[
  {
    "loss": 4.2828514099121096,
    "grad_norm": 3.3195607662200928,
    "learning_rate": 7.82608695652174e-05,
    "entropy": 1.1838731363415718,
    "mean_token_accuracy": 0.452268467284739,
    "num_tokens": 24645.0,
    "epoch": 0.13245033112582782,
    "step": 10
  },
  {
    "eval_loss": 0.5712648034095764,
    "eval_runtime": 8.7607,
    "eval_samples_per_second": 17.122,
    "eval_steps_per_second": 4.338,
    "eval_entropy": 0.5471524494258981,
    "eval_mean_token_accuracy": 0.8050595286645388,
    "eval_num_tokens": 24645.0,
    "epoch": 0.13245033112582782,
    "step": 10
  },
  {
    "loss": 0.46926422119140626,
    "grad_norm": 3.0499935150146484,
    "learning_rate": 0.00016521739130434784,
    "entropy": 0.4832126975059509,
    "mean_token_accuracy": 0.8076777458190918,
    "num_tokens": 48985.0,
    "epoch": 0.26490066225165565,
    "step": 20
  },
  {
    "eval_loss": 0.5358471870422363,
    "eval_runtime": 8.8684,
    "eval_samples_per_second": 16.914,
    "eval_steps_per_second": 4.285,
    "eval_entropy": 0.4200292041427211,
    "eval_mean_token_accuracy": 0.7469924860878995,
    "eval_num_tokens": 48985.0,
    "epoch": 0.26490066225165565,
    "step": 20
  },
  {
    "loss": 0.39472386837005613,
    "grad_norm": 2.067092180252075,
    "learning_rate": 0.00019414634146341464,
    "entropy": 0.3605545647442341,
    "mean_token_accuracy": 0.8380664870142936,
    "num_tokens": 73226.0,
    "epoch": 0.3973509933774834,
    "step": 30
  },
  {
    "eval_loss": 0.5321611166000366,
    "eval_runtime": 8.8949,
    "eval_samples_per_second": 16.864,
    "eval_steps_per_second": 4.272,
    "eval_entropy": 0.36939919249791847,
    "eval_mean_token_accuracy": 0.8029448617445795,
    "eval_num_tokens": 73226.0,
    "epoch": 0.3973509933774834,
    "step": 30
  },
  {
    "loss": 0.39537787437438965,
    "grad_norm": 1.270966649055481,
    "learning_rate": 0.00018439024390243903,
    "entropy": 0.3851592864841223,
    "mean_token_accuracy": 0.8292971581220627,
    "num_tokens": 97170.0,
    "epoch": 0.5298013245033113,
    "step": 40
  },
  {
    "eval_loss": 0.512241542339325,
    "eval_runtime": 8.856,
    "eval_samples_per_second": 16.938,
    "eval_steps_per_second": 4.291,
    "eval_entropy": 0.3763781021300115,
    "eval_mean_token_accuracy": 0.7811560207291653,
    "eval_num_tokens": 97170.0,
    "epoch": 0.5298013245033113,
    "step": 40
  },
  {
    "loss": 0.3449979066848755,
    "grad_norm": 1.178625226020813,
    "learning_rate": 0.00017463414634146342,
    "entropy": 0.3603811550885439,
    "mean_token_accuracy": 0.8657324358820915,
    "num_tokens": 121386.0,
    "epoch": 0.6622516556291391,
    "step": 50
  },
  {
    "eval_loss": 0.6070794463157654,
    "eval_runtime": 8.9125,
    "eval_samples_per_second": 16.83,
    "eval_steps_per_second": 4.264,
    "eval_entropy": 0.36585210116678163,
    "eval_mean_token_accuracy": 0.7500313332206324,
    "eval_num_tokens": 121386.0,
    "epoch": 0.6622516556291391,
    "step": 50
  },
  {
    "loss": 0.30911884307861326,
    "grad_norm": 2.8266186714172363,
    "learning_rate": 0.00016487804878048782,
    "entropy": 0.2615261286497116,
    "mean_token_accuracy": 0.8826999247074128,
    "num_tokens": 145580.0,
    "epoch": 0.7947019867549668,
    "step": 60
  },
  {
    "eval_loss": 0.7182475924491882,
    "eval_runtime": 8.8882,
    "eval_samples_per_second": 16.876,
    "eval_steps_per_second": 4.275,
    "eval_entropy": 0.2628652969081151,
