{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 49023,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03,
      "learning_rate": 5e-05,
      "loss": 4.8278,
      "step": 500
    },
    {
      "epoch": 0.06,
      "learning_rate": 5e-05,
      "loss": 3.5561,
      "step": 1000
    },
    {
      "epoch": 0.09,
      "learning_rate": 5e-05,
      "loss": 3.0434,
      "step": 1500
    },
    {
      "epoch": 0.12,
      "learning_rate": 5e-05,
      "loss": 2.7336,
      "step": 2000
    },
    {
      "epoch": 0.15,
      "learning_rate": 5e-05,
      "loss": 2.5149,
      "step": 2500
    },
    {
      "epoch": 0.18,
      "learning_rate": 5e-05,
      "loss": 2.3549,
      "step": 3000
    },
    {
      "epoch": 0.21,
      "learning_rate": 5e-05,
      "loss": 2.2287,
      "step": 3500
    },
    {
      "epoch": 0.24,
      "learning_rate": 5e-05,
      "loss": 2.1257,
      "step": 4000
    },
    {
      "epoch": 0.28,
      "learning_rate": 5e-05,
      "loss": 2.0463,
      "step": 4500
    },
    {
      "epoch": 0.31,
      "learning_rate": 5e-05,
      "loss": 1.9848,
      "step": 5000
    },
    {
      "epoch": 0.34,
      "learning_rate": 5e-05,
      "loss": 1.9228,
      "step": 5500
    },
    {
      "epoch": 0.37,
      "learning_rate": 5e-05,
      "loss": 1.8806,
      "step": 6000
    },
    {
      "epoch": 0.4,
      "learning_rate": 5e-05,
      "loss": 1.8365,
      "step": 6500
    },
    {
      "epoch": 0.43,
      "learning_rate": 5e-05,
      "loss": 1.801,
      "step": 7000
    },
    {
      "epoch": 0.46,
      "learning_rate": 5e-05,
      "loss": 1.7703,
      "step": 7500
    },
    {
      "epoch": 0.49,
      "learning_rate": 5e-05,
      "loss": 1.7434,
      "step": 8000
    },
    {
      "epoch": 0.52,
      "learning_rate": 5e-05,
      "loss": 1.7187,
      "step": 8500
    },
    {
      "epoch": 0.55,
      "learning_rate": 5e-05,
      "loss": 1.697,
      "step": 9000
    },
    {
      "epoch": 0.58,
      "learning_rate": 5e-05,
      "loss": 1.673,
      "step": 9500
    },
    {
      "epoch": 0.61,
      "learning_rate": 5e-05,
      "loss": 1.6555,
      "step": 10000
    },
    {
      "epoch": 0.64,
      "learning_rate": 5e-05,
      "loss": 1.6399,
      "step": 10500
    },
    {
      "epoch": 0.67,
      "learning_rate": 5e-05,
      "loss": 1.6233,
      "step": 11000
    },
    {
      "epoch": 0.7,
      "learning_rate": 5e-05,
      "loss": 1.6087,
      "step": 11500
    },
    {
      "epoch": 0.73,
      "learning_rate": 5e-05,
      "loss": 1.597,
      "step": 12000
    },
    {
      "epoch": 0.76,
      "learning_rate": 5e-05,
      "loss": 1.5816,
      "step": 12500
    },
    {
      "epoch": 0.8,
      "learning_rate": 5e-05,
      "loss": 1.5684,
      "step": 13000
    },
    {
      "epoch": 0.83,
      "learning_rate": 5e-05,
      "loss": 1.5602,
      "step": 13500
    },
    {
      "epoch": 0.86,
      "learning_rate": 5e-05,
      "loss": 1.5453,
      "step": 14000
    },
    {
      "epoch": 0.89,
      "learning_rate": 5e-05,
      "loss": 1.5366,
      "step": 14500
    },
    {
      "epoch": 0.92,
      "learning_rate": 5e-05,
      "loss": 1.5276,
      "step": 15000
    },
    {
      "epoch": 0.95,
      "learning_rate": 5e-05,
      "loss": 1.5156,
      "step": 15500
    },
    {
      "epoch": 0.98,
      "learning_rate": 5e-05,
      "loss": 1.5075,
      "step": 16000
    },
    {
      "epoch": 1.01,
      "learning_rate": 5e-05,
      "loss": 1.4988,
      "step": 16500
    },
    {
      "epoch": 1.04,
      "learning_rate": 5e-05,
