{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.006003482019571352,
  "eval_steps": 9,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 6.003482019571351e-05,
      "grad_norm": 9.387470245361328,
      "learning_rate": 1e-05,
      "loss": 2.8454,
      "step": 1
    },
    {
      "epoch": 6.003482019571351e-05,
      "eval_loss": NaN,
      "eval_runtime": 1747.2967,
      "eval_samples_per_second": 8.028,
      "eval_steps_per_second": 1.004,
      "step": 1
    },
    {
      "epoch": 0.00012006964039142703,
      "grad_norm": 8.35964584350586,
      "learning_rate": 2e-05,
      "loss": 1.8863,
      "step": 2
    },
    {
      "epoch": 0.00018010446058714055,
      "grad_norm": 6.206863880157471,
      "learning_rate": 3e-05,
      "loss": 1.2171,
      "step": 3
    },
    {
      "epoch": 0.00024013928078285405,
      "grad_norm": 4.522266864776611,
      "learning_rate": 4e-05,
      "loss": 0.9352,
      "step": 4
    },
    {
      "epoch": 0.0003001741009785676,
      "grad_norm": 5.3215155601501465,
      "learning_rate": 5e-05,
      "loss": 2.12,
      "step": 5
    },
    {
      "epoch": 0.0003602089211742811,
      "grad_norm": 10.670658111572266,
      "learning_rate": 6e-05,
      "loss": 1.4522,
      "step": 6
    },
    {
      "epoch": 0.0004202437413699946,
      "grad_norm": 14.341257095336914,
      "learning_rate": 7e-05,
      "loss": 2.9383,
      "step": 7
    },
    {
      "epoch": 0.0004802785615657081,
      "grad_norm": 13.476455688476562,
      "learning_rate": 8e-05,
      "loss": 1.5796,
      "step": 8
    },
    {
      "epoch": 0.0005403133817614216,
      "grad_norm": 6.012520790100098,
      "learning_rate": 9e-05,
      "loss": 0.7204,
      "step": 9
    },
    {
      "epoch": 0.0005403133817614216,
      "eval_loss": NaN,
      "eval_runtime": 1748.6263,
      "eval_samples_per_second": 8.022,
      "eval_steps_per_second": 1.003,
      "step": 9
    },
    {
      "epoch": 0.0006003482019571352,
      "grad_norm": 4.6742472648620605,
      "learning_rate": 0.0001,
      "loss": 0.9866,
      "step": 10
    },
    {
      "epoch": 0.0006603830221528486,
      "grad_norm": 0.0,
      "learning_rate": 9.99695413509548e-05,
      "loss": 0.0,
      "step": 11
    },
    {
      "epoch": 0.0007204178423485622,
      "grad_norm": 6.591522216796875,
      "learning_rate": 9.987820251299122e-05,
      "loss": 0.7643,
      "step": 12
    },
    {
      "epoch": 0.0007804526625442757,
      "grad_norm": 5.335558891296387,
      "learning_rate": 9.972609476841367e-05,
      "loss": 0.2543,
      "step": 13
    },
    {
      "epoch": 0.0008404874827399892,
      "grad_norm": 3.498502731323242,
      "learning_rate": 9.951340343707852e-05,
      "loss": 0.1858,
      "step": 14
    },
    {
      "epoch": 0.0009005223029357027,
      "grad_norm": 0.0,
      "learning_rate": 9.924038765061042e-05,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.0009605571231314162,
      "grad_norm": 1.4111747741699219,
      "learning_rate": 9.890738003669029e-05,
      "loss": 0.0441,
      "step": 16
    },
    {
      "epoch": 0.0010205919433271297,
      "grad_norm": 0.43236783146858215,
      "learning_rate": 9.851478631379982e-05,
      "loss": 0.0093,
      "step": 17
    },
    {
