{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.03128421711246676,
  "eval_steps": 9,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00031284217112466763,
      "grad_norm": 4.1400957107543945,
      "learning_rate": 1e-05,
      "loss": 2.2276,
      "step": 1
    },
    {
      "epoch": 0.00031284217112466763,
      "eval_loss": 2.383758544921875,
      "eval_runtime": 295.0715,
      "eval_samples_per_second": 9.123,
      "eval_steps_per_second": 1.142,
      "step": 1
    },
    {
      "epoch": 0.0006256843422493353,
      "grad_norm": 5.985204219818115,
      "learning_rate": 2e-05,
      "loss": 2.3866,
      "step": 2
    },
    {
      "epoch": 0.0009385265133740028,
      "grad_norm": 5.53966760635376,
      "learning_rate": 3e-05,
      "loss": 2.3058,
      "step": 3
    },
    {
      "epoch": 0.0012513686844986705,
      "grad_norm": 5.031322479248047,
      "learning_rate": 4e-05,
      "loss": 2.2235,
      "step": 4
    },
    {
      "epoch": 0.001564210855623338,
      "grad_norm": 5.630000114440918,
      "learning_rate": 5e-05,
      "loss": 2.4042,
      "step": 5
    },
    {
      "epoch": 0.0018770530267480056,
      "grad_norm": 5.713146209716797,
      "learning_rate": 6e-05,
      "loss": 1.7265,
      "step": 6
    },
    {
      "epoch": 0.0021898951978726734,
      "grad_norm": 4.937361240386963,
      "learning_rate": 7e-05,
      "loss": 1.235,
      "step": 7
    },
    {
      "epoch": 0.002502737368997341,
      "grad_norm": 4.11467981338501,
      "learning_rate": 8e-05,
      "loss": 1.4563,
      "step": 8
    },
    {
      "epoch": 0.0028155795401220087,
      "grad_norm": 3.5735771656036377,
      "learning_rate": 9e-05,
      "loss": 1.0858,
      "step": 9
    },
    {
      "epoch": 0.0028155795401220087,
      "eval_loss": 0.9501042366027832,
      "eval_runtime": 295.5459,
      "eval_samples_per_second": 9.109,
      "eval_steps_per_second": 1.14,
      "step": 9
    },
    {
      "epoch": 0.003128421711246676,
      "grad_norm": 3.4599525928497314,
      "learning_rate": 0.0001,
      "loss": 0.9475,
      "step": 10
    },
    {
      "epoch": 0.0034412638823713435,
      "grad_norm": 3.0647709369659424,
      "learning_rate": 9.99695413509548e-05,
      "loss": 0.5548,
      "step": 11
    },
    {
      "epoch": 0.003754106053496011,
      "grad_norm": 2.7285585403442383,
      "learning_rate": 9.987820251299122e-05,
      "loss": 0.6622,
      "step": 12
    },
    {
      "epoch": 0.004066948224620679,
      "grad_norm": 3.6022748947143555,
      "learning_rate": 9.972609476841367e-05,
      "loss": 0.842,
      "step": 13
    },
    {
      "epoch": 0.004379790395745347,
      "grad_norm": 2.55991792678833,
      "learning_rate": 9.951340343707852e-05,
      "loss": 0.5361,
      "step": 14
    },
    {
      "epoch": 0.004692632566870014,
      "grad_norm": 2.4798598289489746,
      "learning_rate": 9.924038765061042e-05,
      "loss": 0.8465,
      "step": 15
    },
    {
      "epoch": 0.005005474737994682,
      "grad_norm": 2.140596389770508,
      "learning_rate": 9.890738003669029e-05,
      "loss": 0.3703,
      "step": 16
    },
    {
      "epoch": 0.005318316909119349,
      "grad_norm": 1.6395519971847534,
      "learning_rate": 9.851478631379982e-05,
      "loss": 0.3486,
