distilbert_lda_v1_book / trainer_state.json
gokulsrinivasagan's picture
End of training
ee850a1 verified
raw
history blame
224 kB
{
"best_metric": 3.1041817665100098,
"best_model_checkpoint": "distilbert_lda_v1_book/checkpoint-580000",
"epoch": 25.0,
"eval_steps": 10000,
"global_step": 593075,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.021076592336551025,
"grad_norm": 2.107733726501465,
"learning_rate": 5e-06,
"loss": 11.3124,
"step": 500
},
{
"epoch": 0.04215318467310205,
"grad_norm": 1.1342570781707764,
"learning_rate": 1e-05,
"loss": 9.0068,
"step": 1000
},
{
"epoch": 0.06322977700965308,
"grad_norm": 0.893055260181427,
"learning_rate": 1.5e-05,
"loss": 8.3785,
"step": 1500
},
{
"epoch": 0.0843063693462041,
"grad_norm": 1.155956506729126,
"learning_rate": 2e-05,
"loss": 8.2415,
"step": 2000
},
{
"epoch": 0.10538296168275513,
"grad_norm": 1.0789344310760498,
"learning_rate": 2.5e-05,
"loss": 8.1725,
"step": 2500
},
{
"epoch": 0.12645955401930617,
"grad_norm": 0.7853618860244751,
"learning_rate": 3e-05,
"loss": 8.1287,
"step": 3000
},
{
"epoch": 0.14753614635585718,
"grad_norm": 0.9027785658836365,
"learning_rate": 3.5e-05,
"loss": 8.0988,
"step": 3500
},
{
"epoch": 0.1686127386924082,
"grad_norm": 1.1292823553085327,
"learning_rate": 4e-05,
"loss": 8.0716,
"step": 4000
},
{
"epoch": 0.18968933102895924,
"grad_norm": 0.9900161027908325,
"learning_rate": 4.5e-05,
"loss": 8.0463,
"step": 4500
},
{
"epoch": 0.21076592336551025,
"grad_norm": 0.9558222889900208,
"learning_rate": 5e-05,
"loss": 8.0297,
"step": 5000
},
{
"epoch": 0.2318425157020613,
"grad_norm": 0.8645071983337402,
"learning_rate": 5.500000000000001e-05,
"loss": 8.019,
"step": 5500
},
{
"epoch": 0.25291910803861234,
"grad_norm": 1.006055474281311,
"learning_rate": 6e-05,
"loss": 8.0,
"step": 6000
},
{
"epoch": 0.27399570037516335,
"grad_norm": 1.3248836994171143,
"learning_rate": 6.500000000000001e-05,
"loss": 7.9933,
"step": 6500
},
{
"epoch": 0.29507229271171437,
"grad_norm": 1.1509044170379639,
"learning_rate": 7e-05,
"loss": 7.9734,
"step": 7000
},
{
"epoch": 0.3161488850482654,
"grad_norm": 1.0638937950134277,
"learning_rate": 7.500000000000001e-05,
"loss": 7.9447,
"step": 7500
},
{
"epoch": 0.3372254773848164,
"grad_norm": 1.1935096979141235,
"learning_rate": 8e-05,
"loss": 7.9122,
"step": 8000
},
{
"epoch": 0.35830206972136747,
"grad_norm": 1.4608166217803955,
"learning_rate": 8.5e-05,
"loss": 7.8288,
"step": 8500
},
{
"epoch": 0.3793786620579185,
"grad_norm": 1.3763691186904907,
"learning_rate": 9e-05,
"loss": 7.7164,
"step": 9000
},
{
"epoch": 0.4004552543944695,
"grad_norm": 1.2726778984069824,
"learning_rate": 9.5e-05,
"loss": 7.6801,
"step": 9500
},
{
"epoch": 0.4215318467310205,
"grad_norm": 1.1782846450805664,
"learning_rate": 0.0001,
"loss": 7.6644,
"step": 10000
},
{
"epoch": 0.4215318467310205,
"eval_accuracy": 0.16089565495321737,
"eval_loss": 7.487222194671631,
"eval_runtime": 178.5862,
"eval_samples_per_second": 675.898,
"eval_steps_per_second": 7.044,
"step": 10000
},
{
"epoch": 0.4426084390675716,
"grad_norm": 1.1251558065414429,
"learning_rate": 9.99142477382841e-05,
"loss": 7.6494,
"step": 10500
},
{
"epoch": 0.4636850314041226,
"grad_norm": 1.000657558441162,
"learning_rate": 9.98284954765682e-05,
"loss": 7.6406,
"step": 11000
},
{
"epoch": 0.4847616237406736,
"grad_norm": 1.188158392906189,
"learning_rate": 9.97427432148523e-05,
"loss": 7.6293,
"step": 11500
},
{
"epoch": 0.5058382160772247,
"grad_norm": 1.1732149124145508,
"learning_rate": 9.96569909531364e-05,
"loss": 7.5828,
"step": 12000
},
{
"epoch": 0.5269148084137757,
"grad_norm": 1.894943118095398,
"learning_rate": 9.957123869142048e-05,
"loss": 7.4725,
"step": 12500
},
{
"epoch": 0.5479914007503267,
"grad_norm": 2.070441722869873,
"learning_rate": 9.94854864297046e-05,
"loss": 6.8515,
"step": 13000
},
{
"epoch": 0.5690679930868777,
"grad_norm": 1.6918480396270752,
"learning_rate": 9.939973416798869e-05,
"loss": 6.223,
"step": 13500
},
{
"epoch": 0.5901445854234287,
"grad_norm": 1.4557104110717773,
"learning_rate": 9.931398190627278e-05,
"loss": 5.7572,
"step": 14000
},
{
"epoch": 0.6112211777599798,
"grad_norm": 1.472951889038086,
"learning_rate": 9.922822964455688e-05,
"loss": 5.4906,
"step": 14500
},
{
"epoch": 0.6322977700965308,
"grad_norm": 1.4528071880340576,
"learning_rate": 9.914247738284099e-05,
"loss": 5.3312,
"step": 15000
},
{
"epoch": 0.6533743624330818,
"grad_norm": 1.452592372894287,
"learning_rate": 9.905672512112507e-05,
"loss": 5.2129,
"step": 15500
},
{
"epoch": 0.6744509547696328,
"grad_norm": 1.3783841133117676,
"learning_rate": 9.897097285940917e-05,
"loss": 5.1247,
"step": 16000
},
{
"epoch": 0.6955275471061839,
"grad_norm": 1.3407448530197144,
"learning_rate": 9.888522059769327e-05,
"loss": 5.0431,
"step": 16500
},
{
"epoch": 0.7166041394427349,
"grad_norm": 1.7458844184875488,
"learning_rate": 9.879946833597737e-05,
"loss": 4.967,
"step": 17000
},
{
"epoch": 0.737680731779286,
"grad_norm": 1.3584396839141846,
"learning_rate": 9.871371607426147e-05,
"loss": 4.9018,
"step": 17500
},
{
"epoch": 0.758757324115837,
"grad_norm": 1.366771936416626,
"learning_rate": 9.862796381254556e-05,
"loss": 4.8515,
"step": 18000
},
{
"epoch": 0.779833916452388,
"grad_norm": 1.3086378574371338,
"learning_rate": 9.854221155082965e-05,
"loss": 4.7911,
"step": 18500
},
{
"epoch": 0.800910508788939,
"grad_norm": 1.2679400444030762,
"learning_rate": 9.845645928911376e-05,
"loss": 4.7468,
"step": 19000
},
{
"epoch": 0.82198710112549,
"grad_norm": 1.37302565574646,
"learning_rate": 9.837070702739786e-05,
"loss": 4.701,
"step": 19500
},
{
"epoch": 0.843063693462041,
"grad_norm": 1.305145263671875,
"learning_rate": 9.828495476568194e-05,
"loss": 4.6569,
"step": 20000
},
{
"epoch": 0.843063693462041,
"eval_accuracy": 0.5608786407559289,
"eval_loss": 4.298525333404541,
"eval_runtime": 163.3147,
"eval_samples_per_second": 739.101,
"eval_steps_per_second": 7.703,
"step": 20000
},
{
"epoch": 0.864140285798592,
"grad_norm": 1.5340673923492432,
"learning_rate": 9.819920250396604e-05,
"loss": 4.6167,
"step": 20500
},
{
"epoch": 0.8852168781351432,
"grad_norm": 1.2236896753311157,
"learning_rate": 9.811345024225015e-05,
"loss": 4.5843,
"step": 21000
},
{
"epoch": 0.9062934704716942,
"grad_norm": 1.1396021842956543,
"learning_rate": 9.802769798053424e-05,
"loss": 4.5588,
"step": 21500
},
{
"epoch": 0.9273700628082452,
"grad_norm": 1.4041922092437744,
"learning_rate": 9.794194571881834e-05,
"loss": 4.5198,
"step": 22000
},
{
"epoch": 0.9484466551447962,
"grad_norm": 1.2980437278747559,
"learning_rate": 9.785619345710244e-05,
"loss": 4.4955,
"step": 22500
},
{
"epoch": 0.9695232474813472,
"grad_norm": 1.3988012075424194,
"learning_rate": 9.777044119538653e-05,
"loss": 4.4693,
"step": 23000
},
{
"epoch": 0.9905998398178982,
"grad_norm": 1.220626711845398,
"learning_rate": 9.768468893367063e-05,
"loss": 4.448,
"step": 23500
},
{
"epoch": 1.0116764321544494,
"grad_norm": 1.3394639492034912,
"learning_rate": 9.759893667195473e-05,
"loss": 4.428,
"step": 24000
},
{
"epoch": 1.0327530244910004,
"grad_norm": 1.4303207397460938,
"learning_rate": 9.751318441023883e-05,
"loss": 4.4089,
"step": 24500
},
{
"epoch": 1.0538296168275514,
"grad_norm": 1.2835074663162231,
"learning_rate": 9.742743214852293e-05,
"loss": 4.3906,
"step": 25000
},
{
"epoch": 1.0749062091641024,
"grad_norm": 1.300036072731018,
"learning_rate": 9.734167988680703e-05,
"loss": 4.3704,
"step": 25500
},
{
"epoch": 1.0959828015006534,
"grad_norm": 1.3186469078063965,
"learning_rate": 9.725592762509111e-05,
"loss": 4.3511,
"step": 26000
},
{
"epoch": 1.1170593938372044,
"grad_norm": 1.3646124601364136,
"learning_rate": 9.717017536337522e-05,
"loss": 4.335,
"step": 26500
},
{
"epoch": 1.1381359861737554,
"grad_norm": 1.3514286279678345,
"learning_rate": 9.708442310165932e-05,
"loss": 4.327,
"step": 27000
},
{
"epoch": 1.1592125785103065,
"grad_norm": 1.4459589719772339,
"learning_rate": 9.69986708399434e-05,
"loss": 4.3088,
"step": 27500
},
{
"epoch": 1.1802891708468575,
"grad_norm": 1.2839070558547974,
"learning_rate": 9.69129185782275e-05,
"loss": 4.2999,
"step": 28000
},
{
"epoch": 1.2013657631834085,
"grad_norm": 1.1478468179702759,
"learning_rate": 9.68271663165116e-05,
"loss": 4.2814,
"step": 28500
},
{
"epoch": 1.2224423555199595,
"grad_norm": 1.2773290872573853,
"learning_rate": 9.67414140547957e-05,
"loss": 4.2687,
"step": 29000
},
{
"epoch": 1.2435189478565105,
"grad_norm": 1.2651296854019165,
"learning_rate": 9.66556617930798e-05,
"loss": 4.2494,
"step": 29500
},
{
"epoch": 1.2645955401930615,
"grad_norm": 1.3133939504623413,
"learning_rate": 9.656990953136388e-05,
"loss": 4.2453,
"step": 30000
},
{
"epoch": 1.2645955401930615,
"eval_accuracy": 0.6093566271880783,
"eval_loss": 3.9352123737335205,
"eval_runtime": 163.9191,
"eval_samples_per_second": 736.375,
"eval_steps_per_second": 7.675,
"step": 30000
},
{
"epoch": 1.2856721325296125,
"grad_norm": 1.3148829936981201,
"learning_rate": 9.6484157269648e-05,
"loss": 4.2275,
"step": 30500
},
{
"epoch": 1.3067487248661636,
"grad_norm": 1.1539913415908813,
"learning_rate": 9.639840500793209e-05,
"loss": 4.2245,
"step": 31000
},
{
"epoch": 1.3278253172027146,
"grad_norm": 1.2084167003631592,
"learning_rate": 9.631265274621618e-05,
"loss": 4.2101,
"step": 31500
},
{
"epoch": 1.3489019095392658,
"grad_norm": 1.2116655111312866,
"learning_rate": 9.622690048450028e-05,
"loss": 4.2046,
"step": 32000
},
{
"epoch": 1.3699785018758166,
"grad_norm": 1.3656361103057861,
"learning_rate": 9.614114822278439e-05,
"loss": 4.1932,
"step": 32500
},
{
"epoch": 1.3910550942123678,
"grad_norm": 1.48634934425354,
"learning_rate": 9.605539596106847e-05,
"loss": 4.1868,
"step": 33000
},
{
"epoch": 1.4121316865489189,
"grad_norm": 1.2326738834381104,
"learning_rate": 9.596964369935257e-05,
"loss": 4.1732,
"step": 33500
},
{
"epoch": 1.4332082788854699,
"grad_norm": 1.2274714708328247,
"learning_rate": 9.588389143763667e-05,
"loss": 4.1679,
"step": 34000
},
{
"epoch": 1.4542848712220209,
"grad_norm": 1.335710883140564,
"learning_rate": 9.579813917592077e-05,
"loss": 4.1557,
"step": 34500
},
{
"epoch": 1.475361463558572,
"grad_norm": 1.19460928440094,
"learning_rate": 9.571238691420487e-05,
"loss": 4.1504,
"step": 35000
},
{
"epoch": 1.496438055895123,
"grad_norm": 1.2459183931350708,
"learning_rate": 9.562663465248896e-05,
"loss": 4.1397,
"step": 35500
},
{
"epoch": 1.517514648231674,
"grad_norm": 1.213924765586853,
"learning_rate": 9.554088239077306e-05,
"loss": 4.1259,
"step": 36000
},
{
"epoch": 1.538591240568225,
"grad_norm": 1.2519876956939697,
"learning_rate": 9.545513012905716e-05,
"loss": 4.1217,
"step": 36500
},
{
"epoch": 1.559667832904776,
"grad_norm": 1.2918117046356201,
"learning_rate": 9.536937786734126e-05,
"loss": 4.1214,
"step": 37000
},
{
"epoch": 1.580744425241327,
"grad_norm": 1.1987943649291992,
"learning_rate": 9.528362560562534e-05,
"loss": 4.1083,
"step": 37500
},
{
"epoch": 1.601821017577878,
"grad_norm": 1.256113052368164,
"learning_rate": 9.519787334390946e-05,
"loss": 4.1006,
"step": 38000
},
{
"epoch": 1.622897609914429,
"grad_norm": 1.2943336963653564,
"learning_rate": 9.511212108219355e-05,
"loss": 4.091,
"step": 38500
},
{
"epoch": 1.64397420225098,
"grad_norm": 1.2057653665542603,
"learning_rate": 9.502636882047764e-05,
"loss": 4.0837,
"step": 39000
},
{
"epoch": 1.6650507945875312,
"grad_norm": 1.322171688079834,
"learning_rate": 9.494061655876174e-05,
"loss": 4.0766,
"step": 39500
},
{
"epoch": 1.686127386924082,
"grad_norm": 1.2400922775268555,
"learning_rate": 9.485486429704585e-05,
"loss": 4.0748,
"step": 40000
},
{
"epoch": 1.686127386924082,
"eval_accuracy": 0.6315438785704416,
"eval_loss": 3.7739579677581787,
"eval_runtime": 162.2899,
"eval_samples_per_second": 743.768,
"eval_steps_per_second": 7.752,
"step": 40000
},
{
"epoch": 1.7072039792606333,
"grad_norm": 1.2719427347183228,
"learning_rate": 9.476911203532993e-05,
"loss": 4.0623,
"step": 40500
},
{
"epoch": 1.728280571597184,
"grad_norm": 1.2506862878799438,
"learning_rate": 9.468335977361403e-05,
"loss": 4.0602,
"step": 41000
},
{
"epoch": 1.7493571639337353,
"grad_norm": 1.2794588804244995,
"learning_rate": 9.459760751189813e-05,
"loss": 4.0521,
"step": 41500
},
{
"epoch": 1.770433756270286,
"grad_norm": 1.3543503284454346,
"learning_rate": 9.451185525018223e-05,
"loss": 4.0446,
"step": 42000
},
{
"epoch": 1.7915103486068373,
"grad_norm": 1.2771344184875488,
"learning_rate": 9.442610298846633e-05,
"loss": 4.0368,
"step": 42500
},
{
"epoch": 1.8125869409433881,
"grad_norm": 1.3094180822372437,
"learning_rate": 9.434035072675043e-05,
"loss": 4.0274,
"step": 43000
},
{
"epoch": 1.8336635332799394,
"grad_norm": 1.214279055595398,
"learning_rate": 9.425459846503451e-05,
"loss": 4.0177,
"step": 43500
},
{
"epoch": 1.8547401256164904,
"grad_norm": 1.1497358083724976,
"learning_rate": 9.416884620331862e-05,
"loss": 4.0117,
"step": 44000
},
{
"epoch": 1.8758167179530414,
"grad_norm": 1.2871726751327515,
"learning_rate": 9.408309394160272e-05,
"loss": 4.0074,
"step": 44500
},
{
"epoch": 1.8968933102895924,
"grad_norm": 1.1654359102249146,
"learning_rate": 9.39973416798868e-05,
"loss": 3.9974,
"step": 45000
},
{
"epoch": 1.9179699026261434,
"grad_norm": 1.275567650794983,
"learning_rate": 9.39115894181709e-05,
"loss": 3.993,
"step": 45500
},
{
"epoch": 1.9390464949626944,
"grad_norm": 1.1679468154907227,
"learning_rate": 9.382583715645502e-05,
"loss": 3.9842,
"step": 46000
},
{
"epoch": 1.9601230872992454,
"grad_norm": 1.3072179555892944,
"learning_rate": 9.37400848947391e-05,
"loss": 3.9777,
"step": 46500
},
{
"epoch": 1.9811996796357965,
"grad_norm": 1.298688292503357,
"learning_rate": 9.36543326330232e-05,
"loss": 3.9784,
"step": 47000
},
{
"epoch": 2.0022762719723475,
"grad_norm": 1.3788747787475586,
"learning_rate": 9.35685803713073e-05,
"loss": 3.9762,
"step": 47500
},
{
"epoch": 2.0233528643088987,
"grad_norm": 1.2336324453353882,
"learning_rate": 9.34828281095914e-05,
"loss": 3.9634,
"step": 48000
},
{
"epoch": 2.0444294566454495,
"grad_norm": 1.2191146612167358,
"learning_rate": 9.339707584787549e-05,
"loss": 3.9581,
"step": 48500
},
{
"epoch": 2.0655060489820007,
"grad_norm": 1.2671812772750854,
"learning_rate": 9.331132358615959e-05,
"loss": 3.9507,
"step": 49000
},
{
"epoch": 2.0865826413185515,
"grad_norm": 1.2426018714904785,
"learning_rate": 9.322557132444369e-05,
"loss": 3.9453,
"step": 49500
},
{
"epoch": 2.1076592336551028,
"grad_norm": 1.2382694482803345,
"learning_rate": 9.313981906272779e-05,
"loss": 3.9464,
"step": 50000
},
{
"epoch": 2.1076592336551028,
"eval_accuracy": 0.6452668995104948,
"eval_loss": 3.6555023193359375,
"eval_runtime": 158.2162,
"eval_samples_per_second": 762.918,
"eval_steps_per_second": 7.951,
"step": 50000
},
{
"epoch": 2.1287358259916536,
"grad_norm": 1.3298990726470947,
"learning_rate": 9.305406680101189e-05,
"loss": 3.9444,
"step": 50500
},
{
"epoch": 2.149812418328205,
"grad_norm": 1.1494505405426025,
"learning_rate": 9.296831453929597e-05,
"loss": 3.9376,
"step": 51000
},
{
"epoch": 2.1708890106647556,
"grad_norm": 1.1897977590560913,
"learning_rate": 9.288256227758008e-05,
"loss": 3.9294,
"step": 51500
},
{
"epoch": 2.191965603001307,
"grad_norm": 1.1522610187530518,
"learning_rate": 9.279681001586418e-05,
"loss": 3.9283,
"step": 52000
},
{
"epoch": 2.2130421953378576,
"grad_norm": 1.252293348312378,
"learning_rate": 9.271105775414827e-05,
"loss": 3.9211,
"step": 52500
},
{
"epoch": 2.234118787674409,
"grad_norm": 1.1200158596038818,
"learning_rate": 9.262530549243236e-05,
"loss": 3.9232,
"step": 53000
},
{
"epoch": 2.2551953800109596,
"grad_norm": 1.2841140031814575,
"learning_rate": 9.253955323071646e-05,
"loss": 3.9169,
"step": 53500
},
{
"epoch": 2.276271972347511,
"grad_norm": 1.3341553211212158,
"learning_rate": 9.245380096900056e-05,
"loss": 3.9182,
"step": 54000
},
{
"epoch": 2.2973485646840617,
"grad_norm": 1.2523657083511353,
"learning_rate": 9.236804870728466e-05,
"loss": 3.906,
"step": 54500
},
{
"epoch": 2.318425157020613,
"grad_norm": 1.2127001285552979,
"learning_rate": 9.228229644556876e-05,
"loss": 3.9066,
"step": 55000
},
{
"epoch": 2.339501749357164,
"grad_norm": 1.3288698196411133,
"learning_rate": 9.219654418385286e-05,
"loss": 3.8972,
"step": 55500
},
{
"epoch": 2.360578341693715,
"grad_norm": 1.37288498878479,
"learning_rate": 9.211079192213695e-05,
"loss": 3.9039,
"step": 56000
},
{
"epoch": 2.381654934030266,
"grad_norm": 1.3524497747421265,
"learning_rate": 9.202503966042105e-05,
"loss": 3.8965,
"step": 56500
},
{
"epoch": 2.402731526366817,
"grad_norm": 1.1850680112838745,
"learning_rate": 9.193928739870514e-05,
"loss": 3.8891,
"step": 57000
},
{
"epoch": 2.423808118703368,
"grad_norm": 1.2855035066604614,
"learning_rate": 9.185353513698925e-05,
"loss": 3.8866,
"step": 57500
},
{
"epoch": 2.444884711039919,
"grad_norm": 1.1468712091445923,
"learning_rate": 9.176778287527335e-05,
"loss": 3.8831,
"step": 58000
},
{
"epoch": 2.4659613033764702,
"grad_norm": 1.179479718208313,
"learning_rate": 9.168203061355743e-05,
"loss": 3.8826,
"step": 58500
},
{
"epoch": 2.487037895713021,
"grad_norm": 1.21729576587677,
"learning_rate": 9.159627835184153e-05,
"loss": 3.8786,
"step": 59000
},
{
"epoch": 2.5081144880495723,
"grad_norm": 1.3547087907791138,
"learning_rate": 9.151052609012564e-05,
"loss": 3.8773,
"step": 59500
},
{
"epoch": 2.529191080386123,
"grad_norm": 1.1954435110092163,
"learning_rate": 9.142477382840973e-05,
"loss": 3.8728,
"step": 60000
},
{
"epoch": 2.529191080386123,
"eval_accuracy": 0.6566415687607815,
"eval_loss": 3.58086895942688,
"eval_runtime": 159.421,
"eval_samples_per_second": 757.152,
"eval_steps_per_second": 7.891,
"step": 60000
},
{
"epoch": 2.5502676727226743,
"grad_norm": 1.2969269752502441,
"learning_rate": 9.133902156669383e-05,
"loss": 3.8669,
"step": 60500
},
{
"epoch": 2.571344265059225,
"grad_norm": 1.2322068214416504,
"learning_rate": 9.125326930497792e-05,
"loss": 3.8667,
"step": 61000
},
{
"epoch": 2.5924208573957763,
"grad_norm": 1.2509596347808838,
"learning_rate": 9.116751704326202e-05,
"loss": 3.8605,
"step": 61500
},
{
"epoch": 2.613497449732327,
"grad_norm": 1.3609365224838257,
"learning_rate": 9.108176478154612e-05,
"loss": 3.8562,
"step": 62000
},
{
"epoch": 2.6345740420688784,
"grad_norm": 1.2502727508544922,
"learning_rate": 9.09960125198302e-05,
"loss": 3.861,
"step": 62500
},
{
"epoch": 2.655650634405429,
"grad_norm": 1.1428087949752808,
"learning_rate": 9.091026025811432e-05,
"loss": 3.8535,
"step": 63000
},
{
"epoch": 2.6767272267419804,
"grad_norm": 1.3560972213745117,
"learning_rate": 9.082450799639842e-05,
"loss": 3.853,
"step": 63500
},
{
"epoch": 2.6978038190785316,
"grad_norm": 1.1236178874969482,
"learning_rate": 9.07387557346825e-05,
"loss": 3.8507,
"step": 64000
},
{
"epoch": 2.7188804114150824,
"grad_norm": 1.2019386291503906,
"learning_rate": 9.06530034729666e-05,
"loss": 3.8453,
"step": 64500
},
{
"epoch": 2.739957003751633,
"grad_norm": 1.153360366821289,
"learning_rate": 9.05672512112507e-05,
"loss": 3.8422,
"step": 65000
},
{
"epoch": 2.7610335960881844,
"grad_norm": 1.3679053783416748,
"learning_rate": 9.04814989495348e-05,
"loss": 3.8405,
"step": 65500
},
{
"epoch": 2.7821101884247357,
"grad_norm": 1.250867486000061,
"learning_rate": 9.03957466878189e-05,
"loss": 3.8404,
"step": 66000
},
{
"epoch": 2.8031867807612865,
"grad_norm": 1.2153122425079346,
"learning_rate": 9.030999442610299e-05,
"loss": 3.8356,
"step": 66500
},
{
"epoch": 2.8242633730978377,
"grad_norm": 1.2121467590332031,
"learning_rate": 9.022424216438709e-05,
"loss": 3.8297,
"step": 67000
},
{
"epoch": 2.8453399654343885,
"grad_norm": 1.4479998350143433,
"learning_rate": 9.013848990267119e-05,
"loss": 3.8263,
"step": 67500
},
{
"epoch": 2.8664165577709397,
"grad_norm": 1.2643030881881714,
"learning_rate": 9.005273764095529e-05,
"loss": 3.8239,
"step": 68000
},
{
"epoch": 2.8874931501074905,
"grad_norm": 1.163683533668518,
"learning_rate": 8.996698537923937e-05,
"loss": 3.8277,
"step": 68500
},
{
"epoch": 2.9085697424440418,
"grad_norm": 1.1674327850341797,
"learning_rate": 8.988123311752348e-05,
"loss": 3.8178,
"step": 69000
},
{
"epoch": 2.9296463347805926,
"grad_norm": 1.1838338375091553,
"learning_rate": 8.979548085580758e-05,
"loss": 3.819,
"step": 69500
},
{
"epoch": 2.950722927117144,
"grad_norm": 1.238379955291748,
"learning_rate": 8.970972859409167e-05,
"loss": 3.814,
"step": 70000
},
{
"epoch": 2.950722927117144,
"eval_accuracy": 0.6635372992403469,
"eval_loss": 3.5363569259643555,
"eval_runtime": 158.8369,
"eval_samples_per_second": 759.937,
"eval_steps_per_second": 7.92,
"step": 70000
},
{
"epoch": 2.9717995194536946,
"grad_norm": 1.2264717817306519,
"learning_rate": 8.962397633237576e-05,
"loss": 3.8126,
"step": 70500
},
{
"epoch": 2.992876111790246,
"grad_norm": 1.1951265335083008,
"learning_rate": 8.953822407065988e-05,
"loss": 3.8097,
"step": 71000
},
{
"epoch": 3.0139527041267966,
"grad_norm": 1.1838797330856323,
"learning_rate": 8.945247180894396e-05,
"loss": 3.8064,
"step": 71500
},
{
"epoch": 3.035029296463348,
"grad_norm": 1.2217252254486084,
"learning_rate": 8.936671954722806e-05,
"loss": 3.8038,
"step": 72000
},
{
"epoch": 3.0561058887998986,
"grad_norm": 1.2377454042434692,
"learning_rate": 8.928096728551216e-05,
"loss": 3.8042,
"step": 72500
},
{
"epoch": 3.07718248113645,
"grad_norm": 1.3134173154830933,
"learning_rate": 8.919521502379626e-05,
"loss": 3.7967,
"step": 73000
},
{
"epoch": 3.0982590734730007,
"grad_norm": 1.2436655759811401,
"learning_rate": 8.910946276208035e-05,
"loss": 3.7993,
"step": 73500
},
{
"epoch": 3.119335665809552,
"grad_norm": 1.1703335046768188,
"learning_rate": 8.902371050036445e-05,
"loss": 3.7908,
"step": 74000
},
{
"epoch": 3.140412258146103,
"grad_norm": 1.1417837142944336,
"learning_rate": 8.893795823864855e-05,
"loss": 3.7954,
"step": 74500
},
{
"epoch": 3.161488850482654,
"grad_norm": 1.2112427949905396,
"learning_rate": 8.885220597693265e-05,
"loss": 3.7918,