      "loss": 1.4935,
      "step": 17000
    },
    {
      "epoch": 1.07,
      "learning_rate": 5e-05,
      "loss": 1.4818,
      "step": 17500
    },
    {
      "epoch": 1.1,
      "learning_rate": 5e-05,
      "loss": 1.4733,
      "step": 18000
    },
    {
      "epoch": 1.13,
      "learning_rate": 5e-05,
      "loss": 1.4661,
      "step": 18500
    },
    {
      "epoch": 1.16,
      "learning_rate": 5e-05,
      "loss": 1.4603,
      "step": 19000
    },
    {
      "epoch": 1.19,
      "learning_rate": 5e-05,
      "loss": 1.4554,
      "step": 19500
    },
    {
      "epoch": 1.22,
      "learning_rate": 5e-05,
      "loss": 1.449,
      "step": 20000
    },
    {
      "epoch": 1.25,
      "learning_rate": 5e-05,
      "loss": 1.4409,
      "step": 20500
    },
    {
      "epoch": 1.29,
      "learning_rate": 5e-05,
      "loss": 1.4382,
      "step": 21000
    },
    {
      "epoch": 1.32,
      "learning_rate": 5e-05,
      "loss": 1.4298,
      "step": 21500
    },
    {
      "epoch": 1.35,
      "learning_rate": 5e-05,
      "loss": 1.4251,
      "step": 22000
    },
    {
      "epoch": 1.38,
      "learning_rate": 5e-05,
      "loss": 1.4225,
      "step": 22500
    },
    {
      "epoch": 1.41,
      "learning_rate": 5e-05,
      "loss": 1.4122,
      "step": 23000
    },
    {
      "epoch": 1.44,
      "learning_rate": 5e-05,
      "loss": 1.4089,
      "step": 23500
    },
    {
      "epoch": 1.47,
      "learning_rate": 5e-05,
      "loss": 1.4041,
      "step": 24000
    },
    {
      "epoch": 1.5,
      "learning_rate": 5e-05,
      "loss": 1.3998,
      "step": 24500
    },
    {
      "epoch": 1.53,
      "learning_rate": 5e-05,
      "loss": 1.3933,
      "step": 25000
    },
    {
      "epoch": 1.56,
      "learning_rate": 5e-05,
      "loss": 1.3897,
      "step": 25500
    },
    {
      "epoch": 1.59,
      "learning_rate": 5e-05,
      "loss": 1.3874,
      "step": 26000
    },
    {
      "epoch": 1.62,
      "learning_rate": 5e-05,
      "loss": 1.3812,
      "step": 26500
    },
    {
      "epoch": 1.65,
      "learning_rate": 5e-05,
      "loss": 1.3745,
      "step": 27000
    },
    {
      "epoch": 1.68,
      "learning_rate": 5e-05,
      "loss": 1.3728,
      "step": 27500
    },
    {
      "epoch": 1.71,
      "learning_rate": 5e-05,
      "loss": 1.3706,
      "step": 28000
    },
    {
      "epoch": 1.74,
      "learning_rate": 5e-05,
      "loss": 1.3673,
      "step": 28500
    },
    {
      "epoch": 1.77,
      "learning_rate": 5e-05,
      "loss": 1.3605,
      "step": 29000
    },
    {
      "epoch": 1.81,
      "learning_rate": 5e-05,
      "loss": 1.3574,
      "step": 29500
    },
    {
      "epoch": 1.84,
      "learning_rate": 5e-05,
      "loss": 1.3514,
      "step": 30000
    },
    {
      "epoch": 1.87,
      "learning_rate": 5e-05,
      "loss": 1.351,
      "step": 30500
    },
    {
      "epoch": 1.9,
      "learning_rate": 5e-05,
      "loss": 1.3457,
      "step": 31000
    },
    {
      "epoch": 1.93,
      "learning_rate": 5e-05,
      "loss": 1.3451,
      "step": 31500
    },
    {
      "epoch": 1.96,
      "learning_rate": 5e-05,
      "loss": 1.3403,
      "step": 32000
    },
    {
      "epoch": 1.99,
      "learning_rate": 5e-05,
      "loss": 1.3389,
      "step": 32500
    },
    {
      "epoch": 2.02,
      "learning_rate": 5e-05,
      "loss": 1.3346,
      "step": 33000
    },
    {
      "epoch": 2.05,
      "learning_rate": 5e-05,
      "loss": 1.3313,
      "step": 33500
    },
    {
      "epoch": 2.08,
      "learning_rate": 5e-05,
      "loss": 1.3278,
      "step": 34000
