{
  "best_metric": 18.4811,
  "best_model_checkpoint": "ckpt_mt5/google/mt5-small/iwslt2017_de_en/lr0.0001_e50/checkpoint-309168",
  "epoch": 32.0,
  "global_step": 412224,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.04,
      "learning_rate": 0.0001,
      "loss": 6.8681,
      "step": 500
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0001,
      "loss": 3.2786,
      "step": 1000
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0001,
      "loss": 2.9644,
      "step": 1500
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0001,
      "loss": 2.7926,
      "step": 2000
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0001,
      "loss": 2.6636,
      "step": 2500
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0001,
      "loss": 2.5968,
      "step": 3000
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0001,
      "loss": 2.5392,
      "step": 3500
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0001,
      "loss": 2.4696,
      "step": 4000
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0001,
      "loss": 2.4416,
      "step": 4500
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0001,
      "loss": 2.397,
      "step": 5000
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0001,
      "loss": 2.3655,
      "step": 5500
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0001,
      "loss": 2.341,
      "step": 6000
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0001,
      "loss": 2.3054,
      "step": 6500
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0001,
      "loss": 2.2786,
      "step": 7000
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0001,
      "loss": 2.2708,
      "step": 7500
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0001,
      "loss": 2.2475,
      "step": 8000
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0001,
      "loss": 2.2151,
      "step": 8500
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0001,
      "loss": 2.1984,
      "step": 9000
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0001,
      "loss": 2.1732,
      "step": 9500
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0001,
      "loss": 2.1515,
      "step": 10000
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.0001,
      "loss": 2.1661,
      "step": 10500
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.0001,
      "loss": 2.134,
      "step": 11000
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.0001,
      "loss": 2.1317,
      "step": 11500
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.0001,
      "loss": 2.1034,
      "step": 12000
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.0001,
      "loss": 2.0958,
      "step": 12500
    },
    {
      "epoch": 1.0,
      "eval_bleu": 13.4839,
      "eval_gen_len": 16.8176,
      "eval_loss": 1.5927859544754028,
      "eval_runtime": 14.0068,
      "eval_samples_per_second": 63.398,
      "eval_steps_per_second": 3.998,
      "step": 12882
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.0001,
      "loss": 2.0854,
      "step": 13000
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.0001,
      "loss": 2.0328,
      "step": 13500
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0001,
      "loss": 2.0437,
      "step": 14000
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.0001,
      "loss": 2.019,
      "step": 14500
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.0001,
      "loss": 1.9916,
      "step": 15000
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0001,
      "loss": 1.9991,
      "step": 15500
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0001,
      "loss": 1.9934,
      "step": 16000
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.0001,
      "loss": 1.9953,
      "step": 16500
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.0001,
      "loss": 1.9664,
      "step": 17000
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.0001,
      "loss": 1.9691,
      "step": 17500
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.0001,
      "loss": 1.9586,
      "step": 18000
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.0001,
      "loss": 1.9485,
      "step": 18500
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.0001,
      "loss": 1.9447,
      "step": 19000
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.0001,
      "loss": 1.9269,
      "step": 19500
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.0001,
      "loss": 1.9187,
      "step": 20000
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0001,
      "loss": 1.9436,
      "step": 20500
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0001,
      "loss": 1.933,
      "step": 21000
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0001,
      "loss": 1.9067,
      "step": 21500
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.0001,
      "loss": 1.9198,
      "step": 22000
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0001,
      "loss": 1.9039,
      "step": 22500
    },
    {
      "epoch": 1.79,
      "learning_rate": 0.0001,
      "loss": 1.8867,
      "step": 23000
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.0001,
      "loss": 1.8959,
      "step": 23500
    },
    {
      "epoch": 1.86,
      "learning_rate": 0.0001,
      "loss": 1.895,
      "step": 24000
    },
    {
      "epoch": 1.9,
      "learning_rate": 0.0001,
      "loss": 1.8804,
      "step": 24500
    },
    {
      "epoch": 1.94,
      "learning_rate": 0.0001,
      "loss": 1.8732,
      "step": 25000
    },
    {
      "epoch": 1.98,
      "learning_rate": 0.0001,
      "loss": 1.8627,
      "step": 25500
    },
    {
      "epoch": 2.0,
      "eval_bleu": 14.8216,
      "eval_gen_len": 16.8311,
      "eval_loss": 1.4675899744033813,
      "eval_runtime": 14.1723,
      "eval_samples_per_second": 62.657,
      "eval_steps_per_second": 3.951,
      "step": 25764
    },
    {
      "epoch": 2.02,
      "learning_rate": 0.0001,
      "loss": 1.8391,
      "step": 26000
    },
    {
      "epoch": 2.06,
      "learning_rate": 0.0001,
      "loss": 1.8187,
      "step": 26500
    },
    {
      "epoch": 2.1,
      "learning_rate": 0.0001,
      "loss": 1.834,
      "step": 27000
    },
    {
      "epoch": 2.13,
      "learning_rate": 0.0001,
      "loss": 1.8193,
      "step": 27500
    },
    {
      "epoch": 2.17,
      "learning_rate": 0.0001,
      "loss": 1.8134,
      "step": 28000
    },
    {
      "epoch": 2.21,
      "learning_rate": 0.0001,
      "loss": 1.8209,
      "step": 28500
    },
    {
      "epoch": 2.25,
      "learning_rate": 0.0001,
      "loss": 1.8186,
      "step": 29000
    },
    {
      "epoch": 2.29,
      "learning_rate": 0.0001,
      "loss": 1.7822,
      "step": 29500
    },
    {
      "epoch": 2.33,
      "learning_rate": 0.0001,
      "loss": 1.8017,
      "step": 30000
    },
    {
      "epoch": 2.37,
      "learning_rate": 0.0001,
      "loss": 1.7953,
      "step": 30500
    },
    {
      "epoch": 2.41,
      "learning_rate": 0.0001,
      "loss": 1.7979,
      "step": 31000
    },
    {
      "epoch": 2.45,
      "learning_rate": 0.0001,
      "loss": 1.7886,
      "step": 31500
    },
    {
      "epoch": 2.48,
      "learning_rate": 0.0001,
      "loss": 1.7835,
      "step": 32000
    },
    {
      "epoch": 2.52,
      "learning_rate": 0.0001,
      "loss": 1.7881,
      "step": 32500
    },
    {
      "epoch": 2.56,
      "learning_rate": 0.0001,
      "loss": 1.7839,
      "step": 33000
    },
    {
      "epoch": 2.6,
      "learning_rate": 0.0001,
      "loss": 1.7697,
      "step": 33500
    },
    {
      "epoch": 2.64,
      "learning_rate": 0.0001,
      "loss": 1.772,
      "step": 34000
    },
    {
      "epoch": 2.68,
      "learning_rate": 0.0001,
      "loss": 1.7674,
      "step": 34500
    },
    {
      "epoch": 2.72,
      "learning_rate": 0.0001,
      "loss": 1.7538,
      "step": 35000
    },
    {
      "epoch": 2.76,
      "learning_rate": 0.0001,
      "loss": 1.7578,
      "step": 35500
    },
    {
      "epoch": 2.79,
      "learning_rate": 0.0001,
      "loss": 1.7539,
      "step": 36000
    },
    {
      "epoch": 2.83,
      "learning_rate": 0.0001,
      "loss": 1.7745,
      "step": 36500
    },
    {
      "epoch": 2.87,
      "learning_rate": 0.0001,
      "loss": 1.7529,
      "step": 37000
    },
    {
      "epoch": 2.91,
      "learning_rate": 0.0001,
      "loss": 1.7632,
      "step": 37500
    },
    {
      "epoch": 2.95,
      "learning_rate": 0.0001,
      "loss": 1.738,
      "step": 38000
    },
    {
      "epoch": 2.99,
      "learning_rate": 0.0001,
      "loss": 1.7471,
      "step": 38500
    },
    {
      "epoch": 3.0,
      "eval_bleu": 15.6618,
      "eval_gen_len": 16.8018,
      "eval_loss": 1.4083540439605713,
      "eval_runtime": 14.9957,
      "eval_samples_per_second": 59.217,
      "eval_steps_per_second": 3.734,
      "step": 38646
    },
    {
      "epoch": 3.03,
      "learning_rate": 0.0001,
      "loss": 1.7076,
      "step": 39000
    },
    {
      "epoch": 3.07,
      "learning_rate": 0.0001,
      "loss": 1.7085,
      "step": 39500
    },
    {
      "epoch": 3.11,
      "learning_rate": 0.0001,
      "loss": 1.6977,
      "step": 40000
    },
    {
      "epoch": 3.14,
      "learning_rate": 0.0001,
      "loss": 1.7124,
      "step": 40500
    },
    {
      "epoch": 3.18,
      "learning_rate": 0.0001,
      "loss": 1.692,
      "step": 41000
    },
    {
      "epoch": 3.22,
      "learning_rate": 0.0001,
      "loss": 1.7019,
      "step": 41500
    },
    {
      "epoch": 3.26,
      "learning_rate": 0.0001,
      "loss": 1.6821,
      "step": 42000
    },
    {
      "epoch": 3.3,
      "learning_rate": 0.0001,
      "loss": 1.6962,
      "step": 42500
    },
    {
      "epoch": 3.34,
      "learning_rate": 0.0001,
      "loss": 1.6862,
      "step": 43000
    },
    {
      "epoch": 3.38,
      "learning_rate": 0.0001,
      "loss": 1.6819,
      "step": 43500
    },
    {
      "epoch": 3.42,
      "learning_rate": 0.0001,
      "loss": 1.6857,
      "step": 44000
    },
    {
      "epoch": 3.45,
      "learning_rate": 0.0001,
      "loss": 1.6892,
      "step": 44500
    },
    {
      "epoch": 3.49,
      "learning_rate": 0.0001,
      "loss": 1.6814,
      "step": 45000
    },
    {
      "epoch": 3.53,
      "learning_rate": 0.0001,
      "loss": 1.6839,
      "step": 45500
    },
    {
      "epoch": 3.57,
      "learning_rate": 0.0001,
      "loss": 1.6872,
      "step": 46000
    },
    {
      "epoch": 3.61,
      "learning_rate": 0.0001,
      "loss": 1.6696,
      "step": 46500
    },
    {
      "epoch": 3.65,
      "learning_rate": 0.0001,
      "loss": 1.6673,
      "step": 47000
    },
    {
      "epoch": 3.69,
      "learning_rate": 0.0001,
      "loss": 1.6645,
      "step": 47500
    },
    {
      "epoch": 3.73,
      "learning_rate": 0.0001,
      "loss": 1.665,
      "step": 48000
    },
    {
      "epoch": 3.76,
      "learning_rate": 0.0001,
      "loss": 1.6807,
      "step": 48500
    },
    {
      "epoch": 3.8,
      "learning_rate": 0.0001,
      "loss": 1.6788,
      "step": 49000
    },
    {
      "epoch": 3.84,
      "learning_rate": 0.0001,
      "loss": 1.6623,
      "step": 49500
    },
    {
      "epoch": 3.88,
      "learning_rate": 0.0001,
      "loss": 1.6563,
      "step": 50000
    },
    {
      "epoch": 3.92,
      "learning_rate": 0.0001,
      "loss": 1.6697,
      "step": 50500
    },
    {
      "epoch": 3.96,
      "learning_rate": 0.0001,
      "loss": 1.6651,
      "step": 51000
    },
    {
      "epoch": 4.0,
      "learning_rate": 0.0001,
      "loss": 1.6605,
      "step": 51500
    },
    {
      "epoch": 4.0,
      "eval_bleu": 16.0376,
      "eval_gen_len": 16.7725,
      "eval_loss": 1.3691717386245728,
      "eval_runtime": 18.2082,
      "eval_samples_per_second": 48.769,
      "eval_steps_per_second": 3.076,
      "step": 51528
    },
    {
      "epoch": 4.04,
      "learning_rate": 0.0001,
      "loss": 1.635,
      "step": 52000
    },
    {
      "epoch": 4.08,
      "learning_rate": 0.0001,
      "loss": 1.5957,
      "step": 52500
    },
    {