"step": 75000
},
{
"epoch": 3.182565442819205,
"grad_norm": 1.2339280843734741,
"learning_rate": 8.876645371521675e-05,
"loss": 3.7859,
"step": 75500
},
{
"epoch": 3.203642035155756,
"grad_norm": 1.201372742652893,
"learning_rate": 8.868070145350083e-05,
"loss": 3.7879,
"step": 76000
},
{
"epoch": 3.224718627492307,
"grad_norm": 1.3188670873641968,
"learning_rate": 8.859494919178493e-05,
"loss": 3.7866,
"step": 76500
},
{
"epoch": 3.245795219828858,
"grad_norm": 1.194527268409729,
"learning_rate": 8.850919693006904e-05,
"loss": 3.78,
"step": 77000
},
{
"epoch": 3.2668718121654092,
"grad_norm": 1.1953904628753662,
"learning_rate": 8.842344466835313e-05,
"loss": 3.7775,
"step": 77500
},
{
"epoch": 3.28794840450196,
"grad_norm": 1.2014292478561401,
"learning_rate": 8.833769240663723e-05,
"loss": 3.7755,
"step": 78000
},
{
"epoch": 3.3090249968385113,
"grad_norm": 1.2356904745101929,
"learning_rate": 8.825194014492132e-05,
"loss": 3.7751,
"step": 78500
},
{
"epoch": 3.330101589175062,
"grad_norm": 1.3132128715515137,
"learning_rate": 8.816618788320542e-05,
"loss": 3.7743,
"step": 79000
},
{
"epoch": 3.3511781815116133,
"grad_norm": 1.2548147439956665,
"learning_rate": 8.808043562148952e-05,
"loss": 3.7731,
"step": 79500
},
{
"epoch": 3.372254773848164,
"grad_norm": 1.226163625717163,
"learning_rate": 8.799468335977362e-05,
"loss": 3.771,
"step": 80000
},
{
"epoch": 3.372254773848164,
"eval_accuracy": 0.6699893855551783,
"eval_loss": 3.4921743869781494,
"eval_runtime": 158.932,
"eval_samples_per_second": 759.482,
"eval_steps_per_second": 7.915,
"step": 80000
},
{
"epoch": 3.3933313661847153,
"grad_norm": 1.2400171756744385,
"learning_rate": 8.790893109805772e-05,
"loss": 3.771,
"step": 80500
},
{
"epoch": 3.414407958521266,
"grad_norm": 1.215498447418213,
"learning_rate": 8.782317883634182e-05,
"loss": 3.767,
"step": 81000
},
{
"epoch": 3.4354845508578173,
"grad_norm": 1.1667732000350952,
"learning_rate": 8.773742657462591e-05,
"loss": 3.7634,
"step": 81500
},
{
"epoch": 3.456561143194368,
"grad_norm": 1.2694814205169678,
"learning_rate": 8.765167431291e-05,
"loss": 3.7609,
"step": 82000
},
{
"epoch": 3.4776377355309194,
"grad_norm": 1.2780115604400635,
"learning_rate": 8.756592205119411e-05,
"loss": 3.7595,
"step": 82500
},
{
"epoch": 3.4987143278674706,
"grad_norm": 1.2303638458251953,
"learning_rate": 8.748016978947821e-05,
"loss": 3.7605,
"step": 83000
},
{
"epoch": 3.5197909202040214,
"grad_norm": 1.2029612064361572,
"learning_rate": 8.73944175277623e-05,
"loss": 3.7592,
"step": 83500
},
{
"epoch": 3.540867512540572,
"grad_norm": 1.2176700830459595,
"learning_rate": 8.730866526604639e-05,
"loss": 3.7598,
"step": 84000
},
{
"epoch": 3.5619441048771234,
"grad_norm": 1.3242027759552002,
"learning_rate": 8.72229130043305e-05,
"loss": 3.7565,
"step": 84500
},
{
"epoch": 3.5830206972136747,
"grad_norm": 1.2809795141220093,
"learning_rate": 8.713716074261459e-05,
"loss": 3.7529,
"step": 85000
},
{
"epoch": 3.6040972895502255,
"grad_norm": 1.2092963457107544,
"learning_rate": 8.705140848089869e-05,
"loss": 3.7537,
"step": 85500
},
{
"epoch": 3.6251738818867767,
"grad_norm": 1.2070380449295044,
"learning_rate": 8.696565621918278e-05,
"loss": 3.7462,
"step": 86000
},
{
"epoch": 3.6462504742233275,
"grad_norm": 1.1980082988739014,
"learning_rate": 8.687990395746688e-05,
"loss": 3.7492,
"step": 86500
},
{
"epoch": 3.6673270665598787,
"grad_norm": 1.1856951713562012,
"learning_rate": 8.679415169575098e-05,
"loss": 3.7467,
"step": 87000
},
{
"epoch": 3.6884036588964295,
"grad_norm": 1.238400936126709,
"learning_rate": 8.670839943403508e-05,
"loss": 3.744,
"step": 87500
},
{
"epoch": 3.7094802512329808,
"grad_norm": 1.2197458744049072,
"learning_rate": 8.662264717231916e-05,
"loss": 3.7434,
"step": 88000
},
{
"epoch": 3.7305568435695315,
"grad_norm": 1.3021600246429443,
"learning_rate": 8.653689491060328e-05,
"loss": 3.7421,
"step": 88500
},
{
"epoch": 3.751633435906083,
"grad_norm": 1.1675195693969727,
"learning_rate": 8.645114264888737e-05,
"loss": 3.7412,
"step": 89000
},
{
"epoch": 3.7727100282426336,
"grad_norm": 1.3250484466552734,
"learning_rate": 8.636539038717146e-05,
"loss": 3.7345,
"step": 89500
},
{
"epoch": 3.793786620579185,
"grad_norm": 1.1882705688476562,
"learning_rate": 8.627963812545556e-05,
"loss": 3.735,
"step": 90000
},
{
"epoch": 3.793786620579185,
"eval_accuracy": 0.6752798484428797,
"eval_loss": 3.4582436084747314,
"eval_runtime": 158.8483,
"eval_samples_per_second": 759.882,
"eval_steps_per_second": 7.92,
"step": 90000
},
{
"epoch": 3.8148632129157356,
"grad_norm": 1.1858010292053223,
"learning_rate": 8.619388586373967e-05,
"loss": 3.7363,
"step": 90500
},
{
"epoch": 3.835939805252287,
"grad_norm": 1.2616113424301147,
"learning_rate": 8.610813360202375e-05,
"loss": 3.7302,
"step": 91000
},
{
"epoch": 3.857016397588838,
"grad_norm": 1.188483476638794,
"learning_rate": 8.602238134030785e-05,
"loss": 3.7338,
"step": 91500
},
{
"epoch": 3.878092989925389,
"grad_norm": 1.3678734302520752,
"learning_rate": 8.593662907859195e-05,
"loss": 3.7232,
"step": 92000
},
{
"epoch": 3.8991695822619397,
"grad_norm": 1.2638288736343384,
"learning_rate": 8.585087681687605e-05,
"loss": 3.7288,
"step": 92500
},
{
"epoch": 3.920246174598491,
"grad_norm": 1.1952755451202393,
"learning_rate": 8.576512455516015e-05,
"loss": 3.7278,
"step": 93000
},
{
"epoch": 3.941322766935042,
"grad_norm": 1.1661992073059082,
"learning_rate": 8.567937229344425e-05,
"loss": 3.7231,
"step": 93500
},
{
"epoch": 3.962399359271593,
"grad_norm": 1.196839690208435,
"learning_rate": 8.559362003172834e-05,
"loss": 3.7245,
"step": 94000
},
{
"epoch": 3.983475951608144,
"grad_norm": 1.2251155376434326,
"learning_rate": 8.550786777001244e-05,
"loss": 3.7221,
"step": 94500
},
{
"epoch": 4.004552543944695,
"grad_norm": 1.2336207628250122,
"learning_rate": 8.542211550829654e-05,
"loss": 3.7186,
"step": 95000
},
{
"epoch": 4.025629136281246,
"grad_norm": 1.2250827550888062,
"learning_rate": 8.533636324658063e-05,
"loss": 3.712,
"step": 95500
},
{
"epoch": 4.046705728617797,
"grad_norm": 1.2166626453399658,
"learning_rate": 8.525061098486474e-05,
"loss": 3.7136,
"step": 96000
},
{
"epoch": 4.067782320954348,
"grad_norm": 1.2013036012649536,
"learning_rate": 8.516485872314884e-05,
"loss": 3.7097,
"step": 96500
},
{
"epoch": 4.088858913290899,
"grad_norm": 1.161387324333191,
"learning_rate": 8.507910646143292e-05,
"loss": 3.7098,
"step": 97000
},
{
"epoch": 4.10993550562745,
"grad_norm": 1.2835077047348022,
"learning_rate": 8.499335419971702e-05,
"loss": 3.7079,
"step": 97500
},
{
"epoch": 4.1310120979640015,
"grad_norm": 1.2008073329925537,
"learning_rate": 8.490760193800112e-05,
"loss": 3.7059,
"step": 98000
},
{
"epoch": 4.152088690300552,
"grad_norm": 1.1533210277557373,
"learning_rate": 8.482184967628522e-05,
"loss": 3.7043,
"step": 98500
},
{
"epoch": 4.173165282637103,
"grad_norm": 1.1885192394256592,
"learning_rate": 8.473609741456931e-05,
"loss": 3.7086,
"step": 99000
},
{
"epoch": 4.194241874973654,
"grad_norm": 1.2948914766311646,
"learning_rate": 8.46503451528534e-05,
"loss": 3.7079,
"step": 99500
},
{
"epoch": 4.2153184673102055,
"grad_norm": 1.2169519662857056,
"learning_rate": 8.456459289113751e-05,
"loss": 3.7016,
"step": 100000
},
{
"epoch": 4.2153184673102055,
"eval_accuracy": 0.6790343033963083,
"eval_loss": 3.4345057010650635,
"eval_runtime": 165.3163,
"eval_samples_per_second": 730.152,
"eval_steps_per_second": 7.61,
"step": 100000
},
{
"epoch": 4.236395059646756,
"grad_norm": 1.2250070571899414,
"learning_rate": 8.447884062942161e-05,
"loss": 3.7069,
"step": 100500
},
{
"epoch": 4.257471651983307,
"grad_norm": 1.203892469406128,
"learning_rate": 8.43930883677057e-05,
"loss": 3.7032,
"step": 101000
},
{
"epoch": 4.278548244319858,
"grad_norm": 1.2498092651367188,
"learning_rate": 8.430733610598979e-05,
"loss": 3.6977,
"step": 101500
},
{
"epoch": 4.29962483665641,
"grad_norm": 1.249043583869934,
"learning_rate": 8.42215838442739e-05,
"loss": 3.6988,
"step": 102000
},
{
"epoch": 4.32070142899296,
"grad_norm": 1.1411231756210327,
"learning_rate": 8.413583158255799e-05,
"loss": 3.698,
"step": 102500
},
{
"epoch": 4.341778021329511,
"grad_norm": 1.2004824876785278,
"learning_rate": 8.405007932084209e-05,
"loss": 3.6963,
"step": 103000
},
{
"epoch": 4.362854613666062,
"grad_norm": 1.1714224815368652,
"learning_rate": 8.396432705912619e-05,
"loss": 3.6949,
"step": 103500
},
{
"epoch": 4.383931206002614,
"grad_norm": 1.2591228485107422,
"learning_rate": 8.387857479741028e-05,
"loss": 3.6985,
"step": 104000
},
{
"epoch": 4.405007798339165,
"grad_norm": 1.2099586725234985,
"learning_rate": 8.379282253569438e-05,
"loss": 3.6936,
"step": 104500
},
{
"epoch": 4.426084390675715,
"grad_norm": 1.2398302555084229,
"learning_rate": 8.370707027397848e-05,
"loss": 3.6921,
"step": 105000
},
{
"epoch": 4.4471609830122665,
"grad_norm": 1.2352367639541626,
"learning_rate": 8.362131801226258e-05,
"loss": 3.6899,
"step": 105500
},
{
"epoch": 4.468237575348818,
"grad_norm": 1.124231219291687,
"learning_rate": 8.353556575054668e-05,
"loss": 3.6876,
"step": 106000
},
{
"epoch": 4.489314167685369,
"grad_norm": 1.2524739503860474,
"learning_rate": 8.344981348883077e-05,
"loss": 3.6899,
"step": 106500
},
{
"epoch": 4.510390760021919,
"grad_norm": 1.1850067377090454,
"learning_rate": 8.336406122711486e-05,
"loss": 3.687,
"step": 107000
},
{
"epoch": 4.5314673523584705,
"grad_norm": 1.2721275091171265,
"learning_rate": 8.327830896539897e-05,
"loss": 3.6911,
"step": 107500
},
{
"epoch": 4.552543944695022,
"grad_norm": 1.2511844635009766,
"learning_rate": 8.319255670368307e-05,
"loss": 3.6859,
"step": 108000
},
{
"epoch": 4.573620537031573,
"grad_norm": 1.2308571338653564,
"learning_rate": 8.310680444196715e-05,
"loss": 3.6833,
"step": 108500
},
{
"epoch": 4.594697129368123,
"grad_norm": 1.2480345964431763,
"learning_rate": 8.302105218025125e-05,
"loss": 3.6765,
"step": 109000
},
{
"epoch": 4.615773721704675,
"grad_norm": 1.2648814916610718,
"learning_rate": 8.293529991853536e-05,
"loss": 3.6853,
"step": 109500
},
{
"epoch": 4.636850314041226,
"grad_norm": 1.2957087755203247,
"learning_rate": 8.284954765681945e-05,
"loss": 3.681,
"step": 110000
},
{
"epoch": 4.636850314041226,
"eval_accuracy": 0.6824005657231067,
"eval_loss": 3.4122893810272217,
"eval_runtime": 160.307,
"eval_samples_per_second": 752.968,
"eval_steps_per_second": 7.847,
"step": 110000
},
{
"epoch": 4.657926906377777,
"grad_norm": 1.2255547046661377,
"learning_rate": 8.276379539510355e-05,
"loss": 3.6801,
"step": 110500
},
{
"epoch": 4.679003498714328,
"grad_norm": 1.2668893337249756,
"learning_rate": 8.267804313338765e-05,
"loss": 3.6788,
"step": 111000
},
{
"epoch": 4.700080091050879,
"grad_norm": 1.4156382083892822,
"learning_rate": 8.259229087167174e-05,
"loss": 3.6774,
"step": 111500
},
{
"epoch": 4.72115668338743,
"grad_norm": 1.2249494791030884,
"learning_rate": 8.250653860995584e-05,
"loss": 3.6758,
"step": 112000
},
{
"epoch": 4.742233275723981,
"grad_norm": 1.3053946495056152,
"learning_rate": 8.242078634823994e-05,
"loss": 3.6743,
"step": 112500
},
{
"epoch": 4.763309868060532,
"grad_norm": 1.1843082904815674,
"learning_rate": 8.233503408652403e-05,
"loss": 3.6745,
"step": 113000
},
{
"epoch": 4.784386460397083,
"grad_norm": 1.2605531215667725,
"learning_rate": 8.224928182480814e-05,
"loss": 3.6755,
"step": 113500
},
{
"epoch": 4.805463052733634,
"grad_norm": 1.189759612083435,
"learning_rate": 8.216352956309224e-05,
"loss": 3.6665,
"step": 114000
},
{
"epoch": 4.826539645070185,
"grad_norm": 1.2134628295898438,
"learning_rate": 8.207777730137632e-05,
"loss": 3.6689,
"step": 114500
},
{
"epoch": 4.847616237406736,
"grad_norm": 1.2298855781555176,
"learning_rate": 8.199202503966042e-05,
"loss": 3.6683,
"step": 115000
},
{
"epoch": 4.868692829743287,
"grad_norm": 1.203617811203003,
"learning_rate": 8.190627277794453e-05,
"loss": 3.6683,
"step": 115500
},
{
"epoch": 4.889769422079838,
"grad_norm": 1.2415810823440552,
"learning_rate": 8.182052051622862e-05,
"loss": 3.6653,
"step": 116000
},
{
"epoch": 4.910846014416389,
"grad_norm": 1.257602334022522,
"learning_rate": 8.173476825451271e-05,
"loss": 3.6677,
"step": 116500
},
{
"epoch": 4.9319226067529405,
"grad_norm": 1.2657731771469116,
"learning_rate": 8.164901599279681e-05,
"loss": 3.6649,
"step": 117000
},
{
"epoch": 4.952999199089491,
"grad_norm": 1.1760540008544922,
"learning_rate": 8.156326373108091e-05,
"loss": 3.6671,
"step": 117500
},
{
"epoch": 4.974075791426042,
"grad_norm": 1.2068182229995728,
"learning_rate": 8.147751146936501e-05,
"loss": 3.666,
"step": 118000
},
{
"epoch": 4.995152383762593,
"grad_norm": 1.2154312133789062,
"learning_rate": 8.139175920764911e-05,
"loss": 3.6586,
"step": 118500
},
{
"epoch": 5.0162289760991445,
"grad_norm": 1.227812647819519,
"learning_rate": 8.13060069459332e-05,
"loss": 3.6601,
"step": 119000
},
{
"epoch": 5.037305568435695,
"grad_norm": 1.1852141618728638,
"learning_rate": 8.12202546842173e-05,
"loss": 3.6554,
"step": 119500
},
{
"epoch": 5.058382160772246,
"grad_norm": 1.2056678533554077,
"learning_rate": 8.11345024225014e-05,
"loss": 3.6573,
"step": 120000
},
{
"epoch": 5.058382160772246,
"eval_accuracy": 0.6861152077714642,
"eval_loss": 3.3853940963745117,
"eval_runtime": 157.5548,
"eval_samples_per_second": 766.121,
"eval_steps_per_second": 7.985,
"step": 120000
},
{
"epoch": 5.079458753108797,
"grad_norm": 1.2582329511642456,
"learning_rate": 8.104875016078549e-05,
"loss": 3.6519,
"step": 120500
},
{
"epoch": 5.100535345445349,
"grad_norm": 1.2994945049285889,
"learning_rate": 8.09629978990696e-05,
"loss": 3.6511,
"step": 121000
},
{
"epoch": 5.1216119377819,
"grad_norm": 1.160012125968933,
"learning_rate": 8.08772456373537e-05,
"loss": 3.6541,
"step": 121500
},
{
"epoch": 5.14268853011845,
"grad_norm": 1.2531156539916992,
"learning_rate": 8.079149337563778e-05,
"loss": 3.6503,
"step": 122000
},
{
"epoch": 5.163765122455001,
"grad_norm": 1.1808799505233765,
"learning_rate": 8.070574111392188e-05,
"loss": 3.6487,
"step": 122500
},
{
"epoch": 5.184841714791553,
"grad_norm": 1.200656771659851,
"learning_rate": 8.061998885220599e-05,
"loss": 3.6482,
"step": 123000
},
{
"epoch": 5.205918307128104,
"grad_norm": 1.2239103317260742,
"learning_rate": 8.053423659049008e-05,
"loss": 3.6498,
"step": 123500
},
{
"epoch": 5.226994899464654,
"grad_norm": 1.284049153327942,
"learning_rate": 8.044848432877418e-05,
"loss": 3.6489,
"step": 124000
},
{
"epoch": 5.2480714918012055,
"grad_norm": 1.2771472930908203,
"learning_rate": 8.036273206705827e-05,
"loss": 3.6516,
"step": 124500
},
{
"epoch": 5.269148084137757,
"grad_norm": 1.1889935731887817,
"learning_rate": 8.027697980534237e-05,
"loss": 3.6453,
"step": 125000
},
{
"epoch": 5.290224676474308,
"grad_norm": 1.253517508506775,
"learning_rate": 8.019122754362647e-05,
"loss": 3.6469,
"step": 125500
},
{
"epoch": 5.311301268810858,
"grad_norm": 1.3170056343078613,
"learning_rate": 8.010547528191057e-05,
"loss": 3.6446,
"step": 126000
},
{
"epoch": 5.3323778611474095,
"grad_norm": 1.2436405420303345,
"learning_rate": 8.001972302019465e-05,
"loss": 3.6411,
"step": 126500
},
{
"epoch": 5.353454453483961,
"grad_norm": 1.2107585668563843,
"learning_rate": 7.993397075847876e-05,
"loss": 3.6413,
"step": 127000
},
{
"epoch": 5.374531045820512,
"grad_norm": 1.3058685064315796,
"learning_rate": 7.984821849676286e-05,
"loss": 3.6464,
"step": 127500
},
{
"epoch": 5.395607638157063,
"grad_norm": 1.1944959163665771,
"learning_rate": 7.976246623504695e-05,
"loss": 3.6415,
"step": 128000
},
{
"epoch": 5.416684230493614,
"grad_norm": 1.2887589931488037,
"learning_rate": 7.967671397333105e-05,
"loss": 3.6368,
"step": 128500
},
{
"epoch": 5.437760822830165,
"grad_norm": 1.1942954063415527,
"learning_rate": 7.959096171161516e-05,
"loss": 3.6372,
"step": 129000
},
{
"epoch": 5.458837415166716,
"grad_norm": 1.2082514762878418,
"learning_rate": 7.950520944989924e-05,
"loss": 3.6371,
"step": 129500
},
{
"epoch": 5.479914007503266,
"grad_norm": 1.2680776119232178,
"learning_rate": 7.941945718818334e-05,
"loss": 3.6373,
"step": 130000
},
{
"epoch": 5.479914007503266,
"eval_accuracy": 0.6888820669997842,
"eval_loss": 3.3675577640533447,
"eval_runtime": 158.2062,
"eval_samples_per_second": 762.966,
"eval_steps_per_second": 7.952,
"step": 130000
},
{
"epoch": 5.500990599839818,
"grad_norm": 1.2110999822616577,
"learning_rate": 7.933370492646744e-05,
"loss": 3.6405,
"step": 130500
},
{
"epoch": 5.522067192176369,
"grad_norm": 1.281000018119812,
"learning_rate": 7.924795266475154e-05,
"loss": 3.6339,
"step": 131000
},
{
"epoch": 5.54314378451292,
"grad_norm": 1.2875875234603882,
"learning_rate": 7.916220040303564e-05,
"loss": 3.6287,
"step": 131500
},
{
"epoch": 5.564220376849471,
"grad_norm": 1.2870848178863525,
"learning_rate": 7.907644814131972e-05,
"loss": 3.6363,
"step": 132000
},
{
"epoch": 5.585296969186022,
"grad_norm": 1.235945463180542,
"learning_rate": 7.899069587960383e-05,
"loss": 3.6303,
"step": 132500
},
{
"epoch": 5.606373561522573,
"grad_norm": 1.1811497211456299,
"learning_rate": 7.890494361788793e-05,
"loss": 3.6316,
"step": 133000
},
{
"epoch": 5.627450153859124,
"grad_norm": 1.3203915357589722,
"learning_rate": 7.881919135617202e-05,
"loss": 3.627,
"step": 133500
},
{
"epoch": 5.648526746195675,
"grad_norm": 1.2780276536941528,
"learning_rate": 7.873343909445611e-05,
"loss": 3.623,
"step": 134000
},
{
"epoch": 5.669603338532226,
"grad_norm": 1.2244418859481812,
"learning_rate": 7.864768683274023e-05,
"loss": 3.6341,
"step": 134500
},
{
"epoch": 5.690679930868777,
"grad_norm": 1.2857935428619385,
"learning_rate": 7.856193457102431e-05,
"loss": 3.633,
"step": 135000
},
{
"epoch": 5.711756523205328,
"grad_norm": 1.2075480222702026,
"learning_rate": 7.847618230930841e-05,
"loss": 3.629,
"step": 135500
},
{
"epoch": 5.7328331155418795,
"grad_norm": 1.18943190574646,
"learning_rate": 7.839043004759251e-05,
"loss": 3.6264,
"step": 136000
},
{
"epoch": 5.75390970787843,
"grad_norm": 1.2121959924697876,
"learning_rate": 7.83046777858766e-05,
"loss": 3.6249,
"step": 136500
},
{
"epoch": 5.774986300214981,
"grad_norm": 1.293273687362671,
"learning_rate": 7.82189255241607e-05,
"loss": 3.6245,
"step": 137000
},
{
"epoch": 5.796062892551532,
"grad_norm": 1.245897650718689,
"learning_rate": 7.81331732624448e-05,
"loss": 3.6218,
"step": 137500
},
{
"epoch": 5.8171394848880835,
"grad_norm": 1.1534141302108765,
"learning_rate": 7.804742100072889e-05,
"loss": 3.6222,
"step": 138000
},
{
"epoch": 5.838216077224635,
"grad_norm": 1.2734718322753906,
"learning_rate": 7.7961668739013e-05,
"loss": 3.6258,
"step": 138500
},
{
"epoch": 5.859292669561185,
"grad_norm": 1.2768186330795288,
"learning_rate": 7.78759164772971e-05,
"loss": 3.6212,
"step": 139000
},
{
"epoch": 5.880369261897736,
"grad_norm": 1.3364906311035156,
"learning_rate": 7.779016421558118e-05,
"loss": 3.619,
"step": 139500
},
{
"epoch": 5.901445854234288,
"grad_norm": 1.3427903652191162,
"learning_rate": 7.770441195386528e-05,
"loss": 3.6238,
"step": 140000
},
{
"epoch": 5.901445854234288,
"eval_accuracy": 0.6914522029141235,
"eval_loss": 3.350059986114502,
"eval_runtime": 157.9869,
"eval_samples_per_second": 764.025,
"eval_steps_per_second": 7.963,
"step": 140000
},
{
"epoch": 5.922522446570838,
"grad_norm": 1.2633213996887207,
"learning_rate": 7.761865969214939e-05,
"loss": 3.6155,
"step": 140500
},
{
"epoch": 5.943599038907389,
"grad_norm": 1.2555463314056396,
"learning_rate": 7.753290743043348e-05,
"loss": 3.6196,
"step": 141000
},
{
"epoch": 5.96467563124394,
"grad_norm": 1.3542109727859497,
"learning_rate": 7.744715516871758e-05,
"loss": 3.6196,
"step": 141500
},
{
"epoch": 5.985752223580492,
"grad_norm": 1.2034767866134644,
"learning_rate": 7.736140290700167e-05,
"loss": 3.6154,
"step": 142000
},
{
"epoch": 6.006828815917043,
"grad_norm": 1.3052647113800049,
"learning_rate": 7.727565064528577e-05,
"loss": 3.6134,
"step": 142500
},
{
"epoch": 6.027905408253593,
"grad_norm": 1.2425764799118042,
"learning_rate": 7.718989838356987e-05,
"loss": 3.6112,
"step": 143000
},
{
"epoch": 6.0489820005901445,
"grad_norm": 1.2847294807434082,
"learning_rate": 7.710414612185397e-05,
"loss": 3.615,
"step": 143500
},
{
"epoch": 6.070058592926696,
"grad_norm": 1.2556947469711304,
"learning_rate": 7.701839386013807e-05,
"loss": 3.6109,
"step": 144000
},
{
"epoch": 6.091135185263247,
"grad_norm": 1.1879955530166626,
"learning_rate": 7.693264159842217e-05,
"loss": 3.608,
"step": 144500
},
{
"epoch": 6.112211777599797,
"grad_norm": 1.28110933303833,
"learning_rate": 7.684688933670626e-05,
"loss": 3.6072,
"step": 145000
},
{
"epoch": 6.1332883699363485,
"grad_norm": 1.2078018188476562,
"learning_rate": 7.676113707499035e-05,
"loss": 3.6093,
"step": 145500
},
{
"epoch": 6.1543649622729,
"grad_norm": 1.2363221645355225,
"learning_rate": 7.667538481327446e-05,
"loss": 3.6079,
"step": 146000
},
{
"epoch": 6.175441554609451,
"grad_norm": 1.2876498699188232,
"learning_rate": 7.658963255155856e-05,
"loss": 3.6088,
"step": 146500
},
{
"epoch": 6.196518146946001,
"grad_norm": 1.2517874240875244,
"learning_rate": 7.650388028984264e-05,
"loss": 3.6063,
"step": 147000
},
{
"epoch": 6.217594739282553,
"grad_norm": 1.2875258922576904,
"learning_rate": 7.641812802812674e-05,
"loss": 3.6046,
"step": 147500
},
{
"epoch": 6.238671331619104,
"grad_norm": 1.292263150215149,
"learning_rate": 7.633237576641084e-05,
"loss": 3.603,
"step": 148000
},
{
"epoch": 6.259747923955655,
"grad_norm": 1.2617777585983276,
"learning_rate": 7.624662350469494e-05,
"loss": 3.6056,
"step": 148500
},
{
"epoch": 6.280824516292206,
"grad_norm": 1.2769538164138794,
"learning_rate": 7.616087124297904e-05,
"loss": 3.6046,
"step": 149000
},
{
"epoch": 6.301901108628757,
"grad_norm": 1.1972134113311768,
"learning_rate": 7.607511898126313e-05,
"loss": 3.6038,
"step": 149500
},
{
"epoch": 6.322977700965308,
"grad_norm": 1.2098448276519775,
"learning_rate": 7.598936671954723e-05,
"loss": 3.6004,
"step": 150000
},
{
