{
  "best_global_step": 123,
  "best_metric": 0.09478498,
  "best_model_checkpoint": "/media/datasets/function_call/output/fc_sft/sft_72B/0807_1/v0-20250807-112957/checkpoint-123",
  "epoch": 2.9397590361445785,
  "eval_steps": 200.0,
  "global_step": 123,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.024096385542168676,
      "grad_norm": 7.1235077109741285,
      "learning_rate": 7.142857142857143e-07,
      "loss": 0.37839245796203613,
      "memory(GiB)": 66.87,
      "step": 1,
      "token_acc": 0.9095865515674694,
      "train_speed(iter/s)": 0.008945
    },
    {
      "epoch": 0.04819277108433735,
      "grad_norm": 5.216817496713617,
      "learning_rate": 1.4285714285714286e-06,
      "loss": 0.28986841440200806,
      "memory(GiB)": 66.87,
      "step": 2,
      "token_acc": 0.9272056356927205,
      "train_speed(iter/s)": 0.00959
    },
    {
      "epoch": 0.07228915662650602,
      "grad_norm": 5.495376872120821,
      "learning_rate": 2.1428571428571427e-06,
      "loss": 0.31616902351379395,
      "memory(GiB)": 66.87,
      "step": 3,
      "token_acc": 0.9123665480427047,
      "train_speed(iter/s)": 0.00983
    },
    {
      "epoch": 0.0963855421686747,
      "grad_norm": 4.089251029148955,
      "learning_rate": 2.8571428571428573e-06,
      "loss": 0.27855414152145386,
      "memory(GiB)": 66.87,
      "step": 4,
      "token_acc": 0.9210363972856261,
      "train_speed(iter/s)": 0.009983
    },
    {
      "epoch": 0.12048192771084337,
      "grad_norm": 2.9710535096364135,
      "learning_rate": 3.5714285714285718e-06,
      "loss": 0.26068973541259766,
      "memory(GiB)": 66.87,
      "step": 5,
      "token_acc": 0.9541648303217276,
      "train_speed(iter/s)": 0.010075
    },
    {
      "epoch": 0.14457831325301204,
      "grad_norm": 2.865412266601039,
      "learning_rate": 4.2857142857142855e-06,
      "loss": 0.2701705992221832,
      "memory(GiB)": 66.87,
      "step": 6,
      "token_acc": 0.9072621888851006,
      "train_speed(iter/s)": 0.010132
    },
    {
      "epoch": 0.1686746987951807,
      "grad_norm": 3.8674699062256286,
      "learning_rate": 5e-06,
      "loss": 0.2653254568576813,
      "memory(GiB)": 66.88,
      "step": 7,
      "token_acc": 0.9280945043656909,
      "train_speed(iter/s)": 0.010175
    },
    {
      "epoch": 0.1927710843373494,
      "grad_norm": 2.520783928060602,
      "learning_rate": 4.999083215558211e-06,
      "loss": 0.2684745192527771,
      "memory(GiB)": 66.88,
      "step": 8,
      "token_acc": 0.9066886870355079,
      "train_speed(iter/s)": 0.010185
    },
    {
      "epoch": 0.21686746987951808,
      "grad_norm": 3.649986588139491,
      "learning_rate": 4.99633353462781e-06,
      "loss": 0.25809189677238464,
      "memory(GiB)": 66.88,
      "step": 9,
      "token_acc": 0.9039509536784741,
      "train_speed(iter/s)": 0.010214
    },
    {
      "epoch": 0.24096385542168675,
      "grad_norm": 5.104585903169255,
      "learning_rate": 4.991752973900558e-06,
      "loss": 0.2833491563796997,
      "memory(GiB)": 66.88,
      "step": 10,
      "token_acc": 0.8818365968939905,
      "train_speed(iter/s)": 0.01022
    },
    {
      "epoch": 0.26506024096385544,
      "grad_norm": 2.5530044047159945,
      "learning_rate": 4.985344892885899e-06,
      "loss": 0.23303940892219543,
      "memory(GiB)": 66.88,
      "step": 11,
      "token_acc": 0.9077690070627337,
      "train_speed(iter/s)": 0.010238
    },
    {
      "epoch": 0.2891566265060241,
      "grad_norm": 1.9050596145427132,
      "learning_rate": 4.977113991447017e-06,
      "loss": 0.22975286841392517,
      "memory(GiB)": 66.88,
      "step": 12,
      "token_acc": 0.9186147186147187,
      "train_speed(iter/s)": 0.010244
    },
    {
      "epoch": 0.3132530120481928,
      "grad_norm": 1.722949431226048,
      "learning_rate": 4.967066306353816e-06,
      "loss": 0.22108271718025208,
      "memory(GiB)": 66.88,
      "step": 13,
      "token_acc": 0.8714975845410629,
      "train_speed(iter/s)": 0.010261
    },
    {
      "epoch": 0.3373493975903614,
      "grad_norm": 2.877534376680796,
      "learning_rate": 4.95520920685539e-06,
      "loss": 0.2381371259689331,
      "memory(GiB)": 66.88,
      "step": 14,
      "token_acc": 0.9098256061250531,
      "train_speed(iter/s)": 0.010269
    },
    {
      "epoch": 0.3614457831325301,
      "grad_norm": 1.676750043812651,
      "learning_rate": 4.941551389275217e-06,
      "loss": 0.20479455590248108,
      "memory(GiB)": 66.88,
      "step": 15,
      "token_acc": 0.8998109640831758,
      "train_speed(iter/s)": 0.010288
    },
    {
