{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.734827264239029,
  "eval_steps": 50,
  "global_step": 4000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.05,
      "eval_loss": 1.6599863767623901,
      "eval_runtime": 3.0665,
      "eval_samples_per_second": 32.611,
      "eval_steps_per_second": 4.239,
      "step": 50
    },
    {
      "epoch": 0.09,
      "eval_loss": 1.5268713235855103,
      "eval_runtime": 3.0649,
      "eval_samples_per_second": 32.627,
      "eval_steps_per_second": 4.242,
      "step": 100
    },
    {
      "epoch": 0.14,
      "eval_loss": 1.4272453784942627,
      "eval_runtime": 3.0703,
      "eval_samples_per_second": 32.57,
      "eval_steps_per_second": 4.234,
      "step": 150
    },
    {
      "epoch": 0.19,
      "eval_loss": 1.3691519498825073,
      "eval_runtime": 3.0755,
      "eval_samples_per_second": 32.515,
      "eval_steps_per_second": 4.227,
      "step": 200
    },
    {
      "epoch": 0.23,
      "eval_loss": 1.3439632654190063,
      "eval_runtime": 3.0781,
      "eval_samples_per_second": 32.488,
      "eval_steps_per_second": 4.223,
      "step": 250
    },
    {
      "epoch": 0.28,
      "eval_loss": 1.2951898574829102,
      "eval_runtime": 3.0608,
      "eval_samples_per_second": 32.671,
      "eval_steps_per_second": 4.247,
      "step": 300
    },
    {
      "epoch": 0.33,
      "eval_loss": 1.2598179578781128,
      "eval_runtime": 3.0651,
      "eval_samples_per_second": 32.626,
      "eval_steps_per_second": 4.241,
      "step": 350
    },
    {
      "epoch": 0.37,
      "eval_loss": 1.2034798860549927,
      "eval_runtime": 3.0667,
      "eval_samples_per_second": 32.608,
      "eval_steps_per_second": 4.239,
      "step": 400
    },
    {
      "epoch": 0.42,
      "eval_loss": 1.1628668308258057,
      "eval_runtime": 3.0674,
      "eval_samples_per_second": 32.601,
      "eval_steps_per_second": 4.238,
      "step": 450
    },
    {
      "epoch": 0.47,
      "learning_rate": 8.75e-05,
      "loss": 1.6585,
      "step": 500
    },
    {
      "epoch": 0.47,
      "eval_loss": 1.1156766414642334,
      "eval_runtime": 3.0714,
      "eval_samples_per_second": 32.558,
      "eval_steps_per_second": 4.233,
      "step": 500
    },
    {
      "epoch": 0.51,
      "eval_loss": 1.0700278282165527,
      "eval_runtime": 3.0584,
      "eval_samples_per_second": 32.696,
      "eval_steps_per_second": 4.251,
      "step": 550
    },
    {
      "epoch": 0.56,
      "eval_loss": 1.0359002351760864,
      "eval_runtime": 3.0694,
      "eval_samples_per_second": 32.58,
      "eval_steps_per_second": 4.235,
      "step": 600
    },
    {
      "epoch": 0.61,
      "eval_loss": 1.019242286682129,
      "eval_runtime": 3.0562,
      "eval_samples_per_second": 32.72,
      "eval_steps_per_second": 4.254,
      "step": 650
    },
    {
      "epoch": 0.65,
      "eval_loss": 0.9841110706329346,
      "eval_runtime": 3.0664,
      "eval_samples_per_second": 32.611,
      "eval_steps_per_second": 4.239,
      "step": 700
    },
    {
      "epoch": 0.7,
      "eval_loss": 0.993296205997467,
      "eval_runtime": 3.074,
      "eval_samples_per_second": 32.531,
      "eval_steps_per_second": 4.229,
      "step": 750
    },
    {
      "epoch": 0.75,
      "eval_loss": 0.9330427646636963,
      "eval_runtime": 3.0586,
      "eval_samples_per_second": 32.695,
      "eval_steps_per_second": 4.25,
      "step": 800
    },
    {
