{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.7105774324562955,
  "global_step": 10001,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02,
      "learning_rate": 2e-05,
      "loss": 1.8914,
      "step": 50
    },
    {
      "epoch": 0.05,
      "learning_rate": 4e-05,
      "loss": 1.6734,
      "step": 100
    },
    {
      "epoch": 0.07,
      "learning_rate": 6e-05,
      "loss": 1.5728,
      "step": 150
    },
    {
      "epoch": 0.09,
      "learning_rate": 8e-05,
      "loss": 1.528,
      "step": 200
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0001,
      "loss": 1.4775,
      "step": 250
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00012,
      "loss": 1.4352,
      "step": 300
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.00014,
      "loss": 1.4069,
      "step": 350
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00016,
      "loss": 1.375,
      "step": 400
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00018,
      "loss": 1.367,
      "step": 450
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0002,
      "loss": 1.3565,
      "step": 500
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.00022000000000000003,
      "loss": 1.3423,
      "step": 550
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00024,
      "loss": 1.3311,
      "step": 600
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00026000000000000003,
      "loss": 1.3245,
      "step": 650
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00028,
      "loss": 1.3214,
      "step": 700
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00030000000000000003,
      "loss": 1.3247,
      "step": 750
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00032,
      "loss": 1.3089,
      "step": 800
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00034,
      "loss": 1.3194,
      "step": 850
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00036,
      "loss": 1.3225,
      "step": 900
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00038,
      "loss": 1.3254,
      "step": 950
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0004,
      "loss": 1.3139,
      "step": 1000
    },
    {
      "epoch": 0.47,
      "eval_loss": 1.5397893190383911,
      "eval_runtime": 119.9433,
      "eval_samples_per_second": 1511.548,
      "eval_steps_per_second": 5.911,
      "step": 1000
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0003977777777777778,
      "loss": 1.3313,
      "step": 1050
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00039555555555555557,
      "loss": 1.318,
      "step": 1100
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0003933333333333333,
      "loss": 1.3092,
      "step": 1150
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0003911111111111111,
      "loss": 1.3036,
      "step": 1200
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0003888888888888889,
      "loss": 1.2931,
      "step": 1250
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00038666666666666667,
      "loss": 1.2986,
      "step": 1300
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.00038444444444444447,
      "loss": 1.4597,
      "step": 1350
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0003822222222222223,
      "loss": 1.8909,
      "step": 1400
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.00038,
      "loss": 1.347,
      "step": 1450
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.00037777777777777777,
      "loss": 1.3039,
      "step": 1500
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.00037555555555555557,
      "loss": 1.2929,
      "step": 1550
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.0003733333333333334,
      "loss": 1.2783,
      "step": 1600
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0003711111111111111,
      "loss": 1.2644,
      "step": 1650
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0003688888888888889,
      "loss": 1.255,
      "step": 1700
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.00036666666666666667,
      "loss": 1.2509,
      "step": 1750
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.00036444444444444447,
      "loss": 1.254,
      "step": 1800
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.0003622222222222222,
      "loss": 1.2436,
