{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 16875,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.008888888888888889,
      "grad_norm": 69.5,
      "learning_rate": 2.8994082840236688e-06,
      "loss": 0.6647,
      "step": 50
    },
    {
      "epoch": 0.017777777777777778,
      "grad_norm": 44.75,
      "learning_rate": 5.857988165680474e-06,
      "loss": 0.6167,
      "step": 100
    },
    {
      "epoch": 0.02666666666666667,
      "grad_norm": 93.0,
      "learning_rate": 8.816568047337279e-06,
      "loss": 0.5663,
      "step": 150
    },
    {
      "epoch": 0.035555555555555556,
      "grad_norm": 49.5,
      "learning_rate": 1.1775147928994083e-05,
      "loss": 0.5885,
      "step": 200
    },
    {
      "epoch": 0.044444444444444446,
      "grad_norm": 44.0,
      "learning_rate": 1.4733727810650888e-05,
      "loss": 0.5075,
      "step": 250
    },
    {
      "epoch": 0.05333333333333334,
      "grad_norm": 39.0,
      "learning_rate": 1.7692307692307694e-05,
      "loss": 0.4993,
      "step": 300
    },
    {
      "epoch": 0.06222222222222222,
      "grad_norm": 39.0,
      "learning_rate": 1.9986696498760357e-05,
      "loss": 0.525,
      "step": 350
    },
    {
      "epoch": 0.07111111111111111,
      "grad_norm": 66.0,
      "learning_rate": 1.9926226038580155e-05,
      "loss": 0.5397,
      "step": 400
    },
    {
      "epoch": 0.08,
      "grad_norm": 50.25,
      "learning_rate": 1.9865755578399954e-05,
      "loss": 0.5969,
      "step": 450
    },
    {
      "epoch": 0.08888888888888889,
      "grad_norm": 44.25,
      "learning_rate": 1.980528511821975e-05,
      "loss": 0.5574,
      "step": 500
    },
    {
      "epoch": 0.08888888888888889,
      "eval_loss": 0.04251113533973694,
      "eval_runtime": 0.8673,
      "eval_samples_per_second": 11.53,
      "eval_steps_per_second": 2.306,
      "step": 500
    },
    {
      "epoch": 0.09777777777777778,
      "grad_norm": 47.75,
      "learning_rate": 1.9744814658039547e-05,
      "loss": 0.5719,
      "step": 550
    },
    {
      "epoch": 0.10666666666666667,
      "grad_norm": 39.75,
      "learning_rate": 1.9684344197859345e-05,
      "loss": 0.575,
      "step": 600
    },
    {
      "epoch": 0.11555555555555555,
      "grad_norm": 51.5,
      "learning_rate": 1.9623873737679144e-05,
      "loss": 0.582,
      "step": 650
    },
    {
      "epoch": 0.12444444444444444,
      "grad_norm": 48.5,
      "learning_rate": 1.9563403277498942e-05,
      "loss": 0.5276,
      "step": 700
    },
    {
      "epoch": 0.13333333333333333,
      "grad_norm": 65.0,
      "learning_rate": 1.950293281731874e-05,
      "loss": 0.5876,
      "step": 750
    },
    {
      "epoch": 0.14222222222222222,
      "grad_norm": 43.5,
      "learning_rate": 1.944246235713854e-05,
      "loss": 0.5616,
      "step": 800
    },
    {
      "epoch": 0.1511111111111111,
      "grad_norm": 55.25,
      "learning_rate": 1.9381991896958337e-05,
      "loss": 0.5332,
      "step": 850
    },
    {
      "epoch": 0.16,
      "grad_norm": 59.75,
      "learning_rate": 1.9321521436778135e-05,
      "loss": 0.5586,
      "step": 900
    },
    {
      "epoch": 0.1688888888888889,
      "grad_norm": 44.75,
      "learning_rate": 1.9261050976597933e-05,
      "loss": 0.5412,
      "step": 950
    },
    {
      "epoch": 0.17777777777777778,
      "grad_norm": 74.0,
      "learning_rate": 1.9200580516417732e-05,
      "loss": 0.5757,
      "step": 1000
    },
    {
      "epoch": 0.17777777777777778,
      "eval_loss": 0.040884535759687424,
      "eval_runtime": 0.2448,
      "eval_samples_per_second": 40.844,
      "eval_steps_per_second": 8.169,
      "step": 1000
    },
    {
      "epoch": 0.18666666666666668,
      "grad_norm": 63.25,
      "learning_rate": 1.914011005623753e-05,
      "loss": 0.5731,
      "step": 1050
    },
    {
      "epoch": 0.19555555555555557,
      "grad_norm": 46.0,
      "learning_rate": 1.9079639596057325e-05,
      "loss": 0.5527,
      "step": 1100
    },
    {
      "epoch": 0.20444444444444446,
      "grad_norm": 49.25,
      "learning_rate": 1.9019169135877123e-05,
      "loss": 0.5854,
      "step": 1150
    },
    {
      "epoch": 0.21333333333333335,
      "grad_norm": 39.75,
      "learning_rate": 1.895869867569692e-05,
      "loss": 0.5437,
      "step": 1200
    },
    {
      "epoch": 0.2222222222222222,
      "grad_norm": 49.75,
      "learning_rate": 1.889822821551672e-05,
      "loss": 0.5386,
      "step": 1250
    },
    {
      "epoch": 0.2311111111111111,
      "grad_norm": 73.0,
      "learning_rate": 1.8837757755336518e-05,
      "loss": 0.5514,
      "step": 1300
    },
    {
      "epoch": 0.24,
      "grad_norm": 40.25,
      "learning_rate": 1.8777287295156316e-05,
      "loss": 0.5508,
      "step": 1350
    },
    {
      "epoch": 0.24888888888888888,
      "grad_norm": 33.25,
      "learning_rate": 1.8716816834976115e-05,
      "loss": 0.5709,
      "step": 1400
    },
    {
      "epoch": 0.2577777777777778,
      "grad_norm": 31.875,
      "learning_rate": 1.8656346374795913e-05,
      "loss": 0.5281,
      "step": 1450
    },
    {
      "epoch": 0.26666666666666666,
      "grad_norm": 44.0,
      "learning_rate": 1.859587591461571e-05,
      "loss": 0.589,
      "step": 1500
    },
    {
      "epoch": 0.26666666666666666,
      "eval_loss": 0.02658979222178459,
      "eval_runtime": 0.2453,
      "eval_samples_per_second": 40.774,
      "eval_steps_per_second": 8.155,
      "step": 1500
    },
    {
      "epoch": 0.27555555555555555,
      "grad_norm": 42.0,
      "learning_rate": 1.853540545443551e-05,
      "loss": 0.5987,
      "step": 1550
    },
    {
      "epoch": 0.28444444444444444,
      "grad_norm": 46.0,
      "learning_rate": 1.8474934994255308e-05,
      "loss": 0.531,
      "step": 1600
    },
    {
      "epoch": 0.29333333333333333,
      "grad_norm": 72.0,
      "learning_rate": 1.8414464534075106e-05,
      "loss": 0.5679,
      "step": 1650
    },
    {
      "epoch": 0.3022222222222222,
      "grad_norm": 65.0,
      "learning_rate": 1.83539940738949e-05,
      "loss": 0.5447,
      "step": 1700
    },
    {
      "epoch": 0.3111111111111111,
      "grad_norm": 50.75,
      "learning_rate": 1.82935236137147e-05,
      "loss": 0.6193,
      "step": 1750
    },
    {
      "epoch": 0.32,
      "grad_norm": 46.75,
      "learning_rate": 1.8233053153534498e-05,
      "loss": 0.5606,
      "step": 1800
    },
    {
      "epoch": 0.3288888888888889,
      "grad_norm": 52.5,
      "learning_rate": 1.8172582693354296e-05,
      "loss": 0.5106,
      "step": 1850
    },
    {
      "epoch": 0.3377777777777778,
      "grad_norm": 37.75,
      "learning_rate": 1.8112112233174094e-05,
      "loss": 0.5475,
      "step": 1900
    },
    {
      "epoch": 0.3466666666666667,
      "grad_norm": 52.5,
      "learning_rate": 1.8051641772993893e-05,
      "loss": 0.5578,
      "step": 1950
    },
    {
      "epoch": 0.35555555555555557,
      "grad_norm": 35.0,
      "learning_rate": 1.799117131281369e-05,
      "loss": 0.5796,
      "step": 2000
    },
    {
      "epoch": 0.35555555555555557,
      "eval_loss": 0.01918725296854973,
      "eval_runtime": 0.2325,
      "eval_samples_per_second": 43.013,
      "eval_steps_per_second": 8.603,
      "step": 2000
    },
    {
      "epoch": 0.36444444444444446,
      "grad_norm": 47.25,
      "learning_rate": 1.793070085263349e-05,
      "loss": 0.5821,
      "step": 2050
    },
    {
      "epoch": 0.37333333333333335,
      "grad_norm": 45.75,
      "learning_rate": 1.7870230392453288e-05,
      "loss": 0.6066,
      "step": 2100
