trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 10.0,
  "global_step": 3980,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.26,
      "learning_rate": 4.3718592964824124e-05,
      "loss": 2.2285,
      "step": 500
    },
    {
      "epoch": 2.51,
      "learning_rate": 3.7437185929648245e-05,
      "loss": 0.9853,
      "step": 1000
    },
    {
      "epoch": 3.77,
      "learning_rate": 3.1155778894472366e-05,
      "loss": 0.854,
      "step": 1500
    },
    {
      "epoch": 5.03,
      "learning_rate": 2.4874371859296484e-05,
      "loss": 0.8085,
      "step": 2000
    },
    {
      "epoch": 6.28,
      "learning_rate": 1.8592964824120602e-05,
      "loss": 0.768,
      "step": 2500
    },
    {
      "epoch": 7.54,
      "learning_rate": 1.2311557788944725e-05,
      "loss": 0.7459,
      "step": 3000
    },
    {
      "epoch": 8.79,
      "learning_rate": 6.030150753768844e-06,
      "loss": 0.7298,
      "step": 3500
    },
    {
      "epoch": 10.0,
      "step": 3980,
      "total_flos": 2215243269734400.0,
      "train_loss": 0.9812813993674426,
      "train_runtime": 514.4973,
      "train_samples_per_second": 123.499,
      "train_steps_per_second": 7.736
    }
  ],
  "max_steps": 3980,
  "num_train_epochs": 10,
  "total_flos": 2215243269734400.0,
  "trial_name": null,
  "trial_params": null
}