{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 48626,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02,
      "learning_rate": 2.9691523053510468e-05,
      "loss": 0.5127,
      "step": 500
    },
    {
      "epoch": 0.04,
      "learning_rate": 2.9383046107020938e-05,
      "loss": 0.4648,
      "step": 1000
    },
    {
      "epoch": 0.06,
      "learning_rate": 2.9074569160531405e-05,
      "loss": 0.4986,
      "step": 1500
    },
    {
      "epoch": 0.08,
      "learning_rate": 2.8766092214041872e-05,
      "loss": 0.8909,
      "step": 2000
    },
    {
      "epoch": 0.1,
      "learning_rate": 2.845761526755234e-05,
      "loss": 0.6387,
      "step": 2500
    },
    {
      "epoch": 0.12,
      "learning_rate": 2.8149138321062806e-05,
      "loss": 0.5202,
      "step": 3000
    },
    {
      "epoch": 0.14,
      "learning_rate": 2.7840661374573273e-05,
      "loss": 0.5361,
      "step": 3500
    },
    {
      "epoch": 0.16,
      "learning_rate": 2.753218442808374e-05,
      "loss": 0.5472,
      "step": 4000
    },
    {
      "epoch": 0.19,
      "learning_rate": 2.722370748159421e-05,
      "loss": 0.6779,
      "step": 4500
    },
    {
      "epoch": 0.21,
      "learning_rate": 2.6915230535104677e-05,
      "loss": 0.5864,
      "step": 5000
    },
    {
      "epoch": 0.23,
      "learning_rate": 2.6606753588615144e-05,
      "loss": 0.4837,
      "step": 5500
    },
    {
      "epoch": 0.25,
      "learning_rate": 2.6298276642125614e-05,
      "loss": 0.6121,
      "step": 6000
    },
    {
      "epoch": 0.27,
      "learning_rate": 2.598979969563608e-05,
      "loss": 0.447,
      "step": 6500
    },
    {
      "epoch": 0.29,
      "learning_rate": 2.5681322749146548e-05,
      "loss": 0.473,
      "step": 7000
    },
    {
      "epoch": 0.31,
      "learning_rate": 2.5372845802657015e-05,
      "loss": 0.3499,
      "step": 7500
    },
    {
      "epoch": 0.33,
      "learning_rate": 2.5064368856167485e-05,
      "loss": 0.456,
      "step": 8000
    },
    {
      "epoch": 0.35,
      "learning_rate": 2.4755891909677952e-05,
      "loss": 0.4446,
      "step": 8500
    },
    {
      "epoch": 0.37,
      "learning_rate": 2.4447414963188416e-05,
      "loss": 0.3856,
      "step": 9000
    },
    {
      "epoch": 0.39,
      "learning_rate": 2.4138938016698886e-05,
      "loss": 0.4949,
      "step": 9500
    },
    {
      "epoch": 0.41,
      "learning_rate": 2.3830461070209353e-05,
      "loss": 0.4161,
      "step": 10000
    },
    {
      "epoch": 0.43,
      "learning_rate": 2.352198412371982e-05,
      "loss": 0.4024,
      "step": 10500
    },
    {
      "epoch": 0.45,
      "learning_rate": 2.321350717723029e-05,
      "loss": 0.4108,
      "step": 11000
    },
    {
      "epoch": 0.47,
      "learning_rate": 2.2905030230740757e-05,
      "loss": 0.4503,
      "step": 11500
    },
    {
      "epoch": 0.49,
      "learning_rate": 2.2596553284251224e-05,
      "loss": 0.3935,
      "step": 12000
    },
    {
      "epoch": 0.51,
      "learning_rate": 2.228807633776169e-05,
      "loss": 0.4171,
      "step": 12500
    },
    {
      "epoch": 0.53,
      "learning_rate": 2.197959939127216e-05,
      "loss": 0.4416,
      "step": 13000
    },
    {
      "epoch": 0.56,
      "learning_rate": 2.167112244478263e-05,
      "loss": 0.526,
      "step": 13500
    },
    {
      "epoch": 0.58,
      "learning_rate": 2.1362645498293095e-05,
      "loss": 0.4721,
      "step": 14000
    },
    {
      "epoch": 0.6,
      "learning_rate": 2.1054168551803562e-05,
      "loss": 0.3916,
      "step": 14500
    },
    {
      "epoch": 0.62,
      "learning_rate": 2.074569160531403e-05,
      "loss": 0.3965,
      "step": 15000
    },
    {
      "epoch": 0.64,
      "learning_rate": 2.0437214658824496e-05,
      "loss": 0.3889,
      "step": 15500
    },
    {
      "epoch": 0.66,
      "learning_rate": 2.0128737712334963e-05,
      "loss": 0.3888,
      "step": 16000
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.9820260765845433e-05,
      "loss": 0.4885,
