{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.004443259575224384,
  "eval_steps": 9,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 4.443259575224385e-05,
      "grad_norm": 74.77045440673828,
      "learning_rate": 1e-05,
      "loss": 12.5149,
      "step": 1
    },
    {
      "epoch": 4.443259575224385e-05,
      "eval_loss": 12.948126792907715,
      "eval_runtime": 2085.2718,
      "eval_samples_per_second": 9.089,
      "eval_steps_per_second": 1.137,
      "step": 1
    },
    {
      "epoch": 8.88651915044877e-05,
      "grad_norm": 73.22039031982422,
      "learning_rate": 2e-05,
      "loss": 13.3496,
      "step": 2
    },
    {
      "epoch": 0.00013329778725673153,
      "grad_norm": 72.73692321777344,
      "learning_rate": 3e-05,
      "loss": 12.6915,
      "step": 3
    },
    {
      "epoch": 0.0001777303830089754,
      "grad_norm": 72.64486694335938,
      "learning_rate": 4e-05,
      "loss": 12.016,
      "step": 4
    },
    {
      "epoch": 0.00022216297876121924,
      "grad_norm": 76.35221862792969,
      "learning_rate": 5e-05,
      "loss": 10.4084,
      "step": 5
    },
    {
      "epoch": 0.00026659557451346307,
      "grad_norm": 61.80672836303711,
      "learning_rate": 6e-05,
      "loss": 7.6542,
      "step": 6
    },
    {
      "epoch": 0.00031102817026570695,
      "grad_norm": 40.792171478271484,
      "learning_rate": 7e-05,
      "loss": 4.2684,
      "step": 7
    },
    {
      "epoch": 0.0003554607660179508,
      "grad_norm": 17.3669490814209,
      "learning_rate": 8e-05,
      "loss": 3.3804,
      "step": 8
    },
    {
      "epoch": 0.0003998933617701946,
      "grad_norm": 21.140520095825195,
      "learning_rate": 9e-05,
      "loss": 2.5972,
      "step": 9
    },
    {
      "epoch": 0.0003998933617701946,
      "eval_loss": 1.9437395334243774,
      "eval_runtime": 2088.5806,
      "eval_samples_per_second": 9.075,
      "eval_steps_per_second": 1.135,
      "step": 9
    },
    {
      "epoch": 0.0004443259575224385,
      "grad_norm": 18.79267692565918,
      "learning_rate": 0.0001,
      "loss": 2.2143,
      "step": 10
    },
    {
      "epoch": 0.0004887585532746823,
      "grad_norm": 8.969635963439941,
      "learning_rate": 9.99695413509548e-05,
      "loss": 1.6568,
      "step": 11
    },
    {
      "epoch": 0.0005331911490269261,
      "grad_norm": 8.22108268737793,
      "learning_rate": 9.987820251299122e-05,
      "loss": 1.6399,
      "step": 12
    },
    {
      "epoch": 0.00057762374477917,
      "grad_norm": 6.7568230628967285,
      "learning_rate": 9.972609476841367e-05,
      "loss": 1.3787,
      "step": 13
    },
    {
      "epoch": 0.0006220563405314139,
      "grad_norm": 22.37034034729004,
      "learning_rate": 9.951340343707852e-05,
      "loss": 1.3557,
      "step": 14
    },
    {
      "epoch": 0.0006664889362836577,
      "grad_norm": 6.20902156829834,
      "learning_rate": 9.924038765061042e-05,
      "loss": 1.5788,
      "step": 15
    },
    {
      "epoch": 0.0007109215320359015,
      "grad_norm": 5.11328649520874,
      "learning_rate": 9.890738003669029e-05,
      "loss": 1.319,
      "step": 16
    },
    {
      "epoch": 0.0007553541277881454,
      "grad_norm": 7.359101295471191,
      "learning_rate": 9.851478631379982e-05,
      "loss": 1.0957,
      "step": 17
