{
  "best_metric": 0.51153946,
  "best_model_checkpoint": "/export/home2/zli/kc/mm_rag/Qwen2.5-72B-Instruct_lora/checkpoint-513",
  "epoch": 0.9995129079396006,
  "eval_steps": 100,
  "global_step": 513,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.001948368241597662,
      "grad_norm": 0.7911198735237122,
      "learning_rate": 3.846153846153847e-06,
      "loss": 1.1104155778884888,
      "memory(GiB)": 239.17,
      "step": 1,
      "token_acc": 0.7912871287128713,
      "train_speed(iter/s)": 0.00699
    },
    {
      "epoch": 0.00974184120798831,
      "grad_norm": 1.048244833946228,
      "learning_rate": 1.923076923076923e-05,
      "loss": 1.0879460573196411,
      "memory(GiB)": 257.56,
      "step": 5,
      "token_acc": 0.7876893738743511,
      "train_speed(iter/s)": 0.006513
    },
    {
      "epoch": 0.01948368241597662,
      "grad_norm": 0.8897583484649658,
      "learning_rate": 3.846153846153846e-05,
      "loss": 1.179165744781494,
      "memory(GiB)": 267.89,
      "step": 10,
      "token_acc": 0.7578167359479486,
      "train_speed(iter/s)": 0.006891
    },
    {
      "epoch": 0.029225523623964928,
      "grad_norm": 0.3992486298084259,
      "learning_rate": 5.769230769230769e-05,
      "loss": 0.866185188293457,
      "memory(GiB)": 267.89,
      "step": 15,
      "token_acc": 0.7692150449713818,
      "train_speed(iter/s)": 0.006796
    },
    {
      "epoch": 0.03896736483195324,
      "grad_norm": 0.50007164478302,
      "learning_rate": 7.692307692307693e-05,
      "loss": 0.7950119972229004,
      "memory(GiB)": 267.89,
      "step": 20,
      "token_acc": 0.7725610358802016,
      "train_speed(iter/s)": 0.006865
    },
    {
      "epoch": 0.04870920603994155,
      "grad_norm": 0.2926434874534607,
      "learning_rate": 9.615384615384617e-05,
      "loss": 0.6839358806610107,
      "memory(GiB)": 267.89,
      "step": 25,
      "token_acc": 0.7947733918128655,
      "train_speed(iter/s)": 0.006974
    },
    {
      "epoch": 0.058451047247929856,
      "grad_norm": 0.4686521887779236,
      "learning_rate": 9.998335523311734e-05,
      "loss": 0.6574328422546387,
      "memory(GiB)": 267.89,
      "step": 30,
      "token_acc": 0.8028529939352524,
      "train_speed(iter/s)": 0.006948
    },
    {
      "epoch": 0.06819288845591817,
      "grad_norm": 0.2973114252090454,
      "learning_rate": 9.991575486016592e-05,
      "loss": 0.6306457042694091,
      "memory(GiB)": 267.89,
      "step": 35,
      "token_acc": 0.8055379603655778,
      "train_speed(iter/s)": 0.006928
    },
    {
      "epoch": 0.07793472966390648,
      "grad_norm": 0.2803875803947449,
      "learning_rate": 9.979622885143301e-05,
      "loss": 0.6483866691589355,
      "memory(GiB)": 267.89,
      "step": 40,
      "token_acc": 0.8065050756163248,
      "train_speed(iter/s)": 0.006967
    },
    {
      "epoch": 0.08767657087189479,
      "grad_norm": 0.486200213432312,
      "learning_rate": 9.962490154569727e-05,
      "loss": 0.6592767715454102,
      "memory(GiB)": 279.45,
      "step": 45,
      "token_acc": 0.7944840703756538,
      "train_speed(iter/s)": 0.006889
    },
    {
      "epoch": 0.0974184120798831,
      "grad_norm": 0.3190113604068756,
      "learning_rate": 9.940195116883755e-05,
      "loss": 0.6488125324249268,
      "memory(GiB)": 279.45,
      "step": 50,
      "token_acc": 0.7989409984871407,
      "train_speed(iter/s)": 0.006919
    },
    {
      "epoch": 0.1071602532878714,
      "grad_norm": 0.2820813059806824,
      "learning_rate": 9.91276096484306e-05,
      "loss": 0.6796814918518066,
      "memory(GiB)": 279.45,
      "step": 55,
      "token_acc": 0.7929543782192788,
      "train_speed(iter/s)": 0.006811
    },
    {
      "epoch": 0.11690209449585971,
      "grad_norm": 0.37464168667793274,
      "learning_rate": 9.880216237248481e-05,
      "loss": 0.7066696166992188,
      "memory(GiB)": 279.45,
      "step": 60,
      "token_acc": 0.7865912142720123,
