trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.17107309486780714,
  "eval_steps": 500,
  "global_step": 110,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03,
      "learning_rate": 5e-05,
      "loss": 2.2096,
      "step": 20
    },
    {
      "epoch": 0.06,
      "learning_rate": 5e-05,
      "loss": 1.8708,
      "step": 40
    },
    {
      "epoch": 0.09,
      "learning_rate": 5e-05,
      "loss": 1.8354,
      "step": 60
    },
    {
      "epoch": 0.12,
      "learning_rate": 5e-05,
      "loss": 1.7561,
      "step": 80
    },
    {
      "epoch": 0.16,
      "learning_rate": 5e-05,
      "loss": 1.6541,
      "step": 100
    }
  ],
  "logging_steps": 20,
  "max_steps": 400,
  "num_train_epochs": 1,
  "save_steps": 10,
  "total_flos": 3439132066529280.0,
  "trial_name": null,
  "trial_params": null
}