"epoch": 6.322977700965308,
"eval_accuracy": 0.6939296571729416,
"eval_loss": 3.3354380130767822,
"eval_runtime": 159.1742,
"eval_samples_per_second": 758.327,
"eval_steps_per_second": 7.903,
"step": 150000
},
{
"epoch": 6.344054293301859,
"grad_norm": 1.3688713312149048,
"learning_rate": 7.590361445783133e-05,
"loss": 3.6011,
"step": 150500
},
{
"epoch": 6.36513088563841,
"grad_norm": 1.233596920967102,
"learning_rate": 7.581786219611543e-05,
"loss": 3.6018,
"step": 151000
},
{
"epoch": 6.386207477974961,
"grad_norm": 1.217858910560608,
"learning_rate": 7.573210993439951e-05,
"loss": 3.6009,
"step": 151500
},
{
"epoch": 6.407284070311512,
"grad_norm": 1.1927746534347534,
"learning_rate": 7.564635767268363e-05,
"loss": 3.6011,
"step": 152000
},
{
"epoch": 6.428360662648063,
"grad_norm": 1.2303208112716675,
"learning_rate": 7.556060541096772e-05,
"loss": 3.5956,
"step": 152500
},
{
"epoch": 6.449437254984614,
"grad_norm": 1.2555283308029175,
"learning_rate": 7.547485314925181e-05,
"loss": 3.599,
"step": 153000
},
{
"epoch": 6.470513847321165,
"grad_norm": 1.1678221225738525,
"learning_rate": 7.538910088753591e-05,
"loss": 3.596,
"step": 153500
},
{
"epoch": 6.491590439657716,
"grad_norm": 1.3094114065170288,
"learning_rate": 7.530334862582002e-05,
"loss": 3.5965,
"step": 154000
},
{
"epoch": 6.512667031994267,
"grad_norm": 1.2501640319824219,
"learning_rate": 7.52175963641041e-05,
"loss": 3.5981,
"step": 154500
},
{
"epoch": 6.5337436243308185,
"grad_norm": 1.2383755445480347,
"learning_rate": 7.51318441023882e-05,
"loss": 3.5994,
"step": 155000
},
{
"epoch": 6.55482021666737,
"grad_norm": 1.208776593208313,
"learning_rate": 7.50460918406723e-05,
"loss": 3.5976,
"step": 155500
},
{
"epoch": 6.57589680900392,
"grad_norm": 1.2515255212783813,
"learning_rate": 7.49603395789564e-05,
"loss": 3.5961,
"step": 156000
},
{
"epoch": 6.596973401340471,
"grad_norm": 1.1996874809265137,
"learning_rate": 7.48745873172405e-05,
"loss": 3.5901,
"step": 156500
},
{
"epoch": 6.6180499936770225,
"grad_norm": 1.1982700824737549,
"learning_rate": 7.47888350555246e-05,
"loss": 3.5864,
"step": 157000
},
{
"epoch": 6.639126586013573,
"grad_norm": 1.2439733743667603,
"learning_rate": 7.47030827938087e-05,
"loss": 3.5904,
"step": 157500
},
{
"epoch": 6.660203178350124,
"grad_norm": 1.2724612951278687,
"learning_rate": 7.461733053209279e-05,
"loss": 3.5933,
"step": 158000
},
{
"epoch": 6.681279770686675,
"grad_norm": 1.2363367080688477,
"learning_rate": 7.453157827037689e-05,
"loss": 3.5958,
"step": 158500
},
{
"epoch": 6.702356363023227,
"grad_norm": 1.233451008796692,
"learning_rate": 7.444582600866098e-05,
"loss": 3.5876,
"step": 159000
},
{
"epoch": 6.723432955359778,
"grad_norm": 1.3065727949142456,
"learning_rate": 7.436007374694507e-05,
"loss": 3.5925,
"step": 159500
},
{
"epoch": 6.744509547696328,
"grad_norm": 1.2875949144363403,
"learning_rate": 7.427432148522919e-05,
"loss": 3.5931,
"step": 160000
},
{
"epoch": 6.744509547696328,
"eval_accuracy": 0.6958908603747329,
"eval_loss": 3.3241372108459473,
"eval_runtime": 147.0421,
"eval_samples_per_second": 820.894,
"eval_steps_per_second": 8.555,
"step": 160000
},
{
"epoch": 6.765586140032879,
"grad_norm": 1.2521882057189941,
"learning_rate": 7.418856922351327e-05,
"loss": 3.5884,
"step": 160500
},
{
"epoch": 6.786662732369431,
"grad_norm": 1.3088583946228027,
"learning_rate": 7.410281696179737e-05,
"loss": 3.5868,
"step": 161000
},
{
"epoch": 6.807739324705982,
"grad_norm": 1.2584905624389648,
"learning_rate": 7.401706470008147e-05,
"loss": 3.5844,
"step": 161500
},
{
"epoch": 6.828815917042532,
"grad_norm": 1.1866470575332642,
"learning_rate": 7.393131243836557e-05,
"loss": 3.5842,
"step": 162000
},
{
"epoch": 6.8498925093790834,
"grad_norm": 1.2755358219146729,
"learning_rate": 7.384556017664966e-05,
"loss": 3.5827,
"step": 162500
},
{
"epoch": 6.870969101715635,
"grad_norm": 1.2329550981521606,
"learning_rate": 7.375980791493376e-05,
"loss": 3.5823,
"step": 163000
},
{
"epoch": 6.892045694052186,
"grad_norm": 1.2301194667816162,
"learning_rate": 7.367405565321786e-05,
"loss": 3.5804,
"step": 163500
},
{
"epoch": 6.913122286388736,
"grad_norm": 1.233413815498352,
"learning_rate": 7.358830339150196e-05,
"loss": 3.5841,
"step": 164000
},
{
"epoch": 6.9341988787252875,
"grad_norm": 1.248624324798584,
"learning_rate": 7.350255112978606e-05,
"loss": 3.5858,
"step": 164500
},
{
"epoch": 6.955275471061839,
"grad_norm": 1.254355788230896,
"learning_rate": 7.341679886807014e-05,
"loss": 3.5804,
"step": 165000
},
{
"epoch": 6.97635206339839,
"grad_norm": 1.297410488128662,
"learning_rate": 7.333104660635425e-05,
"loss": 3.5782,
"step": 165500
},
{
"epoch": 6.997428655734941,
"grad_norm": 1.3060461282730103,
"learning_rate": 7.324529434463835e-05,
"loss": 3.5797,
"step": 166000
},
{
"epoch": 7.018505248071492,
"grad_norm": 1.2634865045547485,
"learning_rate": 7.315954208292244e-05,
"loss": 3.5781,
"step": 166500
},
{
"epoch": 7.039581840408043,
"grad_norm": 1.2272111177444458,
"learning_rate": 7.307378982120653e-05,
"loss": 3.573,
"step": 167000
},
{
"epoch": 7.060658432744594,
"grad_norm": 1.2437827587127686,
"learning_rate": 7.298803755949063e-05,
"loss": 3.5735,
"step": 167500
},
{
"epoch": 7.081735025081145,
"grad_norm": 1.249191164970398,
"learning_rate": 7.290228529777473e-05,
"loss": 3.5751,
"step": 168000
},
{
"epoch": 7.102811617417696,
"grad_norm": 1.2280375957489014,
"learning_rate": 7.281653303605883e-05,
"loss": 3.5757,
"step": 168500
},
{
"epoch": 7.123888209754247,
"grad_norm": 1.2603150606155396,
"learning_rate": 7.273078077434293e-05,
"loss": 3.5782,
"step": 169000
},
{
"epoch": 7.144964802090798,
"grad_norm": 1.235575795173645,
"learning_rate": 7.264502851262703e-05,
"loss": 3.5709,
"step": 169500
},
{
"epoch": 7.166041394427349,
"grad_norm": 1.2476295232772827,
"learning_rate": 7.255927625091112e-05,
"loss": 3.5703,
"step": 170000
},
{
"epoch": 7.166041394427349,
"eval_accuracy": 0.6985823728460414,
"eval_loss": 3.307676315307617,
"eval_runtime": 147.0783,
"eval_samples_per_second": 820.692,
"eval_steps_per_second": 8.553,
"step": 170000
},
{
"epoch": 7.1871179867639,
"grad_norm": 1.236409306526184,
"learning_rate": 7.247352398919521e-05,
"loss": 3.5729,
"step": 170500
},
{
"epoch": 7.208194579100451,
"grad_norm": 1.3185795545578003,
"learning_rate": 7.238777172747931e-05,
"loss": 3.572,
"step": 171000
},
{
"epoch": 7.229271171437002,
"grad_norm": 1.2679409980773926,
"learning_rate": 7.230201946576342e-05,
"loss": 3.5721,
"step": 171500
},
{
"epoch": 7.250347763773553,
"grad_norm": 1.2387568950653076,
"learning_rate": 7.22162672040475e-05,
"loss": 3.5715,
"step": 172000
},
{
"epoch": 7.271424356110104,
"grad_norm": 1.2054314613342285,
"learning_rate": 7.21305149423316e-05,
"loss": 3.569,
"step": 172500
},
{
"epoch": 7.292500948446655,
"grad_norm": 1.2389501333236694,
"learning_rate": 7.20447626806157e-05,
"loss": 3.5658,
"step": 173000
},
{
"epoch": 7.313577540783206,
"grad_norm": 1.265933632850647,
"learning_rate": 7.19590104188998e-05,
"loss": 3.5653,
"step": 173500
},
{
"epoch": 7.3346541331197574,
"grad_norm": 1.206266164779663,
"learning_rate": 7.18732581571839e-05,
"loss": 3.5676,
"step": 174000
},
{
"epoch": 7.355730725456308,
"grad_norm": 1.2907836437225342,
"learning_rate": 7.1787505895468e-05,
"loss": 3.5652,
"step": 174500
},
{
"epoch": 7.376807317792859,
"grad_norm": 1.2357051372528076,
"learning_rate": 7.17017536337521e-05,
"loss": 3.5664,
"step": 175000
},
{
"epoch": 7.39788391012941,
"grad_norm": 1.1954466104507446,
"learning_rate": 7.161600137203619e-05,
"loss": 3.5693,
"step": 175500
},
{
"epoch": 7.4189605024659615,
"grad_norm": 1.3017399311065674,
"learning_rate": 7.153024911032029e-05,
"loss": 3.5634,
"step": 176000
},
{
"epoch": 7.440037094802513,
"grad_norm": 1.259626865386963,
"learning_rate": 7.144449684860438e-05,
"loss": 3.5668,
"step": 176500
},
{
"epoch": 7.461113687139063,
"grad_norm": 1.2639764547348022,
"learning_rate": 7.135874458688849e-05,
"loss": 3.5678,
"step": 177000
},
{
"epoch": 7.482190279475614,
"grad_norm": 1.340766429901123,
"learning_rate": 7.127299232517259e-05,
"loss": 3.5672,
"step": 177500
},
{
"epoch": 7.503266871812166,
"grad_norm": 1.2179548740386963,
"learning_rate": 7.118724006345667e-05,
"loss": 3.5655,
"step": 178000
},
{
"epoch": 7.524343464148717,
"grad_norm": 1.2619128227233887,
"learning_rate": 7.110148780174077e-05,
"loss": 3.5612,
"step": 178500
},
{
"epoch": 7.545420056485267,
"grad_norm": 1.3326951265335083,
"learning_rate": 7.101573554002488e-05,
"loss": 3.5611,
"step": 179000
},
{
"epoch": 7.566496648821818,
"grad_norm": 1.2543479204177856,
"learning_rate": 7.092998327830897e-05,
"loss": 3.5622,
"step": 179500
},
{
"epoch": 7.58757324115837,
"grad_norm": 1.2507449388504028,
"learning_rate": 7.084423101659306e-05,
"loss": 3.5616,
"step": 180000
},
{
"epoch": 7.58757324115837,
"eval_accuracy": 0.6993166824355962,
"eval_loss": 3.3020989894866943,
"eval_runtime": 147.124,
"eval_samples_per_second": 820.437,
"eval_steps_per_second": 8.551,
"step": 180000
},
{
"epoch": 7.608649833494921,
"grad_norm": 1.1974754333496094,
"learning_rate": 7.075847875487716e-05,
"loss": 3.5596,
"step": 180500
},
{
"epoch": 7.629726425831471,
"grad_norm": 1.2690682411193848,
"learning_rate": 7.067272649316126e-05,
"loss": 3.5617,
"step": 181000
},
{
"epoch": 7.650803018168022,
"grad_norm": 1.2875919342041016,
"learning_rate": 7.058697423144536e-05,
"loss": 3.5565,
"step": 181500
},
{
"epoch": 7.671879610504574,
"grad_norm": 1.2431461811065674,
"learning_rate": 7.050122196972946e-05,
"loss": 3.5563,
"step": 182000
},
{
"epoch": 7.692956202841125,
"grad_norm": 1.2310460805892944,
"learning_rate": 7.041546970801354e-05,
"loss": 3.5585,
"step": 182500
},
{
"epoch": 7.714032795177676,
"grad_norm": 1.2813167572021484,
"learning_rate": 7.032971744629765e-05,
"loss": 3.5578,
"step": 183000
},
{
"epoch": 7.7351093875142265,
"grad_norm": 1.2496862411499023,
"learning_rate": 7.024396518458175e-05,
"loss": 3.5612,
"step": 183500
},
{
"epoch": 7.756185979850778,
"grad_norm": 1.2193994522094727,
"learning_rate": 7.015821292286584e-05,
"loss": 3.5577,
"step": 184000
},
{
"epoch": 7.777262572187329,
"grad_norm": 1.2853484153747559,
"learning_rate": 7.007246066114993e-05,
"loss": 3.5577,
"step": 184500
},
{
"epoch": 7.798339164523879,
"grad_norm": 1.2962794303894043,
"learning_rate": 6.998670839943405e-05,
"loss": 3.5562,
"step": 185000
},
{
"epoch": 7.8194157568604306,
"grad_norm": 1.2816979885101318,
"learning_rate": 6.990095613771813e-05,
"loss": 3.5547,
"step": 185500
},
{
"epoch": 7.840492349196982,
"grad_norm": 1.2759281396865845,
"learning_rate": 6.981520387600223e-05,
"loss": 3.5552,
"step": 186000
},
{
"epoch": 7.861568941533533,
"grad_norm": 1.2163993120193481,
"learning_rate": 6.972945161428633e-05,
"loss": 3.5536,
"step": 186500
},
{
"epoch": 7.882645533870084,
"grad_norm": 1.237457275390625,
"learning_rate": 6.964369935257043e-05,
"loss": 3.5543,
"step": 187000
},
{
"epoch": 7.903722126206635,
"grad_norm": 1.2803075313568115,
"learning_rate": 6.955794709085452e-05,
"loss": 3.5547,
"step": 187500
},
{
"epoch": 7.924798718543186,
"grad_norm": 1.2613801956176758,
"learning_rate": 6.947219482913862e-05,
"loss": 3.5435,
"step": 188000
},
{
"epoch": 7.945875310879737,
"grad_norm": 1.2788318395614624,
"learning_rate": 6.938644256742272e-05,
"loss": 3.5545,
"step": 188500
},
{
"epoch": 7.966951903216288,
"grad_norm": 1.2725120782852173,
"learning_rate": 6.930069030570682e-05,
"loss": 3.5518,
"step": 189000
},
{
"epoch": 7.988028495552839,
"grad_norm": 1.312692403793335,
"learning_rate": 6.921493804399092e-05,
"loss": 3.548,
"step": 189500
},
{
"epoch": 8.00910508788939,
"grad_norm": 1.2593975067138672,
"learning_rate": 6.9129185782275e-05,
"loss": 3.5502,
"step": 190000
},
{
"epoch": 8.00910508788939,
"eval_accuracy": 0.7014315881742103,
"eval_loss": 3.289201498031616,
"eval_runtime": 147.0267,
"eval_samples_per_second": 820.98,
"eval_steps_per_second": 8.556,
"step": 190000
},
{
"epoch": 8.03018168022594,
"grad_norm": 1.2814980745315552,
"learning_rate": 6.904343352055911e-05,
"loss": 3.544,
"step": 190500
},
{
"epoch": 8.051258272562492,
"grad_norm": 1.301543951034546,
"learning_rate": 6.895768125884321e-05,
"loss": 3.5479,
"step": 191000
},
{
"epoch": 8.072334864899043,
"grad_norm": 1.2298086881637573,
"learning_rate": 6.88719289971273e-05,
"loss": 3.5465,
"step": 191500
},
{
"epoch": 8.093411457235595,
"grad_norm": 1.3238123655319214,
"learning_rate": 6.87861767354114e-05,
"loss": 3.5413,
"step": 192000
},
{
"epoch": 8.114488049572145,
"grad_norm": 1.2859206199645996,
"learning_rate": 6.870042447369551e-05,
"loss": 3.5481,
"step": 192500
},
{
"epoch": 8.135564641908696,
"grad_norm": 1.2894060611724854,
"learning_rate": 6.861467221197959e-05,
"loss": 3.5442,
"step": 193000
},
{
"epoch": 8.156641234245248,
"grad_norm": 1.2969906330108643,
"learning_rate": 6.852891995026369e-05,
"loss": 3.5461,
"step": 193500
},
{
"epoch": 8.177717826581798,
"grad_norm": 1.2161030769348145,
"learning_rate": 6.844316768854779e-05,
"loss": 3.5469,
"step": 194000
},
{
"epoch": 8.19879441891835,
"grad_norm": 1.2727272510528564,
"learning_rate": 6.835741542683189e-05,
"loss": 3.545,
"step": 194500
},
{
"epoch": 8.2198710112549,
"grad_norm": 1.2221840620040894,
"learning_rate": 6.827166316511599e-05,
"loss": 3.5405,
"step": 195000
},
{
"epoch": 8.24094760359145,
"grad_norm": 1.2282825708389282,
"learning_rate": 6.818591090340008e-05,
"loss": 3.5439,
"step": 195500
},
{
"epoch": 8.262024195928003,
"grad_norm": 1.2006886005401611,
"learning_rate": 6.810015864168417e-05,
"loss": 3.5397,
"step": 196000
},
{
"epoch": 8.283100788264553,
"grad_norm": 1.2151827812194824,
"learning_rate": 6.801440637996828e-05,
"loss": 3.5397,
"step": 196500
},
{
"epoch": 8.304177380601104,
"grad_norm": 1.280169129371643,
"learning_rate": 6.792865411825238e-05,
"loss": 3.5423,
"step": 197000
},
{
"epoch": 8.325253972937656,
"grad_norm": 1.252305507659912,
"learning_rate": 6.784290185653646e-05,
"loss": 3.5387,
"step": 197500
},
{
"epoch": 8.346330565274206,
"grad_norm": 1.3033969402313232,
"learning_rate": 6.775714959482056e-05,
"loss": 3.5386,
"step": 198000
},
{
"epoch": 8.367407157610758,
"grad_norm": 1.2630615234375,
"learning_rate": 6.767139733310467e-05,
"loss": 3.5403,
"step": 198500
},
{
"epoch": 8.388483749947309,
"grad_norm": 1.273779034614563,
"learning_rate": 6.758564507138876e-05,
"loss": 3.5431,
"step": 199000
},
{
"epoch": 8.409560342283859,
"grad_norm": 1.2285422086715698,
"learning_rate": 6.749989280967286e-05,
"loss": 3.5429,
"step": 199500
},
{
"epoch": 8.430636934620411,
"grad_norm": 1.2629889249801636,
"learning_rate": 6.741414054795696e-05,
"loss": 3.5388,
"step": 200000
},
{
"epoch": 8.430636934620411,
"eval_accuracy": 0.703292521515008,
"eval_loss": 3.278486728668213,
"eval_runtime": 147.161,
"eval_samples_per_second": 820.231,
"eval_steps_per_second": 8.548,
"step": 200000
},
{
"epoch": 8.451713526956961,
"grad_norm": 1.2859611511230469,
"learning_rate": 6.732838828624105e-05,
"loss": 3.5385,
"step": 200500
},
{
"epoch": 8.472790119293512,
"grad_norm": 1.278258204460144,
"learning_rate": 6.724263602452515e-05,
"loss": 3.5396,
"step": 201000
},
{
"epoch": 8.493866711630064,
"grad_norm": 1.2605531215667725,
"learning_rate": 6.715688376280924e-05,
"loss": 3.5369,
"step": 201500
},
{
"epoch": 8.514943303966614,
"grad_norm": 1.322540044784546,
"learning_rate": 6.707113150109335e-05,
"loss": 3.5358,
"step": 202000
},
{
"epoch": 8.536019896303166,
"grad_norm": 1.2791666984558105,
"learning_rate": 6.698537923937745e-05,
"loss": 3.5407,
"step": 202500
},
{
"epoch": 8.557096488639717,
"grad_norm": 1.2757481336593628,
"learning_rate": 6.689962697766153e-05,
"loss": 3.5371,
"step": 203000
},
{
"epoch": 8.578173080976267,
"grad_norm": 1.272605061531067,
"learning_rate": 6.681387471594563e-05,
"loss": 3.5369,
"step": 203500
},
{
"epoch": 8.59924967331282,
"grad_norm": 1.2608590126037598,
"learning_rate": 6.672812245422974e-05,
"loss": 3.5352,
"step": 204000
},
{
"epoch": 8.62032626564937,
"grad_norm": 1.2239645719528198,
"learning_rate": 6.664237019251383e-05,
"loss": 3.5343,
"step": 204500
},
{
"epoch": 8.64140285798592,
"grad_norm": 1.3531590700149536,
"learning_rate": 6.655661793079792e-05,
"loss": 3.5355,
"step": 205000
},
{
"epoch": 8.662479450322472,
"grad_norm": 1.3026645183563232,
"learning_rate": 6.647086566908202e-05,
"loss": 3.531,
"step": 205500
},
{
"epoch": 8.683556042659022,
"grad_norm": 1.2277380228042603,
"learning_rate": 6.638511340736612e-05,
"loss": 3.5316,
"step": 206000
},
{
"epoch": 8.704632634995574,
"grad_norm": 1.237056851387024,
"learning_rate": 6.629936114565022e-05,
"loss": 3.5285,
"step": 206500
},
{
"epoch": 8.725709227332125,
"grad_norm": 1.289618968963623,
"learning_rate": 6.621360888393432e-05,
"loss": 3.5304,
"step": 207000
},
{
"epoch": 8.746785819668675,
"grad_norm": 1.2640162706375122,
"learning_rate": 6.61278566222184e-05,
"loss": 3.5237,
"step": 207500
},
{
"epoch": 8.767862412005227,
"grad_norm": 1.257179856300354,
"learning_rate": 6.604210436050251e-05,
"loss": 3.5292,
"step": 208000
},
{
"epoch": 8.788939004341778,
"grad_norm": 1.2502237558364868,
"learning_rate": 6.595635209878661e-05,
"loss": 3.5301,
"step": 208500
},
{
"epoch": 8.81001559667833,
"grad_norm": 1.245835542678833,
"learning_rate": 6.58705998370707e-05,
"loss": 3.5292,
"step": 209000
},
{
"epoch": 8.83109218901488,
"grad_norm": 1.2822822332382202,
"learning_rate": 6.57848475753548e-05,
"loss": 3.5293,
"step": 209500
},
{
"epoch": 8.85216878135143,
"grad_norm": 1.2632429599761963,
"learning_rate": 6.569909531363891e-05,
"loss": 3.5264,
"step": 210000
},
{
"epoch": 8.85216878135143,
"eval_accuracy": 0.7045577430509905,
"eval_loss": 3.2708382606506348,
"eval_runtime": 146.9197,
"eval_samples_per_second": 821.578,
"eval_steps_per_second": 8.562,
"step": 210000
},
{
"epoch": 8.873245373687983,
"grad_norm": 1.2645542621612549,
"learning_rate": 6.561334305192299e-05,
"loss": 3.5278,
"step": 210500
},
{
"epoch": 8.894321966024533,
"grad_norm": 1.338228464126587,
"learning_rate": 6.552759079020709e-05,
"loss": 3.5304,
"step": 211000
},
{
"epoch": 8.915398558361083,
"grad_norm": 1.3031375408172607,
"learning_rate": 6.544183852849119e-05,
"loss": 3.5255,
"step": 211500
},
{
"epoch": 8.936475150697635,
"grad_norm": 1.2210642099380493,
"learning_rate": 6.535608626677529e-05,
"loss": 3.5313,
"step": 212000
},
{
"epoch": 8.957551743034186,
"grad_norm": 1.231114149093628,
"learning_rate": 6.527033400505939e-05,
"loss": 3.529,
"step": 212500
},
{
"epoch": 8.978628335370738,
"grad_norm": 1.2242404222488403,
"learning_rate": 6.518458174334348e-05,
"loss": 3.5302,
"step": 213000
},
{
"epoch": 8.999704927707288,
"grad_norm": 1.2443705797195435,
"learning_rate": 6.509882948162758e-05,
"loss": 3.529,
"step": 213500
},
{
"epoch": 9.020781520043839,
"grad_norm": 1.3141529560089111,
"learning_rate": 6.501307721991168e-05,
"loss": 3.5232,
"step": 214000
},
{
"epoch": 9.04185811238039,
"grad_norm": 1.296618938446045,
"learning_rate": 6.492732495819578e-05,
"loss": 3.5203,
"step": 214500
},
{
"epoch": 9.062934704716941,
"grad_norm": 1.2757512331008911,
"learning_rate": 6.484157269647986e-05,
"loss": 3.5244,
"step": 215000
},
{
"epoch": 9.084011297053493,
"grad_norm": 1.2778700590133667,
"learning_rate": 6.475582043476398e-05,
"loss": 3.5221,
"step": 215500
},
{
"epoch": 9.105087889390044,
"grad_norm": 1.3288307189941406,
"learning_rate": 6.467006817304807e-05,
"loss": 3.5189,
"step": 216000
},
{
"epoch": 9.126164481726594,
"grad_norm": 1.2778294086456299,
"learning_rate": 6.458431591133216e-05,
"loss": 3.515,
"step": 216500
},
{
"epoch": 9.147241074063146,
"grad_norm": 1.3161003589630127,
"learning_rate": 6.449856364961626e-05,
"loss": 3.5199,
"step": 217000
},
{
"epoch": 9.168317666399696,
"grad_norm": 1.2818279266357422,
"learning_rate": 6.441281138790037e-05,
"loss": 3.518,
"step": 217500
},
{
"epoch": 9.189394258736247,
"grad_norm": 1.253825306892395,
"learning_rate": 6.432705912618445e-05,
"loss": 3.5182,
"step": 218000
},
{
"epoch": 9.210470851072799,
"grad_norm": 1.258705973625183,
"learning_rate": 6.424130686446855e-05,
"loss": 3.5221,
"step": 218500
},
{
"epoch": 9.23154744340935,
"grad_norm": 1.2923662662506104,
"learning_rate": 6.415555460275265e-05,
"loss": 3.5185,
"step": 219000
},
{
"epoch": 9.252624035745901,
"grad_norm": 1.2419333457946777,
"learning_rate": 6.406980234103675e-05,
"loss": 3.5193,
"step": 219500
},
{
"epoch": 9.273700628082452,
"grad_norm": 1.2986774444580078,
"learning_rate": 6.398405007932085e-05,
"loss": 3.5212,
"step": 220000
},
{
"epoch": 9.273700628082452,
"eval_accuracy": 0.7060866067296248,
"eval_loss": 3.2597711086273193,
"eval_runtime": 146.9785,
"eval_samples_per_second": 821.249,
"eval_steps_per_second": 8.559,
"step": 220000
},
{
"epoch": 9.294777220419002,
"grad_norm": 1.2577719688415527,
"learning_rate": 6.389829781760495e-05,
"loss": 3.518,
"step": 220500
},
{
"epoch": 9.315853812755554,
"grad_norm": 1.3110220432281494,
"learning_rate": 6.381254555588903e-05,
"loss": 3.5204,
"step": 221000
},
{
"epoch": 9.336930405092104,
"grad_norm": 1.2881900072097778,
"learning_rate": 6.372679329417314e-05,
"loss": 3.5153,
"step": 221500
},
{
"epoch": 9.358006997428657,
"grad_norm": 1.2712870836257935,
"learning_rate": 6.364104103245724e-05,
"loss": 3.5123,
"step": 222000
},
{
"epoch": 9.379083589765207,
"grad_norm": 1.3054606914520264,
"learning_rate": 6.355528877074132e-05,
"loss": 3.5177,
"step": 222500
},
{
"epoch": 9.400160182101757,
"grad_norm": 1.3274202346801758,
"learning_rate": 6.346953650902542e-05,
"loss": 3.5156,
"step": 223000
},
{
"epoch": 9.42123677443831,
"grad_norm": 1.2802754640579224,
"learning_rate": 6.338378424730954e-05,
"loss": 3.5158,
"step": 223500
},
{
"epoch": 9.44231336677486,
"grad_norm": 1.3183026313781738,
"learning_rate": 6.329803198559362e-05,
"loss": 3.5134,
"step": 224000
},
{
"epoch": 9.46338995911141,
"grad_norm": 1.246785044670105,
"learning_rate": 6.321227972387772e-05,
