{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "global_step": 8400,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.18,
      "learning_rate": 2.8214285714285714e-05,
      "loss": 2.1218,
      "step": 500
    },
    {
      "epoch": 0.36,
      "learning_rate": 2.6428571428571428e-05,
      "loss": 1.254,
      "step": 1000
    },
    {
      "epoch": 0.54,
      "learning_rate": 2.464285714285714e-05,
      "loss": 1.1716,
      "step": 1500
    },
    {
      "epoch": 0.71,
      "learning_rate": 2.2857142857142858e-05,
      "loss": 1.0937,
      "step": 2000
    },
    {
      "epoch": 0.89,
      "learning_rate": 2.107142857142857e-05,
      "loss": 1.0459,
      "step": 2500
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.928571428571429e-05,
      "loss": 0.9711,
      "step": 3000
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.7500000000000002e-05,
      "loss": 0.8414,
      "step": 3500
    },
    {
      "epoch": 1.43,
      "learning_rate": 1.5714285714285715e-05,
      "loss": 0.8165,
      "step": 4000
    },
    {
      "epoch": 1.61,
      "learning_rate": 1.3928571428571429e-05,
      "loss": 0.8204,
      "step": 4500
    },
    {
      "epoch": 1.79,
      "learning_rate": 1.2142857142857144e-05,
      "loss": 0.8327,
      "step": 5000
    },
    {
      "epoch": 1.96,
      "learning_rate": 1.0357142857142857e-05,
      "loss": 0.8118,
      "step": 5500
    },
    {
      "epoch": 2.14,
      "learning_rate": 8.571428571428571e-06,
      "loss": 0.6869,
      "step": 6000
    },
    {
      "epoch": 2.32,
      "learning_rate": 6.785714285714286e-06,
      "loss": 0.6618,
      "step": 6500
    },
    {
      "epoch": 2.5,
      "learning_rate": 4.9999999999999996e-06,
      "loss": 0.6381,
      "step": 7000
    },
    {
      "epoch": 2.68,
      "learning_rate": 3.2142857142857143e-06,
      "loss": 0.6532,
      "step": 7500
    },
    {
      "epoch": 2.86,
      "learning_rate": 1.4285714285714286e-06,
      "loss": 0.6188,
      "step": 8000
    },
    {
      "epoch": 3.0,
      "step": 8400,
      "total_flos": 5.267566240486963e+16,
      "train_loss": 0.9259976813906715,
      "train_runtime": 7183.2279,
      "train_samples_per_second": 37.419,
      "train_steps_per_second": 1.169
    }
  ],
  "max_steps": 8400,
  "num_train_epochs": 3,
  "total_flos": 5.267566240486963e+16,
  "trial_name": null,
  "trial_params": null
}