      "epoch": 0.0010806267635228432,
      "grad_norm": 0.7553333640098572,
      "learning_rate": 9.806308479691595e-05,
      "loss": 0.0163,
      "step": 18
    },
    {
      "epoch": 0.0010806267635228432,
      "eval_loss": NaN,
      "eval_runtime": 1747.826,
      "eval_samples_per_second": 8.025,
      "eval_steps_per_second": 1.004,
      "step": 18
    },
    {
      "epoch": 0.0011406615837185568,
      "grad_norm": 3.277437925338745,
      "learning_rate": 9.755282581475769e-05,
      "loss": 0.0186,
      "step": 19
    },
    {
      "epoch": 0.0012006964039142703,
      "grad_norm": 0.09917615354061127,
      "learning_rate": 9.698463103929542e-05,
      "loss": 0.0016,
      "step": 20
    },
    {
      "epoch": 0.0012607312241099838,
      "grad_norm": 1.3171261548995972,
      "learning_rate": 9.635919272833938e-05,
      "loss": 0.0108,
      "step": 21
    },
    {
      "epoch": 0.0013207660443056973,
      "grad_norm": 0.7155139446258545,
      "learning_rate": 9.567727288213005e-05,
      "loss": 0.0045,
      "step": 22
    },
    {
      "epoch": 0.0013808008645014107,
      "grad_norm": 0.26169103384017944,
      "learning_rate": 9.493970231495835e-05,
      "loss": 0.0033,
      "step": 23
    },
    {
      "epoch": 0.0014408356846971244,
      "grad_norm": 7.987000465393066,
      "learning_rate": 9.414737964294636e-05,
      "loss": 0.302,
      "step": 24
    },
    {
      "epoch": 0.001500870504892838,
      "grad_norm": 0.0,
      "learning_rate": 9.330127018922194e-05,
      "loss": 0.0,
      "step": 25
    },
    {
      "epoch": 0.0015609053250885514,
      "grad_norm": 0.009821158833801746,
      "learning_rate": 9.24024048078213e-05,
      "loss": 0.0003,
      "step": 26
    },
    {
      "epoch": 0.0016209401452842648,
      "grad_norm": 0.01827707327902317,
      "learning_rate": 9.145187862775209e-05,
      "loss": 0.0005,
      "step": 27
    },
    {
      "epoch": 0.0016209401452842648,
      "eval_loss": NaN,
      "eval_runtime": 1747.1761,
      "eval_samples_per_second": 8.028,
      "eval_steps_per_second": 1.004,
      "step": 27
    },
    {
      "epoch": 0.0016809749654799783,
      "grad_norm": 0.12341640144586563,
      "learning_rate": 9.045084971874738e-05,
      "loss": 0.0023,
      "step": 28
    },
    {
      "epoch": 0.001741009785675692,
      "grad_norm": 0.04023746773600578,
      "learning_rate": 8.940053768033609e-05,
      "loss": 0.0003,
      "step": 29
    },
    {
      "epoch": 0.0018010446058714055,
      "grad_norm": 0.0,
      "learning_rate": 8.83022221559489e-05,
      "loss": 0.0,
      "step": 30
    },
    {
      "epoch": 0.001861079426067119,
      "grad_norm": 0.4180465638637543,
      "learning_rate": 8.715724127386972e-05,
      "loss": 0.002,
      "step": 31
    },
    {
      "epoch": 0.0019211142462628324,
      "grad_norm": 0.028819318860769272,
      "learning_rate": 8.596699001693255e-05,
      "loss": 0.0004,
      "step": 32
    },
    {
      "epoch": 0.001981149066458546,
      "grad_norm": 28.845888137817383,
      "learning_rate": 8.473291852294987e-05,
      "loss": 1.3233,
      "step": 33
    },
    {
      "epoch": 0.0020411838866542594,
      "grad_norm": 0.015622777864336967,
      "learning_rate": 8.345653031794292e-05,
      "loss": 0.0005,
