{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0165809981760902,
  "eval_steps": 9,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.000165809981760902,
      "grad_norm": 14.182147026062012,
      "learning_rate": 1e-05,
      "loss": 14.468,
      "step": 1
    },
    {
      "epoch": 0.000165809981760902,
      "eval_loss": 6.990629196166992,
      "eval_runtime": 1574.821,
      "eval_samples_per_second": 3.225,
      "eval_steps_per_second": 0.403,
      "step": 1
    },
    {
      "epoch": 0.000331619963521804,
      "grad_norm": 8.581724166870117,
      "learning_rate": 2e-05,
      "loss": 15.5583,
      "step": 2
    },
    {
      "epoch": 0.000497429945282706,
      "grad_norm": 7.9157562255859375,
      "learning_rate": 3e-05,
      "loss": 12.8037,
      "step": 3
    },
    {
      "epoch": 0.000663239927043608,
      "grad_norm": 10.532652854919434,
      "learning_rate": 4e-05,
      "loss": 12.7289,
      "step": 4
    },
    {
      "epoch": 0.00082904990880451,
      "grad_norm": 11.54885196685791,
      "learning_rate": 5e-05,
      "loss": 13.4162,
      "step": 5
    },
    {
      "epoch": 0.000994859890565412,
      "grad_norm": 13.328164100646973,
      "learning_rate": 6e-05,
      "loss": 13.5196,
      "step": 6
    },
    {
      "epoch": 0.001160669872326314,
      "grad_norm": 13.098895072937012,
      "learning_rate": 7e-05,
      "loss": 13.5123,
      "step": 7
    },
    {
      "epoch": 0.001326479854087216,
      "grad_norm": 17.912372589111328,
      "learning_rate": 8e-05,
      "loss": 15.0786,
      "step": 8
    },
    {
      "epoch": 0.001492289835848118,
      "grad_norm": 15.202522277832031,
      "learning_rate": 9e-05,
      "loss": 12.5367,
      "step": 9
    },
    {
      "epoch": 0.001492289835848118,
      "eval_loss": 5.437175273895264,
      "eval_runtime": 1576.7592,
      "eval_samples_per_second": 3.221,
      "eval_steps_per_second": 0.403,
      "step": 9
    },
    {
      "epoch": 0.00165809981760902,
      "grad_norm": 12.19020938873291,
      "learning_rate": 0.0001,
      "loss": 11.8082,
      "step": 10
    },
    {
      "epoch": 0.0018239097993699221,
      "grad_norm": 8.83342170715332,
      "learning_rate": 9.99695413509548e-05,
      "loss": 8.8364,
      "step": 11
    },
    {
      "epoch": 0.001989719781130824,
      "grad_norm": 9.820218086242676,
      "learning_rate": 9.987820251299122e-05,
      "loss": 7.822,
      "step": 12
    },
    {
      "epoch": 0.002155529762891726,
      "grad_norm": 10.754770278930664,
      "learning_rate": 9.972609476841367e-05,
      "loss": 7.0536,
      "step": 13
    },
    {
      "epoch": 0.002321339744652628,
      "grad_norm": 7.895251750946045,
      "learning_rate": 9.951340343707852e-05,
      "loss": 8.1585,
      "step": 14
    },
    {
      "epoch": 0.00248714972641353,
      "grad_norm": 7.87087345123291,
      "learning_rate": 9.924038765061042e-05,
      "loss": 6.6577,
      "step": 15
    },
    {
      "epoch": 0.002652959708174432,
      "grad_norm": 6.906998634338379,
      "learning_rate": 9.890738003669029e-05,
      "loss": 5.3785,
      "step": 16
    },
    {
      "epoch": 0.002818769689935334,
      "grad_norm": 8.356192588806152,
      "learning_rate": 9.851478631379982e-05,
      "loss": 4.6551,
      "step": 17
