{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 75.27846767517121,
  "global_step": 250000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.15,
      "learning_rate": 1e-05,
      "learning_rate_embeddings": 1e-05,
      "loss": 7.3599,
      "step": 500
    },
    {
      "epoch": 0.3,
      "learning_rate": 2e-05,
      "learning_rate_embeddings": 2e-05,
      "loss": 6.1511,
      "step": 1000
    },
    {
      "epoch": 0.45,
      "learning_rate": 3e-05,
      "learning_rate_embeddings": 3e-05,
      "loss": 5.6955,
      "step": 1500
    },
    {
      "epoch": 0.6,
      "learning_rate": 4e-05,
      "learning_rate_embeddings": 4e-05,
      "loss": 5.3836,
      "step": 2000
    },
    {
      "epoch": 0.75,
      "learning_rate": 5e-05,
      "learning_rate_embeddings": 5e-05,
      "loss": 5.1255,
      "step": 2500
    },
    {
      "epoch": 0.9,
      "learning_rate": 6e-05,
      "learning_rate_embeddings": 6e-05,
      "loss": 4.9319,
      "step": 3000
    },
    {
      "epoch": 1.05,
      "learning_rate": 7.000000000000001e-05,
      "learning_rate_embeddings": 7.000000000000001e-05,
      "loss": 4.7633,
      "step": 3500
    },
    {
      "epoch": 1.2,
      "learning_rate": 8e-05,
      "learning_rate_embeddings": 8e-05,
      "loss": 4.6153,
      "step": 4000
    },
    {
      "epoch": 1.35,
      "learning_rate": 8.999999999999999e-05,
      "learning_rate_embeddings": 8.999999999999999e-05,
      "loss": 4.504,
      "step": 4500
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.0001,
      "learning_rate_embeddings": 0.0001,
      "loss": 4.4162,
      "step": 5000
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.00011,
      "learning_rate_embeddings": 0.00011,
      "loss": 4.3292,
      "step": 5500
    },
    {
      "epoch": 1.81,
      "learning_rate": 0.00012,
      "learning_rate_embeddings": 0.00012,
      "loss": 4.2659,
      "step": 6000
    },
    {
      "epoch": 1.96,
      "learning_rate": 0.00013000000000000002,
      "learning_rate_embeddings": 0.00013000000000000002,
      "loss": 4.2079,
      "step": 6500
    },
    {
      "epoch": 2.11,
      "learning_rate": 0.00014000000000000001,
      "learning_rate_embeddings": 0.00014000000000000001,
      "loss": 4.151,
      "step": 7000
    },
    {
      "epoch": 2.26,
      "learning_rate": 0.00015,
      "learning_rate_embeddings": 0.00015,
      "loss": 4.0975,
      "step": 7500
    },
    {
      "epoch": 2.41,
      "learning_rate": 0.00016,
      "learning_rate_embeddings": 0.00016,
      "loss": 4.0709,
      "step": 8000
    },
    {
      "epoch": 2.56,
      "learning_rate": 0.00017,
      "learning_rate_embeddings": 0.00017,
      "loss": 4.0275,
      "step": 8500
    },
    {
      "epoch": 2.71,
      "learning_rate": 0.00017999999999999998,
      "learning_rate_embeddings": 0.00017999999999999998,
      "loss": 3.9927,
      "step": 9000
    },
    {
      "epoch": 2.86,
      "learning_rate": 0.00019,
      "learning_rate_embeddings": 0.00019,
      "loss": 3.9632,
      "step": 9500
    },
    {
      "epoch": 3.01,
      "learning_rate": 0.0002,
      "learning_rate_embeddings": 0.0002,
      "loss": 3.9414,
      "step": 10000
    },
    {
      "epoch": 3.16,
      "learning_rate": 0.00021,
      "learning_rate_embeddings": 0.00021,
      "loss": 3.9043,
      "step": 10500
    },
    {
      "epoch": 3.31,
      "learning_rate": 0.00022,
      "learning_rate_embeddings": 0.00022,
      "loss": 3.8788,
      "step": 11000
    },
    {
      "epoch": 3.46,
      "learning_rate": 0.00023,
      "learning_rate_embeddings": 0.00023,
      "loss": 3.8635,
      "step": 11500
    },
    {
      "epoch": 3.61,
      "learning_rate": 0.00024,
      "learning_rate_embeddings": 0.00024,
      "loss": 3.8494,
      "step": 12000
    },
    {
      "epoch": 3.76,
      "learning_rate": 0.00025,
      "learning_rate_embeddings": 0.00025,
      "loss": 3.8293,
      "step": 12500
    },
    {
      "epoch": 3.76,
      "eval_loss": 3.760099172592163,
      "eval_runtime": 352.8255,
      "eval_samples_per_second": 481.133,
      "eval_steps_per_second": 3.761,
      "step": 12500
    },
    {
      "epoch": 3.91,
      "learning_rate": 0.00026000000000000003,
      "learning_rate_embeddings": 0.00026000000000000003,
      "loss": 3.8128,
      "step": 13000
    },
    {
      "epoch": 4.07,
      "learning_rate": 0.00027,
      "learning_rate_embeddings": 0.00027,
      "loss": 3.794,
      "step": 13500
    },
    {
      "epoch": 4.22,
      "learning_rate": 0.00028000000000000003,
      "learning_rate_embeddings": 0.00028000000000000003,
      "loss": 3.7704,
      "step": 14000
    },
    {
      "epoch": 4.37,
      "learning_rate": 0.00029,
      "learning_rate_embeddings": 0.00029,
      "loss": 3.7616,
      "step": 14500
    },
    {
      "epoch": 4.52,
      "learning_rate": 0.0003,
      "learning_rate_embeddings": 0.0003,
      "loss": 3.7506,
      "step": 15000
    },
    {
      "epoch": 4.67,
      "learning_rate": 0.00031,
      "learning_rate_embeddings": 0.00031,
      "loss": 3.7454,
      "step": 15500
    },
    {
      "epoch": 4.82,
      "learning_rate": 0.00032,
      "learning_rate_embeddings": 0.00032,
      "loss": 3.7273,
      "step": 16000
    },
    {
      "epoch": 4.97,
      "learning_rate": 0.00033,
      "learning_rate_embeddings": 0.00033,
      "loss": 3.7151,
      "step": 16500
    },
    {
      "epoch": 5.12,
      "learning_rate": 0.00034,
      "learning_rate_embeddings": 0.00034,
      "loss": 3.7035,
      "step": 17000
    },
    {
      "epoch": 5.27,
      "learning_rate": 0.00035,
      "learning_rate_embeddings": 0.00035,
      "loss": 3.6791,
      "step": 17500
    },
    {
      "epoch": 5.42,
      "learning_rate": 0.00035999999999999997,
      "learning_rate_embeddings": 0.00035999999999999997,
      "loss": 3.6797,
      "step": 18000
    },
    {
      "epoch": 5.57,
      "learning_rate": 0.00037,
      "learning_rate_embeddings": 0.00037,
      "loss": 3.6741,
      "step": 18500
    },
    {
      "epoch": 5.72,
      "learning_rate": 0.00038,
      "learning_rate_embeddings": 0.00038,
      "loss": 3.6601,
      "step": 19000
    },
    {
      "epoch": 5.87,
      "learning_rate": 0.00039000000000000005,
      "learning_rate_embeddings": 0.00039000000000000005,
      "loss": 3.6603,
      "step": 19500
    },
    {
      "epoch": 6.02,
      "learning_rate": 0.0004,
      "learning_rate_embeddings": 0.0004,
      "loss": 3.6538,
      "step": 20000
    },
    {
      "epoch": 6.17,
      "learning_rate": 0.00041,
      "learning_rate_embeddings": 0.00041,
      "loss": 3.6284,
      "step": 20500
    },
    {
      "epoch": 6.32,
      "learning_rate": 0.00042,
      "learning_rate_embeddings": 0.00042,
      "loss": 3.6258,
      "step": 21000
    },
    {
      "epoch": 6.47,
      "learning_rate": 0.00043,
      "learning_rate_embeddings": 0.00043,
      "loss": 3.6147,
      "step": 21500
    },
    {
      "epoch": 6.62,
      "learning_rate": 0.00044,
      "learning_rate_embeddings": 0.00044,
      "loss": 3.6086,
      "step": 22000
    },
    {
      "epoch": 6.77,
      "learning_rate": 0.00045000000000000004,
      "learning_rate_embeddings": 0.00045000000000000004,
      "loss": 3.6107,
      "step": 22500
    },
    {
      "epoch": 6.93,
      "learning_rate": 0.00046,
      "learning_rate_embeddings": 0.00046,
      "loss": 3.5963,
      "step": 23000
    },
    {
      "epoch": 7.08,
      "learning_rate": 0.00047,
      "learning_rate_embeddings": 0.00047,
      "loss": 3.5954,
      "step": 23500
    },
    {
      "epoch": 7.23,
      "learning_rate": 0.00048,
      "learning_rate_embeddings": 0.00048,
      "loss": 3.5742,
      "step": 24000
    },
    {
      "epoch": 7.38,
      "learning_rate": 0.00049,
      "learning_rate_embeddings": 0.00049,
      "loss": 3.5729,
      "step": 24500
    },
    {
      "epoch": 7.53,
      "learning_rate": 0.0005,
      "learning_rate_embeddings": 0.0005,
      "loss": 3.5783,
      "step": 25000
    },
    {
      "epoch": 7.53,
      "eval_loss": 3.5366692543029785,
      "eval_runtime": 339.3978,
      "eval_samples_per_second": 500.168,
      "eval_steps_per_second": 3.91,
      "step": 25000
    },
    {
      "epoch": 7.68,
      "learning_rate": 0.0004988888888888889,
      "learning_rate_embeddings": 0.0004988888888888889,
      "loss": 3.5695,
      "step": 25500
    },
    {
      "epoch": 7.83,
      "learning_rate": 0.0004977777777777778,
      "learning_rate_embeddings": 0.0004977777777777778,
      "loss": 3.5695,
      "step": 26000
    },
    {
      "epoch": 7.98,
      "learning_rate": 0.0004966666666666666,
      "learning_rate_embeddings": 0.0004966666666666666,
      "loss": 3.5605,
      "step": 26500
    },
    {
      "epoch": 8.13,
      "learning_rate": 0.0004955555555555556,
      "learning_rate_embeddings": 0.0004955555555555556,
      "loss": 3.5402,
      "step": 27000
    },
    {
      "epoch": 8.28,
      "learning_rate": 0.0004944444444444445,
      "learning_rate_embeddings": 0.0004944444444444445,
      "loss": 3.5331,
      "step": 27500
    },
    {
      "epoch": 8.43,
      "learning_rate": 0.0004933333333333334,
      "learning_rate_embeddings": 0.0004933333333333334,
      "loss": 3.5318,
      "step": 28000
    },
    {
      "epoch": 8.58,
      "learning_rate": 0.0004922222222222222,
      "learning_rate_embeddings": 0.0004922222222222222,
      "loss": 3.5283,
      "step": 28500
    },
    {
      "epoch": 8.73,
      "learning_rate": 0.0004911111111111111,
      "learning_rate_embeddings": 0.0004911111111111111,
      "loss": 3.5284,
      "step": 29000
    },
    {
      "epoch": 8.88,
      "learning_rate": 0.00049,
      "learning_rate_embeddings": 0.00049,
      "loss": 3.5301,
      "step": 29500
    },
    {
      "epoch": 9.03,
      "learning_rate": 0.0004888888888888889,
      "learning_rate_embeddings": 0.0004888888888888889,
      "loss": 3.5205,
      "step": 30000
    },
    {
      "epoch": 9.18,
      "learning_rate": 0.0004877777777777778,
      "learning_rate_embeddings": 0.0004877777777777778,
      "loss": 3.5172,
      "step": 30500
    },
    {
      "epoch": 9.33,
      "learning_rate": 0.0004866666666666667,
      "learning_rate_embeddings": 0.0004866666666666667,
      "loss": 3.4955,
      "step": 31000
    },
    {
      "epoch": 9.48,
      "learning_rate": 0.0004855555555555556,
      "learning_rate_embeddings": 0.0004855555555555556,
      "loss": 3.5095,
      "step": 31500
    },
    {
      "epoch": 9.64,
      "learning_rate": 0.00048444444444444446,
      "learning_rate_embeddings": 0.00048444444444444446,
      "loss": 3.5006,
      "step": 32000
    },
    {
      "epoch": 9.79,
      "learning_rate": 0.00048333333333333334,
      "learning_rate_embeddings": 0.00048333333333333334,
      "loss": 3.5042,
      "step": 32500
    },
    {
      "epoch": 9.94,
      "learning_rate": 0.0004822222222222222,
      "learning_rate_embeddings": 0.0004822222222222222,
      "loss": 3.5016,
      "step": 33000
    },
    {
      "epoch": 10.09,
      "learning_rate": 0.0004811111111111111,
      "learning_rate_embeddings": 0.0004811111111111111,
      "loss": 3.4872,
      "step": 33500
    },
    {
      "epoch": 10.24,
      "learning_rate": 0.00048,
      "learning_rate_embeddings": 0.00048,
      "loss": 3.4749,
      "step": 34000
    },
    {
      "epoch": 10.39,
      "learning_rate": 0.0004788888888888889,
