trainer_state.json

{
  "best_metric": 0.8829333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0644/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": 19.687484741210938,
      "learning_rate": 7e-05,
      "loss": 0.7669,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8162666666666667,
      "eval_loss": 0.6370453238487244,
      "eval_runtime": 12.568,
      "eval_samples_per_second": 298.377,
      "eval_steps_per_second": 4.694,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.6520094871520996,
      "learning_rate": 7e-05,
      "loss": 0.2844,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8621333333333333,
      "eval_loss": 0.4854559302330017,
      "eval_runtime": 13.0055,
      "eval_samples_per_second": 288.339,
      "eval_steps_per_second": 4.537,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.0791161060333252,
      "learning_rate": 7e-05,
      "loss": 0.1769,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8645333333333334,
      "eval_loss": 0.47670337557792664,
      "eval_runtime": 13.4367,
      "eval_samples_per_second": 279.086,
      "eval_steps_per_second": 4.391,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.08308298885822296,
      "learning_rate": 7e-05,
      "loss": 0.1347,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8749333333333333,
      "eval_loss": 0.4772578477859497,
      "eval_runtime": 13.568,
      "eval_samples_per_second": 276.386,
      "eval_steps_per_second": 4.348,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": Infinity,
      "learning_rate": 7e-05,
      "loss": 0.1145,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8829333333333333,
      "eval_loss": 0.4492763578891754,
      "eval_runtime": 12.6475,
      "eval_samples_per_second": 296.501,
      "eval_steps_per_second": 4.665,
      "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
}