{ "best_metric": 0.25880399, "best_model_checkpoint": "/workspace/output/molmo-7b-d/v0-20250103-184047/checkpoint-3600", "epoch": 5.0, "eval_steps": 200, "global_step": 3600, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "acc": 0.21068719, "epoch": 0.001388888888888889, "grad_norm": 87.10760137929194, "learning_rate": 0.0, "loss": 4.65820312, "memory(GiB)": 57.09, "step": 1, "train_speed(iter/s)": 0.023892 }, { "acc": 0.2150097, "epoch": 0.006944444444444444, "grad_norm": 69.93924445657471, "learning_rate": 2.994148005679757e-06, "loss": 4.37573242, "memory(GiB)": 130.0, "step": 5, "train_speed(iter/s)": 0.094301 }, { "acc": 0.47089963, "epoch": 0.013888888888888888, "grad_norm": 15.718174458508388, "learning_rate": 4.2836573631282295e-06, "loss": 2.69121094, "memory(GiB)": 130.0, "step": 10, "train_speed(iter/s)": 0.150516 }, { "acc": 0.62566862, "epoch": 0.020833333333333332, "grad_norm": 11.952512545843422, "learning_rate": 5.037971981564619e-06, "loss": 1.37431641, "memory(GiB)": 130.0, "step": 15, "train_speed(iter/s)": 0.188994 }, { "acc": 0.69504013, "epoch": 0.027777777777777776, "grad_norm": 10.664167724305369, "learning_rate": 5.5731667205767005e-06, "loss": 0.96235352, "memory(GiB)": 130.0, "step": 20, "train_speed(iter/s)": 0.215114 }, { "acc": 0.69391584, "epoch": 0.034722222222222224, "grad_norm": 15.301457987834718, "learning_rate": 5.988296011359514e-06, "loss": 0.91097412, "memory(GiB)": 133.48, "step": 25, "train_speed(iter/s)": 0.230533 }, { "acc": 0.71258931, "epoch": 0.041666666666666664, "grad_norm": 5.236783763280491, "learning_rate": 6.327481339013091e-06, "loss": 0.8465271, "memory(GiB)": 100.81, "step": 30, "train_speed(iter/s)": 0.245112 }, { "acc": 0.71047335, "epoch": 0.04861111111111111, "grad_norm": 13.907732912979608, "learning_rate": 6.614258447352063e-06, "loss": 0.89771729, "memory(GiB)": 100.81, "step": 35, "train_speed(iter/s)": 0.256837 }, { "acc": 0.69757376, "epoch": 0.05555555555555555, "grad_norm": 4.835405359715248, "learning_rate": 6.862676078025172e-06, "loss": 0.85203247, "memory(GiB)": 100.81, "step": 40, "train_speed(iter/s)": 0.263983 }, { "acc": 0.7098877, "epoch": 0.0625, "grad_norm": 12.652916823049557, "learning_rate": 7.08179595744948e-06, "loss": 0.84803467, "memory(GiB)": 100.81, "step": 45, "train_speed(iter/s)": 0.272195 }, { "acc": 0.71628637, "epoch": 0.06944444444444445, "grad_norm": 4.0802053942772885, "learning_rate": 7.2778053688079864e-06, "loss": 0.79383545, "memory(GiB)": 100.81, "step": 50, "train_speed(iter/s)": 0.279194 }, { "acc": 0.7037159, "epoch": 0.0763888888888889, "grad_norm": 4.652972869912193, "learning_rate": 7.455117449365667e-06, "loss": 0.81118164, "memory(GiB)": 100.81, "step": 55, "train_speed(iter/s)": 0.285782 }, { "acc": 0.72002425, "epoch": 0.08333333333333333, "grad_norm": 5.532418910043181, "learning_rate": 7.616990696461561e-06, "loss": 0.78905029, "memory(GiB)": 100.81, "step": 60, "train_speed(iter/s)": 0.291011 }, { "acc": 0.70535297, "epoch": 0.09027777777777778, "grad_norm": 4.67406422391853, "learning_rate": 7.765899648896681e-06, "loss": 0.80931396, "memory(GiB)": 100.81, "step": 65, "train_speed(iter/s)": 0.295571 }, { "acc": 0.7147357, "epoch": 0.09722222222222222, "grad_norm": 3.612299097599364, "learning_rate": 7.903767804800537e-06, "loss": 0.80908203, "memory(GiB)": 131.07, "step": 70, "train_speed(iter/s)": 0.300188 }, { "acc": 0.72479091, "epoch": 0.10416666666666667, "grad_norm": 4.568791046839639, "learning_rate": 8.032119987244375e-06, "loss": 0.76064138, "memory(GiB)": 131.07, "step": 75, "train_speed(iter/s)": 0.303671 }, { "acc": 0.73130875, "epoch": 0.1111111111111111, "grad_norm": 5.76867394533742, "learning_rate": 8.152185435473643e-06, "loss": 0.74737253, "memory(GiB)": 131.07, "step": 80, "train_speed(iter/s)": 0.306462 }, { "acc": 0.73129368, "epoch": 0.11805555555555555, "grad_norm": 4.88314712919246, "learning_rate": 8.264969587694988e-06, "loss": 0.7456665, "memory(GiB)": 131.07, "step": 85, "train_speed(iter/s)": 0.308971 }, { "acc": 0.71992922, "epoch": 0.125, "grad_norm": 3.4318023660327444, "learning_rate": 8.371305314897952e-06, "loss": 0.75183105, "memory(GiB)": 131.07, "step": 90, "train_speed(iter/s)": 0.31115 }, { "acc": 0.74187808, "epoch": 0.13194444444444445, "grad_norm": 9.486749512593935, "learning_rate": 8.471890284028081e-06, "loss": 0.73335419, "memory(GiB)": 131.07, "step": 95, "train_speed(iter/s)": 0.31205 }, { "acc": 0.73988142, "epoch": 0.1388888888888889, "grad_norm": 6.046415731830522, "learning_rate": 8.567314726256459e-06, "loss": 0.71626534, "memory(GiB)": 131.07, "step": 100, "train_speed(iter/s)": 0.312061 }, { "acc": 0.74326582, "epoch": 0.14583333333333334, "grad_norm": 6.80033824229626, "learning_rate": 8.658082423236924e-06, "loss": 0.70533233, "memory(GiB)": 131.07, "step": 105, "train_speed(iter/s)": 0.314121 }, { "acc": 0.75279951, "epoch": 0.1527777777777778, "grad_norm": 10.065172541901923, "learning_rate": 8.744626806814138e-06, "loss": 0.67003098, "memory(GiB)": 131.07, "step": 110, "train_speed(iter/s)": 0.315442 }, { "acc": 0.77138567, "epoch": 0.1597222222222222, "grad_norm": 9.015516536250718, "learning_rate": 8.827323477013188e-06, "loss": 0.61771393, "memory(GiB)": 131.07, "step": 115, "train_speed(iter/s)": 0.317097 }, { "acc": 0.76300774, "epoch": 0.16666666666666666, "grad_norm": 6.062711667468016, "learning_rate": 8.906500053910034e-06, "loss": 0.67458191, "memory(GiB)": 131.07, "step": 120, "train_speed(iter/s)": 0.318466 }, { "acc": 0.78507404, "epoch": 0.1736111111111111, "grad_norm": 10.245281951490513, "learning_rate": 8.982444017039273e-06, "loss": 0.61902437, "memory(GiB)": 131.07, "step": 125, "train_speed(iter/s)": 0.319474 }, { "acc": 0.78214965, "epoch": 0.18055555555555555, "grad_norm": 40.92578755154591, "learning_rate": 9.055409006345152e-06, "loss": 0.59720135, "memory(GiB)": 131.07, "step": 130, "train_speed(iter/s)": 0.319858 }, { "acc": 0.78999453, "epoch": 0.1875, "grad_norm": 20.862392861347566, "learning_rate": 9.12561993333434e-06, "loss": 0.58254638, "memory(GiB)": 131.07, "step": 135, "train_speed(iter/s)": 0.320176 }, { "acc": 0.78094759, "epoch": 0.19444444444444445, "grad_norm": 9.395948807027711, "learning_rate": 9.193277162249006e-06, "loss": 0.60503025, "memory(GiB)": 131.07, "step": 140, "train_speed(iter/s)": 0.320895 }, { "acc": 0.78587742, "epoch": 0.2013888888888889, "grad_norm": 13.560643513057176, "learning_rate": 9.2585599572036e-06, "loss": 0.58811264, "memory(GiB)": 131.07, "step": 145, "train_speed(iter/s)": 0.321441 }, { "acc": 0.78708205, "epoch": 0.20833333333333334, "grad_norm": 17.39551162469508, "learning_rate": 9.321629344692848e-06, "loss": 0.56763268, "memory(GiB)": 131.07, "step": 150, "train_speed(iter/s)": 0.321943 }, { "acc": 0.78839636, "epoch": 0.2152777777777778, "grad_norm": 6.0862237740769185, "learning_rate": 9.382630506560327e-06, "loss": 0.56501522, "memory(GiB)": 131.07, "step": 155, "train_speed(iter/s)": 0.322849 }, { "acc": 0.79003353, "epoch": 0.2222222222222222, "grad_norm": 9.332000809791898, "learning_rate": 9.441694792922116e-06, "loss": 0.57846107, "memory(GiB)": 131.07, "step": 160, "train_speed(iter/s)": 0.32378 }, { "acc": 0.80862875, "epoch": 0.22916666666666666, "grad_norm": 7.80508528579547, "learning_rate": 9.498941425250527e-06, "loss": 0.53986168, "memory(GiB)": 131.07, "step": 165, "train_speed(iter/s)": 0.324054 }, { "acc": 0.79791403, "epoch": 0.2361111111111111, "grad_norm": 7.346945680337047, "learning_rate": 9.55447894514346e-06, "loss": 0.55143967, "memory(GiB)": 131.07, "step": 170, "train_speed(iter/s)": 0.325134 }, { "acc": 0.79466968, "epoch": 0.24305555555555555, "grad_norm": 6.541304671392371, "learning_rate": 9.60840645303182e-06, "loss": 0.55263672, "memory(GiB)": 131.07, "step": 175, "train_speed(iter/s)": 0.326197 }, { "acc": 0.79599619, "epoch": 0.25, "grad_norm": 5.061340427236746, "learning_rate": 9.660814672346424e-06, "loss": 0.54171772, "memory(GiB)": 131.07, "step": 180, "train_speed(iter/s)": 0.325379 }, { "acc": 0.80109653, "epoch": 0.2569444444444444, "grad_norm": 4.386102069443633, "learning_rate": 9.711786867849723e-06, "loss": 0.532901, "memory(GiB)": 131.07, "step": 185, "train_speed(iter/s)": 0.325641 }, { "acc": 0.788484, "epoch": 0.2638888888888889, "grad_norm": 9.386068799809824, "learning_rate": 9.761399641476552e-06, "loss": 0.56219482, "memory(GiB)": 131.07, "step": 190, "train_speed(iter/s)": 0.326463 }, { "acc": 0.78657551, "epoch": 0.2708333333333333, "grad_norm": 30.57227532397844, "learning_rate": 9.809723624781542e-06, "loss": 0.59577637, "memory(GiB)": 131.07, "step": 195, "train_speed(iter/s)": 0.327403 }, { "acc": 0.77691326, "epoch": 0.2777777777777778, "grad_norm": 24.227230729653417, "learning_rate": 9.856824083704928e-06, "loss": 0.63096924, "memory(GiB)": 131.07, "step": 200, "train_speed(iter/s)": 0.328235 }, { "acc": 0.79410772, "epoch": 0.2847222222222222, "grad_norm": 12.207454896527581, "learning_rate": 9.902761448651595e-06, "loss": 0.55623169, "memory(GiB)": 131.07, "step": 205, "train_speed(iter/s)": 0.329318 }, { "acc": 0.78385544, "epoch": 0.2916666666666667, "grad_norm": 114.41313028354135, "learning_rate": 9.947591780685397e-06, "loss": 0.60689435, "memory(GiB)": 131.07, "step": 210, "train_speed(iter/s)": 0.33027 }, { "acc": 0.78210087, "epoch": 0.2986111111111111, "grad_norm": 20.141587824624402, "learning_rate": 9.99136718286095e-06, "loss": 0.58652344, "memory(GiB)": 131.07, "step": 215, "train_speed(iter/s)": 0.330942 }, { "acc": 0.7920527, "epoch": 0.3055555555555556, "grad_norm": 8.421958562452112, "learning_rate": 9.999976563040932e-06, "loss": 0.55961218, "memory(GiB)": 131.07, "step": 220, "train_speed(iter/s)": 0.331831 }, { "acc": 0.79507227, "epoch": 0.3125, "grad_norm": 118.8243683029417, "learning_rate": 9.999881350771313e-06, "loss": 0.54424796, "memory(GiB)": 131.07, "step": 225, "train_speed(iter/s)": 0.332761 }, { "acc": 0.78946581, "epoch": 0.3194444444444444, "grad_norm": 36.81477295658913, "learning_rate": 9.999712899774951e-06, "loss": 0.57321162, "memory(GiB)": 131.07, "step": 230, "train_speed(iter/s)": 0.333425 }, { "acc": 0.78228483, "epoch": 0.3263888888888889, "grad_norm": 25.713677437539047, "learning_rate": 9.999471212519574e-06, "loss": 0.56825256, "memory(GiB)": 131.07, "step": 235, "train_speed(iter/s)": 0.334216 }, { "acc": 0.79108438, "epoch": 0.3333333333333333, "grad_norm": 37.784241249057, "learning_rate": 9.999156292545797e-06, "loss": 0.58692875, "memory(GiB)": 131.07, "step": 240, "train_speed(iter/s)": 0.3346 }, { "acc": 0.7929399, "epoch": 0.3402777777777778, "grad_norm": 8.535961270899605, "learning_rate": 9.998768144467059e-06, "loss": 0.57177811, "memory(GiB)": 131.07, "step": 245, "train_speed(iter/s)": 0.335122 }, { "acc": 0.78859062, "epoch": 0.3472222222222222, "grad_norm": 10.099044260879502, "learning_rate": 9.998306773969554e-06, "loss": 0.55594349, "memory(GiB)": 131.07, "step": 250, "train_speed(iter/s)": 0.33582 }, { "acc": 0.80772114, "epoch": 0.3541666666666667, "grad_norm": 37.10201152587716, "learning_rate": 9.997772187812157e-06, "loss": 0.54929962, "memory(GiB)": 131.07, "step": 255, "train_speed(iter/s)": 0.336151 }, { "acc": 0.80398197, "epoch": 0.3611111111111111, "grad_norm": 24.33760888028203, "learning_rate": 9.997164393826322e-06, "loss": 0.54323769, "memory(GiB)": 131.07, "step": 260, "train_speed(iter/s)": 0.336382 }, { "acc": 0.80659904, "epoch": 0.3680555555555556, "grad_norm": 20.76174518011458, "learning_rate": 9.996483400915958e-06, "loss": 0.53362389, "memory(GiB)": 131.07, "step": 265, "train_speed(iter/s)": 0.336602 }, { "acc": 0.80532684, "epoch": 0.375, "grad_norm": 11.08433867212367, "learning_rate": 9.995729219057312e-06, "loss": 0.52986441, "memory(GiB)": 131.07, "step": 270, "train_speed(iter/s)": 0.337078 }, { "acc": 0.80152454, "epoch": 0.3819444444444444, "grad_norm": 10.445600868874815, "learning_rate": 9.994901859298815e-06, "loss": 0.5497118, "memory(GiB)": 131.07, "step": 275, "train_speed(iter/s)": 0.337587 }, { "acc": 0.80949535, "epoch": 0.3888888888888889, "grad_norm": 6.916524475949725, "learning_rate": 9.994001333760923e-06, "loss": 0.5162343, "memory(GiB)": 131.07, "step": 280, "train_speed(iter/s)": 0.338119 }, { "acc": 0.81486473, "epoch": 0.3958333333333333, "grad_norm": 12.150336722267383, "learning_rate": 9.993027655635934e-06, "loss": 0.51019297, "memory(GiB)": 131.07, "step": 285, "train_speed(iter/s)": 0.338392 }, { "acc": 0.80092564, "epoch": 0.4027777777777778, "grad_norm": 7.653355098250675, "learning_rate": 9.991980839187803e-06, "loss": 0.52064533, "memory(GiB)": 131.07, "step": 290, "train_speed(iter/s)": 0.338816 }, { "acc": 0.80846672, "epoch": 0.4097222222222222, "grad_norm": 51.23149633705719, "learning_rate": 9.99086089975193e-06, "loss": 0.49964476, "memory(GiB)": 131.07, "step": 295, "train_speed(iter/s)": 0.339359 }, { "acc": 0.81497822, "epoch": 0.4166666666666667, "grad_norm": 5.848437545743278, "learning_rate": 9.989667853734933e-06, "loss": 0.51864281, "memory(GiB)": 131.07, "step": 300, "train_speed(iter/s)": 0.339412 }, { "acc": 0.81019306, "epoch": 0.4236111111111111, "grad_norm": 9.554808379783061, "learning_rate": 9.988401718614406e-06, "loss": 0.49421182, "memory(GiB)": 131.07, "step": 305, "train_speed(iter/s)": 0.339728 }, { "acc": 0.81571865, "epoch": 0.4305555555555556, "grad_norm": 3.1593614398505028, "learning_rate": 9.98706251293867e-06, "loss": 0.49002447, "memory(GiB)": 131.07, "step": 310, "train_speed(iter/s)": 0.33982 }, { "acc": 0.8122633, "epoch": 0.4375, "grad_norm": 4.285266854421746, "learning_rate": 9.985650256326495e-06, "loss": 0.49020014, "memory(GiB)": 131.07, "step": 315, "train_speed(iter/s)": 0.339074 }, { "acc": 0.79069309, "epoch": 0.4444444444444444, "grad_norm": 5.341762395729506, "learning_rate": 9.984164969466818e-06, "loss": 0.52460327, "memory(GiB)": 131.07, "step": 320, "train_speed(iter/s)": 0.33833 }, { "acc": 0.80525522, "epoch": 0.4513888888888889, "grad_norm": 66.73254513067435, "learning_rate": 9.982606674118437e-06, "loss": 0.5414197, "memory(GiB)": 131.07, "step": 325, "train_speed(iter/s)": 0.338259 }, { "acc": 0.78918257, "epoch": 0.4583333333333333, "grad_norm": 6.91220975909936, "learning_rate": 9.980975393109683e-06, "loss": 0.54792948, "memory(GiB)": 131.07, "step": 330, "train_speed(iter/s)": 0.338625 }, { "acc": 0.81037483, "epoch": 0.4652777777777778, "grad_norm": 4.168771473436414, "learning_rate": 9.979271150338108e-06, "loss": 0.51318879, "memory(GiB)": 131.07, "step": 335, "train_speed(iter/s)": 0.338171 }, { "acc": 0.81083908, "epoch": 0.4722222222222222, "grad_norm": 6.532286933835978, "learning_rate": 9.977493970770112e-06, "loss": 0.50585566, "memory(GiB)": 131.07, "step": 340, "train_speed(iter/s)": 0.338229 }, { "acc": 0.82154369, "epoch": 0.4791666666666667, "grad_norm": 7.835610655269442, "learning_rate": 9.975643880440592e-06, "loss": 0.45731974, "memory(GiB)": 131.07, "step": 345, "train_speed(iter/s)": 0.338112 }, { "acc": 0.80105, "epoch": 0.4861111111111111, "grad_norm": 4.636723486070957, "learning_rate": 9.973720906452551e-06, "loss": 0.50835361, "memory(GiB)": 131.07, "step": 350, "train_speed(iter/s)": 0.338446 }, { "acc": 0.80463896, "epoch": 0.4930555555555556, "grad_norm": 2.7838654689037057, "learning_rate": 9.97172507697671e-06, "loss": 0.52188816, "memory(GiB)": 131.07, "step": 355, "train_speed(iter/s)": 0.338437 }, { "acc": 0.81386986, "epoch": 0.5, "grad_norm": 3.344156589788971, "learning_rate": 9.96965642125109e-06, "loss": 0.47512083, "memory(GiB)": 133.41, "step": 360, "train_speed(iter/s)": 0.338594 }, { "acc": 0.80757666, "epoch": 0.5069444444444444, "grad_norm": 6.585961659135812, "learning_rate": 9.967514969580579e-06, "loss": 0.50586038, "memory(GiB)": 100.72, "step": 365, "train_speed(iter/s)": 0.33889 }, { "acc": 0.81812487, "epoch": 0.5138888888888888, "grad_norm": 3.8878748110567316, "learning_rate": 9.965300753336498e-06, "loss": 0.49280443, "memory(GiB)": 100.72, "step": 370, "train_speed(iter/s)": 0.339136 }, { "acc": 0.81244354, "epoch": 0.5208333333333334, "grad_norm": 3.4377647556680513, "learning_rate": 9.96301380495614e-06, "loss": 0.50686035, "memory(GiB)": 100.72, "step": 375, "train_speed(iter/s)": 0.339232 }, { "acc": 0.81433783, "epoch": 0.5277777777777778, "grad_norm": 4.090164954347545, "learning_rate": 9.960654157942285e-06, "loss": 0.46210661, "memory(GiB)": 101.73, "step": 380, "train_speed(iter/s)": 0.339316 }, { "acc": 0.80798006, "epoch": 0.5347222222222222, "grad_norm": 2.544503744487103, "learning_rate": 9.958221846862717e-06, "loss": 0.47319975, "memory(GiB)": 101.73, "step": 385, "train_speed(iter/s)": 0.33958 }, { "acc": 0.81903315, "epoch": 0.5416666666666666, "grad_norm": 5.042497046029212, "learning_rate": 9.95571690734972e-06, "loss": 0.49428267, "memory(GiB)": 101.73, "step": 390, "train_speed(iter/s)": 0.339733 }, { "acc": 0.80402279, "epoch": 0.5486111111111112, "grad_norm": 4.123141501535671, "learning_rate": 9.95313937609955e-06, "loss": 0.54107056, "memory(GiB)": 101.73, "step": 395, "train_speed(iter/s)": 0.339838 }, { "acc": 0.81687107, "epoch": 0.5555555555555556, "grad_norm": 14.206970724335013, "learning_rate": 9.950489290871902e-06, "loss": 0.47818375, "memory(GiB)": 101.73, "step": 400, "train_speed(iter/s)": 0.339764 }, { "acc": 0.83043404, "epoch": 0.5625, "grad_norm": 13.479652065488521, "learning_rate": 9.947766690489351e-06, "loss": 0.45714722, "memory(GiB)": 101.73, "step": 405, "train_speed(iter/s)": 0.33976 }, { "acc": 0.82783804, "epoch": 0.5694444444444444, "grad_norm": 6.994687177716389, "learning_rate": 9.944971614836791e-06, "loss": 0.44925947, "memory(GiB)": 101.73, "step": 410, "train_speed(iter/s)": 0.340059 }, { "acc": 0.82454166, "epoch": 0.5763888888888888, "grad_norm": 6.474005259951495, "learning_rate": 9.942104104860843e-06, "loss": 0.46208277, "memory(GiB)": 101.73, "step": 415, "train_speed(iter/s)": 0.340149 }, { "acc": 0.82342644, "epoch": 0.5833333333333334, "grad_norm": 4.144488937559482, "learning_rate": 9.93916420256926e-06, "loss": 0.46902466, "memory(GiB)": 101.73, "step": 420, "train_speed(iter/s)": 0.340079 }, { "acc": 0.81452484, "epoch": 0.5902777777777778, "grad_norm": 4.534407651664684, "learning_rate": 9.936151951030313e-06, "loss": 0.47105713, "memory(GiB)": 101.73, "step": 425, "train_speed(iter/s)": 0.340222 }, { "acc": 0.81685143, "epoch": 0.5972222222222222, "grad_norm": 4.2541381378013545, "learning_rate": 9.933067394372155e-06, "loss": 0.45794678, "memory(GiB)": 101.73, "step": 430, "train_speed(iter/s)": 0.340113 }, { "acc": 0.82499523, "epoch": 0.6041666666666666, "grad_norm": 4.236882311093862, "learning_rate": 9.929910577782175e-06, "loss": 0.44803181, "memory(GiB)": 101.73, "step": 435, "train_speed(iter/s)": 0.340332 }, { "acc": 0.80982151, "epoch": 0.6111111111111112, "grad_norm": 4.6523720622403495, "learning_rate": 9.926681547506343e-06, "loss": 0.45672607, "memory(GiB)": 101.73, "step": 440, "train_speed(iter/s)": 0.340214 }, { "acc": 0.83155994, "epoch": 0.6180555555555556, "grad_norm": 3.539856352290064, "learning_rate": 9.923380350848525e-06, "loss": 0.43849535, "memory(GiB)": 101.73, "step": 445, "train_speed(iter/s)": 0.340238 }, { "acc": 0.82845173, "epoch": 0.625, "grad_norm": 