    "eval_mean_token_accuracy": 0.7899749420191112,
    "eval_num_tokens": 145580.0,
    "epoch": 0.7947019867549668,
    "step": 60
  },
  {
    "loss": 0.24180982112884522,
    "grad_norm": 1.6042029857635498,
    "learning_rate": 0.0001551219512195122,
    "entropy": 0.2673916194587946,
    "mean_token_accuracy": 0.911859568953514,
    "num_tokens": 170643.0,
    "epoch": 0.9271523178807947,
    "step": 70
  },
  {
    "eval_loss": 0.821165919303894,
    "eval_runtime": 8.8971,
    "eval_samples_per_second": 16.859,
    "eval_steps_per_second": 4.271,
    "eval_entropy": 0.27867767418195544,
    "eval_mean_token_accuracy": 0.7433740669175198,
    "eval_num_tokens": 170643.0,
    "epoch": 0.9271523178807947,
    "step": 70
  },
  {
    "loss": 0.22081174850463867,
    "grad_norm": 2.7884373664855957,
    "learning_rate": 0.0001453658536585366,
    "entropy": 0.2026200779762707,
    "mean_token_accuracy": 0.9230157701592696,
    "num_tokens": 193746.0,
    "epoch": 1.0529801324503312,
    "step": 80
  },
  {
    "eval_loss": 0.8150429725646973,
    "eval_runtime": 8.9086,
    "eval_samples_per_second": 16.838,
    "eval_steps_per_second": 4.266,
    "eval_entropy": 0.22414806843000024,
    "eval_mean_token_accuracy": 0.8047775735980586,
    "eval_num_tokens": 193746.0,
    "epoch": 1.0529801324503312,
    "step": 80
  },
  {
    "loss": 0.14040154218673706,
    "grad_norm": 1.3541489839553833,
    "learning_rate": 0.000135609756097561,
    "entropy": 0.16504043247550726,
    "mean_token_accuracy": 0.9462000757455826,
    "num_tokens": 218576.0,
    "epoch": 1.185430463576159,
    "step": 90
  },
  {
    "eval_loss": 0.8339773416519165,
    "eval_runtime": 8.973,
    "eval_samples_per_second": 16.717,
    "eval_steps_per_second": 4.235,
    "eval_entropy": 0.2065272086060998,
    "eval_mean_token_accuracy": 0.7746240653489765,
    "eval_num_tokens": 218576.0,
    "epoch": 1.185430463576159,
    "step": 90
  },
  {
    "loss": 0.12467809915542602,
    "grad_norm": 4.511893272399902,
    "learning_rate": 0.00012585365853658536,
    "entropy": 0.10107915503322147,
    "mean_token_accuracy": 0.9587799295783043,
    "num_tokens": 243344.0,
    "epoch": 1.3178807947019868,
    "step": 100
  },
  {
    "eval_loss": 0.7112035155296326,
    "eval_runtime": 8.9282,
    "eval_samples_per_second": 16.801,
    "eval_steps_per_second": 4.256,
    "eval_entropy": 0.15474749381016745,
    "eval_mean_token_accuracy": 0.8157424864016081,
    "eval_num_tokens": 243344.0,
    "epoch": 1.3178807947019868,
    "step": 100
  },
  {
    "loss": 0.12197786569595337,
    "grad_norm": 1.6361427307128906,
    "learning_rate": 0.00011609756097560975,
    "entropy": 0.07562960358336568,
    "mean_token_accuracy": 0.9517370820045471,
    "num_tokens": 267786.0,
    "epoch": 1.4503311258278146,
    "step": 110
  },
  {
    "eval_loss": 1.0470056533813477,
    "eval_runtime": 8.9724,
    "eval_samples_per_second": 16.718,
    "eval_steps_per_second": 4.235,
    "eval_entropy": 0.14818870221681305,
    "eval_mean_token_accuracy": 0.7478383512873399,
    "eval_num_tokens": 267786.0,
    "epoch": 1.4503311258278146,
    "step": 110
  },
  {
    "loss": 0.10809319019317627,
    "grad_norm": 1.6775206327438354,
    "learning_rate": 0.00010634146341463416,
    "entropy": 0.08924289270071313,
    "mean_token_accuracy": 0.9604727029800415,
    "num_tokens": 292317.0,
    "epoch": 1.5827814569536423,
    "step": 120
