trainer_state.json

{
  "best_metric": 0.8264,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0570/checkpoints/checkpoint-2997",
  "epoch": 9.0,
  "eval_steps": 500,
  "global_step": 2997,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 143.32086181640625,
      "learning_rate": 0.0001,
      "loss": 0.9516,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8037333333333333,
      "eval_loss": 0.6087485551834106,
      "eval_runtime": 10.6351,
      "eval_samples_per_second": 352.607,
      "eval_steps_per_second": 5.548,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0001,
      "loss": 0.3541,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8232,
      "eval_loss": 0.5680658221244812,
      "eval_runtime": 10.5787,
      "eval_samples_per_second": 354.487,
      "eval_steps_per_second": 5.577,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.04232461750507355,
      "learning_rate": 0.0001,
      "loss": 0.2198,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8,
      "eval_loss": 0.7273663878440857,
      "eval_runtime": 10.4871,
      "eval_samples_per_second": 357.583,
      "eval_steps_per_second": 5.626,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 4.487899303436279,
      "learning_rate": 0.0001,
      "loss": 0.174,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8218666666666666,
      "eval_loss": 0.6846040487289429,
      "eval_runtime": 10.6034,
      "eval_samples_per_second": 353.661,
      "eval_steps_per_second": 5.564,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.075191020965576,
      "learning_rate": 0.0001,
      "loss": 0.1394,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8154666666666667,
      "eval_loss": 0.7771618962287903,
      "eval_runtime": 10.3514,
      "eval_samples_per_second": 362.27,
      "eval_steps_per_second": 5.7,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 9.628615379333496,
      "learning_rate": 0.0001,
      "loss": 0.1197,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8237333333333333,
      "eval_loss": 0.702775776386261,
      "eval_runtime": 10.566,
      "eval_samples_per_second": 354.912,
      "eval_steps_per_second": 5.584,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.00021399883553385735,
      "learning_rate": 0.0001,
      "loss": 0.109,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8138666666666666,
      "eval_loss": 0.803549587726593,
      "eval_runtime": 10.633,
      "eval_samples_per_second": 352.674,
      "eval_steps_per_second": 5.549,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.6938982605934143,
      "learning_rate": 0.0001,
      "loss": 0.0994,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8045333333333333,
      "eval_loss": 0.8593702912330627,
      "eval_runtime": 10.4167,
      "eval_samples_per_second": 360.0,
      "eval_steps_per_second": 5.664,
      "step": 2664
    },
    {
      "epoch": 9.0,
      "grad_norm": 0.18702276051044464,
      "learning_rate": 0.0001,
      "loss": 0.0887,
      "step": 2997
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.8264,
      "eval_loss": 0.7629392147064209,
      "eval_runtime": 10.0991,
      "eval_samples_per_second": 371.319,
      "eval_steps_per_second": 5.842,
      "step": 2997
    }
  ],
  "logging_steps": 500,
  "max_steps": 2997,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "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": 1.482671885603328e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}