trainer_state.json

{
  "best_metric": 0.8392,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0688/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": Infinity,
      "learning_rate": 9e-05,
      "loss": 0.8786,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8114666666666667,
      "eval_loss": 0.5753352642059326,
      "eval_runtime": 10.9996,
      "eval_samples_per_second": 340.922,
      "eval_steps_per_second": 5.364,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.3440783023834229,
      "learning_rate": 9e-05,
      "loss": 0.3717,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8168,
      "eval_loss": 0.6258988380432129,
      "eval_runtime": 11.2741,
      "eval_samples_per_second": 332.621,
      "eval_steps_per_second": 5.233,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 56.5678596496582,
      "learning_rate": 9e-05,
      "loss": 0.2534,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8013333333333333,
      "eval_loss": 0.7489505410194397,
      "eval_runtime": 10.9626,
      "eval_samples_per_second": 342.073,
      "eval_steps_per_second": 5.382,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.009689576923847198,
      "learning_rate": 9e-05,
      "loss": 0.1922,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8272,
      "eval_loss": 0.6436048150062561,
      "eval_runtime": 11.3216,
      "eval_samples_per_second": 331.225,
      "eval_steps_per_second": 5.211,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.756009578704834,
      "learning_rate": 9e-05,
      "loss": 0.1622,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8213333333333334,
      "eval_loss": 0.6780934929847717,
      "eval_runtime": 11.1609,
      "eval_samples_per_second": 335.995,
      "eval_steps_per_second": 5.286,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.05606804043054581,
      "learning_rate": 9e-05,
      "loss": 0.1391,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8392,
      "eval_loss": 0.6631086468696594,
      "eval_runtime": 10.8323,
      "eval_samples_per_second": 346.187,
      "eval_steps_per_second": 5.447,
      "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
}