 3.1047,
"step": 42150
},
{
"epoch": 11.356373815676141,
"grad_norm": 0.33268919587135315,
"learning_rate": 0.00042994883533055065,
"loss": 3.1103,
"step": 42200
},
{
"epoch": 11.369832041343669,
"grad_norm": 0.38763168454170227,
"learning_rate": 0.0004297468695300929,
"loss": 3.1093,
"step": 42250
},
{
"epoch": 11.383290267011198,
"grad_norm": 0.38409027457237244,
"learning_rate": 0.0004295449037296351,
"loss": 3.1165,
"step": 42300
},
{
"epoch": 11.396748492678725,
"grad_norm": 0.33871376514434814,
"learning_rate": 0.0004293429379291773,
"loss": 3.1139,
"step": 42350
},
{
"epoch": 11.410206718346252,
"grad_norm": 0.3507962226867676,
"learning_rate": 0.00042914097212871954,
"loss": 3.1039,
"step": 42400
},
{
"epoch": 11.423664944013781,
"grad_norm": 0.36563482880592346,
"learning_rate": 0.00042893900632826173,
"loss": 3.1202,
"step": 42450
},
{
"epoch": 11.437123169681309,
"grad_norm": 0.35809335112571716,
"learning_rate": 0.00042873704052780393,
"loss": 3.1157,
"step": 42500
},
{
"epoch": 11.450581395348838,
"grad_norm": 0.3515256345272064,
"learning_rate": 0.0004285350747273461,
"loss": 3.1138,
"step": 42550
},
{
"epoch": 11.464039621016365,
"grad_norm": 0.36794790625572205,
"learning_rate": 0.0004283331089268883,
"loss": 3.1178,
"step": 42600
},
{
"epoch": 11.477497846683892,
"grad_norm": 0.32427987456321716,
"learning_rate": 0.00042813114312643057,
"loss": 3.1236,
"step": 42650
},
{
"epoch": 11.490956072351421,
"grad_norm": 0.3686062693595886,
"learning_rate": 0.00042792917732597277,
"loss": 3.1194,
"step": 42700
},
{
"epoch": 11.504414298018949,
"grad_norm": 0.3622916340827942,
"learning_rate": 0.00042772721152551496,
"loss": 3.1253,
"step": 42750
},
{
"epoch": 11.517872523686478,
"grad_norm": 0.3762475848197937,
"learning_rate": 0.00042752524572505716,
"loss": 3.1187,
"step": 42800
},
{
"epoch": 11.531330749354005,
"grad_norm": 0.35526514053344727,
"learning_rate": 0.00042732327992459936,
"loss": 3.1194,
"step": 42850
},
{
"epoch": 11.544788975021532,
"grad_norm": 0.3318747878074646,
"learning_rate": 0.0004271213141241416,
"loss": 3.1224,
"step": 42900
},
{
"epoch": 11.558247200689062,
"grad_norm": 0.35735106468200684,
"learning_rate": 0.0004269193483236838,
"loss": 3.1281,
"step": 42950
},
{
"epoch": 11.571705426356589,
"grad_norm": 0.3457440137863159,
"learning_rate": 0.000426717382523226,
"loss": 3.1164,
"step": 43000
},
{
"epoch": 11.571705426356589,
"eval_accuracy": 0.3923172766318432,
"eval_loss": 3.298119068145752,
"eval_runtime": 82.8384,
"eval_samples_per_second": 217.435,
"eval_steps_per_second": 13.593,
"step": 43000
},
{
"epoch": 11.585163652024118,
"grad_norm": 0.36481255292892456,
"learning_rate": 0.0004265154167227682,
"loss": 3.1324,
"step": 43050
},
{
"epoch": 11.598621877691645,
"grad_norm": 0.34653860330581665,
"learning_rate": 0.0004263134509223105,
"loss": 3.1315,
"step": 43100
},
{
"epoch": 11.612080103359173,
"grad_norm": 0.32992926239967346,
"learning_rate": 0.0004261114851218527,
"loss": 3.123,
"step": 43150
},
{
"epoch": 11.625538329026702,
"grad_norm": 0.3616025447845459,
"learning_rate": 0.0004259095193213949,
"loss": 3.1376,
"step": 43200
},
{
"epoch": 11.638996554694229,
"grad_norm": 0.33698633313179016,
"learning_rate": 0.0004257075535209371,
"loss": 3.1265,
"step": 43250
},
{
"epoch": 11.652454780361758,
"grad_norm": 0.3581075072288513,
"learning_rate": 0.00042550558772047934,
"loss": 3.1311,
"step": 43300
},
{
"epoch": 11.665913006029285,
"grad_norm": 0.35785016417503357,
"learning_rate": 0.00042530362192002153,
"loss": 3.1208,
"step": 43350
},
{
"epoch": 11.679371231696813,
"grad_norm": 0.331623375415802,
"learning_rate": 0.00042510165611956373,
"loss": 3.1139,
"step": 43400
},
{
"epoch": 11.692829457364342,
"grad_norm": 0.34417787194252014,
"learning_rate": 0.0004248996903191059,
"loss": 3.1337,
"step": 43450
},
{
"epoch": 11.706287683031869,
"grad_norm": 0.3683578372001648,
"learning_rate": 0.0004246977245186481,
"loss": 3.1348,
"step": 43500
},
{
"epoch": 11.719745908699396,
"grad_norm": 0.35859552025794983,
"learning_rate": 0.0004244957587181904,
"loss": 3.1222,
"step": 43550
},
{
"epoch": 11.733204134366925,
"grad_norm": 0.3777923285961151,
"learning_rate": 0.00042429379291773257,
"loss": 3.1266,
"step": 43600
},
{
"epoch": 11.746662360034453,
"grad_norm": 0.3375019431114197,
"learning_rate": 0.00042409182711727477,
"loss": 3.1221,
"step": 43650
},
{
"epoch": 11.760120585701982,
"grad_norm": 0.3501497507095337,
"learning_rate": 0.00042388986131681696,
"loss": 3.1278,
"step": 43700
},
{
"epoch": 11.773578811369509,
"grad_norm": 0.3535447418689728,
"learning_rate": 0.00042368789551635916,
"loss": 3.1269,
"step": 43750
},
{
"epoch": 11.787037037037036,
"grad_norm": 0.3669780492782593,
"learning_rate": 0.0004234859297159014,
"loss": 3.1266,
"step": 43800
},
{
"epoch": 11.800495262704565,
"grad_norm": 0.35164570808410645,
"learning_rate": 0.0004232839639154436,
"loss": 3.1338,
"step": 43850
},
{
"epoch": 11.813953488372093,
"grad_norm": 0.33973053097724915,
"learning_rate": 0.0004230819981149858,
"loss": 3.1328,
"step": 43900
},
{
"epoch": 11.827411714039622,
"grad_norm": 0.3561455309391022,
"learning_rate": 0.000422880032314528,
"loss": 3.1305,
"step": 43950
},
{
"epoch": 11.840869939707149,
"grad_norm": 0.352586567401886,
"learning_rate": 0.0004226780665140703,
"loss": 3.1439,
"step": 44000
},
{
"epoch": 11.840869939707149,
"eval_accuracy": 0.3925921531764212,
"eval_loss": 3.292262077331543,
"eval_runtime": 82.7924,
"eval_samples_per_second": 217.556,
"eval_steps_per_second": 13.6,
"step": 44000
},
{
"epoch": 11.854328165374676,
"grad_norm": 0.3365379869937897,
"learning_rate": 0.0004224761007136125,
"loss": 3.1369,
"step": 44050
},
{
"epoch": 11.867786391042205,
"grad_norm": 0.33547067642211914,
"learning_rate": 0.0004222741349131547,
"loss": 3.1347,
"step": 44100
},
{
"epoch": 11.881244616709733,
"grad_norm": 0.3591451644897461,
"learning_rate": 0.0004220721691126969,
"loss": 3.1392,
"step": 44150
},
{
"epoch": 11.89470284237726,
"grad_norm": 0.3269728720188141,
"learning_rate": 0.0004218702033122391,
"loss": 3.1342,
"step": 44200
},
{
"epoch": 11.90816106804479,
"grad_norm": 0.36158931255340576,
"learning_rate": 0.00042166823751178134,
"loss": 3.1366,
"step": 44250
},
{
"epoch": 11.921619293712316,
"grad_norm": 0.3639954626560211,
"learning_rate": 0.00042146627171132353,
"loss": 3.1482,
"step": 44300
},
{
"epoch": 11.935077519379846,
"grad_norm": 0.331939697265625,
"learning_rate": 0.00042126430591086573,
"loss": 3.1299,
"step": 44350
},
{
"epoch": 11.948535745047373,
"grad_norm": 0.34648680686950684,
"learning_rate": 0.0004210623401104079,
"loss": 3.1335,
"step": 44400
},
{
"epoch": 11.9619939707149,
"grad_norm": 0.36366772651672363,
"learning_rate": 0.0004208603743099501,
"loss": 3.149,
"step": 44450
},
{
"epoch": 11.97545219638243,
"grad_norm": 0.3582664430141449,
"learning_rate": 0.00042065840850949237,
"loss": 3.1362,
"step": 44500
},
{
"epoch": 11.988910422049956,
"grad_norm": 0.33306729793548584,
"learning_rate": 0.00042045644270903457,
"loss": 3.1433,
"step": 44550
},
{
"epoch": 12.002153316106805,
"grad_norm": 0.3570893704891205,
"learning_rate": 0.00042025447690857676,
"loss": 3.1241,
"step": 44600
},
{
"epoch": 12.015611541774332,
"grad_norm": 0.3521556556224823,
"learning_rate": 0.00042005251110811896,
"loss": 3.0449,
"step": 44650
},
{
"epoch": 12.029069767441861,
"grad_norm": 0.3729991018772125,
"learning_rate": 0.00041985054530766115,
"loss": 3.0456,
"step": 44700
},
{
"epoch": 12.042527993109388,
"grad_norm": 0.3906972110271454,
"learning_rate": 0.0004196485795072034,
"loss": 3.0451,
"step": 44750
},
{
"epoch": 12.055986218776916,
"grad_norm": 0.350547194480896,
"learning_rate": 0.0004194466137067456,
"loss": 3.0532,
"step": 44800
},
{
"epoch": 12.069444444444445,
"grad_norm": 0.3621971309185028,
"learning_rate": 0.0004192446479062878,
"loss": 3.0489,
"step": 44850
},
{
"epoch": 12.082902670111972,
"grad_norm": 0.3549562990665436,
"learning_rate": 0.00041904268210583005,
"loss": 3.0597,
"step": 44900
},
{
"epoch": 12.096360895779501,
"grad_norm": 0.35590115189552307,
"learning_rate": 0.0004188407163053723,
"loss": 3.0646,
"step": 44950
},
{
"epoch": 12.109819121447028,
"grad_norm": 0.36715343594551086,
"learning_rate": 0.0004186387505049145,
"loss": 3.0632,
"step": 45000
},
{
"epoch": 12.109819121447028,
"eval_accuracy": 0.39225317498706014,
"eval_loss": 3.3006749153137207,
"eval_runtime": 82.7457,
"eval_samples_per_second": 217.679,
"eval_steps_per_second": 13.608,
"step": 45000
},
{
"epoch": 12.123277347114556,
"grad_norm": 0.36294618248939514,
"learning_rate": 0.0004184367847044567,
"loss": 3.0651,
"step": 45050
},
{
"epoch": 12.136735572782085,
"grad_norm": 0.33780601620674133,
"learning_rate": 0.0004182348189039989,
"loss": 3.0628,
"step": 45100
},
{
"epoch": 12.150193798449612,
"grad_norm": 0.34479013085365295,
"learning_rate": 0.00041803285310354114,
"loss": 3.0603,
"step": 45150
},
{
"epoch": 12.163652024117141,
"grad_norm": 0.3604039251804352,
"learning_rate": 0.00041783088730308333,
"loss": 3.0646,
"step": 45200
},
{
"epoch": 12.177110249784668,
"grad_norm": 0.3572409749031067,
"learning_rate": 0.00041762892150262553,
"loss": 3.0722,
"step": 45250
},
{
"epoch": 12.190568475452196,
"grad_norm": 0.3671632409095764,
"learning_rate": 0.0004174269557021677,
"loss": 3.0728,
"step": 45300
},
{
"epoch": 12.204026701119725,
"grad_norm": 0.3517717123031616,
"learning_rate": 0.0004172249899017099,
"loss": 3.0864,
"step": 45350
},
{
"epoch": 12.217484926787252,
"grad_norm": 0.3621057868003845,
"learning_rate": 0.00041702302410125217,
"loss": 3.0702,
"step": 45400
},
{
"epoch": 12.230943152454781,
"grad_norm": 0.35359835624694824,
"learning_rate": 0.00041682105830079437,
"loss": 3.079,
"step": 45450
},
{
"epoch": 12.244401378122308,
"grad_norm": 0.35617008805274963,
"learning_rate": 0.00041661909250033656,
"loss": 3.0818,
"step": 45500
},
{
"epoch": 12.257859603789836,
"grad_norm": 0.3963318169116974,
"learning_rate": 0.00041641712669987876,
"loss": 3.0721,
"step": 45550
},
{
"epoch": 12.271317829457365,
"grad_norm": 0.3492389917373657,
"learning_rate": 0.00041621516089942096,
"loss": 3.0764,
"step": 45600
},
{
"epoch": 12.284776055124892,
"grad_norm": 0.34680983424186707,
"learning_rate": 0.0004160131950989632,
"loss": 3.0913,
"step": 45650
},
{
"epoch": 12.298234280792421,
"grad_norm": 0.3710663318634033,
"learning_rate": 0.0004158112292985054,
"loss": 3.0873,
"step": 45700
},
{
"epoch": 12.311692506459949,
"grad_norm": 0.36970847845077515,
"learning_rate": 0.0004156092634980476,
"loss": 3.0923,
"step": 45750
},
{
"epoch": 12.325150732127476,
"grad_norm": 0.3592429757118225,
"learning_rate": 0.00041540729769758985,
"loss": 3.0875,
"step": 45800
},
{
"epoch": 12.338608957795005,
"grad_norm": 0.33475932478904724,
"learning_rate": 0.0004152053318971321,
"loss": 3.0897,
"step": 45850
},
{
"epoch": 12.352067183462532,
"grad_norm": 0.3651215136051178,
"learning_rate": 0.0004150033660966743,
"loss": 3.0953,
"step": 45900
},
{
"epoch": 12.36552540913006,
"grad_norm": 0.339554101228714,
"learning_rate": 0.0004148014002962165,
"loss": 3.0796,
"step": 45950
},
{
"epoch": 12.378983634797589,
"grad_norm": 0.36364108324050903,
"learning_rate": 0.0004145994344957587,
"loss": 3.0919,
"step": 46000
},
{
"epoch": 12.378983634797589,
"eval_accuracy": 0.39245123820475414,
"eval_loss": 3.3007659912109375,
"eval_runtime": 82.8309,
"eval_samples_per_second": 217.455,
"eval_steps_per_second": 13.594,
"step": 46000
},
{
"epoch": 12.392441860465116,
"grad_norm": 0.34883204102516174,
"learning_rate": 0.0004143974686953009,
"loss": 3.103,
"step": 46050
},
{
"epoch": 12.405900086132645,
"grad_norm": 0.3627978265285492,
"learning_rate": 0.00041419550289484313,
"loss": 3.0913,
"step": 46100
},
{
"epoch": 12.419358311800172,
"grad_norm": 0.3386411964893341,
"learning_rate": 0.00041399353709438533,
"loss": 3.0991,
"step": 46150
},
{
"epoch": 12.4328165374677,
"grad_norm": 0.35564661026000977,
"learning_rate": 0.0004137915712939275,
"loss": 3.1039,
"step": 46200
},
{
"epoch": 12.446274763135229,
"grad_norm": 0.3770778775215149,
"learning_rate": 0.0004135896054934697,
"loss": 3.1024,
"step": 46250
},
{
"epoch": 12.459732988802756,
"grad_norm": 0.357135146856308,
"learning_rate": 0.0004133876396930119,
"loss": 3.1046,
"step": 46300
},
{
"epoch": 12.473191214470285,
"grad_norm": 0.3489512503147125,
"learning_rate": 0.00041318567389255417,
"loss": 3.0958,
"step": 46350
},
{
"epoch": 12.486649440137812,
"grad_norm": 0.3788180351257324,
"learning_rate": 0.00041298370809209636,
"loss": 3.1045,
"step": 46400
},
{
"epoch": 12.50010766580534,
"grad_norm": 0.3740611970424652,
"learning_rate": 0.00041278174229163856,
"loss": 3.0954,
"step": 46450
},
{
"epoch": 12.513565891472869,
"grad_norm": 0.3422425091266632,
"learning_rate": 0.00041257977649118076,
"loss": 3.0872,
"step": 46500
},
{
"epoch": 12.527024117140396,
"grad_norm": 0.3587570786476135,
"learning_rate": 0.00041237781069072295,
"loss": 3.1046,
"step": 46550
},
{
"epoch": 12.540482342807923,
"grad_norm": 0.33857262134552,
"learning_rate": 0.0004121758448902652,
"loss": 3.1131,
"step": 46600
},
{
"epoch": 12.553940568475452,
"grad_norm": 0.3531918227672577,
"learning_rate": 0.0004119738790898074,
"loss": 3.1021,
"step": 46650
},
{
"epoch": 12.56739879414298,
"grad_norm": 0.36435598134994507,
"learning_rate": 0.00041177191328934965,
"loss": 3.1062,
"step": 46700
},
{
"epoch": 12.580857019810509,
"grad_norm": 0.34663739800453186,
"learning_rate": 0.00041156994748889185,
"loss": 3.1095,
"step": 46750
},
{
"epoch": 12.594315245478036,
"grad_norm": 0.38091322779655457,
"learning_rate": 0.0004113679816884341,
"loss": 3.0985,
"step": 46800
},
{
"epoch": 12.607773471145563,
"grad_norm": 0.3616071343421936,
"learning_rate": 0.0004111660158879763,
"loss": 3.0941,
"step": 46850
},
{
"epoch": 12.621231696813092,
"grad_norm": 0.3553824722766876,
"learning_rate": 0.0004109640500875185,
"loss": 3.1063,
"step": 46900
},
{
"epoch": 12.63468992248062,
"grad_norm": 0.3609797954559326,
"learning_rate": 0.0004107620842870607,
"loss": 3.1042,
"step": 46950
},
{
"epoch": 12.648148148148149,
"grad_norm": 0.3720819652080536,
"learning_rate": 0.0004105601184866029,
"loss": 3.1095,
"step": 47000
},
{
"epoch": 12.648148148148149,
"eval_accuracy": 0.39278391487649245,
"eval_loss": 3.2952816486358643,
"eval_runtime": 83.049,
"eval_samples_per_second": 216.884,
"eval_steps_per_second": 13.558,
"step": 47000
},
{
"epoch": 12.661606373815676,
"grad_norm": 0.3690645396709442,
"learning_rate": 0.00041035815268614513,
"loss": 3.1115,
"step": 47050
},
{
"epoch": 12.675064599483203,
"grad_norm": 0.3253531754016876,
"learning_rate": 0.0004101561868856873,
"loss": 3.1132,
"step": 47100
},
{
"epoch": 12.688522825150732,
"grad_norm": 0.33710092306137085,
"learning_rate": 0.0004099542210852295,
"loss": 3.1103,
"step": 47150
},
{
"epoch": 12.70198105081826,
"grad_norm": 0.37352409958839417,
"learning_rate": 0.0004097522552847717,
"loss": 3.1083,
"step": 47200
},
{
"epoch": 12.715439276485789,
"grad_norm": 0.3545651137828827,
"learning_rate": 0.00040955028948431397,
"loss": 3.1114,
"step": 47250
},
{
"epoch": 12.728897502153316,
"grad_norm": 0.3472464382648468,
"learning_rate": 0.00040934832368385616,
"loss": 3.1107,
"step": 47300
},
{
"epoch": 12.742355727820843,
"grad_norm": 0.38446712493896484,
"learning_rate": 0.00040914635788339836,
"loss": 3.1136,
"step": 47350
},
{
"epoch": 12.755813953488373,
"grad_norm": 0.3541007339954376,
"learning_rate": 0.00040894439208294056,
"loss": 3.1149,
"step": 47400
},
{
"epoch": 12.7692721791559,
"grad_norm": 0.3710373342037201,
"learning_rate": 0.00040874242628248275,
"loss": 3.1158,
"step": 47450
},
{
"epoch": 12.782730404823429,
"grad_norm": 0.353493332862854,
"learning_rate": 0.000408540460482025,
"loss": 3.1182,
"step": 47500
},
{
"epoch": 12.796188630490956,
"grad_norm": 0.3581356108188629,
"learning_rate": 0.00040833849468156725,
"loss": 3.1173,
"step": 47550
},
{
"epoch": 12.809646856158484,
"grad_norm": 0.3793380558490753,
"learning_rate": 0.00040813652888110945,
"loss": 3.1165,
"step": 47600
},
{
"epoch": 12.823105081826013,
"grad_norm": 0.3635779321193695,
"learning_rate": 0.00040793456308065165,
"loss": 3.117,
"step": 47650
},
{
"epoch": 12.83656330749354,
"grad_norm": 0.3415304720401764,
"learning_rate": 0.0004077325972801939,
"loss": 3.1243,
"step": 47700
},
{
"epoch": 12.850021533161069,
"grad_norm": 0.3555426597595215,
"learning_rate": 0.0004075306314797361,
"loss": 3.1296,
"step": 47750
},
{
"epoch": 12.863479758828596,
"grad_norm": 0.35386669635772705,
"learning_rate": 0.0004073286656792783,
"loss": 3.1166,
"step": 47800
},
{
"epoch": 12.876937984496124,
"grad_norm": 0.38950422406196594,
"learning_rate": 0.0004071266998788205,
"loss": 3.1193,
"step": 47850
},
{
"epoch": 12.890396210163653,
"grad_norm": 0.35458096861839294,
"learning_rate": 0.0004069247340783627,
"loss": 3.1197,
"step": 47900
},
{
"epoch": 12.90385443583118,
"grad_norm": 0.3504520654678345,
"learning_rate": 0.00040672276827790493,
"loss": 3.1118,
"step": 47950
},
{
"epoch": 12.917312661498707,
"grad_norm": 0.3642725944519043,
"learning_rate": 0.0004065208024774471,
"loss": 3.1233,
"step": 48000
},
{
"epoch": 12.917312661498707,
"eval_accuracy": 0.39366156417574194,
"eval_loss": 3.2877914905548096,
"eval_runtime": 82.7887,
"eval_samples_per_second": 217.566,
"eval_steps_per_second": 13.601,
"step": 48000
},
{
"epoch": 12.930770887166236,
"grad_norm": 0.3657045364379883,
"learning_rate": 0.0004063188366769893,
"loss": 3.122,
"step": 48050
},
{
"epoch": 12.944229112833764,
"grad_norm": 0.34448593854904175,
"learning_rate": 0.0004061168708765315,
"loss": 3.1224,
"step": 48100
},
{
"epoch": 12.957687338501293,
"grad_norm": 0.38365674018859863,
"learning_rate": 0.0004059149050760737,
"loss": 3.1203,
"step": 48150
},
{
"epoch": 12.97114556416882,
"grad_norm": 0.3415539562702179,
"learning_rate": 0.00040571293927561597,
"loss": 3.1285,
"step": 48200
},
{
"epoch": 12.984603789836347,
"grad_norm": 0.3627220094203949,
"learning_rate": 0.00040551097347515816,
"loss": 3.1224,
"step": 48250
},
{
"epoch": 12.998062015503876,
"grad_norm": 0.3412235677242279,
"learning_rate": 0.00040530900767470036,
"loss": 3.1219,
"step": 48300
},
{
"epoch": 13.011304909560723,
"grad_norm": 0.35048797726631165,
"learning_rate": 0.00040510704187424255,
"loss": 3.033,
"step": 48350
},
{
"epoch": 13.024763135228252,
"grad_norm": 0.3464799225330353,
"learning_rate": 0.00040490507607378475,
"loss": 3.0172,
"step": 48400
},
{
"epoch": 13.038221360895779,
"grad_norm": 0.3651260435581207,
"learning_rate": 0.00040470311027332705,
"loss": 3.0304,
"step": 48450
},
{
"epoch": 13.051679586563308,
"grad_norm": 0.3380488455295563,
"learning_rate": 0.00040450114447286925,
"loss": 3.0358,
"step": 48500
},
{
"epoch": 13.065137812230835,
"grad_norm": 0.3696550726890564,
"learning_rate": 0.00040429917867241145,
"loss": 3.0379,
"step": 48550
},
{
"epoch": 13.078596037898363,
"grad_norm": 0.347341388463974,
"learning_rate": 0.00040409721287195364,
"loss": 3.0388,
"step": 48600
},
{
"epoch": 13.092054263565892,
"grad_norm": 0.34176427125930786,
"learning_rate": 0.0004038952470714959,
"loss": 3.0401,
"step": 48650
},
{
"epoch": 13.10551248923342,
"grad_norm": 0.36627480387687683,
"learning_rate": 0.0004036932812710381,
"loss": 3.0455,
"step": 48700
},
{
"epoch": 13.118970714900948,
"grad_norm": 0.34817516803741455,
"learning_rate": 0.0004034913154705803,
"loss": 3.0414,
"step": 48750
},
{
"epoch": 13.132428940568476,
"grad_norm": 0.3537481725215912,
"learning_rate": 0.0004032893496701225,
"loss": 3.0523,
"step": 48800
},
{
"epoch": 13.145887166236003,
"grad_norm": 0.36091747879981995,
"learning_rate": 0.0004030873838696647,
"loss": 3.0354,
"step": 48850
},
{
"epoch": 13.159345391903532,
"grad_norm": 0.34992292523384094,
"learning_rate": 0.00040288541806920693,
"loss": 3.0552,
"step": 48900
},
{
"epoch": 13.17280361757106,
"grad_norm": 0.41581109166145325,
"learning_rate": 0.0004026834522687491,
"loss": 3.0606,
"step": 48950
},
{
"epoch": 13.186261843238588,
"grad_norm": 0.35484784841537476,
"learning_rate": 0.0004024814864682913,
"loss": 3.0494,
"step": 49000
},
{
"epoch": 13.186261843238588,
"eval_accuracy": 0.3928094468875501,
"eval_loss": 3.302499771118164,
"eval_runtime": 83.312,
"eval_samples_per_second": 216.199,
"eval_steps_per_second": 13.515,
"step": 49000
},
{
"epoch": 13.199720068906116,
"grad_norm": 0.3928823173046112,
"learning_rate": 0.0004022795206678335,
"loss": 3.0609,
"step": 49050
},
{
"epoch": 13.213178294573643,
"grad_norm": 0.3489074110984802,
"learning_rate": 0.00040207755486737577,
"loss": 3.0493,
"step": 49100
},
{
"epoch": 13.226636520241172,
"grad_norm": 0.3712620139122009,
"learning_rate": 0.00040187558906691796,
"loss": 3.0701,
"step": 49150
},
{
"epoch": 13.2400947459087,
"grad_norm": 0.3899565637111664,
"learning_rate": 0.00040167362326646016,
"loss": 3.0707,
"step": 49200
},
{
"epoch": 13.253552971576227,
"grad_norm": 0.36969685554504395,
