{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "global_step": 409890,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 5e-05,
      "loss": 2.0235,
      "step": 500
    },
    {
      "epoch": 0.01,
      "learning_rate": 5e-05,
      "loss": 1.871,
      "step": 1000
    },
    {
      "epoch": 0.01,
      "learning_rate": 5e-05,
      "loss": 1.7723,
      "step": 1500
    },
    {
      "epoch": 0.01,
      "learning_rate": 5e-05,
      "loss": 1.7105,
      "step": 2000
    },
    {
      "epoch": 0.02,
      "learning_rate": 5e-05,
      "loss": 1.6615,
      "step": 2500
    },
    {
      "epoch": 0.02,
      "learning_rate": 5e-05,
      "loss": 1.6266,
      "step": 3000
    },
    {
      "epoch": 0.03,
      "learning_rate": 5e-05,
      "loss": 1.5964,
      "step": 3500
    },
    {
      "epoch": 0.03,
      "learning_rate": 5e-05,
      "loss": 1.5695,
      "step": 4000
    },
    {
      "epoch": 0.03,
      "learning_rate": 5e-05,
      "loss": 1.5454,
      "step": 4500
    },
    {
      "epoch": 0.04,
      "learning_rate": 5e-05,
      "loss": 1.529,
      "step": 5000
    },
    {
      "epoch": 0.04,
      "learning_rate": 5e-05,
      "loss": 1.505,
      "step": 5500
    },
    {
      "epoch": 0.04,
      "learning_rate": 5e-05,
      "loss": 1.4823,
      "step": 6000
    },
    {
      "epoch": 0.05,
      "learning_rate": 5e-05,
      "loss": 1.463,
      "step": 6500
    },
    {
      "epoch": 0.05,
      "learning_rate": 5e-05,
      "loss": 1.4563,
      "step": 7000
    },
    {
      "epoch": 0.05,
      "learning_rate": 5e-05,
      "loss": 1.4428,
      "step": 7500
    },
    {
      "epoch": 0.06,
      "learning_rate": 5e-05,
      "loss": 1.4384,
      "step": 8000
    },
    {
      "epoch": 0.06,
      "learning_rate": 5e-05,
      "loss": 1.4163,
      "step": 8500
    },
    {
      "epoch": 0.07,
      "learning_rate": 5e-05,
      "loss": 1.4174,
      "step": 9000
    },
    {
      "epoch": 0.07,
      "learning_rate": 5e-05,
      "loss": 1.407,
      "step": 9500
    },
    {
      "epoch": 0.07,
      "learning_rate": 5e-05,
      "loss": 1.3931,
      "step": 10000
    },
    {
      "epoch": 0.08,
      "learning_rate": 5e-05,
      "loss": 1.3871,
      "step": 10500
    },
    {
      "epoch": 0.08,
      "learning_rate": 5e-05,
      "loss": 1.3877,
      "step": 11000
    },
    {
      "epoch": 0.08,
      "learning_rate": 5e-05,
      "loss": 1.3757,
      "step": 11500
    },
    {
      "epoch": 0.09,
      "learning_rate": 5e-05,
      "loss": 1.3676,
      "step": 12000
    },
    {
      "epoch": 0.09,
      "learning_rate": 5e-05,
      "loss": 1.3567,
      "step": 12500
    },
    {
      "epoch": 0.1,
      "learning_rate": 5e-05,
      "loss": 1.3547,
      "step": 13000
    },
    {
      "epoch": 0.1,
      "learning_rate": 5e-05,
      "loss": 1.3508,
      "step": 13500
    },
    {
      "epoch": 0.1,
      "learning_rate": 5e-05,
      "loss": 1.3469,
      "step": 14000
    },
    {
      "epoch": 0.11,
      "learning_rate": 5e-05,
      "loss": 1.3389,
      "step": 14500
    },
    {
      "epoch": 0.11,
      "learning_rate": 5e-05,
      "loss": 1.3319,
      "step": 15000
    },
    {
      "epoch": 0.11,
      "learning_rate": 5e-05,
      "loss": 1.3231,
      "step": 15500
    },
    {
      "epoch": 0.12,
      "learning_rate": 5e-05,
      "loss": 1.325,
      "step": 16000
    },
    {
      "epoch": 0.12,
      "learning_rate": 5e-05,
      "loss": 1.3141,
      "step": 16500
    },
    {
      "epoch": 0.12,
      "learning_rate": 5e-05,
      "loss": 1.3211,
      "step": 17000
    },
    {
      "epoch": 0.13,
      "learning_rate": 5e-05,
      "loss": 1.3068,
      "step": 17500
    },
    {
      "epoch": 0.13,
      "learning_rate": 5e-05,
      "loss": 1.3025,
      "step": 18000
    },
    {
      "epoch": 0.14,
      "learning_rate": 5e-05,
      "loss": 1.2992,
      "step": 18500
    },
    {
      "epoch": 0.14,
      "learning_rate": 5e-05,
      "loss": 1.3018,
      "step": 19000
    },
    {
      "epoch": 0.14,
      "learning_rate": 5e-05,
      "loss": 1.2995,
      "step": 19500
    },
    {
      "epoch": 0.15,
      "learning_rate": 5e-05,
      "loss": 1.2837,
      "step": 20000
    },
    {
      "epoch": 0.15,
      "learning_rate": 5e-05,
      "loss": 1.282,
      "step": 20500
    },
    {
      "epoch": 0.15,
      "learning_rate": 5e-05,
      "loss": 1.2885,
      "step": 21000
    },
    {
      "epoch": 0.16,
      "learning_rate": 5e-05,
      "loss": 1.2807,
      "step": 21500
    },
    {
      "epoch": 0.16,
      "learning_rate": 5e-05,
      "loss": 1.2813,
      "step": 22000
    },
    {
      "epoch": 0.16,
      "learning_rate": 5e-05,
      "loss": 1.2784,
      "step": 22500
    },
    {
      "epoch": 0.17,
      "learning_rate": 5e-05,
      "loss": 1.2701,
      "step": 23000
    },
    {
      "epoch": 0.17,
      "learning_rate": 5e-05,
      "loss": 1.2683,
      "step": 23500
    },
    {
      "epoch": 0.18,
      "learning_rate": 5e-05,
      "loss": 1.2728,
      "step": 24000
    },
    {
      "epoch": 0.18,
      "learning_rate": 5e-05,
      "loss": 1.2631,
      "step": 24500
    },
    {
      "epoch": 0.18,
      "learning_rate": 5e-05,
      "loss": 1.2591,
      "step": 25000
    },
    {
      "epoch": 0.19,
      "learning_rate": 5e-05,
      "loss": 1.2545,
      "step": 25500
    },
    {
      "epoch": 0.19,
      "learning_rate": 5e-05,
      "loss": 1.2521,
      "step": 26000
    },
    {
      "epoch": 0.19,
      "learning_rate": 5e-05,
      "loss": 1.2515,
      "step": 26500
    },
    {
      "epoch": 0.2,
      "learning_rate": 5e-05,
      "loss": 1.2482,
      "step": 27000
    },
    {
      "epoch": 0.2,
      "learning_rate": 5e-05,
      "loss": 1.2478,
      "step": 27500
    },
    {
      "epoch": 0.2,
      "learning_rate": 5e-05,
      "loss": 1.2418,
      "step": 28000
    },
    {
      "epoch": 0.21,
      "learning_rate": 5e-05,
      "loss": 1.2429,
      "step": 28500
    },
    {
      "epoch": 0.21,
      "learning_rate": 5e-05,
      "loss": 1.2419,
      "step": 29000
    },
    {
      "epoch": 0.22,
      "learning_rate": 5e-05,
      "loss": 1.2349,
      "step": 29500
    },
    {
      "epoch": 0.22,
      "learning_rate": 5e-05,
      "loss": 1.2397,
      "step": 30000
    },
    {
      "epoch": 0.22,
      "learning_rate": 5e-05,
      "loss": 1.2308,
      "step": 30500
    },
    {
      "epoch": 0.23,
      "learning_rate": 5e-05,
      "loss": 1.2292,
      "step": 31000
    },
    {
      "epoch": 0.23,
      "learning_rate": 5e-05,
      "loss": 1.2251,
      "step": 31500
    },
    {
      "epoch": 0.23,
      "learning_rate": 5e-05,
      "loss": 1.2331,
      "step": 32000
    },
    {
      "epoch": 0.24,
      "learning_rate": 5e-05,
      "loss": 1.2259,
      "step": 32500
    },
    {
      "epoch": 0.24,
      "learning_rate": 5e-05,
      "loss": 1.2273,
      "step": 33000
    },
    {
      "epoch": 0.25,
      "learning_rate": 5e-05,
      "loss": 1.2195,
      "step": 33500
    },
    {
      "epoch": 0.25,
      "learning_rate": 5e-05,
      "loss": 1.2194,
      "step": 34000
    },
    {
      "epoch": 0.25,
      "learning_rate": 5e-05,
      "loss": 1.2178,
      "step": 34500
    },
    {
      "epoch": 0.26,
      "learning_rate": 5e-05,
      "loss": 1.2169,
      "step": 35000
    },
    {
      "epoch": 0.26,
      "learning_rate": 5e-05,
      "loss": 1.2171,
      "step": 35500
    },
    {
      "epoch": 0.26,
      "learning_rate": 5e-05,
      "loss": 1.2041,
      "step": 36000
    },
    {
      "epoch": 0.27,
      "learning_rate": 5e-05,
      "loss": 1.2183,
      "step": 36500
    },
    {
      "epoch": 0.27,
      "learning_rate": 5e-05,
      "loss": 1.2112,
      "step": 37000
    },
    {
      "epoch": 0.27,
      "learning_rate": 5e-05,
      "loss": 1.206,
      "step": 37500
    },
    {
      "epoch": 0.28,
      "learning_rate": 5e-05,
      "loss": 1.1992,
      "step": 38000
    },
    {
      "epoch": 0.28,
      "learning_rate": 5e-05,
      "loss": 1.2065,
      "step": 38500
    },
    {
      "epoch": 0.29,
      "learning_rate": 5e-05,
      "loss": 1.2097,
      "step": 39000
    },
    {
      "epoch": 0.29,
      "learning_rate": 5e-05,
      "loss": 1.2004,
      "step": 39500
    },
    {
      "epoch": 0.29,
      "learning_rate": 5e-05,
      "loss": 1.2041,
      "step": 40000
    },
    {
      "epoch": 0.3,
      "learning_rate": 5e-05,
      "loss": 1.2011,
      "step": 40500
    },
    {
      "epoch": 0.3,
      "learning_rate": 5e-05,
      "loss": 1.2008,
      "step": 41000
    },
    {
      "epoch": 0.3,
      "learning_rate": 5e-05,
      "loss": 1.2017,
      "step": 41500
    },
    {
      "epoch": 0.31,
      "learning_rate": 5e-05,
      "loss": 1.2022,
      "step": 42000
    },
    {
      "epoch": 0.31,
      "learning_rate": 5e-05,
      "loss": 1.1948,
      "step": 42500
    },
    {
      "epoch": 0.31,
      "learning_rate": 5e-05,
      "loss": 1.197,
      "step": 43000
    },
    {
      "epoch": 0.32,
      "learning_rate": 5e-05,
      "loss": 1.1886,
      "step": 43500
    },
    {
      "epoch": 0.32,
      "learning_rate": 5e-05,
      "loss": 1.1906,
      "step": 44000
    },
    {
      "epoch": 0.33,
      "learning_rate": 5e-05,
      "loss": 1.1881,
      "step": 44500
    },
    {
      "epoch": 0.33,
      "learning_rate": 5e-05,
      "loss": 1.1908,
      "step": 45000
    },
    {
      "epoch": 0.33,
      "learning_rate": 5e-05,
      "loss": 1.1871,
      "step": 45500
    },
    {
      "epoch": 0.34,
      "learning_rate": 5e-05,
      "loss": 1.1857,
      "step": 46000
    },
    {
      "epoch": 0.34,
      "learning_rate": 5e-05,
      "loss": 1.1821,
      "step": 46500
    },
    {
      "epoch": 0.34,
      "learning_rate": 5e-05,
      "loss": 1.1818,
      "step": 47000
    },
    {
      "epoch": 0.35,
      "learning_rate": 5e-05,
      "loss": 1.1781,
      "step": 47500
    },
    {
      "epoch": 0.35,
      "learning_rate": 5e-05,
      "loss": 1.1797,
      "step": 48000
    },
    {
      "epoch": 0.35,
      "learning_rate": 5e-05,
      "loss": 1.1739,
      "step": 48500
    },
    {
      "epoch": 0.36,
      "learning_rate": 5e-05,
      "loss": 1.1817,
      "step": 49000
    },
    {
      "epoch": 0.36,
      "learning_rate": 5e-05,
      "loss": 1.1794,
      "step": 49500
    },
    {
      "epoch": 0.37,
      "learning_rate": 5e-05,
      "loss": 1.1757,
      "step": 50000
    },
    {
      "epoch": 0.37,
      "learning_rate": 5e-05,
      "loss": 1.1701,
      "step": 50500
    },
    {
      "epoch": 0.37,
      "learning_rate": 5e-05,
      "loss": 1.1687,
      "step": 51000
    },
    {
      "epoch": 0.38,
      "learning_rate": 5e-05,
      "loss": 1.1744,
      "step": 51500
    },
    {
      "epoch": 0.38,
      "learning_rate": 5e-05,
      "loss": 1.1692,
      "step": 52000
    },
    {
      "epoch": 0.38,
      "learning_rate": 5e-05,
      "loss": 1.1636,
