{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 18550,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 0.0003,
      "loss": 1.0781,
      "step": 50
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0003,
      "loss": 0.2803,
      "step": 100
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0003,
      "loss": 0.2085,
      "step": 150
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0003,
      "loss": 0.1611,
      "step": 200
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0003,
      "loss": 0.1625,
      "step": 250
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0003,
      "loss": 0.1371,
      "step": 300
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0003,
      "loss": 0.1533,
      "step": 350
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0003,
      "loss": 0.1224,
      "step": 400
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0003,
      "loss": 0.1488,
      "step": 450
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0003,
      "loss": 0.0981,
      "step": 500
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0003,
      "loss": 0.0973,
      "step": 550
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0003,
      "loss": 0.0992,
      "step": 600
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0003,
      "loss": 0.0867,
      "step": 650
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0003,
      "loss": 0.0933,
      "step": 700
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0003,
      "loss": 0.1006,
      "step": 750
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0003,
      "loss": 0.0694,
      "step": 800
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0003,
      "loss": 0.0655,
      "step": 850
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0003,
      "loss": 0.0775,
      "step": 900
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0003,
      "loss": 0.0666,
      "step": 950
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0003,
      "loss": 0.0539,
      "step": 1000
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0003,
      "loss": 0.0726,
      "step": 1050
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0003,
      "loss": 0.0699,
      "step": 1100
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0003,
      "loss": 0.0786,
      "step": 1150
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0003,
      "loss": 0.0773,
      "step": 1200
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0003,
      "loss": 0.073,
      "step": 1250
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0003,
      "loss": 0.0543,
      "step": 1300
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0003,
      "loss": 0.0572,
      "step": 1350
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0003,
      "loss": 0.089,
      "step": 1400
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0003,
      "loss": 0.0516,
      "step": 1450
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0003,
      "loss": 0.0643,
      "step": 1500
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0003,
      "loss": 0.0364,
      "step": 1550
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0003,
      "loss": 0.0698,
      "step": 1600
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0003,
      "loss": 0.0431,
      "step": 1650
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0003,
      "loss": 0.0393,
      "step": 1700
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0003,
      "loss": 0.0682,
      "step": 1750
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0003,
      "loss": 0.0454,
      "step": 1800
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0003,
      "loss": 0.0534,
      "step": 1850
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0003,
      "loss": 0.0519,
      "step": 1900
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0003,
      "loss": 0.0675,
      "step": 1950
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0003,
      "loss": 0.046,
      "step": 2000
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0003,
      "loss": 0.0527,
      "step": 2050
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0003,
      "loss": 0.0588,
      "step": 2100
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0003,
      "loss": 0.0625,
      "step": 2150
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0003,
      "loss": 0.039,
      "step": 2200
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0003,
      "loss": 0.0394,
      "step": 2250
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0003,
      "loss": 0.041,
      "step": 2300
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0003,
      "loss": 0.0363,
      "step": 2350
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.0003,
      "loss": 0.0524,
      "step": 2400
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0003,
      "loss": 0.0469,
      "step": 2450
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0003,
      "loss": 0.0427,
      "step": 2500