      "learning_rate_embeddings": 0.0004788888888888889,
      "loss": 3.4825,
      "step": 34500
    },
    {
      "epoch": 10.54,
      "learning_rate": 0.0004777777777777778,
      "learning_rate_embeddings": 0.0004777777777777778,
      "loss": 3.4735,
      "step": 35000
    },
    {
      "epoch": 10.69,
      "learning_rate": 0.0004766666666666667,
      "learning_rate_embeddings": 0.0004766666666666667,
      "loss": 3.4758,
      "step": 35500
    },
    {
      "epoch": 10.84,
      "learning_rate": 0.00047555555555555556,
      "learning_rate_embeddings": 0.00047555555555555556,
      "loss": 3.476,
      "step": 36000
    },
    {
      "epoch": 10.99,
      "learning_rate": 0.00047444444444444444,
      "learning_rate_embeddings": 0.00047444444444444444,
      "loss": 3.4827,
      "step": 36500
    },
    {
      "epoch": 11.14,
      "learning_rate": 0.00047333333333333336,
      "learning_rate_embeddings": 0.00047333333333333336,
      "loss": 3.4621,
      "step": 37000
    },
    {
      "epoch": 11.29,
      "learning_rate": 0.00047222222222222224,
      "learning_rate_embeddings": 0.00047222222222222224,
      "loss": 3.46,
      "step": 37500
    },
    {
      "epoch": 11.29,
      "eval_loss": 3.450908660888672,
      "eval_runtime": 339.3072,
      "eval_samples_per_second": 500.302,
      "eval_steps_per_second": 3.911,
      "step": 37500
    },
    {
      "epoch": 11.44,
      "learning_rate": 0.0004711111111111111,
      "learning_rate_embeddings": 0.0004711111111111111,
      "loss": 3.4578,
      "step": 38000
    },
    {
      "epoch": 11.59,
      "learning_rate": 0.00047,
      "learning_rate_embeddings": 0.00047,
      "loss": 3.464,
      "step": 38500
    },
    {
      "epoch": 11.74,
      "learning_rate": 0.0004688888888888889,
      "learning_rate_embeddings": 0.0004688888888888889,
      "loss": 3.4597,
      "step": 39000
    },
    {
      "epoch": 11.89,
      "learning_rate": 0.0004677777777777778,
      "learning_rate_embeddings": 0.0004677777777777778,
      "loss": 3.4515,
      "step": 39500
    },
    {
      "epoch": 12.04,
      "learning_rate": 0.00046666666666666666,
      "learning_rate_embeddings": 0.00046666666666666666,
      "loss": 3.4635,
      "step": 40000
    },
    {
      "epoch": 12.2,
      "learning_rate": 0.0004655555555555556,
      "learning_rate_embeddings": 0.0004655555555555556,
      "loss": 3.4398,
      "step": 40500
    },
    {
      "epoch": 12.35,
      "learning_rate": 0.00046444444444444446,
      "learning_rate_embeddings": 0.00046444444444444446,
      "loss": 3.4433,
      "step": 41000
    },
    {
      "epoch": 12.5,
      "learning_rate": 0.00046333333333333334,
      "learning_rate_embeddings": 0.00046333333333333334,
      "loss": 3.4479,
      "step": 41500
    },
    {
      "epoch": 12.65,
      "learning_rate": 0.0004622222222222222,
      "learning_rate_embeddings": 0.0004622222222222222,
      "loss": 3.444,
      "step": 42000
    },
    {
      "epoch": 12.8,
      "learning_rate": 0.00046111111111111114,
      "learning_rate_embeddings": 0.00046111111111111114,
      "loss": 3.4601,
      "step": 42500
    },
    {
      "epoch": 12.95,
      "learning_rate": 0.00046,
      "learning_rate_embeddings": 0.00046,
      "loss": 3.4457,
      "step": 43000
    },
    {
      "epoch": 13.1,
      "learning_rate": 0.0004588888888888889,
      "learning_rate_embeddings": 0.0004588888888888889,
      "loss": 3.4304,
      "step": 43500
    },
    {
      "epoch": 13.25,
      "learning_rate": 0.0004577777777777778,
      "learning_rate_embeddings": 0.0004577777777777778,
      "loss": 3.4221,
      "step": 44000
    },
    {
      "epoch": 13.4,
      "learning_rate": 0.0004566666666666667,
      "learning_rate_embeddings": 0.0004566666666666667,
      "loss": 3.4291,
      "step": 44500
    },
    {
      "epoch": 13.55,
      "learning_rate": 0.00045555555555555556,
      "learning_rate_embeddings": 0.00045555555555555556,
      "loss": 3.4317,
      "step": 45000
    },
    {
      "epoch": 13.7,
      "learning_rate": 0.00045444444444444444,
      "learning_rate_embeddings": 0.00045444444444444444,
      "loss": 3.4365,
      "step": 45500
    },
    {
      "epoch": 13.85,
      "learning_rate": 0.0004533333333333333,
      "learning_rate_embeddings": 0.0004533333333333333,
      "loss": 3.4375,
      "step": 46000
    },
    {
      "epoch": 14.0,
      "learning_rate": 0.00045222222222222224,
      "learning_rate_embeddings": 0.00045222222222222224,
      "loss": 3.4398,
      "step": 46500
    },
    {
      "epoch": 14.15,
      "learning_rate": 0.0004511111111111111,
      "learning_rate_embeddings": 0.0004511111111111111,
      "loss": 3.4177,
      "step": 47000
    },
    {
      "epoch": 14.3,
      "learning_rate": 0.00045000000000000004,
      "learning_rate_embeddings": 0.00045000000000000004,
      "loss": 3.4112,
      "step": 47500
    },
    {
      "epoch": 14.45,
      "learning_rate": 0.0004488888888888889,
      "learning_rate_embeddings": 0.0004488888888888889,
      "loss": 3.4147,
      "step": 48000
    },
    {
      "epoch": 14.6,
      "learning_rate": 0.0004477777777777778,
      "learning_rate_embeddings": 0.0004477777777777778,
      "loss": 3.4309,
      "step": 48500
    },
    {
      "epoch": 14.75,
      "learning_rate": 0.00044666666666666666,
      "learning_rate_embeddings": 0.00044666666666666666,
      "loss": 3.4239,
      "step": 49000
    },
    {
      "epoch": 14.9,
      "learning_rate": 0.00044555555555555554,
      "learning_rate_embeddings": 0.00044555555555555554,
      "loss": 3.4223,
      "step": 49500
    },
    {
      "epoch": 15.06,
      "learning_rate": 0.0004444444444444444,
      "learning_rate_embeddings": 0.0004444444444444444,
      "loss": 3.4191,
      "step": 50000
    },
    {
      "epoch": 15.06,
      "eval_loss": 3.415663242340088,
      "eval_runtime": 339.5814,
      "eval_samples_per_second": 499.898,
      "eval_steps_per_second": 3.908,
      "step": 50000
    },
    {
      "epoch": 15.21,
      "learning_rate": 0.00044333333333333334,
      "learning_rate_embeddings": 0.00044333333333333334,
      "loss": 3.3965,
      "step": 50500
    },
    {
      "epoch": 15.36,
      "learning_rate": 0.00044222222222222227,
      "learning_rate_embeddings": 0.00044222222222222227,
      "loss": 3.4077,
      "step": 51000
    },
    {
      "epoch": 15.51,
      "learning_rate": 0.00044111111111111114,
      "learning_rate_embeddings": 0.00044111111111111114,
      "loss": 3.4123,
      "step": 51500
    },
    {
      "epoch": 15.66,
      "learning_rate": 0.00044,
      "learning_rate_embeddings": 0.00044,
      "loss": 3.412,
      "step": 52000
    },
    {
      "epoch": 15.81,
      "learning_rate": 0.0004388888888888889,
      "learning_rate_embeddings": 0.0004388888888888889,
      "loss": 3.4212,
      "step": 52500
    },
    {
      "epoch": 15.96,
      "learning_rate": 0.00043777777777777776,
      "learning_rate_embeddings": 0.00043777777777777776,
      "loss": 3.419,
      "step": 53000
    },
    {
      "epoch": 16.11,
      "learning_rate": 0.00043666666666666664,
      "learning_rate_embeddings": 0.00043666666666666664,
      "loss": 3.3985,
      "step": 53500
    },
    {
      "epoch": 16.26,
      "learning_rate": 0.0004355555555555555,
      "learning_rate_embeddings": 0.0004355555555555555,
      "loss": 3.3984,
      "step": 54000
    },
    {
      "epoch": 16.41,
      "learning_rate": 0.0004344444444444445,
      "learning_rate_embeddings": 0.0004344444444444445,
      "loss": 3.4066,
      "step": 54500
    },
    {
      "epoch": 16.56,
      "learning_rate": 0.00043333333333333337,
      "learning_rate_embeddings": 0.00043333333333333337,
      "loss": 3.4093,
      "step": 55000
    },
    {
      "epoch": 16.71,
      "learning_rate": 0.00043222222222222224,
      "learning_rate_embeddings": 0.00043222222222222224,
      "loss": 3.4045,
      "step": 55500
    },
    {
      "epoch": 16.86,
      "learning_rate": 0.0004311111111111111,
      "learning_rate_embeddings": 0.0004311111111111111,
      "loss": 3.4125,
      "step": 56000
    },
    {
      "epoch": 17.01,
      "learning_rate": 0.00043,
      "learning_rate_embeddings": 0.00043,
      "loss": 3.4052,
      "step": 56500
    },
    {
      "epoch": 17.16,
      "learning_rate": 0.00042888888888888886,
      "learning_rate_embeddings": 0.00042888888888888886,
      "loss": 3.3912,
      "step": 57000
    },
    {
      "epoch": 17.31,
      "learning_rate": 0.0004277777777777778,
      "learning_rate_embeddings": 0.0004277777777777778,
      "loss": 3.3955,
      "step": 57500
    },
    {
      "epoch": 17.46,
      "learning_rate": 0.0004266666666666667,
      "learning_rate_embeddings": 0.0004266666666666667,
      "loss": 3.3963,
      "step": 58000
    },
    {
      "epoch": 17.62,
      "learning_rate": 0.0004255555555555556,
      "learning_rate_embeddings": 0.0004255555555555556,
      "loss": 3.3943,
      "step": 58500
    },
    {
      "epoch": 17.77,
      "learning_rate": 0.00042444444444444447,
      "learning_rate_embeddings": 0.00042444444444444447,
      "loss": 3.4028,
      "step": 59000
    },
    {
      "epoch": 17.92,
      "learning_rate": 0.00042333333333333334,
      "learning_rate_embeddings": 0.00042333333333333334,
      "loss": 3.3974,
      "step": 59500
    },
    {
      "epoch": 18.07,
      "learning_rate": 0.0004222222222222222,
      "learning_rate_embeddings": 0.0004222222222222222,
      "loss": 3.394,
      "step": 60000
    },
    {
      "epoch": 18.22,
      "learning_rate": 0.0004211111111111111,
      "learning_rate_embeddings": 0.0004211111111111111,
      "loss": 3.3731,
      "step": 60500
    },
    {
      "epoch": 18.37,
      "learning_rate": 0.00042,
      "learning_rate_embeddings": 0.00042,
      "loss": 3.3879,
      "step": 61000
    },
    {
      "epoch": 18.52,
      "learning_rate": 0.0004188888888888889,
      "learning_rate_embeddings": 0.0004188888888888889,
      "loss": 3.4023,
      "step": 61500
    },
    {
      "epoch": 18.67,
      "learning_rate": 0.0004177777777777778,
      "learning_rate_embeddings": 0.0004177777777777778,
      "loss": 3.3873,
      "step": 62000
    },
    {
      "epoch": 18.82,
      "learning_rate": 0.0004166666666666667,
      "learning_rate_embeddings": 0.0004166666666666667,
      "loss": 3.3954,
      "step": 62500
    },
    {
      "epoch": 18.82,
      "eval_loss": 3.3931150436401367,
      "eval_runtime": 339.4313,
      "eval_samples_per_second": 500.119,
      "eval_steps_per_second": 3.909,
      "step": 62500
    },
    {
      "epoch": 18.97,
      "learning_rate": 0.00041555555555555557,
      "learning_rate_embeddings": 0.00041555555555555557,
      "loss": 3.394,
      "step": 63000
    },
    {
      "epoch": 19.12,
      "learning_rate": 0.00041444444444444444,
      "learning_rate_embeddings": 0.00041444444444444444,
      "loss": 3.3877,
      "step": 63500
    },
    {
      "epoch": 19.27,
      "learning_rate": 0.0004133333333333333,
      "learning_rate_embeddings": 0.0004133333333333333,
      "loss": 3.3839,
      "step": 64000
    },
    {
      "epoch": 19.42,
      "learning_rate": 0.00041222222222222224,
      "learning_rate_embeddings": 0.00041222222222222224,
      "loss": 3.3876,
      "step": 64500
    },
    {
      "epoch": 19.57,
      "learning_rate": 0.0004111111111111111,
      "learning_rate_embeddings": 0.0004111111111111111,
      "loss": 3.3876,
      "step": 65000