      "step": 52500
    },
    {
      "epoch": 0.39,
      "learning_rate": 5e-05,
      "loss": 1.1683,
      "step": 53000
    },
    {
      "epoch": 0.39,
      "learning_rate": 5e-05,
      "loss": 1.1601,
      "step": 53500
    },
    {
      "epoch": 0.4,
      "learning_rate": 5e-05,
      "loss": 1.1664,
      "step": 54000
    },
    {
      "epoch": 0.4,
      "learning_rate": 5e-05,
      "loss": 1.1636,
      "step": 54500
    },
    {
      "epoch": 0.4,
      "learning_rate": 5e-05,
      "loss": 1.1635,
      "step": 55000
    },
    {
      "epoch": 0.41,
      "learning_rate": 5e-05,
      "loss": 1.1616,
      "step": 55500
    },
    {
      "epoch": 0.41,
      "learning_rate": 5e-05,
      "loss": 1.1593,
      "step": 56000
    },
    {
      "epoch": 0.41,
      "learning_rate": 5e-05,
      "loss": 1.1598,
      "step": 56500
    },
    {
      "epoch": 0.42,
      "learning_rate": 5e-05,
      "loss": 1.1582,
      "step": 57000
    },
    {
      "epoch": 0.42,
      "learning_rate": 5e-05,
      "loss": 1.1611,
      "step": 57500
    },
    {
      "epoch": 0.42,
      "learning_rate": 5e-05,
      "loss": 1.1571,
      "step": 58000
    },
    {
      "epoch": 0.43,
      "learning_rate": 5e-05,
      "loss": 1.1539,
      "step": 58500
    },
    {
      "epoch": 0.43,
      "learning_rate": 5e-05,
      "loss": 1.1506,
      "step": 59000
    },
    {
      "epoch": 0.44,
      "learning_rate": 5e-05,
      "loss": 1.154,
      "step": 59500
    },
    {
      "epoch": 0.44,
      "learning_rate": 5e-05,
      "loss": 1.1522,
      "step": 60000
    },
    {
      "epoch": 0.44,
      "learning_rate": 5e-05,
      "loss": 1.1549,
      "step": 60500
    },
    {
      "epoch": 0.45,
      "learning_rate": 5e-05,
      "loss": 1.1501,
      "step": 61000
    },
    {
      "epoch": 0.45,
      "learning_rate": 5e-05,
      "loss": 1.1459,
      "step": 61500
    },
    {
      "epoch": 0.45,
      "learning_rate": 5e-05,
      "loss": 1.1544,
      "step": 62000
    },
    {
      "epoch": 0.46,
      "learning_rate": 5e-05,
      "loss": 1.1529,
      "step": 62500
    },
    {
      "epoch": 0.46,
      "learning_rate": 5e-05,
      "loss": 1.1539,
      "step": 63000
    },
    {
      "epoch": 0.46,
      "learning_rate": 5e-05,
      "loss": 1.151,
      "step": 63500
    },
    {
      "epoch": 0.47,
      "learning_rate": 5e-05,
      "loss": 1.1491,
      "step": 64000
    },
    {
      "epoch": 0.47,
      "eval_accuracy": 0.7652064452221591,
      "eval_loss": 1.1646534204483032,
      "eval_runtime": 32.5535,
      "eval_samples_per_second": 412.121,
      "eval_steps_per_second": 8.601,
      "step": 64000
    },
    {
      "epoch": 0.47,
      "learning_rate": 5e-05,
      "loss": 1.1535,
      "step": 64500
    },
    {
      "epoch": 0.48,
      "learning_rate": 5e-05,
      "loss": 1.1521,
      "step": 65000
    },
    {
      "epoch": 0.48,
      "learning_rate": 5e-05,
      "loss": 1.1486,
      "step": 65500
    },
    {
      "epoch": 0.48,
      "learning_rate": 5e-05,
      "loss": 1.1498,
      "step": 66000
    },
    {
      "epoch": 0.49,
      "learning_rate": 5e-05,
      "loss": 1.1513,
      "step": 66500
    },
    {
      "epoch": 0.49,
      "learning_rate": 5e-05,
      "loss": 1.144,
      "step": 67000
    },
    {
      "epoch": 0.49,
      "learning_rate": 5e-05,
      "loss": 1.1417,
      "step": 67500
    },
    {
      "epoch": 0.5,
      "learning_rate": 5e-05,
      "loss": 1.1414,
      "step": 68000
    },
    {
      "epoch": 0.5,
      "learning_rate": 5e-05,
      "loss": 1.1416,
      "step": 68500
    },
    {
      "epoch": 0.51,
      "learning_rate": 5e-05,
      "loss": 1.1459,
      "step": 69000
    },
    {
      "epoch": 0.51,
      "learning_rate": 5e-05,
      "loss": 1.1476,
      "step": 69500
    },
    {
      "epoch": 0.51,
      "learning_rate": 5e-05,
      "loss": 1.1444,
      "step": 70000
    },
    {
      "epoch": 0.52,
      "learning_rate": 5e-05,
      "loss": 1.1402,
      "step": 70500
    },
    {
      "epoch": 0.52,
      "learning_rate": 5e-05,
      "loss": 1.1439,
      "step": 71000
    },
    {
      "epoch": 0.52,
      "learning_rate": 5e-05,
      "loss": 1.1459,
      "step": 71500
    },
    {
      "epoch": 0.53,
      "learning_rate": 5e-05,
      "loss": 1.1338,
      "step": 72000
    },
    {
      "epoch": 0.53,
      "learning_rate": 5e-05,
      "loss": 1.137,
      "step": 72500
    },
    {
      "epoch": 0.53,
      "learning_rate": 5e-05,
      "loss": 1.1375,
      "step": 73000
    },
    {
      "epoch": 0.54,
      "learning_rate": 5e-05,
      "loss": 1.1322,
      "step": 73500
    },
    {
      "epoch": 0.54,
      "learning_rate": 5e-05,
      "loss": 1.1387,
      "step": 74000
    },
    {
      "epoch": 0.55,
      "learning_rate": 5e-05,
      "loss": 1.1421,
      "step": 74500
    },
    {
      "epoch": 0.55,
      "learning_rate": 5e-05,
      "loss": 1.1318,
      "step": 75000
    },
    {
      "epoch": 0.55,
      "learning_rate": 5e-05,
      "loss": 1.1247,
      "step": 75500
    },
    {
      "epoch": 0.56,
      "learning_rate": 5e-05,
      "loss": 1.1313,
      "step": 76000
    },
    {
      "epoch": 0.56,
      "learning_rate": 5e-05,
      "loss": 1.1323,
      "step": 76500
    },
    {
      "epoch": 0.56,
      "learning_rate": 5e-05,
      "loss": 1.1308,
      "step": 77000
    },
    {
      "epoch": 0.57,
      "learning_rate": 5e-05,
      "loss": 1.1294,
      "step": 77500
    },
    {
      "epoch": 0.57,
      "learning_rate": 5e-05,
      "loss": 1.1265,
      "step": 78000
    },
    {
      "epoch": 0.57,
      "learning_rate": 5e-05,
      "loss": 1.1334,
      "step": 78500
    },
    {
      "epoch": 0.58,
      "learning_rate": 5e-05,
      "loss": 1.1273,
      "step": 79000
    },
    {
      "epoch": 0.58,
      "learning_rate": 5e-05,
      "loss": 1.133,
      "step": 79500
    },
    {
      "epoch": 0.59,
      "learning_rate": 5e-05,
      "loss": 1.1316,
      "step": 80000
    },
    {
      "epoch": 0.59,
      "learning_rate": 5e-05,
      "loss": 1.1279,
      "step": 80500
    },
    {
      "epoch": 0.59,
      "learning_rate": 5e-05,
      "loss": 1.1339,
      "step": 81000
    },
    {
      "epoch": 0.6,
      "learning_rate": 5e-05,
      "loss": 1.1336,
      "step": 81500
    },
    {
      "epoch": 0.6,
      "learning_rate": 5e-05,
      "loss": 1.135,
      "step": 82000
    },
    {
      "epoch": 0.6,
      "learning_rate": 5e-05,
      "loss": 1.1345,
      "step": 82500
    },
    {
      "epoch": 0.61,
      "learning_rate": 5e-05,
      "loss": 1.1287,
      "step": 83000
    },
    {
      "epoch": 0.61,
      "learning_rate": 5e-05,
      "loss": 1.1237,
      "step": 83500
    },
    {
      "epoch": 0.61,
      "learning_rate": 5e-05,
      "loss": 1.1314,
      "step": 84000
    },
    {
      "epoch": 0.62,
      "learning_rate": 5e-05,
      "loss": 1.1262,
      "step": 84500
    },
    {
      "epoch": 0.62,
      "learning_rate": 5e-05,
      "loss": 1.1251,
      "step": 85000
    },
    {
      "epoch": 0.63,
      "learning_rate": 5e-05,
      "loss": 1.1226,
      "step": 85500
    },
    {
      "epoch": 0.63,
      "learning_rate": 5e-05,
      "loss": 1.1208,
      "step": 86000
    },
    {
      "epoch": 0.63,
      "learning_rate": 5e-05,
      "loss": 1.1155,
      "step": 86500
    },
    {
      "epoch": 0.64,
      "learning_rate": 5e-05,
      "loss": 1.1196,
      "step": 87000
    },
    {
      "epoch": 0.64,
      "learning_rate": 5e-05,
      "loss": 1.1153,
      "step": 87500
    },
    {
      "epoch": 0.64,
      "learning_rate": 5e-05,
      "loss": 1.1168,
      "step": 88000
    },
    {
      "epoch": 0.65,
      "learning_rate": 5e-05,
      "loss": 1.1131,
      "step": 88500
    },
    {
      "epoch": 0.65,
      "learning_rate": 5e-05,
      "loss": 1.1106,
      "step": 89000
    },
    {
      "epoch": 0.66,
      "learning_rate": 5e-05,
      "loss": 1.1238,
      "step": 89500
    },
    {
      "epoch": 0.66,
      "learning_rate": 5e-05,
      "loss": 1.1121,
      "step": 90000
    },
    {
      "epoch": 0.66,
      "learning_rate": 5e-05,
      "loss": 1.108,
      "step": 90500
    },
    {
      "epoch": 0.67,
      "learning_rate": 5e-05,
      "loss": 1.1046,
      "step": 91000
    },
    {
      "epoch": 0.67,
      "learning_rate": 5e-05,
      "loss": 1.115,
      "step": 91500
    },
    {
      "epoch": 0.67,
      "learning_rate": 5e-05,
      "loss": 1.1154,
      "step": 92000
    },
    {
      "epoch": 0.68,
      "learning_rate": 5e-05,
      "loss": 1.1146,
      "step": 92500
    },
    {
      "epoch": 0.68,
      "learning_rate": 5e-05,
      "loss": 1.117,
      "step": 93000
    },
    {
      "epoch": 0.68,
      "learning_rate": 5e-05,
      "loss": 1.1117,
      "step": 93500
    },
    {
      "epoch": 0.69,
      "learning_rate": 5e-05,
      "loss": 1.1052,
      "step": 94000
    },
    {
      "epoch": 0.69,
      "learning_rate": 5e-05,
      "loss": 1.1126,
      "step": 94500
    },
    {
      "epoch": 0.7,
      "learning_rate": 5e-05,
      "loss": 1.1102,
      "step": 95000
    },
    {
      "epoch": 0.7,
      "learning_rate": 5e-05,
      "loss": 1.1076,
      "step": 95500
    },
    {
      "epoch": 0.7,
      "learning_rate": 5e-05,
      "loss": 1.1124,
      "step": 96000
    },
    {
      "epoch": 0.71,
      "learning_rate": 5e-05,
      "loss": 1.1104,
      "step": 96500
    },
    {
      "epoch": 0.71,
      "learning_rate": 5e-05,
      "loss": 1.115,
      "step": 97000
    },
    {
      "epoch": 0.71,
      "learning_rate": 5e-05,
      "loss": 1.1071,
      "step": 97500
    },
    {
      "epoch": 0.72,
      "learning_rate": 5e-05,
      "loss": 1.1108,
      "step": 98000
    },
    {
      "epoch": 0.72,
      "learning_rate": 5e-05,
      "loss": 1.1079,
      "step": 98500
    },
    {
      "epoch": 0.72,
      "learning_rate": 5e-05,
      "loss": 1.1019,
      "step": 99000
    },
    {
      "epoch": 0.73,
      "learning_rate": 5e-05,
      "loss": 1.103,
      "step": 99500
    },
    {
      "epoch": 0.73,
      "learning_rate": 5e-05,
      "loss": 1.1,
      "step": 100000
    },
    {
      "epoch": 0.74,
      "learning_rate": 5e-05,
      "loss": 1.1006,
      "step": 100500
    },
    {
      "epoch": 0.74,
      "learning_rate": 5e-05,
      "loss": 1.103,
      "step": 101000
    },
    {
      "epoch": 0.74,
      "learning_rate": 5e-05,
      "loss": 1.1005,
      "step": 101500
    },
    {
      "epoch": 0.75,
      "learning_rate": 5e-05,
      "loss": 1.1016,
      "step": 102000
    },
    {
      "epoch": 0.75,
      "learning_rate": 5e-05,
      "loss": 1.0999,
      "step": 102500
    },
    {
      "epoch": 0.75,
      "learning_rate": 5e-05,
      "loss": 1.101,
      "step": 103000
    },
    {
      "epoch": 0.76,
      "learning_rate": 5e-05,
      "loss": 1.1041,
      "step": 103500
    },
    {
      "epoch": 0.76,
      "learning_rate": 5e-05,
      "loss": 1.1,
