trainer_state.json

{
  "best_metric": 0.8938666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0683/checkpoints/checkpoint-1998",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 1998,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 7.16422176361084,
      "learning_rate": 5e-05,
      "loss": 0.7441,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.88,
      "eval_loss": 0.37753036618232727,
      "eval_runtime": 10.322,
      "eval_samples_per_second": 363.302,
      "eval_steps_per_second": 5.716,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.990835189819336,
      "learning_rate": 5e-05,
      "loss": 0.1879,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8736,
      "eval_loss": 0.43866026401519775,
      "eval_runtime": 10.6422,
      "eval_samples_per_second": 352.37,
      "eval_steps_per_second": 5.544,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.5260799527168274,
      "learning_rate": 5e-05,
      "loss": 0.1091,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8882666666666666,
      "eval_loss": 0.4079267680644989,
      "eval_runtime": 10.4093,
      "eval_samples_per_second": 360.254,
      "eval_steps_per_second": 5.668,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.2471596002578735,
      "learning_rate": 5e-05,
      "loss": 0.076,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8890666666666667,
      "eval_loss": 0.40939465165138245,
      "eval_runtime": 10.2466,
      "eval_samples_per_second": 365.976,
      "eval_steps_per_second": 5.758,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.005934775806963444,
      "learning_rate": 5e-05,
      "loss": 0.0616,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8810666666666667,
      "eval_loss": 0.48430871963500977,
      "eval_runtime": 10.2621,
      "eval_samples_per_second": 365.424,
      "eval_steps_per_second": 5.749,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.5435875058174133,
      "learning_rate": 5e-05,
      "loss": 0.0528,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8938666666666667,
      "eval_loss": 0.4448767602443695,
      "eval_runtime": 10.3973,
      "eval_samples_per_second": 360.672,
      "eval_steps_per_second": 5.675,
      "step": 1998
    }
  ],
  "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": 9.88447923735552e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}