{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.353319057815845,
  "global_step": 120000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02,
      "learning_rate": 2.3999999999999997e-05,
      "loss": 0.8845,
      "step": 500
    },
    {
      "epoch": 0.04,
      "learning_rate": 4.7999999999999994e-05,
      "loss": 0.689,
      "step": 1000
    },
    {
      "epoch": 0.04,
      "eval_loss": 0.6793121099472046,
      "eval_runtime": 2.3939,
      "eval_samples_per_second": 959.513,
      "eval_steps_per_second": 15.038,
      "step": 1000
    },
    {
      "epoch": 0.07,
      "learning_rate": 7.199999999999999e-05,
      "loss": 0.6805,
      "step": 1500
    },
    {
      "epoch": 0.09,
      "learning_rate": 9.599999999999999e-05,
      "loss": 0.6802,
      "step": 2000
    },
    {
      "epoch": 0.09,
      "eval_loss": 0.6786620616912842,
      "eval_runtime": 2.1637,
      "eval_samples_per_second": 1061.605,
      "eval_steps_per_second": 16.638,
      "step": 2000
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00011999999999999999,
      "loss": 0.6799,
      "step": 2500
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00014399999999999998,
      "loss": 0.6795,
      "step": 3000
    },
    {
      "epoch": 0.13,
      "eval_loss": 0.6788004040718079,
      "eval_runtime": 2.1379,
      "eval_samples_per_second": 1074.407,
      "eval_steps_per_second": 16.839,
      "step": 3000
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.000168,
      "loss": 0.6791,
      "step": 3500
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00019199999999999998,
      "loss": 0.679,
      "step": 4000
    },
    {
      "epoch": 0.18,
      "eval_loss": 0.6781774163246155,
      "eval_runtime": 2.1652,
      "eval_samples_per_second": 1060.868,
      "eval_steps_per_second": 16.627,
      "step": 4000
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00021599999999999996,
      "loss": 0.6788,
      "step": 4500
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00023999999999999998,
      "loss": 0.6787,
      "step": 5000
    },
    {
      "epoch": 0.22,
      "eval_loss": 0.6782493591308594,
      "eval_runtime": 2.1305,
      "eval_samples_per_second": 1078.133,
      "eval_steps_per_second": 16.897,
      "step": 5000
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00026399999999999997,
      "loss": 0.6786,
      "step": 5500
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.00028799999999999995,
      "loss": 0.6786,
      "step": 6000
    },
    {
      "epoch": 0.27,
      "eval_loss": 0.6780672073364258,
      "eval_runtime": 2.1865,
      "eval_samples_per_second": 1050.561,
      "eval_steps_per_second": 16.465,
      "step": 6000
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.000312,
      "loss": 0.6785,
      "step": 6500
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.000336,
      "loss": 0.6784,
      "step": 7000
    },
    {
      "epoch": 0.31,
      "eval_loss": 0.6781116724014282,
      "eval_runtime": 2.1436,
      "eval_samples_per_second": 1071.581,
      "eval_steps_per_second": 16.794,
      "step": 7000
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00035999999999999997,
      "loss": 0.6783,
      "step": 7500
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00038399999999999996,
      "loss": 0.6783,
      "step": 8000
    },
    {
      "epoch": 0.36,
      "eval_loss": 0.6780590415000916,
      "eval_runtime": 2.0716,
      "eval_samples_per_second": 1108.795,
      "eval_steps_per_second": 17.378,
      "step": 8000
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.000408,
      "loss": 0.6783,
      "step": 8500
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00043199999999999993,
      "loss": 0.6781,
      "step": 9000
    },
    {
      "epoch": 0.4,
      "eval_loss": 0.677251398563385,
      "eval_runtime": 2.0351,
      "eval_samples_per_second": 1128.706,
      "eval_steps_per_second": 17.69,
      "step": 9000
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00045599999999999997,
      "loss": 0.6776,
      "step": 9500
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00047999999999999996,
      "loss": 0.6775,
      "step": 10000
    },
    {
      "epoch": 0.45,
      "eval_loss": 0.6778020262718201,
      "eval_runtime": 2.2347,
      "eval_samples_per_second": 1027.899,
      "eval_steps_per_second": 16.11,
      "step": 10000
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0005039999999999999,
      "loss": 0.6775,
      "step": 10500
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0005279999999999999,
      "loss": 0.6775,
      "step": 11000
    },
    {
      "epoch": 0.49,
      "eval_loss": 0.6769479513168335,
      "eval_runtime": 2.177,
      "eval_samples_per_second": 1055.099,
      "eval_steps_per_second": 16.536,
      "step": 11000
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.000552,
      "loss": 0.6773,
      "step": 11500
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0005759999999999999,
      "loss": 0.6773,
      "step": 12000
    },
    {
      "epoch": 0.54,
      "eval_loss": 0.6773238182067871,
      "eval_runtime": 2.1281,
      "eval_samples_per_second": 1079.366,
      "eval_steps_per_second": 16.916,
      "step": 12000
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0006,
      "loss": 0.6773,
      "step": 12500
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0005999935478721662,
      "loss": 0.6774,
      "step": 13000
    },
    {
      "epoch": 0.58,
      "eval_loss": 0.677127480506897,
      "eval_runtime": 2.1773,
      "eval_samples_per_second": 1054.986,
      "eval_steps_per_second": 16.534,
      "step": 13000
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.000599974191770902,
      "loss": 0.6773,
      "step": 13500
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0005999419325429058,
      "loss": 0.6773,
      "step": 14000
    },
    {
      "epoch": 0.62,
      "eval_loss": 0.6771531105041504,
      "eval_runtime": 2.1173,
      "eval_samples_per_second": 1084.86,
      "eval_steps_per_second": 17.003,
      "step": 14000
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.0005998967715993009,
      "loss": 0.6773,
      "step": 14500
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0005998387109155732,
      "loss": 0.6773,
      "step": 15000
    },
    {
      "epoch": 0.67,
      "eval_loss": 0.6771678924560547,
      "eval_runtime": 2.1919,
      "eval_samples_per_second": 1047.963,
      "eval_steps_per_second": 16.424,
      "step": 15000
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.000599767753031485,
      "loss": 0.6773,
