{
  "best_metric": 16.4534,
  "best_model_checkpoint": "ckpt_mt5/google/mt5-large/wmt14_de_en/lr0.0001_e4/checkpoint-430000",
  "epoch": 1.5326011355571327,
  "global_step": 430000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 0.0001,
      "loss": 4.2692,
      "step": 500
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0001,
      "loss": 2.2422,
      "step": 1000
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0001,
      "loss": 2.1497,
      "step": 1500
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0001,
      "loss": 2.1255,
      "step": 2000
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0001,
      "loss": 2.0609,
      "step": 2500
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0001,
      "loss": 2.0446,
      "step": 3000
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0001,
      "loss": 2.0455,
      "step": 3500
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0001,
      "loss": 2.0007,
      "step": 4000
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0001,
      "loss": 1.9637,
      "step": 4500
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0001,
      "loss": 1.9725,
      "step": 5000
    },
    {
      "epoch": 0.02,
      "eval_bleu": 13.3952,
      "eval_gen_len": 17.202,
      "eval_loss": 1.3810369968414307,
      "eval_runtime": 178.0846,
      "eval_samples_per_second": 16.846,
      "eval_steps_per_second": 1.056,
      "step": 5000
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0001,
      "loss": 1.9656,
      "step": 5500
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0001,
      "loss": 1.9423,
      "step": 6000
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0001,
      "loss": 1.9296,
      "step": 6500
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0001,
      "loss": 1.9221,
      "step": 7000
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0001,
      "loss": 1.9079,
      "step": 7500
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0001,
      "loss": 1.8868,
      "step": 8000
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0001,
      "loss": 1.878,
      "step": 8500
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0001,
      "loss": 1.8997,
      "step": 9000
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0001,
      "loss": 1.8681,
      "step": 9500
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0001,
      "loss": 1.8882,
      "step": 10000
    },
    {
      "epoch": 0.04,
      "eval_bleu": 14.0607,
      "eval_gen_len": 17.2097,
      "eval_loss": 1.3503270149230957,
      "eval_runtime": 199.3786,
      "eval_samples_per_second": 15.047,
      "eval_steps_per_second": 0.943,
      "step": 10000
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0001,
      "loss": 1.8884,
      "step": 10500
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0001,
      "loss": 1.8282,
      "step": 11000
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0001,
      "loss": 1.8765,
      "step": 11500
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0001,
      "loss": 1.8361,
      "step": 12000
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0001,
      "loss": 1.8474,
      "step": 12500
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0001,
      "loss": 1.8516,
      "step": 13000
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0001,
      "loss": 1.8209,
      "step": 13500
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0001,
      "loss": 1.8265,
      "step": 14000
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0001,
      "loss": 1.8116,
      "step": 14500
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0001,
      "loss": 1.829,
      "step": 15000
    },
    {
      "epoch": 0.05,
      "eval_bleu": 14.165,
      "eval_gen_len": 17.2263,
      "eval_loss": 1.3259068727493286,
      "eval_runtime": 180.0944,
      "eval_samples_per_second": 16.658,
      "eval_steps_per_second": 1.044,
      "step": 15000
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0001,
      "loss": 1.8267,
      "step": 15500
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0001,
      "loss": 1.8083,
      "step": 16000
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0001,
      "loss": 1.8099,
      "step": 16500
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0001,
      "loss": 1.7982,
      "step": 17000
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0001,
      "loss": 1.8108,
      "step": 17500
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0001,
      "loss": 1.8229,
      "step": 18000
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0001,
      "loss": 1.7963,
      "step": 18500
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0001,
      "loss": 1.8219,
      "step": 19000
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0001,
      "loss": 1.7778,
      "step": 19500
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0001,
      "loss": 1.7674,
      "step": 20000
    },
    {
      "epoch": 0.07,
      "eval_bleu": 14.7226,
      "eval_gen_len": 17.241,
      "eval_loss": 1.31859290599823,
      "eval_runtime": 177.8933,
      "eval_samples_per_second": 16.864,
      "eval_steps_per_second": 1.057,
      "step": 20000
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0001,
      "loss": 1.758,
      "step": 20500
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0001,
      "loss": 1.7783,
      "step": 21000
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0001,
      "loss": 1.7843,
      "step": 21500
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0001,
      "loss": 1.7796,
      "step": 22000
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0001,
      "loss": 1.7659,
      "step": 22500
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0001,
      "loss": 1.7389,
      "step": 23000
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0001,
      "loss": 1.7551,
      "step": 23500
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0001,
      "loss": 1.7697,
      "step": 24000
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0001,
      "loss": 1.7384,
      "step": 24500
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0001,
      "loss": 1.7522,
      "step": 25000
    },
    {
      "epoch": 0.09,
      "eval_bleu": 14.7929,
      "eval_gen_len": 17.2517,
      "eval_loss": 1.2968846559524536,
      "eval_runtime": 180.2133,
      "eval_samples_per_second": 16.647,
      "eval_steps_per_second": 1.043,
      "step": 25000
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0001,
      "loss": 1.7378,
      "step": 25500
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0001,
      "loss": 1.726,
      "step": 26000
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0001,
      "loss": 1.7374,
      "step": 26500
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0001,
      "loss": 1.7317,
      "step": 27000
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0001,
      "loss": 1.7411,
      "step": 27500
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0001,
      "loss": 1.733,
      "step": 28000
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0001,
      "loss": 1.7171,
      "step": 28500
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0001,
      "loss": 1.753,
      "step": 29000
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0001,
      "loss": 1.744,
      "step": 29500
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0001,
      "loss": 1.7527,
      "step": 30000
    },
    {
      "epoch": 0.11,
      "eval_bleu": 15.029,
      "eval_gen_len": 17.241,
      "eval_loss": 1.2927813529968262,
      "eval_runtime": 179.3354,
      "eval_samples_per_second": 16.728,
      "eval_steps_per_second": 1.048,
      "step": 30000
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0001,
      "loss": 1.7275,
      "step": 30500
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0001,
      "loss": 1.7347,
      "step": 31000
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0001,
      "loss": 1.7252,
      "step": 31500
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0001,
      "loss": 1.6989,
      "step": 32000
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0001,
      "loss": 1.7232,
      "step": 32500
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0001,
      "loss": 1.7156,
      "step": 33000
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0001,
      "loss": 1.7356,
      "step": 33500
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0001,
      "loss": 1.7046,
      "step": 34000
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0001,
      "loss": 1.7046,
      "step": 34500
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0001,
      "loss": 1.7087,
      "step": 35000
    },
    {
      "epoch": 0.12,
      "eval_bleu": 15.0278,
      "eval_gen_len": 17.2407,
      "eval_loss": 1.2754088640213013,
      "eval_runtime": 178.3942,
      "eval_samples_per_second": 16.817,
      "eval_steps_per_second": 1.054,
      "step": 35000
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0001,
      "loss": 1.7322,
      "step": 35500
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0001,
      "loss": 1.7191,
      "step": 36000
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0001,
      "loss": 1.7086,
      "step": 36500
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0001,
      "loss": 1.6837,
      "step": 37000
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0001,
      "loss": 1.7175,
      "step": 37500
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0001,
      "loss": 1.7335,
      "step": 38000
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0001,
      "loss": 1.6954,
      "step": 38500
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0001,
      "loss": 1.7104,
      "step": 39000
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0001,
      "loss": 1.6984,
      "step": 39500
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0001,
      "loss": 1.7062,
      "step": 40000
    },
    {
      "epoch": 0.14,
      "eval_bleu": 15.1151,
      "eval_gen_len": 17.2407,
      "eval_loss": 1.2676228284835815,
      "eval_runtime": 179.8087,
      "eval_samples_per_second": 16.684,
      "eval_steps_per_second": 1.046,
      "step": 40000
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0001,
      "loss": 1.6969,
      "step": 40500
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0001,
      "loss": 1.6802,
      "step": 41000
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0001,
      "loss": 1.6884,
      "step": 41500
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0001,
      "loss": 1.6812,
      "step": 42000
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0001,
      "loss": 1.6566,
      "step": 42500
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0001,
      "loss": 1.6617,
      "step": 43000
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0001,
      "loss": 1.6766,
      "step": 43500
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0001,
      "loss": 1.6937,
      "step": 44000
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0001,
      "loss": 1.6992,
      "step": 44500
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0001,
      "loss": 1.6676,
      "step": 45000
    },
    {
      "epoch": 0.16,
      "eval_bleu": 15.1803,
      "eval_gen_len": 17.249,
      "eval_loss": 1.2574199438095093,
      "eval_runtime": 179.3005,
      "eval_samples_per_second": 16.732,
      "eval_steps_per_second": 1.049,
      "step": 45000
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0001,
      "loss": 1.7226,
      "step": 45500
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0001,
      "loss": 1.684,
      "step": 46000
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0001,
      "loss": 1.6845,
      "step": 46500
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0001,
      "loss": 1.6816,
      "step": 47000
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0001,
      "loss": 1.685,
      "step": 47500
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0001,
      "loss": 1.6906,
      "step": 48000
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0001,
      "loss": 1.6751,
      "step": 48500
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0001,
      "loss": 1.6744,
      "step": 49000
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0001,
      "loss": 1.6454,
      "step": 49500
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0001,
      "loss": 1.6738,
      "step": 50000
    },
    {
      "epoch": 0.18,
      "eval_bleu": 14.9842,
      "eval_gen_len": 17.2427,
      "eval_loss": 1.2550103664398193,
      "eval_runtime": 180.4381,
      "eval_samples_per_second": 16.626,
      "eval_steps_per_second": 1.042,
      "step": 50000
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0001,
      "loss": 1.6721,
      "step": 50500
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0001,
      "loss": 1.6572,
      "step": 51000
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0001,
      "loss": 1.6674,
      "step": 51500
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0001,
      "loss": 1.6598,
      "step": 52000
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0001,
      "loss": 1.678,
      "step": 52500
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0001,
      "loss": 1.6769,
      "step": 53000
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0001,
      "loss": 1.6684,
      "step": 53500
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0001,
      "loss": 1.6537,
      "step": 54000
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0001,
      "loss": 1.6466,
      "step": 54500
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0001,
