{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 918,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.032679738562091505,
      "grad_norm": 13.884983365234152,
      "learning_rate": 5e-06,
      "loss": 0.9132,
      "step": 10
    },
    {
      "epoch": 0.06535947712418301,
      "grad_norm": 4.396307073532539,
      "learning_rate": 5e-06,
      "loss": 0.793,
      "step": 20
    },
    {
      "epoch": 0.09803921568627451,
      "grad_norm": 1.5701715316993357,
      "learning_rate": 5e-06,
      "loss": 0.7554,
      "step": 30
    },
    {
      "epoch": 0.13071895424836602,
      "grad_norm": 1.0469284618172925,
      "learning_rate": 5e-06,
      "loss": 0.7308,
      "step": 40
    },
    {
      "epoch": 0.16339869281045752,
      "grad_norm": 1.025316533729498,
      "learning_rate": 5e-06,
      "loss": 0.7074,
      "step": 50
    },
    {
      "epoch": 0.19607843137254902,
      "grad_norm": 0.8007907881089358,
      "learning_rate": 5e-06,
      "loss": 0.6935,
      "step": 60
    },
    {
      "epoch": 0.22875816993464052,
      "grad_norm": 0.7336515686610362,
      "learning_rate": 5e-06,
      "loss": 0.6863,
      "step": 70
    },
    {
      "epoch": 0.26143790849673204,
      "grad_norm": 0.6654762709275407,
      "learning_rate": 5e-06,
      "loss": 0.6737,
      "step": 80
    },
    {
      "epoch": 0.29411764705882354,
      "grad_norm": 0.6752715837026988,
      "learning_rate": 5e-06,
      "loss": 0.678,
      "step": 90
    },
    {
      "epoch": 0.32679738562091504,
      "grad_norm": 0.6495908393250515,
      "learning_rate": 5e-06,
      "loss": 0.6625,
      "step": 100
    },
    {
      "epoch": 0.35947712418300654,
      "grad_norm": 0.5513534888680728,
      "learning_rate": 5e-06,
      "loss": 0.6728,
      "step": 110
    },
    {
      "epoch": 0.39215686274509803,
      "grad_norm": 0.5987482687514573,
      "learning_rate": 5e-06,
      "loss": 0.6605,
      "step": 120
    },
    {
      "epoch": 0.42483660130718953,
      "grad_norm": 0.8982453869254353,
      "learning_rate": 5e-06,
      "loss": 0.6576,
      "step": 130
    },
    {
      "epoch": 0.45751633986928103,
      "grad_norm": 0.5729452075955007,
      "learning_rate": 5e-06,
      "loss": 0.6581,
      "step": 140
    },
    {
      "epoch": 0.49019607843137253,
      "grad_norm": 0.6356665135318063,
      "learning_rate": 5e-06,
      "loss": 0.6488,
      "step": 150
    },
    {
      "epoch": 0.5228758169934641,
      "grad_norm": 0.8565590645061701,
      "learning_rate": 5e-06,
      "loss": 0.6611,
      "step": 160
    },
    {
      "epoch": 0.5555555555555556,
      "grad_norm": 0.7114568602994797,
      "learning_rate": 5e-06,
      "loss": 0.6448,
      "step": 170
    },
    {
      "epoch": 0.5882352941176471,
      "grad_norm": 0.6611333193380935,
      "learning_rate": 5e-06,
      "loss": 0.6478,
      "step": 180
    },
    {
      "epoch": 0.6209150326797386,
      "grad_norm": 0.6896881110018678,
      "learning_rate": 5e-06,
      "loss": 0.6495,
      "step": 190
    },
    {
      "epoch": 0.6535947712418301,
      "grad_norm": 0.6559022818956797,
      "learning_rate": 5e-06,
      "loss": 0.6501,
      "step": 200
    },
    {
      "epoch": 0.6862745098039216,
      "grad_norm": 0.557250562288118,
      "learning_rate": 5e-06,
      "loss": 0.6462,
      "step": 210
    },
    {
      "epoch": 0.7189542483660131,
      "grad_norm": 0.546010994925319,
      "learning_rate": 5e-06,
      "loss": 0.6447,
      "step": 220
    },
    {
      "epoch": 0.7516339869281046,
      "grad_norm": 0.6551946905014553,
      "learning_rate": 5e-06,
      "loss": 0.6441,
      "step": 230
    },
    {
      "epoch": 0.7843137254901961,
      "grad_norm": 0.7098356396395941,
      "learning_rate": 5e-06,
      "loss": 0.6404,
