trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 3,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.3333333333333333,
      "grad_norm": 104.34296088799998,
      "learning_rate": 1e-05,
      "loss": 1.0512,
      "step": 1
    },
    {
      "epoch": 0.6666666666666666,
      "grad_norm": 98.47481125967262,
      "learning_rate": 5e-06,
      "loss": 1.0267,
      "step": 2
    },
    {
      "epoch": 1.0,
      "grad_norm": 7.664500232897243,
      "learning_rate": 0.0,
      "loss": 0.3927,
      "step": 3
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.4052319824695587,
      "eval_runtime": 1.4338,
      "eval_samples_per_second": 2.092,
      "eval_steps_per_second": 0.697,
      "step": 3
    },
    {
      "epoch": 1.0,
      "step": 3,
      "total_flos": 1256277934080.0,
      "train_loss": 0.8235548933347067,
      "train_runtime": 340.6476,
      "train_samples_per_second": 0.208,
      "train_steps_per_second": 0.009
    }
  ],
  "logging_steps": 1,
  "max_steps": 3,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "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": 1256277934080.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}