trainer_state.json

{
  "best_metric": 0.8826666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0854/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": 54.33023452758789,
      "learning_rate": 5e-05,
      "loss": 0.7808,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8549333333333333,
      "eval_loss": 0.4628535211086273,
      "eval_runtime": 10.0206,
      "eval_samples_per_second": 374.231,
      "eval_steps_per_second": 5.888,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 5e-05,
      "loss": 0.2223,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8733333333333333,
      "eval_loss": 0.40670332312583923,
      "eval_runtime": 10.1811,
      "eval_samples_per_second": 368.33,
      "eval_steps_per_second": 5.795,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.001318420167081058,
      "learning_rate": 5e-05,
      "loss": 0.1149,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8773333333333333,
      "eval_loss": 0.439779132604599,
      "eval_runtime": 10.0229,
      "eval_samples_per_second": 374.145,
      "eval_steps_per_second": 5.887,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.018561743199825287,
      "learning_rate": 5e-05,
      "loss": 0.0824,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8818666666666667,
      "eval_loss": 0.44915181398391724,
      "eval_runtime": 10.1802,
      "eval_samples_per_second": 368.361,
      "eval_steps_per_second": 5.796,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.07076983153820038,
      "learning_rate": 5e-05,
      "loss": 0.0586,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8826666666666667,
      "eval_loss": 0.5294551253318787,
      "eval_runtime": 10.5798,
      "eval_samples_per_second": 354.448,
      "eval_steps_per_second": 5.577,
      "step": 1665
    }
  ],
  "logging_steps": 500,
  "max_steps": 2331,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "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": 8.2370660311296e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}