      "step": 1850
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.00036,
      "loss": 1.2499,
      "step": 1900
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.0003577777777777778,
      "loss": 1.5927,
      "step": 1950
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.00035555555555555557,
      "loss": 1.4569,
      "step": 2000
    },
    {
      "epoch": 0.94,
      "eval_loss": 1.661188006401062,
      "eval_runtime": 106.3952,
      "eval_samples_per_second": 1704.024,
      "eval_steps_per_second": 6.664,
      "step": 2000
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.0003533333333333333,
      "loss": 1.333,
      "step": 2050
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.0003511111111111111,
      "loss": 1.4076,
      "step": 2100
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.0003488888888888889,
      "loss": 1.4328,
      "step": 2150
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.00034666666666666667,
      "loss": 1.3461,
      "step": 2200
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.0003444444444444445,
      "loss": 1.4831,
      "step": 2250
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.0003422222222222222,
      "loss": 1.4935,
      "step": 2300
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.00034,
      "loss": 1.4034,
      "step": 2350
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.00033777777777777777,
      "loss": 1.3652,
      "step": 2400
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.0003355555555555556,
      "loss": 1.2921,
      "step": 2450
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.0003333333333333334,
      "loss": 1.2564,
      "step": 2500
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0003311111111111111,
      "loss": 1.2344,
      "step": 2550
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.00032888888888888887,
      "loss": 1.2705,
      "step": 2600
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.0003266666666666667,
      "loss": 1.3214,
      "step": 2650
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.0003244444444444445,
      "loss": 1.2318,
      "step": 2700
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.0003222222222222222,
      "loss": 1.2029,
      "step": 2750
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.00032,
      "loss": 1.2056,
      "step": 2800
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.0003177777777777778,
      "loss": 1.1931,
      "step": 2850
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.0003155555555555556,
      "loss": 1.1812,
      "step": 2900
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.0003133333333333333,
      "loss": 1.1863,
      "step": 2950
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.0003111111111111111,
      "loss": 1.1795,
      "step": 3000
    },
    {
      "epoch": 1.41,
      "eval_loss": 1.5454002618789673,
      "eval_runtime": 104.0653,
      "eval_samples_per_second": 1742.176,
      "eval_steps_per_second": 6.813,
      "step": 3000
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.00030888888888888893,
      "loss": 1.1856,
      "step": 3050
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0003066666666666667,
      "loss": 1.1765,
      "step": 3100
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.0003044444444444444,
      "loss": 1.3244,
      "step": 3150
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.0003022222222222222,
      "loss": 1.2179,
      "step": 3200
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.00030000000000000003,
      "loss": 1.2114,
      "step": 3250
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.0002977777777777778,
      "loss": 1.1968,
      "step": 3300
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.0002955555555555556,
      "loss": 1.1798,
      "step": 3350
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0002933333333333333,
      "loss": 1.1851,
      "step": 3400
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.00029111111111111113,
      "loss": 1.2182,
      "step": 3450
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0002888888888888889,
      "loss": 1.1791,
      "step": 3500
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0002866666666666667,
      "loss": 1.1842,