"learning_rate": 0.00040147165746600235,
"loss": 3.0602,
"step": 49250
},
{
"epoch": 13.267011197243756,
"grad_norm": 0.3676225244998932,
"learning_rate": 0.00040126969166554455,
"loss": 3.0706,
"step": 49300
},
{
"epoch": 13.280469422911283,
"grad_norm": 0.351186603307724,
"learning_rate": 0.00040106772586508686,
"loss": 3.0626,
"step": 49350
},
{
"epoch": 13.293927648578812,
"grad_norm": 0.3545724153518677,
"learning_rate": 0.00040086576006462905,
"loss": 3.072,
"step": 49400
},
{
"epoch": 13.30738587424634,
"grad_norm": 0.3635029196739197,
"learning_rate": 0.00040066379426417125,
"loss": 3.0702,
"step": 49450
},
{
"epoch": 13.320844099913867,
"grad_norm": 0.4017230272293091,
"learning_rate": 0.00040046182846371344,
"loss": 3.0667,
"step": 49500
},
{
"epoch": 13.334302325581396,
"grad_norm": 0.36741799116134644,
"learning_rate": 0.0004002598626632557,
"loss": 3.0786,
"step": 49550
},
{
"epoch": 13.347760551248923,
"grad_norm": 0.34630507230758667,
"learning_rate": 0.0004000578968627979,
"loss": 3.0818,
"step": 49600
},
{
"epoch": 13.361218776916452,
"grad_norm": 0.3642801344394684,
"learning_rate": 0.0003998559310623401,
"loss": 3.0822,
"step": 49650
},
{
"epoch": 13.37467700258398,
"grad_norm": 0.3653230667114258,
"learning_rate": 0.0003996539652618823,
"loss": 3.0851,
"step": 49700
},
{
"epoch": 13.388135228251507,
"grad_norm": 0.3546346127986908,
"learning_rate": 0.0003994519994614245,
"loss": 3.0844,
"step": 49750
},
{
"epoch": 13.401593453919036,
"grad_norm": 0.3661574423313141,
"learning_rate": 0.00039925003366096673,
"loss": 3.0784,
"step": 49800
},
{
"epoch": 13.415051679586563,
"grad_norm": 0.3491849899291992,
"learning_rate": 0.0003990480678605089,
"loss": 3.07,
"step": 49850
},
{
"epoch": 13.428509905254092,
"grad_norm": 0.3728206753730774,
"learning_rate": 0.0003988461020600511,
"loss": 3.0728,
"step": 49900
},
{
"epoch": 13.44196813092162,
"grad_norm": 0.3626510798931122,
"learning_rate": 0.0003986441362595933,
"loss": 3.0888,
"step": 49950
},
{
"epoch": 13.455426356589147,
"grad_norm": 0.3639829158782959,
"learning_rate": 0.0003984421704591355,
"loss": 3.0752,
"step": 50000
},
{
"epoch": 13.455426356589147,
"eval_accuracy": 0.3933113953602578,
"eval_loss": 3.297682523727417,
"eval_runtime": 82.644,
"eval_samples_per_second": 217.947,
"eval_steps_per_second": 13.625,
"step": 50000
},
{
"epoch": 13.468884582256676,
"grad_norm": 0.3690478801727295,
"learning_rate": 0.00039824020465867776,
"loss": 3.0884,
"step": 50050
},
{
"epoch": 13.482342807924203,
"grad_norm": 0.34633222222328186,
"learning_rate": 0.00039803823885821996,
"loss": 3.0855,
"step": 50100
},
{
"epoch": 13.49580103359173,
"grad_norm": 0.34675320982933044,
"learning_rate": 0.00039783627305776216,
"loss": 3.0845,
"step": 50150
},
{
"epoch": 13.50925925925926,
"grad_norm": 0.3690127432346344,
"learning_rate": 0.00039763430725730435,
"loss": 3.0893,
"step": 50200
},
{
"epoch": 13.522717484926787,
"grad_norm": 0.3766573965549469,
"learning_rate": 0.00039743234145684666,
"loss": 3.0884,
"step": 50250
},
{
"epoch": 13.536175710594316,
"grad_norm": 0.3568640649318695,
"learning_rate": 0.00039723037565638885,
"loss": 3.0839,
"step": 50300
},
{
"epoch": 13.549633936261843,
"grad_norm": 0.36566534638404846,
"learning_rate": 0.00039702840985593105,
"loss": 3.0833,
"step": 50350
},
{
"epoch": 13.56309216192937,
"grad_norm": 0.35633590817451477,
"learning_rate": 0.00039682644405547324,
"loss": 3.085,
"step": 50400
},
{
"epoch": 13.5765503875969,
"grad_norm": 0.3928929269313812,
"learning_rate": 0.00039662447825501544,
"loss": 3.0975,
"step": 50450
},
{
"epoch": 13.590008613264427,
"grad_norm": 0.3681849539279938,
"learning_rate": 0.0003964225124545577,
"loss": 3.0919,
"step": 50500
},
{
"epoch": 13.603466838931956,
"grad_norm": 0.3847604990005493,
"learning_rate": 0.0003962205466540999,
"loss": 3.0878,
"step": 50550
},
{
"epoch": 13.616925064599483,
"grad_norm": 0.3474828898906708,
"learning_rate": 0.0003960185808536421,
"loss": 3.0933,
"step": 50600
},
{
"epoch": 13.63038329026701,
"grad_norm": 0.3648919463157654,
"learning_rate": 0.0003958166150531843,
"loss": 3.0886,
"step": 50650
},
{
"epoch": 13.64384151593454,
"grad_norm": 0.3626580238342285,
"learning_rate": 0.0003956146492527265,
"loss": 3.0951,
"step": 50700
},
{
"epoch": 13.657299741602067,
"grad_norm": 0.33922359347343445,
"learning_rate": 0.0003954126834522687,
"loss": 3.0892,
"step": 50750
},
{
"epoch": 13.670757967269594,
"grad_norm": 0.34805557131767273,
"learning_rate": 0.0003952107176518109,
"loss": 3.092,
"step": 50800
},
{
"epoch": 13.684216192937123,
"grad_norm": 0.34412679076194763,
"learning_rate": 0.0003950087518513531,
"loss": 3.0895,
"step": 50850
},
{
"epoch": 13.69767441860465,
"grad_norm": 0.3760313093662262,
"learning_rate": 0.0003948067860508953,
"loss": 3.0894,
"step": 50900
},
{
"epoch": 13.71113264427218,
"grad_norm": 0.3731016218662262,
"learning_rate": 0.0003946048202504375,
"loss": 3.0995,
"step": 50950
},
{
"epoch": 13.724590869939707,
"grad_norm": 0.3803709149360657,
"learning_rate": 0.00039440285444997976,
"loss": 3.1014,
"step": 51000
},
{
"epoch": 13.724590869939707,
"eval_accuracy": 0.39367764391036547,
"eval_loss": 3.289649486541748,
"eval_runtime": 82.7039,
"eval_samples_per_second": 217.789,
"eval_steps_per_second": 13.615,
"step": 51000
},
{
"epoch": 13.738049095607234,
"grad_norm": 0.37446117401123047,
"learning_rate": 0.00039420088864952196,
"loss": 3.0925,
"step": 51050
},
{
"epoch": 13.751507321274763,
"grad_norm": 0.336206316947937,
"learning_rate": 0.00039399892284906415,
"loss": 3.0961,
"step": 51100
},
{
"epoch": 13.76496554694229,
"grad_norm": 0.3748316168785095,
"learning_rate": 0.00039379695704860646,
"loss": 3.1044,
"step": 51150
},
{
"epoch": 13.77842377260982,
"grad_norm": 0.34574997425079346,
"learning_rate": 0.00039359499124814865,
"loss": 3.1094,
"step": 51200
},
{
"epoch": 13.791881998277347,
"grad_norm": 0.36065474152565,
"learning_rate": 0.00039339302544769085,
"loss": 3.1039,
"step": 51250
},
{
"epoch": 13.805340223944874,
"grad_norm": 0.33803504705429077,
"learning_rate": 0.00039319105964723305,
"loss": 3.1111,
"step": 51300
},
{
"epoch": 13.818798449612403,
"grad_norm": 0.34584841132164,
"learning_rate": 0.00039298909384677524,
"loss": 3.1087,
"step": 51350
},
{
"epoch": 13.83225667527993,
"grad_norm": 0.38044601678848267,
"learning_rate": 0.0003927871280463175,
"loss": 3.1027,
"step": 51400
},
{
"epoch": 13.84571490094746,
"grad_norm": 0.3633703291416168,
"learning_rate": 0.0003925851622458597,
"loss": 3.1058,
"step": 51450
},
{
"epoch": 13.859173126614987,
"grad_norm": 0.35668858885765076,
"learning_rate": 0.0003923831964454019,
"loss": 3.1078,
"step": 51500
},
{
"epoch": 13.872631352282514,
"grad_norm": 0.33576884865760803,
"learning_rate": 0.0003921812306449441,
"loss": 3.1077,
"step": 51550
},
{
"epoch": 13.886089577950044,
"grad_norm": 0.3916233777999878,
"learning_rate": 0.0003919792648444863,
"loss": 3.1119,
"step": 51600
},
{
"epoch": 13.89954780361757,
"grad_norm": 0.35365134477615356,
"learning_rate": 0.0003917772990440285,
"loss": 3.1072,
"step": 51650
},
{
"epoch": 13.9130060292851,
"grad_norm": 0.3590000867843628,
"learning_rate": 0.0003915753332435707,
"loss": 3.1036,
"step": 51700
},
{
"epoch": 13.926464254952627,
"grad_norm": 0.3824634552001953,
"learning_rate": 0.0003913733674431129,
"loss": 3.1052,
"step": 51750
},
{
"epoch": 13.939922480620154,
"grad_norm": 0.33821046352386475,
"learning_rate": 0.0003911714016426551,
"loss": 3.0999,
"step": 51800
},
{
"epoch": 13.953380706287684,
"grad_norm": 0.36889487504959106,
"learning_rate": 0.0003909694358421973,
"loss": 3.1013,
"step": 51850
},
{
"epoch": 13.96683893195521,
"grad_norm": 0.39322492480278015,
"learning_rate": 0.00039076747004173956,
"loss": 3.1148,
"step": 51900
},
{
"epoch": 13.98029715762274,
"grad_norm": 0.34424659609794617,
"learning_rate": 0.00039056550424128176,
"loss": 3.1029,
"step": 51950
},
{
"epoch": 13.993755383290267,
"grad_norm": 0.35424885153770447,
"learning_rate": 0.00039036353844082395,
"loss": 3.0974,
"step": 52000
},
{
"epoch": 13.993755383290267,
"eval_accuracy": 0.3943771123664893,
"eval_loss": 3.282249689102173,
"eval_runtime": 82.6696,
"eval_samples_per_second": 217.879,
"eval_steps_per_second": 13.62,
"step": 52000
},
{
"epoch": 14.006998277347115,
"grad_norm": 0.37737324833869934,
"learning_rate": 0.0003901615726403662,
"loss": 3.0603,
"step": 52050
},
{
"epoch": 14.020456503014643,
"grad_norm": 0.34262603521347046,
"learning_rate": 0.00038995960683990845,
"loss": 3.001,
"step": 52100
},
{
"epoch": 14.03391472868217,
"grad_norm": 0.3641819357872009,
"learning_rate": 0.00038975764103945065,
"loss": 3.0202,
"step": 52150
},
{
"epoch": 14.047372954349699,
"grad_norm": 0.3927568793296814,
"learning_rate": 0.00038955567523899285,
"loss": 3.0162,
"step": 52200
},
{
"epoch": 14.060831180017226,
"grad_norm": 0.37641867995262146,
"learning_rate": 0.00038935370943853504,
"loss": 3.0113,
"step": 52250
},
{
"epoch": 14.074289405684755,
"grad_norm": 0.37262749671936035,
"learning_rate": 0.00038915174363807724,
"loss": 3.0111,
"step": 52300
},
{
"epoch": 14.087747631352283,
"grad_norm": 0.35890135169029236,
"learning_rate": 0.0003889497778376195,
"loss": 3.0309,
"step": 52350
},
{
"epoch": 14.10120585701981,
"grad_norm": 0.38959529995918274,
"learning_rate": 0.0003887478120371617,
"loss": 3.0302,
"step": 52400
},
{
"epoch": 14.114664082687339,
"grad_norm": 0.375625342130661,
"learning_rate": 0.0003885458462367039,
"loss": 3.0322,
"step": 52450
},
{
"epoch": 14.128122308354866,
"grad_norm": 0.382854700088501,
"learning_rate": 0.0003883438804362461,
"loss": 3.0432,
"step": 52500
},
{
"epoch": 14.141580534022394,
"grad_norm": 0.37905988097190857,
"learning_rate": 0.0003881419146357883,
"loss": 3.033,
"step": 52550
},
{
"epoch": 14.155038759689923,
"grad_norm": 0.34364014863967896,
"learning_rate": 0.0003879399488353305,
"loss": 3.0415,
"step": 52600
},
{
"epoch": 14.16849698535745,
"grad_norm": 0.3485009968280792,
"learning_rate": 0.0003877379830348727,
"loss": 3.0404,
"step": 52650
},
{
"epoch": 14.18195521102498,
"grad_norm": 0.4055265486240387,
"learning_rate": 0.0003875360172344149,
"loss": 3.0426,
"step": 52700
},
{
"epoch": 14.195413436692506,
"grad_norm": 0.38214313983917236,
"learning_rate": 0.0003873340514339571,
"loss": 3.0396,
"step": 52750
},
{
"epoch": 14.208871662360034,
"grad_norm": 0.4019845128059387,
"learning_rate": 0.0003871320856334993,
"loss": 3.0401,
"step": 52800
},
{
"epoch": 14.222329888027563,
"grad_norm": 0.3736121952533722,
"learning_rate": 0.00038693011983304156,
"loss": 3.0511,
"step": 52850
},
{
"epoch": 14.23578811369509,
"grad_norm": 0.3655436933040619,
"learning_rate": 0.0003867281540325838,
"loss": 3.0594,
"step": 52900
},
{
"epoch": 14.24924633936262,
"grad_norm": 0.37499451637268066,
"learning_rate": 0.000386526188232126,
"loss": 3.0484,
"step": 52950
},
{
"epoch": 14.262704565030146,
"grad_norm": 0.368104487657547,
"learning_rate": 0.00038632422243166825,
"loss": 3.0547,
"step": 53000
},
{
"epoch": 14.262704565030146,
"eval_accuracy": 0.3935944204190031,
"eval_loss": 3.2954745292663574,
"eval_runtime": 82.7996,
"eval_samples_per_second": 217.537,
"eval_steps_per_second": 13.599,
"step": 53000
},
{
"epoch": 14.276162790697674,
"grad_norm": 0.3793664574623108,
"learning_rate": 0.00038612225663121045,
"loss": 3.065,
"step": 53050
},
{
"epoch": 14.289621016365203,
"grad_norm": 0.38424718379974365,
"learning_rate": 0.00038592029083075265,
"loss": 3.0533,
"step": 53100
},
{
"epoch": 14.30307924203273,
"grad_norm": 0.3868301212787628,
"learning_rate": 0.00038571832503029484,
"loss": 3.0526,
"step": 53150
},
{
"epoch": 14.31653746770026,
"grad_norm": 0.38837793469429016,
"learning_rate": 0.00038551635922983704,
"loss": 3.0573,
"step": 53200
},
{
"epoch": 14.329995693367787,
"grad_norm": 0.37440481781959534,
"learning_rate": 0.0003853143934293793,
"loss": 3.0683,
"step": 53250
},
{
"epoch": 14.343453919035314,
"grad_norm": 0.3686114549636841,
"learning_rate": 0.0003851124276289215,
"loss": 3.0665,
"step": 53300
},
{
"epoch": 14.356912144702843,
"grad_norm": 0.3653067350387573,
"learning_rate": 0.0003849104618284637,
"loss": 3.0595,
"step": 53350
},
{
"epoch": 14.37037037037037,
"grad_norm": 0.34597083926200867,
"learning_rate": 0.0003847084960280059,
"loss": 3.0541,
"step": 53400
},
{
"epoch": 14.383828596037898,
"grad_norm": 0.3539198935031891,
"learning_rate": 0.0003845065302275481,
"loss": 3.0708,
"step": 53450
},
{
"epoch": 14.397286821705427,
"grad_norm": 0.36307668685913086,
"learning_rate": 0.0003843045644270903,
"loss": 3.0603,
"step": 53500
},
{
"epoch": 14.410745047372954,
"grad_norm": 0.3748219609260559,
"learning_rate": 0.0003841025986266325,
"loss": 3.061,
"step": 53550
},
{
"epoch": 14.424203273040483,
"grad_norm": 0.37004929780960083,
"learning_rate": 0.0003839006328261747,
"loss": 3.0597,
"step": 53600
},
{
"epoch": 14.43766149870801,
"grad_norm": 0.36938241124153137,
"learning_rate": 0.0003836986670257169,
"loss": 3.0595,
"step": 53650
},
{
"epoch": 14.451119724375538,
"grad_norm": 0.3579943776130676,
"learning_rate": 0.0003834967012252591,
"loss": 3.0673,
"step": 53700
},
{
"epoch": 14.464577950043067,
"grad_norm": 0.3780539035797119,
"learning_rate": 0.00038329473542480136,
"loss": 3.0678,
"step": 53750
},
{
"epoch": 14.478036175710594,
"grad_norm": 0.3607889711856842,
"learning_rate": 0.0003830927696243436,
"loss": 3.0726,
"step": 53800
},
{
"epoch": 14.491494401378123,
"grad_norm": 0.3579919934272766,
"learning_rate": 0.0003828908038238858,
"loss": 3.0605,
"step": 53850
},
{
"epoch": 14.50495262704565,
"grad_norm": 0.3993847668170929,
"learning_rate": 0.000382688838023428,
"loss": 3.0717,
"step": 53900
},
{
"epoch": 14.518410852713178,
"grad_norm": 0.3736123740673065,
"learning_rate": 0.00038248687222297025,
"loss": 3.0722,
"step": 53950
},
{
"epoch": 14.531869078380707,
"grad_norm": 0.3538348078727722,
"learning_rate": 0.00038228490642251245,
"loss": 3.0659,
"step": 54000
},
{
"epoch": 14.531869078380707,
"eval_accuracy": 0.39390895306586216,
"eval_loss": 3.2927801609039307,
"eval_runtime": 82.7392,
"eval_samples_per_second": 217.696,
"eval_steps_per_second": 13.609,
"step": 54000
},
{
"epoch": 14.545327304048234,
"grad_norm": 0.3570793569087982,
"learning_rate": 0.00038208294062205464,
"loss": 3.07,
"step": 54050
},
{
"epoch": 14.558785529715763,
"grad_norm": 0.3689643144607544,
"learning_rate": 0.00038188097482159684,
"loss": 3.0698,
"step": 54100
},
{
"epoch": 14.57224375538329,
"grad_norm": 0.3582077622413635,
"learning_rate": 0.00038167900902113904,
"loss": 3.0764,
"step": 54150
},
{
"epoch": 14.585701981050818,
"grad_norm": 0.3726467788219452,
"learning_rate": 0.0003814770432206813,
"loss": 3.0704,
"step": 54200
},
{
"epoch": 14.599160206718347,
"grad_norm": 0.36264288425445557,
"learning_rate": 0.0003812750774202235,
"loss": 3.0713,
"step": 54250
},
{
"epoch": 14.612618432385874,
"grad_norm": 0.38736483454704285,
"learning_rate": 0.0003810731116197657,
"loss": 3.0799,
"step": 54300
},
{
"epoch": 14.626076658053403,
"grad_norm": 0.3551369309425354,
"learning_rate": 0.0003808711458193079,
"loss": 3.0703,
"step": 54350
},
{
"epoch": 14.63953488372093,
"grad_norm": 0.3681671917438507,
"learning_rate": 0.00038066918001885007,
"loss": 3.0842,
"step": 54400
},
{
"epoch": 14.652993109388458,
"grad_norm": 0.3957917094230652,
"learning_rate": 0.0003804672142183923,
"loss": 3.0878,
"step": 54450
},
{
"epoch": 14.666451335055987,
"grad_norm": 0.3708997964859009,
"learning_rate": 0.0003802652484179345,
"loss": 3.0789,
"step": 54500
},
{
"epoch": 14.679909560723514,
"grad_norm": 0.36182284355163574,
"learning_rate": 0.0003800632826174767,
"loss": 3.0742,
"step": 54550
},
{
"epoch": 14.693367786391041,
"grad_norm": 0.3811604082584381,
"learning_rate": 0.0003798613168170189,
"loss": 3.0918,
"step": 54600
},
{
"epoch": 14.70682601205857,
"grad_norm": 0.34608572721481323,
"learning_rate": 0.0003796593510165611,
"loss": 3.083,
"step": 54650
},
{
"epoch": 14.720284237726098,
"grad_norm": 0.3658161163330078,
"learning_rate": 0.0003794573852161034,
"loss": 3.098,
"step": 54700
},
{
"epoch": 14.733742463393627,
"grad_norm": 0.3467399477958679,
"learning_rate": 0.0003792554194156456,
"loss": 3.0894,
"step": 54750
},
{
"epoch": 14.747200689061154,
"grad_norm": 0.36559516191482544,
"learning_rate": 0.0003790534536151878,
"loss": 3.0876,
"step": 54800
},
{
"epoch": 14.760658914728682,
"grad_norm": 0.3837985694408417,
"learning_rate": 0.00037885148781473005,
"loss": 3.0921,
"step": 54850
},
{
"epoch": 14.77411714039621,
"grad_norm": 0.35983505845069885,
"learning_rate": 0.00037864952201427225,
"loss": 3.0782,
"step": 54900
},
{
"epoch": 14.787575366063738,
"grad_norm": 0.38079988956451416,
"learning_rate": 0.00037844755621381444,
"loss": 3.0856,
"step": 54950
},
{
"epoch": 14.801033591731267,
"grad_norm": 0.33372536301612854,
"learning_rate": 0.00037824559041335664,
"loss": 3.0821,
"step": 55000
},
{
"epoch": 14.801033591731267,
"eval_accuracy": 0.3943551657016653,
"eval_loss": 3.2850284576416016,
"eval_runtime": 82.6268,
"eval_samples_per_second": 217.992,
"eval_steps_per_second": 13.628,
"step": 55000
},
{
"epoch": 14.814491817398794,
"grad_norm": 0.38492551445961,
"learning_rate": 0.00037804362461289884,
"loss": 3.0885,
"step": 55050
},
{
"epoch": 14.827950043066322,
"grad_norm": 0.35808756947517395,
"learning_rate": 0.0003778416588124411,
"loss": 3.0847,
"step": 55100
},
{
"epoch": 14.84140826873385,
"grad_norm": 0.3854195177555084,
"learning_rate": 0.0003776396930119833,
"loss": 3.0853,
"step": 55150
},
{
"epoch": 14.854866494401378,
"grad_norm": 0.3732593059539795,
"learning_rate": 0.0003774377272115255,
"loss": 3.0818,
"step": 55200
},
{
"epoch": 14.868324720068905,
"grad_norm": 0.36063703894615173,
"learning_rate": 0.0003772357614110677,
"loss": 3.0917,
"step": 55250
},
{
"epoch": 14.881782945736434,
"grad_norm": 0.36341142654418945,
"learning_rate": 0.00037703379561060987,
"loss": 3.0835,
"step": 55300
},
{
"epoch": 14.895241171403962,
"grad_norm": 0.3607381582260132,
"learning_rate": 0.0003768318298101521,
"loss": 3.0937,
"step": 55350
},
{
"epoch": 14.90869939707149,
"grad_norm": 0.3653995096683502,
"learning_rate": 0.0003766298640096943,
"loss": 3.0867,
"step": 55400
},
{
"epoch": 14.922157622739018,
"grad_norm": 0.36560046672821045,
"learning_rate": 0.0003764278982092365,
"loss": 3.1018,
"step": 55450
},
{
"epoch": 14.935615848406545,
"grad_norm": 0.355100154876709,
"learning_rate": 0.0003762259324087787,
"loss": 3.087,
"step": 55500
},
{
"epoch": 14.949074074074074,
"grad_norm": 0.35742226243019104,
"learning_rate": 0.0003760239666083209,
"loss": 3.0823,
"step": 55550
},
{
"epoch": 14.962532299741602,
"grad_norm": 0.3546431362628937,
"learning_rate": 0.0003758220008078632,
"loss": 3.0988,
"step": 55600
},
{
"epoch": 14.97599052540913,
"grad_norm": 0.34668391942977905,
"learning_rate": 0.0003756200350074054,
"loss": 3.09,
"step": 55650
},
{
"epoch": 14.989448751076658,
"grad_norm": 0.3574998676776886,
"learning_rate": 0.0003754180692069476,
"loss": 3.0992,
"step": 55700
},
{
"epoch": 15.002691645133506,
"grad_norm": 0.3662489056587219,
"learning_rate": 0.0003752161034064898,
"loss": 3.0738,
"step": 55750
},
{
"epoch": 15.016149870801033,
"grad_norm": 0.3715851306915283,
"learning_rate": 0.00037501413760603205,
"loss": 2.9931,
"step": 55800
},
{
"epoch": 15.02960809646856,
"grad_norm": 0.36159178614616394,
"learning_rate": 0.00037481217180557425,
"loss": 3.0076,
"step": 55850
},
{
"epoch": 15.04306632213609,
"grad_norm": 0.39454397559165955,
"learning_rate": 0.00037461020600511644,
"loss": 2.9987,
"step": 55900
},
{
"epoch": 15.056524547803617,
"grad_norm": 0.3761124312877655,
"learning_rate": 0.00037440824020465864,
"loss": 3.0101,
"step": 55950
},
{
"epoch": 15.069982773471146,
"grad_norm": 0.3900396227836609,
"learning_rate": 0.00037420627440420083,
"loss": 3.0104,
"step": 56000
},
{
"epoch": 15.069982773471146,
"eval_accuracy": 0.39382768521789996,
"eval_loss": 3.29386830329895,
"eval_runtime": 82.8587,
"eval_samples_per_second": 217.382,
"eval_steps_per_second": 13.589,
"step": 56000
},
{
"epoch": 15.083440999138674,
"grad_norm": 0.36524370312690735,
"learning_rate": 0.0003740043086037431,
"loss": 3.0153,
"step": 56050
},
{
"epoch": 15.0968992248062,
"grad_norm": 0.3930032551288605,
"learning_rate": 0.0003738023428032853,
"loss": 3.0205,
"step": 56100
},
{
"epoch": 15.11035745047373,
"grad_norm": 0.3790472447872162,
"learning_rate": 0.0003736003770028275,
"loss": 3.0126,
"step": 56150
},
{
"epoch": 15.123815676141257,
"grad_norm": 0.37529274821281433,
"learning_rate": 0.00037339841120236967,
"loss": 3.0251,
"step": 56200
},
{
"epoch": 15.137273901808786,
"grad_norm": 0.3832322061061859,
"learning_rate": 0.00037319644540191187,
"loss": 3.0261,
"step": 56250
},
{
"epoch": 15.150732127476314,
"grad_norm": 0.389457643032074,
"learning_rate": 0.0003729944796014541,
"loss": 3.0065,
"step": 56300
},
{
"epoch": 15.164190353143841,