    },
    {
      "epoch": 0.38222222222222224,
      "grad_norm": 95.0,
      "learning_rate": 1.7809759932273086e-05,
      "loss": 0.5818,
      "step": 2150
    },
    {
      "epoch": 0.39111111111111113,
      "grad_norm": 50.25,
      "learning_rate": 1.7749289472092884e-05,
      "loss": 0.555,
      "step": 2200
    },
    {
      "epoch": 0.4,
      "grad_norm": 57.75,
      "learning_rate": 1.7688819011912683e-05,
      "loss": 0.6096,
      "step": 2250
    },
    {
      "epoch": 0.4088888888888889,
      "grad_norm": 50.0,
      "learning_rate": 1.762834855173248e-05,
      "loss": 0.5591,
      "step": 2300
    },
    {
      "epoch": 0.4177777777777778,
      "grad_norm": 38.0,
      "learning_rate": 1.7567878091552276e-05,
      "loss": 0.5534,
      "step": 2350
    },
    {
      "epoch": 0.4266666666666667,
      "grad_norm": 55.25,
      "learning_rate": 1.7507407631372074e-05,
      "loss": 0.5954,
      "step": 2400
    },
    {
      "epoch": 0.43555555555555553,
      "grad_norm": 61.25,
      "learning_rate": 1.7446937171191872e-05,
      "loss": 0.5246,
      "step": 2450
    },
    {
      "epoch": 0.4444444444444444,
      "grad_norm": 37.5,
      "learning_rate": 1.738646671101167e-05,
      "loss": 0.5513,
      "step": 2500
    },
    {
      "epoch": 0.4444444444444444,
      "eval_loss": 0.019861677661538124,
      "eval_runtime": 0.2317,
      "eval_samples_per_second": 43.161,
      "eval_steps_per_second": 8.632,
      "step": 2500
    },
    {
      "epoch": 0.4533333333333333,
      "grad_norm": 50.5,
      "learning_rate": 1.732599625083147e-05,
      "loss": 0.5717,
      "step": 2550
    },
    {
      "epoch": 0.4622222222222222,
      "grad_norm": 41.75,
      "learning_rate": 1.7265525790651267e-05,
      "loss": 0.5213,
      "step": 2600
    },
    {
      "epoch": 0.4711111111111111,
      "grad_norm": 59.75,
      "learning_rate": 1.7205055330471066e-05,
      "loss": 0.5403,
      "step": 2650
    },
    {
      "epoch": 0.48,
      "grad_norm": 34.75,
      "learning_rate": 1.7144584870290864e-05,
      "loss": 0.5731,
      "step": 2700
    },
    {
      "epoch": 0.4888888888888889,
      "grad_norm": 53.0,
      "learning_rate": 1.7084114410110662e-05,
      "loss": 0.5839,
      "step": 2750
    },
    {
      "epoch": 0.49777777777777776,
      "grad_norm": 60.0,
      "learning_rate": 1.702364394993046e-05,
      "loss": 0.5833,
      "step": 2800
    },
    {
      "epoch": 0.5066666666666667,
      "grad_norm": 55.75,
      "learning_rate": 1.696317348975026e-05,
      "loss": 0.5696,
      "step": 2850
    },
    {
      "epoch": 0.5155555555555555,
      "grad_norm": 57.75,
      "learning_rate": 1.6902703029570057e-05,
      "loss": 0.5804,
      "step": 2900
    },
    {
      "epoch": 0.5244444444444445,
      "grad_norm": 47.5,
      "learning_rate": 1.6842232569389852e-05,
      "loss": 0.5935,
      "step": 2950
    },
    {
      "epoch": 0.5333333333333333,
      "grad_norm": 39.25,
      "learning_rate": 1.678176210920965e-05,
      "loss": 0.516,
      "step": 3000
    },
    {
      "epoch": 0.5333333333333333,
      "eval_loss": 0.016407368704676628,
      "eval_runtime": 0.2357,
      "eval_samples_per_second": 42.428,
      "eval_steps_per_second": 8.486,
      "step": 3000
    },
    {
      "epoch": 0.5422222222222223,
      "grad_norm": 80.5,
      "learning_rate": 1.672129164902945e-05,
      "loss": 0.6103,
      "step": 3050
    },
    {
      "epoch": 0.5511111111111111,
      "grad_norm": 65.5,
      "learning_rate": 1.6660821188849247e-05,
      "loss": 0.4698,
      "step": 3100
    },
    {
      "epoch": 0.56,
      "grad_norm": 60.0,
      "learning_rate": 1.6600350728669045e-05,
      "loss": 0.6191,
      "step": 3150
    },
    {
      "epoch": 0.5688888888888889,
      "grad_norm": 53.25,
      "learning_rate": 1.6539880268488844e-05,
      "loss": 0.518,
      "step": 3200
    },
    {
      "epoch": 0.5777777777777777,
      "grad_norm": 74.0,
      "learning_rate": 1.6479409808308642e-05,
      "loss": 0.5538,
      "step": 3250
    },
    {
      "epoch": 0.5866666666666667,
      "grad_norm": 42.25,
      "learning_rate": 1.641893934812844e-05,
      "loss": 0.6062,
      "step": 3300
    },
    {
      "epoch": 0.5955555555555555,
      "grad_norm": 48.25,
      "learning_rate": 1.635846888794824e-05,
      "loss": 0.5821,
      "step": 3350
    },
    {
      "epoch": 0.6044444444444445,
      "grad_norm": 48.25,
      "learning_rate": 1.6297998427768037e-05,
      "loss": 0.5769,
      "step": 3400
    },
    {
      "epoch": 0.6133333333333333,
      "grad_norm": 53.75,
      "learning_rate": 1.6237527967587835e-05,
      "loss": 0.5545,
      "step": 3450
    },
    {
      "epoch": 0.6222222222222222,
      "grad_norm": 44.5,
      "learning_rate": 1.6177057507407634e-05,
      "loss": 0.5537,
      "step": 3500
    },
    {
      "epoch": 0.6222222222222222,
      "eval_loss": 0.0168689526617527,
      "eval_runtime": 0.2329,
      "eval_samples_per_second": 42.943,
      "eval_steps_per_second": 8.589,
      "step": 3500
    },
    {
      "epoch": 0.6311111111111111,
      "grad_norm": 72.0,
      "learning_rate": 1.611658704722743e-05,
      "loss": 0.5742,
      "step": 3550
    },
    {
      "epoch": 0.64,
      "grad_norm": 35.0,
      "learning_rate": 1.6056116587047227e-05,
      "loss": 0.5335,
      "step": 3600
    },
    {
      "epoch": 0.6488888888888888,
      "grad_norm": 38.25,
      "learning_rate": 1.5995646126867025e-05,
      "loss": 0.5515,
      "step": 3650
    },
    {
      "epoch": 0.6577777777777778,
      "grad_norm": 42.25,
      "learning_rate": 1.5935175666686823e-05,
      "loss": 0.5416,
      "step": 3700
    },
    {
      "epoch": 0.6666666666666666,
      "grad_norm": 37.75,
      "learning_rate": 1.587470520650662e-05,
      "loss": 0.5504,
      "step": 3750
    },
    {
      "epoch": 0.6755555555555556,
      "grad_norm": 69.0,
      "learning_rate": 1.581423474632642e-05,
      "loss": 0.5644,
      "step": 3800
    },
    {
      "epoch": 0.6844444444444444,
      "grad_norm": 39.5,
      "learning_rate": 1.575376428614622e-05,
      "loss": 0.5868,
      "step": 3850
    },
    {
      "epoch": 0.6933333333333334,
      "grad_norm": 41.0,
      "learning_rate": 1.5693293825966017e-05,
      "loss": 0.572,
      "step": 3900
    },
    {
      "epoch": 0.7022222222222222,
      "grad_norm": 44.5,
      "learning_rate": 1.5632823365785815e-05,
      "loss": 0.5448,
      "step": 3950
    },
    {
      "epoch": 0.7111111111111111,
      "grad_norm": 35.0,
      "learning_rate": 1.5572352905605613e-05,
      "loss": 0.5392,
      "step": 4000
    },
    {
      "epoch": 0.7111111111111111,
      "eval_loss": 0.013791784644126892,
      "eval_runtime": 0.2368,
      "eval_samples_per_second": 42.229,
      "eval_steps_per_second": 8.446,
      "step": 4000
    },
    {
      "epoch": 0.72,
      "grad_norm": 43.75,
      "learning_rate": 1.551188244542541e-05,
      "loss": 0.4832,
      "step": 4050
    },
    {
      "epoch": 0.7288888888888889,
      "grad_norm": 43.0,
      "learning_rate": 1.545141198524521e-05,
      "loss": 0.5232,
      "step": 4100
    },
    {
      "epoch": 0.7377777777777778,
      "grad_norm": 44.25,
      "learning_rate": 1.5390941525065005e-05,
      "loss": 0.5381,
      "step": 4150
    },
    {
      "epoch": 0.7466666666666667,
      "grad_norm": 52.5,
      "learning_rate": 1.5330471064884803e-05,
      "loss": 0.5427,
      "step": 4200
    },
    {
      "epoch": 0.7555555555555555,
      "grad_norm": 51.25,
      "learning_rate": 1.52700006047046e-05,