      "epoch": 0.79,
      "eval_loss": 0.9203641414642334,
      "eval_runtime": 3.0859,
      "eval_samples_per_second": 32.405,
      "eval_steps_per_second": 4.213,
      "step": 850
    },
    {
      "epoch": 0.84,
      "eval_loss": 0.9058796763420105,
      "eval_runtime": 3.068,
      "eval_samples_per_second": 32.595,
      "eval_steps_per_second": 4.237,
      "step": 900
    },
    {
      "epoch": 0.89,
      "eval_loss": 0.9017823934555054,
      "eval_runtime": 3.0852,
      "eval_samples_per_second": 32.413,
      "eval_steps_per_second": 4.214,
      "step": 950
    },
    {
      "epoch": 0.93,
      "learning_rate": 7.500000000000001e-05,
      "loss": 1.3142,
      "step": 1000
    },
    {
      "epoch": 0.93,
      "eval_loss": 0.8834471702575684,
      "eval_runtime": 3.0917,
      "eval_samples_per_second": 32.344,
      "eval_steps_per_second": 4.205,
      "step": 1000
    },
    {
      "epoch": 0.98,
      "eval_loss": 0.8460087776184082,
      "eval_runtime": 3.0708,
      "eval_samples_per_second": 32.565,
      "eval_steps_per_second": 4.233,
      "step": 1050
    },
    {
      "epoch": 1.03,
      "eval_loss": 0.8248443007469177,
      "eval_runtime": 3.062,
      "eval_samples_per_second": 32.658,
      "eval_steps_per_second": 4.246,
      "step": 1100
    },
    {
      "epoch": 1.07,
      "eval_loss": 0.8010109066963196,
      "eval_runtime": 3.062,
      "eval_samples_per_second": 32.659,
      "eval_steps_per_second": 4.246,
      "step": 1150
    },
    {
      "epoch": 1.12,
      "eval_loss": 0.7615901231765747,
      "eval_runtime": 3.0705,
      "eval_samples_per_second": 32.568,
      "eval_steps_per_second": 4.234,
      "step": 1200
    },
    {
      "epoch": 1.17,
      "eval_loss": 0.750758945941925,
      "eval_runtime": 3.0726,
      "eval_samples_per_second": 32.545,
      "eval_steps_per_second": 4.231,
      "step": 1250
    },
    {
      "epoch": 1.21,
      "eval_loss": 0.7553930878639221,
      "eval_runtime": 3.0632,
      "eval_samples_per_second": 32.645,
      "eval_steps_per_second": 4.244,
      "step": 1300
    },
    {
      "epoch": 1.26,
      "eval_loss": 0.714480459690094,
      "eval_runtime": 3.0588,
      "eval_samples_per_second": 32.693,
      "eval_steps_per_second": 4.25,
      "step": 1350
    },
    {
      "epoch": 1.31,
      "eval_loss": 0.6865142583847046,
      "eval_runtime": 3.0601,
      "eval_samples_per_second": 32.678,
      "eval_steps_per_second": 4.248,
      "step": 1400
    },
    {
      "epoch": 1.35,
      "eval_loss": 0.6855456829071045,
      "eval_runtime": 3.0743,
      "eval_samples_per_second": 32.527,
      "eval_steps_per_second": 4.229,
      "step": 1450
    },
    {
      "epoch": 1.4,
      "learning_rate": 6.25e-05,
      "loss": 0.9985,
      "step": 1500
    },
    {
      "epoch": 1.4,
      "eval_loss": 0.6646093726158142,
      "eval_runtime": 3.0899,
      "eval_samples_per_second": 32.363,
      "eval_steps_per_second": 4.207,
      "step": 1500
    },
    {
      "epoch": 1.45,
      "eval_loss": 0.6673924922943115,
      "eval_runtime": 3.0744,
      "eval_samples_per_second": 32.526,
      "eval_steps_per_second": 4.228,
      "step": 1550
    },
    {
      "epoch": 1.49,
      "eval_loss": 0.6436119675636292,
      "eval_runtime": 3.0612,
      "eval_samples_per_second": 32.667,
      "eval_steps_per_second": 4.247,
      "step": 1600
    },
    {
      "epoch": 1.54,
