{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 9.728622631848438,
  "eval_steps": 99999999,
  "global_step": 19000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.051203277009728626,
      "grad_norm": 4.247892379760742,
      "learning_rate": 2.6315789473684212e-05,
      "loss": 7.4934,
      "step": 100
    },
    {
      "epoch": 0.10240655401945725,
      "grad_norm": 3.517791748046875,
      "learning_rate": 4.997341839447103e-05,
      "loss": 5.579,
      "step": 200
    },
    {
      "epoch": 0.10240655401945725,
      "eval_accuracy": 0.31893968280934165,
      "eval_loss": 4.767695903778076,
      "eval_runtime": 94.8557,
      "eval_samples_per_second": 353.632,
      "eval_steps_per_second": 22.107,
      "step": 200
    },
    {
      "epoch": 0.15360983102918588,
      "grad_norm": 2.3053529262542725,
      "learning_rate": 4.970760233918128e-05,
      "loss": 4.9703,
      "step": 300
    },
    {
      "epoch": 0.2048131080389145,
      "grad_norm": 2.0182411670684814,
      "learning_rate": 4.944178628389155e-05,
      "loss": 4.7716,
      "step": 400
    },
    {
      "epoch": 0.2048131080389145,
      "eval_accuracy": 0.354354102424925,
      "eval_loss": 4.33852481842041,
      "eval_runtime": 95.6535,
      "eval_samples_per_second": 350.682,
      "eval_steps_per_second": 21.923,
      "step": 400
    },
    {
      "epoch": 0.2560163850486431,
      "grad_norm": 1.6917961835861206,
      "learning_rate": 4.917597022860181e-05,
      "loss": 4.5491,
      "step": 500
    },
    {
      "epoch": 0.30721966205837176,
      "grad_norm": 1.3638228178024292,
      "learning_rate": 4.8910154173312074e-05,
      "loss": 4.5162,
      "step": 600
    },
    {
      "epoch": 0.30721966205837176,
      "eval_accuracy": 0.35930259313000934,
      "eval_loss": 4.177200794219971,
      "eval_runtime": 95.6652,
      "eval_samples_per_second": 350.64,
      "eval_steps_per_second": 21.92,
      "step": 600
    },
    {
      "epoch": 0.35842293906810035,
      "grad_norm": 1.5296661853790283,
      "learning_rate": 4.8644338118022334e-05,
      "loss": 4.5039,
      "step": 700
    },
    {
      "epoch": 0.409626216077829,
      "grad_norm": 1.1248077154159546,
      "learning_rate": 4.837852206273259e-05,
      "loss": 4.4056,
      "step": 800
    },
    {
      "epoch": 0.409626216077829,
      "eval_accuracy": 0.36928340870045734,
      "eval_loss": 4.075441837310791,
      "eval_runtime": 96.0011,
      "eval_samples_per_second": 349.412,
      "eval_steps_per_second": 21.843,
      "step": 800
    },
    {
      "epoch": 0.4608294930875576,
      "grad_norm": 1.6162877082824707,
      "learning_rate": 4.811270600744285e-05,
      "loss": 4.3569,
      "step": 900
    },
    {
      "epoch": 0.5120327700972862,
      "grad_norm": 1.6072983741760254,
      "learning_rate": 4.784688995215311e-05,
      "loss": 4.3138,
      "step": 1000
    },
    {
      "epoch": 0.5120327700972862,
      "eval_accuracy": 0.3626070708659433,
      "eval_loss": 4.014294624328613,
      "eval_runtime": 95.7344,
      "eval_samples_per_second": 350.386,
      "eval_steps_per_second": 21.904,
      "step": 1000
    },
    {
      "epoch": 0.5632360471070148,
      "grad_norm": 1.4564248323440552,
      "learning_rate": 4.758107389686337e-05,
      "loss": 4.2611,
      "step": 1100
    },
    {
      "epoch": 0.6144393241167435,
      "grad_norm": 1.2788530588150024,
      "learning_rate": 4.731525784157364e-05,
      "loss": 4.2148,
      "step": 1200
    },
    {
      "epoch": 0.6144393241167435,
      "eval_accuracy": 0.35539686636426476,
      "eval_loss": 3.9600954055786133,
      "eval_runtime": 95.926,
      "eval_samples_per_second": 349.686,
      "eval_steps_per_second": 21.861,
      "step": 1200
    },
    {
      "epoch": 0.6656426011264721,
      "grad_norm": 1.3278919458389282,
      "learning_rate": 4.7049441786283896e-05,
      "loss": 4.2095,
      "step": 1300
    },
    {
      "epoch": 0.7168458781362007,
      "grad_norm": 1.6991857290267944,
      "learning_rate": 4.678362573099415e-05,
      "loss": 4.1925,
      "step": 1400
    },
    {
      "epoch": 0.7168458781362007,
      "eval_accuracy": 0.3723133981106336,
      "eval_loss": 3.901859998703003,
      "eval_runtime": 95.2539,
      "eval_samples_per_second": 352.154,
      "eval_steps_per_second": 22.015,
      "step": 1400
    },
    {
      "epoch": 0.7680491551459293,
      "grad_norm": 1.7155863046646118,
      "learning_rate": 4.6517809675704415e-05,
      "loss": 4.1556,
      "step": 1500
    },
    {
      "epoch": 0.819252432155658,
      "grad_norm": 1.6039901971817017,
      "learning_rate": 4.6251993620414674e-05,
      "loss": 4.0293,
      "step": 1600
    },
    {
      "epoch": 0.819252432155658,
