trainer_state.json

{
  "best_metric": 0.9088,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0078/checkpoints/checkpoint-1665",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 1665,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 2.2568795680999756,
      "learning_rate": 4.525311529382417e-05,
      "loss": 0.7844,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8562666666666666,
      "eval_loss": 0.4407110810279846,
      "eval_runtime": 5.3109,
      "eval_samples_per_second": 706.089,
      "eval_steps_per_second": 11.109,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 6.8746562004089355,
      "learning_rate": 3.277027342564428e-05,
      "loss": 0.2018,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8776,
      "eval_loss": 0.38891273736953735,
      "eval_runtime": 5.6839,
      "eval_samples_per_second": 659.76,
      "eval_steps_per_second": 10.38,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.0048262388445436954,
      "learning_rate": 1.731945121074836e-05,
      "loss": 0.0692,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8893333333333333,
      "eval_loss": 0.3749508559703827,
      "eval_runtime": 5.5723,
      "eval_samples_per_second": 672.972,
      "eval_steps_per_second": 10.588,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.006027453113347292,
      "learning_rate": 4.802337581094707e-06,
      "loss": 0.0124,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9066666666666666,
      "eval_loss": 0.33745884895324707,
      "eval_runtime": 5.2688,
      "eval_samples_per_second": 711.744,
      "eval_steps_per_second": 11.198,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.008978858590126038,
      "learning_rate": 4.4502166460236305e-11,
      "loss": 0.0021,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9088,
      "eval_loss": 0.33641311526298523,
      "eval_runtime": 5.7308,
      "eval_samples_per_second": 654.354,
      "eval_steps_per_second": 10.295,
      "step": 1665
    }
  ],
  "logging_steps": 500,
  "max_steps": 1665,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 8.2370660311296e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}