  },
  {
    "eval_loss": 1.0492334365844727,
    "eval_runtime": 8.8734,
    "eval_samples_per_second": 16.904,
    "eval_steps_per_second": 4.282,
    "eval_entropy": 0.14410611607232376,
    "eval_mean_token_accuracy": 0.8007205567861858,
    "eval_num_tokens": 292317.0,
    "epoch": 1.5827814569536423,
    "step": 120
  },
  {
    "loss": 0.03908039927482605,
    "grad_norm": 2.2686848640441895,
    "learning_rate": 9.658536585365854e-05,
    "entropy": 0.06353283888893202,
    "mean_token_accuracy": 0.9815119966864586,
    "num_tokens": 316521.0,
    "epoch": 1.7152317880794703,
    "step": 130
  },
  {
    "eval_loss": 1.4305915832519531,
    "eval_runtime": 8.8602,
    "eval_samples_per_second": 16.93,
    "eval_steps_per_second": 4.289,
    "eval_entropy": 0.0935905459689246,
    "eval_mean_token_accuracy": 0.7892387258379083,
    "eval_num_tokens": 316521.0,
    "epoch": 1.7152317880794703,
    "step": 130
  },
  {
    "loss": 0.04100652039051056,
    "grad_norm": 5.218770980834961,
    "learning_rate": 8.682926829268293e-05,
    "entropy": 0.03381892999459524,
    "mean_token_accuracy": 0.984375512599945,
    "num_tokens": 340919.0,
    "epoch": 1.847682119205298,
    "step": 140
  },
  {
    "eval_loss": 1.5555816888809204,
    "eval_runtime": 8.7739,
    "eval_samples_per_second": 17.096,
    "eval_steps_per_second": 4.331,
    "eval_entropy": 0.07794101885673099,
    "eval_mean_token_accuracy": 0.8050282001495361,
    "eval_num_tokens": 340919.0,
    "epoch": 1.847682119205298,
    "step": 140
  },
  {
    "loss": 0.03830587565898895,
    "grad_norm": 0.13190802931785583,
    "learning_rate": 7.707317073170732e-05,
    "entropy": 0.026670162624213845,
    "mean_token_accuracy": 0.9837620735168457,
    "num_tokens": 364991.0,
    "epoch": 1.980132450331126,
    "step": 150
  },
  {
    "eval_loss": 1.2507771253585815,
    "eval_runtime": 8.8746,
    "eval_samples_per_second": 16.902,
    "eval_steps_per_second": 4.282,
    "eval_entropy": 0.0817602070930757,
    "eval_mean_token_accuracy": 0.7987468728893682,
    "eval_num_tokens": 364991.0,
    "epoch": 1.980132450331126,
    "step": 150
  },
  {
    "loss": 0.07177287340164185,
    "grad_norm": 2.2161507606506348,
    "learning_rate": 6.731707317073171e-05,
    "entropy": 0.01958650479240245,
    "mean_token_accuracy": 0.9884510479475322,
    "num_tokens": 387642.0,
    "epoch": 2.1059602649006623,
    "step": 160
  },
  {
    "eval_loss": 1.3437883853912354,
    "eval_runtime": 8.8478,
    "eval_samples_per_second": 16.953,
    "eval_steps_per_second": 4.295,
    "eval_entropy": 0.06764962662648606,
    "eval_mean_token_accuracy": 0.7787750677058571,
    "eval_num_tokens": 387642.0,
    "epoch": 2.1059602649006623,
    "step": 160
  },
  {
    "loss": 0.009113987535238266,
    "grad_norm": 0.039817001670598984,
    "learning_rate": 5.756097560975609e-05,
    "entropy": 0.01661088205728447,
    "mean_token_accuracy": 0.9937208637595176,
    "num_tokens": 412538.0,
    "epoch": 2.23841059602649,
    "step": 170
  },
  {
    "eval_loss": 1.5150843858718872,
    "eval_runtime": 8.7479,
    "eval_samples_per_second": 17.147,
    "eval_steps_per_second": 4.344,
    "eval_entropy": 0.06736165528357226,
    "eval_mean_token_accuracy": 0.7647556442963449,
    "eval_num_tokens": 412538.0,
    "epoch": 2.23841059602649,
    "step": 170
  },
  {
    "loss": 0.003525285795331001,
    "grad_norm": 0.00853405799716711,
    "learning_rate": 4.7804878048780485e-05,
