{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 981,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03058103975535168,
      "grad_norm": 4.613314070863576,
      "learning_rate": 5e-06,
      "loss": 0.8051,
      "step": 10
    },
    {
      "epoch": 0.06116207951070336,
      "grad_norm": 1.8900295063815196,
      "learning_rate": 5e-06,
      "loss": 0.7288,
      "step": 20
    },
    {
      "epoch": 0.09174311926605505,
      "grad_norm": 2.828490629262087,
      "learning_rate": 5e-06,
      "loss": 0.709,
      "step": 30
    },
    {
      "epoch": 0.12232415902140673,
      "grad_norm": 1.3871149266999225,
      "learning_rate": 5e-06,
      "loss": 0.6966,
      "step": 40
    },
    {
      "epoch": 0.1529051987767584,
      "grad_norm": 1.5967435207002258,
      "learning_rate": 5e-06,
      "loss": 0.6812,
      "step": 50
    },
    {
      "epoch": 0.1834862385321101,
      "grad_norm": 1.882517891139545,
      "learning_rate": 5e-06,
      "loss": 0.6728,
      "step": 60
    },
    {
      "epoch": 0.21406727828746178,
      "grad_norm": 0.6460792834326635,
      "learning_rate": 5e-06,
      "loss": 0.661,
      "step": 70
    },
    {
      "epoch": 0.24464831804281345,
      "grad_norm": 0.8345961449379203,
      "learning_rate": 5e-06,
      "loss": 0.6527,
      "step": 80
    },
    {
      "epoch": 0.27522935779816515,
      "grad_norm": 0.6632736237849782,
      "learning_rate": 5e-06,
      "loss": 0.65,
      "step": 90
    },
    {
      "epoch": 0.3058103975535168,
      "grad_norm": 0.6414397366786443,
      "learning_rate": 5e-06,
      "loss": 0.6416,
      "step": 100
    },
    {
      "epoch": 0.3363914373088685,
      "grad_norm": 0.6410573252326648,
      "learning_rate": 5e-06,
      "loss": 0.6423,
      "step": 110
    },
    {
      "epoch": 0.3669724770642202,
      "grad_norm": 0.6301173144669027,
      "learning_rate": 5e-06,
      "loss": 0.6431,
      "step": 120
    },
    {
      "epoch": 0.39755351681957185,
      "grad_norm": 0.6700047532457443,
      "learning_rate": 5e-06,
      "loss": 0.6328,
      "step": 130
    },
    {
      "epoch": 0.42813455657492355,
      "grad_norm": 0.7253886574556052,
      "learning_rate": 5e-06,
      "loss": 0.6379,
      "step": 140
    },
    {
      "epoch": 0.45871559633027525,
      "grad_norm": 0.6468641147049369,
      "learning_rate": 5e-06,
      "loss": 0.6226,
      "step": 150
    },
    {
      "epoch": 0.4892966360856269,
      "grad_norm": 0.5346182726002575,
      "learning_rate": 5e-06,
      "loss": 0.6297,
      "step": 160
    },
    {
      "epoch": 0.5198776758409785,
      "grad_norm": 0.5313857610798393,
      "learning_rate": 5e-06,
      "loss": 0.6315,
      "step": 170
    },
    {
      "epoch": 0.5504587155963303,
      "grad_norm": 1.055453437070951,
      "learning_rate": 5e-06,
      "loss": 0.6319,
      "step": 180
    },
    {
      "epoch": 0.581039755351682,
      "grad_norm": 1.0346764494649112,
      "learning_rate": 5e-06,
      "loss": 0.6336,
      "step": 190
    },
    {
      "epoch": 0.6116207951070336,
      "grad_norm": 0.5618281272496094,
      "learning_rate": 5e-06,
      "loss": 0.6219,
      "step": 200
    },
    {
      "epoch": 0.6422018348623854,
      "grad_norm": 1.3519206618987067,