      "step": 240
    },
    {
      "epoch": 0.8169934640522876,
      "grad_norm": 0.6955204100907911,
      "learning_rate": 5e-06,
      "loss": 0.6328,
      "step": 250
    },
    {
      "epoch": 0.8496732026143791,
      "grad_norm": 0.8535008368638369,
      "learning_rate": 5e-06,
      "loss": 0.6376,
      "step": 260
    },
    {
      "epoch": 0.8823529411764706,
      "grad_norm": 0.9070401054614784,
      "learning_rate": 5e-06,
      "loss": 0.639,
      "step": 270
    },
    {
      "epoch": 0.9150326797385621,
      "grad_norm": 1.565789969572099,
      "learning_rate": 5e-06,
      "loss": 0.6359,
      "step": 280
    },
    {
      "epoch": 0.9477124183006536,
      "grad_norm": 1.2559271663411875,
      "learning_rate": 5e-06,
      "loss": 0.634,
      "step": 290
    },
    {
      "epoch": 0.9803921568627451,
      "grad_norm": 0.7809173302366459,
      "learning_rate": 5e-06,
      "loss": 0.6319,
      "step": 300
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.6343755125999451,
      "eval_runtime": 30.4377,
      "eval_samples_per_second": 270.75,
      "eval_steps_per_second": 1.084,
      "step": 306
    },
    {
      "epoch": 1.0130718954248366,
      "grad_norm": 0.8552427081362785,
      "learning_rate": 5e-06,
      "loss": 0.613,
      "step": 310
    },
    {
      "epoch": 1.0457516339869282,
      "grad_norm": 0.7483473424289742,
      "learning_rate": 5e-06,
      "loss": 0.5907,
      "step": 320
    },
    {
      "epoch": 1.0784313725490196,
      "grad_norm": 0.954467001493486,
      "learning_rate": 5e-06,
      "loss": 0.5879,
      "step": 330
    },
    {
      "epoch": 1.1111111111111112,
      "grad_norm": 0.5442657939315798,
      "learning_rate": 5e-06,
      "loss": 0.5885,
      "step": 340
    },
    {
      "epoch": 1.1437908496732025,
      "grad_norm": 0.7256618392119529,
      "learning_rate": 5e-06,
      "loss": 0.5913,
      "step": 350
    },
    {
      "epoch": 1.1764705882352942,
      "grad_norm": 0.6093376203471469,
      "learning_rate": 5e-06,
      "loss": 0.5941,
      "step": 360
    },
    {
      "epoch": 1.2091503267973855,
      "grad_norm": 0.563669838714874,
      "learning_rate": 5e-06,
      "loss": 0.5906,
      "step": 370
    },
    {
      "epoch": 1.2418300653594772,
      "grad_norm": 0.6235844786429067,
      "learning_rate": 5e-06,
      "loss": 0.5912,
      "step": 380
    },
    {
      "epoch": 1.2745098039215685,
      "grad_norm": 0.7954936323443693,
      "learning_rate": 5e-06,
      "loss": 0.5854,
      "step": 390
    },
    {
      "epoch": 1.3071895424836601,
      "grad_norm": 0.5900417117740443,
      "learning_rate": 5e-06,
      "loss": 0.5902,
      "step": 400
    },
    {
      "epoch": 1.3398692810457518,
      "grad_norm": 0.5649519787360927,
      "learning_rate": 5e-06,
      "loss": 0.5929,
      "step": 410
    },
    {
      "epoch": 1.3725490196078431,
      "grad_norm": 0.5532481138751179,
      "learning_rate": 5e-06,
      "loss": 0.5949,
      "step": 420
    },
    {
      "epoch": 1.4052287581699345,
      "grad_norm": 0.5796446258489036,
      "learning_rate": 5e-06,
      "loss": 0.591,
      "step": 430
    },
    {
      "epoch": 1.4379084967320261,
      "grad_norm": 0.5180305142972244,
      "learning_rate": 5e-06,
      "loss": 0.5955,
      "step": 440
    },
    {
      "epoch": 1.4705882352941178,
      "grad_norm": 0.5316504409753056,
      "learning_rate": 5e-06,
      "loss": 0.5916,
      "step": 450
    },
    {
      "epoch": 1.5032679738562091,
      "grad_norm": 0.5669201774818233,
      "learning_rate": 5e-06,
      "loss": 0.5884,
      "step": 460
    },
    {
      "epoch": 1.5359477124183005,
      "grad_norm": 0.5474083628106629,
      "learning_rate": 5e-06,
      "loss": 0.5896,
      "step": 470
    },
    {