"grad_norm": 0.3528468608856201,
"learning_rate": 0.0003727925138009963,
"loss": 3.0271,
"step": 56350
},
{
"epoch": 15.17764857881137,
"grad_norm": 0.3809889554977417,
"learning_rate": 0.0003725905480005385,
"loss": 3.0291,
"step": 56400
},
{
"epoch": 15.191106804478897,
"grad_norm": 0.4030018448829651,
"learning_rate": 0.0003723885822000807,
"loss": 3.0339,
"step": 56450
},
{
"epoch": 15.204565030146426,
"grad_norm": 0.3894319236278534,
"learning_rate": 0.000372186616399623,
"loss": 3.0337,
"step": 56500
},
{
"epoch": 15.218023255813954,
"grad_norm": 0.40263044834136963,
"learning_rate": 0.0003719846505991652,
"loss": 3.0321,
"step": 56550
},
{
"epoch": 15.231481481481481,
"grad_norm": 0.37681061029434204,
"learning_rate": 0.0003717826847987074,
"loss": 3.0431,
"step": 56600
},
{
"epoch": 15.24493970714901,
"grad_norm": 0.3457631766796112,
"learning_rate": 0.0003715807189982496,
"loss": 3.0384,
"step": 56650
},
{
"epoch": 15.258397932816537,
"grad_norm": 0.3707561790943146,
"learning_rate": 0.00037137875319779185,
"loss": 3.0348,
"step": 56700
},
{
"epoch": 15.271856158484065,
"grad_norm": 0.3550037741661072,
"learning_rate": 0.00037117678739733405,
"loss": 3.0465,
"step": 56750
},
{
"epoch": 15.285314384151594,
"grad_norm": 0.37704595923423767,
"learning_rate": 0.00037097482159687624,
"loss": 3.0369,
"step": 56800
},
{
"epoch": 15.298772609819121,
"grad_norm": 0.3636750876903534,
"learning_rate": 0.00037077285579641844,
"loss": 3.0359,
"step": 56850
},
{
"epoch": 15.31223083548665,
"grad_norm": 0.37704896926879883,
"learning_rate": 0.00037057088999596063,
"loss": 3.054,
"step": 56900
},
{
"epoch": 15.325689061154177,
"grad_norm": 0.3902929127216339,
"learning_rate": 0.0003703689241955029,
"loss": 3.0415,
"step": 56950
},
{
"epoch": 15.339147286821705,
"grad_norm": 0.3807653486728668,
"learning_rate": 0.0003701669583950451,
"loss": 3.0444,
"step": 57000
},
{
"epoch": 15.339147286821705,
"eval_accuracy": 0.3942507560734679,
"eval_loss": 3.2938787937164307,
"eval_runtime": 82.9429,
"eval_samples_per_second": 217.161,
"eval_steps_per_second": 13.576,
"step": 57000
},
{
"epoch": 15.352605512489234,
"grad_norm": 0.36937519907951355,
"learning_rate": 0.0003699649925945873,
"loss": 3.0483,
"step": 57050
},
{
"epoch": 15.366063738156761,
"grad_norm": 0.37506166100502014,
"learning_rate": 0.0003697630267941295,
"loss": 3.0422,
"step": 57100
},
{
"epoch": 15.37952196382429,
"grad_norm": 0.36554110050201416,
"learning_rate": 0.00036956106099367167,
"loss": 3.0486,
"step": 57150
},
{
"epoch": 15.392980189491817,
"grad_norm": 0.37356069684028625,
"learning_rate": 0.0003693590951932139,
"loss": 3.039,
"step": 57200
},
{
"epoch": 15.406438415159345,
"grad_norm": 0.3713846504688263,
"learning_rate": 0.0003691571293927561,
"loss": 3.0464,
"step": 57250
},
{
"epoch": 15.419896640826874,
"grad_norm": 0.36668819189071655,
"learning_rate": 0.0003689551635922983,
"loss": 3.0582,
"step": 57300
},
{
"epoch": 15.433354866494401,
"grad_norm": 0.36244866251945496,
"learning_rate": 0.00036875319779184056,
"loss": 3.0594,
"step": 57350
},
{
"epoch": 15.44681309216193,
"grad_norm": 0.3832623362541199,
"learning_rate": 0.0003685512319913828,
"loss": 3.045,
"step": 57400
},
{
"epoch": 15.460271317829458,
"grad_norm": 0.391477108001709,
"learning_rate": 0.000368349266190925,
"loss": 3.0551,
"step": 57450
},
{
"epoch": 15.473729543496985,
"grad_norm": 0.3837999105453491,
"learning_rate": 0.0003681473003904672,
"loss": 3.0555,
"step": 57500
},
{
"epoch": 15.487187769164514,
"grad_norm": 0.37777388095855713,
"learning_rate": 0.0003679453345900094,
"loss": 3.0433,
"step": 57550
},
{
"epoch": 15.500645994832041,
"grad_norm": 0.34823834896087646,
"learning_rate": 0.0003677433687895516,
"loss": 3.0545,
"step": 57600
},
{
"epoch": 15.514104220499568,
"grad_norm": 0.36889156699180603,
"learning_rate": 0.00036754140298909385,
"loss": 3.0613,
"step": 57650
},
{
"epoch": 15.527562446167098,
"grad_norm": 0.37089234590530396,
"learning_rate": 0.00036733943718863604,
"loss": 3.0623,
"step": 57700
},
{
"epoch": 15.541020671834625,
"grad_norm": 0.3649154305458069,
"learning_rate": 0.00036713747138817824,
"loss": 3.0612,
"step": 57750
},
{
"epoch": 15.554478897502154,
"grad_norm": 0.3620425760746002,
"learning_rate": 0.00036693550558772044,
"loss": 3.062,
"step": 57800
},
{
"epoch": 15.567937123169681,
"grad_norm": 0.393711119890213,
"learning_rate": 0.00036673353978726263,
"loss": 3.0615,
"step": 57850
},
{
"epoch": 15.581395348837209,
"grad_norm": 0.3537093997001648,
"learning_rate": 0.0003665315739868049,
"loss": 3.0635,
"step": 57900
},
{
"epoch": 15.594853574504738,
"grad_norm": 0.38703078031539917,
"learning_rate": 0.0003663296081863471,
"loss": 3.0656,
"step": 57950
},
{
"epoch": 15.608311800172265,
"grad_norm": 0.37441444396972656,
"learning_rate": 0.0003661276423858893,
"loss": 3.0586,
"step": 58000
},
{
"epoch": 15.608311800172265,
"eval_accuracy": 0.39471554732157255,
"eval_loss": 3.2850182056427,
"eval_runtime": 82.7416,
"eval_samples_per_second": 217.69,
"eval_steps_per_second": 13.609,
"step": 58000
},
{
"epoch": 15.621770025839794,
"grad_norm": 0.40338388085365295,
"learning_rate": 0.00036592567658543147,
"loss": 3.0704,
"step": 58050
},
{
"epoch": 15.635228251507321,
"grad_norm": 0.3834456503391266,
"learning_rate": 0.00036572371078497367,
"loss": 3.0599,
"step": 58100
},
{
"epoch": 15.648686477174849,
"grad_norm": 0.3643941879272461,
"learning_rate": 0.0003655217449845159,
"loss": 3.0619,
"step": 58150
},
{
"epoch": 15.662144702842378,
"grad_norm": 0.3755214810371399,
"learning_rate": 0.0003653197791840581,
"loss": 3.0668,
"step": 58200
},
{
"epoch": 15.675602928509905,
"grad_norm": 0.36048299074172974,
"learning_rate": 0.00036511781338360036,
"loss": 3.0635,
"step": 58250
},
{
"epoch": 15.689061154177434,
"grad_norm": 0.3875759243965149,
"learning_rate": 0.00036491584758314256,
"loss": 3.0602,
"step": 58300
},
{
"epoch": 15.702519379844961,
"grad_norm": 0.3753551244735718,
"learning_rate": 0.0003647138817826848,
"loss": 3.0702,
"step": 58350
},
{
"epoch": 15.715977605512489,
"grad_norm": 0.3766033947467804,
"learning_rate": 0.000364511915982227,
"loss": 3.0657,
"step": 58400
},
{
"epoch": 15.729435831180018,
"grad_norm": 0.38142356276512146,
"learning_rate": 0.0003643099501817692,
"loss": 3.057,
"step": 58450
},
{
"epoch": 15.742894056847545,
"grad_norm": 0.37450477480888367,
"learning_rate": 0.0003641079843813114,
"loss": 3.0701,
"step": 58500
},
{
"epoch": 15.756352282515074,
"grad_norm": 0.3933105170726776,
"learning_rate": 0.0003639060185808536,
"loss": 3.0829,
"step": 58550
},
{
"epoch": 15.769810508182601,
"grad_norm": 0.3845747709274292,
"learning_rate": 0.00036370405278039584,
"loss": 3.0614,
"step": 58600
},
{
"epoch": 15.783268733850129,
"grad_norm": 0.3942335247993469,
"learning_rate": 0.00036350208697993804,
"loss": 3.0653,
"step": 58650
},
{
"epoch": 15.796726959517658,
"grad_norm": 0.3737879693508148,
"learning_rate": 0.00036330012117948024,
"loss": 3.0657,
"step": 58700
},
{
"epoch": 15.810185185185185,
"grad_norm": 0.36122778058052063,
"learning_rate": 0.00036309815537902243,
"loss": 3.0849,
"step": 58750
},
{
"epoch": 15.823643410852712,
"grad_norm": 0.3577321767807007,
"learning_rate": 0.0003628961895785647,
"loss": 3.0734,
"step": 58800
},
{
"epoch": 15.837101636520241,
"grad_norm": 0.39552953839302063,
"learning_rate": 0.0003626942237781069,
"loss": 3.0707,
"step": 58850
},
{
"epoch": 15.850559862187769,
"grad_norm": 0.37515097856521606,
"learning_rate": 0.0003624922579776491,
"loss": 3.0798,
"step": 58900
},
{
"epoch": 15.864018087855298,
"grad_norm": 0.3599167764186859,
"learning_rate": 0.00036229029217719127,
"loss": 3.0642,
"step": 58950
},
{
"epoch": 15.877476313522825,
"grad_norm": 0.38238927721977234,
"learning_rate": 0.00036208832637673347,
"loss": 3.0778,
"step": 59000
},
{
"epoch": 15.877476313522825,
"eval_accuracy": 0.39493870796290187,
"eval_loss": 3.2786478996276855,
"eval_runtime": 82.7649,
"eval_samples_per_second": 217.628,
"eval_steps_per_second": 13.605,
"step": 59000
},
{
"epoch": 15.890934539190352,
"grad_norm": 0.3646547198295593,
"learning_rate": 0.0003618863605762757,
"loss": 3.0778,
"step": 59050
},
{
"epoch": 15.904392764857882,
"grad_norm": 0.4022713601589203,
"learning_rate": 0.0003616843947758179,
"loss": 3.0706,
"step": 59100
},
{
"epoch": 15.917850990525409,
"grad_norm": 0.3673117756843567,
"learning_rate": 0.00036148242897536016,
"loss": 3.0955,
"step": 59150
},
{
"epoch": 15.931309216192938,
"grad_norm": 0.3503372073173523,
"learning_rate": 0.00036128046317490236,
"loss": 3.0771,
"step": 59200
},
{
"epoch": 15.944767441860465,
"grad_norm": 0.3684360384941101,
"learning_rate": 0.0003610784973744446,
"loss": 3.0836,
"step": 59250
},
{
"epoch": 15.958225667527993,
"grad_norm": 0.3571036458015442,
"learning_rate": 0.0003608765315739868,
"loss": 3.0791,
"step": 59300
},
{
"epoch": 15.971683893195522,
"grad_norm": 0.36907097697257996,
"learning_rate": 0.000360674565773529,
"loss": 3.0725,
"step": 59350
},
{
"epoch": 15.985142118863049,
"grad_norm": 0.3785278797149658,
"learning_rate": 0.0003604725999730712,
"loss": 3.0668,
"step": 59400
},
{
"epoch": 15.998600344530576,
"grad_norm": 0.36876171827316284,
"learning_rate": 0.0003602706341726134,
"loss": 3.0888,
"step": 59450
},
{
"epoch": 16.011843238587424,
"grad_norm": 0.37389466166496277,
"learning_rate": 0.00036006866837215564,
"loss": 3.0042,
"step": 59500
},
{
"epoch": 16.02530146425495,
"grad_norm": 0.3862817585468292,
"learning_rate": 0.00035986670257169784,
"loss": 2.9862,
"step": 59550
},
{
"epoch": 16.03875968992248,
"grad_norm": 0.367305189371109,
"learning_rate": 0.00035966473677124004,
"loss": 2.9901,
"step": 59600
},
{
"epoch": 16.05221791559001,
"grad_norm": 0.3664626181125641,
"learning_rate": 0.00035946277097078223,
"loss": 3.0109,
"step": 59650
},
{
"epoch": 16.065676141257537,
"grad_norm": 0.37476664781570435,
"learning_rate": 0.00035926080517032443,
"loss": 3.0011,
"step": 59700
},
{
"epoch": 16.079134366925064,
"grad_norm": 0.4099160432815552,
"learning_rate": 0.0003590588393698667,
"loss": 3.007,
"step": 59750
},
{
"epoch": 16.09259259259259,
"grad_norm": 0.37961021065711975,
"learning_rate": 0.0003588568735694089,
"loss": 2.9941,
"step": 59800
},
{
"epoch": 16.10605081826012,
"grad_norm": 0.41159889101982117,
"learning_rate": 0.00035865490776895107,
"loss": 2.9912,
"step": 59850
},
{
"epoch": 16.11950904392765,
"grad_norm": 0.3856586515903473,
"learning_rate": 0.00035845294196849327,
"loss": 3.0091,
"step": 59900
},
{
"epoch": 16.132967269595177,
"grad_norm": 0.3681037425994873,
"learning_rate": 0.00035825097616803546,
"loss": 3.0059,
"step": 59950
},
{
"epoch": 16.146425495262704,
"grad_norm": 0.37587231397628784,
"learning_rate": 0.0003580490103675777,
"loss": 3.0106,
"step": 60000
},
{
"epoch": 16.146425495262704,
"eval_accuracy": 0.39467035022965774,
"eval_loss": 3.290004014968872,
"eval_runtime": 82.9197,
"eval_samples_per_second": 217.222,
"eval_steps_per_second": 13.579,
"step": 60000
},
{
"epoch": 16.15988372093023,
"grad_norm": 0.3772297501564026,
"learning_rate": 0.00035784704456711996,
"loss": 3.0115,
"step": 60050
},
{
"epoch": 16.17334194659776,
"grad_norm": 0.371023565530777,
"learning_rate": 0.00035764507876666216,
"loss": 3.0112,
"step": 60100
},
{
"epoch": 16.18680017226529,
"grad_norm": 0.37951380014419556,
"learning_rate": 0.00035744311296620436,
"loss": 3.0172,
"step": 60150
},
{
"epoch": 16.200258397932817,
"grad_norm": 0.36302000284194946,
"learning_rate": 0.0003572411471657466,
"loss": 3.0079,
"step": 60200
},
{
"epoch": 16.213716623600344,
"grad_norm": 0.48842325806617737,
"learning_rate": 0.0003570391813652888,
"loss": 3.0246,
"step": 60250
},
{
"epoch": 16.227174849267872,
"grad_norm": 0.396522581577301,
"learning_rate": 0.000356837215564831,
"loss": 3.0235,
"step": 60300
},
{
"epoch": 16.2406330749354,
"grad_norm": 0.4003185033798218,
"learning_rate": 0.0003566352497643732,
"loss": 3.0324,
"step": 60350
},
{
"epoch": 16.25409130060293,
"grad_norm": 0.39070069789886475,
"learning_rate": 0.0003564332839639154,
"loss": 3.0308,
"step": 60400
},
{
"epoch": 16.267549526270457,
"grad_norm": 0.38743722438812256,
"learning_rate": 0.00035623131816345764,
"loss": 3.0221,
"step": 60450
},
{
"epoch": 16.281007751937985,
"grad_norm": 0.3731786012649536,
"learning_rate": 0.00035602935236299984,
"loss": 3.0211,
"step": 60500
},
{
"epoch": 16.294465977605512,
"grad_norm": 0.3834023177623749,
"learning_rate": 0.00035582738656254203,
"loss": 3.0271,
"step": 60550
},
{
"epoch": 16.30792420327304,
"grad_norm": 0.399309903383255,
"learning_rate": 0.00035562542076208423,
"loss": 3.0244,
"step": 60600
},
{
"epoch": 16.32138242894057,
"grad_norm": 0.3901798129081726,
"learning_rate": 0.0003554234549616264,
"loss": 3.0333,
"step": 60650
},
{
"epoch": 16.334840654608097,
"grad_norm": 0.3769051730632782,
"learning_rate": 0.0003552214891611687,
"loss": 3.0227,
"step": 60700
},
{
"epoch": 16.348298880275625,
"grad_norm": 0.38951775431632996,
"learning_rate": 0.00035501952336071087,
"loss": 3.0226,
"step": 60750
},
{
"epoch": 16.361757105943152,
"grad_norm": 0.38057705760002136,
"learning_rate": 0.00035481755756025307,
"loss": 3.0359,
"step": 60800
},
{
"epoch": 16.37521533161068,
"grad_norm": 0.36813488602638245,
"learning_rate": 0.00035461559175979526,
"loss": 3.046,
"step": 60850
},
{
"epoch": 16.38867355727821,
"grad_norm": 0.3978293240070343,
"learning_rate": 0.0003544136259593375,
"loss": 3.0291,
"step": 60900
},
{
"epoch": 16.402131782945737,
"grad_norm": 0.37667974829673767,
"learning_rate": 0.00035421166015887977,
"loss": 3.0379,
"step": 60950
},
{
"epoch": 16.415590008613265,
"grad_norm": 0.4081231653690338,
"learning_rate": 0.00035400969435842196,
"loss": 3.0364,
"step": 61000
},
{
"epoch": 16.415590008613265,
"eval_accuracy": 0.394879060839197,
"eval_loss": 3.290673017501831,
"eval_runtime": 82.605,
"eval_samples_per_second": 218.05,
"eval_steps_per_second": 13.631,
"step": 61000
},
{
"epoch": 16.429048234280792,
"grad_norm": 0.42379987239837646,
"learning_rate": 0.00035380772855796416,
"loss": 3.0365,
"step": 61050
},
{
"epoch": 16.44250645994832,
"grad_norm": 0.38925445079803467,
"learning_rate": 0.0003536057627575064,
"loss": 3.0468,
"step": 61100
},
{
"epoch": 16.45596468561585,
"grad_norm": 0.3812239170074463,
"learning_rate": 0.0003534037969570486,
"loss": 3.0406,
"step": 61150
},
{
"epoch": 16.469422911283377,
"grad_norm": 0.3832698464393616,
"learning_rate": 0.0003532018311565908,
"loss": 3.039,
"step": 61200
},
{
"epoch": 16.482881136950905,
"grad_norm": 0.3748410940170288,
"learning_rate": 0.000352999865356133,
"loss": 3.0382,
"step": 61250
},
{
"epoch": 16.496339362618432,
"grad_norm": 0.3736821115016937,
"learning_rate": 0.0003527978995556752,
"loss": 3.0498,
"step": 61300
},
{
"epoch": 16.50979758828596,
"grad_norm": 0.3920403718948364,
"learning_rate": 0.00035259593375521744,
"loss": 3.0368,
"step": 61350
},
{
"epoch": 16.52325581395349,
"grad_norm": 0.38232672214508057,
"learning_rate": 0.00035239396795475964,
"loss": 3.0489,
"step": 61400
},
{
"epoch": 16.536714039621017,
"grad_norm": 0.4139941930770874,
"learning_rate": 0.00035219200215430183,
"loss": 3.0483,
"step": 61450
},
{
"epoch": 16.550172265288545,
"grad_norm": 0.4045242667198181,
"learning_rate": 0.00035199003635384403,
"loss": 3.0601,
"step": 61500
},
{
"epoch": 16.563630490956072,
"grad_norm": 0.37240657210350037,
"learning_rate": 0.0003517880705533862,
"loss": 3.0416,
"step": 61550
},
{
"epoch": 16.5770887166236,
"grad_norm": 0.38306596875190735,
"learning_rate": 0.0003515861047529285,
"loss": 3.0553,
"step": 61600
},
{
"epoch": 16.590546942291127,
"grad_norm": 0.3954761326313019,
"learning_rate": 0.0003513841389524707,
"loss": 3.0507,
"step": 61650
},
{
"epoch": 16.604005167958658,
"grad_norm": 0.36340826749801636,
"learning_rate": 0.00035118217315201287,
"loss": 3.0484,
"step": 61700
},
{
"epoch": 16.617463393626185,
"grad_norm": 0.38736069202423096,
"learning_rate": 0.00035098020735155507,
"loss": 3.0563,
"step": 61750
},
{
"epoch": 16.630921619293712,
"grad_norm": 0.37580689787864685,
"learning_rate": 0.00035077824155109737,
"loss": 3.0497,
"step": 61800
},
{
"epoch": 16.64437984496124,
"grad_norm": 0.4007303714752197,
"learning_rate": 0.00035057627575063957,
"loss": 3.0435,
"step": 61850
},
{
"epoch": 16.657838070628767,
"grad_norm": 0.390712708234787,
"learning_rate": 0.00035037430995018176,
"loss": 3.0599,
"step": 61900
},
{
"epoch": 16.671296296296298,
"grad_norm": 0.392600953578949,
"learning_rate": 0.00035017234414972396,
"loss": 3.0557,
"step": 61950
},
{
"epoch": 16.684754521963825,
"grad_norm": 0.36124250292778015,
"learning_rate": 0.00034997037834926615,
"loss": 3.0432,
"step": 62000
},
{
"epoch": 16.684754521963825,
"eval_accuracy": 0.39560047596014486,
"eval_loss": 3.2811717987060547,
"eval_runtime": 82.9299,
"eval_samples_per_second": 217.196,
"eval_steps_per_second": 13.578,
"step": 62000
},
{
"epoch": 16.698212747631352,
"grad_norm": 0.3844659626483917,
"learning_rate": 0.0003497684125488084,
"loss": 3.0551,
"step": 62050
},
{
"epoch": 16.71167097329888,
"grad_norm": 0.37635818123817444,
"learning_rate": 0.0003495664467483506,
"loss": 3.0498,
"step": 62100
},
{
"epoch": 16.725129198966407,
"grad_norm": 0.37683260440826416,
"learning_rate": 0.0003493644809478928,
"loss": 3.0627,
"step": 62150
},
{
"epoch": 16.738587424633938,
"grad_norm": 0.3610328137874603,
"learning_rate": 0.000349162515147435,
"loss": 3.0578,
"step": 62200
},
{
"epoch": 16.752045650301465,
"grad_norm": 0.358863890171051,
"learning_rate": 0.0003489605493469772,
"loss": 3.0428,
"step": 62250
},
{
"epoch": 16.765503875968992,
"grad_norm": 0.37700700759887695,
"learning_rate": 0.00034875858354651944,
"loss": 3.0497,
"step": 62300
},
{
"epoch": 16.77896210163652,
"grad_norm": 0.4090152084827423,
"learning_rate": 0.00034855661774606164,
"loss": 3.0566,
"step": 62350
},
{
"epoch": 16.792420327304047,
"grad_norm": 0.38179752230644226,
"learning_rate": 0.00034835465194560383,
"loss": 3.0592,
"step": 62400
},
{
"epoch": 16.805878552971578,
"grad_norm": 0.3984977602958679,
"learning_rate": 0.00034815268614514603,
"loss": 3.0518,
"step": 62450
},
{
"epoch": 16.819336778639105,
"grad_norm": 0.3879537880420685,
"learning_rate": 0.0003479507203446882,
"loss": 3.0729,
"step": 62500
},
{
"epoch": 16.832795004306632,
"grad_norm": 0.38321301341056824,
"learning_rate": 0.0003477487545442305,
"loss": 3.0636,
"step": 62550
},
{
"epoch": 16.84625322997416,
"grad_norm": 0.42007800936698914,
"learning_rate": 0.00034754678874377267,
"loss": 3.0597,
"step": 62600
},
{
"epoch": 16.859711455641687,
"grad_norm": 0.4137289822101593,
"learning_rate": 0.00034734482294331487,
"loss": 3.0598,
"step": 62650
},
{
"epoch": 16.873169681309218,
"grad_norm": 0.37830615043640137,
"learning_rate": 0.00034714285714285717,
"loss": 3.0595,
"step": 62700
},
{
"epoch": 16.886627906976745,
"grad_norm": 0.3817092776298523,
"learning_rate": 0.00034694089134239937,
"loss": 3.0686,
"step": 62750
},
{
"epoch": 16.900086132644272,
"grad_norm": 0.38101911544799805,
"learning_rate": 0.00034673892554194156,
"loss": 3.0662,
"step": 62800
},
{
"epoch": 16.9135443583118,
"grad_norm": 0.39121511578559875,
"learning_rate": 0.00034653695974148376,
"loss": 3.066,
"step": 62850
},
{
"epoch": 16.927002583979327,
"grad_norm": 0.39795464277267456,
"learning_rate": 0.00034633499394102595,
"loss": 3.0658,
"step": 62900
},
{
"epoch": 16.940460809646858,
"grad_norm": 0.4153473675251007,
"learning_rate": 0.0003461330281405682,
"loss": 3.0704,
"step": 62950
},
{
"epoch": 16.953919035314385,
"grad_norm": 0.38049232959747314,
"learning_rate": 0.0003459310623401104,
"loss": 3.0664,
"step": 63000
},
{
"epoch": 16.953919035314385,
"eval_accuracy": 0.3956848945669184,
"eval_loss": 3.2770028114318848,
"eval_runtime": 82.9065,
"eval_samples_per_second": 217.257,
"eval_steps_per_second": 13.582,
"step": 63000
},
{
"epoch": 16.967377260981912,
"grad_norm": 0.39290884137153625,
"learning_rate": 0.0003457290965396526,
"loss": 3.0583,
"step": 63050
},
{
"epoch": 16.98083548664944,
"grad_norm": 0.3828415274620056,
"learning_rate": 0.0003455271307391948,
"loss": 3.0745,
"step": 63100
},
{
"epoch": 16.994293712316967,
"grad_norm": 0.3886496126651764,
"learning_rate": 0.000345325164938737,
"loss": 3.0718,
"step": 63150
},
{
"epoch": 17.007536606373815,
"grad_norm": 0.37772977352142334,
"learning_rate": 0.00034512319913827924,
"loss": 3.0066,
"step": 63200
},
{
"epoch": 17.020994832041342,
"grad_norm": 0.39448246359825134,
"learning_rate": 0.00034492123333782144,
"loss": 2.9732,
"step": 63250
},
{
"epoch": 17.034453057708873,
"grad_norm": 0.36992695927619934,
"learning_rate": 0.00034471926753736363,
"loss": 2.9737,
"step": 63300
},
{
"epoch": 17.0479112833764,
"grad_norm": 0.37510907649993896,
"learning_rate": 0.00034451730173690583,
"loss": 2.9838,
"step": 63350
},
{
"epoch": 17.061369509043928,
"grad_norm": 0.40669476985931396,
"learning_rate": 0.000344315335936448,
"loss": 2.994,
"step": 63400
},
{
