{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 969,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.030959752321981424,
      "grad_norm": 9.26792445696984,
      "learning_rate": 5e-06,
      "loss": 0.9296,
      "step": 10
    },
    {
      "epoch": 0.06191950464396285,
      "grad_norm": 1.9224039875777081,
      "learning_rate": 5e-06,
      "loss": 0.826,
      "step": 20
    },
    {
      "epoch": 0.09287925696594428,
      "grad_norm": 4.074240624021427,
      "learning_rate": 5e-06,
      "loss": 0.7818,
      "step": 30
    },
    {
      "epoch": 0.1238390092879257,
      "grad_norm": 1.5813292349658419,
      "learning_rate": 5e-06,
      "loss": 0.7624,
      "step": 40
    },
    {
      "epoch": 0.15479876160990713,
      "grad_norm": 0.8536213234663932,
      "learning_rate": 5e-06,
      "loss": 0.7313,
      "step": 50
    },
    {
      "epoch": 0.18575851393188855,
      "grad_norm": 1.087523888330213,
      "learning_rate": 5e-06,
      "loss": 0.7307,
      "step": 60
    },
    {
      "epoch": 0.21671826625386997,
      "grad_norm": 0.8453603539631231,
      "learning_rate": 5e-06,
      "loss": 0.7126,
      "step": 70
    },
    {
      "epoch": 0.2476780185758514,
      "grad_norm": 0.9824688751534248,
      "learning_rate": 5e-06,
      "loss": 0.7113,
      "step": 80
    },
    {
      "epoch": 0.2786377708978328,
      "grad_norm": 0.9326537359571924,
      "learning_rate": 5e-06,
      "loss": 0.7037,
      "step": 90
    },
    {
      "epoch": 0.30959752321981426,
      "grad_norm": 0.548855695013045,
      "learning_rate": 5e-06,
      "loss": 0.6988,
      "step": 100
    },
    {
      "epoch": 0.34055727554179566,
      "grad_norm": 0.5127345076323682,
      "learning_rate": 5e-06,
      "loss": 0.6969,
      "step": 110
    },
    {
      "epoch": 0.3715170278637771,
      "grad_norm": 0.5346034637159197,
      "learning_rate": 5e-06,
      "loss": 0.6932,
      "step": 120
    },
    {
      "epoch": 0.4024767801857585,
      "grad_norm": 0.6366003373202664,
      "learning_rate": 5e-06,
      "loss": 0.6862,
      "step": 130
    },
    {
      "epoch": 0.43343653250773995,
      "grad_norm": 0.659232587562787,
      "learning_rate": 5e-06,
      "loss": 0.6861,
      "step": 140
    },
    {
      "epoch": 0.46439628482972134,
      "grad_norm": 0.7626299453869156,
      "learning_rate": 5e-06,
      "loss": 0.6867,
      "step": 150
    },
    {
      "epoch": 0.4953560371517028,
      "grad_norm": 0.600108061207952,
      "learning_rate": 5e-06,
      "loss": 0.6761,
      "step": 160
    },
    {
      "epoch": 0.5263157894736842,
      "grad_norm": 0.6562920403508146,
      "learning_rate": 5e-06,
      "loss": 0.6822,
      "step": 170
    },
    {
      "epoch": 0.5572755417956656,
      "grad_norm": 0.554500335402875,
      "learning_rate": 5e-06,
      "loss": 0.6771,
      "step": 180
    },
    {
      "epoch": 0.5882352941176471,
      "grad_norm": 0.7448258243980852,
      "learning_rate": 5e-06,
      "loss": 0.6725,
      "step": 190
    },
    {
      "epoch": 0.6191950464396285,
      "grad_norm": 0.7212261495226227,
      "learning_rate": 5e-06,
      "loss": 0.6704,
      "step": 200
    },
    {
      "epoch": 0.6501547987616099,
      "grad_norm": 0.650918538222657,
      "learning_rate": 5e-06,
      "loss": 0.6747,
      "step": 210
    },
    {
      "epoch": 0.6811145510835913,
      "grad_norm": 0.5979672431899267,
      "learning_rate": 5e-06,
      "loss": 0.6721,
      "step": 220
    },
    {
      "epoch": 0.7120743034055728,
      "grad_norm": 0.5587950559181737,
      "learning_rate": 5e-06,
      "loss": 0.6775,
      "step": 230
    },
    {
      "epoch": 0.7430340557275542,
      "grad_norm": 0.7504748626653679,
      "learning_rate": 5e-06,
      "loss": 0.6772,
      "step": 240
    },
    {
      "epoch": 0.7739938080495357,
      "grad_norm": 0.5672124597726503,
      "learning_rate": 5e-06,
      "loss": 0.6649,
      "step": 250
    },
    {
      "epoch": 0.804953560371517,
