bert_base_lda_20_v1_book / trainer_state.json
gokulsrinivasagan's picture
End of training
a2e374a verified
{
"best_metric": 3.409717559814453,
"best_model_checkpoint": "bert_base_lda_20_v1_book/checkpoint-580000",
"epoch": 25.0,
"eval_steps": 10000,
"global_step": 593075,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.021076592336551025,
"grad_norm": 2.357170820236206,
"learning_rate": 5e-06,
"loss": 11.859,
"step": 500
},
{
"epoch": 0.04215318467310205,
"grad_norm": 1.0412052869796753,
"learning_rate": 1e-05,
"loss": 9.664,
"step": 1000
},
{
"epoch": 0.06322977700965308,
"grad_norm": 1.1233024597167969,
"learning_rate": 1.5e-05,
"loss": 9.0294,
"step": 1500
},
{
"epoch": 0.0843063693462041,
"grad_norm": 1.14918053150177,
"learning_rate": 2e-05,
"loss": 8.9094,
"step": 2000
},
{
"epoch": 0.10538296168275513,
"grad_norm": 1.027387022972107,
"learning_rate": 2.5e-05,
"loss": 8.8511,
"step": 2500
},
{
"epoch": 0.12645955401930617,
"grad_norm": 0.9729121327400208,
"learning_rate": 3e-05,
"loss": 8.8121,
"step": 3000
},
{
"epoch": 0.14753614635585718,
"grad_norm": 0.9469872713088989,
"learning_rate": 3.5e-05,
"loss": 8.7827,
"step": 3500
},
{
"epoch": 0.1686127386924082,
"grad_norm": 0.9876546859741211,
"learning_rate": 4e-05,
"loss": 8.7554,
"step": 4000
},
{
"epoch": 0.18968933102895924,
"grad_norm": 1.062038540840149,
"learning_rate": 4.5e-05,
"loss": 8.7308,
"step": 4500
},
{
"epoch": 0.21076592336551025,
"grad_norm": 1.0452961921691895,
"learning_rate": 5e-05,
"loss": 8.7149,
"step": 5000
},
{
"epoch": 0.2318425157020613,
"grad_norm": 0.9361803531646729,
"learning_rate": 5.500000000000001e-05,
"loss": 8.7038,
"step": 5500
},
{
"epoch": 0.25291910803861234,
"grad_norm": 1.0404396057128906,
"learning_rate": 6e-05,
"loss": 8.685,
"step": 6000
},
{
"epoch": 0.27399570037516335,
"grad_norm": 0.9452637434005737,
"learning_rate": 6.500000000000001e-05,
"loss": 8.6653,
"step": 6500
},
{
"epoch": 0.29507229271171437,
"grad_norm": 1.1130284070968628,
"learning_rate": 7e-05,
"loss": 8.6363,
"step": 7000
},
{
"epoch": 0.3161488850482654,
"grad_norm": 1.2457925081253052,
"learning_rate": 7.500000000000001e-05,
"loss": 8.5986,
"step": 7500
},
{
"epoch": 0.3372254773848164,
"grad_norm": 1.1786009073257446,
"learning_rate": 8e-05,
"loss": 8.5247,
"step": 8000
},
{
"epoch": 0.35830206972136747,
"grad_norm": 1.6338282823562622,
"learning_rate": 8.5e-05,
"loss": 8.3813,
"step": 8500
},
{
"epoch": 0.3793786620579185,
"grad_norm": 1.7593787908554077,
"learning_rate": 9e-05,
"loss": 8.2689,
"step": 9000
},
{
"epoch": 0.4004552543944695,
"grad_norm": 1.4493446350097656,
"learning_rate": 9.5e-05,
"loss": 8.217,
"step": 9500
},
{
"epoch": 0.4215318467310205,
"grad_norm": 1.5118393898010254,
"learning_rate": 0.0001,
"loss": 8.1966,
"step": 10000
},
{
"epoch": 0.4215318467310205,
"eval_accuracy": 0.16399807200913816,
"eval_loss": 8.00486946105957,
"eval_runtime": 343.0099,
"eval_samples_per_second": 351.902,
"eval_steps_per_second": 3.668,
"step": 10000
},
{
"epoch": 0.4426084390675716,
"grad_norm": 0.9923507571220398,
"learning_rate": 9.99142477382841e-05,
"loss": 8.1755,
"step": 10500
},
{
"epoch": 0.4636850314041226,
"grad_norm": 1.1836919784545898,
"learning_rate": 9.98284954765682e-05,
"loss": 8.1626,
"step": 11000
},
{
"epoch": 0.4847616237406736,
"grad_norm": 1.1801451444625854,
"learning_rate": 9.97427432148523e-05,
"loss": 8.1573,
"step": 11500
},
{
"epoch": 0.5058382160772247,
"grad_norm": 1.0431355237960815,
"learning_rate": 9.96569909531364e-05,
"loss": 8.1442,
"step": 12000
},
{
"epoch": 0.5269148084137757,
"grad_norm": 1.138790488243103,
"learning_rate": 9.957123869142048e-05,
"loss": 8.1365,
"step": 12500
},
{
"epoch": 0.5479914007503267,
"grad_norm": 0.9565016627311707,
"learning_rate": 9.94854864297046e-05,
"loss": 8.1285,
"step": 13000
},
{
"epoch": 0.5690679930868777,
"grad_norm": 0.9410801529884338,
"learning_rate": 9.939973416798869e-05,
"loss": 8.1173,
"step": 13500
},
{
"epoch": 0.5901445854234287,
"grad_norm": 0.8389434218406677,
"learning_rate": 9.931398190627278e-05,
"loss": 8.112,
"step": 14000
},
{
"epoch": 0.6112211777599798,
"grad_norm": 1.0888636112213135,
"learning_rate": 9.922822964455688e-05,
"loss": 8.1028,
"step": 14500
},
{
"epoch": 0.6322977700965308,
"grad_norm": 0.9818998575210571,
"learning_rate": 9.914247738284099e-05,
"loss": 8.0992,
"step": 15000
},
{
"epoch": 0.6533743624330818,
"grad_norm": 0.9017472863197327,
"learning_rate": 9.905672512112507e-05,
"loss": 8.0873,
"step": 15500
},
{
"epoch": 0.6744509547696328,
"grad_norm": 0.8610674142837524,
"learning_rate": 9.897097285940917e-05,
"loss": 8.0862,
"step": 16000
},
{
"epoch": 0.6955275471061839,
"grad_norm": 0.9559035897254944,
"learning_rate": 9.888522059769327e-05,
"loss": 8.0787,
"step": 16500
},
{
"epoch": 0.7166041394427349,
"grad_norm": 0.8893306851387024,
"learning_rate": 9.879946833597737e-05,
"loss": 8.0724,
"step": 17000
},
{
"epoch": 0.737680731779286,
"grad_norm": 0.9684920310974121,
"learning_rate": 9.871371607426147e-05,
"loss": 8.0634,
"step": 17500
},
{
"epoch": 0.758757324115837,
"grad_norm": 0.9390775561332703,
"learning_rate": 9.862796381254556e-05,
"loss": 8.0674,
"step": 18000
},
{
"epoch": 0.779833916452388,
"grad_norm": 0.8043727874755859,
"learning_rate": 9.854221155082965e-05,
"loss": 8.0574,
"step": 18500
},
{
"epoch": 0.800910508788939,
"grad_norm": 0.8214758038520813,
"learning_rate": 9.845645928911376e-05,
"loss": 8.0585,
"step": 19000
},
{
"epoch": 0.82198710112549,
"grad_norm": 0.9023016095161438,
"learning_rate": 9.837070702739786e-05,
"loss": 8.0537,
"step": 19500
},
{
"epoch": 0.843063693462041,
"grad_norm": 0.8249906301498413,
"learning_rate": 9.828495476568194e-05,
"loss": 8.0473,
"step": 20000
},
{
"epoch": 0.843063693462041,
"eval_accuracy": 0.16556523822650848,
"eval_loss": 7.8938775062561035,
"eval_runtime": 334.5883,
"eval_samples_per_second": 360.76,
"eval_steps_per_second": 3.76,
"step": 20000
},
{
"epoch": 0.864140285798592,
"grad_norm": 0.8509041666984558,
"learning_rate": 9.819920250396604e-05,
"loss": 8.041,
"step": 20500
},
{
"epoch": 0.8852168781351432,
"grad_norm": 0.7805004715919495,
"learning_rate": 9.811345024225015e-05,
"loss": 8.0358,
"step": 21000
},
{
"epoch": 0.9062934704716942,
"grad_norm": 0.8931642770767212,
"learning_rate": 9.802769798053424e-05,
"loss": 8.0331,
"step": 21500
},
{
"epoch": 0.9273700628082452,
"grad_norm": 0.8356170654296875,
"learning_rate": 9.794194571881834e-05,
"loss": 8.0289,
"step": 22000
},
{
"epoch": 0.9484466551447962,
"grad_norm": 0.9137856960296631,
"learning_rate": 9.785619345710244e-05,
"loss": 8.0234,
"step": 22500
},
{
"epoch": 0.9695232474813472,
"grad_norm": 0.8502079844474792,
"learning_rate": 9.777044119538653e-05,
"loss": 8.0183,
"step": 23000
},
{
"epoch": 0.9905998398178982,
"grad_norm": 0.8751811385154724,
"learning_rate": 9.768468893367063e-05,
"loss": 8.0137,
"step": 23500
},
{
"epoch": 1.0116764321544494,
"grad_norm": 0.8679917454719543,
"learning_rate": 9.759893667195473e-05,
"loss": 8.0114,
"step": 24000
},
{
"epoch": 1.0327530244910004,
"grad_norm": 0.8936434388160706,
"learning_rate": 9.751318441023883e-05,
"loss": 8.0072,
"step": 24500
},
{
"epoch": 1.0538296168275514,
"grad_norm": 0.7955121397972107,
"learning_rate": 9.742743214852293e-05,
"loss": 8.0047,
"step": 25000
},
{
"epoch": 1.0749062091641024,
"grad_norm": 1.005022406578064,
"learning_rate": 9.734167988680703e-05,
"loss": 7.9957,
"step": 25500
},
{
"epoch": 1.0959828015006534,
"grad_norm": 0.847980260848999,
"learning_rate": 9.725592762509111e-05,
"loss": 7.9941,
"step": 26000
},
{
"epoch": 1.1170593938372044,
"grad_norm": 0.7947870492935181,
"learning_rate": 9.717017536337522e-05,
"loss": 7.9916,
"step": 26500
},
{
"epoch": 1.1381359861737554,
"grad_norm": 0.8341640830039978,
"learning_rate": 9.708442310165932e-05,
"loss": 7.9932,
"step": 27000
},
{
"epoch": 1.1592125785103065,
"grad_norm": 0.714772641658783,
"learning_rate": 9.69986708399434e-05,
"loss": 7.9868,
"step": 27500
},
{
"epoch": 1.1802891708468575,
"grad_norm": 0.8826730251312256,
"learning_rate": 9.69129185782275e-05,
"loss": 7.9889,
"step": 28000
},
{
"epoch": 1.2013657631834085,
"grad_norm": 0.8229885101318359,
"learning_rate": 9.68271663165116e-05,
"loss": 7.9781,
"step": 28500
},
{
"epoch": 1.2224423555199595,
"grad_norm": 1.1180338859558105,
"learning_rate": 9.67414140547957e-05,
"loss": 7.8554,
"step": 29000
},
{
"epoch": 1.2435189478565105,
"grad_norm": 1.3187551498413086,
"learning_rate": 9.66556617930798e-05,
"loss": 7.2134,
"step": 29500
},
{
"epoch": 1.2645955401930615,
"grad_norm": 1.364199161529541,
"learning_rate": 9.656990953136388e-05,
"loss": 6.9645,
"step": 30000
},
{
"epoch": 1.2645955401930615,
"eval_accuracy": 0.3351688747410192,
"eval_loss": 6.468147277832031,
"eval_runtime": 349.8603,
"eval_samples_per_second": 345.012,
"eval_steps_per_second": 3.596,
"step": 30000
},
{
"epoch": 1.2856721325296125,
"grad_norm": 1.3416740894317627,
"learning_rate": 9.6484157269648e-05,
"loss": 6.7817,
"step": 30500
},
{
"epoch": 1.3067487248661636,
"grad_norm": 1.3637932538986206,
"learning_rate": 9.639840500793209e-05,
"loss": 6.6448,
"step": 31000
},
{
"epoch": 1.3278253172027146,
"grad_norm": 1.4058624505996704,
"learning_rate": 9.631265274621618e-05,
"loss": 6.5347,
"step": 31500
},
{
"epoch": 1.3489019095392658,
"grad_norm": 1.3866232633590698,
"learning_rate": 9.622690048450028e-05,
"loss": 6.4448,
"step": 32000
},
{
"epoch": 1.3699785018758166,
"grad_norm": 1.4489511251449585,
"learning_rate": 9.614114822278439e-05,
"loss": 6.3541,
"step": 32500
},
{
"epoch": 1.3910550942123678,
"grad_norm": 1.507181167602539,
"learning_rate": 9.605539596106847e-05,
"loss": 6.2785,
"step": 33000
},
{
"epoch": 1.4121316865489189,
"grad_norm": 1.4161540269851685,
"learning_rate": 9.596964369935257e-05,
"loss": 6.1937,
"step": 33500
},
{
"epoch": 1.4332082788854699,
"grad_norm": 1.4161708354949951,
"learning_rate": 9.588389143763667e-05,
"loss": 6.0939,
"step": 34000
},
{
"epoch": 1.4542848712220209,
"grad_norm": 1.41823410987854,
"learning_rate": 9.579813917592077e-05,
"loss": 5.9064,
"step": 34500
},
{
"epoch": 1.475361463558572,
"grad_norm": 1.3344123363494873,
"learning_rate": 9.571238691420487e-05,
"loss": 5.6962,
"step": 35000
},
{
"epoch": 1.496438055895123,
"grad_norm": 1.2387702465057373,
"learning_rate": 9.562663465248896e-05,
"loss": 5.5429,
"step": 35500
},
{
"epoch": 1.517514648231674,
"grad_norm": 1.3192780017852783,
"learning_rate": 9.554088239077306e-05,
"loss": 5.4148,
"step": 36000
},
{
"epoch": 1.538591240568225,
"grad_norm": 1.2223302125930786,
"learning_rate": 9.545513012905716e-05,
"loss": 5.3172,
"step": 36500
},
{
"epoch": 1.559667832904776,
"grad_norm": 1.2394053936004639,
"learning_rate": 9.536937786734126e-05,
"loss": 5.2415,
"step": 37000
},
{
"epoch": 1.580744425241327,
"grad_norm": 1.3220828771591187,
"learning_rate": 9.528362560562534e-05,
"loss": 5.1674,
"step": 37500
},
{
"epoch": 1.601821017577878,
"grad_norm": 1.2945057153701782,
"learning_rate": 9.519787334390946e-05,
"loss": 5.1052,
"step": 38000
},
{
"epoch": 1.622897609914429,
"grad_norm": 1.278004765510559,
"learning_rate": 9.511212108219355e-05,
"loss": 5.05,
"step": 38500
},
{
"epoch": 1.64397420225098,
"grad_norm": 1.2774810791015625,
"learning_rate": 9.502636882047764e-05,
"loss": 5.0028,
"step": 39000
},
{
"epoch": 1.6650507945875312,
"grad_norm": 1.2342504262924194,
"learning_rate": 9.494061655876174e-05,
"loss": 4.9636,
"step": 39500
},
{
"epoch": 1.686127386924082,
"grad_norm": 1.199815273284912,
"learning_rate": 9.485486429704585e-05,
"loss": 4.9295,
"step": 40000
},
{
"epoch": 1.686127386924082,
"eval_accuracy": 0.5784786088377104,
"eval_loss": 4.570897102355957,
"eval_runtime": 370.9847,
"eval_samples_per_second": 325.367,
"eval_steps_per_second": 3.391,
"step": 40000
},
{
"epoch": 1.7072039792606333,
"grad_norm": 1.3310173749923706,
"learning_rate": 9.476911203532993e-05,
"loss": 4.8914,
"step": 40500
},
{
"epoch": 1.728280571597184,
"grad_norm": 1.2115612030029297,
"learning_rate": 9.468335977361403e-05,
"loss": 4.8619,
"step": 41000
},
{
"epoch": 1.7493571639337353,
"grad_norm": 1.2032768726348877,
"learning_rate": 9.459760751189813e-05,
"loss": 4.8321,
"step": 41500
},
{
"epoch": 1.770433756270286,
"grad_norm": 1.1947598457336426,
"learning_rate": 9.451185525018223e-05,
"loss": 4.8028,
"step": 42000
},
{
"epoch": 1.7915103486068373,
"grad_norm": 1.1941403150558472,
"learning_rate": 9.442610298846633e-05,
"loss": 4.7791,
"step": 42500
},
{
"epoch": 1.8125869409433881,
"grad_norm": 1.2083755731582642,
"learning_rate": 9.434035072675043e-05,
"loss": 4.7542,
"step": 43000
},
{
"epoch": 1.8336635332799394,
"grad_norm": 1.098036527633667,
"learning_rate": 9.425459846503451e-05,
"loss": 4.7311,
"step": 43500
},
{
"epoch": 1.8547401256164904,
"grad_norm": 1.128735065460205,
"learning_rate": 9.416884620331862e-05,
"loss": 4.7107,
"step": 44000
},
{
"epoch": 1.8758167179530414,
"grad_norm": 1.221762776374817,
"learning_rate": 9.408309394160272e-05,
"loss": 4.6956,
"step": 44500
},
{
"epoch": 1.8968933102895924,
"grad_norm": 1.217251181602478,
"learning_rate": 9.39973416798868e-05,
"loss": 4.6711,
"step": 45000
},
{
"epoch": 1.9179699026261434,
"grad_norm": 1.153186321258545,
"learning_rate": 9.39115894181709e-05,
"loss": 4.6598,
"step": 45500
},
{
"epoch": 1.9390464949626944,
"grad_norm": 1.197913408279419,
"learning_rate": 9.382583715645502e-05,
"loss": 4.6402,
"step": 46000
},
{
"epoch": 1.9601230872992454,
"grad_norm": 1.1512069702148438,
"learning_rate": 9.37400848947391e-05,
"loss": 4.6238,
"step": 46500
},
{
"epoch": 1.9811996796357965,
"grad_norm": 1.2417347431182861,
"learning_rate": 9.36543326330232e-05,
"loss": 4.6145,
"step": 47000
},
{
"epoch": 2.0022762719723475,
"grad_norm": 1.1664767265319824,
"learning_rate": 9.35685803713073e-05,
"loss": 4.6027,
"step": 47500
},
{
"epoch": 2.0233528643088987,
"grad_norm": 1.143926739692688,
"learning_rate": 9.34828281095914e-05,
"loss": 4.5802,
"step": 48000
},
{
"epoch": 2.0444294566454495,
"grad_norm": 1.1860380172729492,
"learning_rate": 9.339707584787549e-05,
"loss": 4.5707,
"step": 48500
},
{
"epoch": 2.0655060489820007,
"grad_norm": 1.1656445264816284,
"learning_rate": 9.331132358615959e-05,
"loss": 4.5542,
"step": 49000
},
{
"epoch": 2.0865826413185515,
"grad_norm": 1.1591306924819946,
"learning_rate": 9.322557132444369e-05,
"loss": 4.5432,
"step": 49500
},
{
"epoch": 2.1076592336551028,
"grad_norm": 1.2601778507232666,
"learning_rate": 9.313981906272779e-05,
"loss": 4.5377,
"step": 50000
},
{
"epoch": 2.1076592336551028,
"eval_accuracy": 0.6289472464552175,
"eval_loss": 4.234973907470703,
"eval_runtime": 460.9963,
"eval_samples_per_second": 261.837,
"eval_steps_per_second": 2.729,
"step": 50000
},
{
"epoch": 2.1287358259916536,
"grad_norm": 1.1379075050354004,
"learning_rate": 9.305406680101189e-05,
"loss": 4.5304,
"step": 50500
},
{
"epoch": 2.149812418328205,
"grad_norm": 1.157738208770752,
"learning_rate": 9.296831453929597e-05,
"loss": 4.5159,
"step": 51000
},
{
"epoch": 2.1708890106647556,
"grad_norm": 1.0959041118621826,
"learning_rate": 9.288256227758008e-05,
"loss": 4.5034,
"step": 51500
},
{
"epoch": 2.191965603001307,
"grad_norm": 1.145795464515686,
"learning_rate": 9.279681001586418e-05,
"loss": 4.4973,
"step": 52000
},
{
"epoch": 2.2130421953378576,
"grad_norm": 1.0979059934616089,
"learning_rate": 9.271105775414827e-05,
"loss": 4.4828,
"step": 52500
},
{
"epoch": 2.234118787674409,
"grad_norm": 1.1925787925720215,
"learning_rate": 9.262530549243236e-05,
"loss": 4.4783,
"step": 53000
},
{
"epoch": 2.2551953800109596,
"grad_norm": 1.0924298763275146,
"learning_rate": 9.253955323071646e-05,
"loss": 4.4704,
"step": 53500
},
{
"epoch": 2.276271972347511,
"grad_norm": 1.1050338745117188,
"learning_rate": 9.245380096900056e-05,
"loss": 4.4642,
"step": 54000
},
{
"epoch": 2.2973485646840617,
"grad_norm": 1.1446146965026855,
"learning_rate": 9.236804870728466e-05,
"loss": 4.4472,
"step": 54500
},
{
"epoch": 2.318425157020613,
"grad_norm": 1.0954008102416992,
"learning_rate": 9.228229644556876e-05,
"loss": 4.4452,
"step": 55000
},
{
"epoch": 2.339501749357164,
"grad_norm": 1.1642777919769287,
"learning_rate": 9.219654418385286e-05,
"loss": 4.4328,
"step": 55500
},
{
"epoch": 2.360578341693715,
"grad_norm": 1.138856291770935,
"learning_rate": 9.211079192213695e-05,
"loss": 4.4329,
"step": 56000
},
{
"epoch": 2.381654934030266,
"grad_norm": 1.106980562210083,
"learning_rate": 9.202503966042105e-05,
"loss": 4.4231,
"step": 56500
},
{
"epoch": 2.402731526366817,
"grad_norm": 1.1022531986236572,
"learning_rate": 9.193928739870514e-05,
"loss": 4.4125,
"step": 57000
},
{
"epoch": 2.423808118703368,
"grad_norm": 1.1264935731887817,
"learning_rate": 9.185353513698925e-05,
"loss": 4.4072,
"step": 57500
},
{
"epoch": 2.444884711039919,
"grad_norm": 1.1125894784927368,
"learning_rate": 9.176778287527335e-05,
"loss": 4.4011,
"step": 58000
},
{
"epoch": 2.4659613033764702,
"grad_norm": 1.1464273929595947,
"learning_rate": 9.168203061355743e-05,
"loss": 4.3953,
"step": 58500
},
{
"epoch": 2.487037895713021,
"grad_norm": 1.0863112211227417,
"learning_rate": 9.159627835184153e-05,
"loss": 4.3881,
"step": 59000
},
{
"epoch": 2.5081144880495723,
"grad_norm": 1.1034303903579712,
"learning_rate": 9.151052609012564e-05,
"loss": 4.3846,
"step": 59500
},
{
"epoch": 2.529191080386123,
"grad_norm": 1.1298364400863647,
"learning_rate": 9.142477382840973e-05,
"loss": 4.3734,
"step": 60000
},
{
"epoch": 2.529191080386123,
"eval_accuracy": 0.6526964298927005,
"eval_loss": 4.085425853729248,
"eval_runtime": 343.4581,
"eval_samples_per_second": 351.443,
"eval_steps_per_second": 3.663,
"step": 60000
},
{
"epoch": 2.5502676727226743,
"grad_norm": 1.1539239883422852,
"learning_rate": 9.133902156669383e-05,
"loss": 4.3669,
"step": 60500
},
{
"epoch": 2.571344265059225,
"grad_norm": 1.099463701248169,
"learning_rate": 9.125326930497792e-05,
"loss": 4.3615,
"step": 61000
},
{
"epoch": 2.5924208573957763,
"grad_norm": 1.05633544921875,
"learning_rate": 9.116751704326202e-05,
"loss": 4.3543,
"step": 61500
},
{
"epoch": 2.613497449732327,
"grad_norm": 1.1771094799041748,
"learning_rate": 9.108176478154612e-05,
"loss": 4.3447,
"step": 62000
},
{
"epoch": 2.6345740420688784,
"grad_norm": 1.1169254779815674,
"learning_rate": 9.09960125198302e-05,
"loss": 4.3497,
"step": 62500
},
{
"epoch": 2.655650634405429,
"grad_norm": 1.068060278892517,
"learning_rate": 9.091026025811432e-05,
"loss": 4.3384,
"step": 63000
},
{
"epoch": 2.6767272267419804,
"grad_norm": 1.1156476736068726,
"learning_rate": 9.082450799639842e-05,
"loss": 4.3327,
"step": 63500
},
{
"epoch": 2.6978038190785316,
"grad_norm": 1.0696433782577515,
"learning_rate": 9.07387557346825e-05,
"loss": 4.3306,
"step": 64000
},
{
"epoch": 2.7188804114150824,
"grad_norm": 1.098692536354065,
"learning_rate": 9.06530034729666e-05,
"loss": 4.3231,
"step": 64500
},
{
"epoch": 2.739957003751633,
"grad_norm": 1.094072937965393,
"learning_rate": 9.05672512112507e-05,
"loss": 4.317,
"step": 65000
},
{
"epoch": 2.7610335960881844,
"grad_norm": 1.0858944654464722,
"learning_rate": 9.04814989495348e-05,
"loss": 4.3122,
"step": 65500
},
{
"epoch": 2.7821101884247357,
"grad_norm": 1.1064774990081787,
"learning_rate": 9.03957466878189e-05,
"loss": 4.3121,
"step": 66000
},
{
"epoch": 2.8031867807612865,
"grad_norm": 1.0845835208892822,
"learning_rate": 9.030999442610299e-05,
"loss": 4.3012,
"step": 66500
},
{
"epoch": 2.8242633730978377,
"grad_norm": 1.1014231443405151,
"learning_rate": 9.022424216438709e-05,
"loss": 4.2945,
"step": 67000
},
{
"epoch": 2.8453399654343885,
"grad_norm": 1.0546036958694458,
"learning_rate": 9.013848990267119e-05,
"loss": 4.2909,
"step": 67500
},
{
"epoch": 2.8664165577709397,
"grad_norm": 1.0463727712631226,
"learning_rate": 9.005273764095529e-05,
"loss": 4.2854,
"step": 68000
},
{
"epoch": 2.8874931501074905,
"grad_norm": 1.094387173652649,
"learning_rate": 8.996698537923937e-05,
"loss": 4.2867,
"step": 68500
},
{
"epoch": 2.9085697424440418,
"grad_norm": 1.048219084739685,
"learning_rate": 8.988123311752348e-05,
"loss": 4.2782,
"step": 69000
},
{
"epoch": 2.9296463347805926,
"grad_norm": 1.079384684562683,
"learning_rate": 8.979548085580758e-05,
"loss": 4.2752,
"step": 69500
},
{
"epoch": 2.950722927117144,
"grad_norm": 1.056179404258728,
"learning_rate": 8.970972859409167e-05,
"loss": 4.2676,
"step": 70000
},
{
"epoch": 2.950722927117144,
"eval_accuracy": 0.666747325090597,
"eval_loss": 3.9893879890441895,
"eval_runtime": 361.2833,
"eval_samples_per_second": 334.103,
"eval_steps_per_second": 3.482,
"step": 70000
},
{
"epoch": 2.9717995194536946,
"grad_norm": 1.0606900453567505,
"learning_rate": 8.962397633237576e-05,
"loss": 4.2668,
"step": 70500
},
{
"epoch": 2.992876111790246,
"grad_norm": 1.0872896909713745,
"learning_rate": 8.953822407065988e-05,
"loss": 4.2605,
"step": 71000
},
{
"epoch": 3.0139527041267966,
"grad_norm": 1.0797181129455566,
"learning_rate": 8.945247180894396e-05,
"loss": 4.2569,
"step": 71500
},
{
"epoch": 3.035029296463348,
"grad_norm": 1.0705859661102295,
"learning_rate": 8.936671954722806e-05,
"loss": 4.2493,
"step": 72000
},
{
"epoch": 3.0561058887998986,
"grad_norm": 1.1240357160568237,
"learning_rate": 8.928096728551216e-05,
"loss": 4.2503,
"step": 72500
},
{
"epoch": 3.07718248113645,
"grad_norm": 1.0993402004241943,
"learning_rate": 8.919521502379626e-05,
"loss": 4.2413,
"step": 73000
},
{
"epoch": 3.0982590734730007,
"grad_norm": 1.0940102338790894,
"learning_rate": 8.910946276208035e-05,
"loss": 4.2422,
"step": 73500
},
{
"epoch": 3.119335665809552,
"grad_norm": 1.0469743013381958,
"learning_rate": 8.902371050036445e-05,
"loss": 4.2312,
"step": 74000
},
{
"epoch": 3.140412258146103,
"grad_norm": 1.0616792440414429,
"learning_rate": 8.893795823864855e-05,
"loss": 4.235,
"step": 74500
},
{
"epoch": 3.161488850482654,
"grad_norm": 1.1091012954711914,
"learning_rate": 8.885220597693265e-05,
"loss": 4.2289,
