bert_base_train_book / trainer_state.json
gokulsrinivasagan's picture
End of training
05535cc verified
{
"best_metric": 1.0764633417129517,
"best_model_checkpoint": "bert_base_train_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": 2.2031943798065186,
"learning_rate": 5e-06,
"loss": 8.8802,
"step": 500
},
{
"epoch": 0.04215318467310205,
"grad_norm": 1.0319809913635254,
"learning_rate": 1e-05,
"loss": 6.6526,
"step": 1000
},
{
"epoch": 0.06322977700965308,
"grad_norm": 1.146312952041626,
"learning_rate": 1.5e-05,
"loss": 6.0188,
"step": 1500
},
{
"epoch": 0.0843063693462041,
"grad_norm": 0.8987782001495361,
"learning_rate": 2e-05,
"loss": 5.9028,
"step": 2000
},
{
"epoch": 0.10538296168275513,
"grad_norm": 0.7788671851158142,
"learning_rate": 2.5e-05,
"loss": 5.8469,
"step": 2500
},
{
"epoch": 0.12645955401930617,
"grad_norm": 0.6918198466300964,
"learning_rate": 3e-05,
"loss": 5.8091,
"step": 3000
},
{
"epoch": 0.14753614635585718,
"grad_norm": 0.7891680002212524,
"learning_rate": 3.5e-05,
"loss": 5.7804,
"step": 3500
},
{
"epoch": 0.1686127386924082,
"grad_norm": 0.696956217288971,
"learning_rate": 4e-05,
"loss": 5.7546,
"step": 4000
},
{
"epoch": 0.18968933102895924,
"grad_norm": 0.8337274193763733,
"learning_rate": 4.5e-05,
"loss": 5.7314,
"step": 4500
},
{
"epoch": 0.21076592336551025,
"grad_norm": 0.6991694569587708,
"learning_rate": 5e-05,
"loss": 5.7164,
"step": 5000
},
{
"epoch": 0.2318425157020613,
"grad_norm": 0.6847338676452637,
"learning_rate": 5.500000000000001e-05,
"loss": 5.7051,
"step": 5500
},
{
"epoch": 0.25291910803861234,
"grad_norm": 0.6237599849700928,
"learning_rate": 6e-05,
"loss": 5.6873,
"step": 6000
},
{
"epoch": 0.27399570037516335,
"grad_norm": 0.6555923223495483,
"learning_rate": 6.500000000000001e-05,
"loss": 5.6826,
"step": 6500
},
{
"epoch": 0.29507229271171437,
"grad_norm": 0.6365340948104858,
"learning_rate": 7e-05,
"loss": 5.6685,
"step": 7000
},
{
"epoch": 0.3161488850482654,
"grad_norm": 0.5584865808486938,
"learning_rate": 7.500000000000001e-05,
"loss": 5.6616,
"step": 7500
},
{
"epoch": 0.3372254773848164,
"grad_norm": 0.5515204071998596,
"learning_rate": 8e-05,
"loss": 5.6604,
"step": 8000
},
{
"epoch": 0.35830206972136747,
"grad_norm": 0.5933423042297363,
"learning_rate": 8.5e-05,
"loss": 5.6481,
"step": 8500
},
{
"epoch": 0.3793786620579185,
"grad_norm": 0.6046615839004517,
"learning_rate": 9e-05,
"loss": 5.6393,
"step": 9000
},
{
"epoch": 0.4004552543944695,
"grad_norm": 0.5836647748947144,
"learning_rate": 9.5e-05,
"loss": 5.6329,
"step": 9500
},
{
"epoch": 0.4215318467310205,
"grad_norm": 0.5906765460968018,
"learning_rate": 0.0001,
"loss": 5.6277,
"step": 10000
},
{
"epoch": 0.4215318467310205,
"eval_accuracy": 0.16478660657712904,
"eval_loss": 5.467868328094482,
"eval_runtime": 178.9268,
"eval_samples_per_second": 674.611,
"eval_steps_per_second": 7.031,
"step": 10000
},
{
"epoch": 0.4426084390675716,
"grad_norm": 0.5030741691589355,
"learning_rate": 9.99142477382841e-05,
"loss": 5.6192,
"step": 10500
},
{
"epoch": 0.4636850314041226,
"grad_norm": 0.5009177923202515,
"learning_rate": 9.98284954765682e-05,
"loss": 5.6146,
"step": 11000
},
{
"epoch": 0.4847616237406736,
"grad_norm": 0.6013538241386414,
"learning_rate": 9.97427432148523e-05,
"loss": 5.6135,
"step": 11500
},
{
"epoch": 0.5058382160772247,
"grad_norm": 0.5553770065307617,
"learning_rate": 9.96569909531364e-05,
"loss": 5.6035,
"step": 12000
},
{
"epoch": 0.5269148084137757,
"grad_norm": 0.5942257642745972,
"learning_rate": 9.957123869142048e-05,
"loss": 5.5983,
"step": 12500
},
{
"epoch": 0.5479914007503267,
"grad_norm": 0.5645352602005005,
"learning_rate": 9.94854864297046e-05,
"loss": 5.5938,
"step": 13000
},
{
"epoch": 0.5690679930868777,
"grad_norm": 0.5677357912063599,
"learning_rate": 9.939973416798869e-05,
"loss": 5.5844,
"step": 13500
},
{
"epoch": 0.5901445854234287,
"grad_norm": 0.5426216125488281,
"learning_rate": 9.931398190627278e-05,
"loss": 5.5807,
"step": 14000
},
{
"epoch": 0.6112211777599798,
"grad_norm": 0.5221067070960999,
"learning_rate": 9.922822964455688e-05,
"loss": 5.573,
"step": 14500
},
{
"epoch": 0.6322977700965308,
"grad_norm": 0.5111722946166992,
"learning_rate": 9.914247738284099e-05,
"loss": 5.571,
"step": 15000
},
{
"epoch": 0.6533743624330818,
"grad_norm": 0.4875262379646301,
"learning_rate": 9.905672512112507e-05,
"loss": 5.5609,
"step": 15500
},
{
"epoch": 0.6744509547696328,
"grad_norm": 0.49939367175102234,
"learning_rate": 9.897097285940917e-05,
"loss": 5.5603,
"step": 16000
},
{
"epoch": 0.6955275471061839,
"grad_norm": 0.501538872718811,
"learning_rate": 9.888522059769327e-05,
"loss": 5.5525,
"step": 16500
},
{
"epoch": 0.7166041394427349,
"grad_norm": 0.5125389099121094,
"learning_rate": 9.879946833597737e-05,
"loss": 5.5493,
"step": 17000
},
{
"epoch": 0.737680731779286,
"grad_norm": 0.510980486869812,
"learning_rate": 9.871371607426147e-05,
"loss": 5.5418,
"step": 17500
},
{
"epoch": 0.758757324115837,
"grad_norm": 0.5121689438819885,
"learning_rate": 9.862796381254556e-05,
"loss": 5.5463,
"step": 18000
},
{
"epoch": 0.779833916452388,
"grad_norm": 0.4904923141002655,
"learning_rate": 9.854221155082965e-05,
"loss": 5.538,
"step": 18500
},
{
"epoch": 0.800910508788939,
"grad_norm": 0.5604308843612671,
"learning_rate": 9.845645928911376e-05,
"loss": 5.5394,
"step": 19000
},
{
"epoch": 0.82198710112549,
"grad_norm": 0.5072571039199829,
"learning_rate": 9.837070702739786e-05,
"loss": 5.5362,
"step": 19500
},
{
"epoch": 0.843063693462041,
"grad_norm": 0.4983499050140381,
"learning_rate": 9.828495476568194e-05,
"loss": 5.5308,
"step": 20000
},
{
"epoch": 0.843063693462041,
"eval_accuracy": 0.1655790521794567,
"eval_loss": 5.39214563369751,
"eval_runtime": 178.7595,
"eval_samples_per_second": 675.242,
"eval_steps_per_second": 7.037,
"step": 20000
},
{
"epoch": 0.864140285798592,
"grad_norm": 0.48933616280555725,
"learning_rate": 9.819920250396604e-05,
"loss": 5.5265,
"step": 20500
},
{
"epoch": 0.8852168781351432,
"grad_norm": 0.49544665217399597,
"learning_rate": 9.811345024225015e-05,
"loss": 5.5225,
"step": 21000
},
{
"epoch": 0.9062934704716942,
"grad_norm": 0.5077598094940186,
"learning_rate": 9.802769798053424e-05,
"loss": 5.5221,
"step": 21500
},
{
"epoch": 0.9273700628082452,
"grad_norm": 0.5252205729484558,
"learning_rate": 9.794194571881834e-05,
"loss": 5.5157,
"step": 22000
},
{
"epoch": 0.9484466551447962,
"grad_norm": 0.5447311401367188,
"learning_rate": 9.785619345710244e-05,
"loss": 5.5143,
"step": 22500
},
{
"epoch": 0.9695232474813472,
"grad_norm": 0.8156083226203918,
"learning_rate": 9.777044119538653e-05,
"loss": 5.5106,
"step": 23000
},
{
"epoch": 0.9905998398178982,
"grad_norm": 0.597063422203064,
"learning_rate": 9.768468893367063e-05,
"loss": 5.507,
"step": 23500
},
{
"epoch": 1.0116764321544494,
"grad_norm": 0.5273151993751526,
"learning_rate": 9.759893667195473e-05,
"loss": 5.5058,
"step": 24000
},
{
"epoch": 1.0327530244910004,
"grad_norm": 0.5183907151222229,
"learning_rate": 9.751318441023883e-05,
"loss": 5.5044,
"step": 24500
},
{
"epoch": 1.0538296168275514,
"grad_norm": 0.5336084365844727,
"learning_rate": 9.742743214852293e-05,
"loss": 5.5026,
"step": 25000
},
{
"epoch": 1.0749062091641024,
"grad_norm": 0.49004632234573364,
"learning_rate": 9.734167988680703e-05,
"loss": 5.4948,
"step": 25500
},
{
"epoch": 1.0959828015006534,
"grad_norm": 0.49460968375205994,
"learning_rate": 9.725592762509111e-05,
"loss": 5.4947,
"step": 26000
},
{
"epoch": 1.1170593938372044,
"grad_norm": 0.5054150223731995,
"learning_rate": 9.717017536337522e-05,
"loss": 5.4915,
"step": 26500
},
{
"epoch": 1.1381359861737554,
"grad_norm": 0.5127531886100769,
"learning_rate": 9.708442310165932e-05,
"loss": 5.4948,
"step": 27000
},
{
"epoch": 1.1592125785103065,
"grad_norm": 0.5035219192504883,
"learning_rate": 9.69986708399434e-05,
"loss": 5.4894,
"step": 27500
},
{
"epoch": 1.1802891708468575,
"grad_norm": 0.4969281554222107,
"learning_rate": 9.69129185782275e-05,
"loss": 5.4926,
"step": 28000
},
{
"epoch": 1.2013657631834085,
"grad_norm": 0.5207239985466003,
"learning_rate": 9.68271663165116e-05,
"loss": 5.4842,
"step": 28500
},
{
"epoch": 1.2224423555199595,
"grad_norm": 0.522996187210083,
"learning_rate": 9.67414140547957e-05,
"loss": 5.4829,
"step": 29000
},
{
"epoch": 1.2435189478565105,
"grad_norm": 0.5490512251853943,
"learning_rate": 9.66556617930798e-05,
"loss": 5.4801,
"step": 29500
},
{
"epoch": 1.2645955401930615,
"grad_norm": 0.5085314512252808,
"learning_rate": 9.656990953136388e-05,
"loss": 5.4819,
"step": 30000
},
{
"epoch": 1.2645955401930615,
"eval_accuracy": 0.1668335840864588,
"eval_loss": 5.355852127075195,
"eval_runtime": 197.45,
"eval_samples_per_second": 611.324,
"eval_steps_per_second": 6.371,
"step": 30000
},
{
"epoch": 1.2856721325296125,
"grad_norm": 0.47031304240226746,
"learning_rate": 9.6484157269648e-05,
"loss": 5.4744,
"step": 30500
},
{
"epoch": 1.3067487248661636,
"grad_norm": 0.4906262755393982,
"learning_rate": 9.639840500793209e-05,
"loss": 5.4786,
"step": 31000
},
{
"epoch": 1.3278253172027146,
"grad_norm": 0.472972571849823,
"learning_rate": 9.631265274621618e-05,
"loss": 5.4754,
"step": 31500
},
{
"epoch": 1.3489019095392658,
"grad_norm": 0.4908204972743988,
"learning_rate": 9.622690048450028e-05,
"loss": 5.4776,
"step": 32000
},
{
"epoch": 1.3699785018758166,
"grad_norm": 0.4790041744709015,
"learning_rate": 9.614114822278439e-05,
"loss": 5.4716,
"step": 32500
},
{
"epoch": 1.3910550942123678,
"grad_norm": 0.49558982253074646,
"learning_rate": 9.605539596106847e-05,
"loss": 5.4761,
"step": 33000
},
{
"epoch": 1.4121316865489189,
"grad_norm": 0.4688015580177307,
"learning_rate": 9.596964369935257e-05,
"loss": 5.4711,
"step": 33500
},
{
"epoch": 1.4332082788854699,
"grad_norm": 0.5014123320579529,
"learning_rate": 9.588389143763667e-05,
"loss": 5.4683,
"step": 34000
},
{
"epoch": 1.4542848712220209,
"grad_norm": 0.49400103092193604,
"learning_rate": 9.579813917592077e-05,
"loss": 5.4692,
"step": 34500
},
{
"epoch": 1.475361463558572,
"grad_norm": 0.5063204169273376,
"learning_rate": 9.571238691420487e-05,
"loss": 5.4681,
"step": 35000
},
{
"epoch": 1.496438055895123,
"grad_norm": 1.2554806470870972,
"learning_rate": 9.562663465248896e-05,
"loss": 5.463,
"step": 35500
},
{
"epoch": 1.517514648231674,
"grad_norm": 0.5049524307250977,
"learning_rate": 9.554088239077306e-05,
"loss": 5.4581,
"step": 36000
},
{
"epoch": 1.538591240568225,
"grad_norm": 0.46852052211761475,
"learning_rate": 9.545513012905716e-05,
"loss": 5.4614,
"step": 36500
},
{
"epoch": 1.559667832904776,
"grad_norm": 0.48980557918548584,
"learning_rate": 9.536937786734126e-05,
"loss": 5.4637,
"step": 37000
},
{
"epoch": 1.580744425241327,
"grad_norm": 0.48720183968544006,
"learning_rate": 9.528362560562534e-05,
"loss": 5.4599,
"step": 37500
},
{
"epoch": 1.601821017577878,
"grad_norm": 0.49417823553085327,
"learning_rate": 9.519787334390946e-05,
"loss": 5.4588,
"step": 38000
},
{
"epoch": 1.622897609914429,
"grad_norm": 0.4943360686302185,
"learning_rate": 9.511212108219355e-05,
"loss": 5.4556,
"step": 38500
},
{
"epoch": 1.64397420225098,
"grad_norm": 0.4666002094745636,
"learning_rate": 9.502636882047764e-05,
"loss": 5.456,
"step": 39000
},
{
"epoch": 1.6650507945875312,
"grad_norm": 0.46367841958999634,
"learning_rate": 9.494061655876174e-05,
"loss": 5.4571,
"step": 39500
},
{
"epoch": 1.686127386924082,
"grad_norm": 0.4495871365070343,
"learning_rate": 9.485486429704585e-05,
"loss": 5.4576,
"step": 40000
},
{
"epoch": 1.686127386924082,
"eval_accuracy": 0.16687209812776124,
"eval_loss": 5.332687854766846,
"eval_runtime": 194.9046,
"eval_samples_per_second": 619.308,
"eval_steps_per_second": 6.454,
"step": 40000
},
{
"epoch": 1.7072039792606333,
"grad_norm": 0.5109173059463501,
"learning_rate": 9.476911203532993e-05,
"loss": 5.45,
"step": 40500
},
{
"epoch": 1.728280571597184,
"grad_norm": 0.46426981687545776,
"learning_rate": 9.468335977361403e-05,
"loss": 5.452,
"step": 41000
},
{
"epoch": 1.7493571639337353,
"grad_norm": 0.4695836305618286,
"learning_rate": 9.459760751189813e-05,
"loss": 5.4536,
"step": 41500
},
{
"epoch": 1.770433756270286,
"grad_norm": 0.49657121300697327,
"learning_rate": 9.451185525018223e-05,
"loss": 5.4524,
"step": 42000
},
{
"epoch": 1.7915103486068373,
"grad_norm": 0.4876340925693512,
"learning_rate": 9.442610298846633e-05,
"loss": 5.4484,
"step": 42500
},
{
"epoch": 1.8125869409433881,
"grad_norm": 0.46910014748573303,
"learning_rate": 9.434035072675043e-05,
"loss": 5.4462,
"step": 43000
},
{
"epoch": 1.8336635332799394,
"grad_norm": 0.45760712027549744,
"learning_rate": 9.425459846503451e-05,
"loss": 5.4441,
"step": 43500
},
{
"epoch": 1.8547401256164904,
"grad_norm": 0.46959230303764343,
"learning_rate": 9.416884620331862e-05,
"loss": 5.4458,
"step": 44000
},
{
"epoch": 1.8758167179530414,
"grad_norm": 0.4537522792816162,
"learning_rate": 9.408309394160272e-05,
"loss": 5.4491,
"step": 44500
},
{
"epoch": 1.8968933102895924,
"grad_norm": 0.4474141299724579,
"learning_rate": 9.39973416798868e-05,
"loss": 5.441,
"step": 45000
},
{
"epoch": 1.9179699026261434,
"grad_norm": 0.4627447724342346,
"learning_rate": 9.39115894181709e-05,
"loss": 5.4411,
"step": 45500
},
{
"epoch": 1.9390464949626944,
"grad_norm": 0.4754473865032196,
"learning_rate": 9.382583715645502e-05,
"loss": 5.4404,
"step": 46000
},
{
"epoch": 1.9601230872992454,
"grad_norm": 0.4617035388946533,
"learning_rate": 9.37400848947391e-05,
"loss": 5.441,
"step": 46500
},
{
"epoch": 1.9811996796357965,
"grad_norm": 0.45778870582580566,
"learning_rate": 9.36543326330232e-05,
"loss": 5.449,
"step": 47000
},
{
"epoch": 2.0022762719723475,
"grad_norm": 0.46125879883766174,
"learning_rate": 9.35685803713073e-05,
"loss": 5.4479,
"step": 47500
},
{
"epoch": 2.0233528643088987,
"grad_norm": 0.46388059854507446,
"learning_rate": 9.34828281095914e-05,
"loss": 5.4362,
"step": 48000
},
{
"epoch": 2.0444294566454495,
"grad_norm": 0.4571990966796875,
"learning_rate": 9.339707584787549e-05,
"loss": 5.4359,
"step": 48500
},
{
"epoch": 2.0655060489820007,
"grad_norm": 0.44776904582977295,
"learning_rate": 9.331132358615959e-05,
"loss": 5.4338,
"step": 49000
},
{
"epoch": 2.0865826413185515,
"grad_norm": 0.470447838306427,
"learning_rate": 9.322557132444369e-05,
"loss": 5.433,
"step": 49500
},
{
"epoch": 2.1076592336551028,
"grad_norm": 0.44955974817276,
"learning_rate": 9.313981906272779e-05,
"loss": 5.434,
"step": 50000
},
{
"epoch": 2.1076592336551028,
"eval_accuracy": 0.16714483173892664,
"eval_loss": 5.319274425506592,
"eval_runtime": 207.8191,
"eval_samples_per_second": 580.822,
"eval_steps_per_second": 6.053,
"step": 50000
},
{
"epoch": 2.1287358259916536,
"grad_norm": 0.45122262835502625,
"learning_rate": 9.305406680101189e-05,
"loss": 5.4359,
"step": 50500
},
{
"epoch": 2.149812418328205,
"grad_norm": 0.4674409329891205,
"learning_rate": 9.296831453929597e-05,
"loss": 5.4365,
"step": 51000
},
{
"epoch": 2.1708890106647556,
"grad_norm": 0.46792227029800415,
"learning_rate": 9.288256227758008e-05,
"loss": 5.4323,
"step": 51500
},
{
"epoch": 2.191965603001307,
"grad_norm": 0.4598369300365448,
"learning_rate": 9.279681001586418e-05,
"loss": 5.4299,
"step": 52000
},
{
"epoch": 2.2130421953378576,
"grad_norm": 0.4470933973789215,
"learning_rate": 9.271105775414827e-05,
"loss": 5.4283,
"step": 52500
},
{
"epoch": 2.234118787674409,
"grad_norm": 0.4459787607192993,
"learning_rate": 9.262530549243236e-05,
"loss": 5.4291,
"step": 53000
},
{
"epoch": 2.2551953800109596,
"grad_norm": 0.44385844469070435,
"learning_rate": 9.253955323071646e-05,
"loss": 5.4308,
"step": 53500
},
{
"epoch": 2.276271972347511,
"grad_norm": 0.48831382393836975,
"learning_rate": 9.245380096900056e-05,
"loss": 5.4357,
"step": 54000
},
{
"epoch": 2.2973485646840617,
"grad_norm": 0.45131176710128784,
"learning_rate": 9.236804870728466e-05,
"loss": 5.4262,
"step": 54500
},
{
"epoch": 2.318425157020613,
"grad_norm": 0.45769938826560974,
"learning_rate": 9.228229644556876e-05,
"loss": 5.4277,
"step": 55000
},
{
"epoch": 2.339501749357164,
"grad_norm": 0.4372427761554718,
"learning_rate": 9.219654418385286e-05,
"loss": 5.4242,
"step": 55500
},
{
"epoch": 2.360578341693715,
"grad_norm": 0.4551551640033722,
"learning_rate": 9.211079192213695e-05,
"loss": 5.4306,
"step": 56000
},
{
"epoch": 2.381654934030266,
"grad_norm": 0.4342139959335327,
"learning_rate": 9.202503966042105e-05,
"loss": 5.4254,
"step": 56500
},
{
"epoch": 2.402731526366817,
"grad_norm": 0.43151190876960754,
"learning_rate": 9.193928739870514e-05,
"loss": 5.4264,
"step": 57000
},
{
"epoch": 2.423808118703368,
"grad_norm": 0.4663863480091095,
"learning_rate": 9.185353513698925e-05,
"loss": 5.4225,
"step": 57500
},
{
"epoch": 2.444884711039919,
"grad_norm": 0.4561541676521301,
"learning_rate": 9.176778287527335e-05,
"loss": 5.4219,
"step": 58000
},
{
"epoch": 2.4659613033764702,
"grad_norm": 0.4475264847278595,
"learning_rate": 9.168203061355743e-05,
"loss": 5.4222,
"step": 58500
},
{
"epoch": 2.487037895713021,
"grad_norm": 0.4333769381046295,
"learning_rate": 9.159627835184153e-05,
"loss": 5.4236,
"step": 59000
},
{
"epoch": 2.5081144880495723,
"grad_norm": 0.45078328251838684,
"learning_rate": 9.151052609012564e-05,
"loss": 5.4239,
"step": 59500
},
{
"epoch": 2.529191080386123,
"grad_norm": 0.43709686398506165,
"learning_rate": 9.142477382840973e-05,
"loss": 5.423,
"step": 60000
},
{
"epoch": 2.529191080386123,
"eval_accuracy": 0.16761679148354858,
"eval_loss": 5.306380271911621,
"eval_runtime": 205.9221,
"eval_samples_per_second": 586.173,
"eval_steps_per_second": 6.109,
"step": 60000
},
{
"epoch": 2.5502676727226743,
"grad_norm": 0.453595906496048,
"learning_rate": 9.133902156669383e-05,
"loss": 5.42,
"step": 60500
},
{
"epoch": 2.571344265059225,
"grad_norm": 0.4419400095939636,
"learning_rate": 9.125326930497792e-05,
"loss": 5.4213,
"step": 61000
},
{
"epoch": 2.5924208573957763,
"grad_norm": 0.4547574520111084,
"learning_rate": 9.116751704326202e-05,
"loss": 5.4182,
"step": 61500
},
{
"epoch": 2.613497449732327,
"grad_norm": 0.425094336271286,
"learning_rate": 9.108176478154612e-05,
"loss": 5.4162,
"step": 62000
},
{
"epoch": 2.6345740420688784,
"grad_norm": 0.4348224401473999,
"learning_rate": 9.09960125198302e-05,
"loss": 5.4232,
"step": 62500
},
{
"epoch": 2.655650634405429,
"grad_norm": 0.43422162532806396,
"learning_rate": 9.091026025811432e-05,
"loss": 5.4193,
"step": 63000
},
{
"epoch": 2.6767272267419804,
"grad_norm": 0.47666481137275696,
"learning_rate": 9.082450799639842e-05,
"loss": 5.4159,
"step": 63500
},
{
"epoch": 2.6978038190785316,
"grad_norm": 0.43942445516586304,
"learning_rate": 9.07387557346825e-05,
"loss": 5.4181,
"step": 64000
},
{
"epoch": 2.7188804114150824,
"grad_norm": 0.41471990942955017,
"learning_rate": 9.06530034729666e-05,
"loss": 5.4172,
"step": 64500
},
{
"epoch": 2.739957003751633,
"grad_norm": 0.4733496308326721,
"learning_rate": 9.05672512112507e-05,
"loss": 5.4169,
"step": 65000
},
{
"epoch": 2.7610335960881844,
"grad_norm": 0.43646612763404846,
"learning_rate": 9.04814989495348e-05,
"loss": 5.4112,
"step": 65500
},
{
"epoch": 2.7821101884247357,
"grad_norm": 0.44608986377716064,
"learning_rate": 9.03957466878189e-05,
"loss": 5.42,
"step": 66000
},
{
"epoch": 2.8031867807612865,
"grad_norm": 0.4417068064212799,
"learning_rate": 9.030999442610299e-05,
"loss": 5.4163,
"step": 66500
},
{
"epoch": 2.8242633730978377,
"grad_norm": 0.4196968376636505,
"learning_rate": 9.022424216438709e-05,
"loss": 5.4099,
"step": 67000
},
{
"epoch": 2.8453399654343885,
"grad_norm": 0.4380576014518738,
"learning_rate": 9.013848990267119e-05,
"loss": 5.4113,
"step": 67500
},
{
"epoch": 2.8664165577709397,
"grad_norm": 0.4311133027076721,
"learning_rate": 9.005273764095529e-05,
"loss": 5.4091,
"step": 68000
},
{
"epoch": 2.8874931501074905,
"grad_norm": 0.45071834325790405,
"learning_rate": 8.996698537923937e-05,
"loss": 5.4155,
"step": 68500
},
{
"epoch": 2.9085697424440418,
"grad_norm": 0.43786630034446716,
"learning_rate": 8.988123311752348e-05,
"loss": 5.4102,
"step": 69000
},
{
"epoch": 2.9296463347805926,
"grad_norm": 0.4358339309692383,
"learning_rate": 8.979548085580758e-05,
"loss": 5.4106,
"step": 69500
},
{
"epoch": 2.950722927117144,
"grad_norm": 0.439919114112854,
"learning_rate": 8.970972859409167e-05,
"loss": 5.4078,
"step": 70000
},
{
"epoch": 2.950722927117144,
"eval_accuracy": 0.16701517385580744,
"eval_loss": 5.301105499267578,
"eval_runtime": 200.0013,
"eval_samples_per_second": 603.526,
"eval_steps_per_second": 6.29,
"step": 70000
},
{
"epoch": 2.9717995194536946,
"grad_norm": 0.44717276096343994,
"learning_rate": 8.962397633237576e-05,
"loss": 5.4101,
"step": 70500
},
{
"epoch": 2.992876111790246,
"grad_norm": 0.43964412808418274,
"learning_rate": 8.953822407065988e-05,
"loss": 5.4072,
"step": 71000
},
{
"epoch": 3.0139527041267966,
"grad_norm": 0.4269196689128876,
"learning_rate": 8.945247180894396e-05,
"loss": 5.4065,
"step": 71500
},
{
"epoch": 3.035029296463348,
"grad_norm": 0.46939900517463684,
"learning_rate": 8.936671954722806e-05,
"loss": 5.4059,
"step": 72000
},
{
"epoch": 3.0561058887998986,
"grad_norm": 0.42411503195762634,
"learning_rate": 8.928096728551216e-05,
"loss": 5.4079,
"step": 72500
},
{
"epoch": 3.07718248113645,
"grad_norm": 0.43144094944000244,
"learning_rate": 8.919521502379626e-05,
"loss": 5.4054,
"step": 73000
},
{
"epoch": 3.0982590734730007,
"grad_norm": 0.45157086849212646,
"learning_rate": 8.910946276208035e-05,
"loss": 5.4058,
"step": 73500
},
{
"epoch": 3.119335665809552,
"grad_norm": 0.42591992020606995,
"learning_rate": 8.902371050036445e-05,
"loss": 5.4014,
"step": 74000
},
{
"epoch": 3.140412258146103,
"grad_norm": 0.4491913616657257,
"learning_rate": 8.893795823864855e-05,
"loss": 5.4062,
"step": 74500
},
{
"epoch": 3.161488850482654,
"grad_norm": 0.4271566569805145,
"learning_rate": 8.885220597693265e-05,
"loss": 5.4065,
"step": 75000
},
{
"epoch": 3.182565442819205,
"grad_norm": 0.44306832551956177,
