{
  "best_global_step": 1300,
  "best_metric": 1.090305209159851,
  "best_model_checkpoint": "models/checkpoint-1000",
  "epoch": 2.962962962962963,
  "eval_steps": 100,
  "global_step": 2000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.014814814814814815,
      "grad_norm": 1.2762936353683472,
      "learning_rate": 4.4334975369458135e-06,
      "loss": 1.928,
      "step": 10
    },
    {
      "epoch": 0.02962962962962963,
      "grad_norm": 0.9871466159820557,
      "learning_rate": 9.359605911330049e-06,
      "loss": 1.9285,
      "step": 20
    },
    {
      "epoch": 0.044444444444444446,
      "grad_norm": 0.6386047005653381,
      "learning_rate": 1.4285714285714285e-05,
      "loss": 1.6919,
      "step": 30
    },
    {
      "epoch": 0.05925925925925926,
      "grad_norm": 0.6393513679504395,
      "learning_rate": 1.921182266009852e-05,
      "loss": 1.7251,
      "step": 40
    },
    {
      "epoch": 0.07407407407407407,
      "grad_norm": 0.5770019888877869,
      "learning_rate": 2.413793103448276e-05,
      "loss": 1.5024,
      "step": 50
    },
    {
      "epoch": 0.08888888888888889,
      "grad_norm": 0.616218090057373,
      "learning_rate": 2.9064039408866993e-05,
      "loss": 1.4639,
      "step": 60
    },
    {
      "epoch": 0.1037037037037037,
      "grad_norm": 0.4487189054489136,
      "learning_rate": 3.399014778325123e-05,
      "loss": 1.5255,
      "step": 70
    },
    {
      "epoch": 0.11851851851851852,
      "grad_norm": 0.5754390358924866,
      "learning_rate": 3.891625615763547e-05,
      "loss": 1.3286,
      "step": 80
    },
    {
      "epoch": 0.13333333333333333,
      "grad_norm": 0.4768243730068207,
      "learning_rate": 4.384236453201971e-05,
      "loss": 1.5381,
      "step": 90
    },
    {
      "epoch": 0.14814814814814814,
      "grad_norm": 0.5551646947860718,
      "learning_rate": 4.876847290640394e-05,
      "loss": 1.5627,
      "step": 100
    },
    {
      "epoch": 0.14814814814814814,
      "eval_loss": 1.1925339698791504,
      "eval_runtime": 6.088,
      "eval_samples_per_second": 10.841,
      "eval_steps_per_second": 10.841,
      "step": 100
    },
    {
      "epoch": 0.16296296296296298,
      "grad_norm": 0.5339038968086243,
      "learning_rate": 5.3694581280788184e-05,
      "loss": 1.5624,
      "step": 110
    },
    {
      "epoch": 0.17777777777777778,
      "grad_norm": 0.5209947824478149,
      "learning_rate": 5.862068965517241e-05,
      "loss": 1.5489,
      "step": 120
    },
    {
      "epoch": 0.1925925925925926,
      "grad_norm": 0.482608437538147,
      "learning_rate": 6.354679802955665e-05,
      "loss": 1.5728,
      "step": 130
    },
    {
      "epoch": 0.2074074074074074,
      "grad_norm": 0.5582935810089111,
      "learning_rate": 6.84729064039409e-05,
      "loss": 1.2552,
      "step": 140
    },
    {
      "epoch": 0.2222222222222222,
      "grad_norm": 0.48873287439346313,
      "learning_rate": 7.339901477832512e-05,
      "loss": 1.3067,
      "step": 150
    },
    {
      "epoch": 0.23703703703703705,
      "grad_norm": 0.46673381328582764,
      "learning_rate": 7.832512315270936e-05,
      "loss": 1.3016,
      "step": 160
    },
    {
      "epoch": 0.2518518518518518,
      "grad_norm": 0.5195438265800476,
      "learning_rate": 8.325123152709359e-05,
      "loss": 1.438,
      "step": 170
    },
    {
      "epoch": 0.26666666666666666,
      "grad_norm": 0.4232109487056732,
      "learning_rate": 8.817733990147783e-05,
      "loss": 1.4887,
      "step": 180
    },
    {
      "epoch": 0.2814814814814815,
      "grad_norm": 0.49931880831718445,
      "learning_rate": 9.310344827586207e-05,
      "loss": 1.5654,
      "step": 190
    },
    {
      "epoch": 0.2962962962962963,
      "grad_norm": 0.5176865458488464,
      "learning_rate": 9.802955665024632e-05,
      "loss": 1.5047,
      "step": 200
    },
    {
      "epoch": 0.2962962962962963,
      "eval_loss": 1.166479468345642,
      "eval_runtime": 6.1258,
      "eval_samples_per_second": 10.774,
      "eval_steps_per_second": 10.774,
      "step": 200
    },
    {
      "epoch": 0.3111111111111111,
      "grad_norm": 0.41957828402519226,
      "learning_rate": 9.99973242740166e-05,
      "loss": 1.6237,
      "step": 210
    },
    {
      "epoch": 0.32592592592592595,
      "grad_norm": 0.4282706081867218,
      "learning_rate": 9.998097365231531e-05,
      "loss": 1.3151,
      "step": 220
    },
    {
      "epoch": 0.34074074074074073,
      "grad_norm": 0.3993554711341858,
      "learning_rate": 9.994976377846524e-05,
      "loss": 1.4522,
      "step": 230
    },
    {
      "epoch": 0.35555555555555557,
      "grad_norm": 0.4324376583099365,
      "learning_rate": 9.990370393110484e-05,
      "loss": 1.4027,
      "step": 240
    },
    {
      "epoch": 0.37037037037037035,
      "grad_norm": 0.3874889314174652,