      "loss": 1.6523,
      "step": 55000
    },
    {
      "epoch": 0.2,
      "eval_bleu": 15.1756,
      "eval_gen_len": 17.2313,
      "eval_loss": 1.2486598491668701,
      "eval_runtime": 179.8049,
      "eval_samples_per_second": 16.685,
      "eval_steps_per_second": 1.046,
      "step": 55000
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0001,
      "loss": 1.674,
      "step": 55500
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0001,
      "loss": 1.6552,
      "step": 56000
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0001,
      "loss": 1.6744,
      "step": 56500
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0001,
      "loss": 1.6871,
      "step": 57000
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0001,
      "loss": 1.6472,
      "step": 57500
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0001,
      "loss": 1.6515,
      "step": 58000
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0001,
      "loss": 1.6661,
      "step": 58500
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0001,
      "loss": 1.6553,
      "step": 59000
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0001,
      "loss": 1.6395,
      "step": 59500
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0001,
      "loss": 1.6291,
      "step": 60000
    },
    {
      "epoch": 0.21,
      "eval_bleu": 15.0648,
      "eval_gen_len": 17.219,
      "eval_loss": 1.244527816772461,
      "eval_runtime": 207.2334,
      "eval_samples_per_second": 14.476,
      "eval_steps_per_second": 0.907,
      "step": 60000
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0001,
      "loss": 1.6343,
      "step": 60500
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0001,
      "loss": 1.6572,
      "step": 61000
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0001,
      "loss": 1.6294,
      "step": 61500
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0001,
      "loss": 1.6153,
      "step": 62000
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0001,
      "loss": 1.6162,
      "step": 62500
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0001,
      "loss": 1.6246,
      "step": 63000
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0001,
      "loss": 1.6182,
      "step": 63500
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0001,
      "loss": 1.6602,
      "step": 64000
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0001,
      "loss": 1.6453,
      "step": 64500
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0001,
      "loss": 1.6385,
      "step": 65000
    },
    {
      "epoch": 0.23,
      "eval_bleu": 15.5441,
      "eval_gen_len": 17.244,
      "eval_loss": 1.241575002670288,
      "eval_runtime": 180.0227,
      "eval_samples_per_second": 16.665,
      "eval_steps_per_second": 1.044,
      "step": 65000
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0001,
      "loss": 1.6316,
      "step": 65500
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0001,
      "loss": 1.6446,
      "step": 66000
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0001,
      "loss": 1.6274,
      "step": 66500
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0001,
      "loss": 1.6157,
      "step": 67000
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0001,
      "loss": 1.6382,
      "step": 67500
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0001,
      "loss": 1.6289,
      "step": 68000
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0001,
      "loss": 1.617,
      "step": 68500
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0001,
      "loss": 1.636,
      "step": 69000
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0001,
      "loss": 1.6263,
      "step": 69500
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0001,
      "loss": 1.6194,
      "step": 70000
    },
    {
      "epoch": 0.25,
      "eval_bleu": 15.2435,
      "eval_gen_len": 17.259,
      "eval_loss": 1.2337630987167358,
      "eval_runtime": 180.4671,
      "eval_samples_per_second": 16.624,
      "eval_steps_per_second": 1.042,
      "step": 70000
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0001,
      "loss": 1.6384,
      "step": 70500
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0001,
      "loss": 1.6302,
      "step": 71000
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0001,
      "loss": 1.622,
      "step": 71500
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0001,
      "loss": 1.6206,
      "step": 72000
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0001,
      "loss": 1.6328,
      "step": 72500
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0001,
      "loss": 1.6376,
      "step": 73000
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0001,
      "loss": 1.6289,
      "step": 73500
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0001,
      "loss": 1.6021,
      "step": 74000
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0001,
      "loss": 1.6124,
      "step": 74500
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0001,
      "loss": 1.6141,
      "step": 75000
    },
    {
      "epoch": 0.27,
      "eval_bleu": 15.4302,
      "eval_gen_len": 17.2277,
      "eval_loss": 1.2314013242721558,
      "eval_runtime": 180.135,
      "eval_samples_per_second": 16.654,
      "eval_steps_per_second": 1.044,
      "step": 75000
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0001,
      "loss": 1.6149,
      "step": 75500
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0001,
      "loss": 1.6188,
      "step": 76000
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0001,
      "loss": 1.6111,
      "step": 76500
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0001,
      "loss": 1.633,
      "step": 77000
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0001,
      "loss": 1.6185,
      "step": 77500
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0001,
      "loss": 1.6207,
      "step": 78000
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0001,
      "loss": 1.6149,
      "step": 78500
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0001,
      "loss": 1.6201,
      "step": 79000
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0001,
      "loss": 1.5922,
      "step": 79500
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0001,
      "loss": 1.6265,
      "step": 80000
    },
    {
      "epoch": 0.29,
      "eval_bleu": 15.3863,
      "eval_gen_len": 17.2173,
      "eval_loss": 1.223164439201355,
      "eval_runtime": 179.7055,
      "eval_samples_per_second": 16.694,
      "eval_steps_per_second": 1.046,
      "step": 80000
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0001,
      "loss": 1.6201,
      "step": 80500
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0001,
      "loss": 1.6108,
      "step": 81000
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0001,
      "loss": 1.589,
      "step": 81500
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0001,
      "loss": 1.6056,
      "step": 82000
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0001,
      "loss": 1.6169,
      "step": 82500
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0001,
      "loss": 1.5993,
      "step": 83000
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0001,
      "loss": 1.5922,
      "step": 83500
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0001,
      "loss": 1.6088,
      "step": 84000
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0001,
      "loss": 1.627,
      "step": 84500
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0001,
      "loss": 1.5995,
      "step": 85000
    },
    {
      "epoch": 0.3,
      "eval_bleu": 15.4069,
      "eval_gen_len": 17.2217,
      "eval_loss": 1.2199046611785889,
      "eval_runtime": 180.5862,
      "eval_samples_per_second": 16.613,
      "eval_steps_per_second": 1.041,
      "step": 85000
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0001,
      "loss": 1.6084,
      "step": 85500
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0001,
      "loss": 1.616,
      "step": 86000
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0001,
      "loss": 1.609,
      "step": 86500
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0001,
      "loss": 1.5885,
      "step": 87000
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0001,
      "loss": 1.5936,
      "step": 87500
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0001,
      "loss": 1.5915,
      "step": 88000
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0001,
      "loss": 1.6144,
      "step": 88500
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0001,
      "loss": 1.5775,
      "step": 89000
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0001,
      "loss": 1.5962,
      "step": 89500
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0001,
      "loss": 1.5999,
      "step": 90000
    },
    {
      "epoch": 0.32,
      "eval_bleu": 15.3272,
      "eval_gen_len": 17.236,
      "eval_loss": 1.2174800634384155,
      "eval_runtime": 184.3219,
      "eval_samples_per_second": 16.276,
      "eval_steps_per_second": 1.02,
      "step": 90000
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0001,
      "loss": 1.589,
      "step": 90500
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0001,
      "loss": 1.6045,
      "step": 91000
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0001,
      "loss": 1.5934,
      "step": 91500
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0001,
      "loss": 1.5798,
      "step": 92000
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0001,
      "loss": 1.6128,
      "step": 92500
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0001,
      "loss": 1.6137,
      "step": 93000
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0001,
      "loss": 1.6001,
      "step": 93500
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0001,
      "loss": 1.5834,
      "step": 94000
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0001,
      "loss": 1.6074,
      "step": 94500
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0001,
      "loss": 1.5923,
      "step": 95000
    },
    {
      "epoch": 0.34,
      "eval_bleu": 15.5346,
      "eval_gen_len": 17.2437,
      "eval_loss": 1.2179254293441772,
      "eval_runtime": 181.6046,
      "eval_samples_per_second": 16.519,
      "eval_steps_per_second": 1.035,
      "step": 95000
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0001,
      "loss": 1.5922,
      "step": 95500
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0001,
      "loss": 1.6042,
      "step": 96000
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0001,
      "loss": 1.5979,
      "step": 96500
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0001,
      "loss": 1.576,
      "step": 97000
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0001,
      "loss": 1.5968,
      "step": 97500
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0001,
      "loss": 1.5997,
      "step": 98000
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0001,
      "loss": 1.5873,
      "step": 98500
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0001,
      "loss": 1.5909,
      "step": 99000
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0001,
      "loss": 1.5808,
      "step": 99500
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0001,
      "loss": 1.5522,
      "step": 100000
    },
    {
      "epoch": 0.36,
      "eval_bleu": 15.2192,
      "eval_gen_len": 17.2457,
      "eval_loss": 1.2111108303070068,
      "eval_runtime": 180.3013,
      "eval_samples_per_second": 16.639,
      "eval_steps_per_second": 1.043,
      "step": 100000
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0001,
      "loss": 1.564,
      "step": 100500
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0001,
      "loss": 1.5779,
      "step": 101000
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0001,
      "loss": 1.575,
      "step": 101500
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0001,
      "loss": 1.5916,
      "step": 102000
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0001,
      "loss": 1.5819,
      "step": 102500
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0001,
      "loss": 1.575,
      "step": 103000
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0001,
      "loss": 1.5682,
      "step": 103500
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0001,
      "loss": 1.5785,
      "step": 104000
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0001,
      "loss": 1.5915,
      "step": 104500
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0001,
      "loss": 1.5916,
      "step": 105000
    },
    {
      "epoch": 0.37,
      "eval_bleu": 15.4399,
      "eval_gen_len": 17.2013,
      "eval_loss": 1.2112374305725098,
      "eval_runtime": 179.7121,
      "eval_samples_per_second": 16.693,
      "eval_steps_per_second": 1.046,
      "step": 105000
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0001,
      "loss": 1.5754,
      "step": 105500
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0001,
      "loss": 1.577,
      "step": 106000
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0001,
      "loss": 1.5795,
      "step": 106500
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0001,
      "loss": 1.572,
      "step": 107000
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0001,
      "loss": 1.579,
      "step": 107500
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0001,
      "loss": 1.5741,
      "step": 108000
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0001,
      "loss": 1.5873,
      "step": 108500
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0001,
      "loss": 1.5864,
      "step": 109000
