{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 37.51465416178195,
  "global_step": 32000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.12,
      "learning_rate": 5e-06,
      "loss": 0.8323,
      "step": 100
    },
    {
      "epoch": 0.23,
      "learning_rate": 5e-06,
      "loss": 0.793,
      "step": 200
    },
    {
      "epoch": 0.35,
      "learning_rate": 5e-06,
      "loss": 0.7508,
      "step": 300
    },
    {
      "epoch": 0.47,
      "learning_rate": 5e-06,
      "loss": 0.7777,
      "step": 400
    },
    {
      "epoch": 0.59,
      "learning_rate": 5e-06,
      "loss": 0.7741,
      "step": 500
    },
    {
      "epoch": 0.7,
      "learning_rate": 5e-06,
      "loss": 0.7435,
      "step": 600
    },
    {
      "epoch": 0.82,
      "learning_rate": 5e-06,
      "loss": 0.7681,
      "step": 700
    },
    {
      "epoch": 0.94,
      "learning_rate": 5e-06,
      "loss": 0.7505,
      "step": 800
    },
    {
      "epoch": 1.06,
      "learning_rate": 5e-06,
      "loss": 0.7285,
      "step": 900
    },
    {
      "epoch": 1.17,
      "learning_rate": 5e-06,
      "loss": 0.7243,
      "step": 1000
    },
    {
      "epoch": 1.29,
      "learning_rate": 5e-06,
      "loss": 0.7562,
      "step": 1100
    },
    {
      "epoch": 1.41,
      "learning_rate": 5e-06,
      "loss": 0.743,
      "step": 1200
    },
    {
      "epoch": 1.52,
      "learning_rate": 5e-06,
      "loss": 0.7086,
      "step": 1300
    },
    {
      "epoch": 1.64,
      "learning_rate": 5e-06,
      "loss": 0.7177,
      "step": 1400
    },
    {
      "epoch": 1.76,
      "learning_rate": 5e-06,
      "loss": 0.727,
      "step": 1500
    },
    {
      "epoch": 1.76,
      "eval_loss": 0.6887246966362,
      "eval_runtime": 70.0491,
      "eval_samples_per_second": 3.041,
      "eval_steps_per_second": 3.041,
      "step": 1500
    },
    {
      "epoch": 1.88,
      "learning_rate": 5e-06,
      "loss": 0.7781,
      "step": 1600
    },
    {
      "epoch": 1.99,
      "learning_rate": 5e-06,
      "loss": 0.7509,
      "step": 1700
    },
    {
      "epoch": 2.11,
      "learning_rate": 5e-06,
      "loss": 0.7057,
      "step": 1800
    },
    {
      "epoch": 2.23,
      "learning_rate": 5e-06,
      "loss": 0.7292,
      "step": 1900
    },
    {
      "epoch": 2.34,
      "learning_rate": 5e-06,
      "loss": 0.7037,
      "step": 2000
    },
    {
      "epoch": 2.46,
      "learning_rate": 5e-06,
      "loss": 0.7005,
      "step": 2100
    },
    {
      "epoch": 2.58,
      "learning_rate": 5e-06,
      "loss": 0.7501,
      "step": 2200
    },
    {
      "epoch": 2.7,
      "learning_rate": 5e-06,
      "loss": 0.7162,
      "step": 2300
    },
    {
      "epoch": 2.81,
      "learning_rate": 5e-06,
      "loss": 0.7428,
      "step": 2400
    },
    {
      "epoch": 2.93,
      "learning_rate": 5e-06,
      "loss": 0.7403,
      "step": 2500
    },
    {
      "epoch": 3.05,
      "learning_rate": 5e-06,
      "loss": 0.7286,
      "step": 2600
    },
    {
      "epoch": 3.17,
      "learning_rate": 5e-06,
      "loss": 0.72,
      "step": 2700
    },
    {
      "epoch": 3.28,
      "learning_rate": 5e-06,
      "loss": 0.6998,
      "step": 2800
    },
    {
      "epoch": 3.4,
      "learning_rate": 5e-06,
      "loss": 0.7515,
      "step": 2900
    },
    {
      "epoch": 3.52,
      "learning_rate": 5e-06,
      "loss": 0.7263,
      "step": 3000
    },
    {
      "epoch": 3.52,
      "eval_loss": 0.6762834191322327,
      "eval_runtime": 70.0592,
      "eval_samples_per_second": 3.04,
      "eval_steps_per_second": 3.04,
      "step": 3000
    },
    {
      "epoch": 3.63,
      "learning_rate": 5e-06,
      "loss": 0.6931,
      "step": 3100
    },
    {
      "epoch": 3.75,
      "learning_rate": 5e-06,
      "loss": 0.6907,
      "step": 3200
    },
    {
      "epoch": 3.87,
      "learning_rate": 5e-06,
      "loss": 0.6818,
      "step": 3300
    },
    {
      "epoch": 3.99,
      "learning_rate": 5e-06,
      "loss": 0.7417,
      "step": 3400
    },
    {
      "epoch": 4.1,
      "learning_rate": 5e-06,
      "loss": 0.6987,
      "step": 3500
    },
    {
      "epoch": 4.22,
      "learning_rate": 5e-06,
      "loss": 0.6854,
      "step": 3600
    },
    {
      "epoch": 4.34,
      "learning_rate": 5e-06,
      "loss": 0.7236,
      "step": 3700
    },
    {
      "epoch": 4.45,
      "learning_rate": 5e-06,
      "loss": 0.7045,
      "step": 3800
    },
    {
      "epoch": 4.57,
      "learning_rate": 5e-06,
      "loss": 0.699,
      "step": 3900
    },
    {
      "epoch": 4.69,
      "learning_rate": 5e-06,
      "loss": 0.7179,
      "step": 4000
    },
    {
      "epoch": 4.81,
      "learning_rate": 5e-06,
      "loss": 0.7066,
      "step": 4100
    },