      "step": 15500
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.0005996839010509641,
      "loss": 0.6772,
      "step": 16000
    },
    {
      "epoch": 0.71,
      "eval_loss": 0.6776318550109863,
      "eval_runtime": 2.199,
      "eval_samples_per_second": 1044.559,
      "eval_steps_per_second": 16.371,
      "step": 16000
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0005995871586419678,
      "loss": 0.6773,
      "step": 16500
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.0005994775300363225,
      "loss": 0.6773,
      "step": 17000
    },
    {
      "epoch": 0.76,
      "eval_loss": 0.676984429359436,
      "eval_runtime": 2.1946,
      "eval_samples_per_second": 1046.652,
      "eval_steps_per_second": 16.404,
      "step": 17000
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0005993550200295384,
      "loss": 0.6772,
      "step": 17500
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0005992196339806002,
      "loss": 0.6772,
      "step": 18000
    },
    {
      "epoch": 0.8,
      "eval_loss": 0.6774880290031433,
      "eval_runtime": 2.1027,
      "eval_samples_per_second": 1092.415,
      "eval_steps_per_second": 17.121,
      "step": 18000
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.0005990713778117324,
      "loss": 0.6773,
      "step": 18500
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.0005989102580081398,
      "loss": 0.6772,
      "step": 19000
    },
    {
      "epoch": 0.85,
      "eval_loss": 0.676984965801239,
      "eval_runtime": 2.1686,
      "eval_samples_per_second": 1059.223,
      "eval_steps_per_second": 16.601,
      "step": 19000
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.0005987362816177249,
      "loss": 0.6773,
      "step": 19500
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.0005985494562507783,
      "loss": 0.6774,
      "step": 20000
    },
    {
      "epoch": 0.89,
      "eval_loss": 0.6769698262214661,
      "eval_runtime": 2.1456,
      "eval_samples_per_second": 1070.586,
      "eval_steps_per_second": 16.779,
      "step": 20000
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.000598349790079647,
      "loss": 0.6773,
      "step": 20500
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.000598137291838376,
      "loss": 0.6772,
      "step": 21000
    },
    {
      "epoch": 0.94,
      "eval_loss": 0.6761835217475891,
      "eval_runtime": 2.1367,
      "eval_samples_per_second": 1075.027,
      "eval_steps_per_second": 16.848,
      "step": 21000
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.000597911970822327,
      "loss": 0.6772,
      "step": 21500
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.000597673836887771,
      "loss": 0.6773,
      "step": 22000
    },
    {
      "epoch": 0.98,
      "eval_loss": 0.6774830222129822,
      "eval_runtime": 2.1107,
      "eval_samples_per_second": 1088.282,
      "eval_steps_per_second": 17.056,
      "step": 22000
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.0005974229004514577,
      "loss": 0.6773,
      "step": 22500
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.0005971591724901598,
      "loss": 0.6773,
      "step": 23000
    },
    {
      "epoch": 1.03,
      "eval_loss": 0.6764113903045654,
      "eval_runtime": 2.206,
      "eval_samples_per_second": 1041.236,
      "eval_steps_per_second": 16.319,
      "step": 23000
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.0005968826645401927,
      "loss": 0.6772,
      "step": 23500
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.00059659338869691,
      "loss": 0.6772,
      "step": 24000
    },
    {
      "epoch": 1.07,
      "eval_loss": 0.6768463253974915,
      "eval_runtime": 2.2173,
      "eval_samples_per_second": 1035.939,
      "eval_steps_per_second": 16.236,
      "step": 24000
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0005962913576141742,
      "loss": 0.6773,
      "step": 24500
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.000595976584503803,
      "loss": 0.6772,
      "step": 25000
    },
    {
      "epoch": 1.12,
      "eval_loss": 0.6769193410873413,
      "eval_runtime": 2.2077,
      "eval_samples_per_second": 1040.465,
      "eval_steps_per_second": 16.307,
      "step": 25000
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.0005956490831349923,
      "loss": 0.6773,
      "step": 25500
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.0005953088678337129,
      "loss": 0.6772,
      "step": 26000
    },
    {
      "epoch": 1.16,
      "eval_loss": 0.6775402426719666,
      "eval_runtime": 2.2094,
      "eval_samples_per_second": 1039.671,
      "eval_steps_per_second": 16.294,
      "step": 26000
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.0005949559534820841,
      "loss": 0.6772,
      "step": 26500
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0005945903555177229,
      "loss": 0.6772,
      "step": 27000
    },
    {
      "epoch": 1.2,
      "eval_loss": 0.6775718331336975,
      "eval_runtime": 2.1372,
      "eval_samples_per_second": 1074.76,
      "eval_steps_per_second": 16.844,
      "step": 27000
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.0005942120899330687,
      "loss": 0.6772,
      "step": 27500
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.0005938211732746836,
      "loss": 0.6772,
      "step": 28000
    },
    {
      "epoch": 1.25,
      "eval_loss": 0.6771659255027771,
      "eval_runtime": 2.191,
      "eval_samples_per_second": 1048.378,
      "eval_steps_per_second": 16.431,
      "step": 28000
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.0005934176226425286,
      "loss": 0.6772,
      "step": 28500
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.0005930014556892158,
      "loss": 0.6772,
      "step": 29000
    },
    {
      "epoch": 1.29,
      "eval_loss": 0.6768732666969299,
      "eval_runtime": 2.2912,
      "eval_samples_per_second": 1002.553,
      "eval_steps_per_second": 15.713,
      "step": 29000
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.0005925726906192357,
      "loss": 0.6772,
      "step": 29500
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.0005921313461881617,
      "loss": 0.6773,
      "step": 30000
    },
    {
      "epoch": 1.34,
      "eval_loss": 0.6772350072860718,
      "eval_runtime": 2.2465,
      "eval_samples_per_second": 1022.498,
      "eval_steps_per_second": 16.025,
      "step": 30000
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.0005916774417018287,
      "loss": 0.6772,
      "step": 30500
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0005912109970154897,
      "loss": 0.6772,
      "step": 31000
    },
    {