"learning_rate": 8.876645371521675e-05,
"loss": 5.404,
"step": 75500
},
{
"epoch": 3.203642035155756,
"grad_norm": 0.43039774894714355,
"learning_rate": 8.868070145350083e-05,
"loss": 5.4038,
"step": 76000
},
{
"epoch": 3.224718627492307,
"grad_norm": 0.4212704300880432,
"learning_rate": 8.859494919178493e-05,
"loss": 5.4043,
"step": 76500
},
{
"epoch": 3.245795219828858,
"grad_norm": 0.4221978783607483,
"learning_rate": 8.850919693006904e-05,
"loss": 5.4004,
"step": 77000
},
{
"epoch": 3.2668718121654092,
"grad_norm": 0.42982056736946106,
"learning_rate": 8.842344466835313e-05,
"loss": 5.3988,
"step": 77500
},
{
"epoch": 3.28794840450196,
"grad_norm": 0.436352401971817,
"learning_rate": 8.833769240663723e-05,
"loss": 5.3969,
"step": 78000
},
{
"epoch": 3.3090249968385113,
"grad_norm": 0.45014292001724243,
"learning_rate": 8.825194014492132e-05,
"loss": 5.4013,
"step": 78500
},
{
"epoch": 3.330101589175062,
"grad_norm": 0.43769505620002747,
"learning_rate": 8.816618788320542e-05,
"loss": 5.4004,
"step": 79000
},
{
"epoch": 3.3511781815116133,
"grad_norm": 0.45284420251846313,
"learning_rate": 8.808043562148952e-05,
"loss": 5.4033,
"step": 79500
},
{
"epoch": 3.372254773848164,
"grad_norm": 0.4484846889972687,
"learning_rate": 8.799468335977362e-05,
"loss": 5.3996,
"step": 80000
},
{
"epoch": 3.372254773848164,
"eval_accuracy": 0.16752518254092066,
"eval_loss": 5.289068698883057,
"eval_runtime": 201.8403,
"eval_samples_per_second": 598.027,
"eval_steps_per_second": 6.233,
"step": 80000
},
{
"epoch": 3.3933313661847153,
"grad_norm": 0.4387897253036499,
"learning_rate": 8.790893109805772e-05,
"loss": 5.4002,
"step": 80500
},
{
"epoch": 3.414407958521266,
"grad_norm": 0.42578163743019104,
"learning_rate": 8.782317883634182e-05,
"loss": 5.397,
"step": 81000
},
{
"epoch": 3.4354845508578173,
"grad_norm": 0.4302797317504883,
"learning_rate": 8.773742657462591e-05,
"loss": 5.394,
"step": 81500
},
{
"epoch": 3.456561143194368,
"grad_norm": 0.4212122857570648,
"learning_rate": 8.765167431291e-05,
"loss": 5.3978,
"step": 82000
},
{
"epoch": 3.4776377355309194,
"grad_norm": 0.42556172609329224,
"learning_rate": 8.756592205119411e-05,
"loss": 5.3969,
"step": 82500
},
{
"epoch": 3.4987143278674706,
"grad_norm": 0.45187705755233765,
"learning_rate": 8.748016978947821e-05,
"loss": 5.3968,
"step": 83000
},
{
"epoch": 3.5197909202040214,
"grad_norm": 0.4345729649066925,
"learning_rate": 8.73944175277623e-05,
"loss": 5.3957,
"step": 83500
},
{
"epoch": 3.540867512540572,
"grad_norm": 0.42922207713127136,
"learning_rate": 8.730866526604639e-05,
"loss": 5.3991,
"step": 84000
},
{
"epoch": 3.5619441048771234,
"grad_norm": 0.44002217054367065,
"learning_rate": 8.72229130043305e-05,
"loss": 5.3952,
"step": 84500
},
{
"epoch": 3.5830206972136747,
"grad_norm": 0.4443035125732422,
"learning_rate": 8.713716074261459e-05,
"loss": 5.3988,
"step": 85000
},
{
"epoch": 3.6040972895502255,
"grad_norm": 0.4227435290813446,
"learning_rate": 8.705140848089869e-05,
"loss": 5.3931,
"step": 85500
},
{
"epoch": 3.6251738818867767,
"grad_norm": 0.43842363357543945,
"learning_rate": 8.696565621918278e-05,
"loss": 5.3913,
"step": 86000
},
{
"epoch": 3.6462504742233275,
"grad_norm": 0.4260132312774658,
"learning_rate": 8.687990395746688e-05,
"loss": 5.395,
"step": 86500
},
{
"epoch": 3.6673270665598787,
"grad_norm": 0.43637609481811523,
"learning_rate": 8.679415169575098e-05,
"loss": 5.3916,
"step": 87000
},
{
"epoch": 3.6884036588964295,
"grad_norm": 0.44304344058036804,
"learning_rate": 8.670839943403508e-05,
"loss": 5.3945,
"step": 87500
},
{
"epoch": 3.7094802512329808,
"grad_norm": 0.41229087114334106,
"learning_rate": 8.662264717231916e-05,
"loss": 5.3926,
"step": 88000
},
{
"epoch": 3.7305568435695315,
"grad_norm": 0.4697146415710449,
"learning_rate": 8.653689491060328e-05,
"loss": 5.3938,
"step": 88500
},
{
"epoch": 3.751633435906083,
"grad_norm": 0.4178410768508911,
"learning_rate": 8.645114264888737e-05,
"loss": 5.3964,
"step": 89000
},
{
"epoch": 3.7727100282426336,
"grad_norm": 0.4398083984851837,
"learning_rate": 8.636539038717146e-05,
"loss": 5.3892,
"step": 89500
},
{
"epoch": 3.793786620579185,
"grad_norm": 0.45152032375335693,
"learning_rate": 8.627963812545556e-05,
"loss": 5.3864,
"step": 90000
},
{
"epoch": 3.793786620579185,
"eval_accuracy": 0.16719885564497222,
"eval_loss": 5.2806315422058105,
"eval_runtime": 197.588,
"eval_samples_per_second": 610.897,
"eval_steps_per_second": 6.367,
"step": 90000
},
{
"epoch": 3.8148632129157356,
"grad_norm": 0.41108986735343933,
"learning_rate": 8.619388586373967e-05,
"loss": 5.3907,
"step": 90500
},
{
"epoch": 3.835939805252287,
"grad_norm": 0.42306339740753174,
"learning_rate": 8.610813360202375e-05,
"loss": 5.3875,
"step": 91000
},
{
"epoch": 3.857016397588838,
"grad_norm": 0.44638413190841675,
"learning_rate": 8.602238134030785e-05,
"loss": 5.3881,
"step": 91500
},
{
"epoch": 3.878092989925389,
"grad_norm": 0.4359639286994934,
"learning_rate": 8.593662907859195e-05,
"loss": 5.3865,
"step": 92000
},
{
"epoch": 3.8991695822619397,
"grad_norm": 0.41788849234580994,
"learning_rate": 8.585087681687605e-05,
"loss": 5.3894,
"step": 92500
},
{
"epoch": 3.920246174598491,
"grad_norm": 0.4148942828178406,
"learning_rate": 8.576512455516015e-05,
"loss": 5.3913,
"step": 93000
},
{
"epoch": 3.941322766935042,
"grad_norm": 0.4180818200111389,
"learning_rate": 8.567937229344425e-05,
"loss": 5.3857,
"step": 93500
},
{
"epoch": 3.962399359271593,
"grad_norm": 0.416759192943573,
"learning_rate": 8.559362003172834e-05,
"loss": 5.3895,
"step": 94000
},
{
"epoch": 3.983475951608144,
"grad_norm": 0.434945285320282,
"learning_rate": 8.550786777001244e-05,
"loss": 5.3864,
"step": 94500
},
{
"epoch": 4.004552543944695,
"grad_norm": 0.47486943006515503,
"learning_rate": 8.542211550829654e-05,
"loss": 5.3838,
"step": 95000
},
{
"epoch": 4.025629136281246,
"grad_norm": 0.42085471749305725,
"learning_rate": 8.533636324658063e-05,
"loss": 5.3776,
"step": 95500
},
{
"epoch": 4.046705728617797,
"grad_norm": 0.4197317957878113,
"learning_rate": 8.525061098486474e-05,
"loss": 5.3848,
"step": 96000
},
{
"epoch": 4.067782320954348,
"grad_norm": 0.42979541420936584,
"learning_rate": 8.516485872314884e-05,
"loss": 5.3765,
"step": 96500
},
{
"epoch": 4.088858913290899,
"grad_norm": 0.41765832901000977,
"learning_rate": 8.507910646143292e-05,
"loss": 5.3788,
"step": 97000
},
{
"epoch": 4.10993550562745,
"grad_norm": 0.4354574978351593,
"learning_rate": 8.499335419971702e-05,
"loss": 5.3779,
"step": 97500
},
{
"epoch": 4.1310120979640015,
"grad_norm": 0.41537976264953613,
"learning_rate": 8.490760193800112e-05,
"loss": 5.3765,
"step": 98000
},
{
"epoch": 4.152088690300552,
"grad_norm": 0.41980552673339844,
"learning_rate": 8.482184967628522e-05,
"loss": 5.3774,
"step": 98500
},
{
"epoch": 4.173165282637103,
"grad_norm": 0.4153006374835968,
"learning_rate": 8.473609741456931e-05,
"loss": 5.3808,
"step": 99000
},
{
"epoch": 4.194241874973654,
"grad_norm": 0.44889509677886963,
"learning_rate": 8.46503451528534e-05,
"loss": 5.3814,
"step": 99500
},
{
"epoch": 4.2153184673102055,
"grad_norm": 0.4466044008731842,
"learning_rate": 8.456459289113751e-05,
"loss": 5.3883,
"step": 100000
},
{
"epoch": 4.2153184673102055,
"eval_accuracy": 0.16407263627369081,
"eval_loss": 5.2893853187561035,
"eval_runtime": 200.7171,
"eval_samples_per_second": 601.374,
"eval_steps_per_second": 6.268,
"step": 100000
},
{
"epoch": 4.236395059646756,
"grad_norm": 0.44025611877441406,
"learning_rate": 8.447884062942161e-05,
"loss": 5.3935,
"step": 100500
},
{
"epoch": 4.257471651983307,
"grad_norm": 0.4311734735965729,
"learning_rate": 8.43930883677057e-05,
"loss": 5.3953,
"step": 101000
},
{
"epoch": 4.278548244319858,
"grad_norm": 0.43135353922843933,
"learning_rate": 8.430733610598979e-05,
"loss": 5.379,
"step": 101500
},
{
"epoch": 4.29962483665641,
"grad_norm": 0.4436395466327667,
"learning_rate": 8.42215838442739e-05,
"loss": 5.379,
"step": 102000
},
{
"epoch": 4.32070142899296,
"grad_norm": 0.43717285990715027,
"learning_rate": 8.413583158255799e-05,
"loss": 5.3802,
"step": 102500
},
{
"epoch": 4.341778021329511,
"grad_norm": 0.4251772463321686,
"learning_rate": 8.405007932084209e-05,
"loss": 5.3759,
"step": 103000
},
{
"epoch": 4.362854613666062,
"grad_norm": 0.4443017840385437,
"learning_rate": 8.396432705912619e-05,
"loss": 5.3801,
"step": 103500
},
{
"epoch": 4.383931206002614,
"grad_norm": 0.43594592809677124,
"learning_rate": 8.387857479741028e-05,
"loss": 5.3789,
"step": 104000
},
{
"epoch": 4.405007798339165,
"grad_norm": 0.421587198972702,
"learning_rate": 8.379282253569438e-05,
"loss": 5.3759,
"step": 104500
},
{
"epoch": 4.426084390675715,
"grad_norm": 0.4281757175922394,
"learning_rate": 8.370707027397848e-05,
"loss": 5.3755,
"step": 105000
},
{
"epoch": 4.4471609830122665,
"grad_norm": 0.42158573865890503,
"learning_rate": 8.362131801226258e-05,
"loss": 5.3756,
"step": 105500
},
{
"epoch": 4.468237575348818,
"grad_norm": 0.43424269556999207,
"learning_rate": 8.353556575054668e-05,
"loss": 5.3748,
"step": 106000
},
{
"epoch": 4.489314167685369,
"grad_norm": 0.41975584626197815,
"learning_rate": 8.344981348883077e-05,
"loss": 5.378,
"step": 106500
},
{
"epoch": 4.510390760021919,
"grad_norm": 0.4232282340526581,
"learning_rate": 8.336406122711486e-05,
"loss": 5.3727,
"step": 107000
},
{
"epoch": 4.5314673523584705,
"grad_norm": 0.4501512348651886,
"learning_rate": 8.327830896539897e-05,
"loss": 5.3769,
"step": 107500
},
{
"epoch": 4.552543944695022,
"grad_norm": 0.430493026971817,
"learning_rate": 8.319255670368307e-05,
"loss": 5.3727,
"step": 108000
},
{
"epoch": 4.573620537031573,
"grad_norm": 0.42426082491874695,
"learning_rate": 8.310680444196715e-05,
"loss": 5.3739,
"step": 108500
},
{
"epoch": 4.594697129368123,
"grad_norm": 0.4224434792995453,
"learning_rate": 8.302105218025125e-05,
"loss": 5.3708,
"step": 109000
},
{
"epoch": 4.615773721704675,
"grad_norm": 0.4398936927318573,
"learning_rate": 8.293529991853536e-05,
"loss": 5.3738,
"step": 109500
},
{
"epoch": 4.636850314041226,
"grad_norm": 0.4452034831047058,
"learning_rate": 8.284954765681945e-05,
"loss": 5.3743,
"step": 110000
},
{
"epoch": 4.636850314041226,
"eval_accuracy": 0.16777100673964437,
"eval_loss": 5.2662177085876465,
"eval_runtime": 203.6452,
"eval_samples_per_second": 592.727,
"eval_steps_per_second": 6.177,
"step": 110000
},
{
"epoch": 4.657926906377777,
"grad_norm": 0.4356929361820221,
"learning_rate": 8.276379539510355e-05,
"loss": 5.3733,
"step": 110500
},
{
"epoch": 4.679003498714328,
"grad_norm": 0.4274049997329712,
"learning_rate": 8.267804313338765e-05,
"loss": 5.3713,
"step": 111000
},
{
"epoch": 4.700080091050879,
"grad_norm": 0.43538618087768555,
"learning_rate": 8.259229087167174e-05,
"loss": 5.3744,
"step": 111500
},
{
"epoch": 4.72115668338743,
"grad_norm": 0.4315907955169678,
"learning_rate": 8.250653860995584e-05,
"loss": 5.3707,
"step": 112000
},
{
"epoch": 4.742233275723981,
"grad_norm": 0.418369323015213,
"learning_rate": 8.242078634823994e-05,
"loss": 5.3696,
"step": 112500
},
{
"epoch": 4.763309868060532,
"grad_norm": 0.4267233610153198,
"learning_rate": 8.233503408652403e-05,
"loss": 5.3707,
"step": 113000
},
{
"epoch": 4.784386460397083,
"grad_norm": 0.42834270000457764,
"learning_rate": 8.224928182480814e-05,
"loss": 5.3703,
"step": 113500
},
{
"epoch": 4.805463052733634,
"grad_norm": 0.4220888018608093,
"learning_rate": 8.216352956309224e-05,
"loss": 5.3643,
"step": 114000
},
{
"epoch": 4.826539645070185,
"grad_norm": 0.4265943467617035,
"learning_rate": 8.207777730137632e-05,
"loss": 5.3678,
"step": 114500
},
{
"epoch": 4.847616237406736,
"grad_norm": 0.4487956166267395,
"learning_rate": 8.199202503966042e-05,
"loss": 5.3664,
"step": 115000
},
{
"epoch": 4.868692829743287,
"grad_norm": 0.434289813041687,
"learning_rate": 8.190627277794453e-05,
"loss": 5.3674,
"step": 115500
},
{
"epoch": 4.889769422079838,
"grad_norm": 0.42209193110466003,
"learning_rate": 8.182052051622862e-05,
"loss": 5.3671,
"step": 116000
},
{
"epoch": 4.910846014416389,
"grad_norm": 0.435409277677536,
"learning_rate": 8.173476825451271e-05,
"loss": 5.3686,
"step": 116500
},
{
"epoch": 4.9319226067529405,
"grad_norm": 0.4433590769767761,
"learning_rate": 8.164901599279681e-05,
"loss": 5.3698,
"step": 117000
},
{
"epoch": 4.952999199089491,
"grad_norm": 0.42689043283462524,
"learning_rate": 8.156326373108091e-05,
"loss": 5.3678,
"step": 117500
},
{
"epoch": 4.974075791426042,
"grad_norm": 0.4452136158943176,
"learning_rate": 8.147751146936501e-05,
"loss": 5.3781,
"step": 118000
},
{
"epoch": 4.995152383762593,
"grad_norm": 0.4457673132419586,
"learning_rate": 8.139175920764911e-05,
"loss": 5.365,
"step": 118500
},
{
"epoch": 5.0162289760991445,
"grad_norm": 0.46507585048675537,
"learning_rate": 8.13060069459332e-05,
"loss": 5.3636,
"step": 119000
},
{
"epoch": 5.037305568435695,
"grad_norm": 0.439755380153656,
"learning_rate": 8.12202546842173e-05,
"loss": 5.3642,
"step": 119500
},
{
"epoch": 5.058382160772246,
"grad_norm": 0.44612136483192444,
"learning_rate": 8.11345024225014e-05,
"loss": 5.3614,
"step": 120000
},
{
"epoch": 5.058382160772246,
"eval_accuracy": 0.1676593370022191,
"eval_loss": 5.249516010284424,
"eval_runtime": 198.3508,
"eval_samples_per_second": 608.548,
"eval_steps_per_second": 6.342,
"step": 120000
},
{
"epoch": 5.079458753108797,
"grad_norm": 0.44694289565086365,
"learning_rate": 8.104875016078549e-05,
"loss": 5.3533,
"step": 120500
},
{
"epoch": 5.100535345445349,
"grad_norm": 0.46488720178604126,
"learning_rate": 8.09629978990696e-05,
"loss": 5.3548,
"step": 121000
},
{
"epoch": 5.1216119377819,
"grad_norm": 0.4434725344181061,
"learning_rate": 8.08772456373537e-05,
"loss": 5.3516,
"step": 121500
},
{
"epoch": 5.14268853011845,
"grad_norm": 0.494030237197876,
"learning_rate": 8.079149337563778e-05,
"loss": 5.3534,
"step": 122000
},
{
"epoch": 5.163765122455001,
"grad_norm": 0.4555765986442566,
"learning_rate": 8.070574111392188e-05,
"loss": 5.3509,
"step": 122500
},
{
"epoch": 5.184841714791553,
"grad_norm": 0.45734184980392456,
"learning_rate": 8.061998885220599e-05,
"loss": 5.3458,
"step": 123000
},
{
"epoch": 5.205918307128104,
"grad_norm": 0.4740845561027527,
"learning_rate": 8.053423659049008e-05,
"loss": 5.3439,
"step": 123500
},
{
"epoch": 5.226994899464654,
"grad_norm": 0.48670533299446106,
"learning_rate": 8.044848432877418e-05,
"loss": 5.3388,
"step": 124000
},
{
"epoch": 5.2480714918012055,
"grad_norm": 0.5606753826141357,
"learning_rate": 8.036273206705827e-05,
"loss": 5.3375,
"step": 124500
},
{
"epoch": 5.269148084137757,
"grad_norm": 0.5068137645721436,
"learning_rate": 8.027697980534237e-05,
"loss": 5.3219,
"step": 125000
},
{
"epoch": 5.290224676474308,
"grad_norm": 0.5216848254203796,
"learning_rate": 8.019122754362647e-05,
"loss": 5.3124,
"step": 125500
},
{
"epoch": 5.311301268810858,
"grad_norm": 0.5404259562492371,
"learning_rate": 8.010547528191057e-05,
"loss": 5.2944,
"step": 126000
},
{
"epoch": 5.3323778611474095,
"grad_norm": 0.6292199492454529,
"learning_rate": 8.001972302019465e-05,
"loss": 5.2686,
"step": 126500
},
{
"epoch": 5.353454453483961,
"grad_norm": 1.331845760345459,
"learning_rate": 7.993397075847876e-05,
"loss": 4.8172,
"step": 127000
},
{
"epoch": 5.374531045820512,
"grad_norm": 1.1069200038909912,
"learning_rate": 7.984821849676286e-05,
"loss": 3.6635,
"step": 127500
},
{
"epoch": 5.395607638157063,
"grad_norm": 13.520956993103027,
"learning_rate": 7.976246623504695e-05,
"loss": 3.2959,
"step": 128000
},
{
"epoch": 5.416684230493614,
"grad_norm": 1.1996490955352783,
"learning_rate": 7.967671397333105e-05,
"loss": 3.1061,
"step": 128500
},
{
"epoch": 5.437760822830165,
"grad_norm": 1.0518244504928589,
"learning_rate": 7.959096171161516e-05,
"loss": 2.9841,
"step": 129000
},
{
"epoch": 5.458837415166716,
"grad_norm": 1.0427472591400146,
"learning_rate": 7.950520944989924e-05,
"loss": 2.8732,
"step": 129500
},
{
"epoch": 5.479914007503266,
"grad_norm": 0.9907004833221436,
"learning_rate": 7.941945718818334e-05,
"loss": 2.7786,
"step": 130000
},
{
"epoch": 5.479914007503266,
"eval_accuracy": 0.5314471897112323,
"eval_loss": 2.4131810665130615,
"eval_runtime": 202.8803,
"eval_samples_per_second": 594.962,
"eval_steps_per_second": 6.201,
"step": 130000
},
{
"epoch": 5.500990599839818,
"grad_norm": 0.9749962687492371,
"learning_rate": 7.933370492646744e-05,
"loss": 2.7133,
"step": 130500
},
{
"epoch": 5.522067192176369,
"grad_norm": 0.9573583006858826,
"learning_rate": 7.924795266475154e-05,
"loss": 2.6445,
"step": 131000
},
{
"epoch": 5.54314378451292,
"grad_norm": 0.9834604859352112,
"learning_rate": 7.916220040303564e-05,
"loss": 2.59,
"step": 131500
},
{
"epoch": 5.564220376849471,
"grad_norm": 0.9948740005493164,
"learning_rate": 7.907644814131972e-05,
"loss": 2.5516,
"step": 132000
},
{
"epoch": 5.585296969186022,
"grad_norm": 0.9803892970085144,
"learning_rate": 7.899069587960383e-05,
"loss": 2.5092,
"step": 132500
},
{
"epoch": 5.606373561522573,
"grad_norm": 2.3991427421569824,
"learning_rate": 7.890494361788793e-05,
"loss": 2.476,
"step": 133000
},
{
"epoch": 5.627450153859124,
"grad_norm": 0.9743901491165161,
"learning_rate": 7.881919135617202e-05,
"loss": 2.44,
"step": 133500
},
{
"epoch": 5.648526746195675,
"grad_norm": 0.9845961928367615,
"learning_rate": 7.873343909445611e-05,
"loss": 2.4079,
"step": 134000
},
{
"epoch": 5.669603338532226,
"grad_norm": 0.9659436941146851,
"learning_rate": 7.864768683274023e-05,
"loss": 2.3931,
"step": 134500
},
{
"epoch": 5.690679930868777,
"grad_norm": 0.978263795375824,
"learning_rate": 7.856193457102431e-05,
"loss": 2.3675,
"step": 135000
},
{
"epoch": 5.711756523205328,
"grad_norm": 0.9540228843688965,
"learning_rate": 7.847618230930841e-05,
"loss": 2.3427,
"step": 135500
},
{
"epoch": 5.7328331155418795,
"grad_norm": 0.9464851021766663,
"learning_rate": 7.839043004759251e-05,
"loss": 2.3207,
"step": 136000
},
{
"epoch": 5.75390970787843,
"grad_norm": 0.9881648421287537,
"learning_rate": 7.83046777858766e-05,
"loss": 2.2985,
"step": 136500
},
{
"epoch": 5.774986300214981,
"grad_norm": 0.9680384397506714,
"learning_rate": 7.82189255241607e-05,
"loss": 2.2804,
"step": 137000
},
{
"epoch": 5.796062892551532,
"grad_norm": 1.3068596124649048,
"learning_rate": 7.81331732624448e-05,
"loss": 2.2613,
"step": 137500
},
{
"epoch": 5.8171394848880835,
"grad_norm": 0.9852378368377686,
"learning_rate": 7.804742100072889e-05,
"loss": 2.251,
"step": 138000
},
{
"epoch": 5.838216077224635,
"grad_norm": 0.9898777604103088,
"learning_rate": 7.7961668739013e-05,
"loss": 2.2406,
"step": 138500
},
{
"epoch": 5.859292669561185,
"grad_norm": 1.076651930809021,
"learning_rate": 7.78759164772971e-05,
"loss": 2.2156,
"step": 139000
},
{
"epoch": 5.880369261897736,
"grad_norm": 0.9574313163757324,
"learning_rate": 7.779016421558118e-05,
"loss": 2.2002,
"step": 139500
},
{
"epoch": 5.901445854234288,
"grad_norm": 0.9755016565322876,
"learning_rate": 7.770441195386528e-05,
"loss": 2.191,
"step": 140000
},
{
"epoch": 5.901445854234288,
"eval_accuracy": 0.6134978338329169,
"eval_loss": 1.8930723667144775,
"eval_runtime": 199.6414,
"eval_samples_per_second": 604.614,
"eval_steps_per_second": 6.301,
"step": 140000
},
{
"epoch": 5.922522446570838,
"grad_norm": 0.9787560105323792,
"learning_rate": 7.761865969214939e-05,
"loss": 2.173,
"step": 140500
},
{
"epoch": 5.943599038907389,
"grad_norm": 0.9598313570022583,
"learning_rate": 7.753290743043348e-05,
"loss": 2.169,
"step": 141000
},
{
"epoch": 5.96467563124394,
"grad_norm": 0.9919191002845764,
"learning_rate": 7.744715516871758e-05,
"loss": 2.1604,
"step": 141500
},
{
"epoch": 5.985752223580492,
"grad_norm": 0.9667888283729553,
"learning_rate": 7.736140290700167e-05,
"loss": 2.1436,
"step": 142000
},
{
"epoch": 6.006828815917043,
"grad_norm": 1.0379912853240967,
"learning_rate": 7.727565064528577e-05,
"loss": 2.1305,
"step": 142500
},
{
"epoch": 6.027905408253593,
"grad_norm": 1.0193846225738525,
"learning_rate": 7.718989838356987e-05,
"loss": 2.1202,
"step": 143000
},
{
"epoch": 6.0489820005901445,
"grad_norm": 0.9990450143814087,
"learning_rate": 7.710414612185397e-05,
"loss": 2.113,
"step": 143500
},
{
"epoch": 6.070058592926696,
"grad_norm": 0.965611457824707,
"learning_rate": 7.701839386013807e-05,
"loss": 2.0983,
"step": 144000
},
{
"epoch": 6.091135185263247,
"grad_norm": 0.9573085904121399,
"learning_rate": 7.693264159842217e-05,
"loss": 2.0837,
"step": 144500
},
{
"epoch": 6.112211777599797,
"grad_norm": 0.9624812602996826,
"learning_rate": 7.684688933670626e-05,
"loss": 2.0747,
"step": 145000
},
{
"epoch": 6.1332883699363485,
"grad_norm": 0.951323926448822,
"learning_rate": 7.676113707499035e-05,
"loss": 2.066,
"step": 145500
},
{
"epoch": 6.1543649622729,
"grad_norm": 0.9279555678367615,
"learning_rate": 7.667538481327446e-05,
"loss": 2.0563,
"step": 146000
},
{
"epoch": 6.175441554609451,
"grad_norm": 0.9143189787864685,
"learning_rate": 7.658963255155856e-05,
"loss": 2.0499,
"step": 146500
},
{
"epoch": 6.196518146946001,
"grad_norm": 0.9248863458633423,
"learning_rate": 7.650388028984264e-05,
"loss": 2.0393,
"step": 147000
},
{
"epoch": 6.217594739282553,
"grad_norm": 0.9433860778808594,
"learning_rate": 7.641812802812674e-05,
"loss": 2.0303,
"step": 147500
},
{
"epoch": 6.238671331619104,
"grad_norm": 0.9305623173713684,
"learning_rate": 7.633237576641084e-05,
"loss": 2.0206,
"step": 148000
},
{
"epoch": 6.259747923955655,
"grad_norm": 0.9531417489051819,
"learning_rate": 7.624662350469494e-05,
"loss": 2.017,
"step": 148500
},
{
"epoch": 6.280824516292206,
"grad_norm": 0.9727780818939209,
"learning_rate": 7.616087124297904e-05,
"loss": 2.012,
"step": 149000
},
{
"epoch": 6.301901108628757,
"grad_norm": 0.960665762424469,
"learning_rate": 7.607511898126313e-05,
"loss": 2.0035,
"step": 149500
},
{
"epoch": 6.322977700965308,
"grad_norm": 0.9659542441368103,
"learning_rate": 7.598936671954723e-05,
"loss": 1.997,
"step": 150000
},
{
