trainer_state.json

{
  "best_metric": 0.9413333333333334,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0600/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": 7.840142250061035,
      "learning_rate": 2.7151869176294504e-05,
      "loss": 0.77,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8850666666666667,
      "eval_loss": 0.36660000681877136,
      "eval_runtime": 12.0748,
      "eval_samples_per_second": 310.564,
      "eval_steps_per_second": 4.886,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.26226305961608887,
      "learning_rate": 1.9662164055386567e-05,
      "loss": 0.1938,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9226666666666666,
      "eval_loss": 0.2471093088388443,
      "eval_runtime": 12.2451,
      "eval_samples_per_second": 306.245,
      "eval_steps_per_second": 4.818,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.736433982849121,
      "learning_rate": 1.0391670726449015e-05,
      "loss": 0.0692,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9282666666666667,
      "eval_loss": 0.24807706475257874,
      "eval_runtime": 11.7818,
      "eval_samples_per_second": 318.287,
      "eval_steps_per_second": 5.008,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 4.8820905685424805,
      "learning_rate": 2.881402548656824e-06,
      "loss": 0.0202,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9370666666666667,
      "eval_loss": 0.23002243041992188,
      "eval_runtime": 11.8259,
      "eval_samples_per_second": 317.102,
      "eval_steps_per_second": 4.989,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.27799883484840393,
      "learning_rate": 2.670129987614178e-11,
      "loss": 0.0057,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9413333333333334,
      "eval_loss": 0.2178334891796112,
      "eval_runtime": 11.822,
      "eval_samples_per_second": 317.204,
      "eval_steps_per_second": 4.991,
      "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
}