      "train_speed(iter/s)": 0.006893
    },
    {
      "epoch": 0.12664393570384802,
      "grad_norm": 0.3829931318759918,
      "learning_rate": 9.842594789256103e-05,
      "loss": 0.5756209850311279,
      "memory(GiB)": 279.45,
      "step": 65,
      "token_acc": 0.8192581553237291,
      "train_speed(iter/s)": 0.0069
    },
    {
      "epoch": 0.13638577691183634,
      "grad_norm": 0.3420902490615845,
      "learning_rate": 9.799935757158891e-05,
      "loss": 0.7137881755828858,
      "memory(GiB)": 279.45,
      "step": 70,
      "token_acc": 0.7774079467443312,
      "train_speed(iter/s)": 0.006961
    },
    {
      "epoch": 0.14612761811982464,
      "grad_norm": 0.2829829454421997,
      "learning_rate": 9.752283517674575e-05,
      "loss": 0.5826292037963867,
      "memory(GiB)": 279.45,
      "step": 75,
      "token_acc": 0.8175895765472313,
      "train_speed(iter/s)": 0.006986
    },
    {
      "epoch": 0.15586945932781296,
      "grad_norm": 0.4237753748893738,
      "learning_rate": 9.699687641782067e-05,
      "loss": 0.5978005886077881,
      "memory(GiB)": 279.45,
      "step": 80,
      "token_acc": 0.8125563774129533,
      "train_speed(iter/s)": 0.007002
    },
    {
      "epoch": 0.16561130053580125,
      "grad_norm": 0.6381401419639587,
      "learning_rate": 9.642202843154491e-05,
      "loss": 0.6211316585540771,
      "memory(GiB)": 279.45,
      "step": 85,
      "token_acc": 0.8153757510906248,
      "train_speed(iter/s)": 0.007036
    },
    {
      "epoch": 0.17535314174378958,
      "grad_norm": 0.3631569743156433,
      "learning_rate": 9.579888921242439e-05,
      "loss": 0.6291302680969239,
      "memory(GiB)": 279.45,
      "step": 90,
      "token_acc": 0.8056357749190514,
      "train_speed(iter/s)": 0.00706
    },
    {
      "epoch": 0.1850949829517779,
      "grad_norm": 0.31778669357299805,
      "learning_rate": 9.512810699066667e-05,
      "loss": 0.6138756752014161,
      "memory(GiB)": 279.45,
      "step": 95,
      "token_acc": 0.8115504301665751,
      "train_speed(iter/s)": 0.007048
    },
    {
      "epoch": 0.1948368241597662,
      "grad_norm": 0.3409082889556885,
      "learning_rate": 9.441037955784944e-05,
      "loss": 0.640412425994873,
      "memory(GiB)": 279.45,
      "step": 100,
      "token_acc": 0.7977538518656023,
      "train_speed(iter/s)": 0.007066
    },
    {
      "epoch": 0.1948368241597662,
      "eval_loss": 0.6026075482368469,
      "eval_runtime": 26.8674,
      "eval_samples_per_second": 0.149,
      "eval_steps_per_second": 0.074,
      "step": 100
    },
    {
      "epoch": 0.20457866536775451,
      "grad_norm": 0.5213446021080017,
      "learning_rate": 9.364645354103206e-05,
      "loss": 0.6617852210998535,
      "memory(GiB)": 279.45,
      "step": 105,
      "token_acc": 0.7954326459818848,
      "train_speed(iter/s)": 0.007023
    },
    {
      "epoch": 0.2143205065757428,
      "grad_norm": 0.3881484270095825,
      "learning_rate": 9.28371236260652e-05,
      "loss": 0.6231249809265137,
      "memory(GiB)": 279.45,
      "step": 110,
      "token_acc": 0.8025908424311553,
      "train_speed(iter/s)": 0.007042
    },
    {
      "epoch": 0.22406234778373113,
      "grad_norm": 0.34954214096069336,
      "learning_rate": 9.198323173090663e-05,
      "loss": 0.6497172355651856,
      "memory(GiB)": 279.45,
      "step": 115,
      "token_acc": 0.794488631798827,
      "train_speed(iter/s)": 0.007052
    },
    {
      "epoch": 0.23380418899171942,
      "grad_norm": 0.3073936402797699,
      "learning_rate": 9.108566612980298e-05,
      "loss": 0.6538199424743653,
      "memory(GiB)": 279.45,
      "step": 120,
      "token_acc": 0.7920408356177332,
      "train_speed(iter/s)": 0.00707
    },
    {
      "epoch": 0.24354603019970775,
      "grad_norm": 0.3140193819999695,
      "learning_rate": 9.014536052924883e-05,
      "loss": 0.5965206623077393,
      "memory(GiB)": 279.45,
      "step": 125,
