trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9259259259259259,
  "global_step": 200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.12,
      "learning_rate": 0.0002,
      "loss": 1.9564,
      "step": 25
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0002,
      "loss": 0.4866,
      "step": 50
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0002,
      "loss": 0.4995,
      "step": 75
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0002,
      "loss": 0.3118,
      "step": 100
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0002,
      "loss": 0.4503,
      "step": 125
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.0002,
      "loss": 0.2724,
      "step": 150
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.0002,
      "loss": 0.4099,
      "step": 175
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.0002,
      "loss": 0.2531,
      "step": 200
    }
  ],
  "max_steps": 216,
  "num_train_epochs": 1,
  "total_flos": 7233885092118528.0,
  "trial_name": null,
  "trial_params": null
}