"epoch": 6.322977700965308,
"eval_accuracy": 0.6414069347552657,
"eval_loss": 1.7233749628067017,
"eval_runtime": 198.6134,
"eval_samples_per_second": 607.743,
"eval_steps_per_second": 6.334,
"step": 150000
},
{
"epoch": 6.344054293301859,
"grad_norm": 0.9654056429862976,
"learning_rate": 7.590361445783133e-05,
"loss": 1.9928,
"step": 150500
},
{
"epoch": 6.36513088563841,
"grad_norm": 0.9228722453117371,
"learning_rate": 7.581786219611543e-05,
"loss": 1.9844,
"step": 151000
},
{
"epoch": 6.386207477974961,
"grad_norm": 0.9529992341995239,
"learning_rate": 7.573210993439951e-05,
"loss": 1.979,
"step": 151500
},
{
"epoch": 6.407284070311512,
"grad_norm": 0.9145802855491638,
"learning_rate": 7.564635767268363e-05,
"loss": 1.9694,
"step": 152000
},
{
"epoch": 6.428360662648063,
"grad_norm": 0.968990683555603,
"learning_rate": 7.556060541096772e-05,
"loss": 1.961,
"step": 152500
},
{
"epoch": 6.449437254984614,
"grad_norm": 0.9550362229347229,
"learning_rate": 7.547485314925181e-05,
"loss": 1.9598,
"step": 153000
},
{
"epoch": 6.470513847321165,
"grad_norm": 0.9265795350074768,
"learning_rate": 7.538910088753591e-05,
"loss": 1.9526,
"step": 153500
},
{
"epoch": 6.491590439657716,
"grad_norm": 0.9825773239135742,
"learning_rate": 7.530334862582002e-05,
"loss": 1.9475,
"step": 154000
},
{
"epoch": 6.512667031994267,
"grad_norm": 0.9446262121200562,
"learning_rate": 7.52175963641041e-05,
"loss": 1.9448,
"step": 154500
},
{
"epoch": 6.5337436243308185,
"grad_norm": 0.9423317909240723,
"learning_rate": 7.51318441023882e-05,
"loss": 1.9408,
"step": 155000
},
{
"epoch": 6.55482021666737,
"grad_norm": 1.9699567556381226,
"learning_rate": 7.50460918406723e-05,
"loss": 1.9321,
"step": 155500
},
{
"epoch": 6.57589680900392,
"grad_norm": 0.9240116477012634,
"learning_rate": 7.49603395789564e-05,
"loss": 1.9277,
"step": 156000
},
{
"epoch": 6.596973401340471,
"grad_norm": 0.9303151369094849,
"learning_rate": 7.48745873172405e-05,
"loss": 1.9199,
"step": 156500
},
{
"epoch": 6.6180499936770225,
"grad_norm": 0.9519091248512268,
"learning_rate": 7.47888350555246e-05,
"loss": 1.9116,
"step": 157000
},
{
"epoch": 6.639126586013573,
"grad_norm": 0.9786426424980164,
"learning_rate": 7.47030827938087e-05,
"loss": 1.9107,
"step": 157500
},
{
"epoch": 6.660203178350124,
"grad_norm": 0.928004264831543,
"learning_rate": 7.461733053209279e-05,
"loss": 1.9052,
"step": 158000
},
{
"epoch": 6.681279770686675,
"grad_norm": 0.9402891397476196,
"learning_rate": 7.453157827037689e-05,
"loss": 1.9045,
"step": 158500
},
{
"epoch": 6.702356363023227,
"grad_norm": 0.931020975112915,
"learning_rate": 7.444582600866098e-05,
"loss": 1.8925,
"step": 159000
},
{
"epoch": 6.723432955359778,
"grad_norm": 0.9521540999412537,
"learning_rate": 7.436007374694507e-05,
"loss": 1.892,
"step": 159500
},
{
"epoch": 6.744509547696328,
"grad_norm": 0.9463508129119873,
"learning_rate": 7.427432148522919e-05,
"loss": 1.8894,
"step": 160000
},
{
"epoch": 6.744509547696328,
"eval_accuracy": 0.6582351844575826,
"eval_loss": 1.620818018913269,
"eval_runtime": 198.2666,
"eval_samples_per_second": 608.806,
"eval_steps_per_second": 6.345,
"step": 160000
},
{
"epoch": 6.765586140032879,
"grad_norm": 0.9433901309967041,
"learning_rate": 7.418856922351327e-05,
"loss": 1.8817,
"step": 160500
},
{
"epoch": 6.786662732369431,
"grad_norm": 0.9463580250740051,
"learning_rate": 7.410281696179737e-05,
"loss": 1.8732,
"step": 161000
},
{
"epoch": 6.807739324705982,
"grad_norm": 0.9400925040245056,
"learning_rate": 7.401706470008147e-05,
"loss": 1.8686,
"step": 161500
},
{
"epoch": 6.828815917042532,
"grad_norm": 0.9394701719284058,
"learning_rate": 7.393131243836557e-05,
"loss": 1.8652,
"step": 162000
},
{
"epoch": 6.8498925093790834,
"grad_norm": 0.9406070113182068,
"learning_rate": 7.384556017664966e-05,
"loss": 1.8602,
"step": 162500
},
{
"epoch": 6.870969101715635,
"grad_norm": 0.9383716583251953,
"learning_rate": 7.375980791493376e-05,
"loss": 1.8554,
"step": 163000
},
{
"epoch": 6.892045694052186,
"grad_norm": 1.2340610027313232,
"learning_rate": 7.367405565321786e-05,
"loss": 1.8524,
"step": 163500
},
{
"epoch": 6.913122286388736,
"grad_norm": 1.5135760307312012,
"learning_rate": 7.358830339150196e-05,
"loss": 1.8508,
"step": 164000
},
{
"epoch": 6.9341988787252875,
"grad_norm": 1.0064946413040161,
"learning_rate": 7.350255112978606e-05,
"loss": 1.8526,
"step": 164500
},
{
"epoch": 6.955275471061839,
"grad_norm": 3.200838804244995,
"learning_rate": 7.341679886807014e-05,
"loss": 1.8451,
"step": 165000
},
{
"epoch": 6.97635206339839,
"grad_norm": 10.596697807312012,
"learning_rate": 7.333104660635425e-05,
"loss": 1.8398,
"step": 165500
},
{
"epoch": 6.997428655734941,
"grad_norm": 0.920864462852478,
"learning_rate": 7.324529434463835e-05,
"loss": 1.8356,
"step": 166000
},
{
"epoch": 7.018505248071492,
"grad_norm": 0.9243833422660828,
"learning_rate": 7.315954208292244e-05,
"loss": 1.8332,
"step": 166500
},
{
"epoch": 7.039581840408043,
"grad_norm": 0.9314616322517395,
"learning_rate": 7.307378982120653e-05,
"loss": 1.8239,
"step": 167000
},
{
"epoch": 7.060658432744594,
"grad_norm": 0.9733430743217468,
"learning_rate": 7.298803755949063e-05,
"loss": 1.8223,
"step": 167500
},
{
"epoch": 7.081735025081145,
"grad_norm": 1.5428160429000854,
"learning_rate": 7.290228529777473e-05,
"loss": 1.8183,
"step": 168000
},
{
"epoch": 7.102811617417696,
"grad_norm": 0.9416449069976807,
"learning_rate": 7.281653303605883e-05,
"loss": 1.8163,
"step": 168500
},
{
"epoch": 7.123888209754247,
"grad_norm": 0.9595532417297363,
"learning_rate": 7.273078077434293e-05,
"loss": 1.8181,
"step": 169000
},
{
"epoch": 7.144964802090798,
"grad_norm": 0.9339419603347778,
"learning_rate": 7.264502851262703e-05,
"loss": 1.8071,
"step": 169500
},
{
"epoch": 7.166041394427349,
"grad_norm": 0.9371300935745239,
"learning_rate": 7.255927625091112e-05,
"loss": 1.801,
"step": 170000
},
{
"epoch": 7.166041394427349,
"eval_accuracy": 0.6709391446696715,
"eval_loss": 1.546624779701233,
"eval_runtime": 197.1157,
"eval_samples_per_second": 612.361,
"eval_steps_per_second": 6.382,
"step": 170000
},
{
"epoch": 7.1871179867639,
"grad_norm": 0.9237951636314392,
"learning_rate": 7.247352398919521e-05,
"loss": 1.8024,
"step": 170500
},
{
"epoch": 7.208194579100451,
"grad_norm": 0.9215587973594666,
"learning_rate": 7.238777172747931e-05,
"loss": 1.7978,
"step": 171000
},
{
"epoch": 7.229271171437002,
"grad_norm": 0.9249234795570374,
"learning_rate": 7.230201946576342e-05,
"loss": 1.7971,
"step": 171500
},
{
"epoch": 7.250347763773553,
"grad_norm": 2.2585291862487793,
"learning_rate": 7.22162672040475e-05,
"loss": 1.7923,
"step": 172000
},
{
"epoch": 7.271424356110104,
"grad_norm": 0.9211357235908508,
"learning_rate": 7.21305149423316e-05,
"loss": 1.7869,
"step": 172500
},
{
"epoch": 7.292500948446655,
"grad_norm": 0.9230213165283203,
"learning_rate": 7.20447626806157e-05,
"loss": 1.7805,
"step": 173000
},
{
"epoch": 7.313577540783206,
"grad_norm": 0.9403547644615173,
"learning_rate": 7.19590104188998e-05,
"loss": 1.779,
"step": 173500
},
{
"epoch": 7.3346541331197574,
"grad_norm": 0.9296572804450989,
"learning_rate": 7.18732581571839e-05,
"loss": 1.7775,
"step": 174000
},
{
"epoch": 7.355730725456308,
"grad_norm": 0.9451926350593567,
"learning_rate": 7.1787505895468e-05,
"loss": 1.7745,
"step": 174500
},
{
"epoch": 7.376807317792859,
"grad_norm": 0.9225753545761108,
"learning_rate": 7.17017536337521e-05,
"loss": 1.7712,
"step": 175000
},
{
"epoch": 7.39788391012941,
"grad_norm": 0.9093494415283203,
"learning_rate": 7.161600137203619e-05,
"loss": 1.7712,
"step": 175500
},
{
"epoch": 7.4189605024659615,
"grad_norm": 0.9403925538063049,
"learning_rate": 7.153024911032029e-05,
"loss": 1.7628,
"step": 176000
},
{
"epoch": 7.440037094802513,
"grad_norm": 0.9464032649993896,
"learning_rate": 7.144449684860438e-05,
"loss": 1.764,
"step": 176500
},
{
"epoch": 7.461113687139063,
"grad_norm": 0.9203761219978333,
"learning_rate": 7.135874458688849e-05,
"loss": 1.7641,
"step": 177000
},
{
"epoch": 7.482190279475614,
"grad_norm": 0.9569200873374939,
"learning_rate": 7.127299232517259e-05,
"loss": 1.7608,
"step": 177500
},
{
"epoch": 7.503266871812166,
"grad_norm": 0.9131455421447754,
"learning_rate": 7.118724006345667e-05,
"loss": 1.7564,
"step": 178000
},
{
"epoch": 7.524343464148717,
"grad_norm": 0.9599666595458984,
"learning_rate": 7.110148780174077e-05,
"loss": 1.7499,
"step": 178500
},
{
"epoch": 7.545420056485267,
"grad_norm": 0.9368263483047485,
"learning_rate": 7.101573554002488e-05,
"loss": 1.7489,
"step": 179000
},
{
"epoch": 7.566496648821818,
"grad_norm": 0.957270085811615,
"learning_rate": 7.092998327830897e-05,
"loss": 1.7488,
"step": 179500
},
{
"epoch": 7.58757324115837,
"grad_norm": 0.9329002499580383,
"learning_rate": 7.084423101659306e-05,
"loss": 1.7429,
"step": 180000
},
{
"epoch": 7.58757324115837,
"eval_accuracy": 0.6795182322111257,
"eval_loss": 1.495853066444397,
"eval_runtime": 197.0462,
"eval_samples_per_second": 612.577,
"eval_steps_per_second": 6.384,
"step": 180000
},
{
"epoch": 7.608649833494921,
"grad_norm": 0.9357250332832336,
"learning_rate": 7.075847875487716e-05,
"loss": 1.7428,
"step": 180500
},
{
"epoch": 7.629726425831471,
"grad_norm": 0.9291309118270874,
"learning_rate": 7.067272649316126e-05,
"loss": 1.7449,
"step": 181000
},
{
"epoch": 7.650803018168022,
"grad_norm": 0.9331381320953369,
"learning_rate": 7.058697423144536e-05,
"loss": 1.7346,
"step": 181500
},
{
"epoch": 7.671879610504574,
"grad_norm": 0.9019379615783691,
"learning_rate": 7.050122196972946e-05,
"loss": 1.7322,
"step": 182000
},
{
"epoch": 7.692956202841125,
"grad_norm": 0.9203329682350159,
"learning_rate": 7.041546970801354e-05,
"loss": 1.7336,
"step": 182500
},
{
"epoch": 7.714032795177676,
"grad_norm": 0.9358742833137512,
"learning_rate": 7.032971744629765e-05,
"loss": 1.7319,
"step": 183000
},
{
"epoch": 7.7351093875142265,
"grad_norm": 0.930351197719574,
"learning_rate": 7.024396518458175e-05,
"loss": 1.732,
"step": 183500
},
{
"epoch": 7.756185979850778,
"grad_norm": 0.8871574997901917,
"learning_rate": 7.015821292286584e-05,
"loss": 1.7257,
"step": 184000
},
{
"epoch": 7.777262572187329,
"grad_norm": 0.9592623710632324,
"learning_rate": 7.007246066114993e-05,
"loss": 1.7232,
"step": 184500
},
{
"epoch": 7.798339164523879,
"grad_norm": 0.9514606595039368,
"learning_rate": 6.998670839943405e-05,
"loss": 1.7221,
"step": 185000
},
{
"epoch": 7.8194157568604306,
"grad_norm": 0.9376581311225891,
"learning_rate": 6.990095613771813e-05,
"loss": 1.7172,
"step": 185500
},
{
"epoch": 7.840492349196982,
"grad_norm": 0.9399397373199463,
"learning_rate": 6.981520387600223e-05,
"loss": 1.7151,
"step": 186000
},
{
"epoch": 7.861568941533533,
"grad_norm": 0.9271183609962463,
"learning_rate": 6.972945161428633e-05,
"loss": 1.7144,
"step": 186500
},
{
"epoch": 7.882645533870084,
"grad_norm": 0.941444456577301,
"learning_rate": 6.964369935257043e-05,
"loss": 1.7129,
"step": 187000
},
{
"epoch": 7.903722126206635,
"grad_norm": 0.9133756756782532,
"learning_rate": 6.955794709085452e-05,
"loss": 1.7114,
"step": 187500
},
{
"epoch": 7.924798718543186,
"grad_norm": 0.9308948516845703,
"learning_rate": 6.947219482913862e-05,
"loss": 1.6978,
"step": 188000
},
{
"epoch": 7.945875310879737,
"grad_norm": 0.9190275073051453,
"learning_rate": 6.938644256742272e-05,
"loss": 1.7065,
"step": 188500
},
{
"epoch": 7.966951903216288,
"grad_norm": 0.9275525212287903,
"learning_rate": 6.930069030570682e-05,
"loss": 1.7022,
"step": 189000
},
{
"epoch": 7.988028495552839,
"grad_norm": 0.9276502132415771,
"learning_rate": 6.921493804399092e-05,
"loss": 1.6975,
"step": 189500
},
{
"epoch": 8.00910508788939,
"grad_norm": 0.9298079609870911,
"learning_rate": 6.9129185782275e-05,
"loss": 1.6988,
"step": 190000
},
{
"epoch": 8.00910508788939,
"eval_accuracy": 0.6867483972435415,
"eval_loss": 1.452094316482544,
"eval_runtime": 200.074,
"eval_samples_per_second": 603.307,
"eval_steps_per_second": 6.288,
"step": 190000
},
{
"epoch": 8.03018168022594,
"grad_norm": 0.9634761810302734,
"learning_rate": 6.904343352055911e-05,
"loss": 1.6915,
"step": 190500
},
{
"epoch": 8.051258272562492,
"grad_norm": 0.93244469165802,
"learning_rate": 6.895768125884321e-05,
"loss": 1.693,
"step": 191000
},
{
"epoch": 8.072334864899043,
"grad_norm": 0.9434279799461365,
"learning_rate": 6.88719289971273e-05,
"loss": 1.6888,
"step": 191500
},
{
"epoch": 8.093411457235595,
"grad_norm": 0.9718308448791504,
"learning_rate": 6.87861767354114e-05,
"loss": 1.6832,
"step": 192000
},
{
"epoch": 8.114488049572145,
"grad_norm": 0.9499397277832031,
"learning_rate": 6.870042447369551e-05,
"loss": 1.6869,
"step": 192500
},
{
"epoch": 8.135564641908696,
"grad_norm": 0.922795832157135,
"learning_rate": 6.861467221197959e-05,
"loss": 1.6843,
"step": 193000
},
{
"epoch": 8.156641234245248,
"grad_norm": 0.9441269040107727,
"learning_rate": 6.852891995026369e-05,
"loss": 1.6837,
"step": 193500
},
{
"epoch": 8.177717826581798,
"grad_norm": 0.9146692156791687,
"learning_rate": 6.844316768854779e-05,
"loss": 1.6829,
"step": 194000
},
{
"epoch": 8.19879441891835,
"grad_norm": 0.945397675037384,
"learning_rate": 6.835741542683189e-05,
"loss": 1.6782,
"step": 194500
},
{
"epoch": 8.2198710112549,
"grad_norm": 0.9383412599563599,
"learning_rate": 6.827166316511599e-05,
"loss": 1.6737,
"step": 195000
},
{
"epoch": 8.24094760359145,
"grad_norm": 0.9231926798820496,
"learning_rate": 6.818591090340008e-05,
"loss": 1.6754,
"step": 195500
},
{
"epoch": 8.262024195928003,
"grad_norm": 0.9157910346984863,
"learning_rate": 6.810015864168417e-05,
"loss": 1.6707,
"step": 196000
},
{
"epoch": 8.283100788264553,
"grad_norm": 0.921350359916687,
"learning_rate": 6.801440637996828e-05,
"loss": 1.6687,
"step": 196500
},
{
"epoch": 8.304177380601104,
"grad_norm": 0.9354209899902344,
"learning_rate": 6.792865411825238e-05,
"loss": 1.6701,
"step": 197000
},
{
"epoch": 8.325253972937656,
"grad_norm": 0.9416276812553406,
"learning_rate": 6.784290185653646e-05,
"loss": 1.6664,
"step": 197500
},
{
"epoch": 8.346330565274206,
"grad_norm": 0.9663906097412109,
"learning_rate": 6.775714959482056e-05,
"loss": 1.6643,
"step": 198000
},
{
"epoch": 8.367407157610758,
"grad_norm": 0.9193472266197205,
"learning_rate": 6.767139733310467e-05,
"loss": 1.6638,
"step": 198500
},
{
"epoch": 8.388483749947309,
"grad_norm": 0.9506089091300964,
"learning_rate": 6.758564507138876e-05,
"loss": 1.6662,
"step": 199000
},
{
"epoch": 8.409560342283859,
"grad_norm": 0.92261803150177,
"learning_rate": 6.749989280967286e-05,
"loss": 1.6637,
"step": 199500
},
{
"epoch": 8.430636934620411,
"grad_norm": 0.9438258409500122,
"learning_rate": 6.741414054795696e-05,
"loss": 1.6587,
"step": 200000
},
{
"epoch": 8.430636934620411,
"eval_accuracy": 0.6930334237131099,
"eval_loss": 1.41603684425354,
"eval_runtime": 196.6626,
"eval_samples_per_second": 613.772,
"eval_steps_per_second": 6.397,
"step": 200000
},
{
"epoch": 8.451713526956961,
"grad_norm": 0.9642271995544434,
"learning_rate": 6.732838828624105e-05,
"loss": 1.6575,
"step": 200500
},
{
"epoch": 8.472790119293512,
"grad_norm": 0.9528006911277771,
"learning_rate": 6.724263602452515e-05,
"loss": 1.6583,
"step": 201000
},
{
"epoch": 8.493866711630064,
"grad_norm": 0.9815327525138855,
"learning_rate": 6.715688376280924e-05,
"loss": 1.6545,
"step": 201500
},
{
"epoch": 8.514943303966614,
"grad_norm": 0.9433075189590454,
"learning_rate": 6.707113150109335e-05,
"loss": 1.6534,
"step": 202000
},
{
"epoch": 8.536019896303166,
"grad_norm": 0.9370937943458557,
"learning_rate": 6.698537923937745e-05,
"loss": 1.6555,
"step": 202500
},
{
"epoch": 8.557096488639717,
"grad_norm": 0.9300549626350403,
"learning_rate": 6.689962697766153e-05,
"loss": 1.6489,
"step": 203000
},
{
"epoch": 8.578173080976267,
"grad_norm": 0.9458557367324829,
"learning_rate": 6.681387471594563e-05,
"loss": 1.6475,
"step": 203500
},
{
"epoch": 8.59924967331282,
"grad_norm": 0.943783164024353,
"learning_rate": 6.672812245422974e-05,
"loss": 1.6443,
"step": 204000
},
{
"epoch": 8.62032626564937,
"grad_norm": 0.9351052641868591,
"learning_rate": 6.664237019251383e-05,
"loss": 1.6448,
"step": 204500
},
{
"epoch": 8.64140285798592,
"grad_norm": 0.9596394300460815,
"learning_rate": 6.655661793079792e-05,
"loss": 1.6432,
"step": 205000
},
{
"epoch": 8.662479450322472,
"grad_norm": 0.9447216391563416,
"learning_rate": 6.647086566908202e-05,
"loss": 1.637,
"step": 205500
},
{
"epoch": 8.683556042659022,
"grad_norm": 0.9592314958572388,
"learning_rate": 6.638511340736612e-05,
"loss": 1.6374,
"step": 206000
},
{
"epoch": 8.704632634995574,
"grad_norm": 0.9421446919441223,
"learning_rate": 6.629936114565022e-05,
"loss": 1.6336,
"step": 206500
},
{
"epoch": 8.725709227332125,
"grad_norm": 0.9309420585632324,
"learning_rate": 6.621360888393432e-05,
"loss": 1.6347,
"step": 207000
},
{
"epoch": 8.746785819668675,
"grad_norm": 0.925533652305603,
"learning_rate": 6.61278566222184e-05,
"loss": 1.6268,
"step": 207500
},
{
"epoch": 8.767862412005227,
"grad_norm": 0.9245041012763977,
"learning_rate": 6.604210436050251e-05,
"loss": 1.6317,
"step": 208000
},
{
"epoch": 8.788939004341778,
"grad_norm": 0.9388335943222046,
"learning_rate": 6.595635209878661e-05,
"loss": 1.6308,
"step": 208500
},
{
"epoch": 8.81001559667833,
"grad_norm": 0.9266485571861267,
"learning_rate": 6.58705998370707e-05,
"loss": 1.6308,
"step": 209000
},
{
"epoch": 8.83109218901488,
"grad_norm": 0.9375306367874146,
"learning_rate": 6.57848475753548e-05,
"loss": 1.63,
"step": 209500
},
{
"epoch": 8.85216878135143,
"grad_norm": 0.9538915753364563,
"learning_rate": 6.569909531363891e-05,
"loss": 1.6247,
"step": 210000
},
{
"epoch": 8.85216878135143,
"eval_accuracy": 0.6976776809219075,
"eval_loss": 1.388360857963562,
"eval_runtime": 198.7109,
"eval_samples_per_second": 607.445,
"eval_steps_per_second": 6.331,
"step": 210000
},
{
"epoch": 8.873245373687983,
"grad_norm": 0.9307466745376587,
"learning_rate": 6.561334305192299e-05,
"loss": 1.6253,
"step": 210500
},
{
"epoch": 8.894321966024533,
"grad_norm": 0.9394568800926208,
"learning_rate": 6.552759079020709e-05,
"loss": 1.638,
"step": 211000
},
{
"epoch": 8.915398558361083,
"grad_norm": 0.9375216960906982,
"learning_rate": 6.544183852849119e-05,
"loss": 1.6227,
"step": 211500
},
{
"epoch": 8.936475150697635,
"grad_norm": 0.9225177764892578,
"learning_rate": 6.535608626677529e-05,
"loss": 1.6256,
"step": 212000
},
{
"epoch": 8.957551743034186,
"grad_norm": 0.9250805974006653,
"learning_rate": 6.527033400505939e-05,
"loss": 1.6228,
"step": 212500
},
{
"epoch": 8.978628335370738,
"grad_norm": 0.958017110824585,
"learning_rate": 6.518458174334348e-05,
"loss": 1.6222,
"step": 213000
},
{
"epoch": 8.999704927707288,
"grad_norm": 0.94196617603302,
"learning_rate": 6.509882948162758e-05,
"loss": 1.6194,
"step": 213500
},
{
"epoch": 9.020781520043839,
"grad_norm": 0.9748749732971191,
"learning_rate": 6.501307721991168e-05,
"loss": 1.6131,
"step": 214000
},
{
"epoch": 9.04185811238039,
"grad_norm": 0.9482659101486206,
"learning_rate": 6.492732495819578e-05,
"loss": 1.6105,
"step": 214500
},
{
"epoch": 9.062934704716941,
"grad_norm": 0.967606246471405,
"learning_rate": 6.484157269647986e-05,
"loss": 1.6111,
"step": 215000
},
{
"epoch": 9.084011297053493,
"grad_norm": 0.9490759968757629,
"learning_rate": 6.475582043476398e-05,
"loss": 1.6098,
"step": 215500
},
{
"epoch": 9.105087889390044,
"grad_norm": 1.0035717487335205,
"learning_rate": 6.467006817304807e-05,
"loss": 1.606,
"step": 216000
},
{
"epoch": 9.126164481726594,
"grad_norm": 0.9482417106628418,
"learning_rate": 6.458431591133216e-05,
"loss": 1.601,
"step": 216500
},
{
"epoch": 9.147241074063146,
"grad_norm": 0.9601401090621948,
"learning_rate": 6.449856364961626e-05,
"loss": 1.6046,
"step": 217000
},
{
"epoch": 9.168317666399696,
"grad_norm": 0.9478941559791565,
"learning_rate": 6.441281138790037e-05,
"loss": 1.6014,
"step": 217500
},
{
"epoch": 9.189394258736247,
"grad_norm": 0.9348861575126648,
"learning_rate": 6.432705912618445e-05,
"loss": 1.6016,
"step": 218000
},
{
"epoch": 9.210470851072799,
"grad_norm": 0.9269882440567017,
"learning_rate": 6.424130686446855e-05,
"loss": 1.6022,
"step": 218500
},
{
"epoch": 9.23154744340935,
"grad_norm": 0.9308039546012878,
"learning_rate": 6.415555460275265e-05,
"loss": 1.5997,
"step": 219000
},
{
"epoch": 9.252624035745901,
"grad_norm": 0.9314356446266174,
"learning_rate": 6.406980234103675e-05,
"loss": 1.5988,
"step": 219500
},
{
"epoch": 9.273700628082452,
"grad_norm": 0.9690579771995544,
"learning_rate": 6.398405007932085e-05,
"loss": 1.5996,
"step": 220000
},
{
"epoch": 9.273700628082452,
"eval_accuracy": 0.7022905750435018,
"eval_loss": 1.3622843027114868,
"eval_runtime": 198.6682,
"eval_samples_per_second": 607.576,
"eval_steps_per_second": 6.332,
"step": 220000
},
{
"epoch": 9.294777220419002,
"grad_norm": 0.9188095927238464,
"learning_rate": 6.389829781760495e-05,
"loss": 1.5954,
"step": 220500
},
{
"epoch": 9.315853812755554,
"grad_norm": 0.960027277469635,
"learning_rate": 6.381254555588903e-05,
"loss": 1.5978,
"step": 221000
},
{
"epoch": 9.336930405092104,
"grad_norm": 0.9335135221481323,
"learning_rate": 6.372679329417314e-05,
"loss": 1.5915,
"step": 221500
},
{
"epoch": 9.358006997428657,
"grad_norm": 0.9714720845222473,
"learning_rate": 6.364104103245724e-05,
"loss": 1.5887,
"step": 222000
},
{
"epoch": 9.379083589765207,
"grad_norm": 0.947618842124939,
"learning_rate": 6.355528877074132e-05,
"loss": 1.5929,
"step": 222500
},
{
"epoch": 9.400160182101757,
"grad_norm": 0.9696279168128967,
"learning_rate": 6.346953650902542e-05,
"loss": 1.5896,
"step": 223000
},
{
"epoch": 9.42123677443831,
"grad_norm": 0.9514039754867554,
"learning_rate": 6.338378424730954e-05,
"loss": 1.5887,
"step": 223500
},
{
"epoch": 9.44231336677486,
"grad_norm": 0.9479451179504395,
"learning_rate": 6.329803198559362e-05,
"loss": 1.5858,
"step": 224000
},
{
"epoch": 9.46338995911141,
"grad_norm": 0.954346239566803,