      "epoch": 0.3855421686746988,
      "grad_norm": 1.566969000551355,
      "learning_rate": 4.926102870633029e-06,
      "loss": 0.21772323548793793,
      "memory(GiB)": 66.88,
      "step": 16,
      "token_acc": 0.9114414996767938,
      "train_speed(iter/s)": 0.010299
    },
    {
      "epoch": 0.40963855421686746,
      "grad_norm": 1.474775133905508,
      "learning_rate": 4.908874981298058e-06,
      "loss": 0.22780515253543854,
      "memory(GiB)": 66.88,
      "step": 17,
      "token_acc": 0.9121887287024901,
      "train_speed(iter/s)": 0.01031
    },
    {
      "epoch": 0.43373493975903615,
      "grad_norm": 1.1686868238081405,
      "learning_rate": 4.88988035667903e-06,
      "loss": 0.19028934836387634,
      "memory(GiB)": 66.88,
      "step": 18,
      "token_acc": 0.9370277078085643,
      "train_speed(iter/s)": 0.010309
    },
    {
      "epoch": 0.4578313253012048,
      "grad_norm": 1.025204511386218,
      "learning_rate": 4.869132927957007e-06,
      "loss": 0.18262633681297302,
      "memory(GiB)": 66.88,
      "step": 19,
      "token_acc": 0.9488215488215488,
      "train_speed(iter/s)": 0.010315
    },
    {
      "epoch": 0.4819277108433735,
      "grad_norm": 1.0162596411054252,
      "learning_rate": 4.846647911867877e-06,
      "loss": 0.196017324924469,
      "memory(GiB)": 66.88,
      "step": 20,
      "token_acc": 0.9034635224760501,
      "train_speed(iter/s)": 0.010313
    },
    {
      "epoch": 0.5060240963855421,
      "grad_norm": 1.1383788259333585,
      "learning_rate": 4.822441799541979e-06,
      "loss": 0.16589009761810303,
      "memory(GiB)": 66.88,
      "step": 21,
      "token_acc": 0.9454836985569214,
      "train_speed(iter/s)": 0.010311
    },
    {
      "epoch": 0.5301204819277109,
      "grad_norm": 1.04542227175653,
      "learning_rate": 4.796532344409055e-06,
      "loss": 0.1764078438282013,
      "memory(GiB)": 66.88,
      "step": 22,
      "token_acc": 0.9518427518427518,
      "train_speed(iter/s)": 0.010325
    },
    {
      "epoch": 0.5542168674698795,
      "grad_norm": 0.8856085586603806,
      "learning_rate": 4.7689385491773934e-06,
      "loss": 0.1612783968448639,
      "memory(GiB)": 66.88,
      "step": 23,
      "token_acc": 0.9456440804169769,
      "train_speed(iter/s)": 0.010334
    },
    {
      "epoch": 0.5783132530120482,
      "grad_norm": 0.8636049971993279,
      "learning_rate": 4.739680651896721e-06,
      "loss": 0.1550418734550476,
      "memory(GiB)": 66.88,
      "step": 24,
      "token_acc": 0.9572158365261814,
      "train_speed(iter/s)": 0.010331
    },
    {
      "epoch": 0.6024096385542169,
      "grad_norm": 0.9928165075449613,
      "learning_rate": 4.708780111115058e-06,
      "loss": 0.15980197489261627,
      "memory(GiB)": 66.88,
      "step": 25,
      "token_acc": 0.9279279279279279,
      "train_speed(iter/s)": 0.010336
    },
    {
      "epoch": 0.6265060240963856,
      "grad_norm": 0.8966252290088227,
      "learning_rate": 4.676259590140431e-06,
      "loss": 0.15649080276489258,
      "memory(GiB)": 66.88,
      "step": 26,
      "token_acc": 0.9375,
      "train_speed(iter/s)": 0.010336
    },
    {
      "epoch": 0.6506024096385542,
      "grad_norm": 0.9239434371130526,
      "learning_rate": 4.642142940418973e-06,
      "loss": 0.15065842866897583,
      "memory(GiB)": 66.88,
      "step": 27,
      "token_acc": 0.9709302325581395,
      "train_speed(iter/s)": 0.010343
    },
    {
      "epoch": 0.6746987951807228,
      "grad_norm": 0.8608562567014957,
      "learning_rate": 4.606455184041623e-06,
      "loss": 0.16093730926513672,
      "memory(GiB)": 66.88,
      "step": 28,
      "token_acc": 0.9602459016393443,
      "train_speed(iter/s)": 0.010345
    },
    {
      "epoch": 0.6987951807228916,
      "grad_norm": 0.8471041034553904,
      "learning_rate": 4.569222495392227e-06,
      "loss": 0.13956543803215027,
      "memory(GiB)": 66.88,
      "step": 29,
      "token_acc": 0.9462642620537357,
      "train_speed(iter/s)": 0.010346
    },
    {
      "epoch": 0.7228915662650602,
      "grad_norm": 0.9549475802923806,
      "learning_rate": 4.530472181950528e-06,
      "loss": 0.1541374921798706,
      "memory(GiB)": 66.88,
      "step": 30,
      "token_acc": 0.9293809938971229,
      "train_speed(iter/s)": 0.010345
    },
    {
      "epoch": 0.7469879518072289,
      "grad_norm": 0.7983871599131706,
      "learning_rate": 4.49023266426411e-06,
      "loss": 0.16580992937088013,
      "memory(GiB)": 66.88,
      "step": 31,
      "token_acc": 0.9374701670644391,
      "train_speed(iter/s)": 0.010343
    },
    {