    },
    {
      "epoch": 19.72,
      "learning_rate": 0.00041,
      "learning_rate_embeddings": 0.00041,
      "loss": 3.392,
      "step": 65500
    },
    {
      "epoch": 19.87,
      "learning_rate": 0.0004088888888888889,
      "learning_rate_embeddings": 0.0004088888888888889,
      "loss": 3.383,
      "step": 66000
    },
    {
      "epoch": 20.02,
      "learning_rate": 0.0004077777777777778,
      "learning_rate_embeddings": 0.0004077777777777778,
      "loss": 3.3901,
      "step": 66500
    },
    {
      "epoch": 20.17,
      "learning_rate": 0.00040666666666666667,
      "learning_rate_embeddings": 0.00040666666666666667,
      "loss": 3.3707,
      "step": 67000
    },
    {
      "epoch": 20.33,
      "learning_rate": 0.00040555555555555554,
      "learning_rate_embeddings": 0.00040555555555555554,
      "loss": 3.3775,
      "step": 67500
    },
    {
      "epoch": 20.48,
      "learning_rate": 0.00040444444444444447,
      "learning_rate_embeddings": 0.00040444444444444447,
      "loss": 3.3807,
      "step": 68000
    },
    {
      "epoch": 20.63,
      "learning_rate": 0.00040333333333333334,
      "learning_rate_embeddings": 0.00040333333333333334,
      "loss": 3.3787,
      "step": 68500
    },
    {
      "epoch": 20.78,
      "learning_rate": 0.0004022222222222222,
      "learning_rate_embeddings": 0.0004022222222222222,
      "loss": 3.3778,
      "step": 69000
    },
    {
      "epoch": 20.93,
      "learning_rate": 0.0004011111111111111,
      "learning_rate_embeddings": 0.0004011111111111111,
      "loss": 3.3895,
      "step": 69500
    },
    {
      "epoch": 21.08,
      "learning_rate": 0.0004,
      "learning_rate_embeddings": 0.0004,
      "loss": 3.3752,
      "step": 70000
    },
    {
      "epoch": 21.23,
      "learning_rate": 0.0003988888888888889,
      "learning_rate_embeddings": 0.0003988888888888889,
      "loss": 3.3688,
      "step": 70500
    },
    {
      "epoch": 21.38,
      "learning_rate": 0.00039777777777777777,
      "learning_rate_embeddings": 0.00039777777777777777,
      "loss": 3.3726,
      "step": 71000
    },
    {
      "epoch": 21.53,
      "learning_rate": 0.0003966666666666667,
      "learning_rate_embeddings": 0.0003966666666666667,
      "loss": 3.3728,
      "step": 71500
    },
    {
      "epoch": 21.68,
      "learning_rate": 0.00039555555555555557,
      "learning_rate_embeddings": 0.00039555555555555557,
      "loss": 3.3753,
      "step": 72000
    },
    {
      "epoch": 21.83,
      "learning_rate": 0.00039444444444444444,
      "learning_rate_embeddings": 0.00039444444444444444,
      "loss": 3.3732,
      "step": 72500
    },
    {
      "epoch": 21.98,
      "learning_rate": 0.0003933333333333333,
      "learning_rate_embeddings": 0.0003933333333333333,
      "loss": 3.3757,
      "step": 73000
    },
    {
      "epoch": 22.13,
      "learning_rate": 0.00039222222222222225,
      "learning_rate_embeddings": 0.00039222222222222225,
      "loss": 3.3616,
      "step": 73500
    },
    {
      "epoch": 22.28,
      "learning_rate": 0.0003911111111111111,
      "learning_rate_embeddings": 0.0003911111111111111,
      "loss": 3.3688,
      "step": 74000
    },
    {
      "epoch": 22.43,
      "learning_rate": 0.00039000000000000005,
      "learning_rate_embeddings": 0.00039000000000000005,
      "loss": 3.375,
      "step": 74500
    },
    {
      "epoch": 22.58,
      "learning_rate": 0.0003888888888888889,
      "learning_rate_embeddings": 0.0003888888888888889,
      "loss": 3.3718,
      "step": 75000
    },
    {
      "epoch": 22.58,
      "eval_loss": 3.380474090576172,
      "eval_runtime": 339.4022,
      "eval_samples_per_second": 500.162,
      "eval_steps_per_second": 3.91,
      "step": 75000
    },
    {
      "epoch": 22.73,
      "learning_rate": 0.0003877777777777778,
      "learning_rate_embeddings": 0.0003877777777777778,
      "loss": 3.3739,
      "step": 75500
    },
    {
      "epoch": 22.88,
      "learning_rate": 0.00038666666666666667,
      "learning_rate_embeddings": 0.00038666666666666667,
      "loss": 3.3765,
      "step": 76000
    },
    {
      "epoch": 23.04,
      "learning_rate": 0.00038555555555555554,
      "learning_rate_embeddings": 0.00038555555555555554,
      "loss": 3.3752,
      "step": 76500
    },
    {
      "epoch": 23.19,
      "learning_rate": 0.0003844444444444444,
      "learning_rate_embeddings": 0.0003844444444444444,
      "loss": 3.353,
      "step": 77000
    },
    {
      "epoch": 23.34,
      "learning_rate": 0.00038333333333333334,
      "learning_rate_embeddings": 0.00038333333333333334,
      "loss": 3.3647,
      "step": 77500
    },
    {
      "epoch": 23.49,
      "learning_rate": 0.0003822222222222223,
      "learning_rate_embeddings": 0.0003822222222222223,
      "loss": 3.3578,
      "step": 78000
    },
    {
      "epoch": 23.64,
      "learning_rate": 0.00038111111111111115,
      "learning_rate_embeddings": 0.00038111111111111115,
      "loss": 3.3731,
      "step": 78500
    },
    {
      "epoch": 23.79,
      "learning_rate": 0.00038,
      "learning_rate_embeddings": 0.00038,
      "loss": 3.3645,
      "step": 79000
    },
    {
      "epoch": 23.94,
      "learning_rate": 0.0003788888888888889,
      "learning_rate_embeddings": 0.0003788888888888889,
      "loss": 3.3735,
      "step": 79500
    },
    {
      "epoch": 24.09,
      "learning_rate": 0.00037777777777777777,
      "learning_rate_embeddings": 0.00037777777777777777,
      "loss": 3.3648,
      "step": 80000
    },
    {
      "epoch": 24.24,
      "learning_rate": 0.00037666666666666664,
      "learning_rate_embeddings": 0.00037666666666666664,
      "loss": 3.3551,
      "step": 80500
    },
    {
      "epoch": 24.39,
      "learning_rate": 0.0003755555555555555,
      "learning_rate_embeddings": 0.0003755555555555555,
      "loss": 3.3605,
      "step": 81000
    },
    {
      "epoch": 24.54,
      "learning_rate": 0.0003744444444444445,
      "learning_rate_embeddings": 0.0003744444444444445,
      "loss": 3.3619,
      "step": 81500
    },
    {
      "epoch": 24.69,
      "learning_rate": 0.0003733333333333334,
      "learning_rate_embeddings": 0.0003733333333333334,
      "loss": 3.3657,
      "step": 82000
    },
    {
      "epoch": 24.84,
      "learning_rate": 0.00037222222222222225,
      "learning_rate_embeddings": 0.00037222222222222225,
      "loss": 3.3676,
      "step": 82500
    },
    {
      "epoch": 24.99,
      "learning_rate": 0.0003711111111111111,
      "learning_rate_embeddings": 0.0003711111111111111,
      "loss": 3.3665,
      "step": 83000
    },
    {
      "epoch": 25.14,
      "learning_rate": 0.00037,
      "learning_rate_embeddings": 0.00037,
      "loss": 3.3486,
      "step": 83500
    },
    {
      "epoch": 25.29,
      "learning_rate": 0.00036888888888888887,
      "learning_rate_embeddings": 0.00036888888888888887,
      "loss": 3.3545,
      "step": 84000
    },
    {
      "epoch": 25.44,
      "learning_rate": 0.00036777777777777774,
      "learning_rate_embeddings": 0.00036777777777777774,
      "loss": 3.3605,
      "step": 84500
    },
    {
      "epoch": 25.59,
      "learning_rate": 0.00036666666666666667,
      "learning_rate_embeddings": 0.00036666666666666667,
      "loss": 3.3613,
      "step": 85000
    },
    {
      "epoch": 25.75,
      "learning_rate": 0.0003655555555555556,
      "learning_rate_embeddings": 0.0003655555555555556,
      "loss": 3.3639,
      "step": 85500
    },
    {
      "epoch": 25.9,
      "learning_rate": 0.00036444444444444447,
      "learning_rate_embeddings": 0.00036444444444444447,
      "loss": 3.3596,
      "step": 86000
    },
    {
      "epoch": 26.05,
      "learning_rate": 0.00036333333333333335,
      "learning_rate_embeddings": 0.00036333333333333335,
      "loss": 3.3658,
      "step": 86500
    },
    {
      "epoch": 26.2,
      "learning_rate": 0.0003622222222222222,
      "learning_rate_embeddings": 0.0003622222222222222,
      "loss": 3.3485,
      "step": 87000
    },
    {
      "epoch": 26.35,
      "learning_rate": 0.0003611111111111111,
      "learning_rate_embeddings": 0.0003611111111111111,
      "loss": 3.3543,
      "step": 87500
    },
    {
      "epoch": 26.35,
      "eval_loss": 3.3701975345611572,
      "eval_runtime": 339.7498,
      "eval_samples_per_second": 499.65,
      "eval_steps_per_second": 3.906,
      "step": 87500
    },
    {
      "epoch": 26.5,
      "learning_rate": 0.00035999999999999997,
      "learning_rate_embeddings": 0.00035999999999999997,
      "loss": 3.3524,
      "step": 88000
    },
    {
      "epoch": 26.65,
      "learning_rate": 0.0003588888888888889,
      "learning_rate_embeddings": 0.0003588888888888889,
      "loss": 3.3594,
      "step": 88500
    },
    {
      "epoch": 26.8,
      "learning_rate": 0.00035777777777777777,
      "learning_rate_embeddings": 0.00035777777777777777,
      "loss": 3.3552,
      "step": 89000
    },
    {
      "epoch": 26.95,
      "learning_rate": 0.0003566666666666667,
      "learning_rate_embeddings": 0.0003566666666666667,
      "loss": 3.3616,
      "step": 89500
    },
    {
      "epoch": 27.1,
      "learning_rate": 0.00035555555555555557,
      "learning_rate_embeddings": 0.00035555555555555557,
      "loss": 3.3461,
      "step": 90000
    },
    {
      "epoch": 27.25,
      "learning_rate": 0.00035444444444444445,
      "learning_rate_embeddings": 0.00035444444444444445,
      "loss": 3.3368,
      "step": 90500
    },
    {
      "epoch": 27.4,
      "learning_rate": 0.0003533333333333333,
      "learning_rate_embeddings": 0.0003533333333333333,
      "loss": 3.3526,
      "step": 91000
    },
    {
      "epoch": 27.55,
      "learning_rate": 0.00035222222222222225,
      "learning_rate_embeddings": 0.00035222222222222225,
      "loss": 3.3525,
      "step": 91500
    },
    {
      "epoch": 27.7,
      "learning_rate": 0.0003511111111111111,
      "learning_rate_embeddings": 0.0003511111111111111,
      "loss": 3.3537,
      "step": 92000
    },
    {
      "epoch": 27.85,
      "learning_rate": 0.00035,
      "learning_rate_embeddings": 0.00035,
      "loss": 3.358,
      "step": 92500
    },
    {
      "epoch": 28.0,
      "learning_rate": 0.0003488888888888889,
      "learning_rate_embeddings": 0.0003488888888888889,
      "loss": 3.3605,
      "step": 93000
    },
    {
      "epoch": 28.15,
      "learning_rate": 0.0003477777777777778,
      "learning_rate_embeddings": 0.0003477777777777778,
      "loss": 3.3352,
      "step": 93500
    },
    {
      "epoch": 28.3,
      "learning_rate": 0.00034666666666666667,
      "learning_rate_embeddings": 0.00034666666666666667,
      "loss": 3.3506,
      "step": 94000
    },
    {
      "epoch": 28.46,
      "learning_rate": 0.00034555555555555555,
      "learning_rate_embeddings": 0.00034555555555555555,
      "loss": 3.3492,
      "step": 94500
    },
    {
      "epoch": 28.61,
      "learning_rate": 0.0003444444444444445,
      "learning_rate_embeddings": 0.0003444444444444445,
      "loss": 3.346,
      "step": 95000
    },
    {
      "epoch": 28.76,
      "learning_rate": 0.00034333333333333335,
      "learning_rate_embeddings": 0.00034333333333333335,
      "loss": 3.3543,
      "step": 95500
    },
    {
      "epoch": 28.91,
      "learning_rate": 0.0003422222222222222,
      "learning_rate_embeddings": 0.0003422222222222222,
      "loss": 3.3494,
      "step": 96000
    },
    {
      "epoch": 29.06,
      "learning_rate": 0.0003411111111111111,