      "grad_norm": 0.6667059518291187,
      "learning_rate": 5e-06,
      "loss": 0.6628,
      "step": 260
    },
    {
      "epoch": 0.8359133126934984,
      "grad_norm": 0.5557591111686754,
      "learning_rate": 5e-06,
      "loss": 0.666,
      "step": 270
    },
    {
      "epoch": 0.8668730650154799,
      "grad_norm": 0.7400262670967761,
      "learning_rate": 5e-06,
      "loss": 0.6608,
      "step": 280
    },
    {
      "epoch": 0.8978328173374613,
      "grad_norm": 0.5053622688793449,
      "learning_rate": 5e-06,
      "loss": 0.6657,
      "step": 290
    },
    {
      "epoch": 0.9287925696594427,
      "grad_norm": 0.611108366609724,
      "learning_rate": 5e-06,
      "loss": 0.6612,
      "step": 300
    },
    {
      "epoch": 0.9597523219814241,
      "grad_norm": 0.5223978557328981,
      "learning_rate": 5e-06,
      "loss": 0.6585,
      "step": 310
    },
    {
      "epoch": 0.9907120743034056,
      "grad_norm": 0.5084066168996301,
      "learning_rate": 5e-06,
      "loss": 0.6582,
      "step": 320
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.6614387631416321,
      "eval_runtime": 31.0197,
      "eval_samples_per_second": 280.015,
      "eval_steps_per_second": 1.096,
      "step": 323
    },
    {
      "epoch": 1.021671826625387,
      "grad_norm": 0.8185084818131129,
      "learning_rate": 5e-06,
      "loss": 0.6318,
      "step": 330
    },
    {
      "epoch": 1.0526315789473684,
      "grad_norm": 0.7247142810746562,
      "learning_rate": 5e-06,
      "loss": 0.6153,
      "step": 340
    },
    {
      "epoch": 1.08359133126935,
      "grad_norm": 0.574027398851242,
      "learning_rate": 5e-06,
      "loss": 0.6182,
      "step": 350
    },
    {
      "epoch": 1.1145510835913313,
      "grad_norm": 0.5449783832911268,
      "learning_rate": 5e-06,
      "loss": 0.6155,
      "step": 360
    },
    {
      "epoch": 1.1455108359133126,
      "grad_norm": 0.5025194777117987,
      "learning_rate": 5e-06,
      "loss": 0.6181,
      "step": 370
    },
    {
      "epoch": 1.1764705882352942,
      "grad_norm": 0.5884698444538693,
      "learning_rate": 5e-06,
      "loss": 0.6145,
      "step": 380
    },
    {
      "epoch": 1.2074303405572755,
      "grad_norm": 0.6402773293709152,
      "learning_rate": 5e-06,
      "loss": 0.6182,
      "step": 390
    },
    {
      "epoch": 1.238390092879257,
      "grad_norm": 0.7376501065513348,
      "learning_rate": 5e-06,
      "loss": 0.6078,
      "step": 400
    },
    {
      "epoch": 1.2693498452012384,
      "grad_norm": 0.5604731743792086,
      "learning_rate": 5e-06,
      "loss": 0.6178,
      "step": 410
    },
    {
      "epoch": 1.3003095975232197,
      "grad_norm": 0.5374880157799365,
      "learning_rate": 5e-06,
      "loss": 0.6118,
      "step": 420
    },
    {
      "epoch": 1.3312693498452013,
      "grad_norm": 0.5183594439426741,
      "learning_rate": 5e-06,
      "loss": 0.6152,
      "step": 430
    },
    {
      "epoch": 1.3622291021671826,
      "grad_norm": 0.7429617462855133,
      "learning_rate": 5e-06,
      "loss": 0.6189,
      "step": 440
    },
    {
      "epoch": 1.3931888544891642,
      "grad_norm": 0.5637315725076051,
      "learning_rate": 5e-06,
      "loss": 0.6143,
      "step": 450
    },
    {
      "epoch": 1.4241486068111455,
      "grad_norm": 0.47806274705798535,
      "learning_rate": 5e-06,
      "loss": 0.6184,
      "step": 460
    },
    {
      "epoch": 1.4551083591331269,
      "grad_norm": 0.7013859457502354,
      "learning_rate": 5e-06,
      "loss": 0.6146,
      "step": 470
    },
    {
      "epoch": 1.4860681114551084,
      "grad_norm": 0.7728057436850783,
      "learning_rate": 5e-06,
      "loss": 0.6235,
      "step": 480
    },
    {
      "epoch": 1.5170278637770898,
      "grad_norm": 0.5540033221172351,
      "learning_rate": 5e-06,
      "loss": 0.6146,
      "step": 490
    },
    {
      "epoch": 1.5479876160990713,
      "grad_norm": 0.4989101370327553,
      "learning_rate": 5e-06,
