trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 96.0,
  "global_step": 120,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 8.0,
      "learning_rate": 0.0002,
      "loss": 3.2324,
      "step": 10
    },
    {
      "epoch": 16.0,
      "learning_rate": 0.0002,
      "loss": 1.4712,
      "step": 20
    },
    {
      "epoch": 24.0,
      "learning_rate": 0.0002,
      "loss": 0.3733,
      "step": 30
    },
    {
      "epoch": 32.0,
      "learning_rate": 0.0002,
      "loss": 0.0691,
      "step": 40
    },
    {
      "epoch": 40.0,
      "learning_rate": 0.0002,
      "loss": 0.0563,
      "step": 50
    },
    {
      "epoch": 48.0,
      "learning_rate": 0.0002,
      "loss": 0.0523,
      "step": 60
    },
    {
      "epoch": 56.0,
      "learning_rate": 0.0002,
      "loss": 0.0511,
      "step": 70
    },
    {
      "epoch": 64.0,
      "learning_rate": 0.0002,
      "loss": 0.0516,
      "step": 80
    },
    {
      "epoch": 72.0,
      "learning_rate": 0.0002,
      "loss": 0.0518,
      "step": 90
    },
    {
      "epoch": 80.0,
      "learning_rate": 0.0002,
      "loss": 0.0504,
      "step": 100
    },
    {
      "epoch": 88.0,
      "learning_rate": 0.0002,
      "loss": 0.0503,
      "step": 110
    },
    {
      "epoch": 96.0,
      "learning_rate": 0.0002,
      "loss": 0.0508,
      "step": 120
    }
  ],
  "max_steps": 500,
  "num_train_epochs": 500,
  "total_flos": 4188956688451584.0,
  "trial_name": null,
  "trial_params": null
}