      "eval_accuracy": 0.37494784430112066,
      "eval_loss": 3.857882499694824,
      "eval_runtime": 95.7224,
      "eval_samples_per_second": 350.43,
      "eval_steps_per_second": 21.907,
      "step": 1600
    },
    {
      "epoch": 0.8704557091653866,
      "grad_norm": 1.6892971992492676,
      "learning_rate": 4.598617756512494e-05,
      "loss": 3.939,
      "step": 1700
    },
    {
      "epoch": 0.9216589861751152,
      "grad_norm": 1.5192146301269531,
      "learning_rate": 4.57203615098352e-05,
      "loss": 3.9407,
      "step": 1800
    },
    {
      "epoch": 0.9216589861751152,
      "eval_accuracy": 0.37819625757774467,
      "eval_loss": 3.8101186752319336,
      "eval_runtime": 95.9641,
      "eval_samples_per_second": 349.547,
      "eval_steps_per_second": 21.852,
      "step": 1800
    },
    {
      "epoch": 0.9728622631848438,
      "grad_norm": 1.6878364086151123,
      "learning_rate": 4.545454545454546e-05,
      "loss": 3.9125,
      "step": 1900
    },
    {
      "epoch": 1.0240655401945724,
      "grad_norm": 1.6546193361282349,
      "learning_rate": 4.518872939925572e-05,
      "loss": 3.8371,
      "step": 2000
    },
    {
      "epoch": 1.0240655401945724,
      "eval_accuracy": 0.37209030706755225,
      "eval_loss": 3.787036180496216,
      "eval_runtime": 95.7788,
      "eval_samples_per_second": 350.224,
      "eval_steps_per_second": 21.894,
      "step": 2000
    },
    {
      "epoch": 1.075268817204301,
      "grad_norm": 1.744416356086731,
      "learning_rate": 4.492291334396598e-05,
      "loss": 3.8068,
      "step": 2100
    },
    {
      "epoch": 1.1264720942140296,
      "grad_norm": 1.6214791536331177,
      "learning_rate": 4.4657097288676236e-05,
      "loss": 3.7642,
      "step": 2200
    },
    {
      "epoch": 1.1776753712237582,
      "grad_norm": 1.5377932786941528,
      "learning_rate": 4.43912812333865e-05,
      "loss": 3.7599,
      "step": 2300
    },
    {
      "epoch": 1.228878648233487,
      "grad_norm": 1.8464128971099854,
      "learning_rate": 4.412546517809676e-05,
      "loss": 3.6853,
      "step": 2400
    },
    {
      "epoch": 1.2800819252432156,
      "grad_norm": 1.901739239692688,
      "learning_rate": 4.3859649122807014e-05,
      "loss": 3.6537,
      "step": 2500
    },
    {
      "epoch": 1.3312852022529442,
      "grad_norm": 1.936339020729065,
      "learning_rate": 4.359383306751728e-05,
      "loss": 3.6377,
      "step": 2600
    },
    {
      "epoch": 1.3824884792626728,
      "grad_norm": 1.8641879558563232,
      "learning_rate": 4.332801701222754e-05,
      "loss": 3.5523,
      "step": 2700
    },
    {
      "epoch": 1.4336917562724014,
      "grad_norm": 2.087099075317383,
      "learning_rate": 4.3062200956937806e-05,
      "loss": 3.5007,
      "step": 2800
    },
    {
      "epoch": 1.48489503328213,
      "grad_norm": 1.7824835777282715,
      "learning_rate": 4.2796384901648065e-05,
      "loss": 3.4422,
      "step": 2900
    },
    {
      "epoch": 1.5360983102918588,
      "grad_norm": 2.0298142433166504,
      "learning_rate": 4.253056884635832e-05,
      "loss": 3.4254,
      "step": 3000
    },
    {
      "epoch": 1.5873015873015874,
      "grad_norm": 2.227325439453125,
      "learning_rate": 4.2264752791068584e-05,
      "loss": 3.4087,
      "step": 3100
    },
    {
      "epoch": 1.638504864311316,
      "grad_norm": 2.2700014114379883,
      "learning_rate": 4.199893673577884e-05,
      "loss": 3.4023,
      "step": 3200
    },
    {
      "epoch": 1.6897081413210446,
      "grad_norm": 2.290431022644043,
      "learning_rate": 4.17331206804891e-05,
      "loss": 3.3579,
      "step": 3300
    },
    {
      "epoch": 1.7409114183307732,
      "grad_norm": 2.5916295051574707,
      "learning_rate": 4.146730462519937e-05,
      "loss": 3.2602,
      "step": 3400
    },
    {
      "epoch": 1.7921146953405018,
      "grad_norm": 2.0242862701416016,
      "learning_rate": 4.120148856990963e-05,
      "loss": 3.1941,
      "step": 3500
    },
    {
      "epoch": 1.8433179723502304,
      "grad_norm": 1.8922170400619507,
      "learning_rate": 4.093567251461988e-05,
      "loss": 3.1774,
      "step": 3600
    },
    {
      "epoch": 1.894521249359959,
      "grad_norm": 2.2697248458862305,
      "learning_rate": 4.0669856459330146e-05,
      "loss": 3.2001,
      "step": 3700
    },
    {
      "epoch": 1.9457245263696876,
      "grad_norm": 2.68994402885437,
      "learning_rate": 4.0404040404040405e-05,
      "loss": 3.1232,
      "step": 3800
    },
    {
      "epoch": 1.9969278033794162,
      "grad_norm": 2.16379451751709,
      "learning_rate": 4.0138224348750665e-05,
      "loss": 3.101,
      "step": 3900
    },
    {
      "epoch": 2.048131080389145,
      "grad_norm": 1.908273458480835,