      "step": 104000
    },
    {
      "epoch": 0.76,
      "learning_rate": 5e-05,
      "loss": 1.109,
      "step": 104500
    },
    {
      "epoch": 0.77,
      "learning_rate": 5e-05,
      "loss": 1.105,
      "step": 105000
    },
    {
      "epoch": 0.77,
      "learning_rate": 5e-05,
      "loss": 1.1031,
      "step": 105500
    },
    {
      "epoch": 0.78,
      "learning_rate": 5e-05,
      "loss": 1.0965,
      "step": 106000
    },
    {
      "epoch": 0.78,
      "learning_rate": 5e-05,
      "loss": 1.0958,
      "step": 106500
    },
    {
      "epoch": 0.78,
      "learning_rate": 5e-05,
      "loss": 1.0959,
      "step": 107000
    },
    {
      "epoch": 0.79,
      "learning_rate": 5e-05,
      "loss": 1.1022,
      "step": 107500
    },
    {
      "epoch": 0.79,
      "learning_rate": 5e-05,
      "loss": 1.0999,
      "step": 108000
    },
    {
      "epoch": 0.79,
      "learning_rate": 5e-05,
      "loss": 1.1004,
      "step": 108500
    },
    {
      "epoch": 0.8,
      "learning_rate": 5e-05,
      "loss": 1.103,
      "step": 109000
    },
    {
      "epoch": 0.8,
      "learning_rate": 5e-05,
      "loss": 1.0974,
      "step": 109500
    },
    {
      "epoch": 0.81,
      "learning_rate": 5e-05,
      "loss": 1.0986,
      "step": 110000
    },
    {
      "epoch": 0.81,
      "learning_rate": 5e-05,
      "loss": 1.098,
      "step": 110500
    },
    {
      "epoch": 0.81,
      "learning_rate": 5e-05,
      "loss": 1.0951,
      "step": 111000
    },
    {
      "epoch": 0.82,
      "learning_rate": 5e-05,
      "loss": 1.0915,
      "step": 111500
    },
    {
      "epoch": 0.82,
      "learning_rate": 5e-05,
      "loss": 1.0952,
      "step": 112000
    },
    {
      "epoch": 0.82,
      "learning_rate": 5e-05,
      "loss": 1.1,
      "step": 112500
    },
    {
      "epoch": 0.83,
      "learning_rate": 5e-05,
      "loss": 1.099,
      "step": 113000
    },
    {
      "epoch": 0.83,
      "learning_rate": 5e-05,
      "loss": 1.0868,
      "step": 113500
    },
    {
      "epoch": 0.83,
      "learning_rate": 5e-05,
      "loss": 1.0906,
      "step": 114000
    },
    {
      "epoch": 0.84,
      "learning_rate": 5e-05,
      "loss": 1.0844,
      "step": 114500
    },
    {
      "epoch": 0.84,
      "learning_rate": 5e-05,
      "loss": 1.0868,
      "step": 115000
    },
    {
      "epoch": 0.85,
      "learning_rate": 5e-05,
      "loss": 1.0802,
      "step": 115500
    },
    {
      "epoch": 0.85,
      "learning_rate": 5e-05,
      "loss": 1.0863,
      "step": 116000
    },
    {
      "epoch": 0.85,
      "learning_rate": 5e-05,
      "loss": 1.0922,
      "step": 116500
    },
    {
      "epoch": 0.86,
      "learning_rate": 5e-05,
      "loss": 1.0927,
      "step": 117000
    },
    {
      "epoch": 0.86,
      "learning_rate": 5e-05,
      "loss": 1.0822,
      "step": 117500
    },
    {
      "epoch": 0.86,
      "learning_rate": 5e-05,
      "loss": 1.0885,
      "step": 118000
    },
    {
      "epoch": 0.87,
      "learning_rate": 5e-05,
      "loss": 1.0944,
      "step": 118500
    },
    {
      "epoch": 0.87,
      "learning_rate": 5e-05,
      "loss": 1.0775,
      "step": 119000
    },
    {
      "epoch": 0.87,
      "learning_rate": 5e-05,
      "loss": 1.088,
      "step": 119500
    },
    {
      "epoch": 0.88,
      "learning_rate": 5e-05,
      "loss": 1.0893,
      "step": 120000
    },
    {
      "epoch": 0.88,
      "learning_rate": 5e-05,
      "loss": 1.084,
      "step": 120500
    },
    {
      "epoch": 0.89,
      "learning_rate": 5e-05,
      "loss": 1.0937,
      "step": 121000
    },
    {
      "epoch": 0.89,
      "learning_rate": 5e-05,
      "loss": 1.0857,
      "step": 121500
    },
    {
      "epoch": 0.89,
      "learning_rate": 5e-05,
      "loss": 1.089,
      "step": 122000
    },
    {
      "epoch": 0.9,
      "learning_rate": 5e-05,
      "loss": 1.0913,
      "step": 122500
    },
    {
      "epoch": 0.9,
      "learning_rate": 5e-05,
      "loss": 1.0835,
      "step": 123000
    },
    {
      "epoch": 0.9,
      "learning_rate": 5e-05,
      "loss": 1.0889,
      "step": 123500
    },
    {
      "epoch": 0.91,
      "learning_rate": 5e-05,
      "loss": 1.0848,
      "step": 124000
    },
    {
      "epoch": 0.91,
      "learning_rate": 5e-05,
      "loss": 1.0824,
      "step": 124500
    },
    {
      "epoch": 0.91,
      "learning_rate": 5e-05,
      "loss": 1.0849,
      "step": 125000
    },
    {
      "epoch": 0.92,
      "learning_rate": 5e-05,
      "loss": 1.0788,
      "step": 125500
    },
    {
      "epoch": 0.92,
      "learning_rate": 5e-05,
      "loss": 1.0886,
      "step": 126000
    },
    {
      "epoch": 0.93,
      "learning_rate": 5e-05,
      "loss": 1.0891,
      "step": 126500
    },
    {
      "epoch": 0.93,
      "learning_rate": 5e-05,
      "loss": 1.0835,
      "step": 127000
    },
    {
      "epoch": 0.93,
      "learning_rate": 5e-05,
      "loss": 1.0814,
      "step": 127500
    },
    {
      "epoch": 0.94,
      "learning_rate": 5e-05,
      "loss": 1.0833,
      "step": 128000
    },
    {
      "epoch": 0.94,
      "eval_accuracy": 0.7811326318893699,
      "eval_loss": 1.0721635818481445,
      "eval_runtime": 32.739,
      "eval_samples_per_second": 409.787,
      "eval_steps_per_second": 8.553,
      "step": 128000
    },
    {
      "epoch": 0.94,
      "learning_rate": 5e-05,
      "loss": 1.0762,
      "step": 128500
    },
    {
      "epoch": 0.94,
      "learning_rate": 5e-05,
      "loss": 1.0737,
      "step": 129000
    },
    {
      "epoch": 0.95,
      "learning_rate": 5e-05,
      "loss": 1.077,
      "step": 129500
    },
    {
      "epoch": 0.95,
      "learning_rate": 5e-05,
      "loss": 1.0795,
      "step": 130000
    },
    {
      "epoch": 0.96,
      "learning_rate": 5e-05,
      "loss": 1.0785,
      "step": 130500
    },
    {
      "epoch": 0.96,
      "learning_rate": 5e-05,
      "loss": 1.0784,
      "step": 131000
    },
    {
      "epoch": 0.96,
      "learning_rate": 5e-05,
      "loss": 1.0768,
      "step": 131500
    },
    {
      "epoch": 0.97,
      "learning_rate": 5e-05,
      "loss": 1.0787,
      "step": 132000
    },
    {
      "epoch": 0.97,
      "learning_rate": 5e-05,
      "loss": 1.0761,
      "step": 132500
    },
    {
      "epoch": 0.97,
      "learning_rate": 5e-05,
      "loss": 1.0734,
      "step": 133000
    },
    {
      "epoch": 0.98,
      "learning_rate": 5e-05,
      "loss": 1.0781,
      "step": 133500
    },
    {
      "epoch": 0.98,
      "learning_rate": 5e-05,
      "loss": 1.0689,
      "step": 134000
    },
    {
      "epoch": 0.98,
      "learning_rate": 5e-05,
      "loss": 1.0794,
      "step": 134500
    },
    {
      "epoch": 0.99,
      "learning_rate": 5e-05,
      "loss": 1.0747,
      "step": 135000
    },
    {
      "epoch": 0.99,
      "learning_rate": 5e-05,
      "loss": 1.0769,
      "step": 135500
    },
    {
      "epoch": 1.0,
      "learning_rate": 5e-05,
      "loss": 1.0722,
      "step": 136000
    },
    {
      "epoch": 1.0,
      "learning_rate": 5e-05,
      "loss": 1.068,
      "step": 136500
    },
    {
      "epoch": 1.0,
      "learning_rate": 5e-05,
      "loss": 1.0729,
      "step": 137000
    },
    {
      "epoch": 1.01,
      "learning_rate": 5e-05,
      "loss": 1.0686,
      "step": 137500
    },
    {
      "epoch": 1.01,
      "learning_rate": 5e-05,
      "loss": 1.0706,
      "step": 138000
    },
    {
      "epoch": 1.01,
      "learning_rate": 5e-05,
      "loss": 1.0744,
      "step": 138500
    },
    {
      "epoch": 1.02,
      "learning_rate": 5e-05,
      "loss": 1.0701,
      "step": 139000
    },
    {
      "epoch": 1.02,
      "learning_rate": 5e-05,
      "loss": 1.07,
      "step": 139500
    },
    {
      "epoch": 1.02,
      "learning_rate": 5e-05,
      "loss": 1.0699,
      "step": 140000
    },
    {
      "epoch": 1.03,
      "learning_rate": 5e-05,
      "loss": 1.069,
      "step": 140500
    },
    {
      "epoch": 1.03,
      "learning_rate": 5e-05,
      "loss": 1.0741,
      "step": 141000
    },
    {
      "epoch": 1.04,
      "learning_rate": 5e-05,
      "loss": 1.0693,
      "step": 141500
    },
    {
      "epoch": 1.04,
      "learning_rate": 5e-05,
      "loss": 1.0725,
      "step": 142000
    },
    {
      "epoch": 1.04,
      "learning_rate": 5e-05,
      "loss": 1.073,
      "step": 142500
    },
    {
      "epoch": 1.05,
      "learning_rate": 5e-05,
      "loss": 1.0739,
      "step": 143000
    },
    {
      "epoch": 1.05,
      "learning_rate": 5e-05,
      "loss": 1.068,
      "step": 143500
    },
    {
      "epoch": 1.05,
      "learning_rate": 5e-05,
      "loss": 1.0678,
      "step": 144000
    },
    {
      "epoch": 1.06,
      "learning_rate": 5e-05,
      "loss": 1.0679,
      "step": 144500
    },
    {
      "epoch": 1.06,
      "learning_rate": 5e-05,
      "loss": 1.069,
      "step": 145000
    },
    {
      "epoch": 1.06,
      "learning_rate": 5e-05,
      "loss": 1.0707,
      "step": 145500
    },
    {
      "epoch": 1.07,
      "learning_rate": 5e-05,
      "loss": 1.0687,
      "step": 146000
    },
    {
      "epoch": 1.07,
      "learning_rate": 5e-05,
      "loss": 1.073,
      "step": 146500
    },
    {
      "epoch": 1.08,
      "learning_rate": 5e-05,
      "loss": 1.0643,
      "step": 147000
    },
    {
      "epoch": 1.08,
      "learning_rate": 5e-05,
      "loss": 1.0647,
      "step": 147500
    },
    {
      "epoch": 1.08,
      "learning_rate": 5e-05,
      "loss": 1.0671,
      "step": 148000
    },
    {
      "epoch": 1.09,
      "learning_rate": 5e-05,
      "loss": 1.0722,
      "step": 148500
    },
    {
      "epoch": 1.09,
      "learning_rate": 5e-05,
      "loss": 1.0692,
      "step": 149000
    },
    {
      "epoch": 1.09,
      "learning_rate": 5e-05,
      "loss": 1.071,
      "step": 149500
    },
    {
      "epoch": 1.1,
      "learning_rate": 5e-05,
      "loss": 1.071,
      "step": 150000
    },
    {
      "epoch": 1.1,
      "learning_rate": 5e-05,
      "loss": 1.0705,
      "step": 150500
    },
    {
      "epoch": 1.11,
      "learning_rate": 5e-05,
      "loss": 1.0657,
      "step": 151000
    },
    {
      "epoch": 1.11,
      "learning_rate": 5e-05,
      "loss": 1.0744,
      "step": 151500
    },
    {
      "epoch": 1.11,
      "learning_rate": 5e-05,
      "loss": 1.0655,
      "step": 152000
    },
    {
      "epoch": 1.12,
      "learning_rate": 5e-05,
      "loss": 1.0688,
      "step": 152500
    },
    {
      "epoch": 1.12,
      "learning_rate": 5e-05,
      "loss": 1.0594,
      "step": 153000
    },
    {
      "epoch": 1.12,
      "learning_rate": 5e-05,
      "loss": 1.0704,
      "step": 153500
    },
    {
      "epoch": 1.13,
      "learning_rate": 5e-05,
      "loss": 1.0613,
      "step": 154000
    },
    {
      "epoch": 1.13,
      "learning_rate": 5e-05,
      "loss": 1.0671,
      "step": 154500
    },
    {
      "epoch": 1.13,
      "learning_rate": 5e-05,
      "loss": 1.0646,
      "step": 155000