      "step": 3550
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0002844444444444445,
      "loss": 1.2017,
      "step": 3600
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.00028222222222222223,
      "loss": 1.2376,
      "step": 3650
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.00028,
      "loss": 1.3329,
      "step": 3700
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0002777777777777778,
      "loss": 1.2929,
      "step": 3750
    },
    {
      "epoch": 1.79,
      "learning_rate": 0.0002755555555555556,
      "loss": 1.2387,
      "step": 3800
    },
    {
      "epoch": 1.81,
      "learning_rate": 0.00027333333333333333,
      "loss": 1.1944,
      "step": 3850
    },
    {
      "epoch": 1.84,
      "learning_rate": 0.00027111111111111113,
      "loss": 1.196,
      "step": 3900
    },
    {
      "epoch": 1.86,
      "learning_rate": 0.00026888888888888893,
      "loss": 1.1817,
      "step": 3950
    },
    {
      "epoch": 1.88,
      "learning_rate": 0.0002666666666666667,
      "loss": 1.1769,
      "step": 4000
    },
    {
      "epoch": 1.88,
      "eval_loss": 1.526174545288086,
      "eval_runtime": 104.8904,
      "eval_samples_per_second": 1728.471,
      "eval_steps_per_second": 6.759,
      "step": 4000
    },
    {
      "epoch": 1.91,
      "learning_rate": 0.00026444444444444443,
      "loss": 1.1606,
      "step": 4050
    },
    {
      "epoch": 1.93,
      "learning_rate": 0.00026222222222222223,
      "loss": 1.2703,
      "step": 4100
    },
    {
      "epoch": 1.95,
      "learning_rate": 0.00026000000000000003,
      "loss": 1.2824,
      "step": 4150
    },
    {
      "epoch": 1.98,
      "learning_rate": 0.00025777777777777783,
      "loss": 1.2228,
      "step": 4200
    },
    {
      "epoch": 2.0,
      "learning_rate": 0.00025555555555555553,
      "loss": 1.2338,
      "step": 4250
    },
    {
      "epoch": 2.03,
      "learning_rate": 0.00025333333333333333,
      "loss": 1.163,
      "step": 4300
    },
    {
      "epoch": 2.05,
      "learning_rate": 0.00025111111111111113,
      "loss": 1.1332,
      "step": 4350
    },
    {
      "epoch": 2.07,
      "learning_rate": 0.0002488888888888889,
      "loss": 1.1206,
      "step": 4400
    },
    {
      "epoch": 2.1,
      "learning_rate": 0.0002466666666666667,
      "loss": 1.1597,
      "step": 4450
    },
    {
      "epoch": 2.12,
      "learning_rate": 0.0002444444444444445,
      "loss": 1.1259,
      "step": 4500
    },
    {
      "epoch": 2.14,
      "learning_rate": 0.0002422222222222222,
      "loss": 1.1178,
      "step": 4550
    },
    {
      "epoch": 2.17,
      "learning_rate": 0.00024,
      "loss": 1.1119,
      "step": 4600
    },
    {
      "epoch": 2.19,
      "learning_rate": 0.00023777777777777778,
      "loss": 1.1104,
      "step": 4650
    },
    {
      "epoch": 2.21,
      "learning_rate": 0.00023555555555555558,
      "loss": 1.1064,
      "step": 4700
    },
    {
      "epoch": 2.24,
      "learning_rate": 0.00023333333333333336,
      "loss": 1.1111,
      "step": 4750
    },
    {
      "epoch": 2.26,
      "learning_rate": 0.0002311111111111111,
      "loss": 1.0994,
      "step": 4800
    },
    {
      "epoch": 2.28,
      "learning_rate": 0.00022888888888888888,
      "loss": 1.1002,
      "step": 4850
    },
    {
      "epoch": 2.31,
      "learning_rate": 0.00022666666666666668,
      "loss": 1.1486,
      "step": 4900
    },
    {
      "epoch": 2.33,
      "learning_rate": 0.00022444444444444446,
      "loss": 1.1115,
      "step": 4950
    },
    {
      "epoch": 2.36,
      "learning_rate": 0.00022222222222222223,
      "loss": 1.1026,
      "step": 5000
    },
    {
      "epoch": 2.36,
      "eval_loss": 1.4830693006515503,
      "eval_runtime": 104.1599,
      "eval_samples_per_second": 1740.593,
      "eval_steps_per_second": 6.807,
      "step": 5000
    },
    {
      "epoch": 2.38,
      "learning_rate": 0.00022000000000000003,
      "loss": 1.1098,
      "step": 5050
    },
    {
      "epoch": 2.4,
      "learning_rate": 0.00021777777777777778,
      "loss": 1.0984,
      "step": 5100
    },
    {
      "epoch": 2.43,
      "learning_rate": 0.00021555555555555556,
      "loss": 1.0858,
      "step": 5150
    },
    {
      "epoch": 2.45,
      "learning_rate": 0.00021333333333333333,