      "epoch": 0.7710843373493976,
      "grad_norm": 0.7801842283566953,
      "learning_rate": 4.448533455103979e-06,
      "loss": 0.15877921879291534,
      "memory(GiB)": 66.88,
      "step": 32,
      "token_acc": 0.9319371727748691,
      "train_speed(iter/s)": 0.010347
    },
    {
      "epoch": 0.7951807228915663,
      "grad_norm": 0.7462651699964246,
      "learning_rate": 4.4054051378190915e-06,
      "loss": 0.16641965508460999,
      "memory(GiB)": 66.88,
      "step": 33,
      "token_acc": 0.9523539421440727,
      "train_speed(iter/s)": 0.010346
    },
    {
      "epoch": 0.8192771084337349,
      "grad_norm": 0.751310839015754,
      "learning_rate": 4.360879343905677e-06,
      "loss": 0.1544458270072937,
      "memory(GiB)": 66.88,
      "step": 34,
      "token_acc": 0.9319052987598647,
      "train_speed(iter/s)": 0.010344
    },
    {
      "epoch": 0.8433734939759037,
      "grad_norm": 0.9944959671354999,
      "learning_rate": 4.3149887298078275e-06,
      "loss": 0.15687474608421326,
      "memory(GiB)": 66.88,
      "step": 35,
      "token_acc": 0.9662415603900976,
      "train_speed(iter/s)": 0.010343
    },
    {
      "epoch": 0.8674698795180723,
      "grad_norm": 0.7619739781041097,
      "learning_rate": 4.267766952966369e-06,
      "loss": 0.14246270060539246,
      "memory(GiB)": 66.88,
      "step": 36,
      "token_acc": 0.9526484751203852,
      "train_speed(iter/s)": 0.010342
    },
    {
      "epoch": 0.891566265060241,
      "grad_norm": 1.064985568075927,
      "learning_rate": 4.219248647133559e-06,
      "loss": 0.15572598576545715,
      "memory(GiB)": 66.88,
      "step": 37,
      "token_acc": 0.9555444555444556,
      "train_speed(iter/s)": 0.010339
    },
    {
      "epoch": 0.9156626506024096,
      "grad_norm": 1.0976178039462479,
      "learning_rate": 4.169469396971739e-06,
      "loss": 0.14152270555496216,
      "memory(GiB)": 66.88,
      "step": 38,
      "token_acc": 0.9561653599429794,
      "train_speed(iter/s)": 0.010342
    },
    {
      "epoch": 0.9397590361445783,
      "grad_norm": 0.9042268662231224,
      "learning_rate": 4.11846571195457e-06,
      "loss": 0.15433737635612488,
      "memory(GiB)": 66.88,
      "step": 39,
      "token_acc": 0.9299969296898987,
      "train_speed(iter/s)": 0.010341
    },
    {
      "epoch": 0.963855421686747,
      "grad_norm": 0.7527693734358947,
      "learning_rate": 4.066274999589967e-06,
      "loss": 0.1318328082561493,
      "memory(GiB)": 66.88,
      "step": 40,
      "token_acc": 0.953932214544258,
      "train_speed(iter/s)": 0.010347
    },
    {
      "epoch": 0.9879518072289156,
      "grad_norm": 0.7622165327787616,
      "learning_rate": 4.012935537984414e-06,
      "loss": 0.12496253103017807,
      "memory(GiB)": 66.88,
      "step": 41,
      "token_acc": 0.9602356406480118,
      "train_speed(iter/s)": 0.01035
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.9529525949598356,
      "learning_rate": 3.958486447768736e-06,
      "loss": 0.12535671889781952,
      "memory(GiB)": 66.88,
      "step": 42,
      "token_acc": 0.9687979539641943,
      "train_speed(iter/s)": 0.010474
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.10762699693441391,
      "eval_runtime": 6.183,
      "eval_samples_per_second": 4.043,
      "eval_steps_per_second": 0.162,
      "eval_token_acc": 0.9530713275409174,
      "step": 42
    },
    {
      "epoch": 1.0240963855421688,
      "grad_norm": 0.675205383537535,
      "learning_rate": 3.9029676634059565e-06,
      "loss": 0.12080010771751404,
      "memory(GiB)": 75.03,
      "step": 43,
      "token_acc": 0.9646790733004178,
      "train_speed(iter/s)": 0.009981
    },
    {
      "epoch": 1.0481927710843373,
      "grad_norm": 0.9830892622190128,
      "learning_rate": 3.846419903902261e-06,
      "loss": 0.12768766283988953,
      "memory(GiB)": 75.03,
      "step": 44,
      "token_acc": 0.9567686967497633,
      "train_speed(iter/s)": 0.009987
    },
    {
      "epoch": 1.072289156626506,
      "grad_norm": 1.0178401937269919,
      "learning_rate": 3.788884642942555e-06,
      "loss": 0.11760720610618591,
      "memory(GiB)": 75.03,
      "step": 45,
      "token_acc": 0.9577819785759294,
      "train_speed(iter/s)": 0.009993
    },
    {
      "epoch": 1.0963855421686748,
      "grad_norm": 0.8225408471293726,
      "learning_rate": 3.7304040784725183e-06,
      "loss": 0.11947129666805267,
      "memory(GiB)": 75.03,
      "step": 46,
      "token_acc": 0.9695869269178393,
      "train_speed(iter/s)": 0.010002
    },
    {
      "epoch": 1.1204819277108433,
      "grad_norm": 0.7569828869018084,