    },
    {
      "epoch": 1.14,
      "learning_rate": 5e-05,
      "loss": 1.0616,
      "step": 155500
    },
    {
      "epoch": 1.14,
      "learning_rate": 5e-05,
      "loss": 1.0631,
      "step": 156000
    },
    {
      "epoch": 1.15,
      "learning_rate": 5e-05,
      "loss": 1.0581,
      "step": 156500
    },
    {
      "epoch": 1.15,
      "learning_rate": 5e-05,
      "loss": 1.0626,
      "step": 157000
    },
    {
      "epoch": 1.15,
      "learning_rate": 5e-05,
      "loss": 1.054,
      "step": 157500
    },
    {
      "epoch": 1.16,
      "learning_rate": 5e-05,
      "loss": 1.0635,
      "step": 158000
    },
    {
      "epoch": 1.16,
      "learning_rate": 5e-05,
      "loss": 1.0572,
      "step": 158500
    },
    {
      "epoch": 1.16,
      "learning_rate": 5e-05,
      "loss": 1.0596,
      "step": 159000
    },
    {
      "epoch": 1.17,
      "learning_rate": 5e-05,
      "loss": 1.0627,
      "step": 159500
    },
    {
      "epoch": 1.17,
      "learning_rate": 5e-05,
      "loss": 1.0647,
      "step": 160000
    },
    {
      "epoch": 1.17,
      "learning_rate": 5e-05,
      "loss": 1.0556,
      "step": 160500
    },
    {
      "epoch": 1.18,
      "learning_rate": 5e-05,
      "loss": 1.0541,
      "step": 161000
    },
    {
      "epoch": 1.18,
      "learning_rate": 5e-05,
      "loss": 1.0534,
      "step": 161500
    },
    {
      "epoch": 1.19,
      "learning_rate": 5e-05,
      "loss": 1.0516,
      "step": 162000
    },
    {
      "epoch": 1.19,
      "learning_rate": 5e-05,
      "loss": 1.0564,
      "step": 162500
    },
    {
      "epoch": 1.19,
      "learning_rate": 5e-05,
      "loss": 1.0497,
      "step": 163000
    },
    {
      "epoch": 1.2,
      "learning_rate": 5e-05,
      "loss": 1.0643,
      "step": 163500
    },
    {
      "epoch": 1.2,
      "learning_rate": 5e-05,
      "loss": 1.0552,
      "step": 164000
    },
    {
      "epoch": 1.2,
      "learning_rate": 5e-05,
      "loss": 1.058,
      "step": 164500
    },
    {
      "epoch": 1.21,
      "learning_rate": 5e-05,
      "loss": 1.0515,
      "step": 165000
    },
    {
      "epoch": 1.21,
      "learning_rate": 5e-05,
      "loss": 1.0529,
      "step": 165500
    },
    {
      "epoch": 1.21,
      "learning_rate": 5e-05,
      "loss": 1.0567,
      "step": 166000
    },
    {
      "epoch": 1.22,
      "learning_rate": 5e-05,
      "loss": 1.0586,
      "step": 166500
    },
    {
      "epoch": 1.22,
      "learning_rate": 5e-05,
      "loss": 1.0576,
      "step": 167000
    },
    {
      "epoch": 1.23,
      "learning_rate": 5e-05,
      "loss": 1.0514,
      "step": 167500
    },
    {
      "epoch": 1.23,
      "learning_rate": 5e-05,
      "loss": 1.054,
      "step": 168000
    },
    {
      "epoch": 1.23,
      "learning_rate": 5e-05,
      "loss": 1.0559,
      "step": 168500
    },
    {
      "epoch": 1.24,
      "learning_rate": 5e-05,
      "loss": 1.0512,
      "step": 169000
    },
    {
      "epoch": 1.24,
      "learning_rate": 5e-05,
      "loss": 1.0609,
      "step": 169500
    },
    {
      "epoch": 1.24,
      "learning_rate": 5e-05,
      "loss": 1.0571,
      "step": 170000
    },
    {
      "epoch": 1.25,
      "learning_rate": 5e-05,
      "loss": 1.0535,
      "step": 170500
    },
    {
      "epoch": 1.25,
      "learning_rate": 5e-05,
      "loss": 1.0533,
      "step": 171000
    },
    {
      "epoch": 1.26,
      "learning_rate": 5e-05,
      "loss": 1.0436,
      "step": 171500
    },
    {
      "epoch": 1.26,
      "learning_rate": 5e-05,
      "loss": 1.0507,
      "step": 172000
    },
    {
      "epoch": 1.26,
      "learning_rate": 5e-05,
      "loss": 1.0522,
      "step": 172500
    },
    {
      "epoch": 1.27,
      "learning_rate": 5e-05,
      "loss": 1.0623,
      "step": 173000
    },
    {
      "epoch": 1.27,
      "learning_rate": 5e-05,
      "loss": 1.0552,
      "step": 173500
    },
    {
      "epoch": 1.27,
      "learning_rate": 5e-05,
      "loss": 1.0512,
      "step": 174000
    },
    {
      "epoch": 1.28,
      "learning_rate": 5e-05,
      "loss": 1.0561,
      "step": 174500
    },
    {
      "epoch": 1.28,
      "learning_rate": 5e-05,
      "loss": 1.0534,
      "step": 175000
    },
    {
      "epoch": 1.28,
      "learning_rate": 5e-05,
      "loss": 1.0453,
      "step": 175500
    },
    {
      "epoch": 1.29,
      "learning_rate": 5e-05,
      "loss": 1.0523,
      "step": 176000
    },
    {
      "epoch": 1.29,
      "learning_rate": 5e-05,
      "loss": 1.0526,
      "step": 176500
    },
    {
      "epoch": 1.3,
      "learning_rate": 5e-05,
      "loss": 1.0498,
      "step": 177000
    },
    {
      "epoch": 1.3,
      "learning_rate": 5e-05,
      "loss": 1.0523,
      "step": 177500
    },
    {
      "epoch": 1.3,
      "learning_rate": 5e-05,
      "loss": 1.0508,
      "step": 178000
    },
    {
      "epoch": 1.31,
      "learning_rate": 5e-05,
      "loss": 1.0492,
      "step": 178500
    },
    {
      "epoch": 1.31,
      "learning_rate": 5e-05,
      "loss": 1.0492,
      "step": 179000
    },
    {
      "epoch": 1.31,
      "learning_rate": 5e-05,
      "loss": 1.0475,
      "step": 179500
    },
    {
      "epoch": 1.32,
      "learning_rate": 5e-05,
      "loss": 1.0491,
      "step": 180000
    },
    {
      "epoch": 1.32,
      "learning_rate": 5e-05,
      "loss": 1.047,
      "step": 180500
    },
    {
      "epoch": 1.32,
      "learning_rate": 5e-05,
      "loss": 1.0509,
      "step": 181000
    },
    {
      "epoch": 1.33,
      "learning_rate": 5e-05,
      "loss": 1.044,
      "step": 181500
    },
    {
      "epoch": 1.33,
      "learning_rate": 5e-05,
      "loss": 1.0461,
      "step": 182000
    },
    {
      "epoch": 1.34,
      "learning_rate": 5e-05,
      "loss": 1.0459,
      "step": 182500
    },
    {
      "epoch": 1.34,
      "learning_rate": 5e-05,
      "loss": 1.0479,
      "step": 183000
    },
    {
      "epoch": 1.34,
      "learning_rate": 5e-05,
      "loss": 1.0478,
      "step": 183500
    },
    {
      "epoch": 1.35,
      "learning_rate": 5e-05,
      "loss": 1.0499,
      "step": 184000
    },
    {
      "epoch": 1.35,
      "learning_rate": 5e-05,
      "loss": 1.0499,
      "step": 184500
    },
    {
      "epoch": 1.35,
      "learning_rate": 5e-05,
      "loss": 1.0542,
      "step": 185000
    },
    {
      "epoch": 1.36,
      "learning_rate": 5e-05,
      "loss": 1.0539,
      "step": 185500
    },
    {
      "epoch": 1.36,
      "learning_rate": 5e-05,
      "loss": 1.0562,
      "step": 186000
    },
    {
      "epoch": 1.37,
      "learning_rate": 5e-05,
      "loss": 1.0478,
      "step": 186500
    },
    {
      "epoch": 1.37,
      "learning_rate": 5e-05,
      "loss": 1.0491,
      "step": 187000
    },
    {
      "epoch": 1.37,
      "learning_rate": 5e-05,
      "loss": 1.0481,
      "step": 187500
    },
    {
      "epoch": 1.38,
      "learning_rate": 5e-05,
      "loss": 1.0525,
      "step": 188000
    },
    {
      "epoch": 1.38,
      "learning_rate": 5e-05,
      "loss": 1.0487,
      "step": 188500
    },
    {
      "epoch": 1.38,
      "learning_rate": 5e-05,
      "loss": 1.0478,
      "step": 189000
    },
    {
      "epoch": 1.39,
      "learning_rate": 5e-05,
      "loss": 1.0454,
      "step": 189500
    },
    {
      "epoch": 1.39,
      "learning_rate": 5e-05,
      "loss": 1.0514,
      "step": 190000
    },
    {
      "epoch": 1.39,
      "learning_rate": 5e-05,
      "loss": 1.0464,
      "step": 190500
    },
    {
      "epoch": 1.4,
      "learning_rate": 5e-05,
      "loss": 1.0503,
      "step": 191000
    },
    {
      "epoch": 1.4,
      "learning_rate": 5e-05,
      "loss": 1.0444,
      "step": 191500
    },
    {
      "epoch": 1.41,
      "learning_rate": 5e-05,
      "loss": 1.0459,
      "step": 192000
    },
    {
      "epoch": 1.41,
      "eval_accuracy": 0.7880748504192896,
      "eval_loss": 1.0307050943374634,
      "eval_runtime": 32.7044,
      "eval_samples_per_second": 410.22,
      "eval_steps_per_second": 8.562,
      "step": 192000
    },
    {
      "epoch": 1.41,
      "learning_rate": 5e-05,
      "loss": 1.0412,
      "step": 192500
    },
    {
      "epoch": 1.41,
      "learning_rate": 5e-05,
      "loss": 1.0555,
      "step": 193000
    },
    {
      "epoch": 1.42,
      "learning_rate": 5e-05,
      "loss": 1.0413,
      "step": 193500
    },
    {
      "epoch": 1.42,
      "learning_rate": 5e-05,
      "loss": 1.0481,
      "step": 194000
    },
    {
      "epoch": 1.42,
      "learning_rate": 5e-05,
      "loss": 1.0471,
      "step": 194500
    },
    {
      "epoch": 1.43,
      "learning_rate": 5e-05,
      "loss": 1.046,
      "step": 195000
    },
    {
      "epoch": 1.43,
      "learning_rate": 5e-05,
      "loss": 1.0478,
      "step": 195500
    },
    {
      "epoch": 1.43,
      "learning_rate": 5e-05,
      "loss": 1.0436,
      "step": 196000
    },
    {
      "epoch": 1.44,
      "learning_rate": 5e-05,
      "loss": 1.0442,
      "step": 196500
    },
    {
      "epoch": 1.44,
      "learning_rate": 5e-05,
      "loss": 1.0473,
      "step": 197000
    },
    {
      "epoch": 1.45,
      "learning_rate": 5e-05,
      "loss": 1.0386,
      "step": 197500
    },
    {
      "epoch": 1.45,
      "learning_rate": 5e-05,
      "loss": 1.0419,
      "step": 198000
    },
    {
      "epoch": 1.45,
      "learning_rate": 5e-05,
      "loss": 1.0435,
      "step": 198500
    },
    {
      "epoch": 1.46,
      "learning_rate": 5e-05,
      "loss": 1.0436,
      "step": 199000
    },
    {
      "epoch": 1.46,
      "learning_rate": 5e-05,
      "loss": 1.0451,
      "step": 199500
    },
    {
      "epoch": 1.46,
      "learning_rate": 5e-05,
      "loss": 1.048,
      "step": 200000
    },
    {
      "epoch": 1.47,
      "learning_rate": 5e-05,
      "loss": 1.0477,
      "step": 200500
    },
    {
      "epoch": 1.47,
      "learning_rate": 5e-05,
      "loss": 1.0453,
      "step": 201000
    },
    {
      "epoch": 1.47,
      "learning_rate": 5e-05,
      "loss": 1.0435,
      "step": 201500
    },
    {
      "epoch": 1.48,
      "learning_rate": 5e-05,
      "loss": 1.0467,
      "step": 202000
    },
    {
      "epoch": 1.48,
      "learning_rate": 5e-05,
      "loss": 1.044,
      "step": 202500
    },
    {
      "epoch": 1.49,
      "learning_rate": 5e-05,
      "loss": 1.0495,
      "step": 203000
    },
    {
      "epoch": 1.49,
      "learning_rate": 5e-05,
      "loss": 1.0374,
      "step": 203500
    },
    {
      "epoch": 1.49,
      "learning_rate": 5e-05,
      "loss": 1.0421,
      "step": 204000
    },
    {
      "epoch": 1.5,
      "learning_rate": 5e-05,
      "loss": 1.037,
      "step": 204500
    },
    {
      "epoch": 1.5,
      "learning_rate": 5e-05,
      "loss": 1.0383,
      "step": 205000
    },
    {
      "epoch": 1.5,
      "learning_rate": 5e-05,
      "loss": 1.038,
      "step": 205500
    },
    {
      "epoch": 1.51,
      "learning_rate": 5e-05,
      "loss": 1.038,
      "step": 206000
    },
    {