      "eval_loss": 0.6476317644119263,
      "eval_runtime": 3.0724,
      "eval_samples_per_second": 32.548,
      "eval_steps_per_second": 4.231,
      "step": 1650
    },
    {
      "epoch": 1.59,
      "eval_loss": 0.6251476407051086,
      "eval_runtime": 3.0506,
      "eval_samples_per_second": 32.78,
      "eval_steps_per_second": 4.261,
      "step": 1700
    },
    {
      "epoch": 1.63,
      "eval_loss": 0.6191129088401794,
      "eval_runtime": 3.0634,
      "eval_samples_per_second": 32.643,
      "eval_steps_per_second": 4.244,
      "step": 1750
    },
    {
      "epoch": 1.68,
      "eval_loss": 0.6157693862915039,
      "eval_runtime": 3.084,
      "eval_samples_per_second": 32.426,
      "eval_steps_per_second": 4.215,
      "step": 1800
    },
    {
      "epoch": 1.73,
      "eval_loss": 0.5934409499168396,
      "eval_runtime": 3.0678,
      "eval_samples_per_second": 32.597,
      "eval_steps_per_second": 4.238,
      "step": 1850
    },
    {
      "epoch": 1.77,
      "eval_loss": 0.5992363691329956,
      "eval_runtime": 3.0628,
      "eval_samples_per_second": 32.649,
      "eval_steps_per_second": 4.244,
      "step": 1900
    },
    {
      "epoch": 1.82,
      "eval_loss": 0.5798881649971008,
      "eval_runtime": 3.0658,
      "eval_samples_per_second": 32.618,
      "eval_steps_per_second": 4.24,
      "step": 1950
    },
    {
      "epoch": 1.87,
      "learning_rate": 5e-05,
      "loss": 0.8817,
      "step": 2000
    },
    {
      "epoch": 1.87,
      "eval_loss": 0.5694944858551025,
      "eval_runtime": 3.0588,
      "eval_samples_per_second": 32.692,
      "eval_steps_per_second": 4.25,
      "step": 2000
    },
    {
      "epoch": 1.91,
      "eval_loss": 0.5600036382675171,
      "eval_runtime": 3.073,
      "eval_samples_per_second": 32.542,
      "eval_steps_per_second": 4.23,
      "step": 2050
    },
    {
      "epoch": 1.96,
      "eval_loss": 0.5443231463432312,
      "eval_runtime": 3.0723,
      "eval_samples_per_second": 32.549,
      "eval_steps_per_second": 4.231,
      "step": 2100
    },
    {
      "epoch": 2.01,
      "eval_loss": 0.5324158668518066,
      "eval_runtime": 3.0759,
      "eval_samples_per_second": 32.511,
      "eval_steps_per_second": 4.226,
      "step": 2150
    },
    {
      "epoch": 2.05,
      "eval_loss": 0.5168842077255249,
      "eval_runtime": 3.0777,
      "eval_samples_per_second": 32.492,
      "eval_steps_per_second": 4.224,
      "step": 2200
    },
    {
      "epoch": 2.1,
      "eval_loss": 0.51568603515625,
      "eval_runtime": 3.0636,
      "eval_samples_per_second": 32.641,
      "eval_steps_per_second": 4.243,
      "step": 2250
    },
    {
      "epoch": 2.15,
      "eval_loss": 0.506223738193512,
      "eval_runtime": 3.0803,
      "eval_samples_per_second": 32.464,
      "eval_steps_per_second": 4.22,
      "step": 2300
    },
    {
      "epoch": 2.19,
      "eval_loss": 0.5046885013580322,
      "eval_runtime": 3.0565,
      "eval_samples_per_second": 32.717,
      "eval_steps_per_second": 4.253,
      "step": 2350
    },
    {
      "epoch": 2.24,
      "eval_loss": 0.4832480251789093,
      "eval_runtime": 3.0606,
      "eval_samples_per_second": 32.673,
      "eval_steps_per_second": 4.247,
      "step": 2400
    },
    {
      "epoch": 2.29,
      "eval_loss": 0.4789283275604248,
      "eval_runtime": 3.0598,
      "eval_samples_per_second": 32.682,
      "eval_steps_per_second": 4.249,