      "learning_rate": 9.98428078037426e-05,
      "loss": 1.4661,
      "step": 250
    },
    {
      "epoch": 0.3851851851851852,
      "grad_norm": 0.42624586820602417,
      "learning_rate": 9.976709350068592e-05,
      "loss": 1.3995,
      "step": 260
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.4548305571079254,
      "learning_rate": 9.967658353165876e-05,
      "loss": 1.4277,
      "step": 270
    },
    {
      "epoch": 0.4148148148148148,
      "grad_norm": 0.5003868937492371,
      "learning_rate": 9.957130480510955e-05,
      "loss": 1.3731,
      "step": 280
    },
    {
      "epoch": 0.42962962962962964,
      "grad_norm": 0.4589303135871887,
      "learning_rate": 9.945128862021134e-05,
      "loss": 1.4744,
      "step": 290
    },
    {
      "epoch": 0.4444444444444444,
      "grad_norm": 0.4564014971256256,
      "learning_rate": 9.931657065755662e-05,
      "loss": 1.3788,
      "step": 300
    },
    {
      "epoch": 0.4444444444444444,
      "eval_loss": 1.154167652130127,
      "eval_runtime": 6.1524,
      "eval_samples_per_second": 10.727,
      "eval_steps_per_second": 10.727,
      "step": 300
    },
    {
      "epoch": 0.45925925925925926,
      "grad_norm": 0.4226451516151428,
      "learning_rate": 9.916719096854956e-05,
      "loss": 1.3056,
      "step": 310
    },
    {
      "epoch": 0.4740740740740741,
      "grad_norm": 0.426888108253479,
      "learning_rate": 9.900319396349875e-05,
      "loss": 1.5306,
      "step": 320
    },
    {
      "epoch": 0.4888888888888889,
      "grad_norm": 0.4118010401725769,
      "learning_rate": 9.88246283984142e-05,
      "loss": 1.5763,
      "step": 330
    },
    {
      "epoch": 0.5037037037037037,
      "grad_norm": 0.3842439651489258,
      "learning_rate": 9.863154736051214e-05,
      "loss": 1.3344,
      "step": 340
    },
    {
      "epoch": 0.5185185185185185,
      "grad_norm": 0.39596307277679443,
      "learning_rate": 9.842400825243237e-05,
      "loss": 1.2285,
      "step": 350
    },
    {
      "epoch": 0.5333333333333333,
      "grad_norm": 0.3774282932281494,
      "learning_rate": 9.820207277517253e-05,
      "loss": 1.4183,
      "step": 360
    },
    {
      "epoch": 0.5481481481481482,
      "grad_norm": 0.40762439370155334,
      "learning_rate": 9.796580690974454e-05,
      "loss": 1.4657,
      "step": 370
    },
    {
      "epoch": 0.562962962962963,
      "grad_norm": 0.43203601241111755,
      "learning_rate": 9.771528089755848e-05,
      "loss": 1.4277,
      "step": 380
    },
    {
      "epoch": 0.5777777777777777,
      "grad_norm": 0.4396969676017761,
      "learning_rate": 9.74505692195401e-05,
      "loss": 1.2714,
      "step": 390
    },
    {
      "epoch": 0.5925925925925926,
      "grad_norm": 0.4056893289089203,
      "learning_rate": 9.71717505739876e-05,
      "loss": 1.3499,
      "step": 400
    },
    {
      "epoch": 0.5925925925925926,
      "eval_loss": 1.1402429342269897,
      "eval_runtime": 6.0899,
      "eval_samples_per_second": 10.838,
      "eval_steps_per_second": 10.838,
      "step": 400
    },
    {
      "epoch": 0.6074074074074074,
      "grad_norm": 0.4428366422653198,
      "learning_rate": 9.687890785317488e-05,
      "loss": 1.3947,
      "step": 410
    },
    {
      "epoch": 0.6222222222222222,
      "grad_norm": 0.3946215510368347,
      "learning_rate": 9.657212811870783e-05,
      "loss": 1.3849,
      "step": 420
    },
    {
      "epoch": 0.6370370370370371,
      "grad_norm": 0.4270579516887665,
      "learning_rate": 9.625150257564097e-05,
      "loss": 1.4044,
      "step": 430
    },
    {
      "epoch": 0.6518518518518519,
      "grad_norm": 0.3932274281978607,
      "learning_rate": 9.591712654536247e-05,
      "loss": 1.4943,
      "step": 440
    },
    {
      "epoch": 0.6666666666666666,
      "grad_norm": 0.4396262764930725,
      "learning_rate": 9.55690994372552e-05,
      "loss": 1.4112,
      "step": 450
    },
    {
      "epoch": 0.6814814814814815,
      "grad_norm": 0.4467144310474396,
      "learning_rate": 9.520752471914254e-05,
      "loss": 1.338,
      "step": 460
    },
    {
      "epoch": 0.6962962962962963,
      "grad_norm": 0.409839928150177,
      "learning_rate": 9.483250988652756e-05,
      "loss": 1.4603,
      "step": 470
    },
    {
      "epoch": 0.7111111111111111,
      "grad_norm": 0.4538469910621643,
      "learning_rate": 9.444416643063485e-05,
      "loss": 1.4134,
      "step": 480
    },
    {
      "epoch": 0.725925925925926,
      "grad_norm": 0.43856948614120483,
      "learning_rate": 9.404260980526429e-05,
      "loss": 1.3097,
      "step": 490
    },
    {
      "epoch": 0.7407407407407407,
      "grad_norm": 0.367565780878067,
      "learning_rate": 9.362795939246694e-05,
      "loss": 1.3562,
      "step": 500
    },
    {
      "epoch": 0.7407407407407407,
      "eval_loss": 1.1292359828948975,
      "eval_runtime": 6.1291,
