trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 180.0,
  "global_step": 180,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 10.0,
      "learning_rate": 0.0002,
      "loss": 0.4973,
      "step": 10
    },
    {
      "epoch": 20.0,
      "learning_rate": 0.0002,
      "loss": 0.1772,
      "step": 20
    },
    {
      "epoch": 30.0,
      "learning_rate": 0.0002,
      "loss": 0.0441,
      "step": 30
    },
    {
      "epoch": 40.0,
      "learning_rate": 0.0002,
      "loss": 0.0221,
      "step": 40
    },
    {
      "epoch": 50.0,
      "learning_rate": 0.0002,
      "loss": 0.0174,
      "step": 50
    },
    {
      "epoch": 60.0,
      "learning_rate": 0.0002,
      "loss": 0.0132,
      "step": 60
    },
    {
      "epoch": 70.0,
      "learning_rate": 0.0002,
      "loss": 0.0088,
      "step": 70
    },
    {
      "epoch": 80.0,
      "learning_rate": 0.0002,
      "loss": 0.0071,
      "step": 80
    },
    {
      "epoch": 90.0,
      "learning_rate": 0.0002,
      "loss": 0.0061,
      "step": 90
    },
    {
      "epoch": 100.0,
      "learning_rate": 0.0002,
      "loss": 0.0056,
      "step": 100
    },
    {
      "epoch": 110.0,
      "learning_rate": 0.0002,
      "loss": 0.0055,
      "step": 110
    },
    {
      "epoch": 120.0,
      "learning_rate": 0.0002,
      "loss": 0.0055,
      "step": 120
    },
    {
      "epoch": 130.0,
      "learning_rate": 0.0002,
      "loss": 0.0055,
      "step": 130
    },
    {
      "epoch": 140.0,
      "learning_rate": 0.0002,
      "loss": 0.0055,
      "step": 140
    },
    {
      "epoch": 150.0,
      "learning_rate": 0.0002,
      "loss": 0.0055,
      "step": 150
    },
    {
      "epoch": 160.0,
      "learning_rate": 0.0002,
      "loss": 0.0055,
      "step": 160
    },
    {
      "epoch": 170.0,
      "learning_rate": 0.0002,
      "loss": 0.0055,
      "step": 170
    },
    {
      "epoch": 180.0,
      "learning_rate": 0.0002,
      "loss": 0.0055,
      "step": 180
    }
  ],
  "max_steps": 500,
  "num_train_epochs": 500,
  "total_flos": 3.046874617479168e+16,
  "trial_name": null,
  "trial_params": null
}