4.628916992182938, "learning_rate": 9.920007036169793e-06, "loss": 0.44665036, "memory(GiB)": 101.73, "step": 450, "train_speed(iter/s)": 0.340465 }, { "acc": 0.827981, "epoch": 0.6319444444444444, "grad_norm": 2.7639289668011635, "learning_rate": 9.916561652887715e-06, "loss": 0.44553595, "memory(GiB)": 101.73, "step": 455, "train_speed(iter/s)": 0.340695 }, { "acc": 0.82418242, "epoch": 0.6388888888888888, "grad_norm": 2.5699138622519477, "learning_rate": 9.913044251475634e-06, "loss": 0.46772599, "memory(GiB)": 101.73, "step": 460, "train_speed(iter/s)": 0.34074 }, { "acc": 0.82081413, "epoch": 0.6458333333333334, "grad_norm": 3.819421823928973, "learning_rate": 9.909454883461921e-06, "loss": 0.4476263, "memory(GiB)": 101.73, "step": 465, "train_speed(iter/s)": 0.340745 }, { "acc": 0.81621161, "epoch": 0.6527777777777778, "grad_norm": 11.434255980801682, "learning_rate": 9.905793601429239e-06, "loss": 0.45358963, "memory(GiB)": 101.73, "step": 470, "train_speed(iter/s)": 0.340866 }, { "acc": 0.80989876, "epoch": 0.6597222222222222, "grad_norm": 3.3992412181651006, "learning_rate": 9.902060459013744e-06, "loss": 0.47856612, "memory(GiB)": 101.73, "step": 475, "train_speed(iter/s)": 0.341055 }, { "acc": 0.81908178, "epoch": 0.6666666666666666, "grad_norm": 4.325601425130281, "learning_rate": 9.898255510904326e-06, "loss": 0.44092093, "memory(GiB)": 101.73, "step": 480, "train_speed(iter/s)": 0.341185 }, { "acc": 0.82690163, "epoch": 0.6736111111111112, "grad_norm": 4.011942792509904, "learning_rate": 9.894378812841793e-06, "loss": 0.43541508, "memory(GiB)": 101.73, "step": 485, "train_speed(iter/s)": 0.341317 }, { "acc": 0.83012829, "epoch": 0.6805555555555556, "grad_norm": 11.20428097827798, "learning_rate": 9.890430421618054e-06, "loss": 0.45354671, "memory(GiB)": 101.73, "step": 490, "train_speed(iter/s)": 0.34136 }, { "acc": 0.81567917, "epoch": 0.6875, "grad_norm": 6.011176841797077, "learning_rate": 9.886410395075299e-06, "loss": 0.47600689, "memory(GiB)": 101.73, "step": 495, "train_speed(iter/s)": 0.34137 }, { "acc": 0.83390331, "epoch": 0.6944444444444444, "grad_norm": 2.5421269797009463, "learning_rate": 9.882318792105139e-06, "loss": 0.4320755, "memory(GiB)": 101.73, "step": 500, "train_speed(iter/s)": 0.341535 }, { "acc": 0.82440739, "epoch": 0.7013888888888888, "grad_norm": 5.398499470086722, "learning_rate": 9.878155672647745e-06, "loss": 0.45568981, "memory(GiB)": 101.73, "step": 505, "train_speed(iter/s)": 0.341835 }, { "acc": 0.82466736, "epoch": 0.7083333333333334, "grad_norm": 2.9253942613017605, "learning_rate": 9.873921097690983e-06, "loss": 0.44401636, "memory(GiB)": 101.73, "step": 510, "train_speed(iter/s)": 0.34167 }, { "acc": 0.83045635, "epoch": 0.7152777777777778, "grad_norm": 2.411332274219069, "learning_rate": 9.8696151292695e-06, "loss": 0.4418056, "memory(GiB)": 101.73, "step": 515, "train_speed(iter/s)": 0.341759 }, { "acc": 0.82254047, "epoch": 0.7222222222222222, "grad_norm": 5.941833912171894, "learning_rate": 9.86523783046383e-06, "loss": 0.45153627, "memory(GiB)": 101.73, "step": 520, "train_speed(iter/s)": 0.341971 }, { "acc": 0.82107563, "epoch": 0.7291666666666666, "grad_norm": 2.8671436433087694, "learning_rate": 9.860789265399467e-06, "loss": 0.44395323, "memory(GiB)": 101.73, "step": 525, "train_speed(iter/s)": 0.342045 }, { "acc": 0.8169796, "epoch": 0.7361111111111112, "grad_norm": 3.8842098252434423, "learning_rate": 9.856269499245921e-06, "loss": 0.45956945, "memory(GiB)": 101.73, "step": 530, "train_speed(iter/s)": 0.342256 }, { "acc": 0.83679924, "epoch": 0.7430555555555556, "grad_norm": 3.3239182012648243, "learning_rate": 9.85167859821577e-06, "loss": 0.43197575, "memory(GiB)": 101.73, "step": 535, "train_speed(iter/s)": 0.34245 }, { "acc": 0.82799988, "epoch": 0.75, "grad_norm": 5.816793260610907, "learning_rate": 9.847016629563683e-06, "loss": 0.42782841, "memory(GiB)": 101.73, "step": 540, "train_speed(iter/s)": 0.342663 }, { "acc": 0.81784801, "epoch": 0.7569444444444444, "grad_norm": 5.182686505017899, "learning_rate": 9.842283661585442e-06, "loss": 0.46612892, "memory(GiB)": 101.73, "step": 545, "train_speed(iter/s)": 0.342689 }, { "acc": 0.82886944, "epoch": 0.7638888888888888, "grad_norm": 2.5996284531124867, "learning_rate": 9.837479763616932e-06, "loss": 0.43436565, "memory(GiB)": 101.73, "step": 550, "train_speed(iter/s)": 0.342815 }, { "acc": 0.83412199, "epoch": 0.7708333333333334, "grad_norm": 3.4156004394784296, "learning_rate": 9.832605006033138e-06, "loss": 0.40990982, "memory(GiB)": 101.73, "step": 555, "train_speed(iter/s)": 0.342947 }, { "acc": 0.82610512, "epoch": 0.7777777777777778, "grad_norm": 2.6539226671509453, "learning_rate": 9.8276594602471e-06, "loss": 0.41491723, "memory(GiB)": 101.73, "step": 560, "train_speed(iter/s)": 0.342421 }, { "acc": 0.84313641, "epoch": 0.7847222222222222, "grad_norm": 5.876967933487755, "learning_rate": 9.822643198708884e-06, "loss": 0.41127276, "memory(GiB)": 101.73, "step": 565, "train_speed(iter/s)": 0.342369 }, { "acc": 0.83134546, "epoch": 0.7916666666666666, "grad_norm": 4.833766788934407, "learning_rate": 9.817556294904497e-06, "loss": 0.4230135, "memory(GiB)": 101.73, "step": 570, "train_speed(iter/s)": 0.342429 }, { "acc": 0.8203537, "epoch": 0.7986111111111112, "grad_norm": 3.95800471385031, "learning_rate": 9.812398823354835e-06, "loss": 0.45338583, "memory(GiB)": 101.73, "step": 575, "train_speed(iter/s)": 0.342453 }, { "acc": 0.84011059, "epoch": 0.8055555555555556, "grad_norm": 2.7356802434548295, "learning_rate": 9.807170859614574e-06, "loss": 0.41599312, "memory(GiB)": 101.73, "step": 580, "train_speed(iter/s)": 0.342508 }, { "acc": 0.82861423, "epoch": 0.8125, "grad_norm": 2.789741600661146, "learning_rate": 9.801872480271075e-06, "loss": 0.43512211, "memory(GiB)": 101.73, "step": 585, "train_speed(iter/s)": 0.342606 }, { "acc": 0.82745457, "epoch": 0.8194444444444444, "grad_norm": 5.351426683388521, "learning_rate": 9.796503762943248e-06, "loss": 0.42212791, "memory(GiB)": 101.73, "step": 590, "train_speed(iter/s)": 0.342504 }, { "acc": 0.83506317, "epoch": 0.8263888888888888, "grad_norm": 2.915645519453413, "learning_rate": 9.791064786280432e-06, "loss": 0.42344589, "memory(GiB)": 101.73, "step": 595, "train_speed(iter/s)": 0.342576 }, { "acc": 0.83208055, "epoch": 0.8333333333333334, "grad_norm": 3.5213580401660156, "learning_rate": 9.785555629961232e-06, "loss": 0.43331146, "memory(GiB)": 101.73, "step": 600, "train_speed(iter/s)": 0.342686 }, { "acc": 0.83166571, "epoch": 0.8402777777777778, "grad_norm": 4.361044463747454, "learning_rate": 9.779976374692353e-06, "loss": 0.45408182, "memory(GiB)": 101.73, "step": 605, "train_speed(iter/s)": 0.342638 }, { "acc": 0.82415371, "epoch": 0.8472222222222222, "grad_norm": 10.626940787117386, "learning_rate": 9.774327102207413e-06, "loss": 0.42070689, "memory(GiB)": 101.73, "step": 610, "train_speed(iter/s)": 0.342691 }, { "acc": 0.84025803, "epoch": 0.8541666666666666, "grad_norm": 2.662755839878228, "learning_rate": 9.76860789526576e-06, "loss": 0.4043293, "memory(GiB)": 101.73, "step": 615, "train_speed(iter/s)": 0.342709 }, { "acc": 0.83348093, "epoch": 0.8611111111111112, "grad_norm": 2.588929018249861, "learning_rate": 9.76281883765125e-06, "loss": 0.41577873, "memory(GiB)": 101.73, "step": 620, "train_speed(iter/s)": 0.342562 }, { "acc": 0.83077507, "epoch": 0.8680555555555556, "grad_norm": 4.12654917117445, "learning_rate": 9.756960014171012e-06, "loss": 0.43080907, "memory(GiB)": 101.73, "step": 625, "train_speed(iter/s)": 0.342671 }, { "acc": 0.827841, "epoch": 0.875, "grad_norm": 2.8606706635597354, "learning_rate": 9.751031510654226e-06, "loss": 0.43506193, "memory(GiB)": 101.73, "step": 630, "train_speed(iter/s)": 0.342731 }, { "acc": 0.83022375, "epoch": 0.8819444444444444, "grad_norm": 6.151953684994658, "learning_rate": 9.745033413950843e-06, "loss": 0.42204194, "memory(GiB)": 101.73, "step": 635, "train_speed(iter/s)": 0.342804 }, { "acc": 0.82415304, "epoch": 0.8888888888888888, "grad_norm": 3.2518580197951725, "learning_rate": 9.738965811930332e-06, "loss": 0.47604675, "memory(GiB)": 101.73, "step": 640, "train_speed(iter/s)": 0.342907 }, { "acc": 0.84281521, "epoch": 0.8958333333333334, "grad_norm": 2.603532205880358, "learning_rate": 9.732828793480376e-06, "loss": 0.41225729, "memory(GiB)": 101.73, "step": 645, "train_speed(iter/s)": 0.342817 }, { "acc": 0.83370028, "epoch": 0.9027777777777778, "grad_norm": 3.805039366288295, "learning_rate": 9.726622448505587e-06, "loss": 0.41593208, "memory(GiB)": 101.73, "step": 650, "train_speed(iter/s)": 0.342892 }, { "acc": 0.82898254, "epoch": 0.9097222222222222, "grad_norm": 3.402485563840372, "learning_rate": 9.720346867926172e-06, "loss": 0.43477345, "memory(GiB)": 101.73, "step": 655, "train_speed(iter/s)": 0.343066 }, { "acc": 0.85120811, "epoch": 0.9166666666666666, "grad_norm": 3.0253302648838534, "learning_rate": 9.714002143676614e-06, "loss": 0.38636012, "memory(GiB)": 101.73, "step": 660, "train_speed(iter/s)": 0.343136 }, { "acc": 0.83650703, "epoch": 0.9236111111111112, "grad_norm": 2.6221009587089696, "learning_rate": 9.707588368704318e-06, "loss": 0.40802522, "memory(GiB)": 101.73, "step": 665, "train_speed(iter/s)": 0.34312 }, { "acc": 0.84927387, "epoch": 0.9305555555555556, "grad_norm": 4.89731074876133, "learning_rate": 9.701105636968253e-06, "loss": 0.40131931, "memory(GiB)": 101.73, "step": 670, "train_speed(iter/s)": 0.343309 }, { "acc": 0.84856434, "epoch": 0.9375, "grad_norm": 5.437382145928593, "learning_rate": 9.69455404343757e-06, "loss": 0.40545149, "memory(GiB)": 101.73, "step": 675, "train_speed(iter/s)": 0.343307 }, { "acc": 0.83657379, "epoch": 0.9444444444444444, "grad_norm": 4.422858302197234, "learning_rate": 9.68793368409022e-06, "loss": 0.42424588, "memory(GiB)": 101.73, "step": 680, "train_speed(iter/s)": 0.343325 }, { "acc": 0.84768372, "epoch": 0.9513888888888888, "grad_norm": 4.425571264161584, "learning_rate": 9.681244655911542e-06, "loss": 0.41174183, "memory(GiB)": 101.73, "step": 685, "train_speed(iter/s)": 0.343398 }, { "acc": 0.84553738, "epoch": 0.9583333333333334, "grad_norm": 4.109087326995114, "learning_rate": 9.674487056892841e-06, "loss": 0.40106497, "memory(GiB)": 101.73, "step": 690, "train_speed(iter/s)": 0.343384 }, { "acc": 0.83130951, "epoch": 0.9652777777777778, "grad_norm": 4.058647388023358, "learning_rate": 9.667660986029956e-06, "loss": 0.43747225, "memory(GiB)": 101.73, "step": 695, "train_speed(iter/s)": 0.34334 }, { "acc": 0.82660599, "epoch": 0.9722222222222222, "grad_norm": 3.012549675607617, "learning_rate": 9.660766543321804e-06, "loss": 0.43917799, "memory(GiB)": 101.73, "step": 700, "train_speed(iter/s)": 0.34345 }, { "acc": 0.85258808, "epoch": 0.9791666666666666, "grad_norm": 2.317800113030072, "learning_rate": 9.653803829768927e-06, "loss": 0.37653742, "memory(GiB)": 101.73, "step": 705, "train_speed(iter/s)": 0.343541 }, { "acc": 0.84237957, "epoch": 0.9861111111111112, "grad_norm": 3.46822185449162, "learning_rate": 9.646772947371998e-06, "loss": 0.40804148, "memory(GiB)": 101.73, "step": 710, "train_speed(iter/s)": 0.343123 }, { "acc": 0.84040251, "epoch": 0.9930555555555556, "grad_norm": 2.7021645387276267, "learning_rate": 9.639673999130342e-06, "loss": 0.42324858, "memory(GiB)": 101.73, "step": 715, "train_speed(iter/s)": 0.343237 }, { "acc": 0.83597126, "epoch": 1.0, "grad_norm": 4.66166531475433, "learning_rate": 9.632507089040402e-06, "loss": 0.42435303, "memory(GiB)": 101.73, "step": 720, "train_speed(iter/s)": 0.343017 }, { "epoch": 1.0, "eval_acc": 0.8472242146033567, "eval_loss": 0.4218238890171051, "eval_runtime": 26.0678, "eval_samples_per_second": 13.158, "eval_steps_per_second": 1.65, "step": 720 }, { "acc": 0.84835815, "epoch": 1.0069444444444444, "grad_norm": 14.07543102393827, "learning_rate": 9.62527232209425e-06, "loss": 0.38356934, "memory(GiB)": 101.73, "step": 725, "train_speed(iter/s)": 0.332859 }, { "acc": 0.8468194, "epoch": 1.0138888888888888, "grad_norm": 4.1338785176242805, "learning_rate": 9.617969804278023e-06, "loss": 0.40347471, "memory(GiB)": 101.73, "step": 730, "train_speed(iter/s)": 0.332833 }, { "acc": 0.83594532, "epoch": 1.0208333333333333, "grad_norm": 6.53482454047143, "learning_rate": 9.610599642570378e-06, "loss": 0.41150756, "memory(GiB)": 101.73, "step": 735, "train_speed(iter/s)": 0.333063 }, { "acc": 0.83995285, "epoch": 1.0277777777777777, "grad_norm": 2.0874194148799954, "learning_rate": 9.603161944940925e-06, "loss": 0.40456595, "memory(GiB)": 101.73, "step": 740, "train_speed(iter/s)": 0.333087 }, { "acc": 0.84285097, "epoch": 1.0347222222222223, "grad_norm": 2.918185410713743, "learning_rate": 9.595656820348646e-06, "loss": 0.39343414, "memory(GiB)": 101.73, "step": 745, "train_speed(iter/s)": 0.333167 }, { "acc": 0.84659653, "epoch": 1.0416666666666667, "grad_norm": 3.3524326036503043, "learning_rate": 9.5880843787403e-06, "loss": 0.39015245, "memory(GiB)": 101.73, "step": 750, "train_speed(iter/s)": 0.333291 }, { "acc": 0.832335, "epoch": 1.0486111111111112, "grad_norm": 9.360098551984148, "learning_rate": 9.58044473104881e-06, "loss": 0.44525146, "memory(GiB)": 101.73, "step": 755, "train_speed(iter/s)": 0.33302 }, { "acc": 0.83618774, "epoch": 1.0555555555555556, "grad_norm": 7.009746416299999, "learning_rate": 9.572737989191634e-06, "loss": 0.42118731, "memory(GiB)": 101.73, "step": 760, "train_speed(iter/s)": 0.332604 }, { "acc": 0.85030022, "epoch": 1.0625, "grad_norm": 3.395766338969053, "learning_rate": 9.564964266069136e-06, "loss": 0.38766785, "memory(GiB)": 101.73, "step": 765, "train_speed(iter/s)": 0.332648 }, { "acc": 0.84878769, "epoch": 1.0694444444444444, "grad_norm": 4.049117510071738, "learning_rate": 9.557123675562923e-06, "loss": 0.39821975, "memory(GiB)": 101.73, "step": 770, "train_speed(iter/s)": 0.332717 }, { "acc": 0.8435914, "epoch": 1.0763888888888888, "grad_norm": 4.578339574266821, "learning_rate": 9.54921633253418e-06, "loss": 0.39716926, "memory(GiB)": 101.73, "step": 775, "train_speed(iter/s)": 0.332748 }, { "acc": 0.8482192, "epoch": 1.0833333333333333, "grad_norm": 4.848980267588248, "learning_rate": 9.541242352821985e-06, "loss": 0.40872383, "memory(GiB)": 101.73, "step": 780, "train_speed(iter/s)": 0.332671 }, { "acc": 0.84455881, "epoch": 1.0902777777777777, "grad_norm": 3.124037944279378, "learning_rate": 9.533201853241619e-06, "loss": 0.39232595, "memory(GiB)": 101.73, "step": 785, "train_speed(iter/s)": 0.332358 }, { "acc": 0.83510704, "epoch": 1.0972222222222223, "grad_norm": 2.8547950538457703, "learning_rate": 9.525094951582842e-06, "loss": 0.39581535, "memory(GiB)": 101.73, "step": 790, "train_speed(iter/s)": 0.332478 }, { "acc": 0.86455097, "epoch": 1.1041666666666667, "grad_norm": 2.5623476913367824, "learning_rate": 9.516921766608186e-06, "loss": 0.36006021, "memory(GiB)": 101.73, "step": 795, "train_speed(iter/s)": 0.332532 }, { "acc": 0.84903641, "epoch": 1.1111111111111112, "grad_norm": 2.3914228118955396, "learning_rate": 9.508682418051192e-06, "loss": 0.37949579, "memory(GiB)": 101.73, "step": 800, "train_speed(iter/s)": 0.332636 }, { "acc": 0.84685268, "epoch": 1.1180555555555556, "grad_norm": 4.498655619786845, "learning_rate": 9.500377026614675e-06, "loss": 0.39767621, "memory(GiB)": 101.73, "step": 805, "train_speed(iter/s)": 0.332765 }, { "acc": 0.84496756, "epoch": 1.125, "grad_norm": 3.7161374341586946, "learning_rate": 9.492005713968949e-06, "loss": 0.39117696, "memory(GiB)": 101.73, "step": 810, "train_speed(iter/s)": 0.332767 }, { "acc": 0.84031639, "epoch": 1.1319444444444444, "grad_norm": 3.868526550953586, "learning_rate": 9.483568602750044e-06, "loss": 0.40724792, "memory(GiB)": 101.73, "step": 815, "train_speed(iter/s)": 0.332759 }, { "acc": 0.84572315, "epoch": 1.1388888888888888, "grad_norm": 4.011321792664004, "learning_rate": 9.47506581655791e-06, "loss": 0.41854248, "memory(GiB)": 101.73, "step": 820, "train_speed(iter/s)": 0.332793 }, { "acc": 0.84330578, "epoch": 1.1458333333333333, "grad_norm": 2.841002351571773, "learning_rate": 9.466497479954604e-06, "loss": 0.41209116, "memory(GiB)": 101.73, "step": 825, "train_speed(iter/s)": 0.332879 }, { "acc": 0.85965681, "epoch": 1.1527777777777777, "grad_norm": 2.9466565908740807, "learning_rate": 9.457863718462472e-06, "loss": 0.35483818, "memory(GiB)": 101.73, "step": 830, "train_speed(iter/s)": 0.333018 }, { "acc": 0.84121952, "epoch": 1.1597222222222223, "grad_norm": 9.17125823782359, "learning_rate": 9.449164658562302e-06, "loss": 0.41710243, "memory(GiB)": 101.73, "step": 835, "train_speed(iter/s)": 0.333006 }, { "acc": 0.85266848, "epoch": 1.1666666666666667, "grad_norm": 4.337289079629721, "learning_rate": 9.440400427691474e-06, "loss": 0.38912995, "memory(GiB)": 101.73, "step": 840, "train_speed(iter/s)": 0.333157 }, { "acc": 0.85404491, "epoch": 1.1736111111111112, "grad_norm": 4.704538014865495, "learning_rate": 9.4315711542421e-06, "loss": 0.38565831, "memory(GiB)": 101.73, "step": 845, "train_speed(iter/s)": 0.33319 }, { "acc": 0.85194864, "epoch": 1.1805555555555556, "grad_norm": 7.104191972117062, "learning_rate": 9.422676967559129e-06, "loss": 0.38472157, "memory(GiB)": 101.73, "step": 850, "train_speed(iter/s)": 0.333352 }, { "acc": 0.84838581, "epoch": 1.1875, "grad_norm": 5.432606499558096, "learning_rate": 9.413717997938466e-06, "loss": 0.38313432, "memory(GiB)": 101.73, "step": 855, "train_speed(iter/s)": 0.333499 }, { "acc": 0.8506237, "epoch": 1.1944444444444444, "grad_norm": 4.500519797720098, "learning_rate": 9.404694376625057e-06, "loss": 0.37346766, "memory(GiB)": 101.73, "step": 860, "train_speed(iter/s)": 0.333789 }, { "acc": 0.8471118, "epoch": 1.2013888888888888, "grad_norm": 3.0647330027852053, "learning_rate": 9.395606235810962e-06, "loss": 0.3891892, "memory(GiB)": 101.73, "step": 865, "train_speed(iter/s)": 0.333964 }, { "acc": 0.84303493, "epoch": 1.2083333333333333, "grad_norm": 3.1169217151603488, "learning_rate": 9.386453708633429e-06, "loss": 0.38980885, "memory(GiB)": 101.73, "step": 870, "train_speed(iter/s)": 0.334173 }, { "acc": 0.83465099, "epoch": 1.2152777777777777, "grad_norm": 3.2398441704006395, "learning_rate": 9.377236929172933e-06, "loss": 0.42389526, "memory(GiB)": 101.73, "step": 875, "train_speed(iter/s)": 0.334394 }, { "acc": 0.85308475, "epoch": 1.2222222222222223, "grad_norm": 3.2099925668997313, "learning_rate": 9.36795603245122e-06, "loss": 0.35237427, "memory(GiB)": 101.73, "step": 880, "train_speed(iter/s)": 0.334616 }, { "acc": 0.84736195, "epoch": 1.2291666666666667, "grad_norm": 3.462100332631022, "learning_rate": 9.358611154429325e-06, "loss": 0.37894335, "memory(GiB)": 101.73, "step": 885, "train_speed(iter/s)": 0.334822 }, { "acc": 0.83803349, "epoch": 1.2361111111111112, "grad_norm": 11.39827473436759, "learning_rate": 9.349202432005577e-06, "loss": 0.43113008, "memory(GiB)": 101.73, "step": 890, "train_speed(iter/s)": 0.334963 }, { "acc": 0.84819775, "epoch": 1.2430555555555556, "grad_norm": 3.928967319430374, "learning_rate": 9.339730003013604e-06, "loss": 0.38550997, "memory(GiB)": 101.73, "step": 895, "train_speed(iter/s)": 0.335242 }, { "acc": 0.85249701, "epoch": 1.25, "grad_norm": 2.149347355481868, "learning_rate": 9.330194006220303e-06, "loss": 0.37847512, "memory(GiB)": 