      "learning_rate": 5e-06,
      "loss": 0.6231,
      "step": 210
    },
    {
      "epoch": 0.672782874617737,
      "grad_norm": 0.7749478594608031,
      "learning_rate": 5e-06,
      "loss": 0.6242,
      "step": 220
    },
    {
      "epoch": 0.7033639143730887,
      "grad_norm": 0.5893825196950665,
      "learning_rate": 5e-06,
      "loss": 0.6241,
      "step": 230
    },
    {
      "epoch": 0.7339449541284404,
      "grad_norm": 0.6773691040863971,
      "learning_rate": 5e-06,
      "loss": 0.6222,
      "step": 240
    },
    {
      "epoch": 0.764525993883792,
      "grad_norm": 0.4827169638012845,
      "learning_rate": 5e-06,
      "loss": 0.619,
      "step": 250
    },
    {
      "epoch": 0.7951070336391437,
      "grad_norm": 0.6490833986094754,
      "learning_rate": 5e-06,
      "loss": 0.6214,
      "step": 260
    },
    {
      "epoch": 0.8256880733944955,
      "grad_norm": 0.4630327930179835,
      "learning_rate": 5e-06,
      "loss": 0.6234,
      "step": 270
    },
    {
      "epoch": 0.8562691131498471,
      "grad_norm": 0.5519953643760132,
      "learning_rate": 5e-06,
      "loss": 0.6238,
      "step": 280
    },
    {
      "epoch": 0.8868501529051988,
      "grad_norm": 0.6014046385653471,
      "learning_rate": 5e-06,
      "loss": 0.6205,
      "step": 290
    },
    {
      "epoch": 0.9174311926605505,
      "grad_norm": 0.9905714959978613,
      "learning_rate": 5e-06,
      "loss": 0.614,
      "step": 300
    },
    {
      "epoch": 0.9480122324159022,
      "grad_norm": 0.5722101718286907,
      "learning_rate": 5e-06,
      "loss": 0.6174,
      "step": 310
    },
    {
      "epoch": 0.9785932721712538,
      "grad_norm": 0.49394030115956855,
      "learning_rate": 5e-06,
      "loss": 0.6318,
      "step": 320
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.6179984211921692,
      "eval_runtime": 175.4761,
      "eval_samples_per_second": 50.218,
      "eval_steps_per_second": 0.393,
      "step": 327
    },
    {
      "epoch": 1.0091743119266054,
      "grad_norm": 0.7689781158724909,
      "learning_rate": 5e-06,
      "loss": 0.6099,
      "step": 330
    },
    {
      "epoch": 1.039755351681957,
      "grad_norm": 0.6771992725638727,
      "learning_rate": 5e-06,
      "loss": 0.5636,
      "step": 340
    },
    {
      "epoch": 1.070336391437309,
      "grad_norm": 0.8428533582721723,
      "learning_rate": 5e-06,
      "loss": 0.5702,
      "step": 350
    },
    {
      "epoch": 1.1009174311926606,
      "grad_norm": 0.6417949273070751,
      "learning_rate": 5e-06,
      "loss": 0.5724,
      "step": 360
    },
    {
      "epoch": 1.1314984709480123,
      "grad_norm": 0.4739060770421859,
      "learning_rate": 5e-06,
      "loss": 0.5667,
      "step": 370
    },
    {
      "epoch": 1.162079510703364,
      "grad_norm": 0.8290714956133294,
      "learning_rate": 5e-06,
      "loss": 0.5723,
      "step": 380
    },
    {
      "epoch": 1.1926605504587156,
      "grad_norm": 0.4621451043452794,
      "learning_rate": 5e-06,
      "loss": 0.5712,
      "step": 390
    },
    {
      "epoch": 1.2232415902140672,
      "grad_norm": 0.5143870327303872,
      "learning_rate": 5e-06,
      "loss": 0.5665,
      "step": 400
    },
    {
      "epoch": 1.2538226299694188,
