trainer_state.json

{
  "best_metric": 0.3370666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0376/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": 20.61095428466797,
      "learning_rate": 0.0005,
      "loss": 3.4731,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.17733333333333334,
      "eval_loss": 3.124823808670044,
      "eval_runtime": 6.1928,
      "eval_samples_per_second": 605.544,
      "eval_steps_per_second": 9.527,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 18.951875686645508,
      "learning_rate": 0.0005,
      "loss": 3.0222,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.20826666666666666,
      "eval_loss": 2.9572927951812744,
      "eval_runtime": 5.7812,
      "eval_samples_per_second": 648.652,
      "eval_steps_per_second": 10.205,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 19.094789505004883,
      "learning_rate": 0.0005,
      "loss": 2.8804,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.22986666666666666,
      "eval_loss": 2.842059373855591,
      "eval_runtime": 5.7848,
      "eval_samples_per_second": 648.246,
      "eval_steps_per_second": 10.199,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 16.52971076965332,
      "learning_rate": 0.0005,
      "loss": 2.7598,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.26106666666666667,
      "eval_loss": 2.7495648860931396,
      "eval_runtime": 5.9383,
      "eval_samples_per_second": 631.493,
      "eval_steps_per_second": 9.935,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 14.526013374328613,
      "learning_rate": 0.0005,
      "loss": 2.6508,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.2992,
      "eval_loss": 2.57963490486145,
      "eval_runtime": 5.7149,
      "eval_samples_per_second": 656.178,
      "eval_steps_per_second": 10.324,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 15.393631935119629,
      "learning_rate": 0.0005,
      "loss": 2.5638,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.31146666666666667,
      "eval_loss": 2.4839441776275635,
      "eval_runtime": 5.7086,
      "eval_samples_per_second": 656.906,
      "eval_steps_per_second": 10.335,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 16.094772338867188,
      "learning_rate": 0.0005,
      "loss": 2.4723,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.31493333333333334,
      "eval_loss": 2.5068202018737793,
      "eval_runtime": 5.9509,
      "eval_samples_per_second": 630.159,
      "eval_steps_per_second": 9.915,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 16.462902069091797,
      "learning_rate": 0.0005,
      "loss": 2.3795,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.3370666666666667,
      "eval_loss": 2.398714542388916,
      "eval_runtime": 6.7794,
      "eval_samples_per_second": 553.147,
      "eval_steps_per_second": 8.703,
      "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
}