{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9963235294117647,
  "eval_steps": 500,
  "global_step": 271,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01838235294117647,
      "grad_norm": 40.77878189086914,
      "learning_rate": 4.917142857142858e-06,
      "loss": 4.3405,
      "mean_token_accuracy": 0.6079744696617126,
      "num_tokens": 396575.0,
      "step": 5
    },
    {
      "epoch": 0.03676470588235294,
      "grad_norm": 8.723130226135254,
      "learning_rate": 1.106357142857143e-05,
      "loss": 1.5241,
      "mean_token_accuracy": 0.7754200756549835,
      "num_tokens": 784215.0,
      "step": 10
    },
    {
      "epoch": 0.05514705882352941,
      "grad_norm": 2.6150617599487305,
      "learning_rate": 1.7210000000000003e-05,
      "loss": 0.2611,
      "mean_token_accuracy": 0.9413737654685974,
      "num_tokens": 1175812.0,
      "step": 15
    },
    {
      "epoch": 0.07352941176470588,
      "grad_norm": 10.302824974060059,
      "learning_rate": 2.3356428571428574e-05,
      "loss": 0.0494,
      "mean_token_accuracy": 0.9854643404483795,
      "num_tokens": 1568846.0,
      "step": 20
    },
    {
      "epoch": 0.09191176470588236,
      "grad_norm": 0.31368523836135864,
      "learning_rate": 2.950285714285715e-05,
      "loss": 0.0329,
      "mean_token_accuracy": 0.9888089418411254,
      "num_tokens": 1953541.0,
      "step": 25
    },
    {
      "epoch": 0.11029411764705882,
      "grad_norm": 0.36221420764923096,
      "learning_rate": 3.5649285714285724e-05,
      "loss": 0.022,
      "mean_token_accuracy": 0.9924580335617066,
      "num_tokens": 2343960.0,
      "step": 30
    },
    {
      "epoch": 0.12867647058823528,
      "grad_norm": 0.24492664635181427,
      "learning_rate": 4.179571428571429e-05,
      "loss": 0.0179,
      "mean_token_accuracy": 0.9932528793811798,
      "num_tokens": 2734906.0,
      "step": 35
    },
    {
      "epoch": 0.14705882352941177,
      "grad_norm": 0.27344170212745667,
      "learning_rate": 4.3002325201486e-05,
      "loss": 0.0125,
      "mean_token_accuracy": 0.9953317403793335,
      "num_tokens": 3119789.0,
      "step": 40
    },
    {
      "epoch": 0.16544117647058823,
      "grad_norm": 0.25588464736938477,
      "learning_rate": 4.291031805163528e-05,
      "loss": 0.012,
      "mean_token_accuracy": 0.9955625712871552,
      "num_tokens": 3511752.0,
      "step": 45
    },
    {
      "epoch": 0.18382352941176472,
      "grad_norm": 0.28750279545783997,
      "learning_rate": 4.274796508682312e-05,
      "loss": 0.0084,
      "mean_token_accuracy": 0.9966519951820374,
      "num_tokens": 3897740.0,
      "step": 50
    },
    {
      "epoch": 0.20220588235294118,
      "grad_norm": 0.25286757946014404,
      "learning_rate": 4.251597923279562e-05,
      "loss": 0.0114,
      "mean_token_accuracy": 0.9952130198478699,
      "num_tokens": 4285139.0,
      "step": 55
    },
    {
      "epoch": 0.22058823529411764,
      "grad_norm": 0.14998362958431244,
      "learning_rate": 4.221537918784545e-05,
      "loss": 0.0135,
      "mean_token_accuracy": 0.9948176205158233,
      "num_tokens": 4682575.0,
      "step": 60
    },
    {
      "epoch": 0.23897058823529413,
      "grad_norm": 0.12603914737701416,
      "learning_rate": 4.184748494949514e-05,
      "loss": 0.0096,
      "mean_token_accuracy": 0.9962626159191131,
      "num_tokens": 5066810.0,
      "step": 65
    },
    {
      "epoch": 0.25735294117647056,
      "grad_norm": 0.19672255218029022,