"learning_rate": 6.321227972387772e-05,
"loss": 1.5852,
"step": 224500
},
{
"epoch": 9.484466551447962,
"grad_norm": 0.9417235851287842,
"learning_rate": 6.312652746216182e-05,
"loss": 1.5806,
"step": 225000
},
{
"epoch": 9.505543143784513,
"grad_norm": 0.9479976296424866,
"learning_rate": 6.304077520044591e-05,
"loss": 1.5877,
"step": 225500
},
{
"epoch": 9.526619736121065,
"grad_norm": 0.9549857974052429,
"learning_rate": 6.295502293873001e-05,
"loss": 1.5817,
"step": 226000
},
{
"epoch": 9.547696328457615,
"grad_norm": 0.9648552536964417,
"learning_rate": 6.286927067701411e-05,
"loss": 1.5773,
"step": 226500
},
{
"epoch": 9.568772920794165,
"grad_norm": 0.9564752578735352,
"learning_rate": 6.278351841529821e-05,
"loss": 1.5811,
"step": 227000
},
{
"epoch": 9.589849513130718,
"grad_norm": 0.9471539855003357,
"learning_rate": 6.269776615358231e-05,
"loss": 1.5793,
"step": 227500
},
{
"epoch": 9.610926105467268,
"grad_norm": 0.9404253363609314,
"learning_rate": 6.26120138918664e-05,
"loss": 1.5767,
"step": 228000
},
{
"epoch": 9.63200269780382,
"grad_norm": 0.9346477389335632,
"learning_rate": 6.252626163015049e-05,
"loss": 1.5764,
"step": 228500
},
{
"epoch": 9.65307929014037,
"grad_norm": 0.9503133893013,
"learning_rate": 6.24405093684346e-05,
"loss": 1.5731,
"step": 229000
},
{
"epoch": 9.67415588247692,
"grad_norm": 0.9383291602134705,
"learning_rate": 6.23547571067187e-05,
"loss": 1.571,
"step": 229500
},
{
"epoch": 9.695232474813473,
"grad_norm": 0.9274345636367798,
"learning_rate": 6.226900484500279e-05,
"loss": 1.5686,
"step": 230000
},
{
"epoch": 9.695232474813473,
"eval_accuracy": 0.7061599202201626,
"eval_loss": 1.3386777639389038,
"eval_runtime": 200.3485,
"eval_samples_per_second": 602.48,
"eval_steps_per_second": 6.279,
"step": 230000
},
{
"epoch": 9.716309067150023,
"grad_norm": 0.9494515061378479,
"learning_rate": 6.218325258328688e-05,
"loss": 1.5749,
"step": 230500
},
{
"epoch": 9.737385659486574,
"grad_norm": 0.936905026435852,
"learning_rate": 6.209750032157098e-05,
"loss": 1.5719,
"step": 231000
},
{
"epoch": 9.758462251823126,
"grad_norm": 0.9716399908065796,
"learning_rate": 6.201174805985508e-05,
"loss": 1.5673,
"step": 231500
},
{
"epoch": 9.779538844159676,
"grad_norm": 0.947190523147583,
"learning_rate": 6.192599579813918e-05,
"loss": 1.5691,
"step": 232000
},
{
"epoch": 9.800615436496226,
"grad_norm": 0.9388655424118042,
"learning_rate": 6.184024353642328e-05,
"loss": 1.5651,
"step": 232500
},
{
"epoch": 9.821692028832778,
"grad_norm": 0.9481565952301025,
"learning_rate": 6.175449127470738e-05,
"loss": 1.563,
"step": 233000
},
{
"epoch": 9.842768621169329,
"grad_norm": 0.95245361328125,
"learning_rate": 6.166873901299147e-05,
"loss": 1.5683,
"step": 233500
},
{
"epoch": 9.863845213505881,
"grad_norm": 0.9404584169387817,
"learning_rate": 6.158298675127557e-05,
"loss": 1.562,
"step": 234000
},
{
"epoch": 9.884921805842431,
"grad_norm": 0.9723191857337952,
"learning_rate": 6.149723448955966e-05,
"loss": 1.5647,
"step": 234500
},
{
"epoch": 9.905998398178982,
"grad_norm": 0.933776319026947,
"learning_rate": 6.141148222784377e-05,
"loss": 1.5585,
"step": 235000
},
{
"epoch": 9.927074990515534,
"grad_norm": 0.9403916597366333,
"learning_rate": 6.132572996612787e-05,
"loss": 1.5575,
"step": 235500
},
{
"epoch": 9.948151582852084,
"grad_norm": 0.9502066373825073,
"learning_rate": 6.123997770441195e-05,
"loss": 1.5612,
"step": 236000
},
{
"epoch": 9.969228175188636,
"grad_norm": 0.959568440914154,
"learning_rate": 6.115422544269605e-05,
"loss": 1.5621,
"step": 236500
},
{
"epoch": 9.990304767525187,
"grad_norm": 0.9627484083175659,
"learning_rate": 6.106847318098015e-05,
"loss": 1.5586,
"step": 237000
},
{
"epoch": 10.011381359861737,
"grad_norm": 0.9339560270309448,
"learning_rate": 6.098272091926425e-05,
"loss": 1.5562,
"step": 237500
},
{
"epoch": 10.032457952198289,
"grad_norm": 0.9296694993972778,
"learning_rate": 6.0896968657548345e-05,
"loss": 1.5548,
"step": 238000
},
{
"epoch": 10.05353454453484,
"grad_norm": 0.973708987236023,
"learning_rate": 6.081121639583244e-05,
"loss": 1.5502,
"step": 238500
},
{
"epoch": 10.07461113687139,
"grad_norm": 0.9420864582061768,
"learning_rate": 6.072546413411654e-05,
"loss": 1.5519,
"step": 239000
},
{
"epoch": 10.095687729207942,
"grad_norm": 0.9263563752174377,
"learning_rate": 6.063971187240064e-05,
"loss": 1.5507,
"step": 239500
},
{
"epoch": 10.116764321544492,
"grad_norm": 0.9529481530189514,
"learning_rate": 6.055395961068473e-05,
"loss": 1.5445,
"step": 240000
},
{
"epoch": 10.116764321544492,
"eval_accuracy": 0.709864268776927,
"eval_loss": 1.3200523853302002,
"eval_runtime": 198.6305,
"eval_samples_per_second": 607.691,
"eval_steps_per_second": 6.333,
"step": 240000
},
{
"epoch": 10.137840913881044,
"grad_norm": 0.9865480661392212,
"learning_rate": 6.046820734896883e-05,
"loss": 1.5457,
"step": 240500
},
{
"epoch": 10.158917506217595,
"grad_norm": 0.9718133211135864,
"learning_rate": 6.038245508725293e-05,
"loss": 1.5475,
"step": 241000
},
{
"epoch": 10.179994098554145,
"grad_norm": 0.9525668025016785,
"learning_rate": 6.029670282553702e-05,
"loss": 1.5448,
"step": 241500
},
{
"epoch": 10.201070690890697,
"grad_norm": 0.959688663482666,
"learning_rate": 6.0210950563821125e-05,
"loss": 1.5483,
"step": 242000
},
{
"epoch": 10.222147283227248,
"grad_norm": 0.961243212223053,
"learning_rate": 6.0125198302105223e-05,
"loss": 1.5457,
"step": 242500
},
{
"epoch": 10.2432238755638,
"grad_norm": 0.9684420228004456,
"learning_rate": 6.0039446040389315e-05,
"loss": 1.5423,
"step": 243000
},
{
"epoch": 10.26430046790035,
"grad_norm": 0.9492542147636414,
"learning_rate": 5.995369377867341e-05,
"loss": 1.547,
"step": 243500
},
{
"epoch": 10.2853770602369,
"grad_norm": 0.9532374143600464,
"learning_rate": 5.986794151695752e-05,
"loss": 1.5371,
"step": 244000
},
{
"epoch": 10.306453652573452,
"grad_norm": 0.9588803052902222,
"learning_rate": 5.97821892552416e-05,
"loss": 1.5435,
"step": 244500
},
{
"epoch": 10.327530244910003,
"grad_norm": 0.9393641352653503,
"learning_rate": 5.969643699352571e-05,
"loss": 1.5421,
"step": 245000
},
{
"epoch": 10.348606837246553,
"grad_norm": 0.9422447085380554,
"learning_rate": 5.9610684731809806e-05,
"loss": 1.5399,
"step": 245500
},
{
"epoch": 10.369683429583105,
"grad_norm": 0.9279852509498596,
"learning_rate": 5.95249324700939e-05,
"loss": 1.5372,
"step": 246000
},
{
"epoch": 10.390760021919656,
"grad_norm": 0.9491358399391174,
"learning_rate": 5.9439180208377996e-05,
"loss": 1.5384,
"step": 246500
},
{
"epoch": 10.411836614256208,
"grad_norm": 0.9157857894897461,
"learning_rate": 5.93534279466621e-05,
"loss": 1.5354,
"step": 247000
},
{
"epoch": 10.432913206592758,
"grad_norm": 0.9285153150558472,
"learning_rate": 5.9267675684946186e-05,
"loss": 1.5371,
"step": 247500
},
{
"epoch": 10.453989798929308,
"grad_norm": 0.960303783416748,
"learning_rate": 5.918192342323029e-05,
"loss": 1.5315,
"step": 248000
},
{
"epoch": 10.47506639126586,
"grad_norm": 0.9521673321723938,
"learning_rate": 5.909617116151439e-05,
"loss": 1.5317,
"step": 248500
},
{
"epoch": 10.496142983602411,
"grad_norm": 0.9762712717056274,
"learning_rate": 5.901041889979848e-05,
"loss": 1.5302,
"step": 249000
},
{
"epoch": 10.517219575938963,
"grad_norm": 0.9871359467506409,
"learning_rate": 5.892466663808258e-05,
"loss": 1.5323,
"step": 249500
},
{
"epoch": 10.538296168275513,
"grad_norm": 0.9665715098381042,
"learning_rate": 5.8838914376366684e-05,
"loss": 1.5316,
"step": 250000
},
{
"epoch": 10.538296168275513,
"eval_accuracy": 0.7127684666397748,
"eval_loss": 1.3001984357833862,
"eval_runtime": 199.6945,
"eval_samples_per_second": 604.453,
"eval_steps_per_second": 6.3,
"step": 250000
},
{
"epoch": 10.559372760612064,
"grad_norm": 0.9583009481430054,
"learning_rate": 5.8753162114650776e-05,
"loss": 1.5347,
"step": 250500
},
{
"epoch": 10.580449352948616,
"grad_norm": 0.9629523158073425,
"learning_rate": 5.8667409852934874e-05,
"loss": 1.5273,
"step": 251000
},
{
"epoch": 10.601525945285166,
"grad_norm": 0.9398017525672913,
"learning_rate": 5.858165759121897e-05,
"loss": 1.5241,
"step": 251500
},
{
"epoch": 10.622602537621717,
"grad_norm": 0.9691195487976074,
"learning_rate": 5.8495905329503064e-05,
"loss": 1.5278,
"step": 252000
},
{
"epoch": 10.643679129958269,
"grad_norm": 0.9120731353759766,
"learning_rate": 5.841015306778716e-05,
"loss": 1.527,
"step": 252500
},
{
"epoch": 10.664755722294819,
"grad_norm": 0.9803792238235474,
"learning_rate": 5.832440080607127e-05,
"loss": 1.5264,
"step": 253000
},
{
"epoch": 10.685832314631371,
"grad_norm": 0.9461917877197266,
"learning_rate": 5.823864854435536e-05,
"loss": 1.5205,
"step": 253500
},
{
"epoch": 10.706908906967922,
"grad_norm": 0.95630943775177,
"learning_rate": 5.815289628263946e-05,
"loss": 1.5236,
"step": 254000
},
{
"epoch": 10.727985499304472,
"grad_norm": 0.9591798782348633,
"learning_rate": 5.8067144020923556e-05,
"loss": 1.5269,
"step": 254500
},
{
"epoch": 10.749062091641024,
"grad_norm": 0.9686102867126465,
"learning_rate": 5.798139175920765e-05,
"loss": 1.522,
"step": 255000
},
{
"epoch": 10.770138683977574,
"grad_norm": 0.9659392833709717,
"learning_rate": 5.789563949749175e-05,
"loss": 1.5231,
"step": 255500
},
{
"epoch": 10.791215276314126,
"grad_norm": 0.9424290060997009,
"learning_rate": 5.780988723577585e-05,
"loss": 1.5212,
"step": 256000
},
{
"epoch": 10.812291868650677,
"grad_norm": 0.933353066444397,
"learning_rate": 5.772413497405994e-05,
"loss": 1.5212,
"step": 256500
},
{
"epoch": 10.833368460987227,
"grad_norm": 0.9382802248001099,
"learning_rate": 5.763838271234404e-05,
"loss": 1.5175,
"step": 257000
},
{
"epoch": 10.85444505332378,
"grad_norm": 0.9676141142845154,
"learning_rate": 5.7552630450628146e-05,
"loss": 1.5184,
"step": 257500
},
{
"epoch": 10.87552164566033,
"grad_norm": 0.9440897107124329,
"learning_rate": 5.746687818891223e-05,
"loss": 1.5157,
"step": 258000
},
{
"epoch": 10.89659823799688,
"grad_norm": 0.9669468402862549,
"learning_rate": 5.7381125927196335e-05,
"loss": 1.5174,
"step": 258500
},
{
"epoch": 10.917674830333432,
"grad_norm": 0.9550685882568359,
"learning_rate": 5.7295373665480434e-05,
"loss": 1.517,
"step": 259000
},
{
"epoch": 10.938751422669982,
"grad_norm": 0.9737645983695984,
"learning_rate": 5.7209621403764525e-05,
"loss": 1.5151,
"step": 259500
},
{
"epoch": 10.959828015006533,
"grad_norm": 0.9722259044647217,
"learning_rate": 5.7123869142048624e-05,
"loss": 1.51,
"step": 260000
},
{
"epoch": 10.959828015006533,
"eval_accuracy": 0.7156418557643824,
"eval_loss": 1.2850353717803955,
"eval_runtime": 199.7964,
"eval_samples_per_second": 604.145,
"eval_steps_per_second": 6.296,
"step": 260000
},
{
"epoch": 10.980904607343085,
"grad_norm": 0.9562275409698486,
"learning_rate": 5.703811688033273e-05,
"loss": 1.5087,
"step": 260500
},
{
"epoch": 11.001981199679635,
"grad_norm": 0.9654918313026428,
"learning_rate": 5.695236461861681e-05,
"loss": 1.5111,
"step": 261000
},
{
"epoch": 11.023057792016187,
"grad_norm": 0.9665001034736633,
"learning_rate": 5.686661235690092e-05,
"loss": 1.5046,
"step": 261500
},
{
"epoch": 11.044134384352738,
"grad_norm": 0.965803325176239,
"learning_rate": 5.678086009518502e-05,
"loss": 1.5099,
"step": 262000
},
{
"epoch": 11.065210976689288,
"grad_norm": 0.9833449125289917,
"learning_rate": 5.669510783346911e-05,
"loss": 1.511,
"step": 262500
},
{
"epoch": 11.08628756902584,
"grad_norm": 0.9571903944015503,
"learning_rate": 5.6609355571753207e-05,
"loss": 1.504,
"step": 263000
},
{
"epoch": 11.10736416136239,
"grad_norm": 0.9918251633644104,
"learning_rate": 5.652360331003731e-05,
"loss": 1.5068,
"step": 263500
},
{
"epoch": 11.128440753698943,
"grad_norm": 0.9502744078636169,
"learning_rate": 5.6437851048321396e-05,
"loss": 1.5021,
"step": 264000
},
{
"epoch": 11.149517346035493,
"grad_norm": 0.9744539260864258,
"learning_rate": 5.63520987866055e-05,
"loss": 1.5063,
"step": 264500
},
{
"epoch": 11.170593938372043,
"grad_norm": 0.9713169932365417,
"learning_rate": 5.62663465248896e-05,
"loss": 1.5033,
"step": 265000
},
{
"epoch": 11.191670530708596,
"grad_norm": 0.9731369018554688,
"learning_rate": 5.618059426317369e-05,
"loss": 1.5006,
"step": 265500
},
{
"epoch": 11.212747123045146,
"grad_norm": 0.9724516868591309,
"learning_rate": 5.609484200145779e-05,
"loss": 1.5046,
"step": 266000
},
{
"epoch": 11.233823715381696,
"grad_norm": 0.954821765422821,
"learning_rate": 5.6009089739741895e-05,
"loss": 1.5023,
"step": 266500
},
{
"epoch": 11.254900307718248,
"grad_norm": 0.9708914756774902,
"learning_rate": 5.5923337478025986e-05,
"loss": 1.499,
"step": 267000
},
{
"epoch": 11.275976900054799,
"grad_norm": 0.9710102677345276,
"learning_rate": 5.5837585216310085e-05,
"loss": 1.5027,
"step": 267500
},
{
"epoch": 11.29705349239135,
"grad_norm": 1.0020370483398438,
"learning_rate": 5.575183295459418e-05,
"loss": 1.4984,
"step": 268000
},
{
"epoch": 11.318130084727901,
"grad_norm": 0.9543415904045105,
"learning_rate": 5.5666080692878274e-05,
"loss": 1.4997,
"step": 268500
},
{
"epoch": 11.339206677064452,
"grad_norm": 0.9699862599372864,
"learning_rate": 5.558032843116238e-05,
"loss": 1.4993,
"step": 269000
},
{
"epoch": 11.360283269401004,
"grad_norm": 0.9720067977905273,
"learning_rate": 5.549457616944648e-05,
"loss": 1.4976,
"step": 269500
},
{
"epoch": 11.381359861737554,
"grad_norm": 0.9743959307670593,
"learning_rate": 5.540882390773057e-05,
"loss": 1.4938,
"step": 270000
},
{
"epoch": 11.381359861737554,
"eval_accuracy": 0.7177530388598284,
"eval_loss": 1.272760033607483,
"eval_runtime": 198.8845,
"eval_samples_per_second": 606.915,
"eval_steps_per_second": 6.325,
"step": 270000
},
{
"epoch": 11.402436454074106,
"grad_norm": 0.9850079417228699,
"learning_rate": 5.532307164601467e-05,
"loss": 1.4967,
"step": 270500
},
{
"epoch": 11.423513046410656,
"grad_norm": 0.9674860239028931,
"learning_rate": 5.5237319384298766e-05,
"loss": 1.4934,
"step": 271000
},
{
"epoch": 11.444589638747207,
"grad_norm": 0.9405706524848938,
"learning_rate": 5.515156712258286e-05,
"loss": 1.4919,
"step": 271500
},
{
"epoch": 11.465666231083759,
"grad_norm": 0.9367933869361877,
"learning_rate": 5.506581486086696e-05,
"loss": 1.4932,
"step": 272000
},
{
"epoch": 11.48674282342031,
"grad_norm": 0.9718525409698486,
"learning_rate": 5.498006259915105e-05,
"loss": 1.4926,
"step": 272500
},
{
"epoch": 11.50781941575686,
"grad_norm": 0.9609485268592834,
"learning_rate": 5.489431033743515e-05,
"loss": 1.4932,
"step": 273000
},
{
"epoch": 11.528896008093412,
"grad_norm": 0.9474647641181946,
"learning_rate": 5.480855807571925e-05,
"loss": 1.4932,
"step": 273500
},
{
"epoch": 11.549972600429962,
"grad_norm": 0.9643217921257019,
"learning_rate": 5.472280581400334e-05,
"loss": 1.4883,
"step": 274000
},
{
"epoch": 11.571049192766514,
"grad_norm": 0.9918298721313477,
"learning_rate": 5.463705355228744e-05,
"loss": 1.4892,
"step": 274500
},
{
"epoch": 11.592125785103065,
"grad_norm": 0.9625241756439209,
"learning_rate": 5.4551301290571546e-05,
"loss": 1.4926,
"step": 275000
},
{
"epoch": 11.613202377439615,
"grad_norm": 0.9726247191429138,
"learning_rate": 5.446554902885563e-05,
"loss": 1.489,
"step": 275500
},
{
"epoch": 11.634278969776167,
"grad_norm": 0.9522711038589478,
"learning_rate": 5.4379796767139736e-05,
"loss": 1.4859,
"step": 276000
},
{
"epoch": 11.655355562112717,
"grad_norm": 0.9779261350631714,
"learning_rate": 5.4294044505423834e-05,
"loss": 1.4853,
"step": 276500
},
{
"epoch": 11.67643215444927,
"grad_norm": 0.9578256011009216,
"learning_rate": 5.4208292243707925e-05,
"loss": 1.4865,
"step": 277000
},
{
"epoch": 11.69750874678582,
"grad_norm": 0.9562909007072449,
"learning_rate": 5.4122539981992024e-05,
"loss": 1.4866,
"step": 277500
},
{
"epoch": 11.71858533912237,
"grad_norm": 0.9364312887191772,
"learning_rate": 5.403678772027613e-05,
"loss": 1.4814,
"step": 278000
},
{
"epoch": 11.739661931458922,
"grad_norm": 0.9668327569961548,
"learning_rate": 5.395103545856022e-05,
"loss": 1.4848,
"step": 278500
},
{
"epoch": 11.760738523795473,
"grad_norm": 0.9704906344413757,
"learning_rate": 5.386528319684432e-05,
"loss": 1.4822,
"step": 279000
},
{
"epoch": 11.781815116132023,
"grad_norm": 0.9373304843902588,
"learning_rate": 5.377953093512842e-05,
"loss": 1.4787,
"step": 279500
},
{
"epoch": 11.802891708468575,
"grad_norm": 0.9551854729652405,
"learning_rate": 5.369377867341251e-05,
"loss": 1.4864,
"step": 280000
},
{
"epoch": 11.802891708468575,
"eval_accuracy": 0.7204604516824319,
"eval_loss": 1.2573603391647339,
"eval_runtime": 201.7495,
"eval_samples_per_second": 598.296,
"eval_steps_per_second": 6.235,
"step": 280000
},
{
"epoch": 11.823968300805126,
"grad_norm": 0.9775806069374084,
"learning_rate": 5.3608026411696614e-05,
"loss": 1.486,
"step": 280500
},
{
"epoch": 11.845044893141678,
"grad_norm": 0.9645451307296753,
"learning_rate": 5.352227414998071e-05,
"loss": 1.4754,
"step": 281000
},
{
"epoch": 11.866121485478228,
"grad_norm": 0.9879536032676697,
"learning_rate": 5.34365218882648e-05,
"loss": 1.4801,
"step": 281500
},
{
"epoch": 11.887198077814778,
"grad_norm": 0.9586070775985718,
"learning_rate": 5.33507696265489e-05,
"loss": 1.474,
"step": 282000
},
{
"epoch": 11.90827467015133,
"grad_norm": 0.9457542300224304,
"learning_rate": 5.3265017364833e-05,
"loss": 1.4809,
"step": 282500
},
{
"epoch": 11.92935126248788,
"grad_norm": 0.9921045303344727,
"learning_rate": 5.317926510311709e-05,
"loss": 1.4767,
"step": 283000
},
{
"epoch": 11.950427854824433,
"grad_norm": 0.9832741022109985,
"learning_rate": 5.3093512841401197e-05,
"loss": 1.4755,
"step": 283500
},
{
"epoch": 11.971504447160983,
"grad_norm": 0.9533354640007019,
"learning_rate": 5.3007760579685295e-05,
"loss": 1.4778,
"step": 284000
},
{
"epoch": 11.992581039497534,
"grad_norm": 0.9728748202323914,
"learning_rate": 5.2922008317969386e-05,
"loss": 1.4791,
"step": 284500
},
{
"epoch": 12.013657631834086,
"grad_norm": 0.9833368062973022,
"learning_rate": 5.2836256056253485e-05,
"loss": 1.4737,
"step": 285000
},
{
"epoch": 12.034734224170636,
"grad_norm": 0.9677203893661499,
"learning_rate": 5.275050379453759e-05,
"loss": 1.4749,
"step": 285500
},
{
"epoch": 12.055810816507186,
"grad_norm": 0.9336782097816467,
"learning_rate": 5.2664751532821675e-05,
"loss": 1.471,
"step": 286000
},
{
"epoch": 12.076887408843739,
"grad_norm": 0.9662377238273621,
"learning_rate": 5.257899927110578e-05,
"loss": 1.4712,
"step": 286500
},
{
"epoch": 12.097964001180289,
"grad_norm": 0.9482128620147705,
"learning_rate": 5.249324700938988e-05,
"loss": 1.4667,
"step": 287000
},
{
"epoch": 12.119040593516841,
"grad_norm": 0.9476494789123535,
"learning_rate": 5.240749474767397e-05,
"loss": 1.4669,
"step": 287500
},
{
"epoch": 12.140117185853391,
"grad_norm": 0.9918713569641113,
"learning_rate": 5.232174248595807e-05,
"loss": 1.4702,
"step": 288000
},
{
"epoch": 12.161193778189942,
"grad_norm": 0.9539222717285156,
"learning_rate": 5.223599022424217e-05,
"loss": 1.4708,
"step": 288500
},
{
"epoch": 12.182270370526494,
"grad_norm": 0.9776086211204529,
"learning_rate": 5.215023796252626e-05,
"loss": 1.4741,
"step": 289000
},
{
"epoch": 12.203346962863044,
"grad_norm": 0.9755340218544006,
"learning_rate": 5.206448570081036e-05,
"loss": 1.4647,
"step": 289500
},
{
"epoch": 12.224423555199595,
"grad_norm": 0.9956245422363281,
"learning_rate": 5.197873343909446e-05,
"loss": 1.4641,
"step": 290000
},
{
"epoch": 12.224423555199595,
"eval_accuracy": 0.7227630966105364,
"eval_loss": 1.2452895641326904,
"eval_runtime": 198.9275,
"eval_samples_per_second": 606.784,
"eval_steps_per_second": 6.324,
"step": 290000
},
{
"epoch": 12.245500147536147,
"grad_norm": 0.977632462978363,
"learning_rate": 5.189298117737855e-05,
"loss": 1.466,
"step": 290500
},
{
"epoch": 12.266576739872697,
"grad_norm": 0.9593811631202698,
"learning_rate": 5.180722891566265e-05,
"loss": 1.4641,
"step": 291000
},
{
"epoch": 12.28765333220925,
"grad_norm": 0.980034589767456,
"learning_rate": 5.1721476653946756e-05,
"loss": 1.4609,
"step": 291500
},
{
"epoch": 12.3087299245458,
"grad_norm": 0.971948504447937,
"learning_rate": 5.163572439223085e-05,
"loss": 1.4644,
"step": 292000
},
{
"epoch": 12.32980651688235,
"grad_norm": 0.9738345146179199,
"learning_rate": 5.1549972130514946e-05,
"loss": 1.4642,
"step": 292500
},
{
"epoch": 12.350883109218902,
"grad_norm": 0.9824228286743164,
"learning_rate": 5.1464219868799044e-05,
"loss": 1.4619,
"step": 293000
},
{
"epoch": 12.371959701555452,
"grad_norm": 0.992312490940094,
"learning_rate": 5.1378467607083136e-05,
"loss": 1.4611,
"step": 293500
},
{
"epoch": 12.393036293892003,
"grad_norm": 0.9575759172439575,
"learning_rate": 5.1292715345367234e-05,
"loss": 1.4615,
"step": 294000
},
{
"epoch": 12.414112886228555,
"grad_norm": 0.9596296548843384,
"learning_rate": 5.120696308365134e-05,
"loss": 1.4585,
"step": 294500
},
{
"epoch": 12.435189478565105,
"grad_norm": 0.9747764468193054,
"learning_rate": 5.112121082193543e-05,
"loss": 1.4618,
"step": 295000
},
{
"epoch": 12.456266070901657,
"grad_norm": 0.96734219789505,
"learning_rate": 5.103545856021953e-05,
"loss": 1.4601,
"step": 295500
},
{
"epoch": 12.477342663238208,
"grad_norm": 0.9791637063026428,
"learning_rate": 5.094970629850363e-05,
"loss": 1.4594,
"step": 296000
},
{
"epoch": 12.498419255574758,
"grad_norm": 0.9625028371810913,
"learning_rate": 5.086395403678772e-05,
"loss": 1.4588,
"step": 296500
},
{
"epoch": 12.51949584791131,
"grad_norm": 0.943059504032135,
"learning_rate": 5.0778201775071824e-05,
"loss": 1.4629,
"step": 297000
},
{
"epoch": 12.54057244024786,
"grad_norm": 0.9959601163864136,
"learning_rate": 5.069244951335592e-05,
"loss": 1.4611,
"step": 297500
},
{
"epoch": 12.561649032584413,
"grad_norm": 0.9991387128829956,
"learning_rate": 5.0606697251640014e-05,
"loss": 1.4547,
"step": 298000
},
{
"epoch": 12.582725624920963,
"grad_norm": 1.0090376138687134,
"learning_rate": 5.052094498992411e-05,
"loss": 1.4592,
"step": 298500
},
{
"epoch": 12.603802217257513,