      "loss": 1.1243,
      "step": 5200
    },
    {
      "epoch": 2.47,
      "learning_rate": 0.00021111111111111113,
      "loss": 1.125,
      "step": 5250
    },
    {
      "epoch": 2.5,
      "learning_rate": 0.0002088888888888889,
      "loss": 1.108,
      "step": 5300
    },
    {
      "epoch": 2.52,
      "learning_rate": 0.00020666666666666668,
      "loss": 1.0821,
      "step": 5350
    },
    {
      "epoch": 2.54,
      "learning_rate": 0.00020444444444444443,
      "loss": 1.087,
      "step": 5400
    },
    {
      "epoch": 2.57,
      "learning_rate": 0.00020222222222222223,
      "loss": 1.0814,
      "step": 5450
    },
    {
      "epoch": 2.59,
      "learning_rate": 0.0002,
      "loss": 1.0799,
      "step": 5500
    },
    {
      "epoch": 2.61,
      "learning_rate": 0.00019777777777777778,
      "loss": 1.0771,
      "step": 5550
    },
    {
      "epoch": 2.64,
      "learning_rate": 0.00019555555555555556,
      "loss": 1.0777,
      "step": 5600
    },
    {
      "epoch": 2.66,
      "learning_rate": 0.00019333333333333333,
      "loss": 1.0765,
      "step": 5650
    },
    {
      "epoch": 2.68,
      "learning_rate": 0.00019111111111111114,
      "loss": 1.0703,
      "step": 5700
    },
    {
      "epoch": 2.71,
      "learning_rate": 0.00018888888888888888,
      "loss": 1.0722,
      "step": 5750
    },
    {
      "epoch": 2.73,
      "learning_rate": 0.0001866666666666667,
      "loss": 1.0917,
      "step": 5800
    },
    {
      "epoch": 2.76,
      "learning_rate": 0.00018444444444444446,
      "loss": 1.0741,
      "step": 5850
    },
    {
      "epoch": 2.78,
      "learning_rate": 0.00018222222222222224,
      "loss": 1.077,
      "step": 5900
    },
    {
      "epoch": 2.8,
      "learning_rate": 0.00018,
      "loss": 1.0683,
      "step": 5950
    },
    {
      "epoch": 2.83,
      "learning_rate": 0.00017777777777777779,
      "loss": 1.0711,
      "step": 6000
    },
    {
      "epoch": 2.83,
      "eval_loss": 1.4865516424179077,
      "eval_runtime": 104.1916,
      "eval_samples_per_second": 1740.063,
      "eval_steps_per_second": 6.805,
      "step": 6000
    },
    {
      "epoch": 2.85,
      "learning_rate": 0.00017555555555555556,
      "loss": 1.0955,
      "step": 6050
    },
    {
      "epoch": 2.87,
      "learning_rate": 0.00017333333333333334,
      "loss": 1.1421,
      "step": 6100
    },
    {
      "epoch": 2.9,
      "learning_rate": 0.0001711111111111111,
      "loss": 1.1308,
      "step": 6150
    },
    {
      "epoch": 2.92,
      "learning_rate": 0.00016888888888888889,
      "loss": 1.0915,
      "step": 6200
    },
    {
      "epoch": 2.94,
      "learning_rate": 0.0001666666666666667,
      "loss": 1.0693,
      "step": 6250
    },
    {
      "epoch": 2.97,
      "learning_rate": 0.00016444444444444444,
      "loss": 1.0804,
      "step": 6300
    },
    {
      "epoch": 2.99,
      "learning_rate": 0.00016222222222222224,
      "loss": 1.0684,
      "step": 6350
    },
    {
      "epoch": 3.01,
      "learning_rate": 0.00016,
      "loss": 1.0886,
      "step": 6400
    },
    {
      "epoch": 3.04,
      "learning_rate": 0.0001577777777777778,
      "loss": 1.0576,
      "step": 6450
    },
    {
      "epoch": 3.06,
      "learning_rate": 0.00015555555555555556,
      "loss": 1.0446,
      "step": 6500
    },
    {
      "epoch": 3.09,
      "learning_rate": 0.00015333333333333334,
      "loss": 1.0426,
      "step": 6550
    },
    {
      "epoch": 3.11,
      "learning_rate": 0.0001511111111111111,
      "loss": 1.0404,
      "step": 6600
    },
    {
      "epoch": 3.13,
      "learning_rate": 0.0001488888888888889,
      "loss": 1.0446,
      "step": 6650
    },
    {
      "epoch": 3.16,
      "learning_rate": 0.00014666666666666666,
      "loss": 1.0564,
      "step": 6700
    },
    {
      "epoch": 3.18,
      "learning_rate": 0.00014444444444444444,
      "loss": 1.0452,
      "step": 6750
    },
    {
      "epoch": 3.2,
      "learning_rate": 0.00014222222222222224,
      "loss": 1.0327,
      "step": 6800
    },
    {
      "epoch": 3.23,
      "learning_rate": 0.00014,
      "loss": 1.0353,
      "step": 6850
    },
    {
      "epoch": 3.25,
      "learning_rate": 0.0001377777777777778,
      "loss": 1.0259,
