{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 987,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.030395136778115502,
      "grad_norm": 16.64022370418662,
      "learning_rate": 5e-06,
      "loss": 0.9277,
      "step": 10
    },
    {
      "epoch": 0.060790273556231005,
      "grad_norm": 1.5999241644765967,
      "learning_rate": 5e-06,
      "loss": 0.825,
      "step": 20
    },
    {
      "epoch": 0.0911854103343465,
      "grad_norm": 0.9838744519445904,
      "learning_rate": 5e-06,
      "loss": 0.7817,
      "step": 30
    },
    {
      "epoch": 0.12158054711246201,
      "grad_norm": 0.876398417764137,
      "learning_rate": 5e-06,
      "loss": 0.7545,
      "step": 40
    },
    {
      "epoch": 0.1519756838905775,
      "grad_norm": 0.9156799635831374,
      "learning_rate": 5e-06,
      "loss": 0.7358,
      "step": 50
    },
    {
      "epoch": 0.182370820668693,
      "grad_norm": 0.808319513356986,
      "learning_rate": 5e-06,
      "loss": 0.7226,
      "step": 60
    },
    {
      "epoch": 0.2127659574468085,
      "grad_norm": 0.8914678061831848,
      "learning_rate": 5e-06,
      "loss": 0.7122,
      "step": 70
    },
    {
      "epoch": 0.24316109422492402,
      "grad_norm": 0.6828512024884075,
      "learning_rate": 5e-06,
      "loss": 0.7016,
      "step": 80
    },
    {
      "epoch": 0.2735562310030395,
      "grad_norm": 0.6845268978851149,
      "learning_rate": 5e-06,
      "loss": 0.6982,
      "step": 90
    },
    {
      "epoch": 0.303951367781155,
      "grad_norm": 0.7312914857408909,
      "learning_rate": 5e-06,
      "loss": 0.6865,
      "step": 100
    },
    {
      "epoch": 0.3343465045592705,
      "grad_norm": 0.7352540765774106,
      "learning_rate": 5e-06,
      "loss": 0.6907,
      "step": 110
    },
    {
      "epoch": 0.364741641337386,
      "grad_norm": 0.7102916388595696,
      "learning_rate": 5e-06,
      "loss": 0.6824,
      "step": 120
    },
    {
      "epoch": 0.3951367781155015,
      "grad_norm": 0.751885584367901,
      "learning_rate": 5e-06,
      "loss": 0.6879,
      "step": 130
    },
    {
      "epoch": 0.425531914893617,
      "grad_norm": 0.589815976371268,
      "learning_rate": 5e-06,
      "loss": 0.6824,
      "step": 140
    },
    {
      "epoch": 0.45592705167173253,
      "grad_norm": 0.5807246304943029,
      "learning_rate": 5e-06,
      "loss": 0.6785,
      "step": 150
    },
    {
      "epoch": 0.48632218844984804,
      "grad_norm": 0.594562932961839,
      "learning_rate": 5e-06,
      "loss": 0.6857,
      "step": 160
    },
    {
      "epoch": 0.5167173252279635,
      "grad_norm": 0.5466511252566243,
      "learning_rate": 5e-06,
      "loss": 0.6769,
      "step": 170
    },
    {
      "epoch": 0.547112462006079,
      "grad_norm": 0.5857763993093636,
      "learning_rate": 5e-06,
      "loss": 0.6853,
      "step": 180
    },
    {
      "epoch": 0.5775075987841946,
      "grad_norm": 0.564419379546199,
      "learning_rate": 5e-06,
      "loss": 0.6716,
      "step": 190
    },
    {
      "epoch": 0.60790273556231,
      "grad_norm": 0.5133760076952175,
      "learning_rate": 5e-06,
      "loss": 0.6726,
      "step": 200
    },
    {
      "epoch": 0.6382978723404256,
      "grad_norm": 0.7195067283606177,