"loss": 3.5142,
"step": 224500
},
{
"epoch": 9.484466551447962,
"grad_norm": 1.2537015676498413,
"learning_rate": 6.312652746216182e-05,
"loss": 3.51,
"step": 225000
},
{
"epoch": 9.505543143784513,
"grad_norm": 1.2763338088989258,
"learning_rate": 6.304077520044591e-05,
"loss": 3.5187,
"step": 225500
},
{
"epoch": 9.526619736121065,
"grad_norm": 1.2788455486297607,
"learning_rate": 6.295502293873001e-05,
"loss": 3.5135,
"step": 226000
},
{
"epoch": 9.547696328457615,
"grad_norm": 1.2547003030776978,
"learning_rate": 6.286927067701411e-05,
"loss": 3.5096,
"step": 226500
},
{
"epoch": 9.568772920794165,
"grad_norm": 1.3502486944198608,
"learning_rate": 6.278351841529821e-05,
"loss": 3.5136,
"step": 227000
},
{
"epoch": 9.589849513130718,
"grad_norm": 1.3184937238693237,
"learning_rate": 6.269776615358231e-05,
"loss": 3.5122,
"step": 227500
},
{
"epoch": 9.610926105467268,
"grad_norm": 1.3496809005737305,
"learning_rate": 6.26120138918664e-05,
"loss": 3.5107,
"step": 228000
},
{
"epoch": 9.63200269780382,
"grad_norm": 1.2101354598999023,
"learning_rate": 6.252626163015049e-05,
"loss": 3.5109,
"step": 228500
},
{
"epoch": 9.65307929014037,
"grad_norm": 1.2941261529922485,
"learning_rate": 6.24405093684346e-05,
"loss": 3.5077,
"step": 229000
},
{
"epoch": 9.67415588247692,
"grad_norm": 1.209496259689331,
"learning_rate": 6.23547571067187e-05,
"loss": 3.5092,
"step": 229500
},
{
"epoch": 9.695232474813473,
"grad_norm": 1.3175601959228516,
"learning_rate": 6.226900484500279e-05,
"loss": 3.5045,
"step": 230000
},
{
"epoch": 9.695232474813473,
"eval_accuracy": 0.7073127359352447,
"eval_loss": 3.252577781677246,
"eval_runtime": 147.7056,
"eval_samples_per_second": 817.206,
"eval_steps_per_second": 8.517,
"step": 230000
},
{
"epoch": 9.716309067150023,
"grad_norm": 1.2992041110992432,
"learning_rate": 6.218325258328688e-05,
"loss": 3.5124,
"step": 230500
},
{
"epoch": 9.737385659486574,
"grad_norm": 1.2638344764709473,
"learning_rate": 6.209750032157098e-05,
"loss": 3.5113,
"step": 231000
},
{
"epoch": 9.758462251823126,
"grad_norm": 1.2856242656707764,
"learning_rate": 6.201174805985508e-05,
"loss": 3.5046,
"step": 231500
},
{
"epoch": 9.779538844159676,
"grad_norm": 1.307008147239685,
"learning_rate": 6.192599579813918e-05,
"loss": 3.5078,
"step": 232000
},
{
"epoch": 9.800615436496226,
"grad_norm": 1.278564214706421,
"learning_rate": 6.184024353642328e-05,
"loss": 3.5057,
"step": 232500
},
{
"epoch": 9.821692028832778,
"grad_norm": 1.3658078908920288,
"learning_rate": 6.175449127470738e-05,
"loss": 3.5043,
"step": 233000
},
{
"epoch": 9.842768621169329,
"grad_norm": 1.2370128631591797,
"learning_rate": 6.166873901299147e-05,
"loss": 3.5099,
"step": 233500
},
{
"epoch": 9.863845213505881,
"grad_norm": 1.2495208978652954,
"learning_rate": 6.158298675127557e-05,
"loss": 3.5041,
"step": 234000
},
{
"epoch": 9.884921805842431,
"grad_norm": 1.3482348918914795,
"learning_rate": 6.149723448955966e-05,
"loss": 3.5076,
"step": 234500
},
{
"epoch": 9.905998398178982,
"grad_norm": 1.2756807804107666,
"learning_rate": 6.141148222784377e-05,
"loss": 3.5038,
"step": 235000
},
{
"epoch": 9.927074990515534,
"grad_norm": 1.2510552406311035,
"learning_rate": 6.132572996612787e-05,
"loss": 3.5023,
"step": 235500
},
{
"epoch": 9.948151582852084,
"grad_norm": 1.317958116531372,
"learning_rate": 6.123997770441195e-05,
"loss": 3.5073,
"step": 236000
},
{
"epoch": 9.969228175188636,
"grad_norm": 1.3200846910476685,
"learning_rate": 6.115422544269605e-05,
"loss": 3.5077,
"step": 236500
},
{
"epoch": 9.990304767525187,
"grad_norm": 1.3323860168457031,
"learning_rate": 6.106847318098015e-05,
"loss": 3.5063,
"step": 237000
},
{
"epoch": 10.011381359861737,
"grad_norm": 1.298541784286499,
"learning_rate": 6.098272091926425e-05,
"loss": 3.5049,
"step": 237500
},
{
"epoch": 10.032457952198289,
"grad_norm": 1.3729757070541382,
"learning_rate": 6.0896968657548345e-05,
"loss": 3.5024,
"step": 238000
},
{
"epoch": 10.05353454453484,
"grad_norm": 1.287295937538147,
"learning_rate": 6.081121639583244e-05,
"loss": 3.5003,
"step": 238500
},
{
"epoch": 10.07461113687139,
"grad_norm": 1.2559477090835571,
"learning_rate": 6.072546413411654e-05,
"loss": 3.4999,
"step": 239000
},
{
"epoch": 10.095687729207942,
"grad_norm": 1.2970126867294312,
"learning_rate": 6.063971187240064e-05,
"loss": 3.5002,
"step": 239500
},
{
"epoch": 10.116764321544492,
"grad_norm": 1.3419015407562256,
"learning_rate": 6.055395961068473e-05,
"loss": 3.4939,
"step": 240000
},
{
"epoch": 10.116764321544492,
"eval_accuracy": 0.7087457916472529,
"eval_loss": 3.2483291625976562,
"eval_runtime": 147.4222,
"eval_samples_per_second": 818.777,
"eval_steps_per_second": 8.533,
"step": 240000
},
{
"epoch": 10.137840913881044,
"grad_norm": 1.2927186489105225,
"learning_rate": 6.046820734896883e-05,
"loss": 3.4969,
"step": 240500
},
{
"epoch": 10.158917506217595,
"grad_norm": 1.282796025276184,
"learning_rate": 6.038245508725293e-05,
"loss": 3.4987,
"step": 241000
},
{
"epoch": 10.179994098554145,
"grad_norm": 1.3105281591415405,
"learning_rate": 6.029670282553702e-05,
"loss": 3.4975,
"step": 241500
},
{
"epoch": 10.201070690890697,
"grad_norm": 1.2927544116973877,
"learning_rate": 6.0210950563821125e-05,
"loss": 3.4995,
"step": 242000
},
{
"epoch": 10.222147283227248,
"grad_norm": 1.2876425981521606,
"learning_rate": 6.0125198302105223e-05,
"loss": 3.499,
"step": 242500
},
{
"epoch": 10.2432238755638,
"grad_norm": 1.2782328128814697,
"learning_rate": 6.0039446040389315e-05,
"loss": 3.4955,
"step": 243000
},
{
"epoch": 10.26430046790035,
"grad_norm": 1.3352009057998657,
"learning_rate": 5.995369377867341e-05,
"loss": 3.5,
"step": 243500
},
{
"epoch": 10.2853770602369,
"grad_norm": 1.2464860677719116,
"learning_rate": 5.986794151695752e-05,
"loss": 3.491,
"step": 244000
},
{
"epoch": 10.306453652573452,
"grad_norm": 1.3195289373397827,
"learning_rate": 5.97821892552416e-05,
"loss": 3.4999,
"step": 244500
},
{
"epoch": 10.327530244910003,
"grad_norm": 1.3341643810272217,
"learning_rate": 5.969643699352571e-05,
"loss": 3.4985,
"step": 245000
},
{
"epoch": 10.348606837246553,
"grad_norm": 1.2749632596969604,
"learning_rate": 5.9610684731809806e-05,
"loss": 3.4966,
"step": 245500
},
{
"epoch": 10.369683429583105,
"grad_norm": 1.2920303344726562,
"learning_rate": 5.95249324700939e-05,
"loss": 3.4972,
"step": 246000
},
{
"epoch": 10.390760021919656,
"grad_norm": 1.2580958604812622,
"learning_rate": 5.9439180208377996e-05,
"loss": 3.4953,
"step": 246500
},
{
"epoch": 10.411836614256208,
"grad_norm": 1.311607837677002,
"learning_rate": 5.93534279466621e-05,
"loss": 3.4938,
"step": 247000
},
{
"epoch": 10.432913206592758,
"grad_norm": 1.270140290260315,
"learning_rate": 5.9267675684946186e-05,
"loss": 3.4973,
"step": 247500
},
{
"epoch": 10.453989798929308,
"grad_norm": 1.247283697128296,
"learning_rate": 5.918192342323029e-05,
"loss": 3.4932,
"step": 248000
},
{
"epoch": 10.47506639126586,
"grad_norm": 1.259935736656189,
"learning_rate": 5.909617116151439e-05,
"loss": 3.4936,
"step": 248500
},
{
"epoch": 10.496142983602411,
"grad_norm": 1.341447353363037,
"learning_rate": 5.901041889979848e-05,
"loss": 3.49,
"step": 249000
},
{
"epoch": 10.517219575938963,
"grad_norm": 1.3087300062179565,
"learning_rate": 5.892466663808258e-05,
"loss": 3.4923,
"step": 249500
},
{
"epoch": 10.538296168275513,
"grad_norm": 1.3003551959991455,
"learning_rate": 5.8838914376366684e-05,
"loss": 3.4934,
"step": 250000
},
{
"epoch": 10.538296168275513,
"eval_accuracy": 0.7100635488174574,
"eval_loss": 3.2361464500427246,
"eval_runtime": 147.6054,
"eval_samples_per_second": 817.761,
"eval_steps_per_second": 8.523,
"step": 250000
},
{
"epoch": 10.559372760612064,
"grad_norm": 1.282348871231079,
"learning_rate": 5.8753162114650776e-05,
"loss": 3.4974,
"step": 250500
},
{
"epoch": 10.580449352948616,
"grad_norm": 1.3179106712341309,
"learning_rate": 5.8667409852934874e-05,
"loss": 3.4916,
"step": 251000
},
{
"epoch": 10.601525945285166,
"grad_norm": 1.2755032777786255,
"learning_rate": 5.858165759121897e-05,
"loss": 3.4895,
"step": 251500
},
{
"epoch": 10.622602537621717,
"grad_norm": 1.2626327276229858,
"learning_rate": 5.8495905329503064e-05,
"loss": 3.4922,
"step": 252000
},
{
"epoch": 10.643679129958269,
"grad_norm": 1.2951860427856445,
"learning_rate": 5.841015306778716e-05,
"loss": 3.4922,
"step": 252500
},
{
"epoch": 10.664755722294819,
"grad_norm": 1.2879743576049805,
"learning_rate": 5.832440080607127e-05,
"loss": 3.4914,
"step": 253000
},
{
"epoch": 10.685832314631371,
"grad_norm": 1.335240125656128,
"learning_rate": 5.823864854435536e-05,
"loss": 3.4865,
"step": 253500
},
{
"epoch": 10.706908906967922,
"grad_norm": 1.2877485752105713,
"learning_rate": 5.815289628263946e-05,
"loss": 3.491,
"step": 254000
},
{
"epoch": 10.727985499304472,
"grad_norm": 1.317150592803955,
"learning_rate": 5.8067144020923556e-05,
"loss": 3.493,
"step": 254500
},
{
"epoch": 10.749062091641024,
"grad_norm": 1.3220484256744385,
"learning_rate": 5.798139175920765e-05,
"loss": 3.4896,
"step": 255000
},
{
"epoch": 10.770138683977574,
"grad_norm": 1.2945040464401245,
"learning_rate": 5.789563949749175e-05,
"loss": 3.4907,
"step": 255500
},
{
"epoch": 10.791215276314126,
"grad_norm": 1.290663480758667,
"learning_rate": 5.780988723577585e-05,
"loss": 3.488,
"step": 256000
},
{
"epoch": 10.812291868650677,
"grad_norm": 1.3087289333343506,
"learning_rate": 5.772413497405994e-05,
"loss": 3.4912,
"step": 256500
},
{
"epoch": 10.833368460987227,
"grad_norm": 1.270817518234253,
"learning_rate": 5.763838271234404e-05,
"loss": 3.4854,
"step": 257000
},
{
"epoch": 10.85444505332378,
"grad_norm": 1.3200535774230957,
"learning_rate": 5.7552630450628146e-05,
"loss": 3.4881,
"step": 257500
},
{
"epoch": 10.87552164566033,
"grad_norm": 1.2482467889785767,
"learning_rate": 5.746687818891223e-05,
"loss": 3.4858,
"step": 258000
},
{
"epoch": 10.89659823799688,
"grad_norm": 1.293032169342041,
"learning_rate": 5.7381125927196335e-05,
"loss": 3.4875,
"step": 258500
},
{
"epoch": 10.917674830333432,
"grad_norm": 1.382114052772522,
"learning_rate": 5.7295373665480434e-05,
"loss": 3.4879,
"step": 259000
},
{
"epoch": 10.938751422669982,
"grad_norm": 1.3330762386322021,
"learning_rate": 5.7209621403764525e-05,
"loss": 3.4873,
"step": 259500
},
{
"epoch": 10.959828015006533,
"grad_norm": 1.2735592126846313,
"learning_rate": 5.7123869142048624e-05,
"loss": 3.4833,
"step": 260000
},
{
"epoch": 10.959828015006533,
"eval_accuracy": 0.7110673510074895,
"eval_loss": 3.2300615310668945,
"eval_runtime": 147.3347,
"eval_samples_per_second": 819.264,
"eval_steps_per_second": 8.538,
"step": 260000
},
{
"epoch": 10.980904607343085,
"grad_norm": 1.282272219657898,
"learning_rate": 5.703811688033273e-05,
"loss": 3.4798,
"step": 260500
},
{
"epoch": 11.001981199679635,
"grad_norm": 1.2903333902359009,
"learning_rate": 5.695236461861681e-05,
"loss": 3.4833,
"step": 261000
},
{
"epoch": 11.023057792016187,
"grad_norm": 1.2990723848342896,
"learning_rate": 5.686661235690092e-05,
"loss": 3.4779,
"step": 261500
},
{
"epoch": 11.044134384352738,
"grad_norm": 1.2750413417816162,
"learning_rate": 5.678086009518502e-05,
"loss": 3.4846,
"step": 262000
},
{
"epoch": 11.065210976689288,
"grad_norm": 1.3147681951522827,
"learning_rate": 5.669510783346911e-05,
"loss": 3.4852,
"step": 262500
},
{
"epoch": 11.08628756902584,
"grad_norm": 1.310168743133545,
"learning_rate": 5.6609355571753207e-05,
"loss": 3.4812,
"step": 263000
},
{
"epoch": 11.10736416136239,
"grad_norm": 1.3368253707885742,
"learning_rate": 5.652360331003731e-05,
"loss": 3.4821,
"step": 263500
},
{
"epoch": 11.128440753698943,
"grad_norm": 1.2728437185287476,
"learning_rate": 5.6437851048321396e-05,
"loss": 3.4786,
"step": 264000
},
{
"epoch": 11.149517346035493,
"grad_norm": 1.3024355173110962,
"learning_rate": 5.63520987866055e-05,
"loss": 3.4833,
"step": 264500
},
{
"epoch": 11.170593938372043,
"grad_norm": 1.291304111480713,
"learning_rate": 5.62663465248896e-05,
"loss": 3.4782,
"step": 265000
},
{
"epoch": 11.191670530708596,
"grad_norm": 1.3226574659347534,
"learning_rate": 5.618059426317369e-05,
"loss": 3.4772,
"step": 265500
},
{
"epoch": 11.212747123045146,
"grad_norm": 1.2693146467208862,
"learning_rate": 5.609484200145779e-05,
"loss": 3.4822,
"step": 266000
},
{
"epoch": 11.233823715381696,
"grad_norm": 1.2951483726501465,
"learning_rate": 5.6009089739741895e-05,
"loss": 3.4804,
"step": 266500
},
{
"epoch": 11.254900307718248,
"grad_norm": 1.3409785032272339,
"learning_rate": 5.5923337478025986e-05,
"loss": 3.4774,
"step": 267000
},
{
"epoch": 11.275976900054799,
"grad_norm": 1.352249026298523,
"learning_rate": 5.5837585216310085e-05,
"loss": 3.4817,
"step": 267500
},
{
"epoch": 11.29705349239135,
"grad_norm": 1.375322937965393,
"learning_rate": 5.575183295459418e-05,
"loss": 3.4762,
"step": 268000
},
{
"epoch": 11.318130084727901,
"grad_norm": 1.2968153953552246,
"learning_rate": 5.5666080692878274e-05,
"loss": 3.4805,
"step": 268500
},
{
"epoch": 11.339206677064452,
"grad_norm": 1.2999241352081299,
"learning_rate": 5.558032843116238e-05,
"loss": 3.4788,
"step": 269000
},
{
"epoch": 11.360283269401004,
"grad_norm": 1.3537373542785645,
"learning_rate": 5.549457616944648e-05,
"loss": 3.4792,
"step": 269500
},
{
"epoch": 11.381359861737554,
"grad_norm": 1.298156976699829,
"learning_rate": 5.540882390773057e-05,
"loss": 3.4747,
"step": 270000
},
{
"epoch": 11.381359861737554,
"eval_accuracy": 0.712007146792482,
"eval_loss": 3.2252111434936523,
"eval_runtime": 147.8713,
"eval_samples_per_second": 816.291,
"eval_steps_per_second": 8.507,
"step": 270000
},
{
"epoch": 11.402436454074106,
"grad_norm": 1.3433160781860352,
"learning_rate": 5.532307164601467e-05,
"loss": 3.4769,
"step": 270500
},
{
"epoch": 11.423513046410656,
"grad_norm": 1.2658536434173584,
"learning_rate": 5.5237319384298766e-05,
"loss": 3.4758,
"step": 271000
},
{
"epoch": 11.444589638747207,
"grad_norm": 1.2452342510223389,
"learning_rate": 5.515156712258286e-05,
"loss": 3.4731,
"step": 271500
},
{
"epoch": 11.465666231083759,
"grad_norm": 1.2627519369125366,
"learning_rate": 5.506581486086696e-05,
"loss": 3.4747,
"step": 272000
},
{
"epoch": 11.48674282342031,
"grad_norm": 1.2426422834396362,
"learning_rate": 5.498006259915105e-05,
"loss": 3.4756,
"step": 272500
},
{
"epoch": 11.50781941575686,
"grad_norm": 1.3024731874465942,
"learning_rate": 5.489431033743515e-05,
"loss": 3.4747,
"step": 273000
},
{
"epoch": 11.528896008093412,
"grad_norm": 1.2484766244888306,
"learning_rate": 5.480855807571925e-05,
"loss": 3.4781,
"step": 273500
},
{
"epoch": 11.549972600429962,
"grad_norm": 1.2800825834274292,
"learning_rate": 5.472280581400334e-05,
"loss": 3.4748,
"step": 274000
},
{
"epoch": 11.571049192766514,
"grad_norm": 1.2782175540924072,
"learning_rate": 5.463705355228744e-05,
"loss": 3.4739,
"step": 274500
},
{
"epoch": 11.592125785103065,
"grad_norm": 1.2547928094863892,
"learning_rate": 5.4551301290571546e-05,
"loss": 3.4768,
"step": 275000
},
{
"epoch": 11.613202377439615,
"grad_norm": 1.2704449892044067,
"learning_rate": 5.446554902885563e-05,
"loss": 3.4736,
"step": 275500
},
{
"epoch": 11.634278969776167,
"grad_norm": 1.246609091758728,
"learning_rate": 5.4379796767139736e-05,
"loss": 3.4737,
"step": 276000
},
{
"epoch": 11.655355562112717,
"grad_norm": 1.3096011877059937,
"learning_rate": 5.4294044505423834e-05,
"loss": 3.4723,
"step": 276500
},
{
"epoch": 11.67643215444927,
"grad_norm": 1.276078701019287,
"learning_rate": 5.4208292243707925e-05,
"loss": 3.4733,
"step": 277000
},
{
"epoch": 11.69750874678582,
"grad_norm": 1.2850346565246582,
"learning_rate": 5.4122539981992024e-05,
"loss": 3.4732,
"step": 277500
},
{
"epoch": 11.71858533912237,
"grad_norm": 1.300667643547058,
"learning_rate": 5.403678772027613e-05,
"loss": 3.4675,
"step": 278000
},
{
"epoch": 11.739661931458922,
"grad_norm": 1.3072844743728638,
"learning_rate": 5.395103545856022e-05,
"loss": 3.471,
"step": 278500
},
{
"epoch": 11.760738523795473,
"grad_norm": 1.3241676092147827,
"learning_rate": 5.386528319684432e-05,
"loss": 3.4695,
"step": 279000
},
{
"epoch": 11.781815116132023,
"grad_norm": 1.301119327545166,
"learning_rate": 5.377953093512842e-05,
"loss": 3.4657,
"step": 279500
},
{
"epoch": 11.802891708468575,
"grad_norm": 1.3054083585739136,
"learning_rate": 5.369377867341251e-05,
"loss": 3.4753,
"step": 280000
},
{
"epoch": 11.802891708468575,
"eval_accuracy": 0.712901069110058,
"eval_loss": 3.2171669006347656,
"eval_runtime": 149.9438,
"eval_samples_per_second": 805.008,
"eval_steps_per_second": 8.39,
"step": 280000
},
{
"epoch": 11.823968300805126,
"grad_norm": 1.3228172063827515,
"learning_rate": 5.3608026411696614e-05,
"loss": 3.4753,
"step": 280500
},
{
"epoch": 11.845044893141678,
"grad_norm": 1.331397533416748,
"learning_rate": 5.352227414998071e-05,
"loss": 3.4652,
"step": 281000
},
{
"epoch": 11.866121485478228,
"grad_norm": 1.334751009941101,
"learning_rate": 5.34365218882648e-05,
"loss": 3.4689,
"step": 281500
},
{
"epoch": 11.887198077814778,
"grad_norm": 1.3301459550857544,
"learning_rate": 5.33507696265489e-05,
"loss": 3.4636,
"step": 282000
},
{
"epoch": 11.90827467015133,
"grad_norm": 1.2765735387802124,
"learning_rate": 5.3265017364833e-05,
"loss": 3.4726,
"step": 282500
},
{
"epoch": 11.92935126248788,
"grad_norm": 1.3712774515151978,
"learning_rate": 5.317926510311709e-05,
"loss": 3.4668,
"step": 283000
},
{
"epoch": 11.950427854824433,
"grad_norm": 1.2732222080230713,
"learning_rate": 5.3093512841401197e-05,
"loss": 3.4675,
"step": 283500
},
{
"epoch": 11.971504447160983,
"grad_norm": 1.3536001443862915,
"learning_rate": 5.3007760579685295e-05,
"loss": 3.47,
"step": 284000
},
{
"epoch": 11.992581039497534,
"grad_norm": 1.2974553108215332,
"learning_rate": 5.2922008317969386e-05,
"loss": 3.4723,
"step": 284500
},
{
"epoch": 12.013657631834086,
"grad_norm": 1.286482334136963,
"learning_rate": 5.2836256056253485e-05,
"loss": 3.4671,
"step": 285000
},
{
"epoch": 12.034734224170636,
"grad_norm": 1.3137521743774414,
"learning_rate": 5.275050379453759e-05,
"loss": 3.4688,
"step": 285500
},
{
"epoch": 12.055810816507186,
"grad_norm": 1.2509976625442505,
"learning_rate": 5.2664751532821675e-05,
"loss": 3.4643,
"step": 286000
},
{
"epoch": 12.076887408843739,
"grad_norm": 1.3011324405670166,
"learning_rate": 5.257899927110578e-05,
"loss": 3.4635,
"step": 286500
},
{
"epoch": 12.097964001180289,
"grad_norm": 1.2985036373138428,
"learning_rate": 5.249324700938988e-05,
"loss": 3.4622,
"step": 287000
},
{
"epoch": 12.119040593516841,
"grad_norm": 1.2630912065505981,
"learning_rate": 5.240749474767397e-05,
"loss": 3.4598,
"step": 287500
},
{
"epoch": 12.140117185853391,
"grad_norm": 1.35895836353302,
"learning_rate": 5.232174248595807e-05,
"loss": 3.4645,
"step": 288000
},
{
"epoch": 12.161193778189942,
"grad_norm": 1.255813479423523,
"learning_rate": 5.223599022424217e-05,
"loss": 3.4662,
"step": 288500
},
{
"epoch": 12.182270370526494,
"grad_norm": 1.3383607864379883,
"learning_rate": 5.215023796252626e-05,
"loss": 3.4694,
"step": 289000
},
{
"epoch": 12.203346962863044,
"grad_norm": 1.274146318435669,
"learning_rate": 5.206448570081036e-05,
"loss": 3.4609,
"step": 289500
},
{
"epoch": 12.224423555199595,
"grad_norm": 1.307592511177063,
"learning_rate": 5.197873343909446e-05,
"loss": 3.46,
"step": 290000
},
{
"epoch": 12.224423555199595,
"eval_accuracy": 0.714113845608957,
"eval_loss": 3.2101588249206543,
"eval_runtime": 147.2751,
"eval_samples_per_second": 819.595,
"eval_steps_per_second": 8.542,
"step": 290000
},
{
"epoch": 12.245500147536147,
"grad_norm": 1.346899151802063,
"learning_rate": 5.189298117737855e-05,
"loss": 3.4614,
"step": 290500
},
{
"epoch": 12.266576739872697,
"grad_norm": 1.3044058084487915,
"learning_rate": 5.180722891566265e-05,
"loss": 3.4594,
"step": 291000
},
{
"epoch": 12.28765333220925,
"grad_norm": 1.326993703842163,
"learning_rate": 5.1721476653946756e-05,
"loss": 3.4588,
"step": 291500
},
{
"epoch": 12.3087299245458,
"grad_norm": 1.3968826532363892,
"learning_rate": 5.163572439223085e-05,
"loss": 3.4611,
"step": 292000
},
{
"epoch": 12.32980651688235,
"grad_norm": 1.366346836090088,
"learning_rate": 5.1549972130514946e-05,
"loss": 3.4621,
"step": 292500
},
{
"epoch": 12.350883109218902,
"grad_norm": 1.3223849534988403,
"learning_rate": 5.1464219868799044e-05,
"loss": 3.4614,
"step": 293000
},
{
"epoch": 12.371959701555452,
"grad_norm": 1.3395236730575562,
"learning_rate": 5.1378467607083136e-05,
"loss": 3.4599,
"step": 293500
},
{
"epoch": 12.393036293892003,
"grad_norm": 1.2722975015640259,
"learning_rate": 5.1292715345367234e-05,
"loss": 3.4605,
"step": 294000
},
{
"epoch": 12.414112886228555,
"grad_norm": 1.2914930582046509,
"learning_rate": 5.120696308365134e-05,
"loss": 3.4568,
"step": 294500
},
{
"epoch": 12.435189478565105,
"grad_norm": 1.2855923175811768,
"learning_rate": 5.112121082193543e-05,
"loss": 3.4618,
"step": 295000
},
{
"epoch": 12.456266070901657,
"grad_norm": 1.2978049516677856,
"learning_rate": 5.103545856021953e-05,
"loss": 3.4586,
"step": 295500
},
{
"epoch": 12.477342663238208,
"grad_norm": 1.3203643560409546,
"learning_rate": 5.094970629850363e-05,
"loss": 3.4588,
"step": 296000
},
{
"epoch": 12.498419255574758,
"grad_norm": 1.3291692733764648,
"learning_rate": 5.086395403678772e-05,
"loss": 3.459,
"step": 296500
},
{
"epoch": 12.51949584791131,
"grad_norm": 1.3098422288894653,
"learning_rate": 5.0778201775071824e-05,
"loss": 3.4653,
"step": 297000
},
{
"epoch": 12.54057244024786,
"grad_norm": 1.4105501174926758,
"learning_rate": 5.069244951335592e-05,
"loss": 3.4613,
"step": 297500
},
{
"epoch": 12.561649032584413,
"grad_norm": 1.3970270156860352,
"learning_rate": 5.0606697251640014e-05,
"loss": 3.4547,
"step": 298000
},
{
"epoch": 12.582725624920963,
"grad_norm": 1.3700406551361084,
"learning_rate": 5.052094498992411e-05,
"loss": 3.4596,
"step": 298500
},
{
"epoch": 12.603802217257513,
"grad_norm": 1.3638668060302734,
"learning_rate": 5.043519272820822e-05,