      "epoch": 4.11,
      "learning_rate": 0.0001,
      "loss": 1.625,
      "step": 53000
    },
    {
      "epoch": 4.15,
      "learning_rate": 0.0001,
      "loss": 1.6288,
      "step": 53500
    },
    {
      "epoch": 4.19,
      "learning_rate": 0.0001,
      "loss": 1.6191,
      "step": 54000
    },
    {
      "epoch": 4.23,
      "learning_rate": 0.0001,
      "loss": 1.6191,
      "step": 54500
    },
    {
      "epoch": 4.27,
      "learning_rate": 0.0001,
      "loss": 1.6116,
      "step": 55000
    },
    {
      "epoch": 4.31,
      "learning_rate": 0.0001,
      "loss": 1.622,
      "step": 55500
    },
    {
      "epoch": 4.35,
      "learning_rate": 0.0001,
      "loss": 1.6097,
      "step": 56000
    },
    {
      "epoch": 4.39,
      "learning_rate": 0.0001,
      "loss": 1.6044,
      "step": 56500
    },
    {
      "epoch": 4.42,
      "learning_rate": 0.0001,
      "loss": 1.61,
      "step": 57000
    },
    {
      "epoch": 4.46,
      "learning_rate": 0.0001,
      "loss": 1.6213,
      "step": 57500
    },
    {
      "epoch": 4.5,
      "learning_rate": 0.0001,
      "loss": 1.6078,
      "step": 58000
    },
    {
      "epoch": 4.54,
      "learning_rate": 0.0001,
      "loss": 1.605,
      "step": 58500
    },
    {
      "epoch": 4.58,
      "learning_rate": 0.0001,
      "loss": 1.6027,
      "step": 59000
    },
    {
      "epoch": 4.62,
      "learning_rate": 0.0001,
      "loss": 1.613,
      "step": 59500
    },
    {
      "epoch": 4.66,
      "learning_rate": 0.0001,
      "loss": 1.5998,
      "step": 60000
    },
    {
      "epoch": 4.7,
      "learning_rate": 0.0001,
      "loss": 1.6038,
      "step": 60500
    },
    {
      "epoch": 4.74,
      "learning_rate": 0.0001,
      "loss": 1.5999,
      "step": 61000
    },
    {
      "epoch": 4.77,
      "learning_rate": 0.0001,
      "loss": 1.5953,
      "step": 61500
    },
    {
      "epoch": 4.81,
      "learning_rate": 0.0001,
      "loss": 1.596,
      "step": 62000
    },
    {
      "epoch": 4.85,
      "learning_rate": 0.0001,
      "loss": 1.6096,
      "step": 62500
    },
    {
      "epoch": 4.89,
      "learning_rate": 0.0001,
      "loss": 1.5909,
      "step": 63000
    },
    {
      "epoch": 4.93,
      "learning_rate": 0.0001,
      "loss": 1.604,
      "step": 63500
    },
    {
      "epoch": 4.97,
      "learning_rate": 0.0001,
      "loss": 1.5936,
      "step": 64000
    },
    {
      "epoch": 5.0,
      "eval_bleu": 16.4937,
      "eval_gen_len": 16.7883,
      "eval_loss": 1.3442012071609497,
      "eval_runtime": 15.5957,
      "eval_samples_per_second": 56.939,
      "eval_steps_per_second": 3.591,
      "step": 64410
    },
    {
      "epoch": 5.01,
      "learning_rate": 0.0001,
      "loss": 1.5869,
      "step": 64500
    },
    {
      "epoch": 5.05,
      "learning_rate": 0.0001,
      "loss": 1.5558,
      "step": 65000
    },
    {
      "epoch": 5.08,
      "learning_rate": 0.0001,
      "loss": 1.5421,
      "step": 65500
    },
    {
      "epoch": 5.12,
      "learning_rate": 0.0001,
      "loss": 1.5559,
      "step": 66000
    },
    {
      "epoch": 5.16,
      "learning_rate": 0.0001,
      "loss": 1.5598,
      "step": 66500
    },
    {
      "epoch": 5.2,
      "learning_rate": 0.0001,
      "loss": 1.5552,
      "step": 67000
    },
    {
      "epoch": 5.24,
      "learning_rate": 0.0001,
      "loss": 1.5606,
      "step": 67500
    },
    {
      "epoch": 5.28,
      "learning_rate": 0.0001,
      "loss": 1.5585,
      "step": 68000
    },
    {
      "epoch": 5.32,
      "learning_rate": 0.0001,
      "loss": 1.55,
      "step": 68500
    },
    {
      "epoch": 5.36,
      "learning_rate": 0.0001,
      "loss": 1.5535,
      "step": 69000
    },
    {
      "epoch": 5.4,
      "learning_rate": 0.0001,
      "loss": 1.5524,
      "step": 69500
    },
    {
      "epoch": 5.43,
      "learning_rate": 0.0001,
      "loss": 1.5537,
      "step": 70000
    },
    {
      "epoch": 5.47,
      "learning_rate": 0.0001,
      "loss": 1.5519,
      "step": 70500
    },
    {
      "epoch": 5.51,
      "learning_rate": 0.0001,
      "loss": 1.5548,
      "step": 71000
    },
    {
      "epoch": 5.55,
      "learning_rate": 0.0001,
      "loss": 1.5586,
      "step": 71500
    },
    {
      "epoch": 5.59,
      "learning_rate": 0.0001,
      "loss": 1.5431,
      "step": 72000
    },
    {
      "epoch": 5.63,
      "learning_rate": 0.0001,
      "loss": 1.5502,
      "step": 72500
    },
    {
      "epoch": 5.67,
      "learning_rate": 0.0001,
      "loss": 1.5385,
      "step": 73000
    },
    {
      "epoch": 5.71,
      "learning_rate": 0.0001,
      "loss": 1.5495,
      "step": 73500
    },
    {
      "epoch": 5.74,
      "learning_rate": 0.0001,
      "loss": 1.547,
      "step": 74000
    },
    {
      "epoch": 5.78,
      "learning_rate": 0.0001,
      "loss": 1.5469,
      "step": 74500
    },
    {
      "epoch": 5.82,
      "learning_rate": 0.0001,
      "loss": 1.5509,
      "step": 75000
    },
    {
      "epoch": 5.86,
      "learning_rate": 0.0001,
      "loss": 1.5311,
      "step": 75500
    },
    {
      "epoch": 5.9,
      "learning_rate": 0.0001,
      "loss": 1.5358,
      "step": 76000
    },
    {
      "epoch": 5.94,
      "learning_rate": 0.0001,
      "loss": 1.5507,
      "step": 76500
    },
    {
      "epoch": 5.98,
      "learning_rate": 0.0001,
      "loss": 1.5365,
      "step": 77000
    },
    {
      "epoch": 6.0,
      "eval_bleu": 16.7335,
      "eval_gen_len": 16.7962,
      "eval_loss": 1.3152118921279907,
      "eval_runtime": 15.4124,
      "eval_samples_per_second": 57.616,
      "eval_steps_per_second": 3.633,
      "step": 77292
    },
    {
      "epoch": 6.02,
      "learning_rate": 0.0001,
      "loss": 1.5246,
      "step": 77500
    },
    {
      "epoch": 6.05,
      "learning_rate": 0.0001,
      "loss": 1.4913,
      "step": 78000
    },
    {
      "epoch": 6.09,
      "learning_rate": 0.0001,
      "loss": 1.5096,
      "step": 78500
    },
    {
      "epoch": 6.13,
      "learning_rate": 0.0001,
      "loss": 1.5072,
      "step": 79000
    },
    {
      "epoch": 6.17,
      "learning_rate": 0.0001,
      "loss": 1.515,
      "step": 79500
    },
    {
      "epoch": 6.21,
      "learning_rate": 0.0001,
      "loss": 1.5047,
      "step": 80000
    },
    {
      "epoch": 6.25,
      "learning_rate": 0.0001,
      "loss": 1.4956,
      "step": 80500
    },
    {
      "epoch": 6.29,
      "learning_rate": 0.0001,
      "loss": 1.517,
      "step": 81000
    },
    {
      "epoch": 6.33,
      "learning_rate": 0.0001,
      "loss": 1.5118,
      "step": 81500
    },
    {
      "epoch": 6.37,
      "learning_rate": 0.0001,
      "loss": 1.4976,
      "step": 82000
    },
    {
      "epoch": 6.4,
      "learning_rate": 0.0001,
      "loss": 1.5008,
      "step": 82500
    },
    {
      "epoch": 6.44,
      "learning_rate": 0.0001,
      "loss": 1.4977,
      "step": 83000
    },
    {
      "epoch": 6.48,
      "learning_rate": 0.0001,
      "loss": 1.4969,
      "step": 83500
    },
    {
      "epoch": 6.52,
      "learning_rate": 0.0001,
      "loss": 1.4927,
      "step": 84000
    },
    {
      "epoch": 6.56,
      "learning_rate": 0.0001,
      "loss": 1.4977,
      "step": 84500
    },
    {
      "epoch": 6.6,
      "learning_rate": 0.0001,
      "loss": 1.4968,
      "step": 85000
    },
    {
      "epoch": 6.64,
      "learning_rate": 0.0001,
      "loss": 1.5029,
      "step": 85500
    },
    {
      "epoch": 6.68,
      "learning_rate": 0.0001,
      "loss": 1.4973,
      "step": 86000
    },
    {
      "epoch": 6.71,
      "learning_rate": 0.0001,
      "loss": 1.5114,
      "step": 86500
    },
    {
      "epoch": 6.75,
      "learning_rate": 0.0001,
      "loss": 1.5043,
      "step": 87000
    },
    {
      "epoch": 6.79,
      "learning_rate": 0.0001,
      "loss": 1.5064,
      "step": 87500
    },
    {
      "epoch": 6.83,
      "learning_rate": 0.0001,
      "loss": 1.4971,
      "step": 88000
    },
    {
      "epoch": 6.87,
      "learning_rate": 0.0001,
      "loss": 1.483,
      "step": 88500
    },
    {
      "epoch": 6.91,
      "learning_rate": 0.0001,
      "loss": 1.4963,
      "step": 89000
    },
    {
      "epoch": 6.95,
      "learning_rate": 0.0001,
      "loss": 1.5032,
      "step": 89500
    },
    {
      "epoch": 6.99,
      "learning_rate": 0.0001,
      "loss": 1.4955,
      "step": 90000
    },
    {
      "epoch": 7.0,
      "eval_bleu": 16.9384,
      "eval_gen_len": 16.8255,
      "eval_loss": 1.3002352714538574,
      "eval_runtime": 15.6135,
      "eval_samples_per_second": 56.874,
      "eval_steps_per_second": 3.587,
      "step": 90174
    },
    {
      "epoch": 7.03,
      "learning_rate": 0.0001,
      "loss": 1.4769,
      "step": 90500
    },
    {
      "epoch": 7.06,
      "learning_rate": 0.0001,
      "loss": 1.4537,
      "step": 91000
    },
    {
      "epoch": 7.1,
      "learning_rate": 0.0001,
      "loss": 1.4728,
      "step": 91500
    },
    {
      "epoch": 7.14,
      "learning_rate": 0.0001,
      "loss": 1.4545,
      "step": 92000
    },
    {
      "epoch": 7.18,
      "learning_rate": 0.0001,
      "loss": 1.4648,
      "step": 92500
    },
    {
      "epoch": 7.22,
      "learning_rate": 0.0001,
      "loss": 1.4609,
      "step": 93000
    },
    {
      "epoch": 7.26,
      "learning_rate": 0.0001,
      "loss": 1.4535,
      "step": 93500
    },
    {
      "epoch": 7.3,
      "learning_rate": 0.0001,
      "loss": 1.4677,
      "step": 94000
    },
    {
      "epoch": 7.34,
      "learning_rate": 0.0001,
      "loss": 1.454,
      "step": 94500
    },
    {
      "epoch": 7.37,
      "learning_rate": 0.0001,
      "loss": 1.4727,
      "step": 95000
    },
    {
      "epoch": 7.41,
      "learning_rate": 0.0001,
      "loss": 1.4576,
      "step": 95500
    },
    {
      "epoch": 7.45,
      "learning_rate": 0.0001,
      "loss": 1.4601,
      "step": 96000
    },
    {
      "epoch": 7.49,
      "learning_rate": 0.0001,
      "loss": 1.4641,
      "step": 96500
    },
    {
      "epoch": 7.53,
      "learning_rate": 0.0001,
      "loss": 1.4663,
      "step": 97000
    },
    {
      "epoch": 7.57,
      "learning_rate": 0.0001,
      "loss": 1.4601,
      "step": 97500
    },
    {
      "epoch": 7.61,
      "learning_rate": 0.0001,
      "loss": 1.4517,
      "step": 98000
    },
    {
      "epoch": 7.65,
      "learning_rate": 0.0001,
      "loss": 1.4485,
      "step": 98500
    },
    {
      "epoch": 7.69,
      "learning_rate": 0.0001,
      "loss": 1.4629,
      "step": 99000
    },
    {
      "epoch": 7.72,
      "learning_rate": 0.0001,
      "loss": 1.4509,
      "step": 99500
    },
    {
      "epoch": 7.76,
      "learning_rate": 0.0001,
      "loss": 1.4568,
      "step": 100000
    },
    {
      "epoch": 7.8,
      "learning_rate": 0.0001,
      "loss": 1.4634,
      "step": 100500
    },
    {
      "epoch": 7.84,
      "learning_rate": 0.0001,
      "loss": 1.449,
      "step": 101000
    },
    {
      "epoch": 7.88,
      "learning_rate": 0.0001,
      "loss": 1.4615,
      "step": 101500
    },
    {
      "epoch": 7.92,
      "learning_rate": 0.0001,
      "loss": 1.4681,
      "step": 102000
    },
    {
      "epoch": 7.96,
      "learning_rate": 0.0001,
      "loss": 1.4626,
      "step": 102500
    },
    {
      "epoch": 8.0,
      "learning_rate": 0.0001,
      "loss": 1.4576,
      "step": 103000
    },
    {
      "epoch": 8.0,
      "eval_bleu": 17.3767,
      "eval_gen_len": 16.8063,
      "eval_loss": 1.2891216278076172,
      "eval_runtime": 15.4362,
      "eval_samples_per_second": 57.527,
      "eval_steps_per_second": 3.628,
      "step": 103056
    },
    {
      "epoch": 8.03,
      "learning_rate": 0.0001,
      "loss": 1.4132,
      "step": 103500
    },
    {
      "epoch": 8.07,
      "learning_rate": 0.0001,
      "loss": 1.4191,
      "step": 104000
    },
    {
      "epoch": 8.11,
      "learning_rate": 0.0001,
      "loss": 1.4139,
      "step": 104500
    },
    {
      "epoch": 8.15,
      "learning_rate": 0.0001,
