{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.8011204481792715,
  "eval_steps": 50,
  "global_step": 3000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.05,
      "eval_loss": 22.8203182220459,
      "eval_runtime": 3.0437,
      "eval_samples_per_second": 32.854,
      "eval_steps_per_second": 4.271,
      "step": 50
    },
    {
      "epoch": 0.09,
      "eval_loss": 21.029277801513672,
      "eval_runtime": 3.063,
      "eval_samples_per_second": 32.648,
      "eval_steps_per_second": 4.244,
      "step": 100
    },
    {
      "epoch": 0.14,
      "eval_loss": 20.389341354370117,
      "eval_runtime": 3.0675,
      "eval_samples_per_second": 32.6,
      "eval_steps_per_second": 4.238,
      "step": 150
    },
    {
      "epoch": 0.19,
      "eval_loss": 19.8281307220459,
      "eval_runtime": 3.071,
      "eval_samples_per_second": 32.563,
      "eval_steps_per_second": 4.233,
      "step": 200
    },
    {
      "epoch": 0.23,
      "eval_loss": 18.816303253173828,
      "eval_runtime": 3.0799,
      "eval_samples_per_second": 32.468,
      "eval_steps_per_second": 4.221,
      "step": 250
    },
    {
      "epoch": 0.28,
      "eval_loss": 18.291460037231445,
      "eval_runtime": 3.0838,
      "eval_samples_per_second": 32.427,
      "eval_steps_per_second": 4.216,
      "step": 300
    },
    {
      "epoch": 0.33,
      "eval_loss": 17.624303817749023,
      "eval_runtime": 3.0609,
      "eval_samples_per_second": 32.67,
      "eval_steps_per_second": 4.247,
      "step": 350
    },
    {
      "epoch": 0.37,
      "eval_loss": 17.041908264160156,
      "eval_runtime": 3.0724,
      "eval_samples_per_second": 32.547,
      "eval_steps_per_second": 4.231,
      "step": 400
    },
    {
      "epoch": 0.42,
      "eval_loss": 16.60989761352539,
      "eval_runtime": 3.0682,
      "eval_samples_per_second": 32.592,
      "eval_steps_per_second": 4.237,
      "step": 450
    },
    {
      "epoch": 0.47,
      "learning_rate": 8.75e-05,
      "loss": 20.2413,
      "step": 500
    },
    {
      "epoch": 0.47,
      "eval_loss": 15.918641090393066,
      "eval_runtime": 3.0677,
      "eval_samples_per_second": 32.598,
      "eval_steps_per_second": 4.238,
      "step": 500
    },
    {
      "epoch": 0.51,
      "eval_loss": 15.275712013244629,
      "eval_runtime": 3.0784,
      "eval_samples_per_second": 32.485,
      "eval_steps_per_second": 4.223,
      "step": 550
    },
    {
      "epoch": 0.56,
      "eval_loss": 14.903023719787598,
      "eval_runtime": 3.0814,
      "eval_samples_per_second": 32.453,
      "eval_steps_per_second": 4.219,
      "step": 600
    },
    {
      "epoch": 0.61,
      "eval_loss": 14.478271484375,
      "eval_runtime": 3.0668,
      "eval_samples_per_second": 32.608,
      "eval_steps_per_second": 4.239,
      "step": 650
    },
    {
      "epoch": 0.65,
      "eval_loss": 14.302887916564941,
      "eval_runtime": 3.0701,
      "eval_samples_per_second": 32.572,
      "eval_steps_per_second": 4.234,
      "step": 700
    },
    {
      "epoch": 0.7,
      "eval_loss": 13.83132553100586,
      "eval_runtime": 3.0703,
      "eval_samples_per_second": 32.57,
      "eval_steps_per_second": 4.234,
      "step": 750
    },
    {