      "eval_samples_per_second": 10.768,
      "eval_steps_per_second": 10.768,
      "step": 500
    },
    {
      "epoch": 0.7555555555555555,
      "grad_norm": 0.3978855609893799,
      "learning_rate": 9.320033846705294e-05,
      "loss": 1.3375,
      "step": 510
    },
    {
      "epoch": 0.7703703703703704,
      "grad_norm": 0.42060592770576477,
      "learning_rate": 9.275987415994216e-05,
      "loss": 1.3213,
      "step": 520
    },
    {
      "epoch": 0.7851851851851852,
      "grad_norm": 0.4226202964782715,
      "learning_rate": 9.230669742036829e-05,
      "loss": 1.4621,
      "step": 530
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.44786694645881653,
      "learning_rate": 9.184094297694807e-05,
      "loss": 1.5007,
      "step": 540
    },
    {
      "epoch": 0.8148148148148148,
      "grad_norm": 0.43062612414360046,
      "learning_rate": 9.136274929762656e-05,
      "loss": 1.6064,
      "step": 550
    },
    {
      "epoch": 0.8296296296296296,
      "grad_norm": 0.39290761947631836,
      "learning_rate": 9.087225854851094e-05,
      "loss": 1.3576,
      "step": 560
    },
    {
      "epoch": 0.8444444444444444,
      "grad_norm": 0.3607810437679291,
      "learning_rate": 9.036961655160473e-05,
      "loss": 1.3658,
      "step": 570
    },
    {
      "epoch": 0.8592592592592593,
      "grad_norm": 0.36818262934684753,
      "learning_rate": 8.985497274145531e-05,
      "loss": 1.4767,
      "step": 580
    },
    {
      "epoch": 0.8740740740740741,
      "grad_norm": 0.4232269525527954,
      "learning_rate": 8.932848012072712e-05,
      "loss": 1.3843,
      "step": 590
    },
    {
      "epoch": 0.8888888888888888,
      "grad_norm": 0.38398849964141846,
      "learning_rate": 8.879029521471439e-05,
      "loss": 1.4844,
      "step": 600
    },
    {
      "epoch": 0.8888888888888888,
      "eval_loss": 1.1228456497192383,
      "eval_runtime": 6.139,
      "eval_samples_per_second": 10.751,
      "eval_steps_per_second": 10.751,
      "step": 600
    },
    {
      "epoch": 0.9037037037037037,
      "grad_norm": 0.3791407644748688,
      "learning_rate": 8.824057802480636e-05,
      "loss": 1.6602,
      "step": 610
    },
    {
      "epoch": 0.9185185185185185,
      "grad_norm": 0.40019121766090393,
      "learning_rate": 8.767949198091926e-05,
      "loss": 1.4272,
      "step": 620
    },
    {
      "epoch": 0.9333333333333333,
      "grad_norm": 0.40539026260375977,
      "learning_rate": 8.710720389290877e-05,
      "loss": 1.5542,
      "step": 630
    },
    {
      "epoch": 0.9481481481481482,
      "grad_norm": 0.4100588858127594,
      "learning_rate": 8.652388390097788e-05,
      "loss": 1.5269,
      "step": 640
    },
    {
      "epoch": 0.9629629629629629,
      "grad_norm": 0.3802511394023895,
      "learning_rate": 8.592970542509439e-05,
      "loss": 1.4225,
      "step": 650
    },
    {
      "epoch": 0.9777777777777777,
      "grad_norm": 0.37948888540267944,
      "learning_rate": 8.532484511343362e-05,
      "loss": 1.4258,
      "step": 660
    },
    {
      "epoch": 0.9925925925925926,
      "grad_norm": 0.3505156636238098,
      "learning_rate": 8.47094827898611e-05,
      "loss": 1.3502,
      "step": 670
    },
    {
      "epoch": 1.0074074074074073,
      "grad_norm": 0.3641723394393921,
      "learning_rate": 8.408380140047134e-05,
      "loss": 1.1749,
      "step": 680
    },
    {
      "epoch": 1.0222222222222221,
      "grad_norm": 0.36800840497016907,
      "learning_rate": 8.344798695919825e-05,
      "loss": 1.324,
      "step": 690
    },
    {
      "epoch": 1.037037037037037,
      "grad_norm": 0.39767882227897644,
      "learning_rate": 8.280222849251351e-05,
      "loss": 1.2107,
      "step": 700
    },
    {
      "epoch": 1.037037037037037,
      "eval_loss": 1.1203594207763672,
      "eval_runtime": 6.1063,
      "eval_samples_per_second": 10.808,
      "eval_steps_per_second": 10.808,
      "step": 700
    },
    {
      "epoch": 1.0518518518518518,
      "grad_norm": 0.4282814562320709,
      "learning_rate": 8.214671798322948e-05,
      "loss": 1.3683,
      "step": 710
    },
    {
      "epoch": 1.0666666666666667,
      "grad_norm": 0.42193305492401123,
      "learning_rate": 8.148165031342289e-05,
      "loss": 1.3319,
      "step": 720
    },
    {
      "epoch": 1.0814814814814815,
      "grad_norm": 0.4662479758262634,
      "learning_rate": 8.080722320649688e-05,
      "loss": 1.2601,
      "step": 730
    },
    {
      "epoch": 1.0962962962962963,
      "grad_norm": 0.5258338451385498,
      "learning_rate": 8.012363716839809e-05,
      "loss": 1.4298,
      "step": 740
    },
    {
      "epoch": 1.1111111111111112,
      "grad_norm": 0.4117671549320221,
      "learning_rate": 7.94310954280067e-05,
      "loss": 1.3073,
      "step": 750
    },
    {
      "epoch": 1.125925925925926,