"loss": 3.4609,
"step": 299000
},
{
"epoch": 12.624878809594065,
"grad_norm": 1.4138298034667969,
"learning_rate": 5.03494404664923e-05,
"loss": 3.4538,
"step": 299500
},
{
"epoch": 12.645955401930616,
"grad_norm": 1.3525958061218262,
"learning_rate": 5.026368820477641e-05,
"loss": 3.457,
"step": 300000
},
{
"epoch": 12.645955401930616,
"eval_accuracy": 0.7153695036676233,
"eval_loss": 3.2041077613830566,
"eval_runtime": 147.7442,
"eval_samples_per_second": 816.993,
"eval_steps_per_second": 8.515,
"step": 300000
},
{
"epoch": 12.667031994267166,
"grad_norm": 1.3221020698547363,
"learning_rate": 5.0177935943060505e-05,
"loss": 3.452,
"step": 300500
},
{
"epoch": 12.688108586603718,
"grad_norm": 1.3619519472122192,
"learning_rate": 5.00921836813446e-05,
"loss": 3.4602,
"step": 301000
},
{
"epoch": 12.709185178940269,
"grad_norm": 1.3002300262451172,
"learning_rate": 5.0006431419628695e-05,
"loss": 3.4565,
"step": 301500
},
{
"epoch": 12.73026177127682,
"grad_norm": 1.3111191987991333,
"learning_rate": 4.992067915791279e-05,
"loss": 3.4563,
"step": 302000
},
{
"epoch": 12.751338363613371,
"grad_norm": 1.307437539100647,
"learning_rate": 4.9834926896196885e-05,
"loss": 3.4507,
"step": 302500
},
{
"epoch": 12.772414955949921,
"grad_norm": 1.3757301568984985,
"learning_rate": 4.974917463448099e-05,
"loss": 3.4585,
"step": 303000
},
{
"epoch": 12.793491548286474,
"grad_norm": 1.3252778053283691,
"learning_rate": 4.966342237276508e-05,
"loss": 3.4582,
"step": 303500
},
{
"epoch": 12.814568140623024,
"grad_norm": 1.3005766868591309,
"learning_rate": 4.957767011104918e-05,
"loss": 3.4559,
"step": 304000
},
{
"epoch": 12.835644732959576,
"grad_norm": 1.3172825574874878,
"learning_rate": 4.949191784933328e-05,
"loss": 3.4542,
"step": 304500
},
{
"epoch": 12.856721325296126,
"grad_norm": 1.3249845504760742,
"learning_rate": 4.9406165587617376e-05,
"loss": 3.4537,
"step": 305000
},
{
"epoch": 12.877797917632677,
"grad_norm": 1.3311492204666138,
"learning_rate": 4.932041332590147e-05,
"loss": 3.4576,
"step": 305500
},
{
"epoch": 12.898874509969229,
"grad_norm": 1.3406810760498047,
"learning_rate": 4.923466106418557e-05,
"loss": 3.4519,
"step": 306000
},
{
"epoch": 12.91995110230578,
"grad_norm": 1.3221213817596436,
"learning_rate": 4.9148908802469665e-05,
"loss": 3.4508,
"step": 306500
},
{
"epoch": 12.94102769464233,
"grad_norm": 1.294512152671814,
"learning_rate": 4.906315654075376e-05,
"loss": 3.4518,
"step": 307000
},
{
"epoch": 12.962104286978882,
"grad_norm": 1.3368299007415771,
"learning_rate": 4.897740427903786e-05,
"loss": 3.4582,
"step": 307500
},
{
"epoch": 12.983180879315432,
"grad_norm": 1.3361867666244507,
"learning_rate": 4.889165201732196e-05,
"loss": 3.4578,
"step": 308000
},
{
"epoch": 13.004257471651984,
"grad_norm": 1.3266749382019043,
"learning_rate": 4.880589975560606e-05,
"loss": 3.4526,
"step": 308500
},
{
"epoch": 13.025334063988534,
"grad_norm": 1.3374311923980713,
"learning_rate": 4.8720147493890156e-05,
"loss": 3.4457,
"step": 309000
},
{
"epoch": 13.046410656325085,
"grad_norm": 1.3623027801513672,
"learning_rate": 4.8634395232174254e-05,
"loss": 3.4516,
"step": 309500
},
{
"epoch": 13.067487248661637,
"grad_norm": 1.295546293258667,
"learning_rate": 4.8548642970458346e-05,
"loss": 3.4464,
"step": 310000
},
{
"epoch": 13.067487248661637,
"eval_accuracy": 0.7163273673839559,
"eval_loss": 3.198410749435425,
"eval_runtime": 147.5661,
"eval_samples_per_second": 817.979,
"eval_steps_per_second": 8.525,
"step": 310000
},
{
"epoch": 13.088563840998187,
"grad_norm": 1.3205565214157104,
"learning_rate": 4.846289070874245e-05,
"loss": 3.4497,
"step": 310500
},
{
"epoch": 13.109640433334738,
"grad_norm": 1.3347554206848145,
"learning_rate": 4.837713844702654e-05,
"loss": 3.451,
"step": 311000
},
{
"epoch": 13.13071702567129,
"grad_norm": 1.359351634979248,
"learning_rate": 4.829138618531064e-05,
"loss": 3.4462,
"step": 311500
},
{
"epoch": 13.15179361800784,
"grad_norm": 1.357680320739746,
"learning_rate": 4.820563392359474e-05,
"loss": 3.4477,
"step": 312000
},
{
"epoch": 13.172870210344392,
"grad_norm": 1.3265962600708008,
"learning_rate": 4.811988166187884e-05,
"loss": 3.4458,
"step": 312500
},
{
"epoch": 13.193946802680943,
"grad_norm": 1.3435533046722412,
"learning_rate": 4.803412940016293e-05,
"loss": 3.4466,
"step": 313000
},
{
"epoch": 13.215023395017493,
"grad_norm": 1.3250727653503418,
"learning_rate": 4.7948377138447034e-05,
"loss": 3.4491,
"step": 313500
},
{
"epoch": 13.236099987354045,
"grad_norm": 1.3472198247909546,
"learning_rate": 4.7862624876731126e-05,
"loss": 3.4442,
"step": 314000
},
{
"epoch": 13.257176579690595,
"grad_norm": 1.302666187286377,
"learning_rate": 4.7776872615015224e-05,
"loss": 3.446,
"step": 314500
},
{
"epoch": 13.278253172027147,
"grad_norm": 1.2951545715332031,
"learning_rate": 4.7691120353299315e-05,
"loss": 3.4427,
"step": 315000
},
{
"epoch": 13.299329764363698,
"grad_norm": 1.2792922258377075,
"learning_rate": 4.760536809158342e-05,
"loss": 3.4505,
"step": 315500
},
{
"epoch": 13.320406356700248,
"grad_norm": 1.3151403665542603,
"learning_rate": 4.751961582986751e-05,
"loss": 3.4481,
"step": 316000
},
{
"epoch": 13.3414829490368,
"grad_norm": 1.3079912662506104,
"learning_rate": 4.743386356815161e-05,
"loss": 3.4487,
"step": 316500
},
{
"epoch": 13.36255954137335,
"grad_norm": 1.2841126918792725,
"learning_rate": 4.734811130643571e-05,
"loss": 3.4467,
"step": 317000
},
{
"epoch": 13.383636133709901,
"grad_norm": 1.3274542093276978,
"learning_rate": 4.726235904471981e-05,
"loss": 3.4425,
"step": 317500
},
{
"epoch": 13.404712726046453,
"grad_norm": 1.3739765882492065,
"learning_rate": 4.71766067830039e-05,
"loss": 3.446,
"step": 318000
},
{
"epoch": 13.425789318383003,
"grad_norm": 1.3088661432266235,
"learning_rate": 4.7090854521288004e-05,
"loss": 3.4486,
"step": 318500
},
{
"epoch": 13.446865910719556,
"grad_norm": 1.3954898118972778,
"learning_rate": 4.7005102259572095e-05,
"loss": 3.4444,
"step": 319000
},
{
"epoch": 13.467942503056106,
"grad_norm": 1.3715546131134033,
"learning_rate": 4.6919349997856193e-05,
"loss": 3.4462,
"step": 319500
},
{
"epoch": 13.489019095392656,
"grad_norm": 1.3462576866149902,
"learning_rate": 4.683359773614029e-05,
"loss": 3.4446,
"step": 320000
},
{
"epoch": 13.489019095392656,
"eval_accuracy": 0.7170561937471841,
"eval_loss": 3.19331955909729,
"eval_runtime": 147.4841,
"eval_samples_per_second": 818.434,
"eval_steps_per_second": 8.53,
"step": 320000
},
{
"epoch": 13.510095687729208,
"grad_norm": 1.3218398094177246,
"learning_rate": 4.674784547442439e-05,
"loss": 3.4484,
"step": 320500
},
{
"epoch": 13.531172280065759,
"grad_norm": 1.3490931987762451,
"learning_rate": 4.666209321270849e-05,
"loss": 3.4445,
"step": 321000
},
{
"epoch": 13.55224887240231,
"grad_norm": 1.3118476867675781,
"learning_rate": 4.657634095099259e-05,
"loss": 3.4448,
"step": 321500
},
{
"epoch": 13.573325464738861,
"grad_norm": 1.3598856925964355,
"learning_rate": 4.6490588689276685e-05,
"loss": 3.445,
"step": 322000
},
{
"epoch": 13.594402057075412,
"grad_norm": 1.3177522420883179,
"learning_rate": 4.6404836427560777e-05,
"loss": 3.4463,
"step": 322500
},
{
"epoch": 13.615478649411964,
"grad_norm": 1.3502224683761597,
"learning_rate": 4.6319084165844875e-05,
"loss": 3.4385,
"step": 323000
},
{
"epoch": 13.636555241748514,
"grad_norm": 1.3368273973464966,
"learning_rate": 4.623333190412897e-05,
"loss": 3.4454,
"step": 323500
},
{
"epoch": 13.657631834085064,
"grad_norm": 1.3694145679473877,
"learning_rate": 4.614757964241307e-05,
"loss": 3.4412,
"step": 324000
},
{
"epoch": 13.678708426421617,
"grad_norm": 1.3193176984786987,
"learning_rate": 4.606182738069717e-05,
"loss": 3.4435,
"step": 324500
},
{
"epoch": 13.699785018758167,
"grad_norm": 1.3092237710952759,
"learning_rate": 4.597607511898127e-05,
"loss": 3.4434,
"step": 325000
},
{
"epoch": 13.720861611094719,
"grad_norm": 1.32658851146698,
"learning_rate": 4.589032285726536e-05,
"loss": 3.4415,
"step": 325500
},
{
"epoch": 13.74193820343127,
"grad_norm": 1.2967160940170288,
"learning_rate": 4.5804570595549465e-05,
"loss": 3.4426,
"step": 326000
},
{
"epoch": 13.76301479576782,
"grad_norm": 1.2876520156860352,
"learning_rate": 4.5718818333833556e-05,
"loss": 3.4455,
"step": 326500
},
{
"epoch": 13.784091388104372,
"grad_norm": 1.3992048501968384,
"learning_rate": 4.5633066072117655e-05,
"loss": 3.4445,
"step": 327000
},
{
"epoch": 13.805167980440922,
"grad_norm": 1.2999306917190552,
"learning_rate": 4.554731381040175e-05,
"loss": 3.4385,
"step": 327500
},
{
"epoch": 13.826244572777473,
"grad_norm": 1.345624327659607,
"learning_rate": 4.546156154868585e-05,
"loss": 3.4423,
"step": 328000
},
{
"epoch": 13.847321165114025,
"grad_norm": 1.3467533588409424,
"learning_rate": 4.537580928696994e-05,
"loss": 3.4402,
"step": 328500
},
{
"epoch": 13.868397757450575,
"grad_norm": 1.3431648015975952,
"learning_rate": 4.529005702525405e-05,
"loss": 3.4406,
"step": 329000
},
{
"epoch": 13.889474349787127,
"grad_norm": 1.3668323755264282,
"learning_rate": 4.520430476353814e-05,
"loss": 3.4384,
"step": 329500
},
{
"epoch": 13.910550942123677,
"grad_norm": 1.3732104301452637,
"learning_rate": 4.511855250182224e-05,
"loss": 3.4398,
"step": 330000
},
{
"epoch": 13.910550942123677,
"eval_accuracy": 0.7174077938302867,
"eval_loss": 3.1897125244140625,
"eval_runtime": 147.6608,
"eval_samples_per_second": 817.455,
"eval_steps_per_second": 8.52,
"step": 330000
},
{
"epoch": 13.931627534460228,
"grad_norm": 1.3022862672805786,
"learning_rate": 4.5032800240106336e-05,
"loss": 3.4413,
"step": 330500
},
{
"epoch": 13.95270412679678,
"grad_norm": 1.3187073469161987,
"learning_rate": 4.4947047978390434e-05,
"loss": 3.4436,
"step": 331000
},
{
"epoch": 13.97378071913333,
"grad_norm": 1.3251093626022339,
"learning_rate": 4.4861295716674526e-05,
"loss": 3.4406,
"step": 331500
},
{
"epoch": 13.994857311469882,
"grad_norm": 1.3421084880828857,
"learning_rate": 4.4775543454958624e-05,
"loss": 3.442,
"step": 332000
},
{
"epoch": 14.015933903806433,
"grad_norm": 1.332471251487732,
"learning_rate": 4.468979119324272e-05,
"loss": 3.4362,
"step": 332500
},
{
"epoch": 14.037010496142983,
"grad_norm": 1.3337472677230835,
"learning_rate": 4.460403893152682e-05,
"loss": 3.4357,
"step": 333000
},
{
"epoch": 14.058087088479535,
"grad_norm": 1.2947136163711548,
"learning_rate": 4.451828666981092e-05,
"loss": 3.4365,
"step": 333500
},
{
"epoch": 14.079163680816086,
"grad_norm": 1.3384618759155273,
"learning_rate": 4.443253440809502e-05,
"loss": 3.4364,
"step": 334000
},
{
"epoch": 14.100240273152636,
"grad_norm": 1.3033937215805054,
"learning_rate": 4.4346782146379116e-05,
"loss": 3.4354,
"step": 334500
},
{
"epoch": 14.121316865489188,
"grad_norm": 1.3284544944763184,
"learning_rate": 4.426102988466321e-05,
"loss": 3.4324,
"step": 335000
},
{
"epoch": 14.142393457825738,
"grad_norm": 1.3674243688583374,
"learning_rate": 4.4175277622947305e-05,
"loss": 3.4282,
"step": 335500
},
{
"epoch": 14.16347005016229,
"grad_norm": 1.3452162742614746,
"learning_rate": 4.4089525361231404e-05,
"loss": 3.4343,
"step": 336000
},
{
"epoch": 14.184546642498841,
"grad_norm": 1.3089090585708618,
"learning_rate": 4.40037730995155e-05,
"loss": 3.4296,
"step": 336500
},
{
"epoch": 14.205623234835391,
"grad_norm": 1.4065284729003906,
"learning_rate": 4.39180208377996e-05,
"loss": 3.434,
"step": 337000
},
{
"epoch": 14.226699827171943,
"grad_norm": 1.3471928834915161,
"learning_rate": 4.38322685760837e-05,
"loss": 3.4343,
"step": 337500
},
{
"epoch": 14.247776419508494,
"grad_norm": 1.3715170621871948,
"learning_rate": 4.374651631436779e-05,
"loss": 3.4344,
"step": 338000
},
{
"epoch": 14.268853011845044,
"grad_norm": 1.3081505298614502,
"learning_rate": 4.3660764052651895e-05,
"loss": 3.4359,
"step": 338500
},
{
"epoch": 14.289929604181596,
"grad_norm": 1.3681291341781616,
"learning_rate": 4.357501179093599e-05,
"loss": 3.4343,
"step": 339000
},
{
"epoch": 14.311006196518147,
"grad_norm": 1.422668695449829,
"learning_rate": 4.3489259529220085e-05,
"loss": 3.4339,
"step": 339500
},
{
"epoch": 14.332082788854699,
"grad_norm": 1.4056037664413452,
"learning_rate": 4.3403507267504183e-05,
"loss": 3.436,
"step": 340000
},
{
"epoch": 14.332082788854699,
"eval_accuracy": 0.7184626252970217,
"eval_loss": 3.1837987899780273,
"eval_runtime": 147.8542,
"eval_samples_per_second": 816.385,
"eval_steps_per_second": 8.508,
"step": 340000
},
{
"epoch": 14.353159381191249,
"grad_norm": 1.358860731124878,
"learning_rate": 4.331775500578828e-05,
"loss": 3.433,
"step": 340500
},
{
"epoch": 14.3742359735278,
"grad_norm": 1.3141694068908691,
"learning_rate": 4.323200274407237e-05,
"loss": 3.4315,
"step": 341000
},
{
"epoch": 14.395312565864351,
"grad_norm": 1.3611981868743896,
"learning_rate": 4.314625048235648e-05,
"loss": 3.4309,
"step": 341500
},
{
"epoch": 14.416389158200902,
"grad_norm": 1.3787122964859009,
"learning_rate": 4.306049822064057e-05,
"loss": 3.4341,
"step": 342000
},
{
"epoch": 14.437465750537454,
"grad_norm": 1.3799214363098145,
"learning_rate": 4.297474595892467e-05,
"loss": 3.434,
"step": 342500
},
{
"epoch": 14.458542342874004,
"grad_norm": 1.3481985330581665,
"learning_rate": 4.2888993697208767e-05,
"loss": 3.4289,
"step": 343000
},
{
"epoch": 14.479618935210555,
"grad_norm": 1.3024696111679077,
"learning_rate": 4.2803241435492865e-05,
"loss": 3.433,
"step": 343500
},
{
"epoch": 14.500695527547107,
"grad_norm": 1.3263978958129883,
"learning_rate": 4.2717489173776956e-05,
"loss": 3.4323,
"step": 344000
},
{
"epoch": 14.521772119883657,
"grad_norm": 1.3351951837539673,
"learning_rate": 4.263173691206106e-05,
"loss": 3.4302,
"step": 344500
},
{
"epoch": 14.542848712220207,
"grad_norm": 1.3777052164077759,
"learning_rate": 4.254598465034515e-05,
"loss": 3.4326,
"step": 345000
},
{
"epoch": 14.56392530455676,
"grad_norm": 1.3416668176651,
"learning_rate": 4.246023238862925e-05,
"loss": 3.4288,
"step": 345500
},
{
"epoch": 14.58500189689331,
"grad_norm": 1.3714169263839722,
"learning_rate": 4.237448012691335e-05,
"loss": 3.4303,
"step": 346000
},
{
"epoch": 14.606078489229862,
"grad_norm": 1.4569975137710571,
"learning_rate": 4.228872786519745e-05,
"loss": 3.4291,
"step": 346500
},
{
"epoch": 14.627155081566412,
"grad_norm": 1.342958688735962,
"learning_rate": 4.220297560348154e-05,
"loss": 3.4319,
"step": 347000
},
{
"epoch": 14.648231673902963,
"grad_norm": 1.375974178314209,
"learning_rate": 4.2117223341765645e-05,
"loss": 3.4298,
"step": 347500
},
{
"epoch": 14.669308266239515,
"grad_norm": 1.3035759925842285,
"learning_rate": 4.2031471080049736e-05,
"loss": 3.4287,
"step": 348000
},
{
"epoch": 14.690384858576065,
"grad_norm": 1.3770883083343506,
"learning_rate": 4.1945718818333834e-05,
"loss": 3.4292,
"step": 348500
},
{
"epoch": 14.711461450912616,
"grad_norm": 1.3159732818603516,
"learning_rate": 4.185996655661793e-05,
"loss": 3.4313,
"step": 349000
},
{
"epoch": 14.732538043249168,
"grad_norm": 1.313312292098999,
"learning_rate": 4.177421429490203e-05,
"loss": 3.4321,
"step": 349500
},
{
"epoch": 14.753614635585718,
"grad_norm": 1.385988712310791,
"learning_rate": 4.168846203318613e-05,
"loss": 3.4289,
"step": 350000
},
{
"epoch": 14.753614635585718,
"eval_accuracy": 0.719255176728657,
"eval_loss": 3.178434133529663,
"eval_runtime": 147.5431,
"eval_samples_per_second": 818.107,
"eval_steps_per_second": 8.526,
"step": 350000
},
{
"epoch": 14.77469122792227,
"grad_norm": 1.309218406677246,
"learning_rate": 4.160270977147022e-05,
"loss": 3.4279,
"step": 350500
},
{
"epoch": 14.79576782025882,
"grad_norm": 1.325492262840271,
"learning_rate": 4.1516957509754326e-05,
"loss": 3.4306,
"step": 351000
},
{
"epoch": 14.81684441259537,
"grad_norm": 1.2865612506866455,
"learning_rate": 4.143120524803842e-05,
"loss": 3.4254,
"step": 351500
},
{
"epoch": 14.837921004931923,
"grad_norm": 1.3227702379226685,
"learning_rate": 4.1345452986322516e-05,
"loss": 3.4276,
"step": 352000
},
{
"epoch": 14.858997597268473,
"grad_norm": 1.3126128911972046,
"learning_rate": 4.1259700724606614e-05,
"loss": 3.4267,
"step": 352500
},
{
"epoch": 14.880074189605025,
"grad_norm": 1.3655269145965576,
"learning_rate": 4.117394846289071e-05,
"loss": 3.4256,
"step": 353000
},
{
"epoch": 14.901150781941576,
"grad_norm": 1.3058969974517822,
"learning_rate": 4.1088196201174804e-05,
"loss": 3.4284,
"step": 353500
},
{
"epoch": 14.922227374278126,
"grad_norm": 1.3704140186309814,
"learning_rate": 4.100244393945891e-05,
"loss": 3.4267,
"step": 354000
},
{
"epoch": 14.943303966614678,
"grad_norm": 1.3632330894470215,
"learning_rate": 4.0916691677743e-05,
"loss": 3.4323,
"step": 354500
},
{
"epoch": 14.964380558951229,
"grad_norm": 1.247710108757019,
"learning_rate": 4.08309394160271e-05,
"loss": 3.4286,
"step": 355000
},
{
"epoch": 14.985457151287779,
"grad_norm": 1.3704109191894531,
"learning_rate": 4.07451871543112e-05,
"loss": 3.4263,
"step": 355500
},
{
"epoch": 15.006533743624331,
"grad_norm": 1.3445990085601807,
"learning_rate": 4.0659434892595295e-05,
"loss": 3.4219,
"step": 356000
},
{
"epoch": 15.027610335960881,
"grad_norm": 1.3786925077438354,
"learning_rate": 4.057368263087939e-05,
"loss": 3.4188,
"step": 356500
},
{
"epoch": 15.048686928297434,
"grad_norm": 1.3428871631622314,
"learning_rate": 4.048793036916349e-05,
"loss": 3.4231,
"step": 357000
},
{
"epoch": 15.069763520633984,
"grad_norm": 1.3536442518234253,
"learning_rate": 4.0402178107447584e-05,
"loss": 3.4252,
"step": 357500
},
{
"epoch": 15.090840112970534,
"grad_norm": 1.3826103210449219,
"learning_rate": 4.031642584573168e-05,
"loss": 3.4243,
"step": 358000
},
{
"epoch": 15.111916705307086,
"grad_norm": 1.3791228532791138,
"learning_rate": 4.023067358401578e-05,
"loss": 3.4198,
"step": 358500
},
{
"epoch": 15.132993297643637,
"grad_norm": 1.393920660018921,
"learning_rate": 4.014492132229988e-05,
"loss": 3.4246,
"step": 359000
},
{
"epoch": 15.154069889980189,
"grad_norm": 1.355263590812683,
"learning_rate": 4.005916906058397e-05,
"loss": 3.4217,
"step": 359500
},
{
"epoch": 15.17514648231674,
"grad_norm": 1.3142739534378052,
"learning_rate": 3.9973416798868075e-05,
"loss": 3.4223,
"step": 360000
},
{
"epoch": 15.17514648231674,
"eval_accuracy": 0.7198432125887803,
"eval_loss": 3.1747541427612305,
"eval_runtime": 147.163,
"eval_samples_per_second": 820.22,
"eval_steps_per_second": 8.548,
"step": 360000
},
{
"epoch": 15.19622307465329,
"grad_norm": 1.3669103384017944,
"learning_rate": 3.988766453715217e-05,
"loss": 3.4251,
"step": 360500
},
{
"epoch": 15.217299666989842,
"grad_norm": 1.3445587158203125,
"learning_rate": 3.9801912275436265e-05,
"loss": 3.4228,
"step": 361000
},
{
"epoch": 15.238376259326392,
"grad_norm": 1.3381707668304443,
"learning_rate": 3.971616001372036e-05,
"loss": 3.4232,
"step": 361500
},
{
"epoch": 15.259452851662942,
"grad_norm": 1.38563871383667,
"learning_rate": 3.963040775200446e-05,
"loss": 3.4215,
"step": 362000
},
{
"epoch": 15.280529443999495,
"grad_norm": 1.3533085584640503,
"learning_rate": 3.954465549028856e-05,
"loss": 3.4207,
"step": 362500
},
{
"epoch": 15.301606036336045,
"grad_norm": 1.3468925952911377,
"learning_rate": 3.945890322857266e-05,
"loss": 3.4219,
"step": 363000
},
{
"epoch": 15.322682628672597,
"grad_norm": 1.3068883419036865,
"learning_rate": 3.9373150966856756e-05,
"loss": 3.4181,
"step": 363500
},
{
"epoch": 15.343759221009147,
"grad_norm": 1.3848633766174316,
"learning_rate": 3.928739870514085e-05,
"loss": 3.4214,
"step": 364000
},
{
"epoch": 15.364835813345698,
"grad_norm": 1.3424211740493774,
"learning_rate": 3.920164644342495e-05,
"loss": 3.418,
"step": 364500
},
{
"epoch": 15.38591240568225,
"grad_norm": 1.3467830419540405,
"learning_rate": 3.9115894181709045e-05,
"loss": 3.4198,
"step": 365000
},
{
"epoch": 15.4069889980188,
"grad_norm": 1.338295578956604,
"learning_rate": 3.903014191999314e-05,
"loss": 3.4174,
"step": 365500
},
{
"epoch": 15.42806559035535,
"grad_norm": 1.3510428667068481,
"learning_rate": 3.894438965827724e-05,
"loss": 3.4216,
"step": 366000
},
{
"epoch": 15.449142182691903,
"grad_norm": 1.3427857160568237,
"learning_rate": 3.885863739656134e-05,
"loss": 3.4197,
"step": 366500
},
{
"epoch": 15.470218775028453,
"grad_norm": 1.3441849946975708,
"learning_rate": 3.877288513484543e-05,
"loss": 3.4188,
"step": 367000
},
{
"epoch": 15.491295367365005,
"grad_norm": 1.3698375225067139,
"learning_rate": 3.868713287312953e-05,
"loss": 3.4212,
"step": 367500
},
{
"epoch": 15.512371959701555,
"grad_norm": 1.3777368068695068,
"learning_rate": 3.860138061141363e-05,
"loss": 3.4192,
"step": 368000
},
{
"epoch": 15.533448552038106,
"grad_norm": 1.31421959400177,
"learning_rate": 3.8515628349697726e-05,
"loss": 3.4209,
"step": 368500
},
{
"epoch": 15.554525144374658,
"grad_norm": 1.3805230855941772,
"learning_rate": 3.842987608798182e-05,
"loss": 3.4179,
"step": 369000
},
{
"epoch": 15.575601736711208,
"grad_norm": 1.4013229608535767,
"learning_rate": 3.834412382626592e-05,
"loss": 3.4148,
"step": 369500
},
{
"epoch": 15.596678329047759,
"grad_norm": 1.3337624073028564,
"learning_rate": 3.8258371564550014e-05,
"loss": 3.4187,
"step": 370000
},
{
"epoch": 15.596678329047759,
"eval_accuracy": 0.7208415299584509,
"eval_loss": 3.1676342487335205,
"eval_runtime": 147.5879,
"eval_samples_per_second": 817.858,
"eval_steps_per_second": 8.524,
"step": 370000
},
{
"epoch": 15.61775492138431,
"grad_norm": 1.3824350833892822,
"learning_rate": 3.817261930283411e-05,
"loss": 3.415,
"step": 370500
},
{
"epoch": 15.638831513720861,
"grad_norm": 1.4466607570648193,
"learning_rate": 3.808686704111821e-05,
"loss": 3.4175,
"step": 371000
},
{
"epoch": 15.659908106057413,
"grad_norm": 1.3693633079528809,
"learning_rate": 3.800111477940231e-05,
"loss": 3.418,
"step": 371500
},
{
"epoch": 15.680984698393964,
"grad_norm": 1.3618525266647339,
"learning_rate": 3.79153625176864e-05,
"loss": 3.4193,
"step": 372000
},
{
"epoch": 15.702061290730514,
"grad_norm": 1.332354187965393,
"learning_rate": 3.7829610255970506e-05,