"step": 75000
},
{
"epoch": 3.182565442819205,
"grad_norm": 1.0863980054855347,
"learning_rate": 8.876645371521675e-05,
"loss": 4.2239,
"step": 75500
},
{
"epoch": 3.203642035155756,
"grad_norm": 1.0484535694122314,
"learning_rate": 8.868070145350083e-05,
"loss": 4.2219,
"step": 76000
},
{
"epoch": 3.224718627492307,
"grad_norm": 1.0933194160461426,
"learning_rate": 8.859494919178493e-05,
"loss": 4.2209,
"step": 76500
},
{
"epoch": 3.245795219828858,
"grad_norm": 1.0987322330474854,
"learning_rate": 8.850919693006904e-05,
"loss": 4.2145,
"step": 77000
},
{
"epoch": 3.2668718121654092,
"grad_norm": 1.0621806383132935,
"learning_rate": 8.842344466835313e-05,
"loss": 4.2105,
"step": 77500
},
{
"epoch": 3.28794840450196,
"grad_norm": 1.0601756572723389,
"learning_rate": 8.833769240663723e-05,
"loss": 4.2033,
"step": 78000
},
{
"epoch": 3.3090249968385113,
"grad_norm": 1.0532214641571045,
"learning_rate": 8.825194014492132e-05,
"loss": 4.2042,
"step": 78500
},
{
"epoch": 3.330101589175062,
"grad_norm": 1.0821919441223145,
"learning_rate": 8.816618788320542e-05,
"loss": 4.2022,
"step": 79000
},
{
"epoch": 3.3511781815116133,
"grad_norm": 1.1643075942993164,
"learning_rate": 8.808043562148952e-05,
"loss": 4.1992,
"step": 79500
},
{
"epoch": 3.372254773848164,
"grad_norm": 1.0541510581970215,
"learning_rate": 8.799468335977362e-05,
"loss": 4.1937,
"step": 80000
},
{
"epoch": 3.372254773848164,
"eval_accuracy": 0.6772322911600334,
"eval_loss": 3.921835422515869,
"eval_runtime": 394.8934,
"eval_samples_per_second": 305.667,
"eval_steps_per_second": 3.186,
"step": 80000
},
{
"epoch": 3.3933313661847153,
"grad_norm": 1.0754112005233765,
"learning_rate": 8.790893109805772e-05,
"loss": 4.1917,
"step": 80500
},
{
"epoch": 3.414407958521266,
"grad_norm": 1.0529427528381348,
"learning_rate": 8.782317883634182e-05,
"loss": 4.1886,
"step": 81000
},
{
"epoch": 3.4354845508578173,
"grad_norm": 1.0568559169769287,
"learning_rate": 8.773742657462591e-05,
"loss": 4.1847,
"step": 81500
},
{
"epoch": 3.456561143194368,
"grad_norm": 1.0922819375991821,
"learning_rate": 8.765167431291e-05,
"loss": 4.1836,
"step": 82000
},
{
"epoch": 3.4776377355309194,
"grad_norm": 1.0594482421875,
"learning_rate": 8.756592205119411e-05,
"loss": 4.178,
"step": 82500
},
{
"epoch": 3.4987143278674706,
"grad_norm": 1.0379809141159058,
"learning_rate": 8.748016978947821e-05,
"loss": 4.1772,
"step": 83000
},
{
"epoch": 3.5197909202040214,
"grad_norm": 1.0656322240829468,
"learning_rate": 8.73944175277623e-05,
"loss": 4.1744,
"step": 83500
},
{
"epoch": 3.540867512540572,
"grad_norm": 1.0557605028152466,
"learning_rate": 8.730866526604639e-05,
"loss": 4.175,
"step": 84000
},
{
"epoch": 3.5619441048771234,
"grad_norm": 1.0651109218597412,
"learning_rate": 8.72229130043305e-05,
"loss": 4.1699,
"step": 84500
},
{
"epoch": 3.5830206972136747,
"grad_norm": 1.0366450548171997,
"learning_rate": 8.713716074261459e-05,
"loss": 4.1669,
"step": 85000
},
{
"epoch": 3.6040972895502255,
"grad_norm": 1.0723545551300049,
"learning_rate": 8.705140848089869e-05,
"loss": 4.1656,
"step": 85500
},
{
"epoch": 3.6251738818867767,
"grad_norm": 1.0765188932418823,
"learning_rate": 8.696565621918278e-05,
"loss": 4.1566,
"step": 86000
},
{
"epoch": 3.6462504742233275,
"grad_norm": 1.0278693437576294,
"learning_rate": 8.687990395746688e-05,
"loss": 4.1581,
"step": 86500
},
{
"epoch": 3.6673270665598787,
"grad_norm": 1.0665415525436401,
"learning_rate": 8.679415169575098e-05,
"loss": 4.1553,
"step": 87000
},
{
"epoch": 3.6884036588964295,
"grad_norm": 1.1068204641342163,
"learning_rate": 8.670839943403508e-05,
"loss": 4.1522,
"step": 87500
},
{
"epoch": 3.7094802512329808,
"grad_norm": 1.0887433290481567,
"learning_rate": 8.662264717231916e-05,
"loss": 4.1515,
"step": 88000
},
{
"epoch": 3.7305568435695315,
"grad_norm": 1.0822304487228394,
"learning_rate": 8.653689491060328e-05,
"loss": 4.151,
"step": 88500
},
{
"epoch": 3.751633435906083,
"grad_norm": 1.0249097347259521,
"learning_rate": 8.645114264888737e-05,
"loss": 4.1483,
"step": 89000
},
{
"epoch": 3.7727100282426336,
"grad_norm": 1.1057443618774414,
"learning_rate": 8.636539038717146e-05,
"loss": 4.1396,
"step": 89500
},
{
"epoch": 3.793786620579185,
"grad_norm": 1.0370334386825562,
"learning_rate": 8.627963812545556e-05,
"loss": 4.1383,
"step": 90000
},
{
"epoch": 3.793786620579185,
"eval_accuracy": 0.6854677570734358,
"eval_loss": 3.8696236610412598,
"eval_runtime": 380.3986,
"eval_samples_per_second": 317.315,
"eval_steps_per_second": 3.307,
"step": 90000
},
{
"epoch": 3.8148632129157356,
"grad_norm": 1.0184780359268188,
"learning_rate": 8.619388586373967e-05,
"loss": 4.1415,
"step": 90500
},
{
"epoch": 3.835939805252287,
"grad_norm": 1.0725204944610596,
"learning_rate": 8.610813360202375e-05,
"loss": 4.1326,
"step": 91000
},
{
"epoch": 3.857016397588838,
"grad_norm": 1.0468461513519287,
"learning_rate": 8.602238134030785e-05,
"loss": 4.1342,
"step": 91500
},
{
"epoch": 3.878092989925389,
"grad_norm": 1.034056305885315,
"learning_rate": 8.593662907859195e-05,
"loss": 4.1255,
"step": 92000
},
{
"epoch": 3.8991695822619397,
"grad_norm": 1.0013164281845093,
"learning_rate": 8.585087681687605e-05,
"loss": 4.1285,
"step": 92500
},
{
"epoch": 3.920246174598491,
"grad_norm": 1.031897783279419,
"learning_rate": 8.576512455516015e-05,
"loss": 4.1259,
"step": 93000
},
{
"epoch": 3.941322766935042,
"grad_norm": 1.0514020919799805,
"learning_rate": 8.567937229344425e-05,
"loss": 4.1222,
"step": 93500
},
{
"epoch": 3.962399359271593,
"grad_norm": 1.0204012393951416,
"learning_rate": 8.559362003172834e-05,
"loss": 4.1192,
"step": 94000
},
{
"epoch": 3.983475951608144,
"grad_norm": 1.0408470630645752,
"learning_rate": 8.550786777001244e-05,
"loss": 4.1196,
"step": 94500
},
{
"epoch": 4.004552543944695,
"grad_norm": 1.0704323053359985,
"learning_rate": 8.542211550829654e-05,
"loss": 4.113,
"step": 95000
},
{
"epoch": 4.025629136281246,
"grad_norm": 1.0480895042419434,
"learning_rate": 8.533636324658063e-05,
"loss": 4.104,
"step": 95500
},
{
"epoch": 4.046705728617797,
"grad_norm": 1.0005662441253662,
"learning_rate": 8.525061098486474e-05,
"loss": 4.1058,
"step": 96000
},
{
"epoch": 4.067782320954348,
"grad_norm": 1.0851818323135376,
"learning_rate": 8.516485872314884e-05,
"loss": 4.1009,
"step": 96500
},
{
"epoch": 4.088858913290899,
"grad_norm": 1.0529694557189941,
"learning_rate": 8.507910646143292e-05,
"loss": 4.1006,
"step": 97000
},
{
"epoch": 4.10993550562745,
"grad_norm": 1.096611499786377,
"learning_rate": 8.499335419971702e-05,
"loss": 4.0969,
"step": 97500
},
{
"epoch": 4.1310120979640015,
"grad_norm": 1.03652822971344,
"learning_rate": 8.490760193800112e-05,
"loss": 4.0958,
"step": 98000
},
{
"epoch": 4.152088690300552,
"grad_norm": 1.0392345190048218,
"learning_rate": 8.482184967628522e-05,
"loss": 4.0927,
"step": 98500
},
{
"epoch": 4.173165282637103,
"grad_norm": 1.0482780933380127,
"learning_rate": 8.473609741456931e-05,
"loss": 4.096,
"step": 99000
},
{
"epoch": 4.194241874973654,
"grad_norm": 1.0446045398712158,
"learning_rate": 8.46503451528534e-05,
"loss": 4.0946,
"step": 99500
},
{
"epoch": 4.2153184673102055,
"grad_norm": 1.073503851890564,
"learning_rate": 8.456459289113751e-05,
"loss": 4.0903,
"step": 100000
},
{
"epoch": 4.2153184673102055,
"eval_accuracy": 0.6917764799272825,
"eval_loss": 3.830518960952759,
"eval_runtime": 371.7472,
"eval_samples_per_second": 324.699,
"eval_steps_per_second": 3.384,
"step": 100000
},
{
"epoch": 4.236395059646756,
"grad_norm": 1.0526291131973267,
"learning_rate": 8.447884062942161e-05,
"loss": 4.0917,
"step": 100500
},
{
"epoch": 4.257471651983307,
"grad_norm": 1.0446993112564087,
"learning_rate": 8.43930883677057e-05,
"loss": 4.0884,
"step": 101000
},
{
"epoch": 4.278548244319858,
"grad_norm": 1.0257527828216553,
"learning_rate": 8.430733610598979e-05,
"loss": 4.0835,
"step": 101500
},
{
"epoch": 4.29962483665641,
"grad_norm": 1.0733942985534668,
"learning_rate": 8.42215838442739e-05,
"loss": 4.0818,
"step": 102000
},
{
"epoch": 4.32070142899296,
"grad_norm": 1.0312360525131226,
"learning_rate": 8.413583158255799e-05,
"loss": 4.0804,
"step": 102500
},
{
"epoch": 4.341778021329511,
"grad_norm": 1.0220298767089844,
"learning_rate": 8.405007932084209e-05,
"loss": 4.0777,
"step": 103000
},
{
"epoch": 4.362854613666062,
"grad_norm": 1.0675746202468872,
"learning_rate": 8.396432705912619e-05,
"loss": 4.0779,
"step": 103500
},
{
"epoch": 4.383931206002614,
"grad_norm": 1.0180693864822388,
"learning_rate": 8.387857479741028e-05,
"loss": 4.0783,
"step": 104000
},
{
"epoch": 4.405007798339165,
"grad_norm": 1.0288758277893066,
"learning_rate": 8.379282253569438e-05,
"loss": 4.0767,
"step": 104500
},
{
"epoch": 4.426084390675715,
"grad_norm": 1.043699860572815,
"learning_rate": 8.370707027397848e-05,
"loss": 4.0731,
"step": 105000
},
{
"epoch": 4.4471609830122665,
"grad_norm": 1.0536242723464966,
"learning_rate": 8.362131801226258e-05,
"loss": 4.0704,
"step": 105500
},
{
"epoch": 4.468237575348818,
"grad_norm": 1.0329852104187012,
"learning_rate": 8.353556575054668e-05,
"loss": 4.0675,
"step": 106000
},
{
"epoch": 4.489314167685369,
"grad_norm": 1.056791067123413,
"learning_rate": 8.344981348883077e-05,
"loss": 4.0693,
"step": 106500
},
{
"epoch": 4.510390760021919,
"grad_norm": 1.0604842901229858,
"learning_rate": 8.336406122711486e-05,
"loss": 4.0636,
"step": 107000
},
{
"epoch": 4.5314673523584705,
"grad_norm": 1.0269449949264526,
"learning_rate": 8.327830896539897e-05,
"loss": 4.0673,
"step": 107500
},
{
"epoch": 4.552543944695022,
"grad_norm": 1.0168564319610596,
"learning_rate": 8.319255670368307e-05,
"loss": 4.0612,
"step": 108000
},
{
"epoch": 4.573620537031573,
"grad_norm": 1.0361183881759644,
"learning_rate": 8.310680444196715e-05,
"loss": 4.0602,
"step": 108500
},
{
"epoch": 4.594697129368123,
"grad_norm": 1.0568174123764038,
"learning_rate": 8.302105218025125e-05,
"loss": 4.0551,
"step": 109000
},
{
"epoch": 4.615773721704675,
"grad_norm": 1.0392801761627197,
"learning_rate": 8.293529991853536e-05,
"loss": 4.0615,
"step": 109500
},
{
"epoch": 4.636850314041226,
"grad_norm": 1.0607601404190063,
"learning_rate": 8.284954765681945e-05,
"loss": 4.0556,
"step": 110000
},
{
"epoch": 4.636850314041226,
"eval_accuracy": 0.6967576585229756,
"eval_loss": 3.7986457347869873,
"eval_runtime": 400.0641,
"eval_samples_per_second": 301.717,
"eval_steps_per_second": 3.144,
"step": 110000
},
{
"epoch": 4.657926906377777,
"grad_norm": 1.0523509979248047,
"learning_rate": 8.276379539510355e-05,
"loss": 4.0542,
"step": 110500
},
{
"epoch": 4.679003498714328,
"grad_norm": 1.0676110982894897,
"learning_rate": 8.267804313338765e-05,
"loss": 4.0599,
"step": 111000
},
{
"epoch": 4.700080091050879,
"grad_norm": 1.0824007987976074,
"learning_rate": 8.259229087167174e-05,
"loss": 4.0507,
"step": 111500
},
{
"epoch": 4.72115668338743,
"grad_norm": 1.017728328704834,
"learning_rate": 8.250653860995584e-05,
"loss": 4.0482,
"step": 112000
},
{
"epoch": 4.742233275723981,
"grad_norm": 1.0395528078079224,
"learning_rate": 8.242078634823994e-05,
"loss": 4.0482,
"step": 112500
},
{
"epoch": 4.763309868060532,
"grad_norm": 1.0375548601150513,
"learning_rate": 8.233503408652403e-05,
"loss": 4.0471,
"step": 113000
},
{
"epoch": 4.784386460397083,
"grad_norm": 1.0234251022338867,
"learning_rate": 8.224928182480814e-05,
"loss": 4.0489,
"step": 113500
},
{
"epoch": 4.805463052733634,
"grad_norm": 1.0382918119430542,
"learning_rate": 8.216352956309224e-05,
"loss": 4.0407,
"step": 114000
},
{
"epoch": 4.826539645070185,
"grad_norm": 1.058995246887207,
"learning_rate": 8.207777730137632e-05,
"loss": 4.0403,
"step": 114500
},
{
"epoch": 4.847616237406736,
"grad_norm": 1.0435962677001953,
"learning_rate": 8.199202503966042e-05,
"loss": 4.0399,
"step": 115000
},
{
"epoch": 4.868692829743287,
"grad_norm": 1.0620546340942383,
"learning_rate": 8.190627277794453e-05,
"loss": 4.0391,
"step": 115500
},
{
"epoch": 4.889769422079838,
"grad_norm": 1.0268526077270508,
"learning_rate": 8.182052051622862e-05,
"loss": 4.0354,
"step": 116000
},
{
"epoch": 4.910846014416389,
"grad_norm": 1.0625653266906738,
"learning_rate": 8.173476825451271e-05,
"loss": 4.0357,
"step": 116500
},
{
"epoch": 4.9319226067529405,
"grad_norm": 1.0372706651687622,
"learning_rate": 8.164901599279681e-05,
"loss": 4.037,
"step": 117000
},
{
"epoch": 4.952999199089491,
"grad_norm": 1.092017412185669,
"learning_rate": 8.156326373108091e-05,
"loss": 4.0341,
"step": 117500
},
{
"epoch": 4.974075791426042,
"grad_norm": 1.0153024196624756,
"learning_rate": 8.147751146936501e-05,
"loss": 4.032,
"step": 118000
},
{
"epoch": 4.995152383762593,
"grad_norm": 1.0225738286972046,
"learning_rate": 8.139175920764911e-05,
"loss": 4.0256,
"step": 118500
},
{
"epoch": 5.0162289760991445,
"grad_norm": 1.0809932947158813,
"learning_rate": 8.13060069459332e-05,
"loss": 4.0253,
"step": 119000
},
{
"epoch": 5.037305568435695,
"grad_norm": 1.0159709453582764,
"learning_rate": 8.12202546842173e-05,
"loss": 4.0231,
"step": 119500
},
{
"epoch": 5.058382160772246,
"grad_norm": 1.0638495683670044,
"learning_rate": 8.11345024225014e-05,
"loss": 4.0202,
"step": 120000
},
{
"epoch": 5.058382160772246,
"eval_accuracy": 0.7016311993721935,
"eval_loss": 3.762713670730591,
"eval_runtime": 509.8567,
"eval_samples_per_second": 236.745,
"eval_steps_per_second": 2.467,
"step": 120000
},
{
"epoch": 5.079458753108797,
"grad_norm": 1.0722323656082153,
"learning_rate": 8.104875016078549e-05,
"loss": 4.0149,
"step": 120500
},
{
"epoch": 5.100535345445349,
"grad_norm": 1.0524057149887085,
"learning_rate": 8.09629978990696e-05,
"loss": 4.0158,
"step": 121000
},
{
"epoch": 5.1216119377819,
"grad_norm": 1.06330144405365,
"learning_rate": 8.08772456373537e-05,
"loss": 4.0163,
"step": 121500
},
{
"epoch": 5.14268853011845,
"grad_norm": 1.132163405418396,
"learning_rate": 8.079149337563778e-05,
"loss": 4.0145,
"step": 122000
},
{
"epoch": 5.163765122455001,
"grad_norm": 1.0527724027633667,
"learning_rate": 8.070574111392188e-05,
"loss": 4.0123,
"step": 122500
},
{
"epoch": 5.184841714791553,
"grad_norm": 1.055350661277771,
"learning_rate": 8.061998885220599e-05,
"loss": 4.0115,
"step": 123000
},
{
"epoch": 5.205918307128104,
"grad_norm": 1.0471426248550415,
"learning_rate": 8.053423659049008e-05,
"loss": 4.0118,
"step": 123500
},
{
"epoch": 5.226994899464654,
"grad_norm": 1.0376275777816772,
"learning_rate": 8.044848432877418e-05,
"loss": 4.0108,
"step": 124000
},
{
"epoch": 5.2480714918012055,
"grad_norm": 1.0201088190078735,
"learning_rate": 8.036273206705827e-05,
"loss": 4.014,
"step": 124500
},
{
"epoch": 5.269148084137757,
"grad_norm": 1.0333786010742188,
"learning_rate": 8.027697980534237e-05,
"loss": 4.0062,
"step": 125000
},
{
"epoch": 5.290224676474308,
"grad_norm": 1.0627624988555908,
"learning_rate": 8.019122754362647e-05,
"loss": 4.0082,
"step": 125500
},
{
"epoch": 5.311301268810858,
"grad_norm": 1.055114984512329,
"learning_rate": 8.010547528191057e-05,
"loss": 4.0066,
"step": 126000
},
{
"epoch": 5.3323778611474095,
"grad_norm": 1.1041890382766724,
"learning_rate": 8.001972302019465e-05,
"loss": 4.0017,
"step": 126500
},
{
"epoch": 5.353454453483961,
"grad_norm": 1.0464231967926025,
"learning_rate": 7.993397075847876e-05,
"loss": 4.0011,
"step": 127000
},
{
"epoch": 5.374531045820512,
"grad_norm": 1.0426297187805176,
"learning_rate": 7.984821849676286e-05,
"loss": 4.0025,
"step": 127500
},
{
"epoch": 5.395607638157063,
"grad_norm": 1.0767418146133423,
"learning_rate": 7.976246623504695e-05,
"loss": 3.9998,
"step": 128000
},
{
"epoch": 5.416684230493614,
"grad_norm": 1.0890185832977295,
"learning_rate": 7.967671397333105e-05,
"loss": 3.9968,
"step": 128500
},
{
"epoch": 5.437760822830165,
"grad_norm": 1.0306752920150757,
"learning_rate": 7.959096171161516e-05,
"loss": 3.9956,
"step": 129000
},
{
"epoch": 5.458837415166716,
"grad_norm": 1.007009744644165,
"learning_rate": 7.950520944989924e-05,
"loss": 3.9954,
"step": 129500
},
{
"epoch": 5.479914007503266,
"grad_norm": 1.061187744140625,
"learning_rate": 7.941945718818334e-05,
"loss": 3.9954,
"step": 130000
},
{
"epoch": 5.479914007503266,
"eval_accuracy": 0.7054268467563853,
"eval_loss": 3.7403712272644043,
"eval_runtime": 382.9213,
"eval_samples_per_second": 315.224,
"eval_steps_per_second": 3.285,
"step": 130000
},
{
"epoch": 5.500990599839818,
"grad_norm": 1.0632169246673584,
"learning_rate": 7.933370492646744e-05,
"loss": 3.999,
"step": 130500
},
{
"epoch": 5.522067192176369,
"grad_norm": 1.0378221273422241,
"learning_rate": 7.924795266475154e-05,
"loss": 3.99,
"step": 131000
},
{
"epoch": 5.54314378451292,
"grad_norm": 1.0668591260910034,
"learning_rate": 7.916220040303564e-05,
"loss": 3.986,
"step": 131500
},
{
"epoch": 5.564220376849471,
"grad_norm": 1.055148720741272,
"learning_rate": 7.907644814131972e-05,
"loss": 3.9929,
"step": 132000
},
{
"epoch": 5.585296969186022,
"grad_norm": 1.0379340648651123,
"learning_rate": 7.899069587960383e-05,
"loss": 3.9846,
"step": 132500
},
{
"epoch": 5.606373561522573,
"grad_norm": 1.033990740776062,
"learning_rate": 7.890494361788793e-05,
"loss": 3.9863,
"step": 133000
},
{
"epoch": 5.627450153859124,
"grad_norm": 1.0608588457107544,
"learning_rate": 7.881919135617202e-05,
"loss": 3.9834,
"step": 133500
},
{
"epoch": 5.648526746195675,
"grad_norm": 1.0459972620010376,
"learning_rate": 7.873343909445611e-05,
"loss": 3.9795,
"step": 134000
},
{
"epoch": 5.669603338532226,
"grad_norm": 1.0605372190475464,
"learning_rate": 7.864768683274023e-05,
"loss": 3.9886,
"step": 134500
},
{
"epoch": 5.690679930868777,
"grad_norm": 1.0254322290420532,
"learning_rate": 7.856193457102431e-05,
"loss": 3.9853,
"step": 135000
},
{
"epoch": 5.711756523205328,
"grad_norm": 1.0655874013900757,
"learning_rate": 7.847618230930841e-05,
"loss": 3.9823,
"step": 135500
},
{
"epoch": 5.7328331155418795,
"grad_norm": 1.0847058296203613,
"learning_rate": 7.839043004759251e-05,
"loss": 3.9826,
"step": 136000
},
{
"epoch": 5.75390970787843,
"grad_norm": 1.033071517944336,
"learning_rate": 7.83046777858766e-05,
"loss": 3.9783,
"step": 136500
},
{
"epoch": 5.774986300214981,
"grad_norm": 1.0303494930267334,
"learning_rate": 7.82189255241607e-05,
"loss": 3.9783,
"step": 137000
},
{
"epoch": 5.796062892551532,
"grad_norm": 1.0557832717895508,
"learning_rate": 7.81331732624448e-05,
"loss": 3.9773,
"step": 137500
},
{
"epoch": 5.8171394848880835,
"grad_norm": 1.020628571510315,
"learning_rate": 7.804742100072889e-05,
"loss": 3.9737,
"step": 138000
},
{
"epoch": 5.838216077224635,
"grad_norm": 1.0511630773544312,
"learning_rate": 7.7961668739013e-05,
"loss": 3.9785,
"step": 138500
},
{
"epoch": 5.859292669561185,
"grad_norm": 1.0364848375320435,
"learning_rate": 7.78759164772971e-05,
"loss": 3.9731,
"step": 139000
},
{
"epoch": 5.880369261897736,
"grad_norm": 1.0542054176330566,
"learning_rate": 7.779016421558118e-05,
"loss": 3.9709,
"step": 139500
},
{
"epoch": 5.901445854234288,
"grad_norm": 1.0658164024353027,
"learning_rate": 7.770441195386528e-05,
"loss": 3.9765,
"step": 140000
},
{
"epoch": 5.901445854234288,
"eval_accuracy": 0.7092709700337916,
"eval_loss": 3.715715169906616,
"eval_runtime": 357.3663,
"eval_samples_per_second": 337.765,
"eval_steps_per_second": 3.52,
"step": 140000
},
{
"epoch": 5.922522446570838,
"grad_norm": 1.0625362396240234,
"learning_rate": 7.761865969214939e-05,
"loss": 3.9662,
"step": 140500
},
{
"epoch": 5.943599038907389,
"grad_norm": 1.0466790199279785,
"learning_rate": 7.753290743043348e-05,
"loss": 3.9716,
"step": 141000
},
{
"epoch": 5.96467563124394,
"grad_norm": 1.029168725013733,
"learning_rate": 7.744715516871758e-05,
"loss": 3.9708,
"step": 141500
},
{
"epoch": 5.985752223580492,
"grad_norm": 1.0432370901107788,
"learning_rate": 7.736140290700167e-05,
"loss": 3.9646,
"step": 142000
},
{
"epoch": 6.006828815917043,
"grad_norm": 1.088691234588623,
"learning_rate": 7.727565064528577e-05,
"loss": 3.9643,
"step": 142500
},
{
"epoch": 6.027905408253593,
"grad_norm": 1.0371322631835938,
"learning_rate": 7.718989838356987e-05,
"loss": 3.9607,
"step": 143000
},
{
"epoch": 6.0489820005901445,
"grad_norm": 1.1907387971878052,
"learning_rate": 7.710414612185397e-05,
"loss": 3.9629,
"step": 143500
},
{
"epoch": 6.070058592926696,
"grad_norm": 1.0450600385665894,
"learning_rate": 7.701839386013807e-05,
"loss": 3.9587,
"step": 144000
},
{
"epoch": 6.091135185263247,
"grad_norm": 1.0403995513916016,
"learning_rate": 7.693264159842217e-05,
"loss": 3.9576,
"step": 144500
},
{
"epoch": 6.112211777599797,
"grad_norm": 1.0670762062072754,
"learning_rate": 7.684688933670626e-05,
"loss": 3.9575,
"step": 145000
},
{
"epoch": 6.1332883699363485,
"grad_norm": 1.052686095237732,
"learning_rate": 7.676113707499035e-05,
"loss": 3.9579,
"step": 145500
},
{
"epoch": 6.1543649622729,
"grad_norm": 1.0293817520141602,
"learning_rate": 7.667538481327446e-05,
"loss": 3.9567,
"step": 146000
},
{
"epoch": 6.175441554609451,
"grad_norm": 1.0518465042114258,
"learning_rate": 7.658963255155856e-05,
"loss": 3.9583,
"step": 146500
},
{
"epoch": 6.196518146946001,
"grad_norm": 1.0499862432479858,
"learning_rate": 7.650388028984264e-05,
"loss": 3.9615,
"step": 147000
},
{
"epoch": 6.217594739282553,
"grad_norm": 1.1056877374649048,
"learning_rate": 7.641812802812674e-05,
"loss": 3.9562,
"step": 147500
},
{
"epoch": 6.238671331619104,
"grad_norm": 1.024328351020813,
"learning_rate": 7.633237576641084e-05,
"loss": 3.9514,
"step": 148000
},
{
"epoch": 6.259747923955655,
"grad_norm": 1.082366943359375,
"learning_rate": 7.624662350469494e-05,
"loss": 3.9533,
"step": 148500
},
{
"epoch": 6.280824516292206,
"grad_norm": 1.0425817966461182,
"learning_rate": 7.616087124297904e-05,
"loss": 3.9516,
"step": 149000
},
{
"epoch": 6.301901108628757,
"grad_norm": 1.0365710258483887,
"learning_rate": 7.607511898126313e-05,
"loss": 3.9513,
"step": 149500
},
{
"epoch": 6.322977700965308,
"grad_norm": 1.0373635292053223,
"learning_rate": 7.598936671954723e-05,
"loss": 3.9475,
"step": 150000
