{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 100,
  "global_step": 2000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0125,
      "grad_norm": 0.14092598855495453,
      "learning_rate": 0.0004,
      "loss": 1.1529,
      "step": 25
    },
    {
      "epoch": 0.025,
      "grad_norm": 0.14281609654426575,
      "learning_rate": 0.0004998852503731983,
      "loss": 1.0472,
      "step": 50
    },
    {
      "epoch": 0.0375,
      "grad_norm": 0.24040694534778595,
      "learning_rate": 0.0004993848168027977,
      "loss": 0.8532,
      "step": 75
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.14735238254070282,
      "learning_rate": 0.0004984880506341147,
      "loss": 0.9761,
      "step": 100
    },
    {
      "epoch": 0.05,
      "eval_loss": 0.9448406100273132,
      "eval_runtime": 322.186,
      "eval_samples_per_second": 3.405,
      "eval_steps_per_second": 0.056,
      "step": 100
    },
    {
      "epoch": 0.0625,
      "grad_norm": 0.1359478235244751,
      "learning_rate": 0.0004971963770447935,
      "loss": 0.9568,
      "step": 125
    },
    {
      "epoch": 0.075,
      "grad_norm": 0.17497147619724274,
      "learning_rate": 0.0004955118488155782,
      "loss": 0.7111,
      "step": 150
    },
    {
      "epoch": 0.0875,
      "grad_norm": 0.14382271468639374,
      "learning_rate": 0.0004934371430679492,
      "loss": 0.9413,
      "step": 175
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.16079047322273254,
      "learning_rate": 0.0004909755570095319,
      "loss": 0.8981,
      "step": 200
    },
    {
      "epoch": 0.1,
      "eval_loss": 0.9723050594329834,
      "eval_runtime": 322.7576,
      "eval_samples_per_second": 3.399,
      "eval_steps_per_second": 0.056,
      "step": 200
    },
    {
      "epoch": 0.1125,
      "grad_norm": 0.14729101955890656,
      "learning_rate": 0.0004881310026940389,
      "loss": 0.6378,
      "step": 225
    },
    {
      "epoch": 0.125,
      "grad_norm": 0.15030288696289062,
      "learning_rate": 0.0004849080008040734,
      "loss": 0.9271,
      "step": 250
    },
    {
      "epoch": 0.1375,
      "grad_norm": 0.1613348424434662,
      "learning_rate": 0.00048131167346667446,
      "loss": 0.8457,
      "step": 275
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.15532569587230682,
      "learning_rate": 0.00047734773611302284,
      "loss": 0.603,
      "step": 300
    },
    {
      "epoch": 0.15,
      "eval_loss": 1.006589651107788,
      "eval_runtime": 323.2105,
      "eval_samples_per_second": 3.394,
      "eval_steps_per_second": 0.056,
      "step": 300
    },
    {
      "epoch": 0.1625,
      "grad_norm": 0.16015686094760895,
      "learning_rate": 0.0004730224883952422,
      "loss": 0.9036,
      "step": 325
    },
    {
      "epoch": 0.175,
      "grad_norm": 0.15767253935337067,
      "learning_rate": 0.0004683428041747334,
      "loss": 0.8283,
      "step": 350
    },
    {
      "epoch": 0.1875,
      "grad_norm": 0.17757417261600494,
      "learning_rate": 0.0004633161205979517,
      "loss": 0.5945,
      "step": 375
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.17248600721359253,
      "learning_rate": 0.0004579504262769877,
      "loss": 0.8655,
      "step": 400
    },
    {
      "epoch": 0.2,
      "eval_loss": 1.0158599615097046,
      "eval_runtime": 323.8034,
      "eval_samples_per_second": 3.388,
      "eval_steps_per_second": 0.056,
      "step": 400
    },
    {
      "epoch": 0.2125,
      "grad_norm": 0.17826460301876068,
      "learning_rate": 0.0004522542485937369,
      "loss": 0.8079,
      "step": 425
    },
    {
      "epoch": 0.225,
      "grad_norm": 0.19307631254196167,
      "learning_rate": 0.00044623664014783386,
      "loss": 0.5737,
      "step": 450
    },
    {
      "epoch": 0.2375,
      "grad_norm": 0.1877959966659546,
      "learning_rate": 0.00043990716436988924,
      "loss": 0.8605,
      "step": 475
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.15268854796886444,
      "learning_rate": 0.0004332758803228925,
      "loss": 0.7674,
      "step": 500
    },
    {
      "epoch": 0.25,
      "eval_loss": 1.0431231260299683,
      "eval_runtime": 323.1376,
      "eval_samples_per_second": 3.395,
      "eval_steps_per_second": 0.056,
      "step": 500
    },
    {