      "step": 17
    },
    {
      "epoch": 0.005631159080244017,
      "grad_norm": 2.1203126907348633,
      "learning_rate": 9.806308479691595e-05,
      "loss": 0.5283,
      "step": 18
    },
    {
      "epoch": 0.005631159080244017,
      "eval_loss": 0.6347249746322632,
      "eval_runtime": 295.3996,
      "eval_samples_per_second": 9.113,
      "eval_steps_per_second": 1.141,
      "step": 18
    },
    {
      "epoch": 0.0059440012513686845,
      "grad_norm": 3.085947275161743,
      "learning_rate": 9.755282581475769e-05,
      "loss": 0.6992,
      "step": 19
    },
    {
      "epoch": 0.006256843422493352,
      "grad_norm": 2.1405577659606934,
      "learning_rate": 9.698463103929542e-05,
      "loss": 0.6296,
      "step": 20
    },
    {
      "epoch": 0.00656968559361802,
      "grad_norm": 2.090209722518921,
      "learning_rate": 9.635919272833938e-05,
      "loss": 0.5551,
      "step": 21
    },
    {
      "epoch": 0.006882527764742687,
      "grad_norm": 2.086621046066284,
      "learning_rate": 9.567727288213005e-05,
      "loss": 0.4639,
      "step": 22
    },
    {
      "epoch": 0.007195369935867355,
      "grad_norm": 1.945451259613037,
      "learning_rate": 9.493970231495835e-05,
      "loss": 0.483,
      "step": 23
    },
    {
      "epoch": 0.007508212106992022,
      "grad_norm": 2.0879640579223633,
      "learning_rate": 9.414737964294636e-05,
      "loss": 0.5471,
      "step": 24
    },
    {
      "epoch": 0.00782105427811669,
      "grad_norm": 2.2464821338653564,
      "learning_rate": 9.330127018922194e-05,
      "loss": 0.661,
      "step": 25
    },
    {
      "epoch": 0.008133896449241357,
      "grad_norm": 2.419508934020996,
      "learning_rate": 9.24024048078213e-05,
      "loss": 0.6773,
      "step": 26
    },
    {
      "epoch": 0.008446738620366025,
      "grad_norm": 1.4543237686157227,
      "learning_rate": 9.145187862775209e-05,
      "loss": 0.4666,
      "step": 27
    },
    {
      "epoch": 0.008446738620366025,
      "eval_loss": 0.5433617830276489,
      "eval_runtime": 295.6491,
      "eval_samples_per_second": 9.105,
      "eval_steps_per_second": 1.14,
      "step": 27
    },
    {
      "epoch": 0.008759580791490694,
      "grad_norm": 1.9048430919647217,
      "learning_rate": 9.045084971874738e-05,
      "loss": 0.6091,
      "step": 28
    },
    {
      "epoch": 0.00907242296261536,
      "grad_norm": 1.299614667892456,
      "learning_rate": 8.940053768033609e-05,
      "loss": 0.4118,
      "step": 29
    },
    {
      "epoch": 0.009385265133740028,
      "grad_norm": 1.730533480644226,
      "learning_rate": 8.83022221559489e-05,
      "loss": 0.5465,
      "step": 30
    },
    {
      "epoch": 0.009698107304864695,
      "grad_norm": 2.2634334564208984,
      "learning_rate": 8.715724127386972e-05,
      "loss": 0.3919,
      "step": 31
    },
    {
      "epoch": 0.010010949475989364,
      "grad_norm": 1.92788565158844,
      "learning_rate": 8.596699001693255e-05,
      "loss": 0.7134,
      "step": 32
    },
    {
      "epoch": 0.010323791647114031,
      "grad_norm": 1.7542908191680908,
      "learning_rate": 8.473291852294987e-05,
      "loss": 0.4754,
      "step": 33
    },
    {
      "epoch": 0.010636633818238699,
      "grad_norm": 1.591113567352295,