      "epoch": 1.38,
      "eval_loss": 0.6777171492576599,
      "eval_runtime": 2.1569,
      "eval_samples_per_second": 1064.935,
      "eval_steps_per_second": 16.69,
      "step": 31000
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.0005907320325329461,
      "loss": 0.6772,
      "step": 31500
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.0005902405692056561,
      "loss": 0.6772,
      "step": 32000
    },
    {
      "epoch": 1.43,
      "eval_loss": 0.676947832107544,
      "eval_runtime": 2.3119,
      "eval_samples_per_second": 993.573,
      "eval_steps_per_second": 15.572,
      "step": 32000
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.0005897366285318178,
      "loss": 0.6772,
      "step": 32500
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.0005892202325554288,
      "loss": 0.6773,
      "step": 33000
    },
    {
      "epoch": 1.47,
      "eval_loss": 0.6766595840454102,
      "eval_runtime": 2.1462,
      "eval_samples_per_second": 1070.248,
      "eval_steps_per_second": 16.774,
      "step": 33000
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.0005886914038653217,
      "loss": 0.6772,
      "step": 33500
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.0005881501655941771,
      "loss": 0.677,
      "step": 34000
    },
    {
      "epoch": 1.52,
      "eval_loss": 0.6765649914741516,
      "eval_runtime": 2.1369,
      "eval_samples_per_second": 1074.929,
      "eval_steps_per_second": 16.847,
      "step": 34000
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.00058759654141751,
      "loss": 0.6768,
      "step": 34500
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.0005870305555526355,
      "loss": 0.6765,
      "step": 35000
    },
    {
      "epoch": 1.56,
      "eval_loss": 0.676632821559906,
      "eval_runtime": 2.2745,
      "eval_samples_per_second": 1009.888,
      "eval_steps_per_second": 15.828,
      "step": 35000
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.0005864522327576088,
      "loss": 0.6764,
      "step": 35500
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0005858615983301424,
      "loss": 0.6763,
      "step": 36000
    },
    {
      "epoch": 1.61,
      "eval_loss": 0.6765820980072021,
      "eval_runtime": 2.1766,
      "eval_samples_per_second": 1055.304,
      "eval_steps_per_second": 16.539,
      "step": 36000
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0005852586781064997,
      "loss": 0.6763,
      "step": 36500
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0005846434984603645,
      "loss": 0.6764,
      "step": 37000
    },
    {
      "epoch": 1.65,
      "eval_loss": 0.6758345365524292,
      "eval_runtime": 2.237,
      "eval_samples_per_second": 1026.821,
      "eval_steps_per_second": 16.093,
      "step": 37000
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0005840160863016872,
      "loss": 0.6763,
      "step": 37500
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0005833764690755083,
      "loss": 0.6764,
      "step": 38000
    },
    {
      "epoch": 1.7,
      "eval_loss": 0.6762025952339172,
      "eval_runtime": 2.1722,
      "eval_samples_per_second": 1057.45,
      "eval_steps_per_second": 16.573,
      "step": 38000
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0005827246747607574,
      "loss": 0.6765,
      "step": 38500
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0005820607318690293,
      "loss": 0.6758,
      "step": 39000
    },
    {
      "epoch": 1.74,
      "eval_loss": 0.6771443486213684,
      "eval_runtime": 2.2147,
      "eval_samples_per_second": 1037.153,
      "eval_steps_per_second": 16.255,
      "step": 39000
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0005813846694433368,
      "loss": 0.6773,
      "step": 39500
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.0005806965170568409,
      "loss": 0.6772,
      "step": 40000
    },
    {
      "epoch": 1.78,
      "eval_loss": 0.6770426630973816,
      "eval_runtime": 2.1711,
      "eval_samples_per_second": 1057.979,
      "eval_steps_per_second": 16.581,
      "step": 40000
    },
    {
      "epoch": 1.81,
      "learning_rate": 0.0005799963048115559,
      "loss": 0.6757,
      "step": 40500
    },
    {
      "epoch": 1.83,
      "learning_rate": 0.0005792840633370341,
      "loss": 0.6575,
      "step": 41000
    },
    {
      "epoch": 1.83,
      "eval_loss": 0.6465174555778503,
      "eval_runtime": 2.2243,
      "eval_samples_per_second": 1032.694,
      "eval_steps_per_second": 16.185,
      "step": 41000
    },
    {
      "epoch": 1.85,
      "learning_rate": 0.0005785598237890247,
      "loss": 0.6461,
      "step": 41500
    },
    {
      "epoch": 1.87,
      "learning_rate": 0.0005778236178481119,
      "loss": 0.6373,
      "step": 42000
    },
    {
      "epoch": 1.87,
      "eval_loss": 0.631793737411499,
      "eval_runtime": 2.22,
      "eval_samples_per_second": 1034.694,
      "eval_steps_per_second": 16.216,
      "step": 42000
    },
    {
      "epoch": 1.9,
      "learning_rate": 0.0005770754777183285,
      "loss": 0.6309,
      "step": 42500
    },
    {
      "epoch": 1.92,
      "learning_rate": 0.0005763154361257473,
      "loss": 0.6257,
      "step": 43000
    },
    {
      "epoch": 1.92,
      "eval_loss": 0.6184197664260864,
      "eval_runtime": 2.2848,
      "eval_samples_per_second": 1005.32,
      "eval_steps_per_second": 15.756,
      "step": 43000
    },
    {
      "epoch": 1.94,
      "learning_rate": 0.0005755435263170498,
      "loss": 0.623,
      "step": 43500
    },
    {
      "epoch": 1.96,
      "learning_rate": 0.0005747597820580717,
      "loss": 0.621,
      "step": 44000
    },
    {
      "epoch": 1.96,
      "eval_loss": 0.6136025786399841,
      "eval_runtime": 2.2706,
      "eval_samples_per_second": 1011.621,
      "eval_steps_per_second": 15.855,
      "step": 44000
    },
    {
      "epoch": 1.99,
      "learning_rate": 0.000573964237632326,
      "loss": 0.6195,
      "step": 44500
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0005731569278395029,
      "loss": 0.6183,
      "step": 45000
    },
    {
      "epoch": 2.01,
      "eval_loss": 0.6127315163612366,
      "eval_runtime": 2.1317,
      "eval_samples_per_second": 1077.566,
      "eval_steps_per_second": 16.888,
      "step": 45000
    },
    {
      "epoch": 2.03,
      "learning_rate": 0.0005723378879939481,
      "loss": 0.6172,
      "step": 45500
    },
    {
      "epoch": 2.05,
      "learning_rate": 0.0005715071539231178,
      "loss": 0.6165,
      "step": 46000
    },
    {
      "epoch": 2.05,
      "eval_loss": 0.6103396415710449,
      "eval_runtime": 2.1687,