"grad_norm": 0.9994741678237915,
"learning_rate": 5.043519272820822e-05,
"loss": 1.4591,
"step": 299000
},
{
"epoch": 12.624878809594065,
"grad_norm": 0.969109296798706,
"learning_rate": 5.03494404664923e-05,
"loss": 1.4527,
"step": 299500
},
{
"epoch": 12.645955401930616,
"grad_norm": 0.965074360370636,
"learning_rate": 5.026368820477641e-05,
"loss": 1.4549,
"step": 300000
},
{
"epoch": 12.645955401930616,
"eval_accuracy": 0.7249531846077395,
"eval_loss": 1.2323747873306274,
"eval_runtime": 198.1382,
"eval_samples_per_second": 609.201,
"eval_steps_per_second": 6.349,
"step": 300000
},
{
"epoch": 12.667031994267166,
"grad_norm": 0.9808958768844604,
"learning_rate": 5.0177935943060505e-05,
"loss": 1.4509,
"step": 300500
},
{
"epoch": 12.688108586603718,
"grad_norm": 0.9450556039810181,
"learning_rate": 5.00921836813446e-05,
"loss": 1.4565,
"step": 301000
},
{
"epoch": 12.709185178940269,
"grad_norm": 0.9691680669784546,
"learning_rate": 5.0006431419628695e-05,
"loss": 1.4534,
"step": 301500
},
{
"epoch": 12.73026177127682,
"grad_norm": 1.0097588300704956,
"learning_rate": 4.992067915791279e-05,
"loss": 1.453,
"step": 302000
},
{
"epoch": 12.751338363613371,
"grad_norm": 0.988410234451294,
"learning_rate": 4.9834926896196885e-05,
"loss": 1.449,
"step": 302500
},
{
"epoch": 12.772414955949921,
"grad_norm": 1.0279239416122437,
"learning_rate": 4.974917463448099e-05,
"loss": 1.4542,
"step": 303000
},
{
"epoch": 12.793491548286474,
"grad_norm": 0.9693096280097961,
"learning_rate": 4.966342237276508e-05,
"loss": 1.4529,
"step": 303500
},
{
"epoch": 12.814568140623024,
"grad_norm": 0.960210382938385,
"learning_rate": 4.957767011104918e-05,
"loss": 1.4509,
"step": 304000
},
{
"epoch": 12.835644732959576,
"grad_norm": 0.9725775122642517,
"learning_rate": 4.949191784933328e-05,
"loss": 1.4503,
"step": 304500
},
{
"epoch": 12.856721325296126,
"grad_norm": 0.9743361473083496,
"learning_rate": 4.9406165587617376e-05,
"loss": 1.4484,
"step": 305000
},
{
"epoch": 12.877797917632677,
"grad_norm": 0.9725016355514526,
"learning_rate": 4.932041332590147e-05,
"loss": 1.4517,
"step": 305500
},
{
"epoch": 12.898874509969229,
"grad_norm": 0.9798526763916016,
"learning_rate": 4.923466106418557e-05,
"loss": 1.4467,
"step": 306000
},
{
"epoch": 12.91995110230578,
"grad_norm": 0.9605358839035034,
"learning_rate": 4.9148908802469665e-05,
"loss": 1.4447,
"step": 306500
},
{
"epoch": 12.94102769464233,
"grad_norm": 0.960102915763855,
"learning_rate": 4.906315654075376e-05,
"loss": 1.4468,
"step": 307000
},
{
"epoch": 12.962104286978882,
"grad_norm": 1.0101896524429321,
"learning_rate": 4.897740427903786e-05,
"loss": 1.4504,
"step": 307500
},
{
"epoch": 12.983180879315432,
"grad_norm": 0.9817771911621094,
"learning_rate": 4.889165201732196e-05,
"loss": 1.4496,
"step": 308000
},
{
"epoch": 13.004257471651984,
"grad_norm": 0.9653884768486023,
"learning_rate": 4.880589975560606e-05,
"loss": 1.4449,
"step": 308500
},
{
"epoch": 13.025334063988534,
"grad_norm": 0.9859559535980225,
"learning_rate": 4.8720147493890156e-05,
"loss": 1.4392,
"step": 309000
},
{
"epoch": 13.046410656325085,
"grad_norm": 0.9902440309524536,
"learning_rate": 4.8634395232174254e-05,
"loss": 1.4429,
"step": 309500
},
{
"epoch": 13.067487248661637,
"grad_norm": 0.9645095467567444,
"learning_rate": 4.8548642970458346e-05,
"loss": 1.4394,
"step": 310000
},
{
"epoch": 13.067487248661637,
"eval_accuracy": 0.7269956651556853,
"eval_loss": 1.2211833000183105,
"eval_runtime": 196.8382,
"eval_samples_per_second": 613.224,
"eval_steps_per_second": 6.391,
"step": 310000
},
{
"epoch": 13.088563840998187,
"grad_norm": 0.9788839817047119,
"learning_rate": 4.846289070874245e-05,
"loss": 1.4412,
"step": 310500
},
{
"epoch": 13.109640433334738,
"grad_norm": 0.9942716360092163,
"learning_rate": 4.837713844702654e-05,
"loss": 1.4424,
"step": 311000
},
{
"epoch": 13.13071702567129,
"grad_norm": 1.0036206245422363,
"learning_rate": 4.829138618531064e-05,
"loss": 1.4373,
"step": 311500
},
{
"epoch": 13.15179361800784,
"grad_norm": 0.9784274697303772,
"learning_rate": 4.820563392359474e-05,
"loss": 1.4399,
"step": 312000
},
{
"epoch": 13.172870210344392,
"grad_norm": 1.0180612802505493,
"learning_rate": 4.811988166187884e-05,
"loss": 1.4371,
"step": 312500
},
{
"epoch": 13.193946802680943,
"grad_norm": 0.9884026646614075,
"learning_rate": 4.803412940016293e-05,
"loss": 1.4369,
"step": 313000
},
{
"epoch": 13.215023395017493,
"grad_norm": 0.99305260181427,
"learning_rate": 4.7948377138447034e-05,
"loss": 1.4388,
"step": 313500
},
{
"epoch": 13.236099987354045,
"grad_norm": 0.989567756652832,
"learning_rate": 4.7862624876731126e-05,
"loss": 1.4341,
"step": 314000
},
{
"epoch": 13.257176579690595,
"grad_norm": 0.9717433452606201,
"learning_rate": 4.7776872615015224e-05,
"loss": 1.437,
"step": 314500
},
{
"epoch": 13.278253172027147,
"grad_norm": 1.0006060600280762,
"learning_rate": 4.7691120353299315e-05,
"loss": 1.4325,
"step": 315000
},
{
"epoch": 13.299329764363698,
"grad_norm": 0.9722633957862854,
"learning_rate": 4.760536809158342e-05,
"loss": 1.4386,
"step": 315500
},
{
"epoch": 13.320406356700248,
"grad_norm": 0.9683562517166138,
"learning_rate": 4.751961582986751e-05,
"loss": 1.4387,
"step": 316000
},
{
"epoch": 13.3414829490368,
"grad_norm": 0.9635105133056641,
"learning_rate": 4.743386356815161e-05,
"loss": 1.4369,
"step": 316500
},
{
"epoch": 13.36255954137335,
"grad_norm": 0.9786714911460876,
"learning_rate": 4.734811130643571e-05,
"loss": 1.4342,
"step": 317000
},
{
"epoch": 13.383636133709901,
"grad_norm": 0.985614538192749,
"learning_rate": 4.726235904471981e-05,
"loss": 1.4314,
"step": 317500
},
{
"epoch": 13.404712726046453,
"grad_norm": 1.0081607103347778,
"learning_rate": 4.71766067830039e-05,
"loss": 1.4334,
"step": 318000
},
{
"epoch": 13.425789318383003,
"grad_norm": 0.9815026521682739,
"learning_rate": 4.7090854521288004e-05,
"loss": 1.4362,
"step": 318500
},
{
"epoch": 13.446865910719556,
"grad_norm": 0.9840613007545471,
"learning_rate": 4.7005102259572095e-05,
"loss": 1.4325,
"step": 319000
},
{
"epoch": 13.467942503056106,
"grad_norm": 1.0086958408355713,
"learning_rate": 4.6919349997856193e-05,
"loss": 1.4338,
"step": 319500
},
{
"epoch": 13.489019095392656,
"grad_norm": 0.9997546076774597,
"learning_rate": 4.683359773614029e-05,
"loss": 1.4298,
"step": 320000
},
{
"epoch": 13.489019095392656,
"eval_accuracy": 0.7283895508873508,
"eval_loss": 1.213509440422058,
"eval_runtime": 196.3074,
"eval_samples_per_second": 614.883,
"eval_steps_per_second": 6.408,
"step": 320000
},
{
"epoch": 13.510095687729208,
"grad_norm": 1.0275838375091553,
"learning_rate": 4.674784547442439e-05,
"loss": 1.4365,
"step": 320500
},
{
"epoch": 13.531172280065759,
"grad_norm": 0.9975188970565796,
"learning_rate": 4.666209321270849e-05,
"loss": 1.4311,
"step": 321000
},
{
"epoch": 13.55224887240231,
"grad_norm": 0.9739210605621338,
"learning_rate": 4.657634095099259e-05,
"loss": 1.4311,
"step": 321500
},
{
"epoch": 13.573325464738861,
"grad_norm": 0.9941136240959167,
"learning_rate": 4.6490588689276685e-05,
"loss": 1.4305,
"step": 322000
},
{
"epoch": 13.594402057075412,
"grad_norm": 0.9770312309265137,
"learning_rate": 4.6404836427560777e-05,
"loss": 1.4325,
"step": 322500
},
{
"epoch": 13.615478649411964,
"grad_norm": 1.0075894594192505,
"learning_rate": 4.6319084165844875e-05,
"loss": 1.4256,
"step": 323000
},
{
"epoch": 13.636555241748514,
"grad_norm": 0.9543708562850952,
"learning_rate": 4.623333190412897e-05,
"loss": 1.4305,
"step": 323500
},
{
"epoch": 13.657631834085064,
"grad_norm": 1.0083528757095337,
"learning_rate": 4.614757964241307e-05,
"loss": 1.4256,
"step": 324000
},
{
"epoch": 13.678708426421617,
"grad_norm": 0.9921474456787109,
"learning_rate": 4.606182738069717e-05,
"loss": 1.4273,
"step": 324500
},
{
"epoch": 13.699785018758167,
"grad_norm": 0.9553293585777283,
"learning_rate": 4.597607511898127e-05,
"loss": 1.4269,
"step": 325000
},
{
"epoch": 13.720861611094719,
"grad_norm": 0.9785659313201904,
"learning_rate": 4.589032285726536e-05,
"loss": 1.4278,
"step": 325500
},
{
"epoch": 13.74193820343127,
"grad_norm": 0.975077748298645,
"learning_rate": 4.5804570595549465e-05,
"loss": 1.4254,
"step": 326000
},
{
"epoch": 13.76301479576782,
"grad_norm": 0.9565373063087463,
"learning_rate": 4.5718818333833556e-05,
"loss": 1.4275,
"step": 326500
},
{
"epoch": 13.784091388104372,
"grad_norm": 0.9806529879570007,
"learning_rate": 4.5633066072117655e-05,
"loss": 1.4286,
"step": 327000
},
{
"epoch": 13.805167980440922,
"grad_norm": 0.9639936685562134,
"learning_rate": 4.554731381040175e-05,
"loss": 1.4211,
"step": 327500
},
{
"epoch": 13.826244572777473,
"grad_norm": 1.018277645111084,
"learning_rate": 4.546156154868585e-05,
"loss": 1.4246,
"step": 328000
},
{
"epoch": 13.847321165114025,
"grad_norm": 1.0246856212615967,
"learning_rate": 4.537580928696994e-05,
"loss": 1.4243,
"step": 328500
},
{
"epoch": 13.868397757450575,
"grad_norm": 1.0436735153198242,
"learning_rate": 4.529005702525405e-05,
"loss": 1.4235,
"step": 329000
},
{
"epoch": 13.889474349787127,
"grad_norm": 1.0222002267837524,
"learning_rate": 4.520430476353814e-05,
"loss": 1.4224,
"step": 329500
},
{
"epoch": 13.910550942123677,
"grad_norm": 0.9968163371086121,
"learning_rate": 4.511855250182224e-05,
"loss": 1.4227,
"step": 330000
},
{
"epoch": 13.910550942123677,
"eval_accuracy": 0.7299162820604502,
"eval_loss": 1.2043737173080444,
"eval_runtime": 187.6003,
"eval_samples_per_second": 643.421,
"eval_steps_per_second": 6.706,
"step": 330000
},
{
"epoch": 13.931627534460228,
"grad_norm": 0.9595262408256531,
"learning_rate": 4.5032800240106336e-05,
"loss": 1.4225,
"step": 330500
},
{
"epoch": 13.95270412679678,
"grad_norm": 0.9709582328796387,
"learning_rate": 4.4947047978390434e-05,
"loss": 1.425,
"step": 331000
},
{
"epoch": 13.97378071913333,
"grad_norm": 0.9848549962043762,
"learning_rate": 4.4861295716674526e-05,
"loss": 1.4233,
"step": 331500
},
{
"epoch": 13.994857311469882,
"grad_norm": 0.9817208647727966,
"learning_rate": 4.4775543454958624e-05,
"loss": 1.4235,
"step": 332000
},
{
"epoch": 14.015933903806433,
"grad_norm": 0.9909298419952393,
"learning_rate": 4.468979119324272e-05,
"loss": 1.4161,
"step": 332500
},
{
"epoch": 14.037010496142983,
"grad_norm": 0.9916654229164124,
"learning_rate": 4.460403893152682e-05,
"loss": 1.4162,
"step": 333000
},
{
"epoch": 14.058087088479535,
"grad_norm": 0.990878701210022,
"learning_rate": 4.451828666981092e-05,
"loss": 1.4156,
"step": 333500
},
{
"epoch": 14.079163680816086,
"grad_norm": 0.987936794757843,
"learning_rate": 4.443253440809502e-05,
"loss": 1.4154,
"step": 334000
},
{
"epoch": 14.100240273152636,
"grad_norm": 0.959426999092102,
"learning_rate": 4.4346782146379116e-05,
"loss": 1.4148,
"step": 334500
},
{
"epoch": 14.121316865489188,
"grad_norm": 1.000077724456787,
"learning_rate": 4.426102988466321e-05,
"loss": 1.4133,
"step": 335000
},
{
"epoch": 14.142393457825738,
"grad_norm": 0.9661691188812256,
"learning_rate": 4.4175277622947305e-05,
"loss": 1.4085,
"step": 335500
},
{
"epoch": 14.16347005016229,
"grad_norm": 1.0082225799560547,
"learning_rate": 4.4089525361231404e-05,
"loss": 1.4135,
"step": 336000
},
{
"epoch": 14.184546642498841,
"grad_norm": 0.9595646262168884,
"learning_rate": 4.40037730995155e-05,
"loss": 1.4088,
"step": 336500
},
{
"epoch": 14.205623234835391,
"grad_norm": 0.9916826486587524,
"learning_rate": 4.39180208377996e-05,
"loss": 1.4141,
"step": 337000
},
{
"epoch": 14.226699827171943,
"grad_norm": 0.9973074793815613,
"learning_rate": 4.38322685760837e-05,
"loss": 1.4142,
"step": 337500
},
{
"epoch": 14.247776419508494,
"grad_norm": 1.0315946340560913,
"learning_rate": 4.374651631436779e-05,
"loss": 1.4126,
"step": 338000
},
{
"epoch": 14.268853011845044,
"grad_norm": 0.978600263595581,
"learning_rate": 4.3660764052651895e-05,
"loss": 1.4131,
"step": 338500
},
{
"epoch": 14.289929604181596,
"grad_norm": 0.9818589091300964,
"learning_rate": 4.357501179093599e-05,
"loss": 1.4112,
"step": 339000
},
{
"epoch": 14.311006196518147,
"grad_norm": 1.0111372470855713,
"learning_rate": 4.3489259529220085e-05,
"loss": 1.4105,
"step": 339500
},
{
"epoch": 14.332082788854699,
"grad_norm": 1.0445629358291626,
"learning_rate": 4.3403507267504183e-05,
"loss": 1.414,
"step": 340000
},
{
"epoch": 14.332082788854699,
"eval_accuracy": 0.7319426499344681,
"eval_loss": 1.1946022510528564,
"eval_runtime": 188.1393,
"eval_samples_per_second": 641.578,
"eval_steps_per_second": 6.687,
"step": 340000
},
{
"epoch": 14.353159381191249,
"grad_norm": 1.0045791864395142,
"learning_rate": 4.331775500578828e-05,
"loss": 1.4097,
"step": 340500
},
{
"epoch": 14.3742359735278,
"grad_norm": 1.0097055435180664,
"learning_rate": 4.323200274407237e-05,
"loss": 1.409,
"step": 341000
},
{
"epoch": 14.395312565864351,
"grad_norm": 1.0181173086166382,
"learning_rate": 4.314625048235648e-05,
"loss": 1.4084,
"step": 341500
},
{
"epoch": 14.416389158200902,
"grad_norm": 1.0041073560714722,
"learning_rate": 4.306049822064057e-05,
"loss": 1.4123,
"step": 342000
},
{
"epoch": 14.437465750537454,
"grad_norm": 0.995884120464325,
"learning_rate": 4.297474595892467e-05,
"loss": 1.4102,
"step": 342500
},
{
"epoch": 14.458542342874004,
"grad_norm": 0.9988816380500793,
"learning_rate": 4.2888993697208767e-05,
"loss": 1.4061,
"step": 343000
},
{
"epoch": 14.479618935210555,
"grad_norm": 0.9720391035079956,
"learning_rate": 4.2803241435492865e-05,
"loss": 1.409,
"step": 343500
},
{
"epoch": 14.500695527547107,
"grad_norm": 1.0000778436660767,
"learning_rate": 4.2717489173776956e-05,
"loss": 1.408,
"step": 344000
},
{
"epoch": 14.521772119883657,
"grad_norm": 0.981829047203064,
"learning_rate": 4.263173691206106e-05,
"loss": 1.406,
"step": 344500
},
{
"epoch": 14.542848712220207,
"grad_norm": 0.9933907985687256,
"learning_rate": 4.254598465034515e-05,
"loss": 1.4071,
"step": 345000
},
{
"epoch": 14.56392530455676,
"grad_norm": 0.9865861535072327,
"learning_rate": 4.246023238862925e-05,
"loss": 1.406,
"step": 345500
},
{
"epoch": 14.58500189689331,
"grad_norm": 1.0130919218063354,
"learning_rate": 4.237448012691335e-05,
"loss": 1.4051,
"step": 346000
},
{
"epoch": 14.606078489229862,
"grad_norm": 1.0143005847930908,
"learning_rate": 4.228872786519745e-05,
"loss": 1.4041,
"step": 346500
},
{
"epoch": 14.627155081566412,
"grad_norm": 1.0110406875610352,
"learning_rate": 4.220297560348154e-05,
"loss": 1.4074,
"step": 347000
},
{
"epoch": 14.648231673902963,
"grad_norm": 1.0253915786743164,
"learning_rate": 4.2117223341765645e-05,
"loss": 1.4035,
"step": 347500
},
{
"epoch": 14.669308266239515,
"grad_norm": 1.0000860691070557,
"learning_rate": 4.2031471080049736e-05,
"loss": 1.4031,
"step": 348000
},
{
"epoch": 14.690384858576065,
"grad_norm": 1.0090601444244385,
"learning_rate": 4.1945718818333834e-05,
"loss": 1.4036,
"step": 348500
},
{
"epoch": 14.711461450912616,
"grad_norm": 0.9827486276626587,
"learning_rate": 4.185996655661793e-05,
"loss": 1.4042,
"step": 349000
},
{
"epoch": 14.732538043249168,
"grad_norm": 0.9991499781608582,
"learning_rate": 4.177421429490203e-05,
"loss": 1.406,
"step": 349500
},
{
"epoch": 14.753614635585718,
"grad_norm": 1.0147457122802734,
"learning_rate": 4.168846203318613e-05,
"loss": 1.4028,
"step": 350000
},
{
"epoch": 14.753614635585718,
"eval_accuracy": 0.7332691942927423,
"eval_loss": 1.185509204864502,
"eval_runtime": 187.7522,
"eval_samples_per_second": 642.901,
"eval_steps_per_second": 6.7,
"step": 350000
},
{
"epoch": 14.77469122792227,
"grad_norm": 1.00407075881958,
"learning_rate": 4.160270977147022e-05,
"loss": 1.4013,
"step": 350500
},
{
"epoch": 14.79576782025882,
"grad_norm": 0.9930475354194641,
"learning_rate": 4.1516957509754326e-05,
"loss": 1.405,
"step": 351000
},
{
"epoch": 14.81684441259537,
"grad_norm": 0.9812691807746887,
"learning_rate": 4.143120524803842e-05,
"loss": 1.399,
"step": 351500
},
{
"epoch": 14.837921004931923,
"grad_norm": 1.0133179426193237,
"learning_rate": 4.1345452986322516e-05,
"loss": 1.4007,
"step": 352000
},
{
"epoch": 14.858997597268473,
"grad_norm": 1.0065988302230835,
"learning_rate": 4.1259700724606614e-05,
"loss": 1.3995,
"step": 352500
},
{
"epoch": 14.880074189605025,
"grad_norm": 0.9896745085716248,
"learning_rate": 4.117394846289071e-05,
"loss": 1.3995,
"step": 353000
},
{
"epoch": 14.901150781941576,
"grad_norm": 0.9915310740470886,
"learning_rate": 4.1088196201174804e-05,
"loss": 1.4009,
"step": 353500
},
{
"epoch": 14.922227374278126,
"grad_norm": 0.9896999001502991,
"learning_rate": 4.100244393945891e-05,
"loss": 1.3966,
"step": 354000
},
{
"epoch": 14.943303966614678,
"grad_norm": 1.0219876766204834,
"learning_rate": 4.0916691677743e-05,
"loss": 1.4039,
"step": 354500
},
{
"epoch": 14.964380558951229,
"grad_norm": 0.9672915935516357,
"learning_rate": 4.08309394160271e-05,
"loss": 1.3987,
"step": 355000
},
{
"epoch": 14.985457151287779,
"grad_norm": 0.9991599321365356,
"learning_rate": 4.07451871543112e-05,
"loss": 1.3989,
"step": 355500
},
{
"epoch": 15.006533743624331,
"grad_norm": 0.9834728837013245,
"learning_rate": 4.0659434892595295e-05,
"loss": 1.3939,
"step": 356000
},
{
"epoch": 15.027610335960881,
"grad_norm": 1.0096129179000854,
"learning_rate": 4.057368263087939e-05,
"loss": 1.3905,
"step": 356500
},
{
"epoch": 15.048686928297434,
"grad_norm": 0.9921718239784241,
"learning_rate": 4.048793036916349e-05,
"loss": 1.3937,
"step": 357000
},
{
"epoch": 15.069763520633984,
"grad_norm": 1.0086896419525146,
"learning_rate": 4.0402178107447584e-05,
"loss": 1.3959,
"step": 357500
},
{
"epoch": 15.090840112970534,
"grad_norm": 1.0355802774429321,
"learning_rate": 4.031642584573168e-05,
"loss": 1.3935,
"step": 358000
},
{
"epoch": 15.111916705307086,
"grad_norm": 1.011924147605896,
"learning_rate": 4.023067358401578e-05,
"loss": 1.3906,
"step": 358500
},
{
"epoch": 15.132993297643637,
"grad_norm": 1.0120903253555298,
"learning_rate": 4.014492132229988e-05,
"loss": 1.3938,
"step": 359000
},
{
"epoch": 15.154069889980189,
"grad_norm": 0.9963767528533936,
"learning_rate": 4.005916906058397e-05,
"loss": 1.3913,
"step": 359500
},
{
"epoch": 15.17514648231674,
"grad_norm": 1.0020921230316162,
"learning_rate": 3.9973416798868075e-05,
"loss": 1.3929,
"step": 360000
},
{
"epoch": 15.17514648231674,
"eval_accuracy": 0.7344073547389167,
"eval_loss": 1.179434061050415,
"eval_runtime": 187.8248,
"eval_samples_per_second": 642.652,
"eval_steps_per_second": 6.698,
"step": 360000
},
{
"epoch": 15.19622307465329,
"grad_norm": 1.0280019044876099,
"learning_rate": 3.988766453715217e-05,
"loss": 1.3941,
"step": 360500
},
{
"epoch": 15.217299666989842,
"grad_norm": 1.0051079988479614,
"learning_rate": 3.9801912275436265e-05,
"loss": 1.3928,
"step": 361000
},
{
"epoch": 15.238376259326392,
"grad_norm": 0.9903327822685242,
"learning_rate": 3.971616001372036e-05,
"loss": 1.3925,
"step": 361500
},
{
"epoch": 15.259452851662942,
"grad_norm": 1.0269783735275269,
"learning_rate": 3.963040775200446e-05,
"loss": 1.3902,
"step": 362000
},
{
"epoch": 15.280529443999495,
"grad_norm": 1.0500859022140503,
"learning_rate": 3.954465549028856e-05,
"loss": 1.3895,
"step": 362500
},
{
"epoch": 15.301606036336045,
"grad_norm": 1.0062679052352905,
"learning_rate": 3.945890322857266e-05,
"loss": 1.3912,
"step": 363000
},
{
"epoch": 15.322682628672597,
"grad_norm": 0.9857174158096313,
"learning_rate": 3.9373150966856756e-05,
"loss": 1.3883,
"step": 363500
},
{
"epoch": 15.343759221009147,
"grad_norm": 1.0087082386016846,
"learning_rate": 3.928739870514085e-05,
"loss": 1.3898,
"step": 364000
},
{
"epoch": 15.364835813345698,
"grad_norm": 1.0169665813446045,
"learning_rate": 3.920164644342495e-05,
"loss": 1.388,
"step": 364500
},
{
"epoch": 15.38591240568225,
"grad_norm": 1.0175210237503052,
"learning_rate": 3.9115894181709045e-05,
"loss": 1.3885,
"step": 365000
},
{
"epoch": 15.4069889980188,
"grad_norm": 1.0041024684906006,
"learning_rate": 3.903014191999314e-05,
"loss": 1.3848,
"step": 365500
},
{
"epoch": 15.42806559035535,
"grad_norm": 1.040848731994629,
"learning_rate": 3.894438965827724e-05,
"loss": 1.3891,
"step": 366000
},
{
"epoch": 15.449142182691903,
"grad_norm": 0.9882203340530396,
"learning_rate": 3.885863739656134e-05,
"loss": 1.3876,
"step": 366500
},
{
"epoch": 15.470218775028453,
"grad_norm": 1.0042024850845337,
"learning_rate": 3.877288513484543e-05,
"loss": 1.386,
"step": 367000
},
{
"epoch": 15.491295367365005,
"grad_norm": 1.0103739500045776,
"learning_rate": 3.868713287312953e-05,
"loss": 1.3876,
"step": 367500
},
{
"epoch": 15.512371959701555,
"grad_norm": 1.0214111804962158,
"learning_rate": 3.860138061141363e-05,
"loss": 1.3871,
"step": 368000
},
{
"epoch": 15.533448552038106,
"grad_norm": 0.986108124256134,
"learning_rate": 3.8515628349697726e-05,
"loss": 1.3884,
"step": 368500
},
{
"epoch": 15.554525144374658,
"grad_norm": 1.0125185251235962,
"learning_rate": 3.842987608798182e-05,
"loss": 1.386,
"step": 369000
},
{
"epoch": 15.575601736711208,
"grad_norm": 1.0472725629806519,
"learning_rate": 3.834412382626592e-05,
"loss": 1.3827,
"step": 369500
},
{
"epoch": 15.596678329047759,
"grad_norm": 1.0223616361618042,
"learning_rate": 3.8258371564550014e-05,
"loss": 1.3863,
"step": 370000
},
{
"epoch": 15.596678329047759,
"eval_accuracy": 0.7359706295545975,
"eval_loss": 1.169610619544983,
"eval_runtime": 187.3888,
"eval_samples_per_second": 644.147,
"eval_steps_per_second": 6.713,
"step": 370000
},
{
"epoch": 15.61775492138431,
"grad_norm": 1.0223511457443237,
"learning_rate": 3.817261930283411e-05,
"loss": 1.3828,
"step": 370500
},
{
"epoch": 15.638831513720861,
"grad_norm": 1.0540897846221924,
"learning_rate": 3.808686704111821e-05,
"loss": 1.3843,
"step": 371000
},
{
"epoch": 15.659908106057413,
"grad_norm": 1.0288928747177124,
"learning_rate": 3.800111477940231e-05,
"loss": 1.3827,
"step": 371500
},
{
"epoch": 15.680984698393964,
"grad_norm": 1.011778712272644,
"learning_rate": 3.79153625176864e-05,
"loss": 1.3859,
"step": 372000
},
{
"epoch": 15.702061290730514,