      "learning_rate": 3.671021101749476e-06,
      "loss": 0.11065046489238739,
      "memory(GiB)": 75.03,
      "step": 47,
      "token_acc": 0.9740143369175627,
      "train_speed(iter/s)": 0.01001
    },
    {
      "epoch": 1.144578313253012,
      "grad_norm": 0.677966017743133,
      "learning_rate": 3.6107792658847597e-06,
      "loss": 0.11033762991428375,
      "memory(GiB)": 75.03,
      "step": 48,
      "token_acc": 0.9390243902439024,
      "train_speed(iter/s)": 0.010016
    },
    {
      "epoch": 1.1686746987951806,
      "grad_norm": 0.7227539404594095,
      "learning_rate": 3.549722753900662e-06,
      "loss": 0.11875811219215393,
      "memory(GiB)": 75.03,
      "step": 49,
      "token_acc": 0.9473170731707317,
      "train_speed(iter/s)": 0.010021
    },
    {
      "epoch": 1.1927710843373494,
      "grad_norm": 0.7823579854659256,
      "learning_rate": 3.487896346325389e-06,
      "loss": 0.12093456089496613,
      "memory(GiB)": 75.03,
      "step": 50,
      "token_acc": 0.9775,
      "train_speed(iter/s)": 0.010026
    },
    {
      "epoch": 1.216867469879518,
      "grad_norm": 0.7006981730271753,
      "learning_rate": 3.425345388349787e-06,
      "loss": 0.10969439893960953,
      "memory(GiB)": 75.03,
      "step": 51,
      "token_acc": 0.9723674383868559,
      "train_speed(iter/s)": 0.010032
    },
    {
      "epoch": 1.2409638554216866,
      "grad_norm": 0.7667508271941734,
      "learning_rate": 3.3621157565699265e-06,
      "loss": 0.10067746043205261,
      "memory(GiB)": 75.03,
      "step": 52,
      "token_acc": 0.9554854539081669,
      "train_speed(iter/s)": 0.010036
    },
    {
      "epoch": 1.2650602409638554,
      "grad_norm": 0.8645109799323007,
      "learning_rate": 3.29825382533995e-06,
      "loss": 0.10228439420461655,
      "memory(GiB)": 75.03,
      "step": 53,
      "token_acc": 0.9547717842323652,
      "train_speed(iter/s)": 0.01004
    },
    {
      "epoch": 1.2891566265060241,
      "grad_norm": 0.7249771413415558,
      "learning_rate": 3.233806432759837e-06,
      "loss": 0.10691750049591064,
      "memory(GiB)": 75.03,
      "step": 54,
      "token_acc": 0.9375,
      "train_speed(iter/s)": 0.010043
    },
    {
      "epoch": 1.3132530120481927,
      "grad_norm": 0.6960467932963366,
      "learning_rate": 3.168820846323053e-06,
      "loss": 0.10835510492324829,
      "memory(GiB)": 75.03,
      "step": 55,
      "token_acc": 0.971547125074096,
      "train_speed(iter/s)": 0.010046
    },
    {
      "epoch": 1.3373493975903614,
      "grad_norm": 0.7542564468335227,
      "learning_rate": 3.1033447282492645e-06,
      "loss": 0.10559524595737457,
      "memory(GiB)": 75.03,
      "step": 56,
      "token_acc": 0.9476978095663835,
      "train_speed(iter/s)": 0.01005
    },
    {
      "epoch": 1.3614457831325302,
      "grad_norm": 0.6655120768950162,
      "learning_rate": 3.0374261005275606e-06,
      "loss": 0.09428796917200089,
      "memory(GiB)": 75.03,
      "step": 57,
      "token_acc": 0.957510729613734,
      "train_speed(iter/s)": 0.010055
    },
    {
      "epoch": 1.3855421686746987,
      "grad_norm": 0.6989464425598889,
      "learning_rate": 2.971113309695796e-06,
      "loss": 0.10525140911340714,
      "memory(GiB)": 75.03,
      "step": 58,
      "token_acc": 0.9479818078453667,
      "train_speed(iter/s)": 0.010061
    },
    {
      "epoch": 1.4096385542168675,
      "grad_norm": 0.746960091364125,
      "learning_rate": 2.9044549913819125e-06,
      "loss": 0.10597580671310425,
      "memory(GiB)": 75.03,
      "step": 59,
      "token_acc": 0.9496567505720824,
      "train_speed(iter/s)": 0.010069
    },
    {
      "epoch": 1.4337349397590362,
      "grad_norm": 0.7231703263838865,
      "learning_rate": 2.8375000346332256e-06,
      "loss": 0.1139477789402008,
      "memory(GiB)": 75.03,
      "step": 60,
      "token_acc": 0.981311706629055,
      "train_speed(iter/s)": 0.010074
    },
    {
      "epoch": 1.4578313253012047,
      "grad_norm": 0.8089868314868126,
      "learning_rate": 2.7702975460598545e-06,
      "loss": 0.11910184472799301,
      "memory(GiB)": 75.03,
      "step": 61,
      "token_acc": 0.9380378657487092,
      "train_speed(iter/s)": 0.010079
    },
    {
      "epoch": 1.4819277108433735,
      "grad_norm": 0.7567376876370934,
      "learning_rate": 2.7028968138185783e-06,
      "loss": 0.10363476723432541,
      "memory(GiB)": 75.03,
      "step": 62,
      "token_acc": 0.9650677789363921,
      "train_speed(iter/s)": 0.010083
    },
    {
      "epoch": 1.5060240963855422,
      "grad_norm": 0.7194977444577725,