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.0003,
      "loss": 0.0482,
      "step": 2550
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0003,
      "loss": 0.0319,
      "step": 2600
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.0003,
      "loss": 0.0371,
      "step": 2650
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0003,
      "loss": 0.0446,
      "step": 2700
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0003,
      "loss": 0.0436,
      "step": 2750
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.0003,
      "loss": 0.0308,
      "step": 2800
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0003,
      "loss": 0.0472,
      "step": 2850
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0003,
      "loss": 0.0219,
      "step": 2900
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0003,
      "loss": 0.0346,
      "step": 2950
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.0003,
      "loss": 0.0351,
      "step": 3000
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.0003,
      "loss": 0.0326,
      "step": 3050
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.0003,
      "loss": 0.0405,
      "step": 3100
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.0003,
      "loss": 0.0346,
      "step": 3150
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.0003,
      "loss": 0.0329,
      "step": 3200
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.0003,
      "loss": 0.0407,
      "step": 3250
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.0003,
      "loss": 0.0446,
      "step": 3300
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.0003,
      "loss": 0.0447,
      "step": 3350
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.0003,
      "loss": 0.0393,
      "step": 3400
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.0003,
      "loss": 0.0306,
      "step": 3450
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.0003,
      "loss": 0.0389,
      "step": 3500
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.0003,
      "loss": 0.0326,
      "step": 3550
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.0003,
      "loss": 0.0343,
      "step": 3600
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.0003,
      "loss": 0.0264,
      "step": 3650
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.0003,
      "loss": 0.0313,
      "step": 3700
    },
    {
      "epoch": 1.0,
      "eval_f1": 22.2222,
      "eval_gen_len": 29.002111729698598,
      "eval_loss": 0.045024264603853226,
      "eval_runtime": 25.7329,
      "eval_samples_per_second": 809.701,
      "eval_steps_per_second": 25.337,
      "step": 3710
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.0003,
      "loss": 0.033,
      "step": 3750
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.0003,
      "loss": 0.0269,
      "step": 3800
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.0003,
      "loss": 0.0199,
      "step": 3850
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.0003,
      "loss": 0.0221,
      "step": 3900
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.0003,
      "loss": 0.0138,
      "step": 3950
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.0003,
      "loss": 0.023,
      "step": 4000
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0003,
      "loss": 0.0171,
      "step": 4050
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.0003,
      "loss": 0.0178,
      "step": 4100
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.0003,
      "loss": 0.0193,
      "step": 4150
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.0003,
      "loss": 0.0129,
      "step": 4200
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.0003,
      "loss": 0.0172,
      "step": 4250
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.0003,
      "loss": 0.0265,
      "step": 4300
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0003,
      "loss": 0.0208,
      "step": 4350
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.0003,
      "loss": 0.0179,
      "step": 4400
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0003,
      "loss": 0.0142,
      "step": 4450
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.0003,
      "loss": 0.0232,
      "step": 4500
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.0003,
      "loss": 0.0257,
      "step": 4550
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0003,
      "loss": 0.0159,
      "step": 4600
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.0003,
      "loss": 0.017,
      "step": 4650
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.0003,
      "loss": 0.0221,
      "step": 4700
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.0003,
      "loss": 0.0244,
      "step": 4750
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.0003,
      "loss": 0.0222,
      "step": 4800
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0003,
      "loss": 0.0129,
      "step": 4850
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.0003,