      "token_acc": 0.8178110129163834,
      "train_speed(iter/s)": 0.007102
    },
    {
      "epoch": 0.25328787140769604,
      "grad_norm": 0.9085149168968201,
      "learning_rate": 8.916329309668397e-05,
      "loss": 0.5718743324279785,
      "memory(GiB)": 279.45,
      "step": 130,
      "token_acc": 0.8312764304824373,
      "train_speed(iter/s)": 0.007101
    },
    {
      "epoch": 0.26302971261568436,
      "grad_norm": 0.4713858664035797,
      "learning_rate": 8.814048544293965e-05,
      "loss": 0.582437801361084,
      "memory(GiB)": 279.45,
      "step": 135,
      "token_acc": 0.8134561744569352,
      "train_speed(iter/s)": 0.007088
    },
    {
      "epoch": 0.2727715538236727,
      "grad_norm": 0.39269959926605225,
      "learning_rate": 8.707800155949217e-05,
      "loss": 0.5708662986755371,
      "memory(GiB)": 279.45,
      "step": 140,
      "token_acc": 0.8203411790110128,
      "train_speed(iter/s)": 0.007075
    },
    {
      "epoch": 0.282513395031661,
      "grad_norm": 0.38881248235702515,
      "learning_rate": 8.597694671162921e-05,
      "loss": 0.6288607597351075,
      "memory(GiB)": 279.45,
      "step": 145,
      "token_acc": 0.8012965630788228,
      "train_speed(iter/s)": 0.007075
    },
    {
      "epoch": 0.2922552362396493,
      "grad_norm": 1.00093412399292,
      "learning_rate": 8.483846628868055e-05,
      "loss": 0.6310084819793701,
      "memory(GiB)": 279.45,
      "step": 150,
      "token_acc": 0.8051169868746433,
      "train_speed(iter/s)": 0.007102
    },
    {
      "epoch": 0.3019970774476376,
      "grad_norm": 0.5331721901893616,
      "learning_rate": 8.366374461250916e-05,
      "loss": 0.5937428951263428,
      "memory(GiB)": 279.45,
      "step": 155,
      "token_acc": 0.8136619109381755,
      "train_speed(iter/s)": 0.007062
    },
    {
      "epoch": 0.3117389186556259,
      "grad_norm": 0.6358391642570496,
      "learning_rate": 8.245400370550198e-05,
      "loss": 0.6008272171020508,
      "memory(GiB)": 279.45,
      "step": 160,
      "token_acc": 0.8157670857358919,
      "train_speed(iter/s)": 0.007051
    },
    {
      "epoch": 0.32148075986361424,
      "grad_norm": 0.40399816632270813,
      "learning_rate": 8.121050201934235e-05,
      "loss": 0.6029816627502441,
      "memory(GiB)": 279.45,
      "step": 165,
      "token_acc": 0.8082507259437268,
      "train_speed(iter/s)": 0.007056
    },
    {
      "epoch": 0.3312226010716025,
      "grad_norm": 0.4126909673213959,
      "learning_rate": 7.993453312588607e-05,
      "loss": 0.6517924785614013,
      "memory(GiB)": 279.45,
      "step": 170,
      "token_acc": 0.7958906764876914,
      "train_speed(iter/s)": 0.007056
    },
    {
      "epoch": 0.34096444227959083,
      "grad_norm": 0.6618629097938538,
      "learning_rate": 7.862742437150336e-05,
      "loss": 0.6141845226287842,
      "memory(GiB)": 279.45,
      "step": 175,
      "token_acc": 0.8037235096058626,
      "train_speed(iter/s)": 0.007045
    },
    {
      "epoch": 0.35070628348757915,
      "grad_norm": 0.5881444215774536,
      "learning_rate": 7.729053549628622e-05,
      "loss": 0.6001346588134766,
      "memory(GiB)": 279.45,
      "step": 180,
      "token_acc": 0.8113048368953881,
      "train_speed(iter/s)": 0.007061
    },
    {
      "epoch": 0.3604481246955675,
      "grad_norm": 1.692840576171875,
      "learning_rate": 7.592525721955786e-05,
      "loss": 0.574149227142334,
      "memory(GiB)": 279.45,
      "step": 185,
      "token_acc": 0.8219587430113746,
      "train_speed(iter/s)": 0.007067
    },
    {
      "epoch": 0.3701899659035558,
      "grad_norm": 0.38788875937461853,
      "learning_rate": 7.45330097931553e-05,
      "loss": 0.6197400093078613,
      "memory(GiB)": 279.45,
      "step": 190,
      "token_acc": 0.8041193060422186,
      "train_speed(iter/s)": 0.007079
    },
    {
      "epoch": 0.37993180711154406,
      "grad_norm": 0.48055437207221985,
      "learning_rate": 7.311524152399054e-05,