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0001,
      "loss": 1.5847,
      "step": 109500
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0001,
      "loss": 1.5624,
      "step": 110000
    },
    {
      "epoch": 0.39,
      "eval_bleu": 15.5417,
      "eval_gen_len": 17.2057,
      "eval_loss": 1.209841012954712,
      "eval_runtime": 188.3104,
      "eval_samples_per_second": 15.931,
      "eval_steps_per_second": 0.998,
      "step": 110000
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0001,
      "loss": 1.5928,
      "step": 110500
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0001,
      "loss": 1.5838,
      "step": 111000
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0001,
      "loss": 1.5656,
      "step": 111500
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0001,
      "loss": 1.5418,
      "step": 112000
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0001,
      "loss": 1.5752,
      "step": 112500
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0001,
      "loss": 1.5707,
      "step": 113000
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0001,
      "loss": 1.5425,
      "step": 113500
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0001,
      "loss": 1.5444,
      "step": 114000
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0001,
      "loss": 1.5447,
      "step": 114500
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0001,
      "loss": 1.5729,
      "step": 115000
    },
    {
      "epoch": 0.41,
      "eval_bleu": 15.4886,
      "eval_gen_len": 17.2267,
      "eval_loss": 1.204936146736145,
      "eval_runtime": 178.5945,
      "eval_samples_per_second": 16.798,
      "eval_steps_per_second": 1.053,
      "step": 115000
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0001,
      "loss": 1.5681,
      "step": 115500
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0001,
      "loss": 1.5655,
      "step": 116000
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0001,
      "loss": 1.5725,
      "step": 116500
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0001,
      "loss": 1.5828,
      "step": 117000
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0001,
      "loss": 1.558,
      "step": 117500
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0001,
      "loss": 1.5509,
      "step": 118000
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0001,
      "loss": 1.5776,
      "step": 118500
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0001,
      "loss": 1.5497,
      "step": 119000
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0001,
      "loss": 1.5742,
      "step": 119500
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0001,
      "loss": 1.5728,
      "step": 120000
    },
    {
      "epoch": 0.43,
      "eval_bleu": 15.7267,
      "eval_gen_len": 17.2187,
      "eval_loss": 1.1999534368515015,
      "eval_runtime": 179.4954,
      "eval_samples_per_second": 16.714,
      "eval_steps_per_second": 1.047,
      "step": 120000
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0001,
      "loss": 1.5444,
      "step": 120500
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0001,
      "loss": 1.5641,
      "step": 121000
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0001,
      "loss": 1.5499,
      "step": 121500
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0001,
      "loss": 1.5659,
      "step": 122000
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0001,
      "loss": 1.5453,
      "step": 122500
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0001,
      "loss": 1.5564,
      "step": 123000
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0001,
      "loss": 1.538,
      "step": 123500
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0001,
      "loss": 1.5585,
      "step": 124000
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0001,
      "loss": 1.5503,
      "step": 124500
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0001,
      "loss": 1.5773,
      "step": 125000
    },
    {
      "epoch": 0.45,
      "eval_bleu": 15.5274,
      "eval_gen_len": 17.2397,
      "eval_loss": 1.2015475034713745,
      "eval_runtime": 180.6014,
      "eval_samples_per_second": 16.611,
      "eval_steps_per_second": 1.041,
      "step": 125000
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0001,
      "loss": 1.5736,
      "step": 125500
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0001,
      "loss": 1.5747,
      "step": 126000
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0001,
      "loss": 1.5688,
      "step": 126500
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0001,
      "loss": 1.5445,
      "step": 127000
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0001,
      "loss": 1.5318,
      "step": 127500
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0001,
      "loss": 1.5682,
      "step": 128000
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0001,
      "loss": 1.5483,
      "step": 128500
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0001,
      "loss": 1.5509,
      "step": 129000
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0001,
      "loss": 1.5519,
      "step": 129500
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0001,
      "loss": 1.5616,
      "step": 130000
    },
    {
      "epoch": 0.46,
      "eval_bleu": 15.4371,
      "eval_gen_len": 17.2293,
      "eval_loss": 1.1969568729400635,
      "eval_runtime": 179.7833,
      "eval_samples_per_second": 16.687,
      "eval_steps_per_second": 1.046,
      "step": 130000
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0001,
      "loss": 1.5416,
      "step": 130500
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0001,
      "loss": 1.5243,
      "step": 131000
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0001,
      "loss": 1.5602,
      "step": 131500
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0001,
      "loss": 1.5628,
      "step": 132000
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0001,
      "loss": 1.5665,
      "step": 132500
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0001,
      "loss": 1.5422,
      "step": 133000
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0001,
      "loss": 1.5489,
      "step": 133500
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0001,
      "loss": 1.5533,
      "step": 134000
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0001,
      "loss": 1.5726,
      "step": 134500
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0001,
      "loss": 1.5229,
      "step": 135000
    },
    {
      "epoch": 0.48,
      "eval_bleu": 15.4862,
      "eval_gen_len": 17.2447,
      "eval_loss": 1.1961296796798706,
      "eval_runtime": 201.6549,
      "eval_samples_per_second": 14.877,
      "eval_steps_per_second": 0.932,
      "step": 135000
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0001,
      "loss": 1.5315,
      "step": 135500
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0001,
      "loss": 1.5542,
      "step": 136000
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0001,
      "loss": 1.5591,
      "step": 136500
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0001,
      "loss": 1.5599,
      "step": 137000
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0001,
      "loss": 1.5483,
      "step": 137500
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0001,
      "loss": 1.5619,
      "step": 138000
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0001,
      "loss": 1.5566,
      "step": 138500
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0001,
      "loss": 1.5435,
      "step": 139000
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0001,
      "loss": 1.553,
      "step": 139500
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0001,
      "loss": 1.5259,
      "step": 140000
    },
    {
      "epoch": 0.5,
      "eval_bleu": 15.5223,
      "eval_gen_len": 17.2253,
      "eval_loss": 1.1915968656539917,
      "eval_runtime": 178.4591,
      "eval_samples_per_second": 16.811,
      "eval_steps_per_second": 1.053,
      "step": 140000
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0001,
      "loss": 1.5456,
      "step": 140500
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0001,
      "loss": 1.5442,
      "step": 141000
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0001,
      "loss": 1.5679,
      "step": 141500
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0001,
      "loss": 1.5812,
      "step": 142000
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0001,
      "loss": 1.5428,
      "step": 142500
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0001,
      "loss": 1.5256,
      "step": 143000
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0001,
      "loss": 1.54,
      "step": 143500
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0001,
      "loss": 1.5559,
      "step": 144000
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0001,
      "loss": 1.503,
      "step": 144500
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0001,
      "loss": 1.5406,
      "step": 145000
    },
    {
      "epoch": 0.52,
      "eval_bleu": 15.5976,
      "eval_gen_len": 17.2333,
      "eval_loss": 1.1902729272842407,
      "eval_runtime": 182.1398,
      "eval_samples_per_second": 16.471,
      "eval_steps_per_second": 1.032,
      "step": 145000
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0001,
      "loss": 1.5738,
      "step": 145500
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0001,
      "loss": 1.5451,
      "step": 146000
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0001,
      "loss": 1.566,
      "step": 146500
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0001,
      "loss": 1.5401,
      "step": 147000
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0001,
      "loss": 1.5642,
      "step": 147500
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0001,
      "loss": 1.5369,
      "step": 148000
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0001,
      "loss": 1.5137,
      "step": 148500
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0001,
      "loss": 1.5646,
      "step": 149000
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0001,
      "loss": 1.5441,
      "step": 149500
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0001,
      "loss": 1.5345,
      "step": 150000
    },
    {
      "epoch": 0.53,
      "eval_bleu": 15.8131,
      "eval_gen_len": 17.236,
      "eval_loss": 1.1884739398956299,
      "eval_runtime": 180.9284,
      "eval_samples_per_second": 16.581,
      "eval_steps_per_second": 1.039,
      "step": 150000
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0001,
      "loss": 1.556,
      "step": 150500
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0001,
      "loss": 1.5581,
      "step": 151000
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0001,
      "loss": 1.5223,
      "step": 151500
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0001,
      "loss": 1.5758,
      "step": 152000
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0001,
      "loss": 1.5187,
      "step": 152500
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0001,
      "loss": 1.5495,
      "step": 153000
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0001,
      "loss": 1.5467,
      "step": 153500
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0001,
      "loss": 1.5691,
      "step": 154000
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0001,
      "loss": 1.5224,
      "step": 154500
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0001,
      "loss": 1.5639,
      "step": 155000
    },
    {
      "epoch": 0.55,
      "eval_bleu": 15.7128,
      "eval_gen_len": 17.2447,
      "eval_loss": 1.192060112953186,
      "eval_runtime": 180.919,
      "eval_samples_per_second": 16.582,
      "eval_steps_per_second": 1.039,
      "step": 155000
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0001,
      "loss": 1.5359,
      "step": 155500
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0001,
      "loss": 1.5498,
      "step": 156000
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0001,
      "loss": 1.5419,
      "step": 156500
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0001,
      "loss": 1.5382,
      "step": 157000
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0001,
      "loss": 1.5337,
      "step": 157500
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0001,
      "loss": 1.5364,
      "step": 158000
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0001,
      "loss": 1.5415,
      "step": 158500
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0001,
      "loss": 1.5509,
      "step": 159000
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0001,
      "loss": 1.537,
      "step": 159500
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0001,
      "loss": 1.5187,
      "step": 160000
    },
    {
      "epoch": 0.57,
      "eval_bleu": 15.8426,
      "eval_gen_len": 17.2283,
      "eval_loss": 1.1836034059524536,
      "eval_runtime": 179.3082,
      "eval_samples_per_second": 16.731,
      "eval_steps_per_second": 1.048,
      "step": 160000
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0001,
      "loss": 1.5252,
      "step": 160500
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0001,
      "loss": 1.5334,
      "step": 161000
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0001,
      "loss": 1.5654,
      "step": 161500
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0001,
      "loss": 1.5402,
      "step": 162000
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0001,
      "loss": 1.5236,
      "step": 162500
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0001,