      "eval_samples_per_second": 1059.145,
      "eval_steps_per_second": 16.6,
      "step": 46000
    },
    {
      "epoch": 2.07,
      "learning_rate": 0.0005706647619660116,
      "loss": 0.6155,
      "step": 46500
    },
    {
      "epoch": 2.1,
      "learning_rate": 0.0005698107489715823,
      "loss": 0.612,
      "step": 47000
    },
    {
      "epoch": 2.1,
      "eval_loss": 0.6013069748878479,
      "eval_runtime": 2.2612,
      "eval_samples_per_second": 1015.821,
      "eval_steps_per_second": 15.921,
      "step": 47000
    },
    {
      "epoch": 2.12,
      "learning_rate": 0.0005689451522971252,
      "loss": 0.6068,
      "step": 47500
    },
    {
      "epoch": 2.14,
      "learning_rate": 0.0005680680098066429,
      "loss": 0.6037,
      "step": 48000
    },
    {
      "epoch": 2.14,
      "eval_loss": 0.5942632555961609,
      "eval_runtime": 2.2238,
      "eval_samples_per_second": 1032.912,
      "eval_steps_per_second": 16.188,
      "step": 48000
    },
    {
      "epoch": 2.16,
      "learning_rate": 0.0005671793598691895,
      "loss": 0.6015,
      "step": 48500
    },
    {
      "epoch": 2.19,
      "learning_rate": 0.0005662792413571921,
      "loss": 0.6,
      "step": 49000
    },
    {
      "epoch": 2.19,
      "eval_loss": 0.5915176868438721,
      "eval_runtime": 2.2389,
      "eval_samples_per_second": 1025.951,
      "eval_steps_per_second": 16.079,
      "step": 49000
    },
    {
      "epoch": 2.21,
      "learning_rate": 0.0005653676936447504,
      "loss": 0.5986,
      "step": 49500
    },
    {
      "epoch": 2.23,
      "learning_rate": 0.0005644447566059142,
      "loss": 0.5973,
      "step": 50000
    },
    {
      "epoch": 2.23,
      "eval_loss": 0.5880685448646545,
      "eval_runtime": 2.2056,
      "eval_samples_per_second": 1041.459,
      "eval_steps_per_second": 16.322,
      "step": 50000
    },
    {
      "epoch": 2.25,
      "learning_rate": 0.0005635104706129397,
      "loss": 0.5958,
      "step": 50500
    },
    {
      "epoch": 2.28,
      "learning_rate": 0.0005625648765345228,
      "loss": 0.5924,
      "step": 51000
    },
    {
      "epoch": 2.28,
      "eval_loss": 0.5799189209938049,
      "eval_runtime": 2.2525,
      "eval_samples_per_second": 1019.737,
      "eval_steps_per_second": 15.982,
      "step": 51000
    },
    {
      "epoch": 2.3,
      "learning_rate": 0.0005616080157340118,
      "loss": 0.5872,
      "step": 51500
    },
    {
      "epoch": 2.32,
      "learning_rate": 0.0005606399300675978,
      "loss": 0.5817,
      "step": 52000
    },
    {
      "epoch": 2.32,
      "eval_loss": 0.5670354962348938,
      "eval_runtime": 2.3437,
      "eval_samples_per_second": 980.093,
      "eval_steps_per_second": 15.361,
      "step": 52000
    },
    {
      "epoch": 2.34,
      "learning_rate": 0.0005596606618824843,
      "loss": 0.5766,
      "step": 52500
    },
    {
      "epoch": 2.36,
      "learning_rate": 0.0005586702540150338,
      "loss": 0.5719,
      "step": 53000
    },
    {
      "epoch": 2.36,
      "eval_loss": 0.5556795597076416,
      "eval_runtime": 2.1734,
      "eval_samples_per_second": 1056.846,
      "eval_steps_per_second": 16.564,
      "step": 53000
    },
    {
      "epoch": 2.39,
      "learning_rate": 0.000557668749788895,
      "loss": 0.5681,
      "step": 53500
    },
    {
      "epoch": 2.41,
      "learning_rate": 0.0005566561930131072,
      "loss": 0.5651,
      "step": 54000
    },
    {
      "epoch": 2.41,
      "eval_loss": 0.5476705431938171,
      "eval_runtime": 2.2374,
      "eval_samples_per_second": 1026.619,
      "eval_steps_per_second": 16.09,
      "step": 54000
    },
    {
      "epoch": 2.43,
      "learning_rate": 0.000555632627980184,
      "loss": 0.562,
      "step": 54500
    },
    {
      "epoch": 2.45,
      "learning_rate": 0.0005545980994641758,
      "loss": 0.5592,
      "step": 55000
    },
    {
      "epoch": 2.45,
      "eval_loss": 0.5407888889312744,
      "eval_runtime": 2.2932,
      "eval_samples_per_second": 1001.641,
      "eval_steps_per_second": 15.698,
      "step": 55000
    },
    {
      "epoch": 2.48,
      "learning_rate": 0.0005535526527187115,
      "loss": 0.5563,
      "step": 55500
    },
    {
      "epoch": 2.5,
      "learning_rate": 0.0005524963334750183,
      "loss": 0.5534,
      "step": 56000
    },
    {
      "epoch": 2.5,
      "eval_loss": 0.536229133605957,
      "eval_runtime": 2.1818,
      "eval_samples_per_second": 1052.81,
      "eval_steps_per_second": 16.5,
      "step": 56000
    },
    {
      "epoch": 2.52,
      "learning_rate": 0.0005514291879399219,
      "loss": 0.5496,
      "step": 56500
    },
    {
      "epoch": 2.54,
      "learning_rate": 0.000550351262793825,
      "loss": 0.5446,
      "step": 57000
    },
    {
      "epoch": 2.54,
      "eval_loss": 0.5251114964485168,
      "eval_runtime": 2.298,
      "eval_samples_per_second": 999.556,
      "eval_steps_per_second": 15.666,
      "step": 57000
    },
    {
      "epoch": 2.57,
      "learning_rate": 0.0005492626051886659,
      "loss": 0.5392,
      "step": 57500
    },
    {
      "epoch": 2.59,
      "learning_rate": 0.0005481632627458546,
      "loss": 0.5342,
      "step": 58000
    },
    {
      "epoch": 2.59,
      "eval_loss": 0.5129896402359009,
      "eval_runtime": 2.2194,
      "eval_samples_per_second": 1034.968,
      "eval_steps_per_second": 16.221,
      "step": 58000
    },
    {
      "epoch": 2.61,
      "learning_rate": 0.0005470532835541911,
      "loss": 0.529,
      "step": 58500
    },
    {
      "epoch": 2.63,
      "learning_rate": 0.000545932716167761,
      "loss": 0.5239,
      "step": 59000
    },
    {
      "epoch": 2.63,
      "eval_loss": 0.5024312734603882,
      "eval_runtime": 2.205,
      "eval_samples_per_second": 1041.743,
      "eval_steps_per_second": 16.327,
      "step": 59000
    },
    {
      "epoch": 2.65,
      "learning_rate": 0.000544801609603812,
      "loss": 0.5193,
      "step": 59500
    },
    {
      "epoch": 2.68,
      "learning_rate": 0.0005436600133406095,
      "loss": 0.5147,
      "step": 60000
    },
    {
      "epoch": 2.68,
      "eval_loss": 0.49465516209602356,
      "eval_runtime": 2.2427,
      "eval_samples_per_second": 1024.23,
      "eval_steps_per_second": 16.052,
      "step": 60000
    },
    {
      "epoch": 2.7,
      "learning_rate": 0.0005425079773152721,
      "loss": 0.5102,
      "step": 60500
    },
    {
      "epoch": 2.72,
      "learning_rate": 0.0005413455519215879,
      "loss": 0.5061,
      "step": 61000
    },
    {
      "epoch": 2.72,
      "eval_loss": 0.4848201274871826,
      "eval_runtime": 2.1727,
