run-1/checkpoint-10/trainer_state.json

{
  "best_metric": 0.125,
  "best_model_checkpoint": "model/image_classification/run-1/checkpoint-10",
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 10,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.9903935194015503,
      "learning_rate": 5.668350980311898e-06,
      "loss": 2.0775,
      "step": 10
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.125,
      "eval_loss": 2.0746703147888184,
      "eval_runtime": 0.7378,
      "eval_samples_per_second": 216.85,
      "eval_steps_per_second": 13.553,
      "step": 10
    }
  ],
  "logging_steps": 10,
  "max_steps": 30,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4.959754037231616e+16,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 7.652273823421063e-06,
    "per_device_train_batch_size": 32
  }
}