      "loss": 1.5005,
      "step": 163000
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0001,
      "loss": 1.5373,
      "step": 163500
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0001,
      "loss": 1.5308,
      "step": 164000
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0001,
      "loss": 1.5422,
      "step": 164500
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0001,
      "loss": 1.5149,
      "step": 165000
    },
    {
      "epoch": 0.59,
      "eval_bleu": 15.7588,
      "eval_gen_len": 17.2353,
      "eval_loss": 1.1842249631881714,
      "eval_runtime": 178.6735,
      "eval_samples_per_second": 16.79,
      "eval_steps_per_second": 1.052,
      "step": 165000
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0001,
      "loss": 1.5349,
      "step": 165500
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0001,
      "loss": 1.5496,
      "step": 166000
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0001,
      "loss": 1.5328,
      "step": 166500
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0001,
      "loss": 1.5511,
      "step": 167000
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0001,
      "loss": 1.5434,
      "step": 167500
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0001,
      "loss": 1.5329,
      "step": 168000
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0001,
      "loss": 1.5294,
      "step": 168500
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0001,
      "loss": 1.5171,
      "step": 169000
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0001,
      "loss": 1.5434,
      "step": 169500
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0001,
      "loss": 1.5073,
      "step": 170000
    },
    {
      "epoch": 0.61,
      "eval_bleu": 15.7586,
      "eval_gen_len": 17.2317,
      "eval_loss": 1.1863584518432617,
      "eval_runtime": 184.5174,
      "eval_samples_per_second": 16.259,
      "eval_steps_per_second": 1.019,
      "step": 170000
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0001,
      "loss": 1.5228,
      "step": 170500
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0001,
      "loss": 1.5297,
      "step": 171000
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0001,
      "loss": 1.5259,
      "step": 171500
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0001,
      "loss": 1.515,
      "step": 172000
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0001,
      "loss": 1.5264,
      "step": 172500
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0001,
      "loss": 1.5189,
      "step": 173000
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0001,
      "loss": 1.5254,
      "step": 173500
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0001,
      "loss": 1.5196,
      "step": 174000
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0001,
      "loss": 1.5165,
      "step": 174500
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0001,
      "loss": 1.5459,
      "step": 175000
    },
    {
      "epoch": 0.62,
      "eval_bleu": 15.4818,
      "eval_gen_len": 17.2423,
      "eval_loss": 1.183072566986084,
      "eval_runtime": 179.1119,
      "eval_samples_per_second": 16.749,
      "eval_steps_per_second": 1.05,
      "step": 175000
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0001,
      "loss": 1.525,
      "step": 175500
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0001,
      "loss": 1.5162,
      "step": 176000
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0001,
      "loss": 1.5415,
      "step": 176500
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0001,
      "loss": 1.4929,
      "step": 177000
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0001,
      "loss": 1.5233,
      "step": 177500
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0001,
      "loss": 1.5441,
      "step": 178000
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.0001,
      "loss": 1.5371,
      "step": 178500
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.0001,
      "loss": 1.5126,
      "step": 179000
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.0001,
      "loss": 1.5386,
      "step": 179500
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.0001,
      "loss": 1.5196,
      "step": 180000
    },
    {
      "epoch": 0.64,
      "eval_bleu": 15.8913,
      "eval_gen_len": 17.2113,
      "eval_loss": 1.1787080764770508,
      "eval_runtime": 180.1583,
      "eval_samples_per_second": 16.652,
      "eval_steps_per_second": 1.044,
      "step": 180000
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.0001,
      "loss": 1.5001,
      "step": 180500
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.0001,
      "loss": 1.5134,
      "step": 181000
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.0001,
      "loss": 1.5385,
      "step": 181500
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.0001,
      "loss": 1.5281,
      "step": 182000
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.0001,
      "loss": 1.5295,
      "step": 182500
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.0001,
      "loss": 1.507,
      "step": 183000
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.0001,
      "loss": 1.5265,
      "step": 183500
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0001,
      "loss": 1.5324,
      "step": 184000
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0001,
      "loss": 1.5213,
      "step": 184500
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0001,
      "loss": 1.4978,
      "step": 185000
    },
    {
      "epoch": 0.66,
      "eval_bleu": 15.768,
      "eval_gen_len": 17.2457,
      "eval_loss": 1.176733136177063,
      "eval_runtime": 179.609,
      "eval_samples_per_second": 16.703,
      "eval_steps_per_second": 1.047,
      "step": 185000
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0001,
      "loss": 1.5038,
      "step": 185500
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0001,
      "loss": 1.5334,
      "step": 186000
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0001,
      "loss": 1.5266,
      "step": 186500
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0001,
      "loss": 1.527,
      "step": 187000
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0001,
      "loss": 1.5125,
      "step": 187500
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0001,
      "loss": 1.5372,
      "step": 188000
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0001,
      "loss": 1.5223,
      "step": 188500
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0001,
      "loss": 1.5177,
      "step": 189000
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.0001,
      "loss": 1.5128,
      "step": 189500
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.0001,
      "loss": 1.5064,
      "step": 190000
    },
    {
      "epoch": 0.68,
      "eval_bleu": 15.8041,
      "eval_gen_len": 17.2323,
      "eval_loss": 1.1802318096160889,
      "eval_runtime": 177.0229,
      "eval_samples_per_second": 16.947,
      "eval_steps_per_second": 1.062,
      "step": 190000
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.0001,
      "loss": 1.5104,
      "step": 190500
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.0001,
      "loss": 1.5162,
      "step": 191000
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.0001,
      "loss": 1.5182,
      "step": 191500
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.0001,
      "loss": 1.521,
      "step": 192000
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.0001,
      "loss": 1.4983,
      "step": 192500
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.0001,
      "loss": 1.5128,
      "step": 193000
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.0001,
      "loss": 1.5018,
      "step": 193500
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.0001,
      "loss": 1.4949,
      "step": 194000
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.0001,
      "loss": 1.5163,
      "step": 194500
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0001,
      "loss": 1.522,
      "step": 195000
    },
    {
      "epoch": 0.7,
      "eval_bleu": 15.8856,
      "eval_gen_len": 17.2357,
      "eval_loss": 1.1744434833526611,
      "eval_runtime": 180.611,
      "eval_samples_per_second": 16.61,
      "eval_steps_per_second": 1.041,
      "step": 195000
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0001,
      "loss": 1.5308,
      "step": 195500
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0001,
      "loss": 1.5102,
      "step": 196000
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0001,
      "loss": 1.5227,
      "step": 196500
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0001,
      "loss": 1.5014,
      "step": 197000
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0001,
      "loss": 1.5338,
      "step": 197500
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.0001,
      "loss": 1.5028,
      "step": 198000
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.0001,
      "loss": 1.5199,
      "step": 198500
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.0001,
      "loss": 1.5112,
      "step": 199000
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.0001,
      "loss": 1.5182,
      "step": 199500
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.0001,
      "loss": 1.5297,
      "step": 200000
    },
    {
      "epoch": 0.71,
      "eval_bleu": 15.5916,
      "eval_gen_len": 17.2447,
      "eval_loss": 1.179445743560791,
      "eval_runtime": 178.9024,
      "eval_samples_per_second": 16.769,
      "eval_steps_per_second": 1.051,
      "step": 200000
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.0001,
      "loss": 1.5248,
      "step": 200500
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.0001,
      "loss": 1.4947,
      "step": 201000
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.0001,
      "loss": 1.5224,
      "step": 201500
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.0001,
      "loss": 1.5359,
      "step": 202000
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.0001,
      "loss": 1.5086,
      "step": 202500
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.0001,
      "loss": 1.5268,
      "step": 203000
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0001,
      "loss": 1.5151,
      "step": 203500
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0001,
      "loss": 1.4821,
      "step": 204000
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0001,
      "loss": 1.5173,
      "step": 204500
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0001,
      "loss": 1.4964,
      "step": 205000
    },
    {
      "epoch": 0.73,
      "eval_bleu": 15.7873,
      "eval_gen_len": 17.23,
      "eval_loss": 1.171249270439148,
      "eval_runtime": 179.2351,
      "eval_samples_per_second": 16.738,
      "eval_steps_per_second": 1.049,
      "step": 205000
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0001,
      "loss": 1.4873,
      "step": 205500
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0001,
      "loss": 1.4984,
      "step": 206000
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0001,
      "loss": 1.5069,
      "step": 206500
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0001,
      "loss": 1.5168,
      "step": 207000
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0001,
      "loss": 1.5034,
      "step": 207500
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0001,
      "loss": 1.4952,
      "step": 208000
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0001,
      "loss": 1.509,
      "step": 208500
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0001,
      "loss": 1.4856,
      "step": 209000
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.0001,
      "loss": 1.5194,
      "step": 209500
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.0001,
      "loss": 1.4995,
      "step": 210000
    },
    {
      "epoch": 0.75,
      "eval_bleu": 15.9841,
      "eval_gen_len": 17.2263,
      "eval_loss": 1.1729360818862915,
      "eval_runtime": 198.1261,
      "eval_samples_per_second": 15.142,
      "eval_steps_per_second": 0.949,
      "step": 210000
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.0001,
      "loss": 1.5071,
      "step": 210500
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.0001,
      "loss": 1.5121,
      "step": 211000
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.0001,
      "loss": 1.5154,
      "step": 211500
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.0001,
      "loss": 1.4939,
      "step": 212000
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.0001,
      "loss": 1.4962,
      "step": 212500
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.0001,
      "loss": 1.4911,
      "step": 213000
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.0001,
      "loss": 1.5165,
      "step": 213500
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.0001,
      "loss": 1.5029,
      "step": 214000
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.0001,
      "loss": 1.4866,
      "step": 214500
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0001,
      "loss": 1.5019,
      "step": 215000
    },
    {
      "epoch": 0.77,
      "eval_bleu": 15.8913,
      "eval_gen_len": 17.2423,
      "eval_loss": 1.1722415685653687,
      "eval_runtime": 179.987,
      "eval_samples_per_second": 16.668,
      "eval_steps_per_second": 1.045,
      "step": 215000
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0001,
      "loss": 1.5028,
      "step": 215500
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0001,
      "loss": 1.5034,
      "step": 216000
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0001,
