{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 20.0,
  "global_step": 5600,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8955357074737549,
      "eval_loss": 0.33214062452316284,
      "eval_runtime": 7.5153,
      "eval_samples_per_second": 149.029,
      "eval_steps_per_second": 18.629,
      "step": 280
    },
    {
      "epoch": 1.79,
      "learning_rate": 2.743006199096114e-05,
      "loss": 0.3501,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9142857193946838,
      "eval_loss": 0.29908323287963867,
      "eval_runtime": 7.7634,
      "eval_samples_per_second": 144.266,
      "eval_steps_per_second": 18.033,
      "step": 560
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9008928537368774,
      "eval_loss": 0.2781011164188385,
      "eval_runtime": 7.6341,
      "eval_samples_per_second": 146.71,
      "eval_steps_per_second": 18.339,
      "step": 840
    },
    {
      "epoch": 3.57,
      "learning_rate": 2.474084022714142e-05,
      "loss": 0.1794,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9107142686843872,
      "eval_loss": 0.37903639674186707,
      "eval_runtime": 7.6349,
      "eval_samples_per_second": 146.696,
      "eval_steps_per_second": 18.337,
      "step": 1120
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9214285612106323,
      "eval_loss": 0.40712061524391174,
      "eval_runtime": 7.4748,
      "eval_samples_per_second": 149.837,
      "eval_steps_per_second": 18.73,
      "step": 1400
    },
    {
      "epoch": 5.36,
      "learning_rate": 2.2051618463321703e-05,
      "loss": 0.0871,
      "step": 1500
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9160714149475098,
      "eval_loss": 0.576979398727417,
      "eval_runtime": 7.7299,
      "eval_samples_per_second": 144.892,
      "eval_steps_per_second": 18.112,
      "step": 1680
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9116071462631226,
      "eval_loss": 0.6520564556121826,
      "eval_runtime": 7.614,
      "eval_samples_per_second": 147.098,
      "eval_steps_per_second": 18.387,
      "step": 1960
    },
    {
      "epoch": 7.14,
      "learning_rate": 1.9362396699501985e-05,
      "loss": 0.0437,
      "step": 2000
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9196428656578064,
      "eval_loss": 0.6204714179039001,
      "eval_runtime": 7.8377,
      "eval_samples_per_second": 142.899,
      "eval_steps_per_second": 17.862,
      "step": 2240
    },
    {
      "epoch": 8.93,
      "learning_rate": 1.6673174935682264e-05,
      "loss": 0.036,
      "step": 2500
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.9169642925262451,
      "eval_loss": 0.5968001484870911,
      "eval_runtime": 7.7266,
      "eval_samples_per_second": 144.953,
      "eval_steps_per_second": 18.119,
      "step": 2520
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.9214285612106323,
      "eval_loss": 0.6099338531494141,
      "eval_runtime": 7.5404,
      "eval_samples_per_second": 148.534,
      "eval_steps_per_second": 18.567,
      "step": 2800
    },
    {
      "epoch": 10.71,
      "learning_rate": 1.3983953171862545e-05,
      "loss": 0.0199,
      "step": 3000
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.9169642925262451,
      "eval_loss": 0.6376703381538391,
      "eval_runtime": 7.6099,
      "eval_samples_per_second": 147.177,
      "eval_steps_per_second": 18.397,
      "step": 3080
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.9205357432365417,
      "eval_loss": 0.6757161021232605,
      "eval_runtime": 7.6257,
      "eval_samples_per_second": 146.872,
      "eval_steps_per_second": 18.359,
      "step": 3360
    },
    {
      "epoch": 12.5,
      "learning_rate": 1.1294731408042823e-05,
      "loss": 0.0198,
      "step": 3500
    },
    {
      "epoch": 13.0,
      "eval_accuracy": 0.9205357432365417,
      "eval_loss": 0.6298871636390686,
      "eval_runtime": 7.5908,
      "eval_samples_per_second": 147.548,
      "eval_steps_per_second": 18.443,
      "step": 3640
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.9125000238418579,
      "eval_loss": 0.693723201751709,
      "eval_runtime": 7.7616,
      "eval_samples_per_second": 144.301,
      "eval_steps_per_second": 18.038,
      "step": 3920
    },
    {
      "epoch": 14.29,
      "learning_rate": 8.605509644223102e-06,
      "loss": 0.0119,
      "step": 4000
    },
    {
      "epoch": 15.0,
      "eval_accuracy": 0.9169642925262451,
      "eval_loss": 0.6261208653450012,
      "eval_runtime": 7.7494,
      "eval_samples_per_second": 144.528,
      "eval_steps_per_second": 18.066,
      "step": 4200
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.9160714149475098,
      "eval_loss": 0.7174915671348572,
      "eval_runtime": 7.862,
      "eval_samples_per_second": 142.457,
      "eval_steps_per_second": 17.807,
      "step": 4480
    },
    {
      "epoch": 16.07,
      "learning_rate": 5.916287880403384e-06,
      "loss": 0.0065,
      "step": 4500
    },
    {
      "epoch": 17.0,
      "eval_accuracy": 0.9169642925262451,
      "eval_loss": 0.7007285952568054,
      "eval_runtime": 7.7124,
      "eval_samples_per_second": 145.221,
      "eval_steps_per_second": 18.153,
      "step": 4760
    },
    {
      "epoch": 17.86,
      "learning_rate": 3.2270661165836636e-06,
      "loss": 0.0069,
      "step": 5000
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.918749988079071,
      "eval_loss": 0.7041569352149963,
      "eval_runtime": 7.8136,
      "eval_samples_per_second": 143.339,
      "eval_steps_per_second": 17.917,
      "step": 5040
    },
    {
      "epoch": 19.0,
      "eval_accuracy": 0.9107142686843872,
      "eval_loss": 0.732833981513977,
      "eval_runtime": 7.5262,
      "eval_samples_per_second": 148.814,
      "eval_steps_per_second": 18.602,
      "step": 5320
    },
    {
      "epoch": 19.64,
      "learning_rate": 5.378443527639439e-07,
      "loss": 0.0034,
      "step": 5500
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.9116071462631226,
      "eval_loss": 0.7286208271980286,
      "eval_runtime": 7.6221,
      "eval_samples_per_second": 146.941,
      "eval_steps_per_second": 18.368,
      "step": 5600
    },
    {
      "epoch": 20.0,
      "step": 5600,
      "total_flos": 5893793336524800.0,
      "train_loss": 0.06829224105924368,
      "train_runtime": 4007.6255,
      "train_samples_per_second": 22.357,
      "train_steps_per_second": 1.397
    }
  ],
  "max_steps": 5600,
  "num_train_epochs": 20,
  "total_flos": 5893793336524800.0,
  "trial_name": null,
  "trial_params": null
}