trainer_state.json

{
  "best_metric": 0.8738666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0208/checkpoints/checkpoint-1998",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 1998,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": Infinity,
      "learning_rate": 3e-05,
      "loss": 2.354,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7285333333333334,
      "eval_loss": 1.2054487466812134,
      "eval_runtime": 4.5475,
      "eval_samples_per_second": 824.632,
      "eval_steps_per_second": 12.974,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 5.737593650817871,
      "learning_rate": 3e-05,
      "loss": 0.8286,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8445333333333334,
      "eval_loss": 0.6268100142478943,
      "eval_runtime": 4.4505,
      "eval_samples_per_second": 842.603,
      "eval_steps_per_second": 13.257,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 14.035255432128906,
      "learning_rate": 3e-05,
      "loss": 0.4807,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8578666666666667,
      "eval_loss": 0.49961724877357483,
      "eval_runtime": 4.6053,
      "eval_samples_per_second": 814.272,
      "eval_steps_per_second": 12.811,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 5.713260650634766,
      "learning_rate": 3e-05,
      "loss": 0.3255,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8570666666666666,
      "eval_loss": 0.48741331696510315,
      "eval_runtime": 4.4151,
      "eval_samples_per_second": 849.35,
      "eval_steps_per_second": 13.363,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.6296706795692444,
      "learning_rate": 3e-05,
      "loss": 0.2385,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8688,
      "eval_loss": 0.45458170771598816,
      "eval_runtime": 4.4873,
      "eval_samples_per_second": 835.683,
      "eval_steps_per_second": 13.148,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 2.1861374378204346,
      "learning_rate": 3e-05,
      "loss": 0.1767,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8738666666666667,
      "eval_loss": 0.4372554123401642,
      "eval_runtime": 4.4378,
      "eval_samples_per_second": 845.02,
      "eval_steps_per_second": 13.295,
      "step": 1998
    }
  ],
  "logging_steps": 500,
  "max_steps": 1998,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "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": 9.88447923735552e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}