    },
    {
      "epoch": 0.0007997867235403892,
      "grad_norm": 4.124218940734863,
      "learning_rate": 9.806308479691595e-05,
      "loss": 1.1526,
      "step": 18
    },
    {
      "epoch": 0.0007997867235403892,
      "eval_loss": 1.082924246788025,
      "eval_runtime": 2088.8216,
      "eval_samples_per_second": 9.074,
      "eval_steps_per_second": 1.135,
      "step": 18
    },
    {
      "epoch": 0.000844219319292633,
      "grad_norm": 4.501346588134766,
      "learning_rate": 9.755282581475769e-05,
      "loss": 0.922,
      "step": 19
    },
    {
      "epoch": 0.000888651915044877,
      "grad_norm": 5.430482387542725,
      "learning_rate": 9.698463103929542e-05,
      "loss": 0.8881,
      "step": 20
    },
    {
      "epoch": 0.0009330845107971208,
      "grad_norm": 2.8988702297210693,
      "learning_rate": 9.635919272833938e-05,
      "loss": 0.9686,
      "step": 21
    },
    {
      "epoch": 0.0009775171065493646,
      "grad_norm": 2.7791342735290527,
      "learning_rate": 9.567727288213005e-05,
      "loss": 0.7266,
      "step": 22
    },
    {
      "epoch": 0.0010219497023016084,
      "grad_norm": 4.415706157684326,
      "learning_rate": 9.493970231495835e-05,
      "loss": 1.4575,
      "step": 23
    },
    {
      "epoch": 0.0010663822980538523,
      "grad_norm": 3.8715085983276367,
      "learning_rate": 9.414737964294636e-05,
      "loss": 1.239,
      "step": 24
    },
    {
      "epoch": 0.001110814893806096,
      "grad_norm": 4.328126430511475,
      "learning_rate": 9.330127018922194e-05,
      "loss": 1.2685,
      "step": 25
    },
    {
      "epoch": 0.00115524748955834,
      "grad_norm": 4.6381072998046875,
      "learning_rate": 9.24024048078213e-05,
      "loss": 1.0216,
      "step": 26
    },
    {
      "epoch": 0.0011996800853105837,
      "grad_norm": 3.818336248397827,
      "learning_rate": 9.145187862775209e-05,
      "loss": 1.2808,
      "step": 27
    },
    {
      "epoch": 0.0011996800853105837,
      "eval_loss": 1.0154649019241333,
      "eval_runtime": 2090.5291,
      "eval_samples_per_second": 9.066,
      "eval_steps_per_second": 1.134,
      "step": 27
    },
    {
      "epoch": 0.0012441126810628278,
      "grad_norm": 4.990555286407471,
      "learning_rate": 9.045084971874738e-05,
      "loss": 1.2793,
      "step": 28
    },
    {
      "epoch": 0.0012885452768150716,
      "grad_norm": 3.8898091316223145,
      "learning_rate": 8.940053768033609e-05,
      "loss": 1.0337,
      "step": 29
    },
    {
      "epoch": 0.0013329778725673154,
      "grad_norm": 4.613339424133301,
      "learning_rate": 8.83022221559489e-05,
      "loss": 1.2245,
      "step": 30
    },
    {
      "epoch": 0.0013774104683195593,
      "grad_norm": 4.101949691772461,
      "learning_rate": 8.715724127386972e-05,
      "loss": 0.9874,
      "step": 31
    },
    {
      "epoch": 0.001421843064071803,
      "grad_norm": 3.4817678928375244,
      "learning_rate": 8.596699001693255e-05,
      "loss": 0.8011,
      "step": 32
    },
    {
      "epoch": 0.001466275659824047,
      "grad_norm": 4.340565204620361,
      "learning_rate": 8.473291852294987e-05,
      "loss": 0.9708,
      "step": 33
    },
    {
      "epoch": 0.0015107082555762907,
      "grad_norm": 4.810932159423828,
      "learning_rate": 8.345653031794292e-05,
