{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.998438719750195,
  "eval_steps": 500,
  "global_step": 960,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0312256049960968,
      "grad_norm": 2.3905091254417177,
      "learning_rate": 5e-06,
      "loss": 0.8071,
      "step": 10
    },
    {
      "epoch": 0.0624512099921936,
      "grad_norm": 1.3958881588356253,
      "learning_rate": 5e-06,
      "loss": 0.7197,
      "step": 20
    },
    {
      "epoch": 0.0936768149882904,
      "grad_norm": 1.5094847169199592,
      "learning_rate": 5e-06,
      "loss": 0.6985,
      "step": 30
    },
    {
      "epoch": 0.1249024199843872,
      "grad_norm": 1.4080817727177422,
      "learning_rate": 5e-06,
      "loss": 0.6855,
      "step": 40
    },
    {
      "epoch": 0.156128024980484,
      "grad_norm": 0.8437914162126238,
      "learning_rate": 5e-06,
      "loss": 0.6727,
      "step": 50
    },
    {
      "epoch": 0.1873536299765808,
      "grad_norm": 0.8642540743285999,
      "learning_rate": 5e-06,
      "loss": 0.6576,
      "step": 60
    },
    {
      "epoch": 0.2185792349726776,
      "grad_norm": 0.6633530231329323,
      "learning_rate": 5e-06,
      "loss": 0.6514,
      "step": 70
    },
    {
      "epoch": 0.2498048399687744,
      "grad_norm": 0.8262205915204606,
      "learning_rate": 5e-06,
      "loss": 0.6457,
      "step": 80
    },
    {
      "epoch": 0.2810304449648712,
      "grad_norm": 0.5613248571253571,
      "learning_rate": 5e-06,
      "loss": 0.6412,
      "step": 90
    },
    {
      "epoch": 0.312256049960968,
      "grad_norm": 0.9150986102724331,
      "learning_rate": 5e-06,
      "loss": 0.6384,
      "step": 100
    },
    {
      "epoch": 0.3434816549570648,
      "grad_norm": 0.7465781739155327,
      "learning_rate": 5e-06,
      "loss": 0.6335,
      "step": 110
    },
    {
      "epoch": 0.3747072599531616,
      "grad_norm": 0.8250370829383081,
      "learning_rate": 5e-06,
      "loss": 0.6302,
      "step": 120
    },
    {
      "epoch": 0.4059328649492584,
      "grad_norm": 0.6341139387710243,
      "learning_rate": 5e-06,
      "loss": 0.634,
      "step": 130
    },
    {
      "epoch": 0.4371584699453552,
      "grad_norm": 0.4964005156113376,
      "learning_rate": 5e-06,
      "loss": 0.6211,
      "step": 140
    },
    {
      "epoch": 0.468384074941452,
      "grad_norm": 0.9475290669111363,
      "learning_rate": 5e-06,
      "loss": 0.6271,
      "step": 150
    },
    {
      "epoch": 0.4996096799375488,
      "grad_norm": 0.7811648794629471,
      "learning_rate": 5e-06,
      "loss": 0.6238,
      "step": 160
    },
    {
      "epoch": 0.5308352849336456,
      "grad_norm": 0.516293302775752,
      "learning_rate": 5e-06,
      "loss": 0.621,
      "step": 170
    },
    {
      "epoch": 0.5620608899297423,
      "grad_norm": 0.471912073011228,
      "learning_rate": 5e-06,
      "loss": 0.6215,
      "step": 180
    },
    {
      "epoch": 0.5932864949258392,
      "grad_norm": 0.5719925024660438,
      "learning_rate": 5e-06,
      "loss": 0.6193,
      "step": 190
    },
    {
      "epoch": 0.624512099921936,
      "grad_norm": 0.5059415320269443,
      "learning_rate": 5e-06,
      "loss": 0.6175,