      "step": 34
    },
    {
      "epoch": 0.002101218706849973,
      "grad_norm": 8.221495628356934,
      "learning_rate": 8.213938048432697e-05,
      "loss": 0.0868,
      "step": 35
    },
    {
      "epoch": 0.0021612535270456863,
      "grad_norm": 0.7402325868606567,
      "learning_rate": 8.07830737662829e-05,
      "loss": 0.0035,
      "step": 36
    },
    {
      "epoch": 0.0021612535270456863,
      "eval_loss": NaN,
      "eval_runtime": 1747.9827,
      "eval_samples_per_second": 8.025,
      "eval_steps_per_second": 1.003,
      "step": 36
    },
    {
      "epoch": 0.0022212883472414,
      "grad_norm": 0.1636962592601776,
      "learning_rate": 7.938926261462366e-05,
      "loss": 0.002,
      "step": 37
    },
    {
      "epoch": 0.0022813231674371137,
      "grad_norm": 4.500185012817383,
      "learning_rate": 7.795964517353735e-05,
      "loss": 0.1112,
      "step": 38
    },
    {
      "epoch": 0.002341357987632827,
      "grad_norm": 0.0,
      "learning_rate": 7.649596321166024e-05,
      "loss": 0.0,
      "step": 39
    },
    {
      "epoch": 0.0024013928078285406,
      "grad_norm": 7.916851043701172,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.754,
      "step": 40
    },
    {
      "epoch": 0.002461427628024254,
      "grad_norm": 0.376034140586853,
      "learning_rate": 7.347357813929454e-05,
      "loss": 0.0019,
      "step": 41
    },
    {
      "epoch": 0.0025214624482199676,
      "grad_norm": 0.1723334640264511,
      "learning_rate": 7.191855733945387e-05,
      "loss": 0.002,
      "step": 42
    },
    {
      "epoch": 0.002581497268415681,
      "grad_norm": 7.519275188446045,
      "learning_rate": 7.033683215379002e-05,
      "loss": 0.0085,
      "step": 43
    },
    {
      "epoch": 0.0026415320886113945,
      "grad_norm": 1.5027763843536377,
      "learning_rate": 6.873032967079561e-05,
      "loss": 0.0151,
      "step": 44
    },
    {
      "epoch": 0.002701566908807108,
      "grad_norm": 0.010223852470517159,
      "learning_rate": 6.710100716628344e-05,
      "loss": 0.0003,
      "step": 45
    },
    {
      "epoch": 0.002701566908807108,
      "eval_loss": NaN,
      "eval_runtime": 1748.0845,
      "eval_samples_per_second": 8.024,
      "eval_steps_per_second": 1.003,
      "step": 45
    },
    {
      "epoch": 0.0027616017290028215,
      "grad_norm": 0.003923419397324324,
      "learning_rate": 6.545084971874738e-05,
      "loss": 0.0002,
      "step": 46
    },
    {
      "epoch": 0.002821636549198535,
      "grad_norm": 0.033521488308906555,
      "learning_rate": 6.378186779084995e-05,
      "loss": 0.0006,
      "step": 47
    },
    {
      "epoch": 0.002881671369394249,
      "grad_norm": 0.006656797137111425,
      "learning_rate": 6.209609477998338e-05,
      "loss": 0.0002,
      "step": 48
    },
    {
      "epoch": 0.0029417061895899623,
      "grad_norm": 0.0,
      "learning_rate": 6.0395584540887963e-05,
      "loss": 0.0,
      "step": 49
    },
    {
      "epoch": 0.003001741009785676,
      "grad_norm": 0.007206365000456572,
      "learning_rate": 5.868240888334653e-05,
      "loss": 0.0002,
      "step": 50
    },
    {
      "epoch": 0.0030617758299813893,
      "grad_norm": 2.403134346008301,
      "learning_rate": 5.695865504800327e-05,