      "epoch": 1.5686274509803921,
      "grad_norm": 0.5404778686275579,
      "learning_rate": 5e-06,
      "loss": 0.593,
      "step": 480
    },
    {
      "epoch": 1.6013071895424837,
      "grad_norm": 0.5872446910510764,
      "learning_rate": 5e-06,
      "loss": 0.5921,
      "step": 490
    },
    {
      "epoch": 1.6339869281045751,
      "grad_norm": 0.6115573952876731,
      "learning_rate": 5e-06,
      "loss": 0.5916,
      "step": 500
    },
    {
      "epoch": 1.6666666666666665,
      "grad_norm": 0.6081257294896645,
      "learning_rate": 5e-06,
      "loss": 0.5903,
      "step": 510
    },
    {
      "epoch": 1.6993464052287581,
      "grad_norm": 0.5308744191347566,
      "learning_rate": 5e-06,
      "loss": 0.5859,
      "step": 520
    },
    {
      "epoch": 1.7320261437908497,
      "grad_norm": 0.5659221276589558,
      "learning_rate": 5e-06,
      "loss": 0.5916,
      "step": 530
    },
    {
      "epoch": 1.7647058823529411,
      "grad_norm": 0.5192160551754689,
      "learning_rate": 5e-06,
      "loss": 0.5914,
      "step": 540
    },
    {
      "epoch": 1.7973856209150327,
      "grad_norm": 0.6988165837234015,
      "learning_rate": 5e-06,
      "loss": 0.5866,
      "step": 550
    },
    {
      "epoch": 1.8300653594771243,
      "grad_norm": 0.5487420924541063,
      "learning_rate": 5e-06,
      "loss": 0.5884,
      "step": 560
    },
    {
      "epoch": 1.8627450980392157,
      "grad_norm": 0.5682797543154547,
      "learning_rate": 5e-06,
      "loss": 0.5909,
      "step": 570
    },
    {
      "epoch": 1.8954248366013071,
      "grad_norm": 0.5077833002974906,
      "learning_rate": 5e-06,
      "loss": 0.5973,
      "step": 580
    },
    {
      "epoch": 1.9281045751633987,
      "grad_norm": 0.5129649155292699,
      "learning_rate": 5e-06,
      "loss": 0.5906,
      "step": 590
    },
    {
      "epoch": 1.9607843137254903,
      "grad_norm": 0.6199185848699229,
      "learning_rate": 5e-06,
      "loss": 0.5897,
      "step": 600
    },
    {
      "epoch": 1.9934640522875817,
      "grad_norm": 0.6367349679762161,
      "learning_rate": 5e-06,
      "loss": 0.5863,
      "step": 610
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.6249033808708191,
      "eval_runtime": 29.8328,
      "eval_samples_per_second": 276.24,
      "eval_steps_per_second": 1.106,
      "step": 612
    },
    {
      "epoch": 2.026143790849673,
      "grad_norm": 0.7411071608832002,
      "learning_rate": 5e-06,
      "loss": 0.5586,
      "step": 620
    },
    {
      "epoch": 2.0588235294117645,
      "grad_norm": 0.5717129772857855,
      "learning_rate": 5e-06,
      "loss": 0.5407,
      "step": 630
    },
    {
      "epoch": 2.0915032679738563,
      "grad_norm": 0.6346534892777344,
      "learning_rate": 5e-06,
      "loss": 0.5442,
      "step": 640
    },
    {
      "epoch": 2.1241830065359477,
      "grad_norm": 0.5360417918450974,
      "learning_rate": 5e-06,
      "loss": 0.5411,
      "step": 650
    },
    {
      "epoch": 2.156862745098039,
      "grad_norm": 0.6070614915525901,
      "learning_rate": 5e-06,
      "loss": 0.5495,
      "step": 660
    },
    {
      "epoch": 2.189542483660131,
      "grad_norm": 0.5484881674300732,
      "learning_rate": 5e-06,
      "loss": 0.5477,
      "step": 670
    },
    {
      "epoch": 2.2222222222222223,
      "grad_norm": 0.5559727309246413,
      "learning_rate": 5e-06,
      "loss": 0.5453,
      "step": 680
    },
    {
      "epoch": 2.2549019607843137,
      "grad_norm": 0.6149254917908307,
      "learning_rate": 5e-06,
      "loss": 0.5491,
      "step": 690
    },
    {
      "epoch": 2.287581699346405,
      "grad_norm": 0.7434819330252584,
      "learning_rate": 5e-06,
      "loss": 0.5411,
      "step": 700
    },
    {
      "epoch": 2.3202614379084965,