      "learning_rate": 2.6353472714635443e-06,
      "loss": 0.10554774105548859,
      "memory(GiB)": 75.03,
      "step": 63,
      "token_acc": 0.9589958158995816,
      "train_speed(iter/s)": 0.010085
    },
    {
      "epoch": 1.5301204819277108,
      "grad_norm": 0.6975689338170576,
      "learning_rate": 2.5676984616903367e-06,
      "loss": 0.1106470376253128,
      "memory(GiB)": 75.03,
      "step": 64,
      "token_acc": 0.969179127250534,
      "train_speed(iter/s)": 0.010089
    },
    {
      "epoch": 1.5542168674698795,
      "grad_norm": 0.7121745712128734,
      "learning_rate": 2.5e-06,
      "loss": 0.09374865889549255,
      "memory(GiB)": 75.03,
      "step": 65,
      "token_acc": 0.9835020622422197,
      "train_speed(iter/s)": 0.010094
    },
    {
      "epoch": 1.5783132530120483,
      "grad_norm": 0.7213534497613325,
      "learning_rate": 2.4323015383096645e-06,
      "loss": 0.11029407382011414,
      "memory(GiB)": 75.03,
      "step": 66,
      "token_acc": 0.9419250347061545,
      "train_speed(iter/s)": 0.010096
    },
    {
      "epoch": 1.6024096385542168,
      "grad_norm": 0.7061016465436212,
      "learning_rate": 2.3646527285364565e-06,
      "loss": 0.11015942692756653,
      "memory(GiB)": 75.03,
      "step": 67,
      "token_acc": 0.9552529182879378,
      "train_speed(iter/s)": 0.010102
    },
    {
      "epoch": 1.6265060240963856,
      "grad_norm": 0.7015643852305689,
      "learning_rate": 2.2971031861814225e-06,
      "loss": 0.10165303200483322,
      "memory(GiB)": 75.03,
      "step": 68,
      "token_acc": 0.9652042360060514,
      "train_speed(iter/s)": 0.010104
    },
    {
      "epoch": 1.6506024096385543,
      "grad_norm": 0.7330570412944479,
      "learning_rate": 2.2297024539401463e-06,
      "loss": 0.0985989049077034,
      "memory(GiB)": 75.03,
      "step": 69,
      "token_acc": 0.9681857265692175,
      "train_speed(iter/s)": 0.010107
    },
    {
      "epoch": 1.6746987951807228,
      "grad_norm": 0.6169691822087203,
      "learning_rate": 2.162499965366775e-06,
      "loss": 0.09456663578748703,
      "memory(GiB)": 75.03,
      "step": 70,
      "token_acc": 0.9653179190751445,
      "train_speed(iter/s)": 0.010113
    },
    {
      "epoch": 1.6987951807228916,
      "grad_norm": 0.650609693511802,
      "learning_rate": 2.0955450086180883e-06,
      "loss": 0.10181058198213577,
      "memory(GiB)": 75.03,
      "step": 71,
      "token_acc": 0.9512961508248232,
      "train_speed(iter/s)": 0.010117
    },
    {
      "epoch": 1.7228915662650603,
      "grad_norm": 0.894180076929431,
      "learning_rate": 2.0288866903042055e-06,
      "loss": 0.10520657896995544,
      "memory(GiB)": 75.03,
      "step": 72,
      "token_acc": 0.9808724832214765,
      "train_speed(iter/s)": 0.01012
    },
    {
      "epoch": 1.7469879518072289,
      "grad_norm": 0.7239458919863225,
      "learning_rate": 1.96257389947244e-06,
      "loss": 0.10151943564414978,
      "memory(GiB)": 75.03,
      "step": 73,
      "token_acc": 0.9522144522144522,
      "train_speed(iter/s)": 0.010124
    },
    {
      "epoch": 1.7710843373493976,
      "grad_norm": 0.6615927229655446,
      "learning_rate": 1.8966552717507364e-06,
      "loss": 0.10701332241296768,
      "memory(GiB)": 75.03,
      "step": 74,
      "token_acc": 0.9559075342465754,
      "train_speed(iter/s)": 0.010127
    },
    {
      "epoch": 1.7951807228915664,
      "grad_norm": 0.6707891181338171,
      "learning_rate": 1.8311791536769485e-06,
      "loss": 0.09327825903892517,
      "memory(GiB)": 75.03,
      "step": 75,
      "token_acc": 0.9833333333333333,
      "train_speed(iter/s)": 0.010132
    },
    {
      "epoch": 1.819277108433735,
      "grad_norm": 0.6705154749935576,
      "learning_rate": 1.7661935672401635e-06,
      "loss": 0.11036819964647293,
      "memory(GiB)": 75.03,
      "step": 76,
      "token_acc": 0.9554263565891473,
      "train_speed(iter/s)": 0.010134
    },
    {
      "epoch": 1.8433734939759037,
      "grad_norm": 0.7309781303962557,
      "learning_rate": 1.7017461746600506e-06,
      "loss": 0.10548846423625946,
      "memory(GiB)": 75.03,
      "step": 77,
      "token_acc": 0.9674674674674675,
      "train_speed(iter/s)": 0.010138
    },
    {
      "epoch": 1.8674698795180724,
      "grad_norm": 0.616731022618215,
      "learning_rate": 1.6378842434300746e-06,
      "loss": 0.09253889322280884,
      "memory(GiB)": 75.03,
      "step": 78,
      "token_acc": 0.9505640345056403,
      "train_speed(iter/s)": 0.010139
    },
    {
      "epoch": 1.891566265060241,
      "grad_norm": 0.7836770271904427,
