trainer_state.json

{
  "best_metric": 0.25333333333333335,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0342/checkpoints/checkpoint-1665",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 1665,
  "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": 0.0005,
      "loss": 3.4852,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.1792,
      "eval_loss": 3.1245391368865967,
      "eval_runtime": 8.2707,
      "eval_samples_per_second": 453.406,
      "eval_steps_per_second": 7.134,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 20.937910079956055,
      "learning_rate": 0.0005,
      "loss": 3.0451,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.1832,
      "eval_loss": 3.036055326461792,
      "eval_runtime": 11.4795,
      "eval_samples_per_second": 326.671,
      "eval_steps_per_second": 5.14,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 17.25104331970215,
      "learning_rate": 0.0005,
      "loss": 2.8813,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.22773333333333334,
      "eval_loss": 2.900686025619507,
      "eval_runtime": 8.2695,
      "eval_samples_per_second": 453.473,
      "eval_steps_per_second": 7.135,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 17.61591148376465,
      "learning_rate": 0.0005,
      "loss": 2.7635,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.232,
      "eval_loss": 2.874490976333618,
      "eval_runtime": 7.3631,
      "eval_samples_per_second": 509.294,
      "eval_steps_per_second": 8.013,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 16.097349166870117,
      "learning_rate": 0.0005,
      "loss": 2.6679,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.25333333333333335,
      "eval_loss": 2.7662136554718018,
      "eval_runtime": 7.0395,
      "eval_samples_per_second": 532.705,
      "eval_steps_per_second": 8.381,
      "step": 1665
    }
  ],
  "logging_steps": 500,
  "max_steps": 1665,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 8.2370660311296e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}