"epoch": 17.074827734711455,
"grad_norm": 0.3946589231491089,
"learning_rate": 0.0003441133701359903,
"loss": 2.9841,
"step": 63450
},
{
"epoch": 17.088285960378983,
"grad_norm": 0.38869068026542664,
"learning_rate": 0.00034391140433553247,
"loss": 2.9811,
"step": 63500
},
{
"epoch": 17.101744186046513,
"grad_norm": 0.3955644965171814,
"learning_rate": 0.00034370943853507467,
"loss": 2.9954,
"step": 63550
},
{
"epoch": 17.11520241171404,
"grad_norm": 0.37245795130729675,
"learning_rate": 0.0003435074727346169,
"loss": 2.994,
"step": 63600
},
{
"epoch": 17.128660637381568,
"grad_norm": 0.3830540180206299,
"learning_rate": 0.00034330550693415917,
"loss": 2.9917,
"step": 63650
},
{
"epoch": 17.142118863049095,
"grad_norm": 0.3846103847026825,
"learning_rate": 0.00034310354113370136,
"loss": 2.9921,
"step": 63700
},
{
"epoch": 17.155577088716623,
"grad_norm": 0.4047485291957855,
"learning_rate": 0.00034290157533324356,
"loss": 2.9926,
"step": 63750
},
{
"epoch": 17.16903531438415,
"grad_norm": 0.35702529549598694,
"learning_rate": 0.00034269960953278576,
"loss": 3.0001,
"step": 63800
},
{
"epoch": 17.18249354005168,
"grad_norm": 0.37685129046440125,
"learning_rate": 0.00034249764373232795,
"loss": 2.9892,
"step": 63850
},
{
"epoch": 17.195951765719208,
"grad_norm": 0.38626858592033386,
"learning_rate": 0.0003422956779318702,
"loss": 3.0065,
"step": 63900
},
{
"epoch": 17.209409991386735,
"grad_norm": 0.40105676651000977,
"learning_rate": 0.0003420937121314124,
"loss": 3.0024,
"step": 63950
},
{
"epoch": 17.222868217054263,
"grad_norm": 0.38862723112106323,
"learning_rate": 0.0003418917463309546,
"loss": 3.0044,
"step": 64000
},
{
"epoch": 17.222868217054263,
"eval_accuracy": 0.39483679721238246,
"eval_loss": 3.292349100112915,
"eval_runtime": 82.6383,
"eval_samples_per_second": 217.962,
"eval_steps_per_second": 13.626,
"step": 64000
},
{
"epoch": 17.23632644272179,
"grad_norm": 0.39989951252937317,
"learning_rate": 0.0003416897805304968,
"loss": 3.0191,
"step": 64050
},
{
"epoch": 17.24978466838932,
"grad_norm": 0.4279472231864929,
"learning_rate": 0.000341487814730039,
"loss": 3.0136,
"step": 64100
},
{
"epoch": 17.263242894056848,
"grad_norm": 0.391873300075531,
"learning_rate": 0.00034128584892958124,
"loss": 3.0149,
"step": 64150
},
{
"epoch": 17.276701119724375,
"grad_norm": 0.39389368891716003,
"learning_rate": 0.00034108388312912343,
"loss": 3.0151,
"step": 64200
},
{
"epoch": 17.290159345391903,
"grad_norm": 0.4009759724140167,
"learning_rate": 0.00034088191732866563,
"loss": 3.0097,
"step": 64250
},
{
"epoch": 17.30361757105943,
"grad_norm": 0.36697378754615784,
"learning_rate": 0.0003406799515282078,
"loss": 3.0112,
"step": 64300
},
{
"epoch": 17.31707579672696,
"grad_norm": 0.36988386511802673,
"learning_rate": 0.00034047798572775,
"loss": 3.0044,
"step": 64350
},
{
"epoch": 17.330534022394488,
"grad_norm": 0.36829185485839844,
"learning_rate": 0.00034027601992729227,
"loss": 3.0189,
"step": 64400
},
{
"epoch": 17.343992248062015,
"grad_norm": 0.37610191106796265,
"learning_rate": 0.00034007405412683447,
"loss": 3.0195,
"step": 64450
},
{
"epoch": 17.357450473729543,
"grad_norm": 0.36737823486328125,
"learning_rate": 0.0003398720883263767,
"loss": 3.0231,
"step": 64500
},
{
"epoch": 17.37090869939707,
"grad_norm": 0.35848408937454224,
"learning_rate": 0.00033967012252591897,
"loss": 3.0301,
"step": 64550
},
{
"epoch": 17.3843669250646,
"grad_norm": 0.41312986612319946,
"learning_rate": 0.00033946815672546116,
"loss": 3.0221,
"step": 64600
},
{
"epoch": 17.397825150732128,
"grad_norm": 0.40311887860298157,
"learning_rate": 0.00033926619092500336,
"loss": 3.0239,
"step": 64650
},
{
"epoch": 17.411283376399656,
"grad_norm": 0.3915446400642395,
"learning_rate": 0.00033906422512454556,
"loss": 3.0192,
"step": 64700
},
{
"epoch": 17.424741602067183,
"grad_norm": 0.38295456767082214,
"learning_rate": 0.00033886225932408775,
"loss": 3.0306,
"step": 64750
},
{
"epoch": 17.43819982773471,
"grad_norm": 0.3868366777896881,
"learning_rate": 0.00033866029352363,
"loss": 3.0291,
"step": 64800
},
{
"epoch": 17.45165805340224,
"grad_norm": 0.38612237572669983,
"learning_rate": 0.0003384583277231722,
"loss": 3.0268,
"step": 64850
},
{
"epoch": 17.46511627906977,
"grad_norm": 0.3738773465156555,
"learning_rate": 0.0003382563619227144,
"loss": 3.0367,
"step": 64900
},
{
"epoch": 17.478574504737296,
"grad_norm": 0.3759799003601074,
"learning_rate": 0.0003380543961222566,
"loss": 3.0289,
"step": 64950
},
{
"epoch": 17.492032730404823,
"grad_norm": 0.40272256731987,
"learning_rate": 0.0003378524303217988,
"loss": 3.0323,
"step": 65000
},
{
"epoch": 17.492032730404823,
"eval_accuracy": 0.3955145362973934,
"eval_loss": 3.283403158187866,
"eval_runtime": 82.5842,
"eval_samples_per_second": 218.105,
"eval_steps_per_second": 13.635,
"step": 65000
},
{
"epoch": 17.50549095607235,
"grad_norm": 0.4135473668575287,
"learning_rate": 0.00033765046452134104,
"loss": 3.0318,
"step": 65050
},
{
"epoch": 17.51894918173988,
"grad_norm": 0.3919709622859955,
"learning_rate": 0.00033744849872088323,
"loss": 3.0312,
"step": 65100
},
{
"epoch": 17.53240740740741,
"grad_norm": 0.5064506530761719,
"learning_rate": 0.00033724653292042543,
"loss": 3.0347,
"step": 65150
},
{
"epoch": 17.545865633074936,
"grad_norm": 0.393545001745224,
"learning_rate": 0.0003370445671199676,
"loss": 3.0386,
"step": 65200
},
{
"epoch": 17.559323858742463,
"grad_norm": 0.4120955765247345,
"learning_rate": 0.0003368426013195098,
"loss": 3.0329,
"step": 65250
},
{
"epoch": 17.57278208440999,
"grad_norm": 0.3882588744163513,
"learning_rate": 0.00033664063551905207,
"loss": 3.0234,
"step": 65300
},
{
"epoch": 17.58624031007752,
"grad_norm": 0.3937736451625824,
"learning_rate": 0.00033643866971859427,
"loss": 3.044,
"step": 65350
},
{
"epoch": 17.59969853574505,
"grad_norm": 0.3903486728668213,
"learning_rate": 0.0003362367039181365,
"loss": 3.0347,
"step": 65400
},
{
"epoch": 17.613156761412576,
"grad_norm": 0.37988489866256714,
"learning_rate": 0.0003360347381176787,
"loss": 3.0352,
"step": 65450
},
{
"epoch": 17.626614987080103,
"grad_norm": 0.40244200825691223,
"learning_rate": 0.00033583277231722097,
"loss": 3.0317,
"step": 65500
},
{
"epoch": 17.64007321274763,
"grad_norm": 0.3838312327861786,
"learning_rate": 0.00033563080651676316,
"loss": 3.0472,
"step": 65550
},
{
"epoch": 17.653531438415158,
"grad_norm": 0.4004960060119629,
"learning_rate": 0.00033542884071630536,
"loss": 3.0536,
"step": 65600
},
{
"epoch": 17.66698966408269,
"grad_norm": 0.37788906693458557,
"learning_rate": 0.00033522687491584755,
"loss": 3.0373,
"step": 65650
},
{
"epoch": 17.680447889750216,
"grad_norm": 0.3725600838661194,
"learning_rate": 0.00033502490911538975,
"loss": 3.0339,
"step": 65700
},
{
"epoch": 17.693906115417743,
"grad_norm": 0.3981366753578186,
"learning_rate": 0.000334822943314932,
"loss": 3.0433,
"step": 65750
},
{
"epoch": 17.70736434108527,
"grad_norm": 0.371738463640213,
"learning_rate": 0.0003346209775144742,
"loss": 3.0439,
"step": 65800
},
{
"epoch": 17.720822566752798,
"grad_norm": 0.37119337916374207,
"learning_rate": 0.0003344190117140164,
"loss": 3.0477,
"step": 65850
},
{
"epoch": 17.73428079242033,
"grad_norm": 0.3815227150917053,
"learning_rate": 0.0003342170459135586,
"loss": 3.046,
"step": 65900
},
{
"epoch": 17.747739018087856,
"grad_norm": 0.3605468273162842,
"learning_rate": 0.0003340150801131008,
"loss": 3.0446,
"step": 65950
},
{
"epoch": 17.761197243755383,
"grad_norm": 0.371143102645874,
"learning_rate": 0.00033381311431264303,
"loss": 3.0495,
"step": 66000
},
{
"epoch": 17.761197243755383,
"eval_accuracy": 0.39534874119580204,
"eval_loss": 3.2809362411499023,
"eval_runtime": 82.6691,
"eval_samples_per_second": 217.881,
"eval_steps_per_second": 13.621,
"step": 66000
},
{
"epoch": 17.77465546942291,
"grad_norm": 0.36951732635498047,
"learning_rate": 0.00033361114851218523,
"loss": 3.0535,
"step": 66050
},
{
"epoch": 17.788113695090438,
"grad_norm": 0.4117947518825531,
"learning_rate": 0.0003334091827117274,
"loss": 3.0582,
"step": 66100
},
{
"epoch": 17.80157192075797,
"grad_norm": 0.382714182138443,
"learning_rate": 0.0003332072169112696,
"loss": 3.0409,
"step": 66150
},
{
"epoch": 17.815030146425496,
"grad_norm": 0.3954039514064789,
"learning_rate": 0.0003330052511108118,
"loss": 3.0502,
"step": 66200
},
{
"epoch": 17.828488372093023,
"grad_norm": 0.38930243253707886,
"learning_rate": 0.0003328032853103541,
"loss": 3.0567,
"step": 66250
},
{
"epoch": 17.84194659776055,
"grad_norm": 0.38286498188972473,
"learning_rate": 0.0003326013195098963,
"loss": 3.0503,
"step": 66300
},
{
"epoch": 17.855404823428078,
"grad_norm": 0.3883371651172638,
"learning_rate": 0.0003323993537094385,
"loss": 3.0459,
"step": 66350
},
{
"epoch": 17.86886304909561,
"grad_norm": 0.4157749116420746,
"learning_rate": 0.00033219738790898077,
"loss": 3.0479,
"step": 66400
},
{
"epoch": 17.882321274763136,
"grad_norm": 0.38203850388526917,
"learning_rate": 0.00033199542210852296,
"loss": 3.0582,
"step": 66450
},
{
"epoch": 17.895779500430663,
"grad_norm": 0.4011441171169281,
"learning_rate": 0.00033179345630806516,
"loss": 3.0555,
"step": 66500
},
{
"epoch": 17.90923772609819,
"grad_norm": 0.4033907949924469,
"learning_rate": 0.00033159149050760735,
"loss": 3.0464,
"step": 66550
},
{
"epoch": 17.922695951765718,
"grad_norm": 0.3633027672767639,
"learning_rate": 0.00033138952470714955,
"loss": 3.0553,
"step": 66600
},
{
"epoch": 17.93615417743325,
"grad_norm": 0.4008675813674927,
"learning_rate": 0.0003311875589066918,
"loss": 3.0529,
"step": 66650
},
{
"epoch": 17.949612403100776,
"grad_norm": 0.3831356167793274,
"learning_rate": 0.000330985593106234,
"loss": 3.061,
"step": 66700
},
{
"epoch": 17.963070628768303,
"grad_norm": 0.43848901987075806,
"learning_rate": 0.0003307836273057762,
"loss": 3.056,
"step": 66750
},
{
"epoch": 17.97652885443583,
"grad_norm": 0.41613638401031494,
"learning_rate": 0.0003305816615053184,
"loss": 3.0605,
"step": 66800
},
{
"epoch": 17.989987080103358,
"grad_norm": 0.4012289047241211,
"learning_rate": 0.0003303796957048606,
"loss": 3.0506,
"step": 66850
},
{
"epoch": 18.003229974160206,
"grad_norm": 0.4024728238582611,
"learning_rate": 0.00033017772990440284,
"loss": 3.0275,
"step": 66900
},
{
"epoch": 18.016688199827733,
"grad_norm": 0.40273720026016235,
"learning_rate": 0.00032997576410394503,
"loss": 2.9475,
"step": 66950
},
{
"epoch": 18.030146425495264,
"grad_norm": 0.40914386510849,
"learning_rate": 0.00032977379830348723,
"loss": 2.9769,
"step": 67000
},
{
"epoch": 18.030146425495264,
"eval_accuracy": 0.39520706569614605,
"eval_loss": 3.2891383171081543,
"eval_runtime": 82.5906,
"eval_samples_per_second": 218.088,
"eval_steps_per_second": 13.634,
"step": 67000
},
{
"epoch": 18.04360465116279,
"grad_norm": 0.39871081709861755,
"learning_rate": 0.0003295718325030294,
"loss": 2.9628,
"step": 67050
},
{
"epoch": 18.05706287683032,
"grad_norm": 0.4168705344200134,
"learning_rate": 0.0003293698667025716,
"loss": 2.966,
"step": 67100
},
{
"epoch": 18.070521102497846,
"grad_norm": 0.3725696802139282,
"learning_rate": 0.0003291679009021139,
"loss": 2.9785,
"step": 67150
},
{
"epoch": 18.083979328165373,
"grad_norm": 0.4275813698768616,
"learning_rate": 0.0003289659351016561,
"loss": 2.9727,
"step": 67200
},
{
"epoch": 18.097437553832904,
"grad_norm": 0.40893566608428955,
"learning_rate": 0.0003287639693011983,
"loss": 2.9686,
"step": 67250
},
{
"epoch": 18.11089577950043,
"grad_norm": 0.415647029876709,
"learning_rate": 0.0003285620035007405,
"loss": 2.9779,
"step": 67300
},
{
"epoch": 18.12435400516796,
"grad_norm": 0.3884516954421997,
"learning_rate": 0.00032836003770028276,
"loss": 2.9692,
"step": 67350
},
{
"epoch": 18.137812230835486,
"grad_norm": 0.418523907661438,
"learning_rate": 0.00032815807189982496,
"loss": 2.9822,
"step": 67400
},
{
"epoch": 18.151270456503013,
"grad_norm": 0.40410879254341125,
"learning_rate": 0.00032795610609936715,
"loss": 2.9871,
"step": 67450
},
{
"epoch": 18.164728682170544,
"grad_norm": 0.426695853471756,
"learning_rate": 0.00032775414029890935,
"loss": 2.9943,
"step": 67500
},
{
"epoch": 18.17818690783807,
"grad_norm": 0.39134326577186584,
"learning_rate": 0.00032755217449845155,
"loss": 2.9876,
"step": 67550
},
{
"epoch": 18.1916451335056,
"grad_norm": 0.3778131306171417,
"learning_rate": 0.0003273502086979938,
"loss": 2.9863,
"step": 67600
},
{
"epoch": 18.205103359173126,
"grad_norm": 0.38964495062828064,
"learning_rate": 0.000327148242897536,
"loss": 2.9871,
"step": 67650
},
{
"epoch": 18.218561584840653,
"grad_norm": 0.38008835911750793,
"learning_rate": 0.0003269462770970782,
"loss": 3.0008,
"step": 67700
},
{
"epoch": 18.232019810508184,
"grad_norm": 0.45064225792884827,
"learning_rate": 0.0003267443112966204,
"loss": 3.0048,
"step": 67750
},
{
"epoch": 18.24547803617571,
"grad_norm": 0.39705249667167664,
"learning_rate": 0.0003265423454961626,
"loss": 3.0072,
"step": 67800
},
{
"epoch": 18.25893626184324,
"grad_norm": 0.41091370582580566,
"learning_rate": 0.00032634037969570483,
"loss": 3.0,
"step": 67850
},
{
"epoch": 18.272394487510766,
"grad_norm": 0.3867056369781494,
"learning_rate": 0.00032613841389524703,
"loss": 2.9791,
"step": 67900
},
{
"epoch": 18.285852713178294,
"grad_norm": 0.42417794466018677,
"learning_rate": 0.0003259364480947892,
"loss": 2.9986,
"step": 67950
},
{
"epoch": 18.29931093884582,
"grad_norm": 0.39678674936294556,
"learning_rate": 0.0003257344822943314,
"loss": 3.0038,
"step": 68000
},
{
"epoch": 18.29931093884582,
"eval_accuracy": 0.39511449857520514,
"eval_loss": 3.289222002029419,
"eval_runtime": 82.6938,
"eval_samples_per_second": 217.816,
"eval_steps_per_second": 13.617,
"step": 68000
},
{
"epoch": 18.31276916451335,
"grad_norm": 0.4160973131656647,
"learning_rate": 0.0003255325164938737,
"loss": 3.0021,
"step": 68050
},
{
"epoch": 18.32622739018088,
"grad_norm": 0.38626375794410706,
"learning_rate": 0.0003253305506934159,
"loss": 3.0122,
"step": 68100
},
{
"epoch": 18.339685615848406,
"grad_norm": 0.40224888920783997,
"learning_rate": 0.0003251285848929581,
"loss": 3.0088,
"step": 68150
},
{
"epoch": 18.353143841515934,
"grad_norm": 0.3748827576637268,
"learning_rate": 0.0003249266190925003,
"loss": 3.0071,
"step": 68200
},
{
"epoch": 18.36660206718346,
"grad_norm": 0.4188649654388428,
"learning_rate": 0.0003247246532920425,
"loss": 3.0084,
"step": 68250
},
{
"epoch": 18.38006029285099,
"grad_norm": 0.3962478041648865,
"learning_rate": 0.00032452268749158476,
"loss": 3.0045,
"step": 68300
},
{
"epoch": 18.39351851851852,
"grad_norm": 0.39891430735588074,
"learning_rate": 0.00032432072169112696,
"loss": 3.0178,
"step": 68350
},
{
"epoch": 18.406976744186046,
"grad_norm": 0.42630842328071594,
"learning_rate": 0.00032411875589066915,
"loss": 3.0289,
"step": 68400
},
{
"epoch": 18.420434969853574,
"grad_norm": 0.4050421118736267,
"learning_rate": 0.00032391679009021135,
"loss": 3.0146,
"step": 68450
},
{
"epoch": 18.4338931955211,
"grad_norm": 0.398600697517395,
"learning_rate": 0.0003237148242897536,
"loss": 3.0169,
"step": 68500
},
{
"epoch": 18.447351421188632,
"grad_norm": 0.3660748600959778,
"learning_rate": 0.0003235128584892958,
"loss": 3.0194,
"step": 68550
},
{
"epoch": 18.46080964685616,
"grad_norm": 0.37340208888053894,
"learning_rate": 0.000323310892688838,
"loss": 3.0204,
"step": 68600
},
{
"epoch": 18.474267872523686,
"grad_norm": 0.36936208605766296,
"learning_rate": 0.0003231089268883802,
"loss": 3.013,
"step": 68650
},
{
"epoch": 18.487726098191214,
"grad_norm": 0.3875936269760132,
"learning_rate": 0.0003229069610879224,
"loss": 3.0274,
"step": 68700
},
{
"epoch": 18.50118432385874,
"grad_norm": 0.4014323353767395,
"learning_rate": 0.00032270499528746463,
"loss": 3.0211,
"step": 68750
},
{
"epoch": 18.514642549526272,
"grad_norm": 0.38240736722946167,
"learning_rate": 0.00032250302948700683,
"loss": 3.0167,
"step": 68800
},
{
"epoch": 18.5281007751938,
"grad_norm": 0.3955930173397064,
"learning_rate": 0.000322301063686549,
"loss": 3.0169,
"step": 68850
},
{
"epoch": 18.541559000861326,
"grad_norm": 0.3775823712348938,
"learning_rate": 0.0003220990978860912,
"loss": 3.0182,
"step": 68900
},
{
"epoch": 18.555017226528854,
"grad_norm": 0.4066317081451416,
"learning_rate": 0.0003218971320856335,
"loss": 3.033,
"step": 68950
},
{
"epoch": 18.56847545219638,
"grad_norm": 0.41924071311950684,
"learning_rate": 0.0003216951662851757,
"loss": 3.0233,
"step": 69000
},
{
"epoch": 18.56847545219638,
"eval_accuracy": 0.395824940363741,
"eval_loss": 3.283914089202881,
"eval_runtime": 82.8099,
"eval_samples_per_second": 217.51,
"eval_steps_per_second": 13.597,
"step": 69000
},
{
"epoch": 18.581933677863912,
"grad_norm": 0.4270181655883789,
"learning_rate": 0.0003214932004847179,
"loss": 3.024,
"step": 69050
},
{
"epoch": 18.59539190353144,
"grad_norm": 0.37374916672706604,
"learning_rate": 0.0003212912346842601,
"loss": 3.0252,
"step": 69100
},
{
"epoch": 18.608850129198967,
"grad_norm": 0.4003197252750397,
"learning_rate": 0.0003210892688838023,
"loss": 3.028,
"step": 69150
},
{
"epoch": 18.622308354866494,
"grad_norm": 0.3980466425418854,
"learning_rate": 0.00032088730308334456,
"loss": 3.0375,
"step": 69200
},
{
"epoch": 18.63576658053402,
"grad_norm": 0.4112990200519562,
"learning_rate": 0.00032068533728288676,
"loss": 3.0368,
"step": 69250
},
{
"epoch": 18.649224806201552,
"grad_norm": 0.3959102928638458,
"learning_rate": 0.00032048337148242895,
"loss": 3.0288,
"step": 69300
},
{
"epoch": 18.66268303186908,
"grad_norm": 0.37589550018310547,
"learning_rate": 0.00032028140568197115,
"loss": 3.027,
"step": 69350
},
{
"epoch": 18.676141257536607,
"grad_norm": 0.40340784192085266,
"learning_rate": 0.00032007943988151334,
"loss": 3.0276,
"step": 69400
},
{
"epoch": 18.689599483204134,
"grad_norm": 0.39021047949790955,
"learning_rate": 0.0003198774740810556,
"loss": 3.0326,
"step": 69450
},
{
"epoch": 18.70305770887166,
"grad_norm": 0.37334316968917847,
"learning_rate": 0.0003196755082805978,
"loss": 3.0385,
"step": 69500
},
{
"epoch": 18.716515934539192,
"grad_norm": 0.41578933596611023,
"learning_rate": 0.00031947354248014,
"loss": 3.0303,
"step": 69550
},
{
"epoch": 18.72997416020672,
"grad_norm": 0.4012208580970764,
"learning_rate": 0.0003192715766796822,
"loss": 3.0441,
"step": 69600
},
{
"epoch": 18.743432385874247,
"grad_norm": 0.3983551859855652,
"learning_rate": 0.0003190696108792244,
"loss": 3.0393,
"step": 69650
},
{
"epoch": 18.756890611541774,
"grad_norm": 0.40729406476020813,
"learning_rate": 0.00031886764507876663,
"loss": 3.0287,
"step": 69700
},
{
"epoch": 18.7703488372093,
"grad_norm": 0.3794533908367157,
"learning_rate": 0.0003186656792783088,
"loss": 3.0286,
"step": 69750
},
{
"epoch": 18.783807062876832,
"grad_norm": 0.3862394094467163,
"learning_rate": 0.000318463713477851,
"loss": 3.038,
"step": 69800
},
{
"epoch": 18.79726528854436,
"grad_norm": 0.4186899662017822,
"learning_rate": 0.00031826174767739327,
"loss": 3.038,
"step": 69850
},
{
"epoch": 18.810723514211887,
"grad_norm": 0.4026969075202942,
"learning_rate": 0.0003180597818769355,
"loss": 3.0385,
"step": 69900
},
{
"epoch": 18.824181739879414,
"grad_norm": 0.3879188597202301,
"learning_rate": 0.0003178578160764777,
"loss": 3.0388,
"step": 69950
},
{
"epoch": 18.83763996554694,
"grad_norm": 0.390610933303833,
"learning_rate": 0.0003176558502760199,
"loss": 3.0461,
"step": 70000
},
{
"epoch": 18.83763996554694,
"eval_accuracy": 0.39624898904100897,
"eval_loss": 3.2779414653778076,
"eval_runtime": 82.6293,
"eval_samples_per_second": 217.986,
"eval_steps_per_second": 13.627,
"step": 70000
},
{
"epoch": 18.85109819121447,
"grad_norm": 0.3776017427444458,
"learning_rate": 0.0003174538844755621,
"loss": 3.0429,
"step": 70050
},
{
"epoch": 18.864556416882,
"grad_norm": 0.4212494194507599,
"learning_rate": 0.0003172519186751043,
"loss": 3.0385,
"step": 70100
},
{
"epoch": 18.878014642549527,
"grad_norm": 0.40709665417671204,
"learning_rate": 0.00031704995287464656,
"loss": 3.0427,
"step": 70150
},
{
"epoch": 18.891472868217054,
"grad_norm": 0.390013724565506,
"learning_rate": 0.00031684798707418875,
"loss": 3.0285,
"step": 70200
},
{
"epoch": 18.90493109388458,
"grad_norm": 0.4204060137271881,
"learning_rate": 0.00031664602127373095,
"loss": 3.0569,
"step": 70250
},
{
"epoch": 18.91838931955211,
"grad_norm": 0.3872334659099579,
"learning_rate": 0.00031644405547327315,
"loss": 3.0413,
"step": 70300
},
{
"epoch": 18.93184754521964,
"grad_norm": 0.45403924584388733,
"learning_rate": 0.00031624208967281534,
"loss": 3.0426,
"step": 70350
},
{
"epoch": 18.945305770887167,
"grad_norm": 0.3969587981700897,
"learning_rate": 0.0003160401238723576,
"loss": 3.0269,
"step": 70400
},
{
"epoch": 18.958763996554694,
"grad_norm": 0.40780627727508545,
"learning_rate": 0.0003158381580718998,
"loss": 3.0336,
"step": 70450
},
{
"epoch": 18.97222222222222,
"grad_norm": 0.38685640692710876,
"learning_rate": 0.000315636192271442,
"loss": 3.0353,
"step": 70500
},
{
"epoch": 18.98568044788975,