      "learning_rate": 3.987240829346093e-05,
      "loss": 3.0659,
      "step": 4000
    },
    {
      "epoch": 2.048131080389145,
      "eval_accuracy": 0.4085060110901451,
      "eval_loss": 3.467172384262085,
      "eval_runtime": 95.3375,
      "eval_samples_per_second": 351.845,
      "eval_steps_per_second": 21.996,
      "step": 4000
    },
    {
      "epoch": 2.0993343573988734,
      "grad_norm": 2.217175006866455,
      "learning_rate": 3.960659223817118e-05,
      "loss": 3.009,
      "step": 4100
    },
    {
      "epoch": 2.150537634408602,
      "grad_norm": 2.744868755340576,
      "learning_rate": 3.934077618288145e-05,
      "loss": 3.0265,
      "step": 4200
    },
    {
      "epoch": 2.2017409114183306,
      "grad_norm": 2.3460192680358887,
      "learning_rate": 3.907496012759171e-05,
      "loss": 2.9379,
      "step": 4300
    },
    {
      "epoch": 2.252944188428059,
      "grad_norm": 2.120131015777588,
      "learning_rate": 3.880914407230197e-05,
      "loss": 2.9437,
      "step": 4400
    },
    {
      "epoch": 2.3041474654377883,
      "grad_norm": 2.3475334644317627,
      "learning_rate": 3.8543328017012234e-05,
      "loss": 2.933,
      "step": 4500
    },
    {
      "epoch": 2.3553507424475164,
      "grad_norm": 2.450538396835327,
      "learning_rate": 3.8277511961722486e-05,
      "loss": 2.9619,
      "step": 4600
    },
    {
      "epoch": 2.4065540194572455,
      "grad_norm": 2.027897596359253,
      "learning_rate": 3.8011695906432746e-05,
      "loss": 2.9179,
      "step": 4700
    },
    {
      "epoch": 2.457757296466974,
      "grad_norm": 2.3188517093658447,
      "learning_rate": 3.774587985114301e-05,
      "loss": 2.8655,
      "step": 4800
    },
    {
      "epoch": 2.5089605734767026,
      "grad_norm": 1.909847617149353,
      "learning_rate": 3.748006379585327e-05,
      "loss": 2.8649,
      "step": 4900
    },
    {
      "epoch": 2.5601638504864312,
      "grad_norm": 1.8613382577896118,
      "learning_rate": 3.721424774056353e-05,
      "loss": 2.8555,
      "step": 5000
    },
    {
      "epoch": 2.61136712749616,
      "grad_norm": 1.7689636945724487,
      "learning_rate": 3.6948431685273796e-05,
      "loss": 2.8957,
      "step": 5100
    },
    {
      "epoch": 2.6625704045058884,
      "grad_norm": 1.8662793636322021,
      "learning_rate": 3.668261562998405e-05,
      "loss": 2.8132,
      "step": 5200
    },
    {
      "epoch": 2.713773681515617,
      "grad_norm": 1.9725576639175415,
      "learning_rate": 3.6416799574694315e-05,
      "loss": 2.8121,
      "step": 5300
    },
    {
      "epoch": 2.7649769585253456,
      "grad_norm": 1.7306541204452515,
      "learning_rate": 3.6150983519404574e-05,
      "loss": 2.8154,
      "step": 5400
    },
    {
      "epoch": 2.8161802355350742,
      "grad_norm": 1.824539065361023,
      "learning_rate": 3.5885167464114834e-05,
      "loss": 2.776,
      "step": 5500
    },
    {
      "epoch": 2.867383512544803,
      "grad_norm": 1.9677956104278564,
      "learning_rate": 3.56193514088251e-05,
      "loss": 2.8155,
      "step": 5600
    },
    {
      "epoch": 2.9185867895545314,
      "grad_norm": 1.9770982265472412,
      "learning_rate": 3.535353535353535e-05,
      "loss": 2.7638,
      "step": 5700
    },
    {
      "epoch": 2.96979006656426,
      "grad_norm": 2.1280829906463623,
      "learning_rate": 3.508771929824561e-05,
      "loss": 2.7771,
      "step": 5800
    },
    {
      "epoch": 3.0209933435739886,
      "grad_norm": 1.7677674293518066,
      "learning_rate": 3.482190324295588e-05,
      "loss": 2.7382,
      "step": 5900
    },
    {
      "epoch": 3.0721966205837172,
      "grad_norm": 1.6372385025024414,
      "learning_rate": 3.455608718766614e-05,
      "loss": 2.6866,
      "step": 6000
    },
    {
      "epoch": 3.0721966205837172,
      "eval_accuracy": 0.4316442393272171,
      "eval_loss": 3.2849764823913574,
      "eval_runtime": 95.1067,
      "eval_samples_per_second": 352.699,
      "eval_steps_per_second": 22.049,
      "step": 6000
    },
    {
      "epoch": 3.123399897593446,
      "grad_norm": 1.9155375957489014,
      "learning_rate": 3.4290271132376396e-05,
      "loss": 2.6794,
      "step": 6100
    },
    {
      "epoch": 3.1746031746031744,
      "grad_norm": 1.5887531042099,
      "learning_rate": 3.402445507708666e-05,
      "loss": 2.6785,
      "step": 6200
    },
    {
      "epoch": 3.225806451612903,
      "grad_norm": 1.7831392288208008,
      "learning_rate": 3.3758639021796915e-05,
      "loss": 2.682,
      "step": 6300
    },
    {
      "epoch": 3.277009728622632,
      "grad_norm": 1.6347112655639648,
      "learning_rate": 3.349282296650718e-05,
      "loss": 2.6161,
      "step": 6400
    },
    {
