trainer_state.json

{
  "best_metric": 0.3565333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0851/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": Infinity,
      "learning_rate": 0.0004533575357652462,
      "loss": 3.4202,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.19013333333333332,
      "eval_loss": 3.076448917388916,
      "eval_runtime": 11.7029,
      "eval_samples_per_second": 320.433,
      "eval_steps_per_second": 5.041,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 25.75981903076172,
      "learning_rate": 0.00032904652505153707,
      "loss": 2.9136,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.24613333333333334,
      "eval_loss": 2.8488078117370605,
      "eval_runtime": 11.8718,
      "eval_samples_per_second": 315.874,
      "eval_steps_per_second": 4.97,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 26.89653205871582,
      "learning_rate": 0.00017454242846071084,
      "loss": 2.6459,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.30373333333333336,
      "eval_loss": 2.584139108657837,
      "eval_runtime": 11.8576,
      "eval_samples_per_second": 316.253,
      "eval_steps_per_second": 4.976,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 29.357566833496094,
      "learning_rate": 4.886055948936635e-05,
      "loss": 2.3478,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.3453333333333333,
      "eval_loss": 2.412684917449951,
      "eval_runtime": 11.9317,
      "eval_samples_per_second": 314.288,
      "eval_steps_per_second": 4.945,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 41.464515686035156,
      "learning_rate": 7.12031494640164e-09,
      "loss": 2.0706,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.3565333333333333,
      "eval_loss": 2.3905811309814453,
      "eval_runtime": 12.1362,
      "eval_samples_per_second": 308.993,
      "eval_steps_per_second": 4.861,
      "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
}