      "learning_rate_embeddings": 0.0003411111111111111,
      "loss": 3.358,
      "step": 96500
    },
    {
      "epoch": 29.21,
      "learning_rate": 0.00034,
      "learning_rate_embeddings": 0.00034,
      "loss": 3.3407,
      "step": 97000
    },
    {
      "epoch": 29.36,
      "learning_rate": 0.0003388888888888889,
      "learning_rate_embeddings": 0.0003388888888888889,
      "loss": 3.3448,
      "step": 97500
    },
    {
      "epoch": 29.51,
      "learning_rate": 0.00033777777777777777,
      "learning_rate_embeddings": 0.00033777777777777777,
      "loss": 3.3537,
      "step": 98000
    },
    {
      "epoch": 29.66,
      "learning_rate": 0.0003366666666666667,
      "learning_rate_embeddings": 0.0003366666666666667,
      "loss": 3.3533,
      "step": 98500
    },
    {
      "epoch": 29.81,
      "learning_rate": 0.0003355555555555556,
      "learning_rate_embeddings": 0.0003355555555555556,
      "loss": 3.3539,
      "step": 99000
    },
    {
      "epoch": 29.96,
      "learning_rate": 0.00033444444444444445,
      "learning_rate_embeddings": 0.00033444444444444445,
      "loss": 3.3497,
      "step": 99500
    },
    {
      "epoch": 30.11,
      "learning_rate": 0.0003333333333333333,
      "learning_rate_embeddings": 0.0003333333333333333,
      "loss": 3.3353,
      "step": 100000
    },
    {
      "epoch": 30.11,
      "eval_loss": 3.360551357269287,
      "eval_runtime": 339.4078,
      "eval_samples_per_second": 500.153,
      "eval_steps_per_second": 3.91,
      "step": 100000
    },
    {
      "epoch": 30.26,
      "learning_rate": 0.0003322222222222222,
      "learning_rate_embeddings": 0.0003322222222222222,
      "loss": 3.3458,
      "step": 100500
    },
    {
      "epoch": 30.41,
      "learning_rate": 0.0003311111111111111,
      "learning_rate_embeddings": 0.0003311111111111111,
      "loss": 3.3401,
      "step": 101000
    },
    {
      "epoch": 30.56,
      "learning_rate": 0.00033,
      "learning_rate_embeddings": 0.00033,
      "loss": 3.3402,
      "step": 101500
    },
    {
      "epoch": 30.71,
      "learning_rate": 0.0003288888888888889,
      "learning_rate_embeddings": 0.0003288888888888889,
      "loss": 3.3472,
      "step": 102000
    },
    {
      "epoch": 30.86,
      "learning_rate": 0.0003277777777777778,
      "learning_rate_embeddings": 0.0003277777777777778,
      "loss": 3.346,
      "step": 102500
    },
    {
      "epoch": 31.01,
      "learning_rate": 0.0003266666666666667,
      "learning_rate_embeddings": 0.0003266666666666667,
      "loss": 3.3464,
      "step": 103000
    },
    {
      "epoch": 31.17,
      "learning_rate": 0.00032555555555555555,
      "learning_rate_embeddings": 0.00032555555555555555,
      "loss": 3.3269,
      "step": 103500
    },
    {
      "epoch": 31.32,
      "learning_rate": 0.0003244444444444444,
      "learning_rate_embeddings": 0.0003244444444444444,
      "loss": 3.3285,
      "step": 104000
    },
    {
      "epoch": 31.47,
      "learning_rate": 0.0003233333333333333,
      "learning_rate_embeddings": 0.0003233333333333333,
      "loss": 3.3384,
      "step": 104500
    },
    {
      "epoch": 31.62,
      "learning_rate": 0.0003222222222222222,
      "learning_rate_embeddings": 0.0003222222222222222,
      "loss": 3.336,
      "step": 105000
    },
    {
      "epoch": 31.77,
      "learning_rate": 0.00032111111111111115,
      "learning_rate_embeddings": 0.00032111111111111115,
      "loss": 3.3437,
      "step": 105500
    },
    {
      "epoch": 31.92,
      "learning_rate": 0.00032,
      "learning_rate_embeddings": 0.00032,
      "loss": 3.3459,
      "step": 106000
    },
    {
      "epoch": 32.07,
      "learning_rate": 0.0003188888888888889,
      "learning_rate_embeddings": 0.0003188888888888889,
      "loss": 3.3486,
      "step": 106500
    },
    {
      "epoch": 32.22,
      "learning_rate": 0.0003177777777777778,
      "learning_rate_embeddings": 0.0003177777777777778,
      "loss": 3.3274,
      "step": 107000
    },
    {
      "epoch": 32.37,
      "learning_rate": 0.00031666666666666665,
      "learning_rate_embeddings": 0.00031666666666666665,
      "loss": 3.3359,
      "step": 107500
    },
    {
      "epoch": 32.52,
      "learning_rate": 0.0003155555555555555,
      "learning_rate_embeddings": 0.0003155555555555555,
      "loss": 3.3383,
      "step": 108000
    },
    {
      "epoch": 32.67,
      "learning_rate": 0.0003144444444444445,
      "learning_rate_embeddings": 0.0003144444444444445,
      "loss": 3.3394,
      "step": 108500
    },
    {
      "epoch": 32.82,
      "learning_rate": 0.0003133333333333334,
      "learning_rate_embeddings": 0.0003133333333333334,
      "loss": 3.3386,
      "step": 109000
    },
    {
      "epoch": 32.97,
      "learning_rate": 0.00031222222222222225,
      "learning_rate_embeddings": 0.00031222222222222225,
      "loss": 3.3499,
      "step": 109500
    },
    {
      "epoch": 33.12,
      "learning_rate": 0.0003111111111111111,
      "learning_rate_embeddings": 0.0003111111111111111,
      "loss": 3.3333,
      "step": 110000
    },
    {
      "epoch": 33.27,
      "learning_rate": 0.00031,
      "learning_rate_embeddings": 0.00031,
      "loss": 3.3281,
      "step": 110500
    },
    {
      "epoch": 33.42,
      "learning_rate": 0.0003088888888888889,
      "learning_rate_embeddings": 0.0003088888888888889,
      "loss": 3.3347,
      "step": 111000
    },
    {
      "epoch": 33.57,
      "learning_rate": 0.00030777777777777775,
      "learning_rate_embeddings": 0.00030777777777777775,
      "loss": 3.3386,
      "step": 111500
    },
    {
      "epoch": 33.72,
      "learning_rate": 0.0003066666666666667,
      "learning_rate_embeddings": 0.0003066666666666667,
      "loss": 3.331,
      "step": 112000
    },
    {
      "epoch": 33.88,
      "learning_rate": 0.0003055555555555556,
      "learning_rate_embeddings": 0.0003055555555555556,
      "loss": 3.3403,
      "step": 112500
    },
    {
      "epoch": 33.88,
      "eval_loss": 3.3534741401672363,
      "eval_runtime": 338.9476,
      "eval_samples_per_second": 500.833,
      "eval_steps_per_second": 3.915,
      "step": 112500
    },
    {
      "epoch": 34.03,
      "learning_rate": 0.0003044444444444445,
      "learning_rate_embeddings": 0.0003044444444444445,
      "loss": 3.3425,
      "step": 113000
    },
    {
      "epoch": 34.18,
      "learning_rate": 0.00030333333333333335,
      "learning_rate_embeddings": 0.00030333333333333335,
      "loss": 3.3207,
      "step": 113500
    },
    {
      "epoch": 34.33,
      "learning_rate": 0.0003022222222222222,
      "learning_rate_embeddings": 0.0003022222222222222,
      "loss": 3.3287,
      "step": 114000
    },
    {
      "epoch": 34.48,
      "learning_rate": 0.0003011111111111111,
      "learning_rate_embeddings": 0.0003011111111111111,
      "loss": 3.3288,
      "step": 114500
    },
    {
      "epoch": 34.63,
      "learning_rate": 0.0003,
      "learning_rate_embeddings": 0.0003,
      "loss": 3.3309,
      "step": 115000
    },
    {
      "epoch": 34.78,
      "learning_rate": 0.0002988888888888889,
      "learning_rate_embeddings": 0.0002988888888888889,
      "loss": 3.3387,
      "step": 115500
    },
    {
      "epoch": 34.93,
      "learning_rate": 0.0002977777777777778,
      "learning_rate_embeddings": 0.0002977777777777778,
      "loss": 3.3403,
      "step": 116000
    },
    {
      "epoch": 35.08,
      "learning_rate": 0.0002966666666666667,
      "learning_rate_embeddings": 0.0002966666666666667,
      "loss": 3.329,
      "step": 116500
    },
    {
      "epoch": 35.23,
      "learning_rate": 0.0002955555555555556,
      "learning_rate_embeddings": 0.0002955555555555556,
      "loss": 3.3297,
      "step": 117000
    },
    {
      "epoch": 35.38,
      "learning_rate": 0.00029444444444444445,
      "learning_rate_embeddings": 0.00029444444444444445,
      "loss": 3.3262,
      "step": 117500
    },
    {
      "epoch": 35.53,
      "learning_rate": 0.0002933333333333333,
      "learning_rate_embeddings": 0.0002933333333333333,
      "loss": 3.325,
      "step": 118000
    },
    {
      "epoch": 35.68,
      "learning_rate": 0.0002922222222222222,
      "learning_rate_embeddings": 0.0002922222222222222,
      "loss": 3.3286,
      "step": 118500
    },
    {
      "epoch": 35.83,
      "learning_rate": 0.00029111111111111113,
      "learning_rate_embeddings": 0.00029111111111111113,
      "loss": 3.3337,
      "step": 119000
    },
    {
      "epoch": 35.98,
      "learning_rate": 0.00029,
      "learning_rate_embeddings": 0.00029,
      "loss": 3.334,
      "step": 119500
    },
    {
      "epoch": 36.13,
      "learning_rate": 0.0002888888888888889,
      "learning_rate_embeddings": 0.0002888888888888889,
      "loss": 3.322,
      "step": 120000
    },
    {
      "epoch": 36.28,
      "learning_rate": 0.0002877777777777778,
      "learning_rate_embeddings": 0.0002877777777777778,
      "loss": 3.3269,
      "step": 120500
    },
    {
      "epoch": 36.43,
      "learning_rate": 0.0002866666666666667,
      "learning_rate_embeddings": 0.0002866666666666667,
      "loss": 3.321,
      "step": 121000
    },
    {
      "epoch": 36.59,
      "learning_rate": 0.00028555555555555555,
      "learning_rate_embeddings": 0.00028555555555555555,
      "loss": 3.3275,
      "step": 121500
    },
    {
      "epoch": 36.74,
      "learning_rate": 0.0002844444444444444,
      "learning_rate_embeddings": 0.0002844444444444444,
      "loss": 3.334,
      "step": 122000
    },
    {
      "epoch": 36.89,
      "learning_rate": 0.00028333333333333335,
      "learning_rate_embeddings": 0.00028333333333333335,
      "loss": 3.3259,
      "step": 122500
    },
    {
      "epoch": 37.04,
      "learning_rate": 0.00028222222222222223,
      "learning_rate_embeddings": 0.00028222222222222223,
      "loss": 3.3373,
      "step": 123000
    },
    {
      "epoch": 37.19,
      "learning_rate": 0.0002811111111111111,
      "learning_rate_embeddings": 0.0002811111111111111,
      "loss": 3.3228,
      "step": 123500
    },
    {
      "epoch": 37.34,
      "learning_rate": 0.00028000000000000003,
      "learning_rate_embeddings": 0.00028000000000000003,
      "loss": 3.3244,
      "step": 124000
    },
    {
      "epoch": 37.49,
      "learning_rate": 0.0002788888888888889,
      "learning_rate_embeddings": 0.0002788888888888889,
      "loss": 3.3212,
      "step": 124500
    },
    {
      "epoch": 37.64,
      "learning_rate": 0.0002777777777777778,
      "learning_rate_embeddings": 0.0002777777777777778,
      "loss": 3.3288,
      "step": 125000
    },
    {
      "epoch": 37.64,
      "eval_loss": 3.348264455795288,
      "eval_runtime": 339.1916,
      "eval_samples_per_second": 500.472,
      "eval_steps_per_second": 3.912,
      "step": 125000
    },
    {
      "epoch": 37.79,
      "learning_rate": 0.00027666666666666665,
      "learning_rate_embeddings": 0.00027666666666666665,
      "loss": 3.3295,
      "step": 125500
    },
    {
      "epoch": 37.94,
      "learning_rate": 0.0002755555555555556,
      "learning_rate_embeddings": 0.0002755555555555556,
      "loss": 3.3343,
      "step": 126000
    },
    {
      "epoch": 38.09,
      "learning_rate": 0.00027444444444444445,
      "learning_rate_embeddings": 0.00027444444444444445,
      "loss": 3.3282,
      "step": 126500
    },
    {
      "epoch": 38.24,
      "learning_rate": 0.00027333333333333333,
      "learning_rate_embeddings": 0.00027333333333333333,
      "loss": 3.3191,