      "loss": 1.5097,
      "step": 216500
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0001,
      "loss": 1.4972,
      "step": 217000
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0001,
      "loss": 1.5038,
      "step": 217500
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0001,
      "loss": 1.5003,
      "step": 218000
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0001,
      "loss": 1.5155,
      "step": 218500
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0001,
      "loss": 1.5075,
      "step": 219000
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0001,
      "loss": 1.4918,
      "step": 219500
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0001,
      "loss": 1.4771,
      "step": 220000
    },
    {
      "epoch": 0.78,
      "eval_bleu": 16.0441,
      "eval_gen_len": 17.247,
      "eval_loss": 1.1694059371948242,
      "eval_runtime": 180.3281,
      "eval_samples_per_second": 16.636,
      "eval_steps_per_second": 1.043,
      "step": 220000
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.0001,
      "loss": 1.4949,
      "step": 220500
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.0001,
      "loss": 1.4991,
      "step": 221000
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.0001,
      "loss": 1.4803,
      "step": 221500
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.0001,
      "loss": 1.4929,
      "step": 222000
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.0001,
      "loss": 1.5001,
      "step": 222500
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.0001,
      "loss": 1.505,
      "step": 223000
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0001,
      "loss": 1.4978,
      "step": 223500
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0001,
      "loss": 1.4951,
      "step": 224000
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0001,
      "loss": 1.5015,
      "step": 224500
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0001,
      "loss": 1.4926,
      "step": 225000
    },
    {
      "epoch": 0.8,
      "eval_bleu": 15.9435,
      "eval_gen_len": 17.225,
      "eval_loss": 1.166946291923523,
      "eval_runtime": 178.5659,
      "eval_samples_per_second": 16.801,
      "eval_steps_per_second": 1.053,
      "step": 225000
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0001,
      "loss": 1.4956,
      "step": 225500
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.0001,
      "loss": 1.507,
      "step": 226000
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.0001,
      "loss": 1.5063,
      "step": 226500
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.0001,
      "loss": 1.5017,
      "step": 227000
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.0001,
      "loss": 1.4815,
      "step": 227500
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.0001,
      "loss": 1.5154,
      "step": 228000
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.0001,
      "loss": 1.4737,
      "step": 228500
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.0001,
      "loss": 1.5014,
      "step": 229000
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.0001,
      "loss": 1.5002,
      "step": 229500
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.0001,
      "loss": 1.5037,
      "step": 230000
    },
    {
      "epoch": 0.82,
      "eval_bleu": 16.1227,
      "eval_gen_len": 17.229,
      "eval_loss": 1.1636000871658325,
      "eval_runtime": 179.6956,
      "eval_samples_per_second": 16.695,
      "eval_steps_per_second": 1.046,
      "step": 230000
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.0001,
      "loss": 1.4883,
      "step": 230500
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.0001,
      "loss": 1.5005,
      "step": 231000
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.0001,
      "loss": 1.4975,
      "step": 231500
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.0001,
      "loss": 1.4812,
      "step": 232000
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.0001,
      "loss": 1.5066,
      "step": 232500
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.0001,
      "loss": 1.5036,
      "step": 233000
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.0001,
      "loss": 1.4684,
      "step": 233500
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.0001,
      "loss": 1.486,
      "step": 234000
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.0001,
      "loss": 1.4904,
      "step": 234500
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.0001,
      "loss": 1.4821,
      "step": 235000
    },
    {
      "epoch": 0.84,
      "eval_bleu": 15.8818,
      "eval_gen_len": 17.2463,
      "eval_loss": 1.1675198078155518,
      "eval_runtime": 179.5855,
      "eval_samples_per_second": 16.705,
      "eval_steps_per_second": 1.047,
      "step": 235000
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.0001,
      "loss": 1.496,
      "step": 235500
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.0001,
      "loss": 1.5012,
      "step": 236000
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.0001,
      "loss": 1.4687,
      "step": 236500
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.0001,
      "loss": 1.468,
      "step": 237000
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.0001,
      "loss": 1.5017,
      "step": 237500
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.0001,
      "loss": 1.5057,
      "step": 238000
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.0001,
      "loss": 1.4869,
      "step": 238500
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.0001,
      "loss": 1.4915,
      "step": 239000
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.0001,
      "loss": 1.5138,
      "step": 239500
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.0001,
      "loss": 1.4726,
      "step": 240000
    },
    {
      "epoch": 0.86,
      "eval_bleu": 15.9348,
      "eval_gen_len": 17.217,
      "eval_loss": 1.1675784587860107,
      "eval_runtime": 188.1276,
      "eval_samples_per_second": 15.947,
      "eval_steps_per_second": 0.999,
      "step": 240000
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.0001,
      "loss": 1.4937,
      "step": 240500
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.0001,
      "loss": 1.492,
      "step": 241000
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.0001,
      "loss": 1.4985,
      "step": 241500
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.0001,
      "loss": 1.4947,
      "step": 242000
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.0001,
      "loss": 1.4905,
      "step": 242500
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.0001,
      "loss": 1.4984,
      "step": 243000
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.0001,
      "loss": 1.4674,
      "step": 243500
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.0001,
      "loss": 1.4829,
      "step": 244000
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.0001,
      "loss": 1.501,
      "step": 244500
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.0001,
      "loss": 1.4954,
      "step": 245000
    },
    {
      "epoch": 0.87,
      "eval_bleu": 16.0221,
      "eval_gen_len": 17.2257,
      "eval_loss": 1.164318323135376,
      "eval_runtime": 190.8915,
      "eval_samples_per_second": 15.716,
      "eval_steps_per_second": 0.985,
      "step": 245000
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.0001,
      "loss": 1.4979,
      "step": 245500
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.0001,
      "loss": 1.4687,
      "step": 246000
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.0001,
      "loss": 1.4622,
      "step": 246500
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.0001,
      "loss": 1.4976,
      "step": 247000
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.0001,
      "loss": 1.4894,
      "step": 247500
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.0001,
      "loss": 1.4793,
      "step": 248000
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.0001,
      "loss": 1.4691,
      "step": 248500
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.0001,
      "loss": 1.4748,
      "step": 249000
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.0001,
      "loss": 1.4732,
      "step": 249500
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.0001,
      "loss": 1.4928,
      "step": 250000
    },
    {
      "epoch": 0.89,
      "eval_bleu": 16.0874,
      "eval_gen_len": 17.2277,
      "eval_loss": 1.1611835956573486,
      "eval_runtime": 169.4356,
      "eval_samples_per_second": 17.706,
      "eval_steps_per_second": 1.11,
      "step": 250000
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.0001,
      "loss": 1.4972,
      "step": 250500
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.0001,
      "loss": 1.4892,
      "step": 251000
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.0001,
      "loss": 1.4605,
      "step": 251500
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.0001,
      "loss": 1.4837,
      "step": 252000
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.0001,
      "loss": 1.4865,
      "step": 252500
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.0001,
      "loss": 1.487,
      "step": 253000
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.0001,
      "loss": 1.4995,
      "step": 253500
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.0001,
      "loss": 1.486,
      "step": 254000
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.0001,
      "loss": 1.4879,
      "step": 254500
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.0001,
      "loss": 1.4915,
      "step": 255000
    },
    {
      "epoch": 0.91,
      "eval_bleu": 15.9305,
      "eval_gen_len": 17.218,
      "eval_loss": 1.1611851453781128,
      "eval_runtime": 167.8847,
      "eval_samples_per_second": 17.869,
      "eval_steps_per_second": 1.12,
      "step": 255000
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.0001,
      "loss": 1.5044,
      "step": 255500
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.0001,
      "loss": 1.4901,
      "step": 256000
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.0001,
      "loss": 1.4946,
      "step": 256500
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.0001,
      "loss": 1.4898,
      "step": 257000
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.0001,
      "loss": 1.4816,
      "step": 257500
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.0001,
      "loss": 1.4798,
      "step": 258000
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.0001,
      "loss": 1.4685,
      "step": 258500
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.0001,
      "loss": 1.4988,
      "step": 259000
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.0001,
      "loss": 1.4788,
      "step": 259500
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.0001,
      "loss": 1.4723,
      "step": 260000
    },
    {
      "epoch": 0.93,
      "eval_bleu": 15.9979,
      "eval_gen_len": 17.2283,
      "eval_loss": 1.1573797464370728,
      "eval_runtime": 196.0191,
      "eval_samples_per_second": 15.305,
      "eval_steps_per_second": 0.959,
      "step": 260000
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.0001,
      "loss": 1.4815,
      "step": 260500
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.0001,
      "loss": 1.4643,
      "step": 261000
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.0001,
      "loss": 1.4973,
      "step": 261500
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.0001,
      "loss": 1.4889,
      "step": 262000
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.0001,
      "loss": 1.47,
      "step": 262500
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.0001,
      "loss": 1.4708,
      "step": 263000
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.0001,
      "loss": 1.478,
      "step": 263500
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.0001,
      "loss": 1.493,
      "step": 264000
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.0001,
      "loss": 1.4793,
      "step": 264500
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.0001,
      "loss": 1.4799,
      "step": 265000
    },
    {
      "epoch": 0.94,
      "eval_bleu": 15.7827,
      "eval_gen_len": 17.2477,
      "eval_loss": 1.158940315246582,
      "eval_runtime": 171.0024,
      "eval_samples_per_second": 17.544,
      "eval_steps_per_second": 1.099,
      "step": 265000
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.0001,
      "loss": 1.4849,
      "step": 265500
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.0001,
      "loss": 1.4808,
      "step": 266000
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.0001,
      "loss": 1.4676,
      "step": 266500
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.0001,
      "loss": 1.4638,
      "step": 267000
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.0001,
      "loss": 1.5045,
      "step": 267500
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.0001,
      "loss": 1.4971,
      "step": 268000
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.0001,
      "loss": 1.4726,
      "step": 268500
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.0001,
      "loss": 1.4786,
      "step": 269000
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.0001,
      "loss": 1.5031,
      "step": 269500
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.0001,
      "loss": 1.4887,
      "step": 270000
    },
    {
      "epoch": 0.96,
      "eval_bleu": 16.0626,
      "eval_gen_len": 17.201,
      "eval_loss": 1.1597611904144287,
      "eval_runtime": 168.7388,