      "learning_rate": 1.574654611650214e-06,
      "loss": 0.09159648418426514,
      "memory(GiB)": 75.03,
      "step": 79,
      "token_acc": 0.9469214437367304,
      "train_speed(iter/s)": 0.010141
    },
    {
      "epoch": 1.9156626506024095,
      "grad_norm": 0.5958099492524951,
      "learning_rate": 1.5121036536746119e-06,
      "loss": 0.09998425841331482,
      "memory(GiB)": 75.03,
      "step": 80,
      "token_acc": 0.9657097288676236,
      "train_speed(iter/s)": 0.010144
    },
    {
      "epoch": 1.9397590361445785,
      "grad_norm": 0.6355843480799813,
      "learning_rate": 1.4502772460993387e-06,
      "loss": 0.1011456847190857,
      "memory(GiB)": 75.03,
      "step": 81,
      "token_acc": 0.9753347427766033,
      "train_speed(iter/s)": 0.010147
    },
    {
      "epoch": 1.963855421686747,
      "grad_norm": 0.5918036358439085,
      "learning_rate": 1.3892207341152416e-06,
      "loss": 0.09216572344303131,
      "memory(GiB)": 75.03,
      "step": 82,
      "token_acc": 0.9726095617529881,
      "train_speed(iter/s)": 0.010149
    },
    {
      "epoch": 1.9879518072289155,
      "grad_norm": 0.6184297641234107,
      "learning_rate": 1.328978898250525e-06,
      "loss": 0.08801873028278351,
      "memory(GiB)": 75.03,
      "step": 83,
      "token_acc": 0.9716621253405995,
      "train_speed(iter/s)": 0.010153
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.9863985514790247,
      "learning_rate": 1.2695959215274817e-06,
      "loss": 0.07605957239866257,
      "memory(GiB)": 75.03,
      "step": 84,
      "token_acc": 0.975669099756691,
      "train_speed(iter/s)": 0.010215
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.09668351709842682,
      "eval_runtime": 6.1433,
      "eval_samples_per_second": 4.069,
      "eval_steps_per_second": 0.163,
      "eval_token_acc": 0.9575166700343504,
      "step": 84
    },
    {
      "epoch": 2.0240963855421685,
      "grad_norm": 0.5948680904279247,
      "learning_rate": 1.2111153570574454e-06,
      "loss": 0.08509032428264618,
      "memory(GiB)": 75.03,
      "step": 85,
      "token_acc": 0.9708614864864865,
      "train_speed(iter/s)": 0.009988
    },
    {
      "epoch": 2.0481927710843375,
      "grad_norm": 1.1027068517185152,
      "learning_rate": 1.1535800960977398e-06,
      "loss": 0.06897895038127899,
      "memory(GiB)": 75.03,
      "step": 86,
      "token_acc": 0.9619883040935673,
      "train_speed(iter/s)": 0.009992
    },
    {
      "epoch": 2.072289156626506,
      "grad_norm": 0.6795426764304612,
      "learning_rate": 1.0970323365940443e-06,
      "loss": 0.07773903012275696,
      "memory(GiB)": 75.03,
      "step": 87,
      "token_acc": 0.9551307062036676,
      "train_speed(iter/s)": 0.009997
    },
    {
      "epoch": 2.0963855421686746,
      "grad_norm": 0.6159758641709957,
      "learning_rate": 1.041513552231265e-06,
      "loss": 0.07336698472499847,
      "memory(GiB)": 75.03,
      "step": 88,
      "token_acc": 0.9743249225320938,
      "train_speed(iter/s)": 0.01
    },
    {
      "epoch": 2.1204819277108435,
      "grad_norm": 0.719141034819639,
      "learning_rate": 9.870644620155878e-07,
      "loss": 0.0716508999466896,
      "memory(GiB)": 75.03,
      "step": 89,
      "token_acc": 0.9761323569297532,
      "train_speed(iter/s)": 0.010003
    },
    {
      "epoch": 2.144578313253012,
      "grad_norm": 0.6096112815462216,
      "learning_rate": 9.337250004100337e-07,
      "loss": 0.07127302885055542,
      "memory(GiB)": 75.03,
      "step": 90,
      "token_acc": 0.9869207772795217,
      "train_speed(iter/s)": 0.010007
    },
    {
      "epoch": 2.1686746987951806,
      "grad_norm": 0.5951192545176354,
      "learning_rate": 8.815342880454312e-07,
      "loss": 0.07116035372018814,
      "memory(GiB)": 75.03,
      "step": 91,
      "token_acc": 0.9435483870967742,
      "train_speed(iter/s)": 0.010012
    },
    {
      "epoch": 2.1927710843373496,
      "grad_norm": 0.6191831144088478,
      "learning_rate": 8.305306030282618e-07,
      "loss": 0.07278469204902649,
      "memory(GiB)": 75.03,
      "step": 92,
      "token_acc": 0.9500645716745587,
      "train_speed(iter/s)": 0.010017
    },
    {
      "epoch": 2.216867469879518,
      "grad_norm": 0.6707704813216955,
      "learning_rate": 7.807513528664415e-07,
      "loss": 0.06988117843866348,
      "memory(GiB)": 75.03,
      "step": 93,
      "token_acc": 0.9820839226779821,
      "train_speed(iter/s)": 0.010023
    },
    {
      "epoch": 2.2409638554216866,
      "grad_norm": 0.7333364217183101,
      "learning_rate": 7.322330470336314e-07,