    },
    {
      "epoch": 0.002984579671696236,
      "grad_norm": 16.607709884643555,
      "learning_rate": 9.806308479691595e-05,
      "loss": 3.9705,
      "step": 18
    },
    {
      "epoch": 0.002984579671696236,
      "eval_loss": 1.9645012617111206,
      "eval_runtime": 1576.7584,
      "eval_samples_per_second": 3.221,
      "eval_steps_per_second": 0.403,
      "step": 18
    },
    {
      "epoch": 0.003150389653457138,
      "grad_norm": 15.812430381774902,
      "learning_rate": 9.755282581475769e-05,
      "loss": 3.9633,
      "step": 19
    },
    {
      "epoch": 0.00331619963521804,
      "grad_norm": 6.731838226318359,
      "learning_rate": 9.698463103929542e-05,
      "loss": 2.7544,
      "step": 20
    },
    {
      "epoch": 0.003482009616978942,
      "grad_norm": 7.793019771575928,
      "learning_rate": 9.635919272833938e-05,
      "loss": 2.309,
      "step": 21
    },
    {
      "epoch": 0.0036478195987398442,
      "grad_norm": 9.872495651245117,
      "learning_rate": 9.567727288213005e-05,
      "loss": 2.5922,
      "step": 22
    },
    {
      "epoch": 0.0038136295805007462,
      "grad_norm": 4.2099223136901855,
      "learning_rate": 9.493970231495835e-05,
      "loss": 1.186,
      "step": 23
    },
    {
      "epoch": 0.003979439562261648,
      "grad_norm": 10.548074722290039,
      "learning_rate": 9.414737964294636e-05,
      "loss": 1.5299,
      "step": 24
    },
    {
      "epoch": 0.00414524954402255,
      "grad_norm": 13.979026794433594,
      "learning_rate": 9.330127018922194e-05,
      "loss": 1.2871,
      "step": 25
    },
    {
      "epoch": 0.004311059525783452,
      "grad_norm": 7.218410968780518,
      "learning_rate": 9.24024048078213e-05,
      "loss": 1.1803,
      "step": 26
    },
    {
      "epoch": 0.004476869507544354,
      "grad_norm": 7.529348850250244,
      "learning_rate": 9.145187862775209e-05,
      "loss": 1.1097,
      "step": 27
    },
    {
      "epoch": 0.004476869507544354,
      "eval_loss": 0.4603639543056488,
      "eval_runtime": 1575.6175,
      "eval_samples_per_second": 3.223,
      "eval_steps_per_second": 0.403,
      "step": 27
    },
    {
      "epoch": 0.004642679489305256,
      "grad_norm": 3.8550705909729004,
      "learning_rate": 9.045084971874738e-05,
      "loss": 0.7259,
      "step": 28
    },
    {
      "epoch": 0.004808489471066158,
      "grad_norm": 4.565075874328613,
      "learning_rate": 8.940053768033609e-05,
      "loss": 0.865,
      "step": 29
    },
    {
      "epoch": 0.00497429945282706,
      "grad_norm": 6.3500285148620605,
      "learning_rate": 8.83022221559489e-05,
      "loss": 0.7888,
      "step": 30
    },
    {
      "epoch": 0.005140109434587962,
      "grad_norm": 5.463298320770264,
      "learning_rate": 8.715724127386972e-05,
      "loss": 0.8169,
      "step": 31
    },
    {
      "epoch": 0.005305919416348864,
      "grad_norm": 4.480053901672363,
      "learning_rate": 8.596699001693255e-05,
      "loss": 0.7127,
      "step": 32
    },
    {
      "epoch": 0.005471729398109766,
      "grad_norm": 7.863976001739502,
      "learning_rate": 8.473291852294987e-05,
      "loss": 1.048,
      "step": 33
    },
    {
      "epoch": 0.005637539379870668,
      "grad_norm": 4.669947624206543,
      "learning_rate": 8.345653031794292e-05,