      "eval_samples_per_second": 1057.234,
      "eval_steps_per_second": 16.57,
      "step": 61000
    },
    {
      "epoch": 2.74,
      "learning_rate": 0.0005401727880078093,
      "loss": 0.502,
      "step": 61500
    },
    {
      "epoch": 2.77,
      "learning_rate": 0.0005389897368744289,
      "loss": 0.4981,
      "step": 62000
    },
    {
      "epoch": 2.77,
      "eval_loss": 0.4745917320251465,
      "eval_runtime": 2.1553,
      "eval_samples_per_second": 1065.724,
      "eval_steps_per_second": 16.703,
      "step": 62000
    },
    {
      "epoch": 2.79,
      "learning_rate": 0.0005377964502719361,
      "loss": 0.4972,
      "step": 62500
    },
    {
      "epoch": 2.81,
      "learning_rate": 0.0005365929803985524,
      "loss": 0.4912,
      "step": 63000
    },
    {
      "epoch": 2.81,
      "eval_loss": 0.46808210015296936,
      "eval_runtime": 2.2693,
      "eval_samples_per_second": 1012.199,
      "eval_steps_per_second": 15.864,
      "step": 63000
    },
    {
      "epoch": 2.83,
      "learning_rate": 0.0005353793798979489,
      "loss": 0.4882,
      "step": 63500
    },
    {
      "epoch": 2.86,
      "learning_rate": 0.000534155701856943,
      "loss": 0.4847,
      "step": 64000
    },
    {
      "epoch": 2.86,
      "eval_loss": 0.45994389057159424,
      "eval_runtime": 2.2357,
      "eval_samples_per_second": 1027.41,
      "eval_steps_per_second": 16.102,
      "step": 64000
    },
    {
      "epoch": 2.88,
      "learning_rate": 0.0005329219998031763,
      "loss": 0.4818,
      "step": 64500
    },
    {
      "epoch": 2.9,
      "learning_rate": 0.0005316783277027734,
      "loss": 0.4792,
      "step": 65000
    },
    {
      "epoch": 2.9,
      "eval_loss": 0.4537006914615631,
      "eval_runtime": 2.1952,
      "eval_samples_per_second": 1046.372,
      "eval_steps_per_second": 16.399,
      "step": 65000
    },
    {
      "epoch": 2.92,
      "learning_rate": 0.0005304247399579808,
      "loss": 0.4766,
      "step": 65500
    },
    {
      "epoch": 2.94,
      "learning_rate": 0.0005291612914047876,
      "loss": 0.474,
      "step": 66000
    },
    {
      "epoch": 2.94,
      "eval_loss": 0.44910311698913574,
      "eval_runtime": 2.304,
      "eval_samples_per_second": 996.96,
      "eval_steps_per_second": 15.625,
      "step": 66000
    },
    {
      "epoch": 2.97,
      "learning_rate": 0.0005278880373105263,
      "loss": 0.4713,
      "step": 66500
    },
    {
      "epoch": 2.99,
      "learning_rate": 0.0005266050333714561,
      "loss": 0.4688,
      "step": 67000
    },
    {
      "epoch": 2.99,
      "eval_loss": 0.4437292516231537,
      "eval_runtime": 2.2404,
      "eval_samples_per_second": 1025.255,
      "eval_steps_per_second": 16.068,
      "step": 67000
    },
    {
      "epoch": 3.01,
      "learning_rate": 0.0005253123357103253,
      "loss": 0.4665,
      "step": 67500
    },
    {
      "epoch": 3.03,
      "learning_rate": 0.0005240100008739177,
      "loss": 0.464,
      "step": 68000
    },
    {
      "epoch": 3.03,
      "eval_loss": 0.4391787052154541,
      "eval_runtime": 2.1873,
      "eval_samples_per_second": 1050.17,
      "eval_steps_per_second": 16.459,
      "step": 68000
    },
    {
      "epoch": 3.06,
      "learning_rate": 0.0005226980858305778,
      "loss": 0.4616,
      "step": 68500
    },
    {
      "epoch": 3.08,
      "learning_rate": 0.0005213766479677197,
      "loss": 0.4592,
      "step": 69000
    },
    {
      "epoch": 3.08,
      "eval_loss": 0.43239352107048035,
      "eval_runtime": 2.1819,
      "eval_samples_per_second": 1052.771,
      "eval_steps_per_second": 16.5,
      "step": 69000
    },
    {
      "epoch": 3.1,
      "learning_rate": 0.0005200457450893163,
      "loss": 0.457,
      "step": 69500
    },
    {
      "epoch": 3.12,
      "learning_rate": 0.0005187054354133712,
      "loss": 0.4547,
      "step": 70000
    },
    {
      "epoch": 3.12,
      "eval_loss": 0.4284396469593048,
      "eval_runtime": 2.1519,
      "eval_samples_per_second": 1067.405,
      "eval_steps_per_second": 16.729,
      "step": 70000
    },
    {
      "epoch": 3.15,
      "learning_rate": 0.0005173557775693715,
      "loss": 0.4528,
      "step": 70500
    },
    {
      "epoch": 3.17,
      "learning_rate": 0.0005159968305957235,
      "loss": 0.4507,
      "step": 71000
    },
    {
      "epoch": 3.17,
      "eval_loss": 0.42597973346710205,
      "eval_runtime": 2.1768,
      "eval_samples_per_second": 1055.242,
      "eval_steps_per_second": 16.538,
      "step": 71000
    },
    {
      "epoch": 3.19,
      "learning_rate": 0.0005146286539371703,
      "loss": 0.4489,
      "step": 71500
    },
    {
      "epoch": 3.21,
      "learning_rate": 0.0005132513074421913,
      "loss": 0.4468,
      "step": 72000
    },
    {
      "epoch": 3.21,
      "eval_loss": 0.4192256033420563,
      "eval_runtime": 2.2165,
      "eval_samples_per_second": 1036.301,
      "eval_steps_per_second": 16.242,
      "step": 72000
    },
    {
      "epoch": 3.23,
      "learning_rate": 0.0005118648513603841,
      "loss": 0.445,
      "step": 72500
    },
    {
      "epoch": 3.26,
      "learning_rate": 0.0005104693463398293,
      "loss": 0.4432,
      "step": 73000
    },
    {
      "epoch": 3.26,
      "eval_loss": 0.4161074459552765,
      "eval_runtime": 2.2735,
      "eval_samples_per_second": 1010.341,
      "eval_steps_per_second": 15.835,
      "step": 73000
    },
    {
      "epoch": 3.28,
      "learning_rate": 0.0005090648534244371,
      "loss": 0.4415,
      "step": 73500
    },
    {
      "epoch": 3.3,
      "learning_rate": 0.0005076514340512776,
      "loss": 0.44,
      "step": 74000
    },
    {
      "epoch": 3.3,
      "eval_loss": 0.41528481245040894,
      "eval_runtime": 2.1341,
      "eval_samples_per_second": 1076.31,
      "eval_steps_per_second": 16.869,
      "step": 74000
    },
    {
      "epoch": 3.32,
      "learning_rate": 0.0005062291500478931,
      "loss": 0.4386,
      "step": 74500
    },
    {
      "epoch": 3.35,
      "learning_rate": 0.0005047980636295937,
      "loss": 0.4367,
      "step": 75000
    },
    {
      "epoch": 3.35,
      "eval_loss": 0.4101610779762268,
      "eval_runtime": 2.2225,
      "eval_samples_per_second": 1033.52,
      "eval_steps_per_second": 16.198,
      "step": 75000
    },
    {
      "epoch": 3.37,
      "learning_rate": 0.0005033582373967348,
      "loss": 0.4351,
      "step": 75500
    },
    {
      "epoch": 3.39,
      "learning_rate": 0.0005019097343319809,
      "loss": 0.4337,
      "step": 76000
    },
    {
      "epoch": 3.39,
      "eval_loss": 0.40619799494743347,