      "loss": 0.07439117133617401,
      "memory(GiB)": 75.03,
      "step": 94,
      "token_acc": 0.9460376194881283,
      "train_speed(iter/s)": 0.010026
    },
    {
      "epoch": 2.2650602409638556,
      "grad_norm": 0.6438557650699052,
      "learning_rate": 6.850112701921735e-07,
      "loss": 0.07780344784259796,
      "memory(GiB)": 75.03,
      "step": 95,
      "token_acc": 0.9549695740365112,
      "train_speed(iter/s)": 0.010029
    },
    {
      "epoch": 2.289156626506024,
      "grad_norm": 0.920087977282289,
      "learning_rate": 6.391206560943241e-07,
      "loss": 0.06655862927436829,
      "memory(GiB)": 75.03,
      "step": 96,
      "token_acc": 0.9930362116991643,
      "train_speed(iter/s)": 0.010032
    },
    {
      "epoch": 2.3132530120481927,
      "grad_norm": 0.6170891932263252,
      "learning_rate": 5.945948621809092e-07,
      "loss": 0.07646074891090393,
      "memory(GiB)": 75.03,
      "step": 97,
      "token_acc": 0.9858607281724991,
      "train_speed(iter/s)": 0.010034
    },
    {
      "epoch": 2.337349397590361,
      "grad_norm": 0.8623684564618234,
      "learning_rate": 5.51466544896021e-07,
      "loss": 0.07112358510494232,
      "memory(GiB)": 75.03,
      "step": 98,
      "token_acc": 0.9751176866173503,
      "train_speed(iter/s)": 0.010038
    },
    {
      "epoch": 2.36144578313253,
      "grad_norm": 0.6212426867999661,
      "learning_rate": 5.097673357358906e-07,
      "loss": 0.0641506090760231,
      "memory(GiB)": 75.03,
      "step": 99,
      "token_acc": 0.9784991568296796,
      "train_speed(iter/s)": 0.01004
    },
    {
      "epoch": 2.3855421686746987,
      "grad_norm": 0.6193686418789898,
      "learning_rate": 4.695278180494725e-07,
      "loss": 0.07258988916873932,
      "memory(GiB)": 75.03,
      "step": 100,
      "token_acc": 0.979624450659209,
      "train_speed(iter/s)": 0.010044
    },
    {
      "epoch": 2.4096385542168672,
      "grad_norm": 0.546469608395153,
      "learning_rate": 4.307775046077739e-07,
      "loss": 0.0687943696975708,
      "memory(GiB)": 75.03,
      "step": 101,
      "token_acc": 0.9673798667134339,
      "train_speed(iter/s)": 0.010048
    },
    {
      "epoch": 2.433734939759036,
      "grad_norm": 0.5621305058472025,
      "learning_rate": 3.935448159583774e-07,
      "loss": 0.06319835036993027,
      "memory(GiB)": 75.03,
      "step": 102,
      "token_acc": 0.9662430500397141,
      "train_speed(iter/s)": 0.010051
    },
    {
      "epoch": 2.4578313253012047,
      "grad_norm": 0.5424762825078937,
      "learning_rate": 3.578570595810274e-07,
      "loss": 0.06497755646705627,
      "memory(GiB)": 75.03,
      "step": 103,
      "token_acc": 0.9771836007130125,
      "train_speed(iter/s)": 0.010053
    },
    {
      "epoch": 2.4819277108433733,
      "grad_norm": 0.6674986563454252,
      "learning_rate": 3.2374040985957005e-07,
      "loss": 0.06723719835281372,
      "memory(GiB)": 75.03,
      "step": 104,
      "token_acc": 0.9736164736164736,
      "train_speed(iter/s)": 0.010056
    },
    {
      "epoch": 2.5060240963855422,
      "grad_norm": 0.6543787100521129,
      "learning_rate": 2.9121988888494297e-07,
      "loss": 0.07685531675815582,
      "memory(GiB)": 75.03,
      "step": 105,
      "token_acc": 0.9603563474387528,
      "train_speed(iter/s)": 0.01006
    },
    {
      "epoch": 2.5301204819277108,
      "grad_norm": 0.5591453440194828,
      "learning_rate": 2.6031934810328006e-07,
      "loss": 0.06915157288312912,
      "memory(GiB)": 75.03,
      "step": 106,
      "token_acc": 0.977219777375097,
      "train_speed(iter/s)": 0.010063
    },
    {
      "epoch": 2.5542168674698793,
      "grad_norm": 0.7683742479099575,
      "learning_rate": 2.3106145082260777e-07,
      "loss": 0.0714510977268219,
      "memory(GiB)": 75.03,
      "step": 107,
      "token_acc": 0.9779321618308132,
      "train_speed(iter/s)": 0.010066
    },
    {
      "epoch": 2.5783132530120483,
      "grad_norm": 0.8136011188969886,
      "learning_rate": 2.0346765559094566e-07,
      "loss": 0.07178479433059692,
      "memory(GiB)": 75.03,
      "step": 108,
      "token_acc": 0.9674718196457327,
      "train_speed(iter/s)": 0.010068
    },
    {
      "epoch": 2.602409638554217,
      "grad_norm": 0.8210355736998771,
      "learning_rate": 1.7755820045802146e-07,
      "loss": 0.0685107558965683,
      "memory(GiB)": 75.03,
      "step": 109,
      "token_acc": 0.9678714859437751,
      "train_speed(iter/s)": 0.010071
    },
    {
      "epoch": 2.6265060240963853,
      "grad_norm": 0.7441349079946483,
      "learning_rate": 1.5335208813212376e-07,