101.73, "step": 900, "train_speed(iter/s)": 0.33542 }, { "acc": 0.85284424, "epoch": 1.2569444444444444, "grad_norm": 2.354529081975499, "learning_rate": 9.320594581323808e-06, "loss": 0.41307311, "memory(GiB)": 101.73, "step": 905, "train_speed(iter/s)": 0.335631 }, { "acc": 0.85289736, "epoch": 1.2638888888888888, "grad_norm": 5.563333560479702, "learning_rate": 9.310931868951452e-06, "loss": 0.38515811, "memory(GiB)": 101.73, "step": 910, "train_speed(iter/s)": 0.335843 }, { "acc": 0.84706001, "epoch": 1.2708333333333333, "grad_norm": 2.7380569878551895, "learning_rate": 9.3012060106577e-06, "loss": 0.38339992, "memory(GiB)": 101.73, "step": 915, "train_speed(iter/s)": 0.336121 }, { "acc": 0.86188374, "epoch": 1.2777777777777777, "grad_norm": 4.874347425169184, "learning_rate": 9.291417148922079e-06, "loss": 0.38147278, "memory(GiB)": 101.73, "step": 920, "train_speed(iter/s)": 0.336324 }, { "acc": 0.8550766, "epoch": 1.2847222222222223, "grad_norm": 3.690731628108375, "learning_rate": 9.28156542714708e-06, "loss": 0.3768013, "memory(GiB)": 101.73, "step": 925, "train_speed(iter/s)": 0.336539 }, { "acc": 0.84941578, "epoch": 1.2916666666666667, "grad_norm": 5.416546502861204, "learning_rate": 9.271650989656078e-06, "loss": 0.38423877, "memory(GiB)": 101.73, "step": 930, "train_speed(iter/s)": 0.336758 }, { "acc": 0.83739882, "epoch": 1.2986111111111112, "grad_norm": 3.162933088324115, "learning_rate": 9.261673981691197e-06, "loss": 0.41248555, "memory(GiB)": 101.73, "step": 935, "train_speed(iter/s)": 0.336932 }, { "acc": 0.84417152, "epoch": 1.3055555555555556, "grad_norm": 3.5871262320118564, "learning_rate": 9.251634549411193e-06, "loss": 0.37667794, "memory(GiB)": 101.73, "step": 940, "train_speed(iter/s)": 0.336712 }, { "acc": 0.85208969, "epoch": 1.3125, "grad_norm": 3.743582531730351, "learning_rate": 9.24153283988931e-06, "loss": 0.39746375, "memory(GiB)": 101.73, "step": 945, "train_speed(iter/s)": 0.336799 }, { "acc": 0.83472309, "epoch": 1.3194444444444444, "grad_norm": 3.9063355350220172, "learning_rate": 9.23136900111113e-06, "loss": 0.42442837, "memory(GiB)": 101.73, "step": 950, "train_speed(iter/s)": 0.336919 }, { "acc": 0.83945732, "epoch": 1.3263888888888888, "grad_norm": 4.080061233293152, "learning_rate": 9.221143181972396e-06, "loss": 0.40141983, "memory(GiB)": 101.73, "step": 955, "train_speed(iter/s)": 0.337029 }, { "acc": 0.84612141, "epoch": 1.3333333333333333, "grad_norm": 3.2821130233749876, "learning_rate": 9.210855532276836e-06, "loss": 0.39341011, "memory(GiB)": 101.73, "step": 960, "train_speed(iter/s)": 0.337125 }, { "acc": 0.86109161, "epoch": 1.3402777777777777, "grad_norm": 1.6240104354039693, "learning_rate": 9.200506202733972e-06, "loss": 0.38223677, "memory(GiB)": 101.73, "step": 965, "train_speed(iter/s)": 0.337223 }, { "acc": 0.84814119, "epoch": 1.3472222222222223, "grad_norm": 35.878702050120395, "learning_rate": 9.190095344956909e-06, "loss": 0.40973086, "memory(GiB)": 101.73, "step": 970, "train_speed(iter/s)": 0.337269 }, { "acc": 0.84968128, "epoch": 1.3541666666666667, "grad_norm": 6.990269095516143, "learning_rate": 9.179623111460109e-06, "loss": 0.3797636, "memory(GiB)": 101.73, "step": 975, "train_speed(iter/s)": 0.337305 }, { "acc": 0.84383955, "epoch": 1.3611111111111112, "grad_norm": 32.14581208739412, "learning_rate": 9.169089655657162e-06, "loss": 0.38918655, "memory(GiB)": 101.73, "step": 980, "train_speed(iter/s)": 0.337382 }, { "acc": 0.85203476, "epoch": 1.3680555555555556, "grad_norm": 3.496050971602842, "learning_rate": 9.158495131858542e-06, "loss": 0.40532894, "memory(GiB)": 101.73, "step": 985, "train_speed(iter/s)": 0.3374 }, { "acc": 0.85004854, "epoch": 1.375, "grad_norm": 3.8694772100362327, "learning_rate": 9.147839695269337e-06, "loss": 0.39978607, "memory(GiB)": 101.73, "step": 990, "train_speed(iter/s)": 0.337439 }, { "acc": 0.8479394, "epoch": 1.3819444444444444, "grad_norm": 6.070730712147665, "learning_rate": 9.137123501986982e-06, "loss": 0.39016724, "memory(GiB)": 101.73, "step": 995, "train_speed(iter/s)": 0.337428 }, { "acc": 0.85400057, "epoch": 1.3888888888888888, "grad_norm": 1.7024422206195924, "learning_rate": 9.126346708998974e-06, "loss": 0.37917585, "memory(GiB)": 101.73, "step": 1000, "train_speed(iter/s)": 0.337377 }, { "acc": 0.84060793, "epoch": 1.3958333333333333, "grad_norm": 5.552604074545073, "learning_rate": 9.115509474180566e-06, "loss": 0.39503174, "memory(GiB)": 101.73, "step": 1005, "train_speed(iter/s)": 0.337516 }, { "acc": 0.85068932, "epoch": 1.4027777777777777, "grad_norm": 3.0259426862916965, "learning_rate": 9.104611956292457e-06, "loss": 0.38301041, "memory(GiB)": 101.73, "step": 1010, "train_speed(iter/s)": 0.337602 }, { "acc": 0.8504097, "epoch": 1.4097222222222223, "grad_norm": 3.5344707343383868, "learning_rate": 9.093654314978463e-06, "loss": 0.37292523, "memory(GiB)": 101.73, "step": 1015, "train_speed(iter/s)": 0.337474 }, { "acc": 0.85743666, "epoch": 1.4166666666666667, "grad_norm": 2.630949673383209, "learning_rate": 9.08263671076319e-06, "loss": 0.36461799, "memory(GiB)": 101.73, "step": 1020, "train_speed(iter/s)": 0.337425 }, { "acc": 0.86220655, "epoch": 1.4236111111111112, "grad_norm": 3.9855410550591905, "learning_rate": 9.071559305049667e-06, "loss": 0.35792432, "memory(GiB)": 101.73, "step": 1025, "train_speed(iter/s)": 0.337411 }, { "acc": 0.84792109, "epoch": 1.4305555555555556, "grad_norm": 3.7922263544393404, "learning_rate": 9.060422260116992e-06, "loss": 0.38916125, "memory(GiB)": 101.73, "step": 1030, "train_speed(iter/s)": 0.337365 }, { "acc": 0.85099182, "epoch": 1.4375, "grad_norm": 3.70846293904974, "learning_rate": 9.049225739117948e-06, "loss": 0.36477528, "memory(GiB)": 101.73, "step": 1035, "train_speed(iter/s)": 0.33741 }, { "acc": 0.86404428, "epoch": 1.4444444444444444, "grad_norm": 2.37261103672829, "learning_rate": 9.03796990607662e-06, "loss": 0.34631798, "memory(GiB)": 101.73, "step": 1040, "train_speed(iter/s)": 0.337456 }, { "acc": 0.86004276, "epoch": 1.4513888888888888, "grad_norm": 2.908735431664467, "learning_rate": 9.026654925885986e-06, "loss": 0.36338158, "memory(GiB)": 101.73, "step": 1045, "train_speed(iter/s)": 0.337461 }, { "acc": 0.85583038, "epoch": 1.4583333333333333, "grad_norm": 4.465703141418222, "learning_rate": 9.015280964305504e-06, "loss": 0.34531968, "memory(GiB)": 101.73, "step": 1050, "train_speed(iter/s)": 0.337494 }, { "acc": 0.85433445, "epoch": 1.4652777777777777, "grad_norm": 3.770405959716388, "learning_rate": 9.003848187958681e-06, "loss": 0.38607841, "memory(GiB)": 101.73, "step": 1055, "train_speed(iter/s)": 0.337509 }, { "acc": 0.858428, "epoch": 1.4722222222222223, "grad_norm": 3.379640478027579, "learning_rate": 8.99235676433064e-06, "loss": 0.35510893, "memory(GiB)": 101.73, "step": 1060, "train_speed(iter/s)": 0.337503 }, { "acc": 0.85026894, "epoch": 1.4791666666666667, "grad_norm": 2.2957543163398397, "learning_rate": 8.980806861765652e-06, "loss": 0.37486589, "memory(GiB)": 101.73, "step": 1065, "train_speed(iter/s)": 0.337572 }, { "acc": 0.85720415, "epoch": 1.4861111111111112, "grad_norm": 3.2746297247572738, "learning_rate": 8.969198649464691e-06, "loss": 0.37236695, "memory(GiB)": 101.73, "step": 1070, "train_speed(iter/s)": 0.337652 }, { "acc": 0.84375439, "epoch": 1.4930555555555556, "grad_norm": 5.001943381636829, "learning_rate": 8.95753229748293e-06, "loss": 0.39222441, "memory(GiB)": 101.73, "step": 1075, "train_speed(iter/s)": 0.337681 }, { "acc": 0.84540091, "epoch": 1.5, "grad_norm": 6.181217782470258, "learning_rate": 8.94580797672727e-06, "loss": 0.39223666, "memory(GiB)": 101.73, "step": 1080, "train_speed(iter/s)": 0.337607 }, { "acc": 0.85473719, "epoch": 1.5069444444444444, "grad_norm": 2.678403983509847, "learning_rate": 8.934025858953828e-06, "loss": 0.3707407, "memory(GiB)": 101.73, "step": 1085, "train_speed(iter/s)": 0.337461 }, { "acc": 0.86338501, "epoch": 1.5138888888888888, "grad_norm": 3.9439070272171293, "learning_rate": 8.92218611676542e-06, "loss": 0.35309997, "memory(GiB)": 101.73, "step": 1090, "train_speed(iter/s)": 0.337552 }, { "acc": 0.84891472, "epoch": 1.5208333333333335, "grad_norm": 5.46194063471783, "learning_rate": 8.910288923609034e-06, "loss": 0.39501739, "memory(GiB)": 101.73, "step": 1095, "train_speed(iter/s)": 0.337604 }, { "acc": 0.85439892, "epoch": 1.5277777777777777, "grad_norm": 2.6521402168678265, "learning_rate": 8.898334453773292e-06, "loss": 0.37683649, "memory(GiB)": 101.73, "step": 1100, "train_speed(iter/s)": 0.337666 }, { "acc": 0.85185204, "epoch": 1.5347222222222223, "grad_norm": 2.635454666402325, "learning_rate": 8.886322882385894e-06, "loss": 0.37046089, "memory(GiB)": 101.73, "step": 1105, "train_speed(iter/s)": 0.337711 }, { "acc": 0.85055122, "epoch": 1.5416666666666665, "grad_norm": 3.3497933627360177, "learning_rate": 8.874254385411048e-06, "loss": 0.37285306, "memory(GiB)": 101.73, "step": 1110, "train_speed(iter/s)": 0.3377 }, { "acc": 0.84177113, "epoch": 1.5486111111111112, "grad_norm": 2.6691808877766103, "learning_rate": 8.8621291396469e-06, "loss": 0.39528844, "memory(GiB)": 101.73, "step": 1115, "train_speed(iter/s)": 0.337819 }, { "acc": 0.85968094, "epoch": 1.5555555555555556, "grad_norm": 3.9807492163771157, "learning_rate": 8.849947322722941e-06, "loss": 0.37127504, "memory(GiB)": 101.73, "step": 1120, "train_speed(iter/s)": 0.337813 }, { "acc": 0.8640234, "epoch": 1.5625, "grad_norm": 4.904076026536717, "learning_rate": 8.837709113097406e-06, "loss": 0.3620749, "memory(GiB)": 101.73, "step": 1125, "train_speed(iter/s)": 0.337775 }, { "acc": 0.85505867, "epoch": 1.5694444444444444, "grad_norm": 1.7650593378125794, "learning_rate": 8.825414690054652e-06, "loss": 0.37128978, "memory(GiB)": 101.73, "step": 1130, "train_speed(iter/s)": 0.337724 }, { "acc": 0.84300156, "epoch": 1.5763888888888888, "grad_norm": 6.075752458033056, "learning_rate": 8.813064233702543e-06, "loss": 0.39115798, "memory(GiB)": 101.73, "step": 1135, "train_speed(iter/s)": 0.337764 }, { "acc": 0.84721622, "epoch": 1.5833333333333335, "grad_norm": 5.222470197877979, "learning_rate": 8.800657924969805e-06, "loss": 0.37910523, "memory(GiB)": 101.73, "step": 1140, "train_speed(iter/s)": 0.33781 }, { "acc": 0.85609684, "epoch": 1.5902777777777777, "grad_norm": 2.89223156856068, "learning_rate": 8.788195945603379e-06, "loss": 0.37958903, "memory(GiB)": 101.73, "step": 1145, "train_speed(iter/s)": 0.337714 }, { "acc": 0.85490465, "epoch": 1.5972222222222223, "grad_norm": 2.104304085343702, "learning_rate": 8.775678478165751e-06, "loss": 0.36199951, "memory(GiB)": 101.73, "step": 1150, "train_speed(iter/s)": 0.3376 }, { "acc": 0.86297607, "epoch": 1.6041666666666665, "grad_norm": 3.9067520290967486, "learning_rate": 8.763105706032287e-06, "loss": 0.33438387, "memory(GiB)": 101.73, "step": 1155, "train_speed(iter/s)": 0.337662 }, { "acc": 0.85664263, "epoch": 1.6111111111111112, "grad_norm": 56.10959937314398, "learning_rate": 8.750477813388537e-06, "loss": 0.3823755, "memory(GiB)": 101.73, "step": 1160, "train_speed(iter/s)": 0.337668 }, { "acc": 0.84644604, "epoch": 1.6180555555555556, "grad_norm": 3.1303485574935124, "learning_rate": 8.737794985227552e-06, "loss": 0.37934666, "memory(GiB)": 101.73, "step": 1165, "train_speed(iter/s)": 0.337782 }, { "acc": 0.86156435, "epoch": 1.625, "grad_norm": 3.779291654725943, "learning_rate": 8.725057407347151e-06, "loss": 0.34583051, "memory(GiB)": 101.73, "step": 1170, "train_speed(iter/s)": 0.337815 }, { "acc": 0.86057777, "epoch": 1.6319444444444444, "grad_norm": 6.305331333525041, "learning_rate": 8.712265266347225e-06, "loss": 0.3509572, "memory(GiB)": 101.73, "step": 1175, "train_speed(iter/s)": 0.337875 }, { "acc": 0.84179935, "epoch": 1.6388888888888888, "grad_norm": 3.534057697831123, "learning_rate": 8.699418749626983e-06, "loss": 0.38306279, "memory(GiB)": 101.73, "step": 1180, "train_speed(iter/s)": 0.337986 }, { "acc": 0.86022358, "epoch": 1.6458333333333335, "grad_norm": 3.596616194421774, "learning_rate": 8.686518045382216e-06, "loss": 0.35667768, "memory(GiB)": 101.73, "step": 1185, "train_speed(iter/s)": 0.337903 }, { "acc": 0.84735928, "epoch": 1.6527777777777777, "grad_norm": 2.8042991260508248, "learning_rate": 8.673563342602538e-06, "loss": 0.37387271, "memory(GiB)": 101.73, "step": 1190, "train_speed(iter/s)": 0.337983 }, { "acc": 0.85692234, "epoch": 1.6597222222222223, "grad_norm": 2.7613649008585712, "learning_rate": 8.660554831068615e-06, "loss": 0.3682164, "memory(GiB)": 101.73, "step": 1195, "train_speed(iter/s)": 0.33803 }, { "acc": 0.850665, "epoch": 1.6666666666666665, "grad_norm": 6.354642055126395, "learning_rate": 8.647492701349395e-06, "loss": 0.36811004, "memory(GiB)": 101.73, "step": 1200, "train_speed(iter/s)": 0.338156 }, { "acc": 0.84379854, "epoch": 1.6736111111111112, "grad_norm": 3.1757762546543247, "learning_rate": 8.634377144799303e-06, "loss": 0.39622905, "memory(GiB)": 101.73, "step": 1205, "train_speed(iter/s)": 0.338155 }, { "acc": 0.8438199, "epoch": 1.6805555555555556, "grad_norm": 2.9686876574121994, "learning_rate": 8.621208353555442e-06, "loss": 0.37581418, "memory(GiB)": 101.73, "step": 1210, "train_speed(iter/s)": 0.338178 }, { "acc": 0.85528498, "epoch": 1.6875, "grad_norm": 2.5489253707665847, "learning_rate": 8.607986520534785e-06, "loss": 0.37263567, "memory(GiB)": 101.73, "step": 1215, "train_speed(iter/s)": 0.338207 }, { "acc": 0.85820246, "epoch": 1.6944444444444444, "grad_norm": 4.887891679393261, "learning_rate": 8.594711839431341e-06, "loss": 0.37658699, "memory(GiB)": 101.73, "step": 1220, "train_speed(iter/s)": 0.338112 }, { "acc": 0.85092402, "epoch": 1.7013888888888888, "grad_norm": 2.360910524678318, "learning_rate": 8.581384504713316e-06, "loss": 0.36587248, "memory(GiB)": 101.73, "step": 1225, "train_speed(iter/s)": 0.338187 }, { "acc": 0.85391512, "epoch": 1.7083333333333335, "grad_norm": 2.463848635972625, "learning_rate": 8.568004711620276e-06, "loss": 0.37998319, "memory(GiB)": 101.73, "step": 1230, "train_speed(iter/s)": 0.338201 }, { "acc": 0.86517754, "epoch": 1.7152777777777777, "grad_norm": 2.939085237700951, "learning_rate": 8.554572656160276e-06, "loss": 0.34979258, "memory(GiB)": 101.73, "step": 1235, "train_speed(iter/s)": 0.338314 }, { "acc": 0.86364365, "epoch": 1.7222222222222223, "grad_norm": 3.3111152008785436, "learning_rate": 8.541088535106987e-06, "loss": 0.35193083, "memory(GiB)": 101.73, "step": 1240, "train_speed(iter/s)": 0.338441 }, { "acc": 0.85455751, "epoch": 1.7291666666666665, "grad_norm": 2.8670527839907876, "learning_rate": 8.527552545996823e-06, "loss": 0.37321618, "memory(GiB)": 101.73, "step": 1245, "train_speed(iter/s)": 0.33858 }, { "acc": 0.86149797, "epoch": 1.7361111111111112, "grad_norm": 6.333170412320581, "learning_rate": 8.513964887126042e-06, "loss": 0.34765677, "memory(GiB)": 101.73, "step": 1250, "train_speed(iter/s)": 0.338722 }, { "acc": 0.86266346, "epoch": 1.7430555555555556, "grad_norm": 6.15251081063916, "learning_rate": 8.500325757547837e-06, "loss": 0.35451782, "memory(GiB)": 101.73, "step": 1255, "train_speed(iter/s)": 0.338819 }, { "acc": 0.86682644, "epoch": 1.75, "grad_norm": 3.3602407157735326, "learning_rate": 8.486635357069431e-06, "loss": 0.35656066, "memory(GiB)": 101.73, "step": 1260, "train_speed(iter/s)": 0.338924 }, { "acc": 0.87807121, "epoch": 1.7569444444444444, "grad_norm": 6.308557312315135, "learning_rate": 8.472893886249137e-06, "loss": 0.32822404, "memory(GiB)": 101.73, "step": 1265, "train_speed(iter/s)": 0.338995 }, { "acc": 0.8561429, "epoch": 1.7638888888888888, "grad_norm": 18.368177846975684, "learning_rate": 8.459101546393425e-06, "loss": 0.37018564, "memory(GiB)": 101.73, "step": 1270, "train_speed(iter/s)": 0.339068 }, { "acc": 0.85693264, "epoch": 1.7708333333333335, "grad_norm": 4.368985948399491, "learning_rate": 8.44525853955398e-06, "loss": 0.36033721, "memory(GiB)": 101.73, "step": 1275, "train_speed(iter/s)": 0.339157 }, { "acc": 0.85860252, "epoch": 1.7777777777777777, "grad_norm": 6.537391483785996, "learning_rate": 8.431365068524735e-06, "loss": 0.34542758, "memory(GiB)": 101.73, "step": 1280, "train_speed(iter/s)": 0.339218 }, { "acc": 0.86011286, "epoch": 1.7847222222222223, "grad_norm": 3.6615393374801717, "learning_rate": 8.417421336838895e-06, "loss": 0.35445812, "memory(GiB)": 101.73, "step": 1285, "train_speed(iter/s)": 0.339293 }, { "acc": 0.86690083, "epoch": 1.7916666666666665, "grad_norm": 9.494908561151547, "learning_rate": 8.403427548765964e-06, "loss": 0.35471287, "memory(GiB)": 101.73, "step": 1290, "train_speed(iter/s)": 0.339365 }, { "acc": 0.86261654, "epoch": 1.7986111111111112, "grad_norm": 4.086474959822119, "learning_rate": 8.389383909308754e-06, "loss": 0.36231318, "memory(GiB)": 101.73, "step": 1295, "train_speed(iter/s)": 0.339445 }, { "acc": 0.86633072, "epoch": 1.8055555555555556, "grad_norm": 3.1457008532368587, "learning_rate": 8.375290624200375e-06, "loss": 0.33784354, "memory(GiB)": 101.73, "step": 1300, "train_speed(iter/s)": 0.339499 }, { "acc": 0.85676422, "epoch": 1.8125, "grad_norm": 3.0219513188022797, "learning_rate": 8.361147899901222e-06, "loss": 0.35047636, "memory(GiB)": 101.73, "step": 1305, "train_speed(iter/s)": 0.339614 }, { "acc": 0.87211819, "epoch": 1.8194444444444444, "grad_norm": 2.162908540556724, "learning_rate": 8.346955943595957e-06, "loss": 0.34178371, "memory(GiB)": 101.73, "step": 1310, "train_speed(iter/s)": 0.339668 }, { "acc": 0.86086178, "epoch": 1.8263888888888888, "grad_norm": 5.106198424264787, "learning_rate": 8.332714963190462e-06, "loss": 0.36008306, "memory(GiB)": 101.73, "step": 1315, "train_speed(iter/s)": 0.339762 }, { "acc": 0.86688232, "epoch": 1.8333333333333335, "grad_norm": 5.20553175585457, "learning_rate": 8.318425167308806e-06, "loss": 0.34904175, "memory(GiB)": 101.73, "step": 1320, "train_speed(iter/s)": 0.339817 }, { "acc": 0.86953564, "epoch": 1.8402777777777777, "grad_norm": 1.9169409393919037, "learning_rate": 8.304086765290184e-06, "loss": 0.32462049, "memory(GiB)": 101.73, "step": 1325, "train_speed(iter/s)": 0.339894 }, { "acc": 0.85959816, "epoch": 1.8472222222222223, "grad_norm": 2.919120395900884, "learning_rate": 8.289699967185843e-06, "loss": 0.34003651, "memory(GiB)": 101.73, "step": 1330, "train_speed(iter/s)": 0.339982 }, { "acc": 0.86576004, "epoch": 1.8541666666666665, "grad_norm": 3.339716130934134, "learning_rate": 8.27526498375602e-06, "loss": 0.34242401, "memory(GiB)": 101.73, "step": 1335, "train_speed(iter/s)": 0.340129 }, { "acc": 0.8572998, "epoch": 1.8611111111111112, "grad_norm": 3.4253838339731804, "learning_rate": 8.260782026466838e-06, "loss": 0.36056826, "memory(GiB)": 101.73, "step": 1340, "train_speed(iter/s)": 0.340256 }, { "acc": 0.85873718, "epoch": 1.8680555555555556, "grad_norm": 3.329467071510601, "learning_rate": 8.24625130748722e-06, "loss": 0.34768081, "memory(GiB)": 101.73, "step": 1345, "train_speed(iter/s)": 0.340386 }, { "acc": 0.85568542, "epoch": 1.875, "grad_norm": 3.4184372446872815, "learning_rate": 8.231673039685774e-06, "loss": 0.37388916, "memory(GiB)": 101.73, "step": 1350, "train_speed(iter/s)": 0.340521 }, { "acc": 0.85642204, "epoch": 1.8819444444444444, "grad_norm": 2.461666257667224, "learning_rate": 8.217047436627683e-06, "loss": 