      "learning_rate": 8.345653031794292e-05,
      "loss": 0.3841,
      "step": 34
    },
    {
      "epoch": 0.010949475989363366,
      "grad_norm": 1.8184356689453125,
      "learning_rate": 8.213938048432697e-05,
      "loss": 0.4843,
      "step": 35
    },
    {
      "epoch": 0.011262318160488035,
      "grad_norm": 1.6940315961837769,
      "learning_rate": 8.07830737662829e-05,
      "loss": 0.4915,
      "step": 36
    },
    {
      "epoch": 0.011262318160488035,
      "eval_loss": 0.4910644590854645,
      "eval_runtime": 295.7718,
      "eval_samples_per_second": 9.102,
      "eval_steps_per_second": 1.139,
      "step": 36
    },
    {
      "epoch": 0.011575160331612702,
      "grad_norm": 1.6678466796875,
      "learning_rate": 7.938926261462366e-05,
      "loss": 0.5623,
      "step": 37
    },
    {
      "epoch": 0.011888002502737369,
      "grad_norm": 1.7581409215927124,
      "learning_rate": 7.795964517353735e-05,
      "loss": 0.4988,
      "step": 38
    },
    {
      "epoch": 0.012200844673862036,
      "grad_norm": 1.9118449687957764,
      "learning_rate": 7.649596321166024e-05,
      "loss": 0.7088,
      "step": 39
    },
    {
      "epoch": 0.012513686844986703,
      "grad_norm": 1.7057359218597412,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.6083,
      "step": 40
    },
    {
      "epoch": 0.012826529016111372,
      "grad_norm": 1.2241131067276,
      "learning_rate": 7.347357813929454e-05,
      "loss": 0.2621,
      "step": 41
    },
    {
      "epoch": 0.01313937118723604,
      "grad_norm": 1.8495560884475708,
      "learning_rate": 7.191855733945387e-05,
      "loss": 0.6354,
      "step": 42
    },
    {
      "epoch": 0.013452213358360707,
      "grad_norm": 1.4551031589508057,
      "learning_rate": 7.033683215379002e-05,
      "loss": 0.4695,
      "step": 43
    },
    {
      "epoch": 0.013765055529485374,
      "grad_norm": 1.7759051322937012,
      "learning_rate": 6.873032967079561e-05,
      "loss": 0.6415,
      "step": 44
    },
    {
      "epoch": 0.014077897700610043,
      "grad_norm": 1.549106240272522,
      "learning_rate": 6.710100716628344e-05,
      "loss": 0.4468,
      "step": 45
    },
    {
      "epoch": 0.014077897700610043,
      "eval_loss": 0.45826274156570435,
      "eval_runtime": 295.6162,
      "eval_samples_per_second": 9.106,
      "eval_steps_per_second": 1.14,
      "step": 45
    },
    {
      "epoch": 0.01439073987173471,
      "grad_norm": 2.1609103679656982,
      "learning_rate": 6.545084971874738e-05,
      "loss": 0.4882,
      "step": 46
    },
    {
      "epoch": 0.014703582042859377,
      "grad_norm": 1.3817501068115234,
      "learning_rate": 6.378186779084995e-05,
      "loss": 0.4709,
      "step": 47
    },
    {
      "epoch": 0.015016424213984044,
      "grad_norm": 1.4052536487579346,
      "learning_rate": 6.209609477998338e-05,
      "loss": 0.3078,
      "step": 48
    },
    {
      "epoch": 0.015329266385108713,
      "grad_norm": 1.2172939777374268,
      "learning_rate": 6.0395584540887963e-05,
      "loss": 0.2555,
      "step": 49
    },
    {
      "epoch": 0.01564210855623338,
      "grad_norm": 1.2417101860046387,
      "learning_rate": 5.868240888334653e-05,
      "loss": 0.5631,
      "step": 50
    },
    {
      "epoch": 0.015954950727358048,