      "epoch": 3.32821300563236,
      "grad_norm": 1.774921178817749,
      "learning_rate": 3.322700691121744e-05,
      "loss": 2.6727,
      "step": 6500
    },
    {
      "epoch": 3.3794162826420893,
      "grad_norm": 1.7162151336669922,
      "learning_rate": 3.29611908559277e-05,
      "loss": 2.692,
      "step": 6600
    },
    {
      "epoch": 3.430619559651818,
      "grad_norm": 1.7720268964767456,
      "learning_rate": 3.2695374800637965e-05,
      "loss": 2.6795,
      "step": 6700
    },
    {
      "epoch": 3.4818228366615465,
      "grad_norm": 1.7839425802230835,
      "learning_rate": 3.242955874534822e-05,
      "loss": 2.6487,
      "step": 6800
    },
    {
      "epoch": 3.533026113671275,
      "grad_norm": 1.7593103647232056,
      "learning_rate": 3.216374269005848e-05,
      "loss": 2.6821,
      "step": 6900
    },
    {
      "epoch": 3.5842293906810037,
      "grad_norm": 1.838611125946045,
      "learning_rate": 3.189792663476874e-05,
      "loss": 2.6547,
      "step": 7000
    },
    {
      "epoch": 3.6354326676907323,
      "grad_norm": 1.5734059810638428,
      "learning_rate": 3.1632110579479e-05,
      "loss": 2.6343,
      "step": 7100
    },
    {
      "epoch": 3.686635944700461,
      "grad_norm": 1.7309696674346924,
      "learning_rate": 3.136629452418926e-05,
      "loss": 2.6379,
      "step": 7200
    },
    {
      "epoch": 3.7378392217101895,
      "grad_norm": 1.837202787399292,
      "learning_rate": 3.110047846889952e-05,
      "loss": 2.6094,
      "step": 7300
    },
    {
      "epoch": 3.789042498719918,
      "grad_norm": 1.6872186660766602,
      "learning_rate": 3.083466241360978e-05,
      "loss": 2.5951,
      "step": 7400
    },
    {
      "epoch": 3.8402457757296466,
      "grad_norm": 1.7772648334503174,
      "learning_rate": 3.056884635832004e-05,
      "loss": 2.5935,
      "step": 7500
    },
    {
      "epoch": 3.8914490527393752,
      "grad_norm": 1.7157689332962036,
      "learning_rate": 3.0303030303030306e-05,
      "loss": 2.6076,
      "step": 7600
    },
    {
      "epoch": 3.942652329749104,
      "grad_norm": 1.7484747171401978,
      "learning_rate": 3.0037214247740565e-05,
      "loss": 2.6294,
      "step": 7700
    },
    {
      "epoch": 3.9938556067588324,
      "grad_norm": 1.6567468643188477,
      "learning_rate": 2.9771398192450828e-05,
      "loss": 2.603,
      "step": 7800
    },
    {
      "epoch": 4.0450588837685615,
      "grad_norm": 1.6986713409423828,
      "learning_rate": 2.9505582137161087e-05,
      "loss": 2.5767,
      "step": 7900
    },
    {
      "epoch": 4.09626216077829,
      "grad_norm": 1.9400880336761475,
      "learning_rate": 2.9239766081871346e-05,
      "loss": 2.5063,
      "step": 8000
    },
    {
      "epoch": 4.09626216077829,
      "eval_accuracy": 0.4372182483806064,
      "eval_loss": 3.1963303089141846,
      "eval_runtime": 95.395,
      "eval_samples_per_second": 351.633,
      "eval_steps_per_second": 21.982,
      "step": 8000
    },
    {
      "epoch": 4.147465437788019,
      "grad_norm": 1.7278245687484741,
      "learning_rate": 2.8976608187134503e-05,
      "loss": 2.5088,
      "step": 8100
    },
    {
      "epoch": 4.198668714797747,
      "grad_norm": 1.778096318244934,
      "learning_rate": 2.8710792131844765e-05,
      "loss": 2.5284,
      "step": 8200
    },
    {
      "epoch": 4.249871991807476,
      "grad_norm": 1.990983009338379,
      "learning_rate": 2.8444976076555025e-05,
      "loss": 2.5181,
      "step": 8300
    },
    {
      "epoch": 4.301075268817204,
      "grad_norm": 1.577682375907898,
      "learning_rate": 2.8179160021265287e-05,
      "loss": 2.4916,
      "step": 8400
    },
    {
      "epoch": 4.352278545826933,
      "grad_norm": 1.6727075576782227,
      "learning_rate": 2.7913343965975547e-05,
      "loss": 2.5346,
      "step": 8500
    },
    {
      "epoch": 4.403481822836661,
      "grad_norm": 1.6889344453811646,
      "learning_rate": 2.7647527910685806e-05,
      "loss": 2.5164,
      "step": 8600
    },
    {
      "epoch": 4.45468509984639,
      "grad_norm": 1.763092041015625,
      "learning_rate": 2.738171185539607e-05,
      "loss": 2.5073,
      "step": 8700
    },
    {
      "epoch": 4.505888376856118,
      "grad_norm": 1.6232457160949707,
      "learning_rate": 2.7115895800106328e-05,
      "loss": 2.515,
      "step": 8800
    },
    {
      "epoch": 4.5570916538658475,
      "grad_norm": 1.7539464235305786,
      "learning_rate": 2.6850079744816587e-05,
      "loss": 2.538,
      "step": 8900
    },
    {
      "epoch": 4.6082949308755765,
      "grad_norm": 1.729683518409729,
      "learning_rate": 2.658426368952685e-05,
      "loss": 2.5301,
      "step": 9000
    },
    {
      "epoch": 4.659498207885305,