      "eval_runtime": 2.1631,
      "eval_samples_per_second": 1061.919,
      "eval_steps_per_second": 16.643,
      "step": 76000
    },
    {
      "epoch": 3.41,
      "learning_rate": 0.0005004526177975481,
      "loss": 0.4322,
      "step": 76500
    },
    {
      "epoch": 3.44,
      "learning_rate": 0.0004989869515324342,
      "loss": 0.4311,
      "step": 77000
    },
    {
      "epoch": 3.44,
      "eval_loss": 0.4019022583961487,
      "eval_runtime": 2.2694,
      "eval_samples_per_second": 1012.161,
      "eval_steps_per_second": 15.863,
      "step": 77000
    },
    {
      "epoch": 3.46,
      "learning_rate": 0.0004975127996496297,
      "loss": 0.4294,
      "step": 77500
    },
    {
      "epoch": 3.48,
      "learning_rate": 0.0004960302266333135,
      "loss": 0.4286,
      "step": 78000
    },
    {
      "epoch": 3.48,
      "eval_loss": 0.40067020058631897,
      "eval_runtime": 2.31,
      "eval_samples_per_second": 994.37,
      "eval_steps_per_second": 15.584,
      "step": 78000
    },
    {
      "epoch": 3.5,
      "learning_rate": 0.0004945392973360323,
      "loss": 0.427,
      "step": 78500
    },
    {
      "epoch": 3.52,
      "learning_rate": 0.0004930400769758634,
      "loss": 0.4259,
      "step": 79000
    },
    {
      "epoch": 3.52,
      "eval_loss": 0.3996644616127014,
      "eval_runtime": 2.1788,
      "eval_samples_per_second": 1054.255,
      "eval_steps_per_second": 16.523,
      "step": 79000
    },
    {
      "epoch": 3.55,
      "learning_rate": 0.0004915326311335622,
      "loss": 0.425,
      "step": 79500
    },
    {
      "epoch": 3.57,
      "learning_rate": 0.0004900170257496933,
      "loss": 0.4239,
      "step": 80000
    },
    {
      "epoch": 3.57,
      "eval_loss": 0.3968483507633209,
      "eval_runtime": 2.2257,
      "eval_samples_per_second": 1032.02,
      "eval_steps_per_second": 16.174,
      "step": 80000
    },
    {
      "epoch": 3.59,
      "learning_rate": 0.0004884933271217461,
      "loss": 0.4229,
      "step": 80500
    },
    {
      "epoch": 3.61,
      "learning_rate": 0.0004869616019012347,
      "loss": 0.4218,
      "step": 81000
    },
    {
      "epoch": 3.61,
      "eval_loss": 0.3948546051979065,
      "eval_runtime": 2.2856,
      "eval_samples_per_second": 1005.007,
      "eval_steps_per_second": 15.751,
      "step": 81000
    },
    {
      "epoch": 3.64,
      "learning_rate": 0.00048542191709078226,
      "loss": 0.4209,
      "step": 81500
    },
    {
      "epoch": 3.66,
      "learning_rate": 0.00048387434004119027,
      "loss": 0.4201,
      "step": 82000
    },
    {
      "epoch": 3.66,
      "eval_loss": 0.39345651865005493,
      "eval_runtime": 2.2871,
      "eval_samples_per_second": 1004.335,
      "eval_steps_per_second": 15.741,
      "step": 82000
    },
    {
      "epoch": 3.68,
      "learning_rate": 0.0004823189384484924,
      "loss": 0.4206,
      "step": 82500
    },
    {
      "epoch": 3.7,
      "learning_rate": 0.000480755780350993,
      "loss": 0.4182,
      "step": 83000
    },
    {
      "epoch": 3.7,
      "eval_loss": 0.39259859919548035,
      "eval_runtime": 2.2778,
      "eval_samples_per_second": 1008.432,
      "eval_steps_per_second": 15.805,
      "step": 83000
    },
    {
      "epoch": 3.73,
      "learning_rate": 0.0004791849341262914,
      "loss": 0.4176,
      "step": 83500
    },
    {
      "epoch": 3.75,
      "learning_rate": 0.0004776064684882901,
      "loss": 0.4168,
      "step": 84000
    },
    {
      "epoch": 3.75,
      "eval_loss": 0.38789093494415283,
      "eval_runtime": 2.2568,
      "eval_samples_per_second": 1017.832,
      "eval_steps_per_second": 15.952,
      "step": 84000
    },
    {
      "epoch": 3.77,
      "learning_rate": 0.0004760204524841897,
      "loss": 0.416,
      "step": 84500
    },
    {
      "epoch": 3.79,
      "learning_rate": 0.0004744269554914683,
      "loss": 0.4155,
      "step": 85000
    },
    {
      "epoch": 3.79,
      "eval_loss": 0.3885124921798706,
      "eval_runtime": 2.2513,
      "eval_samples_per_second": 1020.282,
      "eval_steps_per_second": 15.99,
      "step": 85000
    },
    {
      "epoch": 3.81,
      "learning_rate": 0.00047282604721484643,
      "loss": 0.4144,
      "step": 85500
    },
    {
      "epoch": 3.84,
      "learning_rate": 0.0004712177976832385,
      "loss": 0.4136,
      "step": 86000
    },
    {
      "epoch": 3.84,
      "eval_loss": 0.38439127802848816,
      "eval_runtime": 2.2526,
      "eval_samples_per_second": 1019.704,
      "eval_steps_per_second": 15.981,
      "step": 86000
    },
    {
      "epoch": 3.86,
      "learning_rate": 0.0004696022772466888,
      "loss": 0.413,
      "step": 86500
    },
    {
      "epoch": 3.88,
      "learning_rate": 0.00046797955657329487,
      "loss": 0.4124,
      "step": 87000
    },
    {
      "epoch": 3.88,
      "eval_loss": 0.3855268955230713,
      "eval_runtime": 2.2218,
      "eval_samples_per_second": 1033.851,
      "eval_steps_per_second": 16.203,
      "step": 87000
    },
    {
      "epoch": 3.9,
      "learning_rate": 0.000466349706646116,
      "loss": 0.4118,
      "step": 87500
    },
    {
      "epoch": 3.93,
      "learning_rate": 0.00046471279876006763,
      "loss": 0.4116,
      "step": 88000
    },
    {
      "epoch": 3.93,
      "eval_loss": 0.3830411732196808,
      "eval_runtime": 2.2947,
      "eval_samples_per_second": 1001.012,
      "eval_steps_per_second": 15.688,
      "step": 88000
    },
    {
      "epoch": 3.95,
      "learning_rate": 0.00046306890451880395,
      "loss": 0.4107,
      "step": 88500
    },
    {
      "epoch": 3.97,
      "learning_rate": 0.0004614180958315844,
      "loss": 0.4098,
      "step": 89000
    },
    {
      "epoch": 3.97,
      "eval_loss": 0.38374021649360657,
      "eval_runtime": 2.1424,
      "eval_samples_per_second": 1072.163,
      "eval_steps_per_second": 16.804,
      "step": 89000
    },
    {
      "epoch": 3.99,
      "learning_rate": 0.00045976044491012884,
      "loss": 0.4094,
      "step": 89500
    },
    {
      "epoch": 4.01,
      "learning_rate": 0.00045809602426545847,
      "loss": 0.4087,
      "step": 90000
    },
    {
      "epoch": 4.01,
      "eval_loss": 0.3802391290664673,
      "eval_runtime": 2.1802,
      "eval_samples_per_second": 1053.586,
      "eval_steps_per_second": 16.512,
      "step": 90000
    },
    {
      "epoch": 4.04,
      "learning_rate": 0.00045642490670472436,
      "loss": 0.4084,
      "step": 90500
    },
    {
      "epoch": 4.06,
      "learning_rate": 0.0004547471653280225,
      "loss": 0.4078,
      "step": 91000
    },
    {
      "epoch": 4.06,
      "eval_loss": 0.37994059920310974,