      "loss": 0.6016897678375244,
      "memory(GiB)": 279.45,
      "step": 195,
      "token_acc": 0.8031721765717562,
      "train_speed(iter/s)": 0.007068
    },
    {
      "epoch": 0.3896736483195324,
      "grad_norm": 0.4017133116722107,
      "learning_rate": 7.167342726742685e-05,
      "loss": 0.5860675811767578,
      "memory(GiB)": 279.45,
      "step": 200,
      "token_acc": 0.8191920830629461,
      "train_speed(iter/s)": 0.007059
    },
    {
      "epoch": 0.3896736483195324,
      "eval_loss": 0.5398527979850769,
      "eval_runtime": 26.925,
      "eval_samples_per_second": 0.149,
      "eval_steps_per_second": 0.074,
      "step": 200
    },
    {
      "epoch": 0.3994154895275207,
      "grad_norm": 0.48819786310195923,
      "learning_rate": 7.020906689303766e-05,
      "loss": 0.60306715965271,
      "memory(GiB)": 279.45,
      "step": 205,
      "token_acc": 0.8102000173175167,
      "train_speed(iter/s)": 0.007055
    },
    {
      "epoch": 0.40915733073550903,
      "grad_norm": 0.5116552114486694,
      "learning_rate": 6.872368372434416e-05,
      "loss": 0.6070918083190918,
      "memory(GiB)": 279.45,
      "step": 210,
      "token_acc": 0.804848606780002,
      "train_speed(iter/s)": 0.007061
    },
    {
      "epoch": 0.4188991719434973,
      "grad_norm": 0.42464399337768555,
      "learning_rate": 6.721882295415425e-05,
      "loss": 0.633366060256958,
      "memory(GiB)": 279.45,
      "step": 215,
      "token_acc": 0.8031126254719587,
      "train_speed(iter/s)": 0.007047
    },
    {
      "epoch": 0.4286410131514856,
      "grad_norm": 0.43728524446487427,
      "learning_rate": 6.569605003715201e-05,
      "loss": 0.5457853317260742,
      "memory(GiB)": 279.45,
      "step": 220,
      "token_acc": 0.8268458881732474,
      "train_speed(iter/s)": 0.007041
    },
    {
      "epoch": 0.43838285435947394,
      "grad_norm": 0.526282548904419,
      "learning_rate": 6.41569490614092e-05,
      "loss": 0.5758168220520019,
      "memory(GiB)": 291.53,
      "step": 225,
      "token_acc": 0.8169302630500697,
      "train_speed(iter/s)": 0.007022
    },
    {
      "epoch": 0.44812469556746226,
      "grad_norm": 0.4335802495479584,
      "learning_rate": 6.260312110051312e-05,
      "loss": 0.5597887992858886,
      "memory(GiB)": 291.53,
      "step": 230,
      "token_acc": 0.8260340632603407,
      "train_speed(iter/s)": 0.007017
    },
    {
      "epoch": 0.4578665367754506,
      "grad_norm": 0.46290066838264465,
      "learning_rate": 6.103618254802511e-05,
      "loss": 0.5268624782562256,
      "memory(GiB)": 291.53,
      "step": 235,
      "token_acc": 0.8337082864641919,
      "train_speed(iter/s)": 0.007024
    },
    {
      "epoch": 0.46760837798343885,
      "grad_norm": 0.44755905866622925,
      "learning_rate": 5.945776343600207e-05,
      "loss": 0.6087980270385742,
      "memory(GiB)": 291.53,
      "step": 240,
      "token_acc": 0.8013279748383715,
      "train_speed(iter/s)": 0.007003
    },
    {
      "epoch": 0.47735021919142717,
      "grad_norm": 0.35961049795150757,
      "learning_rate": 5.7869505739330546e-05,
      "loss": 0.6320755958557129,
      "memory(GiB)": 291.53,
      "step": 245,
      "token_acc": 0.7990379032926931,
      "train_speed(iter/s)": 0.006988
    },
    {
      "epoch": 0.4870920603994155,
      "grad_norm": 0.43942877650260925,
      "learning_rate": 5.627306166763684e-05,
      "loss": 0.5911942958831787,
      "memory(GiB)": 291.53,
      "step": 250,
      "token_acc": 0.8108974358974359,
      "train_speed(iter/s)": 0.006984
    },
    {
      "epoch": 0.4968339016074038,
      "grad_norm": 0.3216777443885803,
      "learning_rate": 5.467009194655045e-05,
      "loss": 0.5380321025848389,
      "memory(GiB)": 291.53,
      "step": 255,
      "token_acc": 0.8256673332753673,
      "train_speed(iter/s)": 0.006954
    },
    {
      "epoch": 0.5065757428153921,
      "grad_norm": 0.3348521590232849,