},
{
"epoch": 6.322977700965308,
"eval_accuracy": 0.7121062470089218,
"eval_loss": 3.6981301307678223,
"eval_runtime": 387.5714,
"eval_samples_per_second": 311.442,
"eval_steps_per_second": 3.246,
"step": 150000
},
{
"epoch": 6.344054293301859,
"grad_norm": 1.0629771947860718,
"learning_rate": 7.590361445783133e-05,
"loss": 3.9472,
"step": 150500
},
{
"epoch": 6.36513088563841,
"grad_norm": 1.0040942430496216,
"learning_rate": 7.581786219611543e-05,
"loss": 3.9464,
"step": 151000
},
{
"epoch": 6.386207477974961,
"grad_norm": 1.0757560729980469,
"learning_rate": 7.573210993439951e-05,
"loss": 3.9467,
"step": 151500
},
{
"epoch": 6.407284070311512,
"grad_norm": 1.0498614311218262,
"learning_rate": 7.564635767268363e-05,
"loss": 3.9449,
"step": 152000
},
{
"epoch": 6.428360662648063,
"grad_norm": 1.0598831176757812,
"learning_rate": 7.556060541096772e-05,
"loss": 3.9406,
"step": 152500
},
{
"epoch": 6.449437254984614,
"grad_norm": 1.059706211090088,
"learning_rate": 7.547485314925181e-05,
"loss": 3.9442,
"step": 153000
},
{
"epoch": 6.470513847321165,
"grad_norm": 1.0292398929595947,
"learning_rate": 7.538910088753591e-05,
"loss": 3.9419,
"step": 153500
},
{
"epoch": 6.491590439657716,
"grad_norm": 1.0779485702514648,
"learning_rate": 7.530334862582002e-05,
"loss": 3.942,
"step": 154000
},
{
"epoch": 6.512667031994267,
"grad_norm": 1.0499645471572876,
"learning_rate": 7.52175963641041e-05,
"loss": 3.9432,
"step": 154500
},
{
"epoch": 6.5337436243308185,
"grad_norm": 1.0329405069351196,
"learning_rate": 7.51318441023882e-05,
"loss": 3.9435,
"step": 155000
},
{
"epoch": 6.55482021666737,
"grad_norm": 1.0641345977783203,
"learning_rate": 7.50460918406723e-05,
"loss": 3.9391,
"step": 155500
},
{
"epoch": 6.57589680900392,
"grad_norm": 1.0542525053024292,
"learning_rate": 7.49603395789564e-05,
"loss": 3.9376,
"step": 156000
},
{
"epoch": 6.596973401340471,
"grad_norm": 1.0034940242767334,
"learning_rate": 7.48745873172405e-05,
"loss": 3.9326,
"step": 156500
},
{
"epoch": 6.6180499936770225,
"grad_norm": 1.082963466644287,
"learning_rate": 7.47888350555246e-05,
"loss": 3.9306,
"step": 157000
},
{
"epoch": 6.639126586013573,
"grad_norm": 1.0648419857025146,
"learning_rate": 7.47030827938087e-05,
"loss": 3.9336,
"step": 157500
},
{
"epoch": 6.660203178350124,
"grad_norm": 1.0365084409713745,
"learning_rate": 7.461733053209279e-05,
"loss": 3.9337,
"step": 158000
},
{
"epoch": 6.681279770686675,
"grad_norm": 1.041916847229004,
"learning_rate": 7.453157827037689e-05,
"loss": 3.937,
"step": 158500
},
{
"epoch": 6.702356363023227,
"grad_norm": 1.0709151029586792,
"learning_rate": 7.444582600866098e-05,
"loss": 3.9302,
"step": 159000
},
{
"epoch": 6.723432955359778,
"grad_norm": 1.0640544891357422,
"learning_rate": 7.436007374694507e-05,
"loss": 3.9344,
"step": 159500
},
{
"epoch": 6.744509547696328,
"grad_norm": 1.0779755115509033,
"learning_rate": 7.427432148522919e-05,
"loss": 3.934,
"step": 160000
},
{
"epoch": 6.744509547696328,
"eval_accuracy": 0.7152027964295028,
"eval_loss": 3.6797358989715576,
"eval_runtime": 372.3311,
"eval_samples_per_second": 324.19,
"eval_steps_per_second": 3.379,
"step": 160000
},
{
"epoch": 6.765586140032879,
"grad_norm": 1.0497291088104248,
"learning_rate": 7.418856922351327e-05,
"loss": 3.9309,
"step": 160500
},
{
"epoch": 6.786662732369431,
"grad_norm": 1.0648410320281982,
"learning_rate": 7.410281696179737e-05,
"loss": 3.9271,
"step": 161000
},
{
"epoch": 6.807739324705982,
"grad_norm": 1.0459328889846802,
"learning_rate": 7.401706470008147e-05,
"loss": 3.9246,
"step": 161500
},
{
"epoch": 6.828815917042532,
"grad_norm": 1.0476388931274414,
"learning_rate": 7.393131243836557e-05,
"loss": 3.9249,
"step": 162000
},
{
"epoch": 6.8498925093790834,
"grad_norm": 1.063467025756836,
"learning_rate": 7.384556017664966e-05,
"loss": 3.9225,
"step": 162500
},
{
"epoch": 6.870969101715635,
"grad_norm": 1.0450025796890259,
"learning_rate": 7.375980791493376e-05,
"loss": 3.9231,
"step": 163000
},
{
"epoch": 6.892045694052186,
"grad_norm": 1.060628056526184,
"learning_rate": 7.367405565321786e-05,
"loss": 3.9205,
"step": 163500
},
{
"epoch": 6.913122286388736,
"grad_norm": 1.0600574016571045,
"learning_rate": 7.358830339150196e-05,
"loss": 3.921,
"step": 164000
},
{
"epoch": 6.9341988787252875,
"grad_norm": 1.0973201990127563,
"learning_rate": 7.350255112978606e-05,
"loss": 3.9261,
"step": 164500
},
{
"epoch": 6.955275471061839,
"grad_norm": 1.0583112239837646,
"learning_rate": 7.341679886807014e-05,
"loss": 3.9218,
"step": 165000
},
{
"epoch": 6.97635206339839,
"grad_norm": 1.0847876071929932,
"learning_rate": 7.333104660635425e-05,
"loss": 3.9186,
"step": 165500
},
{
"epoch": 6.997428655734941,
"grad_norm": 1.0278633832931519,
"learning_rate": 7.324529434463835e-05,
"loss": 3.9192,
"step": 166000
},
{
"epoch": 7.018505248071492,
"grad_norm": 1.0276769399642944,
"learning_rate": 7.315954208292244e-05,
"loss": 3.9155,
"step": 166500
},
{
"epoch": 7.039581840408043,
"grad_norm": 1.0618516206741333,
"learning_rate": 7.307378982120653e-05,
"loss": 3.9118,
"step": 167000
},
{
"epoch": 7.060658432744594,
"grad_norm": 1.0534616708755493,
"learning_rate": 7.298803755949063e-05,
"loss": 3.9119,
"step": 167500
},
{
"epoch": 7.081735025081145,
"grad_norm": 1.0465365648269653,
"learning_rate": 7.290228529777473e-05,
"loss": 3.9108,
"step": 168000
},
{
"epoch": 7.102811617417696,
"grad_norm": 1.0513288974761963,
"learning_rate": 7.281653303605883e-05,
"loss": 3.9118,
"step": 168500
},
{
"epoch": 7.123888209754247,
"grad_norm": 1.0785105228424072,
"learning_rate": 7.273078077434293e-05,
"loss": 3.9155,
"step": 169000
},
{
"epoch": 7.144964802090798,
"grad_norm": 1.0616313219070435,
"learning_rate": 7.264502851262703e-05,
"loss": 3.9079,
"step": 169500
},
{
"epoch": 7.166041394427349,
"grad_norm": 1.0441594123840332,
"learning_rate": 7.255927625091112e-05,
"loss": 3.9051,
"step": 170000
},
{
"epoch": 7.166041394427349,
"eval_accuracy": 0.7181761424176181,
"eval_loss": 3.66120982170105,
"eval_runtime": 474.5878,
"eval_samples_per_second": 254.339,
"eval_steps_per_second": 2.651,
"step": 170000
},
{
"epoch": 7.1871179867639,
"grad_norm": 1.0246762037277222,
"learning_rate": 7.247352398919521e-05,
"loss": 3.9102,
"step": 170500
},
{
"epoch": 7.208194579100451,
"grad_norm": 1.009925127029419,
"learning_rate": 7.238777172747931e-05,
"loss": 3.9097,
"step": 171000
},
{
"epoch": 7.229271171437002,
"grad_norm": 1.0212653875350952,
"learning_rate": 7.230201946576342e-05,
"loss": 3.9093,
"step": 171500
},
{
"epoch": 7.250347763773553,
"grad_norm": 1.0637668371200562,
"learning_rate": 7.22162672040475e-05,
"loss": 3.9077,
"step": 172000
},
{
"epoch": 7.271424356110104,
"grad_norm": 1.0469483137130737,
"learning_rate": 7.21305149423316e-05,
"loss": 3.9058,
"step": 172500
},
{
"epoch": 7.292500948446655,
"grad_norm": 1.05317223072052,
"learning_rate": 7.20447626806157e-05,
"loss": 3.9024,
"step": 173000
},
{
"epoch": 7.313577540783206,
"grad_norm": 1.0696461200714111,
"learning_rate": 7.19590104188998e-05,
"loss": 3.9031,
"step": 173500
},
{
"epoch": 7.3346541331197574,
"grad_norm": 1.0131109952926636,
"learning_rate": 7.18732581571839e-05,
"loss": 3.903,
"step": 174000
},
{
"epoch": 7.355730725456308,
"grad_norm": 1.0420342683792114,
"learning_rate": 7.1787505895468e-05,
"loss": 3.9021,
"step": 174500
},
{
"epoch": 7.376807317792859,
"grad_norm": 1.0206871032714844,
"learning_rate": 7.17017536337521e-05,
"loss": 3.9041,
"step": 175000
},
{
"epoch": 7.39788391012941,
"grad_norm": 1.0302414894104004,
"learning_rate": 7.161600137203619e-05,
"loss": 3.9029,
"step": 175500
},
{
"epoch": 7.4189605024659615,
"grad_norm": 1.0818959474563599,
"learning_rate": 7.153024911032029e-05,
"loss": 3.9,
"step": 176000
},
{
"epoch": 7.440037094802513,
"grad_norm": 1.0600553750991821,
"learning_rate": 7.144449684860438e-05,
"loss": 3.9085,
"step": 176500
},
{
"epoch": 7.461113687139063,
"grad_norm": 1.06443452835083,
"learning_rate": 7.135874458688849e-05,
"loss": 3.9026,
"step": 177000
},
{
"epoch": 7.482190279475614,
"grad_norm": 1.0992209911346436,
"learning_rate": 7.127299232517259e-05,
"loss": 3.9046,
"step": 177500
},
{
"epoch": 7.503266871812166,
"grad_norm": 1.0224735736846924,
"learning_rate": 7.118724006345667e-05,
"loss": 3.9034,
"step": 178000
},
{
"epoch": 7.524343464148717,
"grad_norm": 1.0876059532165527,
"learning_rate": 7.110148780174077e-05,
"loss": 3.8991,
"step": 178500
},
{
"epoch": 7.545420056485267,
"grad_norm": 1.094197154045105,
"learning_rate": 7.101573554002488e-05,
"loss": 3.8991,
"step": 179000
},
{
"epoch": 7.566496648821818,
"grad_norm": 1.0587884187698364,
"learning_rate": 7.092998327830897e-05,
"loss": 3.9014,
"step": 179500
},
{
"epoch": 7.58757324115837,
"grad_norm": 1.0567125082015991,
"learning_rate": 7.084423101659306e-05,
"loss": 3.8964,
"step": 180000
},
{
"epoch": 7.58757324115837,
"eval_accuracy": 0.7175588548151539,
"eval_loss": 3.6574795246124268,
"eval_runtime": 369.7866,
"eval_samples_per_second": 326.421,
"eval_steps_per_second": 3.402,
"step": 180000
},
{
"epoch": 7.608649833494921,
"grad_norm": 1.0750635862350464,
"learning_rate": 7.075847875487716e-05,
"loss": 3.8965,
"step": 180500
},
{
"epoch": 7.629726425831471,
"grad_norm": 1.05023992061615,
"learning_rate": 7.067272649316126e-05,
"loss": 3.898,
"step": 181000
},
{
"epoch": 7.650803018168022,
"grad_norm": 1.0411617755889893,
"learning_rate": 7.058697423144536e-05,
"loss": 3.8917,
"step": 181500
},
{
"epoch": 7.671879610504574,
"grad_norm": 1.034397006034851,
"learning_rate": 7.050122196972946e-05,
"loss": 3.8937,
"step": 182000
},
{
"epoch": 7.692956202841125,
"grad_norm": 1.0402940511703491,
"learning_rate": 7.041546970801354e-05,
"loss": 3.8969,
"step": 182500
},
{
"epoch": 7.714032795177676,
"grad_norm": 1.0802668333053589,
"learning_rate": 7.032971744629765e-05,
"loss": 3.8953,
"step": 183000
},
{
"epoch": 7.7351093875142265,
"grad_norm": 1.0716372728347778,
"learning_rate": 7.024396518458175e-05,
"loss": 3.8969,
"step": 183500
},
{
"epoch": 7.756185979850778,
"grad_norm": 1.022306203842163,
"learning_rate": 7.015821292286584e-05,
"loss": 3.8929,
"step": 184000
},
{
"epoch": 7.777262572187329,
"grad_norm": 1.0923848152160645,
"learning_rate": 7.007246066114993e-05,
"loss": 3.8903,
"step": 184500
},
{
"epoch": 7.798339164523879,
"grad_norm": 1.0825234651565552,
"learning_rate": 6.998670839943405e-05,
"loss": 3.8914,
"step": 185000
},
{
"epoch": 7.8194157568604306,
"grad_norm": 1.06827712059021,
"learning_rate": 6.990095613771813e-05,
"loss": 3.8899,
"step": 185500
},
{
"epoch": 7.840492349196982,
"grad_norm": 1.0871261358261108,
"learning_rate": 6.981520387600223e-05,
"loss": 3.8886,
"step": 186000
},
{
"epoch": 7.861568941533533,
"grad_norm": 1.0542433261871338,
"learning_rate": 6.972945161428633e-05,
"loss": 3.8911,
"step": 186500
},
{
"epoch": 7.882645533870084,
"grad_norm": 1.032782793045044,
"learning_rate": 6.964369935257043e-05,
"loss": 3.8897,
"step": 187000
},
{
"epoch": 7.903722126206635,
"grad_norm": 1.0737985372543335,
"learning_rate": 6.955794709085452e-05,
"loss": 3.8901,
"step": 187500
},
{
"epoch": 7.924798718543186,
"grad_norm": 1.0826590061187744,
"learning_rate": 6.947219482913862e-05,
"loss": 3.8798,
"step": 188000
},
{
"epoch": 7.945875310879737,
"grad_norm": 1.0376962423324585,
"learning_rate": 6.938644256742272e-05,
"loss": 3.888,
"step": 188500
},
{
"epoch": 7.966951903216288,
"grad_norm": 1.0698912143707275,
"learning_rate": 6.930069030570682e-05,
"loss": 3.886,
"step": 189000
},
{
"epoch": 7.988028495552839,
"grad_norm": 1.0499728918075562,
"learning_rate": 6.921493804399092e-05,
"loss": 3.8809,
"step": 189500
},
{
"epoch": 8.00910508788939,
"grad_norm": 1.0652636289596558,
"learning_rate": 6.9129185782275e-05,
"loss": 3.8816,
"step": 190000
},
{
"epoch": 8.00910508788939,
"eval_accuracy": 0.7197998617635173,
"eval_loss": 3.641106605529785,
"eval_runtime": 345.0309,
"eval_samples_per_second": 349.841,
"eval_steps_per_second": 3.646,
"step": 190000
},
{
"epoch": 8.03018168022594,
"grad_norm": 1.0654042959213257,
"learning_rate": 6.904343352055911e-05,
"loss": 3.8779,
"step": 190500
},
{
"epoch": 8.051258272562492,
"grad_norm": 1.0701051950454712,
"learning_rate": 6.895768125884321e-05,
"loss": 3.8787,
"step": 191000
},
{
"epoch": 8.072334864899043,
"grad_norm": 1.0810424089431763,
"learning_rate": 6.88719289971273e-05,
"loss": 3.8778,
"step": 191500
},
{
"epoch": 8.093411457235595,
"grad_norm": 1.0826048851013184,
"learning_rate": 6.87861767354114e-05,
"loss": 3.8732,
"step": 192000
},
{
"epoch": 8.114488049572145,
"grad_norm": 1.0865975618362427,
"learning_rate": 6.870042447369551e-05,
"loss": 3.8794,
"step": 192500
},
{
"epoch": 8.135564641908696,
"grad_norm": 1.0548961162567139,
"learning_rate": 6.861467221197959e-05,
"loss": 3.8781,
"step": 193000
},
{
"epoch": 8.156641234245248,
"grad_norm": 1.0833022594451904,
"learning_rate": 6.852891995026369e-05,
"loss": 3.8791,
"step": 193500
},
{
"epoch": 8.177717826581798,
"grad_norm": 1.0577205419540405,
"learning_rate": 6.844316768854779e-05,
"loss": 3.8791,
"step": 194000
},
{
"epoch": 8.19879441891835,
"grad_norm": 1.0607467889785767,
"learning_rate": 6.835741542683189e-05,
"loss": 3.8763,
"step": 194500
},
{
"epoch": 8.2198710112549,
"grad_norm": 1.0553841590881348,
"learning_rate": 6.827166316511599e-05,
"loss": 3.8736,
"step": 195000
},
{
"epoch": 8.24094760359145,
"grad_norm": 1.062528371810913,
"learning_rate": 6.818591090340008e-05,
"loss": 3.8752,
"step": 195500
},
{
"epoch": 8.262024195928003,
"grad_norm": 1.0462160110473633,
"learning_rate": 6.810015864168417e-05,
"loss": 3.8723,
"step": 196000
},
{
"epoch": 8.283100788264553,
"grad_norm": 1.0442641973495483,
"learning_rate": 6.801440637996828e-05,
"loss": 3.8704,
"step": 196500
},
{
"epoch": 8.304177380601104,
"grad_norm": 1.058424949645996,
"learning_rate": 6.792865411825238e-05,
"loss": 3.8726,
"step": 197000
},
{
"epoch": 8.325253972937656,
"grad_norm": 1.1068661212921143,
"learning_rate": 6.784290185653646e-05,
"loss": 3.8717,
"step": 197500
},
{
"epoch": 8.346330565274206,
"grad_norm": 1.094680666923523,
"learning_rate": 6.775714959482056e-05,
"loss": 3.8748,
"step": 198000
},
{
"epoch": 8.367407157610758,
"grad_norm": 1.0505435466766357,
"learning_rate": 6.767139733310467e-05,
"loss": 3.87,
"step": 198500
},
{
"epoch": 8.388483749947309,
"grad_norm": 1.0593727827072144,
"learning_rate": 6.758564507138876e-05,
"loss": 3.8718,
"step": 199000
},
{
"epoch": 8.409560342283859,
"grad_norm": 2.013014793395996,
"learning_rate": 6.749989280967286e-05,
"loss": 3.8706,
"step": 199500
},
{
"epoch": 8.430636934620411,
"grad_norm": 1.0492117404937744,
"learning_rate": 6.741414054795696e-05,
"loss": 3.866,
"step": 200000
},
{
"epoch": 8.430636934620411,
"eval_accuracy": 0.7243159449234394,
"eval_loss": 3.623344659805298,
"eval_runtime": 365.6088,
"eval_samples_per_second": 330.151,
"eval_steps_per_second": 3.441,
"step": 200000
},
{
"epoch": 8.451713526956961,
"grad_norm": 1.0984382629394531,
"learning_rate": 6.732838828624105e-05,
"loss": 3.8656,
"step": 200500
},
{
"epoch": 8.472790119293512,
"grad_norm": 1.087336778640747,
"learning_rate": 6.724263602452515e-05,
"loss": 3.8677,
"step": 201000
},
{
"epoch": 8.493866711630064,
"grad_norm": 1.1131242513656616,
"learning_rate": 6.715688376280924e-05,
"loss": 3.8643,
"step": 201500
},
{
"epoch": 8.514943303966614,
"grad_norm": 1.064294695854187,
"learning_rate": 6.707113150109335e-05,
"loss": 3.8639,
"step": 202000
},
{
"epoch": 8.536019896303166,
"grad_norm": 1.079256296157837,
"learning_rate": 6.698537923937745e-05,
"loss": 3.8673,
"step": 202500
},
{
"epoch": 8.557096488639717,
"grad_norm": 1.0575650930404663,
"learning_rate": 6.689962697766153e-05,
"loss": 3.8641,
"step": 203000
},
{
"epoch": 8.578173080976267,
"grad_norm": 1.0841542482376099,
"learning_rate": 6.681387471594563e-05,
"loss": 3.8632,
"step": 203500
},
{
"epoch": 8.59924967331282,
"grad_norm": 1.274940013885498,
"learning_rate": 6.672812245422974e-05,
"loss": 3.8605,
"step": 204000
},
{
"epoch": 8.62032626564937,
"grad_norm": 1.0720659494400024,
"learning_rate": 6.664237019251383e-05,
"loss": 3.8614,
"step": 204500
},
{
"epoch": 8.64140285798592,
"grad_norm": 1.0526363849639893,
"learning_rate": 6.655661793079792e-05,
"loss": 3.8626,
"step": 205000
},
{
"epoch": 8.662479450322472,
"grad_norm": 1.0941940546035767,
"learning_rate": 6.647086566908202e-05,
"loss": 3.8574,
"step": 205500
},
{
"epoch": 8.683556042659022,
"grad_norm": 1.0506997108459473,
"learning_rate": 6.638511340736612e-05,
"loss": 3.8574,
"step": 206000
},
{
"epoch": 8.704632634995574,
"grad_norm": 1.0625174045562744,
"learning_rate": 6.629936114565022e-05,
"loss": 3.8553,
"step": 206500
},
{
"epoch": 8.725709227332125,
"grad_norm": 1.0743738412857056,
"learning_rate": 6.621360888393432e-05,
"loss": 3.8567,
"step": 207000
},
{
"epoch": 8.746785819668675,
"grad_norm": 1.058609127998352,
"learning_rate": 6.61278566222184e-05,
"loss": 3.8495,
"step": 207500
},
{
"epoch": 8.767862412005227,
"grad_norm": 1.0571857690811157,
"learning_rate": 6.604210436050251e-05,
"loss": 3.8562,
"step": 208000
},
{
"epoch": 8.788939004341778,
"grad_norm": 1.0458927154541016,
"learning_rate": 6.595635209878661e-05,
"loss": 3.8573,
"step": 208500
},
{
"epoch": 8.81001559667833,
"grad_norm": 1.0677621364593506,
"learning_rate": 6.58705998370707e-05,
"loss": 3.8566,
"step": 209000
},
{
"epoch": 8.83109218901488,
"grad_norm": 1.0612725019454956,
"learning_rate": 6.57848475753548e-05,
"loss": 3.855,
"step": 209500
},
{
"epoch": 8.85216878135143,
"grad_norm": 1.063797116279602,
"learning_rate": 6.569909531363891e-05,
"loss": 3.8514,
"step": 210000
},
{
"epoch": 8.85216878135143,
"eval_accuracy": 0.7259153491244752,
"eval_loss": 3.612929344177246,
"eval_runtime": 345.3441,
"eval_samples_per_second": 349.524,
"eval_steps_per_second": 3.643,
"step": 210000
},
{
"epoch": 8.873245373687983,
"grad_norm": 1.0621840953826904,
"learning_rate": 6.561334305192299e-05,
"loss": 3.8538,
"step": 210500
},
{
"epoch": 8.894321966024533,
"grad_norm": 1.0608069896697998,
"learning_rate": 6.552759079020709e-05,
"loss": 3.8536,
"step": 211000
},
{
"epoch": 8.915398558361083,
"grad_norm": 1.0877318382263184,
"learning_rate": 6.544183852849119e-05,
"loss": 3.8508,
"step": 211500
},
{
"epoch": 8.936475150697635,
"grad_norm": 1.0724061727523804,
"learning_rate": 6.535608626677529e-05,
"loss": 3.8559,
"step": 212000
},
{
"epoch": 8.957551743034186,
"grad_norm": 1.043297529220581,
"learning_rate": 6.527033400505939e-05,
"loss": 3.8541,
"step": 212500
},
{
"epoch": 8.978628335370738,
"grad_norm": 1.0555343627929688,
"learning_rate": 6.518458174334348e-05,
"loss": 3.8541,
"step": 213000
},
{
"epoch": 8.999704927707288,
"grad_norm": 1.0732320547103882,
"learning_rate": 6.509882948162758e-05,
"loss": 3.8516,
"step": 213500
},
{
"epoch": 9.020781520043839,
"grad_norm": 1.097643256187439,
"learning_rate": 6.501307721991168e-05,
"loss": 3.8458,
"step": 214000
},
{
"epoch": 9.04185811238039,
"grad_norm": 1.0917260646820068,
"learning_rate": 6.492732495819578e-05,
"loss": 3.8443,
"step": 214500
},
{
"epoch": 9.062934704716941,
"grad_norm": 1.0898720026016235,
"learning_rate": 6.484157269647986e-05,
"loss": 3.8459,
"step": 215000
},
{
"epoch": 9.084011297053493,
"grad_norm": 1.0779426097869873,
"learning_rate": 6.475582043476398e-05,
"loss": 3.8471,
"step": 215500
},
{
"epoch": 9.105087889390044,
"grad_norm": 1.0846961736679077,
"learning_rate": 6.467006817304807e-05,
"loss": 3.844,
"step": 216000
},
{
"epoch": 9.126164481726594,
"grad_norm": 1.0814062356948853,
"learning_rate": 6.458431591133216e-05,
"loss": 3.8365,
"step": 216500
},
{
"epoch": 9.147241074063146,
"grad_norm": 1.077813982963562,
"learning_rate": 6.449856364961626e-05,
"loss": 3.8434,
"step": 217000
},
{
"epoch": 9.168317666399696,
"grad_norm": 1.0718421936035156,
"learning_rate": 6.441281138790037e-05,
"loss": 3.841,
"step": 217500
},
{
"epoch": 9.189394258736247,
"grad_norm": 1.0522727966308594,
"learning_rate": 6.432705912618445e-05,
"loss": 3.8415,
"step": 218000
},
{
"epoch": 9.210470851072799,
"grad_norm": 1.05539870262146,
"learning_rate": 6.424130686446855e-05,
"loss": 3.8469,
"step": 218500
},
{
"epoch": 9.23154744340935,
"grad_norm": 1.0870774984359741,
"learning_rate": 6.415555460275265e-05,
"loss": 3.8413,
"step": 219000
},
{
"epoch": 9.252624035745901,
"grad_norm": 1.086264967918396,
"learning_rate": 6.406980234103675e-05,
"loss": 3.8419,
"step": 219500
},
{
"epoch": 9.273700628082452,
"grad_norm": 1.1172820329666138,
"learning_rate": 6.398405007932085e-05,
"loss": 3.8458,
"step": 220000
},
{
"epoch": 9.273700628082452,
"eval_accuracy": 0.7278197078249661,
"eval_loss": 3.602896213531494,
"eval_runtime": 584.3664,
"eval_samples_per_second": 206.559,
"eval_steps_per_second": 2.153,
"step": 220000
},
{
"epoch": 9.294777220419002,
"grad_norm": 1.0921056270599365,
"learning_rate": 6.389829781760495e-05,
"loss": 3.8395,
"step": 220500
},
{
"epoch": 9.315853812755554,
"grad_norm": 1.106183648109436,
"learning_rate": 6.381254555588903e-05,
"loss": 3.8415,
"step": 221000
},
{
"epoch": 9.336930405092104,
"grad_norm": 1.0834317207336426,
"learning_rate": 6.372679329417314e-05,
"loss": 3.8379,
"step": 221500
},
{
"epoch": 9.358006997428657,
"grad_norm": 1.0805422067642212,
"learning_rate": 6.364104103245724e-05,
"loss": 3.8333,
"step": 222000
},
{
"epoch": 9.379083589765207,
"grad_norm": 1.0940710306167603,
"learning_rate": 6.355528877074132e-05,
"loss": 3.84,
"step": 222500
},
{
"epoch": 9.400160182101757,
"grad_norm": 1.0865525007247925,
"learning_rate": 6.346953650902542e-05,
"loss": 3.8374,
"step": 223000
},
{
"epoch": 9.42123677443831,
"grad_norm": 1.0788134336471558,
"learning_rate": 6.338378424730954e-05,
"loss": 3.8369,
"step": 223500
},
{
"epoch": 9.44231336677486,
"grad_norm": 1.0724178552627563,
"learning_rate": 6.329803198559362e-05,
"loss": 3.8348,
"step": 224000
},
{
"epoch": 9.46338995911141,
"grad_norm": 1.0867884159088135,