      "grad_norm": 1.4559011459350586,
      "learning_rate": 5.695865504800327e-05,
      "loss": 0.5721,
      "step": 51
    },
    {
      "epoch": 0.016267792898482715,
      "grad_norm": 1.5319404602050781,
      "learning_rate": 5.522642316338268e-05,
      "loss": 0.4725,
      "step": 52
    },
    {
      "epoch": 0.016580635069607382,
      "grad_norm": 1.462174654006958,
      "learning_rate": 5.348782368720626e-05,
      "loss": 0.4748,
      "step": 53
    },
    {
      "epoch": 0.01689347724073205,
      "grad_norm": 1.3655391931533813,
      "learning_rate": 5.174497483512506e-05,
      "loss": 0.2604,
      "step": 54
    },
    {
      "epoch": 0.01689347724073205,
      "eval_loss": 0.44369691610336304,
      "eval_runtime": 295.2225,
      "eval_samples_per_second": 9.119,
      "eval_steps_per_second": 1.142,
      "step": 54
    },
    {
      "epoch": 0.017206319411856717,
      "grad_norm": 1.3586876392364502,
      "learning_rate": 5e-05,
      "loss": 0.4544,
      "step": 55
    },
    {
      "epoch": 0.017519161582981387,
      "grad_norm": 1.2819782495498657,
      "learning_rate": 4.825502516487497e-05,
      "loss": 0.4211,
      "step": 56
    },
    {
      "epoch": 0.017832003754106054,
      "grad_norm": 0.91957688331604,
      "learning_rate": 4.6512176312793736e-05,
      "loss": 0.2017,
      "step": 57
    },
    {
      "epoch": 0.01814484592523072,
      "grad_norm": 1.4410914182662964,
      "learning_rate": 4.477357683661734e-05,
      "loss": 0.3749,
      "step": 58
    },
    {
      "epoch": 0.01845768809635539,
      "grad_norm": 1.5840656757354736,
      "learning_rate": 4.3041344951996746e-05,
      "loss": 0.5614,
      "step": 59
    },
    {
      "epoch": 0.018770530267480056,
      "grad_norm": 1.4971287250518799,
      "learning_rate": 4.131759111665349e-05,
      "loss": 0.453,
      "step": 60
    },
    {
      "epoch": 0.019083372438604723,
      "grad_norm": 1.9114784002304077,
      "learning_rate": 3.960441545911204e-05,
      "loss": 0.4862,
      "step": 61
    },
    {
      "epoch": 0.01939621460972939,
      "grad_norm": 1.614539623260498,
      "learning_rate": 3.790390522001662e-05,
      "loss": 0.4331,
      "step": 62
    },
    {
      "epoch": 0.019709056780854058,
      "grad_norm": 2.0431880950927734,
      "learning_rate": 3.6218132209150045e-05,
      "loss": 0.5591,
      "step": 63
    },
    {
      "epoch": 0.019709056780854058,
      "eval_loss": 0.43188172578811646,
      "eval_runtime": 295.1825,
      "eval_samples_per_second": 9.12,
      "eval_steps_per_second": 1.142,
      "step": 63
    },
    {
      "epoch": 0.020021898951978728,
      "grad_norm": 1.3552390336990356,
      "learning_rate": 3.4549150281252636e-05,
      "loss": 0.3512,
      "step": 64
    },
    {
      "epoch": 0.020334741123103395,
      "grad_norm": 1.4610017538070679,
      "learning_rate": 3.289899283371657e-05,
      "loss": 0.331,
      "step": 65
    },
    {
      "epoch": 0.020647583294228063,
      "grad_norm": 1.2008382081985474,
      "learning_rate": 3.12696703292044e-05,
      "loss": 0.3855,
      "step": 66
    },
    {
      "epoch": 0.02096042546535273,
      "grad_norm": 0.9755795001983643,
      "learning_rate": 2.9663167846209998e-05,
      "loss": 0.2343,
      "step": 67