      "loss": 0.2376,
      "step": 51
    },
    {
      "epoch": 0.0031218106501771027,
      "grad_norm": 0.07276768982410431,
      "learning_rate": 5.522642316338268e-05,
      "loss": 0.0009,
      "step": 52
    },
    {
      "epoch": 0.003181845470372816,
      "grad_norm": 14.310860633850098,
      "learning_rate": 5.348782368720626e-05,
      "loss": 0.0216,
      "step": 53
    },
    {
      "epoch": 0.0032418802905685297,
      "grad_norm": 0.9602569937705994,
      "learning_rate": 5.174497483512506e-05,
      "loss": 0.0081,
      "step": 54
    },
    {
      "epoch": 0.0032418802905685297,
      "eval_loss": NaN,
      "eval_runtime": 1747.7961,
      "eval_samples_per_second": 8.026,
      "eval_steps_per_second": 1.004,
      "step": 54
    },
    {
      "epoch": 0.003301915110764243,
      "grad_norm": 0.0,
      "learning_rate": 5e-05,
      "loss": 0.0,
      "step": 55
    },
    {
      "epoch": 0.0033619499309599566,
      "grad_norm": 0.16967162489891052,
      "learning_rate": 4.825502516487497e-05,
      "loss": 0.002,
      "step": 56
    },
    {
      "epoch": 0.00342198475115567,
      "grad_norm": 0.03777099773287773,
      "learning_rate": 4.6512176312793736e-05,
      "loss": 0.0006,
      "step": 57
    },
    {
      "epoch": 0.003482019571351384,
      "grad_norm": 0.0,
      "learning_rate": 4.477357683661734e-05,
      "loss": 0.0,
      "step": 58
    },
    {
      "epoch": 0.0035420543915470975,
      "grad_norm": 0.15030822157859802,
      "learning_rate": 4.3041344951996746e-05,
      "loss": 0.0018,
      "step": 59
    },
    {
      "epoch": 0.003602089211742811,
      "grad_norm": 0.0023677474819123745,
      "learning_rate": 4.131759111665349e-05,
      "loss": 0.0001,
      "step": 60
    },
    {
      "epoch": 0.0036621240319385244,
      "grad_norm": 0.01745045743882656,
      "learning_rate": 3.960441545911204e-05,
      "loss": 0.0002,
      "step": 61
    },
    {
      "epoch": 0.003722158852134238,
      "grad_norm": 0.018494900315999985,
      "learning_rate": 3.790390522001662e-05,
      "loss": 0.0002,
      "step": 62
    },
    {
      "epoch": 0.0037821936723299514,
      "grad_norm": 0.1796453297138214,
      "learning_rate": 3.6218132209150045e-05,
      "loss": 0.0019,
      "step": 63
    },
    {
      "epoch": 0.0037821936723299514,
      "eval_loss": NaN,
      "eval_runtime": 1748.782,
      "eval_samples_per_second": 8.021,
      "eval_steps_per_second": 1.003,
      "step": 63
    },
    {
      "epoch": 0.003842228492525665,
      "grad_norm": 0.0,
      "learning_rate": 3.4549150281252636e-05,
      "loss": 0.0,
      "step": 64
    },
    {
      "epoch": 0.0039022633127213783,
      "grad_norm": 0.014414871111512184,
      "learning_rate": 3.289899283371657e-05,
      "loss": 0.0004,
      "step": 65
    },
    {
      "epoch": 0.003962298132917092,
      "grad_norm": 0.0,
      "learning_rate": 3.12696703292044e-05,
      "loss": 0.0,
      "step": 66
    },
    {
      "epoch": 0.004022332953112806,
      "grad_norm": 0.0,
      "learning_rate": 2.9663167846209998e-05,
      "loss": 0.0,
      "step": 67
    },
    {
      "epoch": 0.004082367773308519,
      "grad_norm": 0.25538694858551025,
      "learning_rate": 2.8081442660546125e-05,