"learning_rate": 6.321227972387772e-05,
"loss": 3.8365,
"step": 224500
},
{
"epoch": 9.484466551447962,
"grad_norm": 1.0692847967147827,
"learning_rate": 6.312652746216182e-05,
"loss": 3.8326,
"step": 225000
},
{
"epoch": 9.505543143784513,
"grad_norm": 1.0527898073196411,
"learning_rate": 6.304077520044591e-05,
"loss": 3.8391,
"step": 225500
},
{
"epoch": 9.526619736121065,
"grad_norm": 1.0909944772720337,
"learning_rate": 6.295502293873001e-05,
"loss": 3.8343,
"step": 226000
},
{
"epoch": 9.547696328457615,
"grad_norm": 1.102407693862915,
"learning_rate": 6.286927067701411e-05,
"loss": 3.8318,
"step": 226500
},
{
"epoch": 9.568772920794165,
"grad_norm": 1.0843193531036377,
"learning_rate": 6.278351841529821e-05,
"loss": 3.8352,
"step": 227000
},
{
"epoch": 9.589849513130718,
"grad_norm": 1.0669035911560059,
"learning_rate": 6.269776615358231e-05,
"loss": 3.8348,
"step": 227500
},
{
"epoch": 9.610926105467268,
"grad_norm": 1.0418211221694946,
"learning_rate": 6.26120138918664e-05,
"loss": 3.8324,
"step": 228000
},
{
"epoch": 9.63200269780382,
"grad_norm": 6.3972039222717285,
"learning_rate": 6.252626163015049e-05,
"loss": 3.8328,
"step": 228500
},
{
"epoch": 9.65307929014037,
"grad_norm": 1.092854619026184,
"learning_rate": 6.24405093684346e-05,
"loss": 3.8297,
"step": 229000
},
{
"epoch": 9.67415588247692,
"grad_norm": 1.2428663969039917,
"learning_rate": 6.23547571067187e-05,
"loss": 3.8303,
"step": 229500
},
{
"epoch": 9.695232474813473,
"grad_norm": 1.0658833980560303,
"learning_rate": 6.226900484500279e-05,
"loss": 3.8295,
"step": 230000
},
{
"epoch": 9.695232474813473,
"eval_accuracy": 0.7293444066912721,
"eval_loss": 3.5904197692871094,
"eval_runtime": 396.5823,
"eval_samples_per_second": 304.366,
"eval_steps_per_second": 3.172,
"step": 230000
},
{
"epoch": 9.716309067150023,
"grad_norm": 1.07084321975708,
"learning_rate": 6.218325258328688e-05,
"loss": 3.8357,
"step": 230500
},
{
"epoch": 9.737385659486574,
"grad_norm": 1.078423023223877,
"learning_rate": 6.209750032157098e-05,
"loss": 3.8333,
"step": 231000
},
{
"epoch": 9.758462251823126,
"grad_norm": 1.0778084993362427,
"learning_rate": 6.201174805985508e-05,
"loss": 3.8291,
"step": 231500
},
{
"epoch": 9.779538844159676,
"grad_norm": 1.106953740119934,
"learning_rate": 6.192599579813918e-05,
"loss": 3.8294,
"step": 232000
},
{
"epoch": 9.800615436496226,
"grad_norm": 1.097403883934021,
"learning_rate": 6.184024353642328e-05,
"loss": 3.8293,
"step": 232500
},
{
"epoch": 9.821692028832778,
"grad_norm": 1.0718828439712524,
"learning_rate": 6.175449127470738e-05,
"loss": 3.8249,
"step": 233000
},
{
"epoch": 9.842768621169329,
"grad_norm": 1.0728009939193726,
"learning_rate": 6.166873901299147e-05,
"loss": 3.8312,
"step": 233500
},
{
"epoch": 9.863845213505881,
"grad_norm": 1.0513832569122314,
"learning_rate": 6.158298675127557e-05,
"loss": 3.8254,
"step": 234000
},
{
"epoch": 9.884921805842431,
"grad_norm": 1.10185968875885,
"learning_rate": 6.149723448955966e-05,
"loss": 3.8308,
"step": 234500
},
{
"epoch": 9.905998398178982,
"grad_norm": 1.2493257522583008,
"learning_rate": 6.141148222784377e-05,
"loss": 3.8226,
"step": 235000
},
{
"epoch": 9.927074990515534,
"grad_norm": 1.0535004138946533,
"learning_rate": 6.132572996612787e-05,
"loss": 3.8242,
"step": 235500
},
{
"epoch": 9.948151582852084,
"grad_norm": 1.0883808135986328,
"learning_rate": 6.123997770441195e-05,
"loss": 3.827,
"step": 236000
},
{
"epoch": 9.969228175188636,
"grad_norm": 1.1752756834030151,
"learning_rate": 6.115422544269605e-05,
"loss": 3.8314,
"step": 236500
},
{
"epoch": 9.990304767525187,
"grad_norm": 1.2416428327560425,
"learning_rate": 6.106847318098015e-05,
"loss": 3.8284,
"step": 237000
},
{
"epoch": 10.011381359861737,
"grad_norm": 1.0622965097427368,
"learning_rate": 6.098272091926425e-05,
"loss": 3.829,
"step": 237500
},
{
"epoch": 10.032457952198289,
"grad_norm": 1.3819490671157837,
"learning_rate": 6.0896968657548345e-05,
"loss": 3.8364,
"step": 238000
},
{
"epoch": 10.05353454453484,
"grad_norm": 1.110809087753296,
"learning_rate": 6.081121639583244e-05,
"loss": 3.8249,
"step": 238500
},
{
"epoch": 10.07461113687139,
"grad_norm": 1.0956617593765259,
"learning_rate": 6.072546413411654e-05,
"loss": 3.8234,
"step": 239000
},
{
"epoch": 10.095687729207942,
"grad_norm": 1.1442569494247437,
"learning_rate": 6.063971187240064e-05,
"loss": 3.824,
"step": 239500
},
{
"epoch": 10.116764321544492,
"grad_norm": 1.0540778636932373,
"learning_rate": 6.055395961068473e-05,
"loss": 3.8176,
"step": 240000
},
{
"epoch": 10.116764321544492,
"eval_accuracy": 0.7307194769841474,
"eval_loss": 3.5856189727783203,
"eval_runtime": 511.3956,
"eval_samples_per_second": 236.033,
"eval_steps_per_second": 2.46,
"step": 240000
},
{
"epoch": 10.137840913881044,
"grad_norm": 6.282252788543701,
"learning_rate": 6.046820734896883e-05,
"loss": 3.8198,
"step": 240500
},
{
"epoch": 10.158917506217595,
"grad_norm": 1.1303387880325317,
"learning_rate": 6.038245508725293e-05,
"loss": 3.8226,
"step": 241000
},
{
"epoch": 10.179994098554145,
"grad_norm": 1.0606968402862549,
"learning_rate": 6.029670282553702e-05,
"loss": 3.8198,
"step": 241500
},
{
"epoch": 10.201070690890697,
"grad_norm": 1.2748091220855713,
"learning_rate": 6.0210950563821125e-05,
"loss": 3.8226,
"step": 242000
},
{
"epoch": 10.222147283227248,
"grad_norm": 1.7449324131011963,
"learning_rate": 6.0125198302105223e-05,
"loss": 3.817,
"step": 242500
},
{
"epoch": 10.2432238755638,
"grad_norm": 1.2308754920959473,
"learning_rate": 6.0039446040389315e-05,
"loss": 3.8177,
"step": 243000
},
{
"epoch": 10.26430046790035,
"grad_norm": 1.1201279163360596,
"learning_rate": 5.995369377867341e-05,
"loss": 3.8234,
"step": 243500
},
{
"epoch": 10.2853770602369,
"grad_norm": 1.0678560733795166,
"learning_rate": 5.986794151695752e-05,
"loss": 3.8152,
"step": 244000
},
{
"epoch": 10.306453652573452,
"grad_norm": 1.0624662637710571,
"learning_rate": 5.97821892552416e-05,
"loss": 3.8208,
"step": 244500
},
{
"epoch": 10.327530244910003,
"grad_norm": 1.703698992729187,
"learning_rate": 5.969643699352571e-05,
"loss": 3.8171,
"step": 245000
},
{
"epoch": 10.348606837246553,
"grad_norm": 1.4088839292526245,
"learning_rate": 5.9610684731809806e-05,
"loss": 3.8173,
"step": 245500
},
{
"epoch": 10.369683429583105,
"grad_norm": 1.1175094842910767,
"learning_rate": 5.95249324700939e-05,
"loss": 3.8211,
"step": 246000
},
{
"epoch": 10.390760021919656,
"grad_norm": 1.1760483980178833,
"learning_rate": 5.9439180208377996e-05,
"loss": 3.8189,
"step": 246500
},
{
"epoch": 10.411836614256208,
"grad_norm": 1.075919270515442,
"learning_rate": 5.93534279466621e-05,
"loss": 3.8188,
"step": 247000
},
{
"epoch": 10.432913206592758,
"grad_norm": 1.2069417238235474,
"learning_rate": 5.9267675684946186e-05,
"loss": 3.821,
"step": 247500
},
{
"epoch": 10.453989798929308,
"grad_norm": 1.113855004310608,
"learning_rate": 5.918192342323029e-05,
"loss": 3.8146,
"step": 248000
},
{
"epoch": 10.47506639126586,
"grad_norm": 1.0659838914871216,
"learning_rate": 5.909617116151439e-05,
"loss": 3.8154,
"step": 248500
},
{
"epoch": 10.496142983602411,
"grad_norm": 1.1139541864395142,
"learning_rate": 5.901041889979848e-05,
"loss": 3.8131,
"step": 249000
},
{
"epoch": 10.517219575938963,
"grad_norm": 1.0830203294754028,
"learning_rate": 5.892466663808258e-05,
"loss": 3.8137,
"step": 249500
},
{
"epoch": 10.538296168275513,
"grad_norm": 1.0743952989578247,
"learning_rate": 5.8838914376366684e-05,
"loss": 3.8131,
"step": 250000
},
{
"epoch": 10.538296168275513,
"eval_accuracy": 0.7319061087100139,
"eval_loss": 3.575120210647583,
"eval_runtime": 377.7939,
"eval_samples_per_second": 319.502,
"eval_steps_per_second": 3.33,
"step": 250000
},
{
"epoch": 10.559372760612064,
"grad_norm": 1.058448076248169,
"learning_rate": 5.8753162114650776e-05,
"loss": 3.8169,
"step": 250500
},
{
"epoch": 10.580449352948616,
"grad_norm": 1.0951435565948486,
"learning_rate": 5.8667409852934874e-05,
"loss": 3.8093,
"step": 251000
},
{
"epoch": 10.601525945285166,
"grad_norm": 1.0522403717041016,
"learning_rate": 5.858165759121897e-05,
"loss": 3.808,
"step": 251500
},
{
"epoch": 10.622602537621717,
"grad_norm": 1.0751514434814453,
"learning_rate": 5.8495905329503064e-05,
"loss": 3.8114,
"step": 252000
},
{
"epoch": 10.643679129958269,
"grad_norm": 1.0634427070617676,
"learning_rate": 5.841015306778716e-05,
"loss": 3.8103,
"step": 252500
},
{
"epoch": 10.664755722294819,
"grad_norm": 1.0979136228561401,
"learning_rate": 5.832440080607127e-05,
"loss": 3.8109,
"step": 253000
},
{
"epoch": 10.685832314631371,
"grad_norm": 1.1229182481765747,
"learning_rate": 5.823864854435536e-05,
"loss": 3.805,
"step": 253500
},
{
"epoch": 10.706908906967922,
"grad_norm": 1.0724881887435913,
"learning_rate": 5.815289628263946e-05,
"loss": 3.8077,
"step": 254000
},
{
"epoch": 10.727985499304472,
"grad_norm": 1.2875444889068604,
"learning_rate": 5.8067144020923556e-05,
"loss": 3.813,
"step": 254500
},
{
"epoch": 10.749062091641024,
"grad_norm": 1.3536533117294312,
"learning_rate": 5.798139175920765e-05,
"loss": 3.8165,
"step": 255000
},
{
"epoch": 10.770138683977574,
"grad_norm": 1.0910779237747192,
"learning_rate": 5.789563949749175e-05,
"loss": 3.8209,
"step": 255500
},
{
"epoch": 10.791215276314126,
"grad_norm": 1.1108232736587524,
"learning_rate": 5.780988723577585e-05,
"loss": 3.8109,
"step": 256000
},
{
"epoch": 10.812291868650677,
"grad_norm": 1.10238516330719,
"learning_rate": 5.772413497405994e-05,
"loss": 3.8113,
"step": 256500
},
{
"epoch": 10.833368460987227,
"grad_norm": 1.0434685945510864,
"learning_rate": 5.763838271234404e-05,
"loss": 3.8057,
"step": 257000
},
{
"epoch": 10.85444505332378,
"grad_norm": 1.102758765220642,
"learning_rate": 5.7552630450628146e-05,
"loss": 3.8075,
"step": 257500
},
{
"epoch": 10.87552164566033,
"grad_norm": 1.1036570072174072,
"learning_rate": 5.746687818891223e-05,
"loss": 3.8051,
"step": 258000
},
{
"epoch": 10.89659823799688,
"grad_norm": 1.0561070442199707,
"learning_rate": 5.7381125927196335e-05,
"loss": 3.8055,
"step": 258500
},
{
"epoch": 10.917674830333432,
"grad_norm": 1.0995664596557617,
"learning_rate": 5.7295373665480434e-05,
"loss": 3.8048,
"step": 259000
},
{
"epoch": 10.938751422669982,
"grad_norm": 1.0765480995178223,
"learning_rate": 5.7209621403764525e-05,
"loss": 3.8033,
"step": 259500
},
{
"epoch": 10.959828015006533,
"grad_norm": 1.0961521863937378,
"learning_rate": 5.7123869142048624e-05,
"loss": 3.7999,
"step": 260000
},
{
"epoch": 10.959828015006533,
"eval_accuracy": 0.7337286177132516,
"eval_loss": 3.565173625946045,
"eval_runtime": 355.376,
"eval_samples_per_second": 339.657,
"eval_steps_per_second": 3.54,
"step": 260000
},
{
"epoch": 10.980904607343085,
"grad_norm": 1.0867959260940552,
"learning_rate": 5.703811688033273e-05,
"loss": 3.799,
"step": 260500
},
{
"epoch": 11.001981199679635,
"grad_norm": 1.0832949876785278,
"learning_rate": 5.695236461861681e-05,
"loss": 3.7993,
"step": 261000
},
{
"epoch": 11.023057792016187,
"grad_norm": 1.0857367515563965,
"learning_rate": 5.686661235690092e-05,
"loss": 3.794,
"step": 261500
},
{
"epoch": 11.044134384352738,
"grad_norm": 1.0635234117507935,
"learning_rate": 5.678086009518502e-05,
"loss": 3.8007,
"step": 262000
},
{
"epoch": 11.065210976689288,
"grad_norm": 1.075434684753418,
"learning_rate": 5.669510783346911e-05,
"loss": 3.8002,
"step": 262500
},
{
"epoch": 11.08628756902584,
"grad_norm": 1.068023920059204,
"learning_rate": 5.6609355571753207e-05,
"loss": 3.7963,
"step": 263000
},
{
"epoch": 11.10736416136239,
"grad_norm": 1.1307628154754639,
"learning_rate": 5.652360331003731e-05,
"loss": 3.7976,
"step": 263500
},
{
"epoch": 11.128440753698943,
"grad_norm": 1.081861138343811,
"learning_rate": 5.6437851048321396e-05,
"loss": 3.7941,
"step": 264000
},
{
"epoch": 11.149517346035493,
"grad_norm": 1.220935344696045,
"learning_rate": 5.63520987866055e-05,
"loss": 3.7962,
"step": 264500
},
{
"epoch": 11.170593938372043,
"grad_norm": 1.0999609231948853,
"learning_rate": 5.62663465248896e-05,
"loss": 3.794,
"step": 265000
},
{
"epoch": 11.191670530708596,
"grad_norm": 1.1112908124923706,
"learning_rate": 5.618059426317369e-05,
"loss": 3.7932,
"step": 265500
},
{
"epoch": 11.212747123045146,
"grad_norm": 1.0829813480377197,
"learning_rate": 5.609484200145779e-05,
"loss": 3.796,
"step": 266000
},
{
"epoch": 11.233823715381696,
"grad_norm": 1.0715502500534058,
"learning_rate": 5.6009089739741895e-05,
"loss": 3.7958,
"step": 266500
},
{
"epoch": 11.254900307718248,
"grad_norm": 1.090786099433899,
"learning_rate": 5.5923337478025986e-05,
"loss": 3.7928,
"step": 267000
},
{
"epoch": 11.275976900054799,
"grad_norm": 1.0999945402145386,
"learning_rate": 5.5837585216310085e-05,
"loss": 3.7951,
"step": 267500
},
{
"epoch": 11.29705349239135,
"grad_norm": 1.1373705863952637,
"learning_rate": 5.575183295459418e-05,
"loss": 3.7896,
"step": 268000
},
{
"epoch": 11.318130084727901,
"grad_norm": 1.0846405029296875,
"learning_rate": 5.5666080692878274e-05,
"loss": 3.7935,
"step": 268500
},
{
"epoch": 11.339206677064452,
"grad_norm": 1.069260597229004,
"learning_rate": 5.558032843116238e-05,
"loss": 3.7936,
"step": 269000
},
{
"epoch": 11.360283269401004,
"grad_norm": 1.0948710441589355,
"learning_rate": 5.549457616944648e-05,
"loss": 3.7915,
"step": 269500
},
{
"epoch": 11.381359861737554,
"grad_norm": 1.1011143922805786,
"learning_rate": 5.540882390773057e-05,
"loss": 3.7893,
"step": 270000
},
{
"epoch": 11.381359861737554,
"eval_accuracy": 0.7347749568391873,
"eval_loss": 3.5576703548431396,
"eval_runtime": 735.4632,
"eval_samples_per_second": 164.122,
"eval_steps_per_second": 1.71,
"step": 270000
},
{
"epoch": 11.402436454074106,
"grad_norm": 1.1210061311721802,
"learning_rate": 5.532307164601467e-05,
"loss": 3.7916,
"step": 270500
},
{
"epoch": 11.423513046410656,
"grad_norm": 1.1134895086288452,
"learning_rate": 5.5237319384298766e-05,
"loss": 3.7901,
"step": 271000
},
{
"epoch": 11.444589638747207,
"grad_norm": 1.0865955352783203,
"learning_rate": 5.515156712258286e-05,
"loss": 3.7858,
"step": 271500
},
{
"epoch": 11.465666231083759,
"grad_norm": 1.1761401891708374,
"learning_rate": 5.506581486086696e-05,
"loss": 3.7902,
"step": 272000
},
{
"epoch": 11.48674282342031,
"grad_norm": 1.096869707107544,
"learning_rate": 5.498006259915105e-05,
"loss": 3.7899,
"step": 272500
},
{
"epoch": 11.50781941575686,
"grad_norm": 1.0905576944351196,
"learning_rate": 5.489431033743515e-05,
"loss": 3.7883,
"step": 273000
},
{
"epoch": 11.528896008093412,
"grad_norm": 1.0559996366500854,
"learning_rate": 5.480855807571925e-05,
"loss": 3.7911,
"step": 273500
},
{
"epoch": 11.549972600429962,
"grad_norm": 1.0809942483901978,
"learning_rate": 5.472280581400334e-05,
"loss": 3.7868,
"step": 274000
},
{
"epoch": 11.571049192766514,
"grad_norm": 1.232422947883606,
"learning_rate": 5.463705355228744e-05,
"loss": 3.7868,
"step": 274500
},
{
"epoch": 11.592125785103065,
"grad_norm": 1.1020326614379883,
"learning_rate": 5.4551301290571546e-05,
"loss": 3.7896,
"step": 275000
},
{
"epoch": 11.613202377439615,
"grad_norm": 1.0979630947113037,
"learning_rate": 5.446554902885563e-05,
"loss": 3.787,
"step": 275500
},
{
"epoch": 11.634278969776167,
"grad_norm": 1.0672019720077515,
"learning_rate": 5.4379796767139736e-05,
"loss": 3.7846,
"step": 276000
},
{
"epoch": 11.655355562112717,
"grad_norm": 1.1061978340148926,
"learning_rate": 5.4294044505423834e-05,
"loss": 3.7847,
"step": 276500
},
{
"epoch": 11.67643215444927,
"grad_norm": 1.0684341192245483,
"learning_rate": 5.4208292243707925e-05,
"loss": 3.7848,
"step": 277000
},
{
"epoch": 11.69750874678582,
"grad_norm": 1.0970689058303833,
"learning_rate": 5.4122539981992024e-05,
"loss": 3.786,
"step": 277500
},
{
"epoch": 11.71858533912237,
"grad_norm": 1.0648521184921265,
"learning_rate": 5.403678772027613e-05,
"loss": 3.7807,
"step": 278000
},
{
"epoch": 11.739661931458922,
"grad_norm": 1.1179702281951904,
"learning_rate": 5.395103545856022e-05,
"loss": 3.7857,
"step": 278500
},
{
"epoch": 11.760738523795473,
"grad_norm": 1.0751231908798218,
"learning_rate": 5.386528319684432e-05,
"loss": 3.7831,
"step": 279000
},
{
"epoch": 11.781815116132023,
"grad_norm": 1.0658329725265503,
"learning_rate": 5.377953093512842e-05,
"loss": 3.7803,
"step": 279500
},
{
"epoch": 11.802891708468575,
"grad_norm": 1.0978705883026123,
"learning_rate": 5.369377867341251e-05,
"loss": 3.7844,
"step": 280000
},
{
"epoch": 11.802891708468575,
"eval_accuracy": 0.7365688754206415,
"eval_loss": 3.548048496246338,
"eval_runtime": 385.0835,
"eval_samples_per_second": 313.454,
"eval_steps_per_second": 3.267,
"step": 280000
},
{
"epoch": 11.823968300805126,
"grad_norm": 1.116504192352295,
"learning_rate": 5.3608026411696614e-05,
"loss": 3.7861,
"step": 280500
},
{
"epoch": 11.845044893141678,
"grad_norm": 1.0924879312515259,
"learning_rate": 5.352227414998071e-05,
"loss": 3.7775,
"step": 281000
},
{
"epoch": 11.866121485478228,
"grad_norm": 1.142781138420105,
"learning_rate": 5.34365218882648e-05,
"loss": 3.7798,
"step": 281500
},
{
"epoch": 11.887198077814778,
"grad_norm": 1.06149423122406,
"learning_rate": 5.33507696265489e-05,
"loss": 3.776,
"step": 282000
},
{
"epoch": 11.90827467015133,
"grad_norm": 1.1316848993301392,
"learning_rate": 5.3265017364833e-05,
"loss": 3.7841,
"step": 282500
},
{
"epoch": 11.92935126248788,
"grad_norm": 1.1298831701278687,
"learning_rate": 5.317926510311709e-05,
"loss": 3.778,
"step": 283000
},
{
"epoch": 11.950427854824433,
"grad_norm": 1.1162455081939697,
"learning_rate": 5.3093512841401197e-05,
"loss": 3.7759,
"step": 283500
},
{
"epoch": 11.971504447160983,
"grad_norm": 1.1231080293655396,
"learning_rate": 5.3007760579685295e-05,
"loss": 3.7815,
"step": 284000
},
{
"epoch": 11.992581039497534,
"grad_norm": 1.096587896347046,
"learning_rate": 5.2922008317969386e-05,
"loss": 3.7827,
"step": 284500
},
{
"epoch": 12.013657631834086,
"grad_norm": 1.1026700735092163,
"learning_rate": 5.2836256056253485e-05,
"loss": 3.7755,
"step": 285000
},
{
"epoch": 12.034734224170636,
"grad_norm": 1.117059350013733,
"learning_rate": 5.275050379453759e-05,
"loss": 3.7777,
"step": 285500
},
{
"epoch": 12.055810816507186,
"grad_norm": 1.040801763534546,
"learning_rate": 5.2664751532821675e-05,
"loss": 3.775,
"step": 286000
},
{
"epoch": 12.076887408843739,
"grad_norm": 1.0776019096374512,
"learning_rate": 5.257899927110578e-05,
"loss": 3.773,
"step": 286500
},
{
"epoch": 12.097964001180289,
"grad_norm": 1.09062659740448,
"learning_rate": 5.249324700938988e-05,
"loss": 3.7721,
"step": 287000
},
{
"epoch": 12.119040593516841,
"grad_norm": 1.0833544731140137,
"learning_rate": 5.240749474767397e-05,
"loss": 3.7704,
"step": 287500
},
{
"epoch": 12.140117185853391,
"grad_norm": 1.1047108173370361,
"learning_rate": 5.232174248595807e-05,
"loss": 3.7744,
"step": 288000
},
{
"epoch": 12.161193778189942,
"grad_norm": 1.0666608810424805,
"learning_rate": 5.223599022424217e-05,
"loss": 3.7747,
"step": 288500
},
{
"epoch": 12.182270370526494,
"grad_norm": 1.1739741563796997,
"learning_rate": 5.215023796252626e-05,
"loss": 3.7791,
"step": 289000
},
{
"epoch": 12.203346962863044,
"grad_norm": 1.096304178237915,
"learning_rate": 5.206448570081036e-05,
"loss": 3.7693,
"step": 289500
},
{
"epoch": 12.224423555199595,
"grad_norm": 1.1129555702209473,
"learning_rate": 5.197873343909446e-05,
"loss": 3.7698,
"step": 290000
},
{
"epoch": 12.224423555199595,
"eval_accuracy": 0.7377126335821094,
"eval_loss": 3.541141986846924,
"eval_runtime": 334.4548,
"eval_samples_per_second": 360.904,
"eval_steps_per_second": 3.761,
"step": 290000
},
{
"epoch": 12.245500147536147,
"grad_norm": 1.115224003791809,
"learning_rate": 5.189298117737855e-05,
"loss": 3.7714,
"step": 290500
},
{
"epoch": 12.266576739872697,
"grad_norm": 1.080935001373291,
"learning_rate": 5.180722891566265e-05,
"loss": 3.769,
"step": 291000
},
{
"epoch": 12.28765333220925,
"grad_norm": 1.1065585613250732,
"learning_rate": 5.1721476653946756e-05,
"loss": 3.7678,
"step": 291500
},
{
"epoch": 12.3087299245458,
"grad_norm": 1.1061121225357056,
"learning_rate": 5.163572439223085e-05,
"loss": 3.7715,
"step": 292000
},
{
"epoch": 12.32980651688235,
"grad_norm": 1.0574543476104736,
"learning_rate": 5.1549972130514946e-05,
"loss": 3.7714,
"step": 292500
},
{
"epoch": 12.350883109218902,
"grad_norm": 1.1120624542236328,
"learning_rate": 5.1464219868799044e-05,
"loss": 3.7688,
"step": 293000
},
{
"epoch": 12.371959701555452,
"grad_norm": 1.1174743175506592,
"learning_rate": 5.1378467607083136e-05,
"loss": 3.7668,
"step": 293500
},
{
"epoch": 12.393036293892003,
"grad_norm": 1.085571050643921,
"learning_rate": 5.1292715345367234e-05,
"loss": 3.768,
"step": 294000
},
{
"epoch": 12.414112886228555,
"grad_norm": 1.145038366317749,
"learning_rate": 5.120696308365134e-05,
"loss": 3.7659,
"step": 294500
},
{
"epoch": 12.435189478565105,
"grad_norm": 1.1140490770339966,
"learning_rate": 5.112121082193543e-05,
"loss": 3.7702,
"step": 295000
},
{
"epoch": 12.456266070901657,
"grad_norm": 1.0689771175384521,
"learning_rate": 5.103545856021953e-05,
"loss": 3.7672,
"step": 295500
},
{
"epoch": 12.477342663238208,
"grad_norm": 1.1414870023727417,
"learning_rate": 5.094970629850363e-05,
"loss": 3.772,
"step": 296000
},
{
"epoch": 12.498419255574758,
"grad_norm": 1.116091012954712,
"learning_rate": 5.086395403678772e-05,
"loss": 3.7691,
"step": 296500
},
{
"epoch": 12.51949584791131,
"grad_norm": 1.0675506591796875,
"learning_rate": 5.0778201775071824e-05,
"loss": 3.7738,
"step": 297000
},
{
"epoch": 12.54057244024786,
"grad_norm": 1.1332987546920776,
"learning_rate": 5.069244951335592e-05,
"loss": 3.7689,
"step": 297500
},
{
"epoch": 12.561649032584413,
"grad_norm": 1.1234970092773438,
"learning_rate": 5.0606697251640014e-05,
"loss": 3.7674,
"step": 298000
},
{
"epoch": 12.582725624920963,
"grad_norm": 1.1834415197372437,
"learning_rate": 5.052094498992411e-05,
"loss": 3.7699,
"step": 298500
},
{
"epoch": 12.603802217257513,
"grad_norm": 1.0973575115203857,