      "loss": 0.6693,
      "step": 34
    },
    {
      "epoch": 0.00580334936163157,
      "grad_norm": 3.4031758308410645,
      "learning_rate": 8.213938048432697e-05,
      "loss": 0.8024,
      "step": 35
    },
    {
      "epoch": 0.005969159343392472,
      "grad_norm": 4.576297760009766,
      "learning_rate": 8.07830737662829e-05,
      "loss": 0.9314,
      "step": 36
    },
    {
      "epoch": 0.005969159343392472,
      "eval_loss": 0.4227474331855774,
      "eval_runtime": 1576.0515,
      "eval_samples_per_second": 3.223,
      "eval_steps_per_second": 0.403,
      "step": 36
    },
    {
      "epoch": 0.006134969325153374,
      "grad_norm": 4.564301490783691,
      "learning_rate": 7.938926261462366e-05,
      "loss": 1.1414,
      "step": 37
    },
    {
      "epoch": 0.006300779306914276,
      "grad_norm": 2.7618212699890137,
      "learning_rate": 7.795964517353735e-05,
      "loss": 0.6578,
      "step": 38
    },
    {
      "epoch": 0.006466589288675178,
      "grad_norm": 2.4349262714385986,
      "learning_rate": 7.649596321166024e-05,
      "loss": 0.6903,
      "step": 39
    },
    {
      "epoch": 0.00663239927043608,
      "grad_norm": 2.0584237575531006,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.6042,
      "step": 40
    },
    {
      "epoch": 0.006798209252196982,
      "grad_norm": 3.166882038116455,
      "learning_rate": 7.347357813929454e-05,
      "loss": 0.8912,
      "step": 41
    },
    {
      "epoch": 0.006964019233957884,
      "grad_norm": 3.8364858627319336,
      "learning_rate": 7.191855733945387e-05,
      "loss": 1.1009,
      "step": 42
    },
    {
      "epoch": 0.0071298292157187864,
      "grad_norm": 3.1330361366271973,
      "learning_rate": 7.033683215379002e-05,
      "loss": 0.8003,
      "step": 43
    },
    {
      "epoch": 0.0072956391974796884,
      "grad_norm": 2.2796144485473633,
      "learning_rate": 6.873032967079561e-05,
      "loss": 0.7073,
      "step": 44
    },
    {
      "epoch": 0.0074614491792405905,
      "grad_norm": 2.880812406539917,
      "learning_rate": 6.710100716628344e-05,
      "loss": 0.6036,
      "step": 45
    },
    {
      "epoch": 0.0074614491792405905,
      "eval_loss": 0.3314964473247528,
      "eval_runtime": 1575.6793,
      "eval_samples_per_second": 3.223,
      "eval_steps_per_second": 0.403,
      "step": 45
    },
    {
      "epoch": 0.0076272591610014925,
      "grad_norm": 2.4614827632904053,
      "learning_rate": 6.545084971874738e-05,
      "loss": 0.636,
      "step": 46
    },
    {
      "epoch": 0.0077930691427623945,
      "grad_norm": 1.988437294960022,
      "learning_rate": 6.378186779084995e-05,
      "loss": 0.5319,
      "step": 47
    },
    {
      "epoch": 0.007958879124523296,
      "grad_norm": 2.654332160949707,
      "learning_rate": 6.209609477998338e-05,
      "loss": 0.4373,
      "step": 48
    },
    {
      "epoch": 0.008124689106284198,
      "grad_norm": 1.7638773918151855,
      "learning_rate": 6.0395584540887963e-05,
      "loss": 0.4448,
      "step": 49
    },
    {
      "epoch": 0.0082904990880451,
      "grad_norm": 3.3107423782348633,
      "learning_rate": 5.868240888334653e-05,
      "loss": 0.5151,
      "step": 50
    },
    {
      "epoch": 0.008456309069806002,
      "grad_norm": 2.6409878730773926,