0.35878954, "memory(GiB)": 101.73, "step": 1355, "train_speed(iter/s)": 0.340639 }, { "acc": 0.86499329, "epoch": 1.8888888888888888, "grad_norm": 4.259794208029547, "learning_rate": 8.20237471257156e-06, "loss": 0.33968384, "memory(GiB)": 101.73, "step": 1360, "train_speed(iter/s)": 0.340762 }, { "acc": 0.85926418, "epoch": 1.8958333333333335, "grad_norm": 2.3889340744047813, "learning_rate": 8.18765508246633e-06, "loss": 0.34691594, "memory(GiB)": 101.73, "step": 1365, "train_speed(iter/s)": 0.340846 }, { "acc": 0.86262779, "epoch": 1.9027777777777777, "grad_norm": 3.6312299823676577, "learning_rate": 8.172888761948066e-06, "loss": 0.34867449, "memory(GiB)": 101.73, "step": 1370, "train_speed(iter/s)": 0.340903 }, { "acc": 0.86736736, "epoch": 1.9097222222222223, "grad_norm": 4.0784838300339725, "learning_rate": 8.158075967336838e-06, "loss": 0.35361021, "memory(GiB)": 101.73, "step": 1375, "train_speed(iter/s)": 0.340966 }, { "acc": 0.86585445, "epoch": 1.9166666666666665, "grad_norm": 2.758741475011602, "learning_rate": 8.143216915633535e-06, "loss": 0.34491754, "memory(GiB)": 101.73, "step": 1380, "train_speed(iter/s)": 0.341008 }, { "acc": 0.85726538, "epoch": 1.9236111111111112, "grad_norm": 3.5673127167589462, "learning_rate": 8.1283118245167e-06, "loss": 0.36892872, "memory(GiB)": 101.73, "step": 1385, "train_speed(iter/s)": 0.341079 }, { "acc": 0.8536458, "epoch": 1.9305555555555556, "grad_norm": 3.1994250752022806, "learning_rate": 8.113360912339326e-06, "loss": 0.36794338, "memory(GiB)": 101.73, "step": 1390, "train_speed(iter/s)": 0.341193 }, { "acc": 0.87299528, "epoch": 1.9375, "grad_norm": 5.391090918639991, "learning_rate": 8.09836439812567e-06, "loss": 0.33371003, "memory(GiB)": 101.73, "step": 1395, "train_speed(iter/s)": 0.341264 }, { "acc": 0.86591825, "epoch": 1.9444444444444444, "grad_norm": 3.3763410094981285, "learning_rate": 8.083322501568032e-06, "loss": 0.34277546, "memory(GiB)": 101.73, "step": 1400, "train_speed(iter/s)": 0.341332 }, { "acc": 0.86980515, "epoch": 1.9513888888888888, "grad_norm": 2.7363409438146022, "learning_rate": 8.06823544302355e-06, "loss": 0.32296598, "memory(GiB)": 101.73, "step": 1405, "train_speed(iter/s)": 0.341398 }, { "acc": 0.86319895, "epoch": 1.9583333333333335, "grad_norm": 3.0147390689828484, "learning_rate": 8.053103443510962e-06, "loss": 0.33763702, "memory(GiB)": 101.73, "step": 1410, "train_speed(iter/s)": 0.341474 }, { "acc": 0.8724597, "epoch": 1.9652777777777777, "grad_norm": 3.6283792947150166, "learning_rate": 8.037926724707367e-06, "loss": 0.32401137, "memory(GiB)": 101.73, "step": 1415, "train_speed(iter/s)": 0.341572 }, { "acc": 0.87336702, "epoch": 1.9722222222222223, "grad_norm": 3.231102866648495, "learning_rate": 8.022705508944994e-06, "loss": 0.32258878, "memory(GiB)": 101.73, "step": 1420, "train_speed(iter/s)": 0.341648 }, { "acc": 0.87259836, "epoch": 1.9791666666666665, "grad_norm": 4.190611839214796, "learning_rate": 8.007440019207919e-06, "loss": 0.33295143, "memory(GiB)": 101.73, "step": 1425, "train_speed(iter/s)": 0.341692 }, { "acc": 0.86225185, "epoch": 1.9861111111111112, "grad_norm": 3.999782034857841, "learning_rate": 7.992130479128823e-06, "loss": 0.33923955, "memory(GiB)": 101.73, "step": 1430, "train_speed(iter/s)": 0.34176 }, { "acc": 0.88750267, "epoch": 1.9930555555555556, "grad_norm": 3.0649487093465306, "learning_rate": 7.976777112985696e-06, "loss": 0.29414239, "memory(GiB)": 101.73, "step": 1435, "train_speed(iter/s)": 0.341782 }, { "acc": 0.86065731, "epoch": 2.0, "grad_norm": 4.492541294620969, "learning_rate": 7.96138014569857e-06, "loss": 0.35968423, "memory(GiB)": 101.73, "step": 1440, "train_speed(iter/s)": 0.34183 }, { "epoch": 2.0, "eval_acc": 0.8741930856405107, "eval_loss": 0.32394787669181824, "eval_runtime": 25.9731, "eval_samples_per_second": 13.206, "eval_steps_per_second": 1.656, "step": 1440 }, { "acc": 0.87566061, "epoch": 2.0069444444444446, "grad_norm": 3.2988239118803357, "learning_rate": 7.945939802826204e-06, "loss": 0.33297586, "memory(GiB)": 101.73, "step": 1445, "train_speed(iter/s)": 0.336817 }, { "acc": 0.87367706, "epoch": 2.013888888888889, "grad_norm": 3.9864890899544867, "learning_rate": 7.930456310562798e-06, "loss": 0.32236509, "memory(GiB)": 101.73, "step": 1450, "train_speed(iter/s)": 0.336936 }, { "acc": 0.88408442, "epoch": 2.0208333333333335, "grad_norm": 3.142019377538419, "learning_rate": 7.914929895734667e-06, "loss": 0.30712235, "memory(GiB)": 101.73, "step": 1455, "train_speed(iter/s)": 0.337046 }, { "acc": 0.87545471, "epoch": 2.0277777777777777, "grad_norm": 21.579465515202426, "learning_rate": 7.899360785796927e-06, "loss": 0.31951995, "memory(GiB)": 101.73, "step": 1460, "train_speed(iter/s)": 0.337169 }, { "acc": 0.869561, "epoch": 2.0347222222222223, "grad_norm": 3.2772620999715767, "learning_rate": 7.883749208830157e-06, "loss": 0.32944577, "memory(GiB)": 101.73, "step": 1465, "train_speed(iter/s)": 0.337198 }, { "acc": 0.86731529, "epoch": 2.0416666666666665, "grad_norm": 5.482996431081107, "learning_rate": 7.868095393537055e-06, "loss": 0.33758581, "memory(GiB)": 101.73, "step": 1470, "train_speed(iter/s)": 0.337249 }, { "acc": 0.87085705, "epoch": 2.048611111111111, "grad_norm": 3.542698342334349, "learning_rate": 7.852399569239099e-06, "loss": 0.32252483, "memory(GiB)": 101.73, "step": 1475, "train_speed(iter/s)": 0.337307 }, { "acc": 0.86159992, "epoch": 2.0555555555555554, "grad_norm": 2.853821845059801, "learning_rate": 7.836661965873173e-06, "loss": 0.34277472, "memory(GiB)": 101.73, "step": 1480, "train_speed(iter/s)": 0.337372 }, { "acc": 0.86977177, "epoch": 2.0625, "grad_norm": 2.335202080690499, "learning_rate": 7.820882813988209e-06, "loss": 0.35118761, "memory(GiB)": 101.73, "step": 1485, "train_speed(iter/s)": 0.337422 }, { "acc": 0.8722353, "epoch": 2.0694444444444446, "grad_norm": 3.1769141565351586, "learning_rate": 7.805062344741807e-06, "loss": 0.32959652, "memory(GiB)": 101.73, "step": 1490, "train_speed(iter/s)": 0.337452 }, { "acc": 0.87156506, "epoch": 2.076388888888889, "grad_norm": 5.028143683249058, "learning_rate": 7.789200789896853e-06, "loss": 0.33031771, "memory(GiB)": 101.73, "step": 1495, "train_speed(iter/s)": 0.337511 }, { "acc": 0.86954565, "epoch": 2.0833333333333335, "grad_norm": 3.7873337590539076, "learning_rate": 7.773298381818106e-06, "loss": 0.32032595, "memory(GiB)": 101.73, "step": 1500, "train_speed(iter/s)": 0.337583 }, { "acc": 0.87048397, "epoch": 2.0902777777777777, "grad_norm": 2.9980547146112047, "learning_rate": 7.757355353468819e-06, "loss": 0.32236178, "memory(GiB)": 101.73, "step": 1505, "train_speed(iter/s)": 0.337636 }, { "acc": 0.85826788, "epoch": 2.0972222222222223, "grad_norm": 4.2342096618271166, "learning_rate": 7.74137193840731e-06, "loss": 0.33629158, "memory(GiB)": 101.73, "step": 1510, "train_speed(iter/s)": 0.337623 }, { "acc": 0.86459484, "epoch": 2.1041666666666665, "grad_norm": 3.477001853963199, "learning_rate": 7.725348370783541e-06, "loss": 0.33757076, "memory(GiB)": 101.73, "step": 1515, "train_speed(iter/s)": 0.337658 }, { "acc": 0.86884823, "epoch": 2.111111111111111, "grad_norm": 4.148406829850161, "learning_rate": 7.709284885335696e-06, "loss": 0.34878542, "memory(GiB)": 101.73, "step": 1520, "train_speed(iter/s)": 0.337642 }, { "acc": 0.88847275, "epoch": 2.1180555555555554, "grad_norm": 5.21167919519557, "learning_rate": 7.693181717386736e-06, "loss": 0.31671143, "memory(GiB)": 101.73, "step": 1525, "train_speed(iter/s)": 0.33774 }, { "acc": 0.86802654, "epoch": 2.125, "grad_norm": 6.351118021037492, "learning_rate": 7.677039102840951e-06, "loss": 0.34130049, "memory(GiB)": 101.73, "step": 1530, "train_speed(iter/s)": 0.337771 }, { "acc": 0.87035189, "epoch": 2.1319444444444446, "grad_norm": 3.1462563267386203, "learning_rate": 7.66085727818051e-06, "loss": 0.32847002, "memory(GiB)": 101.73, "step": 1535, "train_speed(iter/s)": 0.337739 }, { "acc": 0.87579575, "epoch": 2.138888888888889, "grad_norm": 7.036404367525274, "learning_rate": 7.644636480461992e-06, "loss": 0.33579338, "memory(GiB)": 101.73, "step": 1540, "train_speed(iter/s)": 0.337808 }, { "acc": 0.86966209, "epoch": 2.1458333333333335, "grad_norm": 2.8028767779169232, "learning_rate": 7.62837694731291e-06, "loss": 0.34096689, "memory(GiB)": 101.73, "step": 1545, "train_speed(iter/s)": 0.337722 }, { "acc": 0.86123219, "epoch": 2.1527777777777777, "grad_norm": 4.916942622404778, "learning_rate": 7.612078916928237e-06, "loss": 0.34751384, "memory(GiB)": 101.73, "step": 1550, "train_speed(iter/s)": 0.337803 }, { "acc": 0.87225456, "epoch": 2.1597222222222223, "grad_norm": 3.0874061367866665, "learning_rate": 7.595742628066913e-06, "loss": 0.31606097, "memory(GiB)": 101.73, "step": 1555, "train_speed(iter/s)": 0.337863 }, { "acc": 0.8683342, "epoch": 2.1666666666666665, "grad_norm": 2.7476245391903102, "learning_rate": 7.579368320048353e-06, "loss": 0.33775635, "memory(GiB)": 101.73, "step": 1560, "train_speed(iter/s)": 0.337934 }, { "acc": 0.87580605, "epoch": 2.173611111111111, "grad_norm": 3.781738855028497, "learning_rate": 7.562956232748927e-06, "loss": 0.31923892, "memory(GiB)": 101.73, "step": 1565, "train_speed(iter/s)": 0.338011 }, { "acc": 0.87004614, "epoch": 2.1805555555555554, "grad_norm": 3.83368998118028, "learning_rate": 7.5465066065984585e-06, "loss": 0.32810178, "memory(GiB)": 101.73, "step": 1570, "train_speed(iter/s)": 0.338074 }, { "acc": 0.872332, "epoch": 2.1875, "grad_norm": 5.221738940753842, "learning_rate": 7.530019682576701e-06, "loss": 0.35446784, "memory(GiB)": 101.73, "step": 1575, "train_speed(iter/s)": 0.33796 }, { "acc": 0.87468405, "epoch": 2.1944444444444446, "grad_norm": 3.7537602299326096, "learning_rate": 7.5134957022098e-06, "loss": 0.33260701, "memory(GiB)": 101.73, "step": 1580, "train_speed(iter/s)": 0.33798 }, { "acc": 0.88080168, "epoch": 2.201388888888889, "grad_norm": 3.2382032822743336, "learning_rate": 7.496934907566764e-06, "loss": 0.30253038, "memory(GiB)": 101.73, "step": 1585, "train_speed(iter/s)": 0.338059 }, { "acc": 0.87064123, "epoch": 2.2083333333333335, "grad_norm": 2.957663451463655, "learning_rate": 7.480337541255917e-06, "loss": 0.33216987, "memory(GiB)": 101.73, "step": 1590, "train_speed(iter/s)": 0.338127 }, { "acc": 0.87469072, "epoch": 2.2152777777777777, "grad_norm": 2.9386365599515294, "learning_rate": 7.463703846421336e-06, "loss": 0.32307091, "memory(GiB)": 101.73, "step": 1595, "train_speed(iter/s)": 0.338151 }, { "acc": 0.86811104, "epoch": 2.2222222222222223, "grad_norm": 2.9834997972872155, "learning_rate": 7.447034066739297e-06, "loss": 0.33986754, "memory(GiB)": 101.73, "step": 1600, "train_speed(iter/s)": 0.338221 }, { "acc": 0.87765865, "epoch": 2.2291666666666665, "grad_norm": 5.244667526922747, "learning_rate": 7.430328446414703e-06, "loss": 0.32832396, "memory(GiB)": 101.73, "step": 1605, "train_speed(iter/s)": 0.338293 }, { "acc": 0.87501507, "epoch": 2.236111111111111, "grad_norm": 3.484787576558788, "learning_rate": 7.413587230177507e-06, "loss": 0.32662356, "memory(GiB)": 101.73, "step": 1610, "train_speed(iter/s)": 0.338366 }, { "acc": 0.86249599, "epoch": 2.2430555555555554, "grad_norm": 3.004746548822643, "learning_rate": 7.396810663279127e-06, "loss": 0.32253542, "memory(GiB)": 101.73, "step": 1615, "train_speed(iter/s)": 0.338412 }, { "acc": 0.87308559, "epoch": 2.25, "grad_norm": 7.1702579448431845, "learning_rate": 7.3799989914888506e-06, "loss": 0.33047514, "memory(GiB)": 101.73, "step": 1620, "train_speed(iter/s)": 0.338494 }, { "acc": 0.87881336, "epoch": 2.2569444444444446, "grad_norm": 6.139283863744863, "learning_rate": 7.363152461090238e-06, "loss": 0.32465723, "memory(GiB)": 101.73, "step": 1625, "train_speed(iter/s)": 0.338549 }, { "acc": 0.8805707, "epoch": 2.263888888888889, "grad_norm": 4.84624185818044, "learning_rate": 7.346271318877514e-06, "loss": 0.28650289, "memory(GiB)": 101.73, "step": 1630, "train_speed(iter/s)": 0.338616 }, { "acc": 0.87669907, "epoch": 2.2708333333333335, "grad_norm": 4.041825405450047, "learning_rate": 7.329355812151946e-06, "loss": 0.30587997, "memory(GiB)": 101.73, "step": 1635, "train_speed(iter/s)": 0.33868 }, { "acc": 0.88121891, "epoch": 2.2777777777777777, "grad_norm": 28.99476820699483, "learning_rate": 7.312406188718233e-06, "loss": 0.30925446, "memory(GiB)": 101.73, "step": 1640, "train_speed(iter/s)": 0.338738 }, { "acc": 0.87650862, "epoch": 2.2847222222222223, "grad_norm": 2.331146023060369, "learning_rate": 7.295422696880864e-06, "loss": 0.32519441, "memory(GiB)": 101.73, "step": 1645, "train_speed(iter/s)": 0.338753 }, { "acc": 0.8740799, "epoch": 2.2916666666666665, "grad_norm": 2.552893511565151, "learning_rate": 7.2784055854404875e-06, "loss": 0.33000691, "memory(GiB)": 101.73, "step": 1650, "train_speed(iter/s)": 0.338754 }, { "acc": 0.86075459, "epoch": 2.298611111111111, "grad_norm": 3.661826860906401, "learning_rate": 7.261355103690264e-06, "loss": 0.33271484, "memory(GiB)": 101.73, "step": 1655, "train_speed(iter/s)": 0.33882 }, { "acc": 0.87411423, "epoch": 2.3055555555555554, "grad_norm": 2.840348895811668, "learning_rate": 7.244271501412212e-06, "loss": 0.32931597, "memory(GiB)": 101.73, "step": 1660, "train_speed(iter/s)": 0.3388 }, { "acc": 0.87788906, "epoch": 2.3125, "grad_norm": 2.8561468198714377, "learning_rate": 7.227155028873552e-06, "loss": 0.32008338, "memory(GiB)": 101.73, "step": 1665, "train_speed(iter/s)": 0.33883 }, { "acc": 0.85829659, "epoch": 2.3194444444444446, "grad_norm": 4.015346543317703, "learning_rate": 7.210005936823042e-06, "loss": 0.35633571, "memory(GiB)": 101.73, "step": 1670, "train_speed(iter/s)": 0.338871 }, { "acc": 0.87775173, "epoch": 2.326388888888889, "grad_norm": 2.037276521515456, "learning_rate": 7.1928244764873025e-06, "loss": 0.33058481, "memory(GiB)": 101.73, "step": 1675, "train_speed(iter/s)": 0.338868 }, { "acc": 0.87097912, "epoch": 2.3333333333333335, "grad_norm": 3.4288232154502065, "learning_rate": 7.175610899567126e-06, "loss": 0.31878319, "memory(GiB)": 101.73, "step": 1680, "train_speed(iter/s)": 0.338923 }, { "acc": 0.87697048, "epoch": 2.3402777777777777, "grad_norm": 2.853430207452135, "learning_rate": 7.158365458233809e-06, "loss": 0.31011829, "memory(GiB)": 101.73, "step": 1685, "train_speed(iter/s)": 0.338967 }, { "acc": 0.87737989, "epoch": 2.3472222222222223, "grad_norm": 3.7194790748849873, "learning_rate": 7.14108840512544e-06, "loss": 0.3310442, "memory(GiB)": 101.73, "step": 1690, "train_speed(iter/s)": 0.338878 }, { "acc": 0.87455406, "epoch": 2.3541666666666665, "grad_norm": 3.266143323356604, "learning_rate": 7.1237799933432136e-06, "loss": 0.31600342, "memory(GiB)": 101.73, "step": 1695, "train_speed(iter/s)": 0.338867 }, { "acc": 0.87975502, "epoch": 2.361111111111111, "grad_norm": 3.1176281977267815, "learning_rate": 7.10644047644771e-06, "loss": 0.31154928, "memory(GiB)": 101.73, "step": 1700, "train_speed(iter/s)": 0.3389 }, { "acc": 0.87202072, "epoch": 2.3680555555555554, "grad_norm": 3.474595062226248, "learning_rate": 7.089070108455184e-06, "loss": 0.31377418, "memory(GiB)": 101.73, "step": 1705, "train_speed(iter/s)": 0.33847 }, { "acc": 0.8784668, "epoch": 2.375, "grad_norm": 2.7456142238340653, "learning_rate": 7.071669143833848e-06, "loss": 0.31594782, "memory(GiB)": 101.73, "step": 1710, "train_speed(iter/s)": 0.338516 }, { "acc": 0.87724466, "epoch": 2.3819444444444446, "grad_norm": 2.049751609642953, "learning_rate": 7.054237837500145e-06, "loss": 0.30634799, "memory(GiB)": 101.73, "step": 1715, "train_speed(iter/s)": 0.338499 }, { "acc": 0.86771536, "epoch": 2.388888888888889, "grad_norm": 3.4852538358109664, "learning_rate": 7.036776444815005e-06, "loss": 0.33399673, "memory(GiB)": 101.73, "step": 1720, "train_speed(iter/s)": 0.338497 }, { "acc": 0.8677083, "epoch": 2.3958333333333335, "grad_norm": 5.7536631363091, "learning_rate": 7.019285221580112e-06, "loss": 0.34250536, "memory(GiB)": 101.73, "step": 1725, "train_speed(iter/s)": 0.338477 }, { "acc": 0.87606449, "epoch": 2.4027777777777777, "grad_norm": 3.887814129103067, "learning_rate": 7.001764424034153e-06, "loss": 0.32155147, "memory(GiB)": 101.73, "step": 1730, "train_speed(iter/s)": 0.338492 }, { "acc": 0.87954998, "epoch": 2.4097222222222223, "grad_norm": 6.759207980891451, "learning_rate": 6.984214308849067e-06, "loss": 0.31039286, "memory(GiB)": 101.73, "step": 1735, "train_speed(iter/s)": 0.338528 }, { "acc": 0.87517338, "epoch": 2.4166666666666665, "grad_norm": 5.138693499418354, "learning_rate": 6.966635133126286e-06, "loss": 0.32624524, "memory(GiB)": 101.73, "step": 1740, "train_speed(iter/s)": 0.33861 }, { "acc": 0.88268986, "epoch": 2.423611111111111, "grad_norm": 3.522173712318652, "learning_rate": 6.94902715439296e-06, "loss": 0.30495658, "memory(GiB)": 101.73, "step": 1745, "train_speed(iter/s)": 0.338647 }, { "acc": 0.87460957, "epoch": 2.4305555555555554, "grad_norm": 3.5422026180142496, "learning_rate": 6.9313906305981945e-06, "loss": 0.30730667, "memory(GiB)": 101.73, "step": 1750, "train_speed(iter/s)": 0.338681 }, { "acc": 0.8688343, "epoch": 2.4375, "grad_norm": 4.479475459465593, "learning_rate": 6.913725820109267e-06, "loss": 0.33697248, "memory(GiB)": 101.73, "step": 1755, "train_speed(iter/s)": 0.33868 }, { "acc": 0.87335644, "epoch": 2.4444444444444446, "grad_norm": 3.2070844419147213, "learning_rate": 6.896032981707842e-06, "loss": 0.31748641, "memory(GiB)": 101.73, "step": 1760, "train_speed(iter/s)": 0.33873 }, { "acc": 0.87981138, "epoch": 2.451388888888889, "grad_norm": 4.026589843097103, "learning_rate": 6.878312374586182e-06, "loss": 0.31501851, "memory(GiB)": 101.73, "step": 1765, "train_speed(iter/s)": 0.338777 }, { "acc": 0.87052975, "epoch": 2.4583333333333335, "grad_norm": 4.243492199155926, "learning_rate": 6.860564258343344e-06, "loss": 0.32996852, "memory(GiB)": 101.73, "step": 1770, "train_speed(iter/s)": 0.338812 }, { "acc": 0.88656845, "epoch": 2.4652777777777777, "grad_norm": 6.009909931782156, "learning_rate": 6.842788892981389e-06, "loss": 0.29692478, "memory(GiB)": 101.73, "step": 1775, "train_speed(iter/s)": 0.33886 }, { "acc": 0.87949772, "epoch": 2.4722222222222223, "grad_norm": 2.486941299073106, "learning_rate": 6.82498653890156e-06, "loss": 0.30589669, "memory(GiB)": 101.73, "step": 1780, "train_speed(iter/s)": 0.338925 }, { "acc": 0.88737869, "epoch": 2.4791666666666665, "grad_norm": 2.630062639364896, "learning_rate": 6.807157456900474e-06, "loss": 0.2931669, "memory(GiB)": 101.73, "step": 1785, "train_speed(iter/s)": 0.33891 }, { "acc": 0.88413935, "epoch": 2.486111111111111, "grad_norm": 3.2978337938073405, "learning_rate": 6.7893019081663015e-06, "loss": 0.2928961, "memory(GiB)": 101.73, "step": 1790, "train_speed(iter/s)": 0.338994 }, { "acc": 0.87757177, "epoch": 2.4930555555555554, "grad_norm": 2.0263121641711304, "learning_rate": 6.77142015427494e-06, "loss": 0.30921671, "memory(GiB)": 101.73, "step": 1795, "train_speed(iter/s)": 0.339122 }, { "acc": 0.87810173, "epoch": 2.5, "grad_norm": 4.3757059206609625, "learning_rate": 6.753512457186176e-06, "loss": 0.30782709, "memory(GiB)": 101.73, "step": 1800, "train_speed(iter/s)": 0.339211 }, { "acc": 0.87414551, "epoch": 2.5069444444444446, "grad_norm": 2.2068974670317694, "learning_rate": 6.735579079239856e-06, "loss": 