trainer_state.json

{
  "best_metric": 0.9248,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0443/checkpoints/checkpoint-1998",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 1998,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 41.71299743652344,
      "learning_rate": 4.1691691691691694e-05,
      "loss": 0.7679,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8608,
      "eval_loss": 0.4693056046962738,
      "eval_runtime": 6.2313,
      "eval_samples_per_second": 601.797,
      "eval_steps_per_second": 9.468,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.02235090173780918,
      "learning_rate": 3.335835835835836e-05,
      "loss": 0.1828,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.888,
      "eval_loss": 0.38889190554618835,
      "eval_runtime": 6.1618,
      "eval_samples_per_second": 608.587,
      "eval_steps_per_second": 9.575,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.010264151729643345,
      "learning_rate": 2.502502502502503e-05,
      "loss": 0.0699,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9016,
      "eval_loss": 0.3674212098121643,
      "eval_runtime": 5.9464,
      "eval_samples_per_second": 630.63,
      "eval_steps_per_second": 9.922,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": Infinity,
      "learning_rate": 1.6716716716716717e-05,
      "loss": 0.0281,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9138666666666667,
      "eval_loss": 0.35397329926490784,
      "eval_runtime": 6.024,
      "eval_samples_per_second": 622.509,
      "eval_steps_per_second": 9.794,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.002710888395085931,
      "learning_rate": 8.383383383383384e-06,
      "loss": 0.0079,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9218666666666666,
      "eval_loss": 0.33068209886550903,
      "eval_runtime": 5.9664,
      "eval_samples_per_second": 628.522,
      "eval_steps_per_second": 9.889,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.012941241264343262,
      "learning_rate": 5.005005005005005e-08,
      "loss": 0.0032,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9248,
      "eval_loss": 0.3245643377304077,
      "eval_runtime": 6.0645,
      "eval_samples_per_second": 618.353,
      "eval_steps_per_second": 9.729,
      "step": 1998
    }
  ],
  "logging_steps": 500,
  "max_steps": 1998,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "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": 9.88447923735552e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}