      "learning_rate": 5.695865504800327e-05,
      "loss": 0.5612,
      "step": 51
    },
    {
      "epoch": 0.008622119051566905,
      "grad_norm": 3.065234422683716,
      "learning_rate": 5.522642316338268e-05,
      "loss": 0.8439,
      "step": 52
    },
    {
      "epoch": 0.008787929033327806,
      "grad_norm": 3.303943395614624,
      "learning_rate": 5.348782368720626e-05,
      "loss": 0.9554,
      "step": 53
    },
    {
      "epoch": 0.008953739015088709,
      "grad_norm": 1.6492938995361328,
      "learning_rate": 5.174497483512506e-05,
      "loss": 0.5354,
      "step": 54
    },
    {
      "epoch": 0.008953739015088709,
      "eval_loss": 0.340312123298645,
      "eval_runtime": 1576.0534,
      "eval_samples_per_second": 3.223,
      "eval_steps_per_second": 0.403,
      "step": 54
    },
    {
      "epoch": 0.00911954899684961,
      "grad_norm": 3.8954522609710693,
      "learning_rate": 5e-05,
      "loss": 1.0515,
      "step": 55
    },
    {
      "epoch": 0.009285358978610513,
      "grad_norm": 3.292879343032837,
      "learning_rate": 4.825502516487497e-05,
      "loss": 0.8962,
      "step": 56
    },
    {
      "epoch": 0.009451168960371414,
      "grad_norm": 2.53454327583313,
      "learning_rate": 4.6512176312793736e-05,
      "loss": 0.4575,
      "step": 57
    },
    {
      "epoch": 0.009616978942132317,
      "grad_norm": 1.6766483783721924,
      "learning_rate": 4.477357683661734e-05,
      "loss": 0.4632,
      "step": 58
    },
    {
      "epoch": 0.009782788923893218,
      "grad_norm": 1.7836233377456665,
      "learning_rate": 4.3041344951996746e-05,
      "loss": 0.4928,
      "step": 59
    },
    {
      "epoch": 0.00994859890565412,
      "grad_norm": 2.2400903701782227,
      "learning_rate": 4.131759111665349e-05,
      "loss": 0.6358,
      "step": 60
    },
    {
      "epoch": 0.010114408887415022,
      "grad_norm": 1.9944827556610107,
      "learning_rate": 3.960441545911204e-05,
      "loss": 0.5992,
      "step": 61
    },
    {
      "epoch": 0.010280218869175925,
      "grad_norm": 2.788837194442749,
      "learning_rate": 3.790390522001662e-05,
      "loss": 0.5151,
      "step": 62
    },
    {
      "epoch": 0.010446028850936826,
      "grad_norm": 1.742371916770935,
      "learning_rate": 3.6218132209150045e-05,
      "loss": 0.4412,
      "step": 63
    },
    {
      "epoch": 0.010446028850936826,
      "eval_loss": 0.2952826917171478,
      "eval_runtime": 1575.7706,
      "eval_samples_per_second": 3.223,
      "eval_steps_per_second": 0.403,
      "step": 63
    },
    {
      "epoch": 0.010611838832697729,
      "grad_norm": 2.136950969696045,
      "learning_rate": 3.4549150281252636e-05,
      "loss": 0.4605,
      "step": 64
    },
    {
      "epoch": 0.01077764881445863,
      "grad_norm": 2.725506067276001,
      "learning_rate": 3.289899283371657e-05,
      "loss": 0.626,
      "step": 65
    },
    {
      "epoch": 0.010943458796219533,
      "grad_norm": 1.9136492013931274,
      "learning_rate": 3.12696703292044e-05,
      "loss": 0.5691,
      "step": 66
    },
    {
      "epoch": 0.011109268777980434,
      "grad_norm": 2.0167644023895264,
      "learning_rate": 2.9663167846209998e-05,
      "loss": 0.481,
      "step": 67
    },
    {
      "epoch": 0.011275078759741337,