0.31782236, "memory(GiB)": 101.73, "step": 1805, "train_speed(iter/s)": 0.33937 }, { "acc": 0.8663393, "epoch": 2.513888888888889, "grad_norm": 3.9925548505742894, "learning_rate": 6.717620283152043e-06, "loss": 0.32476821, "memory(GiB)": 101.73, "step": 1810, "train_speed(iter/s)": 0.339424 }, { "acc": 0.8749526, "epoch": 2.5208333333333335, "grad_norm": 4.0296739697695365, "learning_rate": 6.699636332011156e-06, "loss": 0.2988498, "memory(GiB)": 101.73, "step": 1815, "train_speed(iter/s)": 0.339512 }, { "acc": 0.88200283, "epoch": 2.5277777777777777, "grad_norm": 4.4011471119110395, "learning_rate": 6.681627489274131e-06, "loss": 0.2915164, "memory(GiB)": 101.73, "step": 1820, "train_speed(iter/s)": 0.339549 }, { "acc": 0.8799408, "epoch": 2.5347222222222223, "grad_norm": 3.0590205668391266, "learning_rate": 6.663594018762553e-06, "loss": 0.30144064, "memory(GiB)": 101.73, "step": 1825, "train_speed(iter/s)": 0.339641 }, { "acc": 0.88366776, "epoch": 2.5416666666666665, "grad_norm": 3.997001096827074, "learning_rate": 6.645536184658794e-06, "loss": 0.29560423, "memory(GiB)": 101.73, "step": 1830, "train_speed(iter/s)": 0.339716 }, { "acc": 0.87788258, "epoch": 2.548611111111111, "grad_norm": 3.4793648553253167, "learning_rate": 6.627454251502139e-06, "loss": 0.3051878, "memory(GiB)": 101.73, "step": 1835, "train_speed(iter/s)": 0.339811 }, { "acc": 0.86607494, "epoch": 2.5555555555555554, "grad_norm": 2.963649888950391, "learning_rate": 6.609348484184916e-06, "loss": 0.32486575, "memory(GiB)": 101.73, "step": 1840, "train_speed(iter/s)": 0.339872 }, { "acc": 0.88934765, "epoch": 2.5625, "grad_norm": 3.0696594994479844, "learning_rate": 6.591219147948616e-06, "loss": 0.29468005, "memory(GiB)": 101.73, "step": 1845, "train_speed(iter/s)": 0.339887 }, { "acc": 0.88128185, "epoch": 2.5694444444444446, "grad_norm": 2.8091039075490536, "learning_rate": 6.573066508379994e-06, "loss": 0.30400395, "memory(GiB)": 101.73, "step": 1850, "train_speed(iter/s)": 0.339965 }, { "acc": 0.88058329, "epoch": 2.576388888888889, "grad_norm": 4.0255629531145045, "learning_rate": 6.554890831407199e-06, "loss": 0.31955268, "memory(GiB)": 101.73, "step": 1855, "train_speed(iter/s)": 0.340015 }, { "acc": 0.88047981, "epoch": 2.5833333333333335, "grad_norm": 2.4218935283386487, "learning_rate": 6.536692383295863e-06, "loss": 0.3094146, "memory(GiB)": 101.73, "step": 1860, "train_speed(iter/s)": 0.340104 }, { "acc": 0.87942057, "epoch": 2.5902777777777777, "grad_norm": 3.7938712045310554, "learning_rate": 6.518471430645206e-06, "loss": 0.33601379, "memory(GiB)": 101.73, "step": 1865, "train_speed(iter/s)": 0.340142 }, { "acc": 0.87505875, "epoch": 2.5972222222222223, "grad_norm": 4.029822102213215, "learning_rate": 6.50022824038413e-06, "loss": 0.30280771, "memory(GiB)": 101.73, "step": 1870, "train_speed(iter/s)": 0.340218 }, { "acc": 0.88756218, "epoch": 2.6041666666666665, "grad_norm": 2.6352853592698313, "learning_rate": 6.481963079767307e-06, "loss": 0.30299997, "memory(GiB)": 101.73, "step": 1875, "train_speed(iter/s)": 0.340247 }, { "acc": 0.87335434, "epoch": 2.611111111111111, "grad_norm": 5.405459671127271, "learning_rate": 6.463676216371266e-06, "loss": 0.32759867, "memory(GiB)": 101.73, "step": 1880, "train_speed(iter/s)": 0.34017 }, { "acc": 0.8905899, "epoch": 2.6180555555555554, "grad_norm": 4.051931136735315, "learning_rate": 6.4453679180904725e-06, "loss": 0.2927712, "memory(GiB)": 101.73, "step": 1885, "train_speed(iter/s)": 0.340225 }, { "acc": 0.88374195, "epoch": 2.625, "grad_norm": 3.045321365849001, "learning_rate": 6.427038453133403e-06, "loss": 0.30915735, "memory(GiB)": 101.73, "step": 1890, "train_speed(iter/s)": 0.340313 }, { "acc": 0.88091192, "epoch": 2.6319444444444446, "grad_norm": 5.40816330790365, "learning_rate": 6.4086880900186135e-06, "loss": 0.30887535, "memory(GiB)": 101.73, "step": 1895, "train_speed(iter/s)": 0.340392 }, { "acc": 0.8715085, "epoch": 2.638888888888889, "grad_norm": 4.947011374969508, "learning_rate": 6.3903170975708165e-06, "loss": 0.33568881, "memory(GiB)": 101.73, "step": 1900, "train_speed(iter/s)": 0.340437 }, { "acc": 0.88775291, "epoch": 2.6458333333333335, "grad_norm": 4.3365471091357, "learning_rate": 6.371925744916924e-06, "loss": 0.29443359, "memory(GiB)": 101.73, "step": 1905, "train_speed(iter/s)": 0.340519 }, { "acc": 0.87607994, "epoch": 2.6527777777777777, "grad_norm": 8.098944228234716, "learning_rate": 6.353514301482126e-06, "loss": 0.31685441, "memory(GiB)": 101.73, "step": 1910, "train_speed(iter/s)": 0.340593 }, { "acc": 0.87877178, "epoch": 2.6597222222222223, "grad_norm": 2.7987374198103026, "learning_rate": 6.3350830369859315e-06, "loss": 0.31020045, "memory(GiB)": 101.73, "step": 1915, "train_speed(iter/s)": 0.340604 }, { "acc": 0.87018623, "epoch": 2.6666666666666665, "grad_norm": 4.328884885182, "learning_rate": 6.316632221438214e-06, "loss": 0.32079277, "memory(GiB)": 101.73, "step": 1920, "train_speed(iter/s)": 0.340662 }, { "acc": 0.8835763, "epoch": 2.673611111111111, "grad_norm": 2.4783741723058497, "learning_rate": 6.298162125135268e-06, "loss": 0.30358009, "memory(GiB)": 101.73, "step": 1925, "train_speed(iter/s)": 0.340706 }, { "acc": 0.883319, "epoch": 2.6805555555555554, "grad_norm": 4.354308492198902, "learning_rate": 6.279673018655836e-06, "loss": 0.31000361, "memory(GiB)": 101.73, "step": 1930, "train_speed(iter/s)": 0.340781 }, { "acc": 0.8759573, "epoch": 2.6875, "grad_norm": 9.467690371837232, "learning_rate": 6.2611651728571564e-06, "loss": 0.31353772, "memory(GiB)": 101.73, "step": 1935, "train_speed(iter/s)": 0.340841 }, { "acc": 0.88470545, "epoch": 2.6944444444444446, "grad_norm": 3.215878487764247, "learning_rate": 6.242638858870988e-06, "loss": 0.28434479, "memory(GiB)": 101.73, "step": 1940, "train_speed(iter/s)": 0.34093 }, { "acc": 0.87855301, "epoch": 2.701388888888889, "grad_norm": 3.3912993182694864, "learning_rate": 6.224094348099642e-06, "loss": 0.29945769, "memory(GiB)": 101.73, "step": 1945, "train_speed(iter/s)": 0.341019 }, { "acc": 0.8707325, "epoch": 2.7083333333333335, "grad_norm": 4.58702527797853, "learning_rate": 6.205531912212003e-06, "loss": 0.29754431, "memory(GiB)": 101.73, "step": 1950, "train_speed(iter/s)": 0.341102 }, { "acc": 0.88574533, "epoch": 2.7152777777777777, "grad_norm": 3.699483186725238, "learning_rate": 6.186951823139551e-06, "loss": 0.30343163, "memory(GiB)": 101.73, "step": 1955, "train_speed(iter/s)": 0.341216 }, { "acc": 0.87052441, "epoch": 2.7222222222222223, "grad_norm": 3.799468816636163, "learning_rate": 6.168354353072376e-06, "loss": 0.31728518, "memory(GiB)": 101.73, "step": 1960, "train_speed(iter/s)": 0.341301 }, { "acc": 0.88283157, "epoch": 2.7291666666666665, "grad_norm": 5.022537915160293, "learning_rate": 6.149739774455192e-06, "loss": 0.29690897, "memory(GiB)": 101.73, "step": 1965, "train_speed(iter/s)": 0.341336 }, { "acc": 0.88846045, "epoch": 2.736111111111111, "grad_norm": 3.963520423961829, "learning_rate": 6.131108359983352e-06, "loss": 0.28321784, "memory(GiB)": 101.73, "step": 1970, "train_speed(iter/s)": 0.341424 }, { "acc": 0.88815289, "epoch": 2.7430555555555554, "grad_norm": 4.477823960200574, "learning_rate": 6.112460382598838e-06, "loss": 0.28913431, "memory(GiB)": 101.73, "step": 1975, "train_speed(iter/s)": 0.341507 }, { "acc": 0.87810936, "epoch": 2.75, "grad_norm": 5.001346704079185, "learning_rate": 6.093796115486277e-06, "loss": 0.30853729, "memory(GiB)": 101.73, "step": 1980, "train_speed(iter/s)": 0.341576 }, { "acc": 0.87556248, "epoch": 2.7569444444444446, "grad_norm": 2.6486962898058937, "learning_rate": 6.075115832068929e-06, "loss": 0.29869452, "memory(GiB)": 101.73, "step": 1985, "train_speed(iter/s)": 0.341652 }, { "acc": 0.88768387, "epoch": 2.763888888888889, "grad_norm": 2.647997997832992, "learning_rate": 6.056419806004689e-06, "loss": 0.28908353, "memory(GiB)": 101.73, "step": 1990, "train_speed(iter/s)": 0.341729 }, { "acc": 0.88060932, "epoch": 2.7708333333333335, "grad_norm": 3.279715246610578, "learning_rate": 6.037708311182076e-06, "loss": 0.29757085, "memory(GiB)": 101.73, "step": 1995, "train_speed(iter/s)": 0.341816 }, { "acc": 0.89519711, "epoch": 2.7777777777777777, "grad_norm": 2.350006963183419, "learning_rate": 6.018981621716216e-06, "loss": 0.29250665, "memory(GiB)": 101.73, "step": 2000, "train_speed(iter/s)": 0.341904 }, { "acc": 0.8793849, "epoch": 2.7847222222222223, "grad_norm": 7.413019755920625, "learning_rate": 6.0002400119448315e-06, "loss": 0.29689407, "memory(GiB)": 101.73, "step": 2005, "train_speed(iter/s)": 0.341989 }, { "acc": 0.88535957, "epoch": 2.7916666666666665, "grad_norm": 3.0590690331077215, "learning_rate": 5.981483756424219e-06, "loss": 0.28531508, "memory(GiB)": 101.73, "step": 2010, "train_speed(iter/s)": 0.342074 }, { "acc": 0.88893423, "epoch": 2.798611111111111, "grad_norm": 6.003641582203187, "learning_rate": 5.962713129925233e-06, "loss": 0.29722118, "memory(GiB)": 101.73, "step": 2015, "train_speed(iter/s)": 0.342134 }, { "acc": 0.8882515, "epoch": 2.8055555555555554, "grad_norm": 5.313173888563081, "learning_rate": 5.943928407429251e-06, "loss": 0.31478307, "memory(GiB)": 101.73, "step": 2020, "train_speed(iter/s)": 0.342214 }, { "acc": 0.88190327, "epoch": 2.8125, "grad_norm": 4.986780722952902, "learning_rate": 5.9251298641241565e-06, "loss": 0.30542126, "memory(GiB)": 101.73, "step": 2025, "train_speed(iter/s)": 0.342286 }, { "acc": 0.89250832, "epoch": 2.8194444444444446, "grad_norm": 5.018536108237801, "learning_rate": 5.906317775400298e-06, "loss": 0.26341515, "memory(GiB)": 101.73, "step": 2030, "train_speed(iter/s)": 0.342356 }, { "acc": 0.87831106, "epoch": 2.826388888888889, "grad_norm": 5.146920963485941, "learning_rate": 5.887492416846459e-06, "loss": 0.30127568, "memory(GiB)": 101.73, "step": 2035, "train_speed(iter/s)": 0.34243 }, { "acc": 0.88257599, "epoch": 2.8333333333333335, "grad_norm": 3.0687672661039476, "learning_rate": 5.8686540642458204e-06, "loss": 0.29559946, "memory(GiB)": 101.73, "step": 2040, "train_speed(iter/s)": 0.3425 }, { "acc": 0.88266973, "epoch": 2.8402777777777777, "grad_norm": 5.535020326656706, "learning_rate": 5.849802993571917e-06, "loss": 0.30087221, "memory(GiB)": 101.73, "step": 2045, "train_speed(iter/s)": 0.342532 }, { "acc": 0.89446363, "epoch": 2.8472222222222223, "grad_norm": 4.612098734690172, "learning_rate": 5.830939480984605e-06, "loss": 0.27696779, "memory(GiB)": 101.73, "step": 2050, "train_speed(iter/s)": 0.342586 }, { "acc": 0.87654648, "epoch": 2.8541666666666665, "grad_norm": 5.014707434986776, "learning_rate": 5.812063802826005e-06, "loss": 0.31633348, "memory(GiB)": 101.73, "step": 2055, "train_speed(iter/s)": 0.342664 }, { "acc": 0.88131304, "epoch": 2.861111111111111, "grad_norm": 3.870152061287398, "learning_rate": 5.793176235616455e-06, "loss": 0.32585454, "memory(GiB)": 101.73, "step": 2060, "train_speed(iter/s)": 0.342728 }, { "acc": 0.88249874, "epoch": 2.8680555555555554, "grad_norm": 3.1435056987806993, "learning_rate": 5.774277056050467e-06, "loss": 0.29018085, "memory(GiB)": 101.73, "step": 2065, "train_speed(iter/s)": 0.342823 }, { "acc": 0.88479662, "epoch": 2.875, "grad_norm": 5.000934426273714, "learning_rate": 5.755366540992666e-06, "loss": 0.30050292, "memory(GiB)": 101.73, "step": 2070, "train_speed(iter/s)": 0.342871 }, { "acc": 0.88433819, "epoch": 2.8819444444444446, "grad_norm": 6.225049436263428, "learning_rate": 5.736444967473736e-06, "loss": 0.28958912, "memory(GiB)": 101.73, "step": 2075, "train_speed(iter/s)": 0.342876 }, { "acc": 0.88738003, "epoch": 2.888888888888889, "grad_norm": 5.829606140426472, "learning_rate": 5.71751261268637e-06, "loss": 0.28439336, "memory(GiB)": 101.73, "step": 2080, "train_speed(iter/s)": 0.342927 }, { "acc": 0.87550907, "epoch": 2.8958333333333335, "grad_norm": 4.927935596858437, "learning_rate": 5.698569753981194e-06, "loss": 0.30479741, "memory(GiB)": 101.73, "step": 2085, "train_speed(iter/s)": 0.342954 }, { "acc": 0.88370075, "epoch": 2.9027777777777777, "grad_norm": 4.982074354249153, "learning_rate": 5.67961666886271e-06, "loss": 0.29437392, "memory(GiB)": 101.73, "step": 2090, "train_speed(iter/s)": 0.342977 }, { "acc": 0.89414291, "epoch": 2.9097222222222223, "grad_norm": 2.699118597588968, "learning_rate": 5.6606536349852425e-06, "loss": 0.26323729, "memory(GiB)": 101.73, "step": 2095, "train_speed(iter/s)": 0.34301 }, { "acc": 0.89436359, "epoch": 2.9166666666666665, "grad_norm": 4.450064476738222, "learning_rate": 5.64168093014885e-06, "loss": 0.27156312, "memory(GiB)": 101.73, "step": 2100, "train_speed(iter/s)": 0.343044 }, { "acc": 0.89274464, "epoch": 2.923611111111111, "grad_norm": 2.9247570340950912, "learning_rate": 5.622698832295274e-06, "loss": 0.28357964, "memory(GiB)": 101.73, "step": 2105, "train_speed(iter/s)": 0.343072 }, { "acc": 0.89117432, "epoch": 2.9305555555555554, "grad_norm": 3.9442967825811723, "learning_rate": 5.603707619503858e-06, "loss": 0.29273071, "memory(GiB)": 101.73, "step": 2110, "train_speed(iter/s)": 0.34312 }, { "acc": 0.88488503, "epoch": 2.9375, "grad_norm": 3.9784644222883245, "learning_rate": 5.584707569987471e-06, "loss": 0.27594287, "memory(GiB)": 101.73, "step": 2115, "train_speed(iter/s)": 0.343143 }, { "acc": 0.88866596, "epoch": 2.9444444444444446, "grad_norm": 15.436705597346922, "learning_rate": 5.565698962088441e-06, "loss": 0.29711914, "memory(GiB)": 101.73, "step": 2120, "train_speed(iter/s)": 0.343188 }, { "acc": 0.88177891, "epoch": 2.951388888888889, "grad_norm": 5.662244591316013, "learning_rate": 5.54668207427447e-06, "loss": 0.29541435, "memory(GiB)": 101.73, "step": 2125, "train_speed(iter/s)": 0.343239 }, { "acc": 0.89320126, "epoch": 2.9583333333333335, "grad_norm": 2.4843357852192427, "learning_rate": 5.527657185134556e-06, "loss": 0.28223519, "memory(GiB)": 101.73, "step": 2130, "train_speed(iter/s)": 0.343282 }, { "acc": 0.89105587, "epoch": 2.9652777777777777, "grad_norm": 2.910744869592209, "learning_rate": 5.508624573374919e-06, "loss": 0.27124515, "memory(GiB)": 101.73, "step": 2135, "train_speed(iter/s)": 0.343292 }, { "acc": 0.88511724, "epoch": 2.9722222222222223, "grad_norm": 2.73890828638729, "learning_rate": 5.4895845178149045e-06, "loss": 0.29908915, "memory(GiB)": 101.73, "step": 2140, "train_speed(iter/s)": 0.343284 }, { "acc": 0.89417515, "epoch": 2.9791666666666665, "grad_norm": 3.936833986713451, "learning_rate": 5.470537297382911e-06, "loss": 0.28417382, "memory(GiB)": 101.73, "step": 2145, "train_speed(iter/s)": 0.343315 }, { "acc": 0.88898296, "epoch": 2.986111111111111, "grad_norm": 3.8512956350719354, "learning_rate": 5.451483191112295e-06, "loss": 0.28061602, "memory(GiB)": 101.73, "step": 2150, "train_speed(iter/s)": 0.343201 }, { "acc": 0.89546566, "epoch": 2.9930555555555554, "grad_norm": 5.810228058013454, "learning_rate": 5.432422478137295e-06, "loss": 0.27272539, "memory(GiB)": 101.73, "step": 2155, "train_speed(iter/s)": 0.343197 }, { "acc": 0.88720446, "epoch": 3.0, "grad_norm": 3.580049072617942, "learning_rate": 5.413355437688926e-06, "loss": 0.27387962, "memory(GiB)": 101.73, "step": 2160, "train_speed(iter/s)": 0.343171 }, { "epoch": 3.0, "eval_acc": 0.8911203557595754, "eval_loss": 0.29212549328804016, "eval_runtime": 25.6208, "eval_samples_per_second": 13.388, "eval_steps_per_second": 1.678, "step": 2160 }, { "acc": 0.89589024, "epoch": 3.0069444444444446, "grad_norm": 4.637545293434605, "learning_rate": 5.394282349090905e-06, "loss": 0.25944438, "memory(GiB)": 101.73, "step": 2165, "train_speed(iter/s)": 0.340007 }, { "acc": 0.88720493, "epoch": 3.013888888888889, "grad_norm": 7.39139114160879, "learning_rate": 5.375203491755545e-06, "loss": 0.31616793, "memory(GiB)": 101.73, "step": 2170, "train_speed(iter/s)": 0.340103 }, { "acc": 0.88987474, "epoch": 3.0208333333333335, "grad_norm": 4.064763480331921, "learning_rate": 5.356119145179673e-06, "loss": 0.27186308, "memory(GiB)": 101.73, "step": 2175, "train_speed(iter/s)": 0.340144 }, { "acc": 0.88327274, "epoch": 3.0277777777777777, "grad_norm": 3.2940720072242153, "learning_rate": 5.33702958894053e-06, "loss": 0.29375, "memory(GiB)": 101.73, "step": 2180, "train_speed(iter/s)": 0.340236 }, { "acc": 0.89080715, "epoch": 3.0347222222222223, "grad_norm": 2.801423990050791, "learning_rate": 5.317935102691677e-06, "loss": 0.28847058, "memory(GiB)": 101.73, "step": 2185, "train_speed(iter/s)": 0.340293 }, { "acc": 0.89707794, "epoch": 3.0416666666666665, "grad_norm": 3.3077023232108407, "learning_rate": 5.298835966158896e-06, "loss": 0.25797377, "memory(GiB)": 101.73, "step": 2190, "train_speed(iter/s)": 0.340319 }, { "acc": 0.88909359, "epoch": 3.048611111111111, "grad_norm": 3.9604632504058146, "learning_rate": 5.279732459136095e-06, "loss": 0.27923727, "memory(GiB)": 101.73, "step": 2195, "train_speed(iter/s)": 0.340391 }, { "acc": 0.8788662, "epoch": 3.0555555555555554, "grad_norm": 3.407305740491057, "learning_rate": 5.260624861481205e-06, "loss": 0.29525146, "memory(GiB)": 101.73, "step": 2200, "train_speed(iter/s)": 0.340393 }, { "acc": 0.88607311, "epoch": 3.0625, "grad_norm": 3.4070141014765807, "learning_rate": 5.241513453112088e-06, "loss": 0.29450035, "memory(GiB)": 101.73, "step": 2205, "train_speed(iter/s)": 0.340409 }, { "acc": 0.89028587, "epoch": 3.0694444444444446, "grad_norm": 6.119391343010178, "learning_rate": 5.222398514002424e-06, "loss": 0.27855229, "memory(GiB)": 101.73, "step": 2210, "train_speed(iter/s)": 0.34041 }, { "acc": 0.90690746, "epoch": 3.076388888888889, "grad_norm": 4.189972170403801, "learning_rate": 5.203280324177623e-06, "loss": 0.24123404, "memory(GiB)": 101.73, "step": 2215, "train_speed(iter/s)": 0.340429 }, { "acc": 0.89223652, "epoch": 3.0833333333333335, "grad_norm": 6.1687310886506825, "learning_rate": 5.184159163710717e-06, "loss": 0.26408105, "memory(GiB)": 101.73, "step": 2220, "train_speed(iter/s)": 0.340459 }, { "acc": 0.88885937, "epoch": 3.0902777777777777, "grad_norm": 3.6370135452975982, "learning_rate": 5.16503531271825e-06, "loss": 0.28215942, "memory(GiB)": 101.73, "step": 2225, "train_speed(iter/s)": 0.340481 }, { "acc": 0.8864089, "epoch": 3.0972222222222223, "grad_norm": 2.6915181888376987, "learning_rate": 5.145909051356187e-06, "loss": 0.27783501, "memory(GiB)": 101.73, "step": 2230, "train_speed(iter/s)": 0.340506 }, { "acc": 0.88523932, "epoch": 3.1041666666666665, "grad_norm": 2.6201181743435873, "learning_rate": 5.1267806598157995e-06, "loss": 0.26578436, "memory(GiB)": 101.73, "step": 2235, "train_speed(iter/s)": 0.340551 }, { "acc": 0.88679752, "epoch": 3.111111111111111, "grad_norm": 2.3757078771167235, "learning_rate": 5.1076504183195726e-06, "loss": 0.2708406, "memory(GiB)": 101.73, "step": 2240, "train_speed(iter/s)": 0.340577 }, { "acc": 0.89169292, "epoch": 3.1180555555555554, "grad_norm": 4.281469899344993, "learning_rate": 5.088518607117084e-06, "loss": 0.26542788, "memory(GiB)": 101.73, "step": 2245, "train_speed(iter/s)": 0.340582 }, { "acc": 0.88440208, "epoch": 3.125, "grad_norm": 6.327375242332073, "learning_rate": 5.0693855064809104e-06, "loss": 0.28328218, "memory(GiB)": 101.73, "step": 2250, "train_speed(iter/s)": 0.340592 }, { "acc": 0.88965473, "epoch": 3.1319444444444446, "grad_norm": 3.335606378012737, "learning_rate": 