      "grad_norm": 1.753625750541687,
      "learning_rate": 2.631844763423711e-05,
      "loss": 2.5267,
      "step": 9100
    },
    {
      "epoch": 4.710701484895033,
      "grad_norm": 1.7453761100769043,
      "learning_rate": 2.605263157894737e-05,
      "loss": 2.5207,
      "step": 9200
    },
    {
      "epoch": 4.761904761904762,
      "grad_norm": 1.9901796579360962,
      "learning_rate": 2.578681552365763e-05,
      "loss": 2.4987,
      "step": 9300
    },
    {
      "epoch": 4.813108038914491,
      "grad_norm": 1.6312249898910522,
      "learning_rate": 2.552099946836789e-05,
      "loss": 2.4943,
      "step": 9400
    },
    {
      "epoch": 4.864311315924219,
      "grad_norm": 1.7069945335388184,
      "learning_rate": 2.525518341307815e-05,
      "loss": 2.4666,
      "step": 9500
    },
    {
      "epoch": 4.915514592933948,
      "grad_norm": 1.7387930154800415,
      "learning_rate": 2.4989367357788412e-05,
      "loss": 2.4932,
      "step": 9600
    },
    {
      "epoch": 4.966717869943676,
      "grad_norm": 1.6212061643600464,
      "learning_rate": 2.472355130249867e-05,
      "loss": 2.52,
      "step": 9700
    },
    {
      "epoch": 5.017921146953405,
      "grad_norm": 1.7624274492263794,
      "learning_rate": 2.445773524720893e-05,
      "loss": 2.4938,
      "step": 9800
    },
    {
      "epoch": 5.0691244239631335,
      "grad_norm": 1.824257254600525,
      "learning_rate": 2.4191919191919194e-05,
      "loss": 2.4275,
      "step": 9900
    },
    {
      "epoch": 5.1203277009728625,
      "grad_norm": 1.616184115409851,
      "learning_rate": 2.3926103136629456e-05,
      "loss": 2.4139,
      "step": 10000
    },
    {
      "epoch": 5.1203277009728625,
      "eval_accuracy": 0.4442466080127022,
      "eval_loss": 3.1405584812164307,
      "eval_runtime": 95.5986,
      "eval_samples_per_second": 350.884,
      "eval_steps_per_second": 21.935,
      "step": 10000
    },
    {
      "epoch": 5.171530977982591,
      "grad_norm": 1.764162302017212,
      "learning_rate": 2.3660287081339712e-05,
      "loss": 2.4878,
      "step": 10100
    },
    {
      "epoch": 5.22273425499232,
      "grad_norm": 1.6120530366897583,
      "learning_rate": 2.3394471026049975e-05,
      "loss": 2.4391,
      "step": 10200
    },
    {
      "epoch": 5.273937532002048,
      "grad_norm": 1.6949564218521118,
      "learning_rate": 2.3128654970760234e-05,
      "loss": 2.4131,
      "step": 10300
    },
    {
      "epoch": 5.325140809011777,
      "grad_norm": 1.667853832244873,
      "learning_rate": 2.2862838915470493e-05,
      "loss": 2.4112,
      "step": 10400
    },
    {
      "epoch": 5.376344086021505,
      "grad_norm": 1.427404522895813,
      "learning_rate": 2.2597022860180756e-05,
      "loss": 2.4242,
      "step": 10500
    },
    {
      "epoch": 5.427547363031234,
      "grad_norm": 1.6098235845565796,
      "learning_rate": 2.2331206804891015e-05,
      "loss": 2.429,
      "step": 10600
    },
    {
      "epoch": 5.478750640040962,
      "grad_norm": 1.8480441570281982,
      "learning_rate": 2.2065390749601278e-05,
      "loss": 2.4439,
      "step": 10700
    },
    {
      "epoch": 5.529953917050691,
      "grad_norm": 1.8656944036483765,
      "learning_rate": 2.1799574694311537e-05,
      "loss": 2.4164,
      "step": 10800
    },
    {
      "epoch": 5.58115719406042,
      "grad_norm": 1.9880622625350952,
      "learning_rate": 2.1533758639021797e-05,
      "loss": 2.465,
      "step": 10900
    },
    {
      "epoch": 5.6323604710701485,
      "grad_norm": 1.8432066440582275,
      "learning_rate": 2.126794258373206e-05,
      "loss": 2.4216,
      "step": 11000
    },
    {
      "epoch": 5.683563748079877,
      "grad_norm": 1.7552499771118164,
      "learning_rate": 2.100212652844232e-05,
      "loss": 2.379,
      "step": 11100
    },
    {
      "epoch": 5.734767025089606,
      "grad_norm": 1.7043852806091309,
      "learning_rate": 2.0736310473152578e-05,
      "loss": 2.4308,
      "step": 11200
    },
    {
      "epoch": 5.785970302099335,
      "grad_norm": 1.8842682838439941,
      "learning_rate": 2.047049441786284e-05,
      "loss": 2.412,
      "step": 11300
    },
    {
      "epoch": 5.837173579109063,
      "grad_norm": 1.6765940189361572,
      "learning_rate": 2.02046783625731e-05,
      "loss": 2.4299,
      "step": 11400
    },
    {
      "epoch": 5.888376856118792,
      "grad_norm": 1.7585973739624023,
      "learning_rate": 1.993886230728336e-05,
      "loss": 2.4078,
      "step": 11500
    },
    {
      "epoch": 5.93958013312852,
      "grad_norm": 1.788201928138733,
      "learning_rate": 1.9673046251993622e-05,
      "loss": 2.4186,
      "step": 11600
    },
    {
      "epoch": 5.990783410138249,