      "step": 16500
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.95117838193559e-05,
      "loss": 0.4247,
      "step": 17000
    },
    {
      "epoch": 0.72,
      "learning_rate": 1.9203306872866367e-05,
      "loss": 0.3837,
      "step": 17500
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.8894829926376838e-05,
      "loss": 0.2933,
      "step": 18000
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.8586352979887305e-05,
      "loss": 0.2972,
      "step": 18500
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.827787603339777e-05,
      "loss": 0.4361,
      "step": 19000
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.7969399086908242e-05,
      "loss": 0.3345,
      "step": 19500
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.7660922140418705e-05,
      "loss": 0.4121,
      "step": 20000
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.7352445193929172e-05,
      "loss": 0.3455,
      "step": 20500
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.704396824743964e-05,
      "loss": 0.3316,
      "step": 21000
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.673549130095011e-05,
      "loss": 0.3792,
      "step": 21500
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.6427014354460576e-05,
      "loss": 0.3612,
      "step": 22000
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.6118537407971043e-05,
      "loss": 0.2907,
      "step": 22500
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.5810060461481514e-05,
      "loss": 0.3382,
      "step": 23000
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.550158351499198e-05,
      "loss": 0.347,
      "step": 23500
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.5193106568502448e-05,
      "loss": 0.3149,
      "step": 24000
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.4884629622012915e-05,
      "loss": 0.2851,
      "step": 24500
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.4576152675523383e-05,
      "loss": 0.1878,
      "step": 25000
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.426767572903385e-05,
      "loss": 0.2873,
      "step": 25500
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.3959198782544319e-05,
      "loss": 0.2119,
      "step": 26000
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.3650721836054786e-05,
      "loss": 0.3332,
      "step": 26500
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.3342244889565253e-05,
      "loss": 0.2739,
      "step": 27000
    },
    {
      "epoch": 1.13,
      "learning_rate": 1.3033767943075721e-05,
      "loss": 0.2526,
      "step": 27500
    },
    {
      "epoch": 1.15,
      "learning_rate": 1.2725290996586188e-05,
      "loss": 0.2085,
      "step": 28000
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.2416814050096657e-05,
      "loss": 0.2634,
      "step": 28500
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.2108337103607124e-05,
      "loss": 0.2051,
      "step": 29000
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.1799860157117592e-05,
      "loss": 0.2745,
      "step": 29500
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.149138321062806e-05,
      "loss": 0.2785,
      "step": 30000
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.1182906264138526e-05,
      "loss": 0.2079,
      "step": 30500
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.0874429317648995e-05,
      "loss": 0.1581,
      "step": 31000
    },
    {
      "epoch": 1.3,
      "learning_rate": 1.0565952371159462e-05,
      "loss": 0.2053,
      "step": 31500
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.025747542466993e-05,
      "loss": 0.271,
      "step": 32000
    },
    {
      "epoch": 1.34,
      "learning_rate": 9.948998478180399e-06,
      "loss": 0.226,
      "step": 32500
    },
    {
      "epoch": 1.36,
      "learning_rate": 9.640521531690864e-06,
      "loss": 0.2281,
      "step": 33000
    },
    {
      "epoch": 1.38,
      "learning_rate": 9.332044585201333e-06,
      "loss": 0.2202,
      "step": 33500
    },
    {
      "epoch": 1.4,