"grad_norm": 0.38661956787109375,
"learning_rate": 0.0003154342264709842,
"loss": 3.0438,
"step": 70550
},
{
"epoch": 18.99913867355728,
"grad_norm": 0.3754230737686157,
"learning_rate": 0.00031523226067052643,
"loss": 3.0398,
"step": 70600
},
{
"epoch": 19.012381567614124,
"grad_norm": 0.7405663132667542,
"learning_rate": 0.0003150302948700686,
"loss": 2.9552,
"step": 70650
},
{
"epoch": 19.025839793281655,
"grad_norm": 0.392824649810791,
"learning_rate": 0.0003148283290696108,
"loss": 2.9487,
"step": 70700
},
{
"epoch": 19.039298018949182,
"grad_norm": 0.4475252628326416,
"learning_rate": 0.0003146263632691531,
"loss": 2.9565,
"step": 70750
},
{
"epoch": 19.05275624461671,
"grad_norm": 0.40361085534095764,
"learning_rate": 0.0003144243974686953,
"loss": 2.9636,
"step": 70800
},
{
"epoch": 19.066214470284237,
"grad_norm": 0.3969842493534088,
"learning_rate": 0.0003142224316682375,
"loss": 2.9552,
"step": 70850
},
{
"epoch": 19.079672695951764,
"grad_norm": 0.43156635761260986,
"learning_rate": 0.0003140204658677797,
"loss": 2.9578,
"step": 70900
},
{
"epoch": 19.093130921619295,
"grad_norm": 0.3937775492668152,
"learning_rate": 0.0003138185000673219,
"loss": 2.955,
"step": 70950
},
{
"epoch": 19.106589147286822,
"grad_norm": 0.405343621969223,
"learning_rate": 0.0003136165342668641,
"loss": 2.9654,
"step": 71000
},
{
"epoch": 19.106589147286822,
"eval_accuracy": 0.39571814050472115,
"eval_loss": 3.287280797958374,
"eval_runtime": 82.7144,
"eval_samples_per_second": 217.761,
"eval_steps_per_second": 13.613,
"step": 71000
},
{
"epoch": 19.12004737295435,
"grad_norm": 0.39757075905799866,
"learning_rate": 0.00031341456846640636,
"loss": 2.9649,
"step": 71050
},
{
"epoch": 19.133505598621877,
"grad_norm": 0.40162545442581177,
"learning_rate": 0.00031321260266594855,
"loss": 2.9706,
"step": 71100
},
{
"epoch": 19.146963824289404,
"grad_norm": 0.41259846091270447,
"learning_rate": 0.00031301063686549075,
"loss": 2.9781,
"step": 71150
},
{
"epoch": 19.160422049956935,
"grad_norm": 0.3921213746070862,
"learning_rate": 0.00031280867106503295,
"loss": 2.9711,
"step": 71200
},
{
"epoch": 19.173880275624462,
"grad_norm": 0.4037973880767822,
"learning_rate": 0.00031260670526457514,
"loss": 2.9728,
"step": 71250
},
{
"epoch": 19.18733850129199,
"grad_norm": 0.40468141436576843,
"learning_rate": 0.0003124047394641174,
"loss": 2.9775,
"step": 71300
},
{
"epoch": 19.200796726959517,
"grad_norm": 0.41887059807777405,
"learning_rate": 0.0003122027736636596,
"loss": 2.9831,
"step": 71350
},
{
"epoch": 19.214254952627044,
"grad_norm": 0.3991416096687317,
"learning_rate": 0.0003120008078632018,
"loss": 2.9792,
"step": 71400
},
{
"epoch": 19.227713178294575,
"grad_norm": 0.3977629840373993,
"learning_rate": 0.000311798842062744,
"loss": 2.9892,
"step": 71450
},
{
"epoch": 19.241171403962102,
"grad_norm": 0.41393712162971497,
"learning_rate": 0.0003115968762622862,
"loss": 2.9968,
"step": 71500
},
{
"epoch": 19.25462962962963,
"grad_norm": 0.40587764978408813,
"learning_rate": 0.00031139491046182843,
"loss": 2.9918,
"step": 71550
},
{
"epoch": 19.268087855297157,
"grad_norm": 0.41805192828178406,
"learning_rate": 0.0003111929446613707,
"loss": 2.9922,
"step": 71600
},
{
"epoch": 19.281546080964684,
"grad_norm": 0.39887356758117676,
"learning_rate": 0.0003109909788609129,
"loss": 2.9918,
"step": 71650
},
{
"epoch": 19.295004306632215,
"grad_norm": 0.3872390389442444,
"learning_rate": 0.00031078901306045507,
"loss": 2.9958,
"step": 71700
},
{
"epoch": 19.308462532299743,
"grad_norm": 0.38069647550582886,
"learning_rate": 0.0003105870472599973,
"loss": 2.9893,
"step": 71750
},
{
"epoch": 19.32192075796727,
"grad_norm": 0.42005452513694763,
"learning_rate": 0.0003103850814595395,
"loss": 3.0031,
"step": 71800
},
{
"epoch": 19.335378983634797,
"grad_norm": 0.41195428371429443,
"learning_rate": 0.0003101831156590817,
"loss": 2.9958,
"step": 71850
},
{
"epoch": 19.348837209302324,
"grad_norm": 0.4174123704433441,
"learning_rate": 0.0003099811498586239,
"loss": 2.998,
"step": 71900
},
{
"epoch": 19.362295434969855,
"grad_norm": 0.39601394534111023,
"learning_rate": 0.0003097791840581661,
"loss": 2.9992,
"step": 71950
},
{
"epoch": 19.375753660637383,
"grad_norm": 0.39685019850730896,
"learning_rate": 0.00030957721825770836,
"loss": 2.9866,
"step": 72000
},
{
"epoch": 19.375753660637383,
"eval_accuracy": 0.39592619923312705,
"eval_loss": 3.2863242626190186,
"eval_runtime": 82.6531,
"eval_samples_per_second": 217.923,
"eval_steps_per_second": 13.623,
"step": 72000
},
{
"epoch": 19.38921188630491,
"grad_norm": 0.3975599706172943,
"learning_rate": 0.00030937525245725055,
"loss": 2.9954,
"step": 72050
},
{
"epoch": 19.402670111972437,
"grad_norm": 0.3908389210700989,
"learning_rate": 0.00030917328665679275,
"loss": 2.9847,
"step": 72100
},
{
"epoch": 19.416128337639964,
"grad_norm": 0.41009944677352905,
"learning_rate": 0.00030897132085633494,
"loss": 3.0025,
"step": 72150
},
{
"epoch": 19.429586563307495,
"grad_norm": 0.41253045201301575,
"learning_rate": 0.00030876935505587714,
"loss": 3.016,
"step": 72200
},
{
"epoch": 19.443044788975023,
"grad_norm": 0.42970511317253113,
"learning_rate": 0.0003085673892554194,
"loss": 3.0036,
"step": 72250
},
{
"epoch": 19.45650301464255,
"grad_norm": 0.4200711250305176,
"learning_rate": 0.0003083654234549616,
"loss": 3.0139,
"step": 72300
},
{
"epoch": 19.469961240310077,
"grad_norm": 0.38626357913017273,
"learning_rate": 0.0003081634576545038,
"loss": 3.007,
"step": 72350
},
{
"epoch": 19.483419465977605,
"grad_norm": 0.4040568768978119,
"learning_rate": 0.000307961491854046,
"loss": 3.0104,
"step": 72400
},
{
"epoch": 19.496877691645132,
"grad_norm": 0.4221293032169342,
"learning_rate": 0.00030775952605358823,
"loss": 3.0168,
"step": 72450
},
{
"epoch": 19.510335917312663,
"grad_norm": 0.44530126452445984,
"learning_rate": 0.0003075575602531305,
"loss": 3.0123,
"step": 72500
},
{
"epoch": 19.52379414298019,
"grad_norm": 0.41686683893203735,
"learning_rate": 0.0003073555944526727,
"loss": 3.0089,
"step": 72550
},
{
"epoch": 19.537252368647717,
"grad_norm": 0.4098789095878601,
"learning_rate": 0.00030715362865221487,
"loss": 3.01,
"step": 72600
},
{
"epoch": 19.550710594315245,
"grad_norm": 0.4215698838233948,
"learning_rate": 0.0003069516628517571,
"loss": 3.0068,
"step": 72650
},
{
"epoch": 19.564168819982772,
"grad_norm": 0.4023432731628418,
"learning_rate": 0.0003067496970512993,
"loss": 3.0227,
"step": 72700
},
{
"epoch": 19.577627045650303,
"grad_norm": 0.41406768560409546,
"learning_rate": 0.0003065477312508415,
"loss": 3.0193,
"step": 72750
},
{
"epoch": 19.59108527131783,
"grad_norm": 0.39187172055244446,
"learning_rate": 0.0003063457654503837,
"loss": 3.0061,
"step": 72800
},
{
"epoch": 19.604543496985357,
"grad_norm": 0.3999687433242798,
"learning_rate": 0.0003061437996499259,
"loss": 3.0158,
"step": 72850
},
{
"epoch": 19.618001722652885,
"grad_norm": 0.4067454934120178,
"learning_rate": 0.00030594183384946816,
"loss": 3.0155,
"step": 72900
},
{
"epoch": 19.631459948320412,
"grad_norm": 0.4182448983192444,
"learning_rate": 0.00030573986804901035,
"loss": 3.0148,
"step": 72950
},
{
"epoch": 19.644918173987943,
"grad_norm": 0.3980155289173126,
"learning_rate": 0.00030553790224855255,
"loss": 3.0191,
"step": 73000
},
{
"epoch": 19.644918173987943,
"eval_accuracy": 0.3962462728696199,
"eval_loss": 3.2805590629577637,
"eval_runtime": 82.696,
"eval_samples_per_second": 217.81,
"eval_steps_per_second": 13.616,
"step": 73000
},
{
"epoch": 19.65837639965547,
"grad_norm": 0.4325604736804962,
"learning_rate": 0.00030533593644809474,
"loss": 3.014,
"step": 73050
},
{
"epoch": 19.671834625322997,
"grad_norm": 0.39533132314682007,
"learning_rate": 0.00030513397064763694,
"loss": 3.015,
"step": 73100
},
{
"epoch": 19.685292850990525,
"grad_norm": 0.4273897707462311,
"learning_rate": 0.0003049320048471792,
"loss": 3.0249,
"step": 73150
},
{
"epoch": 19.698751076658052,
"grad_norm": 0.40684744715690613,
"learning_rate": 0.0003047300390467214,
"loss": 3.0173,
"step": 73200
},
{
"epoch": 19.712209302325583,
"grad_norm": 0.4010809063911438,
"learning_rate": 0.0003045280732462636,
"loss": 3.0136,
"step": 73250
},
{
"epoch": 19.72566752799311,
"grad_norm": 0.4373585283756256,
"learning_rate": 0.0003043261074458058,
"loss": 3.0239,
"step": 73300
},
{
"epoch": 19.739125753660637,
"grad_norm": 0.40056490898132324,
"learning_rate": 0.000304124141645348,
"loss": 3.0152,
"step": 73350
},
{
"epoch": 19.752583979328165,
"grad_norm": 0.4025675356388092,
"learning_rate": 0.0003039221758448903,
"loss": 3.0224,
"step": 73400
},
{
"epoch": 19.766042204995692,
"grad_norm": 0.42092180252075195,
"learning_rate": 0.0003037202100444325,
"loss": 3.0345,
"step": 73450
},
{
"epoch": 19.779500430663223,
"grad_norm": 0.4239610731601715,
"learning_rate": 0.00030351824424397467,
"loss": 3.03,
"step": 73500
},
{
"epoch": 19.79295865633075,
"grad_norm": 0.4084107577800751,
"learning_rate": 0.00030331627844351687,
"loss": 3.0209,
"step": 73550
},
{
"epoch": 19.806416881998278,
"grad_norm": 0.3832364082336426,
"learning_rate": 0.0003031143126430591,
"loss": 3.0323,
"step": 73600
},
{
"epoch": 19.819875107665805,
"grad_norm": 0.40344634652137756,
"learning_rate": 0.0003029123468426013,
"loss": 3.0328,
"step": 73650
},
{
"epoch": 19.833333333333332,
"grad_norm": 0.39347928762435913,
"learning_rate": 0.0003027103810421435,
"loss": 3.0273,
"step": 73700
},
{
"epoch": 19.846791559000863,
"grad_norm": 0.40947094559669495,
"learning_rate": 0.0003025084152416857,
"loss": 3.0278,
"step": 73750
},
{
"epoch": 19.86024978466839,
"grad_norm": 0.4086831510066986,
"learning_rate": 0.0003023064494412279,
"loss": 3.0283,
"step": 73800
},
{
"epoch": 19.873708010335918,
"grad_norm": 0.41457435488700867,
"learning_rate": 0.00030210448364077015,
"loss": 3.0201,
"step": 73850
},
{
"epoch": 19.887166236003445,
"grad_norm": 0.4160161316394806,
"learning_rate": 0.00030190251784031235,
"loss": 3.0306,
"step": 73900
},
{
"epoch": 19.900624461670972,
"grad_norm": 0.39771005511283875,
"learning_rate": 0.00030170055203985454,
"loss": 3.0362,
"step": 73950
},
{
"epoch": 19.9140826873385,
"grad_norm": 0.4261379837989807,
"learning_rate": 0.00030149858623939674,
"loss": 3.0214,
"step": 74000
},
{
"epoch": 19.9140826873385,
"eval_accuracy": 0.3966103484826163,
"eval_loss": 3.275501012802124,
"eval_runtime": 82.6912,
"eval_samples_per_second": 217.822,
"eval_steps_per_second": 13.617,
"step": 74000
},
{
"epoch": 19.92754091300603,
"grad_norm": 0.399692177772522,
"learning_rate": 0.00030129662043893894,
"loss": 3.029,
"step": 74050
},
{
"epoch": 19.940999138673558,
"grad_norm": 0.41963309049606323,
"learning_rate": 0.0003010946546384812,
"loss": 3.0368,
"step": 74100
},
{
"epoch": 19.954457364341085,
"grad_norm": 0.3987467885017395,
"learning_rate": 0.0003008926888380234,
"loss": 3.0319,
"step": 74150
},
{
"epoch": 19.967915590008612,
"grad_norm": 0.43355318903923035,
"learning_rate": 0.0003006907230375656,
"loss": 3.0359,
"step": 74200
},
{
"epoch": 19.98137381567614,
"grad_norm": 0.3953664302825928,
"learning_rate": 0.0003004887572371078,
"loss": 3.03,
"step": 74250
},
{
"epoch": 19.99483204134367,
"grad_norm": 0.38316938281059265,
"learning_rate": 0.0003002867914366501,
"loss": 3.0284,
"step": 74300
},
{
"epoch": 20.00807493540052,
"grad_norm": 0.4180009663105011,
"learning_rate": 0.0003000848256361923,
"loss": 2.9777,
"step": 74350
},
{
"epoch": 20.021533161068046,
"grad_norm": 0.4076843559741974,
"learning_rate": 0.00029988285983573447,
"loss": 2.9305,
"step": 74400
},
{
"epoch": 20.034991386735573,
"grad_norm": 0.41239139437675476,
"learning_rate": 0.00029968089403527667,
"loss": 2.9511,
"step": 74450
},
{
"epoch": 20.0484496124031,
"grad_norm": 0.42153266072273254,
"learning_rate": 0.0002994789282348189,
"loss": 2.9476,
"step": 74500
},
{
"epoch": 20.061907838070628,
"grad_norm": 0.3956698477268219,
"learning_rate": 0.0002992769624343611,
"loss": 2.9463,
"step": 74550
},
{
"epoch": 20.075366063738155,
"grad_norm": 0.4250684976577759,
"learning_rate": 0.0002990749966339033,
"loss": 2.9578,
"step": 74600
},
{
"epoch": 20.088824289405686,
"grad_norm": 0.39493197202682495,
"learning_rate": 0.0002988730308334455,
"loss": 2.9589,
"step": 74650
},
{
"epoch": 20.102282515073213,
"grad_norm": 0.4212232530117035,
"learning_rate": 0.0002986710650329877,
"loss": 2.9497,
"step": 74700
},
{
"epoch": 20.11574074074074,
"grad_norm": 0.42299792170524597,
"learning_rate": 0.00029846909923252995,
"loss": 2.9628,
"step": 74750
},
{
"epoch": 20.129198966408268,
"grad_norm": 0.41732659935951233,
"learning_rate": 0.00029826713343207215,
"loss": 2.9621,
"step": 74800
},
{
"epoch": 20.142657192075795,
"grad_norm": 0.44259753823280334,
"learning_rate": 0.00029806516763161435,
"loss": 2.956,
"step": 74850
},
{
"epoch": 20.156115417743326,
"grad_norm": 0.39583277702331543,
"learning_rate": 0.00029786320183115654,
"loss": 2.9717,
"step": 74900
},
{
"epoch": 20.169573643410853,
"grad_norm": 0.40104466676712036,
"learning_rate": 0.0002976612360306988,
"loss": 2.9783,
"step": 74950
},
{
"epoch": 20.18303186907838,
"grad_norm": 0.4311688244342804,
"learning_rate": 0.000297459270230241,
"loss": 2.9697,
"step": 75000
},
{
"epoch": 20.18303186907838,
"eval_accuracy": 0.39598725876595425,
"eval_loss": 3.2868382930755615,
"eval_runtime": 82.7286,
"eval_samples_per_second": 217.724,
"eval_steps_per_second": 13.611,
"step": 75000
},
{
"epoch": 20.196490094745908,
"grad_norm": 0.42635202407836914,
"learning_rate": 0.0002972573044297832,
"loss": 2.9699,
"step": 75050
},
{
"epoch": 20.209948320413435,
"grad_norm": 0.4375520646572113,
"learning_rate": 0.00029705533862932543,
"loss": 2.9781,
"step": 75100
},
{
"epoch": 20.223406546080966,
"grad_norm": 0.4293828010559082,
"learning_rate": 0.00029685337282886763,
"loss": 2.9741,
"step": 75150
},
{
"epoch": 20.236864771748493,
"grad_norm": 0.44403374195098877,
"learning_rate": 0.0002966514070284098,
"loss": 2.9841,
"step": 75200
},
{
"epoch": 20.25032299741602,
"grad_norm": 0.4646786153316498,
"learning_rate": 0.000296449441227952,
"loss": 2.9805,
"step": 75250
},
{
"epoch": 20.263781223083548,
"grad_norm": 0.40680402517318726,
"learning_rate": 0.0002962474754274942,
"loss": 2.9632,
"step": 75300
},
{
"epoch": 20.277239448751075,
"grad_norm": 0.4195573627948761,
"learning_rate": 0.00029604550962703647,
"loss": 2.9789,
"step": 75350
},
{
"epoch": 20.290697674418606,
"grad_norm": 0.431867390871048,
"learning_rate": 0.00029584354382657867,
"loss": 2.9831,
"step": 75400
},
{
"epoch": 20.304155900086133,
"grad_norm": 0.40752989053726196,
"learning_rate": 0.0002956415780261209,
"loss": 2.9912,
"step": 75450
},
{
"epoch": 20.31761412575366,
"grad_norm": 0.4081822633743286,
"learning_rate": 0.0002954396122256631,
"loss": 2.9773,
"step": 75500
},
{
"epoch": 20.331072351421188,
"grad_norm": 0.4295860826969147,
"learning_rate": 0.0002952376464252053,
"loss": 2.984,
"step": 75550
},
{
"epoch": 20.344530577088715,
"grad_norm": 0.4347155690193176,
"learning_rate": 0.0002950356806247475,
"loss": 2.9897,
"step": 75600
},
{
"epoch": 20.357988802756246,
"grad_norm": 0.4089062511920929,
"learning_rate": 0.0002948337148242897,
"loss": 2.9882,
"step": 75650
},
{
"epoch": 20.371447028423773,
"grad_norm": 0.3969566226005554,
"learning_rate": 0.00029463174902383195,
"loss": 2.9894,
"step": 75700
},
{
"epoch": 20.3849052540913,
"grad_norm": 0.42615869641304016,
"learning_rate": 0.00029442978322337415,
"loss": 2.9842,
"step": 75750
},
{
"epoch": 20.398363479758828,
"grad_norm": 0.38849231600761414,
"learning_rate": 0.00029422781742291634,
"loss": 2.9922,
"step": 75800
},
{
"epoch": 20.411821705426355,
"grad_norm": 0.440918505191803,
"learning_rate": 0.0002940258516224586,
"loss": 2.9941,
"step": 75850
},
{
"epoch": 20.425279931093886,
"grad_norm": 0.42064759135246277,
"learning_rate": 0.0002938238858220008,
"loss": 2.9912,
"step": 75900
},
{
"epoch": 20.438738156761413,
"grad_norm": 0.4000251293182373,
"learning_rate": 0.000293621920021543,
"loss": 2.9921,
"step": 75950
},
{
"epoch": 20.45219638242894,
"grad_norm": 0.4173610806465149,
"learning_rate": 0.0002934199542210852,
"loss": 2.9951,
"step": 76000
},
{
"epoch": 20.45219638242894,
"eval_accuracy": 0.3962431221108085,
"eval_loss": 3.283843994140625,
"eval_runtime": 82.8612,
"eval_samples_per_second": 217.375,
"eval_steps_per_second": 13.589,
"step": 76000
},
{
"epoch": 20.465654608096468,
"grad_norm": 0.41962048411369324,
"learning_rate": 0.00029321798842062743,
"loss": 2.9937,
"step": 76050
},
{
"epoch": 20.479112833763995,
"grad_norm": 0.4084608554840088,
"learning_rate": 0.00029301602262016963,
"loss": 3.003,
"step": 76100
},
{
"epoch": 20.492571059431526,
"grad_norm": 0.4410095512866974,
"learning_rate": 0.0002928140568197118,
"loss": 2.9963,
"step": 76150
},
{
"epoch": 20.506029285099054,
"grad_norm": 0.4109020233154297,
"learning_rate": 0.000292612091019254,
"loss": 2.9895,
"step": 76200
},
{
"epoch": 20.51948751076658,
"grad_norm": 0.42620041966438293,
"learning_rate": 0.0002924101252187962,
"loss": 2.9864,
"step": 76250
},
{
"epoch": 20.532945736434108,
"grad_norm": 0.4074179232120514,
"learning_rate": 0.00029220815941833847,
"loss": 2.996,
"step": 76300
},
{
"epoch": 20.546403962101635,
"grad_norm": 0.4069136381149292,
"learning_rate": 0.0002920061936178807,
"loss": 2.9997,
"step": 76350
},
{
"epoch": 20.559862187769163,
"grad_norm": 0.4035022258758545,
"learning_rate": 0.0002918042278174229,
"loss": 2.9918,
"step": 76400
},
{
"epoch": 20.573320413436694,
"grad_norm": 0.41788581013679504,
"learning_rate": 0.0002916022620169651,
"loss": 3.0028,
"step": 76450
},
{
"epoch": 20.58677863910422,
"grad_norm": 0.4092438220977783,
"learning_rate": 0.0002914002962165073,
"loss": 3.0067,
"step": 76500
},
{
"epoch": 20.600236864771748,
"grad_norm": 0.4183063507080078,
"learning_rate": 0.0002911983304160495,
"loss": 3.0002,
"step": 76550
},
{
"epoch": 20.613695090439276,
"grad_norm": 0.4091934561729431,
"learning_rate": 0.00029099636461559175,
"loss": 3.004,
"step": 76600
},
{
"epoch": 20.627153316106803,
"grad_norm": 0.4188084900379181,
"learning_rate": 0.00029079439881513395,
"loss": 3.0032,
"step": 76650
},
{
"epoch": 20.640611541774334,
"grad_norm": 0.41433677077293396,
"learning_rate": 0.00029059243301467614,
"loss": 3.0022,
"step": 76700
},
{
"epoch": 20.65406976744186,
"grad_norm": 0.4140937328338623,
"learning_rate": 0.0002903904672142184,
"loss": 3.0157,
"step": 76750
},
{
"epoch": 20.66752799310939,
"grad_norm": 0.42956727743148804,
"learning_rate": 0.0002901885014137606,
"loss": 3.0138,
"step": 76800
},
{
"epoch": 20.680986218776916,
"grad_norm": 0.4086523950099945,
"learning_rate": 0.0002899865356133028,
"loss": 3.0111,
"step": 76850
},
{
"epoch": 20.694444444444443,
"grad_norm": 0.4188919961452484,
"learning_rate": 0.000289784569812845,
"loss": 3.0053,
"step": 76900
},
{
"epoch": 20.707902670111974,
"grad_norm": 0.40631499886512756,
"learning_rate": 0.00028958260401238723,
"loss": 3.0076,
"step": 76950
},
{
"epoch": 20.7213608957795,
"grad_norm": 0.4247263967990875,
"learning_rate": 0.00028938063821192943,
"loss": 3.0042,
"step": 77000
},
{
"epoch": 20.7213608957795,
"eval_accuracy": 0.396532340040321,
"eval_loss": 3.2751851081848145,
"eval_runtime": 82.7107,
"eval_samples_per_second": 217.771,
"eval_steps_per_second": 13.614,
"step": 77000
},
{
"epoch": 20.73481912144703,
"grad_norm": 0.4177243709564209,
"learning_rate": 0.0002891786724114716,
"loss": 3.0108,
"step": 77050
},
{
"epoch": 20.748277347114556,
"grad_norm": 0.4227748513221741,
"learning_rate": 0.0002889767066110138,
"loss": 3.0043,
"step": 77100
},
{
"epoch": 20.761735572782083,
"grad_norm": 0.3870146572589874,
"learning_rate": 0.000288774740810556,
"loss": 3.0221,
"step": 77150
},
{
"epoch": 20.775193798449614,
"grad_norm": 0.42427289485931396,
"learning_rate": 0.00028857277501009827,
"loss": 3.0145,
"step": 77200
},
{
"epoch": 20.78865202411714,
"grad_norm": 0.4343383312225342,
"learning_rate": 0.00028837080920964046,
"loss": 3.0135,
"step": 77250
},
{
"epoch": 20.80211024978467,
"grad_norm": 0.4120021462440491,
"learning_rate": 0.0002881688434091827,
"loss": 3.0126,
"step": 77300
},
{
"epoch": 20.815568475452196,
"grad_norm": 0.392022967338562,
"learning_rate": 0.0002879668776087249,
"loss": 3.0093,
"step": 77350
},
{
"epoch": 20.829026701119723,
"grad_norm": 0.4146294593811035,
"learning_rate": 0.0002877649118082671,
"loss": 3.0097,
"step": 77400
},
{
"epoch": 20.842484926787254,
"grad_norm": 0.39745283126831055,
"learning_rate": 0.0002875629460078093,
"loss": 3.0189,
"step": 77450
},
{
"epoch": 20.85594315245478,
"grad_norm": 0.42291516065597534,
"learning_rate": 0.0002873609802073515,
"loss": 3.016,
"step": 77500
},
{
"epoch": 20.86940137812231,
"grad_norm": 0.440886527299881,
"learning_rate": 0.00028715901440689375,
"loss": 3.0192,
"step": 77550
},
{
"epoch": 20.882859603789836,
"grad_norm": 0.4255584478378296,
"learning_rate": 0.00028695704860643594,
"loss": 3.0171,
"step": 77600
},
{
"epoch": 20.896317829457363,