      "step": 200
    },
    {
      "epoch": 0.6557377049180327,
      "grad_norm": 0.5195408058121892,
      "learning_rate": 5e-06,
      "loss": 0.622,
      "step": 210
    },
    {
      "epoch": 0.6869633099141296,
      "grad_norm": 0.6014889400609209,
      "learning_rate": 5e-06,
      "loss": 0.6158,
      "step": 220
    },
    {
      "epoch": 0.7181889149102264,
      "grad_norm": 0.5096070261428851,
      "learning_rate": 5e-06,
      "loss": 0.6177,
      "step": 230
    },
    {
      "epoch": 0.7494145199063232,
      "grad_norm": 0.6028623695390841,
      "learning_rate": 5e-06,
      "loss": 0.6148,
      "step": 240
    },
    {
      "epoch": 0.78064012490242,
      "grad_norm": 0.5451038071079088,
      "learning_rate": 5e-06,
      "loss": 0.6164,
      "step": 250
    },
    {
      "epoch": 0.8118657298985168,
      "grad_norm": 0.4708236433706893,
      "learning_rate": 5e-06,
      "loss": 0.6094,
      "step": 260
    },
    {
      "epoch": 0.8430913348946136,
      "grad_norm": 0.46109612782168113,
      "learning_rate": 5e-06,
      "loss": 0.6103,
      "step": 270
    },
    {
      "epoch": 0.8743169398907104,
      "grad_norm": 0.502648205452055,
      "learning_rate": 5e-06,
      "loss": 0.6095,
      "step": 280
    },
    {
      "epoch": 0.9055425448868072,
      "grad_norm": 0.4489395079927774,
      "learning_rate": 5e-06,
      "loss": 0.6065,
      "step": 290
    },
    {
      "epoch": 0.936768149882904,
      "grad_norm": 0.6477411963812875,
      "learning_rate": 5e-06,
      "loss": 0.6132,
      "step": 300
    },
    {
      "epoch": 0.9679937548790007,
      "grad_norm": 0.5302907770786253,
      "learning_rate": 5e-06,
      "loss": 0.6111,
      "step": 310
    },
    {
      "epoch": 0.9992193598750976,
      "grad_norm": 0.45517099030938496,
      "learning_rate": 5e-06,
      "loss": 0.5986,
      "step": 320
    },
    {
      "epoch": 0.9992193598750976,
      "eval_loss": 0.6128131151199341,
      "eval_runtime": 341.0462,
      "eval_samples_per_second": 25.304,
      "eval_steps_per_second": 0.396,
      "step": 320
    },
    {
      "epoch": 1.0308352849336455,
      "grad_norm": 0.8438882865201324,
      "learning_rate": 5e-06,
      "loss": 0.6164,
      "step": 330
    },
    {
      "epoch": 1.0620608899297423,
      "grad_norm": 0.4821646175445636,
      "learning_rate": 5e-06,
      "loss": 0.5662,
      "step": 340
    },
    {
      "epoch": 1.0932864949258392,
      "grad_norm": 0.5058980149763423,
      "learning_rate": 5e-06,
      "loss": 0.5606,
      "step": 350
    },
    {
      "epoch": 1.124512099921936,
      "grad_norm": 0.4781831185352073,
      "learning_rate": 5e-06,
      "loss": 0.5648,
      "step": 360
    },
    {
      "epoch": 1.1557377049180328,
      "grad_norm": 0.5048559047058323,
      "learning_rate": 5e-06,
      "loss": 0.5582,
      "step": 370
    },
    {
      "epoch": 1.1869633099141297,
      "grad_norm": 0.44414824193518654,
      "learning_rate": 5e-06,
      "loss": 0.5584,
      "step": 380
    },
    {
      "epoch": 1.2181889149102263,
      "grad_norm": 0.4928423351798681,
      "learning_rate": 5e-06,
      "loss": 0.5515,
      "step": 390
    },
    {
      "epoch": 1.2494145199063231,