    },
    {
      "epoch": 0.021273267636477397,
      "grad_norm": 1.3964710235595703,
      "learning_rate": 2.8081442660546125e-05,
      "loss": 0.4639,
      "step": 68
    },
    {
      "epoch": 0.021586109807602064,
      "grad_norm": 1.635550856590271,
      "learning_rate": 2.6526421860705473e-05,
      "loss": 0.4901,
      "step": 69
    },
    {
      "epoch": 0.02189895197872673,
      "grad_norm": 1.1316291093826294,
      "learning_rate": 2.500000000000001e-05,
      "loss": 0.2346,
      "step": 70
    },
    {
      "epoch": 0.0222117941498514,
      "grad_norm": 1.0997239351272583,
      "learning_rate": 2.350403678833976e-05,
      "loss": 0.2798,
      "step": 71
    },
    {
      "epoch": 0.02252463632097607,
      "grad_norm": 1.0223768949508667,
      "learning_rate": 2.2040354826462668e-05,
      "loss": 0.198,
      "step": 72
    },
    {
      "epoch": 0.02252463632097607,
      "eval_loss": 0.42166417837142944,
      "eval_runtime": 295.1343,
      "eval_samples_per_second": 9.121,
      "eval_steps_per_second": 1.142,
      "step": 72
    },
    {
      "epoch": 0.022837478492100736,
      "grad_norm": 1.3145177364349365,
      "learning_rate": 2.061073738537635e-05,
      "loss": 0.5564,
      "step": 73
    },
    {
      "epoch": 0.023150320663225404,
      "grad_norm": 1.4026906490325928,
      "learning_rate": 1.9216926233717085e-05,
      "loss": 0.4435,
      "step": 74
    },
    {
      "epoch": 0.02346316283435007,
      "grad_norm": 1.1671936511993408,
      "learning_rate": 1.7860619515673033e-05,
      "loss": 0.4185,
      "step": 75
    },
    {
      "epoch": 0.023776005005474738,
      "grad_norm": 1.5238317251205444,
      "learning_rate": 1.6543469682057106e-05,
      "loss": 0.4236,
      "step": 76
    },
    {
      "epoch": 0.024088847176599405,
      "grad_norm": 1.646121621131897,
      "learning_rate": 1.526708147705013e-05,
      "loss": 0.4858,
      "step": 77
    },
    {
      "epoch": 0.024401689347724072,
      "grad_norm": 1.8075958490371704,
      "learning_rate": 1.4033009983067452e-05,
      "loss": 0.4913,
      "step": 78
    },
    {
      "epoch": 0.02471453151884874,
      "grad_norm": 1.2012605667114258,
      "learning_rate": 1.2842758726130283e-05,
      "loss": 0.3215,
      "step": 79
    },
    {
      "epoch": 0.025027373689973407,
      "grad_norm": 1.520753026008606,
      "learning_rate": 1.1697777844051105e-05,
      "loss": 0.3818,
      "step": 80
    },
    {
      "epoch": 0.025340215861098078,
      "grad_norm": 1.7736116647720337,
      "learning_rate": 1.0599462319663905e-05,
      "loss": 0.62,
      "step": 81
    },
    {
      "epoch": 0.025340215861098078,
      "eval_loss": 0.41683319211006165,
      "eval_runtime": 295.3198,
      "eval_samples_per_second": 9.116,
      "eval_steps_per_second": 1.141,
      "step": 81
    },
    {
      "epoch": 0.025653058032222745,
      "grad_norm": 1.5321831703186035,
      "learning_rate": 9.549150281252633e-06,
      "loss": 0.3879,
      "step": 82
    },
    {
      "epoch": 0.025965900203347412,
      "grad_norm": 1.1673277616500854,
      "learning_rate": 8.548121372247918e-06,
      "loss": 0.4108,
      "step": 83
    },
    {
      "epoch": 0.02627874237447208,
      "grad_norm": 1.4989150762557983,