    },
    {
      "epoch": 2.11,
      "learning_rate": 5e-05,
      "loss": 1.3249,
      "step": 34500
    },
    {
      "epoch": 2.14,
      "learning_rate": 5e-05,
      "loss": 1.3212,
      "step": 35000
    },
    {
      "epoch": 2.17,
      "learning_rate": 5e-05,
      "loss": 1.3191,
      "step": 35500
    },
    {
      "epoch": 2.2,
      "learning_rate": 5e-05,
      "loss": 1.3158,
      "step": 36000
    },
    {
      "epoch": 2.23,
      "learning_rate": 5e-05,
      "loss": 1.3135,
      "step": 36500
    },
    {
      "epoch": 2.26,
      "learning_rate": 5e-05,
      "loss": 1.3086,
      "step": 37000
    },
    {
      "epoch": 2.29,
      "learning_rate": 5e-05,
      "loss": 1.3077,
      "step": 37500
    },
    {
      "epoch": 2.33,
      "learning_rate": 5e-05,
      "loss": 1.3045,
      "step": 38000
    },
    {
      "epoch": 2.36,
      "learning_rate": 5e-05,
      "loss": 1.3013,
      "step": 38500
    },
    {
      "epoch": 2.39,
      "learning_rate": 5e-05,
      "loss": 1.3017,
      "step": 39000
    },
    {
      "epoch": 2.42,
      "learning_rate": 5e-05,
      "loss": 1.2989,
      "step": 39500
    },
    {
      "epoch": 2.45,
      "learning_rate": 5e-05,
      "loss": 1.2963,
      "step": 40000
    },
    {
      "epoch": 2.48,
      "learning_rate": 5e-05,
      "loss": 1.2916,
      "step": 40500
    },
    {
      "epoch": 2.51,
      "learning_rate": 5e-05,
      "loss": 1.2935,
      "step": 41000
    },
    {
      "epoch": 2.54,
      "learning_rate": 5e-05,
      "loss": 1.2922,
      "step": 41500
    },
    {
      "epoch": 2.57,
      "learning_rate": 5e-05,
      "loss": 1.2867,
      "step": 42000
    },
    {
      "epoch": 2.6,
      "learning_rate": 5e-05,
      "loss": 1.2847,
      "step": 42500
    },
    {
      "epoch": 2.63,
      "learning_rate": 5e-05,
      "loss": 1.2831,
      "step": 43000
    },
    {
      "epoch": 2.66,
      "learning_rate": 5e-05,
      "loss": 1.2802,
      "step": 43500
    },
    {
      "epoch": 2.69,
      "learning_rate": 5e-05,
      "loss": 1.2784,
      "step": 44000
    },
    {
      "epoch": 2.72,
      "learning_rate": 5e-05,
      "loss": 1.2748,
      "step": 44500
    },
    {
      "epoch": 2.75,
      "learning_rate": 5e-05,
      "loss": 1.2779,
      "step": 45000
    },
    {
      "epoch": 2.78,
      "learning_rate": 5e-05,
      "loss": 1.2712,
      "step": 45500
    },
    {
      "epoch": 2.82,
      "learning_rate": 5e-05,
      "loss": 1.2695,
      "step": 46000
    },
    {
      "epoch": 2.85,
      "learning_rate": 5e-05,
      "loss": 1.2677,
      "step": 46500
    },
    {
      "epoch": 2.88,
      "learning_rate": 5e-05,
      "loss": 1.2682,
      "step": 47000
    },
    {
      "epoch": 2.91,
      "learning_rate": 5e-05,
      "loss": 1.2651,
      "step": 47500
    },
    {
      "epoch": 2.94,
      "learning_rate": 5e-05,
      "loss": 1.2635,
      "step": 48000
    },
    {
      "epoch": 2.97,
      "learning_rate": 5e-05,
      "loss": 1.2626,
      "step": 48500
    },
    {
      "epoch": 3.0,
      "learning_rate": 5e-05,
      "loss": 1.2583,
      "step": 49000
    },
    {
      "epoch": 3.0,
      "step": 49023,
      "total_flos": 9.937271302396051e+18,
      "train_loss": 1.5609709015213173,
      "train_runtime": 628686.7825,
      "train_samples_per_second": 59.886,
      "train_steps_per_second": 0.078
    }
  ],
  "logging_steps": 500,
  "max_steps": 49023,
  "num_train_epochs": 3,
  "save_steps": 10000,
  "total_flos": 9.937271302396051e+18,
  "trial_name": null,
  "trial_params": null
}