trainer_state.json

{
  "best_metric": 0.9208,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0210/checkpoints/checkpoint-2664",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 2664,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 7.526144504547119,
      "learning_rate": 8.659485808097889e-05,
      "loss": 0.8821,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8181333333333334,
      "eval_loss": 0.5843281745910645,
      "eval_runtime": 6.3279,
      "eval_samples_per_second": 592.616,
      "eval_steps_per_second": 9.324,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 27.355009078979492,
      "learning_rate": 7.689476527896773e-05,
      "loss": 0.3014,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8512,
      "eval_loss": 0.5184677839279175,
      "eval_runtime": 6.6255,
      "eval_samples_per_second": 565.992,
      "eval_steps_per_second": 8.905,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 3.8983495235443115,
      "learning_rate": 6.231876231770601e-05,
      "loss": 0.146,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8818666666666667,
      "eval_loss": 0.42638665437698364,
      "eval_runtime": 6.227,
      "eval_samples_per_second": 602.215,
      "eval_steps_per_second": 9.475,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.08753003180027008,
      "learning_rate": 4.510613478854492e-05,
      "loss": 0.0705,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8896,
      "eval_loss": 0.42884334921836853,
      "eval_runtime": 7.2673,
      "eval_samples_per_second": 516.011,
      "eval_steps_per_second": 8.119,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.0006916002603247762,
      "learning_rate": 2.7877349199942123e-05,
      "loss": 0.0263,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9018666666666667,
      "eval_loss": 0.4086630940437317,
      "eval_runtime": 6.5703,
      "eval_samples_per_second": 570.753,
      "eval_steps_per_second": 8.98,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.02760840766131878,
      "learning_rate": 1.3255331978432106e-05,
      "loss": 0.0076,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9152,
      "eval_loss": 0.3757134675979614,
      "eval_runtime": 6.1843,
      "eval_samples_per_second": 606.37,
      "eval_steps_per_second": 9.54,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.00023615392274223268,
      "learning_rate": 3.4661526970736e-06,
      "loss": 0.0013,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9205333333333333,
      "eval_loss": 0.33984512090682983,
      "eval_runtime": 6.245,
      "eval_samples_per_second": 600.482,
      "eval_steps_per_second": 9.448,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.0002217637374997139,
      "learning_rate": 1.2516232227643975e-10,
      "loss": 0.0004,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9208,
      "eval_loss": 0.3396505117416382,
      "eval_runtime": 6.178,
      "eval_samples_per_second": 606.992,
      "eval_steps_per_second": 9.55,
      "step": 2664
    }
  ],
  "logging_steps": 500,
  "max_steps": 2664,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "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": 1.317930564980736e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}