      "grad_norm": 0.5064189451582637,
      "learning_rate": 5e-06,
      "loss": 0.5637,
      "step": 400
    },
    {
      "epoch": 1.28064012490242,
      "grad_norm": 0.44193713470343654,
      "learning_rate": 5e-06,
      "loss": 0.5618,
      "step": 410
    },
    {
      "epoch": 1.3118657298985168,
      "grad_norm": 0.4650381211562015,
      "learning_rate": 5e-06,
      "loss": 0.5554,
      "step": 420
    },
    {
      "epoch": 1.3430913348946136,
      "grad_norm": 0.5544428065241478,
      "learning_rate": 5e-06,
      "loss": 0.5547,
      "step": 430
    },
    {
      "epoch": 1.3743169398907105,
      "grad_norm": 0.48005595474790913,
      "learning_rate": 5e-06,
      "loss": 0.5523,
      "step": 440
    },
    {
      "epoch": 1.4055425448868073,
      "grad_norm": 0.4974548951913249,
      "learning_rate": 5e-06,
      "loss": 0.5666,
      "step": 450
    },
    {
      "epoch": 1.436768149882904,
      "grad_norm": 0.4923658625750441,
      "learning_rate": 5e-06,
      "loss": 0.5558,
      "step": 460
    },
    {
      "epoch": 1.4679937548790007,
      "grad_norm": 0.5272663506589431,
      "learning_rate": 5e-06,
      "loss": 0.5584,
      "step": 470
    },
    {
      "epoch": 1.4992193598750976,
      "grad_norm": 0.5304464959914178,
      "learning_rate": 5e-06,
      "loss": 0.5643,
      "step": 480
    },
    {
      "epoch": 1.5304449648711944,
      "grad_norm": 0.5773543616559265,
      "learning_rate": 5e-06,
      "loss": 0.5598,
      "step": 490
    },
    {
      "epoch": 1.561670569867291,
      "grad_norm": 0.4558348320273449,
      "learning_rate": 5e-06,
      "loss": 0.5591,
      "step": 500
    },
    {
      "epoch": 1.5928961748633879,
      "grad_norm": 0.5072303901122793,
      "learning_rate": 5e-06,
      "loss": 0.5626,
      "step": 510
    },
    {
      "epoch": 1.6241217798594847,
      "grad_norm": 0.5369887998410667,
      "learning_rate": 5e-06,
      "loss": 0.5556,
      "step": 520
    },
    {
      "epoch": 1.6553473848555815,
      "grad_norm": 0.5556757682627291,
      "learning_rate": 5e-06,
      "loss": 0.5572,
      "step": 530
    },
    {
      "epoch": 1.6865729898516784,
      "grad_norm": 0.5337242705677901,
      "learning_rate": 5e-06,
      "loss": 0.557,
      "step": 540
    },
    {
      "epoch": 1.7177985948477752,
      "grad_norm": 0.46280527938706506,
      "learning_rate": 5e-06,
      "loss": 0.5617,
      "step": 550
    },
    {
      "epoch": 1.749024199843872,
      "grad_norm": 0.45608832514525505,
      "learning_rate": 5e-06,
      "loss": 0.5581,
      "step": 560
    },
    {
      "epoch": 1.7802498048399689,
      "grad_norm": 0.48374355780746187,
      "learning_rate": 5e-06,
      "loss": 0.5564,
      "step": 570
    },
    {
      "epoch": 1.8114754098360657,
      "grad_norm": 0.5029705354009028,
      "learning_rate": 5e-06,
      "loss": 0.559,
      "step": 580
    },
    {
      "epoch": 1.8427010148321625,
      "grad_norm": 0.46966476792976214,
      "learning_rate": 5e-06,
      "loss": 0.5616,
      "step": 590
    },
    {
      "epoch": 1.8739266198282591,
      "grad_norm": 0.446283124549817,
      "learning_rate": 5e-06,
      "loss": 0.553,
      "step": 600