      "step": 2450
    },
    {
      "epoch": 2.33,
      "learning_rate": 3.7500000000000003e-05,
      "loss": 0.7454,
      "step": 2500
    },
    {
      "epoch": 2.33,
      "eval_loss": 0.4778507947921753,
      "eval_runtime": 3.0634,
      "eval_samples_per_second": 32.644,
      "eval_steps_per_second": 4.244,
      "step": 2500
    },
    {
      "epoch": 2.38,
      "eval_loss": 0.4714885652065277,
      "eval_runtime": 3.0662,
      "eval_samples_per_second": 32.614,
      "eval_steps_per_second": 4.24,
      "step": 2550
    },
    {
      "epoch": 2.43,
      "eval_loss": 0.4641421437263489,
      "eval_runtime": 3.0498,
      "eval_samples_per_second": 32.789,
      "eval_steps_per_second": 4.263,
      "step": 2600
    },
    {
      "epoch": 2.47,
      "eval_loss": 0.45793575048446655,
      "eval_runtime": 3.0673,
      "eval_samples_per_second": 32.602,
      "eval_steps_per_second": 4.238,
      "step": 2650
    },
    {
      "epoch": 2.52,
      "eval_loss": 0.44898709654808044,
      "eval_runtime": 3.0751,
      "eval_samples_per_second": 32.519,
      "eval_steps_per_second": 4.227,
      "step": 2700
    },
    {
      "epoch": 2.57,
      "eval_loss": 0.44389522075653076,
      "eval_runtime": 3.0534,
      "eval_samples_per_second": 32.751,
      "eval_steps_per_second": 4.258,
      "step": 2750
    },
    {
      "epoch": 2.61,
      "eval_loss": 0.4495103061199188,
      "eval_runtime": 3.0657,
      "eval_samples_per_second": 32.619,
      "eval_steps_per_second": 4.24,
      "step": 2800
    },
    {
      "epoch": 2.66,
      "eval_loss": 0.4390113055706024,
      "eval_runtime": 3.0609,
      "eval_samples_per_second": 32.67,
      "eval_steps_per_second": 4.247,
      "step": 2850
    },
    {
      "epoch": 2.71,
      "eval_loss": 0.43222054839134216,
      "eval_runtime": 3.0574,
      "eval_samples_per_second": 32.707,
      "eval_steps_per_second": 4.252,
      "step": 2900
    },
    {
      "epoch": 2.75,
      "eval_loss": 0.4251094460487366,
      "eval_runtime": 3.0796,
      "eval_samples_per_second": 32.472,
      "eval_steps_per_second": 4.221,
      "step": 2950
    },
    {
      "epoch": 2.8,
      "learning_rate": 2.5e-05,
      "loss": 0.6646,
      "step": 3000
    },
    {
      "epoch": 2.8,
      "eval_loss": 0.42413151264190674,
      "eval_runtime": 3.0651,
      "eval_samples_per_second": 32.626,
      "eval_steps_per_second": 4.241,
      "step": 3000
    },
    {
      "epoch": 2.85,
      "eval_loss": 0.42399895191192627,
      "eval_runtime": 3.0493,
      "eval_samples_per_second": 32.794,
      "eval_steps_per_second": 4.263,
      "step": 3050
    },
    {
      "epoch": 2.89,
      "eval_loss": 0.41698771715164185,
      "eval_runtime": 3.0915,
      "eval_samples_per_second": 32.346,
      "eval_steps_per_second": 4.205,
      "step": 3100
    },
    {
      "epoch": 2.94,
      "eval_loss": 0.41834643483161926,
      "eval_runtime": 3.0681,
      "eval_samples_per_second": 32.594,
      "eval_steps_per_second": 4.237,
      "step": 3150
    },
    {
      "epoch": 2.99,
      "eval_loss": 0.4123823940753937,
      "eval_runtime": 3.0706,
      "eval_samples_per_second": 32.567,
      "eval_steps_per_second": 4.234,
      "step": 3200
    },
    {
      "epoch": 3.03,
      "eval_loss": 0.4095262885093689,
      "eval_runtime": 3.0601,
      "eval_samples_per_second": 32.679,
      "eval_steps_per_second": 4.248,