    {
      "epoch": 4.92,
      "learning_rate": 5e-06,
      "loss": 0.713,
      "step": 4200
    },
    {
      "epoch": 5.04,
      "learning_rate": 5e-06,
      "loss": 0.6749,
      "step": 4300
    },
    {
      "epoch": 5.16,
      "learning_rate": 5e-06,
      "loss": 0.6882,
      "step": 4400
    },
    {
      "epoch": 5.28,
      "learning_rate": 5e-06,
      "loss": 0.6809,
      "step": 4500
    },
    {
      "epoch": 5.28,
      "eval_loss": 0.6606600284576416,
      "eval_runtime": 69.4571,
      "eval_samples_per_second": 3.067,
      "eval_steps_per_second": 3.067,
      "step": 4500
    },
    {
      "epoch": 5.39,
      "learning_rate": 5e-06,
      "loss": 0.7433,
      "step": 4600
    },
    {
      "epoch": 5.51,
      "learning_rate": 5e-06,
      "loss": 0.6881,
      "step": 4700
    },
    {
      "epoch": 5.63,
      "learning_rate": 5e-06,
      "loss": 0.7343,
      "step": 4800
    },
    {
      "epoch": 5.74,
      "learning_rate": 5e-06,
      "loss": 0.6675,
      "step": 4900
    },
    {
      "epoch": 5.86,
      "learning_rate": 5e-06,
      "loss": 0.7187,
      "step": 5000
    },
    {
      "epoch": 5.98,
      "learning_rate": 5e-06,
      "loss": 0.6837,
      "step": 5100
    },
    {
      "epoch": 6.1,
      "learning_rate": 5e-06,
      "loss": 0.6825,
      "step": 5200
    },
    {
      "epoch": 6.21,
      "learning_rate": 5e-06,
      "loss": 0.6976,
      "step": 5300
    },
    {
      "epoch": 6.33,
      "learning_rate": 5e-06,
      "loss": 0.7055,
      "step": 5400
    },
    {
      "epoch": 6.45,
      "learning_rate": 5e-06,
      "loss": 0.6584,
      "step": 5500
    },
    {
      "epoch": 6.57,
      "learning_rate": 5e-06,
      "loss": 0.6819,
      "step": 5600
    },
    {
      "epoch": 6.68,
      "learning_rate": 5e-06,
      "loss": 0.6652,
      "step": 5700
    },
    {
      "epoch": 6.8,
      "learning_rate": 5e-06,
      "loss": 0.6728,
      "step": 5800
    },
    {
      "epoch": 6.92,
      "learning_rate": 5e-06,
      "loss": 0.6916,
      "step": 5900
    },
    {
      "epoch": 7.03,
      "learning_rate": 5e-06,
      "loss": 0.657,
      "step": 6000
    },
    {
      "epoch": 7.03,
      "eval_loss": 0.6467106938362122,
      "eval_runtime": 69.8327,
      "eval_samples_per_second": 3.05,
      "eval_steps_per_second": 3.05,
      "step": 6000
    },
    {
      "epoch": 7.15,
      "learning_rate": 5e-06,
      "loss": 0.7331,
      "step": 6100
    },
    {
      "epoch": 7.27,
      "learning_rate": 5e-06,
      "loss": 0.6598,
      "step": 6200
    },
    {
      "epoch": 7.39,
      "learning_rate": 5e-06,
      "loss": 0.6602,
      "step": 6300
    },
    {
      "epoch": 7.5,
      "learning_rate": 5e-06,
      "loss": 0.6819,
      "step": 6400
    },
    {
      "epoch": 7.62,
      "learning_rate": 5e-06,
      "loss": 0.6764,
      "step": 6500
    },
    {
      "epoch": 7.74,
      "learning_rate": 5e-06,
      "loss": 0.6674,
      "step": 6600
    },
    {
      "epoch": 7.85,
      "learning_rate": 5e-06,
      "loss": 0.6848,
      "step": 6700
    },
    {
      "epoch": 7.97,
      "learning_rate": 5e-06,
      "loss": 0.6446,
      "step": 6800
    },
    {
      "epoch": 8.09,
      "learning_rate": 5e-06,
      "loss": 0.6601,
      "step": 6900
    },
    {
      "epoch": 8.21,
      "learning_rate": 5e-06,
      "loss": 0.6649,
      "step": 7000
    },
    {
      "epoch": 8.32,
      "learning_rate": 5e-06,
      "loss": 0.657,
      "step": 7100
    },
    {
      "epoch": 8.44,
      "learning_rate": 5e-06,
      "loss": 0.6326,
      "step": 7200
    },
    {
      "epoch": 8.56,
      "learning_rate": 5e-06,
      "loss": 0.6541,
      "step": 7300
    },
    {
      "epoch": 8.68,
      "learning_rate": 5e-06,
      "loss": 0.6579,
      "step": 7400
    },
    {
      "epoch": 8.79,
      "learning_rate": 5e-06,
      "loss": 0.6784,
      "step": 7500
    },
    {
      "epoch": 8.79,
      "eval_loss": 0.6341073513031006,
      "eval_runtime": 70.1746,
      "eval_samples_per_second": 3.035,
      "eval_steps_per_second": 3.035,
      "step": 7500
    },
    {
      "epoch": 8.91,
      "learning_rate": 5e-06,
      "loss": 0.6889,
      "step": 7600
    },
    {
      "epoch": 9.03,
      "learning_rate": 5e-06,
      "loss": 0.6752,
      "step": 7700
    },
    {
      "epoch": 9.14,
      "learning_rate": 5e-06,
      "loss": 0.6654,
      "step": 7800
    },
    {
      "epoch": 9.26,
      "learning_rate": 5e-06,
      "loss": 0.6516,
      "step": 7900
    },
    {
      "epoch": 9.38,
      "learning_rate": 5e-06,
      "loss": 0.6847,
      "step": 8000
    },
    {
      "epoch": 9.5,
      "learning_rate": 5e-06,
      "loss": 0.6396,
      "step": 8100
    },
    {
      "epoch": 9.61,