      "loss": 1.0238,
      "step": 34
    },
    {
      "epoch": 0.0015551408513285346,
      "grad_norm": 3.2785961627960205,
      "learning_rate": 8.213938048432697e-05,
      "loss": 1.0832,
      "step": 35
    },
    {
      "epoch": 0.0015995734470807784,
      "grad_norm": 3.1490390300750732,
      "learning_rate": 8.07830737662829e-05,
      "loss": 0.6921,
      "step": 36
    },
    {
      "epoch": 0.0015995734470807784,
      "eval_loss": 0.938591718673706,
      "eval_runtime": 2088.9887,
      "eval_samples_per_second": 9.073,
      "eval_steps_per_second": 1.135,
      "step": 36
    },
    {
      "epoch": 0.0016440060428330222,
      "grad_norm": 3.129164934158325,
      "learning_rate": 7.938926261462366e-05,
      "loss": 0.7297,
      "step": 37
    },
    {
      "epoch": 0.001688438638585266,
      "grad_norm": 3.497979164123535,
      "learning_rate": 7.795964517353735e-05,
      "loss": 0.6793,
      "step": 38
    },
    {
      "epoch": 0.00173287123433751,
      "grad_norm": 3.4145689010620117,
      "learning_rate": 7.649596321166024e-05,
      "loss": 0.9505,
      "step": 39
    },
    {
      "epoch": 0.001777303830089754,
      "grad_norm": 4.804287910461426,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.9092,
      "step": 40
    },
    {
      "epoch": 0.0018217364258419977,
      "grad_norm": 3.785067081451416,
      "learning_rate": 7.347357813929454e-05,
      "loss": 0.7816,
      "step": 41
    },
    {
      "epoch": 0.0018661690215942416,
      "grad_norm": 3.534381628036499,
      "learning_rate": 7.191855733945387e-05,
      "loss": 1.0754,
      "step": 42
    },
    {
      "epoch": 0.0019106016173464854,
      "grad_norm": 5.234914302825928,
      "learning_rate": 7.033683215379002e-05,
      "loss": 0.9232,
      "step": 43
    },
    {
      "epoch": 0.0019550342130987292,
      "grad_norm": 6.870372295379639,
      "learning_rate": 6.873032967079561e-05,
      "loss": 1.3589,
      "step": 44
    },
    {
      "epoch": 0.0019994668088509733,
      "grad_norm": 2.744567394256592,
      "learning_rate": 6.710100716628344e-05,
      "loss": 0.6761,
      "step": 45
    },
    {
      "epoch": 0.0019994668088509733,
      "eval_loss": 0.899674654006958,
      "eval_runtime": 2090.0954,
      "eval_samples_per_second": 9.068,
      "eval_steps_per_second": 1.134,
      "step": 45
    },
    {
      "epoch": 0.002043899404603217,
      "grad_norm": 3.99665904045105,
      "learning_rate": 6.545084971874738e-05,
      "loss": 0.9011,
      "step": 46
    },
    {
      "epoch": 0.002088332000355461,
      "grad_norm": 3.316638231277466,
      "learning_rate": 6.378186779084995e-05,
      "loss": 0.9587,
      "step": 47
    },
    {
      "epoch": 0.0021327645961077045,
      "grad_norm": 3.783149480819702,
      "learning_rate": 6.209609477998338e-05,
      "loss": 1.0353,
      "step": 48
    },
    {
      "epoch": 0.0021771971918599486,
      "grad_norm": 3.442662000656128,
      "learning_rate": 6.0395584540887963e-05,
      "loss": 0.6331,
      "step": 49
    },
    {
      "epoch": 0.002221629787612192,
      "grad_norm": 3.411076545715332,
      "learning_rate": 5.868240888334653e-05,
      "loss": 1.1829,
      "step": 50
    },
    {
      "epoch": 0.0022660623833644362,
      "grad_norm": 3.211519479751587,