      "loss": 0.0137,
      "step": 4900
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.0003,
      "loss": 0.01,
      "step": 4950
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.0003,
      "loss": 0.0293,
      "step": 5000
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.0003,
      "loss": 0.0155,
      "step": 5050
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.0003,
      "loss": 0.0103,
      "step": 5100
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.0003,
      "loss": 0.0104,
      "step": 5150
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.0003,
      "loss": 0.0316,
      "step": 5200
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.0003,
      "loss": 0.0173,
      "step": 5250
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.0003,
      "loss": 0.0224,
      "step": 5300
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.0003,
      "loss": 0.0181,
      "step": 5350
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0003,
      "loss": 0.0144,
      "step": 5400
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.0003,
      "loss": 0.0328,
      "step": 5450
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.0003,
      "loss": 0.0182,
      "step": 5500
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.0003,
      "loss": 0.0162,
      "step": 5550
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.0003,
      "loss": 0.0274,
      "step": 5600
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.0003,
      "loss": 0.0126,
      "step": 5650
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.0003,
      "loss": 0.0239,
      "step": 5700
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.0003,
      "loss": 0.0138,
      "step": 5750
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.0003,
      "loss": 0.0085,
      "step": 5800
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.0003,
      "loss": 0.0225,
      "step": 5850
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0003,
      "loss": 0.0138,
      "step": 5900
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0003,
      "loss": 0.0245,
      "step": 5950
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0003,
      "loss": 0.0146,
      "step": 6000
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0003,
      "loss": 0.0176,
      "step": 6050
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0003,
      "loss": 0.008,
      "step": 6100
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0003,
      "loss": 0.0169,
      "step": 6150
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0003,
      "loss": 0.0254,
      "step": 6200
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0003,
      "loss": 0.0057,
      "step": 6250
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0003,
      "loss": 0.0159,
      "step": 6300
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.0003,
      "loss": 0.0265,
      "step": 6350
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0003,
      "loss": 0.0075,
      "step": 6400
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0003,
      "loss": 0.0132,
      "step": 6450
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0003,
      "loss": 0.0095,
      "step": 6500
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0003,
      "loss": 0.0217,
      "step": 6550
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.0003,
      "loss": 0.0081,
      "step": 6600
    },
    {
      "epoch": 1.79,
      "learning_rate": 0.0003,
      "loss": 0.0095,
      "step": 6650
    },
    {
      "epoch": 1.81,
      "learning_rate": 0.0003,
      "loss": 0.0176,
      "step": 6700
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.0003,
      "loss": 0.0122,
      "step": 6750
    },
    {
      "epoch": 1.83,
      "learning_rate": 0.0003,
      "loss": 0.0111,
      "step": 6800
    },
    {
      "epoch": 1.85,
      "learning_rate": 0.0003,
      "loss": 0.026,
      "step": 6850
    },
    {
      "epoch": 1.86,
      "learning_rate": 0.0003,
      "loss": 0.0101,
      "step": 6900
    },
    {
      "epoch": 1.87,
      "learning_rate": 0.0003,
      "loss": 0.0155,
      "step": 6950
    },
    {
      "epoch": 1.89,
      "learning_rate": 0.0003,
      "loss": 0.0146,
      "step": 7000
    },
    {
      "epoch": 1.9,
      "learning_rate": 0.0003,
      "loss": 0.018,
      "step": 7050
    },
    {
      "epoch": 1.91,
      "learning_rate": 0.0003,
      "loss": 0.0159,
      "step": 7100
    },
    {
      "epoch": 1.93,
      "learning_rate": 0.0003,
      "loss": 0.0178,
      "step": 7150
    },
    {
      "epoch": 1.94,
      "learning_rate": 0.0003,
      "loss": 0.0151,
      "step": 7200
    },
    {
      "epoch": 1.95,
      "learning_rate": 0.0003,
      "loss": 0.0185,
      "step": 7250
    },
    {
      "epoch": 1.97,
      "learning_rate": 0.0003,
      "loss": 0.0055,
      "step": 7300
    },
    {
      "epoch": 1.98,
      "learning_rate": 0.0003,