      "learning_rate": 5e-06,
      "loss": 0.6484,
      "step": 8200
    },
    {
      "epoch": 9.73,
      "learning_rate": 5e-06,
      "loss": 0.6396,
      "step": 8300
    },
    {
      "epoch": 9.85,
      "learning_rate": 5e-06,
      "loss": 0.6951,
      "step": 8400
    },
    {
      "epoch": 9.96,
      "learning_rate": 5e-06,
      "loss": 0.641,
      "step": 8500
    },
    {
      "epoch": 10.08,
      "learning_rate": 5e-06,
      "loss": 0.6379,
      "step": 8600
    },
    {
      "epoch": 10.2,
      "learning_rate": 5e-06,
      "loss": 0.6264,
      "step": 8700
    },
    {
      "epoch": 10.32,
      "learning_rate": 5e-06,
      "loss": 0.6364,
      "step": 8800
    },
    {
      "epoch": 10.43,
      "learning_rate": 5e-06,
      "loss": 0.676,
      "step": 8900
    },
    {
      "epoch": 10.55,
      "learning_rate": 5e-06,
      "loss": 0.6756,
      "step": 9000
    },
    {
      "epoch": 10.55,
      "eval_loss": 0.6359875202178955,
      "eval_runtime": 70.4519,
      "eval_samples_per_second": 3.023,
      "eval_steps_per_second": 3.023,
      "step": 9000
    },
    {
      "epoch": 10.67,
      "learning_rate": 5e-06,
      "loss": 0.6641,
      "step": 9100
    },
    {
      "epoch": 10.79,
      "learning_rate": 5e-06,
      "loss": 0.6126,
      "step": 9200
    },
    {
      "epoch": 10.9,
      "learning_rate": 5e-06,
      "loss": 0.6538,
      "step": 9300
    },
    {
      "epoch": 11.02,
      "learning_rate": 5e-06,
      "loss": 0.641,
      "step": 9400
    },
    {
      "epoch": 11.14,
      "learning_rate": 5e-06,
      "loss": 0.6501,
      "step": 9500
    },
    {
      "epoch": 11.25,
      "learning_rate": 5e-06,
      "loss": 0.647,
      "step": 9600
    },
    {
      "epoch": 11.37,
      "learning_rate": 5e-06,
      "loss": 0.6463,
      "step": 9700
    },
    {
      "epoch": 11.49,
      "learning_rate": 5e-06,
      "loss": 0.6507,
      "step": 9800
    },
    {
      "epoch": 11.61,
      "learning_rate": 5e-06,
      "loss": 0.6525,
      "step": 9900
    },
    {
      "epoch": 11.72,
      "learning_rate": 5e-06,
      "loss": 0.6194,
      "step": 10000
    },
    {
      "epoch": 11.84,
      "learning_rate": 5e-06,
      "loss": 0.6338,
      "step": 10100
    },
    {
      "epoch": 11.96,
      "learning_rate": 5e-06,
      "loss": 0.6492,
      "step": 10200
    },
    {
      "epoch": 12.08,
      "learning_rate": 5e-06,
      "loss": 0.6531,
      "step": 10300
    },
    {
      "epoch": 12.19,
      "learning_rate": 5e-06,
      "loss": 0.6073,
      "step": 10400
    },
    {
      "epoch": 12.31,
      "learning_rate": 5e-06,
      "loss": 0.6307,
      "step": 10500
    },
    {
      "epoch": 12.31,
      "eval_loss": 0.6309817433357239,
      "eval_runtime": 70.2077,
      "eval_samples_per_second": 3.034,
      "eval_steps_per_second": 3.034,
      "step": 10500
    },
    {
      "epoch": 12.43,
      "learning_rate": 5e-06,
      "loss": 0.6608,
      "step": 10600
    },
    {
      "epoch": 12.54,
      "learning_rate": 5e-06,
      "loss": 0.6252,
      "step": 10700
    },
    {
      "epoch": 12.66,
      "learning_rate": 5e-06,
      "loss": 0.6258,
      "step": 10800
    },
    {
      "epoch": 12.78,
      "learning_rate": 5e-06,
      "loss": 0.6504,
      "step": 10900
    },
    {
      "epoch": 12.9,
      "learning_rate": 5e-06,
      "loss": 0.6281,
      "step": 11000
    },
    {
      "epoch": 13.01,
      "learning_rate": 5e-06,
      "loss": 0.6398,
      "step": 11100
    },
    {
      "epoch": 13.13,
      "learning_rate": 5e-06,
      "loss": 0.6318,
      "step": 11200
    },
    {
      "epoch": 13.25,
      "learning_rate": 5e-06,
      "loss": 0.6162,
      "step": 11300
    },
    {
      "epoch": 13.36,
      "learning_rate": 5e-06,
      "loss": 0.6101,
      "step": 11400
    },
    {
      "epoch": 13.48,
      "learning_rate": 5e-06,
      "loss": 0.6124,
      "step": 11500
    },
    {
      "epoch": 13.6,
      "learning_rate": 5e-06,
      "loss": 0.5994,
      "step": 11600
    },
    {
      "epoch": 13.72,
      "learning_rate": 5e-06,
      "loss": 0.6599,
      "step": 11700
    },
    {
      "epoch": 13.83,
      "learning_rate": 5e-06,
      "loss": 0.6192,
      "step": 11800
    },
    {
      "epoch": 13.95,
      "learning_rate": 5e-06,
      "loss": 0.6341,
      "step": 11900
    },
    {
      "epoch": 14.07,
      "learning_rate": 5e-06,
      "loss": 0.6155,
      "step": 12000
    },
    {
      "epoch": 14.07,
      "eval_loss": 0.63248211145401,
      "eval_runtime": 70.8191,
      "eval_samples_per_second": 3.008,
      "eval_steps_per_second": 3.008,
      "step": 12000
    },
    {
      "epoch": 14.19,
      "learning_rate": 5e-06,
      "loss": 0.6562,