      "grad_norm": 0.5164679975473679,
      "learning_rate": 5e-06,
      "loss": 0.5738,
      "step": 410
    },
    {
      "epoch": 1.2844036697247707,
      "grad_norm": 0.8444328502616898,
      "learning_rate": 5e-06,
      "loss": 0.568,
      "step": 420
    },
    {
      "epoch": 1.3149847094801224,
      "grad_norm": 0.5068551444160476,
      "learning_rate": 5e-06,
      "loss": 0.5685,
      "step": 430
    },
    {
      "epoch": 1.345565749235474,
      "grad_norm": 0.5272187758950088,
      "learning_rate": 5e-06,
      "loss": 0.5719,
      "step": 440
    },
    {
      "epoch": 1.3761467889908257,
      "grad_norm": 0.6560982973750454,
      "learning_rate": 5e-06,
      "loss": 0.5631,
      "step": 450
    },
    {
      "epoch": 1.4067278287461773,
      "grad_norm": 0.4484289597201794,
      "learning_rate": 5e-06,
      "loss": 0.5662,
      "step": 460
    },
    {
      "epoch": 1.4373088685015292,
      "grad_norm": 0.5845602559047488,
      "learning_rate": 5e-06,
      "loss": 0.5555,
      "step": 470
    },
    {
      "epoch": 1.4678899082568808,
      "grad_norm": 0.5019202622500104,
      "learning_rate": 5e-06,
      "loss": 0.572,
      "step": 480
    },
    {
      "epoch": 1.4984709480122325,
      "grad_norm": 0.5453352197296611,
      "learning_rate": 5e-06,
      "loss": 0.5678,
      "step": 490
    },
    {
      "epoch": 1.529051987767584,
      "grad_norm": 0.5096577153134583,
      "learning_rate": 5e-06,
      "loss": 0.5747,
      "step": 500
    },
    {
      "epoch": 1.5596330275229358,
      "grad_norm": 0.6175776252130769,
      "learning_rate": 5e-06,
      "loss": 0.5664,
      "step": 510
    },
    {
      "epoch": 1.5902140672782874,
      "grad_norm": 0.5104602945006693,
      "learning_rate": 5e-06,
      "loss": 0.5634,
      "step": 520
    },
    {
      "epoch": 1.620795107033639,
      "grad_norm": 0.596086026271991,
      "learning_rate": 5e-06,
      "loss": 0.5751,
      "step": 530
    },
    {
      "epoch": 1.6513761467889907,
      "grad_norm": 0.5308994737717756,
      "learning_rate": 5e-06,
      "loss": 0.5733,
      "step": 540
    },
    {
      "epoch": 1.6819571865443423,
      "grad_norm": 0.4845901344757882,
      "learning_rate": 5e-06,
      "loss": 0.5717,
      "step": 550
    },
    {
      "epoch": 1.7125382262996942,
      "grad_norm": 0.5671326796569592,
      "learning_rate": 5e-06,
      "loss": 0.5724,
      "step": 560
    },
    {
      "epoch": 1.7431192660550459,
      "grad_norm": 0.4598174188757565,
      "learning_rate": 5e-06,
      "loss": 0.5683,
      "step": 570
    },
    {
      "epoch": 1.7737003058103975,
      "grad_norm": 0.5112157139948377,
      "learning_rate": 5e-06,
      "loss": 0.5681,
      "step": 580
    },
    {
      "epoch": 1.8042813455657494,
      "grad_norm": 0.5586531166082738,
      "learning_rate": 5e-06,
      "loss": 0.5684,
      "step": 590
    },
    {
      "epoch": 1.834862385321101,
      "grad_norm": 0.4688565551712795,
      "learning_rate": 5e-06,
      "loss": 0.5644,
      "step": 600
    },
    {
      "epoch": 1.8654434250764527,
      "grad_norm": 0.5068561602224454,
      "learning_rate": 5e-06,
      "loss": 0.5634,
      "step": 610
    },
    {
      "epoch": 1.8960244648318043,