      "grad_norm": 1.6128724813461304,
      "learning_rate": 1.940723019670388e-05,
      "loss": 2.3844,
      "step": 11700
    },
    {
      "epoch": 6.041986687147977,
      "grad_norm": 1.7053714990615845,
      "learning_rate": 1.9141414141414144e-05,
      "loss": 2.3299,
      "step": 11800
    },
    {
      "epoch": 6.093189964157706,
      "grad_norm": 2.06257963180542,
      "learning_rate": 1.8875598086124403e-05,
      "loss": 2.3666,
      "step": 11900
    },
    {
      "epoch": 6.1443932411674345,
      "grad_norm": 1.8419615030288696,
      "learning_rate": 1.8609782030834662e-05,
      "loss": 2.3246,
      "step": 12000
    },
    {
      "epoch": 6.1443932411674345,
      "eval_accuracy": 0.44839520298251256,
      "eval_loss": 3.115219831466675,
      "eval_runtime": 95.2063,
      "eval_samples_per_second": 352.329,
      "eval_steps_per_second": 22.026,
      "step": 12000
    },
    {
      "epoch": 6.1955965181771635,
      "grad_norm": 1.4872585535049438,
      "learning_rate": 1.834662413609782e-05,
      "loss": 2.3234,
      "step": 12100
    },
    {
      "epoch": 6.246799795186892,
      "grad_norm": 1.6765344142913818,
      "learning_rate": 1.808080808080808e-05,
      "loss": 2.353,
      "step": 12200
    },
    {
      "epoch": 6.298003072196621,
      "grad_norm": 1.611560583114624,
      "learning_rate": 1.7814992025518344e-05,
      "loss": 2.3758,
      "step": 12300
    },
    {
      "epoch": 6.349206349206349,
      "grad_norm": 1.7853387594223022,
      "learning_rate": 1.75491759702286e-05,
      "loss": 2.3232,
      "step": 12400
    },
    {
      "epoch": 6.400409626216078,
      "grad_norm": 1.614193320274353,
      "learning_rate": 1.7283359914938863e-05,
      "loss": 2.3759,
      "step": 12500
    },
    {
      "epoch": 6.451612903225806,
      "grad_norm": 1.8242110013961792,
      "learning_rate": 1.7017543859649125e-05,
      "loss": 2.4106,
      "step": 12600
    },
    {
      "epoch": 6.502816180235535,
      "grad_norm": 1.674735188484192,
      "learning_rate": 1.6751727804359385e-05,
      "loss": 2.3834,
      "step": 12700
    },
    {
      "epoch": 6.554019457245264,
      "grad_norm": 1.8332616090774536,
      "learning_rate": 1.6485911749069644e-05,
      "loss": 2.3631,
      "step": 12800
    },
    {
      "epoch": 6.605222734254992,
      "grad_norm": 1.8917430639266968,
      "learning_rate": 1.6220095693779903e-05,
      "loss": 2.3549,
      "step": 12900
    },
    {
      "epoch": 6.65642601126472,
      "grad_norm": 1.9428924322128296,
      "learning_rate": 1.5954279638490166e-05,
      "loss": 2.3205,
      "step": 13000
    },
    {
      "epoch": 6.7076292882744495,
      "grad_norm": 1.639167308807373,
      "learning_rate": 1.568846358320043e-05,
      "loss": 2.3646,
      "step": 13100
    },
    {
      "epoch": 6.7588325652841785,
      "grad_norm": 1.6185379028320312,
      "learning_rate": 1.5422647527910685e-05,
      "loss": 2.3757,
      "step": 13200
    },
    {
      "epoch": 6.810035842293907,
      "grad_norm": 1.6808964014053345,
      "learning_rate": 1.5156831472620947e-05,
      "loss": 2.3542,
      "step": 13300
    },
    {
      "epoch": 6.861239119303636,
      "grad_norm": 1.4692785739898682,
      "learning_rate": 1.4891015417331208e-05,
      "loss": 2.3476,
      "step": 13400
    },
    {
      "epoch": 6.912442396313364,
      "grad_norm": 1.6999351978302002,
      "learning_rate": 1.4625199362041468e-05,
      "loss": 2.3441,
      "step": 13500
    },
    {
      "epoch": 6.963645673323093,
      "grad_norm": 1.7041444778442383,
      "learning_rate": 1.4359383306751729e-05,
      "loss": 2.3725,
      "step": 13600
    },
    {
      "epoch": 7.014848950332821,
      "grad_norm": 1.773606538772583,
      "learning_rate": 1.409356725146199e-05,
      "loss": 2.3379,
      "step": 13700
    },
    {
      "epoch": 7.06605222734255,
      "grad_norm": 1.7857049703598022,
      "learning_rate": 1.382775119617225e-05,
      "loss": 2.3471,
      "step": 13800
    },
    {
      "epoch": 7.117255504352278,
      "grad_norm": 1.7172592878341675,
      "learning_rate": 1.356193514088251e-05,
      "loss": 2.2639,
      "step": 13900
    },
    {
      "epoch": 7.168458781362007,
      "grad_norm": 2.0828044414520264,
      "learning_rate": 1.329611908559277e-05,
      "loss": 2.3111,
      "step": 14000
    },
    {
      "epoch": 7.168458781362007,
      "eval_accuracy": 0.44894773835621105,
      "eval_loss": 3.0879344940185547,
      "eval_runtime": 95.1844,
      "eval_samples_per_second": 352.411,
      "eval_steps_per_second": 22.031,
      "step": 14000
    },
    {
      "epoch": 7.2196620583717355,
      "grad_norm": 1.7171036005020142,
      "learning_rate": 1.3032961190855927e-05,