      "learning_rate": 4.141391201811243e-05,
      "loss": 0.0089,
      "mean_token_accuracy": 0.9965241849422455,
      "num_tokens": 5463729.0,
      "step": 70
    },
    {
      "epoch": 0.2757352941176471,
      "grad_norm": 0.12892723083496094,
      "learning_rate": 4.0916564302911154e-05,
      "loss": 0.008,
      "mean_token_accuracy": 0.9969832301139832,
      "num_tokens": 5850128.0,
      "step": 75
    },
    {
      "epoch": 0.29411764705882354,
      "grad_norm": 0.13913273811340332,
      "learning_rate": 4.035762576148849e-05,
      "loss": 0.0078,
      "mean_token_accuracy": 0.9973669826984406,
      "num_tokens": 6244002.0,
      "step": 80
    },
    {
      "epoch": 0.3125,
      "grad_norm": 0.09209585934877396,
      "learning_rate": 3.973955080961153e-05,
      "loss": 0.0067,
      "mean_token_accuracy": 0.9973081052303314,
      "num_tokens": 6620382.0,
      "step": 85
    },
    {
      "epoch": 0.33088235294117646,
      "grad_norm": 0.1415279358625412,
      "learning_rate": 3.9065053543365454e-05,
      "loss": 0.0065,
      "mean_token_accuracy": 0.9977279663085937,
      "num_tokens": 7000568.0,
      "step": 90
    },
    {
      "epoch": 0.3492647058823529,
      "grad_norm": 0.09117847681045532,
      "learning_rate": 3.833709582099125e-05,
      "loss": 0.006,
      "mean_token_accuracy": 0.9978210866451264,
      "num_tokens": 7386388.0,
      "step": 95
    },
    {
      "epoch": 0.36764705882352944,
      "grad_norm": 0.08553037792444229,
      "learning_rate": 3.7558874256748195e-05,
      "loss": 0.0052,
      "mean_token_accuracy": 0.9980245590209961,
      "num_tokens": 7786489.0,
      "step": 100
    },
    {
      "epoch": 0.3860294117647059,
      "grad_norm": 0.08987542241811752,
      "learning_rate": 3.673380618391356e-05,
      "loss": 0.0065,
      "mean_token_accuracy": 0.9980957508087158,
      "num_tokens": 8179170.0,
      "step": 105
    },
    {
      "epoch": 0.40441176470588236,
      "grad_norm": 0.12981808185577393,
      "learning_rate": 3.5865514648559164e-05,
      "loss": 0.0064,
      "mean_token_accuracy": 0.9980520904064178,
      "num_tokens": 8576123.0,
      "step": 110
    },
    {
      "epoch": 0.4227941176470588,
      "grad_norm": 0.049204424023628235,
      "learning_rate": 3.495781250000001e-05,
      "loss": 0.0067,
      "mean_token_accuracy": 0.9976665556430817,
      "num_tokens": 8977664.0,
      "step": 115
    },
    {
      "epoch": 0.4411764705882353,
      "grad_norm": 0.09803196042776108,
      "learning_rate": 3.401468564777736e-05,
      "loss": 0.0052,
      "mean_token_accuracy": 0.9982431530952454,
      "num_tokens": 9371923.0,
      "step": 120
    },
    {
      "epoch": 0.45955882352941174,
      "grad_norm": 0.064445860683918,
      "learning_rate": 3.304027555869822e-05,
      "loss": 0.0045,
      "mean_token_accuracy": 0.9983674705028533,
      "num_tokens": 9747750.0,
      "step": 125
    },
    {
      "epoch": 0.47794117647058826,
      "grad_norm": 0.11150972545146942,
      "learning_rate": 3.203886107079038e-05,
      "loss": 0.0049,
      "mean_token_accuracy": 0.9983975529670716,
      "num_tokens": 10138347.0,
      "step": 130
    },
    {
      "epoch": 0.4963235294117647,
      "grad_norm": 0.06595049798488617,
      "learning_rate": 3.101483960403179e-05,
      "loss": 0.0038,
      "mean_token_accuracy": 0.9987940490245819,
      "num_tokens": 10519930.0,
      "step": 135
    },
    {
      "epoch": 0.5147058823529411,