      "loss": 0.0046,
      "step": 7350
    },
    {
      "epoch": 1.99,
      "learning_rate": 0.0003,
      "loss": 0.0093,
      "step": 7400
    },
    {
      "epoch": 2.0,
      "eval_f1": 22.2222,
      "eval_gen_len": 29.002111729698598,
      "eval_loss": 0.06078604236245155,
      "eval_runtime": 25.7417,
      "eval_samples_per_second": 809.427,
      "eval_steps_per_second": 25.329,
      "step": 7420
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0003,
      "loss": 0.0159,
      "step": 7450
    },
    {
      "epoch": 2.02,
      "learning_rate": 0.0003,
      "loss": 0.0088,
      "step": 7500
    },
    {
      "epoch": 2.04,
      "learning_rate": 0.0003,
      "loss": 0.0055,
      "step": 7550
    },
    {
      "epoch": 2.05,
      "learning_rate": 0.0003,
      "loss": 0.0074,
      "step": 7600
    },
    {
      "epoch": 2.06,
      "learning_rate": 0.0003,
      "loss": 0.0086,
      "step": 7650
    },
    {
      "epoch": 2.08,
      "learning_rate": 0.0003,
      "loss": 0.0168,
      "step": 7700
    },
    {
      "epoch": 2.09,
      "learning_rate": 0.0003,
      "loss": 0.0072,
      "step": 7750
    },
    {
      "epoch": 2.1,
      "learning_rate": 0.0003,
      "loss": 0.0038,
      "step": 7800
    },
    {
      "epoch": 2.12,
      "learning_rate": 0.0003,
      "loss": 0.0119,
      "step": 7850
    },
    {
      "epoch": 2.13,
      "learning_rate": 0.0003,
      "loss": 0.012,
      "step": 7900
    },
    {
      "epoch": 2.14,
      "learning_rate": 0.0003,
      "loss": 0.0053,
      "step": 7950
    },
    {
      "epoch": 2.16,
      "learning_rate": 0.0003,
      "loss": 0.0093,
      "step": 8000
    },
    {
      "epoch": 2.17,
      "learning_rate": 0.0003,
      "loss": 0.0138,
      "step": 8050
    },
    {
      "epoch": 2.18,
      "learning_rate": 0.0003,
      "loss": 0.0179,
      "step": 8100
    },
    {
      "epoch": 2.2,
      "learning_rate": 0.0003,
      "loss": 0.0055,
      "step": 8150
    },
    {
      "epoch": 2.21,
      "learning_rate": 0.0003,
      "loss": 0.0039,
      "step": 8200
    },
    {
      "epoch": 2.22,
      "learning_rate": 0.0003,
      "loss": 0.0066,
      "step": 8250
    },
    {
      "epoch": 2.24,
      "learning_rate": 0.0003,
      "loss": 0.0079,
      "step": 8300
    },
    {
      "epoch": 2.25,
      "learning_rate": 0.0003,
      "loss": 0.0042,
      "step": 8350
    },
    {
      "epoch": 2.26,
      "learning_rate": 0.0003,
      "loss": 0.004,
      "step": 8400
    },
    {
      "epoch": 2.28,
      "learning_rate": 0.0003,
      "loss": 0.0108,
      "step": 8450
    },
    {
      "epoch": 2.29,
      "learning_rate": 0.0003,
      "loss": 0.0077,
      "step": 8500
    },
    {
      "epoch": 2.3,
      "learning_rate": 0.0003,
      "loss": 0.0115,
      "step": 8550
    },
    {
      "epoch": 2.32,
      "learning_rate": 0.0003,
      "loss": 0.0067,
      "step": 8600
    },
    {
      "epoch": 2.33,
      "learning_rate": 0.0003,
      "loss": 0.0172,
      "step": 8650
    },
    {
      "epoch": 2.35,
      "learning_rate": 0.0003,
      "loss": 0.0039,
      "step": 8700
    },
    {
      "epoch": 2.36,
      "learning_rate": 0.0003,
      "loss": 0.0067,
      "step": 8750
    },
    {
      "epoch": 2.37,
      "learning_rate": 0.0003,
      "loss": 0.0051,
      "step": 8800
    },
    {
      "epoch": 2.39,
      "learning_rate": 0.0003,
      "loss": 0.0163,
      "step": 8850
    },
    {
      "epoch": 2.4,
      "learning_rate": 0.0003,
      "loss": 0.0077,
      "step": 8900
    },
    {
      "epoch": 2.41,
      "learning_rate": 0.0003,
      "loss": 0.0096,
      "step": 8950
    },
    {
      "epoch": 2.43,
      "learning_rate": 0.0003,
      "loss": 0.0044,
      "step": 9000
    },
    {
      "epoch": 2.44,
      "learning_rate": 0.0003,
      "loss": 0.0028,
      "step": 9050
    },
    {
      "epoch": 2.45,
      "learning_rate": 0.0003,
      "loss": 0.0094,
      "step": 9100
    },
    {
      "epoch": 2.47,
      "learning_rate": 0.0003,
      "loss": 0.0034,
      "step": 9150
    },
    {
      "epoch": 2.48,
      "learning_rate": 0.0003,
      "loss": 0.0097,
      "step": 9200
    },
    {
      "epoch": 2.49,
      "learning_rate": 0.0003,
      "loss": 0.0018,
      "step": 9250
    },
    {
      "epoch": 2.51,
      "learning_rate": 0.0003,
      "loss": 0.0036,
      "step": 9300
    },
    {
      "epoch": 2.52,
      "learning_rate": 0.0003,
      "loss": 0.0083,
      "step": 9350
    },
    {
      "epoch": 2.53,
      "learning_rate": 0.0003,
      "loss": 0.0044,
      "step": 9400
    },
    {
      "epoch": 2.55,
      "learning_rate": 0.0003,
      "loss": 0.0088,
      "step": 9450
    },
    {
      "epoch": 2.56,
      "learning_rate": 0.0003,
      "loss": 0.0071,
      "step": 9500
    },
    {
      "epoch": 2.57,
      "learning_rate": 0.0003,
      "loss": 0.0124,
      "step": 9550
    },
    {
      "epoch": 2.59,
      "learning_rate": 0.0003,
      "loss": 0.0044,