      "loss": 1.4263,
      "step": 105000
    },
    {
      "epoch": 8.19,
      "learning_rate": 0.0001,
      "loss": 1.4243,
      "step": 105500
    },
    {
      "epoch": 8.23,
      "learning_rate": 0.0001,
      "loss": 1.418,
      "step": 106000
    },
    {
      "epoch": 8.27,
      "learning_rate": 0.0001,
      "loss": 1.4326,
      "step": 106500
    },
    {
      "epoch": 8.31,
      "learning_rate": 0.0001,
      "loss": 1.425,
      "step": 107000
    },
    {
      "epoch": 8.34,
      "learning_rate": 0.0001,
      "loss": 1.4222,
      "step": 107500
    },
    {
      "epoch": 8.38,
      "learning_rate": 0.0001,
      "loss": 1.4213,
      "step": 108000
    },
    {
      "epoch": 8.42,
      "learning_rate": 0.0001,
      "loss": 1.4312,
      "step": 108500
    },
    {
      "epoch": 8.46,
      "learning_rate": 0.0001,
      "loss": 1.4182,
      "step": 109000
    },
    {
      "epoch": 8.5,
      "learning_rate": 0.0001,
      "loss": 1.4212,
      "step": 109500
    },
    {
      "epoch": 8.54,
      "learning_rate": 0.0001,
      "loss": 1.4269,
      "step": 110000
    },
    {
      "epoch": 8.58,
      "learning_rate": 0.0001,
      "loss": 1.4172,
      "step": 110500
    },
    {
      "epoch": 8.62,
      "learning_rate": 0.0001,
      "loss": 1.4246,
      "step": 111000
    },
    {
      "epoch": 8.66,
      "learning_rate": 0.0001,
      "loss": 1.4246,
      "step": 111500
    },
    {
      "epoch": 8.69,
      "learning_rate": 0.0001,
      "loss": 1.4239,
      "step": 112000
    },
    {
      "epoch": 8.73,
      "learning_rate": 0.0001,
      "loss": 1.4251,
      "step": 112500
    },
    {
      "epoch": 8.77,
      "learning_rate": 0.0001,
      "loss": 1.4164,
      "step": 113000
    },
    {
      "epoch": 8.81,
      "learning_rate": 0.0001,
      "loss": 1.4263,
      "step": 113500
    },
    {
      "epoch": 8.85,
      "learning_rate": 0.0001,
      "loss": 1.4201,
      "step": 114000
    },
    {
      "epoch": 8.89,
      "learning_rate": 0.0001,
      "loss": 1.4265,
      "step": 114500
    },
    {
      "epoch": 8.93,
      "learning_rate": 0.0001,
      "loss": 1.4339,
      "step": 115000
    },
    {
      "epoch": 8.97,
      "learning_rate": 0.0001,
      "loss": 1.4183,
      "step": 115500
    },
    {
      "epoch": 9.0,
      "eval_bleu": 17.4296,
      "eval_gen_len": 16.7838,
      "eval_loss": 1.2837210893630981,
      "eval_runtime": 14.5647,
      "eval_samples_per_second": 60.969,
      "eval_steps_per_second": 3.845,
      "step": 115938
    },
    {
      "epoch": 9.0,
      "learning_rate": 0.0001,
      "loss": 1.4209,
      "step": 116000
    },
    {
      "epoch": 9.04,
      "learning_rate": 0.0001,
      "loss": 1.3891,
      "step": 116500
    },
    {
      "epoch": 9.08,
      "learning_rate": 0.0001,
      "loss": 1.3783,
      "step": 117000
    },
    {
      "epoch": 9.12,
      "learning_rate": 0.0001,
      "loss": 1.377,
      "step": 117500
    },
    {
      "epoch": 9.16,
      "learning_rate": 0.0001,
      "loss": 1.3986,
      "step": 118000
    },
    {
      "epoch": 9.2,
      "learning_rate": 0.0001,
      "loss": 1.3933,
      "step": 118500
    },
    {
      "epoch": 9.24,
      "learning_rate": 0.0001,
      "loss": 1.3749,
      "step": 119000
    },
    {
      "epoch": 9.28,
      "learning_rate": 0.0001,
      "loss": 1.3683,
      "step": 119500
    },
    {
      "epoch": 9.32,
      "learning_rate": 0.0001,
      "loss": 1.3863,
      "step": 120000
    },
    {
      "epoch": 9.35,
      "learning_rate": 0.0001,
      "loss": 1.3873,
      "step": 120500
    },
    {
      "epoch": 9.39,
      "learning_rate": 0.0001,
      "loss": 1.387,
      "step": 121000
    },
    {
      "epoch": 9.43,
      "learning_rate": 0.0001,
      "loss": 1.3925,
      "step": 121500
    },
    {
      "epoch": 9.47,
      "learning_rate": 0.0001,
      "loss": 1.4003,
      "step": 122000
    },
    {
      "epoch": 9.51,
      "learning_rate": 0.0001,
      "loss": 1.3969,
      "step": 122500
    },
    {
      "epoch": 9.55,
      "learning_rate": 0.0001,
      "loss": 1.4046,
      "step": 123000
    },
    {
      "epoch": 9.59,
      "learning_rate": 0.0001,
      "loss": 1.3959,
      "step": 123500
    },
    {
      "epoch": 9.63,
      "learning_rate": 0.0001,
      "loss": 1.3851,
      "step": 124000
    },
    {
      "epoch": 9.66,
      "learning_rate": 0.0001,
      "loss": 1.3809,
      "step": 124500
    },
    {
      "epoch": 9.7,
      "learning_rate": 0.0001,
      "loss": 1.3903,
      "step": 125000
    },
    {
      "epoch": 9.74,
      "learning_rate": 0.0001,
      "loss": 1.3801,
      "step": 125500
    },
    {
      "epoch": 9.78,
      "learning_rate": 0.0001,
      "loss": 1.4103,
      "step": 126000
    },
    {
      "epoch": 9.82,
      "learning_rate": 0.0001,
      "loss": 1.3926,
      "step": 126500
    },
    {
      "epoch": 9.86,
      "learning_rate": 0.0001,
      "loss": 1.3913,
      "step": 127000
    },
    {
      "epoch": 9.9,
      "learning_rate": 0.0001,
      "loss": 1.3922,
      "step": 127500
    },
    {
      "epoch": 9.94,
      "learning_rate": 0.0001,
      "loss": 1.4018,
      "step": 128000
    },
    {
      "epoch": 9.98,
      "learning_rate": 0.0001,
      "loss": 1.404,
      "step": 128500
    },
    {
      "epoch": 10.0,
      "eval_bleu": 17.8832,
      "eval_gen_len": 16.7928,
      "eval_loss": 1.2739335298538208,
      "eval_runtime": 17.7078,
      "eval_samples_per_second": 50.147,
      "eval_steps_per_second": 3.162,
      "step": 128820
    },
    {
      "epoch": 10.01,
      "learning_rate": 0.0001,
      "loss": 1.3823,
      "step": 129000
    },
    {
      "epoch": 10.05,
      "learning_rate": 0.0001,
      "loss": 1.3571,
      "step": 129500
    },
    {
      "epoch": 10.09,
      "learning_rate": 0.0001,
      "loss": 1.356,
      "step": 130000
    },
    {
      "epoch": 10.13,
      "learning_rate": 0.0001,
      "loss": 1.3567,
      "step": 130500
    },
    {
      "epoch": 10.17,
      "learning_rate": 0.0001,
      "loss": 1.3508,
      "step": 131000
    },
    {
      "epoch": 10.21,
      "learning_rate": 0.0001,
      "loss": 1.3586,
      "step": 131500
    },
    {
      "epoch": 10.25,
      "learning_rate": 0.0001,
      "loss": 1.347,
      "step": 132000
    },
    {
      "epoch": 10.29,
      "learning_rate": 0.0001,
      "loss": 1.3699,
      "step": 132500
    },
    {
      "epoch": 10.32,
      "learning_rate": 0.0001,
      "loss": 1.3485,
      "step": 133000
    },
    {
      "epoch": 10.36,
      "learning_rate": 0.0001,
      "loss": 1.3646,
      "step": 133500
    },
    {
      "epoch": 10.4,
      "learning_rate": 0.0001,
      "loss": 1.3679,
      "step": 134000
    },
    {
      "epoch": 10.44,
      "learning_rate": 0.0001,
      "loss": 1.3605,
      "step": 134500
    },
    {
      "epoch": 10.48,
      "learning_rate": 0.0001,
      "loss": 1.3657,
      "step": 135000
    },
    {
      "epoch": 10.52,
      "learning_rate": 0.0001,
      "loss": 1.3617,
      "step": 135500
    },
    {
      "epoch": 10.56,
      "learning_rate": 0.0001,
      "loss": 1.3626,
      "step": 136000
    },
    {
      "epoch": 10.6,
      "learning_rate": 0.0001,
      "loss": 1.3749,
      "step": 136500
    },
    {
      "epoch": 10.63,
      "learning_rate": 0.0001,
      "loss": 1.3509,
      "step": 137000
    },
    {
      "epoch": 10.67,
      "learning_rate": 0.0001,
      "loss": 1.3638,
      "step": 137500
    },
    {
      "epoch": 10.71,
      "learning_rate": 0.0001,
      "loss": 1.3637,
      "step": 138000
    },
    {
      "epoch": 10.75,
      "learning_rate": 0.0001,
      "loss": 1.3584,
      "step": 138500
    },
    {
      "epoch": 10.79,
      "learning_rate": 0.0001,
      "loss": 1.3553,
      "step": 139000
    },
    {
      "epoch": 10.83,
      "learning_rate": 0.0001,
      "loss": 1.3813,
      "step": 139500
    },
    {
      "epoch": 10.87,
      "learning_rate": 0.0001,
      "loss": 1.3632,
      "step": 140000
    },
    {
      "epoch": 10.91,
      "learning_rate": 0.0001,
      "loss": 1.3649,
      "step": 140500
    },
    {
      "epoch": 10.95,
      "learning_rate": 0.0001,
      "loss": 1.3561,
      "step": 141000
    },
    {
      "epoch": 10.98,
      "learning_rate": 0.0001,
      "loss": 1.3637,
      "step": 141500
    },
    {
      "epoch": 11.0,
      "eval_bleu": 17.984,
      "eval_gen_len": 16.7928,
      "eval_loss": 1.2685599327087402,
      "eval_runtime": 14.6211,
      "eval_samples_per_second": 60.734,
      "eval_steps_per_second": 3.83,
      "step": 141702
    },
    {
      "epoch": 11.02,
      "learning_rate": 0.0001,
      "loss": 1.3437,
      "step": 142000
    },
    {
      "epoch": 11.06,
      "learning_rate": 0.0001,
      "loss": 1.3294,
      "step": 142500
    },
    {
      "epoch": 11.1,
      "learning_rate": 0.0001,
      "loss": 1.3409,
      "step": 143000
    },
    {
      "epoch": 11.14,
      "learning_rate": 0.0001,
      "loss": 1.3218,
      "step": 143500
    },
    {
      "epoch": 11.18,
      "learning_rate": 0.0001,
      "loss": 1.3243,
      "step": 144000
    },
    {
      "epoch": 11.22,
      "learning_rate": 0.0001,
      "loss": 1.3325,
      "step": 144500
    },
    {
      "epoch": 11.26,
      "learning_rate": 0.0001,
      "loss": 1.3259,
      "step": 145000
    },
    {
      "epoch": 11.29,
      "learning_rate": 0.0001,
      "loss": 1.3284,
      "step": 145500
    },
    {
      "epoch": 11.33,
      "learning_rate": 0.0001,
      "loss": 1.341,
      "step": 146000
    },
    {
      "epoch": 11.37,
      "learning_rate": 0.0001,
      "loss": 1.3337,
      "step": 146500
    },
    {
      "epoch": 11.41,
      "learning_rate": 0.0001,
      "loss": 1.3348,
      "step": 147000
    },
    {
      "epoch": 11.45,
      "learning_rate": 0.0001,
      "loss": 1.3365,
      "step": 147500
    },
    {
      "epoch": 11.49,
      "learning_rate": 0.0001,
      "loss": 1.3414,
      "step": 148000
    },
    {
      "epoch": 11.53,
      "learning_rate": 0.0001,
      "loss": 1.3314,
      "step": 148500
    },
    {
      "epoch": 11.57,
      "learning_rate": 0.0001,
      "loss": 1.3407,
      "step": 149000
    },
    {
      "epoch": 11.61,
      "learning_rate": 0.0001,
      "loss": 1.3343,
      "step": 149500
    },
    {
      "epoch": 11.64,
      "learning_rate": 0.0001,
      "loss": 1.35,
      "step": 150000
    },
    {
      "epoch": 11.68,
      "learning_rate": 0.0001,
      "loss": 1.3287,
      "step": 150500
    },
    {
      "epoch": 11.72,
      "learning_rate": 0.0001,
      "loss": 1.3362,
      "step": 151000
    },
    {
      "epoch": 11.76,
      "learning_rate": 0.0001,
      "loss": 1.3376,
      "step": 151500
    },
    {
      "epoch": 11.8,
      "learning_rate": 0.0001,
      "loss": 1.3463,
      "step": 152000
    },
    {
      "epoch": 11.84,
      "learning_rate": 0.0001,
      "loss": 1.3395,
      "step": 152500
    },
    {
      "epoch": 11.88,
      "learning_rate": 0.0001,
      "loss": 1.3302,
      "step": 153000
    },
    {
      "epoch": 11.92,
      "learning_rate": 0.0001,
      "loss": 1.329,
      "step": 153500
    },
    {
      "epoch": 11.95,
      "learning_rate": 0.0001,
      "loss": 1.342,
      "step": 154000
    },
    {
      "epoch": 11.99,
      "learning_rate": 0.0001,
      "loss": 1.3394,
      "step": 154500
    },
    {
      "epoch": 12.0,
      "eval_bleu": 17.9459,
      "eval_gen_len": 16.7489,
      "eval_loss": 1.2629603147506714,
      "eval_runtime": 15.8099,
      "eval_samples_per_second": 56.167,
      "eval_steps_per_second": 3.542,
      "step": 154584
    },
    {
      "epoch": 12.03,
      "learning_rate": 0.0001,
      "loss": 1.3062,
      "step": 155000
    },
    {
      "epoch": 12.07,
      "learning_rate": 0.0001,
      "loss": 1.3171,
      "step": 155500
    },
    {
      "epoch": 12.11,
      "learning_rate": 0.0001,
      "loss": 1.3054,
      "step": 156000
    },
    {
      "epoch": 12.15,
      "learning_rate": 0.0001,
