{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "global_step": 9552,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.16,
      "learning_rate": 4.738274706867672e-05,
      "loss": 1.5983,
      "step": 500
    },
    {
      "epoch": 0.31,
      "learning_rate": 4.4765494137353433e-05,
      "loss": 1.0414,
      "step": 1000
    },
    {
      "epoch": 0.47,
      "learning_rate": 4.2148241206030156e-05,
      "loss": 0.8918,
      "step": 1500
    },
    {
      "epoch": 0.63,
      "learning_rate": 3.953098827470687e-05,
      "loss": 0.8029,
      "step": 2000
    },
    {
      "epoch": 0.79,
      "learning_rate": 3.691373534338359e-05,
      "loss": 0.7394,
      "step": 2500
    },
    {
      "epoch": 0.94,
      "learning_rate": 3.42964824120603e-05,
      "loss": 0.7008,
      "step": 3000
    },
    {
      "epoch": 1.1,
      "learning_rate": 3.1679229480737025e-05,
      "loss": 0.6665,
      "step": 3500
    },
    {
      "epoch": 1.26,
      "learning_rate": 2.906197654941374e-05,
      "loss": 0.6283,
      "step": 4000
    },
    {
      "epoch": 1.41,
      "learning_rate": 2.6444723618090456e-05,
      "loss": 0.6072,
      "step": 4500
    },
    {
      "epoch": 1.57,
      "learning_rate": 2.382747068676717e-05,
      "loss": 0.5773,
      "step": 5000
    },
    {
      "epoch": 1.73,
      "learning_rate": 2.1210217755443887e-05,
      "loss": 0.5638,
      "step": 5500
    },
    {
      "epoch": 1.88,
      "learning_rate": 1.8592964824120602e-05,
      "loss": 0.5515,
      "step": 6000
    },
    {
      "epoch": 2.04,
      "learning_rate": 1.597571189279732e-05,
      "loss": 0.529,
      "step": 6500
    },
    {
      "epoch": 2.2,
      "learning_rate": 1.3358458961474038e-05,
      "loss": 0.5156,
      "step": 7000
    },
    {
      "epoch": 2.36,
      "learning_rate": 1.0741206030150754e-05,
      "loss": 0.501,
      "step": 7500
    },
    {
      "epoch": 2.51,
      "learning_rate": 8.123953098827471e-06,
      "loss": 0.4886,
      "step": 8000
    },
    {
      "epoch": 2.67,
      "learning_rate": 5.5067001675041875e-06,
      "loss": 0.4748,
      "step": 8500
    },
    {
      "epoch": 2.83,
      "learning_rate": 2.8894472361809047e-06,
      "loss": 0.4751,
      "step": 9000
    },
    {
      "epoch": 2.98,
      "learning_rate": 2.7219430485762144e-07,
      "loss": 0.4687,
      "step": 9500
    },
    {
      "epoch": 3.0,
      "step": 9552,
      "total_flos": 8.908025376224051e+16,
      "train_loss": 0.6736920107549159,
      "train_runtime": 5995.2753,
      "train_samples_per_second": 15.931,
      "train_steps_per_second": 1.593
    }
  ],
  "max_steps": 9552,
  "num_train_epochs": 3,
  "total_flos": 8.908025376224051e+16,
  "trial_name": null,
  "trial_params": null
}