      "learning_rate": 5.695865504800327e-05,
      "loss": 0.9899,
      "step": 51
    },
    {
      "epoch": 0.00231049497911668,
      "grad_norm": 2.8170390129089355,
      "learning_rate": 5.522642316338268e-05,
      "loss": 0.7881,
      "step": 52
    },
    {
      "epoch": 0.002354927574868924,
      "grad_norm": 2.29962420463562,
      "learning_rate": 5.348782368720626e-05,
      "loss": 0.7825,
      "step": 53
    },
    {
      "epoch": 0.0023993601706211675,
      "grad_norm": 3.250339984893799,
      "learning_rate": 5.174497483512506e-05,
      "loss": 0.8462,
      "step": 54
    },
    {
      "epoch": 0.0023993601706211675,
      "eval_loss": 0.8770561814308167,
      "eval_runtime": 2090.379,
      "eval_samples_per_second": 9.067,
      "eval_steps_per_second": 1.134,
      "step": 54
    },
    {
      "epoch": 0.0024437927663734115,
      "grad_norm": 3.0480387210845947,
      "learning_rate": 5e-05,
      "loss": 0.8317,
      "step": 55
    },
    {
      "epoch": 0.0024882253621256556,
      "grad_norm": 2.246021032333374,
      "learning_rate": 4.825502516487497e-05,
      "loss": 0.5275,
      "step": 56
    },
    {
      "epoch": 0.002532657957877899,
      "grad_norm": 3.443877696990967,
      "learning_rate": 4.6512176312793736e-05,
      "loss": 1.0023,
      "step": 57
    },
    {
      "epoch": 0.0025770905536301432,
      "grad_norm": 3.9084110260009766,
      "learning_rate": 4.477357683661734e-05,
      "loss": 0.9394,
      "step": 58
    },
    {
      "epoch": 0.002621523149382387,
      "grad_norm": 3.4571526050567627,
      "learning_rate": 4.3041344951996746e-05,
      "loss": 1.1275,
      "step": 59
    },
    {
      "epoch": 0.002665955745134631,
      "grad_norm": 2.5637829303741455,
      "learning_rate": 4.131759111665349e-05,
      "loss": 0.7408,
      "step": 60
    },
    {
      "epoch": 0.0027103883408868745,
      "grad_norm": 2.948639154434204,
      "learning_rate": 3.960441545911204e-05,
      "loss": 0.811,
      "step": 61
    },
    {
      "epoch": 0.0027548209366391185,
      "grad_norm": 2.9560561180114746,
      "learning_rate": 3.790390522001662e-05,
      "loss": 0.6138,
      "step": 62
    },
    {
      "epoch": 0.002799253532391362,
      "grad_norm": 2.9435060024261475,
      "learning_rate": 3.6218132209150045e-05,
      "loss": 0.8675,
      "step": 63
    },
    {
      "epoch": 0.002799253532391362,
      "eval_loss": 0.8335610628128052,
      "eval_runtime": 2088.3473,
      "eval_samples_per_second": 9.076,
      "eval_steps_per_second": 1.135,
      "step": 63
    },
    {
      "epoch": 0.002843686128143606,
      "grad_norm": 1.8505394458770752,
      "learning_rate": 3.4549150281252636e-05,
      "loss": 0.4345,
      "step": 64
    },
    {
      "epoch": 0.00288811872389585,
      "grad_norm": 3.3062446117401123,
      "learning_rate": 3.289899283371657e-05,
      "loss": 0.9831,
      "step": 65
    },
    {
      "epoch": 0.002932551319648094,
      "grad_norm": 3.808541774749756,
      "learning_rate": 3.12696703292044e-05,
      "loss": 0.906,
      "step": 66
    },
    {
      "epoch": 0.002976983915400338,
      "grad_norm": 3.104768753051758,
      "learning_rate": 2.9663167846209998e-05,
      "loss": 0.9008,
      "step": 67
    },
    {
      "epoch": 0.0030214165111525815,