      "grad_norm": 2.068988561630249,
      "learning_rate": 2.8081442660546125e-05,
      "loss": 0.5006,
      "step": 68
    },
    {
      "epoch": 0.011440888741502238,
      "grad_norm": 1.7421566247940063,
      "learning_rate": 2.6526421860705473e-05,
      "loss": 0.4953,
      "step": 69
    },
    {
      "epoch": 0.01160669872326314,
      "grad_norm": 1.764000415802002,
      "learning_rate": 2.500000000000001e-05,
      "loss": 0.4209,
      "step": 70
    },
    {
      "epoch": 0.011772508705024042,
      "grad_norm": 1.9363012313842773,
      "learning_rate": 2.350403678833976e-05,
      "loss": 0.37,
      "step": 71
    },
    {
      "epoch": 0.011938318686784945,
      "grad_norm": 3.3709402084350586,
      "learning_rate": 2.2040354826462668e-05,
      "loss": 0.5696,
      "step": 72
    },
    {
      "epoch": 0.011938318686784945,
      "eval_loss": 0.30603596568107605,
      "eval_runtime": 1575.9112,
      "eval_samples_per_second": 3.223,
      "eval_steps_per_second": 0.403,
      "step": 72
    },
    {
      "epoch": 0.012104128668545846,
      "grad_norm": 4.627692699432373,
      "learning_rate": 2.061073738537635e-05,
      "loss": 0.9927,
      "step": 73
    },
    {
      "epoch": 0.012269938650306749,
      "grad_norm": 3.018941640853882,
      "learning_rate": 1.9216926233717085e-05,
      "loss": 0.643,
      "step": 74
    },
    {
      "epoch": 0.01243574863206765,
      "grad_norm": 2.8828186988830566,
      "learning_rate": 1.7860619515673033e-05,
      "loss": 0.6989,
      "step": 75
    },
    {
      "epoch": 0.012601558613828553,
      "grad_norm": 3.3432273864746094,
      "learning_rate": 1.6543469682057106e-05,
      "loss": 0.625,
      "step": 76
    },
    {
      "epoch": 0.012767368595589454,
      "grad_norm": 3.2553389072418213,
      "learning_rate": 1.526708147705013e-05,
      "loss": 0.8164,
      "step": 77
    },
    {
      "epoch": 0.012933178577350357,
      "grad_norm": 2.3025598526000977,
      "learning_rate": 1.4033009983067452e-05,
      "loss": 0.6151,
      "step": 78
    },
    {
      "epoch": 0.013098988559111258,
      "grad_norm": 1.768186330795288,
      "learning_rate": 1.2842758726130283e-05,
      "loss": 0.4743,
      "step": 79
    },
    {
      "epoch": 0.01326479854087216,
      "grad_norm": 1.6566768884658813,
      "learning_rate": 1.1697777844051105e-05,
      "loss": 0.3514,
      "step": 80
    },
    {
      "epoch": 0.013430608522633062,
      "grad_norm": 2.122018575668335,
      "learning_rate": 1.0599462319663905e-05,
      "loss": 0.464,
      "step": 81
    },
    {
      "epoch": 0.013430608522633062,
      "eval_loss": 0.27632373571395874,
      "eval_runtime": 1575.8525,
      "eval_samples_per_second": 3.223,
      "eval_steps_per_second": 0.403,
      "step": 81
    },
    {
      "epoch": 0.013596418504393965,
      "grad_norm": 1.748769998550415,
      "learning_rate": 9.549150281252633e-06,
      "loss": 0.3143,
      "step": 82
    },
    {
      "epoch": 0.013762228486154866,
      "grad_norm": 3.251952648162842,
      "learning_rate": 8.548121372247918e-06,
      "loss": 1.0136,
      "step": 83
    },
    {
      "epoch": 0.013928038467915769,
      "grad_norm": 1.9073454141616821,
      "learning_rate": 7.597595192178702e-06,
      "loss": 0.4326,