      "loss": 0.5796,
      "step": 4250
    },
    {
      "epoch": 0.7644444444444445,
      "grad_norm": 42.25,
      "learning_rate": 1.52095301445244e-05,
      "loss": 0.5196,
      "step": 4300
    },
    {
      "epoch": 0.7733333333333333,
      "grad_norm": 34.75,
      "learning_rate": 1.5149059684344198e-05,
      "loss": 0.5521,
      "step": 4350
    },
    {
      "epoch": 0.7822222222222223,
      "grad_norm": 26.75,
      "learning_rate": 1.5088589224163996e-05,
      "loss": 0.5732,
      "step": 4400
    },
    {
      "epoch": 0.7911111111111111,
      "grad_norm": 53.25,
      "learning_rate": 1.5028118763983795e-05,
      "loss": 0.5951,
      "step": 4450
    },
    {
      "epoch": 0.8,
      "grad_norm": 27.875,
      "learning_rate": 1.4967648303803593e-05,
      "loss": 0.5393,
      "step": 4500
    },
    {
      "epoch": 0.8,
      "eval_loss": 0.015166623517870903,
      "eval_runtime": 0.2317,
      "eval_samples_per_second": 43.162,
      "eval_steps_per_second": 8.632,
      "step": 4500
    },
    {
      "epoch": 0.8088888888888889,
      "grad_norm": 31.25,
      "learning_rate": 1.4907177843623391e-05,
      "loss": 0.5691,
      "step": 4550
    },
    {
      "epoch": 0.8177777777777778,
      "grad_norm": 28.125,
      "learning_rate": 1.4846707383443188e-05,
      "loss": 0.5535,
      "step": 4600
    },
    {
      "epoch": 0.8266666666666667,
      "grad_norm": 32.75,
      "learning_rate": 1.4786236923262986e-05,
      "loss": 0.5656,
      "step": 4650
    },
    {
      "epoch": 0.8355555555555556,
      "grad_norm": 56.25,
      "learning_rate": 1.4725766463082784e-05,
      "loss": 0.5505,
      "step": 4700
    },
    {
      "epoch": 0.8444444444444444,
      "grad_norm": 73.0,
      "learning_rate": 1.4665296002902583e-05,
      "loss": 0.585,
      "step": 4750
    },
    {
      "epoch": 0.8533333333333334,
      "grad_norm": 55.5,
      "learning_rate": 1.4604825542722381e-05,
      "loss": 0.5366,
      "step": 4800
    },
    {
      "epoch": 0.8622222222222222,
      "grad_norm": 48.75,
      "learning_rate": 1.454435508254218e-05,
      "loss": 0.5466,
      "step": 4850
    },
    {
      "epoch": 0.8711111111111111,
      "grad_norm": 47.75,
      "learning_rate": 1.4483884622361976e-05,
      "loss": 0.5458,
      "step": 4900
    },
    {
      "epoch": 0.88,
      "grad_norm": 42.75,
      "learning_rate": 1.4423414162181774e-05,
      "loss": 0.5499,
      "step": 4950
    },
    {
      "epoch": 0.8888888888888888,
      "grad_norm": 62.0,
      "learning_rate": 1.4362943702001573e-05,
      "loss": 0.5606,
      "step": 5000
    },
    {
      "epoch": 0.8888888888888888,
      "eval_loss": 0.014741213992238045,
      "eval_runtime": 0.2391,
      "eval_samples_per_second": 41.821,
      "eval_steps_per_second": 8.364,
      "step": 5000
    },
    {
      "epoch": 0.8977777777777778,
      "grad_norm": 49.5,
      "learning_rate": 1.4302473241821371e-05,
      "loss": 0.542,
      "step": 5050
    },
    {
      "epoch": 0.9066666666666666,
      "grad_norm": 61.0,
      "learning_rate": 1.424200278164117e-05,
      "loss": 0.5337,
      "step": 5100
    },
    {
      "epoch": 0.9155555555555556,
      "grad_norm": 27.625,
      "learning_rate": 1.4181532321460968e-05,
      "loss": 0.4814,
      "step": 5150
    },
    {
      "epoch": 0.9244444444444444,
      "grad_norm": 51.75,
      "learning_rate": 1.4121061861280764e-05,
      "loss": 0.5558,
      "step": 5200
    },
    {
      "epoch": 0.9333333333333333,
      "grad_norm": 38.0,
      "learning_rate": 1.4060591401100562e-05,
      "loss": 0.563,
      "step": 5250
    },
    {
      "epoch": 0.9422222222222222,
      "grad_norm": 53.5,
      "learning_rate": 1.400012094092036e-05,
      "loss": 0.5361,
      "step": 5300
    },
    {
      "epoch": 0.9511111111111111,
      "grad_norm": 55.5,
      "learning_rate": 1.3939650480740159e-05,
      "loss": 0.5272,
      "step": 5350
    },
    {
      "epoch": 0.96,
      "grad_norm": 39.5,
      "learning_rate": 1.3879180020559957e-05,
      "loss": 0.5477,
      "step": 5400
    },
    {
      "epoch": 0.9688888888888889,
      "grad_norm": 44.25,
      "learning_rate": 1.3818709560379756e-05,
      "loss": 0.5576,
      "step": 5450
    },
    {
      "epoch": 0.9777777777777777,
      "grad_norm": 49.25,
      "learning_rate": 1.3758239100199552e-05,
      "loss": 0.5449,
      "step": 5500
    },
    {
      "epoch": 0.9777777777777777,
      "eval_loss": 0.017370671033859253,
      "eval_runtime": 0.2343,
      "eval_samples_per_second": 42.681,
      "eval_steps_per_second": 8.536,
      "step": 5500
    },
    {
      "epoch": 0.9866666666666667,
      "grad_norm": 51.5,
      "learning_rate": 1.369776864001935e-05,
      "loss": 0.5486,
      "step": 5550
    },
    {
      "epoch": 0.9955555555555555,
      "grad_norm": 45.0,
      "learning_rate": 1.3637298179839149e-05,
      "loss": 0.5165,
      "step": 5600
    },
    {
      "epoch": 1.0044444444444445,
      "grad_norm": 31.375,
      "learning_rate": 1.3576827719658947e-05,
      "loss": 0.4315,
      "step": 5650
    },
    {
      "epoch": 1.0133333333333334,
      "grad_norm": 59.0,
      "learning_rate": 1.3516357259478746e-05,
      "loss": 0.2956,
      "step": 5700
    },
    {
      "epoch": 1.0222222222222221,
      "grad_norm": 43.25,
      "learning_rate": 1.3455886799298544e-05,
      "loss": 0.2986,
      "step": 5750
    },
    {
      "epoch": 1.031111111111111,
      "grad_norm": 34.25,
      "learning_rate": 1.339541633911834e-05,
      "loss": 0.2607,
      "step": 5800
    },
    {
      "epoch": 1.04,
      "grad_norm": 39.5,
      "learning_rate": 1.3334945878938139e-05,
      "loss": 0.321,
      "step": 5850
    },
    {
      "epoch": 1.048888888888889,
      "grad_norm": 32.25,
      "learning_rate": 1.3274475418757937e-05,
      "loss": 0.3027,
      "step": 5900
    },
    {
      "epoch": 1.0577777777777777,
      "grad_norm": 60.25,
      "learning_rate": 1.3214004958577735e-05,
      "loss": 0.3313,
      "step": 5950
    },
    {
      "epoch": 1.0666666666666667,
      "grad_norm": 43.75,
      "learning_rate": 1.3153534498397534e-05,
      "loss": 0.302,
      "step": 6000
    },
    {
      "epoch": 1.0666666666666667,
      "eval_loss": 0.013352945446968079,
      "eval_runtime": 0.2372,
      "eval_samples_per_second": 42.164,
      "eval_steps_per_second": 8.433,
      "step": 6000
    },
    {
      "epoch": 1.0755555555555556,
      "grad_norm": 32.75,
      "learning_rate": 1.3093064038217332e-05,
      "loss": 0.3092,
      "step": 6050
    },
    {
      "epoch": 1.0844444444444445,
      "grad_norm": 28.75,
      "learning_rate": 1.3032593578037129e-05,
      "loss": 0.3224,
      "step": 6100
    },
    {
      "epoch": 1.0933333333333333,
      "grad_norm": 48.0,
      "learning_rate": 1.2972123117856927e-05,
      "loss": 0.3313,
      "step": 6150
    },
    {
      "epoch": 1.1022222222222222,
      "grad_norm": 39.0,
      "learning_rate": 1.2911652657676725e-05,
      "loss": 0.321,
      "step": 6200
    },
    {
      "epoch": 1.1111111111111112,
      "grad_norm": 608.0,
      "learning_rate": 1.2851182197496524e-05,
      "loss": 0.3081,
      "step": 6250
    },
    {
      "epoch": 1.12,
      "grad_norm": 33.75,
      "learning_rate": 1.2790711737316322e-05,
      "loss": 0.3186,
      "step": 6300
    },
    {
      "epoch": 1.1288888888888888,
      "grad_norm": 24.625,
      "learning_rate": 1.273024127713612e-05,
      "loss": 0.3168,
      "step": 6350
    },
    {
      "epoch": 1.1377777777777778,