      "grad_norm": 0.4276946187019348,
      "learning_rate": 7.872980387671686e-05,
      "loss": 1.2172,
      "step": 760
    },
    {
      "epoch": 1.1407407407407408,
      "grad_norm": 0.46661466360092163,
      "learning_rate": 7.801997100722542e-05,
      "loss": 1.0951,
      "step": 770
    },
    {
      "epoch": 1.1555555555555554,
      "grad_norm": 0.4148588478565216,
      "learning_rate": 7.730180785154759e-05,
      "loss": 1.2391,
      "step": 780
    },
    {
      "epoch": 1.1703703703703703,
      "grad_norm": 0.44193753600120544,
      "learning_rate": 7.657552791827744e-05,
      "loss": 1.2424,
      "step": 790
    },
    {
      "epoch": 1.1851851851851851,
      "grad_norm": 0.4419997036457062,
      "learning_rate": 7.584134712911217e-05,
      "loss": 1.1991,
      "step": 800
    },
    {
      "epoch": 1.1851851851851851,
      "eval_loss": 1.118480920791626,
      "eval_runtime": 6.1053,
      "eval_samples_per_second": 10.81,
      "eval_steps_per_second": 10.81,
      "step": 800
    },
    {
      "epoch": 1.2,
      "grad_norm": 0.4470985233783722,
      "learning_rate": 7.5099483754659e-05,
      "loss": 1.4003,
      "step": 810
    },
    {
      "epoch": 1.2148148148148148,
      "grad_norm": 0.4362390637397766,
      "learning_rate": 7.435015834954374e-05,
      "loss": 1.3019,
      "step": 820
    },
    {
      "epoch": 1.2296296296296296,
      "grad_norm": 0.43836158514022827,
      "learning_rate": 7.359359368684027e-05,
      "loss": 1.545,
      "step": 830
    },
    {
      "epoch": 1.2444444444444445,
      "grad_norm": 0.4291227459907532,
      "learning_rate": 7.283001469184052e-05,
      "loss": 1.2174,
      "step": 840
    },
    {
      "epoch": 1.2592592592592593,
      "grad_norm": 0.45286813378334045,
      "learning_rate": 7.20596483751846e-05,
      "loss": 1.2915,
      "step": 850
    },
    {
      "epoch": 1.2740740740740741,
      "grad_norm": 0.4440063536167145,
      "learning_rate": 7.128272376537097e-05,
      "loss": 1.3337,
      "step": 860
    },
    {
      "epoch": 1.2888888888888888,
      "grad_norm": 0.43544214963912964,
      "learning_rate": 7.049947184066663e-05,
      "loss": 1.3124,
      "step": 870
    },
    {
      "epoch": 1.3037037037037038,
      "grad_norm": 0.4050310552120209,
      "learning_rate": 6.971012546043774e-05,
      "loss": 1.4816,
      "step": 880
    },
    {
      "epoch": 1.3185185185185184,
      "grad_norm": 0.4690093398094177,
      "learning_rate": 6.891491929592096e-05,
      "loss": 1.3281,
      "step": 890
    },
    {
      "epoch": 1.3333333333333333,
      "grad_norm": 0.5653554201126099,
      "learning_rate": 6.811408976045613e-05,
      "loss": 1.214,
      "step": 900
    },
    {
      "epoch": 1.3333333333333333,
      "eval_loss": 1.1083061695098877,
      "eval_runtime": 6.1327,
      "eval_samples_per_second": 10.762,
      "eval_steps_per_second": 10.762,
      "step": 900
    },
    {
      "epoch": 1.348148148148148,
      "grad_norm": 0.5084799528121948,
      "learning_rate": 6.730787493920096e-05,
      "loss": 1.1765,
      "step": 910
    },
    {
      "epoch": 1.362962962962963,
      "grad_norm": 0.4370996952056885,
      "learning_rate": 6.649651451834885e-05,
      "loss": 1.0374,
      "step": 920
    },
    {
      "epoch": 1.3777777777777778,
      "grad_norm": 0.4599282741546631,
      "learning_rate": 6.568024971387048e-05,
      "loss": 1.2552,
      "step": 930
    },
    {
      "epoch": 1.3925925925925926,
      "grad_norm": 0.49177879095077515,
      "learning_rate": 6.485932319980088e-05,
      "loss": 1.3265,
      "step": 940
    },
    {
      "epoch": 1.4074074074074074,
      "grad_norm": 0.44238558411598206,
      "learning_rate": 6.40339790360928e-05,
      "loss": 1.2497,
      "step": 950
    },
    {
      "epoch": 1.4222222222222223,
      "grad_norm": 0.4685762822628021,
      "learning_rate": 6.320446259605814e-05,
      "loss": 1.4278,
      "step": 960
    },
    {
      "epoch": 1.4370370370370371,
      "grad_norm": 0.4614316523075104,
      "learning_rate": 6.237102049341898e-05,
      "loss": 1.3879,
      "step": 970
    },
    {
      "epoch": 1.4518518518518517,
      "grad_norm": 0.422168105840683,
      "learning_rate": 6.153390050898968e-05,
      "loss": 1.1758,
      "step": 980
    },
    {
      "epoch": 1.4666666666666668,
      "grad_norm": 0.4396195411682129,
      "learning_rate": 6.069335151701214e-05,
      "loss": 1.2809,
      "step": 990
    },
    {
      "epoch": 1.4814814814814814,
      "grad_norm": 0.4107734262943268,
      "learning_rate": 5.9849623411165934e-05,
      "loss": 1.3601,
      "step": 1000
    },
    {
      "epoch": 1.4814814814814814,
      "eval_loss": 1.1034804582595825,
      "eval_runtime": 6.1262,
      "eval_samples_per_second": 10.773,
      "eval_steps_per_second": 10.773,
      "step": 1000
    },
    {