      "epoch": 1.51,
      "learning_rate": 5e-05,
      "loss": 1.046,
      "step": 206500
    },
    {
      "epoch": 1.52,
      "learning_rate": 5e-05,
      "loss": 1.0428,
      "step": 207000
    },
    {
      "epoch": 1.52,
      "learning_rate": 5e-05,
      "loss": 1.0392,
      "step": 207500
    },
    {
      "epoch": 1.52,
      "learning_rate": 5e-05,
      "loss": 1.0415,
      "step": 208000
    },
    {
      "epoch": 1.53,
      "learning_rate": 5e-05,
      "loss": 1.0426,
      "step": 208500
    },
    {
      "epoch": 1.53,
      "learning_rate": 5e-05,
      "loss": 1.0419,
      "step": 209000
    },
    {
      "epoch": 1.53,
      "learning_rate": 5e-05,
      "loss": 1.0495,
      "step": 209500
    },
    {
      "epoch": 1.54,
      "learning_rate": 5e-05,
      "loss": 1.0492,
      "step": 210000
    },
    {
      "epoch": 1.54,
      "learning_rate": 5e-05,
      "loss": 1.0462,
      "step": 210500
    },
    {
      "epoch": 1.54,
      "learning_rate": 5e-05,
      "loss": 1.0383,
      "step": 211000
    },
    {
      "epoch": 1.55,
      "learning_rate": 5e-05,
      "loss": 1.0402,
      "step": 211500
    },
    {
      "epoch": 1.55,
      "learning_rate": 5e-05,
      "loss": 1.0465,
      "step": 212000
    },
    {
      "epoch": 1.56,
      "learning_rate": 5e-05,
      "loss": 1.044,
      "step": 212500
    },
    {
      "epoch": 1.56,
      "learning_rate": 5e-05,
      "loss": 1.0362,
      "step": 213000
    },
    {
      "epoch": 1.56,
      "learning_rate": 5e-05,
      "loss": 1.0354,
      "step": 213500
    },
    {
      "epoch": 1.57,
      "learning_rate": 5e-05,
      "loss": 1.0389,
      "step": 214000
    },
    {
      "epoch": 1.57,
      "learning_rate": 5e-05,
      "loss": 1.0459,
      "step": 214500
    },
    {
      "epoch": 1.57,
      "learning_rate": 5e-05,
      "loss": 1.0503,
      "step": 215000
    },
    {
      "epoch": 1.58,
      "learning_rate": 5e-05,
      "loss": 1.0489,
      "step": 215500
    },
    {
      "epoch": 1.58,
      "learning_rate": 5e-05,
      "loss": 1.0399,
      "step": 216000
    },
    {
      "epoch": 1.58,
      "learning_rate": 5e-05,
      "loss": 1.0444,
      "step": 216500
    },
    {
      "epoch": 1.59,
      "learning_rate": 5e-05,
      "loss": 1.0413,
      "step": 217000
    },
    {
      "epoch": 1.59,
      "learning_rate": 5e-05,
      "loss": 1.0431,
      "step": 217500
    },
    {
      "epoch": 1.6,
      "learning_rate": 5e-05,
      "loss": 1.0456,
      "step": 218000
    },
    {
      "epoch": 1.6,
      "learning_rate": 5e-05,
      "loss": 1.0424,
      "step": 218500
    },
    {
      "epoch": 1.6,
      "learning_rate": 5e-05,
      "loss": 1.0382,
      "step": 219000
    },
    {
      "epoch": 1.61,
      "learning_rate": 5e-05,
      "loss": 1.0418,
      "step": 219500
    },
    {
      "epoch": 1.61,
      "learning_rate": 5e-05,
      "loss": 1.0369,
      "step": 220000
    },
    {
      "epoch": 1.61,
      "learning_rate": 5e-05,
      "loss": 1.0398,
      "step": 220500
    },
    {
      "epoch": 1.62,
      "learning_rate": 5e-05,
      "loss": 1.032,
      "step": 221000
    },
    {
      "epoch": 1.62,
      "learning_rate": 5e-05,
      "loss": 1.0388,
      "step": 221500
    },
    {
      "epoch": 1.62,
      "learning_rate": 5e-05,
      "loss": 1.042,
      "step": 222000
    },
    {
      "epoch": 1.63,
      "learning_rate": 5e-05,
      "loss": 1.0375,
      "step": 222500
    },
    {
      "epoch": 1.63,
      "learning_rate": 5e-05,
      "loss": 1.0454,
      "step": 223000
    },
    {
      "epoch": 1.64,
      "learning_rate": 5e-05,
      "loss": 1.0387,
      "step": 223500
    },
    {
      "epoch": 1.64,
      "learning_rate": 5e-05,
      "loss": 1.0312,
      "step": 224000
    },
    {
      "epoch": 1.64,
      "learning_rate": 5e-05,
      "loss": 1.0339,
      "step": 224500
    },
    {
      "epoch": 1.65,
      "learning_rate": 5e-05,
      "loss": 1.041,
      "step": 225000
    },
    {
      "epoch": 1.65,
      "learning_rate": 5e-05,
      "loss": 1.035,
      "step": 225500
    },
    {
      "epoch": 1.65,
      "learning_rate": 5e-05,
      "loss": 1.0393,
      "step": 226000
    },
    {
      "epoch": 1.66,
      "learning_rate": 5e-05,
      "loss": 1.0321,
      "step": 226500
    },
    {
      "epoch": 1.66,
      "learning_rate": 5e-05,
      "loss": 1.0369,
      "step": 227000
    },
    {
      "epoch": 1.67,
      "learning_rate": 5e-05,
      "loss": 1.036,
      "step": 227500
    },
    {
      "epoch": 1.67,
      "learning_rate": 5e-05,
      "loss": 1.0383,
      "step": 228000
    },
    {
      "epoch": 1.67,
      "learning_rate": 5e-05,
      "loss": 1.0354,
      "step": 228500
    },
    {
      "epoch": 1.68,
      "learning_rate": 5e-05,
      "loss": 1.03,
      "step": 229000
    },
    {
      "epoch": 1.68,
      "learning_rate": 5e-05,
      "loss": 1.0383,
      "step": 229500
    },
    {
      "epoch": 1.68,
      "learning_rate": 5e-05,
      "loss": 1.0335,
      "step": 230000
    },
    {
      "epoch": 1.69,
      "learning_rate": 5e-05,
      "loss": 1.0317,
      "step": 230500
    },
    {
      "epoch": 1.69,
      "learning_rate": 5e-05,
      "loss": 1.0361,
      "step": 231000
    },
    {
      "epoch": 1.69,
      "learning_rate": 5e-05,
      "loss": 1.0406,
      "step": 231500
    },
    {
      "epoch": 1.7,
      "learning_rate": 5e-05,
      "loss": 1.0328,
      "step": 232000
    },
    {
      "epoch": 1.7,
      "learning_rate": 5e-05,
      "loss": 1.034,
      "step": 232500
    },
    {
      "epoch": 1.71,
      "learning_rate": 5e-05,
      "loss": 1.035,
      "step": 233000
    },
    {
      "epoch": 1.71,
      "learning_rate": 5e-05,
      "loss": 1.0353,
      "step": 233500
    },
    {
      "epoch": 1.71,
      "learning_rate": 5e-05,
      "loss": 1.0288,
      "step": 234000
    },
    {
      "epoch": 1.72,
      "learning_rate": 5e-05,
      "loss": 1.031,
      "step": 234500
    },
    {
      "epoch": 1.72,
      "learning_rate": 5e-05,
      "loss": 1.0332,
      "step": 235000
    },
    {
      "epoch": 1.72,
      "learning_rate": 5e-05,
      "loss": 1.0359,
      "step": 235500
    },
    {
      "epoch": 1.73,
      "learning_rate": 5e-05,
      "loss": 1.033,
      "step": 236000
    },
    {
      "epoch": 1.73,
      "learning_rate": 5e-05,
      "loss": 1.0382,
      "step": 236500
    },
    {
      "epoch": 1.73,
      "learning_rate": 5e-05,
      "loss": 1.038,
      "step": 237000
    },
    {
      "epoch": 1.74,
      "learning_rate": 5e-05,
      "loss": 1.0391,
      "step": 237500
    },
    {
      "epoch": 1.74,
      "learning_rate": 5e-05,
      "loss": 1.0317,
      "step": 238000
    },
    {
      "epoch": 1.75,
      "learning_rate": 5e-05,
      "loss": 1.0423,
      "step": 238500
    },
    {
      "epoch": 1.75,
      "learning_rate": 5e-05,
      "loss": 1.0369,
      "step": 239000
    },
    {
      "epoch": 1.75,
      "learning_rate": 5e-05,
      "loss": 1.0419,
      "step": 239500
    },
    {
      "epoch": 1.76,
      "learning_rate": 5e-05,
      "loss": 1.0388,
      "step": 240000
    },
    {
      "epoch": 1.76,
      "learning_rate": 5e-05,
      "loss": 1.0349,
      "step": 240500
    },
    {
      "epoch": 1.76,
      "learning_rate": 5e-05,
      "loss": 1.034,
      "step": 241000
    },
    {
      "epoch": 1.77,
      "learning_rate": 5e-05,
      "loss": 1.0359,
      "step": 241500
    },
    {
      "epoch": 1.77,
      "learning_rate": 5e-05,
      "loss": 1.0423,
      "step": 242000
    },
    {
      "epoch": 1.77,
      "learning_rate": 5e-05,
      "loss": 1.0341,
      "step": 242500
    },
    {
      "epoch": 1.78,
      "learning_rate": 5e-05,
      "loss": 1.0287,
      "step": 243000
    },
    {
      "epoch": 1.78,
      "learning_rate": 5e-05,
      "loss": 1.0347,
      "step": 243500
    },
    {
      "epoch": 1.79,
      "learning_rate": 5e-05,
      "loss": 1.0304,
      "step": 244000
    },
    {
      "epoch": 1.79,
      "learning_rate": 5e-05,
      "loss": 1.0311,
      "step": 244500
    },
    {
      "epoch": 1.79,
      "learning_rate": 5e-05,
      "loss": 1.0364,
      "step": 245000
    },
    {
      "epoch": 1.8,
      "learning_rate": 5e-05,
      "loss": 1.0271,
      "step": 245500
    },
    {
      "epoch": 1.8,
      "learning_rate": 5e-05,
      "loss": 1.0308,
      "step": 246000
    },
    {
      "epoch": 1.8,
      "learning_rate": 5e-05,
      "loss": 1.0319,
      "step": 246500
    },
    {
      "epoch": 1.81,
      "learning_rate": 5e-05,
      "loss": 1.0345,
      "step": 247000
    },
    {
      "epoch": 1.81,
      "learning_rate": 5e-05,
      "loss": 1.0334,
      "step": 247500
    },
    {
      "epoch": 1.82,
      "learning_rate": 5e-05,
      "loss": 1.0353,
      "step": 248000
    },
    {
      "epoch": 1.82,
      "learning_rate": 5e-05,
      "loss": 1.0274,
      "step": 248500
    },
    {
      "epoch": 1.82,
      "learning_rate": 5e-05,
      "loss": 1.032,
      "step": 249000
    },
    {
      "epoch": 1.83,
      "learning_rate": 5e-05,
      "loss": 1.0336,
      "step": 249500
    },
    {
      "epoch": 1.83,
      "learning_rate": 5e-05,
      "loss": 1.029,
      "step": 250000
    },
    {
      "epoch": 1.83,
      "learning_rate": 5e-05,
      "loss": 1.0278,
      "step": 250500
    },
    {
      "epoch": 1.84,
      "learning_rate": 5e-05,
      "loss": 1.0325,
      "step": 251000
    },
    {
      "epoch": 1.84,
      "learning_rate": 5e-05,
      "loss": 1.0329,
      "step": 251500
    },
    {
      "epoch": 1.84,
      "learning_rate": 5e-05,
      "loss": 1.0312,
      "step": 252000
    },
    {
      "epoch": 1.85,
      "learning_rate": 5e-05,
      "loss": 1.0384,
      "step": 252500
    },
    {
      "epoch": 1.85,
      "learning_rate": 5e-05,
      "loss": 1.0322,
      "step": 253000
    },
    {
      "epoch": 1.86,
      "learning_rate": 5e-05,
      "loss": 1.0296,
      "step": 253500
    },
    {
      "epoch": 1.86,
      "learning_rate": 5e-05,
      "loss": 1.0309,
      "step": 254000
    },
    {
      "epoch": 1.86,
      "learning_rate": 5e-05,
      "loss": 1.0259,
      "step": 254500
    },
    {
      "epoch": 1.87,
      "learning_rate": 5e-05,
      "loss": 1.0361,
      "step": 255000
    },
    {
      "epoch": 1.87,
      "learning_rate": 5e-05,
      "loss": 1.0302,
      "step": 255500
    },
    {
      "epoch": 1.87,
      "learning_rate": 5e-05,
      "loss": 1.0314,
      "step": 256000
    },
    {
      "epoch": 1.87,
      "eval_accuracy": 0.7907370033516936,
      "eval_loss": 1.0169389247894287,
      "eval_runtime": 32.7988,
      "eval_samples_per_second": 409.04,
      "eval_steps_per_second": 8.537,
      "step": 256000
    },
    {
      "epoch": 1.88,
      "learning_rate": 5e-05,
      "loss": 1.0349,
      "step": 256500
    },
    {
      "epoch": 1.88,
      "learning_rate": 5e-05,
      "loss": 1.0369,
      "step": 257000
    },
    {