      "loss": 2.3203,
      "step": 14100
    },
    {
      "epoch": 7.2708653353814645,
      "grad_norm": 1.8851195573806763,
      "learning_rate": 1.2767145135566188e-05,
      "loss": 2.3111,
      "step": 14200
    },
    {
      "epoch": 7.322068612391193,
      "grad_norm": 1.8158164024353027,
      "learning_rate": 1.250132908027645e-05,
      "loss": 2.3227,
      "step": 14300
    },
    {
      "epoch": 7.373271889400922,
      "grad_norm": 1.833143711090088,
      "learning_rate": 1.223551302498671e-05,
      "loss": 2.2796,
      "step": 14400
    },
    {
      "epoch": 7.42447516641065,
      "grad_norm": 1.5996191501617432,
      "learning_rate": 1.196969696969697e-05,
      "loss": 2.2753,
      "step": 14500
    },
    {
      "epoch": 7.475678443420379,
      "grad_norm": 1.7543715238571167,
      "learning_rate": 1.170388091440723e-05,
      "loss": 2.3191,
      "step": 14600
    },
    {
      "epoch": 7.526881720430108,
      "grad_norm": 1.8224458694458008,
      "learning_rate": 1.1438064859117491e-05,
      "loss": 2.3362,
      "step": 14700
    },
    {
      "epoch": 7.578084997439836,
      "grad_norm": 1.825265884399414,
      "learning_rate": 1.1172248803827752e-05,
      "loss": 2.2788,
      "step": 14800
    },
    {
      "epoch": 7.629288274449565,
      "grad_norm": 1.6064096689224243,
      "learning_rate": 1.0906432748538012e-05,
      "loss": 2.3287,
      "step": 14900
    },
    {
      "epoch": 7.680491551459293,
      "grad_norm": 1.9973961114883423,
      "learning_rate": 1.0640616693248273e-05,
      "loss": 2.3054,
      "step": 15000
    },
    {
      "epoch": 7.731694828469022,
      "grad_norm": 1.9129321575164795,
      "learning_rate": 1.0374800637958534e-05,
      "loss": 2.2897,
      "step": 15100
    },
    {
      "epoch": 7.7828981054787505,
      "grad_norm": 1.7142568826675415,
      "learning_rate": 1.0108984582668795e-05,
      "loss": 2.2763,
      "step": 15200
    },
    {
      "epoch": 7.8341013824884795,
      "grad_norm": 1.8948873281478882,
      "learning_rate": 9.843168527379054e-06,
      "loss": 2.2831,
      "step": 15300
    },
    {
      "epoch": 7.885304659498208,
      "grad_norm": 1.671781301498413,
      "learning_rate": 9.577352472089313e-06,
      "loss": 2.301,
      "step": 15400
    },
    {
      "epoch": 7.936507936507937,
      "grad_norm": 1.8178682327270508,
      "learning_rate": 9.311536416799576e-06,
      "loss": 2.3297,
      "step": 15500
    },
    {
      "epoch": 7.987711213517665,
      "grad_norm": 1.7934144735336304,
      "learning_rate": 9.045720361509835e-06,
      "loss": 2.2703,
      "step": 15600
    },
    {
      "epoch": 8.038914490527393,
      "grad_norm": 1.7798469066619873,
      "learning_rate": 8.779904306220096e-06,
      "loss": 2.3071,
      "step": 15700
    },
    {
      "epoch": 8.090117767537123,
      "grad_norm": 2.005934238433838,
      "learning_rate": 8.514088250930355e-06,
      "loss": 2.2579,
      "step": 15800
    },
    {
      "epoch": 8.141321044546851,
      "grad_norm": 1.7841377258300781,
      "learning_rate": 8.248272195640618e-06,
      "loss": 2.2942,
      "step": 15900
    },
    {
      "epoch": 8.19252432155658,
      "grad_norm": 1.814175009727478,
      "learning_rate": 7.982456140350877e-06,
      "loss": 2.2761,
      "step": 16000
    },
    {
      "epoch": 8.19252432155658,
      "eval_accuracy": 0.45419183636131977,
      "eval_loss": 3.0668067932128906,
      "eval_runtime": 95.2855,
      "eval_samples_per_second": 352.037,
      "eval_steps_per_second": 22.008,
      "step": 16000
    },
    {
      "epoch": 8.243727598566307,
      "grad_norm": 1.7196694612503052,
      "learning_rate": 7.719298245614036e-06,
      "loss": 2.2475,
      "step": 16100
    },
    {
      "epoch": 8.294930875576037,
      "grad_norm": 1.780168890953064,
      "learning_rate": 7.453482190324296e-06,
      "loss": 2.2623,
      "step": 16200
    },
    {
      "epoch": 8.346134152585766,
      "grad_norm": 1.7353187799453735,
      "learning_rate": 7.187666135034557e-06,
      "loss": 2.2436,
      "step": 16300
    },
    {
      "epoch": 8.397337429595494,
      "grad_norm": 1.8766587972640991,
      "learning_rate": 6.921850079744817e-06,
      "loss": 2.2514,
      "step": 16400
    },
    {
      "epoch": 8.448540706605222,
      "grad_norm": 1.8156400918960571,
      "learning_rate": 6.656034024455077e-06,
      "loss": 2.2647,
      "step": 16500
    },
    {
      "epoch": 8.499743983614952,
      "grad_norm": 1.7604610919952393,
      "learning_rate": 6.390217969165338e-06,
      "loss": 2.2633,
      "step": 16600
    },
    {
      "epoch": 8.55094726062468,
      "grad_norm": 1.9719775915145874,
      "learning_rate": 6.124401913875599e-06,
      "loss": 2.2729,
