trainer_state.json

{
  "best_metric": 0.3584,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0777/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": 24.9738826751709,
      "learning_rate": 0.0003,
      "loss": 3.342,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.204,
      "eval_loss": 3.014453649520874,
      "eval_runtime": 13.5081,
      "eval_samples_per_second": 277.611,
      "eval_steps_per_second": 4.368,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 20.792720794677734,
      "learning_rate": 0.0003,
      "loss": 2.8015,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.2557333333333333,
      "eval_loss": 2.7573230266571045,
      "eval_runtime": 13.4736,
      "eval_samples_per_second": 278.322,
      "eval_steps_per_second": 4.379,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 21.414352416992188,
      "learning_rate": 0.0003,
      "loss": 2.5985,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.2976,
      "eval_loss": 2.5624492168426514,
      "eval_runtime": 13.6063,
      "eval_samples_per_second": 275.607,
      "eval_steps_per_second": 4.336,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 28.414926528930664,
      "learning_rate": 0.0003,
      "loss": 2.438,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.324,
      "eval_loss": 2.4382827281951904,
      "eval_runtime": 12.9972,
      "eval_samples_per_second": 288.524,
      "eval_steps_per_second": 4.539,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 24.37203598022461,
      "learning_rate": 0.0003,
      "loss": 2.3182,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.3450666666666667,
      "eval_loss": 2.383817672729492,
      "eval_runtime": 13.4496,
      "eval_samples_per_second": 278.819,
      "eval_steps_per_second": 4.387,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 22.887929916381836,
      "learning_rate": 0.0003,
      "loss": 2.1925,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.3584,
      "eval_loss": 2.325667142868042,
      "eval_runtime": 13.4144,
      "eval_samples_per_second": 279.551,
      "eval_steps_per_second": 4.398,
      "step": 1998
    }
  ],
  "logging_steps": 500,
  "max_steps": 2331,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "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
}