"grad_norm": 0.9989762902259827,
"learning_rate": 3.7829610255970506e-05,
"loss": 1.3871,
"step": 372500
},
{
"epoch": 15.723137883067066,
"grad_norm": 1.007678508758545,
"learning_rate": 3.77438579942546e-05,
"loss": 1.3808,
"step": 373000
},
{
"epoch": 15.744214475403616,
"grad_norm": 1.0293598175048828,
"learning_rate": 3.7658105732538696e-05,
"loss": 1.3787,
"step": 373500
},
{
"epoch": 15.765291067740169,
"grad_norm": 0.9732978940010071,
"learning_rate": 3.7572353470822794e-05,
"loss": 1.3827,
"step": 374000
},
{
"epoch": 15.786367660076719,
"grad_norm": 1.0195307731628418,
"learning_rate": 3.748660120910689e-05,
"loss": 1.3786,
"step": 374500
},
{
"epoch": 15.80744425241327,
"grad_norm": 1.0174545049667358,
"learning_rate": 3.740084894739099e-05,
"loss": 1.3765,
"step": 375000
},
{
"epoch": 15.828520844749821,
"grad_norm": 1.0052070617675781,
"learning_rate": 3.731509668567509e-05,
"loss": 1.3783,
"step": 375500
},
{
"epoch": 15.849597437086372,
"grad_norm": 1.0047810077667236,
"learning_rate": 3.722934442395919e-05,
"loss": 1.3816,
"step": 376000
},
{
"epoch": 15.870674029422922,
"grad_norm": 0.9908145666122437,
"learning_rate": 3.714359216224328e-05,
"loss": 1.3807,
"step": 376500
},
{
"epoch": 15.891750621759474,
"grad_norm": 1.011812686920166,
"learning_rate": 3.705783990052738e-05,
"loss": 1.3815,
"step": 377000
},
{
"epoch": 15.912827214096025,
"grad_norm": 1.0190160274505615,
"learning_rate": 3.6972087638811475e-05,
"loss": 1.3774,
"step": 377500
},
{
"epoch": 15.933903806432577,
"grad_norm": 0.9915936589241028,
"learning_rate": 3.6886335377095574e-05,
"loss": 1.3799,
"step": 378000
},
{
"epoch": 15.954980398769127,
"grad_norm": 0.976587176322937,
"learning_rate": 3.680058311537967e-05,
"loss": 1.3788,
"step": 378500
},
{
"epoch": 15.976056991105677,
"grad_norm": 1.0209779739379883,
"learning_rate": 3.671483085366377e-05,
"loss": 1.3807,
"step": 379000
},
{
"epoch": 15.99713358344223,
"grad_norm": 0.9853180646896362,
"learning_rate": 3.662907859194786e-05,
"loss": 1.376,
"step": 379500
},
{
"epoch": 16.01821017577878,
"grad_norm": 1.0516399145126343,
"learning_rate": 3.654332633023197e-05,
"loss": 1.3762,
"step": 380000
},
{
"epoch": 16.01821017577878,
"eval_accuracy": 0.7372482864771231,
"eval_loss": 1.1627013683319092,
"eval_runtime": 191.2878,
"eval_samples_per_second": 631.018,
"eval_steps_per_second": 6.576,
"step": 380000
},
{
"epoch": 16.039286768115332,
"grad_norm": 0.9836915135383606,
"learning_rate": 3.645757406851606e-05,
"loss": 1.3738,
"step": 380500
},
{
"epoch": 16.06036336045188,
"grad_norm": 1.0355842113494873,
"learning_rate": 3.6371821806800157e-05,
"loss": 1.3728,
"step": 381000
},
{
"epoch": 16.081439952788433,
"grad_norm": 0.9951421618461609,
"learning_rate": 3.6286069545084255e-05,
"loss": 1.377,
"step": 381500
},
{
"epoch": 16.102516545124985,
"grad_norm": 1.0165185928344727,
"learning_rate": 3.620031728336835e-05,
"loss": 1.3746,
"step": 382000
},
{
"epoch": 16.123593137461537,
"grad_norm": 1.0084277391433716,
"learning_rate": 3.6114565021652445e-05,
"loss": 1.368,
"step": 382500
},
{
"epoch": 16.144669729798085,
"grad_norm": 1.0118244886398315,
"learning_rate": 3.602881275993655e-05,
"loss": 1.3707,
"step": 383000
},
{
"epoch": 16.165746322134638,
"grad_norm": 1.018411636352539,
"learning_rate": 3.594306049822064e-05,
"loss": 1.3759,
"step": 383500
},
{
"epoch": 16.18682291447119,
"grad_norm": 1.0341806411743164,
"learning_rate": 3.585730823650474e-05,
"loss": 1.371,
"step": 384000
},
{
"epoch": 16.20789950680774,
"grad_norm": 0.953336238861084,
"learning_rate": 3.577155597478883e-05,
"loss": 1.3719,
"step": 384500
},
{
"epoch": 16.22897609914429,
"grad_norm": 0.9999815821647644,
"learning_rate": 3.5685803713072936e-05,
"loss": 1.3673,
"step": 385000
},
{
"epoch": 16.250052691480843,
"grad_norm": 1.0058740377426147,
"learning_rate": 3.560005145135703e-05,
"loss": 1.3728,
"step": 385500
},
{
"epoch": 16.27112928381739,
"grad_norm": 1.012632131576538,
"learning_rate": 3.5514299189641126e-05,
"loss": 1.3855,
"step": 386000
},
{
"epoch": 16.292205876153943,
"grad_norm": 1.0225788354873657,
"learning_rate": 3.5428546927925224e-05,
"loss": 1.3735,
"step": 386500
},
{
"epoch": 16.313282468490495,
"grad_norm": 1.0026121139526367,
"learning_rate": 3.534279466620932e-05,
"loss": 1.3683,
"step": 387000
},
{
"epoch": 16.334359060827044,
"grad_norm": 1.014241099357605,
"learning_rate": 3.525704240449342e-05,
"loss": 1.369,
"step": 387500
},
{
"epoch": 16.355435653163596,
"grad_norm": 1.019674301147461,
"learning_rate": 3.517129014277752e-05,
"loss": 1.3686,
"step": 388000
},
{
"epoch": 16.376512245500148,
"grad_norm": 1.0139714479446411,
"learning_rate": 3.508553788106161e-05,
"loss": 1.3683,
"step": 388500
},
{
"epoch": 16.3975888378367,
"grad_norm": 1.02431321144104,
"learning_rate": 3.499978561934571e-05,
"loss": 1.3668,
"step": 389000
},
{
"epoch": 16.41866543017325,
"grad_norm": 0.9997479915618896,
"learning_rate": 3.491403335762981e-05,
"loss": 1.3675,
"step": 389500
},
{
"epoch": 16.4397420225098,
"grad_norm": 1.0277302265167236,
"learning_rate": 3.4828281095913906e-05,
"loss": 1.3697,
"step": 390000
},
{
"epoch": 16.4397420225098,
"eval_accuracy": 0.7386636222407444,
"eval_loss": 1.1562188863754272,
"eval_runtime": 192.9524,
"eval_samples_per_second": 625.574,
"eval_steps_per_second": 6.52,
"step": 390000
},
{
"epoch": 16.460818614846353,
"grad_norm": 1.0323355197906494,
"learning_rate": 3.4742528834198004e-05,
"loss": 1.3663,
"step": 390500
},
{
"epoch": 16.4818952071829,
"grad_norm": 0.9885368347167969,
"learning_rate": 3.46567765724821e-05,
"loss": 1.3701,
"step": 391000
},
{
"epoch": 16.502971799519454,
"grad_norm": 1.043430209159851,
"learning_rate": 3.45710243107662e-05,
"loss": 1.3671,
"step": 391500
},
{
"epoch": 16.524048391856006,
"grad_norm": 1.021682620048523,
"learning_rate": 3.448527204905029e-05,
"loss": 1.3704,
"step": 392000
},
{
"epoch": 16.545124984192555,
"grad_norm": 1.007081389427185,
"learning_rate": 3.43995197873344e-05,
"loss": 1.3644,
"step": 392500
},
{
"epoch": 16.566201576529107,
"grad_norm": 1.0181798934936523,
"learning_rate": 3.431376752561849e-05,
"loss": 1.3669,
"step": 393000
},
{
"epoch": 16.58727816886566,
"grad_norm": 0.9851253628730774,
"learning_rate": 3.422801526390259e-05,
"loss": 1.3691,
"step": 393500
},
{
"epoch": 16.608354761202207,
"grad_norm": 1.0028140544891357,
"learning_rate": 3.4142263002186686e-05,
"loss": 1.3675,
"step": 394000
},
{
"epoch": 16.62943135353876,
"grad_norm": 1.0049411058425903,
"learning_rate": 3.4056510740470784e-05,
"loss": 1.3707,
"step": 394500
},
{
"epoch": 16.65050794587531,
"grad_norm": 1.015480875968933,
"learning_rate": 3.3970758478754875e-05,
"loss": 1.3673,
"step": 395000
},
{
"epoch": 16.67158453821186,
"grad_norm": 1.013917088508606,
"learning_rate": 3.388500621703898e-05,
"loss": 1.3668,
"step": 395500
},
{
"epoch": 16.692661130548412,
"grad_norm": 1.0468978881835938,
"learning_rate": 3.379925395532307e-05,
"loss": 1.3665,
"step": 396000
},
{
"epoch": 16.713737722884964,
"grad_norm": 0.950524091720581,
"learning_rate": 3.371350169360717e-05,
"loss": 1.363,
"step": 396500
},
{
"epoch": 16.734814315221517,
"grad_norm": 1.015105128288269,
"learning_rate": 3.362774943189127e-05,
"loss": 1.3676,
"step": 397000
},
{
"epoch": 16.755890907558065,
"grad_norm": 0.9856720566749573,
"learning_rate": 3.354199717017537e-05,
"loss": 1.3639,
"step": 397500
},
{
"epoch": 16.776967499894617,
"grad_norm": 1.0424240827560425,
"learning_rate": 3.345624490845946e-05,
"loss": 1.3638,
"step": 398000
},
{
"epoch": 16.79804409223117,
"grad_norm": 1.0016916990280151,
"learning_rate": 3.3370492646743564e-05,
"loss": 1.3618,
"step": 398500
},
{
"epoch": 16.819120684567718,
"grad_norm": 0.9998739361763,
"learning_rate": 3.3284740385027655e-05,
"loss": 1.3669,
"step": 399000
},
{
"epoch": 16.84019727690427,
"grad_norm": 0.9806528091430664,
"learning_rate": 3.319898812331175e-05,
"loss": 1.3604,
"step": 399500
},
{
"epoch": 16.861273869240822,
"grad_norm": 1.0413306951522827,
"learning_rate": 3.311323586159585e-05,
"loss": 1.36,
"step": 400000
},
{
"epoch": 16.861273869240822,
"eval_accuracy": 0.7394921877914268,
"eval_loss": 1.1513326168060303,
"eval_runtime": 187.1236,
"eval_samples_per_second": 645.06,
"eval_steps_per_second": 6.723,
"step": 400000
},
{
"epoch": 16.88235046157737,
"grad_norm": 1.036800742149353,
"learning_rate": 3.302748359987995e-05,
"loss": 1.3586,
"step": 400500
},
{
"epoch": 16.903427053913923,
"grad_norm": 1.0093916654586792,
"learning_rate": 3.294173133816404e-05,
"loss": 1.3619,
"step": 401000
},
{
"epoch": 16.924503646250475,
"grad_norm": 1.0519369840621948,
"learning_rate": 3.2855979076448147e-05,
"loss": 1.3648,
"step": 401500
},
{
"epoch": 16.945580238587024,
"grad_norm": 1.0250507593154907,
"learning_rate": 3.277022681473224e-05,
"loss": 1.3617,
"step": 402000
},
{
"epoch": 16.966656830923576,
"grad_norm": 1.0354464054107666,
"learning_rate": 3.2684474553016336e-05,
"loss": 1.3602,
"step": 402500
},
{
"epoch": 16.987733423260128,
"grad_norm": 1.0398974418640137,
"learning_rate": 3.2598722291300435e-05,
"loss": 1.3623,
"step": 403000
},
{
"epoch": 17.00881001559668,
"grad_norm": 1.0283706188201904,
"learning_rate": 3.251297002958453e-05,
"loss": 1.3585,
"step": 403500
},
{
"epoch": 17.02988660793323,
"grad_norm": 1.0299946069717407,
"learning_rate": 3.242721776786863e-05,
"loss": 1.3589,
"step": 404000
},
{
"epoch": 17.05096320026978,
"grad_norm": 1.00074303150177,
"learning_rate": 3.234146550615272e-05,
"loss": 1.3561,
"step": 404500
},
{
"epoch": 17.072039792606333,
"grad_norm": 1.0427615642547607,
"learning_rate": 3.225571324443683e-05,
"loss": 1.3571,
"step": 405000
},
{
"epoch": 17.09311638494288,
"grad_norm": 1.038236379623413,
"learning_rate": 3.216996098272092e-05,
"loss": 1.3566,
"step": 405500
},
{
"epoch": 17.114192977279433,
"grad_norm": 1.0303802490234375,
"learning_rate": 3.208420872100502e-05,
"loss": 1.3575,
"step": 406000
},
{
"epoch": 17.135269569615986,
"grad_norm": 1.016390323638916,
"learning_rate": 3.1998456459289116e-05,
"loss": 1.3521,
"step": 406500
},
{
"epoch": 17.156346161952534,
"grad_norm": 1.0257891416549683,
"learning_rate": 3.1912704197573214e-05,
"loss": 1.3531,
"step": 407000
},
{
"epoch": 17.177422754289086,
"grad_norm": 1.0196011066436768,
"learning_rate": 3.1826951935857306e-05,
"loss": 1.3524,
"step": 407500
},
{
"epoch": 17.19849934662564,
"grad_norm": 1.0268886089324951,
"learning_rate": 3.174119967414141e-05,
"loss": 1.3571,
"step": 408000
},
{
"epoch": 17.219575938962187,
"grad_norm": 1.0497263669967651,
"learning_rate": 3.16554474124255e-05,
"loss": 1.3547,
"step": 408500
},
{
"epoch": 17.24065253129874,
"grad_norm": 1.0018360614776611,
"learning_rate": 3.15696951507096e-05,
"loss": 1.3568,
"step": 409000
},
{
"epoch": 17.26172912363529,
"grad_norm": 1.026289701461792,
"learning_rate": 3.14839428889937e-05,
"loss": 1.3531,
"step": 409500
},
{
"epoch": 17.282805715971843,
"grad_norm": 1.0473647117614746,
"learning_rate": 3.13981906272778e-05,
"loss": 1.3566,
"step": 410000
},
{
"epoch": 17.282805715971843,
"eval_accuracy": 0.741055892936262,
"eval_loss": 1.1424680948257446,
"eval_runtime": 189.428,
"eval_samples_per_second": 637.213,
"eval_steps_per_second": 6.641,
"step": 410000
},
{
"epoch": 17.303882308308392,
"grad_norm": 1.0567522048950195,
"learning_rate": 3.131243836556189e-05,
"loss": 1.3576,
"step": 410500
},
{
"epoch": 17.324958900644944,
"grad_norm": 1.0679912567138672,
"learning_rate": 3.1226686103845994e-05,
"loss": 1.3511,
"step": 411000
},
{
"epoch": 17.346035492981496,
"grad_norm": 1.037283182144165,
"learning_rate": 3.1140933842130086e-05,
"loss": 1.3527,
"step": 411500
},
{
"epoch": 17.367112085318045,
"grad_norm": 1.0383124351501465,
"learning_rate": 3.1055181580414184e-05,
"loss": 1.355,
"step": 412000
},
{
"epoch": 17.388188677654597,
"grad_norm": 1.0149115324020386,
"learning_rate": 3.096942931869828e-05,
"loss": 1.3561,
"step": 412500
},
{
"epoch": 17.40926526999115,
"grad_norm": 0.9910415410995483,
"learning_rate": 3.088367705698238e-05,
"loss": 1.3506,
"step": 413000
},
{
"epoch": 17.430341862327698,
"grad_norm": 0.9862610697746277,
"learning_rate": 3.079792479526647e-05,
"loss": 1.3517,
"step": 413500
},
{
"epoch": 17.45141845466425,
"grad_norm": 1.0038148164749146,
"learning_rate": 3.071217253355058e-05,
"loss": 1.3509,
"step": 414000
},
{
"epoch": 17.472495047000802,
"grad_norm": 1.024957537651062,
"learning_rate": 3.062642027183467e-05,
"loss": 1.3501,
"step": 414500
},
{
"epoch": 17.49357163933735,
"grad_norm": 1.0283193588256836,
"learning_rate": 3.054066801011877e-05,
"loss": 1.3509,
"step": 415000
},
{
"epoch": 17.514648231673903,
"grad_norm": 1.073493242263794,
"learning_rate": 3.045491574840287e-05,
"loss": 1.3526,
"step": 415500
},
{
"epoch": 17.535724824010455,
"grad_norm": 1.0310298204421997,
"learning_rate": 3.0369163486686964e-05,
"loss": 1.3499,
"step": 416000
},
{
"epoch": 17.556801416347007,
"grad_norm": 1.040345549583435,
"learning_rate": 3.028341122497106e-05,
"loss": 1.3486,
"step": 416500
},
{
"epoch": 17.577878008683555,
"grad_norm": 1.0382001399993896,
"learning_rate": 3.019765896325516e-05,
"loss": 1.3532,
"step": 417000
},
{
"epoch": 17.598954601020107,
"grad_norm": 1.0507255792617798,
"learning_rate": 3.0111906701539255e-05,
"loss": 1.3512,
"step": 417500
},
{
"epoch": 17.62003119335666,
"grad_norm": 1.0532575845718384,
"learning_rate": 3.002615443982335e-05,
"loss": 1.3488,
"step": 418000
},
{
"epoch": 17.641107785693208,
"grad_norm": 1.0296951532363892,
"learning_rate": 2.9940402178107452e-05,
"loss": 1.3485,
"step": 418500
},
{
"epoch": 17.66218437802976,
"grad_norm": 1.0440666675567627,
"learning_rate": 2.9854649916391547e-05,
"loss": 1.3472,
"step": 419000
},
{
"epoch": 17.683260970366312,
"grad_norm": 1.026249647140503,
"learning_rate": 2.976889765467564e-05,
"loss": 1.3527,
"step": 419500
},
{
"epoch": 17.70433756270286,
"grad_norm": 1.0334538221359253,
"learning_rate": 2.9683145392959737e-05,
"loss": 1.3482,
"step": 420000
},
{
"epoch": 17.70433756270286,
"eval_accuracy": 0.7417368963431007,
"eval_loss": 1.138806700706482,
"eval_runtime": 186.5181,
"eval_samples_per_second": 647.154,
"eval_steps_per_second": 6.745,
"step": 420000
},
{
"epoch": 17.725414155039413,
"grad_norm": 1.0414074659347534,
"learning_rate": 2.9597393131243838e-05,
"loss": 1.3493,
"step": 420500
},
{
"epoch": 17.746490747375965,
"grad_norm": 1.0584443807601929,
"learning_rate": 2.9511640869527933e-05,
"loss": 1.3452,
"step": 421000
},
{
"epoch": 17.767567339712514,
"grad_norm": 1.0507086515426636,
"learning_rate": 2.942588860781203e-05,
"loss": 1.3461,
"step": 421500
},
{
"epoch": 17.788643932049066,
"grad_norm": 1.0374517440795898,
"learning_rate": 2.934013634609613e-05,
"loss": 1.3483,
"step": 422000
},
{
"epoch": 17.809720524385618,
"grad_norm": 1.010377287864685,
"learning_rate": 2.9254384084380225e-05,
"loss": 1.3492,
"step": 422500
},
{
"epoch": 17.830797116722167,
"grad_norm": 1.040511131286621,
"learning_rate": 2.9168631822664323e-05,
"loss": 1.3478,
"step": 423000
},
{
"epoch": 17.85187370905872,
"grad_norm": 1.00983726978302,
"learning_rate": 2.908287956094842e-05,
"loss": 1.3476,
"step": 423500
},
{
"epoch": 17.87295030139527,
"grad_norm": 1.0438557863235474,
"learning_rate": 2.899712729923252e-05,
"loss": 1.3482,
"step": 424000
},
{
"epoch": 17.894026893731823,
"grad_norm": 1.0267603397369385,
"learning_rate": 2.8911375037516615e-05,
"loss": 1.3469,
"step": 424500
},
{
"epoch": 17.91510348606837,
"grad_norm": 1.009652018547058,
"learning_rate": 2.8825622775800716e-05,
"loss": 1.3486,
"step": 425000
},
{
"epoch": 17.936180078404924,
"grad_norm": 1.0519484281539917,
"learning_rate": 2.873987051408481e-05,
"loss": 1.3464,
"step": 425500
},
{
"epoch": 17.957256670741476,
"grad_norm": 1.0304384231567383,
"learning_rate": 2.8654118252368906e-05,
"loss": 1.3446,
"step": 426000
},
{
"epoch": 17.978333263078024,
"grad_norm": 1.0165801048278809,
"learning_rate": 2.8568365990653008e-05,
"loss": 1.3432,
"step": 426500
},
{
"epoch": 17.999409855414576,
"grad_norm": 1.026118278503418,
"learning_rate": 2.8482613728937103e-05,
"loss": 1.3479,
"step": 427000
},
{
"epoch": 18.02048644775113,
"grad_norm": 1.0402462482452393,
"learning_rate": 2.8396861467221198e-05,
"loss": 1.3419,
"step": 427500
},
{
"epoch": 18.041563040087677,
"grad_norm": 1.054831624031067,
"learning_rate": 2.83111092055053e-05,
"loss": 1.3399,
"step": 428000
},
{
"epoch": 18.06263963242423,
"grad_norm": 0.997891902923584,
"learning_rate": 2.8225356943789394e-05,
"loss": 1.3423,
"step": 428500
},
{
"epoch": 18.08371622476078,
"grad_norm": 1.0122929811477661,
"learning_rate": 2.813960468207349e-05,
"loss": 1.3396,
"step": 429000
},
{
"epoch": 18.10479281709733,
"grad_norm": 1.0179975032806396,
"learning_rate": 2.805385242035759e-05,
"loss": 1.3403,
"step": 429500
},
{
"epoch": 18.125869409433882,
"grad_norm": 1.073257327079773,
"learning_rate": 2.7968100158641686e-05,
"loss": 1.3398,
"step": 430000
},
{
"epoch": 18.125869409433882,
"eval_accuracy": 0.7429558403069044,
"eval_loss": 1.133116364479065,
"eval_runtime": 190.4734,
"eval_samples_per_second": 633.716,
"eval_steps_per_second": 6.605,
"step": 430000
},
{
"epoch": 18.146946001770434,
"grad_norm": 1.0547127723693848,
"learning_rate": 2.788234789692578e-05,
"loss": 1.3411,
"step": 430500
},
{
"epoch": 18.168022594106986,
"grad_norm": 1.065807580947876,
"learning_rate": 2.7796595635209882e-05,
"loss": 1.3404,
"step": 431000
},
{
"epoch": 18.189099186443535,
"grad_norm": 1.025649905204773,
"learning_rate": 2.7710843373493977e-05,
"loss": 1.343,
"step": 431500
},
{
"epoch": 18.210175778780087,
"grad_norm": 1.0318825244903564,
"learning_rate": 2.7625091111778072e-05,
"loss": 1.3433,
"step": 432000
},
{
"epoch": 18.23125237111664,
"grad_norm": 1.0540595054626465,
"learning_rate": 2.7539338850062174e-05,
"loss": 1.34,
"step": 432500
},
{
"epoch": 18.252328963453188,
"grad_norm": 1.0002199411392212,
"learning_rate": 2.745358658834627e-05,
"loss": 1.3413,
"step": 433000
},
{
"epoch": 18.27340555578974,
"grad_norm": 1.0451005697250366,
"learning_rate": 2.7367834326630364e-05,
"loss": 1.3379,
"step": 433500
},
{
"epoch": 18.294482148126292,
"grad_norm": 1.0272574424743652,
"learning_rate": 2.7282082064914465e-05,
"loss": 1.3387,
"step": 434000
},
{
"epoch": 18.31555874046284,
"grad_norm": 1.0345803499221802,
"learning_rate": 2.719632980319856e-05,
"loss": 1.3377,
"step": 434500
},
{
"epoch": 18.336635332799393,
"grad_norm": 1.0111515522003174,
"learning_rate": 2.7110577541482655e-05,
"loss": 1.3373,
"step": 435000
},
{
"epoch": 18.357711925135945,
"grad_norm": 1.0496395826339722,
"learning_rate": 2.7024825279766757e-05,
"loss": 1.3399,
"step": 435500
},
{
"epoch": 18.378788517472493,
"grad_norm": 1.0501689910888672,
"learning_rate": 2.6939073018050852e-05,
"loss": 1.3389,
"step": 436000
},
{
"epoch": 18.399865109809046,
"grad_norm": 1.027983546257019,
"learning_rate": 2.685332075633495e-05,
"loss": 1.3392,
"step": 436500
},
{
"epoch": 18.420941702145598,
"grad_norm": 1.0174145698547363,
"learning_rate": 2.6767568494619045e-05,
"loss": 1.3378,
"step": 437000
},
{
"epoch": 18.44201829448215,
"grad_norm": 1.0536024570465088,
"learning_rate": 2.6681816232903143e-05,
"loss": 1.3358,
"step": 437500
},
{
"epoch": 18.4630948868187,
"grad_norm": 1.0644031763076782,
"learning_rate": 2.6596063971187242e-05,
"loss": 1.3333,
"step": 438000
},
{
"epoch": 18.48417147915525,
"grad_norm": 1.1041563749313354,
"learning_rate": 2.6510311709471337e-05,
"loss": 1.3339,
"step": 438500
},
{
"epoch": 18.505248071491803,
"grad_norm": 1.0172464847564697,
"learning_rate": 2.642455944775544e-05,
"loss": 1.336,
"step": 439000
},
{
"epoch": 18.52632466382835,
"grad_norm": 1.038860559463501,
"learning_rate": 2.6338807186039533e-05,
"loss": 1.3353,
"step": 439500
},
{
"epoch": 18.547401256164903,
"grad_norm": 1.0317773818969727,
"learning_rate": 2.6253054924323628e-05,
"loss": 1.3332,
"step": 440000
},
{
"epoch": 18.547401256164903,
"eval_accuracy": 0.7435791374405416,
"eval_loss": 1.1294691562652588,
"eval_runtime": 181.273,
"eval_samples_per_second": 665.88,
"eval_steps_per_second": 6.94,
"step": 440000
},
{
"epoch": 18.568477848501455,
"grad_norm": 1.0559191703796387,
"learning_rate": 2.616730266260773e-05,
"loss": 1.3335,
"step": 440500
},
{
"epoch": 18.589554440838004,
"grad_norm": 1.0526764392852783,
"learning_rate": 2.6081550400891825e-05,
"loss": 1.3365,
"step": 441000
},
{
"epoch": 18.610631033174556,
"grad_norm": 1.0439352989196777,
"learning_rate": 2.599579813917592e-05,
"loss": 1.3365,
"step": 441500
},
{
"epoch": 18.63170762551111,
"grad_norm": 1.056220293045044,
"learning_rate": 2.591004587746002e-05,
"loss": 1.3345,
"step": 442000
},
{
"epoch": 18.652784217847657,
"grad_norm": 1.0541622638702393,
"learning_rate": 2.5824293615744116e-05,
"loss": 1.3348,
"step": 442500
},
{
"epoch": 18.67386081018421,
"grad_norm": 1.0538071393966675,
"learning_rate": 2.573854135402821e-05,
"loss": 1.3362,
"step": 443000
},
{
"epoch": 18.69493740252076,
"grad_norm": 1.0437824726104736,
"learning_rate": 2.5652789092312313e-05,
"loss": 1.3334,
"step": 443500
},
{
"epoch": 18.716013994857313,
"grad_norm": 1.0523426532745361,
"learning_rate": 2.5567036830596408e-05,
"loss": 1.3327,
"step": 444000
},
{
"epoch": 18.737090587193862,
"grad_norm": 1.0338430404663086,
"learning_rate": 2.5481284568880503e-05,
"loss": 1.3347,
"step": 444500
},
{
"epoch": 18.758167179530414,
"grad_norm": 1.049448847770691,
"learning_rate": 2.5395532307164605e-05,
"loss": 1.3296,
"step": 445000
},
{
"epoch": 18.779243771866966,
"grad_norm": 1.0741432905197144,
"learning_rate": 2.53097800454487e-05,
"loss": 1.3369,
"step": 445500
},
{
"epoch": 18.800320364203515,
"grad_norm": 1.0242787599563599,
"learning_rate": 2.5224027783732794e-05,
"loss": 1.335,
"step": 446000
},
{
"epoch": 18.821396956540067,
"grad_norm": 1.0983304977416992,
"learning_rate": 2.5138275522016896e-05,
"loss": 1.327,