      "step": 6900
    },
    {
      "epoch": 3.27,
      "learning_rate": 0.00013555555555555556,
      "loss": 1.0328,
      "step": 6950
    },
    {
      "epoch": 3.3,
      "learning_rate": 0.00013333333333333334,
      "loss": 1.0321,
      "step": 7000
    },
    {
      "epoch": 3.3,
      "eval_loss": 1.4409459829330444,
      "eval_runtime": 104.0221,
      "eval_samples_per_second": 1742.9,
      "eval_steps_per_second": 6.816,
      "step": 7000
    },
    {
      "epoch": 3.32,
      "learning_rate": 0.00013111111111111111,
      "loss": 1.0295,
      "step": 7050
    },
    {
      "epoch": 3.34,
      "learning_rate": 0.00012888888888888892,
      "loss": 1.0262,
      "step": 7100
    },
    {
      "epoch": 3.37,
      "learning_rate": 0.00012666666666666666,
      "loss": 1.0278,
      "step": 7150
    },
    {
      "epoch": 3.39,
      "learning_rate": 0.00012444444444444444,
      "loss": 1.0227,
      "step": 7200
    },
    {
      "epoch": 3.41,
      "learning_rate": 0.00012222222222222224,
      "loss": 1.0262,
      "step": 7250
    },
    {
      "epoch": 3.44,
      "learning_rate": 0.00012,
      "loss": 1.0205,
      "step": 7300
    },
    {
      "epoch": 3.46,
      "learning_rate": 0.00011777777777777779,
      "loss": 1.0107,
      "step": 7350
    },
    {
      "epoch": 3.49,
      "learning_rate": 0.00011555555555555555,
      "loss": 1.0218,
      "step": 7400
    },
    {
      "epoch": 3.51,
      "learning_rate": 0.00011333333333333334,
      "loss": 1.0205,
      "step": 7450
    },
    {
      "epoch": 3.53,
      "learning_rate": 0.00011111111111111112,
      "loss": 1.0196,
      "step": 7500
    },
    {
      "epoch": 3.56,
      "learning_rate": 0.00010888888888888889,
      "loss": 1.0197,
      "step": 7550
    },
    {
      "epoch": 3.58,
      "learning_rate": 0.00010666666666666667,
      "loss": 1.0139,
      "step": 7600
    },
    {
      "epoch": 3.6,
      "learning_rate": 0.00010444444444444445,
      "loss": 1.0011,
      "step": 7650
    },
    {
      "epoch": 3.63,
      "learning_rate": 0.00010222222222222222,
      "loss": 1.0056,
      "step": 7700
    },
    {
      "epoch": 3.65,
      "learning_rate": 0.0001,
      "loss": 1.0152,
      "step": 7750
    },
    {
      "epoch": 3.67,
      "learning_rate": 9.777777777777778e-05,
      "loss": 1.0078,
      "step": 7800
    },
    {
      "epoch": 3.7,
      "learning_rate": 9.555555555555557e-05,
      "loss": 1.0101,
      "step": 7850
    },
    {
      "epoch": 3.72,
      "learning_rate": 9.333333333333334e-05,
      "loss": 0.9983,
      "step": 7900
    },
    {
      "epoch": 3.74,
      "learning_rate": 9.111111111111112e-05,
      "loss": 0.9922,
      "step": 7950
    },
    {
      "epoch": 3.77,
      "learning_rate": 8.888888888888889e-05,
      "loss": 0.9994,
      "step": 8000
    },
    {
      "epoch": 3.77,
      "eval_loss": 1.420628547668457,
      "eval_runtime": 104.1119,
      "eval_samples_per_second": 1741.396,
      "eval_steps_per_second": 6.81,
      "step": 8000
    },
    {
      "epoch": 3.79,
      "learning_rate": 8.666666666666667e-05,
      "loss": 1.0066,
      "step": 8050
    },
    {
      "epoch": 3.82,
      "learning_rate": 8.444444444444444e-05,
      "loss": 1.0028,
      "step": 8100
    },
    {
      "epoch": 3.84,
      "learning_rate": 8.222222222222222e-05,
      "loss": 1.0007,
      "step": 8150
    },
    {
      "epoch": 3.86,
      "learning_rate": 8e-05,
      "loss": 1.0081,
      "step": 8200
    },
    {
      "epoch": 3.89,
      "learning_rate": 7.777777777777778e-05,
      "loss": 0.9953,
      "step": 8250
    },
    {
      "epoch": 3.91,
      "learning_rate": 7.555555555555556e-05,
      "loss": 0.9866,
      "step": 8300
    },
    {
      "epoch": 3.93,
      "learning_rate": 7.333333333333333e-05,
      "loss": 0.9961,
      "step": 8350
    },
    {
      "epoch": 3.96,
      "learning_rate": 7.111111111111112e-05,
      "loss": 0.9891,
      "step": 8400
    },
    {
      "epoch": 3.98,
      "learning_rate": 6.88888888888889e-05,
      "loss": 0.9917,
      "step": 8450
    },
    {
      "epoch": 4.0,
      "learning_rate": 6.666666666666667e-05,
      "loss": 1.0035,
      "step": 8500
    },
    {
      "epoch": 4.03,