      "step": 12100
    },
    {
      "epoch": 14.3,
      "learning_rate": 5e-06,
      "loss": 0.633,
      "step": 12200
    },
    {
      "epoch": 14.42,
      "learning_rate": 5e-06,
      "loss": 0.6169,
      "step": 12300
    },
    {
      "epoch": 14.54,
      "learning_rate": 5e-06,
      "loss": 0.6312,
      "step": 12400
    },
    {
      "epoch": 14.65,
      "learning_rate": 5e-06,
      "loss": 0.6401,
      "step": 12500
    },
    {
      "epoch": 14.77,
      "learning_rate": 5e-06,
      "loss": 0.6365,
      "step": 12600
    },
    {
      "epoch": 14.89,
      "learning_rate": 5e-06,
      "loss": 0.6286,
      "step": 12700
    },
    {
      "epoch": 15.01,
      "learning_rate": 5e-06,
      "loss": 0.5877,
      "step": 12800
    },
    {
      "epoch": 15.12,
      "learning_rate": 5e-06,
      "loss": 0.6334,
      "step": 12900
    },
    {
      "epoch": 15.24,
      "learning_rate": 5e-06,
      "loss": 0.5785,
      "step": 13000
    },
    {
      "epoch": 15.36,
      "learning_rate": 5e-06,
      "loss": 0.6155,
      "step": 13100
    },
    {
      "epoch": 15.47,
      "learning_rate": 5e-06,
      "loss": 0.6404,
      "step": 13200
    },
    {
      "epoch": 15.59,
      "learning_rate": 5e-06,
      "loss": 0.6302,
      "step": 13300
    },
    {
      "epoch": 15.71,
      "learning_rate": 5e-06,
      "loss": 0.6154,
      "step": 13400
    },
    {
      "epoch": 15.83,
      "learning_rate": 5e-06,
      "loss": 0.6119,
      "step": 13500
    },
    {
      "epoch": 15.83,
      "eval_loss": 0.6228322386741638,
      "eval_runtime": 70.1345,
      "eval_samples_per_second": 3.037,
      "eval_steps_per_second": 3.037,
      "step": 13500
    },
    {
      "epoch": 15.94,
      "learning_rate": 5e-06,
      "loss": 0.6193,
      "step": 13600
    },
    {
      "epoch": 16.06,
      "learning_rate": 5e-06,
      "loss": 0.6161,
      "step": 13700
    },
    {
      "epoch": 16.18,
      "learning_rate": 5e-06,
      "loss": 0.5785,
      "step": 13800
    },
    {
      "epoch": 16.3,
      "learning_rate": 5e-06,
      "loss": 0.6043,
      "step": 13900
    },
    {
      "epoch": 16.41,
      "learning_rate": 5e-06,
      "loss": 0.6205,
      "step": 14000
    },
    {
      "epoch": 16.53,
      "learning_rate": 5e-06,
      "loss": 0.6321,
      "step": 14100
    },
    {
      "epoch": 16.65,
      "learning_rate": 5e-06,
      "loss": 0.5996,
      "step": 14200
    },
    {
      "epoch": 16.76,
      "learning_rate": 5e-06,
      "loss": 0.6232,
      "step": 14300
    },
    {
      "epoch": 16.88,
      "learning_rate": 5e-06,
      "loss": 0.6148,
      "step": 14400
    },
    {
      "epoch": 17.0,
      "learning_rate": 5e-06,
      "loss": 0.602,
      "step": 14500
    },
    {
      "epoch": 17.12,
      "learning_rate": 5e-06,
      "loss": 0.5716,
      "step": 14600
    },
    {
      "epoch": 17.23,
      "learning_rate": 5e-06,
      "loss": 0.629,
      "step": 14700
    },
    {
      "epoch": 17.35,
      "learning_rate": 5e-06,
      "loss": 0.6134,
      "step": 14800
    },
    {
      "epoch": 17.47,
      "learning_rate": 5e-06,
      "loss": 0.6023,
      "step": 14900
    },
    {
      "epoch": 17.58,
      "learning_rate": 5e-06,
      "loss": 0.5943,
      "step": 15000
    },
    {
      "epoch": 17.58,
      "eval_loss": 0.623548686504364,
      "eval_runtime": 70.4978,
      "eval_samples_per_second": 3.021,
      "eval_steps_per_second": 3.021,
      "step": 15000
    },
    {
      "epoch": 17.7,
      "learning_rate": 5e-06,
      "loss": 0.6092,
      "step": 15100
    },
    {
      "epoch": 17.82,
      "learning_rate": 5e-06,
      "loss": 0.6172,
      "step": 15200
    },
    {
      "epoch": 17.94,
      "learning_rate": 5e-06,
      "loss": 0.6247,
      "step": 15300
    },
    {
      "epoch": 18.05,
      "learning_rate": 5e-06,
      "loss": 0.6043,
      "step": 15400
    },
    {
      "epoch": 18.17,
      "learning_rate": 5e-06,
      "loss": 0.6186,
      "step": 15500
    },
    {
      "epoch": 18.29,
      "learning_rate": 5e-06,
      "loss": 0.5877,
      "step": 15600
    },
    {
      "epoch": 18.41,
      "learning_rate": 5e-06,
      "loss": 0.5993,
      "step": 15700
    },
    {
      "epoch": 18.52,
      "learning_rate": 5e-06,
      "loss": 0.5949,
      "step": 15800
    },
    {
      "epoch": 18.64,
      "learning_rate": 5e-06,
      "loss": 0.5775,
      "step": 15900
    },
    {
      "epoch": 18.76,
      "learning_rate": 5e-06,
      "loss": 0.6147,
      "step": 16000
    },
    {
      "epoch": 18.87,
      "learning_rate": 5e-06,
      "loss": 0.5973,
      "step": 16100
    },
    {
      "epoch": 18.99,
      "learning_rate": 5e-06,
      "loss": 0.6103,
      "step": 16200
    },