"step": 446500
},
{
"epoch": 18.84247354887662,
"grad_norm": 1.0492666959762573,
"learning_rate": 2.505252326030099e-05,
"loss": 1.3296,
"step": 447000
},
{
"epoch": 18.863550141213167,
"grad_norm": 1.0496268272399902,
"learning_rate": 2.496677099858509e-05,
"loss": 1.3306,
"step": 447500
},
{
"epoch": 18.88462673354972,
"grad_norm": 1.009260654449463,
"learning_rate": 2.4881018736869184e-05,
"loss": 1.3337,
"step": 448000
},
{
"epoch": 18.90570332588627,
"grad_norm": 1.05097496509552,
"learning_rate": 2.4795266475153283e-05,
"loss": 1.3301,
"step": 448500
},
{
"epoch": 18.92677991822282,
"grad_norm": 1.0488065481185913,
"learning_rate": 2.470951421343738e-05,
"loss": 1.3346,
"step": 449000
},
{
"epoch": 18.947856510559372,
"grad_norm": 1.0175256729125977,
"learning_rate": 2.4623761951721476e-05,
"loss": 1.329,
"step": 449500
},
{
"epoch": 18.968933102895924,
"grad_norm": 1.0717912912368774,
"learning_rate": 2.4538009690005574e-05,
"loss": 1.3316,
"step": 450000
},
{
"epoch": 18.968933102895924,
"eval_accuracy": 0.744795795556412,
"eval_loss": 1.1221095323562622,
"eval_runtime": 180.7259,
"eval_samples_per_second": 667.895,
"eval_steps_per_second": 6.961,
"step": 450000
},
{
"epoch": 18.990009695232473,
"grad_norm": 1.0513113737106323,
"learning_rate": 2.4452257428289672e-05,
"loss": 1.3287,
"step": 450500
},
{
"epoch": 19.011086287569025,
"grad_norm": 1.0699896812438965,
"learning_rate": 2.436650516657377e-05,
"loss": 1.3278,
"step": 451000
},
{
"epoch": 19.032162879905577,
"grad_norm": 1.0617879629135132,
"learning_rate": 2.428075290485787e-05,
"loss": 1.3266,
"step": 451500
},
{
"epoch": 19.05323947224213,
"grad_norm": 1.0463162660598755,
"learning_rate": 2.4195000643141964e-05,
"loss": 1.3276,
"step": 452000
},
{
"epoch": 19.074316064578678,
"grad_norm": 1.0622825622558594,
"learning_rate": 2.4109248381426062e-05,
"loss": 1.3296,
"step": 452500
},
{
"epoch": 19.09539265691523,
"grad_norm": 1.020668625831604,
"learning_rate": 2.402349611971016e-05,
"loss": 1.3275,
"step": 453000
},
{
"epoch": 19.116469249251782,
"grad_norm": 0.9923982620239258,
"learning_rate": 2.3937743857994255e-05,
"loss": 1.3248,
"step": 453500
},
{
"epoch": 19.13754584158833,
"grad_norm": 1.0654940605163574,
"learning_rate": 2.3851991596278354e-05,
"loss": 1.3282,
"step": 454000
},
{
"epoch": 19.158622433924883,
"grad_norm": 1.0939205884933472,
"learning_rate": 2.3766239334562452e-05,
"loss": 1.325,
"step": 454500
},
{
"epoch": 19.179699026261435,
"grad_norm": 1.0714713335037231,
"learning_rate": 2.3680487072846547e-05,
"loss": 1.3234,
"step": 455000
},
{
"epoch": 19.200775618597984,
"grad_norm": 1.0567889213562012,
"learning_rate": 2.3594734811130645e-05,
"loss": 1.3232,
"step": 455500
},
{
"epoch": 19.221852210934536,
"grad_norm": 1.0443955659866333,
"learning_rate": 2.3508982549414744e-05,
"loss": 1.3232,
"step": 456000
},
{
"epoch": 19.242928803271088,
"grad_norm": 1.0499762296676636,
"learning_rate": 2.342323028769884e-05,
"loss": 1.3261,
"step": 456500
},
{
"epoch": 19.264005395607636,
"grad_norm": 1.0240346193313599,
"learning_rate": 2.3337478025982937e-05,
"loss": 1.3232,
"step": 457000
},
{
"epoch": 19.28508198794419,
"grad_norm": 1.074677586555481,
"learning_rate": 2.3251725764267035e-05,
"loss": 1.3248,
"step": 457500
},
{
"epoch": 19.30615858028074,
"grad_norm": 1.0480748414993286,
"learning_rate": 2.316597350255113e-05,
"loss": 1.3269,
"step": 458000
},
{
"epoch": 19.327235172617293,
"grad_norm": 1.0506702661514282,
"learning_rate": 2.308022124083523e-05,
"loss": 1.3248,
"step": 458500
},
{
"epoch": 19.34831176495384,
"grad_norm": 1.074256420135498,
"learning_rate": 2.2994468979119323e-05,
"loss": 1.3258,
"step": 459000
},
{
"epoch": 19.369388357290394,
"grad_norm": 1.058158040046692,
"learning_rate": 2.290871671740342e-05,
"loss": 1.3236,
"step": 459500
},
{
"epoch": 19.390464949626946,
"grad_norm": 1.0356804132461548,
"learning_rate": 2.282296445568752e-05,
"loss": 1.3235,
"step": 460000
},
{
"epoch": 19.390464949626946,
"eval_accuracy": 0.745667395627653,
"eval_loss": 1.117651104927063,
"eval_runtime": 180.1282,
"eval_samples_per_second": 670.112,
"eval_steps_per_second": 6.984,
"step": 460000
},
{
"epoch": 19.411541541963494,
"grad_norm": 1.0651278495788574,
"learning_rate": 2.2737212193971615e-05,
"loss": 1.3221,
"step": 460500
},
{
"epoch": 19.432618134300046,
"grad_norm": 1.071193814277649,
"learning_rate": 2.2651459932255713e-05,
"loss": 1.327,
"step": 461000
},
{
"epoch": 19.4536947266366,
"grad_norm": 1.030908465385437,
"learning_rate": 2.256570767053981e-05,
"loss": 1.3219,
"step": 461500
},
{
"epoch": 19.474771318973147,
"grad_norm": 1.0321614742279053,
"learning_rate": 2.2479955408823906e-05,
"loss": 1.3211,
"step": 462000
},
{
"epoch": 19.4958479113097,
"grad_norm": 1.059685468673706,
"learning_rate": 2.2394203147108005e-05,
"loss": 1.3225,
"step": 462500
},
{
"epoch": 19.51692450364625,
"grad_norm": 1.0375324487686157,
"learning_rate": 2.2308450885392103e-05,
"loss": 1.3217,
"step": 463000
},
{
"epoch": 19.5380010959828,
"grad_norm": 1.0744473934173584,
"learning_rate": 2.22226986236762e-05,
"loss": 1.3206,
"step": 463500
},
{
"epoch": 19.559077688319352,
"grad_norm": 1.0338191986083984,
"learning_rate": 2.21369463619603e-05,
"loss": 1.3228,
"step": 464000
},
{
"epoch": 19.580154280655904,
"grad_norm": 1.0830291509628296,
"learning_rate": 2.2051194100244394e-05,
"loss": 1.3215,
"step": 464500
},
{
"epoch": 19.601230872992456,
"grad_norm": 1.096269965171814,
"learning_rate": 2.1965441838528493e-05,
"loss": 1.3218,
"step": 465000
},
{
"epoch": 19.622307465329005,
"grad_norm": 1.0255447626113892,
"learning_rate": 2.187968957681259e-05,
"loss": 1.3219,
"step": 465500
},
{
"epoch": 19.643384057665557,
"grad_norm": 1.0073683261871338,
"learning_rate": 2.179393731509669e-05,
"loss": 1.3227,
"step": 466000
},
{
"epoch": 19.66446065000211,
"grad_norm": 1.058016061782837,
"learning_rate": 2.1708185053380784e-05,
"loss": 1.3205,
"step": 466500
},
{
"epoch": 19.685537242338658,
"grad_norm": 1.050594449043274,
"learning_rate": 2.1622432791664883e-05,
"loss": 1.3229,
"step": 467000
},
{
"epoch": 19.70661383467521,
"grad_norm": 1.0627409219741821,
"learning_rate": 2.1536680529948978e-05,
"loss": 1.3206,
"step": 467500
},
{
"epoch": 19.727690427011762,
"grad_norm": 1.0585235357284546,
"learning_rate": 2.1450928268233076e-05,
"loss": 1.3195,
"step": 468000
},
{
"epoch": 19.74876701934831,
"grad_norm": 1.025846004486084,
"learning_rate": 2.1365176006517174e-05,
"loss": 1.3194,
"step": 468500
},
{
"epoch": 19.769843611684863,
"grad_norm": 1.0703961849212646,
"learning_rate": 2.127942374480127e-05,
"loss": 1.3203,
"step": 469000
},
{
"epoch": 19.790920204021415,
"grad_norm": 1.057073712348938,
"learning_rate": 2.1193671483085367e-05,
"loss": 1.3235,
"step": 469500
},
{
"epoch": 19.811996796357963,
"grad_norm": 1.059766173362732,
"learning_rate": 2.1107919221369466e-05,
"loss": 1.321,
"step": 470000
},
{
"epoch": 19.811996796357963,
"eval_accuracy": 0.7464482678301888,
"eval_loss": 1.1127008199691772,
"eval_runtime": 179.9291,
"eval_samples_per_second": 670.853,
"eval_steps_per_second": 6.992,
"step": 470000
},
{
"epoch": 19.833073388694515,
"grad_norm": 1.0303518772125244,
"learning_rate": 2.102216695965356e-05,
"loss": 1.3201,
"step": 470500
},
{
"epoch": 19.854149981031068,
"grad_norm": 1.0270575284957886,
"learning_rate": 2.093641469793766e-05,
"loss": 1.3188,
"step": 471000
},
{
"epoch": 19.87522657336762,
"grad_norm": 1.039878010749817,
"learning_rate": 2.0850662436221757e-05,
"loss": 1.3173,
"step": 471500
},
{
"epoch": 19.89630316570417,
"grad_norm": 1.070297360420227,
"learning_rate": 2.0764910174505852e-05,
"loss": 1.3177,
"step": 472000
},
{
"epoch": 19.91737975804072,
"grad_norm": 1.0523431301116943,
"learning_rate": 2.067915791278995e-05,
"loss": 1.3191,
"step": 472500
},
{
"epoch": 19.938456350377272,
"grad_norm": 1.0907182693481445,
"learning_rate": 2.059340565107405e-05,
"loss": 1.316,
"step": 473000
},
{
"epoch": 19.95953294271382,
"grad_norm": 1.0330580472946167,
"learning_rate": 2.0507653389358144e-05,
"loss": 1.3171,
"step": 473500
},
{
"epoch": 19.980609535050373,
"grad_norm": 1.10728919506073,
"learning_rate": 2.0421901127642242e-05,
"loss": 1.3153,
"step": 474000
},
{
"epoch": 20.001686127386925,
"grad_norm": 1.0791987180709839,
"learning_rate": 2.033614886592634e-05,
"loss": 1.3221,
"step": 474500
},
{
"epoch": 20.022762719723474,
"grad_norm": 1.0551412105560303,
"learning_rate": 2.0250396604210435e-05,
"loss": 1.3172,
"step": 475000
},
{
"epoch": 20.043839312060026,
"grad_norm": 1.1041675806045532,
"learning_rate": 2.0164644342494534e-05,
"loss": 1.3162,
"step": 475500
},
{
"epoch": 20.064915904396578,
"grad_norm": 1.065263271331787,
"learning_rate": 2.007889208077863e-05,
"loss": 1.3105,
"step": 476000
},
{
"epoch": 20.085992496733127,
"grad_norm": 1.0644158124923706,
"learning_rate": 1.9993139819062727e-05,
"loss": 1.3124,
"step": 476500
},
{
"epoch": 20.10706908906968,
"grad_norm": 1.055672526359558,
"learning_rate": 1.9907387557346825e-05,
"loss": 1.3147,
"step": 477000
},
{
"epoch": 20.12814568140623,
"grad_norm": 1.0548309087753296,
"learning_rate": 1.9821635295630923e-05,
"loss": 1.3132,
"step": 477500
},
{
"epoch": 20.14922227374278,
"grad_norm": 1.0814296007156372,
"learning_rate": 1.9735883033915022e-05,
"loss": 1.3125,
"step": 478000
},
{
"epoch": 20.17029886607933,
"grad_norm": 1.0717817544937134,
"learning_rate": 1.965013077219912e-05,
"loss": 1.3154,
"step": 478500
},
{
"epoch": 20.191375458415884,
"grad_norm": 1.0595901012420654,
"learning_rate": 1.9564378510483215e-05,
"loss": 1.3123,
"step": 479000
},
{
"epoch": 20.212452050752436,
"grad_norm": 1.0718363523483276,
"learning_rate": 1.9478626248767313e-05,
"loss": 1.315,
"step": 479500
},
{
"epoch": 20.233528643088984,
"grad_norm": 1.071467638015747,
"learning_rate": 1.939287398705141e-05,
"loss": 1.3123,
"step": 480000
},
{
"epoch": 20.233528643088984,
"eval_accuracy": 0.747397249951712,
"eval_loss": 1.1086502075195312,
"eval_runtime": 179.27,
"eval_samples_per_second": 673.32,
"eval_steps_per_second": 7.017,
"step": 480000
},
{
"epoch": 20.254605235425537,
"grad_norm": 1.0354539155960083,
"learning_rate": 1.9307121725335506e-05,
"loss": 1.3098,
"step": 480500
},
{
"epoch": 20.27568182776209,
"grad_norm": 1.0836129188537598,
"learning_rate": 1.9221369463619605e-05,
"loss": 1.3101,
"step": 481000
},
{
"epoch": 20.296758420098637,
"grad_norm": 1.0656843185424805,
"learning_rate": 1.9135617201903703e-05,
"loss": 1.31,
"step": 481500
},
{
"epoch": 20.31783501243519,
"grad_norm": 1.076956033706665,
"learning_rate": 1.9049864940187798e-05,
"loss": 1.3101,
"step": 482000
},
{
"epoch": 20.33891160477174,
"grad_norm": 1.0736594200134277,
"learning_rate": 1.8964112678471896e-05,
"loss": 1.3115,
"step": 482500
},
{
"epoch": 20.35998819710829,
"grad_norm": 1.053231954574585,
"learning_rate": 1.8878360416755995e-05,
"loss": 1.3133,
"step": 483000
},
{
"epoch": 20.381064789444842,
"grad_norm": 1.0727981328964233,
"learning_rate": 1.879260815504009e-05,
"loss": 1.3114,
"step": 483500
},
{
"epoch": 20.402141381781394,
"grad_norm": 1.0519009828567505,
"learning_rate": 1.8706855893324188e-05,
"loss": 1.3084,
"step": 484000
},
{
"epoch": 20.423217974117943,
"grad_norm": 1.0419996976852417,
"learning_rate": 1.8621103631608283e-05,
"loss": 1.3112,
"step": 484500
},
{
"epoch": 20.444294566454495,
"grad_norm": 1.0779378414154053,
"learning_rate": 1.853535136989238e-05,
"loss": 1.3105,
"step": 485000
},
{
"epoch": 20.465371158791047,
"grad_norm": 1.0957493782043457,
"learning_rate": 1.844959910817648e-05,
"loss": 1.3079,
"step": 485500
},
{
"epoch": 20.4864477511276,
"grad_norm": 1.1035734415054321,
"learning_rate": 1.8363846846460574e-05,
"loss": 1.313,
"step": 486000
},
{
"epoch": 20.507524343464148,
"grad_norm": 1.0734623670578003,
"learning_rate": 1.8278094584744673e-05,
"loss": 1.3077,
"step": 486500
},
{
"epoch": 20.5286009358007,
"grad_norm": 1.088228464126587,
"learning_rate": 1.819234232302877e-05,
"loss": 1.3112,
"step": 487000
},
{
"epoch": 20.549677528137252,
"grad_norm": 1.0551313161849976,
"learning_rate": 1.8106590061312866e-05,
"loss": 1.3118,
"step": 487500
},
{
"epoch": 20.5707541204738,
"grad_norm": 1.083943486213684,
"learning_rate": 1.8020837799596964e-05,
"loss": 1.3114,
"step": 488000
},
{
"epoch": 20.591830712810353,
"grad_norm": 1.0358083248138428,
"learning_rate": 1.7935085537881062e-05,
"loss": 1.3108,
"step": 488500
},
{
"epoch": 20.612907305146905,
"grad_norm": 1.0961809158325195,
"learning_rate": 1.7849333276165157e-05,
"loss": 1.3087,
"step": 489000
},
{
"epoch": 20.633983897483454,
"grad_norm": 1.0800755023956299,
"learning_rate": 1.7763581014449256e-05,
"loss": 1.3083,
"step": 489500
},
{
"epoch": 20.655060489820006,
"grad_norm": 1.0686836242675781,
"learning_rate": 1.7677828752733354e-05,
"loss": 1.3069,
"step": 490000
},
{
"epoch": 20.655060489820006,
"eval_accuracy": 0.7480371067441939,
"eval_loss": 1.1046335697174072,
"eval_runtime": 179.5593,
"eval_samples_per_second": 672.235,
"eval_steps_per_second": 7.006,
"step": 490000
},
{
"epoch": 20.676137082156558,
"grad_norm": 1.0932430028915405,
"learning_rate": 1.7592076491017452e-05,
"loss": 1.3098,
"step": 490500
},
{
"epoch": 20.697213674493106,
"grad_norm": 1.1011919975280762,
"learning_rate": 1.7506324229301547e-05,
"loss": 1.3116,
"step": 491000
},
{
"epoch": 20.71829026682966,
"grad_norm": 1.06903076171875,
"learning_rate": 1.7420571967585646e-05,
"loss": 1.3102,
"step": 491500
},
{
"epoch": 20.73936685916621,
"grad_norm": 1.0857702493667603,
"learning_rate": 1.7334819705869744e-05,
"loss": 1.3062,
"step": 492000
},
{
"epoch": 20.760443451502763,
"grad_norm": 1.0666908025741577,
"learning_rate": 1.7249067444153842e-05,
"loss": 1.3118,
"step": 492500
},
{
"epoch": 20.78152004383931,
"grad_norm": 1.0573458671569824,
"learning_rate": 1.7163315182437937e-05,
"loss": 1.3054,
"step": 493000
},
{
"epoch": 20.802596636175863,
"grad_norm": 1.07243812084198,
"learning_rate": 1.7077562920722035e-05,
"loss": 1.3068,
"step": 493500
},
{
"epoch": 20.823673228512416,
"grad_norm": 1.076859712600708,
"learning_rate": 1.6991810659006134e-05,
"loss": 1.3085,
"step": 494000
},
{
"epoch": 20.844749820848964,
"grad_norm": 1.0283046960830688,
"learning_rate": 1.690605839729023e-05,
"loss": 1.3076,
"step": 494500
},
{
"epoch": 20.865826413185516,
"grad_norm": 1.0457956790924072,
"learning_rate": 1.6820306135574327e-05,
"loss": 1.3065,
"step": 495000
},
{
"epoch": 20.88690300552207,
"grad_norm": 1.1020958423614502,
"learning_rate": 1.6734553873858425e-05,
"loss": 1.3084,
"step": 495500
},
{
"epoch": 20.907979597858617,
"grad_norm": 1.0991090536117554,
"learning_rate": 1.664880161214252e-05,
"loss": 1.3067,
"step": 496000
},
{
"epoch": 20.92905619019517,
"grad_norm": 1.0590287446975708,
"learning_rate": 1.656304935042662e-05,
"loss": 1.3089,
"step": 496500
},
{
"epoch": 20.95013278253172,
"grad_norm": 1.0456629991531372,
"learning_rate": 1.6477297088710717e-05,
"loss": 1.306,
"step": 497000
},
{
"epoch": 20.97120937486827,
"grad_norm": 1.0750702619552612,
"learning_rate": 1.639154482699481e-05,
"loss": 1.306,
"step": 497500
},
{
"epoch": 20.992285967204822,
"grad_norm": 1.0466045141220093,
"learning_rate": 1.630579256527891e-05,
"loss": 1.3049,
"step": 498000
},
{
"epoch": 21.013362559541374,
"grad_norm": 1.0760300159454346,
"learning_rate": 1.6220040303563008e-05,
"loss": 1.3037,
"step": 498500
},
{
"epoch": 21.034439151877926,
"grad_norm": 1.0749800205230713,
"learning_rate": 1.6134288041847103e-05,
"loss": 1.3063,
"step": 499000
},
{
"epoch": 21.055515744214475,
"grad_norm": 1.076637864112854,
"learning_rate": 1.60485357801312e-05,
"loss": 1.3028,
"step": 499500
},
{
"epoch": 21.076592336551027,
"grad_norm": 1.036298155784607,
"learning_rate": 1.59627835184153e-05,
"loss": 1.3016,
"step": 500000
},
{
"epoch": 21.076592336551027,
"eval_accuracy": 0.7486180490782286,
"eval_loss": 1.0993659496307373,
"eval_runtime": 180.2427,
"eval_samples_per_second": 669.686,
"eval_steps_per_second": 6.979,
"step": 500000
},
{
"epoch": 21.09766892888758,
"grad_norm": 1.0629420280456543,
"learning_rate": 1.5877031256699395e-05,
"loss": 1.3016,
"step": 500500
},
{
"epoch": 21.118745521224128,
"grad_norm": 1.0522983074188232,
"learning_rate": 1.5791278994983493e-05,
"loss": 1.3036,
"step": 501000
},
{
"epoch": 21.13982211356068,
"grad_norm": 1.049883246421814,
"learning_rate": 1.570552673326759e-05,
"loss": 1.3007,
"step": 501500
},
{
"epoch": 21.16089870589723,
"grad_norm": 1.0550237894058228,
"learning_rate": 1.5619774471551686e-05,
"loss": 1.3038,
"step": 502000
},
{
"epoch": 21.18197529823378,
"grad_norm": 1.0725326538085938,
"learning_rate": 1.5534022209835785e-05,
"loss": 1.3025,
"step": 502500
},
{
"epoch": 21.203051890570332,
"grad_norm": 1.0421857833862305,
"learning_rate": 1.544826994811988e-05,
"loss": 1.303,
"step": 503000
},
{
"epoch": 21.224128482906885,
"grad_norm": 1.0802655220031738,
"learning_rate": 1.5362517686403978e-05,
"loss": 1.3029,
"step": 503500
},
{
"epoch": 21.245205075243433,
"grad_norm": 1.0664763450622559,
"learning_rate": 1.5276765424688076e-05,
"loss": 1.3029,
"step": 504000
},
{
"epoch": 21.266281667579985,
"grad_norm": 1.1007957458496094,
"learning_rate": 1.5191013162972173e-05,
"loss": 1.3007,
"step": 504500
},
{
"epoch": 21.287358259916537,
"grad_norm": 1.0769048929214478,
"learning_rate": 1.5105260901256271e-05,
"loss": 1.3015,
"step": 505000
},
{
"epoch": 21.308434852253086,
"grad_norm": 1.0929840803146362,
"learning_rate": 1.501950863954037e-05,
"loss": 1.3018,
"step": 505500
},
{
"epoch": 21.329511444589638,
"grad_norm": 1.0656461715698242,
"learning_rate": 1.4933756377824464e-05,
"loss": 1.2989,
"step": 506000
},
{
"epoch": 21.35058803692619,
"grad_norm": 1.0714404582977295,
"learning_rate": 1.4848004116108563e-05,
"loss": 1.2969,
"step": 506500
},
{
"epoch": 21.371664629262742,
"grad_norm": 1.042749285697937,
"learning_rate": 1.4762251854392661e-05,
"loss": 1.3032,
"step": 507000
},
{
"epoch": 21.39274122159929,
"grad_norm": 1.05530846118927,
"learning_rate": 1.4676499592676758e-05,
"loss": 1.3031,
"step": 507500
},
{
"epoch": 21.413817813935843,
"grad_norm": 1.0559828281402588,
"learning_rate": 1.4590747330960856e-05,
"loss": 1.2997,
"step": 508000
},
{
"epoch": 21.434894406272395,
"grad_norm": 1.0788999795913696,
"learning_rate": 1.4504995069244952e-05,
"loss": 1.2976,
"step": 508500
},
{
"epoch": 21.455970998608944,
"grad_norm": 1.0418285131454468,
"learning_rate": 1.4419242807529049e-05,
"loss": 1.3011,
"step": 509000
},
{
"epoch": 21.477047590945496,
"grad_norm": 1.027953028678894,
"learning_rate": 1.4333490545813147e-05,
"loss": 1.2999,
"step": 509500
},
{
"epoch": 21.498124183282048,
"grad_norm": 1.0732427835464478,
"learning_rate": 1.4247738284097246e-05,
"loss": 1.2977,
"step": 510000
},
{
"epoch": 21.498124183282048,
"eval_accuracy": 0.7496748157268738,
"eval_loss": 1.0951884984970093,
"eval_runtime": 180.1728,
"eval_samples_per_second": 669.946,
"eval_steps_per_second": 6.982,
"step": 510000
},
{
"epoch": 21.519200775618597,
"grad_norm": 1.054063081741333,
"learning_rate": 1.416198602238134e-05,
"loss": 1.3012,
"step": 510500
},
{
"epoch": 21.54027736795515,
"grad_norm": 1.0439032316207886,
"learning_rate": 1.4076233760665439e-05,
"loss": 1.2996,
"step": 511000
},
{
"epoch": 21.5613539602917,
"grad_norm": 1.0697977542877197,
"learning_rate": 1.3990481498949534e-05,
"loss": 1.297,
"step": 511500
},
{
"epoch": 21.58243055262825,
"grad_norm": 1.0741539001464844,
"learning_rate": 1.3904729237233632e-05,
"loss": 1.3005,
"step": 512000
},
{
"epoch": 21.6035071449648,
"grad_norm": 1.0577858686447144,
"learning_rate": 1.381897697551773e-05,
"loss": 1.2976,
"step": 512500
},
{
"epoch": 21.624583737301354,
"grad_norm": 1.0526641607284546,
"learning_rate": 1.3733224713801825e-05,
"loss": 1.3017,
"step": 513000
},
{
"epoch": 21.645660329637906,
"grad_norm": 1.0861785411834717,
"learning_rate": 1.3647472452085924e-05,
"loss": 1.3022,
"step": 513500
},
{
"epoch": 21.666736921974454,
"grad_norm": 1.0773048400878906,
"learning_rate": 1.3561720190370022e-05,
"loss": 1.2992,
"step": 514000
},
{
"epoch": 21.687813514311006,
"grad_norm": 1.0931907892227173,
"learning_rate": 1.3475967928654119e-05,
"loss": 1.2969,
"step": 514500
},
{
"epoch": 21.70889010664756,
"grad_norm": 1.0796507596969604,
"learning_rate": 1.3390215666938217e-05,
"loss": 1.3002,
"step": 515000
},
{
"epoch": 21.729966698984107,
"grad_norm": 1.0816377401351929,
"learning_rate": 1.3304463405222315e-05,
"loss": 1.2992,
"step": 515500
},
{
"epoch": 21.75104329132066,
"grad_norm": 1.0915684700012207,
"learning_rate": 1.321871114350641e-05,
"loss": 1.3002,
"step": 516000
},
{
"epoch": 21.77211988365721,
"grad_norm": 1.098949670791626,
"learning_rate": 1.3132958881790508e-05,
"loss": 1.2963,
"step": 516500
},
{
"epoch": 21.79319647599376,
"grad_norm": 1.0689001083374023,
"learning_rate": 1.3047206620074607e-05,
"loss": 1.2961,
"step": 517000
},
{
"epoch": 21.814273068330312,
"grad_norm": 1.09818434715271,
"learning_rate": 1.2961454358358702e-05,
"loss": 1.2974,
"step": 517500
},
{
"epoch": 21.835349660666864,
"grad_norm": 1.1238771677017212,
"learning_rate": 1.28757020966428e-05,
"loss": 1.295,
"step": 518000
},
{
"epoch": 21.856426253003413,
"grad_norm": 1.067001223564148,
"learning_rate": 1.2789949834926898e-05,
"loss": 1.2945,
"step": 518500
},
{
"epoch": 21.877502845339965,
"grad_norm": 1.0604243278503418,
"learning_rate": 1.2704197573210993e-05,
"loss": 1.2997,
"step": 519000
},
{
"epoch": 21.898579437676517,
"grad_norm": 1.0653491020202637,
"learning_rate": 1.2618445311495092e-05,
"loss": 1.2942,
"step": 519500
},
{
"epoch": 21.91965603001307,
"grad_norm": 1.0872427225112915,
"learning_rate": 1.2532693049779186e-05,
"loss": 1.2929,
"step": 520000
},
{
"epoch": 21.91965603001307,
"eval_accuracy": 0.7499933660656343,
"eval_loss": 1.0932220220565796,
"eval_runtime": 179.7957,
"eval_samples_per_second": 671.351,