      "epoch": 0.2625,
      "grad_norm": 0.16773808002471924,
      "learning_rate": 0.00042635332671593575,
      "loss": 0.5884,
      "step": 525
    },
    {
      "epoch": 0.275,
      "grad_norm": 0.15766142308712006,
      "learning_rate": 0.00041915050515566445,
      "loss": 0.8178,
      "step": 550
    },
    {
      "epoch": 0.2875,
      "grad_norm": 0.16790153086185455,
      "learning_rate": 0.00041167886266207167,
      "loss": 0.7797,
      "step": 575
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.15149210393428802,
      "learning_rate": 0.0004039502734764241,
      "loss": 0.7334,
      "step": 600
    },
    {
      "epoch": 0.3,
      "eval_loss": 1.0655592679977417,
      "eval_runtime": 321.6139,
      "eval_samples_per_second": 3.411,
      "eval_steps_per_second": 0.056,
      "step": 600
    },
    {
      "epoch": 0.3125,
      "grad_norm": 0.15414434671401978,
      "learning_rate": 0.0003959770201902294,
      "loss": 0.744,
      "step": 625
    },
    {
      "epoch": 0.325,
      "grad_norm": 0.1529635190963745,
      "learning_rate": 0.0003877717742252371,
      "loss": 0.6345,
      "step": 650
    },
    {
      "epoch": 0.3375,
      "grad_norm": 0.16185611486434937,
      "learning_rate": 0.00037934757569549495,
      "loss": 0.7354,
      "step": 675
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.16656433045864105,
      "learning_rate": 0.00037071781268346345,
      "loss": 0.7455,
      "step": 700
    },
    {
      "epoch": 0.35,
      "eval_loss": 1.0835301876068115,
      "eval_runtime": 322.4593,
      "eval_samples_per_second": 3.402,
      "eval_steps_per_second": 0.056,
      "step": 700
    },
    {
      "epoch": 0.3625,
      "grad_norm": 0.15590643882751465,
      "learning_rate": 0.00036189619996312495,
      "loss": 0.5972,
      "step": 725
    },
    {
      "epoch": 0.375,
      "grad_norm": 0.1837926208972931,
      "learning_rate": 0.00035289675720390174,
      "loss": 0.7592,
      "step": 750
    },
    {
      "epoch": 0.3875,
      "grad_norm": 0.1620703488588333,
      "learning_rate": 0.00034373378669002105,
      "loss": 0.736,
      "step": 775
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.16613048315048218,
      "learning_rate": 0.00033442185059073706,
      "loss": 0.564,
      "step": 800
    },
    {
      "epoch": 0.4,
      "eval_loss": 1.096523642539978,
      "eval_runtime": 321.7553,
      "eval_samples_per_second": 3.409,
      "eval_steps_per_second": 0.056,
      "step": 800
    },
    {
      "epoch": 0.4125,
      "grad_norm": 0.1587529182434082,
      "learning_rate": 0.00032497574781753367,
      "loss": 0.7598,
      "step": 825
    },
    {
      "epoch": 0.425,
      "grad_norm": 0.16457463800907135,
      "learning_rate": 0.000315410490505086,
      "loss": 0.7292,
      "step": 850
    },
    {
      "epoch": 0.4375,
      "grad_norm": 0.16429653763771057,
      "learning_rate": 0.0003057412801533589,
      "loss": 0.5329,
      "step": 875
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.16424661874771118,
      "learning_rate": 0.0002959834834687587,
      "loss": 0.7785,
      "step": 900
    },
    {
      "epoch": 0.45,
      "eval_loss": 1.0958806276321411,
      "eval_runtime": 322.3489,
      "eval_samples_per_second": 3.403,
      "eval_steps_per_second": 0.056,
      "step": 900
    },
    {
      "epoch": 0.4625,
      "grad_norm": 0.16132202744483948,
      "learning_rate": 0.00028615260794273236,
      "loss": 0.7259,
      "step": 925
    },
    {
      "epoch": 0.475,
      "grad_norm": 0.16069433093070984,
      "learning_rate": 0.00027626427720662416,
      "loss": 0.4996,
      "step": 950
    },
    {
      "epoch": 0.4875,
      "grad_norm": 0.16033512353897095,
      "learning_rate": 0.00026633420620195917,
      "loss": 0.7768,
      "step": 975
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.16236689686775208,
      "learning_rate": 0.00025637817620561263,
      "loss": 0.7225,
      "step": 1000
    },
    {
      "epoch": 0.5,
      "eval_loss": 1.1207813024520874,
      "eval_runtime": 322.5881,
      "eval_samples_per_second": 3.401,
      "eval_steps_per_second": 0.056,
      "step": 1000
    },
    {
      "epoch": 0.5125,
      "grad_norm": 0.15189126133918762,
      "learning_rate": 0.0002464120097495559,