"learning_rate": 5.043519272820822e-05,
"loss": 3.7697,
"step": 299000
},
{
"epoch": 12.624878809594065,
"grad_norm": 1.2088514566421509,
"learning_rate": 5.03494404664923e-05,
"loss": 3.7632,
"step": 299500
},
{
"epoch": 12.645955401930616,
"grad_norm": 1.1433393955230713,
"learning_rate": 5.026368820477641e-05,
"loss": 3.7647,
"step": 300000
},
{
"epoch": 12.645955401930616,
"eval_accuracy": 0.7389871693682019,
"eval_loss": 3.532167434692383,
"eval_runtime": 349.0903,
"eval_samples_per_second": 345.773,
"eval_steps_per_second": 3.604,
"step": 300000
},
{
"epoch": 12.667031994267166,
"grad_norm": 1.1161082983016968,
"learning_rate": 5.0177935943060505e-05,
"loss": 3.7608,
"step": 300500
},
{
"epoch": 12.688108586603718,
"grad_norm": 1.0812859535217285,
"learning_rate": 5.00921836813446e-05,
"loss": 3.7692,
"step": 301000
},
{
"epoch": 12.709185178940269,
"grad_norm": 1.1342896223068237,
"learning_rate": 5.0006431419628695e-05,
"loss": 3.7649,
"step": 301500
},
{
"epoch": 12.73026177127682,
"grad_norm": 1.1247891187667847,
"learning_rate": 4.992067915791279e-05,
"loss": 3.7655,
"step": 302000
},
{
"epoch": 12.751338363613371,
"grad_norm": 1.1611120700836182,
"learning_rate": 4.9834926896196885e-05,
"loss": 3.7608,
"step": 302500
},
{
"epoch": 12.772414955949921,
"grad_norm": 1.1434266567230225,
"learning_rate": 4.974917463448099e-05,
"loss": 3.7667,
"step": 303000
},
{
"epoch": 12.793491548286474,
"grad_norm": 1.1008976697921753,
"learning_rate": 4.966342237276508e-05,
"loss": 3.766,
"step": 303500
},
{
"epoch": 12.814568140623024,
"grad_norm": 1.5098308324813843,
"learning_rate": 4.957767011104918e-05,
"loss": 3.7639,
"step": 304000
},
{
"epoch": 12.835644732959576,
"grad_norm": 1.1050877571105957,
"learning_rate": 4.949191784933328e-05,
"loss": 3.7626,
"step": 304500
},
{
"epoch": 12.856721325296126,
"grad_norm": 1.0757510662078857,
"learning_rate": 4.9406165587617376e-05,
"loss": 3.7615,
"step": 305000
},
{
"epoch": 12.877797917632677,
"grad_norm": 1.0879818201065063,
"learning_rate": 4.932041332590147e-05,
"loss": 3.7646,
"step": 305500
},
{
"epoch": 12.898874509969229,
"grad_norm": 1.1412475109100342,
"learning_rate": 4.923466106418557e-05,
"loss": 3.7617,
"step": 306000
},
{
"epoch": 12.91995110230578,
"grad_norm": 1.0685700178146362,
"learning_rate": 4.9148908802469665e-05,
"loss": 3.7582,
"step": 306500
},
{
"epoch": 12.94102769464233,
"grad_norm": 1.0802620649337769,
"learning_rate": 4.906315654075376e-05,
"loss": 3.7621,
"step": 307000
},
{
"epoch": 12.962104286978882,
"grad_norm": 1.1091217994689941,
"learning_rate": 4.897740427903786e-05,
"loss": 3.7661,
"step": 307500
},
{
"epoch": 12.983180879315432,
"grad_norm": 1.1094380617141724,
"learning_rate": 4.889165201732196e-05,
"loss": 3.7634,
"step": 308000
},
{
"epoch": 13.004257471651984,
"grad_norm": 1.0787266492843628,
"learning_rate": 4.880589975560606e-05,
"loss": 3.7582,
"step": 308500
},
{
"epoch": 13.025334063988534,
"grad_norm": 1.091692566871643,
"learning_rate": 4.8720147493890156e-05,
"loss": 3.7516,
"step": 309000
},
{
"epoch": 13.046410656325085,
"grad_norm": 1.0909366607666016,
"learning_rate": 4.8634395232174254e-05,
"loss": 3.7565,
"step": 309500
},
{
"epoch": 13.067487248661637,
"grad_norm": 1.0891413688659668,
"learning_rate": 4.8548642970458346e-05,
"loss": 3.752,
"step": 310000
},
{
"epoch": 13.067487248661637,
"eval_accuracy": 0.7402996715330387,
"eval_loss": 3.525369167327881,
"eval_runtime": 389.336,
"eval_samples_per_second": 310.03,
"eval_steps_per_second": 3.231,
"step": 310000
},
{
"epoch": 13.088563840998187,
"grad_norm": 1.1016366481781006,
"learning_rate": 4.846289070874245e-05,
"loss": 3.7558,
"step": 310500
},
{
"epoch": 13.109640433334738,
"grad_norm": 1.106675386428833,
"learning_rate": 4.837713844702654e-05,
"loss": 3.7563,
"step": 311000
},
{
"epoch": 13.13071702567129,
"grad_norm": 1.1256815195083618,
"learning_rate": 4.829138618531064e-05,
"loss": 3.752,
"step": 311500
},
{
"epoch": 13.15179361800784,
"grad_norm": 1.0983445644378662,
"learning_rate": 4.820563392359474e-05,
"loss": 3.7564,
"step": 312000
},
{
"epoch": 13.172870210344392,
"grad_norm": 1.1477258205413818,
"learning_rate": 4.811988166187884e-05,
"loss": 3.7533,
"step": 312500
},
{
"epoch": 13.193946802680943,
"grad_norm": 1.132095217704773,
"learning_rate": 4.803412940016293e-05,
"loss": 3.7544,
"step": 313000
},
{
"epoch": 13.215023395017493,
"grad_norm": 1.116005301475525,
"learning_rate": 4.7948377138447034e-05,
"loss": 3.7607,
"step": 313500
},
{
"epoch": 13.236099987354045,
"grad_norm": 1.1024689674377441,
"learning_rate": 4.7862624876731126e-05,
"loss": 3.7569,
"step": 314000
},
{
"epoch": 13.257176579690595,
"grad_norm": 1.0978257656097412,
"learning_rate": 4.7776872615015224e-05,
"loss": 3.7533,
"step": 314500
},
{
"epoch": 13.278253172027147,
"grad_norm": 1.1053153276443481,
"learning_rate": 4.7691120353299315e-05,
"loss": 3.7486,
"step": 315000
},
{
"epoch": 13.299329764363698,
"grad_norm": 1.0764636993408203,
"learning_rate": 4.760536809158342e-05,
"loss": 3.7557,
"step": 315500
},
{
"epoch": 13.320406356700248,
"grad_norm": 1.6036978960037231,
"learning_rate": 4.751961582986751e-05,
"loss": 3.7524,
"step": 316000
},
{
"epoch": 13.3414829490368,
"grad_norm": 1.1007312536239624,
"learning_rate": 4.743386356815161e-05,
"loss": 3.7528,
"step": 316500
},
{
"epoch": 13.36255954137335,
"grad_norm": 1.0897459983825684,
"learning_rate": 4.734811130643571e-05,
"loss": 3.75,
"step": 317000
},
{
"epoch": 13.383636133709901,
"grad_norm": 1.2348536252975464,
"learning_rate": 4.726235904471981e-05,
"loss": 3.7485,
"step": 317500
},
{
"epoch": 13.404712726046453,
"grad_norm": 1.1708979606628418,
"learning_rate": 4.71766067830039e-05,
"loss": 3.7529,
"step": 318000
},
{
"epoch": 13.425789318383003,
"grad_norm": 1.1295363903045654,
"learning_rate": 4.7090854521288004e-05,
"loss": 3.754,
"step": 318500
},
{
"epoch": 13.446865910719556,
"grad_norm": 1.1427693367004395,
"learning_rate": 4.7005102259572095e-05,
"loss": 3.7502,
"step": 319000
},
{
"epoch": 13.467942503056106,
"grad_norm": 1.1225385665893555,
"learning_rate": 4.6919349997856193e-05,
"loss": 3.7523,
"step": 319500
},
{
"epoch": 13.489019095392656,
"grad_norm": 1.163130760192871,
"learning_rate": 4.683359773614029e-05,
"loss": 3.7505,
"step": 320000
},
{
"epoch": 13.489019095392656,
"eval_accuracy": 0.7411624469054623,
"eval_loss": 3.5206940174102783,
"eval_runtime": 440.56,
"eval_samples_per_second": 273.983,
"eval_steps_per_second": 2.855,
"step": 320000
},
{
"epoch": 13.510095687729208,
"grad_norm": 1.1969181299209595,
"learning_rate": 4.674784547442439e-05,
"loss": 3.7541,
"step": 320500
},
{
"epoch": 13.531172280065759,
"grad_norm": 1.1104196310043335,
"learning_rate": 4.666209321270849e-05,
"loss": 3.7506,
"step": 321000
},
{
"epoch": 13.55224887240231,
"grad_norm": 1.1563313007354736,
"learning_rate": 4.657634095099259e-05,
"loss": 3.7504,
"step": 321500
},
{
"epoch": 13.573325464738861,
"grad_norm": 1.1158488988876343,
"learning_rate": 4.6490588689276685e-05,
"loss": 3.7487,
"step": 322000
},
{
"epoch": 13.594402057075412,
"grad_norm": 1.162798523902893,
"learning_rate": 4.6404836427560777e-05,
"loss": 3.7509,
"step": 322500
},
{
"epoch": 13.615478649411964,
"grad_norm": 1.1283701658248901,
"learning_rate": 4.6319084165844875e-05,
"loss": 3.7453,
"step": 323000
},
{
"epoch": 13.636555241748514,
"grad_norm": 1.097607970237732,
"learning_rate": 4.623333190412897e-05,
"loss": 3.7504,
"step": 323500
},
{
"epoch": 13.657631834085064,
"grad_norm": 1.1341294050216675,
"learning_rate": 4.614757964241307e-05,
"loss": 3.7455,
"step": 324000
},
{
"epoch": 13.678708426421617,
"grad_norm": 1.096516728401184,
"learning_rate": 4.606182738069717e-05,
"loss": 3.7485,
"step": 324500
},
{
"epoch": 13.699785018758167,
"grad_norm": 1.1244505643844604,
"learning_rate": 4.597607511898127e-05,
"loss": 3.7466,
"step": 325000
},
{
"epoch": 13.720861611094719,
"grad_norm": 1.0886595249176025,
"learning_rate": 4.589032285726536e-05,
"loss": 3.7476,
"step": 325500
},
{
"epoch": 13.74193820343127,
"grad_norm": 1.0960170030593872,
"learning_rate": 4.5804570595549465e-05,
"loss": 3.7456,
"step": 326000
},
{
"epoch": 13.76301479576782,
"grad_norm": 1.102042317390442,
"learning_rate": 4.5718818333833556e-05,
"loss": 3.7488,
"step": 326500
},
{
"epoch": 13.784091388104372,
"grad_norm": 1.0951591730117798,
"learning_rate": 4.5633066072117655e-05,
"loss": 3.7488,
"step": 327000
},
{
"epoch": 13.805167980440922,
"grad_norm": 1.08651864528656,
"learning_rate": 4.554731381040175e-05,
"loss": 3.7427,
"step": 327500
},
{
"epoch": 13.826244572777473,
"grad_norm": 1.1274892091751099,
"learning_rate": 4.546156154868585e-05,
"loss": 3.7452,
"step": 328000
},
{
"epoch": 13.847321165114025,
"grad_norm": 1.133682131767273,
"learning_rate": 4.537580928696994e-05,
"loss": 3.7455,
"step": 328500
},
{
"epoch": 13.868397757450575,
"grad_norm": 1.1238220930099487,
"learning_rate": 4.529005702525405e-05,
"loss": 3.7447,
"step": 329000
},
{
"epoch": 13.889474349787127,
"grad_norm": 1.1423314809799194,
"learning_rate": 4.520430476353814e-05,
"loss": 3.7421,
"step": 329500
},
{
"epoch": 13.910550942123677,
"grad_norm": 1.1446346044540405,
"learning_rate": 4.511855250182224e-05,
"loss": 3.7444,
"step": 330000
},
{
"epoch": 13.910550942123677,
"eval_accuracy": 0.7420820016514994,
"eval_loss": 3.514549732208252,
"eval_runtime": 356.0255,
"eval_samples_per_second": 339.038,
"eval_steps_per_second": 3.533,
"step": 330000
},
{
"epoch": 13.931627534460228,
"grad_norm": 1.092546820640564,
"learning_rate": 4.5032800240106336e-05,
"loss": 3.7556,
"step": 330500
},
{
"epoch": 13.95270412679678,
"grad_norm": 1.082696557044983,
"learning_rate": 4.4947047978390434e-05,
"loss": 3.7534,
"step": 331000
},
{
"epoch": 13.97378071913333,
"grad_norm": 1.1129013299942017,
"learning_rate": 4.4861295716674526e-05,
"loss": 3.7432,
"step": 331500
},
{
"epoch": 13.994857311469882,
"grad_norm": 1.1127337217330933,
"learning_rate": 4.4775543454958624e-05,
"loss": 3.7454,
"step": 332000
},
{
"epoch": 14.015933903806433,
"grad_norm": 1.1370004415512085,
"learning_rate": 4.468979119324272e-05,
"loss": 3.7368,
"step": 332500
},
{
"epoch": 14.037010496142983,
"grad_norm": 1.1192593574523926,
"learning_rate": 4.460403893152682e-05,
"loss": 3.7372,
"step": 333000
},
{
"epoch": 14.058087088479535,
"grad_norm": 1.125972867012024,
"learning_rate": 4.451828666981092e-05,
"loss": 3.7373,
"step": 333500
},
{
"epoch": 14.079163680816086,
"grad_norm": 1.1059080362319946,
"learning_rate": 4.443253440809502e-05,
"loss": 3.7375,
"step": 334000
},
{
"epoch": 14.100240273152636,
"grad_norm": 1.0693610906600952,
"learning_rate": 4.4346782146379116e-05,
"loss": 3.7373,
"step": 334500
},
{
"epoch": 14.121316865489188,
"grad_norm": 1.121158242225647,
"learning_rate": 4.426102988466321e-05,
"loss": 3.7359,
"step": 335000
},
{
"epoch": 14.142393457825738,
"grad_norm": 1.1104836463928223,
"learning_rate": 4.4175277622947305e-05,
"loss": 3.7315,
"step": 335500
},
{
"epoch": 14.16347005016229,
"grad_norm": 1.1174156665802002,
"learning_rate": 4.4089525361231404e-05,
"loss": 3.737,
"step": 336000
},
{
"epoch": 14.184546642498841,
"grad_norm": 1.2723459005355835,
"learning_rate": 4.40037730995155e-05,
"loss": 3.7313,
"step": 336500
},
{
"epoch": 14.205623234835391,
"grad_norm": 1.0968804359436035,
"learning_rate": 4.39180208377996e-05,
"loss": 3.7371,
"step": 337000
},
{
"epoch": 14.226699827171943,
"grad_norm": 1.1313152313232422,
"learning_rate": 4.38322685760837e-05,
"loss": 3.7364,
"step": 337500
},
{
"epoch": 14.247776419508494,
"grad_norm": 1.1184526681900024,
"learning_rate": 4.374651631436779e-05,
"loss": 3.7356,
"step": 338000
},
{
"epoch": 14.268853011845044,
"grad_norm": 1.112886905670166,
"learning_rate": 4.3660764052651895e-05,
"loss": 3.7369,
"step": 338500
},
{
"epoch": 14.289929604181596,
"grad_norm": 1.1022179126739502,
"learning_rate": 4.357501179093599e-05,
"loss": 3.7355,
"step": 339000
},
{
"epoch": 14.311006196518147,
"grad_norm": 1.1573325395584106,
"learning_rate": 4.3489259529220085e-05,
"loss": 3.7352,
"step": 339500
},
{
"epoch": 14.332082788854699,
"grad_norm": 1.1866533756256104,
"learning_rate": 4.3403507267504183e-05,
"loss": 3.737,
"step": 340000
},
{
"epoch": 14.332082788854699,
"eval_accuracy": 0.7435341379804632,
"eval_loss": 3.5067989826202393,
"eval_runtime": 389.8926,
"eval_samples_per_second": 309.588,
"eval_steps_per_second": 3.227,
"step": 340000
},
{
"epoch": 14.353159381191249,
"grad_norm": 1.0904620885849,
"learning_rate": 4.331775500578828e-05,
"loss": 3.7319,
"step": 340500
},
{
"epoch": 14.3742359735278,
"grad_norm": 1.1007230281829834,
"learning_rate": 4.323200274407237e-05,
"loss": 3.7314,
"step": 341000
},
{
"epoch": 14.395312565864351,
"grad_norm": 1.2009572982788086,
"learning_rate": 4.314625048235648e-05,
"loss": 3.7329,
"step": 341500
},
{
"epoch": 14.416389158200902,
"grad_norm": 1.118922472000122,
"learning_rate": 4.306049822064057e-05,
"loss": 3.7367,
"step": 342000
},
{
"epoch": 14.437465750537454,
"grad_norm": 1.1224857568740845,
"learning_rate": 4.297474595892467e-05,
"loss": 3.7355,
"step": 342500
},
{
"epoch": 14.458542342874004,
"grad_norm": 1.1095330715179443,
"learning_rate": 4.2888993697208767e-05,
"loss": 3.7307,
"step": 343000
},
{
"epoch": 14.479618935210555,
"grad_norm": 1.4460396766662598,
"learning_rate": 4.2803241435492865e-05,
"loss": 3.7337,
"step": 343500
},
{
"epoch": 14.500695527547107,
"grad_norm": 1.0933984518051147,
"learning_rate": 4.2717489173776956e-05,
"loss": 3.7339,
"step": 344000
},
{
"epoch": 14.521772119883657,
"grad_norm": 1.0958056449890137,
"learning_rate": 4.263173691206106e-05,
"loss": 3.729,
"step": 344500
},
{
"epoch": 14.542848712220207,
"grad_norm": 1.1270827054977417,
"learning_rate": 4.254598465034515e-05,
"loss": 3.7325,
"step": 345000
},
{
"epoch": 14.56392530455676,
"grad_norm": 1.1075594425201416,
"learning_rate": 4.246023238862925e-05,
"loss": 3.7317,
"step": 345500
},
{
"epoch": 14.58500189689331,
"grad_norm": 1.1267136335372925,
"learning_rate": 4.237448012691335e-05,
"loss": 3.7312,
"step": 346000
},
{
"epoch": 14.606078489229862,
"grad_norm": 1.1254174709320068,
"learning_rate": 4.228872786519745e-05,
"loss": 3.7312,
"step": 346500
},
{
"epoch": 14.627155081566412,
"grad_norm": 1.2097035646438599,
"learning_rate": 4.220297560348154e-05,
"loss": 3.7322,
"step": 347000
},
{
"epoch": 14.648231673902963,
"grad_norm": 1.1356642246246338,
"learning_rate": 4.2117223341765645e-05,
"loss": 3.7302,
"step": 347500
},
{
"epoch": 14.669308266239515,
"grad_norm": 1.162704348564148,
"learning_rate": 4.2031471080049736e-05,
"loss": 3.7288,
"step": 348000
},
{
"epoch": 14.690384858576065,
"grad_norm": 1.1293377876281738,
"learning_rate": 4.1945718818333834e-05,
"loss": 3.7305,
"step": 348500
},
{
"epoch": 14.711461450912616,
"grad_norm": 1.0934334993362427,
"learning_rate": 4.185996655661793e-05,
"loss": 3.7321,
"step": 349000
},
{
"epoch": 14.732538043249168,
"grad_norm": 1.0934746265411377,
"learning_rate": 4.177421429490203e-05,
"loss": 3.7317,
"step": 349500
},
{
"epoch": 14.753614635585718,
"grad_norm": 1.1567370891571045,
"learning_rate": 4.168846203318613e-05,
"loss": 3.728,
"step": 350000
},
{
"epoch": 14.753614635585718,
"eval_accuracy": 0.7442072772903893,
"eval_loss": 3.5009634494781494,
"eval_runtime": 370.3713,
"eval_samples_per_second": 325.905,
"eval_steps_per_second": 3.397,
"step": 350000
},
{
"epoch": 14.77469122792227,
"grad_norm": 1.1248700618743896,
"learning_rate": 4.160270977147022e-05,
"loss": 3.7276,
"step": 350500
},
{
"epoch": 14.79576782025882,
"grad_norm": 1.0951999425888062,
"learning_rate": 4.1516957509754326e-05,
"loss": 3.7298,
"step": 351000
},
{
"epoch": 14.81684441259537,
"grad_norm": 1.1136871576309204,
"learning_rate": 4.143120524803842e-05,
"loss": 3.7269,
"step": 351500
},
{
"epoch": 14.837921004931923,
"grad_norm": 1.1177966594696045,
"learning_rate": 4.1345452986322516e-05,
"loss": 3.7275,
"step": 352000
},
{
"epoch": 14.858997597268473,
"grad_norm": 1.113468885421753,
"learning_rate": 4.1259700724606614e-05,
"loss": 3.7284,
"step": 352500
},
{
"epoch": 14.880074189605025,
"grad_norm": 1.2254393100738525,
"learning_rate": 4.117394846289071e-05,
"loss": 3.7296,
"step": 353000
},
{
"epoch": 14.901150781941576,
"grad_norm": 1.1033222675323486,
"learning_rate": 4.1088196201174804e-05,
"loss": 3.7292,
"step": 353500
},
{
"epoch": 14.922227374278126,
"grad_norm": 1.1160097122192383,
"learning_rate": 4.100244393945891e-05,
"loss": 3.7264,
"step": 354000
},
{
"epoch": 14.943303966614678,
"grad_norm": 1.1194636821746826,
"learning_rate": 4.0916691677743e-05,
"loss": 3.731,
"step": 354500
},
{
"epoch": 14.964380558951229,
"grad_norm": 1.1060283184051514,
"learning_rate": 4.08309394160271e-05,
"loss": 3.7288,
"step": 355000
},
{
"epoch": 14.985457151287779,
"grad_norm": 1.1302695274353027,
"learning_rate": 4.07451871543112e-05,
"loss": 3.7272,
"step": 355500
},
{
"epoch": 15.006533743624331,
"grad_norm": 1.1052415370941162,
"learning_rate": 4.0659434892595295e-05,
"loss": 3.7216,
"step": 356000
},
{
"epoch": 15.027610335960881,
"grad_norm": 1.1193491220474243,
"learning_rate": 4.057368263087939e-05,
"loss": 3.7165,
"step": 356500
},
{
"epoch": 15.048686928297434,
"grad_norm": 1.1366230249404907,
"learning_rate": 4.048793036916349e-05,
"loss": 3.7211,
"step": 357000
},
{
"epoch": 15.069763520633984,
"grad_norm": 1.6845085620880127,
"learning_rate": 4.0402178107447584e-05,
"loss": 3.7252,
"step": 357500
},
{
"epoch": 15.090840112970534,
"grad_norm": 1.1622151136398315,
"learning_rate": 4.031642584573168e-05,
"loss": 3.7212,
"step": 358000
},
{
"epoch": 15.111916705307086,
"grad_norm": 1.1454503536224365,
"learning_rate": 4.023067358401578e-05,
"loss": 3.7196,
"step": 358500
},
{
"epoch": 15.132993297643637,
"grad_norm": 1.1333832740783691,
"learning_rate": 4.014492132229988e-05,
"loss": 3.7219,
"step": 359000
},
{
"epoch": 15.154069889980189,
"grad_norm": 1.1307975053787231,
"learning_rate": 4.005916906058397e-05,
"loss": 3.7206,
"step": 359500
},
{
"epoch": 15.17514648231674,
"grad_norm": 1.1331675052642822,
"learning_rate": 3.9973416798868075e-05,
"loss": 3.7219,
"step": 360000
},
{
"epoch": 15.17514648231674,
"eval_accuracy": 0.7449767223207922,
"eval_loss": 3.495884656906128,
"eval_runtime": 377.7091,
"eval_samples_per_second": 319.574,
"eval_steps_per_second": 3.331,
"step": 360000
},
{
"epoch": 15.19622307465329,
"grad_norm": 1.13162362575531,
"learning_rate": 3.988766453715217e-05,
"loss": 3.7232,
"step": 360500
},
{
"epoch": 15.217299666989842,
"grad_norm": 1.1096625328063965,
"learning_rate": 3.9801912275436265e-05,
"loss": 3.722,
"step": 361000
},
{
"epoch": 15.238376259326392,
"grad_norm": 1.132102608680725,
"learning_rate": 3.971616001372036e-05,
"loss": 3.7213,
"step": 361500
},
{
"epoch": 15.259452851662942,
"grad_norm": 1.1606813669204712,
"learning_rate": 3.963040775200446e-05,
"loss": 3.719,
"step": 362000
},
{
"epoch": 15.280529443999495,
"grad_norm": 1.1512383222579956,
"learning_rate": 3.954465549028856e-05,
"loss": 3.7184,
"step": 362500
},
{
"epoch": 15.301606036336045,
"grad_norm": 1.1273061037063599,
"learning_rate": 3.945890322857266e-05,
"loss": 3.7216,
"step": 363000
},
{
"epoch": 15.322682628672597,
"grad_norm": 1.0954068899154663,
"learning_rate": 3.9373150966856756e-05,
"loss": 3.7178,
"step": 363500
},
{
"epoch": 15.343759221009147,
"grad_norm": 1.1459099054336548,
"learning_rate": 3.928739870514085e-05,
"loss": 3.7191,
"step": 364000
},
{
"epoch": 15.364835813345698,
"grad_norm": 1.1425992250442505,
"learning_rate": 3.920164644342495e-05,
"loss": 3.7171,
"step": 364500
},
{
"epoch": 15.38591240568225,
"grad_norm": 1.129887342453003,
"learning_rate": 3.9115894181709045e-05,
"loss": 3.7191,
"step": 365000
},
{
"epoch": 15.4069889980188,
"grad_norm": 1.1196808815002441,
"learning_rate": 3.903014191999314e-05,
"loss": 3.7151,
"step": 365500
},
{
"epoch": 15.42806559035535,
"grad_norm": 1.1615278720855713,
"learning_rate": 3.894438965827724e-05,
"loss": 3.7205,
"step": 366000
},
{
"epoch": 15.449142182691903,
"grad_norm": 1.1187829971313477,
"learning_rate": 3.885863739656134e-05,
"loss": 3.7168,
"step": 366500
},
{
"epoch": 15.470218775028453,
"grad_norm": 1.1491782665252686,
"learning_rate": 3.877288513484543e-05,
"loss": 3.7173,
"step": 367000
},
{
"epoch": 15.491295367365005,
"grad_norm": 1.1315094232559204,
"learning_rate": 3.868713287312953e-05,
"loss": 3.7181,
"step": 367500
},
{
"epoch": 15.512371959701555,
"grad_norm": 1.147562026977539,
"learning_rate": 3.860138061141363e-05,
"loss": 3.7174,
"step": 368000
},
{
"epoch": 15.533448552038106,
"grad_norm": 1.1303573846817017,
"learning_rate": 3.8515628349697726e-05,
"loss": 3.7176,
"step": 368500
},
{
"epoch": 15.554525144374658,
"grad_norm": 1.1642251014709473,
"learning_rate": 3.842987608798182e-05,
"loss": 3.7164,
"step": 369000
},
{
"epoch": 15.575601736711208,
"grad_norm": 1.1601874828338623,
"learning_rate": 3.834412382626592e-05,
"loss": 3.714,
"step": 369500
},
{
"epoch": 15.596678329047759,
"grad_norm": 1.134791374206543,
"learning_rate": 3.8258371564550014e-05,
"loss": 3.7175,
"step": 370000
},
{
"epoch": 15.596678329047759,
"eval_accuracy": 0.7461383565870549,
"eval_loss": 3.4885029792785645,
"eval_runtime": 366.2261,
"eval_samples_per_second": 329.594,
"eval_steps_per_second": 3.435,
"step": 370000
},
{
"epoch": 15.61775492138431,
"grad_norm": 1.1372766494750977,
"learning_rate": 3.817261930283411e-05,
"loss": 3.7131,
"step": 370500
},
{
"epoch": 15.638831513720861,
"grad_norm": 1.1637376546859741,
"learning_rate": 3.808686704111821e-05,
"loss": 3.7157,
"step": 371000
},
{
"epoch": 15.659908106057413,
"grad_norm": 1.1574280261993408,
"learning_rate": 3.800111477940231e-05,
"loss": 3.7156,
"step": 371500
},
{
"epoch": 15.680984698393964,
"grad_norm": 1.119712471961975,
"learning_rate": 3.79153625176864e-05,
"loss": 3.7158,
"step": 372000
},
{
"epoch": 15.702061290730514,
"grad_norm": 1.1127405166625977,