5.050251396702519e-06, "loss": 0.27447596, "memory(GiB)": 101.73, "step": 2255, "train_speed(iter/s)": 0.340564 }, { "acc": 0.89447746, "epoch": 3.138888888888889, "grad_norm": 3.971413493473237, "learning_rate": 5.0311165580881585e-06, "loss": 0.26328266, "memory(GiB)": 101.73, "step": 2260, "train_speed(iter/s)": 0.340575 }, { "acc": 0.88526878, "epoch": 3.1458333333333335, "grad_norm": 3.9913556594484008, "learning_rate": 5.0119812709547566e-06, "loss": 0.26721802, "memory(GiB)": 101.73, "step": 2265, "train_speed(iter/s)": 0.340561 }, { "acc": 0.89905844, "epoch": 3.1527777777777777, "grad_norm": 3.6292317674402117, "learning_rate": 4.9928458156258105e-06, "loss": 0.24952831, "memory(GiB)": 101.73, "step": 2270, "train_speed(iter/s)": 0.340598 }, { "acc": 0.90703621, "epoch": 3.1597222222222223, "grad_norm": 2.5916058110248397, "learning_rate": 4.973710472427283e-06, "loss": 0.24961276, "memory(GiB)": 101.73, "step": 2275, "train_speed(iter/s)": 0.34063 }, { "acc": 0.8791647, "epoch": 3.1666666666666665, "grad_norm": 4.9755713822504175, "learning_rate": 4.954575521683491e-06, "loss": 0.29728041, "memory(GiB)": 101.73, "step": 2280, "train_speed(iter/s)": 0.340679 }, { "acc": 0.89638615, "epoch": 3.173611111111111, "grad_norm": 4.765953277092038, "learning_rate": 4.935441243713005e-06, "loss": 0.26370394, "memory(GiB)": 101.73, "step": 2285, "train_speed(iter/s)": 0.340579 }, { "acc": 0.89993305, "epoch": 3.1805555555555554, "grad_norm": 4.796855317724758, "learning_rate": 4.916307918824538e-06, "loss": 0.25945036, "memory(GiB)": 101.73, "step": 2290, "train_speed(iter/s)": 0.340624 }, { "acc": 0.89749622, "epoch": 3.1875, "grad_norm": 4.077903509861586, "learning_rate": 4.897175827312842e-06, "loss": 0.25814972, "memory(GiB)": 101.73, "step": 2295, "train_speed(iter/s)": 0.340634 }, { "acc": 0.89532013, "epoch": 3.1944444444444446, "grad_norm": 2.9930679640780458, "learning_rate": 4.878045249454601e-06, "loss": 0.25680308, "memory(GiB)": 101.73, "step": 2300, "train_speed(iter/s)": 0.340629 }, { "acc": 0.89404106, "epoch": 3.201388888888889, "grad_norm": 9.044532042244569, "learning_rate": 4.858916465504323e-06, "loss": 0.26445313, "memory(GiB)": 101.73, "step": 2305, "train_speed(iter/s)": 0.340667 }, { "acc": 0.88623104, "epoch": 3.2083333333333335, "grad_norm": 6.440284323833954, "learning_rate": 4.839789755690235e-06, "loss": 0.27099028, "memory(GiB)": 101.73, "step": 2310, "train_speed(iter/s)": 0.340689 }, { "acc": 0.88883505, "epoch": 3.2152777777777777, "grad_norm": 4.008447674699076, "learning_rate": 4.820665400210183e-06, "loss": 0.26926422, "memory(GiB)": 101.73, "step": 2315, "train_speed(iter/s)": 0.34073 }, { "acc": 0.89209347, "epoch": 3.2222222222222223, "grad_norm": 2.9462752953211084, "learning_rate": 4.801543679227519e-06, "loss": 0.28077393, "memory(GiB)": 101.73, "step": 2320, "train_speed(iter/s)": 0.340768 }, { "acc": 0.90544291, "epoch": 3.2291666666666665, "grad_norm": 3.2070127351402573, "learning_rate": 4.782424872867004e-06, "loss": 0.25720313, "memory(GiB)": 101.73, "step": 2325, "train_speed(iter/s)": 0.340812 }, { "acc": 0.89043808, "epoch": 3.236111111111111, "grad_norm": 4.963263502548416, "learning_rate": 4.763309261210697e-06, "loss": 0.26552734, "memory(GiB)": 101.73, "step": 2330, "train_speed(iter/s)": 0.340869 }, { "acc": 0.89086895, "epoch": 3.2430555555555554, "grad_norm": 5.996678984257977, "learning_rate": 4.744197124293861e-06, "loss": 0.29014907, "memory(GiB)": 101.73, "step": 2335, "train_speed(iter/s)": 0.340916 }, { "acc": 0.88824921, "epoch": 3.25, "grad_norm": 4.661978038167385, "learning_rate": 4.725088742100851e-06, "loss": 0.27598886, "memory(GiB)": 101.73, "step": 2340, "train_speed(iter/s)": 0.340975 }, { "acc": 0.90345469, "epoch": 3.2569444444444446, "grad_norm": 5.067431282964852, "learning_rate": 4.70598439456102e-06, "loss": 0.2427031, "memory(GiB)": 101.73, "step": 2345, "train_speed(iter/s)": 0.341045 }, { "acc": 0.91135798, "epoch": 3.263888888888889, "grad_norm": 4.374947699458002, "learning_rate": 4.686884361544615e-06, "loss": 0.24237332, "memory(GiB)": 101.73, "step": 2350, "train_speed(iter/s)": 0.34109 }, { "acc": 0.88529987, "epoch": 3.2708333333333335, "grad_norm": 3.2497593782681626, "learning_rate": 4.667788922858675e-06, "loss": 0.27918091, "memory(GiB)": 101.73, "step": 2355, "train_speed(iter/s)": 0.341135 }, { "acc": 0.8951086, "epoch": 3.2777777777777777, "grad_norm": 4.703813669876629, "learning_rate": 4.648698358242937e-06, "loss": 0.27369049, "memory(GiB)": 101.73, "step": 2360, "train_speed(iter/s)": 0.341181 }, { "acc": 0.89912701, "epoch": 3.2847222222222223, "grad_norm": 3.0732915304334663, "learning_rate": 4.629612947365731e-06, "loss": 0.26150217, "memory(GiB)": 101.73, "step": 2365, "train_speed(iter/s)": 0.341223 }, { "acc": 0.90068092, "epoch": 3.2916666666666665, "grad_norm": 5.055103907548479, "learning_rate": 4.61053296981989e-06, "loss": 0.25842612, "memory(GiB)": 101.73, "step": 2370, "train_speed(iter/s)": 0.341244 }, { "acc": 0.89772367, "epoch": 3.298611111111111, "grad_norm": 6.636649309949125, "learning_rate": 4.591458705118651e-06, "loss": 0.26917741, "memory(GiB)": 101.73, "step": 2375, "train_speed(iter/s)": 0.341284 }, { "acc": 0.90206804, "epoch": 3.3055555555555554, "grad_norm": 3.898114339474827, "learning_rate": 4.5723904326915605e-06, "loss": 0.25737305, "memory(GiB)": 101.73, "step": 2380, "train_speed(iter/s)": 0.341313 }, { "acc": 0.90095797, "epoch": 3.3125, "grad_norm": 7.406618692458744, "learning_rate": 4.55332843188038e-06, "loss": 0.24617977, "memory(GiB)": 101.73, "step": 2385, "train_speed(iter/s)": 0.341359 }, { "acc": 0.90679264, "epoch": 3.3194444444444446, "grad_norm": 3.2278717071765546, "learning_rate": 4.534272981934993e-06, "loss": 0.26191123, "memory(GiB)": 101.73, "step": 2390, "train_speed(iter/s)": 0.341438 }, { "acc": 0.89675159, "epoch": 3.326388888888889, "grad_norm": 6.232358660573751, "learning_rate": 4.51522436200932e-06, "loss": 0.26368232, "memory(GiB)": 101.73, "step": 2395, "train_speed(iter/s)": 0.341492 }, { "acc": 0.89320259, "epoch": 3.3333333333333335, "grad_norm": 9.336217090017287, "learning_rate": 4.4961828511572195e-06, "loss": 0.26878896, "memory(GiB)": 101.73, "step": 2400, "train_speed(iter/s)": 0.341537 }, { "acc": 0.90293465, "epoch": 3.3402777777777777, "grad_norm": 5.771989254299704, "learning_rate": 4.477148728328411e-06, "loss": 0.23618774, "memory(GiB)": 101.73, "step": 2405, "train_speed(iter/s)": 0.341597 }, { "acc": 0.89542847, "epoch": 3.3472222222222223, "grad_norm": 8.890216489036503, "learning_rate": 4.4581222723643775e-06, "loss": 0.27369397, "memory(GiB)": 101.73, "step": 2410, "train_speed(iter/s)": 0.341663 }, { "acc": 0.89897604, "epoch": 3.3541666666666665, "grad_norm": 10.209516705903122, "learning_rate": 4.439103761994287e-06, "loss": 0.2468132, "memory(GiB)": 101.73, "step": 2415, "train_speed(iter/s)": 0.341703 }, { "acc": 0.88786516, "epoch": 3.361111111111111, "grad_norm": 3.844549530620871, "learning_rate": 4.420093475830912e-06, "loss": 0.25787063, "memory(GiB)": 101.73, "step": 2420, "train_speed(iter/s)": 0.341747 }, { "acc": 0.90001011, "epoch": 3.3680555555555554, "grad_norm": 3.6744603092340515, "learning_rate": 4.401091692366538e-06, "loss": 0.25619226, "memory(GiB)": 101.73, "step": 2425, "train_speed(iter/s)": 0.341777 }, { "acc": 0.8973815, "epoch": 3.375, "grad_norm": 7.238511447008848, "learning_rate": 4.382098689968897e-06, "loss": 0.2451417, "memory(GiB)": 101.73, "step": 2430, "train_speed(iter/s)": 0.341821 }, { "acc": 0.89963903, "epoch": 3.3819444444444446, "grad_norm": 4.463477465284448, "learning_rate": 4.363114746877074e-06, "loss": 0.2622735, "memory(GiB)": 101.73, "step": 2435, "train_speed(iter/s)": 0.341849 }, { "acc": 0.8982995, "epoch": 3.388888888888889, "grad_norm": 4.584390700541224, "learning_rate": 4.3441401411974445e-06, "loss": 0.25706239, "memory(GiB)": 101.73, "step": 2440, "train_speed(iter/s)": 0.341872 }, { "acc": 0.88739204, "epoch": 3.3958333333333335, "grad_norm": 7.2700204144048035, "learning_rate": 4.325175150899594e-06, "loss": 0.27985995, "memory(GiB)": 101.73, "step": 2445, "train_speed(iter/s)": 0.34188 }, { "acc": 0.90591974, "epoch": 3.4027777777777777, "grad_norm": 4.701025788285595, "learning_rate": 4.306220053812247e-06, "loss": 0.25586658, "memory(GiB)": 101.73, "step": 2450, "train_speed(iter/s)": 0.341863 }, { "acc": 0.89671564, "epoch": 3.4097222222222223, "grad_norm": 3.5024092817845762, "learning_rate": 4.287275127619194e-06, "loss": 0.26711457, "memory(GiB)": 101.73, "step": 2455, "train_speed(iter/s)": 0.341912 }, { "acc": 0.89187412, "epoch": 3.4166666666666665, "grad_norm": 3.1331350139936176, "learning_rate": 4.268340649855233e-06, "loss": 0.26458747, "memory(GiB)": 101.73, "step": 2460, "train_speed(iter/s)": 0.34197 }, { "acc": 0.89410534, "epoch": 3.423611111111111, "grad_norm": 2.6386667139020323, "learning_rate": 4.24941689790209e-06, "loss": 0.2519309, "memory(GiB)": 101.73, "step": 2465, "train_speed(iter/s)": 0.342002 }, { "acc": 0.91391573, "epoch": 3.4305555555555554, "grad_norm": 3.3930313458101087, "learning_rate": 4.230504148984366e-06, "loss": 0.22323613, "memory(GiB)": 101.73, "step": 2470, "train_speed(iter/s)": 0.342031 }, { "acc": 0.88966484, "epoch": 3.4375, "grad_norm": 3.576144793306567, "learning_rate": 4.211602680165474e-06, "loss": 0.26663389, "memory(GiB)": 101.73, "step": 2475, "train_speed(iter/s)": 0.342056 }, { "acc": 0.90185118, "epoch": 3.4444444444444446, "grad_norm": 3.0351618188615954, "learning_rate": 4.192712768343573e-06, "loss": 0.25376158, "memory(GiB)": 101.73, "step": 2480, "train_speed(iter/s)": 0.342085 }, { "acc": 0.89549732, "epoch": 3.451388888888889, "grad_norm": 3.3992051611479863, "learning_rate": 4.1738346902475225e-06, "loss": 0.26019917, "memory(GiB)": 101.73, "step": 2485, "train_speed(iter/s)": 0.342127 }, { "acc": 0.90261774, "epoch": 3.4583333333333335, "grad_norm": 3.37600766474198, "learning_rate": 4.154968722432823e-06, "loss": 0.2450819, "memory(GiB)": 101.73, "step": 2490, "train_speed(iter/s)": 0.342165 }, { "acc": 0.89998322, "epoch": 3.4652777777777777, "grad_norm": 5.760093894583458, "learning_rate": 4.136115141277559e-06, "loss": 0.24672766, "memory(GiB)": 101.73, "step": 2495, "train_speed(iter/s)": 0.342183 }, { "acc": 0.89386778, "epoch": 3.4722222222222223, "grad_norm": 5.61319151711386, "learning_rate": 4.117274222978364e-06, "loss": 0.25242987, "memory(GiB)": 101.73, "step": 2500, "train_speed(iter/s)": 0.342213 }, { "acc": 0.90841942, "epoch": 3.4791666666666665, "grad_norm": 3.5190202823007595, "learning_rate": 4.098446243546361e-06, "loss": 0.22757564, "memory(GiB)": 101.73, "step": 2505, "train_speed(iter/s)": 0.342248 }, { "acc": 0.88912029, "epoch": 3.486111111111111, "grad_norm": 5.241339952219938, "learning_rate": 4.0796314788031235e-06, "loss": 0.27620914, "memory(GiB)": 101.73, "step": 2510, "train_speed(iter/s)": 0.342288 }, { "acc": 0.89256954, "epoch": 3.4930555555555554, "grad_norm": 7.171857875636105, "learning_rate": 4.060830204376641e-06, "loss": 0.26655228, "memory(GiB)": 101.73, "step": 2515, "train_speed(iter/s)": 0.342324 }, { "acc": 0.89562893, "epoch": 3.5, "grad_norm": 2.751263422359721, "learning_rate": 4.04204269569727e-06, "loss": 0.25634003, "memory(GiB)": 101.73, "step": 2520, "train_speed(iter/s)": 0.342361 }, { "acc": 0.90156956, "epoch": 3.5069444444444446, "grad_norm": 3.8262571455543024, "learning_rate": 4.023269227993709e-06, "loss": 0.2426084, "memory(GiB)": 101.73, "step": 2525, "train_speed(iter/s)": 0.34241 }, { "acc": 0.89592133, "epoch": 3.513888888888889, "grad_norm": 2.7239237491184, "learning_rate": 4.004510076288959e-06, "loss": 0.27324376, "memory(GiB)": 101.73, "step": 2530, "train_speed(iter/s)": 0.34243 }, { "acc": 0.90064278, "epoch": 3.5208333333333335, "grad_norm": 2.440659371146051, "learning_rate": 3.9857655153963e-06, "loss": 0.25882859, "memory(GiB)": 101.73, "step": 2535, "train_speed(iter/s)": 0.342409 }, { "acc": 0.90220985, "epoch": 3.5277777777777777, "grad_norm": 6.173071253776259, "learning_rate": 3.967035819915265e-06, "loss": 0.24151213, "memory(GiB)": 101.73, "step": 2540, "train_speed(iter/s)": 0.342436 }, { "acc": 0.89599705, "epoch": 3.5347222222222223, "grad_norm": 3.2005386510871596, "learning_rate": 3.9483212642276105e-06, "loss": 0.27071168, "memory(GiB)": 101.73, "step": 2545, "train_speed(iter/s)": 0.342476 }, { "acc": 0.89755955, "epoch": 3.5416666666666665, "grad_norm": 9.15052836433047, "learning_rate": 3.929622122493306e-06, "loss": 0.26333673, "memory(GiB)": 101.73, "step": 2550, "train_speed(iter/s)": 0.342504 }, { "acc": 0.89241247, "epoch": 3.548611111111111, "grad_norm": 2.8551585858243733, "learning_rate": 3.910938668646511e-06, "loss": 0.26134133, "memory(GiB)": 101.73, "step": 2555, "train_speed(iter/s)": 0.342537 }, { "acc": 0.90300541, "epoch": 3.5555555555555554, "grad_norm": 4.200634570714865, "learning_rate": 3.892271176391565e-06, "loss": 0.24932976, "memory(GiB)": 101.73, "step": 2560, "train_speed(iter/s)": 0.342576 }, { "acc": 0.90070772, "epoch": 3.5625, "grad_norm": 4.098223237386411, "learning_rate": 3.873619919198978e-06, "loss": 0.25703783, "memory(GiB)": 101.73, "step": 2565, "train_speed(iter/s)": 0.342659 }, { "acc": 0.89807205, "epoch": 3.5694444444444446, "grad_norm": 9.637838056707626, "learning_rate": 3.85498517030142e-06, "loss": 0.251478, "memory(GiB)": 101.73, "step": 2570, "train_speed(iter/s)": 0.342709 }, { "acc": 0.89927998, "epoch": 3.576388888888889, "grad_norm": 5.293025893334484, "learning_rate": 3.836367202689728e-06, "loss": 0.24165745, "memory(GiB)": 101.73, "step": 2575, "train_speed(iter/s)": 0.342741 }, { "acc": 0.90661182, "epoch": 3.5833333333333335, "grad_norm": 4.126703368737641, "learning_rate": 3.817766289108898e-06, "loss": 0.24157546, "memory(GiB)": 101.73, "step": 2580, "train_speed(iter/s)": 0.342775 }, { "acc": 0.89383621, "epoch": 3.5902777777777777, "grad_norm": 4.074196832478406, "learning_rate": 3.7991827020540883e-06, "loss": 0.25025237, "memory(GiB)": 101.73, "step": 2585, "train_speed(iter/s)": 0.342811 }, { "acc": 0.90283613, "epoch": 3.5972222222222223, "grad_norm": 7.616151830363701, "learning_rate": 3.7806167137666337e-06, "loss": 0.24257934, "memory(GiB)": 101.73, "step": 2590, "train_speed(iter/s)": 0.342896 }, { "acc": 0.9043417, "epoch": 3.6041666666666665, "grad_norm": 4.013042948818498, "learning_rate": 3.7620685962300583e-06, "loss": 0.22736707, "memory(GiB)": 101.73, "step": 2595, "train_speed(iter/s)": 0.342962 }, { "acc": 0.90302153, "epoch": 3.611111111111111, "grad_norm": 7.480361593683446, "learning_rate": 3.7435386211660825e-06, "loss": 0.25048995, "memory(GiB)": 101.73, "step": 2600, "train_speed(iter/s)": 0.343004 }, { "acc": 0.90426874, "epoch": 3.6180555555555554, "grad_norm": 3.14198386783033, "learning_rate": 3.7250270600306497e-06, "loss": 0.23340836, "memory(GiB)": 101.73, "step": 2605, "train_speed(iter/s)": 0.343045 }, { "acc": 0.90514717, "epoch": 3.625, "grad_norm": 4.930192268094544, "learning_rate": 3.7065341840099483e-06, "loss": 0.24507568, "memory(GiB)": 101.73, "step": 2610, "train_speed(iter/s)": 0.343078 }, { "acc": 0.91004333, "epoch": 3.6319444444444446, "grad_norm": 5.405558679120437, "learning_rate": 3.6880602640164364e-06, "loss": 0.23129289, "memory(GiB)": 101.73, "step": 2615, "train_speed(iter/s)": 0.34312 }, { "acc": 0.90500717, "epoch": 3.638888888888889, "grad_norm": 6.283844606464733, "learning_rate": 3.6696055706848732e-06, "loss": 0.24272099, "memory(GiB)": 101.73, "step": 2620, "train_speed(iter/s)": 0.34316 }, { "acc": 0.88391743, "epoch": 3.6458333333333335, "grad_norm": 5.93726524128595, "learning_rate": 3.6511703743683613e-06, "loss": 0.26117985, "memory(GiB)": 101.73, "step": 2625, "train_speed(iter/s)": 0.343217 }, { "acc": 0.90749083, "epoch": 3.6527777777777777, "grad_norm": 4.149177960068219, "learning_rate": 3.632754945134376e-06, "loss": 0.24308465, "memory(GiB)": 101.73, "step": 2630, "train_speed(iter/s)": 0.343259 }, { "acc": 0.90750189, "epoch": 3.6597222222222223, "grad_norm": 7.101346023538212, "learning_rate": 3.6143595527608133e-06, "loss": 0.21127968, "memory(GiB)": 101.73, "step": 2635, "train_speed(iter/s)": 0.343293 }, { "acc": 0.89547405, "epoch": 3.6666666666666665, "grad_norm": 5.390934613472564, "learning_rate": 3.5959844667320416e-06, "loss": 0.27649164, "memory(GiB)": 101.73, "step": 2640, "train_speed(iter/s)": 0.343339 }, { "acc": 0.89665613, "epoch": 3.673611111111111, "grad_norm": 3.174570568300612, "learning_rate": 3.577629956234946e-06, "loss": 0.2634002, "memory(GiB)": 101.73, "step": 2645, "train_speed(iter/s)": 0.34339 }, { "acc": 0.89330215, "epoch": 3.6805555555555554, "grad_norm": 4.468851842008194, "learning_rate": 3.5592962901549925e-06, "loss": 0.25232086, "memory(GiB)": 101.73, "step": 2650, "train_speed(iter/s)": 0.343396 }, { "acc": 0.90548496, "epoch": 3.6875, "grad_norm": 3.8524850872461407, "learning_rate": 3.5409837370722865e-06, "loss": 0.22683721, "memory(GiB)": 101.73, "step": 2655, "train_speed(iter/s)": 0.343441 }, { "acc": 0.9049942, "epoch": 3.6944444444444446, "grad_norm": 4.301665754833759, "learning_rate": 3.5226925652576337e-06, "loss": 0.25169647, "memory(GiB)": 101.73, "step": 2660, "train_speed(iter/s)": 0.34344 }, { "acc": 0.90362396, "epoch": 3.701388888888889, "grad_norm": 3.975117604181611, "learning_rate": 3.504423042668615e-06, "loss": 0.23812282, "memory(GiB)": 101.73, "step": 2665, "train_speed(iter/s)": 0.343477 }, { "acc": 0.90102139, "epoch": 3.7083333333333335, "grad_norm": 4.731994289393552, "learning_rate": 3.486175436945659e-06, "loss": 0.26122446, "memory(GiB)": 101.73, "step": 2670, "train_speed(iter/s)": 0.343463 }, { "acc": 0.8942503, "epoch": 3.7152777777777777, "grad_norm": 3.740054982289041, "learning_rate": 3.4679500154081264e-06, "loss": 0.25515299, "memory(GiB)": 101.73, "step": 2675, "train_speed(iter/s)": 0.343461 }, { "acc": 0.90634022, "epoch": 3.7222222222222223, "grad_norm": 3.1762395131780345, "learning_rate": 3.449747045050383e-06, "loss": 0.22432833, "memory(GiB)": 101.73, "step": 2680, "train_speed(iter/s)": 0.343507 }, { "acc": 0.90033245, "epoch": 3.7291666666666665, "grad_norm": 1.4522470812039046, "learning_rate": 3.4315667925378982e-06, "loss": 0.23518414, "memory(GiB)": 101.73, "step": 2685, "train_speed(iter/s)": 0.343527 }, { "acc": 0.90143833, "epoch": 3.736111111111111, "grad_norm": 3.1122492784066718, "learning_rate": 3.4134095242033354e-06, "loss": 0.25242925, "memory(GiB)": 101.73, "step": 2690, "train_speed(iter/s)": 0.343516 }, { "acc": 0.90106049, "epoch": 3.7430555555555554, "grad_norm": 3.6040857168011065, "learning_rate": 3.395275506042648e-06, "loss": 0.23500984, "memory(GiB)": 101.73, "step": 2695, "train_speed(iter/s)": 0.343541 }, { "acc": 0.89810867, "epoch": 3.75, "grad_norm": 2.726266348660559, "learning_rate": 3.377165003711185e-06, "loss": 0.26144593, "memory(GiB)": 101.73, "step": 2700, "train_speed(iter/s)": 0.343574 }, { "acc": 0.89270897, "epoch": 3.7569444444444446, "grad_norm": 3.5864113282789534, "learning_rate": 3.359078282519802e-06, "loss": 0.25928802, "memory(GiB)": 101.73, "step": 2705, "train_speed(iter/s)": 0.343605 }, { "acc": 0.90779352, "epoch": 3.763888888888889, "grad_norm": 