      "epoch": 0.75,
      "eval_loss": 13.52160358428955,
      "eval_runtime": 3.0694,
      "eval_samples_per_second": 32.579,
      "eval_steps_per_second": 4.235,
      "step": 800
    },
    {
      "epoch": 0.79,
      "eval_loss": 12.77853012084961,
      "eval_runtime": 3.0776,
      "eval_samples_per_second": 32.493,
      "eval_steps_per_second": 4.224,
      "step": 850
    },
    {
      "epoch": 0.84,
      "eval_loss": 12.551227569580078,
      "eval_runtime": 3.0749,
      "eval_samples_per_second": 32.521,
      "eval_steps_per_second": 4.228,
      "step": 900
    },
    {
      "epoch": 0.89,
      "eval_loss": 12.139355659484863,
      "eval_runtime": 3.0686,
      "eval_samples_per_second": 32.588,
      "eval_steps_per_second": 4.236,
      "step": 950
    },
    {
      "epoch": 0.93,
      "learning_rate": 7.500000000000001e-05,
      "loss": 16.164,
      "step": 1000
    },
    {
      "epoch": 0.93,
      "eval_loss": 11.858689308166504,
      "eval_runtime": 3.0577,
      "eval_samples_per_second": 32.704,
      "eval_steps_per_second": 4.252,
      "step": 1000
    },
    {
      "epoch": 0.98,
      "eval_loss": 11.4910249710083,
      "eval_runtime": 3.071,
      "eval_samples_per_second": 32.562,
      "eval_steps_per_second": 4.233,
      "step": 1050
    },
    {
      "epoch": 1.03,
      "eval_loss": 11.197161674499512,
      "eval_runtime": 3.0645,
      "eval_samples_per_second": 32.632,
      "eval_steps_per_second": 4.242,
      "step": 1100
    },
    {
      "epoch": 1.07,
      "eval_loss": 10.903576850891113,
      "eval_runtime": 3.0551,
      "eval_samples_per_second": 32.732,
      "eval_steps_per_second": 4.255,
      "step": 1150
    },
    {
      "epoch": 1.12,
      "eval_loss": 10.947243690490723,
      "eval_runtime": 3.0615,
      "eval_samples_per_second": 32.664,
      "eval_steps_per_second": 4.246,
      "step": 1200
    },
    {
      "epoch": 1.17,
      "eval_loss": 10.560125350952148,
      "eval_runtime": 3.0709,
      "eval_samples_per_second": 32.563,
      "eval_steps_per_second": 4.233,
      "step": 1250
    },
    {
      "epoch": 1.21,
      "eval_loss": 10.332932472229004,
      "eval_runtime": 3.0883,
      "eval_samples_per_second": 32.38,
      "eval_steps_per_second": 4.209,
      "step": 1300
    },
    {
      "epoch": 1.26,
      "eval_loss": 9.66699504852295,
      "eval_runtime": 3.0701,
      "eval_samples_per_second": 32.572,
      "eval_steps_per_second": 4.234,
      "step": 1350
    },
    {
      "epoch": 1.31,
      "eval_loss": 9.532146453857422,
      "eval_runtime": 3.0603,
      "eval_samples_per_second": 32.676,
      "eval_steps_per_second": 4.248,
      "step": 1400
    },
    {
      "epoch": 1.35,
      "eval_loss": 9.379703521728516,
      "eval_runtime": 3.0723,
      "eval_samples_per_second": 32.549,
      "eval_steps_per_second": 4.231,
      "step": 1450
    },
    {
      "epoch": 1.4,
      "learning_rate": 6.25e-05,
      "loss": 12.3127,
      "step": 1500
    },
    {
      "epoch": 1.4,
      "eval_loss": 8.9696683883667,
      "eval_runtime": 3.0645,
      "eval_samples_per_second": 32.632,
      "eval_steps_per_second": 4.242,
      "step": 1500
    },
    {
      "epoch": 1.45,