      "learning_rate": 6.444444444444446e-05,
      "loss": 0.9899,
      "step": 8550
    },
    {
      "epoch": 4.05,
      "learning_rate": 6.222222222222222e-05,
      "loss": 0.9769,
      "step": 8600
    },
    {
      "epoch": 4.07,
      "learning_rate": 6e-05,
      "loss": 0.9781,
      "step": 8650
    },
    {
      "epoch": 4.1,
      "learning_rate": 5.7777777777777776e-05,
      "loss": 0.9802,
      "step": 8700
    },
    {
      "epoch": 4.12,
      "learning_rate": 5.555555555555556e-05,
      "loss": 0.9727,
      "step": 8750
    },
    {
      "epoch": 4.15,
      "learning_rate": 5.333333333333333e-05,
      "loss": 0.9742,
      "step": 8800
    },
    {
      "epoch": 4.17,
      "learning_rate": 5.111111111111111e-05,
      "loss": 0.9723,
      "step": 8850
    },
    {
      "epoch": 4.19,
      "learning_rate": 4.888888888888889e-05,
      "loss": 0.9809,
      "step": 8900
    },
    {
      "epoch": 4.22,
      "learning_rate": 4.666666666666667e-05,
      "loss": 0.9753,
      "step": 8950
    },
    {
      "epoch": 4.24,
      "learning_rate": 4.4444444444444447e-05,
      "loss": 0.9774,
      "step": 9000
    },
    {
      "epoch": 4.24,
      "eval_loss": 1.3941148519515991,
      "eval_runtime": 104.2747,
      "eval_samples_per_second": 1738.677,
      "eval_steps_per_second": 6.799,
      "step": 9000
    },
    {
      "epoch": 4.26,
      "learning_rate": 4.222222222222222e-05,
      "loss": 0.9603,
      "step": 9050
    },
    {
      "epoch": 4.29,
      "learning_rate": 4e-05,
      "loss": 0.9632,
      "step": 9100
    },
    {
      "epoch": 4.31,
      "learning_rate": 3.777777777777778e-05,
      "loss": 0.9701,
      "step": 9150
    },
    {
      "epoch": 4.33,
      "learning_rate": 3.555555555555556e-05,
      "loss": 0.966,
      "step": 9200
    },
    {
      "epoch": 4.36,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.9725,
      "step": 9250
    },
    {
      "epoch": 4.38,
      "learning_rate": 3.111111111111111e-05,
      "loss": 0.9644,
      "step": 9300
    },
    {
      "epoch": 4.4,
      "learning_rate": 2.8888888888888888e-05,
      "loss": 0.9613,
      "step": 9350
    },
    {
      "epoch": 4.43,
      "learning_rate": 2.6666666666666667e-05,
      "loss": 0.971,
      "step": 9400
    },
    {
      "epoch": 4.45,
      "learning_rate": 2.4444444444444445e-05,
      "loss": 0.9592,
      "step": 9450
    },
    {
      "epoch": 4.47,
      "learning_rate": 2.2222222222222223e-05,
      "loss": 0.9692,
      "step": 9500
    },
    {
      "epoch": 4.5,
      "learning_rate": 2e-05,
      "loss": 0.9639,
      "step": 9550
    },
    {
      "epoch": 4.52,
      "learning_rate": 1.777777777777778e-05,
      "loss": 0.9562,
      "step": 9600
    },
    {
      "epoch": 4.55,
      "learning_rate": 1.5555555555555555e-05,
      "loss": 0.9645,
      "step": 9650
    },
    {
      "epoch": 4.57,
      "learning_rate": 1.3333333333333333e-05,
      "loss": 0.9638,
      "step": 9700
    },
    {
      "epoch": 4.59,
      "learning_rate": 1.1111111111111112e-05,
      "loss": 0.9614,
      "step": 9750
    },
    {
      "epoch": 4.62,
      "learning_rate": 8.88888888888889e-06,
      "loss": 0.9607,
      "step": 9800
    },
    {
      "epoch": 4.64,
      "learning_rate": 6.666666666666667e-06,
      "loss": 0.9602,
      "step": 9850
    },
    {
      "epoch": 4.66,
      "learning_rate": 4.444444444444445e-06,
      "loss": 0.9643,
      "step": 9900
    },
    {
      "epoch": 4.69,
      "learning_rate": 2.2222222222222225e-06,
      "loss": 0.9525,
      "step": 9950
    },
    {
      "epoch": 4.71,
      "learning_rate": 0.0,
      "loss": 0.9618,
      "step": 10000
    },
    {
      "epoch": 4.71,
      "eval_loss": 1.3830920457839966,
      "eval_runtime": 104.4628,
      "eval_samples_per_second": 1735.547,
      "eval_steps_per_second": 6.787,
      "step": 10000
    },
    {
      "epoch": 4.71,
      "step": 10001,
      "total_flos": 1.685090114171044e+17,
      "train_loss": 0.00010499673156234791,
      "train_runtime": 1054.6912,
      "train_samples_per_second": 19418.005,
      "train_steps_per_second": 9.481
    }
  ],
  "max_steps": 10000,
  "num_train_epochs": 5,
  "total_flos": 1.685090114171044e+17,
  "trial_name": null,
  "trial_params": null
}