3.4821077640595215, "learning_rate": 3.341015607430968e-06, "loss": 0.23630223, "memory(GiB)": 101.73, "step": 2710, "train_speed(iter/s)": 0.343634 }, { "acc": 0.88402081, "epoch": 3.7708333333333335, "grad_norm": 5.402082590204051, "learning_rate": 3.3229772430548872e-06, "loss": 0.2842896, "memory(GiB)": 101.73, "step": 2715, "train_speed(iter/s)": 0.343644 }, { "acc": 0.89697504, "epoch": 3.7777777777777777, "grad_norm": 3.1203680320024945, "learning_rate": 3.304963453645624e-06, "loss": 0.24786057, "memory(GiB)": 101.73, "step": 2720, "train_speed(iter/s)": 0.343653 }, { "acc": 0.89681606, "epoch": 3.7847222222222223, "grad_norm": 4.784880833886988, "learning_rate": 3.28697450309723e-06, "loss": 0.24739251, "memory(GiB)": 101.73, "step": 2725, "train_speed(iter/s)": 0.343679 }, { "acc": 0.90272732, "epoch": 3.7916666666666665, "grad_norm": 3.354351512168527, "learning_rate": 3.2690106549398786e-06, "loss": 0.24098086, "memory(GiB)": 101.73, "step": 2730, "train_speed(iter/s)": 0.343663 }, { "acc": 0.89669085, "epoch": 3.798611111111111, "grad_norm": 4.562914813279376, "learning_rate": 3.2510721723360044e-06, "loss": 0.24971335, "memory(GiB)": 101.73, "step": 2735, "train_speed(iter/s)": 0.343704 }, { "acc": 0.89800167, "epoch": 3.8055555555555554, "grad_norm": 4.232092460708063, "learning_rate": 3.233159318076448e-06, "loss": 0.24750371, "memory(GiB)": 101.73, "step": 2740, "train_speed(iter/s)": 0.343708 }, { "acc": 0.91989727, "epoch": 3.8125, "grad_norm": 4.324705162370445, "learning_rate": 3.2152723545766056e-06, "loss": 0.19653138, "memory(GiB)": 101.73, "step": 2745, "train_speed(iter/s)": 0.343737 }, { "acc": 0.90480156, "epoch": 3.8194444444444446, "grad_norm": 4.094118128093458, "learning_rate": 3.197411543872585e-06, "loss": 0.25692635, "memory(GiB)": 101.73, "step": 2750, "train_speed(iter/s)": 0.34372 }, { "acc": 0.89521971, "epoch": 3.826388888888889, "grad_norm": 7.01995940066227, "learning_rate": 3.1795771476173653e-06, "loss": 0.2531599, "memory(GiB)": 101.73, "step": 2755, "train_speed(iter/s)": 0.343745 }, { "acc": 0.90467262, "epoch": 3.8333333333333335, "grad_norm": 3.9849535755868195, "learning_rate": 3.1617694270769713e-06, "loss": 0.23331397, "memory(GiB)": 101.73, "step": 2760, "train_speed(iter/s)": 0.343762 }, { "acc": 0.91151018, "epoch": 3.8402777777777777, "grad_norm": 3.6084761076635457, "learning_rate": 3.1439886431266347e-06, "loss": 0.21082807, "memory(GiB)": 101.73, "step": 2765, "train_speed(iter/s)": 0.34378 }, { "acc": 0.90110893, "epoch": 3.8472222222222223, "grad_norm": 4.76524514339417, "learning_rate": 3.1262350562469808e-06, "loss": 0.25923443, "memory(GiB)": 101.73, "step": 2770, "train_speed(iter/s)": 0.343818 }, { "acc": 0.89796429, "epoch": 3.8541666666666665, "grad_norm": 3.9287144228778024, "learning_rate": 3.1085089265202095e-06, "loss": 0.24114075, "memory(GiB)": 101.73, "step": 2775, "train_speed(iter/s)": 0.343812 }, { "acc": 0.9085268, "epoch": 3.861111111111111, "grad_norm": 3.0751701860282314, "learning_rate": 3.0908105136262826e-06, "loss": 0.22855995, "memory(GiB)": 101.73, "step": 2780, "train_speed(iter/s)": 0.343845 }, { "acc": 0.90738831, "epoch": 3.8680555555555554, "grad_norm": 3.1413952239104104, "learning_rate": 3.073140076839129e-06, "loss": 0.23103039, "memory(GiB)": 101.73, "step": 2785, "train_speed(iter/s)": 0.343876 }, { "acc": 0.90736532, "epoch": 3.875, "grad_norm": 2.989880541422952, "learning_rate": 3.055497875022835e-06, "loss": 0.23236108, "memory(GiB)": 101.73, "step": 2790, "train_speed(iter/s)": 0.3439 }, { "acc": 0.91404114, "epoch": 3.8819444444444446, "grad_norm": 10.059317904857696, "learning_rate": 3.0378841666278565e-06, "loss": 0.22054443, "memory(GiB)": 101.73, "step": 2795, "train_speed(iter/s)": 0.343934 }, { "acc": 0.91268139, "epoch": 3.888888888888889, "grad_norm": 3.663419727544536, "learning_rate": 3.020299209687237e-06, "loss": 0.21329353, "memory(GiB)": 101.73, "step": 2800, "train_speed(iter/s)": 0.343851 }, { "acc": 0.90505514, "epoch": 3.8958333333333335, "grad_norm": 3.7817002877666255, "learning_rate": 3.0027432618128194e-06, "loss": 0.24037421, "memory(GiB)": 101.73, "step": 2805, "train_speed(iter/s)": 0.34389 }, { "acc": 0.90936232, "epoch": 3.9027777777777777, "grad_norm": 3.505578498849252, "learning_rate": 2.9852165801914818e-06, "loss": 0.21331539, "memory(GiB)": 101.73, "step": 2810, "train_speed(iter/s)": 0.343878 }, { "acc": 0.90463762, "epoch": 3.9097222222222223, "grad_norm": 5.035890509179841, "learning_rate": 2.9677194215813628e-06, "loss": 0.2340332, "memory(GiB)": 101.73, "step": 2815, "train_speed(iter/s)": 0.343918 }, { "acc": 0.91466646, "epoch": 3.9166666666666665, "grad_norm": 4.901427224659314, "learning_rate": 2.950252042308099e-06, "loss": 0.22736809, "memory(GiB)": 101.73, "step": 2820, "train_speed(iter/s)": 0.343945 }, { "acc": 0.91305885, "epoch": 3.923611111111111, "grad_norm": 4.762675724207692, "learning_rate": 2.9328146982610757e-06, "loss": 0.21405044, "memory(GiB)": 101.73, "step": 2825, "train_speed(iter/s)": 0.343936 }, { "acc": 0.91384068, "epoch": 3.9305555555555554, "grad_norm": 4.4556363032336055, "learning_rate": 2.915407644889676e-06, "loss": 0.19862232, "memory(GiB)": 101.73, "step": 2830, "train_speed(iter/s)": 0.343963 }, { "acc": 0.91425467, "epoch": 3.9375, "grad_norm": 4.48354142391771, "learning_rate": 2.898031137199535e-06, "loss": 0.22390165, "memory(GiB)": 101.73, "step": 2835, "train_speed(iter/s)": 0.343944 }, { "acc": 0.90526085, "epoch": 3.9444444444444446, "grad_norm": 4.984884813064181, "learning_rate": 2.880685429748809e-06, "loss": 0.23374908, "memory(GiB)": 101.73, "step": 2840, "train_speed(iter/s)": 0.343948 }, { "acc": 0.917663, "epoch": 3.951388888888889, "grad_norm": 3.1978275940506933, "learning_rate": 2.863370776644444e-06, "loss": 0.21864316, "memory(GiB)": 101.73, "step": 2845, "train_speed(iter/s)": 0.34398 }, { "acc": 0.90146713, "epoch": 3.9583333333333335, "grad_norm": 5.338078987116394, "learning_rate": 2.846087431538455e-06, "loss": 0.23189349, "memory(GiB)": 101.73, "step": 2850, "train_speed(iter/s)": 0.343856 }, { "acc": 0.90521212, "epoch": 3.9652777777777777, "grad_norm": 2.6961159090915525, "learning_rate": 2.8288356476242067e-06, "loss": 0.2326869, "memory(GiB)": 101.73, "step": 2855, "train_speed(iter/s)": 0.343864 }, { "acc": 0.90831709, "epoch": 3.9722222222222223, "grad_norm": 5.369767010977922, "learning_rate": 2.811615677632709e-06, "loss": 0.22528372, "memory(GiB)": 101.73, "step": 2860, "train_speed(iter/s)": 0.343895 }, { "acc": 0.91458817, "epoch": 3.9791666666666665, "grad_norm": 6.308934660163883, "learning_rate": 2.794427773828907e-06, "loss": 0.2268847, "memory(GiB)": 101.73, "step": 2865, "train_speed(iter/s)": 0.343899 }, { "acc": 0.91015339, "epoch": 3.986111111111111, "grad_norm": 3.6572394482166724, "learning_rate": 2.777272188007996e-06, "loss": 0.21870656, "memory(GiB)": 101.73, "step": 2870, "train_speed(iter/s)": 0.343924 }, { "acc": 0.91265316, "epoch": 3.9930555555555554, "grad_norm": 6.6960528916888675, "learning_rate": 2.7601491714917223e-06, "loss": 0.22325075, "memory(GiB)": 101.73, "step": 2875, "train_speed(iter/s)": 0.34395 }, { "acc": 0.90769653, "epoch": 4.0, "grad_norm": 3.8850794368257997, "learning_rate": 2.74305897512471e-06, "loss": 0.22835503, "memory(GiB)": 101.73, "step": 2880, "train_speed(iter/s)": 0.343939 }, { "epoch": 4.0, "eval_acc": 0.8967149619853679, "eval_loss": 0.26612961292266846, "eval_runtime": 24.7985, "eval_samples_per_second": 13.831, "eval_steps_per_second": 1.734, "step": 2880 }, { "acc": 0.92103558, "epoch": 4.006944444444445, "grad_norm": 5.906430563555609, "learning_rate": 2.726001849270778e-06, "loss": 0.19677427, "memory(GiB)": 101.73, "step": 2885, "train_speed(iter/s)": 0.341573 }, { "acc": 0.91259556, "epoch": 4.013888888888889, "grad_norm": 3.7879137611063816, "learning_rate": 2.7089780438092844e-06, "loss": 0.21481938, "memory(GiB)": 101.73, "step": 2890, "train_speed(iter/s)": 0.341624 }, { "acc": 0.9053381, "epoch": 4.020833333333333, "grad_norm": 6.231784630607876, "learning_rate": 2.6919878081314494e-06, "loss": 0.21756485, "memory(GiB)": 101.73, "step": 2895, "train_speed(iter/s)": 0.341666 }, { "acc": 0.91989698, "epoch": 4.027777777777778, "grad_norm": 6.530126465541856, "learning_rate": 2.675031391136717e-06, "loss": 0.20109849, "memory(GiB)": 101.73, "step": 2900, "train_speed(iter/s)": 0.34169 }, { "acc": 0.92158413, "epoch": 4.034722222222222, "grad_norm": 8.19113289280511, "learning_rate": 2.658109041229097e-06, "loss": 0.18494437, "memory(GiB)": 101.73, "step": 2905, "train_speed(iter/s)": 0.341747 }, { "acc": 0.91528072, "epoch": 4.041666666666667, "grad_norm": 5.196788929456166, "learning_rate": 2.641221006313542e-06, "loss": 0.2032238, "memory(GiB)": 101.73, "step": 2910, "train_speed(iter/s)": 0.341775 }, { "acc": 0.92303152, "epoch": 4.048611111111111, "grad_norm": 7.864716805294142, "learning_rate": 2.624367533792291e-06, "loss": 0.19096355, "memory(GiB)": 101.73, "step": 2915, "train_speed(iter/s)": 0.341785 }, { "acc": 0.91825953, "epoch": 4.055555555555555, "grad_norm": 5.335075751721026, "learning_rate": 2.607548870561271e-06, "loss": 0.21585865, "memory(GiB)": 101.73, "step": 2920, "train_speed(iter/s)": 0.341815 }, { "acc": 0.90057945, "epoch": 4.0625, "grad_norm": 4.833951672917176, "learning_rate": 2.5907652630064588e-06, "loss": 0.22249794, "memory(GiB)": 101.73, "step": 2925, "train_speed(iter/s)": 0.341826 }, { "acc": 0.91943073, "epoch": 4.069444444444445, "grad_norm": 8.019697338393213, "learning_rate": 2.5740169570002882e-06, "loss": 0.20015411, "memory(GiB)": 101.73, "step": 2930, "train_speed(iter/s)": 0.341804 }, { "acc": 0.91776352, "epoch": 4.076388888888889, "grad_norm": 8.514964900470892, "learning_rate": 2.55730419789804e-06, "loss": 0.19529152, "memory(GiB)": 101.73, "step": 2935, "train_speed(iter/s)": 0.341815 }, { "acc": 0.91629219, "epoch": 4.083333333333333, "grad_norm": 6.08663782319064, "learning_rate": 2.5406272305342438e-06, "loss": 0.20550578, "memory(GiB)": 101.73, "step": 2940, "train_speed(iter/s)": 0.341842 }, { "acc": 0.92428894, "epoch": 4.090277777777778, "grad_norm": 2.5039357125914274, "learning_rate": 2.5239862992191023e-06, "loss": 0.19628391, "memory(GiB)": 101.73, "step": 2945, "train_speed(iter/s)": 0.341877 }, { "acc": 0.91116905, "epoch": 4.097222222222222, "grad_norm": 7.318180727257995, "learning_rate": 2.507381647734899e-06, "loss": 0.21234784, "memory(GiB)": 101.73, "step": 2950, "train_speed(iter/s)": 0.341914 }, { "acc": 0.91188145, "epoch": 4.104166666666667, "grad_norm": 6.6612949734360765, "learning_rate": 2.490813519332442e-06, "loss": 0.20844579, "memory(GiB)": 101.73, "step": 2955, "train_speed(iter/s)": 0.341935 }, { "acc": 0.9142437, "epoch": 4.111111111111111, "grad_norm": 6.349873204038989, "learning_rate": 2.4742821567274825e-06, "loss": 0.20826674, "memory(GiB)": 101.73, "step": 2960, "train_speed(iter/s)": 0.34197 }, { "acc": 0.91423683, "epoch": 4.118055555555555, "grad_norm": 8.317774783727135, "learning_rate": 2.4577878020971822e-06, "loss": 0.22406006, "memory(GiB)": 101.73, "step": 2965, "train_speed(iter/s)": 0.342009 }, { "acc": 0.91010065, "epoch": 4.125, "grad_norm": 10.126802668666569, "learning_rate": 2.4413306970765393e-06, "loss": 0.22186904, "memory(GiB)": 101.73, "step": 2970, "train_speed(iter/s)": 0.342033 }, { "acc": 0.92021332, "epoch": 4.131944444444445, "grad_norm": 6.474786305104566, "learning_rate": 2.424911082754871e-06, "loss": 0.20578272, "memory(GiB)": 101.73, "step": 2975, "train_speed(iter/s)": 0.342068 }, { "acc": 0.91911068, "epoch": 4.138888888888889, "grad_norm": 6.152212251038718, "learning_rate": 2.4085291996722655e-06, "loss": 0.21330719, "memory(GiB)": 101.73, "step": 2980, "train_speed(iter/s)": 0.342098 }, { "acc": 0.90562267, "epoch": 4.145833333333333, "grad_norm": 6.434270724550823, "learning_rate": 2.3921852878160708e-06, "loss": 0.22529688, "memory(GiB)": 101.73, "step": 2985, "train_speed(iter/s)": 0.342124 }, { "acc": 0.92362442, "epoch": 4.152777777777778, "grad_norm": 4.610860381392509, "learning_rate": 2.375879586617367e-06, "loss": 0.18123511, "memory(GiB)": 101.73, "step": 2990, "train_speed(iter/s)": 0.342154 }, { "acc": 0.91461487, "epoch": 4.159722222222222, "grad_norm": 4.105012494189029, "learning_rate": 2.3596123349474727e-06, "loss": 0.1987551, "memory(GiB)": 101.73, "step": 2995, "train_speed(iter/s)": 0.342196 }, { "acc": 0.92359133, "epoch": 4.166666666666667, "grad_norm": 4.699484309075114, "learning_rate": 2.34338377111443e-06, "loss": 0.19864633, "memory(GiB)": 101.73, "step": 3000, "train_speed(iter/s)": 0.342246 }, { "acc": 0.9230299, "epoch": 4.173611111111111, "grad_norm": 7.962603744103441, "learning_rate": 2.327194132859526e-06, "loss": 0.20071533, "memory(GiB)": 101.73, "step": 3005, "train_speed(iter/s)": 0.342296 }, { "acc": 0.912883, "epoch": 4.180555555555555, "grad_norm": 6.7436017503812815, "learning_rate": 2.3110436573538076e-06, "loss": 0.21834412, "memory(GiB)": 101.73, "step": 3010, "train_speed(iter/s)": 0.342324 }, { "acc": 0.92828503, "epoch": 4.1875, "grad_norm": 4.260639944998096, "learning_rate": 2.2949325811945946e-06, "loss": 0.2033814, "memory(GiB)": 101.73, "step": 3015, "train_speed(iter/s)": 0.342358 }, { "acc": 0.92300644, "epoch": 4.194444444444445, "grad_norm": 2.91708878590044, "learning_rate": 2.2788611404020382e-06, "loss": 0.195956, "memory(GiB)": 101.73, "step": 3020, "train_speed(iter/s)": 0.342391 }, { "acc": 0.91175823, "epoch": 4.201388888888889, "grad_norm": 3.2469308089973485, "learning_rate": 2.262829570415637e-06, "loss": 0.2216738, "memory(GiB)": 101.73, "step": 3025, "train_speed(iter/s)": 0.342444 }, { "acc": 0.92101059, "epoch": 4.208333333333333, "grad_norm": 5.199265904450231, "learning_rate": 2.246838106090807e-06, "loss": 0.20876896, "memory(GiB)": 101.73, "step": 3030, "train_speed(iter/s)": 0.342479 }, { "acc": 0.92571077, "epoch": 4.215277777777778, "grad_norm": 5.1589147484364775, "learning_rate": 2.230886981695427e-06, "loss": 0.19227347, "memory(GiB)": 101.73, "step": 3035, "train_speed(iter/s)": 0.342511 }, { "acc": 0.91746311, "epoch": 4.222222222222222, "grad_norm": 4.523533624609451, "learning_rate": 2.2149764309064233e-06, "loss": 0.20252769, "memory(GiB)": 101.73, "step": 3040, "train_speed(iter/s)": 0.342529 }, { "acc": 0.92907982, "epoch": 4.229166666666667, "grad_norm": 4.484898357270931, "learning_rate": 2.1991066868063264e-06, "loss": 0.17676392, "memory(GiB)": 101.73, "step": 3045, "train_speed(iter/s)": 0.34255 }, { "acc": 0.92075233, "epoch": 4.236111111111111, "grad_norm": 6.798845521797008, "learning_rate": 2.183277981879876e-06, "loss": 0.19941406, "memory(GiB)": 101.73, "step": 3050, "train_speed(iter/s)": 0.342564 }, { "acc": 0.91766424, "epoch": 4.243055555555555, "grad_norm": 4.950483312812647, "learning_rate": 2.1674905480105994e-06, "loss": 0.21676636, "memory(GiB)": 101.73, "step": 3055, "train_speed(iter/s)": 0.342603 }, { "acc": 0.91581535, "epoch": 4.25, "grad_norm": 5.727602416491945, "learning_rate": 2.151744616477426e-06, "loss": 0.21338763, "memory(GiB)": 101.73, "step": 3060, "train_speed(iter/s)": 0.342631 }, { "acc": 0.9153821, "epoch": 4.256944444444445, "grad_norm": 8.155391373213389, "learning_rate": 2.1360404179512874e-06, "loss": 0.19959717, "memory(GiB)": 101.73, "step": 3065, "train_speed(iter/s)": 0.342676 }, { "acc": 0.91343565, "epoch": 4.263888888888889, "grad_norm": 7.534324022230486, "learning_rate": 2.120378182491756e-06, "loss": 0.21784058, "memory(GiB)": 101.73, "step": 3070, "train_speed(iter/s)": 0.342716 }, { "acc": 0.91763706, "epoch": 4.270833333333333, "grad_norm": 5.838501760901444, "learning_rate": 2.1047581395436537e-06, "loss": 0.20123901, "memory(GiB)": 101.73, "step": 3075, "train_speed(iter/s)": 0.34275 }, { "acc": 0.91569424, "epoch": 4.277777777777778, "grad_norm": 5.912931399886815, "learning_rate": 2.089180517933705e-06, "loss": 0.2089062, "memory(GiB)": 101.73, "step": 3080, "train_speed(iter/s)": 0.342768 }, { "acc": 0.91127586, "epoch": 4.284722222222222, "grad_norm": 6.321848607015057, "learning_rate": 2.0736455458671825e-06, "loss": 0.21258488, "memory(GiB)": 101.73, "step": 3085, "train_speed(iter/s)": 0.342756 }, { "acc": 0.90627985, "epoch": 4.291666666666667, "grad_norm": 4.4561601387109135, "learning_rate": 2.0581534509245563e-06, "loss": 0.2243315, "memory(GiB)": 101.73, "step": 3090, "train_speed(iter/s)": 0.34278 }, { "acc": 0.91781731, "epoch": 4.298611111111111, "grad_norm": 4.589217713951676, "learning_rate": 2.0427044600581726e-06, "loss": 0.19114343, "memory(GiB)": 101.73, "step": 3095, "train_speed(iter/s)": 0.342801 }, { "acc": 0.92404423, "epoch": 4.305555555555555, "grad_norm": 5.960490626167477, "learning_rate": 2.0272987995889132e-06, "loss": 0.20025024, "memory(GiB)": 101.73, "step": 3100, "train_speed(iter/s)": 0.342852 }, { "acc": 0.91189528, "epoch": 4.3125, "grad_norm": 3.4901104254439446, "learning_rate": 2.011936695202899e-06, "loss": 0.20672469, "memory(GiB)": 101.73, "step": 3105, "train_speed(iter/s)": 0.342859 }, { "acc": 0.92345657, "epoch": 4.319444444444445, "grad_norm": 5.200062911542391, "learning_rate": 1.9966183719481635e-06, "loss": 0.19884933, "memory(GiB)": 101.73, "step": 3110, "train_speed(iter/s)": 0.342894 }, { "acc": 0.93076115, "epoch": 4.326388888888889, "grad_norm": 4.310948513552987, "learning_rate": 1.9813440542313746e-06, "loss": 0.18369787, "memory(GiB)": 101.73, "step": 3115, "train_speed(iter/s)": 0.342939 }, { "acc": 0.92673454, "epoch": 4.333333333333333, "grad_norm": 5.518304658975087, "learning_rate": 1.9661139658145288e-06, "loss": 0.19188232, "memory(GiB)": 101.73, "step": 3120, "train_speed(iter/s)": 0.342888 }, { "acc": 0.93909798, "epoch": 4.340277777777778, "grad_norm": 4.7517608824503474, "learning_rate": 1.9509283298116973e-06, "loss": 0.16544604, "memory(GiB)": 101.73, "step": 3125, "train_speed(iter/s)": 0.342917 }, { "acc": 0.91507568, "epoch": 4.347222222222222, "grad_norm": 9.923491852945716, "learning_rate": 1.935787368685728e-06, "loss": 0.1952302, "memory(GiB)": 101.73, "step": 3130, "train_speed(iter/s)": 0.342943 }, { "acc": 0.91850033, "epoch": 4.354166666666667, "grad_norm": 6.0061062470300115, "learning_rate": 1.9206913042450114e-06, "loss": 0.21029849, "memory(GiB)": 101.73, "step": 3135, "train_speed(iter/s)": 0.342954 }, { "acc": 0.91360073, "epoch": 4.361111111111111, "grad_norm": 10.312396157334698, "learning_rate": 1.9056403576402159e-06, "loss": 0.22273135, "memory(GiB)": 101.73, "step": 3140, "train_speed(iter/s)": 0.342984 }, { "acc": 0.92476015, "epoch": 4.368055555555555, "grad_norm": 4.78373838533618, "learning_rate": 1.890634749361058e-06, "loss": 0.18518678, "memory(GiB)": 101.73, "step": 3145, "train_speed(iter/s)": 0.343 }, { "acc": 0.91752625, "epoch": 4.375, "grad_norm": 7.091436711506266, "learning_rate": 1.8756746992330627e-06, "loss": 0.20546875, "memory(GiB)": 101.73, "step": 3150, "train_speed(iter/s)": 0.343011 }, { "acc": 0.90989933, "epoch": 4.381944444444445, "grad_norm": 4.850150236316559, "learning_rate": 1.8607604264143531e-06, "loss": 0.21127014, "memory(GiB)": 101.73, "step": 3155, "train_speed(iter/s)": 0.343013 }, { "acc": 0.93822994, "epoch": 4.388888888888889, "grad_norm": 