      "learning_rate": 5.306226409010855e-05,
      "loss": 0.5896577358245849,
      "memory(GiB)": 291.53,
      "step": 260,
      "token_acc": 0.8136957057229658,
      "train_speed(iter/s)": 0.00696
    },
    {
      "epoch": 0.5163175840233805,
      "grad_norm": 0.4316939115524292,
      "learning_rate": 5.145125066609877e-05,
      "loss": 0.5456296920776367,
      "memory(GiB)": 291.53,
      "step": 265,
      "token_acc": 0.8252879627187197,
      "train_speed(iter/s)": 0.006968
    },
    {
      "epoch": 0.5260594252313687,
      "grad_norm": 0.4729405343532562,
      "learning_rate": 4.983872755614461e-05,
      "loss": 0.5579260349273681,
      "memory(GiB)": 291.53,
      "step": 270,
      "token_acc": 0.8236703843672228,
      "train_speed(iter/s)": 0.006963
    },
    {
      "epoch": 0.535801266439357,
      "grad_norm": 0.4944189786911011,
      "learning_rate": 4.8226372212343726e-05,
      "loss": 0.5667993545532226,
      "memory(GiB)": 291.53,
      "step": 275,
      "token_acc": 0.8146758992113868,
      "train_speed(iter/s)": 0.006964
    },
    {
      "epoch": 0.5455431076473454,
      "grad_norm": 0.8255349397659302,
      "learning_rate": 4.661586191227247e-05,
      "loss": 0.5635212898254395,
      "memory(GiB)": 291.53,
      "step": 280,
      "token_acc": 0.8212789502806103,
      "train_speed(iter/s)": 0.006981
    },
    {
      "epoch": 0.5552849488553336,
      "grad_norm": 0.4205448627471924,
      "learning_rate": 4.500887201417187e-05,
      "loss": 0.6006342887878418,
      "memory(GiB)": 291.53,
      "step": 285,
      "token_acc": 0.8099815913274698,
      "train_speed(iter/s)": 0.007001
    },
    {
      "epoch": 0.565026790063322,
      "grad_norm": 0.39303454756736755,
      "learning_rate": 4.3407074214130446e-05,
      "loss": 0.5548898220062256,
      "memory(GiB)": 291.53,
      "step": 290,
      "token_acc": 0.8264817150063052,
      "train_speed(iter/s)": 0.007007
    },
    {
      "epoch": 0.5747686312713103,
      "grad_norm": 0.5799729824066162,
      "learning_rate": 4.181213480707637e-05,
      "loss": 0.5500263214111328,
      "memory(GiB)": 291.53,
      "step": 295,
      "token_acc": 0.8245773863751392,
      "train_speed(iter/s)": 0.007016
    },
    {
      "epoch": 0.5845104724792985,
      "grad_norm": 0.5087621212005615,
      "learning_rate": 4.0225712953388494e-05,
      "loss": 0.5443117618560791,
      "memory(GiB)": 291.53,
      "step": 300,
      "token_acc": 0.8282135401133314,
      "train_speed(iter/s)": 0.007015
    },
    {
      "epoch": 0.5845104724792985,
      "eval_loss": 0.5304377675056458,
      "eval_runtime": 26.9276,
      "eval_samples_per_second": 0.149,
      "eval_steps_per_second": 0.074,
      "step": 300
    },
    {
      "epoch": 0.5942523136872869,
      "grad_norm": 0.5076440572738647,
      "learning_rate": 3.864945895292908e-05,
      "loss": 0.6255448818206787,
      "memory(GiB)": 291.53,
      "step": 305,
      "token_acc": 0.8085856491082538,
      "train_speed(iter/s)": 0.007003
    },
    {
      "epoch": 0.6039941548952752,
      "grad_norm": 0.5984292030334473,
      "learning_rate": 3.708501252829386e-05,
      "loss": 0.5672544956207275,
      "memory(GiB)": 291.53,
      "step": 310,
      "token_acc": 0.8221675325859084,
      "train_speed(iter/s)": 0.007004
    },
    {
      "epoch": 0.6137359961032636,
      "grad_norm": 0.4321894645690918,
      "learning_rate": 3.553400111906523e-05,
      "loss": 0.5947960376739502,
      "memory(GiB)": 291.53,
      "step": 315,
      "token_acc": 0.8143758155719878,
      "train_speed(iter/s)": 0.007012
    },
    {
      "epoch": 0.6234778373112518,
      "grad_norm": 0.7790753245353699,
      "learning_rate": 3.399803818884311e-05,
      "loss": 0.5950922489166259,
      "memory(GiB)": 291.53,
      "step": 320,
      "token_acc": 0.8138864880437914,
      "train_speed(iter/s)": 0.007019
    },
    {
      "epoch": 0.6332196785192401,
