trainer_state.json

{
  "best_metric": 0.9029333333333334,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0468/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": 12.01774787902832,
      "learning_rate": 3e-05,
      "loss": 0.7805,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8978666666666667,
      "eval_loss": 0.34581413865089417,
      "eval_runtime": 12.3408,
      "eval_samples_per_second": 303.869,
      "eval_steps_per_second": 4.781,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 3e-05,
      "loss": 0.1842,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.892,
      "eval_loss": 0.3449442684650421,
      "eval_runtime": 11.0168,
      "eval_samples_per_second": 340.391,
      "eval_steps_per_second": 5.355,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.08555743098258972,
      "learning_rate": 3e-05,
      "loss": 0.0764,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8994666666666666,
      "eval_loss": 0.3483298122882843,
      "eval_runtime": 11.3075,
      "eval_samples_per_second": 331.638,
      "eval_steps_per_second": 5.218,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.052189912647008896,
      "learning_rate": 3e-05,
      "loss": 0.05,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8992,
      "eval_loss": 0.39224061369895935,
      "eval_runtime": 11.293,
      "eval_samples_per_second": 332.066,
      "eval_steps_per_second": 5.224,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.11208364367485046,
      "learning_rate": 3e-05,
      "loss": 0.0367,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9029333333333334,
      "eval_loss": 0.4006660282611847,
      "eval_runtime": 11.0655,
      "eval_samples_per_second": 338.893,
      "eval_steps_per_second": 5.332,
      "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
}