"grad_norm": 0.4507913589477539,
"learning_rate": 0.0002867550828059782,
"loss": 3.0184,
"step": 77650
},
{
"epoch": 20.909776055124894,
"grad_norm": 0.4259226620197296,
"learning_rate": 0.0002865531170055204,
"loss": 3.0238,
"step": 77700
},
{
"epoch": 20.92323428079242,
"grad_norm": 0.44342297315597534,
"learning_rate": 0.0002863511512050626,
"loss": 3.012,
"step": 77750
},
{
"epoch": 20.93669250645995,
"grad_norm": 0.3984871208667755,
"learning_rate": 0.0002861491854046048,
"loss": 3.0149,
"step": 77800
},
{
"epoch": 20.950150732127476,
"grad_norm": 0.42094311118125916,
"learning_rate": 0.000285947219604147,
"loss": 3.0184,
"step": 77850
},
{
"epoch": 20.963608957795003,
"grad_norm": 0.4364471435546875,
"learning_rate": 0.00028574525380368923,
"loss": 3.0233,
"step": 77900
},
{
"epoch": 20.977067183462534,
"grad_norm": 0.44835424423217773,
"learning_rate": 0.0002855432880032314,
"loss": 3.0139,
"step": 77950
},
{
"epoch": 20.99052540913006,
"grad_norm": 0.425870805978775,
"learning_rate": 0.0002853413222027736,
"loss": 3.027,
"step": 78000
},
{
"epoch": 20.99052540913006,
"eval_accuracy": 0.3969768143264351,
"eval_loss": 3.2710349559783936,
"eval_runtime": 82.6182,
"eval_samples_per_second": 218.015,
"eval_steps_per_second": 13.629,
"step": 78000
},
{
"epoch": 21.00376830318691,
"grad_norm": 0.4237120747566223,
"learning_rate": 0.00028513935640231587,
"loss": 2.9965,
"step": 78050
},
{
"epoch": 21.017226528854437,
"grad_norm": 0.41989701986312866,
"learning_rate": 0.00028493739060185807,
"loss": 2.9233,
"step": 78100
},
{
"epoch": 21.030684754521964,
"grad_norm": 0.43977585434913635,
"learning_rate": 0.00028473542480140026,
"loss": 2.9283,
"step": 78150
},
{
"epoch": 21.04414298018949,
"grad_norm": 0.4168258011341095,
"learning_rate": 0.0002845334590009425,
"loss": 2.9319,
"step": 78200
},
{
"epoch": 21.05760120585702,
"grad_norm": 0.4202430248260498,
"learning_rate": 0.0002843314932004847,
"loss": 2.933,
"step": 78250
},
{
"epoch": 21.07105943152455,
"grad_norm": 0.414730966091156,
"learning_rate": 0.0002841295274000269,
"loss": 2.9526,
"step": 78300
},
{
"epoch": 21.084517657192077,
"grad_norm": 0.4064372181892395,
"learning_rate": 0.0002839275615995691,
"loss": 2.9348,
"step": 78350
},
{
"epoch": 21.097975882859604,
"grad_norm": 0.43984556198120117,
"learning_rate": 0.0002837255957991113,
"loss": 2.9432,
"step": 78400
},
{
"epoch": 21.11143410852713,
"grad_norm": 0.41910257935523987,
"learning_rate": 0.00028352362999865355,
"loss": 2.9463,
"step": 78450
},
{
"epoch": 21.12489233419466,
"grad_norm": 0.42459166049957275,
"learning_rate": 0.00028332166419819574,
"loss": 2.9536,
"step": 78500
},
{
"epoch": 21.13835055986219,
"grad_norm": 0.4346620738506317,
"learning_rate": 0.000283119698397738,
"loss": 2.9387,
"step": 78550
},
{
"epoch": 21.151808785529717,
"grad_norm": 0.42380502820014954,
"learning_rate": 0.0002829177325972802,
"loss": 2.9534,
"step": 78600
},
{
"epoch": 21.165267011197244,
"grad_norm": 0.42337992787361145,
"learning_rate": 0.0002827157667968224,
"loss": 2.9589,
"step": 78650
},
{
"epoch": 21.17872523686477,
"grad_norm": 0.4302671551704407,
"learning_rate": 0.0002825138009963646,
"loss": 2.9528,
"step": 78700
},
{
"epoch": 21.1921834625323,
"grad_norm": 0.4296605587005615,
"learning_rate": 0.0002823118351959068,
"loss": 2.9575,
"step": 78750
},
{
"epoch": 21.205641688199826,
"grad_norm": 0.41230344772338867,
"learning_rate": 0.00028210986939544903,
"loss": 2.9551,
"step": 78800
},
{
"epoch": 21.219099913867357,
"grad_norm": 0.4141388535499573,
"learning_rate": 0.0002819079035949912,
"loss": 2.9761,
"step": 78850
},
{
"epoch": 21.232558139534884,
"grad_norm": 0.4193553328514099,
"learning_rate": 0.0002817059377945334,
"loss": 2.9654,
"step": 78900
},
{
"epoch": 21.24601636520241,
"grad_norm": 0.4132939577102661,
"learning_rate": 0.00028150397199407567,
"loss": 2.9713,
"step": 78950
},
{
"epoch": 21.25947459086994,
"grad_norm": 0.408407062292099,
"learning_rate": 0.00028130200619361787,
"loss": 2.973,
"step": 79000
},
{
"epoch": 21.25947459086994,
"eval_accuracy": 0.3961210030451541,
"eval_loss": 3.2851531505584717,
"eval_runtime": 82.6758,
"eval_samples_per_second": 217.863,
"eval_steps_per_second": 13.619,
"step": 79000
},
{
"epoch": 21.272932816537466,
"grad_norm": 0.4197745621204376,
"learning_rate": 0.00028110004039316006,
"loss": 2.9851,
"step": 79050
},
{
"epoch": 21.286391042204997,
"grad_norm": 0.5212720632553101,
"learning_rate": 0.00028089807459270226,
"loss": 2.9733,
"step": 79100
},
{
"epoch": 21.299849267872524,
"grad_norm": 0.4182852506637573,
"learning_rate": 0.0002806961087922445,
"loss": 2.9694,
"step": 79150
},
{
"epoch": 21.31330749354005,
"grad_norm": 0.4595959782600403,
"learning_rate": 0.0002804941429917867,
"loss": 2.9633,
"step": 79200
},
{
"epoch": 21.32676571920758,
"grad_norm": 0.42938849329948425,
"learning_rate": 0.0002802921771913289,
"loss": 2.9826,
"step": 79250
},
{
"epoch": 21.340223944875106,
"grad_norm": 0.46772459149360657,
"learning_rate": 0.0002800902113908711,
"loss": 2.9785,
"step": 79300
},
{
"epoch": 21.353682170542637,
"grad_norm": 0.41134586930274963,
"learning_rate": 0.0002798882455904133,
"loss": 2.9794,
"step": 79350
},
{
"epoch": 21.367140396210164,
"grad_norm": 0.42376479506492615,
"learning_rate": 0.00027968627978995555,
"loss": 2.9778,
"step": 79400
},
{
"epoch": 21.38059862187769,
"grad_norm": 0.39155057072639465,
"learning_rate": 0.00027948431398949774,
"loss": 2.9799,
"step": 79450
},
{
"epoch": 21.39405684754522,
"grad_norm": 0.43734702467918396,
"learning_rate": 0.00027928234818904,
"loss": 2.9694,
"step": 79500
},
{
"epoch": 21.407515073212746,
"grad_norm": 0.4178415834903717,
"learning_rate": 0.0002790803823885822,
"loss": 2.973,
"step": 79550
},
{
"epoch": 21.420973298880277,
"grad_norm": 0.4427221119403839,
"learning_rate": 0.0002788784165881244,
"loss": 2.9938,
"step": 79600
},
{
"epoch": 21.434431524547804,
"grad_norm": 0.4959544241428375,
"learning_rate": 0.0002786764507876666,
"loss": 2.9921,
"step": 79650
},
{
"epoch": 21.44788975021533,
"grad_norm": 0.4457259774208069,
"learning_rate": 0.0002784744849872088,
"loss": 2.9783,
"step": 79700
},
{
"epoch": 21.46134797588286,
"grad_norm": 0.43760278820991516,
"learning_rate": 0.000278272519186751,
"loss": 2.9942,
"step": 79750
},
{
"epoch": 21.474806201550386,
"grad_norm": 0.42622968554496765,
"learning_rate": 0.0002780705533862932,
"loss": 2.9781,
"step": 79800
},
{
"epoch": 21.488264427217917,
"grad_norm": 0.41008955240249634,
"learning_rate": 0.0002778685875858355,
"loss": 2.9747,
"step": 79850
},
{
"epoch": 21.501722652885444,
"grad_norm": 0.433695524930954,
"learning_rate": 0.00027766662178537767,
"loss": 2.9927,
"step": 79900
},
{
"epoch": 21.51518087855297,
"grad_norm": 0.43649762868881226,
"learning_rate": 0.00027746465598491987,
"loss": 2.9798,
"step": 79950
},
{
"epoch": 21.5286391042205,
"grad_norm": 0.41045019030570984,
"learning_rate": 0.00027726269018446206,
"loss": 2.9926,
"step": 80000
},
{
"epoch": 21.5286391042205,
"eval_accuracy": 0.3964692162172381,
"eval_loss": 3.282101631164551,
"eval_runtime": 82.7128,
"eval_samples_per_second": 217.766,
"eval_steps_per_second": 13.613,
"step": 80000
},
{
"epoch": 21.542097329888026,
"grad_norm": 0.4404677748680115,
"learning_rate": 0.00027706072438400426,
"loss": 2.9904,
"step": 80050
},
{
"epoch": 21.555555555555557,
"grad_norm": 0.4134572446346283,
"learning_rate": 0.0002768587585835465,
"loss": 2.9932,
"step": 80100
},
{
"epoch": 21.569013781223084,
"grad_norm": 0.40927648544311523,
"learning_rate": 0.0002766567927830887,
"loss": 2.9877,
"step": 80150
},
{
"epoch": 21.58247200689061,
"grad_norm": 0.4398091435432434,
"learning_rate": 0.0002764548269826309,
"loss": 2.9902,
"step": 80200
},
{
"epoch": 21.59593023255814,
"grad_norm": 0.3910978436470032,
"learning_rate": 0.0002762528611821731,
"loss": 2.9976,
"step": 80250
},
{
"epoch": 21.609388458225666,
"grad_norm": 0.42031171917915344,
"learning_rate": 0.00027605089538171535,
"loss": 2.9992,
"step": 80300
},
{
"epoch": 21.622846683893197,
"grad_norm": 0.4312560558319092,
"learning_rate": 0.00027584892958125754,
"loss": 2.9938,
"step": 80350
},
{
"epoch": 21.636304909560724,
"grad_norm": 0.3953777849674225,
"learning_rate": 0.0002756469637807998,
"loss": 2.9956,
"step": 80400
},
{
"epoch": 21.649763135228252,
"grad_norm": 0.4320078492164612,
"learning_rate": 0.000275444997980342,
"loss": 3.0005,
"step": 80450
},
{
"epoch": 21.66322136089578,
"grad_norm": 0.4328147768974304,
"learning_rate": 0.0002752430321798842,
"loss": 2.9935,
"step": 80500
},
{
"epoch": 21.676679586563306,
"grad_norm": 0.4226149618625641,
"learning_rate": 0.0002750410663794264,
"loss": 3.0015,
"step": 80550
},
{
"epoch": 21.690137812230837,
"grad_norm": 0.4088149070739746,
"learning_rate": 0.0002748391005789686,
"loss": 2.9897,
"step": 80600
},
{
"epoch": 21.703596037898365,
"grad_norm": 0.4195426106452942,
"learning_rate": 0.00027463713477851083,
"loss": 2.9933,
"step": 80650
},
{
"epoch": 21.717054263565892,
"grad_norm": 0.4243777394294739,
"learning_rate": 0.000274435168978053,
"loss": 2.9952,
"step": 80700
},
{
"epoch": 21.73051248923342,
"grad_norm": 0.46793031692504883,
"learning_rate": 0.0002742332031775953,
"loss": 2.9951,
"step": 80750
},
{
"epoch": 21.743970714900946,
"grad_norm": 0.41916611790657043,
"learning_rate": 0.00027403123737713747,
"loss": 3.0011,
"step": 80800
},
{
"epoch": 21.757428940568474,
"grad_norm": 0.415775865316391,
"learning_rate": 0.00027382927157667967,
"loss": 2.9959,
"step": 80850
},
{
"epoch": 21.770887166236005,
"grad_norm": 0.4347650110721588,
"learning_rate": 0.00027362730577622186,
"loss": 3.0038,
"step": 80900
},
{
"epoch": 21.784345391903532,
"grad_norm": 0.421785444021225,
"learning_rate": 0.00027342533997576406,
"loss": 3.0099,
"step": 80950
},
{
"epoch": 21.79780361757106,
"grad_norm": 0.41916942596435547,
"learning_rate": 0.0002732233741753063,
"loss": 3.0082,
"step": 81000
},
{
"epoch": 21.79780361757106,
"eval_accuracy": 0.39685675955103644,
"eval_loss": 3.2761969566345215,
"eval_runtime": 82.7323,
"eval_samples_per_second": 217.714,
"eval_steps_per_second": 13.61,
"step": 81000
},
{
"epoch": 21.811261843238587,
"grad_norm": 0.40934279561042786,
"learning_rate": 0.0002730214083748485,
"loss": 2.9914,
"step": 81050
},
{
"epoch": 21.824720068906114,
"grad_norm": 0.42549100518226624,
"learning_rate": 0.0002728194425743907,
"loss": 3.0078,
"step": 81100
},
{
"epoch": 21.838178294573645,
"grad_norm": 0.44911614060401917,
"learning_rate": 0.0002726174767739329,
"loss": 3.002,
"step": 81150
},
{
"epoch": 21.851636520241172,
"grad_norm": 0.4416077733039856,
"learning_rate": 0.00027241551097347515,
"loss": 3.0069,
"step": 81200
},
{
"epoch": 21.8650947459087,
"grad_norm": 0.41148641705513,
"learning_rate": 0.00027221354517301734,
"loss": 3.0077,
"step": 81250
},
{
"epoch": 21.878552971576227,
"grad_norm": 0.4299815893173218,
"learning_rate": 0.00027201157937255954,
"loss": 3.0209,
"step": 81300
},
{
"epoch": 21.892011197243754,
"grad_norm": 0.4402017891407013,
"learning_rate": 0.0002718096135721018,
"loss": 3.0,
"step": 81350
},
{
"epoch": 21.905469422911285,
"grad_norm": 0.42350611090660095,
"learning_rate": 0.000271607647771644,
"loss": 3.0014,
"step": 81400
},
{
"epoch": 21.918927648578812,
"grad_norm": 0.4433906376361847,
"learning_rate": 0.0002714056819711862,
"loss": 3.0105,
"step": 81450
},
{
"epoch": 21.93238587424634,
"grad_norm": 0.41976824402809143,
"learning_rate": 0.0002712037161707284,
"loss": 3.0113,
"step": 81500
},
{
"epoch": 21.945844099913867,
"grad_norm": 0.38680437207221985,
"learning_rate": 0.0002710017503702706,
"loss": 3.0052,
"step": 81550
},
{
"epoch": 21.959302325581394,
"grad_norm": 0.45060545206069946,
"learning_rate": 0.0002707997845698128,
"loss": 2.9957,
"step": 81600
},
{
"epoch": 21.972760551248925,
"grad_norm": 0.43828296661376953,
"learning_rate": 0.000270597818769355,
"loss": 3.0206,
"step": 81650
},
{
"epoch": 21.986218776916452,
"grad_norm": 0.42935317754745483,
"learning_rate": 0.00027039585296889727,
"loss": 3.0022,
"step": 81700
},
{
"epoch": 21.99967700258398,
"grad_norm": 0.4025055766105652,
"learning_rate": 0.00027019388716843947,
"loss": 3.0176,
"step": 81750
},
{
"epoch": 22.012919896640827,
"grad_norm": 0.409675657749176,
"learning_rate": 0.00026999192136798166,
"loss": 2.9137,
"step": 81800
},
{
"epoch": 22.026378122308355,
"grad_norm": 0.4098537862300873,
"learning_rate": 0.00026978995556752386,
"loss": 2.9174,
"step": 81850
},
{
"epoch": 22.039836347975882,
"grad_norm": 0.4596457779407501,
"learning_rate": 0.00026958798976706606,
"loss": 2.931,
"step": 81900
},
{
"epoch": 22.05329457364341,
"grad_norm": 0.41698116064071655,
"learning_rate": 0.0002693860239666083,
"loss": 2.9375,
"step": 81950
},
{
"epoch": 22.06675279931094,
"grad_norm": 0.4248860478401184,
"learning_rate": 0.0002691840581661505,
"loss": 2.9219,
"step": 82000
},
{
"epoch": 22.06675279931094,
"eval_accuracy": 0.3962550732649206,
"eval_loss": 3.2888433933258057,
"eval_runtime": 82.6869,
"eval_samples_per_second": 217.834,
"eval_steps_per_second": 13.618,
"step": 82000
},
{
"epoch": 22.080211024978468,
"grad_norm": 0.4336738586425781,
"learning_rate": 0.0002689820923656927,
"loss": 2.9384,
"step": 82050
},
{
"epoch": 22.093669250645995,
"grad_norm": 0.40990057587623596,
"learning_rate": 0.00026878012656523495,
"loss": 2.9495,
"step": 82100
},
{
"epoch": 22.107127476313522,
"grad_norm": 0.44524773955345154,
"learning_rate": 0.00026857816076477714,
"loss": 2.9423,
"step": 82150
},
{
"epoch": 22.12058570198105,
"grad_norm": 0.4511680006980896,
"learning_rate": 0.00026837619496431934,
"loss": 2.9384,
"step": 82200
},
{
"epoch": 22.13404392764858,
"grad_norm": 0.4383542239665985,
"learning_rate": 0.0002681742291638616,
"loss": 2.9381,
"step": 82250
},
{
"epoch": 22.147502153316108,
"grad_norm": 0.43761157989501953,
"learning_rate": 0.0002679722633634038,
"loss": 2.9498,
"step": 82300
},
{
"epoch": 22.160960378983635,
"grad_norm": 0.43207573890686035,
"learning_rate": 0.000267770297562946,
"loss": 2.9412,
"step": 82350
},
{
"epoch": 22.174418604651162,
"grad_norm": 0.4764629602432251,
"learning_rate": 0.0002675683317624882,
"loss": 2.9548,
"step": 82400
},
{
"epoch": 22.18787683031869,
"grad_norm": 0.41554775834083557,
"learning_rate": 0.0002673663659620304,
"loss": 2.9427,
"step": 82450
},
{
"epoch": 22.20133505598622,
"grad_norm": 0.4455977976322174,
"learning_rate": 0.0002671644001615726,
"loss": 2.9521,
"step": 82500
},
{
"epoch": 22.214793281653748,
"grad_norm": 0.41451287269592285,
"learning_rate": 0.0002669624343611148,
"loss": 2.9567,
"step": 82550
},
{
"epoch": 22.228251507321275,
"grad_norm": 0.4103204309940338,
"learning_rate": 0.00026676046856065707,
"loss": 2.9552,
"step": 82600
},
{
"epoch": 22.241709732988802,
"grad_norm": 0.4221997559070587,
"learning_rate": 0.00026655850276019927,
"loss": 2.9584,
"step": 82650
},
{
"epoch": 22.25516795865633,
"grad_norm": 0.4298080503940582,
"learning_rate": 0.00026635653695974146,
"loss": 2.9514,
"step": 82700
},
{
"epoch": 22.26862618432386,
"grad_norm": 0.44201743602752686,
"learning_rate": 0.00026615457115928366,
"loss": 2.956,
"step": 82750
},
{
"epoch": 22.282084409991388,
"grad_norm": 0.43245506286621094,
"learning_rate": 0.00026595260535882586,
"loss": 2.9673,
"step": 82800
},
{
"epoch": 22.295542635658915,
"grad_norm": 0.42273321747779846,
"learning_rate": 0.0002657506395583681,
"loss": 2.9617,
"step": 82850
},
{
"epoch": 22.309000861326442,
"grad_norm": 0.4414304196834564,
"learning_rate": 0.0002655486737579103,
"loss": 2.9648,
"step": 82900
},
{
"epoch": 22.32245908699397,
"grad_norm": 0.468494176864624,
"learning_rate": 0.00026534670795745255,
"loss": 2.967,
"step": 82950
},
{
"epoch": 22.3359173126615,
"grad_norm": 0.44611600041389465,
"learning_rate": 0.00026514474215699475,
"loss": 2.9694,
"step": 83000
},
{
"epoch": 22.3359173126615,
"eval_accuracy": 0.39639348935890967,
"eval_loss": 3.284945487976074,
"eval_runtime": 82.8378,
"eval_samples_per_second": 217.437,
"eval_steps_per_second": 13.593,
"step": 83000
},
{
"epoch": 22.349375538329028,
"grad_norm": 0.432601660490036,
"learning_rate": 0.00026494277635653694,
"loss": 2.9673,
"step": 83050
},
{
"epoch": 22.362833763996555,
"grad_norm": 0.407348096370697,
"learning_rate": 0.00026474081055607914,
"loss": 2.9589,
"step": 83100
},
{
"epoch": 22.376291989664082,
"grad_norm": 0.4226687252521515,
"learning_rate": 0.00026453884475562134,
"loss": 2.9666,
"step": 83150
},
{
"epoch": 22.38975021533161,
"grad_norm": 0.4381425678730011,
"learning_rate": 0.0002643368789551636,
"loss": 2.9649,
"step": 83200
},
{
"epoch": 22.403208440999137,
"grad_norm": 0.4228724539279938,
"learning_rate": 0.0002641349131547058,
"loss": 2.9756,
"step": 83250
},
{
"epoch": 22.416666666666668,
"grad_norm": 0.4300466775894165,
"learning_rate": 0.000263932947354248,
"loss": 2.9689,
"step": 83300
},
{
"epoch": 22.430124892334195,
"grad_norm": 0.4339345693588257,
"learning_rate": 0.0002637309815537902,
"loss": 2.9815,
"step": 83350
},
{
"epoch": 22.443583118001722,
"grad_norm": 0.44847241044044495,
"learning_rate": 0.0002635290157533324,
"loss": 2.974,
"step": 83400
},
{
"epoch": 22.45704134366925,
"grad_norm": 0.43475624918937683,
"learning_rate": 0.0002633270499528746,
"loss": 2.9716,
"step": 83450
},
{
"epoch": 22.470499569336777,
"grad_norm": 0.4242154657840729,
"learning_rate": 0.0002631250841524168,
"loss": 2.9722,
"step": 83500
},
{
"epoch": 22.483957795004308,
"grad_norm": 0.46649372577667236,
"learning_rate": 0.00026292311835195907,
"loss": 2.9873,
"step": 83550
},
{
"epoch": 22.497416020671835,
"grad_norm": 0.4434148371219635,
"learning_rate": 0.00026272115255150126,
"loss": 2.9709,
"step": 83600
},
{
"epoch": 22.510874246339363,
"grad_norm": 0.4270596206188202,
"learning_rate": 0.00026251918675104346,
"loss": 2.9725,
"step": 83650
},
{
"epoch": 22.52433247200689,
"grad_norm": 0.4376223385334015,
"learning_rate": 0.00026231722095058566,
"loss": 2.9651,
"step": 83700
},
{
"epoch": 22.537790697674417,
"grad_norm": 0.4081477224826813,
"learning_rate": 0.00026211525515012785,
"loss": 2.9876,
"step": 83750
},
{
"epoch": 22.551248923341948,
"grad_norm": 0.43901485204696655,
"learning_rate": 0.0002619132893496701,
"loss": 2.979,
"step": 83800
},
{
"epoch": 22.564707149009475,
"grad_norm": 0.4190357029438019,
"learning_rate": 0.0002617113235492123,
"loss": 2.9699,
"step": 83850
},
{
"epoch": 22.578165374677003,
"grad_norm": 0.41881316900253296,
"learning_rate": 0.00026150935774875455,
"loss": 2.985,
"step": 83900
},
{
"epoch": 22.59162360034453,
"grad_norm": 0.4245586395263672,
"learning_rate": 0.00026130739194829675,
"loss": 2.9691,
"step": 83950
},
{
"epoch": 22.605081826012057,
"grad_norm": 0.44662144780158997,
"learning_rate": 0.00026110542614783894,
"loss": 2.9864,
"step": 84000
},
{
"epoch": 22.605081826012057,
"eval_accuracy": 0.3968706663485487,
"eval_loss": 3.277745008468628,
"eval_runtime": 82.6562,
"eval_samples_per_second": 217.915,
"eval_steps_per_second": 13.623,
"step": 84000
},
{
"epoch": 22.618540051679588,
"grad_norm": 0.4233132600784302,
"learning_rate": 0.00026090346034738114,
"loss": 2.9753,
"step": 84050
},
{
"epoch": 22.631998277347115,
"grad_norm": 0.438705176115036,
"learning_rate": 0.0002607014945469234,
"loss": 2.9964,
"step": 84100
},
{
"epoch": 22.645456503014643,
"grad_norm": 0.44637560844421387,
"learning_rate": 0.0002604995287464656,
"loss": 2.9885,
"step": 84150
},
{
"epoch": 22.65891472868217,
"grad_norm": 0.4181239604949951,
"learning_rate": 0.0002602975629460078,
"loss": 2.9766,
"step": 84200
},
{
"epoch": 22.672372954349697,
"grad_norm": 0.4466797411441803,
"learning_rate": 0.00026009559714555,
"loss": 2.987,
"step": 84250
},
{
"epoch": 22.685831180017228,
"grad_norm": 0.4136148989200592,
"learning_rate": 0.0002598936313450922,
"loss": 2.9818,
"step": 84300
},
{
"epoch": 22.699289405684755,
"grad_norm": 0.44519859552383423,
"learning_rate": 0.0002596916655446344,
"loss": 2.996,
"step": 84350
},
{
"epoch": 22.712747631352283,
"grad_norm": 0.44434547424316406,
"learning_rate": 0.0002594896997441766,
"loss": 2.9943,
"step": 84400
},
{
"epoch": 22.72620585701981,
"grad_norm": 0.4297696352005005,
"learning_rate": 0.00025928773394371887,
"loss": 2.9901,
"step": 84450
},
{
"epoch": 22.739664082687337,
"grad_norm": 0.44606852531433105,
"learning_rate": 0.00025908576814326107,
"loss": 2.9926,
"step": 84500
},
{
"epoch": 22.753122308354868,
"grad_norm": 0.46659067273139954,
"learning_rate": 0.00025888380234280326,
"loss": 2.9843,
"step": 84550
},
{
"epoch": 22.766580534022395,
"grad_norm": 0.40941062569618225,
"learning_rate": 0.00025868183654234546,
"loss": 2.9816,
"step": 84600
},
{
"epoch": 22.780038759689923,
"grad_norm": 0.42072591185569763,
"learning_rate": 0.00025847987074188765,
"loss": 2.9775,
"step": 84650
},
{
"epoch": 22.79349698535745,
"grad_norm": 0.41729506850242615,
"learning_rate": 0.0002582779049414299,
"loss": 2.979,
"step": 84700
},
{
"epoch": 22.806955211024977,