"loss": 3.4203,
"step": 372500
},
{
"epoch": 15.723137883067066,
"grad_norm": 1.3403964042663574,
"learning_rate": 3.77438579942546e-05,
"loss": 3.4153,
"step": 373000
},
{
"epoch": 15.744214475403616,
"grad_norm": 1.3866877555847168,
"learning_rate": 3.7658105732538696e-05,
"loss": 3.4139,
"step": 373500
},
{
"epoch": 15.765291067740169,
"grad_norm": 1.317996621131897,
"learning_rate": 3.7572353470822794e-05,
"loss": 3.4172,
"step": 374000
},
{
"epoch": 15.786367660076719,
"grad_norm": 1.4216530323028564,
"learning_rate": 3.748660120910689e-05,
"loss": 3.4124,
"step": 374500
},
{
"epoch": 15.80744425241327,
"grad_norm": 1.3848369121551514,
"learning_rate": 3.740084894739099e-05,
"loss": 3.4114,
"step": 375000
},
{
"epoch": 15.828520844749821,
"grad_norm": 1.306089162826538,
"learning_rate": 3.731509668567509e-05,
"loss": 3.4142,
"step": 375500
},
{
"epoch": 15.849597437086372,
"grad_norm": 1.3504360914230347,
"learning_rate": 3.722934442395919e-05,
"loss": 3.4173,
"step": 376000
},
{
"epoch": 15.870674029422922,
"grad_norm": 1.3742388486862183,
"learning_rate": 3.714359216224328e-05,
"loss": 3.4158,
"step": 376500
},
{
"epoch": 15.891750621759474,
"grad_norm": 1.3441376686096191,
"learning_rate": 3.705783990052738e-05,
"loss": 3.4174,
"step": 377000
},
{
"epoch": 15.912827214096025,
"grad_norm": 1.3735742568969727,
"learning_rate": 3.6972087638811475e-05,
"loss": 3.4128,
"step": 377500
},
{
"epoch": 15.933903806432577,
"grad_norm": 1.3164881467819214,
"learning_rate": 3.6886335377095574e-05,
"loss": 3.4159,
"step": 378000
},
{
"epoch": 15.954980398769127,
"grad_norm": 1.3117059469223022,
"learning_rate": 3.680058311537967e-05,
"loss": 3.4141,
"step": 378500
},
{
"epoch": 15.976056991105677,
"grad_norm": 1.3854739665985107,
"learning_rate": 3.671483085366377e-05,
"loss": 3.4164,
"step": 379000
},
{
"epoch": 15.99713358344223,
"grad_norm": 1.3398085832595825,
"learning_rate": 3.662907859194786e-05,
"loss": 3.4138,
"step": 379500
},
{
"epoch": 16.01821017577878,
"grad_norm": 1.395106315612793,
"learning_rate": 3.654332633023197e-05,
"loss": 3.414,
"step": 380000
},
{
"epoch": 16.01821017577878,
"eval_accuracy": 0.7215908958856491,
"eval_loss": 3.165067672729492,
"eval_runtime": 147.3578,
"eval_samples_per_second": 819.135,
"eval_steps_per_second": 8.537,
"step": 380000
},
{
"epoch": 16.039286768115332,
"grad_norm": 1.3090741634368896,
"learning_rate": 3.645757406851606e-05,
"loss": 3.4119,
"step": 380500
},
{
"epoch": 16.06036336045188,
"grad_norm": 1.3652530908584595,
"learning_rate": 3.6371821806800157e-05,
"loss": 3.41,
"step": 381000
},
{
"epoch": 16.081439952788433,
"grad_norm": 1.300609827041626,
"learning_rate": 3.6286069545084255e-05,
"loss": 3.4138,
"step": 381500
},
{
"epoch": 16.102516545124985,
"grad_norm": 1.4015551805496216,
"learning_rate": 3.620031728336835e-05,
"loss": 3.4112,
"step": 382000
},
{
"epoch": 16.123593137461537,
"grad_norm": 1.3337161540985107,
"learning_rate": 3.6114565021652445e-05,
"loss": 3.4047,
"step": 382500
},
{
"epoch": 16.144669729798085,
"grad_norm": 1.3469023704528809,
"learning_rate": 3.602881275993655e-05,
"loss": 3.4077,
"step": 383000
},
{
"epoch": 16.165746322134638,
"grad_norm": 1.4076045751571655,
"learning_rate": 3.594306049822064e-05,
"loss": 3.413,
"step": 383500
},
{
"epoch": 16.18682291447119,
"grad_norm": 1.4201496839523315,
"learning_rate": 3.585730823650474e-05,
"loss": 3.4092,
"step": 384000
},
{
"epoch": 16.20789950680774,
"grad_norm": 1.362922191619873,
"learning_rate": 3.577155597478883e-05,
"loss": 3.4113,
"step": 384500
},
{
"epoch": 16.22897609914429,
"grad_norm": 1.365928292274475,
"learning_rate": 3.5685803713072936e-05,
"loss": 3.4048,
"step": 385000
},
{
"epoch": 16.250052691480843,
"grad_norm": 1.3617658615112305,
"learning_rate": 3.560005145135703e-05,
"loss": 3.4101,
"step": 385500
},
{
"epoch": 16.27112928381739,
"grad_norm": 1.3825161457061768,
"learning_rate": 3.5514299189641126e-05,
"loss": 3.4059,
"step": 386000
},
{
"epoch": 16.292205876153943,
"grad_norm": 1.3895723819732666,
"learning_rate": 3.5428546927925224e-05,
"loss": 3.4098,
"step": 386500
},
{
"epoch": 16.313282468490495,
"grad_norm": 1.3294148445129395,
"learning_rate": 3.534279466620932e-05,
"loss": 3.4056,
"step": 387000
},
{
"epoch": 16.334359060827044,
"grad_norm": 1.3587225675582886,
"learning_rate": 3.525704240449342e-05,
"loss": 3.408,
"step": 387500
},
{
"epoch": 16.355435653163596,
"grad_norm": 1.3962146043777466,
"learning_rate": 3.517129014277752e-05,
"loss": 3.4051,
"step": 388000
},
{
"epoch": 16.376512245500148,
"grad_norm": 1.4265996217727661,
"learning_rate": 3.508553788106161e-05,
"loss": 3.4073,
"step": 388500
},
{
"epoch": 16.3975888378367,
"grad_norm": 1.3339958190917969,
"learning_rate": 3.499978561934571e-05,
"loss": 3.4039,
"step": 389000
},
{
"epoch": 16.41866543017325,
"grad_norm": 1.3336193561553955,
"learning_rate": 3.491403335762981e-05,
"loss": 3.4064,
"step": 389500
},
{
"epoch": 16.4397420225098,
"grad_norm": 1.4063923358917236,
"learning_rate": 3.4828281095913906e-05,
"loss": 3.409,
"step": 390000
},
{
"epoch": 16.4397420225098,
"eval_accuracy": 0.7222337986061381,
"eval_loss": 3.1609442234039307,
"eval_runtime": 147.0322,
"eval_samples_per_second": 820.949,
"eval_steps_per_second": 8.556,
"step": 390000
},
{
"epoch": 16.460818614846353,
"grad_norm": 1.3616081476211548,
"learning_rate": 3.4742528834198004e-05,
"loss": 3.4057,
"step": 390500
},
{
"epoch": 16.4818952071829,
"grad_norm": 1.3544275760650635,
"learning_rate": 3.46567765724821e-05,
"loss": 3.4083,
"step": 391000
},
{
"epoch": 16.502971799519454,
"grad_norm": 1.404502034187317,
"learning_rate": 3.45710243107662e-05,
"loss": 3.4058,
"step": 391500
},
{
"epoch": 16.524048391856006,
"grad_norm": 1.3368873596191406,
"learning_rate": 3.448527204905029e-05,
"loss": 3.4111,
"step": 392000
},
{
"epoch": 16.545124984192555,
"grad_norm": 1.3857684135437012,
"learning_rate": 3.43995197873344e-05,
"loss": 3.405,
"step": 392500
},
{
"epoch": 16.566201576529107,
"grad_norm": 1.3709105253219604,
"learning_rate": 3.431376752561849e-05,
"loss": 3.4071,
"step": 393000
},
{
"epoch": 16.58727816886566,
"grad_norm": 1.3408377170562744,
"learning_rate": 3.422801526390259e-05,
"loss": 3.41,
"step": 393500
},
{
"epoch": 16.608354761202207,
"grad_norm": 1.365715742111206,
"learning_rate": 3.4142263002186686e-05,
"loss": 3.4082,
"step": 394000
},
{
"epoch": 16.62943135353876,
"grad_norm": 1.3507157564163208,
"learning_rate": 3.4056510740470784e-05,
"loss": 3.4102,
"step": 394500
},
{
"epoch": 16.65050794587531,
"grad_norm": 1.3823806047439575,
"learning_rate": 3.3970758478754875e-05,
"loss": 3.407,
"step": 395000
},
{
"epoch": 16.67158453821186,
"grad_norm": 1.3520289659500122,
"learning_rate": 3.388500621703898e-05,
"loss": 3.4071,
"step": 395500
},
{
"epoch": 16.692661130548412,
"grad_norm": 1.4460705518722534,
"learning_rate": 3.379925395532307e-05,
"loss": 3.4089,
"step": 396000
},
{
"epoch": 16.713737722884964,
"grad_norm": 1.335172414779663,
"learning_rate": 3.371350169360717e-05,
"loss": 3.4038,
"step": 396500
},
{
"epoch": 16.734814315221517,
"grad_norm": 1.3653130531311035,
"learning_rate": 3.362774943189127e-05,
"loss": 3.4083,
"step": 397000
},
{
"epoch": 16.755890907558065,
"grad_norm": 1.3100826740264893,
"learning_rate": 3.354199717017537e-05,
"loss": 3.4045,
"step": 397500
},
{
"epoch": 16.776967499894617,
"grad_norm": 1.3340774774551392,
"learning_rate": 3.345624490845946e-05,
"loss": 3.4043,
"step": 398000
},
{
"epoch": 16.79804409223117,
"grad_norm": 1.386819839477539,
"learning_rate": 3.3370492646743564e-05,
"loss": 3.4036,
"step": 398500
},
{
"epoch": 16.819120684567718,
"grad_norm": 1.3353698253631592,
"learning_rate": 3.3284740385027655e-05,
"loss": 3.4099,
"step": 399000
},
{
"epoch": 16.84019727690427,
"grad_norm": 1.3014289140701294,
"learning_rate": 3.319898812331175e-05,
"loss": 3.4019,
"step": 399500
},
{
"epoch": 16.861273869240822,
"grad_norm": 1.3777759075164795,
"learning_rate": 3.311323586159585e-05,
"loss": 3.4022,
"step": 400000
},
{
"epoch": 16.861273869240822,
"eval_accuracy": 0.7226163594133288,
"eval_loss": 3.158400535583496,
"eval_runtime": 147.1487,
"eval_samples_per_second": 820.299,
"eval_steps_per_second": 8.549,
"step": 400000
},
{
"epoch": 16.88235046157737,
"grad_norm": 1.4272950887680054,
"learning_rate": 3.302748359987995e-05,
"loss": 3.4018,
"step": 400500
},
{
"epoch": 16.903427053913923,
"grad_norm": 1.3526352643966675,
"learning_rate": 3.294173133816404e-05,
"loss": 3.403,
"step": 401000
},
{
"epoch": 16.924503646250475,
"grad_norm": 1.4073326587677002,
"learning_rate": 3.2855979076448147e-05,
"loss": 3.4074,
"step": 401500
},
{
"epoch": 16.945580238587024,
"grad_norm": 1.3411641120910645,
"learning_rate": 3.277022681473224e-05,
"loss": 3.4043,
"step": 402000
},
{
"epoch": 16.966656830923576,
"grad_norm": 1.3726180791854858,
"learning_rate": 3.2684474553016336e-05,
"loss": 3.4032,
"step": 402500
},
{
"epoch": 16.987733423260128,
"grad_norm": 1.349877119064331,
"learning_rate": 3.2598722291300435e-05,
"loss": 3.4043,
"step": 403000
},
{
"epoch": 17.00881001559668,
"grad_norm": 1.3466535806655884,
"learning_rate": 3.251297002958453e-05,
"loss": 3.4022,
"step": 403500
},
{
"epoch": 17.02988660793323,
"grad_norm": 1.376395344734192,
"learning_rate": 3.242721776786863e-05,
"loss": 3.403,
"step": 404000
},
{
"epoch": 17.05096320026978,
"grad_norm": 1.328911304473877,
"learning_rate": 3.234146550615272e-05,
"loss": 3.399,
"step": 404500
},
{
"epoch": 17.072039792606333,
"grad_norm": 1.4076844453811646,
"learning_rate": 3.225571324443683e-05,
"loss": 3.4006,
"step": 405000
},
{
"epoch": 17.09311638494288,
"grad_norm": 1.4381500482559204,
"learning_rate": 3.216996098272092e-05,
"loss": 3.4005,
"step": 405500
},
{
"epoch": 17.114192977279433,
"grad_norm": 1.3696494102478027,
"learning_rate": 3.208420872100502e-05,
"loss": 3.4009,
"step": 406000
},
{
"epoch": 17.135269569615986,
"grad_norm": 1.373154878616333,
"learning_rate": 3.1998456459289116e-05,
"loss": 3.3968,
"step": 406500
},
{
"epoch": 17.156346161952534,
"grad_norm": 1.3724151849746704,
"learning_rate": 3.1912704197573214e-05,
"loss": 3.3973,
"step": 407000
},
{
"epoch": 17.177422754289086,
"grad_norm": 1.4304693937301636,
"learning_rate": 3.1826951935857306e-05,
"loss": 3.3967,
"step": 407500
},
{
"epoch": 17.19849934662564,
"grad_norm": 1.4361759424209595,
"learning_rate": 3.174119967414141e-05,
"loss": 3.4007,
"step": 408000
},
{
"epoch": 17.219575938962187,
"grad_norm": 1.4238133430480957,
"learning_rate": 3.16554474124255e-05,
"loss": 3.3985,
"step": 408500
},
{
"epoch": 17.24065253129874,
"grad_norm": 1.3805296421051025,
"learning_rate": 3.15696951507096e-05,
"loss": 3.401,
"step": 409000
},
{
"epoch": 17.26172912363529,
"grad_norm": 1.3923993110656738,
"learning_rate": 3.14839428889937e-05,
"loss": 3.399,
"step": 409500
},
{
"epoch": 17.282805715971843,
"grad_norm": 1.4540752172470093,
"learning_rate": 3.13981906272778e-05,
"loss": 3.4019,
"step": 410000
},
{
"epoch": 17.282805715971843,
"eval_accuracy": 0.7237547668812884,
"eval_loss": 3.1511106491088867,
"eval_runtime": 146.7189,
"eval_samples_per_second": 822.703,
"eval_steps_per_second": 8.574,
"step": 410000
},
{
"epoch": 17.303882308308392,
"grad_norm": 1.3966001272201538,
"learning_rate": 3.131243836556189e-05,
"loss": 3.4018,
"step": 410500
},
{
"epoch": 17.324958900644944,
"grad_norm": 1.357434630393982,
"learning_rate": 3.1226686103845994e-05,
"loss": 3.3955,
"step": 411000
},
{
"epoch": 17.346035492981496,
"grad_norm": 1.3540213108062744,
"learning_rate": 3.1140933842130086e-05,
"loss": 3.3969,
"step": 411500
},
{
"epoch": 17.367112085318045,
"grad_norm": 1.354892611503601,
"learning_rate": 3.1055181580414184e-05,
"loss": 3.3995,
"step": 412000
},
{
"epoch": 17.388188677654597,
"grad_norm": 1.3504061698913574,
"learning_rate": 3.096942931869828e-05,
"loss": 3.4021,
"step": 412500
},
{
"epoch": 17.40926526999115,
"grad_norm": 1.384518027305603,
"learning_rate": 3.088367705698238e-05,
"loss": 3.3949,
"step": 413000
},
{
"epoch": 17.430341862327698,
"grad_norm": 1.305583119392395,
"learning_rate": 3.079792479526647e-05,
"loss": 3.3986,
"step": 413500
},
{
"epoch": 17.45141845466425,
"grad_norm": 1.3656197786331177,
"learning_rate": 3.071217253355058e-05,
"loss": 3.3954,
"step": 414000
},
{
"epoch": 17.472495047000802,
"grad_norm": 1.368380069732666,
"learning_rate": 3.062642027183467e-05,
"loss": 3.3969,
"step": 414500
},
{
"epoch": 17.49357163933735,
"grad_norm": 1.3824825286865234,
"learning_rate": 3.054066801011877e-05,
"loss": 3.3965,
"step": 415000
},
{
"epoch": 17.514648231673903,
"grad_norm": 1.3974982500076294,
"learning_rate": 3.045491574840287e-05,
"loss": 3.3988,
"step": 415500
},
{
"epoch": 17.535724824010455,
"grad_norm": 1.4187037944793701,
"learning_rate": 3.0369163486686964e-05,
"loss": 3.394,
"step": 416000
},
{
"epoch": 17.556801416347007,
"grad_norm": 1.4104770421981812,
"learning_rate": 3.028341122497106e-05,
"loss": 3.3957,
"step": 416500
},
{
"epoch": 17.577878008683555,
"grad_norm": 1.4317139387130737,
"learning_rate": 3.019765896325516e-05,
"loss": 3.3994,
"step": 417000
},
{
"epoch": 17.598954601020107,
"grad_norm": 1.4008325338363647,
"learning_rate": 3.0111906701539255e-05,
"loss": 3.398,
"step": 417500
},
{
"epoch": 17.62003119335666,
"grad_norm": 1.38494873046875,
"learning_rate": 3.002615443982335e-05,
"loss": 3.396,
"step": 418000
},
{
"epoch": 17.641107785693208,
"grad_norm": 1.3881726264953613,
"learning_rate": 2.9940402178107452e-05,
"loss": 3.3939,
"step": 418500
},
{
"epoch": 17.66218437802976,
"grad_norm": 1.3903405666351318,
"learning_rate": 2.9854649916391547e-05,
"loss": 3.3944,
"step": 419000
},
{
"epoch": 17.683260970366312,
"grad_norm": 1.4606748819351196,
"learning_rate": 2.976889765467564e-05,
"loss": 3.4001,
"step": 419500
},
{
"epoch": 17.70433756270286,
"grad_norm": 1.3841146230697632,
"learning_rate": 2.9683145392959737e-05,
"loss": 3.395,
"step": 420000
},
{
"epoch": 17.70433756270286,
"eval_accuracy": 0.7241483790724939,
"eval_loss": 3.1483395099639893,
"eval_runtime": 147.3352,
"eval_samples_per_second": 819.261,
"eval_steps_per_second": 8.538,
"step": 420000
},
{
"epoch": 17.725414155039413,
"grad_norm": 1.394323468208313,
"learning_rate": 2.9597393131243838e-05,
"loss": 3.3947,
"step": 420500
},
{
"epoch": 17.746490747375965,
"grad_norm": 1.3667819499969482,
"learning_rate": 2.9511640869527933e-05,
"loss": 3.3926,
"step": 421000
},
{
"epoch": 17.767567339712514,
"grad_norm": 1.3990609645843506,
"learning_rate": 2.942588860781203e-05,
"loss": 3.3939,
"step": 421500
},
{
"epoch": 17.788643932049066,
"grad_norm": 1.3643574714660645,
"learning_rate": 2.934013634609613e-05,
"loss": 3.3955,
"step": 422000
},
{
"epoch": 17.809720524385618,
"grad_norm": 1.370139241218567,
"learning_rate": 2.9254384084380225e-05,
"loss": 3.3974,
"step": 422500
},
{
"epoch": 17.830797116722167,
"grad_norm": 1.451165795326233,
"learning_rate": 2.9168631822664323e-05,
"loss": 3.3958,
"step": 423000
},
{
"epoch": 17.85187370905872,
"grad_norm": 1.369847297668457,
"learning_rate": 2.908287956094842e-05,
"loss": 3.3954,
"step": 423500
},
{
"epoch": 17.87295030139527,
"grad_norm": 1.388010025024414,
"learning_rate": 2.899712729923252e-05,
"loss": 3.3934,
"step": 424000
},
{
"epoch": 17.894026893731823,
"grad_norm": 1.3533453941345215,
"learning_rate": 2.8911375037516615e-05,
"loss": 3.3933,
"step": 424500
},
{
"epoch": 17.91510348606837,
"grad_norm": 1.415744662284851,
"learning_rate": 2.8825622775800716e-05,
"loss": 3.3969,
"step": 425000
},
{
"epoch": 17.936180078404924,
"grad_norm": 1.384810209274292,
"learning_rate": 2.873987051408481e-05,
"loss": 3.3938,
"step": 425500
},
{
"epoch": 17.957256670741476,
"grad_norm": 1.395005464553833,
"learning_rate": 2.8654118252368906e-05,
"loss": 3.3917,
"step": 426000
},
{
"epoch": 17.978333263078024,
"grad_norm": 1.3720922470092773,
"learning_rate": 2.8568365990653008e-05,
"loss": 3.3914,
"step": 426500
},
{
"epoch": 17.999409855414576,
"grad_norm": 1.337225317955017,
"learning_rate": 2.8482613728937103e-05,
"loss": 3.3953,
"step": 427000
},
{
"epoch": 18.02048644775113,
"grad_norm": 1.406703233718872,
"learning_rate": 2.8396861467221198e-05,
"loss": 3.3922,
"step": 427500
},
{
"epoch": 18.041563040087677,
"grad_norm": 1.457181453704834,
"learning_rate": 2.83111092055053e-05,
"loss": 3.3886,
"step": 428000
},
{
"epoch": 18.06263963242423,
"grad_norm": 1.3616580963134766,
"learning_rate": 2.8225356943789394e-05,
"loss": 3.39,
"step": 428500
},
{
"epoch": 18.08371622476078,
"grad_norm": 1.365246295928955,
"learning_rate": 2.813960468207349e-05,
"loss": 3.3889,
"step": 429000
},
{
"epoch": 18.10479281709733,
"grad_norm": 1.375732183456421,
"learning_rate": 2.805385242035759e-05,
"loss": 3.3879,
"step": 429500
},
{
"epoch": 18.125869409433882,
"grad_norm": 1.4346765279769897,
"learning_rate": 2.7968100158641686e-05,
"loss": 3.3878,
"step": 430000
},
{
"epoch": 18.125869409433882,
"eval_accuracy": 0.7248121437168892,
"eval_loss": 3.1472625732421875,
"eval_runtime": 147.36,
"eval_samples_per_second": 819.123,
"eval_steps_per_second": 8.537,
"step": 430000
},
{
"epoch": 18.146946001770434,
"grad_norm": 1.3857396841049194,
"learning_rate": 2.788234789692578e-05,
"loss": 3.3897,
"step": 430500
},
{
"epoch": 18.168022594106986,
"grad_norm": 1.4410696029663086,
"learning_rate": 2.7796595635209882e-05,
"loss": 3.3905,
"step": 431000
},
{
"epoch": 18.189099186443535,
"grad_norm": 1.423904299736023,
"learning_rate": 2.7710843373493977e-05,
"loss": 3.392,
"step": 431500
},
{
"epoch": 18.210175778780087,
"grad_norm": 1.359594464302063,
"learning_rate": 2.7625091111778072e-05,
"loss": 3.3943,
"step": 432000
},
{
"epoch": 18.23125237111664,
"grad_norm": 1.4528799057006836,
"learning_rate": 2.7539338850062174e-05,
"loss": 3.3892,
"step": 432500
},
{
"epoch": 18.252328963453188,
"grad_norm": 1.3452966213226318,
"learning_rate": 2.745358658834627e-05,
"loss": 3.3899,
"step": 433000
},
{
"epoch": 18.27340555578974,
"grad_norm": 1.3954037427902222,
"learning_rate": 2.7367834326630364e-05,
"loss": 3.3877,
"step": 433500
},
{
"epoch": 18.294482148126292,
"grad_norm": 1.362208366394043,
"learning_rate": 2.7282082064914465e-05,
"loss": 3.3887,
"step": 434000
},
{
"epoch": 18.31555874046284,
"grad_norm": 1.4276126623153687,
"learning_rate": 2.719632980319856e-05,
"loss": 3.3866,
"step": 434500
},
{
"epoch": 18.336635332799393,
"grad_norm": 1.3818186521530151,
"learning_rate": 2.7110577541482655e-05,
"loss": 3.3881,
"step": 435000
},
{
"epoch": 18.357711925135945,
"grad_norm": 1.3792022466659546,
"learning_rate": 2.7024825279766757e-05,
"loss": 3.3917,
"step": 435500
},
{
"epoch": 18.378788517472493,
"grad_norm": 1.431623101234436,
"learning_rate": 2.6939073018050852e-05,
"loss": 3.3902,
"step": 436000
},
{
"epoch": 18.399865109809046,
"grad_norm": 1.3408091068267822,
"learning_rate": 2.685332075633495e-05,
"loss": 3.3892,
"step": 436500
},
{
"epoch": 18.420941702145598,
"grad_norm": 1.3476356267929077,
"learning_rate": 2.6767568494619045e-05,
"loss": 3.3895,
"step": 437000
},
{
"epoch": 18.44201829448215,
"grad_norm": 1.3712995052337646,
"learning_rate": 2.6681816232903143e-05,
"loss": 3.3871,
"step": 437500
},
{
"epoch": 18.4630948868187,
"grad_norm": 1.4418641328811646,
"learning_rate": 2.6596063971187242e-05,
"loss": 3.3815,
"step": 438000
},
{
"epoch": 18.48417147915525,
"grad_norm": 1.425835371017456,
"learning_rate": 2.6510311709471337e-05,
"loss": 3.384,
"step": 438500
},
{
"epoch": 18.505248071491803,
"grad_norm": 1.3902266025543213,
"learning_rate": 2.642455944775544e-05,
"loss": 3.388,
"step": 439000
},
{
"epoch": 18.52632466382835,
"grad_norm": 1.3683359622955322,
"learning_rate": 2.6338807186039533e-05,
"loss": 3.3859,
"step": 439500
},
{
"epoch": 18.547401256164903,
"grad_norm": 1.3700156211853027,
"learning_rate": 2.6253054924323628e-05,
"loss": 3.3833,
"step": 440000
},
{
"epoch": 18.547401256164903,
"eval_accuracy": 0.7250475111122142,
"eval_loss": 3.1439208984375,
"eval_runtime": 148.0157,
"eval_samples_per_second": 815.495,
"eval_steps_per_second": 8.499,
"step": 440000
},
{
"epoch": 18.568477848501455,
"grad_norm": 1.4098507165908813,
"learning_rate": 2.616730266260773e-05,
"loss": 3.3846,
"step": 440500
},
{
"epoch": 18.589554440838004,
"grad_norm": 1.396986484527588,
"learning_rate": 2.6081550400891825e-05,
"loss": 3.3858,
"step": 441000
},
{
"epoch": 18.610631033174556,
"grad_norm": 1.3911192417144775,
"learning_rate": 2.599579813917592e-05,
"loss": 3.3857,
"step": 441500
},
{
"epoch": 18.63170762551111,
"grad_norm": 1.373984932899475,
"learning_rate": 2.591004587746002e-05,
"loss": 3.3863,
"step": 442000
},
{
"epoch": 18.652784217847657,
"grad_norm": 1.3671016693115234,
"learning_rate": 2.5824293615744116e-05,
"loss": 3.3874,
"step": 442500
},
{
"epoch": 18.67386081018421,
"grad_norm": 1.4883065223693848,
"learning_rate": 2.573854135402821e-05,
"loss": 3.3897,
"step": 443000
},
{
"epoch": 18.69493740252076,
"grad_norm": 1.378732681274414,
"learning_rate": 2.5652789092312313e-05,
"loss": 3.3865,
"step": 443500
},
{
"epoch": 18.716013994857313,
"grad_norm": 1.4070626497268677,
"learning_rate": 2.5567036830596408e-05,
"loss": 3.3848,
"step": 444000
},
{
"epoch": 18.737090587193862,
"grad_norm": 1.4219104051589966,
"learning_rate": 2.5481284568880503e-05,
"loss": 3.3871,
"step": 444500
},
{
"epoch": 18.758167179530414,
"grad_norm": 1.4142497777938843,
"learning_rate": 2.5395532307164605e-05,
"loss": 3.3805,
"step": 445000
},
{
"epoch": 18.779243771866966,
"grad_norm": 1.4998488426208496,
"learning_rate": 2.53097800454487e-05,
"loss": 3.3883,
"step": 445500
},
{
"epoch": 18.800320364203515,
"grad_norm": 1.3575749397277832,
"learning_rate": 2.5224027783732794e-05,
"loss": 3.3861,
"step": 446000
},
{
"epoch": 18.821396956540067,
"grad_norm": 1.4566185474395752,
"learning_rate": 2.5138275522016896e-05,
"loss": 3.3789,
"step": 446500
},
{
"epoch": 18.84247354887662,
"grad_norm": 1.4653058052062988,