    "entropy": 0.00830267872515833,
    "mean_token_accuracy": 0.9980769231915474,
    "num_tokens": 437055.0,
    "epoch": 2.370860927152318,
    "step": 180
  },
  {
    "eval_loss": 1.644209384918213,
    "eval_runtime": 8.7784,
    "eval_samples_per_second": 17.087,
    "eval_steps_per_second": 4.329,
    "eval_entropy": 0.06734515197259545,
    "eval_mean_token_accuracy": 0.7691416097314734,
    "eval_num_tokens": 437055.0,
    "epoch": 2.370860927152318,
    "step": 180
  },
  {
    "loss": 0.001157384179532528,
    "grad_norm": 0.09711030125617981,
    "learning_rate": 3.804878048780488e-05,
    "entropy": 0.004432732032000786,
    "mean_token_accuracy": 1.0,
    "num_tokens": 461493.0,
    "epoch": 2.5033112582781456,
    "step": 190
  },
  {
    "eval_loss": 1.7609623670578003,
    "eval_runtime": 8.7515,
    "eval_samples_per_second": 17.14,
    "eval_steps_per_second": 4.342,
    "eval_entropy": 0.06280427154553381,
    "eval_mean_token_accuracy": 0.7669486277981808,
    "eval_num_tokens": 461493.0,
    "epoch": 2.5033112582781456,
    "step": 190
  },
  {
    "loss": 0.0030027683824300766,
    "grad_norm": 0.012400347739458084,
    "learning_rate": 2.8292682926829267e-05,
    "entropy": 0.004392967079911614,
    "mean_token_accuracy": 0.997916667163372,
    "num_tokens": 485528.0,
    "epoch": 2.6357615894039736,
    "step": 200
  },
  {
    "eval_loss": 1.8287278413772583,
    "eval_runtime": 8.7245,
    "eval_samples_per_second": 17.193,
    "eval_steps_per_second": 4.356,
    "eval_entropy": 0.06022026199015703,
    "eval_mean_token_accuracy": 0.7735275751666019,
    "eval_num_tokens": 485528.0,
    "epoch": 2.6357615894039736,
    "step": 200
  },
  {
    "loss": 0.0011992882937192917,
    "grad_norm": 0.004404825158417225,
    "learning_rate": 1.853658536585366e-05,
    "entropy": 0.0030429765494773163,
    "mean_token_accuracy": 1.0,
    "num_tokens": 510081.0,
    "epoch": 2.7682119205298013,
    "step": 210
  },
  {
    "eval_loss": 1.8526947498321533,
    "eval_runtime": 8.6297,
    "eval_samples_per_second": 17.382,
    "eval_steps_per_second": 4.403,
    "eval_entropy": 0.057951420073789596,
    "eval_mean_token_accuracy": 0.7669486277981808,
    "eval_num_tokens": 510081.0,
    "epoch": 2.7682119205298013,
    "step": 210
  },
  {
    "loss": 0.007075458765029907,
    "grad_norm": 0.029240623116493225,
    "learning_rate": 8.780487804878048e-06,
    "entropy": 0.006888014055220992,
    "mean_token_accuracy": 0.9928314402699471,
    "num_tokens": 534422.0,
    "epoch": 2.9006622516556293,
    "step": 220
  },
  {
    "eval_loss": 1.8728781938552856,
    "eval_runtime": 8.6887,
    "eval_samples_per_second": 17.264,
    "eval_steps_per_second": 4.374,
    "eval_entropy": 0.05773907397892081,
    "eval_mean_token_accuracy": 0.7735275751666019,
    "eval_num_tokens": 534422.0,
    "epoch": 2.9006622516556293,
    "step": 220
  },
  {
    "eval_loss": 1.882822871208191,
    "eval_runtime": 8.6056,
    "eval_samples_per_second": 17.43,
    "eval_steps_per_second": 4.416,
    "eval_entropy": 0.05690596842377042,
    "eval_mean_token_accuracy": 0.7669486277981808,
    "eval_num_tokens": 552555.0,
    "epoch": 3.0,
    "step": 228
  },
  {
    "train_runtime": 1087.8226,
    "train_samples_per_second": 3.326,
    "train_steps_per_second": 0.21,
    "total_flos": 3.178031783683891e+16,
    "train_loss": 0.3232832907233387,
    "epoch": 3.0,
    "step": 228
  }
]