trainer_state.json

{
  "best_metric": 0.9210666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0139/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": 8.661507935327383e-05,
      "loss": 0.8801,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8192,
      "eval_loss": 0.5924332141876221,
      "eval_runtime": 6.1726,
      "eval_samples_per_second": 607.528,
      "eval_steps_per_second": 9.558,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 62.28591537475586,
      "learning_rate": 7.689476527896773e-05,
      "loss": 0.3177,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8472,
      "eval_loss": 0.5134756565093994,
      "eval_runtime": 6.2222,
      "eval_samples_per_second": 602.685,
      "eval_steps_per_second": 9.482,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.0022745367605239153,
      "learning_rate": 6.231876231770601e-05,
      "loss": 0.1697,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8693333333333333,
      "eval_loss": 0.44837644696235657,
      "eval_runtime": 6.1119,
      "eval_samples_per_second": 613.557,
      "eval_steps_per_second": 9.653,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.015315541997551918,
      "learning_rate": 4.510613478854492e-05,
      "loss": 0.0816,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8818666666666667,
      "eval_loss": 0.4373076558113098,
      "eval_runtime": 6.6792,
      "eval_samples_per_second": 561.445,
      "eval_steps_per_second": 8.833,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.0015707158017903566,
      "learning_rate": 2.7877349199942123e-05,
      "loss": 0.0351,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9016,
      "eval_loss": 0.39958763122558594,
      "eval_runtime": 6.1263,
      "eval_samples_per_second": 612.11,
      "eval_steps_per_second": 9.631,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.00036011863267049193,
      "learning_rate": 1.3255331978432106e-05,
      "loss": 0.0078,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9112,
      "eval_loss": 0.3591609299182892,
      "eval_runtime": 6.1049,
      "eval_samples_per_second": 614.261,
      "eval_steps_per_second": 9.664,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 20.443323135375977,
      "learning_rate": 3.4661526970736e-06,
      "loss": 0.0012,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9165333333333333,
      "eval_loss": 0.3426171541213989,
      "eval_runtime": 6.299,
      "eval_samples_per_second": 595.33,
      "eval_steps_per_second": 9.367,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.0006112630944699049,
      "learning_rate": 1.2516232227643975e-10,
      "loss": 0.0007,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9210666666666667,
      "eval_loss": 0.3279791474342346,
      "eval_runtime": 7.7018,
      "eval_samples_per_second": 486.896,
      "eval_steps_per_second": 7.66,
      "step": 2664
    }
  ],
  "logging_steps": 500,
  "max_steps": 2664,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "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": 1.317930564980736e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}