      "epoch": 1.88,
      "learning_rate": 5e-05,
      "loss": 1.0345,
      "step": 257500
    },
    {
      "epoch": 1.89,
      "learning_rate": 5e-05,
      "loss": 1.0275,
      "step": 258000
    },
    {
      "epoch": 1.89,
      "learning_rate": 5e-05,
      "loss": 1.0342,
      "step": 258500
    },
    {
      "epoch": 1.9,
      "learning_rate": 5e-05,
      "loss": 1.0287,
      "step": 259000
    },
    {
      "epoch": 1.9,
      "learning_rate": 5e-05,
      "loss": 1.0238,
      "step": 259500
    },
    {
      "epoch": 1.9,
      "learning_rate": 5e-05,
      "loss": 1.0258,
      "step": 260000
    },
    {
      "epoch": 1.91,
      "learning_rate": 5e-05,
      "loss": 1.029,
      "step": 260500
    },
    {
      "epoch": 1.91,
      "learning_rate": 5e-05,
      "loss": 1.0277,
      "step": 261000
    },
    {
      "epoch": 1.91,
      "learning_rate": 5e-05,
      "loss": 1.026,
      "step": 261500
    },
    {
      "epoch": 1.92,
      "learning_rate": 5e-05,
      "loss": 1.0265,
      "step": 262000
    },
    {
      "epoch": 1.92,
      "learning_rate": 5e-05,
      "loss": 1.0283,
      "step": 262500
    },
    {
      "epoch": 1.92,
      "learning_rate": 5e-05,
      "loss": 1.0283,
      "step": 263000
    },
    {
      "epoch": 1.93,
      "learning_rate": 5e-05,
      "loss": 1.0268,
      "step": 263500
    },
    {
      "epoch": 1.93,
      "learning_rate": 5e-05,
      "loss": 1.032,
      "step": 264000
    },
    {
      "epoch": 1.94,
      "learning_rate": 5e-05,
      "loss": 1.0233,
      "step": 264500
    },
    {
      "epoch": 1.94,
      "learning_rate": 5e-05,
      "loss": 1.0257,
      "step": 265000
    },
    {
      "epoch": 1.94,
      "learning_rate": 5e-05,
      "loss": 1.0286,
      "step": 265500
    },
    {
      "epoch": 1.95,
      "learning_rate": 5e-05,
      "loss": 1.0226,
      "step": 266000
    },
    {
      "epoch": 1.95,
      "learning_rate": 5e-05,
      "loss": 1.0265,
      "step": 266500
    },
    {
      "epoch": 1.95,
      "learning_rate": 5e-05,
      "loss": 1.0245,
      "step": 267000
    },
    {
      "epoch": 1.96,
      "learning_rate": 5e-05,
      "loss": 1.0231,
      "step": 267500
    },
    {
      "epoch": 1.96,
      "learning_rate": 5e-05,
      "loss": 1.0259,
      "step": 268000
    },
    {
      "epoch": 1.97,
      "learning_rate": 5e-05,
      "loss": 1.0302,
      "step": 268500
    },
    {
      "epoch": 1.97,
      "learning_rate": 5e-05,
      "loss": 1.0316,
      "step": 269000
    },
    {
      "epoch": 1.97,
      "learning_rate": 5e-05,
      "loss": 1.0314,
      "step": 269500
    },
    {
      "epoch": 1.98,
      "learning_rate": 5e-05,
      "loss": 1.0329,
      "step": 270000
    },
    {
      "epoch": 1.98,
      "learning_rate": 5e-05,
      "loss": 1.0275,
      "step": 270500
    },
    {
      "epoch": 1.98,
      "learning_rate": 5e-05,
      "loss": 1.0308,
      "step": 271000
    },
    {
      "epoch": 1.99,
      "learning_rate": 5e-05,
      "loss": 1.022,
      "step": 271500
    },
    {
      "epoch": 1.99,
      "learning_rate": 5e-05,
      "loss": 1.0346,
      "step": 272000
    },
    {
      "epoch": 1.99,
      "learning_rate": 5e-05,
      "loss": 1.0275,
      "step": 272500
    },
    {
      "epoch": 2.0,
      "learning_rate": 5e-05,
      "loss": 1.0283,
      "step": 273000
    },
    {
      "epoch": 2.0,
      "learning_rate": 5e-05,
      "loss": 1.0255,
      "step": 273500
    },
    {
      "epoch": 2.01,
      "learning_rate": 5e-05,
      "loss": 1.0261,
      "step": 274000
    },
    {
      "epoch": 2.01,
      "learning_rate": 5e-05,
      "loss": 1.0219,
      "step": 274500
    },
    {
      "epoch": 2.01,
      "learning_rate": 5e-05,
      "loss": 1.0295,
      "step": 275000
    },
    {
      "epoch": 2.02,
      "learning_rate": 5e-05,
      "loss": 1.0239,
      "step": 275500
    },
    {
      "epoch": 2.02,
      "learning_rate": 5e-05,
      "loss": 1.0221,
      "step": 276000
    },
    {
      "epoch": 2.02,
      "learning_rate": 5e-05,
      "loss": 1.0216,
      "step": 276500
    },
    {
      "epoch": 2.03,
      "learning_rate": 5e-05,
      "loss": 1.0266,
      "step": 277000
    },
    {
      "epoch": 2.03,
      "learning_rate": 5e-05,
      "loss": 1.027,
      "step": 277500
    },
    {
      "epoch": 2.03,
      "learning_rate": 5e-05,
      "loss": 1.0235,
      "step": 278000
    },
    {
      "epoch": 2.04,
      "learning_rate": 5e-05,
      "loss": 1.0259,
      "step": 278500
    },
    {
      "epoch": 2.04,
      "learning_rate": 5e-05,
      "loss": 1.0209,
      "step": 279000
    },
    {
      "epoch": 2.05,
      "learning_rate": 5e-05,
      "loss": 1.0202,
      "step": 279500
    },
    {
      "epoch": 2.05,
      "learning_rate": 5e-05,
      "loss": 1.0295,
      "step": 280000
    },
    {
      "epoch": 2.05,
      "learning_rate": 5e-05,
      "loss": 1.0236,
      "step": 280500
    },
    {
      "epoch": 2.06,
      "learning_rate": 5e-05,
      "loss": 1.0292,
      "step": 281000
    },
    {
      "epoch": 2.06,
      "learning_rate": 5e-05,
      "loss": 1.0226,
      "step": 281500
    },
    {
      "epoch": 2.06,
      "learning_rate": 5e-05,
      "loss": 1.0308,
      "step": 282000
    },
    {
      "epoch": 2.07,
      "learning_rate": 5e-05,
      "loss": 1.0219,
      "step": 282500
    },
    {
      "epoch": 2.07,
      "learning_rate": 5e-05,
      "loss": 1.023,
      "step": 283000
    },
    {
      "epoch": 2.07,
      "learning_rate": 5e-05,
      "loss": 1.0251,
      "step": 283500
    },
    {
      "epoch": 2.08,
      "learning_rate": 5e-05,
      "loss": 1.0301,
      "step": 284000
    },
    {
      "epoch": 2.08,
      "learning_rate": 5e-05,
      "loss": 1.0293,
      "step": 284500
    },
    {
      "epoch": 2.09,
      "learning_rate": 5e-05,
      "loss": 1.0169,
      "step": 285000
    },
    {
      "epoch": 2.09,
      "learning_rate": 5e-05,
      "loss": 1.0204,
      "step": 285500
    },
    {
      "epoch": 2.09,
      "learning_rate": 5e-05,
      "loss": 1.0182,
      "step": 286000
    },
    {
      "epoch": 2.1,
      "learning_rate": 5e-05,
      "loss": 1.016,
      "step": 286500
    },
    {
      "epoch": 2.1,
      "learning_rate": 5e-05,
      "loss": 1.0174,
      "step": 287000
    },
    {
      "epoch": 2.1,
      "learning_rate": 5e-05,
      "loss": 1.0148,
      "step": 287500
    },
    {
      "epoch": 2.11,
      "learning_rate": 5e-05,
      "loss": 1.0224,
      "step": 288000
    },
    {
      "epoch": 2.11,
      "learning_rate": 5e-05,
      "loss": 1.0201,
      "step": 288500
    },
    {
      "epoch": 2.12,
      "learning_rate": 5e-05,
      "loss": 1.0219,
      "step": 289000
    },
    {
      "epoch": 2.12,
      "learning_rate": 5e-05,
      "loss": 1.0172,
      "step": 289500
    },
    {
      "epoch": 2.12,
      "learning_rate": 5e-05,
      "loss": 1.0165,
      "step": 290000
    },
    {
      "epoch": 2.13,
      "learning_rate": 5e-05,
      "loss": 1.0227,
      "step": 290500
    },
    {
      "epoch": 2.13,
      "learning_rate": 5e-05,
      "loss": 1.0217,
      "step": 291000
    },
    {
      "epoch": 2.13,
      "learning_rate": 5e-05,
      "loss": 1.0239,
      "step": 291500
    },
    {
      "epoch": 2.14,
      "learning_rate": 5e-05,
      "loss": 1.0192,
      "step": 292000
    },
    {
      "epoch": 2.14,
      "learning_rate": 5e-05,
      "loss": 1.0196,
      "step": 292500
    },
    {
      "epoch": 2.14,
      "learning_rate": 5e-05,
      "loss": 1.0214,
      "step": 293000
    },
    {
      "epoch": 2.15,
      "learning_rate": 5e-05,
      "loss": 1.0215,
      "step": 293500
    },
    {
      "epoch": 2.15,
      "learning_rate": 5e-05,
      "loss": 1.0198,
      "step": 294000
    },
    {
      "epoch": 2.16,
      "learning_rate": 5e-05,
      "loss": 1.0244,
      "step": 294500
    },
    {
      "epoch": 2.16,
      "learning_rate": 5e-05,
      "loss": 1.0201,
      "step": 295000
    },
    {
      "epoch": 2.16,
      "learning_rate": 5e-05,
      "loss": 1.0123,
      "step": 295500
    },
    {
      "epoch": 2.17,
      "learning_rate": 5e-05,
      "loss": 1.0124,
      "step": 296000
    },
    {
      "epoch": 2.17,
      "learning_rate": 5e-05,
      "loss": 1.0179,
      "step": 296500
    },
    {
      "epoch": 2.17,
      "learning_rate": 5e-05,
      "loss": 1.0177,
      "step": 297000
    },
    {
      "epoch": 2.18,
      "learning_rate": 5e-05,
      "loss": 1.0145,
      "step": 297500
    },
    {
      "epoch": 2.18,
      "learning_rate": 5e-05,
      "loss": 1.0174,
      "step": 298000
    },
    {
      "epoch": 2.18,
      "learning_rate": 5e-05,
      "loss": 1.0144,
      "step": 298500
    },
    {
      "epoch": 2.19,
      "learning_rate": 5e-05,
      "loss": 1.0209,
      "step": 299000
    },
    {
      "epoch": 2.19,
      "learning_rate": 5e-05,
      "loss": 1.0219,
      "step": 299500
    },
    {
      "epoch": 2.2,
      "learning_rate": 5e-05,
      "loss": 1.0247,
      "step": 300000
    },
    {
      "epoch": 2.2,
      "learning_rate": 5e-05,
      "loss": 1.018,
      "step": 300500
    },
    {
      "epoch": 2.2,
      "learning_rate": 5e-05,
      "loss": 1.0213,
      "step": 301000
    },
    {
      "epoch": 2.21,
      "learning_rate": 5e-05,
      "loss": 1.026,
      "step": 301500
    },
    {
      "epoch": 2.21,
      "learning_rate": 5e-05,
      "loss": 1.0211,
      "step": 302000
    },
    {
      "epoch": 2.21,
      "learning_rate": 5e-05,
      "loss": 1.0174,
      "step": 302500
    },
    {
      "epoch": 2.22,
      "learning_rate": 5e-05,
      "loss": 1.0153,
      "step": 303000
    },
    {
      "epoch": 2.22,
      "learning_rate": 5e-05,
      "loss": 1.0168,
      "step": 303500
    },
    {
      "epoch": 2.22,
      "learning_rate": 5e-05,
      "loss": 1.0128,
      "step": 304000
    },
    {
      "epoch": 2.23,
      "learning_rate": 5e-05,
      "loss": 1.0127,
      "step": 304500
    },
    {
      "epoch": 2.23,
      "learning_rate": 5e-05,
      "loss": 1.0144,
      "step": 305000
    },
    {
      "epoch": 2.24,
      "learning_rate": 5e-05,
      "loss": 1.0126,
      "step": 305500
    },
    {
      "epoch": 2.24,
      "learning_rate": 5e-05,
      "loss": 1.0119,
      "step": 306000
    },
    {
      "epoch": 2.24,
      "learning_rate": 5e-05,
      "loss": 1.007,
      "step": 306500
    },
    {
      "epoch": 2.25,
      "learning_rate": 5e-05,
      "loss": 1.0151,
      "step": 307000
    },
    {
      "epoch": 2.25,
      "learning_rate": 5e-05,
      "loss": 1.0186,
      "step": 307500
    },
    {
      "epoch": 2.25,
      "learning_rate": 5e-05,
      "loss": 1.0147,
      "step": 308000
    },
    {
      "epoch": 2.26,
      "learning_rate": 5e-05,
      "loss": 1.0116,
      "step": 308500
    },
    {
      "epoch": 2.26,
      "learning_rate": 5e-05,
      "loss": 1.0095,
      "step": 309000
    },
    {
      "epoch": 2.27,
      "learning_rate": 5e-05,