"learning_rate": 3.7829610255970506e-05,
"loss": 3.718,
"step": 372500
},
{
"epoch": 15.723137883067066,
"grad_norm": 1.1472201347351074,
"learning_rate": 3.77438579942546e-05,
"loss": 3.7129,
"step": 373000
},
{
"epoch": 15.744214475403616,
"grad_norm": 1.154315710067749,
"learning_rate": 3.7658105732538696e-05,
"loss": 3.7095,
"step": 373500
},
{
"epoch": 15.765291067740169,
"grad_norm": 1.0962966680526733,
"learning_rate": 3.7572353470822794e-05,
"loss": 3.7145,
"step": 374000
},
{
"epoch": 15.786367660076719,
"grad_norm": 1.1130129098892212,
"learning_rate": 3.748660120910689e-05,
"loss": 3.7105,
"step": 374500
},
{
"epoch": 15.80744425241327,
"grad_norm": 1.1453133821487427,
"learning_rate": 3.740084894739099e-05,
"loss": 3.7074,
"step": 375000
},
{
"epoch": 15.828520844749821,
"grad_norm": 1.1095906496047974,
"learning_rate": 3.731509668567509e-05,
"loss": 3.7105,
"step": 375500
},
{
"epoch": 15.849597437086372,
"grad_norm": 1.1000834703445435,
"learning_rate": 3.722934442395919e-05,
"loss": 3.7126,
"step": 376000
},
{
"epoch": 15.870674029422922,
"grad_norm": 1.114425778388977,
"learning_rate": 3.714359216224328e-05,
"loss": 3.7124,
"step": 376500
},
{
"epoch": 15.891750621759474,
"grad_norm": 1.1223325729370117,
"learning_rate": 3.705783990052738e-05,
"loss": 3.717,
"step": 377000
},
{
"epoch": 15.912827214096025,
"grad_norm": 1.1239268779754639,
"learning_rate": 3.6972087638811475e-05,
"loss": 3.7119,
"step": 377500
},
{
"epoch": 15.933903806432577,
"grad_norm": 1.0810458660125732,
"learning_rate": 3.6886335377095574e-05,
"loss": 3.7137,
"step": 378000
},
{
"epoch": 15.954980398769127,
"grad_norm": 1.094726800918579,
"learning_rate": 3.680058311537967e-05,
"loss": 3.7119,
"step": 378500
},
{
"epoch": 15.976056991105677,
"grad_norm": 1.1389211416244507,
"learning_rate": 3.671483085366377e-05,
"loss": 3.7129,
"step": 379000
},
{
"epoch": 15.99713358344223,
"grad_norm": 1.1204620599746704,
"learning_rate": 3.662907859194786e-05,
"loss": 3.7078,
"step": 379500
},
{
"epoch": 16.01821017577878,
"grad_norm": 1.1532717943191528,
"learning_rate": 3.654332633023197e-05,
"loss": 3.7095,
"step": 380000
},
{
"epoch": 16.01821017577878,
"eval_accuracy": 0.7471560345864163,
"eval_loss": 3.482637643814087,
"eval_runtime": 362.2614,
"eval_samples_per_second": 333.201,
"eval_steps_per_second": 3.473,
"step": 380000
},
{
"epoch": 16.039286768115332,
"grad_norm": 1.0982561111450195,
"learning_rate": 3.645757406851606e-05,
"loss": 3.7053,
"step": 380500
},
{
"epoch": 16.06036336045188,
"grad_norm": 1.1232151985168457,
"learning_rate": 3.6371821806800157e-05,
"loss": 3.7057,
"step": 381000
},
{
"epoch": 16.081439952788433,
"grad_norm": 1.1275469064712524,
"learning_rate": 3.6286069545084255e-05,
"loss": 3.7106,
"step": 381500
},
{
"epoch": 16.102516545124985,
"grad_norm": 1.1390022039413452,
"learning_rate": 3.620031728336835e-05,
"loss": 3.7067,
"step": 382000
},
{
"epoch": 16.123593137461537,
"grad_norm": 1.1465953588485718,
"learning_rate": 3.6114565021652445e-05,
"loss": 3.6992,
"step": 382500
},
{
"epoch": 16.144669729798085,
"grad_norm": 1.1362589597702026,
"learning_rate": 3.602881275993655e-05,
"loss": 3.7043,
"step": 383000
},
{
"epoch": 16.165746322134638,
"grad_norm": 1.1211862564086914,
"learning_rate": 3.594306049822064e-05,
"loss": 3.7077,
"step": 383500
},
{
"epoch": 16.18682291447119,
"grad_norm": 1.1453872919082642,
"learning_rate": 3.585730823650474e-05,
"loss": 3.7055,
"step": 384000
},
{
"epoch": 16.20789950680774,
"grad_norm": 1.1214009523391724,
"learning_rate": 3.577155597478883e-05,
"loss": 3.7053,
"step": 384500
},
{
"epoch": 16.22897609914429,
"grad_norm": 1.1315155029296875,
"learning_rate": 3.5685803713072936e-05,
"loss": 3.7022,
"step": 385000
},
{
"epoch": 16.250052691480843,
"grad_norm": 1.118340015411377,
"learning_rate": 3.560005145135703e-05,
"loss": 3.7042,
"step": 385500
},
{
"epoch": 16.27112928381739,
"grad_norm": 1.1325634717941284,
"learning_rate": 3.5514299189641126e-05,
"loss": 3.7014,
"step": 386000
},
{
"epoch": 16.292205876153943,
"grad_norm": 1.1475262641906738,
"learning_rate": 3.5428546927925224e-05,
"loss": 3.7073,
"step": 386500
},
{
"epoch": 16.313282468490495,
"grad_norm": 1.1211719512939453,
"learning_rate": 3.534279466620932e-05,
"loss": 3.7013,
"step": 387000
},
{
"epoch": 16.334359060827044,
"grad_norm": 1.1212246417999268,
"learning_rate": 3.525704240449342e-05,
"loss": 3.7025,
"step": 387500
},
{
"epoch": 16.355435653163596,
"grad_norm": 1.140494704246521,
"learning_rate": 3.517129014277752e-05,
"loss": 3.7014,
"step": 388000
},
{
"epoch": 16.376512245500148,
"grad_norm": 1.139985203742981,
"learning_rate": 3.508553788106161e-05,
"loss": 3.7023,
"step": 388500
},
{
"epoch": 16.3975888378367,
"grad_norm": 1.1217272281646729,
"learning_rate": 3.499978561934571e-05,
"loss": 3.7007,
"step": 389000
},
{
"epoch": 16.41866543017325,
"grad_norm": 1.1122268438339233,
"learning_rate": 3.491403335762981e-05,
"loss": 3.7021,
"step": 389500
},
{
"epoch": 16.4397420225098,
"grad_norm": 1.114155650138855,
"learning_rate": 3.4828281095913906e-05,
"loss": 3.7041,
"step": 390000
},
{
"epoch": 16.4397420225098,
"eval_accuracy": 0.748099679513518,
"eval_loss": 3.4774372577667236,
"eval_runtime": 669.3348,
"eval_samples_per_second": 180.337,
"eval_steps_per_second": 1.879,
"step": 390000
},
{
"epoch": 16.460818614846353,
"grad_norm": 1.1597012281417847,
"learning_rate": 3.4742528834198004e-05,
"loss": 3.6993,
"step": 390500
},
{
"epoch": 16.4818952071829,
"grad_norm": 1.1105221509933472,
"learning_rate": 3.46567765724821e-05,
"loss": 3.7026,
"step": 391000
},
{
"epoch": 16.502971799519454,
"grad_norm": 1.1412173509597778,
"learning_rate": 3.45710243107662e-05,
"loss": 3.7036,
"step": 391500
},
{
"epoch": 16.524048391856006,
"grad_norm": 1.130205512046814,
"learning_rate": 3.448527204905029e-05,
"loss": 3.705,
"step": 392000
},
{
"epoch": 16.545124984192555,
"grad_norm": 1.1421148777008057,
"learning_rate": 3.43995197873344e-05,
"loss": 3.6982,
"step": 392500
},
{
"epoch": 16.566201576529107,
"grad_norm": 1.1718064546585083,
"learning_rate": 3.431376752561849e-05,
"loss": 3.7012,
"step": 393000
},
{
"epoch": 16.58727816886566,
"grad_norm": 1.098997950553894,
"learning_rate": 3.422801526390259e-05,
"loss": 3.7049,
"step": 393500
},
{
"epoch": 16.608354761202207,
"grad_norm": 1.116553544998169,
"learning_rate": 3.4142263002186686e-05,
"loss": 3.7025,
"step": 394000
},
{
"epoch": 16.62943135353876,
"grad_norm": 1.1532702445983887,
"learning_rate": 3.4056510740470784e-05,
"loss": 3.7057,
"step": 394500
},
{
"epoch": 16.65050794587531,
"grad_norm": 1.138427734375,
"learning_rate": 3.3970758478754875e-05,
"loss": 3.7021,
"step": 395000
},
{
"epoch": 16.67158453821186,
"grad_norm": 1.1160695552825928,
"learning_rate": 3.388500621703898e-05,
"loss": 3.7033,
"step": 395500
},
{
"epoch": 16.692661130548412,
"grad_norm": 1.1632444858551025,
"learning_rate": 3.379925395532307e-05,
"loss": 3.7014,
"step": 396000
},
{
"epoch": 16.713737722884964,
"grad_norm": 1.0868161916732788,
"learning_rate": 3.371350169360717e-05,
"loss": 3.6987,
"step": 396500
},
{
"epoch": 16.734814315221517,
"grad_norm": 1.1202818155288696,
"learning_rate": 3.362774943189127e-05,
"loss": 3.7007,
"step": 397000
},
{
"epoch": 16.755890907558065,
"grad_norm": 1.0992971658706665,
"learning_rate": 3.354199717017537e-05,
"loss": 3.7,
"step": 397500
},
{
"epoch": 16.776967499894617,
"grad_norm": 1.1405789852142334,
"learning_rate": 3.345624490845946e-05,
"loss": 3.6995,
"step": 398000
},
{
"epoch": 16.79804409223117,
"grad_norm": 1.1510382890701294,
"learning_rate": 3.3370492646743564e-05,
"loss": 3.6975,
"step": 398500
},
{
"epoch": 16.819120684567718,
"grad_norm": 1.123204231262207,
"learning_rate": 3.3284740385027655e-05,
"loss": 3.7028,
"step": 399000
},
{
"epoch": 16.84019727690427,
"grad_norm": 1.0934505462646484,
"learning_rate": 3.319898812331175e-05,
"loss": 3.6959,
"step": 399500
},
{
"epoch": 16.861273869240822,
"grad_norm": 1.1632798910140991,
"learning_rate": 3.311323586159585e-05,
"loss": 3.6965,
"step": 400000
},
{
"epoch": 16.861273869240822,
"eval_accuracy": 0.7487736915983122,
"eval_loss": 3.474999189376831,
"eval_runtime": 346.3172,
"eval_samples_per_second": 348.542,
"eval_steps_per_second": 3.633,
"step": 400000
},
{
"epoch": 16.88235046157737,
"grad_norm": 1.1358646154403687,
"learning_rate": 3.302748359987995e-05,
"loss": 3.6953,
"step": 400500
},
{
"epoch": 16.903427053913923,
"grad_norm": 1.1432658433914185,
"learning_rate": 3.294173133816404e-05,
"loss": 3.6969,
"step": 401000
},
{
"epoch": 16.924503646250475,
"grad_norm": 1.144667625427246,
"learning_rate": 3.2855979076448147e-05,
"loss": 3.7012,
"step": 401500
},
{
"epoch": 16.945580238587024,
"grad_norm": 1.1570831537246704,
"learning_rate": 3.277022681473224e-05,
"loss": 3.6965,
"step": 402000
},
{
"epoch": 16.966656830923576,
"grad_norm": 1.1515319347381592,
"learning_rate": 3.2684474553016336e-05,
"loss": 3.6991,
"step": 402500
},
{
"epoch": 16.987733423260128,
"grad_norm": 1.152614712715149,
"learning_rate": 3.2598722291300435e-05,
"loss": 3.6975,
"step": 403000
},
{
"epoch": 17.00881001559668,
"grad_norm": 1.1435205936431885,
"learning_rate": 3.251297002958453e-05,
"loss": 3.6948,
"step": 403500
},
{
"epoch": 17.02988660793323,
"grad_norm": 1.151934027671814,
"learning_rate": 3.242721776786863e-05,
"loss": 3.6945,
"step": 404000
},
{
"epoch": 17.05096320026978,
"grad_norm": 1.1233102083206177,
"learning_rate": 3.234146550615272e-05,
"loss": 3.6914,
"step": 404500
},
{
"epoch": 17.072039792606333,
"grad_norm": 1.1564433574676514,
"learning_rate": 3.225571324443683e-05,
"loss": 3.6923,
"step": 405000
},
{
"epoch": 17.09311638494288,
"grad_norm": 1.154978632926941,
"learning_rate": 3.216996098272092e-05,
"loss": 3.6934,
"step": 405500
},
{
"epoch": 17.114192977279433,
"grad_norm": 1.1324048042297363,
"learning_rate": 3.208420872100502e-05,
"loss": 3.6925,
"step": 406000
},
{
"epoch": 17.135269569615986,
"grad_norm": 1.160428762435913,
"learning_rate": 3.1998456459289116e-05,
"loss": 3.6894,
"step": 406500
},
{
"epoch": 17.156346161952534,
"grad_norm": 1.14723801612854,
"learning_rate": 3.1912704197573214e-05,
"loss": 3.6893,
"step": 407000
},
{
"epoch": 17.177422754289086,
"grad_norm": 1.142676830291748,
"learning_rate": 3.1826951935857306e-05,
"loss": 3.689,
"step": 407500
},
{
"epoch": 17.19849934662564,
"grad_norm": 1.1348414421081543,
"learning_rate": 3.174119967414141e-05,
"loss": 3.6944,
"step": 408000
},
{
"epoch": 17.219575938962187,
"grad_norm": 1.1707022190093994,
"learning_rate": 3.16554474124255e-05,
"loss": 3.6907,
"step": 408500
},
{
"epoch": 17.24065253129874,
"grad_norm": 1.1302919387817383,
"learning_rate": 3.15696951507096e-05,
"loss": 3.6938,
"step": 409000
},
{
"epoch": 17.26172912363529,
"grad_norm": 1.1420087814331055,
"learning_rate": 3.14839428889937e-05,
"loss": 3.6907,
"step": 409500
},
{
"epoch": 17.282805715971843,
"grad_norm": 1.1732131242752075,
"learning_rate": 3.13981906272778e-05,
"loss": 3.6936,
"step": 410000
},
{
"epoch": 17.282805715971843,
"eval_accuracy": 0.7495819962265959,
"eval_loss": 3.467454433441162,
"eval_runtime": 347.9304,
"eval_samples_per_second": 346.926,
"eval_steps_per_second": 3.616,
"step": 410000
},
{
"epoch": 17.303882308308392,
"grad_norm": 1.1984071731567383,
"learning_rate": 3.131243836556189e-05,
"loss": 3.6933,
"step": 410500
},
{
"epoch": 17.324958900644944,
"grad_norm": 1.1304235458374023,
"learning_rate": 3.1226686103845994e-05,
"loss": 3.6873,
"step": 411000
},
{
"epoch": 17.346035492981496,
"grad_norm": 1.1922729015350342,
"learning_rate": 3.1140933842130086e-05,
"loss": 3.6891,
"step": 411500
},
{
"epoch": 17.367112085318045,
"grad_norm": 1.1361206769943237,
"learning_rate": 3.1055181580414184e-05,
"loss": 3.6919,
"step": 412000
},
{
"epoch": 17.388188677654597,
"grad_norm": 1.1486762762069702,
"learning_rate": 3.096942931869828e-05,
"loss": 3.694,
"step": 412500
},
{
"epoch": 17.40926526999115,
"grad_norm": 1.4625214338302612,
"learning_rate": 3.088367705698238e-05,
"loss": 3.6867,
"step": 413000
},
{
"epoch": 17.430341862327698,
"grad_norm": 1.0977376699447632,
"learning_rate": 3.079792479526647e-05,
"loss": 3.6914,
"step": 413500
},
{
"epoch": 17.45141845466425,
"grad_norm": 1.0892326831817627,
"learning_rate": 3.071217253355058e-05,
"loss": 3.689,
"step": 414000
},
{
"epoch": 17.472495047000802,
"grad_norm": 1.1485940217971802,
"learning_rate": 3.062642027183467e-05,
"loss": 3.6872,
"step": 414500
},
{
"epoch": 17.49357163933735,
"grad_norm": 1.1337236166000366,
"learning_rate": 3.054066801011877e-05,
"loss": 3.6889,
"step": 415000
},
{
"epoch": 17.514648231673903,
"grad_norm": 1.1493059396743774,
"learning_rate": 3.045491574840287e-05,
"loss": 3.6902,
"step": 415500
},
{
"epoch": 17.535724824010455,
"grad_norm": 1.159186601638794,
"learning_rate": 3.0369163486686964e-05,
"loss": 3.6886,
"step": 416000
},
{
"epoch": 17.556801416347007,
"grad_norm": 1.2178006172180176,
"learning_rate": 3.028341122497106e-05,
"loss": 3.6861,
"step": 416500
},
{
"epoch": 17.577878008683555,
"grad_norm": 1.1744892597198486,
"learning_rate": 3.019765896325516e-05,
"loss": 3.6932,
"step": 417000
},
{
"epoch": 17.598954601020107,
"grad_norm": 1.1790454387664795,
"learning_rate": 3.0111906701539255e-05,
"loss": 3.6896,
"step": 417500
},
{
"epoch": 17.62003119335666,
"grad_norm": 1.1799150705337524,
"learning_rate": 3.002615443982335e-05,
"loss": 3.6871,
"step": 418000
},
{
"epoch": 17.641107785693208,
"grad_norm": 1.1559871435165405,
"learning_rate": 2.9940402178107452e-05,
"loss": 3.6872,
"step": 418500
},
{
"epoch": 17.66218437802976,
"grad_norm": 1.1544427871704102,
"learning_rate": 2.9854649916391547e-05,
"loss": 3.6861,
"step": 419000
},
{
"epoch": 17.683260970366312,
"grad_norm": 1.1825308799743652,
"learning_rate": 2.976889765467564e-05,
"loss": 3.6898,
"step": 419500
},
{
"epoch": 17.70433756270286,
"grad_norm": 1.1446431875228882,
"learning_rate": 2.9683145392959737e-05,
"loss": 3.6885,
"step": 420000
},
{
"epoch": 17.70433756270286,
"eval_accuracy": 0.7504733197332878,
"eval_loss": 3.464151620864868,
"eval_runtime": 269.985,
"eval_samples_per_second": 447.084,
"eval_steps_per_second": 4.66,
"step": 420000
},
{
"epoch": 17.725414155039413,
"grad_norm": 1.170128345489502,
"learning_rate": 2.9597393131243838e-05,
"loss": 3.6884,
"step": 420500
},
{
"epoch": 17.746490747375965,
"grad_norm": 1.1436148881912231,
"learning_rate": 2.9511640869527933e-05,
"loss": 3.6846,
"step": 421000
},
{
"epoch": 17.767567339712514,
"grad_norm": 1.150434970855713,
"learning_rate": 2.942588860781203e-05,
"loss": 3.6844,
"step": 421500
},
{
"epoch": 17.788643932049066,
"grad_norm": 1.1505721807479858,
"learning_rate": 2.934013634609613e-05,
"loss": 3.6865,
"step": 422000
},
{
"epoch": 17.809720524385618,
"grad_norm": 1.1380311250686646,
"learning_rate": 2.9254384084380225e-05,
"loss": 3.6887,
"step": 422500
},
{
"epoch": 17.830797116722167,
"grad_norm": 1.2679435014724731,
"learning_rate": 2.9168631822664323e-05,
"loss": 3.6863,
"step": 423000
},
{
"epoch": 17.85187370905872,
"grad_norm": 1.1537624597549438,
"learning_rate": 2.908287956094842e-05,
"loss": 3.6882,
"step": 423500
},
{
"epoch": 17.87295030139527,
"grad_norm": 1.1346724033355713,
"learning_rate": 2.899712729923252e-05,
"loss": 3.6877,
"step": 424000
},
{
"epoch": 17.894026893731823,
"grad_norm": 1.145811915397644,
"learning_rate": 2.8911375037516615e-05,
"loss": 3.6859,
"step": 424500
},
{
"epoch": 17.91510348606837,
"grad_norm": 1.1415505409240723,
"learning_rate": 2.8825622775800716e-05,
"loss": 3.6883,
"step": 425000
},
{
"epoch": 17.936180078404924,
"grad_norm": 1.1721705198287964,
"learning_rate": 2.873987051408481e-05,
"loss": 3.6848,
"step": 425500
},
{
"epoch": 17.957256670741476,
"grad_norm": 1.162593960762024,
"learning_rate": 2.8654118252368906e-05,
"loss": 3.6833,
"step": 426000
},
{
"epoch": 17.978333263078024,
"grad_norm": 1.2089393138885498,
"learning_rate": 2.8568365990653008e-05,
"loss": 3.6832,
"step": 426500
},
{
"epoch": 17.999409855414576,
"grad_norm": 1.1597883701324463,
"learning_rate": 2.8482613728937103e-05,
"loss": 3.6876,
"step": 427000
},
{
"epoch": 18.02048644775113,
"grad_norm": 1.1553784608840942,
"learning_rate": 2.8396861467221198e-05,
"loss": 3.6807,
"step": 427500
},
{
"epoch": 18.041563040087677,
"grad_norm": 1.1713021993637085,
"learning_rate": 2.83111092055053e-05,
"loss": 3.6803,
"step": 428000
},
{
"epoch": 18.06263963242423,
"grad_norm": 1.128185749053955,
"learning_rate": 2.8225356943789394e-05,
"loss": 3.6818,
"step": 428500
},
{
"epoch": 18.08371622476078,
"grad_norm": 1.1165796518325806,
"learning_rate": 2.813960468207349e-05,
"loss": 3.6788,
"step": 429000
},
{
"epoch": 18.10479281709733,
"grad_norm": 1.1402311325073242,
"learning_rate": 2.805385242035759e-05,
"loss": 3.6792,
"step": 429500
},
{
"epoch": 18.125869409433882,
"grad_norm": 1.1764500141143799,
"learning_rate": 2.7968100158641686e-05,
"loss": 3.679,
"step": 430000
},
{
"epoch": 18.125869409433882,
"eval_accuracy": 0.7512112571064025,
"eval_loss": 3.460817813873291,
"eval_runtime": 255.0938,
"eval_samples_per_second": 473.183,
"eval_steps_per_second": 4.932,
"step": 430000
},
{
"epoch": 18.146946001770434,
"grad_norm": 1.165667176246643,
"learning_rate": 2.788234789692578e-05,
"loss": 3.6797,
"step": 430500
},
{
"epoch": 18.168022594106986,
"grad_norm": 1.1702324151992798,
"learning_rate": 2.7796595635209882e-05,
"loss": 3.6803,
"step": 431000
},
{
"epoch": 18.189099186443535,
"grad_norm": 1.1576842069625854,
"learning_rate": 2.7710843373493977e-05,
"loss": 3.6817,
"step": 431500
},
{
"epoch": 18.210175778780087,
"grad_norm": 1.148067593574524,
"learning_rate": 2.7625091111778072e-05,
"loss": 3.6835,
"step": 432000
},
{
"epoch": 18.23125237111664,
"grad_norm": 1.1909527778625488,
"learning_rate": 2.7539338850062174e-05,
"loss": 3.6798,
"step": 432500
},
{
"epoch": 18.252328963453188,
"grad_norm": 1.1437509059906006,
"learning_rate": 2.745358658834627e-05,
"loss": 3.6799,
"step": 433000
},
{
"epoch": 18.27340555578974,
"grad_norm": 1.1543158292770386,
"learning_rate": 2.7367834326630364e-05,
"loss": 3.6779,
"step": 433500
},
{
"epoch": 18.294482148126292,
"grad_norm": 1.1683281660079956,
"learning_rate": 2.7282082064914465e-05,
"loss": 3.6794,
"step": 434000
},
{
"epoch": 18.31555874046284,
"grad_norm": 1.1762720346450806,
"learning_rate": 2.719632980319856e-05,
"loss": 3.6777,
"step": 434500
},
{
"epoch": 18.336635332799393,
"grad_norm": 1.1727948188781738,
"learning_rate": 2.7110577541482655e-05,
"loss": 3.6766,
"step": 435000
},
{
"epoch": 18.357711925135945,
"grad_norm": 1.1614731550216675,
"learning_rate": 2.7024825279766757e-05,
"loss": 3.6786,
"step": 435500
},
{
"epoch": 18.378788517472493,
"grad_norm": 1.1607666015625,
"learning_rate": 2.6939073018050852e-05,
"loss": 3.6796,
"step": 436000
},
{
"epoch": 18.399865109809046,
"grad_norm": 1.142889142036438,
"learning_rate": 2.685332075633495e-05,
"loss": 3.6774,
"step": 436500
},
{
"epoch": 18.420941702145598,
"grad_norm": 1.143833875656128,
"learning_rate": 2.6767568494619045e-05,
"loss": 3.6786,
"step": 437000
},
{
"epoch": 18.44201829448215,
"grad_norm": 1.17220139503479,
"learning_rate": 2.6681816232903143e-05,
"loss": 3.6776,
"step": 437500
},
{
"epoch": 18.4630948868187,
"grad_norm": 1.1844310760498047,
"learning_rate": 2.6596063971187242e-05,
"loss": 3.6746,
"step": 438000
},
{
"epoch": 18.48417147915525,
"grad_norm": 1.2015862464904785,
"learning_rate": 2.6510311709471337e-05,
"loss": 3.6762,
"step": 438500
},
{
"epoch": 18.505248071491803,
"grad_norm": 1.117239236831665,
"learning_rate": 2.642455944775544e-05,
"loss": 3.6771,
"step": 439000
},
{
"epoch": 18.52632466382835,
"grad_norm": 1.1687533855438232,
"learning_rate": 2.6338807186039533e-05,
"loss": 3.6746,
"step": 439500
},
{
"epoch": 18.547401256164903,
"grad_norm": 1.137424111366272,
"learning_rate": 2.6253054924323628e-05,
"loss": 3.6735,
"step": 440000
},
{
"epoch": 18.547401256164903,
"eval_accuracy": 0.7517400566575286,
"eval_loss": 3.4563968181610107,
"eval_runtime": 257.722,
"eval_samples_per_second": 468.357,
"eval_steps_per_second": 4.881,
"step": 440000
},
{
"epoch": 18.568477848501455,
"grad_norm": 1.1586599349975586,
"learning_rate": 2.616730266260773e-05,
"loss": 3.6736,
"step": 440500
},
{
"epoch": 18.589554440838004,
"grad_norm": 1.1611191034317017,
"learning_rate": 2.6081550400891825e-05,
"loss": 3.6762,
"step": 441000
},
{
"epoch": 18.610631033174556,
"grad_norm": 1.166339635848999,
"learning_rate": 2.599579813917592e-05,
"loss": 3.6755,
"step": 441500
},
{
"epoch": 18.63170762551111,
"grad_norm": 1.1848188638687134,
"learning_rate": 2.591004587746002e-05,
"loss": 3.675,
"step": 442000
},
{
"epoch": 18.652784217847657,
"grad_norm": 1.1607309579849243,
"learning_rate": 2.5824293615744116e-05,
"loss": 3.6755,
"step": 442500
},
{
"epoch": 18.67386081018421,
"grad_norm": 1.175057291984558,
"learning_rate": 2.573854135402821e-05,
"loss": 3.6778,
"step": 443000
},
{
"epoch": 18.69493740252076,
"grad_norm": 1.1734440326690674,
"learning_rate": 2.5652789092312313e-05,
"loss": 3.6734,
"step": 443500
},
{
"epoch": 18.716013994857313,
"grad_norm": 1.177460789680481,
"learning_rate": 2.5567036830596408e-05,
"loss": 3.6741,
"step": 444000
},
{
"epoch": 18.737090587193862,
"grad_norm": 1.1912283897399902,
"learning_rate": 2.5481284568880503e-05,
"loss": 3.6761,
"step": 444500
},
{
"epoch": 18.758167179530414,
"grad_norm": 1.1738898754119873,
"learning_rate": 2.5395532307164605e-05,
"loss": 3.6698,
"step": 445000
},
{
"epoch": 18.779243771866966,
"grad_norm": 1.1905044317245483,
"learning_rate": 2.53097800454487e-05,
"loss": 3.6782,
"step": 445500
},
{
"epoch": 18.800320364203515,
"grad_norm": 1.1484711170196533,
"learning_rate": 2.5224027783732794e-05,
"loss": 3.6759,
"step": 446000
},
{
"epoch": 18.821396956540067,
"grad_norm": 1.218059778213501,
"learning_rate": 2.5138275522016896e-05,
"loss": 3.6672,
"step": 446500
},