      "loss": 0.003,
      "step": 68
    },
    {
      "epoch": 0.004142402593504233,
      "grad_norm": 0.007418240420520306,
      "learning_rate": 2.6526421860705473e-05,
      "loss": 0.0002,
      "step": 69
    },
    {
      "epoch": 0.004202437413699946,
      "grad_norm": 0.002056136028841138,
      "learning_rate": 2.500000000000001e-05,
      "loss": 0.0001,
      "step": 70
    },
    {
      "epoch": 0.00426247223389566,
      "grad_norm": 0.04461194574832916,
      "learning_rate": 2.350403678833976e-05,
      "loss": 0.0006,
      "step": 71
    },
    {
      "epoch": 0.004322507054091373,
      "grad_norm": 0.10500391572713852,
      "learning_rate": 2.2040354826462668e-05,
      "loss": 0.0008,
      "step": 72
    },
    {
      "epoch": 0.004322507054091373,
      "eval_loss": NaN,
      "eval_runtime": 1748.6975,
      "eval_samples_per_second": 8.021,
      "eval_steps_per_second": 1.003,
      "step": 72
    },
    {
      "epoch": 0.0043825418742870865,
      "grad_norm": 0.16719873249530792,
      "learning_rate": 2.061073738537635e-05,
      "loss": 0.0029,
      "step": 73
    },
    {
      "epoch": 0.0044425766944828,
      "grad_norm": 0.0035197085235267878,
      "learning_rate": 1.9216926233717085e-05,
      "loss": 0.0001,
      "step": 74
    },
    {
      "epoch": 0.0045026115146785135,
      "grad_norm": 0.005952209699898958,
      "learning_rate": 1.7860619515673033e-05,
      "loss": 0.0001,
      "step": 75
    },
    {
      "epoch": 0.004562646334874227,
      "grad_norm": 0.0,
      "learning_rate": 1.6543469682057106e-05,
      "loss": 0.0,
      "step": 76
    },
    {
      "epoch": 0.00462268115506994,
      "grad_norm": 0.0,
      "learning_rate": 1.526708147705013e-05,
      "loss": 0.0,
      "step": 77
    },
    {
      "epoch": 0.004682715975265654,
      "grad_norm": 0.0021064886823296547,
      "learning_rate": 1.4033009983067452e-05,
      "loss": 0.0001,
      "step": 78
    },
    {
      "epoch": 0.004742750795461367,
      "grad_norm": 0.016838403418660164,
      "learning_rate": 1.2842758726130283e-05,
      "loss": 0.0004,
      "step": 79
    },
    {
      "epoch": 0.004802785615657081,
      "grad_norm": 0.0046536740846931934,
      "learning_rate": 1.1697777844051105e-05,
      "loss": 0.0001,
      "step": 80
    },
    {
      "epoch": 0.004862820435852794,
      "grad_norm": 0.01591491512954235,
      "learning_rate": 1.0599462319663905e-05,
      "loss": 0.0003,
      "step": 81
    },
    {
      "epoch": 0.004862820435852794,
      "eval_loss": NaN,
      "eval_runtime": 1749.4659,
      "eval_samples_per_second": 8.018,
      "eval_steps_per_second": 1.003,
      "step": 81
    },
    {
      "epoch": 0.004922855256048508,
      "grad_norm": 0.0,
      "learning_rate": 9.549150281252633e-06,
      "loss": 0.0,
      "step": 82
    },
    {
      "epoch": 0.004982890076244221,
      "grad_norm": 0.0,
      "learning_rate": 8.548121372247918e-06,
      "loss": 0.0,
      "step": 83
    },
    {
      "epoch": 0.005042924896439935,
      "grad_norm": 0.003679132554680109,
      "learning_rate": 7.597595192178702e-06,
      "loss": 0.0002,
      "step": 84
    },
    {
      "epoch": 0.005102959716635649,
      "grad_norm": 0.05891428515315056,
      "learning_rate": 6.698729810778065e-06,