      "loss": 0.4932,
      "step": 1025
    },
    {
      "epoch": 0.525,
      "grad_norm": 0.1648908108472824,
      "learning_rate": 0.00023645154547503855,
      "loss": 0.7902,
      "step": 1050
    },
    {
      "epoch": 0.5375,
      "grad_norm": 0.16001655161380768,
      "learning_rate": 0.00022651261296116894,
      "loss": 0.7003,
      "step": 1075
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.1633903682231903,
      "learning_rate": 0.00021661100756789666,
      "loss": 0.4709,
      "step": 1100
    },
    {
      "epoch": 0.55,
      "eval_loss": 1.1372770071029663,
      "eval_runtime": 323.5516,
      "eval_samples_per_second": 3.39,
      "eval_steps_per_second": 0.056,
      "step": 1100
    },
    {
      "epoch": 0.5625,
      "grad_norm": 0.1652471274137497,
      "learning_rate": 0.00020676246533337764,
      "loss": 0.8076,
      "step": 1125
    },
    {
      "epoch": 0.575,
      "grad_norm": 0.17891553044319153,
      "learning_rate": 0.00019698263796561526,
      "loss": 0.7156,
      "step": 1150
    },
    {
      "epoch": 0.5875,
      "grad_norm": 0.16443009674549103,
      "learning_rate": 0.00018728706796812333,
      "loss": 0.6316,
      "step": 1175
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.16446325182914734,
      "learning_rate": 0.00017769116393914037,
      "loss": 0.6956,
      "step": 1200
    },
    {
      "epoch": 0.6,
      "eval_loss": 1.1236783266067505,
      "eval_runtime": 320.7164,
      "eval_samples_per_second": 3.42,
      "eval_steps_per_second": 0.056,
      "step": 1200
    },
    {
      "epoch": 0.6125,
      "grad_norm": 0.1620441973209381,
      "learning_rate": 0.00016821017608365264,
      "loss": 0.6163,
      "step": 1225
    },
    {
      "epoch": 0.625,
      "grad_norm": 0.17003227770328522,
      "learning_rate": 0.00015885917197714112,
      "loss": 0.6232,
      "step": 1250
    },
    {
      "epoch": 0.6375,
      "grad_norm": 0.17415954172611237,
      "learning_rate": 0.00014965301261957238,
      "loss": 0.6991,
      "step": 1275
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.1617245227098465,
      "learning_rate": 0.00014060632881768558,
      "loss": 0.599,
      "step": 1300
    },
    {
      "epoch": 0.65,
      "eval_loss": 1.1370735168457031,
      "eval_runtime": 320.7099,
      "eval_samples_per_second": 3.421,
      "eval_steps_per_second": 0.056,
      "step": 1300
    },
    {
      "epoch": 0.6625,
      "grad_norm": 0.1753346472978592,
      "learning_rate": 0.00013173349793311424,
      "loss": 0.6611,
      "step": 1325
    },
    {
      "epoch": 0.675,
      "grad_norm": 0.17510834336280823,
      "learning_rate": 0.0001230486210332916,
      "loss": 0.6814,
      "step": 1350
    },
    {
      "epoch": 0.6875,
      "grad_norm": 0.17805688083171844,
      "learning_rate": 0.00011456550048145536,
      "loss": 0.5757,
      "step": 1375
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.17829716205596924,
      "learning_rate": 0.00010629761800136473,
      "loss": 0.6646,
      "step": 1400
    },
    {
      "epoch": 0.7,
      "eval_loss": 1.1476235389709473,
      "eval_runtime": 321.7747,
      "eval_samples_per_second": 3.409,
      "eval_steps_per_second": 0.056,
      "step": 1400
    },
    {
      "epoch": 0.7125,
      "grad_norm": 0.18059992790222168,
      "learning_rate": 9.82581132515907e-05,
      "loss": 0.6797,
      "step": 1425
    },
    {
      "epoch": 0.725,
      "grad_norm": 0.15256117284297943,
      "learning_rate": 9.045976294343145e-05,
      "loss": 0.5596,
      "step": 1450
    },
    {
      "epoch": 0.7375,
      "grad_norm": 0.17262427508831024,
      "learning_rate": 8.291496053563699e-05,
      "loss": 0.6905,
      "step": 1475
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.17382751405239105,
      "learning_rate": 7.563569653821565e-05,
      "loss": 0.6772,
      "step": 1500
    },
    {
      "epoch": 0.75,
      "eval_loss": 1.1411069631576538,
      "eval_runtime": 323.0865,
      "eval_samples_per_second": 3.395,
      "eval_steps_per_second": 0.056,
      "step": 1500
    },
    {
      "epoch": 0.7625,
      "grad_norm": 0.15558482706546783,
      "learning_rate": 6.863353945662288e-05,
      "loss": 0.5172,
      "step": 1525
    },
