trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 6.337254901960784,
  "global_step": 101,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.63,
      "learning_rate": 0.0002,
      "loss": 1.0302,
      "step": 10
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.0002,
      "loss": 0.887,
      "step": 20
    },
    {
      "epoch": 1.88,
      "learning_rate": 0.0002,
      "loss": 0.8287,
      "step": 30
    },
    {
      "epoch": 2.51,
      "learning_rate": 0.0002,
      "loss": 0.7763,
      "step": 40
    },
    {
      "epoch": 3.14,
      "learning_rate": 0.0002,
      "loss": 0.7513,
      "step": 50
    },
    {
      "epoch": 3.76,
      "learning_rate": 0.0002,
      "loss": 0.6717,
      "step": 60
    },
    {
      "epoch": 4.39,
      "learning_rate": 0.0002,
      "loss": 0.6378,
      "step": 70
    },
    {
      "epoch": 5.02,
      "learning_rate": 0.0002,
      "loss": 0.6025,
      "step": 80
    },
    {
      "epoch": 5.65,
      "learning_rate": 0.0002,
      "loss": 0.5327,
      "step": 90
    },
    {
      "epoch": 6.27,
      "learning_rate": 0.0002,
      "loss": 0.4946,
      "step": 100
    },
    {
      "epoch": 6.34,
      "step": 101,
      "total_flos": 4.4403317821771776e+17,
      "train_loss": 0.7186767674318635,
      "train_runtime": 12449.8782,
      "train_samples_per_second": 0.519,
      "train_steps_per_second": 0.008
    }
  ],
  "max_steps": 101,
  "num_train_epochs": 7,
  "total_flos": 4.4403317821771776e+17,
  "trial_name": null,
  "trial_params": null
}