      "grad_norm": 0.37229499220848083,
      "learning_rate": 3.247872154681439e-05,
      "loss": 0.5663117408752442,
      "memory(GiB)": 291.53,
      "step": 325,
      "token_acc": 0.8152720218675602,
      "train_speed(iter/s)": 0.007009
    },
    {
      "epoch": 0.6429615197272285,
      "grad_norm": 0.5171226859092712,
      "learning_rate": 3.097763168560741e-05,
      "loss": 0.5201399803161622,
      "memory(GiB)": 291.53,
      "step": 330,
      "token_acc": 0.8316269396896496,
      "train_speed(iter/s)": 0.007019
    },
    {
      "epoch": 0.6527033609352167,
      "grad_norm": 0.579806923866272,
      "learning_rate": 2.949633013715982e-05,
      "loss": 0.576199722290039,
      "memory(GiB)": 291.53,
      "step": 335,
      "token_acc": 0.8226395939086294,
      "train_speed(iter/s)": 0.007018
    },
    {
      "epoch": 0.662445202143205,
      "grad_norm": 0.5151726007461548,
      "learning_rate": 2.8036357848311012e-05,
      "loss": 0.5100949287414551,
      "memory(GiB)": 291.53,
      "step": 340,
      "token_acc": 0.835439610037913,
      "train_speed(iter/s)": 0.007019
    },
    {
      "epoch": 0.6721870433511934,
      "grad_norm": 0.6563583612442017,
      "learning_rate": 2.659923357780828e-05,
      "loss": 0.5569330215454101,
      "memory(GiB)": 291.53,
      "step": 345,
      "token_acc": 0.8189111231296197,
      "train_speed(iter/s)": 0.007026
    },
    {
      "epoch": 0.6819288845591817,
      "grad_norm": 0.4978316128253937,
      "learning_rate": 2.518645231639457e-05,
      "loss": 0.5415875434875488,
      "memory(GiB)": 291.53,
      "step": 350,
      "token_acc": 0.8224528817263043,
      "train_speed(iter/s)": 0.007018
    },
    {
      "epoch": 0.69167072576717,
      "grad_norm": 0.3655760586261749,
      "learning_rate": 2.3799483731621237e-05,
      "loss": 0.5323990821838379,
      "memory(GiB)": 291.53,
      "step": 355,
      "token_acc": 0.8286348880915769,
      "train_speed(iter/s)": 0.007019
    },
    {
      "epoch": 0.7014125669751583,
      "grad_norm": 0.5881947875022888,
      "learning_rate": 2.2439770639003627e-05,
      "loss": 0.5367710590362549,
      "memory(GiB)": 291.53,
      "step": 360,
      "token_acc": 0.8259178446067735,
      "train_speed(iter/s)": 0.007019
    },
    {
      "epoch": 0.7111544081831466,
      "grad_norm": 0.4462352991104126,
      "learning_rate": 2.110872750110996e-05,
      "loss": 0.6012316703796386,
      "memory(GiB)": 291.53,
      "step": 365,
      "token_acc": 0.8113236655353286,
      "train_speed(iter/s)": 0.007025
    },
    {
      "epoch": 0.720896249391135,
      "grad_norm": 0.5265180468559265,
      "learning_rate": 1.980773895614481e-05,
      "loss": 0.5925416946411133,
      "memory(GiB)": 291.53,
      "step": 370,
      "token_acc": 0.8087358252834943,
      "train_speed(iter/s)": 0.007037
    },
    {
      "epoch": 0.7306380905991232,
      "grad_norm": 0.5589653253555298,
      "learning_rate": 1.8538158377557702e-05,
      "loss": 0.554245138168335,
      "memory(GiB)": 291.53,
      "step": 375,
      "token_acc": 0.8206760127289928,
      "train_speed(iter/s)": 0.007042
    },
    {
      "epoch": 0.7403799318071116,
      "grad_norm": 0.49020129442214966,
      "learning_rate": 1.7301306466175533e-05,
      "loss": 0.5485538482666016,
      "memory(GiB)": 291.53,
      "step": 380,
      "token_acc": 0.8174893357708715,
      "train_speed(iter/s)": 0.007043
    },
    {
      "epoch": 0.7501217730150999,
      "grad_norm": 0.8308594226837158,
      "learning_rate": 1.6098469876323093e-05,
      "loss": 0.5580153465270996,
      "memory(GiB)": 291.53,
      "step": 385,
      "token_acc": 0.8252497729336966,
      "train_speed(iter/s)": 0.007044
    },
    {
      "epoch": 0.7598636142230881,
      "grad_norm": 0.9425000548362732,
      "learning_rate": 1.4930899877361015e-05,
      "loss": 0.5678109169006348,
      "memory(GiB)": 291.53,
      "step": 390,