"learning_rate": 2.505252326030099e-05,
"loss": 3.3816,
"step": 447000
},
{
"epoch": 18.863550141213167,
"grad_norm": 1.3904739618301392,
"learning_rate": 2.496677099858509e-05,
"loss": 3.3836,
"step": 447500
},
{
"epoch": 18.88462673354972,
"grad_norm": 1.3994755744934082,
"learning_rate": 2.4881018736869184e-05,
"loss": 3.3858,
"step": 448000
},
{
"epoch": 18.90570332588627,
"grad_norm": 1.406556487083435,
"learning_rate": 2.4795266475153283e-05,
"loss": 3.3827,
"step": 448500
},
{
"epoch": 18.92677991822282,
"grad_norm": 1.4133650064468384,
"learning_rate": 2.470951421343738e-05,
"loss": 3.3864,
"step": 449000
},
{
"epoch": 18.947856510559372,
"grad_norm": 1.3789119720458984,
"learning_rate": 2.4623761951721476e-05,
"loss": 3.3807,
"step": 449500
},
{
"epoch": 18.968933102895924,
"grad_norm": 1.4109866619110107,
"learning_rate": 2.4538009690005574e-05,
"loss": 3.3828,
"step": 450000
},
{
"epoch": 18.968933102895924,
"eval_accuracy": 0.726023687529618,
"eval_loss": 3.1380982398986816,
"eval_runtime": 147.7603,
"eval_samples_per_second": 816.904,
"eval_steps_per_second": 8.514,
"step": 450000
},
{
"epoch": 18.990009695232473,
"grad_norm": 1.404995083808899,
"learning_rate": 2.4452257428289672e-05,
"loss": 3.3812,
"step": 450500
},
{
"epoch": 19.011086287569025,
"grad_norm": 1.4986629486083984,
"learning_rate": 2.436650516657377e-05,
"loss": 3.3815,
"step": 451000
},
{
"epoch": 19.032162879905577,
"grad_norm": 1.4102296829223633,
"learning_rate": 2.428075290485787e-05,
"loss": 3.3788,
"step": 451500
},
{
"epoch": 19.05323947224213,
"grad_norm": 1.3935768604278564,
"learning_rate": 2.4195000643141964e-05,
"loss": 3.3806,
"step": 452000
},
{
"epoch": 19.074316064578678,
"grad_norm": 1.3869986534118652,
"learning_rate": 2.4109248381426062e-05,
"loss": 3.383,
"step": 452500
},
{
"epoch": 19.09539265691523,
"grad_norm": 1.39316725730896,
"learning_rate": 2.402349611971016e-05,
"loss": 3.3826,
"step": 453000
},
{
"epoch": 19.116469249251782,
"grad_norm": 1.334660291671753,
"learning_rate": 2.3937743857994255e-05,
"loss": 3.378,
"step": 453500
},
{
"epoch": 19.13754584158833,
"grad_norm": 1.4034367799758911,
"learning_rate": 2.3851991596278354e-05,
"loss": 3.3815,
"step": 454000
},
{
"epoch": 19.158622433924883,
"grad_norm": 1.4678161144256592,
"learning_rate": 2.3766239334562452e-05,
"loss": 3.3769,
"step": 454500
},
{
"epoch": 19.179699026261435,
"grad_norm": 1.4519423246383667,
"learning_rate": 2.3680487072846547e-05,
"loss": 3.3771,
"step": 455000
},
{
"epoch": 19.200775618597984,
"grad_norm": 1.408634901046753,
"learning_rate": 2.3594734811130645e-05,
"loss": 3.3774,
"step": 455500
},
{
"epoch": 19.221852210934536,
"grad_norm": 1.403300404548645,
"learning_rate": 2.3508982549414744e-05,
"loss": 3.3776,
"step": 456000
},
{
"epoch": 19.242928803271088,
"grad_norm": 1.418412685394287,
"learning_rate": 2.342323028769884e-05,
"loss": 3.3821,
"step": 456500
},
{
"epoch": 19.264005395607636,
"grad_norm": 1.3968007564544678,
"learning_rate": 2.3337478025982937e-05,
"loss": 3.3776,
"step": 457000
},
{
"epoch": 19.28508198794419,
"grad_norm": 1.4605085849761963,
"learning_rate": 2.3251725764267035e-05,
"loss": 3.3786,
"step": 457500
},
{
"epoch": 19.30615858028074,
"grad_norm": 1.4881045818328857,
"learning_rate": 2.316597350255113e-05,
"loss": 3.3816,
"step": 458000
},
{
"epoch": 19.327235172617293,
"grad_norm": 1.4012296199798584,
"learning_rate": 2.308022124083523e-05,
"loss": 3.3788,
"step": 458500
},
{
"epoch": 19.34831176495384,
"grad_norm": 1.3988802433013916,
"learning_rate": 2.2994468979119323e-05,
"loss": 3.381,
"step": 459000
},
{
"epoch": 19.369388357290394,
"grad_norm": 1.4261513948440552,
"learning_rate": 2.290871671740342e-05,
"loss": 3.3781,
"step": 459500
},
{
"epoch": 19.390464949626946,
"grad_norm": 1.4033757448196411,
"learning_rate": 2.282296445568752e-05,
"loss": 3.3795,
"step": 460000
},
{
"epoch": 19.390464949626946,
"eval_accuracy": 0.7264809777390042,
"eval_loss": 3.1348958015441895,
"eval_runtime": 147.6494,
"eval_samples_per_second": 817.518,
"eval_steps_per_second": 8.52,
"step": 460000
},
{
"epoch": 19.411541541963494,
"grad_norm": 1.4191155433654785,
"learning_rate": 2.2737212193971615e-05,
"loss": 3.3779,
"step": 460500
},
{
"epoch": 19.432618134300046,
"grad_norm": 1.3670237064361572,
"learning_rate": 2.2651459932255713e-05,
"loss": 3.3823,
"step": 461000
},
{
"epoch": 19.4536947266366,
"grad_norm": 1.3874393701553345,
"learning_rate": 2.256570767053981e-05,
"loss": 3.3751,
"step": 461500
},
{
"epoch": 19.474771318973147,
"grad_norm": 1.4305390119552612,
"learning_rate": 2.2479955408823906e-05,
"loss": 3.3775,
"step": 462000
},
{
"epoch": 19.4958479113097,
"grad_norm": 1.4339715242385864,
"learning_rate": 2.2394203147108005e-05,
"loss": 3.3774,
"step": 462500
},
{
"epoch": 19.51692450364625,
"grad_norm": 1.3992252349853516,
"learning_rate": 2.2308450885392103e-05,
"loss": 3.3774,
"step": 463000
},
{
"epoch": 19.5380010959828,
"grad_norm": 1.4240254163742065,
"learning_rate": 2.22226986236762e-05,
"loss": 3.3747,
"step": 463500
},
{
"epoch": 19.559077688319352,
"grad_norm": 1.3786145448684692,
"learning_rate": 2.21369463619603e-05,
"loss": 3.3782,
"step": 464000
},
{
"epoch": 19.580154280655904,
"grad_norm": 1.467140793800354,
"learning_rate": 2.2051194100244394e-05,
"loss": 3.3765,
"step": 464500
},
{
"epoch": 19.601230872992456,
"grad_norm": 1.4346853494644165,
"learning_rate": 2.1965441838528493e-05,
"loss": 3.3779,
"step": 465000
},
{
"epoch": 19.622307465329005,
"grad_norm": 1.4213042259216309,
"learning_rate": 2.187968957681259e-05,
"loss": 3.3761,
"step": 465500
},
{
"epoch": 19.643384057665557,
"grad_norm": 1.38515305519104,
"learning_rate": 2.179393731509669e-05,
"loss": 3.3773,
"step": 466000
},
{
"epoch": 19.66446065000211,
"grad_norm": 1.4008880853652954,
"learning_rate": 2.1708185053380784e-05,
"loss": 3.3753,
"step": 466500
},
{
"epoch": 19.685537242338658,
"grad_norm": 1.4652386903762817,
"learning_rate": 2.1622432791664883e-05,
"loss": 3.3789,
"step": 467000
},
{
"epoch": 19.70661383467521,
"grad_norm": 1.3977175951004028,
"learning_rate": 2.1536680529948978e-05,
"loss": 3.377,
"step": 467500
},
{
"epoch": 19.727690427011762,
"grad_norm": 1.4201745986938477,
"learning_rate": 2.1450928268233076e-05,
"loss": 3.376,
"step": 468000
},
{
"epoch": 19.74876701934831,
"grad_norm": 1.3562321662902832,
"learning_rate": 2.1365176006517174e-05,
"loss": 3.3768,
"step": 468500
},
{
"epoch": 19.769843611684863,
"grad_norm": 1.5074714422225952,
"learning_rate": 2.127942374480127e-05,
"loss": 3.3739,
"step": 469000
},
{
"epoch": 19.790920204021415,
"grad_norm": 1.4173215627670288,
"learning_rate": 2.1193671483085367e-05,
"loss": 3.3799,
"step": 469500
},
{
"epoch": 19.811996796357963,
"grad_norm": 1.430889368057251,
"learning_rate": 2.1107919221369466e-05,
"loss": 3.3746,
"step": 470000
},
{
"epoch": 19.811996796357963,
"eval_accuracy": 0.7271663967554408,
"eval_loss": 3.1318140029907227,
"eval_runtime": 147.8409,
"eval_samples_per_second": 816.459,
"eval_steps_per_second": 8.509,
"step": 470000
},
{
"epoch": 19.833073388694515,
"grad_norm": 1.3997249603271484,
"learning_rate": 2.102216695965356e-05,
"loss": 3.3781,
"step": 470500
},
{
"epoch": 19.854149981031068,
"grad_norm": 1.4380711317062378,
"learning_rate": 2.093641469793766e-05,
"loss": 3.3751,
"step": 471000
},
{
"epoch": 19.87522657336762,
"grad_norm": 1.3691411018371582,
"learning_rate": 2.0850662436221757e-05,
"loss": 3.3746,
"step": 471500
},
{
"epoch": 19.89630316570417,
"grad_norm": 1.391831874847412,
"learning_rate": 2.0764910174505852e-05,
"loss": 3.375,
"step": 472000
},
{
"epoch": 19.91737975804072,
"grad_norm": 1.4064886569976807,
"learning_rate": 2.067915791278995e-05,
"loss": 3.3758,
"step": 472500
},
{
"epoch": 19.938456350377272,
"grad_norm": 1.436800241470337,
"learning_rate": 2.059340565107405e-05,
"loss": 3.3705,
"step": 473000
},
{
"epoch": 19.95953294271382,
"grad_norm": 1.4297164678573608,
"learning_rate": 2.0507653389358144e-05,
"loss": 3.3747,
"step": 473500
},
{
"epoch": 19.980609535050373,
"grad_norm": 1.4841296672821045,
"learning_rate": 2.0421901127642242e-05,
"loss": 3.3722,
"step": 474000
},
{
"epoch": 20.001686127386925,
"grad_norm": 1.4472925662994385,
"learning_rate": 2.033614886592634e-05,
"loss": 3.3808,
"step": 474500
},
{
"epoch": 20.022762719723474,
"grad_norm": 1.469746708869934,
"learning_rate": 2.0250396604210435e-05,
"loss": 3.374,
"step": 475000
},
{
"epoch": 20.043839312060026,
"grad_norm": 1.4777560234069824,
"learning_rate": 2.0164644342494534e-05,
"loss": 3.3731,
"step": 475500
},
{
"epoch": 20.064915904396578,
"grad_norm": 1.407267451286316,
"learning_rate": 2.007889208077863e-05,
"loss": 3.3653,
"step": 476000
},
{
"epoch": 20.085992496733127,
"grad_norm": 1.418859839439392,
"learning_rate": 1.9993139819062727e-05,
"loss": 3.3699,
"step": 476500
},
{
"epoch": 20.10706908906968,
"grad_norm": 1.4226833581924438,
"learning_rate": 1.9907387557346825e-05,
"loss": 3.372,
"step": 477000
},
{
"epoch": 20.12814568140623,
"grad_norm": 1.365512728691101,
"learning_rate": 1.9821635295630923e-05,
"loss": 3.3701,
"step": 477500
},
{
"epoch": 20.14922227374278,
"grad_norm": 1.4760130643844604,
"learning_rate": 1.9735883033915022e-05,
"loss": 3.3699,
"step": 478000
},
{
"epoch": 20.17029886607933,
"grad_norm": 1.4753940105438232,
"learning_rate": 1.965013077219912e-05,
"loss": 3.3723,
"step": 478500
},
{
"epoch": 20.191375458415884,
"grad_norm": 1.4325244426727295,
"learning_rate": 1.9564378510483215e-05,
"loss": 3.371,
"step": 479000
},
{
"epoch": 20.212452050752436,
"grad_norm": 1.3958473205566406,
"learning_rate": 1.9478626248767313e-05,
"loss": 3.3739,
"step": 479500
},
{
"epoch": 20.233528643088984,
"grad_norm": 1.4405913352966309,
"learning_rate": 1.939287398705141e-05,
"loss": 3.3704,
"step": 480000
},
{
"epoch": 20.233528643088984,
"eval_accuracy": 0.7275380735031856,
"eval_loss": 3.1286561489105225,
"eval_runtime": 147.1416,
"eval_samples_per_second": 820.339,
"eval_steps_per_second": 8.55,
"step": 480000
},
{
"epoch": 20.254605235425537,
"grad_norm": 1.387924313545227,
"learning_rate": 1.9307121725335506e-05,
"loss": 3.3674,
"step": 480500
},
{
"epoch": 20.27568182776209,
"grad_norm": 1.5385538339614868,
"learning_rate": 1.9221369463619605e-05,
"loss": 3.3667,
"step": 481000
},
{
"epoch": 20.296758420098637,
"grad_norm": 1.379767656326294,
"learning_rate": 1.9135617201903703e-05,
"loss": 3.3668,
"step": 481500
},
{
"epoch": 20.31783501243519,
"grad_norm": 1.425859808921814,
"learning_rate": 1.9049864940187798e-05,
"loss": 3.3692,
"step": 482000
},
{
"epoch": 20.33891160477174,
"grad_norm": 1.4766411781311035,
"learning_rate": 1.8964112678471896e-05,
"loss": 3.3714,
"step": 482500
},
{
"epoch": 20.35998819710829,
"grad_norm": 1.4334497451782227,
"learning_rate": 1.8878360416755995e-05,
"loss": 3.3732,
"step": 483000
},
{
"epoch": 20.381064789444842,
"grad_norm": 1.398589015007019,
"learning_rate": 1.879260815504009e-05,
"loss": 3.3694,
"step": 483500
},
{
"epoch": 20.402141381781394,
"grad_norm": 1.4148585796356201,
"learning_rate": 1.8706855893324188e-05,
"loss": 3.3676,
"step": 484000
},
{
"epoch": 20.423217974117943,
"grad_norm": 1.4488985538482666,
"learning_rate": 1.8621103631608283e-05,
"loss": 3.3707,
"step": 484500
},
{
"epoch": 20.444294566454495,
"grad_norm": 1.4300071001052856,
"learning_rate": 1.853535136989238e-05,
"loss": 3.368,
"step": 485000
},
{
"epoch": 20.465371158791047,
"grad_norm": 1.4643628597259521,
"learning_rate": 1.844959910817648e-05,
"loss": 3.3663,
"step": 485500
},
{
"epoch": 20.4864477511276,
"grad_norm": 1.421117901802063,
"learning_rate": 1.8363846846460574e-05,
"loss": 3.3702,
"step": 486000
},
{
"epoch": 20.507524343464148,
"grad_norm": 1.449509620666504,
"learning_rate": 1.8278094584744673e-05,
"loss": 3.3656,
"step": 486500
},
{
"epoch": 20.5286009358007,
"grad_norm": 1.4595346450805664,
"learning_rate": 1.819234232302877e-05,
"loss": 3.3699,
"step": 487000
},
{
"epoch": 20.549677528137252,
"grad_norm": 1.4136524200439453,
"learning_rate": 1.8106590061312866e-05,
"loss": 3.3718,
"step": 487500
},
{
"epoch": 20.5707541204738,
"grad_norm": 1.4270710945129395,
"learning_rate": 1.8020837799596964e-05,
"loss": 3.3712,
"step": 488000
},
{
"epoch": 20.591830712810353,
"grad_norm": 1.4298735857009888,
"learning_rate": 1.7935085537881062e-05,
"loss": 3.37,
"step": 488500
},
{
"epoch": 20.612907305146905,
"grad_norm": 1.4872045516967773,
"learning_rate": 1.7849333276165157e-05,
"loss": 3.3675,
"step": 489000
},
{
"epoch": 20.633983897483454,
"grad_norm": 1.4392427206039429,
"learning_rate": 1.7763581014449256e-05,
"loss": 3.3674,
"step": 489500
},
{
"epoch": 20.655060489820006,
"grad_norm": 1.4235326051712036,
"learning_rate": 1.7677828752733354e-05,
"loss": 3.366,
"step": 490000
},
{
"epoch": 20.655060489820006,
"eval_accuracy": 0.7282561124007456,
"eval_loss": 3.1247944831848145,
"eval_runtime": 147.4869,
"eval_samples_per_second": 818.418,
"eval_steps_per_second": 8.53,
"step": 490000
},
{
"epoch": 20.676137082156558,
"grad_norm": 1.477043867111206,
"learning_rate": 1.7592076491017452e-05,
"loss": 3.3689,
"step": 490500
},
{
"epoch": 20.697213674493106,
"grad_norm": 1.4441403150558472,
"learning_rate": 1.7506324229301547e-05,
"loss": 3.3699,
"step": 491000
},
{
"epoch": 20.71829026682966,
"grad_norm": 1.450617790222168,
"learning_rate": 1.7420571967585646e-05,
"loss": 3.3709,
"step": 491500
},
{
"epoch": 20.73936685916621,
"grad_norm": 1.4334152936935425,
"learning_rate": 1.7334819705869744e-05,
"loss": 3.3666,
"step": 492000
},
{
"epoch": 20.760443451502763,
"grad_norm": 1.4084230661392212,
"learning_rate": 1.7249067444153842e-05,
"loss": 3.3701,
"step": 492500
},
{
"epoch": 20.78152004383931,
"grad_norm": 1.4242982864379883,
"learning_rate": 1.7163315182437937e-05,
"loss": 3.3658,
"step": 493000
},
{
"epoch": 20.802596636175863,
"grad_norm": 1.451520323753357,
"learning_rate": 1.7077562920722035e-05,
"loss": 3.3652,
"step": 493500
},
{
"epoch": 20.823673228512416,
"grad_norm": 1.4471306800842285,
"learning_rate": 1.6991810659006134e-05,
"loss": 3.3674,
"step": 494000
},
{
"epoch": 20.844749820848964,
"grad_norm": 1.4390236139297485,
"learning_rate": 1.690605839729023e-05,
"loss": 3.367,
"step": 494500
},
{
"epoch": 20.865826413185516,
"grad_norm": 1.340144157409668,
"learning_rate": 1.6820306135574327e-05,
"loss": 3.3675,
"step": 495000
},
{
"epoch": 20.88690300552207,
"grad_norm": 1.4810863733291626,
"learning_rate": 1.6734553873858425e-05,
"loss": 3.3687,
"step": 495500
},
{
"epoch": 20.907979597858617,
"grad_norm": 1.4605376720428467,
"learning_rate": 1.664880161214252e-05,
"loss": 3.3673,
"step": 496000
},
{
"epoch": 20.92905619019517,
"grad_norm": 1.4020469188690186,
"learning_rate": 1.656304935042662e-05,
"loss": 3.3683,
"step": 496500
},
{
"epoch": 20.95013278253172,
"grad_norm": 1.3742103576660156,
"learning_rate": 1.6477297088710717e-05,
"loss": 3.366,
"step": 497000
},
{
"epoch": 20.97120937486827,
"grad_norm": 1.4163638353347778,
"learning_rate": 1.639154482699481e-05,
"loss": 3.3665,
"step": 497500
},
{
"epoch": 20.992285967204822,
"grad_norm": 1.4138046503067017,
"learning_rate": 1.630579256527891e-05,
"loss": 3.3651,
"step": 498000
},
{
"epoch": 21.013362559541374,
"grad_norm": 1.4007530212402344,
"learning_rate": 1.6220040303563008e-05,
"loss": 3.3642,
"step": 498500
},
{
"epoch": 21.034439151877926,
"grad_norm": 1.4674230813980103,
"learning_rate": 1.6134288041847103e-05,
"loss": 3.3678,
"step": 499000
},
{
"epoch": 21.055515744214475,
"grad_norm": 1.4210869073867798,
"learning_rate": 1.60485357801312e-05,
"loss": 3.3627,
"step": 499500
},
{
"epoch": 21.076592336551027,
"grad_norm": 1.437940239906311,
"learning_rate": 1.59627835184153e-05,
"loss": 3.3621,
"step": 500000
},
{
"epoch": 21.076592336551027,
"eval_accuracy": 0.7286221508134925,
"eval_loss": 3.1214497089385986,
"eval_runtime": 147.1716,
"eval_samples_per_second": 820.172,
"eval_steps_per_second": 8.548,
"step": 500000
},
{
"epoch": 21.09766892888758,
"grad_norm": 1.4531985521316528,
"learning_rate": 1.5877031256699395e-05,
"loss": 3.3627,
"step": 500500
},
{
"epoch": 21.118745521224128,
"grad_norm": 1.3674176931381226,
"learning_rate": 1.5791278994983493e-05,
"loss": 3.366,
"step": 501000
},
{
"epoch": 21.13982211356068,
"grad_norm": 1.3786336183547974,
"learning_rate": 1.570552673326759e-05,
"loss": 3.3607,
"step": 501500
},
{
"epoch": 21.16089870589723,
"grad_norm": 1.3946843147277832,
"learning_rate": 1.5619774471551686e-05,
"loss": 3.3657,
"step": 502000
},
{
"epoch": 21.18197529823378,
"grad_norm": 1.4062042236328125,
"learning_rate": 1.5534022209835785e-05,
"loss": 3.3649,
"step": 502500
},
{
"epoch": 21.203051890570332,
"grad_norm": 1.3415184020996094,
"learning_rate": 1.544826994811988e-05,
"loss": 3.3637,
"step": 503000
},
{
"epoch": 21.224128482906885,
"grad_norm": 1.4797886610031128,
"learning_rate": 1.5362517686403978e-05,
"loss": 3.3632,
"step": 503500
},
{
"epoch": 21.245205075243433,
"grad_norm": 1.4170137643814087,
"learning_rate": 1.5276765424688076e-05,
"loss": 3.3627,
"step": 504000
},
{
"epoch": 21.266281667579985,
"grad_norm": 1.47367525100708,
"learning_rate": 1.5191013162972173e-05,
"loss": 3.3627,
"step": 504500
},
{
"epoch": 21.287358259916537,
"grad_norm": 1.4060405492782593,
"learning_rate": 1.5105260901256271e-05,
"loss": 3.3635,
"step": 505000
},
{
"epoch": 21.308434852253086,
"grad_norm": 1.4364203214645386,
"learning_rate": 1.501950863954037e-05,
"loss": 3.3631,
"step": 505500
},
{
"epoch": 21.329511444589638,
"grad_norm": 1.4189989566802979,
"learning_rate": 1.4933756377824464e-05,
"loss": 3.3606,
"step": 506000
},
{
"epoch": 21.35058803692619,
"grad_norm": 1.401957631111145,
"learning_rate": 1.4848004116108563e-05,
"loss": 3.358,
"step": 506500
},
{
"epoch": 21.371664629262742,
"grad_norm": 1.3962433338165283,
"learning_rate": 1.4762251854392661e-05,
"loss": 3.3653,
"step": 507000
},
{
"epoch": 21.39274122159929,
"grad_norm": 1.4039461612701416,
"learning_rate": 1.4676499592676758e-05,
"loss": 3.3639,
"step": 507500
},
{
"epoch": 21.413817813935843,
"grad_norm": 1.4362659454345703,
"learning_rate": 1.4590747330960856e-05,
"loss": 3.3613,
"step": 508000
},
{
"epoch": 21.434894406272395,
"grad_norm": 1.4081591367721558,
"learning_rate": 1.4504995069244952e-05,
"loss": 3.3571,
"step": 508500
},
{
"epoch": 21.455970998608944,
"grad_norm": 1.3960293531417847,
"learning_rate": 1.4419242807529049e-05,
"loss": 3.3613,
"step": 509000
},
{
"epoch": 21.477047590945496,
"grad_norm": 1.3907986879348755,
"learning_rate": 1.4333490545813147e-05,
"loss": 3.3613,
"step": 509500
},
{
"epoch": 21.498124183282048,
"grad_norm": 1.4456912279129028,
"learning_rate": 1.4247738284097246e-05,
"loss": 3.3582,
"step": 510000
},
{
"epoch": 21.498124183282048,
"eval_accuracy": 0.7291358383089758,
"eval_loss": 3.1188690662384033,
"eval_runtime": 147.7369,
"eval_samples_per_second": 817.033,
"eval_steps_per_second": 8.515,
"step": 510000
},
{
"epoch": 21.519200775618597,
"grad_norm": 1.3979663848876953,
"learning_rate": 1.416198602238134e-05,
"loss": 3.3615,
"step": 510500
},
{
"epoch": 21.54027736795515,
"grad_norm": 1.345090389251709,
"learning_rate": 1.4076233760665439e-05,
"loss": 3.3611,
"step": 511000
},
{
"epoch": 21.5613539602917,
"grad_norm": 1.4260592460632324,
"learning_rate": 1.3990481498949534e-05,
"loss": 3.3586,
"step": 511500
},
{
"epoch": 21.58243055262825,
"grad_norm": 1.441012978553772,
"learning_rate": 1.3904729237233632e-05,
"loss": 3.3629,
"step": 512000
},
{
"epoch": 21.6035071449648,
"grad_norm": 1.379099726676941,
"learning_rate": 1.381897697551773e-05,
"loss": 3.3593,
"step": 512500
},
{
"epoch": 21.624583737301354,
"grad_norm": 1.4386441707611084,
"learning_rate": 1.3733224713801825e-05,
"loss": 3.3632,
"step": 513000
},
{
"epoch": 21.645660329637906,
"grad_norm": 1.4349092245101929,
"learning_rate": 1.3647472452085924e-05,
"loss": 3.3675,
"step": 513500
},
{
"epoch": 21.666736921974454,
"grad_norm": 1.4554965496063232,
"learning_rate": 1.3561720190370022e-05,
"loss": 3.3611,
"step": 514000
},
{
"epoch": 21.687813514311006,
"grad_norm": 1.42093825340271,
"learning_rate": 1.3475967928654119e-05,
"loss": 3.3575,
"step": 514500
},
{
"epoch": 21.70889010664756,
"grad_norm": 1.4610778093338013,
"learning_rate": 1.3390215666938217e-05,
"loss": 3.362,
"step": 515000
},
{
"epoch": 21.729966698984107,
"grad_norm": 1.4008283615112305,
"learning_rate": 1.3304463405222315e-05,
"loss": 3.3614,
"step": 515500
},
{
"epoch": 21.75104329132066,
"grad_norm": 1.456808090209961,
"learning_rate": 1.321871114350641e-05,
"loss": 3.3626,
"step": 516000
},
{
"epoch": 21.77211988365721,
"grad_norm": 1.4532802104949951,
"learning_rate": 1.3132958881790508e-05,
"loss": 3.3572,
"step": 516500
},
{
"epoch": 21.79319647599376,
"grad_norm": 1.3922919034957886,
"learning_rate": 1.3047206620074607e-05,
"loss": 3.358,
"step": 517000
},
{
"epoch": 21.814273068330312,
"grad_norm": 1.4842333793640137,
"learning_rate": 1.2961454358358702e-05,
"loss": 3.3602,
"step": 517500
},
{
"epoch": 21.835349660666864,
"grad_norm": 1.4687374830245972,
"learning_rate": 1.28757020966428e-05,
"loss": 3.3555,
"step": 518000
},
{
"epoch": 21.856426253003413,
"grad_norm": 1.4305050373077393,
"learning_rate": 1.2789949834926898e-05,
"loss": 3.3569,
"step": 518500
},
{
"epoch": 21.877502845339965,
"grad_norm": 1.4337642192840576,
"learning_rate": 1.2704197573210993e-05,
"loss": 3.3631,
"step": 519000
},
{
"epoch": 21.898579437676517,
"grad_norm": 1.3712856769561768,
"learning_rate": 1.2618445311495092e-05,
"loss": 3.3564,
"step": 519500
},
{
"epoch": 21.91965603001307,
"grad_norm": 1.4119799137115479,
"learning_rate": 1.2532693049779186e-05,
"loss": 3.3547,
"step": 520000
},
{
"epoch": 21.91965603001307,
"eval_accuracy": 0.7293545542601414,
"eval_loss": 3.117441415786743,
"eval_runtime": 147.5826,
"eval_samples_per_second": 817.888,
"eval_steps_per_second": 8.524,
"step": 520000
},
{
"epoch": 21.940732622349618,
"grad_norm": 1.5094081163406372,