      "grad_norm": 0.6277100245140601,
      "learning_rate": 5e-06,
      "loss": 0.539,
      "step": 710
    },
    {
      "epoch": 2.3529411764705883,
      "grad_norm": 0.6071082517695824,
      "learning_rate": 5e-06,
      "loss": 0.546,
      "step": 720
    },
    {
      "epoch": 2.3856209150326797,
      "grad_norm": 0.5763605350200824,
      "learning_rate": 5e-06,
      "loss": 0.5424,
      "step": 730
    },
    {
      "epoch": 2.418300653594771,
      "grad_norm": 0.6503780611807637,
      "learning_rate": 5e-06,
      "loss": 0.5467,
      "step": 740
    },
    {
      "epoch": 2.450980392156863,
      "grad_norm": 0.6953693200719112,
      "learning_rate": 5e-06,
      "loss": 0.551,
      "step": 750
    },
    {
      "epoch": 2.4836601307189543,
      "grad_norm": 0.582324537988359,
      "learning_rate": 5e-06,
      "loss": 0.5454,
      "step": 760
    },
    {
      "epoch": 2.5163398692810457,
      "grad_norm": 0.5697978800379555,
      "learning_rate": 5e-06,
      "loss": 0.546,
      "step": 770
    },
    {
      "epoch": 2.549019607843137,
      "grad_norm": 0.588747491644582,
      "learning_rate": 5e-06,
      "loss": 0.5434,
      "step": 780
    },
    {
      "epoch": 2.581699346405229,
      "grad_norm": 0.5761332349094056,
      "learning_rate": 5e-06,
      "loss": 0.5515,
      "step": 790
    },
    {
      "epoch": 2.6143790849673203,
      "grad_norm": 0.6003606940754802,
      "learning_rate": 5e-06,
      "loss": 0.5513,
      "step": 800
    },
    {
      "epoch": 2.6470588235294117,
      "grad_norm": 0.5450155999410424,
      "learning_rate": 5e-06,
      "loss": 0.5472,
      "step": 810
    },
    {
      "epoch": 2.6797385620915035,
      "grad_norm": 0.7041726525142917,
      "learning_rate": 5e-06,
      "loss": 0.5496,
      "step": 820
    },
    {
      "epoch": 2.712418300653595,
      "grad_norm": 0.5257674189970212,
      "learning_rate": 5e-06,
      "loss": 0.5444,
      "step": 830
    },
    {
      "epoch": 2.7450980392156863,
      "grad_norm": 0.5960484874019031,
      "learning_rate": 5e-06,
      "loss": 0.5501,
      "step": 840
    },
    {
      "epoch": 2.7777777777777777,
      "grad_norm": 0.5362794145960937,
      "learning_rate": 5e-06,
      "loss": 0.5477,
      "step": 850
    },
    {
      "epoch": 2.810457516339869,
      "grad_norm": 0.5503663269672819,
      "learning_rate": 5e-06,
      "loss": 0.5472,
      "step": 860
    },
    {
      "epoch": 2.843137254901961,
      "grad_norm": 0.5500519539060107,
      "learning_rate": 5e-06,
      "loss": 0.5512,
      "step": 870
    },
    {
      "epoch": 2.8758169934640523,
      "grad_norm": 0.6512147682006916,
      "learning_rate": 5e-06,
      "loss": 0.5521,
      "step": 880
    },
    {
      "epoch": 2.9084967320261437,
      "grad_norm": 0.5740057468439662,
      "learning_rate": 5e-06,
      "loss": 0.5449,
      "step": 890
    },
    {
      "epoch": 2.9411764705882355,
      "grad_norm": 0.5364033145980043,
      "learning_rate": 5e-06,
      "loss": 0.5498,
      "step": 900
    },
    {
      "epoch": 2.973856209150327,
      "grad_norm": 0.5484821254955965,
      "learning_rate": 5e-06,
      "loss": 0.5495,
      "step": 910
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.628399670124054,
      "eval_runtime": 29.5588,
      "eval_samples_per_second": 278.8,
      "eval_steps_per_second": 1.116,
      "step": 918
    },
    {
      "epoch": 3.0,
      "step": 918,
      "total_flos": 1537684191313920.0,
      "train_loss": 0.6034856663290452,
      "train_runtime": 5898.5619,
      "train_samples_per_second": 79.628,
      "train_steps_per_second": 0.156
    }
  ],
  "logging_steps": 10,
  "max_steps": 918,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1537684191313920.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}