"grad_norm": 0.42029449343681335,
"learning_rate": 0.0002580759391409721,
"loss": 2.9977,
"step": 84750
},
{
"epoch": 22.820413436692505,
"grad_norm": 0.43667352199554443,
"learning_rate": 0.00025787397334051435,
"loss": 2.9965,
"step": 84800
},
{
"epoch": 22.833871662360036,
"grad_norm": 0.4144902527332306,
"learning_rate": 0.00025767200754005655,
"loss": 3.0019,
"step": 84850
},
{
"epoch": 22.847329888027563,
"grad_norm": 0.48112985491752625,
"learning_rate": 0.00025747004173959874,
"loss": 2.9991,
"step": 84900
},
{
"epoch": 22.86078811369509,
"grad_norm": 0.42465952038764954,
"learning_rate": 0.00025726807593914094,
"loss": 2.9965,
"step": 84950
},
{
"epoch": 22.874246339362617,
"grad_norm": 0.4174148142337799,
"learning_rate": 0.00025706611013868313,
"loss": 2.9913,
"step": 85000
},
{
"epoch": 22.874246339362617,
"eval_accuracy": 0.3973353489497978,
"eval_loss": 3.2742652893066406,
"eval_runtime": 82.6265,
"eval_samples_per_second": 217.993,
"eval_steps_per_second": 13.628,
"step": 85000
},
{
"epoch": 22.887704565030145,
"grad_norm": 0.4097732603549957,
"learning_rate": 0.0002568641443382254,
"loss": 3.0042,
"step": 85050
},
{
"epoch": 22.901162790697676,
"grad_norm": 0.4230758249759674,
"learning_rate": 0.0002566621785377676,
"loss": 2.9956,
"step": 85100
},
{
"epoch": 22.914621016365203,
"grad_norm": 0.47532251477241516,
"learning_rate": 0.0002564602127373098,
"loss": 2.9993,
"step": 85150
},
{
"epoch": 22.92807924203273,
"grad_norm": 0.42103108763694763,
"learning_rate": 0.00025625824693685203,
"loss": 2.9799,
"step": 85200
},
{
"epoch": 22.941537467700257,
"grad_norm": 0.45972493290901184,
"learning_rate": 0.0002560562811363942,
"loss": 3.0015,
"step": 85250
},
{
"epoch": 22.954995693367785,
"grad_norm": 0.4185284972190857,
"learning_rate": 0.0002558543153359364,
"loss": 2.9986,
"step": 85300
},
{
"epoch": 22.968453919035316,
"grad_norm": 0.4455835521221161,
"learning_rate": 0.0002556523495354786,
"loss": 2.9877,
"step": 85350
},
{
"epoch": 22.981912144702843,
"grad_norm": 0.43334561586380005,
"learning_rate": 0.00025545038373502087,
"loss": 2.9954,
"step": 85400
},
{
"epoch": 22.99537037037037,
"grad_norm": 0.41968095302581787,
"learning_rate": 0.00025524841793456306,
"loss": 3.0005,
"step": 85450
},
{
"epoch": 23.00861326442722,
"grad_norm": 0.45201441645622253,
"learning_rate": 0.00025504645213410526,
"loss": 2.9449,
"step": 85500
},
{
"epoch": 23.022071490094746,
"grad_norm": 0.45828717947006226,
"learning_rate": 0.00025484448633364745,
"loss": 2.9045,
"step": 85550
},
{
"epoch": 23.035529715762273,
"grad_norm": 0.46304261684417725,
"learning_rate": 0.00025464252053318965,
"loss": 2.9076,
"step": 85600
},
{
"epoch": 23.0489879414298,
"grad_norm": 0.4227404296398163,
"learning_rate": 0.0002544405547327319,
"loss": 2.9183,
"step": 85650
},
{
"epoch": 23.06244616709733,
"grad_norm": 0.4482329487800598,
"learning_rate": 0.0002542385889322741,
"loss": 2.9184,
"step": 85700
},
{
"epoch": 23.07590439276486,
"grad_norm": 0.453664094209671,
"learning_rate": 0.00025403662313181635,
"loss": 2.9095,
"step": 85750
},
{
"epoch": 23.089362618432386,
"grad_norm": 0.43255752325057983,
"learning_rate": 0.00025383465733135854,
"loss": 2.9182,
"step": 85800
},
{
"epoch": 23.102820844099913,
"grad_norm": 0.44747617840766907,
"learning_rate": 0.00025363269153090074,
"loss": 2.9322,
"step": 85850
},
{
"epoch": 23.11627906976744,
"grad_norm": 0.46070900559425354,
"learning_rate": 0.00025343072573044294,
"loss": 2.9315,
"step": 85900
},
{
"epoch": 23.12973729543497,
"grad_norm": 0.46046116948127747,
"learning_rate": 0.00025322875992998513,
"loss": 2.9247,
"step": 85950
},
{
"epoch": 23.1431955211025,
"grad_norm": 0.46017971634864807,
"learning_rate": 0.0002530267941295274,
"loss": 2.9319,
"step": 86000
},
{
"epoch": 23.1431955211025,
"eval_accuracy": 0.3965518964743226,
"eval_loss": 3.288102626800537,
"eval_runtime": 82.7004,
"eval_samples_per_second": 217.798,
"eval_steps_per_second": 13.615,
"step": 86000
},
{
"epoch": 23.156653746770026,
"grad_norm": 0.4487834572792053,
"learning_rate": 0.0002528248283290696,
"loss": 2.9384,
"step": 86050
},
{
"epoch": 23.170111972437553,
"grad_norm": 0.4324404001235962,
"learning_rate": 0.00025262286252861183,
"loss": 2.926,
"step": 86100
},
{
"epoch": 23.18357019810508,
"grad_norm": 0.44153496623039246,
"learning_rate": 0.000252420896728154,
"loss": 2.9352,
"step": 86150
},
{
"epoch": 23.19702842377261,
"grad_norm": 0.42381033301353455,
"learning_rate": 0.0002522189309276962,
"loss": 2.9407,
"step": 86200
},
{
"epoch": 23.21048664944014,
"grad_norm": 0.4593563377857208,
"learning_rate": 0.0002520169651272384,
"loss": 2.934,
"step": 86250
},
{
"epoch": 23.223944875107666,
"grad_norm": 0.44144001603126526,
"learning_rate": 0.00025181499932678067,
"loss": 2.9462,
"step": 86300
},
{
"epoch": 23.237403100775193,
"grad_norm": 0.43457597494125366,
"learning_rate": 0.00025161303352632286,
"loss": 2.9448,
"step": 86350
},
{
"epoch": 23.25086132644272,
"grad_norm": 0.4174328148365021,
"learning_rate": 0.00025141106772586506,
"loss": 2.9509,
"step": 86400
},
{
"epoch": 23.26431955211025,
"grad_norm": 0.4494629204273224,
"learning_rate": 0.00025120910192540726,
"loss": 2.9471,
"step": 86450
},
{
"epoch": 23.27777777777778,
"grad_norm": 0.43323251605033875,
"learning_rate": 0.00025100713612494945,
"loss": 2.961,
"step": 86500
},
{
"epoch": 23.291236003445306,
"grad_norm": 0.42934539914131165,
"learning_rate": 0.0002508051703244917,
"loss": 2.9512,
"step": 86550
},
{
"epoch": 23.304694229112833,
"grad_norm": 0.41573357582092285,
"learning_rate": 0.0002506032045240339,
"loss": 2.953,
"step": 86600
},
{
"epoch": 23.31815245478036,
"grad_norm": 0.4408019185066223,
"learning_rate": 0.00025040123872357615,
"loss": 2.9619,
"step": 86650
},
{
"epoch": 23.33161068044789,
"grad_norm": 0.461113303899765,
"learning_rate": 0.00025019927292311834,
"loss": 2.9428,
"step": 86700
},
{
"epoch": 23.34506890611542,
"grad_norm": 0.4499945044517517,
"learning_rate": 0.00024999730712266054,
"loss": 2.9575,
"step": 86750
},
{
"epoch": 23.358527131782946,
"grad_norm": 0.4402410686016083,
"learning_rate": 0.00024979534132220274,
"loss": 2.9602,
"step": 86800
},
{
"epoch": 23.371985357450473,
"grad_norm": 0.44360268115997314,
"learning_rate": 0.00024959337552174493,
"loss": 2.9546,
"step": 86850
},
{
"epoch": 23.385443583118,
"grad_norm": 0.4170439839363098,
"learning_rate": 0.0002493914097212872,
"loss": 2.9595,
"step": 86900
},
{
"epoch": 23.39890180878553,
"grad_norm": 0.4547288715839386,
"learning_rate": 0.0002491894439208294,
"loss": 2.9571,
"step": 86950
},
{
"epoch": 23.41236003445306,
"grad_norm": 0.41132912039756775,
"learning_rate": 0.00024898747812037163,
"loss": 2.952,
"step": 87000
},
{
"epoch": 23.41236003445306,
"eval_accuracy": 0.3970523238910524,
"eval_loss": 3.2812342643737793,
"eval_runtime": 82.7806,
"eval_samples_per_second": 217.587,
"eval_steps_per_second": 13.602,
"step": 87000
},
{
"epoch": 23.425818260120586,
"grad_norm": 0.45564085245132446,
"learning_rate": 0.0002487855123199138,
"loss": 2.9624,
"step": 87050
},
{
"epoch": 23.439276485788113,
"grad_norm": 0.4767475724220276,
"learning_rate": 0.000248583546519456,
"loss": 2.9655,
"step": 87100
},
{
"epoch": 23.45273471145564,
"grad_norm": 0.4390631318092346,
"learning_rate": 0.0002483815807189982,
"loss": 2.9505,
"step": 87150
},
{
"epoch": 23.466192937123168,
"grad_norm": 0.418514221906662,
"learning_rate": 0.0002481796149185404,
"loss": 2.9622,
"step": 87200
},
{
"epoch": 23.4796511627907,
"grad_norm": 0.43617865443229675,
"learning_rate": 0.00024797764911808266,
"loss": 2.9642,
"step": 87250
},
{
"epoch": 23.493109388458226,
"grad_norm": 0.45152929425239563,
"learning_rate": 0.00024777568331762486,
"loss": 2.9621,
"step": 87300
},
{
"epoch": 23.506567614125753,
"grad_norm": 0.4509811997413635,
"learning_rate": 0.00024757371751716706,
"loss": 2.9654,
"step": 87350
},
{
"epoch": 23.52002583979328,
"grad_norm": 0.43976351618766785,
"learning_rate": 0.0002473717517167093,
"loss": 2.9646,
"step": 87400
},
{
"epoch": 23.533484065460808,
"grad_norm": 0.42602604627609253,
"learning_rate": 0.0002471697859162515,
"loss": 2.9636,
"step": 87450
},
{
"epoch": 23.54694229112834,
"grad_norm": 0.44689950346946716,
"learning_rate": 0.0002469678201157937,
"loss": 2.9673,
"step": 87500
},
{
"epoch": 23.560400516795866,
"grad_norm": 0.47383999824523926,
"learning_rate": 0.0002467658543153359,
"loss": 2.9699,
"step": 87550
},
{
"epoch": 23.573858742463393,
"grad_norm": 0.4508107602596283,
"learning_rate": 0.00024656388851487815,
"loss": 2.9796,
"step": 87600
},
{
"epoch": 23.58731696813092,
"grad_norm": 0.4487416744232178,
"learning_rate": 0.00024636192271442034,
"loss": 2.9795,
"step": 87650
},
{
"epoch": 23.600775193798448,
"grad_norm": 0.4085511863231659,
"learning_rate": 0.00024615995691396254,
"loss": 2.9755,
"step": 87700
},
{
"epoch": 23.61423341946598,
"grad_norm": 0.46656128764152527,
"learning_rate": 0.00024595799111350473,
"loss": 2.9774,
"step": 87750
},
{
"epoch": 23.627691645133506,
"grad_norm": 0.46008649468421936,
"learning_rate": 0.00024575602531304693,
"loss": 2.9658,
"step": 87800
},
{
"epoch": 23.641149870801033,
"grad_norm": 0.4458731710910797,
"learning_rate": 0.0002455540595125892,
"loss": 2.9904,
"step": 87850
},
{
"epoch": 23.65460809646856,
"grad_norm": 0.4410950839519501,
"learning_rate": 0.00024535209371213143,
"loss": 2.9755,
"step": 87900
},
{
"epoch": 23.668066322136088,
"grad_norm": 0.42997658252716064,
"learning_rate": 0.0002451501279116736,
"loss": 2.98,
"step": 87950
},
{
"epoch": 23.68152454780362,
"grad_norm": 0.4559027850627899,
"learning_rate": 0.0002449481621112158,
"loss": 2.9787,
"step": 88000
},
{
"epoch": 23.68152454780362,
"eval_accuracy": 0.39748115302996523,
"eval_loss": 3.275696277618408,
"eval_runtime": 82.6679,
"eval_samples_per_second": 217.884,
"eval_steps_per_second": 13.621,
"step": 88000
},
{
"epoch": 23.694982773471146,
"grad_norm": 0.42745450139045715,
"learning_rate": 0.000244746196310758,
"loss": 2.9734,
"step": 88050
},
{
"epoch": 23.708440999138674,
"grad_norm": 0.44881296157836914,
"learning_rate": 0.0002445442305103002,
"loss": 2.9811,
"step": 88100
},
{
"epoch": 23.7218992248062,
"grad_norm": 0.43561914563179016,
"learning_rate": 0.00024434226470984246,
"loss": 2.9786,
"step": 88150
},
{
"epoch": 23.735357450473728,
"grad_norm": 0.4392639994621277,
"learning_rate": 0.00024414029890938463,
"loss": 2.9716,
"step": 88200
},
{
"epoch": 23.74881567614126,
"grad_norm": 0.44505080580711365,
"learning_rate": 0.00024393833310892686,
"loss": 2.9821,
"step": 88250
},
{
"epoch": 23.762273901808786,
"grad_norm": 0.45526963472366333,
"learning_rate": 0.0002437363673084691,
"loss": 2.9898,
"step": 88300
},
{
"epoch": 23.775732127476314,
"grad_norm": 0.422258585691452,
"learning_rate": 0.0002435344015080113,
"loss": 2.9774,
"step": 88350
},
{
"epoch": 23.78919035314384,
"grad_norm": 0.4553403854370117,
"learning_rate": 0.0002433324357075535,
"loss": 2.9826,
"step": 88400
},
{
"epoch": 23.802648578811368,
"grad_norm": 0.4607415795326233,
"learning_rate": 0.00024313046990709572,
"loss": 2.9926,
"step": 88450
},
{
"epoch": 23.8161068044789,
"grad_norm": 0.4590122401714325,
"learning_rate": 0.00024292850410663792,
"loss": 2.9783,
"step": 88500
},
{
"epoch": 23.829565030146426,
"grad_norm": 0.43685176968574524,
"learning_rate": 0.00024272653830618014,
"loss": 2.979,
"step": 88550
},
{
"epoch": 23.843023255813954,
"grad_norm": 0.44387924671173096,
"learning_rate": 0.00024252457250572234,
"loss": 2.9809,
"step": 88600
},
{
"epoch": 23.85648148148148,
"grad_norm": 0.4566636383533478,
"learning_rate": 0.00024232260670526453,
"loss": 2.9842,
"step": 88650
},
{
"epoch": 23.86993970714901,
"grad_norm": 0.4601520299911499,
"learning_rate": 0.00024212064090480676,
"loss": 2.9842,
"step": 88700
},
{
"epoch": 23.88339793281654,
"grad_norm": 0.47891494631767273,
"learning_rate": 0.00024191867510434898,
"loss": 2.9833,
"step": 88750
},
{
"epoch": 23.896856158484066,
"grad_norm": 0.4479354918003082,
"learning_rate": 0.0002417167093038912,
"loss": 2.9829,
"step": 88800
},
{
"epoch": 23.910314384151594,
"grad_norm": 0.4588629901409149,
"learning_rate": 0.0002415147435034334,
"loss": 2.9901,
"step": 88850
},
{
"epoch": 23.92377260981912,
"grad_norm": 0.44897961616516113,
"learning_rate": 0.00024131277770297562,
"loss": 2.9946,
"step": 88900
},
{
"epoch": 23.93723083548665,
"grad_norm": 0.4773879647254944,
"learning_rate": 0.00024111081190251782,
"loss": 2.9888,
"step": 88950
},
{
"epoch": 23.95068906115418,
"grad_norm": 0.4188065826892853,
"learning_rate": 0.00024090884610206002,
"loss": 2.9899,
"step": 89000
},
{
"epoch": 23.95068906115418,
"eval_accuracy": 0.39772517386756295,
"eval_loss": 3.272411823272705,
"eval_runtime": 82.7384,
"eval_samples_per_second": 217.698,
"eval_steps_per_second": 13.609,
"step": 89000
},
{
"epoch": 23.964147286821706,
"grad_norm": 0.4670105576515198,
"learning_rate": 0.00024070688030160224,
"loss": 2.9861,
"step": 89050
},
{
"epoch": 23.977605512489234,
"grad_norm": 0.4304860532283783,
"learning_rate": 0.00024050491450114443,
"loss": 2.9893,
"step": 89100
},
{
"epoch": 23.99106373815676,
"grad_norm": 0.4615049362182617,
"learning_rate": 0.00024030294870068666,
"loss": 2.9832,
"step": 89150
},
{
"epoch": 24.00430663221361,
"grad_norm": 0.4380539357662201,
"learning_rate": 0.00024010098290022888,
"loss": 2.9569,
"step": 89200
},
{
"epoch": 24.017764857881136,
"grad_norm": 0.44901078939437866,
"learning_rate": 0.0002398990170997711,
"loss": 2.9137,
"step": 89250
},
{
"epoch": 24.031223083548664,
"grad_norm": 0.4563383460044861,
"learning_rate": 0.0002396970512993133,
"loss": 2.9082,
"step": 89300
},
{
"epoch": 24.044681309216195,
"grad_norm": 0.43932202458381653,
"learning_rate": 0.00023949508549885552,
"loss": 2.9013,
"step": 89350
},
{
"epoch": 24.058139534883722,
"grad_norm": 0.4604346752166748,
"learning_rate": 0.00023929311969839772,
"loss": 2.9157,
"step": 89400
},
{
"epoch": 24.07159776055125,
"grad_norm": 0.4781767725944519,
"learning_rate": 0.00023909115389793992,
"loss": 2.9113,
"step": 89450
},
{
"epoch": 24.085055986218777,
"grad_norm": 0.4381226897239685,
"learning_rate": 0.00023888918809748214,
"loss": 2.9135,
"step": 89500
},
{
"epoch": 24.098514211886304,
"grad_norm": 0.45084530115127563,
"learning_rate": 0.00023868722229702433,
"loss": 2.915,
"step": 89550
},
{
"epoch": 24.11197243755383,
"grad_norm": 0.43952709436416626,
"learning_rate": 0.00023848525649656656,
"loss": 2.9171,
"step": 89600
},
{
"epoch": 24.125430663221362,
"grad_norm": 0.46405529975891113,
"learning_rate": 0.00023828329069610878,
"loss": 2.9256,
"step": 89650
},
{
"epoch": 24.13888888888889,
"grad_norm": 0.4647674858570099,
"learning_rate": 0.000238081324895651,
"loss": 2.9221,
"step": 89700
},
{
"epoch": 24.152347114556417,
"grad_norm": 0.47676974534988403,
"learning_rate": 0.0002378793590951932,
"loss": 2.9244,
"step": 89750
},
{
"epoch": 24.165805340223944,
"grad_norm": 0.45089593529701233,
"learning_rate": 0.0002376773932947354,
"loss": 2.9174,
"step": 89800
},
{
"epoch": 24.17926356589147,
"grad_norm": 0.45035508275032043,
"learning_rate": 0.00023747542749427762,
"loss": 2.9313,
"step": 89850
},
{
"epoch": 24.192721791559002,
"grad_norm": 0.43069884181022644,
"learning_rate": 0.00023727346169381982,
"loss": 2.9317,
"step": 89900
},
{
"epoch": 24.20618001722653,
"grad_norm": 0.49231401085853577,
"learning_rate": 0.00023707149589336204,
"loss": 2.9325,
"step": 89950
},
{
"epoch": 24.219638242894057,
"grad_norm": 0.4412052631378174,
"learning_rate": 0.00023686953009290424,
"loss": 2.9305,
"step": 90000
},
{
"epoch": 24.219638242894057,
"eval_accuracy": 0.3968511099145471,
"eval_loss": 3.285769462585449,
"eval_runtime": 82.6514,
"eval_samples_per_second": 217.927,
"eval_steps_per_second": 13.623,
"step": 90000
},
{
"epoch": 24.233096468561584,
"grad_norm": 0.45433464646339417,
"learning_rate": 0.00023666756429244643,
"loss": 2.9363,
"step": 90050
},
{
"epoch": 24.24655469422911,
"grad_norm": 0.42987197637557983,
"learning_rate": 0.00023646559849198868,
"loss": 2.9434,
"step": 90100
},
{
"epoch": 24.260012919896642,
"grad_norm": 0.45912572741508484,
"learning_rate": 0.00023626363269153088,
"loss": 2.9382,
"step": 90150
},
{
"epoch": 24.27347114556417,
"grad_norm": 0.4451676607131958,
"learning_rate": 0.0002360616668910731,
"loss": 2.928,
"step": 90200
},
{
"epoch": 24.286929371231697,
"grad_norm": 0.42687520384788513,
"learning_rate": 0.0002358597010906153,
"loss": 2.9474,
"step": 90250
},
{
"epoch": 24.300387596899224,
"grad_norm": 0.4559120833873749,
"learning_rate": 0.00023565773529015752,
"loss": 2.9339,
"step": 90300
},
{
"epoch": 24.31384582256675,
"grad_norm": 0.465943306684494,
"learning_rate": 0.00023545576948969972,
"loss": 2.9422,
"step": 90350
},
{
"epoch": 24.327304048234282,
"grad_norm": 0.426479697227478,
"learning_rate": 0.00023525380368924194,
"loss": 2.9517,
"step": 90400
},
{
"epoch": 24.34076227390181,
"grad_norm": 0.4488561451435089,
"learning_rate": 0.00023505183788878414,
"loss": 2.9545,
"step": 90450
},
{
"epoch": 24.354220499569337,
"grad_norm": 0.4385835528373718,
"learning_rate": 0.00023484987208832633,
"loss": 2.9563,
"step": 90500
},
{
"epoch": 24.367678725236864,
"grad_norm": 0.4604963958263397,
"learning_rate": 0.00023464790628786858,
"loss": 2.947,
"step": 90550
},
{
"epoch": 24.38113695090439,
"grad_norm": 0.4424059987068176,
"learning_rate": 0.00023444594048741078,
"loss": 2.9477,
"step": 90600
},
{
"epoch": 24.394595176571922,
"grad_norm": 0.4392193853855133,
"learning_rate": 0.000234243974686953,
"loss": 2.9602,
"step": 90650
},
{
"epoch": 24.40805340223945,
"grad_norm": 0.428774893283844,
"learning_rate": 0.0002340420088864952,
"loss": 2.9426,
"step": 90700
},
{
"epoch": 24.421511627906977,
"grad_norm": 0.4254078269004822,
"learning_rate": 0.00023384004308603742,
"loss": 2.9465,
"step": 90750
},
{
"epoch": 24.434969853574504,
"grad_norm": 0.4557410776615143,
"learning_rate": 0.00023363807728557962,
"loss": 2.9546,
"step": 90800
},
{
"epoch": 24.44842807924203,
"grad_norm": 0.48022007942199707,
"learning_rate": 0.0002334361114851218,
"loss": 2.9565,
"step": 90850
},
{
"epoch": 24.461886304909562,
"grad_norm": 0.43262091279029846,
"learning_rate": 0.00023323414568466404,
"loss": 2.9482,
"step": 90900
},
{
"epoch": 24.47534453057709,
"grad_norm": 0.4592428505420685,
"learning_rate": 0.00023303217988420623,
"loss": 2.9533,
"step": 90950
},
{
"epoch": 24.488802756244617,
"grad_norm": 0.44655975699424744,
"learning_rate": 0.00023283021408374848,
"loss": 2.9429,
"step": 91000
},
{
"epoch": 24.488802756244617,
"eval_accuracy": 0.3976547707051572,
"eval_loss": 3.2799808979034424,
"eval_runtime": 82.6791,
"eval_samples_per_second": 217.854,
"eval_steps_per_second": 13.619,
"step": 91000
},
{
"epoch": 24.502260981912144,
"grad_norm": 0.4476975202560425,
"learning_rate": 0.00023262824828329068,
"loss": 2.958,
"step": 91050
},
{
"epoch": 24.51571920757967,
"grad_norm": 0.45505842566490173,
"learning_rate": 0.0002324262824828329,
"loss": 2.965,
"step": 91100
},
{
"epoch": 24.529177433247202,
"grad_norm": 0.4326203465461731,
"learning_rate": 0.0002322243166823751,
"loss": 2.9637,
"step": 91150
},
{
"epoch": 24.54263565891473,
"grad_norm": 0.4207751750946045,
"learning_rate": 0.0002320223508819173,
"loss": 2.9586,
"step": 91200
},
{
"epoch": 24.556093884582257,
"grad_norm": 0.42772597074508667,
"learning_rate": 0.00023182038508145952,
"loss": 2.9473,
"step": 91250
},
{
"epoch": 24.569552110249784,
"grad_norm": 0.4410114586353302,
"learning_rate": 0.0002316184192810017,
"loss": 2.9567,
"step": 91300
},
{
"epoch": 24.58301033591731,
"grad_norm": 0.45146074891090393,
"learning_rate": 0.00023141645348054394,
"loss": 2.9611,
"step": 91350
},
{
"epoch": 24.596468561584842,
"grad_norm": 0.4386666715145111,
"learning_rate": 0.00023121448768008613,
"loss": 2.9561,
"step": 91400
},
{
"epoch": 24.60992678725237,
"grad_norm": 0.4680837094783783,
"learning_rate": 0.00023101252187962838,
"loss": 2.9581,
"step": 91450
},
{
"epoch": 24.623385012919897,
"grad_norm": 0.4437674880027771,
"learning_rate": 0.00023081055607917058,
"loss": 2.9641,
"step": 91500
},
{
"epoch": 24.636843238587424,
"grad_norm": 0.4646398425102234,
"learning_rate": 0.0002306085902787128,
"loss": 2.9592,
"step": 91550
},
{
"epoch": 24.65030146425495,
"grad_norm": 0.4523963928222656,
"learning_rate": 0.000230406624478255,
"loss": 2.9613,
"step": 91600
},
{
"epoch": 24.66375968992248,
"grad_norm": 0.45717960596084595,
"learning_rate": 0.0002302046586777972,
"loss": 2.9687,
"step": 91650
},
{
"epoch": 24.67721791559001,
"grad_norm": 0.44318801164627075,
"learning_rate": 0.00023000269287733942,
"loss": 2.9701,
"step": 91700
},
{
"epoch": 24.690676141257537,
"grad_norm": 0.46825453639030457,
"learning_rate": 0.00022980072707688161,
"loss": 2.9664,
"step": 91750
},
{
"epoch": 24.704134366925064,
"grad_norm": 0.46073204278945923,
"learning_rate": 0.00022959876127642384,
"loss": 2.963,
"step": 91800