      "grad_norm": 35.25,
      "learning_rate": 1.2669770816955917e-05,
      "loss": 0.2926,
      "step": 6400
    },
    {
      "epoch": 1.1466666666666667,
      "grad_norm": 36.0,
      "learning_rate": 1.2609300356775715e-05,
      "loss": 0.3467,
      "step": 6450
    },
    {
      "epoch": 1.1555555555555554,
      "grad_norm": 32.0,
      "learning_rate": 1.2548829896595513e-05,
      "loss": 0.2906,
      "step": 6500
    },
    {
      "epoch": 1.1555555555555554,
      "eval_loss": 0.012937337160110474,
      "eval_runtime": 0.2433,
      "eval_samples_per_second": 41.104,
      "eval_steps_per_second": 8.221,
      "step": 6500
    },
    {
      "epoch": 1.1644444444444444,
      "grad_norm": 36.5,
      "learning_rate": 1.2488359436415312e-05,
      "loss": 0.3198,
      "step": 6550
    },
    {
      "epoch": 1.1733333333333333,
      "grad_norm": 31.625,
      "learning_rate": 1.242788897623511e-05,
      "loss": 0.3089,
      "step": 6600
    },
    {
      "epoch": 1.1822222222222223,
      "grad_norm": 19.0,
      "learning_rate": 1.2367418516054908e-05,
      "loss": 0.3189,
      "step": 6650
    },
    {
      "epoch": 1.1911111111111112,
      "grad_norm": 44.0,
      "learning_rate": 1.2306948055874705e-05,
      "loss": 0.2821,
      "step": 6700
    },
    {
      "epoch": 1.2,
      "grad_norm": 26.375,
      "learning_rate": 1.2246477595694503e-05,
      "loss": 0.3345,
      "step": 6750
    },
    {
      "epoch": 1.208888888888889,
      "grad_norm": 34.0,
      "learning_rate": 1.2186007135514301e-05,
      "loss": 0.2933,
      "step": 6800
    },
    {
      "epoch": 1.2177777777777778,
      "grad_norm": 51.0,
      "learning_rate": 1.21255366753341e-05,
      "loss": 0.3073,
      "step": 6850
    },
    {
      "epoch": 1.2266666666666666,
      "grad_norm": 36.75,
      "learning_rate": 1.2065066215153898e-05,
      "loss": 0.2922,
      "step": 6900
    },
    {
      "epoch": 1.2355555555555555,
      "grad_norm": 30.875,
      "learning_rate": 1.2004595754973696e-05,
      "loss": 0.346,
      "step": 6950
    },
    {
      "epoch": 1.2444444444444445,
      "grad_norm": 39.75,
      "learning_rate": 1.1944125294793493e-05,
      "loss": 0.2795,
      "step": 7000
    },
    {
      "epoch": 1.2444444444444445,
      "eval_loss": 0.013419872149825096,
      "eval_runtime": 0.2304,
      "eval_samples_per_second": 43.394,
      "eval_steps_per_second": 8.679,
      "step": 7000
    },
    {
      "epoch": 1.2533333333333334,
      "grad_norm": 26.375,
      "learning_rate": 1.1883654834613291e-05,
      "loss": 0.3054,
      "step": 7050
    },
    {
      "epoch": 1.2622222222222224,
      "grad_norm": 39.75,
      "learning_rate": 1.182318437443309e-05,
      "loss": 0.3053,
      "step": 7100
    },
    {
      "epoch": 1.271111111111111,
      "grad_norm": 46.25,
      "learning_rate": 1.1762713914252888e-05,
      "loss": 0.3227,
      "step": 7150
    },
    {
      "epoch": 1.28,
      "grad_norm": 20.25,
      "learning_rate": 1.1702243454072686e-05,
      "loss": 0.3362,
      "step": 7200
    },
    {
      "epoch": 1.2888888888888888,
      "grad_norm": 53.5,
      "learning_rate": 1.1641772993892485e-05,
      "loss": 0.2854,
      "step": 7250
    },
    {
      "epoch": 1.2977777777777777,
      "grad_norm": 63.25,
      "learning_rate": 1.1581302533712281e-05,
      "loss": 0.3259,
      "step": 7300
    },
    {
      "epoch": 1.3066666666666666,
      "grad_norm": 35.25,
      "learning_rate": 1.152083207353208e-05,
      "loss": 0.3024,
      "step": 7350
    },
    {
      "epoch": 1.3155555555555556,
      "grad_norm": 41.5,
      "learning_rate": 1.1460361613351878e-05,
      "loss": 0.3036,
      "step": 7400
    },
    {
      "epoch": 1.3244444444444445,
      "grad_norm": 30.75,
      "learning_rate": 1.1399891153171676e-05,
      "loss": 0.3375,
      "step": 7450
    },
    {
      "epoch": 1.3333333333333333,
      "grad_norm": 40.25,
      "learning_rate": 1.1339420692991474e-05,
      "loss": 0.304,
      "step": 7500
    },
    {
      "epoch": 1.3333333333333333,
      "eval_loss": 0.011497000232338905,
      "eval_runtime": 0.2306,
      "eval_samples_per_second": 43.364,
      "eval_steps_per_second": 8.673,
      "step": 7500
    },
    {
      "epoch": 1.3422222222222222,
      "grad_norm": 31.875,
      "learning_rate": 1.1278950232811273e-05,
      "loss": 0.3073,
      "step": 7550
    },
    {
      "epoch": 1.3511111111111112,
      "grad_norm": 27.0,
      "learning_rate": 1.121847977263107e-05,
      "loss": 0.3235,
      "step": 7600
    },
    {
      "epoch": 1.3599999999999999,
      "grad_norm": 43.75,
      "learning_rate": 1.1158009312450868e-05,
      "loss": 0.3392,
      "step": 7650
    },
    {
      "epoch": 1.3688888888888888,
      "grad_norm": 45.75,
      "learning_rate": 1.1097538852270666e-05,
      "loss": 0.3393,
      "step": 7700
    },
    {
      "epoch": 1.3777777777777778,
      "grad_norm": 58.25,
      "learning_rate": 1.1037068392090464e-05,
      "loss": 0.342,
      "step": 7750
    },
    {
      "epoch": 1.3866666666666667,
      "grad_norm": 40.5,
      "learning_rate": 1.0976597931910263e-05,
      "loss": 0.3167,
      "step": 7800
    },
    {
      "epoch": 1.3955555555555557,
      "grad_norm": 40.25,
      "learning_rate": 1.0916127471730061e-05,
      "loss": 0.2775,
      "step": 7850
    },
    {
      "epoch": 1.4044444444444444,
      "grad_norm": 31.0,
      "learning_rate": 1.0855657011549857e-05,
      "loss": 0.2936,
      "step": 7900
    },
    {
      "epoch": 1.4133333333333333,
      "grad_norm": 41.5,
      "learning_rate": 1.0795186551369656e-05,
      "loss": 0.3132,
      "step": 7950
    },
    {
      "epoch": 1.4222222222222223,
      "grad_norm": 79.5,
      "learning_rate": 1.0734716091189454e-05,
      "loss": 0.3007,
      "step": 8000
    },
    {
      "epoch": 1.4222222222222223,
      "eval_loss": 0.010556383058428764,
      "eval_runtime": 0.2401,
      "eval_samples_per_second": 41.645,
      "eval_steps_per_second": 8.329,
      "step": 8000
    },
    {
      "epoch": 1.431111111111111,
      "grad_norm": 23.875,
      "learning_rate": 1.0674245631009252e-05,
      "loss": 0.3166,
      "step": 8050
    },
    {
      "epoch": 1.44,
      "grad_norm": 26.625,
      "learning_rate": 1.061377517082905e-05,
      "loss": 0.3079,
      "step": 8100
    },
    {
      "epoch": 1.448888888888889,
      "grad_norm": 52.5,
      "learning_rate": 1.0553304710648849e-05,
      "loss": 0.3107,
      "step": 8150
    },
    {
      "epoch": 1.4577777777777778,
      "grad_norm": 54.0,
      "learning_rate": 1.0492834250468646e-05,
      "loss": 0.3267,
      "step": 8200
    },
    {
      "epoch": 1.4666666666666668,
      "grad_norm": 41.25,
      "learning_rate": 1.0432363790288444e-05,
      "loss": 0.3177,
      "step": 8250
    },
    {
      "epoch": 1.4755555555555555,
      "grad_norm": 42.5,
      "learning_rate": 1.0371893330108242e-05,
      "loss": 0.3303,
      "step": 8300
    },
    {
      "epoch": 1.4844444444444445,
      "grad_norm": 37.5,
      "learning_rate": 1.031142286992804e-05,
      "loss": 0.325,
      "step": 8350
    },
    {
      "epoch": 1.4933333333333334,
      "grad_norm": 36.25,
      "learning_rate": 1.0250952409747839e-05,
      "loss": 0.2704,
      "step": 8400
    },
    {
      "epoch": 1.5022222222222221,
      "grad_norm": 30.625,
      "learning_rate": 1.0190481949567637e-05,
      "loss": 0.3716,
      "step": 8450
    },
    {
      "epoch": 1.511111111111111,
      "grad_norm": 57.25,
      "learning_rate": 1.0130011489387434e-05,
