trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "global_step": 9820,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.1,
      "learning_rate": 4.9693251533742335e-05,
      "loss": 19748.602,
      "step": 500
    },
    {
      "epoch": 0.2,
      "learning_rate": 4.918200408997955e-05,
      "loss": 3106.5375,
      "step": 1000
    },
    {
      "epoch": 0.31,
      "learning_rate": 4.867075664621677e-05,
      "loss": 791.6922,
      "step": 1500
    },
    {
      "epoch": 0.41,
      "learning_rate": 4.815950920245399e-05,
      "loss": 409.7595,
      "step": 2000
    },
    {
      "epoch": 0.51,
      "learning_rate": 4.764826175869121e-05,
      "loss": 253.3151,
      "step": 2500
    },
    {
      "epoch": 0.61,
      "learning_rate": 4.7137014314928426e-05,
      "loss": 189.49,
      "step": 3000
    },
    {
      "epoch": 0.71,
      "learning_rate": 4.6625766871165645e-05,
      "loss": 148.3558,
      "step": 3500
    },
    {
      "epoch": 0.81,
      "learning_rate": 4.611451942740286e-05,
      "loss": 126.5535,
      "step": 4000
    },
    {
      "epoch": 0.92,
      "learning_rate": 4.560327198364008e-05,
      "loss": 114.7683,
      "step": 4500
    },
    {
      "epoch": 1.02,
      "learning_rate": 4.5092024539877307e-05,
      "loss": 100.4415,
      "step": 5000
    },
    {
      "epoch": 1.12,
      "learning_rate": 4.4580777096114525e-05,
      "loss": 84.655,
      "step": 5500
    },
    {
      "epoch": 1.22,
      "learning_rate": 4.4069529652351736e-05,
      "loss": 79.8442,
      "step": 6000
    },
    {
      "epoch": 1.32,
      "learning_rate": 4.3558282208588955e-05,
      "loss": 81.4306,
      "step": 6500
    },
    {
      "epoch": 1.43,
      "learning_rate": 4.304703476482618e-05,
      "loss": 74.3729,
      "step": 7000
    },
    {
      "epoch": 1.53,
      "learning_rate": 4.25357873210634e-05,
      "loss": 73.7614,
      "step": 7500
    },
    {
      "epoch": 1.63,
      "learning_rate": 4.2024539877300617e-05,
      "loss": 73.2727,
      "step": 8000
    },
    {
      "epoch": 1.73,
      "learning_rate": 4.1513292433537835e-05,
      "loss": 68.4054,
      "step": 8500
    },
    {
      "epoch": 1.83,
      "learning_rate": 4.100204498977505e-05,
      "loss": 69.6723,
      "step": 9000
    },
    {
      "epoch": 1.93,
      "learning_rate": 4.049079754601227e-05,
      "loss": 67.092,
      "step": 9500
    }
  ],
  "max_steps": 49100,
  "num_train_epochs": 10,
  "total_flos": 6.617906200146989e+16,
  "trial_name": null,
  "trial_params": null
}