      "eval_samples_per_second": 17.779,
      "eval_steps_per_second": 1.114,
      "step": 270000
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.0001,
      "loss": 1.4725,
      "step": 270500
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.0001,
      "loss": 1.4835,
      "step": 271000
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.0001,
      "loss": 1.4692,
      "step": 271500
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.0001,
      "loss": 1.499,
      "step": 272000
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.0001,
      "loss": 1.4662,
      "step": 272500
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.0001,
      "loss": 1.4583,
      "step": 273000
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.0001,
      "loss": 1.4754,
      "step": 273500
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.0001,
      "loss": 1.4654,
      "step": 274000
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.0001,
      "loss": 1.4931,
      "step": 274500
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.0001,
      "loss": 1.484,
      "step": 275000
    },
    {
      "epoch": 0.98,
      "eval_bleu": 15.8914,
      "eval_gen_len": 17.2293,
      "eval_loss": 1.1583878993988037,
      "eval_runtime": 167.8424,
      "eval_samples_per_second": 17.874,
      "eval_steps_per_second": 1.12,
      "step": 275000
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.0001,
      "loss": 1.4657,
      "step": 275500
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.0001,
      "loss": 1.4628,
      "step": 276000
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.0001,
      "loss": 1.4776,
      "step": 276500
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.0001,
      "loss": 1.4804,
      "step": 277000
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.0001,
      "loss": 1.4813,
      "step": 277500
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.0001,
      "loss": 1.4811,
      "step": 278000
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.0001,
      "loss": 1.4677,
      "step": 278500
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.0001,
      "loss": 1.4727,
      "step": 279000
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.0001,
      "loss": 1.486,
      "step": 279500
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.0001,
      "loss": 1.4631,
      "step": 280000
    },
    {
      "epoch": 1.0,
      "eval_bleu": 15.973,
      "eval_gen_len": 17.2537,
      "eval_loss": 1.1608219146728516,
      "eval_runtime": 169.0186,
      "eval_samples_per_second": 17.75,
      "eval_steps_per_second": 1.112,
      "step": 280000
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.0001,
      "loss": 1.4864,
      "step": 280500
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.0001,
      "loss": 1.4195,
      "step": 281000
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.0001,
      "loss": 1.4412,
      "step": 281500
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.0001,
      "loss": 1.425,
      "step": 282000
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.0001,
      "loss": 1.4133,
      "step": 282500
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.0001,
      "loss": 1.4102,
      "step": 283000
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.0001,
      "loss": 1.4137,
      "step": 283500
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.0001,
      "loss": 1.4081,
      "step": 284000
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.0001,
      "loss": 1.408,
      "step": 284500
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.0001,
      "loss": 1.4393,
      "step": 285000
    },
    {
      "epoch": 1.02,
      "eval_bleu": 15.9418,
      "eval_gen_len": 17.221,
      "eval_loss": 1.1621432304382324,
      "eval_runtime": 169.2373,
      "eval_samples_per_second": 17.727,
      "eval_steps_per_second": 1.111,
      "step": 285000
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.0001,
      "loss": 1.4186,
      "step": 285500
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.0001,
      "loss": 1.4215,
      "step": 286000
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.0001,
      "loss": 1.4244,
      "step": 286500
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.0001,
      "loss": 1.433,
      "step": 287000
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.0001,
      "loss": 1.4203,
      "step": 287500
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.0001,
      "loss": 1.4159,
      "step": 288000
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.0001,
      "loss": 1.3987,
      "step": 288500
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.0001,
      "loss": 1.4258,
      "step": 289000
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.0001,
      "loss": 1.4122,
      "step": 289500
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.0001,
      "loss": 1.3953,
      "step": 290000
    },
    {
      "epoch": 1.03,
      "eval_bleu": 15.9209,
      "eval_gen_len": 17.241,
      "eval_loss": 1.1581294536590576,
      "eval_runtime": 169.6325,
      "eval_samples_per_second": 17.685,
      "eval_steps_per_second": 1.108,
      "step": 290000
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.0001,
      "loss": 1.3846,
      "step": 290500
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.0001,
      "loss": 1.4382,
      "step": 291000
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.0001,
      "loss": 1.4267,
      "step": 291500
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.0001,
      "loss": 1.4243,
      "step": 292000
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.0001,
      "loss": 1.4144,
      "step": 292500
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.0001,
      "loss": 1.4273,
      "step": 293000
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.0001,
      "loss": 1.4193,
      "step": 293500
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.0001,
      "loss": 1.411,
      "step": 294000
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.0001,
      "loss": 1.4247,
      "step": 294500
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.0001,
      "loss": 1.3903,
      "step": 295000
    },
    {
      "epoch": 1.05,
      "eval_bleu": 15.9687,
      "eval_gen_len": 17.2437,
      "eval_loss": 1.1619981527328491,
      "eval_runtime": 168.2889,
      "eval_samples_per_second": 17.826,
      "eval_steps_per_second": 1.117,
      "step": 295000
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.0001,
      "loss": 1.4059,
      "step": 295500
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.0001,
      "loss": 1.4246,
      "step": 296000
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.0001,
      "loss": 1.4187,
      "step": 296500
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.0001,
      "loss": 1.4099,
      "step": 297000
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.0001,
      "loss": 1.4201,
      "step": 297500
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.0001,
      "loss": 1.4157,
      "step": 298000
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.0001,
      "loss": 1.4329,
      "step": 298500
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0001,
      "loss": 1.423,
      "step": 299000
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0001,
      "loss": 1.4464,
      "step": 299500
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0001,
      "loss": 1.4138,
      "step": 300000
    },
    {
      "epoch": 1.07,
      "eval_bleu": 16.0157,
      "eval_gen_len": 17.229,
      "eval_loss": 1.1558936834335327,
      "eval_runtime": 168.1206,
      "eval_samples_per_second": 17.844,
      "eval_steps_per_second": 1.118,
      "step": 300000
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0001,
      "loss": 1.4312,
      "step": 300500
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0001,
      "loss": 1.4277,
      "step": 301000
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0001,
      "loss": 1.4429,
      "step": 301500
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.0001,
      "loss": 1.4106,
      "step": 302000
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.0001,
      "loss": 1.3988,
      "step": 302500
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.0001,
      "loss": 1.4232,
      "step": 303000
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.0001,
      "loss": 1.442,
      "step": 303500
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.0001,
      "loss": 1.42,
      "step": 304000
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0001,
      "loss": 1.416,
      "step": 304500
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0001,
      "loss": 1.4108,
      "step": 305000
    },
    {
      "epoch": 1.09,
      "eval_bleu": 15.9367,
      "eval_gen_len": 17.2437,
      "eval_loss": 1.155571460723877,
      "eval_runtime": 166.8072,
      "eval_samples_per_second": 17.985,
      "eval_steps_per_second": 1.127,
      "step": 305000
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0001,
      "loss": 1.4125,
      "step": 305500
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0001,
      "loss": 1.4191,
      "step": 306000
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0001,
      "loss": 1.4443,
      "step": 306500
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0001,
      "loss": 1.3871,
      "step": 307000
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0001,
      "loss": 1.4162,
      "step": 307500
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0001,
      "loss": 1.4242,
      "step": 308000
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0001,
      "loss": 1.4097,
      "step": 308500
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0001,
      "loss": 1.3992,
      "step": 309000
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0001,
      "loss": 1.4104,
      "step": 309500
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0001,
      "loss": 1.429,
      "step": 310000
    },
    {
      "epoch": 1.1,
      "eval_bleu": 16.0064,
      "eval_gen_len": 17.2253,
      "eval_loss": 1.1538498401641846,
      "eval_runtime": 179.5358,
      "eval_samples_per_second": 16.71,
      "eval_steps_per_second": 1.047,
      "step": 310000
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.0001,
      "loss": 1.4135,
      "step": 310500
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.0001,
      "loss": 1.417,
      "step": 311000
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.0001,
      "loss": 1.4075,
      "step": 311500
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.0001,
      "loss": 1.4386,
      "step": 312000
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.0001,
      "loss": 1.4089,
      "step": 312500
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.0001,
      "loss": 1.394,
      "step": 313000
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.0001,
      "loss": 1.4325,
      "step": 313500
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.0001,
      "loss": 1.4032,
      "step": 314000
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.0001,
      "loss": 1.3985,
      "step": 314500
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.0001,
      "loss": 1.4353,
      "step": 315000
    },
    {
      "epoch": 1.12,
      "eval_bleu": 16.1492,
      "eval_gen_len": 17.221,
      "eval_loss": 1.151426911354065,
      "eval_runtime": 166.4584,
      "eval_samples_per_second": 18.023,
      "eval_steps_per_second": 1.129,
      "step": 315000
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.0001,
      "loss": 1.4511,
      "step": 315500
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.0001,
      "loss": 1.4427,
      "step": 316000
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.0001,
      "loss": 1.4378,
      "step": 316500
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.0001,
      "loss": 1.4181,
      "step": 317000
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.0001,
      "loss": 1.4421,
      "step": 317500
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.0001,
      "loss": 1.411,
      "step": 318000
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.0001,
      "loss": 1.4305,
      "step": 318500
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.0001,
      "loss": 1.4287,
      "step": 319000
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.0001,
      "loss": 1.423,
      "step": 319500
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.0001,
      "loss": 1.4132,
      "step": 320000
    },
    {
      "epoch": 1.14,
      "eval_bleu": 16.1148,
      "eval_gen_len": 17.224,
      "eval_loss": 1.155116319656372,
      "eval_runtime": 167.2951,
      "eval_samples_per_second": 17.932,
      "eval_steps_per_second": 1.124,
      "step": 320000
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.0001,
      "loss": 1.4139,
      "step": 320500
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.0001,
      "loss": 1.4224,
      "step": 321000
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.0001,
      "loss": 1.4304,
      "step": 321500
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.0001,
      "loss": 1.4017,
      "step": 322000
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.0001,
      "loss": 1.4213,
      "step": 322500
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.0001,
      "loss": 1.4226,
      "step": 323000
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.0001,
      "loss": 1.4201,
      "step": 323500
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.0001,
      "loss": 1.4136,
      "step": 324000