},
{
"epoch": 24.71759259259259,
"grad_norm": 0.4572125971317291,
"learning_rate": 0.00022939679547596603,
"loss": 2.9601,
"step": 91850
},
{
"epoch": 24.73105081826012,
"grad_norm": 0.49562516808509827,
"learning_rate": 0.00022919482967550828,
"loss": 2.9684,
"step": 91900
},
{
"epoch": 24.74450904392765,
"grad_norm": 0.447395920753479,
"learning_rate": 0.00022899286387505048,
"loss": 2.9656,
"step": 91950
},
{
"epoch": 24.757967269595177,
"grad_norm": 0.4500346779823303,
"learning_rate": 0.00022879089807459268,
"loss": 2.975,
"step": 92000
},
{
"epoch": 24.757967269595177,
"eval_accuracy": 0.39792139008871236,
"eval_loss": 3.273996114730835,
"eval_runtime": 82.8261,
"eval_samples_per_second": 217.468,
"eval_steps_per_second": 13.595,
"step": 92000
},
{
"epoch": 24.771425495262704,
"grad_norm": 0.4537627696990967,
"learning_rate": 0.0002285889322741349,
"loss": 2.9661,
"step": 92050
},
{
"epoch": 24.78488372093023,
"grad_norm": 0.475948691368103,
"learning_rate": 0.0002283869664736771,
"loss": 2.9804,
"step": 92100
},
{
"epoch": 24.79834194659776,
"grad_norm": 0.4436039924621582,
"learning_rate": 0.00022818500067321932,
"loss": 2.9692,
"step": 92150
},
{
"epoch": 24.81180017226529,
"grad_norm": 0.4599752426147461,
"learning_rate": 0.00022798303487276151,
"loss": 2.9689,
"step": 92200
},
{
"epoch": 24.825258397932817,
"grad_norm": 0.45375698804855347,
"learning_rate": 0.0002277810690723037,
"loss": 2.9691,
"step": 92250
},
{
"epoch": 24.838716623600344,
"grad_norm": 0.5722225308418274,
"learning_rate": 0.00022757910327184596,
"loss": 2.9689,
"step": 92300
},
{
"epoch": 24.852174849267872,
"grad_norm": 0.4496856927871704,
"learning_rate": 0.00022737713747138818,
"loss": 2.9685,
"step": 92350
},
{
"epoch": 24.8656330749354,
"grad_norm": 0.4757240414619446,
"learning_rate": 0.00022717517167093038,
"loss": 2.9729,
"step": 92400
},
{
"epoch": 24.87909130060293,
"grad_norm": 0.45742520689964294,
"learning_rate": 0.00022697320587047258,
"loss": 2.9725,
"step": 92450
},
{
"epoch": 24.892549526270457,
"grad_norm": 0.45312777161598206,
"learning_rate": 0.0002267712400700148,
"loss": 2.9722,
"step": 92500
},
{
"epoch": 24.906007751937985,
"grad_norm": 0.4436255395412445,
"learning_rate": 0.000226569274269557,
"loss": 2.9842,
"step": 92550
},
{
"epoch": 24.919465977605512,
"grad_norm": 0.4487346410751343,
"learning_rate": 0.00022636730846909922,
"loss": 2.9722,
"step": 92600
},
{
"epoch": 24.93292420327304,
"grad_norm": 0.4670848250389099,
"learning_rate": 0.00022616534266864141,
"loss": 2.9774,
"step": 92650
},
{
"epoch": 24.94638242894057,
"grad_norm": 0.4423047602176666,
"learning_rate": 0.0002259633768681836,
"loss": 2.9676,
"step": 92700
},
{
"epoch": 24.959840654608097,
"grad_norm": 0.47778424620628357,
"learning_rate": 0.00022576141106772586,
"loss": 2.9872,
"step": 92750
},
{
"epoch": 24.973298880275625,
"grad_norm": 0.46354302763938904,
"learning_rate": 0.00022555944526726806,
"loss": 2.9793,
"step": 92800
},
{
"epoch": 24.986757105943152,
"grad_norm": 0.4492092430591583,
"learning_rate": 0.00022535747946681028,
"loss": 2.9776,
"step": 92850
},
{
"epoch": 25.0,
"grad_norm": 0.9362195730209351,
"learning_rate": 0.00022515551366635248,
"loss": 2.9797,
"step": 92900
},
{
"epoch": 25.013458225667527,
"grad_norm": 0.45706066489219666,
"learning_rate": 0.0002249535478658947,
"loss": 2.8972,
"step": 92950
},
{
"epoch": 25.026916451335055,
"grad_norm": 0.4540911018848419,
"learning_rate": 0.0002247515820654369,
"loss": 2.8901,
"step": 93000
},
{
"epoch": 25.026916451335055,
"eval_accuracy": 0.39698463690003577,
"eval_loss": 3.286383628845215,
"eval_runtime": 82.8226,
"eval_samples_per_second": 217.477,
"eval_steps_per_second": 13.595,
"step": 93000
},
{
"epoch": 25.040374677002585,
"grad_norm": 0.43824413418769836,
"learning_rate": 0.0002245496162649791,
"loss": 2.8921,
"step": 93050
},
{
"epoch": 25.053832902670113,
"grad_norm": 0.4552362263202667,
"learning_rate": 0.00022434765046452131,
"loss": 2.9038,
"step": 93100
},
{
"epoch": 25.06729112833764,
"grad_norm": 0.4600333571434021,
"learning_rate": 0.0002241456846640635,
"loss": 2.9022,
"step": 93150
},
{
"epoch": 25.080749354005167,
"grad_norm": 0.45745813846588135,
"learning_rate": 0.00022394371886360576,
"loss": 2.8946,
"step": 93200
},
{
"epoch": 25.094207579672695,
"grad_norm": 0.4775872230529785,
"learning_rate": 0.00022374175306314796,
"loss": 2.9063,
"step": 93250
},
{
"epoch": 25.107665805340226,
"grad_norm": 0.9106134176254272,
"learning_rate": 0.00022353978726269018,
"loss": 2.9124,
"step": 93300
},
{
"epoch": 25.121124031007753,
"grad_norm": 0.48400700092315674,
"learning_rate": 0.00022333782146223238,
"loss": 2.9129,
"step": 93350
},
{
"epoch": 25.13458225667528,
"grad_norm": 0.4530729353427887,
"learning_rate": 0.0002231358556617746,
"loss": 2.9046,
"step": 93400
},
{
"epoch": 25.148040482342807,
"grad_norm": 0.4515216648578644,
"learning_rate": 0.0002229338898613168,
"loss": 2.908,
"step": 93450
},
{
"epoch": 25.161498708010335,
"grad_norm": 0.4834233820438385,
"learning_rate": 0.000222731924060859,
"loss": 2.9184,
"step": 93500
},
{
"epoch": 25.174956933677866,
"grad_norm": 0.4750334322452545,
"learning_rate": 0.00022252995826040122,
"loss": 2.9181,
"step": 93550
},
{
"epoch": 25.188415159345393,
"grad_norm": 0.452582985162735,
"learning_rate": 0.0002223279924599434,
"loss": 2.9235,
"step": 93600
},
{
"epoch": 25.20187338501292,
"grad_norm": 0.4878632128238678,
"learning_rate": 0.00022212602665948566,
"loss": 2.9299,
"step": 93650
},
{
"epoch": 25.215331610680447,
"grad_norm": 0.4488566815853119,
"learning_rate": 0.00022192406085902786,
"loss": 2.9295,
"step": 93700
},
{
"epoch": 25.228789836347975,
"grad_norm": 0.45161136984825134,
"learning_rate": 0.00022172209505857008,
"loss": 2.9245,
"step": 93750
},
{
"epoch": 25.242248062015506,
"grad_norm": 0.43922969698905945,
"learning_rate": 0.00022152012925811228,
"loss": 2.9277,
"step": 93800
},
{
"epoch": 25.255706287683033,
"grad_norm": 0.46489647030830383,
"learning_rate": 0.00022131816345765447,
"loss": 2.925,
"step": 93850
},
{
"epoch": 25.26916451335056,
"grad_norm": 0.44657576084136963,
"learning_rate": 0.0002211161976571967,
"loss": 2.9238,
"step": 93900
},
{
"epoch": 25.282622739018088,
"grad_norm": 0.46452194452285767,
"learning_rate": 0.0002209142318567389,
"loss": 2.9308,
"step": 93950
},
{
"epoch": 25.296080964685615,
"grad_norm": 0.45336994528770447,
"learning_rate": 0.00022071226605628112,
"loss": 2.9296,
"step": 94000
},
{
"epoch": 25.296080964685615,
"eval_accuracy": 0.39716346962429483,
"eval_loss": 3.2845382690429688,
"eval_runtime": 82.6694,
"eval_samples_per_second": 217.88,
"eval_steps_per_second": 13.621,
"step": 94000
},
{
"epoch": 25.309539190353142,
"grad_norm": 0.46927860379219055,
"learning_rate": 0.0002205103002558233,
"loss": 2.9301,
"step": 94050
},
{
"epoch": 25.322997416020673,
"grad_norm": 0.4543541669845581,
"learning_rate": 0.00022030833445536556,
"loss": 2.9262,
"step": 94100
},
{
"epoch": 25.3364556416882,
"grad_norm": 0.455096572637558,
"learning_rate": 0.00022010636865490776,
"loss": 2.9265,
"step": 94150
},
{
"epoch": 25.349913867355728,
"grad_norm": 0.48218822479248047,
"learning_rate": 0.00021990440285444998,
"loss": 2.9406,
"step": 94200
},
{
"epoch": 25.363372093023255,
"grad_norm": 0.4825000762939453,
"learning_rate": 0.00021970243705399218,
"loss": 2.9392,
"step": 94250
},
{
"epoch": 25.376830318690782,
"grad_norm": 0.45533639192581177,
"learning_rate": 0.00021950047125353437,
"loss": 2.9465,
"step": 94300
},
{
"epoch": 25.390288544358313,
"grad_norm": 0.44987037777900696,
"learning_rate": 0.0002192985054530766,
"loss": 2.9416,
"step": 94350
},
{
"epoch": 25.40374677002584,
"grad_norm": 0.46171754598617554,
"learning_rate": 0.0002190965396526188,
"loss": 2.9325,
"step": 94400
},
{
"epoch": 25.417204995693368,
"grad_norm": 0.4504666328430176,
"learning_rate": 0.00021889457385216102,
"loss": 2.9331,
"step": 94450
},
{
"epoch": 25.430663221360895,
"grad_norm": 0.4376462697982788,
"learning_rate": 0.0002186926080517032,
"loss": 2.9435,
"step": 94500
},
{
"epoch": 25.444121447028422,
"grad_norm": 0.4670478105545044,
"learning_rate": 0.00021849064225124546,
"loss": 2.933,
"step": 94550
},
{
"epoch": 25.457579672695953,
"grad_norm": 0.4690640866756439,
"learning_rate": 0.00021828867645078766,
"loss": 2.9533,
"step": 94600
},
{
"epoch": 25.47103789836348,
"grad_norm": 0.44130662083625793,
"learning_rate": 0.00021808671065032985,
"loss": 2.9538,
"step": 94650
},
{
"epoch": 25.484496124031008,
"grad_norm": 0.4500773549079895,
"learning_rate": 0.00021788474484987208,
"loss": 2.9391,
"step": 94700
},
{
"epoch": 25.497954349698535,
"grad_norm": 0.4685859978199005,
"learning_rate": 0.00021768277904941427,
"loss": 2.9436,
"step": 94750
},
{
"epoch": 25.511412575366062,
"grad_norm": 0.4648078680038452,
"learning_rate": 0.0002174808132489565,
"loss": 2.9398,
"step": 94800
},
{
"epoch": 25.524870801033593,
"grad_norm": 0.4338512420654297,
"learning_rate": 0.0002172788474484987,
"loss": 2.9461,
"step": 94850
},
{
"epoch": 25.53832902670112,
"grad_norm": 0.47696825861930847,
"learning_rate": 0.0002170768816480409,
"loss": 2.9545,
"step": 94900
},
{
"epoch": 25.551787252368648,
"grad_norm": 0.45933452248573303,
"learning_rate": 0.0002168749158475831,
"loss": 2.9519,
"step": 94950
},
{
"epoch": 25.565245478036175,
"grad_norm": 0.4370841383934021,
"learning_rate": 0.00021667295004712534,
"loss": 2.9454,
"step": 95000
},
{
"epoch": 25.565245478036175,
"eval_accuracy": 0.397293628557261,
"eval_loss": 3.2789294719696045,
"eval_runtime": 82.6934,
"eval_samples_per_second": 217.817,
"eval_steps_per_second": 13.617,
"step": 95000
},
{
"epoch": 25.578703703703702,
"grad_norm": 0.4755338728427887,
"learning_rate": 0.00021647098424666756,
"loss": 2.9466,
"step": 95050
},
{
"epoch": 25.592161929371233,
"grad_norm": 0.4773982763290405,
"learning_rate": 0.00021626901844620976,
"loss": 2.959,
"step": 95100
},
{
"epoch": 25.60562015503876,
"grad_norm": 0.4488808810710907,
"learning_rate": 0.00021606705264575198,
"loss": 2.9503,
"step": 95150
},
{
"epoch": 25.619078380706288,
"grad_norm": 0.4498634934425354,
"learning_rate": 0.00021586508684529417,
"loss": 2.9471,
"step": 95200
},
{
"epoch": 25.632536606373815,
"grad_norm": 0.457423597574234,
"learning_rate": 0.0002156631210448364,
"loss": 2.9389,
"step": 95250
},
{
"epoch": 25.645994832041342,
"grad_norm": 0.4621497690677643,
"learning_rate": 0.0002154611552443786,
"loss": 2.9492,
"step": 95300
},
{
"epoch": 25.659453057708873,
"grad_norm": 0.49652400612831116,
"learning_rate": 0.0002152591894439208,
"loss": 2.9494,
"step": 95350
},
{
"epoch": 25.6729112833764,
"grad_norm": 0.47319862246513367,
"learning_rate": 0.000215057223643463,
"loss": 2.951,
"step": 95400
},
{
"epoch": 25.686369509043928,
"grad_norm": 0.44089022278785706,
"learning_rate": 0.00021485525784300524,
"loss": 2.956,
"step": 95450
},
{
"epoch": 25.699827734711455,
"grad_norm": 0.48533281683921814,
"learning_rate": 0.00021465329204254746,
"loss": 2.9537,
"step": 95500
},
{
"epoch": 25.713285960378983,
"grad_norm": 0.5042552351951599,
"learning_rate": 0.00021445132624208966,
"loss": 2.9599,
"step": 95550
},
{
"epoch": 25.72674418604651,
"grad_norm": 0.45634913444519043,
"learning_rate": 0.00021424936044163188,
"loss": 2.9622,
"step": 95600
},
{
"epoch": 25.74020241171404,
"grad_norm": 0.5739097595214844,
"learning_rate": 0.00021404739464117407,
"loss": 2.9636,
"step": 95650
},
{
"epoch": 25.753660637381568,
"grad_norm": 0.47575172781944275,
"learning_rate": 0.00021384542884071627,
"loss": 2.9617,
"step": 95700
},
{
"epoch": 25.767118863049095,
"grad_norm": 0.43450653553009033,
"learning_rate": 0.0002136434630402585,
"loss": 2.9622,
"step": 95750
},
{
"epoch": 25.780577088716623,
"grad_norm": 0.4463081359863281,
"learning_rate": 0.0002134414972398007,
"loss": 2.961,
"step": 95800
},
{
"epoch": 25.794035314384153,
"grad_norm": 0.455655574798584,
"learning_rate": 0.0002132395314393429,
"loss": 2.9591,
"step": 95850
},
{
"epoch": 25.80749354005168,
"grad_norm": 0.46448880434036255,
"learning_rate": 0.00021303756563888514,
"loss": 2.9601,
"step": 95900
},
{
"epoch": 25.820951765719208,
"grad_norm": 0.4840283989906311,
"learning_rate": 0.00021283559983842736,
"loss": 2.9706,
"step": 95950
},
{
"epoch": 25.834409991386735,
"grad_norm": 0.4429430365562439,
"learning_rate": 0.00021263363403796956,
"loss": 2.9691,
"step": 96000
},
{
"epoch": 25.834409991386735,
"eval_accuracy": 0.3979101994625892,
"eval_loss": 3.274695634841919,
"eval_runtime": 82.8766,
"eval_samples_per_second": 217.335,
"eval_steps_per_second": 13.586,
"step": 96000
},
{
"epoch": 25.847868217054263,
"grad_norm": 0.4571455121040344,
"learning_rate": 0.00021243166823751175,
"loss": 2.966,
"step": 96050
},
{
"epoch": 25.86132644272179,
"grad_norm": 0.47416019439697266,
"learning_rate": 0.00021222970243705398,
"loss": 2.9526,
"step": 96100
},
{
"epoch": 25.87478466838932,
"grad_norm": 0.47117966413497925,
"learning_rate": 0.00021202773663659617,
"loss": 2.9595,
"step": 96150
},
{
"epoch": 25.888242894056848,
"grad_norm": 1.3879706859588623,
"learning_rate": 0.0002118257708361384,
"loss": 2.9662,
"step": 96200
},
{
"epoch": 25.901701119724375,
"grad_norm": 0.46109259128570557,
"learning_rate": 0.0002116238050356806,
"loss": 2.9716,
"step": 96250
},
{
"epoch": 25.915159345391903,
"grad_norm": 0.47239986062049866,
"learning_rate": 0.00021142183923522281,
"loss": 2.9754,
"step": 96300
},
{
"epoch": 25.92861757105943,
"grad_norm": 0.4863017499446869,
"learning_rate": 0.00021121987343476504,
"loss": 2.9681,
"step": 96350
},
{
"epoch": 25.94207579672696,
"grad_norm": 0.529517650604248,
"learning_rate": 0.00021101790763430726,
"loss": 2.9665,
"step": 96400
},
{
"epoch": 25.955534022394488,
"grad_norm": 0.43884140253067017,
"learning_rate": 0.00021081594183384946,
"loss": 2.9727,
"step": 96450
},
{
"epoch": 25.968992248062015,
"grad_norm": 0.45245248079299927,
"learning_rate": 0.00021061397603339165,
"loss": 2.9666,
"step": 96500
},
{
"epoch": 25.982450473729543,
"grad_norm": 0.490133136510849,
"learning_rate": 0.00021041201023293388,
"loss": 2.9653,
"step": 96550
},
{
"epoch": 25.99590869939707,
"grad_norm": 0.4798160493373871,
"learning_rate": 0.00021021004443247607,
"loss": 2.9614,
"step": 96600
},
{
"epoch": 26.009151593453918,
"grad_norm": 0.4801061153411865,
"learning_rate": 0.0002100080786320183,
"loss": 2.9261,
"step": 96650
},
{
"epoch": 26.022609819121445,
"grad_norm": 0.45648154616355896,
"learning_rate": 0.0002098061128315605,
"loss": 2.8788,
"step": 96700
},
{
"epoch": 26.036068044788976,
"grad_norm": 0.46171003580093384,
"learning_rate": 0.00020960414703110274,
"loss": 2.889,
"step": 96750
},
{
"epoch": 26.049526270456504,
"grad_norm": 0.4955553412437439,
"learning_rate": 0.00020940218123064494,
"loss": 2.8797,
"step": 96800
},
{
"epoch": 26.06298449612403,
"grad_norm": 0.47606784105300903,
"learning_rate": 0.00020920021543018713,
"loss": 2.9118,
"step": 96850
},
{
"epoch": 26.076442721791558,
"grad_norm": 0.46941566467285156,
"learning_rate": 0.00020899824962972936,
"loss": 2.8898,
"step": 96900
},
{
"epoch": 26.089900947459086,
"grad_norm": 0.4896952509880066,
"learning_rate": 0.00020879628382927155,
"loss": 2.8976,
"step": 96950
},
{
"epoch": 26.103359173126616,
"grad_norm": 0.47314926981925964,
"learning_rate": 0.00020859431802881378,
"loss": 2.9068,
"step": 97000
},
{
"epoch": 26.103359173126616,
"eval_accuracy": 0.3969564973644446,
"eval_loss": 3.286790609359741,
"eval_runtime": 83.0019,
"eval_samples_per_second": 217.007,
"eval_steps_per_second": 13.566,
"step": 97000
},
{
"epoch": 26.116817398794144,
"grad_norm": 0.4595525860786438,
"learning_rate": 0.00020839235222835597,
"loss": 2.9027,
"step": 97050
},
{
"epoch": 26.13027562446167,
"grad_norm": 0.45305392146110535,
"learning_rate": 0.00020819038642789817,
"loss": 2.9026,
"step": 97100
},
{
"epoch": 26.1437338501292,
"grad_norm": 0.4702953100204468,
"learning_rate": 0.0002079884206274404,
"loss": 2.907,
"step": 97150
},
{
"epoch": 26.157192075796726,
"grad_norm": 0.47354647517204285,
"learning_rate": 0.00020778645482698264,
"loss": 2.9104,
"step": 97200
},
{
"epoch": 26.170650301464256,
"grad_norm": 0.48177456855773926,
"learning_rate": 0.00020758448902652484,
"loss": 2.9084,
"step": 97250
},
{
"epoch": 26.184108527131784,
"grad_norm": 0.47766661643981934,
"learning_rate": 0.00020738252322606703,
"loss": 2.9046,
"step": 97300
},
{
"epoch": 26.19756675279931,
"grad_norm": 0.4453447461128235,
"learning_rate": 0.00020718055742560926,
"loss": 2.9093,
"step": 97350
},
{
"epoch": 26.21102497846684,
"grad_norm": 0.4752526581287384,
"learning_rate": 0.00020697859162515145,
"loss": 2.9208,
"step": 97400
},
{
"epoch": 26.224483204134366,
"grad_norm": 0.45564496517181396,
"learning_rate": 0.00020677662582469368,
"loss": 2.9177,
"step": 97450
},
{
"epoch": 26.237941429801896,
"grad_norm": 0.48327958583831787,
"learning_rate": 0.00020657466002423587,
"loss": 2.9131,
"step": 97500
},
{
"epoch": 26.251399655469424,
"grad_norm": 0.4772711396217346,
"learning_rate": 0.00020637269422377807,
"loss": 2.9171,
"step": 97550
},
{
"epoch": 26.26485788113695,
"grad_norm": 0.44969451427459717,
"learning_rate": 0.0002061707284233203,
"loss": 2.922,
"step": 97600
},
{
"epoch": 26.27831610680448,
"grad_norm": 0.4639694094657898,
"learning_rate": 0.00020596876262286251,
"loss": 2.9169,
"step": 97650
},
{
"epoch": 26.291774332472006,
"grad_norm": 0.4834367632865906,
"learning_rate": 0.00020576679682240474,
"loss": 2.9194,
"step": 97700
},
{
"epoch": 26.305232558139537,
"grad_norm": 0.4687032699584961,
"learning_rate": 0.00020556483102194693,
"loss": 2.9196,
"step": 97750
},
{
"epoch": 26.318690783807064,
"grad_norm": 0.4695006012916565,
"learning_rate": 0.00020536286522148916,
"loss": 2.9218,
"step": 97800
},
{
"epoch": 26.33214900947459,
"grad_norm": 0.5058684349060059,
"learning_rate": 0.00020516089942103135,
"loss": 2.9245,
"step": 97850
},
{
"epoch": 26.34560723514212,
"grad_norm": 0.4534665048122406,
"learning_rate": 0.00020495893362057355,
"loss": 2.9305,
"step": 97900
},
{
"epoch": 26.359065460809646,
"grad_norm": 0.4632345139980316,
"learning_rate": 0.00020475696782011577,
"loss": 2.9335,
"step": 97950
},
{
"epoch": 26.372523686477177,
"grad_norm": 0.5274003744125366,
"learning_rate": 0.00020455500201965797,
"loss": 2.9336,
"step": 98000
},
{
"epoch": 26.372523686477177,
"eval_accuracy": 0.39755014378324866,
"eval_loss": 3.2820188999176025,
"eval_runtime": 82.7608,
"eval_samples_per_second": 217.639,
"eval_steps_per_second": 13.605,
"step": 98000
},
{
"epoch": 26.372523686477177,
"step": 98000,
"total_flos": 2.048094740938752e+18,
"train_loss": 3.1790810820910393,
"train_runtime": 65510.0983,
"train_samples_per_second": 181.478,
"train_steps_per_second": 2.269
}
],
"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": 12
}
},
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 2.048094740938752e+18,
"train_batch_size": 16,
"trial_name": null,
"trial_params": null
}
{
"best_global_step": 78000,
"best_metric": 3.2710349559783936,
"best_model_checkpoint": "/scratch/cl5625/exceptions/models/high_2000_634/checkpoint-20000",
"epoch": 26.372523686477177,
"eval_steps": 1000,
"global_step": 98000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.013458225667527994,
"grad_norm": 1.26742684841156,
"learning_rate": 0.000294,
"loss": 8.4484,
"step": 50
},
{
"epoch": 0.026916451335055987,
"grad_norm": 1.158722996711731,
"learning_rate": 0.0005939999999999999,
"loss": 6.7858,
"step": 100
},
{
"epoch": 0.04037467700258398,
"grad_norm": 1.5116316080093384,
"learning_rate": 0.0005998020735155513,
"loss": 6.2921,
"step": 150
},
{
"epoch": 0.053832902670111975,
"grad_norm": 0.9892672300338745,
"learning_rate": 0.0005996001077150935,
"loss": 6.0248,
"step": 200
},
{
"epoch": 0.06729112833763996,
"grad_norm": 1.1211016178131104,
"learning_rate": 0.0005993981419146358,
"loss": 5.8647,
"step": 250
},
{
"epoch": 0.08074935400516796,
"grad_norm": 1.6260143518447876,
"learning_rate": 0.0005991961761141779,
"loss": 5.7301,
"step": 300
},
{
"epoch": 0.09420757967269595,
"grad_norm": 1.0242689847946167,
"learning_rate": 0.0005989942103137202,
"loss": 5.6218,
"step": 350
},
{
"epoch": 0.10766580534022395,
"grad_norm": 1.1451252698898315,
"learning_rate": 0.0005987922445132624,
"loss": 5.4895,
"step": 400
},
{
"epoch": 0.12112403100775193,
"grad_norm": 1.1215616464614868,
"learning_rate": 0.0005985902787128047,
"loss": 5.3563,
"step": 450
},
{
"epoch": 0.13458225667527993,
"grad_norm": 1.6647250652313232,
"learning_rate": 0.0005983883129123468,
"loss": 5.253,
"step": 500
},
{
"epoch": 0.14804048234280792,
"grad_norm": 1.2564589977264404,
"learning_rate": 0.000598186347111889,
"loss": 5.1438,
"step": 550
},
{
"epoch": 0.16149870801033592,
"grad_norm": 1.2288016080856323,
"learning_rate": 0.0005979843813114312,
"loss": 5.0377,
"step": 600
},
{
"epoch": 0.1749569336778639,
"grad_norm": 1.4052221775054932,
"learning_rate": 0.0005977824155109734,
"loss": 4.964,