trainer_state.json

{
  "best_metric": 0.9208,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0545/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": 5.84034034034034e-05,
      "loss": 0.8694,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8434666666666667,
      "eval_loss": 0.4954388737678528,
      "eval_runtime": 11.1944,
      "eval_samples_per_second": 334.988,
      "eval_steps_per_second": 5.27,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.008236194960772991,
      "learning_rate": 4.6736736736736736e-05,
      "loss": 0.2426,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8797333333333334,
      "eval_loss": 0.41863977909088135,
      "eval_runtime": 13.8939,
      "eval_samples_per_second": 269.902,
      "eval_steps_per_second": 4.246,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 7.113038063049316,
      "learning_rate": 3.507007007007007e-05,
      "loss": 0.105,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8810666666666667,
      "eval_loss": 0.44186532497406006,
      "eval_runtime": 11.0567,
      "eval_samples_per_second": 339.161,
      "eval_steps_per_second": 5.336,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.002348785288631916,
      "learning_rate": 2.3403403403403404e-05,
      "loss": 0.0463,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8994666666666666,
      "eval_loss": 0.37570565938949585,
      "eval_runtime": 11.0268,
      "eval_samples_per_second": 340.081,
      "eval_steps_per_second": 5.351,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.015271172858774662,
      "learning_rate": 1.1736736736736736e-05,
      "loss": 0.0148,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9056,
      "eval_loss": 0.37674063444137573,
      "eval_runtime": 10.9551,
      "eval_samples_per_second": 342.307,
      "eval_steps_per_second": 5.386,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.00011422853276599199,
      "learning_rate": 7.007007007007007e-08,
      "loss": 0.0046,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9208,
      "eval_loss": 0.3395988941192627,
      "eval_runtime": 12.9728,
      "eval_samples_per_second": 289.066,
      "eval_steps_per_second": 4.548,
      "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
}