      "loss": 1.3102,
      "step": 156500
    },
    {
      "epoch": 12.19,
      "learning_rate": 0.0001,
      "loss": 1.3038,
      "step": 157000
    },
    {
      "epoch": 12.23,
      "learning_rate": 0.0001,
      "loss": 1.3101,
      "step": 157500
    },
    {
      "epoch": 12.27,
      "learning_rate": 0.0001,
      "loss": 1.3029,
      "step": 158000
    },
    {
      "epoch": 12.3,
      "learning_rate": 0.0001,
      "loss": 1.3094,
      "step": 158500
    },
    {
      "epoch": 12.34,
      "learning_rate": 0.0001,
      "loss": 1.3118,
      "step": 159000
    },
    {
      "epoch": 12.38,
      "learning_rate": 0.0001,
      "loss": 1.3101,
      "step": 159500
    },
    {
      "epoch": 12.42,
      "learning_rate": 0.0001,
      "loss": 1.3082,
      "step": 160000
    },
    {
      "epoch": 12.46,
      "learning_rate": 0.0001,
      "loss": 1.3243,
      "step": 160500
    },
    {
      "epoch": 12.5,
      "learning_rate": 0.0001,
      "loss": 1.3215,
      "step": 161000
    },
    {
      "epoch": 12.54,
      "learning_rate": 0.0001,
      "loss": 1.3113,
      "step": 161500
    },
    {
      "epoch": 12.58,
      "learning_rate": 0.0001,
      "loss": 1.3082,
      "step": 162000
    },
    {
      "epoch": 12.61,
      "learning_rate": 0.0001,
      "loss": 1.311,
      "step": 162500
    },
    {
      "epoch": 12.65,
      "learning_rate": 0.0001,
      "loss": 1.3107,
      "step": 163000
    },
    {
      "epoch": 12.69,
      "learning_rate": 0.0001,
      "loss": 1.3205,
      "step": 163500
    },
    {
      "epoch": 12.73,
      "learning_rate": 0.0001,
      "loss": 1.3238,
      "step": 164000
    },
    {
      "epoch": 12.77,
      "learning_rate": 0.0001,
      "loss": 1.3082,
      "step": 164500
    },
    {
      "epoch": 12.81,
      "learning_rate": 0.0001,
      "loss": 1.3093,
      "step": 165000
    },
    {
      "epoch": 12.85,
      "learning_rate": 0.0001,
      "loss": 1.3088,
      "step": 165500
    },
    {
      "epoch": 12.89,
      "learning_rate": 0.0001,
      "loss": 1.3183,
      "step": 166000
    },
    {
      "epoch": 12.93,
      "learning_rate": 0.0001,
      "loss": 1.3148,
      "step": 166500
    },
    {
      "epoch": 12.96,
      "learning_rate": 0.0001,
      "loss": 1.2985,
      "step": 167000
    },
    {
      "epoch": 13.0,
      "eval_bleu": 17.94,
      "eval_gen_len": 16.8029,
      "eval_loss": 1.2564747333526611,
      "eval_runtime": 17.4465,
      "eval_samples_per_second": 50.898,
      "eval_steps_per_second": 3.21,
      "step": 167466
    },
    {
      "epoch": 13.0,
      "learning_rate": 0.0001,
      "loss": 1.2996,
      "step": 167500
    },
    {
      "epoch": 13.04,
      "learning_rate": 0.0001,
      "loss": 1.2734,
      "step": 168000
    },
    {
      "epoch": 13.08,
      "learning_rate": 0.0001,
      "loss": 1.2883,
      "step": 168500
    },
    {
      "epoch": 13.12,
      "learning_rate": 0.0001,
      "loss": 1.2891,
      "step": 169000
    },
    {
      "epoch": 13.16,
      "learning_rate": 0.0001,
      "loss": 1.2815,
      "step": 169500
    },
    {
      "epoch": 13.2,
      "learning_rate": 0.0001,
      "loss": 1.2755,
      "step": 170000
    },
    {
      "epoch": 13.24,
      "learning_rate": 0.0001,
      "loss": 1.2908,
      "step": 170500
    },
    {
      "epoch": 13.27,
      "learning_rate": 0.0001,
      "loss": 1.2796,
      "step": 171000
    },
    {
      "epoch": 13.31,
      "learning_rate": 0.0001,
      "loss": 1.2934,
      "step": 171500
    },
    {
      "epoch": 13.35,
      "learning_rate": 0.0001,
      "loss": 1.2884,
      "step": 172000
    },
    {
      "epoch": 13.39,
      "learning_rate": 0.0001,
      "loss": 1.2746,
      "step": 172500
    },
    {
      "epoch": 13.43,
      "learning_rate": 0.0001,
      "loss": 1.2857,
      "step": 173000
    },
    {
      "epoch": 13.47,
      "learning_rate": 0.0001,
      "loss": 1.2778,
      "step": 173500
    },
    {
      "epoch": 13.51,
      "learning_rate": 0.0001,
      "loss": 1.2878,
      "step": 174000
    },
    {
      "epoch": 13.55,
      "learning_rate": 0.0001,
      "loss": 1.2977,
      "step": 174500
    },
    {
      "epoch": 13.58,
      "learning_rate": 0.0001,
      "loss": 1.2902,
      "step": 175000
    },
    {
      "epoch": 13.62,
      "learning_rate": 0.0001,
      "loss": 1.2941,
      "step": 175500
    },
    {
      "epoch": 13.66,
      "learning_rate": 0.0001,
      "loss": 1.2984,
      "step": 176000
    },
    {
      "epoch": 13.7,
      "learning_rate": 0.0001,
      "loss": 1.2916,
      "step": 176500
    },
    {
      "epoch": 13.74,
      "learning_rate": 0.0001,
      "loss": 1.2986,
      "step": 177000
    },
    {
      "epoch": 13.78,
      "learning_rate": 0.0001,
      "loss": 1.2846,
      "step": 177500
    },
    {
      "epoch": 13.82,
      "learning_rate": 0.0001,
      "loss": 1.2846,
      "step": 178000
    },
    {
      "epoch": 13.86,
      "learning_rate": 0.0001,
      "loss": 1.2952,
      "step": 178500
    },
    {
      "epoch": 13.9,
      "learning_rate": 0.0001,
      "loss": 1.29,
      "step": 179000
    },
    {
      "epoch": 13.93,
      "learning_rate": 0.0001,
      "loss": 1.2898,
      "step": 179500
    },
    {
      "epoch": 13.97,
      "learning_rate": 0.0001,
      "loss": 1.2831,
      "step": 180000
    },
    {
      "epoch": 14.0,
      "eval_bleu": 18.1011,
      "eval_gen_len": 16.7804,
      "eval_loss": 1.2606099843978882,
      "eval_runtime": 14.6084,
      "eval_samples_per_second": 60.787,
      "eval_steps_per_second": 3.833,
      "step": 180348
    },
    {
      "epoch": 14.01,
      "learning_rate": 0.0001,
      "loss": 1.2847,
      "step": 180500
    },
    {
      "epoch": 14.05,
      "learning_rate": 0.0001,
      "loss": 1.257,
      "step": 181000
    },
    {
      "epoch": 14.09,
      "learning_rate": 0.0001,
      "loss": 1.2509,
      "step": 181500
    },
    {
      "epoch": 14.13,
      "learning_rate": 0.0001,
      "loss": 1.2439,
      "step": 182000
    },
    {
      "epoch": 14.17,
      "learning_rate": 0.0001,
      "loss": 1.2469,
      "step": 182500
    },
    {
      "epoch": 14.21,
      "learning_rate": 0.0001,
      "loss": 1.2629,
      "step": 183000
    },
    {
      "epoch": 14.24,
      "learning_rate": 0.0001,
      "loss": 1.2571,
      "step": 183500
    },
    {
      "epoch": 14.28,
      "learning_rate": 0.0001,
      "loss": 1.2549,
      "step": 184000
    },
    {
      "epoch": 14.32,
      "learning_rate": 0.0001,
      "loss": 1.2677,
      "step": 184500
    },
    {
      "epoch": 14.36,
      "learning_rate": 0.0001,
      "loss": 1.2664,
      "step": 185000
    },
    {
      "epoch": 14.4,
      "learning_rate": 0.0001,
      "loss": 1.2657,
      "step": 185500
    },
    {
      "epoch": 14.44,
      "learning_rate": 0.0001,
      "loss": 1.2644,
      "step": 186000
    },
    {
      "epoch": 14.48,
      "learning_rate": 0.0001,
      "loss": 1.2741,
      "step": 186500
    },
    {
      "epoch": 14.52,
      "learning_rate": 0.0001,
      "loss": 1.2747,
      "step": 187000
    },
    {
      "epoch": 14.56,
      "learning_rate": 0.0001,
      "loss": 1.2797,
      "step": 187500
    },
    {
      "epoch": 14.59,
      "learning_rate": 0.0001,
      "loss": 1.2698,
      "step": 188000
    },
    {
      "epoch": 14.63,
      "learning_rate": 0.0001,
      "loss": 1.2551,
      "step": 188500
    },
    {
      "epoch": 14.67,
      "learning_rate": 0.0001,
      "loss": 1.2723,
      "step": 189000
    },
    {
      "epoch": 14.71,
      "learning_rate": 0.0001,
      "loss": 1.2712,
      "step": 189500
    },
    {
      "epoch": 14.75,
      "learning_rate": 0.0001,
      "loss": 1.2745,
      "step": 190000
    },
    {
      "epoch": 14.79,
      "learning_rate": 0.0001,
      "loss": 1.2761,
      "step": 190500
    },
    {
      "epoch": 14.83,
      "learning_rate": 0.0001,
      "loss": 1.2692,
      "step": 191000
    },
    {
      "epoch": 14.87,
      "learning_rate": 0.0001,
      "loss": 1.284,
      "step": 191500
    },
    {
      "epoch": 14.9,
      "learning_rate": 0.0001,
      "loss": 1.2679,
      "step": 192000
    },
    {
      "epoch": 14.94,
      "learning_rate": 0.0001,
      "loss": 1.2735,
      "step": 192500
    },
    {
      "epoch": 14.98,
      "learning_rate": 0.0001,
      "loss": 1.2739,
      "step": 193000
    },
    {
      "epoch": 15.0,
      "eval_bleu": 18.0819,
      "eval_gen_len": 16.8176,
      "eval_loss": 1.2597845792770386,
      "eval_runtime": 14.6833,
      "eval_samples_per_second": 60.477,
      "eval_steps_per_second": 3.814,
      "step": 193230
    },
    {
      "epoch": 15.02,
      "learning_rate": 0.0001,
      "loss": 1.2476,
      "step": 193500
    },
    {
      "epoch": 15.06,
      "learning_rate": 0.0001,
      "loss": 1.229,
      "step": 194000
    },
    {
      "epoch": 15.1,
      "learning_rate": 0.0001,
      "loss": 1.238,
      "step": 194500
    },
    {
      "epoch": 15.14,
      "learning_rate": 0.0001,
      "loss": 1.2385,
      "step": 195000
    },
    {
      "epoch": 15.18,
      "learning_rate": 0.0001,
      "loss": 1.244,
      "step": 195500
    },
    {
      "epoch": 15.22,
      "learning_rate": 0.0001,
      "loss": 1.2324,
      "step": 196000
    },
    {
      "epoch": 15.25,
      "learning_rate": 0.0001,
      "loss": 1.2433,
      "step": 196500
    },
    {
      "epoch": 15.29,
      "learning_rate": 0.0001,
      "loss": 1.2337,
      "step": 197000
    },
    {
      "epoch": 15.33,
      "learning_rate": 0.0001,
      "loss": 1.2408,
      "step": 197500
    },
    {
      "epoch": 15.37,
      "learning_rate": 0.0001,
      "loss": 1.2505,
      "step": 198000
    },
    {
      "epoch": 15.41,
      "learning_rate": 0.0001,
      "loss": 1.2442,
      "step": 198500
    },
    {
      "epoch": 15.45,
      "learning_rate": 0.0001,
      "loss": 1.2459,
      "step": 199000
    },
    {
      "epoch": 15.49,
      "learning_rate": 0.0001,
      "loss": 1.2494,
      "step": 199500
    },
    {
      "epoch": 15.53,
      "learning_rate": 0.0001,
      "loss": 1.2551,
      "step": 200000
    },
    {
      "epoch": 15.56,
      "learning_rate": 0.0001,
      "loss": 1.2425,
      "step": 200500
    },
    {
      "epoch": 15.6,
      "learning_rate": 0.0001,
      "loss": 1.2507,
      "step": 201000
    },
    {
      "epoch": 15.64,
      "learning_rate": 0.0001,
      "loss": 1.2517,
      "step": 201500
    },
    {
      "epoch": 15.68,
      "learning_rate": 0.0001,
      "loss": 1.2627,
      "step": 202000
    },
    {
      "epoch": 15.72,
      "learning_rate": 0.0001,
      "loss": 1.2552,
      "step": 202500
    },
    {
      "epoch": 15.76,
      "learning_rate": 0.0001,
      "loss": 1.246,
      "step": 203000
    },
    {
      "epoch": 15.8,
      "learning_rate": 0.0001,
      "loss": 1.2496,
      "step": 203500
    },
    {
      "epoch": 15.84,
      "learning_rate": 0.0001,
      "loss": 1.259,
      "step": 204000
    },
    {
      "epoch": 15.87,
      "learning_rate": 0.0001,
      "loss": 1.2492,
      "step": 204500
    },
    {
      "epoch": 15.91,
      "learning_rate": 0.0001,
      "loss": 1.2527,
      "step": 205000
    },
    {
      "epoch": 15.95,
      "learning_rate": 0.0001,
      "loss": 1.2537,
      "step": 205500
    },
    {
      "epoch": 15.99,
      "learning_rate": 0.0001,
      "loss": 1.2521,
      "step": 206000
    },
    {
      "epoch": 16.0,
      "eval_bleu": 18.2473,
      "eval_gen_len": 16.8232,
      "eval_loss": 1.2520203590393066,
      "eval_runtime": 14.5261,
      "eval_samples_per_second": 61.131,
      "eval_steps_per_second": 3.855,
      "step": 206112
    },
    {
      "epoch": 16.03,
      "learning_rate": 0.0001,
      "loss": 1.2204,
      "step": 206500
    },
    {
      "epoch": 16.07,
      "learning_rate": 0.0001,
      "loss": 1.2186,
      "step": 207000
    },
    {
      "epoch": 16.11,
      "learning_rate": 0.0001,
      "loss": 1.2252,
      "step": 207500
    },
    {