5.46370191663923, "learning_rate": 1.8458921493924342e-06, "loss": 0.14929545, "memory(GiB)": 101.73, "step": 3160, "train_speed(iter/s)": 0.34303 }, { "acc": 0.91957836, "epoch": 4.395833333333333, "grad_norm": 5.524837271666093, "learning_rate": 1.831070085980988e-06, "loss": 0.19637758, "memory(GiB)": 101.73, "step": 3165, "train_speed(iter/s)": 0.343038 }, { "acc": 0.92123384, "epoch": 4.402777777777778, "grad_norm": 7.926793678931113, "learning_rate": 1.8162944533166954e-06, "loss": 0.19701744, "memory(GiB)": 101.73, "step": 3170, "train_speed(iter/s)": 0.342961 }, { "acc": 0.92286034, "epoch": 4.409722222222222, "grad_norm": 7.847698734973718, "learning_rate": 1.8015654678560371e-06, "loss": 0.19722157, "memory(GiB)": 101.73, "step": 3175, "train_speed(iter/s)": 0.342964 }, { "acc": 0.91846771, "epoch": 4.416666666666667, "grad_norm": 11.602785837281019, "learning_rate": 1.7868833453721465e-06, "loss": 0.20651245, "memory(GiB)": 101.73, "step": 3180, "train_speed(iter/s)": 0.342968 }, { "acc": 0.92178154, "epoch": 4.423611111111111, "grad_norm": 8.927190820316273, "learning_rate": 1.7722483009516225e-06, "loss": 0.19756188, "memory(GiB)": 101.73, "step": 3185, "train_speed(iter/s)": 0.342953 }, { "acc": 0.92359142, "epoch": 4.430555555555555, "grad_norm": 7.0620744798059025, "learning_rate": 1.7576605489913989e-06, "loss": 0.19887724, "memory(GiB)": 101.73, "step": 3190, "train_speed(iter/s)": 0.342998 }, { "acc": 0.9229085, "epoch": 4.4375, "grad_norm": 5.903111514033937, "learning_rate": 1.7431203031955906e-06, "loss": 0.19516015, "memory(GiB)": 101.73, "step": 3195, "train_speed(iter/s)": 0.34304 }, { "acc": 0.92671881, "epoch": 4.444444444444445, "grad_norm": 3.573204950885832, "learning_rate": 1.7286277765723736e-06, "loss": 0.18694823, "memory(GiB)": 101.73, "step": 3200, "train_speed(iter/s)": 0.343078 }, { "acc": 0.92473421, "epoch": 4.451388888888889, "grad_norm": 4.159136138411481, "learning_rate": 1.7141831814308532e-06, "loss": 0.19077295, "memory(GiB)": 101.73, "step": 3205, "train_speed(iter/s)": 0.34308 }, { "acc": 0.91383629, "epoch": 4.458333333333333, "grad_norm": 7.8697493954658455, "learning_rate": 1.6997867293779663e-06, "loss": 0.23260565, "memory(GiB)": 101.73, "step": 3210, "train_speed(iter/s)": 0.343117 }, { "acc": 0.92381392, "epoch": 4.465277777777778, "grad_norm": 6.152018205562397, "learning_rate": 1.6854386313153674e-06, "loss": 0.19314412, "memory(GiB)": 101.73, "step": 3215, "train_speed(iter/s)": 0.343116 }, { "acc": 0.92322865, "epoch": 4.472222222222222, "grad_norm": 5.385841801425564, "learning_rate": 1.6711390974363546e-06, "loss": 0.19609743, "memory(GiB)": 101.73, "step": 3220, "train_speed(iter/s)": 0.343166 }, { "acc": 0.92354794, "epoch": 4.479166666666667, "grad_norm": 4.4092587657079, "learning_rate": 1.656888337222774e-06, "loss": 0.17454972, "memory(GiB)": 101.73, "step": 3225, "train_speed(iter/s)": 0.343198 }, { "acc": 0.91775503, "epoch": 4.486111111111111, "grad_norm": 4.013579134507486, "learning_rate": 1.6426865594419644e-06, "loss": 0.19643557, "memory(GiB)": 101.73, "step": 3230, "train_speed(iter/s)": 0.343239 }, { "acc": 0.93061867, "epoch": 4.493055555555555, "grad_norm": 5.6294114253186756, "learning_rate": 1.6285339721436927e-06, "loss": 0.17754095, "memory(GiB)": 101.73, "step": 3235, "train_speed(iter/s)": 0.343269 }, { "acc": 0.92582932, "epoch": 4.5, "grad_norm": 5.737199620676354, "learning_rate": 1.6144307826571068e-06, "loss": 0.18249105, "memory(GiB)": 101.73, "step": 3240, "train_speed(iter/s)": 0.343266 }, { "acc": 0.92780704, "epoch": 4.506944444444445, "grad_norm": 5.65833700507881, "learning_rate": 1.6003771975876987e-06, "loss": 0.17639775, "memory(GiB)": 101.73, "step": 3245, "train_speed(iter/s)": 0.343315 }, { "acc": 0.92351227, "epoch": 4.513888888888889, "grad_norm": 6.25866794334907, "learning_rate": 1.5863734228142751e-06, "loss": 0.19747633, "memory(GiB)": 101.73, "step": 3250, "train_speed(iter/s)": 0.343343 }, { "acc": 0.92094841, "epoch": 4.520833333333333, "grad_norm": 6.011894829639706, "learning_rate": 1.572419663485948e-06, "loss": 0.20139146, "memory(GiB)": 101.73, "step": 3255, "train_speed(iter/s)": 0.343337 }, { "acc": 0.92008724, "epoch": 4.527777777777778, "grad_norm": 5.54010697697139, "learning_rate": 1.5585161240191207e-06, "loss": 0.1791935, "memory(GiB)": 101.73, "step": 3260, "train_speed(iter/s)": 0.343371 }, { "acc": 0.9245883, "epoch": 4.534722222222222, "grad_norm": 6.420422985273098, "learning_rate": 1.5446630080945027e-06, "loss": 0.18729727, "memory(GiB)": 101.73, "step": 3265, "train_speed(iter/s)": 0.343372 }, { "acc": 0.92342873, "epoch": 4.541666666666667, "grad_norm": 4.290515420275203, "learning_rate": 1.5308605186541159e-06, "loss": 0.18176882, "memory(GiB)": 101.73, "step": 3270, "train_speed(iter/s)": 0.343363 }, { "acc": 0.9153471, "epoch": 4.548611111111111, "grad_norm": 6.6889348025132795, "learning_rate": 1.5171088578983317e-06, "loss": 0.19240036, "memory(GiB)": 101.73, "step": 3275, "train_speed(iter/s)": 0.343414 }, { "acc": 0.93120098, "epoch": 4.555555555555555, "grad_norm": 6.708275992166459, "learning_rate": 1.503408227282898e-06, "loss": 0.18204656, "memory(GiB)": 101.73, "step": 3280, "train_speed(iter/s)": 0.343446 }, { "acc": 0.91508112, "epoch": 4.5625, "grad_norm": 7.03757740085225, "learning_rate": 1.489758827515998e-06, "loss": 0.19290162, "memory(GiB)": 101.73, "step": 3285, "train_speed(iter/s)": 0.343441 }, { "acc": 0.92076712, "epoch": 4.569444444444445, "grad_norm": 5.257125987275666, "learning_rate": 1.4761608585553047e-06, "loss": 0.18960032, "memory(GiB)": 101.73, "step": 3290, "train_speed(iter/s)": 0.343449 }, { "acc": 0.9261817, "epoch": 4.576388888888889, "grad_norm": 7.586985807718556, "learning_rate": 1.4626145196050539e-06, "loss": 0.18427305, "memory(GiB)": 101.73, "step": 3295, "train_speed(iter/s)": 0.343448 }, { "acc": 0.93572044, "epoch": 4.583333333333333, "grad_norm": 5.174975068258436, "learning_rate": 1.4491200091131203e-06, "loss": 0.17530704, "memory(GiB)": 101.73, "step": 3300, "train_speed(iter/s)": 0.343476 }, { "acc": 0.92142067, "epoch": 4.590277777777778, "grad_norm": 2.7766176350174474, "learning_rate": 1.4356775247681192e-06, "loss": 0.1819458, "memory(GiB)": 101.73, "step": 3305, "train_speed(iter/s)": 0.343469 }, { "acc": 0.92758675, "epoch": 4.597222222222222, "grad_norm": 8.424408900106021, "learning_rate": 1.422287263496501e-06, "loss": 0.1895298, "memory(GiB)": 101.73, "step": 3310, "train_speed(iter/s)": 0.343506 }, { "acc": 0.93491707, "epoch": 4.604166666666667, "grad_norm": 9.687761901503304, "learning_rate": 1.408949421459676e-06, "loss": 0.18075891, "memory(GiB)": 101.73, "step": 3315, "train_speed(iter/s)": 0.343512 }, { "acc": 0.9381012, "epoch": 4.611111111111111, "grad_norm": 5.905868947420923, "learning_rate": 1.3956641940511347e-06, "loss": 0.16116495, "memory(GiB)": 101.73, "step": 3320, "train_speed(iter/s)": 0.343525 }, { "acc": 0.92054367, "epoch": 4.618055555555555, "grad_norm": 6.1885747944492735, "learning_rate": 1.3824317758935832e-06, "loss": 0.19729955, "memory(GiB)": 101.73, "step": 3325, "train_speed(iter/s)": 0.343538 }, { "acc": 0.92909966, "epoch": 4.625, "grad_norm": 3.7114859695136033, "learning_rate": 1.369252360836101e-06, "loss": 0.18774183, "memory(GiB)": 101.73, "step": 3330, "train_speed(iter/s)": 0.343562 }, { "acc": 0.93170919, "epoch": 4.631944444444445, "grad_norm": 6.9177632464094865, "learning_rate": 1.3561261419512895e-06, "loss": 0.17090454, "memory(GiB)": 101.73, "step": 3335, "train_speed(iter/s)": 0.343532 }, { "acc": 0.92585773, "epoch": 4.638888888888889, "grad_norm": 6.828565679411712, "learning_rate": 1.343053311532454e-06, "loss": 0.18238465, "memory(GiB)": 101.73, "step": 3340, "train_speed(iter/s)": 0.343557 }, { "acc": 0.91903534, "epoch": 4.645833333333333, "grad_norm": 5.367138133814305, "learning_rate": 1.3300340610907807e-06, "loss": 0.19298458, "memory(GiB)": 101.73, "step": 3345, "train_speed(iter/s)": 0.343583 }, { "acc": 0.91593809, "epoch": 4.652777777777778, "grad_norm": 4.388889958742242, "learning_rate": 1.3170685813525359e-06, "loss": 0.19910928, "memory(GiB)": 101.73, "step": 3350, "train_speed(iter/s)": 0.343596 }, { "acc": 0.9239089, "epoch": 4.659722222222222, "grad_norm": 7.979392327032548, "learning_rate": 1.3041570622562624e-06, "loss": 0.18677392, "memory(GiB)": 101.73, "step": 3355, "train_speed(iter/s)": 0.343626 }, { "acc": 0.92999897, "epoch": 4.666666666666667, "grad_norm": 23.61376185166529, "learning_rate": 1.2912996929500105e-06, "loss": 0.18426642, "memory(GiB)": 101.73, "step": 3360, "train_speed(iter/s)": 0.343637 }, { "acc": 0.93519926, "epoch": 4.673611111111111, "grad_norm": 6.4377028887779915, "learning_rate": 1.2784966617885538e-06, "loss": 0.17984772, "memory(GiB)": 101.73, "step": 3365, "train_speed(iter/s)": 0.343666 }, { "acc": 0.93129597, "epoch": 4.680555555555555, "grad_norm": 6.124547911246043, "learning_rate": 1.2657481563306414e-06, "loss": 0.18247683, "memory(GiB)": 101.73, "step": 3370, "train_speed(iter/s)": 0.343693 }, { "acc": 0.92232943, "epoch": 4.6875, "grad_norm": 5.931941944106171, "learning_rate": 1.253054363336241e-06, "loss": 0.17325442, "memory(GiB)": 101.73, "step": 3375, "train_speed(iter/s)": 0.343734 }, { "acc": 0.93359146, "epoch": 4.694444444444445, "grad_norm": 7.753520923917529, "learning_rate": 1.2404154687638103e-06, "loss": 0.16628423, "memory(GiB)": 101.73, "step": 3380, "train_speed(iter/s)": 0.343729 }, { "acc": 0.93307018, "epoch": 4.701388888888889, "grad_norm": 7.53046348429032, "learning_rate": 1.2278316577675646e-06, "loss": 0.17817822, "memory(GiB)": 101.73, "step": 3385, "train_speed(iter/s)": 0.343758 }, { "acc": 0.93201313, "epoch": 4.708333333333333, "grad_norm": 9.025531908827059, "learning_rate": 1.2153031146947749e-06, "loss": 0.17697582, "memory(GiB)": 101.73, "step": 3390, "train_speed(iter/s)": 0.34378 }, { "acc": 0.93053436, "epoch": 4.715277777777778, "grad_norm": 7.021234732278172, "learning_rate": 1.2028300230830572e-06, "loss": 0.16738009, "memory(GiB)": 101.73, "step": 3395, "train_speed(iter/s)": 0.3438 }, { "acc": 0.93569164, "epoch": 4.722222222222222, "grad_norm": 15.616139180655933, "learning_rate": 1.1904125656576904e-06, "loss": 0.16758065, "memory(GiB)": 101.73, "step": 3400, "train_speed(iter/s)": 0.34383 }, { "acc": 0.92955055, "epoch": 4.729166666666667, "grad_norm": 6.701798595359445, "learning_rate": 1.1780509243289383e-06, "loss": 0.18826933, "memory(GiB)": 101.73, "step": 3405, "train_speed(iter/s)": 0.343858 }, { "acc": 0.92169542, "epoch": 4.736111111111111, "grad_norm": 7.2038512053503645, "learning_rate": 1.1657452801893784e-06, "loss": 0.19082313, "memory(GiB)": 101.73, "step": 3410, "train_speed(iter/s)": 0.343886 }, { "acc": 0.92840328, "epoch": 4.743055555555555, "grad_norm": 9.858417301395926, "learning_rate": 1.1534958135112606e-06, "loss": 0.18543091, "memory(GiB)": 101.73, "step": 3415, "train_speed(iter/s)": 0.343916 }, { "acc": 0.93052006, "epoch": 4.75, "grad_norm": 4.6042916720288565, "learning_rate": 1.141302703743854e-06, "loss": 0.17618411, "memory(GiB)": 101.73, "step": 3420, "train_speed(iter/s)": 0.34394 }, { "acc": 0.93616848, "epoch": 4.756944444444445, "grad_norm": 7.4778490207199075, "learning_rate": 1.1291661295108284e-06, "loss": 0.17016323, "memory(GiB)": 101.73, "step": 3425, "train_speed(iter/s)": 0.343962 }, { "acc": 0.93410454, "epoch": 4.763888888888889, "grad_norm": 10.22618327699479, "learning_rate": 1.1170862686076295e-06, "loss": 0.17755842, "memory(GiB)": 101.73, "step": 3430, "train_speed(iter/s)": 0.34399 }, { "acc": 0.92950439, "epoch": 4.770833333333333, "grad_norm": 6.508153021167871, "learning_rate": 1.1050632979988816e-06, "loss": 0.16775516, "memory(GiB)": 101.73, "step": 3435, "train_speed(iter/s)": 0.344006 }, { "acc": 0.93873663, "epoch": 4.777777777777778, "grad_norm": 6.399803664609742, "learning_rate": 1.0930973938157865e-06, "loss": 0.14967173, "memory(GiB)": 101.73, "step": 3440, "train_speed(iter/s)": 0.344032 }, { "acc": 0.93606501, "epoch": 4.784722222222222, "grad_norm": 9.143976403019813, "learning_rate": 1.081188731353554e-06, "loss": 0.17374879, "memory(GiB)": 101.73, "step": 3445, "train_speed(iter/s)": 0.344058 }, { "acc": 0.9375165, "epoch": 4.791666666666667, "grad_norm": 7.754683265166291, "learning_rate": 1.0693374850688204e-06, "loss": 0.15017707, "memory(GiB)": 101.73, "step": 3450, "train_speed(iter/s)": 0.344073 }, { "acc": 0.92773657, "epoch": 4.798611111111111, "grad_norm": 13.340523234777597, "learning_rate": 1.05754382857711e-06, "loss": 0.180286, "memory(GiB)": 101.73, "step": 3455, "train_speed(iter/s)": 0.344093 }, { "acc": 0.92894754, "epoch": 4.805555555555555, "grad_norm": 3.62871529288943, "learning_rate": 1.0458079346502735e-06, "loss": 0.17606659, "memory(GiB)": 101.73, "step": 3460, "train_speed(iter/s)": 0.344124 }, { "acc": 0.92550812, "epoch": 4.8125, "grad_norm": 7.453371763696296, "learning_rate": 1.0341299752139704e-06, "loss": 0.1886322, "memory(GiB)": 101.73, "step": 3465, "train_speed(iter/s)": 0.344152 }, { "acc": 0.92362499, "epoch": 4.819444444444445, "grad_norm": 6.774667028179097, "learning_rate": 1.0225101213451426e-06, "loss": 0.17885201, "memory(GiB)": 101.73, "step": 3470, "train_speed(iter/s)": 0.34418 }, { "acc": 0.9255188, "epoch": 4.826388888888889, "grad_norm": 6.197957042298724, "learning_rate": 1.0109485432695131e-06, "loss": 0.18616344, "memory(GiB)": 101.73, "step": 3475, "train_speed(iter/s)": 0.344207 }, { "acc": 0.93224564, "epoch": 4.833333333333333, "grad_norm": 6.644413934310758, "learning_rate": 9.994454103590919e-07, "loss": 0.16555772, "memory(GiB)": 101.73, "step": 3480, "train_speed(iter/s)": 0.34424 }, { "acc": 0.92744942, "epoch": 4.840277777777778, "grad_norm": 4.351916474227688, "learning_rate": 9.880008911296872e-07, "loss": 0.20068693, "memory(GiB)": 101.73, "step": 3485, "train_speed(iter/s)": 0.344267 }, { "acc": 0.9230197, "epoch": 4.847222222222222, "grad_norm": 9.964081878004457, "learning_rate": 9.7661515323845e-07, "loss": 0.17565002, "memory(GiB)": 101.73, "step": 3490, "train_speed(iter/s)": 0.344291 }, { "acc": 0.93543081, "epoch": 4.854166666666667, "grad_norm": 6.8395540650956805, "learning_rate": 9.652883634814041e-07, "loss": 0.17975049, "memory(GiB)": 101.73, "step": 3495, "train_speed(iter/s)": 0.344308 }, { "acc": 0.93403215, "epoch": 4.861111111111111, "grad_norm": 5.701715874653675, "learning_rate": 9.540206877910153e-07, "loss": 0.16942759, "memory(GiB)": 101.73, "step": 3500, "train_speed(iter/s)": 0.344335 }, { "acc": 0.92589064, "epoch": 4.868055555555555, "grad_norm": 7.704105561595292, "learning_rate": 9.428122912337475e-07, "loss": 0.17456805, "memory(GiB)": 101.73, "step": 3505, "train_speed(iter/s)": 0.344363 }, { "acc": 0.94150906, "epoch": 4.875, "grad_norm": 3.438431889948542, "learning_rate": 9.316633380076606e-07, "loss": 0.15039229, "memory(GiB)": 101.73, "step": 3510, "train_speed(iter/s)": 0.344396 }, { "acc": 0.93364429, "epoch": 4.881944444444445, "grad_norm": 5.602784744395754, "learning_rate": 9.205739914399868e-07, "loss": 0.16423728, "memory(GiB)": 101.73, "step": 3515, "train_speed(iter/s)": 0.34443 }, { "acc": 0.93492546, "epoch": 4.888888888888889, "grad_norm": 6.450634106440373, "learning_rate": 9.095444139847534e-07, "loss": 0.16065657, "memory(GiB)": 101.73, "step": 3520, "train_speed(iter/s)": 0.344454 }, { "acc": 0.9399188, "epoch": 4.895833333333333, "grad_norm": 7.621591927258962, "learning_rate": 8.985747672203926e-07, "loss": 0.16348395, "memory(GiB)": 101.73, "step": 3525, "train_speed(iter/s)": 0.34448 }, { "acc": 0.93001232, "epoch": 4.902777777777778, "grad_norm": 4.912482966701962, "learning_rate": 8.876652118473818e-07, "loss": 0.1870851, "memory(GiB)": 101.73, "step": 3530, "train_speed(iter/s)": 0.344504 }, { "acc": 0.92510281, "epoch": 4.909722222222222, "grad_norm": 6.857031735241769, "learning_rate": 8.768159076858827e-07, "loss": 0.18478702, "memory(GiB)": 101.73, "step": 3535, "train_speed(iter/s)": 0.344538 }, { "acc": 0.93646317, "epoch": 4.916666666666667, "grad_norm": 5.432910962320387, "learning_rate": 8.660270136734077e-07, "loss": 0.16789238, "memory(GiB)": 101.73, "step": 3540, "train_speed(iter/s)": 0.344567 }, { "acc": 0.92513866, "epoch": 4.923611111111111, "grad_norm": 14.001241817054742, "learning_rate": 8.552986878624815e-07, "loss": 0.17173724, "memory(GiB)": 101.73, "step": 3545, "train_speed(iter/s)": 0.344582 }, { "acc": 0.93663769, "epoch": 4.930555555555555, "grad_norm": 9.603326526795332, "learning_rate": 8.446310874183368e-07, "loss": 0.16478382, "memory(GiB)": 101.73, "step": 3550, "train_speed(iter/s)": 0.344601 }, { "acc": 0.93988123, "epoch": 4.9375, "grad_norm": 9.381389112980651, "learning_rate": 8.340243686166037e-07, "loss": 0.16926775, "memory(GiB)": 101.73, "step": 3555, "train_speed(iter/s)": 0.344638 }, { "acc": 0.93035545, "epoch": 4.944444444444445, "grad_norm": 7.110265271158136, "learning_rate": 8.234786868410236e-07, "loss": 0.16944885, "memory(GiB)": 101.73, "step": 3560, "train_speed(iter/s)": 0.344667 }, { "acc": 0.9245862, "epoch": 4.951388888888889, "grad_norm": 6.202865636689749, "learning_rate": 8.129941965811745e-07, "loss": 0.18254293, "memory(GiB)": 101.73, "step": 3565, "train_speed(iter/s)": 0.344693 }, { "acc": 0.9371048, "epoch": 4.958333333333333, "grad_norm": 11.61144011390846, "learning_rate": 8.025710514302019e-07, "loss": 0.16599008, "memory(GiB)": 101.73, "step": 3570, "train_speed(iter/s)": 0.344715 }, { "acc": 0.92540922, "epoch": 4.965277777777778, "grad_norm": 5.9675388419806294, "learning_rate": 7.922094040825753e-07, "loss": 0.17711488, "memory(GiB)": 101.73, "step": 3575, "train_speed(iter/s)": 0.344748 }, { "acc": 0.93437986, "epoch": 4.972222222222222, "grad_norm": 4.777475006904249, "learning_rate": 7.819094063318475e-07, "loss": 0.16441042, "memory(GiB)": 101.73, "step": 3580, "train_speed(iter/s)": 0.344783 }, { "acc": 0.92917442, "epoch": 4.979166666666667, "grad_norm": 9.660056870992797, "learning_rate": 7.716712090684326e-07, "loss": 0.18292786, "memory(GiB)": 101.73, "step": 3585, "train_speed(iter/s)": 0.344803 }, { "acc": 0.94042187, "epoch": 4.986111111111111, "grad_norm": 4.423451670274828, "learning_rate": 7.614949622773942e-07, "loss": 0.14605323, "memory(GiB)": 101.73, "step": 3590, "train_speed(iter/s)": 0.344817 }, { "acc": 0.93927879, "epoch": 4.993055555555555, "grad_norm": 8.31218232668258, "learning_rate": 7.5138081503625e-07, "loss": 0.15975434, "memory(GiB)": 101.73, "step": 3595, "train_speed(iter/s)": 0.344846 }, { "acc": 0.93250046, "epoch": 5.0, "grad_norm": 5.5702806703726075, "learning_rate": 7.413289155127845e-07, "loss": 0.16792555, "memory(GiB)": 101.73, "step": 3600, "train_speed(iter/s)": 0.34485 }, { "epoch": 5.0, "eval_acc": 0.9066131114617701, "eval_loss": 0.25880399346351624, "eval_runtime": 25.0627, "eval_samples_per_second": 13.686, "eval_steps_per_second": 1.716, "step": 3600 } ], "logging_steps": 5, "max_steps": 4320, "num_input_tokens_seen": 0, "num_train_epochs": 6, "save_steps": 200, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": false }, "attributes": {} } }, "total_flos": 1.9491571697188864e+17, "train_batch_size": 1, "trial_name": null, "trial_params": null }