      "eval_loss": 8.936990737915039,
      "eval_runtime": 3.0799,
      "eval_samples_per_second": 32.468,
      "eval_steps_per_second": 4.221,
      "step": 1550
    },
    {
      "epoch": 1.49,
      "eval_loss": 8.862835884094238,
      "eval_runtime": 3.0649,
      "eval_samples_per_second": 32.627,
      "eval_steps_per_second": 4.242,
      "step": 1600
    },
    {
      "epoch": 1.54,
      "eval_loss": 8.701496124267578,
      "eval_runtime": 3.0772,
      "eval_samples_per_second": 32.497,
      "eval_steps_per_second": 4.225,
      "step": 1650
    },
    {
      "epoch": 1.59,
      "eval_loss": 8.539968490600586,
      "eval_runtime": 3.0598,
      "eval_samples_per_second": 32.682,
      "eval_steps_per_second": 4.249,
      "step": 1700
    },
    {
      "epoch": 1.63,
      "eval_loss": 8.266422271728516,
      "eval_runtime": 3.0852,
      "eval_samples_per_second": 32.413,
      "eval_steps_per_second": 4.214,
      "step": 1750
    },
    {
      "epoch": 1.68,
      "eval_loss": 8.135841369628906,
      "eval_runtime": 3.0678,
      "eval_samples_per_second": 32.596,
      "eval_steps_per_second": 4.238,
      "step": 1800
    },
    {
      "epoch": 1.73,
      "eval_loss": 7.678401470184326,
      "eval_runtime": 3.0876,
      "eval_samples_per_second": 32.388,
      "eval_steps_per_second": 4.21,
      "step": 1850
    },
    {
      "epoch": 1.77,
      "eval_loss": 7.719874382019043,
      "eval_runtime": 3.0607,
      "eval_samples_per_second": 32.672,
      "eval_steps_per_second": 4.247,
      "step": 1900
    },
    {
      "epoch": 1.82,
      "eval_loss": 7.3947248458862305,
      "eval_runtime": 3.0791,
      "eval_samples_per_second": 32.477,
      "eval_steps_per_second": 4.222,
      "step": 1950
    },
    {
      "epoch": 1.87,
      "learning_rate": 5e-05,
      "loss": 10.9628,
      "step": 2000
    },
    {
      "epoch": 1.87,
      "eval_loss": 7.449285984039307,
      "eval_runtime": 3.0645,
      "eval_samples_per_second": 32.631,
      "eval_steps_per_second": 4.242,
      "step": 2000
    },
    {
      "epoch": 1.91,
      "eval_loss": 7.32260274887085,
      "eval_runtime": 3.0717,
      "eval_samples_per_second": 32.555,
      "eval_steps_per_second": 4.232,
      "step": 2050
    },
    {
      "epoch": 1.96,
      "eval_loss": 7.2264933586120605,
      "eval_runtime": 3.0789,
      "eval_samples_per_second": 32.479,
      "eval_steps_per_second": 4.222,
      "step": 2100
    },
    {
      "epoch": 2.01,
      "eval_loss": 7.041122913360596,
      "eval_runtime": 3.0753,
      "eval_samples_per_second": 32.517,
      "eval_steps_per_second": 4.227,
      "step": 2150
    },
    {
      "epoch": 2.05,
      "eval_loss": 6.81175422668457,
      "eval_runtime": 3.0744,
      "eval_samples_per_second": 32.527,
      "eval_steps_per_second": 4.229,
      "step": 2200
    },
    {
      "epoch": 2.1,
      "eval_loss": 6.902734279632568,
      "eval_runtime": 3.0892,
      "eval_samples_per_second": 32.371,
      "eval_steps_per_second": 4.208,
      "step": 2250
    },
    {
      "epoch": 2.15,
      "eval_loss": 6.792438507080078,
      "eval_runtime": 3.0726,
      "eval_samples_per_second": 32.545,
      "eval_steps_per_second": 4.231,