      "loss": 0.3312,
      "step": 8500
    },
    {
      "epoch": 1.511111111111111,
      "eval_loss": 0.008495492860674858,
      "eval_runtime": 0.2373,
      "eval_samples_per_second": 42.14,
      "eval_steps_per_second": 8.428,
      "step": 8500
    },
    {
      "epoch": 1.52,
      "grad_norm": 37.5,
      "learning_rate": 1.0069541029207232e-05,
      "loss": 0.3069,
      "step": 8550
    },
    {
      "epoch": 1.528888888888889,
      "grad_norm": 37.25,
      "learning_rate": 1.000907056902703e-05,
      "loss": 0.3316,
      "step": 8600
    },
    {
      "epoch": 1.537777777777778,
      "grad_norm": 35.25,
      "learning_rate": 9.948600108846829e-06,
      "loss": 0.306,
      "step": 8650
    },
    {
      "epoch": 1.5466666666666666,
      "grad_norm": 41.75,
      "learning_rate": 9.888129648666627e-06,
      "loss": 0.3417,
      "step": 8700
    },
    {
      "epoch": 1.5555555555555556,
      "grad_norm": 154.0,
      "learning_rate": 9.827659188486425e-06,
      "loss": 0.3217,
      "step": 8750
    },
    {
      "epoch": 1.5644444444444443,
      "grad_norm": 42.0,
      "learning_rate": 9.767188728306224e-06,
      "loss": 0.304,
      "step": 8800
    },
    {
      "epoch": 1.5733333333333333,
      "grad_norm": 26.125,
      "learning_rate": 9.70671826812602e-06,
      "loss": 0.3119,
      "step": 8850
    },
    {
      "epoch": 1.5822222222222222,
      "grad_norm": 24.375,
      "learning_rate": 9.646247807945819e-06,
      "loss": 0.3162,
      "step": 8900
    },
    {
      "epoch": 1.5911111111111111,
      "grad_norm": 17.75,
      "learning_rate": 9.585777347765617e-06,
      "loss": 0.3566,
      "step": 8950
    },
    {
      "epoch": 1.6,
      "grad_norm": 28.75,
      "learning_rate": 9.525306887585415e-06,
      "loss": 0.2844,
      "step": 9000
    },
    {
      "epoch": 1.6,
      "eval_loss": 0.010613194666802883,
      "eval_runtime": 0.2357,
      "eval_samples_per_second": 42.431,
      "eval_steps_per_second": 8.486,
      "step": 9000
    },
    {
      "epoch": 1.608888888888889,
      "grad_norm": 25.625,
      "learning_rate": 9.464836427405213e-06,
      "loss": 0.2979,
      "step": 9050
    },
    {
      "epoch": 1.6177777777777778,
      "grad_norm": 33.0,
      "learning_rate": 9.404365967225012e-06,
      "loss": 0.3155,
      "step": 9100
    },
    {
      "epoch": 1.6266666666666667,
      "grad_norm": 25.625,
      "learning_rate": 9.343895507044808e-06,
      "loss": 0.3136,
      "step": 9150
    },
    {
      "epoch": 1.6355555555555554,
      "grad_norm": 38.75,
      "learning_rate": 9.283425046864607e-06,
      "loss": 0.3223,
      "step": 9200
    },
    {
      "epoch": 1.6444444444444444,
      "grad_norm": 40.5,
      "learning_rate": 9.222954586684405e-06,
      "loss": 0.3562,
      "step": 9250
    },
    {
      "epoch": 1.6533333333333333,
      "grad_norm": 29.375,
      "learning_rate": 9.162484126504203e-06,
      "loss": 0.3067,
      "step": 9300
    },
    {
      "epoch": 1.6622222222222223,
      "grad_norm": 44.25,
      "learning_rate": 9.102013666324002e-06,
      "loss": 0.3332,
      "step": 9350
    },
    {
      "epoch": 1.6711111111111112,
      "grad_norm": 28.5,
      "learning_rate": 9.0415432061438e-06,
      "loss": 0.2975,
      "step": 9400
    },
    {
      "epoch": 1.6800000000000002,
      "grad_norm": 40.25,
      "learning_rate": 8.981072745963597e-06,
      "loss": 0.3248,
      "step": 9450
    },
    {
      "epoch": 1.6888888888888889,
      "grad_norm": 28.375,
      "learning_rate": 8.920602285783395e-06,
      "loss": 0.3053,
      "step": 9500
    },
    {
      "epoch": 1.6888888888888889,
      "eval_loss": 0.010935052298009396,
      "eval_runtime": 0.2331,
      "eval_samples_per_second": 42.905,
      "eval_steps_per_second": 8.581,
      "step": 9500
    },
    {
      "epoch": 1.6977777777777778,
      "grad_norm": 28.875,
      "learning_rate": 8.860131825603193e-06,
      "loss": 0.3015,
      "step": 9550
    },
    {
      "epoch": 1.7066666666666666,
      "grad_norm": 30.75,
      "learning_rate": 8.799661365422991e-06,
      "loss": 0.3051,
      "step": 9600
    },
    {
      "epoch": 1.7155555555555555,
      "grad_norm": 41.25,
      "learning_rate": 8.73919090524279e-06,
      "loss": 0.2987,
      "step": 9650
    },
    {
      "epoch": 1.7244444444444444,
      "grad_norm": 24.125,
      "learning_rate": 8.678720445062588e-06,
      "loss": 0.3083,
      "step": 9700
    },
    {
      "epoch": 1.7333333333333334,
      "grad_norm": 41.75,
      "learning_rate": 8.618249984882385e-06,
      "loss": 0.2924,
      "step": 9750
    },
    {
      "epoch": 1.7422222222222223,
      "grad_norm": 35.5,
      "learning_rate": 8.557779524702183e-06,
      "loss": 0.3256,
      "step": 9800
    },
    {
      "epoch": 1.751111111111111,
      "grad_norm": 26.875,
      "learning_rate": 8.497309064521981e-06,
      "loss": 0.317,
      "step": 9850
    },
    {
      "epoch": 1.76,
      "grad_norm": 32.25,
      "learning_rate": 8.43683860434178e-06,
      "loss": 0.3248,
      "step": 9900
    },
    {
      "epoch": 1.7688888888888887,
      "grad_norm": 28.875,
      "learning_rate": 8.376368144161578e-06,
      "loss": 0.311,
      "step": 9950
    },
    {
      "epoch": 1.7777777777777777,
      "grad_norm": 36.0,
      "learning_rate": 8.315897683981376e-06,
      "loss": 0.3339,
      "step": 10000
    },
    {
      "epoch": 1.7777777777777777,
      "eval_loss": 0.008871009573340416,
      "eval_runtime": 0.238,
      "eval_samples_per_second": 42.021,
      "eval_steps_per_second": 8.404,
      "step": 10000
    },
    {
      "epoch": 1.7866666666666666,
      "grad_norm": 30.5,
      "learning_rate": 8.255427223801173e-06,
      "loss": 0.3088,
      "step": 10050
    },
    {
      "epoch": 1.7955555555555556,
      "grad_norm": 21.875,
      "learning_rate": 8.194956763620971e-06,
      "loss": 0.3369,
      "step": 10100
    },
    {
      "epoch": 1.8044444444444445,
      "grad_norm": 30.125,
      "learning_rate": 8.13448630344077e-06,
      "loss": 0.2748,
      "step": 10150
    },
    {
      "epoch": 1.8133333333333335,
      "grad_norm": 39.25,
      "learning_rate": 8.074015843260568e-06,
      "loss": 0.3141,
      "step": 10200
    },
    {
      "epoch": 1.8222222222222222,
      "grad_norm": 72.5,
      "learning_rate": 8.013545383080366e-06,
      "loss": 0.2874,
      "step": 10250
    },
    {
      "epoch": 1.8311111111111111,
      "grad_norm": 36.5,
      "learning_rate": 7.953074922900164e-06,
      "loss": 0.2908,
      "step": 10300
    },
    {
      "epoch": 1.8399999999999999,
      "grad_norm": 51.0,
      "learning_rate": 7.892604462719961e-06,
      "loss": 0.281,
      "step": 10350
    },
    {
      "epoch": 1.8488888888888888,
      "grad_norm": 34.5,
      "learning_rate": 7.83213400253976e-06,
      "loss": 0.3403,
      "step": 10400
    },
    {
      "epoch": 1.8577777777777778,
      "grad_norm": 33.5,
      "learning_rate": 7.771663542359558e-06,
      "loss": 0.2971,
      "step": 10450
    },
    {
      "epoch": 1.8666666666666667,
      "grad_norm": 47.5,
      "learning_rate": 7.711193082179356e-06,
      "loss": 0.3233,
      "step": 10500
    },
    {
      "epoch": 1.8666666666666667,
      "eval_loss": 0.008435798808932304,
      "eval_runtime": 0.2404,
      "eval_samples_per_second": 41.595,
      "eval_steps_per_second": 8.319,
      "step": 10500
    },
    {
      "epoch": 1.8755555555555556,
      "grad_norm": 45.75,
      "learning_rate": 7.650722621999154e-06,
      "loss": 0.307,