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.0001,
      "loss": 1.4073,
      "step": 324500
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.0001,
      "loss": 1.4367,
      "step": 325000
    },
    {
      "epoch": 1.16,
      "eval_bleu": 15.9087,
      "eval_gen_len": 17.211,
      "eval_loss": 1.1528856754302979,
      "eval_runtime": 166.7183,
      "eval_samples_per_second": 17.994,
      "eval_steps_per_second": 1.128,
      "step": 325000
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.0001,
      "loss": 1.4218,
      "step": 325500
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.0001,
      "loss": 1.3805,
      "step": 326000
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.0001,
      "loss": 1.4221,
      "step": 326500
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0001,
      "loss": 1.4126,
      "step": 327000
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0001,
      "loss": 1.4524,
      "step": 327500
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0001,
      "loss": 1.4151,
      "step": 328000
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0001,
      "loss": 1.4214,
      "step": 328500
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0001,
      "loss": 1.3809,
      "step": 329000
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0001,
      "loss": 1.4167,
      "step": 329500
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.0001,
      "loss": 1.4191,
      "step": 330000
    },
    {
      "epoch": 1.18,
      "eval_bleu": 16.1278,
      "eval_gen_len": 17.221,
      "eval_loss": 1.1542410850524902,
      "eval_runtime": 168.2051,
      "eval_samples_per_second": 17.835,
      "eval_steps_per_second": 1.118,
      "step": 330000
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.0001,
      "loss": 1.3929,
      "step": 330500
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.0001,
      "loss": 1.4365,
      "step": 331000
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.0001,
      "loss": 1.401,
      "step": 331500
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.0001,
      "loss": 1.3796,
      "step": 332000
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.0001,
      "loss": 1.413,
      "step": 332500
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.0001,
      "loss": 1.3946,
      "step": 333000
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.0001,
      "loss": 1.4317,
      "step": 333500
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.0001,
      "loss": 1.4078,
      "step": 334000
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.0001,
      "loss": 1.4409,
      "step": 334500
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.0001,
      "loss": 1.4164,
      "step": 335000
    },
    {
      "epoch": 1.19,
      "eval_bleu": 16.2028,
      "eval_gen_len": 17.2027,
      "eval_loss": 1.1532447338104248,
      "eval_runtime": 167.2569,
      "eval_samples_per_second": 17.936,
      "eval_steps_per_second": 1.124,
      "step": 335000
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0001,
      "loss": 1.4241,
      "step": 335500
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0001,
      "loss": 1.3915,
      "step": 336000
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0001,
      "loss": 1.4431,
      "step": 336500
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0001,
      "loss": 1.4064,
      "step": 337000
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0001,
      "loss": 1.4402,
      "step": 337500
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0001,
      "loss": 1.4319,
      "step": 338000
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.0001,
      "loss": 1.4513,
      "step": 338500
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.0001,
      "loss": 1.4253,
      "step": 339000
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.0001,
      "loss": 1.4028,
      "step": 339500
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.0001,
      "loss": 1.4247,
      "step": 340000
    },
    {
      "epoch": 1.21,
      "eval_bleu": 16.1903,
      "eval_gen_len": 17.2157,
      "eval_loss": 1.149227499961853,
      "eval_runtime": 169.898,
      "eval_samples_per_second": 17.658,
      "eval_steps_per_second": 1.107,
      "step": 340000
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.0001,
      "loss": 1.4259,
      "step": 340500
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.0001,
      "loss": 1.4381,
      "step": 341000
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.0001,
      "loss": 1.4045,
      "step": 341500
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.0001,
      "loss": 1.4231,
      "step": 342000
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.0001,
      "loss": 1.4088,
      "step": 342500
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.0001,
      "loss": 1.3943,
      "step": 343000
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.0001,
      "loss": 1.4262,
      "step": 343500
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.0001,
      "loss": 1.4117,
      "step": 344000
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.0001,
      "loss": 1.4205,
      "step": 344500
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.0001,
      "loss": 1.416,
      "step": 345000
    },
    {
      "epoch": 1.23,
      "eval_bleu": 15.8815,
      "eval_gen_len": 17.234,
      "eval_loss": 1.1508541107177734,
      "eval_runtime": 168.8909,
      "eval_samples_per_second": 17.763,
      "eval_steps_per_second": 1.113,
      "step": 345000
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.0001,
      "loss": 1.4326,
      "step": 345500
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.0001,
      "loss": 1.4182,
      "step": 346000
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.0001,
      "loss": 1.4223,
      "step": 346500
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0001,
      "loss": 1.3952,
      "step": 347000
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0001,
      "loss": 1.4154,
      "step": 347500
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0001,
      "loss": 1.4296,
      "step": 348000
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0001,
      "loss": 1.4166,
      "step": 348500
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0001,
      "loss": 1.425,
      "step": 349000
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.0001,
      "loss": 1.4098,
      "step": 349500
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.0001,
      "loss": 1.3947,
      "step": 350000
    },
    {
      "epoch": 1.25,
      "eval_bleu": 15.8977,
      "eval_gen_len": 17.2367,
      "eval_loss": 1.1482059955596924,
      "eval_runtime": 168.2127,
      "eval_samples_per_second": 17.835,
      "eval_steps_per_second": 1.118,
      "step": 350000
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.0001,
      "loss": 1.4147,
      "step": 350500
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.0001,
      "loss": 1.3816,
      "step": 351000
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.0001,
      "loss": 1.4114,
      "step": 351500
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.0001,
      "loss": 1.4049,
      "step": 352000
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.0001,
      "loss": 1.4225,
      "step": 352500
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.0001,
      "loss": 1.4298,
      "step": 353000
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.0001,
      "loss": 1.4194,
      "step": 353500
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.0001,
      "loss": 1.4222,
      "step": 354000
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.0001,
      "loss": 1.4023,
      "step": 354500
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.0001,
      "loss": 1.4083,
      "step": 355000
    },
    {
      "epoch": 1.27,
      "eval_bleu": 16.1412,
      "eval_gen_len": 17.2037,
      "eval_loss": 1.1469227075576782,
      "eval_runtime": 168.482,
      "eval_samples_per_second": 17.806,
      "eval_steps_per_second": 1.116,
      "step": 355000
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.0001,
      "loss": 1.4014,
      "step": 355500
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.0001,
      "loss": 1.423,
      "step": 356000
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.0001,
      "loss": 1.4247,
      "step": 356500
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.0001,
      "loss": 1.4228,
      "step": 357000
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.0001,
      "loss": 1.422,
      "step": 357500
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.0001,
      "loss": 1.3857,
      "step": 358000
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.0001,
      "loss": 1.4052,
      "step": 358500
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.0001,
      "loss": 1.4068,
      "step": 359000
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.0001,
      "loss": 1.4264,
      "step": 359500
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.0001,
      "loss": 1.4153,
      "step": 360000
    },
    {
      "epoch": 1.28,
      "eval_bleu": 15.9324,
      "eval_gen_len": 17.239,
      "eval_loss": 1.1481246948242188,
      "eval_runtime": 168.4647,
      "eval_samples_per_second": 17.808,
      "eval_steps_per_second": 1.116,
      "step": 360000
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.0001,
      "loss": 1.4019,
      "step": 360500
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.0001,
      "loss": 1.3826,
      "step": 361000
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.0001,
      "loss": 1.4,
      "step": 361500
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.0001,
      "loss": 1.4046,
      "step": 362000
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.0001,
      "loss": 1.4166,
      "step": 362500
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.0001,
      "loss": 1.4263,
      "step": 363000
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.0001,
      "loss": 1.3992,
      "step": 363500
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.0001,
      "loss": 1.4354,
      "step": 364000
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.0001,
      "loss": 1.4123,
      "step": 364500
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.0001,
      "loss": 1.4161,
      "step": 365000
    },
    {
      "epoch": 1.3,
      "eval_bleu": 16.2851,
      "eval_gen_len": 17.237,
      "eval_loss": 1.1478354930877686,
      "eval_runtime": 166.1965,
      "eval_samples_per_second": 18.051,
      "eval_steps_per_second": 1.131,
      "step": 365000
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.0001,
      "loss": 1.4076,
      "step": 365500
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.0001,
      "loss": 1.3988,
      "step": 366000
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0001,
      "loss": 1.3985,
      "step": 366500
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0001,
      "loss": 1.412,
      "step": 367000
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0001,
      "loss": 1.4278,
      "step": 367500
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0001,
      "loss": 1.4212,
      "step": 368000
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0001,
      "loss": 1.4111,
      "step": 368500
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.0001,
      "loss": 1.4072,
      "step": 369000
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.0001,
      "loss": 1.4285,
      "step": 369500
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.0001,
      "loss": 1.4195,
      "step": 370000
    },
    {
      "epoch": 1.32,
      "eval_bleu": 16.1431,
      "eval_gen_len": 17.247,
      "eval_loss": 1.1499695777893066,
      "eval_runtime": 168.7396,
      "eval_samples_per_second": 17.779,
      "eval_steps_per_second": 1.114,
      "step": 370000
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.0001,
      "loss": 1.4188,
      "step": 370500
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.0001,
      "loss": 1.3934,
      "step": 371000
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.0001,
      "loss": 1.416,
      "step": 371500
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.0001,
      "loss": 1.4295,
      "step": 372000
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.0001,
      "loss": 1.3999,
      "step": 372500
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.0001,
      "loss": 1.4044,
      "step": 373000
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.0001,
      "loss": 1.4271,
      "step": 373500
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.0001,
      "loss": 1.4233,
      "step": 374000
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.0001,
      "loss": 1.4151,
      "step": 374500
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.0001,
      "loss": 1.4197,
      "step": 375000
    },
    {
      "epoch": 1.34,
      "eval_bleu": 16.0361,
      "eval_gen_len": 17.2437,
      "eval_loss": 1.1444387435913086,
      "eval_runtime": 170.1455,
      "eval_samples_per_second": 17.632,
      "eval_steps_per_second": 1.105,
      "step": 375000
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.0001,
      "loss": 1.4001,
      "step": 375500
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.0001,
      "loss": 1.4263,
      "step": 376000
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.0001,
      "loss": 1.4085,
      "step": 376500
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.0001,
      "loss": 1.3903,
      "step": 377000
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.0001,
      "loss": 1.4001,