      "grad_norm": 2.730449914932251,
      "learning_rate": 2.8081442660546125e-05,
      "loss": 0.7801,
      "step": 68
    },
    {
      "epoch": 0.0030658491069048255,
      "grad_norm": 3.0719079971313477,
      "learning_rate": 2.6526421860705473e-05,
      "loss": 0.623,
      "step": 69
    },
    {
      "epoch": 0.003110281702657069,
      "grad_norm": 2.975539445877075,
      "learning_rate": 2.500000000000001e-05,
      "loss": 0.9251,
      "step": 70
    },
    {
      "epoch": 0.003154714298409313,
      "grad_norm": 2.507993459701538,
      "learning_rate": 2.350403678833976e-05,
      "loss": 0.5249,
      "step": 71
    },
    {
      "epoch": 0.003199146894161557,
      "grad_norm": 2.6667215824127197,
      "learning_rate": 2.2040354826462668e-05,
      "loss": 0.4238,
      "step": 72
    },
    {
      "epoch": 0.003199146894161557,
      "eval_loss": 0.8086918592453003,
      "eval_runtime": 2088.3743,
      "eval_samples_per_second": 9.075,
      "eval_steps_per_second": 1.135,
      "step": 72
    },
    {
      "epoch": 0.003243579489913801,
      "grad_norm": 2.936882495880127,
      "learning_rate": 2.061073738537635e-05,
      "loss": 0.8522,
      "step": 73
    },
    {
      "epoch": 0.0032880120856660444,
      "grad_norm": 3.8667545318603516,
      "learning_rate": 1.9216926233717085e-05,
      "loss": 0.8983,
      "step": 74
    },
    {
      "epoch": 0.0033324446814182885,
      "grad_norm": 3.9333572387695312,
      "learning_rate": 1.7860619515673033e-05,
      "loss": 1.0196,
      "step": 75
    },
    {
      "epoch": 0.003376877277170532,
      "grad_norm": 2.326871871948242,
      "learning_rate": 1.6543469682057106e-05,
      "loss": 0.7437,
      "step": 76
    },
    {
      "epoch": 0.003421309872922776,
      "grad_norm": 2.7183380126953125,
      "learning_rate": 1.526708147705013e-05,
      "loss": 0.6249,
      "step": 77
    },
    {
      "epoch": 0.00346574246867502,
      "grad_norm": 2.6709814071655273,
      "learning_rate": 1.4033009983067452e-05,
      "loss": 0.668,
      "step": 78
    },
    {
      "epoch": 0.003510175064427264,
      "grad_norm": 2.335897207260132,
      "learning_rate": 1.2842758726130283e-05,
      "loss": 0.4872,
      "step": 79
    },
    {
      "epoch": 0.003554607660179508,
      "grad_norm": 3.3873088359832764,
      "learning_rate": 1.1697777844051105e-05,
      "loss": 1.0067,
      "step": 80
    },
    {
      "epoch": 0.0035990402559317515,
      "grad_norm": 3.065892457962036,
      "learning_rate": 1.0599462319663905e-05,
      "loss": 0.8042,
      "step": 81
    },
    {
      "epoch": 0.0035990402559317515,
      "eval_loss": 0.7969595789909363,
      "eval_runtime": 2089.1579,
      "eval_samples_per_second": 9.072,
      "eval_steps_per_second": 1.134,
      "step": 81
    },
    {
      "epoch": 0.0036434728516839955,
      "grad_norm": 3.0394890308380127,
      "learning_rate": 9.549150281252633e-06,
      "loss": 0.8169,
      "step": 82
    },
    {
      "epoch": 0.003687905447436239,
      "grad_norm": 2.319058418273926,
      "learning_rate": 8.548121372247918e-06,
      "loss": 0.2704,
      "step": 83
    },
    {
      "epoch": 0.003732338043188483,
      "grad_norm": 3.6886699199676514,
      "learning_rate": 7.597595192178702e-06,
      "loss": 0.8313,
