{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.9954337899543377,
  "eval_steps": 500,
  "global_step": 984,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.030441400304414,
      "grad_norm": 4.619646397901246,
      "learning_rate": 5e-06,
      "loss": 0.801,
      "step": 10
    },
    {
      "epoch": 0.060882800608828,
      "grad_norm": 2.4758507718701606,
      "learning_rate": 5e-06,
      "loss": 0.7282,
      "step": 20
    },
    {
      "epoch": 0.091324200913242,
      "grad_norm": 3.479812913897804,
      "learning_rate": 5e-06,
      "loss": 0.7045,
      "step": 30
    },
    {
      "epoch": 0.121765601217656,
      "grad_norm": 2.9360047427282385,
      "learning_rate": 5e-06,
      "loss": 0.6875,
      "step": 40
    },
    {
      "epoch": 0.15220700152207,
      "grad_norm": 0.973776971269499,
      "learning_rate": 5e-06,
      "loss": 0.6818,
      "step": 50
    },
    {
      "epoch": 0.182648401826484,
      "grad_norm": 0.800750811910126,
      "learning_rate": 5e-06,
      "loss": 0.6695,
      "step": 60
    },
    {
      "epoch": 0.213089802130898,
      "grad_norm": 0.7737759634387463,
      "learning_rate": 5e-06,
      "loss": 0.6519,
      "step": 70
    },
    {
      "epoch": 0.243531202435312,
      "grad_norm": 1.3731039805269152,
      "learning_rate": 5e-06,
      "loss": 0.6423,
      "step": 80
    },
    {
      "epoch": 0.273972602739726,
      "grad_norm": 0.6785230486405546,
      "learning_rate": 5e-06,
      "loss": 0.6501,
      "step": 90
    },
    {
      "epoch": 0.30441400304414,
      "grad_norm": 1.054981943761996,
      "learning_rate": 5e-06,
      "loss": 0.6403,
      "step": 100
    },
    {
      "epoch": 0.334855403348554,
      "grad_norm": 0.7312141288913386,
      "learning_rate": 5e-06,
      "loss": 0.6327,
      "step": 110
    },
    {
      "epoch": 0.365296803652968,
      "grad_norm": 0.611778635862804,
      "learning_rate": 5e-06,
      "loss": 0.6311,
      "step": 120
    },
    {
      "epoch": 0.395738203957382,
      "grad_norm": 0.6581343673138518,
      "learning_rate": 5e-06,
      "loss": 0.6383,
      "step": 130
    },
    {
      "epoch": 0.426179604261796,
      "grad_norm": 0.8961158759762995,
      "learning_rate": 5e-06,
      "loss": 0.6289,
      "step": 140
    },
    {
      "epoch": 0.45662100456621,
      "grad_norm": 0.5571046772698393,
      "learning_rate": 5e-06,
      "loss": 0.6315,
      "step": 150
    },
    {
      "epoch": 0.487062404870624,
      "grad_norm": 0.5225558082134438,
      "learning_rate": 5e-06,
      "loss": 0.6292,
      "step": 160
    },
    {
      "epoch": 0.517503805175038,
      "grad_norm": 0.5700430081742105,
      "learning_rate": 5e-06,
      "loss": 0.6274,
      "step": 170
    },
    {
      "epoch": 0.547945205479452,
      "grad_norm": 1.3638017063276748,
      "learning_rate": 5e-06,
      "loss": 0.6147,
      "step": 180
    },
    {
      "epoch": 0.578386605783866,
      "grad_norm": 1.1474194001541684,
      "learning_rate": 5e-06,
      "loss": 0.6214,
      "step": 190
    },
    {
      "epoch": 0.60882800608828,
      "grad_norm": 0.5983544869985015,
      "learning_rate": 5e-06,
      "loss": 0.6211,
      "step": 200
    },
    {
      "epoch": 0.639269406392694,