      "loss": 0.6109,
      "step": 500
    },
    {
      "epoch": 1.5789473684210527,
      "grad_norm": 0.6201415740790441,
      "learning_rate": 5e-06,
      "loss": 0.6127,
      "step": 510
    },
    {
      "epoch": 1.609907120743034,
      "grad_norm": 0.6182247247832351,
      "learning_rate": 5e-06,
      "loss": 0.6158,
      "step": 520
    },
    {
      "epoch": 1.6408668730650153,
      "grad_norm": 0.5273934388255005,
      "learning_rate": 5e-06,
      "loss": 0.6152,
      "step": 530
    },
    {
      "epoch": 1.671826625386997,
      "grad_norm": 0.5657886835824731,
      "learning_rate": 5e-06,
      "loss": 0.6135,
      "step": 540
    },
    {
      "epoch": 1.7027863777089784,
      "grad_norm": 0.6873412871431165,
      "learning_rate": 5e-06,
      "loss": 0.615,
      "step": 550
    },
    {
      "epoch": 1.7337461300309598,
      "grad_norm": 0.5877043924887657,
      "learning_rate": 5e-06,
      "loss": 0.6125,
      "step": 560
    },
    {
      "epoch": 1.7647058823529411,
      "grad_norm": 0.6037122620569243,
      "learning_rate": 5e-06,
      "loss": 0.6118,
      "step": 570
    },
    {
      "epoch": 1.7956656346749225,
      "grad_norm": 0.4753962140527953,
      "learning_rate": 5e-06,
      "loss": 0.6128,
      "step": 580
    },
    {
      "epoch": 1.826625386996904,
      "grad_norm": 0.5268653628592879,
      "learning_rate": 5e-06,
      "loss": 0.6172,
      "step": 590
    },
    {
      "epoch": 1.8575851393188856,
      "grad_norm": 1.002687895644436,
      "learning_rate": 5e-06,
      "loss": 0.615,
      "step": 600
    },
    {
      "epoch": 1.888544891640867,
      "grad_norm": 0.5100009376935558,
      "learning_rate": 5e-06,
      "loss": 0.6182,
      "step": 610
    },
    {
      "epoch": 1.9195046439628483,
      "grad_norm": 0.691746404073842,
      "learning_rate": 5e-06,
      "loss": 0.6131,
      "step": 620
    },
    {
      "epoch": 1.9504643962848296,
      "grad_norm": 0.5303891186235742,
      "learning_rate": 5e-06,
      "loss": 0.6152,
      "step": 630
    },
    {
      "epoch": 1.9814241486068112,
      "grad_norm": 0.5949259061449034,
      "learning_rate": 5e-06,
      "loss": 0.6073,
      "step": 640
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.6524380445480347,
      "eval_runtime": 31.1666,
      "eval_samples_per_second": 278.696,
      "eval_steps_per_second": 1.091,
      "step": 646
    },
    {
      "epoch": 2.0123839009287927,
      "grad_norm": 0.9446095499864162,
      "learning_rate": 5e-06,
      "loss": 0.591,
      "step": 650
    },
    {
      "epoch": 2.043343653250774,
      "grad_norm": 0.6258477650873515,
      "learning_rate": 5e-06,
      "loss": 0.5646,
      "step": 660
    },
    {
      "epoch": 2.0743034055727554,
      "grad_norm": 0.604477009985834,
      "learning_rate": 5e-06,
      "loss": 0.5693,
      "step": 670
    },
    {
      "epoch": 2.1052631578947367,
      "grad_norm": 0.5940857975430127,
      "learning_rate": 5e-06,
      "loss": 0.5621,
      "step": 680
    },
    {
      "epoch": 2.136222910216718,
      "grad_norm": 0.6310773073567004,
      "learning_rate": 5e-06,
      "loss": 0.5664,
      "step": 690
    },
    {
      "epoch": 2.1671826625387,
      "grad_norm": 0.6735784589983905,
      "learning_rate": 5e-06,
      "loss": 0.5659,
      "step": 700
    },
    {
      "epoch": 2.198142414860681,
      "grad_norm": 0.6737551186617048,
      "learning_rate": 5e-06,
      "loss": 0.5744,
      "step": 710
    },
    {
      "epoch": 2.2291021671826625,
      "grad_norm": 0.538363932283787,
      "learning_rate": 5e-06,
      "loss": 0.5634,
      "step": 720
    },
    {
      "epoch": 2.260061919504644,
      "grad_norm": 0.6696764784705668,
      "learning_rate": 5e-06,
      "loss": 0.5633,
      "step": 730
    },
    {
      "epoch": 2.291021671826625,
      "grad_norm": 0.5541664073886028,
      "learning_rate": 5e-06,
      "loss": 0.572,
      "step": 740
    },
    {
      "epoch": 2.321981424148607,