    },
    {
      "epoch": 1.905152224824356,
      "grad_norm": 0.4745527474098281,
      "learning_rate": 5e-06,
      "loss": 0.5589,
      "step": 610
    },
    {
      "epoch": 1.9363778298204528,
      "grad_norm": 0.501609279464785,
      "learning_rate": 5e-06,
      "loss": 0.5628,
      "step": 620
    },
    {
      "epoch": 1.9676034348165494,
      "grad_norm": 0.49320626859834116,
      "learning_rate": 5e-06,
      "loss": 0.5522,
      "step": 630
    },
    {
      "epoch": 1.9988290398126463,
      "grad_norm": 0.4324557011242181,
      "learning_rate": 5e-06,
      "loss": 0.5596,
      "step": 640
    },
    {
      "epoch": 1.9988290398126463,
      "eval_loss": 0.6045193076133728,
      "eval_runtime": 340.5129,
      "eval_samples_per_second": 25.344,
      "eval_steps_per_second": 0.396,
      "step": 640
    },
    {
      "epoch": 2.030444964871194,
      "grad_norm": 0.6006527642113036,
      "learning_rate": 5e-06,
      "loss": 0.5662,
      "step": 650
    },
    {
      "epoch": 2.061670569867291,
      "grad_norm": 0.5376147888211947,
      "learning_rate": 5e-06,
      "loss": 0.5018,
      "step": 660
    },
    {
      "epoch": 2.092896174863388,
      "grad_norm": 0.5448017881956769,
      "learning_rate": 5e-06,
      "loss": 0.5057,
      "step": 670
    },
    {
      "epoch": 2.1241217798594847,
      "grad_norm": 0.6095347029172922,
      "learning_rate": 5e-06,
      "loss": 0.5036,
      "step": 680
    },
    {
      "epoch": 2.1553473848555815,
      "grad_norm": 0.5281790301882382,
      "learning_rate": 5e-06,
      "loss": 0.5066,
      "step": 690
    },
    {
      "epoch": 2.1865729898516784,
      "grad_norm": 0.543025537124188,
      "learning_rate": 5e-06,
      "loss": 0.5086,
      "step": 700
    },
    {
      "epoch": 2.217798594847775,
      "grad_norm": 0.6024294613229594,
      "learning_rate": 5e-06,
      "loss": 0.508,
      "step": 710
    },
    {
      "epoch": 2.249024199843872,
      "grad_norm": 0.5261160691218546,
      "learning_rate": 5e-06,
      "loss": 0.5073,
      "step": 720
    },
    {
      "epoch": 2.280249804839969,
      "grad_norm": 0.4878879224650377,
      "learning_rate": 5e-06,
      "loss": 0.5126,
      "step": 730
    },
    {
      "epoch": 2.3114754098360657,
      "grad_norm": 0.5298908191049263,
      "learning_rate": 5e-06,
      "loss": 0.5098,
      "step": 740
    },
    {
      "epoch": 2.3427010148321625,
      "grad_norm": 0.4963375261761113,
      "learning_rate": 5e-06,
      "loss": 0.5084,
      "step": 750
    },
    {
      "epoch": 2.3739266198282594,
      "grad_norm": 0.476625155447844,
      "learning_rate": 5e-06,
      "loss": 0.5097,
      "step": 760
    },
    {
      "epoch": 2.4051522248243558,
      "grad_norm": 0.5208071390082176,
      "learning_rate": 5e-06,
      "loss": 0.5028,
      "step": 770
    },
    {
      "epoch": 2.4363778298204526,
      "grad_norm": 0.4800697229604007,
      "learning_rate": 5e-06,
      "loss": 0.5102,
      "step": 780
    },
    {
      "epoch": 2.4676034348165494,
      "grad_norm": 0.5837948115948769,
      "learning_rate": 5e-06,
      "loss": 0.5068,
      "step": 790
    },
    {
      "epoch": 2.4988290398126463,
