trainer_state.json

{
  "best_metric": 0.28613333333333335,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0629/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": 0.0005,
      "loss": 3.4583,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.17653333333333332,
      "eval_loss": 3.140031337738037,
      "eval_runtime": 10.1988,
      "eval_samples_per_second": 367.689,
      "eval_steps_per_second": 5.785,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0005,
      "loss": 3.0252,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.2112,
      "eval_loss": 2.946561574935913,
      "eval_runtime": 10.4178,
      "eval_samples_per_second": 359.959,
      "eval_steps_per_second": 5.663,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 21.62390899658203,
      "learning_rate": 0.0005,
      "loss": 2.8713,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.24693333333333334,
      "eval_loss": 2.8636884689331055,
      "eval_runtime": 11.4434,
      "eval_samples_per_second": 327.7,
      "eval_steps_per_second": 5.156,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 24.78705406188965,
      "learning_rate": 0.0005,
      "loss": 2.7299,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.2624,
      "eval_loss": 2.726522445678711,
      "eval_runtime": 11.0794,
      "eval_samples_per_second": 338.466,
      "eval_steps_per_second": 5.325,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 14.857529640197754,
      "learning_rate": 0.0005,
      "loss": 2.6152,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.28613333333333335,
      "eval_loss": 2.631291389465332,
      "eval_runtime": 11.0699,
      "eval_samples_per_second": 338.757,
      "eval_steps_per_second": 5.33,
      "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
}