      "eval_runtime": 2.2046,
      "eval_samples_per_second": 1041.911,
      "eval_steps_per_second": 16.329,
      "step": 91000
    },
    {
      "epoch": 4.08,
      "learning_rate": 0.00045306287352519543,
      "loss": 0.407,
      "step": 91500
    },
    {
      "epoch": 4.1,
      "learning_rate": 0.00045137210497262333,
      "loss": 0.4068,
      "step": 92000
    },
    {
      "epoch": 4.1,
      "eval_loss": 0.3793868124485016,
      "eval_runtime": 2.2708,
      "eval_samples_per_second": 1011.534,
      "eval_steps_per_second": 15.853,
      "step": 92000
    },
    {
      "epoch": 4.13,
      "learning_rate": 0.0004496749336299999,
      "loss": 0.4063,
      "step": 92500
    },
    {
      "epoch": 4.15,
      "learning_rate": 0.0004479714337370977,
      "loss": 0.4057,
      "step": 93000
    },
    {
      "epoch": 4.15,
      "eval_loss": 0.37835967540740967,
      "eval_runtime": 2.237,
      "eval_samples_per_second": 1026.834,
      "eval_steps_per_second": 16.093,
      "step": 93000
    },
    {
      "epoch": 4.17,
      "learning_rate": 0.00044626167981052036,
      "loss": 0.4052,
      "step": 93500
    },
    {
      "epoch": 4.19,
      "learning_rate": 0.00044454574664044404,
      "loss": 0.4047,
      "step": 94000
    },
    {
      "epoch": 4.19,
      "eval_loss": 0.37884432077407837,
      "eval_runtime": 2.2678,
      "eval_samples_per_second": 1012.88,
      "eval_steps_per_second": 15.874,
      "step": 94000
    },
    {
      "epoch": 4.22,
      "learning_rate": 0.000442823709287344,
      "loss": 0.4044,
      "step": 94500
    },
    {
      "epoch": 4.24,
      "learning_rate": 0.0004410956430787129,
      "loss": 0.4047,
      "step": 95000
    },
    {
      "epoch": 4.24,
      "eval_loss": 0.37695789337158203,
      "eval_runtime": 2.1458,
      "eval_samples_per_second": 1070.466,
      "eval_steps_per_second": 16.777,
      "step": 95000
    },
    {
      "epoch": 4.26,
      "learning_rate": 0.0004393616236057647,
      "loss": 0.4036,
      "step": 95500
    },
    {
      "epoch": 4.28,
      "learning_rate": 0.00043762172672012875,
      "loss": 0.4029,
      "step": 96000
    },
    {
      "epoch": 4.28,
      "eval_loss": 0.37500157952308655,
      "eval_runtime": 2.2301,
      "eval_samples_per_second": 1029.999,
      "eval_steps_per_second": 16.143,
      "step": 96000
    },
    {
      "epoch": 4.3,
      "learning_rate": 0.0004358760285305312,
      "loss": 0.4025,
      "step": 96500
    },
    {
      "epoch": 4.33,
      "learning_rate": 0.0004341246053994663,
      "loss": 0.4022,
      "step": 97000
    },
    {
      "epoch": 4.33,
      "eval_loss": 0.37471264600753784,
      "eval_runtime": 2.2149,
      "eval_samples_per_second": 1037.069,
      "eval_steps_per_second": 16.254,
      "step": 97000
    },
    {
      "epoch": 4.35,
      "learning_rate": 0.00043236753393985534,
      "loss": 0.4019,
      "step": 97500
    },
    {
      "epoch": 4.37,
      "learning_rate": 0.0004306048910116964,
      "loss": 0.4015,
      "step": 98000
    },
    {
      "epoch": 4.37,
      "eval_loss": 0.3735538125038147,
      "eval_runtime": 2.2361,
      "eval_samples_per_second": 1027.233,
      "eval_steps_per_second": 16.099,
      "step": 98000
    },
    {
      "epoch": 4.39,
      "learning_rate": 0.0004288367537187012,
      "loss": 0.4011,
      "step": 98500
    },
    {
      "epoch": 4.42,
      "learning_rate": 0.00042706319940492284,
      "loss": 0.4007,
      "step": 99000
    },
    {
      "epoch": 4.42,
      "eval_loss": 0.3751888573169708,
      "eval_runtime": 2.2595,
      "eval_samples_per_second": 1016.613,
      "eval_steps_per_second": 15.933,
      "step": 99000
    },
    {
      "epoch": 4.44,
      "learning_rate": 0.00042528430565137254,
      "loss": 0.4003,
      "step": 99500
    },
    {
      "epoch": 4.46,
      "learning_rate": 0.00042350015027262593,
      "loss": 0.4,
      "step": 100000
    },
    {
      "epoch": 4.46,
      "eval_loss": 0.37434616684913635,
      "eval_runtime": 2.2061,
      "eval_samples_per_second": 1041.22,
      "eval_steps_per_second": 16.319,
      "step": 100000
    },
    {
      "epoch": 4.48,
      "learning_rate": 0.00042171081131341917,
      "loss": 0.3996,
      "step": 100500
    },
    {
      "epoch": 4.51,
      "learning_rate": 0.00041991636704523497,
      "loss": 0.3995,
      "step": 101000
    },
    {
      "epoch": 4.51,
      "eval_loss": 0.37412840127944946,
      "eval_runtime": 2.2118,
      "eval_samples_per_second": 1038.514,
      "eval_steps_per_second": 16.276,
      "step": 101000
    },
    {
      "epoch": 4.53,
      "learning_rate": 0.00041811689596287893,
      "loss": 0.3989,
      "step": 101500
    },
    {
      "epoch": 4.55,
      "learning_rate": 0.0004163124767810454,
      "loss": 0.3985,
      "step": 102000
    },
    {
      "epoch": 4.55,
      "eval_loss": 0.37019699811935425,
      "eval_runtime": 2.207,
      "eval_samples_per_second": 1040.78,
      "eval_steps_per_second": 16.312,
      "step": 102000
    },
    {
      "epoch": 4.57,
      "learning_rate": 0.00041450318843087506,
      "loss": 0.3983,
      "step": 102500
    },
    {
      "epoch": 4.59,
      "learning_rate": 0.00041268911005650166,
      "loss": 0.3981,
      "step": 103000
    },
    {
      "epoch": 4.59,
      "eval_loss": 0.3799527883529663,
      "eval_runtime": 2.7041,
      "eval_samples_per_second": 849.448,
      "eval_steps_per_second": 13.313,
      "step": 103000
    },
    {
      "epoch": 4.62,
      "learning_rate": 0.00041087032101159006,
      "loss": 0.3978,
      "step": 103500
    },
    {
      "epoch": 4.64,
      "learning_rate": 0.00040904690085586515,
      "loss": 0.3986,
      "step": 104000
    },
    {
      "epoch": 4.64,
      "eval_loss": 0.3734039068222046,
      "eval_runtime": 2.2569,
      "eval_samples_per_second": 1017.749,
      "eval_steps_per_second": 15.951,
      "step": 104000
    },
    {
      "epoch": 4.66,
      "learning_rate": 0.0004072189293516316,
      "loss": 0.3969,
      "step": 104500
    },
    {
      "epoch": 4.68,
      "learning_rate": 0.0004053864864602847,
      "loss": 0.3966,
      "step": 105000
    },
    {
      "epoch": 4.68,
      "eval_loss": 0.37050846219062805,
      "eval_runtime": 2.2224,
      "eval_samples_per_second": 1033.581,
      "eval_steps_per_second": 16.199,
      "step": 105000
    },
    {
      "epoch": 4.71,
      "learning_rate": 0.00040354965233881297,
      "loss": 0.3961,
      "step": 105500
    },
    {
      "epoch": 4.73,
      "learning_rate": 0.0004017085073362913,
      "loss": 0.3957,
      "step": 106000
