trainer_state.json

{
  "best_metric": 0.9266666666666666,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0029/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": 23.200597763061523,
      "learning_rate": 2.85232922088503e-05,
      "loss": 0.8396,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8802666666666666,
      "eval_loss": 0.3961934447288513,
      "eval_runtime": 7.1344,
      "eval_samples_per_second": 525.625,
      "eval_steps_per_second": 8.27,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.0222754068672657,
      "learning_rate": 2.4368144166557088e-05,
      "loss": 0.1824,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8992,
      "eval_loss": 0.3207838535308838,
      "eval_runtime": 6.666,
      "eval_samples_per_second": 562.552,
      "eval_steps_per_second": 8.851,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.1332422494888306,
      "learning_rate": 1.8357520279328874e-05,
      "loss": 0.0603,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9098666666666667,
      "eval_loss": 0.31589362025260925,
      "eval_runtime": 6.1687,
      "eval_samples_per_second": 607.912,
      "eval_steps_per_second": 9.564,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.004292855504900217,
      "learning_rate": 1.1681898323495617e-05,
      "loss": 0.0187,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9165333333333333,
      "eval_loss": 0.33510422706604004,
      "eval_runtime": 6.3739,
      "eval_samples_per_second": 588.333,
      "eval_steps_per_second": 9.256,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.35071006417274475,
      "learning_rate": 5.6634670985405755e-06,
      "loss": 0.0053,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9266666666666666,
      "eval_loss": 0.29088523983955383,
      "eval_runtime": 6.1673,
      "eval_samples_per_second": 608.05,
      "eval_steps_per_second": 9.567,
      "step": 1665
    }
  ],
  "logging_steps": 500,
  "max_steps": 2331,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 8.2370660311296e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}