      "step": 127000
    },
    {
      "epoch": 38.39,
      "learning_rate": 0.0002722222222222222,
      "learning_rate_embeddings": 0.0002722222222222222,
      "loss": 3.3225,
      "step": 127500
    },
    {
      "epoch": 38.54,
      "learning_rate": 0.00027111111111111113,
      "learning_rate_embeddings": 0.00027111111111111113,
      "loss": 3.3233,
      "step": 128000
    },
    {
      "epoch": 38.69,
      "learning_rate": 0.00027,
      "learning_rate_embeddings": 0.00027,
      "loss": 3.3248,
      "step": 128500
    },
    {
      "epoch": 38.84,
      "learning_rate": 0.00026888888888888893,
      "learning_rate_embeddings": 0.00026888888888888893,
      "loss": 3.3238,
      "step": 129000
    },
    {
      "epoch": 38.99,
      "learning_rate": 0.0002677777777777778,
      "learning_rate_embeddings": 0.0002677777777777778,
      "loss": 3.3335,
      "step": 129500
    },
    {
      "epoch": 39.14,
      "learning_rate": 0.0002666666666666667,
      "learning_rate_embeddings": 0.0002666666666666667,
      "loss": 3.3204,
      "step": 130000
    },
    {
      "epoch": 39.3,
      "learning_rate": 0.00026555555555555555,
      "learning_rate_embeddings": 0.00026555555555555555,
      "loss": 3.3155,
      "step": 130500
    },
    {
      "epoch": 39.45,
      "learning_rate": 0.00026444444444444443,
      "learning_rate_embeddings": 0.00026444444444444443,
      "loss": 3.3218,
      "step": 131000
    },
    {
      "epoch": 39.6,
      "learning_rate": 0.0002633333333333333,
      "learning_rate_embeddings": 0.0002633333333333333,
      "loss": 3.3211,
      "step": 131500
    },
    {
      "epoch": 39.75,
      "learning_rate": 0.00026222222222222223,
      "learning_rate_embeddings": 0.00026222222222222223,
      "loss": 3.3292,
      "step": 132000
    },
    {
      "epoch": 39.9,
      "learning_rate": 0.00026111111111111116,
      "learning_rate_embeddings": 0.00026111111111111116,
      "loss": 3.3256,
      "step": 132500
    },
    {
      "epoch": 40.05,
      "learning_rate": 0.00026000000000000003,
      "learning_rate_embeddings": 0.00026000000000000003,
      "loss": 3.3204,
      "step": 133000
    },
    {
      "epoch": 40.2,
      "learning_rate": 0.0002588888888888889,
      "learning_rate_embeddings": 0.0002588888888888889,
      "loss": 3.3087,
      "step": 133500
    },
    {
      "epoch": 40.35,
      "learning_rate": 0.0002577777777777778,
      "learning_rate_embeddings": 0.0002577777777777778,
      "loss": 3.3138,
      "step": 134000
    },
    {
      "epoch": 40.5,
      "learning_rate": 0.00025666666666666665,
      "learning_rate_embeddings": 0.00025666666666666665,
      "loss": 3.3227,
      "step": 134500
    },
    {
      "epoch": 40.65,
      "learning_rate": 0.00025555555555555553,
      "learning_rate_embeddings": 0.00025555555555555553,
      "loss": 3.3172,
      "step": 135000
    },
    {
      "epoch": 40.8,
      "learning_rate": 0.0002544444444444444,
      "learning_rate_embeddings": 0.0002544444444444444,
      "loss": 3.3333,
      "step": 135500
    },
    {
      "epoch": 40.95,
      "learning_rate": 0.0002533333333333334,
      "learning_rate_embeddings": 0.0002533333333333334,
      "loss": 3.3318,
      "step": 136000
    },
    {
      "epoch": 41.1,
      "learning_rate": 0.00025222222222222226,
      "learning_rate_embeddings": 0.00025222222222222226,
      "loss": 3.3243,
      "step": 136500
    },
    {
      "epoch": 41.25,
      "learning_rate": 0.00025111111111111113,
      "learning_rate_embeddings": 0.00025111111111111113,
      "loss": 3.3072,
      "step": 137000
    },
    {
      "epoch": 41.4,
      "learning_rate": 0.00025,
      "learning_rate_embeddings": 0.00025,
      "loss": 3.3202,
      "step": 137500
    },
    {
      "epoch": 41.4,
      "eval_loss": 3.3424789905548096,
      "eval_runtime": 339.2553,
      "eval_samples_per_second": 500.378,
      "eval_steps_per_second": 3.912,
      "step": 137500
    },
    {
      "epoch": 41.55,
      "learning_rate": 0.0002488888888888889,
      "learning_rate_embeddings": 0.0002488888888888889,
      "loss": 3.3259,
      "step": 138000
    },
    {
      "epoch": 41.7,
      "learning_rate": 0.0002477777777777778,
      "learning_rate_embeddings": 0.0002477777777777778,
      "loss": 3.3193,
      "step": 138500
    },
    {
      "epoch": 41.85,
      "learning_rate": 0.0002466666666666667,
      "learning_rate_embeddings": 0.0002466666666666667,
      "loss": 3.3199,
      "step": 139000
    },
    {
      "epoch": 42.01,
      "learning_rate": 0.00024555555555555556,
      "learning_rate_embeddings": 0.00024555555555555556,
      "loss": 3.3275,
      "step": 139500
    },
    {
      "epoch": 42.16,
      "learning_rate": 0.00024444444444444443,
      "learning_rate_embeddings": 0.00024444444444444443,
      "loss": 3.3066,
      "step": 140000
    },
    {
      "epoch": 42.31,
      "learning_rate": 0.00024333333333333336,
      "learning_rate_embeddings": 0.00024333333333333336,
      "loss": 3.3145,
      "step": 140500
    },
    {
      "epoch": 42.46,
      "learning_rate": 0.00024222222222222223,
      "learning_rate_embeddings": 0.00024222222222222223,
      "loss": 3.3146,
      "step": 141000
    },
    {
      "epoch": 42.61,
      "learning_rate": 0.0002411111111111111,
      "learning_rate_embeddings": 0.0002411111111111111,
      "loss": 3.3169,
      "step": 141500
    },
    {
      "epoch": 42.76,
      "learning_rate": 0.00024,
      "learning_rate_embeddings": 0.00024,
      "loss": 3.3166,
      "step": 142000
    },
    {
      "epoch": 42.91,
      "learning_rate": 0.0002388888888888889,
      "learning_rate_embeddings": 0.0002388888888888889,
      "loss": 3.3196,
      "step": 142500
    },
    {
      "epoch": 43.06,
      "learning_rate": 0.00023777777777777778,
      "learning_rate_embeddings": 0.00023777777777777778,
      "loss": 3.3171,
      "step": 143000
    },
    {
      "epoch": 43.21,
      "learning_rate": 0.00023666666666666668,
      "learning_rate_embeddings": 0.00023666666666666668,
      "loss": 3.3046,
      "step": 143500
    },
    {
      "epoch": 43.36,
      "learning_rate": 0.00023555555555555556,
      "learning_rate_embeddings": 0.00023555555555555556,
      "loss": 3.3128,
      "step": 144000
    },
    {
      "epoch": 43.51,
      "learning_rate": 0.00023444444444444446,
      "learning_rate_embeddings": 0.00023444444444444446,
      "loss": 3.3092,
      "step": 144500
    },
    {
      "epoch": 43.66,
      "learning_rate": 0.00023333333333333333,
      "learning_rate_embeddings": 0.00023333333333333333,
      "loss": 3.3234,
      "step": 145000
    },
    {
      "epoch": 43.81,
      "learning_rate": 0.00023222222222222223,
      "learning_rate_embeddings": 0.00023222222222222223,
      "loss": 3.3222,
      "step": 145500
    },
    {
      "epoch": 43.96,
      "learning_rate": 0.0002311111111111111,
      "learning_rate_embeddings": 0.0002311111111111111,
      "loss": 3.3211,
      "step": 146000
    },
    {
      "epoch": 44.11,
      "learning_rate": 0.00023,
      "learning_rate_embeddings": 0.00023,
      "loss": 3.3126,
      "step": 146500
    },
    {
      "epoch": 44.26,
      "learning_rate": 0.0002288888888888889,
      "learning_rate_embeddings": 0.0002288888888888889,
      "loss": 3.3047,
      "step": 147000
    },
    {
      "epoch": 44.41,
      "learning_rate": 0.00022777777777777778,
      "learning_rate_embeddings": 0.00022777777777777778,
      "loss": 3.3081,
      "step": 147500
    },
    {
      "epoch": 44.56,
      "learning_rate": 0.00022666666666666666,
      "learning_rate_embeddings": 0.00022666666666666666,
      "loss": 3.3218,
      "step": 148000
    },
    {
      "epoch": 44.72,
      "learning_rate": 0.00022555555555555556,
      "learning_rate_embeddings": 0.00022555555555555556,
      "loss": 3.3085,
      "step": 148500
    },
    {
      "epoch": 44.87,
      "learning_rate": 0.00022444444444444446,
      "learning_rate_embeddings": 0.00022444444444444446,
      "loss": 3.319,
      "step": 149000
    },
    {
      "epoch": 45.02,
      "learning_rate": 0.00022333333333333333,
      "learning_rate_embeddings": 0.00022333333333333333,
      "loss": 3.3178,
      "step": 149500
    },
    {
      "epoch": 45.17,
      "learning_rate": 0.0002222222222222222,
      "learning_rate_embeddings": 0.0002222222222222222,
      "loss": 3.3086,
      "step": 150000
    },
    {
      "epoch": 45.17,
      "eval_loss": 3.337905168533325,
      "eval_runtime": 339.221,
      "eval_samples_per_second": 500.429,
      "eval_steps_per_second": 3.912,
      "step": 150000
    },
    {
      "epoch": 45.32,
      "learning_rate": 0.00022111111111111113,
      "learning_rate_embeddings": 0.00022111111111111113,
      "loss": 3.302,
      "step": 150500
    },
    {
      "epoch": 45.47,
      "learning_rate": 0.00022,
      "learning_rate_embeddings": 0.00022,
      "loss": 3.3006,
      "step": 151000
    },
    {
      "epoch": 45.62,
      "learning_rate": 0.00021888888888888888,
      "learning_rate_embeddings": 0.00021888888888888888,
      "loss": 3.313,
      "step": 151500
    },
    {
      "epoch": 45.77,
      "learning_rate": 0.00021777777777777776,
      "learning_rate_embeddings": 0.00021777777777777776,
      "loss": 3.3155,
      "step": 152000
    },
    {
      "epoch": 45.92,
      "learning_rate": 0.00021666666666666668,
      "learning_rate_embeddings": 0.00021666666666666668,
      "loss": 3.309,
      "step": 152500
    },
    {
      "epoch": 46.07,
      "learning_rate": 0.00021555555555555556,
      "learning_rate_embeddings": 0.00021555555555555556,
      "loss": 3.3075,
      "step": 153000
    },
    {
      "epoch": 46.22,
      "learning_rate": 0.00021444444444444443,
      "learning_rate_embeddings": 0.00021444444444444443,
      "loss": 3.3064,
      "step": 153500
    },
    {
      "epoch": 46.37,
      "learning_rate": 0.00021333333333333336,
      "learning_rate_embeddings": 0.00021333333333333336,
      "loss": 3.3059,
      "step": 154000
    },
    {
      "epoch": 46.52,
      "learning_rate": 0.00021222222222222223,
      "learning_rate_embeddings": 0.00021222222222222223,
      "loss": 3.3095,
      "step": 154500
    },
    {
      "epoch": 46.67,
      "learning_rate": 0.0002111111111111111,
      "learning_rate_embeddings": 0.0002111111111111111,
      "loss": 3.317,
      "step": 155000
    },
    {
      "epoch": 46.82,
      "learning_rate": 0.00021,
      "learning_rate_embeddings": 0.00021,
      "loss": 3.3098,
      "step": 155500
    },
    {
      "epoch": 46.97,
      "learning_rate": 0.0002088888888888889,
      "learning_rate_embeddings": 0.0002088888888888889,
      "loss": 3.3066,
      "step": 156000
    },
    {
      "epoch": 47.12,
      "learning_rate": 0.00020777777777777778,
      "learning_rate_embeddings": 0.00020777777777777778,
      "loss": 3.3093,
      "step": 156500
    },
    {
      "epoch": 47.27,
      "learning_rate": 0.00020666666666666666,
      "learning_rate_embeddings": 0.00020666666666666666,
      "loss": 3.3059,
      "step": 157000
    },
    {
      "epoch": 47.43,
      "learning_rate": 0.00020555555555555556,
      "learning_rate_embeddings": 0.00020555555555555556,
      "loss": 3.3065,
      "step": 157500
    },
    {
      "epoch": 47.58,
      "learning_rate": 0.00020444444444444446,
      "learning_rate_embeddings": 0.00020444444444444446,
      "loss": 3.3094,
      "step": 158000
    },
    {
      "epoch": 47.73,