      "learning_rate": 5e-06,
      "loss": 0.6747,
      "step": 210
    },
    {
      "epoch": 0.668693009118541,
      "grad_norm": 0.6535399849598605,
      "learning_rate": 5e-06,
      "loss": 0.673,
      "step": 220
    },
    {
      "epoch": 0.6990881458966566,
      "grad_norm": 0.5983112715015926,
      "learning_rate": 5e-06,
      "loss": 0.6701,
      "step": 230
    },
    {
      "epoch": 0.729483282674772,
      "grad_norm": 0.6315519097135667,
      "learning_rate": 5e-06,
      "loss": 0.6699,
      "step": 240
    },
    {
      "epoch": 0.7598784194528876,
      "grad_norm": 0.752213223333162,
      "learning_rate": 5e-06,
      "loss": 0.6649,
      "step": 250
    },
    {
      "epoch": 0.790273556231003,
      "grad_norm": 0.5498218531784089,
      "learning_rate": 5e-06,
      "loss": 0.6697,
      "step": 260
    },
    {
      "epoch": 0.8206686930091185,
      "grad_norm": 0.8862901715920243,
      "learning_rate": 5e-06,
      "loss": 0.6639,
      "step": 270
    },
    {
      "epoch": 0.851063829787234,
      "grad_norm": 0.5808746614523029,
      "learning_rate": 5e-06,
      "loss": 0.6624,
      "step": 280
    },
    {
      "epoch": 0.8814589665653495,
      "grad_norm": 0.5459654532749203,
      "learning_rate": 5e-06,
      "loss": 0.6606,
      "step": 290
    },
    {
      "epoch": 0.9118541033434651,
      "grad_norm": 0.5838459305444006,
      "learning_rate": 5e-06,
      "loss": 0.6632,
      "step": 300
    },
    {
      "epoch": 0.9422492401215805,
      "grad_norm": 0.7036382282933219,
      "learning_rate": 5e-06,
      "loss": 0.6576,
      "step": 310
    },
    {
      "epoch": 0.9726443768996961,
      "grad_norm": 0.5259058744593071,
      "learning_rate": 5e-06,
      "loss": 0.6591,
      "step": 320
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.6591953635215759,
      "eval_runtime": 31.7982,
      "eval_samples_per_second": 278.664,
      "eval_steps_per_second": 0.566,
      "step": 329
    },
    {
      "epoch": 1.0030395136778116,
      "grad_norm": 0.8394036975692434,
      "learning_rate": 5e-06,
      "loss": 0.6601,
      "step": 330
    },
    {
      "epoch": 1.033434650455927,
      "grad_norm": 0.8451859406825153,
      "learning_rate": 5e-06,
      "loss": 0.6224,
      "step": 340
    },
    {
      "epoch": 1.0638297872340425,
      "grad_norm": 0.5812996686128133,
      "learning_rate": 5e-06,
      "loss": 0.615,
      "step": 350
    },
    {
      "epoch": 1.094224924012158,
      "grad_norm": 0.5503154879132033,
      "learning_rate": 5e-06,
      "loss": 0.6236,
      "step": 360
    },
    {
      "epoch": 1.1246200607902737,
      "grad_norm": 0.7833862166535409,
      "learning_rate": 5e-06,
      "loss": 0.6177,
      "step": 370
    },
    {
      "epoch": 1.155015197568389,
      "grad_norm": 0.6294679579661799,
      "learning_rate": 5e-06,
      "loss": 0.6186,
      "step": 380
    },
    {
      "epoch": 1.1854103343465046,
      "grad_norm": 0.5626057919504072,
      "learning_rate": 5e-06,
      "loss": 0.6131,
      "step": 390
    },
    {
      "epoch": 1.21580547112462,
      "grad_norm": 0.5655023836689654,
      "learning_rate": 5e-06,
      "loss": 0.6133,
      "step": 400
    },
    {
      "epoch": 1.2462006079027357,
