trainer_state.json

{
  "best_metric": 0.932,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0849/checkpoints/checkpoint-2664",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 2664,
  "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": 2.6272522522522523e-05,
      "loss": 0.7825,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8909333333333334,
      "eval_loss": 0.3434804677963257,
      "eval_runtime": 10.3361,
      "eval_samples_per_second": 362.807,
      "eval_steps_per_second": 5.708,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.6966972351074219,
      "learning_rate": 2.2522522522522523e-05,
      "loss": 0.1487,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9002666666666667,
      "eval_loss": 0.3227052688598633,
      "eval_runtime": 10.1598,
      "eval_samples_per_second": 369.1,
      "eval_steps_per_second": 5.807,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 9.850299835205078,
      "learning_rate": 1.8772522522522523e-05,
      "loss": 0.053,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9144,
      "eval_loss": 0.29001492261886597,
      "eval_runtime": 10.2141,
      "eval_samples_per_second": 367.139,
      "eval_steps_per_second": 5.776,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.032224901020526886,
      "learning_rate": 1.5022522522522522e-05,
      "loss": 0.019,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9224,
      "eval_loss": 0.2970980405807495,
      "eval_runtime": 10.1544,
      "eval_samples_per_second": 369.297,
      "eval_steps_per_second": 5.81,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.013558613136410713,
      "learning_rate": 1.1272522522522524e-05,
      "loss": 0.0038,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9258666666666666,
      "eval_loss": 0.30903157591819763,
      "eval_runtime": 10.2672,
      "eval_samples_per_second": 365.242,
      "eval_steps_per_second": 5.746,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.0034554717130959034,
      "learning_rate": 7.522522522522523e-06,
      "loss": 0.0012,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9298666666666666,
      "eval_loss": 0.2996990382671356,
      "eval_runtime": 10.2178,
      "eval_samples_per_second": 367.007,
      "eval_steps_per_second": 5.774,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.0032182452268898487,
      "learning_rate": 3.772522522522523e-06,
      "loss": 0.0005,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9309333333333333,
      "eval_loss": 0.2987803518772125,
      "eval_runtime": 10.1277,
      "eval_samples_per_second": 370.27,
      "eval_steps_per_second": 5.826,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.017047923058271408,
      "learning_rate": 2.2522522522522524e-08,
      "loss": 0.0002,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.932,
      "eval_loss": 0.2998943328857422,
      "eval_runtime": 10.227,
      "eval_samples_per_second": 366.677,
      "eval_steps_per_second": 5.769,
      "step": 2664
    }
  ],
  "logging_steps": 500,
  "max_steps": 2664,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "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": 1.317930564980736e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}