      "grad_norm": 0.5280421758640109,
      "learning_rate": 5e-06,
      "loss": 0.5141,
      "step": 800
    },
    {
      "epoch": 2.530054644808743,
      "grad_norm": 0.4594714496886714,
      "learning_rate": 5e-06,
      "loss": 0.5081,
      "step": 810
    },
    {
      "epoch": 2.56128024980484,
      "grad_norm": 0.51076427145537,
      "learning_rate": 5e-06,
      "loss": 0.5123,
      "step": 820
    },
    {
      "epoch": 2.5925058548009368,
      "grad_norm": 0.5309550406289082,
      "learning_rate": 5e-06,
      "loss": 0.5067,
      "step": 830
    },
    {
      "epoch": 2.6237314597970336,
      "grad_norm": 0.6033885635557941,
      "learning_rate": 5e-06,
      "loss": 0.519,
      "step": 840
    },
    {
      "epoch": 2.6549570647931304,
      "grad_norm": 0.5271243446375676,
      "learning_rate": 5e-06,
      "loss": 0.5048,
      "step": 850
    },
    {
      "epoch": 2.6861826697892273,
      "grad_norm": 0.5270298610894952,
      "learning_rate": 5e-06,
      "loss": 0.5165,
      "step": 860
    },
    {
      "epoch": 2.717408274785324,
      "grad_norm": 0.5222507557357616,
      "learning_rate": 5e-06,
      "loss": 0.5154,
      "step": 870
    },
    {
      "epoch": 2.748633879781421,
      "grad_norm": 0.49639435328740067,
      "learning_rate": 5e-06,
      "loss": 0.5091,
      "step": 880
    },
    {
      "epoch": 2.7798594847775178,
      "grad_norm": 0.4828475074772525,
      "learning_rate": 5e-06,
      "loss": 0.5112,
      "step": 890
    },
    {
      "epoch": 2.8110850897736146,
      "grad_norm": 0.5145958529566682,
      "learning_rate": 5e-06,
      "loss": 0.5109,
      "step": 900
    },
    {
      "epoch": 2.8423106947697114,
      "grad_norm": 0.528104324477883,
      "learning_rate": 5e-06,
      "loss": 0.5166,
      "step": 910
    },
    {
      "epoch": 2.873536299765808,
      "grad_norm": 0.47920251362694366,
      "learning_rate": 5e-06,
      "loss": 0.519,
      "step": 920
    },
    {
      "epoch": 2.9047619047619047,
      "grad_norm": 0.538906478147928,
      "learning_rate": 5e-06,
      "loss": 0.5181,
      "step": 930
    },
    {
      "epoch": 2.9359875097580015,
      "grad_norm": 0.5243949264804789,
      "learning_rate": 5e-06,
      "loss": 0.5134,
      "step": 940
    },
    {
      "epoch": 2.9672131147540983,
      "grad_norm": 0.47727618067883554,
      "learning_rate": 5e-06,
      "loss": 0.512,
      "step": 950
    },
    {
      "epoch": 2.998438719750195,
      "grad_norm": 0.5110204465597075,
      "learning_rate": 5e-06,
      "loss": 0.5083,
      "step": 960
    },
    {
      "epoch": 2.998438719750195,
      "eval_loss": 0.608026921749115,
      "eval_runtime": 339.6647,
      "eval_samples_per_second": 25.407,
      "eval_steps_per_second": 0.397,
      "step": 960
    },
    {
      "epoch": 2.998438719750195,
      "step": 960,
      "total_flos": 1607826375966720.0,
      "train_loss": 0.5702028140425682,
      "train_runtime": 56737.9664,
      "train_samples_per_second": 8.669,
      "train_steps_per_second": 0.017
    }
  ],
  "logging_steps": 10,
  "max_steps": 960,
  "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": 1607826375966720.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}