      "step": 84
    },
    {
      "epoch": 0.0037767706389407268,
      "grad_norm": 2.6520419120788574,
      "learning_rate": 6.698729810778065e-06,
      "loss": 0.8752,
      "step": 85
    },
    {
      "epoch": 0.003821203234692971,
      "grad_norm": 2.6148107051849365,
      "learning_rate": 5.852620357053651e-06,
      "loss": 0.7429,
      "step": 86
    },
    {
      "epoch": 0.0038656358304452144,
      "grad_norm": 2.711278200149536,
      "learning_rate": 5.060297685041659e-06,
      "loss": 0.5209,
      "step": 87
    },
    {
      "epoch": 0.0039100684261974585,
      "grad_norm": 2.5776398181915283,
      "learning_rate": 4.322727117869951e-06,
      "loss": 0.6768,
      "step": 88
    },
    {
      "epoch": 0.0039545010219497025,
      "grad_norm": 2.9126386642456055,
      "learning_rate": 3.6408072716606346e-06,
      "loss": 0.8562,
      "step": 89
    },
    {
      "epoch": 0.0039989336177019465,
      "grad_norm": 3.8028640747070312,
      "learning_rate": 3.0153689607045845e-06,
      "loss": 0.8226,
      "step": 90
    },
    {
      "epoch": 0.0039989336177019465,
      "eval_loss": 0.7885631918907166,
      "eval_runtime": 2088.9791,
      "eval_samples_per_second": 9.073,
      "eval_steps_per_second": 1.135,
      "step": 90
    },
    {
      "epoch": 0.00404336621345419,
      "grad_norm": 3.353768825531006,
      "learning_rate": 2.4471741852423237e-06,
      "loss": 0.7805,
      "step": 91
    },
    {
      "epoch": 0.004087798809206434,
      "grad_norm": 3.188265800476074,
      "learning_rate": 1.9369152030840556e-06,
      "loss": 0.8433,
      "step": 92
    },
    {
      "epoch": 0.004132231404958678,
      "grad_norm": 4.046318531036377,
      "learning_rate": 1.4852136862001764e-06,
      "loss": 0.9672,
      "step": 93
    },
    {
      "epoch": 0.004176664000710922,
      "grad_norm": 2.8643298149108887,
      "learning_rate": 1.0926199633097157e-06,
      "loss": 0.6082,
      "step": 94
    },
    {
      "epoch": 0.004221096596463165,
      "grad_norm": 2.820201873779297,
      "learning_rate": 7.596123493895991e-07,
      "loss": 0.5979,
      "step": 95
    },
    {
      "epoch": 0.004265529192215409,
      "grad_norm": 3.394400119781494,
      "learning_rate": 4.865965629214819e-07,
      "loss": 0.9704,
      "step": 96
    },
    {
      "epoch": 0.004309961787967653,
      "grad_norm": 2.8901283740997314,
      "learning_rate": 2.7390523158633554e-07,
      "loss": 0.7127,
      "step": 97
    },
    {
      "epoch": 0.004354394383719897,
      "grad_norm": 4.210619926452637,
      "learning_rate": 1.2179748700879012e-07,
      "loss": 0.9518,
      "step": 98
    },
    {
      "epoch": 0.004398826979472141,
      "grad_norm": 4.082587718963623,
      "learning_rate": 3.04586490452119e-08,
      "loss": 0.6261,
      "step": 99
    },
    {
      "epoch": 0.004398826979472141,
      "eval_loss": 0.7863272428512573,
      "eval_runtime": 2088.7224,
      "eval_samples_per_second": 9.074,
      "eval_steps_per_second": 1.135,
      "step": 99
    },
    {
      "epoch": 0.004443259575224384,
      "grad_norm": 3.289276123046875,
      "learning_rate": 0.0,
      "loss": 0.7214,
      "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": 7.41887283560448e+16,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}