      "grad_norm": 0.6419841359149339,
      "learning_rate": 5e-06,
      "loss": 0.6117,
      "step": 410
    },
    {
      "epoch": 1.2765957446808511,
      "grad_norm": 0.9501856765863967,
      "learning_rate": 5e-06,
      "loss": 0.6154,
      "step": 420
    },
    {
      "epoch": 1.3069908814589666,
      "grad_norm": 0.598897751556055,
      "learning_rate": 5e-06,
      "loss": 0.6162,
      "step": 430
    },
    {
      "epoch": 1.337386018237082,
      "grad_norm": 0.7429134385869901,
      "learning_rate": 5e-06,
      "loss": 0.6167,
      "step": 440
    },
    {
      "epoch": 1.3677811550151975,
      "grad_norm": 0.5294832137073032,
      "learning_rate": 5e-06,
      "loss": 0.6169,
      "step": 450
    },
    {
      "epoch": 1.3981762917933132,
      "grad_norm": 0.8585569002266061,
      "learning_rate": 5e-06,
      "loss": 0.6134,
      "step": 460
    },
    {
      "epoch": 1.4285714285714286,
      "grad_norm": 0.5980490230617989,
      "learning_rate": 5e-06,
      "loss": 0.6148,
      "step": 470
    },
    {
      "epoch": 1.458966565349544,
      "grad_norm": 0.5763371418857346,
      "learning_rate": 5e-06,
      "loss": 0.6173,
      "step": 480
    },
    {
      "epoch": 1.4893617021276595,
      "grad_norm": 0.5080499060023709,
      "learning_rate": 5e-06,
      "loss": 0.6139,
      "step": 490
    },
    {
      "epoch": 1.5197568389057752,
      "grad_norm": 0.599791892872901,
      "learning_rate": 5e-06,
      "loss": 0.62,
      "step": 500
    },
    {
      "epoch": 1.5501519756838906,
      "grad_norm": 0.561193296506177,
      "learning_rate": 5e-06,
      "loss": 0.6151,
      "step": 510
    },
    {
      "epoch": 1.580547112462006,
      "grad_norm": 0.55216025845368,
      "learning_rate": 5e-06,
      "loss": 0.6142,
      "step": 520
    },
    {
      "epoch": 1.6109422492401215,
      "grad_norm": 0.5020053919351462,
      "learning_rate": 5e-06,
      "loss": 0.6138,
      "step": 530
    },
    {
      "epoch": 1.641337386018237,
      "grad_norm": 0.804833961940261,
      "learning_rate": 5e-06,
      "loss": 0.6065,
      "step": 540
    },
    {
      "epoch": 1.6717325227963524,
      "grad_norm": 0.5098656593355803,
      "learning_rate": 5e-06,
      "loss": 0.6128,
      "step": 550
    },
    {
      "epoch": 1.702127659574468,
      "grad_norm": 0.5672353564854465,
      "learning_rate": 5e-06,
      "loss": 0.6113,
      "step": 560
    },
    {
      "epoch": 1.7325227963525835,
      "grad_norm": 0.5373878158094728,
      "learning_rate": 5e-06,
      "loss": 0.6165,
      "step": 570
    },
    {
      "epoch": 1.7629179331306992,
      "grad_norm": 0.6400981957262623,
      "learning_rate": 5e-06,
      "loss": 0.6135,
      "step": 580
    },
    {
      "epoch": 1.7933130699088147,
      "grad_norm": 0.5784668286610699,
      "learning_rate": 5e-06,
      "loss": 0.6127,
      "step": 590
    },
    {
      "epoch": 1.8237082066869301,
      "grad_norm": 0.4829691384216596,
      "learning_rate": 5e-06,
      "loss": 0.6165,
      "step": 600
    },
    {
      "epoch": 1.8541033434650456,
      "grad_norm": 0.5607220085804383,
      "learning_rate": 5e-06,
      "loss": 0.6142,
      "step": 610
    },
    {
      "epoch": 1.884498480243161,