      "step": 3250
    },
    {
      "epoch": 3.08,
      "eval_loss": 0.40909576416015625,
      "eval_runtime": 3.0757,
      "eval_samples_per_second": 32.513,
      "eval_steps_per_second": 4.227,
      "step": 3300
    },
    {
      "epoch": 3.13,
      "eval_loss": 0.4018322229385376,
      "eval_runtime": 3.0657,
      "eval_samples_per_second": 32.619,
      "eval_steps_per_second": 4.24,
      "step": 3350
    },
    {
      "epoch": 3.17,
      "eval_loss": 0.40012139081954956,
      "eval_runtime": 3.0562,
      "eval_samples_per_second": 32.72,
      "eval_steps_per_second": 4.254,
      "step": 3400
    },
    {
      "epoch": 3.22,
      "eval_loss": 0.39826804399490356,
      "eval_runtime": 3.0643,
      "eval_samples_per_second": 32.633,
      "eval_steps_per_second": 4.242,
      "step": 3450
    },
    {
      "epoch": 3.27,
      "learning_rate": 1.25e-05,
      "loss": 0.6115,
      "step": 3500
    },
    {
      "epoch": 3.27,
      "eval_loss": 0.3965628147125244,
      "eval_runtime": 3.0689,
      "eval_samples_per_second": 32.585,
      "eval_steps_per_second": 4.236,
      "step": 3500
    },
    {
      "epoch": 3.31,
      "eval_loss": 0.39603424072265625,
      "eval_runtime": 3.0553,
      "eval_samples_per_second": 32.73,
      "eval_steps_per_second": 4.255,
      "step": 3550
    },
    {
      "epoch": 3.36,
      "eval_loss": 0.395506352186203,
      "eval_runtime": 3.0847,
      "eval_samples_per_second": 32.418,
      "eval_steps_per_second": 4.214,
      "step": 3600
    },
    {
      "epoch": 3.41,
      "eval_loss": 0.38715580105781555,
      "eval_runtime": 3.0716,
      "eval_samples_per_second": 32.556,
      "eval_steps_per_second": 4.232,
      "step": 3650
    },
    {
      "epoch": 3.45,
      "eval_loss": 0.3873971104621887,
      "eval_runtime": 3.068,
      "eval_samples_per_second": 32.595,
      "eval_steps_per_second": 4.237,
      "step": 3700
    },
    {
      "epoch": 3.5,
      "eval_loss": 0.3866007626056671,
      "eval_runtime": 3.0802,
      "eval_samples_per_second": 32.466,
      "eval_steps_per_second": 4.221,
      "step": 3750
    },
    {
      "epoch": 3.55,
      "eval_loss": 0.3842252492904663,
      "eval_runtime": 3.0659,
      "eval_samples_per_second": 32.617,
      "eval_steps_per_second": 4.24,
      "step": 3800
    },
    {
      "epoch": 3.59,
      "eval_loss": 0.3832497298717499,
      "eval_runtime": 3.0807,
      "eval_samples_per_second": 32.46,
      "eval_steps_per_second": 4.22,
      "step": 3850
    },
    {
      "epoch": 3.64,
      "eval_loss": 0.38256534934043884,
      "eval_runtime": 3.0625,
      "eval_samples_per_second": 32.653,
      "eval_steps_per_second": 4.245,
      "step": 3900
    },
    {
      "epoch": 3.69,
      "eval_loss": 0.381301611661911,
      "eval_runtime": 3.0615,
      "eval_samples_per_second": 32.664,
      "eval_steps_per_second": 4.246,
      "step": 3950
    },
    {
      "epoch": 3.73,
      "learning_rate": 0.0,
      "loss": 0.569,
      "step": 4000
    },
    {
      "epoch": 3.73,
      "eval_loss": 0.38127923011779785,
      "eval_runtime": 3.0687,
      "eval_samples_per_second": 32.587,
      "eval_steps_per_second": 4.236,
      "step": 4000
    }
  ],
  "logging_steps": 500,
  "max_steps": 4000,
  "num_train_epochs": 4,
  "save_steps": 1000,
  "total_flos": 5.477031683555328e+16,
  "trial_name": null,
  "trial_params": null
}