      "learning_rate": 7.597595192178702e-06,
      "loss": 0.5533,
      "step": 84
    },
    {
      "epoch": 0.026591584545596746,
      "grad_norm": 1.5603235960006714,
      "learning_rate": 6.698729810778065e-06,
      "loss": 0.6211,
      "step": 85
    },
    {
      "epoch": 0.026904426716721414,
      "grad_norm": 1.4086264371871948,
      "learning_rate": 5.852620357053651e-06,
      "loss": 0.2919,
      "step": 86
    },
    {
      "epoch": 0.02721726888784608,
      "grad_norm": 1.7630650997161865,
      "learning_rate": 5.060297685041659e-06,
      "loss": 0.7224,
      "step": 87
    },
    {
      "epoch": 0.027530111058970748,
      "grad_norm": 1.5985987186431885,
      "learning_rate": 4.322727117869951e-06,
      "loss": 0.5151,
      "step": 88
    },
    {
      "epoch": 0.02784295323009542,
      "grad_norm": 1.4477020502090454,
      "learning_rate": 3.6408072716606346e-06,
      "loss": 0.3901,
      "step": 89
    },
    {
      "epoch": 0.028155795401220086,
      "grad_norm": 1.2764562368392944,
      "learning_rate": 3.0153689607045845e-06,
      "loss": 0.3503,
      "step": 90
    },
    {
      "epoch": 0.028155795401220086,
      "eval_loss": 0.4124401807785034,
      "eval_runtime": 295.306,
      "eval_samples_per_second": 9.116,
      "eval_steps_per_second": 1.141,
      "step": 90
    },
    {
      "epoch": 0.028468637572344753,
      "grad_norm": 1.822226643562317,
      "learning_rate": 2.4471741852423237e-06,
      "loss": 0.7625,
      "step": 91
    },
    {
      "epoch": 0.02878147974346942,
      "grad_norm": 1.6981431245803833,
      "learning_rate": 1.9369152030840556e-06,
      "loss": 0.4653,
      "step": 92
    },
    {
      "epoch": 0.029094321914594087,
      "grad_norm": 1.470794916152954,
      "learning_rate": 1.4852136862001764e-06,
      "loss": 0.4396,
      "step": 93
    },
    {
      "epoch": 0.029407164085718755,
      "grad_norm": 1.1165651082992554,
      "learning_rate": 1.0926199633097157e-06,
      "loss": 0.2726,
      "step": 94
    },
    {
      "epoch": 0.02972000625684342,
      "grad_norm": 1.0978894233703613,
      "learning_rate": 7.596123493895991e-07,
      "loss": 0.3345,
      "step": 95
    },
    {
      "epoch": 0.03003284842796809,
      "grad_norm": 0.906370222568512,
      "learning_rate": 4.865965629214819e-07,
      "loss": 0.21,
      "step": 96
    },
    {
      "epoch": 0.030345690599092756,
      "grad_norm": 1.5754798650741577,
      "learning_rate": 2.7390523158633554e-07,
      "loss": 0.5051,
      "step": 97
    },
    {
      "epoch": 0.030658532770217427,
      "grad_norm": 1.6001464128494263,
      "learning_rate": 1.2179748700879012e-07,
      "loss": 0.4832,
      "step": 98
    },
    {
      "epoch": 0.030971374941342094,
      "grad_norm": 1.4036067724227905,
      "learning_rate": 3.04586490452119e-08,
      "loss": 0.3839,
      "step": 99
    },
    {
      "epoch": 0.030971374941342094,
      "eval_loss": 0.41170284152030945,
      "eval_runtime": 295.3304,
      "eval_samples_per_second": 9.115,
      "eval_steps_per_second": 1.141,
      "step": 99
    },
    {
      "epoch": 0.03128421711246676,
      "grad_norm": 1.121339201927185,
      "learning_rate": 0.0,
      "loss": 0.3074,
      "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
}