      "epoch": 16.15,
      "learning_rate": 0.0001,
      "loss": 1.2187,
      "step": 208000
    },
    {
      "epoch": 16.19,
      "learning_rate": 0.0001,
      "loss": 1.2226,
      "step": 208500
    },
    {
      "epoch": 16.22,
      "learning_rate": 0.0001,
      "loss": 1.2227,
      "step": 209000
    },
    {
      "epoch": 16.26,
      "learning_rate": 0.0001,
      "loss": 1.2196,
      "step": 209500
    },
    {
      "epoch": 16.3,
      "learning_rate": 0.0001,
      "loss": 1.2259,
      "step": 210000
    },
    {
      "epoch": 16.34,
      "learning_rate": 0.0001,
      "loss": 1.2259,
      "step": 210500
    },
    {
      "epoch": 16.38,
      "learning_rate": 0.0001,
      "loss": 1.2134,
      "step": 211000
    },
    {
      "epoch": 16.42,
      "learning_rate": 0.0001,
      "loss": 1.2315,
      "step": 211500
    },
    {
      "epoch": 16.46,
      "learning_rate": 0.0001,
      "loss": 1.2313,
      "step": 212000
    },
    {
      "epoch": 16.5,
      "learning_rate": 0.0001,
      "loss": 1.2361,
      "step": 212500
    },
    {
      "epoch": 16.53,
      "learning_rate": 0.0001,
      "loss": 1.2379,
      "step": 213000
    },
    {
      "epoch": 16.57,
      "learning_rate": 0.0001,
      "loss": 1.2346,
      "step": 213500
    },
    {
      "epoch": 16.61,
      "learning_rate": 0.0001,
      "loss": 1.2276,
      "step": 214000
    },
    {
      "epoch": 16.65,
      "learning_rate": 0.0001,
      "loss": 1.2308,
      "step": 214500
    },
    {
      "epoch": 16.69,
      "learning_rate": 0.0001,
      "loss": 1.2392,
      "step": 215000
    },
    {
      "epoch": 16.73,
      "learning_rate": 0.0001,
      "loss": 1.2351,
      "step": 215500
    },
    {
      "epoch": 16.77,
      "learning_rate": 0.0001,
      "loss": 1.232,
      "step": 216000
    },
    {
      "epoch": 16.81,
      "learning_rate": 0.0001,
      "loss": 1.2262,
      "step": 216500
    },
    {
      "epoch": 16.85,
      "learning_rate": 0.0001,
      "loss": 1.2316,
      "step": 217000
    },
    {
      "epoch": 16.88,
      "learning_rate": 0.0001,
      "loss": 1.232,
      "step": 217500
    },
    {
      "epoch": 16.92,
      "learning_rate": 0.0001,
      "loss": 1.2333,
      "step": 218000
    },
    {
      "epoch": 16.96,
      "learning_rate": 0.0001,
      "loss": 1.2467,
      "step": 218500
    },
    {
      "epoch": 17.0,
      "eval_bleu": 18.4357,
      "eval_gen_len": 16.8052,
      "eval_loss": 1.2525311708450317,
      "eval_runtime": 15.7073,
      "eval_samples_per_second": 56.534,
      "eval_steps_per_second": 3.565,
      "step": 218994
    },
    {
      "epoch": 17.0,
      "learning_rate": 0.0001,
      "loss": 1.2252,
      "step": 219000
    },
    {
      "epoch": 17.04,
      "learning_rate": 0.0001,
      "loss": 1.2032,
      "step": 219500
    },
    {
      "epoch": 17.08,
      "learning_rate": 0.0001,
      "loss": 1.2093,
      "step": 220000
    },
    {
      "epoch": 17.12,
      "learning_rate": 0.0001,
      "loss": 1.1994,
      "step": 220500
    },
    {
      "epoch": 17.16,
      "learning_rate": 0.0001,
      "loss": 1.2054,
      "step": 221000
    },
    {
      "epoch": 17.19,
      "learning_rate": 0.0001,
      "loss": 1.2008,
      "step": 221500
    },
    {
      "epoch": 17.23,
      "learning_rate": 0.0001,
      "loss": 1.2087,
      "step": 222000
    },
    {
      "epoch": 17.27,
      "learning_rate": 0.0001,
      "loss": 1.2107,
      "step": 222500
    },
    {
      "epoch": 17.31,
      "learning_rate": 0.0001,
      "loss": 1.2042,
      "step": 223000
    },
    {
      "epoch": 17.35,
      "learning_rate": 0.0001,
      "loss": 1.2033,
      "step": 223500
    },
    {
      "epoch": 17.39,
      "learning_rate": 0.0001,
      "loss": 1.2048,
      "step": 224000
    },
    {
      "epoch": 17.43,
      "learning_rate": 0.0001,
      "loss": 1.2097,
      "step": 224500
    },
    {
      "epoch": 17.47,
      "learning_rate": 0.0001,
      "loss": 1.2034,
      "step": 225000
    },
    {
      "epoch": 17.51,
      "learning_rate": 0.0001,
      "loss": 1.2124,
      "step": 225500
    },
    {
      "epoch": 17.54,
      "learning_rate": 0.0001,
      "loss": 1.2062,
      "step": 226000
    },
    {
      "epoch": 17.58,
      "learning_rate": 0.0001,
      "loss": 1.2174,
      "step": 226500
    },
    {
      "epoch": 17.62,
      "learning_rate": 0.0001,
      "loss": 1.2292,
      "step": 227000
    },
    {
      "epoch": 17.66,
      "learning_rate": 0.0001,
      "loss": 1.2289,
      "step": 227500
    },
    {
      "epoch": 17.7,
      "learning_rate": 0.0001,
      "loss": 1.2146,
      "step": 228000
    },
    {
      "epoch": 17.74,
      "learning_rate": 0.0001,
      "loss": 1.2199,
      "step": 228500
    },
    {
      "epoch": 17.78,
      "learning_rate": 0.0001,
      "loss": 1.2193,
      "step": 229000
    },
    {
      "epoch": 17.82,
      "learning_rate": 0.0001,
      "loss": 1.2053,
      "step": 229500
    },
    {
      "epoch": 17.85,
      "learning_rate": 0.0001,
      "loss": 1.2153,
      "step": 230000
    },
    {
      "epoch": 17.89,
      "learning_rate": 0.0001,
      "loss": 1.2123,
      "step": 230500
    },
    {
      "epoch": 17.93,
      "learning_rate": 0.0001,
      "loss": 1.2085,
      "step": 231000
    },
    {
      "epoch": 17.97,
      "learning_rate": 0.0001,
      "loss": 1.2176,
      "step": 231500
    },
    {
      "epoch": 18.0,
      "eval_bleu": 18.211,
      "eval_gen_len": 16.8221,
      "eval_loss": 1.244187355041504,
      "eval_runtime": 14.5564,
      "eval_samples_per_second": 61.004,
      "eval_steps_per_second": 3.847,
      "step": 231876
    },
    {
      "epoch": 18.01,
      "learning_rate": 0.0001,
      "loss": 1.2041,
      "step": 232000
    },
    {
      "epoch": 18.05,
      "learning_rate": 0.0001,
      "loss": 1.1793,
      "step": 232500
    },
    {
      "epoch": 18.09,
      "learning_rate": 0.0001,
      "loss": 1.1872,
      "step": 233000
    },
    {
      "epoch": 18.13,
      "learning_rate": 0.0001,
      "loss": 1.1749,
      "step": 233500
    },
    {
      "epoch": 18.16,
      "learning_rate": 0.0001,
      "loss": 1.1858,
      "step": 234000
    },
    {
      "epoch": 18.2,
      "learning_rate": 0.0001,
      "loss": 1.1815,
      "step": 234500
    },
    {
      "epoch": 18.24,
      "learning_rate": 0.0001,
      "loss": 1.1865,
      "step": 235000
    },
    {
      "epoch": 18.28,
      "learning_rate": 0.0001,
      "loss": 1.1888,
      "step": 235500
    },
    {
      "epoch": 18.32,
      "learning_rate": 0.0001,
      "loss": 1.1891,
      "step": 236000
    },
    {
      "epoch": 18.36,
      "learning_rate": 0.0001,
      "loss": 1.197,
      "step": 236500
    },
    {
      "epoch": 18.4,
      "learning_rate": 0.0001,
      "loss": 1.2004,
      "step": 237000
    },
    {
      "epoch": 18.44,
      "learning_rate": 0.0001,
      "loss": 1.1939,
      "step": 237500
    },
    {
      "epoch": 18.48,
      "learning_rate": 0.0001,
      "loss": 1.1939,
      "step": 238000
    },
    {
      "epoch": 18.51,
      "learning_rate": 0.0001,
      "loss": 1.1918,
      "step": 238500
    },
    {
      "epoch": 18.55,
      "learning_rate": 0.0001,
      "loss": 1.1939,
      "step": 239000
    },
    {
      "epoch": 18.59,
      "learning_rate": 0.0001,
      "loss": 1.1951,
      "step": 239500
    },
    {
      "epoch": 18.63,
      "learning_rate": 0.0001,
      "loss": 1.204,
      "step": 240000
    },
    {
      "epoch": 18.67,
      "learning_rate": 0.0001,
      "loss": 1.1944,
      "step": 240500
    },
    {
      "epoch": 18.71,
      "learning_rate": 0.0001,
      "loss": 1.1878,
      "step": 241000
    },
    {
      "epoch": 18.75,
      "learning_rate": 0.0001,
      "loss": 1.1954,
      "step": 241500
    },
    {
      "epoch": 18.79,
      "learning_rate": 0.0001,
      "loss": 1.1992,
      "step": 242000
    },
    {
      "epoch": 18.82,
      "learning_rate": 0.0001,
      "loss": 1.2055,
      "step": 242500
    },
    {
      "epoch": 18.86,
      "learning_rate": 0.0001,
      "loss": 1.2059,
      "step": 243000
    },
    {
      "epoch": 18.9,
      "learning_rate": 0.0001,
      "loss": 1.2074,
      "step": 243500
    },
    {
      "epoch": 18.94,
      "learning_rate": 0.0001,
      "loss": 1.1933,
      "step": 244000
    },
    {
      "epoch": 18.98,
      "learning_rate": 0.0001,
      "loss": 1.2035,
      "step": 244500
    },
    {
      "epoch": 19.0,
      "eval_bleu": 18.1322,
      "eval_gen_len": 16.8007,
      "eval_loss": 1.2485110759735107,
      "eval_runtime": 14.5368,
      "eval_samples_per_second": 61.086,
      "eval_steps_per_second": 3.852,
      "step": 244758
    },
    {
      "epoch": 19.02,
      "learning_rate": 0.0001,
      "loss": 1.1817,
      "step": 245000
    },
    {
      "epoch": 19.06,
      "learning_rate": 0.0001,
      "loss": 1.1702,
      "step": 245500
    },
    {
      "epoch": 19.1,
      "learning_rate": 0.0001,
      "loss": 1.1744,
      "step": 246000
    },
    {
      "epoch": 19.14,
      "learning_rate": 0.0001,
      "loss": 1.1675,
      "step": 246500
    },
    {
      "epoch": 19.17,
      "learning_rate": 0.0001,
      "loss": 1.1711,
      "step": 247000
    },
    {
      "epoch": 19.21,
      "learning_rate": 0.0001,
      "loss": 1.1691,
      "step": 247500
    },
    {
      "epoch": 19.25,
      "learning_rate": 0.0001,
      "loss": 1.1747,
      "step": 248000
    },
    {
      "epoch": 19.29,
      "learning_rate": 0.0001,
      "loss": 1.1743,
      "step": 248500
    },
    {
      "epoch": 19.33,
      "learning_rate": 0.0001,
      "loss": 1.1893,
      "step": 249000
    },
    {
      "epoch": 19.37,
      "learning_rate": 0.0001,
      "loss": 1.1713,
      "step": 249500
    },
    {
      "epoch": 19.41,
      "learning_rate": 0.0001,
      "loss": 1.174,
      "step": 250000
    },
    {
      "epoch": 19.45,
      "learning_rate": 0.0001,
      "loss": 1.1902,
      "step": 250500
    },
    {
      "epoch": 19.48,
      "learning_rate": 0.0001,
      "loss": 1.1832,
      "step": 251000
    },
    {
      "epoch": 19.52,
      "learning_rate": 0.0001,
      "loss": 1.1735,
      "step": 251500
    },
    {
      "epoch": 19.56,
      "learning_rate": 0.0001,
      "loss": 1.1791,
      "step": 252000
    },
    {
      "epoch": 19.6,
      "learning_rate": 0.0001,
      "loss": 1.1861,
      "step": 252500
    },
    {
      "epoch": 19.64,
      "learning_rate": 0.0001,
      "loss": 1.177,
      "step": 253000
    },
    {
      "epoch": 19.68,
      "learning_rate": 0.0001,
      "loss": 1.1679,
      "step": 253500
    },
    {
      "epoch": 19.72,
      "learning_rate": 0.0001,
      "loss": 1.1826,
      "step": 254000
    },
    {
      "epoch": 19.76,
      "learning_rate": 0.0001,
      "loss": 1.1812,
      "step": 254500
    },
    {
      "epoch": 19.8,
      "learning_rate": 0.0001,
      "loss": 1.1782,
      "step": 255000
    },
    {
      "epoch": 19.83,
      "learning_rate": 0.0001,
      "loss": 1.1848,
      "step": 255500
    },
    {
      "epoch": 19.87,
      "learning_rate": 0.0001,
      "loss": 1.1813,
      "step": 256000
    },
    {
      "epoch": 19.91,
      "learning_rate": 0.0001,
      "loss": 1.1975,
      "step": 256500
    },
    {
      "epoch": 19.95,
      "learning_rate": 0.0001,
      "loss": 1.1883,
      "step": 257000
    },
    {
      "epoch": 19.99,
      "learning_rate": 0.0001,
      "loss": 1.1813,
      "step": 257500
    },
    {
      "epoch": 20.0,
      "eval_bleu": 18.3148,
      "eval_gen_len": 16.8502,
      "eval_loss": 1.2500028610229492,
      "eval_runtime": 15.5104,
      "eval_samples_per_second": 57.252,
      "eval_steps_per_second": 3.61,
      "step": 257640
    },
    {
      "epoch": 20.03,
      "learning_rate": 0.0001,
      "loss": 1.1539,
      "step": 258000
    },
    {
      "epoch": 20.07,
      "learning_rate": 0.0001,
      "loss": 1.1499,
      "step": 258500
    },
    {
      "epoch": 20.11,
      "learning_rate": 0.0001,
