{
  "best_global_step": 1000,
  "best_metric": 1.1034804582595825,
  "best_model_checkpoint": "models/checkpoint-1000",
  "epoch": 1.4814814814814814,
  "eval_steps": 100,
  "global_step": 1000,
  "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
    }
  ],
  "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": 7.2834303232128e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}