trainer_state.json

{
  "best_metric": 0.9301333333333334,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0676/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": 3.971456289291382,
      "learning_rate": 5.604204204204204e-05,
      "loss": 0.7516,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8629333333333333,
      "eval_loss": 0.42521101236343384,
      "eval_runtime": 13.1738,
      "eval_samples_per_second": 284.656,
      "eval_steps_per_second": 4.479,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.6782960295677185,
      "learning_rate": 4.2042042042042036e-05,
      "loss": 0.2381,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8877333333333334,
      "eval_loss": 0.4008244276046753,
      "eval_runtime": 12.79,
      "eval_samples_per_second": 293.198,
      "eval_steps_per_second": 4.613,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.001374923507682979,
      "learning_rate": 2.8042042042042038e-05,
      "loss": 0.1055,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9082666666666667,
      "eval_loss": 0.3230113685131073,
      "eval_runtime": 12.6996,
      "eval_samples_per_second": 295.285,
      "eval_steps_per_second": 4.646,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 3.1614341735839844,
      "learning_rate": 1.404204204204204e-05,
      "loss": 0.041,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9229333333333334,
      "eval_loss": 0.2985336482524872,
      "eval_runtime": 12.6247,
      "eval_samples_per_second": 297.038,
      "eval_steps_per_second": 4.673,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.2876746952533722,
      "learning_rate": 4.204204204204204e-08,
      "loss": 0.0074,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9301333333333334,
      "eval_loss": 0.2803119122982025,
      "eval_runtime": 12.2079,
      "eval_samples_per_second": 307.178,
      "eval_steps_per_second": 4.833,
      "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
}