      "step": 9600
    },
    {
      "epoch": 2.6,
      "learning_rate": 0.0003,
      "loss": 0.0084,
      "step": 9650
    },
    {
      "epoch": 2.61,
      "learning_rate": 0.0003,
      "loss": 0.0088,
      "step": 9700
    },
    {
      "epoch": 2.63,
      "learning_rate": 0.0003,
      "loss": 0.011,
      "step": 9750
    },
    {
      "epoch": 2.64,
      "learning_rate": 0.0003,
      "loss": 0.0068,
      "step": 9800
    },
    {
      "epoch": 2.65,
      "learning_rate": 0.0003,
      "loss": 0.0084,
      "step": 9850
    },
    {
      "epoch": 2.67,
      "learning_rate": 0.0003,
      "loss": 0.0079,
      "step": 9900
    },
    {
      "epoch": 2.68,
      "learning_rate": 0.0003,
      "loss": 0.0104,
      "step": 9950
    },
    {
      "epoch": 2.7,
      "learning_rate": 0.0003,
      "loss": 0.0047,
      "step": 10000
    },
    {
      "epoch": 2.71,
      "learning_rate": 0.0003,
      "loss": 0.0132,
      "step": 10050
    },
    {
      "epoch": 2.72,
      "learning_rate": 0.0003,
      "loss": 0.0066,
      "step": 10100
    },
    {
      "epoch": 2.74,
      "learning_rate": 0.0003,
      "loss": 0.0108,
      "step": 10150
    },
    {
      "epoch": 2.75,
      "learning_rate": 0.0003,
      "loss": 0.0085,
      "step": 10200
    },
    {
      "epoch": 2.76,
      "learning_rate": 0.0003,
      "loss": 0.0164,
      "step": 10250
    },
    {
      "epoch": 2.78,
      "learning_rate": 0.0003,
      "loss": 0.0135,
      "step": 10300
    },
    {
      "epoch": 2.79,
      "learning_rate": 0.0003,
      "loss": 0.0106,
      "step": 10350
    },
    {
      "epoch": 2.8,
      "learning_rate": 0.0003,
      "loss": 0.0063,
      "step": 10400
    },
    {
      "epoch": 2.82,
      "learning_rate": 0.0003,
      "loss": 0.0035,
      "step": 10450
    },
    {
      "epoch": 2.83,
      "learning_rate": 0.0003,
      "loss": 0.0078,
      "step": 10500
    },
    {
      "epoch": 2.84,
      "learning_rate": 0.0003,
      "loss": 0.0188,
      "step": 10550
    },
    {
      "epoch": 2.86,
      "learning_rate": 0.0003,
      "loss": 0.0092,
      "step": 10600
    },
    {
      "epoch": 2.87,
      "learning_rate": 0.0003,
      "loss": 0.0079,
      "step": 10650
    },
    {
      "epoch": 2.88,
      "learning_rate": 0.0003,
      "loss": 0.0059,
      "step": 10700
    },
    {
      "epoch": 2.9,
      "learning_rate": 0.0003,
      "loss": 0.006,
      "step": 10750
    },
    {
      "epoch": 2.91,
      "learning_rate": 0.0003,
      "loss": 0.0061,
      "step": 10800
    },
    {
      "epoch": 2.92,
      "learning_rate": 0.0003,
      "loss": 0.0104,
      "step": 10850
    },
    {
      "epoch": 2.94,
      "learning_rate": 0.0003,
      "loss": 0.0084,
      "step": 10900
    },
    {
      "epoch": 2.95,
      "learning_rate": 0.0003,
      "loss": 0.0093,
      "step": 10950
    },
    {
      "epoch": 2.96,
      "learning_rate": 0.0003,
      "loss": 0.0053,
      "step": 11000
    },
    {
      "epoch": 2.98,
      "learning_rate": 0.0003,
      "loss": 0.0109,
      "step": 11050
    },
    {
      "epoch": 2.99,
      "learning_rate": 0.0003,
      "loss": 0.0084,
      "step": 11100
    },
    {
      "epoch": 3.0,
      "eval_f1": 22.2222,
      "eval_gen_len": 29.002111729698598,
      "eval_loss": 0.07568035274744034,
      "eval_runtime": 25.7985,
      "eval_samples_per_second": 807.643,
      "eval_steps_per_second": 25.273,
      "step": 11130
    },
    {
      "epoch": 3.01,
      "learning_rate": 0.0003,
      "loss": 0.0069,
      "step": 11150
    },
    {
      "epoch": 3.02,
      "learning_rate": 0.0003,
      "loss": 0.0035,
      "step": 11200
    },
    {
      "epoch": 3.03,
      "learning_rate": 0.0003,
      "loss": 0.0057,
      "step": 11250
    },
    {
      "epoch": 3.05,
      "learning_rate": 0.0003,
      "loss": 0.0041,
      "step": 11300
    },
    {
      "epoch": 3.06,
      "learning_rate": 0.0003,
      "loss": 0.0047,
      "step": 11350
    },
    {
      "epoch": 3.07,
      "learning_rate": 0.0003,
      "loss": 0.0025,
      "step": 11400
    },
    {
      "epoch": 3.09,
      "learning_rate": 0.0003,
      "loss": 0.006,
      "step": 11450
    },
    {
      "epoch": 3.1,
      "learning_rate": 0.0003,
      "loss": 0.0033,
      "step": 11500
    },
    {
      "epoch": 3.11,
      "learning_rate": 0.0003,
      "loss": 0.0046,
      "step": 11550
    },
    {
      "epoch": 3.13,
      "learning_rate": 0.0003,
      "loss": 0.0065,
      "step": 11600
    },
    {
      "epoch": 3.14,
      "learning_rate": 0.0003,
      "loss": 0.0058,
      "step": 11650
    },
    {
      "epoch": 3.15,
      "learning_rate": 0.0003,
      "loss": 0.0056,
      "step": 11700
    },
    {
      "epoch": 3.17,
      "learning_rate": 0.0003,
      "loss": 0.0094,
      "step": 11750
    },
    {
      "epoch": 3.18,
      "learning_rate": 0.0003,
      "loss": 0.0035,
      "step": 11800
    },
    {
      "epoch": 3.19,