      "loss": 0.0015,
      "step": 85
    },
    {
      "epoch": 0.005162994536831362,
      "grad_norm": 0.004211484454572201,
      "learning_rate": 5.852620357053651e-06,
      "loss": 0.0001,
      "step": 86
    },
    {
      "epoch": 0.005223029357027076,
      "grad_norm": 0.016904192045331,
      "learning_rate": 5.060297685041659e-06,
      "loss": 0.0004,
      "step": 87
    },
    {
      "epoch": 0.005283064177222789,
      "grad_norm": 0.05485766381025314,
      "learning_rate": 4.322727117869951e-06,
      "loss": 0.0007,
      "step": 88
    },
    {
      "epoch": 0.005343098997418503,
      "grad_norm": 3.945460557937622,
      "learning_rate": 3.6408072716606346e-06,
      "loss": 0.2777,
      "step": 89
    },
    {
      "epoch": 0.005403133817614216,
      "grad_norm": 0.0012216288596391678,
      "learning_rate": 3.0153689607045845e-06,
      "loss": 0.0001,
      "step": 90
    },
    {
      "epoch": 0.005403133817614216,
      "eval_loss": NaN,
      "eval_runtime": 1749.58,
      "eval_samples_per_second": 8.017,
      "eval_steps_per_second": 1.003,
      "step": 90
    },
    {
      "epoch": 0.00546316863780993,
      "grad_norm": 0.0035565090365707874,
      "learning_rate": 2.4471741852423237e-06,
      "loss": 0.0001,
      "step": 91
    },
    {
      "epoch": 0.005523203458005643,
      "grad_norm": 0.0,
      "learning_rate": 1.9369152030840556e-06,
      "loss": 0.0,
      "step": 92
    },
    {
      "epoch": 0.005583238278201357,
      "grad_norm": 0.04645892232656479,
      "learning_rate": 1.4852136862001764e-06,
      "loss": 0.0006,
      "step": 93
    },
    {
      "epoch": 0.00564327309839707,
      "grad_norm": 0.0014873318141326308,
      "learning_rate": 1.0926199633097157e-06,
      "loss": 0.0001,
      "step": 94
    },
    {
      "epoch": 0.005703307918592784,
      "grad_norm": 0.2146553099155426,
      "learning_rate": 7.596123493895991e-07,
      "loss": 0.003,
      "step": 95
    },
    {
      "epoch": 0.005763342738788498,
      "grad_norm": 16.49122428894043,
      "learning_rate": 4.865965629214819e-07,
      "loss": 1.4834,
      "step": 96
    },
    {
      "epoch": 0.005823377558984211,
      "grad_norm": 0.004562173970043659,
      "learning_rate": 2.7390523158633554e-07,
      "loss": 0.0001,
      "step": 97
    },
    {
      "epoch": 0.005883412379179925,
      "grad_norm": 0.0012735759373754263,
      "learning_rate": 1.2179748700879012e-07,
      "loss": 0.0001,
      "step": 98
    },
    {
      "epoch": 0.005943447199375638,
      "grad_norm": 0.0040828571654856205,
      "learning_rate": 3.04586490452119e-08,
      "loss": 0.0001,
      "step": 99
    },
    {
      "epoch": 0.005943447199375638,
      "eval_loss": NaN,
      "eval_runtime": 1749.7589,
      "eval_samples_per_second": 8.017,
      "eval_steps_per_second": 1.002,
      "step": 99
    },
    {
      "epoch": 0.006003482019571352,
      "grad_norm": 0.0023373435251414776,
      "learning_rate": 0.0,
      "loss": 0.0001,
      "step": 100
    }
  ],
  "logging_steps": 1,
  "max_steps": 100,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 25,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 7.715631936621773e+16,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}