      "grad_norm": 0.06900100409984589,
      "learning_rate": 2.997270785036188e-05,
      "loss": 0.0045,
      "mean_token_accuracy": 0.9984328985214234,
      "num_tokens": 10923657.0,
      "step": 140
    },
    {
      "epoch": 0.5330882352941176,
      "grad_norm": 0.08975087106227875,
      "learning_rate": 2.891704202776889e-05,
      "loss": 0.003,
      "mean_token_accuracy": 0.9988918602466583,
      "num_tokens": 11308146.0,
      "step": 145
    },
    {
      "epoch": 0.5514705882352942,
      "grad_norm": 0.06893062591552734,
      "learning_rate": 2.7852477785161814e-05,
      "loss": 0.0045,
      "mean_token_accuracy": 0.9988150656223297,
      "num_tokens": 11698084.0,
      "step": 150
    },
    {
      "epoch": 0.5698529411764706,
      "grad_norm": 0.1488070785999298,
      "learning_rate": 2.6783689846268743e-05,
      "loss": 0.0037,
      "mean_token_accuracy": 0.9985724210739135,
      "num_tokens": 12080500.0,
      "step": 155
    },
    {
      "epoch": 0.5882352941176471,
      "grad_norm": 0.11203882098197937,
      "learning_rate": 2.5715371481949467e-05,
      "loss": 0.006,
      "mean_token_accuracy": 0.9976416766643524,
      "num_tokens": 12475518.0,
      "step": 160
    },
    {
      "epoch": 0.6066176470588235,
      "grad_norm": 0.08506222069263458,
      "learning_rate": 2.4652213901063884e-05,
      "loss": 0.0039,
      "mean_token_accuracy": 0.9987370252609253,
      "num_tokens": 12871598.0,
      "step": 165
    },
    {
      "epoch": 0.625,
      "grad_norm": 0.06337340176105499,
      "learning_rate": 2.3598885650394917e-05,
      "loss": 0.0048,
      "mean_token_accuracy": 0.9986328899860382,
      "num_tokens": 13272176.0,
      "step": 170
    },
    {
      "epoch": 0.6433823529411765,
      "grad_norm": 0.08344393223524094,
      "learning_rate": 2.256001211408547e-05,
      "loss": 0.0034,
      "mean_token_accuracy": 0.9988697469234467,
      "num_tokens": 13673743.0,
      "step": 175
    },
    {
      "epoch": 0.6617647058823529,
      "grad_norm": 0.08593422174453735,
      "learning_rate": 2.1540155202611485e-05,
      "loss": 0.0041,
      "mean_token_accuracy": 0.9986071050167084,
      "num_tokens": 14066523.0,
      "step": 180
    },
    {
      "epoch": 0.6801470588235294,
      "grad_norm": 0.10229282081127167,
      "learning_rate": 2.054379332048112e-05,
      "loss": 0.0037,
      "mean_token_accuracy": 0.9987092971801758,
      "num_tokens": 14462689.0,
      "step": 185
    },
    {
      "epoch": 0.6985294117647058,
      "grad_norm": 0.11607314646244049,
      "learning_rate": 1.957530170062605e-05,
      "loss": 0.0042,
      "mean_token_accuracy": 0.9986791372299194,
      "num_tokens": 14847962.0,
      "step": 190
    },
    {
      "epoch": 0.7169117647058824,
      "grad_norm": 0.02692285180091858,
      "learning_rate": 1.863893319184073e-05,
      "loss": 0.0036,
      "mean_token_accuracy": 0.9986750185489655,
      "num_tokens": 15237420.0,
      "step": 195
    },
    {
      "epoch": 0.7352941176470589,
      "grad_norm": 0.053456008434295654,
      "learning_rate": 1.7738799583635873e-05,
      "loss": 0.0034,
      "mean_token_accuracy": 0.9988762140274048,
      "num_tokens": 15626132.0,
      "step": 200
    },
    {
      "epoch": 0.7536764705882353,
      "grad_norm": 0.05752914026379585,
      "learning_rate": 1.6878853550512764e-05,
      "loss": 0.0041,
      "mean_token_accuracy": 0.9984098911285401,
      "num_tokens": 16018063.0,
