trainer_state.json

{
  "best_metric": 0.9248,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0368/checkpoints/checkpoint-2664",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 2664,
  "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": 6.7914265912926e-05,
      "loss": 0.7765,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8688,
      "eval_loss": 0.4552498459815979,
      "eval_runtime": 12.554,
      "eval_samples_per_second": 298.709,
      "eval_steps_per_second": 4.7,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 8.733572006225586,
      "learning_rate": 6.185866251124284e-05,
      "loss": 0.2545,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8810666666666667,
      "eval_loss": 0.3962556719779968,
      "eval_runtime": 12.5766,
      "eval_samples_per_second": 298.172,
      "eval_steps_per_second": 4.691,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 12.276375770568848,
      "learning_rate": 5.2563508019062023e-05,
      "loss": 0.1329,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8944,
      "eval_loss": 0.381111204624176,
      "eval_runtime": 12.6382,
      "eval_samples_per_second": 296.721,
      "eval_steps_per_second": 4.668,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.002898092847317457,
      "learning_rate": 4.1149935250010576e-05,
      "loss": 0.0703,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8968,
      "eval_loss": 0.42370837926864624,
      "eval_runtime": 12.5177,
      "eval_samples_per_second": 299.577,
      "eval_steps_per_second": 4.713,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.00017890050366986543,
      "learning_rate": 2.8994589526430138e-05,
      "loss": 0.0323,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9088,
      "eval_loss": 0.37604784965515137,
      "eval_runtime": 12.5997,
      "eval_samples_per_second": 297.627,
      "eval_steps_per_second": 4.683,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.0012296010972931981,
      "learning_rate": 1.756358493638142e-05,
      "loss": 0.0116,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.92,
      "eval_loss": 0.3510739803314209,
      "eval_runtime": 12.9846,
      "eval_samples_per_second": 288.804,
      "eval_steps_per_second": 4.544,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.005171077325940132,
      "learning_rate": 8.235669337084578e-06,
      "loss": 0.0026,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9218666666666666,
      "eval_loss": 0.33891016244888306,
      "eval_runtime": 13.1083,
      "eval_samples_per_second": 286.079,
      "eval_steps_per_second": 4.501,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.0006612282013520598,
      "learning_rate": 2.135927015187301e-06,
      "loss": 0.0007,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9248,
      "eval_loss": 0.33268091082572937,
      "eval_runtime": 12.5743,
      "eval_samples_per_second": 298.227,
      "eval_steps_per_second": 4.692,
      "step": 2664
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.317930564980736e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}