    {
      "epoch": 19.11,
      "learning_rate": 5e-06,
      "loss": 0.6024,
      "step": 16300
    },
    {
      "epoch": 19.23,
      "learning_rate": 5e-06,
      "loss": 0.5729,
      "step": 16400
    },
    {
      "epoch": 19.34,
      "learning_rate": 5e-06,
      "loss": 0.6012,
      "step": 16500
    },
    {
      "epoch": 19.34,
      "eval_loss": 0.6155942678451538,
      "eval_runtime": 70.5596,
      "eval_samples_per_second": 3.019,
      "eval_steps_per_second": 3.019,
      "step": 16500
    },
    {
      "epoch": 19.46,
      "learning_rate": 5e-06,
      "loss": 0.6123,
      "step": 16600
    },
    {
      "epoch": 19.58,
      "learning_rate": 5e-06,
      "loss": 0.5937,
      "step": 16700
    },
    {
      "epoch": 19.7,
      "learning_rate": 5e-06,
      "loss": 0.5824,
      "step": 16800
    },
    {
      "epoch": 19.81,
      "learning_rate": 5e-06,
      "loss": 0.6433,
      "step": 16900
    },
    {
      "epoch": 19.93,
      "learning_rate": 5e-06,
      "loss": 0.5799,
      "step": 17000
    },
    {
      "epoch": 20.05,
      "learning_rate": 5e-06,
      "loss": 0.593,
      "step": 17100
    },
    {
      "epoch": 20.16,
      "learning_rate": 5e-06,
      "loss": 0.5909,
      "step": 17200
    },
    {
      "epoch": 20.28,
      "learning_rate": 5e-06,
      "loss": 0.5918,
      "step": 17300
    },
    {
      "epoch": 20.4,
      "learning_rate": 5e-06,
      "loss": 0.5908,
      "step": 17400
    },
    {
      "epoch": 20.52,
      "learning_rate": 5e-06,
      "loss": 0.5932,
      "step": 17500
    },
    {
      "epoch": 20.63,
      "learning_rate": 5e-06,
      "loss": 0.6085,
      "step": 17600
    },
    {
      "epoch": 20.75,
      "learning_rate": 5e-06,
      "loss": 0.5737,
      "step": 17700
    },
    {
      "epoch": 20.87,
      "learning_rate": 5e-06,
      "loss": 0.5926,
      "step": 17800
    },
    {
      "epoch": 20.98,
      "learning_rate": 5e-06,
      "loss": 0.606,
      "step": 17900
    },
    {
      "epoch": 21.1,
      "learning_rate": 5e-06,
      "loss": 0.5834,
      "step": 18000
    },
    {
      "epoch": 21.1,
      "eval_loss": 0.6064698100090027,
      "eval_runtime": 70.551,
      "eval_samples_per_second": 3.019,
      "eval_steps_per_second": 3.019,
      "step": 18000
    },
    {
      "epoch": 21.22,
      "learning_rate": 5e-06,
      "loss": 0.57,
      "step": 18100
    },
    {
      "epoch": 21.34,
      "learning_rate": 5e-06,
      "loss": 0.5878,
      "step": 18200
    },
    {
      "epoch": 21.45,
      "learning_rate": 5e-06,
      "loss": 0.5623,
      "step": 18300
    },
    {
      "epoch": 21.57,
      "learning_rate": 5e-06,
      "loss": 0.5978,
      "step": 18400
    },
    {
      "epoch": 21.69,
      "learning_rate": 5e-06,
      "loss": 0.594,
      "step": 18500
    },
    {
      "epoch": 21.81,
      "learning_rate": 5e-06,
      "loss": 0.6013,
      "step": 18600
    },
    {
      "epoch": 21.92,
      "learning_rate": 5e-06,
      "loss": 0.5576,
      "step": 18700
    },
    {
      "epoch": 22.04,
      "learning_rate": 5e-06,
      "loss": 0.5794,
      "step": 18800
    },
    {
      "epoch": 22.16,
      "learning_rate": 5e-06,
      "loss": 0.5863,
      "step": 18900
    },
    {
      "epoch": 22.27,
      "learning_rate": 5e-06,
      "loss": 0.5956,
      "step": 19000
    },
    {
      "epoch": 22.39,
      "learning_rate": 5e-06,
      "loss": 0.5849,
      "step": 19100
    },
    {
      "epoch": 22.51,
      "learning_rate": 5e-06,
      "loss": 0.5705,
      "step": 19200
    },
    {
      "epoch": 22.63,
      "learning_rate": 5e-06,
      "loss": 0.5945,
      "step": 19300
    },
    {
      "epoch": 22.74,
      "learning_rate": 5e-06,
      "loss": 0.5673,
      "step": 19400
    },
    {
      "epoch": 22.86,
      "learning_rate": 5e-06,
      "loss": 0.5942,
      "step": 19500
    },
    {
      "epoch": 22.86,
      "eval_loss": 0.6019883155822754,
      "eval_runtime": 70.6262,
      "eval_samples_per_second": 3.016,
      "eval_steps_per_second": 3.016,
      "step": 19500
    },
    {
      "epoch": 22.98,
      "learning_rate": 5e-06,
      "loss": 0.556,
      "step": 19600
    },
    {
      "epoch": 23.09,
      "learning_rate": 5e-06,
      "loss": 0.5919,
      "step": 19700
    },
    {
      "epoch": 23.21,
      "learning_rate": 5e-06,
      "loss": 0.554,
      "step": 19800
    },
    {
      "epoch": 23.33,
      "learning_rate": 5e-06,
      "loss": 0.5708,
      "step": 19900
    },
    {
      "epoch": 23.45,
      "learning_rate": 5e-06,
      "loss": 0.5555,
      "step": 20000
    },
    {
      "epoch": 23.56,
      "learning_rate": 5e-06,
      "loss": 0.6004,
      "step": 20100
    },
    {
      "epoch": 23.68,