      "loss": 0.07153232395648956,
      "memory(GiB)": 75.03,
      "step": 110,
      "token_acc": 0.9674844812296778,
      "train_speed(iter/s)": 0.010072
    },
    {
      "epoch": 2.6506024096385543,
      "grad_norm": 0.6604976986430176,
      "learning_rate": 1.3086707204299415e-07,
      "loss": 0.06870590150356293,
      "memory(GiB)": 75.03,
      "step": 111,
      "token_acc": 0.9767591994835377,
      "train_speed(iter/s)": 0.010077
    },
    {
      "epoch": 2.674698795180723,
      "grad_norm": 0.5429714194652937,
      "learning_rate": 1.1011964332097114e-07,
      "loss": 0.06135660782456398,
      "memory(GiB)": 75.03,
      "step": 112,
      "token_acc": 0.9758283541553504,
      "train_speed(iter/s)": 0.01008
    },
    {
      "epoch": 2.6987951807228914,
      "grad_norm": 0.7071715603757774,
      "learning_rate": 9.112501870194273e-08,
      "loss": 0.06300944089889526,
      "memory(GiB)": 75.03,
      "step": 113,
      "token_acc": 0.9884735202492212,
      "train_speed(iter/s)": 0.010084
    },
    {
      "epoch": 2.7228915662650603,
      "grad_norm": 0.65456034983842,
      "learning_rate": 7.38971293669713e-08,
      "loss": 0.07272332906723022,
      "memory(GiB)": 75.03,
      "step": 114,
      "token_acc": 0.9806610407876231,
      "train_speed(iter/s)": 0.010086
    },
    {
      "epoch": 2.746987951807229,
      "grad_norm": 0.5781826596184406,
      "learning_rate": 5.844861072478336e-08,
      "loss": 0.06247750297188759,
      "memory(GiB)": 75.03,
      "step": 115,
      "token_acc": 0.9747267759562842,
      "train_speed(iter/s)": 0.010088
    },
    {
      "epoch": 2.7710843373493974,
      "grad_norm": 0.5540943563848175,
      "learning_rate": 4.47907931446101e-08,
      "loss": 0.07139578461647034,
      "memory(GiB)": 75.03,
      "step": 116,
      "token_acc": 0.9693449499838658,
      "train_speed(iter/s)": 0.010088
    },
    {
      "epoch": 2.7951807228915664,
      "grad_norm": 0.5181461682647335,
      "learning_rate": 3.293369364618465e-08,
      "loss": 0.06446193158626556,
      "memory(GiB)": 75.03,
      "step": 117,
      "token_acc": 0.9868049010367578,
      "train_speed(iter/s)": 0.010091
    },
    {
      "epoch": 2.819277108433735,
      "grad_norm": 0.6809547597187211,
      "learning_rate": 2.2886008552983064e-08,
      "loss": 0.07164957374334335,
      "memory(GiB)": 75.03,
      "step": 118,
      "token_acc": 0.9815526518063028,
      "train_speed(iter/s)": 0.010093
    },
    {
      "epoch": 2.8433734939759034,
      "grad_norm": 0.5408062773807014,
      "learning_rate": 1.4655107114101008e-08,
      "loss": 0.05888580530881882,
      "memory(GiB)": 75.03,
      "step": 119,
      "token_acc": 0.9835369091874668,
      "train_speed(iter/s)": 0.010095
    },
    {
      "epoch": 2.8674698795180724,
      "grad_norm": 0.520571861677298,
      "learning_rate": 8.247026099443279e-09,
      "loss": 0.06711478531360626,
      "memory(GiB)": 75.03,
      "step": 120,
      "token_acc": 0.9729596853490659,
      "train_speed(iter/s)": 0.010097
    },
    {
      "epoch": 2.891566265060241,
      "grad_norm": 0.5429001804267765,
      "learning_rate": 3.666465372190453e-09,
      "loss": 0.06482349336147308,
      "memory(GiB)": 75.03,
      "step": 121,
      "token_acc": 0.9841530054644809,
      "train_speed(iter/s)": 0.010099
    },
    {
      "epoch": 2.9156626506024095,
      "grad_norm": 0.638863253266101,
      "learning_rate": 9.167844417901084e-10,
      "loss": 0.06565111875534058,
      "memory(GiB)": 75.03,
      "step": 122,
      "token_acc": 0.9656954887218046,
      "train_speed(iter/s)": 0.010101
    },
    {
      "epoch": 2.9397590361445785,
      "grad_norm": 0.7549296882606735,
      "learning_rate": 0.0,
      "loss": 0.06485162675380707,
      "memory(GiB)": 75.03,
      "step": 123,
      "token_acc": 0.9698492462311558,
      "train_speed(iter/s)": 0.010103
    },
    {
      "epoch": 2.9397590361445785,
      "eval_loss": 0.09478497505187988,
      "eval_runtime": 6.5517,
      "eval_samples_per_second": 3.816,
      "eval_steps_per_second": 0.153,
      "eval_token_acc": 0.9597393412810669,
      "step": 123
    },
    {
      "epoch": 2.9397590361445785,
      "eval_loss": 0.09478497505187988,
      "eval_runtime": 6.7,
      "eval_samples_per_second": 3.731,
      "eval_steps_per_second": 0.149,
      "eval_token_acc": 0.9597393412810669,
      "step": 123
    }
  ],
  "logging_steps": 1,
  "max_steps": 123,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1083906603876352.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}