trainer_state.json

{
  "best_metric": 0.5416,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0266/checkpoints/checkpoint-2664",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 2664,
  "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": 0.0002627252252252252,
      "loss": 3.1805,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.2594666666666667,
      "eval_loss": 2.733779191970825,
      "eval_runtime": 11.1394,
      "eval_samples_per_second": 336.642,
      "eval_steps_per_second": 5.296,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 19.42656135559082,
      "learning_rate": 0.0002252252252252252,
      "loss": 2.5743,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.3373333333333333,
      "eval_loss": 2.391808271408081,
      "eval_runtime": 11.4864,
      "eval_samples_per_second": 326.473,
      "eval_steps_per_second": 5.137,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 24.83740997314453,
      "learning_rate": 0.00018772522522522518,
      "loss": 2.2795,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.3701333333333333,
      "eval_loss": 2.2521259784698486,
      "eval_runtime": 11.3107,
      "eval_samples_per_second": 331.545,
      "eval_steps_per_second": 5.216,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 18.17445182800293,
      "learning_rate": 0.0001502252252252252,
      "loss": 2.0364,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.4288,
      "eval_loss": 2.057361364364624,
      "eval_runtime": 11.6032,
      "eval_samples_per_second": 323.188,
      "eval_steps_per_second": 5.085,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 16.651958465576172,
      "learning_rate": 0.00011272522522522522,
      "loss": 1.8113,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.4610666666666667,
      "eval_loss": 1.8833966255187988,
      "eval_runtime": 11.9353,
      "eval_samples_per_second": 314.195,
      "eval_steps_per_second": 4.943,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 40.779823303222656,
      "learning_rate": 7.522522522522523e-05,
      "loss": 1.5631,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.5008,
      "eval_loss": 1.8044787645339966,
      "eval_runtime": 11.6721,
      "eval_samples_per_second": 321.278,
      "eval_steps_per_second": 5.055,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 20.69434928894043,
      "learning_rate": 3.7725225225225225e-05,
      "loss": 1.2965,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.5256,
      "eval_loss": 1.7276753187179565,
      "eval_runtime": 11.4001,
      "eval_samples_per_second": 328.943,
      "eval_steps_per_second": 5.175,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": Infinity,
      "learning_rate": 3.378378378378378e-07,
      "loss": 1.0335,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.5416,
      "eval_loss": 1.7681281566619873,
      "eval_runtime": 11.297,
      "eval_samples_per_second": 331.948,
      "eval_steps_per_second": 5.223,
      "step": 2664
    }
  ],
  "logging_steps": 500,
  "max_steps": 2664,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "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": 1.317930564980736e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}