      "grad_norm": 0.554409072476674,
      "learning_rate": 5e-06,
      "loss": 0.6108,
      "step": 620
    },
    {
      "epoch": 1.9148936170212765,
      "grad_norm": 0.4666111631306896,
      "learning_rate": 5e-06,
      "loss": 0.6054,
      "step": 630
    },
    {
      "epoch": 1.9452887537993921,
      "grad_norm": 0.6239667480279937,
      "learning_rate": 5e-06,
      "loss": 0.611,
      "step": 640
    },
    {
      "epoch": 1.9756838905775076,
      "grad_norm": 0.7028679873862859,
      "learning_rate": 5e-06,
      "loss": 0.6102,
      "step": 650
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.6488233208656311,
      "eval_runtime": 32.2972,
      "eval_samples_per_second": 274.358,
      "eval_steps_per_second": 0.557,
      "step": 658
    },
    {
      "epoch": 2.0060790273556233,
      "grad_norm": 0.7027126564788553,
      "learning_rate": 5e-06,
      "loss": 0.6004,
      "step": 660
    },
    {
      "epoch": 2.0364741641337387,
      "grad_norm": 0.6726728809115222,
      "learning_rate": 5e-06,
      "loss": 0.5684,
      "step": 670
    },
    {
      "epoch": 2.066869300911854,
      "grad_norm": 0.7037164967207233,
      "learning_rate": 5e-06,
      "loss": 0.5618,
      "step": 680
    },
    {
      "epoch": 2.0972644376899696,
      "grad_norm": 0.5634457537187532,
      "learning_rate": 5e-06,
      "loss": 0.5656,
      "step": 690
    },
    {
      "epoch": 2.127659574468085,
      "grad_norm": 0.7190014758588847,
      "learning_rate": 5e-06,
      "loss": 0.5669,
      "step": 700
    },
    {
      "epoch": 2.1580547112462005,
      "grad_norm": 0.6182141475596381,
      "learning_rate": 5e-06,
      "loss": 0.567,
      "step": 710
    },
    {
      "epoch": 2.188449848024316,
      "grad_norm": 0.6494642619820201,
      "learning_rate": 5e-06,
      "loss": 0.5625,
      "step": 720
    },
    {
      "epoch": 2.2188449848024314,
      "grad_norm": 0.5352624075289281,
      "learning_rate": 5e-06,
      "loss": 0.5674,
      "step": 730
    },
    {
      "epoch": 2.2492401215805473,
      "grad_norm": 0.5779884492140585,
      "learning_rate": 5e-06,
      "loss": 0.5637,
      "step": 740
    },
    {
      "epoch": 2.2796352583586628,
      "grad_norm": 0.6212702567540722,
      "learning_rate": 5e-06,
      "loss": 0.5755,
      "step": 750
    },
    {
      "epoch": 2.310030395136778,
      "grad_norm": 0.6675573139313373,
      "learning_rate": 5e-06,
      "loss": 0.5715,
      "step": 760
    },
    {
      "epoch": 2.3404255319148937,
      "grad_norm": 1.112080252593875,
      "learning_rate": 5e-06,
      "loss": 0.5673,
      "step": 770
    },
    {
      "epoch": 2.370820668693009,
      "grad_norm": 0.5430387056330088,
      "learning_rate": 5e-06,
      "loss": 0.5703,
      "step": 780
    },
    {
      "epoch": 2.4012158054711246,
      "grad_norm": 0.5730573646763748,
      "learning_rate": 5e-06,
      "loss": 0.5716,
      "step": 790
    },
    {
      "epoch": 2.43161094224924,
      "grad_norm": 0.5782525038445755,
      "learning_rate": 5e-06,
      "loss": 0.5673,
      "step": 800
    },
    {
      "epoch": 2.4620060790273555,
      "grad_norm": 0.5515752300271801,
      "learning_rate": 5e-06,
      "loss": 0.5729,
      "step": 810
    },