      "loss": 1.009,
      "step": 309500
    },
    {
      "epoch": 2.27,
      "learning_rate": 5e-05,
      "loss": 1.0073,
      "step": 310000
    },
    {
      "epoch": 2.27,
      "learning_rate": 5e-05,
      "loss": 1.0147,
      "step": 310500
    },
    {
      "epoch": 2.28,
      "learning_rate": 5e-05,
      "loss": 1.0102,
      "step": 311000
    },
    {
      "epoch": 2.28,
      "learning_rate": 5e-05,
      "loss": 1.0063,
      "step": 311500
    },
    {
      "epoch": 2.28,
      "learning_rate": 5e-05,
      "loss": 1.0043,
      "step": 312000
    },
    {
      "epoch": 2.29,
      "learning_rate": 5e-05,
      "loss": 1.0066,
      "step": 312500
    },
    {
      "epoch": 2.29,
      "learning_rate": 5e-05,
      "loss": 1.0122,
      "step": 313000
    },
    {
      "epoch": 2.29,
      "learning_rate": 5e-05,
      "loss": 1.0172,
      "step": 313500
    },
    {
      "epoch": 2.3,
      "learning_rate": 5e-05,
      "loss": 1.0115,
      "step": 314000
    },
    {
      "epoch": 2.3,
      "learning_rate": 5e-05,
      "loss": 1.0113,
      "step": 314500
    },
    {
      "epoch": 2.31,
      "learning_rate": 5e-05,
      "loss": 1.008,
      "step": 315000
    },
    {
      "epoch": 2.31,
      "learning_rate": 5e-05,
      "loss": 1.0094,
      "step": 315500
    },
    {
      "epoch": 2.31,
      "learning_rate": 5e-05,
      "loss": 1.0179,
      "step": 316000
    },
    {
      "epoch": 2.32,
      "learning_rate": 5e-05,
      "loss": 1.0109,
      "step": 316500
    },
    {
      "epoch": 2.32,
      "learning_rate": 5e-05,
      "loss": 1.0127,
      "step": 317000
    },
    {
      "epoch": 2.32,
      "learning_rate": 5e-05,
      "loss": 1.0131,
      "step": 317500
    },
    {
      "epoch": 2.33,
      "learning_rate": 5e-05,
      "loss": 1.0109,
      "step": 318000
    },
    {
      "epoch": 2.33,
      "learning_rate": 5e-05,
      "loss": 1.0117,
      "step": 318500
    },
    {
      "epoch": 2.33,
      "learning_rate": 5e-05,
      "loss": 1.0176,
      "step": 319000
    },
    {
      "epoch": 2.34,
      "learning_rate": 5e-05,
      "loss": 1.009,
      "step": 319500
    },
    {
      "epoch": 2.34,
      "learning_rate": 5e-05,
      "loss": 1.013,
      "step": 320000
    },
    {
      "epoch": 2.34,
      "eval_accuracy": 0.794592400932352,
      "eval_loss": 0.9925464987754822,
      "eval_runtime": 32.6868,
      "eval_samples_per_second": 410.441,
      "eval_steps_per_second": 8.566,
      "step": 320000
    },
    {
      "epoch": 2.35,
      "learning_rate": 5e-05,
      "loss": 1.0104,
      "step": 320500
    },
    {
      "epoch": 2.35,
      "learning_rate": 5e-05,
      "loss": 1.014,
      "step": 321000
    },
    {
      "epoch": 2.35,
      "learning_rate": 5e-05,
      "loss": 1.0136,
      "step": 321500
    },
    {
      "epoch": 2.36,
      "learning_rate": 5e-05,
      "loss": 1.0077,
      "step": 322000
    },
    {
      "epoch": 2.36,
      "learning_rate": 5e-05,
      "loss": 1.0157,
      "step": 322500
    },
    {
      "epoch": 2.36,
      "learning_rate": 5e-05,
      "loss": 1.0133,
      "step": 323000
    },
    {
      "epoch": 2.37,
      "learning_rate": 5e-05,
      "loss": 1.0109,
      "step": 323500
    },
    {
      "epoch": 2.37,
      "learning_rate": 5e-05,
      "loss": 1.0082,
      "step": 324000
    },
    {
      "epoch": 2.38,
      "learning_rate": 5e-05,
      "loss": 1.0085,
      "step": 324500
    },
    {
      "epoch": 2.38,
      "learning_rate": 5e-05,
      "loss": 1.0052,
      "step": 325000
    },
    {
      "epoch": 2.38,
      "learning_rate": 5e-05,
      "loss": 1.0053,
      "step": 325500
    },
    {
      "epoch": 2.39,
      "learning_rate": 5e-05,
      "loss": 1.0065,
      "step": 326000
    },
    {
      "epoch": 2.39,
      "learning_rate": 5e-05,
      "loss": 1.0095,
      "step": 326500
    },
    {
      "epoch": 2.39,
      "learning_rate": 5e-05,
      "loss": 1.0104,
      "step": 327000
    },
    {
      "epoch": 2.4,
      "learning_rate": 5e-05,
      "loss": 1.0096,
      "step": 327500
    },
    {
      "epoch": 2.4,
      "learning_rate": 5e-05,
      "loss": 1.0092,
      "step": 328000
    },
    {
      "epoch": 2.4,
      "learning_rate": 5e-05,
      "loss": 1.0091,
      "step": 328500
    },
    {
      "epoch": 2.41,
      "learning_rate": 5e-05,
      "loss": 1.0063,
      "step": 329000
    },
    {
      "epoch": 2.41,
      "learning_rate": 5e-05,
      "loss": 1.0108,
      "step": 329500
    },
    {
      "epoch": 2.42,
      "learning_rate": 5e-05,
      "loss": 1.0093,
      "step": 330000
    },
    {
      "epoch": 2.42,
      "learning_rate": 5e-05,
      "loss": 1.0087,
      "step": 330500
    },
    {
      "epoch": 2.42,
      "learning_rate": 5e-05,
      "loss": 1.0118,
      "step": 331000
    },
    {
      "epoch": 2.43,
      "learning_rate": 5e-05,
      "loss": 1.011,
      "step": 331500
    },
    {
      "epoch": 2.43,
      "learning_rate": 5e-05,
      "loss": 1.009,
      "step": 332000
    },
    {
      "epoch": 2.43,
      "learning_rate": 5e-05,
      "loss": 1.0128,
      "step": 332500
    },
    {
      "epoch": 2.44,
      "learning_rate": 5e-05,
      "loss": 1.0027,
      "step": 333000
    },
    {
      "epoch": 2.44,
      "learning_rate": 5e-05,
      "loss": 1.008,
      "step": 333500
    },
    {
      "epoch": 2.44,
      "learning_rate": 5e-05,
      "loss": 1.0017,
      "step": 334000
    },
    {
      "epoch": 2.45,
      "learning_rate": 5e-05,
      "loss": 1.0039,
      "step": 334500
    },
    {
      "epoch": 2.45,
      "learning_rate": 5e-05,
      "loss": 1.0058,
      "step": 335000
    },
    {
      "epoch": 2.46,
      "learning_rate": 5e-05,
      "loss": 1.0124,
      "step": 335500
    },
    {
      "epoch": 2.46,
      "learning_rate": 5e-05,
      "loss": 1.0042,
      "step": 336000
    },
    {
      "epoch": 2.46,
      "learning_rate": 5e-05,
      "loss": 1.0025,
      "step": 336500
    },
    {
      "epoch": 2.47,
      "learning_rate": 5e-05,
      "loss": 1.003,
      "step": 337000
    },
    {
      "epoch": 2.47,
      "learning_rate": 5e-05,
      "loss": 1.0034,
      "step": 337500
    },
    {
      "epoch": 2.47,
      "learning_rate": 5e-05,
      "loss": 1.008,
      "step": 338000
    },
    {
      "epoch": 2.48,
      "learning_rate": 5e-05,
      "loss": 1.0097,
      "step": 338500
    },
    {
      "epoch": 2.48,
      "learning_rate": 5e-05,
      "loss": 1.0007,
      "step": 339000
    },
    {
      "epoch": 2.48,
      "learning_rate": 5e-05,
      "loss": 0.9979,
      "step": 339500
    },
    {
      "epoch": 2.49,
      "learning_rate": 5e-05,
      "loss": 1.0044,
      "step": 340000
    },
    {
      "epoch": 2.49,
      "learning_rate": 5e-05,
      "loss": 1.001,
      "step": 340500
    },
    {
      "epoch": 2.5,
      "learning_rate": 5e-05,
      "loss": 1.0032,
      "step": 341000
    },
    {
      "epoch": 2.5,
      "learning_rate": 5e-05,
      "loss": 1.0039,
      "step": 341500
    },
    {
      "epoch": 2.5,
      "learning_rate": 5e-05,
      "loss": 1.0087,
      "step": 342000
    },
    {
      "epoch": 2.51,
      "learning_rate": 5e-05,
      "loss": 1.0036,
      "step": 342500
    },
    {
      "epoch": 2.51,
      "learning_rate": 5e-05,
      "loss": 1.0086,
      "step": 343000
    },
    {
      "epoch": 2.51,
      "learning_rate": 5e-05,
      "loss": 1.0025,
      "step": 343500
    },
    {
      "epoch": 2.52,
      "learning_rate": 5e-05,
      "loss": 1.0081,
      "step": 344000
    },
    {
      "epoch": 2.52,
      "learning_rate": 5e-05,
      "loss": 1.0109,
      "step": 344500
    },
    {
      "epoch": 2.53,
      "learning_rate": 5e-05,
      "loss": 1.0124,
      "step": 345000
    },
    {
      "epoch": 2.53,
      "learning_rate": 5e-05,
      "loss": 1.0137,
      "step": 345500
    },
    {
      "epoch": 2.53,
      "learning_rate": 5e-05,
      "loss": 1.0027,
      "step": 346000
    },
    {
      "epoch": 2.54,
      "learning_rate": 5e-05,
      "loss": 1.0042,
      "step": 346500
    },
    {
      "epoch": 2.54,
      "learning_rate": 5e-05,
      "loss": 1.0065,
      "step": 347000
    },
    {
      "epoch": 2.54,
      "learning_rate": 5e-05,
      "loss": 1.0166,
      "step": 347500
    },
    {
      "epoch": 2.55,
      "learning_rate": 5e-05,
      "loss": 1.0162,
      "step": 348000
    },
    {
      "epoch": 2.55,
      "learning_rate": 5e-05,
      "loss": 1.0116,
      "step": 348500
    },
    {
      "epoch": 2.55,
      "learning_rate": 5e-05,
      "loss": 1.009,
      "step": 349000
    },
    {
      "epoch": 2.56,
      "learning_rate": 5e-05,
      "loss": 1.0118,
      "step": 349500
    },
    {
      "epoch": 2.56,
      "learning_rate": 5e-05,
      "loss": 1.0088,
      "step": 350000
    },
    {
      "epoch": 2.57,
      "learning_rate": 5e-05,
      "loss": 1.002,
      "step": 350500
    },
    {
      "epoch": 2.57,
      "learning_rate": 5e-05,
      "loss": 1.009,
      "step": 351000
    },
    {
      "epoch": 2.57,
      "learning_rate": 5e-05,
      "loss": 1.0109,
      "step": 351500
    },
    {
      "epoch": 2.58,
      "learning_rate": 5e-05,
      "loss": 1.0129,
      "step": 352000
    },
    {
      "epoch": 2.58,
      "learning_rate": 5e-05,
      "loss": 1.0085,
      "step": 352500
    },
    {
      "epoch": 2.58,
      "learning_rate": 5e-05,
      "loss": 1.0026,
      "step": 353000
    },
    {
      "epoch": 2.59,
      "learning_rate": 5e-05,
      "loss": 1.0009,
      "step": 353500
    },
    {
      "epoch": 2.59,
      "learning_rate": 5e-05,
      "loss": 1.0081,
      "step": 354000
    },
    {
      "epoch": 2.59,
      "learning_rate": 5e-05,
      "loss": 1.0086,
      "step": 354500
    },
    {
      "epoch": 2.6,
      "learning_rate": 5e-05,
      "loss": 1.0068,
      "step": 355000
    },
    {
      "epoch": 2.6,
      "learning_rate": 5e-05,
      "loss": 1.0118,
      "step": 355500
    },
    {
      "epoch": 2.61,
      "learning_rate": 5e-05,
      "loss": 1.0039,
      "step": 356000
    },
    {
      "epoch": 2.61,
      "learning_rate": 5e-05,
      "loss": 1.0015,
      "step": 356500
    },
    {
      "epoch": 2.61,
      "learning_rate": 5e-05,
      "loss": 1.0046,
      "step": 357000
    },
    {
      "epoch": 2.62,
      "learning_rate": 5e-05,
      "loss": 1.0015,
      "step": 357500
    },
    {
      "epoch": 2.62,
      "learning_rate": 5e-05,
      "loss": 1.0091,
      "step": 358000
    },
    {
      "epoch": 2.62,
      "learning_rate": 5e-05,
      "loss": 1.0128,
      "step": 358500
    },
    {
      "epoch": 2.63,
      "learning_rate": 5e-05,
      "loss": 1.005,
      "step": 359000
    },
    {
      "epoch": 2.63,
      "learning_rate": 5e-05,
      "loss": 1.0034,
      "step": 359500
    },
    {
      "epoch": 2.63,
      "learning_rate": 5e-05,
      "loss": 1.0061,
      "step": 360000
    },
    {
      "epoch": 2.64,
      "learning_rate": 5e-05,