      "step": 10550
    },
    {
      "epoch": 1.8844444444444446,
      "grad_norm": 34.0,
      "learning_rate": 7.590252161818952e-06,
      "loss": 0.3293,
      "step": 10600
    },
    {
      "epoch": 1.8933333333333333,
      "grad_norm": 51.25,
      "learning_rate": 7.52978170163875e-06,
      "loss": 0.3265,
      "step": 10650
    },
    {
      "epoch": 1.9022222222222223,
      "grad_norm": 33.25,
      "learning_rate": 7.4693112414585474e-06,
      "loss": 0.3356,
      "step": 10700
    },
    {
      "epoch": 1.911111111111111,
      "grad_norm": 45.25,
      "learning_rate": 7.408840781278346e-06,
      "loss": 0.303,
      "step": 10750
    },
    {
      "epoch": 1.92,
      "grad_norm": 40.75,
      "learning_rate": 7.348370321098144e-06,
      "loss": 0.3251,
      "step": 10800
    },
    {
      "epoch": 1.9288888888888889,
      "grad_norm": 32.0,
      "learning_rate": 7.2878998609179415e-06,
      "loss": 0.2962,
      "step": 10850
    },
    {
      "epoch": 1.9377777777777778,
      "grad_norm": 29.0,
      "learning_rate": 7.22742940073774e-06,
      "loss": 0.3057,
      "step": 10900
    },
    {
      "epoch": 1.9466666666666668,
      "grad_norm": 34.25,
      "learning_rate": 7.166958940557538e-06,
      "loss": 0.3112,
      "step": 10950
    },
    {
      "epoch": 1.9555555555555557,
      "grad_norm": 27.5,
      "learning_rate": 7.106488480377336e-06,
      "loss": 0.323,
      "step": 11000
    },
    {
      "epoch": 1.9555555555555557,
      "eval_loss": 0.008191155269742012,
      "eval_runtime": 0.2402,
      "eval_samples_per_second": 41.624,
      "eval_steps_per_second": 8.325,
      "step": 11000
    },
    {
      "epoch": 1.9644444444444444,
      "grad_norm": 48.5,
      "learning_rate": 7.046018020197134e-06,
      "loss": 0.3373,
      "step": 11050
    },
    {
      "epoch": 1.9733333333333334,
      "grad_norm": 26.375,
      "learning_rate": 6.985547560016932e-06,
      "loss": 0.307,
      "step": 11100
    },
    {
      "epoch": 1.982222222222222,
      "grad_norm": 29.125,
      "learning_rate": 6.92507709983673e-06,
      "loss": 0.3005,
      "step": 11150
    },
    {
      "epoch": 1.991111111111111,
      "grad_norm": 44.25,
      "learning_rate": 6.864606639656528e-06,
      "loss": 0.3017,
      "step": 11200
    },
    {
      "epoch": 2.0,
      "grad_norm": 35.0,
      "learning_rate": 6.804136179476326e-06,
      "loss": 0.3092,
      "step": 11250
    },
    {
      "epoch": 2.008888888888889,
      "grad_norm": 30.625,
      "learning_rate": 6.743665719296124e-06,
      "loss": 0.2326,
      "step": 11300
    },
    {
      "epoch": 2.017777777777778,
      "grad_norm": 25.625,
      "learning_rate": 6.683195259115922e-06,
      "loss": 0.2262,
      "step": 11350
    },
    {
      "epoch": 2.026666666666667,
      "grad_norm": 37.5,
      "learning_rate": 6.62272479893572e-06,
      "loss": 0.2507,
      "step": 11400
    },
    {
      "epoch": 2.0355555555555553,
      "grad_norm": 28.75,
      "learning_rate": 6.562254338755518e-06,
      "loss": 0.2335,
      "step": 11450
    },
    {
      "epoch": 2.0444444444444443,
      "grad_norm": 33.5,
      "learning_rate": 6.501783878575316e-06,
      "loss": 0.2481,
      "step": 11500
    },
    {
      "epoch": 2.0444444444444443,
      "eval_loss": 0.008659908547997475,
      "eval_runtime": 0.2316,
      "eval_samples_per_second": 43.172,
      "eval_steps_per_second": 8.634,
      "step": 11500
    },
    {
      "epoch": 2.0533333333333332,
      "grad_norm": 19.375,
      "learning_rate": 6.4413134183951144e-06,
      "loss": 0.2288,
      "step": 11550
    },
    {
      "epoch": 2.062222222222222,
      "grad_norm": 33.25,
      "learning_rate": 6.380842958214912e-06,
      "loss": 0.2372,
      "step": 11600
    },
    {
      "epoch": 2.071111111111111,
      "grad_norm": 17.25,
      "learning_rate": 6.32037249803471e-06,
      "loss": 0.2476,
      "step": 11650
    },
    {
      "epoch": 2.08,
      "grad_norm": 29.5,
      "learning_rate": 6.2599020378545085e-06,
      "loss": 0.2467,
      "step": 11700
    },
    {
      "epoch": 2.088888888888889,
      "grad_norm": 22.375,
      "learning_rate": 6.199431577674306e-06,
      "loss": 0.2503,
      "step": 11750
    },
    {
      "epoch": 2.097777777777778,
      "grad_norm": 35.5,
      "learning_rate": 6.138961117494104e-06,
      "loss": 0.249,
      "step": 11800
    },
    {
      "epoch": 2.1066666666666665,
      "grad_norm": 38.25,
      "learning_rate": 6.078490657313903e-06,
      "loss": 0.2392,
      "step": 11850
    },
    {
      "epoch": 2.1155555555555554,
      "grad_norm": 40.75,
      "learning_rate": 6.0180201971337e-06,
      "loss": 0.2451,
      "step": 11900
    },
    {
      "epoch": 2.1244444444444444,
      "grad_norm": 17.875,
      "learning_rate": 5.957549736953498e-06,
      "loss": 0.2744,
      "step": 11950
    },
    {
      "epoch": 2.1333333333333333,
      "grad_norm": 18.25,
      "learning_rate": 5.897079276773297e-06,
      "loss": 0.2577,
      "step": 12000
    },
    {
      "epoch": 2.1333333333333333,
      "eval_loss": 0.0083710215985775,
      "eval_runtime": 0.2415,
      "eval_samples_per_second": 41.406,
      "eval_steps_per_second": 8.281,
      "step": 12000
    },
    {
      "epoch": 2.1422222222222222,
      "grad_norm": 18.0,
      "learning_rate": 5.836608816593094e-06,
      "loss": 0.2294,
      "step": 12050
    },
    {
      "epoch": 2.151111111111111,
      "grad_norm": 42.5,
      "learning_rate": 5.776138356412892e-06,
      "loss": 0.2363,
      "step": 12100
    },
    {
      "epoch": 2.16,
      "grad_norm": 26.5,
      "learning_rate": 5.715667896232691e-06,
      "loss": 0.2335,
      "step": 12150
    },
    {
      "epoch": 2.168888888888889,
      "grad_norm": 41.5,
      "learning_rate": 5.655197436052488e-06,
      "loss": 0.259,
      "step": 12200
    },
    {
      "epoch": 2.1777777777777776,
      "grad_norm": 29.875,
      "learning_rate": 5.5947269758722865e-06,
      "loss": 0.2568,
      "step": 12250
    },
    {
      "epoch": 2.1866666666666665,
      "grad_norm": 57.0,
      "learning_rate": 5.534256515692085e-06,
      "loss": 0.2489,
      "step": 12300
    },
    {
      "epoch": 2.1955555555555555,
      "grad_norm": 49.0,
      "learning_rate": 5.473786055511883e-06,
      "loss": 0.2464,
      "step": 12350
    },
    {
      "epoch": 2.2044444444444444,
      "grad_norm": 27.375,
      "learning_rate": 5.4133155953316806e-06,
      "loss": 0.2362,
      "step": 12400
    },
    {
      "epoch": 2.2133333333333334,
      "grad_norm": 41.25,
      "learning_rate": 5.352845135151479e-06,
      "loss": 0.2397,
      "step": 12450
    },
    {
      "epoch": 2.2222222222222223,
      "grad_norm": 38.0,
      "learning_rate": 5.292374674971277e-06,
      "loss": 0.2569,
      "step": 12500
    },
    {
      "epoch": 2.2222222222222223,
      "eval_loss": 0.008943794295191765,
      "eval_runtime": 0.2377,
      "eval_samples_per_second": 42.063,
      "eval_steps_per_second": 8.413,
      "step": 12500
    },
    {
      "epoch": 2.2311111111111113,
      "grad_norm": 14.75,
      "learning_rate": 5.231904214791075e-06,
      "loss": 0.252,
      "step": 12550
    },
    {
      "epoch": 2.24,
      "grad_norm": 20.125,
      "learning_rate": 5.171433754610873e-06,
      "loss": 0.2718,
      "step": 12600
    },
    {
      "epoch": 2.2488888888888887,
      "grad_norm": 30.125,
      "learning_rate": 5.110963294430671e-06,
      "loss": 0.2743,
      "step": 12650
    },
    {
      "epoch": 2.2577777777777777,
      "grad_norm": 45.75,