"learning_rate": 1.2446940788063285e-05,
"loss": 3.3597,
"step": 520500
},
{
"epoch": 21.96180921468617,
"grad_norm": 1.436201572418213,
"learning_rate": 1.2361188526347383e-05,
"loss": 3.3581,
"step": 521000
},
{
"epoch": 21.982885807022722,
"grad_norm": 1.4114795923233032,
"learning_rate": 1.227543626463148e-05,
"loss": 3.3588,
"step": 521500
},
{
"epoch": 22.00396239935927,
"grad_norm": 1.4875950813293457,
"learning_rate": 1.2189684002915578e-05,
"loss": 3.3551,
"step": 522000
},
{
"epoch": 22.025038991695823,
"grad_norm": 1.4479924440383911,
"learning_rate": 1.2103931741199675e-05,
"loss": 3.3537,
"step": 522500
},
{
"epoch": 22.046115584032375,
"grad_norm": 1.4202848672866821,
"learning_rate": 1.2018179479483773e-05,
"loss": 3.3561,
"step": 523000
},
{
"epoch": 22.067192176368923,
"grad_norm": 1.4268715381622314,
"learning_rate": 1.193242721776787e-05,
"loss": 3.3569,
"step": 523500
},
{
"epoch": 22.088268768705476,
"grad_norm": 1.5013192892074585,
"learning_rate": 1.1846674956051966e-05,
"loss": 3.3589,
"step": 524000
},
{
"epoch": 22.109345361042028,
"grad_norm": 1.4787311553955078,
"learning_rate": 1.1760922694336064e-05,
"loss": 3.3537,
"step": 524500
},
{
"epoch": 22.130421953378576,
"grad_norm": 1.4780042171478271,
"learning_rate": 1.1675170432620161e-05,
"loss": 3.3592,
"step": 525000
},
{
"epoch": 22.15149854571513,
"grad_norm": 1.3958930969238281,
"learning_rate": 1.1589418170904258e-05,
"loss": 3.3519,
"step": 525500
},
{
"epoch": 22.17257513805168,
"grad_norm": 1.4388000965118408,
"learning_rate": 1.1503665909188354e-05,
"loss": 3.3536,
"step": 526000
},
{
"epoch": 22.193651730388233,
"grad_norm": 1.4664348363876343,
"learning_rate": 1.1417913647472453e-05,
"loss": 3.3549,
"step": 526500
},
{
"epoch": 22.21472832272478,
"grad_norm": 1.4486054182052612,
"learning_rate": 1.133216138575655e-05,
"loss": 3.3567,
"step": 527000
},
{
"epoch": 22.235804915061333,
"grad_norm": 1.462202548980713,
"learning_rate": 1.1246409124040646e-05,
"loss": 3.3526,
"step": 527500
},
{
"epoch": 22.256881507397885,
"grad_norm": 1.4373327493667603,
"learning_rate": 1.1160656862324744e-05,
"loss": 3.3571,
"step": 528000
},
{
"epoch": 22.277958099734434,
"grad_norm": 1.4135340452194214,
"learning_rate": 1.1074904600608842e-05,
"loss": 3.3553,
"step": 528500
},
{
"epoch": 22.299034692070986,
"grad_norm": 1.4653147459030151,
"learning_rate": 1.0989152338892939e-05,
"loss": 3.353,
"step": 529000
},
{
"epoch": 22.320111284407538,
"grad_norm": 1.5179762840270996,
"learning_rate": 1.0903400077177037e-05,
"loss": 3.3568,
"step": 529500
},
{
"epoch": 22.341187876744087,
"grad_norm": 1.4130831956863403,
"learning_rate": 1.0817647815461134e-05,
"loss": 3.3561,
"step": 530000
},
{
"epoch": 22.341187876744087,
"eval_accuracy": 0.7298220648055258,
"eval_loss": 3.1151580810546875,
"eval_runtime": 147.5867,
"eval_samples_per_second": 817.865,
"eval_steps_per_second": 8.524,
"step": 530000
},
{
"epoch": 22.36226446908064,
"grad_norm": 1.4461263418197632,
"learning_rate": 1.073189555374523e-05,
"loss": 3.3566,
"step": 530500
},
{
"epoch": 22.38334106141719,
"grad_norm": 1.5368952751159668,
"learning_rate": 1.0646143292029327e-05,
"loss": 3.3591,
"step": 531000
},
{
"epoch": 22.40441765375374,
"grad_norm": 1.4540859460830688,
"learning_rate": 1.0560391030313425e-05,
"loss": 3.3529,
"step": 531500
},
{
"epoch": 22.42549424609029,
"grad_norm": 1.4295954704284668,
"learning_rate": 1.0474638768597522e-05,
"loss": 3.3518,
"step": 532000
},
{
"epoch": 22.446570838426844,
"grad_norm": 1.4346060752868652,
"learning_rate": 1.0388886506881619e-05,
"loss": 3.3535,
"step": 532500
},
{
"epoch": 22.467647430763392,
"grad_norm": 1.441426157951355,
"learning_rate": 1.0303134245165717e-05,
"loss": 3.3561,
"step": 533000
},
{
"epoch": 22.488724023099945,
"grad_norm": 1.4384709596633911,
"learning_rate": 1.0217381983449814e-05,
"loss": 3.3539,
"step": 533500
},
{
"epoch": 22.509800615436497,
"grad_norm": 1.4117240905761719,
"learning_rate": 1.013162972173391e-05,
"loss": 3.3552,
"step": 534000
},
{
"epoch": 22.53087720777305,
"grad_norm": 1.4357889890670776,
"learning_rate": 1.0045877460018009e-05,
"loss": 3.3571,
"step": 534500
},
{
"epoch": 22.551953800109597,
"grad_norm": 1.4541184902191162,
"learning_rate": 9.960125198302105e-06,
"loss": 3.3591,
"step": 535000
},
{
"epoch": 22.57303039244615,
"grad_norm": 1.3997505903244019,
"learning_rate": 9.874372936586203e-06,
"loss": 3.3539,
"step": 535500
},
{
"epoch": 22.5941069847827,
"grad_norm": 1.4746195077896118,
"learning_rate": 9.7886206748703e-06,
"loss": 3.3526,
"step": 536000
},
{
"epoch": 22.61518357711925,
"grad_norm": 1.4516388177871704,
"learning_rate": 9.702868413154398e-06,
"loss": 3.3526,
"step": 536500
},
{
"epoch": 22.636260169455802,
"grad_norm": 1.4937301874160767,
"learning_rate": 9.617116151438495e-06,
"loss": 3.3515,
"step": 537000
},
{
"epoch": 22.657336761792354,
"grad_norm": 1.4457532167434692,
"learning_rate": 9.531363889722592e-06,
"loss": 3.3535,
"step": 537500
},
{
"epoch": 22.678413354128903,
"grad_norm": 1.4052773714065552,
"learning_rate": 9.44561162800669e-06,
"loss": 3.3518,
"step": 538000
},
{
"epoch": 22.699489946465455,
"grad_norm": 1.4570778608322144,
"learning_rate": 9.359859366290787e-06,
"loss": 3.3547,
"step": 538500
},
{
"epoch": 22.720566538802007,
"grad_norm": 1.3718531131744385,
"learning_rate": 9.274107104574883e-06,
"loss": 3.3552,
"step": 539000
},
{
"epoch": 22.741643131138556,
"grad_norm": 1.4549061059951782,
"learning_rate": 9.18835484285898e-06,
"loss": 3.3497,
"step": 539500
},
{
"epoch": 22.762719723475108,
"grad_norm": 1.4438334703445435,
"learning_rate": 9.102602581143078e-06,
"loss": 3.3516,
"step": 540000
},
{
"epoch": 22.762719723475108,
"eval_accuracy": 0.7300271968264377,
"eval_loss": 3.1145029067993164,
"eval_runtime": 147.5062,
"eval_samples_per_second": 818.312,
"eval_steps_per_second": 8.528,
"step": 540000
},
{
"epoch": 22.78379631581166,
"grad_norm": 1.5209364891052246,
"learning_rate": 9.016850319427175e-06,
"loss": 3.355,
"step": 540500
},
{
"epoch": 22.804872908148212,
"grad_norm": 1.4285752773284912,
"learning_rate": 8.931098057711271e-06,
"loss": 3.3535,
"step": 541000
},
{
"epoch": 22.82594950048476,
"grad_norm": 1.4603443145751953,
"learning_rate": 8.84534579599537e-06,
"loss": 3.3531,
"step": 541500
},
{
"epoch": 22.847026092821313,
"grad_norm": 1.478726863861084,
"learning_rate": 8.759593534279468e-06,
"loss": 3.3536,
"step": 542000
},
{
"epoch": 22.868102685157865,
"grad_norm": 1.4591944217681885,
"learning_rate": 8.673841272563565e-06,
"loss": 3.3532,
"step": 542500
},
{
"epoch": 22.889179277494414,
"grad_norm": 1.4354801177978516,
"learning_rate": 8.588089010847663e-06,
"loss": 3.3535,
"step": 543000
},
{
"epoch": 22.910255869830966,
"grad_norm": 1.4454721212387085,
"learning_rate": 8.50233674913176e-06,
"loss": 3.3531,
"step": 543500
},
{
"epoch": 22.931332462167518,
"grad_norm": 1.4588005542755127,
"learning_rate": 8.416584487415856e-06,
"loss": 3.3475,
"step": 544000
},
{
"epoch": 22.952409054504066,
"grad_norm": 1.5083816051483154,
"learning_rate": 8.330832225699953e-06,
"loss": 3.3475,
"step": 544500
},
{
"epoch": 22.97348564684062,
"grad_norm": 1.48488187789917,
"learning_rate": 8.245079963984051e-06,
"loss": 3.3471,
"step": 545000
},
{
"epoch": 22.99456223917717,
"grad_norm": 1.463814616203308,
"learning_rate": 8.159327702268148e-06,
"loss": 3.3539,
"step": 545500
},
{
"epoch": 23.01563883151372,
"grad_norm": 1.47548508644104,
"learning_rate": 8.073575440552244e-06,
"loss": 3.3508,
"step": 546000
},
{
"epoch": 23.03671542385027,
"grad_norm": 1.4441149234771729,
"learning_rate": 7.987823178836343e-06,
"loss": 3.3513,
"step": 546500
},
{
"epoch": 23.057792016186824,
"grad_norm": 1.4644056558609009,
"learning_rate": 7.90207091712044e-06,
"loss": 3.3493,
"step": 547000
},
{
"epoch": 23.078868608523376,
"grad_norm": 1.483940601348877,
"learning_rate": 7.816318655404536e-06,
"loss": 3.3474,
"step": 547500
},
{
"epoch": 23.099945200859924,
"grad_norm": 1.492734670639038,
"learning_rate": 7.730566393688634e-06,
"loss": 3.3519,
"step": 548000
},
{
"epoch": 23.121021793196476,
"grad_norm": 1.4302022457122803,
"learning_rate": 7.64481413197273e-06,
"loss": 3.3482,
"step": 548500
},
{
"epoch": 23.14209838553303,
"grad_norm": 1.4348735809326172,
"learning_rate": 7.559061870256828e-06,
"loss": 3.3504,
"step": 549000
},
{
"epoch": 23.163174977869577,
"grad_norm": 1.4341139793395996,
"learning_rate": 7.473309608540925e-06,
"loss": 3.3486,
"step": 549500
},
{
"epoch": 23.18425157020613,
"grad_norm": 1.4635459184646606,
"learning_rate": 7.387557346825023e-06,
"loss": 3.3517,
"step": 550000
},
{
"epoch": 23.18425157020613,
"eval_accuracy": 0.730335906390004,
"eval_loss": 3.110952138900757,
"eval_runtime": 146.6074,
"eval_samples_per_second": 823.328,
"eval_steps_per_second": 8.581,
"step": 550000
},
{
"epoch": 23.20532816254268,
"grad_norm": 1.4879865646362305,
"learning_rate": 7.3018050851091205e-06,
"loss": 3.346,
"step": 550500
},
{
"epoch": 23.22640475487923,
"grad_norm": 1.4445784091949463,
"learning_rate": 7.216052823393217e-06,
"loss": 3.3502,
"step": 551000
},
{
"epoch": 23.247481347215782,
"grad_norm": 1.4836221933364868,
"learning_rate": 7.1303005616773155e-06,
"loss": 3.3492,
"step": 551500
},
{
"epoch": 23.268557939552334,
"grad_norm": 1.4737675189971924,
"learning_rate": 7.044548299961412e-06,
"loss": 3.3531,
"step": 552000
},
{
"epoch": 23.289634531888883,
"grad_norm": 1.4925107955932617,
"learning_rate": 6.958796038245509e-06,
"loss": 3.3432,
"step": 552500
},
{
"epoch": 23.310711124225435,
"grad_norm": 1.49608314037323,
"learning_rate": 6.873043776529606e-06,
"loss": 3.3542,
"step": 553000
},
{
"epoch": 23.331787716561987,
"grad_norm": 1.504751443862915,
"learning_rate": 6.787291514813704e-06,
"loss": 3.3466,
"step": 553500
},
{
"epoch": 23.35286430889854,
"grad_norm": 1.4608049392700195,
"learning_rate": 6.701539253097801e-06,
"loss": 3.3478,
"step": 554000
},
{
"epoch": 23.373940901235088,
"grad_norm": 1.5101016759872437,
"learning_rate": 6.615786991381898e-06,
"loss": 3.3475,
"step": 554500
},
{
"epoch": 23.39501749357164,
"grad_norm": 1.4229103326797485,
"learning_rate": 6.530034729665996e-06,
"loss": 3.349,
"step": 555000
},
{
"epoch": 23.416094085908192,
"grad_norm": 1.43767511844635,
"learning_rate": 6.444282467950093e-06,
"loss": 3.3472,
"step": 555500
},
{
"epoch": 23.43717067824474,
"grad_norm": 1.4638793468475342,
"learning_rate": 6.358530206234189e-06,
"loss": 3.3443,
"step": 556000
},
{
"epoch": 23.458247270581293,
"grad_norm": 1.435266375541687,
"learning_rate": 6.272777944518287e-06,
"loss": 3.3499,
"step": 556500
},
{
"epoch": 23.479323862917845,
"grad_norm": 1.4639055728912354,
"learning_rate": 6.187025682802384e-06,
"loss": 3.3504,
"step": 557000
},
{
"epoch": 23.500400455254393,
"grad_norm": 1.4363114833831787,
"learning_rate": 6.101273421086482e-06,
"loss": 3.3483,
"step": 557500
},
{
"epoch": 23.521477047590945,
"grad_norm": 1.4339805841445923,
"learning_rate": 6.015521159370579e-06,
"loss": 3.347,
"step": 558000
},
{
"epoch": 23.542553639927497,
"grad_norm": 1.4509868621826172,
"learning_rate": 5.929768897654676e-06,
"loss": 3.3466,
"step": 558500
},
{
"epoch": 23.563630232264046,
"grad_norm": 1.484214186668396,
"learning_rate": 5.844016635938773e-06,
"loss": 3.3476,
"step": 559000
},
{
"epoch": 23.584706824600598,
"grad_norm": 1.4402806758880615,
"learning_rate": 5.75826437422287e-06,
"loss": 3.3473,
"step": 559500
},
{
"epoch": 23.60578341693715,
"grad_norm": 1.4699591398239136,
"learning_rate": 5.672512112506967e-06,
"loss": 3.349,
"step": 560000
},
{
"epoch": 23.60578341693715,
"eval_accuracy": 0.7308820009464293,
"eval_loss": 3.1086690425872803,
"eval_runtime": 147.3495,
"eval_samples_per_second": 819.181,
"eval_steps_per_second": 8.538,
"step": 560000
},
{
"epoch": 23.6268600092737,
"grad_norm": 1.4755828380584717,
"learning_rate": 5.5867598507910655e-06,
"loss": 3.3478,
"step": 560500
},
{
"epoch": 23.64793660161025,
"grad_norm": 1.473509669303894,
"learning_rate": 5.501007589075162e-06,
"loss": 3.3485,
"step": 561000
},
{
"epoch": 23.669013193946803,
"grad_norm": 1.5190178155899048,
"learning_rate": 5.41525532735926e-06,
"loss": 3.3495,
"step": 561500
},
{
"epoch": 23.690089786283355,
"grad_norm": 1.511405348777771,
"learning_rate": 5.329503065643356e-06,
"loss": 3.3439,
"step": 562000
},
{
"epoch": 23.711166378619904,
"grad_norm": 1.5435580015182495,
"learning_rate": 5.243750803927454e-06,
"loss": 3.3451,
"step": 562500
},
{
"epoch": 23.732242970956456,
"grad_norm": 1.5060734748840332,
"learning_rate": 5.157998542211551e-06,
"loss": 3.3472,
"step": 563000
},
{
"epoch": 23.753319563293008,
"grad_norm": 1.4195207357406616,
"learning_rate": 5.072246280495649e-06,
"loss": 3.3444,
"step": 563500
},
{
"epoch": 23.774396155629557,
"grad_norm": 1.4588980674743652,
"learning_rate": 4.986494018779746e-06,
"loss": 3.3467,
"step": 564000
},
{
"epoch": 23.79547274796611,
"grad_norm": 1.4943127632141113,
"learning_rate": 4.900741757063843e-06,
"loss": 3.3461,
"step": 564500
},
{
"epoch": 23.81654934030266,
"grad_norm": 1.4811856746673584,
"learning_rate": 4.81498949534794e-06,
"loss": 3.344,
"step": 565000
},
{
"epoch": 23.83762593263921,
"grad_norm": 1.4292244911193848,
"learning_rate": 4.729237233632038e-06,
"loss": 3.3453,
"step": 565500
},
{
"epoch": 23.85870252497576,
"grad_norm": 1.4507802724838257,
"learning_rate": 4.643484971916134e-06,
"loss": 3.346,
"step": 566000
},
{
"epoch": 23.879779117312314,
"grad_norm": 1.4574283361434937,
"learning_rate": 4.557732710200232e-06,
"loss": 3.3453,
"step": 566500
},
{
"epoch": 23.900855709648862,
"grad_norm": 1.451655626296997,
"learning_rate": 4.471980448484329e-06,
"loss": 3.3466,
"step": 567000
},
{
"epoch": 23.921932301985414,
"grad_norm": 1.4816192388534546,
"learning_rate": 4.386228186768427e-06,
"loss": 3.3444,
"step": 567500
},
{
"epoch": 23.943008894321967,
"grad_norm": 1.4634467363357544,
"learning_rate": 4.300475925052524e-06,
"loss": 3.3455,
"step": 568000
},
{
"epoch": 23.96408548665852,
"grad_norm": 1.434103012084961,
"learning_rate": 4.214723663336621e-06,
"loss": 3.3486,
"step": 568500
},
{
"epoch": 23.985162078995067,
"grad_norm": 1.5023547410964966,
"learning_rate": 4.128971401620718e-06,
"loss": 3.345,
"step": 569000
},
{
"epoch": 24.00623867133162,
"grad_norm": 1.4730761051177979,
"learning_rate": 4.043219139904815e-06,
"loss": 3.3457,
"step": 569500
},
{
"epoch": 24.02731526366817,
"grad_norm": 1.4018100500106812,
"learning_rate": 3.957466878188912e-06,
"loss": 3.3446,
"step": 570000
},
{
"epoch": 24.02731526366817,
"eval_accuracy": 0.731118374607573,
"eval_loss": 3.1079559326171875,
"eval_runtime": 147.2532,
"eval_samples_per_second": 819.717,
"eval_steps_per_second": 8.543,
"step": 570000
},
{
"epoch": 24.04839185600472,
"grad_norm": 1.510614037513733,
"learning_rate": 3.87171461647301e-06,
"loss": 3.3491,
"step": 570500
},
{
"epoch": 24.069468448341272,
"grad_norm": 1.4496036767959595,
"learning_rate": 3.7859623547571067e-06,
"loss": 3.342,
"step": 571000
},
{
"epoch": 24.090545040677824,
"grad_norm": 1.500940203666687,
"learning_rate": 3.7002100930412046e-06,
"loss": 3.3407,
"step": 571500
},
{
"epoch": 24.111621633014373,
"grad_norm": 1.4029479026794434,
"learning_rate": 3.614457831325301e-06,
"loss": 3.3479,
"step": 572000
},
{
"epoch": 24.132698225350925,
"grad_norm": 1.4565962553024292,
"learning_rate": 3.5287055696093987e-06,
"loss": 3.3455,
"step": 572500
},
{
"epoch": 24.153774817687477,
"grad_norm": 1.4220225811004639,
"learning_rate": 3.4429533078934957e-06,
"loss": 3.3457,
"step": 573000
},
{
"epoch": 24.174851410024026,
"grad_norm": 1.4809374809265137,
"learning_rate": 3.357201046177593e-06,
"loss": 3.3462,
"step": 573500
},
{
"epoch": 24.195928002360578,
"grad_norm": 1.448825478553772,
"learning_rate": 3.2714487844616906e-06,
"loss": 3.3444,
"step": 574000
},
{
"epoch": 24.21700459469713,
"grad_norm": 1.517209768295288,
"learning_rate": 3.1856965227457872e-06,
"loss": 3.3453,
"step": 574500
},
{
"epoch": 24.238081187033682,
"grad_norm": 1.4786149263381958,
"learning_rate": 3.099944261029885e-06,
"loss": 3.349,
"step": 575000
},
{
"epoch": 24.25915777937023,
"grad_norm": 1.4762290716171265,
"learning_rate": 3.014191999313982e-06,
"loss": 3.3433,
"step": 575500
},
{
"epoch": 24.280234371706783,
"grad_norm": 1.4942190647125244,
"learning_rate": 2.928439737598079e-06,
"loss": 3.3425,
"step": 576000
},
{
"epoch": 24.301310964043335,
"grad_norm": 1.4617185592651367,
"learning_rate": 2.8426874758821767e-06,
"loss": 3.3441,
"step": 576500
},
{
"epoch": 24.322387556379883,
"grad_norm": 1.4893302917480469,
"learning_rate": 2.7569352141662737e-06,
"loss": 3.3397,
"step": 577000
},
{
"epoch": 24.343464148716436,
"grad_norm": 1.44925856590271,
"learning_rate": 2.6711829524503707e-06,
"loss": 3.34,
"step": 577500
},
{
"epoch": 24.364540741052988,
"grad_norm": 1.507400631904602,
"learning_rate": 2.585430690734468e-06,
"loss": 3.3417,
"step": 578000
},
{
"epoch": 24.385617333389536,
"grad_norm": 1.4852689504623413,
"learning_rate": 2.4996784290185657e-06,
"loss": 3.3436,
"step": 578500
},
{
"epoch": 24.40669392572609,
"grad_norm": 1.4584376811981201,
"learning_rate": 2.4139261673026627e-06,
"loss": 3.3399,
"step": 579000
},
{
"epoch": 24.42777051806264,
"grad_norm": 1.4095585346221924,
"learning_rate": 2.3281739055867597e-06,
"loss": 3.3422,
"step": 579500
},
{
"epoch": 24.44884711039919,
"grad_norm": 1.450351595878601,
"learning_rate": 2.242421643870857e-06,
"loss": 3.342,
"step": 580000
},
{
"epoch": 24.44884711039919,
"eval_accuracy": 0.7316728773368888,
"eval_loss": 3.1041817665100098,
"eval_runtime": 148.1057,
"eval_samples_per_second": 814.999,
"eval_steps_per_second": 8.494,
"step": 580000
},
{
"epoch": 24.46992370273574,
"grad_norm": 1.5080937147140503,
"learning_rate": 2.1566693821549547e-06,
"loss": 3.3427,
"step": 580500
},
{
"epoch": 24.491000295072293,
"grad_norm": 1.4717568159103394,
"learning_rate": 2.0709171204390517e-06,
"loss": 3.3456,
"step": 581000
},
{
"epoch": 24.512076887408845,
"grad_norm": 1.4973249435424805,
"learning_rate": 1.985164858723149e-06,
"loss": 3.3404,
"step": 581500
},
{
"epoch": 24.533153479745394,
"grad_norm": 1.4677600860595703,
"learning_rate": 1.8994125970072462e-06,
"loss": 3.3457,
"step": 582000
},
{
"epoch": 24.554230072081946,
"grad_norm": 1.4844106435775757,
"learning_rate": 1.8136603352913432e-06,
"loss": 3.3389,
"step": 582500
},
{
"epoch": 24.5753066644185,
"grad_norm": 1.5207314491271973,
"learning_rate": 1.7279080735754405e-06,
"loss": 3.3458,
"step": 583000
},
{
"epoch": 24.596383256755047,
"grad_norm": 1.5356721878051758,
"learning_rate": 1.642155811859538e-06,
"loss": 3.3446,
"step": 583500
},
{
"epoch": 24.6174598490916,
"grad_norm": 1.4863603115081787,
"learning_rate": 1.5564035501436352e-06,
"loss": 3.3413,
"step": 584000
},
{
"epoch": 24.63853644142815,
"grad_norm": 1.5266973972320557,
"learning_rate": 1.4706512884277324e-06,
"loss": 3.3437,
"step": 584500
},
{
"epoch": 24.6596130337647,
"grad_norm": 1.4621362686157227,
"learning_rate": 1.3848990267118295e-06,
"loss": 3.3397,
"step": 585000
},
{
"epoch": 24.680689626101252,
"grad_norm": 1.4747660160064697,
"learning_rate": 1.299146764995927e-06,
"loss": 3.3415,
"step": 585500
},
{
"epoch": 24.701766218437804,
"grad_norm": 1.470448613166809,
"learning_rate": 1.213394503280024e-06,
"loss": 3.3409,
"step": 586000
},
{
"epoch": 24.722842810774353,
"grad_norm": 1.4936798810958862,
"learning_rate": 1.1276422415641212e-06,
"loss": 3.3404,
"step": 586500
},
{
"epoch": 24.743919403110905,
"grad_norm": 1.4461981058120728,
"learning_rate": 1.0418899798482187e-06,
"loss": 3.3428,
"step": 587000
},
{
"epoch": 24.764995995447457,
"grad_norm": 1.4625365734100342,
"learning_rate": 9.561377181323157e-07,
"loss": 3.3404,
"step": 587500
},
{
"epoch": 24.786072587784005,
"grad_norm": 1.469041347503662,
"learning_rate": 8.70385456416413e-07,
"loss": 3.3411,
"step": 588000
},
{
"epoch": 24.807149180120557,
"grad_norm": 1.4133771657943726,
"learning_rate": 7.846331947005103e-07,
"loss": 3.3431,
"step": 588500
},
{
"epoch": 24.82822577245711,
"grad_norm": 1.4636611938476562,
"learning_rate": 6.988809329846075e-07,
"loss": 3.3388,
"step": 589000
},
{
"epoch": 24.84930236479366,
"grad_norm": 1.5076528787612915,
"learning_rate": 6.131286712687047e-07,
"loss": 3.3423,
"step": 589500
},
{
"epoch": 24.87037895713021,
"grad_norm": 1.4504926204681396,
"learning_rate": 5.27376409552802e-07,
"loss": 3.3397,
"step": 590000
},
{
"epoch": 24.87037895713021,
"eval_accuracy": 0.7314052805426616,
"eval_loss": 3.1047677993774414,
"eval_runtime": 147.6671,
"eval_samples_per_second": 817.419,
"eval_steps_per_second": 8.519,
"step": 590000
},
{
"epoch": 24.891455549466762,
"grad_norm": 1.4451123476028442,
"learning_rate": 4.4162414783689917e-07,
"loss": 3.3421,
"step": 590500
},
{
"epoch": 24.912532141803315,
"grad_norm": 1.4576573371887207,
"learning_rate": 3.558718861209964e-07,
"loss": 3.34,
"step": 591000
},
{
"epoch": 24.933608734139863,
"grad_norm": 1.4591474533081055,
"learning_rate": 2.701196244050937e-07,
"loss": 3.3413,
"step": 591500
},
{
"epoch": 24.954685326476415,
"grad_norm": 1.5335863828659058,
"learning_rate": 1.8436736268919094e-07,
"loss": 3.3415,
"step": 592000
},
{
"epoch": 24.975761918812967,
"grad_norm": 1.4356646537780762,
"learning_rate": 9.861510097328816e-08,
"loss": 3.3401,
"step": 592500
},
{
"epoch": 24.996838511149516,
"grad_norm": 1.4528509378433228,
"learning_rate": 1.2862839257385414e-08,
"loss": 3.3406,
"step": 593000
},
{
"epoch": 25.0,
"step": 593075,
"total_flos": 7.65181665004032e+18,
"train_loss": 3.640720207871391,
"train_runtime": 132653.8084,
"train_samples_per_second": 429.189,
"train_steps_per_second": 4.471
}
],
"logging_steps": 500,
"max_steps": 593075,
"num_input_tokens_seen": 0,
"num_train_epochs": 25,
"save_steps": 10000,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 7.65181665004032e+18,
"train_batch_size": 96,
"trial_name": null,
"trial_params": null
}