      "grad_norm": 0.5030983225008678,
      "learning_rate": 5e-06,
      "loss": 0.6269,
      "step": 210
    },
    {
      "epoch": 0.669710806697108,
      "grad_norm": 0.5355894732987774,
      "learning_rate": 5e-06,
      "loss": 0.6186,
      "step": 220
    },
    {
      "epoch": 0.700152207001522,
      "grad_norm": 0.5535858496033256,
      "learning_rate": 5e-06,
      "loss": 0.6174,
      "step": 230
    },
    {
      "epoch": 0.730593607305936,
      "grad_norm": 0.4757423728889985,
      "learning_rate": 5e-06,
      "loss": 0.6157,
      "step": 240
    },
    {
      "epoch": 0.76103500761035,
      "grad_norm": 0.5846472040453055,
      "learning_rate": 5e-06,
      "loss": 0.6119,
      "step": 250
    },
    {
      "epoch": 0.791476407914764,
      "grad_norm": 0.5064988157906138,
      "learning_rate": 5e-06,
      "loss": 0.6174,
      "step": 260
    },
    {
      "epoch": 0.821917808219178,
      "grad_norm": 0.5603859566155508,
      "learning_rate": 5e-06,
      "loss": 0.6115,
      "step": 270
    },
    {
      "epoch": 0.852359208523592,
      "grad_norm": 0.5340207450065109,
      "learning_rate": 5e-06,
      "loss": 0.618,
      "step": 280
    },
    {
      "epoch": 0.882800608828006,
      "grad_norm": 0.5176245531200292,
      "learning_rate": 5e-06,
      "loss": 0.6141,
      "step": 290
    },
    {
      "epoch": 0.91324200913242,
      "grad_norm": 0.5686518744643043,
      "learning_rate": 5e-06,
      "loss": 0.61,
      "step": 300
    },
    {
      "epoch": 0.943683409436834,
      "grad_norm": 0.5557582588776627,
      "learning_rate": 5e-06,
      "loss": 0.6086,
      "step": 310
    },
    {
      "epoch": 0.974124809741248,
      "grad_norm": 0.48632226958782565,
      "learning_rate": 5e-06,
      "loss": 0.6134,
      "step": 320
    },
    {
      "epoch": 0.9984779299847792,
      "eval_loss": 0.6086177825927734,
      "eval_runtime": 177.842,
      "eval_samples_per_second": 49.758,
      "eval_steps_per_second": 0.394,
      "step": 328
    },
    {
      "epoch": 1.004566210045662,
      "grad_norm": 0.5349618575586927,
      "learning_rate": 5e-06,
      "loss": 0.6066,
      "step": 330
    },
    {
      "epoch": 1.035007610350076,
      "grad_norm": 0.7085072078930811,
      "learning_rate": 5e-06,
      "loss": 0.5583,
      "step": 340
    },
    {
      "epoch": 1.06544901065449,
      "grad_norm": 0.5086842387297956,
      "learning_rate": 5e-06,
      "loss": 0.5577,
      "step": 350
    },
    {
      "epoch": 1.095890410958904,
      "grad_norm": 0.5618625623080041,
      "learning_rate": 5e-06,
      "loss": 0.5692,
      "step": 360
    },
    {
      "epoch": 1.126331811263318,
      "grad_norm": 0.8882124898558976,
      "learning_rate": 5e-06,
      "loss": 0.5622,
      "step": 370
    },
    {
      "epoch": 1.156773211567732,
      "grad_norm": 0.8010806701357303,
      "learning_rate": 5e-06,
      "loss": 0.5626,
      "step": 380
    },
    {
      "epoch": 1.187214611872146,
      "grad_norm": 0.6044300304393334,
      "learning_rate": 5e-06,
      "loss": 0.564,
      "step": 390
    },
    {
      "epoch": 1.21765601217656,
      "grad_norm": 0.6478561685693183,
      "learning_rate": 5e-06,
      "loss": 0.5599,
      "step": 400
    },
    {