      "learning_rate": 5e-06,
      "loss": 0.5894,
      "step": 20200
    },
    {
      "epoch": 23.8,
      "learning_rate": 5e-06,
      "loss": 0.5718,
      "step": 20300
    },
    {
      "epoch": 23.92,
      "learning_rate": 5e-06,
      "loss": 0.5744,
      "step": 20400
    },
    {
      "epoch": 24.03,
      "learning_rate": 5e-06,
      "loss": 0.5602,
      "step": 20500
    },
    {
      "epoch": 24.15,
      "learning_rate": 5e-06,
      "loss": 0.5656,
      "step": 20600
    },
    {
      "epoch": 24.27,
      "learning_rate": 5e-06,
      "loss": 0.5657,
      "step": 20700
    },
    {
      "epoch": 24.38,
      "learning_rate": 5e-06,
      "loss": 0.5553,
      "step": 20800
    },
    {
      "epoch": 24.5,
      "learning_rate": 5e-06,
      "loss": 0.5962,
      "step": 20900
    },
    {
      "epoch": 24.62,
      "learning_rate": 5e-06,
      "loss": 0.5982,
      "step": 21000
    },
    {
      "epoch": 24.62,
      "eval_loss": 0.5987153053283691,
      "eval_runtime": 70.2465,
      "eval_samples_per_second": 3.032,
      "eval_steps_per_second": 3.032,
      "step": 21000
    },
    {
      "epoch": 24.74,
      "learning_rate": 5e-06,
      "loss": 0.5946,
      "step": 21100
    },
    {
      "epoch": 24.85,
      "learning_rate": 5e-06,
      "loss": 0.5473,
      "step": 21200
    },
    {
      "epoch": 24.97,
      "learning_rate": 5e-06,
      "loss": 0.5605,
      "step": 21300
    },
    {
      "epoch": 25.09,
      "learning_rate": 5e-06,
      "loss": 0.5953,
      "step": 21400
    },
    {
      "epoch": 25.21,
      "learning_rate": 5e-06,
      "loss": 0.5697,
      "step": 21500
    },
    {
      "epoch": 25.32,
      "learning_rate": 5e-06,
      "loss": 0.5627,
      "step": 21600
    },
    {
      "epoch": 25.44,
      "learning_rate": 5e-06,
      "loss": 0.567,
      "step": 21700
    },
    {
      "epoch": 25.56,
      "learning_rate": 5e-06,
      "loss": 0.5394,
      "step": 21800
    },
    {
      "epoch": 25.67,
      "learning_rate": 5e-06,
      "loss": 0.5461,
      "step": 21900
    },
    {
      "epoch": 25.79,
      "learning_rate": 5e-06,
      "loss": 0.5615,
      "step": 22000
    },
    {
      "epoch": 25.91,
      "learning_rate": 5e-06,
      "loss": 0.5547,
      "step": 22100
    },
    {
      "epoch": 26.03,
      "learning_rate": 5e-06,
      "loss": 0.5534,
      "step": 22200
    },
    {
      "epoch": 26.14,
      "learning_rate": 5e-06,
      "loss": 0.5494,
      "step": 22300
    },
    {
      "epoch": 26.26,
      "learning_rate": 5e-06,
      "loss": 0.569,
      "step": 22400
    },
    {
      "epoch": 26.38,
      "learning_rate": 5e-06,
      "loss": 0.5352,
      "step": 22500
    },
    {
      "epoch": 26.38,
      "eval_loss": 0.6042129397392273,
      "eval_runtime": 69.7858,
      "eval_samples_per_second": 3.052,
      "eval_steps_per_second": 3.052,
      "step": 22500
    },
    {
      "epoch": 26.49,
      "learning_rate": 5e-06,
      "loss": 0.5754,
      "step": 22600
    },
    {
      "epoch": 26.61,
      "learning_rate": 5e-06,
      "loss": 0.5443,
      "step": 22700
    },
    {
      "epoch": 26.73,
      "learning_rate": 5e-06,
      "loss": 0.5765,
      "step": 22800
    },
    {
      "epoch": 26.85,
      "learning_rate": 5e-06,
      "loss": 0.5494,
      "step": 22900
    },
    {
      "epoch": 26.96,
      "learning_rate": 5e-06,
      "loss": 0.5598,
      "step": 23000
    },
    {
      "epoch": 27.08,
      "learning_rate": 5e-06,
      "loss": 0.5634,
      "step": 23100
    },
    {
      "epoch": 27.2,
      "learning_rate": 5e-06,
      "loss": 0.534,
      "step": 23200
    },
    {
      "epoch": 27.32,
      "learning_rate": 5e-06,
      "loss": 0.5626,
      "step": 23300
    },
    {
      "epoch": 27.43,
      "learning_rate": 5e-06,
      "loss": 0.5681,
      "step": 23400
    },
    {
      "epoch": 27.55,
      "learning_rate": 5e-06,
      "loss": 0.5735,
      "step": 23500
    },
    {
      "epoch": 27.67,
      "learning_rate": 5e-06,
      "loss": 0.5464,
      "step": 23600
    },
    {
      "epoch": 27.78,
      "learning_rate": 5e-06,
      "loss": 0.529,
      "step": 23700
    },
    {
      "epoch": 27.9,
      "learning_rate": 5e-06,
      "loss": 0.548,
      "step": 23800
    },
    {
      "epoch": 28.02,
      "learning_rate": 5e-06,
      "loss": 0.5699,
      "step": 23900
    },
    {
      "epoch": 28.14,
      "learning_rate": 5e-06,
      "loss": 0.5746,
      "step": 24000
    },
    {
      "epoch": 28.14,
      "eval_loss": 0.5994372367858887,
      "eval_runtime": 70.1931,
      "eval_samples_per_second": 3.034,
      "eval_steps_per_second": 3.034,
      "step": 24000
    },
    {
      "epoch": 28.25,
      "learning_rate": 5e-06,
      "loss": 0.5537,