"eval_steps_per_second": 6.997,
"step": 520000
},
{
"epoch": 21.940732622349618,
"grad_norm": 1.108196496963501,
"learning_rate": 1.2446940788063285e-05,
"loss": 1.2966,
"step": 520500
},
{
"epoch": 21.96180921468617,
"grad_norm": 1.0955814123153687,
"learning_rate": 1.2361188526347383e-05,
"loss": 1.296,
"step": 521000
},
{
"epoch": 21.982885807022722,
"grad_norm": 1.059950351715088,
"learning_rate": 1.227543626463148e-05,
"loss": 1.297,
"step": 521500
},
{
"epoch": 22.00396239935927,
"grad_norm": 1.1138300895690918,
"learning_rate": 1.2189684002915578e-05,
"loss": 1.2926,
"step": 522000
},
{
"epoch": 22.025038991695823,
"grad_norm": 1.1060651540756226,
"learning_rate": 1.2103931741199675e-05,
"loss": 1.2916,
"step": 522500
},
{
"epoch": 22.046115584032375,
"grad_norm": 1.077918529510498,
"learning_rate": 1.2018179479483773e-05,
"loss": 1.2936,
"step": 523000
},
{
"epoch": 22.067192176368923,
"grad_norm": 1.0522549152374268,
"learning_rate": 1.193242721776787e-05,
"loss": 1.2943,
"step": 523500
},
{
"epoch": 22.088268768705476,
"grad_norm": 1.092625617980957,
"learning_rate": 1.1846674956051966e-05,
"loss": 1.2953,
"step": 524000
},
{
"epoch": 22.109345361042028,
"grad_norm": 1.1038691997528076,
"learning_rate": 1.1760922694336064e-05,
"loss": 1.2896,
"step": 524500
},
{
"epoch": 22.130421953378576,
"grad_norm": 1.1169365644454956,
"learning_rate": 1.1675170432620161e-05,
"loss": 1.2956,
"step": 525000
},
{
"epoch": 22.15149854571513,
"grad_norm": 1.0569047927856445,
"learning_rate": 1.1589418170904258e-05,
"loss": 1.2894,
"step": 525500
},
{
"epoch": 22.17257513805168,
"grad_norm": 1.049979567527771,
"learning_rate": 1.1503665909188354e-05,
"loss": 1.2913,
"step": 526000
},
{
"epoch": 22.193651730388233,
"grad_norm": 1.1312675476074219,
"learning_rate": 1.1417913647472453e-05,
"loss": 1.2921,
"step": 526500
},
{
"epoch": 22.21472832272478,
"grad_norm": 1.074652075767517,
"learning_rate": 1.133216138575655e-05,
"loss": 1.2919,
"step": 527000
},
{
"epoch": 22.235804915061333,
"grad_norm": 1.0824129581451416,
"learning_rate": 1.1246409124040646e-05,
"loss": 1.29,
"step": 527500
},
{
"epoch": 22.256881507397885,
"grad_norm": 1.0799014568328857,
"learning_rate": 1.1160656862324744e-05,
"loss": 1.2938,
"step": 528000
},
{
"epoch": 22.277958099734434,
"grad_norm": 1.0733681917190552,
"learning_rate": 1.1074904600608842e-05,
"loss": 1.2928,
"step": 528500
},
{
"epoch": 22.299034692070986,
"grad_norm": 1.095476746559143,
"learning_rate": 1.0989152338892939e-05,
"loss": 1.2886,
"step": 529000
},
{
"epoch": 22.320111284407538,
"grad_norm": 1.1352216005325317,
"learning_rate": 1.0903400077177037e-05,
"loss": 1.2921,
"step": 529500
},
{
"epoch": 22.341187876744087,
"grad_norm": 1.0702844858169556,
"learning_rate": 1.0817647815461134e-05,
"loss": 1.2924,
"step": 530000
},
{
"epoch": 22.341187876744087,
"eval_accuracy": 0.7505405882368048,
"eval_loss": 1.0898773670196533,
"eval_runtime": 181.9801,
"eval_samples_per_second": 663.292,
"eval_steps_per_second": 6.913,
"step": 530000
},
{
"epoch": 22.36226446908064,
"grad_norm": 1.0668058395385742,
"learning_rate": 1.073189555374523e-05,
"loss": 1.2937,
"step": 530500
},
{
"epoch": 22.38334106141719,
"grad_norm": 1.1220370531082153,
"learning_rate": 1.0646143292029327e-05,
"loss": 1.2943,
"step": 531000
},
{
"epoch": 22.40441765375374,
"grad_norm": 1.0763940811157227,
"learning_rate": 1.0560391030313425e-05,
"loss": 1.288,
"step": 531500
},
{
"epoch": 22.42549424609029,
"grad_norm": 1.059950828552246,
"learning_rate": 1.0474638768597522e-05,
"loss": 1.2878,
"step": 532000
},
{
"epoch": 22.446570838426844,
"grad_norm": 1.0889326333999634,
"learning_rate": 1.0388886506881619e-05,
"loss": 1.29,
"step": 532500
},
{
"epoch": 22.467647430763392,
"grad_norm": 1.0938808917999268,
"learning_rate": 1.0303134245165717e-05,
"loss": 1.2921,
"step": 533000
},
{
"epoch": 22.488724023099945,
"grad_norm": 1.1028900146484375,
"learning_rate": 1.0217381983449814e-05,
"loss": 1.2895,
"step": 533500
},
{
"epoch": 22.509800615436497,
"grad_norm": 1.0475518703460693,
"learning_rate": 1.013162972173391e-05,
"loss": 1.29,
"step": 534000
},
{
"epoch": 22.53087720777305,
"grad_norm": 1.0876102447509766,
"learning_rate": 1.0045877460018009e-05,
"loss": 1.2918,
"step": 534500
},
{
"epoch": 22.551953800109597,
"grad_norm": 1.0975545644760132,
"learning_rate": 9.960125198302105e-06,
"loss": 1.2939,
"step": 535000
},
{
"epoch": 22.57303039244615,
"grad_norm": 1.0940485000610352,
"learning_rate": 9.874372936586203e-06,
"loss": 1.29,
"step": 535500
},
{
"epoch": 22.5941069847827,
"grad_norm": 1.0592381954193115,
"learning_rate": 9.7886206748703e-06,
"loss": 1.2886,
"step": 536000
},
{
"epoch": 22.61518357711925,
"grad_norm": 1.0931566953659058,
"learning_rate": 9.702868413154398e-06,
"loss": 1.2883,
"step": 536500
},
{
"epoch": 22.636260169455802,
"grad_norm": 1.0770177841186523,
"learning_rate": 9.617116151438495e-06,
"loss": 1.2871,
"step": 537000
},
{
"epoch": 22.657336761792354,
"grad_norm": 1.0931721925735474,
"learning_rate": 9.531363889722592e-06,
"loss": 1.2901,
"step": 537500
},
{
"epoch": 22.678413354128903,
"grad_norm": 1.0444447994232178,
"learning_rate": 9.44561162800669e-06,
"loss": 1.2877,
"step": 538000
},
{
"epoch": 22.699489946465455,
"grad_norm": 1.082780122756958,
"learning_rate": 9.359859366290787e-06,
"loss": 1.2893,
"step": 538500
},
{
"epoch": 22.720566538802007,
"grad_norm": 1.0417447090148926,
"learning_rate": 9.274107104574883e-06,
"loss": 1.2907,
"step": 539000
},
{
"epoch": 22.741643131138556,
"grad_norm": 1.0530136823654175,
"learning_rate": 9.18835484285898e-06,
"loss": 1.2851,
"step": 539500
},
{
"epoch": 22.762719723475108,
"grad_norm": 1.0858839750289917,
"learning_rate": 9.102602581143078e-06,
"loss": 1.2862,
"step": 540000
},
{
"epoch": 22.762719723475108,
"eval_accuracy": 0.750952696061676,
"eval_loss": 1.0886858701705933,
"eval_runtime": 180.21,
"eval_samples_per_second": 669.808,
"eval_steps_per_second": 6.981,
"step": 540000
},
{
"epoch": 22.78379631581166,
"grad_norm": 1.1260205507278442,
"learning_rate": 9.016850319427175e-06,
"loss": 1.2894,
"step": 540500
},
{
"epoch": 22.804872908148212,
"grad_norm": 1.0872058868408203,
"learning_rate": 8.931098057711271e-06,
"loss": 1.2883,
"step": 541000
},
{
"epoch": 22.82594950048476,
"grad_norm": 1.1029294729232788,
"learning_rate": 8.84534579599537e-06,
"loss": 1.2888,
"step": 541500
},
{
"epoch": 22.847026092821313,
"grad_norm": 1.1280102729797363,
"learning_rate": 8.759593534279468e-06,
"loss": 1.2881,
"step": 542000
},
{
"epoch": 22.868102685157865,
"grad_norm": 1.0686568021774292,
"learning_rate": 8.673841272563565e-06,
"loss": 1.2863,
"step": 542500
},
{
"epoch": 22.889179277494414,
"grad_norm": 1.0763654708862305,
"learning_rate": 8.588089010847663e-06,
"loss": 1.2888,
"step": 543000
},
{
"epoch": 22.910255869830966,
"grad_norm": 1.0961731672286987,
"learning_rate": 8.50233674913176e-06,
"loss": 1.2877,
"step": 543500
},
{
"epoch": 22.931332462167518,
"grad_norm": 1.096727728843689,
"learning_rate": 8.416584487415856e-06,
"loss": 1.2842,
"step": 544000
},
{
"epoch": 22.952409054504066,
"grad_norm": 1.1517711877822876,
"learning_rate": 8.330832225699953e-06,
"loss": 1.2834,
"step": 544500
},
{
"epoch": 22.97348564684062,
"grad_norm": 1.106178879737854,
"learning_rate": 8.245079963984051e-06,
"loss": 1.2832,
"step": 545000
},
{
"epoch": 22.99456223917717,
"grad_norm": 1.0865274667739868,
"learning_rate": 8.159327702268148e-06,
"loss": 1.2888,
"step": 545500
},
{
"epoch": 23.01563883151372,
"grad_norm": 1.0775015354156494,
"learning_rate": 8.073575440552244e-06,
"loss": 1.2851,
"step": 546000
},
{
"epoch": 23.03671542385027,
"grad_norm": 1.0845338106155396,
"learning_rate": 7.987823178836343e-06,
"loss": 1.2853,
"step": 546500
},
{
"epoch": 23.057792016186824,
"grad_norm": 1.1126538515090942,
"learning_rate": 7.90207091712044e-06,
"loss": 1.2836,
"step": 547000
},
{
"epoch": 23.078868608523376,
"grad_norm": 1.127252459526062,
"learning_rate": 7.816318655404536e-06,
"loss": 1.2837,
"step": 547500
},
{
"epoch": 23.099945200859924,
"grad_norm": 1.0950137376785278,
"learning_rate": 7.730566393688634e-06,
"loss": 1.2856,
"step": 548000
},
{
"epoch": 23.121021793196476,
"grad_norm": 1.0910077095031738,
"learning_rate": 7.64481413197273e-06,
"loss": 1.2849,
"step": 548500
},
{
"epoch": 23.14209838553303,
"grad_norm": 1.0781633853912354,
"learning_rate": 7.559061870256828e-06,
"loss": 1.2844,
"step": 549000
},
{
"epoch": 23.163174977869577,
"grad_norm": 1.112038493156433,
"learning_rate": 7.473309608540925e-06,
"loss": 1.283,
"step": 549500
},
{
"epoch": 23.18425157020613,
"grad_norm": 1.0995217561721802,
"learning_rate": 7.387557346825023e-06,
"loss": 1.2853,
"step": 550000
},
{
"epoch": 23.18425157020613,
"eval_accuracy": 0.7516806449801912,
"eval_loss": 1.0846529006958008,
"eval_runtime": 180.2259,
"eval_samples_per_second": 669.748,
"eval_steps_per_second": 6.98,
"step": 550000
},
{
"epoch": 23.20532816254268,
"grad_norm": 1.120358943939209,
"learning_rate": 7.3018050851091205e-06,
"loss": 1.2811,
"step": 550500
},
{
"epoch": 23.22640475487923,
"grad_norm": 1.0734331607818604,
"learning_rate": 7.216052823393217e-06,
"loss": 1.2845,
"step": 551000
},
{
"epoch": 23.247481347215782,
"grad_norm": 1.119028091430664,
"learning_rate": 7.1303005616773155e-06,
"loss": 1.2842,
"step": 551500
},
{
"epoch": 23.268557939552334,
"grad_norm": 1.1106165647506714,
"learning_rate": 7.044548299961412e-06,
"loss": 1.2884,
"step": 552000
},
{
"epoch": 23.289634531888883,
"grad_norm": 1.1227283477783203,
"learning_rate": 6.958796038245509e-06,
"loss": 1.2784,
"step": 552500
},
{
"epoch": 23.310711124225435,
"grad_norm": 1.203613042831421,
"learning_rate": 6.873043776529606e-06,
"loss": 1.2872,
"step": 553000
},
{
"epoch": 23.331787716561987,
"grad_norm": 1.105718731880188,
"learning_rate": 6.787291514813704e-06,
"loss": 1.281,
"step": 553500
},
{
"epoch": 23.35286430889854,
"grad_norm": 1.1004852056503296,
"learning_rate": 6.701539253097801e-06,
"loss": 1.2833,
"step": 554000
},
{
"epoch": 23.373940901235088,
"grad_norm": 1.124861240386963,
"learning_rate": 6.615786991381898e-06,
"loss": 1.2835,
"step": 554500
},
{
"epoch": 23.39501749357164,
"grad_norm": 1.0674879550933838,
"learning_rate": 6.530034729665996e-06,
"loss": 1.2826,
"step": 555000
},
{
"epoch": 23.416094085908192,
"grad_norm": 1.1134259700775146,
"learning_rate": 6.444282467950093e-06,
"loss": 1.2823,
"step": 555500
},
{
"epoch": 23.43717067824474,
"grad_norm": 1.071090817451477,
"learning_rate": 6.358530206234189e-06,
"loss": 1.2808,
"step": 556000
},
{
"epoch": 23.458247270581293,
"grad_norm": 1.0834676027297974,
"learning_rate": 6.272777944518287e-06,
"loss": 1.2833,
"step": 556500
},
{
"epoch": 23.479323862917845,
"grad_norm": 1.1171835660934448,
"learning_rate": 6.187025682802384e-06,
"loss": 1.2842,
"step": 557000
},
{
"epoch": 23.500400455254393,
"grad_norm": 1.1138626337051392,
"learning_rate": 6.101273421086482e-06,
"loss": 1.2819,
"step": 557500
},
{
"epoch": 23.521477047590945,
"grad_norm": 1.1007425785064697,
"learning_rate": 6.015521159370579e-06,
"loss": 1.2821,
"step": 558000
},
{
"epoch": 23.542553639927497,
"grad_norm": 1.0929934978485107,
"learning_rate": 5.929768897654676e-06,
"loss": 1.2797,
"step": 558500
},
{
"epoch": 23.563630232264046,
"grad_norm": 1.0950853824615479,
"learning_rate": 5.844016635938773e-06,
"loss": 1.2823,
"step": 559000
},
{
"epoch": 23.584706824600598,
"grad_norm": 1.075896143913269,
"learning_rate": 5.75826437422287e-06,
"loss": 1.2813,
"step": 559500
},
{
"epoch": 23.60578341693715,
"grad_norm": 1.121459722518921,
"learning_rate": 5.672512112506967e-06,
"loss": 1.2827,
"step": 560000
},
{
"epoch": 23.60578341693715,
"eval_accuracy": 0.7522949320827662,
"eval_loss": 1.0812584161758423,
"eval_runtime": 180.9746,
"eval_samples_per_second": 666.977,
"eval_steps_per_second": 6.951,
"step": 560000
},
{
"epoch": 23.6268600092737,
"grad_norm": 1.1021052598953247,
"learning_rate": 5.5867598507910655e-06,
"loss": 1.2821,
"step": 560500
},
{
"epoch": 23.64793660161025,
"grad_norm": 1.105753779411316,
"learning_rate": 5.501007589075162e-06,
"loss": 1.2812,
"step": 561000
},
{
"epoch": 23.669013193946803,
"grad_norm": 1.1438204050064087,
"learning_rate": 5.41525532735926e-06,
"loss": 1.2824,
"step": 561500
},
{
"epoch": 23.690089786283355,
"grad_norm": 1.1509230136871338,
"learning_rate": 5.329503065643356e-06,
"loss": 1.2778,
"step": 562000
},
{
"epoch": 23.711166378619904,
"grad_norm": 1.1568360328674316,
"learning_rate": 5.243750803927454e-06,
"loss": 1.2789,
"step": 562500
},
{
"epoch": 23.732242970956456,
"grad_norm": 1.119217038154602,
"learning_rate": 5.157998542211551e-06,
"loss": 1.2808,
"step": 563000
},
{
"epoch": 23.753319563293008,
"grad_norm": 1.093076229095459,
"learning_rate": 5.072246280495649e-06,
"loss": 1.2775,
"step": 563500
},
{
"epoch": 23.774396155629557,
"grad_norm": 1.0950493812561035,
"learning_rate": 4.986494018779746e-06,
"loss": 1.281,
"step": 564000
},
{
"epoch": 23.79547274796611,
"grad_norm": 1.1132159233093262,
"learning_rate": 4.900741757063843e-06,
"loss": 1.2805,
"step": 564500
},
{
"epoch": 23.81654934030266,
"grad_norm": 1.103472352027893,
"learning_rate": 4.81498949534794e-06,
"loss": 1.2781,
"step": 565000
},
{
"epoch": 23.83762593263921,
"grad_norm": 1.1023170948028564,
"learning_rate": 4.729237233632038e-06,
"loss": 1.2777,
"step": 565500
},
{
"epoch": 23.85870252497576,
"grad_norm": 1.0755901336669922,
"learning_rate": 4.643484971916134e-06,
"loss": 1.2806,
"step": 566000
},
{
"epoch": 23.879779117312314,
"grad_norm": 1.0853792428970337,
"learning_rate": 4.557732710200232e-06,
"loss": 1.2786,
"step": 566500
},
{
"epoch": 23.900855709648862,
"grad_norm": 1.1134154796600342,
"learning_rate": 4.471980448484329e-06,
"loss": 1.2793,
"step": 567000
},
{
"epoch": 23.921932301985414,
"grad_norm": 1.1309688091278076,
"learning_rate": 4.386228186768427e-06,
"loss": 1.2781,
"step": 567500
},
{
"epoch": 23.943008894321967,
"grad_norm": 1.1107990741729736,
"learning_rate": 4.300475925052524e-06,
"loss": 1.2792,
"step": 568000
},
{
"epoch": 23.96408548665852,
"grad_norm": 1.097574234008789,
"learning_rate": 4.214723663336621e-06,
"loss": 1.2827,
"step": 568500
},
{
"epoch": 23.985162078995067,
"grad_norm": 1.1118608713150024,
"learning_rate": 4.128971401620718e-06,
"loss": 1.2775,
"step": 569000
},
{
"epoch": 24.00623867133162,
"grad_norm": 1.0952686071395874,
"learning_rate": 4.043219139904815e-06,
"loss": 1.28,
"step": 569500
},
{
"epoch": 24.02731526366817,
"grad_norm": 1.0767697095870972,
"learning_rate": 3.957466878188912e-06,
"loss": 1.2787,
"step": 570000
},
{
"epoch": 24.02731526366817,
"eval_accuracy": 0.752429292210267,
"eval_loss": 1.080482840538025,
"eval_runtime": 180.2755,
"eval_samples_per_second": 669.564,
"eval_steps_per_second": 6.978,
"step": 570000
},
{
"epoch": 24.04839185600472,
"grad_norm": 1.1088603734970093,
"learning_rate": 3.87171461647301e-06,
"loss": 1.2805,
"step": 570500
},
{
"epoch": 24.069468448341272,
"grad_norm": 1.078587293624878,
"learning_rate": 3.7859623547571067e-06,
"loss": 1.2754,
"step": 571000
},
{
"epoch": 24.090545040677824,
"grad_norm": 1.1069971323013306,
"learning_rate": 3.7002100930412046e-06,
"loss": 1.2753,
"step": 571500
},
{
"epoch": 24.111621633014373,
"grad_norm": 1.0973984003067017,
"learning_rate": 3.614457831325301e-06,
"loss": 1.2818,
"step": 572000
},
{
"epoch": 24.132698225350925,
"grad_norm": 1.1187726259231567,
"learning_rate": 3.5287055696093987e-06,
"loss": 1.2781,
"step": 572500
},
{
"epoch": 24.153774817687477,
"grad_norm": 1.0782705545425415,
"learning_rate": 3.4429533078934957e-06,
"loss": 1.2782,
"step": 573000
},
{
"epoch": 24.174851410024026,
"grad_norm": 1.078142523765564,
"learning_rate": 3.357201046177593e-06,
"loss": 1.2806,
"step": 573500
},
{
"epoch": 24.195928002360578,
"grad_norm": 1.076899528503418,
"learning_rate": 3.2714487844616906e-06,
"loss": 1.2799,
"step": 574000
},
{
"epoch": 24.21700459469713,
"grad_norm": 1.0901553630828857,
"learning_rate": 3.1856965227457872e-06,
"loss": 1.2789,
"step": 574500
},
{
"epoch": 24.238081187033682,
"grad_norm": 1.0841399431228638,
"learning_rate": 3.099944261029885e-06,
"loss": 1.2819,
"step": 575000
},
{
"epoch": 24.25915777937023,
"grad_norm": 1.1338623762130737,
"learning_rate": 3.014191999313982e-06,
"loss": 1.2781,
"step": 575500
},
{
"epoch": 24.280234371706783,
"grad_norm": 1.0909664630889893,
"learning_rate": 2.928439737598079e-06,
"loss": 1.2757,
"step": 576000
},
{
"epoch": 24.301310964043335,
"grad_norm": 1.092181921005249,
"learning_rate": 2.8426874758821767e-06,
"loss": 1.2778,
"step": 576500
},
{
"epoch": 24.322387556379883,
"grad_norm": 1.1049392223358154,
"learning_rate": 2.7569352141662737e-06,
"loss": 1.2739,
"step": 577000
},
{
"epoch": 24.343464148716436,
"grad_norm": 1.1252498626708984,
"learning_rate": 2.6711829524503707e-06,
"loss": 1.2748,
"step": 577500
},
{
"epoch": 24.364540741052988,
"grad_norm": 1.1361205577850342,
"learning_rate": 2.585430690734468e-06,
"loss": 1.2737,
"step": 578000
},
{
"epoch": 24.385617333389536,
"grad_norm": 1.105175256729126,
"learning_rate": 2.4996784290185657e-06,
"loss": 1.2766,
"step": 578500
},
{
"epoch": 24.40669392572609,
"grad_norm": 1.115120530128479,
"learning_rate": 2.4139261673026627e-06,
"loss": 1.2742,
"step": 579000
},
{
"epoch": 24.42777051806264,
"grad_norm": 1.1030393838882446,
"learning_rate": 2.3281739055867597e-06,
"loss": 1.2755,
"step": 579500
},
{
"epoch": 24.44884711039919,
"grad_norm": 1.100075364112854,
"learning_rate": 2.242421643870857e-06,
"loss": 1.276,
"step": 580000
},
{
"epoch": 24.44884711039919,
"eval_accuracy": 0.7532042741672275,
"eval_loss": 1.0764633417129517,
"eval_runtime": 179.8688,
"eval_samples_per_second": 671.078,
"eval_steps_per_second": 6.994,
"step": 580000
},
{
"epoch": 24.46992370273574,
"grad_norm": 1.1108529567718506,
"learning_rate": 2.1566693821549547e-06,
"loss": 1.2753,
"step": 580500
},
{
"epoch": 24.491000295072293,
"grad_norm": 1.1284890174865723,
"learning_rate": 2.0709171204390517e-06,
"loss": 1.2782,
"step": 581000
},
{
"epoch": 24.512076887408845,
"grad_norm": 1.110996127128601,
"learning_rate": 1.985164858723149e-06,
"loss": 1.2733,
"step": 581500
},
{
"epoch": 24.533153479745394,
"grad_norm": 1.1034977436065674,
"learning_rate": 1.8994125970072462e-06,
"loss": 1.2784,
"step": 582000
},
{
"epoch": 24.554230072081946,
"grad_norm": 1.100600242614746,
"learning_rate": 1.8136603352913432e-06,
"loss": 1.2742,
"step": 582500
},
{
"epoch": 24.5753066644185,
"grad_norm": 1.0803606510162354,
"learning_rate": 1.7279080735754405e-06,
"loss": 1.2778,
"step": 583000
},
{
"epoch": 24.596383256755047,
"grad_norm": 1.1201647520065308,
"learning_rate": 1.642155811859538e-06,
"loss": 1.2771,
"step": 583500
},
{
"epoch": 24.6174598490916,
"grad_norm": 1.1072803735733032,
"learning_rate": 1.5564035501436352e-06,
"loss": 1.275,
"step": 584000
},
{
"epoch": 24.63853644142815,
"grad_norm": 1.1152465343475342,
"learning_rate": 1.4706512884277324e-06,
"loss": 1.2766,
"step": 584500
},
{
"epoch": 24.6596130337647,
"grad_norm": 1.0824260711669922,
"learning_rate": 1.3848990267118295e-06,
"loss": 1.2729,
"step": 585000
},
{
"epoch": 24.680689626101252,
"grad_norm": 1.0788400173187256,
"learning_rate": 1.299146764995927e-06,
"loss": 1.2752,
"step": 585500
},
{
"epoch": 24.701766218437804,
"grad_norm": 1.1227959394454956,
"learning_rate": 1.213394503280024e-06,
"loss": 1.2735,
"step": 586000
},
{
"epoch": 24.722842810774353,
"grad_norm": 1.0858204364776611,
"learning_rate": 1.1276422415641212e-06,
"loss": 1.2757,
"step": 586500
},
{
"epoch": 24.743919403110905,
"grad_norm": 1.0969204902648926,
"learning_rate": 1.0418899798482187e-06,
"loss": 1.2769,
"step": 587000
},
{
"epoch": 24.764995995447457,
"grad_norm": 1.1148242950439453,
"learning_rate": 9.561377181323157e-07,
"loss": 1.2727,
"step": 587500
},
{
"epoch": 24.786072587784005,
"grad_norm": 1.107650637626648,
"learning_rate": 8.70385456416413e-07,
"loss": 1.2741,
"step": 588000
},
{
"epoch": 24.807149180120557,
"grad_norm": 1.0647169351577759,
"learning_rate": 7.846331947005103e-07,
"loss": 1.2764,
"step": 588500
},
{
"epoch": 24.82822577245711,
"grad_norm": 1.1068612337112427,
"learning_rate": 6.988809329846075e-07,
"loss": 1.2714,
"step": 589000
},
{
"epoch": 24.84930236479366,
"grad_norm": 1.1318004131317139,
"learning_rate": 6.131286712687047e-07,
"loss": 1.2755,
"step": 589500
},
{
"epoch": 24.87037895713021,
"grad_norm": 1.0909773111343384,
"learning_rate": 5.27376409552802e-07,
"loss": 1.2732,
"step": 590000
},
{
"epoch": 24.87037895713021,
"eval_accuracy": 0.7529894922525033,
"eval_loss": 1.0770493745803833,
"eval_runtime": 176.7111,
"eval_samples_per_second": 683.07,
"eval_steps_per_second": 7.119,
"step": 590000
},
{
"epoch": 24.891455549466762,
"grad_norm": 1.0952293872833252,
"learning_rate": 4.4162414783689917e-07,
"loss": 1.2738,
"step": 590500
},
{
"epoch": 24.912532141803315,
"grad_norm": 1.0905896425247192,
"learning_rate": 3.558718861209964e-07,
"loss": 1.2737,
"step": 591000
},
{
"epoch": 24.933608734139863,
"grad_norm": 1.117954969406128,
"learning_rate": 2.701196244050937e-07,
"loss": 1.2745,
"step": 591500
},
{
"epoch": 24.954685326476415,
"grad_norm": 1.145927906036377,
"learning_rate": 1.8436736268919094e-07,
"loss": 1.2745,
"step": 592000
},
{
"epoch": 24.975761918812967,
"grad_norm": 1.0901849269866943,
"learning_rate": 9.861510097328816e-08,
"loss": 1.2732,
"step": 592500
},
{
"epoch": 24.996838511149516,
"grad_norm": 1.0881909132003784,
"learning_rate": 1.2862839257385414e-08,
"loss": 1.2742,
"step": 593000
},
{
"epoch": 25.0,
"step": 593075,
"total_flos": 1.498518436978944e+19,
"train_loss": 2.334269580478457,
"train_runtime": 190366.3414,
"train_samples_per_second": 299.074,
"train_steps_per_second": 3.115
}
],
"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": 1.498518436978944e+19,
"train_batch_size": 96,
"trial_name": null,
"trial_params": null
}