      "learning_rate": 5.050492834250469e-06,
      "loss": 0.273,
      "step": 12700
    },
    {
      "epoch": 2.2666666666666666,
      "grad_norm": 29.25,
      "learning_rate": 4.990022374070267e-06,
      "loss": 0.2208,
      "step": 12750
    },
    {
      "epoch": 2.2755555555555556,
      "grad_norm": 50.25,
      "learning_rate": 4.929551913890065e-06,
      "loss": 0.2514,
      "step": 12800
    },
    {
      "epoch": 2.2844444444444445,
      "grad_norm": 30.125,
      "learning_rate": 4.869081453709863e-06,
      "loss": 0.2246,
      "step": 12850
    },
    {
      "epoch": 2.2933333333333334,
      "grad_norm": 60.5,
      "learning_rate": 4.808610993529661e-06,
      "loss": 0.2303,
      "step": 12900
    },
    {
      "epoch": 2.3022222222222224,
      "grad_norm": 37.25,
      "learning_rate": 4.748140533349459e-06,
      "loss": 0.2476,
      "step": 12950
    },
    {
      "epoch": 2.311111111111111,
      "grad_norm": 26.875,
      "learning_rate": 4.687670073169257e-06,
      "loss": 0.2241,
      "step": 13000
    },
    {
      "epoch": 2.311111111111111,
      "eval_loss": 0.008464875631034374,
      "eval_runtime": 0.2287,
      "eval_samples_per_second": 43.727,
      "eval_steps_per_second": 8.745,
      "step": 13000
    },
    {
      "epoch": 2.32,
      "grad_norm": 37.0,
      "learning_rate": 4.627199612989055e-06,
      "loss": 0.2368,
      "step": 13050
    },
    {
      "epoch": 2.328888888888889,
      "grad_norm": 30.375,
      "learning_rate": 4.5667291528088535e-06,
      "loss": 0.2178,
      "step": 13100
    },
    {
      "epoch": 2.3377777777777777,
      "grad_norm": 45.5,
      "learning_rate": 4.506258692628651e-06,
      "loss": 0.2493,
      "step": 13150
    },
    {
      "epoch": 2.3466666666666667,
      "grad_norm": 36.25,
      "learning_rate": 4.445788232448449e-06,
      "loss": 0.235,
      "step": 13200
    },
    {
      "epoch": 2.3555555555555556,
      "grad_norm": 32.25,
      "learning_rate": 4.3853177722682476e-06,
      "loss": 0.2474,
      "step": 13250
    },
    {
      "epoch": 2.3644444444444446,
      "grad_norm": 27.875,
      "learning_rate": 4.324847312088045e-06,
      "loss": 0.2374,
      "step": 13300
    },
    {
      "epoch": 2.3733333333333335,
      "grad_norm": 13.0625,
      "learning_rate": 4.264376851907843e-06,
      "loss": 0.2404,
      "step": 13350
    },
    {
      "epoch": 2.3822222222222225,
      "grad_norm": 34.25,
      "learning_rate": 4.203906391727642e-06,
      "loss": 0.2331,
      "step": 13400
    },
    {
      "epoch": 2.391111111111111,
      "grad_norm": 21.375,
      "learning_rate": 4.143435931547439e-06,
      "loss": 0.2332,
      "step": 13450
    },
    {
      "epoch": 2.4,
      "grad_norm": 36.5,
      "learning_rate": 4.082965471367237e-06,
      "loss": 0.2423,
      "step": 13500
    },
    {
      "epoch": 2.4,
      "eval_loss": 0.008836803957819939,
      "eval_runtime": 0.2285,
      "eval_samples_per_second": 43.761,
      "eval_steps_per_second": 8.752,
      "step": 13500
    },
    {
      "epoch": 2.408888888888889,
      "grad_norm": 24.625,
      "learning_rate": 4.022495011187036e-06,
      "loss": 0.2503,
      "step": 13550
    },
    {
      "epoch": 2.417777777777778,
      "grad_norm": 35.75,
      "learning_rate": 3.962024551006833e-06,
      "loss": 0.2484,
      "step": 13600
    },
    {
      "epoch": 2.4266666666666667,
      "grad_norm": 41.75,
      "learning_rate": 3.9015540908266315e-06,
      "loss": 0.26,
      "step": 13650
    },
    {
      "epoch": 2.4355555555555557,
      "grad_norm": 33.25,
      "learning_rate": 3.84108363064643e-06,
      "loss": 0.2402,
      "step": 13700
    },
    {
      "epoch": 2.4444444444444446,
      "grad_norm": 46.25,
      "learning_rate": 3.7806131704662272e-06,
      "loss": 0.2083,
      "step": 13750
    },
    {
      "epoch": 2.453333333333333,
      "grad_norm": 27.875,
      "learning_rate": 3.7201427102860256e-06,
      "loss": 0.214,
      "step": 13800
    },
    {
      "epoch": 2.462222222222222,
      "grad_norm": 21.625,
      "learning_rate": 3.6596722501058234e-06,
      "loss": 0.2677,
      "step": 13850
    },
    {
      "epoch": 2.471111111111111,
      "grad_norm": 30.75,
      "learning_rate": 3.5992017899256213e-06,
      "loss": 0.2371,
      "step": 13900
    },
    {
      "epoch": 2.48,
      "grad_norm": 38.75,
      "learning_rate": 3.5387313297454196e-06,
      "loss": 0.2461,
      "step": 13950
    },
    {
      "epoch": 2.488888888888889,
      "grad_norm": 25.75,
      "learning_rate": 3.4782608695652175e-06,
      "loss": 0.2414,
      "step": 14000
    },
    {
      "epoch": 2.488888888888889,
      "eval_loss": 0.00847357977181673,
      "eval_runtime": 0.2327,
      "eval_samples_per_second": 42.968,
      "eval_steps_per_second": 8.594,
      "step": 14000
    },
    {
      "epoch": 2.497777777777778,
      "grad_norm": 63.0,
      "learning_rate": 3.417790409385016e-06,
      "loss": 0.2657,
      "step": 14050
    },
    {
      "epoch": 2.506666666666667,
      "grad_norm": 49.0,
      "learning_rate": 3.3573199492048137e-06,
      "loss": 0.2417,
      "step": 14100
    },
    {
      "epoch": 2.5155555555555553,
      "grad_norm": 31.625,
      "learning_rate": 3.2968494890246116e-06,
      "loss": 0.2388,
      "step": 14150
    },
    {
      "epoch": 2.5244444444444447,
      "grad_norm": 38.25,
      "learning_rate": 3.23637902884441e-06,
      "loss": 0.2432,
      "step": 14200
    },
    {
      "epoch": 2.533333333333333,
      "grad_norm": 25.75,
      "learning_rate": 3.1759085686642078e-06,
      "loss": 0.2342,
      "step": 14250
    },
    {
      "epoch": 2.542222222222222,
      "grad_norm": 29.875,
      "learning_rate": 3.1154381084840057e-06,
      "loss": 0.2453,
      "step": 14300
    },
    {
      "epoch": 2.551111111111111,
      "grad_norm": 26.875,
      "learning_rate": 3.054967648303804e-06,
      "loss": 0.229,
      "step": 14350
    },
    {
      "epoch": 2.56,
      "grad_norm": 22.0,
      "learning_rate": 2.994497188123602e-06,
      "loss": 0.2452,
      "step": 14400
    },
    {
      "epoch": 2.568888888888889,
      "grad_norm": 38.25,
      "learning_rate": 2.9340267279433997e-06,
      "loss": 0.2435,
      "step": 14450
    },
    {
      "epoch": 2.5777777777777775,
      "grad_norm": 20.0,
      "learning_rate": 2.873556267763198e-06,
      "loss": 0.2353,
      "step": 14500
    },
    {
      "epoch": 2.5777777777777775,
      "eval_loss": 0.008571269921958447,
      "eval_runtime": 0.2385,
      "eval_samples_per_second": 41.921,
      "eval_steps_per_second": 8.384,
      "step": 14500
    },
    {
      "epoch": 2.586666666666667,
      "grad_norm": 37.0,
      "learning_rate": 2.813085807582996e-06,
      "loss": 0.2355,
      "step": 14550
    },
    {
      "epoch": 2.5955555555555554,
      "grad_norm": 25.25,
      "learning_rate": 2.752615347402794e-06,
      "loss": 0.2666,
      "step": 14600
    },
    {
      "epoch": 2.6044444444444443,
      "grad_norm": 51.25,
      "learning_rate": 2.692144887222592e-06,
      "loss": 0.2361,
      "step": 14650
    },
    {
      "epoch": 2.6133333333333333,
      "grad_norm": 21.75,
      "learning_rate": 2.63167442704239e-06,
      "loss": 0.2313,
      "step": 14700
    },
    {
      "epoch": 2.6222222222222222,
      "grad_norm": 36.25,
      "learning_rate": 2.571203966862188e-06,
      "loss": 0.2238,
      "step": 14750
    },
    {
      "epoch": 2.631111111111111,
      "grad_norm": 27.5,
      "learning_rate": 2.510733506681986e-06,
      "loss": 0.2526,