      "loss": 1.0004,
      "step": 360500
    },
    {
      "epoch": 2.64,
      "learning_rate": 5e-05,
      "loss": 1.009,
      "step": 361000
    },
    {
      "epoch": 2.65,
      "learning_rate": 5e-05,
      "loss": 1.0038,
      "step": 361500
    },
    {
      "epoch": 2.65,
      "learning_rate": 5e-05,
      "loss": 1.0017,
      "step": 362000
    },
    {
      "epoch": 2.65,
      "learning_rate": 5e-05,
      "loss": 1.0012,
      "step": 362500
    },
    {
      "epoch": 2.66,
      "learning_rate": 5e-05,
      "loss": 1.0019,
      "step": 363000
    },
    {
      "epoch": 2.66,
      "learning_rate": 5e-05,
      "loss": 0.9962,
      "step": 363500
    },
    {
      "epoch": 2.66,
      "learning_rate": 5e-05,
      "loss": 1.0034,
      "step": 364000
    },
    {
      "epoch": 2.67,
      "learning_rate": 5e-05,
      "loss": 1.0062,
      "step": 364500
    },
    {
      "epoch": 2.67,
      "learning_rate": 5e-05,
      "loss": 1.0059,
      "step": 365000
    },
    {
      "epoch": 2.68,
      "learning_rate": 5e-05,
      "loss": 1.0079,
      "step": 365500
    },
    {
      "epoch": 2.68,
      "learning_rate": 5e-05,
      "loss": 1.0083,
      "step": 366000
    },
    {
      "epoch": 2.68,
      "learning_rate": 5e-05,
      "loss": 1.0113,
      "step": 366500
    },
    {
      "epoch": 2.69,
      "learning_rate": 5e-05,
      "loss": 1.0042,
      "step": 367000
    },
    {
      "epoch": 2.69,
      "learning_rate": 5e-05,
      "loss": 1.0047,
      "step": 367500
    },
    {
      "epoch": 2.69,
      "learning_rate": 5e-05,
      "loss": 1.0012,
      "step": 368000
    },
    {
      "epoch": 2.7,
      "learning_rate": 5e-05,
      "loss": 1.0021,
      "step": 368500
    },
    {
      "epoch": 2.7,
      "learning_rate": 5e-05,
      "loss": 1.0043,
      "step": 369000
    },
    {
      "epoch": 2.7,
      "learning_rate": 5e-05,
      "loss": 1.0016,
      "step": 369500
    },
    {
      "epoch": 2.71,
      "learning_rate": 5e-05,
      "loss": 1.0022,
      "step": 370000
    },
    {
      "epoch": 2.71,
      "learning_rate": 5e-05,
      "loss": 0.9942,
      "step": 370500
    },
    {
      "epoch": 2.72,
      "learning_rate": 5e-05,
      "loss": 1.0004,
      "step": 371000
    },
    {
      "epoch": 2.72,
      "learning_rate": 5e-05,
      "loss": 0.998,
      "step": 371500
    },
    {
      "epoch": 2.72,
      "learning_rate": 5e-05,
      "loss": 0.9981,
      "step": 372000
    },
    {
      "epoch": 2.73,
      "learning_rate": 5e-05,
      "loss": 0.9983,
      "step": 372500
    },
    {
      "epoch": 2.73,
      "learning_rate": 5e-05,
      "loss": 0.996,
      "step": 373000
    },
    {
      "epoch": 2.73,
      "learning_rate": 5e-05,
      "loss": 0.9905,
      "step": 373500
    },
    {
      "epoch": 2.74,
      "learning_rate": 5e-05,
      "loss": 0.9982,
      "step": 374000
    },
    {
      "epoch": 2.74,
      "learning_rate": 5e-05,
      "loss": 0.992,
      "step": 374500
    },
    {
      "epoch": 2.74,
      "learning_rate": 5e-05,
      "loss": 1.0035,
      "step": 375000
    },
    {
      "epoch": 2.75,
      "learning_rate": 5e-05,
      "loss": 0.9998,
      "step": 375500
    },
    {
      "epoch": 2.75,
      "learning_rate": 5e-05,
      "loss": 0.9997,
      "step": 376000
    },
    {
      "epoch": 2.76,
      "learning_rate": 5e-05,
      "loss": 0.995,
      "step": 376500
    },
    {
      "epoch": 2.76,
      "learning_rate": 5e-05,
      "loss": 0.9972,
      "step": 377000
    },
    {
      "epoch": 2.76,
      "learning_rate": 5e-05,
      "loss": 0.9965,
      "step": 377500
    },
    {
      "epoch": 2.77,
      "learning_rate": 5e-05,
      "loss": 0.9999,
      "step": 378000
    },
    {
      "epoch": 2.77,
      "learning_rate": 5e-05,
      "loss": 0.9981,
      "step": 378500
    },
    {
      "epoch": 2.77,
      "learning_rate": 5e-05,
      "loss": 1.0018,
      "step": 379000
    },
    {
      "epoch": 2.78,
      "learning_rate": 5e-05,
      "loss": 0.9977,
      "step": 379500
    },
    {
      "epoch": 2.78,
      "learning_rate": 5e-05,
      "loss": 0.9947,
      "step": 380000
    },
    {
      "epoch": 2.78,
      "learning_rate": 5e-05,
      "loss": 0.9954,
      "step": 380500
    },
    {
      "epoch": 2.79,
      "learning_rate": 5e-05,
      "loss": 0.9954,
      "step": 381000
    },
    {
      "epoch": 2.79,
      "learning_rate": 5e-05,
      "loss": 1.002,
      "step": 381500
    },
    {
      "epoch": 2.8,
      "learning_rate": 5e-05,
      "loss": 0.9944,
      "step": 382000
    },
    {
      "epoch": 2.8,
      "learning_rate": 5e-05,
      "loss": 0.9964,
      "step": 382500
    },
    {
      "epoch": 2.8,
      "learning_rate": 5e-05,
      "loss": 1.0025,
      "step": 383000
    },
    {
      "epoch": 2.81,
      "learning_rate": 5e-05,
      "loss": 1.0045,
      "step": 383500
    },
    {
      "epoch": 2.81,
      "learning_rate": 5e-05,
      "loss": 0.9984,
      "step": 384000
    },
    {
      "epoch": 2.81,
      "eval_accuracy": 0.7979610372060427,
      "eval_loss": 0.9764020442962646,
      "eval_runtime": 32.733,
      "eval_samples_per_second": 409.862,
      "eval_steps_per_second": 8.554,
      "step": 384000
    },
    {
      "epoch": 2.81,
      "learning_rate": 5e-05,
      "loss": 0.9979,
      "step": 384500
    },
    {
      "epoch": 2.82,
      "learning_rate": 5e-05,
      "loss": 0.991,
      "step": 385000
    },
    {
      "epoch": 2.82,
      "learning_rate": 5e-05,
      "loss": 0.9959,
      "step": 385500
    },
    {
      "epoch": 2.83,
      "learning_rate": 5e-05,
      "loss": 0.999,
      "step": 386000
    },
    {
      "epoch": 2.83,
      "learning_rate": 5e-05,
      "loss": 0.9982,
      "step": 386500
    },
    {
      "epoch": 2.83,
      "learning_rate": 5e-05,
      "loss": 1.0048,
      "step": 387000
    },
    {
      "epoch": 2.84,
      "learning_rate": 5e-05,
      "loss": 0.9964,
      "step": 387500
    },
    {
      "epoch": 2.84,
      "learning_rate": 5e-05,
      "loss": 0.9956,
      "step": 388000
    },
    {
      "epoch": 2.84,
      "learning_rate": 5e-05,
      "loss": 0.997,
      "step": 388500
    },
    {
      "epoch": 2.85,
      "learning_rate": 5e-05,
      "loss": 0.9915,
      "step": 389000
    },
    {
      "epoch": 2.85,
      "learning_rate": 5e-05,
      "loss": 0.9912,
      "step": 389500
    },
    {
      "epoch": 2.85,
      "learning_rate": 5e-05,
      "loss": 0.9944,
      "step": 390000
    },
    {
      "epoch": 2.86,
      "learning_rate": 5e-05,
      "loss": 0.9965,
      "step": 390500
    },
    {
      "epoch": 2.86,
      "learning_rate": 5e-05,
      "loss": 0.9899,
      "step": 391000
    },
    {
      "epoch": 2.87,
      "learning_rate": 5e-05,
      "loss": 1.0008,
      "step": 391500
    },
    {
      "epoch": 2.87,
      "learning_rate": 5e-05,
      "loss": 0.9898,
      "step": 392000
    },
    {
      "epoch": 2.87,
      "learning_rate": 5e-05,
      "loss": 0.9999,
      "step": 392500
    },
    {
      "epoch": 2.88,
      "learning_rate": 5e-05,
      "loss": 0.9974,
      "step": 393000
    },
    {
      "epoch": 2.88,
      "learning_rate": 5e-05,
      "loss": 0.9946,
      "step": 393500
    },
    {
      "epoch": 2.88,
      "learning_rate": 5e-05,
      "loss": 0.996,
      "step": 394000
    },
    {
      "epoch": 2.89,
      "learning_rate": 5e-05,
      "loss": 0.9921,
      "step": 394500
    },
    {
      "epoch": 2.89,
      "learning_rate": 5e-05,
      "loss": 0.9977,
      "step": 395000
    },
    {
      "epoch": 2.89,
      "learning_rate": 5e-05,
      "loss": 1.0009,
      "step": 395500
    },
    {
      "epoch": 2.9,
      "learning_rate": 5e-05,
      "loss": 0.9879,
      "step": 396000
    },
    {
      "epoch": 2.9,
      "learning_rate": 5e-05,
      "loss": 0.999,
      "step": 396500
    },
    {
      "epoch": 2.91,
      "learning_rate": 5e-05,
      "loss": 0.9994,
      "step": 397000
    },
    {
      "epoch": 2.91,
      "learning_rate": 5e-05,
      "loss": 0.9971,
      "step": 397500
    },
    {
      "epoch": 2.91,
      "learning_rate": 5e-05,
      "loss": 0.9913,
      "step": 398000
    },
    {
      "epoch": 2.92,
      "learning_rate": 5e-05,
      "loss": 0.9971,
      "step": 398500
    },
    {
      "epoch": 2.92,
      "learning_rate": 5e-05,
      "loss": 1.003,
      "step": 399000
    },
    {
      "epoch": 2.92,
      "learning_rate": 5e-05,
      "loss": 0.9903,
      "step": 399500
    },
    {
      "epoch": 2.93,
      "learning_rate": 5e-05,
      "loss": 0.9929,
      "step": 400000
    },
    {
      "epoch": 2.93,
      "learning_rate": 5e-05,
      "loss": 0.9959,
      "step": 400500
    },
    {
      "epoch": 2.93,
      "learning_rate": 5e-05,
      "loss": 0.9972,
      "step": 401000
    },
    {
      "epoch": 2.94,
      "learning_rate": 5e-05,
      "loss": 0.9964,
      "step": 401500
    },
    {
      "epoch": 2.94,
      "learning_rate": 5e-05,
      "loss": 0.9963,
      "step": 402000
    },
    {
      "epoch": 2.95,
      "learning_rate": 5e-05,
      "loss": 0.9954,
      "step": 402500
    },
    {
      "epoch": 2.95,
      "learning_rate": 5e-05,
      "loss": 0.993,
      "step": 403000
    },
    {
      "epoch": 2.95,
      "learning_rate": 5e-05,
      "loss": 0.9961,
      "step": 403500
    },
    {
      "epoch": 2.96,
      "learning_rate": 5e-05,
      "loss": 0.9967,
      "step": 404000
    },
    {
      "epoch": 2.96,
      "learning_rate": 5e-05,
      "loss": 1.0006,
      "step": 404500
    },
    {
      "epoch": 2.96,
      "learning_rate": 5e-05,
      "loss": 0.9901,
      "step": 405000
    },
    {
      "epoch": 2.97,
      "learning_rate": 5e-05,
      "loss": 0.9988,
      "step": 405500
    },
    {
      "epoch": 2.97,
      "learning_rate": 5e-05,
      "loss": 1.0004,
      "step": 406000
    },
    {
      "epoch": 2.98,
      "learning_rate": 5e-05,
      "loss": 1.0004,
      "step": 406500
    },
    {
      "epoch": 2.98,
      "learning_rate": 5e-05,
      "loss": 0.991,
      "step": 407000
    },
    {
      "epoch": 2.98,
      "learning_rate": 5e-05,
      "loss": 0.9978,
      "step": 407500
    },
    {
      "epoch": 2.99,
      "learning_rate": 5e-05,
      "loss": 0.9928,
      "step": 408000
    },
    {
      "epoch": 2.99,
      "learning_rate": 5e-05,
      "loss": 1.0039,
      "step": 408500
    },
    {
      "epoch": 2.99,
      "learning_rate": 5e-05,
      "loss": 0.993,
      "step": 409000
    },
    {
      "epoch": 3.0,
      "learning_rate": 5e-05,
      "loss": 0.9899,
      "step": 409500
    },
    {
      "epoch": 3.0,
      "step": 409890,
      "total_flos": 3.6701198802265047e+19,
      "train_loss": 0.06289897245738281,
      "train_runtime": 14736.1316,
      "train_samples_per_second": 2670.259,
      "train_steps_per_second": 27.815
    }
  ],
  "max_steps": 409890,
  "num_train_epochs": 3,
  "total_flos": 3.6701198802265047e+19,
  "trial_name": null,
  "trial_params": null
}