      "loss": 1.1511,
      "step": 259000
    },
    {
      "epoch": 20.14,
      "learning_rate": 0.0001,
      "loss": 1.1543,
      "step": 259500
    },
    {
      "epoch": 20.18,
      "learning_rate": 0.0001,
      "loss": 1.1562,
      "step": 260000
    },
    {
      "epoch": 20.22,
      "learning_rate": 0.0001,
      "loss": 1.1448,
      "step": 260500
    },
    {
      "epoch": 20.26,
      "learning_rate": 0.0001,
      "loss": 1.152,
      "step": 261000
    },
    {
      "epoch": 20.3,
      "learning_rate": 0.0001,
      "loss": 1.1541,
      "step": 261500
    },
    {
      "epoch": 20.34,
      "learning_rate": 0.0001,
      "loss": 1.1638,
      "step": 262000
    },
    {
      "epoch": 20.38,
      "learning_rate": 0.0001,
      "loss": 1.1674,
      "step": 262500
    },
    {
      "epoch": 20.42,
      "learning_rate": 0.0001,
      "loss": 1.163,
      "step": 263000
    },
    {
      "epoch": 20.45,
      "learning_rate": 0.0001,
      "loss": 1.1728,
      "step": 263500
    },
    {
      "epoch": 20.49,
      "learning_rate": 0.0001,
      "loss": 1.1561,
      "step": 264000
    },
    {
      "epoch": 20.53,
      "learning_rate": 0.0001,
      "loss": 1.1711,
      "step": 264500
    },
    {
      "epoch": 20.57,
      "learning_rate": 0.0001,
      "loss": 1.1616,
      "step": 265000
    },
    {
      "epoch": 20.61,
      "learning_rate": 0.0001,
      "loss": 1.1648,
      "step": 265500
    },
    {
      "epoch": 20.65,
      "learning_rate": 0.0001,
      "loss": 1.1701,
      "step": 266000
    },
    {
      "epoch": 20.69,
      "learning_rate": 0.0001,
      "loss": 1.166,
      "step": 266500
    },
    {
      "epoch": 20.73,
      "learning_rate": 0.0001,
      "loss": 1.1677,
      "step": 267000
    },
    {
      "epoch": 20.77,
      "learning_rate": 0.0001,
      "loss": 1.1582,
      "step": 267500
    },
    {
      "epoch": 20.8,
      "learning_rate": 0.0001,
      "loss": 1.1709,
      "step": 268000
    },
    {
      "epoch": 20.84,
      "learning_rate": 0.0001,
      "loss": 1.1627,
      "step": 268500
    },
    {
      "epoch": 20.88,
      "learning_rate": 0.0001,
      "loss": 1.1811,
      "step": 269000
    },
    {
      "epoch": 20.92,
      "learning_rate": 0.0001,
      "loss": 1.1711,
      "step": 269500
    },
    {
      "epoch": 20.96,
      "learning_rate": 0.0001,
      "loss": 1.1743,
      "step": 270000
    },
    {
      "epoch": 21.0,
      "learning_rate": 0.0001,
      "loss": 1.1628,
      "step": 270500
    },
    {
      "epoch": 21.0,
      "eval_bleu": 18.4603,
      "eval_gen_len": 16.7872,
      "eval_loss": 1.246074914932251,
      "eval_runtime": 15.4589,
      "eval_samples_per_second": 57.443,
      "eval_steps_per_second": 3.623,
      "step": 270522
    },
    {
      "epoch": 21.04,
      "learning_rate": 0.0001,
      "loss": 1.135,
      "step": 271000
    },
    {
      "epoch": 21.08,
      "learning_rate": 0.0001,
      "loss": 1.1393,
      "step": 271500
    },
    {
      "epoch": 21.11,
      "learning_rate": 0.0001,
      "loss": 1.134,
      "step": 272000
    },
    {
      "epoch": 21.15,
      "learning_rate": 0.0001,
      "loss": 1.1386,
      "step": 272500
    },
    {
      "epoch": 21.19,
      "learning_rate": 0.0001,
      "loss": 1.1391,
      "step": 273000
    },
    {
      "epoch": 21.23,
      "learning_rate": 0.0001,
      "loss": 1.1443,
      "step": 273500
    },
    {
      "epoch": 21.27,
      "learning_rate": 0.0001,
      "loss": 1.1424,
      "step": 274000
    },
    {
      "epoch": 21.31,
      "learning_rate": 0.0001,
      "loss": 1.1406,
      "step": 274500
    },
    {
      "epoch": 21.35,
      "learning_rate": 0.0001,
      "loss": 1.1465,
      "step": 275000
    },
    {
      "epoch": 21.39,
      "learning_rate": 0.0001,
      "loss": 1.1468,
      "step": 275500
    },
    {
      "epoch": 21.43,
      "learning_rate": 0.0001,
      "loss": 1.1541,
      "step": 276000
    },
    {
      "epoch": 21.46,
      "learning_rate": 0.0001,
      "loss": 1.1498,
      "step": 276500
    },
    {
      "epoch": 21.5,
      "learning_rate": 0.0001,
      "loss": 1.146,
      "step": 277000
    },
    {
      "epoch": 21.54,
      "learning_rate": 0.0001,
      "loss": 1.1417,
      "step": 277500
    },
    {
      "epoch": 21.58,
      "learning_rate": 0.0001,
      "loss": 1.1502,
      "step": 278000
    },
    {
      "epoch": 21.62,
      "learning_rate": 0.0001,
      "loss": 1.1581,
      "step": 278500
    },
    {
      "epoch": 21.66,
      "learning_rate": 0.0001,
      "loss": 1.1504,
      "step": 279000
    },
    {
      "epoch": 21.7,
      "learning_rate": 0.0001,
      "loss": 1.1599,
      "step": 279500
    },
    {
      "epoch": 21.74,
      "learning_rate": 0.0001,
      "loss": 1.1513,
      "step": 280000
    },
    {
      "epoch": 21.77,
      "learning_rate": 0.0001,
      "loss": 1.1591,
      "step": 280500
    },
    {
      "epoch": 21.81,
      "learning_rate": 0.0001,
      "loss": 1.1591,
      "step": 281000
    },
    {
      "epoch": 21.85,
      "learning_rate": 0.0001,
      "loss": 1.1582,
      "step": 281500
    },
    {
      "epoch": 21.89,
      "learning_rate": 0.0001,
      "loss": 1.1624,
      "step": 282000
    },
    {
      "epoch": 21.93,
      "learning_rate": 0.0001,
      "loss": 1.1493,
      "step": 282500
    },
    {
      "epoch": 21.97,
      "learning_rate": 0.0001,
      "loss": 1.1509,
      "step": 283000
    },
    {
      "epoch": 22.0,
      "eval_bleu": 18.0522,
      "eval_gen_len": 16.8097,
      "eval_loss": 1.2541053295135498,
      "eval_runtime": 17.6518,
      "eval_samples_per_second": 50.306,
      "eval_steps_per_second": 3.172,
      "step": 283404
    },
    {
      "epoch": 22.01,
      "learning_rate": 0.0001,
      "loss": 1.1512,
      "step": 283500
    },
    {
      "epoch": 22.05,
      "learning_rate": 0.0001,
      "loss": 1.1157,
      "step": 284000
    },
    {
      "epoch": 22.09,
      "learning_rate": 0.0001,
      "loss": 1.1264,
      "step": 284500
    },
    {
      "epoch": 22.12,
      "learning_rate": 0.0001,
      "loss": 1.1271,
      "step": 285000
    },
    {
      "epoch": 22.16,
      "learning_rate": 0.0001,
      "loss": 1.1294,
      "step": 285500
    },
    {
      "epoch": 22.2,
      "learning_rate": 0.0001,
      "loss": 1.1248,
      "step": 286000
    },
    {
      "epoch": 22.24,
      "learning_rate": 0.0001,
      "loss": 1.1311,
      "step": 286500
    },
    {
      "epoch": 22.28,
      "learning_rate": 0.0001,
      "loss": 1.1287,
      "step": 287000
    },
    {
      "epoch": 22.32,
      "learning_rate": 0.0001,
      "loss": 1.1249,
      "step": 287500
    },
    {
      "epoch": 22.36,
      "learning_rate": 0.0001,
      "loss": 1.1297,
      "step": 288000
    },
    {
      "epoch": 22.4,
      "learning_rate": 0.0001,
      "loss": 1.1308,
      "step": 288500
    },
    {
      "epoch": 22.43,
      "learning_rate": 0.0001,
      "loss": 1.1333,
      "step": 289000
    },
    {
      "epoch": 22.47,
      "learning_rate": 0.0001,
      "loss": 1.1229,
      "step": 289500
    },
    {
      "epoch": 22.51,
      "learning_rate": 0.0001,
      "loss": 1.1344,
      "step": 290000
    },
    {
      "epoch": 22.55,
      "learning_rate": 0.0001,
      "loss": 1.1564,
      "step": 290500
    },
    {
      "epoch": 22.59,
      "learning_rate": 0.0001,
      "loss": 1.1309,
      "step": 291000
    },
    {
      "epoch": 22.63,
      "learning_rate": 0.0001,
      "loss": 1.1418,
      "step": 291500
    },
    {
      "epoch": 22.67,
      "learning_rate": 0.0001,
      "loss": 1.1381,
      "step": 292000
    },
    {
      "epoch": 22.71,
      "learning_rate": 0.0001,
      "loss": 1.1315,
      "step": 292500
    },
    {
      "epoch": 22.74,
      "learning_rate": 0.0001,
      "loss": 1.1344,
      "step": 293000
    },
    {
      "epoch": 22.78,
      "learning_rate": 0.0001,
      "loss": 1.1384,
      "step": 293500
    },
    {
      "epoch": 22.82,
      "learning_rate": 0.0001,
      "loss": 1.14,
      "step": 294000
    },
    {
      "epoch": 22.86,
      "learning_rate": 0.0001,
      "loss": 1.1396,
      "step": 294500
    },
    {
      "epoch": 22.9,
      "learning_rate": 0.0001,
      "loss": 1.1428,
      "step": 295000
    },
    {
      "epoch": 22.94,
      "learning_rate": 0.0001,
      "loss": 1.1477,
      "step": 295500
    },
    {
      "epoch": 22.98,
      "learning_rate": 0.0001,
      "loss": 1.1361,
      "step": 296000
    },
    {
      "epoch": 23.0,
      "eval_bleu": 18.2043,
      "eval_gen_len": 16.8266,
      "eval_loss": 1.2469897270202637,
      "eval_runtime": 15.5569,
      "eval_samples_per_second": 57.081,
      "eval_steps_per_second": 3.6,
      "step": 296286
    },
    {
      "epoch": 23.02,
      "learning_rate": 0.0001,
      "loss": 1.1359,
      "step": 296500
    },
    {
      "epoch": 23.06,
      "learning_rate": 0.0001,
      "loss": 1.1016,
      "step": 297000
    },
    {
      "epoch": 23.09,
      "learning_rate": 0.0001,
      "loss": 1.1042,
      "step": 297500
    },
    {
      "epoch": 23.13,
      "learning_rate": 0.0001,
      "loss": 1.12,
      "step": 298000
    },
    {
      "epoch": 23.17,
      "learning_rate": 0.0001,
      "loss": 1.1116,
      "step": 298500
    },
    {
      "epoch": 23.21,
      "learning_rate": 0.0001,
      "loss": 1.1175,
      "step": 299000
    },
    {
      "epoch": 23.25,
      "learning_rate": 0.0001,
      "loss": 1.1042,
      "step": 299500
    },
    {
      "epoch": 23.29,
      "learning_rate": 0.0001,
      "loss": 1.112,
      "step": 300000
    },
    {
      "epoch": 23.33,
      "learning_rate": 0.0001,
      "loss": 1.1219,
      "step": 300500
    },
    {
      "epoch": 23.37,
      "learning_rate": 0.0001,
      "loss": 1.1264,
      "step": 301000
    },
    {
      "epoch": 23.4,
      "learning_rate": 0.0001,
      "loss": 1.105,
      "step": 301500
    },
    {
      "epoch": 23.44,
      "learning_rate": 0.0001,
      "loss": 1.1212,
      "step": 302000
    },
    {
      "epoch": 23.48,
      "learning_rate": 0.0001,
      "loss": 1.1232,
      "step": 302500
    },
    {
      "epoch": 23.52,
      "learning_rate": 0.0001,
      "loss": 1.1249,
      "step": 303000
    },
    {
      "epoch": 23.56,
      "learning_rate": 0.0001,
      "loss": 1.117,
      "step": 303500
    },
    {
      "epoch": 23.6,
      "learning_rate": 0.0001,
      "loss": 1.1332,
      "step": 304000
    },
    {
      "epoch": 23.64,
      "learning_rate": 0.0001,
      "loss": 1.123,
      "step": 304500
    },
    {
      "epoch": 23.68,
      "learning_rate": 0.0001,
      "loss": 1.1189,
      "step": 305000
    },
    {
      "epoch": 23.72,
      "learning_rate": 0.0001,
      "loss": 1.1307,
      "step": 305500
    },
    {
      "epoch": 23.75,
      "learning_rate": 0.0001,
      "loss": 1.134,
      "step": 306000
    },
    {
      "epoch": 23.79,
      "learning_rate": 0.0001,
      "loss": 1.1269,
      "step": 306500
    },
    {
      "epoch": 23.83,
      "learning_rate": 0.0001,
      "loss": 1.1268,
      "step": 307000
    },
    {
      "epoch": 23.87,
      "learning_rate": 0.0001,
      "loss": 1.1235,
      "step": 307500
    },
    {
      "epoch": 23.91,
      "learning_rate": 0.0001,
      "loss": 1.1232,
      "step": 308000
    },
    {
      "epoch": 23.95,
      "learning_rate": 0.0001,
      "loss": 1.1303,
      "step": 308500
    },
    {
      "epoch": 23.99,
      "learning_rate": 0.0001,
      "loss": 1.1336,
      "step": 309000
    },
    {
      "epoch": 24.0,
      "eval_bleu": 18.4811,
      "eval_gen_len": 16.8435,
      "eval_loss": 1.254769206047058,
      "eval_runtime": 15.6599,
      "eval_samples_per_second": 56.705,
      "eval_steps_per_second": 3.576,
      "step": 309168
    },
    {
      "epoch": 24.03,
      "learning_rate": 0.0001,
      "loss": 1.1055,
      "step": 309500
    },
    {
      "epoch": 24.06,
      "learning_rate": 0.0001,
      "loss": 1.0985,
      "step": 310000
    },
    {
      "epoch": 24.1,