      "step": 2300
    },
    {
      "epoch": 2.19,
      "eval_loss": 6.703750133514404,
      "eval_runtime": 3.0901,
      "eval_samples_per_second": 32.361,
      "eval_steps_per_second": 4.207,
      "step": 2350
    },
    {
      "epoch": 2.24,
      "eval_loss": 6.512704849243164,
      "eval_runtime": 3.0957,
      "eval_samples_per_second": 32.303,
      "eval_steps_per_second": 4.199,
      "step": 2400
    },
    {
      "epoch": 2.29,
      "eval_loss": 6.426401138305664,
      "eval_runtime": 3.0916,
      "eval_samples_per_second": 32.346,
      "eval_steps_per_second": 4.205,
      "step": 2450
    },
    {
      "epoch": 2.33,
      "learning_rate": 3.7500000000000003e-05,
      "loss": 9.3224,
      "step": 2500
    },
    {
      "epoch": 2.33,
      "eval_loss": 6.311591148376465,
      "eval_runtime": 3.0884,
      "eval_samples_per_second": 32.379,
      "eval_steps_per_second": 4.209,
      "step": 2500
    },
    {
      "epoch": 2.38,
      "eval_loss": 6.2699294090271,
      "eval_runtime": 3.0935,
      "eval_samples_per_second": 32.326,
      "eval_steps_per_second": 4.202,
      "step": 2550
    },
    {
      "epoch": 2.43,
      "eval_loss": 6.2037458419799805,
      "eval_runtime": 3.0934,
      "eval_samples_per_second": 32.327,
      "eval_steps_per_second": 4.202,
      "step": 2600
    },
    {
      "epoch": 2.47,
      "eval_loss": 6.107393264770508,
      "eval_runtime": 3.0876,
      "eval_samples_per_second": 32.388,
      "eval_steps_per_second": 4.21,
      "step": 2650
    },
    {
      "epoch": 2.52,
      "eval_loss": 5.982547760009766,
      "eval_runtime": 3.0732,
      "eval_samples_per_second": 32.54,
      "eval_steps_per_second": 4.23,
      "step": 2700
    },
    {
      "epoch": 2.57,
      "eval_loss": 5.9501118659973145,
      "eval_runtime": 3.0562,
      "eval_samples_per_second": 32.72,
      "eval_steps_per_second": 4.254,
      "step": 2750
    },
    {
      "epoch": 2.61,
      "eval_loss": 5.896381855010986,
      "eval_runtime": 3.0684,
      "eval_samples_per_second": 32.59,
      "eval_steps_per_second": 4.237,
      "step": 2800
    },
    {
      "epoch": 2.66,
      "eval_loss": 5.893676280975342,
      "eval_runtime": 3.076,
      "eval_samples_per_second": 32.51,
      "eval_steps_per_second": 4.226,
      "step": 2850
    },
    {
      "epoch": 2.71,
      "eval_loss": 5.809429168701172,
      "eval_runtime": 3.0738,
      "eval_samples_per_second": 32.533,
      "eval_steps_per_second": 4.229,
      "step": 2900
    },
    {
      "epoch": 2.75,
      "eval_loss": 5.65841817855835,
      "eval_runtime": 3.0612,
      "eval_samples_per_second": 32.667,
      "eval_steps_per_second": 4.247,
      "step": 2950
    },
    {
      "epoch": 2.8,
      "learning_rate": 2.5e-05,
      "loss": 8.3258,
      "step": 3000
    },
    {
      "epoch": 2.8,
      "eval_loss": 5.634002208709717,
      "eval_runtime": 3.078,
      "eval_samples_per_second": 32.488,
      "eval_steps_per_second": 4.224,
      "step": 3000
    }
  ],
  "logging_steps": 500,
  "max_steps": 4000,
  "num_train_epochs": 4,
  "save_steps": 1000,
  "total_flos": 4.107859357335552e+16,
  "trial_name": null,
  "trial_params": null
}