trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "global_step": 188,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 0.0005,
      "loss": 5.9448,
      "step": 94
    },
    {
      "epoch": 1.0,
      "eval_loss": 4.646627902984619,
      "eval_runtime": 70.2506,
      "eval_samples_per_second": 2.847,
      "eval_steps_per_second": 0.356,
      "step": 94
    },
    {
      "epoch": 2.0,
      "learning_rate": 0.0,
      "loss": 4.6291,
      "step": 188
    },
    {
      "epoch": 2.0,
      "eval_loss": 4.621755123138428,
      "eval_runtime": 70.6527,
      "eval_samples_per_second": 2.831,
      "eval_steps_per_second": 0.354,
      "step": 188
    },
    {
      "epoch": 2.0,
      "step": 188,
      "total_flos": 6.168280896e+17,
      "train_loss": 5.2869415283203125,
      "train_runtime": 1145.3249,
      "train_samples_per_second": 1.31,
      "train_steps_per_second": 0.164
    }
  ],
  "max_steps": 188,
  "num_train_epochs": 2,
  "total_flos": 6.168280896e+17,
  "trial_name": null,
  "trial_params": null
}