      "step": 24100
    },
    {
      "epoch": 28.37,
      "learning_rate": 5e-06,
      "loss": 0.5479,
      "step": 24200
    },
    {
      "epoch": 28.49,
      "learning_rate": 5e-06,
      "loss": 0.5643,
      "step": 24300
    },
    {
      "epoch": 28.6,
      "learning_rate": 5e-06,
      "loss": 0.5273,
      "step": 24400
    },
    {
      "epoch": 28.72,
      "learning_rate": 5e-06,
      "loss": 0.544,
      "step": 24500
    },
    {
      "epoch": 28.84,
      "learning_rate": 5e-06,
      "loss": 0.5172,
      "step": 24600
    },
    {
      "epoch": 28.96,
      "learning_rate": 5e-06,
      "loss": 0.5658,
      "step": 24700
    },
    {
      "epoch": 29.07,
      "learning_rate": 5e-06,
      "loss": 0.5343,
      "step": 24800
    },
    {
      "epoch": 29.19,
      "learning_rate": 5e-06,
      "loss": 0.5307,
      "step": 24900
    },
    {
      "epoch": 29.31,
      "learning_rate": 5e-06,
      "loss": 0.5386,
      "step": 25000
    },
    {
      "epoch": 29.43,
      "learning_rate": 5e-06,
      "loss": 0.5553,
      "step": 25100
    },
    {
      "epoch": 29.54,
      "learning_rate": 5e-06,
      "loss": 0.5309,
      "step": 25200
    },
    {
      "epoch": 29.66,
      "learning_rate": 5e-06,
      "loss": 0.5323,
      "step": 25300
    },
    {
      "epoch": 29.78,
      "learning_rate": 5e-06,
      "loss": 0.5477,
      "step": 25400
    },
    {
      "epoch": 29.89,
      "learning_rate": 5e-06,
      "loss": 0.5618,
      "step": 25500
    },
    {
      "epoch": 29.89,
      "eval_loss": 0.5992656350135803,
      "eval_runtime": 70.8526,
      "eval_samples_per_second": 3.006,
      "eval_steps_per_second": 3.006,
      "step": 25500
    },
    {
      "epoch": 30.01,
      "learning_rate": 5e-06,
      "loss": 0.5368,
      "step": 25600
    },
    {
      "epoch": 30.13,
      "learning_rate": 5e-06,
      "loss": 0.55,
      "step": 25700
    },
    {
      "epoch": 30.25,
      "learning_rate": 5e-06,
      "loss": 0.5138,
      "step": 25800
    },
    {
      "epoch": 30.36,
      "learning_rate": 5e-06,
      "loss": 0.5266,
      "step": 25900
    },
    {
      "epoch": 30.48,
      "learning_rate": 5e-06,
      "loss": 0.5539,
      "step": 26000
    },
    {
      "epoch": 30.6,
      "learning_rate": 5e-06,
      "loss": 0.536,
      "step": 26100
    },
    {
      "epoch": 30.72,
      "learning_rate": 5e-06,
      "loss": 0.5427,
      "step": 26200
    },
    {
      "epoch": 30.83,
      "learning_rate": 5e-06,
      "loss": 0.5496,
      "step": 26300
    },
    {
      "epoch": 30.95,
      "learning_rate": 5e-06,
      "loss": 0.5127,
      "step": 26400
    },
    {
      "epoch": 31.07,
      "learning_rate": 5e-06,
      "loss": 0.5569,
      "step": 26500
    },
    {
      "epoch": 31.18,
      "learning_rate": 5e-06,
      "loss": 0.5196,
      "step": 26600
    },
    {
      "epoch": 31.3,
      "learning_rate": 5e-06,
      "loss": 0.5268,
      "step": 26700
    },
    {
      "epoch": 31.42,
      "learning_rate": 5e-06,
      "loss": 0.5419,
      "step": 26800
    },
    {
      "epoch": 31.54,
      "learning_rate": 5e-06,
      "loss": 0.5087,
      "step": 26900
    },
    {
      "epoch": 31.65,
      "learning_rate": 5e-06,
      "loss": 0.5254,
      "step": 27000
    },
    {
      "epoch": 31.65,
      "eval_loss": 0.5909192562103271,
      "eval_runtime": 70.5089,
      "eval_samples_per_second": 3.021,
      "eval_steps_per_second": 3.021,
      "step": 27000
    },
    {
      "epoch": 31.77,
      "learning_rate": 5e-06,
      "loss": 0.5346,
      "step": 27100
    },
    {
      "epoch": 31.89,
      "learning_rate": 5e-06,
      "loss": 0.5279,
      "step": 27200
    },
    {
      "epoch": 32.0,
      "learning_rate": 5e-06,
      "loss": 0.5711,
      "step": 27300
    },
    {
      "epoch": 32.12,
      "learning_rate": 5e-06,
      "loss": 0.5079,
      "step": 27400
    },
    {
      "epoch": 32.24,
      "learning_rate": 5e-06,
      "loss": 0.5303,
      "step": 27500
    },
    {
      "epoch": 32.36,
      "learning_rate": 5e-06,
      "loss": 0.5347,
      "step": 27600
    },
    {
      "epoch": 32.47,
      "learning_rate": 5e-06,
      "loss": 0.4936,
      "step": 27700
    },
    {
      "epoch": 32.59,
      "learning_rate": 5e-06,
      "loss": 0.5303,
      "step": 27800
    },
    {
      "epoch": 32.71,
      "learning_rate": 5e-06,
      "loss": 0.5543,
      "step": 27900
    },
    {
      "epoch": 32.83,
      "learning_rate": 5e-06,
      "loss": 0.5266,
      "step": 28000
    },
    {
      "epoch": 32.94,
      "learning_rate": 5e-06,
      "loss": 0.5258,
      "step": 28100
    },
    {
      "epoch": 33.06,
      "learning_rate": 5e-06,
      "loss": 0.5559,
      "step": 28200