    },
    {
      "epoch": 4.73,
      "eval_loss": 0.36802881956100464,
      "eval_runtime": 2.2556,
      "eval_samples_per_second": 1018.364,
      "eval_steps_per_second": 15.96,
      "step": 106000
    },
    {
      "epoch": 4.75,
      "learning_rate": 0.00039986313199036664,
      "loss": 0.3956,
      "step": 106500
    },
    {
      "epoch": 4.77,
      "learning_rate": 0.00039801360702373484,
      "loss": 0.3957,
      "step": 107000
    },
    {
      "epoch": 4.77,
      "eval_loss": 0.3662741482257843,
      "eval_runtime": 2.5144,
      "eval_samples_per_second": 913.525,
      "eval_steps_per_second": 14.317,
      "step": 107000
    },
    {
      "epoch": 4.8,
      "learning_rate": 0.00039616001334060954,
      "loss": 0.3951,
      "step": 107500
    },
    {
      "epoch": 4.82,
      "learning_rate": 0.00039430243202318314,
      "loss": 0.3948,
      "step": 108000
    },
    {
      "epoch": 4.82,
      "eval_loss": 0.36831599473953247,
      "eval_runtime": 2.2146,
      "eval_samples_per_second": 1037.206,
      "eval_steps_per_second": 16.256,
      "step": 108000
    },
    {
      "epoch": 4.84,
      "learning_rate": 0.00039244094432808034,
      "loss": 0.3945,
      "step": 108500
    },
    {
      "epoch": 4.86,
      "learning_rate": 0.0003905756316828033,
      "loss": 0.3943,
      "step": 109000
    },
    {
      "epoch": 4.86,
      "eval_loss": 0.36974549293518066,
      "eval_runtime": 2.1759,
      "eval_samples_per_second": 1055.667,
      "eval_steps_per_second": 16.545,
      "step": 109000
    },
    {
      "epoch": 4.88,
      "learning_rate": 0.00038870657568216963,
      "loss": 0.3939,
      "step": 109500
    },
    {
      "epoch": 4.91,
      "learning_rate": 0.00038683385808474416,
      "loss": 0.3936,
      "step": 110000
    },
    {
      "epoch": 4.91,
      "eval_loss": 0.3671797811985016,
      "eval_runtime": 2.2064,
      "eval_samples_per_second": 1041.083,
      "eval_steps_per_second": 16.316,
      "step": 110000
    },
    {
      "epoch": 4.93,
      "learning_rate": 0.00038495756080926107,
      "loss": 0.3933,
      "step": 110500
    },
    {
      "epoch": 4.95,
      "learning_rate": 0.0003830777659310416,
      "loss": 0.3932,
      "step": 111000
    },
    {
      "epoch": 4.95,
      "eval_loss": 0.36492955684661865,
      "eval_runtime": 2.2207,
      "eval_samples_per_second": 1034.354,
      "eval_steps_per_second": 16.211,
      "step": 111000
    },
    {
      "epoch": 4.97,
      "learning_rate": 0.0003811945556784033,
      "loss": 0.3931,
      "step": 111500
    },
    {
      "epoch": 5.0,
      "learning_rate": 0.00037930801242906366,
      "loss": 0.3925,
      "step": 112000
    },
    {
      "epoch": 5.0,
      "eval_loss": 0.36507272720336914,
      "eval_runtime": 2.3005,
      "eval_samples_per_second": 998.488,
      "eval_steps_per_second": 15.649,
      "step": 112000
    },
    {
      "epoch": 5.02,
      "learning_rate": 0.00037741821870653576,
      "loss": 0.3924,
      "step": 112500
    },
    {
      "epoch": 5.04,
      "learning_rate": 0.00037552525717651905,
      "loss": 0.3919,
      "step": 113000
    },
    {
      "epoch": 5.04,
      "eval_loss": 0.3649584650993347,
      "eval_runtime": 2.1025,
      "eval_samples_per_second": 1092.533,
      "eval_steps_per_second": 17.123,
      "step": 113000
    },
    {
      "epoch": 5.06,
      "learning_rate": 0.0003736292106432834,
      "loss": 0.3918,
      "step": 113500
    },
    {
      "epoch": 5.09,
      "learning_rate": 0.00037173016204604647,
      "loss": 0.3915,
      "step": 114000
    },
    {
      "epoch": 5.09,
      "eval_loss": 0.36355069279670715,
      "eval_runtime": 2.1224,
      "eval_samples_per_second": 1082.26,
      "eval_steps_per_second": 16.962,
      "step": 114000
    },
    {
      "epoch": 5.11,
      "learning_rate": 0.0003698281944553456,
      "loss": 0.3915,
      "step": 114500
    },
    {
      "epoch": 5.13,
      "learning_rate": 0.0003679233910694053,
      "loss": 0.3911,
      "step": 115000
    },
    {
      "epoch": 5.13,
      "eval_loss": 0.36546093225479126,
      "eval_runtime": 2.202,
      "eval_samples_per_second": 1043.148,
      "eval_steps_per_second": 16.349,
      "step": 115000
    },
    {
      "epoch": 5.15,
      "learning_rate": 0.000366015835210496,
      "loss": 0.3911,
      "step": 115500
    },
    {
      "epoch": 5.17,
      "learning_rate": 0.0003641056103212908,
      "loss": 0.3905,
      "step": 116000
    },
    {
      "epoch": 5.17,
      "eval_loss": 0.36497965455055237,
      "eval_runtime": 2.1541,
      "eval_samples_per_second": 1066.328,
      "eval_steps_per_second": 16.712,
      "step": 116000
    },
    {
      "epoch": 5.2,
      "learning_rate": 0.00036219279996121446,
      "loss": 0.3904,
      "step": 116500
    },
    {
      "epoch": 5.22,
      "learning_rate": 0.0003602774878027888,
      "loss": 0.3905,
      "step": 117000
    },
    {
      "epoch": 5.22,
      "eval_loss": 0.40538668632507324,
      "eval_runtime": 2.1607,
      "eval_samples_per_second": 1063.067,
      "eval_steps_per_second": 16.661,
      "step": 117000
    },
    {
      "epoch": 5.24,
      "learning_rate": 0.00035835975762797245,
      "loss": 0.3917,
      "step": 117500
    },
    {
      "epoch": 5.26,
      "learning_rate": 0.0003564396933244957,
      "loss": 0.3894,
      "step": 118000
    },
    {
      "epoch": 5.26,
      "eval_loss": 0.36090952157974243,
      "eval_runtime": 2.2235,
      "eval_samples_per_second": 1033.034,
      "eval_steps_per_second": 16.19,
      "step": 118000
    },
    {
      "epoch": 5.29,
      "learning_rate": 0.0003545173788821915,
      "loss": 0.3893,
      "step": 118500
    },
    {
      "epoch": 5.31,
      "learning_rate": 0.00035259289838932104,
      "loss": 0.3889,
      "step": 119000
    },
    {
      "epoch": 5.31,
      "eval_loss": 0.3599041998386383,
      "eval_runtime": 2.1374,
      "eval_samples_per_second": 1074.669,
      "eval_steps_per_second": 16.843,
      "step": 119000
    },
    {
      "epoch": 5.33,
      "learning_rate": 0.0003506663360288954,
      "loss": 0.3889,
      "step": 119500
    },
    {
      "epoch": 5.35,
      "learning_rate": 0.00034873777607499334,
      "loss": 0.3888,
      "step": 120000
    },
    {
      "epoch": 5.35,
      "eval_loss": 0.3593074679374695,
      "eval_runtime": 2.1457,
      "eval_samples_per_second": 1070.506,
      "eval_steps_per_second": 16.778,
      "step": 120000
    }
  ],
  "max_steps": 250000,
  "num_train_epochs": 12,
  "total_flos": 1.9219803744787812e+21,
  "trial_name": null,
  "trial_params": null
}