      "grad_norm": 0.5177020231748777,
      "learning_rate": 5e-06,
      "loss": 0.5693,
      "step": 620
    },
    {
      "epoch": 1.926605504587156,
      "grad_norm": 0.5480741145181502,
      "learning_rate": 5e-06,
      "loss": 0.5613,
      "step": 630
    },
    {
      "epoch": 1.9571865443425076,
      "grad_norm": 0.4783567812818659,
      "learning_rate": 5e-06,
      "loss": 0.5585,
      "step": 640
    },
    {
      "epoch": 1.9877675840978593,
      "grad_norm": 0.5551088225058829,
      "learning_rate": 5e-06,
      "loss": 0.5674,
      "step": 650
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.6093349456787109,
      "eval_runtime": 174.9969,
      "eval_samples_per_second": 50.355,
      "eval_steps_per_second": 0.394,
      "step": 654
    },
    {
      "epoch": 2.018348623853211,
      "grad_norm": 0.6215337411502556,
      "learning_rate": 5e-06,
      "loss": 0.5384,
      "step": 660
    },
    {
      "epoch": 2.0489296636085625,
      "grad_norm": 0.6523651145276358,
      "learning_rate": 5e-06,
      "loss": 0.5183,
      "step": 670
    },
    {
      "epoch": 2.079510703363914,
      "grad_norm": 0.4938873169688457,
      "learning_rate": 5e-06,
      "loss": 0.5154,
      "step": 680
    },
    {
      "epoch": 2.1100917431192663,
      "grad_norm": 0.5700577847392374,
      "learning_rate": 5e-06,
      "loss": 0.513,
      "step": 690
    },
    {
      "epoch": 2.140672782874618,
      "grad_norm": 0.6865444356940279,
      "learning_rate": 5e-06,
      "loss": 0.5107,
      "step": 700
    },
    {
      "epoch": 2.1712538226299696,
      "grad_norm": 0.49230256627015,
      "learning_rate": 5e-06,
      "loss": 0.5109,
      "step": 710
    },
    {
      "epoch": 2.2018348623853212,
      "grad_norm": 0.8680096117870334,
      "learning_rate": 5e-06,
      "loss": 0.5201,
      "step": 720
    },
    {
      "epoch": 2.232415902140673,
      "grad_norm": 0.5282977989250981,
      "learning_rate": 5e-06,
      "loss": 0.5176,
      "step": 730
    },
    {
      "epoch": 2.2629969418960245,
      "grad_norm": 0.5641604277626704,
      "learning_rate": 5e-06,
      "loss": 0.5175,
      "step": 740
    },
    {
      "epoch": 2.293577981651376,
      "grad_norm": 0.5627994676639944,
      "learning_rate": 5e-06,
      "loss": 0.5233,
      "step": 750
    },
    {
      "epoch": 2.324159021406728,
      "grad_norm": 0.5351783170372003,
      "learning_rate": 5e-06,
      "loss": 0.5193,
      "step": 760
    },
    {
      "epoch": 2.3547400611620795,
      "grad_norm": 0.5159357728539045,
      "learning_rate": 5e-06,
      "loss": 0.5145,
      "step": 770
    },
    {
      "epoch": 2.385321100917431,
      "grad_norm": 0.6104068286820499,
      "learning_rate": 5e-06,
      "loss": 0.5164,
      "step": 780
    },
    {
      "epoch": 2.4159021406727827,
      "grad_norm": 0.677498087908613,
      "learning_rate": 5e-06,
      "loss": 0.5262,
      "step": 790
    },
    {
      "epoch": 2.4464831804281344,
      "grad_norm": 0.5283675690505885,
      "learning_rate": 5e-06,
      "loss": 0.5149,
      "step": 800
    },
    {
      "epoch": 2.477064220183486,
      "grad_norm": 0.5044083108738047,
      "learning_rate": 5e-06,
      "loss": 0.5176,
      "step": 810