      "epoch": 1.4962962962962962,
      "grad_norm": 0.44464996457099915,
      "learning_rate": 5.900296703027541e-05,
      "loss": 1.2977,
      "step": 1010
    },
    {
      "epoch": 1.511111111111111,
      "grad_norm": 0.4580194056034088,
      "learning_rate": 5.815363408373572e-05,
      "loss": 1.2059,
      "step": 1020
    },
    {
      "epoch": 1.525925925925926,
      "grad_norm": 0.5082858204841614,
      "learning_rate": 5.7301877076680075e-05,
      "loss": 1.4162,
      "step": 1030
    },
    {
      "epoch": 1.5407407407407407,
      "grad_norm": 0.44900521636009216,
      "learning_rate": 5.644794923491048e-05,
      "loss": 1.3725,
      "step": 1040
    },
    {
      "epoch": 1.5555555555555556,
      "grad_norm": 0.49669963121414185,
      "learning_rate": 5.559210442961411e-05,
      "loss": 1.2814,
      "step": 1050
    },
    {
      "epoch": 1.5703703703703704,
      "grad_norm": 0.4526650607585907,
      "learning_rate": 5.473459710188791e-05,
      "loss": 1.2945,
      "step": 1060
    },
    {
      "epoch": 1.585185185185185,
      "grad_norm": 0.46277210116386414,
      "learning_rate": 5.387568218709368e-05,
      "loss": 1.3592,
      "step": 1070
    },
    {
      "epoch": 1.6,
      "grad_norm": 0.467378705739975,
      "learning_rate": 5.3015615039066314e-05,
      "loss": 1.3909,
      "step": 1080
    },
    {
      "epoch": 1.6148148148148147,
      "grad_norm": 0.41982775926589966,
      "learning_rate": 5.215465135419754e-05,
      "loss": 1.3904,
      "step": 1090
    },
    {
      "epoch": 1.6296296296296298,
      "grad_norm": 0.5024394989013672,
      "learning_rate": 5.1293047095417835e-05,
      "loss": 1.2574,
      "step": 1100
    },
    {
      "epoch": 1.6296296296296298,
      "eval_loss": 1.1013927459716797,
      "eval_runtime": 6.1032,
      "eval_samples_per_second": 10.814,
      "eval_steps_per_second": 10.814,
      "step": 1100
    },
    {
      "epoch": 1.6444444444444444,
      "grad_norm": 0.464005708694458,
      "learning_rate": 5.043105841609912e-05,
      "loss": 1.34,
      "step": 1110
    },
    {
      "epoch": 1.6592592592592592,
      "grad_norm": 0.4217549264431,
      "learning_rate": 4.9568941583900886e-05,
      "loss": 1.372,
      "step": 1120
    },
    {
      "epoch": 1.674074074074074,
      "grad_norm": 0.4997546374797821,
      "learning_rate": 4.870695290458217e-05,
      "loss": 1.3459,
      "step": 1130
    },
    {
      "epoch": 1.6888888888888889,
      "grad_norm": 0.5140387415885925,
      "learning_rate": 4.784534864580247e-05,
      "loss": 1.1885,
      "step": 1140
    },
    {
      "epoch": 1.7037037037037037,
      "grad_norm": 0.5225329399108887,
      "learning_rate": 4.698438496093369e-05,
      "loss": 1.2256,
      "step": 1150
    },
    {
      "epoch": 1.7185185185185186,
      "grad_norm": 0.588337779045105,
      "learning_rate": 4.612431781290632e-05,
      "loss": 1.2276,
      "step": 1160
    },
    {
      "epoch": 1.7333333333333334,
      "grad_norm": 0.5051802396774292,
      "learning_rate": 4.526540289811211e-05,
      "loss": 1.335,
      "step": 1170
    },
    {
      "epoch": 1.748148148148148,
      "grad_norm": 0.4533883035182953,
      "learning_rate": 4.440789557038591e-05,
      "loss": 1.2393,
      "step": 1180
    },
    {
      "epoch": 1.762962962962963,
      "grad_norm": 0.5078204870223999,
      "learning_rate": 4.3552050765089526e-05,
      "loss": 1.3826,
      "step": 1190
    },
    {
      "epoch": 1.7777777777777777,
      "grad_norm": 0.5195481181144714,
      "learning_rate": 4.269812292331993e-05,
      "loss": 1.2652,
      "step": 1200
    },
    {
      "epoch": 1.7777777777777777,
      "eval_loss": 1.0943422317504883,
      "eval_runtime": 6.1154,
      "eval_samples_per_second": 10.792,
      "eval_steps_per_second": 10.792,
      "step": 1200
    },
    {
      "epoch": 1.7925925925925927,
      "grad_norm": 0.5239336490631104,
      "learning_rate": 4.184636591626429e-05,
      "loss": 1.346,
      "step": 1210
    },
    {
      "epoch": 1.8074074074074074,
      "grad_norm": 0.4464518129825592,
      "learning_rate": 4.099703296972458e-05,
      "loss": 1.1956,
      "step": 1220
    },
    {
      "epoch": 1.8222222222222222,
      "grad_norm": 0.4705686867237091,
      "learning_rate": 4.015037658883407e-05,
      "loss": 1.4122,
      "step": 1230
    },
    {
      "epoch": 1.837037037037037,
      "grad_norm": 0.455498069524765,
      "learning_rate": 3.930664848298788e-05,
      "loss": 1.1995,
      "step": 1240
    },
    {
      "epoch": 1.8518518518518519,
      "grad_norm": 0.46564143896102905,
      "learning_rate": 3.8466099491010334e-05,
      "loss": 1.2045,
      "step": 1250
    },
    {
      "epoch": 1.8666666666666667,
      "grad_norm": 0.5201877951622009,
      "learning_rate": 3.7628979506581035e-05,
      "loss": 1.3013,
      "step": 1260