      "learning_rate": 0.0003,
      "loss": 0.0016,
      "step": 11850
    },
    {
      "epoch": 3.21,
      "learning_rate": 0.0003,
      "loss": 0.0047,
      "step": 11900
    },
    {
      "epoch": 3.22,
      "learning_rate": 0.0003,
      "loss": 0.0054,
      "step": 11950
    },
    {
      "epoch": 3.23,
      "learning_rate": 0.0003,
      "loss": 0.0022,
      "step": 12000
    },
    {
      "epoch": 3.25,
      "learning_rate": 0.0003,
      "loss": 0.0052,
      "step": 12050
    },
    {
      "epoch": 3.26,
      "learning_rate": 0.0003,
      "loss": 0.0046,
      "step": 12100
    },
    {
      "epoch": 3.27,
      "learning_rate": 0.0003,
      "loss": 0.0091,
      "step": 12150
    },
    {
      "epoch": 3.29,
      "learning_rate": 0.0003,
      "loss": 0.0053,
      "step": 12200
    },
    {
      "epoch": 3.3,
      "learning_rate": 0.0003,
      "loss": 0.0081,
      "step": 12250
    },
    {
      "epoch": 3.32,
      "learning_rate": 0.0003,
      "loss": 0.0088,
      "step": 12300
    },
    {
      "epoch": 3.33,
      "learning_rate": 0.0003,
      "loss": 0.0086,
      "step": 12350
    },
    {
      "epoch": 3.34,
      "learning_rate": 0.0003,
      "loss": 0.0083,
      "step": 12400
    },
    {
      "epoch": 3.36,
      "learning_rate": 0.0003,
      "loss": 0.0032,
      "step": 12450
    },
    {
      "epoch": 3.37,
      "learning_rate": 0.0003,
      "loss": 0.0084,
      "step": 12500
    },
    {
      "epoch": 3.38,
      "learning_rate": 0.0003,
      "loss": 0.0067,
      "step": 12550
    },
    {
      "epoch": 3.4,
      "learning_rate": 0.0003,
      "loss": 0.0071,
      "step": 12600
    },
    {
      "epoch": 3.41,
      "learning_rate": 0.0003,
      "loss": 0.0062,
      "step": 12650
    },
    {
      "epoch": 3.42,
      "learning_rate": 0.0003,
      "loss": 0.0089,
      "step": 12700
    },
    {
      "epoch": 3.44,
      "learning_rate": 0.0003,
      "loss": 0.0043,
      "step": 12750
    },
    {
      "epoch": 3.45,
      "learning_rate": 0.0003,
      "loss": 0.0075,
      "step": 12800
    },
    {
      "epoch": 3.46,
      "learning_rate": 0.0003,
      "loss": 0.0055,
      "step": 12850
    },
    {
      "epoch": 3.48,
      "learning_rate": 0.0003,
      "loss": 0.0044,
      "step": 12900
    },
    {
      "epoch": 3.49,
      "learning_rate": 0.0003,
      "loss": 0.0063,
      "step": 12950
    },
    {
      "epoch": 3.5,
      "learning_rate": 0.0003,
      "loss": 0.0105,
      "step": 13000
    },
    {
      "epoch": 3.52,
      "learning_rate": 0.0003,
      "loss": 0.0088,
      "step": 13050
    },
    {
      "epoch": 3.53,
      "learning_rate": 0.0003,
      "loss": 0.0108,
      "step": 13100
    },
    {
      "epoch": 3.54,
      "learning_rate": 0.0003,
      "loss": 0.003,
      "step": 13150
    },
    {
      "epoch": 3.56,
      "learning_rate": 0.0003,
      "loss": 0.0064,
      "step": 13200
    },
    {
      "epoch": 3.57,
      "learning_rate": 0.0003,
      "loss": 0.0083,
      "step": 13250
    },
    {
      "epoch": 3.58,
      "learning_rate": 0.0003,
      "loss": 0.0024,
      "step": 13300
    },
    {
      "epoch": 3.6,
      "learning_rate": 0.0003,
      "loss": 0.0038,
      "step": 13350
    },
    {
      "epoch": 3.61,
      "learning_rate": 0.0003,
      "loss": 0.0122,
      "step": 13400
    },
    {
      "epoch": 3.63,
      "learning_rate": 0.0003,
      "loss": 0.0126,
      "step": 13450
    },
    {
      "epoch": 3.64,
      "learning_rate": 0.0003,
      "loss": 0.0035,
      "step": 13500
    },
    {
      "epoch": 3.65,
      "learning_rate": 0.0003,
      "loss": 0.0056,
      "step": 13550
    },
    {
      "epoch": 3.67,
      "learning_rate": 0.0003,
      "loss": 0.0034,
      "step": 13600
    },
    {
      "epoch": 3.68,
      "learning_rate": 0.0003,
      "loss": 0.003,
      "step": 13650
    },
    {
      "epoch": 3.69,
      "learning_rate": 0.0003,
      "loss": 0.0034,
      "step": 13700
    },
    {
      "epoch": 3.71,
      "learning_rate": 0.0003,
      "loss": 0.0089,
      "step": 13750
    },
    {
      "epoch": 3.72,
      "learning_rate": 0.0003,
      "loss": 0.0065,
      "step": 13800
    },
    {
      "epoch": 3.73,
      "learning_rate": 0.0003,
      "loss": 0.0038,
      "step": 13850
    },
    {
      "epoch": 3.75,
      "learning_rate": 0.0003,
      "loss": 0.0046,
      "step": 13900
    },
    {
      "epoch": 3.76,
      "learning_rate": 0.0003,
      "loss": 0.0041,
      "step": 13950
    },
    {
      "epoch": 3.77,
      "learning_rate": 0.0003,
      "loss": 0.0063,
      "step": 14000
    },
    {
      "epoch": 3.79,
      "learning_rate": 0.0003,
      "loss": 0.0072,
      "step": 14050
    },
    {
      "epoch": 3.8,
      "learning_rate": 0.0003,
      "loss": 0.0025,
      "step": 14100
    },
    {
      "epoch": 3.81,
      "learning_rate": 0.0003,
      "loss": 0.0037,
      "step": 14150
    },
    {
      "epoch": 3.83,
      "learning_rate": 0.0003,