      "grad_norm": 0.7344897597565605,
      "learning_rate": 5e-06,
      "loss": 0.5686,
      "step": 750
    },
    {
      "epoch": 2.3529411764705883,
      "grad_norm": 0.5507126279939225,
      "learning_rate": 5e-06,
      "loss": 0.5729,
      "step": 760
    },
    {
      "epoch": 2.3839009287925697,
      "grad_norm": 0.6344649027391833,
      "learning_rate": 5e-06,
      "loss": 0.5665,
      "step": 770
    },
    {
      "epoch": 2.414860681114551,
      "grad_norm": 0.6106920196828644,
      "learning_rate": 5e-06,
      "loss": 0.5702,
      "step": 780
    },
    {
      "epoch": 2.4458204334365323,
      "grad_norm": 0.6238727356689014,
      "learning_rate": 5e-06,
      "loss": 0.5706,
      "step": 790
    },
    {
      "epoch": 2.476780185758514,
      "grad_norm": 0.6326270787372308,
      "learning_rate": 5e-06,
      "loss": 0.5739,
      "step": 800
    },
    {
      "epoch": 2.5077399380804954,
      "grad_norm": 0.6392551743906308,
      "learning_rate": 5e-06,
      "loss": 0.5707,
      "step": 810
    },
    {
      "epoch": 2.538699690402477,
      "grad_norm": 0.5293919880888138,
      "learning_rate": 5e-06,
      "loss": 0.5694,
      "step": 820
    },
    {
      "epoch": 2.569659442724458,
      "grad_norm": 0.5315004573871298,
      "learning_rate": 5e-06,
      "loss": 0.5715,
      "step": 830
    },
    {
      "epoch": 2.6006191950464395,
      "grad_norm": 0.6021834459451176,
      "learning_rate": 5e-06,
      "loss": 0.5686,
      "step": 840
    },
    {
      "epoch": 2.6315789473684212,
      "grad_norm": 0.7264827387006807,
      "learning_rate": 5e-06,
      "loss": 0.5711,
      "step": 850
    },
    {
      "epoch": 2.6625386996904026,
      "grad_norm": 0.5432440378468479,
      "learning_rate": 5e-06,
      "loss": 0.5682,
      "step": 860
    },
    {
      "epoch": 2.693498452012384,
      "grad_norm": 0.4887763215481647,
      "learning_rate": 5e-06,
      "loss": 0.5727,
      "step": 870
    },
    {
      "epoch": 2.7244582043343653,
      "grad_norm": 0.6219072904402111,
      "learning_rate": 5e-06,
      "loss": 0.5676,
      "step": 880
    },
    {
      "epoch": 2.7554179566563466,
      "grad_norm": 0.5596606256063349,
      "learning_rate": 5e-06,
      "loss": 0.5748,
      "step": 890
    },
    {
      "epoch": 2.7863777089783284,
      "grad_norm": 0.6091604993895532,
      "learning_rate": 5e-06,
      "loss": 0.5639,
      "step": 900
    },
    {
      "epoch": 2.8173374613003097,
      "grad_norm": 0.5920477606163769,
      "learning_rate": 5e-06,
      "loss": 0.5717,
      "step": 910
    },
    {
      "epoch": 2.848297213622291,
      "grad_norm": 0.6807900156580523,
      "learning_rate": 5e-06,
      "loss": 0.5721,
      "step": 920
    },
    {
      "epoch": 2.8792569659442724,
      "grad_norm": 0.6050955818661157,
      "learning_rate": 5e-06,
      "loss": 0.5748,
      "step": 930
    },
    {
      "epoch": 2.9102167182662537,
      "grad_norm": 0.6165235050612774,
      "learning_rate": 5e-06,
      "loss": 0.5732,
      "step": 940
    },
    {
      "epoch": 2.9411764705882355,
      "grad_norm": 0.6346237356986199,
      "learning_rate": 5e-06,
      "loss": 0.5743,
      "step": 950
    },
    {
      "epoch": 2.972136222910217,
      "grad_norm": 0.6125475070291018,
      "learning_rate": 5e-06,
      "loss": 0.5733,
      "step": 960
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.6580318212509155,
      "eval_runtime": 31.0741,
      "eval_samples_per_second": 279.526,
      "eval_steps_per_second": 1.094,
      "step": 969
    },
    {
      "epoch": 3.0,
      "step": 969,
      "total_flos": 1623111090831360.0,
      "train_loss": 0.628350810984955,
      "train_runtime": 6073.5579,
      "train_samples_per_second": 81.516,
      "train_steps_per_second": 0.16
    }
  ],
  "logging_steps": 10,
  "max_steps": 969,
  "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": 1623111090831360.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}