{
"epoch": 18.84247354887662,
"grad_norm": 1.1417310237884521,
"learning_rate": 2.505252326030099e-05,
"loss": 3.6731,
"step": 447000
},
{
"epoch": 18.863550141213167,
"grad_norm": 1.1796529293060303,
"learning_rate": 2.496677099858509e-05,
"loss": 3.6714,
"step": 447500
},
{
"epoch": 18.88462673354972,
"grad_norm": 1.1395533084869385,
"learning_rate": 2.4881018736869184e-05,
"loss": 3.6755,
"step": 448000
},
{
"epoch": 18.90570332588627,
"grad_norm": 1.1728287935256958,
"learning_rate": 2.4795266475153283e-05,
"loss": 3.671,
"step": 448500
},
{
"epoch": 18.92677991822282,
"grad_norm": 1.1570111513137817,
"learning_rate": 2.470951421343738e-05,
"loss": 3.6768,
"step": 449000
},
{
"epoch": 18.947856510559372,
"grad_norm": 1.1332066059112549,
"learning_rate": 2.4623761951721476e-05,
"loss": 3.67,
"step": 449500
},
{
"epoch": 18.968933102895924,
"grad_norm": 1.1674747467041016,
"learning_rate": 2.4538009690005574e-05,
"loss": 3.6726,
"step": 450000
},
{
"epoch": 18.968933102895924,
"eval_accuracy": 0.7527566413942954,
"eval_loss": 3.4499077796936035,
"eval_runtime": 256.0489,
"eval_samples_per_second": 471.418,
"eval_steps_per_second": 4.913,
"step": 450000
},
{
"epoch": 18.990009695232473,
"grad_norm": 1.1675111055374146,
"learning_rate": 2.4452257428289672e-05,
"loss": 3.6689,
"step": 450500
},
{
"epoch": 19.011086287569025,
"grad_norm": 1.198738694190979,
"learning_rate": 2.436650516657377e-05,
"loss": 3.6692,
"step": 451000
},
{
"epoch": 19.032162879905577,
"grad_norm": 1.1789628267288208,
"learning_rate": 2.428075290485787e-05,
"loss": 3.6676,
"step": 451500
},
{
"epoch": 19.05323947224213,
"grad_norm": 1.1698846817016602,
"learning_rate": 2.4195000643141964e-05,
"loss": 3.6699,
"step": 452000
},
{
"epoch": 19.074316064578678,
"grad_norm": 1.1759222745895386,
"learning_rate": 2.4109248381426062e-05,
"loss": 3.6715,
"step": 452500
},
{
"epoch": 19.09539265691523,
"grad_norm": 1.1428979635238647,
"learning_rate": 2.402349611971016e-05,
"loss": 3.6698,
"step": 453000
},
{
"epoch": 19.116469249251782,
"grad_norm": 1.133567452430725,
"learning_rate": 2.3937743857994255e-05,
"loss": 3.6678,
"step": 453500
},
{
"epoch": 19.13754584158833,
"grad_norm": 1.1864464282989502,
"learning_rate": 2.3851991596278354e-05,
"loss": 3.6694,
"step": 454000
},
{
"epoch": 19.158622433924883,
"grad_norm": 1.2190135717391968,
"learning_rate": 2.3766239334562452e-05,
"loss": 3.6662,
"step": 454500
},
{
"epoch": 19.179699026261435,
"grad_norm": 1.1743699312210083,
"learning_rate": 2.3680487072846547e-05,
"loss": 3.6634,
"step": 455000
},
{
"epoch": 19.200775618597984,
"grad_norm": 1.181689977645874,
"learning_rate": 2.3594734811130645e-05,
"loss": 3.6643,
"step": 455500
},
{
"epoch": 19.221852210934536,
"grad_norm": 1.1571180820465088,
"learning_rate": 2.3508982549414744e-05,
"loss": 3.665,
"step": 456000
},
{
"epoch": 19.242928803271088,
"grad_norm": 1.1632943153381348,
"learning_rate": 2.342323028769884e-05,
"loss": 3.667,
"step": 456500
},
{
"epoch": 19.264005395607636,
"grad_norm": 1.1506397724151611,
"learning_rate": 2.3337478025982937e-05,
"loss": 3.664,
"step": 457000
},
{
"epoch": 19.28508198794419,
"grad_norm": 1.2267950773239136,
"learning_rate": 2.3251725764267035e-05,
"loss": 3.6675,
"step": 457500
},
{
"epoch": 19.30615858028074,
"grad_norm": 1.1446926593780518,
"learning_rate": 2.316597350255113e-05,
"loss": 3.6702,
"step": 458000
},
{
"epoch": 19.327235172617293,
"grad_norm": 1.1562714576721191,
"learning_rate": 2.308022124083523e-05,
"loss": 3.668,
"step": 458500
},
{
"epoch": 19.34831176495384,
"grad_norm": 1.1942178010940552,
"learning_rate": 2.2994468979119323e-05,
"loss": 3.6675,
"step": 459000
},
{
"epoch": 19.369388357290394,
"grad_norm": 1.1838908195495605,
"learning_rate": 2.290871671740342e-05,
"loss": 3.6655,
"step": 459500
},
{
"epoch": 19.390464949626946,
"grad_norm": 1.162261724472046,
"learning_rate": 2.282296445568752e-05,
"loss": 3.6668,
"step": 460000
},
{
"epoch": 19.390464949626946,
"eval_accuracy": 0.7533798338516956,
"eval_loss": 3.4467790126800537,
"eval_runtime": 256.3045,
"eval_samples_per_second": 470.948,
"eval_steps_per_second": 4.908,
"step": 460000
},
{
"epoch": 19.411541541963494,
"grad_norm": 1.183319330215454,
"learning_rate": 2.2737212193971615e-05,
"loss": 3.6637,
"step": 460500
},
{
"epoch": 19.432618134300046,
"grad_norm": 1.1829698085784912,
"learning_rate": 2.2651459932255713e-05,
"loss": 3.6686,
"step": 461000
},
{
"epoch": 19.4536947266366,
"grad_norm": 1.1465206146240234,
"learning_rate": 2.256570767053981e-05,
"loss": 3.6654,
"step": 461500
},
{
"epoch": 19.474771318973147,
"grad_norm": 1.1733912229537964,
"learning_rate": 2.2479955408823906e-05,
"loss": 3.6638,
"step": 462000
},
{
"epoch": 19.4958479113097,
"grad_norm": 1.1979892253875732,
"learning_rate": 2.2394203147108005e-05,
"loss": 3.6652,
"step": 462500
},
{
"epoch": 19.51692450364625,
"grad_norm": 1.2293606996536255,
"learning_rate": 2.2308450885392103e-05,
"loss": 3.6629,
"step": 463000
},
{
"epoch": 19.5380010959828,
"grad_norm": 1.175973892211914,
"learning_rate": 2.22226986236762e-05,
"loss": 3.663,
"step": 463500
},
{
"epoch": 19.559077688319352,
"grad_norm": 1.1402400732040405,
"learning_rate": 2.21369463619603e-05,
"loss": 3.6658,
"step": 464000
},
{
"epoch": 19.580154280655904,
"grad_norm": 1.2262070178985596,
"learning_rate": 2.2051194100244394e-05,
"loss": 3.6638,
"step": 464500
},
{
"epoch": 19.601230872992456,
"grad_norm": 1.2079238891601562,
"learning_rate": 2.1965441838528493e-05,
"loss": 3.6638,
"step": 465000
},
{
"epoch": 19.622307465329005,
"grad_norm": 1.1621575355529785,
"learning_rate": 2.187968957681259e-05,
"loss": 3.6634,
"step": 465500
},
{
"epoch": 19.643384057665557,
"grad_norm": 1.126846432685852,
"learning_rate": 2.179393731509669e-05,
"loss": 3.6643,
"step": 466000
},
{
"epoch": 19.66446065000211,
"grad_norm": 1.1841880083084106,
"learning_rate": 2.1708185053380784e-05,
"loss": 3.6633,
"step": 466500
},
{
"epoch": 19.685537242338658,
"grad_norm": 1.1684527397155762,
"learning_rate": 2.1622432791664883e-05,
"loss": 3.6653,
"step": 467000
},
{
"epoch": 19.70661383467521,
"grad_norm": 1.163293480873108,
"learning_rate": 2.1536680529948978e-05,
"loss": 3.6625,
"step": 467500
},
{
"epoch": 19.727690427011762,
"grad_norm": 1.1829620599746704,
"learning_rate": 2.1450928268233076e-05,
"loss": 3.6655,
"step": 468000
},
{
"epoch": 19.74876701934831,
"grad_norm": 1.1562832593917847,
"learning_rate": 2.1365176006517174e-05,
"loss": 3.6624,
"step": 468500
},
{
"epoch": 19.769843611684863,
"grad_norm": 1.1888495683670044,
"learning_rate": 2.127942374480127e-05,
"loss": 3.6628,
"step": 469000
},
{
"epoch": 19.790920204021415,
"grad_norm": 1.236480951309204,
"learning_rate": 2.1193671483085367e-05,
"loss": 3.6661,
"step": 469500
},
{
"epoch": 19.811996796357963,
"grad_norm": 1.1794041395187378,
"learning_rate": 2.1107919221369466e-05,
"loss": 3.663,
"step": 470000
},
{
"epoch": 19.811996796357963,
"eval_accuracy": 0.7540568910506241,
"eval_loss": 3.442244052886963,
"eval_runtime": 255.0055,
"eval_samples_per_second": 473.347,
"eval_steps_per_second": 4.933,
"step": 470000
},
{
"epoch": 19.833073388694515,
"grad_norm": 1.1411206722259521,
"learning_rate": 2.102216695965356e-05,
"loss": 3.6648,
"step": 470500
},
{
"epoch": 19.854149981031068,
"grad_norm": 1.1823804378509521,
"learning_rate": 2.093641469793766e-05,
"loss": 3.6626,
"step": 471000
},
{
"epoch": 19.87522657336762,
"grad_norm": 1.1452839374542236,
"learning_rate": 2.0850662436221757e-05,
"loss": 3.6615,
"step": 471500
},
{
"epoch": 19.89630316570417,
"grad_norm": 1.196459174156189,
"learning_rate": 2.0764910174505852e-05,
"loss": 3.6608,
"step": 472000
},
{
"epoch": 19.91737975804072,
"grad_norm": 1.1701470613479614,
"learning_rate": 2.067915791278995e-05,
"loss": 3.6644,
"step": 472500
},
{
"epoch": 19.938456350377272,
"grad_norm": 1.234171986579895,
"learning_rate": 2.059340565107405e-05,
"loss": 3.6577,
"step": 473000
},
{
"epoch": 19.95953294271382,
"grad_norm": 1.1414179801940918,
"learning_rate": 2.0507653389358144e-05,
"loss": 3.6618,
"step": 473500
},
{
"epoch": 19.980609535050373,
"grad_norm": 1.187326431274414,
"learning_rate": 2.0421901127642242e-05,
"loss": 3.6596,
"step": 474000
},
{
"epoch": 20.001686127386925,
"grad_norm": 1.2053556442260742,
"learning_rate": 2.033614886592634e-05,
"loss": 3.6662,
"step": 474500
},
{
"epoch": 20.022762719723474,
"grad_norm": 1.1715282201766968,
"learning_rate": 2.0250396604210435e-05,
"loss": 3.6597,
"step": 475000
},
{
"epoch": 20.043839312060026,
"grad_norm": 1.2438955307006836,
"learning_rate": 2.0164644342494534e-05,
"loss": 3.6601,
"step": 475500
},
{
"epoch": 20.064915904396578,
"grad_norm": 1.228310227394104,
"learning_rate": 2.007889208077863e-05,
"loss": 3.6539,
"step": 476000
},
{
"epoch": 20.085992496733127,
"grad_norm": 1.1748828887939453,
"learning_rate": 1.9993139819062727e-05,
"loss": 3.6566,
"step": 476500
},
{
"epoch": 20.10706908906968,
"grad_norm": 1.1964792013168335,
"learning_rate": 1.9907387557346825e-05,
"loss": 3.6582,
"step": 477000
},
{
"epoch": 20.12814568140623,
"grad_norm": 1.1523553133010864,
"learning_rate": 1.9821635295630923e-05,
"loss": 3.6576,
"step": 477500
},
{
"epoch": 20.14922227374278,
"grad_norm": 1.1931657791137695,
"learning_rate": 1.9735883033915022e-05,
"loss": 3.6557,
"step": 478000
},
{
"epoch": 20.17029886607933,
"grad_norm": 1.1677428483963013,
"learning_rate": 1.965013077219912e-05,
"loss": 3.6597,
"step": 478500
},
{
"epoch": 20.191375458415884,
"grad_norm": 1.2224293947219849,
"learning_rate": 1.9564378510483215e-05,
"loss": 3.6552,
"step": 479000
},
{
"epoch": 20.212452050752436,
"grad_norm": 1.1597537994384766,
"learning_rate": 1.9478626248767313e-05,
"loss": 3.6585,
"step": 479500
},
{
"epoch": 20.233528643088984,
"grad_norm": 1.2317051887512207,
"learning_rate": 1.939287398705141e-05,
"loss": 3.656,
"step": 480000
},
{
"epoch": 20.233528643088984,
"eval_accuracy": 0.7546488273694968,
"eval_loss": 3.4387309551239014,
"eval_runtime": 256.4484,
"eval_samples_per_second": 470.683,
"eval_steps_per_second": 4.905,
"step": 480000
},
{
"epoch": 20.254605235425537,
"grad_norm": 1.135244607925415,
"learning_rate": 1.9307121725335506e-05,
"loss": 3.6523,
"step": 480500
},
{
"epoch": 20.27568182776209,
"grad_norm": 1.2045842409133911,
"learning_rate": 1.9221369463619605e-05,
"loss": 3.6539,
"step": 481000
},
{
"epoch": 20.296758420098637,
"grad_norm": 1.1729049682617188,
"learning_rate": 1.9135617201903703e-05,
"loss": 3.6532,
"step": 481500
},
{
"epoch": 20.31783501243519,
"grad_norm": 1.1982076168060303,
"learning_rate": 1.9049864940187798e-05,
"loss": 3.6548,
"step": 482000
},
{
"epoch": 20.33891160477174,
"grad_norm": 1.1967219114303589,
"learning_rate": 1.8964112678471896e-05,
"loss": 3.6564,
"step": 482500
},
{
"epoch": 20.35998819710829,
"grad_norm": 1.200156331062317,
"learning_rate": 1.8878360416755995e-05,
"loss": 3.6558,
"step": 483000
},
{
"epoch": 20.381064789444842,
"grad_norm": 1.1751731634140015,
"learning_rate": 1.879260815504009e-05,
"loss": 3.6543,
"step": 483500
},
{
"epoch": 20.402141381781394,
"grad_norm": 1.1636532545089722,
"learning_rate": 1.8706855893324188e-05,
"loss": 3.6508,
"step": 484000
},
{
"epoch": 20.423217974117943,
"grad_norm": 1.1666349172592163,
"learning_rate": 1.8621103631608283e-05,
"loss": 3.6547,
"step": 484500
},
{
"epoch": 20.444294566454495,
"grad_norm": 1.1803529262542725,
"learning_rate": 1.853535136989238e-05,
"loss": 3.6545,
"step": 485000
},
{
"epoch": 20.465371158791047,
"grad_norm": 1.1866021156311035,
"learning_rate": 1.844959910817648e-05,
"loss": 3.6528,
"step": 485500
},
{
"epoch": 20.4864477511276,
"grad_norm": 1.2033007144927979,
"learning_rate": 1.8363846846460574e-05,
"loss": 3.6549,
"step": 486000
},
{
"epoch": 20.507524343464148,
"grad_norm": 1.1976654529571533,
"learning_rate": 1.8278094584744673e-05,
"loss": 3.6513,
"step": 486500
},
{
"epoch": 20.5286009358007,
"grad_norm": 1.2172843217849731,
"learning_rate": 1.819234232302877e-05,
"loss": 3.6558,
"step": 487000
},
{
"epoch": 20.549677528137252,
"grad_norm": 1.1715760231018066,
"learning_rate": 1.8106590061312866e-05,
"loss": 3.6552,
"step": 487500
},
{
"epoch": 20.5707541204738,
"grad_norm": 1.1693028211593628,
"learning_rate": 1.8020837799596964e-05,
"loss": 3.6557,
"step": 488000
},
{
"epoch": 20.591830712810353,
"grad_norm": 1.1565238237380981,
"learning_rate": 1.7935085537881062e-05,
"loss": 3.6555,
"step": 488500
},
{
"epoch": 20.612907305146905,
"grad_norm": 1.1624950170516968,
"learning_rate": 1.7849333276165157e-05,
"loss": 3.6536,
"step": 489000
},
{
"epoch": 20.633983897483454,
"grad_norm": 1.2025693655014038,
"learning_rate": 1.7763581014449256e-05,
"loss": 3.6524,
"step": 489500
},
{
"epoch": 20.655060489820006,
"grad_norm": 1.1723138093948364,
"learning_rate": 1.7677828752733354e-05,
"loss": 3.6499,
"step": 490000
},
{
"epoch": 20.655060489820006,
"eval_accuracy": 0.7554408709551229,
"eval_loss": 3.435163736343384,
"eval_runtime": 255.5913,
"eval_samples_per_second": 472.262,
"eval_steps_per_second": 4.922,
"step": 490000
},
{
"epoch": 20.676137082156558,
"grad_norm": 1.218497395515442,
"learning_rate": 1.7592076491017452e-05,
"loss": 3.654,
"step": 490500
},
{
"epoch": 20.697213674493106,
"grad_norm": 1.1935006380081177,
"learning_rate": 1.7506324229301547e-05,
"loss": 3.6555,
"step": 491000
},
{
"epoch": 20.71829026682966,
"grad_norm": 1.20321786403656,
"learning_rate": 1.7420571967585646e-05,
"loss": 3.6546,
"step": 491500
},
{
"epoch": 20.73936685916621,
"grad_norm": 1.2118149995803833,
"learning_rate": 1.7334819705869744e-05,
"loss": 3.651,
"step": 492000
},
{
"epoch": 20.760443451502763,
"grad_norm": 1.2187227010726929,
"learning_rate": 1.7249067444153842e-05,
"loss": 3.6547,
"step": 492500
},
{
"epoch": 20.78152004383931,
"grad_norm": 1.1892955303192139,
"learning_rate": 1.7163315182437937e-05,
"loss": 3.6482,
"step": 493000
},
{
"epoch": 20.802596636175863,
"grad_norm": 1.1712920665740967,
"learning_rate": 1.7077562920722035e-05,
"loss": 3.6516,
"step": 493500
},
{
"epoch": 20.823673228512416,
"grad_norm": 1.1569304466247559,
"learning_rate": 1.6991810659006134e-05,
"loss": 3.6528,
"step": 494000
},
{
"epoch": 20.844749820848964,
"grad_norm": 1.1458454132080078,
"learning_rate": 1.690605839729023e-05,
"loss": 3.6525,
"step": 494500
},
{
"epoch": 20.865826413185516,
"grad_norm": 1.1724021434783936,
"learning_rate": 1.6820306135574327e-05,
"loss": 3.6514,
"step": 495000
},
{
"epoch": 20.88690300552207,
"grad_norm": 1.217012643814087,
"learning_rate": 1.6734553873858425e-05,
"loss": 3.6522,
"step": 495500
},
{
"epoch": 20.907979597858617,
"grad_norm": 1.1850290298461914,
"learning_rate": 1.664880161214252e-05,
"loss": 3.651,
"step": 496000
},
{
"epoch": 20.92905619019517,
"grad_norm": 1.1969726085662842,
"learning_rate": 1.656304935042662e-05,
"loss": 3.6543,
"step": 496500
},
{
"epoch": 20.95013278253172,
"grad_norm": 1.168127179145813,
"learning_rate": 1.6477297088710717e-05,
"loss": 3.6493,
"step": 497000
},
{
"epoch": 20.97120937486827,
"grad_norm": 1.1778851747512817,
"learning_rate": 1.639154482699481e-05,
"loss": 3.6498,
"step": 497500
},
{
"epoch": 20.992285967204822,
"grad_norm": 1.1762205362319946,
"learning_rate": 1.630579256527891e-05,
"loss": 3.6505,
"step": 498000
},
{
"epoch": 21.013362559541374,
"grad_norm": 1.1878337860107422,
"learning_rate": 1.6220040303563008e-05,
"loss": 3.6493,
"step": 498500
},
{
"epoch": 21.034439151877926,
"grad_norm": 1.1805577278137207,
"learning_rate": 1.6134288041847103e-05,
"loss": 3.6504,
"step": 499000
},
{
"epoch": 21.055515744214475,
"grad_norm": 1.185826063156128,
"learning_rate": 1.60485357801312e-05,
"loss": 3.6471,
"step": 499500
},
{
"epoch": 21.076592336551027,
"grad_norm": 1.1767042875289917,
"learning_rate": 1.59627835184153e-05,
"loss": 3.6462,
"step": 500000
},
{
"epoch": 21.076592336551027,
"eval_accuracy": 0.7558517294855452,
"eval_loss": 3.430654525756836,
"eval_runtime": 254.8879,
"eval_samples_per_second": 473.565,
"eval_steps_per_second": 4.936,
"step": 500000
},
{
"epoch": 21.09766892888758,
"grad_norm": 1.1800049543380737,
"learning_rate": 1.5877031256699395e-05,
"loss": 3.6464,
"step": 500500
},
{
"epoch": 21.118745521224128,
"grad_norm": 1.157139778137207,
"learning_rate": 1.5791278994983493e-05,
"loss": 3.6481,
"step": 501000
},
{
"epoch": 21.13982211356068,
"grad_norm": 1.1528403759002686,
"learning_rate": 1.570552673326759e-05,
"loss": 3.6451,
"step": 501500
},
{
"epoch": 21.16089870589723,
"grad_norm": 1.171179175376892,
"learning_rate": 1.5619774471551686e-05,
"loss": 3.6478,
"step": 502000
},
{
"epoch": 21.18197529823378,
"grad_norm": 1.137410044670105,
"learning_rate": 1.5534022209835785e-05,
"loss": 3.649,
"step": 502500
},
{
"epoch": 21.203051890570332,
"grad_norm": 1.1530464887619019,
"learning_rate": 1.544826994811988e-05,
"loss": 3.6469,
"step": 503000
},
{
"epoch": 21.224128482906885,
"grad_norm": 1.1945103406906128,
"learning_rate": 1.5362517686403978e-05,
"loss": 3.647,
"step": 503500
},
{
"epoch": 21.245205075243433,
"grad_norm": 1.1731829643249512,
"learning_rate": 1.5276765424688076e-05,
"loss": 3.6452,
"step": 504000
},
{
"epoch": 21.266281667579985,
"grad_norm": 1.2280714511871338,
"learning_rate": 1.5191013162972173e-05,
"loss": 3.644,
"step": 504500
},
{
"epoch": 21.287358259916537,
"grad_norm": 1.2395403385162354,
"learning_rate": 1.5105260901256271e-05,
"loss": 3.646,
"step": 505000
},
{
"epoch": 21.308434852253086,
"grad_norm": 1.2107930183410645,
"learning_rate": 1.501950863954037e-05,
"loss": 3.6467,
"step": 505500
},
{
"epoch": 21.329511444589638,
"grad_norm": 1.183016061782837,
"learning_rate": 1.4933756377824464e-05,
"loss": 3.6426,
"step": 506000
},
{
"epoch": 21.35058803692619,
"grad_norm": 1.1768831014633179,
"learning_rate": 1.4848004116108563e-05,
"loss": 3.6411,
"step": 506500
},
{
"epoch": 21.371664629262742,
"grad_norm": 1.1800588369369507,
"learning_rate": 1.4762251854392661e-05,
"loss": 3.6478,
"step": 507000
},
{
"epoch": 21.39274122159929,
"grad_norm": 1.1567245721817017,
"learning_rate": 1.4676499592676758e-05,
"loss": 3.6473,
"step": 507500
},
{
"epoch": 21.413817813935843,
"grad_norm": 1.1526093482971191,
"learning_rate": 1.4590747330960856e-05,
"loss": 3.6448,
"step": 508000
},
{
"epoch": 21.434894406272395,
"grad_norm": 1.1744813919067383,
"learning_rate": 1.4504995069244952e-05,
"loss": 3.6435,
"step": 508500
},
{
"epoch": 21.455970998608944,
"grad_norm": 1.1538681983947754,
"learning_rate": 1.4419242807529049e-05,
"loss": 3.6463,
"step": 509000
},
{
"epoch": 21.477047590945496,
"grad_norm": 1.137689232826233,
"learning_rate": 1.4333490545813147e-05,
"loss": 3.6456,
"step": 509500
},
{
"epoch": 21.498124183282048,
"grad_norm": 1.205203890800476,
"learning_rate": 1.4247738284097246e-05,
"loss": 3.6423,
"step": 510000
},
{
"epoch": 21.498124183282048,
"eval_accuracy": 0.756579593383006,
"eval_loss": 3.427316427230835,
"eval_runtime": 255.2956,
"eval_samples_per_second": 472.809,
"eval_steps_per_second": 4.928,
"step": 510000
},
{
"epoch": 21.519200775618597,
"grad_norm": 1.1524887084960938,
"learning_rate": 1.416198602238134e-05,
"loss": 3.6457,
"step": 510500
},
{
"epoch": 21.54027736795515,
"grad_norm": 1.1380053758621216,
"learning_rate": 1.4076233760665439e-05,
"loss": 3.643,
"step": 511000
},
{
"epoch": 21.5613539602917,
"grad_norm": 1.1726975440979004,
"learning_rate": 1.3990481498949534e-05,
"loss": 3.6403,
"step": 511500
},
{
"epoch": 21.58243055262825,
"grad_norm": 1.1960371732711792,
"learning_rate": 1.3904729237233632e-05,
"loss": 3.6459,
"step": 512000
},
{
"epoch": 21.6035071449648,
"grad_norm": 1.1531378030776978,
"learning_rate": 1.381897697551773e-05,
"loss": 3.6425,
"step": 512500
},
{
"epoch": 21.624583737301354,
"grad_norm": 1.2037007808685303,
"learning_rate": 1.3733224713801825e-05,
"loss": 3.6464,
"step": 513000
},
{
"epoch": 21.645660329637906,
"grad_norm": 1.2140535116195679,
"learning_rate": 1.3647472452085924e-05,
"loss": 3.6472,
"step": 513500
},
{
"epoch": 21.666736921974454,
"grad_norm": 1.1850353479385376,
"learning_rate": 1.3561720190370022e-05,
"loss": 3.6454,
"step": 514000
},
{
"epoch": 21.687813514311006,
"grad_norm": 1.1957803964614868,
"learning_rate": 1.3475967928654119e-05,
"loss": 3.6432,
"step": 514500
},
{
"epoch": 21.70889010664756,
"grad_norm": 1.1872490644454956,
"learning_rate": 1.3390215666938217e-05,
"loss": 3.6448,
"step": 515000
},
{
"epoch": 21.729966698984107,
"grad_norm": 1.1576117277145386,
"learning_rate": 1.3304463405222315e-05,
"loss": 3.6446,
"step": 515500
},
{
"epoch": 21.75104329132066,
"grad_norm": 1.1819888353347778,
"learning_rate": 1.321871114350641e-05,
"loss": 3.6467,
"step": 516000
},
{
"epoch": 21.77211988365721,
"grad_norm": 1.1919713020324707,
"learning_rate": 1.3132958881790508e-05,
"loss": 3.6396,
"step": 516500
},
{
"epoch": 21.79319647599376,
"grad_norm": 1.1756795644760132,
"learning_rate": 1.3047206620074607e-05,
"loss": 3.6417,
"step": 517000
},
{
"epoch": 21.814273068330312,
"grad_norm": 1.1938257217407227,
"learning_rate": 1.2961454358358702e-05,
"loss": 3.644,
"step": 517500
},
{
"epoch": 21.835349660666864,
"grad_norm": 1.2280433177947998,
"learning_rate": 1.28757020966428e-05,
"loss": 3.6404,
"step": 518000
},
{
"epoch": 21.856426253003413,
"grad_norm": 1.1832849979400635,
"learning_rate": 1.2789949834926898e-05,
"loss": 3.6392,
"step": 518500
},
{
"epoch": 21.877502845339965,
"grad_norm": 1.1834074258804321,
"learning_rate": 1.2704197573210993e-05,
"loss": 3.6465,
"step": 519000
},
{
"epoch": 21.898579437676517,
"grad_norm": 1.1685230731964111,
"learning_rate": 1.2618445311495092e-05,
"loss": 3.6392,
"step": 519500
},
{
"epoch": 21.91965603001307,
"grad_norm": 1.2267236709594727,
"learning_rate": 1.2532693049779186e-05,
"loss": 3.6374,
"step": 520000
},
{
"epoch": 21.91965603001307,
"eval_accuracy": 0.7568859321582689,
"eval_loss": 3.4251790046691895,
"eval_runtime": 253.5146,
"eval_samples_per_second": 476.13,
"eval_steps_per_second": 4.962,