      "loss": 0.0067,
      "step": 14200
    },
    {
      "epoch": 3.84,
      "learning_rate": 0.0003,
      "loss": 0.0016,
      "step": 14250
    },
    {
      "epoch": 3.85,
      "learning_rate": 0.0003,
      "loss": 0.0053,
      "step": 14300
    },
    {
      "epoch": 3.87,
      "learning_rate": 0.0003,
      "loss": 0.0084,
      "step": 14350
    },
    {
      "epoch": 3.88,
      "learning_rate": 0.0003,
      "loss": 0.0054,
      "step": 14400
    },
    {
      "epoch": 3.89,
      "learning_rate": 0.0003,
      "loss": 0.0045,
      "step": 14450
    },
    {
      "epoch": 3.91,
      "learning_rate": 0.0003,
      "loss": 0.0028,
      "step": 14500
    },
    {
      "epoch": 3.92,
      "learning_rate": 0.0003,
      "loss": 0.0081,
      "step": 14550
    },
    {
      "epoch": 3.94,
      "learning_rate": 0.0003,
      "loss": 0.0054,
      "step": 14600
    },
    {
      "epoch": 3.95,
      "learning_rate": 0.0003,
      "loss": 0.0038,
      "step": 14650
    },
    {
      "epoch": 3.96,
      "learning_rate": 0.0003,
      "loss": 0.0051,
      "step": 14700
    },
    {
      "epoch": 3.98,
      "learning_rate": 0.0003,
      "loss": 0.007,
      "step": 14750
    },
    {
      "epoch": 3.99,
      "learning_rate": 0.0003,
      "loss": 0.0046,
      "step": 14800
    },
    {
      "epoch": 4.0,
      "eval_f1": 22.2222,
      "eval_gen_len": 29.002111729698598,
      "eval_loss": 0.07932917773723602,
      "eval_runtime": 25.7288,
      "eval_samples_per_second": 809.832,
      "eval_steps_per_second": 25.341,
      "step": 14840
    },
    {
      "epoch": 4.0,
      "learning_rate": 0.0003,
      "loss": 0.007,
      "step": 14850
    },
    {
      "epoch": 4.02,
      "learning_rate": 0.0003,
      "loss": 0.0036,
      "step": 14900
    },
    {
      "epoch": 4.03,
      "learning_rate": 0.0003,
      "loss": 0.0087,
      "step": 14950
    },
    {
      "epoch": 4.04,
      "learning_rate": 0.0003,
      "loss": 0.0094,
      "step": 15000
    },
    {
      "epoch": 4.06,
      "learning_rate": 0.0003,
      "loss": 0.0023,
      "step": 15050
    },
    {
      "epoch": 4.07,
      "learning_rate": 0.0003,
      "loss": 0.0068,
      "step": 15100
    },
    {
      "epoch": 4.08,
      "learning_rate": 0.0003,
      "loss": 0.0022,
      "step": 15150
    },
    {
      "epoch": 4.1,
      "learning_rate": 0.0003,
      "loss": 0.01,
      "step": 15200
    },
    {
      "epoch": 4.11,
      "learning_rate": 0.0003,
      "loss": 0.0047,
      "step": 15250
    },
    {
      "epoch": 4.12,
      "learning_rate": 0.0003,
      "loss": 0.0086,
      "step": 15300
    },
    {
      "epoch": 4.14,
      "learning_rate": 0.0003,
      "loss": 0.0048,
      "step": 15350
    },
    {
      "epoch": 4.15,
      "learning_rate": 0.0003,
      "loss": 0.0025,
      "step": 15400
    },
    {
      "epoch": 4.16,
      "learning_rate": 0.0003,
      "loss": 0.0031,
      "step": 15450
    },
    {
      "epoch": 4.18,
      "learning_rate": 0.0003,
      "loss": 0.0132,
      "step": 15500
    },
    {
      "epoch": 4.19,
      "learning_rate": 0.0003,
      "loss": 0.0035,
      "step": 15550
    },
    {
      "epoch": 4.2,
      "learning_rate": 0.0003,
      "loss": 0.0022,
      "step": 15600
    },
    {
      "epoch": 4.22,
      "learning_rate": 0.0003,
      "loss": 0.0055,
      "step": 15650
    },
    {
      "epoch": 4.23,
      "learning_rate": 0.0003,
      "loss": 0.0028,
      "step": 15700
    },
    {
      "epoch": 4.25,
      "learning_rate": 0.0003,
      "loss": 0.0062,
      "step": 15750
    },
    {
      "epoch": 4.26,
      "learning_rate": 0.0003,
      "loss": 0.001,
      "step": 15800
    },
    {
      "epoch": 4.27,
      "learning_rate": 0.0003,
      "loss": 0.006,
      "step": 15850
    },
    {
      "epoch": 4.29,
      "learning_rate": 0.0003,
      "loss": 0.0026,
      "step": 15900
    },
    {
      "epoch": 4.3,
      "learning_rate": 0.0003,
      "loss": 0.0042,
      "step": 15950
    },
    {
      "epoch": 4.31,
      "learning_rate": 0.0003,
      "loss": 0.0012,
      "step": 16000
    },
    {
      "epoch": 4.33,
      "learning_rate": 0.0003,
      "loss": 0.0037,
      "step": 16050
    },
    {
      "epoch": 4.34,
      "learning_rate": 0.0003,
      "loss": 0.0037,
      "step": 16100
    },
    {
      "epoch": 4.35,
      "learning_rate": 0.0003,
      "loss": 0.003,
      "step": 16150
    },
    {
      "epoch": 4.37,
      "learning_rate": 0.0003,
      "loss": 0.0055,
      "step": 16200
    },
    {
      "epoch": 4.38,
      "learning_rate": 0.0003,
      "loss": 0.0028,
      "step": 16250
    },
    {
      "epoch": 4.39,
      "learning_rate": 0.0003,
      "loss": 0.0015,
      "step": 16300
    },
    {
      "epoch": 4.41,
      "learning_rate": 0.0003,
      "loss": 0.0034,
      "step": 16350
    },
    {
      "epoch": 4.42,
      "learning_rate": 0.0003,
      "loss": 0.0082,
      "step": 16400
    },
    {
      "epoch": 4.43,
