{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "global_step": 8301,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.18,
      "learning_rate": 2.8644741597397903e-05,
      "loss": 2.0532,
      "step": 500
    },
    {
      "epoch": 0.36,
      "learning_rate": 2.728948319479581e-05,
      "loss": 1.3426,
      "step": 1000
    },
    {
      "epoch": 0.54,
      "learning_rate": 2.5934224792193712e-05,
      "loss": 1.2121,
      "step": 1500
    },
    {
      "epoch": 0.72,
      "learning_rate": 2.4578966389591615e-05,
      "loss": 1.1652,
      "step": 2000
    },
    {
      "epoch": 0.9,
      "learning_rate": 2.3223707986989517e-05,
      "loss": 1.0871,
      "step": 2500
    },
    {
      "epoch": 1.0,
      "eval_exact_match": 79.35666982024598,
      "eval_f1": 87.05079900714725,
      "step": 2767
    },
    {
      "epoch": 1.08,
      "learning_rate": 2.1868449584387424e-05,
      "loss": 0.9397,
      "step": 3000
    },
    {
      "epoch": 1.26,
      "learning_rate": 2.051319118178533e-05,
      "loss": 0.8108,
      "step": 3500
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.9157932779183232e-05,
      "loss": 0.8315,
      "step": 4000
    },
    {
      "epoch": 1.63,
      "learning_rate": 1.7802674376581135e-05,
      "loss": 0.805,
      "step": 4500
    },
    {
      "epoch": 1.81,
      "learning_rate": 1.644741597397904e-05,
      "loss": 0.8106,
      "step": 5000
    },
    {
      "epoch": 1.99,
      "learning_rate": 1.5092157571376944e-05,
      "loss": 0.7997,
      "step": 5500
    },
    {
      "epoch": 2.0,
      "eval_exact_match": 80.13245033112582,
      "eval_f1": 88.1743949836477,
      "step": 5534
    },
    {
      "epoch": 2.17,
      "learning_rate": 1.3736899168774846e-05,
      "loss": 0.5996,
      "step": 6000
    },
    {
      "epoch": 2.35,
      "learning_rate": 1.238164076617275e-05,
      "loss": 0.589,
      "step": 6500
    },
    {
      "epoch": 2.53,
      "learning_rate": 1.1026382363570653e-05,
      "loss": 0.5806,
      "step": 7000
    },
    {
      "epoch": 2.71,
      "learning_rate": 9.671123960968558e-06,
      "loss": 0.5767,
      "step": 7500
    },
    {
      "epoch": 2.89,
      "learning_rate": 8.315865558366462e-06,
      "loss": 0.5806,
      "step": 8000
    },
    {
      "epoch": 3.0,
      "eval_exact_match": 81.07852412488174,
      "eval_f1": 88.40838871409778,
      "step": 8301
    }
  ],
  "max_steps": 11068,
  "num_train_epochs": 4,
  "total_flos": 5.204717742535475e+16,
  "trial_name": null,
  "trial_params": null
}