    {
      "epoch": 2.4924012158054714,
      "grad_norm": 0.6159973777815156,
      "learning_rate": 5e-06,
      "loss": 0.5712,
      "step": 820
    },
    {
      "epoch": 2.522796352583587,
      "grad_norm": 0.5694018418127859,
      "learning_rate": 5e-06,
      "loss": 0.5667,
      "step": 830
    },
    {
      "epoch": 2.5531914893617023,
      "grad_norm": 0.5797243112894562,
      "learning_rate": 5e-06,
      "loss": 0.5687,
      "step": 840
    },
    {
      "epoch": 2.5835866261398177,
      "grad_norm": 0.6616634425335868,
      "learning_rate": 5e-06,
      "loss": 0.5739,
      "step": 850
    },
    {
      "epoch": 2.613981762917933,
      "grad_norm": 0.6133935953312176,
      "learning_rate": 5e-06,
      "loss": 0.5731,
      "step": 860
    },
    {
      "epoch": 2.6443768996960486,
      "grad_norm": 0.6410077762466703,
      "learning_rate": 5e-06,
      "loss": 0.5772,
      "step": 870
    },
    {
      "epoch": 2.674772036474164,
      "grad_norm": 0.6957749590141841,
      "learning_rate": 5e-06,
      "loss": 0.5705,
      "step": 880
    },
    {
      "epoch": 2.7051671732522795,
      "grad_norm": 0.5103295479540869,
      "learning_rate": 5e-06,
      "loss": 0.5718,
      "step": 890
    },
    {
      "epoch": 2.735562310030395,
      "grad_norm": 0.5434510084681313,
      "learning_rate": 5e-06,
      "loss": 0.5635,
      "step": 900
    },
    {
      "epoch": 2.7659574468085104,
      "grad_norm": 0.5490760128674873,
      "learning_rate": 5e-06,
      "loss": 0.5689,
      "step": 910
    },
    {
      "epoch": 2.7963525835866263,
      "grad_norm": 0.5024890606168032,
      "learning_rate": 5e-06,
      "loss": 0.5725,
      "step": 920
    },
    {
      "epoch": 2.8267477203647418,
      "grad_norm": 0.558224951103413,
      "learning_rate": 5e-06,
      "loss": 0.5731,
      "step": 930
    },
    {
      "epoch": 2.857142857142857,
      "grad_norm": 0.5770328368518338,
      "learning_rate": 5e-06,
      "loss": 0.5682,
      "step": 940
    },
    {
      "epoch": 2.8875379939209727,
      "grad_norm": 0.5500792024748634,
      "learning_rate": 5e-06,
      "loss": 0.5691,
      "step": 950
    },
    {
      "epoch": 2.917933130699088,
      "grad_norm": 0.6529577507817819,
      "learning_rate": 5e-06,
      "loss": 0.5768,
      "step": 960
    },
    {
      "epoch": 2.9483282674772036,
      "grad_norm": 0.49823556701097355,
      "learning_rate": 5e-06,
      "loss": 0.5671,
      "step": 970
    },
    {
      "epoch": 2.978723404255319,
      "grad_norm": 0.5712256269882896,
      "learning_rate": 5e-06,
      "loss": 0.5707,
      "step": 980
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.6519396305084229,
      "eval_runtime": 30.3855,
      "eval_samples_per_second": 291.619,
      "eval_steps_per_second": 0.592,
      "step": 987
    },
    {
      "epoch": 3.0,
      "step": 987,
      "total_flos": 1653261761249280.0,
      "train_loss": 0.6273607391233744,
      "train_runtime": 4771.414,
      "train_samples_per_second": 105.851,
      "train_steps_per_second": 0.207
    }
  ],
  "logging_steps": 10,
  "max_steps": 987,
  "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": 1653261761249280.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}