      "epoch": 1.248097412480974,
      "grad_norm": 0.48517644823822886,
      "learning_rate": 5e-06,
      "loss": 0.5629,
      "step": 410
    },
    {
      "epoch": 1.278538812785388,
      "grad_norm": 0.6579745015004641,
      "learning_rate": 5e-06,
      "loss": 0.5688,
      "step": 420
    },
    {
      "epoch": 1.308980213089802,
      "grad_norm": 0.5933025521185841,
      "learning_rate": 5e-06,
      "loss": 0.5625,
      "step": 430
    },
    {
      "epoch": 1.339421613394216,
      "grad_norm": 0.5524692856580558,
      "learning_rate": 5e-06,
      "loss": 0.5645,
      "step": 440
    },
    {
      "epoch": 1.36986301369863,
      "grad_norm": 0.5949700286062661,
      "learning_rate": 5e-06,
      "loss": 0.5667,
      "step": 450
    },
    {
      "epoch": 1.400304414003044,
      "grad_norm": 0.7922305492429499,
      "learning_rate": 5e-06,
      "loss": 0.5671,
      "step": 460
    },
    {
      "epoch": 1.430745814307458,
      "grad_norm": 0.5035984988245502,
      "learning_rate": 5e-06,
      "loss": 0.5699,
      "step": 470
    },
    {
      "epoch": 1.461187214611872,
      "grad_norm": 0.4675577590538898,
      "learning_rate": 5e-06,
      "loss": 0.5622,
      "step": 480
    },
    {
      "epoch": 1.491628614916286,
      "grad_norm": 0.6149051107408182,
      "learning_rate": 5e-06,
      "loss": 0.568,
      "step": 490
    },
    {
      "epoch": 1.5220700152207,
      "grad_norm": 0.624211800852349,
      "learning_rate": 5e-06,
      "loss": 0.5641,
      "step": 500
    },
    {
      "epoch": 1.5525114155251143,
      "grad_norm": 0.9181588884042637,
      "learning_rate": 5e-06,
      "loss": 0.5666,
      "step": 510
    },
    {
      "epoch": 1.582952815829528,
      "grad_norm": 0.4937076242442424,
      "learning_rate": 5e-06,
      "loss": 0.5666,
      "step": 520
    },
    {
      "epoch": 1.6133942161339423,
      "grad_norm": 0.5082102469467535,
      "learning_rate": 5e-06,
      "loss": 0.5574,
      "step": 530
    },
    {
      "epoch": 1.643835616438356,
      "grad_norm": 0.4937565845275598,
      "learning_rate": 5e-06,
      "loss": 0.5689,
      "step": 540
    },
    {
      "epoch": 1.6742770167427703,
      "grad_norm": 0.4848002546906178,
      "learning_rate": 5e-06,
      "loss": 0.5688,
      "step": 550
    },
    {
      "epoch": 1.704718417047184,
      "grad_norm": 0.5230615787294617,
      "learning_rate": 5e-06,
      "loss": 0.5677,
      "step": 560
    },
    {
      "epoch": 1.7351598173515983,
      "grad_norm": 0.4976273028384401,
      "learning_rate": 5e-06,
      "loss": 0.5586,
      "step": 570
    },
    {
      "epoch": 1.765601217656012,
      "grad_norm": 0.49403463868422004,
      "learning_rate": 5e-06,
      "loss": 0.5604,
      "step": 580
    },
    {
      "epoch": 1.7960426179604263,
      "grad_norm": 0.6320569692146091,
      "learning_rate": 5e-06,
      "loss": 0.5578,
      "step": 590
    },
    {
      "epoch": 1.82648401826484,
      "grad_norm": 0.5695898310066485,
      "learning_rate": 5e-06,
      "loss": 0.5593,
      "step": 600
    },
    {
      "epoch": 1.8569254185692543,
      "grad_norm": 0.49537917557389,
      "learning_rate": 5e-06,
      "loss": 0.5627,
      "step": 610
    },
    {
      "epoch": 1.887366818873668,