      "learning_rate": 0.0001,
      "loss": 1.0886,
      "step": 310500
    },
    {
      "epoch": 24.14,
      "learning_rate": 0.0001,
      "loss": 1.0986,
      "step": 311000
    },
    {
      "epoch": 24.18,
      "learning_rate": 0.0001,
      "loss": 1.0899,
      "step": 311500
    },
    {
      "epoch": 24.22,
      "learning_rate": 0.0001,
      "loss": 1.0992,
      "step": 312000
    },
    {
      "epoch": 24.26,
      "learning_rate": 0.0001,
      "loss": 1.1086,
      "step": 312500
    },
    {
      "epoch": 24.3,
      "learning_rate": 0.0001,
      "loss": 1.1009,
      "step": 313000
    },
    {
      "epoch": 24.34,
      "learning_rate": 0.0001,
      "loss": 1.1071,
      "step": 313500
    },
    {
      "epoch": 24.38,
      "learning_rate": 0.0001,
      "loss": 1.0994,
      "step": 314000
    },
    {
      "epoch": 24.41,
      "learning_rate": 0.0001,
      "loss": 1.1143,
      "step": 314500
    },
    {
      "epoch": 24.45,
      "learning_rate": 0.0001,
      "loss": 1.1013,
      "step": 315000
    },
    {
      "epoch": 24.49,
      "learning_rate": 0.0001,
      "loss": 1.1087,
      "step": 315500
    },
    {
      "epoch": 24.53,
      "learning_rate": 0.0001,
      "loss": 1.1149,
      "step": 316000
    },
    {
      "epoch": 24.57,
      "learning_rate": 0.0001,
      "loss": 1.1037,
      "step": 316500
    },
    {
      "epoch": 24.61,
      "learning_rate": 0.0001,
      "loss": 1.1053,
      "step": 317000
    },
    {
      "epoch": 24.65,
      "learning_rate": 0.0001,
      "loss": 1.1215,
      "step": 317500
    },
    {
      "epoch": 24.69,
      "learning_rate": 0.0001,
      "loss": 1.1041,
      "step": 318000
    },
    {
      "epoch": 24.72,
      "learning_rate": 0.0001,
      "loss": 1.1131,
      "step": 318500
    },
    {
      "epoch": 24.76,
      "learning_rate": 0.0001,
      "loss": 1.1172,
      "step": 319000
    },
    {
      "epoch": 24.8,
      "learning_rate": 0.0001,
      "loss": 1.1131,
      "step": 319500
    },
    {
      "epoch": 24.84,
      "learning_rate": 0.0001,
      "loss": 1.1167,
      "step": 320000
    },
    {
      "epoch": 24.88,
      "learning_rate": 0.0001,
      "loss": 1.1263,
      "step": 320500
    },
    {
      "epoch": 24.92,
      "learning_rate": 0.0001,
      "loss": 1.1155,
      "step": 321000
    },
    {
      "epoch": 24.96,
      "learning_rate": 0.0001,
      "loss": 1.1248,
      "step": 321500
    },
    {
      "epoch": 25.0,
      "learning_rate": 0.0001,
      "loss": 1.1201,
      "step": 322000
    },
    {
      "epoch": 25.0,
      "eval_bleu": 18.3582,
      "eval_gen_len": 16.8322,
      "eval_loss": 1.2584398984909058,
      "eval_runtime": 14.3876,
      "eval_samples_per_second": 61.72,
      "eval_steps_per_second": 3.892,
      "step": 322050
    },
    {
      "epoch": 25.03,
      "learning_rate": 0.0001,
      "loss": 1.0814,
      "step": 322500
    },
    {
      "epoch": 25.07,
      "learning_rate": 0.0001,
      "loss": 1.0844,
      "step": 323000
    },
    {
      "epoch": 25.11,
      "learning_rate": 0.0001,
      "loss": 1.0897,
      "step": 323500
    },
    {
      "epoch": 25.15,
      "learning_rate": 0.0001,
      "loss": 1.0872,
      "step": 324000
    },
    {
      "epoch": 25.19,
      "learning_rate": 0.0001,
      "loss": 1.0947,
      "step": 324500
    },
    {
      "epoch": 25.23,
      "learning_rate": 0.0001,
      "loss": 1.0922,
      "step": 325000
    },
    {
      "epoch": 25.27,
      "learning_rate": 0.0001,
      "loss": 1.0899,
      "step": 325500
    },
    {
      "epoch": 25.31,
      "learning_rate": 0.0001,
      "loss": 1.0878,
      "step": 326000
    },
    {
      "epoch": 25.35,
      "learning_rate": 0.0001,
      "loss": 1.085,
      "step": 326500
    },
    {
      "epoch": 25.38,
      "learning_rate": 0.0001,
      "loss": 1.0874,
      "step": 327000
    },
    {
      "epoch": 25.42,
      "learning_rate": 0.0001,
      "loss": 1.0906,
      "step": 327500
    },
    {
      "epoch": 25.46,
      "learning_rate": 0.0001,
      "loss": 1.0846,
      "step": 328000
    },
    {
      "epoch": 25.5,
      "learning_rate": 0.0001,
      "loss": 1.1061,
      "step": 328500
    },
    {
      "epoch": 25.54,
      "learning_rate": 0.0001,
      "loss": 1.104,
      "step": 329000
    },
    {
      "epoch": 25.58,
      "learning_rate": 0.0001,
      "loss": 1.1034,
      "step": 329500
    },
    {
      "epoch": 25.62,
      "learning_rate": 0.0001,
      "loss": 1.1046,
      "step": 330000
    },
    {
      "epoch": 25.66,
      "learning_rate": 0.0001,
      "loss": 1.106,
      "step": 330500
    },
    {
      "epoch": 25.69,
      "learning_rate": 0.0001,
      "loss": 1.1009,
      "step": 331000
    },
    {
      "epoch": 25.73,
      "learning_rate": 0.0001,
      "loss": 1.0916,
      "step": 331500
    },
    {
      "epoch": 25.77,
      "learning_rate": 0.0001,
      "loss": 1.1006,
      "step": 332000
    },
    {
      "epoch": 25.81,
      "learning_rate": 0.0001,
      "loss": 1.1043,
      "step": 332500
    },
    {
      "epoch": 25.85,
      "learning_rate": 0.0001,
      "loss": 1.0963,
      "step": 333000
    },
    {
      "epoch": 25.89,
      "learning_rate": 0.0001,
      "loss": 1.0994,
      "step": 333500
    },
    {
      "epoch": 25.93,
      "learning_rate": 0.0001,
      "loss": 1.0964,
      "step": 334000
    },
    {
      "epoch": 25.97,
      "learning_rate": 0.0001,
      "loss": 1.1056,
      "step": 334500
    },
    {
      "epoch": 26.0,
      "eval_bleu": 18.3882,
      "eval_gen_len": 16.8176,
      "eval_loss": 1.255277395248413,
      "eval_runtime": 17.3981,
      "eval_samples_per_second": 51.04,
      "eval_steps_per_second": 3.219,
      "step": 334932
    },
    {
      "epoch": 26.01,
      "learning_rate": 0.0001,
      "loss": 1.0931,
      "step": 335000
    },
    {
      "epoch": 26.04,
      "learning_rate": 0.0001,
      "loss": 1.0651,
      "step": 335500
    },
    {
      "epoch": 26.08,
      "learning_rate": 0.0001,
      "loss": 1.069,
      "step": 336000
    },
    {
      "epoch": 26.12,
      "learning_rate": 0.0001,
      "loss": 1.0766,
      "step": 336500
    },
    {
      "epoch": 26.16,
      "learning_rate": 0.0001,
      "loss": 1.0728,
      "step": 337000
    },
    {
      "epoch": 26.2,
      "learning_rate": 0.0001,
      "loss": 1.0818,
      "step": 337500
    },
    {
      "epoch": 26.24,
      "learning_rate": 0.0001,
      "loss": 1.0702,
      "step": 338000
    },
    {
      "epoch": 26.28,
      "learning_rate": 0.0001,
      "loss": 1.0771,
      "step": 338500
    },
    {
      "epoch": 26.32,
      "learning_rate": 0.0001,
      "loss": 1.07,
      "step": 339000
    },
    {
      "epoch": 26.35,
      "learning_rate": 0.0001,
      "loss": 1.0824,
      "step": 339500
    },
    {
      "epoch": 26.39,
      "learning_rate": 0.0001,
      "loss": 1.0875,
      "step": 340000
    },
    {
      "epoch": 26.43,
      "learning_rate": 0.0001,
      "loss": 1.089,
      "step": 340500
    },
    {
      "epoch": 26.47,
      "learning_rate": 0.0001,
      "loss": 1.0858,
      "step": 341000
    },
    {
      "epoch": 26.51,
      "learning_rate": 0.0001,
      "loss": 1.0919,
      "step": 341500
    },
    {
      "epoch": 26.55,
      "learning_rate": 0.0001,
      "loss": 1.0851,
      "step": 342000
    },
    {
      "epoch": 26.59,
      "learning_rate": 0.0001,
      "loss": 1.0878,
      "step": 342500
    },
    {
      "epoch": 26.63,
      "learning_rate": 0.0001,
      "loss": 1.0749,
      "step": 343000
    },
    {
      "epoch": 26.67,
      "learning_rate": 0.0001,
      "loss": 1.0765,
      "step": 343500
    },
    {
      "epoch": 26.7,
      "learning_rate": 0.0001,
      "loss": 1.0961,
      "step": 344000
    },
    {
      "epoch": 26.74,
      "learning_rate": 0.0001,
      "loss": 1.0894,
      "step": 344500
    },
    {
      "epoch": 26.78,
      "learning_rate": 0.0001,
      "loss": 1.0909,
      "step": 345000
    },
    {
      "epoch": 26.82,
      "learning_rate": 0.0001,
      "loss": 1.0929,
      "step": 345500
    },
    {
      "epoch": 26.86,
      "learning_rate": 0.0001,
      "loss": 1.0927,
      "step": 346000
    },
    {
      "epoch": 26.9,
      "learning_rate": 0.0001,
      "loss": 1.0996,
      "step": 346500
    },
    {
      "epoch": 26.94,
      "learning_rate": 0.0001,
      "loss": 1.0904,
      "step": 347000
    },
    {
      "epoch": 26.98,
      "learning_rate": 0.0001,
      "loss": 1.0872,
      "step": 347500
    },
    {
      "epoch": 27.0,
      "eval_bleu": 18.4254,
      "eval_gen_len": 16.8964,
      "eval_loss": 1.2588080167770386,
      "eval_runtime": 15.7824,
      "eval_samples_per_second": 56.265,
      "eval_steps_per_second": 3.548,
      "step": 347814
    },
    {
      "epoch": 27.01,
      "learning_rate": 0.0001,
      "loss": 1.0759,
      "step": 348000
    },
    {
      "epoch": 27.05,
      "learning_rate": 0.0001,
      "loss": 1.0545,
      "step": 348500
    },
    {
      "epoch": 27.09,
      "learning_rate": 0.0001,
      "loss": 1.0614,
      "step": 349000
    },
    {
      "epoch": 27.13,
      "learning_rate": 0.0001,
      "loss": 1.0769,
      "step": 349500
    },
    {
      "epoch": 27.17,
      "learning_rate": 0.0001,
      "loss": 1.0558,
      "step": 350000
    },
    {
      "epoch": 27.21,
      "learning_rate": 0.0001,
      "loss": 1.0661,
      "step": 350500
    },
    {
      "epoch": 27.25,
      "learning_rate": 0.0001,
      "loss": 1.0698,
      "step": 351000
    },
    {
      "epoch": 27.29,
      "learning_rate": 0.0001,
      "loss": 1.0634,
      "step": 351500
    },
    {
      "epoch": 27.32,
      "learning_rate": 0.0001,
      "loss": 1.0796,
      "step": 352000
    },
    {
      "epoch": 27.36,
      "learning_rate": 0.0001,
      "loss": 1.0529,
      "step": 352500
    },
    {
      "epoch": 27.4,
      "learning_rate": 0.0001,
      "loss": 1.0817,
      "step": 353000
    },
    {
      "epoch": 27.44,
      "learning_rate": 0.0001,
      "loss": 1.0745,
      "step": 353500
    },
    {
      "epoch": 27.48,
      "learning_rate": 0.0001,
      "loss": 1.0827,
      "step": 354000
    },
    {
      "epoch": 27.52,
      "learning_rate": 0.0001,
      "loss": 1.0686,
      "step": 354500
    },
    {
      "epoch": 27.56,
      "learning_rate": 0.0001,
      "loss": 1.0733,
      "step": 355000
    },
    {
      "epoch": 27.6,
      "learning_rate": 0.0001,
      "loss": 1.0759,
      "step": 355500
    },
    {
      "epoch": 27.64,
      "learning_rate": 0.0001,
      "loss": 1.0803,
      "step": 356000
    },
    {
      "epoch": 27.67,
      "learning_rate": 0.0001,
      "loss": 1.0632,
      "step": 356500
    },
    {
      "epoch": 27.71,
      "learning_rate": 0.0001,
      "loss": 1.0755,
      "step": 357000
    },
    {
      "epoch": 27.75,
      "learning_rate": 0.0001,
      "loss": 1.0736,
      "step": 357500
    },
    {
      "epoch": 27.79,
      "learning_rate": 0.0001,
      "loss": 1.0694,
      "step": 358000
    },
    {
      "epoch": 27.83,
      "learning_rate": 0.0001,
      "loss": 1.0845,
      "step": 358500
    },
    {
      "epoch": 27.87,
      "learning_rate": 0.0001,
      "loss": 1.0786,
      "step": 359000
    },
    {
      "epoch": 27.91,
      "learning_rate": 0.0001,
      "loss": 1.0789,
      "step": 359500
    },
    {
      "epoch": 27.95,
      "learning_rate": 0.0001,
      "loss": 1.0778,
      "step": 360000
    },
    {
      "epoch": 27.98,
      "learning_rate": 0.0001,
      "loss": 1.0773,
      "step": 360500
    },
    {
      "epoch": 28.0,
      "eval_bleu": 18.1015,
      "eval_gen_len": 16.8232,
      "eval_loss": 1.2550303936004639,
      "eval_runtime": 14.6294,
      "eval_samples_per_second": 60.7,
      "eval_steps_per_second": 3.828,
      "step": 360696
    },
    {
      "epoch": 28.02,
      "learning_rate": 0.0001,
      "loss": 1.062,
      "step": 361000
    },
    {
      "epoch": 28.06,
      "learning_rate": 0.0001,
      "loss": 1.0475,