      "step": 84
    },
    {
      "epoch": 0.01409384844967667,
      "grad_norm": 1.4053856134414673,
      "learning_rate": 6.698729810778065e-06,
      "loss": 0.4547,
      "step": 85
    },
    {
      "epoch": 0.014259658431437573,
      "grad_norm": 2.472945213317871,
      "learning_rate": 5.852620357053651e-06,
      "loss": 0.6188,
      "step": 86
    },
    {
      "epoch": 0.014425468413198474,
      "grad_norm": 2.9573841094970703,
      "learning_rate": 5.060297685041659e-06,
      "loss": 0.6689,
      "step": 87
    },
    {
      "epoch": 0.014591278394959377,
      "grad_norm": 1.7082061767578125,
      "learning_rate": 4.322727117869951e-06,
      "loss": 0.394,
      "step": 88
    },
    {
      "epoch": 0.014757088376720278,
      "grad_norm": 1.5923526287078857,
      "learning_rate": 3.6408072716606346e-06,
      "loss": 0.4601,
      "step": 89
    },
    {
      "epoch": 0.014922898358481181,
      "grad_norm": 1.5400145053863525,
      "learning_rate": 3.0153689607045845e-06,
      "loss": 0.4093,
      "step": 90
    },
    {
      "epoch": 0.014922898358481181,
      "eval_loss": 0.26483044028282166,
      "eval_runtime": 1575.8205,
      "eval_samples_per_second": 3.223,
      "eval_steps_per_second": 0.403,
      "step": 90
    },
    {
      "epoch": 0.015088708340242082,
      "grad_norm": 1.6519761085510254,
      "learning_rate": 2.4471741852423237e-06,
      "loss": 0.3969,
      "step": 91
    },
    {
      "epoch": 0.015254518322002985,
      "grad_norm": 2.0877702236175537,
      "learning_rate": 1.9369152030840556e-06,
      "loss": 0.6622,
      "step": 92
    },
    {
      "epoch": 0.015420328303763886,
      "grad_norm": 1.6852989196777344,
      "learning_rate": 1.4852136862001764e-06,
      "loss": 0.5761,
      "step": 93
    },
    {
      "epoch": 0.015586138285524789,
      "grad_norm": 1.3629008531570435,
      "learning_rate": 1.0926199633097157e-06,
      "loss": 0.3565,
      "step": 94
    },
    {
      "epoch": 0.015751948267285692,
      "grad_norm": 1.808484673500061,
      "learning_rate": 7.596123493895991e-07,
      "loss": 0.4087,
      "step": 95
    },
    {
      "epoch": 0.015917758249046593,
      "grad_norm": 1.5466543436050415,
      "learning_rate": 4.865965629214819e-07,
      "loss": 0.4395,
      "step": 96
    },
    {
      "epoch": 0.016083568230807494,
      "grad_norm": 2.435929298400879,
      "learning_rate": 2.7390523158633554e-07,
      "loss": 0.6049,
      "step": 97
    },
    {
      "epoch": 0.016249378212568395,
      "grad_norm": 1.4214541912078857,
      "learning_rate": 1.2179748700879012e-07,
      "loss": 0.4302,
      "step": 98
    },
    {
      "epoch": 0.0164151881943293,
      "grad_norm": 2.556894540786743,
      "learning_rate": 3.04586490452119e-08,
      "loss": 0.5967,
      "step": 99
    },
    {
      "epoch": 0.0164151881943293,
      "eval_loss": 0.26305386424064636,
      "eval_runtime": 1576.2459,
      "eval_samples_per_second": 3.222,
      "eval_steps_per_second": 0.403,
      "step": 99
    },
    {
      "epoch": 0.0165809981760902,
      "grad_norm": 1.7079695463180542,
      "learning_rate": 0.0,
      "loss": 0.6013,
      "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": 2.5391158788096e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}