      "step": 14800
    },
    {
      "epoch": 2.64,
      "grad_norm": 42.5,
      "learning_rate": 2.450263046501784e-06,
      "loss": 0.2189,
      "step": 14850
    },
    {
      "epoch": 2.648888888888889,
      "grad_norm": 26.0,
      "learning_rate": 2.389792586321582e-06,
      "loss": 0.23,
      "step": 14900
    },
    {
      "epoch": 2.6577777777777776,
      "grad_norm": 26.875,
      "learning_rate": 2.3293221261413803e-06,
      "loss": 0.2432,
      "step": 14950
    },
    {
      "epoch": 2.6666666666666665,
      "grad_norm": 42.75,
      "learning_rate": 2.268851665961178e-06,
      "loss": 0.2274,
      "step": 15000
    },
    {
      "epoch": 2.6666666666666665,
      "eval_loss": 0.008635434322059155,
      "eval_runtime": 0.2354,
      "eval_samples_per_second": 42.477,
      "eval_steps_per_second": 8.495,
      "step": 15000
    },
    {
      "epoch": 2.6755555555555555,
      "grad_norm": 61.0,
      "learning_rate": 2.208381205780976e-06,
      "loss": 0.2729,
      "step": 15050
    },
    {
      "epoch": 2.6844444444444444,
      "grad_norm": 54.0,
      "learning_rate": 2.1479107456007743e-06,
      "loss": 0.252,
      "step": 15100
    },
    {
      "epoch": 2.6933333333333334,
      "grad_norm": 36.25,
      "learning_rate": 2.0874402854205722e-06,
      "loss": 0.248,
      "step": 15150
    },
    {
      "epoch": 2.7022222222222223,
      "grad_norm": 36.5,
      "learning_rate": 2.02696982524037e-06,
      "loss": 0.2529,
      "step": 15200
    },
    {
      "epoch": 2.7111111111111112,
      "grad_norm": 55.5,
      "learning_rate": 1.9664993650601684e-06,
      "loss": 0.2213,
      "step": 15250
    },
    {
      "epoch": 2.7199999999999998,
      "grad_norm": 26.375,
      "learning_rate": 1.9060289048799663e-06,
      "loss": 0.2291,
      "step": 15300
    },
    {
      "epoch": 2.728888888888889,
      "grad_norm": 19.625,
      "learning_rate": 1.8455584446997644e-06,
      "loss": 0.2448,
      "step": 15350
    },
    {
      "epoch": 2.7377777777777776,
      "grad_norm": 24.75,
      "learning_rate": 1.7850879845195623e-06,
      "loss": 0.2385,
      "step": 15400
    },
    {
      "epoch": 2.7466666666666666,
      "grad_norm": 18.25,
      "learning_rate": 1.7246175243393604e-06,
      "loss": 0.2213,
      "step": 15450
    },
    {
      "epoch": 2.7555555555555555,
      "grad_norm": 21.375,
      "learning_rate": 1.6641470641591585e-06,
      "loss": 0.2616,
      "step": 15500
    },
    {
      "epoch": 2.7555555555555555,
      "eval_loss": 0.008459658361971378,
      "eval_runtime": 0.2312,
      "eval_samples_per_second": 43.248,
      "eval_steps_per_second": 8.65,
      "step": 15500
    },
    {
      "epoch": 2.7644444444444445,
      "grad_norm": 29.125,
      "learning_rate": 1.6036766039789564e-06,
      "loss": 0.2575,
      "step": 15550
    },
    {
      "epoch": 2.7733333333333334,
      "grad_norm": 22.375,
      "learning_rate": 1.5432061437987544e-06,
      "loss": 0.2523,
      "step": 15600
    },
    {
      "epoch": 2.7822222222222224,
      "grad_norm": 12.9375,
      "learning_rate": 1.4827356836185525e-06,
      "loss": 0.2463,
      "step": 15650
    },
    {
      "epoch": 2.7911111111111113,
      "grad_norm": 26.5,
      "learning_rate": 1.4222652234383504e-06,
      "loss": 0.2429,
      "step": 15700
    },
    {
      "epoch": 2.8,
      "grad_norm": 36.25,
      "learning_rate": 1.3617947632581485e-06,
      "loss": 0.2605,
      "step": 15750
    },
    {
      "epoch": 2.8088888888888888,
      "grad_norm": 42.25,
      "learning_rate": 1.3013243030779466e-06,
      "loss": 0.2566,
      "step": 15800
    },
    {
      "epoch": 2.8177777777777777,
      "grad_norm": 16.25,
      "learning_rate": 1.2408538428977445e-06,
      "loss": 0.2234,
      "step": 15850
    },
    {
      "epoch": 2.8266666666666667,
      "grad_norm": 20.625,
      "learning_rate": 1.1803833827175426e-06,
      "loss": 0.2523,
      "step": 15900
    },
    {
      "epoch": 2.8355555555555556,
      "grad_norm": 39.5,
      "learning_rate": 1.1199129225373407e-06,
      "loss": 0.2528,
      "step": 15950
    },
    {
      "epoch": 2.8444444444444446,
      "grad_norm": 30.875,
      "learning_rate": 1.0594424623571386e-06,
      "loss": 0.2304,
      "step": 16000
    },
    {
      "epoch": 2.8444444444444446,
      "eval_loss": 0.008183728903532028,
      "eval_runtime": 0.2301,
      "eval_samples_per_second": 43.458,
      "eval_steps_per_second": 8.692,
      "step": 16000
    },
    {
      "epoch": 2.8533333333333335,
      "grad_norm": 35.5,
      "learning_rate": 9.989720021769367e-07,
      "loss": 0.2496,
      "step": 16050
    },
    {
      "epoch": 2.862222222222222,
      "grad_norm": 38.75,
      "learning_rate": 9.385015419967347e-07,
      "loss": 0.2264,
      "step": 16100
    },
    {
      "epoch": 2.871111111111111,
      "grad_norm": 44.75,
      "learning_rate": 8.780310818165328e-07,
      "loss": 0.2509,
      "step": 16150
    },
    {
      "epoch": 2.88,
      "grad_norm": 42.0,
      "learning_rate": 8.175606216363307e-07,
      "loss": 0.269,
      "step": 16200
    },
    {
      "epoch": 2.888888888888889,
      "grad_norm": 25.5,
      "learning_rate": 7.570901614561288e-07,
      "loss": 0.2497,
      "step": 16250
    },
    {
      "epoch": 2.897777777777778,
      "grad_norm": 30.75,
      "learning_rate": 6.966197012759268e-07,
      "loss": 0.2629,
      "step": 16300
    },
    {
      "epoch": 2.9066666666666667,
      "grad_norm": 29.875,
      "learning_rate": 6.361492410957248e-07,
      "loss": 0.2537,
      "step": 16350
    },
    {
      "epoch": 2.9155555555555557,
      "grad_norm": 39.5,
      "learning_rate": 5.756787809155228e-07,
      "loss": 0.2633,
      "step": 16400
    },
    {
      "epoch": 2.924444444444444,
      "grad_norm": 25.75,
      "learning_rate": 5.152083207353208e-07,
      "loss": 0.2272,
      "step": 16450
    },
    {
      "epoch": 2.9333333333333336,
      "grad_norm": 31.375,
      "learning_rate": 4.5473786055511884e-07,
      "loss": 0.2259,
      "step": 16500
    },
    {
      "epoch": 2.9333333333333336,
      "eval_loss": 0.008398449048399925,
      "eval_runtime": 0.2374,
      "eval_samples_per_second": 42.13,
      "eval_steps_per_second": 8.426,
      "step": 16500
    },
    {
      "epoch": 2.942222222222222,
      "grad_norm": 30.625,
      "learning_rate": 3.942674003749169e-07,
      "loss": 0.2386,
      "step": 16550
    },
    {
      "epoch": 2.951111111111111,
      "grad_norm": 24.625,
      "learning_rate": 3.3379694019471493e-07,
      "loss": 0.2298,
      "step": 16600
    },
    {
      "epoch": 2.96,
      "grad_norm": 40.25,
      "learning_rate": 2.733264800145129e-07,
      "loss": 0.2364,
      "step": 16650
    },
    {
      "epoch": 2.968888888888889,
      "grad_norm": 27.0,
      "learning_rate": 2.1285601983431096e-07,
      "loss": 0.2397,
      "step": 16700
    },
    {
      "epoch": 2.977777777777778,
      "grad_norm": 36.75,
      "learning_rate": 1.5238555965410898e-07,
      "loss": 0.2443,
      "step": 16750
    },
    {
      "epoch": 2.986666666666667,
      "grad_norm": 31.0,
      "learning_rate": 9.191509947390701e-08,
      "loss": 0.2231,
      "step": 16800
    },
    {
      "epoch": 2.9955555555555557,
      "grad_norm": 26.5,
      "learning_rate": 3.144463929370503e-08,
      "loss": 0.2604,
      "step": 16850
    }
  ],
  "logging_steps": 50,
  "max_steps": 16875,
  "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": 1.4299467032690688e+17,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}