      "token_acc": 0.8189367271414387,
      "train_speed(iter/s)": 0.007047
    },
    {
      "epoch": 0.7696054554310765,
      "grad_norm": 0.46440383791923523,
      "learning_rate": 1.3799811052033467e-05,
      "loss": 0.5055109024047851,
      "memory(GiB)": 291.53,
      "step": 395,
      "token_acc": 0.835208145167912,
      "train_speed(iter/s)": 0.007044
    },
    {
      "epoch": 0.7793472966390648,
      "grad_norm": 0.6887357831001282,
      "learning_rate": 1.2706380032979691e-05,
      "loss": 0.6191585540771485,
      "memory(GiB)": 291.53,
      "step": 400,
      "token_acc": 0.801925584194163,
      "train_speed(iter/s)": 0.007049
    },
    {
      "epoch": 0.7793472966390648,
      "eval_loss": 0.5201972723007202,
      "eval_runtime": 26.7707,
      "eval_samples_per_second": 0.149,
      "eval_steps_per_second": 0.075,
      "step": 400
    },
    {
      "epoch": 0.7890891378470531,
      "grad_norm": 0.6512044072151184,
      "learning_rate": 1.1651744278723687e-05,
      "loss": 0.5115838527679444,
      "memory(GiB)": 291.53,
      "step": 405,
      "token_acc": 0.8362541073384447,
      "train_speed(iter/s)": 0.007043
    },
    {
      "epoch": 0.7988309790550414,
      "grad_norm": 0.36380794644355774,
      "learning_rate": 1.0637000890415388e-05,
      "loss": 0.5678341865539551,
      "memory(GiB)": 291.53,
      "step": 410,
      "token_acc": 0.8218965517241379,
      "train_speed(iter/s)": 0.007056
    },
    {
      "epoch": 0.8085728202630297,
      "grad_norm": 0.524336576461792,
      "learning_rate": 9.663205470554276e-06,
      "loss": 0.5683213233947754,
      "memory(GiB)": 291.54,
      "step": 415,
      "token_acc": 0.8144156614593633,
      "train_speed(iter/s)": 0.007057
    },
    {
      "epoch": 0.8183146614710181,
      "grad_norm": 0.510553777217865,
      "learning_rate": 8.73137102488249e-06,
      "loss": 0.5425020217895508,
      "memory(GiB)": 291.54,
      "step": 420,
      "token_acc": 0.8262775083923909,
      "train_speed(iter/s)": 0.007067
    },
    {
      "epoch": 0.8280565026790063,
      "grad_norm": 0.4342099726200104,
      "learning_rate": 7.842466908590006e-06,
      "loss": 0.4687312602996826,
      "memory(GiB)": 291.54,
      "step": 425,
      "token_acc": 0.8489349775784754,
      "train_speed(iter/s)": 0.007065
    },
    {
      "epoch": 0.8377983438869946,
      "grad_norm": 0.47141674160957336,
      "learning_rate": 6.997417817927865e-06,
      "loss": 0.5220559597015381,
      "memory(GiB)": 291.54,
      "step": 430,
      "token_acc": 0.8345159448658348,
      "train_speed(iter/s)": 0.007065
    },
    {
      "epoch": 0.847540185094983,
      "grad_norm": 0.7017420530319214,
      "learning_rate": 6.197102828278611e-06,
      "loss": 0.4900541305541992,
      "memory(GiB)": 291.54,
      "step": 435,
      "token_acc": 0.846220772158782,
      "train_speed(iter/s)": 0.007061
    },
    {
      "epoch": 0.8572820263029712,
      "grad_norm": 0.9358065724372864,
      "learning_rate": 5.442354479684558e-06,
      "loss": 0.5937717914581299,
      "memory(GiB)": 291.54,
      "step": 440,
      "token_acc": 0.8115085230071928,
      "train_speed(iter/s)": 0.007068
    },
    {
      "epoch": 0.8670238675109596,
      "grad_norm": 0.6291302442550659,
      "learning_rate": 4.733957910785114e-06,
      "loss": 0.5440679073333741,
      "memory(GiB)": 291.54,
      "step": 445,
      "token_acc": 0.8238650458943191,
      "train_speed(iter/s)": 0.007071
    },
    {
      "epoch": 0.8767657087189479,
      "grad_norm": 0.7625430822372437,
      "learning_rate": 4.072650042064174e-06,
      "loss": 0.5928035736083984,
      "memory(GiB)": 291.54,
      "step": 450,
      "token_acc": 0.8072507552870091,
      "train_speed(iter/s)": 0.007064
    },
    {
      "epoch": 0.8865075499269361,
      "grad_norm": 0.5580248832702637,
      "learning_rate": 3.4591188092571893e-06,
      "loss": 0.5281401634216308,
      "memory(GiB)": 291.54,