    },
    {
      "epoch": 2.5076452599388377,
      "grad_norm": 0.49494102207897933,
      "learning_rate": 5e-06,
      "loss": 0.5204,
      "step": 820
    },
    {
      "epoch": 2.5382262996941893,
      "grad_norm": 0.4861352677652072,
      "learning_rate": 5e-06,
      "loss": 0.5337,
      "step": 830
    },
    {
      "epoch": 2.5688073394495414,
      "grad_norm": 0.48851988662021156,
      "learning_rate": 5e-06,
      "loss": 0.5238,
      "step": 840
    },
    {
      "epoch": 2.599388379204893,
      "grad_norm": 0.5226833608668234,
      "learning_rate": 5e-06,
      "loss": 0.5205,
      "step": 850
    },
    {
      "epoch": 2.6299694189602447,
      "grad_norm": 0.5465654006210326,
      "learning_rate": 5e-06,
      "loss": 0.5221,
      "step": 860
    },
    {
      "epoch": 2.6605504587155964,
      "grad_norm": 0.5039029213538379,
      "learning_rate": 5e-06,
      "loss": 0.5196,
      "step": 870
    },
    {
      "epoch": 2.691131498470948,
      "grad_norm": 0.5371018828919037,
      "learning_rate": 5e-06,
      "loss": 0.5237,
      "step": 880
    },
    {
      "epoch": 2.7217125382262997,
      "grad_norm": 0.5383889127160468,
      "learning_rate": 5e-06,
      "loss": 0.5169,
      "step": 890
    },
    {
      "epoch": 2.7522935779816513,
      "grad_norm": 0.5406840391563221,
      "learning_rate": 5e-06,
      "loss": 0.5116,
      "step": 900
    },
    {
      "epoch": 2.782874617737003,
      "grad_norm": 0.5182378062317926,
      "learning_rate": 5e-06,
      "loss": 0.5147,
      "step": 910
    },
    {
      "epoch": 2.8134556574923546,
      "grad_norm": 0.483759876679847,
      "learning_rate": 5e-06,
      "loss": 0.514,
      "step": 920
    },
    {
      "epoch": 2.8440366972477067,
      "grad_norm": 0.5715170886676529,
      "learning_rate": 5e-06,
      "loss": 0.5189,
      "step": 930
    },
    {
      "epoch": 2.8746177370030583,
      "grad_norm": 0.4882822153954844,
      "learning_rate": 5e-06,
      "loss": 0.5131,
      "step": 940
    },
    {
      "epoch": 2.90519877675841,
      "grad_norm": 0.518462889966442,
      "learning_rate": 5e-06,
      "loss": 0.5228,
      "step": 950
    },
    {
      "epoch": 2.9357798165137616,
      "grad_norm": 0.500511829330958,
      "learning_rate": 5e-06,
      "loss": 0.5177,
      "step": 960
    },
    {
      "epoch": 2.9663608562691133,
      "grad_norm": 0.4799796524761594,
      "learning_rate": 5e-06,
      "loss": 0.5283,
      "step": 970
    },
    {
      "epoch": 2.996941896024465,
      "grad_norm": 0.6756287967133653,
      "learning_rate": 5e-06,
      "loss": 0.5228,
      "step": 980
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.6134681701660156,
      "eval_runtime": 174.9719,
      "eval_samples_per_second": 50.362,
      "eval_steps_per_second": 0.394,
      "step": 981
    },
    {
      "epoch": 3.0,
      "step": 981,
      "total_flos": 1642792778465280.0,
      "train_loss": 0.5776342995062272,
      "train_runtime": 29440.9564,
      "train_samples_per_second": 17.06,
      "train_steps_per_second": 0.033
    }
  ],
  "logging_steps": 10,
  "max_steps": 981,
  "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": 1642792778465280.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}