      "step": 377500
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.0001,
      "loss": 1.4263,
      "step": 378000
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.0001,
      "loss": 1.4206,
      "step": 378500
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.0001,
      "loss": 1.4053,
      "step": 379000
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.0001,
      "loss": 1.4201,
      "step": 379500
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.0001,
      "loss": 1.3977,
      "step": 380000
    },
    {
      "epoch": 1.35,
      "eval_bleu": 16.1154,
      "eval_gen_len": 17.222,
      "eval_loss": 1.1454415321350098,
      "eval_runtime": 170.8086,
      "eval_samples_per_second": 17.564,
      "eval_steps_per_second": 1.101,
      "step": 380000
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.0001,
      "loss": 1.4023,
      "step": 380500
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.0001,
      "loss": 1.4031,
      "step": 381000
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.0001,
      "loss": 1.4191,
      "step": 381500
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.0001,
      "loss": 1.423,
      "step": 382000
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.0001,
      "loss": 1.3902,
      "step": 382500
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.0001,
      "loss": 1.4145,
      "step": 383000
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.0001,
      "loss": 1.4165,
      "step": 383500
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.0001,
      "loss": 1.4183,
      "step": 384000
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.0001,
      "loss": 1.4394,
      "step": 384500
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.0001,
      "loss": 1.4208,
      "step": 385000
    },
    {
      "epoch": 1.37,
      "eval_bleu": 16.2118,
      "eval_gen_len": 17.241,
      "eval_loss": 1.1445575952529907,
      "eval_runtime": 169.8554,
      "eval_samples_per_second": 17.662,
      "eval_steps_per_second": 1.107,
      "step": 385000
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.0001,
      "loss": 1.416,
      "step": 385500
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0001,
      "loss": 1.4094,
      "step": 386000
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0001,
      "loss": 1.4065,
      "step": 386500
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0001,
      "loss": 1.4112,
      "step": 387000
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0001,
      "loss": 1.4252,
      "step": 387500
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0001,
      "loss": 1.41,
      "step": 388000
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0001,
      "loss": 1.4171,
      "step": 388500
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.0001,
      "loss": 1.4008,
      "step": 389000
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.0001,
      "loss": 1.4014,
      "step": 389500
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.0001,
      "loss": 1.4071,
      "step": 390000
    },
    {
      "epoch": 1.39,
      "eval_bleu": 15.9784,
      "eval_gen_len": 17.2343,
      "eval_loss": 1.142768144607544,
      "eval_runtime": 167.0855,
      "eval_samples_per_second": 17.955,
      "eval_steps_per_second": 1.125,
      "step": 390000
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.0001,
      "loss": 1.416,
      "step": 390500
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.0001,
      "loss": 1.4108,
      "step": 391000
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.0001,
      "loss": 1.4192,
      "step": 391500
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.0001,
      "loss": 1.4026,
      "step": 392000
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.0001,
      "loss": 1.3872,
      "step": 392500
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.0001,
      "loss": 1.4187,
      "step": 393000
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.0001,
      "loss": 1.425,
      "step": 393500
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.0001,
      "loss": 1.4193,
      "step": 394000
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.0001,
      "loss": 1.4201,
      "step": 394500
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.0001,
      "loss": 1.4101,
      "step": 395000
    },
    {
      "epoch": 1.41,
      "eval_bleu": 16.1531,
      "eval_gen_len": 17.245,
      "eval_loss": 1.1416969299316406,
      "eval_runtime": 167.4533,
      "eval_samples_per_second": 17.915,
      "eval_steps_per_second": 1.123,
      "step": 395000
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.0001,
      "loss": 1.4049,
      "step": 395500
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.0001,
      "loss": 1.4003,
      "step": 396000
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.0001,
      "loss": 1.4256,
      "step": 396500
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.0001,
      "loss": 1.3967,
      "step": 397000
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.0001,
      "loss": 1.3854,
      "step": 397500
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.0001,
      "loss": 1.4274,
      "step": 398000
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.0001,
      "loss": 1.4198,
      "step": 398500
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.0001,
      "loss": 1.4008,
      "step": 399000
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.0001,
      "loss": 1.4243,
      "step": 399500
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.0001,
      "loss": 1.4123,
      "step": 400000
    },
    {
      "epoch": 1.43,
      "eval_bleu": 16.2695,
      "eval_gen_len": 17.2143,
      "eval_loss": 1.1432126760482788,
      "eval_runtime": 166.233,
      "eval_samples_per_second": 18.047,
      "eval_steps_per_second": 1.131,
      "step": 400000
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.0001,
      "loss": 1.4136,
      "step": 400500
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.0001,
      "loss": 1.416,
      "step": 401000
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.0001,
      "loss": 1.4148,
      "step": 401500
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.0001,
      "loss": 1.407,
      "step": 402000
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.0001,
      "loss": 1.3971,
      "step": 402500
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.0001,
      "loss": 1.4253,
      "step": 403000
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.0001,
      "loss": 1.3834,
      "step": 403500
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.0001,
      "loss": 1.414,
      "step": 404000
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.0001,
      "loss": 1.3787,
      "step": 404500
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.0001,
      "loss": 1.391,
      "step": 405000
    },
    {
      "epoch": 1.44,
      "eval_bleu": 16.2607,
      "eval_gen_len": 17.2533,
      "eval_loss": 1.143123745918274,
      "eval_runtime": 167.1752,
      "eval_samples_per_second": 17.945,
      "eval_steps_per_second": 1.125,
      "step": 405000
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.0001,
      "loss": 1.3761,
      "step": 405500
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.0001,
      "loss": 1.4276,
      "step": 406000
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.0001,
      "loss": 1.386,
      "step": 406500
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.0001,
      "loss": 1.4176,
      "step": 407000
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.0001,
      "loss": 1.412,
      "step": 407500
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.0001,
      "loss": 1.4092,
      "step": 408000
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0001,
      "loss": 1.4312,
      "step": 408500
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0001,
      "loss": 1.3859,
      "step": 409000
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0001,
      "loss": 1.3953,
      "step": 409500
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0001,
      "loss": 1.4389,
      "step": 410000
    },
    {
      "epoch": 1.46,
      "eval_bleu": 16.0845,
      "eval_gen_len": 17.2303,
      "eval_loss": 1.1442809104919434,
      "eval_runtime": 166.7494,
      "eval_samples_per_second": 17.991,
      "eval_steps_per_second": 1.127,
      "step": 410000
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0001,
      "loss": 1.3919,
      "step": 410500
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0001,
      "loss": 1.4085,
      "step": 411000
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.0001,
      "loss": 1.4186,
      "step": 411500
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.0001,
      "loss": 1.3884,
      "step": 412000
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.0001,
      "loss": 1.3986,
      "step": 412500
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.0001,
      "loss": 1.4113,
      "step": 413000
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.0001,
      "loss": 1.4169,
      "step": 413500
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.0001,
      "loss": 1.4122,
      "step": 414000
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.0001,
      "loss": 1.3793,
      "step": 414500
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.0001,
      "loss": 1.4326,
      "step": 415000
    },
    {
      "epoch": 1.48,
      "eval_bleu": 16.319,
      "eval_gen_len": 17.245,
      "eval_loss": 1.1430983543395996,
      "eval_runtime": 167.8646,
      "eval_samples_per_second": 17.872,
      "eval_steps_per_second": 1.12,
      "step": 415000
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.0001,
      "loss": 1.4042,
      "step": 415500
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.0001,
      "loss": 1.4103,
      "step": 416000
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.0001,
      "loss": 1.4212,
      "step": 416500
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.0001,
      "loss": 1.3778,
      "step": 417000
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.0001,
      "loss": 1.4011,
      "step": 417500
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.0001,
      "loss": 1.4153,
      "step": 418000
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.0001,
      "loss": 1.3995,
      "step": 418500
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.0001,
      "loss": 1.3995,
      "step": 419000
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.0001,
      "loss": 1.4132,
      "step": 419500
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.0001,
      "loss": 1.4132,
      "step": 420000
    },
    {
      "epoch": 1.5,
      "eval_bleu": 16.2334,
      "eval_gen_len": 17.2527,
      "eval_loss": 1.1432762145996094,
      "eval_runtime": 170.7401,
      "eval_samples_per_second": 17.571,
      "eval_steps_per_second": 1.101,
      "step": 420000
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.0001,
      "loss": 1.3825,
      "step": 420500
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.0001,
      "loss": 1.4013,
      "step": 421000
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.0001,
      "loss": 1.3993,
      "step": 421500
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.0001,
      "loss": 1.3843,
      "step": 422000
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.0001,
      "loss": 1.4022,
      "step": 422500
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.0001,
      "loss": 1.404,
      "step": 423000
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.0001,
      "loss": 1.4008,
      "step": 423500
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.0001,
      "loss": 1.3899,
      "step": 424000
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.0001,
      "loss": 1.3909,
      "step": 424500
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.0001,
      "loss": 1.4039,
      "step": 425000
    },
    {
      "epoch": 1.51,
      "eval_bleu": 16.23,
      "eval_gen_len": 17.222,
      "eval_loss": 1.140016794204712,
      "eval_runtime": 168.2393,
      "eval_samples_per_second": 17.832,
      "eval_steps_per_second": 1.117,
      "step": 425000
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.0001,
      "loss": 1.4175,
      "step": 425500
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.0001,
      "loss": 1.4198,
      "step": 426000
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.0001,
      "loss": 1.3985,
      "step": 426500
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.0001,
      "loss": 1.4103,
      "step": 427000
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.0001,
      "loss": 1.3871,
      "step": 427500
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0001,
      "loss": 1.3833,
      "step": 428000
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0001,
      "loss": 1.4081,
      "step": 428500
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0001,
      "loss": 1.3838,
      "step": 429000
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0001,
      "loss": 1.4182,
      "step": 429500
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0001,
      "loss": 1.3988,
      "step": 430000
    },
    {
      "epoch": 1.53,
      "eval_bleu": 16.4534,
      "eval_gen_len": 17.2187,
      "eval_loss": 1.1407719850540161,
      "eval_runtime": 170.1126,
      "eval_samples_per_second": 17.635,
      "eval_steps_per_second": 1.105,
      "step": 430000
    }
  ],
  "max_steps": 1127200,
  "num_train_epochs": 4,
  "total_flos": 3.6860938066490204e+18,
  "trial_name": null,
  "trial_params": null
}