      "learning_rate": 0.00020333333333333333,
      "learning_rate_embeddings": 0.00020333333333333333,
      "loss": 3.3066,
      "step": 158500
    },
    {
      "epoch": 47.88,
      "learning_rate": 0.00020222222222222223,
      "learning_rate_embeddings": 0.00020222222222222223,
      "loss": 3.3067,
      "step": 159000
    },
    {
      "epoch": 48.03,
      "learning_rate": 0.0002011111111111111,
      "learning_rate_embeddings": 0.0002011111111111111,
      "loss": 3.3138,
      "step": 159500
    },
    {
      "epoch": 48.18,
      "learning_rate": 0.0002,
      "learning_rate_embeddings": 0.0002,
      "loss": 3.3107,
      "step": 160000
    },
    {
      "epoch": 48.33,
      "learning_rate": 0.00019888888888888888,
      "learning_rate_embeddings": 0.00019888888888888888,
      "loss": 3.2933,
      "step": 160500
    },
    {
      "epoch": 48.48,
      "learning_rate": 0.00019777777777777778,
      "learning_rate_embeddings": 0.00019777777777777778,
      "loss": 3.3018,
      "step": 161000
    },
    {
      "epoch": 48.63,
      "learning_rate": 0.00019666666666666666,
      "learning_rate_embeddings": 0.00019666666666666666,
      "loss": 3.3067,
      "step": 161500
    },
    {
      "epoch": 48.78,
      "learning_rate": 0.00019555555555555556,
      "learning_rate_embeddings": 0.00019555555555555556,
      "loss": 3.3059,
      "step": 162000
    },
    {
      "epoch": 48.93,
      "learning_rate": 0.00019444444444444446,
      "learning_rate_embeddings": 0.00019444444444444446,
      "loss": 3.3023,
      "step": 162500
    },
    {
      "epoch": 48.93,
      "eval_loss": 3.330859899520874,
      "eval_runtime": 339.2751,
      "eval_samples_per_second": 500.349,
      "eval_steps_per_second": 3.911,
      "step": 162500
    },
    {
      "epoch": 49.08,
      "learning_rate": 0.00019333333333333333,
      "learning_rate_embeddings": 0.00019333333333333333,
      "loss": 3.305,
      "step": 163000
    },
    {
      "epoch": 49.23,
      "learning_rate": 0.0001922222222222222,
      "learning_rate_embeddings": 0.0001922222222222222,
      "loss": 3.2983,
      "step": 163500
    },
    {
      "epoch": 49.38,
      "learning_rate": 0.00019111111111111114,
      "learning_rate_embeddings": 0.00019111111111111114,
      "loss": 3.2919,
      "step": 164000
    },
    {
      "epoch": 49.53,
      "learning_rate": 0.00019,
      "learning_rate_embeddings": 0.00019,
      "loss": 3.3056,
      "step": 164500
    },
    {
      "epoch": 49.68,
      "learning_rate": 0.00018888888888888888,
      "learning_rate_embeddings": 0.00018888888888888888,
      "loss": 3.3077,
      "step": 165000
    },
    {
      "epoch": 49.83,
      "learning_rate": 0.00018777777777777776,
      "learning_rate_embeddings": 0.00018777777777777776,
      "loss": 3.3122,
      "step": 165500
    },
    {
      "epoch": 49.98,
      "learning_rate": 0.0001866666666666667,
      "learning_rate_embeddings": 0.0001866666666666667,
      "loss": 3.3143,
      "step": 166000
    },
    {
      "epoch": 50.14,
      "learning_rate": 0.00018555555555555556,
      "learning_rate_embeddings": 0.00018555555555555556,
      "loss": 3.3005,
      "step": 166500
    },
    {
      "epoch": 50.29,
      "learning_rate": 0.00018444444444444443,
      "learning_rate_embeddings": 0.00018444444444444443,
      "loss": 3.303,
      "step": 167000
    },
    {
      "epoch": 50.44,
      "learning_rate": 0.00018333333333333334,
      "learning_rate_embeddings": 0.00018333333333333334,
      "loss": 3.3001,
      "step": 167500
    },
    {
      "epoch": 50.59,
      "learning_rate": 0.00018222222222222224,
      "learning_rate_embeddings": 0.00018222222222222224,
      "loss": 3.2962,
      "step": 168000
    },
    {
      "epoch": 50.74,
      "learning_rate": 0.0001811111111111111,
      "learning_rate_embeddings": 0.0001811111111111111,
      "loss": 3.2952,
      "step": 168500
    },
    {
      "epoch": 50.89,
      "learning_rate": 0.00017999999999999998,
      "learning_rate_embeddings": 0.00017999999999999998,
      "loss": 3.307,
      "step": 169000
    },
    {
      "epoch": 51.04,
      "learning_rate": 0.00017888888888888889,
      "learning_rate_embeddings": 0.00017888888888888889,
      "loss": 3.3073,
      "step": 169500
    },
    {
      "epoch": 51.19,
      "learning_rate": 0.00017777777777777779,
      "learning_rate_embeddings": 0.00017777777777777779,
      "loss": 3.2904,
      "step": 170000
    },
    {
      "epoch": 51.34,
      "learning_rate": 0.00017666666666666666,
      "learning_rate_embeddings": 0.00017666666666666666,
      "loss": 3.298,
      "step": 170500
    },
    {
      "epoch": 51.49,
      "learning_rate": 0.00017555555555555556,
      "learning_rate_embeddings": 0.00017555555555555556,
      "loss": 3.3043,
      "step": 171000
    },
    {
      "epoch": 51.64,
      "learning_rate": 0.00017444444444444446,
      "learning_rate_embeddings": 0.00017444444444444446,
      "loss": 3.3007,
      "step": 171500
    },
    {
      "epoch": 51.79,
      "learning_rate": 0.00017333333333333334,
      "learning_rate_embeddings": 0.00017333333333333334,
      "loss": 3.2965,
      "step": 172000
    },
    {
      "epoch": 51.94,
      "learning_rate": 0.00017222222222222224,
      "learning_rate_embeddings": 0.00017222222222222224,
      "loss": 3.2953,
      "step": 172500
    },
    {
      "epoch": 52.09,
      "learning_rate": 0.0001711111111111111,
      "learning_rate_embeddings": 0.0001711111111111111,
      "loss": 3.2967,
      "step": 173000
    },
    {
      "epoch": 52.24,
      "learning_rate": 0.00017,
      "learning_rate_embeddings": 0.00017,
      "loss": 3.2872,
      "step": 173500
    },
    {
      "epoch": 52.39,
      "learning_rate": 0.00016888888888888889,
      "learning_rate_embeddings": 0.00016888888888888889,
      "loss": 3.2876,
      "step": 174000
    },
    {
      "epoch": 52.54,
      "learning_rate": 0.0001677777777777778,
      "learning_rate_embeddings": 0.0001677777777777778,
      "loss": 3.301,
      "step": 174500
    },
    {
      "epoch": 52.69,
      "learning_rate": 0.00016666666666666666,
      "learning_rate_embeddings": 0.00016666666666666666,
      "loss": 3.2948,
      "step": 175000
    },
    {
      "epoch": 52.69,
      "eval_loss": 3.329881191253662,
      "eval_runtime": 339.2854,
      "eval_samples_per_second": 500.334,
      "eval_steps_per_second": 3.911,
      "step": 175000
    },
    {
      "epoch": 52.85,
      "learning_rate": 0.00016555555555555556,
      "learning_rate_embeddings": 0.00016555555555555556,
      "loss": 3.3063,
      "step": 175500
    },
    {
      "epoch": 53.0,
      "learning_rate": 0.00016444444444444446,
      "learning_rate_embeddings": 0.00016444444444444446,
      "loss": 3.3003,
      "step": 176000
    },
    {
      "epoch": 53.15,
      "learning_rate": 0.00016333333333333334,
      "learning_rate_embeddings": 0.00016333333333333334,
      "loss": 3.2956,
      "step": 176500
    },
    {
      "epoch": 53.3,
      "learning_rate": 0.0001622222222222222,
      "learning_rate_embeddings": 0.0001622222222222222,
      "loss": 3.2917,
      "step": 177000
    },
    {
      "epoch": 53.45,
      "learning_rate": 0.0001611111111111111,
      "learning_rate_embeddings": 0.0001611111111111111,
      "loss": 3.2873,
      "step": 177500
    },
    {
      "epoch": 53.6,
      "learning_rate": 0.00016,
      "learning_rate_embeddings": 0.00016,
      "loss": 3.2944,
      "step": 178000
    },
    {
      "epoch": 53.75,
      "learning_rate": 0.0001588888888888889,
      "learning_rate_embeddings": 0.0001588888888888889,
      "loss": 3.2998,
      "step": 178500
    },
    {
      "epoch": 53.9,
      "learning_rate": 0.00015777777777777776,
      "learning_rate_embeddings": 0.00015777777777777776,
      "loss": 3.2969,
      "step": 179000
    },
    {
      "epoch": 54.05,
      "learning_rate": 0.0001566666666666667,
      "learning_rate_embeddings": 0.0001566666666666667,
      "loss": 3.2944,
      "step": 179500
    },
    {
      "epoch": 54.2,
      "learning_rate": 0.00015555555555555556,
      "learning_rate_embeddings": 0.00015555555555555556,
      "loss": 3.2885,
      "step": 180000
    },
    {
      "epoch": 54.35,
      "learning_rate": 0.00015444444444444444,
      "learning_rate_embeddings": 0.00015444444444444444,
      "loss": 3.2935,
      "step": 180500
    },
    {
      "epoch": 54.5,
      "learning_rate": 0.00015333333333333334,
      "learning_rate_embeddings": 0.00015333333333333334,
      "loss": 3.2907,
      "step": 181000
    },
    {
      "epoch": 54.65,
      "learning_rate": 0.00015222222222222224,
      "learning_rate_embeddings": 0.00015222222222222224,
      "loss": 3.2975,
      "step": 181500
    },
    {
      "epoch": 54.8,
      "learning_rate": 0.0001511111111111111,
      "learning_rate_embeddings": 0.0001511111111111111,
      "loss": 3.295,
      "step": 182000
    },
    {
      "epoch": 54.95,
      "learning_rate": 0.00015,
      "learning_rate_embeddings": 0.00015,
      "loss": 3.2902,
      "step": 182500
    },
    {
      "epoch": 55.1,
      "learning_rate": 0.0001488888888888889,
      "learning_rate_embeddings": 0.0001488888888888889,
      "loss": 3.2884,
      "step": 183000
    },
    {
      "epoch": 55.25,
      "learning_rate": 0.0001477777777777778,
      "learning_rate_embeddings": 0.0001477777777777778,
      "loss": 3.2876,
      "step": 183500
    },
    {
      "epoch": 55.4,
      "learning_rate": 0.00014666666666666666,
      "learning_rate_embeddings": 0.00014666666666666666,
      "loss": 3.2898,
      "step": 184000
    },
    {
      "epoch": 55.56,
      "learning_rate": 0.00014555555555555556,
      "learning_rate_embeddings": 0.00014555555555555556,
      "loss": 3.292,
      "step": 184500
    },
    {
      "epoch": 55.71,
      "learning_rate": 0.00014444444444444444,
      "learning_rate_embeddings": 0.00014444444444444444,
      "loss": 3.2974,
      "step": 185000
    },
    {
      "epoch": 55.86,
      "learning_rate": 0.00014333333333333334,
      "learning_rate_embeddings": 0.00014333333333333334,
      "loss": 3.2978,
      "step": 185500
    },
    {
      "epoch": 56.01,
      "learning_rate": 0.0001422222222222222,
      "learning_rate_embeddings": 0.0001422222222222222,
      "loss": 3.2971,
      "step": 186000
    },
    {
      "epoch": 56.16,
      "learning_rate": 0.00014111111111111111,
      "learning_rate_embeddings": 0.00014111111111111111,
      "loss": 3.2802,
      "step": 186500
    },
    {
      "epoch": 56.31,
      "learning_rate": 0.00014000000000000001,
      "learning_rate_embeddings": 0.00014000000000000001,
      "loss": 3.2838,
      "step": 187000
    },
    {
      "epoch": 56.46,
      "learning_rate": 0.0001388888888888889,
      "learning_rate_embeddings": 0.0001388888888888889,
      "loss": 3.2857,
      "step": 187500
    },
    {
      "epoch": 56.46,
      "eval_loss": 3.322549343109131,
      "eval_runtime": 339.2306,
      "eval_samples_per_second": 500.415,
      "eval_steps_per_second": 3.912,
      "step": 187500
    },
    {
      "epoch": 56.61,
      "learning_rate": 0.0001377777777777778,
      "learning_rate_embeddings": 0.0001377777777777778,
      "loss": 3.2874,
      "step": 188000
    },
    {
      "epoch": 56.76,
      "learning_rate": 0.00013666666666666666,
      "learning_rate_embeddings": 0.00013666666666666666,
      "loss": 3.285,
      "step": 188500
    },
    {