      "learning_rate": 9.0235676387118e-06,
      "loss": 0.1796,
      "step": 34000
    },
    {
      "epoch": 1.42,
      "learning_rate": 8.715090692222269e-06,
      "loss": 0.2394,
      "step": 34500
    },
    {
      "epoch": 1.44,
      "learning_rate": 8.406613745732735e-06,
      "loss": 0.2211,
      "step": 35000
    },
    {
      "epoch": 1.46,
      "learning_rate": 8.098136799243202e-06,
      "loss": 0.3344,
      "step": 35500
    },
    {
      "epoch": 1.48,
      "learning_rate": 7.789659852753671e-06,
      "loss": 0.2179,
      "step": 36000
    },
    {
      "epoch": 1.5,
      "learning_rate": 7.481182906264139e-06,
      "loss": 0.1424,
      "step": 36500
    },
    {
      "epoch": 1.52,
      "learning_rate": 7.172705959774607e-06,
      "loss": 0.2399,
      "step": 37000
    },
    {
      "epoch": 1.54,
      "learning_rate": 6.864229013285074e-06,
      "loss": 0.2546,
      "step": 37500
    },
    {
      "epoch": 1.56,
      "learning_rate": 6.555752066795541e-06,
      "loss": 0.2332,
      "step": 38000
    },
    {
      "epoch": 1.58,
      "learning_rate": 6.247275120306009e-06,
      "loss": 0.2694,
      "step": 38500
    },
    {
      "epoch": 1.6,
      "learning_rate": 5.938798173816477e-06,
      "loss": 0.1963,
      "step": 39000
    },
    {
      "epoch": 1.62,
      "learning_rate": 5.630321227326945e-06,
      "loss": 0.1356,
      "step": 39500
    },
    {
      "epoch": 1.65,
      "learning_rate": 5.3218442808374125e-06,
      "loss": 0.243,
      "step": 40000
    },
    {
      "epoch": 1.67,
      "learning_rate": 5.013367334347879e-06,
      "loss": 0.2149,
      "step": 40500
    },
    {
      "epoch": 1.69,
      "learning_rate": 4.704890387858347e-06,
      "loss": 0.2202,
      "step": 41000
    },
    {
      "epoch": 1.71,
      "learning_rate": 4.396413441368815e-06,
      "loss": 0.1844,
      "step": 41500
    },
    {
      "epoch": 1.73,
      "learning_rate": 4.087936494879283e-06,
      "loss": 0.166,
      "step": 42000
    },
    {
      "epoch": 1.75,
      "learning_rate": 3.7794595483897506e-06,
      "loss": 0.1799,
      "step": 42500
    },
    {
      "epoch": 1.77,
      "learning_rate": 3.470982601900218e-06,
      "loss": 0.0921,
      "step": 43000
    },
    {
      "epoch": 1.79,
      "learning_rate": 3.1625056554106857e-06,
      "loss": 0.1504,
      "step": 43500
    },
    {
      "epoch": 1.81,
      "learning_rate": 2.8540287089211535e-06,
      "loss": 0.2312,
      "step": 44000
    },
    {
      "epoch": 1.83,
      "learning_rate": 2.545551762431621e-06,
      "loss": 0.1744,
      "step": 44500
    },
    {
      "epoch": 1.85,
      "learning_rate": 2.2370748159420886e-06,
      "loss": 0.0672,
      "step": 45000
    },
    {
      "epoch": 1.87,
      "learning_rate": 1.9285978694525564e-06,
      "loss": 0.1917,
      "step": 45500
    },
    {
      "epoch": 1.89,
      "learning_rate": 1.6201209229630238e-06,
      "loss": 0.2083,
      "step": 46000
    },
    {
      "epoch": 1.91,
      "learning_rate": 1.3116439764734916e-06,
      "loss": 0.1929,
      "step": 46500
    },
    {
      "epoch": 1.93,
      "learning_rate": 1.0031670299839593e-06,
      "loss": 0.1463,
      "step": 47000
    },
    {
      "epoch": 1.95,
      "learning_rate": 6.946900834944268e-07,
      "loss": 0.2275,
      "step": 47500
    },
    {
      "epoch": 1.97,
      "learning_rate": 3.862131370048945e-07,
      "loss": 0.1238,
      "step": 48000
    },
    {
      "epoch": 1.99,
      "learning_rate": 7.773619051536215e-08,
      "loss": 0.1673,
      "step": 48500
    },
    {
      "epoch": 2.0,
      "step": 48626,
      "total_flos": 9.031922425511117e+16,
      "train_loss": 0.3265183397838395,
      "train_runtime": 46558.184,
      "train_samples_per_second": 2.089,
      "train_steps_per_second": 1.044
    }
  ],
  "logging_steps": 500,
  "max_steps": 48626,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "total_flos": 9.031922425511117e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}