    },
    {
      "epoch": 1.8814814814814815,
      "grad_norm": 0.5213164687156677,
      "learning_rate": 3.679553740394186e-05,
      "loss": 1.4276,
      "step": 1270
    },
    {
      "epoch": 1.8962962962962964,
      "grad_norm": 0.5018309354782104,
      "learning_rate": 3.596602096390721e-05,
      "loss": 1.2728,
      "step": 1280
    },
    {
      "epoch": 1.911111111111111,
      "grad_norm": 0.508551836013794,
      "learning_rate": 3.5140676800199135e-05,
      "loss": 1.4958,
      "step": 1290
    },
    {
      "epoch": 1.925925925925926,
      "grad_norm": 0.43654364347457886,
      "learning_rate": 3.431975028612952e-05,
      "loss": 1.4549,
      "step": 1300
    },
    {
      "epoch": 1.925925925925926,
      "eval_loss": 1.090305209159851,
      "eval_runtime": 6.0868,
      "eval_samples_per_second": 10.843,
      "eval_steps_per_second": 10.843,
      "step": 1300
    },
    {
      "epoch": 1.9407407407407407,
      "grad_norm": 0.5232775807380676,
      "learning_rate": 3.3503485481651165e-05,
      "loss": 1.2891,
      "step": 1310
    },
    {
      "epoch": 1.9555555555555557,
      "grad_norm": 0.5384771227836609,
      "learning_rate": 3.269212506079905e-05,
      "loss": 1.0813,
      "step": 1320
    },
    {
      "epoch": 1.9703703703703703,
      "grad_norm": 0.5372674465179443,
      "learning_rate": 3.188591023954388e-05,
      "loss": 1.3032,
      "step": 1330
    },
    {
      "epoch": 1.9851851851851852,
      "grad_norm": 0.47814837098121643,
      "learning_rate": 3.108508070407905e-05,
      "loss": 1.4167,
      "step": 1340
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.42629411816596985,
      "learning_rate": 3.0289874539562286e-05,
      "loss": 1.33,
      "step": 1350
    },
    {
      "epoch": 2.0148148148148146,
      "grad_norm": 0.45381420850753784,
      "learning_rate": 2.950052815933338e-05,
      "loss": 1.2558,
      "step": 1360
    },
    {
      "epoch": 2.0296296296296297,
      "grad_norm": 0.508139431476593,
      "learning_rate": 2.8717276234629043e-05,
      "loss": 1.2541,
      "step": 1370
    },
    {
      "epoch": 2.0444444444444443,
      "grad_norm": 0.46616795659065247,
      "learning_rate": 2.794035162481541e-05,
      "loss": 1.0489,
      "step": 1380
    },
    {
      "epoch": 2.0592592592592593,
      "grad_norm": 0.5279243588447571,
      "learning_rate": 2.7169985308159485e-05,
      "loss": 1.1235,
      "step": 1390
    },
    {
      "epoch": 2.074074074074074,
      "grad_norm": 0.5149587988853455,
      "learning_rate": 2.6406406313159748e-05,
      "loss": 1.0643,
      "step": 1400
    },
    {
      "epoch": 2.074074074074074,
      "eval_loss": 1.1016974449157715,
      "eval_runtime": 6.1005,
      "eval_samples_per_second": 10.819,
      "eval_steps_per_second": 10.819,
      "step": 1400
    },
    {
      "epoch": 2.088888888888889,
      "grad_norm": 0.495922714471817,
      "learning_rate": 2.5649841650456263e-05,
      "loss": 1.1033,
      "step": 1410
    },
    {
      "epoch": 2.1037037037037036,
      "grad_norm": 0.5315575003623962,
      "learning_rate": 2.4900516245341017e-05,
      "loss": 1.2393,
      "step": 1420
    },
    {
      "epoch": 2.1185185185185187,
      "grad_norm": 0.529092013835907,
      "learning_rate": 2.415865287088784e-05,
      "loss": 1.1497,
      "step": 1430
    },
    {
      "epoch": 2.1333333333333333,
      "grad_norm": 0.5322346687316895,
      "learning_rate": 2.3424472081722554e-05,
      "loss": 1.1443,
      "step": 1440
    },
    {
      "epoch": 2.148148148148148,
      "grad_norm": 0.6136528849601746,
      "learning_rate": 2.2698192148452408e-05,
      "loss": 1.1925,
      "step": 1450
    },
    {
      "epoch": 2.162962962962963,
      "grad_norm": 0.4956046938896179,
      "learning_rate": 2.1980028992774594e-05,
      "loss": 1.2621,
      "step": 1460
    },
    {
      "epoch": 2.1777777777777776,
      "grad_norm": 0.4820000231266022,
      "learning_rate": 2.1270196123283142e-05,
      "loss": 1.1378,
      "step": 1470
    },
    {
      "epoch": 2.1925925925925926,
      "grad_norm": 0.6214703917503357,
      "learning_rate": 2.0568904571993295e-05,
      "loss": 1.2145,
      "step": 1480
    },
    {
      "epoch": 2.2074074074074073,
      "grad_norm": 0.5239953994750977,
      "learning_rate": 1.987636283160193e-05,
      "loss": 1.2098,
      "step": 1490
    },
    {
      "epoch": 2.2222222222222223,
      "grad_norm": 0.47842028737068176,
      "learning_rate": 1.9192776793503133e-05,
      "loss": 1.2472,
      "step": 1500
    },
    {
      "epoch": 2.2222222222222223,
      "eval_loss": 1.0963842868804932,
      "eval_runtime": 6.1014,
      "eval_samples_per_second": 10.817,
      "eval_steps_per_second": 10.817,
      "step": 1500
    },
    {
      "epoch": 2.237037037037037,
      "grad_norm": 0.5109423398971558,