      "grad_norm": 0.6247132495885726,
      "learning_rate": 5e-06,
      "loss": 0.5621,
      "step": 620
    },
    {
      "epoch": 1.9178082191780823,
      "grad_norm": 0.5030338206961641,
      "learning_rate": 5e-06,
      "loss": 0.5598,
      "step": 630
    },
    {
      "epoch": 1.948249619482496,
      "grad_norm": 0.4962258563588259,
      "learning_rate": 5e-06,
      "loss": 0.5562,
      "step": 640
    },
    {
      "epoch": 1.9786910197869103,
      "grad_norm": 0.49909928012252913,
      "learning_rate": 5e-06,
      "loss": 0.5568,
      "step": 650
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.5997375249862671,
      "eval_runtime": 178.4917,
      "eval_samples_per_second": 49.577,
      "eval_steps_per_second": 0.392,
      "step": 657
    },
    {
      "epoch": 2.009132420091324,
      "grad_norm": 0.8861960444337913,
      "learning_rate": 5e-06,
      "loss": 0.5441,
      "step": 660
    },
    {
      "epoch": 2.0395738203957383,
      "grad_norm": 0.5425502389160675,
      "learning_rate": 5e-06,
      "loss": 0.514,
      "step": 670
    },
    {
      "epoch": 2.070015220700152,
      "grad_norm": 0.5562526095704428,
      "learning_rate": 5e-06,
      "loss": 0.5032,
      "step": 680
    },
    {
      "epoch": 2.1004566210045663,
      "grad_norm": 0.49212370632524655,
      "learning_rate": 5e-06,
      "loss": 0.5154,
      "step": 690
    },
    {
      "epoch": 2.13089802130898,
      "grad_norm": 0.534245423835535,
      "learning_rate": 5e-06,
      "loss": 0.5116,
      "step": 700
    },
    {
      "epoch": 2.1613394216133943,
      "grad_norm": 0.5454909388550724,
      "learning_rate": 5e-06,
      "loss": 0.5136,
      "step": 710
    },
    {
      "epoch": 2.191780821917808,
      "grad_norm": 0.8231087538933396,
      "learning_rate": 5e-06,
      "loss": 0.5052,
      "step": 720
    },
    {
      "epoch": 2.2222222222222223,
      "grad_norm": 0.5745911989069736,
      "learning_rate": 5e-06,
      "loss": 0.5157,
      "step": 730
    },
    {
      "epoch": 2.252663622526636,
      "grad_norm": 0.5370023428369475,
      "learning_rate": 5e-06,
      "loss": 0.5142,
      "step": 740
    },
    {
      "epoch": 2.2831050228310503,
      "grad_norm": 0.5417242182114983,
      "learning_rate": 5e-06,
      "loss": 0.507,
      "step": 750
    },
    {
      "epoch": 2.313546423135464,
      "grad_norm": 0.4948430389470553,
      "learning_rate": 5e-06,
      "loss": 0.5071,
      "step": 760
    },
    {
      "epoch": 2.3439878234398783,
      "grad_norm": 0.5823872991278739,
      "learning_rate": 5e-06,
      "loss": 0.5152,
      "step": 770
    },
    {
      "epoch": 2.374429223744292,
      "grad_norm": 0.5749628913643053,
      "learning_rate": 5e-06,
      "loss": 0.5158,
      "step": 780
    },
    {
      "epoch": 2.4048706240487063,
      "grad_norm": 0.523744812578657,
      "learning_rate": 5e-06,
      "loss": 0.5121,
      "step": 790
    },
    {
      "epoch": 2.43531202435312,
      "grad_norm": 0.6430805814437145,
      "learning_rate": 5e-06,
      "loss": 0.5187,
      "step": 800
    },
    {
      "epoch": 2.4657534246575343,
      "grad_norm": 0.6135659154747556,
      "learning_rate": 5e-06,
      "loss": 0.5105,
      "step": 810
    },
    {