      "step": 205
    },
    {
      "epoch": 0.7720588235294118,
      "grad_norm": 0.08336920291185379,
      "learning_rate": 1.6062871294944667e-05,
      "loss": 0.0044,
      "mean_token_accuracy": 0.9985141575336456,
      "num_tokens": 16409175.0,
      "step": 210
    },
    {
      "epoch": 0.7904411764705882,
      "grad_norm": 0.11755795776844025,
      "learning_rate": 1.5294435965283893e-05,
      "loss": 0.0047,
      "mean_token_accuracy": 0.9981832683086396,
      "num_tokens": 16803906.0,
      "step": 215
    },
    {
      "epoch": 0.8088235294117647,
      "grad_norm": 0.08356364071369171,
      "learning_rate": 1.4576921921409594e-05,
      "loss": 0.0047,
      "mean_token_accuracy": 0.9984434008598327,
      "num_tokens": 17187152.0,
      "step": 220
    },
    {
      "epoch": 0.8272058823529411,
      "grad_norm": 0.09949304908514023,
      "learning_rate": 1.3913479917209507e-05,
      "loss": 0.003,
      "mean_token_accuracy": 0.9990458905696868,
      "num_tokens": 17574580.0,
      "step": 225
    },
    {
      "epoch": 0.8455882352941176,
      "grad_norm": 0.0650205984711647,
      "learning_rate": 1.3307023264962129e-05,
      "loss": 0.0033,
      "mean_token_accuracy": 0.9989714860916138,
      "num_tokens": 17961330.0,
      "step": 230
    },
    {
      "epoch": 0.8639705882352942,
      "grad_norm": 0.0479765459895134,
      "learning_rate": 1.2760215042374612e-05,
      "loss": 0.0032,
      "mean_token_accuracy": 0.9988696932792663,
      "num_tokens": 18349709.0,
      "step": 235
    },
    {
      "epoch": 0.8823529411764706,
      "grad_norm": 0.10548855364322662,
      "learning_rate": 1.2275456398452826e-05,
      "loss": 0.0029,
      "mean_token_accuracy": 0.9990268647670746,
      "num_tokens": 18733145.0,
      "step": 240
    },
    {
      "epoch": 0.9007352941176471,
      "grad_norm": 0.030466251075267792,
      "learning_rate": 1.1854876009555089e-05,
      "loss": 0.0029,
      "mean_token_accuracy": 0.9990714967250824,
      "num_tokens": 19123886.0,
      "step": 245
    },
    {
      "epoch": 0.9191176470588235,
      "grad_norm": 0.07559388130903244,
      "learning_rate": 1.1500320731930042e-05,
      "loss": 0.0036,
      "mean_token_accuracy": 0.9988040447235107,
      "num_tokens": 19506834.0,
      "step": 250
    },
    {
      "epoch": 0.9375,
      "grad_norm": 0.11076359450817108,
      "learning_rate": 1.1213347491785512e-05,
      "loss": 0.002,
      "mean_token_accuracy": 0.9992827951908112,
      "num_tokens": 19894291.0,
      "step": 255
    },
    {
      "epoch": 0.9558823529411765,
      "grad_norm": 0.04865700751543045,
      "learning_rate": 1.0995216448500645e-05,
      "loss": 0.002,
      "mean_token_accuracy": 0.9993002355098725,
      "num_tokens": 20272252.0,
      "step": 260
    },
    {
      "epoch": 0.9742647058823529,
      "grad_norm": 0.07477735728025436,
      "learning_rate": 1.0846885461003128e-05,
      "loss": 0.0026,
      "mean_token_accuracy": 0.9991890132427216,
      "num_tokens": 20650206.0,
      "step": 265
    },
    {
      "epoch": 0.9926470588235294,
      "grad_norm": 0.10289261490106583,
      "learning_rate": 1.0769005881610753e-05,
      "loss": 0.0033,
      "mean_token_accuracy": 0.9989068508148193,
      "num_tokens": 21044381.0,
      "step": 270
    }
  ],
  "logging_steps": 5,
  "max_steps": 272,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.7390843485875077e+18,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}