      "learning_rate": 1.8518349686577113e-05,
      "loss": 1.2154,
      "step": 1510
    },
    {
      "epoch": 2.251851851851852,
      "grad_norm": 0.5405035614967346,
      "learning_rate": 1.7853282016770536e-05,
      "loss": 1.23,
      "step": 1520
    },
    {
      "epoch": 2.2666666666666666,
      "grad_norm": 0.551059365272522,
      "learning_rate": 1.7197771507486493e-05,
      "loss": 1.2517,
      "step": 1530
    },
    {
      "epoch": 2.2814814814814817,
      "grad_norm": 0.6016335487365723,
      "learning_rate": 1.655201304080178e-05,
      "loss": 1.1859,
      "step": 1540
    },
    {
      "epoch": 2.2962962962962963,
      "grad_norm": 0.5649178624153137,
      "learning_rate": 1.5916198599528663e-05,
      "loss": 1.1167,
      "step": 1550
    },
    {
      "epoch": 2.311111111111111,
      "grad_norm": 0.4889310598373413,
      "learning_rate": 1.5290517210138888e-05,
      "loss": 1.0467,
      "step": 1560
    },
    {
      "epoch": 2.325925925925926,
      "grad_norm": 0.6229158043861389,
      "learning_rate": 1.4675154886566383e-05,
      "loss": 1.3891,
      "step": 1570
    },
    {
      "epoch": 2.3407407407407406,
      "grad_norm": 0.5472054481506348,
      "learning_rate": 1.4070294574905607e-05,
      "loss": 1.1602,
      "step": 1580
    },
    {
      "epoch": 2.3555555555555556,
      "grad_norm": 0.5209168791770935,
      "learning_rate": 1.3476116099022134e-05,
      "loss": 1.1856,
      "step": 1590
    },
    {
      "epoch": 2.3703703703703702,
      "grad_norm": 0.5474329590797424,
      "learning_rate": 1.2892796107091238e-05,
      "loss": 1.2152,
      "step": 1600
    },
    {
      "epoch": 2.3703703703703702,
      "eval_loss": 1.0988593101501465,
      "eval_runtime": 6.0965,
      "eval_samples_per_second": 10.826,
      "eval_steps_per_second": 10.826,
      "step": 1600
    },
    {
      "epoch": 2.3851851851851853,
      "grad_norm": 0.5262495875358582,
      "learning_rate": 1.232050801908074e-05,
      "loss": 1.1864,
      "step": 1610
    },
    {
      "epoch": 2.4,
      "grad_norm": 0.5173038244247437,
      "learning_rate": 1.1759421975193635e-05,
      "loss": 1.1231,
      "step": 1620
    },
    {
      "epoch": 2.414814814814815,
      "grad_norm": 0.511406421661377,
      "learning_rate": 1.1209704785285624e-05,
      "loss": 1.1723,
      "step": 1630
    },
    {
      "epoch": 2.4296296296296296,
      "grad_norm": 0.5466118454933167,
      "learning_rate": 1.067151987927288e-05,
      "loss": 1.1933,
      "step": 1640
    },
    {
      "epoch": 2.4444444444444446,
      "grad_norm": 0.5360016822814941,
      "learning_rate": 1.0145027258544704e-05,
      "loss": 1.2755,
      "step": 1650
    },
    {
      "epoch": 2.4592592592592593,
      "grad_norm": 0.517094612121582,
      "learning_rate": 9.63038344839527e-06,
      "loss": 1.1326,
      "step": 1660
    },
    {
      "epoch": 2.474074074074074,
      "grad_norm": 0.5082244873046875,
      "learning_rate": 9.127741451489086e-06,
      "loss": 1.1996,
      "step": 1670
    },
    {
      "epoch": 2.488888888888889,
      "grad_norm": 0.5089357495307922,
      "learning_rate": 8.637250702373445e-06,
      "loss": 1.1574,
      "step": 1680
    },
    {
      "epoch": 2.5037037037037035,
      "grad_norm": 0.5557934641838074,
      "learning_rate": 8.159057023051936e-06,
      "loss": 1.2476,
      "step": 1690
    },
    {
      "epoch": 2.5185185185185186,
      "grad_norm": 0.5560778379440308,
      "learning_rate": 7.693302579631712e-06,
      "loss": 1.0941,
      "step": 1700
    },
    {
      "epoch": 2.5185185185185186,
      "eval_loss": 1.0979622602462769,
      "eval_runtime": 6.0857,
      "eval_samples_per_second": 10.845,
      "eval_steps_per_second": 10.845,
      "step": 1700
    },
    {
      "epoch": 2.533333333333333,
      "grad_norm": 0.5187978744506836,
      "learning_rate": 7.24012584005786e-06,
      "loss": 1.3577,
      "step": 1710
    },
    {
      "epoch": 2.5481481481481483,
      "grad_norm": 0.5103093385696411,
      "learning_rate": 6.799661532947049e-06,
      "loss": 1.0282,
      "step": 1720
    },
    {
      "epoch": 2.562962962962963,
      "grad_norm": 0.5039860010147095,
      "learning_rate": 6.372040607533064e-06,
      "loss": 1.1957,
      "step": 1730
    },
    {
      "epoch": 2.5777777777777775,
      "grad_norm": 0.48238325119018555,
      "learning_rate": 5.957390194735723e-06,
      "loss": 1.2012,
      "step": 1740
    },
    {
      "epoch": 2.5925925925925926,
      "grad_norm": 0.615743100643158,
      "learning_rate": 5.55583356936516e-06,
      "loss": 1.2513,
      "step": 1750
    },
    {
      "epoch": 2.6074074074074076,
      "grad_norm": 0.4950842261314392,
      "learning_rate": 5.1674901134724395e-06,
      "loss": 1.1578,
      "step": 1760
    },
    {
      "epoch": 2.6222222222222222,
      "grad_norm": 0.5384798645973206,
