trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 668,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.7986670732498169,
      "learning_rate": 0.00020029985007496252,
      "loss": 0.8294,
      "step": 334
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.6858757138252258,
      "eval_runtime": 18.9055,
      "eval_samples_per_second": 7.881,
      "eval_steps_per_second": 1.005,
      "step": 334
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.891516089439392,
      "learning_rate": 0.0,
      "loss": 0.6325,
      "step": 668
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.6478394865989685,
      "eval_runtime": 18.8168,
      "eval_samples_per_second": 7.918,
      "eval_steps_per_second": 1.01,
      "step": 668
    }
  ],
  "logging_steps": 500,
  "max_steps": 668,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.214694925787136e+16,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}