      "epoch": 56.91,
      "learning_rate": 0.00013555555555555556,
      "learning_rate_embeddings": 0.00013555555555555556,
      "loss": 3.2866,
      "step": 189000
    },
    {
      "epoch": 57.06,
      "learning_rate": 0.00013444444444444447,
      "learning_rate_embeddings": 0.00013444444444444447,
      "loss": 3.2867,
      "step": 189500
    },
    {
      "epoch": 57.21,
      "learning_rate": 0.00013333333333333334,
      "learning_rate_embeddings": 0.00013333333333333334,
      "loss": 3.2914,
      "step": 190000
    },
    {
      "epoch": 57.36,
      "learning_rate": 0.00013222222222222221,
      "learning_rate_embeddings": 0.00013222222222222221,
      "loss": 3.2829,
      "step": 190500
    },
    {
      "epoch": 57.51,
      "learning_rate": 0.00013111111111111111,
      "learning_rate_embeddings": 0.00013111111111111111,
      "loss": 3.2859,
      "step": 191000
    },
    {
      "epoch": 57.66,
      "learning_rate": 0.00013000000000000002,
      "learning_rate_embeddings": 0.00013000000000000002,
      "loss": 3.2886,
      "step": 191500
    },
    {
      "epoch": 57.81,
      "learning_rate": 0.0001288888888888889,
      "learning_rate_embeddings": 0.0001288888888888889,
      "loss": 3.2867,
      "step": 192000
    },
    {
      "epoch": 57.96,
      "learning_rate": 0.00012777777777777776,
      "learning_rate_embeddings": 0.00012777777777777776,
      "loss": 3.2824,
      "step": 192500
    },
    {
      "epoch": 58.11,
      "learning_rate": 0.0001266666666666667,
      "learning_rate_embeddings": 0.0001266666666666667,
      "loss": 3.29,
      "step": 193000
    },
    {
      "epoch": 58.27,
      "learning_rate": 0.00012555555555555557,
      "learning_rate_embeddings": 0.00012555555555555557,
      "loss": 3.2847,
      "step": 193500
    },
    {
      "epoch": 58.42,
      "learning_rate": 0.00012444444444444444,
      "learning_rate_embeddings": 0.00012444444444444444,
      "loss": 3.2864,
      "step": 194000
    },
    {
      "epoch": 58.57,
      "learning_rate": 0.00012333333333333334,
      "learning_rate_embeddings": 0.00012333333333333334,
      "loss": 3.2875,
      "step": 194500
    },
    {
      "epoch": 58.72,
      "learning_rate": 0.00012222222222222221,
      "learning_rate_embeddings": 0.00012222222222222221,
      "loss": 3.2882,
      "step": 195000
    },
    {
      "epoch": 58.87,
      "learning_rate": 0.00012111111111111112,
      "learning_rate_embeddings": 0.00012111111111111112,
      "loss": 3.2879,
      "step": 195500
    },
    {
      "epoch": 59.02,
      "learning_rate": 0.00012,
      "learning_rate_embeddings": 0.00012,
      "loss": 3.297,
      "step": 196000
    },
    {
      "epoch": 59.17,
      "learning_rate": 0.00011888888888888889,
      "learning_rate_embeddings": 0.00011888888888888889,
      "loss": 3.274,
      "step": 196500
    },
    {
      "epoch": 59.32,
      "learning_rate": 0.00011777777777777778,
      "learning_rate_embeddings": 0.00011777777777777778,
      "loss": 3.2839,
      "step": 197000
    },
    {
      "epoch": 59.47,
      "learning_rate": 0.00011666666666666667,
      "learning_rate_embeddings": 0.00011666666666666667,
      "loss": 3.2833,
      "step": 197500
    },
    {
      "epoch": 59.62,
      "learning_rate": 0.00011555555555555555,
      "learning_rate_embeddings": 0.00011555555555555555,
      "loss": 3.2791,
      "step": 198000
    },
    {
      "epoch": 59.77,
      "learning_rate": 0.00011444444444444445,
      "learning_rate_embeddings": 0.00011444444444444445,
      "loss": 3.2831,
      "step": 198500
    },
    {
      "epoch": 59.92,
      "learning_rate": 0.00011333333333333333,
      "learning_rate_embeddings": 0.00011333333333333333,
      "loss": 3.2841,
      "step": 199000
    },
    {
      "epoch": 60.07,
      "learning_rate": 0.00011222222222222223,
      "learning_rate_embeddings": 0.00011222222222222223,
      "loss": 3.2858,
      "step": 199500
    },
    {
      "epoch": 60.22,
      "learning_rate": 0.0001111111111111111,
      "learning_rate_embeddings": 0.0001111111111111111,
      "loss": 3.2791,
      "step": 200000
    },
    {
      "epoch": 60.22,
      "eval_loss": 3.319915771484375,
      "eval_runtime": 339.2313,
      "eval_samples_per_second": 500.414,
      "eval_steps_per_second": 3.912,
      "step": 200000
    },
    {
      "epoch": 60.37,
      "learning_rate": 0.00011,
      "learning_rate_embeddings": 0.00011,
      "loss": 3.2785,
      "step": 200500
    },
    {
      "epoch": 60.52,
      "learning_rate": 0.00010888888888888888,
      "learning_rate_embeddings": 0.00010888888888888888,
      "loss": 3.2823,
      "step": 201000
    },
    {
      "epoch": 60.67,
      "learning_rate": 0.00010777777777777778,
      "learning_rate_embeddings": 0.00010777777777777778,
      "loss": 3.2873,
      "step": 201500
    },
    {
      "epoch": 60.82,
      "learning_rate": 0.00010666666666666668,
      "learning_rate_embeddings": 0.00010666666666666668,
      "loss": 3.286,
      "step": 202000
    },
    {
      "epoch": 60.98,
      "learning_rate": 0.00010555555555555555,
      "learning_rate_embeddings": 0.00010555555555555555,
      "loss": 3.2868,
      "step": 202500
    },
    {
      "epoch": 61.13,
      "learning_rate": 0.00010444444444444445,
      "learning_rate_embeddings": 0.00010444444444444445,
      "loss": 3.2854,
      "step": 203000
    },
    {
      "epoch": 61.28,
      "learning_rate": 0.00010333333333333333,
      "learning_rate_embeddings": 0.00010333333333333333,
      "loss": 3.2732,
      "step": 203500
    },
    {
      "epoch": 61.43,
      "learning_rate": 0.00010222222222222223,
      "learning_rate_embeddings": 0.00010222222222222223,
      "loss": 3.2758,
      "step": 204000
    },
    {
      "epoch": 61.58,
      "learning_rate": 0.00010111111111111112,
      "learning_rate_embeddings": 0.00010111111111111112,
      "loss": 3.2804,
      "step": 204500
    },
    {
      "epoch": 61.73,
      "learning_rate": 0.0001,
      "learning_rate_embeddings": 0.0001,
      "loss": 3.2867,
      "step": 205000
    },
    {
      "epoch": 61.88,
      "learning_rate": 9.888888888888889e-05,
      "learning_rate_embeddings": 9.888888888888889e-05,
      "loss": 3.2849,
      "step": 205500
    },
    {
      "epoch": 62.03,
      "learning_rate": 9.777777777777778e-05,
      "learning_rate_embeddings": 9.777777777777778e-05,
      "loss": 3.2842,
      "step": 206000
    },
    {
      "epoch": 62.18,
      "learning_rate": 9.666666666666667e-05,
      "learning_rate_embeddings": 9.666666666666667e-05,
      "loss": 3.2772,
      "step": 206500
    },
    {
      "epoch": 62.33,
      "learning_rate": 9.555555555555557e-05,
      "learning_rate_embeddings": 9.555555555555557e-05,
      "loss": 3.2764,
      "step": 207000
    },
    {
      "epoch": 62.48,
      "learning_rate": 9.444444444444444e-05,
      "learning_rate_embeddings": 9.444444444444444e-05,
      "loss": 3.2743,
      "step": 207500
    },
    {
      "epoch": 62.63,
      "learning_rate": 9.333333333333334e-05,
      "learning_rate_embeddings": 9.333333333333334e-05,
      "loss": 3.282,
      "step": 208000
    },
    {
      "epoch": 62.78,
      "learning_rate": 9.222222222222222e-05,
      "learning_rate_embeddings": 9.222222222222222e-05,
      "loss": 3.2856,
      "step": 208500
    },
    {
      "epoch": 62.93,
      "learning_rate": 9.111111111111112e-05,
      "learning_rate_embeddings": 9.111111111111112e-05,
      "loss": 3.2817,
      "step": 209000
    },
    {
      "epoch": 63.08,
      "learning_rate": 8.999999999999999e-05,
      "learning_rate_embeddings": 8.999999999999999e-05,
      "loss": 3.2886,
      "step": 209500
    },
    {
      "epoch": 63.23,
      "learning_rate": 8.888888888888889e-05,
      "learning_rate_embeddings": 8.888888888888889e-05,
      "loss": 3.2673,
      "step": 210000
    },
    {
      "epoch": 63.38,
      "learning_rate": 8.777777777777778e-05,
      "learning_rate_embeddings": 8.777777777777778e-05,
      "loss": 3.2758,
      "step": 210500
    },
    {
      "epoch": 63.53,
      "learning_rate": 8.666666666666667e-05,
      "learning_rate_embeddings": 8.666666666666667e-05,
      "loss": 3.2835,
      "step": 211000
    },
    {
      "epoch": 63.69,
      "learning_rate": 8.555555555555556e-05,
      "learning_rate_embeddings": 8.555555555555556e-05,
      "loss": 3.2853,
      "step": 211500
    },
    {
      "epoch": 63.84,
      "learning_rate": 8.444444444444444e-05,
      "learning_rate_embeddings": 8.444444444444444e-05,
      "loss": 3.2735,
      "step": 212000
    },
    {
      "epoch": 63.99,
      "learning_rate": 8.333333333333333e-05,
      "learning_rate_embeddings": 8.333333333333333e-05,
      "loss": 3.2807,
      "step": 212500
    },
    {
      "epoch": 63.99,
      "eval_loss": 3.3131635189056396,
      "eval_runtime": 339.5925,
      "eval_samples_per_second": 499.882,
      "eval_steps_per_second": 3.908,
      "step": 212500
    },
    {
      "epoch": 64.14,
      "learning_rate": 8.222222222222223e-05,
      "learning_rate_embeddings": 8.222222222222223e-05,
      "loss": 3.2741,
      "step": 213000
    },
    {
      "epoch": 64.29,
      "learning_rate": 8.11111111111111e-05,
      "learning_rate_embeddings": 8.11111111111111e-05,
      "loss": 3.2718,
      "step": 213500
    },
    {
      "epoch": 64.44,
      "learning_rate": 8e-05,
      "learning_rate_embeddings": 8e-05,
      "loss": 3.2756,
      "step": 214000
    },
    {
      "epoch": 64.59,
      "learning_rate": 7.888888888888888e-05,
      "learning_rate_embeddings": 7.888888888888888e-05,
      "loss": 3.2743,
      "step": 214500
    },
    {
      "epoch": 64.74,
      "learning_rate": 7.777777777777778e-05,
      "learning_rate_embeddings": 7.777777777777778e-05,
      "loss": 3.268,
      "step": 215000
    },
    {
      "epoch": 64.89,
      "learning_rate": 7.666666666666667e-05,
      "learning_rate_embeddings": 7.666666666666667e-05,
      "loss": 3.2793,
      "step": 215500
    },
    {
      "epoch": 65.04,
      "learning_rate": 7.555555555555556e-05,
      "learning_rate_embeddings": 7.555555555555556e-05,
      "loss": 3.2823,
      "step": 216000
    },
    {
      "epoch": 65.19,
      "learning_rate": 7.444444444444444e-05,
      "learning_rate_embeddings": 7.444444444444444e-05,
      "loss": 3.2718,
      "step": 216500
    },
    {
      "epoch": 65.34,
      "learning_rate": 7.333333333333333e-05,
      "learning_rate_embeddings": 7.333333333333333e-05,
      "loss": 3.2742,
      "step": 217000
    },
    {
      "epoch": 65.49,
      "learning_rate": 7.222222222222222e-05,
      "learning_rate_embeddings": 7.222222222222222e-05,
      "loss": 3.2736,
      "step": 217500
    },
    {
      "epoch": 65.64,
      "learning_rate": 7.11111111111111e-05,
      "learning_rate_embeddings": 7.11111111111111e-05,
      "loss": 3.2769,
      "step": 218000
    },
    {
      "epoch": 65.79,
      "learning_rate": 7.000000000000001e-05,
      "learning_rate_embeddings": 7.000000000000001e-05,
      "loss": 3.2778,
      "step": 218500
    },
    {
      "epoch": 65.94,
      "learning_rate": 6.88888888888889e-05,
      "learning_rate_embeddings": 6.88888888888889e-05,
      "loss": 3.2787,
      "step": 219000
    },
    {
      "epoch": 66.09,
      "learning_rate": 6.777777777777778e-05,
      "learning_rate_embeddings": 6.777777777777778e-05,
      "loss": 3.2723,
      "step": 219500
    },
    {
      "epoch": 66.25,