      "learning_rate": 4.792475280857472e-06,
      "loss": 1.1497,
      "step": 1770
    },
    {
      "epoch": 2.637037037037037,
      "grad_norm": 0.6161677241325378,
      "learning_rate": 4.430900562744805e-06,
      "loss": 1.1602,
      "step": 1780
    },
    {
      "epoch": 2.651851851851852,
      "grad_norm": 0.5426111221313477,
      "learning_rate": 4.082873454637542e-06,
      "loss": 1.1659,
      "step": 1790
    },
    {
      "epoch": 2.6666666666666665,
      "grad_norm": 0.457955002784729,
      "learning_rate": 3.7484974243590375e-06,
      "loss": 1.0319,
      "step": 1800
    },
    {
      "epoch": 2.6666666666666665,
      "eval_loss": 1.0951391458511353,
      "eval_runtime": 6.0852,
      "eval_samples_per_second": 10.846,
      "eval_steps_per_second": 10.846,
      "step": 1800
    },
    {
      "epoch": 2.6814814814814816,
      "grad_norm": 0.547542154788971,
      "learning_rate": 3.4278718812921774e-06,
      "loss": 1.2674,
      "step": 1810
    },
    {
      "epoch": 2.696296296296296,
      "grad_norm": 0.591656506061554,
      "learning_rate": 3.1210921468251166e-06,
      "loss": 1.2564,
      "step": 1820
    },
    {
      "epoch": 2.7111111111111112,
      "grad_norm": 0.4875819683074951,
      "learning_rate": 2.828249426012414e-06,
      "loss": 1.1971,
      "step": 1830
    },
    {
      "epoch": 2.725925925925926,
      "grad_norm": 0.4732377529144287,
      "learning_rate": 2.549430780459905e-06,
      "loss": 1.2736,
      "step": 1840
    },
    {
      "epoch": 2.7407407407407405,
      "grad_norm": 0.5228894948959351,
      "learning_rate": 2.284719102441518e-06,
      "loss": 1.2338,
      "step": 1850
    },
    {
      "epoch": 2.7555555555555555,
      "grad_norm": 0.5478381514549255,
      "learning_rate": 2.0341930902554762e-06,
      "loss": 1.057,
      "step": 1860
    },
    {
      "epoch": 2.7703703703703706,
      "grad_norm": 0.5223204493522644,
      "learning_rate": 1.7979272248274725e-06,
      "loss": 1.1046,
      "step": 1870
    },
    {
      "epoch": 2.785185185185185,
      "grad_norm": 0.5573652386665344,
      "learning_rate": 1.57599174756764e-06,
      "loss": 0.9805,
      "step": 1880
    },
    {
      "epoch": 2.8,
      "grad_norm": 0.5413745045661926,
      "learning_rate": 1.3684526394878604e-06,
      "loss": 1.2528,
      "step": 1890
    },
    {
      "epoch": 2.814814814814815,
      "grad_norm": 0.5002546906471252,
      "learning_rate": 1.1753716015858008e-06,
      "loss": 1.3043,
      "step": 1900
    },
    {
      "epoch": 2.814814814814815,
      "eval_loss": 1.0956275463104248,
      "eval_runtime": 6.1088,
      "eval_samples_per_second": 10.804,
      "eval_steps_per_second": 10.804,
      "step": 1900
    },
    {
      "epoch": 2.8296296296296295,
      "grad_norm": 0.5736653208732605,
      "learning_rate": 9.968060365012466e-07,
      "loss": 1.1591,
      "step": 1910
    },
    {
      "epoch": 2.8444444444444446,
      "grad_norm": 0.5387482643127441,
      "learning_rate": 8.328090314504544e-07,
      "loss": 1.2999,
      "step": 1920
    },
    {
      "epoch": 2.859259259259259,
      "grad_norm": 0.555121123790741,
      "learning_rate": 6.834293424433868e-07,
      "loss": 1.1029,
      "step": 1930
    },
    {
      "epoch": 2.8740740740740742,
      "grad_norm": 0.6326370239257812,
      "learning_rate": 5.487113797886689e-07,
      "loss": 1.0369,
      "step": 1940
    },
    {
      "epoch": 2.888888888888889,
      "grad_norm": 0.5233654975891113,
      "learning_rate": 4.2869519489045493e-07,
      "loss": 1.0585,
      "step": 1950
    },
    {
      "epoch": 2.9037037037037035,
      "grad_norm": 0.5031371116638184,
      "learning_rate": 3.2341646834124773e-07,
      "loss": 1.1165,
      "step": 1960
    },
    {
      "epoch": 2.9185185185185185,
      "grad_norm": 0.5148937106132507,
      "learning_rate": 2.3290649931408947e-07,
      "loss": 1.3139,
      "step": 1970
    },
    {
      "epoch": 2.9333333333333336,
      "grad_norm": 0.5122260451316833,
      "learning_rate": 1.5719219625740523e-07,
      "loss": 1.2775,
      "step": 1980
    },
    {
      "epoch": 2.948148148148148,
      "grad_norm": 0.5385026931762695,
      "learning_rate": 9.629606889516862e-08,
      "loss": 1.1664,
      "step": 1990
    },
    {
      "epoch": 2.962962962962963,
      "grad_norm": 0.5466761589050293,
      "learning_rate": 5.023622153477736e-08,
      "loss": 1.1295,
      "step": 2000
    },
    {
      "epoch": 2.962962962962963,
      "eval_loss": 1.0957329273223877,
      "eval_runtime": 6.1165,
      "eval_samples_per_second": 10.79,
      "eval_steps_per_second": 10.79,
      "step": 2000
    }
  ],
  "logging_steps": 10,
  "max_steps": 2025,
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.4615096406354432e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}