      "step": 361500
    },
    {
      "epoch": 28.1,
      "learning_rate": 0.0001,
      "loss": 1.0451,
      "step": 362000
    },
    {
      "epoch": 28.14,
      "learning_rate": 0.0001,
      "loss": 1.0556,
      "step": 362500
    },
    {
      "epoch": 28.18,
      "learning_rate": 0.0001,
      "loss": 1.0592,
      "step": 363000
    },
    {
      "epoch": 28.22,
      "learning_rate": 0.0001,
      "loss": 1.036,
      "step": 363500
    },
    {
      "epoch": 28.26,
      "learning_rate": 0.0001,
      "loss": 1.0604,
      "step": 364000
    },
    {
      "epoch": 28.3,
      "learning_rate": 0.0001,
      "loss": 1.0486,
      "step": 364500
    },
    {
      "epoch": 28.33,
      "learning_rate": 0.0001,
      "loss": 1.044,
      "step": 365000
    },
    {
      "epoch": 28.37,
      "learning_rate": 0.0001,
      "loss": 1.0577,
      "step": 365500
    },
    {
      "epoch": 28.41,
      "learning_rate": 0.0001,
      "loss": 1.0607,
      "step": 366000
    },
    {
      "epoch": 28.45,
      "learning_rate": 0.0001,
      "loss": 1.0545,
      "step": 366500
    },
    {
      "epoch": 28.49,
      "learning_rate": 0.0001,
      "loss": 1.0699,
      "step": 367000
    },
    {
      "epoch": 28.53,
      "learning_rate": 0.0001,
      "loss": 1.0605,
      "step": 367500
    },
    {
      "epoch": 28.57,
      "learning_rate": 0.0001,
      "loss": 1.0682,
      "step": 368000
    },
    {
      "epoch": 28.61,
      "learning_rate": 0.0001,
      "loss": 1.0586,
      "step": 368500
    },
    {
      "epoch": 28.64,
      "learning_rate": 0.0001,
      "loss": 1.0658,
      "step": 369000
    },
    {
      "epoch": 28.68,
      "learning_rate": 0.0001,
      "loss": 1.0661,
      "step": 369500
    },
    {
      "epoch": 28.72,
      "learning_rate": 0.0001,
      "loss": 1.0637,
      "step": 370000
    },
    {
      "epoch": 28.76,
      "learning_rate": 0.0001,
      "loss": 1.068,
      "step": 370500
    },
    {
      "epoch": 28.8,
      "learning_rate": 0.0001,
      "loss": 1.0626,
      "step": 371000
    },
    {
      "epoch": 28.84,
      "learning_rate": 0.0001,
      "loss": 1.0647,
      "step": 371500
    },
    {
      "epoch": 28.88,
      "learning_rate": 0.0001,
      "loss": 1.0692,
      "step": 372000
    },
    {
      "epoch": 28.92,
      "learning_rate": 0.0001,
      "loss": 1.0622,
      "step": 372500
    },
    {
      "epoch": 28.96,
      "learning_rate": 0.0001,
      "loss": 1.0624,
      "step": 373000
    },
    {
      "epoch": 28.99,
      "learning_rate": 0.0001,
      "loss": 1.0761,
      "step": 373500
    },
    {
      "epoch": 29.0,
      "eval_bleu": 18.3819,
      "eval_gen_len": 16.8086,
      "eval_loss": 1.2643870115280151,
      "eval_runtime": 15.4548,
      "eval_samples_per_second": 57.458,
      "eval_steps_per_second": 3.623,
      "step": 373578
    },
    {
      "epoch": 29.03,
      "learning_rate": 0.0001,
      "loss": 1.0365,
      "step": 374000
    },
    {
      "epoch": 29.07,
      "learning_rate": 0.0001,
      "loss": 1.0438,
      "step": 374500
    },
    {
      "epoch": 29.11,
      "learning_rate": 0.0001,
      "loss": 1.0354,
      "step": 375000
    },
    {
      "epoch": 29.15,
      "learning_rate": 0.0001,
      "loss": 1.0416,
      "step": 375500
    },
    {
      "epoch": 29.19,
      "learning_rate": 0.0001,
      "loss": 1.0458,
      "step": 376000
    },
    {
      "epoch": 29.23,
      "learning_rate": 0.0001,
      "loss": 1.0305,
      "step": 376500
    },
    {
      "epoch": 29.27,
      "learning_rate": 0.0001,
      "loss": 1.0517,
      "step": 377000
    },
    {
      "epoch": 29.3,
      "learning_rate": 0.0001,
      "loss": 1.0425,
      "step": 377500
    },
    {
      "epoch": 29.34,
      "learning_rate": 0.0001,
      "loss": 1.0467,
      "step": 378000
    },
    {
      "epoch": 29.38,
      "learning_rate": 0.0001,
      "loss": 1.033,
      "step": 378500
    },
    {
      "epoch": 29.42,
      "learning_rate": 0.0001,
      "loss": 1.0521,
      "step": 379000
    },
    {
      "epoch": 29.46,
      "learning_rate": 0.0001,
      "loss": 1.049,
      "step": 379500
    },
    {
      "epoch": 29.5,
      "learning_rate": 0.0001,
      "loss": 1.0287,
      "step": 380000
    },
    {
      "epoch": 29.54,
      "learning_rate": 0.0001,
      "loss": 1.0495,
      "step": 380500
    },
    {
      "epoch": 29.58,
      "learning_rate": 0.0001,
      "loss": 1.0427,
      "step": 381000
    },
    {
      "epoch": 29.61,
      "learning_rate": 0.0001,
      "loss": 1.0641,
      "step": 381500
    },
    {
      "epoch": 29.65,
      "learning_rate": 0.0001,
      "loss": 1.06,
      "step": 382000
    },
    {
      "epoch": 29.69,
      "learning_rate": 0.0001,
      "loss": 1.0451,
      "step": 382500
    },
    {
      "epoch": 29.73,
      "learning_rate": 0.0001,
      "loss": 1.055,
      "step": 383000
    },
    {
      "epoch": 29.77,
      "learning_rate": 0.0001,
      "loss": 1.0683,
      "step": 383500
    },
    {
      "epoch": 29.81,
      "learning_rate": 0.0001,
      "loss": 1.0496,
      "step": 384000
    },
    {
      "epoch": 29.85,
      "learning_rate": 0.0001,
      "loss": 1.0716,
      "step": 384500
    },
    {
      "epoch": 29.89,
      "learning_rate": 0.0001,
      "loss": 1.0628,
      "step": 385000
    },
    {
      "epoch": 29.93,
      "learning_rate": 0.0001,
      "loss": 1.0613,
      "step": 385500
    },
    {
      "epoch": 29.96,
      "learning_rate": 0.0001,
      "loss": 1.0619,
      "step": 386000
    },
    {
      "epoch": 30.0,
      "eval_bleu": 18.3628,
      "eval_gen_len": 16.8153,
      "eval_loss": 1.2656418085098267,
      "eval_runtime": 15.5389,
      "eval_samples_per_second": 57.147,
      "eval_steps_per_second": 3.604,
      "step": 386460
    },
    {
      "epoch": 30.0,
      "learning_rate": 0.0001,
      "loss": 1.0543,
      "step": 386500
    },
    {
      "epoch": 30.04,
      "learning_rate": 0.0001,
      "loss": 1.0146,
      "step": 387000
    },
    {
      "epoch": 30.08,
      "learning_rate": 0.0001,
      "loss": 1.0257,
      "step": 387500
    },
    {
      "epoch": 30.12,
      "learning_rate": 0.0001,
      "loss": 1.0338,
      "step": 388000
    },
    {
      "epoch": 30.16,
      "learning_rate": 0.0001,
      "loss": 1.0244,
      "step": 388500
    },
    {
      "epoch": 30.2,
      "learning_rate": 0.0001,
      "loss": 1.03,
      "step": 389000
    },
    {
      "epoch": 30.24,
      "learning_rate": 0.0001,
      "loss": 1.036,
      "step": 389500
    },
    {
      "epoch": 30.27,
      "learning_rate": 0.0001,
      "loss": 1.0357,
      "step": 390000
    },
    {
      "epoch": 30.31,
      "learning_rate": 0.0001,
      "loss": 1.0313,
      "step": 390500
    },
    {
      "epoch": 30.35,
      "learning_rate": 0.0001,
      "loss": 1.0437,
      "step": 391000
    },
    {
      "epoch": 30.39,
      "learning_rate": 0.0001,
      "loss": 1.0414,
      "step": 391500
    },
    {
      "epoch": 30.43,
      "learning_rate": 0.0001,
      "loss": 1.0272,
      "step": 392000
    },
    {
      "epoch": 30.47,
      "learning_rate": 0.0001,
      "loss": 1.0287,
      "step": 392500
    },
    {
      "epoch": 30.51,
      "learning_rate": 0.0001,
      "loss": 1.0362,
      "step": 393000
    },
    {
      "epoch": 30.55,
      "learning_rate": 0.0001,
      "loss": 1.0424,
      "step": 393500
    },
    {
      "epoch": 30.59,
      "learning_rate": 0.0001,
      "loss": 1.0336,
      "step": 394000
    },
    {
      "epoch": 30.62,
      "learning_rate": 0.0001,
      "loss": 1.0393,
      "step": 394500
    },
    {
      "epoch": 30.66,
      "learning_rate": 0.0001,
      "loss": 1.0462,
      "step": 395000
    },
    {
      "epoch": 30.7,
      "learning_rate": 0.0001,
      "loss": 1.0415,
      "step": 395500
    },
    {
      "epoch": 30.74,
      "learning_rate": 0.0001,
      "loss": 1.0397,
      "step": 396000
    },
    {
      "epoch": 30.78,
      "learning_rate": 0.0001,
      "loss": 1.0573,
      "step": 396500
    },
    {
      "epoch": 30.82,
      "learning_rate": 0.0001,
      "loss": 1.0425,
      "step": 397000
    },
    {
      "epoch": 30.86,
      "learning_rate": 0.0001,
      "loss": 1.039,
      "step": 397500
    },
    {
      "epoch": 30.9,
      "learning_rate": 0.0001,
      "loss": 1.0518,
      "step": 398000
    },
    {
      "epoch": 30.93,
      "learning_rate": 0.0001,
      "loss": 1.0485,
      "step": 398500
    },
    {
      "epoch": 30.97,
      "learning_rate": 0.0001,
      "loss": 1.0515,
      "step": 399000
    },
    {
      "epoch": 31.0,
      "eval_bleu": 18.321,
      "eval_gen_len": 16.8018,
      "eval_loss": 1.2628525495529175,
      "eval_runtime": 14.6973,
      "eval_samples_per_second": 60.419,
      "eval_steps_per_second": 3.81,
      "step": 399342
    },
    {
      "epoch": 31.01,
      "learning_rate": 0.0001,
      "loss": 1.0386,
      "step": 399500
    },
    {
      "epoch": 31.05,
      "learning_rate": 0.0001,
      "loss": 1.0253,
      "step": 400000
    },
    {
      "epoch": 31.09,
      "learning_rate": 0.0001,
      "loss": 1.0119,
      "step": 400500
    },
    {
      "epoch": 31.13,
      "learning_rate": 0.0001,
      "loss": 1.0145,
      "step": 401000
    },
    {
      "epoch": 31.17,
      "learning_rate": 0.0001,
      "loss": 1.0214,
      "step": 401500
    },
    {
      "epoch": 31.21,
      "learning_rate": 0.0001,
      "loss": 1.0208,
      "step": 402000
    },
    {
      "epoch": 31.25,
      "learning_rate": 0.0001,
      "loss": 1.017,
      "step": 402500
    },
    {
      "epoch": 31.28,
      "learning_rate": 0.0001,
      "loss": 1.0196,
      "step": 403000
    },
    {
      "epoch": 31.32,
      "learning_rate": 0.0001,
      "loss": 1.0266,
      "step": 403500
    },
    {
      "epoch": 31.36,
      "learning_rate": 0.0001,
      "loss": 1.0236,
      "step": 404000
    },
    {
      "epoch": 31.4,
      "learning_rate": 0.0001,
      "loss": 1.0264,
      "step": 404500
    },
    {
      "epoch": 31.44,
      "learning_rate": 0.0001,
      "loss": 1.0242,
      "step": 405000
    },
    {
      "epoch": 31.48,
      "learning_rate": 0.0001,
      "loss": 1.0223,
      "step": 405500
    },
    {
      "epoch": 31.52,
      "learning_rate": 0.0001,
      "loss": 1.0362,
      "step": 406000
    },
    {
      "epoch": 31.56,
      "learning_rate": 0.0001,
      "loss": 1.0159,
      "step": 406500
    },
    {
      "epoch": 31.59,
      "learning_rate": 0.0001,
      "loss": 1.0264,
      "step": 407000
    },
    {
      "epoch": 31.63,
      "learning_rate": 0.0001,
      "loss": 1.031,
      "step": 407500
    },
    {
      "epoch": 31.67,
      "learning_rate": 0.0001,
      "loss": 1.0355,
      "step": 408000
    },
    {
      "epoch": 31.71,
      "learning_rate": 0.0001,
      "loss": 1.0267,
      "step": 408500
    },
    {
      "epoch": 31.75,
      "learning_rate": 0.0001,
      "loss": 1.0309,
      "step": 409000
    },
    {
      "epoch": 31.79,
      "learning_rate": 0.0001,
      "loss": 1.0416,
      "step": 409500
    },
    {
      "epoch": 31.83,
      "learning_rate": 0.0001,
      "loss": 1.0392,
      "step": 410000
    },
    {
      "epoch": 31.87,
      "learning_rate": 0.0001,
      "loss": 1.0469,
      "step": 410500
    },
    {
      "epoch": 31.9,
      "learning_rate": 0.0001,
      "loss": 1.0465,
      "step": 411000
    },
    {
      "epoch": 31.94,
      "learning_rate": 0.0001,
      "loss": 1.0382,
      "step": 411500
    },
    {
      "epoch": 31.98,
      "learning_rate": 0.0001,
      "loss": 1.0377,
      "step": 412000
    },
    {
      "epoch": 32.0,
      "eval_bleu": 18.3337,
      "eval_gen_len": 16.8637,
      "eval_loss": 1.2663763761520386,
      "eval_runtime": 15.4773,
      "eval_samples_per_second": 57.374,
      "eval_steps_per_second": 3.618,
      "step": 412224
    }
  ],
  "max_steps": 644100,
  "num_train_epochs": 50,
  "total_flos": 5.072492765333422e+17,
  "trial_name": null,
  "trial_params": null
}