      "learning_rate": 6.666666666666667e-05,
      "learning_rate_embeddings": 6.666666666666667e-05,
      "loss": 3.2704,
      "step": 220000
    },
    {
      "epoch": 66.4,
      "learning_rate": 6.555555555555556e-05,
      "learning_rate_embeddings": 6.555555555555556e-05,
      "loss": 3.2658,
      "step": 220500
    },
    {
      "epoch": 66.55,
      "learning_rate": 6.444444444444444e-05,
      "learning_rate_embeddings": 6.444444444444444e-05,
      "loss": 3.2735,
      "step": 221000
    },
    {
      "epoch": 66.7,
      "learning_rate": 6.333333333333335e-05,
      "learning_rate_embeddings": 6.333333333333335e-05,
      "loss": 3.277,
      "step": 221500
    },
    {
      "epoch": 66.85,
      "learning_rate": 6.222222222222222e-05,
      "learning_rate_embeddings": 6.222222222222222e-05,
      "loss": 3.2733,
      "step": 222000
    },
    {
      "epoch": 67.0,
      "learning_rate": 6.111111111111111e-05,
      "learning_rate_embeddings": 6.111111111111111e-05,
      "loss": 3.2696,
      "step": 222500
    },
    {
      "epoch": 67.15,
      "learning_rate": 6e-05,
      "learning_rate_embeddings": 6e-05,
      "loss": 3.2729,
      "step": 223000
    },
    {
      "epoch": 67.3,
      "learning_rate": 5.888888888888889e-05,
      "learning_rate_embeddings": 5.888888888888889e-05,
      "loss": 3.2657,
      "step": 223500
    },
    {
      "epoch": 67.45,
      "learning_rate": 5.7777777777777776e-05,
      "learning_rate_embeddings": 5.7777777777777776e-05,
      "loss": 3.2644,
      "step": 224000
    },
    {
      "epoch": 67.6,
      "learning_rate": 5.6666666666666664e-05,
      "learning_rate_embeddings": 5.6666666666666664e-05,
      "loss": 3.2699,
      "step": 224500
    },
    {
      "epoch": 67.75,
      "learning_rate": 5.555555555555555e-05,
      "learning_rate_embeddings": 5.555555555555555e-05,
      "loss": 3.2662,
      "step": 225000
    },
    {
      "epoch": 67.75,
      "eval_loss": 3.3130714893341064,
      "eval_runtime": 339.1629,
      "eval_samples_per_second": 500.515,
      "eval_steps_per_second": 3.913,
      "step": 225000
    },
    {
      "epoch": 67.9,
      "learning_rate": 5.444444444444444e-05,
      "learning_rate_embeddings": 5.444444444444444e-05,
      "loss": 3.2699,
      "step": 225500
    },
    {
      "epoch": 68.05,
      "learning_rate": 5.333333333333334e-05,
      "learning_rate_embeddings": 5.333333333333334e-05,
      "loss": 3.2828,
      "step": 226000
    },
    {
      "epoch": 68.2,
      "learning_rate": 5.222222222222223e-05,
      "learning_rate_embeddings": 5.222222222222223e-05,
      "loss": 3.2624,
      "step": 226500
    },
    {
      "epoch": 68.35,
      "learning_rate": 5.1111111111111115e-05,
      "learning_rate_embeddings": 5.1111111111111115e-05,
      "loss": 3.2693,
      "step": 227000
    },
    {
      "epoch": 68.5,
      "learning_rate": 5e-05,
      "learning_rate_embeddings": 5e-05,
      "loss": 3.2713,
      "step": 227500
    },
    {
      "epoch": 68.65,
      "learning_rate": 4.888888888888889e-05,
      "learning_rate_embeddings": 4.888888888888889e-05,
      "loss": 3.2695,
      "step": 228000
    },
    {
      "epoch": 68.8,
      "learning_rate": 4.7777777777777784e-05,
      "learning_rate_embeddings": 4.7777777777777784e-05,
      "loss": 3.2716,
      "step": 228500
    },
    {
      "epoch": 68.95,
      "learning_rate": 4.666666666666667e-05,
      "learning_rate_embeddings": 4.666666666666667e-05,
      "loss": 3.2698,
      "step": 229000
    },
    {
      "epoch": 69.11,
      "learning_rate": 4.555555555555556e-05,
      "learning_rate_embeddings": 4.555555555555556e-05,
      "loss": 3.2712,
      "step": 229500
    },
    {
      "epoch": 69.26,
      "learning_rate": 4.4444444444444447e-05,
      "learning_rate_embeddings": 4.4444444444444447e-05,
      "loss": 3.2709,
      "step": 230000
    },
    {
      "epoch": 69.41,
      "learning_rate": 4.3333333333333334e-05,
      "learning_rate_embeddings": 4.3333333333333334e-05,
      "loss": 3.2711,
      "step": 230500
    },
    {
      "epoch": 69.56,
      "learning_rate": 4.222222222222222e-05,
      "learning_rate_embeddings": 4.222222222222222e-05,
      "loss": 3.2664,
      "step": 231000
    },
    {
      "epoch": 69.71,
      "learning_rate": 4.1111111111111116e-05,
      "learning_rate_embeddings": 4.1111111111111116e-05,
      "loss": 3.2646,
      "step": 231500
    },
    {
      "epoch": 69.86,
      "learning_rate": 4e-05,
      "learning_rate_embeddings": 4e-05,
      "loss": 3.2668,
      "step": 232000
    },
    {
      "epoch": 70.01,
      "learning_rate": 3.888888888888889e-05,
      "learning_rate_embeddings": 3.888888888888889e-05,
      "loss": 3.276,
      "step": 232500
    },
    {
      "epoch": 70.16,
      "learning_rate": 3.777777777777778e-05,
      "learning_rate_embeddings": 3.777777777777778e-05,
      "loss": 3.2614,
      "step": 233000
    },
    {
      "epoch": 70.31,
      "learning_rate": 3.6666666666666666e-05,
      "learning_rate_embeddings": 3.6666666666666666e-05,
      "loss": 3.2621,
      "step": 233500
    },
    {
      "epoch": 70.46,
      "learning_rate": 3.555555555555555e-05,
      "learning_rate_embeddings": 3.555555555555555e-05,
      "loss": 3.2599,
      "step": 234000
    },
    {
      "epoch": 70.61,
      "learning_rate": 3.444444444444445e-05,
      "learning_rate_embeddings": 3.444444444444445e-05,
      "loss": 3.2583,
      "step": 234500
    },
    {
      "epoch": 70.76,
      "learning_rate": 3.3333333333333335e-05,
      "learning_rate_embeddings": 3.3333333333333335e-05,
      "loss": 3.2672,
      "step": 235000
    },
    {
      "epoch": 70.91,
      "learning_rate": 3.222222222222222e-05,
      "learning_rate_embeddings": 3.222222222222222e-05,
      "loss": 3.2601,
      "step": 235500
    },
    {
      "epoch": 71.06,
      "learning_rate": 3.111111111111111e-05,
      "learning_rate_embeddings": 3.111111111111111e-05,
      "loss": 3.2676,
      "step": 236000
    },
    {
      "epoch": 71.21,
      "learning_rate": 3e-05,
      "learning_rate_embeddings": 3e-05,
      "loss": 3.256,
      "step": 236500
    },
    {
      "epoch": 71.36,
      "learning_rate": 2.8888888888888888e-05,
      "learning_rate_embeddings": 2.8888888888888888e-05,
      "loss": 3.2629,
      "step": 237000
    },
    {
      "epoch": 71.51,
      "learning_rate": 2.7777777777777776e-05,
      "learning_rate_embeddings": 2.7777777777777776e-05,
      "loss": 3.261,
      "step": 237500
    },
    {
      "epoch": 71.51,
      "eval_loss": 3.306839942932129,
      "eval_runtime": 339.4312,
      "eval_samples_per_second": 500.119,
      "eval_steps_per_second": 3.909,
      "step": 237500
    },
    {
      "epoch": 71.67,
      "learning_rate": 2.666666666666667e-05,
      "learning_rate_embeddings": 2.666666666666667e-05,
      "loss": 3.2595,
      "step": 238000
    },
    {
      "epoch": 71.82,
      "learning_rate": 2.5555555555555557e-05,
      "learning_rate_embeddings": 2.5555555555555557e-05,
      "loss": 3.2579,
      "step": 238500
    },
    {
      "epoch": 71.97,
      "learning_rate": 2.4444444444444445e-05,
      "learning_rate_embeddings": 2.4444444444444445e-05,
      "loss": 3.2662,
      "step": 239000
    },
    {
      "epoch": 72.12,
      "learning_rate": 2.3333333333333336e-05,
      "learning_rate_embeddings": 2.3333333333333336e-05,
      "loss": 3.2709,
      "step": 239500
    },
    {
      "epoch": 72.27,
      "learning_rate": 2.2222222222222223e-05,
      "learning_rate_embeddings": 2.2222222222222223e-05,
      "loss": 3.2614,
      "step": 240000
    },
    {
      "epoch": 72.42,
      "learning_rate": 2.111111111111111e-05,
      "learning_rate_embeddings": 2.111111111111111e-05,
      "loss": 3.259,
      "step": 240500
    },
    {
      "epoch": 72.57,
      "learning_rate": 2e-05,
      "learning_rate_embeddings": 2e-05,
      "loss": 3.2593,
      "step": 241000
    },
    {
      "epoch": 72.72,
      "learning_rate": 1.888888888888889e-05,
      "learning_rate_embeddings": 1.888888888888889e-05,
      "loss": 3.2592,
      "step": 241500
    },
    {
      "epoch": 72.87,
      "learning_rate": 1.7777777777777777e-05,
      "learning_rate_embeddings": 1.7777777777777777e-05,
      "loss": 3.2553,
      "step": 242000
    },
    {
      "epoch": 73.02,
      "learning_rate": 1.6666666666666667e-05,
      "learning_rate_embeddings": 1.6666666666666667e-05,
      "loss": 3.2617,
      "step": 242500
    },
    {
      "epoch": 73.17,
      "learning_rate": 1.5555555555555555e-05,
      "learning_rate_embeddings": 1.5555555555555555e-05,
      "loss": 3.2566,
      "step": 243000
    },
    {
      "epoch": 73.32,
      "learning_rate": 1.4444444444444444e-05,
      "learning_rate_embeddings": 1.4444444444444444e-05,
      "loss": 3.2525,
      "step": 243500
    },
    {
      "epoch": 73.47,
      "learning_rate": 1.3333333333333335e-05,
      "learning_rate_embeddings": 1.3333333333333335e-05,
      "loss": 3.2539,
      "step": 244000
    },
    {
      "epoch": 73.62,
      "learning_rate": 1.2222222222222222e-05,
      "learning_rate_embeddings": 1.2222222222222222e-05,
      "loss": 3.2569,
      "step": 244500
    },
    {
      "epoch": 73.77,
      "learning_rate": 1.1111111111111112e-05,
      "learning_rate_embeddings": 1.1111111111111112e-05,
      "loss": 3.259,
      "step": 245000
    },
    {
      "epoch": 73.92,
      "learning_rate": 1e-05,
      "learning_rate_embeddings": 1e-05,
      "loss": 3.2507,
      "step": 245500
    },
    {
      "epoch": 74.07,
      "learning_rate": 8.888888888888888e-06,
      "learning_rate_embeddings": 8.888888888888888e-06,
      "loss": 3.2603,
      "step": 246000
    },
    {
      "epoch": 74.22,
      "learning_rate": 7.777777777777777e-06,
      "learning_rate_embeddings": 7.777777777777777e-06,
      "loss": 3.2564,
      "step": 246500
    },
    {
      "epoch": 74.38,
      "learning_rate": 6.6666666666666675e-06,
      "learning_rate_embeddings": 6.6666666666666675e-06,
      "loss": 3.2549,
      "step": 247000
    },
    {
      "epoch": 74.53,
      "learning_rate": 5.555555555555556e-06,
      "learning_rate_embeddings": 5.555555555555556e-06,
      "loss": 3.2578,
      "step": 247500
    },
    {
      "epoch": 74.68,
      "learning_rate": 4.444444444444444e-06,
      "learning_rate_embeddings": 4.444444444444444e-06,
      "loss": 3.2499,
      "step": 248000
    },
    {
      "epoch": 74.83,
      "learning_rate": 3.3333333333333337e-06,
      "learning_rate_embeddings": 3.3333333333333337e-06,
      "loss": 3.2504,
      "step": 248500
    },
    {
      "epoch": 74.98,
      "learning_rate": 2.222222222222222e-06,
      "learning_rate_embeddings": 2.222222222222222e-06,
      "loss": 3.2584,
      "step": 249000
    },
    {
      "epoch": 75.13,
      "learning_rate": 1.111111111111111e-06,
      "learning_rate_embeddings": 1.111111111111111e-06,
      "loss": 3.2556,
      "step": 249500
    },
    {
      "epoch": 75.28,
      "learning_rate": 0.0,
      "learning_rate_embeddings": 0.0,
      "loss": 3.2483,
      "step": 250000
    },
    {
      "epoch": 75.28,
      "eval_loss": 3.2883002758026123,
      "eval_runtime": 339.4284,
      "eval_samples_per_second": 500.123,
      "eval_steps_per_second": 3.91,
      "step": 250000
    }
  ],
  "max_steps": 250000,
  "num_train_epochs": 76,
  "total_flos": 4.1816131633152e+18,
  "trial_name": null,
  "trial_params": null
}