    },
    {
      "epoch": 33.18,
      "learning_rate": 5e-06,
      "loss": 0.5096,
      "step": 28300
    },
    {
      "epoch": 33.29,
      "learning_rate": 5e-06,
      "loss": 0.5427,
      "step": 28400
    },
    {
      "epoch": 33.41,
      "learning_rate": 5e-06,
      "loss": 0.5336,
      "step": 28500
    },
    {
      "epoch": 33.41,
      "eval_loss": 0.587517261505127,
      "eval_runtime": 70.6475,
      "eval_samples_per_second": 3.015,
      "eval_steps_per_second": 3.015,
      "step": 28500
    },
    {
      "epoch": 33.53,
      "learning_rate": 5e-06,
      "loss": 0.5419,
      "step": 28600
    },
    {
      "epoch": 33.65,
      "learning_rate": 5e-06,
      "loss": 0.5235,
      "step": 28700
    },
    {
      "epoch": 33.76,
      "learning_rate": 5e-06,
      "loss": 0.5266,
      "step": 28800
    },
    {
      "epoch": 33.88,
      "learning_rate": 5e-06,
      "loss": 0.5308,
      "step": 28900
    },
    {
      "epoch": 34.0,
      "learning_rate": 5e-06,
      "loss": 0.5203,
      "step": 29000
    },
    {
      "epoch": 34.11,
      "learning_rate": 5e-06,
      "loss": 0.4988,
      "step": 29100
    },
    {
      "epoch": 34.23,
      "learning_rate": 5e-06,
      "loss": 0.502,
      "step": 29200
    },
    {
      "epoch": 34.35,
      "learning_rate": 5e-06,
      "loss": 0.5111,
      "step": 29300
    },
    {
      "epoch": 34.47,
      "learning_rate": 5e-06,
      "loss": 0.539,
      "step": 29400
    },
    {
      "epoch": 34.58,
      "learning_rate": 5e-06,
      "loss": 0.5086,
      "step": 29500
    },
    {
      "epoch": 34.7,
      "learning_rate": 5e-06,
      "loss": 0.5285,
      "step": 29600
    },
    {
      "epoch": 34.82,
      "learning_rate": 5e-06,
      "loss": 0.5153,
      "step": 29700
    },
    {
      "epoch": 34.94,
      "learning_rate": 5e-06,
      "loss": 0.5366,
      "step": 29800
    },
    {
      "epoch": 35.05,
      "learning_rate": 5e-06,
      "loss": 0.5307,
      "step": 29900
    },
    {
      "epoch": 35.17,
      "learning_rate": 5e-06,
      "loss": 0.5677,
      "step": 30000
    },
    {
      "epoch": 35.17,
      "eval_loss": 0.591385543346405,
      "eval_runtime": 70.1376,
      "eval_samples_per_second": 3.037,
      "eval_steps_per_second": 3.037,
      "step": 30000
    },
    {
      "epoch": 35.29,
      "learning_rate": 5e-06,
      "loss": 0.5283,
      "step": 30100
    },
    {
      "epoch": 35.4,
      "learning_rate": 5e-06,
      "loss": 0.5281,
      "step": 30200
    },
    {
      "epoch": 35.52,
      "learning_rate": 5e-06,
      "loss": 0.5087,
      "step": 30300
    },
    {
      "epoch": 35.64,
      "learning_rate": 5e-06,
      "loss": 0.4965,
      "step": 30400
    },
    {
      "epoch": 35.76,
      "learning_rate": 5e-06,
      "loss": 0.5085,
      "step": 30500
    },
    {
      "epoch": 35.87,
      "learning_rate": 5e-06,
      "loss": 0.5159,
      "step": 30600
    },
    {
      "epoch": 35.99,
      "learning_rate": 5e-06,
      "loss": 0.5149,
      "step": 30700
    },
    {
      "epoch": 36.11,
      "learning_rate": 5e-06,
      "loss": 0.5281,
      "step": 30800
    },
    {
      "epoch": 36.23,
      "learning_rate": 5e-06,
      "loss": 0.511,
      "step": 30900
    },
    {
      "epoch": 36.34,
      "learning_rate": 5e-06,
      "loss": 0.5327,
      "step": 31000
    },
    {
      "epoch": 36.46,
      "learning_rate": 5e-06,
      "loss": 0.5267,
      "step": 31100
    },
    {
      "epoch": 36.58,
      "learning_rate": 5e-06,
      "loss": 0.5124,
      "step": 31200
    },
    {
      "epoch": 36.69,
      "learning_rate": 5e-06,
      "loss": 0.5069,
      "step": 31300
    },
    {
      "epoch": 36.81,
      "learning_rate": 5e-06,
      "loss": 0.4839,
      "step": 31400
    },
    {
      "epoch": 36.93,
      "learning_rate": 5e-06,
      "loss": 0.5009,
      "step": 31500
    },
    {
      "epoch": 36.93,
      "eval_loss": 0.6007654070854187,
      "eval_runtime": 70.1373,
      "eval_samples_per_second": 3.037,
      "eval_steps_per_second": 3.037,
      "step": 31500
    },
    {
      "epoch": 37.05,
      "learning_rate": 5e-06,
      "loss": 0.5169,
      "step": 31600
    },
    {
      "epoch": 37.16,
      "learning_rate": 5e-06,
      "loss": 0.5003,
      "step": 31700
    },
    {
      "epoch": 37.28,
      "learning_rate": 5e-06,
      "loss": 0.502,
      "step": 31800
    },
    {
      "epoch": 37.4,
      "learning_rate": 5e-06,
      "loss": 0.5149,
      "step": 31900
    },
    {
      "epoch": 37.51,
      "learning_rate": 5e-06,
      "loss": 0.4903,
      "step": 32000
    }
  ],
  "max_steps": 255900,
  "num_train_epochs": 300,
  "total_flos": 3.193778270424269e+19,
  "trial_name": null,
  "trial_params": null
}