      "epoch": 2.496194824961948,
      "grad_norm": 0.5819355389015244,
      "learning_rate": 5e-06,
      "loss": 0.5127,
      "step": 820
    },
    {
      "epoch": 2.5266362252663623,
      "grad_norm": 0.5222652845841953,
      "learning_rate": 5e-06,
      "loss": 0.5172,
      "step": 830
    },
    {
      "epoch": 2.557077625570776,
      "grad_norm": 0.5085292250908287,
      "learning_rate": 5e-06,
      "loss": 0.5154,
      "step": 840
    },
    {
      "epoch": 2.5875190258751903,
      "grad_norm": 0.5051693370140233,
      "learning_rate": 5e-06,
      "loss": 0.5122,
      "step": 850
    },
    {
      "epoch": 2.617960426179604,
      "grad_norm": 0.5219920053064364,
      "learning_rate": 5e-06,
      "loss": 0.5154,
      "step": 860
    },
    {
      "epoch": 2.6484018264840183,
      "grad_norm": 0.5918822792144229,
      "learning_rate": 5e-06,
      "loss": 0.5141,
      "step": 870
    },
    {
      "epoch": 2.678843226788432,
      "grad_norm": 0.522542434998343,
      "learning_rate": 5e-06,
      "loss": 0.5117,
      "step": 880
    },
    {
      "epoch": 2.7092846270928463,
      "grad_norm": 0.5321626854739884,
      "learning_rate": 5e-06,
      "loss": 0.5073,
      "step": 890
    },
    {
      "epoch": 2.73972602739726,
      "grad_norm": 0.5099777411434487,
      "learning_rate": 5e-06,
      "loss": 0.5202,
      "step": 900
    },
    {
      "epoch": 2.7701674277016743,
      "grad_norm": 0.5225721184412385,
      "learning_rate": 5e-06,
      "loss": 0.5204,
      "step": 910
    },
    {
      "epoch": 2.800608828006088,
      "grad_norm": 0.49419482852544927,
      "learning_rate": 5e-06,
      "loss": 0.5117,
      "step": 920
    },
    {
      "epoch": 2.8310502283105023,
      "grad_norm": 0.5580841932281034,
      "learning_rate": 5e-06,
      "loss": 0.5125,
      "step": 930
    },
    {
      "epoch": 2.861491628614916,
      "grad_norm": 0.5459012978192056,
      "learning_rate": 5e-06,
      "loss": 0.5159,
      "step": 940
    },
    {
      "epoch": 2.8919330289193304,
      "grad_norm": 0.4931927553237747,
      "learning_rate": 5e-06,
      "loss": 0.516,
      "step": 950
    },
    {
      "epoch": 2.922374429223744,
      "grad_norm": 0.5125805588681561,
      "learning_rate": 5e-06,
      "loss": 0.5204,
      "step": 960
    },
    {
      "epoch": 2.9528158295281584,
      "grad_norm": 0.5359088405972897,
      "learning_rate": 5e-06,
      "loss": 0.5126,
      "step": 970
    },
    {
      "epoch": 2.983257229832572,
      "grad_norm": 0.5366173084145919,
      "learning_rate": 5e-06,
      "loss": 0.507,
      "step": 980
    },
    {
      "epoch": 2.9954337899543377,
      "eval_loss": 0.6046163439750671,
      "eval_runtime": 178.3759,
      "eval_samples_per_second": 49.609,
      "eval_steps_per_second": 0.392,
      "step": 984
    },
    {
      "epoch": 2.9954337899543377,
      "step": 984,
      "total_flos": 1647817890201600.0,
      "train_loss": 0.5723786247455007,
      "train_runtime": 29605.407,
      "train_samples_per_second": 17.036,
      "train_steps_per_second": 0.033
    }
  ],
  "logging_steps": 10,
  "max_steps": 984,
  "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": 1647817890201600.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}