      "step": 455,
      "token_acc": 0.8284527518172378,
      "train_speed(iter/s)": 0.007059
    },
    {
      "epoch": 0.8962493911349245,
      "grad_norm": 0.6311898827552795,
      "learning_rate": 2.894002447715399e-06,
      "loss": 0.483402156829834,
      "memory(GiB)": 291.54,
      "step": 460,
      "token_acc": 0.846503178928247,
      "train_speed(iter/s)": 0.007059
    },
    {
      "epoch": 0.9059912323429128,
      "grad_norm": 0.47069641947746277,
      "learning_rate": 2.3778888284716193e-06,
      "loss": 0.5357548236846924,
      "memory(GiB)": 291.54,
      "step": 465,
      "token_acc": 0.8302510497632449,
      "train_speed(iter/s)": 0.007062
    },
    {
      "epoch": 0.9157330735509012,
      "grad_norm": 0.6504586935043335,
      "learning_rate": 1.9113148466983254e-06,
      "loss": 0.4945201873779297,
      "memory(GiB)": 291.54,
      "step": 470,
      "token_acc": 0.836600057366861,
      "train_speed(iter/s)": 0.007062
    },
    {
      "epoch": 0.9254749147588894,
      "grad_norm": 0.8726986646652222,
      "learning_rate": 1.4947658631941309e-06,
      "loss": 0.6114956855773925,
      "memory(GiB)": 291.54,
      "step": 475,
      "token_acc": 0.8079200232941861,
      "train_speed(iter/s)": 0.007069
    },
    {
      "epoch": 0.9352167559668777,
      "grad_norm": 0.4940550625324249,
      "learning_rate": 1.1286751994797284e-06,
      "loss": 0.4838570117950439,
      "memory(GiB)": 291.54,
      "step": 480,
      "token_acc": 0.8428108691740147,
      "train_speed(iter/s)": 0.007071
    },
    {
      "epoch": 0.9449585971748661,
      "grad_norm": 0.47085583209991455,
      "learning_rate": 8.134236870284861e-07,
      "loss": 0.5545851230621338,
      "memory(GiB)": 291.54,
      "step": 485,
      "token_acc": 0.8195587462791105,
      "train_speed(iter/s)": 0.007076
    },
    {
      "epoch": 0.9547004383828543,
      "grad_norm": 0.47154033184051514,
      "learning_rate": 5.493392711005796e-07,
      "loss": 0.5688926696777343,
      "memory(GiB)": 291.54,
      "step": 490,
      "token_acc": 0.8178977272727272,
      "train_speed(iter/s)": 0.007077
    },
    {
      "epoch": 0.9644422795908427,
      "grad_norm": 0.49407562613487244,
      "learning_rate": 3.366966695929119e-07,
      "loss": 0.5036429405212403,
      "memory(GiB)": 291.54,
      "step": 495,
      "token_acc": 0.8390853451207072,
      "train_speed(iter/s)": 0.007075
    },
    {
      "epoch": 0.974184120798831,
      "grad_norm": 0.5743114352226257,
      "learning_rate": 1.7571708725953596e-07,
      "loss": 0.514036750793457,
      "memory(GiB)": 291.54,
      "step": 500,
      "token_acc": 0.8316027940343591,
      "train_speed(iter/s)": 0.007081
    },
    {
      "epoch": 0.974184120798831,
      "eval_loss": 0.5129119753837585,
      "eval_runtime": 26.8989,
      "eval_samples_per_second": 0.149,
      "eval_steps_per_second": 0.074,
      "step": 500
    },
    {
      "epoch": 0.9839259620068193,
      "grad_norm": 0.5348408818244934,
      "learning_rate": 6.656798560001343e-08,
      "loss": 0.5319667816162109,
      "memory(GiB)": 291.54,
      "step": 505,
      "token_acc": 0.8256676557863502,
      "train_speed(iter/s)": 0.007071
    },
    {
      "epoch": 0.9936678032148076,
      "grad_norm": 0.5291311144828796,
      "learning_rate": 9.362908654986235e-09,
      "loss": 0.5844589233398437,
      "memory(GiB)": 291.54,
      "step": 510,
      "token_acc": 0.8063741960257271,
      "train_speed(iter/s)": 0.007071
    },
    {
      "epoch": 0.9995129079396006,
      "eval_loss": 0.5115394592285156,
      "eval_runtime": 26.9687,
      "eval_samples_per_second": 0.148,
      "eval_steps_per_second": 0.074,
      "step": 513
    }
  ],
  "logging_steps": 5,
  "max_steps": 513,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 50,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 8.13335087128363e+18,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}