{
  "best_metric": 6.510499000549316,
  "best_model_checkpoint": "/temp/t5_small-qg-aap/checkpoint-4040",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 4040,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.12,
      "learning_rate": 0.004876237623762376,
      "loss": 8.0829,
      "step": 100
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.004752475247524752,
      "loss": 7.6287,
      "step": 200
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.004628712871287129,
      "loss": 7.3201,
      "step": 300
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0045049504950495055,
      "loss": 7.066,
      "step": 400
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.004381188118811882,
      "loss": 6.8463,
      "step": 500
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.004257425742574258,
      "loss": 6.696,
      "step": 600
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.004133663366336634,
      "loss": 6.5656,
      "step": 700
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.00400990099009901,
      "loss": 6.4555,
      "step": 800
    },
    {
      "epoch": 1.0,
      "eval_loss": 8.181524276733398,
      "eval_runtime": 87.6052,
      "eval_samples_per_second": 101.238,
      "eval_steps_per_second": 25.318,
      "step": 808
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.0038861386138613866,
      "loss": 6.3186,
      "step": 900
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0037623762376237627,
      "loss": 6.2484,
      "step": 1000
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.0036386138613861384,
      "loss": 6.1225,
      "step": 1100
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.0035148514851485145,
      "loss": 6.0828,
      "step": 1200
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0033910891089108915,
      "loss": 5.9965,
      "step": 1300
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0032673267326732676,
      "loss": 5.888,
      "step": 1400
    },
    {
      "epoch": 1.86,
      "learning_rate": 0.0031435643564356438,
      "loss": 5.8093,
      "step": 1500
    },
    {
      "epoch": 1.98,
      "learning_rate": 0.00301980198019802,
      "loss": 5.7684,
      "step": 1600
    },
    {
      "epoch": 2.0,
      "eval_loss": 7.345946788787842,
      "eval_runtime": 87.5869,
      "eval_samples_per_second": 101.259,
      "eval_steps_per_second": 25.323,
      "step": 1616
    },
    {
      "epoch": 2.1,
      "learning_rate": 0.002896039603960396,
      "loss": 5.6732,
      "step": 1700
    },
    {
      "epoch": 2.23,
      "learning_rate": 0.0027722772277227726,
      "loss": 5.6518,
      "step": 1800
    },
    {
      "epoch": 2.35,
      "learning_rate": 0.0026485148514851487,
      "loss": 5.5807,
      "step": 1900
    },
    {
      "epoch": 2.48,
      "learning_rate": 0.002524752475247525,
      "loss": 5.5466,
      "step": 2000
    },
    {
      "epoch": 2.6,
      "learning_rate": 0.002400990099009901,
      "loss": 5.5098,
      "step": 2100
    },
    {
      "epoch": 2.72,
      "learning_rate": 0.0022772277227722775,
      "loss": 5.462,
      "step": 2200
    },
    {
      "epoch": 2.85,
      "learning_rate": 0.0021534653465346536,
      "loss": 5.4502,
      "step": 2300
    },
    {
      "epoch": 2.97,
      "learning_rate": 0.0020297029702970298,
      "loss": 5.3707,
      "step": 2400
    },
    {
      "epoch": 3.0,
      "eval_loss": 6.852190971374512,
      "eval_runtime": 87.7618,
      "eval_samples_per_second": 101.058,
      "eval_steps_per_second": 25.273,
      "step": 2424
    },
    {
      "epoch": 3.09,
      "learning_rate": 0.0019059405940594061,
      "loss": 5.3459,
      "step": 2500
    },
    {
      "epoch": 3.22,
      "learning_rate": 0.001782178217821782,
      "loss": 5.3014,
      "step": 2600
    },
    {
      "epoch": 3.34,
      "learning_rate": 0.0016584158415841586,
      "loss": 5.3001,
      "step": 2700
    },
    {
      "epoch": 3.47,
      "learning_rate": 0.0015346534653465347,
      "loss": 5.2818,
      "step": 2800
    },
    {
      "epoch": 3.59,
      "learning_rate": 0.0014108910891089108,
      "loss": 5.23,
      "step": 2900
    },
    {
      "epoch": 3.71,
      "learning_rate": 0.0012871287128712872,
      "loss": 5.185,
      "step": 3000
    },
    {
      "epoch": 3.84,
      "learning_rate": 0.0011633663366336635,
      "loss": 5.176,
      "step": 3100
    },
    {
      "epoch": 3.96,
      "learning_rate": 0.0010396039603960397,
      "loss": 5.1609,
      "step": 3200
    },
    {
      "epoch": 4.0,
      "eval_loss": 6.593112945556641,
      "eval_runtime": 87.448,
      "eval_samples_per_second": 101.42,
      "eval_steps_per_second": 25.364,
      "step": 3232
    },
    {
      "epoch": 4.08,
      "learning_rate": 0.0009158415841584158,
      "loss": 5.1834,
      "step": 3300
    },
    {
      "epoch": 4.21,
      "learning_rate": 0.0007920792079207921,
      "loss": 5.1021,
      "step": 3400
    },
    {
      "epoch": 4.33,
      "learning_rate": 0.0006683168316831684,
      "loss": 5.1443,
      "step": 3500
    },
    {
      "epoch": 4.46,
      "learning_rate": 0.0005445544554455446,
      "loss": 5.0904,
      "step": 3600
    },
    {
      "epoch": 4.58,
      "learning_rate": 0.00042079207920792084,
      "loss": 5.0949,
      "step": 3700
    },
    {
      "epoch": 4.7,
      "learning_rate": 0.000297029702970297,
      "loss": 5.0856,
      "step": 3800
    },
    {
      "epoch": 4.83,
      "learning_rate": 0.00017326732673267329,
      "loss": 5.1251,
      "step": 3900
    },
    {
      "epoch": 4.95,
      "learning_rate": 4.950495049504951e-05,
      "loss": 5.1034,
      "step": 4000
    },
    {
      "epoch": 5.0,
      "eval_loss": 6.510499000549316,
      "eval_runtime": 87.5193,
      "eval_samples_per_second": 101.338,
      "eval_steps_per_second": 25.343,
      "step": 4040
    },
    {
      "epoch": 5.0,
      "step": 4040,
      "total_flos": 3.621491878920192e+16,
      "train_loss": 5.794503840833607,
      "train_runtime": 2995.2119,
      "train_samples_per_second": 86.311,
      "train_steps_per_second": 1.349
    }
  ],
  "logging_steps": 100,
  "max_steps": 4040,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 3.621491878920192e+16,
  "trial_name": null,
  "trial_params": null
}