    {
      "epoch": 0.775,
      "grad_norm": 0.1950470507144928,
      "learning_rate": 6.191961740661687e-05,
      "loss": 0.7039,
      "step": 1550
    },
    {
      "epoch": 0.7875,
      "grad_norm": 0.17224222421646118,
      "learning_rate": 5.550460042899982e-05,
      "loss": 0.6914,
      "step": 1575
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.1652670055627823,
      "learning_rate": 4.9398683532350855e-05,
      "loss": 0.488,
      "step": 1600
    },
    {
      "epoch": 0.8,
      "eval_loss": 1.1589314937591553,
      "eval_runtime": 321.6657,
      "eval_samples_per_second": 3.41,
      "eval_steps_per_second": 0.056,
      "step": 1600
    },
    {
      "epoch": 0.8125,
      "grad_norm": 0.1697286069393158,
      "learning_rate": 4.3611570490698945e-05,
      "loss": 0.7453,
      "step": 1625
    },
    {
      "epoch": 0.825,
      "grad_norm": 0.16859295964241028,
      "learning_rate": 3.815245842188697e-05,
      "loss": 0.6625,
      "step": 1650
    },
    {
      "epoch": 0.8375,
      "grad_norm": 0.16462524235248566,
      "learning_rate": 3.30300231711339e-05,
      "loss": 0.4719,
      "step": 1675
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.17534367740154266,
      "learning_rate": 2.8252405523025106e-05,
      "loss": 0.7468,
      "step": 1700
    },
    {
      "epoch": 0.85,
      "eval_loss": 1.161086916923523,
      "eval_runtime": 322.1848,
      "eval_samples_per_second": 3.405,
      "eval_steps_per_second": 0.056,
      "step": 1700
    },
    {
      "epoch": 0.8625,
      "grad_norm": 0.16568109393119812,
      "learning_rate": 2.3827198263843162e-05,
      "loss": 0.709,
      "step": 1725
    },
    {
      "epoch": 0.875,
      "grad_norm": 0.17694273591041565,
      "learning_rate": 1.9761434114799497e-05,
      "loss": 0.5756,
      "step": 1750
    },
    {
      "epoch": 0.8875,
      "grad_norm": 0.16244478523731232,
      "learning_rate": 1.606157455534535e-05,
      "loss": 0.6545,
      "step": 1775
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.1651735007762909,
      "learning_rate": 1.2733499554322708e-05,
      "loss": 0.6352,
      "step": 1800
    },
    {
      "epoch": 0.9,
      "eval_loss": 1.1563351154327393,
      "eval_runtime": 320.396,
      "eval_samples_per_second": 3.424,
      "eval_steps_per_second": 0.056,
      "step": 1800
    },
    {
      "epoch": 0.9125,
      "grad_norm": 0.17625375092029572,
      "learning_rate": 9.782498225276437e-06,
      "loss": 0.5515,
      "step": 1825
    },
    {
      "epoch": 0.925,
      "grad_norm": 0.17532864212989807,
      "learning_rate": 7.213260420777607e-06,
      "loss": 0.686,
      "step": 1850
    },
    {
      "epoch": 0.9375,
      "grad_norm": 0.15695761144161224,
      "learning_rate": 5.029869279117167e-06,
      "loss": 0.6296,
      "step": 1875
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.17587953805923462,
      "learning_rate": 3.235794735214709e-06,
      "loss": 0.569,
      "step": 1900
    },
    {
      "epoch": 0.95,
      "eval_loss": 1.1628855466842651,
      "eval_runtime": 324.1332,
      "eval_samples_per_second": 3.384,
      "eval_steps_per_second": 0.056,
      "step": 1900
    },
    {
      "epoch": 0.9625,
      "grad_norm": 0.17684130370616913,
      "learning_rate": 1.8338880060553287e-06,
      "loss": 0.6861,
      "step": 1925
    },
    {
      "epoch": 0.975,
      "grad_norm": 0.16606929898262024,
      "learning_rate": 8.263770594185149e-07,
      "loss": 0.6187,
      "step": 1950
    },
    {
      "epoch": 0.9875,
      "grad_norm": 0.1606944501399994,
      "learning_rate": 2.1486307310000787e-07,
      "loss": 0.5915,
      "step": 1975
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.16651476919651031,
      "learning_rate": 3.1789025450867925e-10,
      "loss": 0.6817,
      "step": 2000
    },
    {
      "epoch": 1.0,
      "eval_loss": 1.160001277923584,
      "eval_runtime": 322.869,
      "eval_samples_per_second": 3.398,
      "eval_steps_per_second": 0.056,
      "step": 2000
    }
  ],
  "logging_steps": 25,
  "max_steps": 2000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9223372036854775807,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2.0158630867238912e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}