trainer_state.json

{
  "best_metric": 0.9096,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0874/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": 3e-05,
      "loss": 0.7833,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8714666666666666,
      "eval_loss": 0.4259362816810608,
      "eval_runtime": 11.1415,
      "eval_samples_per_second": 336.579,
      "eval_steps_per_second": 5.296,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.15657202899456024,
      "learning_rate": 3e-05,
      "loss": 0.1846,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8904,
      "eval_loss": 0.3618867099285126,
      "eval_runtime": 11.1184,
      "eval_samples_per_second": 337.278,
      "eval_steps_per_second": 5.307,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.034433525055646896,
      "learning_rate": 3e-05,
      "loss": 0.0881,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9037333333333334,
      "eval_loss": 0.332414448261261,
      "eval_runtime": 11.0673,
      "eval_samples_per_second": 338.835,
      "eval_steps_per_second": 5.331,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.7911471128463745,
      "learning_rate": 3e-05,
      "loss": 0.0562,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8893333333333333,
      "eval_loss": 0.44924014806747437,
      "eval_runtime": 11.015,
      "eval_samples_per_second": 340.445,
      "eval_steps_per_second": 5.356,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.01970021426677704,
      "learning_rate": 3e-05,
      "loss": 0.0407,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9096,
      "eval_loss": 0.3530827462673187,
      "eval_runtime": 11.0521,
      "eval_samples_per_second": 339.301,
      "eval_steps_per_second": 5.338,
      "step": 1665
    }
  ],
  "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": 8.2370660311296e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}