"step": 520000
},
{
"epoch": 21.940732622349618,
"grad_norm": 1.2148746252059937,
"learning_rate": 1.2446940788063285e-05,
"loss": 3.6396,
"step": 520500
},
{
"epoch": 21.96180921468617,
"grad_norm": 1.1989517211914062,
"learning_rate": 1.2361188526347383e-05,
"loss": 3.6415,
"step": 521000
},
{
"epoch": 21.982885807022722,
"grad_norm": 1.1796672344207764,
"learning_rate": 1.227543626463148e-05,
"loss": 3.6418,
"step": 521500
},
{
"epoch": 22.00396239935927,
"grad_norm": 1.2314836978912354,
"learning_rate": 1.2189684002915578e-05,
"loss": 3.6392,
"step": 522000
},
{
"epoch": 22.025038991695823,
"grad_norm": 1.2306203842163086,
"learning_rate": 1.2103931741199675e-05,
"loss": 3.6368,
"step": 522500
},
{
"epoch": 22.046115584032375,
"grad_norm": 1.210378646850586,
"learning_rate": 1.2018179479483773e-05,
"loss": 3.6387,
"step": 523000
},
{
"epoch": 22.067192176368923,
"grad_norm": 1.1683293581008911,
"learning_rate": 1.193242721776787e-05,
"loss": 3.6395,
"step": 523500
},
{
"epoch": 22.088268768705476,
"grad_norm": 1.2244541645050049,
"learning_rate": 1.1846674956051966e-05,
"loss": 3.6402,
"step": 524000
},
{
"epoch": 22.109345361042028,
"grad_norm": 1.2194442749023438,
"learning_rate": 1.1760922694336064e-05,
"loss": 3.6352,
"step": 524500
},
{
"epoch": 22.130421953378576,
"grad_norm": 1.204868197441101,
"learning_rate": 1.1675170432620161e-05,
"loss": 3.6408,
"step": 525000
},
{
"epoch": 22.15149854571513,
"grad_norm": 1.1633797883987427,
"learning_rate": 1.1589418170904258e-05,
"loss": 3.6367,
"step": 525500
},
{
"epoch": 22.17257513805168,
"grad_norm": 1.1597599983215332,
"learning_rate": 1.1503665909188354e-05,
"loss": 3.6368,
"step": 526000
},
{
"epoch": 22.193651730388233,
"grad_norm": 1.242363691329956,
"learning_rate": 1.1417913647472453e-05,
"loss": 3.6373,
"step": 526500
},
{
"epoch": 22.21472832272478,
"grad_norm": 1.2096441984176636,
"learning_rate": 1.133216138575655e-05,
"loss": 3.6379,
"step": 527000
},
{
"epoch": 22.235804915061333,
"grad_norm": 1.2314869165420532,
"learning_rate": 1.1246409124040646e-05,
"loss": 3.6361,
"step": 527500
},
{
"epoch": 22.256881507397885,
"grad_norm": 1.2146505117416382,
"learning_rate": 1.1160656862324744e-05,
"loss": 3.6389,
"step": 528000
},
{
"epoch": 22.277958099734434,
"grad_norm": 1.1914061307907104,
"learning_rate": 1.1074904600608842e-05,
"loss": 3.6354,
"step": 528500
},
{
"epoch": 22.299034692070986,
"grad_norm": 1.2115174531936646,
"learning_rate": 1.0989152338892939e-05,
"loss": 3.6346,
"step": 529000
},
{
"epoch": 22.320111284407538,
"grad_norm": 1.2348052263259888,
"learning_rate": 1.0903400077177037e-05,
"loss": 3.6378,
"step": 529500
},
{
"epoch": 22.341187876744087,
"grad_norm": 1.162584900856018,
"learning_rate": 1.0817647815461134e-05,
"loss": 3.6389,
"step": 530000
},
{
"epoch": 22.341187876744087,
"eval_accuracy": 0.7573940522269202,
"eval_loss": 3.4224727153778076,
"eval_runtime": 255.9944,
"eval_samples_per_second": 471.518,
"eval_steps_per_second": 4.914,
"step": 530000
},
{
"epoch": 22.36226446908064,
"grad_norm": 1.1808884143829346,
"learning_rate": 1.073189555374523e-05,
"loss": 3.6387,
"step": 530500
},
{
"epoch": 22.38334106141719,
"grad_norm": 1.2487386465072632,
"learning_rate": 1.0646143292029327e-05,
"loss": 3.6395,
"step": 531000
},
{
"epoch": 22.40441765375374,
"grad_norm": 1.1953319311141968,
"learning_rate": 1.0560391030313425e-05,
"loss": 3.6317,
"step": 531500
},
{
"epoch": 22.42549424609029,
"grad_norm": 1.1957013607025146,
"learning_rate": 1.0474638768597522e-05,
"loss": 3.6342,
"step": 532000
},
{
"epoch": 22.446570838426844,
"grad_norm": 1.2204194068908691,
"learning_rate": 1.0388886506881619e-05,
"loss": 3.6357,
"step": 532500
},
{
"epoch": 22.467647430763392,
"grad_norm": 1.2203528881072998,
"learning_rate": 1.0303134245165717e-05,
"loss": 3.637,
"step": 533000
},
{
"epoch": 22.488724023099945,
"grad_norm": 1.2093323469161987,
"learning_rate": 1.0217381983449814e-05,
"loss": 3.6338,
"step": 533500
},
{
"epoch": 22.509800615436497,
"grad_norm": 1.1529541015625,
"learning_rate": 1.013162972173391e-05,
"loss": 3.6346,
"step": 534000
},
{
"epoch": 22.53087720777305,
"grad_norm": 1.1957097053527832,
"learning_rate": 1.0045877460018009e-05,
"loss": 3.6375,
"step": 534500
},
{
"epoch": 22.551953800109597,
"grad_norm": 1.2042196989059448,
"learning_rate": 9.960125198302105e-06,
"loss": 3.6391,
"step": 535000
},
{
"epoch": 22.57303039244615,
"grad_norm": 1.1739212274551392,
"learning_rate": 9.874372936586203e-06,
"loss": 3.6369,
"step": 535500
},
{
"epoch": 22.5941069847827,
"grad_norm": 1.195076823234558,
"learning_rate": 9.7886206748703e-06,
"loss": 3.6314,
"step": 536000
},
{
"epoch": 22.61518357711925,
"grad_norm": 1.2037856578826904,
"learning_rate": 9.702868413154398e-06,
"loss": 3.6331,
"step": 536500
},
{
"epoch": 22.636260169455802,
"grad_norm": 1.2251296043395996,
"learning_rate": 9.617116151438495e-06,
"loss": 3.6346,
"step": 537000
},
{
"epoch": 22.657336761792354,
"grad_norm": 1.20695161819458,
"learning_rate": 9.531363889722592e-06,
"loss": 3.6347,
"step": 537500
},
{
"epoch": 22.678413354128903,
"grad_norm": 1.170993685722351,
"learning_rate": 9.44561162800669e-06,
"loss": 3.6326,
"step": 538000
},
{
"epoch": 22.699489946465455,
"grad_norm": 1.206341028213501,
"learning_rate": 9.359859366290787e-06,
"loss": 3.6347,
"step": 538500
},
{
"epoch": 22.720566538802007,
"grad_norm": 1.163734793663025,
"learning_rate": 9.274107104574883e-06,
"loss": 3.6353,
"step": 539000
},
{
"epoch": 22.741643131138556,
"grad_norm": 1.1591821908950806,
"learning_rate": 9.18835484285898e-06,
"loss": 3.6315,
"step": 539500
},
{
"epoch": 22.762719723475108,
"grad_norm": 1.228010892868042,
"learning_rate": 9.102602581143078e-06,
"loss": 3.6316,
"step": 540000
},
{
"epoch": 22.762719723475108,
"eval_accuracy": 0.7576664807399532,
"eval_loss": 3.421459913253784,
"eval_runtime": 255.3398,
"eval_samples_per_second": 472.727,
"eval_steps_per_second": 4.927,
"step": 540000
},
{
"epoch": 22.78379631581166,
"grad_norm": 1.2144652605056763,
"learning_rate": 9.016850319427175e-06,
"loss": 3.6357,
"step": 540500
},
{
"epoch": 22.804872908148212,
"grad_norm": 1.2013120651245117,
"learning_rate": 8.931098057711271e-06,
"loss": 3.6349,
"step": 541000
},
{
"epoch": 22.82594950048476,
"grad_norm": 1.2087434530258179,
"learning_rate": 8.84534579599537e-06,
"loss": 3.6342,
"step": 541500
},
{
"epoch": 22.847026092821313,
"grad_norm": 1.257639765739441,
"learning_rate": 8.759593534279468e-06,
"loss": 3.6352,
"step": 542000
},
{
"epoch": 22.868102685157865,
"grad_norm": 1.1748807430267334,
"learning_rate": 8.673841272563565e-06,
"loss": 3.6323,
"step": 542500
},
{
"epoch": 22.889179277494414,
"grad_norm": 1.2343976497650146,
"learning_rate": 8.588089010847663e-06,
"loss": 3.6351,
"step": 543000
},
{
"epoch": 22.910255869830966,
"grad_norm": 1.2385674715042114,
"learning_rate": 8.50233674913176e-06,
"loss": 3.6328,
"step": 543500
},
{
"epoch": 22.931332462167518,
"grad_norm": 1.2400726079940796,
"learning_rate": 8.416584487415856e-06,
"loss": 3.6302,
"step": 544000
},
{
"epoch": 22.952409054504066,
"grad_norm": 1.2759028673171997,
"learning_rate": 8.330832225699953e-06,
"loss": 3.6283,
"step": 544500
},
{
"epoch": 22.97348564684062,
"grad_norm": 1.2203872203826904,
"learning_rate": 8.245079963984051e-06,
"loss": 3.6303,
"step": 545000
},
{
"epoch": 22.99456223917717,
"grad_norm": 1.193613052368164,
"learning_rate": 8.159327702268148e-06,
"loss": 3.6346,
"step": 545500
},
{
"epoch": 23.01563883151372,
"grad_norm": 1.1948072910308838,
"learning_rate": 8.073575440552244e-06,
"loss": 3.6317,
"step": 546000
},
{
"epoch": 23.03671542385027,
"grad_norm": 1.1924632787704468,
"learning_rate": 7.987823178836343e-06,
"loss": 3.6296,
"step": 546500
},
{
"epoch": 23.057792016186824,
"grad_norm": 1.2195173501968384,
"learning_rate": 7.90207091712044e-06,
"loss": 3.6293,
"step": 547000
},
{
"epoch": 23.078868608523376,
"grad_norm": 1.270365595817566,
"learning_rate": 7.816318655404536e-06,
"loss": 3.6292,
"step": 547500
},
{
"epoch": 23.099945200859924,
"grad_norm": 1.2233268022537231,
"learning_rate": 7.730566393688634e-06,
"loss": 3.6308,
"step": 548000
},
{
"epoch": 23.121021793196476,
"grad_norm": 1.17930006980896,
"learning_rate": 7.64481413197273e-06,
"loss": 3.6306,
"step": 548500
},
{
"epoch": 23.14209838553303,
"grad_norm": 1.2134041786193848,
"learning_rate": 7.559061870256828e-06,
"loss": 3.6315,
"step": 549000
},
{
"epoch": 23.163174977869577,
"grad_norm": 1.199601650238037,
"learning_rate": 7.473309608540925e-06,
"loss": 3.6297,
"step": 549500
},
{
"epoch": 23.18425157020613,
"grad_norm": 1.2457325458526611,
"learning_rate": 7.387557346825023e-06,
"loss": 3.6323,
"step": 550000
},
{
"epoch": 23.18425157020613,
"eval_accuracy": 0.758181719665861,
"eval_loss": 3.4175033569335938,
"eval_runtime": 255.9243,
"eval_samples_per_second": 471.647,
"eval_steps_per_second": 4.916,
"step": 550000
},
{
"epoch": 23.20532816254268,
"grad_norm": 1.2464157342910767,
"learning_rate": 7.3018050851091205e-06,
"loss": 3.6265,
"step": 550500
},
{
"epoch": 23.22640475487923,
"grad_norm": 1.2051788568496704,
"learning_rate": 7.216052823393217e-06,
"loss": 3.6295,
"step": 551000
},
{
"epoch": 23.247481347215782,
"grad_norm": 1.226716160774231,
"learning_rate": 7.1303005616773155e-06,
"loss": 3.6299,
"step": 551500
},
{
"epoch": 23.268557939552334,
"grad_norm": 1.2218393087387085,
"learning_rate": 7.044548299961412e-06,
"loss": 3.6346,
"step": 552000
},
{
"epoch": 23.289634531888883,
"grad_norm": 1.2295660972595215,
"learning_rate": 6.958796038245509e-06,
"loss": 3.6226,
"step": 552500
},
{
"epoch": 23.310711124225435,
"grad_norm": 1.2974913120269775,
"learning_rate": 6.873043776529606e-06,
"loss": 3.6341,
"step": 553000
},
{
"epoch": 23.331787716561987,
"grad_norm": 1.225721001625061,
"learning_rate": 6.787291514813704e-06,
"loss": 3.6257,
"step": 553500
},
{
"epoch": 23.35286430889854,
"grad_norm": 1.2174464464187622,
"learning_rate": 6.701539253097801e-06,
"loss": 3.6286,
"step": 554000
},
{
"epoch": 23.373940901235088,
"grad_norm": 1.2467906475067139,
"learning_rate": 6.615786991381898e-06,
"loss": 3.6287,
"step": 554500
},
{
"epoch": 23.39501749357164,
"grad_norm": 1.2044843435287476,
"learning_rate": 6.530034729665996e-06,
"loss": 3.6287,
"step": 555000
},
{
"epoch": 23.416094085908192,
"grad_norm": 1.198686122894287,
"learning_rate": 6.444282467950093e-06,
"loss": 3.6292,
"step": 555500
},
{
"epoch": 23.43717067824474,
"grad_norm": 1.160306692123413,
"learning_rate": 6.358530206234189e-06,
"loss": 3.626,
"step": 556000
},
{
"epoch": 23.458247270581293,
"grad_norm": 1.214788556098938,
"learning_rate": 6.272777944518287e-06,
"loss": 3.6292,
"step": 556500
},
{
"epoch": 23.479323862917845,
"grad_norm": 1.2000130414962769,
"learning_rate": 6.187025682802384e-06,
"loss": 3.6294,
"step": 557000
},
{
"epoch": 23.500400455254393,
"grad_norm": 1.224147915840149,
"learning_rate": 6.101273421086482e-06,
"loss": 3.6273,
"step": 557500
},
{
"epoch": 23.521477047590945,
"grad_norm": 1.2343361377716064,
"learning_rate": 6.015521159370579e-06,
"loss": 3.6273,
"step": 558000
},
{
"epoch": 23.542553639927497,
"grad_norm": 1.238450050354004,
"learning_rate": 5.929768897654676e-06,
"loss": 3.627,
"step": 558500
},
{
"epoch": 23.563630232264046,
"grad_norm": 1.2036480903625488,
"learning_rate": 5.844016635938773e-06,
"loss": 3.6263,
"step": 559000
},
{
"epoch": 23.584706824600598,
"grad_norm": 1.1617209911346436,
"learning_rate": 5.75826437422287e-06,
"loss": 3.627,
"step": 559500
},
{
"epoch": 23.60578341693715,
"grad_norm": 1.256417989730835,
"learning_rate": 5.672512112506967e-06,
"loss": 3.6287,
"step": 560000
},
{
"epoch": 23.60578341693715,
"eval_accuracy": 0.7588664206634853,
"eval_loss": 3.4145781993865967,
"eval_runtime": 383.3592,
"eval_samples_per_second": 314.864,
"eval_steps_per_second": 3.282,
"step": 560000
},
{
"epoch": 23.6268600092737,
"grad_norm": 1.2074886560440063,
"learning_rate": 5.5867598507910655e-06,
"loss": 3.6269,
"step": 560500
},
{
"epoch": 23.64793660161025,
"grad_norm": 1.2370638847351074,
"learning_rate": 5.501007589075162e-06,
"loss": 3.6262,
"step": 561000
},
{
"epoch": 23.669013193946803,
"grad_norm": 1.277632474899292,
"learning_rate": 5.41525532735926e-06,
"loss": 3.6279,
"step": 561500
},
{
"epoch": 23.690089786283355,
"grad_norm": 1.28183114528656,
"learning_rate": 5.329503065643356e-06,
"loss": 3.6234,
"step": 562000
},
{
"epoch": 23.711166378619904,
"grad_norm": 1.2591907978057861,
"learning_rate": 5.243750803927454e-06,
"loss": 3.6243,
"step": 562500
},
{
"epoch": 23.732242970956456,
"grad_norm": 1.2464503049850464,
"learning_rate": 5.157998542211551e-06,
"loss": 3.6274,
"step": 563000
},
{
"epoch": 23.753319563293008,
"grad_norm": 1.219276785850525,
"learning_rate": 5.072246280495649e-06,
"loss": 3.6229,
"step": 563500
},
{
"epoch": 23.774396155629557,
"grad_norm": 1.2028733491897583,
"learning_rate": 4.986494018779746e-06,
"loss": 3.6256,
"step": 564000
},
{
"epoch": 23.79547274796611,
"grad_norm": 1.2222185134887695,
"learning_rate": 4.900741757063843e-06,
"loss": 3.6241,
"step": 564500
},
{
"epoch": 23.81654934030266,
"grad_norm": 1.221431851387024,
"learning_rate": 4.81498949534794e-06,
"loss": 3.6241,
"step": 565000
},
{
"epoch": 23.83762593263921,
"grad_norm": 1.1969276666641235,
"learning_rate": 4.729237233632038e-06,
"loss": 3.625,
"step": 565500
},
{
"epoch": 23.85870252497576,
"grad_norm": 1.2142211198806763,
"learning_rate": 4.643484971916134e-06,
"loss": 3.6267,
"step": 566000
},
{
"epoch": 23.879779117312314,
"grad_norm": 1.176747441291809,
"learning_rate": 4.557732710200232e-06,
"loss": 3.6243,
"step": 566500
},
{
"epoch": 23.900855709648862,
"grad_norm": 1.2115191221237183,
"learning_rate": 4.471980448484329e-06,
"loss": 3.6243,
"step": 567000
},
{
"epoch": 23.921932301985414,
"grad_norm": 1.2651346921920776,
"learning_rate": 4.386228186768427e-06,
"loss": 3.6235,
"step": 567500
},
{
"epoch": 23.943008894321967,
"grad_norm": 1.2140218019485474,
"learning_rate": 4.300475925052524e-06,
"loss": 3.6253,
"step": 568000
},
{
"epoch": 23.96408548665852,
"grad_norm": 1.208403468132019,
"learning_rate": 4.214723663336621e-06,
"loss": 3.6292,
"step": 568500
},
{
"epoch": 23.985162078995067,
"grad_norm": 1.2602713108062744,
"learning_rate": 4.128971401620718e-06,
"loss": 3.6216,
"step": 569000
},
{
"epoch": 24.00623867133162,
"grad_norm": 1.2241783142089844,
"learning_rate": 4.043219139904815e-06,
"loss": 3.6253,
"step": 569500
},
{
"epoch": 24.02731526366817,
"grad_norm": 1.1893742084503174,
"learning_rate": 3.957466878188912e-06,
"loss": 3.623,
"step": 570000
},
{
"epoch": 24.02731526366817,
"eval_accuracy": 0.7591419517801029,
"eval_loss": 3.4131720066070557,
"eval_runtime": 427.5584,
"eval_samples_per_second": 282.315,
"eval_steps_per_second": 2.942,
"step": 570000
},
{
"epoch": 24.04839185600472,
"grad_norm": 1.230156421661377,
"learning_rate": 3.87171461647301e-06,
"loss": 3.6279,
"step": 570500
},
{
"epoch": 24.069468448341272,
"grad_norm": 1.1888548135757446,
"learning_rate": 3.7859623547571067e-06,
"loss": 3.6209,
"step": 571000
},
{
"epoch": 24.090545040677824,
"grad_norm": 1.2186180353164673,
"learning_rate": 3.7002100930412046e-06,
"loss": 3.6194,
"step": 571500
},
{
"epoch": 24.111621633014373,
"grad_norm": 1.1973209381103516,
"learning_rate": 3.614457831325301e-06,
"loss": 3.6276,
"step": 572000
},
{
"epoch": 24.132698225350925,
"grad_norm": 1.2377920150756836,
"learning_rate": 3.5287055696093987e-06,
"loss": 3.623,
"step": 572500
},
{
"epoch": 24.153774817687477,
"grad_norm": 1.1899082660675049,
"learning_rate": 3.4429533078934957e-06,
"loss": 3.6241,
"step": 573000
},
{
"epoch": 24.174851410024026,
"grad_norm": 1.2315367460250854,
"learning_rate": 3.357201046177593e-06,
"loss": 3.6268,
"step": 573500
},
{
"epoch": 24.195928002360578,
"grad_norm": 1.2045565843582153,
"learning_rate": 3.2714487844616906e-06,
"loss": 3.6252,
"step": 574000
},
{
"epoch": 24.21700459469713,
"grad_norm": 1.210938811302185,
"learning_rate": 3.1856965227457872e-06,
"loss": 3.6243,
"step": 574500
},
{
"epoch": 24.238081187033682,
"grad_norm": 1.1734343767166138,
"learning_rate": 3.099944261029885e-06,
"loss": 3.6286,
"step": 575000
},
{
"epoch": 24.25915777937023,
"grad_norm": 1.254120945930481,
"learning_rate": 3.014191999313982e-06,
"loss": 3.6248,
"step": 575500
},
{
"epoch": 24.280234371706783,
"grad_norm": 1.195219874382019,
"learning_rate": 2.928439737598079e-06,
"loss": 3.6216,
"step": 576000
},
{
"epoch": 24.301310964043335,
"grad_norm": 1.2246639728546143,
"learning_rate": 2.8426874758821767e-06,
"loss": 3.6233,
"step": 576500
},
{
"epoch": 24.322387556379883,
"grad_norm": 1.2106448411941528,
"learning_rate": 2.7569352141662737e-06,
"loss": 3.6183,
"step": 577000
},
{
"epoch": 24.343464148716436,
"grad_norm": 1.2116409540176392,
"learning_rate": 2.6711829524503707e-06,
"loss": 3.6196,
"step": 577500
},
{
"epoch": 24.364540741052988,
"grad_norm": 1.2525883913040161,
"learning_rate": 2.585430690734468e-06,
"loss": 3.6192,
"step": 578000
},
{
"epoch": 24.385617333389536,
"grad_norm": 1.1908764839172363,
"learning_rate": 2.4996784290185657e-06,
"loss": 3.6222,
"step": 578500
},
{
"epoch": 24.40669392572609,
"grad_norm": 1.2558976411819458,
"learning_rate": 2.4139261673026627e-06,
"loss": 3.6187,
"step": 579000
},
{
"epoch": 24.42777051806264,
"grad_norm": 1.2399632930755615,
"learning_rate": 2.3281739055867597e-06,
"loss": 3.6215,
"step": 579500
},
{
"epoch": 24.44884711039919,
"grad_norm": 1.2201695442199707,
"learning_rate": 2.242421643870857e-06,
"loss": 3.6224,
"step": 580000
},
{
"epoch": 24.44884711039919,
"eval_accuracy": 0.7595316861575283,
"eval_loss": 3.409717559814453,
"eval_runtime": 392.6133,
"eval_samples_per_second": 307.442,
"eval_steps_per_second": 3.204,
"step": 580000
},
{
"epoch": 24.46992370273574,
"grad_norm": 1.2455500364303589,
"learning_rate": 2.1566693821549547e-06,
"loss": 3.62,
"step": 580500
},
{
"epoch": 24.491000295072293,
"grad_norm": 1.2218661308288574,
"learning_rate": 2.0709171204390517e-06,
"loss": 3.6232,
"step": 581000
},
{
"epoch": 24.512076887408845,
"grad_norm": 1.2291735410690308,
"learning_rate": 1.985164858723149e-06,
"loss": 3.6181,
"step": 581500
},
{
"epoch": 24.533153479745394,
"grad_norm": 1.252737045288086,
"learning_rate": 1.8994125970072462e-06,
"loss": 3.6251,
"step": 582000
},
{
"epoch": 24.554230072081946,
"grad_norm": 1.191583514213562,
"learning_rate": 1.8136603352913432e-06,
"loss": 3.6203,
"step": 582500
},
{
"epoch": 24.5753066644185,
"grad_norm": 1.2260080575942993,
"learning_rate": 1.7279080735754405e-06,
"loss": 3.6235,
"step": 583000
},
{
"epoch": 24.596383256755047,
"grad_norm": 1.2457807064056396,
"learning_rate": 1.642155811859538e-06,
"loss": 3.6236,
"step": 583500
},
{
"epoch": 24.6174598490916,
"grad_norm": 1.2328829765319824,
"learning_rate": 1.5564035501436352e-06,
"loss": 3.6215,
"step": 584000
},
{
"epoch": 24.63853644142815,
"grad_norm": 1.207205891609192,
"learning_rate": 1.4706512884277324e-06,
"loss": 3.6226,
"step": 584500
},
{
"epoch": 24.6596130337647,
"grad_norm": 1.190905213356018,
"learning_rate": 1.3848990267118295e-06,
"loss": 3.6186,
"step": 585000
},
{
"epoch": 24.680689626101252,
"grad_norm": 1.215759038925171,
"learning_rate": 1.299146764995927e-06,
"loss": 3.6212,
"step": 585500
},
{
"epoch": 24.701766218437804,
"grad_norm": 1.2632070779800415,
"learning_rate": 1.213394503280024e-06,
"loss": 3.6184,
"step": 586000
},
{
"epoch": 24.722842810774353,
"grad_norm": 1.2307192087173462,
"learning_rate": 1.1276422415641212e-06,
"loss": 3.6197,
"step": 586500
},
{
"epoch": 24.743919403110905,
"grad_norm": 1.1997207403182983,
"learning_rate": 1.0418899798482187e-06,
"loss": 3.623,
"step": 587000
},
{
"epoch": 24.764995995447457,
"grad_norm": 1.2403733730316162,
"learning_rate": 9.561377181323157e-07,
"loss": 3.6182,
"step": 587500
},
{
"epoch": 24.786072587784005,
"grad_norm": 1.2494302988052368,
"learning_rate": 8.70385456416413e-07,
"loss": 3.62,
"step": 588000
},
{
"epoch": 24.807149180120557,
"grad_norm": 1.1718021631240845,
"learning_rate": 7.846331947005103e-07,
"loss": 3.6207,
"step": 588500
},
{
"epoch": 24.82822577245711,
"grad_norm": 1.2604660987854004,
"learning_rate": 6.988809329846075e-07,
"loss": 3.6161,
"step": 589000
},
{
"epoch": 24.84930236479366,
"grad_norm": 1.236738681793213,
"learning_rate": 6.131286712687047e-07,
"loss": 3.6218,
"step": 589500
},
{
"epoch": 24.87037895713021,
"grad_norm": 1.2456611394882202,
"learning_rate": 5.27376409552802e-07,
"loss": 3.6193,
"step": 590000
},
{
"epoch": 24.87037895713021,
"eval_accuracy": 0.7595289396838688,
"eval_loss": 3.409834861755371,
"eval_runtime": 388.9121,
"eval_samples_per_second": 310.368,
"eval_steps_per_second": 3.235,
"step": 590000
},
{
"epoch": 24.891455549466762,
"grad_norm": 1.2020431756973267,
"learning_rate": 4.4162414783689917e-07,
"loss": 3.6197,
"step": 590500
},
{
"epoch": 24.912532141803315,
"grad_norm": 1.2167938947677612,
"learning_rate": 3.558718861209964e-07,
"loss": 3.6189,
"step": 591000
},
{
"epoch": 24.933608734139863,
"grad_norm": 1.2430943250656128,
"learning_rate": 2.701196244050937e-07,
"loss": 3.6202,
"step": 591500
},
{
"epoch": 24.954685326476415,
"grad_norm": 1.2530001401901245,
"learning_rate": 1.8436736268919094e-07,
"loss": 3.6212,
"step": 592000
},
{
"epoch": 24.975761918812967,
"grad_norm": 1.2046359777450562,
"learning_rate": 9.861510097328816e-08,
"loss": 3.6179,
"step": 592500
},
{
"epoch": 24.996838511149516,
"grad_norm": 1.1981650590896606,
"learning_rate": 1.2862839257385414e-08,
"loss": 3.6182,
"step": 593000
},
{
"epoch": 25.0,
"step": 593075,
"total_flos": 1.5091168791146803e+19,
"train_loss": 4.081508317309086,
"train_runtime": 296836.3665,
"train_samples_per_second": 191.801,
"train_steps_per_second": 1.998
}
],
"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": 1.5091168791146803e+19,
"train_batch_size": 96,
"trial_name": null,
"trial_params": null
}