      "step": 16700
    },
    {
      "epoch": 8.602150537634408,
      "grad_norm": 1.7817537784576416,
      "learning_rate": 5.858585858585859e-06,
      "loss": 2.2605,
      "step": 16800
    },
    {
      "epoch": 8.653353814644138,
      "grad_norm": 2.052873373031616,
      "learning_rate": 5.59276980329612e-06,
      "loss": 2.2612,
      "step": 16900
    },
    {
      "epoch": 8.704557091653866,
      "grad_norm": 1.7299405336380005,
      "learning_rate": 5.326953748006379e-06,
      "loss": 2.2544,
      "step": 17000
    },
    {
      "epoch": 8.755760368663594,
      "grad_norm": 1.68231999874115,
      "learning_rate": 5.06113769271664e-06,
      "loss": 2.2786,
      "step": 17100
    },
    {
      "epoch": 8.806963645673322,
      "grad_norm": 1.740615725517273,
      "learning_rate": 4.7953216374269005e-06,
      "loss": 2.2514,
      "step": 17200
    },
    {
      "epoch": 8.858166922683052,
      "grad_norm": 1.6444553136825562,
      "learning_rate": 4.5295055821371614e-06,
      "loss": 2.2666,
      "step": 17300
    },
    {
      "epoch": 8.90937019969278,
      "grad_norm": 1.6717356443405151,
      "learning_rate": 4.263689526847422e-06,
      "loss": 2.2459,
      "step": 17400
    },
    {
      "epoch": 8.960573476702509,
      "grad_norm": 1.9801234006881714,
      "learning_rate": 3.9978734715576826e-06,
      "loss": 2.2494,
      "step": 17500
    },
    {
      "epoch": 9.011776753712237,
      "grad_norm": 1.8575870990753174,
      "learning_rate": 3.732057416267943e-06,
      "loss": 2.2438,
      "step": 17600
    },
    {
      "epoch": 9.062980030721967,
      "grad_norm": 1.7382246255874634,
      "learning_rate": 3.4662413609782037e-06,
      "loss": 2.2627,
      "step": 17700
    },
    {
      "epoch": 9.114183307731695,
      "grad_norm": 1.739556074142456,
      "learning_rate": 3.2004253056884642e-06,
      "loss": 2.249,
      "step": 17800
    },
    {
      "epoch": 9.165386584741423,
      "grad_norm": 1.9629456996917725,
      "learning_rate": 2.9346092503987244e-06,
      "loss": 2.2321,
      "step": 17900
    },
    {
      "epoch": 9.216589861751151,
      "grad_norm": 1.9510672092437744,
      "learning_rate": 2.668793195108985e-06,
      "loss": 2.2231,
      "step": 18000
    },
    {
      "epoch": 9.216589861751151,
      "eval_accuracy": 0.4517146740233816,
      "eval_loss": 3.069502830505371,
      "eval_runtime": 95.8923,
      "eval_samples_per_second": 349.809,
      "eval_steps_per_second": 21.868,
      "step": 18000
    },
    {
      "epoch": 9.267793138760881,
      "grad_norm": 1.7088594436645508,
      "learning_rate": 2.4056353003721427e-06,
      "loss": 2.2545,
      "step": 18100
    },
    {
      "epoch": 9.31899641577061,
      "grad_norm": 1.7437766790390015,
      "learning_rate": 2.1398192450824032e-06,
      "loss": 2.232,
      "step": 18200
    },
    {
      "epoch": 9.370199692780337,
      "grad_norm": 1.936566948890686,
      "learning_rate": 1.8740031897926634e-06,
      "loss": 2.2572,
      "step": 18300
    },
    {
      "epoch": 9.421402969790067,
      "grad_norm": 1.9586554765701294,
      "learning_rate": 1.608187134502924e-06,
      "loss": 2.2504,
      "step": 18400
    },
    {
      "epoch": 9.472606246799796,
      "grad_norm": 1.8036390542984009,
      "learning_rate": 1.3423710792131845e-06,
      "loss": 2.2402,
      "step": 18500
    },
    {
      "epoch": 9.523809523809524,
      "grad_norm": 1.866513967514038,
      "learning_rate": 1.076555023923445e-06,
      "loss": 2.2471,
      "step": 18600
    },
    {
      "epoch": 9.575012800819252,
      "grad_norm": 1.9675647020339966,
      "learning_rate": 8.107389686337055e-07,
      "loss": 2.2286,
      "step": 18700
    },
    {
      "epoch": 9.626216077828982,
      "grad_norm": 1.8157950639724731,
      "learning_rate": 5.44922913343966e-07,
      "loss": 2.2222,
      "step": 18800
    },
    {
      "epoch": 9.67741935483871,
      "grad_norm": 1.8483563661575317,
      "learning_rate": 2.7910685805422647e-07,
      "loss": 2.2518,
      "step": 18900
    },
    {
      "epoch": 9.728622631848438,
      "grad_norm": 1.9589660167694092,
      "learning_rate": 1.3290802764486977e-08,
      "loss": 2.2022,
      "step": 19000
    },
    {
      "epoch": 9.728622631848438,
      "step": 19000,
      "total_flos": 7.9413964701696e+16,
      "train_loss": 2.775124670731394,
      "train_runtime": 3271.669,
      "train_samples_per_second": 92.919,
      "train_steps_per_second": 5.807
    }
  ],
  "logging_steps": 100,
  "max_steps": 19000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 99999999,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 7.9413964701696e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}