      "learning_rate": 0.0003,
      "loss": 0.0039,
      "step": 16450
    },
    {
      "epoch": 4.45,
      "learning_rate": 0.0003,
      "loss": 0.0028,
      "step": 16500
    },
    {
      "epoch": 4.46,
      "learning_rate": 0.0003,
      "loss": 0.0032,
      "step": 16550
    },
    {
      "epoch": 4.47,
      "learning_rate": 0.0003,
      "loss": 0.0017,
      "step": 16600
    },
    {
      "epoch": 4.49,
      "learning_rate": 0.0003,
      "loss": 0.0028,
      "step": 16650
    },
    {
      "epoch": 4.5,
      "learning_rate": 0.0003,
      "loss": 0.0084,
      "step": 16700
    },
    {
      "epoch": 4.51,
      "learning_rate": 0.0003,
      "loss": 0.0042,
      "step": 16750
    },
    {
      "epoch": 4.53,
      "learning_rate": 0.0003,
      "loss": 0.014,
      "step": 16800
    },
    {
      "epoch": 4.54,
      "learning_rate": 0.0003,
      "loss": 0.0046,
      "step": 16850
    },
    {
      "epoch": 4.56,
      "learning_rate": 0.0003,
      "loss": 0.0043,
      "step": 16900
    },
    {
      "epoch": 4.57,
      "learning_rate": 0.0003,
      "loss": 0.0029,
      "step": 16950
    },
    {
      "epoch": 4.58,
      "learning_rate": 0.0003,
      "loss": 0.008,
      "step": 17000
    },
    {
      "epoch": 4.6,
      "learning_rate": 0.0003,
      "loss": 0.0014,
      "step": 17050
    },
    {
      "epoch": 4.61,
      "learning_rate": 0.0003,
      "loss": 0.004,
      "step": 17100
    },
    {
      "epoch": 4.62,
      "learning_rate": 0.0003,
      "loss": 0.0012,
      "step": 17150
    },
    {
      "epoch": 4.64,
      "learning_rate": 0.0003,
      "loss": 0.0023,
      "step": 17200
    },
    {
      "epoch": 4.65,
      "learning_rate": 0.0003,
      "loss": 0.0036,
      "step": 17250
    },
    {
      "epoch": 4.66,
      "learning_rate": 0.0003,
      "loss": 0.0082,
      "step": 17300
    },
    {
      "epoch": 4.68,
      "learning_rate": 0.0003,
      "loss": 0.0014,
      "step": 17350
    },
    {
      "epoch": 4.69,
      "learning_rate": 0.0003,
      "loss": 0.0114,
      "step": 17400
    },
    {
      "epoch": 4.7,
      "learning_rate": 0.0003,
      "loss": 0.0087,
      "step": 17450
    },
    {
      "epoch": 4.72,
      "learning_rate": 0.0003,
      "loss": 0.0029,
      "step": 17500
    },
    {
      "epoch": 4.73,
      "learning_rate": 0.0003,
      "loss": 0.0043,
      "step": 17550
    },
    {
      "epoch": 4.74,
      "learning_rate": 0.0003,
      "loss": 0.0017,
      "step": 17600
    },
    {
      "epoch": 4.76,
      "learning_rate": 0.0003,
      "loss": 0.0013,
      "step": 17650
    },
    {
      "epoch": 4.77,
      "learning_rate": 0.0003,
      "loss": 0.0058,
      "step": 17700
    },
    {
      "epoch": 4.78,
      "learning_rate": 0.0003,
      "loss": 0.0019,
      "step": 17750
    },
    {
      "epoch": 4.8,
      "learning_rate": 0.0003,
      "loss": 0.002,
      "step": 17800
    },
    {
      "epoch": 4.81,
      "learning_rate": 0.0003,
      "loss": 0.0043,
      "step": 17850
    },
    {
      "epoch": 4.82,
      "learning_rate": 0.0003,
      "loss": 0.0088,
      "step": 17900
    },
    {
      "epoch": 4.84,
      "learning_rate": 0.0003,
      "loss": 0.0042,
      "step": 17950
    },
    {
      "epoch": 4.85,
      "learning_rate": 0.0003,
      "loss": 0.0011,
      "step": 18000
    },
    {
      "epoch": 4.87,
      "learning_rate": 0.0003,
      "loss": 0.0078,
      "step": 18050
    },
    {
      "epoch": 4.88,
      "learning_rate": 0.0003,
      "loss": 0.0038,
      "step": 18100
    },
    {
      "epoch": 4.89,
      "learning_rate": 0.0003,
      "loss": 0.0041,
      "step": 18150
    },
    {
      "epoch": 4.91,
      "learning_rate": 0.0003,
      "loss": 0.0038,
      "step": 18200
    },
    {
      "epoch": 4.92,
      "learning_rate": 0.0003,
      "loss": 0.0023,
      "step": 18250
    },
    {
      "epoch": 4.93,
      "learning_rate": 0.0003,
      "loss": 0.0057,
      "step": 18300
    },
    {
      "epoch": 4.95,
      "learning_rate": 0.0003,
      "loss": 0.0016,
      "step": 18350
    },
    {
      "epoch": 4.96,
      "learning_rate": 0.0003,
      "loss": 0.0038,
      "step": 18400
    },
    {
      "epoch": 4.97,
      "learning_rate": 0.0003,
      "loss": 0.0028,
      "step": 18450
    },
    {
      "epoch": 4.99,
      "learning_rate": 0.0003,
      "loss": 0.0062,
      "step": 18500
    },
    {
      "epoch": 5.0,
      "learning_rate": 0.0003,
      "loss": 0.0006,
      "step": 18550
    },
    {
      "epoch": 5.0,
      "eval_f1": 22.2222,
      "eval_gen_len": 29.002111729698598,
      "eval_loss": 0.07783858478069305,
      "eval_runtime": 26.143,
      "eval_samples_per_second": 797.002,
      "eval_steps_per_second": 24.94,
      "step": 18550
    }
  ],
  "logging_steps": 50,
  "max_steps": 18550,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 250,
  "total_flos": 4.952289149190144e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}